Split (1)

train · ~9.11M rows (showing the first 93.5k)

identifier stringlengths 11 32	pdf_url stringlengths 17 4.62k ⌀	lang stringclasses 120 values	error stringclasses 1 value	title stringlengths 2 500 ⌀	source_name stringlengths 1 435 ⌀	publication_year float64 1.9k 2.02k	license stringclasses 3 values	word_count int64 0 1.64M	text stringlengths 1 9.75M
https://openalex.org/W2061976857	https://www.openaccessrepository.it/record/27817/files/fulltext.pdf	English	null	Wireless Cooperative Networks	EURASIP Journal on Advances in Signal Processing	2,008	cc-by	1,898	Hindawi Publishing Corporation EURASIP Journal on Advances in Signal Processing Volume 2008, Article ID 810149, 2 pages doi:10.1155/2008/810149 Hindawi Publishing Corporation EURASIP Journal on Advances in Signal Processing Volume 2008, Article ID 810149, 2 pages doi:10.1155/2008/810149 Hindawi Publishing Corporation EURASIP Journal on Advances in Signal Processing Volume 2008, Article ID 810149, 2 pages doi:10.1155/2008/810149 Correspondence should be addressed to Andrea Conti, a.conti@ieee.org Correspondence should be addressed to Andrea Conti, a.conti@ieee.org Received 29 July 2008; Accepted 29 July 2008 Received 29 July 2008; Accepted 29 July 2008 Copyright © 2008 Andrea Conti et al. This is an open access article distributed under the Creative Commons A which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is Copyright © 2008 Andrea Conti et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. outage, delay, energy, scaling laws); cross-layer issues, for example, PHY/MAC/NET interactions, joint source-channel coding, separation theorems; multiterminal information theory; multihop communications; integration of wireless heterogeneous (long- and short-range) systems. Cooperative networks are gaining an increasing interest in information and communications technologies since such networks can improve communication capability and provide a fertile environment for the development of context-aware services. Cooperative communications and networking represent a new paradigm which involves both transmission and distributed processing, promising signiﬁcant increase of capacity and diversity gain in wireless networks. From one hand, the integration of long-range and short-range wireless communication networks (e.g., infras- tructured networks such as 3G, wireless ad hoc networks, and wireless sensor networks) improves the performance in terms of both area coverage and quality of service (QoS). On the other hand, the cooperation among nodes, as in the case of wireless sensor networks, allows a distributed space-time signal processing which enables, among others, environmental monitoring, localization techniques, and distributed measurements, with a reduced complexity or energy consumption per node. The relevance of this topic is also reﬂected by numerous technical sessions in current international conferences as well as by the increasing number of national and international projects on these aspects. Cooperative networks are gaining an increasing interest in information and communications technologies since such networks can improve communication capability and provide a fertile environment for the development of context-aware services. Cooperative communications and networking represent a new paradigm which involves both transmission and distributed processing, promising signiﬁcant increase of capacity and diversity gain in wireless networks. From one hand, the integration of long-range and short-range wireless communication networks (e.g., infras- tructured networks such as 3G, wireless ad hoc networks, and wireless sensor networks) improves the performance in terms of both area coverage and quality of service (QoS). Andrea Conti,1 Jiangzhou Wang,2 Hyundong Shin,3 Ramesh Annavajjala,4 and Moe Z. Win , g g, y g , jj , 1Engineering Department in Ferrara (ENDIF), University of Ferrara, 44100 Ferrara, Italy 2Department of Electronics, University of Kent, Canterbury, Kent CT2 7NT, UK 3Department of Radio Communication Engineering, School of Electronics and Information, Kyung Hee University, Gueonggi-Do 449-701, South Korea 4Mitsubishi Electric Research Laboratories, 201 Broadway, Cambridge, MA 02139, USA 5Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA 02139, USA 1Engineering Department in Ferrara (ENDIF), University of Ferrara, 44100 Ferrara, Italy p f y f y 3Department of Radio Communication Engineering, School of Electronics and Information, Kyung Hee University, Gueonggi-Do 449-701, South Korea p f y f y 3Department of Radio Communication Engineering, School of Electronics and Information, Kyung Hee University Gueonggi-Do 449-701, South Korea gg 4Mitsubishi Electric Research Laboratories, 201 Broadway, Cambridge, MA 02139, USA 5Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA 02139, USA 4Mitsubishi Electric Research Laboratories, 201 Broadway, Cambridge, MA 02139, USA 5Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA 02139, USA Correspondence should be addressed to Andrea Conti, a.conti@ieee.org The underlying idea of the PRCSMA protocol is to modify the basic rules of the IEEE 802.11 MAC protocol to execute a distributed cooperative ARQ scheme in wireless networks to enhance their performance and to extend coverage. destination node directly and indirectly (through multiple relays). It is shown that the best-relay selection reduces the amount of required resources while improving the performance. Authors derive closed form expressions for tight lower bounds on the symbol error probability and outage probability. g p y A double-diﬀerential modulation for the amplify-and- forward protocol over Nakagami-m fading channels with carrier oﬀsets is proposed by M. R. Bhatnagar et al. in “Cooperative communications over ﬂat fading channels with carrier oﬀsets: a double-diﬀerential modulation approach.” They propose an emulated maximum ratio combining (EMRC) decoder, which could be used by the double- diﬀerential receiver in the absence of exact channel knowl- edge. Approximate bit error rate (BER) analysis is performed for the double-diﬀerential modulation-based cooperative communication system. They propose a double-diﬀerential system which is immune to random carrier oﬀsets, whereas the conventional single-diﬀerential modulation-based coop- erative system breaks down, and perform better than training-based cooperative system which utilizes training data to estimate carrier oﬀsets and channel gains. At the scheduling level, “Optimally joint subcarrier matching and power allocation in OFDM multihop system” by W. Wang et al. propose an optimally joint subcarrier matching and power allocation scheme to maximize the channel capacity under total system power constrain of OFDM systems. The problem is formulated as a mixed binary integer programming problem (which is prohibitive to ﬁnd the global optimum in terms of complexity) and then a low-complexity scheme by making use of the equivalent channel power gain for any matched subcarrier pair is proposed. W. Mesbah and T. N. Davidson study, in “Power and resource allocation for orthogonal multiple access relay systems,” the problem of joint power and channel resource allocation for orthogonal multiple access relay (MAR) systems to maximize the achievable rate region. The authors consider four relaying strategies and show that the problem can be formulated as a quasiconvex problem in several cases. Therefore, eﬃcient algorithms can be derived for joint optimal power and channel resource allocation. In “Delay optimization in cooperative relaying with cyclic delay diversity,” S. B. Slimane et al. propose to inserting random delays at the nonregenerative ﬁxed relays to further improve the system performance. Correspondence should be addressed to Andrea Conti, a.conti@ieee.org On the other hand, the cooperation among nodes, as in the case of wireless sensor networks, allows a distributed space-time signal processing which enables, among others, environmental monitoring, localization techniques, and distributed measurements, with a reduced complexity or energy consumption per node. The relevance of this topic is also reﬂected by numerous technical sessions in current international conferences as well as by the increasing number of national and international projects on these aspects. In “Asymptotic analysis of large cooperative relay net- works using random matrix theory” by H. Li et al., coopera- tive relay networks with large number of nodes are analyzed, and in particular the asymptotic performance improvement of cooperative transmission over direct transmission and relay transmission is analyzed using random matrix theory. The key idea is to investigate the eigenvalue distributions related to channel capacity and to analyze the moments of this distribution in large wireless networks. The analysis in this paper provides important tools for the understanding and the design of large cooperative wireless networks. H. Van Khuong and T. Le-Ngoc propose, in the paper “Bandwidth-eﬃcient cooperative relaying schemes with multi-antenna relay,” coded cooperative relaying schemes in which all successfully decoded signals from multiple sources are forwarded simultaneously by a multiantenna relay to a common multiantenna destination to increase bandwidth eﬃciency. These schemes facilitate various retransmission strategies at relay together with single-user and multiuser iterative decoding techniques at destination, suitable for tradeoﬀs between performance, latency, and complexity. This special issue aims to collect cutting-edge research achievements in this area. We solicited papers that present original and unpublished work on topics including, but not limited to, the following: physical layer models, for example, channel models (statistics, fading, MIMO, feed- back); device constraints (power, energy, multiple access, synchronization) and resource management; distributed processing for cooperative networks (e.g., distributed com- pression in wireless sensor networks, channel and network codes design); performance metrics (e.g., capacity, cost, The problem of choosing the best relay node in relaying networks is addressed in “Performance of multiple-relay cooperative diversity systems with best-relay selection over rayleigh fading channels” by S. S. Ikki and M. H. Ahmed. They consider an amplify-and-forward (AF) cooperative diversity system where a source node communicates with a EURASIP Journal on Advances in Signal Processing 2 persistent relay carrier sensing multiple access (PRCSMA) protocol that allows for the execution of a distributed cooperative automatic retransmission request (ARQ) scheme in IEEE 802.11 wireless networks. Correspondence should be addressed to Andrea Conti, a.conti@ieee.org However, random delays result in limited performance gain from multipath diversity. In this paper, two promising delay optimization schemes are introduced for a multicellular OFDM system with cooperative relaying, stationary multiple users, and ﬁxed relays. At networking level, the paper “Resource sharing via planed relay for HWN” by C. Shen et al. presents an improved version of adaptive distributed cross-layer routing algorithm for hybrid wireless network with dedicated relay stations. They verify that the performance of routing proto- col beneﬁts of the hybrid wireless networks nature. y A. Conti, V. Tralli, and M. Chiani address the construc- tion of space-time codes for cooperative communications over block fading channels in the paper “Pragmatic space- time codes for cooperative relaying in block fading channels.” They consider a pragmatic approach based on the concate- nation of convolutional codes and binary/quaternary phase shift keying modulation to obtain cooperative codes for relay networks. The pairwise error probability, an asymptotic bound on the frame error probability, and a design criterion to optimize both diversity and coding gain are also derived. While the implementation of pragmatic space-time codes only requires common convolutional encoders and Viterbi decoders with suitable generators, rate, and branch metric, they perform well in block fading channels, including quasistatic channel, even with a low number of states and relays. Collaboration in heterogeneous wireless networks is addressed by A. Bazzi et al. a in “Multi radio resource management: parallel transmission for higher throughput.” Mobile communication systems beyond the third generation will see the interconnection of heterogeneous radio access networks (UMTS, WiMax, wireless local area networks, etc.) to always provide the best QoS to users with multimode terminals. The issue of parallel transmission over multiple radio access technologies (RATs) is investigated focusing the attention on the QoS perceived by the end users. It shows the real beneﬁt of parallel transmission over multiple RATs and how it is conditioned to the fulﬁlment of some requirements related to the particular kind of RATs, the multiradio resource management strategy, and the transport level protocol behavior. In “Interference mitigation in cooperative SFBC- OFDM,” D. Sreedhar and A. Chockalingam consider coop- erative space-frequency block-coded orthogonal frequency- division multiplexing (SFBC-OFDM) networks with amplify-and-forward (AF) and decode-and-forward (DF) protocols at the relays. They propose an interference cancel- lation algorithm for this system at the destination node, and show that the proposed algorithm eﬀectively mitigates the intersymbol interference and intercarrier interference eﬀects. Correspondence should be addressed to Andrea Conti, a.conti@ieee.org Andrea Conti Jiangzhou Wang Hyundong Shin Ramesh Annavajjala Moe Z. Win For what concerns the MAC layer, J. Alonso-Zrate et al., in “Persistent RCSMA: a MAC protocol for a distributed cooperative ARQ scheme in wireless networks,” present the
W2096102553.txt	https://basepub.dauphine.psl.eu/bitstream/123456789/4578/1/2001-15.pdf	fr		Taxation optimale de la consommation et biens informels	Revue économique/Revue économique	2,002	cc-by-sa	12,482	DOCUMENT DE TRAVAIL DT/2001/15 Taxation optimale de la consommation et biens informels Jean-François GAUTIER TAXATION OPTIMALE DE LA CONSOMMATION ET BIENS INFORMELS 1 Jean-François Gautier (Université Paris IX et DIAL) e-mail : gautier1@caramail.com Document de travail DIAL / Unité de Recherche CIPRE Octobre 2001 RESUME La question posée par ce papier est la suivante : dans quelle mesure est-il souhaitable de taxer le secteur informel ? Un modèle théorique de taxation optimale permet d’abord de fournir une réponse analytique à cette question. Une version empirique du modèle permet ensuite d’estimer les taux de taxation optimaux à Madagascar. On montre ainsi que la taxation optimale des biens informels est toujours très inférieure à celle des biens formels. Néanmoins, pour des degrés de progressivité « raisonnables », le taux de taxation optimal moyen du secteur informel est nettement plus important que le taux effectif actuel. Toutefois, la solution optimale peut être de subventionner le secteur informel pour des degrés élevés de progressivité. ABSTRACT This paper analyses whether it is worth taxing the black market. A theoretical optimal tax model provides a first analytical discussion. An empirical version allows us to estimate an optimal tax system for Madagascar. For reasonable inequality aversion parameters, the taxation’s rate of the informal good is always quite below the formal good’s one. Nevertheless, tax rates on informal goods are above the actual rates. Informal goods could be optimally subsidised for higher inequality aversion parameters. Classification JEL : H21 (Efficiency, Optimal Taxation) ; H26 (Tax Evasion). 1 Je remercie Rachel Ravelosoa, François Roubaud et Charlotte Guénard d’avoir bien voulu me communiquer les données et documents techniques de l’enquête budget - consommation des ménages à Antananarivo (MADIO 1995b), ainsi que l’équipe DIAL pour les commentaires apportés à une version précédente de ce papier. 2 Table des matières INTRODUCTION............................................................................................................................... 4 1. FISCALITE DES BIENS ET CONSOMMATION DES MENAGES .............................. 5 1.1. Taux de taxation effectifs de la TVA................................................................................... 5 1.1.1. Structure des recettes.................................................................................................... 5 2. LE MODELE.......................................................................................................................... 7 2.1. Les hypothèses et les agents du modèle............................................................................... 7 2.1.1. Le marché des biens ..................................................................................................... 7 2.1.2. Les ménages ................................................................................................................. 8 2.1.3. L’Etat............................................................................................................................ 8 2.2. La solution optimale (locale) de ti ........................................................................................ 9 2.2.1. Relation entre ti et la dérivée de la demande compensée aux prix ( S bhi ,bi )................. 11 2.2.2. Relation entre ti et la dérivée de la demande au revenu. ............................................ 11 2.2.3. Relation entre ti et le poids social des ménages ......................................................... 11 2.2.4. Relation entre ti et le taux de couverture du secteur informel β i. ............................... 12 2.3. La solution optimale (locale) de β i..................................................................................... 12 2.3.1. Commentaires, cas d’une taxation optimale du secteur informel.............................. 13 2.3.2. Cas où il est optimal de subventionner le secteur informel........................................ 13 3. MODELE EMPIRIQUE ET SIMULATIONS.................................................................. 14 3.1. Le modèle empirique.......................................................................................................... 14 3.1.1. Poids social des ménages ........................................................................................... 15 3.1.2. Formes fonctionnelles des solutions de ti, β i et λ....................................................... 15 3.2. Les fonctions de demande de biens .................................................................................... 16 3.2.1. Résolution du système................................................................................................ 17 3.2.2. Paramétrage du modèle. ............................................................................................. 17 3.3. Résultats des simulations ................................................................................................... 19 3.3.1. Taxation des biens formels versus taxation des biens informels. .............................. 20 3.3.2. Résultats sur la taxation des biens et services............................................................ 21 CONCLUSION ................................................................................................................................. 23 REFERENCES................................................................................................................................. 25 ANNEXE........................................................................................................................................... 27 Liste des tableaux Tableau n° 1-1 : Taux de pression estimé de la TVA par produits...................................................... 6 Tableau n° 1-2 : Structure de la consommation par quintiles de dépenses....................................... 6 Tableau n° 3-1 : Elasticités prix et revenus et distribution de la demande....................................... 18 Tableau n° 3-2 : Taux de taxation optimaux avec demande endogène ............................................. 23 Liste des Annexes Annexe n° 1 : Variation de la demande et des prix............................................................................ 27 Annexe n° 2 : Structure de la consommation et taux de pression par déciles de dépenses ............... 27 3 INTRODUCTION L’objet de ce papier est d’estimer une structure optimale de taux de taxation sur les biens de consommation et les services. L’originalité est de prendre en compte la dualité sur le marché des biens et services, en distinguant les biens formels des biens informels. L’intérêt d’intégrer cette dualité réside dans le fait que les marchés des biens et services dans les PED sont divisés entre un marché formel, assujetti aux impôts, et un marché informel faiblement taxé. Par ailleurs, la demande relative pour chaque type de bien (formel / informel) varie avec les revenus des consommateurs. Cette dualité entraîne ainsi une distorsion sur les marchés qui théoriquement doit être à l’origine d’une perte d’efficacité de l’économie. Elle entraîne en effet une réallocation « artificielle » des facteurs de production vers le secteur informel et motive de plus les comportements de fraude fiscale. Ainsi, taxer les biens informels permettrait par exemple, à revenu de l’Etat constant, de diminuer les taux de taxation nominaux sur le secteur formel. Cette baisse des taux réduirait la taille du « poids mort » engendré par la fiscalité sur les marchés formels et par conséquent améliorerait l’efficacité du système de taxation. Si du point de vue de l’efficacité des marchés, l’Etat doit rechercher la taxation du secteur informel, le principe de redistribution devrait en revanche l’en écarter. En effet, le bien informel semble être un bien nécessaire, voire inférieur dans les PED, i.e. dont la part dans la consommation décroît avec le revenu des ménages [MADIO (1995b), Roubaud (1994), Cogneau et al. (1996), Boyabé (1999)]. La taxation du bien informel accroîtrait donc principalement la taxation du budget des ménages les plus démunis. D’autre part, les modèles de fiscalité optimale avec fraude [voir notamment Kaplow (1990) et Cremer et Gahvary (1993), voir Ray (1997) pour une revue] montrent que la taxation des marchés avec fraude doit être inférieure à celle des marchés sans fraude. L’existence d’un coût lié à la fraude, croissant avec la taxation, explique ce résultat qui permet alors de réduire le poids mort sur le marché frauduleux (critère d’efficacité). Toutefois ces modèles ne prennent pas en compte la substituabilité des biens entre les marchés formels et informels. La perte d’efficacité sur les marchés formels est alors sous-estimée. En outre, leur cadre d’agent représentatif ne leur permet pas d’étudier les effets redistributifs de la taxation de la consommation. En tenant compte des deux critères d’efficacité et d’équité, nous allons chercher dans ce papier à définir une fiscalité optimale qui permettra d’estimer dans quelle mesure l’Etat doit fiscaliser le secteur informel par rapport au secteur formel. Pour traiter ces questions, nous avons choisi de suivre le plan suivant : Une première partie présente les taux de taxation des biens formels et informels, et la structure de la consommation par revenu, qui servent à calibrer notre modèle. Les statistiques utilisées sont : l’enquête consommation des ménages à Antananarivo et l’enquête sur le secteur informel (MADIO 1995). Nous traitons aussi les données de l’administration fiscale sur les recettes de TVA prélevées dans l’agglomération d’Antananarivo en 1998. Nous construisons ensuite un modèle théorique de taxation optimale avec un vecteur de biens formels et un autre de biens informels. Nous développons analytiquement ce modèle de manière à présenter comment les taux de taxation des secteurs formels et informels sont influencés par la typologie des biens. Une version empirique de ce modèle est ensuite présentée. Elle permet d’estimer un système de taxation optimale de la consommation sur la base d’enquête sur les ménages à Madagascar (MADIO 1995b) et de valeurs d’élasticités prix et revenus tirées de la littérature. 4 1. FISCALITE DES BIENS ET CONSOMMATION DES MENAGES Cette première partie a pour but de présenter le système actuel de TVA à Madagascar. On cherche essentiellement à estimer comment les exonérations et le secteur informel jouent sur le taux de pression effectif des biens (par rapport aux taux prévus par la loi). En utilisant les enquêtes sur la consommation des ménages, on produit alors une première information sur le caractère redistributif de la TVA. 1.1. Taux de taxation effectifs de la TVA 2 1.1.1. Structure des recettes La TVA est actuellement la première source de revenus fiscaux à Madagascar, avec 31 % du total des recettes fiscales de l’administration centrale. Toutefois ces recettes représentaient en 1998 un peu moins de 3 % du PIB. Elle est appliquée à un taux unique de 20 % (depuis la loi de finances de 1996), accompagnée d’exonérations permettant de ne pas augmenter le prix des produits « sensibles ». Le Code Général des Impôts (CGI) prévoit principalement l’exonération des produits alimentaires non transformés, les secteurs de l’éducation et de la santé 3 . Il semble de plus que des exonérations puissent être octroyées à des entreprises de manière discrétionnaire. In fine, on estime à partir des fichiers de l’administration fiscale que 40 % du chiffre d’affaires des entreprises formelles est actuellement exonéré. Bien que le système de TVA prévu par la loi soit relativement simple, l’accumulation des exonérations et autres régimes dérogatoires sont à l’origine de taux de taxation multiples (cf. tableau 1-1). Les exonérations modifient le schéma de la TVA de la manière suivante : - D’une part, elles ont pour effet de réduire le taux de taxation des biens normalement taxés au taux de 20 % (lorsqu’elles sont accordées de manière nominative à des entreprises). Par exemple, le taux de pression moyen sur l’habillement qui devrait normalement être de 16,7 % (avec exonérations sur produits uniquement), ne serait en réalité que de 3,7 % sur les biens formels. - D’autre part, les secteurs exonérés supportent une charge rémanente de TVA dès qu’ils se fournissent auprès d’entreprises assujetties (ces entreprises n’étant pas autorisées à récupérer la TVA payée, cf. Gautier 1999). Par exemple, le tableau 1-1 montre que les céréaliers, l’éducation, la santé qui sont normalement des secteurs exonérés, supportent toutefois une TVA. In fine la somme de ces mécanismes réduit le taux de pression de la TVA sur les biens et services formels. Le taux effectif moyen des produits formels, qui est actuellement de 8,0 %, serait de 11,3 % si le code général des impôts était parfaitement appliqué (sans rémanences et sans exonérations nominales). En outre, le taux de taxation rémanent des biens informels serait de 0,3 %. In fine, le taux de pression moyen, qui devrait être de 7,5 % lorsqu’il est estimé sur la base des termes du CGI, est en réalité de 3,2 % dans l’agglomération d’Antananarivo 4 . 2 La méthode d’estimation des taux effectifs de TVA et celle d’estimation de la consommation des biens informels par les ménages sont présentées en annexe de ce papier. 3 Cf. encadré en annexe pour plus de détail. 4 Ce résultat semble cohérent avec le taux de pression calculé à partir du TOFE et du PIB qui est de 2,8 % sur l’ensemble du territoire. 5 Tableau n° 1-1 : Taux de pression5 estimée de la TVA par produits Biens Riz, céréales Fruits, légumes, élevage Alimentation transformée Habillement Logement Equipement, entretien maison Santé Transport Education Autres biens & services Total Global 0.1 % 0.1 % 6.7 % 2.8 % 0.8 % 6.0 % 0.6 % 8.6 % 0.8 % 5.3 % 3.2 % Taux effectif Sur biens informels 0.0 % 0.0 % 1.1 % 1.2 % 0.5 % 1.5 % 0.0 % 0.0 % 0.0 % 0.3 % 0.3 % Sur biens formels 2.7 % 1.7 % 11.9 % 3.8 % 1.3 % 10.8 % 0.8 % 14.1 % 1.4 % 14.2 % 8.0 % Taux sur la base du code des impôts 0.0 % 0.3 % 12.5 % 16.7 % 6.1 % 16.3 % 1.1 % 16.7 % 0.0 % 11.1 % 7.5 % Sources : Centre Fiscal des Entreprises, MADIO, calculs propres, estimés à partir des données fiscales de 1998. Le tableau 1-2 permet de présenter l’évolution de la structure de consommation des ménages en fonction de leur revenu. On observe que : - La part budgétaire des biens et services informels est une fonction décroissante des dépenses des ménages. Si elle représente un peu moins de 76 % de la consommation du premier quintile de dépense, cette part n’est plus que de 55 % pour le dernier quintile. Les biens informels seraient donc des biens nécessaires, voir inférieures. De fait, la consommation de biens informels doit favoriser la progressivité du système de taxation, ces produits étant moins taxés que les biens formels. - Les préférences des consommateurs évoluent avec leur niveau total de dépenses. Si les biens alimentaires sont des biens nécessaires ; l’habillement, le transport, la santé, l’éducation et les autres biens et services sont visiblement des biens de luxe. La part du logement et des équipements reste stable dans la consommation. Toutefois, l’apparente stabilité du logement peut être biaisée par le fait que les loyers n’ont pas été imputés au budget des ménages propriétaires de leur logement. Ne pas imputer les loyers est justifié dans notre cas puisque nous nous intéressons uniquement aux dépenses monétaires effectivement ou potentiellement taxées. Tableau n° 1-2 : Structure de la consommation par quintiles de dépenses Quintiles de dépenses Riz, céréales Fruits, légumes, élevage Alimentation transformée Habillement Logement Equipement, entretien maison Santé Transport Education Autres biens & services Part des biens informels Taux de taxation moyen 1 31.1 % 18.9 % 15.9 % 5.6 % 13.3 % 5.2 % 3.2 % 4.1 % 1.7 % 1.0 % 75.8 % 2.0 % 2 28.5 % 20.4 % 13.5 % 5.8 % 14.1 % 5.9 % 3.4 % 4.6 % 2.3 % 1.6 % 72.3 % 2.3 % 3 24.3 % 21.9 % 14.2 % 5.7 % 14.5 % 5.6 % 5.1 % 4.2 % 3.3 % 1.4 % 67.7 % 2.3 % 4 20.5 % 21.8 % 14.6 % 6.6 % 13.8 % 5.7 % 3.9 % 6.0 % 4.9 % 2.2 % 63.7 % 2.6 % 5 8.1 % 16.9 % 10.7 % 7.9 % 12.2 % 5.3 % 5.4 % 24.7 % 3.7 % 5.2 % 55.0 % 3.8 % Sources : Centre Fiscal des Entreprises, MADIO, calculs propres, estimés à partir des données fiscales de 1998. * ni les loyers des ménages propriétaires de leur logement, ni l’autoconsommation n’ont été imputés dans les dépenses des ménages. 5 Le taux de pression est calculé sur le prix au consommateur : Tx conso = Tx producteur / (1+ Tx producteur). Pour un taux de 20 % sur le prix producteur équivaut à un taux d’environ 16,7 % sur le prix consommateur. 6 Ainsi, l’évolution des préférences des consommateurs en fonction des revenus entraîne bien une redistribution verticale de la taxation des biens. Le taux moyen de taxation de la consommation des ménages les plus pauvres étant de 2,0 %, lorsqu’il est de 3.8 % pour le quintile le plus riche. 2. LE MODELE 2.1. Les hypothèses et les agents du modèle Notre modèle s’inscrit dans la tradition des modèles de fiscalité indirecte optimale dans un cadre multi-agents (Diamond et Mirrlees 1971). Nous nous inspirons de Kaplow (1990) et de Cremer et Ghavary (1993) pour intégrer une dualité dans notre modèle entre d’une part des biens formels parfaitement taxés et d’autre part des biens informels imparfaitement taxés. Seule l’introduction de cette dualité entre les biens de consommation apporte une modification aux hypothèses standards de la littérature de la fiscalité optimale. 2.1.1. Le marché des biens Nous supposons que le marché des biens est caractérisé par « n » paires de biens « i ». Les paires de biens traduisent la dualité entre les secteurs formels et informels au sein de l’économie. Autrement dit, pour chaque bien « i », il existe un bien (xi) produit par le secteur formel et un autre bien (yi) produit par le secteur informel. Au total, il existe 2.n biens dans l’économie. Nous supposons que les biens formels sont taxés au taux ti. Les biens informels le sont au taux de : β iti. La variable β i représente le taux de fiscalisation (ou de couverture) du secteur informel dans la branche i. Autrement dit, (1-β i ) est le taux de fraude fiscale du secteur informel. Ainsi, si β i = 0, le secteur informel dans la branche i échappe intégralement à la taxation de sa production. L’Etat n’a pas la capacité de taxer le chiffre d’affaires des micro-unités de production. En revanche, si βi tend vers 1, l’Etat s’est donné les moyens de limiter la fraude fiscale des micro-entreprises. La production du « secteur informel » est taxée à un taux proche du taux normal prévu par le code des impôts. Ainsi, sur le marché des biens, les prix au consommateur peuvent s’écrire : Pour les biens xi : Pour les biens yi : qxi = pxi (1 + ti) qyi = pyi (1 + β i ti) Où pxi et pyi sont les prix au producteur. Sous l’hypothèse de concurrence pure et parfaite, les prix au producteur sont égaux au coût marginal et les consommateurs sont « price-taker ». Par simplification, on suppose donc que les prix au producteur sont exogènes. Autrement dit, nous ne tenons pas compte de l’effet produit par la taxation des intrants sur le coût des entreprises (rémanence de la TVA sur le secteur informel et les entreprises exonérées). Pour simplifier notre exposé, nous posons par la suite que les prix au producteur sont normés à 1. Pour la clarté de l’exposé, les notations utilisées sont les suivantes : X est le vecteur colonne des xi, et Y le vecteur colonne des yi. B est la matrice (n , 2) définie par B : (X , Y). Ainsi, on notera bi la matrice ligne définie par la paire de biens (xi , yi). Par ailleurs, pour l’ensemble des n branches de l’économie, on notera β le vecteur colonne formé par l’ensemble des β i. 7 2.1.2. Les ménages L’utilité directe des ménages est une fonction de la consommation de ces paires de biens. Pour simplifier, on suppose que le travail (loisir) et les biens de consommation sont séparables dans la fonction d’utilité du ménage. De même, nous admettrons que les ménages ne font que consommer, i.e. ils n’ont pas d’activités productives ou commerçantes. Ces deux dernières hypothèses permettent de supposer une contrainte budgétaire linéaire par rapport au taux de taxation, facilitant ensuite le traitement analytique du modèle. En outre, on suppose que les biens formels et informels sont imparfaitement substituables dans la demande des ménages. Cette hypothèse se justifie par la différence observée de qualité et de mode de commercialisation des deux biens. On pose Uh l’utilité (Marshalienne) du consommateur h, avec h : (1,…,H). (1) Uh = Uh (X h , Y h , T h ) Avec, pour le ménage h, X h le vecteur des consommations en biens xi , Yh vecteur des yi , et T h le temps de travail. L’utilité indirecte (Hicksienne) s’écrit : Vh = Vh (qx , qy , Ih , w h ) Avec Ih le revenu et w h le taux de salaire du ménage h, qui sont tous deux supposés exogènes. 2.1.3. L’Etat Si β est le taux de fiscalisation des unités de production informelles, son niveau dépend des moyens mis en œuvre par les autorités pour fiscaliser les micro-entreprises informelles 6 . De fait, le coût de la collecte des impôts par l’administration fiscale est une fonction de la variable β. On nomme g(β) la fonction de coût liée au taux de contrôle. Cette fonction est positive et croissante par rapport à β. On pourra aussi faire l’hypothèse qu’elle est convexe. Comme les recettes fiscales sont proportionnelles à β i, la convexité de la fonction de coûts permet ainsi de poser une productivité marginale décroissante des contrôles effectués par l’administration fiscale (i.e. l’efficacité des contrôles est décroissante). On a donc pour résumer : g(β)>0, g’(β)>0 et g’’(β)>0. En théorie, l’objectif de l’Etat est donc de définir les taux de taxation ti et les taux de couverture des unités informelles β i de sorte à maximiser la fonction d’utilité sociale tout en atteignant son objectif de recettes fiscales. Le programme de l’Etat s’écrit alors : La fonction d’utilité sociale s’écrit : (3) W = W(V1 , ……, VH) W est la fonction d’utilité sociale. En supposant que celle-ci est de type « Bergson-Samuelson », les utilités indirectes de chaque ménage (Vh ) sont indépendantes les unes des autres (forme additive). 6 On rappelle que cette fiscalisation n’intervient que par le biais du contrôle des unités de production informelle. Nous ne prenons pas en compte pour l’instant dans le modèle les phénomènes de rémanence qui permettent de taxer indirectement par les intrants la production informelle. 8 Tandis que sa contrainte budgétaire est donnée par : (4) n H i =1 h =1 R0 = ∑ ∑ (t i xih + β i ti yih ) − g ( β ) où R0 est l’objectif de recettes fiscales fixé par l’Etat. Le programme s’écrit : (5)  max W = W (V 1 ,........, V H )  n H  h h  s.c.Ro = ∑ ∑ ( ti x i + β i ti y i ) − g ( β )  i=1 h=1 Programme dont le Lagrangien s’écrit : (6) n H L = W + λ  ∑ ∑ (t i x ih + β i t i y ih ) − g ( β ) − Ro   i =1 h =1  On rappelle que les conditions du premier ordre se calculent par rapport à ti et β i., i.e. les deux variables endogènes du système. Les expressions des solutions optimales de ti et β i uniquement en fonction des variables exogènes prennent une forme trop compliquée pour pouvoir être commentées de manière analytique. A l’instar des articles écrits antérieurement sur la fiscalité optimale avec fraude (cf. Kaplow 1990, Cremer et Ghavary 1993, Ray 1998), nous avons choisi de commenter les solutions en fonction de l’autre variable endogène. L’optimum étudié est donc local, c’est à dire exprimé en fonction de l’optimum de l’autre variable endogène 7 . Les applications numériques permettront par la suite de résoudre ce problème et de calculer des optima « absolus ». 2.2. La solution optimale (locale) de ti La condition du premier ordre par rapport à ti s’écrit : (7)  ∂W  ∂V h ∂ V h     ∂ xih ∂L ∂ yih  2  = −λ (x h + β y h ) +  ti = 0 ⇔ ∑ + β + β t    ∑ ∑ ∑ i k k k i i h  ∂q ∂ti ∂q y i ∂qbk  h  ∂V i h  h  ∂q xi   bk    On notera que ∂qxi = ∂ti et que ∂qyi = β i ∂ti , car les prix au producteur sont supposés exogènes dans le modèle. Grâce à l’identité de Roy, nous savons que les propriétés de la fonction d’utilité indirecte permettent d’écrire : (8) ∂V h ∂V h = − kh x kh = − µ h xkh ∂qx i ∂I (9) ∂V h ∂V h = − kh y hk = − µh y kh ∂q y i ∂I Où µh est l’utilité marginale du revenu du ménage h. Pour simplifier les écritures, nous posons : (10) Φh = ∂W h µ ∂V h 7 Cette méthode, quoique discutable, est celle appliquée dans la littérature : Kaplow (1990) , Cremer et Ghavary (1993) procèdent de cette manière. Ray (1998) calcule la solution optimale qu’uniquement par rapport à t i, considérant β i fixé. 9 Φ h représente l’effet d’une variation marginale de l’utilité revenu d’un ménage sur l’utilité sociale. Selon l’expression de Diamond et Mirrlees (1971), Φ h est « l’utilité sociale marginale du revenu du ménage h ». De plus, l’équation de Slutsky permet d’écrire : (11) ∂xih ∂x h = S bhi ,bk − xkh ih ∂q ∂I bk (12) ∂yih ∂y h = S bh ,bk − y kh ih i ∂qbk ∂I Où S bhi ,bk est la matrice de Slutsky pour chaque ménage, c’est à dire la matrice des élasticités prix directes et croisées pour chaque bien de la demande compensée du ménage h. Etant donné que notre intérêt se porte essentiellement sur la relation entre les biens formels et informels, on peut restreindre notre cadre d’analyse en supposant que les biens ne sont substituables qu’à l’intérieur d’une même paire de biens (xi,yi). La matrice de Slutsky se limite donc aux élasticités prix directs de chaque bien et aux élasticités prix croisés entre les biens informels et les biens formels d’une même catégorie de bien i. Autrement dit, les élasticités prix croisés entre (xi, xk ) ou (xi, yk ) ou (yi, yk ) sont supposées être nulles. Cette hypothèse ne paraît pas être une restriction forte par rapport à la réalité. En effet, les études de Ravelosoa et al. (1999) à Madagascar, ou de Deaton et Grimard (1993) au Pakistan, montrent toutes deux, à partir de systèmes AIDS appliqués à la demande de biens alimentaires, que les élasticités prix croisés sont généralement très faibles, sinon statistiquement non différentes de zéro. Ces résultats renforcent d’autant plus notre hypothèse qu’ils sont obtenus sur des biens alimentaires. Pour lesquels la probabilité d’être substituables est supérieure aux classes agrégées de biens que nous utilisons 8 . La matrice de Slutsky s’écrit : S bhi ,bi . En substituant les expressions (10), (11) et (12) dans l’équation (7), puis en simplifiant, nous obtenons l’expression suivante de la condition du premier ordre de ti : (13)    h ∂ xih ∑ Φ ( xk + β k y k ) = λ  ∑ ( x k + β k y k ) + ∑ ∑  t i Sbi ,bi (1 + β i ) − t i  x k h h h h  h h h i h  h 2  ∂I + β i2 ykh h ∂y ih     ∂ I h   On pose à présent : x k = ∑ x kh et y k = ∑ y kh , en substituant dans (13) et en simplifiant, nous h h obtenons : (14)  Φh  h   h ∂ x ih ∂y ih  2 h h    t x + β y   ( x + β y ) ∑∑ i  k ∂I h i k ∂I h  i i ∑ t i S (1 + β )  ∑h  λ  i  i h   h = − 1 + xi + β i yi xi + β k y i xk + β k yk       h bi , bi 2 i 8 . On peut encore citer Deaton et Grimard (1993, p.18) : « None of these cross-price effects, except possibly that between wheat and rice, are large enough, or precise enough, to change by itself, the way in wich we might think about tax ». 10 L’équation (14) fournit une expression de la solution optimale locale du taux de taxation proche de la formulation de Ramsey (1927) avec plusieurs agents. Les relations entre la solution des ti et les autres variables sont les suivantes : 2.2.1. Relation entre ti et la dérivée de la demande compensée aux prix ( S bhi ,bi ) Le terme de gauche de l’équation (14) représente un critère d’efficacité de la fiscalité. Il est de signe négatif. On observe que ce terme implique que les dérivées de la demande par rapport aux prix compensés (matrice de Slutsky) d’un bien i jouent négativement sur le taux de taxation optimal de ce bien. En cas de fortes élasticités prix, ou lorsque le bien est principalement consommé par les populations riches, les dérivées prix sont élevées. Dans ce cas, la réduction du taux de taxation permet de minimiser le poids mort produit par la taxation des biens. Autrement dit, on retrouve le résultat standard de la théorie de la fiscalité optimale qui est que le taux de taxation optimal d’un bien est une fonction décroissante de la dérivée de sa demande compensée par rapport aux prix. Ce critère est donc régressif du point de vue de la redistribution verticale. 2.2.2. Relation entre ti et la dérivée de la demande au revenu.  ∂x ih  ∂I Le deuxième terme de droite de l’équation (14) ∑ ∑ t i  xkh i h + β i ² y hk h ∂yih  représente l’évolution ∂I h  des recettes fiscales en fonction de la sensibilité de la consommation à la variation des revenus. Ce terme est de signe négatif lorsque les revenus diminuent et sous l’hypothèse que les biens sont « normaux ». Autrement dit, la taxation provoque une diminution des revenus Ih qui entraîne une réduction de la consommation des biens. Si la dérivée de la consommation du bien i par rapport au revenu est forte, alors la consommation diminuera de manière substantielle lorsque les impôts s’élèveront. L’efficacité de la collecte des impôts sera amenuisée par la réduction de l’assiette fiscale. Pour cette raison, le taux de taxation du bien i doit être alors relativement faible pour les biens dont la dérivée de la demande par rapport au revenu est forte, sous l’hypothèse que ces biens sont normaux. La dérivée de la demande par rapport au revenu étant a priori plus forte pour les individus « pauvres », cette dernière relation devrait favoriser la progressivité du système de taxation. 2.2.3. Relation entre ti et le poids social des ménages Le premier terme de droite de l’équation (14) représente le critère d’équité dans la solution optimale de ti. C’est une somme de la consommation taxée des ménages pondérée par le poids social de chaque ménage, moins 1. Il est généralement admis dans la littérature que ce terme est de signe négatif. Par ailleurs, on peut exprimer ce terme d’une manière plus explicite en suivant la méthode proposée par Feldstein (1972) : (15)  Φ h xkh + β k y kh  1  Φ h   Φ h  xkh + β k y kh    = cov +   ,      λ x +β y  H ∑ h  λ  xk + β k y k  h  λ  k k k   ∑  11 1  Φh  ∑   représente l’utilité marginale sociale moyenne de la population9 . De fait, H h  λ  L’expression lorsque la consommation d’un bien i est corrélée positivement avec l’utilité marginale sociale d’un  Φh  h ( xk + β k y hk )  h  λ  − 1 diminue, ce qui entraîne un taux de taxation xk + β k yk ∑  individu, la valeur absolue de relativement faible du bien i. Ainsi, le taux de taxation doit être faible pour les biens qui sont consommés principalement par les individus dont le poids social est important et les individus dont l’utilité marginale des revenus est forte (i.e. les individus à faible revenu). De fait, plus la fonction d’utilité sociale W est concave (i.e. le poids social des catégories pauvres de ménages est élevé), plus le taux de taxation est faible. 2.2.4. Relation entre ti et le taux de couverture du secteur informel β i . Un taux de couverture du secteur informel élevé (pour une solution optimale donnée de βi), permet de réduire le taux de taxation optimal. En effet, on observe dans l’équation (14) que β i est facteur  Φ h  ∂y ih h , h , qui ont toutes une influence négative sur le taux de taxation des variables : S bi ,bi ,  λ   ∂I optimal (t i). Ce résultat semble logique, puisqu’il indique que le taux de taxation normal appliqué à l’économie formelle pourra être réduit par une baisse de l’évasion fiscale du secteur informel, à recettes fiscales inchangées. 2.3. La solution optimale (locale) de β i La condition du premier ordre par rapport à βi du programme de maximisation sous contrainte de l’Etat s’écrit : (16)  h ∂q ∂y h    ∂q  ∂W ∂V h ∂q y i  ∂L yi i   = −λ ∑ ti y ih + ∑ ∑  t i yi ∂xi + β i ti = 0 ⇔ ∑ − g β' i  h ∂β i ∂q y i ∂β i  h ∂β i ∂q yk ∂β i ∂q y k   h  ∂V i h       En utilisant l’identité de Roy et en réutilisant les variables définies précédemment, on obtient : (17)   ∂xih ∑ Φ yi t i = λ  ∑ ti yi + ∑ ∑ .ti  h h  h h h 2 i h  ∂ q yk + βi  ∂ yih  − g 'β i  ∂q y k   Par hypothèse, les biens ne sont substituables qu’entre bien formel et informel (à l’intérieur d’une même paire de biens i). En réarrangeant l’équation (17), nous obtenons l’expression suivante de la solution de β i : (18) ∂y ih  Φh  ∂x h = ∑  − 1  y ih + g β' i − t i ∑ i h ∂q h  λ h ∂q y  i y β iti ∑ i On ne tient pas compte de l’interprétation qui pourrait être faite de λ. On notera toutefois qu’en cas de variation minime de R (les recettes fiscales), alors d’après la formulation du lagrangien, on peut écrire : λ = - ∂W/∂t i . ∂t i/∂R. D’après Ahmad et Stern (1991), λ peut être alors interprété comme l’utilité marginale sociale du revenu du gouvernement, c’est à dire « le prix implicite des dépenses publiques » (traduction de shadow price). 9 12 2.3.1. Commentaires, cas d’une taxation optimale du secteur informel Si les biens informels sont des biens typiques, alors le terme de gauche de l’équation (18) sera de signe négatif. Supposons tout d’abord que le terme de droite soit aussi de signe négatif, de sorte que le taux de couverture du secteur informel soit positif (dans le cas contraire, la solution optimale pour l’Etat serait alors de subventionner le secteur informel). Sous ces hypothèses, la solution optimale de β i dépend de ses arguments de la manière suivante : ∂yi ∂ h qy Le terme de gauche ∑ h représente un critère d’efficacité du système de taxation. Il indique par i exemple qu’une forte élasticité prix de la demande de bien informel doit jouer à la baisse sur le taux de couverture du secteur informel. Les termes de droite impliquent : h β i est une fonction décroissante de l’utilité sociale marginale du ménage ( ∑ Φ ). Autrement dit, le h λ critère d’équité veut que le taux de taxation des biens informels soit réduit si ces biens sont d’avantages consommés par les ménages pauvres, et inversement. Pour une raison d’efficacité du système, une forte convexité ( g 'βi croissant) du coût de la taxation du secteur informel réduit le taux de couverture optimal du secteur informel. En effet une forte convexité du coût de fiscalisation du secteur informel (poids mort de la fraude) se traduit par un accroissement des coûts pour l’état, supérieur à celui de l’assiette fiscale. Par ailleurs, on notera que, sous cette hypothèse de convexité de la fonction de coût de fiscalisation du secteur informel, il est alors impossible que le taux de couverture optimal du secteur informel soit égal à 1 (si β→1 alors g’→∞). Ainsi, le taux de taxation du secteur informel est inférieur à celui du secteur formel, ce qui permet de réduire la charge supplémentaire (i.e. le poids mort) liée à la taxation de ce secteur 10 .  ∂x  En outre, une forte substituabilité entre les biens formels et informels  i > 0 influencera  ∂q yi  h positivement le taux de couverture optimal du secteur informel. Il s’agit là aussi d’un critère d’efficacité. Dans l’hypothèse d’une forte substituabilité entre les deux types de biens, les incitations à la fraude fiscale sont d’autant plus importantes. Dans ce cas, l’écart entre les taux de taxation des deux types de biens (égal à 1-β i) provoque une réduction de la base fiscale. La consommation des biens formels est substituée par celle des biens informels. 2.3.2. Cas où il est optimal de subventionner le secteur informel Sous l’hypothèse standard que le bien informel est un bien normal et typique, le secteur informel doit être taxé (le terme de droite de l’équation 18 est positif, donc β it i l’est aussi) sous deux conditions : - Le coût de taxation du secteur informel ne croît pas trop vite par rapport au nombre d’entreprises assujetties. 10 On retrouve ainsi pour les mêmes raisons les résultats des modèles cités plus haut sur la fiscalité optimale avec fraude (Kaplow 1990, Cremer et Gahvary 1993). 13 - Par ailleurs, la part des individus « non pauvres » consommant des biens informels doit être suffisamment importante (étant donné la concavité de la fonction d’utilité sociale). Dans le cas contraire (terme de droite de l’équation (18) négatif), il serait donc optimal pour l’état de subventionner le secteur informel (β it i négatif). Considérons à présent le cas où le bien informel s’avérerait être un bien «Giffen ». La qualité du bien informel est très faible, lors d’une augmentation des prix du bien informel, l’effet revenu l’emporte, accroissant ainsi sa consommation. ∂y ih h ∂q y Dans ce cas, ∑ est positif. A supposer que le terme de droite de l’équation (18) reste de signe i négatif, β i est alors de signe négatif. Le secteur informel reçoit alors une subvention. Les commentaires sont alors les suivants : - La relation entre β i (en valeur absolue) et les termes de droite de l’équation (18) restent les h mêmes par rapport aux variables ∑ Φ et g’β . h λ - En revanche, sous l’effet de la contrainte budgétaire, le signe de la dérivée entre la demande du bien formel et le prix du bien informel devient négatif. La hausse des dépenses en biens informels doit être compensée par une réduction de la consommation des biens formels, les biens deviennent ainsi complémentaires11 . Dans ce cas, la sensibilité de la demande des biens formels au prix informel diminue la subvention optimale accordée au secteur informel. En résumé, lorsque le bien informel est un bien typique, celui ci doit être taxé s’il est consommé de manière non marginale par les classes non défavorisées de la population, et si la convexité du coût lié à la fiscalisation du secteur informel n’est pas «trop » forte. Dans le cas contraire, le modèle prévoit de subventionner le secteur informel (empiriquement : à ne pas le taxer). 3. MODELE EMPIRIQUE ET SIMULATIONS 3.1. Le modèle empirique Le modèle empirique doit permettre de fournir des estimations numériques des solutions optimales des taux de taxation et des taux de couverture du secteur informel. Pour cela, nous devons dans un premier temps choisir des formes fonctionnelles des fonctions d’utilité sociale et de demande. 11 ∂xih <0 ∂q yi 14 3.1.1. Poids social des ménages De manière standard, nous avons choisi de représenter la fonction d’utilité sociale par la fonction d’Atkinson (1970). Celle-ci s’écrit : (19)  (V h ) (1−ε ) K ∑  (1 − ε ) W (V h ) =  h  ∑ K.ln(V h )  h pour ε ≠ 1 pour ε = 1 L’exposant ε représente l’élasticité de l’utilité sociale par rapport à l’utilité individuelle. C’est le « degré d’aversion aux inégalités de la société ». K est un facteur positif quelconque. Etant donné les caractéristiques de la fonction d’utilité sociale d’Atkinson (1970), on peut alors écrire Φ h : (20) [ ] ∂∂VI −ε Φh = K V h h h Plus ε est élevé, et moins l’Etat accordera d’importance au bien-être des ménages les plus aisés, et inversement. Autrement dit, plus ε est grand, plus la société éprouve de l’aversion pour les inégalités. On rappelle que les caractéristiques de la fonction d’utilité collective d’Atkinson (1970) sont les suivantes : Pour ε = 0 , la société est « utilitariste » (Bentham) : chaque individu est considéré de la même manière, quel que soit son revenu. Pour ε → ∞ , la société est « égalitariste » (Rawls) : l’utilité collective est égale à l’utilité de l’individu le plus défavorisé de la société. Nous choisissons une forme PIGLOG pour représenter la fonction d’utilité indirecte, qui permet ensuite d’être cohérent avec le choix d’une fonction AIDS pour modéliser la demande des consommateurs (cf. infra) : (21) ln V h ( I h , p) = ln( I h / n h ) − ln a h ( p) b h ( p) Où ah (p) et bh (p) représentent des indices de prix du consommateur reflétant ses préférences (Cf. infra). La variable nh est le nombre d’équivalent adulte dans le ménage h. L’échelle d’équivalence retenue est celle de l’OCDE (poids de 1 pour le chef de ménage, 0,7 pour les autres adultes et 0,5 pour les enfants - on définit un adulte comme étant une personne de moins de 16 ans). Afin de simplifier Φ h , on transforme cette expression de sorte à la normaliser à 1 pour le consommateur le plus démuni. On supprime de plus le rapport des utilités marginales du revenu. L’approximation du poids social des individus est alors : (22)  v1  Φ = h v  h ε où v1 est l’utilité de l’individu le plus pauvre. 3.1.2. Formes fonctionnelles des solutions de ti, β i et λ Les revenus compensés ne peuvent pas être observés dans la réalité. Cela implique que les élasticités prix sont estimées à partir des variations totales de la demande, comprenant l’effet substitution et l’effet revenu. Autrement dit, le programme de fiscalité optimale ne peut pas intégrer l’équation de Slutsky de manière empirique. Par ailleurs, ne disposant pas d’informations sur le surcoût lié à la fiscalisation du secteur informel, nous préférons ne pas en tenir compte dans 15 l’estimation empirique du système fiscal optimal. Cette simplification aura bien sûr pour effet d’augmenter le taux de taxation optimal des biens informels. De plus, nous maintenons l’hypothèse d’élasticités prix croisés non nulles uniquement entre les biens formels et informels appartenant à la famille de bien i. La solution optimale de t i est alors représentée par la fonction suivante : (23)   ∂x h ∑ Φ h ( xkh + β k y kh ) = λ ∑ ( x hk + β k ykh ) + t i ∑   h h h i  ∂q x i +  ∂y h ∂x ih ∂y h   + β i2  i + i    ∂q y ∂q x   ∂q y i i i    Celle de β i est donnée par l’équation (17) : (24)   ∂x h   ∂ y h  ∑h Φ h yih = λ ∑h yih + t i ∑h . ∂q i + β i ∂q i  yi yi  La contrainte de l’Etat reste : (25) n H i =1 h =1 R0 = ∑ ∑ (t i xih + β i ti yih ) Les équations (23) à (25) permettent alors de définir un système juste identifié de 2n+1 équations, permettant d’obtenir les solutions des n.t i,, n.β i . La variable Φ h est calculée grâce à l’équation (22). 3.2. Les fonctions de demande de biens Les fonctions de demande vont permettre de prendre en compte les variations de la demande de biens induites par la variation des taxes. La forme retenue est une fonction AIDS (Deaton et Muellbauer 1980). Son choix présente un certain nombre d’avantages par rapport à d’autres formes de fonction de demande couramment utilisées telles que la LES (Linear Expenditure System) ou la CES (Constant Elasticity Substitution) 12 : - L’AIDS permet d’obtenir des courbes d’Engel qui ne sont pas linéaires. Le fait que les préférences ne soient pas homothétiques est important car, dans le cas contraire, la théorie montre (Atkinson 1977) qu’un taux de TVA uniforme est optimal. De plus, on peut remarquer qu’il existe alors une incohérence lorsque l’on estime les élasticités à partir d’une fonction LES ou CES et que l’on calibre le modèle à partir de données d’enquêtes où la part budgétaire des biens pour chaque ménage n’est pas constante. - En outre, les biens sont forcément complémentaires pour une LES. A l’inverse, la fonction de demande AIDS permet de prendre en compte des biens substituables, et donc d’étudier les effets de la substituabilité entre les biens informels et formels. La demande pour un bien i par le ménage h s’écrit : (26) wbih = α bih + ∑ d bhi ,bi ln qbk + c bih ln k (I h / n h ) Q 12 D’un point de vue pratique, des fonctions AIDS ont déjà été estimées dans de nombreux pays en développement, dont Madagascar (Ravelosoa et al. 1999). 16 Où αbi, d bi,bi et cbi sont les paramètres à estimer. wbih est la part budgétaire du bien bi pour le h h h ménage h, qbk la matrice des prix au consommateur, Ih la dépense totale du ménage 13 . Enfin, Q est l’indice des prix à la consommation, estimé par l’indice de Stone : (27) ln Q = ∑ wbk ln qbk k q bh I h avec wbk = ∑ bk h bk Les paramètres de la fonction d’utilité indirecte du consommateur sont alors : (28) h ln a h ( p) = ∑ wbk ln qbk k et h ln b h ( p ) = ∏ p k b i c k 3.2.1. Résolution du système Le programme de l’Etat est représenté par un système de 2n+1 équations. En parallèle, le système de demande des consommateurs est composé de 2n équations (demande de biens formels et informels). Au total, le modèle est donc constitué de 4n+1 équations. A notre connaissance, c’est la première fois qu’un modèle empirique de fiscalité optimale est résolu en tenant compte explicitement des variations de la demande entraînées par celles des taux de taxation (voir Ray 1997 pour une revue). Or, les variations des taux de taxation, et donc des prix, entraînées par le calcul des taxes optimales, ne sont pas toujours « marginales ». Les résultats des modèles risquent alors d’être fortement biaisés sous l’hypothèse de demandes constantes. De manière pratique, le système d’équations (21) à (28), plus les équations de prix au consommateur, sont retranscrits sous le logiciel GAMS. Ceci permet ensuite de déterminer de manière simultanée : les taux de taxation optimaux, le taux de couverture optimale du secteur informel, la demande de biens des ménages et les prix au consommateur. Les élasticités et les dérivées de la demande par rapport aux prix et revenu, ainsi que les paramètres de la fonction AIDS, sont en revanche supposés exogènes, afin de faciliter la résolution du modèle. Nos simulations consistent ensuite à faire varier les indices d’aversion aux inégalités pour montrer quel serait le système de taxation des biens formels et informels qui serait dans ce cas le plus approprié. 3.2.2. Paramétrage du modèle. Etant donné la taille de l’enquête ménage (400 ménages) et de l’unité géographique de l’enquête MADIO (1995b), il est peu aisé d’estimer des systèmes de demande de type AIDS permettant de calculer les élasticités prix requises pour le paramétrage du modèle. La méthode proposée par Deaton (1997) pour estimer les élasticités prix à partir de la variance spatiale des prix est ici difficilement applicable du fait de l’unité spatiale de l’enquête. Ainsi, dans cette enquête, la variance des prix d’un bien proviendra probablement en grande partie des différences de qualités entre les produits. Le risque est alors d’estimer des comportements de choix de qualité, au lieu de comportements de consommation en fonction des prix (Deaton 1997). Contrôler le choix de la qualité accroît encore la complexité la méthode. Pour ces raisons, nous avons choisi d’utiliser des paramètres déjà calculés dans des études antérieures : Pour les produits alimentaires, Ravelosoa 13 Calculer la fonction de demande par rapport au revenu équivalent (Ih/nh) permet ensuite de calculer une agrégation exacte des demandes individuelles (Cf. Deaton et Muellbauer 1980). Les simulations retenues ont été en effet effectuées sur des classes agrégées de ménages (cf. infra). 17 et al. (1999) ont déjà estimé des fonctions de demande de type AIDS à Madagascar. Il existe en revanche peu de résultats pour les produits autres qu’alimentaires dans les pays en développement. Seul Goaïed (1991) nous fournit une estimation des élasticités prix et revenu du logement en Tunisie. Pour les autres biens, les élasticités sont tirées du travail de Nichèle et Robin (1993) pour la France. Ces élasticités sont présentées dans le tableau 3-1. Toutefois, ces études ont estimé les élasticités prix et revenus des produits de manière indifférenciée entre les biens formels et informels. Nous sommes alors obligés d’émettre des hypothèses sur la valeur des paramètres entre les biens formels et informels. Nous nous appuyons pour cela sur les résultats de l’enquête MADIO (1995b), et les travaux de Roubaud (1994) et Cogneau et al. (1996) au Cameroun14 : - Le tableau 1-2 (supra) montre que la part budgétaire des biens et services informels décroît avec le revenu des ménages. Les biens informels seraient donc des biens nécessaires (élasticité revenu inférieure à 1), tandis que les biens formels seraient des biens de luxe (voir aussi Roubaud 1994 ou Cogneau et al. 1996 pour le Cameroun). L’exemple de la consommation du poste « Riz et céréales » est assez parlant de ce point de vue. On observe en effet dans l’enquête que la totalité du riz ordinaire et paddy consommée provient du secteur informel, lorsque le riz de luxe est en revanche entièrement acheté dans le secteur formel. Ainsi, à l’image de Cogneau et al. (1996), on peut supposer que les élasticités revenu des biens informels sont inférieures à celles des biens formels, voire inférieures à 1. - L’enquête auprès des ménages MADIO (1995b, p.20-21) fait ressortir que le secteur informel est choisi pour la modicité des prix, tandis que les biens formels sont choisis pour leur meilleure qualité. L’élasticité prix des biens informels doit être alors inférieure en valeur absolue à celle des biens formels (Cogneau et al. 1996). N’étant pas en mesure d’estimer les élasticités de la consommation, nous écartons le cas où les biens informels seraient des biens « Giffen ». Les élasticités du tableau 3-1 représentent des valeurs moyennes sur l’ensemble des biens. Pour les raisons exposées ci-dessus, les élasticités des biens formels et informels ont été fixées arbitrairement de la manière suivante : les élasticités revenu (prix directs) du tableau 3-1 sont augmentées (diminuées) de 10 % pour les biens formels et diminuées (augmentées) de 5 % pour les biens informels. Tableau n° 3-1 : Elasticités prix et revenus et distribution de la demande Biens et services Riz, céréales (1) Viandes, légumes, (1) Produits transformés (1) Habillement (2) Logement (3) Equipement maison, (3) Santé (2) Transport (4) Education (2) Autres biens et services (4) Total Elasticité revenu 0.20 1.30 1.40 1.17 1.17 1.17 1.09 1.30 1.17 1.20 Elasticité prix directe -0.70 -0.85 -0.87 -0.91 -1.47 -1.47 -1.00 -1.15 -0.58 -1.10 Elasticité prix Distribution des Distribution des croisée biens informels biens formels 0.52 25.3 % 0.9 % 0.70 30.6 % 1.4 % 0.70 8.0 % 19.8 % 0.78 3.3 % 13.2 % 0.86 13.6 % 12.4 % 0.90 4.0 % 7.6 % 0.62 1.0 % 10.8 % 0.90 10.2 % 24.6 % 0.50 2.4 % 5.9 % 0.88 1.8 % 3.4 % 100.0 % 100.0 % Sources : (1) Ravelosoa et al. (1999), (2) Nichèle et Robin (1993), (3)Goaïed (1991), (4) valeur arbitraire. 14 Voir aussi Boyabé (1999) pour une revue. 18 Enfin, il s’agit de choisir les élasticités prix croisées entre les biens informels et formels. Celles-ci sont en revanche clairement positives, les deux biens sont en effet substituables par hypothèse. Ils le sont toutefois imparfaitement du fait de la différence de qualité des produits. On peut supposer que cette imparfaite substituabilité entre les biens se reflète dans la part du bien informel dans la consommation totale d’un bien i15 . Une fois choisies les élasticités prix et revenus, on en déduit facilement la valeur des paramètres d bih ,bi , c bih et d bih ,bk des fonctions de demande, à partir des formules des élasticités des fonctions AIDS 16 . La constante de la fonction de demande est ensuite déterminée comme la différence entre les termes de gauche et de droite des équations de demande. Par ailleurs, l’homogénéité de degré zéro et la « contrainte d’additivité » au sein d’une fonction de demande AIDS impliquent que : ∑ d bih ,bk = 0, ∑ c bih = 0, ∑ a hbi = 1 . Les paramètres n’étant pas estimés, k i i il était alors difficile de les fixer de manière à satisfaire ces contraintes, sans obtenir ensuite des valeurs aberrantes des dérivées de la demande. La contrainte budgétaire est néanmoins respectée en calculant de manière résiduelle la demande des « autres biens et services » informels. Le modèle respecte aussi la symétrie de Slutsky : d xhi , yi = d yhi, xi . 3.3. Résultats des simulations Le paramétrage du modèle nous a amené à émettre des hypothèses sur la typologie des biens informels. Celles-ci ne sont pas sans conséquences sur les résultats du modèle empirique. Il est alors utile de préciser ces liens : - Les élasticités revenu relativement faibles d’une part, et la forte substituabilité entre les biens formels et informels d’autre part, doit accroître (selon les critères d’efficacité) le taux de taxation optimal des biens informels. Toutefois, les dérivées de la demande par rapport aux revenus devraient réduire la taxation optimale des biens informels, ceux-ci étant plus consommés par les ménages à bas revenus. - En revanche, les préférences des catégories défavorisées pour les biens informels tendront à diminuer la pression fiscale sur ces biens (critère d’équité). - Les dérivées de la demande par rapport aux prix (directes) ont des effets contradictoires sur le niveau de taxation des biens informels. D’une part, les élasticités prix des biens informels étant supposées plus importantes que celles des biens formels (pour un même bien i), le taux de taxation du bien informel doit être plus faible (efficacité). En revanche, les biens formels étant principalement consommés par les hauts revenus, les dérivées prix seront probablement plus importantes pour ces biens. Il est en fait plus probable que les dérivées prix (directes) jouent en faveur d’une réduction de la taxation des biens formels. 15 Plus la part des deux biens dans la consommation totale est proche de 50 %, plus les biens sont a priori substituables. On choisit d’estimer les élasticités prix croisées à partir de la formule suivante : 1 − 0 .5 − 16 eR = 1 + cbih wbih (élasticité revenu), eii = −1 + d bih ,bi wbih − cbih q y i yi qy i yi + q xi xi . (élasticité prix directe), et eik = croisée). 19 d bih ,bk wbih − c bih wbih h (élasticité prix wbk Nos simulations permettent à présent d’estimer les taux de taxation optimaux selon le degré de redistribution verticale recherché par l’Etat au travers de son système d’imposition indirecte. Les simulations consistent en effet à calculer les effets des variations de la variable ε sur les variables endogènes du modèle (t, β, q). Dans un premier temps nous avions calculé les taux de taxation optimaux à partir des 400 ménages de notre échantillon. Or les résultats produits de cette manière sont difficilement interprétables du fait d’une forte hétérogénéité des préférences intra-classes. On peut en effet observer dans le tableau 2 (en annexe) que les coefficients de variation des taux de pression intra-déciles sont nettement supérieurs à la variation des taux de pression moyens inter-déciles. Ceci témoigne alors d’une forte variation des préférences entre les individus de même revenu, plus importante que pour les individus ayant des revenus différents. Cette variance intra-déciles produit des mouvements erratiques des taux de taxation optimaux par produits lorsque les indices d’aversion aux inégalités varient. Pour limiter ces effets et «linéariser » nos résultats, nous avons alors calculé les taux de taxation optimaux à partir de classes de dépenses agrégées par vingtiles. 3.3.1. Taxation des biens formels versus taxation des biens informels. La taxation relative des deux secteurs est fonction du degré de redistribution souhaité. L’accroissement du poids social des populations les plus défavorisées tend à accroître l’écart des taux de taxation entre les secteurs formels et informels. A l’opposée, les critères d’efficacité ont pour conséquence d’accroître le taux de pression sur les biens informels. In fine, l’écart entre les taux de taxation optimaux des deux secteurs est plus faible que celui existant dans le système actuel (cf. tableau 3-2). Ainsi, les taux de taxation optimaux moyens des biens et services formels sont inférieurs aux taux actuels (cf. tableau 3-2). A l’opposée, les produits informels subissent en moyenne un taux de pression plus élevé que le taux actuel, pour des valeurs des indices d’aversion de 0 ou de 1. En outre, la redistribution des taux de taxation entre les biens accroît la part budgétaire des biens formels (cf. tableau 1 en annexe). Ainsi, pour un indice d’aversion aux inégalités de 2 par exemple, le taux de taxation moyen du secteur formel est inférieur au taux actuel, bien que le taux de pression moyen des biens informels soit identique. Ainsi, si le taux de taxation optimal moyen du secteur informel reste peu élevé (entre 1,1 % et 0,3 %), la hausse de la taxation de certains biens informels permet de réduire la pression moyenne du secteur formel. Cette évolution du système de taxation devrait alors occasionner un gain d’efficacité. En conclusion, il semblerait optimal d’accroître le taux de pression sur certains produits informels. Pour un indice d’aversion aux inégalités de 1, le taux de pression moyen des biens informels passerait à 1 %, contre un taux de 0,3 % actuellement. A recettes constantes, le remaniement du système de taxation permettrait de réduire le taux de pression moyen des biens formels. Toujours pour un indice d’aversion aux inégalités de 1, le taux de pression optimal sur les biens formels serait 6,6 %, contre 7,6 % actuellement. On note que les taux optimaux moyens sur les secteurs informel et formel sont proches de l’état initial. Ce résultat tient au fait que l’algorithme de résolution employé par GAMS (CONOPT 2 et MINOS 5) est « influencé » par les équilibres de départ dans le sens où la solution des fonctions de demande tend à converger vers l’équilibre initial. Les variations possibles des taux de taxation sont ainsi limitées. Les résultats sont plus saillants si le modèle est résolu sous l’hypothèse d’une demande exogène. Si les valeurs ordinales sont similaires, les écarts entre les taux de pression entre secteurs formel et informel sont en revanche plus amples. Ainsi le taux de pression optimal du secteur informel est de 1,8 % pour ε = 0, et tombe à –3,6 % (subvention sur les prix des biens 20 informels) pour ε = 3. Si les variations avec demande endogène sont moins marquées, elles ont néanmoins l’avantage de mieux prendre en compte les effets substitutions (voir aussi Banks et al. 1996). Effets qui peuvent aussi expliquer la faiblesse des évolutions des taux lorsque la demande est endogène. De plus, cette évolution limitée des taux pour chaque produit permet aux résultats de nos simulations d’être directement une source possible de réformes fiscales. Ainsi, les résultats du modèle avec demande endogène sont plus riches d’enseignements du point de vue empirique. 3.3.2. Résultats sur la taxation des biens et services Bien que les taux de taxation optimaux soit calculés à partir de dépenses agrégées, leurs mouvements restent parfois erratiques lorsque l’indice d’aversion aux inégalités varie. Toutefois, des résultats notables semblent ressortir des simulations (cf. tableau 3-2) : Les taux de taxation optimaux des produits alimentaires divergent peu de ceux du système actuel. Nos simulations font toutefois ressortir qu’il serait optimal d’assurer un taux de taxation entre 6 et 8 %17 des produits alimentaires transformés, qu’ils soient d’origine formelle ou informelle. De même, pour les produits de l’élevage, de la pêche, les fruits et les légumes, il pourrait être optimal de taxer ces biens à des taux compris entre 5 et 7 % pour des niveaux de progressivité élevés (ε =1 et ε = 2). En revanche, les céréales, qui sont des biens nécessaires, sont taxées au taux zéro. Une réforme fiscale consisterait donc à assurer l’application d’un taux zéro sur l’ensemble des céréales. Appliquer un taux de 5 % sur les produits alimentaires (hors produits de base) serait en revanche une réforme possible. On remarque aussi que la taxation des céréales, qui est actuellement due à une rémanence de la TVA pour les biens exonérés, est bien à l’origine d’effets régressifs. L’habillement pourrait être plus amplement taxé pour des critères d’efficacité. Il est intéressant de noter que pour ε = 1, le taux de taxation optimal sur l’habillement est proche du taux normal de la TVA. La réforme pour l’Etat consisterait alors à supprimer les exemptions accordées aux associations de sorte à appliquer de manière universelle le droit fiscal. Nous constatons ensuite que la taxation optimale des secteurs sociaux : éducation, santé n’est pas négligeable. Un taux de pression d’environ 3 % sur la consommation des services sociaux formels (soit un taux de l’ordre de 5 % sur les prix au producteur) pourrait représenter une solution optimale. Ce résultat, a priori étonnant, n’est en fait pas illogique : L’éducation et la santé étant deux biens de luxe, dans le sens où leur part budgétaire croît plus vite que les revenus, et avec des élasticités-prix faibles, il est en fait logique d’obtenir un tel résultat dans un modèle de fiscalité optimale 18 . On peut d’ailleurs s’interroger sur le fait qu’en exonérant les services sociaux de manière indifférenciée, on accorde un cadeau fiscal plus important aux couches les plus favorisées de la population. La solution pourrait être alors de taxer les services sociaux formels, puis de reverser le produit de ces taxes en subventionnant l’accès de ses services pour les populations les plus pauvres. On augmenterait ainsi la progressivité du système de taxes et de dépenses publiques. On notera par ailleurs, que les récentes propositions de réforme fiscale visant à taxer les produits pharmaceutiques à Madagascar (proposition de loi de finances 1999) vont dans le sens d’une fiscalité indirecte optimale. Sachant qu’il existe par ailleurs des centrales d’achat de médicaments où ceux-ci sont subventionnés. Du point de vue pratique, il pourrait être optimal d’introduire des taux différenciés dans la taxation des biens de consommation à Madagascar. Un taux de 5 % sur le prix au producteur des biens 17 soit un taux d’environ 10 % sur le prix au producteur. Prendre en compte les externalités et les effets à long terme créés par les secteurs sociaux limite bien sûr la pertinence d’un tel résultat. 18 21 alimentaires (hors céréales), les services sociaux et le logement serait une réforme envisageable dans ce but. Les exonérations doivent être remplacées par un taux zéro pour les céréales. Les effets de rémanence pourraient être ainsi éliminés. Pour les autres produits, on notera que les taux maximum sont généralement inférieurs au taux normal actuel (soit un taux de 16,7 % rapporté au prix au consommateur ou de 20 % par rapport au prix au producteur). A recettes constantes, un abaissement du taux « normal » sur les prix au producteur autour de 15 % serait envisageable, étant donné la taxation des biens alimentaires et des services sociaux au taux de 5 %. Cette réforme aurait de plus pour effet de réduire les incitations sur la fraude et donc de réduire les distorsions créées par la TVA sur le jeu de la concurrence. L’autre côté de la réforme consisterait à taxer le secteur informel de manière ciblée. Les simulations montrent que le taux de pression devrait être essentiellement accru sur les secteurs de l’« agro-industrie » et de l’habillement de manière à les rapprocher des taux des entreprises formelles. 22 Tableau n° 3-2 : Taux de taxation optimaux avec demande endogène Degré d'aversion aux inégalités Taux de taxation des biens formels Initial ε=0 ε=1 ε=2 ε=3 Riz, céréales Viandes, poisson, légumes, fruits… Alimentation transformée Habillement Logement Equipement, services maisons Santé Education Transport Autres biens et services 2.7 % 0.7 % 10.0 % 3.7 % 3.8 % 11.7 % 1.5 % 1.1 % 13.7 % 10.0 % 1.2 % 0.6 % 5.3 % 9.9 % 1.8 % 7.0 % 13.6 % 6.5 % 6.8 % 6.9 % 0.0 % 0.0 % 7.9 % 16.3 % 4.8 % 10.8 % 3.1 % 2.0 % -0.1 % 8.4 % 0.3 % 7.5 % 6.2 % 4.4 % 2.5 % 10.1 % 3.4 % 1.8 % 12.8 % 19.4 % 1.8 % 4.9 % 11.7 % 2.6 % 4.3 % 10.5 % 1.6 % 15.9 % 8.4 % 8.8 % Moyen 7.8 % 7.0 % 6.6 % 7.1 % 7.6 % Riz, céréales Viandes, poisson, légumes, fruits… Alimentation transformée Habillement Logement Equipement, services maisons Santé Education Transport Autres biens et services 0.0 % 0.0 % 1.1 % 1.2 % 0.5 % 1.5 % 0.0 % 0.0 % 0.0 % 0.3 % 0.1 % 0.1 % -0.7 % 20.8 % 0.3 % 5.0 % 40.7 % 9.8 % -9.9 % 5.6 % 0.0 % 0.0 % 8.3 % 1.3 % 0.8 % 2.8 % -1.2 % -0.3 % 0.1 % -1.1 % 0.0 % -0.4 % 6.0 % 0.5 % 0.3 % -2.5 % 0.2 % -0.4 % 0.0 % -0.1 % 0.2 % 0.4 % 2.3 % 0.8 % 0.7 % 1.8 % 0.0 % -0.7 % -1.8 % -0.1 % Moyen 0.3 % 1.1 % 1.0 % 0.3 % 0.3 % Riz, céréales Viandes, poisson, légumes, fruits… Alimentation transformée Habillement Logement Equipement, services maisons Santé Education Transport Autres biens et services 0.3 % 0.4 % 25.8 % 7.1 % 7.6 % 11.9 % 2.1 % 0.9 % 39.6 % 4.3 % 0.6 % 0.6 % 11.4 % 31.1 % 3.5 % 10.4 % 27.8 % 9.8 % -0.1 % 4.9 % 0.0 % 0.0 % 32.1 % 33.5 % 11.9 % 12.5 % 4.7 % 1.6 % -0.1 % 4.1 % 0.0 % -1.0 % 22.9 % 9.6 % 4.5 % 9.4 % 5.7 % 1.8 % 44.3 % 2.9 % 1.0 % 2.9 % 33.8 % 5.3 % 8.4 % 10.6 % 2.2 % 10.4 % 21.6 % 3.9 % Total 100.0 % 100.0 % 100.0 % 100.0 % 100.0 % Taux de taxation des biens informels Structure des recettes fiscales CONCLUSION Les préférences des consommateurs « pauvres » pour les biens informels et l’élasticité prix élevée de ces biens justifie leur faible taxation. A partir d’un modèle de fiscalité optimale, nous estimons en effet que les taux de taxation des biens informels sont nettement inférieurs à ceux des biens formels. Si nos résultats sont globalement proches de ceux obtenus par les modèles analytiques de fiscalité optimale avec fraude (Kaplow 1990, Cremer et Gahvary 1993), il existe toutefois des nuances importantes dans les raisons d’y parvenir. Dans notre modèle, la faiblesse du taux de taxation des biens informels ne provient pas d’un surcoût lié à la taxation du secteur frauduleux, mais comme nous l’avons vu plus haut de la typologie des biens. Ensuite le fait que la taxation du bien informel 23 soit inférieure à celle du bien formel n’est pas vérifié pour tous les secteurs de l’économie, contrairement à ce qui est conclu par les modèles analytiques. L’introduction dans le modèle empirique d’un surcoût lié à la taxation du secteur informel aurait encore accru l’écart des taux de taxation entre les secteurs formels et informels (réduction du poids mort lié à la taxation de la fraude). Le surcoût lié à la fiscalisation de l’informel ne peut être connu que si l’on dispose de statistiques sur les coûts de fonctionnement de l’administration. Information qui n’est généralement pas disponible dans les PED. En revanche, la modélisation en équilibre général est envisageable. Celle-ci permettrait de tenir compte de deux effets importants. Elle favoriserait d’une part un accroissement de la taxation du secteur informel. En effet, tout écart des taux de taxation entre les secteurs se traduirait par une réallocation des facteurs de production des secteurs formels vers l’informel. Ceci affecterait la contrainte budgétaire de l’Etat, encourageant ainsi la taxation du secteur informel. A l’opposée, la taxation des biens informels doit réduire le prix au producteur informel, et donc le revenu des ménages les moins favorisés. Il serait donc intéressant de développer ce modèle dans un EGC, sachant que nous avons déjà estimé la probabilité pour une entreprise d’être formelle ou informelle en fonction des taux de taxation. Par ailleurs, Fortin et al. (1997) ont déjà endogénéisé le secteur informel dans un modèle EGC. 24 REFERENCES Ahmad E. et Stern N. (1991), «The theory and practice of tax reform in developing countries », Cambridge University Press. Atkinson A.B. (1970), «On the measurement of inequality », Journal of Economic Theory, vol 2, pp. 244-63. Boyabé J.B. (1999), « Marché informel : une lecture critique du modèle d’Akerlof », Revue Tiers Monde, tome XL, n° 157, janvier-mars, pp. 169 – 186. Banks J., Blundell R. et Lewbel R. (1996), «Tax reform and welfare measurement : do we need demand system estimation ? », The Economic Journal, 106, pp. 1227-41. Banks J., Blundell R. et Lewbel R. (1997), « Quadratic Engel Curves and Consumer Demand », The Review of Economics and Statistics, vol. 79, novembre, pp. 527- 38. Cogneau D., Razafindrakoto M. et Roubaud F. (1996), « Le secteur informel urbain et l’ajustement structurel au Cameroun », Revue d’Economie du Développement, n°3, pp. 27-63. Cremer H., Gahvari F. (1993), « Tax evasion and optimal commodity taxation », Journal of Public Economics, vol 50, pp 261 – 275. Deaton A. (1997), The analysis of household surveys : a microeconometric approach to development policy, Ed. The John Hopkins University Press, The World Bank, 497 p. Deaton A. et Grimard F. (1993), « Demand analysis and tax reform in Pakistan », LSMS Working Paper n°85, World Bank, 45 p. Deaton A. et Muellbauer J. (1980), « An Almost Ideal Demand System », The American Economic Review, vol 70, n°3, pp. 312- 26. Fortin B., Marceau N., Savard L. (1997), « Taxation, wage controls and the informal sector », Journal of Public Economics, vol 66, pp 293 – 312. Gautier J-F. (1999), « Les paradoxes de la TVA dans un pays en développement : l’illustration de l’industrie malgache », Tiers Monde, vol 40, n° 157, pp. 187-211. Goaïed M. (1991), « Estimation d’un système de demande des ménages à partir de données groupées des enquêtes budgétaires en Tunisie », Economie et Prévision, n°97, pp. 11-19. Kaplow L. (1990), « Optimal Taxation with costly enforcement and evasion », Journal of Public Economics, vol 43, pp 221 – 236. MADIO (1995a), Le secteur informel dans l’agglomération d’Antananarivo. Performances, Insertion, Perspectives. Enquête 1-2-3, phase 2, 50 p. MADIO (1995b), La consommation et le rôle du secteur informel dans la satisfaction des ménages dans l’agglomération d’Antananarivo, Enquête 1-2-3, phase 3, 33 p. Nichèle V. et Robin J-M. (1993), « Evaluation des effets redistributifs des réformes de la fiscalité indirecte française », Economie et Prévision, n°110-111, pp. 105-28. 25 Ravelosoa J.R., Hagglabe S. et Rajemison H. (1999), « Estimation des élasticités de la demande à Madagascar à partir d’un modèle AIDS », Institut National de la Statistique de Madagascar, octobre, 52 p. Ray R. (1997), « Issues in the design and reform of commodity taxes : analytical results and empirical evidence », Journal of Economic Surveys, vol 11, n°4, pp. 353 – 88. Roubaud F. (1994), « Consommation et condition de vie des ménages à Yaoundé », DIAL –DSCN, n°1994-04E. 26 ANNEXE Annexe n° 1 : Variation de la demande et des prix Part budgétaire des biens formels Initial ε=0 ε=1 ε=2 ε=3 Riz, céréales Viandes, poisson, légumes, fruits… Alimentation transformée Habillement Logement Equipement, services maisons Santé Education Transport Autres biens et services 0.4 % 0.5 % 7.7 % 5.1 % 4.9 % 3.0 % 4.2 % 9.6 % 2.3 % 1.3 % 0.3 % 0.4 % 7.4 % 5.4 % 4.8 % 2.8 % 4.7 % 2.4 % 9.6 % 1.3 % 0.7 % 0.4 % 7.4 % 6.8 % 5.5 % 3.0 % 4.8 % 2.5 % 7.1 % 1.6 % 0.1 % 0.5 % 6.8 % 6.4 % 4.4 % 3.6 % 5.1 % 3.2 % 11.4 % 0.5 % 0.4 % 0.3 % 8.4 % 5.3 % 4.6 % 2.8 % 4.2 % 2.3 % 9.5 % 1.4 % Total 39.0 % 39.1 % 39.8 % 42.0 % 39.2 % Riz, céréales Viandes, poisson, légumes, fruits… Alimentation transformée Habillement Logement Equipement, services maisons Santé Education Transport Autres biens et services 15.4 % 18.6 % 4.9 % 2.0 % 8.3 % 2.5 % 0.6 % 6.2 % 1.4 % 1.1 % 15.4 % 18.5 % 4.8 % 2.4 % 8.1 % 2.5 % 0.9 % 1.6 % 5.6 % 1.1 % 13.8 % 16.3 % 5.2 % 2.2 % 11.2 % 2.6 % 0.6 % 1.4 % 6.1 % 0.8 % 12.2 % 18.0 % 4.9 % 2.9 % 6.7 % 2.3 % 0.7 % 1.6 % 7.7 % 0.8 % 14.7 % 18.3 % 4.9 % 2.5 % 8.5 % 2.6 % 0.9 % 1.5 % 5.9 % 1.0 % Total 61.0 % 60.9 % 60.2 % 57.8 % 60.8 % 1.1 % 4.1 % 8.4 % 0.6 % Part budgétaire des biens informels Variation de l'indice des prix (Stone) Annexe n° 2 : Structure de la consommation et taux de pression par déciles de dépenses Déciles de Dépenses 1 2 3 4 5 6 7 8 9 10 Moyen 2.0 % 1.9 % 2.2 % 2.4 % 2.2 % 2.4 % 2.6 % 2.6 % 2.8 % 4.1 % Taux de TVA effectif Minimum 0.4 % 0.3 % 0.4 % 0.8 % 0.4 % 0.4 % 0.6 % 0.7 % 0.8 % 0.7 % Maximum 6.2 % 5.5 % 5.4 % 6.0 % 5.0 % 6.5 % 4.7 % 5.9 % 7.3 % 10.6 % Sources : Centre Fiscale des Entreprises, MADIO, calculs propres, estimés à partir des données fiscales de l’année 1998. 27 Coefficient de Variation 10.3 % 8.7 % 9.4 % 7.3 % 6.7 % 8.1 % 6.7 % 7.2 % 9.0 % 8.0 %
https://openalex.org/W4312961890	https://materconstrucc.revistas.csic.es/index.php/materconstrucc/article/download/3053/3848	es		Hornos con recuperadores para la fabricación de cemento	Materiales de construcción	1,950	cc-by	629	Anónimo Materiales de Construcción, vol. 13 (1950) 618^12 HOBgQS COH EEOUPIKADORBS P A M LA FA3RICACI0H PB CEMEMTO, Anónimo. Dei "PIT AÎTO QUAEEr% 132^ Enero 1950. La conocida casa constructora de maquinaria Allis-Chalmers acaba de construir un nuevo sistema de hornos con recuperadores de calor para la fabricación de cemento, tanto por vía seca como en húmedo. Este nuevo pro- cedimiento lleva las iniciales A-C-L (Allis-Chalmers-Lellep) y representa un avance de consideración sobre el sistema de horno-recuperador inventado el Dr» 0»G« Lellep haco por unos 22 años, TSn la fig, 5 pueden verse los esquemas correspondientes a ambos t¿ pos de fabricaoiông el de la parte superior corresponde a un proceso r^^- vía Beca mientras que el inferior representa una fabricación por vía 1 .cda^ Bn el nuevo sistema A-C-L, el material entra en forma de bolitas, de unos 12 mm, de diámetro aproximadamente, hechas con el crudo ligeramente humedecido o con la torta de filtración previamente desmenuzada. Estas bolas se llevan - al horno rotatorio por medio de una parrilla móvil a través de una cámara con dos,compartimiento^, - Al atravesar el crudo, en forma de lecho poroso de bolitas, estas cámaras, los gasep calientes del horno rotativo atraviesan d¿ oho lecho dos veces consecutivas, extrayéndose la cantidad máxima de calor de dichos gases y reduciendo el volumen, temperatura y contenido en polvoe de los humos que salen por la chimenea. Cuando el crudo alcanza el horno,ya 00 ha producido un 30^- de la calcinación. Mediante el nuevo procedimiento A-C-L se consiguen ahorros de coflibustible desconocidos hasta la fecha. Se menciona la cifra de 1,030000 Koal, por Tm, de clinker cocido, tanto para el proceso en húmedo como para la vía seca Este bajo consumo de combustible va asociado a unos reducidos costos de instalación, facilidad de emplear cualquier tipo de carbón, economías en - IHSTITÜTO TÉCNICO DE LA CONSTRUCCIÓN Y DEL CEMENTO - (c) Consejo Superior de Investigaciones Científicas Licencia Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0) https://materconstrucc.revistas.csic.es -37 la molturaciôn y preparación de la hulla^ etc. 3e dice^ por otra parte, que los nuevos hornos así equipados^ son una tercera parte más cortos que los o£ diñarlos, lo cual supone ahorros en la instalación» En los esquemas de la fig. 5 los números encerrados en circuios - tienen el significado siguientes (l) Silo grande de alimentaciônj (2) Alimen. tador rotatorio5 (3) Elevador de cangilones5 (4) Rebosaderos del alimentador5 (5) Alimentador de preoisiônj (6) Tambor granulador; (7) Transportador-distribuidorj (8) Tolva de alimentación a la parrilla^ (9) Pa3:^rilla móvil; (10) cámara desecadoraj (11) Cámara de pre-calcinación; (12) Descenso del crudo de la parrilla a la entrada del horno; (13) Horno rotativo; (14) Sistema de fuegos (boquilla, soplante, ét); (15) Enfriador de aire.; (16) Soplante para el enfriador de cliriker; (17) Chimenea del enfriador; (18) Soplante; (19) Gran caja de succión; (20) Caja de succión colocada debajo del compartirnie^ to de secado^ (21) Chimenea auxiliar^ (22) Transportador de cadena; (23) - Transportador do hélice para la recuperación del crudo que cae a través de la parrilla; (24) Piltro de tambor; (25) Transportador para la torta del fi¿ tro; (26) Desmenuzador; (27) Tubo de caída del crudo desmenussado; (2,8) Tilev¿ I:r de cangilones; (29) Silo; (30) Alimentador; (31) Transportador . hélice, oolocado debajo del alimentador; (32) Caída de crudo en operación normal; - (33) Desintegradorej (24) Parrilla de sacudidas; (35) Entrada auxiliar del cru do, .' Dada la brevedad de la nota que comentamos en las presentes líneas, no es posible enj-uicíar qué grado de innovación suponen los sistemas A-C-L de la Allis-Chalmers, sobre los conooidos hornos Lepol, a los cuales se asemejan extraordinariamente» IHSTITüTO TSCIÍICO DE LA COISTHUCCIOI Y DEL CEMENTO - (c) Consejo Superior de Investigaciones Científicas Licencia Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0) https://materconstrucc.revistas.csic.es
https://openalex.org/W3046354620	https://pard.mk.ua/index.php/journal/article/download/197/158	Ukrainian	null	Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону	Publìčne upravlìnnâ ta regìonalʹnij rozvitok	2,020	cc-by-sa	5,276	«Public Administration and Regional Development» https://pard.mk.ua/index.php/journal «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal УДК 353:[351.851:005.21 DOI: 10.34132/PARD2020.08.12 УДК 353:[351.851:005.21 DOI: 10.34132/PARD2020.08.12 ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ Сиченко В.В., ректор, д-р наук з держ. упр., професор, КЗВО «Дніпровська академії неперервної освіти», Дніпро, Україна Хитько М.М., канд. наук з держ. упр., доцент КЗВО «Дніпровська академії неперервної освіти», Дніпро, Україна У статті досліджується необхідність та особливості стратегічного планування освітнього розвитку на регіональному рівні з використанням наявних проблем у якості основи цілепокладання управлінської діяльності. На основі системного аналізу та виокремлення спільного і особливого у тенденціях освітнього розвитку на рівні країни та регіону визначаються базові проблеми стратегічного планування та управління. Здійснюється узагальнюючий ан аліз розвитку регіональної системи освіти на прикладі Дніпропетровської області з виходом на постановку стратегічних цілей її реформування. Показано що основою стратегічного планування розвитку освіти є розроблення нового цілепокладання розвитку української освіти на базі цінностей демократичного суспільства, особистісного розвитку, спрямованості до європейського освітнього простору; створення каркаса нового законодавчого поля функціонування освіти;визначення засад рівного доступу кожного громадянина до якісної освіти всіх рівнів як магістральний напрям її розвитку; формування змісту освіти на основі державних стандартів як важеля управління якістю освіти в країні. 596 Isue DOI: 10.34132/pard Акцентуються проблеми стратегізації управління освітою на регіональному рівні та необхідність пов’язаних з цим інновацій у нормативно-організаційній царині. Показано що в основі 596 Isue DOI: 10.34132/pard Акцентуються проблеми стратегізації управління освітою на регіональному рівні та необхідність пов’язаних з цим інновацій у нормативно-організаційній царині. Показано що в основі 596 Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону освітньої парадигми має лежати завдання забезпечення потреб споживачів освітніх послуг, розширення можливостей доступу до якісної освіти, забезпечення ефективного соціального механізму конкурентоздатності випускників шкіл на сучасному ринку праці. На основі проведеного дослідження робляться пропозиції щодо удосконалення стратегічного планування розвитку основних складових регіональної системи освіти. Ключові слова: стратегічне планування, регіональне управління, регіональна система освіти, оптимізація державного управління, ціле покладання управління. Постановка проблеми у загальному вигляді. Приймаючи виклики сьогодення, українське суспільство прагне модернізуватися на інноваційній основі, аби набути прискорення, характерного для глобального світового поступу. Головним рушієм цього процесу завжди були і незмінно залишаться люди — компетентні і кваліфіковані, конструктивні і конкурентоспроможні, компромісні і консенсусні. У розвитку людського потенціалу, людського капіталу ключову і, головне, безперервно зростаючу роль відіграє освіта. Виключно завдяки освітній складовій в індексах людського розвитку та глобальної конкурентоспроможності Україна за цими показниками утримує серединні позиції в колі багатьох країн світу. Водночас дедалі стає очевидним, що екстенсивний шлях зростання вітчизняної освіти себе повністю вичерпав. На порядку денному досягнення нею нових якісних характеристик, які відповідають вимогам сьогодення. Їх досягнення вимагає чіткого стратегічного планування освітнього розвитку як на національному, так і регіональному рівнях. ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ Аналіз останніх досліджень і публікацій. Сучасний період стратегії модернізації вітчизняної освіти розпочато у 2014 р., у якому новий погляд на якісний рівень та суспільну роль освіти зумовили Революція гідності, а також підписання Угоди про асоціацію України з Європейським Союзом. У державній стратегії пріоритетними є 597 ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal орієнтири щодо людиноцентризму, забезпечення якості і доступ- ності, конкурентоспроможності й ефективності, інтеграції в євро- пейські та світові простори освіти і досліджень, періодичного націо- нального і міжнародного моніторингу, комплексного аналізу стану і перспектив розвитку освіти, стабільного необхідного і достатнього забезпечення фінансовими та інтелектуальним ресурсами. Держав- на освітня стратегія покликана змінювати ставлення суспільства до освіти з огляду на перспективу реформування освітньої сфери, як того вимагають євроінтеграційні та глобалізаційні виклики ХХІ ст. Її окремі складові досліджуються у роботах Я. Мельник, В. Кременя, С. Одайник, А. Ревко, С. Тульчинської та інших авторів. Разом з тим на периферії досліджень залишається регіональний аспект стратегічного планування освітнього розвитку на проблемно- цільовій основі. орієнтири щодо людиноцентризму, забезпечення якості і доступ- ності, конкурентоспроможності й ефективності, інтеграції в євро- пейські та світові простори освіти і досліджень, періодичного націо- нального і міжнародного моніторингу, комплексного аналізу стану і перспектив розвитку освіти, стабільного необхідного і достатнього забезпечення фінансовими та інтелектуальним ресурсами. Держав- на освітня стратегія покликана змінювати ставлення суспільства до освіти з огляду на перспективу реформування освітньої сфери, як того вимагають євроінтеграційні та глобалізаційні виклики ХХІ ст. Її окремі складові досліджуються у роботах Я. Мельник, В. Кременя, С. Одайник, А. Ревко, С. Тульчинської та інших авторів. Разом з тим на периферії досліджень залишається регіональний аспект стратегічного планування освітнього розвитку на проблемно- цільовій основі. Формулювання цілей статті (постановка завдання). Визначення базових проблем регіональної освіти як основи стратегічного планування її розвитку. Виклад основного матеріалу дослідження. Саме для ХХІ ст. стає характерним розуміння того, що освіта не може надалі залишатися у сфері відокремленої галузевої чи відомчої політики, розглядатися як витратне соціальне благо та безповоротна стаття видатків, а є продуктивним чинником й умовою розвитку, відтак повинна набути статусу загальнонаціональної стратегії. Загальноцивілізаційні тенденції викликали появу нової парадигми освіти, її переорієнтацію з держави на людину, на фундаментальні людські цінності, на послідовну демократизацію освітнього процесу й освітньо-педагогічної ідеології загалом [2, с.10]. ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ Сучасна освітня стратегія має реалізовуватись у контексті потреб модернізації країни, згідно зі Стратегією сталого розвитку «Україна — 2020» (2015 р.), Угодою про асоціацію між Україною та Європейським Союзом (2014 р.), іншими стратегічними документами української держави, що визначають її європейський і світовий статус у майбутньому. З урахуванням кризового економічного стану країни потрібно насамперед визначити найважливіші освітні проблеми та обґрунтувати способи їх розв’язання в найближчі 598 Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону роки та в перспективі. Очевидно, що реформування не може бути успішним, якщо здійснюється як потік постійних, локальних змін, часто суперечливих, концептуально не об’єднаних, не має широкої підтримки та розуміння в освітян і суспільстві. у у На загальнодержавному рівні основою стратегічного планування розвитку освіти є розроблення нового цілепокладання розвитку української освіти на базі цінностей демократичного суспільства, особистісного розвитку, спрямованості до європейського освітнього простору; створення каркаса нового законодавчого поля функціонування освіти; визначення засад рівного доступу кожного громадянина до якісної освіти всіх рівнів як магістральний напрям її розвитку; формування змісту освіти на основі державних стандартів як важеля управління якістю освіти в країні; створення ресурсів для переходу на 12-річну загальну середню освіту; перехід до варіативної освіти; запровадження незалежного зовнішнього оцінювання; упровадження нової моделі атестації і державної акредитації; поширення інформаційних та загалом інноваційних освітніх технологій; рух до багатоканального фінансування галузі; становлення державно-громадського управління [4, с.26]. Разом з тим слід вказати і на наявні системні проблеми, які мають стати основою визначення стратегічних напрямів розвитку галузі. Це відсутність системної науково обґрунтованої ідеології розвитку освіти з послабленням її культуротворчої місії; згортання низки освітніх мереж, зокрема, дошкільної, в умовах відсутності прогнозу демографічної ситуації і потреб розвитку освіти на регіональному рівні; невизначеність функцій професійно-технічних навчальних закладів і технікумів, училищ, коледжів у нових ринкових умовах, інноваційній економіці; неадекватне розширення мережі вищої освіти з ризиками здобуття молоддю неякісної освіти; невирішеність проблем корупції і хабарництва; відсутність ефективного моніторингу реформ тощо. Загалом українська освіта фактично не стала загальнонаціональним пріоритетом [11, с.135]. ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 599 Регіональна система освіти розвивається у межах загальнонаціональних тенденцій, досягнень та проблем. Узагальнення статистичних даних дає уявлення про базові тренди ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 599 Регіональна система освіти розвивається у межах загальнонаціональних тенденцій, досягнень та проблем. Узагальнення статистичних даних дає уявлення про базові тренди «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal та підстави стратегічного планування основних складових регіональної системи освіти. та підстави стратегічного планування основних складових регіональної системи освіти. ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ Значущість дошкільної освіти постійно зростає по мірі усталення високого статусу дитинства, наростання кількості міжнародних документів про охорону дитинства, і особливо після проведення у 2010 р. під егідою ЮНЕСКО Всесвітньої конференцію з навчання, виховання і розвитку дітей раннього і дошкільного віку, базові положення якої відображені у законах України «Про дошкільну освіту» та «Про охорону дитинства», у Національній стратегії розвитку освіти України до 2021 р. Однак слід констатувати, що економічні та управлінські негаразди призвели до скорочення кількості дитячих садків, мережа дошкільних навчальних закладів не була збережена див. рис. 1. Рис.1. Кількість закладів дошкільної освіти, од. Джерело:[1] Рис.1. Кількість закладів дошкільної освіти, од. Джерело:[1] 600 Isue DOI: 10.34132/pard У 2010 р. дошкільну освіту п’ятирічних дітей було визнано обов’язковою. Наслідком цих змін стало те, що майже всі діти старшого дошкільного віку охоплені різними формами здобуття дошкільної освіти, однак молодші категорії дошкільнят значною мірою залишаються за межами організованих форм соціалізації. Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону Сьогодні у регіоні досить великою є проблема нестачі місць у дошкільних закладах, збільшується наповненість у групі, що виступає небезпечним фактором зниження якості освіти див. рис. 2 та рис. 3. Рис.2. Кількість місць в закладах дошкільної освіти. Джерело:[1] Рис.2. Кількість місць в закладах дошкільної освіти. Джерело:[1] Рис.3. Кількість осіб в закладах дошкільної освіти. Джерело:[1] Рис.3. Кількість осіб в закладах дошкільної освіти. Джерело:[1] 601 ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal Рис.4. Охоплення дітей в закладах дошкільної освіти, %. Джерело:[1] К ’ б і Рис.4. Охоплення дітей в закладах дошкільної освіти, %. Джерело:[1] Ключовим перспективним розв’язанням проблеми доступності дошкільної освіти може бути впровадження кращого зарубіжного досвіду функціонування мережі дошкільних навчальних закладів різних типів і форм власності: стаціонарні та сезонні ясла, садки з різною тривалістю, дошкільні відділення при початкових класах, материнські школи, майданчики або відкриті дитячі садки [9, с.65]. Від органів управління на місцях потрібні сприяння приватним ініціативам; створення груп з короткотривалим перебуванням у них дітей; відновлення діяльності дошкільних навчальних закладів, що тривалий час використовувались не за призначенням; відкриття на базі загальноосвітніх навчальних закладів навчально-виховних комплексів, все це сприяло б зростанню рівня охопленості дітей в системі дошкільної освіти, див. рис.4. Практична спрямованість підготовки кадрів системи дошкільної освіти потребує значного поглиблення, адресного замовлення, сучасної політики розвитку людського ресурсу. ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ 602 Isue DOI: 10.34132/pard У цілому здійснення стратегічного планування у регіональній системі дошкільної освіти вимагають такі напрями як підпорядкування реформ оптимізації життєдіяльності дитини дошкільного віку, узгодження організаційних аспектів наступності дошкільної та початкової освіти, інтеграція діяльності управлінських інституцій, що опікуються дошкільним 602 Isue DOI: 10.34132/pard У цілому здійснення стратегічного планування у регіональній системі дошкільної освіти вимагають такі напрями як підпорядкування реформ оптимізації життєдіяльності дитини дошкільного віку, узгодження організаційних аспектів наступності дошкільної та початкової освіти, інтеграція діяльності управлінських інституцій, що опікуються дошкільним Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону дитинством, громадська учать у реформуванні дошкільної освіти, упровадження цілісної регіональної програмно-методичної моделі життєзабезпечення, підтримки та супроводу дитячого розвитку, кардинальні зміни системи духовного та фізичного виховання дітей на основі сучасних концепцій та технологій. На регіональному рівні, на нашу думку, необхідно створити єдиний центр комплексного сервісу дошкільної освіти, що забезпечить перехід від вертикального адміністрування діяльності дошкільних закладів на стратегію адаптивного сервісу на принципах поліархічно- горизонтальної взаємодії усіх причетних до дошкільної галузі інституцій регіональної системи освіти. Загальноосвітня школа забезпечує становлення учня як особистості, закладає фундамент для успішної самореалізації людини впродовж життя. І на національному, і на регіональному рівні зазнали істотної трансформації цільова спрямованість загальної середньої освіти, її зміст, організація навчального процесу, дидактико-методичне забезпечення, підходи до оцінювання освітніх результатів у напрямі посилення особистісної орієнтації освіти, її розвивального, компетентнісного, демократичного характеру. Створено умови для розвитку обдарованості дітей, більш повного задоволення освітніх запитів і потреб учнів — гімназії, ліцеї, колегіуми. Разом з тим відбувається оптимізація шкільної мережі, яка не завжди є ефективною внаслідок організаційних та фінансових причин. Протягом 2010-2019 років спостерігається динаміка зменшення кількості ЗСО див. рис. 5., на ряду із зростанням кількості учнів у ЗСО протягом 2016-2020 років, та незначним зростанням кількості учителів (див. рис. 7), у ЗСО протягом 2019-2020 років. Серед нерозв’язаних проблем — недостатня орієнтованість шкільної освіти на практичні потреби учня у контексті особливостей регіонального розвитку; далеке від оптимального оснащення шкіл, інформаційне забезпечення регіонального простору; відсутність єдиної регіональної системи моніторингу якості загальної середньої освіти як основи управління. 603 ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal Рис.5. Кількість закладів загальної середньої освіти, од. Джерело:[1] Рис.6. Кількість учнів в закладах загальної середньої освіти, тис. осіб.Джерело:[1] «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal Рис.5. Кількість закладів загальної середньої освіти, од. Джерело:[1] Рис.5. Кількість закладів загальної середньої освіти, од. Джерело:[1] Рис.6. Кількість учнів в закладах загальної середньої освіти, тис. осіб.Джерело:[1] Рис.6. Кількість учнів в закладах загальної середньої освіти, тис. ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ осіб.Джерело:[1] Профілізація старшої школи як одна із визначальних засад її ефективного функціонування в сучасних умовах відбувається недостатньо інтенсивно та зі значними труднощами, можна говорити про несформованість мережі шкіл, що реально забезпечують профільну спеціалізацію. Рівень комп’ютеризації школи у 5-6 разів нижчий, ніж у Європі. Зберігається нерівність у забезпеченні доступу до якісної освіти учнів сільської місцевості порівняно з міськими школярами. Isue DOI: 10.34132/pard 604 Isue DOI: 10.34132/pard Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону Рис.7. Кількість учителів в закладах загальної середньої освіти, тис. осіб.Джерело:[1] Рис.7. Кількість учителів в закладах загальної середньої освіти, тис. осіб.Джерело:[1] Одним із напрямів реформування регіональної школи слід вважати вирішення проблеми малокомплектних навчальних закладів на основі створення освітніх округів, а також розширення можливостей для широкого використання форм індивідуального та дистанційного навчання. А головне – необхідно на рівні регіону забезпечити інтегрованість та узгодженість вирішення проблеми фінансового, управлінського, організаційного і кадрового забезпечення загальноосвітньої школи. Регіональне стратегічне планування розвитку загальноосвітньої школи має відображати зміни у національній стратегії освітнього розвитку, нові базові цінності, які утверджу- ють соціально та особистісно значущий сенс здобуття освіти, пріоритетність соціальної та інформаційної функцій школи [5, с.9]. У центрі регіонального стратегічного управління освітою мають бути завдання зміни структури школи, місії та цільові освітні спрямування її ступенів, упровадження 12-річного тер- міну навчання, розширення можливостей профільного навчання. Для нашого регіону з його промисловим потенціалом особливо важливим є становлення інноваційної моделі шкільної освіти, 605 ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal формування і розвиток інтегрованого інформаційно-освітнього середовища з використанням дистанційних технологій. Окремої уваги потребує вирішення проблеми оптимізації мережі загальноосвітніх навчальних закладів з метою ефективної реалізації завдань загальної середньої освіти. В її основі має лежати завдання забезпечення потреб споживачів освітніх послуг, розширення можливостей доступу до якісної освіти, забезпечення ефективного соціального механізму конкурентоздатності випускників шкіл на сучасному ринку праці. Разом з тим нагальною є проблема збереження шкіл, особливо в сільській місцевості, як культурних й освітніх осередків. До питання закриття шкіл місцеві органи державної виконавчої влади повинні підходити виважено, враховувати демографічний прогноз, соціально-економічні умови, думку громади [6, с.84]. Важливо забезпечити функціонування у кожному сільському населеному пункті, де є діти шкільного віку, принаймні початкової школи. Вирішення цього питання потребує створення окремої регіональної цільової соціально-освітньої програми на засадах соціального партнерства з широким використанням виховного й освітнього потенціалу сільського соціуму. ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ Доцільно створити й поширити мережу освітніх округів, здатних забезпечити розгалужену інфраструктуру освітнього простору регіону, кращий доступ навчальних закладів до освітніх ресурсів, підвищити якість освітніх послуг і поліпшити умови для реалізації програм профільного навчання у сільських загальноосвітніх навчальних закладах. У професійній освіті регіону, як і країни загалом, відбулися значні трансформації, зокрема перехід до ступеневої системи навчання, запровадження нових форм підготовки з інтегрованих (укрупнених) професій, здійснення незалежної кваліфікаційної атестації, розширення співпраці з роботодавцями тощо. Поряд із цим спостерігалися негативні тенденції: скорочення мережі та контингенту учнів ПТНЗ (див. рис. 8, 9, 10), повільне оновлення переліку напрямів професійної підготовки, зменшення обсягів дер- жавного замовлення та фінансування професійної освіти. Isue DOI: 10.34132/pard 606 Isue DOI: 10.34132/pard Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону Рис.8. Кількість закладів професійної освіти, од. Джерело:[1 ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 607 Рис.8. Кількість закладів професійної освіти, од. Джерело:[1 Рис.9. Кількість учнів в закладах професійної освіти, тис. осіб. Джерело:[1] Рис.10. Кількість викладачів в закладах професійної освіти, тис. осіб.Джерело:[1] Рис.8. Кількість закладів професійної освіти, од. Джерело:[1 Рис.9. Кількість учнів в закладах професійної освіти, тис. осіб. Джерело:[1] Рис.9. Кількість учнів в закладах професійної освіти, тис. осіб. Джерело:[1] Рис.9. Кількість учнів в закладах професійної освіти, тис. осіб. Джерело:[1] ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 607 Джерело:[1] Рис.10. Кількість викладачів в закладах професійної освіти, тис. осіб.Джерело:[1] Р 10 Кі і д і д ф ій ї і ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 607 Рис.10. Кількість викладачів в закладах професійної освіти, тис. осіб.Джерело:[1] 607 «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal Сьогодні якість професійної підготовки кваліфікованих робітників і молодших спеціалістів в регіоні не повною мірою відповідає вимогам сучасного ринку праці. Основні проблеми спричинені збереженням значної централізації в управлінні професійною освітою, недостатньою взаємодією професійних навчальних закладів, роботодавців і науковців у розробленні державних стандартів професійної освіти, багатотипністю мережі ПТНЗ, недосконалістю механізмів фінансування. Неефективною є система професійної орієнтації молоді на робітничі професії та консультування з професійної кар’єри в ринкових умовах. Чимало складних проблем накопичилося в організації професійного навчання на виробництві. Стратегічним напрямом розвитку слід вважати створення багатогалузевих професійних навчальних закладів із сучасною матеріально-технічною базою з орієнтацією на інтегрування професій та прив’язкою підготовки фахівців до потреб національного та регіональних ринків праці. Необхідним є вирішення проблеми фінансування навчальних закладів державної форми власності на регіональному рівні на засадах, які унеможливлюють його неадекватне скорочення. ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ Потребує також урегулювання нормативна база використання майна ПТНЗ, їх перепрофілювання, відчуження приміщень, земельних ділянок, а також скорочення робочих місць для педагогічних працівників. Серед цілей стратегічного планування у даній сфері слід визначити прогнозування потреб регіонального ринку праці, створення дієвих механізмів соціального партнерства, використання економічних інструментів стимулювання розвитку та роботодавців до інвестування у професійну підготовку фахівців [13, с.126]. На нашу думку, необхідне створення маркетингових і профорієнтаційних служб у навчальних закладах. Необхідно створити ефективну систему регіонального замовлення робітничих кадрів на основі ефективної статистики та прогнозів. Потрібно також подальше розширення експериментальної апробації ефективних форм підготовки робітничих кадрів, таких як модель освітньо- професійного кластеру з формування трудового потенціалу в регіоні чи модель управління маркетингом системи ПТО. Достатньо 608 Isue DOI: 10.34132/pard Isue DOI: 10.34132/pard Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону перспективним є створення регіональних центрів маркетингових досліджень у сфері професійної освіти. Особливе значення має забезпечення поетапної оптимізації мережі професійно-технічних навчальних закладів та ефективне управління ними як комунальною власністю з урахуванням науково обґрунтованих критеріїв, особливостей соціально-економічного розвитку регіону і вимог ринку праці. Доцільним є створення єдиної регіональної системи професійної освіти, що поєднує різні типи професійних навчальних закладів з трансформацією існуючих у багаторівневі і багатопрофільні професійні ліцеї та коледжі. До- статньо ефективним напрямом розвитку є формування регіональ- них інтегрованих освітньо-професійних систем у вигляді освітньо- професійних округів чи навчально-науково-виробничих кластерів, що поєднують заклади професійної освіти з регіональними підприємствами та організаціями, науковими установами. Найбільш очевидною ознакою розвитку системи вищої освіти регіону є її екстенсивне зростання, в основному за рахунок радикальної зміни співвідношення університетів, академій та інститутів з одного боку – та коледжів – з іншого. Демографічна та економічна кризи негативно позначалася на кількості закладів вищої освіти (рис. 11), та на якості надання освітніх послуг, конкурентоспроможності вищої освіти на вітчизняному та європейському ринках праці. Рис.11. Кількість закладів вищої освіти, од. Джерело:[1] Рис.11. Кількість закладів вищої освіти, од. Джерело:[1] 609 ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal Рис.12. Кількість студентів в закладах вищої освіти, тис. осіб. Джерело:[1] Рис.13. Кількість прийнятих на навчання в заклади вищої освіти, тис. осіб. Джерело:[1] Слід сказати що мережа закладів вищої освіти державної/ «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal Рис.12. Кількість студентів в закладах вищої освіти, тис. осіб. Джерело:[1] Рис.12. Кількість студентів в закладах вищої освіти, тис. осіб Джерело:[1] Рис.13. ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ Кількість прийнятих на навчання в заклади вищої освіти, тис осіб Джерело:[1] Рис.13. Кількість прийнятих на навчання в заклади вищої освіти, тис. осіб. Джерело:[1] р Слід сказати, що мережа закладів вищої освіти державної/ комунальної форми власності регулювалося здебільшого заходами адміністративного характеру. Такі дії призводили до протестних настроїв чи відкритих протестів керівництва й студентів закладів, як, наприклад, у випадку з Дніпропетровським аграрним університетом. Адже процеси реорганізації майже завжди передбачали не скорочення державного фінансування та інших видатків, утрату усталених традицій. Між тим серед першочергових завдань є створення потужних національних регіональних університетів шляхом об’єднання існуючих закладів, проте на компромісній та обґрунтованій основі. 610 Isue DOI: 10.34132/pard Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону На сьогодні можна констатувати такі риси регіональної системи вищої освіти як розпорошеність мережі закладів, низька конкурентоспроможність більшості з них, повільний розвиток автономії, відсутність сучасних систем внутрішнього забезпечення якості, слабкість дослідницько-інноваційної основи підготовки фахівців, низький рівень інтернаціоналізації, неефективність фінансових механізмів ресурсного забезпечення. Однією з найбільших проблем є відірваність вищої освіти від потреб та інтересів ринку праці. Попит ринку праці на кваліфікованих робітників перевищує попит на фахівців із вищою освітою. Відповідно серед безробітних кількість перших на чверть менша, ніж других. Система освіти цього не відображає, і за останні чверть століття випуск кваліфікованих робітників скоротився вполовину, а випуск фахівців із вищою освітою зріс більш ніж у півтора рази [3, с.12]. Формально від місцевої влади залежить формування державного замовлення на підготовку фахівців з вищою освітою, оскільки саме регіональні державні адміністрації мають повноваження погоджувати пропозиції місцевих закладів вищої освіти про обсяги державного замовлення. Однак місцева влада, не маючи відповідальності за фінансування державного замовлення, як правило, підтримує будь-які пропозиції місцевих закладів про високі обсяги держзамовлення без оцінки реальної ситуації. Очевидно, до цього процесу мають залучатися органи місцевого самоврядування з відповідним перерозподілом функцій управління та фінансування. Це питання має вирішуватися у комплексі із завданням забезпечення реальної фінансової автономії державних закладів вищої освіти, які залишаються у статусі бюджетних установ. ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 611 Очевидно, що стратегічне планування розвитку системи вищої освіти на рівні регіону має підпорядковуватися реформаторським настановам нового Закону України «Про вищу освіту» (набув чинності 6 вересня 2014 р.). На часі реалізація таких напрямів реформ, як унормування сучасної організації вищої освіти, забез- печення автономії і академічної свободи, створення національної та регіональної системи забезпечення якості вищої освіти, 611 «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal вадження нових стандартів вищої освіти та освітніх програм. ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 ПРОБЛЕМНО-ЦІЛЬОВІ ЗАСАДИ СТРАТЕГІЧНОГО ПЛАНУВАННЯ РОЗВИТКУ СИСТЕМИ ОСВІТИ РЕГІОНУ Стратегічними напрямами сучасної модернізації вищої освіти слід вважати упорядкування мережі закладів вищої освіти з концентрацією потенціалу і підвищити конкурентоспроможності. Контингент студентів має бути приведений у відповідність до реальної ресурсної бази вищої освіти регіону. Доцільно обмежити граничну чисельність ліцензованих місць у закладах, а головне — кількість самих закладів вищої освіти. Необхідно укрупнити існуючі вищі навчальні заклади шляхом їх об’єднання в потужні регіональні заклади вищої освіти. Узагальнюючи напрями стратегічного планування розвитку регіональної освіти, слід виділити глобальні, інтегруючі його напрями на основі виокремлення кола пріоритетних проблем. Це упорядкування та осучаснення мережі закладів освіти всіх рівнів; стандартизація нормативів якості освіти; створення нової еко- номіки освіти; концентрація інноваційно-технологічних ресурсів; запровадження державно-громадської моделі управління освітою, розвиток соціального партнерства у розв’язанні освітніх проблем. Щоби регіональна стратегія в освітній сфері забезпечувала доступ до якісної освіти, враховувала вимоги соціально-економічного розвитку регіону, вкрай важливо розробити, цілеспрямовану регіональну програму інноваційного розвитку освіти. Найважливі- шим у її розробленні і впровадженні є досягнення реальної взаємодії чинників: політичного, соціального, економічного, культурного і, власне, освітнього. Таке поєднання дасть змогу вивести проблему якості освіти за межі суто галузевої проблематики, збалансувати пріоритети, створити систему безперервної освіти [7, с.19]. Ця си- стема має розвиватися на основі нормативного і змістового узгод- ження всіх її ланок, організаційного взаємодоповнення формальної, неформальної та інформальної освіти. Стратегічний пріоритет освітньої стратегії — формування регіональної спільноти, яка по- стійно навчається, засвоюючи одночасно демократичні цінності, ро- звиваючи громадянське суспільство, утверджуючи людиноцентризм в освіті. 612 Isue DOI: 10.34132/pard Висновки. Таким чином, стратегічне планування розвитку 12 Isue DOI: 10 34132/pard Висновки. Таким чином, стратегічне планування розвитку Висновки. Таким чином, стратегічне планування розвитку 612 Isue DOI: 10.34132/pard Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону регіональної освіти є ефективним механізмом забезпечення її реформування у контексті потреб громадян та економіки. Його ефективність визначається розумінням пріоритетних проблем освітнього розвитку та їх взаємоузгодженності з соціально- економічним розвитком регіону. Необхідна інтегрована стратегія здійснення інновацій як в усіх сферах освіти, так і в інших сферах життєдіяльності регіону. Це вимагає зростання управлінських функцій регіональної влади та місцевого самоврядування щодо планування та забезпечення освітніх реформ у регіоні. Стаття надійшла до редакції: 05.04.20 ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 613 Стаття надійшла до редакції: 05.04.20 ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 613 «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal PROBLEM-GOAL PRINCIPLES OF STRATEGIC PLANNING OF THE REGIONAL EDUCATION SYSTEM DEVELOPMENT Viktor Sychenko, Doctor of Science in Public Administration, Professor, CIHE “Dnipro Academy of Continuing Education, Dnipro, Ukraine Maiya Khytko, PhD in Public Administration, Docent of the Department of Public Administration and Law of , CIHE “Dnipro Academy of Continuing Education, Dnipro, Ukraine 614 Isue DOI: 10.34132/pard The article examines the need and features of strategic planning of educational development at the regional level using existing problems as a basis for goal-setting management activities. Based on a systematic analysis and separation of common and special in the trends of educational development at the level of the country and the region, the basic problems of strategic planning and management are determined. At the national level, the basis of strategic planning for the development of education is the development of a new goal for the development of Ukrainian education based on the values of a democratic society, personal development, orientation to the European educational space. The generalized analysis of development of regional system of education on an example of the Dnepropetrovsk area with an exit on statement of the strategic purposes of its reforming is carried out. The problems of strategizing education management at the regional level and the need for related innovations in the regulatory and organizational sphere are emphasized. In order for the regional strategy in the ﬁ eld of education to provide access to quality education, take into account the requirements of socio-economic development of the region, it is extremely important to develop a purposeful regional program of innovative development of education. The most important in its development and implementation is to achieve real interaction of factors: political, social, economic, cultural 614 Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону and, in fact, educational. Based on the study, proposals are made to improve the strategic planning of the main components of the regional education system. The strategic priority of the educational strategy is the formation of a regional community that is constantly learning, while mastering democratic values, developing civil society, and afﬁ rming human-centeredness in education. An integrated strategy for innovation is needed in all spheres of education, as well as in other spheres of life in the region. Key words: strategic planning, regional management, regional education system, optimization of public administration. References 1. Demohraﬁ chna ta sotsialna statystyka. Osvita (Demographic and social statistics. Education) Електронний ресурс: http://www.dneprstat.gov. ua/statinfo/o/2019/o1.pdf. Дата звернення: 20.06.2020. [in Ukrainian]. 2. Lyubchuk O.K. (2010). Teoretyko-metodolohichni zasady derzhavnoho upravlinnia neperervnoiu osvitoiu v Ukraini ta yii rehionakh / O.K. Lyubchuk, Kupryianov V. S.// Theoretical and methodological principles of public administration of continuing education in Ukraine and its regions. - Donetsk, 395 р. [in Ukrainian]. 3. Melnyk Y. V. (2015). Teoretychni ta orhanizatsiini osoblyvosti modernizatsii upravlinnia sferoiu osvity / Y. V. Melnyk // Theoretical and organizational features of modernization of education management: regional aspect : author’s ref. dis. ... cand. Sciences of the state. gov., Lviv, 20 p. [in Ukrainian]. 3. Melnyk Y. V. (2015). Teoretychni ta orhanizatsiini osoblyvosti modernizatsii upravlinnia sferoiu osvity / Y. V. Melnyk // Theoretical and organizational features of modernization of education management: regional aspect : author’s ref. dis. ... cand. Sciences of the state. gov., Lviv, 20 p. [in Ukrainian]. 4. Natsionalna dopovid pro stan i perspektyvy rozvytku osvity v Ukraini (National report on the state and prospects of education in Ukraine), NAPNU, Kyiv, (2016), 224 p. [in Ukrainian]. 4. Natsionalna dopovid pro stan i perspektyvy rozvytku osvity v Ukraini (National report on the state and prospects of education in Ukraine), NAPNU, Kyiv, (2016), 224 p. [in Ukrainian]. 5. Odaynyk S.F. (2015). Upravlinnia yakistiu zahalnoi serednoi osvity v rehioni: teoretyko-metodolohichnyi aspekt / S.F. Odaynyk // Management of quality of general secondary education in the region: theoretical and methodological aspect, Aylant, Kherson, 263 p. [in Ukrainian]. 5. Odaynyk S.F. (2015). Upravlinnia yakistiu zahalnoi serednoi osvity v rehioni: teoretyko-metodolohichnyi aspekt / S.F. Odaynyk // Management of quality of general secondary education in the region: theoretical and methodological aspect, Aylant, Kherson, 263 p. [in Ukrainian]. ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 615 6. Pastovensky O.V. (2014). Naukovi osnovy hromadsko-derzhavnoho ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 615 6. Pastovensky O.V. (2014). Naukovi osnovy hromadsko-derzhavnoho 615 «Public Administration and Regional Development» https://pard.mk.ua/index.php/journal upravlinnia zahalnoiu serednoiu osvitoiu v rehioni / O.V. Pastovensky // Scientiﬁ c bases of public-state management of general secondary education in the region, Ruta, Zhytomyr, 383 р. [in Ukrainian]. upravlinnia zahalnoiu serednoiu osvitoiu v rehioni / O.V. Pastovensky // Scientiﬁ c bases of public-state management of general secondary education in the region, Ruta, Zhytomyr, 383 р. [in Ukrainian]. upravlinnia zahalnoiu serednoiu osvitoiu v rehioni / O.V. Відомості про авторів / Information about the Authors Сиченко Віктор Володимирович: Комунальний заклад вищої освіти «Дніпровська академія неперервної освіти», вул. Володимира Антоновича 70, м. Дніпро, 49006, Україна Сиченко Віктор Володимирович: Комунальний заклад вищої освіти «Дніпровська академія неперервної освіти», вул. Володимира Антоновича 70, м. Дніпро, 49006, Україна Viktor Sychenko: Communal Institution of Higher Education “Dnipro Academy of Continuing Education”, st. Antonovich, 70, Dni- pro, 49006, Ukraine ORCID.ORG/0000-0001-9655-2317 ORCID.ORG/0000-0001-9655-2317 Е-mail: sychenko@dano.dp.ua Хитько Майя Миколаївна: Комунальний заклад вищої освіти «Дніпровська академія неперервної освіти», вул. Володимира Анто- новича 70, м. Дніпро, 49006, Україна Хитько Майя Миколаївна: Комунальний заклад вищої освіти «Дніпровська академія неперервної освіти», вул. Володимира Анто- новича 70, м. Дніпро, 49006, Україна Maiya Khytko: Communal Institution of Higher Education “Dnipro Academy of Continuing Education”, st. Antonovich, 70, Dnipro, 49006, Ukraine ISSN 2616-6216. Publ. upr. reg. rozvit. 2020, 8: 596-617 References Pastovensky // Scientiﬁ c bases of public-state management of general secondary education in the region, Ruta, Zhytomyr, 383 р. [in Ukrainian]. 7. 7. Revko A.M. (2019) Osvitnia dominanta u paradyhmalnomu vymiri sotsiohumanitarnoho prostoru rehionu / A.M. Revko, O. V. Brahynets // Educational dominant in the paradigmatic dimension of socio-humanitarian space of the region [vyd.], Chernihiv, 353 р. [in Ukrainian]. 8. Sychenko V.V. (2016). Socialium rationem ac viam nosse status socialis policy in consilio / V.V. Sychenko // Publica administrationem et administratione consuetudines. # II (XV). - p. 77-81. [in Ukrainian]. 9. Sychenko V.V. (2018). Rei publicae consilium Ucraina in agro perdurantes educationem / V.V. Sychenko, S.O. Rybkina, O.M. Martynenko// Publico administrationem et administratione consuetudines. # I (XVIII). - p. 74-79. [in Ukrainian]. 10. Sychenko V.V. (2019). Proprietates opportuna consilio in educatione, sicut instrumentum administrationis publici / V.V. Sychenko, M.M. Khytko// Dissertatio medica contingunt, technologiae provectus, innovations (materiae, ex scientiﬁ c practical colloquium International V). - Nemoros s.r.o.: Praga - p. 117-121 [in Ukrainian]. 11. Tulchynska S.O. (2012). Upravlinnia u sferi osviti i nauki yak faktor zabezpechennia staloho rozvytku derzhavy ta rehioniv / Tulchynska S.O., Melnychenko A.A., Akimova O.A. //Management in the ﬁ eld of education and science as a factor in ensuring sustainable development of the state and regions), Edelveis, Kyiv, 328 р. [in Ukrainian]. 11. Tulchynska S.O. (2012). Upravlinnia u sferi osviti i nauki yak faktor zabezpechennia staloho rozvytku derzhavy ta rehioniv / Tulchynska S.O., Melnychenko A.A., Akimova O.A. //Management in the ﬁ eld of education and science as a factor in ensuring sustainable development of the state and regions), Edelveis, Kyiv, 328 р. [in Ukrainian]. 12. Shevchuk A.V. (2013). Rehionalni osvitni systemy: teoriia, metodolohiia, praktyka innovatsiinoho rozvytku / A.V. Shevchuk // Regional educational systems: theory, methodology, practice of innovation development, Inst. Region. research. NAS of Ukraine, Lviv, 463 p. [in Ukrainian]. 12. Shevchuk A.V. (2013). Rehionalni osvitni systemy: teoriia, metodolohiia, praktyka innovatsiinoho rozvytku / A.V. Shevchuk // Regional educational systems: theory, methodology, practice of innovation development, Inst. Region. research. NAS of Ukraine, Lviv, 463 p. [in Ukrainian]. 13. Yurchuk L.M. (2012). Teoretyko-metodolohichni zasady ta mekhanizmy derzhavno-hromadskoho upravlinnia rehionalnoiu systemoiu osvity / L.M. Yurchuk // Theoretical and methodological principles and mechanisms of state and public administration of the regional education system), Balyuk I.B., Vinnytsia, 398 р. [in Ukrainian]. Isue DOI: 10.34132/pard 616 Проблемно-цільові засади стратегічного планування розвитку системи освіти регіону Відомості про авторів / Information about the Authors ORCID. ORG/ 0000-0002-9216-8876 Е-mail: maiya.hitko@gmail.com 617
W3036341147.txt	http://www.scielo.br/pdf/cr/v50n8/1678-4596-cr-50-08-e20190903.pdf	en		Fungi on irrigated rice seeds produced in the pre-germinated system in the Alto Vale do Itajaí region, Santa Catarina state, Brazil	Ciência rural	2,020	cc-by	4,498	Ciência Rural, Santarice Maria, v.50:8, e20190903, 2020 system in the Alto Vale do Itajaí Fungi on irrigated seeds produced in the pre-germinated region, Santa Catarina state, Brazil. http://doi.org/10.1590/0103-8478cr20190903 1 ISSNe 1678-4596 crop protection Fungi on irrigated rice seeds produced in the pre-germinated system in the Alto Vale do Itajaí region, Santa Catarina state, Brazil Bruno Tabarelli Scheidt1* Juliano Berghetti1 Flávio Chupel Martins1 1 2 Ricardo Trezzi Casa Welliton Recalcatti Valdemir Rossarola2 3 José de Alencar Lemos Vieira Junior Programa de Pós-graduação em Produção Vegetal (PPGPV), Universidade do Estado de Santa Catarina (UDESC), 88520-000, Lages, SC, Brasil. E-mail: brunotabarelli.s@hotmail.com. *Corresponding author. 2 Universidade do Estado de Santa Catarina (UDESC), Lages, SC, Brasil. 3 Universidade do Oeste de Santa Catarina (UNOESC), Campos Novos, SC, Brasil. 1 ABSTRACT: The aim of this research was to identify and quantify fungi infecting irrigated rice seeds produced in the 2015/16, 2016/17 and 2017/18 harvests in the pre-germinated system in the Alto Vale do Itajaí Region, Santa Catarina State, Brazil. A total of 479 lots of eight cultivars were analyzed. Seeds were disinfected and sown in potato-sucrose-agar culture medium with incubation for seven days at 25ºC ± 2ºC and photoperiod of 12 hours. The main fungus detected in the three crops season was Microdochium albescens with 54.9% of average incidence, followed by Alternaria padwikii (7.7%) and Bipolaris oryzae (3.3%). Key words: Microdochium albescens, Oryza sativa, seed health. Fungos em sementes de arroz irrigado produzidas no sistema pré-germinado na região do Alto Vale do Itajaí, estado de Santa Catarina, Brasil RESUMO: O objetivo deste trabalho foi identificar e quantificar fungos infectando sementes de arroz irrigado, produzidas nas safras 2015/16, 2016/17 e 2017/18, no sistema pré-germinado na região do Alto Vale do Itajaí, estado de Santa Catarina. Foram analisados 479 lotes de oito cultivares, sendo as sementes desinfestadas e semeadas em meio de cultura de batata-sacarose-ágar com incubação por sete dias a 25ºC ± 2 ºC e fotoperíodo de 12 horas. O principal fungo detectado nas três safras foi Microdochium albescens com 54,9% de incidência média, seguido de Alternaria padwikii (7,7%) e Bipolaris oryzae (3,3%). Palavras-chave: Microdochium albescens, Oryza sativa, sanidade de sementes. In the state of Santa Catarina (SC), irrigated rice (Oryza sativa L.) is produced in a pregerminated system widely sown in monoculture areas and in regions where long periods of wetness occur, factors that favor the occurrence of diseases (SOSBAI, 2018). In this crop system the pathogenic fungus inoculum is easily dispersed to the panicles with subsequent infection of grains or seeds, which can cause stains and their reduction (MIURA, 2002). Fungi Alternaria padwickii (Ganguly), Bipolaris oryzae (Haan’s Breda), Cercospora janseana (Racib.), Fusarium moniliforme (Sheldon), Microdochium albescens (Thüm.) syn. Microdochium oryzae; Gerlachia oryzae (Hashioka & Yokogi), Pyricularia oryzae (Cav.) and Sarocladium oryzae (Saw) some of the main pathogens associated with rice seed Received 11.12.19 (MEW & GONZALES, 2002; SOSBAI, 2018). Infected seeds may affect physiological quality and constitute an important agent of pathogen dissemination (OU, 1972). Rice seed production in SC focuses mainly on the pursuit of high physiological quality. There is no obvious concern for seed production with high health quality or low incidence of pathogens. The aim of this research was to identify and quantify fungi infecting seeds of irrigated rice cultivars produced in the pre-germinated system in the Alto Vale do Itajaí (AVI), SC, Brazil. Seed samples produced in the 2015/16, 2016/17 and 2017/18 crops season, benefited by Cooperative Cravil, were analyzed in a total of 479 lots of SCS121 CL (165 lots), SCS122 Miura (85), SCS116 Satoru cultivars. (79), EPAGRI 109 (58), Approved 04.20.20 Returned by the author 05.21.20 CR-2019-0903.R1 Ciência Rural, v.50, n.8, 2020. Scheidt et al. 2 SCSBRS Tio Taka (49), SCS118 Marquês (22), SCS117 CL (11) and Primoriso CL (10), from the AVI region, SC. Disease management in these crops followed the indications of the cooperative. For chemical control, a mixture of strobirulin + triazole + benzothiazole was sprayed at the stages of booting and flowering and / or grain formation (SOSBAI, 2018). In the cooperative the seeds were stored in polypropylene bags with humidity and room temperature. Seeds were collected and sent to the Plant Pathology Laboratory Santa Catarina State University (UDESC). Detection of fungi was done in PSA + A culture medium (Potato-Sucrose-Agar + Antibiotic = 200 mg L-1 streptomycin sulfate). Seeds were disinfected with sodium hypochlorite solution (1%) for two minutes and then washed with sterile water. For each lot, 200 seeds were analyzed, four replicates of 50, sown in acrylic Petri dishes and kept in growth chambers for seven days at 25 ºC and 12 hours photoperiod. For detection of P. oryzae the paper substrate incubation method (blotter test) was used. Six lots were randomly selected from each cultivar, totaling 48 lots. The seeds were arranged in a gerbox acrylic box containing two layers of moistened filter paper and incubated in a continuous light growth chamber at 27 ºC. The evaluation was performed at four and seven days. A completely randomized experimental design was used for both detection methods. Were considered infected in which it was possible to identify the colony and / or structures of the fungi, confirming the identification with microscope (BARNETT & HUNTER, 1998). The main pathogenic fungus associated with seeds were M. albescens with 54.9% of the average incidence and 100% of the prevalence, respectively followed by A. padwikii (7.7% / 90.6%), B. oryzae (3.3% / 59.9%), Curvularia sp. (3.1% / 66.5%) and S. oryzae (2.7% / 49.4%) (Table 1 and Table 1 - Average incidence of fungi associated with seeds of irrigated rice cultivars produced in the Alto Vale do Itajaí Region, Santa Catarina State. Lages, 2019. Cultivar Nº Lotes ---------------------------------------------Incidence of fungi (%)-------------------------------------------- Mic Alt Cur Bip Sar Nig Pen Asp Altsp Cer Fus -----------------------------------------------------------------------Safra 2015 / 2016------------------------------------------------------------------------SCS121 CL 57 61.9 6.0 6.6 7.3 2.2 0.8 1.1 0.7 0.1 0.2 0.0 SCS116 Satoru 33 72.3 2.9 1.7 1.6 2.2 1.6 1.3 1.2 0.4 0.3 0.0 SCSBRS Tio Taka 17 55.6 7.7 5.7 7.3 2.7 3.0 0.5 0.7 0.9 0.4 0.4 Epagri 109 11 68.1 2.0 2.0 2.5 3.9 1.2 2.1 2.3 1.0 0.0 0.0 SCS117 CL 11 60.5 9.0 3.9 5.1 2.5 1.4 0.4 0.3 0.5 0.4 0.0 SCS118 Marquês 10 57.2 7.6 9.6 3.3 3.7 0.7 1.8 1.8 0.0 0.2 0.0 Average (%) 62.6 5.8 4.9 4.5 2.8 1.4 1.2 1.2 0.5 0.3 0.1 ----------------------------------------------------------------------Safra 2016 / 2017-------------------------------------------------------------------------SCS121 CL 55 530 7.4 3.0 2.7 4.2 0.9 0.6 1.0 0.7 1.2 0.1 SCS116 Satoru 33 506 9.1 2.4 2.8 2.7 0.7 1.3 1.9 0.9 0.8 0.2 Epagri 109 31 455 11.3 3.0 3.2 3.9 0.6 0.8 1.3 1.5 0.6 0.3 SCSBRS Tio Taka 26 627 6.5 2.5 3.4 3.4 0.7 0.5 0.7 1.0 0.5 0.3 SCS122 Miura 12 431 11.7 5.6 5.1 7.1 0.0 1.2 2.1 2.3 1.1 0.0 SCS118 Marquês 12 453 9.3 3.0 3.3 4.8 1.6 0.4 0.8 1.6 1.1 0.0 Average (%) 500 9.2 3.2 3.4 4.3 0.7 0.8 1.3 1.3 0.9 0.2 ----------------------------------------------------------------------Safra 2017 / 2018-------------------------------------------------------------------------SCS122 Miura 73 535 7.2 1.7 1.6 1.1 0.6 0.1 0.0 0.3 1.2 0.0 SCS121 CL 53 516 5.9 0.8 3.3 0.5 0.8 0.1 0.2 0.4 1.0 0.1 Epagri 109 16 446 14.3 1.7 1.9 0.3 0.7 0.1 0.0 0.8 1.2 0.0 SCS116 Satoru 13 489 8.9 1.2 2.6 0.3 0.0 0.2 0.4 0.7 2.5 0.1 Primoriso CL 10 577 7.2 1.0 0.9 0.9 0.5 0.2 0.0 0.4 2.8 0.0 SCSBRS Tio Taka 6 562 4.5 0.7 2.5 2.2 0.3 0.3 0.0 0.3 1.2 0.0 Average (%) 521 8.0 1.2 2.1 0.9 0.5 0.2 0.1 0.5 1.6 0.0 Overall Average (%) 550 7.7 3.1 3.3 2.7 0.9 0.7 0.9 0.8 0.9 0.1 Mic - Microdochium albescens; Alt - Alternaria padwiicki; Cur - Curvularia sp.; Bip - Bipolaris oryzae; Sar - Sarocladium oryzae; Nig Nigrospora oryza; Pen - Penicillium spp.; Asp - Aspergillus flavus; Altsp - Alternaria sp.; Cer - Cercospora janseana e Fus - Fusarium sp. Ciência Rural, v.50, n.8, 2020. Fungi on irrigated rice seeds produced in the pre-germinated system in the Alto Vale do Itajaí region, Santa Catarina state, Brazil. table 2). Similar results were reported in irrigated rice cultivars in 350 samples analyzed in Rio Grande do Sul State (RS), from 1993 to 1998, also detecting G. oryzae (18.0%), Alternaria sp. (6.3%), C. lunata (4.9%) and B. oryzae (2.6%) (FRANCO et al., 2001). In another study also in RS, in 162 plots of rice from the 2005/06 crop season, Alternaria sp. (9.6%), Bipolaris sp. (9.3%), Gerlachia sp. (4.9%) and Curvularia sp. (3.8%) (FARIAS et al., 2007). The fungus M. albescens was the only with 100% prevalence (Table 2). This pathogen causes scalding in rice leaves, a growing disease in crop in the AVI, SC. The fungus was reported in RS association with seeds ranging from 6 to 31% (average 18%) (FRANCO et al., 2001) and 1 to 33% (average 4.9%) (FARIAS et al., 2007), but with a lower incidence than that obtained in this study (range 43.1 to 72.3% and average 54.9%). This may be related 3 to the detection method, reaction of rice genotypes and cropping systems. MALAVOLTA (2007), using filter paper method detected an average incidence of M. oryzae of 24.6 and 29.8%, respectively in 2003 and 2004, in rice genotype seeds in the of São Paulo (SP) state. PSA medium is more efficient in detecting M. albescens when compared to the filter paper method (GUTIÉRREZ et al. 2009). Analyzing genotypes, a higher incidence was observed in SCS116 Satoru (72.3% in 2015/16 crop) and lower in SCS122 Miura (43.1% in 2016/17 crop). These values are considered high for a possible obtaining of healthy seed for pathogen management, since there is no information on resistance reaction of cultivars for M. albescens in the states of SC and RS (SOSBAI, 2018). The fungus A. padwickii, included in the group of grain spotters, obtained average incidence in the three harvests of 7.7% and prevalence of Table 2 - Average prevalence of fungi associated with seeds of irrigated rice cultivars produced in the Alto Vale do Itajaí Region, Santa Catarina State. Lages, 2019. Cultivar Nº Lotes --------------------------------------------Prevalence of fungi (%)-------------------------------------------- Mic Alt Cur Bip Sar Nig Pen Asp Altsp Cer Fus ----------------------------------------------------------------------------Safra 2015 / 2016--------------------------------------------------------------------SCS121 CL 57 100 84.7 66.7 39.4 63.2 36.4 33.3 48.5 21.2 9.1 0.0 SCS116 Satoru 33 100 100 70.6 76.5 57.6 64.7 29.4 23.5 17.7 17.7 0.0 SCSBRS Tio Taka 17 100 100 90.9 45.5 64.7 54.6 36.4 9.1 18.2 18.2 0.4 Epagri 109 11 100 36.4 45.5 54.6 81.8 27.3 54.6 63.6 27.3 9.1 0.0 SCS117 CL 11 100 89.5 89.5 77.2 72.7 26.3 45.6 22.8 8.8 14.0 0.0 SCS118 Marquês 10 100 90.0 100 50.0 70.0 40.0 50.0 40.0 0.0 100 0.0 Average (%) 100 83.4 77.2 57.2 68.3 41.6 41.6 34.6 15.5 13.0 0.1 -------------------------------------------------------------------------Safra 2016 / 2017-----------------------------------------------------------------------SCS121 CL 55 100 89.1 80.0 63.6 63.6 23.6 27.3 36.4 32.7 40.0 3.6 SCS116 Satoru 33 100 90.9 69.7 60.6 45.5 21.2 48.5 42.4 36.4 21.2 6.1 Epagri 109 31 100 92.3 69.2 53.8 38.5 19.2 21.9 19.2 46.2 34.6 7.7 SCSBRS Tio Taka 26 100 96.8 77.4 67.7 58.1 16.1 32.3 45.2 54.8 32.3 9.7 SCS122 Miura 12 100 100 91.7 100 91.7 0.0 58.3 50.0 75.0 33.3 16.7 SCS118 Marquês 12 100 100 91.7 75.0 50.0 50.0 16.2 25.0 50.0 25.0 16.7 Average (%) 100 94.8 79.9 70.1 57.9 21.7 34.1 36.4 49.2 31.1 10.1 ------------------------------------------------------------------------Safra 2017 / 2018------------------------------------------------------------------------SCS122 Miura 73 100 868 26.4 66.0 18.9 22.6 5.7 7.6 18.9 32.1 5.7 SCS121 CL 53 100 100 53.1 69.2 15.4 23.1 15.4 15.4 30.8 53.8 0.0 Epagri 109 16 100 100 16.6 50.0 0.0 16.7 16.7 0.0 16.7 50.0 0.0 SCS116 Satoru 13 100 937 50.0 56.3 12.5 18.8 6.3 0.0 31.3 50.0 0.0 Primoriso CL 10 100 918 57.5 45.2 34.3 12.3 6.9 1.4 10.9 49.3 2.7 SCSBRS Tio Taka 6 100 90.0 50.0 30.0 50.0 0.0 10.0 0.0 20.0 90.0 0.0 Average (%) 100 93.7 42.3 52.8 21.9 15.6 10.2 4.1 21.4 54.2 1.4 Overall Average (%) 100 90.6 66.5 59.9 49.4 26.3 28.6 25.0 28.7 32.8 11.6 Mic - Microdochium albescens; Alt - Alternaria padwiicki; Cur - Curvularia sp.; Bip - Bipolaris oryzae; Sar - Sarocladium oryzae; Nig Nigrospora oryza; Pen - Penicillium spp.; Asp - Aspergillus flavus; Altsp - Alternaria sp.; Cer - Cercospora janseana e Fus - Fusarium sp. Ciência Rural, v.50, n.8, 2020. Scheidt et al. 4 90.6%. In Egypt, India, Korea, Nepal and Thailand, the fungus was detected in 282 lots of 388 analyzed (MATHUR et al., 1972). In Brazil, in SP in 2003 and 2004, the average incidence of A. padwickii was 8.2% and 1.6%; respectively, in cultivation of genotypes in flood irrigated system (MALAVOLTA, 2007). In RS, FRANCO et al. (2001), detected values from 0.4% to 20.6%, but do not describe the Alternaria species. The fungus B. oryzae presented higher average incidence in cultivar SCS121 CL and SCSBRS Tio Taka (7.3%, 2015/16 crop), and lower average value in cultivar Primoriso CL (0.9%, 2017/18 crop). For this fungus, we highlighted a prevalence of 100% detected in the 2016/17 crop in the cultivar SCS122 Miura. In SP was detected an average incidence of 13.6%, higher than that found in this study (MALAVOLTA, 2007). The fungus Curvularia sp. was detected with incidence and prevalence values similar to B. oryzae. It was not possible to identify the species of Curvularia since no morphological and molecular characterization was performed. The species C. eragrostidis (P. Henn) and C. lunata (Wakker) Boedijn, are detected in rice seeds (LIMA & FURTADO, 2007). Other fungi were detected with lower average incidence, such as: S. oryzae (2.7%), Nigrospora oryzae (0.9%), Penicillium spp. (0.7%), Aspergillus flavus (0.9%), Alternaria sp. (0.8%), C. janseana (0.9%) and Fusarium sp. (0.1%) (Table 1). These are also reported in seeds in RS (FRANCO et al., 2001; FARIAS et al., 2007). The fungus P. oryzae was not detected by the tested. In RS P. oryzae was detected with low incidence (0.04%) in 350 seed samples (FRANCO et al., 2001), but later was not detected in 162 samples analyzed (FARIAS et al., 2007). This was the first survey of fungi associated with irrigated rice seeds in the pre-germinated cultivation system carried out in Santa Catarina State, serving as a basis for future research related to the epidemiological importance of infected seeds. ACKNOWLEDGEMENTS The reserach was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance code 001. DECLARATION INTERESTS OF CONFLICT OF The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results. AUTHORS’ CONTRIBUTIONS BTS and RTC conceived and designed experiments. BTS, VR and FCM performed the experiments, BTS carried out the lab analyses. BTS, JALVJ e JB performed statistical analyses of experimental data. BTS, JB, RTC, FCM, WR, VR and JALVJ prepared the draft of the manuscript. All authors critically revised the manuscript and approved of the final version. REFERENCES BARNETT, H. L.; HUNTER, B. B. Illustred genera of imperfect fungi. The American Phytopathological Society. St. Paul. 1998. 218p. FARIAS, J. R. C. et al. Incidência de fungos associados a sementes de arroz em seis regiões produtoras do Rio Grande do Sul. Revista Brasileira de Agrociência, v.14, p.487-490, 2007. Available from: <https://periodicos.ufpel.edu.br/ojs2/index.php/CAST/article/ viewFile/1395/1381>. Accessed: Jan, 30, 2020. FRANCO, D. F. et al. Fungos associados a sementes de arroz irrigado no Rio Grande do Sul. Revista Brasileira de Agrociência, v.7, p.235-236, 2001. Available from: <http://www.ufpel.tche.br/ faem/agrociencia/v7n3/artigo16.pdf>. Accessed: Jan. 30, 2020. GUTIÉRREZ S. A. et al. Estudio preliminar sobre métodos de detección de Microdochium oryzae em semillas de arroz. Tropical Plant Pathology, v.34, p.042-044, 2009. Available from <http://www.scielo.br/scielo.php?script=sci_ arttext&pid=S198256762009000100007&lng=en&nrm=iso>. Accessed: Feb. 06, 2020. doi: 10.1590/S1982-56762009000100007. LIMA, A. B.; FURTADO, M. Curvularia species (anamorphic fungi: Hyphomycetes) from Santiago island, Cape Vert. Portugaliae Acta Biologica, v.22, p.145-156, 2007. Available from: <https:// www.cabdirect.org/cabdirect/abstract/20083133955>. Accessed: Feb. 23, 2020. MALAVOLTA, V. M. A. et al. Incidência de fungos e quantificação de danos em sementes de genótipos de arroz. Summa Phytopathologica, v.33, p.280-286, 2007. Available from: <https://www.scielo.br/scielo.php?pid=S0100540520070003000 12&script=sci_abstract&tlng=pt>. Accessed: Feb. 25, 2020. doi: 10.1590/S0100-54052007000300012. MATHUR, S. B. et al. Seed-borne infection of Trichoconis padwickii in rice, distribution, and damage to seeds and seedlings. International Seed Testing Association, v.37, p.803-810, 1972. MEW, T. W.; GONZALES, P. A handbook of rice seedborne fungi. Los Baños, (Philippines): International Rice Research Institute. Science Publishers, Inc. 2002, 83p. MIURA, L. Doenças. In: Epagri Arroz irrigado: Sistema prégerminado. Florianóplis, Epagri/ GMC, 2002, 203-227p. OU, S. H. Rice diseases. Surrey: Commonwealth Mycological Institute, 1972. 368p. SOSBAI. Arroz irrigado: recomendações técnicas da pesquisa para o Sul do Brasil. In: XXXII Reunião Técnica da Cultura do Arroz Irrigado, 2018, Farroupilha, RS. Cachoeirinha: Sociedade Sul-Brasileira de Arroz Irrigado, 2018, 205p. Available from: <http://www.sosbai.com.br/docs/Boletim_RT_2016.pdf>. Accessed: Jan. 22, 2020. Ciência Rural, v.50, n.8, 2020. Ciência Rural, Santa Maria, v.51:04, 2021 10.1590/01038478crerr20190903 e10903 Erratum Erratum In the article "Fungi on irrigated rice seeds produced in the pre-germinated system in the Alto Vale do Itajaí region, Santa Catarina state, Brazil" published in Ciência Rural, volume 50, number 8, DOI http://doi.org/10.1590/0103-8478cr20190903. In the table 1, where we read: Table 1 - Average incidence of fungi associated with seeds of irrigated rice cultivars produced in the Alto Vale do Itajaí Region, Santa Catarina State. Lages, 2019. Cultivar Nº Lotes ---------------------------------------------Incidence of fungi (%)-------------------------------------------- Mic Alt Cur Bip Sar Nig Pen Asp Altsp Cer Fus -----------------------------------------------------------------------Safra 2015 / 2016------------------------------------------------------------------------SCS121 CL 57 61.9 6.0 6.6 7.3 2.2 0.8 1.1 0.7 0.1 0.2 0.0 SCS116 Satoru 33 72.3 2.9 1.7 1.6 2.2 1.6 1.3 1.2 0.4 0.3 0.0 SCSBRS Tio Taka 17 55.6 7.7 5.7 7.3 2.7 3.0 0.5 0.7 0.9 0.4 0.4 Epagri 109 11 68.1 2.0 2.0 2.5 3.9 1.2 2.1 2.3 1.0 0.0 0.0 SCS117 CL 11 60.5 9.0 3.9 5.1 2.5 1.4 0.4 0.3 0.5 0.4 0.0 SCS118 Marquês 10 57.2 7.6 9.6 3.3 3.7 0.7 1.8 1.8 0.0 0.2 0.0 Average (%) 62.6 5.8 4.9 4.5 2.8 1.4 1.2 1.2 0.5 0.3 0.1 ----------------------------------------------------------------------Safra 2016 / 2017-------------------------------------------------------------------------SCS121 CL 55 530 7.4 3.0 2.7 4.2 0.9 0.6 1.0 0.7 1.2 0.1 SCS116 Satoru 33 506 9.1 2.4 2.8 2.7 0.7 1.3 1.9 0.9 0.8 0.2 Epagri 109 31 455 11.3 3.0 3.2 3.9 0.6 0.8 1.3 1.5 0.6 0.3 SCSBRS Tio Taka 26 627 6.5 2.5 3.4 3.4 0.7 0.5 0.7 1.0 0.5 0.3 SCS122 Miura 12 431 11.7 5.6 5.1 7.1 0.0 1.2 2.1 2.3 1.1 0.0 SCS118 Marquês 12 453 9.3 3.0 3.3 4.8 1.6 0.4 0.8 1.6 1.1 0.0 Average (%) 500 9.2 3.2 3.4 4.3 0.7 0.8 1.3 1.3 0.9 0.2 ----------------------------------------------------------------------Safra 2017 / 2018-------------------------------------------------------------------------SCS122 Miura 73 535 7.2 1.7 1.6 1.1 0.6 0.1 0.0 0.3 1.2 0.0 SCS121 CL 53 516 5.9 0.8 3.3 0.5 0.8 0.1 0.2 0.4 1.0 0.1 Epagri 109 16 446 14.3 1.7 1.9 0.3 0.7 0.1 0.0 0.8 1.2 0.0 SCS116 Satoru 13 489 8.9 1.2 2.6 0.3 0.0 0.2 0.4 0.7 2.5 0.1 Primoriso CL 10 577 7.2 1.0 0.9 0.9 0.5 0.2 0.0 0.4 2.8 0.0 SCSBRS Tio Taka 6 562 4.5 0.7 2.5 2.2 0.3 0.3 0.0 0.3 1.2 0.0 Average (%) 521 8.0 1.2 2.1 0.9 0.5 0.2 0.1 0.5 1.6 0.0 Overall Average (%) 550 7.7 3.1 3.3 2.7 0.9 0.7 0.9 0.8 0.9 0.1 Mic - Microdochium albescens; Alt - Alternaria padwiicki; Cur - Curvularia sp.; Bip - Bipolaris oryzae; Sar - Sarocladium oryzae; Nig - Nigrospora oryza; Pen - Penicillium spp.; Asp - Aspergillus flavus; Altsp - Alternaria sp.; Cer - Cercospora janseana e Fus Fusarium sp. Read: Table 1 - Average incidence of fungi associated with seeds of irrigated rice cultivars produced in the Alto Vale do Itajaí Region, Santa Catarina State. Lages, 2019. Cultivar Nº Lotes ---------------------------------------------Incidence of fungi (%)-------------------------------------------- Mic Alt Cur Bip Sar Nig Pen Asp Altsp Cer Fus -----------------------------------------------------------------------Safra 2015 / 2016------------------------------------------------------------------------SCS121 CL 57 61.9 6.0 6.6 7.3 2.2 0.8 1.1 0.7 0.1 0.2 0.0 SCS116 Satoru 33 72.3 2.9 1.7 1.6 2.2 1.6 1.3 1.2 0.4 0.3 0.0 SCSBRS Tio Taka 17 55.6 7.7 5.7 7.3 2.7 3.0 0.5 0.7 0.9 0.4 0.4 Epagri 109 11 68.1 2.0 2.0 2.5 3.9 1.2 2.1 2.3 1.0 0.0 0.0 SCS117 CL 11 60.5 9.0 3.9 5.1 2.5 1.4 0.4 0.3 0.5 0.4 0.0 SCS118 Marquês 10 57.2 7.6 9.6 3.3 3.7 0.7 1.8 1.8 0.0 0.2 0.0 Average (%) 62.6 5.8 4.9 4.5 2.8 1.4 1.2 1.2 0.5 0.3 0.1 ----------------------------------------------------------------------Safra 2016 / 2017-------------------------------------------------------------------------SCS121 CL 55 53.0 7.4 3.0 2.7 4.2 0.9 0.6 1.0 0.7 1.2 0.1 SCS116 Satoru 33 50.6 9.1 2.4 2.8 2.7 0.7 1.3 1.9 0.9 0.8 0.2 Epagri 109 31 45.5 11.3 3.0 3.2 3.9 0.6 0.8 1.3 1.5 0.6 0.3 SCSBRS Tio Taka 26 62.7 6.5 2.5 3.4 3.4 0.7 0.5 0.7 1.0 0.5 0.3 SCS122 Miura 12 43.1 11.7 5.6 5.1 7.1 0.0 1.2 2.1 2.3 1.1 0.0 SCS118 Marquês 12 45.3 9.3 3.0 3.3 4.8 1.6 0.4 0.8 1.6 1.1 0.0 Average (%) 50.0 9.2 3.2 3.4 4.3 0.7 0.8 1.3 1.3 0.9 0.2 ----------------------------------------------------------------------Safra 2017 / 2018-------------------------------------------------------------------------SCS122 Miura 73 53.5 7.2 1.7 1.6 1.1 0.6 0.1 0.0 0.3 1.2 0.0 SCS121 CL 53 51.6 5.9 0.8 3.3 0.5 0.8 0.1 0.2 0.4 1.0 0.1 Epagri 109 16 44.6 14.3 1.7 1.9 0.3 0.7 0.1 0.0 0.8 1.2 0.0 SCS116 Satoru 13 48.9 8.9 1.2 2.6 0.3 0.0 0.2 0.4 0.7 2.5 0.1 Primoriso CL 10 57.7 7.2 1.0 0.9 0.9 0.5 0.2 0.0 0.4 2.8 0.0 SCSBRS Tio Taka 6 56.2 4.5 0.7 2.5 2.2 0.3 0.3 0.0 0.3 1.2 0.0 Average (%) 52.1 8.0 1.2 2.1 0.9 0.5 0.2 0.1 0.5 1.6 0.0 Overall Average (%) 55.0 7.7 3.1 3.3 2.7 0.9 0.7 0.9 0.8 0.9 0.1 Mic - Microdochium albescens; Alt - Alternaria padwiicki; Cur - Curvularia sp.; Bip - Bipolaris oryzae; Sar - Sarocladium oryzae; Nig - Nigrospora oryza; Pen - Penicillium spp.; Asp - Aspergillus flavus; Altsp - Alternaria sp.; Cer - Cercospora janseana e Fus Fusarium sp. In the table 2, where we read: Table 2 - Average prevalence of fungi associated with seeds of irrigated rice cultivars produced in the Alto Vale do Itajaí Region, Santa Catarina State. Lages, 2019. Cultivar Nº Lotes --------------------------------------------Prevalence of fungi (%)-------------------------------------------- Mic Alt Cur Bip Sar Nig Pen Asp Altsp Cer Fus ----------------------------------------------------------------------------Safra 2015 / 2016--------------------------------------------------------------------SCS121 CL 57 100 84.7 66.7 39.4 63.2 36.4 33.3 48.5 21.2 9.1 0.0 SCS116 Satoru 33 100 100 70.6 76.5 57.6 64.7 29.4 23.5 17.7 17.7 0.0 SCSBRS Tio Taka 17 100 100 90.9 45.5 64.7 54.6 36.4 9.1 18.2 18.2 0.4 Epagri 109 11 100 36.4 45.5 54.6 81.8 27.3 54.6 63.6 27.3 9.1 0.0 SCS117 CL 11 100 89.5 89.5 77.2 72.7 26.3 45.6 22.8 8.8 14.0 0.0 SCS118 Marquês 10 100 90.0 100 50.0 70.0 40.0 50.0 40.0 0.0 100 0.0 Average (%) 100 83.4 77.2 57.2 68.3 41.6 41.6 34.6 15.5 13.0 0.1 -------------------------------------------------------------------------Safra 2016 / 2017-----------------------------------------------------------------------SCS121 CL 55 100 89.1 80.0 63.6 63.6 23.6 27.3 36.4 32.7 40.0 3.6 SCS116 Satoru 33 100 90.9 69.7 60.6 45.5 21.2 48.5 42.4 36.4 21.2 6.1 Epagri 109 31 100 92.3 69.2 53.8 38.5 19.2 21.9 19.2 46.2 34.6 7.7 SCSBRS Tio Taka 26 100 96.8 77.4 67.7 58.1 16.1 32.3 45.2 54.8 32.3 9.7 SCS122 Miura 12 100 100 91.7 100 91.7 0.0 58.3 50.0 75.0 33.3 16.7 SCS118 Marquês 12 100 100 91.7 75.0 50.0 50.0 16.2 25.0 50.0 25.0 16.7 Average (%) 100 94.8 79.9 70.1 57.9 21.7 34.1 36.4 49.2 31.1 10.1 ------------------------------------------------------------------------Safra 2017 / 2018------------------------------------------------------------------------SCS122 Miura 73 100 868 26.4 66.0 18.9 22.6 5.7 7.6 18.9 32.1 5.7 SCS121 CL 53 100 100 53.1 69.2 15.4 23.1 15.4 15.4 30.8 53.8 0.0 Epagri 109 16 100 100 16.6 50.0 0.0 16.7 16.7 0.0 16.7 50.0 0.0 SCS116 Satoru 13 100 937 50.0 56.3 12.5 18.8 6.3 0.0 31.3 50.0 0.0 Primoriso CL 10 100 918 57.5 45.2 34.3 12.3 6.9 1.4 10.9 49.3 2.7 SCSBRS Tio Taka 6 100 90.0 50.0 30.0 50.0 0.0 10.0 0.0 20.0 90.0 0.0 Average (%) 100 93.7 42.3 52.8 21.9 15.6 10.2 4.1 21.4 54.2 1.4 Overall Average (%) 100 90.6 66.5 59.9 49.4 26.3 28.6 25.0 28.7 32.8 11.6 Mic - Microdochium albescens; Alt - Alternaria padwiicki; Cur - Curvularia sp.; Bip - Bipolaris oryzae; Sar - Sarocladium oryzae; Nig - Nigrospora oryza; Pen - Penicillium spp.; Asp - Aspergillus flavus; Altsp - Alternaria sp.; Cer - Cercospora janseana e Fus Fusarium sp. Read: Table 2 - Average prevalence of fungi associated with seeds of irrigated rice cultivars produced in the Alto Vale do Itajaí Region, Santa Catarina State. Lages, 2019. Cultivar Nº Lotes --------------------------------------------Prevalence of fungi (%)-------------------------------------------- Mic Alt Cur Bip Sar Nig Pen Asp Altsp Cer Fus ----------------------------------------------------------------------------Safra 2015 / 2016--------------------------------------------------------------------SCS121 CL 57 100 84.7 66.7 39.4 63.2 36.4 33.3 48.5 21.2 9.1 0.0 SCS116 Satoru 33 100 100 70.6 76.5 57.6 64.7 29.4 23.5 17.7 17.7 0.0 SCSBRS Tio Taka 17 100 100 90.9 45.5 64.7 54.6 36.4 9.1 18.2 18.2 0.4 Epagri 109 11 100 36.4 45.5 54.6 81.8 27.3 54.6 63.6 27.3 9.1 0.0 SCS117 CL 11 100 89.5 89.5 77.2 72.7 26.3 45.6 22.8 8.8 14.0 0.0 SCS118 Marquês 10 100 90.0 100 50.0 70.0 40.0 50.0 40.0 0.0 100 0.0 Average (%) 100 83.4 77.2 57.2 68.3 41.6 41.6 34.6 15.5 13.0 0.1 -------------------------------------------------------------------------Safra 2016 / 2017-----------------------------------------------------------------------SCS121 CL 55 100 89.1 80.0 63.6 63.6 23.6 27.3 36.4 32.7 40.0 3.6 SCS116 Satoru 33 100 90.9 69.7 60.6 45.5 21.2 48.5 42.4 36.4 21.2 6.1 Epagri 109 31 100 92.3 69.2 53.8 38.5 19.2 21.9 19.2 46.2 34.6 7.7 SCSBRS Tio Taka 26 100 96.8 77.4 67.7 58.1 16.1 32.3 45.2 54.8 32.3 9.7 SCS122 Miura 12 100 100 91.7 100 91.7 0.0 58.3 50.0 75.0 33.3 16.7 SCS118 Marquês 12 100 100 91.7 75.0 50.0 50.0 16.2 25.0 50.0 25.0 16.7 Average (%) 100 94.8 79.9 70.1 57.9 21.7 34.1 36.4 49.2 31.1 10.1 ------------------------------------------------------------------------Safra 2017 / 2018------------------------------------------------------------------------SCS122 Miura 73 100 86.8 26.4 66.0 18.9 22.6 5.7 7.6 18.9 32.1 5.7 SCS121 CL 53 100 100 53.1 69.2 15.4 23.1 15.4 15.4 30.8 53.8 0.0 Epagri 109 16 100 100 16.6 50.0 0.0 16.7 16.7 0.0 16.7 50.0 0.0 SCS116 Satoru 13 100 93.7 50.0 56.3 12.5 18.8 6.3 0.0 31.3 50.0 0.0 Primoriso CL 10 100 91.8 57.5 45.2 34.3 12.3 6.9 1.4 10.9 49.3 2.7 SCSBRS Tio Taka 6 100 90.0 50.0 30.0 50.0 0.0 10.0 0.0 20.0 90.0 0.0 Average (%) 100 93.7 42.3 52.8 21.9 15.6 10.2 4.1 21.4 54.2 1.4 Overall Average (%) 100 90.6 66.5 59.9 49.4 26.3 28.6 25.0 28.7 32.8 11.6 Mic - Microdochium albescens; Alt - Alternaria padwiicki; Cur - Curvularia sp.; Bip - Bipolaris oryzae; Sar - Sarocladium oryzae; Nig - Nigrospora oryza; Pen - Penicillium spp.; Asp - Aspergillus flavus; Altsp - Alternaria sp.; Cer - Cercospora janseana e Fus Fusarium sp.
https://openalex.org/W2923275624	https://digibug.ugr.es/bitstream/10481/59397/1/08669762.pdf	English	null	A Probabilistic Algorithm for Predictive Control With Full-Complexity Models in Non-Residential Buildings	IEEE access	2,019	cc-by	15,253	2169-3536 2019 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. SPECIAL SECTION ON CYBER-PHYSICAL SYSTEMS Received February 19, 2019, accepted March 15, 2019, date of publication March 19, 2019, date of current version April 5, 2019. Received February 19, 2019, accepted March 15, 2019, date of publication March 19, 2019, date of current version April 5, 2019. Digital Object Identifier 10.1109/ACCESS.2019.2906311 AN GÓMEZ-ROMERO 1, CARLOS J. FERNÁNDEZ-BASSO1, M. VICTORIA CAMBRONERO IGUEL MOLINA-SOLANA 3, JESÚS R. CAMPAÑA1, M. DOLORES RUIZ1, ND MARIA J. MARTIN-BAUTISTA1 This work was supported in part by the Universidad de Granada under Grant P9-2014-ING, in part by the Spanish Ministry of Science, Innovation and Universities under Grant TIN2017-91223-EXP, in part by the Spanish Ministry of Economy and Competitiveness under Grant TIN2015-64776-C3-1-R, and in part by the European Union (Energy IN TIME EeB.NMP.2013-4), under Grant 608981. ABSTRACT Despite the increasing capabilities of information technologies for data acquisition and pro- cessing, building energy management systems still require manual conﬁguration and supervision to achieve optimal performance. Model predictive control (MPC) aims to leverage equipment control–particularly heating, ventilation, and air conditioning (HVAC)–by using a model of the building to capture its dynamic characteristics and to predict its response to alternative control scenarios. Usually, MPC approaches are based on simpliﬁed linear models, which support faster computation but also present some limitations regarding interpretability, solution diversiﬁcation, and longer-term optimization. In this paper, we propose a novel MPC algorithm that uses a full-complexity grey-box simulation model to optimize HVAC operation in non-residential buildings. Our system generates hundreds of candidate operation plans, typically for the next day, and evaluates them in terms of consumption and comfort by means of a parallel simulator conﬁgured according to the expected building conditions (weather and occupancy). The system has been implemented and tested in an ofﬁce building in Helsinki, both in a simulated environment and in the real building, yielding energy savings around 35% during the intermediate winter season and 20% in the whole winter season with respect to the current operation of the heating equipment. NDEX TERMS Model predictive control, simulation, control, building energy management system INDEX TERMS Model predictive control, simulation, control, building energy management sy Directive on the Energy Performance of Buildings addressing the target of a 30% increase of energy efﬁciency by 2030 [5]. The associate editor coordinating the review of this manuscript and approving it for publication was Mengchu Zhou. SPECIAL SECTION ON CYBER-PHYSICAL SYSTEMS I. INTRODUCTION ildi f Buildings account for more than one third of the worldwide primary energy consumption [1] and they are an equally important source of CO2 emissions [2]. In western countries, non-residential buildings consume between 30-40% of the energy, mostly during the operational stage and by the HVAC (heating, ventilation, and air conditioning) systems [3]. These ﬁgures are expected to increase in the future due to inefﬁ- ciency of aging infrastructures, impact of climate change in weather, and economic growth in China and India [4]. At the same time, technological advances offer great opportunities to achieve energy savings in new and old buildings. For the latter, the European Union issued in 2016 an update of the There are several complementary strategies to reduce energy consumption in existing buildings. Renovation works and retroﬁtting, making the most of affordable and clean sources, are essential, and to be effective, they must be accompanied by suitable operation protocols to optimize energy management [6]. As a matter of fact, selecting daily optimal setpoints for the HVAC equipment is estimated to lead to savings up to 35%, depending on the climate [7]. New approaches to building energy management systems (BEMS) offer interactive and real-time building monitoring and remote control, and provide support for simulation and optimization [8]–[10]. Still, a great deal of the decision-making is left to the operators, who must analyze available data, estimate energy demand, and propose control The associate editor coordinating the review of this manuscript and approving it for publication was Mengchu Zhou. 38748 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models technologies, which provide support for massive data man- agement for continuous model calibration, distributed execu- tion of simulation software, accurate prediction of building conditions, and remote operation. rules to be implemented in the BEMS. Common a priori control strategies include optimized start/stop of equipment, chiller and boiler optimization, adaptive control, and optimal energy sourcing [11]. In the last decade, several proposals for automating the generation of operational plans based on Model Predictive Control (MPC) have been presented [12], [13]. MPC uses a simulation model of the building to capture its dynamic characteristics and predict its response to alternative con- trol scenarios. It pursues a (conﬂicting) dual target: reduc- ing energy consumption thanks to pre-emptive control and anticipation of the building state while keeping users’ com- fort. I. INTRODUCTION ildi f By establishing a complete sequence of instructions for the building equipment –i.e. the (daily) operational plan–, it overcomes the limitations of homeostatic controllers, which cannot guarantee long-term optimal operation: the ahead time and the timespan of the control instructions can expand to several hours, leading to plans entailing more uncertainty – because of the use of forecasted building conditions (e.g. weather, occupancy)– and more complexity –because of the exponential increase of possible plans–, but also more efﬁ- cient –because of the exploitation of the inertial effects of HVAC equipment. The core of the system is the intelligent operational plan generator (OPG) module, an MPC-like control scheduler supported by a cloud-based extension of the IESVE2 simula- tion software. The OPG algorithm calculates an operational plan (OP) for a future period (typically the next day) after simulating hundreds of candidate plans under the forecasted state of the building (i.e. considering weather and occu- pancy estimations) in order to minimize energy consumption while guaranteeing occupants’ comfort. Eventually, the OP setpoints are automatically applied to the equipment with- out direct involvement of the operator. To the best of our knowledge, this is the ﬁrst proposal using an off-the-shell full-complexity model for predictive control. In this paper, we describe the OPG algorithm design, implementation, and evaluation in the Sanomatalo commer- cial building located in Helsinki (Finland). The control strate- gies for this building focus on optimizing the air supply temperature setpoint and the airﬂow volume setpoints. The main contributions of this research work are the following: MPC is formulated as a combinatorial optimization prob- lem, in which a search algorithm must ﬁnd the best actua- tion plan, in terms of thermal comfort and overall building consumption, in a solution space including all the possi- ble setpoint combinations for a given future period [14]. Nevertheless, most works tend to simplify the models (e.g. by reducing the model differential equations to linear com- binations) or to reduce the search space (e.g. by limit- ing the control to a small part of the building equipment, and by incorporating manually-extracted expert knowledge). This results in short-scope, limited-extensibility and low- performance solutions involving a great deal of manual work • The OPG algorithm, based on probabilistic search, directly provides operational plans for HVAC equipment including on/off and numerical setpoint values that are directly applied through the BEMS –no additional trans- lation from demand estimations into actions is needed. 1The Energy IN TIME project (Simulation-based control for energy efﬁ- ciency building operation and maintenance) was funded by the European Commission within the 7th Framework Programme in 2013-2017. See [66] for a brief description of the overall project results. 2https://www.iesve.com/VE2018 I. INTRODUCTION ildi f • We extend the control horizon compared to usual MPC approaches. The OPG considers setpoints up to a 1-day period, which ﬁts better to the usual building operation (e.g. the operator can validate control for the whole day) and offers more opportunities for longer-term energy saving policies. • We use a full-complexity simulation model out of the box, decoupled from the optimization algorithm and directly interpretable by experts and operators. The sim- ulation model self-recalibrates by using data directly measured from the building and runs on a cloud-based distributed version of IESVE. The departing hypothesis of our research work is that we can exploit the increasing capabilities of massive and parallel data processing technologies to run a large amount of simulations with full-complexity physical models and to assess multiple hypothetical control scenarios to obtain the appropriate setpoints in terms of efﬁciency and comfort. Availability of sensor data allows us to develop more accurate models, since data can be used for calibration, calculation of better predictions of relevant contextual factors (e.g. occu- pancy), and detection of control performance decline. At the same time, physical models are more interpretable and easier to extend; actually, we can use physical models and model development tools out of the box, such as TRNSYS, Energy- Plus or IESVE [15]. • We carry out an evaluation of the system in the simula- tion environment and in the real building; in the latter case, over a longer period of time than related works (30 days), in line with the recommendations in [16]. Comparison with the base control, performed according to the International Performance Measurement and Veriﬁcation Protocol (IPMVP) [17], yielded energy savings above 20%, with peaks above 40% at the end of the winter season. The remainder of the paper is organized as follows. Next, we describe several related works, most of them centered in the use of simpliﬁed simulation models. In Section III, we describe the pilot building, the simulation model, and the evaluation methodology. In Section IV, we detail the design of the OPG algorithm and its features. Section V presents the In the Energy IN TIME project,1 we developed an advanced BEMS for optimized HVAC operation in non- residential buildings. This BEMS is powered by Big Data 38749 38749 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models As an alternative to MILP and related techniques, Katsigarakis et al. II. RELATED WORK MPC was introduced by Mahdavi in 2001 [18], and was initially used ofﬂine to derive an optimized control law from sensor measurements and simulations, and to validate pre- deﬁned control strategies [19], [20]. Associated small-scale experiments, most of them carried out in the simulation envi- ronment, showed that the application of MPC can effectively accomplish a reduction in energy consumption [21]. Further studies characterized and performed a preliminary evaluation of HVAC-related energy management actions that can be exploited in MPC [22]: outside air economizer cycle, pro- grammed start and stop lead time, load reset, and occupied time adaptive control strategy. Additionally, other authors emphasized the need for considering subjective comfort mea- sures beyond indoor temperatures and humidity thresholds, such as predicted mean vote (PMV) [23]. In contrast, current MPC-powered BEMS are not limited to only apply a plan elicited from expert knowledge and conﬁrmed suitable after simulation. They can dynamically generate control instructions by searching an operational plan that, according to the simulation model, satisﬁes the expected energy demand while minimizes consumption. Nevertheless, the calculation of the ﬁtness of a plan by simulation is com- putationally expensive [24]. A subsequent problem of MPC is the accuracy of the simulation model, particularly if a simpliﬁed version is required [50]–[52], or if there is uncertainty in the expected building conditions; e.g., weather forecast and occupancy estimations [53]–[55]. In this regard, Kwak et al. proposed exploiting parallel co-simulation, which is the execution of several simulation models under different conditions to minimize the errors due to uncertainty in input data and unexpected occupancy variations. The authors implemented a general-purpose enthalpy controller that generated control signals starting 15 and 30 minutes later [56], and a daily controller [57]. For the combination of the simulation mod- els –in EnergyPlus and MATLAB–, they used the Building Controls Virtual Test Bed (BCVTB) suite. The system was tested during one day in severe weather conditions in a real building, showing energy savings around 2% in the best case. Bianchini et al. [25] addressed this issue by replacing the full model of the building by a simpliﬁed linear model. The linear model is afterwards solved by using different heuris- tics that reduce the search to a computable mixed integer linear programming (MILP) problem. II. RELATED WORK Although this solution considerably reduces the capability of the algorithm to ﬁnd unknown solutions, it proved to yield good results in a sim- ulation environment when tested for a delimited section of the building. Different proposals using linear and non-linear programming, having different degree of complexity, appli- cation scope, evaluation comprehensiveness and achieved energy savings, can be found in the literature, in particular for non-residential buildings [26]–[34]. 3https://sanoma.ﬁ/en/sanoma-house/ 4https://www.caverion.com/ I. INTRODUCTION ildi f [46] created a surrogate building model by applying Machine Learning techniques. This surrogate model is automatically learnt from pre-computed outcomes of the real model by using a regression technique (e.g. support vector machines), and optimization with it is signiﬁcantly faster than in MILP. Unfortunately, it can be inaccurate or unfeasible if the building state is difﬁcult to model; i.e. when the control scope is too broad, there are too many outputs to estimate, or the variables have complex interdependen- cies. Analogously, Casals et al. used Bayesian networks to simplify the simulation model of a subway station, obtaining good prediction accuracy [47]. Their system does not provide long-term operation plans –and consequently, it does not opti- mize HVAC operation–, yet it achieves considerable energy savings in ventilation and lighting systems –thanks to the use of sophisticated Computer Vision techniques for real-time occupancy estimation. Manjarres et al. trained a predictive black-box model using Random Forests that reproduces the daily behavior of the building and replaces the physical model of the building; however, the control strategies are limited to switching on and off the HVAC systems [48]. Kontes et al. created a surrogate model with support vector machines (SVM) to optimize radiator operation with similar promising results [49]. experimental setup and the results obtained in the simulation environment and in the real building compared to the baseline operation. In Section VI we discuss the contributions of our proposal in terms of energy savings and comfort achievement, as well as possible improvements to the system. Finally, we summarize the conclusions of the work and introduce prospective directions for future research. II. RELATED WORK B. SIMULATION MODEL AND CALIBRATION The accuracy of the simulation model is a crucial aspect of MPC approaches to avoid the generation of control instructions under wrong assumptions [58]–[60]. To this aim, control-oriented models must effectively catch all the inter- actions between HVAC equipment (radiators, heat pumps, etc.) [61]. This is however a difﬁcult and costly process [62]. Grey-box models have showed good performance and cost-beneﬁt ratio [63], [64], even with relatively simple for- mulations and few input variables [65]. This kind of models rely on the existing corpus of expert knowledge to model ther- mal behaviour by using differential equations encoding the physical principles of mass, energy and momentum transfer; and they apply statistical models to tune model outputs based on historical and live data. The accuracy of the simulation model is a crucial aspect of MPC approaches to avoid the generation of control instructions under wrong assumptions [58]–[60]. To this aim, control-oriented models must effectively catch all the inter- actions between HVAC equipment (radiators, heat pumps, etc.) [61]. This is however a difﬁcult and costly process [62]. Grey-box models have showed good performance and cost-beneﬁt ratio [63], [64], even with relatively simple for- mulations and few input variables [65]. This kind of models rely on the existing corpus of expert knowledge to model ther- mal behaviour by using differential equations encoding the physical principles of mass, energy and momentum transfer; and they apply statistical models to tune model outputs based on historical and live data. FIGURE 1. Sanomatalo building: (a) general view; (b) detail of the façade (source: FUNIBER for the Energy IN TIME project). The building is connected to the district heating network and rooms are heated by waterborne radiators and fan coil units. There are four heat exchangers, one of them dedicated to the AHU heating network (power = 550 kW). All areas in the building have mechanical ventilation, which adjusts airﬂow based on room temperature and CO2 concentration. The BEMS is provided by Schneider Electric and allows con- trolling ventilation, heating, and cooling sub-systems from a centralized console. It enables about 2.000 inspection points, as well as an OPC (OLE for Process Control) module that allows remote setpoint writing. A canonical grey-box model –namely, the operational model– was created at system design time with the IESVE software by IES energy experts with the support of Caverion’s building operators. III. MATERIALS AND METHODS A. SANOMATALO BUILDING AND PILOT AREA 3 Sanomatalo3 (Sanoma house, ‘house of the press’) is a multi-purpose building situated in Helsinki and inaugurated in 1999. It was designed by Jan Söderlund and Antti-Matti Siikala, featuring a double glass façade with a steel frame structure to reduce the need for heating. In its 9 ﬂoors and 8227,56 m2, it houses the ofﬁces of the Sanoma media group and offers 2 ﬂoors of covered public space. The building is managed by Caverion,4 a Finnish construction and mainte- nance company. Similarly, MPC solutions have been successfully applied to optimize the use of different energy sources in buildings with mixed supply systems [35]–[37] and to achieve distributed control [38], [39] –enabling extensions to minimize com- munication between network components [40]. To increase the capabilities for solution diversiﬁcation, other search tech- niques have been applied to optimization in MPC, such as genetic algorithms [41]–[43] and particle swarm optimiza- tion [44]. To address the stabilization of the control process, nonlinear MPC solutions with varying horizon have been proposed [45]. 38750 VOLUME 7, 2019 38750 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models FIGURE 1. Sanomatalo building: (a) general view; (b) detail of the façade (source: FUNIBER for the Energy IN TIME project). available at the beginning of the project in 2013, but they were acquired in 2015-2017. For demo purposes, we identiﬁed a pilot area of 2,748.60 m2 encompassing ﬂoors 6th to 8th, which include small-size ofﬁces, meeting rooms, and open poly- valent spaces. The use of the pilot area is the expected one for an ofﬁce building, with ﬂexible working hours between 6am–18pm and an overall ﬂoor space factor of 26.2 m2/person. Total electricity consumption in the pilot area in 2017 from January to April was about 60 MWh, while district heating consumption was about 35 MWh in the same area and period. These ﬂoors are served by a single not-shared air handling unit (AHU), which is conﬁgured by means of a temperature setpoint. This piece of equipment was the main parameter of the energy optimization strategies (see section III.C). In addition, we adjusted the air volume setpoint of three variable air volume (VAV) units serving 8th ﬂoor. FIGURE 1. Sanomatalo building: (a) general view; (b) detail of the façade (source: FUNIBER for the Energy IN TIME project). B. SIMULATION MODEL AND CALIBRATION IESVE comprises a series of individual components including climate, geometric modelling, solar shading, energy and carbon, lighting, airﬂow, thermal mass, value/cost and egress modules that are linked by a single Integrated Data Model (IDM) through a Common User Inter- face (CUI). By combining these modules, we can model and simulate all aspects of a building’s construction, location, geometry, climate, usage, sub-systems and thermal perfor- mance. p g The main challenge in Sanomatalo is minimizing energy consumption (and costs) while guaranteeing comfort (indoor temperature and CO2 concentration) during the heating sea- son –usually between September and May, being the period from January to March the coldest one. Indoor temperatures can be retrieved in real-time through the BEMS, whereas CO2 sensors cannot be remotely accessed –data must be down- loaded ofﬂine. Heating consumption is monitored every hour by a separated sub-system. District heating prices are ﬁxed for each season, amounting to approximately 50e/MWh in the harsh winter period (Jan-Feb), and 45e/MWh in the remainder of the winter period (Mar-May, Nov-Dec). Electricity price is about 77 and 79e/MWh, respectively. No detailed historical records of sensor measurements were The simulation model developed for Sanomatalo included: (a) the passive components of the building (façade, claddings, solar irradiation, etc.), created with the ModelIT and the Sun- Cast modules; (b) the active components (anything producing or consuming electricity, especially in relation to the HVAC system), created with the ApacheHVAC, MacroFlo and Vista modules; (c) the expected building conditions (predicted occupancy and weather forecast). Simulation was performed by the ApacheSim module, which dynamically simulates the interaction between all of the active and passive elements over a selected period of time, taking into account the external 38751 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models inﬂuences (i.e. weather and occupancy) and the internal ther- mal behavior. The results of the simulation were viewed in the VistaPro module for analysis of heating and cooling loads, energy consumption, internal temperatures, thermal comfort, etc. (2) The airﬂow of 3 VAV devices (VAVairﬂowi) in ﬂoor 8th, in the range [50, 200] l/s. The choice of selecting these 3 VAVs was the limited availability of CO2 sensors at the beginning of the project: only the area affected by these 3 VAVs was monitored. The details of the Sanomatalo model are not public and fall out of the scope of this paper. B. SIMULATION MODEL AND CALIBRATION Nevertheless, this should not be seen as a limitation of our proposal. On the contrary, our approach is agnostic to the underlying simulation model, as far as it allows setting operational proﬁles as input. In pre-OPG operation, Tsupply values were manually set by the operators and VAVairﬂow values were automatically set by using presence sensors. The comfort requirements for the new system in the heating period were the following: The parameters of the operational model were con- tinuously adjusted to ﬁt live data measurements with the simulation output. Calibration was implemented as a semi-automatic procedure encompassing two iterative steps: (1) measuring the model accuracy by comparing simulation outputs with measured building data; (2) modifying model parameters to reduce model errors. In addition, IES carried out an entropy analysis to detect which parameters have the greatest inﬂuence in the model output, and therefore should be ﬁrstly modiﬁed. Overall, the calibration procedure resulted in a simulation model yielding errors below 5% [66]. • Indoor air temperature (IAT) must be in the range [20.5, 21.5] ◦C during ofﬁce hours 6:00–18:00. A ﬂexible margin in [20, 22] ◦C is considered acceptable. This temperature was represented by 25 output simulation variables, corresponding to 25 sensors spread across the 3 ﬂoors directly accessible through the BEMS. 3 ﬂoors directly accessible through the BEMS. • CO2 concentration (Con) upper limit is 850 ppm dur- ing ofﬁce hours. This concentration was represented by 4 output simulation variables, corresponding to 4 sensors for which there were no live measurements through the BEMS. The target variables to optimize were the heat and the fan power consumption meters of the pilot area –one of each for the whole pilot area–, which we will call Heat and Fan. They were represented by two output variables in the simulation model. There were no corresponding physical sensors for these variables, but their values can be directly derived from the BEMS temperature and air ﬂow measurements. C. ENERGY OPTIMIZATION STRATEGIES Following the Energy IN TIME terminology, control strate- gies specify the setpoint values allowed for each piece of actionable equipment. Strategies can denote single setpoint restrictions (e.g. setpoint variable range, frequency of change) or cross-parameter restrictions (e.g. two setpoints cannot have speciﬁc values at the same time). Besides, strategies can vary depending on the season. Energy optimization strategies are strategies enriched with heuristic information aimed at improving the energy efﬁciency and maintaining comfort. That is, energy optimization strategies deﬁne additional set- point constraints that can help to reduce energy consumption (e.g. reasonable length of the pre-heating period). Energy optimization strategies can be seen as the instantiation of the Energy Management Control functions proposed in [22] for a particular building. 2) GENERATION OF CANDIDATE PLANS Let us consider a time instant t, and the corresponding set- point value at this time st. For example, in Fig. 3, let us suppose t = 9:00; hence, st = 23 for Tsupply setpoint. We notate setpoint values at time t −1t as st−1t; e.g. if 1t = 4, then st−4 = 20, considering 1-hour intervals for simplicity’s sake. Let us notate the modiﬁcation of a setpoint value s as ˆs = s ± 1s; e.g. decrementing st in 1s = 0.5 give us ˆst = 23 −0.5 = 22.5. The sets {1t} and {1s} are discrete and ordered. We deﬁne a time horizon 1tmax = max {1t} to limit the temporal window of the modiﬁcations, as well as a maximum setpoint change value 1smax = max {1s} . D. EVALUATION METHODOLOGY Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models With this model, we obtained a reliable approximation of the consumption that would have been measured if the system without the OPG had been used during the real test period. More details of this procedure are described in Section V.B. the setpoint means reducing the IAT and the energy consump- tion. We will also center the explanation in type A situa- tions (see below). Nevertheless, the same principle applies to problems involving multiple variables and situations B and C. The explanation can be easily extended to more than one (independent) variable. We also studied comfort in terms of the indoor tempera- tures (IAT) and CO2 concentration (Con) mentioned above, checking that the simulated and measured values were within the acceptable ranges. The overall functioning of the algorithm is depicted in Fig. 2, and its details are covered in the following subsections. A. CONTROL SYSTEM ARCHITECTURE The Operational Plan Generation module is the core of the control system in Energy IN TIME. It encompasses three main stages: 1) 8:00 – 12:00: Heating control results in an IAT above the upper comfort threshold. The previous Tsupply set- points must be reduced to guarantee comfort. 1) Collection of forecasted building data: The OPG retrieves the weather forecast and the occupancy pre- diction for the operation period, usually the next day. Within our project, the weather forecast was obtained from the Weather Analytics API,5 and occupancy pre- dictions were obtained by using an agenda, which iden- tiﬁed working days and average room occupancy per hour. 2) 7:00 – 13:00: Heating control results in an IAT within the comfort range, but it may be possible to reduce the previous Tsupply setpoints while keeping the tempera- ture above the lower bound of the comfort threshold. Note that in both situations setpoint decrement may not be possible if it is already at the minimum value allowed by the equipment. 2) Generation, simulation and evaluation of candidate plans: The OPG runs several simulations to reproduce the expected building behavior, in terms of energy con- sumption and comfort, under different operation plans and according to the forecasted conditions retrieved in the previous stage. The best plan in terms of energy efﬁ- ciency and comfort satisfaction is selected. We explain in Section IV.B how these candidates to best plan are generated and assessed. Analogously, we can identify one situation in which more energy is required, since the comfort requirements are not satisﬁed: 1) 15:00 – 18:00: Heating control results in an IAT below the upper comfort threshold. The Tsupply setpoint must be increased Note that in this case setpoint increment may not be pos- sible if it is already at the maximum value allowed by the equipment. 3) Storage and execution of best plan: The best plan is stored in a database and made available to the setpoint writing component, which eventually sends the OP control instructions to the BEMS. This database also stores the context associated to each selected OP, i.e. the forecasted building data used by the OPG algorithm and the simulation results. This information is useful to explain the rationale of an OP to the building managers, who can revise and modify the setpoint values in real time as well. 5http://dev.weatheranalytics.com D. EVALUATION METHODOLOGY Following the International Performance Measurement and Veriﬁcation Protocol (IPMVP), our evaluation methodology compared energy savings achieved by the OPG with respect to a base case in which it is not used. This process was carried out both in the simulation environment and in the real building: • Evaluation in the simulation environment: We selected 3 days in the 2016-2017 period, respectively corre- sponding to a prototypical average (12-Jan-2016), cold (21-Jan-2016), and warm day (30-Jan-2017) of the win- ter season. The baseline was the real operation of the building for the same days. These data were collected at the beginning of the project. More details of this procedure are described in Section V.A. During the plan generation process, the operational model is cloned and reconﬁgured according to the forecasted occu- pancy and weather conditions –namely, the independent pro- ﬁle variables. As introduced in Section IV.A, the occupancy was measured as the room occupancy % from the building agenda, and the weather was a set of variables including outdoor air temperature (OAT), solar irradiance, etc. • Evaluation in the real building: The OPG was activated in the building during a 30-day period in the late winter season, from April 19th to May 19th 2017. The reason of this choice is that we identiﬁed in the simulation envi- ronment that the OPG can achieve better results in the transitions between seasons –usually, the heating season in Sanomatalo ends in the second week of May. For the baseline, we built a regression model from historical data which estimates the energy consumption of the HVAC system without the OPG from the weather and the occu- pancy values, following the recommendations in [67]. Therefore, to run a simulation, we specify the operational input proﬁles –i.e. the equipment setpoint sequences to be tested in the simulation– and the independent proﬁles –i.e. the occupancy and the weather time series–, in order to get the predicted proﬁles –i.e. the value sequences for indoor temperatures, CO2 concentration, and energy consumption. Energy optimization strategies for the Sanomatalo experi- ments with the OPG solution encompassed: Energy optimization strategies for the Sanomatalo experi- ments with the OPG solution encompassed: (1) The supply temperature of the AHU in the pilot area (Tsupply), in the range [17, 23] ◦C; (1) The supply temperature of the AHU in the pilot area (Tsupply), in the range [17, 23] ◦C; 38752 VOLUME 7, 2019 VOLUME 7, 2019 J. IV. PREDICTIVE CONTROL ALGORITHM A. CONTROL SYSTEM ARCHITECTURE In Fig. 3, we show an optimization scenario in which the IAT is controlled by a single Tsupply setpoint, as in our building. Given an initial plan for Tsupply values, we can simulate it and identify opportunities for energy optimization: B. OPERATIONAL PLAN GENERATION Creation of alternative plans is performed by an iterative algorithm based on a greedy heuristic and extended to balance diversity and local optimization. In this section, we explain the main steps of this stage: (1) identiﬁcation of situations of interest for energy savings or comfort improvement; (2) gen- eration of candidate plans to address these situations; (3) can- didate plans simulation and selection of the best one. The candidate plan generation process starts from the current best plan, which at the beginning can be prede- ﬁned or roughly estimated from outdoors temperatures. Next, it detects a situation of interest by analyzing the simulation of the current best plan; e.g. in our example, situation A at t = 9:00. For this t, the algorithm will propose a few candidate plans by decrementing previous setpoint values. To illustrate the processing in the OPG, we will assume a case with only one Tsupplysetpoint, in which decrementing 38753 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models FIGURE 2. Overall functioning of the OPG algorithm, including main stages: (1) identification of situations of interest (Section IV.B.1); (2) generation of candidate plans (Section IV.B.2); (3) plan simulation (IAT outside the comfort range during office hours is marked with ) and assessment (Section IV.B.3). The second candidate plan is selected, because it has the best comfort ranking. FIGURE 2. Overall functioning of the OPG algorithm, including main stages: (1) identification of situations of interest (Section IV.B.1); (2) generation of candidate plans (Section IV.B.2); (3) plan simulation (IAT outside the comfort range during office hours is marked with ) and assessment (Section IV.B.3). The second candidate plan is selected, because it has the best comfort ranking. 4) The transition probability at each step from ˆst′ to ˆst′−1 is given by the following function (Eq. 1): To model all possible combinations of setpoint modiﬁca- tions in situation A, we deﬁne a lattice graph like the one in Fig. 4. Each vertex of this graph represents a setpoint modiﬁcation at a given previous instant: ˆst−1t = st−1t − 1st−1t. Each directed edge connects a setpoint change with the following setpoint change in reverse time order. p(ˆst′ →ˆst′−1) = ( δ if 1st′ = 1st′−1 1−δ \|{1s}\|−1 otherwise (1) (1) with the diversiﬁcation parameter δ ∈[0, 1]. 6A random walk is a path consisting of a sequence of random steps on a mathematical space. Formally, it can be deﬁned as a sum of a sequence of independent, identically distributed random variables representing move directions, or as a Markov chain over the subjacent state space [68]. 3) CANDIDATE PLANS SIMULATION AND SELECTION Each w ∈W produces a candidate OP, which is built by replacing the appropriate setpoints of the current best plan by ˆst−1, ˆst−2, . . . , ˆst−1tmax . The remaining setpoints outside the [t −1tmax, t −1] interval are not modiﬁed. The algorithm only selects a small random subset of candidate OPs C ⊆ W to be simulated. In the best case, all the OPs in C will be simulated in parallel; therefore, the selection of C may depend on the simulator capabilities (see the experimental setup in Section V). Finally, the algorithm picks the most efﬁcient OP satis- fying comfort requirements. Efﬁciency is calculated as the total energy consumption of the plan, while comfort can be measured in different ways; for example, by using the root-mean-square deviation (RMSD) or the % of time with comfort-related values (e.g. IAT, Con) inside the comfort range, maybe limited to a period of interest (e.g. ofﬁce hours). If there is no such plan, the OPG selects the closest one to meet the requirements. To do so, OPs are ﬁrstly sorted by comfort satisfaction, and secondly by energy consumption. FIGURE 4. Lattice graph representing possible setpoint modifications (situation A, Tsupply). Time intervals are set to 1 hour, setpoint modifications are multiples of 0.5 ◦C. This assumption considerably reduces the number of pos- sible walks, as shown in Fig. 6. Note that this restriction may prevent the algorithm to explore the complete range of allowed {1s} modiﬁcations. Moreover, other heuristics could be incorporated to the process by means of additional walk restrictions encoded in the transition probability function; e.g. to limit how many different 1s can be used in the same walk. The procedure is restarted to identify the next interesting situation (Section IV.B.1), now using the simulation of the new plan as a reference. The algorithm iterates while there are remaining situations to process or when a maximum number of situations have been processed. B. OPERATIONAL PLAN GENERATION This function balances two choices: maintaining the same previous setpoint change (δ) and selecting any setpoint change (1−δ). If δ = 0, the transition probabilities at each step are the same for each allowed direction. If δ ≫0, the setpoints will tend to decrease in the same amount. From this graph, the setpoint modiﬁcations that form a candidate plan are modeled as the result of a random walk6 through this graph w = ⟨ˆst, ˆst−1, ˆst−2, . . . , st−1tmax⟩, with the following properties: 1) The walk starts at (0, 0) node, representing the current setpoint at starting time t (i.e. the current setpoint is not modiﬁed) ′ An identical graph is built in situation B. An analogous graph and a corresponding probability function are deﬁned in situation C to represent setpoint increments. 2) Each step goes from t′ to t ′−1 for any t′ in the sequence (i.e. always moving from right to left in the graph) In Fig. 5, we depict two examples of random walks and the resulting setpoint modiﬁcation sequences w1, w2. 3) The length of each path is \|{1t}\| (i.e. each path is a sequence of setpoint changes from t to t −1tmax) To reduce the number of possible alternatives, we can introduce an additional restriction to the walks: 1) Only moves to closest nodes in horizontal, vertical and diagonal directions are allowed (i.e. differences between time instants of changes of 1s, if any, are small) VOLUME 7, 2019 38754 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models FIGURE 3. Identification of savings opportunities and discomfort in a simulated plan: indoor temperature vs Tsupply setpoint values . The comfort range in [20, 22] ◦C is also shown (dashed line). FIGURE 3. Identification of savings opportunities and discomfort in a simulated plan: indoor temperature vs Tsupply setpoint values . The comfort range in [20, 22] ◦C is also shown (dashed line). FIGURE 4. Lattice graph representing possible setpoint modifications (situation A, Tsupply). Time intervals are set to 1 hour, setpoint modifications are multiples of 0.5 ◦C. C. COMPUTATIONAL PROPERTIES The OPG algorithm cannot guarantee a global optimum in terms of energy consumption for two reasons: (a) only a limited number of setpoint modiﬁcations are explored; (b) the global plan is built from locally pseudo-optimal choices focused on situations of interest A, B, C. Conversely, it yields good solutions in a reasonable time, and allows easy incorpo- ration of heuristics in the setpoint modiﬁcation process. FIGURE 5. Samples of setpoint modification sequences obtained by using random walks, restrictions (A)–(D) apply. Regarding (a), in the general formulation (Fig. 4 and 5), the number of possible setpoint change sequences \|W\| for each situation and independent variable is bounded by \|W\| ≤ \|{1s} + 1\|\|{1t}\|. In the restricted formulation (Fig. 6), the number of possible random walks is bounded by \|W\| < 3\|{1t}\|. The \|W\| for multiple-dimension random walks grows exponentially [68]. In any case, only \|C\| ≪\|W\| candidate OPs will be simulated at each iteration. Therefore, the overall efﬁciency of the algorithm is bounded by the number of sit- uations of interest processed multiplied by the time required to run each batch of simulations of size \|C\|. Note that the execution time of the OP generation process is insigniﬁcant compared to the simulation time. FIGURE 5. Samples of setpoint modification sequences obtained by using random walks, restrictions (A)–(D) apply. The parallel cloud version of IESVE allows running a ﬁxed number p of parallel simulations without performance degra- dation. To increase solution diversity, we can set \|C\| < p, and our implementation will ﬁll the remaining simulation slots with other plans, namely: (a) random variations of the current best OP at any time before t; (b) combinations of previously discarded good OPs; (c) baseline OPs –e.g. for Sanomatalo, OPs based on outdoors temperature. These plans are compared against the plans obtained with the random walks, and can be selected as best current plan for the next iteration in the same conditions. FIGURE 6. Simplified setpoint modification graph and random walk sample, restrictions (A)-(E) apply. Regarding (b), under some realistic assumptions, the OPG algorithm ﬁnds a good approximation to the optimal solu- tion. Speciﬁcally, for Tsupply control in Sanomatalo we can assume that external temperatures and internal occupancy values follow a bell-shaped curve. 4) TRIGGERING THE OPG ALGORITHM Our implementation of the OPG considers two particu- lar situations: pre-conditioning and post-conditioning. Pre- conditioning is performed to achieve comfort at the beginning of the working day, while post-conditioning is performed to save energy by relaxing the comfort requirements at the end of the working day and later. We apply predeﬁned setpoint change strategies for each variable during these intervals, which allow us to reduce the number of required simulations. The OPG algorithm is usually launched before midnight to calculate the setpoints for the next day, allocating enough time to let the process ﬁnish before setpoints are due – a few hours in most cases. The algorithm can run again several times during the day, in order to create a new plan for the remainder of the day using updated weather and occupancy predictions and to recover from control deviations and failures. 38755 VOLUME 7, 2019 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models FIGURE 5. Samples of setpoint modification sequences obtained by using random walks, restrictions (A)–(D) apply. note that such a receding horizon could be implemented just by generating control instructions for a whole period (e.g. 24 hours) each time the OPG algorithm is triggered, instead of generating control instructions until the end of the current day. At the moment, the implemented recalculation process only generates a baseline plan using predeﬁned operation curves when an updated weather forecast signiﬁcantly differs from the initially used one. Enabling a faster and maybe simpliﬁed version of the OPG for quick recovery during the day, triggered by different events –e.g. comfort degradation is detected with live BEMS data–, remains as future work. V. EXPERIMENTS AND RESULTS The OPG algorithm has been implemented in the Python and R programming languages. For the experiments in this section, it ran on a Supermicro SuperServer 6027R-TRF, conﬁgured with 2 processors Intel Xeon E-2600 2.4GHz (2 × 8 cores), 128 GB RAM, 2 × 600 GB magnetic storage. The details of the cloud-based version of the IESVE simulator are not disclosed by IES by conﬁdentiality reasons. The OPG algorithm has been implemented in the Python and R programming languages. For the experiments in this section, it ran on a Supermicro SuperServer 6027R-TRF, conﬁgured with 2 processors Intel Xeon E-2600 2.4GHz (2 × 8 cores), 128 GB RAM, 2 × 600 GB magnetic storage. The details of the cloud-based version of the IESVE simulator are not disclosed by IES by conﬁdentiality reasons. Experimentation based only in the simulation environment was performed in advance to test and tune the deployment of the OPG used in the real building. After some prelimi- nary tests and following the building requirements, the OPG parameters were set to the following values: • Simulation batch size: p = 50, resulting in simulation times below 20 minutes • Ahead period of the OPG: 1 day, no receding horizon GURE 7. Comparison of baseline and OPG plans in terms of setpoi d comfort values (with comfort thresholds) for a Standard winter d mulation environment: (a) Tsupply operation; (b) VAVairflowi opera h b d h l f h l d • OPG starting time: > 2 hours before the ﬁrst setpoint is due • Maximum setpoint change frequency: 15 minutes • Simulation output resolution: 15 minutes • Simulation output resolution: 15 minutes • Only minor setpoint changes are allowed in random walks • {1t}Tsupply, VAVairﬂow = {30, 60, 90, 120} minutes • {1s}VAVairﬂow = {0, 25, 50} l/s • δ = 0.3 (random setpoint modiﬁcations are preferred) • Tsupply and VAV setpoints are optimized independently (ﬁrst Tsupply and then VAVairﬂow) • Comfort satisfaction is measured by using the RSMD from the comfort interval Additionally, the maximum number of simulation batches was restricted in order to establish an upper limit to the execu- tion time. Since an average simulation batch took 20 minutes (with p = 50), we set the maximum number of batches per plan to 6 in order to keep the execution time under 2 hours. V. EXPERIMENTS AND RESULTS This means that 6 A-B-C situations (Figure 3) can be analysed in each run of the OPG. Excluding pre- and post- conditioning, 4 out of 6 were reserved for Tsupply changes, and 2 for VAVairﬂow changes. Situations are sorted by rele- vance at each iteration of the OPG algorithm; e.g. for Tsupply, situations A and C are more important than B. C. COMPUTATIONAL PROPERTIES To guarantee comfort in the winter season, the optimal OP would entail increasing the temperature setpoints in the early morning, then decreasing them around noon, and maybe incrementing them again in the afternoon. Moreover, the low external temperatures favor heat losses, which would in turn require supplying hot air FIGURE 6. Simplified setpoint modification graph and random walk sample, restrictions (A)-(E) apply. This formulation slightly diverges from the receding horizon control typically implemented in canonical MPC, because the prediction horizon is not shifted. However, 38756 VOLUME 7, 2019 38756 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models URE 7. Comparison of baseline and OPG plans in terms of setpoin d comfort values (with comfort thresholds) for a Standard winter d ulation environment: (a) Tsupply operation; (b) VAVairflowi opera the base and the OPG plans for the Tsupply and Vairﬂowi operation –the main working hours are delimi both cases, the largest operation differences correspon frequently. Therefore, it is safe to limit the search to local setpoint increments and decrements. A. SIMULATION ENVIRONMENT As described in Section II.D, we selected three prototypical days of the winter season: average (Standard), cold (Harsh) and warm (Intermediate). Then, we simulated the behaviour of the pilot area of the building according to the setpoints orig- inally applied (i.e. the base plan) and the setpoints calculated by our algorithm (i.e. the OPG plan), in order to check how they compare in terms of comfort and consumption. FIGURE 7. Comparison of baseline and OPG plans in terms of setpoints and comfort values (with comfort thresholds) for a Standard winter day, simulation environment: (a) Tsupply operation; (b) VAVairflowi operation. of the base and the OPG plans for the Tsupply and the VAVairﬂowi operation –the main working hours are delimited. In both cases, the largest operation differences correspond to Fig. 7 depicts the simulation results for a Standard day, corresponding to the most common conditions during the winter season. In the top of the ﬁgure, we show the setpoints VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models FIGURE 8. Comparison of power (kW) of baseline and OPG plans for a STANDARD winter day, simulated environment: (a) heat meter; (b) VAV fan power meter. Table 1 includes the detailed numbers for the three refer- ence days. To obtain the overall energy consumption, we have approximated the integral of the power functions with the area under the curve (AUC). AUC has been computed by: (1) interpolation of data points with a spline; and (2) calculation of the adaptive quadrature of the interpolated function [69]. It can be seen that the OPG reduces the power consumption of the base operation of both the heating and the ventila- tion subsystems. As expected, in the experiments the highest heating savings are achieved in the warmer intermediate day, when there is still room for adjustments. Broadly speaking, the OPG dynamically adapts the operation to the particular conditions of a speciﬁc day without requiring the operator attention, which is convenient in less cold days in the winter season or before transitioning to the spring season. Con- versely, the HVAC system is already operating at (almost) full power during the harsh days to achieve comfort, and therefore there is little room for improvement during working hours. A more detailed discussion on these features is included in Section VI. On the other hand, fan power savings have similar values in different working days. A. SIMULATION ENVIRONMENT The bad results in the intermediate day were the consequence of the misestimation of the occupancy used by the algorithm. B. ON-SITE TEST AND EVALUATION The evaluation in the pilot area of the real building was performed from April 19th to May 19th 2017. These days mostly ﬁt into the Intermediate category studied in the previ- ous section, the one which yielded the highest energy savings. FIGURE 8. Comparison of power (kW) of baseline and OPG plans for a STANDARD winter day, simulated environment: (a) heat meter; (b) VAV fan power meter. The baseline for daily energy consumption was calcu- lated by a generalized linear regression model (glmnet) [70], a method based on lasso analysis (least absolute shrinkage and selection operator). Other prediction techniques, such as linear regression or autoregression, could have also been explored. Source data for the model was obtained from build- ing sensors (energy, OAT and occupancy) logged in the period February-May 2016. the less crowded periods. Note that the Tsupply setpoints of the base plan before 5:00 and after 21:00 are registered but not applied; control is managed by a human-operated switch. In the bottom of the ﬁgure, we show the mean comfort values obtained in simulation in terms of IAT and Con; these values lay within the comfort intervals (also included in the ﬁgure). Fig. 8 shows the power consumption calculated by the sim- ulation model; respectively, the Heat and Fan power meters values. We can observe that most savings are achieved at the borderline hours, that is, at the beginning and at the end of the working day. This is consistent with the pre-conditioning and post-conditioning provisions made by the OPG algorithm. the less crowded periods. Note that the Tsupply setpoints of the base plan before 5:00 and after 21:00 are registered but not applied; control is managed by a human-operated switch. In the bottom of the ﬁgure, we show the mean comfort values obtained in simulation in terms of IAT and Con; these values lay within the comfort intervals (also included in the ﬁgure). More speciﬁcally, we developed two baseline models for prediction of daily consumption of heating equipment and VAV fans, based on the expected heating demand and occu- pancy. Expected daily energy demand (hdd, in heating degree days) was calculated by using integration with base temper- ature set to 18 ◦C and the BEMS OAT [71]. Estimated daily occupancy (occ, in %) was the maximum occupancy value of the ofﬁce agenda. To build the prediction models, we ﬁrstly pre-processed the data, discarding outliers and measurement errors. B. ON-SITE TEST AND EVALUATION Savings have been achieved without compromising users’ comfort. Figure 11 shows the IAT and CO2 concentration values in the pilot area in the evaluation period. The IAT values were calculated as follows: (1) sensor measurements, obtained from the BEMS temperature sensors (25) were FIGURE 9. Comparison of daily energy consumption (kWh) estimated by the baseline models vs historical data in Feb-May 2016: (a) heating; (b) VAV fans. FIGURE 10. Comparison of daily energy consumption (kWh) during the test period vs estimated by the baseline models: (a) heating; (b) VAV fans; (c) savings. Days in red italic font are weekend or holiday days. FIGURE 9. Comparison of daily energy consumption (kWh) estimated by the baseline models vs historical data in Feb-May 2016: (a) heating; (b) VAV fans. FIGURE 10. Comparison of daily energy consumption (kWh) during the test period vs estimated by the baseline models: (a) heating; (b) VAV fans; (c) savings. Days in red italic font are weekend or holiday days. a low R2 value, which means that energy savings calculated with this model should be considered with caution. Heat∗(hdd, occ) = −9.122×hdd+0.703×occ+20.71 (2) Fan∗(hdd, occ) = −0.038×hdd+0.012×occ+5.015 (3) Fig. 10(a) and 10(b) show the comparison of the values of daily energy consumption in the pilot area obtained from the BEMS (+) with the values estimated by the prediction models (o) for the test period in the real building. Fig. 10(c) shows the energy savings achieved in % of the (estimated) consumption before optimization. FIGURE 10. Comparison of daily energy consumption (kWh) during the test period vs estimated by the baseline models: (a) heating; (b) VAV fans; (c) savings. Days in red italic font are weekend or holiday days. TABLE 2. Energy savings (kWh) achieved in the on-site test with the OPG control vs estimated by the baseline models. TABLE 2. Energy savings (kWh) achieved in the on-site test with the OPG control vs estimated by the baseline models. TABLE 2. Energy savings (kWh) achieved in the on-site test with the OPG control vs estimated by the baseline models. and around 20% for the electrical subsystem. Weekends and holidays offer opportunities for higher energy savings, since the OPG adjust the operation to the building occupancy better than the manual operation. Savings have been achieved without compromising users’ comfort. Figure 11 shows the IAT and CO2 concentration values in the pilot area in the evaluation period. B. ON-SITE TEST AND EVALUATION Fig. 8 shows the power consumption calculated by the sim- ulation model; respectively, the Heat and Fan power meters values. We can observe that most savings are achieved at the borderline hours, that is, at the beginning and at the end of the working day. This is consistent with the pre-conditioning and post-conditioning provisions made by the OPG algorithm. TABLE 1. Energy consumption (kWh) for the experiments in the simulation environment. 38758 TABLE 1. Energy consumption (kWh) for the experiments in the simulation environment. 38758 TABLE 1. Energy consumption (kWh) for the experiments in the simulation environment. Fig. 9 compares the energy consumption in February- May 2016 and the values calculated by the heating and the fan consumption prediction models. The parameters of the regression models are given in Eq. 2 and Eq. 3 respectively, yielding correlation coefﬁcient values of R2 = 0.632 and R2 = 0.234. Note that: (1) the heating baseline model slightly overestimates consumption from mid-April to June, which means that energy savings calculated in the next section are slightly overestimated as well; (2) the fan power model has VOLUME 7, 2019 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models FIGURE 9. Comparison of daily energy consumption (kWh) estimated by the baseline models vs historical data in Feb-May 2016: (a) heating; (b) VAV fans. a low R2 value, which means that energy savings calculated with this model should be considered with caution. Heat∗(hdd, occ) = −9.122×hdd+0.703×occ+20.71 (2) Fan∗(hdd, occ) = −0.038×hdd+0.012×occ+5.015 (3) Fig. 10(a) and 10(b) show the comparison of the values of daily energy consumption in the pilot area obtained from the BEMS (+) with the values estimated by the prediction models (o) for the test period in the real building. Fig. 10(c) shows the energy savings achieved in % of the (estimated) consumption before optimization. TABLE 2. Energy savings (kWh) achieved in the on-site test with the OPG control vs estimated by the baseline models. FIGURE 10. Comparison of daily energy consumption (kWh) during the test period vs estimated by the baseline models: (a) heating; (b) VAV fans; (c) savings. Days in red italic font are weekend or holiday days. and around 20% for the electrical subsystem. Weekends and holidays offer opportunities for higher energy savings, since the OPG adjust the operation to the building occupancy better than the manual operation. B. ON-SITE TEST AND EVALUATION The on-site evaluation in the Sanomatalo building, which was carried out at the end of the 2017 heating season, conﬁrmed these assumptions. Energy usage in colder days could be even more optimized by relaxing the comfort temperature restrictions to permit OPs with some minor discomfort for a short period of time. The advantage of our system is that it allows operators to characterize and quantify this discomfort in advance, thus supporting them to make more informed decisions. (Note that this feature was not exploited in the experiments.) FIGURE 11. Daily comfort values achieved in the on-site test with the OPG control, maximum and minimum sensor average values: (a) indoor air temperature (IAT, ◦C), with comfort interval; (b) CO2 concentration (Con, ppm), with comfort threshold. CO2 measurements were only available from April 19th to 14th May. Days in italic red font are weekend or holiday days. obtained. Con values were retrieved from 4 ofﬂine sensors; the remainder of the procedure is the same as for IAT. IAT min values lie within the comfort range during the test period. Actually, it would have been possible to conﬁgure the OPG to reduce Tsupply even more. However, as explained at the beginning of this section, we prioritized optimizing discomfort situations. IAT max values are over the comfort upper threshold by 1 ◦C. A more detailed analysis of these values identiﬁed that discomfort was not sustained and only happened for short time periods (less than 1 hour). Our system reduced the temperature setpoints given by the normal operation of the building between 0.5 and 2 ◦C. During the on-site test, this meant savings in heating above 40% (Table 2) while keeping comfort (see Fig. 11(a)). The algorithm adapted well to workdays and weekends, show- ing slightly better results in the former ones (Fig. 10(c)). A possible explanation for this is that operators have lower availability in weekends and holiday days, and therefore it is not possible for them to create customized plans. The airﬂow consumption was also reduced in a 20% (Table 2) without compromising the CO2 concentration comfort (Fig. 11(b)), despite the lack of a proper model calibration and the smaller number of simulations involving VAVs. Nevertheless, due to the lower accuracy of the baseline model, these results are less precise and should be further analyzed; e.g. by using autoregression to build the baseline [72]. B. ON-SITE TEST AND EVALUATION The IAT values were calculated as follows: (1) sensor measurements, obtained from the BEMS temperature sensors (25), were resampled and interpolated to match the setpoint change frequency parameter (15 minutes); (2) sensor temperatures were averaged at each timestamp; (3) maximum and mini- mum values of timestamps within the working hours were As summarized in Table 2, the average savings per day are, respectively, around 40% for the thermal subsystem VOLUME 7, 2019 As summarized in Table 2, the average savings per day are, respectively, around 40% for the thermal subsystem 38759 38759 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models FIGURE 11. Daily comfort values achieved in the on-site test with the OPG control, maximum and minimum sensor average values: (a) indoor air temperature (IAT, ◦C), with comfort interval; (b) CO2 concentration (Con, ppm), with comfort threshold. CO2 measurements were only available from April 19th to 14th May. Days in italic red font are weekend or holiday days. reduces energy consumption; particularly heating consump- tion. Automatic control allows for more effective plans since it enables a ﬁner-grained and more frequent scheduling of setpoint changes without the supervision of the building oper- ators. The OPG algorithm and software offer a ﬂexible and conﬁgurable framework to generate more efﬁcient operation plans, predicting the building state and adapting energy usage to more realistic demand estimations without compromising users’ comfort. As expected, it has proved to be particularly successful in optimizing temperature setpoints, in which a longer control horizon, accounting for the inertia of the equip- ment, is crucial. The building operators were satisﬁed by the use of the system during the test period. One highlighted system’s feature was the capabilities to validate the plans in advance (and even to modify them) and to provide jus- tiﬁcations of the algorithm decisions –by means of graphi- cal depictions of the simulation results, in a similar way to Fig. 11. As already anticipated by the experiments in the simulation environment (Section V.A), the highest energy savings can be obtained for heating in the mid-season, when it is not necessary to use the heating equipment at full, and particu- larly, in the warmer days (Table 1, Intermediate). At the same time, the system can react to isolated cold days. B. ON-SITE TEST AND EVALUATION Similarly, CO2 concentration values are mostly below the comfort threshold, although with some exceptions. After more detailed analysis, we identiﬁed that the highest values were measured by a single sensor, which in some cases exceeded 950 ppm. However, the levels calculated by the simulation environment were considerably smaller. This is a clear example of the importance of having all the sensor data available through the BEMS. With the CO2 sensors ofﬂine, our system was not able to recalibrate the simulation model –which would have led to better plans–, nor to detect in real time that some plans were not guaranteeing comfort –which would have triggered a correction action. In summary, although the Sanomatalo building was already efﬁciently operated, and considering the limitations of the baseline estimations, the overall savings ﬁgures in the inter- mediate winter are in line with the 30% target of EU energy directive [5] and the 35% savings estimations provided in [7]. The experiments also revealed more opportunities for sav- ings in the future, e.g. by improving the simulation model VI. DISCUSSION VII. CONCLUSIONS AND FUTURE WORK VII. CONCLUSIONS AND FUTURE WORK This paper has presented the design and the implementation of an MPC-based control system aimed at reducing energy consumption in non-residential buildings while guaranteeing occupants’ comfort. The main difference of our proposal with respect to other approaches is that we use a full-complexity simulation model, which runs in parallel in the cloud. This allows using more accurate models and facilitates commu- nication between computer scientists, building operators and simulation developers, exploiting synergies of their joint work. Comprehensive quantitative and qualitative compari- son with MPC approaches using reduced-complexity simu- lation models would be useful to support decision-making between different alternative approaches. In Table 3, we show a rough projection of energy savings in the pilot area for the whole year using: (1) savings calculated in Section V.A (only heating); (2) historical monthly con- sumption values provided by the building operators; (3) esti- mated distribution of day types. We assume that the heating system is not used during the summer, and therefore it does not make sense to quantify savings in this period. It can be seen that the overall energy consumption reduction during the whole winter season is around 20%, larger than the consump- tion of a standard winter month. We can also estimate savings of CO2 emissions: assuming a carbon factor of 206 kgCO2/ MWh for district heating energy in Finland [73], the new system applied in the pilot area can save more than 1.60 Tons of CO2 per year. These ﬁgures could be directly adapted to other estimation of day types (e.g. including savings in the summer) and extended to other sections of the building with similar conﬁguration. Experimentation in the Sanomatalo building, located in Helsinki, both in the simulation environment and in the real building, has shown that important energy savings –up to 40% at the end of the winter season– can be achieved, particularly by optimizing the control of the heating equipment. Note that our approach can be adapted to other scenarios, and speciﬁcally, to cooling equipment. In our experiments we did not consider the energy costs of running our system, which should be deducted from the HVAC savings [82]. These promising ﬁgures can give rise to disruptive models for energy service provision, as we explore in [83]. VI. DISCUSSION The results presented in Section V show that the use of MPC in the ofﬁces of the Sanomatalo building signiﬁcantly VOLUME 7, 2019 38760 38760 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models with online CO2 data (for calibration) and more detailed occupancy predictions (actual agenda data were not very ﬁne-grained). in our experiments were mostly static, while occupancy monitoring and reconﬁguration have shown effective in the past [77], [78]. In this regard, the capabilities and limitations for OP recalculation during the day should also be further explored. Four, a more comprehensive study of energy sav- ings with additional baseline models should be carried out in order to quantify more precisely the return of investing in our solution [79], in particular if only ventilation is addressed. Last but not least, the building setup was relatively simple, with district heating and almost ﬁxed energy costs. It would be interesting to study the applicability and the scalability of the OPG approach to smart grids, including more control variables –some of them affecting the production side– and energy storage equipment; see for example [80], [81]. TABLE 3. Energy savings (heating, MWh) in the pilot area projected for the whole year. TABLE 3. Energy savings (heating, MWh) in the pilot area projected for the whole year. REFERENCES [1] J. Laustsen, ‘‘Policy pathways: Energy performance certiﬁcation of build- ings,’’ International Energy Agency (IEA), Paris, France, Tech. Rep., 2010. [2] D. Urge-Vorsatz, K. Petrichenko, M. Staniec, and J. Eom, ‘‘Energy use in buildings in a long-term perspective,’’ Current Opinion Environ. Sustain- ability, vol. 5, no. 2, pp. 141–151, 2013. [3] L. Pérez-Lombard, J. Ortiz, and C. Pout, ‘‘A review on buildings energy consumption information,’’ Energy Buildings, vol. 40, no. 3, pp. 394–398, 2008. [4] X. Cao, X. Dai, and J. Liu, ‘‘Building energy-consumption status worldwide and the state-of-the-art technologies for zero-energy build- ings during the past decade,’’ Energy Buildings, vol. 128, pp. 198–213, Sep. 2016. p [5] Proposal for a Directive of the European Parliament and of the Council Amending Directive 2010/31/EU on the Energy Performance of Buildings, Eur. Commission, Brussels, Belgium, 2016. [6] A Clean Planet for all—A European Long-Term Strategic Vision for a Prosperous, Modern, Competitive and Climate Neutral Economy, Eur. Commission, Brussels, Belgium, 2018. [7] A. Ghahramani, K. Zhang, K. Dutta, Z. Yang, and B. Becerik-Gerber, ‘‘Energy savings from temperature setpoints and deadband: Quantifying the inﬂuence of building and system properties on savings,’’ Appl. Energy, vol. 165, pp. 930–942, Mar. 2016. pp [8] V. Marinakis, H. Doukas, C. Karakosta, and J. Psarras, ‘‘An integrated sys- tem for buildings’ energy-efﬁcient automation: Application in the tertiary sector,’’ Appl. Energy, vol. 101, pp. 6–14, Jan. 2013. [9] A. Costa, M. M. Keane, J. I. Torrens, and E. Corry, ‘‘Building operation and energy performance: Monitoring, analysis and optimisation toolkit,’’ Appl. Energy, vol. 101, pp. 310–316, Jan. 2013. [10] B. Swords, E. Coyle, and B. Norton, ‘‘An enterprise energy-information system,’’ Appl. Energy, vol. 85, no. 1, pp. 61–69, 2008. [11] Y. Lu, S. Wang, and K. Shan, ‘‘Design optimization and optimal control of grid-connected and standalone nearly/net zero energy buildings,’’ Appl. Energy, vol. 155, pp. 463–477, Oct. 2015. gy pp [12] A. Afram and F. Janabi-Shariﬁ, ‘‘Theory and applications of HVAC control systems—A review of model predictive control (MPC),’’ Building Envi- ron., vol. 72, pp. 343–355, Feb. 2014. [13] A. Mirakhorli and B. Dong, ‘‘Occupancy behavior based model predictive control for building indoor climate—A critical review,’’ Energy Buildings, vol. 129, pp. 499–513, Oct. 2016. [14] G. Serale, M. Fiorentini, A. Capozzoli, D. Bernardini, and A. Bemporad, ‘‘Model Predictive Control (MPC) for enhancing building and HVAC system energy efﬁciency: Problem formulation, applications and oppor- tunities,’’ Energies, vol. 11, no. 3, p. VII. CONCLUSIONS AND FUTURE WORK Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models t Time instant Tsupply AHU supply temperature setpoint value (◦C) VAV Variable air volume unit VAVairﬂowi Airﬂow setpoint value for VAV number i (liters per second, l/s) w Ordered sequence of setpoint changes from t to t –1tmax W Set of all w t Time instant Tsupply AHU supply temperature setpoint value (◦C) VAV Variable air volume unit VAVairﬂowi Airﬂow setpoint value for VAV number i (liters per second, l/s) w Ordered sequence of setpoint changes from t to t –1tmax W Set of all w apply reinforcement learning to reward and reuse particularly efﬁcient OPG plans. Finally, we believe that combining interpretable white/ grey-box models, like the one used in this work, and efﬁcient black-box models, learnt from historical data, is one of the most prospective directions for future work. Faster simulation of such hybrid model would allow for the implementation of more sophisticated optimization and planning techniques. Recent approaches to data-driven black-box models have showed good accuracy, but only for short time periods [84]. Learning more general and precise models would require larger datasets, more computational power, and techniques able to exploit them. Recent advances in the Deep Learning area suggest that this is a feasible goal. VAV VAVairﬂowi ACKNOWLEDGEMENTS The authors would like to thank Vicente Madero (Acciona Construcción S.A.); Catherine Conaghan, Adalberto Guerra, Stephen Earle (Integrated Environmental Solutions Ltd.); Christian Beder (Cork Institute of Technology); and Jukka Heino (Caverion Suomi Oy) for their support and assistance with this research work. NOMENCLATURE AHU Air Handling Unit AUC Area under the curve BCVTB Building Controls Virtual Test Bed BEMS Building Energy Management System C Set of candidate plans considered in an iteration of the OPG Con CO2 concentration (parts per million, ppm) δ diversiﬁcation parameter 1t Time increment / decrement {1t} Set of time increment / decrement values 1tmax Maximum time increment / decrement 1s Setpoint increment / decrement {1s} Set of setpoint increment / decrement values 1smax Maximum setpoint value increment / decrement Fan Energy consumption due to electrical subsystem (kWh) Fan∗ Estimated daily fan energy consumption with the baseline model (kWh) HDD Heating Degree Days hdd Estimated daily demand measured in HDD (integrated) Heat Energy consumption due to thermal subsystem (kWh) Heat∗ Estimated daily heating energy consumption with the baseline model (kWh) HVAC Heating, Ventilation, and Air Conditioning IAT Indoor Air Temperature (◦C) IPMVP International Performance Measurement and Veriﬁcation Protocol MILP Mixed Integer Linear Programming MPC Model Predictive Control OAT Outdoor Air Temperature (◦C) occ Estimated daily occupancy (maximum) (%) PMV Predicted mean value RSMD Root mean square deviation st Setpoint value at time t ˆst Setpoint value at time t modiﬁed 38762 VII. CONCLUSIONS AND FUTURE WORK The implementation of the system in other buildings entails: (a) developing a speciﬁc simulation model, if not available; (b) parametrizing the OPG algorithm, including the deﬁnition of energy optimization strategies; (c) ﬁnd- ing appropriate sources for weather and occupancy fore- casts; (d) adapting the setpoint writing component, if fully-automatic control is enabled; (e) deploying the compu- tational infrastructure to run these components. As a matter of fact, in the context of the Energy IN TIME project we applied modiﬁed versions of the OPG to other scenarios, such as an airport and a hotel – achieving similar results, as brieﬂy described in [66]. Among the tasks required to extend the system to other buildings, developing and tuning the simulation model is the most time-consuming one. gy p p The OPG algorithm opens several opportunities for further research. The current design relies on a variant of heuristic search, which can be hard to scale up if several variables are to be optimized at the same time. In this regard, other search and optimization techniques could be applied. Specif- ically, genetic algorithms allow balancing diversiﬁcation and intensiﬁcation of solution search by adjusting their param- eters. Another possible extension of the OPG would be to incorporate means to deﬁne imprecise comfort ranges, thus formalizing the notion of relaxed comfort into the proce- dure. It would also be interesting to study how to repre- sent energy optimization strategies in a machine-processable language, in such a way that the system could use them for self-conﬁguration. Moreover, self-conﬁguration could be supported by machine learning techniques able to identify successful operation patterns from historical data, and to The collaboration with the Sanomatalo building operators revealed some prospective improvements to the system. First, it would be convenient to offer a better interface for conﬁguration of the OPG and interaction with the generated plans, as in [74]. Second, users’ comfort should be measured beyond thermal and CO2 concentration inter- vals, probably by using PMV, adaptive comfort models and comfort standards [75], [76]. Third, occupancy estimations 38761 38761 VOLUME 7, 2019 J. REFERENCES Salsbury, and P. Xu, ‘‘Demand reduction in building energy systems based on economic model predictive control,’’ Chem. Eng. Sci., vol. 67, no. 1, pp. 92–100, 2012. [51] D. Picard, J. Drgoňa, M. Kvasnica, and L. Helsen, ‘‘Impact of the controller model complexity on model predictive control performance for buildings,’’ Energy Buildings, vol. 152, pp. 739–751, Oct. 2017. pp [30] S. J. Kang, J. Park, K.-Y. Oh, J. G. Noh, and H. Park, ‘‘Scheduling-based real time energy ﬂow control strategy for building energy management system,’’ Energy Buildings, vol. 75, pp. 239–248, Jun. 2014. [52] M. Gruber, A. Trüschel, and J. O. Dalenbäck, ‘‘Model-based controllers for indoor climate control in ofﬁce buildings—Complexity and performance evaluation,’’ Energy Buildings, vol. 68, pp. 213–222, Jan. 2014. [31] I. Hazyuk, C. Ghiaus, and D. Penhouet, ‘‘Optimal temperature con- trol of intermittently heated buildings using model predictive control: Part II—Control algorithm,’’ Buildings Environ., vol. 51, pp. 388–394, May 2012. [53] S. Petersen and K. W. Bundgaard, ‘‘The effect of weather forecast uncer- tainty on a predictive control concept for building systems operation,’’ Appl. Energy, vol. 116, pp. 311–321, Mar. 2014. [54] F. Oldewurtel, D. Sturzenegger, and M. Morari, ‘‘Importance of occu- pancy information for building climate control,’’ Appl. Energy, vol. 101, pp. 521–532, Jan. 2013. [32] R. De Coninck and L. Helsen, ‘‘Practical implementation and evaluation of model predictive control for an ofﬁce building in Brussels,’’ Energy Buildings, vol. 111, pp. 290–298, Jan. 2016. [55] B. Dong and K. P. Lam, ‘‘A real-time model predictive control for building heating and cooling systems based on the occupancy behavior pattern detection and local weather forecasting,’’ Building Simul., vol. 7, no. 1, pp. 89–106, 2014. [33] S. C. Bengea, A. D. Kelman, F. Borrelli, R. Taylor, and S. Narayanan, ‘‘Implementation of model predictive control for an HVAC system in a mid-size commercial building,’’ HVAC&R Res., vol. 20, no. 1, pp. 121–135, 2014. [56] Y. Kwak, J.-H. Huh, and C. Jang, ‘‘Development of a model predictive control framework through real-time building energy management system data,’’ Appl. Energy, vol. 155, pp. 1–13, Oct. 2015. [34] K. Deng et al., ‘‘Model predictive control of central chiller plant with thermal energy storage via dynamic programming and mixed-integer linear programming,’’ IEEE Trans. Autom. Sci. Eng., vol. 12, no. 2, pp. 565–579, Apr. 2015. [57] Y. Kwak and J.-H. Huh, ‘‘Development of a method of real-time build- ing energy simulation for efﬁcient predictive control,’’ Energy Convers. REFERENCES 631, 2018. g p [15] D. B. Crawley, J. W. Hand, M. Kummer, and B. T. Grifﬁth, ‘‘Contrasting the capabilities of building energy performance simulation programs,’’ [15] D. B. Crawley, J. W. Hand, M. Kummer, and B. T. Grifﬁth, ‘‘Contrasting the capabilities of building energy performance simulation programs,’’ Building Environ., vol. 43, no. 4, pp. 661–673, Apr. 2008. Building Environ., vol. 43, no. 4, pp. 661–673, Apr. 2008. [16] P. Rockett and E. A. Hathway, ‘‘Model-predictive control for non-domestic buildings: A critical review and prospects,’’ Build. Res. Inf., vol. 45, no. 5, pp. 556–571, 2017. VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models [17] S. Meyers and S. Kromer, ‘‘Measurement and veriﬁcation strategies for energy savings certiﬁcates: Meeting the challenges of an uncertain world,’’ Energy Efﬁciency, vol. 1, no. 4, pp. 313–321, 2008. [41] H. Pombeiro, M. J. Machado, and C. Silva, ‘‘Dynamic programming and genetic algorithms to control an HVAC system: Maximizing thermal comfort and minimizing cost with PV production and storage,’’ Sustain. Cities Soc., vol. 34, pp. 228–238, Oct. 2017. [18] A. Mahdavi, ‘‘Simulation-based control of building systems operation,’’ Building Environ., vol. 36, no. 6, pp. 789–796, 2001. pp [42] F. Ascione, N. Bianco, C. De Stasio, G. M. Mauro, and G. P. Vanoli, ‘‘Simulation-based model predictive control by the multi-objective opti- mization of building energy performance and thermal comfort,’’ Energy Buildings, vol. 111, pp. 131–144, Jan. 2016. pp [19] J. A. Clarke et al., ‘‘Simulation-assisted control in building energy man- agement systems,’’ Energy Buildings, vol. 34, no. 9, pp. 933–940, 2002. [20] S. Petersen and S. Svendsen, ‘‘Method for simulating predictive control of building systems operation in the early stages of building design,’’ Appl. Energy, vol. 88, no. 12, pp. 4597–4606, 2011. [43] S. Wang and X. Jin, ‘‘Model-based optimal control of VAV air-conditioning system using genetic algorithm,’’ Building Environ., vol. 35, no. 6, pp. 471–487, 2000. [21] M. Killian and M. Kozek, ‘‘Ten questions concerning model predic- tive control for energy efﬁcient buildings,’’ Building Environ., vol. 105, pp. 403–412, Aug. 2016. [44] C. D. Corbin, G. P. Henze, and P. May-Ostendorp, ‘‘A model predictive control optimization environment for real-time commercial building appli- cation,’’ J. Build. Perform. Simul., vol. 6, no. 3, pp. 159–174, 2013. pp g [22] W. Z. Huang, M. Zaheeruddin, and S. H. REFERENCES Cho, ‘‘Dynamic simulation of energy management control functions for HVAC systems in buildings,’’ Energy Convers. Manag., vol. 47, nos. 7–8, pp. 926–943, 2006. [45] D. He, ‘‘Dual-mode nonlinear MPC via terminal control laws with free- parameters,’’ IEEE/CAA J. Autom. Sinica, vol. 4, no. 3, pp. 526–533, Jul. 2017. [23] R. Z. Freire, G. H. C. Oliveira, and N. Mendes, ‘‘Predictive controllers for thermal comfort optimization and energy savings,’’ Energy Buildings, vol. 40, no. 7, pp. 1353–1365, 2008. [46] K. I. Katsigarakis, G. D. Kontes, G. I. Giannakis, and D. V. Rovas, ‘‘Sense-think-act framework for intelligent building energy management,’’ Comput.-Aided Civil Infrastruct. Eng., vol. 31, no. 1, pp. 50–64, 2016. [24] D. Agdas and R. S. Srinivasan, ‘‘Building energy simulation and parallel computing: Opportunities and challenges,’’ in Proc. Winter Simulation Conf., 2014, pp. 3167–3175. [47] M. Casals, M. Gangolells, N. Forcada, M. Macarulla, A. Giretti, and M. Vaccarini, ‘‘SEAM4US: An intelligent energy management sys- tem for underground stations,’’ Appl. Energy, vol. 166, pp. 150–164, Mar. 2016. [25] G. Bianchini, M. Casini, A. Vicino, and D. Zarrilli, ‘‘Demand-response in building heating systems: A model predictive control approach,’’ Appl. Energy, vol. 168, pp. 159–170, Apr. 2016. [48] D. Manjarres, A. Mera, E. Perea, A. Lejarazu, and S. Gil-Lopez, ‘‘An energy-efﬁcient predictive control for HVAC systems applied to tertiary buildings based on regression techniques,’’ Energy Buildings, vol. 152, pp. 409–417, Oct. 2017. [26] J. Figueiredo and J. S. da Costa, ‘‘A SCADA system for energy man- agement in intelligent buildings,’’ Energy Buildings, vol. 49, pp. 85–98, Jun. 2012. [27] J. Široký, F. Oldewurtel, J. Cigler, and S. Prívara, ‘‘Experimental analysis of model predictive control for an energy efﬁcient building heating system,’’ Appl. Energy, vol. 88, no. 9, pp. 3079–3087, 2011. [49] G. D. Kontes, C. Valmaseda, G. I. Giannakis, K. I. Katsigarakis, and D. V. Rovas, ‘‘Intelligent BEMS design using detailed thermal simulation models and surrogate-based stochastic optimization,’’ J. Process Control, vol. 24, no. 6, pp. 846–855, 2014. [28] H. Huang, L. Chen, and E. Hu, ‘‘A new model predictive control scheme for energy and cost savings in commercial buildings: An airport ter- minal building case study,’’ Building Environ., vol. 89, pp. 203–216, Jul. 2015. [50] S. Prívara, J. Cigler, Z. Vána, F. Oldewurtel, C. Sagerschnig, and E. Žáceková, ‘‘Building modeling as a crucial part for building predictive control,’’ Energy Buildings, vol. 56, pp. 8–22, Jan. 2013. [29] J. Ma, J. Qin, T. REFERENCES degree in intel- ligent systems from the Universidad de Granada, in 2004 and 2008, respectively. [64] A. Foucquier, S. Robert, F. Suard, L. Stéphan, and A. Jay, ‘‘State of the art in building modelling and energy performances predic- tion: A review,’’ Renew. Sustain. Energy Rev., vol. 23, pp. 272–288, Jul. 2013. He was a Lecturer with the Applied Artiﬁ- cial Intelligence Group, Universidad Carlos III de Madrid, from 2008 to 2013, and a Research Asso- ciate in the EU FP7 Project Energy IN TIME with the Universidad de Granada, from 2013 to 2017. He was also a Visiting Researcher with the Data Science Institute, Imperial College London, from 2016 to 2017. He has been a Senior Research Fellow with the Computer Science and Artiﬁcial Intelligence Department, Universidad de Granada, since 2016. He has partic- ipated in more than 20 projects in security, ambient intelligence, and energy efﬁciency. His research interests include the use of semantic representation models and machine learning techniques to perform automatic reasoning towards higher-level information fusion. [65] B. Gunay, W. Shen, and G. Newsham, ‘‘Inverse blackbox modeling of the heating and cooling load in ofﬁce buildings,’’ Energy Buildings, vol. 142, pp. 200–210, May 2017. [66] A. Conserva et al., ‘‘Energy in time project: Summary of ﬁnal results,’’ in Proc. 12th Conf. Sustain. Develop. Energy, Water Environ. Syst., 2017, Paper SDEWES2017-0867. [67] Z. Li, Y. Han, and P. Xu, ‘‘Methods for benchmarking building energy consumption against its past or intended performance: An overview,’’ Appl. Energy, vol. 124, pp. 325–334, Jul. 2014. [68] G. F. Lawler and V. Limic, Random Walk: A Modern Introduction. Cambridge, U.K.: Cambridge Univ. Press, 2010. Dr. Gómez-Romero is the Principal Investigator of the projects BIGFUSE: Semantics for Big Data Fusion and Analysis: Improving Energy Efﬁciency in Smart Grids and PROFICIENT: Deep Learning for Energy-Efﬁcient Building Control. [69] R Core Team. (2018). R: A Language and Environment for Statistical Computing. [Online]. Available: http://www.r-project.org/ [70] J. Friedman, T. Hastie, and R. Tibshirani, ‘‘Regularization paths for gen- eralized linear models via coordinate descent,’’ J. Statist. Softw., vol. 33, no. 1, pp. 1–22, 2010. [71] M. S. Al-Homoud, ‘‘Computer-aided building energy analysis tech- niques,’’ Building Environ., vol. 36, no. 4, pp. 421–433, 2001. [72] W. Liang, R. Quinte, X. Jia, and J.-Q. Sun, ‘‘MPC control for improving energy efﬁciency of a building air handler for multi-zone VAVs,’’ Building Environ., vol. 92, pp. 256–268, Oct. 2015. CARLOS J. REFERENCES FERNÁNDEZ-BASSO received the degree in computer science and the M.Sc. degree in data science from the Universidad de Granada, in 2014 and 2015, respectively, where he is cur- rently pursuing the Ph.D. degree in computer sci- ence and energy efﬁciency. [73] B. Kofﬁ, A. Cerutti, M. Duerr, A. Iancu, A. Kona, and G. Janssens-Maenhout, ‘‘CoM default emission factors for the member states of the European union—Version 2017,’’ European Commission, Joint Research Center (JRC), Brussels, Belgium, Tech. Rep., 2017. [74] J. Cigler, P. Tomáško, and J. Široký, ‘‘BuildingLAB: A tool to analyze per- formance of model predictive controllers for buildings,’’ Energy Buildings, vol. 57, pp. 34–41, Feb. 2013. He was a Lead Developer in the EU FP7 Project Energy IN TIME in the topics of building simula- tion and control, data analytics, and machine learn- ing. He also collaborates with the Data Science Institute, Imperial College London, where he has carried out research stays, from 2016 to 2018. He is currently a Research Assistant with the Computer Science and Artiﬁcial Intelligence Department, Universidad de Granada. [75] J. H. Lim, J. T. Kim, S. H. Cho, and G. Y. Yun, ‘‘Development of the adaptive PMV model for improving prediction performances,’’ Energy Buildings, vol. 98, pp. 100–105, Jul. 2015. [76] B. W. Olesen, ‘‘Indoor environmental input parameters for the design and assessment of energy performance of buildings,’’ REHVA J., pp. 17–23, Jan. 2015. [77] T. Ekwevugbe, N. Brown, V. Pakka, and D. Fan, ‘‘Improved occupancy monitoring in non-domestic buildings,’’ Sustain. Cities Soc., vol. 30, pp. 97–107, Apr. 2017. [78] A. Capozzoli, M. S. Piscitelli, A. Gorrino, I. Ballarini, and V. Corrado, ‘‘Data analytics for occupancy pattern learning to reduce the energy con- sumption of HVAC systems in ofﬁce buildings,’’ Sustain. Cities Soc., vol. 35, pp. 191–208, Nov. 2017. M. VICTORIA CAMBRONERO received the degree in industrial engineering with a specializa- tion in construction and industrial facilities and the M.Sc. degree in energy technology for sustainable development from the Polytechnic University of Valencia, in 2008 and 2009, respectively. [79] Q. Meng, M. Mourshed, and S. Wei, ‘‘Going beyond the mean: Distribu- tional degree-day base temperatures for building energy analytics using change point quantile regression,’’ IEEE Access, vol. 6, pp. 39532–39540, 2018. [80] C. Wang, B. Jiao, L. Guo, Z. Tian, J. Niu, and S. Li, ‘‘Robust scheduling of building energy system under uncertainty,’’ Appl. Energy, vol. 167, pp. 366–376, Apr. 2016. REFERENCES Manag., vol. 113, pp. 220–229, Apr. 2016. [35] I. Sharma et al., ‘‘A modeling framework for optimal energy manage- ment of a residential building,’’ Energy Buildings, vol. 130, pp. 55–63, Oct. 2016. [58] P. de Wilde, ‘‘The gap between predicted and measured energy perfor- mance of buildings: A framework for investigation,’’ Autom. Construct., vol. 41, pp. 40–49, May 2014. [36] B. Mayer, M. Killian, and M. Kozek, ‘‘A branch and bound approach for building cooling supply control with hybrid model predictive control,’’ Energy Buildings, vol. 128, pp. 553–566, Sep. 2016. pp y [59] A. C. Menezes, A. Cripps, D. Bouchlaghem, and R. Buswell, ‘‘Predicted vs. actual energy performance of non-domestic buildings: Using post- occupancy evaluation data to reduce the performance gap,’’ Appl. Energy, vol. 97, pp. 355–364, Sep. 2012. [37] S. Salakij, N. Yu, S. Paolucci, and P. Antsaklis, ‘‘Model-based predictive control for building energy management. I: Energy modeling and optimal control,’’ Energy Buildings, vol. 133, pp. 345–358, Dec. 2016. [60] N. Li, Z. Yang, B. Becerik-Gerber, C. Tang, and N. Chen, ‘‘Why is the reliability of building simulation limited as a tool for evaluating energy conservation measures?’’ Appl. Energy, vol. 159, pp. 196–205, Dec. 2015. [38] R. Zafar, A. Mahmood, S. Razzaq, W. Ali, U. Naeem, and K. Shehzad, ‘‘Prosumer based energy management and sharing in smart grid,’’ Renew. Sustain. Energy Rev., vol. 82, pp. 1675–1684, Feb. 2018. [39] T. Bai, S. Li, and Y. Zheng, ‘‘Distributed model predictive control for net- worked plant-wide systems with neighborhood cooperation,’’ IEEE/CAA J. Autom. Sinica, vol. 6, no. 1, pp. 108–117, Jan. 2019. [61] H. Satyavada and S. Baldi, ‘‘An integrated control-oriented modelling for HVAC performance benchmarking,’’ J. Build. Eng., vol. 6, pp. 262–273, Jun. 2016. [40] X. Mi and S. Li, ‘‘Event-triggered MPC design for distributed systems with network communications,’’ IEEE/CAA J. Autom. Sinica, vol. 5, no. 1, pp. 240–250, Jan. 2018. [62] E. Źáčeková, Z. Váña, and J. Cigler, ‘‘Towards the real-life implementa- tion of MPC for an ofﬁce building: Identiﬁcation issues,’’ Appl. Energy, vol. 135, pp. 53–62, Dec. 2014. 38763 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models [63] Z. Afroz, G. M. Shaﬁullah, T. Urmee, and G. Higgins, ‘‘Modeling tech- niques used in building HVAC control systems: A review,’’ Renew. Sustain. Energy Rev., vol. 83, pp. 64–84, Mar. 2018. JUAN GÓMEZ-ROMERO received the degree in computer science and the Ph.D. REFERENCES She was a Project Developer and a Researcher on building management and energy efﬁciency with the Institute of Energy Engineering, Valen- cia, before joining Acciona Infrastructures R&D Department, in 2010. Since 2018, she has been a Project Manager with Acciona Ingeniería. She has a wide experience in research and innovation projects and has collaborated and managed several FP7 and H2020 projects in renewable energies, HVAC systems and storage integration, and control strategies for energy efﬁciency improvement in buildings and districts; e.g., 2DISTRICT, EnergyINTIME, LowUP, Flexynets, FC-DISTRICT, MESSIB, EINSTEIN, CommONEnergy, and COST-EFFECTIVE. She is currently an Industrial Engineer specialized in energy efﬁciency and building simulation. Ms. Cambronero holds a Project Management Professional certiﬁcation granted by the Project Management Institute and a Certiﬁed in Measurement and Veriﬁcation Protocol certiﬁcation granted by the Association of Energy Engineers and Efﬁciency Valuation Organization. [81] H. Thieblemont, F. Haghighat, R. Ooka, and A. Moreau, ‘‘Predictive con- trol strategies based on weather forecast in buildings with energy storage system: A review of the state-of-the art,’’ Energy Buildings, vol. 153, pp. 485–500, Oct. 2017. [82] E. Feller, L. Ramakrishnan, and C. Morin, ‘‘Performance and energy efﬁciency of big data applications in cloud environments: A Hadoop case study,’’ J. Parallel Distrib. Comput., vols. 79–80, pp. 80–89, May 2015. [83] J. Gómez-Romero, M. Molina-Solana, M. Ros, M. D. Ruiz, and M. J. Martin-Bautista, ‘‘Comfort as a Service: A new paradigm for residen- tial environmental quality control,’’ Sustainability, vol. 10, no. 9, p. 3053, Aug. 2018. [84] F. Ferracuti et al., ‘‘Data-driven models for short-term thermal behaviour prediction in real buildings,’’ Appl. Energy, vol. 204, pp. 1375–1387, Oct. 2017. 38764 VOLUME 7, 2019 J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models J. Gómez-Romero et al.: Probabilistic Algorithm for Predictive Control With Full-Complexity Models MIGUEL MOLINA-SOLANA received the degree in computer science and the Ph.D. degree from the Universidad de Granada, in 2007 and 2012, respectively. M. DOLORES RUIZ received the degree in math- ematics and the European Ph.D. degree in com- puter science from the Universidad de Granada, in 2005 and 2010, respectively. From 2012 to 2015, he was a Research Asso- ciate with the Universidad de Granada in the FP7 Project Energy IN TIME. He was a Research Associate on visualization with the Data Science Institute, Imperial College. He is currently a Marie Curie Research Fellow with Imperial College Lon- don, U.K. REFERENCES His research interests include applied work in machine learning and knowledge representation in diverse domains such as music, energy management, and business. She held a non-permanent teaching positions with the Universities of Jaén, Granada, and Cádiz. She has participated in more than ten projects, including the EU FP7 Projects ePOOLICE and Energy IN TIME. She is currently a Research Associate with the Computer Science and Artiﬁ- cial Intelligence Department, Universidad de Granada. Her research interests include data mining, information retrieval, energy efﬁciency, big data, cor- relation statistical measures, sentence quantiﬁcation, and fuzzy sets theory. She has organized several special sessions about Data Mining in inter- national conferences and was part of the organization committee of the FQAS’2013 and SUM’2017 conferences. Dr. Molina-Solana is the Principal Investigator of the H2020 Project DATASOUND–Understanding Data With Sound. Dr. Ruiz belongs to the Approximate Reasoning and Artiﬁcial Intelligence Research Group and the Cybersecurity Lab, Universidad de Granada. She has been the Principal Investigator of the project Exception and anomaly detection by means of fuzzy rules using the RL-theory. Application to fraud detection. MARIA J. MARTIN-BAUTISTA received the degree in computer science and the Ph.D. degree from the Universidad de Granada, in 1996 and 2000, respectively. She has been a Professor with the Computer Sci- ence and Artiﬁcial Intelligent Department, Uni- versidad de Granada, since 2018. She was the Principal Investigator in the FP7 European Project Energy IN TIME with the Universidad de Granada, from 2013 to 2017. She has also been a Principal Investigator of several international and national projects and knowledge transfer contracts with private companies. She has published more than 100 papers in international journals and conferences. Her current research interests include intelligent systems, big data, and knowledge representation with applications to energy, security, and health. Prof. Martin-Bautista is a member of the IEEE Society and the EUSFLAT Society. MARIA J. MARTIN-BAUTISTA received the degree in computer science and the Ph.D. degree from the Universidad de Granada, in 1996 and 2000, respectively. She has been a Professor with the Computer Sci- ence and Artiﬁcial Intelligent Department, Uni- versidad de Granada, since 2018. She was the Principal Investigator in the FP7 European Project Energy IN TIME with the Universidad de Granada, from 2013 to 2017. She has also been a Principal Investigator of several international and national projects and knowledge transfer contracts with private companies. REFERENCES She has published more than 100 papers in international journals and conferences. Her current research interests include intelligent systems, big data, and knowledge representation with applications to energy, security, and health. JESÚS R. CAMPAÑA received the M.Sc. and Ph.D. degrees in computer science from the Uni- versidad de Granada, where he has been a member of the Intelligent Databases and Information Sys- tems Research Group, Department of Computer Science and Artiﬁcial Intelligence, since 2005. From 2013 to 2018, he was a Lecturer with the Universidad de Granada. He has been a member of several research projects related to fuzzy data representation in databases, data mining, text min- ing, and energy efﬁciency, including the FP7 Project Energy IN TIME. His research interests include fuzzy databases, XML, knowledge representation, semantic web, data mining, and text mining. 38765 VOLUME 7, 2019 VOLUME 7, 2019
https://openalex.org/W4387711593	https://jurnal.untirta.ac.id/index.php/JELTS/article/download/21340/11217	English	null	Motivating Mentally Retarded Students by Using the Quizizz Application	Journal of English Language Teaching and Cultural Studies	2,023	cc-by	6,612	Keywords: mentally retarded student, motivation, quizizz application Keywords: mentally retarded student, motivation, quizizz application classroom. Teachers need innovate to improve students' motivation in English language teaching. One option is the Quizizz application, which allows students to complete quizzizes and provide answers in online. This application's appealing design, pleasant backbeat, and real-time design make it harder and more competitive for students. Motivating Mentally Retarded Students by Using the Quizizz Application Tosi Rut Syamsun1, John Pahamzah2, Syafrizal3 1Sekolah Khusus 01 Kota Serang, 2,3 Universitas Sultan Ageng Tirtayasa tosi441@guru.slb.belajar.id Tosi Rut Syamsun1, John Pahamzah2, Syafrizal3 1Sekolah Khusus 01 Kota Serang, 2,3 Universitas Sultan Ageng Tirtayasa tosi441@guru.slb.belajar.id Revised: August 25, 2023 Accepted: October 15, 2023 Revised: August 25, 2023 Revised: August 25, 2023 Submitted: July 27, 2023 Accepted: October 15, 2023 Accepted: October 15, 2023 Submitted: July 27, 2023 Abstract The purpose of this study was to investigate why students in a special needs school who are mentally retarded use the Quizizz application and how teachers utilize the program at special needs school (Sekolah Khusus) 01 Serang city. Case studies made up the research methodology. The instructor and mentally retarded students, a ninth-grade mentally disabled student at Sekolah Khusus Negeri 01 Kota Serang, were watched and interviewed by the researcher. According to the research's results, utilizing Quizizz in the special needs school (SKh) 01 in Serang city began with the instructor introducing or chatting with the ninth-graders about it in order for them to feel more at test using the online assessment technique. The teacher employed Quizizzes extremely useful features in the Quizizz implementation as well. The responses of students toward using Quizizz were based on an interview with mentally retarded students, which revealed that: (1) The majority of students agree that Quizizz is excited motivation, and fun and that Quizizz creates an engaging atmosphere; (2) Quizizz is interesting with lots of useful features; and (3) A competitive surroundings can be created because Quizizz has many difficult features, which encourages the students to be the top scorer on the Quizizz leaderboard. Mentally disabled pupils in English language instruction may benefit from the Quizizz application. Motivation The motivation that grows in students can be a stimulus for students to improve learning outcomes or student achievement, (Kurniati, Sari, & Simaibang, 2023). Motivation in students can improve learning outcomes and achievement. Integrative motivation involves communication with a linguistic community, such as acting, writing scripts, and practicing voice projection. The term "learning motivation" refers to a student's intrinsic motivation and desire to study a foreign language, (Zhu et. al, 2023). Motivation plays a significant role in language acquisition, and understanding learners' uniqueness is crucial for developing effective techniques. In various settings, motivation might influence language acquisition differently. For instance, due to their individual characteristics, students may have varying attitudes toward learning a second language and approaches to doing so in various circumstances (Khatibi & Azam, 2018). This study aims to determine students' motivation for using the Quizizz application for English learning, particularly at special needs schools in Serang City for students who are mentally retarded. The title of the study is Motivating Mentally Retarded Students by Using the Quizizz. The problem in this research: (1) What types of Quizizz are applied by the teacher in the class? (2) How does the teacher apply Quizizz in the class? (3) How does the use of the Quizizz application for students’ motivation in English language teaching for mentally DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 157\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. for English learning, particularly at special needs schools in Serang City for students who are mentally retarded. The title of the study is Motivating Mentally Retarded Students by Using the Quizizz. The problem in this research: (1) What types of Quizizz are applied by the teacher in the class? (2) How does the teacher apply Quizizz in the class? (3) How does the use of the Quizizz application for students’ motivation in English language teaching for mentally their individual characteristics, students may have varying attitudes toward learning a second language and approaches to doing so in various circumstances (Khatibi & Azam, 2018). Quizizz Application An educational program called Quizizz uses an online quiz-like format. The use of the Quizizz application can make teachers more challenged to provide innovation in foreign language teaching (Anwar et al., 2023). This application is also useful in (Fakhruddin & Nurhidayat (2020); Amalia (2020); (Rahmi, 2020). Additionally, students felt that Quizziz is engaging and enjoyable since they cannot cheat throughout the quiz, which fosters a competitive environment and increases the difficulty for the pupils, (Amalia, 2020). However, in this study will be conducted the quizizz application as a learning media for mentally retarded students at special need school in serang city. Previous studies have shown that Quizizz can help students learn English more effectively. Students have positive opinions of the application, seeing it as enjoyable and engaging without putting undue burden on them when completing evaluations. Additionally, students feel that Quizizz is engaging and enjoyable since they cannot cheat throughout the quiz, fostering a competitive environment and increasing difficulty for the students. retarded students at special need schools in Serang city? Introduction Motivation is crucial to a student's success in learning English, as it combines effort and desire to achieve a goal. Teachers should be aware of the importance of motivation in language acquisition, as it is essential for students to be energetic in studying and motivated to study English well. 156\| JELTS Vol. 6 No. 2, 2023 There are some previous studies in this research, from several researches; Quizizz can help students learn English more effectively. Students had three positive opinions of the application. They saw Quizziz as one of the applications that may make learning English enjoyable while without putting an undue load on them when it came to completing the evaluation, DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 156\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. well. Mentally retarded students, such as those at SKh (special school) state in Serang City, often exhibit signs of mental retardation, such as poor executive function, communication skills, and social skills. The students often struggle with language learning due to their slower mental growth and difficulty speaking English in the There are some previous studies in this research, from several researches; Quizizz can help students learn English more effectively. Students had three positive opinions of the application. They saw Quizziz as one of the applications that may make learning English enjoyable while without putting an undue load on them when it came to completing the evaluation, DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 Mentally retarded students, such as those at SKh (special school) state in Serang City, often exhibit signs of mental retardation, such as poor executive function, communication skills, and social skills. The students often struggle with language learning due to their slower mental growth and difficulty speaking English in the DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 157\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. Quizizz Application An educational program called Quizizz uses an online quiz-like format. The use of the Quizizz application can make teachers more challenged to provide innovation in foreign language teaching (Anwar et al., 2023). This application is also useful in DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 157\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 157\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. 157\| JELTS Vol. 6 No. 2, 2023 English. Quizizz used at every level and is suitable for learning a variety of things (Itsnaini et al, 2023). Based on a review by (Yong & Rudolph, 2022), Quizizz includes a lot of helpful features and is usually aesthetically designed. One of these features seeks to lessen students’ anxiety before every test. the game code given to them by their teacher in order to "play" Quizizz. A thorough report on student replies is provided to the teacher when the quiz is finished and may be downloaded and saved. Third, unlike the majority of other free gasification tools, Quizizz includes a number of distinctive characteristics. For instance, Quizizz is made for active learning, not for instructor assistance. Although the teacher dashboard allows teachers to customize learning activities, it also offers superb real-time student updates. Last but not least, there are settings that enable you to randomize the order of the questions, define the time limit and leaderboards, and even offer a summary of the questions and right answers at the conclusion of the quiz. To provide some much-needed fun to quizzes, Quizizz is used. This provides teachers with extremely useful feedback on how well their kids are performing, which is also very significant Mentally Retarded Students Quizizz is an online assessment tool that may be employed in enjoyable multiplayer class activities that allow all students to practice together using their computers, Smartphones, or gadgets. Quizizz has a lot of elements, like an avatar, memes, and music, to make the game more entertaining and fascinating. Many multiple- choice exercises or practices are available in Quizizz as quizzes, allowing game players or students to learn while having fun (Nanda et al., 2018), open assessment materials offer a means of facilitating the extensive changes required to support the growth of IT-enabled assessment throughout the whole range of learning, (Webb & Gibson, 2015). This work is licensed under a Creative Commons Attribution 4.0 International License. DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 and social skills (Sulistyowati & Rukminingsih, 2022). Negeri 01 Kota Serang in studying English by using Quizizz is the goal of this study. The students need some learning media that make them feel enjoy in the classroom. & A major cognitive and developmental disability due to irregularities in a brain structure or function during the course of a person's life is known as mental retardation (Rahmah et al, 2022). The features of students with mental retardation who attend inclusive educational institutions are very different from those of other students (Wiryanto, 2021). Students who are mentally retarded can perform psychomotor tasks like those required of other students and have IQs between 70 and 80, (Nurfitriani & Hidayat, 2020). Students who have mental retardation and attend inclusive schools learn the same material as other students. The study used various data collection methods to collect accurate findings. Observation was conducted in seventh grade at Junior High School SKh Negeri 01 Kota Serang, involving the researcher in the subject's daily activities. The researcher learned about factors affecting students with mild mental retardation's ability to learn English, including the learning principle, teaching methods, media used, teacher roles, obstacles experienced, and students' responses. Interviews were conducted, focusing on the variables influencing the difficulty of learning English for children with mental impairment. Documentation was conducted using letters, memoranda, meeting minutes, diaries, official papers, internal reports, and records from the firm itself. The goal was to bolster the validity of the study findings derived from observations and interviews. DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 Mentally Retarded Students Although the phrase "mental retardation" is a new term for some people, we frequently come across people who exhibit signs of this condition in daily life, such as children who consistently receive poor grades in their academics or individuals who frequently fail to comprehend instructions. If the youngster has mental difficulties, the aforementioned issues could arise. A mentally retarded pupil has a poor executive function, communication skills, There are several ways to put the Quizizz application into practice, Quizizz is a free tool that enables teachers to easily transform educational activities into engaging multiplayer games for students, (Chandler, 2015). This allows instructors to design their own quizzes or utilize ones that have already been made by other educators and work on any device with a browser. No need for a username or password. Second, students only need to browse the Quizizz website and input DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 158\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. Method This study employed a qualitative research. In contrast to experiments, qualitative research used the researcher as a main tool, triangulation in data gathering, and inductive/qualitative data processing, and emphasizes the significance of generalization more (Sugiyono, 2010). This study employed case studies. A case study as research carried out in great detail and depth against a specific organization, institution, or symptom, (Suharsimi, 2010). Case studies were more in-depth and exclusively address a certain topic. An in-depth analysis of the case or issue of the variables influencing the motivation faced by students with mental retardation in junior high school at SKh Based on theoretical investigations, interview guidelines are created. Teachers of students with mental retardation, classroom teachers, companion teachers, and topic teachers are questioned using interviewing techniques to get the essential information. After field data have been gathered and used to address the issue formulation, qualitative research draws its final conclusions. If the evidence gathered during DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 159\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. Observation In the ninth grade at a special needs school in Serang City, the researcher observed on May 9, 2023. As a platform for online testing, Quizizz is employed. The ninth grade of tunagrahita is one of the teacher's classrooms that used Quizizz. IX (tunagrahita) is a class of 5 pupils, 4 of them are male and another is female. In this English class observation, there are two sessions with five students each, and the teacher uses the Quizizz program to administer daily tests in English. The tests include "job/profession" and "things in the classroom". There are seven questions on this test with multiple choice answers, and each one has a 45 second time limit. Using their own device, each student takes the quiz on Quizizz once. (Tunagrahita). Salutations, an absence, and a prayer were the opening. The instructor gets ready to present the content after that. The test was administered by the instructor using the Quizizz app after the lecture. Through instructor-paced lessons, the teacher kept an eye on the class while they completed the exam; the speed was set by the teacher, and everyone in the class worked through each question simultaneously. Student rankings and scores are displayed in instructor-paced lessons. Once they had finished the daily exam, the results showed up on the pupils' screens. The teacher then discusses the daily exam question with the class after evaluating it. After concluding the lesson with a last task, the teacher extends pleasantries. When the researcher is present in the classroom, there are a few students' devices that have issues logging into the Quizizz program and an unreliable signal. The pupils are hampered by these challenges when they are taking the test. the study is used to support the conclusions, they are said to be valid. Question: “How is the Quizizz using in English class for mentally retarded students?” Question: “How is the Quizizz using in English class for mentally retarded students?” Question: Why do you use Quizizz application for learning English learning and test? Question: Why do you use Quizizz application for learning English learning and test? Answer: “The most important details in this text are the socialization and trial of Quizizz, the preparation for the teacher, and the use of Quizizz in a class. The teacher opens the class with greeting, absence, and prayer, prepares tools like an LCD projector and laptop, shares the Quizizz link, begins the test, monitors the students, discusses the question test together, and closes the class with salam.” Answer: “Quizizz is an online platform that was chosen by a teacher at Special need school (SKh) 01 Serang city for assessment. It provides live scores for tests, allowing students to know who has the highest score and which class the student came from. It is more monitorable than Google Form, as it allows for immediate feedback on scores and rankings. Quizizz also helps students stay motivated to study English.” Quizizz was chosen because of its characteristics that aid the instructor in evaluating the pupils since the teacher said that the students frequently get bored when taking a test with Google Classroom. Students Paced Lessons is one of the initial features. With this feature, both students and teachers may keep track of their progress at their own speed. Each question or lesson includes a scoreboard and a real-time result. The second feature is reports, which give the teacher in-depth information about the class and individual student for every quiz so they can track the students' development. Quizizz's design is also game-based and includes aspects from games like avatars, music, scoreboards, etc. When taking an online English test, the pupils are kept from getting bored by these features. Using Quizizz's Student Paced Lessons capabilities, the instructor may monitor the pupils based on her observations. Students appreciate the exam in class because it has game-based components that appear after they respond to a question. The researchers observed Quizizz being used in English class at IX (Tunagrahita) and the teacher's comment, thus this is how the lesson started. Salutations, absences, and prayers were used as the opening. Interview with the teacher Table 1. Result of Observation Items Yes No Well prepared from teacher for class ✓ Well prepared from teacher for test ✓ Used Quizizz application from teacher for test ✓ The teacher controlled the class via instructor paced lesson ✓ The students enjoy using Quizizz ✓ The teacher evaluates the daily test ✓ Table 1. Result of Observation Table 1. Result of Observation The researcher also conducted interview with the English teacher in special need school in Serang City because the researcher wants to gain deep information about the using of Quizizz application in English language teaching at the ninth grade of special need school in Serang City. The researcher did the interview on May, 19th 2023 in special need school with six questions. The teacher’s opening marked the start of teaching and learning activities in Grade IX 160\| JELTS Vol. 6 No. 2, 2023 160\| JELTS Vol. 6 No. 2, 2023 DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 This work is licensed under a Creative C DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 text. I must thus enter the most text possible on Quizizz. I grabbed a screenshot, but it's a little tricky because you have to manually enter the text on Quizizz for questions like "recount" and "narratives" that require screenshots of everything.” about this. Some people object, saying things like "Miss. I didn't finish because the time is so fast," and similar things, because they occasionally forget or concentrate solely on the question. In response to their protests, I promised to create the solution key if the question involved an essay. For instance, even if the students provided the proper response while I capitalized or punctuated the answer in the answer key, the Quizizz response was incorrect since they failed to do so.” Question: “Are all English materials appropriate with the Quizizz application?” Answer: “English literature in the ninth grade mostly uses conditional phrases and the perfect tense, changing only from the positive to the negative to the interrogative. I thus believe that Quizizz can get all contents. For narrative texts, it varies according to the text. I must thus enter the most text possible on Quizizz. I grabbed a screenshot, but it's a little tricky because you have to manually enter the text on Quizizz for questions like "recount" and "narratives" that require screenshots of everything.” Miscommunication between the teacher and the students is a barrier to utilizing Quizizz. This misconception relates to how long a test is supposed to take. Quizizz's system is the other impediment. The appropriate answer for the essay test must match the answer key that the teacher created, which must have capital letters or punctuation, as well as the Quizizz system. The kids may accurately complete the essay test; however, they omitted capitalization or punctuation. According to the researcher's findings, there are certain challenges when utilizing Quizizz. In English classes, the first challenge is that some students have poor or unstable signals while using Quizizz for daily assessments. The second challenge is that some pupils' technology is buggy and sluggish. As a result, when a student has a problem with their gadget, the teacher typically allots them some more time. Question: In your opinion, what are the advantages and disadvantages of using the Quizizz application? The instructor then prepares a few resources, including an LCD projector for the class and a laptop to watch the pupils while they take the quiz. Each student completes the daily test on Quizizz using their own devices after the online daily test has begun. The teacher keeps an eye on the pupils during the daily tests as the class works through each question at the teacher's pace in instructor-paced lessons. The score and ranking of the students are displayed in instructor-paced lessons. There is a scoreboard and an actual result for each exercise or lesson. Question: What are the obstacles when using the Quizizz application English class at IX (Tunagrahita)? Answer: “The issue is that, despite my warnings to Quizizz that they have 15 seconds for each question throughout the exam and that if they don't use them, the score would be zero, they occasionally forget DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 161\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. 161\| JELTS Vol. 6 No. 2, 2023 DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 162\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. Students 3: “Yes of course, using Quizizz makes me more understand about English language.” Most participants agreed that Quizizz is an exciting platform to learn English. They provide several reasons why Quizizz is engaging, such as the picture offered by Quizizz and students can more motivate in answering the question because they could see ranking in their own devices. Student 1: "Because I am interested in using device to answer some quizess, such as there are many colors and pictures in the quiz. We can see the ranking.” Question: In your opinion, what are the advantages and disadvantages of using the Quizizz application? Answer: “Quizizz has the advantages of not requiring test correction and of being able to keep track of test takers' arrival times, delays, and absences. The drawbacks, however, include the possibility that some students will resume the exam, miss it, and request a retake, which will need the creation of new questions by the teacher for the retest.” The benefit is that the exam does not need to be corrected by the teacher because Quizizz contains a summary of each student's scores in its report function. The Quizizz feature known as Students' Paced Lessons allows the teacher to keep track of each student's progress while they are taking a test. Each question or lesson contains a scoreboard and a real-time result. This feature, known as Students' Paced Lesson, Question: “Are all English materials appropriate with the Quizizz application?” Answer: “English literature in the ninth grade mostly uses conditional phrases and the perfect tense, changing only from the positive to the negative to the interrogative. I thus believe that Quizizz can get all contents. For narrative texts, it varies according to the 162\| JELTS Vol. 6 No. 2, 2023 Student 3: “I think this is a really nice to use it as a learning media because it gives us motivation in English learning” demonstrates students' progress at their own speed. The teacher may view live student activity through this tool. Based on the responses, the participants agreed that Quizizz is a helpful, accessible platform and can be used as a tool for learning English because they claimed that Quizizz not only offer exciting features to enhance students’ motivation. On the other hand, Quizizz also motivate them in English learning because any nice picture and music when they use quizizz application on their own device. Question: “How was the condition of the face to face learning and study from home classes when you applied Quizizz?” Answer: “Face to face learning is preferable; however, if instructions are given through a Zoom meeting, direct communication may be dictated. There is also a problem with the voice, as we are unable to hear anything due to the unreliable connection, and some pupils failed to pay attention to the lesson. Unquestionably superior to learning at home is study from home.” Student 1: “I think Quizizz is an interesting media to learn English because there are many pictures, it can motivate us to know more about English language.” Face to face learning lessons are in better condition because the instructor can easily lead the students in a face-to-face setting, and the students can readily comprehend the instructor when taking an online test. Online courses have signal issues, which are problematic. Some students experience signal issues with their devices, making it difficult for them to hear the teacher when she instructs them about the online test during a Zoom conference. In this situation, the teacher will need to repeat the instructions more than once. If a student is having trouble with an online test during an offline lesson, they can speak with the teacher immediately. Student 2: “I think it might be interesting if we learn our English language using Quizizz by answering the quiz and see the ranking.” Students 3: “Yes of course, using Quizizz makes me more understand about English language.” Student 3: “Because there are many nice pictures and feature who trigger us to answer the right option. In last meeting I got second position.” link in LCD projector. Each student takes the quiz using Quizizz on their own devices after the test has begun. The teacher keeps an eye on the students while they complete daily tests through instructor paced lessons. The speed is set by the teacher, and the whole class goes through each question together. The score and ranking of the students are displayed in instructor-paced lessons. When the daily exam is over, the results will show up on the students' screens. The teacher then discusses and examines the daily test question with the class. As the class comes to a conclusion after the last exercise, the teacher extends pleasantries. This question intended to understand why the participants chose Quizizz as a learning media. The result shows that the participants chose Quizizz is because Quizizz are easy to use, accessible, various picture are provided on Quizizz. Furthermore, they more motivate to get the highest score to make them get first position in ranking table. The score and ranking of the students are displayed in instructor-paced lessons. When the daily exam is over, the results will show up on the students' screens. The teacher then discusses and examines the daily test question with the class. As the class comes to a conclusion after the last exercise, the teacher extends pleasantries. In order to teach and learn English at a special needs school city's ninth grade, the study's primary goal is to illustrate how teachers may persuade mentally retarded students to utilize the Quizizz application. After receiving the data's results, the researcher discusses them in this section. Following data analysis, the researcher was able to determine how the ninth-grade teacher at the special needs school utilized the Quizziz application in English. Graphic 1. Result of Quizizz Test 0 2000 4000 6000 8000 10000 Skor Result of Test NA FP RR MZP PY DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 By observing classes and speaking with the English instructor, the researcher discovered that Quizizz was being used to teach English to children with mental disabilities in the ninth grade at the special needs school (SKh). According to what the researcher saw during classroom observations, teaching and learning activities in IX (Tunagrahita) start with an introduction from the instructor. Interview with students Interview with students Student 1: "With music and interesting picture, it is good application for English learning." Student 2: "Because from Quizizz, I can try hard to get first position and keep me motivated." Student 2: “In my opinion, using Quizizz is useful to learn English language” 163\| JELTS Vol. 6 No. 2, 2023 163\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. Student 3: “Because there are many nice pictures and feature who trigger us to answer the right option. In last meeting I got second position.” link in LCD projector. Each student takes the quiz using Quizizz on their own devices after the test has begun. The teacher keeps an eye Student 3: “Because there are many nice pictures and feature who trigger us to answer the right option. In last meeting I got second position.” claimed that Quizizz offers exciting imagery to enhance students’ motivation, as well as motivating them in English learning by providing nice pictures and music when they use the application on their own device. This question was intended to understand why the participants chose Quizizz as a learning media. The results showed that the participants chose Quizizz because it is easy to use, accessible, various picture are provided on Quizizz, and they are more motivated to get the highest score to make them get first position in the ranking table. how to respond to questions and how to use Quizizz, the teacher did not only engage in socialization but also administers a quiz. According to experts, this instructor action is appropriate for pupils who are inexperienced with online testing procedures. Students need training in order to become comfortable with online assessment method or online assessment process, (Fakhruddin & Nurhidayat, 2020). Picture 1. Result of Quizizz Test Based on the result from researcher observation and interviewed with the teacher and students. The students get motivated when the teacher will start test using Quizizz application, some students said “miss we want playing Quizizz”. When they start learning new things, students are always very motivated to learn, (Wu & Tao, 2022). Then the teacher opened the Quizizz application and students opened their mobile phone to open quizizz application. They are excited to join the quizizz, one of student had trouble with his mobile phone and then he called the teacher to check his mobile phone “miss, my phone cannot open the application. May you help me?” The teacher came to the student and tried to check his phone “okay, now you can open the quizizz and join the quiz.” The student felt happy because he can join the quiz like his friends. After all of students joined quiz, the teacher starts the quiz and show the first question. “Okay students, try Picture 1. Result of Quizizz Test Quizizz's benefits and features are related to the claim made by Webb and Gibson that digital media enhanced assessment includes a genuine learning experience with digital media, continuous and unobtrusive performance, learning and measurement of knowledge, detail and high- resolution data record analyzed and displayed in computational ways, and also allows teachers and students to access in real-time, (Webb & Gibson, 2015). This work is licensed under a Creative Commons Attribution 4.0 International License. Greeting, absence, and prayer make up the opening. The teacher then uses Quizlet to help them study for the exam, and from their laptops, they share the test Graphic 1. Result of Quizizz Test DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 164\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. g , discovered that Quizizz was being used to teach English to children with mental disabilities in the ninth grade at the special needs school (SKh). According to what the researcher saw during classroom observations, teaching and learning activities in IX (Tunagrahita) start with an introduction from the instructor. Greeting, absence, and prayer make up the opening. The teacher then uses Quizlet to help them study for the exam, and from their laptops, they share the test Through an interview with an English instructor at the special needs school, the researcher discovered the usage of the Quizizz program for pupils who are mentally retarded in the ninth grade. According to the results of an interview with an English instructor, the teacher is introducing Quizizz to the ninth-grade pupils and showing them how to use it before utilizing it in IX (Tunagrahita). To help the students learn 164\| JELTS Vol. 6 No. 2, 2023 Through an interview with an English instructor at the special needs school, the researcher discovered the usage of the Quizizz program for pupils who are mentally retarded in the ninth grade. According to the results of an interview with an English instructor, the teacher is introducing Quizizz to the ninth-grade pupils and showing them how to use it before utilizing it in IX (Tunagrahita). To help the students learn students always ask the teacher to play quiz in the classroom. Quizizz application made students more excited and motivated to learning English in the classroom because students will get achievement in ranking table after quiz finished, (Kurniati, Sari, & Simaibang, 2023). The result of this study is conforming to previous study, the students felt excited when the teacher did test using Quizizz application in English learning. answer this question with choose one of some options.” All of students had answered the question and the result showed on the monitor. Some students answered the right answer and they were feeling happy with that, “yeeee, I am right”. Then the ranking had showed up, one of student who got first position feeling happy and proud “wow, I got first position”. Other students feeling motivated to get first position and they cannot wait to next question, “miss, next question please”. The teacher moves to next question “okay, be ready for next question”. Move to next question all of students focus on the question and answer. They want to get right answer to boost their point and move their position higher than before”. After all of students answered the question, the result showed up on monitor. Some students got right answer and they are feeling happy with that “yees, I am right”, “yees, my answer is right again “. The ranking table showed up on monitor and the position had change from before. One of student who got second position now move up to first position and she feel happy and proud, “wow, I move up to first position”. The quiz runs for 15 minutes and after all of question had answered by students, the final ranking showed up on monitor. The student who got first position is very happy and proud of her work, “yeees, I am the champion of this quiz”. Other students feel sad and excited for the next quiz. Now in every meeting, DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 The participants agreed that Quizizz is a helpful, accessible platform and can be used as a tool for learning English. They 165\| JELTS Vol. 6 No. 2, 2023 This work is licensed under a Creative Commons Attribution 4.0 International License. DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 This work is licensed under a Creative Commons Attribution 4.0 International License. Conclusion The research found that teachers use Quizizz applications to evaluate students' comprehension in English. The system includes essay and multiple-choice quizzes, with the correct answer key being the teachers. The teaching process begins with an opening and online test with the teacher monitoring students through instructor pace lessons. However, obstacles like misunderstandings, open responses, and bad signals can hinder the process. Quizizzes use at SKh 01 Serang city has made students motivated and happy, especially for mentally retarded students. The application is easy to use, fun, and can stimulate interest and excitement in learning English. The ranking function helps students improve their performance and reduces anxiety. Overall, Quizizz is a valuable tool for improving English learning for mentally retarded students. The teacher has to manage the online test better include preparing the test for open 166\| JELTS Vol. 6 No. 2, 2023 his work is licensed under a Creative Commons Attribution 4.0 International License. responses or essay and control the students so that the students do not feel fatigued when doing online test. The teacher also has to improve her creativity to find the other innovative platform for the students’ assessment in future. Special need school (SKh) 01 Serang city has to provide the proper internet connection to support the teaching and learning process. The good quality internet connection can facilitate the online test more effective. For the future researcher, they can conduct the similar research but they must add the other respondents of teacher in order to see more using of Quizizz as an online assessment platform. The future researcher also can implement classroom action research to conduct the similar research. Fakhruddin & Nurhidayat. (2020). Students’ perception on quizizz as game-based learning in learning grammar in written discourse. Wiralodra English Journal (WEJ). Vol. 4 No. 2. Itsnaini, N. S., Asmara, C. H., & Pongpan, K. (2023, January). Students' perspectives on using" quizizz" apps to assess learning. In Journal Universitas Muhammadiyah Gresik Engineering, Social Science, and Health International Conference (UMGESHIC) (Vol. 2, No. 1, pp. 248- 254). Kurniati, S, & Simaibang. (2023). Correlation among the eighth grade students' learning motivation, parents' involvement, and English speaking skill at SMP IT Ishlahul Ummah Prabumulih. Jurnal Esteem Prodi Pendidikan Bahasa Inggris. Vol. 6 No. 1. Nanda, S. R., Abdul, N. B., & Daddi, H. (2018). The use of quizizz application in improving students’ reading comprehension skill At SMKN 3 Takalar. Conclusion Journal of Computer Interaction in Education, 1(2), 173– 182. DOI: http://dx.doi.org/10.48181/jelts.v6i2.21340 167\| JELTS Vol. 6 No. 2 This work is licensed under a Creative Commons Attribution 4.0 International License. References Anwar, M. S., Hilmi, D., Rachmawaty, R., & Hidayat, A. R. (2023). The effect of Hots-based Qira'ah learning evaluation with quizizz application on students’ learning outcomes (Experimental research at markaz arabiyah course institutions). Al-Fusha: Arabic Language Education Journal, 5(1), 1- 9. Nurfitriani, R., & Hidayat, M. A. (2020). Student management strategy for tunagrahita abk students in the inclusion class. Jurnal Pendidikan Guru Madrasah Ibtidaiyah, 4(2), 79- 92. Arikunto, S. (2010). Dasar-Dasar Evaluasi Pendidikan. Jakarta: Bumi Aksara. Sulistyowati, H., & Rukminingsih, R. (2022, January). Language acquisition of a mentally retarded student at SDLB of Tunas Harapan: Psycholinguistics Study. In 2nd International Conference on Education and Technology (ICETECH 2021) (pp. 53- 58). Atlantis Press. De Silva, A. D. A., Khatibi, A., & Ferdous Azam, S. M. (2018). Do the demographic differences manifest in motivation to learn science and impact on science performance? Evidence from Sri Lanka. International Journal of Science and Mathematics Education, 16, 47-67. Rahmah, A., Rouns, E., & Luck, A. (2022). The effect of self-development program for improving independence in defective students in SLB N 1 Lima 167\| JELTS Vol. 6 No. 2, 2023 Kaum Batusangkar. World Psychology, 1(2), 46-53. Wu & Tao. (2022). Motivation and grit affects undergraduate students' english language performance. European Journal of Educational Research, v11 n2 p781-794. Sugiyono. (2010). Penelitian kuantitatif, kualitatif dan R&D. Bandung: Alfabeta. Yong, A., & Rudolph, J. (2022). A review of Quizizz–a gamified student response system. Journal of Applied Learning and Teaching, 5(1), 146-155. Webb, M. E., & Gibson, D. C. (2015). Technology enhanced assessment in complex collaborative settings. Education and Information Technologies, 20 (4), 675–695. Zhu, L., Peng, P., Lu, Z., & Tian, Y. (2023). Metavim: Meta variationally intrinsic motivated reinforcement learning for decentralized traffic signal control. IEEE Transactions on Knowledge and Data Engineering. Wiryanto, T. Influence of Motivation to Natural Sciences Mental Retardation Students on Learning Outcomes in Schools Organize Inclusive Education. Journal of Disability, 1(1), 37-44. tp://dx.doi.org/10.48181/jelts.v6i2.21340 168\| JELTS Vol. 6 No This work is licensed under a Creative Commons Attribution 4.0 International License. 168\| JELTS Vol. 6 No. 2, 2023
https://openalex.org/W2107346733	https://www.animbiosci.org/upload/pdf/ajas-26-6-788-5.pdf	English	null	Comparison of α1-Antitrypsin, α1-Acid Glycoprotein, Fibrinogen and NOx as Indicator of Subclinical Mastitis in Riverine Buffalo (&lt;italic&gt;Bubalus bubalis&lt;/italic&gt;)	Asian-Australasian journal of animal sciences	2,013	cc-by	4,943	* Corresponding Author: Anirban Guha. Tel: +919836341948, E-mail: archies76@gmail.com 1 Dept. of Biochemistry, College of Basic Sciences, Chaudhary Charan Singh, Haryana Agricultural University, Hisar-125004, Haryana, India. 2 Veterinary Physiology and Biochemistry, Lala Lajpat Rai University of Veterinary and Animal Sciences (erstwhile. Chaudhary Charan Singh Haryana Agricultural University), Hisar - 125004, Haryana, India. Submitted May 8, 2012; Accepted Jul. 17, 2012; Revised Aug. 17, 2012 788 788 Asian Australas. J. Anim. Sci. Vol. 26, No. 6 : 788-794 June 2013 http://dx.doi.org/10.5713/ajas.2012.12261 www.ajas.info pISSN 1011-2367 eISSN 1976-5517 Comparison of 1-Antitrypsin, 1-Acid Glycoprotein, Fibrinogen and NOx as Indicator of Subclinical Mastitis in Riverine Buffalo (Bubalus bubalis) Anirban Guha, Ruby Guha1 and Sandeep Gera2 Dept. of Animal Resource Development and Animal Husbandry, Govt. of West Bengal, India ABSTRACT: Mastitis set apart as clinical and sub clinical is a disease complex of dairy cattle, with sub clinical being the most important economically. Of late, laboratories showed interest in developing biochemical markers to diagnose sub clinical mastitis (SCM) in herds. Many workers reported noteworthy alternation of acute phase proteins (APPs) and nitric oxide, (measured as nitrate+nitrite = NOx) in milk due to intra-mammary inflammation. But, the literature on validation of these parameters as indicators of SCM, particularly in riverine milch buffalo (Bubalus bubalis) milk is inadequate. Hence, the present study focused on comparing several APPs viz. 1- anti trypsin, 1- acid glycoprotein, fibrinogen and NOx as indicators of SCM in buffalo milk. These components in milk were estimated using standardized analytical protocols. Somatic cell count (SCC) was done microscopically. Microbial culture was done on 5% ovine blood agar. Of the 776 buffaloes (3,096 quarters) sampled, only 347 buffaloes comprising 496 quarters were found positive for SCM i.e. milk culture showed growth in blood agar with SCC2105 cells/ml of milk. The cultural examination revealed Gram positive bacteria as the most prevalent etiological agent. It was observed that 1- anti trypsin and NOx had a highly significant (p<0.01) increase in SCM milk, whereas, the increase of 1- acid glycoprotein in infected milk was significant (p<0.05). Fibrinogen was below detection level in both healthy and SCM milk. The percent sensitivity, specificity and accuracy, predictive values and likelihood ratios were calculated taking bacterial culture examination and SCC2105 cells/ml of milk as the benchmark. Udder profile correlation coefficient was also used. Allowing for statistical and epidemiological analysis, it was concluded that 1- anti trypsin indicates SCM irrespective of etiology, whereas 1- acid glycoprotein better diagnosed SCM caused by gram positive bacteria. NOx did not prove to be a good indicator of SCM. It is recommended measuring both 1- anti trypsin and 1- acid glycoprotein in milk to diagnose SCM in buffalo irrespective of etiology. (Key Words: Acute Phase Proteins, Nitric Oxide, Subclinical Mastitis and Buffalo) Copyright © 2013 by Asian-Australasian Journal of Animal Sciences Somatic cell count Macrophages, a somatic cell fraction of milk, are the source of nitric oxide (NOx) in bovines. In intra-mammary infection (IMI), macrophages are the initially predominant cell type to travel from the peripheral circulation to the mammary gland in response to inflammatory insults and contribute to the pathophysiology of the mammary gland. NOx is produced in large amounts by inducible nitric oxide synthase (iNOS) and its derivatives, such as peroxynitrite and nitrogen dioxide, and plays a role in inflammation (De and Mukherjee, 2009). The somatic cell count (SCC) of the milk samples was determined microscopically (Gera and Guha, 2011). Following through mixing, a 4 mm diameter platinum loop was used to evenly spread 0.01 ml of milk over four 1.0 cm2 area template outlines. Slides were stained for 30 s in Newman-Lampert stain, with the composition as follows: Methylene blue 1.2 gm. 95% ethyl alcohol 54 ml. Tetrachloroethane 40 ml. Glacial acetic acid 6 ml. Methylene blue 1.2 gm. 95% ethyl alcohol 54 ml. Tetrachloroethane 40 ml. Glacial acetic acid 6 ml. The diagnostics based on physical and chemical changes in SCM milk is not satisfactory. A confirmatory diagnosis of SCM according to International Dairy Federation (IDF) recommendations is based on the microbiological status and inflammatory reactions i.e., somatic cell count (SCC2105 cells/ml of milk) of the quarter. However, the logistic and financial considerations involved with sampling all animals in a herd have precluded these techniques from being widely adopted (Guha et al., 2010). One of the principles of detecting inflammation within the mammary gland is to study the mammary epithelial integrity (Gera and Guha, 2011). For this reason alternative parameters to indicate inflammation are used to identify trends in the development of the udder health in dairy herd (Guha et al., 2010). Several superior breeds of milch buffaloes are being developed on the Indian sub- continent where buffaloes are foremost dairy animal. Thus, the present study was undertaken to investigate the effectiveness of the aforesaid APPs and NOx in detecting SCM and recognizing them as indicators for bubaline SCM for further development of kit for diagnosing SCM in herds. In the present study their concentration in healthy and SCM milk was analyzed both statistically and epidemiologically and further correlated with Log10SCC. Somatic cells were stained with deep blue nuclei against a light blue background. Estimation of fibrinogen The fibrinogen was estimated by the tyrosine method as described by Varley et al. (1980). The fibrinogen was precipitated with calcium in casein free skimmed milk samples. The blue coloured complex developed due to the reduction of phosphomolybdate and phosphotungstate by tyrosine residues of polypeptide was estimated spectrophotometrically at 680 nm. INTRODUCTION hygienic status. Milk production involves rapid physical, chemical and biological changes right from galactopoiesis to let down. Mastitis, complex multi-factorial inflammatory reaction, which often results from an intra-mammary bacterial infection entails losses due to reduced milk production, treatment costs, increased labor, milk withheld for human consumption due to residues in the form of antibiotics and micro-organisms and pre-mature culling. Consequently, an early detection at the sub clinical stage is necessary to prevent production loss and to enhance prospects of recovery (Guha et al., 2010). Riverine buffalo milk production in the Indian sub continent has long been accepted as the backbone of the rain-fed agrarian socio-economic fabric. Sustainability of buffalo milk production even during dry spells has contributed to a lower suicide rate amongst farmers in drought stricken terrain. The quality of milk lies in its Subclinical mastitis, a herd problem, affects the normal functioning of the mammary gland epithelial cells’ ability to convert circulating nutrients into milk components (Gera and Guha, 2012). It often goes unnoticed due to absence of visually apparent changes in udder and milk. Detection of SCM is also difficult due to pooling of milk for sale from different milk collection points so that the source of SCM Guha et al. (2013) Asian Australas. J. Anim. Sci. 26:788-794 789 Bacterial culture examination The milk samples collected aseptically were shaken thoroughly. A 4 mm diameter platinum loop was used to streak 0.01 ml of the sample on 5% ovine blood agar plates. The plates were incubated aerobically at 37C for 24 h. The resulting growth from the respective plates of media was purified and identified on the basis of morphology, colony characteristics and Gram’s reaction (Gera and Guha, 2011). Somatic cell count The working factor of the microscope was calculated to be 35,400 by using a stage micrometer, calculating diameter of the microscopic field (0.012 cm) and the field per square cm (8850) for the given microscope. Total no. of cells was obtained by multiplying the total no. of cells counted in 25 fields with the working factor. MATERIALS AND METHODS cannot be determined after collection (Gera et al., 2011; Guha et al., 2012). Subclincial mastitis is also a depot of micro-organisms that lead to the spread of infection to the other animals within the herd. Collection of milk samples Fifty ml of milk samples each were collected from 776 Murrah buffaloes over 3096 quarters under aseptic condition in sterile containers. The quarters were marked as right-fore (RF), right-hind (RH), left-fore (LF) and left-hind (LH). Acute phase proteins (APPs) are an assortment of blood hepatic glycoproteins that change in concentration due to external or internal challenges, such as infection, inflammation, surgical trauma, or stress. Quantification of APP concentration in body fluids can provide valuable diagnostic information in the detection, prognosis, and monitoring of disease in several animal species (Gonzalez et al., 2008). The recent recognition, that APPs are produced in the bovine mammary gland in response to bacterial mastitis has made it obligatory to consider them as alternative biomarkers for mastitis. An increase in concentration of APPs precedes the onset of clinical signs even in the absence of macroscopic changes in the ruminant milk (Safi et al., 2009). Estimation of NOx The NOx (nitrate+nitrite) was estimated by Griess reaction as described by Bouchard et al. (1999). Nitrate was converted to nitrite by nitrate reductase. The acidified nitrite produced nitrosating agent which reacted with sulfanilic acid to produce diazonium ions. The diazonium ions coupled with N (1-naphthyl) ethylenediamine to form choromophoric azo-dye whose intensity was measured spectrophotometrically at 550 nm. Etio-prevalence of SCM In the present study, on the basis of bacterial culture examination and SCC it was observed that 347 riverine buffaloes (496 quarters) were SCM positive. Milk samples showing SCC2105 cells/ml and growth in culture media were considered positive for SCM. The SCC was observed to increase significantly (p<0.01) in SCM milk irrespective of the etiological agents (Tables 2 and 3). The mean SCC in SCM milk was 2.050.056 (Table 3). From Table 1 it is evident that the most prevalent etiological agent was Staphylococcus spp. followed by Streptococcus spp. And Escherichia coli. A few instances of mixed infection and (Corynaebacterium spp. and Bacillus spp.) were also encountered during the investigation (Table 1). Together the frequencies of Gram positive infections were >79%. Estimation of 1- antitrypsin The 1-anti trypsin was measured by the Benzoyl arginine p-nitroanilide (BAPNA) method as described by Fritz et al. (1974), with little modification as described by Guha and Gera (2011). Casein and fat were removed by clearing solutions containing rennet and polyethylene glycol. The trypsin residue formed a yellow coloured complex 4 nitroaniline after reacting with BAPNA which was measure spectrophotometrically at 405 nm. The intensity of colour was inversely proportional to 1-anti trypsin concentration. Statistical analysis glycoprotein protein was finally precipitated with phosphotungstic acid. The tyrosine content of the precipitate was estimated by the above mentioned procedure. Analysis of variance (ANOVA) and Duncan Multiple Range Test (DMRT) were carried out to compare the milk components. Comparison of means of estimated concentration of different parameters in healthy and SCM milk, irrespective of the etiology, was done by t-test. SCC was converted to Log10SCC. Pearson’s correlation coefficient among milk components showing substantial alternation in concentration between healthy and SCM milk samples including Log10SCC was also calculated. All statistical analysis was done with SPSS statistical software (Petrie and Watson, 2008). Calculation of percent sensitivity, specificity, accuracy, predictive values and likelihood ratios Percent sensitivity, specificity, accuracy, predictive values and likelihood ratios were found taking bacterial growth in culture media and SCC2105 cells/ml of milk as the benchmark (Katsoulos et al., 2010; Guha et al., 2010; Gera and Guha, 2011). The cut-off values for each significantly altered parameter were obtained from Receiver Operator Characteristic (ROC) analysis curve with the aid of the MedCalc software. The percent sensitivity, specificity were calculated by the formulae of Thrusfield (2005). The percent accuracy was calculated by the formula of Reddy et al. (2001). Percent positive and negative predictive values, likelihood ratios (both positive and negative) were also measured by the methods of Petrie and Watson (2008). Estimation of 1- acid glycoprotein The 1- acid glycoprotein protein was estimated by the tyrosine method as described by Varley et al. (1980). Casein of skimmed milk was removed by acid precipitation and the heat coagulable protein by percholoric acid. The 1- acid Guha et al. (2013) Asian Australas. J. Anim. Sci. 26:788-794 790 Effect of SCM on milk components In the present study there was a statistically significant (p<0.01) increase in the concentration of 1- anti trypsin in infected milk samples irrespective of the causative agents. A significant (p<0.05) increase of 1- acid glycoprotein concentration in the SCM milk was also observed. Fibrinogen was below detection levels in both healthy and infected milk samples. NOx also showed significant increase in SCM milk (Tables 2 and 3). Percent sensitivity, specificity, accuracy, predictive Table 1. Prevalence of bacterial agents in subclinical mastitis milk of riverine buffalo (Bubalus bubalis) Genus Animal Quarters Number Percentage Number Percentage Staphylococcus spp. 146 42.07 210 42.33 Streptococcus spp. 118 34.00 172 34.67 Escherichia coli 69 19.88 95 19.15 Others (Corynaebacterium spp. and Bacillus spp.) +Mixed infection 14 4.05 19 3.85 Total 347 100 496 100 Guha et al. (2013) Asian Australas. J. Anim. Sci. 26:788-794 791 Table 2. Effect of different bacterial agents on acute phase proteins and NOx in healthy and subclinical mastitis milk of riverine buffalo (Bubalus bubalis) Parameters MeanSE of healthy milk (n = 496) MeanSE of subclinical mastitis milk (n = 496) Staphylococcus spp. (n = 210) Sterptococcus spp. (n= 172) Escherichia coli (n = 95) Others+mixed infection (n = 19) Somatic cell count (105 cells/ml) 0.930.007 2.080.020 2.050.013 2.010.015 2.030.012 1-Anti trypsin (U/L) 3,340.43138.54 6,982.13122.72 6,074.22102.87 5,798.7694.78 5,998.7399.34 α1-Acid glycoprotein (mg/ml) 0.176a0.026 3.08b0.02 2.99b0.01 2.88b0.08 2.89b0.03 Fibrinogen (g/dl) Not detectable Not detectable Not detectable Not detectable Not detectable NOx (Nitrate+nitrite) (M) 11.090.19 18.070.14 17.090.11 16.280.14 16.890.24 Mean having different superscripts * and ** horizontally differ significantly (p<0.01). Mean having different superscripts a and b horizontally differ significantly (p<0.05). Table 2. Effect of different bacterial agents on acute phase proteins and NOx in healthy and subclinical mastitis milk of riverine buffalo (Bubalus bubalis) Table 3. MeanSE of SCC, acute phase proteins and NOx in milk from healthy and subclinical mastitis (irrespective of the causative agent) in riverine buffalo to decide the threshold limit (n = 496) Parameter Healthy milk Subclinical mastitis milk Cut-off points Somatic cell count (105 cells/ml) 0.930.007 2.050.056 - 1-Anti trypsin (U/L) 3,340.43138.54 6,397.71*150.99 6,396.22 1-Acid glycoprotein (mg ml-1) 0.176a0.026 2.92b0.032 2.92 Fibrinogen (g/dl) Not detectable Not detectable - NOx (nitrate+nitrite) (M) 11.090.19 17.330.29 17.34 Mean having different superscripts * and ** horizontally differ significantly (p<0.01). Mean having different superscripts a and b horizontally differ significantly (p<0.05). phase proteins and NOx in milk from healthy and subclinical mastitis (irrespective of the causative he threshold limit (n 496) Mean having different superscripts * and horizontally differ significantly (p<0.01) Mean having different superscripts a and b horizontally differ significantly (p<0.05). strongly correlated (p<0.01) with 1-antitrypsin only in SCM milk. With NOx, Log10SCC is correlated at (p<0.05) and (p<0.01) in healthy and SCM milk, respectively. Log10SCC is correlated with 1- acid glycoprotein at p<0.05. values and likelihood ratios After calculating the percent sensitivity, specificity, accuracy, predictive values and likelihood ratios for all the significantly altered parameters, 1- anti trypsin was most in agreement with IDF criteria for SCM (i.e. bacterial growth in culture media and SCC2105 cells/ml), followed by 1- acid glycoprotein. The value for NOx was at par (Table 4). However, the values for the same parameters were high when the causative agent was only Gram positive bacteria (Table 5). Percent sensitivity, specificity, accuracy, predictive NOx is correlated with 1- antitrypsin and 1- acid glycoprotein at p<0.01 and p<0.05, respectively. DISCUSSION The present study was carried out to compare the usefulness of 1-antitrypsin, 1-acid glycoprotein, fibrinogen and NOx in detecting SCM, with special reference to bubaline SCM. Udder profile correlation coefficient From Table 6 it can be observed that Log10SCC From Table 6 it can be observed that Log10SCC is Table 4. Evaluation of 1-anti trypsin, 1-acid glycoprotein and NOx as an indicator for diagnosis of subclinical mastitis in riverine buffalo Name of the parameter Total samples examined (N) Test positive buffaloes (a+b) Test reaction as compared to cultural examination Percent sensitivity a/(a+d) 100 Percent specificity c/(b+c) 100 Percent accuracy (a+c)/N 100 Positive predictive value (%) a/(a+b) 100 Negative predictive value (%) c/(c+d) 100 Likelihood ratio (positive) sensitivity/ (100- specificity) Likelihood ration (negative) (100- sensitivity)/ specificity True positive (a) False positive (b) True negative (c) False negative (d) 1-Anti trypsin 2,048 494 410 84 1,468 86 82.66 94.59 91.70 83.00 94.47 15.28 0.18 1-Acid glycoprotein 2,048 508 376 132 1,420 120 75.81 91.49 87.70 74.02 92.21 8.90 0.26 NOx 2,048 514 358 156 1,396 138 74.18 89.95 85.64 69.64 91.00 7.44 0.29 Positive bacterial culture and SCC2105 cells/ml 2,048 496 496 - 1,552 - 100 100 100 - - - - psin, 1-acid glycoprotein and NOx as an indicator for diagnosis of subclinical mastitis in riverine Table 4. Evaluation of 1-anti trypsin, 1-acid glycoprotein and NOx as an indicator for diagn buffalo Table 4. Evaluation of 1-anti trypsin, 1-acid glycoprotein and NOx as an indicator for diagnosis of subclin Guha et al. (2013) Asian Australas. J. Anim. Sci. 26:788-794 792 Table 5. Udder profile correlation coefficient Evaluation of 1-anti trypsin, 1-acid glycoprotein and NOx as an indicator for diagnosis of subclinical mastitis in riverine buffalo caused by gram positive bacteria only Name of the parameter Total samples examined (N) Test positive buffaloes (a+b) Test reaction as compared to cultural examination Percent sensitivity a/(a+d) 100 Percent specificity c/(b+c) 100 Percent accuracy (a+c)/N 100 Positive predictive value (%) a/(a+b) Negative predictive value (%) c/(c+d) Likelihood ratio (positive) sensitivity/ (100- specificity) Likelihood ration (negative) (100-sensitivity) /specificity True positive (a) False positive (b) True negative (c) False negative (d) 1-Anti trypsin 1,964 429 366 63 1,489 46 88.83 95.94 94.45 85.31 97.00 21.88 0.12 1-Acid glycoprotein 1,964 424 344 80 1,472 68 83.50 94.85 92.46 81.13 95.58 16.21 0.17 NOx 1,964 454 329 125 1,427 60 79.85 91.95 89.41 72.46 95.97 9.92 0.22 Positive bacterial culture and SCC2105 cells/ml 1,964 412 412 - 1,552 - 100 100 100 - - - - trypsin, 1-acid glycoprotein and NOx as an indicator for diagnosis of subclinical mastitis in riverine bacteria only SCM milk samples were those that showed bacterial growth in culture media and had a SCC of 2105 cells/ml (IDF, 2005). Gram positive bacterial agents were the most prevalent (Table 1). Similar observations were reported by Sharma et al. (2010) who attributed the contamination to the presence of organisms in the sub-continent atmosphere. The mean SCC in the SCM milk were significantly (p<0.01) high (Tables 2 and 3) owing to inflammatory reactions (Guha et al., 2010). isoforms of 1-acid glycoprotein, a low MW group (44 kDa), produced in the mammary gland (MG-AGP), and a higher MW group (55 to 70 kDa), produced by somatic cells (SC-AGP), were isolated by Ceciliani et al. (2007). Identical SC-AGP isoforms can be found both in milk and blood polymorpho-nuclear cells. Hence, an increase in the concentration of 1-acid glycoprotein can be attributed to increased synthesis by the somatic cells as well as by mammary gland cells as an immuno-protective measure. Gera and Guha (2011) also reported a similar observation in crossbred cow SCM milk. The significant increase of α1- anti trypsin in SCM milk (Tables 2 and 3) could be due to bacterial infection. The APP showed a substantial increase in SCM milk caused by all types of organisms. Udder profile correlation coefficient (2013) Asian Australas. J. Anim. Sci. 26:788-794 793 (Table 4) for all kind of infections. Our observation concurs with the reports of Gera and Guha (2011) in crossbred cows. These values were more in milk samples infected with Gram positive bacteria. The predictive values and likelihood ratios for positive tests are observed to be greater for 1-anti trypsin (83.00%; 15.28) than 1-acid glycoprotein (74.02%; 8.90) and NOx (69.64%; 7.44) (Table 4). The percent positive predictive values and likelihhod ratio (positive) when calculated in SCM milk infected with Gram positive bacteria for 1-anti trypsin, 1- acid glycoprotein and NOx were found to be 85.13; 21.88, 81.13; 16.21, and 72.46; 9.92, respectively (Table 5). The variation in these values was due to the fact that the concentration of the same parameters were lesser in SCM milk infected with E. coli than those milk samples which were infected with Staphylococcus or Streptococcus, though the level of significance were same for all the cases when compared with healthy milk (Table 2). The elevation of the parameters in gram positive SCM samples might be due to the fact that gram positive bacteria are more pathogenic in destroying the mammary gland epithelia whereas E. coli are relatively less severe on mammary gland cells (Wenz et al., 2006). and p<0.01 respectively in healthy and infected milk (0.546 vs 0.845, healthy vs infected milk). This may be due to the fact that the source of NOx is macrophages, a somatic cell fraction as discussed above. Apart from Log10 SCC, NOx was also significantly correlated with 1-antitrypsin and 1 - acid glycoprotein at p<0.01 and p<0.05, respectively. Similar observations were reported by Gera and Guha (2011) in cow milk. CONCLUSION It can be reasonably concluded that though the concentration of two APPs, viz. 1-antitrypsin and 1-acid glycoprotein as well as NOx was significantly higher in milk during subclinical form of the inflammatory reaction, but, only 1-anti trypsin was in agreement with the IDF criteria for SCM diagnosis for all kind of bacterial infections. The 1-acid glycoprotein indicates SCM when caused by Gram positive bacteria alone. NOx did not attest to be a good indicator of SCM. The threshold value for 1- antitrypsin and 1-acid glycoprotein is fixed at 6,396.22 U/L and 2.92 mg/ml, respectively. Standardizing easy qualitative methods for estimating these indicators is recommended to enable the development of a kit for diagnosing SCM in the field. A pathophysiological explanation of the ascertained association is also recommended for further study. Likelihood test of a positive test result >10 indicates that the test can be used to rule in the disease. Likelihood ratio of negative results describes how much more likely the animal has a negative test result when it has the disease (Petrie and Watson, 2008). The likelihood ratio for a positive test for α1- anti trypsin was found to be greater than 10 irrespective of the bacterial agent causing SCM. For 1- acid glycoprotein the ratio was greater than 10 when SCM causative agents were Gram positive bacteria. The likelihood ratio (positive) for NOx was lesser than 10 irrespective of the mastitogenic agents. The purpose of separately considering Gram positive bacterial agents is that they are the most prevalent mastitogenic agents in the tropical countries as discussed above. To the best of our knowledge no such studies for 1-acid glycoprotein and NOx were conducted previously. This is the first of its type. Hence, it can be considered as a pioneer work with special reference to bubaline SCM. Udder profile correlation coefficient The increase in the concentration of 1-anti trypsin was attributed to breach in the blood milk barrier by the action of inflammatory modulators and bacterial toxins, thus, it is a serum derivative (Gera and Guha, 2011). In the present study, fibrinogen was not detected in either healthy or SCM milk (Table 1). Our observation agrees with Tabrazi et al. (2008) and Gera and Guha (2011); who reported fibrinogen, a mild APP, appears in the milk during acute or chronic stage as a blood clotting factor or indicator of fibrosis. Fibrinogen was not taken up for further investigation. A significant (p<0.05) increase of 1-acid glycoprotein concentration was also observed for all types of infections (Tables 2 and 3). Up to 2006 there was no report of the presence of this 1-acidglycoprotein in healthy or mastitic milk of dairy animals. Mansson et al. (2006) was first to report α1- acid glycoprotein in healthy as well as in milk showing higher SCC in cows. A weaker and negative correlation of 1-acid glycoprotein with SCC was reported by these authors. But, in the present investigation it was observed that the concentration of 1-acid glycoprotein had a strong positive correlation with Log10 SCC. The increase could be due to excess of somatic cells in SCM. Two From Table 3 it can be observed that NOx in the infected milk samples increased significantly (p<0.01). A similar observation was made by Bulbul and Ylmaz (2004) and Gera and Guha (2011). They attributed the increase to increased macrophages, a fraction of SCC. We perused percent sensitivity, specificity, accuracy, predictive values and likelihood ratios of 1-anti trypsin, 1-acid glycoprotein and NOx as predictors of mastitis, taking the IDF criteria as the bench mark. It was observed that % sensitivity, specificity, accuracy were better for 1- anti trypsin, followed by 1-acid glycoprotein and NOx Table 6. Correlation coefficient of milk biochemical components in healthy and SCM milk of riverine buffalo (n = 496) Log10 SCC 1-Anti trypsin 1-Acid glycoprotein NOx (nitrate+nitrite) Healthy milk Infected milk Healthy milk Infected milk Healthy milk Infected milk Healthy milk Infected milk Log10 SCC - - 0.019 0.795 0.098 0.559* 0.546* 0.845 1-Anti trypsin - - - - 0.064 0.028 0.022 0.803 1-Acid glycoprotein - - - - - - 0.056 0.644* NOx (nitrate+nitrite) - - - - - - - - * Indicate significant at p<0.05. ** Indicate significant at p<0.01. Guha et al. mastitis, Intas Polivet 12:9-11. mastitis, Intas Polivet 12:9-11. Gera, S., A. Sharma, R. S. Dabur, V. K. Jain and S. L. Garg. 2006. Studies on changes in milk composition and chemotherapeutic sensitivity in camel (Camelus dromedarius) in subclinical mastitis. In: Proceedings of the Internation Scientific Conference in Camels (Vol-II pp. 937-946) Quassim University Saudi Arabia. Petrie, A. and P. Watson. 2008. Statistics for Veterinary and Animal Science, (Blackwell Publishing, London). Reddy, L. V., P. C. Choudhuri and P. A. Hamza. 2001. Comparative efficacy of different tests in the diagnosis of subclinical mastitis in crossbred cows. Ind. Vet. J. 78:903-905. Gonzalez, F. D. H., F. Tecles, S. Martı´nez-Subiela, A. Tvarijonaviciute, L. Soler and J. J. Cerone. 2008. Acute phase protein response in goats. J. Vet. Diagn. Invest. 20:580-584. Safi, S., A. Khoshvaghti, S. R. Jafarzadeh, M. Bolourchi and I. Nowrouzian. 2009. Acute phase proteins in the diagnosis of bovine subclinical mastitis. Vet. Clin. Pathol. 38:471-496. Guha, A. and S. Gera. 2012. Evaluation of milk trace elements, lactate dehydrogenase, alkaline phosphatase and aspartate aminotransferase activity of subclinical mastitis milk as and indicator of subclinical mastitis in riverine buffalo (Bubalus bubalis). Asian Australas. J. Anim. Sci. 25:353-360. Sharma, N., V. Pandey and N. A. Sudhan. 2010. Comparison of some indirect screening tests for detection of SCM in dairy cows. Bulg. J. Vet. Med. 13:98-103. Tabrazi, A. D., R. A. Batavani, S. A. Rezaei and M. Ahmadi. 2008. Fibrinogen and ceruloplasmin in plasma and milk from dairy cows with subclinical and clinical mastitis. Pak. J. Biol. Sci. 11: 571-576. Guha, A., S. Gera and A. Sharma. 2010. Assessment of chemical and electrolyte profile as an indicator of subclinical mastitis in riverine buffalo (Bubalus bubalis), Har. Vet. 49:19-21. International Dairy Federation (IDF). 2005. Diagnostic potential of California Mastitis Test to detect subclinical mastitis 26. Maastricht, Netherlands. pp. 15-19. Thrusfield, M. 2005. Veterinary epidemiology, 3rd ed, Blackwell Science, United Kingdom., p. 158. Varley, H., A. H. Gowenlock and M. Bell. 1980. Practical clinical biochemistry. 5th ed. William Heinemann Medical Books Ltd., London: United Kingdom. pp. 574-575. Katsoulos, P. D., G. Christodoulopoulos, A. Minas, M. A. Karatzia, K. Pourliotis, K. Spyridon and S. K. Kritas. 2010. The role of lactate dehydrogenase, alkaline phosphatase and aspartate aminotransferase in the diagnosis of subclinical intramammary infections in dairy sheep and goats. J. Dairy Res. 77:107-111. Wenz, J. R., G. M. Barrington, F. B. Garry, R. P. Ellis and R. J. Magnuson. 2006. REFERENCES Bouchard, L., S. Blais, C. Desrosiers, X. Zhao and P. Lacasse. 1999. Nitric oxide production during endotoxin induced mastitis in cows. J. Dairy Sci. 82:2574-2581. Bulbul, A. and B. Ylmaz. 2004. Relationship between the level of nitric oxide and somatic cell count in the milk of cows with mastitis. Veteriner Bilimleri Dergisi 20:95-102. Ceciliani, F., V. Pocacqua, C. Lecchi, R. Fortin, R. Rebucci, G. Avallone, V. Bronzo, F. Cheli and P. Sartorelli. 2007. Differential expression and secretion of alpha1-acid glycoprotein in bovine milk. J. Dairy Res. 74:374-380. De, U. K. and R. Mukherjee. 2009. The inhibitory response of Azadirachta indica extract on nitric oxide production by milk leukocytes during clinical mastitis. Vety. Archiv. 79:41-50. To prevent any ambiguity, double statistical evaluation for each presumed indicator was done in this study by correlating with Log10SCC (Gold Standard test) separately in healthy and SCM milk. From Table 6 it can be observed that 1-anti trypsin was also found strongly correlated (p<0.01) with Log10SCC in SCM milk (0.795), while it was insignificant in healthy milk (0.092). The 1-acid glycoprotein had a positive correlation, significant at p<0.05 (0.098 vs 0.559, healthy vs infected milk) with Log10SCC (Table 6). It is also evident that the correlation between Log10 SCC and NOx were significant at p<0.05 Fritz, H., I. Trautschold and E. Werle. 1974. Methods in Enzymatic Analysis, 2nd ed. Academic Press New York, USA, p. 234. Gera, S. A. Guha. 2011. Assessment of acute phase proteins and nitric oxide as indicators of subclinical mastitis in Holstein Haryana cattle. Ind. J. Anim. Sci. 81:1029-1031. Haryana cattle. Ind. J. Anim. Sci. 81:1029-1031. Gera, S. and A. Guha. 2012. Effect of sub clinical mastitis on milk biochemical constituents in crossbred cows. Ind. Vet. J. 89:33- 34. Gera, S., A. Guha, A. Sharma and V. Manocha. 2011. Evaluation of trace element profile as an indicator of bovine sub-clinical Guha et al. (2013) Asian Australas. J. Anim. Sci. 26:788-794 794 Mansson, H. L., C. Branning, G. Alden and M. Paulsson, 2006. Relationship between somatic cell count individual leukocyte populations and milk components in bovine udder quarter milk. Int. Dairy J. 16:717-727. mastitis, Intas Polivet 12:9-11. Escherichia coli isolates’ serotypes, genotypes and virulence genes and clinical coliform mastitis severity. J. Dairy Sci. 89:3408-3412.
https://openalex.org/W2042116669	https://europepmc.org/articles/pmc3344571?pdf=render	Latin	null	(<i>Z</i>)-1,3-Bis(4-chlorophenyl)-2-(1<i>H</i>-1,2,4-triazol-1-yl)prop-2-en-1-one	Acta crystallographica. Section E	2,012	cc-by	3,307	Ling-Ling Dai, Ben-Tao Yin, Jing-Song Lv, Sheng-Feng Cui and Cheng-He Zhou* Reﬁnement R[F 2 > 2(F 2)] = 0.037 wR(F 2) = 0.105 S = 1.02 3104 reﬂections Reﬁnement R[F 2 > 2(F 2)] = 0.037 wR(F 2) = 0.105 S = 1.02 3104 reﬂections 208 parameters H-atom parameters constrained max = 0.21 e A˚ 3 min = 0.26 e A˚ 3 Laboratory of Bioorganic & Medicinal Chemistry, School of Chemistry and Chemical Engineering, Southwest University, Chongqing 400715, People’s Republic of China Correspondence e-mail: zhouch@swu.edu.cn 208 parameters H-atom parameters constrained max = 0.21 e A˚ 3 min = 0.26 e A˚ 3 Received 8 April 2012; accepted 13 April 2012 Key indicators: single-crystal X-ray study; T = 296 K; mean (C–C) = 0.003 A˚; R factor = 0.037; wR factor = 0.105; data-to-parameter ratio = 14.9. Data collection: APEX2 (Bruker, 2009); cell reﬁnement: SAINT (Bruker, 2009); data reduction: SAINT; program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to reﬁne structure: SHELXL97 (Sheldrick, 2008); molecular graphics: SHELXTL (Sheldrick, 2008); software used to prepare material for publication: SHELXTL. In the title molecule, C17H11Cl2N3O, the C C bond connecting the triazole and 4-chlorophenyl groups adopts a Z geometry. The dihedral angles formed by the triazole ring and the 4-chloro substituted benzene rings are 67.3 (1) and 59.1 (1). The dihedral angle between the two benzene rings is 73.5 (1). In the title molecule, C17H11Cl2N3O, the C C bond connecting the triazole and 4-chlorophenyl groups adopts a Z geometry. The dihedral angles formed by the triazole ring and the 4-chloro substituted benzene rings are 67.3 (1) and 59.1 (1). The dihedral angle between the two benzene rings is 73.5 (1). This work was partially supported by the National Natural Science Foundation of China (No. 21172181), the Key Program of the Natural Science Foundation of Chongqing (CSTC2012jjB10026), the Specialized Research Fund for the Doctoral Program of Higher Education of China (SRFDP 20110182110007) and the Research Funds for the Central Universities (XDJK2011D007, XDJK2012B026). Related literature For the pharmacological activity of triazole compounds, see: Wang & Zhou (2011); Zhou & Wang (2012). For the biological activity of chalcones, see: Jin et al. (2010). For related struc- tures, see: Wang et al. (2009); Yan et al. (2009). For the synthesis, see: Yin et al. (2012). Supplementary data and ﬁgures for this paper are available from the IUCr electronic archives (Reference: LH5453). Experimental Crystal data C17H11Cl2N3O Mr = 344.19 organic compounds Triclinic, P1 a = 5.588 (3) A˚ b = 11.850 (7) A˚ c = 12.653 (8) A˚ = 74.787 (10) = 88.884 (9) = 86.461 (9) V = 807.1 (8) A˚ 3 Z = 2 Mo K radiation = 0.41 mm1 T = 296 K 0.22 0.18 0.15 mm Data collection Bruker APEXII CCD diffractometer Absorption correction: multi-scan (SADABS; Bruker, 2009) Tmin = 0.915, Tmax = 0.941 4414 measured reﬂections 3104 independent reﬂections 2458 reﬂections with I > 2(I) Rint = 0.013 Reﬁnement R[F 2 > 2(F 2)] = 0.037 wR(F 2) = 0.105 S = 1.02 3104 reﬂections 208 parameters H-atom parameters constrained max = 0.21 e A˚ 3 min = 0.26 e A˚ 3 Acta Crystallographica Section E Structure Reports Online ISSN 1600-5368 Acta Crystallographica Section E Structure Reports Online ISSN 1600-5368 Bruker (2009). APEX2, SAINT and SADABS. Bruker AXS Inc., Madison, Wisconsin, USA. Jin, L., Yan, C.-Y., Gan, L.-L. & Zhou, C.-H. (2010). Chin. J. Biochem. Pharm., 31, 358–361. Sheldrick, G. M. (2008). Acta Cryst. A64, 112–122. Wang, G., Lu, Y., Zhou, C. & Zhang, Y. (2009). Acta Cryst. E65, o1113. Wang, Y. & Zhou, C.-H. (2011). Sci. Sin. Chim. 41, 1429–1456. Yan, C.-Y., Wang, G.-Z. & Zhou, C.-H. (2009). Acta Cryst. E65, o2054. Yin, B.-T., Lv, J.-S., Wang, Y. & Zhou, C.-H. (2012). Acta Cryst. E68, o1197. Zhou, C.-H. & Wang, Y. (2012). Curr. Med. Chem. 19, 239–280. (Z)-1,3-Bis(4-chlorophenyl)-2-(1H-1,2,4- triazol-1-yl)prop-2-en-1-one Ling-Ling Dai, Ben-Tao Yin, Jing-Song Lv, Sheng-Feng Cui and Cheng-He Zhou* doi:10.1107/S1600536812016170 Refinement H atoms were placed at calculated position with C—H = 0.93 Å. The Uiso(H) values were set equal to 1.2Ueq(C). C ti d t il H atoms were placed at calculated position with C—H = 0.93 Å. The Uiso(H) values were set equal to 1.2Ueq(C). H atoms were placed at calculated position with C—H = 0.93 Å. The Uiso(H) values were set equal to 1.2Ueq(C). Computing details Comment Triazoles as an important type of five-membered aromatic heterocycles have been paid increasing attention for their broad bioactive spectrum in medicinal chemistry (Wang et al., 2011; Zhou et al., 2012). The incorporation of a triazole ring into chalcone skeletons could largely improve bioactivities like antimicrobial, anticancer, antiviral and anti- inflammatory (Jin et al., 2010). In view of this, we have synthesized and reported some triazolylchalcones (Wang et al., 2009; Yan et al., 2009; Yin et al., 2012). Herein, the crystal structure of title compound (I) is reported. The molecular structure of (I) is shown in Fig. 1. The C8═C11 bond adopts a Z geometry. The atoms in the region of the double bond have an essentially planar arrangement i.e. the r.m.s. deviation the atoms C7/C8/C11/C12/N1 is 0.025 Å. The torsion angles of C12–C11═C8–C7 and C12–C11═C8–N1 are -174.66 (17)° and 5.7 (3)°. The dihedral angles formed by the triazole ring and the 4-chloro-substituted benzene rings are 67.3 (1)° (C1-C6) and 59.1 (1)° (C12-C17), respectively. The dihedral angle between the two benzene rings is 73.5 (1)°. References Experimental Mr = 344.19 o1456 Dai et al. Acta Cryst. (2012). E68, o1456 doi:10.1107/S1600536812016170 supplementary materials Experimental Compound (I) was prepared according to the procedure of Yin et al. (2012). Single crystals were grown by slow evaporation of a solution of (I) in ethyl acetate and petroleum ether (1:3, V/V) at room temperature. Acta Cryst. (2012). E68, o1456 [doi:10.1107/S1600536812016170] Acta Cryst. (2012). E68, o1456 [doi:10.1107/S1600536812016170] Figure 1 (Z)-1,3-Bis(4-chlorophenyl)-2-(1H-1,2,4-triazol-1-yl)prop-2-en- 1-one (Z)-1,3-Bis(4-chlorophenyl)-2-(1H-1,2,4-triazol-1-yl)prop-2-en- 1-one Crystal data C17H11Cl2N3O Mr = 344.19 Triclinic, P1 Hall symbol: -P 1 a = 5.588 (3) Å b = 11.850 (7) Å c = 12.653 (8) Å α = 74.787 (10)° β = 88.884 (9)° γ = 86.461 (9)° V = 807.1 (8) Å3 Z = 2 F(000) = 352 Dx = 1.416 Mg m−3 Mo Kα radiation, λ = 0.71073 Å Cell parameters from 2086 reflections θ = 3.3–27.2° µ = 0.41 mm−1 T = 296 K Block, yellow 0.22 × 0.18 × 0.15 mm Data collection Bruker APEXII CCD diffractometer Radiation source: fine-focus sealed tube Graphite monochromator φ and ω scans Absorption correction: multi-scan (SADABS; Bruker, 2009) Tmin = 0.915, Tmax = 0.941 4414 measured reflections 3104 independent reflections 2458 reflections with I > 2σ(I) Rint = 0.013 θmax = 26.0°, θmin = 3.4° h = −6→6 k = −7→14 l = −14→15 Refinement Refinement on F2 Least-squares matrix: full R[F2 > 2σ(F2)] = 0.037 wR(F2) = 0.105 S = 1.02 3104 reflections 208 parameters 0 restraints Primary atom site location: structure-invariant direct methods Secondary atom site location: difference Fourier map Hydrogen site location: inferred from neighbouring sites H-atom parameters constrained w = 1/[σ2(Fo2) + (0.0515P)2 + 0.1439P] where P = (Fo2 + 2Fc2)/3 (Δ/σ)max = 0.001 Δρmax = 0.21 e Å−3 Δρmin = −0.26 e Å−3 Crystal data C17H11Cl2N3O Mr = 344.19 Triclinic, P1 Hall symbol: -P 1 a = 5.588 (3) Å b = 11.850 (7) Å c = 12.653 (8) Å α = 74.787 (10)° β = 88.884 (9)° γ = 86.461 (9)° V = 807.1 (8) Å3 Z = 2 F(000) = 352 Dx = 1.416 Mg m−3 Mo Kα radiation, λ = 0.71073 Å Cell parameters from 2086 reflections θ = 3.3–27.2° µ = 0.41 mm−1 T = 296 K Block, yellow 0.22 × 0.18 × 0.15 mm Data collection Bruker APEXII CCD diffractometer Radiation source: fine-focus sealed tube Graphite monochromator φ and ω scans Absorption correction: multi-scan (SADABS; Bruker, 2009) Tmin = 0.915, Tmax = 0.941 4414 measured reflections 3104 independent reflections 2458 reflections with I > 2σ(I) Rint = 0.013 θmax = 26.0°, θmin = 3.4° h = −6→6 k = −7→14 l = −14→15 Refinement Refinement on F2 Least-squares matrix: full R[F2 > 2σ(F2)] = 0.037 wR(F2) = 0.105 S = 1.02 3104 reflections 208 parameters 0 restraints Primary atom site location: structure-invariant direct methods Secondary atom site location: differen map Hydrogen site location: inferred from neighbouring sites H-atom parameters constrained w = 1/[σ2(Fo2) + (0.0515P)2 + 0.1439P where P = (Fo2 + 2Fc2)/3 (Δ/σ)max = 0.001 Δρmax = 0.21 e Å−3 Δρmin = −0.26 e Å−3 Crystal data C17H11Cl2N3O Mr = 344.19 Triclinic, P1 Hall symbol: -P 1 a = 5.588 (3) Å b = 11.850 (7) Å c = 12.653 (8) Å α = 74.787 (10)° β = 88.884 (9)° γ = 86.461 (9)° V = 807.1 (8) Å3 Z = 2 F(000) = 352 Dx = 1.416 Mg m−3 Mo Kα radiation, λ = 0.71073 Å Cell parameters from 2086 reflections θ = 3.3–27.2° µ = 0.41 mm−1 T = 296 K Block, yellow 0.22 × 0.18 × 0.15 mm Data collection Bruker APEXII CCD diffractometer Radiation source: fine-focus sealed tube Graphite monochromator φ and ω scans Absorption correction: multi-scan (SADABS; Bruker, 2009) Tmin = 0.915, Tmax = 0.941 4414 measured reflections 3104 independent reflections 2458 reflections with I > 2σ(I) Rint = 0.013 θmax = 26.0°, θmin = 3.4° h = −6→6 k = −7→14 l = −14→15 Refinement Refinement on F2 Least-squares matrix: full R[F2 > 2σ(F2)] = 0.037 wR(F2) = 0.105 S = 1.02 3104 reflections 208 parameters 0 restraints Primary atom site location: structure-invariant direct methods Secondary atom site location: difference Fouri map Hydrogen site location: inferred from neighbouring sites H-atom parameters constrained w = 1/[σ2(Fo2) + (0.0515P)2 + 0.1439P] where P = (Fo2 + 2Fc2)/3 (Δ/σ)max = 0.001 Δρmax = 0.21 e Å−3 Δρmin = −0.26 e Å−3 4414 measured reflections 3104 independent reflections 2458 reflections with I > 2σ(I) Rint = 0.013 θmax = 26.0°, θmin = 3.4° h = −6→6 k = −7→14 l = −14→15 Secondary atom site location: difference Fourier map Hydrogen site location: inferred from neighbouring sites H-atom parameters constrained w = 1/[σ2(Fo2) + (0.0515P)2 + 0.1439P] where P = (Fo2 + 2Fc2)/3 (Δ/σ)max = 0.001 Δρmax = 0.21 e Å−3 Δρmin = −0.26 e Å−3 Secondary atom site location: difference Fourier map Hydrogen site location: inferred from neighbouring sites H-atom parameters constrained w = 1/[σ2(Fo2) + (0.0515P)2 + 0.1439P] where P = (Fo2 + 2Fc2)/3 (Δ/σ)max = 0.001 Δρmax = 0.21 e Å−3 Δρmin = −0.26 e Å−3 Secondary atom site location: difference Fourier map Hydrogen site location: inferred from neighbouring sites H-atom parameters constrained w = 1/[σ2(Fo2) + (0.0515P)2 + 0.1439P] where P = (Fo2 + 2Fc2)/3 (Δ/σ)max = 0.001 Δρmax = 0.21 e Å−3 Δρmin = −0.26 e Å−3 sup-2 Acta Cryst. Computing details Data collection: APEX2 (Bruker, 2009); cell refinement: SAINT (Bruker, 2009); data reduction: SAINT (Bruker, 2009); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: SHELXTL (Sheldrick, 2008); software used to prepare material for publication: SHELXTL (Sheldrick, 2008). sup-1 Acta Cryst. (2012). E68, o1456 supplementary materials Fi 1 Figure 1 Figure 1 Figure 1 The molecular structure of (I) showing the displacement ellipsoids at the 50% probability level Figure 1 The molecular structure of (I), showing the displacement ellipsoids at the 50% probability level. g The molecular structure of (I), showing the displacement ellipsoids at the 50% probability level. supplementary materials Special details Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes. Refinement. Refinement of F2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F2, conventional R-factors R are based on F, with F set to zero for negative F2. The threshold expression of F2 > σ(F2) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F2 are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger. Figure 1 (2012). E68, o1456 supplementary materials supplementary materials Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) x y z Uiso/Ueq Cl1 1.58438 (14) 0.60011 (6) −0.17465 (5) 0.1010 (3) Cl2 0.14916 (9) 0.80981 (5) 0.64259 (4) 0.07445 (19) C1 1.4458 (4) 0.6784 (2) −0.08893 (15) 0.0703 (6) C2 1.5551 (4) 0.7730 (2) −0.07247 (16) 0.0719 (6) H2A 1.6975 0.7964 −0.1085 0.086 C3 1.4505 (4) 0.83280 (18) −0.00168 (16) 0.0654 (5) H3A 1.5235 0.8969 0.0100 0.079* C4 1.2364 (3) 0.79806 (16) 0.05257 (14) 0.0532 (4) C5 1.1286 (4) 0.70284 (18) 0.03344 (15) 0.0637 (5) H5A 0.9861 0.6788 0.0691 0.076* C6 1.2320 (4) 0.6435 (2) −0.03847 (17) 0.0733 (6) H6A 1.1579 0.5808 −0.0525 0.088* C7 1.1296 (3) 0.87101 (16) 0.12331 (15) 0.0557 (4) C8 0.9716 (3) 0.81969 (14) 0.21872 (14) 0.0487 (4) C9 1.1937 (3) 0.63087 (16) 0.31196 (15) 0.0582 (5) H9A 1.3439 0.6594 0.3154 0.070* C10 0.9086 (4) 0.52367 (17) 0.32519 (19) 0.0736 (6) H10A 0.8222 0.4566 0.3425 0.088* C11 0.8102 (3) 0.88957 (15) 0.25517 (14) 0.0522 (4) H11A 0.7981 0.9669 0.2131 0.063* C12 0.6506 (3) 0.86366 (14) 0.35025 (13) 0.0475 (4) C13 0.6915 (3) 0.77059 (16) 0.44322 (14) 0.0554 (4) H13A 0.8248 0.7190 0.4452 0.067* C14 0.5387 (3) 0.75364 (16) 0.53190 (15) 0.0564 (4) H14A 0.5684 0.6911 0.5931 0.068* C15 0.3405 (3) 0.83027 (15) 0.52953 (14) 0.0524 (4) C16 0.2943 (3) 0.92271 (17) 0.43910 (16) 0.0637 (5) H16A 0.1595 0.9733 0.4373 0.076* C17 0.4495 (3) 0.93960 (15) 0.35136 (15) 0.0586 (5) H17A 0.4198 1.0033 0.2911 0.070* N1 0.9986 (2) 0.69650 (11) 0.26747 (11) 0.0466 (3) N2 0.8100 (3) 0.62670 (13) 0.27539 (14) 0.0621 (4) N3 1.1434 (3) 0.52092 (14) 0.34991 (15) 0.0737 (5) O1 1.1701 (3) 0.97469 (12) 0.10500 (12) 0.0776 (4) Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) x y z Uiso*/Ueq Acta Cryst. (2012). Acta Cryst. (2012). E68, o1456 supplementary materials E68, o1456 sup-3 supplementary materials supplementary materials Atomic displacement parameters (Å2) U11 U22 U33 U12 U13 U23 Cl1 0.1303 (6) 0.1091 (5) 0.0559 (3) 0.0454 (4) 0.0020 (3) −0.0198 (3) Cl2 0.0717 (3) 0.0773 (4) 0.0663 (3) 0.0119 (3) 0.0173 (2) −0.0097 (3) C1 0.0846 (14) 0.0750 (14) 0.0403 (10) 0.0208 (11) 0.0000 (9) −0.0016 (9) C2 0.0633 (12) 0.0815 (15) 0.0578 (12) 0.0082 (11) 0.0113 (9) 0.0014 (10) C3 0.0629 (11) 0.0632 (12) 0.0620 (12) −0.0054 (9) 0.0100 (9) −0.0024 (9) C4 0.0524 (10) 0.0568 (10) 0.0446 (9) −0.0020 (8) 0.0029 (7) −0.0033 (8) C5 0.0644 (12) 0.0749 (13) 0.0514 (11) −0.0129 (10) 0.0062 (9) −0.0143 (9) C6 0.0929 (16) 0.0733 (13) 0.0536 (11) −0.0080 (11) −0.0012 (11) −0.0158 (10) C7 0.0547 (10) 0.0540 (11) 0.0542 (10) −0.0096 (8) 0.0019 (8) −0.0052 (8) C8 0.0478 (9) 0.0483 (9) 0.0477 (9) −0.0061 (7) −0.0004 (7) −0.0079 (7) C9 0.0458 (9) 0.0616 (11) 0.0635 (11) 0.0033 (8) 0.0018 (8) −0.0117 (9) C10 0.0772 (14) 0.0486 (11) 0.0915 (16) −0.0111 (10) 0.0164 (12) −0.0116 (10) C11 0.0581 (10) 0.0459 (9) 0.0495 (9) −0.0011 (8) −0.0021 (8) −0.0070 (7) C12 0.0507 (9) 0.0439 (9) 0.0483 (9) −0.0003 (7) −0.0032 (7) −0.0131 (7) C13 0.0527 (10) 0.0569 (10) 0.0523 (10) 0.0133 (8) −0.0012 (8) −0.0102 (8) C14 0.0606 (11) 0.0529 (10) 0.0495 (10) 0.0077 (8) −0.0020 (8) −0.0051 (8) C15 0.0532 (10) 0.0534 (10) 0.0514 (10) 0.0010 (8) 0.0013 (8) −0.0161 (8) C16 0.0636 (11) 0.0584 (11) 0.0634 (12) 0.0193 (9) 0.0022 (9) −0.0113 (9) C17 0.0719 (12) 0.0465 (9) 0.0517 (10) 0.0112 (8) −0.0022 (9) −0.0061 (8) N1 0.0400 (7) 0.0463 (7) 0.0520 (8) −0.0052 (6) 0.0049 (6) −0.0101 (6) N2 0.0496 (8) 0.0545 (9) 0.0815 (11) −0.0141 (7) 0.0071 (8) −0.0146 (8) N3 0.0744 (11) 0.0538 (10) 0.0842 (12) 0.0101 (8) 0.0069 (9) −0.0065 (8) O1 0.0930 (10) 0.0584 (9) 0.0789 (10) −0.0199 (7) 0.0262 (8) −0.0121 (7) Geometric parameters (Å, º) Cl1—C1 1.745 (2) C9—N1 1.342 (2) Cl2—C15 1.743 (2) C9—H9A 0.9300 C1—C2 1.372 (3) C10—N2 1.311 (3) C1—C6 1.379 (3) C10—N3 1.351 (3) C2—C3 1.381 (3) C10—H10A 0.9300 C2—H2A 0.9300 C11—C12 1.461 (2) C3—C4 1.398 (3) C11—H11A 0.9300 C3—H3A 0.9300 C12—C13 1.397 (2) C4—C5 1.389 (3) C12—C17 1.398 (2) C4—C7 1.492 (3) C13—C14 1.377 (2) C5—C6 1.386 (3) C13—H13A 0.9300 C5—H5A 0.9300 C14—C15 1.384 (2) C6—H6A 0.9300 C14—H14A 0.9300 C7—O1 1.224 (2) C15—C16 1.377 (3) C7—C8 1.499 (2) C16—C17 1.375 (3) C8—C11 1.343 (2) C16—H16A 0.9300 C8—N1 1.428 (2) C17—H17A 0.9300 C9—N3 1.311 (3) N1—N2 1.366 (2) C2—C1—C6 121.6 (2) N2—C10—H10A 122.1 C2—C1—Cl1 118.80 (18) N3—C10—H10A 122.1 Atomic displacement parameters (Å2) Geometric parameters (Å, º) Acta Cryst. supplementary materials 119.6 (2) C8—C11—C12 130.38 (16) 119.1 (2) C8—C11—H11A 114.8 120.5 C12—C11—H11A 114.8 120.5 C13—C12—C17 117.31 (16) 120.7 (2) C13—C12—C11 124.22 (15) 119.6 C17—C12—C11 118.39 (15) 119.6 C14—C13—C12 121.34 (16) 118.93 (19) C14—C13—H13A 119.3 123.74 (16) C12—C13—H13A 119.3 117.23 (18) C13—C14—C15 119.61 (16) 120.40 (19) C13—C14—H14A 120.2 119.8 C15—C14—H14A 120.2 119.8 C16—C15—C14 120.58 (17) 119.2 (2) C16—C15—Cl2 119.85 (14) 120.4 C14—C15—Cl2 119.56 (14) 120.4 C17—C16—C15 119.36 (17) 120.59 (16) C17—C16—H16A 120.3 118.20 (17) C15—C16—H16A 120.3 121.21 (16) C16—C17—C12 121.78 (17) 122.27 (15) C16—C17—H17A 119.1 119.71 (16) C12—C17—H17A 119.1 118.02 (14) C9—N1—N2 109.33 (15) 110.62 (17) C9—N1—C8 129.36 (14) 124.7 N2—N1—C8 121.31 (13) 124.7 C10—N2—N1 101.71 (16) 115.89 (18) C9—N3—C10 102.45 (16) 1.4 (3) C17—C12—C13—C14 0.6 (3) −177.75 (14) C11—C12—C13—C14 177.44 (17) 0.0 (3) C12—C13—C14—C15 −0.2 (3) −0.6 (3) C13—C14—C15—C16 0.4 (3) −177.11 (17) C13—C14—C15—Cl2 −179.19 (14) −0.1 (3) C14—C15—C16—C17 −1.1 (3) 176.17 (18) Cl2—C15—C16—C17 178.52 (15) −2.1 (3) C15—C16—C17—C12 1.6 (3) 177.06 (15) C13—C12—C17—C16 −1.3 (3) 1.4 (3) C11—C12—C17—C16 −178.32 (18) −150.2 (2) N3—C9—N1—N2 −0.5 (2) 26.1 (3) N3—C9—N1—C8 178.82 (17) 30.1 (3) C11—C8—N1—C9 −122.4 (2) −153.57 (17) C7—C8—N1—C9 58.0 (2) 25.1 (3) C11—C8—N1—N2 56.9 (2) −155.23 (17) C7—C8—N1—N2 −122.75 (17) −155.21 (17) N3—C10—N2—N1 0.0 (2) 24.5 (2) C9—N1—N2—C10 0.3 (2) 5.7 (3) C8—N1—N2—C10 −179.08 (16) −174.66 (17) N1—C9—N3—C10 0.5 (2) 23.4 (3) N2—C10—N3—C9 −0.3 (3) −159.82 (19) C6—C1—Cl1 119.6 (2) C8—C11—C12 130.38 (16) C1—C2—C3 119.1 (2) C8—C11—H11A 114.8 C1—C2—H2A 120.5 C12—C11—H11A 114.8 C3—C2—H2A 120.5 C13—C12—C17 117.31 (16) C2—C3—C4 120.7 (2) C13—C12—C11 124.22 (15) C2—C3—H3A 119.6 C17—C12—C11 118.39 (15) C4—C3—H3A 119.6 C14—C13—C12 121.34 (16) C5—C4—C3 118.93 (19) C14—C13—H13A 119.3 C5—C4—C7 123.74 (16) C12—C13—H13A 119.3 C3—C4—C7 117.23 (18) C13—C14—C15 119.61 (16) C6—C5—C4 120.40 (19) C13—C14—H14A 120.2 C6—C5—H5A 119.8 C15—C14—H14A 120.2 C4—C5—H5A 119.8 C16—C15—C14 120.58 (17) C1—C6—C5 119.2 (2) C16—C15—Cl2 119.85 (14) C1—C6—H6A 120.4 C14—C15—Cl2 119.56 (14) C5—C6—H6A 120.4 C17—C16—C15 119.36 (17) O1—C7—C4 120.59 (16) C17—C16—H16A 120.3 O1—C7—C8 118.20 (17) C15—C16—H16A 120.3 C4—C7—C8 121.21 (16) C16—C17—C12 121.78 (17) C11—C8—N1 122.27 (15) C16—C17—H17A 119.1 C11—C8—C7 119.71 (16) C12—C17—H17A 119.1 N1—C8—C7 118.02 (14) C9—N1—N2 109.33 (15) N3—C9—N1 110.62 (17) C9—N1—C8 129.36 (14) N3—C9—H9A 124.7 N2—N1—C8 121.31 (13) N1—C9—H9A 124.7 C10—N2—N1 101.71 (16) N2—C10—N3 115.89 (18) C9—N3—C10 102.45 (16) C6—C1—C2—C3 1.4 (3) C17—C12—C13—C14 0.6 (3) Cl1—C1—C2—C3 −177.75 (14) C11—C12—C13—C14 177.44 (17) C1—C2—C3—C4 0.0 (3) C12—C13—C14—C15 −0.2 (3) C2—C3—C4—C5 −0.6 (3) C13—C14—C15—C16 0.4 (3) C2—C3—C4—C7 −177.11 (17) C13—C14—C15—Cl2 −179.19 (14) C3—C4—C5—C6 −0.1 (3) C14—C15—C16—C17 −1.1 (3) C7—C4—C5—C6 176.17 (18) Cl2—C15—C16—C17 178.52 (15) C2—C1—C6—C5 −2.1 (3) C15—C16—C17—C12 1.6 (3) Cl1—C1—C6—C5 177.06 (15) C13—C12—C17—C16 −1.3 (3) C4—C5—C6—C1 1.4 (3) C11—C12—C17—C16 −178.32 (18) C5—C4—C7—O1 −150.2 (2) N3—C9—N1—N2 −0.5 (2) C3—C4—C7—O1 26.1 (3) N3—C9—N1—C8 178.82 (17) C5—C4—C7—C8 30.1 (3) C11—C8—N1—C9 −122.4 (2) C3—C4—C7—C8 −153.57 (17) C7—C8—N1—C9 58.0 (2) O1—C7—C8—C11 25.1 (3) C11—C8—N1—N2 56.9 (2) C4—C7—C8—C11 −155.23 (17) C7—C8—N1—N2 −122.75 (17) O1—C7—C8—N1 −155.21 (17) N3—C10—N2—N1 0.0 (2) C4—C7—C8—N1 24.5 (2) C9—N1—N2—C10 0.3 (2) N1—C8—C11—C12 5.7 (3) C8—N1—N2—C10 −179.08 (16) C7—C8—C11—C12 −174.66 (17) N1—C9—N3—C10 0.5 (2) C8—C11—C12—C13 23.4 (3) N2—C10—N3—C9 −0.3 (3) C8—C11—C12—C17 −159.82 (19) sup-5 Acta Cryst. supplementary materials (2012). E68, o1456 sup-4 supplementary materials supplementary materials (2012). E68, o1456
https://openalex.org/W1990960376	http://journals.iucr.org/f/issues/2005/01/00/gx5034/gx5034.pdf	English	null	On the purification and preliminary crystallographic analysis of isoquinoline 1-oxidoreductase from<i>Brevundimonas diminuta</i>7	Acta crystallographica. Section F, Structural biology and crystallization communications	2,004	cc-by	4,369	1. Introduction ³ Current address: Structural Biology Laboratory, Department Of Chemistry, University of York, Heslington, York YO10 5YW, England. The molybdopterin-dependent enzyme xanthine dehydrogenase (XDH) has been widely studied owing to its pathogenic role in post- ischaemia reperfusion injury, acting as a source of reactive oxygen species (Harrison, 2002), and for its involvement in gout (Monu & Pope, 2004). The enzyme is a member of the mononuclear molyb- denum family of proteins that have been grouped into three sub- families (Hille, 1996), one of which includes XDH and related enzymes. A number of crystal structures of proteins from the XDH family are now available, including the aldehyde dehydrogenases from Desulfovibrio gigas (MOP; Romao et al., 1995; Rebelo et al., 2001) and D. desulfuricans (MOD; Rebelo et al., 2000), xanthine oxidase/dehydrogenases from bovine milk (Enroth et al., 2000) and Rhodobacter capsulatus (Truglio et al., 2002) and quinoline 2-oxidoreductase (QOR) from Pseudomonas putida 86 (Bonin et al., 2004). In addition, the structure of the closely related CO dehy- drogenase from Oligotropha carboxidovorans (Dobbek et al., 1999, 2002) has been solved. The latter is the only enzyme of the XDH family solved to real atomic resolution (1.1 AÊ ; Dobbek et al., 2002). The enzymes of this family consist of a large subunit or domain, which contains a mononuclear Mo at the active site that is bound to a molybdopterin cofactor through a dithiolene moiety, and a small subunit or domain containing two [2Fe±2S] clusters. For some enzymes of this family, including XDH, QOR and CODH, the primary sequence includes another domain containing an FAD molecule that acts as a transient electron acceptor. The Mb atom in the active site has been shown to be pentavalently coordinated, being bound to the two sulfurs of the dithiolene moiety, an oxo and a sul®do ligand and an OHÿ ion or H2O molecule, although the active-site con®guration of CODH contains an additional Cu atom. The crystal structures and extensive electron paramagnetic resonance (EPR; Hille, 1996) and extended X-ray absorption ®ne structure (EXAFS) studies (Cramer et al., 1984, Bordas et al., 1979) have greatly improved the understanding of the mechanism and function of this class of proteins. However, questions about details of the mechanism and substrate speci®city or rather lack of substrate speci®city remain unanswered. For example, the position of the sul®do ligand in the active site remains uncertain. crystallization communications crystallization communications Acta Crystallographica Section F Structural Biology and Crystallization Communications ISSN 1744-3091 D. Roeland Boer,a Axel MuÈller,a³ Susanne Fetzner,b David J. Lowec and Maria JoaÄo RomaÄoa* Isoquinoline 1-oxidoreductase (IOR) from Brevundimonas diminuta is a mononuclear molybdoenzyme of the xanthine-dehydrogenase family of proteins and catalyzes the conversion of isoquinoline to isoquinoline-1-one. Its primary sequence and behaviour, speci®cally in its substrate speci®city and lipophilicity, differ from other members of the family. A crystal structure of the enzyme is expected to provide an explanation for these differences. This paper describes the crystallization and preliminary X-ray diffraction experiments as well as an optimized puri®cation protocol for IOR. Crystallization of IOR was achieved using two different crystallization buffers. Streak-seeding and cross-linking were essential to obtain well diffracting crystals. Suitable cryo-conditions were found and a structure solution was obtained by molecular replacement. However, phases need to be improved in order to obtain a more interpretable electron- density map. aREQUIMTE/CQFB, Departamento de QuõÂmica, Faculdade de CieÃncias e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal, bWestfaÈlische Wilhelms-UniversitaÈt MuÈnster, Institut fuÈr Molekulare Mikrobiologie und Biotechnologie, Corrensstrasse 3, D-48149 MuÈnster, Germany, and cBiological Chemistry Department, John Innes Centre, Colney, Norwich NR4 7UH, England crystallization communications ®nal concentration of 1.5%(v/v). All column resin materials and hardware were from Amersham Pharmacia Biotech (Sweden) unless otherwise stated. All puri®cation steps were performed on an Amersham HPLC containing UV-900, P-900 and Frac-900 units coupled to a PC running Unicorn software v.3.21.02. All steps were performed at 277 K. Hydrophobic interaction was performed using phenyl Sepharose 6 FF resin. Two cation-exchange columns were used: a hand-packed SQ Sepharose Fast Flow resin and a Resource 15S 1 ml prepacked column. Gel chromatography was performed on a pre-packed Superdex 200 column (60 cm). SDS±PAGE was performed using 10% separating gel (SchaÈgger & von Jagow, 1987). Protein purity was determined using speci®c activity measurements and protein concentration was assessed by the Lowry method (Lowry et al., 1951). The pure samples from the gel-chromatography column were combined and the buffer was replaced with 10 mM Tris±HCl pH 8.5 using iterative concentrating steps through centrifugation over a Millipore/Amicon YM-10 microcon with 10 kDa cutoff followed by dilution steps with the new buffer. The protein sample was subse- quently concentrated to 20 mg mlÿ1. The puri®ed protein samples were divided into 25 ml aliquots and stored at 193 K. (Bonin et al., 2004), but resulfuration of MOP resulted in an active- site geometry with the S atom in the apical position (Huber et al., 1996), whereas the crystal structure (Rebelo et al., 2001) suggests there is an oxo ligand instead (desulfo form). The structure of isoquinoline 1-oxidoreductase (IOR) from Brevundimonas diminuta 7, another member of the XDH family, is expected to contribute to the general understanding of these enzymes. In addition, IOR deserves crystallographic characterization because it is in several ways an outlier in the XDH family, as described below. IOR is a heterodimeric protein that catalyses the ®rst step of isoquinoline degradation (Fig. 1). It consists of two subunits, one of 80 kDa containing the Mo active centre and one of 16 kDa containing the two [2Fe±2S] clusters (Lehmann et al., 1994). There are three remarkable differences between IOR and other members of the XDH family. Firstly, a sequence comparison of members within the family (Lehmann et al., 1994, 1995) shows that the N-terminal part of the catalytic subunit of IOR, which consists of residues 39±113 (domain A), aligns well with an amino-acid stretch closer to the C-terminus of XDH. 2.2. Crystallization Suitable crystallization conditions were screened at 277 and 293 K using the hanging-drop vapour-diffusion method and the sparse- matrix-based Hampton Research Crystal Screens 1 and 2 as well as the Memstart membrane protein screen of Molecular Dimensions Inc. Two conditions of the Memstart screen were found to yield crystals at 293 K, the ®rst containing 0.1 M NaCl, 100 mM HEPES pH 7.5, 10% PEG 4000 (crystallization buffer A) and the second 100 mM Tris±HCl pH 8.5, 12.5% PEG 8000 (crystallization buffer B1). Subsequent optimization experiments were performed by varying the pH in the range 6.5±9.0 (in steps of 0.5, using appropriate buffers for each pH) against the precipitant concentration (8±13% for buffer A, 10±15% for buffer B1) at both 277 and 293 K. Optimal conditions were found to correspond to the initial conditions identi®ed in the screens. Hampton Additive Screens were applied in order to improve the morphology of the crystals grown in buffer B1. A 10 mM concentration of BaCl2 gave the required improvement. The opti- mized crystallization conditions were based on the hanging-drop vapour-diffusion method at a temperature of 293 K using either crystallization buffer A or a buffer containing 10 mM BaCl2 (Sigma- Aldrich), 100 mM Tris±HCl pH 8.5, 12.5% PEG 8000 (crystallization buffer B2) using a protein concentration of 20 mg mlÿ1. Streak- seeding was often used to induce crystallization and was performed with either cleaned rat whiskers or human eyelashes. Crystals appeared within hours and reached maximum dimensions after one week. A harvesting buffer containing 0.1 M NaCl, 100 mM HEPES pH 7.5, 12.5% PEG 4000 or 10 mM BaCl2, 100 mM Tris±HCl pH 8.5 and 15% PEG 8000 for buffers A and B2, respectively, was added to the drop containing the crystals after maximum size had been reached. Cross-linking was performed using the harvesting buffer The puri®cation and characterization of IOR has been described previously (Lehmann et al., 1994) and involves heat-precipitation, hydrophobic interaction, cation-exchange and size-exclusion chro- matography. However, attempts to reproduce the published puri®- cation protocol were unsuccessful. For unknown reasons, in some cases the protein was lost during the hydrophobic interaction chro- matography (HIC) step. This problem was encountered previously (personal communication by SF) and depended on the cell culture used in the puri®cation. The puri®cation protocol was therefore reconsidered and different methods were tried in order to improve it. 1. Introduction It has been found in the equatorial position in bovine milk XDH (Okamoto et al., 2004) and in QOR Received 27 October 2004 Accepted 3 December 2004 Online 24 December 2004 Received 27 October 2004 Accepted 3 December 2004 Online 24 December 2004 # 2005 International Union of Crystallography All rights reserved # 2005 International Union of Crystallography All rights reserved 137 doi:10.1107/S1744309104032105 Acta Cryst. (2005). F61, 137±140 crystallization communications As a consequence, the sequence identity between the large subunit of IOR and corresponding units of other members of the XDH family is lower than the identity within those members. Secondly, the enzyme has a persistent `very rapid' signal as determined by EPR spectroscopy, which is generally associated with an enzyme-bound catalytic intermediate (Canne et al., 1999). This indicates that in the absence of an electron acceptor the catalytic cycle cannot be completed and substrate remains bound to the Mo atom. Finally, the protein appears to be very hydrophobic, an observation based on the fact that the protein crystallized in a screening condition designed for membrane proteins and also on its high af®nity for hydrophobic resins employed in the puri®cation. Hydrophobicity is a characteristic that has not been reported for other members of the family. 138 Boer et al. Isoquinoline 1-oxidoreductase 2.2. Crystallization However, the choice of puri®cation methods was severely restricted by the hydrophobic nature of the IOR, its instability at low pH and its complete loss of activity when subject to anion-exchange chromato- graphic methods. In this paper, we describe our efforts towards puri®cation and crystallization of IOR. We report preliminary X-ray analysis and derivatization of the crystals using arsenite inactivation of the molybdenum active site (Boer et al., 2004). 3. Results and discussion The puri®ed protein produced crystals which appeared after 1 or 2 d and reached maximum dimensions after one week. However, spontaneous crystal growth using buffer A only occurred when crystallizing the protein directly after puri®cation. After freezing and thawing the protein sample once, crystal growth had to be induced using (for example) streak-seeding. Crystal growth was more repro- ducible with PEG 8K (buffers B1 and B2) crystallization buffers. However, seeding techniques were found to increase the de®nition of the crystals edges and were therefore applied for all crystallization experiments. Initial freezing trials of the crystals resulted in poorly diffracting crystals with highly anisotropic diffraction spots exhibiting poorly resolved re¯ections. Extensive trials of different cryoprotec- tants for both crystallization conditions did not improve diffraction to an acceptable level. Crystals were mounted in capillaries either using the harvesting buffer or a cryoprotectant solution. In both cases, the crystals did not diffract beyond 3.5±3.0 AÊ on an in-house rotating- anode generator. In contrast, freezing the crystals, both in the cryo- stream as well as in bulk liquid nitrogen, decreased the diffraction and increased the mosaicity and anisotropy to such an extent that useful data could not be obtained. Cross-linking (with glutaraldehyde or betaine monohydrate) proved to be necessary in order to stabilize the crystals upon freezing, which increased the reproducibility of the mounting process and resulted in diffraction comparable with that of crystals mounted in capillaries. As mentioned previously, the reported puri®cation protocol (Lehmann et al., 1994) includes heat precipitation, hydrophobic interaction, cation-exchange and size-exclusion chromatography, with a yield of 68%. However, we were unable to reproduce the reported yield because of dif®culties in eluting IOR in the hydro- phobic interaction step, which led to complete loss of protein. A careful reconsideration of the puri®cation protocol was therefore necessary. Firstly, a procedure omitting both the heat precipitation and hydrophobic interaction steps was tried, leaving only the cation- exchange and size-exclusion chromatography steps after cell lysis. This worked for freshly grown cells, but after storage of the cells at 193 K the cation-exchange step did not retain the protein. A proce- dure was then tried that involved heat precipitation, a cation- exchange step and size-exclusion chromatography. However, the retention problem reoccurred at the ®rst cation-exchange procedure. crystallization communications Table 1 Summary of data sets measured. Values in parentheses are for the highest resolution shell. Native As peak Fe peak Fe in¯ection point Wavelength (AÊ ) 0.976 1.0450 1.7352 1.7415 Temperature (K) 100 100 100 100 Crystal data Space group P212121 P212121 P212121 P212121 Unit-cell parameters (AÊ ) a 67.3 67.1 66.5 66.5 b 122.2 117.0 114.7 114.7 c 239.1 238.1 237.5 237.5 Mosaicity 0.9 0.31 0.26 0.26 Data collection Resolution (AÊ ) 25±2.6 (2.66±2.60) 25±2.9 (3.0±2.9) 20±3.13 (3.32±3.19) 20±3.3 (3.4±3.3) No. observations 1234059 306704 231175 399048 No. unique re¯ections 59051 80049 60990 52462 Rmerge² (%) 11.7 (37.8) 8.5 (34.0) 12.2 (39.8) 13.5 (41) Completeness (%) 95.8 (93.7) 99.7 (100) 99.6 (100) 99.5 (100) I/(I) 5.1 (1.7) 12.1 (4.5) 10.4 (3.8) 13.4 (5.1) ² Rmerge = P h P i jIhi ÿ hIhij=P h P i Ihi. enriched with 1% glutaraldehyde or by simultaneous growth of IOR with betaine (carboxymethyltrimethylammonium) monohydrate at 10 mM initial concentration. Crystals were mounted using a cryo- protectant solution containing 30% PEG 4000 added to the appro- priate harvesting buffer. Values in parentheses are for the highest resolution shell. 2.3. X-ray diffraction experiments Diffraction experiments were carried out at the European Synchrotron Radiation Facility (ESRF) in Grenoble (France). A native data set was measured on a crystal grown in crystallization buffer A on beamline BM30A. Diffraction data were collected at 100 K at a wavelength of 0.976 AÊ to a maximum resolution of 2.5 AÊ . A two-wavelength MAD data set at the absorption K edge of Fe was measured on the ID29 beamline on native crystals grown in crystal- lization buffer B2. Crystals grown in buffer B2 were soaked for 1 h in 10 mM NaAsO2 to achieve inactivation of the protein by ligation of [AsO3]ÿ to Mo (Boer et al., 2004). A data set at the absorption K edge of As was measured at 100 K on the ID14-EH4 beamline. The data are summarized in Table 1. The native data were processed using the DENZO/SCALEPACK package (Otwinowski & Minor, 1997); all other data were processed using the XDS program (Kabsch, 1993). ² Rmerge = P h P i jIhi ÿ hIhij=P h P i Ihi. containing enzymes (Rebelo et al., 2000). The purity of the sample using the procedure described above is comparable to that reported earlier. 2.1. Protein purification Figure 1 Reaction scheme of the conversion of isoquinoline to isoquinoline-1-one, catalyzed by isoquinoline 1-oxidoreductase (IOR). Chemicals were ordered from Merck (Darmstadt, Germany) unless speci®ed otherwise. Cells were grown as described previously (Lehmann et al., 1994). Cells were lysed using an Emulsi¯ex-C5 French Press from Avestin (Canada) at 100 MPa in three passes. Isoquinoline 1-oxidoreductase activity was determined by formation of INT-formazan (Lehmann et al., 1994). Heat precipitation was performed at 328 K for 10 min. Triton X-100 precipitation was performed by adding pure Triton X-100 to the protein solution to a Figure 1 Figure 1 gu e Reaction scheme of the conversion of isoquinoline to isoquinoline-1-one, catalyzed by isoquinoline 1-oxidoreductase (IOR). 138 Boer et al. Isoquinoline 1-oxidoreductase Acta Cryst. (2005). F61, 137±140 3. Results and discussion The retention problems during the cation-exchange step and the increased af®nity in the hydrophobic interaction step were assumed to be a consequence of the high hydrophobicity of IOR, which causes it to interact with the (charged) lipids present in the cell lysate. The protein±lipid complexes that are expected to form have altered electrostatic properties and may not have af®nity for the cation- exchange resin. We hypothesized that binding competition of a non- ionic detergent with the charged lipids might result in protein±lipid complexes with electrostatic behaviour similar to that of the unbound protein and restore protein af®nity for the cation-exchange resin. To test this hypothesis, Triton X-100, a strong non-ionic detergent, was added to the cell lysate. A white ¯aky precipitation was observed and a clear red±brown supernatant solution was obtained after centrifu- gation. After applying the supernatant to the SQ Sepharose column, IOR was found to bind to the cation resin. A second cation-exchange step using a high-resolution Resource S column was used to compensate for the effect of the omission of the heat precipitation and HIC steps on the purity of the ®nal sample. Using this procedure, the yield was 4%. However, the speci®c activity decreased from 13.9 units mgÿ1 as determined by Lehmann to 8.1 units mgÿ1. Although the reason for this difference is not clear, it might be caused by a loss of the Mo atom as is often observed in other molybdopterin- Crystals grown in crystallization buffer A belong to space group P212121, with unit-cell parameters a = 67.3, b = 122.2, c = 239.1 AÊ , while crystals grown in buffer B2 have unit-cell parameters a = 66.5, b = 114.7, c = 237.5 AÊ (Table 1). The BaCl2 in buffer B2 has an impact on the b unit-cell parameter, which changes from 122.2 to 114.7 AÊ , a 6% decrease. This suggests a change in packing and concomitantly indicates non-isomorphism between crystals grown under the two different conditions. For both types of crystals, the Matthews coef®- cient is 2.4 AÊ 3 Daÿ1 and the solvent content is 49% assuming the presence of two molecules in the asymmetric unit. In the self-rotation function, however, only peaks corresponding to space-group symmetry are observed (see below). Although the sequence of IOR has low homology with other proteins of the XDH family, its overall fold is expected to be similar 139 Acta Cryst. (2005). Boer et al. Isoquinoline 1-oxidoreductase References Boer, D. R., Thapper, A., Brondino, C. D., Romao, M. J. & Moura, J. J. (2004). J. Am. Chem. Soc. 126, 8614±8615. Bonin, I., Martins, B. M., Purvanov, V., Fetzner, S., Huber, R. & Dobbek, H. (2004). Structure, 12, 1425±1435. to that found for other members of the family. This is based on the observation that domain A has been relocated in the primary sequence, but can be expected to appear in approximately the same region in space in the three-dimensional structure. Molecular replacement was therefore attempted using four structures of members of the XDH family (i.e. XDH from bovine milk CODH, QOR and MOP) as search models. The structures were stripped of water molecules and superposed using the DALI server (Holm & Sander, 1993), after which parts with low structural similarity were removed from the models by manual inspection. Extensive mole- cular-replacement calculations using different models in a variety of programs did not result in a usable set of starting phases. The molecular-replacement algorithm in PHASER (Storoni et al., 2004) using all four models did yield a clear solution and indicated the presence of two molecules in the asymmetric unit. The non- crystallographic symmetry element relating the two molecules is a twofold rotation axis parallel to the c axis of the unit cell and therefore parallel to the twofold crystallographic screw axis, which explains why no additional peaks are observed in the self-rotation function. Density-modi®cation calculations using the program DM from the CCP4 program package (Collaborative Computational Project, Number 4, 1994), starting from phases from PHASER and including non-crystallographic symmetry derived from the two molecules gave a clear solvent/protein boundary. Combination of the phases from DM and the anomalous signal measured at the wave- length of the arsenite absorption K edge of the arsenite-inhibited enzyme resulted in the anomalous maps shown in Fig. 2. A 14 peak appears close to the molybdenum site similarly located to that found in an analogously inhibited form of MOP that we have analyzed by crystallography (Boer et al., 2004). In addition, anomalous density appears on the Fe atoms of the two [2Fe±2S] clusters. Clearly, an initial set of phases has been obtained that is essentially correct, but the resulting electron-density maps are not of suf®cient quality to assign amino-acid residues along the backbone. Improvement of the Bordas, J., Bray, R. C., Garner, C. D., Gutteridge, S. & Hasnain, S. S. (1979). J. Inorg. References Biochem. 11, 181±186. Canne, C., Lowe, D. J., Fetzner, S., Adams, B., Smith, A. T., Kappl, R., Bray, R C & HÈ J (1999) Bi h i 38 14077 14087 Canne, C., Lowe, D. J., Fetzner, S., Adams, B., Smith, A. T., Kappl, R., Bray, R. C. & HuÈttermann, J. (1999). Biochemistry, 38, 14077±14087. C , C , , J, , S , , , S , , pp , , y, R. C. & HuÈttermann, J. (1999). Biochemistry, 38, 14077±14087. R. C. & HuÈttermann, J. (1999). Biochemistry, 38, 14077±14 Collaborative Computational Project, Number 4 (1994). Acta Cryst. D50, 760± 763. Cramer, S. P., Moura, J. J., Xavier, A. V. & LeGall, J. (1984). J. Inorg. Biochem. 20, 275±280. Dobbek, H., Gremer, L., Kiefersauer, R., Huber, R. & Meyer, O. (2002). Proc. Natl Acad. Sci. USA, 99, 15971±15976. Dobbek, H., Gremer, L., Meyer, O. & Huber, R. (1999). Proc. Natl Acad. Sci. USA, 96, 8884±8889. Enroth, C., Eger, B. T., Okamoto, K., Nishino, T. & Pai, E. F. (2000). Proc. Natl Acad. Sci. USA, 97, 10723±10728. Harrison, R. (2002). Free Radic. Biol. Med. 33, 774±797. Hille, R. (1996). Chem. Rev. 96, 2757±2816. ( ) Holm, L. & Sander, C. (1993). J. Mol. Biol. 233, 123±138. Huber, R., Hof, P., Duarte, R. O., Moura, J. J., Moura, I., Liu, M. Y., LeGall, J., Huber, R., Hof, P., Duarte, R. O., Moura, J. J., Moura, I., Liu, M. Y., LeGall, J., Hille, R., Archer, M. & Romao, M. J. (1996). Proc. Natl Acad. Sci. USA, 93, 8846±8851. Hille, R., Archer, M. & Romao, M. J. (1996). Proc. Natl Acad. Sci. USA, 93, 8846±8851. Kabsch, W. (1993). J. Appl. Cryst. 26, 795±800. Lehmann, M., Tshisuaka, B., Fetzner, S. & Lingens, F. (1995). J. Biol. Chem. 270, 14420±14429. Lehmann, M., Tshisuaka, B., Fetzner, S., Roger, P. & Lingens, F. (1994). J. Biol. Chem. 269, 11254±11260. Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951). J. Biol. Chem. 193, 265±275. , Monu, J. U. & Pope, T. L. Jr (2004). Radiol. Clin. North. Am. 42, 169±184. Okamoto, K., Matsumoto, K., Hille, R., Eger, B. T., Pai, E. F. & Nishino, T. (2004). Proc. Natl Acad. Sci. USA, 101, 7931±7936. ( ) Otwinowski, S. & Minor, W. (1997). Methods Enzymol. 276, 307±326. Rebelo, J. M., Dias, J. M., Huber, R., Moura, J. J. & Romao, M. J. (2001). J. Biol. Inorg. 3. Results and discussion F61, 137±140 Boer et al. Isoquinoline 1-oxidoreductase crystallization communications maps can be expected by including the anomalous differences of data sets measured at the As and Fe absorption edges. Figure 2 Anomalous electron-density map in the asymmetric unit contoured at 3.5 in red, calculated using PHASER phases combined with the anomalous difference of the Bijvoet pairs of the data set measured at the absorption peak of As. The model shown is MOP. The Mo active site and the two [2Fe±2S] clusters are shown as ball- and-stick representations and appear sequentially from the middle of the picture to the bottom left. In conclusion, we have found that applying the non-ionic detergent Triton X-100 in a precipitation step overcomes retention problems of IOR in the cation-exchange puri®cation step. The puri®ed protein was crystallized and initial diffraction experiments have been performed using PEG 4K as cryoprotectant in conjunction with cross- linking of the crystals. Native and anomalous data have been collected, which were used to obtain a molecular-replacement solu- tion and to validate the initial set of phases obtained. Our efforts are now aimed at improving the phases and interpreting the electron- density map, which includes the use of the anomalous data in the density-modi®cation calculations. We thank the EU for funding (HRRN-CT-1999-00084). We thank Alice Pereira, Cristina TimoÂteo and Sergey Bursakov for helpful suggestions regarding puri®cation and Shabir Najmudin for careful reading of the manuscript. We thank Mr Tony Page for growing the B. diminuta cells. DJL was supported by the BBSRC Core Strategic Grant to the John Innes Centre. Figure 2 g Anomalous electron-density map in the asymmetric unit contoured at 3.5 in red, calculated using PHASER phases combined with the anomalous difference of the Bijvoet pairs of the data set measured at the absorption peak of As. The model shown is MOP. The Mo active site and the two [2Fe±2S] clusters are shown as ball- and-stick representations and appear sequentially from the middle of the picture to the bottom left. 140 Boer et al. Isoquinoline 1-oxidoreductase References Chem. 6, 791±800. g Rebelo, J. M., Macieira, S., Dias, J. M., Huber, R., Ascenso, C. S., Rusnak, F., M J J M I & R M J (2000) J M l Bi l 297 135 146 Rebelo, J. M., Macieira, S., Dias, J. M., Huber, R., Ascenso, C. S., Rusnak, F., Moura, J. J., Moura, I. & Romao, M. J. (2000). J. Mol. Biol. 297, 135±146. Romao, M. J., Archer, M., Moura, I., Moura, J. J., LeGall, J., Engh, R., Schneider, M., Hof, P. & Huber, R. (1995). Science, 270, 1170±1176. SchaÈgger, H. & von Jagow, G. (1987). Anal. Biochem. 166, 368±379. Storoni, L. C., McCoy, A. J. & Read, R. J. (2004). Acta Cryst. D60, 432±438. Truglio, J. J., Theis, K., Leimkuhler, S., Rappa, R., Rajagopalan, K. V. & Kisker, C. (2002). Structure, 10, 115±125. Romao, M. J., Archer, M., Moura, I., Moura, J. J., LeGall, J., Engh, R., Romao, M. J., Archer, M., Moura, I., Moura, J. J., LeGall, J., Engh, R., Schneider, M., Hof, P. & Huber, R. (1995). Science, 270, 1170±1176. Romao, M. J., Archer, M., Moura, I., Moura, J. J., LeGall, J., Eng Schneider, M., Hof, P. & Huber, R. (1995). Science, 270, 1170±1176. Schneider, M., Hof, P. & Huber, R. (1995). Science, 270, 1170±1176. agger, H. & von Jagow, G. (1987). Anal. Biochem. 166, 368±37 Storoni, L. C., McCoy, A. J. & Read, R. J. (2004). Acta Cryst. D60, 432±438. Truglio, J. J., Theis, K., Leimkuhler, S., Rappa, R., Rajagopalan, K. V. & Kisker, C. (2002). Structure, 10, 115±125. 140 Boer et al. Isoquinoline 1-oxidoreductase 140 Acta Cryst. (2005). F61, 137±140 Acta Cryst. (2005). F61, 137±140
https://openalex.org/W4389564895	https://www.scielo.br/j/rbepid/a/xJPFvCRzcy3nN3DZN8rH4cc/?lang=en&format=pdf	English	null	Detection of sexually transmitted infections among transvestites and transsexual women in prison in the metropolitan region of Rio de Janeiro, Brazil	Revista brasileira de epidemiologia	2,023	cc-by	7,538	ERROR: type should be string, got "https://doi.org/10.1590/1980-549720230058\nRevista Brasileira de Epidemiologia\nwww.scielo.br/rbepid\nORIGINAL ARTICLE www.scielo.br/rbepid https://doi.org/10.1590/1980-549720230058\nRevista Brasileira de Epidemiologia Detection of sexually transmitted infections \namong transvestites and transsexual women \nin prison in the metropolitan region of \nRio de Janeiro, Brazil\nTestagem rápida de infecções sexualmente transmissíveis \nentre travestis e mulheres transexuais em situação de \nprisão na Região Metropolitana do Rio de Janeiro, Brasil\nCarlos Renato Alves-da-SilvaI,II\n, Claudia BonanI\n, \nSaint Clair dos Santos Gomes JuniorI\n, Rosilene Santarone VieiraI Testagem rápida de infecções sexualmente transmissíveis \nentre travestis e mulheres transexuais em situação de \nprisão na Região Metropolitana do Rio de Janeiro, Brasil\naI,II\n, Claudia BonanI\n, IFundação Oswaldo Cruz. Instituto Nacional de Saúde da Mulher, da Criança e do Adolescente Fernandes Figueira \n– Rio de Janeiro (RJ), Brazil. IIState Secretariat for Penitentiary Administration, LGBTI Health and Citizenship Support Division \n – Rio de Janeiro (RJ), Brazil. ABSTRACT Objective: To evaluate the seropositivity rate of rapid tests for HIV, syphilis and hepatitis B and C among transvestites and \ntransgender women (transfeminine persons) inmates in the metropolitan region of Rio de Janeiro, analyzing the results based on \nsociodemographic, prison profile and access to health technologies to prevent sexually transmitted infections (STIs). Methods: Cross-\nsectional census-type study carried out with transfeminine in eleven male prisons in Rio de Janeiro, between the months of April \nand June 2021. Results: The detection rates found were 34.4% for HIV, and 48.9% for syphilis, and 0.8% for type B and C hepatitis. Seropositivity for more than one infection was verified in 25.4% of participants, and HIV/syphilis was the most prevalent. An increase \nin the level of education (p=0.037) and having a steady partner in prison (p=0.041) were considered protective factors for STIs in \nthis population. Difficulties were identified in accessing STI prevention technologies, such as male condoms, lubricating gel, rapid \ntests, and prophylactic antiretroviral therapies for HIV. Conclusion: HIV and syphilis seropositivity rates were high, but within the \nprofile found in this population in other studies inside or outside prisons. The data found indicates the need to incorporate effective \nstrategies for access to health technologies for the prevention of STIs. The scarcity of scientific publications containing epidemiological \ndata on STIs in the transfeminine prison population limited deeper comparisons of the results obtained in this study. nder persons. Prisons. Syphilis. Hepatitis. HIV seropositivity. Keywords: Transgender persons. Prisons. Syphilis. Hepatitis. HIV seropositivity. CORRESPONDING AUTHOR: Carlos Renato Alves-da-Silva. Avenida Rui Barbosa, 716, Flamengo, CEP: 22250-020, Rio de Janeiro (RJ), Brasil. E-mail: carlos-renato. INTRODUCTION It is observed that transvestites \nand transsexual women, also defined as transfeminine \npeople, make up a target group for greater discrimination \nand violence, both due to the difficulties experienced in ed-\nucational establishments and the lack of opportunities in \nthe formal job market and due to their very existence, since \nthey live in the country with the highest transgender death \nrate in the world6. Some studies show the conditions of deprivation of lib-\nerty of transfeminine women, revealing scenarios of trans-\nphobic practices and violations of these people’s rights to \ncitizenship and health7,8, such as respect for the use of their \nsocial names, the use of clothes and accessories according \nto their gender, the guarantee of hormone treatment, ac-\ncess to and continuity of their education, among other pro-\nvisions described in the National Policy for Comprehensive \nHealth of the LGBT Population9 and in the legislation that \nprovides for the treatment of these people in the Brazilian \npenitentiary system. The spread of sexually transmitted infections (STIs) \nin prisons is one of the problems to be tackled by prison \nhealth teams, faced with environments with conditions \nconducive to unsafe sexual practices, as described in na-\ntional and international studies15-18. However, it is neces-\nsary to know and analyze this reality in order to promote \neffective actions to reduce the risks of STIs, such as com-\nbined strategies, in order to improve access to condoms, \nHIV prophylactic therapies, rapid STI testing regularly ev-\nery six months, among other actions recommended by the \nMinistry of Health (MoH)19. The National Penitentiary Department (Departamen-\nto Penitenciário Nacional – Depen) recently published a \npopulation survey, referring to the second half of 2022, \nrecording that, in Brazil, there were 830 thousand peo-\nple deprived of liberty (PDL), however, of this total, \n650 thousand were serving their criminal sentences in \nprison units and the rest of them were under house \narrest. This census also revealed that around 12,500 \nself-declared LGBTI people were imprisoned in Brazil-\nian prison units1, of which approximately 1,600 were \ntransvestites and transsexual women. The methodology \nadopted in this survey to quantify this population was \nbased on completing a questionnaire sent by Depen \nto the bodies of all federative units responsible for the \npenitentiary system. INTRODUCTION sexual women being deprived of liberty, occupying second \nplace in the national ranking, in absolute number, in custo-\ndy of the transfeminine population10. The majority of prison spaces in Brazil are reported due \nto prison overcrowding, inadequate physical structure, de-\nficiencies in human resources, and specialized materials \nfor health assistance and promotion, among other situa-\ntions mentioned in publications that highlighted the poten-\ntial of these spaces as human rights violators1,2. A technical document, published by the Ministry of \nWomen, Family and Human Rights, in 2020, containing the \nnational diagnosis of the criminal treatment of this popu-\nlation in Brazil, revealed that the imprisonment of trans-\nfeminine women in our country occurs, for the most part, \nin prisons aimed at male inmates and, in large part, at an \nalarming disproportion, with them being a minority of the \ntotal number of people imprisoned11. These conditions \ndrastically affect the overall health of these people, wheth-\ner due to possible psychological damage or harm to their \nphysical health. Imprisonment conditions of specific groups, such as \nwomen, the aged, indigenous people, foreigners, people \nwith disabilities, and people self-declared as lesbian, gay, \nbisexual, transgender, and intersex (LGBTI), are marked \nby the exacerbation of these violations, in the face of \nnon-compliance of the duty of care inherent to the vulner-\nabilities established by each specific group, as defined in \nsome national and international regulations3-5. Health care access and quality in the Brazilian pen-\nitentiary system is one of the greatest challenges to be \nfaced. The National Policy for Comprehensive Penitentiary \nHealth Care (Política Nacional de Atenção Integral à Saúde \nPenitenciária – Pnaisp) has emerged as a powerful manage-\nment strategy to resolve these obstacles in several Brazilian \nstates. This policy aims to guarantee access to comprehen-\nsive primary health care by PDL12,13. The implementation of \nPnaisp, in RJ, intensified, from January 2022, with the acces-\nsion of the capital of Rio de Janeiro, however, during the \nperiod if this study, only four units, of the total of 46 prison \nunits with outpatient care, had Pnaisp teams14. Social vulnerability among people from the LGBTI popu-\nlation is not homogeneous. https://doi.org/10.1590/1980-549720230058\nRevista Brasileira de Epidemiologia\nwww.scielo.br/rbepid\nORIGINAL ARTICLE silva@fiocruz.br HOR: Carlos Renato Alves-da-Silva. Avenida Rui Barbosa, 716, Flamengo, CEP: 22250-020, Rio de Janeiro (RJ), Brasil. E-mail: carlos-renato CONFLICT OF INTERESTS: nothing to declare HOW TO CITE THIS ARTICLE: Alves-da-Silva CR, Bonan C, Gomes Junior SCS, Vieira RS. Detection of sexually transmitted infections among transvestites and \ntranssexual women in prison in the metropolitan region of Rio de Janeiro, Brazil. Rev Bras Epidemiol. 2023; 26: e230058. https://doi.org/10.1590/1980-549720230058 This is an open article distributed under the CC-BY 4.0 license, which allows copying and redistribution of the material in any format and for any purpose as long as \nthe original authorship and publication credits are maintained. This is an open article distributed under the CC-BY 4.0 license, which allows copying and redistribution of the material in any format and for any purpose as long as \nthe original authorship and publication credits are maintained. Received on: 07/12/2023\nReviewed on: 09/27/2023\nAccepted on: 10/04/2023 Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n1 https://doi.org/10.1590/1980-549720230058 www.scielo.br/rbepid INTRODUCTION This study aimed to evaluate the isolated and com-\nbined seropositivity rate for HIV, syphilis, and type B and \nC hepatitis, through the use of rapid tests among trans-\nfeminines deprived of liberty, in the Metropolitan Region \nof Rio de Janeiro and to identify potential associated fac-\ntors, considering sociodemographic characteristics, pris-\non conditions and access to health technologies for the \nprevention of STIs, such as male condoms, lubricants, and \nrapid tests. Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n2 Study design This is a cross-sectional study with a quantitative ap-\nproach in a census format, that is, involving the entire \nself-declared adult transfeminine population deprived of \nliberty in the Metropolitan Region of Rio de Janeiro. The prison population in the state of Rio de Janeiro (RJ), \nrecorded in the same publication, revealed that around 48 \nthousand people were in RJ’s prisons, among which 579 de-\nclared themselves LGBTI, with 176 transvestites and trans- www.scielo.br/rbepid https://doi.org/10.1590/1980-549720230058 Participants Transfeminine PDL were previously selected, by con-\nsulting the Penitentiary Identification System (Sistema de \nIdentificação Penitenciária – Sipen) of the State Secretariat \nfor Penitentiary Administration (Secretaria de Estado de Ad-\nministração Penitenciária – Seap-RJ), a nominal database up-\ndated daily, through information about the people entering \nthe RJ penitentiary system. Access to this database was pos-\nsible, as one of the researchers is a public server pharmacist \nwho held the position of director of the LGBTI Health and \nCitizenship Support Division at Seap (RJ). The study also ad-\nopted the “snowball” method, to identify possible transfem-\ninines not included in Sipen20, when the selected participant \nreported the existence of transfeminines not registered in \nthis database, being subsequently invited to participate in \nthe research. All people who declared themselves trans-\nfeminine were included, using terms such as transvestite, \ntranny, transsexual woman, trans, transex, woman or any \nreference to transfeminine identity. Tran feminine women \nwho had been in prison for less than six months were ex-\ncluded, the time period recommended by the MoH for rap-\nid testing among transvestites and transsexual women in \nprison19. Interviewees who did not have the results of rapid \ntests in the last six months in their medical records and who \nexpressed opposition to carrying out rapid testing after the \ninterview were also excluded. Background Table 1. Socioeconomic, demographic, and prison \ncharacteristics of transfeminines deprived of liberty in \nthe Metropolitan Region of Rio de Janeiro, Brazil, 2021. Eleven male prisons that held transfeminine people \nduring the study period, with their respective locations, \ntotal prison population, number of transfeminine partici-\npants, and the percentage rate between participants and \nthe total prison population (Chart 1). It is clarified that, \nduring the study period, there were no transfeminine pris-\noners in female prison units and that the selected units are \nclassified as neutral, to which LGBTI people are sent in RJ. the Metropolitan Region of Rio de Janeiro, Brazil, 2021. N sample: 131 people. Characteristic – socioeconomic\nn\n%\nAge range (years)\n18–29\n85\n64.9\n30–39\n34\n25.9\n40 and over\n12\n9.2\nRace/color\nNon-white\n113\n86.3\nWhite\n18\n13.7\nEducation\nIncomplete elementary school \n65\n49.6\nComplete elementary school\n4\n3.0\nIncomplete High School\n37\n28.3\nComplete High School \n19\n14.5\nComplete and incomplete higher education\n6\n4.6\nSexual orientation\nHeterosexual\n123\n93.9\nBisexual\n8\n5.3\nPaid activity before imprisonment\nWas not working\n23\n17.6\nSex worker\n55\n42.0\nFormal or informal employment (beauty, \nfashion, food, and others)\n53\n40.4\nUse of illicit substances before the current arrest\nYes\n106\n80.9\nNo\n25\n19.1\nCharacteristic – prison\nLength of stay in prison (years)\nUp to 1 \n59\n45.0\nMore than 1 \n72\n55.0\nProportion between transfeminines and men in the cell\n1/1 to 1/10\n20\n15.3\nOver 1/10\n111\n84.7\nSteady partner in prison\nYes\n77\n58.8\nNo\n54\n41.2\nNumber of partners in prison\nUp to 1\n22\n16.8\nMore than 1\n109\n83.2\nRegistered visitor\nYes\n65\n49.6\nNo\n66\n50.4\nLabor activity in prison\nYes\n23\n17.6\nNo\n108\n82.4\nEducation in prison\nYes\n26\n19.8\nNo\n105\n80.2 RESULTS 151 self-declared transfeminine people were identified, \nhowever 13 people had been in prison for less than six \nmonths and seven did not consent to rapid tests for the \nSTIs analyzed in this study. The database identified 92 of \nthe selected transfeminines (60.1%), the other interview-\nees were reached by the snowball method. Recruitment of \ninterviewees was carried out in 11 prison units (Chart 1), \nand it was found that four prison units had health teams, \nthrough the implementation of Pnaisp, and five prison \nunits had prison schools. Data relating to the sociodemo-\ngraphic and prison profile of the interviewees were collect-\ned from a sample composed of 131 transfeminines (Table \n1), demonstrating that 85 were young adults under the age \nof 29 (64.9%), and, not unlike the Brazilian prison popula-\ntion, the sample consisted of 113 black and brown women \n(86.3%). Education demonstrated that only 25 interviewees \nmanaged to complete high school education (19.1%) and 65 \ndid not complete elementary education (49.6%). The major-\nity, that is, 123 transfeminines, revealed that they were het-\nerosexual (93.9%). Prostitution was the most cited source \nof income for 55 of the interviewees (42.0%), and the use of \nillicit substances before imprisonment was revealed by 106 \ntransfeminines (80.9%). The situation inside prisons also \nrevealed that 72 interviewees had been in prison for more \nthan a year (55.0%) and 111 were incarcerated with cisgen-\nder and heterosexual men, in a proportion greater than \n1/10 (84.7%). It was revealed that 109 of the interviewees \nhad relationships with two or more partners throughout \ntheir imprisonment (83.2%) and, at the time of the inter-\nviews, 77 revealed that they were in a steady relationship \n(58.8%). Among those interviewed, 66 did not have regis-\ntered visitors (50.4%) and few were involved, or had been \ninvolved in resocializing practices, with records that 23 \nwere working or have worked in prisons (17.6%) and only \n26 had the opportunity to enter prison schools (19.8%). Data analysis Collected data were analyzed using the statistical pro-\ngram Jasp (Jeffrey’s Amazing Statistics Program) version \n0.17.1.0, evaluating percentage analyses of sociodemo-\ngraphic, prison and access to health technology variables, \nas well as the results for each rapid STI test evaluated. The variable called “positive result in at least one of the \nSTI tests” was created to evaluate associations with socio-\ndemographic variables, prison conditions, and access to \nhealth care technologies dichotomized to find statistical \nsignificance. The chi-square test was applied in all analyses, \nwith a statistical significance level of 5% (p<0.05). The results on access by transfeminines to male con-\ndoms, lubricants, and rapid testing for STIs (Table 2) \ndemonstrated that 29 had difficulty accessing male con-\ndoms (22.1%) and 95, lubricants (72.5%). It was found that \n37 subjects were on antiretroviral therapy (ART) for HIV \n(28.2%), thus, 94 health records were consulted to evalu-\nate the performance of rapid testing (71.8%), in which 41 Study variables The sociodemographic variables analyzed were: age \ngroup in years (from 18 to 29, from 30 to 39, over 40); race/\ncolor (white, non-white); education (incomplete and com-\nplete elementary school, incomplete and complete high \nschool, higher education); sexual orientation (heterosexu-\nal, bisexual); paid activity before prison (not working, sex \nworker, formal or other informal employment); and use \nof illicit substances before arrest (yes, no). The prison vari-\nables were: length of stay in prison (up to one year, over \none year); proportion between transfeminines and men in \nthe same cell (from 1/1 to 1/10, over 1/10); steady partner \nin prison (yes, no); number of partners in prison (up to one, \nmore than one); registered visitor (yes, no); work activity \nin prison (yes, no); schooling in prison (yes, no). The health \ncare technologies were: access to male condoms (yes, no); \naccess to lubricant (yes, no); carrying out rapid tests in the \nlast six months (yes, no); and rapid test positivity (yes, no). The variables analyzed were considered factors associated \nwith STIs, with some being used in some studies15,21. Data collect Data collect was carried out in the period between April \nand June 2021, using a structured questionnaire containing \nquestions with closed answers and selecting people who \ndeclared themselves transvestites or transsexual women. This information allowed to describe the sociodemographic \nand prison profiles, as well as access to health technologies \nused to prevent STIs: male condoms, lubricants, and rapid \ntesting. However, the variables related to length of stay and \nexistence of registered visitors were extracted from Sipen, \na strategy adopted to improve data accuracy. Data relating \nto the results of serological tests for HIV, syphilis, and hep-\natitis B and C in the last six months, as recommended by \nthe MoH, were extracted from the health records of the in-\nterviewees. When this information was not available, rapid \ntests were requested to be carried out by the prison unit’s \nhealth team, and no situation of rapid test shortages, inac-\ncessibility to health records or refusal by health profession- Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n3 www.scielo.br/rbepid https://doi.org/10.1590/1980-549720230058 36338520.2.0000.5269, authorized by the Criminal Execu-\ntions Court of the Court of Justice of the state of Rio de \nJaneiro and by the School of Penitentiary Management of \nSeap-RJ, through Process SEI-21/087/000987/2019. All eth-\nical and safety issues involved in research with PDL were \nobserved, including informed consents signed by all study \nparticipants. All interviewees with positive rapid test re-\nsults were assisted by the prison unit’s health teams, be-\ning included in institutional clinical protocols for the treat-\nment of STIs. als to carry out rapid tests were observed. These profes-\nsionals are responsible for inputting the results in medical \nrecords, as well as confirming the results, through confir-\nmatory laboratory tests, according to standardized opera-\ntional procedures established for each infection. The rapid \ntests recorded in the medical records and carried out after \nthe interviews were based on immunochromatography \ntechnology, using whole blood finger-prick test. All rapid tests were made available to prison units in Rio \nde Janeiro, through the State Secretariat of Health of Rio \nde Janeiro (Secretaria de Estado da Saúde do Rio de Janeiro \n– SES-RJ) and by the municipal Health Departments of the \nrespective cities where the prison units were located, and \nwhich had health teams hired by Pnaisp. s e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n4 Ethical aspects %\nPresídio Evaristo de Moraes)\nRio de Janeiro\n3,600\n36\n1.0\nCadeia José Antônio Costa Barros† \nRio de Janeiro\n720\n11\n1.5\nInstituto Benjamim de Moraes Filho\nRio de Janeiro\n1,964\n17\n0.9\nPresídio Nelson Hungria \nRio de Janeiro\n830\n8\n1.0\nInstituto Plácido de Sá Carvalho\nRio de Janeiro\n960\n10\n1.0\nPresídio Cotrim Neto† \nJaperi\n1,241\n6\n0.5\nPresídio Milton Dias Moreira\nJaperi\n1,725\n4\n0.2\nCadeia Tiago Silva Teles,† \nSão Gonçalo\n1,303\n8\n0.6\nCadeia Juíza Patrícia Acioli \nSão Gonçalo\n1,583\n8\n0.5\nPresídio Edgar Costa† \nNiterói\n524\n9\n1.7\nPresídio Romeiro Neto† \nMagé\n1,303\n14\n1.1\nTotal\n15,753\n131\n0,8\n\u0003\nSource: Seap-RJ, 2021. existence of the Pnaisp health team; †no prison school. Name of prison unit\nMunicipality\nTotal PDL\nTrans. %\nPresídio Evaristo de Moraes)\nRio de Janeiro\n3,600\n36\n1.0\nCadeia José Antônio Costa Barros† \nRio de Janeiro\n720\n11\n1.5\nInstituto Benjamim de Moraes Filho\nRio de Janeiro\n1,964\n17\n0.9\nPresídio Nelson Hungria \nRio de Janeiro\n830\n8\n1.0\nInstituto Plácido de Sá Carvalho\nRio de Janeiro\n960\n10\n1.0\nPresídio Cotrim Neto† \nJaperi\n1,241\n6\n0.5\nPresídio Milton Dias Moreira\nJaperi\n1,725\n4\n0.2\nCadeia Tiago Silva Teles,† \nSão Gonçalo\n1,303\n8\n0.6\nCadeia Juíza Patrícia Acioli* \nSão Gonçalo\n1,583\n8\n0.5\nPresídio Edgar Costa† \nNiterói\n524\n9\n1.7\nPresídio Romeiro Neto† \nMagé\n1,303\n14\n1.1\nTotal\n15,753\n131\n0,8\n\u0003\nSource: Seap-RJ, 2021. existence of the Pnaisp health team; †no prison school. p\nJ,\nexistence of the Pnaisp health team; †no prison school. profile for hepatitis types B and C proved to be identical \nfor both infections, since, throughout this study, it was ver-\nified that two interviewees were already undergoing treat-\nment for these infections, one for hepatitis B (0.8%) and \none for hepatitis C (0.8%). No positive results were found \nin the 55 medical records with records of testing for hep-\natitis in the last six months (42.0%), as well as no positive \nresults in the 75 rapid tests carried out after the interviews \n(57.2%). The serological STIs profile through rapid testing \nof the interviewees revealed that 71 had positive results \nfor at least one of the four STIs tested in this study (54.2%). Ethical aspects The research was duly approved by the Research \nEthics Committee of the Fernandes Figueira National In-\nstitute of Women, Children and Adolescent Health (Insti-\ntuto Nacional de Saúde da Mulher, da Criança e do Adoles-\ncente Fernandes Figueira – IFF-Fiocruz), under CAAE No. Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n4 www.scielo.br/rbepid https://doi.org/10.1590/1980-549720230058 Chart 1. Map of the Metropolitan Region of Rio de Janeiro and the selected prison units with their respective population \ndata, 2021. Name of prison unit\nMunicipality\nTotal PDL\nTrans. %\nPresídio Evaristo de Moraes)\nRio de Janeiro\n3,600\n36\n1.0\nCadeia José Antônio Costa Barros† \nRio de Janeiro\n720\n11\n1.5\nInstituto Benjamim de Moraes Filho\nRio de Janeiro\n1,964\n17\n0.9\nPresídio Nelson Hungria \nRio de Janeiro\n830\n8\n1.0\nInstituto Plácido de Sá Carvalho\nRio de Janeiro\n960\n10\n1.0\nPresídio Cotrim Neto† \nJaperi\n1,241\n6\n0.5\nPresídio Milton Dias Moreira\nJaperi\n1,725\n4\n0.2\nCadeia Tiago Silva Teles,† \nSão Gonçalo\n1,303\n8\n0.6\nCadeia Juíza Patrícia Acioli \nSão Gonçalo\n1,583\n8\n0.5\nPresídio Edgar Costa† \nNiterói\n524\n9\n1.7\nPresídio Romeiro Neto† \nMagé\n1,303\n14\n1.1\nTotal\n15,753\n131\n0,8\n\u0003\nSource: Seap-RJ, 2021. *existence of the Pnaisp health team; †no prison school. Chart 1. Map of the Metropolitan Region of Rio de Janeiro and the selected prison units with their respective population \ndata, 2021. Chart 1. Map of the Metropolitan Region of Rio de Janeiro and the selected prison units with their respective population \ndata, 2021. Map of the Metropolitan Region of Rio de Janeiro and the selected prison units with their respective\n1 Chart 1. Map of the Metropolitan Region of Rio de Janeiro and the selected prison units with their respective population \ndata, 2021. Chart 1. Map of the Metropolitan Region of Rio de Janeiro and the selected prison units with their respective population \ndata, 2021. Name of prison unit\nMunicipality\nTotal PDL\nTrans. Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n5 Ethical aspects Among these 71 transfeminines, it was found that 19 pre-\nsented more than one positive result in the rapid tests car-\nried out (26.8%), in which the most prevalent combined se- interviewees were tested for HIV (43.6%), among whom, \nthree were positive for the rapid HIV test (2.3%). Among \nthe 53 who underwent rapid testing after the interview, \nfive tested positive for HIV (9.4%). The total of these results \nrevealed that 45 of the transfeminines tested positive for \nHIV (34.4%). At the time of the interview, none were ob-\nserved undergoing treatment for syphilis, and the syphilis \ntesting profile revealed that 56 transfeminines had records \nof testing in the last six months (42.8%). The results of the \nrapid test for syphilis in the last six months from the date \nof the interview revealed that 56 were positive for this in-\nfection (69.6%), while 25 of them had positive results after \nthe interview (33.3%). Thus, among them, it was found that \n64 people were positive for syphilis (48.9%). The testing Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n5 www.scielo.br/rbepid https://doi.org/10.1590/1980-549720230058 ropositivity was HIV/syphilis, which accounted for 18 of the \ncases (94.7%). The associations made between the results \nof rapid tests and sociodemographic, prison characteristics \nand access to health care technologies used to prevent STIs \n(Table 3) demonstrated that, among the 46 transfeminine \nwomen over the age of 29, 31 had positive results for STIs \n(67.4%), compared to 85 aged between 18 and 29 years, \nin which there were 40 positive results (47.1%); (p=0.026). 44 positive tests for STIs were revealed among the 70 in-\nterviewees who only attended elementary school (62.8%), \ncompared to 27 positive results among the 61 who attend-\ned high school and higher education (44.3%); (p=0.033). Of \nthe 54 who declared not having a steady partner, 35 had \nmore positive results for STIs (64.8%), compared to the 36 \npositive results among the 77 who reported having emo-\ntional stability (46.8%); (p=0.041). Table 2. Access to health technologies associated with \nsexually transmitted infections and rapid test results \namong transfeminine women deprived of liberty in the \nMetropolitan Region of Rio de Janeiro, Brazil, 2021. g\np\ny\nMetropolitan Region of Rio de Janeiro, Brazil, 2021. Ethical aspects Characteristic\nn\n%\nAccess to male condoms\nYes\n29/131\n22.1\nNo \n102/131\n77.9\nAccess to lubricants\nYes\n95/131\n72.5\nNo \n36/131\n27.5\nCarrying out the rapid test for HIV\nNo, subject already undergoing treatment\n37/131\n28.2\nRegistered in the medical record, \nwith a positive result \n3/131\n2.3\nRegistered in the medical record, \nwith a negative result\n38/131\n29.0\nCarried out during the interview, \nnegative result\n48/131\n36.6\nCarried out during the interview, \npositive result\n5/131\n3.9\nCarrying out the rapid test for syphilis\nNo, subject already undergoing treatment\n0/131\n0.0\nRegistered in the medical record, \nwith a positive result \n39/131\n29.8\nRegistered in the medical record, \nwith a negative result\n17/131\n13.0\nCarried out during the interview, \nnegative result\n50/131\n38.1\nCarried out during the interview, positive result\n25/131\n19.1\nCarrying out the rapid test for hepatitis B\nNo, subject already undergoing treatment\n1/131\n0.8\nRegistered in the medical record, \nwith a positive result \n0/131\n0.0\nRegistered in the medical record, \nwith a negative result\n55/131\n42.0\nCarried out during the interview, negative result\n75/131\n57.2\nCarried out during the interview, positive result\n0/131\n0.0\nCarrying out the rapid test for hepatitis C\nNo, subject already undergoing treatment\n1/131\n0.8\nRegistered in the medical record, \nwith a positive result \n0/131\n0.0\nRegistered in the medical record, \nwith a negative result\n55/131\n42.0\nCarried out during the interview, negative result\n75/131\n57.2\nCarried out during the interview, positive result\n0/131\n0.0\nPositive result in at least one of the STI tests\nYes\n71/131\n54.2\nNo\n60/131\n45.8\nFrequency of positive results for STIs\nOnly one STI\n52/71\n73.2\nHIV/syphilis\n18/71\n25.4\nHIV/hepatitis B\n1/71\n1.4\nSTI: sexually transmitted infections; HIV: human immunodeficiency virus. Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n6 Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n7 DISCUSSION The observation that more than half of the medical re-\ncords checked did not have records of regular rapid test-\ning for STIs demonstrated the fragility of the health care \noffered to this population. The detection rate for HIV was \nsimilar when compared with other national studies with \nthis population outside prisons22,23 and with a study carried \nout in Mexico24. The majority of interviewees were young \nadults, black or brown, with low education, and with pros-\ntitution as their main source of income before imprison-\nment. A higher level of education, having a steady partner \nin prison, and younger age proved to be protective factors \nfor the STIs analyzed. The absence of records of rapid testing in the medical \nrecords can be translated into the difficulties in accessing \nhealth services reported in the various studies with the \ntransfeminine population, both due to the technical lack of \npreparation of the health professionals who assist these \npeople and the Brazilian penitentiary system itself, an en-\nvironment that enhances violations of the health rights of \nthe population deprived of liberty25,26. The HIV detection rate of 34.4% found in this study, \nwas similar to that found in 2012 in Mexico City24, where \n31.9% of incarcerated transfeminines tested positive for \nHIV in rapid tests. In the systematic review with meta-anal-\nysis published in 201827, which searched for studies on \nSTIs among key incarcerated populations, only two articles \nwere found, one carried out in the United States of America \nand the other in Argentina, signaling the scarcity of studies \nwith this population in the prison environment, as already \nrecorded in research published in 201928. The few epide-\nmiological data published on HIV and the lack of data for \nsyphilis and hepatitis B and C in the transfeminine popu-\nlation in prisons were limiting factors for comparative pur-\nposes. The sociodemographic and prison profile demon-\nstrated a known situation in relation to transfeminines in \nprison7,29,30, a reality composed of factors associated with https://doi.org/10.1590/1980-549720230058 www.scielo.br/rbepid Table 3. Association between the results of rapid testing for sexually transmitted infections and sociodemographic, \nprison characteristics and access to health technologies among transfeminine women deprived of liberty in the \nMetropolitan Region of Rio de Janeiro, Brazil, 2021. Table 3. REFERENCES trans necropolitics, such as early family separation, diffi-\nculties in remaining in school environments and accessing \nthe labor market, prostitution on the streets, surrounded \nby drugs and theft, facilitating conditions that lead them \nto prison or death in the broadest sense31. The conditions \nof incarceration observed in this study, in which transfem-\ninines share spaces with cisgender and heterosexual men, \nare comparable to those portrayed in a study carried out in \nthe state of São Paulo8 and in the national report published \nin 2020, situations favorable to sexual practices motivated \nby several factors, both due to economic favoritism and \nemotional relationships11. 1. Sánchez A, Toledo CRS, Camacho LAB, Larouze B. Mortalidade \ne causas de óbitos nas prisões do Rio de Janeiro, Brasil. Cad Saúde Pública 2021; 37(9): e00224920. http://dx.doi. org/10.1590/0102-311X00224920 2. Rangel FM, Bicalho PPG. Superlotação das prisões brasileiras: \noperador político da racionalidade contemporânea. Estud Psicol (Natal) 2016; 21(4): 415-23. https://doi. org/10.5935/1678-4669.20160040 3. Nota Técnica no 9/2020/DIAMGE/CGCAP/DIRPP/DEPEN/MJ: \ncustódia de pessoas LGBTI. Revista Brasileira Execução Penal \n2021; 2(2): 267-91. https://doi.org/10.1234/rbep.v2i2.395 The abandonment of this population in prisons occurs \nin similar proportions to cisgender women. As reported \nin the study carried out in 2019, around 60% of women \nincarcerated in Brazilian prisons did not receive visits from \nfamily members32. Family abandonment among transfem-\ninines begins, in most cases, before prison, a situation that \noften leads them to prostitution, which, despite being a \nrecognized profession in Brazil, cannot be the only fate for \nthese people6,33,34. 4. Rio de Janeiro (Estado). Resolução da Secretaria de Estado \nde Administração Penitenciária (SEAP) no 558, de 29 de \nmaio de 2015. Estabelece diretrizes e normativas para o \ntratamento da população LGBT no sistema penitenciário \ndo estado do Rio de Janeiro [Internet]. Diário Oficial do \nEstado do Rio de Janeiro de 03 de junho de 2015 [cited \non Jul 7, 2023]. Available at: https://www.ioerj.com.br/\nportal/modules/conteudoonline/view_pdf.php?ie=M-\njI1OTQ=&ip=MTQ=&s=YmYzNTM0MjdmMGNkNWUzN-\njhiZWUzM2YyNmMxYjM5ZTY=&directlink=1&Ori-\ngn=WebIndexer 4. Rio de Janeiro (Estado). Resolução da Secretaria de Estado \nde Administração Penitenciária (SEAP) no 558, de 29 de \nmaio de 2015. Estabelece diretrizes e normativas para o \ntratamento da população LGBT no sistema penitenciário \ndo estado do Rio de Janeiro [Internet]. Diário Oficial do \nEstado do Rio de Janeiro de 03 de junho de 2015 [cited \non Jul 7, 2023]. REFERENCES Available at: https://www.ioerj.com.br/\nportal/modules/conteudoonline/view_pdf.php?ie=M-\njI1OTQ=&ip=MTQ=&s=YmYzNTM0MjdmMGNkNWUzN-\njhiZWUzM2YyNmMxYjM5ZTY=&directlink=1&Ori-\ngn=WebIndexer Schooling was considered a protective factor for STIs, \nas demonstrated in other studies with the transfeminine \npopulation outside prisons35, however, the factors that in-\ndicated lower detection rates for these infections among \nyounger transfeminines and those with a steady partner \nrequire further studies to purposes of comparisons and \nextrapolations. The 5.1% refusal to undergo rapid STI tests \nobserved in this study was higher than that found in a \nstudy carried out in 2007 with cisgender women in prison \nin São Paulo, when 3% of them did not agree to partici-\npate in the tests36. It is important to highlight that, often, \nthe refusal to carry out rapid tests for STIs, especially for \nHIV, is directly related to the fear of receiving difficult news \nand the stigma of the repercussion of this result within the \nprison unit, whether among other transfeminine wom-\nen by pride and power37, or by the violence that can be \nperpetrated by their sexual partners. Ther fore, it is nec-\nessary for health teams to have strategies for an appro-\npriate approach, evaluating the psychological and phys-\nical consequences of carrying out rapid STI tests inside \nprisons. More than revealing epidemiological data, it is \nnecessary to understand social dynamics, knowledge, at-\ntitudes, and practices established within Brazilian prisons \nto adopt effective preventive measures aimed at mitigat-\ning the spread of STIs, which make transfeminine women \neven more vulnerable, as assessed in the systematic re-\nview published in 201938. It is essential that prison health \nteams can adopt combined prevention actions to combat \nthese diseases, through educational campaigns on mea-\nsures to prevent and treat these infections, regular rapid \ntesting and incorporation and promotion of access to pro-\nphylactic technologies to the risk of STIs. And, above all, \nthat these actions can accommodate the specific health \ndemands of this population in the prison reality. 5. Alamino FNP, Del Vecchio VA. Os princípios de Yogyakarta \ne a proteção de direitos fundamentais das minorias de \norientação sexual e de identidade de gênero. R Fac Dir Univ \nSão Paulo 2018; 113: 645-68. https://doi.org/10.11606/\nissn.2318-8235.v113i0p645-668 6. Benevides BG. Dossiê. Assassinatos e violências contra \ntravestis e transexuais brasileiras em 2022 [Internet]. Brasília: Associação Nacional de Travestis e Transexuais; \n2023 [cited on Jul 20, 2023]. Available at: https://antrabrasil. files.wordpress.com/2023/01/dossieantra2023.pdf 7. Ferreira GG. Violência, interseccionalidades e \nseletividade penal na experiência de travestis \npresas. Temporalis 2014; 14(27): 99-117. https://doi. DISCUSSION Association between the results of rapid testing for sexually transmitted infections and sociodemographic, \nprison characteristics and access to health technologies among transfeminine women deprived of liberty in the \nMetropolitan Region of Rio de Janeiro, Brazil, 2021. prison characteristics and access to health technologies among transfeminine women deprived of liberty in the \nMetropolitan Region of Rio de Janeiro, Brazil, 2021. Characteristic\nPositive result for STI\nN\nYes\nNo\np-value\nn\n%\nn\n%\nAge range (years)\n19–29\n85\n40\n47.1\n45\n52.9\n0.026\nOver 29\n46\n31\n67.4\n15\n32.6\nRace/color\nWhite\n113\n62\n54.9\n51\n45.1\n0.700\nNon-white\n18\n9\n50.0\n9\n50.0\nEducation\nElementary school\n70\n44\n62.8\n26\n37.2\n0.033\nHigh School and Higher education\n61\n27\n44.3\n34\n55.7\nSexual orientation\nHeterosexual\n123\n66\n53.7\n57\n46.3\n0.627\nBisexual\n8\n5\n62.5\n3\n37.5\nPaid activity before imprisonment\nSex worker\n55\n30\n54.5\n25\n45.5\n0.946\nFormal or other informal employment\n76\n41\n53.9\n35\n46.1\nUse of illicit substances before imprisonment\nYes\n106\n61\n57.5\n45\n42.5\n0.113\nNo\n25\n10\n40.0\n15\n60.0\nLength of stay in prison (years)\nUp to 1 \n59\n34\n57.6\n25\n42.4\n0.476\nMore than 1 \n72\n37\n51.4\n35\n48.6\nProportion between trans. and men in the cell\n1/1 to 1/10\n20\n11\n55.0\n9\n45.0\n0.938\nOver 1/10\n111\n60\n54.1\n51\n45.9\nSteady partner in prison\nYes\n77\n36\n46.8\n41\n53.2\n0.041\nNo\n54\n35\n64.8\n19\n35.2\nNumber of partners in prison\nUp to 1\n22\n13\n59.1\n9\n40.9\n0.614\nMore than 1\n109\n58\n53.2\n51\n46.8\nRegistered visitor\nYes\n65\n30\n46.2\n35\n53.8\n0.067\nNo\n66\n41\n62.1\n25\n37.9\nEducation in prison\nYes\n26\n15\n57.7\n11\n42.3\n0.690\nNo\n105\n56\n53.3\n49\n46.7\nAccess to male condoms\nYes\n102\n55\n53.9\n47\n46.1\n0.905\nNo \n29\n16\n55.2\n13\n44.8\nAccess to lubricants\nYes\n36\n16\n44.4\n20\n55.6\n0.168\nNo \n95\n55\n57.9\n40\n42.1\nN sample: 131 people. Bold indicates statistically significant p-values. STI: sexually transmitted infections. www.scielo.br/rbepid https://doi.org/10.1590/1980-549720230058 Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058\n8 REFERENCES org/10.22422/2238-1856.2014v14n27p99-117 8. Zamboni M. Travestis e transexuais privadas de liberdade: \na (des) construção de um sujeito de direitos. REA 2016; 2: \n15-23. 9. Brasil. Ministério da Saúde. Secretaria de Gestão Estratégica e \nParticipativa. Departamento de Apoio à Gestão Participativa. Política nacional de saúde integral de lésbicas, gays, \nbissexuais, travestis e transexuais. Brasília: Ministério da \nSaúde; 2013 10. Brasil. Ministério da Justiça e Segurança Pública. Departamento \nPenitenciário Nacional. Coordenação de Atenção às Mulheres \ne Grupos Específicos. Informação nº 95/2022/COAMGE/\nCGCAP/DIRPP/DEPEN. Mapeamento Nacional da População \nLGBTI [Internet]. [cited on Jul 18, 2023]. Available at: https://\nwww.gov.br/senappen/pt-br/servicos/sisdepen/relatorios/\npopulacao-carceraria/presos-lgbti/presos-lgbti-2022.pdf 11. Brasil. Ministério da Mulher, da Família e Direitos Humanos. Secretaria Nacional de Proteção Global. Departamento 11. Brasil. Ministério da Mulher, da Família e Direitos Humanos. Secretaria Nacional de Proteção Global. Departamento www.scielo.br/rbepid https://doi.org/10.1590/1980-549720230058 Healthcare (Basel) 2022; 10(12): 2380. https://doi:10.3390/\nhealthcare10122380 de Promoção dos Direitos de LGBT. LGBT nas prisões \nno Brasil: diagnóstico dos procedimentos institucionais \ne experiências de encarceramento [Internet]. Brasília: \nMinistério da Mulher, da Família e Direitos Humanos; \n2020 [cited on Jul 18, 2023]. Available at: https://www. gov.br/mdh/pt-br/assuntos/noticias/2020-2/fevereiro/\nTratamentopenaldepessoasLGBT.pdf de Promoção dos Direitos de LGBT. LGBT nas prisões \nno Brasil: diagnóstico dos procedimentos institucionais \ne experiências de encarceramento [Internet]. Brasília: \nMinistério da Mulher, da Família e Direitos Humanos; \n2020 [cited on Jul 18, 2023]. Available at: https://www. gov.br/mdh/pt-br/assuntos/noticias/2020-2/fevereiro/\nTratamentopenaldepessoasLGBT.pdf de Promoção dos Direitos de LGBT. LGBT nas prisões \nno Brasil: diagnóstico dos procedimentos institucionais \ne experiências de encarceramento [Internet]. Brasília: \nMinistério da Mulher, da Família e Direitos Humanos; \n2020 [cited on Jul 18, 2023]. Available at: https://www. gov.br/mdh/pt-br/assuntos/noticias/2020-2/fevereiro/\nTratamentopenaldepessoasLGBT.pdf 22. Santos PMR, Santos KC, Magalhães LS, Oliveira BR, Carneiro \nMAS, Souza MM, et al. Travestis and transexual women: \nwho are at higher risk for sexually transmitted infections? Rev Bras Epidemiol 2021; 24: e210017. https://doi. org/10.1590/1980-549720210017 12. Soares Filho MM. Política Nacional de Atenção Integral \nà Saúde das Pessoas Privadas de Liberdade no Sistema \nPrisional (PNAISP): um desafio para o Sistema Único de \nSaúde (SUS) brasileiro. In: Miranda AE, Rangel C, Costa-\nMoura R (orgs). Questões sobre a população prisional no \nBrasil: saúde, justiça e direitos humanos [Internet]. Vitória: \nUFES, Proex; 2016. p. 8-37. 23. Bastos FI, Bastos LS, Coutinho C, Toledo L, Mota JC, \nVelasco-de-Castro CA, et al. HIV, HCV, HBV, and syphilis \namong transgender women from Brazil: assessing \ndifferent methods to adjust infections rates of a hard-\nto-reach, sparse population. Medicine (Baltimore) \n2018; 97(suppl 1): S16-S24. https://doi.org/10.1097/\nmd.0000000000009447 13. Bartos MSH. REFERENCES Política Nacional de Atenção Integral à \nSaúde das Pessoas Privadas de Liberdade no Sistema \nPrisional: uma reflexão sob a ótica da intersetorialidade. Ciênc Saúde Coletiva 2023; 28(4): 1131-8. https://doi. org/10.1590/1413-81232023284.08962022 24. Colchero MA, Cortés-Ortiz MA, Romero-Martínez M, Vega H, \nGonzález A, Román R, et al. HIV prevalence, sociodemographic \ncharacteristics, and sexual behaviors among transwomen \nin Mexico City. Salud Publica Mex 2015; 57 Suppl 2: s99-\n106. https://doi.org/10.21149/spm.v57s2.7596 14. Prefeitura do Rio de Janeiro. Município do Rio de Janeiro \navança na cobertura de saúde no sistema prisional [Internet]. 2023 [cited on Sep 18, 2023]. Available at: https://prefeitura. rio/saude/municipio-do-rio-avanca-na-cobertura-de-saude-\nno-sistema-prisional/ 25. Rocon PC, Wandekoken KD, Barros MEB, Duarte MJO, Sodré \nF. Acesso à saúde pela população trans no Brasil: nas \nentrelinhas da revisão integrativa. Trab Edu Saúde 2020; 18(1): \ne0023469. http://dx.doi.org/10.1590/1981-7746-sol00234 26. Monteiro S, Brigeiro M. Experiências de acesso de mulheres \ntrans/travestis aos serviços de saúde: avanços, limites e \ntensões. Cad Saúde Pública 2019; 35(4): e00111318. https://\ndoi.org/10.1590/0102-311X00111318 15. Sousa KAA, Araújo TME, Teles AS, Rangel EML, Nery IS. Fatores \nassociados à prevalência do vírus da imunodeficiência em \npopulação privada de liberdade. Rev Esc Enferm USP 2017; 51: \ne03274. https://doi.org/10.1590/S1980-220X2016040903274 27. Wirtz AL, Yeh PT, Flath NL, Beyrer C, Dolan K. HIV and viral \nhepatitis among imprisoned key populations. Epidemiol Rev \n2018; 40(1): 12-26. https://doi.org/10.1093/epirev/mxy003 16. Machado F, Becker D, Oliveira CF, Possuelo LG, Renner \nJDP. Soroprevalência de infecção pelo HIV, hepatites B e \nC e sífilis em presidiários da região central do Rio Grande \ndo Sul, Brasil. Mundo Saúde 2019; 43(1): 117-28. htttps://\ndoi.org/10.15343/0104-7809.20194301117128 28. Rodrigues NG, Silva CH, Araújo IS. Visibilidade de pessoas \ntrans na produção científica brasileira. Reciis 2019; 13(3): \n658-70. https://doi.org/10.29397/reciis.v13i3.1723 17. Nasrullah M, Frazier E, Fagan J, Hardnett F, Skarbinski J. The \nassociation of recent incarceration and health outcomes \namong HIV-infected adults receiving care in the United States. Int J Prison Health 2016; 12(3): 135-44. https://10.1108/\nIJPH-04-2016-0010 29. Nascimento FEM. Agrupamento de travestis e transexuais \nencarceradas no Ceará, Brasil. Rev Estud Fem 2020; 28(1): \ne57687. https://doi.org/10.1590/1806-9584-2020v28n157687 30. Sanzovo NM. O lugar das trans na prisão. Belo Horizonte: \nEditora D’Plácido; 2020. 18. Singh GP, Lata S, Swu AK, Virk NS, Singh J, Thakkar S. A \nretrospective study to find prevalence of HIV, HCV and dual \nHIV-HCV infections in the prison inmates. J Family Med Prim \nCare 2022; 11(10): 6250-4. https://doi.org/10.4103/jfmpc. jfmpc_788_22 31. Caravaca-Morera JA, Padilha MI. Necropolítica \ntrans: diálogos sobre dispositivos de poder, morte e \ninvisibilização na contemporaneidade. REFERENCES Texto & \nContexto Enferm 2018; 27(2): e3770017. https://doi. org/10.1590/0104-07072018003770017 19. Brasil. Ministério da Saúde. Guia prático para execução \nde testes rápidos DIAG/DCCI/SVS/MS [Internet]. Brasília: \nMinistério da Saúde; 2022 [cited on Jul 7, 2023]. Available \nat: https://www.gov.br/aids/pt-br/centrais-de-conteudo/\npublicacoes/2022/guia_pratico_execucao_de_testes_\nrapidos-1.pdf 32. Becker A, Spessote DV, Sardinha LS, Matos LGS, Chaves NN, \nBicalho PPG. O cárcere e o abandono: prisão, penalização \ne relações de gênero. Rev Psicol Divers Saúde 2016; 5(2): \n141-54. https://doi.org/10.17267/2317-3394rpds.v5i2.1050 33. Garcia MRV. Prostituição e atividades ilícitas entre travestis \nde baixa renda. Cad Psicol Soc Trab 2008; 11(2): 241-56. 20. Vinuto J. A amostragem em bola de neve em pesquisas \nqualitativas: um debate em aberto. Temáticas 2014; 22(44): \n203-20. https://doi.10.20396/temáticas,v22i44.10977 34. Brasil. Ministério do Trabalho e Emprego. Classificação \nbrasileira de ocupações [Internet]. Brasília: Ministério do \nTrabalho e Emprego; 2010 [cited on Jul 2, 2023]. Available at: \nhttps://portalfat.mte.gov.br/wp-content/uploads/2016/04/\nCBO2002_Liv3.pdf 21. Russoto Y, Micali C, Laganà N, Marino A, Campanella E, \nCelesia BM, et al. Diagnosis, treatment and prevention of \nHIV infections among detainees: a review of the literature. Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058 www.scielo.br/rbepid https://doi.org/10.1590/1980-549720230058 35. Ferreira ACG, Coelho LE, Jalil EM, Luz PM, Friedman RK, \nGuimarães MRC, et al. Transcendendo: a cohort study of \nHIV-infected and uninfected transgender women in Rio de \nJaneiro, Brazil. Transgend Health 2019; 4(1): 107-17. https://\ndoi:10.1089/trgh.2018.0063 37. Teixeira FB, Paulino DB, Raimondi GA, Crovato CAS, Prado \nMAM. Entre o segredo e as possibilidades do cuidado: (re)\npensando os silêncios em torno das narrativas das travestis \nsobre HIV/Aids. Sex, Salud Soc (Rio J) 2018; 29: 373-88. https://doi.org/10.1590/1984-6487.sess.2018.29.17.a 36. Strazza L, Massad E, Azevedo RS, Carvalho HB. Estudo de \ncomportamento associado à infecção pelo HIV e HCV em \ndetentas de um presídio de São Paulo, Brasil. Cad Saúde \nPública 2007; 23(1): 197-205. https://doi.org/10.1590/\nS0102-311X2007000100021 38. Brömdal A, Mullens AB, Phillips TM, Gow J. Experiences \nof transgender prisoners and their knowledge, attitudes, \nand practices regarding sexual behaviors and HIV/STIs: a \nsystematic review. Int J Transgend 2019; 20(1): 4-20. https://\ndoi.org/10.1080/15532739.2018.1538838 RESUMO Objetivo: Avaliar a taxa de soropositividade dos testes rápidos para HIV, sífilis e hepatite B e C entre travestis e mulheres transexuais \n(transfemininas) privadas de liberdade na Região Metropolitana do Rio de Janeiro, analisando os resultados diante do perfil \nsociodemográfico, prisional e acesso às tecnologias de saúde para prevenir infecções sexualmente transmissíveis (IST). REFERENCES Métodos: \nEstudo transversal do tipo censitário realizado com transfemininas em 11 prisões do Rio de Janeiro, entre os meses de abril e \njunho de 2021. Resultados: As taxas de soropositividade encontradas foram de 34,4% para o HIV, 48,9% para sífilis e 0,8% para as \nhepatites do tipo B e C. A soropositividade para mais de uma infecção foi verificada em 25,4% das participantes, e HIV/sífilis foi a mais \nprevalente. O aumento no nível de escolarização (p=0,037), e possuir parceiro fixo na prisão (p=0,041) foram considerados fatores \nde proteção para as IST nessa população. Foram identificadas dificuldades no acesso às tecnologias de prevenção contra IST, como \npreservativo masculino, gel lubrificante, testes rápidos e terapias antirretrovirais profiláticas para o HIV. Conclusão: As taxas de \nsoropositividade para o HIV e sífilis foram elevadas, mas no perfil encontrado nessa população em outros estudos dentro e fora das \nprisões. Os dados encontrados indicam a necessidade de incorporar estratégias efetivas para o acesso às tecnologias em saúde para \na prevenção das IST. A escassez de publicações científicas contendo dados epidemiológicos sobre IST na população transfeminina \nem situação de prisão limitou a realização de comparações mais profundas dos resultados obtidos neste estudo.i alavras-chave: Pessoas transgênero. Prisões. Sífilis. Hepatite. Soropositividade para HIV. ACKNOWLEGMENTS: We are grateful for the support of health professionals from Pnaisp and Seap-RJ, as well as the directors and criminal police officers of the \nprison units visited. We would like to thank the team from Health Management and the School of Penitentiary Management, and in particular, criminal police officer \nLuciana de Souza Resende, assigned to Seap’s Health and Citizenship Support Division for the LGBTI Population, and the transsexual woman activist Alessandra \nRamos Makkeda (in memoriam). AUTHORS’ CONTRIBUTIONS: Alves-da-Silva, CR: Project administration, Data curation, Writing – original draft, Investigation, Methodology. Bonan, C: \nConceptualization, Supervision, Validation. Gomes Junior, SCS: Formal analysis, Funding acquisition, Software. Vieira, RS: Writing – review & editing, Visualization. FUNDING: Research Incentive Program (Programa de Incentivo a Pesquisa – PIP) of Instituto Nacional de Saúde da Mulher, da Criança e do Adolescente Fernandes \nFigueira (IFF-Fiocruz). entive Program (Programa de Incentivo a Pesquisa – PIP) of Instituto Nacional de Saúde da Mulher, da Criança e do Adolescente Fernande Travestis e mulheres transexuais presas e as infecções sexualmente transmissíveis. Rev Bras Epidemiol. 2023; 26: e230058 10\n© 2023 \| Epidemio is a publication of \n Associação Brasileira de Saúde Coletiva - ABRASCO"
https://openalex.org/W2119209728	https://bmcpsychiatry.biomedcentral.com/counter/pdf/10.1186/s12888-015-0406-0	English	null	Emotional disorders among informal caregivers in the general population: target groups for prevention	BMC psychiatry	2,015	cc-by	7,442	RESEARCH ARTICLE Open Access * Correspondence: mtuithof@trimbos.nl Netherlands Institute of Mental Health and Addiction, PO Box 725, 3500 AS Utrecht, the Netherlands © 2015 Tuithof et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Tuithof et al. BMC Psychiatry (2015) 15:23 DOI 10.1186/s12888-015-0406-0 Tuithof et al. BMC Psychiatry (2015) 15:23 DOI 10.1186/s12888-015-0406-0 Emotional disorders among informal caregivers in the general population: target groups for prevention Marlous Tuithof, Margreet ten Have, Saskia van Dorsselaer and Ron de Graaf Background efficient to target prevention measures to those at in- creased risk of emotional disorders [14,24]. Identification of risk indicators facilitates such selection. Ideally, risks indicators should 1) identify a small and feasible target group; 2) be strongly associated with emotional disor- ders among informal caregivers; 3) result in a substantial health gain at population level if their adverse effect is blocked; and 4) require a low number of people who need an intervention to prevent one case of emotional disorder [24]. Most previous research on diminished mental health among informal caregiving focused on the second aspect. Based on these findings, a close relation- ship with the care recipient, [14,25,26] more hours of caregiving [27] and living with the care recipient [12,28] as well as general characteristics such as employment, [28,29] or lack of social support [17,30-32] can be identi- fied as potential risk indicators for emotional disorders among informal caregivers. Moreover, time-consuming activities, such as work or having children at home, in combination with many hours of caregiving may result in particularly high risks of adverse outcomes [29]. In most Western countries, the elderly population is growing and, due to age-related morbidity and disability, health care costs are rising [1-4]. However, as the current economic crisis forces cuts in health care ex- penditure, policy makers are seeking to promote com- munity solutions, such as care and support from family and friends, i.e. informal care [5,6]. At present, approxi- mately one third of the adult general population in Western countries provides informal care [7]. Due to projected demographic developments, these numbers may increase substantially in the future. Previous research has shown that informal caregiving negatively impacts caregivers’ well-being as it affects their daily life and is associated with mental health prob- lems such as depression (for overviews, see [8-11]). However, many studies were performed using conveni- ence samples recruited via patients or service providers. Highly burdened caregivers are likely to be overrepre- sented in such studies. For example, in a study among informal caregivers of dementia patients, the average time spent giving care was almost 90 hours weekly, [12] whereas population-based research gives an average of 20 hours [7]. Another study among caregivers of de- pressed patients observed that 85% of the caregivers lived with the patient, [13] whereas this is only 25% in the general population [14]. Abstract Background: There are indications that informal caregiving negatively impacts caregivers’ mental health, but this was hardly examined using diagnoses of mental disorders and most studies used convenience samples without including non-caregivers as reference group. We examine whether informal caregivers more often have any emotional disorder, i.e. mood or anxiety disorder, than non-caregivers. Identify key risk indicators for any emotional disorder among informal caregivers in the general population. Methods: Data were used from the second wave of the Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2), a nationally representative face-to-face survey (n = 5,303; aged 21–68). Respondents were defined as informal caregiver when they provided unpaid care in the 12 months preceding the second wave to a family member, partner or friend who needed care because of physical or mental problems, or ageing. Twelve-month DSM-IV diagnoses of emotional disorders were assessed using the Composite International Diagnostic Interview 3.0. Key risk indicators were identified using the following aspects: prevalence, odds ratio, attributable risk proportion, and number needed to treat. Sociodemographic, caregiving-related and other characteristics were considered as risk indicators. Results: In the past year, 31.1% of the respondents provided informal care, which ranged in time spent (8 or more hours/week: 32.1%) and duration (longer than 1 year: 48.7%). Informal caregiving was not associated with having any 12-month emotional disorder. Among caregivers, giving care to a first-degree relative, partner or close friend and giving emotional support increased the risk for any emotional disorder. Moreover, using all aspects, target groups were identified for prevention: caregivers without a job, living without a partner, and with a lack of social support. Conclusions: Although informal caregivers do not have an increased risk of emotional disorders, key risk indicators were identified using four aspects. Especially informal caregivers with limited resources (unemployment, living without a partner, lack of social support) may benefit from targeted prevention whereas general prevention measures may be desirable for carers with a burdensome care situation (giving care to a close loved one or providing emotional support). Keywords: Informal care, Mood and anxiety disorders, Risk indicators, General population study Tuithof et al. BMC Psychiatry (2015) 15:23 Page 2 of 8 Background Thus, findings based on convenience samples may overrepresent the negative consequences of caregiving [15,16]. Moreover, many studies did not include a control group (for overview, see [16]) and thus lacked a comparison with non- caregivers. Finally, whereas previous research observed that informal caregiving was associated with higher scores on screeners measuring depressed or anxious mood, [8,17-20] it is uncertain whether informal caregiv- ing is similarly associated with diagnoses of mental dis- orders. To our knowledge, only one community-based study, conducted in 1990 in Ontario, examined the rela- tionship between informal caregiving and mental disor- ders. In this study, about 15% of the adult population aged 15–64 provided informal care and a slightly in- creased risk of mental disorders was observed among these caregivers [21]. Using data from the second wave of the Netherlands Mental Health Survey and Incidence Study-2 (NEM- ESIS-2), a psychiatric epidemiological study among the Dutch general population aged 21 to 68 years at that wave, we examined i) whether informal caregiving is as- sociated with presence of any emotional disorder in the past year; and ii) which characteristics (i.e. sociodemo- graphic, caregiving-related and other characteristics) are risk indicators for any emotional disorder among infor- mal caregivers. Methods NEMESIS-2 is a psychiatric epidemiological cohort study of the Dutch general population. It is based on a multistage, stratified random sampling of households. Based on the most recent birthday at first contact within the household, an individual aged 18–64 years with suffi- cient fluency in the Dutch language was randomly se- lected. The study was approved by the Medical Ethics Review Committee for Institutions on Mental Health Care (METIGG). After having been informed about the study aims, respondents provided written informed con- sent. A comprehensive description of the design is pro- vided in [33]. Informal caregiving was assessed during the second wave and therefore data from this wave were used for the present study. An emotional disorder, i.e. mood or anxiety disorder, may result in substantial consequences in terms of the quality of life of the caregiver [18] and quality of care is likely to be affected. Application of existing measures to prevent emotional disorders among informal caregivers (e.g. psycho-educational interventions or counseling [22]) is therefore important and general practitioners have been targeted as a critical first point of contact to identify the impact of caregiving [23]. However, as the group of informal caregivers is substantial, [7] with only a minority having an emotional disorder, [21] it may be In the first wave (T0), performed from November 2007 to July 2009, a total of 6,646 persons aged 18–64 were interviewed (response rate 65.1%; mean duration of 95 minutes). This sample was nationally representative, although younger subjects were somewhat underrepre- sented [33]. All T0 respondents were approached for follow-up (T1), three years after T0 from November Tuithof et al. BMC Psychiatry (2015) 15:23 Page 3 of 8 2010 to June 2012. Of this group, 5,303 persons were interviewed again (response rate 80.4%, with those de- ceased excluded). Attrition was not significantly related to any 12-month disorder, any anxiety, any mood, or any substance use disorder, nor to the separate disorders, after controlling for sociodemographics [34]. The mean duration of the second interview was 84 minutes. The mean period between both interviews was 3 years and 7 days (1,102 days; standard deviation = 64; ranging from 2 to 4 years). comparison to blinded clinical reappraisal interviews. For the present study, mood and anxiety disorder were com- bined into presence or absence of any emotional disorder. Emotional disorders DSM-IV diagnoses [35] were made using the Composite International Diagnostic Interview (CIDI) 3.0, a fully structured lay-administered diagnostic interview. This instrument was developed and adapted for use in the World Mental Health Survey Initiative [36] The CIDI 3.0 version used in NEMESIS-2 was an improvement of the Dutch one used in this initiative. The disorders con- sidered in this paper include mood (major depression; dysthymia; bipolar disorder) and anxiety disorders (panic disorder; agoraphobia without panic disorder; social pho- bia; generalized anxiety disorder) in the past 12 months, assessed at the second wave. Clinical calibration studies in various countries [37] found that the CIDI 3.0 assesses mood and anxiety disorders with generally good validity in Sociodemographics Age, gender, educational level (primary, basic vocational or lower secondary education vs higher education), employment status, partner status and having children living at home (vs having children not living at home and having no children) were considered. Age, gender, and educational level were measured at T0, other socio- demographics were measured at T1. The face-to-face interviews were mainly held at the re- spondent’s home and were conducted by trained profes- sional interviewers (65 at the second wave) of the fieldwork agency GfK (Growth from Knowledge) Panel Services Benelux, with their team of five supervisors. In- terviewers were selected on their experience with sys- tematic face-to-face data collection, experience with sensitive topics and ability to achieve a good response in other studies. Fieldwork was monitored over the entire data collection period by the NEMESIS-investigators and the fieldwork agency (for more information on qual- ity checks of the data, see [33]). Other characteristics l f Social support from three resources (partner; family or friends; neighbors) was measured with two questions regarding the extent respondents could rely on the resource for help and could open up to them (four response categories ranging from “not at all” to “a lot”). The mean score on the two questions was used to indi- cate social support perceived from a particular resource, taking the respondents’ evaluation of it into account. The total social support measure (ranging from 1 to 4) was calculated as the mean score on the perceived social support from at least two resources, because not all re- spondents had a partner [38]. A dichotomous variable was constructed using a cut-off point of 2.5 (the middle of the scale), thereby separating the lowest twelve per- cent, a group with quite severe lack of social support, versus the rest. Sensitivity analyses with cut-off points of 2 (4%) and 3 (31%) resulted in similar findings. Informal caregiving and caregiving-related characteristics Respondents were defined as informal caregiver when they provided unpaid care in the 12 months preceding the second wave to a family member, partner or friend who needed care because of physical or mental prob- lems, or ageing (n = 1,759). Follow-up questions mea- sured to whom they gave care (first-degree family such as parents, siblings or children; other family such as in- laws or other; partner; friend), whether they lived with the care recipient, reasons for care (physical problems; mental problems; ageing; dementia), kind of care (domestic work; personal care or nursing; emotional support or super- vision; practical care), time spent giving care (0–7 hours weekly; 8 or more hours weekly) and duration of care (less than 1 year; 1 year or longer). A standard checklist assessed presence of 17 chronic physical disorders (asthma; chronic obstructive pulmon- ary disease; chronic bronchitis; emphysema; severe heart disease; heart attack; hypertension; stroke; stomach or intestinal ulcers; severe intestinal disorders; diabetes; thyroid disorder; chronic back pain; arthritis; migraine; impaired vision or hearing; and another chronic physical disorder), treated or monitored by a medical doctor in the prior 12-months [39]. Analyses l Analyses were performed using Stata version 12.1, [40] which enabled to control for the complex sampling and recruitment procedure of the study. The data were weighted to correct for differences in the response rates in several sociodemographic groups at both waves and differences in the probability of selection of respondents within households at baseline. Using the complete sample (n = 5,303), characteristics of informal caregiving were examined using frequencies and logistic regression analyses adjusted for age and gen- der. The association between informal caregiving and presence of 12-month emotional disorder was examined Tuithof et al. BMC Psychiatry (2015) 15:23 Page 4 of 8 Page 4 of 8 Table 1 Characteristics of informal caregivers Non-caregiver (n = 3,544; 68.9%) Informal caregiver (n = 1,759; 31.1%) %a %a ORb 95% CIb Sociodemographics Female gender 45.0 59.6 1.84 1.59-2.12 Younger than 45 54.7 36.2 0.46* 0.41-0.53 Low educational level 30.2 28.0 0.81* 0.69-0.95 Unemployed 22.7 29.5 1.06 0.89-1.26 Living without a partner 31.8 25.9 0.83 0.69-1.00 Children at home 45.7 44.0 0.99 0.87-1.13 Other characteristics Lack of social support 11.8 11.2 0.87 0.69-1.09 Chronic physical disorder 37.1 45.5 1.09 0.93-1.28 p < 0.05; p < 0.001; OR: Odds Ratio; CI: Confidence Intervals. aColumn percentages. bLogistic regression models, adjusted for age and gender. Table 1 Characteristics of informal caregivers with logistic regression analyses adjusted for all sociode- mographics. Sensitivity analyses were performed with stricter definitions of informal care; i.e. eight or more hours weekly, longer than one year, or both. Further analyses were conducted among the informal caregivers (n = 1,759). That is, potential risk indicators of emotional disorders among informal caregivers were first separately investigated with bivariate logistic regression analyses adjusted for age and gender, and then simultan- eously with conventional backstepping procedures to ob- tain a parsimonious set of risk indicators. No collinearity of risk indicators was observed in the final model. Lastly, the Population Attributable Risk Proportion (PARP) and the Number Needed to Treat (NNT) of the final set of significant risk indicators were calculated in univariable analyses to select useful indicators for targeted preven- tion [24]. The PARP represents the proportion of emo- tional disorders in the population of informal caregivers that can be reduced when the adverse effect of the risk indicator is completely eliminated [41]. Characteristics of informal caregivers About one third (31.1%) of the respondents provided unpaid care to a family member or friend in the 12 months preceding the interview. Informal caregivers were more often female, 45 or older and had a higher educational level than non-caregivers (Table 1). Notably, informal caregivers did not differ from non-caregivers with regard to employment status, partner status, having children at home, lack of social support or having a chronic physical disorder. Results for the two types of regression models, i.e. bi- variate analyses adjusted for age and gender and the final backstep model, are shown in Table 2. Notably, findings are largely similar for both types of models. Therefore, hereafter we will describe the results of the most parsi- monious model, the final backstep model. Regarding sociodemographics, emotional disorder among informal caregivers was associated with female gender, being younger than 45, unemployment and living without a Analyses l Although this as- sumption is not realistic as preventive programs are often unable to completely block the adverse effect of the risk indicator and some risk indicators are unmodifi- able (i.e. age, gender), the PARP can be useful as it gives an indication of the highest achievable health gain. The PARP was examined using the punaf procedure of Stata [42]. The NNT indicates how many informal caregivers have to receive an intervention to avoid one case of emotional disorder, assuming that the adverse effect of the risk indicator is completely blocked by the interven- tion [24,43]. The NNT can best be interpreted as a mini- mum of the required effort, as interventions are rarely completely effective. This was calculated as the inverse of the risk difference between informal caregivers with and without the risk indicator, which was estimated with generalized linear models with a binomial distribution and an identity link function. 8.8%; p = 0.15). Logistic regression analysis adjusted for all sociodemographics (not in table) showed no associ- ation between informal caregiving and having an emo- tional disorder (OR = 0.86; 95% CI = 0.66-1.11). When informal caregiving was defined more strictly, as giving care eight or more hours weekly, longer than one year, or both, no association with an emotional disorder was observed (OR = 0.92; 95% CI = 0.64-1.31, OR = 1.22; 95% CI = 0.92-1.63 and OR = 1.17; 95% CI = 0.73-1.87, respectively). 8.8%; p = 0.15). Logistic regression analysis adjusted for all sociodemographics (not in table) showed no associ- ation between informal caregiving and having an emo- tional disorder (OR = 0.86; 95% CI = 0.66-1.11). When informal caregiving was defined more strictly, as giving care eight or more hours weekly, longer than one year, or both, no association with an emotional disorder was observed (OR = 0.92; 95% CI = 0.64-1.31, OR = 1.22; 95% CI = 0.92-1.63 and OR = 1.17; 95% CI = 0.73-1.87, respectively). Risk indicators for 12-month emotional disorder among informal caregivers More than half of the respondents provided care to a first-degree family member (first column of Table 2), al- most one third to other family members, 12% to their partner and 16% to a friend. Approximately 40% of care- givers provided care to more than one person and 17% lived with the care recipient. Physical problems were the most common reason for giving care (61%). Almost all informal caregivers provided emotional support (89.5%) or practical care (77.3%). About one third provided eight or more hours of care weekly, and about half of care- givers provided care for one year or longer. The association between informal caregiving and having a 12-month emotional disorder The prevalence of any emotional disorder was similar among informal caregivers and non-caregivers (7.5% vs Tuithof et al. BMC Psychiatry (2015) 15:23 Page 5 of 8 Page 5 of 8 Table 2 Risk indicators of 12-month emotional disorder among informal caregivers (n = 1,759) Model 1a Model 2b % OR 95% CI OR 95% CI PARP NNT Sociodemographics Female gender 59.6 2.05 1.20-3.50 1.68* 1.00-2.82 37.3 21.2 Younger than 45 36.2 2.00 1.24-3.24 2.53 1.44-4.43 25.3 19.0 Low educational level 28.0 1.14 0.68-1.89 Unemployed 29.5 2.67* 1.56-4.55 2.23 1.29-3.87 25.7 15.3 Living without a partner 25.9 2.34* 1.51-3.65 1.87 1.20-2.91 26.6 13.0 Children at home 44.0 0.72 0.47-1.11 Caregiving characteristics Care recipient First-degree family 55.3 1.33 0.88-2.01 2.04 1.28-3.25 12.7 57.9 Other family 30.0 0.35* 0.20-0.61 Partner 11.8 1.54 0.82-2.88 2.17* 1.10-4.28 3.2 49.1 Friend 16.4 1.92 1.21-3.07 3.46* 1.73-6.92 13.0 16.9 More than one care recipient 39.0 0.81 0.54-1.20 0.53* 0.30-0.96 - - Living with care recipient 16.6 1.40 0.85-2.30 Reason for care Physical problems 61.2 0.71 0.48-1.06 Mental problems 29.9 1.87 1.26-2.79 Ageing 30.7 0.63 0.37-1.07 Dementia 12.2 1.44 0.85-2.45 Kind of care Domestic work 59.1 1.19 0.77-1.82 Personal care or nursing 27.0 1.39 0.89-2.16 Emotional support 89.5 6.83* 2.49-18.76 6.16* 2.18-17.39 83.0 14.4 Practical care 77.3 1.45 0.82-2.57 Eight or more hours per week 32.1 1.19 0.78-1.79 Longer than one year 48.7 1.88 1.20-2.93 1.92 1.19-3.12 22.6 28.6 Other characteristics Lack of social support 11.2 4.55* 2.80-7.40 3.22*** 1.97-5.25 23.7 6.3 Chronic physical disorder 45.5 1.52 0.97-2.38 p < 0.05; p < 0.01; **p < 0.001; OR: Odds Ratio; CI: Confidence Intervals. PARP: Population Attributable Risk Proportion; NNT: Number Needed to Treat. aLogistic regression models, adjusted for age and gender. bFinal backward stepwise multivariable logistic regression model. Table 2 Risk indicators of 12-month emotional disorder among informal caregivers (n = 1,759) tors of 12-month emotional disorder among informal caregivers (n = 1,759) partner, but not with educational level and having chil- dren at home. Of the caregiving characteristics, giving care to a direct family member, partner or close friend, providing emotional support, and giving care for one year or longer were associated with an emotional dis- order. Unexpectedly, providing care to more than one care recipient was associated with a lower chance of an emotional disorder. The association between informal caregiving and having a 12-month emotional disorder Living with the care recipient, rea- son for care, and giving eight or more hours of care weekly were not associated with an emotional disorder. Moreover, neither employment status nor having chil- dren at home strengthened the relationship between hours of care provided and having an emotional dis- order, that is, no additive interaction-effects were ob- served (not in table) [44]. Finally, caregivers who perceived a lack of social support had higher chances of experiencing an emotional disorder. No association be- tween chronic physical disorders and emotional disorder was observed. In the final two columns of Table 2, the PARP and the NNT are displayed for the risk indicators in the final backstep model. Emotional support had the highest PARP (83.0%) and an acceptable NNT (14.4), but the high prevalence rate (89.5%) limits its use for targeted Tuithof et al. BMC Psychiatry (2015) 15:23 Page 6 of 8 educational level than non-caregivers. Caregivers and non-caregivers did not differ in employment status, part- ner status, having children at home, social support and presence of a chronic physical disorder. prevention. Another notable risk indicator is lack of so- cial support, with a substantial PARP (23.7%), the lowest NNT (6.3) and a fairly low prevalence rate (11.2%). Fi- nally, being unemployed and living without a partner had considerable PARPs (25.7% and 26.6%, respectively), relatively low NNTs (15.3 and 13.0) and manageable prevalence rates (29.5% and 25.9%). Providing informal care was not significantly associ- ated with having an emotional disorder and even a lower prevalence of emotional disorder was found among care- givers compared to non-caregivers. In this regard, our findings are at odds with findings from a similar population-based study performed in Canada in 1990, in which informal caregivers were found to have a slightly higher chance of mood and anxiety disorders than non- caregivers. Although that study used a similar age range and the same definition of informal care, [21] the per- centage of informal caregivers in the population was much smaller (15%) than in our study (31%). As that study used the same diagnostic instrument to examine mental disorders (though an earlier version), the discrep- ancy is likely due to societal changes over time (giving informal care might be more common nowadays) or cross-national differences. The association between informal caregiving and having a 12-month emotional disorder The fact that we did not ob- serve an association between informal caregiving and emotional disorder, even when stricter definitions of in- formal care were applied (eight or more hours weekly, longer than one year, or both), might be explained by a continuing selection process, such that those who provide informal care tend to be healthier than those who either do not start or quit caregiving. A similar phenomenon is seen in the workforce, with a lower prevalence of mental disorders among workers than those outside the work- force, [39] dubbed the ‘healthy worker effect’ [45]. Such a phenomenon thwarts examination of the relationship between informal caregiving and emotional disorder in cross-sectional research. Discussion Key findings l h In line with previous cross-national research, [7] we ob- served that informal caregiving was quite common (31.3%) in the general population. Although no associ- ation between informal caregiving and emotional disor- ders was observed, risk indicators were identified. Specifically, informal caregivers with limited resources (lack of social support, unemployment, living without a partner) and those with a burdensome care situation (giving care to a close loved one or providing emotional support) were at higher risk of an emotional disorder. Strengths and limitations The use of a large population-based sample which in- cluded different types of informal caregivers, not limited to heavily burdened caregivers, as well as a reference group of non-caregivers, and the use of diagnoses mea- sured with a valid instrument (CIDI 3.0), are important strengths of our study. To our knowledge, this is the first study to examine risk indicators for emotional disorders in a representative sample of informal caregivers and to identify indicators useful for prevention activities. A few cautionary remarks should be made in relation to our findings. Although the NEMESIS-2 sample was representative of the Dutch population on most parame- ters, people with an insufficient mastery of the Dutch language were underrepresented. Moreover, although NEMESIS-2 had a broad age range, the maximum age of the sample at the second wave was 68, thereby excluding older caregivers. Hence, our findings are not generalizable to older age groups. Furthermore, non-response may have been higher among heavily burdened caregivers with too little time to participate. The extent to which this occurred is unclear. Another limitation is that only the second wave of NEMESIS-2 could be used and our findings were there- fore based on cross-sectional data only. One should thus be careful in interpreting these findings as no causal rela- tionships were examined and reverse-causality may also play a role. That is, an emotional disorder may contribute to presence of the risk indicators. Despite major methodological differences, findings re- garding risk indicators of emotional disorders among in- formal caregivers are largely in line with previous research primarily based on convenience samples [15,16]. Overall, emotional disorders among informal caregivers seem to be related to having limited resources: lack of social sup- port was a strong indicator for emotional disorders among informal caregivers, [17,30-32] as well as unemployment and living alone [46]. Other significant sociodemographic correlates for an emotional disorder were being female and younger age, but not educational level [12,24,46]. Although most of these risk indicators also exist in the general population, [47] they help to identify vulnerable informal caregivers. Among informal caregivers, duration of care was asso- ciated with an emotional disorder, but not time spent giving care. The latter finding was in contrast with previous research [14] and might be explained by positive aspects associated with caregiving, such as increased closeness with the care recipient and personal satisfaction [46,48,49] Authors’ contributions All th t f th All authors are part of the NEMESIS-2 research team. MtH and RdG obtained funding for the NEMESIS-2 study and for studying this particular research topic. All authors (MT, MtH, SvD, RdG) contributed to the conception, design and interpretation of analysis for this manuscript. MT undertook the analysis of this manuscript, with assistance of SvD, and wrote the first draft of the manuscript. All authors discussed the results and implications and commented on the manuscript at all stages. All authors contributed extensively to and have approved the final manuscript. Some other caregiving-related indicators are worth noting. Specifically, giving care to a close loved one, i.e. a first-degree relative, partner or close friend, put infor- mal caregivers at higher risk for an emotional disorder [14]. This might be because in such cases the illness of the care recipient is also burdensome. A previous study observed that giving care to a parent was associated with poorer mental health, but giving care to an in-law was not [26]. It was suggested that this difference is due to type of care provided, with less emotional support and more practical care [26]. Similarly, we observed a higher risk of emotional disorder when the caregiver provided emotional support. References References 1. Carter R. Addressing the caregiving crisis. Prev Chronic Dis. 2008;5:A02. 2. Christensen K, Doblhammer G, Rau R, Vaupel JW. Ageing populations: the challenges ahead. Lancet. 2009;374:1196–208. 3. Lee R. The outlook for population growth. Science. 2011;333:569–73. 4. Alemayehu B, Warner KE. The lifetime distribution of health care costs. Health Serv Res. 2004;39:627–42. 5. Levine C, Halper D, Peist A, Gould DA. Bridging troubled waters: family caregivers, transitions, and long-term care. Health Aff. 2010;29:116–24. 6. Bolin K, Lindgren B, Lundborg P. Your next of kin or your own career? Caring and working among the 50+ of Europe. J Health Econ. 2008;27:718–38. 7. Shahly V, Chatterji S, Gruber MJ, Al-Hamzawi A, Alonso J, Andrade LH, et al. Cross-national differences in the prevalence and correlates of burden among older family caregivers in the World Health Organization World Mental Health (WMH) Surveys. Psychol Med. 2013;43:865–79. 8. Pinquart M, Sorensen S. Differences between caregivers and noncaregivers in psychological health and physical health: a meta-analysis. Psychol Aging. 2003;18:250–67. 9. Cuijpers P. Depressive disorders in caregivers of dementia patients: a systematic review. Aging Ment Health. 2005;9:325–30. 10. Seeher K, Low LF, Reppermund S, Brodaty H. Predictors and outcomes for caregivers of people with mild cognitive impairment: a systematic literature review. Alzheimers Dement. 2013;9:346–55. 1. Carter R. Addressing the caregiving crisis. Prev Chronic Dis. 2008;5:A02. 1. Carter R. Addressing the caregiving crisis. Prev Chronic Dis. 2008;5:A02. 2. Christensen K, Doblhammer G, Rau R, Vaupel JW. Ageing populations: the challenges ahead. Lancet. 2009;374:1196–208. 3. Lee R. The outlook for population growth. Science. 2011;333:569–73. 4. Alemayehu B, Warner KE. The lifetime distribution of health care costs. Health Serv Res. 2004;39:627–42. 5. Levine C, Halper D, Peist A, Gould DA. Bridging troubled waters: family caregivers, transitions, and long-term care. Health Aff. 2010;29:116–24. 6. Bolin K, Lindgren B, Lundborg P. Your next of kin or your own career? Caring and working among the 50+ of Europe. J Health Econ. 2008;27:718–38. 7. Shahly V, Chatterji S, Gruber MJ, Al-Hamzawi A, Alonso J, Andrade LH, et al. Cross-national differences in the prevalence and correlates of burden among older family caregivers in the World Health Organization World Mental Health (WMH) Surveys. Psychol Med. 2013;43:865–79. 8. Pinquart M, Sorensen S. Differences between caregivers and noncaregivers in psychological health and physical health: a meta-analysis. Psychol Aging. 2003;18:250–67. 9. Cuijpers P. Depressive disorders in caregivers of dementia patients: a systematic review. Aging Ment Health. 2005;9:325–30. Acknowledgements The Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2) is conducted by the Netherlands Institute of Mental Health and Addiction (Trimbos Institute) in Utrecht. Financial support has been received from the Ministry of Health, Welfare and Sport, with supplement support from the Netherlands Organization for Health Research and Development (ZonMw) and the Genetic Risk and Outcome of Psychosis (GROUP) investigators. The funding sources had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. Our findings suggest that tailoring prevention measures to people with limited resources (lack of social support, unemployment, living without a partner) can result in health gain (PARP) with relatively low effort (prevalence and NNT). Low-level interventions, such as e-health in- terventions [50] at the first signs of distress, may suffice to prevent an escalation of problems. As informal caregivers may join the care recipient when visiting the general prac- titioner (GP) or consult the GP when first signs of distress appear, GPs could play a key role in identifying caregivers at risk of emotional disorders. They should be particularly attentive when caregivers have limited resources. Close and more intense care situations (giving care to a close loved one or giving emotional support) may also be worthy of their attention. However, as such care situations are common, a more general approach may be appropri- ate, for example providing extra coping tools on websites targeted to informal caregivers. Competing interests The a thors declare tha Competing interests The authors declare that they have no competing interests. Received: 20 September 2014 Accepted: 3 February 2015 Received: 20 September 2014 Accepted: 3 February 2015 Received: 20 September 2014 Accepted: 3 February 2015 Findings f l Al- though a meta-analysis observed a poorer outcome for predominantly spousal caregivers of dementia patients compared with other diseases, [8] we did not observe such a relationship. Although power limitations cannot be ex- cluded, as only 12% of our sample provided care to a de- mentia patient, there is a more likely explanation: our sample was relatively young and the majority of caregivers of dementia patients provided care to a parent, rather than to a partner. As the relationship with the care recipient in combination with a specific illness may affect the outcome for caregivers, [25] sampling differences (predominantly spousal caregivers vs all types of caregivers) may explain the difference in findings. Findings f l Informal caregiving was quite common in the Dutch general population (31.3%). Informal caregivers were more often female, 45 years or older, and had a higher Page 7 of 8 Tuithof et al. BMC Psychiatry (2015) 15:23 Tuithof et al. BMC Psychiatry (2015) 15:23 population, but varies significantly in terms of time spent, duration, and type, and is not necessarily associ- ated with a higher chance of emotional disorders. Al- though this seems to contradict the widely accepted notion that informal caring negatively affects well-being, [51,52] it may also show that findings for convenience samples are not applicable to informal caregiving in the general population. Moreover, informal caregivers who continue giving care may be healthier than those who do not start or quit caregiving. The risk indicators identified here, such as limited resources, giving care to a close loved one, and giving emotional support, could help to target prevention measures to informal caregivers at risk for emotional disorders. Such aspects can positively impact mental health, pro- vided that caregiving activities are not too burdensome [46] Our findings could indicate that the positive effect of such rewards offset the burden of much time spent giving care, but does not overcome the burden of a long duration of care. Future research should examine this point. Al- though a meta-analysis observed a poorer outcome for predominantly spousal caregivers of dementia patients compared with other diseases, [8] we did not observe such a relationship. Although power limitations cannot be ex- cluded, as only 12% of our sample provided care to a de- mentia patient, there is a more likely explanation: our sample was relatively young and the majority of caregivers of dementia patients provided care to a parent, rather than to a partner. As the relationship with the care recipient in combination with a specific illness may affect the outcome for caregivers, [25] sampling differences (predominantly spousal caregivers vs all types of caregivers) may explain the difference in findings. Such aspects can positively impact mental health, pro- vided that caregiving activities are not too burdensome [46] Our findings could indicate that the positive effect of such rewards offset the burden of much time spent giving care, but does not overcome the burden of a long duration of care. Future research should examine this point. 9. Cuijpers P. Depressive disorders in caregivers of dementia patients: a systematic review. Aging Ment Health. 2005;9:325–30. 7. Shahly V, Chatterji S, Gruber MJ, Al-Hamzawi A, Alonso J, Andrade LH, et al. Cross-national differences in the prevalence and correlates of burden among older family caregivers in the World Health Organization World Mental Health (WMH) Surveys. Psychol Med. 2013;43:865–79. 8. Pinquart M, Sorensen S. Differences between caregivers and noncaregivers in psychological health and physical health: a meta-analysis. Psychol Aging. 2003;18:250–67. Conclusions 10. Seeher K, Low LF, Reppermund S, Brodaty H. Predictors and outcomes for caregivers of people with mild cognitive impairment: a systematic literature review. Alzheimers Dement. 2013;9:346–55. Findings from this population-based study show that informal care is frequently provided in the general Page 8 of 8 Page 8 of 8 Tuithof et al. BMC Psychiatry (2015) 15:23 35. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-IV). Washington, DC: American Psychiatric Association; 1994. 11. Awad AG, Voruganti LN. The burden of schizophrenia on caregivers: a review. Pharmacoeconomics. 2008;26:149–62. 12. Covinsky KE, Newcomer R, Fox P, Wood J, Sands L, Dane K, et al. Patient and caregiver characteristics associated with depression in caregivers of patients with dementia. J Gen Intern Med. 2003;18:1006–14. 36. Kessler RC, Üstün TB. The WHO World Mental Health Surveys. Global Perspectives on the Epidemiology of Mental Disorders. New York: Cambridge University Press; 2008. patients with dementia. J Gen Intern Med. 2003;18:1006–14. 13. Van Wijngaarden B, Schene AH, Koeter MW. Family caregiving in depression: impact on caregivers’ daily life, distress, and help seeking. J Affect Disord. 2004;81:211–22. 37. Haro JM, Arbabzadeh-Bouchez S, Brugha TS, De Girolamo G, Guyer ME, Jin R, et al. Concordance of the Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO World Mental Health surveys. Int J Methods Psychiatr Res. 2006;15:167–80. 14. Hirst M. Carer distress: a prospective, population-based study. Soc Sci Med. 2005;61:697–708. 38. Ten Have M, De Graaf R, Van Weeghel J, Van Dorsselaer S. The association between common mental disorders and violence: to what extent is it influenced by prior victimization, negative life events and low levels of social support? Psychol Med. 2014;14:1485–98. 15. Schulz R, Newsom J, Mittelmark M, Burton L, Hirsch C, Jackson S. Health effects of caregiving: the caregiver health effects study: an ancillary study of the Cardiovascular Health Study. Ann Behav Med. 1997;19:110–6. 16. Pinquart M, Sorensen S. Associations of stressors and uplifts of caregiving with caregiver burden and depressive mood: a meta-analysis. J Gerontol B Psychol Sci Soc Sci. 2003;58:112–28. 39. De Graaf R, Tuithof M, Van Dorsselaer S, Ten Have M. Comparing the effects on work performance of mental and physical disorders. Soc Psychiatry Psychiatr Epidemiol. 2012;47:1873–83. 17. Butterworth P, Pymont C, Rodgers B, Windsor TD, Anstey KJ. Factors that explain the poorer mental health of caregivers: Results from a community survey of older Australians. Conclusions Aust N Z J Psychiatry. 2010;44:616–24. y 40. Statacorp. Statacorp. Stata Statistical Software: Release 12. College Station, Texas: Statacorp LP; 2011. 41. Miettinen O. Proportion of disease caused or prevented by a given exposure, trait, or intervention. Am J Psychiatry. 1974;99:325–32. 18. Ho SC, Chan A, Woo J, Chong P, Sham A. Impact of caregiving on health and quality of life: a comparative population-based study of caregivers for elderly persons and noncaregivers. J Gerontol A Biol Sci Med Sci. 2009;64:873–9. 42. Newson R. PUNAF: Stata module to compute population attributable fractions for cohort studies. Boston, USA: Boston College Department of Economics, Statistical Software Components S457193; 2010. 24-12-2010. 19. Cooper C, Katona C, Orrell M, Livingston G. Coping strategies, anxiety and depression in caregivers of people with Alzheimer’s disease. Int J Geriatr Psychiatry. 2008;23:929–36. 43. Cook RJ, Sackett DL. The number needed to treat: a clinically useful measure of treatment effect. BMJ. 1995;310:452–4. 44. Rothman KJ. Epidemiology. An Introduction. New York: Oxford University Press; 2002. 20. Phillips AC, Gallagher S, Hunt K, Der G, Carroll D. Symptoms of depression in non-routine caregivers: the role of caregiver strain and burden. Br J Clin Psychol. 2009;48:335–46. 45. Baillargeon J. Characteristics of the healthy worker effect. Occup Med. 2001;16:359–66. 21. Cochrane JJ, Goering PN, Rogers JM. The mental health of informal caregivers in Ontario: an epidemiological survey. Am J Public Health. 1997;87:2002–7. 46. Buyck JF, Bonnaud S, Boumendil A, Andrieu S, Bonenfant S, Goldberg M, et al. Informal caregiving and self-reported mental and physical health: results from the Gazel Cohort Study. Am J Public Health. 2011;101:1971–9. 22. Pinquart M, Sorensen S. Helping caregivers of persons with dementia: which interventions work and how large are their effects? Int Psychogeriatr. 2006;18:577–95. 47. de Graaf R, Ten Have M, van Gool C, van Dorsselaer S. Prevalence of mental disorders and trends from 1996 to 2009. Results from the Netherlands Mental Health Survey and Incidence Study-2. Soc Psychiatry Psychiatr Epidemiol. 2012;47:203–13. 23. Cameron ID, Aggar C, Robinson AL, Kurrle SE. Assessing and helping carers of older people. BMJ. 2011;343:d5202. 48. Kramer BJ. Gain in the caregiving experience: where are we? What next? Gerontologist. 1997;37:218–32. 24. Joling KJ, Smit F, van Marwijk HW, van der Horst HE, Scheltens P, Schulz R, et al. Identifying target groups for the prevention of depression among caregivers of dementia patients. Int Psychogeriatr. 2012;24:298–306. 49. Ratcliffe J, Lester LH, Couzner L, Crotty M. Conclusions An assessment of the relationship between informal caring and quality of life in older community-dwelling adults - more positives than negatives? Health Soc Care Community. 2013;21:35–46. 25. Hammond T, Weinberg MK, Cummins RA. The dyadic interaction of relationships and disability type on informal carer subjective well-being. Qual Life Res. 2013;23:1535–42. 50. van Straten A, Cuijpers P, Smits N. Effectiveness of a web-based self-help intervention for symptoms of depression, anxiety, and stress: randomized controlled trial. J Med Internet Res. 2008;10:e7. 26. Strauss JR. Caregiving for parents and in-laws: commonalities and differ- ences. J Gerontol Soc Work. 2013;56:49–66. 27. Yates ME, Tennstedt S, Chang BH. Contributors to and mediators of psychological well-being for informal caregivers. J Gerontol B Psychol Sci Soc Sci. 1999;54:12–22. 51. Schulz R, Beach SR. Caregiving as a risk factor for mortality: the Caregiver Health Effects Study. JAMA. 1999;282:2215–9. 51. Schulz R, Beach SR. Caregiving as a risk factor for mortality: the Caregiver Health Effects Study. JAMA. 1999;282:2215–9. 52. Navaie-Waliser M, Spriggs A, Feldman PH. Informal caregiving: differential experiences by gender. Med Care. 2002;40:1249–59. 52. Navaie-Waliser M, Spriggs A, Feldman PH. Informal caregiving: differential experiences by gender. Med Care. 2002;40:1249–59. 28. Robison J, Fortinsky R, Kleppinger A, Shugrue N, Porter M. A broader view o family caregiving: effects of caregiving and caregiver conditions on depressive symptoms, health, work, and social isolation. J Gerontol B Psychol Sci Soc Sci. 2009;64:788–98. 29. Stephens MA, Townsend AL, Martire LM, Druley JA. Balancing parent care with other roles: interrole conflict of adult daughter caregivers. J Gerontol B Psychol Sci Soc Sci. 2001;56:24–34. 29. Stephens MA, Townsend AL, Martire LM, Druley JA. Balancing parent care with other roles: interrole conflict of adult daughter caregivers. J Gerontol B Psychol Sci Soc Sci. 2001;56:24–34. 30. Rodakowski J, Skidmore ER, Rogers JC, Schulz R. Does social support impact depression in caregivers of adults ageing with spinal cord injuries? Clin Rehabil. 2013;27:565–75. 30. Rodakowski J, Skidmore ER, Rogers JC, Schulz R. Does social support impact depression in caregivers of adults ageing with spinal cord injuries? Clin Rehabil. 2013;27:565–75. Conclusions Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit 31. Clay OJ, Roth DL, Wadley VG, Haley WE. Changes in social support and their impact on psychosocial outcome over a 5-year period for African American and White dementia caregivers. Int J Geriatr Psychiatry. 2008;23:857–62. 31. Clay OJ, Roth DL, Wadley VG, Haley WE. Changes in social support and their impact on psychosocial outcome over a 5-year period for African American and White dementia caregivers. Int J Geriatr Psychiatry. 2008;23:857–62. 32. Rodakowski J, Skidmore ER, Rogers JC, Schulz R. Role of social support in predicting caregiver burden. Arch Phys Med Rehabil. 2012;93:2229–36. • Convenient online submission 32. Rodakowski J, Skidmore ER, Rogers JC, Schulz R. Role of social support in predicting caregiver burden. Arch Phys Med Rehabil. 2012;93:2229–36. • Thorough peer review 33. De Graaf R, Ten Have M, Van Dorsselaer S. The Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2): design and methods. Int J Methods Psychiatr Res. 2010;19:125–41. 34. De Graaf R, Van Dorsselaer S, Tuithof M, Ten Have M. Sociodemographic and psychiatric predictors of attrition in a prospective psychiatric epidemiological study among the general population. Result of the Netherlands Mental Health Survey and Incidence Study-2. Compr Psychiatry. 2013;54:1131–9. 34. De Graaf R, Van Dorsselaer S, Tuithof M, Ten Have M. Sociodemographic and psychiatric predictors of attrition in a prospective psychiatric epidemiological study among the general population. Result of the Netherlands Mental Health Survey and Incidence Study-2. Compr Psychiatry. 2013;54:1131–9. Submit your manuscript at www.biomedcentral.com/submit
W4207001350.txt	https://journals.wlb-stuttgart.de/ojs/index.php/zwlg/article/download/1553/1646	de		Rezension von: Hermann Ehmer / Albert de Lange (Hg.): Lebenserinnerungen des Waldenserpfarrers Adolf Märkt (1861 – 1947)	Zeitschrift für württembergische Landesgeschichte	2,022	cc-by	997	568 Buchbesprechungen das das Leben der jungen Frau mit „Wandervögeln“ vergleicht. Anscheinend wurde dieser für die Jugendbewegung zentrale Begriff hier erstmals als Synonym für Sehnsucht ver wendet. Jedenfalls war der Grabstein in Berlin-Dahlem ab 1901 ein Treffpunkt der Berliner Jugendbewegung. Der Verfasser erklärt ausdrücklich, dass er eine Biographie und keine Geschichte der von Branco vertretenen Wissenschaften schreibe. Dies ist zu bedauern. Die sorgfältige Studie würde sicher mehr Leser finden, wenn die Geowissenschaften und ihre universitäre Entwicklung in den Vordergrund gerückt worden wären. Bernd Wunder Hermann Ehmer / Albert de Lange (Hg.), Lebenserinnerungen des Waldenserpfarrers Adolf Märkt (1861 – 1947) (Waldenserstudien, Bd. 6), Ubstadt-Weiher: verlag regionalkultur 2018. 269 S. mit 33 Abb. ISBN 978-3-95505-097-9. € 22,– Den Schwerpunkt des Bandes bildet die von Hermann Ehmer transkribierte Autobio graphie des Pfarrers Adolf Märkt, die hier erstmals vollständig veröffentlicht wird. Märkt wurde 1861 in Böblingen geboren, er stammte aus einfachen Verhältnissen. Nach dem Theologiestudium im Tübinger Stift leistete er sein Vikariat ab in Eberstadt/Gellmersbach bei Weinsberg, Breitenberg/Oberkollwangen bei Calw und Schweindorf bei Aalen. Nachdem er als Pfarrverweser in Ruppertshofen/Spraitbach bei Gaildorf tätig war, bekleidete er von 1888 bis 1901 seine erste Pfarrstelle in Pinache und Serres bei Mühlacker. Seine Begegnung mit der waldensischen Kultur beschrieb er rückblickend: „Meine sonst mehr ruhige und kühle Natur war bald Feuer und Flamme für die Sache.“ Die Beschäftigung mit den Waldensern ließ ihn lebenslang nicht mehr los. Einer seiner Korrespondenzpartner aus dem Piemont formulierte 1899 treffend, Märkt sei in Württemberg „waldensischer als die Waldenser“. Sein unermüdliches Streben galt der Wiedererweckung der Identität der hier lebenden Waldenser. Weitere Pfarrstellen versah Märkt ab 1901 in Hessigheim und ab 1909 in Birkach. Seinen tätigen Ruhestand verbrachte er ab 1929 in Hirsau, ab 1935 in Ludwigsburg, wo er seine Lebenserinnerungen niederschrieb. Diese sind von großer Ausführlichkeit, einem bemerkenswerten Detailreichtum und lebendigem Schreibstil geprägt. Offensichtlich verfügte Märkt über ein gutes Erinnerungsvermögen. Informative Beiträge über Märkts Leben und Wirken ergänzen den Band. Hermann Ehmer gibt einen Überblick über seinen Lebenslauf mit allgemeinen Informationen zur damaligen Ausbildung der Pfarrer und zum Berufsleben. Wie das Leben in den Waldenserdörfern Pinache und Serres aussah, ist den beiden Pfarrberichten Märkts aus den Jahren 1890 und 1901 zu entnehmen, eingeleitet und transkribiert von Friedrich Hörger. Albert de Lange schildert das Ergebnis seiner Untersuchungen des im Archiv der Deutschen Waldenservereinigung (ADWV) liegenden Briefwechsels von Pfarrer Märkt mit den Waldensern in Italien. Daniela Falk beschreibt die in Serres weit verzweigte Familie Gilles, von der sich mehrere Vertreter stark für das Waldensertum engagierten und Märkt beeindruckten. Im Anhang listet Walter Mogk den Inhalt des Adolf-Märkt-Nachlasses im Archiv der Deutschen Waldenservereinigung und im Pfarrarchiv Pinache auf. Die Bibliographie der Veröffentlichungen von Märkt und eine chronologische Auflistung von Märkts Veröffentlichungen, ebenso von Walter Mogk erstellt, runden den Band ab. Die Edition von Märkts Lebenserinnerungen ist verdienstvoll, umso mehr, als sie eine wichtige Quelle zur Geschichte der Waldenser in Württemberg und ihrer Identitätsbildung darstellt. Horizonterweiternd sind die von Albert de Lange angebrachten Fußnoten, dank Familien- und Personengeschichte 569 derer man sich gleich über die im Text genannten Personen, Geschehnisse, Begriffe etc. informieren kann. Der Band gibt zudem tiefe Einblicke in das Leben, Denken und Wirken eines württembergischen Pfarrers in der Zeit zwischen der Reichsgründung 1871 und der nationalsozialistischen Herrschaft. Er ist sorgfältig redigiert, ansprechend bebildert und mit einem Personenregister versehen. Martin Frieß Hans Hildenbrand, Hofphotograph und Pionier der frühen Farbfotografie, hg. vom Haus der Geschichte Baden-Württemberg, bearb. von Hans Christian Adam, Ubstadt-Weiher: verlag regionalkultur 2018. 296 S. mit 265, meist farb. Abb. ISBN 978-3-95505-096-2. € 24,80 Der vorliegende Band geht der frühen Farbfotografie am Beispiel des Stuttgarter Fotografen Hans Hildenbrand (1870 – 1957) in acht Beiträgen nach. Das Haus der Geschichte Baden-Württemberg, der Herausgeber des Buches, besitzt eine umfangreiche und beeindruckende Sammlung seiner Fotografien. Mit dem stattlichen und reich bebilderten Band möchte das Haus der Geschichte an den fast vergessenen „Stuttgarter Pionier der Farbfotografie“ erinnern und sein Schaffen einer breiten Öffentlichkeit bekannt machen, wie es der ehemalige Direktor des Hauses, Thomas Schnabel, im Vorwort formuliert (S. 9). In die „Bildwelten“ Hildenbrands führt Hans Christian Adam ein (S. 11 – 27). Zunehmend löste dieser sich von der Atelierfotografie mit ihren Porträtaufnahmen und widmete sich Landschaften und touristischen Motiven, wobei seine Bilder das Typische, das Schöne und das Ansprechende in den Vordergrund stellen. Solche Aufnahmen vermarktete er vor allem für die damals in Mode kommenden Postkarten, aber auch für Prospekte, Zeitungen und Zeitschriften sowie für 3-D-Bilder. Auch die neue Technik – beispielsweise Flugzeuge –, Blumen und Stillleben sowie aktuelle journalistische Themen waren ihm wichtige Betätigungsfelder. Adam sieht in Hildenbrand keinen Künstler, aber auch keinen reinen Handwerker, sondern er habe „ein wenig die kunstgewerbliche Rolle“ (S. 11) gespielt. Winfried Mönch widmet sich Leben und Werk des Fotografen (S. 28 – 57). „Hofphotograph“ war ein reiner Titel, den Hildenbrand seit der Verleihung 1903 auch nach der Abdankung König Wilhelms II. von Württemberg bis zu seinem Tod 1957 führte. Seine vielseitigen fotografischen Tätigkeiten und Verdienste – Fotograf, Inhaber eines fotografischen Ateliers, Verleger und Produzent von Postkarten sowie Händler von Foto- und Filmartikeln – sind mit diesem Titel aber nicht adäquat erfasst. Rolf H. Krauss behandelt die Aktfotografie des Stuttgarter Fotografen (S. 59 – 79), der er sich nur kurze Zeit (insgesamt 18 Aufnahmen sind bekannt) widmete und die „eine bedeutsame künstlerische Begabung“ (S. 67) erkennen ließ. In einem weiteren Beitrag (S. 80 – 93) stellt er die Fotografien Hildenbrands im amerikanischen „National Geographic Magazine“ vor, die die weltweite Verbreitung seiner Arbeiten belegen. Die Mitarbeit Hildenbrands bei der Farbenphotographischen Gesellschaft und dem Chromoplast-Verlag schildert Dieter Lorenz (S. 95 – 105), Einrichtungen, die sich speziell den Farbaufnahmen widmeten. Hier geht es vor allem um „Stereo-Bilder“ (3-D-Aufnahmen). Dorothea Peters ordnet Hildenbrand in die Frühzeit der Farbfotografie und des Farb druckes ein, die ja auch andere Fotografen nutzten (S. 107 – 137). Hildenbrand war „kein technischer Erneuerer“, sondern „ein früher Anwender“ und gehört damit „zu den Pio nieren der Farbfotographie und ihres Druckes“ (S. 136). Erstaunlich ist für Hildenbrand die Vielzahl der Medien – Ansichtskarten, Stereofotografien, Druckbeilagen, Bücher und
https://openalex.org/W4247766593	https://zenodo.org/records/1561179/files/article.pdf	English	null	Some facts bearing on the structure of atoms, particularly of the helium atom	Journal of the Franklin Institute	1,922	public-domain	492	46 CURRENT TO 46 CURRENT TO [J. F. I. 646 CURRENT TOPICS. Some Facts Bearing on the Structure of Atoms, Particularly of the Helium Atom. R. A. MILLIKAN. (Phys. Roy., Dec., I92I.)--Professor Millikan has developed a wonderful method by which he can measure the electric charge on the fragment of an atom to the same degree of accuracy with which the census enumer- ates the number of people in the United States. When he applies this method to the atoms that have suffered bombardment by alpha rays coming from a bit of polonium he can tell exactly how many electrons have been knocked off, because each detached electron has carried away a definite and known quantity of negative elec- tricity. When oxygen, nitrogen or carbon with their few elec- trons or iodine or mercury with their many electrons are subjected to assaults of the alpha particles, it is found that a single such particle practically never detaches more than one electron. " This, in itself, throws a certain light on atomic structure, for it shows conclusively that the electrons within an atom act quite independently of one another. They are certainly not in rings, of say four or eight or any other number which become unstable when one of their number is removed, or which can in general be shattered as a whole." g With helium it is different. The result is thus picturesquely phrased, " The alpha particle shooting at random through the helium atom at its maximum ionizing speed gets both electrons every sixth shot in which it gets anything." Such a fact does not determine the structure of the helium atom but it does make certain structures impossible. One such structure thus ruled out is the original Bohr arrangement according to which the helium atom has its two electrons at opposite ends of a diameter. Another is that suggested by Sommerfeld in which one electron rotates close to the nucleus while the other is in an orbit much farther away. y To kill the traditional two birds with one stone it is necessary that they be close together at the time of their demise. Similar reasoning leads to the conclusion that the two electrons which get knocked off once out of six must be close together in the atom a considerable part of the time. 46 CURRENT TO This is in consonance with the recent arrangements of Land6 and of Bohr, according to which the elec- trons move in roughly equal orbits whose planes are inclined to each other. " Bohr may also find in it support for his recent contention that the outer shell of heavy atoms possesses few electrons instead of many as postulated in most of the discussions of the " static atom." This article is from the Norman Bridge Laboratory of Physics, Pasadena, California, and may perhaps be regarded as the first fruits of Colonel Millikan's activities in his new surroundings. roundings. G. F. S.
https://openalex.org/W3119973208	https://jfootankleres.biomedcentral.com/track/pdf/10.1186/s13047-020-00444-6	English	null	Technology-enabled remote management of diabetes foot disease and potential for reduction in associated health costs: a pilot study	Journal of foot and ankle research	2,021	cc-by	1,861	© The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. LETTER TO THE EDITOR Open Access Background Diabetes foot ulceration can significantly impact quality of life and up to 20% of infected foot ulcers will require an amputation [1]. Furthermore, it has been recognised that a diabetes foot ulcer and a lower extremity amputa- tion are independent risk factors for premature death [1]. Based on NHS England data It is estimated that more than £80 million is spent on foot ulcers and ampu- tations annually in Scotland alone [2]. Early access to the diabetes foot multi-disciplinary team (MDT) is asso- ciated with a reduction in amputations [3]. However, in the Scottish Highlands local audits highlighted that more than half of patients who subsequently underwent an amputation, were unknown to or had presented late to the diabetes specialist podiatrists / MDT. Accessing spe- cialist services is challenging in rural areas and incurs significant excess travel distances and costs for patients. Main et al. Journal of Foot and Ankle Research (2021) 14:7 https://doi.org/10.1186/s13047-020-00444-6 Main et al. Journal of Foot and Ankle Research (2021) 14:7 https://doi.org/10.1186/s13047-020-00444-6 Fiona Main1, Ania Zubala2, Jane Gorman1, Sandra Jones1, Jenny Hall2, David Macfarlane1 and Sandra MacRury1,2* Fiona Main1, Ania Zubala2, Jane Gorman1, Sandra Jones1, Jenny Hall2, David Macfarlane1 and Sandra MacRury1,2* Abstract Diabetes-related foot disease, particularly when associated with amputation, affects quality of life and has a significant impact on health care costs. A pilot study using enhanced technology to facilitate remote access and video conferencing from rural locations to the diabetes MDT through a new service pathway confirmed high levels of patient satisfaction with 89% of foot ulcers improved or stable and only two minor amputations. A health economic analysis suggested potential for significant cost savings if this was scaled up regionally. Further evaluation of an integrated pathway, impact on lower limb amputation rates and full health economic assessment is recommended. Keywords: Diabetes, Foot disease, Technology-enabled, Pathway, Amputation, Economic analysis Building on our previously developed RAPID pathway (Fig. 1) to triage and manage diabetes foot ulceration in rural areas [4], we conducted a further pilot study that included an additional enhancement of the technology by incorporating ‘Direct Access’ software to enable re- mote access to clinical databases and a new video con- ference platform. We also developed an economic model to assess the potential value of the RAPID intervention project based on this pilot study. * Correspondence: sandra.macrury@ui.ac.uk 1Highland Diabetes Institute, Centre for Health Science, Inverness IV2 3JH, UK 2Division of Rural Health and Wellbeing, University of the Highlands and Islands, Inverness IV2 3JH, UK Methods Two community podiatrists reviewed patients with diabetes foot ulcers in patients’ homes or local clinical settings in two defined rural areas over a six-month period using tech- nology enabled care. A tablet device was used to capture images and, together with ‘Direct Access’ software, allowed remote access to a generic email account and clinical data- bases. An Omnirouter mini™device aided connectivity through prioritisation of cellular networks. The tablet de- vice was used for video-consultations, employing the NHS Near Me (Attend Anywhere) platform. This was facilitated * Correspondence: sandra.macrury@ui.ac.uk 1Highland Diabetes Institute, Centre for Health Science, Inverness IV2 3JH, UK 2Division of Rural Health and Wellbeing, University of the Highlands and Islands, Inverness IV2 3JH, UK Main et al. Journal of Foot and Ankle Research (2021) 14:7 Page 2 of 3 Main et al. Journal of Foot and Ankle Research Fig. 1 RAPID Pathway: process flow for community patient referral and triage to specialist diabetes foot services by the community podiatrist with the patient in their own home. Patients were triaged according to the pathway shown in Fig. 1. Ulcer duration, healing rates, amputations, and patient experience were collected. An economic model was developed to assess the value delivered by the RAPID intervention versus costs attributable to the health, and care system for those without access, based on data from the pilot study and extrapolated to the wider area. Financial proxies were identified to quantify return on investment. Estimation of amputation rates was based on the local dia- betes register (17,567) and 15% lifetime cumulative inci- dence of foot ulceration [1], giving an amputation rate of 9% in Highland region. The modelling relied on a series of assumptions around physical, psychological and social im- pact costs. high satisfaction level was reported by patients with the service pathway and technology, at 99 and 96% respectively. Based on foot ulcer and minor amputation data within our area the economic analysis hypothetical cost Table 1 Patient characteristics and ulcer outcomes Table 1 Patient characteristics and ulcer outcomes Total participants (n= 31) Male n=18 Female n= 13 Age (yr) 71 74 (Mean + range) 51–84 53–89 Location of consultation (%) (n=110) -Home 41.8 26.3 -Local clinic 13.6 8.18 -Care home 4.5 0.9 -Not recorded 1.8 2.7 Number of ulcers per patient (%) (n=31) -1 29.0 25.8 -2 9.7 9.7 -3 16.1 3.2 -4 0 0 -5 3.2 3.2 Outcomes of ulcers (%) (n=55) Improved and healing 38.2 10.9 - Complete healing 25.5 7.3 - Improved 12.7 3.6 Stable 18.1 21.8 Deteriorating 3.6 3.6 Minor toe amputations 0 3.6 Conclusion Remote patient management with a shift away from hospital-based care, to community and home-based care and use of telemedicine and/or VC is becoming increas- ingly common, particularly in relation the COVID-19 pandemic. The use of VC platforms has also been shown to be effective in screening for infection and wound- based assessment in people with diabetes [5]. The new service pathway confirmed that patients with a diabetes foot ulcer can be triaged within 24 h, ensuring timely ac- cess to a specialist foot team. There was a high level of patient satisfaction. An embedded pathway and technol- ogy solution including remote consultation can permit early intervention in patients with foot ulcers, with the potential to reduce amputations. The economic assess- ment was modelled on a small number of patients; an expanded assessment at regional level linked to minor and major amputations numbers will allow improvement in the cost valuation model. 4. MacRury S, Stephen K, Main F, Gorman J, Jones S, Macfarlane D. Reducing amputations in people with diabetes (RAPID): evaluation of a new care pathway. Int J Environ Res Public Health. 2018;15:999. p y 5. Rogers LC, Lavery LA, Joseph WS, Armstrong DA. All feet on deck – the role of podiatry during the COVID-19 pandemic. J Am Podiatr Med Assoc. https://doi.org/10.7547/20-051. Competing interests h h d l The authors declare no competing interests. There are several limitations: predominantly that it was pilot study in a relatively small group of individuals with no comparative control group. Accordingly, the limited number of minor and no major amputations over the short time-period may have been a chance find- ing or related to referral bias. The economic analysis was based on a number of assumptions and evidence-based research and related to a small group intervention. Received: 31 August 2020 Accepted: 26 December 2020 Received: 31 August 2020 Accepted: 26 December 2020 Received: 31 August 2020 Accepted: 26 December 2020 References 1. Armstrong DG, Swerdlow MA, Armstrong AA, Conte MS, Padula WV, Bus SA. Five year mortality and direct costs of care for people with diabetic foot complications are comparable to cancer. J Foot Ankle Res. 2020;13:16. 1. Armstrong DG, Swerdlow MA, Armstrong AA, Conte MS, Padula WV, Bus SA. Five year mortality and direct costs of care for people with diabetic foot complications are comparable to cancer. J Foot Ankle Res. 2020;13:16. 2. Kerr M, Barron, Chadwick P, Evans T, Kong WM, Rayman G, Sutton-Smith M, Todd G. The cost of the diabetic foot ulcer and amputations to the National Health Service in England. Diabet Med. 2019;36(8):995–1002. 3. Schofield CJ, Yu N, Jain AS, Leese GP. Decreasing amputation rates in patients with diabetes – a population-based study. Diabet Med. 2009;26(8): 773–7. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Acknowledgements Acknowledgements Frontline-consultancy for health economic modelling. Authors’ contributions d f h FM conceived of the project, participated in the design and coordination of the project, drafting manuscript; AZ collected and analysed data; JG participated in design and coordination of the project; SJ participated in design and coordination of the project; JH collected and analysed data; DM participated in coordination of the project and drafting manuscript; SM conceived of the project, participated in the design and coordination of the project, drafting manuscript. All authors read and approved the final manuscript. Abbreviations MDT M l idi i MDT: Multidisciplinary team; RAPID: Reducing Amputations in People with Diabetes; NHS: National Health Service; VC: Video-consultation Funding h l The evaluation was funded by TEC programme Scottish Government. The funder had no role in the design of the evaluation, or collection, analysis and interpretation of data or in writing the manuscript. Availability of data and materials Non-applicable. Results Thirty-one patients were referred for specialist advice using the pathway. Following triage, there were 110 community-based podiatry visits (range 1–10 visits/pa- tient), including 45 successful video-consultations with the MDT. The average home appointment time was 45 min, with actual VC times of 6–14 min. There were six failed VC connections due to technology problems. In summary, a total of 55 ft ulcers were treated over the course of the study. A total of 89% of ulcers healed, im- proved, or remained stable with an improvement or heal- ing noted in 27 (49.1%), including full healing in 18 wounds (33.2%) during the course of the project. Of the remainder, 22 (40.0%) ulcers were stable and four (7.3%) deteriorated. There were 2 (3.6%) minor amputations and no major amputations. Results are presented in Table 1. A Page 3 of 3 Page 3 of 3 Main et al. Journal of Foot and Ankle Research (2021) 14:7 Main et al. Journal of Foot and Ankle Research valuation suggested potential cost savings of £138,820 vs £252,124 without RAPID, giving a return on investment ratio of 1:1.8 meaning £1.80 saved for every £1 invested. This is likely to be an underestimate of cost saving at- tributed to major amputations and does not consider improved quality of life or wider social cost savings. Ethics approval and consent to participate Non-applicable. Availability of data and materials N li bl Availability of data and materials Non-applicable. Non-applicable.
https://openalex.org/W4394915829	https://www.mdpi.com/2077-0383/13/8/2354/pdf?version=1713436013	English	null	Impact of Bronchoscopic Lung Volume Reduction with Valves on the Pulmonary Gas Exchange	Journal of clinical medicine	2,024	cc-by	8,371	Academic Editor: Kazuyoshi Imaizumi Academic Editor: Kazuyoshi Imaizumi Journal of Clinical Medicine Journal of Clinical Medicine Journal of Clinical Medicine Journal of Clinical Medicine Article Impact of Bronchoscopic Lung Volume Reduction with Valves on the Pulmonary Gas Exchange Jane Winantea 1,, Katharina Stiehl 1, Ruediger Karpf-Wissel 1 , Faustina Funke 1, Hubertus Birte Schwarz 1, Heinz Steveling 3, Christian Taube 3, Filiz Oezkan 1 and Kaid Darwiche 1 Jane Winantea 1,, Katharina Stiehl 1, Ruediger Karpf-Wissel 1 , Faustina Funke 1, Hubertus Hautzel 2, Birte Schwarz 1, Heinz Steveling 3, Christian Taube 3, Filiz Oezkan 1 and Kaid Darwiche 1 1 Department of Pulmonology, Section of Interventional Pulmonology, University Medicine Essen, 1 Department of Pulmonology, Section of Interventional Pulmonology, University Medicine Essen, Ruhrlandklinik, University Duisburg-Essen, 45239 Essen, Germany; kaid.darwiche@rlk.uk-essen.de (K.D.) 1 Department of Pulmonology, Section of Interventional Pulmonology, University Medicine Essen, Ruhrlandklinik, University Duisburg-Essen, 45239 Essen, Germany; kaid.darwiche@rlk.uk-essen.de (K.D.) 2 Department of Nuclear Medicine, University Hospital Essen, University Duisburg-Essen, * Correspondence: jane.winantea@rlk.uk-essen.de; Tel.: +49-201-433-01 * Correspondence: jane.winantea@rlk.uk-essen.de; Tel.: +49-201-433-01 Abstract: Introduction: Bronchoscopic lung volume reduction (BLVR) with endobronchial valves has been shown to be a safe and effective treatment for patients with severe lung emphysema. Previous studies have reported a benefit in pulmonary function, exercise capacity, and quality of life after BLVR-treatment. The effect of BLVR with valves on the pulmonary gas exchange and its association with clinical outcomes has not been analyzed to date. The primary goal of this study was to investigate the impact of BLVR on the pulmonary gas exchange and the impact of the target lobe selection in patients with discordant target lobes in high-resolution computed tomography (HRCT) scan and perfusion scan on the pulmonary gas exchange and clinical outcomes. Methods: In this single-center study, we retrospectively analyzed pulmonary function tests, 6-min-walk-tests, HRCT scans, perfusion scans, and blood gas analyses in 77 patients over the course of 6 months following BLVR treatment. Results: We observed that complete lobar occlusion with bronchoscopic valves leads to a transient impairment of pulmonary gas exchange. Despite this, an overall positive clinical outcome could be shown in patients treated with endobronchial valves. If the target lobe selection based on HRCT and perfusion scans is discrepant, a selection based on the HRCT scan tends to be associated with a better outcome than a selection based on the perfusion scan. Conclusions: Complete lobar occlusion with bronchoscopic valves leads to a transient impairment of pulmonary gas exchange but nevertheless results in an overall positive clinical outcome. Citation: Winantea, J.; Stiehl, K.; Karpf-Wissel, R.; Funke, F.; Hautzel, H.; Schwarz, B.; Steveling, H.; Taube, C.; Oezkan, F.; Darwiche, K. Impact of Bronchoscopic Lung Volume Reduction with Valves on the Pulmonary Gas Exchange. J. Clin. Med. 2024, 13, 2354. https://doi.org/ 10.3390/jcm13082354 Keywords: bronchoscopic lung volume reduction; endoscopic valves; lung emphysema 1. Introduction Received: 28 February 2024 Revised: 3 April 2024 Accepted: 12 April 2024 Published: 18 April 2024 Chronic obstructive pulmonary disease (COPD) is one of the most common chronic respiratory diseases, affecting 251 million people globally. It is estimated to become the third leading cause of death in 2030 by the World Health Organisation (WHO) [1]. The vast majority of COPD patients with predominant emphysema phenotype suffer from breathlessness due to hyperinflation, leading to notable impairment in exercise capacity and quality of life, despite optimal pharmacological therapy and rehabilitation [2]. The National Emphysema Treatment Trial (NETT) published in 2003 showed that patients with severe emphysema who underwent lung volume reduction surgery benefitted significantly with improvements in pulmonary function, exercise capacity, and quality of life [3]. Providing a less invasive option for patients who might not be suitable for lung volume reduction surgery, bronchoscopic lung volume reduction with endobronchial valves was introduced in Europe in the mid-to-late 2000s. Since then, significant clinical trials and studies further established the role of endobronchial valves. Patients with severe impairment of the pulmonary function (forced expiration in one second less than 40% of the predicted value) One-way emphysema valves (Zephyr, PulmonX, Redwood City, CA, USA) blocking the left upper lobe are shown. One valve is placed in LB 1/2, one valve in LB 3, and one valve in the lingula. Figure 1. Bronchoscopic lung volume reduction of the left upper lobe. One-way emphysema valves (Zephyr, PulmonX, Redwood City, CA, USA) blocking the left upper lobe are shown. One valve is placed in LB 1/2, one valve in LB 3, and one valve in the lingula. Figure 1. Bronchoscopic lung volume reduction of the left upper lobe. One-way emphysema valves (Zephyr, PulmonX, Redwood City, CA, USA) blocking the left upper lobe are shown. One valve is placed in LB 1/2, one valve in LB 3, and one valve in the lingula. Figure 1. Bronchoscopic lung volume reduction of the left upper lobe. One-way emphysema valves (Zephyr, PulmonX, Redwood City, CA, USA) blocking the left upper lobe are shown. One valve is placed in LB 1/2, one valve in LB 3, and one valve in the lingula. The treatment is reversible, and has been shown as a safe and effective therapy with clinically meaningful improvement in pulmonary function, exercise capacity, and quality of life [4,5]. BLVR has been proven to be an effective treatment option both in heterogeneous and homogeneous emphysema [6,7]. The treatment is reversible, and has been shown as a safe and effective therapy with clinically meaningful improvement in pulmonary function, exercise capacity, and quality of life [4,5]. BLVR has been proven to be an effective treatment option both in heterogeneous and homogeneous emphysema [6,7]. g g p y The absence of interlobar collateral ventilation and complete lobar occlusion, resulting in significant volume reduction of the target lobe, have been shown as an important predictor for the clinical response [4,8,9]. While previous studies have analyzed the effect of complete lobar occlusion based on pulmonary function and exercise capacity, it has not yet been examined to what extent the lobar occlusion influences the pulmonary gas exchange. We hypothesized that the pulmonary gas exchange would be affected by the lobar occlusion. As a widely practiced standard, the selection of the target lobe is based on the emphysema distribution in high-resolution computed tomography scan (HRCT) and perfusion scintigraphy [10]. However, in some cases, the most diseased lobe in HRCT and perfusion scintigraphy are not identical. or a perfusion scan on the p 2. Materials and Methods 2. Materials and Methods To test the hypothesis of this study, we analyzed clinical data of patients with severe lung emphysema who underwent bronchoscopic lung volume reduction with one-way To test the hypothesis of this study, we analyzed clinical data of patients with severe lung emphysema who underwent bronchoscopic lung volume reduction with one-way valves in our tertiary care center at Ruhrlandklinik Essen, Germany, retrospectively. The study complies with the Declaration of Helsinki, and the study protocol was approved by the local ethics committee of the University Duisburg-Essen (approval no. 15-6452-BO). In such cases, it is still unclear whether the selection of the target lobe, based on HRCT or perfusion scintigraphy, results in the best g p y The absence of interlobar collateral ventilation and complete lobar occlusion, resulting in significant volume reduction of the target lobe, have been shown as an important predictor for the clinical response [4,8,9]. While previous studies have analyzed the effect of complete lobar occlusion based on pulmonary function and exercise capacity, it has not yet been examined to what extent the lobar occlusion influences the pulmonary gas exchange. We hypothesized that the pulmonary gas exchange would be affected by the lobar occlusion. As a widely practiced standard, the selection of the target lobe is based on the emphysema distribution in high-resolution computed tomography scan (HRCT) and perfusion scintigraphy [10]. However, in some cases, the most diseased lobe in HRCT and perfusion scintigraphy are not identical. In such cases, it is still unclear whether the selection of the target lobe, based on HRCT or perfusion scintigraphy, results in the best outcome. outcome. The primary goal of this study was to investigate the impact of BLVR on the pulmonary gas exchange, and the impact of the target lobe selection based either on HRCT The primary goal of this study was to investigate the impact of BLVR on the pulmonary gas exchange, and the impact of the target lobe selection based either on HRCT or a perfusion scan on the pulmonary gas exchange and clinical outcome. Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/jcm J. Clin. Med. 2024, 13, 2354. https://doi.org/10.3390/jcm13082354 J. Clin. Med. 2024, 13, 2354 2 of 11 40% of n 175% 2 of 11 40% of n 175% and pronounced hyperinflation (residual volume greater than 175% of the predicted value) were shown to benefit significantly from the treatment [4–8]. Based on the results of clinical trials, including the LIBERATE study, endobronchial valve treatment was approved by the Food and Drug Administration for use in the North American market in June 2018 [6].l on the results of clinical trials, including the LIBERATE study, endobronchial valve treatment was approved by the Food and Drug Administration for use in the North American market in June 2018 [6]. Bronchoscopic lung volume reduction (BLVR) with valves aims to reduce and pronounced hyperinflation (residual volume greater than 175% of the predicted value) were shown to benefit significantly from the treatment [4–8]. Based on the results of clinical trials, including the LIBERATE study, endobronchial valve treatment was approved by the Food and Drug Administration for use in the North American market in June 2018 [6]. on the results of clinical trials, including the LIBERATE study, endobronchial valve treatment was approved by the Food and Drug Administration for use in the North American market in June 2018 [6]. Bronchoscopic lung volume reduction (BLVR) with valves aims to reduce Bronchoscopic lung volume reduction (BLVR) with valves aims to reduce hyperinfla- tion by blocking the most diseased lobe of the lung with one-way valves (Figure 1). p g ( ) hyperinflation by blocking the most diseased lobe of the lung with one-way valves (Figure 1). Figure 1. Bronchoscopic lung volume reduction of the left upper lobe. One-way emphysema valves (Zephyr, PulmonX, Redwood City, CA, USA) blocking the left upper lobe are shown. One valve is placed in LB 1/2, one valve in LB 3, and one valve in the lingula. Figure 1. Bronchoscopic lung volume reduction of the left upper lobe. One-way emphysema valves (Zephyr, PulmonX, Redwood City, CA, USA) blocking the left upper lobe are shown. One valve is placed in LB 1/2, one valve in LB 3, and one valve in the lingula. Figure 1. Bronchoscopic lung volume reduction of the left upper lobe. 2.1. Patient Selection and Valve Implantation Patients treated with endobronchial valves in our center between 2015 and 2017 were included in this analysis. Patients, who underwent valve removal, revision, or replacement in the 6-month follow-up (FU) period were excluded. Patients with significant J. Clin. Med. 2024, 13, 2354 3 of 11 3 of 11 hyperinflation (residual volume RV > 200% pred.) and reduced forced expiration in one second (FEV1 ≤40% pred.) in their pre-treatment lung function test were evaluated for BLVR. A 6 min walk test (6-MWT), HRCT, as well as a perfusion scan were performed. Patient and target lobe selection for treatment with endobronchial valves was discussed prior to the treatment in a multidisciplinary emphysema board at our center in consideration of surgical and other endoscopic lung volume reduction techniques. g p g q Moreover, fissure integrity was visually rated by radiologists experienced in HRCT evaluation. Absence of collateral ventilation was confirmed using catheter-based Chartis® evaluation, as described previously [9]. Endoscopic valve implantation was performed in a standard manner described in previous studies [11,12]. In brief, all patients received a complete lobar occlusion of the most emphysematous lobe by implantation of one or several endobronchial valves (Zephyr®, Pulmonx, Inc., Palo Alto, CA, USA) under total intravenous anesthesia. All patients were monitored on the intensive care unit for 24 h. Chest radiography was performed 2 h post-intervention and on the next day to rule out pneumothorax and assess target lobe volume reduction (TLVR). To verify the correct position of the endobrochial valves and rule out valve dislocation, a bronchoscopy was performed two days after valve placement, as well as in the Fus after one month, three, six, and twelve months. 2.3. Selection of the Target Lobe Even though the data was analyzed retrospectively, two interventional pulmonolo- gists (CT1 und CT2) received only the CT scan and not the perfusion scan, while another two physicians specialized in nuclear medicine received only the lung perfusion scan and not the CT scan. All four physicians did not have access to other clinical information. Each of them was instructed to choose the most applicable target lobe, based on a visual quantification of the emphysema severity in each lobe. In the same manner, an interven- tional pulmonologist with a year-long experience in perfusion scan analysis and a nuclear medicine physician (PS1 and PS2) chose the target lobe based on the perfusion scan. TLVR was assessed based on chest radiography one day, three months, and six months after valve placement, as well as using HRCT 3 months post-intervention. Patients with complete lobar atelectasis and partial atelectasis were assigned to the TLVR group, whereas the non-TLVR group consisted of patients with no volume reduction or dystelectasis with negligible volume reduction. 2.2. Clinical and Radiological Data Acquisition Pulmonary function tests (PFTs) and 6-MWTs were evaluated one month, three months, and six months after the intervention. Pulmonary gas exchange was measured using the alveolar-arterial oxygen gradient (AaDO2) in a blood gas analysis at rest. Additionally, we evaluated oxygen desaturation (∆SpO2) in patients with an identical rate of supplemental oxygen during a 6 min walk test before and after the BLVR intervention. Oxygen saturation was recorded continuously during the test. ∆SpO2 was calculated by subtracting the lowest oxygen saturation (SpO2) from the SpO2 at rest. 3.2. Selected Target Lobe Based on HRCT and Perfusion Scan 3.2. Selected Target Lobe Based on HRCT and Perfusion Scan The selected target lobe by CT1 and CT2 matched in all cases. The target lobes selected in the HRCT scans did not match the treated lobe in 7 (9%) of the cases. Selection of the target lobes using PS1 and PS2 did not match in three of the cases, corresponding to an inter-observer agreement of 96.1%. In the event of disagreement upon the target lobe, a third interventional pulmonologist (PS3) was queried to choose a target lobe. The selection of PS3 matched those of PS1 in two cases and PS2 in one case. Overall, the target lobes selected based on the perfusion scan did not match the treated lobe in eight (10%) cases. 2.4. Statistical Analysis Data were statistically analyzed using SPSS Statistics (IBM SPSS Statistics for Win- dows, Version 25). Baseline characteristics are presented as mean values and standard deviation. Paired samples student t-test was used to evaluate changes from baseline to post-intervention values. Not-normally distributed parameters, due to small sam- ple size, were analyzed via Wilcoxon signed-rank-test. p-values < 0.05 were considered statistically significant. J. Clin. Med. 2024, 13, 2354 4 of 11 3. Results 3.1. Patient Characteristics 3.1. Patient Characteristics A total of 77 patients (mean age 62 years, 32 male) were included in the analysis. Baseline data are shown in detail in Table 1. Forty-three patients were on long-term oxygen therapy at baseline. In 25 (32.5%) patients, the left upper lobe was treated; in 23 (29.9%) the left lower lobe was treated; in 1 (1.3%) patient, the right middle lobe was treated; in 16 (20.8%) patients, the right upper lobe was treated; and in 11 (14.3%) patients, the right lower lobe was treated. One patient (1.3%) received valve treatment in both the right upper lobe and middle lobe. Table 1. Patients characteristics at baseline. Gender 32 males (41.6%) 45 females (58.4%) Age (years) 62.36 ± 7.17 BMI (kg/m2) 22.79 ± 4.0 FEV1 (L) 0.77 ± 0.2 FEV1 (%predicted) 28.64 ± 6.6 IVC (L) 2.3 ± 0.8 IVC (%predicted) 66.2 ± 16.1 RV (L) 5.79 ± 1.08 RV (%predicted) 268.29 ± 53.03 KCO (%predicted) 32.8 ± 11.9 AaDO2 (mmHg) 29.09 ± 9.95 Patients on long-term oxygen therapy 43 (55.8%) 6-MWD (m) 286.86 ± 77.2 Table 1. Patients characteristics at baseline. 3.3. Primary Outcome Measures Pulmonary Gas Exchange A significantly worsened pulmonary gas exchange was observed after one month in both the whole group and the TLVR group, shown in Figure 2. However, the pulmonary gas exchange showed no relevant impairment after three and six months. Using the same rate of supplemental oxygen in 24/77 patients, significantly lower desaturation rates were detected in the 6 months-FU (∆SpO2% from 6.71(±4.8)% to 4.96(±3.2)%, p = 0.026). J. Clin. Med. 2024, 13, 2354 J. Clin. Med. 2024, 13, x FOR 5 of 11 Figure 2. AaDO2 at baseline, after 1 month, 3 months, and 6 months. (a) AaDO2 in the all patients at baseline, after 1 month, 3 months, and 6 months, showing a signiﬁcant worsening of the gas exchange in the one-month-follow up, returning to baseline after 3 months. (b) AaDO2 in patients with TLVR, at baseline, after 1 month, 3 months, and 6 months, showing a signiﬁcant worsening of the gas exchange in the 1-month-follow up and 6 month-follow-up. (c) AaDO2 in patients who did not achieve TLVR; no signiﬁcant change was observed after 1 month, 3 months, and 6 months follow-up. p-values compared to baseline. No p-value indicates non-signiﬁcance. Figure 2. AaDO2 at baseline, after 1 month, 3 months, and 6 months. (a) AaDO2 in the all patients at baseline, after 1 month, 3 months, and 6 months, showing a significant worsening of the gas exchange in the one-month-follow up, returning to baseline after 3 months. (b) AaDO2 in patients with TLVR, at baseline, after 1 month, 3 months, and 6 months, showing a significant worsening of the gas exchange in the 1-month-follow up and 6 month-follow-up. (c) AaDO2 in patients who did not achieve TLVR; no significant change was observed after 1 month, 3 months, and 6 months follow-up. p-values compared to baseline. No p-value indicates non-significance. Figure 2. AaDO2 at baseline, after 1 month, 3 months, and 6 months. (a) AaDO2 in the all patients at baseline, after 1 month, 3 months, and 6 months, showing a signiﬁcant worsening of the gas exchange in the one-month-follow up, returning to baseline after 3 months. (b) AaDO2 in patients with TLVR, at baseline, after 1 month, 3 months, and 6 months, showing a signiﬁcant worsening of the gas exchange in the 1-month-follow up and 6 month-follow-up. 3.4.1. Target Lobe Volume Reduction 3.4.1. Target Lobe Volume Reduction A signiﬁcant lung volume reduction was achieved in 48 patients (62%), whereas 29 patients (38%) remained without TLVR. Signiﬁcant lung volume reduction persisted in 87.5% of cases, while six patients lost their initial lung volume reduction in the 6-month U A significant lung volume reduction was achieved in 48 patients (62%), whereas 29 patients (38%) remained without TLVR. Significant lung volume reduction persisted in 87.5% of cases, while six patients lost their initial lung volume reduction in the 6-month FU. 3.3. Primary Outcome Measures Pulmonary Gas Exchange (c) AaDO2 in patients who did not achieve TLVR; no signiﬁcant change was observed after 1 month, 3 months, and 6 months follow-up. p-values compared to baseline. No p-value indicates non-signiﬁcance. Figure 2. AaDO2 at baseline, after 1 month, 3 months, and 6 months. (a) AaDO2 in the all patients at baseline, after 1 month, 3 months, and 6 months, showing a significant worsening of the gas exchange in the one-month-follow up, returning to baseline after 3 months. (b) AaDO2 in patients with TLVR, at baseline, after 1 month, 3 months, and 6 months, showing a significant worsening of the gas exchange in the 1-month-follow up and 6 month-follow-up. (c) AaDO2 in patients who did not achieve TLVR; no significant change was observed after 1 month, 3 months, and 6 months follow-up. p-values compared to baseline. No p-value indicates non-significance. Figure 2. AaDO2 at baseline, after 1 month, 3 months, and 6 months. (a) AaDO2 in the all patients at baseline, after 1 month, 3 months, and 6 months, showing a signiﬁcant worsening of the gas exchange in the one-month-follow up, returning to baseline after 3 months. (b) AaDO2 in patients with TLVR, at baseline, after 1 month, 3 months, and 6 months, showing a signiﬁcant worsening of the gas exchange in the 1-month-follow up and 6 month-follow-up. (c) AaDO2 in patients who did not achieve TLVR; no signiﬁcant change was observed after 1 month, 3 months, and 6 months follow-up. p-values compared to baseline. No p-value indicates non-signiﬁcance. Figure 2. AaDO2 at baseline, after 1 month, 3 months, and 6 months. (a) AaDO2 in the all patients at baseline, after 1 month, 3 months, and 6 months, showing a significant worsening of the gas exchange in the one-month-follow up, returning to baseline after 3 months. (b) AaDO2 in patients with TLVR, at baseline, after 1 month, 3 months, and 6 months, showing a significant worsening of the gas exchange in the 1-month-follow up and 6 month-follow-up. (c) AaDO2 in patients who did not achieve TLVR; no significant change was observed after 1 month, 3 months, and 6 months follow-up. p-values compared to baseline. No p-value indicates non-significance. FU. 3.4.2. Pulmonary Function Test and Exercise Capacity FU. 3.4.2. Pulmonary Function Test and Exercise Capacity FU. 3.4.2. Pulmonary Function Test and Exercise Capacity 3.4.2. Pulmonary Function Test and Exercise Capacity FEV1 and RV at baseline as well as after 1 month, 3 months, and 6 months post- intervention are presented in Figure 3. In the whole sample group, FEV1 increased signiﬁcantly one month post-intervention, with the improvement maintained after three months. Patients with TLVR improved in all three FU both in terms of FEV1 and hyperinﬂation, whereas patients without atelectasis displayed no signiﬁcant changes after valve implantation Exercise tolerance measured using 6-MWT signiﬁcantly improved in FEV1 and RV at baseline as well as after 1 month, 3 months, and 6 months post- intervention are presented in Figure 3. In the whole sample group, FEV1 increased signifi- cantly one month post-intervention, with the improvement maintained after three months. Patients with TLVR improved in all three FU both in terms of FEV1 and hyperinflation, whereas patients without atelectasis displayed no significant changes after valve implanta- tion. Exercise tolerance measured using 6-MWT significantly improved in the whole group after 1 and 3 months. J. Clin. Med. 2024, 13, 2354 6 of 11 J. Clin. Med. 2024, 13, x FOR PEER REVIEW 7 of 12 Figure 3. FEV1 and RV at baseline and after 1 month, 3 months, and 6 months post-intervention. (a) FEV1 and RV in all patients with signiﬁcant improvement of FEV1 and reduction of hyperinﬂation. (b) FEV1 and RV in patients with TLVR, showing signiﬁcant improvement of FEV1 and reduction of hyperinﬂation in all follow ups. (c) FEV1 and RV in patients without TLVR, with no signiﬁcant change. p-values compared to baseline. No p-value indicates non-signiﬁcance. In patients with TLVR, exercise capacity signiﬁcantly improved in all three FUs. In patients without successful lobar occlusion, walking distances did not signiﬁcantly improve (Figure 4). Figure 3. FEV1 and RV at baseline and after 1 month, 3 months, and 6 months (a) FEV1 and RV in all patients with significant improvement of FEV1 and reduction (b) FEV1 and RV in patients with TLVR, showing significant improvement of FE of hyperinflation in all follow ups. (c) FEV1 and RV in patients without TLVR, w change. p-values compared to baseline. No p-value indicates non-significance. lin. Med. 2024, 13, x FOR PEER REVIEW 7 of 1 Figure 3. FEV1 and RV at baseline and after 1 month, 3 months, and 6 months post-intervention. FU. 3.4.2. Pulmonary Function Test and Exercise Capacity (a) FEV1 and RV in all patients with signiﬁcant improvement of FEV1 and reduction of hyperinﬂation. (b) FEV1 and RV in patients with TLVR, showing signiﬁcant improvement of FEV1 and reduction of hyperinﬂation in all follow ups. (c) FEV1 and RV in patients without TLVR, with no signiﬁcant change. p-values compared to baseline. No p-value indicates non-signiﬁcance. In patients with TLVR, exercise capacity signiﬁcantly improved in all three FUs. In patients without successful lobar occlusion, walking distances did not signiﬁcantly i o e (Fi u e 4) Figure 3. FEV1 and RV at baseline and after 1 month, 3 months, and 6 months post-intervention. (a) FEV1 and RV in all patients with significant improvement of FEV1 and reduction of hyperinflation. (b) FEV1 and RV in patients with TLVR, showing significant improvement of FEV1 and reduction of hyperinflation in all follow ups. (c) FEV1 and RV in patients without TLVR, with no significant change. p-values compared to baseline. No p-value indicates non-significance. J. Clin. Med. 2024, 13, 2354 Clin. Med. 2024, 13, 2354 In patients with TLVR, e In patients without successful improve (Figure 4). n. Med. 2024, 13, x FOR PEER REVIEW Figure 4. 6-MWD at baseline, after signiﬁcant improvement after 1 m with TLVR, showing an improvem MWD in patients without TLVR, sh to baseline. No p-value indicates n 3.4.3. Impact of Discrepancy of Outcome A relevant change in the based on either HRCT or perfu increase of FEV1 from baseline selected based on the HRCT signiﬁcant, in patients in which scan but not the HRCT, FEV1 d in patients where the treated lo not the HRCT, no signiﬁcant where the treated lobe matche the gas exchange responded signiﬁcantly, then evening out Figure 4. 6-MWD at baseline, afte with significant improvement afte patients with TLVR, showing an i up. (c) 6-MWD in patients withou compared to baseline. No p-value 3.4.3. Impact of Discrepancy of Clinical Outcome A relevant change in the based on either HRCT or perfus increase of FEV1 from baselin one selected based on the HRC significant, in patients in which scan but not the HRCT, FEV1 d in patients where the treated lob the HRCT, no significant chang the treated lobe matched the o exchange responded exactly lik evening out in the further follo 7 of 11 In patients with TLVR, exercise capacity significantly improved in all three FUs. FU. 3.4.2. Pulmonary Function Test and Exercise Capacity In patients without successful lobar occlusion, walking distances did not significantly improve (Figure 4). EVIEW 8 of Figure 4. 6-MWD at baseline, after 1 month, 3 months, and 6 months. (a) 6-MWD in all patients wi signiﬁcant improvement after 1 month, returning to baseline after 3 months. (b) 6-MWD in patien with TLVR, showing an improvement in the one-month-follow up and six-month-follow up. (c) MWD in patients without TLVR, showing no signiﬁcant change in all follow ups. p-values compare to baseline. No p-value indicates non-signiﬁcance. 3.4.3. Impact of Discrepancy of the Target Lobe in HRCT and Perfusion Scan on Clinical Outcome Figure 4. 6-MWD at baseline, after 1 month, 3 months, and 6 months. (a) 6-MWD in all patients with significant improvement after 1 month, returning to baseline after 3 months. (b) 6-MWD in patients with TLVR, showing an improvement in the one-month-follow up and six-month-follow up. (c) 6-MWD in patients without TLVR, showing no significant change in all follow ups. p-values compared to baseline. No p-value indicates non-significance. 3.4.3. Impact of Discrepancy of the Target Lobe in HRCT and Perfusion Scan on Clinical Outcome Figure 4. 6-MWD at baseline, after 1 month, 3 months, and 6 months. (a) 6-MWD in all patients wi signiﬁcant improvement after 1 month, returning to baseline after 3 months. (b) 6-MWD in patien with TLVR, showing an improvement in the one-month-follow up and six-month-follow up. (c) MWD in patients without TLVR, showing no signiﬁcant change in all follow ups. p-values compare to baseline. No p-value indicates non-signiﬁcance. Figure 4. 6-MWD at baseline, after 1 month, 3 months, and 6 months. (a) 6-MWD in all patients with significant improvement after 1 month, returning to baseline after 3 months. (b) 6-MWD in patients with TLVR, showing an improvement in the one-month-follow up and six-month-follow up. (c) 6-MWD in patients without TLVR, showing no significant change in all follow ups. p-values compared to baseline. No p-value indicates non-significance. Figure 4. 6-MWD at baseline, after 1 month, 3 months, and 6 months. (a) 6-MWD in all patients wi signiﬁcant improvement after 1 month, returning to baseline after 3 months. (b) 6-MWD in patien with TLVR, showing an improvement in the one-month-follow up and six-month-follow up. (c) MWD in patients without TLVR, showing no signiﬁcant change in all follow ups. p-values compare to baseline. No p-value indicates non-signiﬁcance. Figure 4. FU. 3.4.2. Pulmonary Function Test and Exercise Capacity 6-MWD at baseline, after 1 month, 3 months, and 6 months. (a) 6-MWD in all patients with significant improvement after 1 month, returning to baseline after 3 months. (b) 6-MWD in patients with TLVR, showing an improvement in the one-month-follow up and six-month-follow up. (c) 6-MWD in patients without TLVR, showing no significant change in all follow ups. p-values compared to baseline. No p-value indicates non-significance. A relevant change in the clinical outcome in patients with a discrepant target lob based on either HRCT or perfusion scan could be shown. We could demonstrate a relevan increase of FEV1 from baseline after one month, where the treated lobe matched the on selected based on the HRCT (p < 0.05). On the other hand, although statistically n signiﬁcant, in patients in which the treated lobe matched the lobe selected in the perfusio scan but not the HRCT, FEV1 decreased by 0.13 L after one month (p = 0.076). Furthermor in patients where the treated lobe matched the selection based on the perfusion scan bu not the HRCT, no signiﬁcant change in the AaDO2 was observed, whereas in patien where the treated lobe matched the one based on the HRCT but not the perfusion sca the gas exchange responded exactly like in the whole group, initially worsenin A relevant change in the clinical outcome in patients with a discrepant target lobe based on either HRCT or perfusion scan could be shown. We could demonstrate a relevant increase of FEV1 from baseline after one month, where the treated lobe matched the one selected based on the HRCT (p < 0.05). On the other hand, although statistically not significant, in patients in which the treated lobe matched the lobe selected in the perfusion scan but not the HRCT, FEV1 decreased by 0.13 L after one month (p = 0.076). Furthermore, in patients where the treated lobe matched the selection based on the perfusion scan but not the HRCT, no significant change in the AaDO2 was observed, whereas in patients where the treated lobe matched the one based on the HRCT but not the perfusion scan, the gas exchange responded exactly like in the whole group, initially worsening significantly, then evening out in the further follow-ups (Figure 5). J. Clin. Med. 2024, 13, 2354 J. Clin. Med. 2024, 13, x FOR P 8 of 11 9 of 12 Figure 5. FU. 3.4.2. Pulmonary Function Test and Exercise Capacity (a) FEV1 in patients with discrepancy in the target lobe, showing an improvement of FEV1 in patients, where the target lobe matches the HRCT and a worsening of FEV1 in patients, where the target lobe was selected based on the perfusion scan. (b) Gas exchange in patients with discrepancy in the target lobe, showing a more distinct worsening of the AaDO2 in patients, where the target lobe matches the HRCT in the one-month-follow up, evening out in the next follow ups. Figure 5. FEV1 and AaDO2 in patients with discrepancy in the target lobe. (a) FEV1 in patients with discrepancy in the target lobe, showing an improvement of FEV1 in patients, where the target lobe matches the HRCT and a worsening of FEV1 in patients, where the target lobe was selected based on the perfusion scan. (b) Gas exchange in patients with discrepancy in the target lobe, showing a more distinct worsening of the AaDO2 in patients, where the target lobe matches the HRCT in the one-month-follow up, evening out in the next follow ups. FU. 3.4.2. Pulmonary Function Test and Exercise Capacity FEV1 and AaDO2 in patients with discrepancy in the target lobe. (a) FEV1 in patients discrepancy in the target lobe, showing an improvement of FEV1 in patients, where the target matches the HRCT and a worsening of FEV1 in patients, where the target lobe was selected b on the perfusion scan. (b) Gas exchange in patients with discrepancy in the target lobe, showi more distinct worsening of the AaDO2 in patients, where the target lobe matches the HRCT in one-month-follow up, evening out in the next follow ups. 4 Discussion Figure 5. FEV1 and AaDO2 in patients with discrepancy in the target lobe. (a) FEV1 in patients discrepancy in the target lobe, showing an improvement of FEV1 in patients, where the targe matches the HRCT and a worsening of FEV1 in patients, where the target lobe was selected on the perfusion scan. (b) Gas exchange in patients with discrepancy in the target lobe, show more distinct worsening of the AaDO2 in patients, where the target lobe matches the HRCT one-month-follow up, evening out in the next follow ups. Figure 5. FEV1 and AaDO2 in patients with discrepancy in the target lobe. (a) FEV1 in patients with discrepancy in the target lobe, showing an improvement of FEV1 in patients, where the target lobe matches the HRCT and a worsening of FEV1 in patients, where the target lobe was selected based on the perfusion scan. (b) Gas exchange in patients with discrepancy in the target lobe, showing a more distinct worsening of the AaDO2 in patients, where the target lobe matches the HRCT in the one-month-follow up, evening out in the next follow ups. 4 Discussion Figure 5. FEV1 and AaDO2 in patients with discrepancy in the target lobe. (a) FEV1 in patients with discrepancy in the target lobe, showing an improvement of FEV1 in patients, where the target lobe matches the HRCT and a worsening of FEV1 in patients, where the target lobe was selected based on the perfusion scan. (b) Gas exchange in patients with discrepancy in the target lobe, showing a more distinct worsening of the AaDO2 in patients, where the target lobe matches the HRCT in the one-month-follow up, evening out in the next follow ups. Figure 5. FEV1 and AaDO2 in patients with discrepancy in the target lobe. 4. Discussion W h th 4. Discussion We hypothesize that excluding the treated lobe from ventilation aﬀects the overall gas exchange. In this study, we can show a signiﬁcant worsening of the pulmonary gas exchange one month after bronchoscopic valve treatment. The worsening of the pulmonary gas exchange can be explained by a ventilation perfusion mismatch as a result of the lobar collapse in the treated lobe. However, the pulmonary gas exchange returned to baseline after 3 months. The best possible explanation for this is a delay in the reduction We hypothesize that excluding the treated lobe from ventilation affects the overall gas exchange. In this study, we can show a significant worsening of the pulmonary gas exchange one month after bronchoscopic valve treatment. The worsening of the pulmonary gas exchange can be explained by a ventilation perfusion mismatch as a result of the lobar collapse in the treated lobe. However, the pulmonary gas exchange returned to baseline J. Clin. Med. 2024, 13, 2354 9 of 11 9 of 11 after 3 months. The best possible explanation for this is a delay in the reduction of perfusion in the volume-reduced lobe. Additionally, we showed that the worsening of the pulmonary gas exchange 4 weeks after treatment did not co-concur with a worsened exercise capacity measured using 6-MWT. Moreover, a decrease of desaturation despite an identical rate of supplemental oxygen during the 6-MWT was observed after 6 months, presumably because the reduction of pulmonary hyperinflation compensates for the impaired gas exchange. This corresponds to the finding of Pizarro et al. and Chung et al., showing a redistribution of the ventilation and perfusion to the untreated lobe after completed lobar occlusion with valves [13–15].i Our finding implicates that, while the worsening of the gas exchange diminishes, patients benefit further from the reduction of hyperinflation, which has been shown to be associated with an improvement of both breathing mechanics and cardiac function [16–18]. These findings are in accordance with the fact that patients with homogenous emphysema also benefit from valve treatment, despite a relatively lower destruction degree of the target lobe [7]. We assume that the higher the degree of hyperinflation, the less clinically significant the impairment of the gas exchange. This could contribute to patient selection for valve treatment, especially patients needing high amounts of supplemental oxygen before treatment. 4. Discussion W h th 4. Discussion In these patients, valve treatment can be considered despite risk of transient increase of supplemental oxygen if a pronounced hyperinflation is present. In any case, informing patients about a possible transient worsening of the gas exchange, or even a temporary need of supplemental oxygen, should be considered as part of the informed consent. It is conceivable that the transient deterioration of the pulmonary gas exchange is correlated to the degree of destruction of the treated lobe. As quantitative HRCT analysis were not yet consistently available in our center at the time of the valve placement, we were not able to examine whether the change in pulmonary gas exchange correlated with the level of destruction and/or volume of the treated lobe. This therefore remains unanswered and subject to further study. The subsequent selection of the target lobe by experts, visually either based on HRCT or a perfusion scan, blinded to the actual treated lobe, showed a discrepancy in a few cases. In these cases, we could show that clinical outcome seems to be better if the target-lobe is selected based on HRCT scan. Also, the fact that visual selection of the target lobe based on a perfusion scan was in some cases inconsistent between two experts indicates that this may possibly be less reliable than the HRCT. Thomsen et al. showed that the perfusion of the target lobe did not have any effect on the clinical outcomes, which supports our hypothesis [19]. A possible explanation could be the lack of single photon emission computed tomography (SPECT) imaging with lobe-based quantification for the selection of the target lobe. Apart from the retrospective nature of the study, there are some further limitations that should be discussed. Patients in our study showed slightly less improvement in clinical outcomes as compared to those observed in lately published randomized clinical trials [6,7,20,21]. This may be accounted for by lower percentage of patients achieving a TLVR compared to other trials. We speculate that this is explained by the lack of CT-scan based quantitative analysis of the fissure completeness before treatment at that time in our center. As few FU data are missing, it is possible that patients who had a greater benefit after treatment did not keep their FU appointment as regularly and reliably as patients who responded less to the treatment or had exacerbations. 4. Discussion W h th 4. Discussion Hence FU data could be biased, and may under-represent patients with better clinical outcomes. Additionally, pulmonary gas exchange was measured in AaDO2 calculated from capillary blood gas analysis. The retrospective character of the study carries the potential risk of not strictly standardized conditions for the collection of the blood gas analysis. J. Clin. Med. 2024, 13, 2354 10 of 11 10 of 11 References 1. WHO. Chronic Respiratory Diseases. Chronic Obstructive Pulmonary Disease (COPD). Available online: https://www.emro. who.int/health-topics/chronic-obstructive-pulmonary-disease-copd/index.html (accessed on 11 April 2024). p p y p ( p ) 2. Janson, C.; Marks, G.; Buist, S.; Gnatiuc, L.; Gislason, T.; McBurnie, M.A.; Nielsen, R.; Studnicka, M.; Toelle, B.; Benediktsdottir, B.; et al. The impact of COPD on health status: Findings from the BOLD study. Eur. Respir. J. 2013, 42, 1472–1483. [CrossRef] [PubMed] p p y p ( p ) 2. Janson, C.; Marks, G.; Buist, S.; Gnatiuc, L.; Gislason, T.; McBurnie, M.A.; Nielsen, R.; Studnicka, M.; Toelle, B.; Benediktsdottir, B.; et al. The impact of COPD on health status: Findings from the BOLD study. Eur. Respir. J. 2013, 42, 1472–1483. [CrossRef] [PubMed] 3. Criner, G.J.; Sternberg, A.L. National Emphysema Treatment Trial: The major outcomes of lung volume reduction surgery in severe emphysema. Proc. Am. Thorac. Soc. 2008, 5, 393–405. [CrossRef] [PubMed] p y 4. Sciurba, F.C.; Ernst, A.; Herth, F.J.; Strange, C.; Criner, G.J.; Marquette, C.H.; Kovitz, K.L.; Chiacchierini, R.P.; Goldin, J.; McLennan, G.; et al. A randomized study of endobronchial valves for advanced emphysema. N. Engl. J. Med. 2010, 363, 1233–1244. [CrossRef] [PubMed] 5. Shah, P.L.; Herth, F.J. Current status of bronchoscopic lung volume reduction with endobronchial valves. Thorax 2014, 69, 280. [CrossRef] [PubMed] 6. Criner, G.J.; Sue, R.; Wright, S.; Dransfield, M.; Rivas-Perez, H.; Wiese, T.; Sciurba, F.C.; Shah, P.L.; Wahidi, M.M.; de Oliveira, H.G.; et al. A Multicenter RCT of Zephyr(R) Endobronchial Valve Treatment in Heterogeneous Emphysema (LIBERATE). Am. J. Respir. Crit. Care Med. 2018, 198, 1151–1164. [CrossRef] [PubMed] p 7. Valipour, A.; Slebos, D.J.; Herth, F.; Darwiche, K.; Wagner, M.; Ficker, J.H.; Petermann, C.; Hubner, R.H.; Stanzel, F.; Eberhardt, R.; et al. Endobronchial Valve Therapy in Patients with Homogeneous Emphysema. Results from the IMPACT Study. Am. J. Respir. Crit. Care Med. 2016, 194, 1073–1082. [CrossRef] [PubMed] 8. Eberhardt, R.; Gompelmann, D.; Schuhmann, M.; Reinhardt, H.; Ernst, A.; Heussel, C.P.; Herth, F.J.F. Complete unilateral vs partial bilateral endoscopic lung volume reduction in patients with bilateral lung emphysema. Chest 2012, 142, 900–908. [CrossRef] [PubMed] [ ] [ ] 9. Mantri, S.; Macaraeg, C.; Shetty, S.; Aljuri, N.; Freitag, L.; Herth, F.; Eberhardt, R.; Ernst, A. Technical advances: Measurement of collateral flow in the lung with a dedicated endobronchial catheter system. J. Bronchol. Interv. Pulmonol. 2009, 16, 141–144. [CrossRef] [PubMed] [ ] [ ] 10. 5. Conclusions To the best of our knowledge, this is the first study to show that a complete lobar occlusion with bronchoscopic valves leads to a transient impairment of pulmonary gas exchange. Despite the impairment of pulmonary gas exchange, we observed an overall positive clinical outcome of bronchoscopic lung volume reduction, in line with previous publications on this topic. If the target lobe selection based on HRCT and perfusion scans is discrepant, a selection based on the HRCT scan tends to be associated with a better outcome than a selection based on the perfusion scan. Author Contributions: Conceptualization, J.W. and K.D.; Methodology, J.W. and K.D.; Formal analysis, J.W.; Investigation, J.W., K.S. and H.H.; Resources, J.W. and K.D.; Data curation, J.W., K.S., R.K.-W., F.F., H.H., B.S. and H.S.; Writing—original draft, J.W.; Writing—review & editing, F.O. and K.D.; Visualization, J.W.; Supervision, C.T.; Project administration, J.W. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of the University Duisburg-Essen (protocol code 15-6452-BO, date of approval 29 June 2015). Informed Consent Statement: Patient consent was waived due to the retrospective nature of the study. Data Availability Statement: The original contributions presented in the study are included in the article material, further inquiries can be directed to the corresponding authors. Conflicts of Interest: Katharina Stiehl, Hubertus Hautzel, Heinz Steveling and Christian Taube declare no conflict of interest. References [CrossRef] [PubMed] 16 Pi C S h l R H ti l C T l t I Ni k i G Sk h D I t f d i l l d ti 15. Pizarro, C.; Ahmadzadehfar, H.; Essler, M.; Tuleta, I.; Fimmers, R.; Nickenig, G.; Skowasch, D. Effec therapy on pulmonary perfusion and ventilation distribution. PLoS ONE 2015, 10, e0118976. [CrossRe 16 Pizarro C ; Schueler R ; Hammerstingl C ; Tuleta I ; Nickenig G ; Skowasch D Impact of endoscopic l , ; , ; , ; , ; , ; g, ; , therapy on pulmonary perfusion and ventilation distribution. PLoS ONE 2015, 10, e0118976. [CrossRef] [PubMed] 16 Pizarro C ; Schueler R ; Hammerstingl C ; Tuleta I ; Nickenig G ; Skowasch D Impact of endoscopic lung volume reduction on 16. Pizarro, C.; Schueler, R.; Hammerstingl, C.; Tuleta, I.; Nickenig, G.; Skowasch, D. Impact of endoscopic lung volume reduction on right ventricular myocardial function. PLoS ONE 2015, 10, e0121377. [CrossRef] [PubMed] 16. Pizarro, C.; Schueler, R.; Hammerstingl, C.; Tuleta, I.; Nickenig, G.; Skowasch, D. Impact of endo right ventricular myocardial function. PLoS ONE 2015, 10, e0121377. [CrossRef] [PubMed] 16. Pizarro, C.; Schueler, R.; Hammerstingl, C.; Tuleta, I.; Nickenig, G.; Skowasch, D. Impact of endoscopic lung volume reduction on right ventricular myocardial function. PLoS ONE 2015, 10, e0121377. [CrossRef] [PubMed] g g p p g right ventricular myocardial function. PLoS ONE 2015, 10, e0121377. [CrossRef] [PubMed] g y 17. Watz, H.; Waschki, B.; Meyer, T.; Kretschmar, G.; Kirsten, A.; Claussen, M.; Magnussen, H. Decreasin associated heart dysfunction in COPD: Role of hyperinflation. Chest 2010, 138, 32–38. [CrossRef] [P , ; , ; y , ; , ; , ; , ; g , g sociated heart dysfunction in COPD: Role of hyperinflation. Chest 2010, 138, 32–38. [CrossRef] [PubMed] y ypl , , [ ] [ ] 18. Zoumot, Z.; LoMauro, A.; Aliverti, A.; Nelson, C.; Ward, S.; Jordan, S.; Polkey, M.I.; Shah, P.L.; Hopkinson, N.S. Lung Volume Reduction in Emphysema Improves Chest Wall Asynchrony. Chest 2015, 148, 185–195. [CrossRef] [PubMed] y ypl 18. Zoumot, Z.; LoMauro, A.; Aliverti, A.; Nelson, C.; Ward, S.; Jordan, S.; Polkey, M.I.; Shah, P.L.; Hopkinson, N.S. Lung Volume Reduction in Emphysema Improves Chest Wall Asynchrony. Chest 2015, 148, 185–195. [CrossRef] [PubMed] p y p y y 19. References Herth, F.J.; Noppen, M.; Valipour, A.; Leroy, S.; Vergnon, J.M.; Ficker, J.H.; Egan, J.J.; Gasparini, S.; Agusti, C.; Holmes- Higgin, D.; et al. Efficacy predictors of lung volume reduction with Zephyr valves in a European cohort. Eur. Respir. J. 2012, 39, 1334–1342. [CrossRef] [PubMed] [ ] [ ] 11. Herth, F.J.F.; Slebos, D.J.; Criner, G.J.; Valipour, A.; Sciurba, F.; Shah, P.L. Endoscopic Lung Volume Reduction: An Expert Panel Recommendation—Update 2019. Respiration 2019, 97, 548–557. [CrossRef] [PubMed] J. Clin. Med. 2024, 13, 2354 11 of 11 11 of 11 12. Darwiche, K.; Karpf-Wissel, R.; Eisenmann, S.; Aigner, C.; Welter, S.; Zarogoulidis, P.; Hohenforst-Schmidt, W.; Freitag, L.; Oezkan, F. Bronchoscopic Lung Volume Reduction with Endobronchial Valves in Low-FEV1 Patients. Respiration 2016, 92, 414–419. [CrossRef] [PubMed] 13. Chung, S.C.; Peters, M.J.; Chen, S.; Emmett, L.; Ing, A.J. Effect of unilateral endobronchial valve insertion on pulmonary ventilation and perfusion: A pilot study. Respirology 2010, 15, 1079–1083. [CrossRef] [PubMed] p p y p gy 14. Pizarro, C.; Ahmadzadehfar, H.; Essler, M.; Fimmers, R.; Nickenig, G.; Skowasch, D. Volumetric an 14. Pizarro, C.; Ahmadzadehfar, H.; Essler, M.; Fimmers, R.; Nickenig, G.; Skowasch, D. Volumetric and scintigraphic changes following endoscopic lung volume reduction. Eur. Respir. J. 2015, 45, 262–265. [CrossRef] [PubMed] 14. Pizarro, C.; Ahmadzadehfar, H.; Essler, M.; Fimmers, R.; Nickenig, G.; Skowasch, D. Volumetric and scintigraphic changes 14. Pizarro, C.; Ahmadzadehfar, H.; Essler, M.; Fimmers, R.; Nickenig, G.; Skowasch, D. Volumetric an following endoscopic lung volume reduction. Eur. Respir. J. 2015, 45, 262–265. [CrossRef] [PubMed] g following endoscopic lung volume reduction. Eur. Respir. J. 2015, 45, 262–265. [CrossRef] [PubMed] zarro, C.; Ahmadzadehfar, H.; Essler, M.; Tuleta, I.; Fimmers, R.; Nickenig, G.; Skowasch, D. Effect of en 15. Pizarro, C.; Ahmadzadehfar, H.; Essler, M.; Tuleta, I.; Fimmers, R.; Nickenig, G.; Skowasch, D. Effec therapy on pulmonary perfusion and ventilation distribution PLoS ONE 2015 10 e0118976 [CrossRe 15. Pizarro, C.; Ahmadzadehfar, H.; Essler, M.; Tuleta, I.; Fimmers, R.; Nickenig, G.; Skowasch, D. Effect of endobronchial valve therapy on pulmonary perfusion and ventilation distribution. PLoS ONE 2015, 10, e0118976. [CrossRef] [PubMed] 15. Pizarro, C.; Ahmadzadehfar, H.; Essler, M.; Tuleta, I.; Fimmers, R.; Nickenig, G.; Skowasch, D. Effect of endobronchial valve therapy on pulmonary perfusion and ventilation distribution. PLoS ONE 2015, 10, e0118976. References Thomsen, C.; Theilig, D.; Herzog, D.; Poellinger, A.; Doellinger, F.; Schreiter, N.; Schreiter, V.; Schurmann, D.; Temmesfeld- Wollbrueck, B.; Hippenstiel, S.; et al. Lung perfusion and emphysema distribution affect the outcome of endobronchial valve therapy. Int. J. Chronic Obstr. Pulm. Dis. 2016, 11, 1245–1259. 19. Thomsen, C.; Theilig, D.; Herzog, D.; Poellinger, A.; Doellinger, F.; Schreiter, N.; Schreiter, V.; Schurmann, D.; Temmesfeld- Wollbrueck, B.; Hippenstiel, S.; et al. Lung perfusion and emphysema distribution affect the outcome of endobronchial valve therapy. Int. J. Chronic Obstr. Pulm. Dis. 2016, 11, 1245–1259. py 20. Criner, G.J.; Delage, A.; Voelker, K.; Hogarth, D.K.; Majid, A.; Zgoda, M.; Lazarus, D.R.; Casal, R.; Benzaquen, S.B.; Holladay, R.C.; et al. Improving Lung Function in Severe Heterogenous Emphysema with the Spiration Valve System (EMPROVE). A Multicenter, Open- Label Randomized Controlled Clinical Trial. Am. J. Respir. Crit. Care Med. 2019, 200, 1354–1362. [CrossRef] [PubMed] 20. Criner, G.J.; Delage, A.; Voelker, K.; Hogarth, D.K.; Majid, A.; Zgoda, M.; Lazarus, D.R.; Casal, R.; Benzaquen, S.B.; Holladay, R.C.; et al. Improving Lung Function in Severe Heterogenous Emphysema with the Spiration Valve System (EMPROVE). A Multicenter, Open- Label Randomized Controlled Clinical Trial. Am. J. Respir. Crit. Care Med. 2019, 200, 1354–1362. [CrossRef] [PubMed] J p [ ] [ ] 21. Kemp, S.V.; Slebos, D.J.; Kirk, A.; Kornaszewska, M.; Carron, K.; Ek, L.; Broman, G.; Hillerdal, G.; Mal, H.; Pison, C.; et al. A Mul- ticenter Randomized Controlled Trial of Zephyr Endobronchial Valve Treatment in Heterogeneous Emphysema (TRANSFORM). Am. J. Respir. Crit. Care Med. 2017, 196, 1535–1543. [CrossRef] [PubMed] p 21. Kemp, S.V.; Slebos, D.J.; Kirk, A.; Kornaszewska, M.; Carron, K.; Ek, L.; Broman, G.; Hillerdal, G.; Mal, H.; Pison, C.; et al. A Mul- ticenter Randomized Controlled Trial of Zephyr Endobronchial Valve Treatment in Heterogeneous Emphysema (TRANSFORM). Am. J. Respir. Crit. Care Med. 2017, 196, 1535–1543. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://openalex.org/W2132847750	https://europepmc.org/articles/pmc2537571?pdf=render	English	null	Testing for treatment effects on gene ontology	BMC bioinformatics	2,008	cc-by	6,687	BioMed Central BioMed Central This article is available from: http://www.biomedcentral.com/1471-2105/9/S9/S20 This article is available from: http://www.biomedcentral.com/1471-2105/9/S9/S20 © 2008 Lee et al; licensee BioMed Central Ltd. ; This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract In studies that use DNA arrays to assess changes in gene expression, it is preferable to measure the significance of treatment effects on a group of genes from a pathway or functional category such as gene ontology terms (GO terms, http://www.geneontology.org) because this facilitates the interpretation of effects and may markedly increase significance. A modified meta-analysis method to combine p-values was developed to measure the significance of an overall treatment effect on such functionally-defined groups of genes, taking into account the correlation structure among genes. For hypothesis testing that allows gene expression to change in both directions, p-values are calculated under the null distribution generated by a Monte Carlo method. As a test of this procedure, we attempted to distinguish altered pathways in microarray studies performed with Mitochips, oligonucleotide microarrays specific to mitochondrial DNA-encoded transcripts. We found that our analytic method improves the specificity of selection for altered pathways, due to incorporation of the inter-gene correlation structure in each pathway. It is thus a practical method to measure treatment effects on GO groups. In many actual applications, microarray experiments measure treatment effects under complicated design structures and with small sample sizes. For such applications to real data of limited statistical power, and also in computer simulations, we demonstrate that our method gives reasonable test results. BMC Bioinformatics 2008, 9(Suppl 9):S20 doi:10.1186/1471-2105-9-S9-S20 BMC Bioinformatics 2008, 9(Suppl 9):S20 doi:10.1186/1471-2105-9-S9-S20 BMC Bioinformatics 2008, 9(Suppl 9):S20 doi:10.1186/1471-2105-9-S9-S20 Open Acce Proceedings Testing for treatment effects on gene ontology Taewon Lee1, Varsha G Desai2, Cruz Velasco3, Robert J Shmookler Reis4 and Robert R Delongchamp5 Open Access Address: 1Biometry Branch, Division of Personalized Nutrition and Medicine, National Center for Toxicological Research, 3900 NCTR Road, Jefferson, AR 72079, USA, 2Center for Functional Genomics, Division of Systems Toxicology, National Center for Toxicological Research, 3900 NCTR Road, Jefferson, AR 72079, USA, 3Louisiana State University Health Sciences Center, New Orleans, LA 70112, 4Depts. of Geriatrics. Biochemistry & Molecular Biology, and Pharmacology/Toxicology, University of Arkansas for Medical Sciences, and VA Medical Center, Little Rock, AR 72205, USA and 5Department of Epidemiology, University of Arkansas for Medical Sciences, College of Public Health, 4301 W Markham St, #820, Little Rock, AR 72205, USA Email: Taewon Lee - taewon.lee@gmail.com; Varsha G Desai - varsha.desai@fda.hhs.gov; Cruz Velasco - cvelas@lsuhsc.edu; Robert J Shmookler Reis - rjsr@uams.edu; Robert R Delongchamp - RDelongchamp@uams.edu * Corresponding author from Fifth Annual MCBIOS Conference. Systems Biology: Bridging the Omics Oklahoma City, OK, USA. 23–24 February 2008 Published: 12 August 2008 Published: 12 August 2008 creating treatment blocks within these sources of variation [9]. Such cases are not suited to permutation methods. creating treatment blocks within these sources of variation [9]. Such cases are not suited to permutation methods. ity to derive biologically meaningful conclusions from such data, have always been daunting tasks for statisti- cians. Because of the high volume and complex character- istics of microarray data, much of the initial work on their analysis has focused on development of data mining or data reduction methods to identify differentially expressed genes. Typically, the p-value of a test statistic is calculated for each gene, the genes are ranked according to these p-values, and a pre-specified significance criterion, such as the false discovery rate, is used to determine a cut- off which creates a category of differentially expressed genes [1-3]. A modified meta-analysis method was developed by Delongchamp et al. [10] to combine p-values, and thus to measure the significance of an overall treatment effect on a group of genes, while taking into account the inter-genic correlation structure. The method is based on the fact that p-values follow a uniform distribution under the null hypothesis. Inverse-normal transformed p-values have a normal distribution and their sum over a set of genes also would follow a normal distribution, provided that the component p-values are independent. The test we have developed to measure the significance of overall treat- ment effect on genes within a GO category is based on a modification of this statistic, to reflect the actual correla- tion structure among genes sharing a GO term. Attempts to interpret individual genes in a list of signifi- cant genes are demanding and laborious. Therefore, recent efforts have focused on discovery of biological pathways rather than on individual gene function. Gene ontology terms (GO terms, http://www.geneontol ogy.org) reflect gene groupings based on molecular func- tion, biological process, or cellular structure/organelle. The interpretation of differentially expressed GO groups is generally simpler than the presentation of a list of statisti- cally significant genes, and more resistant to erroneous conclusions that can arise from microarray artefacts. In this paper, we extend the method from a simple one- class t-test with the null hypothesis H0 : μ = 0 to a test for pair-wise contrast in a fixed-effects linear model. In the following sections, we describe in detail the extension of the methodology, with validation through computer sim- ulations and application to two toxicogenomics studies designed to evaluate treatment effects on the levels of mRNA transcripts involved in mitochondrial function. creating treatment blocks within these sources of variation [9]. Such cases are not suited to permutation methods. We thus demonstrate that this methodology provides a practical approach to testing the significance of the treat- ment effects on gene classes defined by GO terms, and by extension on any other prior categorization of genes into functional subsets. Because many microarray experiments measure treatment effects under complicated design struc- tures and with small sample sizes [9], we used a simula- tion study to determine whether the method gives reasonable results under these conditions. Several statistical methods that combine the analysis of differential gene expression with biological databases have been proposed and implemented in computer pack- ages for a more rapid interpretation of genome-wide expression data [4]. However, most such methods are based on a series of univariate statistical tests and do not properly account for the complex structure of gene inter- actions. The statistical significance of a GO group is com- monly assessed by comparing the number of statistically significant genes in the group to the number expected by chance using Fisher's exact test, which is based on the hypergeometric distribution [5]. Fisher's exact test is used to compare these proportions to assess overrepresentation of significant genes in functional categories. This approach is not amenable to correction for correlations among p-values, since the test inherently assumes exchangeability among genes, an assumption which is not valid under arbitrary or actual correlation structures [6,7]. Specific applications to toxicogenomics studies showed that the methodology has improved specificity in choos- ing significantly altered pathways or functional categories, and may thus assist in the understanding of molecular mechanisms of mitochondrial toxicity in the liver induced by anti-HIV drugs [11,12] and in assessing effects on mito- chondrial function of weight-reducing dietary supple- ments, such as usnic acid [13]. Hotelling's T2 Statistic and permutation methods address the correlation structure among genes. Hotelling's T2 sta- tistic is not applicable when the sample size is smaller than the number of genes in a GO term [8]. Permutation methods, although quite valuable under appropriate con- ditions, are severely compromised by limited numbers of permutable sample pairs. In many cases, the design of microarray studies has a rather complicated structure intended to manage technical variation associated with differences among probes, dyes, and reagent batches by Introduction to simultaneously determine the expression levels of thousands of genes. However, the interpretation of large amounts of microarray gene expression data, and the abil- Introduction The advent of DNA microarray technology has revolution- ized genomic research and medicine because of its ability Page 1 of 9 (page number not for citation purposes) Page 1 of 9 (page number not for citation purposes) BMC Bioinformatics 2008, 9(Suppl 9):S20 http://www.biomedcentral.com/1471-2105/9/S9/S20 Page 2 of 9 (page number not for citation purposes) Measurement of a treatment effect for each gene † † † == −− −− −− (X X) X y (y X ) (y X 1 ′ ′ = − ′ σ 2 1 n r cov( , ) cov( , ) T T s s t t s t s t = ′ ′ ′ ′ = ′ c c(X X) 1c c c(X X) 1c c( † † σ σ σ σ −− −− 1 X X) 1c c(X X) X y c(X X) Xy c(X X) 1c c(X X 1 1 −− −− −− −− ′ ′ ′ ′ = ′ ′ ′ cov( , ) s s s t 1 σ σ ) X y y X(X X) c c(X X) 1c c(X X) X X( 1 1 1 −− −− −− −− ′ ′ ′ = ′ ′ ′ ′ ′ cov( , ) , s s s t s t σ σ σ X X) c 1 −− ′ = = σ σ σ s t s t rs t , , , In many toxicogenomic studies, the significance of a treat- ment effect is tested under the null hypothesis H0 : cβ = 0, where cβ is a pair-wise contrast among treatments. Under the null hypothesis, has a t-distribution T = c c(X X) 1c ˆ ˆ † σ ′ ′ −− In many toxicogenomic studies, the significance of a treat- ment effect is tested under the null hypothesis H0 : cβ = 0, where cβ is a pair-wise contrast among treatments. Under the null hypothesis, has a t-distribution with n - r degrees of freedom, and the p-value to assess the significance of a treatment is calculated from this statistic. T = c c(X X) 1c ˆ ˆ † σ ′ ′ −− In many toxicogenomic studies, the significance of a treat- ment effect is tested under the null hypothesis H0 : cβ = 0, where cβ is a pair-wise contrast among treatments. Under the null hypothesis, has a t-distribution T = c c(X X) 1c ˆ ˆ † σ ′ ′ −− with n - r degrees of freedom, and the p-value to assess the significance of a treatment is calculated from this statistic. with n - r degrees of freedom, and the p-value to assess the significance of a treatment is calculated from this statistic. Test for a gene group A modified meta-analysis method of combining p-values was developed to measure the significance of an overall treatment effect on any group of genes by a one-class t-test [10]. The p-value calculated from the null hypothesis is a random variable with uniform distribution, which can be transformed to a suitable probability distribution. Inverse-normal transformed p-values, zk = Φ-1 (1 - pk) ~N(0, 1), k = 1, , m have a normal distribution and their sum, , is also normally distributed when p-values are independent. Here, pk represents a p- value for a gene in a GO group comprising m genes. The p-value for the sum of inverse-transformed p-values, , gives the overall significance of the treatment effect on the GO group. We refer to this as the naïve estimate because it assumes independence among p-values. z N k k m m = ∑ 1 0 1 ~ ( , ) p zk k m = −( ) = ∑ 1 1 Φ m where rs,t is the s-th row and t-th column element of R. Therefore, is the appropriate p-value which corrects the naïve p-value, . p = −( ) ′ 1 Φ 1 z 1’R1 p = −( ) ′ 1 Φ 1 z m which corrects the naïve p-value, . p = −( ) ′ 1 Φ 1 z m It follows that , where is the average of off-diagonal elements of R. The correction depends only on the average correlation, , and the cor- rection tends to give a reduced significance when > 0. When R is unknown, we estimated the covariance, to provide an average correlation coefficient for the correction. p m m r = − ( ) ⎛ ⎝⎜ ⎞ ⎠⎟ ′ + − 1 1 1 1 Φ 1 z ( ) r r r ˆ ˆ ˆ ) , σ s t s s t t n r = ′ − 1 (y X ) (y X −− −− † † r It follows that , where is p m m r = − ( ) ⎛ ⎝⎜ ⎞ ⎠⎟ ′ + − 1 1 1 1 Φ 1 z ( ) r the average of off-diagonal elements of R. The correction depends only on the average correlation, , and the cor- rection tends to give a reduced significance when > 0. Measurement of a treatment effect for each gene Measurement of a treatment effect for each gene Under a fixed-effects linear model, gene expression data for an arbitrary gene can be written as y = Xβ + ε, where y and ε are n × 1 random vectors, X is a known n × p design matrix of rank r, and β is a p × 1 vector representing unknown parameters. The vector y denotes an observed measurement of expression, suitably transformed, for n biological samples, and ε is an error vector, distributed as Nn (0, σ2 In), where σ2 denotes the unknown within-treat- Page 2 of 9 (page number not for citation purposes) Page 2 of 9 (page number not for citation purposes) http://www.biomedcentral.com/1471-2105/9/S9/S20 http://www.biomedcentral.com/1471-2105/9/S9/S20 BMC Bioinformatics 2008, 9(Suppl 9):S20 mally distributed and its variance is , where 1 is m vector of 1s and R is the correlation matrix of (y1, y2, , ym). Note that var( ) var( ) cov( , ) ′ = + = = > ∑ ∑ 1 z 1’R1 T T T k k m s t s t 2 1 ment variance among samples. The parameters β and σ2 are assumed to be gene-specific. Statistical analyses are applied to one gene at a time, with a common design matrix, X. The unbiased estimators of β and σ2 are mally ˆ ˆ ˆ ˆ). Test for a gene group When R is unknown, we estimated the covariance, to provide an average correlation coefficient for the correction. r r ˆ ˆ ˆ ) , σ s t s s t t n r = ′ − 1 (y X ) (y X −− −− † † r the treatment effect on the GO group. We refer to this as the naïve estimate because it assumes independence among p-values. The correlation structure of p-values is different for a two- sided t-test, which allows gene expression changes in both directions, than for a one-sided situation. A two-tailed test, in which p = 2(1 - Φ(\|z\|)), requires a different correc- tion method, since the correlation among \|zk\|, k = 1, , m differs from a one-sided test in which p = 1 - Φ(z). The null distribution of the summary statistics 1'\|z\| can be gener- ated through Monte Carlo sampling from the null distri- bution of z, MVN(0, cov(z)). When z1, , zn are random samples from MVN(0, cov(z)), the p-value for the observed value, Ψ = 1'\|z\|, is computed as r In reality, genes in a GO group are likely to be functionally related and thus not independent. When the correlation structure among genes is known, we can make a simple adjustment of the naïve estimate. In this case, the test sta- tistics T still has a standard normal distribution and we denote it as , for the k-th gene in a GO Tk k k = c c(X X) 1c ˆ† σ ′ ′ −− group. A common contrast vector, c, is used through all genes since we are measuring same contrast for each gene. The summary statistic for a GO group, is also nor- ′ = = − ( ) = = − = = ∑ ∑ ∑ 1 z z p T k k m k k m k k m 1 1 1 1 1 Φ , where I(A) is an indicator func- p r l l r = > ′ ( ) =∑ 1 1 I Ψ 1 z \| \| , where I(A) is an indicator func- tion which gives 1 if A is true, or 0 otherwise. Test for a gene group Here, cov(z) p r l l r = > ′ ( ) =∑ 1 1 I Ψ 1 z \| \| , where I(A) is an indicator func- p r l l r = > ′ ( ) =∑ 1 1 I Ψ 1 z \| \| tion which gives 1 if A is true, or 0 otherwise. Here, cov(z) Page 3 of 9 (page number not for citation purposes) (page number not for citation purposes) BMC Bioinformatics 2008, 9(Suppl 9):S20 http://www.biomedcentral.com/1471-2105/9/S9/S20 http://www.biomedcentral.com/1471-2105/9/S9/S20 http://www.biomedcentral.com/1471-2105/9/S9/S20 The simulation is conducted under a fairly common set of conditions for microarray studies, comprising three treat- ments with three samples (arrays) per treatment. Samples are generated from N(μi, Σ) for each treatment, i = 1, 2 3,. In this simulation, samples for two treatment have the same average values, μ1 = μ2 = 1, and samples for the other treatment have twice that average value μ3 = 2. P-values for the pair-wise contrast are calculated under the null hypo- thesis, H0 : μ1 = μ2. A GO term is composed of m = 20 genes which have correlation structure generated ran- domly between 0.36 and 0.55. The standard deviations, σi, i = 1, ,m for the genes, which are the diagonal ele- ments of Σ, are generated randomly between 0.01 and 0.25. We iterated this procedure at least 10,000 times to observe the distribution of calculated p-values. has to be estimated from the data. The estimated correla- tion, and its variation, , which has for off-diago- nal elements, are used to estimate cov(z). ˆR R r http://www.biomedcentral.com/1471-2105/9/S9/S20 http://www.biomedcentral.com/1471-2105/9/S9/S20 BMC Bioinformatics 2008, 9(Suppl 9):S20 than 0.05, indicating that the naïve p-values lead to a very high false-discovery rate. The corrected p-values (dashed blue line) fall very near the diagonal line. The corrected p- values thus have more specificity in choosing altered func- tional gene groups than the naïve p-values. sample size n = 25 was conducted as above, to compare several methods for two sided tests. The empirical distri- butions of p-values were generated from 500,000 itera- tions. We looked at small p-values between 0.1 and 0.001 on a log scale. Figure 3 shows that the Hotelling T2 test gives the smallest difference from the uniform distribu- tion. The Hotelling T2 test is applicable when the number of sample is larger than the number of genes in a group. When we have a reasonably correct estimate of R, the method is a little better than the method which uses an approximation of R. Both the method and the method give quite accurate p-values with reference to the p-values from the true correlation matrix, R. ˆR R ˆR R Figure 2 shows the simulation result for a two-sided case. P-values are calculated from the null distribution gener- ated from Monte Carlo samples from MVN(0, cov(z)). Two estimates of cov(z) are used for the sampling. When is used, the empirical distribution of p-values is closer to a uniform distribution than when is used. The esti- mate of the average correlation, , is more robust than that of each element of when sample size is small (n = 3 for each group). R ˆR r ˆR Examples p We present two real-world examples based on data from Mitochip, a mitochondria-specific mouse oligonucleotide microarray which was developed by Dr. Varsha Desai at Figure 3 shows the distribution of p-values for a case with larger samples. The simulation for one-class t-test with Cumulative distribution of p-values for two-sided test case with sample size n = 9 Figure 2 Cumulative distribution of p-values for two-sided test case with sample size n = 9. P-values calculated from random samples based on (dashed blue line) and (dashed green line) give reliable corrections, while the naïve p-value (dashed red line) overstates the significance of the test. R ˆR Cumulative Figure 2 Results Simulation The derivation of the method presented in the previous section is based on the known correlation matrix of the vector of dependent variable Y. When the correlation matrix is not known, we use an estimate of the correlation matrix. In reality, the correlation matrix is always unknown. The proposed method produces an approxi- mately correct p-value for a group of genes. To demon- strate that the method gives not perfect but acceptably correct p-values, we present simulation results in this sec- tion. The validation is done by checking the cumulative distribution of p-values from the proposed methods under the null distribution. The p-values must have a uni- form distribution, which should form a diagonal line in the following figures if p-values are correctly calculated. Figure 1 plots the cumulative distribution of p-values from a one-sided test when the number of samples is n = 9, i.e., 3 groups with 3 samples for each group. The naïve p-val- ues, shown by the red line, clearly deviate from the diago- nal line. Almost 30% of p-values are estimated to be less Cumulative distribution of p-values for one-sided test case with sample size n = 9 Figure 1 Cumulative distribution of p-values for one-sided test case with sample size n = 9. The naïve p-values (dashed red line) deviate from the diagonal line. Almost 30% of p-values are estimated to be less than 0.05. The corrected p-values (dashed blue line) fall very near the diagonal line. Cumulative distribution of p values for one sided test case with sample size n 9 Figure 1 Cumulative distribution of p-values for one-sided test case with sample size n = 9. The naïve p-values (dashed red line) deviate from the diagonal line. Almost 30% of p-values are estimated to be less than 0.05. The corrected p-values (dashed blue line) fall very near the diagonal line. Cumulative Figure 2 Cumulative distribution of p-values for two-sided test case with sample size n = 9. P-values calculated from random samples based on (dashed blue line) and (dashed green line) give reliable corrections, while the naïve p-value (dashed red line) overstates the significance of the test. R ˆR BMC Bioinformatics 2008, 9(Suppl 9):S20 http://www.biomedcentral.com/1471-2105/9/S9/S20 Comparison of two-sided tests with sample size n = 25 Figure 3 Comparison of two-sided tests with sample size n = 25. Hotelling T2 test (dashed blue line) gives the smallest difference from the uniform distribution. When we have enough number of samples to have a reasonably correct estimate of R, the method (dashed red line) is a little better than the method (dashed green line). Both the method and the method give quite accurate p-values compared to the p-values from the true correlation matrix, R (dashed cyan line). ˆR R ˆR R http://www.biomedcentral.com/1471-2105/9/S9/S20 http://www.biomedcentral.com/1471-2105/9/S9/S20 BMC Bioinformatics 2008, 9(Suppl 9):S20 ent across the inner mitochondrial membrane. An important finding of the study is a lack of usnic acid effect on complex V, despite a significant up-regulation of all four complexes of the electron transport chain. Usnic acid is a known uncoupler that is highly lipophilic in both neutral and anionic forms due to its numerous carbonyl groups that absorb the negative charge of the anion by res- onance stabilization. This lipophilicity of usnic acid and the usniate anion allows easy passage of both entities through the mitochondrial membranes by passive diffu- sion into the matrix where it is ionized, releasing a proton into the matrix. The resulting usniate anion can then dif- fuse back into the inter-membrane space where it binds to the proton on the acidic side of the inner membrane to re- form usnic acid which can then diffuse back into the matrix. The resulting cycle causes proton leakage that eventually can dissipate the proton gradient across the inner membrane, disrupting the tight coupling between electron transport and ATP synthesis. This model would explain the absence of gene-expression changes associated with complex V in usnic acid-treated mice, despite the increased electron transport by complexes I – IV. It may also explain the decline in ATP level in spite of increased oxygen consumption. is reported to produce severe adverse effects, and shows more toxicity when AZT is applied in combination with 3TC. Adverse effects are believed to be due to drug- induced mitochondrial disfunction [14]. Oxidative phosphorylation is a key mitochondrial func- tion that requires the electron transport assembly of four protein complexes (I, II, III, IV) to catalyze sequential oxi- dation/reduction reactions, and complex V to generate ATP. Several clinical and animal studies have investigated the effect of nucleoside reverse transcriptase inhibitors (NRTIs), analogs such as AZT, on mitochondrial respira- tory chain complexes. These studies suggest that there is a deficit in one of the components of complexes I and IV in skeletal muscle of children perinatally exposed to antiret- roviral nucleoside analogues [15]. Table 2 shows p-values indicating the significance of treat- ment effects on the GO groups related to oxidative phos- phorylation and apoptosis. The two-sided correction method detects significant effects on genes encoding com- ponents of complexes III and IV, whereas the naïve method finds that genes in all 5 complexes are signifi- cantly affected. http://www.biomedcentral.com/1471-2105/9/S9/S20 This demonstrates that the two-sided cor- rection method is more specific in finding significantly affected gene groups, although of course the "true" answer is not known a priori. Although Fisher's exact test also detects significant alteration in genes of complex IV, this test appears to be detecting a gene group that is different from the other groups, rather than registering treatment effects directly. The one-sided test is not applicable since it seems that gene expression changes in both directions after AZT and 3TC treatment. In Table 3, only the two-sided correction method enables us to explain the function of usnic acid as described above. The one-sided correction method gives p-values similar to those in the two-sided correction method, but this is likely to be due to most of the gene expression changes entailing up-regulation. When the direction of gene expression change due to a treatment is known, then the one-sided correction method is the appropriate choice; it also needs less computation time than the two- sided correction method which employs Monte Carlo sample generation. Usnic acid is a lichen metabolite used as a weight-loss die- tary supplement due to its uncoupling action on mito- chondria. However, its use has been associated with severe liver disorders in many individuals. Animal studies conducted thus far have evaluated effect of usnic acid on mitochondria, primarily by measuring the rate of oxygen consumption and/or ATP generation. Generation of ATP requires tight coupling of electron transport with oxida- tive phosphorylation, maintained through a proton gradi- Compariso Figure 3 Comparison of two-sided tests with sample size n = 25. Hotelling T2 test (dashed blue line) gives the smallest difference from the uniform distribution. When we have enough number of samples to have a reasonably correct estimate of R, the method (dashed red line) is a little better than the method (dashed green line). Both the method and the method give quite accurate p-values compared to the p-values from the true correlation matrix, R (dashed cyan line). ˆR R ˆR R based on a database from Mouse Genome Informatics (MGI, http://www.informatics.jax.org). the National Center for Toxicological Research [11]. Mito- chip measures the levels of mRNA for 542 mitochondrial and nuclear genes associated with mitochondrial structure and function. Each Mitochip includes 9 housekeeping genes and 9 Arabidopsis plant genes to serve as positive and negative control genes, respectively. We considered 317 relevant GO groups related to mitochondrial functions, Table 1 shows the design of an experiment to test the effects of zidovudine (AZT) and lamivudine (3TC) on mouse-liver gene expression. AZT is an anti-HIV drug used to reduce mother-to-child transmission of the virus. AZT Page 6 of 9 (page number not for citation purposes) Table 1: Experimental design for the AZT and 3TC effects on mouse-liver gene expression. Genotype (+/-) (+/+) Treatment Vehicle AZT 240 mg/kg/d AZT+3TC 160+100 mg/kg/d Vehicle AZT+3TC 160+100 mg/kg/d Batch 1 A1 B1 C1 D1 E1 Batch 2 A2 B2 C2 D2 E2 Batch 3 A3 B3 C3 D3 E3 Fifteen samples are collected and assayed for gene expression using the experimental design Table 1: Experimental design for the AZT and 3TC effects on mouse-liver gene expression. Fifteen samples are collected and assayed for gene expression using the experimental design Discussion However, the correction methods that depend only on the average correlation, , are more robust because the estimation of is more robust. ˆR r r For the one-sided test, the correction for the correlation depends only on , the average of off-diagonal elements r Randomization methods that permute class labels can adjust p-values for the correlation structure among genes. Randomization methods choose a summary statistic (e.g. enrichment score (ES) in [16], average z-score in [17]), which reflects the degree to which a set of genes is enriched. When the significance of the summary statistic is measured by permuting class labels, the method pre- serves gene-gene correlations and when applicable, would give similar result to the presented method. Randomiza- tion methods that permute gene labels, such as Fisher's exact test, do not preserve the correlation structure and misrepresent the group's significance. of the correlation matrix. The corrections using or are equivalent for the one-sided test method. Important points in choosing a correlation estimate for the two- sided test are the following; 1) The correlation estimate should be robust for small sample sizes, and 2) The corre- lation estimate should preserve . satisfies these two conditions. R ˆR r R When the direction of gene expression change is pre-spec- ified, the one-sided test is a good choice since it is easy and fast to calculate p-values. However, the two-sided test is the one we have to use in most cases, because it is usually not possible to pre-specify how individual genes will respond to treatment in the exploratory context. When we have a small number of samples to estimate the correla- tion, the method gives a robust result. Since misrep- resents the true correlation, and gives biased p-values, the method works better for larger sample sizes. This is seen in Figure 3, where the method is better than the method. We hesitate to present a specific threshold sample size as sufficient for a converged correlation esti- mate, since it varies with respect to several conditions, such as the number of genes in a group, the variation of the elements of the correlation matrix, etc. The simulation result in Figure 3 shows that both methods give quite accurate p-values compared to the p-values from the true correlation matrix, R. The Hotelling T2 test is the best choice whenever it is applicable. R R ˆR ˆR R Conclusion h We have presented a method to test the significance of expression changes within a group of genes, while consid- ering the correlation structure among genes in each group. This method will enable the rapid detection of microarray evidence indicating altered cell functions or pathways, and will facilitate the interpretation of microarray out- comes. Application of the method to real data shows that it is an improved, practical method to evaluate the effects of treatments on functional classes of genes such as those based on Gene Ontology descriptors. Discussion In many studies that use microarray data, the number of samples is small as in the first example shown above. While the number of samples in the simulations is only 3 for each group, the distribution of corrected p-values Table 2: Effects of AZT and 3TC on oxidative phosphorylation and apoptosis. Table 2: Effects of AZT and 3TC on oxidative phosphorylation and apoptosis. gene group # genes up # P<0.05 Fisher's exact Not corrected One-sided correction Two-sided correction Complex1 29 18 10 0.460 7.41E-6 0.600 0.049 Complex2 3 2 1 0.688 0.017 0.828 0.043 Complex3 7 5 3 0.401 5.3E-5 0.559 0.002 Complex4 13 8 8 0.026 1.25E-07 0.641 0.001 Complex5 14 6 6 0.273 0.0003 0.982 0.022 apoptosis 18 10 7 0.347 2.92E-5 0.592 0.005 P-values calculated from four methods are presented for a comparison. The number of genes, up-regulated genes, and significantly expressed genes in each gene group are presented P-values calculated from four methods are presented for a comparison. The number of genes, up-regulated genes, and significantly expressed genes in each gene group are presented http://www.biomedcentral.com/1471-2105/9/S9/S20 BMC Bioinformatics 2008, 9(Suppl 9):S20 Table 3: Effects of usnic acid on phosphorylation and apoptosis. gene group # genes up # P<0.05 Fisher's exact Not corrected One-sided correction Two-sided correction Complex1 37 31 12 0.517 1.32E-12 0.027 0.022 Complex2 3 2 1 0.680 0.035 0.029 0.014 Complex3 7 5 2 0.704 0.019 0.055 0.042 Complex4 18 17 11 0.008 7.08E-10 0.006 0.003 Complex5 17 13 4 0.838 0.001 0.044 0.051 apoptosis 19 14 11 0.014 8.05E-10 0.008 0.0004 Table 3: Effects of usnic acid on phosphorylation and apoptosis. mutable pairs. In this case, randomization methods are not applicable. Even though randomization methods inherently take into account the correlation structure among genes, they may not be practical when the design of the experiment is complicated and the number of sam- ples per group is small, reducing the numbers of permut- able pairs. mutable pairs. In this case, randomization methods are not applicable. Even though randomization methods inherently take into account the correlation structure among genes, they may not be practical when the design of the experiment is complicated and the number of sam- ples per group is small, reducing the numbers of permut- able pairs. the correlation, , might not be very close to the true cor- relation, R. Competing interests The authors declare that they have no competing interests. Competing interests The authors declare that they have no competing interests. p g The authors declare that they have no competing interests. The authors declare that they have no competing interests. References 1. Benjamini Y, Hochberg Y: Controlling the false discovery rate: a practical and power-ful approach to multiple testing. Journal of the Royal Statistical Society Series B 1995, 57:289-300. 1. Benjamini Y, Hochberg Y: Controlling the false discovery rate: a practical and power-ful approach to multiple testing. Journal of the Royal Statistical Society Series B 1995, 57:289-300. f y y 2. Storey JD, Taylor JE, Siegmund D: Strong control, conservative point estimation, and simultaneous conservative consistency of false discovery rates: A unified approach. Journal of the Royal Statistical Society, Series B 2004, 66:187-205. 2. Storey JD, Taylor JE, Siegmund D: Strong control, conservative point estimation, and simultaneous conservative consistency of false discovery rates: A unified approach. Journal of the Royal Statistical Society, Series B 2004, 66:187-205. y 3. Allison DB, Gadbury GL, Heo M, Fernandez JR, Lee C-K, Prolla TA, Weindruch R: A mixture model approach for the analysis of microarray gene expression data. Computational Statistics and Data Analysis 2002, 39:1-20. y 4. Khatri P, Draghici S: Ontological analysis of gene expression data: current tools, limitations, and open problems. Bioinfor- matics 2005, 21:3587-3595. y 4. Khatri P, Draghici S: Ontological analysis of gene expression data: current tools, limitations, and open problems. Bioinfor- matics 2005, 21:3587-3595. , 5. Draghci S, Khatri P, Martins RP, Ostermeier GC, Krawetz SA: Global functional profiling of gene expression. Genomics 2003, 81:98-104. 5. Draghci S, Khatri P, Martins RP, Ostermeier GC, Krawetz SA: Global functional profiling of gene expression. Genomics 2003, 81:98-104. 6. Tian L, Greenberg SA, Kong SW, Altschuler J, Kohane IS, Park PJ: Dis- covering statistically significant pathways in expression pro- filing studies. PNAS 2005, 102:13544-13549. g 7. Pavlidis P, Qin J, Arango V, Mann JJ, Sibille E: Using the gene ontol- ogy for microarray data mining: a comparison of methods and application to age effects in human prefrontal cortex. Neurochemical Research 2004, 29:1213-1222. 8. Kong SW, Pu WT, Park PJ: A multivariate approach for inte- grating genome-wide expression data and biological knowl- edge. Bioinformatics 2006, 22:2373-80. g f 9. Delongchamp RR, Velasco C, Desai VG, Lee T, Fuscoe JC: Designing toxicogenomics studies that use DNA array technology. Bio- informatics and Biology Insights 2007 in press. gy g p 10. Delongchamp RR, Lee T, Velasco C: A method for computing the overall statistical significance of a treatment effect among a group of genes. BMC Bioinformatics 7(Suppl 2):S11. References 2006 Sep 6 g p g ( pp ) p 11. Desai VG, Lee T, Delongchamp RR, Moland CL, Branham WS, Fuscoe JC, Leakey JEA: Development of mitochondria-specific mouse oligonucleotide microarray and validation of data by real- time PCR. Mitochondrion 2007, 7(5):322-9. ( ) 12. Desai VG, Lee T, Delongchamp RR, Leakey JEA, Lewis SM, Lee F, Moland CL, Branham WS, Fuscoe JC: NRTI-induced expression profile of mitochondrial genes in the mouse liver. Mitochon- drion 2008, 8(2):181-195. ( ) 13. Joseph A, Lee T, Moland CL, Branham WS, Fuscoe JC, Leakey JEA, Allaben W, Lewis SM, Ali AA, Desai VG: Effect of usnic acid on mitochondrial functions as measured by mitochondria-spe- cific oligonucleotide microarray in liver of B6C3F1 mice. Bio- chemical Pharmacology 2008 in press. Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 9 of 9 (page number not for citation purposes) Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Publish with BioMed Central and every scientist can read your work free of charge gy p 14. Cherry CL, Wesselingh SL: Nucleoside analogues and HIV: the combined cost to mitochondria. J Antimicrob Chemother 2003, 51:1091-1093. 15. Barret B, Tardieu M, Rustin P, Lacroix C, Chabrol B, et al.: For the French perinatal cohort study group. Persistent mitochon- drial dysfunction in HIV-1-exposed but uninfected infants: clinical screening in a large prospective cohort. AIDS 2003, 17:1769-1785. 16. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gil- lette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES, Mesirov JP: A knowledge-based approach for interpreting genome-wide expression profiles. Authors' contributions TL conducted the simulations, analyzed the data and wrote the manuscript. VGD conducted the microarray experiments using Mitochip and made the biological interpretations. CV reviewed the literature. RJSR gave val- uable suggestions on the preparation of the manuscript. RRD directed the methodology development, data analy- In Table 1, the distribution of animals from different treat- ment groups (A-E) in three batches (1–3) gives no per- Page 8 of 9 (page number not for citation purposes) Page 8 of 9 (page number not for citation purposes) http://www.biomedcentral.com/1471-2105/9/S9/S20 http://www.biomedcentral.com/1471-2105/9/S9/S20 BMC Bioinformatics 2008, 9(Suppl 9):S20 sis, and manuscript preparation. All authors read and approved the final manuscript. sis, and manuscript preparation. All authors read and approved the final manuscript. Acknowledgements TL was supported by an Oak Ridge Institute of Science and Education (ORISE) fellowship at NCTR. TL was supported by an Oak Ridge Institute of Science and Education (ORISE) fellowship at NCTR. This article has been published as part of BMC Bioinformatics Volume 9 Sup- plement 9, 2008: Proceedings of the Fifth Annual MCBIOS Conference. Sys- tems Biology: Bridging the Omics. The full contents of the supplement are available online at http://www.biomedcentral.com/1471-2105/9?issue=S9 References PNAS 2005, 102:15545-15550. p p 17. Efron B, Tibshirani R: On testing the significance of sets of genes. Tech report 2006 [http://www-stat.stanford.edu/~tibs/GSA/].
https://openalex.org/W2116013663	https://academicjournals.org/journal/AJB/article-full-text-pdf/480E73811327.pdf	English	null	The relevance of biotechnology in the development of functional foods for improved nutritional and health quality in developing countries	African journal of biotechnology	2,003	cc-by	3,844	African Journal of Biotechnology Vol. 2 (12), pp. 631-635, December 2003 Available online at http://www.academicjournals.org/AJB ISSN 1684–5315 © 2003 Academic Journals Minireview The relevance of biotechnology in the development of functional foods for improved nutritional and health quality in developing countries Lorraine L. Niba Lorraine L. Niba Department of Human Nutrition, Foods and Exercise, 319 Wallace Hall, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0430, USA. Phone: 1-(540) 231-8763. E-mail: lniba@vt.edu. Accepted 21 November 2003 The quality of food and food plants can be modified and optimized to meet the nutritional and health needs of at-risk and compromised populations prevalent in most of the developing countries. High rates of malnutrition, infectious disease as well as diet-related diseases such as diabetes and hypertension are prevalent in many developing countries. These are as a result of compromised immune function, inadequate sources of nutritious and quality foods and limited access to healthy and suitable foods. Biotechnology and genetic modification techniques have been proposed and applied for the improvement of the quality of various food crops. These have typically been geared towards increasing yields and pest resistance of cash crops. There is considerably less emphasis however, toward improving quality with regard to fortification or functionality of foods and food plants. Functional foods have nutritional and physiological benefits and are applicable in disease prevention and management. The application of biotechnology techniques for the development of functional food plants with higher levels of bioactive components or increased availability of nutrients would greatly benefit most populations in developing countries and improve the health and nutritional status overall. Key words: Biotechnology, functional foods, food quality, health, developing countries. African Journal of Biotechnology Vol. 2 (12), pp. 631-635, December 2003 Available online at http://www.academicjournals.org/AJB ISSN 1684–5315 © 2003 Academic Journals Key words: Biotechnology, functional foods, food quality, health, developing countries. INTRODUCTION High rates of malnutrition and escalating rates of diet- related diseases such as diabetes and hypertension in developing countries are attributable to various concerns: inadequate sources of dietary protein, foods with high levels of anti-nutritional components and toxicants, a disproportionate amount of highly digestible, high- High rates of malnutrition and escalating rates of diet- related diseases such as diabetes and hypertension in developing countries are attributable to various concerns: inadequate sources of dietary protein, foods with high levels of anti-nutritional components and toxicants, a disproportionate amount of highly digestible, high- glycemic index carbohydrates which constitute the staple foods such as yams, maize and rice, and limited alternatives. In addition, socio-economic factors such as urbanization and migration to urban areas have led to changes in lifestyles to include imported highly processed foods and modification in eating patterns and food habits. Afr. J. Biotechnol. Afr. J. Biotechnol. 632 food applications among hitherto skeptical consumers. Furthermore, it provides insights into the potential for application of biotechnology in developing improved quality and functional foods for human nutrition and health, rather than simply for use in agricultural technology for improved yields and pest resistance. The production of increased levels of beta-carotene (the precursor to vitamin A) in plants is especially important, as its precursor, lycopene has been shown to have physiological chemo-preventive effects with regard to various cancers (Yan and Kerr, 2002). Furthermore, lycopene, commonly found in various carotenoid containing plants such as tomatoes and carrots, is an essential ingredient in maintaining eye health and vision. In many rural areas, there are challenges of protein malnutrition as well as inadequate vitamin and mineral intake. Furthermore, these problems are compounded by inadequate prenatal nutrition which leads to deficiencies in both mothers and children. Nutritional needs of post- partum nursing mothers are often not met, subsequently leading to inadequate nutrition in children. The quality of staple foods and other foods that are commonly consumed in most of these areas could be optimized to improve their quality. While a recent survey of scientific experts around the world indicated that the top biotechnological need for developing countries was technology for diagnosis of infectious disease, other needs such as vaccines and increased nutrient content of food crops as well as combinatorial chemistry for drug discovery also ranked among the top ten needs (Daar et al., 2002). Starch Modification and Glycemic Index Genetic modification efforts geared at increasing the levels, yields of carbohydrates and dry matter of food plants have been largely successful. There is however, potential for modification and improvement of starch and carbohydrate quality. INTRODUCTION It is purported that without biotechnology, the yields and availability of food crops in developing countries, particularly in sub-Saharan Africa would be heavily compromised, leading to inadequate supplies by the year 2025 (Dyson, 1999; Thompson, 2002). Bouis et al. (2003) suggest that there are various ways in which biotechnology and genetic medication can be exploited for consumer benefit, particularly in developing countries: crop productivity, reduction in use of pesticides and improvement in micronutrient content of foods. Modifications that have been targeted and developed by various biotechnology companies include improvement in the oil content and composition of oil seeds such as legumes (Mazur, 2001; Mazur et al., 1999; Uzogara, 2000). Improvement in soybean oil quality includes stabilization of the unsaturated fatty acids by increasing levels of the antioxidant, vitamin E (Yan and Kerr, 2002). These successes indicate a relevant and important role for biotechnology in improving food quality and developing functional foods, particularly those targeted for needy populations in developing countries, such as children and pre-natal women. This review therefore seeks to describe the potential for modifying and optimizing quality of food crops to improve their functionality for improved health and nutritional status, as this would greatly contribute to disease prevention and management. BIOTECHNOLOGY, FOOD QUALITY AND FUNCTIONAL FOODS In certain areas, biotechnology and genetic modification techniques are being optimized for the production and development of healthy foods, and improvement in the levels and activity of biologically active components in food plants (phytochemicals). Biotechnology techniques in developing countries however have mostly been targeted at increasing yields of cash crops. Food crops or the improvement of food quality and functional foods have garnered much less attention. Root and tuber crops such as cassava and yams are typically high in amylopectin with lower ratios of amylose to amylopectin. This yields highly digestible, high- glycemic index foods. These high-glycemic index starches have been associated with diet-related conditions such as type 2 diabetes and insulin resistance. Modification of starch to increase the amylose to amylopectin ratios would be greatly useful in improving glycemic index and physiological responses to the starchy foods that are prevalent in the diet in many developing countries. Root and tuber crops such as cassava and yams are typically high in amylopectin with lower ratios of amylose to amylopectin. This yields highly digestible, high- glycemic index foods. These high-glycemic index starches have been associated with diet-related conditions such as type 2 diabetes and insulin resistance. Modification of starch to increase the amylose to amylopectin ratios would be greatly useful in improving glycemic index and physiological responses to the starchy foods that are prevalent in the diet in many developing countries. Techniques applied in genetic modification include mutation breeding, improved conventional breeding, transgenic modifications, DNA insertion, gene transfer and somatic hybridization, (Bouis et al., 2003; Christou, 1997; Mazur, 2001; Mackay, 1991; Yan and Kerr, 2002). While there has been some hesitation with regard to the acceptability and adoption of biotechnology products in certain developing countries, achievements such as the development of high-vitamin A rice have greatly increased the acceptability of biotechnology for human Techniques applied in genetic modification include mutation breeding, improved conventional breeding, transgenic modifications, DNA insertion, gene transfer and somatic hybridization, (Bouis et al., 2003; Christou, 1997; Mazur, 2001; Mackay, 1991; Yan and Kerr, 2002). Modifications in the starch synthesis pathway have been effected to modify ratios of amylose to amylopectin in potato and cassava starch (Visser et al., 1991, 1997). Improving the glycemic indices of staple starchy food crops such as cassava, yams, potatoes and grains will improve their nutritional quality, and subsequently have an effect on the prevalence of attendant disease conditions. MICRONUTRIENTS – BIOFORTIFICATION Active compounds such as carotenoids from tomatoes, glucosinolates from Brassica vegetables, phytoestrogens from soybean and phenolics and antioxidants from various plants have been shown to protect and prevent against numerous diseases such as cancers and cardiovascular disease. These antioxidants are also associated with slowing the aging process and improving overall health. Phytoestrogens from Dioscorea species such as wild yam (Dioscorea villosa) have also been shown to possess numerous physiological benefits. Some sweet potato varieties have been shown to exhibit anti-diabetic potential. Depleted soils, rudimentary farming and agricultural techniques that characterize farming practices in poorer areas of most developing countries have led to food crops that are low in required micronutrients and vitamins. This in turn has led to high levels of micronutrient deficiencies and subsequently diseases associated with these deficiencies. The effects of vitamin A, vitamin B12 and iron deficiencies are prevalent. There is a need therefore for better quality foods with adequate micronutrient levels. Commercial and industrial level fortification is used in most developed countries for foods and products in which processing has depleted nutrient levels. This is however expensive and not a feasible option in poorer countries. Most of these bioactive components are plant secondary metabolites. Their levels and expression can be improved and optimized through genetic modification of their synthetic pathways. DellaPenna (1999) suggests that levels of these compounds could be increased in target plants by identifying candidate genes which can then be transferred by transgenic approaches to modify the synthetic pathways of interest. Micronutrient content of products can be increased by insertion of genes that produce the desired nutrient. This is applicable for staples that are typically deficient. Introduction of soybean ferritin gene into rice results in increased iron content of rice (Goto et al., 1999). Ferritin being the primary storage protein, its gene coding sequence is inserted into rice plants, resulting in rice with up to three times the amount of iron (Goto et al., 1999). This has also been used in improving iron content of beans (Gura, 1999). This technique can possibly be examined for application in other grain cereals where iron absorption is compromised by the presence of phytates and other anti-nutritional components. Utilization of plants as ‘bio-factories’ for the production of vaccines in developing countries has been examined (DellaPenna, 1999; Thompson, 2002; Uzogara, 2000). BIOTECHNOLOGY, FOOD QUALITY AND FUNCTIONAL FOODS While there has been some hesitation with regard to the acceptability and adoption of biotechnology products in certain developing countries, achievements such as the development of high-vitamin A rice have greatly increased the acceptability of biotechnology for human Niba 633 MICRONUTRIENTS – BIOFORTIFICATION This is attainable by use of transgenic plants for the production of adequate levels of desired vaccines and facilitates distribution to target populations who may otherwise not have access to such vaccines. Bioactive Components soybean. Datta and Bouis (2000), have increased levels of lysine in soybean and canola by transgenic techniques. Application of hybrids to increase yields of beneficial crops from selected strains has also been examined (Mazur et al., 1999). Various commonly consumed food crops such as many vegetables, legumes and grains contain physiologically beneficial bioactive compounds such as anti-oxidant carotenoids, phenolics, alkaloids, glucosinolates and other phytochemicals. Protein Genetic modification aimed at secondary compounds could also be beneficial in improving nutritional quality of foods. Iron absorption for instance is influenced by the presence of phytates and tannins in grains such as maize. Modified maize in which phytates have been reduced by insertion of the degrading enzyme (phytase) gene in the product, result in greater availability of iron, despite no change in overall content of iron (Bouis et al., 2003; Mendoza et al., 1998; Raboy, 1996). The increased iron absorption from low-phytic acid maize is particularly applicable in areas where maize and maize products are staples. The daily protein requirement for optimal health status is estimated at about 0.8 g/kg of body weight (FAO, 1985; Yan and Kerr, 2002). There is a need for substantial amounts of high-quality dietary protein. Unfortunately, in most developing countries, high quality proteins are scarce and of limited availability. Animal protein is particularly limited in areas of drought and in areas of poor soil quality or limited land. In urban areas, meats and animal products such as milk and eggs are mostly imported or even when produced locally, are generally expensive. As such there is a prevalence of protein malnutrition, particularly among dependent groups such as children and prenatal women who have the highest needs for high quality protein. Other micronutrients with potential for application and improvement are zinc, iodine and potassium. BENEFICIAL FOOD CROPS Genetic modification can be applied for improving the availability and quality of protein in food crops. This is primarily by modification of the amino acid composition and amounts, as has been done in some legumes and in Grain legumes contain high levels of macronutrients such as starch, protein and fiber. In addition, legumes such as Afr. J. Biotechnol. Afr. J. Biotechnol. 634 Phaseolus beans and Vigna species, chick-pea and other such varieties contribute a significant proportion of the dietary energy in many tropical areas. components such as cyanogenic glycosides. Levels of anti-nutritional factors such as haemagglutinins and tannins are depleted by fermentation (Lorri, 1993; Mbugua, 1992). Legume consumption and utilization however is a distant second to grain cereals, roots and tuber crop consumption in most developing countries. Consumption of legumes is on the decline mainly due to lack of interest, limited research and development investment and lack of their exploitation as international cash crops. In addition, various legumes contain anti-nutritional factors such as cyanogenic glycosides and polyphenols. Furthermore, the long cooking times and the tedious cooking procedures required for preparing some legumes also decrease their utility and exploitation. Fermentation markedly improves nutritional and health quality in food products. Protein content of certain grain foods is increased by fermentation. Furthermore, fermentation and staling results in modification of starches to form resistant starches which are associated with protection against colon cancer and gastrointestinal disease (Ahmed et al., 2000). In South Africa, for instance, a disparity in the rates of colon cancer between whites and black populations has been attributed to the consumption of fermented maize porridge which is high in complex and resistant starches (Ahmed et al., 2000). The macronutrients in legumes contribute to enhanced food quality. Legume proteins contain sulfur-containing amino acids that are typically deficient in grain cereals. Legume starches tend to be B-type starches that are less digestible, have lower glycemic index and result in lower corresponding increases in post-prandial blood glucose levels than most grain and root starches. ( ) Improvement and optimization of fermentation and bio- processing in developing countries is applicable in improving quality and functionality of foods. This would simply be geared at improving already existing technologies as well as incorporating new technologies. FERMENTATION AND BIO-PROCESSING Fermented foods constitute an important part of the diet and food ingredients in many developing countries. These are primarily fermented grains and root crops, as well as curdled milks and milk products. BENEFICIAL FOOD CROPS It has been suggested that even though hitherto most fermentation processes used in developing countries do not use innocula or extrinsic cultures, these processes could be improved by using starter cultures, and also by backslopping, which entails application of brine from prior fermentation cycles (Holzapfel, 2002). These modifications could be effected by recombinant DNA technology, and have the advantage of being effective in decreasing the length of fermentation processes, and improving the overall quality and efficacy of fermentation (Holzapfel, 2002). Modifications to improve legume quality would include increasing the yields, increasing protein content and augmenting levels of beneficial and desired amino acids, increasing starch content, reducing anti-nutritional and toxic factors such as cyanogenic glycosides, and development of easy-cooking varieties. While genetic modification and bioengineering techniques have been applied to various crops, such techniques have only been marginally applied to legumes. However, various procedures have been suggested for modification of legumes. These include development of transgenic plants by use of Agrobacterium-mediated gene transfer and particle bombardment techniques (Christou, 1997). These techniques are proposed to be applicable in increasing levels of vitamins such as Vitamin A and B12 which are deficient in Phaseolus species (Christou, 1997). Additional applications include pest resistance in legumes, particularly aflatoxin in peanuts (Christou, 1997). Controlled fermentation processes can be applied for the production of physiologically beneficial components, for increasing amounts of protein and for increasing the levels of complex carbohydrates and other fermentation products. Protein content and quality in grain cereals is improved via fermentation as a result of the fact that trypsin inhibitors are depleted, increasing the digestibility of the proteins as well as availability of various amino acids (Holzapfel, 2002; Lorri, 1993; Mbugua, 1992) Some fermentative bacteria such as Lactobacillus produce beneficial metabolic products which have anti- microbial properties. These include organic acids, hydrogen peroxide and various bacteriocins such as nicin (Holzapfel, 2002). This is an added benefit of optimized fermentation and bio-processing. PROBIOTIC MICRO-ORGANISMS Fermented grain products are processed into staples such as gruels and dumplings, which serve as baby foods as well as for adult consumption. Fermented cassava and yam foods are staples in many parts of Africa. Fermentation is typically effected by bacteria and yeasts and facilitates preservation and storage. In addition, it serves as a means to reduce detrimental Probiotic micro-organisms have great physiological import as their fermentation of complex, non-digestible carbohydrates in the colon produces short chain fatty acids that have numerous benefits. These short chain fatty acids are substrates for colonic cell metabolism and Niba 635 Crittenden R, Laitila A, Forssell P, Matto J, Saarela M, Mattila- Sandholm T, Myllarinen P (2001). Adhesion of bifidobacteria to granular starch and its implications in probiotic technologies. Appl. Environ. Microbiol. 67:3469-3475. improve their resilience. In addition, the lowered pH of the colon acts to protect against proliferation of other bacteria. These short chain fatty acids are reported to play a role in prevention against colon cancer and other diseases. Most important, probiotic bacteria such as Bifidobacteria and various Lactobacilli improve immune status of the colon, and are involved in preventing and depleting pathogenic and infectious bacteria in the gastrointestinal tract such as Helicobacter pylori (Hamilton-Miller, 2003). Furthermore, probiotic bacteria are effective against lactose intolerance and diarrhea (Scheinbach, 1998). Daar AS, Thorsteinsdottir H, Martin DK, Smith AC, Nast S, Singer PA (2002). Top ten biotechnologies for improving health in developing countries. Nat. Genet. 32:229-232. Datta S, Bouis HE (2000). Application of biotechnology to improving the nutritional quality of rice. Food Nutr. Bull. 21:451-456. DellaPenna D (1999). Nutritional genomics: manipulating plant micronutrient to improve human health. Science 285:375-379. Dyson T (1999). World food trends and prospects for 2025. Proc. Natl. Acad. Sci. 96:5929-5936. FAO (1985). Food and Agricultural Organization, World Health Organization, United Nations University. Energy and protein requirement. Report of a Joint FAO/WHO/UNU Expert Consultation. World Health Organization Technical Report Series 724. Geneve, Switzerland: World health Organization. Most commercial probiotics are produced by fermentation (Hansen, 2002). In addition, some strains of Bifidobacteria have been shown to be adherent to starches from maize, potato and oats (Crittenden et al., 2001). Improving cultivation and fermentation processes for the production of probiotic and beneficial bacteria is a viable avenue for dealing with the challenges of diseases such as diarrhea and the compromised status of gastrointestinal health in many developing countries. CONCLUSION Mackay GR (1991). Genetic Manipulation of the potato for improved processing. J Sci Food Agric. 57:449-458. Inadequate nutrition, sub-par quality foods and limited food processing capabilities have led to compromised, sub-par health status and a prevalence of diet-related diseases in many developing countries, most especially among children, prenatal and post-partum women. There is clear potential for the application of biotechnology and genetic modification as tools to combat these challenges and improve the situation of at-risk populations. While there is concerted and dedicated focus on combating more apparent challenges such as infectious disease and producing enough food to feed the growing populations, the quality of the foods produced ought to be an important consideration. The feasibility and cost of applying modern and emerging genetic modification technologies in these areas is certainly a daunting task, as sometimes the primary need is often just survival. However, as scientists and policy makers make progress towards alleviating these debilitating conditions, consideration should also be given to the possibility and potential for augmenting food quality and developing functional foods for improved nutritional and health status. Mazur BJ (2001). Developing transgenic grains with improved oils, proteins and carbohydrates. Novartis Found Symp. 236: 233-239. discussion 240-1. Mazur B, Krebbers E, Tingey S (1999). Gene discovery and product development for grain quality traits. Science 285:372-377. Mendoza C, Viteri FE, Lonnerdal B, Young KA, Raboy V, Brown KH (1998). Effect of genetically modified, low-phytic maize on absorption of iron from tortillas. Am. J. Clin. Nutr. 68:1123-1127. Mbugua SK, Ahrens RH, Kigutha HN, Subrahmanian V (1992). Effect of fermentation, malted flour treatment and drum drying on nutritional quality of uji. Ecol. Food Nutr. 28:271-277. Raboy V (1996). Cereal low phytic mutants: a “global” approach to improving mineral nutritional quality. Micronutrients Agric. 2:15-16. Scheinbach S (1998). Probiotics: Functionality and commercial status. Biotechnol Adv. 16:581-608. Thompson JA (2002). Research needs to improve agricultural productivity and food quality, with emphasis on biotechnology. J. Nutr. 132:3441S - 3442S. Uzogara SG (2000). The impact of genetic modification of human foods in the 21 st century: a review. Biotechnol. Adv. 18:179-206. Visser R, Somhurst I, Kuipers G, Ruys N, Feenstra W, Jacobsen E (1991). Inhibition of the expression of granule-bound starch synthase in potato by anti-sense constructs. Mol. Gen. Genet. 225:289-296. y Visser R, Suurs L, Steeneken P, Jacobsen E (1997). Some physicochemical properties of amylose-free potato starch. Starch/Staerke 49: 443-448. Yan L, Kerr PS (2002). PROBIOTIC MICRO-ORGANISMS Goto F, Yoshihara T, Shigemoto N, Toki S, Takaiwa F (1999). Iron fortification of rice seed by the soybean ferritin gene. Nat. Biotech. 17:282-286. (1999). New genes boost rice nutrients. Science 285:994 Gura T (1999). New genes boost rice nutrients. Science 285:994-995. Hamilton-Miller JM (2003). The role of probiotics in the treatment and prevention of Helicobacter pylori infection. Int. J. Antimicrob. Agents 22:360-366. Hansen EB (2002). Commerical bacterial starter cultures for fermented foods of the future. Int. J. Food Microbiol. 78:119-131. Holzapfel WH (2002). Appropriate starter culture technologies for small- scale fermentation in developing countries. Int. J. Food Microbiol. 75:197-212. CONCLUSION Genetically engineered crops: their potential use for improvement of human nutrition. Nutr. Rev. 60:135-141. Christou P (1997). Biotechnology applied to grain legumes. Field Crops Res. 53:83-97. Bouis HE, Chassy BM, Ochanda JO (2003). Genetically modified food crops and their contribution to human nutrition and food quality. Trends Food. Sci. Tech. 14:191-209. Ahmed R, Segal I, Hassan H (2000). Fermentation of dietary starch in humans. Am. J. Gastroenterol. 95: 1017-1020. REFERENCES Ahmed R, Segal I, Hassan H (2000). Fermentation of dietary starch in humans. Am. J. Gastroenterol. 95: 1017-1020. Bouis HE, Chassy BM, Ochanda JO (2003). Genetically modified food crops and their contribution to human nutrition and food quality. Trends Food. Sci. Tech. 14:191-209. Christou P (1997). Biotechnology applied to grain legumes. Field Crops Res. 53:83-97.
https://openalex.org/W2592776773	http://www.mrforum.com/wp-content/uploads/open_access/9781945291173/26.pdf	English	null	The Residual Stress Effect on the Shape Memory Polymers	Materials research proceedings	2,016	cc-by	4,612	Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Published under license by Materials Research Forum LLC. The Residual Stress Effect on the Shape Memory Polymers A. Kallel1,a, M. Lamraoui2,b, J. Fitoussi2,c, A. Tcharkhtchi2,d, 1 Léonard de Vinci Pôle universitaire, Research center, Paris la Défense, France 2 Laboratoire PIMM, UMR 8006 CNRS, Arts et Métiers ParisTech, Paris 75013, France aachraf.kallel@devinci.fr, bmenad.lamraoui@ensam.eu, cjoseph.fitoussi@ensam.eu, dabbas.tcharkhtchi@ensam.eu Keywords: Shape Memory Polymers, Residual Stress Effect, Driving Force, Blend PCL /SBS Abstract. The current paper presents an experimental study of the residual stress role to recover polymer blend to its original position after deformation. In this study, we use a polymer blend of 40% PolyCaproLactone (PCL) and 60% Styrene-Butadiene-Styrene (SBS). The Shape Memory Polymer (SMP) is able to storing a permanent macroscopic after passing through a temporary one; then under an external stimulus; they can recover their initial shapes. The recovery rate depends on the mechanical property of the mixture and also of the number of the loading cycles. Indeed, one multiplying the number of mechanical loading cycle; generally tensile test; the recovery rate does not change in a linear manner and therefore the residual stress is not added in integer from one cycle to another. In this work we studied the mechanism of the shape as well as property memory effect of the blend under study during a shape memory cycle, to establish a relationship between the “shape memory effect” and the “properties memory effect”. Finally, we measure the “Driving Force” responsible for the shape memory effect by an original method. This study allows establishing a relationship between the rate of recovery and residual stresses introduced into the polymer during its deformation. Residual Stresses 2016: ICRS-10 Residual Stresses 2016: ICRS-10 Materials Research Forum LLC Residual Stresses 2016: ICRS-10 Materials Research Forum LLC Materials Research Proceedings 2 (2016) 151-156 doi: http://dx.doi.org/10.21741/9781945291173-26 56 doi: http://dx.doi.org/10.21741/9781945291173-26 Materials Research Proceedings 2 (2016) 151-156 doi: http://dx.doi.org/10.21741/9781945291173-26 The Residual Stress Effect on the Shape Memory Polymers A. Kallel1,a, M. Lamraoui2,b, J. Fitoussi2,c, A. Tcharkhtchi2,d, 1 Léonard de Vinci Pôle universitaire, Research center, Paris la Défense, France 2 Laboratoire PIMM, UMR 8006 CNRS, Arts et Métiers ParisTech, Paris 75013, France aachraf.kallel@devinci.fr, bmenad.lamraoui@ensam.eu, cjoseph.fitoussi@ensam.eu, dabbas.tcharkhtchi@ensam.eu Introduction Shape memory polymers (SMPs) have the capacity of changing their shapes from a temporary shape to a permanent shape upon an external stimulus. The stimulus could be provided by thermal [1, 2], magnetic or electric sources [3]. In the case of thermal stimulus, the cycle process consists of first a deformation at high temperature, relaxation, fixing and finally recovery [4-6]. SMPs are useful in many domains as medical and biological devices [8, 9], smart textile [10, 11], tubing and actuating materials and packaging components [12-14]. The capacity of SMPs to change shape is generally explained by the presence of at least two different phases with different mechanical properties (hard and soft). Recently different new concepts regarding to certain aspects related to shape memory effect were shown [4-6]. We can consider as concept the Property Memory Effect (PME) in which stress-strain tensile tests and both mono and multi-frequency DMA (Dynamic Mechanical Analyzer) tests on virgin and 100% recovered samples of PU (rubber) revealed that the polymer at the end of the shape memory tests regains 100% of its initial form without regaining some of its physical properties like the glass transition temperature, the tensile modulus, the heat expansion coefficient and the free volume fraction. After recovery tests on a stretched sample, the majority of the polymers regain only a certain percentage of their initial shape. The Degree of Shape Memory Effect (DSME) is defined as the percentage of the residual shape. This denotes the Partial Shape Memory Effect (PSME) concept. The successive cyclic tensile tests on partial shape memory polymers (PSMP) show that the DSME increases with the increase of the number of cycles. This test also showed the shape capacity of polymer increases by the number of cycles. This new concept of shape memory effect is 151 Residual Stresses 2016: ICRS-10 Materials Research Forum LLC Materials Research Proceedings 2 (2016) 151-156 doi: http://dx.doi.org/10.21741/9781945291173-26 Materials Research Forum LLC doi: http://dx.doi.org/10.21741/9781945291173-26 Materials Research Forum LLC doi: http://dx.doi.org/10.21741/9781945291173-26 of very important value. We can in a legitimate way suppose that a polymer without shape memory effect can be transformed into a shape memory polymer. Another new concept concerns the driving force responsible of recovery phenomenon. It has been shown that during shape memory cycling tests, and during fixing test the polymer is cooled down rapidly in order to fix the shape obtained at the end of the test. Experimental Materials. A mixture of polycaprolactone (PCL) (40%) and Styrene-Butadiene-Styrene (SBS) (60%) is used. This blend shows almost a total shape memory effect. We choose this percentage because it gives a good combination of shape recovery ratio and shape fixing ratio [18]. The PCL that was used for this study is PCL CAPA 860 supplied by Perstrop Limited UK with a density of 1. 14 𝑔/𝑐𝑐3 of density and its Tg and Tm are respectively -50°C and 60°C. The used Styrene- Butadiene-Styrene (SBS); provided by the company Sinopec Group, is an amorphous copolymer (Tg =-80°C) between the polybutadiene rubber (PB) and thermoplastic polystyrene (PS). ) p y ( ) p p y y ( ) Determinations of melting point and glass transition temperature. A TA instruments DSC- Q10 V9.0 Build 275 type Differential Scanning Calorimetry has been used to determine the transition temperature of polymers. Specimens have been cooled with nitrogen from room temperature to -80 °C (Minimal temperature that the machine can achieve) and then heated until 80 °C at a constant rate of 3 °C / min. -0,2 -0,15 -0,1 -0,05 0 -80 -60 -40 -20 0 20 40 60 80 Heat flow [W/g] Temperature [°C] Fig. 1: DSC Results of: (a) pure PCL, (b) pure SBS and (c) PCL (40%) /SBS (60%) blend. Pure PCL presents a melting point of 59,2 °C [18]. Pure SBS is an amorphous polymer without any melting point. The blend shows a melting point of 57,6 °C which is almost the same as the pure PCL (Fig. 1). In order to determine the glass transition temperature, dynamical mechanical thermal analysis (DMTA) tests has been carried out using a DMA Q800 TA Instruments. A 1Hz frequency has been performed between -100°C to 0°C with a heating rate of 3°C/min, Fig. 1: DSC Results of: (a) pure PCL, (b) pure SBS and (c) PCL (40%) /SBS (60%) blend. These curves in Fig.2 show that the transition temperature of the SBS and PCL are -82°C and - 42°C, respectively. These results reinforce the fact that the SBS and the PCL are not miscible because their Tg remain the same in the blend. Fig. 3 showed the MEB images of a virgin simple (right image) and a simple after one SM cycle (left image). The bright region was PCL phase and the dark region was SBS phase. Introduction In this period the residual stresses will be locked in the structure. This residual stress is indeed the origin of driving force responsible for the shape memory effect [4-6]. During recovery, at a temperature higher than the glass transition of soft segments, these stresses are released and the SMP regains its initial shape. In the present work, in order to study the origin of this driving force "Driving force", the relationship between the recovery rate and the residual stresses, introduced into the polymer during its deformation, has been determined. Experimental 3: The MEB images of the cross-section of the blend SBS 60% and PCL 40%. Tensile and shape memory test. The stress-strain tests have been done by a uniaxial-tensile loading Instron 5566-type machine with a load cell of 1 kN at room temperature, according to the standard NF ISO 6239 [17]. Tensile and shape memory test. The stress-strain tests have been done by a uniaxial-tensile loading Instron 5566-type machine with a load cell of 1 kN at room temperature, according to the standard NF ISO 6239 [17]. Shape memory test was carried according to the following steps: (1) heating the sample up to 50°C and holding for 10 minutes; (2) extending to the strain of 100% with a speed of 6 mm/min (tensile test) (3) fixing this deformation by cooling down the sample to the room temperature using a fan for 10 min (4) unloading the sample to zero stress and then recording the strain (5) heating up the sample to above the glass transition temperature of the blend and then recording the strain (recovery test). During this step, the length of the sample decreases and thanks to the shape memory effect, the sample recovers all its substantial original shape. It is important to indicate that the recovery test cannot practically be performed isothermally because during heating, the oven from room temperature up to 50°C and before reaching this temperature, the sample begins its deformation for recovering its initial shape. So recovery test was carried out in ramp (constant temperature rate) condition. For multi-cycle tests, successive shape memory cycles were performed on the same sample. At the end of the first recovery cycle test, the sample is subjected to another shape memory cycle, i.e. tensile, fixing and recovery. This cycle was repeated a several times. Experimental Shape memory test was carried according to the following steps: (1) heating the sample up to 50°C and holding for 10 minutes; (2) extending to the strain of 100% with a speed of 6 mm/min (tensile test) (3) fixing this deformation by cooling down the sample to the room temperature using a fan for 10 min (4) unloading the sample to zero stress and then recording the strain (5) heating up the sample to above the glass transition temperature of the blend and then recording the strain (recovery test). During this step, the length of the sample decreases and thanks to the shape memory effect, the sample recovers all its substantial original shape. It is important to indicate that the recovery test cannot practically be performed isothermally because during heating, the oven from room temperature up to 50°C and before reaching this temperature, the sample begins its deformation for recovering its initial shape. So recovery test was carried out in ramp (constant temperature rate) condition. For multi-cycle tests, successive shape memory cycles were performed on the same sample. At the end of the first recovery cycle test, the sample is subjected to another shape memory cycle, i.e. tensile, fixing and recovery. This cycle was repeated a several times. Results and discussion Thermal effects on the mechanical and the shape memory properties. The results of tensile tests on the SBS/PCL blend are shown in Figure 5 and table 2. These results clearly show the influence of temperature on mechanical properties. The stress decreases with increasing temperature. We can note that the thresholds which separate elastic and plastic areas are at 8% of strain (ε). The strain energy can be stored in the polymer structure, which subsequently will be released under the effect of thermal stimulation This stored energy is the source of the driving force that will bring the 0 200 400 600 800 1000 1200 -95 -75 -55 -35 -15 5 25 45 65 Storage Modulus (Mpa) Temperature(°C) Fig. 2: DMTA analysis of virgin sample 0 200 400 600 800 1000 1200 -95 -75 -55 -35 -15 5 25 45 65 Storage Modulus (Mpa) Temperature(°C) Fig. 2: DMTA an Fig. 2: DMTA analysis of virgin sample e blend SBS 60% and PCL 40%. Fig. 3: The MEB images of the cross-section of the blend SBS 60% and PCL 40%. . Experimental That shows the intrinsic immiscibility between the SBS, the soft phase, and the PCL the hard phase. This confirms DMA results showing two different transition temperatures: the lowest corresponds to the soft phase (SBS) and the highest corresponds to the hard phase (PCL). 152 Residual Stresses 2016: ICRS-10 Materials Research Forum LLC Residual Stresses 2016: ICRS-10 Materials Research Forum LLC Materials Research Proceedings 2 (2016) 151-156 doi: http://dx.doi.org/10.21741/9781945291173-26 0 200 400 600 800 1000 1200 -95 -75 -55 -35 -15 5 25 45 65 Storage Modulus (Mpa) Temperature(°C) Fig. 2: DMTA analysis of virgin sample Fig. 3: The MEB images of the cross-section of the blend SBS 60% and PCL 40%. Tensile and shape memory test. The stress-strain tests have been done by a uniaxial-tensile loading Instron 5566-type machine with a load cell of 1 kN at room temperature, according to the standard NF ISO 6239 [17]. Shape memory test was carried according to the following steps: (1) heating the sample up to 50°C and holding for 10 minutes; (2) extending to the strain of 100% with a speed of 6 mm/min (tensile test) (3) fixing this deformation by cooling down the sample to the room temperature using a fan for 10 min (4) unloading the sample to zero stress and then recording the strain (5) heating up the sample to above the glass transition temperature of the blend and then recording the strain (recovery doi: http://dx.doi.org/10.21741/9781945291173-26 Materials Research Proceedings 2 (2016) 151-156 0 200 400 600 800 1000 1200 -95 -75 -55 -35 -15 5 25 45 65 Storage Modulus (Mpa) Temperature(°C) Fig. 2: DMTA analysis of virgin sample Fig. 3: The MEB images of the cross-section of the blend SBS 60% and PCL 40%. Tensile and shape memory test. The stress-strain tests have been done by a uniaxial-tensile loading Instron 5566-type machine with a load cell of 1 kN at room temperature, according to the standard NF ISO 6239 [17]. Results and discussion Thermal effects on the mechanical and the shape memory properties. The results of tensile tests on the SBS/PCL blend are shown in Figure 5 and table 2. These results clearly show the influence of temperature on mechanical properties. The stress decreases with increasing temperature. We can note that the thresholds which separate elastic and plastic areas are at 8% of strain (ε). The strain energy can be stored in the polymer structure, which subsequently will be released under the effect of thermal stimulation. This stored energy is the source of the driving force that will bring the sample to its original shape. 153 Residual Stresses 2016: ICRS-10 Materials Research Forum LL Materials Research Proceedings 2 (2016) 151-156 doi: http://dx.doi.org/10.21741/9781945291173-2 0 1 2 3 4 5 6 0 20 40 60 80 100 Stress [Mpa] Strain [%] 1 2 3 4 Fig. 5: Tensile tests of SBS / PCL blend at different temperatures: (1) 25°C, (2) 35°C, (3) 50°C, (4) 60°C Table 1: Young modulus values and stress at 100% of strain at different temperature T(°C) Modulus [MPa] Stress at 100% strain [MPa] 25 174 5.7 35 103 4.7 50 66 3.1 60 7 0.3 Fig. 5 shows that the blend is less rigid at high temperature. The curve’s initial slope (Young Modulus) as well as the sample’s stiffness is inversely proportional to the temperature. 50°C will be used as a reference temperature for the entire tensile test. The recovery test is carried out in a thermal chamber for a range of temperature after holding at 100% of deformation. The recovery rate and the fixing rate are respectively: Residual Stresses 2016: ICRS-10 Materials Research Forum LLC doi: http://dx.doi.org/10.21741/9781945291173-26 Materials Research Proceedings 2 (2016) 151-156 0 1 2 3 4 5 6 0 20 40 60 80 100 Stress [Mpa] Strain [%] 1 2 3 4 Fig. 5: Tensile tests of SBS / PCL blend at different temperatures: (1) 25°C, (2) 35°C, (3) 50°C, (4) 60°C Table 1: Young modulus values and stress at 100% of strain at different temperature Fig. 5: Tensile tests of SBS / PCL blend at different temperatures: (1) 25°C, (2) 35°C, (3) 50°C, (4) 60°C Fig. 5 shows that the blend is less rigid at high temperature. The curve’s initial slope (Young Modulus) as well as the sample’s stiffness is inversely proportional to the temperature. Results and discussion 50°C will be used as a reference temperature for the entire tensile test. The recovery test is carried out in a thermal chamber for a range of temperature after holding at 100% of deformation. The recovery rate and the fixing rate are respectively: 𝑅𝑟(%) = 𝐿𝑢 −𝐿𝑓 𝐿𝑢 −𝐿𝑖 ∗100 (1), and 𝑅𝑓(%) = 𝐿𝑢 𝐿𝑚∗100 (2) (2) Where Li is the initial sample length, Lm the length after 100% of deformation, Lu the length after tensile test without stress (jaws are release) and Lf is the t length after recovery. Where Li is the initial sample length, Lm the length after 100% of deformation, Lu the length after tensile test without stress (jaws are release) and Lf is the t length after recovery. To calculate the recovery rate, w used Lu, the length after tensile tes without stress, and not Lm, the lengt after 100% of deformation, becaus when we release the jaws th sample’s length changes. The fixin rate is not equal to 100%. s Fig. 6: Recovery rate at different temperature: (1) 75°C, (2) 65°C, (3) 55°C 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 Recovery (%) Time [Min] 1 2 3 Table 2: Recovery tests at different temperature Temperature [°C] Rr [%] Rf [%] 75 97,2 92,9 65 91,1 94,1 55 71,9 93,4 Table 2: Recovery tests at different temperature Temperature [°C] Rr [%] Rf [%] 75 97,2 92,9 65 91,1 94,1 55 71,9 93,4 Fig. 6: Recovery rate at different temperature: (1) 75°C, (2) 65°C, (3) 55°C Molecular mobility increases by rising the temperature. For our blend, 100% recovery is reached in about 20 min with 75 °C while for 55 °C the polymer reached only 71,9% of recovery. For this blend, the SBS chains, the soft phase, ensure the recovery. Multi-cycle effects on mechanical and shape memory properties. Multi-cycle tests were performed to study their effects on the mechanical and shape memory properties. For these assessments, we will use 50°C for tensile test and 55°C for recovery test. The residual stress stored in the sample after each tensile test, need a thermal stimulus to be released. The amount of the internal energy stored during the cycle is more important when the tensile test is made at -30°C than at 23°C. Results and discussion Therefore, for sample which the tensile test is made at - 30°C, the recovery starts at low temperature and can reach 55% at the ambient temperature. The residual stress stored in the sample after each tensile test, need a thermal stimulus to be released. The amount of the internal energy stored during the cycle is more important when the tensile test is made at -30°C than at 23°C. Therefore, for sample which the tensile test is made at - 30°C, the recovery starts at low temperature and can reach 55% at the ambient temperature. 154 154 Residual Stresses 2016: ICRS-10 Materials Research Forum LLC doi: http://dx.doi.org/10.21741/9781945291173-26 Materials Research Proceedings 2 (2016) 151-156 0 0,5 1 1,5 2 2,5 3 3,5 0 20 40 60 80 100 Stress [MPa] Strain[%] 1_cycle 2_cycles 3_cycles 4_cycles 5_cycles 6_cycles 7_cycles 8_cycles 9_cycles 10_cycles Fig. 7: Tensile test for different cycles 0 20 40 60 80 100 0 5 10 15 20 Recovery [%] Time [min] 1_cycle 2_cycles 3_cycles 4_cycles 5_cycles 6_cycles 7_cycles 8_cycles 9_cycles 10_cycles Fig. 8: Recovery rate for different cycles. Table 3: Maximal stress and Modulus for each cycle Strain cycles Stress [MPa] Modulus [MPa] 1 3,1 65,6 2 2,7 46,8 3 2,7 20,5 4 2,6 17,6 5 2,4 17,6 6 2,4 15,2 7 2,3 14,9 8 2,3 13,7 9 2,1 12,3 10 2,1 12,1 0 0,5 1 1,5 2 2,5 3 3,5 0 20 40 60 80 100 Stress [MPa] Strain[%] 1_cycle 2_cycles 3_cycles 4_cycles 5_cycles 6_cycles 7_cycles 8_cycles 9_cycles 10_cycles Fig. 7: Tensile test for different cycles Fig. 7: Tensile test for different cycles Fig. 7: Tensile test for different cycles 0 20 40 60 80 100 0 5 10 15 20 Recovery [%] Time [min] 1_cycle 2_cycles 3_cycles 4_cycles 5_cycles 6_cycles 7_cycles 8_cycles 9_cycles 10_cycles Fig. 8: Recovery rate for different cycles. Table 4: Fixing rate and Recovery rate at different cycles Cycles Rr [%] Rf [%] 1 89,7 94,6 2 76,9 94 3 83,5 93,7 4 90,7 92,2 5 92,2 88,8 6 92 89,1 7 92,4 88,1 8 93,1 87,8 9 93,7 87,7 10 92 88,1 Fig. 8: Recovery rate for different cycles. Tensile tests were carried out on a sample after shape memory cycles. Results are given in figure 8. We can notice that the specimen’s stress decreases from 3.1 Mpa (cycle1) to 2.1 Mpa after 10 shape memory cycles. The Young modulus undergoes a significant decrease because it passes from 65.6 Mpa in the first cycle to 12.1 Mpa in the tenth cycle. Conclusion After a shape memory cycle; the soft material, more sensitive to the deformation, is going to store some strain energy which will be afterward a source for the driving force to return to the initial shape. Whereas the hard phase, during the tensile test, resist to the deformation of the soft chains and this resistance is going to create residual stress in the hard phase which will be released by an external stimulus (heating). The direct measure of the residual stress by diffraction for our example will be very difficult to see impossible because the mixture is not crystalline and does not present any arrays as the case of the steel or other metals. We remark that the blend regains 90% of its initial shape after the first memory test and that the recovery rate varies in a random way until the fifth cycle from which it stabilizes around 92 % until tenth cycles. The fixing rate decrease from 94% in the first cycle to 88% in the fifth one. Then it keeps this rate until the tenth cycle. We can notice that even when the sample regains almost its initial shape after each cycle, it does not have the same mechanicals properties. We can notice that the sample does not remain the same viscoelastic properties even when it regains almost its initial shape after each cycle. Residual Stresses 2016: ICRS-10 Materials Research Forum LLC Materials Research Proceedings 2 (2016) 151-156 doi: http://dx.doi.org/10.21741/9781945291173-26 Materials Research Forum LLC doi: http://dx.doi.org/10.21741/9781945291173-26 References [1] Lendlein A, Kelch S. Shape-memory polymers. Angew Chem Int Ed 2002;41:2034–57. http://dx.doi.org/10.1002/1521-3773(20020617)41:12<2034::AID-ANIE2034>3.0.CO;2-M [2] Behl M, Lendlein A. Shape-memory polymers. Materials Today 2007;10:20–8. http://dx.doi.org/10.1016/S1369-7021(07)70047-0 [1] Lendlein A, Kelch S. Shape-memory polymers. Angew Chem Int Ed 2002;41:2034–57. http://dx.doi.org/10.1002/1521-3773(20020617)41:12<2034::AID-ANIE2034>3.0.CO;2-M hl dl i h l i l d http://dx.doi.org/10.1002/1521 3773(20020617)41:12<2034::AID ANIE2034>3.0.CO;2 M [2] Behl M, Lendlein A. Shape-memory polymers. Materials Today 2007;10:20–8. http://dx.doi.org/10.1016/S1369-7021(07)70047-0 [3] Leng J, Lu H, Liu Y, Du S. Conductive nanoparticles in electro activated shape memory polymer sensor and actuator. Paper presented at the proceedings of SPIE – the international society for optical engineering. Nanosensors and microsensors for bio-systems, San Diego, CA, United States, vol. 6931; 2008. p. 693109. http://dx.doi.org/10.1117/12.775743 [3] Leng J, Lu H, Liu Y, Du S. Conductive nanoparticles in electro activated shape memory polymer sensor and actuator. Paper presented at the proceedings of SPIE – the international society for optical engineering. Nanosensors and microsensors for bio-systems, San Diego, CA, United States, vol. 6931; 2008. p. 693109. http://dx.doi.org/10.1117/12.775743 [3] Leng J, Lu H, Liu Y, Du S. Conductive nanoparticles in electro activated shape memory polymer sensor and actuator. Paper presented at the proceedings of SPIE – the international society for optical engineering. Nanosensors and microsensors for bio-systems, San Diego, CA, United States, vol. 6931; 2008. p. 693109. http://dx.doi.org/10.1117/12.775743 [4] Tcharkhtchi A. and al, Some New Concepts of Shape Memory Effect of Polymers. Polymers 2014, 6, 1144-1163. http://dx.doi.org/10.3390/polym6041144 [4] Tcharkhtchi A. and al, Some New Concepts of Shape Memory Effect of Polymers. Polymers 2014, 6, 1144-1163. http://dx.doi.org/10.3390/polym6041144 [5] Farzaneh, S., Fitoussi, J., Lucas, A., Bocquet, M., Tcharkhtchi, A., "Shape memory effect and properties memory effect of polyurethane", Journal of Applied Polymer Science, Volume 128, Issue 5, pages 3240–3249, 2013. http://dx.doi.org/10.1002/app.38530 [5] Farzaneh, S., Fitoussi, J., Lucas, A., Bocquet, M., Tcharkhtchi, A., "Shape memory effect and properties memory effect of polyurethane", Journal of Applied Polymer Science, Volume 128, Issue 5, pages 3240–3249, 2013. http://dx.doi.org/10.1002/app.38530 [6] S. Abdallah-Elhirtsi, J. Fitoussi, B. J. Rashmi, K. Prashantha, S. Farzaneh, M. F. Lacrampe, P. Krawczak, A.Tcharkhtchi, "Study of partial shape memory effect of polymers by multicycle tests", , Polymer Composites, 36 (6), pp. 1145-1151 (2015). http://dx.doi.org/10.1002/pc.23472 [7] T. Tadaki and al. Shape Memory Alloys. Annual Review of Materials Science. Vol. 18: 25-45 (1988). http://dx.doi.org/10.1146/annurev.ms.18.080188.000325 [8] Lendlein A, Langer R. Biodegradable, elastic shape-memory polymers for potential biomedical applications. Science 2002;96:1673–6. http://dx.doi.org/10.1126/science.1066102 [8] Lendlein A, Langer R. Biodegradable, elastic shape-memory polymers for potential biomedical applications. References Science 2002;96:1673–6. http://dx.doi.org/10.1126/science.1066102 [9] Wache HM, Tartakowska DJ, Hentrich A, Wagner MH. Development of a polymer stent with shape memory effect as a drug delivery system. J Mat. Sci: Mater Med 2004;14:109-12. http://dx.doi.org/10.1023/A:1022007510352 [9] Wache HM, Tartakowska DJ, Hentrich A, Wagner MH. Development of a polymer stent with shape memory effect as a drug delivery system. J Mat. Sci: Mater Med 2004;14:109-12. http://dx.doi.org/10.1023/A:1022007510352 p g [10] Hu JL, Ding XM, Tao XM. Shape memory polymers and their applications to smart textile products. J China Textile Univ 2002;19:89–93. Application textile [10] Hu JL, Ding XM, Tao XM. Shape memory polymers and their applications to smart textile products. J China Textile Univ 2002;19:89–93. Application textile [11] Russel DA, Hayashi S, Yamada T. The potential use of memory film in clothing. Paper presented at the techtextil symposium-new protective textiles (through textile technology index database), April 1999. Application textile [11] Russel DA, Hayashi S, Yamada T. The potential use of memory film in clothing. Paper presented at the techtextil symposium-new protective textiles (through textile technology index database), April 1999. Application textile [12] J. Karger-Kocsis. Recent advances in shape memory polymers and composites: a review. J Mater Sci (2008) 43:254–269. http://dx.doi.org/10.1007/s10853-007-2176-7 [12] J. Karger-Kocsis. Recent advances in shape memory polymers and composites: a review. J Mater Sci (2008) 43:254–269. http://dx.doi.org/10.1007/s10853-007-2176-7 [13] T. Xie. Recent advances in polymer shape memory. Polymer 52 (2011) 4985-5000. http://dx.doi.org/10.1016/j.polymer.2011.08.003 [13] T. Xie. Recent advances in polymer shape memory. Polymer 52 (2011) 4985-5000. http://dx.doi.org/10.1016/j.polymer.2011.08.003 http://dx.doi.org/10.1016/j.polymer.2011.08.003 [14] J. Hu. Recent advances in shape–memory polymers: Structure, mechanism, functionality, modeling and applications. Progress in Polymer Science 37 (2012) 1720– 1763. http://dx.doi.org/10.1016/j.progpolymsci.2012.06.001 [15] Lin JR, Chen LW. Study on shape-memory behavior of polyether-based polyurethanes. I. Influence of the hard-segment content. J Appl Polym Sci 1998;69:1563–74. http://dx.doi.org/10.1002/(SICI)1097-4628(19980822)69:8<1563::AID-APP11>3.0.CO;2-W [16] Lin JR, Chen LW. Study on shape-memory behavior of polyether-based polyurethanes. II. Influence of soft-segment molecular weight. J Appl Polym Sci 1998;69:1575–86. http://dx.doi.org/10.1002/(SICI)1097-4628(19980822)69:8<1575::AID-APP12>3.0.CO;2-U [17] NF ISO 6239:1986. Plastics Determination of Tensile Properties by Use of Small Specimens. Association Française de Normalisation (AFNOR): Seine-Saint-Denis, France, 1986. [18] Zhang H., Wang H., Zhong W. and Du Q. A novel type of shape memory polymer blend and the shape memory mechanism. Polymer 50 (2009) 1596-1601. http://dx doi org/10 1016/j polymer 2009 01 011 [14] J. Hu. Recent advances in shape–memory polymers: Structure, mechanism, functionality, modeling and applications. Progress in Polymer Science 37 (2012) 1720– 1763. 155 155 Residual Stresses 2016: ICRS-10 Materials Research Forum LLC Materials Research Proceedings 2 (2016) 151-156 doi: http://dx.doi.org/10.21741/9781945291173-26 References http://dx.doi.org/10.1016/j.progpolymsci.2012.06.001 [15] Lin JR Chen LW Study on shape memory behavior of polyether based polyurethanes I [14] J. Hu. Recent advances in shape–memory polymers: Structure, mechanism, functionality, modeling and applications. Progress in Polymer Science 37 (2012) 1720– 1763. http://dx.doi.org/10.1016/j.progpolymsci.2012.06.001 ç ( ) [18] Zhang H., Wang H., Zhong W. and Du Q. A novel type of shape memory polymer blend and the shape memory mechanism. Polymer 50 (2009) 1596-1601. http://dx.doi.org/10.1016/j.polymer.2009.01.011 156
https://openalex.org/W2012650972	https://repository.ubn.ru.nl/bitstream/handle/2066/71485/3/71485.pdf	English	null	Peroxisome Proliferator-activated Receptor γ Activation Promotes Infiltration of Alternatively Activated Macrophages into Adipose Tissue	Journal of biological chemistry/The Journal of biological chemistry	2,008	cc-by	8,141	Peroxisome proliferator-activated receptor gamma activation promotes infiltration of alternatively activated macrophages into adipose tissue. Stienstra, R.; Duval, C.; Keshtkar, S.; Laak, J.A.W.M. van der; Kersten, S.; Muller, M. 2008, Article / Letter to editor (Journal of Biological Chemistry, 283, 33, (2008), pp. 22620-7) Doi link to publisher: https://doi.org/10.1074/jbc.M710314200 Version of the following full text: Publisher’s version Downloaded from: http://hdl.handle.net/2066/71485 Download date: 2024-10-24 Peroxisome Proliferator-activated Receptor Activation Promotes Infiltration of Alternatively Activated Macrophages into Adipose Tissue* Rinke Stienstra‡§1, Caroline Duval‡§, Shohreh Keshtkar‡§, Jeroen van der Laak¶, Sander Kersten‡§, andMichaelMu¨ller‡§2 From the ‡Nutrition, Metabolism, and Genomics Group, Division of Human Nutrition, Wageningen University, 6700 EV Wageningen, §Nutrigenomics Consortium, Top Institute of Food and Nutrition, Wageningen, and ¶Department of Pathology, Radboud University Nijmegen Medical Centre, 6525 GA Nijmegen, The Netherlands Obesity is associated with infiltration of macrophages into adi- pose tissue. Adipose macrophages may contribute to an elevated inflammatory status by secreting a variety of proinflammatory mediators, including tumor necrosis factor and interleukin-6 (IL-6). Recent data suggest that during diet-induced obesity the phenotypeofadipose-residentmacrophageschangesfromalterna- tively activated macrophages toward a more classical and pro-in- flammatory phenotype. Here, we explore the effect of peroxisome proliferator-activated receptor activation on obesity-induced inflammation in 129SV mice fed a high fat diet for 20 weeks. High fat feeding increased bodyweight gain, adipose tissue mass, and liver triglycerides. Rosiglitazone treatment further increased adi- pose mass, reduced liver triglycerides, and changed adipose tissue morphologytowardsmalleradipocytes.Surprisingly,rosiglitazone markedly increased the number of macrophages in adipose tissue, as shown by immunohistochemical analysis and quantification of macrophage marker genes CD68 and F4/80. In adipose tissue, markers for classically activated macrophages including IL-18 were down-regulated, whereas markers characteristic for alterna- tively activated macrophages (arginase 1, IL-10) were up-regulated by rosiglitazone. Importantly, conditioned media from rosiglita- zone-treated alternatively activated macrophages neutralized the inhibitory effect of macrophages on 3T3-L1 adipocyte differentia- tion, suggesting that alternatively activated macrophages may be involved in mediating the effects of rosiglitazone on adipose tissue morphology and mass. Our results suggest that short term rosigli- tazone treatment increases infiltration of alternatively activated macrophages in adipose tissue. The alternatively activated macro- phages might play a role in peroxisome proliferator-activated receptor-dependentexpansionandremodelingofadiposetissue. increase morbidity risk, which are collected in the metabolic syndrome and include hypertension, dyslipidemia, and insulin resistance (2). Each serves as an independent risk factor for atherosclerosis and associated coronary heart disease. increase morbidity risk, which are collected in the metabolic syndrome and include hypertension, dyslipidemia, and insulin resistance (2). Each serves as an independent risk factor for atherosclerosis and associated coronary heart disease. y Although the overall negative impact of obesity and meta- bolic syndrome on morbidity is evident, it has been very diffi- cult to get a handle on why some individuals are obese seem- ingly without any damaging consequences for health, whereas others are afflicted by a range of metabolic abnormalities. 3 The abbreviations used are: IL, interleukin; PPAR, peroxisome proliferator- activated receptor; qPCR, quantitative PCR; LFD, low fat diet; HFD, low fat diet; FCS, fetal calf serum. Note: To cite this publication please use the final published version (if applicable). THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 283, NO. 33, pp. 22620–22627, August 15, 2008 © 2008 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A. THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 283, NO. 33, pp. 22620–22627, August 15, 2008 © 2008 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A. y 1 Current address: Dept. of Internal Medicine, Radboud University Nijmegen Medical Centre, 6525 GA Nijmegen, The Netherlands. * This study was supported by the Centre for Human Nutrigenomics and Top Institute of Food and Nutrition. The costs of publication of this article were defrayed in part by the payment of page charges. This article must there- fore be hereby marked “advertisement” in accordance with 18 U.S.C. Sec- tion 1734 solely to indicate this fact. j g 2 To whom correspondence should be addressed: Division of Human Nutri- tion, Wageningen University, P. O. Box 8129, 6700 EV Wageningen, The Netherlands, Tel.: 31-317-482590; Fax: 31-317-483342; E-mail: michael. muller@wur.nl. 22620 JOURNAL OF BIOLOGICAL CHEMISTRY This is an Open Access article under the CC BY license. * This study was supported by the Centre for Human Nutrigenomics and Top Institute of Food and Nutrition. The costs of publication of this article were defrayed in part by the payment of page charges. This article must there- fore be hereby marked “advertisement” in accordance with 18 U.S.C. Sec- tion 1734 solely to indicate this fact. 1 Current address: Dept. of Internal Medicine, Radboud University Nijmegen Medical Centre, 6525 GA Nijmegen, The Netherlands. 2 To whom correspondence should be addressed: Division of Human Nutri- tion, Wageningen University, P. O. Box 8129, 6700 EV Wageningen, The Netherlands, Tel.: 31-317-482590; Fax: 31-317-483342; E-mail: michael. muller@wur.nl. PPAR Activation Promotes Macrophage Infiltration into WAT In the last week of diet intervention, half of the mice receiv- ing the HFD were switched to HFD supplemented with ros- iglitazone (0.01% w/w). At the end of the feeding experiment, blood was collected in EDTA-coated tubes and centrifuged at 10,000 g to collect plasma. Liver and epididymal white adipose tissue were dissected, weighed, and immediately fro- zen in liquid nitrogen. The animal experiments were approved by the animal experimentation committee of Wageningen University. As mentioned above, PPAR is the molecular target for the insulin-sensitizing thiazolidinediones, which effectively lower plasma glucose and insulin levels by promoting insulin sensitiv- ity and, thus, stimulating glucose uptake. Thiazolidinediones also reduce plasma free fatty acid concentrations, although this effect is mainly evident in rodents (14, 15). Microarray Analysis, RNA Isolation, and Quality Control— Total RNA was isolated from adipose tissue or 3T3-L1 adipo- cytes using TRIzol reagent (Invitrogen) according to the man- ufacturer’s instructions. Concentrations and purity of RNA samples were determined on a NanoDrop ND-1000 spectro- photometer (Isogen, Maarssen, The Netherlands). RNA integ- rity was checked on an Agilent 2100 bioanalyzer (Agilent Tech- nologies, Amsterdam, the Netherlands) with 6000 Nano Chips. In addition to its role in adipocyte differentiation and glucose metabolism, PPAR also has potent anti-inflammatory activity. Treatment of mice with rosiglitazone causes a significant decrease in expression of numerous inflammatory mediators, including tumor necrosis factor and IL-6 (16). In adipose tissue, induction of adipocyte differentiation by PPAR is par- alleled by the appearance of smaller adipocytes (17), which may partially account for the inhibitory effect of PPAR on inflam- matory gene expression (18). Furthermore, it has been sug- gested that PPAR may suppress inflammation associated with diet-induced obesity by lowering the number of macrophages present in adipose tissue (5, 19). Pooled RNA samples from five mice per experimental group were used for microarray analysis. Samples were hybridized on Affymetrix GeneChip Mouse Genome 430A arrays. Expression levels were calculated applying the multi-chip modified gamma model for oligonucleotide signal (multi-mgMOS) (20) and a remapped Gene Chip Description (CDF) File (21). Detailed descriptions of the applied methods are available on request. Microarray data were submitted to the NCBI gene expression and hybridization array data repository. The assigned GEO accession number is GSE11295. PPAR Activation Promotes Macrophage Infiltration into WAT The aim of the present study was to examine the effect of rosiglitazone on macrophage infiltration in Sv129 mice ren- dered obese by chronic feeding of a high fat diet. Our data indicate that short term rosiglitazone treatment increases infiltration of alternatively activated macrophages in adipose tissue. Ingenuity Pathway Analysis (Ingenuity Systems, Redwood City, CA) was used to identify the most significantly changed cellular pathways using input criteria of a relative fold change above 2 or below 2 and a p value similar to or below 0.001. PPAR Activation Promotes Macrophage Infiltration into WAT PPAR Activation Promotes Macrophage Infiltration into WAT TABLE 1 Primers sequences used for qPCR Gene Forward primer Reverse primer 36B4 AGCGCGTCCTGGCATTGTGTGG GGGCAGCAGTGGTGGCAGCAGC UCP1 CCTGCCTCTCTCGGAAACAA TGTAGGCTGCCCAATGAACA CPT-1b TCTCCTACCACGGGTGGATGTT CCATGACCGGCTTGATCTCTTC CD36 GAGCAACTGGTGGATGGTTT GCAGAATCAAGGGAGAGCAC Adiponectin GCAGAGATGGCACTCCTGGA CCCTTCAGCTCCTGTCATTCC Glut4 GGAAGGAAAAGGGCTATGCTG TGAGGAACCGTCCAAGAATGA CD68 CCAATTCAGGGTGGAAGAAA CTCGGGCTCTGATGTAGGTC F4/80 CTTTGGCTATGGGCTTCCAGTC GCAAGGAGGACAGAGTTTATCGTG Cyclin A2 CTTTACCCGCAGCAAGAAAAC ACGTTCACTGGCTTGTCTTCTA Cyclin B CTCAGGGTCACTAGGAACACG AGCTCTTCGCTGACTTTATTACC Clec7a AGGTTTTTCTCAGTCTTGCCTTC GGGAGCAGTGTCTCTTACTTCC Arginase 1 TGGCTTGCGAGACGTAGAC GCTCAGGTGAATCGGCCTTTT IL-10 CTGGACAACATACTGCTAACCG GGGCATCACTTCTACCAGGTAA sIL-1ra AAATCTGCTGGGGACCCTAC TGAGCTGGTTGTTTCTCAGG IL-18 GACTCTTGCGTCAACTTCAAGG CAGGCTGTCTTTTGTCAACGA MCP-1 CCCAATGAGTAGGCTGGAGA TCTGGACCCATTCCTTCTTG MSR1 TGAACGAGAGGATGCTGACTG GGAGGGGCCATTTTTAGTGC SAA3 GTTCACGGGACATGGAGCAGAGGA GCAGGCCAGCAGGTGGGAAGTG Fibr- AGTGTTGTGTCCTACGGGATG CTGAGGAGGTATCGGAAACAGA IL-6 CTTCCATCCAGTTGCCTTCTTG AATTAAGCCTCCGACTTGTGAAG Leptin AGAAGATCCCAGGGAGGAAA TGATGAGGGTTTTGGTGTCA PPAR Activation Promotes Macrophage Infiltration into WAT TABLE 1 Primers sequences used for qPCR TABLE 1 TABLE 1 Primers sequences used for qPCR drugs, the thiazolidinediones. In addition, PPAR binds and is activated by polyunsaturated fatty acids and fatty acid-derived molecules. The expression of PPAR is highest in adipose tis- sues, where it plays a pivotal role in the adipocyte differentia- tion and lipid storage (10, 11). PPAR is also relatively well expressed in macrophages. Target genes of PPAR identified so far include fatty acid-binding protein 4, GLUT4, glycogen syn- thase 2, glycerol-3-phosphate dehydrogenase, lipoprotein lipase, glycerol kinase, and aquaporin 7 (12, 13). drugs, the thiazolidinediones. In addition, PPAR binds and is activated by polyunsaturated fatty acids and fatty acid-derived molecules. The expression of PPAR is highest in adipose tis- sues, where it plays a pivotal role in the adipocyte differentia- tion and lipid storage (10, 11). PPAR is also relatively well expressed in macrophages. Target genes of PPAR identified so far include fatty acid-binding protein 4, GLUT4, glycogen syn- thase 2, glycerol-3-phosphate dehydrogenase, lipoprotein lipase, glycerol kinase, and aquaporin 7 (12, 13). (D12450B or D12451, Research Diets, New Brunswick, NJ). The lard component in these diets was replaced by palm oil. In the last week of diet intervention, half of the mice receiv- ing the HFD were switched to HFD supplemented with ros- iglitazone (0.01% w/w). At the end of the feeding experiment, blood was collected in EDTA-coated tubes and centrifuged at 10,000 g to collect plasma. Liver and epididymal white adipose tissue were dissected, weighed, and immediately fro- zen in liquid nitrogen. The animal experiments were approved by the animal experimentation committee of Wageningen University. (D12450B or D12451, Research Diets, New Brunswick, NJ). The lard component in these diets was replaced by palm oil. Peroxisome Proliferator-activated Receptor Activation Promotes Infiltration of Alternatively Activated Macrophages into Adipose Tissue* Clearly, our understanding of the chain of events that leads to metabolic syndrome, although growing, is still remarkably scarce. Recent studies suggest an important role for inflamma- tory processes. Indeed, it has been found that obesity is associ- ated with a state of chronic low grade inflammation which is likely caused by adipocyte hypertrophy together with infiltra- tion of macrophages into adipose tissue (3). As a result, the secretion of pro-inflammatory mediators such as tumor necro- sis factor and IL-63 from adipose tissue is increased, leading to disruption of normal homeostatic control of metabolism either locally or systemically (4–6). Why macrophages infiltrate adi- pose tissue during obesity is currently unclear, although it has been suggested that macrophage localization and infiltration is strongly linked to adipose cell death (7). More recently, it has been proposed that adipose tissue resident macrophages itself undergo phenotypic changes during obesity. In adipose tissue of mice rendered obese by high fat feeding, macrophages appear to be mainly activated according to “classical activa- tion,” whereas macrophages present in adipose tissue of lean mice are “alternatively activated” (8). Classically activated mac- rophages express high levels of pro-inflammatory mediators including tumor necrosis factor , which may contribute to insulin resistance. In contrast, alternatively activated macro- phages are considered anti-inflammatory by expressing genes such as IL-10, IL-1 receptor antagonist, and arginase I (9). The global prevalence rate of obesity is rising steadily (1). Obesity is linked to several metabolic disturbances that greatly The peroxisome proliferator-activated receptor is a ligand- activated transcription factor and member of the superfamily of nuclear receptors. It regulates gene transcription by binding to specific DNA sequences in target genes, leading to chromatin remodeling and activation of gene transcription. PPAR serves as the molecular target for an important class of anti-diabetic * This study was supported by the Centre for Human Nutrigenomics and Top Institute of Food and Nutrition. The costs of publication of this article were defrayed in part by the payment of page charges. This article must there- fore be hereby marked “advertisement” in accordance with 18 U.S.C. Sec- tion 1734 solely to indicate this fact. 22620 JOURNAL OF BIOLOGICAL CHEMISTRY This is an Open Access article under the CC BY license. VOLUME 283•NUMBER 33•AUGUST 15, 2008 PPAR Activation Promotes Macrophage Infiltration into WAT formed. The primers used are listed in Table 1. The mRNA expression of all genes reported was normalized to 36B4 gene expression. formed. The primers used are listed in Table 1. The mRNA expression of all genes reported was normalized to 36B4 gene expression. filter. After centrifugation, the floating cells were collected as adipocytes, and the pelleted cells were collected as stromal vas- cular cells. Both cell fractions were washed with phosphate- buffered saline, and RNA was isolated using Trizol reagent (Invitrogen). Histology/Immunohistochemistry—For detection of macro- phages/monocytes, an F4/80 antibody (Serotec, Oxford, UK) was used. Sections were preincubated with 20% normal goat serum followed by overnight incubation at 4 °C with the pri- mary antibody diluted 1:50 in phosphate-buffered saline, 1% bovine serum albumin. After incubation with the primary anti- body, a goat anti-rat IgG conjugated to horseradish peroxidase (Serotec) was used as a secondary antibody. Visualization of the complex was done using 3,3-diaminobenzidene for 5 min. Negative controls were used by omitting the primary antibody. Hematoxylin and eosin staining of sections was done using standard protocols. The percentage of F4/80-expressing cells for each slide was calculated as the sum of the number of nuclei of F4/80-positive cells divided by the total number of nuclei present in a slide. Four fields per slide were included, and a total of 4 or 5 mice per group were used. g Cell Culture—Mouse RAW264.7 cells were obtained from the European Council of American Chambers OF Commerce (Salisbury, Wiltshire, UK) and grown in Dulbecco’s modified Eagle’s medium containing 10% FCS and antibiotics. Cells were treated with IL-4 (5 ng/ml), rosiglitazone (1 M), or both for 24 h. Medium was removed, and cells were carefully washed using phosphate-buffered saline and re-incubated with fresh medium without the addition of ligands and/or cytokines. 16 h later medium was removed from RAW264.7 cells, centrifuged, and stored at 20 °C until further use. 3T3-L1 fibroblasts were grown in Dulbecco’s modified Eagle’s medium plus 10% calf serum and plated for final differentiation in Dulbecco’s modi- fied Eagle’s medium plus 10% FCS. Two days after reaching confluence, the medium was changed to macrophage-condi- tioned medium, and the following compounds were added: isobutyl methylxanthine (0.5 mM), dexamethasone (1 M), and insulin (5 g/ml). After 2 days the medium was changed to macrophage-conditioned medium plus 10% FCS and insulin (5 g/ml). RESULTS Rosiglitazone Treatment of Obese Mice Increases Adipose Mass Together with Up-regulation of PPAR Target Genes— Sv129 mice were fed a LFD or HFD for 20 weeks. Although energy intake was similar (Fig. 1A), mice on HFD showed a significantly higher bodyweight gain compared with mice on LFD (Fig. 1B). Adipose tissue weight at the end of the study was also significantly higher in the HFD animals (Fig. 1D). One week of rosiglitazone treatment of mice fed the HFD further increased adipose tissue weight (Fig. 1D) and plasma leptin concentrations (Fig. 1E). HFD also significantly increased plasma insulin (Fig. 1F), suggesting development of insulin resistance as well as liver triglycerides (Fig. 1G), which were reversed by rosiglitazone. The increase in adipose tissue weight and decrease in liver triglyceride content strongly suggests redistribution of liver fat toward white adipose tissue after ros- iglitazone treatment. Rosiglitazone also caused a marked increase in the expression of several PPAR target genes in adipose tissue including UCP-1 and CPT-1b (Fig. 1H). In line with previous data, an overall change in adipocyte size was observed after rosiglitazone treatment resulting in the appear- ance of smaller adipocytes (Fig. 1I and J). 4,6-Diamidino-2-phenylindole Staining—To visualize DNA content of adipose tissue, sections were incubated with 300 nM 4,6-diamidino-2-phenylindole (Sigma-Aldrich) for 30 min at room temperature. Sections were analyzed using a fluorescence microscope. Liver Triglycerides—Liver triglycerides were determined in 10% liver homogenates prepared in buffer containing 250 mM sucrose, 1 mM EDTA, and 10 mM Tris-HCl at pH 7.5 using a commercially available kit from Instruchemie (Delfzijl, The Netherlands). Plasma Insulin and Leptin—Plasma levels of insulin were measured using a commercially available kit from Linco (LINCO Research Inc., St. Charles, MO). Plasma levels of leptin were measured using a kit from R&D systems (Abing- don, UK). Fractioning of Adipose Tissue—Freshly isolated epididymal adipose tissue was used for the isolation of adipocytes and stro- mal vascular cells. Minced adipose tissue was digested using collagenase (Sigma-Aldrich) at a concentration of 5 mg/ml dis- solved in Dulbecco’s modified Eagle’s medium with 10% FCS (Lonza, Verviers, Belgium). Tissues were incubated for 45 min at 37 °C and subsequently filtered through a 250 M nylon mesh Rosiglitazone Treatment Promotes Macrophage Infiltration— Besides an increase in adipose mass, rosiglitazone treatment promoted formation of clusters of cells surrounding adipo- cytes, as revealed by 4,6-diamidino-2-phenylindole nuclear staining, suggesting the presence of cell aggregates (Fig. 2A). PPAR Activation Promotes Macrophage Infiltration into WAT After 4 days the medium was changed to macrophage- conditioned medium plus 10% FCS, which was changed every 2 days. Oil Red O staining of 3T3 adipocytes was performed using a standard protocol. Adipocyte Size—Morphometry of individual fat cells was assessed using digital image analysis. Microscopic images were digitized in 24-bit red/green/blue (specimen level pixel size 1.28 1.28 um2). Recognition of fat cells was initially per- formed by applying a region-growing algorithm on manually indicated seed points. If required, results of region growing were interactively corrected. From the resulting regions the area and minimum Feret diameter were calculated. The mini- mum Feret diameter is useful because fat cells are generally (near) convex objects. For each specimen, all fat cells in 3–5 microscopic fields of view were measured. On average, 465 fat cells were measured per specimen (range 335–591). Differ- ences between groups were studied using one-way univariate analysis of variance. Tukey honestly significant difference post hoc testing was applied for multiple comparison testing. All statistics were performed using SPSS software (SPSS Inc, Chi- cago, Il). Statistical Analysis—Statistical significant differences were calculated using Student’s t test. The cut-off for statistical sig- nificance was set at a p value of 0.05 or below. MATERIALS AND METHODS Real-time PCR—1 g of RNA was used for reverse transcrip- tion using the iScript cDNA Synthesis Kit (Bio-Rad). Real-time PCR was done with platinum Taq polymerase (Invitrogen) and SYBR Green using an iCycler PCR machine (Bio-Rad). Melt curve analysis was included to assure a single PCR product was Animal Study—Sv129 male mice were purchased at The Jackson Laboratory (Bar Harbor, ME). Male mice received a low fat diet (LFD) or low fat diet (HFD) for 21 weeks, provid- ing 10 or 45% energy percent in the form of triglycerides JOURNAL OF BIOLOGICAL CHEMISTRY AUGUST 15, 2008•VOLUME 283•NUMBER 33 22621 RESULTS VOLUME 283•NUMBER 33•AUGUST 15, 2008 22622 JOURNAL OF BIOLOGICAL CHEMISTRY 22622 PPAR Activation Promotes Macrophage Infiltration into WAT FIGURE 1. Effect of high fat diet and rosiglitazone on adiposity and metabolic parameters. Sv129 mice were fed a LFD or HFD for 20 weeks. In half of the animals receiving the HFD, rosiglitazone was added to the HFD during the last week of the dietary intervention. A, daily caloric intake in mice fed LFD or HFD. B, percent- age change in bodyweight during the course of the LFD or HFD feeding. C, GTT of LFD and HFD fed mice. D, weight of white adipose tissue (WAT) as a percentage of total bodyweight. E, plasma leptin concentration as measured in individual plasma samples (n 4–5 mice per group). F, plasma insulin concentration. G, liver triglyceride content. H, changes in adipose gene expression after rosiglitazone treatment, as determined by qPCR. I, representative hematoxylin and eosin-staining of white adipose tissue in the three experimental groups. J, quantification of adipocyte size. Error bars represent S.E. Differences were evaluated using Students t test. , p 0.05; , p 0.01; *, p 0.001. n 4–5 animals per group. PPAR Activation Promotes Ma zone increased F4/80 staining, suggesting more macrophages were present. These results were corrob- orated by quantitative analysis of the number of macrophages present in adipose tissue (Fig. 2C) and gene expression analysis of F4/80 and CD68, another macrophage specific gene (Fig. 2D). In agreement with an increased macrophage content, expression of MCP-1 was up-regu- lated by rosiglitazone in total adi- pose tissue (Fig. 2D). These data suggest that short term rosiglita- zone treatment stimulates macro- phage infiltration in adipose tissue of obese mice. Noticeably, plasma concentrations of MCP-1 were not elevated after rosiglitazone treat- ment and even showed a tendency toward a decrease (data not shown), suggesting that MCP-1 expression is increased only locally. To examine the effect of rosiglita- zone and the increased abundance of macrophages on global adipose gene expression, microarray analy- sis was performed in animals fed HFD with or without rosiglitazone. As expected, rosiglitazone signifi- cantly changed genes involved in lipid metabolism and energy pro- duction (Fig. 3). More surprisingly, cell cycle, tissue morphology, and connective tissue development were significantly altered after ros- iglitazone treatment, suggesting changes in adipose tissue structure and composition. RESULTS Rosiglitazone Increases Adipose Abundance of Alternatively Acti- vated Macrophages—Macrophages can either be activated by T helper 1-type responses via interferon and by alternative activation via T helper 2-type cytokines. The latter response, which is stimulated by IL-4 and IL-13, elicits a macrophage phenotype connected with tissue repair and remodeling and block- age of inflammation (9). To investi- gate the effects of rosiglitazone on adipose tissue macrophage polarization, the expression of sev- eral macrophage markers was determined in adipose tissue (Fig. 4). Strikingly, adipose expression of sIL-1ra, Clec7a, argi- nase 1, MSR1, and IL-10, which are markers characteristic of alternatively activated macrophages (9, 22, 23), were all signif- icantly increased by rosiglitazone treatment. In contrast, bolic parameters. Sv129 mice osiglitazone was added to the ce fed LFD or HFD. B, percent- GTT of LFD and HFD fed mice. plasma leptin concentration as insulin concentration. G, liver e treatment, as determined by ue in the three experimental were evaluated using Students FIGURE 1. Effect of high fat diet and rosiglitazone on adiposity and metabolic parameters. Sv129 mice were fed a LFD or HFD for 20 weeks. In half of the animals receiving the HFD, rosiglitazone was added to the HFD during the last week of the dietary intervention. A, daily caloric intake in mice fed LFD or HFD. B, percent- age change in bodyweight during the course of the LFD or HFD feeding. C, GTT of LFD and HFD fed mice. D, weight of white adipose tissue (WAT) as a percentage of total bodyweight. E, plasma leptin concentration as measured in individual plasma samples (n 4–5 mice per group). F, plasma insulin concentration. G, liver triglyceride content. H, changes in adipose gene expression after rosiglitazone treatment, as determined by qPCR. I, representative hematoxylin and eosin-staining of white adipose tissue in the three experimental groups. J, quantification of adipocyte size. Error bars represent S.E. Differences were evaluated using Students t test. , p 0.05; , p 0.01; , p 0.001. n 4–5 animals per group. The clusters of cells strongly resembled so-called Crown-like structures that are formed by aggregated macrophages in adi- pose tissue of obese mice (7). To analyze if more macrophages were present in adipose tissue after rosiglitazone treatment, immunohistochemical staining using the specific macrophage marker F4/80 was performed. As shown in Fig. RESULTS Differences were evaluated using Student’s t test. , p 0.05; , p 0.01; , p 0.001. n 4–5 animals per group. serum amyloid A3, and IL-6 were also down-regulated by ros- iglitazone, although the latter failed to reach statistical signifi- cance. Rosiglitazone did not alter adipose expression levels of CD11c, suggesting no change in the number of pro-inflamma- tory classically polarized macrophages (8). p 0.01; *, p 0.001. n 4–5 animals per group. FIGURE 3. Microarray analysis reveals changes in lipid metabolism, energy production,cellcycle,andtissuemorphologyafterrosiglitazonetreatment. Pooled white adipose tissues from mice fed a HFD or HFD plus rosiglitazone was analyzed using Affymetrix GeneChips. The top 10 list of most significantly changed pathways after rosiglitazone treatment is shown. Pathway analysis was carried out using the Ingenuity program. A -fold change above 2 or below 2 and a p value similar to or smaller then 0.001 were used as cutoff values. To investigate whether the changes in total adipose tissue gene expression were specific for the stromal vascular fraction containing adipose tissue macrophages, adipose tissue from was fractionated into stromal vascular cells and adipocytes. As expected, leptin was predominantly expressed in the adipocyte fraction (Fig. 5A). In the stromal vascular fraction, rosiglitazone significantly up-regulated expression of arginase I and sIL-1ra (Fig. 5B). In contrast, rosiglitazone down-regulated expression of IL-18. Thus, changes in expression of markers genes observed in stromal vascular cells mirrored those in white adi- pose tissue . Taken together, these data suggest that rosiglita- zone increases the presence of alternatively activated macro- phages in adipose tissue. FIGURE 3. Microarray analysis reveals changes in lipid metabolism, energy production,cellcycle,andtissuemorphologyafterrosiglitazonetreatment. Pooled white adipose tissues from mice fed a HFD or HFD plus rosiglitazone was analyzed using Affymetrix GeneChips. The top 10 list of most significantly changed pathways after rosiglitazone treatment is shown. Pathway analysis was carried out using the Ingenuity program. A -fold change above 2 or below 2 and a p value similar to or smaller then 0.001 were used as cutoff values. It has been reported that alternatively activated macrophages stimulate cell proliferation (9). As mentioned above, global analysis of gene expression by microarray indicated that rosigli- tazone altered expression of genes related to cell cycle, prolif- eration, and tissue morphology (Fig. 3). Therefore, we analyzed expression levels of cyclin A2 and cyclin B in total white adipose tissue . RESULTS Rosiglitazone increased the abundance of macrophages in adi- pose tissue of mice fed a HFD. A, representative hematoxylin-eosin (HE) and 4,6-diamidino-2-phenylindole-staining of white adipose tissue in the three experimental groups. B, representative immunohistochemical staining of white adipose tissue using the specific macrophage marker F4/80. Magni- fication is indicated at the right bottom corner. C, quantification of the % of macrophages present in adipose tissue. D, changes in adipose expression of macrophage marker genes after rosiglitazone treatment, as determined by qPCR. Gene expression levels in LFD-fed animals were set at 1. Error bars rep- resent S.E. Differences were evaluated using Student’s t test. , p 0.05; , p 0.01; , p 0.001. n 4–5 animals per group. FIGURE 3. Microarray analysis reveals changes in lipid metabolism, energy production,cellcycle,andtissuemorphologyafterrosiglitazonetreatment. Pooled white adipose tissues from mice fed a HFD or HFD plus rosiglitazone was analyzed using Affymetrix GeneChips. The top 10 list of most significantly changed pathways after rosiglitazone treatment is shown. Pathway analysis was carried out using the Ingenuity program. A -fold change above 2 or below 2 and a p value similar to or smaller then 0.001 were used as cutoff values. PPAR Activation Promotes Macrophage Infiltration PPAR Activation Promotes Macrophage Infiltr FIGURE 4. Rosiglitazone increased the expression of markers for alterna- tively activated macrophages in adipose tissue of mice fed a HFD. Changes in adipose gene expression in all three experimental groups, as determined by qPCR. Gene expression levels in LFD fed animals were set at 1. Error bars represent S.E. Differences are evaluated using Student’s t test. , p 0.01; , p 0.05. n 4–5 animals per group. FIGURE 2. Rosiglitazone increased the abundance of macrophages in adi- pose tissue of mice fed a HFD. A, representative hematoxylin-eosin (HE) and FIGURE 2. Rosiglitazone increased the abundance of macrophages in adi- pose tissue of mice fed a HFD. A, representative hematoxylin-eosin (HE) and 4,6-diamidino-2-phenylindole-staining of white adipose tissue in the three experimental groups. B, representative immunohistochemical staining of white adipose tissue using the specific macrophage marker F4/80. Magni- fication is indicated at the right bottom corner. C, quantification of the % of macrophages present in adipose tissue. D, changes in adipose expression of macrophage marker genes after rosiglitazone treatment, as determined by qPCR. Gene expression levels in LFD-fed animals were set at 1. Error bars rep- resent S.E. RESULTS Thus, changes in expression of markers genes observed in stromal vascular cells mirrored those in white adi- pose tissue . Taken together, these data suggest that rosiglita- zone increases the presence of alternatively activated macro- phages in adipose tissue. It has been reported that alternatively activated macrophages stimulate cell proliferation (9). As mentioned above, global analysis of gene expression by microarray indicated that rosigli- tazone altered expression of genes related to cell cycle, prolif- ti d ti h l (Fi 3) Th f l d FIGURE 2. Rosiglitazone increased the abundance of macrophages in adi- pose tissue of mice fed a HFD. A, representative hematoxylin-eosin (HE) and 4,6-diamidino-2-phenylindole-staining of white adipose tissue in the three experimental groups. B, representative immunohistochemical staining of white adipose tissue using the specific macrophage marker F4/80. Magni- fication is indicated at the right bottom corner. C, quantification of the % of macrophages present in adipose tissue. D, changes in adipose expression of macrophage marker genes after rosiglitazone treatment, as determined by qPCR. Gene expression levels in LFD-fed animals were set at 1. Error bars rep- resent S.E. Differences were evaluated using Student’s t test. , p 0.05; , p 0.01; , p 0.001. n 4–5 animals per group. FIGURE 3. Microarray analysis reveals changes in lipid metabolism, energy production,cellcycle,andtissuemorphologyafterrosiglitazonetreatment. Pooled white adipose tissues from mice fed a HFD or HFD plus rosiglitazone was analyzed using Affymetrix GeneChips. The top 10 list of most significantly changed pathways after rosiglitazone treatment is shown. Pathway analysis was carried out using the Ingenuity program. A -fold change above 2 or below 2 and a p value similar to or smaller then 0.001 were used as cutoff values. FIGURE 4. Rosiglitazone increased the expression of markers for alterna- tively activated macrophages in adipose tissue of mice fed a HFD. Changes in adipose gene expression in all three experimental groups, as determined by qPCR. Gene expression levels in LFD fed animals were set at 1. Error bars represent S.E. Differences are evaluated using Student’s t test. , p 0.01; , p 0.05. n 4–5 animals per group. PPAR Activation Promotes Macrophage Infiltration into WAT PPAR Activation Promotes Macrophage Infiltration into WAT expression of IL-18, a cytokine linked to the classical activation of macrophages via T helper 1 type responses (24), was d d b l l l f f b FIGURE 2. RESULTS 2B, rosiglita- JOURNAL OF BIOLOGICAL CHEMISTRY 22 AUGUST 15, 2008•VOLUME 283•NUMBER 33 22623 serum amyloid A3, and IL-6 were also down-regulated by ros- iglitazone, although the latter failed to reach statistical signifi- cance. Rosiglitazone did not alter adipose expression levels of CD11c, suggesting no change in the number of pro-inflamma- tory classically polarized macrophages (8). T h h h h l d FIGURE 2. Rosiglitazone increased the abundance of macrophages in adi- pose tissue of mice fed a HFD. A, representative hematoxylin-eosin (HE) and 4,6-diamidino-2-phenylindole-staining of white adipose tissue in the three experimental groups. B, representative immunohistochemical staining of white adipose tissue using the specific macrophage marker F4/80. Magni- fication is indicated at the right bottom corner. C, quantification of the % of macrophages present in adipose tissue. D, changes in adipose expression of macrophage marker genes after rosiglitazone treatment, as determined by qPCR. Gene expression levels in LFD-fed animals were set at 1. Error bars rep- resent S.E. Differences were evaluated using Student’s t test. , p 0.05; , p 0.01; , p 0.001. n 4–5 animals per group. FIGURE 4. Rosiglitazone increased the expression of markers for alterna- tively activated macrophages in adipose tissue of mice fed a HFD. Changes in adipose gene expression in all three experimental groups, as determined by qPCR. Gene expression levels in LFD fed animals were set at 1. Error bars represent S.E. Differences are evaluated using Student’s t test. , p 0.01; , p 0.05. n 4–5 animals per group. PPAR Activation Promotes Macrophage Infiltration into WAT i f IL 18 t ki li k d t th l i l ti ti serum amyloid A3, and IL-6 were also down-regulated by ros- iglitazone, although the latter failed to reach statistical signifi- cance. Rosiglitazone did not alter adipose expression levels of CD11c, suggesting no change in the number of pro-inflamma- tory classically polarized macrophages (8). To investigate whether the changes in total adipose tissue gene expression were specific for the stromal vascular fraction containing adipose tissue macrophages, adipose tissue from was fractionated into stromal vascular cells and adipocytes. As expected, leptin was predominantly expressed in the adipocyte fraction (Fig. 5A). In the stromal vascular fraction, rosiglitazone significantly up-regulated expression of arginase I and sIL-1ra (Fig. 5B). In contrast, rosiglitazone down-regulated expression of IL-18. DISCUSSION Obesity is associated with the infiltration of macrophages into adipose tissue, which may contribute to an elevated inflam- matory status by secreting a variety of pro-inflammatory medi- ators, including tumor necrosis factor and IL-6. MCP-1 has been identified as an important chemo-attractant responsible for the recruitment of macrophages into adipose tissue (6, 25). Previous studies suggest that PPAR may counteract obesity- induced inflammation in adipose tissue via several mechanisms (26) including down-regulation of chemo-attractant and pro- inflammatory genes (5), apoptosis of adipose-resident macro- phages (19), and changing the morphology of adipose tissue toward smaller adipocytes (27). In contrast to previous studies, in our study treatment with rosiglitazone was associated with a marked increase in the number of macrophages present in adi- pose tissue, as evidenced by (immuno)histological staining and measurement of expression of macrophages markers genes. In support of these data, we found that rosiglitazone increased expression of MCP-1 in adipose tissue leading to an influx of monocytes and macrophages. This is in contrast to previously reported data showing a reduction of MCP-1 expression in adi- pocytes after PPAR activation (28). However, results in this study were obtained using isolated human whole adipose tissue. In addition, activation of PPAR might have differential effects in obese versus lean animals. Whereas treatment of lean mice with rosiglitazone might reduce MCP-1 expression, short term PPAR activation in obese animals might lead to increased MCP-1 expression and an influx of macrophages. These mac- rophages are possibly involved in remodeling of adipose tissue to accommodate the PPAR-dependent influx of triglycerides originating from other tissues. FIGURE 5. Rosiglitazone increases the expression of markers for alterna- tivelyactivatedmacrophagesinthestromalvascularfractionofmicefed a HFD. A, selective expression of leptin in the adipocyte fraction of white adipose tissue. B, rosiglitazone-induced changes in expression of marker genes for alternatively activated macrophages in the stromal vascular frac- tion of mice fed a HFD, as determined by qPCR. Gene expression levels in HFD fed animals were set at 1. C, changes in adipose expression of cyclin A2 and cyclin B after rosiglitazone treatment, as determined by qPCR. Gene expres- sion levels in LFD-fed animals were set at 1. Error bars represent S.E. Differ- ences were evaluated using Student’s t test. , p 0.05; , p 0.01; *, p 0.001. increased by rosiglitazone, suggesting increased cell cycle pro- gression in adipose tissue of obese mice after PPAR activation (Fig. 5C). RESULTS Remarkably, expression of both genes was significantly expression of IL-18, a cytokine linked to the classical activation of macrophages via T helper 1 type responses (24), was decreased by rosiglitazone. Expression levels of fibrinogen, 22624 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283•NUMBER 33•AUGUST 15, 2008 22624 PPAR Activation Promotes Macrophage Infiltration into WAT FIGURE 5. Rosiglitazone increases the expression of markers for alterna- tivelyactivatedmacrophagesinthestromalvascularfractionofmicefed a HFD. A, selective expression of leptin in the adipocyte fraction of white adipose tissue. B, rosiglitazone-induced changes in expression of marker genes for alternatively activated macrophages in the stromal vascular frac- tion of mice fed a HFD, as determined by qPCR. Gene expression levels in HFD fed animals were set at 1. C, changes in adipose expression of cyclin A2 and cyclin B after rosiglitazone treatment, as determined by qPCR. Gene expres- sion levels in LFD-fed animals were set at 1. Error bars represent S.E. Differ- ences were evaluated using Student’s t test. , p 0.05; , p 0.01; , p 0.001. more pronounced induction was observed using conditioned medium from macrophages treated with both IL-4 and rosigli- tazone. The effect was accompanied by a significant increase in expression of adipogenic marker genes (Fig. 6B). These data suggest that rosiglitazone may deactivate the inhibitory effect of macrophages on adipocyte differentiation via a mechanism involving alternatively activated macrophages. PPAR Activation Promotes Macrophage Infiltration into WAT PPAR Activation Promotes Macrophage Infiltration into WAT ever, after HFD feeding the macro- phage-specific PPAR knock-out mice became more obese, fewer macrophages were present in adi- pose tissue (31). Supporting our data, expression of markers of alter- natively activated macrophages in adipose tissue was strongly decreased in animals lacking PPAR in macrophages, which likely accounted for the increase in inflammatory gene expression, including IL-6 and Nos2. Impor- tantly, the absence of PPAR in macrophages aggravated insulin resistance, especially after HFD. These data suggest that macro- phage PPAR plays a major role in determining macrophage polariza- tion in adipose tissue, thereby sup- pressing production of inflamma- tory mediators as well as in mediating the effect of thiazo- lidinediones on insulin sensitivity. FIGURE6.ConditionedmediumfromIL-4-androsiglitazone-treatedmacrophagesdeactivatestheinhib- itory effect of macrophages on adipogenesis. A, representative Oil Red O staining of 3T3-L1 adipocytes after 6 days of differentiation. During differentiation, 3T3-L1 cells were incubated with conditioned media from RAW264.7cells treated with vehicle, IL-4, rosiglitazone, or IL-4 and rosiglitazone. B, expression of adipogenic markergenes,asdeterminedbyqPCR.Control,mediumfromuntreatedRAW264.7macrophages;IL-4,medium from IL-4-treated RAW264.7 macrophages; Rosi, medium from rosiglitazone-treated RAW264.7 macrophages; IL-4 Rosi, medium from IL-4-and rosiglitazone-treated RAW264.7 macrophages. Gene expression levels from 3T3-L1 cells treated with medium from untreated RAW264.7 macrophages were set at 1. Error bars represent S.E. Differences were evaluated using a Students t test. p 0.05 (), p 0.01 (), and p 0.005 () compared with 3t3 adipocytes treated with control-conditioned medium. FIGURE6.ConditionedmediumfromIL-4-androsiglitazone-treatedmacrophagesdeactivatestheinhib- itory effect of macrophages on adipogenesis. A, representative Oil Red O staining of 3T3-L1 adipocytes after In contrast with previous data showing a decrease in adipose tissue macrophage content after long term rosiglitazone treatment due to apoptosis (19), our data suggest that short term treatment increases the adipose presence of alternatively activated macrophages. Future studies should address the effects of short versus long term rosiglitazone treat- ment. Whereas short term exposure to rosiglitazone induces macrophage infiltration, longer treatment might eventually lead to apoptosis of macrophages after successful PPAR-de- pendent adipose tissue remodeling and expansion. e set at 1. Error bars represent and p 0.005 (*) compared phage phenotype may be partially responsible for the suppres- sion of inflammatory gene expression by rosiglitazone. It can be hypothesized that changes in macrophage popula- tion may contribute to PPAR-dependent remodeling and expansion of adipose tissue. Alternatively activated macro- phages have been linked to tissue repair and cell proliferation. DISCUSSION ( g ) Conditioned Medium from Alternatively and PPAR-acti- vated Macrophages Deactivates the Inhibitory Effect of Macro- phages on Adipogenesis—One week of rosiglitazone treatment was associated with a significant increase in adipose tissue mass yet decreased adipocyte size. To investigate whether the effect of rosiglitazone on adipose tissue morphology and size may be partially mediated via changes in the macrophage population, the effect of alternatively activated macrophages on adipogen- esis was explored. The alternatively activated phenotype was elicited in vitro by treating RAW264.7 cells with IL-4, which induced arginase 1 and PPAR expression (data not shown). In addition, RAW264.7 cells were treated with rosiglitazone. Sub- sequently, conditioned medium from RAW264.7 macrophages treated with either IL-4 and/or rosiglitazone was added to dif- ferentiating 3T3-L1 adipocytes. Medium from untreated mac- rophages strongly inhibited 3T3-L1 adipocyte differentiation and expression of adipogenic marker genes compared with nor- mal medium (data not shown). Remarkably, compared with conditioned medium from untreated RAW264.7 cells, medium from IL-4- or rosiglitazone-treated macrophages markedly enhanced 3T3-L1 adipocyte differentiation (Fig. 6A). An even Recently, it was reported that obesity leads to polarization of adipose-resident macrophages toward a M1 phenotype. M1-polarized macrophages secrete higher levels of pro-inflam- matory genes and likely contribute to obesity-induced inflam- mation (8). The increased expression of arginase I and sIL-1ra and decreased expression of IL-18 in rosiglitazone-treated ani- mals is indicative of re-polarization of adipose macrophages toward an M2 phenotype. In contrast to our results, MCP-1 has been reported to mainly attract M1-polarized macrophages (29). Nevertheless, it has been suggested that M1 and M2 mac- rophages can switch from one phenotype to another (9). The increased adipose expression of MCP-1 after rosiglitazone treatment might indeed lead to a higher influx of M1 macro- phages. However, the local environment created by the activa- tion of PPAR might induce a switch from M1- toward M2-ac- tivated macrophages after infiltration of the adipose tissue. In as much as M2 or alternatively activated macrophages possess mainly anti-inflammatory properties, the switch in macro- JOURNAL OF BIOLOGICAL CHEMISTRY AUGUST 15, 2008•VOLUME 283•NUMBER 33 22625 FIGURE6.ConditionedmediumfromIL-4-androsiglitazone-treatedmacrophagesdeactivatestheinhib- itory effect of macrophages on adipogenesis. A, representative Oil Red O staining of 3T3-L1 adipocytes after 6 days of differentiation. During differentiation, 3T3-L1 cells were incubated with conditioned media from RAW264.7cells treated with vehicle, IL-4, rosiglitazone, or IL-4 and rosiglitazone. Acknowledgment—We thank Rene Bakker for assisting with the ani- mal experiments. PPAR Activation Promotes Macrophage Infiltration into WAT Indeed, co-culture experiments have shown that alternatively activated macrophages can positively influence proliferation of neighboring cells (30). According to our data, rosiglitazone altered expression of genes related to cell cycle, proliferation, and tissue morphology, including cyclin A2 and cyclin B. The increase in cell cycle-related genes is probably the result of increased numbers of alternatively activated macrophages in adipose tissue rather than direct stimulation of gene expression by PPAR. In conclusion, we show that short term PPAR activation in the context of HFD-induced obesity promotes macrophage infiltration into adipose tissue while simultaneously reducing inflammatory gene expression. The majority of the newly recruited macrophages is alternatively activated and may play an important role in PPAR-dependent adipose tissue remod- eling and expansion. Interestingly, we observed that conditioned medium from alternatively activated macrophages treated with rosiglitazone neutralized the inhibitory effect of conditioned medium from untreated macrophages on 3T3-L1 adipogenesis. Accordingly, it suggests that the rosiglitazone-induced switch from classical to alternatively activated macrophages may underlie the effects of rosiglitazone on adipose tissue remodeling and growth. DISCUSSION B, expression of adipogenic markergenes,asdeterminedbyqPCR.Control,mediumfromuntreatedRAW264.7macrophages;IL-4,medium from IL-4-treated RAW264.7 macrophages; Rosi, medium from rosiglitazone-treated RAW264.7 macrophages; IL-4 Rosi, medium from IL-4-and rosiglitazone-treated RAW264.7 macrophages. Gene expression levels from 3T3-L1 cells treated with medium from untreated RAW264.7 macrophages were set at 1. Error bars represent S.E. Differences were evaluated using a Students t test. p 0.05 (), p 0.01 (), and p 0.005 (*) compared with 3t3 adipocytes treated with control-conditioned medium. PPAR Activation Promotes Macrophage Infiltration into WAT 1. Hill, J. O. (2006) Endocr. Rev. 27, 750–761 PPAR Activation Promotes Macrophage Infiltration into WAT Wang, S., Fortier, M., Greenberg, A. S., and Obin, M. S. (2005) J. Lipid Res. 46, 2347–2355 20. Liu, X., Milo, M., Lawrence, N. D., and Rattray, M. (2005) Bioinformatics. 21, 3637–3644 8. Lumeng, C. N., Bodzin, J. L., and Saltiel, A. R. (2007) J. Clin. Investig. 117, 175–184 21. Dai, M., Wang, P., Boyd, A. D., Kostov, G., Athey, B., Jones, E. G., Bunney, W. E., Myers, R. M., Speed, T. P., Akil, H., Watson, S. J., and Meng, F. (2005) Nucleic Acids Res. 33, e175 9. Gordon, S. (2003) Nat. Rev. Immunol. 3, 23–35 22. Mills, C. D., Kincaid, K., Alt, J. M., Heilman, M. J., and Hill, A. M. (2000) J. Immunol. 164, 6166–6173 10. Rosen, E. D., Sarraf, P., Troy, A. E., Bradwin, G., Moore, K., Milstone, D. S., Spiegelman, B. M., and Mortensen, R. M. (1999) Mol. Cell 4, 611–617 11. Koutnikova, H., Cock, T. A., Watanabe, M., Houten, S. M., Champy, M. F., Dierich, A., and Auwerx, J. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 14457–14462 23. Martinez, F. O., Gordon, S., Locati, M., and Mantovani, A. (2006) J. Im- munol. 177, 7303–7311 Dierich, A., and Auwerx, J. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 14457–14462 24. Swain, S. L. (2001) J. Exp. Med. 194, 11–14 12. Lehrke, M., and Lazar, M. A. (2005) Cell 123, 993–999 25. Weisberg, S. P., Hunter, D., Huber, R., Lemieux, J., Slaymaker, S., Vaddi, K., Charo, I., Leibel, R. L., and Ferrante, A. W., Jr. (2006) J. Clin. Investig. 116, 115–124 13. Mandard, S., Stienstra, R., Escher, P., Tan, N. S., Kim, I., Gonzalez, F. J., Wahli, W., Desvergne, B., Muller, M., and Kersten, S. (2007) Cell. Mol. Life Sci. 64, 1145–1157 26. Stienstra, R., Duval, C., Mu¨ller, M., and Kersten, S. (2007) PPAR Res. 2007, 95974–95984 14. Oakes, N. D., Thalen, P. G., Jacinto, S. M., and Ljung, B. (2001) Diabetes 50, 1158–1165 27. Berthiaume, M., Sell, H., Lalonde, J., Gelinas, Y., Tchernof, A., Richard, D., and Deshaies, Y. (2004) Am. J. Physiol. Regul. Integr. Comp. Physiol. 287, 1116–1123 15. Tan, G. D., Fielding, B. A., Currie, J. M., Humphreys, S. M., Desage, M., Frayn, K. N., Laville, M., Vidal, H., and Karpe, F. (2005) Diabetologia 48, 83–95 28. Bruun, J. M., Lihn, A. S., Pedersen, S. B., and Richelsen, B. (2005) J. Clin. Endocrinol. Metab. 90, 2282–2289 16. Hammarstedt, A., Andersson, C. REFERENCES 1. Hill, J. O. (2006) Endocr. Rev. 27, 750–761 2. Reaven, G. M. (2006) Am. J. Clin. Nutr. 83, 1237–1247 PPAR is well expressed in adipocytes and macrophages. From our study it is not possible to derive any conclusions regarding the role of macrophages versus adipocyte PPAR in mediating the effect of rosiglitazone on total macrophage abun- dance as well as on macrophage polarization. While this manu- script was in preparation, two reports were published address- ing the role of macrophage PPAR in the development of obesity-induced inflammation and insulin resistance using macrophage-specific PPAR knock-out mice (31, 32). How- 3. Hotamisligil, G. S. (2006) Nature 444, 860–867 4. Weisberg, S. P., McCann, D., Desai, M., Rosenbaum, M., Leib Ferrante, A. W., Jr. (2003) J. Clin. Investig. 112, 1796–1808 5. Xu, H., Barnes, G. T., Yang, Q., Tan, G., Yang, D., Chou, C. J., Sole, J., Nichols, A., Ross, J. S., Tartaglia, L. A., and Chen, H. (2003) J. Clin. Investig. 112, 1821–1830 6. Kanda, H., Tateya, S., Tamori, Y., Kotani, K., Hiasa, K., Kitazawa, R., Kitazawa, S., Miyachi, H., Maeda, S., Egashira, K., and Kasuga, M. (2006) J. Clin. Investig. 116, 1494–1505 6. Kanda, H., Tateya, S., Tamori, Y., Kotani, K., Hiasa, K., Kitazawa, R., Kitazawa, S., Miyachi, H., Maeda, S., Egashira, K., and Kasuga, M. (2006) J. Clin. Investig. 116, 1494–1505 7. Cinti, S., Mitchell, G., Barbatelli, G., Murano, I., Ceresi, E., Faloia, E., 7. Cinti, S., Mitchell, G., Barbatelli, G., Murano, I., Ceresi, E., Faloia, E., 22626 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283•NUMBER 33•AUGUST 15, 2008 22626 VOLUME 283•NUMBER 33•AUGUST 15, 2008 PPAR Activation Promotes Macrophage Infiltration into WAT X., Rotter, S. V., and Smith, U. (2005) Prostaglandins Leukot. Essent. Fatty Acids 73, 65–75 29. Charo, I. F. (2007) Cell Metab. 6, 96–98 17. Okuno, A., Tamemoto, H., Tobe, K., Ueki, K., Mori, Y., Iwamoto, K., Umesono, K., Akanuma, Y., Fujiwara, T., Horikoshi, H., Yazaki, Y., and Kadowaki, T. (1998) J. Clin. Investig. 101, 1354–1361 30. Song, E., Ouyang, N., Horbelt, M., Antus, B., Wang, M., and Exton, M. S. (2000) Cell Immunol. 204, 19–28 31. Odegaard, J. I., Ricardo-Gonzalez, R. R., Goforth, M. H., Morel, C. R., Subramanian, V., Mukundan, L., Eagle, A. R., Vats, D., Brombacher, F., Ferrante, A. W., and Chawla, A. (2007) Nature 447, 1116–1120 31. Odegaard, J. I., Ricardo-Gonzalez, R. R., Goforth, M. H., Morel, C. R., Subramanian, V., Mukundan, L., Eagle, A. R., Vats, D., Brombacher, F., 18. Jernas, M., Palming, J., Sjoholm, K., Jennische, E., Svensson, P. A., Gabri- elsson, B. G., Levin, M., Sjogren, A., Rudemo, M., Lystig, T. C., Carlsson, B., Carlsson, L. M., and Lonn, M. (2006) FASEB J. 20, 1540–1542 rrante, A. W., and Chawla, A. (2007) Nature 447, 1116–112 32. Hevener, A. L., Olefsky, J. M., Reichart, D., Nguyen, M. T., Bandyopady- hay, G., Leung, H. Y., Watt, M. J., Benner, C., Febbraio, M. A., Nguyen, A. K., Folian, B., Subramaniam, S., Gonzalez, F. J., Glass, C. K., and Ricote, M. (2007) J. Clin. Investig. 117, 1658–1669 19. Bodles, A. M., Varma, V., Yao-Borengasser, A., Phanavanh, B., Peterson, C. A., McGehee, R. E., Jr., Rasouli, N., Wabitsch, M., and Kern, P. A. (2006) J. Lipid Res. 47, 2080–2088 JOURNAL OF BIOLOGICAL CHEMISTRY 22627 AUGUST 15, 2008•VOLUME 283•NUMBER 33
https://openalex.org/W1964272614	https://europepmc.org/articles/pmc2082012?pdf=render	English	null	Dual antiplatelet therapy and drug eluting stents: a marriage of convenience	Thrombosis journal	2,007	cc-by	2,820	Open Review Dual antiplatelet therapy and drug eluting stents: a marriage of convenience Juan Benezet-Mazuecos, Borja Ibanez and Juan J Badimon* Open Access Address: Cardiovascular Biology Research Laboratory, Cardiovascular Institute, Mount Sinai School of Medicine, New York, NY, USA Email: Juan Benezet-Mazuecos - jbenezet@yahoo.es; Borja Ibanez - borja.ibanez@mssm.edu; Juan J Badimon* - juan.badimon@mssm.edu * Corresponding author Received: 11 September 2007 Accepted: 1 October 2007 © 2007 Benezet-Mazuecos et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. less-steel stent with a thin layer of a non-erodable syn- thetic polymer containing either sirolimus or paclitaxel respectively. DES implantation resulted in a rate of reste- nosis below 10% (compared to 30% observed after bare metal stents implantation) [4-6], with a similar rate of in- stent thrombosis. [7,8] It is important to remind that dual antiplatelet therapy was mandatory for at least 3–6 months after DES implantation. However some initial warns were raised by few groups suggesting that DES do not undergo a complete re-endothelization. [9,10] It was just a matter of time to face, with striking surprise, the unexpected high rate of late (>30 days after stent deploy- ment) and especially very late in-stent thrombosis (beyond 12 months from stent implantation) in DES compared to bare metal stents. [11,12] Further cause for concern came from the Basel Stent Cost-effectiveness Trial-Late Thrombotic Events (BASKET-LATE) data, which showed that among 746 DES or BMS patients who had dual antiplatelet therapy discontinued after the first six months, the rate of cardiac death or non-fatal MI over the following year was higher in patients with DES than BMS (4.9% versus 1.3%; p = 0.01), and that this was likely to be related to late stent thrombosis, which presented as death or MI in 88% of cases. [13] After this first report, a heated discussion began over the safety of DES, especially when the protective umbrella of dual antiplatelet therapy is discontinued. Dr. Camenzind presented in the WCC- 2006 in Spain a meta-analysis of first generation DES in comparison to BMS accounting for a total of n = 878 sirolimus eluting stent (SES) vs n = 870 BMS (RAVEL, SIR- IUS, C-SIRIUS and E-SIRIUS) and n = 1685 paclitaxel- eluting stents (PES) vs n = 1675 BMS (TAXUS II, IV-VI). BioMed Central BioMed Central Page 1 of 3 (page number not for citation purposes) http://www.thrombosisjournal.com/content/5/1/15 http://www.thrombosisjournal.com/content/5/1/15 total mortality and Q-wave MI combined were 38% (SES) and 16% (PES) higher in DES as compared to control BMS (p-value: SES vs BMS: 0.03 ; PES vs BMS. 0.68). [14] Over- all, it seems that experience has shown that the success of DES comes at a price: patients using DES could be trading restenosis, which is seldom life-threatening, for in-stent thrombosis, which may lead to death and myocardial inf- arction. Moreover, a number of reports imply that throm- bosis rates of DES may even be higher in the "real world" than in clinical trials where stents are used "on-label". Today it is considered that as much as 60% of coronary percutaneous procedures are "off-label". When DES are used "off-label", it is estimated that the rates of thrombo- sis are higher. On the other hand, although data from large registries and meta-analyses of randomized trials indicate a higher risk for DES thrombosis, others recent studies claim the absence of such increased risk. [15] The controversy is open. genicity. Many factors influence the healing process and vary with individual risk factors. Coating of stents with substances that potentially facilitate reendothelialization may represent a novel therapeutic approach. Significant advances in material engineering are poised to produce biodegradable stents. [18] Several risk factors correlate with "vulnerable/hyperreac- tive blood" (diabetes, smoking, hyperlipidemia, inflam- matory states, cathecholamines...). Also, several patient- related factors have been associated with the development of in-stent thrombosis, including low ejection fraction, diabetes mellitus, advanced age, renal failure, and acute coronary syndromes. In particular, in this last situation it could be due to delayed healing, lack of endothelializa- tion, and presence of a pronounced inflammatory and thrombogenic environment combined by enhanced platelet reactivity. Furthermore, certain lesion characteris- tics (bifurcation complex or in-stent restenosis lesions) are reported to be associated with an increased risk of stent thrombosis. [11,19] Three are the major players involved in the pathological processes leading to late in-stent thrombosis: the stent coated with an antiproliferative drug, the vulnerability (thrombogenicity) of the patient's blood, and the antiplatelet therapy. The majority of the retrospective studies on the rate of late in-stent thrombosis have identified the window of 6 to 9 months post stenting as the peak of more incidence. The majority of the studies using multivariate analysis have identified the discontinuation of dual antithrombotic therapy as the principal parameter responsible for stent thrombosis. http://www.thrombosisjournal.com/content/5/1/15 [20,21] Stents are foreign bodies in the vessel wall and thus induce platelet adhesion and activation of the coagulation cascade. Furthermore, high-pressure implantation induces significant vascular injury, with exposure of thrombogenic molecules of the subintima and media (including plaque material) to the blood stream. As a consequence, only potent platelet inhibition made the procedure feasible. [22] The association between late thrombosis with delayed stent coverage and cessation of dual antithrombotic therapy strongly suggest the need for a longer period of combination therapy. On the other hand, longer administration of dual therapy is hampered by its higher cost, patient compliance and the possibility of increased bleeding complications. The appropriate duration of the long-term antiplatelet regi- men for prevention of DES thrombosis remains to be assessed in randomized prospective trials; at present, a course of 12 months of dual-antiplatelet therapy may be considered. [23] Several factors related with the stents are associated with an increased risk of thrombosis, including the procedure itself (stent malapposition and/or underexpansion, number of implanted stents, length, persistent slow coro- nary blood flow, and dissections) and stent design (mate- rials, strut thickness and polymer type). The coating of stents with antiproliferative (sirolimus) and/or cytostatic (paclitaxel) drugs should have the additional advantage of inhibiting the proliferation of vascular cells. [10] Reendothelialization occurs after vascular injury and sim- ilarly after stent placement. But then two questions arise since damaged and/or dysfunctional endothelium is directly related with the development and progression of atherosclerosis [16,17]: Could the DES selectively inhibit the proliferation and migration of smooth muscle cells without affecting the growth of the endothelial cells? And could it be possible for DES to exert their antirestenotic effects without affecting the neo-endothelialization of the struts? In vitro, rapamycin and paclitaxel not only inhibit proliferation and migration of vascular smooth muscle cells but equally suppress endothelial cells, thereby poten- tially impairing reendothelialization. A post-mortem study comparing coronary segments from patients after DES and BMS implantation revealed delayed arterial heal- ing and poorer endothelialization after DES compared with BMS implantation of similar duration. [10] Thus, current evidence suggests delayed reendothelialization and arterial healing after implantation of DES compared with BMS, resulting in potentially enhanced thrombo- Therefore, in an example of individualized medicine, many factors should be taken into account for the selec- tion of the type of stent to be implanted. Open Review Dual antiplatelet therapy and drug eluting stents: a marriage of convenience Juan Benezet-Mazuecos, Borja Ibanez and Juan J Badimon* The incidence – up to the latest available follow-up – of In 1977, a major breakthrough in cardiovascular medi- cine was introduced by Dr. Andreas Gruentzig. In Septem- ber of that year, Gruentzig performed the very first percutaneous transluminal coronary angioplasty (PTCA) procedure in a human. Twenty-three years later, the coro- nary artery was still patent. [1] Despite the initial enthusi- asm, some complications were clearly associated with the procedure, being restenosis and thrombosis the major problems to deal with. In the early nineties, bare metal stents (BMS) were introduced to avoid the so called "elas- tic recoil of the artery" leading to restenosis. Despite some improvements in the rates of restenosis and thrombosis, they were still very significant. The introduction of dual antiplatelet therapy (thyenopiridine on top of aspirin) substantially decreased the rate of stent thrombosis, being currently a rare complication occurring in less than 1% of the procedures. [2] Despite the "elastic recoil" of the artery was abrogated with the stent deployment, there was still a non-deprecia- ble rate of in-stent restenosis due to the neointimal prolif- eration inside the stents. In an attempt to fight the exaggerated proliferation after coronary stent implanta- tion, our lab first used rapamycin in an animal model of restenosis showing the effectiveness of this antiprolifera- tive drug in preventing this neointimal proliferation. [3] Soon after, stents locally delivering antiproliferative drugs (the so called drug eluting stents [DES]) were designed and tested in humans with impressive results in terms of in-stent restenosis inhibition. Two drug-eluting stents (DES), the Cypher stent (Cordis/Johnson & Johnson) and Taxus stent (Boston Scientific), were introduced. The Cypher and Taxus stents are produced by coating a stain- Page 1 of 3 (page number not for citation purposes) Page 1 of 3 (page number not for citation purposes) Thrombosis Journal 2007, 5:15 References References 1. King SB 3rd, Meier B: Interventional treatment of coronary heart disease and peripheral vascular disease. Circulation 2000, 102:IV81-86. 16. Vanhoutte PM: Endothelial dysfunction and atherosclerosis. Eur Heart J 1997:19-29. J 17. Corti R, Badimon JJ: Value or desirability of hemorheological- hemostatic parameter changes as endpoints in blood lipid- regulating trials. Curr Opin Lipidol 2001, 12:629-37. 2. Bertrand ME, Rupprecht HJ, Urban P, Gershlick AH: Double-blind study of the safety of clopidogrel with and without a loading dose in combination with aspirin compared with ticlopidine in combination with aspirin after coronary stenting: the clopidogrel aspirin stent international cooperative study (CLASSICS). Circulation 2000, 102:624-629. g g p p 18. Luscher TF, Steffel J, Eberli FR, Joner M, Nakazawa G, Tanner FC, Vir- mani R: Drug-eluting stent and coronary thrombosis: biologi- cal mechanisms and clinical implications. Circulation 2007, 115:1051-1058. 19. Park DW, Park SW, Park KH, Lee BK, Kim YH, Lee CW, Hong MK, Kim JJ, Park SJ: Frequency of and risk factors for stent throm- bosis after drug-eluting stent implantation during long-term follow-up. Am J Cardiol 2006, 98:352-356. ( ) 3. Gallo R, Padurean A, Jayaraman T, Marx S, Roque M, Adelman S, Che- sebro J, Fallon J, Fuster V, Marks A, Badimon JJ: Inhibition of intimal thickening after balloon angioplasty in porcine coronary arteries by targeting regulators of the cell cycle. Circulation 1999, 99:2164-2170. p J 20. McFadden EP, Stabile E, Regar E, Cheneau E, Ong AT, Kinnaird T, Sud- dath WO, Weissman NJ, Torguson R, Kent KM, Pichard AD, Satler LF, Waksman R, Serruys PW: Late thrombosis in drug-eluting coronary stents after discontinuation of antiplatelet therapy. Lancet 2004, 364:1519-1521. , 4. Smith SC, Dove JT, Jacobs AK, Kennedy JW, Kereiakes D, Kern MJ, Kuntz RE, Popma JJ, Schaff HV, Williams DO, Gibbons RJ, Alpert JP, Eagle KA, Faxon DP, Fuster V, Gardner TJ, Gregoratos G, Russell RO, Smith SC: ACC/AHA guidelines of percutaneous coronary interventions (revision of the 1993 PTCA guidelines) – exec- utive summary. A report of the American College of Cardi- ology/American Heart Association Task Force on Practice Guidelines (committee to revise the 1993 guidelines for per- cutaneous transluminal coronary angioplasty). J Am Coll Car- diol 2001, 37:2215-2239. 21. Eisenstein EL, Anstrom KJ, Kong DF, Shaw LK, Tuttle RH, Mark DB, Kramer JM, Harrington RA, Matchar DB, Kandzari DE, Peterson ED, Schulman KA, Califf RM: Clopidogrel use and long-term clinical outcomes after drug-eluting stent implantation. http://www.thrombosisjournal.com/content/5/1/15 Among these fac- tors, the patient characteristics (diabetes, left ventricular ejection fraction ...), the severity of the disease and the characteristics of the lesions, and finally the possibility and/or compliance to a prolonged antithrombotic treat- Page 2 of 3 (page number not for citation purposes) Page 2 of 3 (page number not for citation purposes) Thrombosis Journal 2007, 5:15 http://www.thrombosisjournal.com/content/5/1/15 http://www.thrombosisjournal.com/content/5/1/15 ment. What seem to be clear is that a longer period of dual anti-platelet therapy is needed for patients undergoing DES implantation than for those receiving a bare metal stent. However, at this point it is completely unknown how longer that period should be. Despite the current rec- ommendations from the AHA/ACC suggest 12 months of dual antiplatelet therapy, some experts highly suggest a chronic dual antiplatelet therapy until new data with longer follow up periods is available. LATE investigators: Late clinical events after clopidogrel dis- continuation may limit the benefit of drug-eluting stents: an observational study of drug-eluting versus bare-metal stents. J Am Coll Cardiol 2006, 48:2584-91. 14. Camenzind E, Steg PG, Wijns W: Safety of drug-eluting stents: a metaanalysis of 1st generation DES programs. World Congres- sof Cardiology 2006, September 2–6, 2006, Barcelona, Spain [http:www.escardio.org/knowledge/congresses/CongressReports/ hotline sandctus/707009_Camenzind.htm]. 15. 15. Marzocchi A, Saia F, Piovaccari G, Manari A, Aurier E, Benassi A, Cre- monesi A, Percoco G, Varani E, Magnavacchi P, Guastaroba P, Grilli R, Maresta A: Long-term safety and efficacy of drug-eluting stents two-year results of the REAL (Registro angioplastiche dell'Emilia Romagna) multicenter registry. Circulation 2007, 115:3181-3188. References Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Page 3 of 3 (page number not for citation purposes) Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral Publish with BioMed Central and every scientist can read your work free of charge References JAMA 2006, 297:159-168. 22. Wenaweser P, Dorffler-Melly J, Imboden K, Windecker S, Togni M, Meier B, Haeberli A, Hess OM: Stent thrombosis is associated with an impaired response to antiplatelet therapy. J Am Coll Cardiol 2005, 45:1748-1752. 5. 5. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin M, Colombo A, Schuler G, Barragan P, Guagliumi G, Molnar F, Falotico R: A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med 2002, 346:1773-1780. 23. 23. Mehta SR, Yusuf S, Peters RJ, Bertrand ME, Lewis BS, Natarajan MK, Malmberg K, Rupprecht H, Zhao F, Chrolavicius S, Copland I, Fox KA, Clopidogrel in unstable angina to prevent recurrent events trial (CURE) Investigators: Effects of pretreatment with clopidogrel and aspirin followed by long-term therapy in patients under- going percutaneous coronary intervention: the PCI-CURE study. Lancet 2001, 358:527-533. 6. Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O'Shaugh- nessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein PS, Jaeger JL, Kuntz RE: Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003, 349:1315-1323. 7. Moreno R, Fernandez C, Hernandez R, Alfonso F, Angiolillo DJ, Sabate M, Escaned J, Banuelos C, Fernandez-Ortiz A, Macaya C: Drug-eluting stent thrombosis: results from a pooled analy- sis including 10 randomized studies. J Am Coll Cardiol 2005, 45:954-959. 8. Babapulle MN, Joseph L, Belisle P, Brophy JM, Eisenberg MJ: A hier- archical Bayesian meta-analysis of randomised clinical trials of drug-eluting stents. Lancet 2004, 364:583-591. g g 9. Takano M, Ohba T, Inami S, Seimiya K, Sakai S, Mizuno K: Angio- scopic differences in neointimal coverage and in persistence of thrombus between sirolimus-eluting stents and bare metal stents after a 6-month implantation. Eur Heart J 2006, 27:2189-2195. Publish with BioMed Central and every scientist can read your work free of charge 10. Joner M, Finn AV, Farb A, Mont EK, Kolodgie FD, Ladich E, Kutys R, Skorija K, Gold HK, Virmani R: Pathology of drug-eluting stents in humans: delayed healing and late thrombotic risk. J Am Coll Cardiol 2006, 48:193-202. 11. Iakovou I, Schmidt T, Bonizzoni E, Ge L, Sangiorgi GM, Stankovic G, Airoldi F, Chieffo A, Montorfano M, Carlino M, Michev I, Corvaja N, Briguori C, Gerckens U, Grube E, Colombo A: Incidence, predic- tors, and outcome of thrombosis after successful implanta- tion of drug-eluting stents. JAMA 2005, 293:2126-2130. g g J 12. Ong AT, McFadden EP, Regar E, de Jaegere PP, van Domburg RT, Ser- ruys PW: Late angiographic stent thrombosis (LAST) events with drug-eluting stents. J Am Coll Cardiol 2005, 45:2088-2092. g g J 13. Pfisterer M, Brunner-La Rocca HP, Buser PT, Rickenbacher P, Hun- ziker P, Mueller C, Jeger R, Bader F, Osswald S, Kaiser C, BASKET-
https://openalex.org/W4285798759	https://health-policy-systems.biomedcentral.com/counter/pdf/10.1186/s12961-022-00885-4	English	null	The collaborative research and service delivery partnership between the United States healthcare system and the U.S. Military Health System during the COVID-19 pandemic	Health research policy and systems	2,022	cc-by	6,260	© The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Koehlmoos et al. Health Research Policy and Systems (2022) 20:81 https://doi.org/10.1186/s12961-022-00885-4 Koehlmoos et al. Health Research Policy and Systems (2022) 20:81 https://doi.org/10.1186/s12961-022-00885-4 Koehlmoos et al. Health Research Policy and Systems (2022) 20:81 https://doi.org/10.1186/s12961-022-00885-4 Open Access Abstract Objectives: To examine the military-civilian collaborative efforts which addressed the unprecedented challenges of the COVID-19 pandemic, particularly in areas including provision of supplies, patient and provider support, and devel‑ opment and dissemination of new vaccine and drug candidates. Methods: We examined peer reviewed and grey literature from September 2020 to June 2021 to describe the relationship between the U.S. healthcare system and Military Health System (MHS). For analysis, we applied the World Health Organization framework for health systems, which consists of six building blocks. Results: The strongest collaborative efforts occurred in areas of medicine and technology, human resources, and healthcare delivery, most notably in the MHS supplying providers, setting up treatment venues, and participating in development of vaccines and therapeutics. Highlighting that the MHS, with its centralized structure and ability to deploy assets rapidly, is an important contributor to the nation’s ability to provide a coordinated, large-scale response to health emergencies. Conclusions: Continuing the relationship between the two health systems is vital to maintaining the nation’s capa‑ bility to meet future health challenges. Keywords: Military Health System, COVID-19 pandemic response, U.S. Healthcare System Keywords: Military Health System, COVID-19 pandemic response, U.S. Healthcare System which has struggled with bed capacity, provider availabil- ity, and supply chain issues to meet the needs of patients [3]. Provision of civilian health services is one of the largest industries in the U.S., composed of a fragmented delivery system organized at the local level and funded by a mix of private and federally funded insurance pro- grams [4, 5]. While the U.S. civilian healthcare system boasts a multitude of strengths, including its large highly trained workforce, wide range of medical specialists, secondary and tertiary institutions, cutting edge medi- cal equipment, and for some medical services, unparal- leled health outcomes, it is frequently overburdened and Introductionh The U.S.COVID-19 pandemic has caused more than one million deaths in the United States (U.S.) as of June 2022 [1]. It was first declared a public health emergency (PHE) on January, 31, 2020 [2]. Since then, its rapid and con- tinued spread has posed challenges in nearly all facets of the U.S. economy, most notably in the healthcare sector, *Correspondence: miranda.janvrin.ctr@usuhs.edu 1 Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814, United States of America Full list of author information is available at the end of the article 1 Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814, United States of America Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Koehlmoos et al. Health Research Policy and Systems (2022) 20:81 Page 2 of 7 throughout the COVID-19 PHE. The six building blocks of a health system, as defined by the WHO, and shown in Table 1 below, were used to frame our findings and dis- cussion [16]. cost-inefficient even during normal operations [6, 7]. The additional challenges posed by the COVID-19 PHE spurred collaborative efforts by the U.S. civilian health- care sector with the U.S. Military Health System (MHS) in the national response to the COVID-19 pandemic [8, 9]. Such collaboration has traditionally occurred through healthcare training programs, or in times of natural or other disasters, though on a much smaller scale [10–13].h Resources were identified through scanning the electronic search engines (Google, Twitter, PubMed) between September 2020 and June 2021 using terms specific to each health systems building block for the U.S. and MHS (see Additional file 1). Further, snowball sampling was employed during the literature review pro- cess within each building block to identify additional resources for review and inclusion. Studies were organ- ized and described within the building blocks of health systems with an eye at drawing conclusions around the inter-relationship of the MHS and U.S. health systems. Governance In comparison to the centralized governance approach guided by the National Security Strategy, the U.S. health services system is highly fragmented with many organi- zations involved in its governance, including Congress, the Department of Health & Human Services (DHHS), state and local public health departments, and the private sector healthcare insurers and systems [4, 5]. Governance of each body sculpted their response to the COVID-19 PHE, with the U.S. civilian healthcare system response largely shaped at the state or local level for initial deci- sions on lock down/stay at home directives, access to testing, and mask-wearing guidelines and messaging [17, 18]. In contrast, the more centralized nature of the MHS led to more consistent messaging, including the rollout of standardized language and guidelines around Force Health Protection Conditions, using the provider-based Clinical Communities Advisory Council as a bi-weekly sounding board [19]. Additionally, the development of MHS clinical practice guidelines for the treatment of COVID-19 was built on the existing platforms for the Joint Trauma System, enabling rapid development and dissemination of guidelines on a system-wide level [20]. The MHS is a complex system that weaves together healthcare delivery, medical education, public health, private sector partnerships, and cutting edge medical research and development with personnel and infra- structure from the Military Departments (Army, Navy, and Air Force) and the office of the Assistant Secretary of Defense (e.g., the Defense Health Agency [DHA], and the Uniformed Services University of the Health Sci- ences.)[14] The MHS serves approximately 9.6 million beneficiaries, of whom approximately 17% are active- duty service members from every military branch, with the remainder being non-active-duty dependents and retirees, representative of the U.S. population [15]. The MHS is focused on deployment readiness with the ability to mobilize assets and resources efficiently in support of national public health emergencies. This paper explores the inter-relationship between the U.S. healthcare sys- tem and the MHS through the lens of a health systems approach. The World Health Organization’s (WHO) Health Systems Framework breaks down a health system into six building blocks: (1) governance, (2) medicine and technology, (3) human resources, (4) healthcare delivery, (5) finances, and (6) information [16]. A health systems approach enables a comprehensive and detailed review organized according to these building blocks, and illus- trates lessons learned from the MHS support to the U.S. civilian health system during a time of global pandemic. Methods A review of the peer-reviewed and grey literature was conducted to examine the inter-relationship between the U.S. non-federal healthcare services and the MHS Strategic policy frameworks combined with effective oversight, coalition-building, regulation, attention to system design and accountability Equitable access to essential medical products, vaccines and technologies or assured quality, safety, efficacy and cost-effectiveness, and their scientifically sound and cost- effective use All personnel involved in the process of providing care and running healthcare facilities Coverage, access, quality, and safety of health services Funds necessary to ensure people have access to needed health services Technology used to store, transmit, and use healthcare information Medicine and technology Medicine and technology have been vital aspects of the nation’s response to the COVID-19 public health emer- gency, most notably in Operation Warp Speed (OWS), a $10 billion initiative which aimed to forge “a partner- ship between DHHS, the Department of Defense (DoD), and the private sector focused on developing, producing, and distributing materials for testing, prevention, and therapy on a rapid timeline.”[8] Specifically, the role of the DoD as a co-lead was to provide leadership through logistical expertise, including program management and contracting proficiency [22]. The DoD also contributed to research efforts alongside manufacturers from the U.S. healthcare sector, including clinical trial testing at MHS operated military treatment facilities [23]. Notably, the U.S. Army Research Institute for Infectious Diseases, in conjunction with pharmaceutical manufacturer Gilead, performed necessary testing for Gilead’s existing drug Remdesivir to receive Food and Drug Administration approval as a potential treatment for COVID-19 [24, 25]. As of June 2021, three vaccines had been awarded emer- gency use authorizations, and were distributed to adults and children over 12 across the U.S [26]. As of June 2021, OWS contributed to the distribution of over 372 million vaccines nationwide, including 4 million vaccines to DoD personnel [27, 28]. Aside from OWS, the military has also engaged in research of its own showing how medi- cal practices in the military can inform civilian public health practices, a specific example being the use of mili- tary experience with different testing strategies and ways to mitigate the spread of the virus in close quarters and applying such experience to civilian college dormitories, prisons, and sports training environments [29]. The MHS also faced staffing challenges of its own, with the Army reaching out to retired military doctors and medics to return to service and provide care in Military Treatment Facilities (MTFs), while troops were deployed to field settings to combat the spread of COVID-19 both in the U.S. and abroad [38–40]. Additionally, the MHS determined that there were not enough trained personnel to fill key pandemic response leadership roles and that more training was warranted to produce these individu- als [41]. The cooperation between U.S. civilian healthcare workers and military healthcare workers has proven cru- cial in ensuring adequate healthcare provider staffing lev- els throughout the pandemic [42]. Medicine and technology In October 2020, the death toll from COVID-19 among nurses equaled that among nurses in World War I even with data available from only a quarter of countries and the numbers have only increased since [43]. Table 1 Description of WHO Building Blocks Used in this Analysis Table 1 Description of WHO Building Blocks Used in this Analysis Table 1 Description of WHO Building Blocks Used in this Analysis Table 1 Description of WHO Building Blocks Used in this Analysis Table 1 Description of WHO Building Blocks Used in this Analysis Governance Strategic policy frameworks combined with effective oversight, coalition-building, regulation, attention to system design and accountability Medicine and Technology Equitable access to essential medical products, vaccines and technologies or assured quality, safety, efficacy and cost-effectiveness, and their scientifically sound and cost- effective use Human Resources All personnel involved in the process of providing care and running healthcare facilities Healthcare Delivery Coverage, access, quality, and safety of health services Healthcare Financing Funds necessary to ensure people have access to needed health services Healthcare Information Technology used to store, transmit, and use healthcare information Page 3 of 7 Page 3 of 7 Koehlmoos et al. Health Research Policy and Systems (2022) 20:81 The MHS provided coordinated support on request to state and local-level civilian authorities through the joint doctrine Defense Support To Civil Authorities (DSCA) [21]. As a result, the MHS retained its centralized com- munications but generally followed state-level guidance. care team and four, seven-person Rapid Rural Response Teams from the Navy were activated to support efforts in Texas [34]. Additionally, two U.S. Naval Ships, the U.S.N.S. Comfort and U.S.N.S. Mercy, deployed to New York and Los Angeles respectively providing healthcare personnel [35]. The mission was originally to provide care to non-COVID-19 patients, however, the U.S.N.S. Com- fort soon took on infected patients. Governors across numerous states mobilized National Guard members to bolster vaccination efforts [36]. Overall, over 20,000 National Guardsmen deployed in 52 states and territories and other military personnel are prepared to deploy and respond within 48 h’ notice to any requests from DHHS [37]. Healthcare delivery Intricately linked to its role in human resources, the MHS has provided support to the U.S. civilian health- care system through enabling additional delivery of ser- vices. Outpatient appointments, inpatient admissions, and surgeries dramatically decreased among patients with commercial insurance, Medicare, Medicaid, and the MHS-provided TRICARE [44–46]. While this freed some bed capacity to care for COVID-19 patients, the U.S. has fewer physicians and hospital beds per capita compared to other countries already overwhelmed by COVID-19, and a higher population of patients who required intensive care [33, 47]. The MHS was able to aid the civilian healthcare system by setting up field hospitals to care for civilian patients [48].h Human resourcesh The U.S. has a highly skilled workforce, but staff short- ages exist and are expected to grow in the coming years [30, 31]. The COVID-19 pandemic exacerbated this staff- ing shortage due to increased volume of patients and ill- ness among providers [32, 33]. Critical staffing shortages drove state and local civilian authorities to seek federal assistance through DoD provider support to civilian hos- pitals. Approximately 160 Air Force medical and support personnel deployed to California, and 580 Army and Navy medical and support personnel deployed to Texas [34]. An additional team of 85 personnel from the U.S. Army Urban Augmentation Task Force, 44-person acute This support was provided in part through joint doc- trine of DSCA as described in Governance, above [21]. Three major examples were the standup of the field hos- pital at the Jacob J. Javitz Convention Center in New York City in June 2020, which enabled treatment of approxi- mately 1100 patients, the deployment of the two Naval Page 4 of 7 Koehlmoos et al. Health Research Policy and Systems (2022) 20:81 Ships, U.S.N.S. Comfort and U.S.N.S. Mercy, to New York and Los Angeles respectively in May 2020 and the admission of non-COVID-19 patients outside the MHS to William Beaumont Army Medical Center in order to free beds in other El Paso hospital facilities in October 2020 [35, 49–51]. Ships, U.S.N.S. Comfort and U.S.N.S. Mercy, to New York and Los Angeles respectively in May 2020 and the admission of non-COVID-19 patients outside the MHS to William Beaumont Army Medical Center in order to free beds in other El Paso hospital facilities in October 2020 [35, 49–51]. pandemic encouraged more data sharing than previously existed. The Tiberius platform developed by Palantir Technolo- gies specifically for OWS response to the COVID-19 pandemic, is used to collect, correlate, and visualize data across OWS for military and civilian partners spanning the MHS as well as civilian public health and medical networks. This platform integrates data on manufactur- ing, supply chain, allocation, state and territory planning, and administration of the vaccine, however, no identifi- able data or personal health information is contained in the platform [41]. Discussionh The MHS, serves a representative population of Ameri- cans and is a crucial player in the multipronged approach to filling in the gaps in the U.S. civilian healthcare sector that have been exposed during both the PHE and pan- demic as a whole. The timing and variation in the spread of COVID-19 across the U.S. has required substantial agility in the mobilization of resources, which the mili- tary medical forces were equipped to provide. This capa- bility is supported by a well-developed research capacity, which contributed significantly to the development and testing of vaccines and therapeutics, and a dedicated logistics infrastructure which enabled rapid delivery of vaccines during OWS [22, 52]. Overall, the MHS played a vital role in the U.S. civilian healthcare sector response to the COVID-19 pandemic, with the most pronounced collaborative efforts seen in human resources, healthcare delivery, and medicine and technology. Healthcare financingh i g The U.S. civilian healthcare system consumes 17.8% of the GDP and is the most expensive per capita in the world [53]. The MHS normally costs approximately $52 billion per year and has been holding steady at 10% of the DoD total budget since 2010 [54]. The COVID-19 pandemic took a large toll on the financing of the U.S. civilian healthcare system. While demand for specialty care to treat COVID-19 increased, demand for routine services decreased dramatically leading to reduced rev- enues for providers and hospitals. U.S. civilian hospi- tals were estimated to lose $323.1 billion in 2020 [55]. Additionally, many individuals who were laid off due to the pandemic lost their health insurance coverage and were left unable to pay to receive medical care [55]. The MHS also reduced healthcare expenditures during the pandemic due to the reduction in direct and purchased care utilization, but for the MHS this is positive savings [56]. However, there were still issues centered around financial lines of communication within the DHA and service resources. Issues occurred when it was not clear which party was responsible for funding, creating spend- ing plans, submitting and coordinating funding requests, determining the appropriate funding source, or creating expense reports. In some cases, these financial issues caused mission delays [56]. No collaborative efforts were found during this time period between the MHS and the U.S. civilian healthcare system regarding financing. Sharing of EHR data has been critically important when treating complex COVID-19 patients however, the majority of EHR systems are not interconnected or inter- operable. To overcome this, the Joint Health Informa- tion Exchange, which is a military based data exchange, expanded to allow bidirectional data sharing from Com- monWell Health Alliance, a civilian entity spanning over 15,000 hospitals and clinics [58, 59]. This data shar- ing allows for more efficient treatment of COVID-19 patients. Human resourcesh Another useful platform, GeoHealth, a geographic information system application sponsored by the Office of the Assistant Secretary of Defense for Pre- paredness and Response, is used by military and civilian entities to examine burden of disease, population-level risk factors, and bed and ventilator capacity of hospitals at the county level across the U.S [57]. Both platforms have proven invaluable in tracking capacity for treatment and prevention of COVID-19. The MHS partnered with civilian entities through OWS, providing expertise in the areas of operational planning, logistics and supply chains [22, 52]. The MHS leveraged its contracting connections to obtain deals with the six manufacturers, three of which have devel- oped vaccines currently in use [22]. The MHS also sup- ported vaccination distribution logistics and provided manpower [22, 52]. These examples show how the rapid deployment of MHS personnel and military assets pro- vided vital support to the U.S. civilian healthcare system during the pandemic. Abbreviations DHHS: Department of Health & Human Services; DoD: U.S. Department of Defense; DSCA: Defense Support To Civil Authorities; EHR: Electronic health record; GDP: Gross domestic product; MHS: Military Health System; MTF: Military treatment facility; OWS: Operation warp speed; PHE: Public Health Emergency; WHO: World Health Organization; U.S.: United States of America; U.S.N.S.: U.S. Naval Ship. f Within the areas of human resources, healthcare deliv- ery, and medicine and technology, the collaborative efforts between the MHS and U.S. civilian healthcare system provides important lessons for the continued response to the COVID-19 pandemic and future national health emergencies, including the need to continue train- ing new generations of providers to ensure enough are available to respond, retaining capabilities for mobile medical treatment and testing facilities in the case that permanent structures are full, and continued scientific collaborative efforts between the MHS and U.S. civilian healthcare system. Together, both systems have a wealth of knowledge, skill, and ability that can work in tandem to better care for our nation, develop therapeutics, and administer them efficiently. Funding Funding This work was funded through a grant from the Department of Defense, Defense Health Agency (award # HU0001-20-2-0035). Availability of data and materials This study had several limitations. First, the evolving nature of the COVID-19 pandemic means that knowl- edge and new literature are generated daily, and collabo- rative efforts between the U.S. healthcare system and the MHS continue to change in response to emerging priori- ties. Second, while this study presents lessons learned, it does not address hard outcomes in terms of morbidity and mortality, health service utilization, or costs. Finally, this study does not include late-breaking changes driven by the wide release of COVID-19 vaccines. Thus, this paper is intended to serve as a snapshot in time of the pandemic prior to June 2021. All information pertaining to our research is contained within the manuscript; no external datasets were used. Author details 1 Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814, United States of America. 2 The Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc, 6720B Rockledge Drive, Suite 605, Bethesda, MD 20817, United States of America. 3 Yale Univer‑ sity, New Haven, CT 06520, United States of America. Healthcare information Healthcare information technology in the U.S. is com- posed of hundreds of electronic health record (EHR) sys- tems, data repositories, and information systems. There is a need for increased health information exchanges across the country to improve quality of care and the COVID-19 The areas of governance, finance, and information technology presented fewer opportunities for collabo- rative efforts, due to fundamental differences in priori- ties and administration between the two systems. For Koehlmoos et al. Health Research Policy and Systems (2022) 20:81 Page 5 of 7 highlighted the MHS as instrumental to the collaboration needed to address the pandemic. Therefore continuing to further develop and strengthen this partnership is nec- essary in preparation for continued, timely response to future PHEs. instance, the varied state-level responses to the pan- demic highlighted the fragmentation of care inherent throughout the U.S. civilian healthcare sector, contrast- ing with the centralized, mission-focused response of the MHS. Regarding healthcare financing, no collaborative efforts were found during this time period between the MHS and the U.S. civilian healthcare system, though this could change if these efforts continue. Communication between each system’s disparate information systems remains a significant challenge, and future data sharing should be addressed to ease future efforts. Conclusion Th CO The COVID-19 PHE has demonstrated that the U.S. civilian healthcare sector depends on the MHS and its assets to rapidly fill gaps and strengthen its capacity. The COVID-19 PHE demonstrated that the U.S. depended on the MHS and its assets in its response. It needed MHS personnel to provide direct support to hot spots across the country, to provide leadership alongside DHHS in OWS, and support regulatory social distancing. Further, our findings highlight that MHS providers have been essential in supporting their civilian counterparts dur- ing the COVID-19 PHE. Overall, the COVID-19 PHE Received: 27 July 2021 Accepted: 3 July 2022 Received: 27 July 2021 Accepted: 3 July 2022 Received: 27 July 2021 Accepted: 3 July 2022 Consent for publication Not applicable. Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. Author contributions All authors contributed equally to this manuscript. All authors read and approved the final manuscript. 1. Centers for Disease Control and Prevention. COVID Data Tracker: U.S. Department of Health & Human Services; 2020 [updated 2022 June 15; cited 2022 June 16]. Available from: https://covid.cdc.gov/covid-data- tracker/#datatracker-home. Acknowledgements Not applicable. Acknowledgements Not applicable. Ethics approval and consent to participate Ethics approval and consent to participate Not applicable. Supplementary Information The online version contains supplementary material available at https://doi. org/10.1186/s12961-022-00885-4. The online version contains supplementary material available at https://doi. org/10.1186/s12961-022-00885-4. Additional file 1: Complete list of search terms used during data collection. Additional file 1: Complete list of search terms used during data collection. References Is US medical care inefficient? JAMA. 2018;320:971. https://doi. org/10.1001/jama.2018.10779. 7. Fuchs VR. Is US medical care inefficient? JAMA. 2018;320:971. https://doi. org/10.1001/jama.2018.10779. 8. Slaoui M, Hepburn M. Developing safe and effective Covid vac‑ cines—operation warp speed’s strategy and approach. N Engl J Med. 2020;383:1701–3. https://doi.org/10.1056/nejmp2027405. 8. Slaoui M, Hepburn M. Developing safe and effective Covid vac‑ cines—operation warp speed’s strategy and approach. N Engl J Med. 2020;383:1701–3. https://doi.org/10.1056/nejmp2027405. 28. U.S. Department of Defense. DOD Vaccination Administration to DOD Population Defense.gov: U.S. Department of Defense; 2021 [cited 2022 June 15]. Available from: https://www.defense.gov/Spotlights/Coron avirus-DOD-Response/. p g j p 9. Knudson MM, Jacobs LM, Elster EA. How to partner with the military in responding to pandemics—a blueprint for success. JAMA Surg. 2020;155:548. https://doi.org/10.1001/jamasurg.2020.1227. 29. Michael NL. SARS-CoV-2 in the U.S. Military—lessons for civil society. N Engl J Med. 2020;383:2472–3. https://doi.org/10.1056/nejme2032179.fi 10. Knudson MM, Elster EA, Bailey JA, Johannigman JA, Bailey PV, Schwab CW, et al. Military-civilian partnerships in training, sustaining, recruitment, retention, and readiness: proceedings from an exploratory first-steps meeting. J Am Coll Surg. 2018;227:284–92. https://doi.org/10.1016/j. jamcollsurg.2018.04.030. 30. Miceli S. How Can Hospitals Overcome Staffing and Supply Shortages Amid COVID-19 Surges?2020 [cited 2020 September 17]. Available from: https://www.nationalacademies.org/news/2020/08/how-can-hospitals- overcome-staffing-and-supply-shortages-amid-covid-19-surges. f y 31. New AAMC Report Confirms Growing Physician Shortage [press release]. Association of American Medical Colleges, June 26 2020. 11. Marklund LA, Graham AM, Morton PG, Hurst CG, Motola I, Robinson DW, et al. Collaboration between civilian and military healthcare profession‑ als: a better way for planning, preparing, and responding to all hazard domestic events. Prehosp Disaster Med. 2010;25:399–412. https://doi.org/ 10.1017/s1049023x00008451. 32. American Hospital Association. ARCHIVED: Fast Facts on U.S. Hospitals, 2019: American Hospital Association; 2019 [updated Jan 2019; cited 2020 September 15]. Available from: https://www.aha.org/statistics/2020-01- 07-archived-fast-facts-us-hospitals-2019. 12. Mullick H. Miltary Aids Disaster Response Effort in the Aftermath of Hur‑ ricane Florence. Defense News [Internet]. 2018 [cited 2021 February 2]. Available from: https://www.defense.gov/Explore/News/Article/Article/ 1645682/military-aids-disaster-response-effort-in-the-aftermath-of-hurri cane-florence/. 33. Oxford Uo. Weekly New Hospital Admissions. Our World in Data: Univer‑ sity of Oxford; 2020. 34. U.S. Army North Public Affairs. U.S. military deploys to Texas and California in support of COVID-19 operations. Healthmil News [Internet]. 2020 [cited 2020 September 21]. Available from: https://health.mil/News/Articles/ 2020/07/14/U-S-military-deploys-to-Texas-and-California-in-support-of- COVID-19-operations. 13. Huff B. Task Force 51 ensures readiness for hurricane responses. Army Worldwide News [Internet]. 2020 [cited 2021 February 2]. Available from: https://www.army.mil/article/241286/task_force_51_ensures_readiness_ for_hurricane_responses. 35. Ziezulewicz G. The U.S.N.S. Comfort Is Now Taking Covid-19 Patients. References Koehlmoos et al. Health Research Policy and Systems (2022) 20:81 Page 6 of 7 2. Office of the Assistant Secretary for Preparedness and Response. Public Health Emergency Declarations: U.S. Department of Health & Human Ser‑ vices; [updated 2022 May 9; cited 2022 June 16]. Available from: https:// www.phe.gov/emergency/news/healthactions/phe/Pages/default.aspx. 2. Office of the Assistant Secretary for Preparedness and Response. Public Health Emergency Declarations: U.S. Department of Health & Human Ser‑ vices; [updated 2022 May 9; cited 2022 June 16]. Available from: https:// www.phe.gov/emergency/news/healthactions/phe/Pages/default.aspx. 20Department%20of%20Defense%20has,he%20said%20in%20a%20sta tement. 20Department%20of%20Defense%20has,he%20said%20in%20a%20sta tement. 23. Kersey R. JBSA medical facilities support Operation Warp Speed2020 [cited 2022 June 22]. Available from: https://www.59mdw.af.mil/News/ Article/2435159/jbsa-medical-facilities-support-operation-warp-speed/. 3. Steier J, Moxham J. The load and capacity model of healthcare delivery: considerations for the crisis management of the COVID-19 pandemic. J Thorac Dis. 2020;12:3022–30. https://doi.org/10.21037/jtd-2020-054. 3. Steier J, Moxham J. The load and capacity model of healthcare delivery: considerations for the crisis management of the COVID-19 pandemic. J Thorac Dis. 2020;12:3022–30. https://doi.org/10.21037/jtd-2020-054. 24. Swatek G. Effort at Fort Detrick leads to approval for COVID-19 treatment. The Frederick News-Post [Internet]. 2020 [cited 2020 November 2]. Avail‑ able from: https://www.fredericknewspost.com/news/continuing_cover age/coronavirus/effort-at-fort-detrick-leads-to-approval-for-covid-19- treatment/article_6d1d31c0-81cb-56fa-a25a-7c734f953c87.html. p g j 4. De Lew N, Greenberg G, Kinchen K. A layman’s guide to the U.S. health care system. Health Care Financ Rev. 1992;14:151–69. g j 4. De Lew N, Greenberg G, Kinchen K. A layman’s guide to the U.S. health care system. Health Care Financ Rev. 1992;14:151–69. 5. Tikkanen R, Osborn R, Mossialos E, Djordjevic A, Wharton GA. Interna‑ tional Health Care System Profiles: United States2020 [cited 2022 June 16]. Available from: https://www.commonwealthfund.org/international- health-policy-center/countries/united-states. 5. Tikkanen R, Osborn R, Mossialos E, Djordjevic A, Wharton GA. Interna‑ tional Health Care System Profiles: United States2020 [cited 2022 June 16]. Available from: https://www.commonwealthfund.org/international- health-policy-center/countries/united-states. 25. Beigel JH, Tomashek KM, Dodd LE, Mehta AK, Zingman BS, Kalil AC, et al. Remdesivir for the treatment of COVID-19—final report. N Engl J Med. 2020;383:1813–26. https://doi.org/10.1056/nejmoa2007764. 26. FDA Issues Emergency Use Authorization for Third COVID-19 Vaccine [press release]. FDA.gov: U.S. Food and Drug Administration2021. 6. Rice T, Rosenau P, Unruh LY, Barnes AJ, Saltman RB, van Ginneken E. United States of America: health system review. Health Syst Transit. 2013;15:1–431. 27. Centers for Disease Control and Prevention. COVID-19 Vaccinations in the United States USA.gov: Centers for Disease Control and Prevention; 2021 [updated 2022 June 15; cited 2022 June 16]. Available from: https://covid. cdc.gov/covid-data-tracker/#vaccinations_vacc-total-admin-rate-total. 7. Fuchs VR. References Available from: https://www.statnews.com/2020/09/28/opera tion-warp-speed-vast-military-involvement/#:~:text=%E2%80%9CThe% Page 7 of 7 Koehlmoos et al. Health Research Policy and Systems (2022) 20:81 59. Federal Electronic Health Record Connects With More Community Part‑ ners [press release]. U.S. Department of Defense, October 13 2020. 59. Federal Electronic Health Record Connects With More Community Part‑ ners [press release]. U.S. Department of Defense, October 13 2020. 59. Federal Electronic Health Record Connects With More Community Part‑ ners [press release]. U.S. Department of Defense, October 13 2020. 41. Simunaci L. Technology, Expertise Help Determine Vaccine Distribution. Defense News [Internet]. 2020 [cited 2021 February 4]. Available from: https://www.defense.gov/Explore/News/Article/Article/2410195/techn ology-expertise-help-determine-vaccine-distribution/. References Here’s What to Expect: The Navy ship’s mission in New York now comes with a new level of complexity, raising questions about whether it’s equipped for the task. The New York Times [Internet]. 2020 [cited 2020 September 21]. Available from: https://www.nytimes.com/2020/04/08/ magazine/hospital-ship-comfort-new-york-coronavirus.html. 14. Defense Health Agency. About the Military Health System Health.mil [updated 2022 June 6; cited 2022 June 16]. Available from: https://www. health.mil/About-MHS. 15. Adirim T. A military health system for the twenty-first century. Health Aff (Millwood). 2019;38:1268–73. https://doi.org/10.1377/hlthaff.2019.00302. 16. Monitoring the building blocks of health systems: a handbook of indica‑ tors and their measurement strategies. Health statistics and information systems: World Health Organization; 2010. p. 93. 36. Maxouris C. States call in National Guard and train volunteers to help boost vaccination pace as US hits a Covid-19 daily death toll record. CNN [Internet]. 2021 [cited 2021 January 8]. Available from: https://www.wfsb. com/news/states-call-in-national-guard-and-train-volunteers-to-help- boost-vaccination-pace-as-us/article_418e61d2-cbd6-5bc3-9adf-298b8 f081c83.html. y g 17. Gordon SH, Huberfeld N, Jones DK. What Federalism Means for the US Response to Coronavirus Disease 2019. JAMA Health Forum. 2020;1: e200510. https://doi.org/10.1001/jamahealthforum.2020.0510. 37. Everstine BW. DOD Deploying Medical Personnel, More on Standby As COVID-19 Cases Surge. Air Force Magazine [Internet]. 2020 [cited 2020 December 16]. Available from: https://www.airforcemag.com/dod-deplo ying-medical-personnel-more-on-standby-as-covid-19-cases-surge/. 18. Madhani A. What to wear: Feds’ mixed messages on masks sow confu‑ sion. AP News [Internet]. 2020 [cited 2020 November 13]. Available from: https://apnews.com/article/virus-outbreak-donald-trump-ap-top-news- politics-pandemics-26f0cb8ed836a76f0e019357cbea7f58. 38. Grazier D. Ignored Warnings Left the Military Health System Unpre‑ pared2020 [cited 2020 September 21]. Available from: https://www.pogo. org/analysis/2020/05/ignored-warnings-left-the-military-health-system- unprepared/. 19. Esper M. Guidance for Commanders on Risk-Based Changing of Health Protection Condition Levels During the Coronavirus Disease 2019 Pan‑ demic In: Defense Do, editor. Washington, DC: Office of the U.S. Secretary of Defense; 2020. p. 4. 20. Defense Health Agency. About JTS: DoD Center of Excellence for Trauma [updated 2019 March 5; cited 2020 November 2]. Available from: https:// jts.amedd.army.mil/index.cfm/about/about_jts. 39. Bowman T. U.S. Military Is Sending Medical Staff To COVID-19 Hotspots. The Coronavirus Crisis [Internet]. 2020 [cited 2021 February 2]. Available from: https://www.npr.org/sections/coronavirus-live-updates/2020/07/ 13/890553905/u-s-military-is-sending-medical-staff-to-covid-19-hotsp ots. 21. Defense Support of Civil Authorities. JP 3–28 ed. Washington, DC: Joint Chiefs of Staff. 40. Associated Press. U.S. National Guard assists as Czech Republic sounds Covid-19 alarm. NBC World News [Internet]. 2020 [cited 2021 February 2]. Available from: https://www.nbcnews.com/news/world/u-s-national- guard-assists-czech-republic-sounds-covid-19-n1245444. 22. Florko N. New document reveals scope and structure of Operation Warp Speed and underscores vast military involvement2020 [cited 2022 June 16]. Publisher’s Note 42. Hall S. DoD medical personnel work alongside civilian healthcare workers to respond to COVID-19 in El Paso, Texas. Army Worldwide News [Inter‑ net]. 2020 [cited 2020 November 23]. Available from: https://www.army. mil/article/241130/dod_medical_personnel_work_alongside_civilian_ healthcare_workers_to_respond_to_covid_19_in_el_paso_texas. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. 43. Mitchell G. Known COVID-19 nurse death toll ’now equal to that from WWI’. Nursing Times [Internet]. 2020 [cited 2020 November 2]. Available from: https://www.nursingtimes.net/news/coronavirus/known-covid-19- nurse-death-toll-now-equal-that-from-ww1-27-10-2020/. 44. Jones R. Mitigating the impact of Covid-19 on Medicare Advantage risk scores and payment. MedCity News [Internet]. 2020 [cited 2020 October 19]. Available from: https://medcitynews.com/2020/09/mitigating-the- impact-of-covid-19-on-medicare-advantage-risk-scores-and-payment/. 45. McKinsey and Company. COVID-19 impact on surgical volumes. Lecture Presented: Health Systems at the Uniformed Services University of the Health Sciences. Bethesda, MD: McKinsey and Company; 2020. 46. Birkmeyer JD, Barnato A, Birkmeyer N, Bessler R, Skinner J. The impact of the COVID-19 pandemic on hospital admissions in the United States. Health Aff. 2020;39:2010–7. https://doi.org/10.1377/hlthaff.2020.00980. 47. Joyce A. The U.S. Has Fewer Physicians and Hospital Beds Per Capita Than Italy and Other Countries Overwhelmed by COVID-192020 [cited 2020 November 13]. Available from: https://www.kff.org/health-costs/press- release/the-u-s-has-fewer-physicians-and-hospital-beds-per-capita-than- italy-and-other-countries-overwhelmed-by-covid-19/. 48. Yuan L, Sherryn S, Hu P, Chen F. U.S. Field Hospitals: A Study on Public Health Emergency Response to COVID-19. medRxiv. 2020. 49. Yuan J. Amid New York’s 42,400 hospitalizations, the military handled 3 percent. But it helped in immeasurable ways. The New York Times [Inter‑ net]. 2020 [cited 2020 December 17]. Available from: https://www.washi ngtonpost.com/national/new-york-javits-center-usns-comfort/2020/05/ 02/55abfe54-88af-11ea-8ac1-bfb250876b7a_story.html. y 50. Fuentes G. Beyond Mercy: Navy’s COVID-19 Hospital Ship Missions and the Future of Medicine at Sea2020 [cited 2020 December 18]. Available from: https://news.usni.org/2020/05/25/beyond-mercy-navys-covid-19- hospital-ship-missions-and-the-future-of-medicine-at-sea. 51. Report K. Governor Abbott requests use of William Beaumont Army Medical Center for non-COVID patients2020 [cited 2020 October 28]. Available from: https://www.ktsm.com/local/el-paso-news/governor- abbott-requests-use-of-beaumont-medical-center-for-non-covid-patie nts/. 52. Military Health System Communications Office. Hepburn: DOD role in Operation Warp Speed was ‘transformative’. Healthmil News [Internet]. 2020 [cited 2021 June 11]. Available from: https://health.mil/News/Artic les/2020/12/11/Hepburn-DOD-role-in-Operation-Warp-Speed-was-trans formative#:~:text=Matthew%20Hepburn’s%20praise%20for%20DOD,in% 20our%20DNA%E2%80%9D%20at%20DOD. 53. Tikkanen R, Abrams MK. U.S. Health Care from a Global Perspective, 2019: Higher Spending, Worse Outcomes?2020 [cited 2020 October 12]. Available from: https://www.commonwealthfund.org/publications/issue- briefs/2020/jan/us-health-care-global-perspective-2019. 54. Mendez BHP, Kamarck KN, Kapp L, Ott A, Torreon BS. FY2020 National Defense Authorization Act: Selected Military Personnel Issues. Congres‑ sional Research Service, 2020 Sept 18, 2020. Report No.: Contract No.: R46107. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: Publisher’s Note • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: 55. Blumenthal D, Fowler EJ, Abrams M, Collins SR. Covid-19—implications for the Health Care System. N Engl J Med. 2020;383:1483–8. https://doi. org/10.1056/nejmsb2021088. 56. Adirim T. Regarding the Military Health System Before the Senate Appropriations Commitee Defense Subcommitee. United States Senate Committee on Appropriations; 2021. p. 20. 57. U.S. Department of Health & Human Services. Department of Health & Human Services - ASPR GIS: Assistant Secretary for Preparedness and Response; 2017 [cited 2020 November 9]. Available from: https://asprgis. hhs.gov/GISSupport/pages/default.aspx. 58. Defense Health Agency. Joint Health Information Exchange Health.mil: Defense Health Agency; [cited 2020 October 13]. Available from: https:// www.health.mil/Military-Health-Topics/Technology/Joint-HIE. 58. Defense Health Agency. Joint Health Information Exchange Health.mil: Defense Health Agency; [cited 2020 October 13]. Available from: https:// www.health.mil/Military-Health-Topics/Technology/Joint-HIE.
https://openalex.org/W2897622607	https://www.frontiersin.org/articles/10.3389/fpls.2018.01538/pdf	English	null	Prediction of Lime Tolerance in Rhododendron Based on Herbarium Specimen and Geochemical Data	Frontiers in plant science	2,018	cc-by	9,825	Edited by: Erik T. Nilsen, Virginia Tech, United States Reviewed by: Juliana S. Medeiros, The Holden Arboretum, United States Robbie Hart, Missouri Botanical Garden, United States Correspondence: Shusheng Wang shusheng.wang@ilvo.vlaanderen.be Correspondence: Shusheng Wang shusheng.wang@ilvo.vlaanderen.be Specialty section: This article was submitted to Plant Abiotic Stress, a section of the journal Frontiers in Plant Science Prediction of Lime Tolerance in Rhododendron Based on Herbarium Specimen and Geochemical Data Shusheng Wang1,2,3*, Leen Leus1, Marie-Christine Van Labeke2 and Johan Van Huylenbroeck1 1 Plant Sciences Unit, Applied Genetics and Breeding, Flanders Research Institute for Agriculture, Fisheries and Food (ILVO), Melle, Belgium, 2 Department of Plants and Crops, Faculty of Bioscience Engineering, Ghent University, Ghent, Belgium, 3 Lushan Botanical Garden, Jiangxi Province and Chinese Academy of Sciences, Jiujiang, China Rhododendrons are typically known to be calcifuges that cannot grow well in lime soils. Data on lime tolerance of different taxa in Rhododendron are scarce. Habitats of naturally distributed specimens of genus Rhododendron were compiled as Chinese text-based locations from the Chinese Virtual Herbarium. The locations were then geocoded into latitude/longitude pairs and subsequently connected to soil characteristics including pH and CaCO3 from the Harmonized World Soil Database (HWSD). Using the upper quartile values of pH > 7.2 and CaCO3 > 2% weight in topsoil as threshold, we predicted the lime tolerant taxa. A dataset of 31,146 Rhododendron specimens including the information on taxonomy, GPS locations and soil parameters for both top- and subsoil was built. The majority of the specimens were distributed in soils with moderately acidic pH and without presence of CaCO3. 76 taxa with potential lime tolerance were predicted out of 525 taxa. The large scale data analysis based on combined data of geocoded herbarium specimens and HWSD allows identiﬁcation of valuable Rhododendron species, subspecies or botanical varieties with potential tolerance to lime soils with higher pH. The predicted tolerant taxa are valuable resources for an in-depth evaluation of lime tolerance or for further use in horticulture and breeding. ORIGINAL RESEARCH published: 23 October 2018 doi: 10.3389/fpls.2018.01538 Keywords: Rhododendron, lime tolerance, pH, calcium carbonate, herbarium specimen, geocoding INTRODUCTION The tolerance of various plant species to abiotic stresses evolves according to environmental changes in their habitats (Dimichele et al., 1987; Amtmann et al., 2005; Marais and Juenger, 2010). The evolutionary processes inﬂuenced by environmental change as well as the modern regionalization and dispersal of natural habitats have resulted in diverse biogeographical distribution patterns among diﬀerent plants (Wen, 1999; Xing et al., 2015). Predicting plant species’ tolerance to abiotic stresses using distribution and geochemical data has been accepted as a potentially useful approach (Saslis-Lagoudakis et al., 2015). However, collection of large- scale distribution data of plants through ﬁeld study is time-consuming. An alternative method of extracting the distribution information from the specimens in herbaria, which were collected and identiﬁed by botanists and experienced plant hunters including the location data, is eﬀective and meaningful (Hart et al., 2014; Romeiras et al., 2014; Zhang et al., 2015). Received: 21 June 2018 Accepted: 28 September 2018 Published: 23 October 2018 Citation: Wang S, Leus L, Van Labeke M-C and Van Huylenbroeck J (2018) Prediction of Lime Tolerance in Rhododendron Based on Herbarium Specimen and Geochemical Data. Front. Plant Sci. 9:1538. doi: 10.3389/fpls.2018.01538 October 2018 \| Volume 9 \| Article 1538 Frontiers in Plant Science \| www.frontiersin.org 1 Prediction of Lime Tolerance in Rhododendron Wang et al. Rhododendron is the largest genus in the family of Ericaceae, comprising nine subgenera, and with about 1000 species, primarily distributed in Asia, Europe, and North America. In China 571 species, 180 botanical varieties and 72 subspecies are reported (Fang et al., 2005). Although rhododendrons are of high ornamental value, they are typically recognized as calcifugous plants (which cannot grow well on lime/calcareous soils), and they usually grow well in soils with pH of 4.5 to 6.0 (Kinsman, 1999). Rhododendrons growing in pH-neutral or alkaline soils frequently suﬀer from iron (Fe) deﬁciency chlorosis symptoms: interveinal chlorosis in newly formed leaves, shoot and root growth reduction, leaf wilting, defoliation, and ﬁnally, plant death (Demasi et al., 2015a). Fe deﬁciency may be caused by physical or chemical properties of lime soils, which contain high bicarbonate (HCO3−) concentrations in their soil solution (Mengel, 1994). Mordhorst et al. (1993) found that high calcium cation ([Ca2+]) supply (in absence of HCO3−) does not suppress growth in Rhododendron. Other researchers found that the inﬂuence of calcium compounds on the development of Rhododendron micro-cuttings did not depend on the amount of assimilated Ca2+ ions but rather on the type of anions present in given salts (Giel and Bojarczuk, 2002). Further work showed that the major factor limiting Rhododendron growth in calcareous soils is the increase in substrate pH, rather than an increase in the concentration of calcium ions (Giel and Bojarczuk, 2011). contains no delimiters between Chinese words, and limited address reference resources (Tian et al., 2016). In this study, the distribution map of Rhododendron natural taxa including species, subspecies and varieties in China were generated according to the latitude/longitude pairs geocoded from Chinese text-based addresses of herbarium specimens. We connected the obtained GPS data of the herbarium specimens with the HWSD, which enabled us to derive data on soil pH and CaCO3 concentrations for each specimen location. The inﬂuence of soil characteristics on the distributions of Rhododendron taxa and the tolerance potential of rhododendrons to high pH and CaCO3 concentration were evaluated. Citation: By analyzing Rhododendron specimens and geochemical data, we aim to illustrate and predict their potential tolerance to abiotic stress at taxon level. The large scale data analysis, which concerns a multitude of taxa and a large area, allowed us to identify valuable Rhododendron species, subspecies or natural varieties with tolerance potential to lime soils with higher pH. 1http://www.cvh.ac.cn 2https://www.r-project.org/ Data Collection of Herbarium Specimens and Geocoding In this study, Rhododendron species, subspecies, and botanical varieties were regarded as independent taxa (taxonomic units). Data on Rhododendron taxa were collected from CVH. Approximately 90,000 Rhododendron specimens were present in that database. Taxonomic data as well as the Chinese text- based locations of herbarium specimens were extracted in R (R Development Core Team2), using packages “RCurl” (version 1.95-4.8) and “XML” (version 3.98-1.5). The Latin names and subgenus information were subsequently revised and uniformed according to Flora of China (Fang et al., 2005). The Chinese Virtual Herbarium (CVH1), an online portal allowing access to herbarium specimen information, is a collaboration of more than 30 major herbaria in China and consists of more than 6 million specimens. CVH lists about 90 thousand Rhododendron specimens. However, the information on the collection locations of the specimens are mostly recorded in Chinese text instead of Global Position System (GPS) data due to the absence of portable precise positioning devices at the time of collection. Therefore, this information cannot be used directly for visualization of plant distribution and it hampers the automatic connection with other databases such as the Harmonized World Soil Database (HSWD). The HSWD is a 30 arc-second raster database with over 15,000 diﬀerent soil mapping units that combines existing regional and national updates of soil information worldwide (Batjes et al., 2012). The Chinese text-based locations of each specimen were geocoded to GPS latitude/longitude pairs using the R package “REmap” (Lang, 2016) based on Baidu Maps API (Application Program Interface). Soil Parameters and Data Cleaning Soil Parameters and Data Cleaning The Harmonized World Soil Database (HWSD, version 1.21) (Batjes et al., 2012) was used to obtain the soil data information. “MU_GLOBAL” (Global Mapping Unit Identiﬁer) provided the link between the Geographic Information System (GIS) layer and the attribute database. Mainland China can be recognized by “MU_GLOBAL” from 11000 to 11935. For each of these 936 mapping units, there is only one set of soil data with physical and chemical parameters. First “MU_GLOBAL” of the specimens with obtained latitude/longitude pairs were extracted in ArcGIS 10.2 (ESRI, Redlands, CA, United States) based on HWSD projected in spatial reference of WGS_1984 at a resolution of 0.0083 decimal degrees, which covered a grid cell of about 1 km × 1 km. Then soil parameters, including pH measured in a soil-water solution and CaCO3 of both topsoils (0 ∼30 cm) and subsoils Geocoding is often described as the process of converting text-based address data into digital geographic coordinates, most commonly resulting in latitude/longitude pairs (Goldberg, 2011). Geocoding technology is increasingly important in the coming era of big data to bridge the gap between non-spatial and spatial data in various ﬁelds, such as epidemiology (Bergman et al., 2012; Nuvolone et al., 2016), environmental health (Faure et al., 2017), land or forest economy (Johnston et al., 2016; Moeltner et al., 2017), and so on. Because of its great importance, many geocoding methods have been developed including online services, commercial in-house services, as well as no-cost strategies using R (Goldstein et al., 2014; Faure et al., 2017). However, Chinese geocoding faces great challenges due to the complexity of the address string format in Chinese, which October 2018 \| Volume 9 \| Article 1538 Frontiers in Plant Science \| www.frontiersin.org 2 Prediction of Lime Tolerance in Rhododendron Wang et al. (30 ∼100 cm) were obtained by connecting the specimen location data and HWSD data using the same “MU_GLOBAL”. (30 ∼100 cm) were obtained by connecting the specimen location data and HWSD data using the same “MU_GLOBAL”. and the ﬁeld-recorded locations was 31.2 ± 7.5 km (1st quantile 3.7 km, median 7.0 km, 3rd quantile 22.2 km). From these 35,574 specimens, the “MU_GLOBAL” data were extracted based on specimen GPS locations and HWSD. The obtained data showed that “MU_GLOBAL” of 382 specimens were located outside of mainland China. Data Analysis TABLE 2 \| List of top 20 Rhododendron species represented by the highest numbers of specimens. Latin name Number of specimens Rhododendron simsii 1435 Rhododendron decorum 1374 Rhododendron micranthum 826 Rhododendron mariesii 799 Rhododendron racemosum 630 Rhododendron oreodoxa 582 Rhododendron stamineum 573 Rhododendron ovatum 556 Rhododendron rubiginosum 488 Rhododendron pachytrichum 457 Rhododendron yunnanense 447 Rhododendron mariae 431 Rhododendron concinnum 391 Rhododendron microphyton 378 Rhododendron augustinii 376 Rhododendron vernicosum 374 Rhododendron siderophyllum 357 Rhododendron irroratum 342 Rhododendron mucronulatum 322 Rhododendron lutescens 320 TABLE 1 \| Distribution of the different species present in the database in nine subgenera of Rhododendron. Subgenus of Rhododendron Number of species present in our datasetz Number of described species in Chinay Rhododendron 137 184 Pseudazalea 3 6 Pseudorhodorastrum 8 10 Rhodorastrum 2 2 Hymenanthes 181 259 Azaleastrum 18 26 Pentanthera 2 2 Tsutsusi 61 81 Therorhodion 1 1 Total 413 571 zThe number of collected species showed here excludes the 78 botanical varieties and 34 subspecies. yBased on Fang et al. (2005) and also excluding the varieties and subspecies. TABLE 2 \| List of top 20 Rhododendron species represented by the highest numbers of specimens TABLE 1 \| Distribution of the different species present in the database in nine subgenera of Rhododendron. Latin name Number of specimens Rhododendron simsii 1435 Rhododendron decorum 1374 Rhododendron micranthum 826 Rhododendron mariesii 799 Rhododendron racemosum 630 Rhododendron oreodoxa 582 Rhododendron stamineum 573 Rhododendron ovatum 556 Rhododendron rubiginosum 488 Rhododendron pachytrichum 457 Rhododendron yunnanense 447 Rhododendron mariae 431 Rhododendron concinnum 391 Rhododendron microphyton 378 Rhododendron augustinii 376 Rhododendron vernicosum 374 Rhododendron siderophyllum 357 Rhododendron irroratum 342 Rhododendron mucronulatum 322 Rhododendron lutescens 320 Soil Parameters and Data Cleaning Furthermore, 458 specimens were located in ‘inland water,’ ‘rock debris,’ ‘glaciers and permanent snow,’ or ‘urban area’ and 3,588 specimens lacked subsoil data. All of these specimens were discarded from the dataset. Finally, we built a database of 31,146 specimens including the information on taxonomy, GPS locations and soil parameters for both top- and subsoil. This database was used for subsequent data analysis. The specimens from which the calculated location was linked via “MU_GLOBAL” to areas identiﬁed as either ‘inland water,’ ‘rock debris,’ ‘glaciers and permanent snow,’ or ‘urban area’ were removed from the soil parameters dataset. These areas are indicated by the symbols “WR,” “RK,” “GG,” or “UR,” respectively, under the soil unit symbol “SU_SYM90”. Specimens lacking subsoil data were also deleted. Frontiers in Plant Science \| www.frontiersin.org Data Analysis Accuracy of the geocoding algorithm was evaluated by calculating the distance between the automated geocoded locations and the original ﬁeld-recorded locations by “distVincentyEllipsoid()” in R package “geosphere” (Hijmans, 2017). The number of specimens for each taxon were counted and taxa were classiﬁed according to the nine subgenera of Rhododendron listed in Flora of China (Fang et al., 2005). Heat maps (Kernel Density) were generated in ArcGIS 10.2 to visualize the distribution centers of the genus Rhododendron. The number and percentage of grid cells, specimens and taxa were calculated according to the diﬀerent pH and CaCO3 ranges for both top- and subsoil. For pH, ﬁve ranges (<4.5, 4.5–5.5, 5.5–7.2, 7.2–8.5, and >8.5) and for CaCO3 (% weight), four classes (<2, 2–5, 5–15, and >15), were deﬁned according to HWSD. For each taxon with at least 10 specimens in our database, we determined lower quartile (LQ), median and upper quartile (UQ) for both pH and CaCO3 of topsoil. Median values hereby provide the taxon’s central tendency to environmental conditions (pH and CaCO3) in their distributions, while LQ and UQ values represent more extreme conditions that rhododendrons counter within their habitats. Because we aimed to predict the tolerance potential of rhododendrons to lime soils with high pH and CaCO3 concentration, we considered the UQ as tolerance indicator. The taxa with UQ of pH > 7.2 and CaCO3 > 2% weight in topsoils were predicted as being potentially tolerant to lime soils. To evaluate the results of prediction, the number of specimens distributed in lime (pH > 7.2 and CaCO3 > 2%) or non-lime (pH ≤7.2 and CaCO3 ≤2%) top soils of predicted tolerant and non-tolerant taxa were calculated, a Chi-square test in R was then performed. The distribution of the taxa as predicted in this study and based on literature reports of taxa with lime tolerance potential was mapped using ArcGIS 10.2. TABLE 1 \| Distribution of the different species present in the database in nine subgenera of Rhododendron. Subgenus of Rhododendron Number of species present in our datasetz Number of described species in Chinay Rhododendron 137 184 Pseudazalea 3 6 Pseudorhodorastrum 8 10 Rhodorastrum 2 2 Hymenanthes 181 259 Azaleastrum 18 26 Pentanthera 2 2 Tsutsusi 61 81 Therorhodion 1 1 Total 413 571 zThe number of collected species showed here excludes the 78 botanical varieties and 34 subspecies. yBased on Fang et al. (2005) and also excluding the varieties and subspecies. Tolerance Potential of Rhododendron Taxa to Lime Soils Rhododendron Taxa and Distribution The 31,146 specimens of our dataset were divided into 413 species, 78 varieties and 34 subspecies. The species represented 72.3% of the described species in China and covered all nine subgenera (Table 1), while varieties and subspecies represented 43.3 and 47.2% of the described Chinese Rhododendron varieties and subspecies, respectively. For subsequent analyses the varieties and subspecies were treated at the same level as species as separate taxa: in total, 525 taxa were analyzed. The number of specimens for each taxon varied from 1 to 1435 (Supplementary Table S1). For 302 of these, at least 10 specimens were present, while 37 had more than 200 specimens. Rhododendron simsii, one of the most common species in genus Rhododendron, has the most specimens, followed by R. decorum, R. micranthum, and R. mariesii, etc. The 20 species with most specimens represented 36.8% of the complete database (Table 2). The heat map made from the 31,146 specimens showed that genetic resources of genus Rhododendron were mainly distributed in southwest China, with two hotspots in Sichuan and Yunnan province (Figure 1). Using the UQ values of pH > 7.2 and CaCO3 > 2% weight in topsoil as a threshold, 76 Rhododendron taxa were identiﬁed as potentially tolerant to lime soils, which are characterized by a higher pH and higher CaCO3 content (Table 5). The diﬀerent taxa belong to the nine Rhododendron subgenera. In the predicted tolerant taxa, 2,995 and 5,443 specimens were distributed in lime and non-lime soils, respectively. And in the predicted non- tolerant taxa, 3,475 and 19,111 specimens were distributed in lime and non-lime soils, respectively. The Chi-square test showed that the distribution of specimens in lime and non-lime soils was signiﬁcant diﬀerent between predicted tolerant and non-tolerant taxa (X2 was 1503.7, df = 1 and p-value < 2.2 × 10−16). Using our geocoding method, we observed that in 20 taxa, more than half of the specimens were located in soils with pH > 7.2; for 53 taxa, at least one-third of the specimens were located in soils with topsoil pH > 7.2. For 10 taxa, whose tolerance potential was also supported by literature (R. davidsonianum, R. fortunei, R. micranthum, R. nivale, R. phaeochrysum, R. primuliﬂorum, R. telmateium, R. trichocladum, R. vernicosum, R. Database Construction Of the nearly 90,000 Rhododendron specimens present in the CVH, 69,129 specimens containing Chinese text-based locations could be collected. Of those, the locations of 35,574 specimens were geocoded to GPS latitude/longitude pairs with a labeled text that best matched the location text of specimen. As a vast majority of specimens didn’t include the ﬁeld-recorded GPS locations, we randomly selected 60 specimens with GPS information which could be found from the scanned picture of specimens in CVH. The average distance between the automated geocoded locations Frontiers in Plant Science \| www.frontiersin.org October 2018 \| Volume 9 \| Article 1538 3 Prediction of Lime Tolerance in Rhododendron Wang et al. Tolerance Potential of Rhododendron Taxa to Lime Soils DISCUSSION China is considered to be a center of origin for Rhododendron. Chinese herbaria, accessible via the CVH, conserved tens of thousands of specimens collected by botanists and plant hunters since 1884 and this number is still increasing. Using the statistical software R with package “REmap,” we geocoded 51.5% of the Chinese text-based locations into latitude/longitude pairs with an acceptable accuracy (31.2 ± 7.5 km) in terms of a continental- scale dataset. Faure et al. (2017) reported in their epidemiological study 81.4 and 84.4% of geocoded addresses using a free-online geocoding service and an in-house geocoder system, respectively. They concluded that the geocoding accuracy was higher in urban areas compared to rural areas, but comparable for the two automatic geocoding methods. The lower geocoding rate in our study might be due to the quality of the Chinese texts which described the locations. The possibility for geocoding drops when the textual description of a location is written in an irregular manner, or a location has more than one synonym which occurs among diﬀerent Chinese authorities in diﬀerent governmental agencies. Besides, the whole analysis relied upon data being recorded on the herbarium specimen label. If speciﬁc locality information was not recorded, no geocoding scheme could be able to resolve an accurate location. Likewise, Hart et al. (2014) excluded ∼90% of specimens in their study mostly because speciﬁc information, such as altitude and date of collection was not recorded on the herbarium specimen label. Despite availability of commercial and no-cost geocoding strategies (Goldstein et al., 2014; Faure et al., 2017), the automated geocoding of textual documents faces challenges, especially for development of language modeling methods for Effect of Soil pH p A total of 10,804,183 grid cells in China were mainly distributed at pH range of 5.5–7.2 (41.2%) and 7.2–8.5 (38.8%) of topsoil, with more at 7.2–8.5 (44.9%) than 5.5–7.2 (36.0%) of subsoil (Table 3). More than 75% of Rhododendron specimens were located in areas with top- and subsoil pH ranging between 4.5 and 7.2 (Table 3). About 20% of the specimens were linked to a topsoil pH between 7.2 and 8.5. Almost no specimens or taxa were distributed in areas with pH lower than 4.5 or higher than 8.5. In topsoil, grid cells distributed at four humps around pH of 6.5 (12.5%), 8 (11%), 5.7 (7.9%), and 4.8 (6.7) (Figure 2). pH 6.5 grouped the most specimens (9,898, 31.8%), followed by 4.9 (3,858, 12.4%) and 4.8 (3,399, 10.9%), then 7.8 (2,735, 8.8%) and 8 (2,727, 8.8%). Similarly, most taxa (393, 74.9%) distributed in topsoil pH of 6.5, followed by 4.9 (284, 54.1%), 4.8 (248, 47.2%), 7.8 (218, 41.5%), and 8 (202, 38.5%) (Figure 2). Tolerance Potential of Rhododendron Taxa to Lime Soils yunnanense), the geocoded locations of the specimens growing in topsoil with pH > 7.2 and CaCO3 > 2% weight were compared with the location of the other specimens (Figure 4). Effect of Soil CaCO3 More than a half of grid cells in China were linked with CaCO3 concentration less than 2% for both top (59.8%) and subsoils (52.5%), followed by concentration rang of 5–15 and 2–5 (% weight) (Table 4). A vast majority of specimens (>75%) and taxa (>95%) were distributed in top or subsoil with CaCO3 less than 2% weight, followed by CaCO3 ranging between 5– 15 and 2–5 (% weight) (Table 4). Almost no Rhododendron taxa were distributed in soils with CaCO3 higher than 15% weight. In topsoils, 52.9, 5.5, 4.6, 3.8, and 2.1% of grid cells were linked with CaCO3 concentration of 0, 9.3, 7, 3, 0.1 (% weight) respectively (Figure 3). In topsoils without any CaCO3, 20,788 (66.7%) specimens and 497 (94.7%) taxa were distributed, followed by a concentration of 0.1% weight, with 2,730 (8.8%) specimens and 218 (41.5%) taxa, and 9.3% weight with 2,674 (8.6%) specimens and 217 (41.3%) taxa (Figure 3). October 2018 \| Volume 9 \| Article 1538 Frontiers in Plant Science \| www.frontiersin.org 4 Prediction of Lime Tolerance in Rhododendron Wang et al. FIGURE 1 \| Heat map (Kernel Density) showing the distribution in China of the 31,146 Rhododendron specimens present in our dataset. \| Heat map (Kernel Density) showing the distribution in China of the 31,146 Rhododendron specimens present in our dataset. FIGURE 1 \| Heat map (Kernel Density) showing the distribution in China of the 31,146 Rhododendron specimens present in our dataset. TABLE 3 \| Distribution of grid cells, Rhododendron specimens and corresponding taxa (525 in total) according to different pH ranges in top and subsoil. pH(−log(H+)) Topsoil Subsoil Number of grid cells Number of specimens Number of taxa Number of grid cells Number of specimens Number of taxa <4.5 793 (0.007%) 0 0 19, 041 (0.2%) 48 (0.15%) 20 (3.8%) 4.5–5.5 2, 118, 460 (19.6%) 10, 234 (32.9%) 409 (77.9%) 1, 833, 929 (17.0%) 10, 180 (32.7%) 404 (76.9%) 5.5–7.2 4, 447, 296 (41.2%) 14, 442 (46.4%) 420 (80.0%) 3, 886, 312 (36.0%) 13, 602 (43.7%) 408 (77.7%) 7.2–8.5 4, 192, 123 (38.8%) 6, 467 (20.8%) 335 (63.8%) 4, 851, 365 (44.9%) 7, 290 (23.4%) 340 (64.8%) >8.5 45, 511 (0.4%) 3 (0.01%) 2 (0.38%) 213, 536 (2.0%) 26 (0.08%) 10 (1.90%) textual document geocoding (Faure et al., 2017). The complexity of the address string format in Chinese text-based geocoding compounds these challenges (Tian et al., 2016). Frontiers in Plant Science \| www.frontiersin.org Frontiers in Plant Science \| www.frontiersin.org Effect of Soil CaCO3 of 571 species occurring in China by Shrestha et al. (2018). However, the hotspots still had a regional distribution, even within the Himalaya–Hengduan Mountains area. The factors that inﬂuence Rhododendron distribution cannot be only attributed to the climate (Kumar, 2012), soil conditions also inﬂuence Rhododendron growth in a speciﬁc region (Esen et al., 2004; Kamei et al., 2009). By geocoding the specimens, we were able to link specimen location to the soil characteristics of HWSD, thus leading to a metadata analysis of soil conditions for rhododendrons in nature. Rhododendrons have shallow, ﬁbrous root systems that are restricted to the upper soil (Kinsman, 1999; Hales et al., 2009), thus our analysis focused primarily on topsoil (0–30 cm). The majority of the plants were located in topsoils with low pH (below or around 6.5), without presence of CaCO3. Importantly, the pH values reported in the HWSD database are In our ﬁnal extracted dataset of 31,146 specimens, the majority (72.3%) of reported Chinese Rhododendron taxa were present and covered all nine subgenera. This was representative of the genus Rhododendron in China. The heat map showed that the hotspots were mainly distributed in the Himalaya–Hengduan Mountains area in Sichuan and Yunnan province, which is consistent with previously reports of the general distribution of Rhododendron species in China (Milne et al., 2010; Liu et al., 2014; Yan et al., 2015). Furthermore, the heat map (Figure 1) matched well with the spatial patterns of total Rhododendron species richness in China estimated in 50 km × 50 km grid cells resulted from analyzing 556 Rhododendron species out October 2018 \| Volume 9 \| Article 1538 Frontiers in Plant Science \| www.frontiersin.org 5 Wang et al. Prediction of Lime Tolerance in Rhododendron FIGURE 2 \| Percentages of grid cells (10,804,183 in total), Rhododendron specimens (31,146 in total) and taxa (525 in total) in different pH values for topsoil (pH measured in a soil-water solution and different specimens of one taxon could be found at diverse pH values). FIGURE 2 \| Percentages of grid cells (10,804,183 in total), Rhododendron specimens (31,146 in total) and taxa (525 in total) in different pH values for topsoil (pH measured in a soil-water solution and different specimens of one taxon could be found at diverse pH values). TABLE 4 \| Distribution of grid cells, Rhododendron specimens and taxa (525 in total) according to different CaCO3 ranges in top and subsoil. Effect of Soil CaCO3 CaCO3 (% weight) Topsoil Subsoil Number of grid cells Number of specimens Number of taxa Number of grid cells Number of specimens Number of taxa <2 6, 460, 264 (59.8%) 24, 554 (78.8%) 505 (96.2%) 5, 675, 678 (52.5%) 23, 711 (76.1%) 503 (95.8%) 2–5 887, 088 (8.2%) 724 (2.3%) 166 (31.6%) 1, 107, 230 (10.2%) 1, 289 (4.1%) 200 (38.1%) 5–15 2, 611, 756 (24.2%) 5, 865 (18.8%) 303 (57.7%) 3, 066, 378 (28.4%) 6, 140 (19.7%) 304 (57.9%) >15 845, 075 (7.8%) 3 (0.01%) 2 (0.38%) 954, 897 (8.8%) 6 (0.02%) 3 (0.57%) soil-water pH, which is approximately 0.9 unit higher than pH in 1 M KCl (Kabala et al., 2016). The results are in accordance with the general trend that Rhododendron cannot grow well in neutral or alkaline soils and is considered to be a calcifugous genus (Kinsman, 1999; Demasi et al., 2015a). Kinsman (1999) reported that at almost all sites in Northwest Yunnan where Rhododendrons grow in shallow soils overlying limestone, the soils still had pH values of less than 6. The pH values in Kinsman’s report were also measured in soil-water solutions, but in Kinsman’s study most soil samples were taken 3–10 cm below ground surface while the topsoil pH measured in HWSD contains the 0–30 cm layer. The lower pH in topsoils (0–10 cm) can be explained by organic horizon acidiﬁcation and rhizosphere interaction of plant roots (James and Riha, 1987; Wang et al., 2016). Using data analysis, we tried to predict at taxon level (species, and CaCO3 content > 2% weight. According to HWSD, pH values from 7.2 to 8.5 are indicative of carbonate rich soils which chemically form less available carbonates aﬀecting nutrient availability (P, Fe) (Batjes et al., 2012). Further, the bioavailability of trace element cations such as copper (Cu), Zn, nickel (Ni), cadmium (Cd), and lead (Pb) and their concentration in plants is signiﬁcantly reduced at pH > 7.0 (Valentinuzzi et al., 2015). In addition, calcifugous plants are intolerant to high concentrations of Ca2+ when combined with high pH (Vicherova et al., 2015). Using these thresholds, we compiled a list of 76 potentially lime- tolerant taxa. For 61 of these, at least 30% of the specimens were geocoded in locations from which the topsoil pH was above 7.2 and CaCO3 > 2% weight. For several taxa mentioned in our prediction, support was found in literature for their lime tolerance. Effect of Soil CaCO3 Prediction of Lime Tolerance in Rhododendron FIGURE 3 \| Percentages of grid cells (10,804,183 in total), Rhododendron specimens (31,146 in total) and taxa (525 in total) in different concentrations (% weight) of CaCO3 in topsoil (Different specimens of one taxon could be found at diverse CaCO3 concentrations). FIGURE 3 \| Percentages of grid cells (10,804,183 in total), Rhododendron specimens (31,146 in total) and taxa (525 in total) in different concentrations (% weight) of CaCO3 in topsoil (Different specimens of one taxon could be found at diverse CaCO3 concentrations). R. yunnanense was also found on gravely loam soils of high pH in the Sichuan region of China, the same as R. davidsonianum (Reid et al., 1998). Kaisheva (2006) classiﬁed R. phaeochrysum, R. balfourianum, R. primuliﬂorum, R. telmateium, R. yunnanense, and R. trichocladum as lime tolerant species. R. fortunei was mentioned as a promising gene resource for breeding lime tolerant rhododendrons (Shujun et al., 2008). Kinsman (1999) determined R. primuliﬂorum deﬁnitely to be growing under alkaline soil conditions (pH 7.4–7.9), while R. rupicola var. chryseum and R. proteoides, which were also in our prediction list, were found in soil pH values below 6. R. yunnanense was also found on gravely loam soils of high pH in the Sichuan region of China, the same as R. davidsonianum (Reid et al., 1998). Kaisheva (2006) classiﬁed R. phaeochrysum, R. balfourianum, R. primuliﬂorum, R. telmateium, R. yunnanense, and R. trichocladum as lime tolerant species. R. fortunei was mentioned as a promising gene resource for breeding lime tolerant rhododendrons (Shujun et al., 2008). Kinsman (1999) determined R. primuliﬂorum deﬁnitely to be growing under alkaline soil conditions (pH 7.4–7.9), while R. rupicola var. chryseum and R. proteoides, which were also in our prediction list, were found in soil pH values below 6. and R. rupicola var. muliense were predicted with lime tolerance. The bias of geocoding or soil data of HWSD may result in the fact that R. rupicola was not predicted as a lime tolerant taxon in our study. It does, however, not exclude a lime tolerant potential because congeneric species often have similar ecological characteristics and use similar resources. Furthermore, similar interspeciﬁc associations can strengthen their competitive ability and promote local exclusion to non-congeneric species to obtain more living space (Yuan et al., 2018). Field and experimental work should be carried out to conﬁrm this in the future. Effect of Soil CaCO3 McAleese and Rankin (2000) showed that R. primuliﬂorum could grow in soil-water pH, which is approximately 0.9 unit higher than pH in 1 M KCl (Kabala et al., 2016). The results are in accordance with the general trend that Rhododendron cannot grow well in neutral or alkaline soils and is considered to be a calcifugous genus (Kinsman, 1999; Demasi et al., 2015a). Kinsman (1999) reported that at almost all sites in Northwest Yunnan where Rhododendrons grow in shallow soils overlying limestone, the soils still had pH values of less than 6. The pH values in Kinsman’s report were also measured in soil-water solutions, but in Kinsman’s study most soil samples were taken 3–10 cm below ground surface while the topsoil pH measured in HWSD contains the 0–30 cm layer. The lower pH in topsoils (0–10 cm) can be explained by organic horizon acidiﬁcation and rhizosphere interaction of plant roots (James and Riha, 1987; Wang et al., 2016). and CaCO3 content > 2% weight. According to HWSD, pH values from 7.2 to 8.5 are indicative of carbonate rich soils which chemically form less available carbonates aﬀecting nutrient availability (P, Fe) (Batjes et al., 2012). Further, the bioavailability of trace element cations such as copper (Cu), Zn, nickel (Ni), cadmium (Cd), and lead (Pb) and their concentration in plants is signiﬁcantly reduced at pH > 7.0 (Valentinuzzi et al., 2015). In addition, calcifugous plants are intolerant to high concentrations of Ca2+ when combined with high pH (Vicherova et al., 2015). Using these thresholds, we compiled a list of 76 potentially lime- tolerant taxa. For 61 of these, at least 30% of the specimens were geocoded in locations from which the topsoil pH was above 7.2 and CaCO3 > 2% weight. For several taxa mentioned in our prediction, support was found in literature for their lime tolerance. McAleese and Rankin (2000) showed that R. primuliﬂorum could grow in soil with pH > 7 (sampled in topsoil 10–20 cm), as well as R. telmateium and R. yunnanense, followed by R. vernicosum. Using data analysis, we tried to predict at taxon level (species, subspecies or varieties) which potential genetic resources might exhibit lime tolerance. Thresholds were set at UQ for pH > 7.2 October 2018 \| Volume 9 \| Article 1538 Frontiers in Plant Science \| www.frontiersin.org 6 Wang et al. Prediction of Lime Tolerance in Rhododendron Wang et al. Frontiers in Plant Science \| www.frontiersin.org Effect of Soil CaCO3 Evidence for lime tolerance can also be found in species that co-occur in similar habitats. An example is R. nivale, a perennial evergreen undershrub with a height of 30–120 cm, distributed in the northeastern and southeastern areas of the Tibet Autonomous Region of China, Nepal, India, Bhutan, and Sikkim (Guo et al., 2017). This species co-occurred with R. primuliﬂorum as the representative alpine species in the snowy mountains in the northwest of Yunnan (Xu et al., 1996), which grows in limestone crevices shared with Paraquilegia anemonoides, another species associated uniquely with limestone. This shows that R. nivale has a similar habitat as R. primuliﬂorum, indicating R. nivale may also has a good lime tolerance potential, as well as its subspecies R. nivale subsp. austral and R. nivale subsp. boreale. Moreover, R. rupicola was also found together with R. primuliﬂorum in limestone crevices shared with P. anemonoides (McAleese and Rankin, 2000). Although R. rupicola was included in our dataset but not predicted as a lime tolerant taxon, its two varieties R. rupicola var. chryseum In addition, some predicted lime tolerant taxa can also be supported from horticultural and physiological studies. R. micranthum was proven to be able to grow in containers on lime-supplemented media (McAleese and Rankin, 2000). R. x pulchrum is a well-known taxon for landscaping and breeding. It’s cultivar R. x pulchrum ‘Sen-e-oomurasaki’ showed extremely low chlorosis and mortality rates and high ferric chelate reductase activity in high pH hydroponic conditions, resulted in iron eﬃcient genetic resources for azalea cultivation and gardening in calcareous soils (Demasi et al., 2015a,b). This information demonstrates that our strategy to use UQ values of pH > 7.2 and CaCO3 > 2% weight in topsoils at the habitats of Rhododendron taxa can be used as an eﬃcient indicator for prediction of lime tolerance. As a supplement to the identiﬁcation of taxa where lime tolerance was already reported, our work has also identiﬁed some potentially interesting taxa for which (to our knowledge) no information is available on potential lime tolerance. October 2018 \| Volume 9 \| Article 1538 Frontiers in Plant Science \| www.frontiersin.org 7 Prediction of Lime Tolerance in Rhododendron Wang et al. Wang et al. Effect of Soil CaCO3 Prediction of Lime Tolerance in Rhododen TABLE 5 \| Lower quartile (LQ), median and upper quartile (UQ) values of pH and CaCO3 in topsoils and specimen distribution for the 76 taxa with UQ values of topsoil pH > 7.2, CaCO3 > 2% weight and a minimum of 10 specimens. Topsoil pH Topsoil CaCO3 (% weight) Latin name Sub-genus∗ Total number of specimens LQ Median UQ % of specimens LQ Median UQ % of specimens 4.5 ∼5.5 5.5 ∼7.2 7.2 ∼8.5 <2 2 ∼5 5 ∼15 R. aganniphum var. ﬂavorufum 5 16 6.55 6.6 8 12.5 50 37.5 0.05 0.1 7 62.5 0 37.5 R. alutaceum 5 10 6.4 6.45 8 20 40 40 0 0 7 60 0 40 R. amesiae 1 13 5.1 6.5 8 38.5 15.4 46.2 0 0 7 53.8 7.7 38.5 R. anthopogonoides 1 112 7.1 7.3 7.8 0 39.3 60.7 2.6 4.6 5.5 23.2 51.8 25 R. balfourianum var. aganniphoides 5 23 6.5 6.6 8 4.3 56.5 39.1 0 0.1 7 60.9 4.3 34.8 R. brachyanthum 1 10 6.6 7.8 7.8 10 20 70 0.1 9.3 9.3 30 0 70 R. bracteatum 1 50 6.5 7.8 8 2 46 52 0 7 7 48 0 52 R. bulu 1 17 6.5 6.5 7.7 17.6 47.1 35.3 0 0 7.4 64.7 5.9 29.4 R. capitatum 1 90 7.3 7.3 7.8 1.1 20 78.9 4.6 5 6 12.2 42.2 45.6 R. chrysodoron 1 42 4.9 6 7.8 26.2 45.2 28.6 0 0 9.3 71.4 0 28.6 R. concinnum 1 391 6.5 6.5 7.8 15.1 55.5 29.4 0 0 7 70.6 0 29.4 R. crassimedium 8 12 4.9 7.15 7.8 33.3 16.7 50 0 4.7 9.3 50 0 50 R. crinigerum 5 67 6.5 6.8 8 16.4 38.8 44.8 0 0.2 7 55.2 1.5 43.3 R. dauricum 4 242 5.7 6.4 7.8 0.8 69 30.2 0 0 3.4 69.8 11.2 19 R. davidii 5 126 6.5 6.5 7.8 23 46.8 30.2 0 0 7 69.8 1.6 28.6 R. davidsonianum 1 316 6.5 6.5 7.7 15.8 57.6 26.6 0 0 7 73.4 0 26.6 R. excellens 1 56 4.8 5.75 7.8 50 23.2 26.8 0 0 5.95 73.2 1.8 25 R. fastigiatum 1 134 6.5 6.6 7.8 20.9 47 32.1 0 0.1 7 67.9 3.7 28.4 R. ﬂavidum 1 33 6.5 6.5 8 9.1 51.5 39.4 0 0 7 60.6 3 36.4 R. Effect of Soil CaCO3 fortunei 5 319 4.9 6.5 7.8 44.2 29.5 26.3 0 0 2.6 73.7 1.9 24.5 R. hanceanum 1 78 4.9 4.9 7.8 65.4 3.8 30.8 0 0 9.3 69.2 0 30.8 R. hemitrichotum 3 28 7.2 8 8 0 25 75 1.3 7 7 25 10.7 64.3 R. henryi 6 115 4.8 5 7.8 63.5 1.7 34.8 0 0 9.3 65.2 0.9 33.9 R. henryi var. dunnii 6 19 4.8 4.9 7.8 57.9 0 42.1 0 0 9.3 57.9 0 42.1 R. hongkongense 6 18 5 7.8 7.8 44.4 0 55.6 0 9.3 9.3 44.4 0 55.6 R. hunnewellianum 5 36 6.5 6.55 7.8 13.9 44.4 41.7 0 0 9.3 58.3 5.6 36.1 R. impeditum 1 17 6.5 6.6 7.9 17.6 41.2 41.2 0 0.1 7 58.8 5.9 35.3 R. intricatum 1 69 6.5 6.5 8 2.9 56.5 40.6 0 0 7 59.4 1.4 39.1 R. kongboense 1 13 6.5 6.5 8 0 53.8 46.2 0 0 7 53.8 0 46.2 R. longesquamatum 5 38 6.5 6.5 7.8 15.8 57.9 26.3 0 0 2.6 73.7 2.6 23.7 October 2018 \| Volume 9 \| Article 1538 8 Prediction of Lime Tolerance in Rhododendron Wang et al. TABLE 5 \| Continued Topsoil pH Topsoil CaCO3 (% weight) Latin name Sub-genus∗ Total number of specimens LQ Median UQ % of specimens LQ Median UQ % of specimens 4.5 ∼5.5 5.5 ∼7.2 7.2 ∼8.5 <2 2 ∼5 5 ∼15 R. mainlingense 1 10 6.5 6.5 7.7 10 60 30 0 0 7.4 70 0 30 R. micranthum 1 826 6.5 6.6 8 5.3 51.3 43.3 0 0.1 7 54.7 6.3 39 R. molle 7 179 4.9 5.9 7.8 44.1 28.5 27.4 0 0 2.6 72.6 3.9 23.5 R. mucronatum 8 46 4.9 6.6 7.8 32.6 26.1 41.3 0 0.1 9.3 56.5 0 43.5 R. mucronulatum 4 322 6.4 6.5 7.8 0.9 67.1 32 0 0 7.4 63.7 3.4 32.9 R. nanpingense 8 13 4.8 7.8 7.8 46.2 0 53.8 0 9.3 9.3 46.2 0 53.8 R. nivale 1 83 6.4 6.5 7.8 16.9 49.4 33.7 0 0 3.2 62.7 26.5 10.8 R. nivale subsp. australe 1 22 6.5 7.8 8 18.2 27.3 54.5 0 7 7 45.5 0 54.5 R. nivale subsp. boreale 1 55 6.5 6.5 7.85 9.1 63.6 27.3 0 0 3 72.7 3.6 23.6 R. oreodoxa 5 582 6.5 6.5 7.8 3.8 69.4 26.8 0 0 7 73.2 0.7 26.1 R. Effect of Soil CaCO3 oreodoxa var. shensiense 5 14 6.4 6.9 7.8 14.3 42.9 42.9 0 0.1 9.3 57.1 0 42.9 R. ovatum 6 556 4.8 5.1 7.8 59.2 14.6 26.3 0 0 2.6 73.7 3.1 23.2 R. phaeochrysum 5 143 6.5 6.6 8 5.6 59.4 35 0 0.1 7 65 0 35 R. phaeochrysum var. levistratum 5 45 6.5 8 8 4.4 40 55.6 0 7 7 44.4 0 55.6 R. praestans 5 26 4.8 6.5 8 46.2 26.9 26.9 0 0 7 73.1 0 26.9 R. primuliﬂorum 1 69 6.5 6.5 8 24.6 47.8 27.5 0 0 5 69.6 7.2 23.2 R. proteoides 5 19 4.9 4.9 8 52.6 10.5 36.8 0 0 7 63.2 0 36.8 R. przewalskii 5 131 6.5 7.1 7.9 0.8 58 41.2 0 3 6 44.3 27.5 28.2 R. pubescens 3 41 5.1 6.5 7.8 34.1 34.1 31.7 0 0 7 68.3 2.4 29.3 R. purdomii 5 54 6.5 6.9 8 1.9 50 48.1 0 0.15 7 51.9 7.4 40.7 R. radendum 1 24 7.8 7.8 7.8 0 20.8 79.2 5.95 9.3 9.3 20.8 4.2 75 R. redowskianum 9 68 6.5 7.8 7.8 4.4 35.3 60.3 0 9.3 9.3 39.7 0 60.3 R. rigidum 1 29 6.5 6.6 8 0 55.2 44.8 0 0.1 7 55.2 3.4 41.4 R. roxieanum var. cucullatum 5 22 6.8 8 8 4.5 27.3 68.2 0.2 7 7 31.8 0 68.2 October 2018 \| Volume 9 \| Article 1538 9 Prediction of Lime Tolerance in Rhododendron Wang et al. Prediction of Lime Tolerance in Rh TABLE 5 \| Continued Topsoil pH Topsoil CaCO3 (% weight) Latin name Sub-genus∗ Total number of specimens LQ Median UQ % of specimens LQ Median UQ % of specimens 4.5 ∼5.5 5.5 ∼7.2 7.2 ∼8.5 <2 2 ∼5 5 ∼15 R. rufescens 1 67 6.5 6.5 8 4.5 47.8 47.8 0 0 7 52.2 4.5 43.3 R. rufum 5 77 6.4 8 8 6.5 37.7 55.8 0 3 7 41.6 11.7 46.8 R. rupicola var. chryseum 1 24 6.5 7.8 7.8 16.7 29.2 54.2 0 7 9.3 45.8 0 54.2 R. rupicola var. muliense 1 23 6.5 8 8 0 47.8 52.2 0 7 7 47.8 0 52.2 R. sanguineum var. himertum 5 13 6.8 6.8 7.8 23.1 46.2 30.8 0.2 0.2 9.3 69.2 0 30.8 R. sargentianum 1 14 6.5 6.5 7.9 21.4 35.7 42.9 0 0 2.6 57.1 28.6 14.3 R. Effect of Soil CaCO3 saxatile 8 12 7.8 7.8 7.8 0 0 100 9.3 9.3 9.3 0 0 100 R. sphaeroblastum 5 43 6.5 6.5 8 9.3 44.2 46.5 0 0 7 53.5 0 46.5 R. stamineum 6 573 4.9 6.3 7.8 44.9 26 29.1 0 0 9.3 70.9 2.3 26.9 R. tatsienense 1 142 6.5 6.5 7.7 14.8 50.7 34.5 0 0 7 65.5 0 34.5 R. telmateium 1 58 6.5 6.5 8 17.2 56.9 25.9 0 0 7 74.1 0 25.9 R. thymifolium 1 77 7.1 7.3 8 1.3 36.4 62.3 3 6 7 20.8 27.3 51.9 R. trichocladum 2 169 6.5 6.6 8 8.9 53.8 37.3 0 0.1 7 62.7 0 37.3 R. trichostomum 1 135 6.5 6.5 8 2.2 60.7 37 0 0 7 62.2 2.2 35.6 R. vellereum 5 30 6.5 7.1 7.7 13.3 46.7 40 0 5 7.4 46.7 13.3 40 R. vernicosum 5 374 6.5 6.5 8 8.3 57.8 34 0 0 7 66 2.4 31.6 R. vialii 6 16 6.4 6.7 7.8 18.8 37.5 43.8 0 0.15 9.3 56.3 0 43.8 R. wasonii 5 76 6.5 6.5 7.7 14.5 44.7 40.8 0 0 7.4 59.2 1.3 39.5 R. websterianum 1 43 6.5 6.6 8 4.7 48.8 46.5 0 1.6 7 53.5 4.7 41.9 R. x detonsum 5 16 5.1 7.4 8 37.5 12.5 50 0 3.6 7 50 0 50 R. x pulchrum 8 28 4.8 7.8 7.8 42.9 3.6 53.6 0 5.8 9.3 46.4 3.6 50 R. yunnanense 1 447 6.5 6.5 7.7 21.5 52.6 26 0 0 2.6 74 1.8 24.2 ∗1, Rhododendron; 2, Pseudazalea; 3, Pseudorhodorastrum; 4, Rhodorastrum; 5, Hymenanthes; 6, Azaleastrum; 7, Pentanthera; 8, Tsutsusi; 9, Therorhodion. October 2018 \| Volume 9 \| Article 1538 10 Prediction of Lime Tolerance in Rhododendron Wang et al. 4 \| Geocoded locations of herbarium specimens from 10 Rhododendron taxa predicted in this study and supported by literature as being lime tolerant: onianum (A), R. fortunei (B), R. micranthum (C), R. nivale (D), R. phaeochrysum (E), R. primuliﬂorum (F), R. telmateium (G), R. trichocladum (H), sum (I), R. yunnanense (J). ▲: Specimen at a location with pH < 7.2 and CaCO3 content < 2% weight in topsoils. : Specimen at a location with and CaCO3 content > 2% weight in topsoil. FIGURE 4 \| Geocoded locations of herbarium specimens from 10 Rhododendron taxa predicted in this study and supported by literature as being lime tolerant: R. SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2018.01538/ full#supplementary-material TABLE S1 \| The number of specimens for each taxon in 525 Rhododendron taxa. TABLE S1 \| The number of specimens for each taxon in 525 Rhododendron taxa. FUNDING This work was supported by the Chinese Scholarship Council (201608360101). This work was supported by the Chinese Scholarship Council (201608360101). This work was supported by the Chinese Scholarship Council (201608360101). CONCLUSION Our results showed that information present in herbarium specimens might be used to identify potentially interesting genetic resources in Rhododendron. Geocoding of the Chinese text-based locations of plant specimens into latitude/longitude pairs makes it possible to study plant distribution as well as to connect the distribution data to soil database of the habitats. This approach makes use of a large number of plant samples, which increases the reliability of the obtained results. The combination of geocoded specimen information and the soil database led to identiﬁcation of valuable resources at taxon level for tolerance against lime soils characterized by a high pH and high CaCO3 concentrations. The predicted tolerant taxa in this study pave the way for an in depth evaluation of potential resources for lime tolerance and in the long term for using genetic material in breeding or studies of abiotic stress. Moreover, the continental- scale dataset with both comprehensive taxonomic coverage and Effect of Soil CaCO3 davidsonianum (A), R. fortunei (B), R. micranthum (C), R. nivale (D), R. phaeochrysum (E), R. primuliﬂorum (F), R. telmateium (G), R. trichocladum (H), R. vernicosum (I), R. yunnanense (J). ▲: Specimen at a location with pH < 7.2 and CaCO3 content < 2% weight in topsoils. : Specimen at a location with pH > 7.2 and CaCO3 content > 2% weight in topsoil. October 2018 \| Volume 9 \| Article 1538 11 Frontiers in Plant Science \| www.frontiersin.org Prediction of Lime Tolerance in Rhododendron Wang et al. Other studies conﬁrmed the value of herbarium specimens, especially those with detailed locations, as a source of information. Contrasting phenological responses of Rhododendron species to the Himalayan climate were reported by analyzing Rhododendron herbarium specimens located in Lijiang County, Yunnan, China (Hart et al., 2014). Elevational distribution of native orchid species compiled from CVH were investigated and the results illustrated that the elevational pattern of orchid species richness in Yunnan was collectively shaped by several mechanisms related to geometric constraints, size of the land area, and environments (Zhang et al., 2015). The research location of above two studies were at county or provincial level, the geocoded locations of Rhododendron specimens in our study is valuable to extract continental climate, altitude, physical and chemical data of soil, or other environmental databases with GIS layer, and extend the research area to the national level. For instance, a continent-wide dataset of occurrence records with geographical coordinates of Australian grasses were used for predicting species’ tolerance to salinity and alkalinity (Saslis-Lagoudakis et al., 2015). An increasing number of plant specimens, especially those with GPS information, have been collected in recent decades. These herbarium specimens can be used to study the plants’ tolerance to abiotic stresses, their phylogenesis, evolution, conservation eﬀorts for endangered plants, eﬀects of climate change or land/forest economy studies assisted by ﬁeld or experimental work. geocoded locations can be connected with other GIS layers such as the WorldClim database3 for ecological or evolutionary researches. ACKNOWLEDGMENTS Biodiversity occurrence data were provided by: Chengdu Institute of Biology, Chinese Academy of Sciences; Wuhan Botanical Garden, Chinese Academy of Sciences; Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences; Northwest Institute of Plateau Biology, Chinese Academy of Sciences; Guangxi Institute of Botany, Chinese Academy of Sciences; South China Botanical Garden, Chinese Academy of Sciences; Institute of Applied Ecology, Chinese Academy of Sciences; Kunming Institute of Botany, Chinese Academy of Sciences; Lushan Botanical Garden, Jiangxi Province and Chinese Academy of Sciences; Institute of Botany, Jiangsu Province and Chinese Academy of Sciences; Chinese National Herbarium, Institute of Botany, Chinese Academy of Sciences; WUK Herbarium, Northwest Agriculture & Forestry University. [Accessed through Chinese Virtual Herbarium (CVH) data portal, http://www.cvh.ac.cn/ (latest portal: www.cvh.ac.cn), in October, 2016]. AUTHOR CONTRIBUTIONS SW and JVH originally formulated the idea. M-CVL and LL developed methodology. SW performed statistical analyses and wrote the manuscript. All authors edited and approved the ﬁnal manuscript. 3 http://www.worldclim.org/ REFERENCES Res. Earth Surface 114, F03013. doi: 10.1029/2008jf001168 Shrestha, N., Su, X. Y., Xu, X. T., and Wang, Z. H. (2018). The drivers of high Rhododendron diversity in south-west China: does seasonality matter? J. Biogeogr. 45, 438–447. doi: 10.1111/jbi.13136 Hart, R., Salick, J., Ranjitkar, S., and Xu, J. C. (2014). Herbarium specimens show contrasting phenological responses to Himalayan climate. Proc. Natl. Acad. Sci. U.S.A. 111, 10615–10619. doi: 10.1073/pnas.1403376111 p Hijmans, R. J. (2017). geosphere: Spherical Trigonometry. R package version 1.5-7. Shujun, Y., Ximing, C., and Jeongsik, L. E. E. (2008). Alkali tolerance of Rhododendron fortunei in subirrigated ebb & ﬂow bench systems with hydroponics. Acta Hortic. Sin. 35, 715–720. James, B. R., and Riha, S. J. (1987). Forest soil organic horizon acidiﬁcation - eﬀects of temperature, time, and solution soil ratio. Soil Sci. Soc. Am. J. 51, 458–462. doi: 10.2136/sssaj1987.03615995005100020037x Tian, Q., Ren, F., Hu, T., Liu, J. T., and Li, R. C. (2016). Using an optimized Chinese address matching method to develop a geocoding service: a case study of Shenzhen, China. ISPRS Int. J. Geo Inf. 5:65. doi: 10.3390/ijgi50 50065 Johnston, R. J., Holland, B. M., and Yao, L. Y. (2016). Individualized geocoding in stated preference questionnaires: implications for survey design and welfare estimation. Land Econ. 92, 737–759. doi: 10.3368/le.92.4.737 Valentinuzzi, F., Mimmo, T., Cesco, S., Al Mamun, S., and Santner, J. (2015). The eﬀect of lime on the rhizosphere processes and elemental uptake of white lupin. Environ. Exp. Bot. 118, 85–94. doi: 10.1016/j.envexpbot.2015. 06.010 Kabala, C., Musztyfaga, E., Galka, B., Labunska, D., and Manczynska, P. (2016). Conversion of soil pH 1:2.5 KCl and 1:2.5 H2O to 1:5 H2O: Conclusions for soil management, environmental monitoring, and international soil databases. Polish J. Environ. Stud. 25, 647–653. doi: 10.15244/pjoes/ 61549 Vicherova, E., Hajek, M., and Hajek, T. (2015). Calcium intolerance of fen mosses: Physiological evidence, eﬀects of nutrient availability and successional drivers. Persp. Plant Ecol. Evol. Syst. 17, 347–359. doi: 10.1016/j.ppees.2015. 06.005 Kaisheva, M. V. (2006). The Eﬀect of Metals and Soil pH on the Growth of Rhododendron and Other Alpine plants in Limestone Soil. Doctoral thesis, University of Edinburgh, London. Wang, X. J., Tang, C. X., Severi, J., Butterly, C. R., and Baldock, J. A. (2016). Rhizosphere priming eﬀect on soil organic carbon decomposition under plant species diﬀering in soil acidiﬁcation and root exudation. New Phytol. 211, 864–873. doi: 10.1111/nph.13966 Kamei, J., Pandey, H. N., and Barik, S. K. (2009). REFERENCES Bergman, E. J., Nieto, F. J., Bersch, A. J., Buckingham, W. R., and Malecki, K. (2012). Using tax parcel data to compare geocoding accuracy of urban and non-urban addresses in Wisconsin. Am. J. Epidemiol. 175, S116–S116. Amtmann, A., Bohnert, H. J., and Bressan, R. A. (2005). Abiotic stress and plant genome evolution. Search New models. Plant Physiol. 138, 127–130. doi: 10. 1104/pp.105.059972 Demasi, S., Caser, M., Kobayashi, N., Kurashige, Y., and Scariot, V. (2015a). Hydroponic screening for iron deﬁciency tolerance in evergreen azaleas. Notulae Botanicae Horti Agrobotanici Cluj Napoca 43, 210–213. doi: 10.15835/ nbha4319929 Batjes, N., Dijkshoorn, K., Van Engelen, V., Fischer, G., Jones, A., et al. (2012). Harmonized World Soil Database, 1.2 Edn, eds F. Nachtergaele, H. Van Velthuizen, L. Verelst, and D. Wiberg. Rome: Food and Agricultural Organisation of United Nations. Demasi, S., Handa, T., and Scariot, V. (2015b). Ferric chelate reductase activity under iron deﬁciency stress in azalea. Int. J. Hortic. Floric. 3, 157–160. October 2018 \| Volume 9 \| Article 1538 Frontiers in Plant Science \| www.frontiersin.org 12 Prediction of Lime Tolerance in Rhododendron Wang et al. Dimichele, W. A., Phillips, T. L., and Olmstead, R. G. (1987). Opportunistic evolution - abiotic environmental-stress and the fossil record of plants. Rev. Pal. Palynol. 50, 151–178. doi: 10.1016/0034-6667(87) 90044-3 Marais, D. L. D., and Juenger, T. E. (2010). Pleiotropy, plasticity, and the evolution of plant abiotic stress tolerance. Year Evol. Biol. 1206, 56–79. doi: 10.1111/j. 1749-6632.2010.05703.x McAleese, A. J., and Rankin, D. W. H. (2000). Growing rhododendrons on limestone soils: is it really possible? J. Am. Rhododendr. Soc. 54, 126–134. Esen, D., Zedaker, S. M., Kirwan, J. L., and Mou, P. (2004). Soil and site factors inﬂuencing purple-ﬂowered rhododendron (Rhododendron ponticum L.) and eastern beech forests (Fagus orientalis Lipsky) in Turkey. For. Ecol. Manag. 203, 229–240. doi: 10.1016/j.foreco.2004.07.052 Mengel, K. (1994). Iron availability in plant tissues-iron chlorosis on calcareous soils. Plant Soil 165, 275–283. doi: 10.1007/BF00 008070 Fang, M. Y., Fang, R. Z., He, M. Y., Hu, L. Z., Yang, H. B., et al. (2005). “Rhododendron,” in Flora of China, eds Z. Y. Wu, P. H. Raven, and D. Y. Hong (Beijing: Science Press and Missouri Botanical Garden Press), 260–455. Milne, R. I., Davies, C., Prickett, R., Inns, L. H., and Chamberlain, D. F. (2010). Phylogeny of Rhododendron subgenus Hymenanthes based on chloroplast DNA markers: between-lineage hybridisation during adaptive radiation? Plant Syst. Evol. 285, 233–244. REFERENCES doi: 10.1007/s00606-010- 0269-2 Faure, E., Danjou, A. M. N., Clavel-Chapelon, F., Boutron-Ruault, M. C., and Dossus, L. (2017). Accuracy of two geocoding methods for geographic information system-based exposure assessment in epidemiological studies. Environ. Health 16:1. doi: 10.1186/s12940-017-0217-5 Moeltner, K., Blinn, C. E., and Holmes, T. P. (2017). Forest pests and home values: the importance of accuracy in damage assessment and geocoding of properties. J. For. Econ. 26, 46–55. doi: 10.1016/j.jfe.2017.02.002 Giel, P., and Bojarczuk, K. (2002). The eﬀect of high concentration of selected calcium salts on development of microcuttings of rhododendron R. ‘Catawbiense Grandiﬂorum’ in in vitro cultures. Dendrobiology 48, 23–29. Mordhorst, A. P., Kullik, C., and Preil, W. (1993). Ca uptake and distribution in Rhododendron selected for lime tolerance. Gartenbauwissenschaft 58, 111–116. Giel, P., and Bojarczuk, K. (2011). Eﬀects of high concentrations of calcium salts in the substrate and its pH on the growth of selected rhododendron cultivars. Acta Soc. Bot. Pol. 80, 105–114. doi: 10.5586/asbp.2011.021 Nuvolone, D., Santini, M., Pepe, P., and Cipriani, F. (2016). Impacts of geocoding quality in environmental epidemiology studies: two case-studies in Tuscany Region (Central Italy). Epidemiol. Prev. 40, 44–50. doi: 10.19191/EP16.1.P044. 013 Soc. Bot. Pol. 80, 105–114. doi: 10.5586/asbp.2011.021 Goldberg, D. W. (2011). Advances in geocoding research and practice. Trans. GIS 15, 727–733. doi: 10.1111/j.1467-9671.2011.01298.x Reid, M. S., Dodge, L. L., and Zagory, E. M. (1998). High pH tolerant rootstocks for rhododendrons and azaleas. Slosson Rep. 3, 96–98. Goldstein, N. D., Auchincloss, A. H., and Lee, B. K. (2014). A no-cost geocoding strategy using R. Epidemiology 25, 311–313. doi: 10.1097/Ede. 0000000000000052 Romeiras, M. M., Figueira, R., Duarte, M. C., Beja, P., and Darbyshire, I. (2014). Documenting biogeographical patterns of African Timber species using herbarium records: A conservation perspective based on native trees from Angola. PLoS One 9:e103403. doi: 10.1371/journal.pone. 0103403 Guo, X., Shang, X. F., Li, B., Zhou, X. Z., and Wen, H. (2017). Acaricidal activities of the essential oil from Rhododendron nivale Hook. f. and its main compound, delta-cadinene against Psoroptes cuniculi. Vet. Parasitol. 236, 51–54. doi: 10. 1016/j.vetpar.2017.01.028 Saslis-Lagoudakis, C. H., Hua, X., Bui, E., Moray, C., and Bromham, L. (2015). Predicting species’ tolerance to salinity and alkalinity using distribution data and geochemical modelling: a case study using Australian grasses. Ann. Bot. 115, 343–351. doi: 10.1093/aob/mcu248 Hales, T. C., Ford, C. R., Hwang, T., Vose, J. M., and Band, L. E. (2009). Topographic and ecologic controls on root reinforcement. J. Geophys. REFERENCES Tree species distribution and its impact on soil properties, and nitrogen and phosphorus mineralization in a humid subtropical forest ecosystem of northeastern India. Can. J. For. Res. Rev. Canadienne De Recherche Forestiere 39, 36–47. doi: 10.1139/X08-151 Kinsman, D. J. J. (1999). Rhododendrons in yunnan, China - pH of associated soils. J. Am. Rhododendr. Soc. Available at: https://scholar.lib.vt. edu/ejournals/JARS/v54n3/v54n3-mcaleese.htm Wen, J. (1999). Evolution of eastern Asian and eastern North American disjunct distributions in ﬂowering plants. Annu. Rev. Ecol. Syst. 30, 421–455. doi: 10. 1146/annurev.ecolsys.30.1.421 Xing, Y. W., Gandolfo, M. A., and Linder, H. P. (2015). The Cenozoic biogeographical evolution of woody angiosperms inferred from fossil distributions. Glob. Ecol. Biogeogr.24, 1290–1301. doi: 10.1111/geb. 12383 Kumar, P. (2012). Assessment of impact of climate change on Rhododendrons in Sikkim Himalayas using Maxent modelling: limitations and challenges. Biodiv. Conserv. 21, 1251–1266. doi: 10.1007/s10531-012- 0279-1 Xu, L. H., Li, Q. R., and Jiang, C. L. (1996). Diversity of soil actinomycetes in Yunnan. China. Appl. Environ. Microbiol. 62, 244–248. Lang, D. W. (2016). REmap: Create html maps by Echarts. R package version 0.3.2. g ( ) p p y p g Liu, J. Q., Duan, Y. W., Hao, G., Ge, X. J., and Sun, H. (2014). Evolutionary history and underlying adaptation of alpine plants on the Qinghai-Tibet Plateau. J. Syst. Evol. 52, 241–249. doi: 10.1111/jse.12094 Yan, L. J., Liu, J., Moller, M., Zhang, L., and Zhang, X. M. (2015). DNA barcoding of Rhododendron (Ericaceae), the largest Chinese plant genus in biodiversity October 2018 \| Volume 9 \| Article 1538 Frontiers in Plant Science \| www.frontiersin.org 13 Prediction of Lime Tolerance in Rhododendron Wang et al. hotspots of the Himalaya-Hengduan Mountains. Mol. Ecol. Resour. 15, 932– 944. doi: 10.1111/1755-0998.12353 Conﬂict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Yuan, Z. L., Wei, B. L., Chen, Y., Jia, H. R., and Wei, Q. N. (2018). How do similarities in spatial distributions and interspeciﬁc associations aﬀect the coexistence of Quercus species in the Baotianman National Nature Reserve. Henan, China. Ecol. Evol. 8, 2580–2593. doi: 10.1002/ece3. 3863 Copyright © 2018 Wang, Leus, Van Labeke and Van Huylenbroeck. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). Frontiers in Plant Science \| www.frontiersin.org Zhang, S. B., Chen, W. Y., Huang, J. L., Bi, Y. F., and Yang, X. F. (2015). Orchid species richness along elevational and environmental gradients in Yunnan. China. PLoS One 10, e0142621. doi: 10.1371/journal.pone. 0142621 hotspots of the Himalaya-Hengduan Mountains. Mol. Ecol. Resour. 15, 932– 944. doi: 10.1111/1755-0998.12353 REFERENCES The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Zhang, S. B., Chen, W. Y., Huang, J. L., Bi, Y. F., and Yang, X. F. (2015). Orchid species richness along elevational and environmental gradients in Yunnan. China. PLoS One 10, e0142621. doi: 10.1371/journal.pone. 0142621 October 2018 \| Volume 9 \| Article 1538 Frontiers in Plant Science \| www.frontiersin.org 14
https://openalex.org/W1553788181	https://europepmc.org/articles/pmc4515928?pdf=render	English	null	The pain of a heart being broken: pain experience and use of analgesics by caregivers of patients with Alzheimer’s disease	BMC psychiatry	2,015	cc-by	6,948	© 2015 Wojtyna and Popiołek. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http:// creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Wojtyna and Popiołek BMC Psychiatry (2015) 15:176 DOI 10.1186/s12888-015-0571-1 Wojtyna and Popiołek BMC Psychiatry (2015) 15:176 DOI 10.1186/s12888-015-0571-1 * Correspondence: ewa.wojtyna@us.edu.pl 1Institute of Psychology, University of Silesia, Katowice, ul. Grażyńskiego 53, 40-126 Katowice, Poland Full list of author information is available at the end of the article The pain of a heart being broken: pain experience and use of analgesics by caregivers of patients with Alzheimer’s disease Ewa Wojtyna1* and Katarzyna Popiołek2 Abstract Background: It has been observed that psychical suffering (e.g. the feeling of losing a significant person) tends to reduce the physical pain tolerance threshold, as well as to increase the subjective sense of painfulness. The purpose of this study was to assess pain sensation among a group of caregivers of patients with Alzheimer’s disease, and to determine the psychological factors (emotional and relational) that contribute to both pain perception and coping with pain via the use of analgesics. Methods: The study comprised 127 caregivers of patients with Alzheimer’s disease. Questionnaires were used to elicit pain intensity, strength of emotional relationship between caregiver and patient, sense of painfulness of the loss experienced, depression level, and somatic ailments. Results: A large majority (87.4 %) of participants reported pain complaints, while 93 % took analgesics without a doctor’s recommendation at least once a week; 8 % took painkillers daily. The strongest predictors of both perceived pain and tendency to use analgesics were sense of loss and painfulness of loss in relation to the patient’s deteriorating condition. Results: A large majority (87.4 %) of participants reported pain complaints, while 93 % took analgesics without a doctor’s recommendation at least once a week; 8 % took painkillers daily. The strongest predictors of both d d d l f l d f l f l l h p y g p perceived pain and tendency to use analgesics were sense of loss and painfulness of loss in relation to the patient’s deteriorating condition. Conclusions: The pain experienced by caregivers may be connected to social pain resulting from the experience of losing someone they are close to. Caregivers may resort to excessive use of analgesics as a pain-coping strategy. Keywords: Pain, Caregiver, Alzheimer’s disease, Social pain, Analgesic Background unsuitable for the weather, smoking near flammable ob- jects, etc. Other problems that are extremely burden- some for caregivers are disturbances of circadian rhythm as well as patients’ tendency to wander about, sometimes away from their residences, which demands caregivers’ vigilance and application of special safety precautions. Patients may also be indiscriminate in their behavior. Another difficult and potentially painful symptom is that patients may not recognize their close family members, including caregivers themselves. Alzheimer’s disease (AD) is the most frequently occur- ring type of dementia. It is forecasted that in 2050 in Europe alone there will be 14.5 million people suffering from the disease [1]. Dysmnesia may be the most dis- tinctive symptom of AD; yet for those who interact with the patient, the most troublesome symptoms are often behavior disorders, which accompany the deterioration of cognitive functions [2, 3]. Patients begin to behave differently than they used to: often they are aggressive, impulsive, sexually disinhibited or neglectful of their per- sonal hygiene. A grave problem is that they endanger their own lives, as well as the lives of other people, e.g. by turning on gas or water, putting on clothes that are The burden of giving both physical and mental care to the AD patient results in frequent occurrence of symp- toms of depression [4, 5], chronic fatigue, exhaustion, and sorrow [6] among caregivers. Our research to date has demonstrated that an important predictor of depres- sion and exhaustion among people performing as care- givers is the type of bond they have with the person for whom they provide care: the stronger the relationship, * Correspondence: ewa.wojtyna@us.edu.pl 1Institute of Psychology, University of Silesia, Katowice, ul. Grażyńskiego 53, 40-126 Katowice, Poland Full list of author information is available at the end of the article Wojtyna and Popiołek BMC Psychiatry (2015) 15:176 Page 2 of 8 a substantial constituent of suffering, understood as a construct joining the physical, psychic, and existential dimensions of pain [11]. The affective dimension of pain is determined by limbic system structures – the same system referred to earlier as the neuropsychological basis of the social pain mechanism. The generation of social pain requires awareness of a threat to vital interpersonal relations and one’s own position as a participant in these relations. Background The notion of ‘sense of loss’ is used to connote the caregiver’s cognitive assessment of differences between the functioning of the AD patient now versus before the disease manifested. We assume that observations of that change create the basis for triggering the social pain mechanism. We therefore categorize the caregiver’s as- sessment of the patient’s deteriorating condition as the ‘sense of loss,’ and the suffering connected to that assess- ment as the ‘painfulness of loss.’ the greater the emotional exhaustion of the caregiver [6]. Moreover, more profound feelings of loss connected with the deteriorating psychophysical condition of the patient and changes in her or his personality are associ- ated with caregivers’ intensified symptoms of depression and diminished sense of personal achievement related to providing the care. It is interesting to note that a more profound sense of loss, coupled with strong positive emotional bonds and positive experiences with the pa- tient when she or he was still healthy, has been associ- ated with intensified somatic, rather than affective, symptoms of depression [7]. The results described are in alignment with the con- cept of social pain [8, 9]. In the present context, social pain is defined as a sensation of pain that arises not due to any tissue lesion, but as a result of a disturbance in substantial interpersonal relations. Frustration connected with the loss of a significant person and feelings of in- justice or being treated unfairly result in activation of the anterior cingulate cortex, the same central nervous system area that is activated in the process of perceiving physical pain and that is connected with its affective di- mension. It has been observed that social pain tends to reduce the physical pain tolerance threshold, as well as to increase the subjective sense of painfulness [9]. It is therefore reasonable to expect that the sense of losing a significant person as a consequence of AD-induced de- terioration of the individual’s psychophysical condition is connected with intensification of pain sensations and/or a decreased capacity to cope with pain. The aims of the study were to assess pain sensation among a group of caregivers of AD patients, and to de- termine the psychological factors (emotional and rela- tional) that contribute to both pain perception and coping with pain by use of analgesics. Methods The design was cross-sectional. Caregivers obtained data in the form of medical documentation on the severity of dementia symptoms from doctors. The remaining data were collected from caregivers via questionnaire. Partici- pation by caregivers was voluntary, and participants were informed about the opportunity to receive treat- ment from a psychologist or psychotherapist if they felt that this would be helpful to them. Written informed consent for participation in the study was obtained from all participants. Ethics approval for the study was pro- vided by the Ethics Committee of the University of Silesia. Given the current research context that associates a more frequent occurrence of somatic symptoms of de- pression in caregivers who have had positive experiences with the patient and who assess their bond with the pa- tient as very positive and close, we hypothesized that pain sensations of these caregivers would be relatively intensified and the use of analgesics more frequent. Participants Th l Our focus upon pain perception by caregivers looking after AD patients requires introduction of some specific terminology we use to describe that experience. Pain is a multi-dimensional phenomenon, and – in accordance with the definition provided by the International Associ- ation for the Study of Pain – an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage [10]. This definition assumes the possibility that a pain sensa- tion arises via both nociceptive and non-nociceptive mechanisms. In the latter case, pain may have its origins not only in damage suffered by nerve fibers (neuropathic pain), but also in psychogenic background. The sample comprised 127 primary caregivers of pa- tients with AD. The participants were recruited from caregivers who accompanied the persons entrusted to their care for consultation in psycho-geriatric, geriatric, or psychiatric outpatient clinics, as well as in day care centers for patients with AD in several provinces of Poland. The criteria for inclusion in the study were: at least 1 year of providing care, at least 6 h of care every day, and direct contact with the patient, who had at least a moderate level of dementia. Exclusion criteria were: caregiver experiencing serious emotional disturbances and caregiver being a minor. The study was conducted between July and December 2013. Of all the caregivers who met the inclusion criteria (n = 358), 35.5 % agreed to complete the questionnaire. Those who declined to participate cited lack of time (94.4 %) or lack of interest The psychic dimension of pain is associated with at- tachment of an emotionally negative character to pain sensations. Pain is perceived as unpleasant and is con- nected with fear/anxiety or depression. This response is Wojtyna and Popiołek BMC Psychiatry (2015) 15:176 Page 3 of 8 (5.6 %). All those who declined and cited lack of time as their reason were also the sole caregiver of an individual with AD. The sociodemographic and care-related char- acteristics of participants are provided in Table 1. [6] developed by the authors was used. Participants Th l The tool consists of three questions, to which the participant responds by marking her/his choice on a visual-analog scale of 0 to 10: 1) ‘How strong is your emotional bond with the pa- tient?’ (0 = ‘no emotional bond’; 10 = ‘extremely strong emotional bond’); 2) ‘What are your experiences with the patient from the pre-onset time?’ (0 = ‘extremely negative’; 10 = ‘extremely positive’); and 3) ‘Do you feel a sense of loss associated with the deteriorating condition of the patient?’ (0 = ‘no sense of loss’; 10 = ‘extremely se- vere sense of loss’). The results from these three scales Thermometer of Caregiver-Patient Strength of Emotional Relation To characterize the quality of the emotional bond be- tween the caregiver and the AD patient, the Thermom- eter of Caregiver-Patient Strength of Emotional Relation Table 1 Characteristics of participants Characteristics N (%) M SD Range Gender Female 73 (57.5) Male 54 (42.5) Age [years] 53.21 21.84 31–83 Marital status Married 91 (71.7) Widowed 6 (4.7) Single 11 (8.7) Divorced 19 (14.9) Level of education Primary 9 (7.1) Technical/vocational 32 (25.2) Secondary 29 (22.8) University 57 (44.9) Employment status Working 76 (59.8) Living with patients 116 (91.3) Sole caregiver 98 (77.1) Time spent caring [h/day] 10.34 6.87 6–19 Time since onset of AD [months] 26.57 8.42 12–37 Pain Occurrence of pain-related somatic diseases 28 (22.0) Spondylosis 23 (18.1) Rheumatoid arthritis 4 (3.1) Phantom limb pain 1 (0.8) Pain level 4.13 4.28 0–10 Taking analgesics without a doctor’s recommendation [number of doses/week] 3.24 1.99 0–9 Strength of relationship with the patient Strength of emotional relationship 8.02 0.67 4–10 Positive experiences 7.21 1.99 0–10 Sense of loss 7.13 1.64 1–10 ‘Painfulness’ of loss 6.45 2.12 1–10 Depression score 5.58 3.65 0–19 Note: n = 127 Wojtyna and Popiołek BMC Psychiatry (2015) 15:176 Page 4 of 8 almost daily (4 to 7 days a week) use of analgesics by the caregiver, without a doctor’s recommendation. Predictor variables included a strong caregiver–patient relation- ship and substantial ‘painfulness’ of loss (with ‘strong’ and ‘substantial’ defined as scores at least half a standard deviation above the arithmetical means), depression (de- fined as a score > 8 on the HADS-D depression scale [12, 13]), significant perceived pain (defined as a score > 3 on the NRS scale), and the occurrence of somatic dis- eases accompanied by physical pain. were subsequently coded and introduced to statistical analyses as: ‘Strength of emotional relationship’, ‘Positive experiences’, and ‘Sense of loss’. Results It has been found that eighty-two caregivers were off- spring of the patient; 39 were partners or spouses; and six were friends. A vast majority of caregivers were sole carers and also lived with the patient. Female caregivers slightly outnumbered males (57.5 %). Most offspring and all friends who acted as caregivers had at least a second- ary education. Painfulness of Loss The sense of painfulness of loss experienced by the care- giver in relation to the person she or he cares for was assessed in a similar way to the strength of emotional re- lationship between the caregiver and the patient, using the 0 to 10 single-item visual-analog scale, where 0 indi- cated no or minimum painfulness of loss experienced, and 10 signified maximum intensity of that feeling. Somatic Ailments Questionnaire This tool was developed by the authors to collect data on somatic diseases from which the caregiver was suffering, and drugs taken (Additional file 1). Separate questions concerned the frequency of taking analgesics, including those that do not require a prescription and are taken in- dependently, without a doctor’s recommendation. Nearly half of caregivers (45.7 %) had had at least one chronic disease diagnosed: 32.3 % had been treated for hypertension, 14.2 % for circulatory insufficiency, 15.7 % for respiratory tract diseases, 20.4 % for diabetes, and 9.4 % took medication for depression. Among the somatic ailments, pain was distinguished. Participants were asked to state the level of pain they had experienced, on average, during the previous week, using the Numerical Rating Scale (from 0 = no pain, to 10 = maximum pain). Just 28 participants (22 %) had diagnoses of patho- logical states that included pain sensations, whereas 87.4 % reported having pain complaints during the week prior to the study; most often these were backache or headache. The mean intensity of the pain was 4.13 for the group as a whole (SD = 4.28; min = 0; max = 10). Depression To assess the level of depressive symptoms in caregivers, the Polish version of depression scale from the Hospital Anxiety and Depression Scale (HADS-D) was used [12, 13]. Participants express their agreement with each of seven statements, using a 4-point scale wherein higher scores rep- resent greater levels of depressive affect, and a total score above 8 indicates a risk of clinical depression [12, 13]. Cronbach’s alpha for the HADS-D rendered a reliability level of 0.87 in our study. Data on participants’ psychophysical condition and the strength of their relationship to the patient are summa- rized in Table 1. Participants defined their relationships with patients as strong, and most reported that they had had positive experiences with the patient in the past. Most participants declared a substantial sense of loss and painfulness associated with that loss. Depression was present in 35.4 % of the participants. Statistical analysis Only one participant had never used analgesics with- out a doctor’s recommendation (that person had been diagnosed with rheumatoid arthritis, and a daily dose of analgesics was recommended by a doctor). Another par- ticipant took analgesics just once or twice a month. The vast majority of participants (93 %) took analgesics with- out consulting a doctor at least once a week, while 8 % took such medication every day (Fig. 1). Most frequently, participants were using analgesics available without pre- scription, but 9.4 % of the sample admitted to taking the prescribed drugs more often than directed by the doctor. On average, participants were taking about three doses of analgesics per week and were taking them independ- ently, without consulting a doctor (Table 1). The STATISTICA 10 package was used for analysis. De- scriptive data are presented as means and standard devi- ations for continuous data and as frequencies and percentages for categorical data. Kendall’s tau-b correlation coefficients were computed to examine the association of the three key participant variables (strength of the caregiver-patient emotional bond, sense of painfulness of loss connected with the de- terioration of the patient’s condition, and intensification of depression) with the outcome variables of the inten- sity of pain perceived and frequency of taking analgesics without a doctor’s recommendation. Logistic regression was used to model the probability of taking analgesics as a function of caregivers’ psycho- physical condition and the caregiver–patient emotional bond. The dependent variable was established as daily or It was demonstrated that the higher the intensity of physical pain perceived, the higher the assessment of Page 5 of 8 Wojtyna and Popiołek BMC Psychiatry (2015) 15:176 1-2 times a week 29.1% a month 5.5% <1 time a month 1.6% daily 7.9% several times a week 55.9% 1-3 times Fig. 1 Frequency of taking analgesics without a doctor’s recommendation by caregivers of patients with Alzheimer’s disease a month 5.5% 1-3 times 1 Frequency of taking analgesics without a doctor’s recommendation by caregivers of patients with Alzheimer’s disease symptoms. Similarly, symptoms of depression nearly dou- bled the chance of using analgesics without a doctor’s recommendation. painfulness of loss, and the more acute the sense of loss (Table 2). Moreover, pain correlated positively, albeit weakly, with the strength of emotional relationship and intensity of depression. Discussion d This aspect, however, requires further stud- ies on caregivers dealing with AD patients. bonds with the patients more often demonstrate somatic, rather than affective, symptoms related to depression. In this study we have demonstrated a positive correlation be- tween depression and intensification of pain. This suggests that the pain reported by participants in our study could also be a symptom of depression. However, the strength of the correlation between depression and pain was low. Moreover, 35.4 % of the participants had significant symp- toms of depression, whereas as many as 87.4 % reported the presence of pain. Therefore, not all the pain complaints were necessarily related to the previously occurring depres- sive or somatic disturbances. Taking into account the fact that most of the pain complaints may be tension-related, we are reminded here of one of the most significant models of caregiving – the stress process model [14]. In accordance with this model, loss of closeness in the caregiver–patient relationship is a secondary stressor, which in turn may negatively influence psychophysical health. Our study indi- cates that the sense of painfully losing a significant person whom the caregiver has been looking after might consti- tute a source of social pain, which could lead to the occur- rence of pain complaints, by stimulating the anterior cingulate cortex and insula [8, 15–18]. The closeness of the caregiver-patient relationship has emerged as an import predictor of the psychophysical condition of the caregiver. Fauth et al. [22] demonstrated that the sense of being very close and losing this close- ness during the provision of care to the patient is associ- ated with more negative self-assessments of physical health. The results of our study supplement those find- ings by indicating a probable mechanism of origin for this perceived health deterioration: more frequent and profound experiences of pain are an important influence on the general evaluation of one’s health and on the phenomenon of catastrophizing [23, 24]. Catastrophizing is a condition in which an individual perceives her/his ailments as more severe and dangerous, and the possibil- ity of coping with them less probable, than is actually the case. Experiencing unexplained pain – and social pain may be defined as such pain – is conducive to a tendency to catastrophize. A loss can be perceived as particularly painful when the caregiver lacks support and understanding from others; yet support for caregivers is often substantially limited. Discussion d The painfulness of loss connected with deterioration of the psychophysical condition of a patient with AD in- creased the chance of taking analgesics nearly four times more than did the sensation of pain itself, or the occur- rence of a somatic disease that included pain among its Earlier studies [7] have indicated that people providing care for patients with AD who also have stronger emotional Table 2 Correlations of factors related to giving care with pain level and use of analgesics Factors related to giving care Pain level Frequency of taking analgesics (without doctor’s recommendation) Strength of relationship with the patient Strength of emotional relationship 0.20* 0.13 Positive experiences −0.14 −0.09 Sense of loss 0.31* 0.38* ‘Painfulness’ of loss 0.47* 0.51* Depression 0.19* 0.24** Note: n = 127; p < 0.05; p < 0.01; *p < 0.001 of factors related to giving care with pain level and use of analgesics care Pain level Frequency of taking analgesics (without doctor’s recommendati Table 2 Correlations of factors related to giving care with pain level and use of analgesics Factors related to giving care Pain level Frequency of taking analgesics Note: n = 127; p < 0.05; p < 0.01; *p < 0.001 Wojtyna and Popiołek BMC Psychiatry (2015) 15:176 Page 6 of 8 Page 6 of 8 Table 3 Probability of using analgesics as a function of caregivers’ psychophysical condition and emotional relationship with the patient Characteristics of the caregiver Analgesics (without doctor’s recommendation) OR 95 % CI p Strength of relationship with the patient Strength of emotional relationship 1.17 0.22, 2.16 0.012 Positive experiences 0.87 0.09, 1.94 0.035 Sense of loss 3.64 1.85, 7.99 <0.001 ‘Painfulness’ of loss 4.82 2.03, 8.77 <0.001 Depression 1.86 0.51, 3.21 <0.001 Pain 1.26 0.70, 2.01 <0.001 Somatic pain-related diseases 1.13 0.48, 2.12 0.002 Note: n = 127 Table 3 Probability of using analgesics as a function of caregivers’ psychophysical condition and emotional relationship with the others [19, 20]. In Poland, the cultural message that one must sacrifice oneself for other family members is strong. Admitting in public that providing care is tire- some or too difficult, or that the caregiver is unable to cope or already perceives the patient as a stranger, may be met with negative reactions. Frustration connected with lack of understanding, unjust treatment and unfair judgment are other documented sources of social pain [18, 19, 21]. Discussion d The frequency of taking analgesics without a doctor’s recommendation also correlated positively with painful- ness of loss, a sense of loss, and more intense symptoms of depression (Table 2). It is worth noting that the strength of those relationships was greater than that be- tween the same variables and intensity of pain. Our study shows that many caregivers of AD patients report pain complaints and that they frequently take an- algesics without consulting a doctor. The sense of loss and painfulness of loss turned out to be the strongest predictors not only of pain intensity but also of using analgesics. This result confirms our hypothesis that pain sensations are relatively intensified and that the use of analgesics is more frequent among this subset of care- givers of AD patients. Conversely, caregivers who did not perceive the loss as emotionally painful also did not experience as much physical pain. Our results therefore also support the concept of social pain [8, 9]. The logistic regression analysis indicated that the sense of loss and painfulness of loss were the factors that most profoundly increased the probability of using analgesics without consulting a doctor (Table 3). Discussion d The results would be worth verifying in a longitudinal study, to confirm whether the dynamics of the caregiver-patient relation- ship influence the level of pain complaints reported, and whether these complaints are affected by changes in strategy for coping with stress, including changes related to the use of analgesics (which Zhu et al. [29] observed in relation to other drugs). ‘painfulness of loss.’ Apart from that, perceiving the suf- fering of a loved one may result in fear of losing that person, and may thereby trigger the mechanism of social pain, particularly in people with an anxious/preoccupied attachment style [27]. who were town or city dwellers. We should also mention that only 35.5 % of the caregivers invited to participate in the study took part in it. In the future, it would be worth- while to devote attention to the group of caregivers who refused to participate in the study and who cited lack of time as the reason. One can expect that provision of care by those people is connected with specific bio-social costs. The variable availability of medical procedures and social care in Poland does not allow direct extrapolation of the re- sults to the general population of caregivers of AD patients. It would be worthwhile to repeat the study in different cul- tural and economic conditions. In future studies, it would also be useful to consider measures of frustration levels as- sociated with the accessibility of medical and social services. Also, higher levels of depression are related to lower assessments of health [28]. In our study we observed that depression is connected with a more profound ex- perience of pain. This result is important inasmuch as the study demonstrated a fairly frequent occurrence of intensified depression symptoms (in more than one- third of the participants), whereas only 9.4 % of care- givers were receiving anti-depressive treatment. This speaks volumes about the insufficiently met require- ments of support and treatment for caregivers. Our study was cross-sectional. The results would be worth verifying in a longitudinal study, to confirm whether the dynamics of the caregiver-patient relation- ship influence the level of pain complaints reported, and whether these complaints are affected by changes in strategy for coping with stress, including changes related to the use of analgesics (which Zhu et al. [29] observed in relation to other drugs). Zhu et al. Conclusions Pain is a symptom commonly experienced by caregivers of AD patients. This phenomenon may be connected to social pain resulting from the experience of losing some- one to whom they feel close. Caregivers may resort to excessive use of analgesics as a pain-coping strategy. In light of the facts mentioned here, it is justified that caregivers should be given not only adequate pharmaco- logical assistance but also support in the form of coping measures for working through the grief they experience while the patient is still alive. This need is especially crit- ical when one considers that the process of losing a person to AD often takes several years. This type of assistance is likely to reduce the psychical suffering of caregivers, but also to minimize their tendency for excessive reliance on analgesics. Discussion d [29] demonstrated that more intense symp- toms of depression are connected with an increased probability of taking drugs – both those prescribed by a doctor (OR, 1.112), and those available over the counter (OR, 1.117). In their study, analgesics were not listed by participants as the most frequently used drugs (the list included antihypertensive drugs, drugs used for dyslipid- emia, and psychotropic drugs). By contrast, in our study analgesics were the drugs taken most frequently by care- givers. A possible explanation for the discrepancy is that a specialist consultation is relatively difficult to secure, whereas painkillers may be purchased over the counter any time of the day (or night) in close proximity to one’s place of residence [30, 31]. What is more, one can ex- pect analgesics to bring an actual reduction in pain, whether of organic or psychosomatic origin. For ex- ample, DeWall et al. [32] found that paracetamol (Acet- aminophen) could reduce social pain. It is also worth stressing that over-the-counter painkillers are a class of very cheap drugs that bring relief; one pill costs less than a cigarette. The finding that caregivers use painkillers seems, in this situation, to reflect the reality that such drugs are one of the most readily available strategies for coping with social pain. Our study was based on single-item thermometer scales, which render the subjective level of sensations ex- perienced by caregivers. The interpretation of results from such measures carries the risk of errors associated with excessive generalization and confounding of con- cepts. Such factors as, for example, the sense of loss are aggregate variables that may comprise several other con- structs. In a future study it would be worthwhile to ex- plore more comprehensively notions of loss and of being close to the patient. Future studies should also investigate such variables as adequacy of social support, or levels of distress and effi- ciency of strategies for coping with stress, as potential buffers to the costs of providing care to an AD patient. Abbreviations AD: Alzheimer’s Disease; HADS: Hospital Anxiety and Depression Scale; NRS: Numerical Rating Scale. AD: Alzheimer’s Disease; HADS: Hospital Anxiety and Depression Scale; NRS: Numerical Rating Scale. Limitations and advice for future studies Our study has several limitations. The first is that the group of participants was not representative of caregivers at large. Participants were recruited from Polish caregivers Additional file Additional file 1: Somatic Ailments Questionnaire (originally in polish). (DOCX 18 kb) Additional file 1: Somatic Ailments Questionnaire (originally in polish). (DOCX 18 kb) Discussion d This is often a matter of time restrictions, among other factors: caregivers do not have free time to meet other people who could be sources of support (in our study, the caregivers spent an average of 10 h a day providing care). Our participants were often the only person actually providing care to the patient (77.1 %). In the Polish setting, obtaining additional assistance may also be hindered by limited availability of institutional as well as financial support. Another reason for the absence of support may be the limited possibility of honest dis- cussion concerning the difficulties and internal conflicts experienced, for fear of being assessed negatively by Monin and Schulz [11, 25] state that caregivers’ per- ceptions of suffering experienced by their loved ones re- sult in increasing distress and negative health consequences in the caregivers themselves. AD patients surely experience frustration and suffering, yet the sever- ity of these experiences may be overestimated by their caregivers, as demonstrated by Schulz et al. [26]. In our research, we did not include the variable ‘perception of patient’s suffering’, yet we expect that it could be part of the aggregated variable we did study, namely the Wojtyna and Popiołek BMC Psychiatry (2015) 15:176 Page 7 of 8 who were town or city dwellers. We should also mention that only 35.5 % of the caregivers invited to participate in the study took part in it. In the future, it would be worth- while to devote attention to the group of caregivers who refused to participate in the study and who cited lack of time as the reason. One can expect that provision of care by those people is connected with specific bio-social costs. The variable availability of medical procedures and social care in Poland does not allow direct extrapolation of the re- sults to the general population of caregivers of AD patients. It would be worthwhile to repeat the study in different cul- tural and economic conditions. In future studies, it would also be useful to consider measures of frustration levels as- sociated with the accessibility of medical and social services. Our study was cross-sectional. References 24. Wojtyna E. Irrational Suffering – An impact of cognitive behavioural therapy on the depression level and the perception of pain in cancer patients. In: L'Abate L, editor. Mental illnesses - evaluation, treatments and implications. Rijeka: InTech; 2012. p. 227–44. 1. Mura T, Dartigues J-F, Berr C. How many dementia cases in France and Europe? Alternative projections and scenarios 2010–2050. Eur J Neurol. 2010;17:252–9. 2. Cummings J, Zhong K. Treatments for behavioural disorders in neurodegenerative diseases: drug development strategies. Nat Rev Drug Discov. 2006;5(1):64–74. 25. Monin JK, Schulz R. The effects of suffering in chronically ill older adults on the health and well-being of family members involved in their care. GeroPsych. 2010;23(4):207–13. 3. Lanctôt K, Herrmann N, Rothenburg L, Eryavec G. Behavioral correlates of GABAergic disruption in AD. Int Psychogeriatr. 2007;19(1):151–8. 3. Lanctôt K, Herrmann N, Rothenburg L, Eryavec G. Behavioral correlates of GABAergic disruption in AD. Int Psychogeriatr. 2007;19(1):151–8. 26. Schulz R, Monin JK, Czaja SJ, Lingler JH, Beach SR, Martire LM, et al. Measuring the experience and perception of suffering. Gerontologist. 2010;50(6):774–84. 4. Croog S, Burleson J, Sudilovsky A, Baume R. Spouse caregivers of Alzheimer patients: problem responses to caregiver burden. Aging Ment Health. 2006;10(2):87–100. 4. Croog S, Burleson J, Sudilovsky A, Baume R. Spouse caregivers of Alzheimer patients: problem responses to caregiver burden. Aging Ment Health. 2006;10(2):87–100. 27. Frías MT, Shaver PR. The moderating role of attachment insecurities in the association between social and physical pain. J Res Pers. 2014;53:193–200. 5. Mausbach BT, Roepke SK, Chattillion EA, Harmell AL, Moore R, Romero-Moreno R, et al. Multiple mediators of the relations between caregiving stress and depressive symptoms. Aging Ment Health. 2012;16(1):27–38. 5. Mausbach BT, Roepke SK, Chattillion EA, Harmell AL, Moore R, Romero-Moreno R, et al. Multiple mediators of the relations between caregiving stress and depressive symptoms. Aging Ment Health. 2012;16(1):27–38. 28. Cooper C, Katona C, Orrell M, Livingston G. Coping strategies, anxiety and depression in caregivers of people with AD. Int J Geriatr Psychiatry. 2008;23(9):929–36. 6. Wojtyna E, Popiołek K. Character of the relationship with Alzheimer patient and the psychological costs of care. Polish Psychological Bulletin. 2012;43(4):244–52. 6. Wojtyna E, Popiołek K. Character of the relationship with Alzheimer patient and the psychological costs of care. Polish Psychological Bulletin. 2012;43(4):244–52. 29. Zhu CW, Scarmeas N, Ornstein K, Albert M, Brandt J, Blacker D, et al. Competing interests p g The authors declare that they have no competing interests. Page 8 of 8 Page 8 of 8 Wojtyna and Popiołek BMC Psychiatry (2015) 15:176 Received: 30 October 2014 Accepted: 22 July 2015 23. Sullivan MJL. The Communal Coping Model of pain catastrophizing: Clinical and research implications. Can Psychol. 2012;53:32–41. Authors’ contributions 17. DeWall CN, Baumeister RF. Alone but feeling no pain: effects of social rejection on physical pain tolerance and pain threshold, affective forecasting, and interpersonal empathetic concern. J Pers Soc Psychol. 2006;91:1–15. EW participated in the design of the study, collected the data, performed the statistical analysis, interpreted and discussed the data and drafted the manuscript. KP participated in the design of the study, interpreted and discussed the data and drafted the manuscript. Both authors read and approved the final manuscript. 18. Kross E, Berman MG, Mischel W, Smith EE, Wager TD. Social rejection shares somatosensory representations with physical pain. PNAS. 2011;108:6270–5. 19. Nordgren LF, Banas K, MacDonald G. Empathy gap for social pain: Why people underestimate the pain of social suffering. J Pers Soc Psychol. 2011;100:120–8. Acknowledgments h h f ll The authors gratefully acknowledge the time and effort of the caregivers, without whom this study would not have been possible. The authors would like to thank Dr. Janusz Wojtyna for assistance during data acquisition. There were no funding sources for the current study. 20. Wirga M, Wojtyna E. Udręki zdrowego umysłu. Neuropsychologia cierpienia [Torments of a healthy mind. Neuropsychology of suffering]. In: Binnebesel J, Błeszyński J, Domżał Z, editors. Wielowymiarowość cierpienia [Multidimensionality of suffering]. Łódź: Wydawnictwo Naukowe WSEZ; 2010. p. 31–51. Author details 1 21. MacDonald G, Leary MR. Why does social exclusion hurt? The relationship between social and physical pain. Psychol Rev. 2005;131:202–23. 1Institute of Psychology, University of Silesia, Katowice, ul. Grażyńskiego 53, 40-126 Katowice, Poland. 2Faculty in Katowice, University of Social Sciences and Humanities, ul. Techników 9, 40-326 Katowice, Poland. 22. Fauth E, Hess K, Piercy K, Norton M, Corcoran C, Rabins P, et al. Caregivers’ relationship closeness with the person with dementia predicts both positive and negative outcomes for caregivers’ physical health and psychological well-being. Aging Ment Health. 2012;16(6):699–711. Received: 30 October 2014 Accepted: 22 July 2015 References Witte W, Stein C. History, definitions, and contemporary viewpoints. In: Kopf A, Patel NB, editors. Guide to pain management in low-resource settings. Seattle: International Association for the Study of Pain; 2010: p. 3–7 y ; p 11. Monin J, Schulz R. Interpersonal effects of suffering in older adult caregiving relationships. Psychol Aging. 2009;24(3):681–95. y p 11. Monin J, Schulz R. Interpersonal effects of suffering in older adult caregiving relationships. Psychol Aging. 2009;24(3):681–95. 12. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67(6):361–70. 12. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67(6):361–70. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit 13. Majkowicz M. Praktyczna ocena efektywności opieki paliatywnej – wybrane techniki badawcze [Practical assessment of efficiency of palliative care – selected study techniques]. In: de Walden-Gałuszko K, Majkowicz M, editors. Ocena jakości opieki paliatywnej w teorii i praktyce [Assessment of palliative care in theory and practice]. Gdańsk: Akademia Medyczna w Gdańsku; 2000. p. 21–42. 13. Majkowicz M. Praktyczna ocena efektywności opieki paliatywnej – wybrane techniki badawcze [Practical assessment of efficiency of palliative care – selected study techniques]. In: de Walden-Gałuszko K, Majkowicz M, editors. Ocena jakości opieki paliatywnej w teorii i praktyce [Assessment of palliative care in theory and practice]. Gdańsk: Akademia Medyczna w Gdańsku; 2000. p. 21–42. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: References Health-care use and cost in dementia caregivers: Longitudinal results from the Predictors Caregiver Study. Alzheimers Dement. 2015;11(4):444–454. 7. Wojtyna E, Banout M, Popiołek K. Objawy depresyjne a więź emocjonalna z chorym u opiekunów pacjentów z otępieniem alzheimerowskim [Depression and emotional bond with patient among caregivers of Alzheimer patients]. Czasopismo Psychologiczne [Journal of Psychology], 2014; 20(2): 259–265 30. Sussman S, Rohrbach LA, Spruijt-Metz D, Barnett E, Lisha N, Sun P. One-year prediction of pain killer use among at-risk older teens and emerging adults. J Drug Educ. 2012;42:195–210. 31. Wojtyna E. O bólu i radzeniu sobie z nim w kontekście współczesnej mentalności [About pain and coping with it in the context of contemporary mentality]. In: Górnik-Durose M, editor. Kultura współczesna a zdrowie [Contemporary culture and health]. Sopot: Gdańskie Wydawnictwo Psychologiczne; 2013. p. 137–53. 31. Wojtyna E. O bólu i radzeniu sobie z nim w kontekście współczesnej mentalności [About pain and coping with it in the context of contemporary mentality]. In: Górnik-Durose M, editor. Kultura współczesna a zdrowie [Contemporary culture and health]. Sopot: Gdańskie Wydawnictwo Psychologiczne; 2013. p. 137–53. 8. Eisenberger N. The neural bases of social pain: Evidence for shared representations with physical pain. Psychosom Med. 2012;74(2):126–35. 8. Eisenberger N. The neural bases of social pain: Evidence for shared representations with physical pain. Psychosom Med. 2012;74(2):126–35. 9. Eisenberger NI, Jarcho JM, Lieberman MD, Naliboff BD. An experimental study of shared sensitivity to physical pain and social rejection. Pain. 2006;126:132–8. 9. Eisenberger NI, Jarcho JM, Lieberman MD, Naliboff BD. An experimental study of shared sensitivity to physical pain and social rejection. Pain. 2006;126:132–8. 32. DeWall CN, MacDonald G, Webster GD, Masten CL, Baumeister RF, Powell C, et al. Acetaminophen reduces social pain: Behavioral and neural evidence. Psychol Sci. 2010;21:931–7. 32. DeWall CN, MacDonald G, Webster GD, Masten CL, Baumeister RF, Powell C, et al. Acetaminophen reduces social pain: Behavioral and neural evidence. Psychol Sci. 2010;21:931–7. 10. Witte W, Stein C. History, definitions, and contemporary viewpoints. In: Kopf A, Patel NB, editors. Guide to pain management in low-resource settings. Seattle: International Association for the Study of Pain; 2010: p. 3–7 11. Monin J, Schulz R. Interpersonal effects of suffering in older adult caregiving relationships. Psychol Aging. 2009;24(3):681–95. 10. Witte W, Stein C. History, definitions, and contemporary viewpoints. In: Kopf A, Patel NB, editors. Guide to pain management in low-resource settings. Seattle: International Association for the Study of Pain; 2010: p. 3–7 10. Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit • Convenient online submission • Thorough peer review 14. Pearlin LI, Mullan JT, Semple SJ, Skaff MM. Care giving and the stress process: an overview of concepts and their measures. Gerontol. 1990;30:583–94. 14. Pearlin LI, Mullan JT, Semple SJ, Skaff MM. Care giving and the stress process: an overview of concepts and their measures. Gerontol. 1990;30:583–94. 15. Bernstein MJ, Claypool HM. Social exclusion and pain sensitivity: why exclusion sometimes hurts and sometimes numbs. Per Soc Psychol Bull. 2012;38:185–96. 15. Bernstein MJ, Claypool HM. Social exclusion and pain sensitivity: why exclusion sometimes hurts and sometimes numbs. Per Soc Psychol Bull. 2012;38:185–96. 16. Chester DS, Pond Jr RS, Richman SB, DeWall N. The optimal calibration hypothesis: how life history modulates the brain’s social pain network. Front Evol Neurosci. 2012;4:ArtID 10. 16. Chester DS, Pond Jr RS, Richman SB, DeWall N. The optimal calibration hypothesis: how life history modulates the brain’s social pain network. Front Evol Neurosci. 2012;4:ArtID 10. Submit your manuscript at www.biomedcentral.com/submit
https://openalex.org/W4296869471	http://www.arjish.com/index.php/arjish/article/download/457/304	Urdu	null	The dictionary and commentary on the “al-Jahliyyah” in classical Islamic literature	al-Iʿjāz taḥqīqī majallah barāʾe islāmiyyāt va insāniyyāt	2,021	cc-by	10,392	Abstract T There are number of Quranic terms elaborated in various perspectives in Islamic literature of classis era. The term “Al-Jahiliyah” is one of those terms. This is usually related to pre Islamic era of Arabs when idol worshiping was performed on vast level in almost all societies of Arab peninsula. Some classic experts of Arabic language have taken this term into account with respect to the particular acts and customs which were prevailing before Islam in different societies of the world. Same is done by some of Quranic commentators as well. Besides that, there are some language experts and Quranic commentators who have elaborated this term while relating it to the specific time periods. It indicates that the term has come across a prominent evolution of ideological scenario and it is still under consideration of Muslim scholars. [ 80 ] [ 80 ] Dr. Hafiz Muhammad Saleem Saeedi Assistant Professor, Islamic Studies, Mohi-ud-din Islamic University, Nerian Sharif, AJK Dr. Hafiz Muhammad Saleem Saeedi Assistant Professor, Islamic Studies, Mohi-ud-din Islamic University, Nerian Sharif, AJK تغل اور ریسفت ےس قلعتم الکیکیس االسیم ادب ںیم "ااجلہ لی ۃ " ےک دصماق یک نیعت تعلقےفسیرغتیلایکیلاعینیصدےجایں The dictionary and commentary on the “al-Jahliyyah” in classical Islamic literature Received on: 08-10-2021 Accepted on: 10-11-2021 Asad Mehmood Abbasi PhD Scholar, Islamic Studies, Mohi-ud-din Islamic University, Nerian Sharif, A hD Scholar, Islamic Studies, Mohi-ud-din Islamic University, Nerian Sharif, AJK Abstract T اس ےک دورسے قتشم "ااجلہ لی ۃ " ےس رماد لبق از االسم اک زامہن ےہ3۔ رادیہی ےن اس اک وہفمم ایبن رکےن ےک دوران زامےن اک نیعت رکےت وہےئ وتارخی اک ذرک ںیہن ایک ےہ اہتبل ہی واحض ایک ےہ ہک دو ربمغیپ ق وں ےک امنیب زامہن اجتیلہ اک زامہن ےہ4 ۔ہی وہ دہع وہی ےہ سج ںیم ولگ ربمغیپوں یک امیلعتت ےس ارحناف رک ےک رفک اور رشک یک رطف ولٹ اجےت :رآن دیجم ںیم ایس ولہپ یک رطف اش رہ ایک ایگ ےہ ںیہ۔ق فقال املالٔ الذین كفروا من قومه ما نراك إال بشرا مثلنا وما نراك اتبعك إال ال ذین هم أراذلنا بدي الرأي وما نرى لكم علینا من فضلٍ بل نظنكم كاذبني ق۔ ال اي قوم أرأیتم إن كنت على بینةٍ من ريب وآاتين رمحة من عنده فعمیت علیكم أنلزمكموها وأنتم هلا كارهون ۔ و اي قوم ال أسألكم علیه ماال إن أجري إال على هللا وما أان بطارد الذین آمنوا إهنم مالقو رهبم ولكين أراكم قوما جتهلون5۔ 1. ہملک " لہج " الص ںیم ہملک "ملع" یک دض ےہ۔ 2. ایس ےس دورسا ظفل "اہجل " قتشم ےہ سج اک وہفمم "ملع ےک ریغب وکیئ اکم رکی " ےہ ۔ 3. Abstract T Keywords: Islam, Quran, Jahiliyah, Arabic رآن ق دیجم یک ضعب االطصاحت ریثک الج ہت اعمین و افممیہ یک احلم ںیہ اور رمور ای م ےک اسھت ان یک ریبعتات و ی والیت ںیم ونتع اسےنم آی رامی ۔آپﷺ ےن اس یک میلعت و دترسی رآن دیجم یک رشتحی و نییبت یک ذہم داری دے رک وعبمث ق ےہ۔یبن ارکمﷺ وک اہلل اعتیلٰ ےن ق ےک دوران وکیئ ولہپ رامیئ یھت ویکہکن آپﷺ وک ولعمم اھت ہک ان رآین االطصاحت یک نییعت ںیہن ق ہنشت ںیہن وھچڑا اھت اس ےک ی ووجد ضعب ق اافلظ ےک افممیہ دبےتل وہےئ احالت ےک اسھت یئن وصرںیت اایتخر رکےت رںیہ ےگ۔اس یک ای اثمل رضحت رمع افروق ریض اہلل ہنع اک وقل :ےہہک Keywords: Islam, Quran, Jahiliyah, Arabic رآن ق دیجم یک ضعب االطصاحت ریثک الج ہت اعمین و افممیہ یک احلم ںیہ اور رمور ای م ےک اسھت ان یک ریبعتات و ی والیت ںیم ونتع اسےنم آی رامی ۔آپﷺ ےن اس یک میلعت و دترسی رآن دیجم یک رشتحی و نییبت یک ذہم داری دے رک وعبمث ق ےہ۔یبن ارکمﷺ وک اہلل اعتیلٰ ےن ق ےک دوران وکیئ ولہپ رامیئ یھت ویکہکن آپﷺ وک ولعمم اھت ہک ان رآین االطصاحت یک نییعت ںیہن ق ہنشت ںیہن وھچڑا اھت اس ےک ی ووجد ضعب ق اافلظ ےک افممیہ دبےتل وہےئ احالت ےک اسھت یئن وصرںیت اایتخر رکےت رںیہ ےگ۔اس یک ای اثمل رضحت رمع افروق ریض اہلل ہنع اک وقل کہے Keywords: Islam, Quran, Jahiliyah, Arabic نےانویںلاعبیریھاےر Keywords: Islam, Quran, Jahiliyah, Arabic نےانویںلاعبیریھاےر "ثالث الن یکون ال نبیﷺ بین نه لنا احب الی من الدنیا وما اه فی ؛ الکالل ة ،واخلالف ة وابواب الرب1" رام دےتی وت ہی (واضح ) ےھجم داین و ام اہیف ےس زی دہ وبحمب یھت۔الکہل، الخہف اور ری ۔ ارگ روسل اہللﷺ نیت زیچوں یک واضح ق رآین االطصاحت ںیم ےس ای االطصح "ااجلہ لی ۃ " ےہ۔ہی ای ان ق رعموف ہملک ےہ سج ےس قلعتم آی ر و روای ت ںیم وکیئ وصخمص رشتحی ی رامی ےہ۔رضحت اوب ذر ریض اہلل ہنع وک ریبعت نیعتم ںیہن یک یئگ ےہ۔ آپﷺ ےن اس وک ذمومم اعدات ےک نمض ںیم یہ اامعتسل ق : رامی آپﷺ ےن ڈاےتٹن وہےئ ق [ 81 ] Al – Aijaz, Research Journal of Islamic Studies & Humanities Vol. V, No. Abstract T 4 (Oct – Dec 2021) " انک امرؤ فیک اجلا لی ه ة " 2 وت ااسی آدیم ےہ سج ںیم اجتیلہ ےہ۔ اس اقمہل ںیم ہی ولعمم رکےن یک وکشش یک یئگ ےہ ہک الکیکیس دہع یک امہ افتریس اور اغلت ںیم ہملک"ااجلہ لی ۃ " وک وطبر االطصح سک وہفم م ںیم شیپ ایک ایگ ےہ۔اس نمض ںیم اچر اغلت اور اچر افتریس اک ااختنب رک ےک ان ںیم ذموکر افممیہ وک ایبن رک ےک زجتہی ایک ایگ ےہ۔ الکیکیس رعیب اغلت ںیم "ااجلہ لی ۃ " یک ونعمی نییعت :رادیہی (م لیلخ نب ادمح ق175 رادیہی ےن اس ےکفلتخم ولہپ ایبن رکےت وہےئ ھ) یک "اتکب انیعل " رعیب یک یلہپ ی اضہطب تغل ےہ۔ق :واحض ایک ےہ ہک 1. ہملک " لہج " الص ںیم ہملک "ملع" یک دض ےہ۔ 2. ایس ےس دورسا ظفل "اہجل " قتشم ےہ سج اک وہفمم "ملع ےک ریغب وکیئ اکم رکی " ےہ ۔ 3. Abstract T اس ےک دورسے قتشم "ااجلہ لی ۃ " ےس رماد لبق از االسم اک زامہن ےہ3۔ رادیہی ےن اس اک وہفمم ایبن رکےن ےک دوران زامےن اک نیعت رکےت وہےئ وتارخی اک ذرک ںیہن ایک ےہ اہتبل ہی واحض ایک ےہ ہک دو ربمغیپ ق وں ےک امنیب زامہن اجتیلہ اک زامہن ےہ4 ۔ہی وہ دہع وہی ےہ سج ںیم ولگ ربمغیپوں یک امیلعتت ےس ارحناف رک ےک رفک اور رشک یک رطف ولٹ اجےت :رآن دیجم ںیم ایس ولہپ یک رطف اش رہ ایک ایگ ےہ ںیہ۔ق فقال املالٔ الذین كفروا من قومه ما نراك إال بشرا مثلنا وما نراك اتبعك إال ال ذین هم أراذلنا بدي الرأي وما نرى لكم علینا من فضلٍ بل نظنكم كاذبني ق۔ ال اي قوم أرأیتم إن كنت على بینةٍ من ريب وآاتين رمحة من عنده فعمیت علیكم أنلزمكموها وأنتم هلا كارهون ۔ و اي قوم ال أسألكم علیه ماال إن أجري إال على هللا وما أان بطارد الذین آمنوا إهنم مالقو رهبم ولكين أراكم قوما جتهلون5۔ " انک امرؤ فیک اجلا لی ه ة " 2 وت ااسی آدیم ےہ سج ںیم اجتیلہ ےہ۔ اس اقمہل ںیم ہی ولعمم رکےن یک وکشش یک یئگ ےہ ہک الکیکیس دہع یک امہ افتریس اور اغلت ںیم ہملک"ااجلہ لی ۃ " وک وطبر االطصح سک وہفم م ںیم شیپ ایک ایگ ےہ۔اس نمض ںیم اچر اغلت اور اچر افتریس اک ااختنب رک ےک ان ںیم ذموکر افممیہ وک ایبن رک ےک زجتہی ایک ایگ ےہ۔ الکیکیس رعیب اغلت ںیم "ااجلہ لی ۃ " یک ونعمی نییعت :رادیہی (م لیلخ نب ادمح ق175 رادیہی ےن اس ےکفلتخم ولہپ ایبن رکےت وہےئ ھ) یک "اتکب انیعل " رعیب یک یلہپ ی اضہطب تغل ےہ۔ق :واحض ایک ےہ ہک ہمللہلںیہململیدے :واحض ایک ےہ ہک 1. ہملک " لہج " الص ںیم ہملک "ملع" یک دض ےہ۔ 2. ایس ےس دورسا ظفل "اہجل " قتشم ےہ سج اک وہفمم "ملع ےک ریغب وکیئ اکم رکی " ےہ ۔ 3. Abstract T اس ےک دورسے قتشم "ااجلہ لی ۃ " ےس رماد لبق از االسم اک زامہن ےہ3۔ رادیہی ےن اس اک وہفمم ایبن رکےن ےک دوران زامےن اک نیعت رکےت وہےئ وتارخی اک ذرک ںیہن ایک ےہ اہتبل ہی واحض ایک ےہ ہک دو ربمغیپ ق وں ےک امنیب زامہن اجتیلہ اک زامہن ےہ4 ۔ہی وہ دہع وہی ےہ سج ںیم ولگ ربمغیپوں یک امیلعتت ےس ارحناف رک ےک رفک اور رشک یک رطف ولٹ اجےت :آندیجمںیمایسولہپیکرطفاش رہایکایگےہ ںیہ ق :ںیہ۔قرآن دیجم ںیم ایس ولہپ یک رطف اش رہ ایک ایگ ےہ فقال املالٔ الذین كفروا من قومه ما نراك إال بشرا مثلنا وما نراك اتبعك إال ال ذین هم أراذلنا بدي الرأي وما نرى لكم علینا من فضلٍ بل نظنكم كاذبني ق۔ ال اي قوم أرأیتم إن كنت على بینةٍ من ريب وآاتين رمحة من عنده فعمیت علیكم أنلزمكموها وأنتم هلا كارهون ۔ و اي قوم ال أسألكم علیه ماال إن أجري إال على هللا وما أان بطارد الذین آمنوا إهنم مالقو رهبم ولكين أراكم قوما جتهلون5۔ اور اس یک وقم ںیم ےس ان رسداروں ےن ہک وھنجں نے کفر کیا ،اھت مہ ےھجت ںیہن دےتھکی رگم اےنپ جیس ای شب اور مہ ےھجت ںیہن دےتھکی ہک ان ولوگں ےک وسا یسک ےن ریتی ریپوی یک وہ وج ہ رے س ےس رذلی ،ںیہ یحطس راےئ ےک اسھت اور مہ مھا رے ت ےیل اےنپ آپ رپ وکیئ ری ب رب نہیں ،دےتھکی ہکلب مہ ںیھمت وھجےٹ امگن رکےت اس۔ںیہ ےن ہک اے ریمی وقم ! ایک ت ےن داھکی ہک ارگ ںیم اےنپ رب یک رطف ےس ای واحض دلیل رپ (اقمئ )وہں اور اس ےن ےھجم اےنپ ی س ےس ب ری رتمح طع رامیئ ق ،وہ رھپ وہ ت رپ فخم ریھک یئگ ،وہ وت ایک مہ اےس ت رپ زب ردیتس اکپچ دںی ،گے ح ہک ت اےس ی دنسپ رکےن واےل اور ۔وہ اے ریمی وقم ! Abstract T ںیم ت ےس اس رپ یسک امل اک وسال ںیہن ، رکی ریمی م ردوری اہلل ےک وسا یسک رپ ںیہن اور ںیم ان ولوگں وک دور اٹہےن واال ںیہن وج اامین الےئ ،ںیہ انیقی وہ اےنپ رب ےس ےنلم واےل ںیہ اور کیل ںیم ںیھمت اےسی ولگ داتھکی وہں وج اہجل ب رتت ۔وہ [ 82 ] ولو أننا نزلنا إلیهم املالئكة وكلمهم املوتى وحشران علیهم كل شيءٍ قبال ما كانوا لیؤمنوا إال أن یشاء هللا ولكن أكثرهم جيهلو ن6 اور ارگ مہ ان یک رطف رےتش ق ای ر دےتی اور ان ےس رمدے وگتفگ رکےت اور مہ ہ ر زیچ ان ےک ی س اسےنم ال مج رکےت وت یھب وہ اےسی ہن ےھت ہک اامین ےل آےت رگم ہی ہک اہلل اچےہ اور کیل ان ےک ارثک اہجل ب رتت ہیں ۔ اس اک بلطم ہی ولعمم وہی ےہ ہک اجہ لی ۃ ای ںیہن ہکلب آدم ہیلع االسلم ےس ےل رک اب ی دعتمد زامونں اور دعتمد وصروتں ںیم اظہ ر وہ یئ "ےہ۔ان ںیم ہ ر ای زامےن وک" رتفۃ" ہک اجی ےہ۔اتکب انیعل ںیم یھب یہی ہملک زمان الفت ة قبل االسالم" لمعتسم ےہ۔ہملک "رتفۃ" وک :انب ہبیتق(م276ھ) ےن دو ان ء رضحت عی سیٰ ہیلع االسلم اور رضحت دمحمﷺےک درایمین زام ےن ےس ریبعت ایک ےہ7۔ انچہچن ہک اج اتکس ےہ ہک رسیتی دصی رجہی ی "رتفۃ" ےس رماد آپﷺ ےس لبق ےک ھچ وس ب رس رماد ےیل اجےت ےھت اہتبل ذموکرہ دوونں وحاولں ںیم ہی وہفمم ومعتیم ےک اسھت رظن آی ےہ۔اس ںیم ضحم رعب اک رغجاایفیئ ہطخ وصخمص ںیہن ےہ۔ ریق رکیت ےہ اور احالت و وااعقت ےک س اس ےک اافلظ ریثک الج ہت رمور ای م ےک اسھت اسھت زی ن اراقتءےک رمالح ےس زگرےن ےک دعب ب افممیہ اک ااحہط رکےت ںیہ۔رعیب اغلت ںیم اس یک واحض ریظن دیھکی اج یتکس ےہ ۔انچہچن دمحم نب ادمح االزہ ری :(م370 ھ) یک تغل "ذہتی اۃغلل" ںیم "لھج" یک وتحیض "اتکب انیعل" ےک اقمےلب ںیم زی دہ لیصفت ےک اسھت دی یئگ ےہ۔اوھنں ےن لہج وک ضحم ولعمامت ی ملع یک یمک ےک اسھت رموبط ںیہن ایک ےہ ہکلب اس ےس ای دقم آےگ ب رھ رک ہی ومفق یھب شیپ ایک ےہ ہک لھج یلمع زدنیگ ںیم یسک اکم ےس قلعتم رجتےب اور اہمرت ےک دقفا ن وک یھب طیحم ےہ8۔ ررہ وال االسم" ہی ای زامہن ےہ سج ںیم الفت اس ظفل یک وغلی واضح رکےن ےک دعب اوھنں ےن "ااجلہ لی ۃ " یک رعتفی ںیم اھکل ےہ ہک "وق االسم ںیہن ےہ9 رار دی ےہ۔ "۔اوھنں ےن زنیم رپ آامسین دہای یک ریغ وموجدیگ ےک زامےن وک اجتیلہ اک زامہن ق ادتبایئ اغل ت ںیم "ااجلہ لی ۃ " وک االسم ےس لبق ای اقب نییعت زامےن ےک اسھت وجڑا ایگےہ۔سج رطح ج دی د اغلت ںیم لبق از االسم ےک رار دی ایگ ےہ اس رطح اک وہفمم دقمی اغلت ںیم ںیہن اتلم ےہ۔دقمی اغلت ےک اطمقب وہ دور سج ںیم ملع و رعب اعمرشے وک اجیلہ اعمرشہ ق رجتہب اور ااہلیم ر امنہیئ اک دقفان راہ ےہ،وہ "ااجلہ لی ۃ " ےک زرمے ںیم امشر وہی ےہ10۔ :یٹھچ دصی رجہی ںیم اامم زرشحمی (م537) ےن ہملک "ااجلہ لی ۃ " وک ااسس االبلع ۃ ںیم رمبک ےک وطر رپ "ااجلہ لی ۃ ااقلدم " یک وصرت ںیم اامعتسل ایک ےہ۔ "ااقلدہم "اک ینعم رپای زامہن ی زگرا وہا زامہن ےہ ۔اوھنں ےن اس وک یسک دہع ےک اسھت اخص ںیہن ایک ےہ11۔ :دعب ںیم ایس دصی ںیم الجمت رری (م357 ھ) ےن سمش اولعلم ںیم ااجلہ لی ۃ وک ریغب یسک زامین وحاےل ےک ایبن ایک ےہ۔"حمی ة ااجلهلی ة12 " وک اجیلہ انعد، ضغب ،ہنیک اور دینمش ےک وہفمم ںیم شیپ ایک ےہ۔اس ےک دعب اوھنں ےن رامن شیپ ایک ےہ ہک" وج صخش یبن ارکمﷺ اک دنمرہج ذلی ق جح یک ااطتسع رےنھک ےک ی ووجد ریغب جح ےیک رم ایگ،وہ اجتیلہ رپ رما13 "۔ رق ےہ۔الثم زرشحمی اور ریمحی ےن اس اضتد یٹھچ دصی رجہی یک اغلت اور اس لبق یک اغلت ںیم شیپ یک یئگ رعتافیت ںیم ااہتنیئ اقب وغر ق ولو أننا نزلنا إلیهم املالئكة وكلمهم املوتى وحشران علیهم كل شيءٍ قبال ما كانوا لیؤمنوا إال أن یشاء هللا ولكن أكثرهم جيهلو ن6 اور ارگ مہ ان یک رطف رےتش ق ای ر دےتی اور ان ےس رمدے وگتفگ رکےت اور مہ ہ ر زیچ ان ےک ی س اسےنم ال مج رکےت وت یھب وہ اےسی ہن ےھت ہک اامین ےل آےت رگم ہی ہک اہلل اچےہ اور کیل ان ےک ارثک اہجل ب رتت ہیں ۔ اس اک بلطم ہی ولعمم وہی ےہ ہک اجہ لی ۃ ای ںیہن ہکلب آدم ہیلع االسلم ےس ےل رک اب ی دعتمد زامونں اور دعتمد وصروتں ںیم اظہ ر وہ یئ "ےہ۔ان ںیم ہ ر ای زامےن وک" رتفۃ" ہک اجی ےہ۔اتکب انیعل ںیم یھب یہی ہملک زمان الفت ة قبل االسالم" لمعتسم ےہ۔ہملک "رتفۃ" وک :انب ہبیتق(م276ھ) ےن دو ان ء رضحت عی سیٰ ہیلع االسلم اور رضحت دمحمﷺےک درایمین زام ےن ےس ریبعت ایک ےہ7۔ انچہچن ہک اج اتکس ےہ ہک رسیتی دصی رجہی ی "رتفۃ" ےس رماد آپﷺ ےس لبق ےک ھچ وس ب رس رماد ےیل اجےت ےھت اہتبل ذموکرہ دوونں وحاولں ںیم ہی وہفمم ومعتیم ےک اسھت رظن آی ےہ۔اس ںیم ضحم رعب اک رغجاایفیئ ہطخ وصخمص ںیہن ےہ۔ ریق رکیت ےہ اور احالت و وااعقت ےک س اس ےک اافلظ ریثک الج ہت رمور ای م ےک اسھت اسھت زی ن اراقتءےک رمالح ےس زگرےن ےک دعب ب افممیہ اک ااحہط رکےت ںیہ۔رعیب اغلت ںیم اس یک واحض ریظن دیھکی اج یتکس ےہ ۔انچہچن دمحم نب ادمح االزہ ری :(م370 ھ) یک تغل "ذہتی اۃغلل" ںیم "لھج" یک وتحیض "اتکب انیعل" ےک اقمےلب ںیم زی دہ لیصفت ےک اسھت دی یئگ ےہ۔اوھنں ےن لہج وک ضحم ولعمامت ی ملع یک یمک ےک اسھت رموبط ںیہن ایک ےہ ہکلب اس ےس ای دقم آےگ ب رھ رک ہی ومفق یھب شیپ ایک ےہ ہک لھج یلمع زدنیگ ںیم یسک اکم ےس قلعتم رجتےب اور اہمرت ےک دقفا ن وک یھب طیحم ےہ8۔ ظفیوغارےےدعوھےاجیرعفںیاھکےہاالہاہےسںی [ 83 ] Al – Aijaz, Research Journal of Islamic Studies & Humanities Vol. Abstract T V, No. 4 (Oct – Dec 2021) وک اظہ ر ربیت زامین ےک اسھت وسنمب ںیہن ایک ےہ وج لبق از االسم اور دعب از االسم ےک احالت ںیم ےھت کیل اس رعتفی وک یسک وصخمص زامین ب ایک ےہ۔زرشحمی ےن رصف امیض ےک زامےن اک ذرک ایک ےہ کیل یسک وصخمص دور یک نییعت ںیہن یک ےہ۔ریمحی یہ وہ ےلہپ تغل دان ںیہ وہنجں ےن ااجلہ لی ۃ وک یسک دقماری ی یتیمک نمض ںیم ایبن رکےن ےک اجبےئ ایعمری وصرت ںیم االسم دینمش ےک ےیل وطبر لیثمت ایبن ایک ےہ۔اس نمض ںیم ان یک شیپ رکدہ دحی اقب وغر ےہ ۔ اقمہل اگنر وک التش ب ر ےک ی ووجد ان اافلظ ےک اسھت دحی ںیہن لم یکس ےہ۔اس ےک اجبےئ اسہقب ذریخہ اہ : رامی ےئ ااحدی ںیم ای اور دحی یتلم ےہ سج ےک اطمقب یبن ارکمﷺ ےن ق . Abstract T من مل مينعه عن احلج حاجة ظاهرة أو سلطان جائر أو مرض حابس فمات ومل حيج فلیمت إن شاء یهوداي شاء نصرانیا سج صخش وک جح رکےن ےس اظہ ری رضورت ی اظمل ،احمک ی روک دےنی وایل امیبری ہن روےک اور وہ ریغب جح ےئک وہےئ رم ،اجےئ وت اچےہ وت وہ وہیدی یک ومت رمے اور اچےہ رصناین یک ومت رمے14 ۔ ر شیپ رکےن یک وکشش یک ےہ ہک ان ےک دہع ی ااجلہ لی ۃ اک وہفمم ریمحی ےن وہیدی ی اسیعیئ وک اجی لی ت ےس دبتلی رکےت وہےئ اغابل ہی ی ب ااہتنیئ وتعس اایتخر رک ایگ اھت ا ور ان امتم ولوگں ش لم اھت وج رے ےس ی ہ ر وصتر ےیک اجےت ںیہ۔نکمم ےہ ہک ذموکرہ نیت اغلت االسم ےک داب رق وسحمس وہ راہ وہ اسوتںی دصی رجہی ںیم می صۃ وہشد رپ آےن وایل رعموف رعیب تغل "اسلن ںیم ااجلہ لی ۃ یک رعتافیت ےک امنیب ومعیم اس ق ر رکد ارعلب" اس اراقت ںیم ای ومب ار ادا رکیت وہیئ رظن آیت ےہ۔ :انب وظنمر(م711 ھ) ےن ےلہپ االزہ ری یک ایبن رکدہ رعتفی اک ااعدہ ایک ےہ ینعی "ریغ االسیم دہع"۔اس ےک دعب اوجلہ ری ےک وحاےل ےک اسھت ذہتی اۃغلل وک یھب دم رظن رےتھک وہےئ اوھنں ےن اس وک لبق از االسم ےک زامےن ےس وجڑا ےہ"ررۃ وال االسم "ااجلهلی ۃ زنم الفت اس ےک دعب ےتھکل ںیہ ہک"هي احلال التی کانت ه علی ا العرب قبل االسالم من له اجل بلل سبحان ه ورسول ه وشرائع الدین املفاخر ة بالنساب والکرب والتجرب وغی ذلک15 "۔ رہمج:ہی اہلل احبسہن واعتیلٰ،اس ےک روسل اور اس یک رشوتعیں ےس ال ب ملع وہےن یک ہی وہ احل ےہ سج رپ االسم ےس لبق رعب ےک ولگ ےھت۔وہ بسن رپ رخف اور ربکت رکےت ےھت۔اس ےک العوہ وہ ملظ و وجر ےک راےتس رپ رواں ےھت۔ سج رطح االزہ ری ےن اسہقب زامےن وک ااجلہ لی ۃ یک وتحیض ںیم شیپ ایک ےہ ایس رطح انب وظنمر ےن یھب لبق از االسم ےک دہع وک اجتیلہ رار اک زامہن ق دی ےہ کیل اس اجتیلہ اک وکیئ ہطقن آاغز ایبن ںیہن ایک ےہ۔اہتبل انب وظنمر ےن ااجلہ لی ۃ وک ھچک رغجاایفیئ،اعمرشیت اور االخیق ولہپ ؤں ےک اسھت وجڑ رک رعیب تغل ونیسی ںیم یلہپ رمہبت ای ااضیف واضح یک داغ لیب ڈایل ےہ۔ ان یک ایبن رکدہ رعتفی زامن و اکمن ےک اسھت اسھت رعوبں یک اعدات یک اکعیس یھب رکیت ےہ۔اےنپ اعمرصنی یک رطح اوھنں ےن یھب ااجلہ لی ۃ وک ای اےسی زامےن ےک احالت ےک اسھت وجڑا ےہ نج ںیم رےنہ واےل ولوگں یک اعدات ومعیم وطر رپ ی دنسپی دہ ںیہ۔ انب وظنمر یک ایبن رکدہ رعتفی ےس ہی وت ولعمم ںیہن وہی ہک اجتیلہ اک زامہن بک ےس بک ی اک ےہ اہتبل ہی ولعمم وہ اجی ےہ ہک لبق از االسم تغل اور ریسفت ےس قلعتم الکیکیس االسیم ادب ںیم "ااجلہ لی ۃ " ےک دصماق یک نیعت [ 84 ] رعب سک رطح زدنیگ زگار رےہ ےھت۔انچہچن ہی یلہپ تغل ےہ سج ںیم ای ایسی اجعم رعتفی شیپ یک یئگ ےہ سج ںیم اسہقب رعتافیت وک ہن رصف مج رک دی ایگ ےہ ہکلب دعب ںیم شیپ یک اجےن وایل اغلت رامہ رک دی یئگ ےہ۔ ںیم ایبن رکدہ رعتافیت وک ای اینبد ق رآن و دحی اور اجیلہ ادب ےک ز رشتسمنیق اک ایخل ےہ ہک ہک رعیب تغل ےک الکیکیس دہع ںیم اافلظ ےک اعمین ویہ ایبن ےیک اجےت ےھت وج ق ی دہ رام اھت ہک رعیب ا ری ںیہ اور اغابل اس جہنم یک اینبد ںیم ہی رظنہی اکر ق ق افلظ اےنپ اعمین دبتلی ںیہن رکےت ںیہ16۔ ہک اغلت اک قیمع اطمہعل رکےن ےس ولعمم وہی ےہ ہک امہ رنی تغل ےن اینپ داتسن ںیم ااجلہیلہ ےک وہ افممیہ شیپ ےیک ںیہ وج ان ےک اطم ح بق زی دہ حیحص ںیہ۔اافلظ ےک اعمین و افممیہ ںیم اراقتء اور وتعس اک دیپا وہی ای رطفی ارم ےہ اور یہی رصف ااہمت اغلت ںیم یہ ںیہن ہکلب اغلت یک ان اتکوبں ںیم یھب رظن آی ےہ نج یک تیثیح ی ونی ےہ۔ رآین افتریس ںیم ااجلہ لی ۃ یک ونعمی نییعت الکیکیس ق االسیم ی رخی ےک الکیکیس دہع ےک رسفمنی ےن "ااجلہ لی ۃ " یک وتحیض و رشتحی ےک ےیل قیقحت و صیحمت ےس اکم ایل ےہ۔ان یک ی والیت ںیم شیپ رآین افتریس رکدہ ولعمامت اک ان یک اعمرص اغلت ںیم دےی ےئگ وصترات ےک اسھت زجتی یت اقتب رکی نکمم ےہ۔ریثک دعتاد ںیم یھکل یئگ ق ںیم : نب امیلسن (م ےس اقمہل ذہا ںیم اچر افتریس اک ااختنب ایک ایگ ےہ۔اونیل افتریس ںیم ےس اقمب150ھ) یک ریسفت اک ااختنب ایک ایگ ےہ وج :الیلخل(م175 ) یک اتکب انیعل ںیم دےی ےئگ وصتر ےک اسھت اقتب ےک ےیل اامعتسل یک اج یتکس ےہ۔وچیھت دصی رجہی ںیم اامم ربطی :(م310 رنی ےہ، بختنم یک یئگ ےہ۔وچیھت دصی رجہی ںیم ھ) یک ریسفت اجعم اایبلن وج اغابل الکیکیس افتریس ںیم رعموف ب (ریمحی :م357 :ھ) اوریٹھچ دصی رجہی ںیم زرشحمی (م537 رہ ےنیل ےک ےیل زرشحمی یک ریسفت ااشکلف اک ااختنب ) یک اغلت اک اقتیلب اجب :ریبط (م ایک ایگ ےہ۔اسوتںی دصی رجہی ںیم اامم ق671 :ھ) یک ریسفت ااجلعم الاکحم ارقلآن وک انب وظنمر(م711 ھ) یک اسلن ارعلب ےک رے ےک ےیل اقتیلب اجب شیپ رظن راھک ایگ ےہ۔ نب امیلسن یک وتحیض اقمب نبامیلسنےناینپریسفتںیمدواقمامت اقمب17رپ"ااجلہلیۃ "وکیبنارکمﷺےکدوروبنتےسےلہپےکزامےنےکاسھتوسنمبایک رعب سک رطح زدنیگ زگار رےہ ےھت۔انچہچن ہی یلہپ تغل ےہ سج ںیم ای ایسی اجعم رعتفی شیپ یک یئگ ےہ سج ںیم اسہقب رعتافیت وک ہن رصف مج رک دی ایگ ےہ ہکلب دعب ںیم شیپ یک اجےن وایل اغلت رامہ رک دی یئگ ےہ۔ ںیم ایبن رکدہ رعتافیت وک ای اینبد ق رآن و دحی اور اجیلہ ادب ےک ز رشتسمنیق اک ایخل ےہ ہک ہک رعیب تغل ےک الکیکیس دہع ںیم اافلظ ےک اعمین ویہ ایبن ےیک اجےت ےھت وج ق ی دہ رام اھت ہک رعیب ا ری ںیہ اور اغابل اس جہنم یک اینبد ںیم ہی رظنہی اکر ق ق افلظ اےنپ اعمین دبتلی ںیہن رکےت ںیہ16۔ ہک اغلت اک قیمع اطمہعل رکےن ےس ولعمم وہی ےہ ہک امہ رنی تغل ےن اینپ داتسن ںیم ااجلہیلہ ےک وہ افممیہ شیپ ےیک ںیہ وج ان ےک اطم ح بق زی دہ حیحص ںیہ۔اافلظ ےک اعمین و افممیہ ںیم اراقتء اور وتعس اک دیپا وہی ای رطفی ارم ےہ اور یہی رصف ااہمت اغلت ںیم یہ ںیہن ہکلب اغلت یک ان اتکوبں ںیم یھب رظن آی ےہ نج یک تیثیح ی ونی ےہ۔ الیکییافریںیاجیونعنییع عیینعنوی جایںیرفاییکیلا االسیم ی رخی ےک الکیکیس دہع ےک رسفمنی ےن "ااجلہ لی ۃ " یک وتحیض و رشتحی ےک ےیل قیقحت و صیحمت ےس اکم ایل ےہ۔ان یک ی والیت ںیم شیپ رآین افتریس رکدہ ولعمامت اک ان یک اعمرص اغلت ںیم دےی ےئگ وصترات ےک اسھت زجتی یت اقتب رکی نکمم ےہ۔ریثک دعتاد ںیم یھکل یئگ ق ںیم : نب امیلسن (م ےس اقمہل ذہا ںیم اچر افتریس اک ااختنب ایک ایگ ےہ۔اونیل افتریس ںیم ےس اقمب150ھ) یک ریسفت اک ااختنب ایک ایگ ےہ وج :الیلخل(م175 ) یک اتکب انیعل ںیم دےی ےئگ وصتر ےک اسھت اقتب ےک ےیل اامعتسل یک اج یتکس ےہ۔وچیھت دصی رجہی ںیم اامم ربطی :(م310 رنی ےہ، بختنم یک یئگ ےہ۔وچیھت دصی رجہی ںیم ھ) یک ریسفت اجعم اایبلن وج اغابل الکیکیس افتریس ںیم رعموف ب (ریمحی :م357 :ھ) اوریٹھچ دصی رجہی ںیم زرشحمی (م537 رہ ےنیل ےک ےیل زرشحمی یک ریسفت ااشکلف اک ااختنب ) یک اغلت اک اقتیلب اجب :ریبط (م ایک ایگ ےہ۔اسوتںی دصی رجہی ںیم اامم ق671 :ھ) یک ریسفت ااجلعم الاکحم ارقلآن وک انب وظنمر(م711 ھ) یک اسلن ارعلب ےک رے ےک ےیل اقتیلب اجب شیپ رظن راھک ایگ ےہ۔ اقنامیلیوحی رے ےک ےیل اقتیلب اجب اقنامیلیوحیشیپ رظن راھک ایگ ےہ۔ نب امیلسن یک وتحیض اقمب نب امیلسن ےن اینپ ریسفت ںیم دو اقمامت اقمب17 رپ "ااجلہ لی ۃ " وک یبن ارکمﷺ ےک دور وبنت ےس ےلہپ ےک زامےن ےک اسھت وسنمب ایک ےہ18 ےن اس نمض ںیم ااجلہ لی ۃ یک وتحیض شیپ رکےت وہ ۔اینپ اعمرص تغل اتکب انیعل ےک ب ر سکع اقمب ےئ یسک اقب امیپشئ اوروصخمص ر ہی دیپا وہی ےہ ہک ان ےک اطمقب یبنﷺ یک زامےن اک وحاہل ںیہن دی ےہ۔اوھنں ےن اس وک ومعیم وصرت ںیم شیپ ایک ےہ سج ےس ای ی ب یک ریسفت اک قیمع اطمہعل ہی اظہ ر رک تثعب ےس لبق رعب اک اسرا زامہن ااجلہیلہ ےک زرمے ںیم آی ےہ۔کیل اقمب ی ےہ ہک ان ےک اطمقب یبن ریبی زامہن یہ ااجلہ لی ۃ الہکی ےہ ۔ ارکمﷺ یک تثعب اک ق نب امیلسن ےک اطمقب یبنﷺ یک تثعب ےس لبق اور دعب ےک زامےن وک ااجلہیلہ ےک وہفمم ںیم ایبن ایک اج اتکس اتکب انیعل اور ریسفت اقمب ےہ۔اس ںیم رصف تثعب ےس لبق اک زامہن ش لم ںیہن ےہ۔اس اک نب امیلسن ےن "حمی ۃ ااجلہ لی ۃ وبثت ہی ےہ ہک اقمب19 " ےک نمض ںیم ان [ 85 ] Al – Aijaz, Research Journal of Islamic Studies & Humanities Vol. Abstract T V, No. Abstract T 4 (Oct – Dec 2021) رشمنیک ہکم وک شیپ ایک ےہ وہنجں ےن یبن ارکمﷺ یک وبنت و راسل اک ااکنر ایک اور آپﷺ وک ھچ رجہی ںیم رمعہ رکےن یک رغض ےس ہکم ںیم دالخ وہےن ےس روک دی اھت20۔ اوھنں ےن "نظ ااجلہ لی ۃ21 " ےک تحت ان دانمشن االسم اک ذرک ایک ےہ وج رشمک اور اجلہ ےھت۔ان ںیم اوب ایفسن اور اس ےک وہ اسیھت امشر ےیک ںیم ہی امن ایل اور وہشمر رک دی اھت ہک ربمغیپ االسمﷺ دیہش وہ ےکچ ںیہ ےئگ وہنجں ےن نیت ہنس رجہی ںیم ادح یک ج22 ۔ ایس رطح "مکح ااجلہ لی ۃ23 ےن ان اظمل " ےک تحت اقمب رؤاس اک ذرک ایک ےہ وج دمہنی ںیم یبن ارکمﷺ یک رجہت ےس لبق ہلصیف اسزی ےک اتخمر تھے24 ۔ ےن ر ےھت۔ اقمب ےک اطمقب ااجلہ لی ۃ ںیم ان ولوگں یک دعات و ااکفر ش لم ںیہ وج یبن ارکمﷺ یک وبنت ےک نم سپ ولعمم وہی ےہ ہک اقمب لبق از االسم ےک رعب اعمرشے ےک ی رے ںیم رآین آی ت وک وطبر دتسمالت اامعتسل ںیہن ایک ےہ اور ہن یہ اوھنں ےن وگتفگ رکےت وہےئ ق ان ہطقن اہےئرظن وک ایبن ایک ےہ ہک ان ےک اطمقب االسم ےس لبق امتم الہ رعب اجلہ ےھت۔اوھنں ےن ہی یھب ایبن ںیہن ایک ہک االسم ےس لبق ےک رعب ہ ر دہع ںیم اجلہ اور دب االخق ےھت۔اقمب رآن دیجم ےک شیپ رکدہ وصتر اطمقب "ااجلہ لی ۃ " ای وعیس المفہو م االطصح ےہ اور ق راد ےک وحاےل ےس اھجمس اج اتکس ےہ ، اس وک رصف لبق از االسم ےک رعب اعمرشے یک روینش "ااجلہ لی ۃ " وک ریغ االسیم االخایقت ےک احلم اق ںیم انھجمس درس ںیہن ےہ۔ ربطی یک وتحیض " اامم ربطی ےن"ااجلہ لی ۃ " یک ریسفت ںیم ی ریخی وحاولں وک زی دہ ایبن ایک ےہ۔ اجلاہلیۃ االولی25" یک ریسفت ںیم اوھنں ےن اھکل ےہ ہک املع اس رمبک یک یسک ای ریسفت رپ قفتم ںیہن ںیہ۔اوھنں ےن املعء ےک فلتخم اوقال وک یھب ایبن ایک ےہ نج ےک اطمقب "ااجلہ لی ۃ االویلٰ" ےس رما د آدم علیہ االسلم اور ونح ہیلع االسلم، ونح ہیلع االسلم اور ادرسی ہیلع االسلم ی رھپ آدم ہیلع االسلم اور عی سیٰ ہیلع االسلم اک درایمین زام ہن ےہ26 ۔ ردی رضحت عی سیٰ ہیلع االسل اامم ربطی ان امتم اوقال ےک درس وہےن اک اامتحل اظہ ر رکےت وہےئ ان وک وبقل رکےت ںیہ اہتبل ان ےک ب م اور ری ےہ۔اہتبل ہ ر ومفق ےک ی رے ںیم اامم ربطی ےن اینپ ام لبق اغلت ںیم رضحت دمحم ﷺ ےک درایمن اک زامہن اس ےک وہفمم ےک زی دہ ق ایبن رکدہ اجیلہ ادوار وک ان ء یک انمس ےک وحاولں وک وحلمظ راھک ےہ27۔ نب یک رطح ااجلہ لی ۃ ےس قلعتم اقمب ربطی ےن اقمب امیلسن ےک اس ومفق یک یھب ی دیئ یک ےہ سج اک قلعت رشمنیک یک اجی ےس یبن نب امیلسن ےن یبنﷺ اور آپ ےک اخمنیفل ےک امنیب وہےن وایل اشکشک اک ذرک ایک ارکمﷺ یک اخمتفل ےک اسھت ےہ۔سج رطح اقمب ےہ، اامم ربطی ےن یھب اس وک اینپ ریسفت اک ہصح انبی ےہ28 ۔ اہتبل اامم ربطی ےن ےن نظ ااجلہ لی ۃ ےک نمض ںیم رآین رکف اک ہصح انبی ےہ۔الثم: اقمب نب امیلسن ےس زی دہ وعیس امیپےن رپ ق ااجلہ لی ۃ وک اقمب ہک اامم ربطی ےن امتم رشمنیک وک اس ںیم ش لم راھک ےہ رصف اوب ایفسن اور اس ےک اسویھتں اک ذرک ایک ےہ ح29 نب امیلسن ےن ۔اقمب ےن ر ےھت۔ اقمب ےک اطمقب ااجلہ لی ۃ ںیم ان ولوگں یک دعات و ااکفر ش لم ںیہ وج یبن ارکمﷺ یک وبنت ےک نم سپ ولعمم وہی ےہ ہک اقمب لبق از االسم ےک رعب اعمرشے ےک ی رے ںیم رآین آی ت وک وطبر دتسمالت اامعتسل ںیہن ایک ےہ اور ہن یہ اوھنں ےن وگتفگ رکےت وہےئ ق ان ہطقن اہےئرظن وک ایبن ایک ےہ ہک ان ےک اطمقب االسم ےس لبق امتم الہ رعب اجلہ ےھت۔اوھنں ےن ہی یھب ایبن ںیہن ایک ہک االسم ےس لبق ےک رعب ہ ر دہع ںیم اجلہ اور دب االخق ےھت۔اقمب رآن دیجم ےک شیپ رکدہ وصتر اطمقب "ااجلہ لی ۃ " ای وعیس المفہو م االطصح ےہ اور ق راد ےک وحاےل ےس اھجمس اج اتکس ےہ ، اس وک رصف لبق از االسم ےک رعب اعمرشے یک روینش "ااجلہ لی ۃ " وک ریغ االسیم االخایقت ےک احلم اق ںیمانھجمسدرس ںیہنےہ۔ ربیوحیںیم انھجمس درس ںیہن ےہ۔ یحویبر " اامم ربطی ےن"ااجلہ لی ۃ " یک ریسفت ںیم ی ریخی وحاولں وک زی دہ ایبن ایک ےہ۔ اجلاہلیۃ االولی25" یک ریسفت ںیم اوھنں ےن اھکل ےہ ہک املع اس رمبک یک یسک ای ریسفت رپ قفتم ںیہن ںیہ۔اوھنں ےن املعء ےک فلتخم اوقال وک یھب ایبن ایک ےہ نج ےک اطمقب "ااجلہ لی ۃ االویلٰ" ےس رما د آدم علیہ االسلم اور ونح ہیلع االسلم، ونح ہیلع االسلم اور ادرسی ہیلع االسلم ی رھپ آدم ہیلع االسلم اور عی سیٰ ہیلع االسلم اک درایمین زام ہن ےہ26 ۔ ردی رضحت عی سیٰ ہیلع االسل اامم ربطی ان امتم اوقال ےک درس وہےن اک اامتحل اظہ ر رکےت وہےئ ان وک وبقل رکےت ںیہ اہتبل ان ےک ب م اور ری ےہ۔اہتبل ہ ر ومفق ےک ی رے ںیم اامم ربطی ےن اینپ ام لبق اغلت ںیم رضحت دمحم ﷺ ےک درایمن اک زامہن اس ےک وہفمم ےک زی دہ ق ایبن رکدہ اجیلہ ادوار وک ان ء یک انمس ےک وحاولں وک وحلمظ راھک ےہ27۔ نب یک رطح ااجلہ لی ۃ ےس قلعتم اقمب ربطی ےن اقمب امیلسن ےک اس ومفق یک یھب ی دیئ یک ےہ سج اک قلعت رشمنیک یک اجی ےس یبن نب امیلسن ےن یبنﷺ اور آپ ےک اخمنیفل ےک امنیب وہےن وایل اشکشک اک ذرک ایک ارکمﷺ یک اخمتفل ےک اسھت ےہ۔سج رطح اقمب ےہ، اامم ربطی ےن یھب اس وک اینپ ریسفت اک ہصح انبی ےہ28 ۔ اہتبل اامم ربطی ےن ےن نظ ااجلہ لی ۃ ےک نمض ںیم رآین رکف اک ہصح انبی ےہ۔الثم: اقمب نب امیلسن ےس زی دہ وعیس امیپےن رپ ق ااجلہ لی ۃ وک اقمب ہک اامم ربطی ےن امتم رشمنیک وک اس ںیم ش لم راھک ےہ رصف اوب ایفسن اور اس ےک اسویھتں اک ذرک ایک ےہ ح29 نب امیلسن ےن ۔اقمب تغل اور ریسفت ےس قلعتم الکیکیس االسیم ادب ںیم "ااجلہ لی ۃ " ےک دصماق یک نیعت [ 86 ] مکح ااجلہ لی ۃ ےک نمض ہک اامم ربطی ےن اس ںیم وتبں یک وپاج رکےن واےل رشمنیک ںیم وہید دمہنی ےک ریغ افصنمہن دعایتل اظنم اک ذرک ایک ےہ ح رعب وک یھب امشر ایک ےہ30 نب امیلسن ےن حمی ۃ ااجلہ لی ۃ ںیم ان رشمنیک اک ذرک ایک ےہ وہنجں ےن یبن ارکمﷺ اور ۔ایس رطح اقمب آپﷺ ےک ااحصب وک ہکم" ںیم رمعہ یک رغض ےس دالخ وہےن ےس روک دی اھت۔اامم ربطی ےن حمی ۃ ااجلہ لی ۃ وک اخالق الکفر " ےس ریبعت عصی ا ت وک اس ںیم ش لم ایک ےہ رکےت وہےئ امتم ریغ وملسمں ےک ت 31۔ اس ےس ولعمم وہی ےہ ہک اامم ربطی ےک اطمقب ااجلہ لی ۃ وک رصف لبق از االسم الہ رعب یک دح ی ای دہع ی رغجاایفیئ ےطخ ےک ی ویسں رپ قبطنم رکےن یک روای درس ںیہن ےہ۔اس ےک اجبےئ ہ ر دہع ےک امتم ریغ وملسمں ےک طلغ االخایقت وک اس ےک ہطیح ںیم امشر رکی زی دہ درس ےہ۔ ہی دحتی د دعب یک افتریس ںیم رظن آیت ےہ۔اامم ربطی ےک اہں ااجلہ لی ۃ یک شیپ یک یئگ وغلی رعتفی االزہ ر ی یک دی یئگ رعتفی ےک ری ولعمم وہیت ےہ ویکہکن اس ںیم ای دہع اک اش رہ اتلم ےہ اہتبل اہجں اامم ربطی اس ےک وہفمم وک ومعیم رن دے رک شیپ رکےت زی دہ ق ہیں رآین رکفت ںیم وتعس رظن آیت ےہ۔ واہں ان ےک ق زرشحمی اور ریبط ق یک وتحیض سج رطح اغلت یک روینش ںیم ولعمم وہی رآین افتریس و ےہ ااجلہ لی ۃ ےک اعمین و افممیہ ای اراقتء ےس زگر رک ونتع وک اظہ ر رکےت ںیہ ایس رطح ق ریبط یک ریسفت ںیم زام ی والیت ںیم یھب "ااجلہ لی ۃ " ےک وہفمم ےک وحاےل ےس رمور ای م ےک اسھت اراقتء واعق وہی ےہ۔ارگہچ زرشحمی اور ق ین اابتعر ےس رقتابی ڈب رھ دصی اک اف ہلص ےہ کیل اس االطصح یک رشتحی ںیم دوونں ےک امنیب وماتقف ی یئ اجیت ےہ ایس ےیل ان وک ای اسھت ش لم اطمہعل ایک ایگ ےہ۔ ریبط اک اخہص ہی ےہ ہک اوھنں ےن "ااجلہ لی ۃ " وک "م ااجلہ لی ۃ زرشحمی اور ق32 " اور "الہ ااجلہ لی ۃ33 رابیک ےک وطر رپ اامعتسل ایک ےہ۔اینپ " یک ب تفا ریس ںیم دوونں رسفمنی ےن ااجلہ لی ۃ وک ای یہ اعمرشے رپ قبطنم ایک ےہ۔ ےک اہں ااجلہ لی ۃ ےس رماد یبن ارکمﷺ اک اخمفل رگوہ اور ربطی ےک اہں اس ےس رماد امتم رشمنیک اس وک ویں اھجمس اج اتکس ےہ ہک اقمب ین اک لبق ریبط اور زرشحمی ےک اطمقب اس ےس رماد ی رخی ان ہک ق ںیہ۔ح از االسم اک ااسی امتم زامہن اور ولگ ںیہ وج ااہلیم دہای ت رپ لمع ریپا رآن دیجم ںیم ہی رصاح ںیہک یھب وموجد ںیہن ےہ ہک ااجلہ لی ۃ ےس رماد یسک اخص دہع ی ےطخ ےک ولوگں یک وہےن ےس رگب راں رےہ ںیہ۔ق ریبط اک اس اک وخبیب ااسحس ےہ ذہتی و اقثتف ی تشیعم و اعمرشت ےہ۔ق اس ےیل وہ نظ ااجلہ لی ۃ34 یک واضح ںیم اےنپ ومفق یک یقطنم واضح شیپ رکےت وہےئ ےتہک ںیہ ہک "نظ ااجلہ لی ۃ ینعی الہ اجتیلہ یک دب امگین،اس ےک اضمف(الہ) وک ج دف رک دی ایگ ےہ35۔ ااجلہ لی ۃ یک ی رخی ےک نیعت ںیم زرشحمی ےن دو واضحی ی ں شیپ یک ںیہ۔ یلہپ واضح ربطی یک ریسفتی روای یک ریپوی رکیت ےہ سج ےک اطمقب ااجلہ لی ۃ دو ربمغیپوں ےک درایمن ای زامےن اک ی م ےہ36 ۔کیل زرشحمی اک ومعیم ہطقن رظن ہی ےہ ہک ااجلہ لی ۃ ای زامےن ےک اسھت رموبط ےہ سج وک "ااقلدہم" اک ی م دی ایگ ےہ۔ہی وہفمم ان یک رمی رکدہ تغل ںیم یھب وموجد ےہ37۔ تغل اور ریسفت ےس قلعتم الکیکیس االسیم ادب ںیم "ااجلہ لی ۃ " ےک دصماق یک نیعت [ 86 ] مکح ااجلہ لی ۃ ےک نمض ہک اامم ربطی ےن اس ںیم وتبں یک وپاج رکےن واےل رشمنیک ںیم وہید دمہنی ےک ریغ افصنمہن دعایتل اظنم اک ذرک ایک ےہ ح رعب وک یھب امشر ایک ےہ30 نب امیلسن ےن حمی ۃ ااجلہ لی ۃ ںیم ان رشمنیک اک ذرک ایک ےہ وہنجں ےن یبن ارکمﷺ اور ۔ایس رطح اقمب آپﷺ ےک ااحصب وک ہکم" ںیم رمعہ یک رغض ےس دالخ وہےن ےس روک دی اھت۔اامم ربطی ےن حمی ۃ ااجلہ لی ۃ وک اخالق الکفر " ےس ریبعت عصی ا ت وک اس ںیم ش لم ایک ےہ رکےت وہےئ امتم ریغ وملسمں ےک ت 31۔ اس ےس ولعمم وہی ےہ ہک اامم ربطی ےک اطمقب ااجلہ لی ۃ وک رصف لبق از االسم الہ رعب یک دح ی ای دہع ی رغجاایفیئ ےطخ ےک ی ویسں رپ قبطنم رکےن یک روای درس ںیہن ےہ۔اس ےک اجبےئ ہ ر دہع ےک امتم ریغ وملسمں ےک طلغ االخایقت وک اس ےک ہطیح ںیم امشر رکی زی دہ درس ےہ۔ ہی دحتی د دعب یک افتریس ںیم رظن آیت ےہ۔اامم ربطی ےک اہں ااجلہ لی ۃ یک شیپ یک یئگ وغلی رعتفی االزہ ر ی یک دی یئگ رعتفی ےک ری ولعمم وہیت ےہ ویکہکن اس ںیم ای دہع اک اش رہ اتلم ےہ اہتبل اہجں اامم ربطی اس ےک وہفمم وک ومعیم رن دے رک شیپ رکےت زی دہ ق ہیں رآین رکفت ںیم وتعس رظن آیت ےہ۔ واہں ان ےک ق زرشحمی اور ریبط ق یک وتحیض سج رطح اغلت یک روینش ںیم ولعمم وہی رآین افتریس و ےہ ااجلہ لی ۃ ےک اعمین و افممیہ ای اراقتء ےس زگر رک ونتع وک اظہ ر رکےت ںیہ ایس رطح ق ریبط یک ریسفت ںیم زام ی والیت ںیم یھب "ااجلہ لی ۃ " ےک وہفمم ےک وحاےل ےس رمور ای م ےک اسھت اراقتء واعق وہی ےہ۔ارگہچ زرشحمی اور ق ین اابتعر ےس رقتابی ڈب رھ دصی اک اف ہلص ےہ کیل اس االطصح یک رشتحی ںیم دوونں ےک امنیب وماتقف ی یئ اجیت ےہ ایس ےیل ان وک ای اسھت ش لم اطمہعل ایک ایگ ےہ۔ ریبط اک اخہص ہی ےہ ہک اوھنں ےن "ااجلہ لی ۃ " وک "م ااجلہ لی ۃ زرشحمی اور ق32 " اور "الہ ااجلہ لی ۃ33 رابیک ےک وطر رپ اامعتسل ایک ےہ۔اینپ " یک ب تفا ریس ںیم دوونں رسفمنی ےن ااجلہ لی ۃ وک ای یہ اعمرشے رپ قبطنم ایک ےہ۔ ےک اہں ااجلہ لی ۃ ےس رماد یبن ارکمﷺ اک اخمفل رگوہ اور ربطی ےک اہں اس ےس رماد امتم رشمنیک اس وک ویں اھجمس اج اتکس ےہ ہک اقمب ین اک لبق ریبط اور زرشحمی ےک اطمقب اس ےس رماد ی رخی ان ہک ق ںیہ۔ح از االسم اک ااسی امتم زامہن اور ولگ ںیہ وج ااہلیم دہای ت رپ لمع ریپا رآن دیجم ںیم ہی رصاح ںیہک یھب وموجد ںیہن ےہ ہک ااجلہ لی ۃ ےس رماد یسک اخص دہع ی ےطخ ےک ولوگں یک وہےن ےس رگب راں رےہ ںیہ۔ق ریبط اک اس اک وخبیب ااسحس ےہ ذہتی و اقثتف ی تشیعم و اعمرشت ےہ۔ق اس ےیل وہ نظ ااجلہ لی ۃ34 یک واضح ںیم اےنپ ومفق یک یقطنم واضح شیپ رکےت وہےئ ےتہک ںیہ ہک "نظ ااجلہ لی ۃ ینعی الہ اجتیلہ یک دب امگین،اس ےک اضمف(الہ) وک ج دف رک دی ایگ ےہ35۔ ااجلہ لی ۃ یک ی رخی ےک نیعت ںیم زرشحمی ےن دو واضحی ی ں شیپ یک ںیہ۔ یلہپ واضح ربطی یک ریسفتی روای یک ریپوی رکیت ےہ سج ےک اطمقب ااجلہ لی ۃ دو ربمغیپوں ےک درایمن ای زامےن اک ی م ےہ36 ۔کیل زرشحمی اک ومعیم ہطقن رظن ہی ےہ ہک ااجلہ لی ۃ ای زامےن ےک اسھت رموبط ےہ سج وک "ااقلدہم" اک ی م دی ایگ ےہ۔ہی وہفمم ان یک رمی رکدہ تغل ںیم یھب وموجد ےہ37۔ [ 86 ] مکح ااجلہ لی ۃ ےک نمض ہک اامم ربطی ےن اس ںیم وتبں یک وپاج رکےن واےل رشمنیک ںیم وہید دمہنی ےک ریغ افصنمہن دعایتل اظنم اک ذرک ایک ےہ ح رعب وک یھب امشر ایک ےہ30 نب امیلسن ےن حمی ۃ ااجلہ لی ۃ ںیم ان رشمنیک اک ذرک ایک ےہ وہنجں ےن یبن ارکمﷺ اور ۔ایس رطح اقمب آپﷺ ےک ااحصب وک ہکم" ںیم رمعہ یک رغض ےس دالخ وہےن ےس روک دی اھت۔اامم ربطی ےن حمی ۃ ااجلہ لی ۃ وک اخالق الکفر " ےس ریبعت عصی ا ت وک اس ںیم ش لم ایک ےہ رکےت وہےئ امتم ریغ وملسمں ےک ت 31۔ اس ےس ولعمم وہی ےہ ہک اامم ربطی ےک اطمقب ااجلہ لی ۃ وک رصف لبق از االسم الہ رعب یک دح ی ای دہع ی رغجاایفیئ ےطخ ےک ی ویسں رپ قبطنم رکےن یک روای درس ںیہن ےہ۔اس ےک اجبےئ ہ ر دہع ےک امتم ریغ وملسمں ےک طلغ االخایقت وک اس ےک ہطیح ںیم امشر رکی زی دہ درس ےہ۔ ہی دحتی د دعب یک افتریس ںیم رظن آیت ےہ۔اامم ربطی ےک اہں ااجلہ لی ۃ یک شیپ یک یئگ وغلی رعتفی االزہ ر ی یک دی یئگ رعتفی ےک ری ولعمم وہیت ےہ ویکہکن اس ںیم ای دہع اک اش رہ اتلم ےہ اہتبل اہجں اامم ربطی اس ےک وہفمم وک ومعیم رن دے رک شیپ رکےت زی دہ ق ہیں رآین رکفت ںیم وتعس رظن آیت ےہ۔ واہں ان ےک ق محشر طبی یک وحیض حشربییحوی سج رطح اغلت یک روینش ںیم ولعمم وہی رآین افتریس و ےہ ااجلہ لی ۃ ےک اعمین و افممیہ ای اراقتء ےس زگر رک ونتع وک اظہ ر رکےت ںیہ ایس رطح ق ریبط یک ریسفت ںیم زام ی والیت ںیم یھب "ااجلہ لی ۃ " ےک وہفمم ےک وحاےل ےس رمور ای م ےک اسھت اراقتء واعق وہی ےہ۔ارگہچ زرشحمی اور ق ین اابتعر ےس رقتابی ڈب رھ دصی اک اف ہلص ےہ کیل اس االطصح یک رشتحی ںیم دوونں ےک امنیب وماتقف ی یئ اجیت ےہ ایس ےیل ان وک ای اسھت ش لم اطمہعل ایک ایگ ےہ۔ [ 87 ] Al – Aijaz, Research Journal of Islamic Studies & Humanities Vol. Abstract T V, No. Abstract T 4 (Oct – Dec 2021) ریبط ےن ریغب یسک زامین ےصح اک وحاہل شیپ رکےت وہےئ زرشحمی ےک وصتر "ااجلہ لی ۃ ااقل ای وس ب رس دعب ایس آی یک ریسفت ےک تحت اامم ق " دم ےس اافتق ایک ےہ اور دوعیٰ ایک ےہ ہک "ااجلہ لی ۃ " اک االطق لبق از االسم ےک زامہن رپ وہی ےہ38۔ان یک دلیل ہی ےہ ہک لبق از االسم ےک رعشا وک رار دی ایگ ےہ اس ےیل ااجلہ لی ۃ اک ومعیم وہفمم لبق از االسم اک اجلہ ہک اجی اھت،ایس رطح ااحدی ںیم یھب ھچک اوطار وک اجتیلہ ق امیض ےہ۔اس ریبط ےن ان امتم اعدات و اوطار وک ایبن ایک ےہ وج رعب اعمرشے ںیم رموج ںیھت اور نج یک انب رپ لیصفت ںیم اامم ق ان وک اجلہ ہک اجی اھت۔ان رآین آی ت شیپ یک ںیہ ی ہک ان اک حبق واحض وہ ےکس۔اقمہل اگنر ےک اشمدہے ےک اطمقب ہی ولہپ ان ےس لبق یسک اعدات ےس قلعتم اوھنں ےن ق رسفم ےن ںیہن وھکال ےہ۔ہی اامم وموصف اک اانپ مہف ےہ۔الثم وہ عصی ۃ ااجلہ لی ۃ اک ذرک رکےن ےک دعب الہ رعب یک ی رپیتس یک الیصفتت ںیم اجےت ریبط ےن حمی ۃ ااجلہ لی ۃ ےک ںیہ اور ان ےک وبعمدوں ینعی الت،انمت اور زعیٰ وریغہ ےک ی رے ںیم الیصفتت شیپ رکےت ںیہ۔اس ےک دعب اامم ق نمض ںیم رشمنیک رعب اک وتدیح ےس ااکنر شیپ رکےت وہےئ اس ےک فلتخم ولہپؤں رپ روینش ڈایل ےہ39 ۔ ریبط ےن اامم ق مکح ااجلہ لی ۃ ےک تحت ہی واحض ایک ےہ ہک رعب ںیم دو وقانین ےتلچ ےھت۔ ای اقونن زمکور ےک ےیل اھت سج یک رو ےس اس وک ہ ر ج رم ور ےک ےیل اھت سج یک رو ےس اس وک زسا ےس اچبےن اک ہ ر رحہب اامعتسل ایک اج ی اھت ہک ای اقونن اطق یک زسا یتلم یھت ح40 ۔ ریبط ےن ہی یھب اامم ق راءت "مکح" یھب ےہ۔اس ےس آی اک وہفمم اجیلہ ےلصیف ےک اجبےئ اجیلہ ہلصیف اسز ےک وطر اھکل ےہ ہک "مکح" یک ای ق ریبط اس رطح ہی ومفق شیپ رکی اچےتہ ںیہ ہک آی ںیم اجیلہ ہلصیف اسزوں ےس رماد وہ اکنہ اور اضقۃ ںیہ وج رع رپ اسےنم آی ےہ۔اامم ق ب ںیم ریغ افصنمہن اور ریغ یقطنم ہلصیف اسزی ےک رمبکت وہےت ےھت41 ونں اور ی روخیں ےک اجبےئ لبق از االسم رعب ۔اس ےس اجتیلہ ان ےک وصخمص ولوگں یک وصخمص اعدوتں،رحوتکں اور روسم و رواج یک آہنیئ دار نب اجیت ےہ۔ اس ےس ولعمم وہی ےہ ہک یٹھچ دصی وسیعی ی ہی ومفق قفتم ہیلع وہ اکچ اھت ہک االسم یک آدم ےس ےلہپ اجحز اور اس ےک رگد و وناح ںیم رےنہ ہک دعب ےک واےل وہ امتم ابقلئ "ااجلہ لی ۃ " ےک ریپواکر ےھت وج دہای ےس رحموم ےھت۔رشوع ےک رسفمنی ےن اس وک ومعیم وہفمم ںیم ح رسفمنی ےن اس وک وصخمص وہفمم ںیم شیپ ایک ےہ۔ اس نمض ںیم ای لکشم ہی شیپ آیت ےہ ہک رآن دیجم ےن ربتج ااجلہ لی ۃ وہفمم اےنپ اسھت یئک دیچیپہ وساالت یھب ےل آی ےہ۔الثم ق42 ےس عنم ریبط ےن اھکل ےہ ہک لبق از االسم الہ رعب اعمیش وطر رپ مکحتسم ںیہن ےھت ہکلب ان رپ ولفمک ااحلیل اک دور دورہ اھت احال ایک ےہ۔اامم ق ہکن اس آی ںیم رعب یک امیل احل اک وکیئ ریبط ہی واضح ںیہن رکےت ہک اس ذرک وموجد ںیہن ےہ اور ہن یہ یسک اسہقب رسفم ےن اس اک ذرک ایک ےہ۔ اامم ق االفس ںیم زدنیگ زگارےن واےل اعم ولوگں یک وخانیت وک زی و زتنی ےک ےیل زویرات اور درگی اابسب سک رطح رسیم آےت ےھت43۔ زرشحمی ےن یھب ااجلہ لی ۃ ےک نمض ںیم الہ رعب یک یفنم رکف،وسچ،اعدات اور روسم و رواج اک ذرک ایک ےہ۔ان ےک اطمقب رعب زامہن ام لبق " االسم الھواء" " اور اجلھل " ںیم زدنیگ زگار رےہ ےھت44 ۔اوھنں ےن مکح ااجلہ لی ۃ45 وک االسم ےس ےلہپ ےلصیف اصدر رکےن واےل اقض ی و ں ےک وحاےل ےس ایبن ایک ےہ۔الثم رجنان اک ای رعموف رمکحا ن "اعف" اھت۔زرشحمی ےن اس یک ہلصیف اسزی یک الصح وک یبن ارکمﷺ ےک اس نمض ںیم ای لکشم ہی شیپ آیت ےہ ہک رآن دیجم ےن ربتج ااجلہ لی ۃ وہفمم اےنپ اسھت یئک دیچیپہ وساالت یھب ےل آی ےہ۔الثم ق42 ےس عنم ریبط ےن اھکل ےہ ہک لبق از االسم الہ رعب اعمیش وطر رپ مکحتسم ںیہن ےھت ہکلب ان رپ ولفمک ااحلیل اک دور دورہ اھت احال ایک ےہ۔اامم ق ہکن اس آی ںیم رعب یک امیل احل اک وکیئ ریبط ہی واضح ںیہن رکےت ہک اس ذرک وموجد ںیہن ےہ اور ہن یہ یسک اسہقب رسفم ےن اس اک ذرک ایک ےہ۔ اامم ق االفس ںیم زدنیگ زگارےن واےل اعم ولوگں یک وخانیت وک زی و زتنی ےک ےیل زویرات اور درگی اابسب سک رطح رسیم آےت ےھت43۔ زرشحمی ےن یھب ااجلہ لی ۃ ےک نمض ںیم الہ رعب یک یفنم رکف،وسچ،اعدات اور روسم و رواج اک ذرک ایک ےہ۔ان ےک اطمقب رعب زامہن ام لبق " االسم الھواء" " اور اجلھل " ںیم زدنیگ زگار رےہ ےھت44 ۔اوھنں ےن مکح ااجلہ لی ۃ45 وک االسم ےس ےلہپ ےلصیف اصدر رکےن واےل اقض ی و ں ےک وحاےل ےس ایبن ایک ےہ۔الثم رجنان اک ای رعموف رمکحا ن "اعف" اھت۔زرشحمی ےن اس یک ہلصیف اسزی یک الصح وک یبن ارکمﷺ ےک تغل اور ریسفت ےس قلعتم الکیکیس االسیم ادب ںیم "ااجلہ لی ۃ " ےک دصماق یک نیعت [ 88 ] ر ی ی ایک ےہ اظنم دعل و ااصنف ےس مک ب46۔ لبق از االسم ےک احالت ےس اس رطح ےک وحاےل اور رکدار رسیتی دصی رجہی ےک دعب ےھکل ےئگ ادب ںیم ی آاسین لم اجےت ںیہ۔کیل ان رکداروں ےک اسھت اجتیلہ وک وسنمب رکےن اور ان یک رتمکی وک ی ی رکےن اک راحجن تہب مک اتلم ےہ۔ دعب ےک زامونں ںیم زی دہ وعیس اطمہعل اور زجتی یت اہمروتں ےک رتہب وہ اجےن یک انب رپ اجتیلہ ےک وہفمم ںیم وتعس اور رہگایئ دیپا وہیت یئگ یھت۔ تغل اور ریسفت، دوونں ےک اطمہعل ےس واحض وہی ےہ ہک دوونں ںیم یہ "ااجلہ لی ۃ " ےک وہفمم ںیم ای رکفی اراقتء وارد وہا ےہ۔ لبق از االسم رعب ےک ب رے احالت اس اک رمک ری وموضع رےہ ںیہ۔ ہی اراقتء وچدوھںی دصی ی اجری راہ ےہ۔ الہات تاہلا رآین االطصح ااجلہ لی ۃ ےک ونتمع افممیہ املسمن امہ رنی تغل اور رسفمنی ےک اہں ی ےئ اجےت ق ںیہ۔ ضعب املعء ےن اس وک یظفل ،زامین اور ین ےک وحاےل ےس اس یک واضح یک ےہ۔ درگی املعء ےن ان رغجاایفیئ دحود ےک اسھت رموبط رکےت وہےئ ان ء یک تثعب ےک درایمین وق رکدار ںیم وموجد اصخلئ اور اوطار یک رو ےس اس وک شیپ ایک ےہ۔ ی ین اذلرک ےقبط اک شیپ رکدہ وہفمم زی دہ رعموف اور وبقمل ےہ ویکہکن اس رآہین اعمرص احالت رپ ی آاسین قبطنم وہ اجیت ںیہ۔ وہفمم ںیم اجتیعم اور رحتک ی یئ اجیت ےہ سج ےس آی ت ق اتای بیاتا دمحم نب ج رب ر الطت رری ، اجعم اایبلن، دار ارکفل،ریبوت،( 1999 )ء دمحم نب ج رب ر ربطی، ی رخی ارللس واولملک،دار ارتلاث ارعلیب ،،ریبوت )(س ن ،دمحم نب اامسلیع ااخبلری،حیحص اخبری(،دار االسلم،الوہر2004 )ء ،رادیہی ،اتکب انیعل،وزارۃ ااقثلف ، دغباد لیلخ نب ادمح ق( 1980ء) ،اوب دمحم دبع اہلل نب ملسم نب ہبیتق اوکلیف ،ااعملرف،ہبتکم ارشلفی ارلاض،مق،اب ران( 1994ء) دمحم نب ادمح ،االزہ ری،ذہتی اۃغلل،دار ابتکل العلمی ۃ ،ریبوت( 2004ء ) ومحمد نب رمع نب دمحم نب أدمح اوخلارزم ازلرشخمي،، ااسس االبلع ۃ ،دار اصدر،ریبوت( 1992 ء) ی وشنان نب عس د الجمت رری الی می ،،سمش اولعلم،دار ارکفل،دقشم( 1999ء ) ، دبع ارلمحٰ ادلاریم ،ننس داریم ااصنر اۃنسل، یلبپ )زنشیک،الوہر،(س ن ،انب وظنمر، اسلن ارعلب،دار اصدر،ریبوت ( 1990ء) ، نب امیلسن، ریسفت ارقلآن امیظعل،ۃبتکم اایحلءۃ المصررۃی امع للکی ا ب، اقہ رہ اقب( 1979ء1989ء) انب رمع زرشحمی، ااشکلف،دار ابتکل العلمی ۃ ،ریبوت(، 1995ء ) M.G Carter, Arabic Lexicography, Cambridge University Press, (1990) Bernard Weiss, Langusge and tradition in medieval Islam, Der Islam, Vol 61, (1984) [ 89 ] Al – Aijaz, Research Journal of Islamic Studies & Humanities Vol. 17 :االزحاب33 :، ااملدئہ50اقنامیلریسفرقمیظع وحاہل اجت وحہل اجت 1 دمحم نب ج رب ر الطت رری ، اجعم اایبلن، دار ارکفل،ریبوت،( 1999 )ء ج9،ص438 2 ،دمحم نب اامسلیع ااخبلری،حیحص اخبری دار(،االسلم،الوہر2004 )ء اتکب االامین، ی ب ربمن22 3 ،رادیہی ،اتکب انیعل،وزارۃ ااقثلف ، دغباد لیلخ نب ادمح ق( 1980ء)،ج3 ، ص390 4 ااضی، ج8، ص115 5 :وھد27،28،29 6 :االاعنم111 7 ،اوب دمحم دبع اہلل نب ملسم نب ہبیتق اوکلیف ،ااعملرف،ہبتکم ارشلفی ارلاض،مق،اب ران( 1994ء) ، ص54 8 دمحم نب ادمح،االزہ ری، ذہتی اۃغلل،دار ابتکل العلمی ۃ ،ریبوت( 2004ء)، ج4،ص312،313 9 ااضی، ص314 10 اس ےک وتمازی اامم ربطی وج وچیھت دصی رجہی ےک یہ ومرخ اور رسفم ںیہ، ےن اینپ اتکب ی رخی ارللس واولملک ںیم یبن ارکمﷺ یک دوعیت اسم یع ےک زامےن ےک اقمےلب ںیم ،رار دی ےہ۔(دمحم نب ج رب ر ربطی، ی رخی ارللس واولملک،دار ارتلاث ارعلیب،ریبوت رعب ےک لبق از االسم زامےن وک "ااجلہ لی ۃ " ق)(س ن ج1 ،ص 232 ، ج2 ، ص308 ) اس ےک اسھت اسھت اامم ربطی ےن عی سیٰ ہیلع االسلم یک آدم ےس لبق ونب ارسالی ےک احالت وک یھب ایس ہملک ےس وموسم ایک ےہ۔(ی رخی ارللس واولملک،ج1 ، ص 590 ) 11 ومحمد نب رمع نب دمحم نب أدمح اوخلارزم ازلرشخمي،، ااسس االبلع ۃ ،دار اصدر،ریبوت( 1992 ء) ، ص107 12 :احتفل26 13 ی وشنان نب عس د الجمت رری الی می ،،سمش اولعلم،دار ارکفل،دقشم( 1999ء) ،ج2 ، ص1199 14 ، دبع ارلمحٰ ادلاریم ،ننس داریم)ااصنر اۃنسل، یلبپ زنشیک،الوہر،(س ن دحی ربمن1823 15 ،انب وظنمر، اسلن ارعلب،دار اصدر،ریبوت ( 1990ء) ، ج11 ، ص130 16 M.G Carter, Arabic Lexicography, Cambridge University Press, (1990), P. 106 Bernard Weiss, Langusge and tradition in medieval Islam, Der Islam, Vol 61, (1984), P. Abstract T V, No. 4 (Oct – Dec 2021) مر 22 تفسیر ، ج اقمب1 ، ص308امد 23 :ااملدئہ50ریسفاق وحاہل اجت 99 17از33املدئ50 زادما 18 ، نب امیلسن، ریسفت ارقلآن امیظعل،ۃبتکم اایحلءۃ المصررۃی امع للکی ا ب، اقہ رہ اقمب( 1979ء1989ء) ، ج1، ص483۔ج2،ص488 19 :احتفل26 تغل اور ریسفت ےس قلعتم الکیکیس االسیم ادب ںیم "ااجلہ لی ۃ " ےک دصماق یک نیعت [ 90 ] 26 دمحم نب ج رب ر،ربطی، ریسفت اجعم اایبلن،دار ارکفل،ریبوت1999ء، ج22 ، ص6،7 27 ریسفت ربطی، ج22 ، ص7 28 ریسفت ربطی، ج4، ص188 29 ریسفت ربطی، ج4 ، ص190 30 ریسفت ربطی، ج6،ص371 31 تفسیر ربطی،ج26،ص135 32 انب رمع زرشحمی، ااشکلف،دار ابتکل العلمی ۃ ،ریبوت(، 1995ء)، ج1 ، ص420 33 ااشکلف، ج1 ، ص420 ،، ارقلیبط، ااجلعم الاکحم ارقلآن،دار ابتکل العلمی ۃ ،ریبوت( 2000ء)، ج4 ، ص156 34 :آل رمعان154 35 ریبط، ج ریسفت ق4 ، ص156 36 ان ےک ایخل ںیم ہی آدم ہیلع االسلم اور ونح ہیلع االسلم، ونح آدم ہیلع االسلم اور ادرسی آدم ہیلع االسلم ، داؤد آدم ہیلع االسلم اور امیلسن آدم ہیلع االسلم اک درایمین زامہن وہ اتکس ےہ۔دےیھکی ااشکلف، ج3 ، ص521 37 ااشکلف،ج3،521 ، ااسس اۃغلل، ص107 38 ریبط، ج ریسفت ق14 ، ص117 39 ریسفت ریبط،ج ق16 ، ص190 40 ریبط،ج ریسفت ق6،ص139 41 ریبط، ج ریسفت ق6،ص139،140 42 :االزحاب33 43 ریبط، ج ریسفت ق14 ، ص117 44 ااشکلف،ج1 ، ص628 ۔زرشحمی لھج وک ملع ےک اقمےلب ںیم شیپ رکےت ںیہ۔ 45 :ااملدئہ50 46 ااشکلف، ج1،ص628،629 References [ 90 ] 26 دمحم نب ج رب ر،ربطی، ریسفت اجعم اایبلن،دار ارکفل،ریبوت1999ء، ج22 ، ص6،7 27 ریسفت ربطی، ج22 ، ص7 28 ریسفت ربطی، ج4، ص188 29 ریسفت ربطی، ج4 ، ص190 30 ریسفت ربطی، ج6،ص371 31 تفسیر ربطی،ج26،ص135 32 انب رمع زرشحمی، ااشکلف،دار ابتکل العلمی ۃ ،ریبوت(، 1995ء)، ج1 ، ص420 33 ااشکلف، ج1 ، ص420 ،، ارقلیبط، ااجلعم الاکحم ارقلآن،دار ابتکل العلمی ۃ ،ریبوت( 2000ء)، ج4 ، ص156 34 :آل رمعان154 35 ریبط، ج ریسفت ق4 ، ص156 36 ان ےک ایخل ںیم ہی آدم ہیلع االسلم اور ونح ہیلع االسلم، ونح آدم ہیلع االسلم اور ادرسی آدم ہیلع االسلم ، داؤد آدم ہیلع االسلم اور امیلسن آدم ہیلع االسلم اک درایمین زامہن وہ اتکس ےہ۔دےیھکی ااشکلف، ج3 ، ص521 37 ااشکلف،ج3،521 ، ااسس اۃغلل، ص107 38 ریبط، ج ریسفت ق14 ، ص117 39 ریسفت ریبط،ج ق16 ، ص190 40 ریبط،ج ریسفت ق6،ص139 41 ریبط، ج ریسفت ق6،ص139،140 42 :االزحاب33 43 ریبط، ج ریسفت ق14 ، ص117 44 ااشکلف،ج1 ، ص628 ۔زرشحمی لھج وک ملع ےک اقمےلب ںیم شیپ رکےت ںیہ۔ 45 :ااملدئہ50 46 ااشکلف، ج1،ص628،629 R f اشکلف ج1ص628 629 References 1. برفسیر 31 تفسیر ربطی،ج26،ص135رشحاشکبتکنرم وحاہل اجت Muhammad bin Jarir al-Tabari, Jami al-Bayan, Dar-ul-Fikr, Beirut, vol. 9, p. 438 2. Muhammad bin Isma'il al-Bukhari, Sahih Bukhari, Dar-ul-Islam, Lahore, (2004) Kitab-ul-Iman, Chapter 22 3. Khalil bin Ahmad Farahidi, Kitab al-Ain, Wazarat al-Thaqafah, Baghdad, vol. 3, p. 390 4. Ibid., vol. 8, p. 115. 5. Hud: 27, 28, 29 6. Al-Anam:111 7. Abu Muhammad Abdullah bin Muslim bin Qutayba al-Kufi, Al-Ma'ari, Maktaba al-Sharif al-Rida, Qom, Iran, (1994), p. 54 8. Muhammad bin Ahmad al-Azhari, Tahzeeb al-Lagha, Dar-ul-Kutub al-Ilmiyyah, Beirut, vol. 4, p. 312, 313 9. Ibid., p. 314 10. In parallel, Imam Al-Tabari, a historian and commentator dating back to the 4th/12th century, described the pre-Islamic period of Arabia as "al-Jahiliyyah" in contrast to the time of the Prophet's (peace and blessings of Allaah be upon him). (Muhammad bin Jarir Al-Tabari, Tarikh al-Rasal wa'l- وحاہل اجت 99 17 :االزحاب33 :، ااملدئہ50 18 ، نب امیلسن، ریسفت ارقلآن امیظعل،ۃبتکم اایحلءۃ المصررۃی امع للکی ا ب، اقہ رہ اقمب( 1979ء1989ء) ، ج1، ص483۔ج2،ص488 19 :احتفل26 20 ، ج ریسفت اقمب4، ص76 21 :آل رمعان154 22 تفسیر ، ج اقمب1 ، ص308 23 :ااملدئہ50 24 ، ج ریسفت اقمب1 ، ص483،483 25 :االزحاب33 1 دمحم نب ج رب ر الطت رری ، اجعم اایبلن، دار ارکفل،ریبوت،( 1999 )ء ج9،ص438 2 ،دمحم نب اامسلیع ااخبلری،حیحص اخبری دار(،االسلم،الوہر2004 )ء اتکب االامین، ی ب ربمن22 3 ،رادیہی ،اتکب انیعل،وزارۃ ااقثلف ، دغباد لیلخ نب ادمح ق( 1980ء)،ج3 ، ص390 4 ااضی، ج8، ص115 5 :وھد27،28،29 6 :االاعنم111 7 ،اوب دمحم دبع اہلل نب ملسم نب ہبیتق اوکلیف ،ااعملرف،ہبتکم ارشلفی ارلاض،مق،اب ران( 1994ء) ، ص54 8 دمحم نب ادمح،االزہ ری، ذہتی اۃغلل،دار ابتکل العلمی ۃ ،ریبوت( 2004ء)، ج4،ص312،313 9 ااضی، ص314 10 اس ےک وتمازی اامم ربطی وج وچیھت دصی رجہی ےک یہ ومرخ اور رسفم ںیہ، ےن اینپ اتکب ی رخی ارللس واولملک ںیم یبن ارکمﷺ یک دوعیت اسم یع ےک زامےن ےک اقمےلب ںیم ،رار دی ےہ۔(دمحم نب ج رب ر ربطی، ی رخی ارللس واولملک،دار ارتلاث ارعلیب،ریبوت رعب ےک لبق از االسم زامےن وک "ااجلہ لی ۃ " ق)(س ن ج1 ،ص 232 ، ج2 ، ص308 ) اس ےک اسھت اسھت اامم ربطی ےن عی سیٰ ہیلع االسلم یک آدم ےس لبق ونب ارسالی ےک احالت وک یھب ایس ہملک ےس وموسم ایک ےہ۔(ی رخی ارللس واولملک،ج1 ، ص 590 ) 11 ومحمد نب رمع نب دمحم نب أدمح اوخلارزم ازلرشخمي،، ااسس االبلع ۃ ،دار اصدر،ریبوت( 1992 ء) ، ص107 12 :احتفل26 13 ی وشنان نب عس د الجمت رری الی می ،،سمش اولعلم،دار ارکفل،دقشم( 1999ء) ،ج2 ، ص1199 14 ، دبع ارلمحٰ ادلاریم ،ننس داریم)ااصنر اۃنسل، یلبپ زنشیک،الوہر،(س ن دحی ربمن1823 15 ،انب وظنمر، اسلن ارعلب،دار اصدر،ریبوت ( 1990ء) ، ج11 ، ص130 16 M.G Carter, Arabic Lexicography, Cambridge University Press, (1990), P. 106 Bernard Weiss, Langusge and tradition in medieval Islam, Der Islam, Vol 61, (1984), P. برفسیر 30 ریسفت ربطی، ج6،ص371ربریسف References References 1. Muhammad bin Jarir al-Tabari, Jami al-Bayan, Dar-ul-Fikr, Beirut, vol. 9, p. 438 References 1. Muhammad bin Jarir al-Tabari, Jami al-Bayan, Dar-ul-Fikr, Beirut, vol. 9, p. 438 2. Muhammad bin Isma'il al-Bukhari, Sahih Bukhari, Dar-ul-Islam, Lahore, (2004) Kitab-ul-Iman, References 1. Muhammad bin Jarir al-Tabari, Jami al-Bayan, Dar-ul-Fikr, Beirut, vol. 9, p. 438 7. Abu Muhammad Abdullah bin Muslim bin Qutayba al-Kufi, Al-Ma'ari, Maktaba al-Sharif al-Rida, Qom, Iran, (1994), p. 54 ( ) p 8. Muhammad bin Ahmad al-Azhari, Tahzeeb al-Lagha, Dar-ul-Kutub al-Ilmiyyah, Beirut, vol. 4, p. 312, 313 9 Ibid p 314 9. Ibid., p. 314 10. In parallel, Imam Al-Tabari, a historian and commentator dating back to the 4th/12th century, described the pre-Islamic period of Arabia as "al-Jahiliyyah" in contrast to the time of the Prophet's (peace and blessings of Allaah be upon him) (Muhammad bin Jarir Al Tabari Tarikh al Rasal wa'l 9. Ibid., p. 314 10. In parallel, Imam Al-Tabari, a historian and commentator dating back to the 4th/12th century, described the pre-Islamic period of Arabia as "al-Jahiliyyah" in contrast to the time of the Prophet's (peace and blessings of Allaah be upon him) (Muhammad bin Jarir Al-Tabari Tarikh al-Rasal wa'l- [ 91 ] Al – Aijaz, Research Journal of Islamic Studies & Humanities Vol. V, No. 4 (Oct – Dec 2021) Muluk, Dar al Tarath al Arabi, Beirut, vol. 1, p. 232, vol. 2, p. 308) At the same time, Imam Al-Tabari has also named the situation of Banu Israel before the arrival of Jesus (peace be upon him) with the same word. (Tarikh al-Rasal wa'l-Muluk, Vol. 1, p. 590) 11. Mahmud bin Umar bin Muhammad bin Ahmad al-Khwarizmi al-Zamakhshari, Asas al-Balagha, Dar Sadr, Beirut, (1992), p. 107. 12 F th 26 12. Fath:26 12. Fath:26 13. Nishwan bin Sa'id al-Humayri Al-'Alimani, Shams-ul-Ulum, Dar-ul-Fikr, Damascus, (1999), vol. 2, p. 1199 bd al-Rahman al-Darmi, Sunan Darmi, Ansar al-Sunnah, Publications, Lahore, Hadith No. 1823 15. Ibn Manzoor, Lasan al-Arab, Dar Sadr, Beirut, (1990), vol. 11, p. 130 15. Ibn Manzoor, Lasan al-Arab, Dar Sadr, Beirut, (1990), vol. 11, p. 130 ( ) p 16. M.G Carter, Arabic Lexicography, Cambridge University Press, (1990), P. 106 g p y g y ( ) . Bernard Weiss, Langusge and tradition in medieval Islam, Der Islam, Vol 61, (1984), P. 99 Al Ahzab:33 Al Ma'ida:50 g p y g y ( ) 17. 20. Fath:26 20. Fath:26 21. Tafsir Al-Muqatil, vol. 4, p. 76 22. Al-Imran: 154 23. Tafseer Muqatil, vol. 1, p. 308 24. Al-Ma'ida:50 24. Al-Ma'ida:50 25. Tafseer Muqatil, vol. 1, pp. 483-483 26. Al-Ahzab: 33 26. Al Ahzab: 33 27. Muhammad bin JarirTabari, Tafseer Jami al-Bayan, Dar-ul-Fikr, Beirut, 1999, vol. 22, pp. 6,7 28 T f i l T b i l 22 7 27. Muhammad bin JarirTabari, Tafseer Jami al-Bayan, Dar-ul-Fikr, Beirut, 1999, vol. 22, pp. 6,7 27. Muhammad bin JarirTabari, Tafseer Jami al-Bayan, Dar-ul-Fikr, 28. Tafsir al-Tabari, vol. 22, p. 7 29. Tafsir al-Tabari, vol. 4, p. 188 30. Tafsir al-Tabari, vol. 4, p. 190 31. Tafsir al-Tabari, vol. 6, p. 371 32. Tafsir al-Tabari, vol. 26, p. 135 33. Ibn 'Umar Al-Zamhashri, al-Kashaf, Dar-ul- yy p 34. Al-Kashaf, vol. 1, p. 420, al-Qurtawi, al-Jami al-Aqsa al-Qur'an, Dar-ul-Kutub al-Ilmiyyah, Beirut, (2000), vol. 4, p. 156. 34. Al-Kashaf, vol. 1, p. 420, al-Qurtawi, al-Jami al-Aqsa al-Qur'an, Dar-ul-Kutub al-Ilmiyyah, Beirut, (2000) l 4 156 34. Al-Kashaf, vol. 1, p. 420, al-Qurtawi, al-Jami al-Aqsa al-Qur'an, Dar-ul-Kutub al-Ilmiyyah, Beirut, (2000), vol. 4, p. 156. (2000), vol. 4, p. 156. References Bernard Weiss, Langusge and tradition in medieval Islam, Der Islam, Vol 61, (1984), P 17. Bernard Weiss, Langusge and tradition in medieval Islam, Der Islam, Vol 61, (1984), P. 99 18. Al-Ahzab:33, Al-Ma'ida:50 18. Al-Ahzab:33, Al-Ma'ida:50 , 19. Muqatil bin Sulaiman, Tafsir al-Qur'an al-Azim, Maktabat al-Hayyat al-Misriyyah al-'Ama al-Kitab, Cairo (1979-1989), vol. 1, p. 483. Vol. 2, p. 488 20 F h 26 19. Muqatil bin Sulaiman, Tafsir al-Qur'an al-Azim, Maktabat al-Hayyat al-Misriyyah al-'Ama al-Kitab, Cairo (1979-1989), vol. 1, p. 483. Vol. 2, p. 488 (2000), vol. 4, p. 156. ( ) p 35. Al-Imran: 154 36. Tafsir al-Qurtubi, vol. 4, p. 156 Q , , p 37. In their view, this could be the time between Adam and Noah, Noah Adam and Idris Adam, David Adam and Solomon Adam. See al-Kashaf, vol. 3, p. 521. Q , , p 37. In their view, this could be the time between Adam and Noah, Noah Adam and Idris Adam, David Adam and Solomon Adam. See al-Kashaf, vol. 3, p. 521. 38 Al K h f l 3 521 A l L h 107 37. In their view, this could be the time between Adam and Noah, No Adam and Solomon Adam. See al-Kashaf, vol. 3, p. 521. 38. Al-Kashaf, vol. 3, p. 521, Asas al-Lagha, p. 107 38. Al-Kashaf, vol. 3, p. 521, Asas al-Lagha, p. 107 39. Tafsir al-Qurtubi, vol. 14, p. 117 40. Tafseer Al-Qurtubi, vol. 16, p. 190 41. Tafseer Al-Qurtubi, vol. 6, p. 139 42. Tafsir al-Qurtubi, vol. 6, pp. 139,140 43. Al-Ahzab:33 44. Tafsir al-Qurtubi, vol. 14, p. 117 Q p 45. Al-Kashaf, vol. 1, p. 628. Zamhashri presents jhal against knowledge. 46 Al M 'id 50 45. Al-Kashaf, vol. 1, p. 628. Zamhashri presents jhal against knowledg 46. Al-Ma'ida:50 47. Al-Kashaf, vol. 1, pp. 628,629
https://openalex.org/W2111273768	http://ira.lib.polyu.edu.hk/bitstream/10397/5850/1/Yu_application_kernel_density.pdf	English	null	Application of Kernel Density Estimation in Lamb Wave-Based Damage Detection	Mathematical problems in engineering	2,012	cc-by	11,496	Hindawi Publishing Corporation Mathematical Problems in Engineering Volume 2012, Article ID 406521, 24 pages doi:10.1155/2012/406521 Hindawi Publishing Corporation Mathematical Problems in Engineering Volume 2012, Article ID 406521, 24 pages doi:10.1155/2012/406521 Hindawi Publishing Corporation Mathematical Problems in Engineering Volume 2012, Article ID 406521, 24 pages doi:10.1155/2012/406521 1 School of Mechanics and Civil & Architecture, Northwestern Polytechnical University, Xi’an, Shaanxi 710129, China 2 Department of Mechanical Engineering, The Hong Kong Polytechnic University, Hong Kong 1 School of Mechanics and Civil & Architecture, Northwestern Polytechnical University, Xi’an, Shaanxi 710129, China 2 Department of Mechanical Engineering, The Hong Kong Polytechnic University, Hong Kong Correspondence should be addressed to Long Yu, yulong@nwpu.edu.cn Received 13 April 2012; Revised 15 June 2012; Accepted 20 June 2012 Academic Editor: Alessandro Marzani Copyright q 2012 L. Yu and Z. Su. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The present work concerns the estimation of the probability density function p.d.f. of measured data in the Lamb wave-based damage detection. Although there was a number of research work which focused on the consensus algorithm of combining all the results of individual sensors, the p.d.f. of measured data, which was the fundamental part of the probability-based method, was still given by experience in existing work. Based on the analysis about the noise-induced errors in measured data, it was learned that the type of distribution was related with the level of noise. In the case of weak noise, the p.d.f. of measured data could be considered as the normal distribution. The empirical methods could give satisﬁed estimating results. However, in the case of strong noise, the p.d.f. was complex and did not belong to any type of common distribution function. Nonparametric methods, therefore, were needed. As the most popular nonparametric method, kernel density estimation was introduced. In order to demonstrate the performance of the kernel density estimation methods, a numerical model was built to generate the signals of Lamb waves. Three levels of white Gaussian noise were intentionally added into the simulated signals. The estimation results showed that the nonparametric methods outperformed the empirical methods in terms of accuracy. 1. Introduction Structural health monitoring SHM is an emerging technology that merges with a variety of techniques related to diagnostics and prognostics. Monitoring the status of structural health can improve the safety and maintainability of critical structures in many ﬁelds, such as civil engineering, aerospace, and military industry. An ideal SHM system includes several subsystems in which the damage detection methodology is the key part. Therefore, 2 Mathematical Problems in Engineering numerous damage-detection methods have been researched in years 1. The method based on Lamb waves has the apparent advantages of high sensitivity to structural damage compared with methods based on the mode shapes 2 or structure dynamic responses 3. It has been veriﬁed that the Lamb wave-based damage detection methods can detect crack, delamination, surface corrosion, penetrate holes, weld defect, and many other kinds of damage in plate and shell structure 4–6. Consequently, the Lamb wave is widely acknowledged as one of the most encouraging tools for SHM. The relevant research has been conducted intensively since the 1980s 7. The portion of the SHM process that has received the least attention in the technical literature is the development of statistical models for discrimination between features from the undamaged and damaged structures. The algorithms, which analyze statistical distributions of the measured or derived features to enhance the damage identiﬁcation process, have been developed 8, 9. The probability-based diagnostic methods have also been introduced in Lamb wave-based damage detection area in recent years 10, 11. However, the statistical modes using in the existing Lamb wave-based methods are relatively simple. Despite a number of literatures had been published, which focused on the consensus algorithm of combining all the results of individual sensors, the p.d.f. of the measured data was empirically determined. As a key part of statistical model, it is obvious that the accuracy of the p.d.f. has a signiﬁcant eﬀect on the precision of damage-detecting result. Compared with the estimating results by empirical formula, the results of statistical methods will be more accurate and reliable. Hence, the study of using statistical methods to estimate the p.d.f. is necessary in Lamb wave-based damage detection. Elementary parametric estimation method has been adopted under the assumption that the p.d.f. of the measured data is normal distribution 12. However, the assumption in parametric method limits the application of this method. 2.1. Background Lamb waves are a kind of elastic waves propagates in thin plate and shell structure. With a high susceptibility to interference on a propagation path, for example, damage or a boundary, Lamb waves can travel over a long distance even in materials with a high attenuation ratio, and thus a broad area can be quickly examined 13. Lamb waves are made up of a superposition of longitudinal and shear modes, and its propagation characteristics vary with entry angle, excitation, and structural geometry. A Lamb mode can be either symmetric or antisymmetric, formulated by tanqh tanph − 4k2qp k2 −q22 for symmetric modes, 2.1 tanqh tanph − k2 −q22 4k2qp for antisymmetric modes, 2.2 2.1 where p2 w2/c2 L −k2, q2 w2/c2 T −k2, k w/cp, and h, k, cL, cT, cp, ω are the plate thickness, wavenumber, velocities of longitudinal and transverse modes, phase velocity, and wave circular frequency, respectively. Equations 2.1 and 2.2, correlating the propagation velocity with its frequency, imply that Lamb waves, regardless of its mode, are dispersive velocity is dependent on frequency. Lamb waves can be actively excited by a variety of means, such as ultrasonic probe, laser, interdigital transducer, and piezoelectric element. The piezoelectric element can also be used as sensor to collect signals of Lamb waves perfectly. The piezoelectric element is particularly suitable for integration into a host structure as an in situ generator/sensor, for their neglectable mass/volume, easy integration, excellent mechanical strength, wide- frequency responses, low power consumption and acoustic impedance, as well as low cost. Applications of piezoelectric element in Lamb wave-based damage detection are numerous. Lamb mode selection is an important part for damage detection. The basic symmetric mode, S0, and the antisymmetric mode, A0, are normally used in practice. Although S0 is preferred in many of studies 14, utilization of A0 is increasing because that A0 is the highly eﬀective for detecting delamination and transverse ply cracks 15. To implement the Lamb mode selection, a multielement transducer setup was proposed 16 to dominantly generate S0 or A0. The algorithms for Lamb wave-based damage identiﬁcation can be roughly divided into two categories. The ﬁrst category is the algorithms that identify and locate damage by observing the damage-reﬂected Lamb waves, such as Time-of-Flight ToF method 17–19, embedded ultrasonic structural radar 20, and time of diﬀerence method 21. 1. Introduction If the extra assumption is correct, the results produced by parametric method can be more accurate than the results given by empirical formula. While if the assumption is incorrect, parametric methods can be very misleading. Since the type of p.d.f. of measured data from ﬁeld experiments is varied and can hardly be predicted, more robust approach methods should be considered. The nonparametric statistic methods can give the parameters of distribution and do not rely on assumptions that the data are drawn from a given probability distribution. Therefore, introducing the nonparametric statistic methods is crucial in Lamb wave-based damage detection. The aim of this paper is to demonstrate the necessity and feasibility of application of kernel density estimation, which is the most popular nonparametric estimation method in Lamb wave-based damage detection. Two kinds of kernel density estimation methods, the one based on the Gaussian approximation and the one based on the smoothing properties of linear diﬀusion processes, were brieﬂy introduced in this paper. The signals of Lamb waves with diﬀerent levels of white Gaussian noise were acquired by using numerical simulation. The framework of applying nonparametric estimation method in Lamb wave-based damage detection was demonstrated by using the simulated signals. The characteristics of noise- induced error in the arriving time of damage-scattered Lamb waves, which is the index used to locate damage, was analyzed. Based on this analysis, the outcomes of two kinds of kernel density estimation method as well as the parametric estimation methods were compared. The results show that the nonparametric methods outperform the parametric method in terms of accuracy and reliability. Mathematical Problems in Engineering Mathematical Problems in Engineering 3 2.1. Background The second category is the algorithms that analyze the changes in the characteristics of Lamb waves caused by the damage in its propagation path, such as tomography method 22 and virtual- sensing paths method 23. For the algorithms that focus on the damage-reﬂected waves, the arriving time of the Lamb waves is the key index used to locate damage. Since the signal of Lamb waves is wave packet in the form, several methods have been developed to measure the arriving time of Lamb waves, such as threshold method, correlation method, wavelet method 24, Mathematical Problems in Engineering 4 and a novel cross-correlation analysis method based on a wavelet transform 25, 26. Among those methods, the threshold method, which was adopted in this paper, has the advantage of simplicity. In threshold method, a threshold value Vt was ﬁrstly set up on basis of experience. Once the amplitude of one or several peaks exceed Vt, then the corresponding peaks were recorded. Depending on the magnitude of Vt, one or more peaks could be recorded for a wave packet. If only one peak was recorded, the arriving time was the time corresponding to that peak. If more than one peak were recorded, then the arriving time will be the average of all recorded time. Usually, the threshold value is selected to let several peaks belong to one wave packet can be recorded. The beneﬁt of recording several peaks instead of only the strongest peak is that the averaging process itself can reduce noise to some extent. 2.2. Time of Flight Method ToF, deﬁned as the time lag from the moment when a sensor catches the damage reﬂected signal to the moment when the same sensor catches the incident signal, was widely used to locate damage 17–19. Consider a sensor network consisting of N piezoelectric wafers denoted by si i 1, 2, . . . , N. For convenience of discussion, sm −sn hereinafter stands for the sensing path in which sm serves as the actuator and sn as the sensor. The center of the damage, if any, is presumed to be x, y in coordinate system. Then, the ToF can be deﬁned in 2.1 as Ti−j: LA−D VS0 LD−S VSH0−damage −LA−S VS0 Ti−j, 2.3 2.3 In which LA−D, LD−S, and LA−S represent the distance from the actuator si to the damage, from the damage to the sensor sj, and from actuator si to the sensor sj, respectively. VSH0−damage and j j g VS0 are velocities of the damage-converted SH0 mode and the incipient S0 mode, respective Because there are two unknown damage parameters, x, y, in 2.3, the solution of 2.3 will be a root locus, which implies the possible locations of the damage for a certain ToF value. In traditional approaches, the damage location is given by seeking the intersections of two or more loci. As shown in Figure 1a, in the case of using three sensor pairs, there will be three loci, each exhibiting a time delay due to the existence of damage. The point with which all three loci intersect was considered as the location of damage, while the points with which only two loci intersect were considered as pseudodamage location. There is a prerequisite in the traditional approach. That is all of the measured ToF values Tm were accurate. However, errors are always inevitable in any experimental result due to the reasons such as noises. Therefore, as shown in Figure 1b, there is no point with which all three loci intersect if the loci were drawn based on noise contained Tm instead of the theoretical value T. It is suggested that the damage location can be given as the area where the density of intersections of two loci is relatively large. That leads to the research about the probability-based approach method, to give the precise damaged area based on the density of intersections. 2.2. Time of Flight Method 5 5 Mathematical Problems in Engineering sk si sj sl Locus given by si −sj Locus given by si −sk Locus given by si −sl Damage LA−S LD−S LA−D a Damage area sk sj si sl b Figure 1: Damage localization using ToF method in a plate. a Locus based on accurate ToF value, b Locus based on ToF with error. sk si sj sl Locus given by si −sj Locus given by si −sk Locus given by si −sl Damage LA−S LD−S LA−D a Damage sl a Damage area sk sj si sl b 1: Damage localization using ToF method in a plate a Locus based on a Damage area sk sj si sl b Figure 1: Damage localization using ToF method in a plate. a Locus based on accurate ToF value, b sl b Figure 1: Damage localization using ToF method in a plate. a Locus based on accurate ToF value, Locus based on ToF with error. 2.3. Probability-Based ToF Method The detection result in matrix form can be illustrated in an image shown in Figure 2, where the lighter the greyscale, the greater the possibility of damage existing at that pixel each pixel exclusively corresponds to a spatial point of the structure under inspection. It is obvious that the p.d.f. of damage occurrence is the key part of probability based method. Su et al. 10 suggest the p.d.f. can be quantiﬁed in relation to the loci: fzij 1 σij √ 2π exp − z2 ij 2σ2 ij , 2.4 2.4 where fzij is the Gaussian distribution function, representing the p.d.f. of damage occurrence at node Li i 1, . . . , K × K for the structure that is comprised of K × K mesh nodes, perceived by a sensor, sj j 1, . . . , N for the sensor network consisting of N sensors. σij is the standard deviation and where fzij is the Gaussian distribution function, representing the p.d.f. of damage occurrence at node Li i 1, . . . , K × K for the structure that is comprised of K × K mesh nodes, perceived by a sensor, sj j 1, . . . , N for the sensor network consisting of N sensors. σij is the standard deviation and zij χi −μij , 2.5 2.5 where χi is the location vector of node Li and μij is the location vector of the point on the locus provided by sensor sj that has the shortest distance to node Li. j Satisﬁed results have been obtained by using this kind of p.d.f. But it should be noticed that the standard variance σij was selected depending on experience. The concept of probability-based approach was also adopted in some other Lamb wave-based damage detection methods rather than ToF method. Wang et al. 23 combine the concept of probability-based approach with virtual-sensing paths method. The p.d.f. in their work is an empirical formula and the parameters were given by experience. There are mainly two disadvantages in the existing work. First, empirical formula usually are simpler to write down and faster to compute, but it depends heavily on the experimental environment. Any change which is inevitable in experiment may cause a big error in the estimated results. 2.3. Probability-Based ToF Method The concept of probability-based approach was introduced by Zhao et al. 27 to improve the performance of Lamb wave-based method, and then it was adopted by Su et al. 28 in ToF method. In traditional ToF method, only the points on loci are considered as possible damage location. Other points, regardless of its distance to the loci, will all be excluded outside the possible damage location. In fact, due to the existence of errors in Tm, the real damage may not be on the loci which were drawn based on Tm. Therefore, in probability-based approach method, the points absent in the loci are also considered as possible damage location. The possibility of damage occurrence in those points will be determined by its distance to 6 Mathematical Problems in Engineering Mathematical Problems in Engineering the loci. The mesh nodes right located on an above-established locus have the highest degree of probability of damage presence; for the others, the greater the distance to the locus, the lower the probability damage exists there. To quantify the probabilities at all nodes with regard to all loci, a function called as p.d.f. of damage occurrence was introduced. For each loci, a probability distribution map can be given for the detection target plant structure based on p.d.f. of damage occurrence. Combination of all the probability distribution maps can give the ﬁnal damage detection result. g The main frame of data fusion-based method can be divided into two steps. g The main frame of data fusion-based method can be divided into two steps. 1 The inspection area of the structure was evenly meshed. For a certain measured ToF, each mesh node will be evaluated about its possibility for the presence of damage by using a probability density function. 1 The inspection area of the structure was evenly meshed. For a certain measured ToF, each mesh node will be evaluated about its possibility for the presence of damage by using a probability density function. 2 All evaluated results for each measured ToF were combined to give the detection result in a matrix form. Each element of the matrix represents the probability of the presence of damage for one mesh node. Mathematical Problems in Engineering 7 g g 0 0.1 0.2 0.3 0.4 0.5 0.6 0 0.1 0.2 0.3 0.4 0.5 0.6 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 (m) (m) Damage location indicated by high probability area Probability of damage Figure 2: Damage localization result of probability-based method. Figure 2: Damage localization result of probability-based method. function about the location of damage in the plane fDTs, where s \|DT −DTm\|, DT and DTm are the damage location corresponding to T the actual ToF data and Tm the experimental ToF data, respectively. It should be noticed that the damage location cannot be directly measured in experiment. Thus, estimating fDTs directly will be diﬃcult. Based on the estimation of the function fTmt about the distribution of experiment data Tm in time domain, estimating fDTs by using the mapping relationship deﬁned in 2.3 should be a better method. Therefore, probability density estimation methods will be introduced in Section 3. The advantages and feasibility of applying probability density estimation methods in ToF method will be demonstrated. 3. Probability Density Estimation In statistic, density estimation is the method that estimates the parameters of a distribution based on the observed samples. Depending on whether a priori knowledge about the type of the distribution is required, the density estimation methods can be divided into two categories: parametric estimation and nonparametric estimation. 2.3. Probability-Based ToF Method That is, the simplicity of empirical formula makes up for its nonrobustness. Since the data measurement work in the Lamb wave-based damage detection is not time consuming, it is reasonable that the density function should be estimated by using robust statistic method. Second, the p.d.f. used in existing work is the distribution Mathematical Problems in Engineering 0 0.1 0.2 0.3 0.4 0.5 0.6 0 0.1 0.2 0.3 0.4 0.5 0.6 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 (m) (m) Damage location indicated by high probability area Probability of damage Figure 2: Damage localization result of probability-based method. Mathematical Problems in Engineering Mathematical Problems in Engineering 3.1. Parametric Estimation Parametric estimation mainly includes point estimation and interval estimation. In statistics, point estimation is the use of sample data to calculate a single number of possible values of an unknown population parameter, in contrast to interval estimation, which is an interval. Most commonly used point estimation methods are method of moment estimation, maximum likelihood estimation, and Bayesian estimation. For instance, if it is known that the sample data come from a normal distribution, then the two parameters of normal distribution, Mathematical Problems in Engineering 8 expectation and variance, can be calculated by using 3.1 and 3.2, which is derived by using maximum-likelihood estimation method: expectation and variance, can be calculated by using 3.1 and 3.2, which is derived by using maximum-likelihood estimation method: μ 1 N N i1 xi, 3.1 σ2 1 N N i1 xi − μ2, 3.2 μ 1 N N i1 xi, 3.1 σ2 1 N N i1 xi − μ2, 3.2 μ 1 N N i1 xi, 3.1 3.1 σ2 1 N N i1 xi − μ2, 3.2 3.2 where N is the number of samples. where N is the number of samples. 3.2. Nonparametric Estimation Nonparametric estimation is a method that estimates the parameters of an unknown distri- bution while does not rely on assumptions about the type of this distribution. Commonly, nonparametric estimation methods include histogram, nonparametric regression, and kernel density estimation, which is the most popular one. 3.2.1. Kernel Density Estimation Based on the Gaussian Approximation Kernel density estimation is a nonparametric method to estimate the probability density function of a random variable. Kernel density estimation is a fundamental data smoothing problem where inferences about the population are made, based on a ﬁnite-data sample. In some ﬁelds such as signal processing and econometrics, kernel density estimation was also termed as the Parzen-Rosenblatt window method, after Emanuel Parzen and Murray Rosenblatt, who are usually credited with independently creating this method in its current form 29, 30. Let x1, x2, . . . , xn be an independent and identically distributed sample drawn from some distribution with an unknown density f. Estimating the shape of this function f is interested. Its kernel density estimator is fhx 1 n n i1 Khx −xi 1 nh n i1 Kh x −xi h , 3.3 3.3 where K• is the kernel, a symmetric but not necessary positive function that integrates to one; and h is positive and a smoothing parameter called the bandwidth. A kernel with subscript h is called as the scaled kernel and deﬁned as Khx 1/hKx/h. A range of kernel functions are commonly used: uniform, triangular, biweight, triweight, Epanechnikov, normal, and others. As with the kernel regression, the choice of kernel function is not crucial, but the choice of bandwidth is important. where K• is the kernel, a symmetric but not necessary positive function that integrates to one; and h is positive and a smoothing parameter called the bandwidth. A kernel with subscript h is called as the scaled kernel and deﬁned as Khx 1/hKx/h. A range of kernel functions are commonly used: uniform, triangular, biweight, triweight, Epanechnikov, normal, and others. As with the kernel regression, the choice of kernel function is not crucial, but the choice of bandwidth is important. The bandwidth of the kernel is a free parameter which exhibits a strong inﬂuence on the resulting estimate 31, 32. The most common optimality criterion used to select this Mathematical Problems in Engineering 9 Mathematical Problems in Engineering parameter is the expected L2 risk function, also termed as the Mean Integrated Squared Error MISE; MISEh E fhx −fx 2 dx. 3.4 3.4 Under weak assumptions on f and K 29, 30, MISEh AMISEh o1/nh h4, where o is the little o notation of the family of Bachmann-Landau notations. 3.2.1. Kernel Density Estimation Based on the Gaussian Approximation o1/nh h4 denotes the function family in which every function grows much slower that 1/nh h4 33. The AMISE is the asymptotic MISE which consists of the two leading terms AMISEh RK nh 1 4m2K2h4Rf′′, 3.5 3.5 where Rg gx2dx for a function g, m2K x2Kxdx, and f′′ is the second derivative of f. The minimum of this AMISE is the solution to this diﬀerential equation: where Rg gx2dx for a function g, m2K x2Kxdx, and f′′ is the second derivative of f. The minimum of this AMISE is the solution to this diﬀerential equation: ∂ ∂hAMISEh −RK nh2 m2K2h3Rf′′ 0 3.6 3.6 or hAMISE RK1/5 m2K2/5Rf′′1/5n1/5 . 3.7 3.7 Neither the AMISE nor the hAMISE can be used directly since they involve the unknown density function f or its second derivative f′′. Therefore, a variety of automatic, data-based methods have been developed for selecting the bandwidth. If the kernel function is normal and it is assumed that the distribution being estimated is Gaussian, then it can be derived from 3.7 that optimal choice for h is h 4 σ5 3n 1/5 ≈1.06 σn−1/5, 3.8 3.8 where σ is the standard deviation of the samples. This approximation is termed as the normal distribution approximation, Gaussian approximation, or Silverman’s rule of thumb 32. where σ is the standard deviation of the samples. This approximation is termed as the normal distribution approximation, Gaussian approximation, or Silverman’s rule of thumb 32. 3.2.2. Kernel Density Estimation via Diﬀusion Kernel density estimation is an ongoing research topic in statistics. Botev et al. 34 proposed an adaptive kernel density estimation method based on the smoothing properties of linear diﬀusion processes. This novel approach method includes two parts: ﬁrst, a simple and intuitive kernel estimator with substantially reduced asymptotic bias and mean square error, and better boundary bias performance; second, an improved plug-in bandwidth selection method that completely avoids the Gaussian approximation. The new plug-in method is Mathematical Problems in Engineering 10 thus genuinely “nonparametric,” since it does not require a preliminary normal model for the data. thus genuinely “nonparametric,” since it does not require a preliminary normal model for the data. (I) The Diﬀusion Estimator (I) The Diﬀusion Estimator (I) The Diﬀusion Estimator Given N independent realizations χN ≡{X1, . . . , XN} from an unknown continuous p.d.f. f on X, the Gaussian kernel density estimator is deﬁned as fx; h 1 N N i1 φx, Xi; h, 3.9 3.9 where where φx, Xi; h 1 √ 2πh e−x−Xi2/2h 3.10 3.10 is a Gaussian p.d.f. kernel with location Xi and scale √ h. The scale is the bandwidth in kernel density estimation. is a Gaussian p.d.f. kernel with location Xi and scale √ h. The scale is the bandwidth in kernel density estimation. Chaudhuri and Marron 35 had found that there is a link between the Gaussian kernel density estimator and the well-known Fourier heat equation which is a diﬀusion partial diﬀerential equation PDE. The link is the Gaussian kernel density estimator deﬁned in 3.9 in fact is the unique solution to the Fourier heat equation: ∂ ∂t fx; h 1 2 ∂2 ∂x2 fx; h, x ∈χ, h > 0, 3.11 3.11 with χ ≡R and initial condition fx; 0 Δx, where Δx N i1 δx −Xi is the empirical density of the data χN and δx −Xi is the Dirac measure at Xi. In the heat equation interpretation, the Gaussian kernel in 3.9 is the so-called Green’s function 36 for the diﬀusion PDE in 3.11. Thus, the Gaussian kernel density estimator fx; h can be obtained by evolving the solution of 3.11 up to h. Because any bounded domain can be mapped onto 0, 1 by a linear transformation, there is no loss of generality in assuming that the domain of the data is known as χ ≡0, 1. 3.2.2. Kernel Density Estimation via Diﬀusion Then, the analytical solution of PDE 3.11 with initial condition Δx and the Neumann boundary condition in this case is fx; h 1 N N i1 κx, Xi; h, x ∈0, 1, 3.12 3.12 where the kernel k is given by where the kernel k is given by κx, Xi; h ∞ k−∞ φx, 2k Xi; h φx, 2k −Xi; h. 3.13 3.13 Mathematical Problems in Engineering 11 The Neumann boundary condition is ematical Problems in Engineering 11 Mathematical Problems in Engineering Mathematical Problems in Engineering Mathematical Problems in Engineering 11 The Neumann boundary condition is The Neumann boundary condition is ∂ ∂x fx; h x1 ∂ ∂x fx; h x0 0, 3.14 3.14 and the target of this boundary condition is to ensure that 3.12 satisﬁes the requirements of p.d.f., such as f should be a nonnegative Lebesgue-integrable function and integrates to unity. y It has been proved that the estimator given in 3.12 arises as the solution of the diﬀusion PDE is better in boundary bias properties compared with the traditional estimator given in 3.9. Therefore, motivated by the idea of acquiring the estimator from the solution of diﬀusion PDE, Botev proposed that the most general linear time-homogeneous diﬀusion PDE can be a starting point for the construction of a better kernel density estimator. The simple diﬀusion model described in 3.11 can be extended on the basis of the smoothing properties of the linear diﬀusion PDE: ∂ ∂hgx; h Lgx; H, x ∈χ, t > 0, 3.15 3.15 where the linear diﬀerential operator L is of the form 1/2d/dxaxd/dx·/px, and a and p can be any arbitrary positive function on χ with bounded second derivatives, and the initial condition is gx; 0 Δx. where the linear diﬀerential operator L is of the form 1/2d/dxaxd/dx·/px, and a and p can be any arbitrary positive function on χ with bounded second derivatives, and the initial condition is gx; 0 Δx. The solution of 3.15 can be the diﬀusion kernel estimator and written as gx; h 1 N N i1 κx, Xi; h. 3.2.2. Kernel Density Estimation via Diﬀusion 3.16 3.16 There is no analytical expression for the diﬀusion kernel satisfying 3.16, κ can be written in terms of a generalized Fourier series in the case that χ is bounded: κx, Xi; h px ∞ k0 eλκhϕkxϕk y, 3.17 3.17 where {ϕk} and {λk} are the eigenfunctions and eigenvalues of the Sturm-Liouville problem on 0, 1: where {ϕk} and {λk} are the eigenfunctions and eigenvalues of the Sturm-Liouville problem on 0, 1: L∗ϕk λkϕk, k 0, 1, 2, . . . , ϕ′ k0 ϕ′ k1 0, k 0, 1, 2, . . . , 3.18 3.18 where L∗is of the form 1/2py∂/∂yay∂/∂y·; that is, L∗is the adjoint operator of L. where L∗is of the form 1/2py∂/∂yay∂/∂y·; that is, L∗is the adjoint operator of L. 12 Mathematical Problems in Engineering Mathematical Problems in Engineering (II) Improved Plug-In Bandwidth Selection Method The novel plug-in bandwidth selection method for the diﬀusion estimator deﬁned in 3.16 proposed by Botev is based on the improved plug-in bandwidth selection method for the Gaussian kernel density estimator deﬁned in 3.9. The novel plug-in bandwidth selection method for the diﬀusion estimator deﬁned in 3.16 proposed by Botev is based on the improved plug-in bandwidth selection method for the Gaussian kernel density estimator deﬁned in 3.9. Assuming that f′′ is a continuous square-integrable function, the asymptotically optimal value of h for Gaussian kernel density estimator is the minimize of the ﬁrst-order asymptotic approximation of MISE 37 ∗h 1 2N√π f′′2 2/5 . 3.19 3.19 It is clear from 3.19 that to compute the optimal ∗h, one needs to estimate the functional ∥f′′∥2. Consider the problem of estimating ∥fj∥ 2 for any arbitrary integer j ≥1. The identity ∥fj∥ 2 −1jEff2jX suggests two plug-in estimators: It is clear from 3.19 that to compute the optimal ∗h, one needs to estimate the functional ∥f′′∥2. Consider the problem of estimating ∥fj∥ 2 for any arbitrary integer j ≥1. The identity ∥fj∥ 2 −1jEff2jX suggests two plug-in estimators: the ﬁrst one is −1jEf f2jX −1j N2 N k1 N m1 φ2jXk, Xm; hj , the second one is fj 2 : fj·, h 2 −1j N2 N k1 N m1 φ2jXk, Xm; 2hj . 3.2.2. Kernel Density Estimation via Diﬀusion 3.20 3.20 For a given bandwidth, both estimators −1jEff2jX and ∥ fj∥ 2 aim to estimate the same quantity ∥fj∥ 2. Therefore, ∗hj can be selected to make both estimators asymptotically equivalent in the mean square error sense: ∗hj 1 1/2j1/2 3 1 × 3 × 5 × · · · × 2j −1 N π/2 fj12 2/32j . 3.21 3.21 Computation of ∗hj by using 3.21 involves ∥fj1∥ 2 which is unknown. Thus, each ∗hj is estimated by Computation of ∗hj by using 3.21 involves ∥fj1∥ 2 which is unknown. Thus, each ∗hj is estimated by ∗ hj ⎛ ⎜ ⎝1 1/2j1/2 3 1 × 3 × 5 × · · · × 2j −1 N π/2 fj1 2 ⎞ ⎟ ⎠ 2/32j . 3.22 3.22 Computation of ∥ fj1∥ 2 requires the estimation of ∗ hj1, which in turn requires the estimation of ∗ hj2, and so on, as seen from 3.20 and 3.22. There is the problem of estimating the inﬁnite sequence {∗ hjk, k ≥1}. However, for some l > 0, if ∗ hl1 can be given, then all {∗ hj, 1 ≤j ≤l} can be estimated recursively. Based on this idea, the l-stage direct plug-in bandwidth selector 37 has been proposed. ematical Problems in Engineering 13 Mathematical Problems in Engineering 13 Denote the functional dependence of ∗ hj and ∗ hl1 as Denote the functional dependence of ∗ hj and ∗ hl1 as ∗ hj γj ∗ hj1 . 3.23 ∗ hj γj ∗ hj1 . 3.23 It is then obvious that ∗ hj γjγj1∗ hj2 γjγj1γj2∗ hj3 · · · . For simplicity of notation, the composition can be deﬁned as γkh γ1 · · · γk−1 γkh, k ≥1. 3.24 3.24 The estimate of ∗h satisﬁes ∗ h ξ∗ h1γ ξγ1 ∗ h2 ξγ2 ∗ h3 · · · ξγl ∗ hl1 . 3.2.2. Kernel Density Estimation via Diﬀusion 3.25 3.25 Then, for a given integer l > 0, the l-stage direct plug-in bandwidth selector consists of computing Then, for a given integer l > 0, the l-stage direct plug-in bandwidth selector consists of computing ∗ h ξγl∗hl1, 3.26 3.26 where ∗hl1 is estimated by assuming that f in ∥fl2∥ 2 is a normal density with mean and variance estimated from the data. where ∗hl1 is estimated by assuming that f in ∥fl2∥ 2 is a normal density with mean and variance estimated from the data. It is noticed that the assumption in the l-stage direct plug-in bandwidth selector method can lead to arbitrarily bad estimates of ∗h, when, for example, the true f is far from being Gaussian. Therefore, Botev proposed to ﬁnd a solution to the nonlinear equation: h ξγlh, 3.27 3.27 for some l, using either ﬁxed point iteration or Newton’s method with initial guess h 0. The ﬁxed-point iteration version is formalized in the following Improved Sheather-Jones algorithm: 1 Given l > 2, initialize with z0 ε, where ε is machine precision, and n 0; 2 Set zn1 ξγlzn; 3 if \|zn1 −zn\| < ε, stop and set ∗ h zn1; otherwise, set n : n 1 and repeat from step 2; 3 if \|zn1 −zn\| < ε, stop and set ∗ h zn1; otherwise, set n : n 1 and repeat from step 2; 3 if \|zn1 −zn\| < ε, stop and set ∗ h zn1; otherwise, set n : n 1 and repeat from step 2; 3 if \|zn1 −zn\| < ε, stop and set ∗ h zn1; otherwise, set n : n 1 and repeat from step 2; 4 Deliver the Gaussian kernel density estimator in 3.9 evaluated at ∗ h as the ﬁnal estimator of f, and ∗ h2 γl−1zn1 as the bandwidth for the optimal estimation of ∥f′′∥2. 4 Deliver the Gaussian kernel density estimator in 3.9 evaluated at ∗ h as the ﬁnal estimator of f, and ∗ h2 γl−1zn1 as the bandwidth for the optimal estimation of ∥f′′∥2. It has been proved that the recommending setting for l is 5. It has been proved that the recommending setting for l is 5. 3.2.2. Kernel Density Estimation via Diﬀusion 14 Mathematical Problems in Engineering Mathematical Problems in Engineering Mathematical Problems in Engineering Mathematical Problems in Engineering The above section explains how to estimate the bandwidth ∗h of the Gaussian kernel density estimator. Now, the algorithm that estimates the bandwidth √ h∗of the diﬀusion estimator will be introduced. Assuming that f is as many times continuously diﬀerentiable as needed, then it has been proved that the square of the asymptotically optimal bandwidth is h∗ Ef σ−1x 2N√π Lf 2 2/5 . 3.28 3.28 Computation of h∗in 3.28 requires an estimate of ∥Lf∥2 and Efσ−1x. The latter one can be estimated via the unbiased estimator 1/N N i1 σ−1Xi. The identity ∥Lf∥2 EfL∗Lfx suggests two possible estimators. The ﬁrst one is EfL∗Lfx : 1 N N i1 N j1 L∗Lκ x, Xi; h2\|xXj. 3.29 3.29 The second one is Lf 2 : 1 N N i1 N j1 L∗Lκ x, Xi; 2h2\|xXj. 3.30 3.30 Just like the way that ∗h2 is derived for the Gaussian kernel density estimator, h∗ 2 is selected to make both estimators EfL∗Lfx and ∥ Lf∥ 2 have the same asymptotic mean square error: Just like the way that ∗h2 is derived for the Gaussian kernel density estimator, h∗ 2 is selected to make both estimators EfL∗Lfx and ∥ Lf∥ 2 have the same asymptotic mean square error: h∗ 2 8 √ 2 24 −3 √ 2Ef σ−1X 8√πNEf L∗L2fX 2/7 . 3.31 3.31 Note that h∗ 2 has the same rate of convergence to 0 as ∗h2. In fact, since the Gaussian kernel density estimator is a special case of the diﬀusion estimator when px ax 1, the plug- in estimator equation 3.30 for the estimation of ∥Lf∥2 reduces to the plug-in estimator for the estimation of 1/4∥f′′∥2. In addition, the h∗ 2 in 3.31 and ∗h2 are identical when px ax 1. Thus, the bandwidth for the diﬀusion estimator given in 3.16 can be selected by using the following algorithm: 1 Given the data X1,. . .,XN, run the Improved Sheather-Jones algorithm to obtain the Gaussian kernel density estimator deﬁned in 3.9 evaluated at ∗ h and the optimal bandwidth ∗ h2 for the estimation of ∥f′′∥2. This is the pilot estimation step. 3.2.2. Kernel Density Estimation via Diﬀusion 2 Let px be the Gaussian kernel estimator from above step, and let ax pαx for some α ∈0, 1. 2 Let px be the Gaussian kernel estimator from above step, and let ax pαx for some α ∈0, 1. 3 Estimate ∥Lf∥2 via the plug-in estimator given in 3.30 using h∗ 2 ∗ h2 15 Mathematical Problems in Engineering Yes No Set l = 5, z0 = ε, n = 0, where ε is machine precision zn+1 = ξγ[l](zn) n = n + 1 \|zn+1 −zn\| < ε ∗ꉱh = zn+1 h =∗ꉱh, set = γ[l−1](zn+1) Estimate 㐙Lf㐙2 via (3.30) using ꉱh∗ 2 = ∗ꉱh2 ∗ꉱh2 Calculate ꉱh∗using (3.28) Calculate ﬁnal density estimation p(x) = ꉱf, a(x) = pα(x) for some α∈[0,1] Calculate ꉱf using (3.9) with result using ) with h = ꉱh∗ (3.16 Figure 3: Flow chart of kernel density estimation via diﬀusion. Set l = 5, z0 = ε, n = 0, where ε is machine precision \|zn+1 −zn\| < ε Figure 3: Flow chart of kernel density estimation via diﬀusion. 4 Substitute the estimate of ∥Lf∥2 into 3.28 to obtain an estimate for h∗. 5 Deliver the diﬀusion estimator in 3.16 evaluated at h∗as the ﬁnal density estimate. The ﬂow chart of the entire bandwidth selection algorithm was shown in Figure 3. 16 Mathematical Problems in Engineering Damage 600 mm 600 mm 200 mm 200 mm 200 mm 200 mm 100 mm 100 mm s3 s2 s1 s8 s7 s6 s4 s5 Figure 4: Schematic of numerical simulation mode. Figure 4: Schematic of numerical simulation mode. 5.1. The Characteristics of Noise-Induced Error in ToF It can be expected in theory that the nonparametric estimation methods should have a better performance than parametric estimation method when deal with the distribution without a priori knowledge about its type. The advantage of kernel density estimation method will be demonstrated in this paper by estimating fTmt of s4-8. In statistic, the performance of density estimation methods is usually veriﬁed through comparing the estimation results with the bona ﬁde p.d.f of some well-known datasets. That is, in order to show the accuracy of estimation results, one needs to know the real p.d.f. of the distribution to be estimated. It is diﬃcult to give the analytical expression of fTmt about ToF measured by threshold method. However, partial understanding about the characteristics of noise-induced error in ToF still can be obtained by analyzing the process of threshold method. That will be helpful to prove the advantage of nonparametric estimation methods in ToF method. ToF is given by comparing the arriving time of incident waves and damage-scattered waves. Since the incident waves is strong, the errors in arriving time of incident waves can be neglected. Without loss of generality, the errors in ToF was considered to be caused entirely by the errors in the arriving time of damage-scattered waves. As mentioned in Section 2.1, the existence of wave packet is determined by whether the amplitude of signal is bigger than the threshold value. Once a wave packet is detected, the arriving time of entire wave packet is given by the time of recorded peaks. The process of threshold method suggests there are two kinds of noise-induced errors in ToF: Tm T ε1 ε2, 5.1 5.1 Tm T ε1 ε2, where ε1 denotes the variance in the arriving time of single peak, ε2 denotes the error caused by misidentiﬁcation of peaks. While ε1 is easy to understand, ε2 is relatively complex. The signal received by s4-8 which shown in Figure 5 is taken as example to explain the existence of ε2. Noise not only can change the time of peaks, but also can change the relative magnitude relationship of peaks. That means the sequence of peaks on its magnitude may be changed by noise. If there were no noise and the arriving time was measured by recording the strongest peak, then the second peak of the damage-scattered waves shown in Figure 5 should be recorded. 4. Numerical Simulation captured by the four sensor pairs repeated 30 times. That is, there are 30 ToF results for each sensor pair under each level of noise. 4. Numerical Simulation Feasibility of using the kernel density estimation method to estimate the p.d.f. of experiment results was demonstrated in a thin plate structure via ﬁnite-element FE simulation. Eight PZT wafers were surface installed at an aluminium plate. The aluminium plate was 600 mm × 600 mm × 1.5 mm in size, supported with all its four edges. The elastic modulus, poission’s ration, and density of the aluminium are 71e9GPa, 0.35, and 2711 Kg/m3, respectively. The thin plate was three dimensionally modeled using eight-node brick solid elements. To ensure simulation precision, the largest dimension of FE elements was less than 1 mm and the plate was divided into multilayer in thickness, guaranteeing that at least ten elements were allocated per wavelength of the incident diagnostic wave, which has been demonstrated suﬃciently to portray the characteristics of elastic waves in the thin plate 19. A through- thickness hole of 16 mm in diameter was assumed in the plate, 200 mm and 200 mm away from the left and low edges of the plate, respectively Figure 4. The S0 mode of Lamb waves was used to detect damage. Five-cycle Hanning window-modulated sinusoid tone bursts at a central frequency of 300 kHz were activated as the incident diagnostic wave signal. The speed of S0 mode is 5159.5 m/s in this simulation. Gaussian noise is statistical noise that has its probability density function equal to that of the normal distribution, which is also known as the Gaussian distribution. A special case is white Gaussian noise, in which the values at any pairs of times are statistically independent and uncorrelated. It is well known that noise comes from many natural sources is Gaussian noise. Therefore, in order to simulate the environment noise, three signal-to-noise ration SNR levels 20 dB, 30 dB, and 40 dB of white Gaussian noise were intentionally added into the numerical simulated Lamb waves signals. In numerical model, four sensor pairs are used to locate the damage. The sensor pairs are s2-6 formed by sensor 2 and sensor 6; s4-8 by sensor 4 and 8; s3-7 by sensor 3 and 7; s3- 5 by sensor 3 and 5. The process of adding three levels white Gaussian noise in the signals Mathematical Problems in Engineering 17 captured by the four sensor pairs repeated 30 times. That is, there are 30 ToF results for each sensor pair under each level of noise. 5.2. Density Estimation Results Parametric estimation method, the kernel density estimation based on the Gaussian approximation, and the adaptive kernel density estimation via diﬀusion were used to estimate fTmt. The sample data is ToF measured by s4-8 with three levels noise. The estimation results for the signal with 40 dB SNR noise was shown in Figure 6. The symbol “” in Figure 6 and the following Figures 7, 8, and 9 were used to give an intuitive understanding about the distribution of samples. Each “” represented a sample. It could be seen that samples were distributed around the two values. Most of the samples 26 samples of total 30 samples were distributed in the range from 1.1e−5 second to 1.15e−5 second. 4 samples were distributed in the range from 0.82e−5 second to 0.87e−5 second. The p.d.f. given by the kernel density estimation based on the Gaussian approximation and the adaptive kernel density estimation via diﬀusion was the functions with two peaks. The p.d.f. given by parametric estimation method was undeniably a normal density function. Based on the conclusion drawn in the above section about the characteristics of noise-induced errors in ToF, the distribution of samples could be easily understood. Because the noise was weak in this case, most of the samples, which were only aﬀected by ε1, were distributed around the analytic value of ToF 1.117e −5 second. The other 4 samples which were relatively far from the analytic value were aﬀected by both ε1 and ε2. Therefore, it could be learnt that two kinds of kernel density estimation make correct estimating about the p.d.f. of Tm. Because the assumption about the type of distribution to be estimated was incorrect, parametric estimation method was very misleading in this case. The fact that only 4 samples were aﬀected by both ε1 and ε2 in this case could be utilized to learn the characteristic of ε1. Since these samples could be easily distinguished from the samples which were only aﬀected by ε1, these samples could be excluded from the data set. Then, the density function was estimated with the reﬁned dataset. The results were shown in Figure 7. It could be seen that the results of two kinds kernel density estimation methods were similar to normal distribution. Lilliefors test was adopted to check whether the reﬁned samples came from a normal distribution. 5.1. The Characteristics of Noise-Induced Error in ToF However, the strongest peak may change to other peaks, such as the third or the fourth peak, in noise-contaminated signals. The same problem exists in the method of recording several peaks. For example, if there is no noise and the arriving time is measured as the average of four peaks. Then, the ﬁrst four peaks the second, the third, the fourth, and the ﬁfth in this case should be recorded. However, the ﬁrst peak in noise-contaminated signals is likely to become stronger than the ﬁfth peak. That leads to the error ε2 in ToF. It is obvious that ε2 is larger than ε1, but it appears only in strong noise environment. 18 Mathematical Problems in Engineering 0 0.5 1 1.5 Magnitude Incident waves Damage scattered waves Boundary scattered waves Time (s) −2.5 −2 −1.5 −1 −0.5 0 0.5 1 1.5 2 2.5 ×10−7 ×10−4 Figure 5: The signals of Lamb waves received by sensor 8. Figure 5: The signals of Lamb waves received by sensor 8. 5.2. Density Estimation Results In statistics, the Lilliefors test, named after Hubert Lilliefors, was an adaptation of the Kolmogorov-Smirnov test 38. It was used to test the null hypothesis that data 19 Mathematical Problems in Engineering Mathematical Problems in Engineering 0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2 0 1 2 3 4 5 6 Time of ﬂight (s) Probability density KDE via diffusion KDE based on Gaussian approximation Parametric estimation method ×10−5 ×106 Figure 6: p.d.f. estimated results for samples from s4-8 with 40 dB noise. Figure 6: p.d.f. estimated results for samples from s4-8 with 40 dB noise. 1.095 1.1 1.105 1.11 1.115 1.12 1.125 1.13 0 1 2 3 4 5 6 7 8 9 Time of ﬂight (s) Probability density ×106 ×10−5 KDE via diffusion KDE based on Gaussian approximation Parametric estimation method Figure 7: p.d.f. estimated results for reﬁned samples from s4-8 with 40 dB noise. Figure 7: p.d.f. estimated results for reﬁned samples from s4-8 with 40 dB noise. came from a normally distributed population, when the null hypothesis did not specify which normal distribution; that is, it did not specify the expected value and variance of the distribution. came from a normally distributed population, when the null hypothesis did not specify which normal distribution; that is, it did not specify the expected value and variance of the distribution. The calculated value from the Lilliefors test was 0.1373, which was less than the critical value 0.1699 corresponding to 5% signiﬁcance level. The null hypothesis that the reﬁned data came from a normally distributed population was accepted. It explained why the empirical formula given in the previous work was a normal distribution type and why the damage detection results based on the empirical formula was satisﬁed. Since the noise in previous 20 Mathematical Problems in Engineering 0.9 1 1.1 1.2 1.3 1.4 1.5 0 0.5 1 1.5 2 2.5 3 Time of ﬂight (s) Probability density KDE via diffusion KDE based on Gaussian approximation Parametric estimation method ×106 ×10−5 −0.5 Figure 8: p.d.f. estimated results for samples from s4-8 with 30 dB noise. 0.9 1 1.1 1.2 1.3 1.4 1.5 0 0.5 1 1.5 2 2.5 3 Time of ﬂight (s) Probability density KDE via diffusion KDE based on Gaussian approximation Parametric estimation method ×106 ×10−5 −0.5 Figure 8: p.d.f. estimated results for samples from s4-8 with 30 dB noise. 0.9 1 1.1 1.2 1.3 1.4 1.5 0 0.5 1 1.5 2 2.5 3 Time of ﬂight (s) Probability density KDE via diffusion KDE based on Gaussian approximation Parametric estimation method ×106 ×10−5 −0.5 Figure 8: p.d.f. Mathematical Problems in Engineering estimated results for samples from s4-8 with 30 dB noise. 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0 1 2 3 4 5 6 7 KDE via diffusion KDE based on Gaussian approximation Parametric estimation method Time of ﬂight (s) Probability density ×105 ×10−5 Figure 9: p.d.f. estimated results for samples from s4-8 with 20 dB noise. Figure 8: p.d.f. estimated results for samples from s4-8 with 30 dB noise. 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0 1 2 3 4 5 6 7 KDE via diffusion KDE based on Gaussian approximation Parametric estimation method Time of ﬂight (s) Probability density ×105 ×10−5 Figure 9: p.d.f. estimated results for samples from s4-8 with 20 dB noise. Figure 9: p.d.f. estimated results for samples from s4-8 with 20 dB noise. work 12 was weak and the Tm data was only aﬀected by ε1, its distribution was actually normal distribution. The estimation results for the signals with 30 dB SNR noise were shown in Figure 8. It could be seen that as in the case of 20 dB SNR noise, parametric estimation method failed to give correct estimation. g The estimation results for the signals with 20 dB SNR noise were shown in Figure 9. It could be seen that, with the increase of noise level, the kernel density estimation based on the Gaussian approximation, which was traditional kernel density estimation, failed to give correct estimation. Only the novel and completely data-driven method, the kernel density estimation via diﬀusion-, could give correct estimation. Mathematical Problems in Engineering 21 0.1 0.15 0.2 0.25 0.1 0.12 0.14 0.16 0.18 0.2 0.22 0.24 0.26 0.28 0 2 4 6 8 10 12 14 16 18 (m) (m) ×10−3 Probability of damage Figure 10: Damage localization result based on parametric estimate method partial view. Figure 10: Damage localization result based on parametric estimate method partial view. 5.3. Damage Detection Results The damage localization under 20 dB noise environment was selected as the example to show that an accurate estimation was important for the localization result. The p.d.f. estimation results given by three kinds of density estimation methods introduced in Section 2 were used to calculate the location of damage. The results were shown in Figures 10, 11, and 12. It could be seen that the locating process which employed the kernel density estimation via diﬀusion has the most accurate localization result. This indicated that the an accurate estimation could ensure an better localization result. 6. Conclusion The characteristics of noise-induced error in ToF data measured by using threshold method were analyzed. The empirical formula method and the parametric estimation method presented in existing work had the same assumption that the experimental data came from a normal distribution. This assumption had been veriﬁed by real experiments and numerical simulation. The results in this paper revealed that the type of distribution of ToF data was related to the noise level. The empirical formula method and the parametric estimation method were developed in laboratory environment where the noise was weak. It had also been proved in this paper that the ToF data measured from high SNR signal SNR > 40 dB were distributed normally. Therefore, the density estimation method with the normality assumption presented in existing work can work well in laboratory environment. However, the signals of ﬁeld experiment usually contained much more strong noise. The results in this paper showed that even for the signal with 40 dB SNR, the distribution of measured ToF data were not normal distribution. In this case, nonparametric estimation method must be emplyed to estimate the p.d.f. correctly. Further, investigating about the signals with 30 dB and 20 dB noise showed that, with the increasing noise, only the kernel 22 Mathematical Problems in Engineering 0.1 0.15 0.2 0.25 0.3 0.1 0.12 0.14 0.16 0.18 0.2 0.22 0.24 0.26 0.28 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 (m) (m) Probability of damage Figure 11: Damage localization result based on kernel density estimation with Gaussian approximation partial view. 22 Mathematical Problems in Engineering Figure 11: Damage localization result based on kernel density estimation with Gaussian approximation partial view. 0.1 0.15 0.2 0.25 0.3 0.1 0.12 0.14 0.16 0.18 0.2 0.22 0.24 0.26 0.28 0 0.01 0.02 0.03 0.04 0.05 0.06 (m) (m) Probability of damage Figure 12: Damage localization result based on kernel density estimation via diﬀusion partial view. Figure 12: Damage localization result based on kernel density estimation via diﬀusion partial view density estimation via diﬀusion, which is purely data driven, can give a satisﬁed estimating result. The damage localization under 20 dB noise environment had been carried out. Parametric estimation method with the normality assumption, the kernel density estimation based on the Gaussian approximation and the kernel density estimation via diﬀusion were adopted to estimate the p.d.f. of measured data. Three diﬀerent p.d.f. 6. Conclusion were obtained by employing the above-motioned three kinds of density estimation methods. By using each p.d.f, a damage location result can be calculated. Through comparing the three results of damage location, it can be seen that an accurate estimation of p.d.f. has a direct eﬀect on Mathematical Problems in Engineering 23 the accuracy of the results. Applying kernel density estimation in Lamb wave-based damage detection was necessary. the accuracy of the results. Applying kernel density estimation in Lamb wave-based damage detection was necessary. The noise studied in this paper was the white Gaussian noise. The noise in the real ﬁeld experiment was much more complex. Further study was needed to reveal the characteristic of errors in ToF data caused by noise in ﬁeld experiment. However, the complex nature of noise in ﬁeld experiment could not be a trouble for the application of kernel density estimation method, instead, it could be a reason to apply this method. It had been proved that when deal with simple noise, the kernel density estimation method introduced in this paper performed better, in comparison with empirical methods. Since the kernel density estimation method did not rely on any assumption about the distribution to be estimated, it could be expected that the kernel density estimation method could demonstrate a greater advantage in a complex noise environment. Acknowledgments This work was ﬁnancially supported by National Natural Science Foundation of China under grant no. 50905141, the Program for New Century Excellent Talents in University of China, and the NPU Foundation for Fundamental Research under grant no. NPU-FFR-JC20110258. This work was ﬁnancially supported by National Natural Science Foundation of China under grant no. 50905141, the Program for New Century Excellent Talents in University of China, and the NPU Foundation for Fundamental Research under grant no. NPU-FFR-JC20110258. References 1 Y. J. Yan, L. Cheng, Z. Y. Wu, and L. H. Yam, “Development in vibration-based structural damage detection technique,” Mechanical Systems and Signal Processing, vol. 21, no. 5, pp. 2198–2211, 2007. 1 Y. J. Yan, L. Cheng, Z. Y. Wu, and L. H. Yam, “Development in vibration-based structural damage detection technique,” Mechanical Systems and Signal Processing, vol. 21, no. 5, pp. 2198–2211, 2007. 2 Q. Yang, “A new damage identiﬁcation method based on structural ﬂexibility disassembly,” Journal f Vib i d C l l 1 1 1 11 1 Y. J. Yan, L. Cheng, Z. Y. Wu, and L. H. Yam, Development in vibration based structural damage detection technique,” Mechanical Systems and Signal Processing, vol. 21, no. 5, pp. 2198–2211, 2007. 2 Q. Yang, “A new damage identiﬁcation method based on structural ﬂexibility disassembly,” Journal of Vibration and Control, vol. 17, no. 7, pp. 1000–1008, 2011. detection technique, Mechanical Systems and Signal Processing, vol. 21, no. 5, pp. 2198 2211, 2007. 2 Q. Yang, “A new damage identiﬁcation method based on structural ﬂexibility disassembly,” Journal of Vibration and Control, vol. 17, no. 7, pp. 1000–1008, 2011. f pp 3 L. Yu, L. Cheng, L. H. Yam, Y. J. Yan, and J. S. Jiang, “Online damage detection for laminated composite shells partially ﬁlled with ﬂuid,” Composite Structures, vol. 80, no. 3, pp. 334–342, 2007. 4 X. Lu, M. Lu, L. M. Zhou et al., “Evaluation of welding damage in welded tubular steel structures using guided waves and a probability-based imaging approach,” Smart Materials and Structures, vol. 20, no. 1, Article ID 015018, 2011. 5 J. Chen, Z. Su, and L. Cheng, “Identiﬁcation of corrosion damage in submerged structures using fundamental anti-symmetric Lamb waves,” Smart Materials and Structures, vol. 19, no. 1, Article ID 015004, 2010. 6 G. Zumpano and M. Meo, “A new nonlinear elastic time reversal acoustic method for the identiﬁcation and localisation of stress corrosion cracking in welded plate-like structures—a simulation study,” International Journal of Solids and Structures, vol. 44, no. 11-12, pp. 3666–3684, 2007. y f pp 7 Z. Su, L. Ye, and Y. Lu, “Guided Lamb waves for identiﬁcation of damage in composite structures: a review,” Journal of Sound and Vibration, vol. 295, no. 3–5, pp. 753–780, 2006. 8 C. R. Farrar, S. W. Doebling, and D. A. Nix, “Vibration-based structural damage identiﬁcation,” Philosophical Transactions of the Royal Society A, vol. 359, no. 1778, pp. Mathematical Problems in Engineering 24 14 S. G. Pierce, B. Culshaw, G. Manson, K. Worden, and W. J. Staszewski, “Application of ultrasonic Lamb wave techniques to the evaluation of advanced composite structures,” in Smart Structures and Materials 2000—Sensory Phenomena and Measurement Instrumentation for Smart Structures and Materials, vol. 3986 of Proceedings of SPIE, pp. 93–103, March 2000. y 3986 of Proceedings of SPIE, pp. 93–103, March 2000. g f pp 15 D. N. Alleyne and P. Cawley, “The interaction of Lamb waves with defects,” IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 39, no. 3, pp. 381–397, 1992. 16 S. Grondel, C. Paget, C. Delebarre, J. Assaad, and K. Levin, “Design of optimal conﬁguration for generating A0 Lamb mode in a composite plate using piezoceramic transducers,” Journal of the Acoustical Society of America, vol. 112, no. 1, pp. 84–90, 2002. y f pp 17 Z. Su, L. Ye, and X. Bu, “A damage identiﬁcation technique for CF/EP composite laminates using distributed piezoelectric transducers,” Composite Structures, vol. 57, no. 1–4, pp. 465–471, 2002. 18 M. Lemistre and D. L. Balageas, “Structural health monitoring system based on diﬀracted Lamb wave analysis by multiresolution processing,” Smart Materials and Structures, vol. 10, no. 3, pp. 504–511, 2001. 19 Y. Lu, L. Ye, Z. Su, and C. Yang, “Quantitative assessment of through-thickness crack size based on Lamb wave scattering in aluminium plates,” NDT and E International, vol. 41, no. 1, pp. 59–68, 2008. 20 V. Giurgiutiu, J. Bao, and W. Zhao, “Piezoelectric wafer active sensor embedded ultrasonics in beams and plates,” Experimental Mechanics, vol. 43, no. 4, pp. 428–449, 2003. p p pp 21 L. Yu, L. Cheng, and Z. Su, “Correlative sensor array and its applications to identiﬁcation of dama in plate-like structures,” Structural Control and Health Monitoring. In press. p g p 22 D. A. Hutchins, D. P. Jansen, and C. Edwards, “Lamb-wave tomography using non-contact transduction,” Ultrasonics, vol. 31, no. 2, pp. 97–103, 1993. pp 23 D. Wang, L. Ye, Z. Su, Y. Lu, F. Li, and G. Meng, “Probabilistic damage identiﬁcation based on correlation analysis using guided wave signals in aluminum plates,” Structural Health Monitoring, vol. 9, no. 2, pp. 133–144, 2010. pp 24 G. Peng and S. Yuan, “Damage localization on two-dimensional structure based on wavelet transform and active lamb wave-based method,” Materials Science Forum, vol. 475-479, no. 3, pp. 2119–2122, 2005. 25 L. De Marchi, A. Marzani, N. Speciale, and E. References 131–149, 2001. p f y y pp 9 H. Sohn, C. R. Farrar, F. M. Hemez, and J. J. Czarnecki, “A review of structural health monitoring literature: 1996–2001,” in Proceedings of the 3rd World Conference on Structural Control, Como, Italy, April 2002. p 10 Z. Su, L. Cheng, X. Wang, L. Yu, and C. Zhou, “Predicting delamination of composite laminates using an imaging approach,” Smart Materials and Structures, vol. 18, no. 7, Article ID 074002, 2009. g g pp . D. Niri and S. Salamone, “A probabilistic framework for acoustic emission source localization in late-like structures,” Smart Materials and Structures, vol. 21, no. 3, Article ID 035009, 2012. 11 E. D. Niri and S. Salamone, “A probabilistic framework for acoustic emission source lo plate-like structures,” Smart Materials and Structures, vol. 21, no. 3, Article ID 035009, 201 p plate-like structures,” Smart Materials and Structures, vol. 21, no. 3, Article ID 035009, 2012. 12 L. Yu, Z. Su, and X. Liu, “The deﬁnition and measurement of the probability density function in lamb wave damage detection based on data fusion,” in Proceedings of the14th Asia Paciﬁc Vibration Conference, Kowloon, Hong Kong, 2011. g g 13 A. Raghavan and C. E. S. Cesnik, “Review of guided-wave structural health monitoring,” Shock a Vibration Digest, vol. 39, no. 2, pp. 91–114, 2007. Mathematical Problems in Engineering Viola, “A passive monitoring technique based on dispersion compensation to locate impacts in plate-like structures,” Smart Materials and Structures, vol. 20, no. 3, Article ID 035021, 2011. 26 A. Perelli, L. De Marchi, A. Marzani, and N. Speciale, “Acoustic emissions localization in plates with dispersion and reverberations by using sparse PZT sensors in passive mode,” Smart Materials and Structures, vol. 21, no. 2, Article ID 025010, 2012. 27 X. Zhao, H. Gao, G. Zhang et al., “Active health monitoring of an aircraft wing with embedded piezoelectric sensor/actuator network: I. Defect detection, localization and growth monitoring,” Smart Materials and Structures, vol. 16, no. 4, pp. 1208–1217, 2007. pp 28 Z. Su, X. Wang, L. Cheng, L. Yu, and Z. Chen, “On selection of data fusion schemes for structu damage evaluation,” Structural Health Monitoring, vol. 8, no. 3, pp. 223–241, 2009. 29 M. Rosenblatt, “Remarks on some nonparametric estimates of a density function,” Annals of Mathematical Statistics, vol. 27, no. 3, pp. 832–837, 1956. pp 30 E. Parzen, “On estimation of a probability density function and mode,” Annals of Mathematical Statistics, vol. 33, no. 3, pp. 1065–1076, 1962. pp 31 D. W. Scott, “On optimal and data-based histograms,” Biometrika, vol. 66, no. 3, pp. 605–610, 1979. p g pp 32 B. W. Silverman, Density Estimation for Statistics and Data Analysis, Monographs on Statistics and Applied Probability, Chapman & Hall, CRC Press, London, UK, 1998. pp y, p , , , , 33 R. L. Graham, D. E. Knuth, and O. Patashnik, Concrete Mathematics, Addison-Wesley, Reading, Mass, USA, 2nd edition, 1994. 34 Z. I. Botev, J. F. Grotowski, and D. P. Kroese, “Kernel density estimation via diﬀusion,” Annals of Statistics, vol. 38, no. 5, pp. 2916–2957, 2010. pp 35 P. Chaudhuri and J. S. Marron, “Scale space view of curve estimation,” Annals of Statistics, vol. 28, n 2, pp. 408–428, 2000. pp 36 S. Larsson and V. Thom´ee, Partial Diﬀerential Equations with Numerical Methods, vol. 45 of Texts in Applied Mathematics, Springer, Berlin, Germany, 2003. pp , p g , , y, 37 M. P. Wand and M. C. Jones, Kernel Smoothing, vol. 60 of Monographs on Statistics and Applied Probability, Chapman & Hall, London, UK, 1995. 38 H. W. Lilliefors, “On the Kolmogorov-Smirnov test for the exponential distribution with me unknown,” Journal of the American Statistical Association, vol. 64, no. 325, pp. 387–389, 1969. Mathematical Problems in Engineering Submit your manuscripts at http://www.hindawi.com Operations Research Advances in Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Mathematical Problems in Engineering Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Abstract and Applied Analysis ISRN Applied Mathematics Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 International Journal of Combinatorics Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Journal of Function Spaces and Applications International Journal of Mathematics and Mathematical Sciences Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 ISRN Geometry Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Discrete Dynamics in Nature and Society Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Advances in Mathematical Physics ISRN Algebra Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Probability and Statistics Journal of Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 ISRN Mathematical Analysis Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Journal of Applied Mathematics Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Advances in Decision Sciences Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Stochastic Analysis International Journal of Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 The Scientific World Journal Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 ISRN Discrete Mathematics Hindawi Publishing Corporation http://www.hindawi.com Differential Equations International Journal of Volume 2013 Hindawi Publishing Corporation Hindawi Publishing Corporation The Scientific World Journal Hindawi Publishing Corporation h // hi d i V l 2013 Mathematical Problems in Engineering Submit your manuscripts at http://www.hindawi.com Submit your manuscripts at http://www.hindawi.com Operations Research Advances in Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 ISRN Applied Mathematics Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 International Journal of Mathematics and Mathematical Sciences Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 ISRN Geometry Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 ISRN Algebra Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Probability and Statistics Journal of Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 ISRN Mathematical Analysis Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 Stochastic Analysis International Journal of Hindawi Publishing Corporation http://www.hindawi.com Volume 2013 ISRN Discrete Mathematics Hindawi Publishing Corporation http://www.hindawi.com Differential Equations International Journal of Volume 2013 Operations Research Advances in Probability and Statistics Journal of ISRN Geometry Hindawi Publishing Corporation http://www.hindawi.com Volume 2013
https://openalex.org/W2898187310	https://www.frontiersin.org/articles/10.3389/fnins.2018.00760/pdf	English	null	How Is the Neural Response to the Design of Experience Goods Related to Personalized Preference? An Implicit View	Frontiers in neuroscience	2,018	cc-by	7,753	Edited by: Ioan Opris Edited by: Ioan Opris, University of Miami, United States Reviewed by: Xuegang Cui, Beijing Normal University, China Liang M. A., Tsinghua University, China Reviewed by: Xuegang Cui, Beijing Normal University, China Liang M. A., Tsinghua University, China Correspondence: Qingguo Ma maqingguo3669@zju.edu.cn Specialty section: This article was submitted to Neural Technology, a section of the journal Frontiers in Neuroscience Received: 29 June 2018 Accepted: 02 October 2018 Published: 26 October 2018 Keywords: event-related potentials, experience goods, late positive potential, P200, personalized product designs How Is the Neural Response to the Design of Experience Goods Related to Personalized Preference? An Implicit View Yongbin Ma1,2,3, Jia Jin1,2, Wenjun Yu1,2, Wuke Zhang1,2, Zhijiang Xu4 and Qingguo Ma1,2,5 1 Business School, Ningbo University, Ningbo, China, 2 Academy of Neuroeconomics and Neuromanagement, Ningbo University, Ningbo, China, 3 Morden Management Research Centre, Ningbo University, Ningbo, China, 4 College of Information Engineering, Zhejiang University of Technology, Hangzhou, China, 5 Institute of Neuromanagement Science, Zhejiang University of Technology, Hangzhou, China 1 Business School, Ningbo University, Ningbo, China, 2 Academy of Neuroeconomics and Neuromanagement, Ningbo University, Ningbo, China, 3 Morden Management Research Centre, Ningbo University, Ningbo, China, 4 College of Information Engineering, Zhejiang University of Technology, Hangzhou, China, 5 Institute of Neuromanagement Science, Zhejiang University of Technology, Hangzhou, China Understanding the process by which consumers evaluate the designs of experience goods is critical for ﬁrms designing and delivering experience products. As the implicit process involved in this evaluation, and given the possible social desirability bias inherent to traditional methods of product design evaluation in certain conditions, neuroscientiﬁc methods are preferred to gain insight into the neural basis of consumers’ evaluation of experience good designs. We here used event-related potentials (ERPs) and a revised go/no-go paradigm to investigate consumers’ neural responses to experience good designs. Personalized product designs and neutral landscape pictures were randomly presented to 20 student participants; they were asked to view these product designs without making any decisions. The paired t-test and repeated-measures analysis of correlation showed that the P200 and late positive potential (LPP) elicited by the most- preferred experience good designs were signiﬁcantly higher than that elicited by least- preferred designs, and the two ERP components were positively correlated with the personalized rating scores. Thus, P200 and LPP might be the early and late indices of consumers’ evaluation of experience good designs, respectively, and may facilitate an understanding of the temporal course of this evaluation. Furthermore, these two ERP components can be used to identify consumers’ preferences toward experience good designs. In addition, given the use of personalized experimental stimuli, these ﬁndings may help to explain why customized products are preferred by consumers. ORIGINAL RESEARCH published: 26 October 2018 doi: 10.3389/fnins.2018.00760 Citation: Studies have explored the neural processes and brain regions involved in the evaluation of esthetic objectives by means of EPRs, for example, arts (Augustin et al., 2011) and faces (Chatterjee et al., 2009). For product designs, in addition to the esthetic aspect, such as arts and music, the symbolic aspect is also important and is signiﬁcantly related to consumer behavior (Homburg et al., 2015). ) Previous studies have also adopted the ERP technology to study consumers’ preference using product images and have indicated that some ERP components can eﬀectively predict consumer preferences and choices (Junghöfer et al., 2010; Pozharliev et al., 2015; Telpaz et al., 2015). Nevertheless, these studies did not considered diﬀerences between diﬀerent types of products, except for the study by Pozharliev et al. (2015), which explicitly focused on luxury goods. In fact, consumers consider diﬀerent factors when purchasing diﬀerent types of products. For example, for search goods, the objective attributes (e.g., price and functions) are inﬂuential in the decision-making process (Huang et al., 2009). The EEG signals collected when consumers view the product images (subjective attributes) of search goods cannot capture the key information of consumers’ decision-making process, and the predictive relationship of EEG signals and product preferences is also not accurate. For experience goods, the subjective attributes (e.g., product designs) are important (Hoch and Ha, 1986). EEG signals collected when consumers view the product images can capture the process of consumers’ evaluation of product (designs) more accurately. In addition, in these studies, all the subjects were assigned to view the same product design images, and the subjective and penalized demand of consumers for product designs was not appropriately considered. For experience goods, for which subjective attributes are important (Hoch and Ha, 1986), EEG signals collected when viewing the same product design pictures as previous studies may not capture the individualized diﬀerences in consumers’ evaluation of product designs. Thus, by using personalized experimental stimuli and ERP technology, this study focused on the process of evaluation of experience product designs. g g The evaluation of product designs is an esthetic evaluation process, which has been found to be associated with consumers’ cognitive and emotional responses (Silvia and Warburton, 2006), and is inﬂuenced by the existing knowledge and experience of consumers. This makes the consumers’ evaluation of product designs more subjective and personal. Citation: Ma Y, Jin J, Yu W, Zhang W, Xu Z and Ma Q (2018) How Is the Neural Response to the Design of Experience Goods Related to Personalized Preference? An Implicit View. Front. Neurosci. 12:760. doi: 10.3389/fnins.2018.00760 Design is an inherent component of all products, and has a lasting eﬀect on consumers’ loyalty to brands and their purchasing decisions (Reimann et al., 2010; Homburg et al., 2015). Experience goods, such as clothes, wines, and cosmetics, are products whose utility cannot be ascertained before purchase, because of the lack of full information (Nelson, 1970, 1974). Product design is especially important for such goods, because product design is an important source of utility October 2018 \| Volume 12 \| Article 760 Frontiers in Neuroscience \| www.frontiersin.org Neural Response and Product Designs Ma et al. of experience goods. For example, besides the functional dimension, the esthetic and symbolic dimension of T-shirts are also important parts of T-shirt designs (Homburg et al., 2015), and they are also the key factors inﬂuencing consumers’ experience evaluation of T-shirts. More important, product design is the main source of uncertainty in consumers’ evaluation of experience goods. Compared to search attributes of experience products, such as material and price, consumers’ evaluation of experience attributes of these goods, such as product designs, is more subjective (Huang et al., 2009), and their needs for experience attributes are more personalized (Bloch et al., 2003). Therefore, it is diﬃcult for enterprises to understand consumers’ demand for experiential good designs eﬀectively, and for consumers to use the information provided by ﬁrms to judge the ﬁt between their needs and experience good designs, which will increase the uncertainty of consumers’ evaluation of experiential products. In practice, some T-shirt, mobile phone case, and cake ﬁrms have even given up on understanding of the consumers’ needs for experience product designs, and have outsourced the product design task to consumers through personalized customization (Schreier, 2006). The basic premise of personalized customization is to let consumers choose the design elements that they like, while the ﬁrms complete the production and distribution. Therefore, exploring consumers’ process of evaluating experience product designs can shed light on the consumer needs for experience products and provide important information to enterprises for designing and delivering experience goods. and explore the “common scale” that allows comparison of heterogeneous and individualized behavior (Levy and Glimcher, 2012; Camerer and Yoon, 2015). Frontiers in Neuroscience \| www.frontiersin.org Study Materials Considering the familiarities of the participants and the needs for personalized experimental stimuli, we selected the T-shirts as experimental stimuli. Using the three-dimensional scale reported by Weathers et al. (2007), we measured the perceived product type of T-shirts of 30 students (recruited from the same university as those for the main experiment), the average score of the three dimensions was 4.910 (1 = pure search goods, 7 = pure experience goods), which indicate that T-shirts were more likely to be perceived as experience rather than search goods. In the evaluative categorization stage, more conscious and cognitive appraisals are triggered, which result in a more enduring esthetic judgment and emotional response (Kumar and Garg, 2010; Celaconde et al., 2013). The late positive potential (LPP) is a long-lasting ERP component peaking at around 500–700 ms after the onset of stimuli. An enhanced LPP amplitude represents a reliable, replicable, and time-speciﬁc emotional response to stimuli (Hajcak et al., 2006). The LPP amplitude is positively correlated with arousal level, which implies that pictures causing high emotional arousal (pleasant or unpleasant, rather than neutral) elicit augmented LPP (Schupp et al., 2000). The LPP also indicates the sustained enhanced attention allocation and motivational signiﬁcance of emotional visual stimuli during aﬀective perceptual processing (Bradley et al., 2003). As most-preferred product designs of experience goods will elicit more emotional arousal than less-preferred designs, and individuals reliably allocate more attention to the former (Coates, 2003), we speculated that greater positive-going amplitude of LPP would be observed for the most-preferred experience goods designs than that for the least-preferred designs. Moreover, according to neuropsychological models of attention, P200 and LPP are correlated and jointly reﬂect diﬀerent attentional processes for the same visual stimulus (Pozharliev et al., 2015). Therefore, we speculated that the two ERP components would both be evoked during the product design evaluation process of experience goods. The current common practice of implementing “personalized T-shirt design” is as follows: ﬁrst, ﬁrms post design elements on their website; then, consumers choose and combine design elements by themselves; ﬁnally, T-shirts are manufactured and derived according to the consumers’ preference. As this is time- consuming and is not easy to implement in the laboratory, we asked the research assistants to select the images of the T-shirt designs of ﬁve major T-shirt brands from their oﬃcial websites on Tmall.com (choosing the new and personalized designs). Study Materials We also used the Baidu search engine to select 40 neutral landscape images for implicit tasks. All the pictures were processed to have a white background, a resolution of 360 × 270 pixels, and were in pdf format. The size of these pictures was 360 mm × 360 mm. There were no explicit brand names, logo, or other explicit signals on the product design pictures. Experimental Procedures and EEG Recording g The experiment was divided into two stages (Figure 1). In the ﬁrst stage, participants were told that these T-shirt designs were randomly combined using the design elements of T-shirt customization companies on Tmall.com, and they were asked to indicate their preference for these T-shirt design pictures. Their preference was measured using a 0–100 horizontal preference scale (100 = completely prefer the design pictures). In order to increase the accuracy of subjective measurement, participants were told that “there are no right or wrong answers in the evaluation of the T-shirt designs; please give your true preference. Your evaluation will only be used for this study, and we will keep your answers strictly conﬁdential.” We did not record the name and other personal information of these participants. The survey was conducted in an independent behavioral laboratory and there was no interference throughout the process. In this study, we analyzed the neural activities related to the design evaluation of experience goods. We speculated that more positive-going ERP amplitude would be observed for the two ERP components, P200 and LPP, for the most-preferred experience good designs than that for the least-preferred designs. To investigate the personal and subjective nature of product design evaluation, we used personalized product designs rather than the same stimulus for all the participants in the experiment, since personalized stimuli may reﬂect individuals’ preferences for product designs more accurately (Pearce et al., 2016). Citation: Therefore, relatively objective measures are needed to capture the process and characteristics of product design evaluation (Luo et al., 2008; Pearce et al., 2016). Additionally, the implicit information processing involved in this esthetic evaluation process of product designs makes the traditional questionnaire method less eﬀective in this study (Wang and Tseng, 2015). For example, questionnaires can be used to elicit consumers’ explicit preference for experience product designs, but it cannot explain the implicit reasons for this preference. In addition, for products that satisfy consumers’ social communication and status-seeking need (Heine and Phan, 2011), such as luxury goods, and for participants who participate in product preference evaluation in order to obtain monetary rewards (Davidson et al., 2002), directly eliciting consumers’ response to product designs using questionnaire may cause social desirability bias. Product design evaluation is correlated with esthetic evaluation (Bloch, 1995; Silvia and Warburton, 2006). Previous studies have posited that esthetic evaluation consists of two distinct stages: early impression formation and a subsequent evaluative categorization stage (Celaconde et al., 2013). During the early impression formation phase, individuals devote more attentional resources to exploring stimuli. The subconscious processes and visual properties of the stimuli are involved in this stage (Celaconde et al., 2004). P200 is a positive-going waveform with a peak latency at about 200 ms after the onset of stimuli, and is related to early automatic and selective attention (Olofsson et al., 2008). P200 can be elicited by aﬀective stimuli, reﬂecting the initial sensory encoding of emotionally signiﬁcant stimuli, and the onset of a persistent positive shift of the ERP waveform (Olofsson et al., 2008). As individuals pay more visual attention to product designs that they ﬁnd attractive (Coates, 2003), we The event-related potential (ERP) technique can directly measure individuals’ perceptual and cognitive processes in response to stimuli with high temporal resolution (Luck et al., 2000), and can help to record the activities that involve social-desirability biases or are otherwise diﬃcult to report (Cerf et al., 2015). It can also help to discover physiological factors that inﬂuence individual behavior and preferences October 2018 \| Volume 12 \| Article 760 2 Neural Response and Product Designs Ma et al. speculated that a greater positive-going amplitude would be observed in P200 for the most-preferred product designs of experience goods than for the least-preferred designs. October 2018 \| Volume 12 \| Article 760 Study Participants g After scoring, participants were invited to attend the second stage of the experiment, which was conducted in a dimly lit and electrically shielded EEG laboratory. The design of this experiment followed a revised go/no-go paradigm, in which the participants were instructed to “go” (press a key) when a landscape picture was presented, and to “no-go” (refrain from pressing a key) when a T-shirt picture was presented. The ERP experiment consisted of three blocks, and each block contained 40 trials. For each participant, there were two groups of 40 personalized T-shirt pictures (most-preferred and least-preferred product designs), and one group of 40 landscape pictures. The two groups of T-shirt pictures were classiﬁed based on the self- rating scores allotted to these pictures in the ﬁrst stage. The 40 The participants were 20 students recruited from Ningbo University, aged 18–26 years of age (mean age = 24 years, SD = 2.247; 53% women). All the participants reported normal or corrected-to-normal vision and had no history of neurological or psychiatric illness. The participants were paid ¥40 as compensation for participation in the study. Four of the participants were removed from the study due to excessive ERP artifacts. This study was approved by the local institutional ethics committee of the Academy of Neuroeconomics and Neuromanagement at Ningbo University. All participants provided written informed consent before the experiment started. October 2018 \| Volume 12 \| Article 760 Frontiers in Neuroscience \| www.frontiersin.org 3 Neural Response and Product Designs Ma et al. FIGURE 1 \| Example trials of the experiment. (A) T-shirt preference rating task. During each trial, after a ﬁxation cross appeared, participants were presented with a picture of a T-shirt, and then they were instructed to give their preference scores for the T-shirt. (B) Landscape picture and product design presentation in the ERP task. In each trial, after a ﬁxation cross appeared, landscape pictures and product designs were presented randomly. Participants were instructed to view these pictures without making any decisions regarding the product designs and to press a key for landscape pictures. FIGURE 1 \| Example trials of the experiment. (A) T-shirt preference rating task. During each trial, after a ﬁxation cross appeared, participants were presented with a picture of a T-shirt, and then they were instructed to give their preference scores for the T-shirt. (B) Landscape picture and product design presentation in the ERP task. Statistical Analysis landscape pictures were discarded at the data analysis stage. All the pictures were presented in a randomized order. Participants were instructed to watch these pictures without making any overt responses, and to minimize head and body movements during this stage. Previous studies have indicated that the modulation of the P200 amplitude to emotional visual stimuli is most pronounced in the posterior scalp areas (Schupp et al., 2004; Pozharliev et al., 2015), and that the LPP is most pronounced in the superior–posterior scalp areas (Sabatinelli et al., 2006; Pozharliev et al., 2015). These two components, elicited by the emotional visual stimuli, are usually reported in both the left and right hemispheres (Dolcos and Cabeza, 2002). Based on these ﬁndings, for P200, we analyzed signals from the following nines electrodes: C1, CZ, C2, CP1, CPZ, CP2, P1, PZ, and P2. As the nine electrode sites in the posterior area had a similar pattern of results, the data from the region were pooled to obtain a representative value. For the LPP, we focused on the following six electrodes: C1, CZ, C2, CP1, CPZ, and CP2, the data from all the six electrode sites in the central and parietal scalp areas were also pooled for the same reason as for P200. The pictures were presented on a 19-in monitor (1280 × 1024 pixels, 60 Hz), which was connected to a 2 GHz Pentium computer. E-prime 2.0 (Psychology Software tools, Pittsburgh, PA, United States) was used for stimulus presentation and data collection. The pictures were positioned at the center of the screen and viewed from a distance of 100 cm, at a visual angle of 6.27–6.271◦. The background of the screen was gray (RGB: 128, 128, 128). Event-related potential data were recorded using 64 Ag/AgCl electrodes mounted on an elastic cap with a Neuroscan Synamp2 Ampliﬁer (Scan 4.3.1, Neurosoft Labs, Inc.). A forehead location was used for grounding, and the reference was the left mastoid. Using left and right mastoid references, data were transferred to the average oﬄine. Vertical electrooculograms were recorded using a pair of electrodes placed above and below the right eye, while horizontal electrooculograms were recorded using another pair of electrodes placed on the right side of the right eye and left side of the left eye. Both vertical and horizontal pairs of electrodes were placed 10 mm from the eyes. Electrooculogram artifacts were corrected oﬄine. Study Participants In each trial, after a ﬁxation cross appeared, landscape pictures and product designs were presented randomly. Participants were instructed to view these pictures without making any decisions regarding the product designs and to press a key for landscape pictures. Statistical Analysis The experiment started with the impedances of the electrodes less than 5 k. Based on previous studies and visual inspection of grand averages of waveforms (Pozharliev et al., 2015), the time windows chosen for P200 and the LPP were 160–210 and 500–700 ms, respectively. For each participant, we examined the P200 and LPP amplitude in the most- and least-preferred conditions and matched these amplitudes with the average preference rating scores for the personalized product designs under both conditions. We examined whether there were signiﬁcant diﬀerences in the ERP amplitude for diﬀerent preference conditions, and how the ERP components induced by product designs were correlated to the corresponding personalized preference rating scores. As product designs that were presented to the participants were nested within the subjects, the pairwise t-test and repeated measures analysis of correlation were used. Epochs were made beginning 200 ms before stimulus onset and continuing for 800 ms after the onset. The EEG was aligned to a 200-ms baseline, and error-of-commission artifacts were corrected using the method proposed by (Semlitsch et al., 1986). Trials with bursts of electromyography activity, peak-to-peak deﬂection exceeding ± 100 µV, and ampliﬁer clipping were excluded. The averaged ERPs were digitally ﬁltered using a low- pass ﬁlter at 30 Hz (24 dB/octave). The EEG recordings for every participant were averaged separately over each recording site for each of the most- and least-preferred product design conditions. The data were further analyzed for the two experimental conditions. Frontiers in Neuroscience \| www.frontiersin.org RESULTS The average scores for T-shirt designs of the ﬁve brands were 52.71, 51.66, 52.24, 53.23, and 50.47, and repeated measures ANOVA revealed that there was no signiﬁcant diﬀerence between October 2018 \| Volume 12 \| Article 760 Frontiers in Neuroscience \| www.frontiersin.org 4 Neural Response and Product Designs Ma et al. FIGURE 2 \| P200 condition effect. P200 waveforms were averaged from the 16 subjects, and we pooled the data from C1, CZ, C2, CP1, CPZ, CP2, P1, PZ, and P2 electrodes. The time window was 160–210 ms for P200. (A) The ERP grand-average waveforms of P200 component in the most-preferred and the least-preferred product design conditions. (B) The bar chart reﬂecting mean amplitude of P200 in the most-preferred and the least-preferred product design conditions. Error bars indicate standard error. (C) Topographic maps of the most-preferred and the least-preferred product design conditions for P200 amplitude. FIGURE 3 \| The LPP condition effect. The LPP waveforms were averaged from the 16 subjects, and we pooled the data from C1, CZ, C2, CP1, CPZ, and CP2 electrodes. The time window was 500–700 ms for the LPP. (A) The ERP grand-average waveforms of the LPP component in the most-preferred and the least-preferred product design conditions. (B) The bar chart reﬂecting mean amplitude of P200 in the most-preferred and the least-preferred product design conditions. Error bars indicate standard error. (C) Topographic maps of the most-preferred and the least-preferred product design conditions for the LPP amplitude. FIGURE 2 \| P200 condition effect. P200 waveforms were averaged from the 16 subjects, and we pooled the data from C1, CZ, C2, CP1, CPZ, CP2, P1, PZ, and P2 electrodes. The time window was 160–210 ms for P200. (A) The ERP grand-average waveforms of P200 component in the most-preferred and the least-preferred product design conditions. (B) The bar chart reﬂecting mean amplitude of P200 in the most-preferred and the least-preferred product design conditions. Error bars indicate standard error. (C) Topographic maps of the most-preferred and the least-preferred product design conditions for P200 amplitude. FIGURE 2 \| P200 condition effect. P200 waveforms were averaged from the 16 subjects, and we pooled the data from C1, CZ, C2, CP1, CPZ, CP2, P1, PZ, and P2 electrodes. The time window was 160–210 ms for P200. (A) The ERP grand-average waveforms of P200 component in the most-preferred and the least-preferred product design conditions. RESULTS (B) The bar chart reﬂecting mean amplitude of P200 in the most-preferred and the least-preferred product design conditions. Error bars indicate standard error. (C) Topographic maps of the most-preferred and the least-preferred product design conditions for P200 amplitude. FIGURE 3 \| The LPP condition effect. The LPP waveforms were averaged from the 16 subjects, and we pooled the data from C1, CZ, C2, CP1, CPZ, and CP2 electrodes. The time window was 500–700 ms for the LPP. (A) The ERP grand-average waveforms of the LPP component in the most-preferred and the least-preferred product design conditions. (B) The bar chart reﬂecting mean amplitude of P200 in the most-preferred and the least-preferred product design conditions. Error bars indicate standard error. (C) Topographic maps of the most-preferred and the least-preferred product design conditions for the LPP amplitude. FIGURE 3 \| The LPP condition effect. The LPP waveforms were averaged from the 16 subjects, and we pooled the data from C1, CZ, C2, CP1, CPZ, and CP2 electrodes. The time window was 500–700 ms for the LPP. (A) The ERP grand-average waveforms of the LPP component in the most-preferred and the least-preferred product design conditions. (B) The bar chart reﬂecting mean amplitude of P200 in the most-preferred and the least-preferred product design conditions. Error bars indicate standard error. (C) Topographic maps of the most-preferred and the least-preferred product design conditions for the LPP amplitude. these brands [F(4,2964) = 2.279, p > 0.05). In addition, we found that the scores of the same T-shirt design were signiﬁcantly diﬀerent (the SD of the mean scores was 16.858), which indicated the importance of individualized experimental stimuli. Paired t-tests showed that the diﬀerence between consumers’ average preference for the most- and least-preferred experience good designs was signiﬁcant [(M most-preferred = 68.75, M least- preferred = 35.680), t(15) = 13.369, p < 0.01]. the consumer preference rating scores, which implies that the higher the P200 and the LPP amplitudes were, the higher were the personal preference ratings for the product designs of experience goods. Frontiers in Neuroscience \| www.frontiersin.org DISCUSSION A pairwise t-test indicated that the P200 amplitude was larger for the most-preferred than for the least-preferred experience good designs [t(15) = 3.331, p < 0.01, M most-preferred = 5.717, M least-preferred = 4.273; 95% conﬁdence interval [CI]: [0.520 2.367], Figure 2]. For the LPP, a paired t-test also indicated a signiﬁcant diﬀerence between the most- and least-preferred experience good designs [t(15) = 2.372, p < 0.05, M most- preferred = 6.089, M least-preferred = 4.268; 95% CI: [0.185, 3.456], Figure 3]. This study investigated consumers’ neural responses to the most- preferred and least-preferred experience good designs. Using personalized T-shirt pictures, we found that, compared to the least-preferred product designs of experience goods, the P200 and LPP amplitudes were consistently enhanced for the most- preferred designs. Furthermore, the mean amplitudes of P200 and the LPP were signiﬁcantly positively correlated with the consumers’ average preference ratings. From a theoretical standpoint, compared to the study of Telpaz et al. (2015), which only demonstrated a positive relationship between an increased early ERP component (N200) and consumers’ future choice (Telpaz et al., 2015), we demonstrated that, compared to viewing pictures of the least-preferred product designs, viewing pictures of the most- preferred experience good designs elicited larger amplitudes in early and late (P200 and LPP) ERP components. This may be because we focused on experience rather than on search goods. Repeated measures analysis of correlations indicated that the P200 amplitude and average preference rating scores for the experience good designs were signiﬁcantly positively correlated (r = 0.553, p < 0.05, 95% CI = [0.056 0.830]). The LPP amplitude and average preference rating scores for the experience good designs were also signiﬁcantly positively correlated (r = 0.606, p < 0.01, 95% CI = [0.136, 0.853]). These results indicated a positive correlation between the P200 and LPP amplitude and October 2018 \| Volume 12 \| Article 760 Frontiers in Neuroscience \| www.frontiersin.org 5 Neural Response and Product Designs Ma et al. correlated to consumers’ personal preference scores. Based on a previous study (Bradley et al., 2007), we can infer that the most- preferred product designs of experience goods evoked greater emotional arousal than the least-preferred designs. DISCUSSION For highly arousing emotional stimuli (whether positive or negative), the amplitude of the LPP was steadily increased as compared to neutral stimuli (Schupp et al., 2000, 2004). Minnix et al. (2013) have also showed that images with high-emotional arousal induce higher LPP amplitude than images with low-emotional arousal, while the LPP amplitude did not diﬀer between images without emotional arousal (neutral objective images). In addition, other studies have also showed that the esthetic experience evaluation process includes two diﬀerent stages: early impression formation and post-evaluation classiﬁcation (Celaconde et al., 2013). P200 and the LPP can represent the early attention arousal and the late emotional cognition assessment, respectively (Parasuraman and Beatty, 1980). Based on these studies, it is believed that, for product designs with high emotional arousal, later ERP components will be enhanced. In the process of evaluating search good designs, consumers focus on search attributes (e.g., functions, shape, and dimensions). The emotional arousal level induced by these search attributes was lower than by the experience attributes. Therefore, there was no diﬀerence between the LPP component in the preferred and unreferred product conditions. Experience good designs are related to symbolic and esthetical dimensions of product design, which are related to emotional experience and arousal (Homburg et al., 2015). This explains why the amplitude of early and late ERP components were both enhanced for experience goods. p g p g At a practical level, this study established that the two ERP components, P200 and the LPP, were positively correlated with the evaluation of experience good designs, implying that ERPs can be used to predict consumers’ personal preferences for product designs of experience goods. In addition, as this study used the personalized experimental stimuli, the results can also help to understand the consumer’s evaluation of personalized product designs. This study further indicated that ERPs can be used to predict consumers’ preference for experience good designs, without them making actual decisions. EEG signals are of great use when rating data are not available or are limited and may help to reduce responses and inference biases in a practical context. For the early components, this study indicated that, compared to viewing the pictures of the least-preferred experience good designs, viewing pictures of the most-preferred designs elicited a larger P200, rather than N200, amplitude. This result indicates the role of early attention in the evaluation of experience good designs. DISCUSSION Previous studies have indicated that early ERP components (such as P200) reﬂect the perceptual processing of visual stimuli (Parasuraman and Beatty, 1980); a larger degree of visual stimulation (such as the most-preferred product designs) evoked more attentional arousal. However, consumers who purchase search goods will more likely to consider the objective attributes (e.g., price and function) and economic rewards of these attributes (Huang et al., 2009). As search goods (attributes) that are not preferred by consumers will result in lower reward outcomes than they anticipated (Baker and Holroyd, 2011), the N200 component in the early stage will be enhanced to represent this mismatch (Pozharliev et al., 2015). Therefore, ERP amplitudes enhancement diﬀered for search and experience goods. In addition, previous studies have indicated that P200 not only reﬂects the initial response to esthetic stimuli, but is also related to the subsequent approach and withdrawal behavior (Schapkin et al., 2000). For example, early neural activities (such as P200) were related to later subjective behavior during environmental risk evaluation (Qin and Han, 2009). These results support the conclusion that the P200 component was positively related to the subjective preference evaluation in our study. Compared to previous studies, we used subjective rather than objective ratings (for example, sales or other measures) to assess consumers’ preferences for experience good designs. Such subjective measures have been recommended to capture the personal and subjective nature of esthetic evaluation (Chatterjee et al., 2009; Conway and Rehding, 2013; Pearce et al., 2016). We assigned experimental stimuli based on the individual preferences of the participants. We found that participants’ preferences for the same experience product design varied greatly. Therefore, if we had not used personalized materials, we could not have observed diﬀerences in the EEGs during the process of evaluating experience product designs. In addition, by using passive viewing rather than direct assessment of experience good designs, our results indicated that the ERP components evoked by experience good designs were related to consumers’ preferences. These results are consistent with those of previous studies, which have indicated that even without paying explicit attention to pictures (not required to provide assessment), participants can still exhibit a neural response to the pictures (Chatterjee et al., 2009; Pozharliev et al., 2015). This study had some limitations. First, although this study focused on experiential products, we only used T-shirts as experimental stimuli. Frontiers in Neuroscience \| www.frontiersin.org DISCUSSION Besides the esthetic dimension, the enhancement of the LPP component in this paper may also be related to the symbolic dimension of experiential product designs, because the symbolic meaning of the product design is also related to aﬀective product attributes (Homburg et al., 2015). The increased LPP amplitude also reﬂected motivated attentional processing and motivational intensity (Ferrari et al., 2011), implying enhanced allocation of relevant resources to promoting and speeding up the processes leading to a suitable response to the stimuli (Lang et al., 1997). Previous studies have shown that enhanced LPP reﬂects the increased positive emotional motivation for a preferred brand and represents a positive buying willingness for the brand (Bosshard et al., 2016). Therefore, in line with previous studies, we observed that LPP was related to behavioral responses to product designs of experience goods. Previous studies have shown that the late component, LPP, was associated with the arousal of emotional stimuli. For highly arousing emotional stimuli (whether positive or negative), the amplitude of the LPP was steadily increased as compared to neutral stimuli (Schupp et al., 2000, 2004). Minnix et al. (2013) have also showed that images with high-emotional arousal induce higher LPP amplitude than images with low-emotional arousal, while the LPP amplitude did not diﬀer between images without emotional arousal (neutral objective images). In addition, other studies have also showed that the esthetic experience evaluation process includes two diﬀerent stages: early impression formation and post-evaluation classiﬁcation (Celaconde et al., 2013). P200 and the LPP can represent the early attention arousal and the late emotional cognition assessment, respectively (Parasuraman and Beatty, 1980). Based on these studies, it is believed that, for product designs with high emotional arousal, later ERP components will be enhanced. In the process of evaluating search good designs, consumers focus on search attributes (e.g., functions, shape, and dimensions). The emotional arousal level induced by these search attributes was lower than by the experience attributes. Therefore, there was no diﬀerence between the LPP component in the preferred and unreferred product conditions. Experience good designs are related to symbolic and esthetical dimensions of product design, which are related to emotional experience and arousal (Homburg et al., 2015). This explains why the amplitude of early and late ERP components were both enhanced for experience goods. Previous studies have shown that the late component, LPP, was associated with the arousal of emotional stimuli. REFERENCES Camerer, C., and Yoon, C. (2015). Introduction to the Journal of Marketing Research special issue on euroscience and marketing. J. Mark. Res. 52, 423–426. doi: 10.1509/0022-2437-52.4.423 Augustin, M. D., Defranceschi, B., Fuchs, H. K., Carbon, C., and Hutzler, F. (2011). The neural time course of art perception: an ERP study on the processing of style versus content in art. Neuropsychologia 49, 2071–2081. doi: 10.1016/j. neuropsychologia.2011.03.038 Celaconde, C. J., Garciaprieto, J., Ramasco, J. J., Mirasso, C. R., Bajo, R., Munar, E., et al. (2013). Dynamics of brain networks in the aesthetic appreciation. Proc. Natl. Acad. Sci. U.S.A. 110, 10454–10461. doi: 10.1073/pnas.1302855110 et al. (2013). Dynamics of brain networks in the aesthetic appreciation. Proc. Natl. Acad. Sci. U.S.A. 110, 10454–10461. doi: 10.1073/pnas.1302855110 Baker, T. E., and Holroyd, C. B. (2011). Dissociated roles of the anterior cingulate cortex in reward and conﬂict processing as revealed by the feedback error- related negativity and N200. Biol. Psychol. 87, 25–34. doi: 10.1016/j.biopsycho. 2011.01.010 Celaconde, C. J., Marty, G., Maestu, F., Ortiz, T., Munar, E., Fernandez, A., et al. (2004). Activation of the prefrontal cortex in the human visual aesthetic perception. Proc. Natl. Acad. Sci. U.S.A. 101, 6321–6325. doi: 10.1073/pnas. 0401427101 Bloch, P. H. (1995). Seeking the ideal form: product design and consumer response. J. Mark. 59, 16–29. doi: 10.2307/1252116 Cerf, M., Greenleaf, E. A., Meyvis, T., and Morwitz, V. G. (2015). Using single- neuron recording in marketing: opportunities, challenges, and an application to fear enhancement in communications. J. Mark. Res. 52, 530–545. doi: 10. 1509/jmr.13.0606 Bloch, P. H., Brunel, F. F., and Arnold, T. J. (2003). Individual diﬀerences in the centrality of visual product aesthetics: concept and measurement. J. Consum. Res. 29, 551–565. doi: 10.1086/346250 j Chatterjee, A., Thomas, A., Smith, S. E., and Aguirre, G. K. (2009). The neural response to facial attractiveness. Neuropsychology 23, 135–143. doi: 10.1037/ a0014430 Bosshard, S. S., Bourke, J. D., Kunaharan, S., Koller, M., and Walla, P. (2016). Established liked versus disliked brands: brain activity, implicit associations and explicit responses. Cogent Psychol. 3:1176691. doi: 10.1080/23311908.2016. 1176691 Coates, D. (2003). Watches Tell More than Time: Product Design, Information, and the Quest for Elegance. London: McGraw-Hill. Bradley, M. M., Hamby, S., Low, A., and Lang, P. J. (2007). Brain potentials in perception: picture complexity and emotional arousal. Psychophysiology 44, 364–373. doi: 10.1111/j.1469-8986.2007.00520.x Conway, B. R., and Rehding, A. (2013). Neuroaesthetics and the trouble with beauty. PLoS. Biol. 11:e1001504. doi: 10.1371/journal.pbio. FUNDING This study was supported by grant LY17G020013 from the Natural Science Foundation of Zhejiang Province, grant 18YJA630080 and 18YJC630232 from Humanities and Social Sciences Projects of Ministry of Education, grant 71302084 from the National Natural Science Foundation of China, grant 17BGL092 from the National Social Science Foundation of China, and Fund of Academy of Neuroeconomics and Neuromanagement at Ningbo University. This study was supported by grant LY17G020013 from the Natural Science Foundation of Zhejiang Province, grant 18YJA630080 and 18YJC630232 from Humanities and Social Sciences Projects of Ministry of Education, grant 71302084 from the National Natural Science Foundation of China, grant 17BGL092 from the National Social Science Foundation of China, and Fund of Academy of Neuroeconomics and Neuromanagement at Ningbo University. ACKNOWLEDGMENTS This study was carried out in accordance with the recommendations of the Ethics Committee of the Academy of This study was carried out in accordance with the recommendations of the Ethics Committee of the Academy of Neuroeconomics and Neuromanagement at Ningbo University We thank Hao Ding, Menglin Zhao, Aiming Yang, and Jiangli Gao for their help with experiments. y Neuroeconomics and Neuromanagement at Ningbo University DISCUSSION Whether the results of this study can be extended to other experience goods requires further attention. Second, in order to capture consumers’ subjective As for the LPP, we found a more enhanced LPP amplitude at around 600 ms for the most-preferred product designs than for the least-preferred designs after exposure to experience good designs. We also found that the LPP amplitudes were positively October 2018 \| Volume 12 \| Article 760 6 Neural Response and Product Designs Ma et al. and personalized characteristics of product design evaluation, we used the subjective preference scores of each participant to classify experimental stimuli into two groups. Although we used various methods to maximize the accuracy of subjective measurement, a social desirability bias may still remain. Third, we did not directly compare the ERP response between experience and search goods. Although we focused on experiential products, the concept of experience products was proposed relative to search goods (Nelson, 1970, 1974). It may be better to compare the diﬀerences in ERP responses between the two types of product designs directly. with written informed consent from all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the Ethics Committee of the Academy of Neuroeconomics and Neuromanagement at Ningbo University. AUTHOR CONTRIBUTIONS YM made substantial contributions and participated in all aspects of the paper, conducted the experiment, analyzed the data, and wrote the manuscript. JJ and WY made substantial contributions to the work and participated in all aspects of the paper. WZ and ZX participated in the data acquisition and data interpretation stage. QM oversaw the study and managed every part of research. All authors read and approved the ﬁnal manuscript. CONCLUSION This study explored individuals’ neural responses to experience good designs and their relation to personal preferences. Using personalized T-shirt designs as stimuli, the results indicated that two ERP components, P200 and the LPP, were both enhanced in response to the most-preferred product designs of experience goods as compared to the least-preferred designs, when participants simply viewed the product designs without making actual decisions. Both of the ERP components were positively correlated with the consumers’ preference scores for these experience good designs. The results indicated that ERP signals may provide important information regarding consumer preferences for experience good designs and can shed light on why consumers like customized products. REFERENCES 60, 17–28. Huang, P., Lurie, N. H., and Mitra, S. (2009). Searching for experience on the web: an empirical examination of consumer behavior for search and experience goods. J. Mark. 73, 55–69. doi: 10.1509/jmkg.73.2.55 Schreier, M. (2006). The value increment of mass-customized products: an empirical assessment. J. Consum. Behav. 5, 317–327. doi: 10.1002/cb.183 Junghöfer, M., Kissler, J., Schupp, H., Putsche, C., Elling, L., and Dobel, C. (2010). A fast neural signature of motivated attention to consumer goods separates the sexes. Front. Hum. Neurosci. 4:179. doi: 10.3389/fnhum.2010.00179 Schupp, H., Cuthbert, B., Bradley, M., Hillman, C., Hamm, A., and Lang, P. (2004). Brain processes in emotional perception: motivated attention. Cogn. Emot. 18, 593–611. doi: 10.1080/02699930341000239 Kumar, M., and Garg, N. (2010). Aesthetic principles and cognitive emotion appraisals: how much of the beauty lies in the eye of the beholder? J. Consum. Psychol. 20, 485–494. doi: 10.1016/j.jcps.2010.06.015 Schupp, H. T., Cuthbert, B. N., Bradley, M. M., Cacioppo, J. T., Ito, T. A., and Lang, P. J. (2000). Aﬀective picture processing: the late positive potential is modulated by motivational relevance. Psychophysiology 37, 257–261. doi: 10.1111/1469- 8986.3720257 Lang, P. J., Bradley, M. M., and Cuthbert, B. N. (1997). “Motivated attention: aﬀect, activation, and action,” in Attention and Orienting: Sensory and Motivational Processes, eds P. J. Lang, R. F. Simons, and M. T. Balaban (Hillsdale, NJ: Erlbaum), 97–136. Semlitsch, H. V., Anderer, P., Schuster, P., and Presslich, O. (1986). A solution for reliable and valid reduction of ocular artifacts, applied to the P300 ERP. Psychophysiology 23, 695–703. doi: 10.1111/j.1469-8986.1986.tb00696.x Levy, D. J., and Glimcher, P. W. (2012). The root of all value: a neural common currency for choice. Curr. Opin. Neurobiol. 22, 1027–1038. doi: 10.1016/j.conb. 2012.06.001 Silvia, P. J., and Warburton, J. B. (2006). “Positive and negative aﬀect: bridging states and traits,” in Personality and Everyday Functioning, Vol. 1, eds D. L. Segal and J. C. Thomas (New York, NY: Wiley), 268–284. Luck, S. J., Woodman, G. F., and Vogel, E. K. (2000). Event-related potential studies of attention. Trends Cogn. Sci. 4, 432–440. doi: 10.1016/S1364-6613(00)01545- X Telpaz, A., Webb, R., and Levy, D. J. (2015). Using EEG to predict consumers’ future choices. J. Mark. Res. 52, 511–529. doi: 10.1509/jmr.13.0564 Wang, Y., and Tseng, M. M. (2015). A Naïve Bayes approach to map customer requirements to product variants. J. Intell. Manuf. 26, 501–509. doi: 10.1007/ s10845-013-0806-2 Luo, L., Kannan, P., and Ratchford, B. T. (2008). REFERENCES 1001504 Davidson, R. J., Pizzagalli, D., Nitschke, J. B., and Putnam, K. (2002). Depression: perspectives from aﬀective neuroscience. Annu. Rev. Psychol. 53, 545–574. doi: 10.1146/annurev.psych.53.100901.135148 Bradley, M. M., Sabatinelli, D., Lang, P. J., Fitzsimmons, J. R., King, W., and Desai, P. (2003). Activation of the visual cortex in motivated attention. Behav. Neurosci. 117, 369–380. doi: 10.1037/0735-7044.117.2.369 October 2018 \| Volume 12 \| Article 760 Frontiers in Neuroscience \| www.frontiersin.org 7 Neural Response and Product Designs Ma et al. Dolcos, F., and Cabeza, R. (2002). Event-related potentials of emotional memory: encoding pleasant, unpleasant, and neutral pictures. Cogn. Aﬀect. Behav. Neurosci. 2, 252–263. doi: 10.3758/CABN.2.3.252 Pearce, M. T., Zaidel, D. W., Vartanian, O., Skov, M., Leder, H., Chatterjee, A., et al. (2016). Neuroaesthetics: the cognitive neuroscience of aesthetic experience. Perspect. Psychol. Sci. 11, 265–279. doi: 10.1177/1745691615621274 Ferrari, V., Bradley, M. M., Codispoti, M., and Lang, P. J. (2011). Repetitive exposure: brain and reﬂex measures of emotion and attention. Psychophysiology 48, 515–522. doi: 10.1111/j.1469-8986.2010.01083.x Pozharliev, R., Verbeke, W. J. M. I., Van Strien, J. W., and Bagozzi, R. P. (2015). Merely being with you increases my attention to luxury products: using EEG to understand consumers’ emotional experience with luxury branded products. J. Mark. Res. 52, 546–558. doi: 10.1509/jmr.13.0560 Hajcak, G., Moser, J. S., and Simons, R. F. (2006). Attending to aﬀect: appraisal strategies modulate the electrocortical response to arousing pictures. Emotion 6, 517–522. doi: 10.1037/1528-3542.6.3.517 Qin, J., and Han, S. (2009). Neurocognitive mechanisms underlying identiﬁcation of environmental risks. Neuropsychologia 47, 397–405. doi: 10.1016/j. neuropsychologia.2008.09.010 Heine, K., and Phan, M. (2011). Trading-up mass-market goods to luxury products. Aust. Mark. J. 19, 108–114. doi: 10.1016/j.ausmj.2011.03.001 Reimann, M., Zaichkowsky, J. L., Neuhaus, C., Bender, T., and Weber, B. (2010). Aesthetic package design: a behavioral, neural, and psychological investigation. J. Consum. Psychol. 20, 431–441. Aust. Mark. J. 19, 108–114. doi: 10.1016/j.ausmj.2011.03.001 Hoch, S. J., and Ha, Y.-W. (1986). Consumer Learning: advertising and the ambiguity of product experience. J. Consum. Res. 13, 221–233. doi: 10.1086/ 209062 Sabatinelli, D., Lang, P. J., Keil, A., and Bradley, M. M. (2006). Emotional perception: correlation of functional MRI and event-related potentials. Cereb. Cortex 17, 1085–1091. doi: 10.1093/cercor/bhl017 Homburg, C., Schwemmle, M., and Kuehnl, C. (2015). New product design: concept, measurement, and consequences. J. Mark. 79, 41–56. doi: 10.1509/jm. 14.0199 Schapkin, S. A., Gusev, A. N., and Kuhl, J. (2000). Categorization of unilaterally presented emotional words: an ERP analysis. Acta Neurobiol. Exp. REFERENCES Incorporating subjective characteristics in product design and evaluations. J. Mark. Res. 45, 182–194. doi: 10.1509/jmkr.45.2.182 Weathers, D., Sharma, S., and Wood, S. L. (2007). Eﬀects of online communication practices on consumer perceptions of performance uncertainty for search and experience goods. J. Retail. 83, 393–401. doi: 10.1016/j.jretai.2007.03.009 Minnix, J. A., Versace, F., Robinson, J. D., Lam, C. Y., Engelmann, J. M., Cui, Y., et al. (2013). The late positive potential (LPP) in response to varying types of emotional and cigarette stimuli in smokers: a content comparison. Int. J. Psychophysiol. 89, 18–25. doi: 10.1016/j.ijpsycho.2013.04.019 Conﬂict of Interest Statement: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Psychophysiol. 89, 18–25. doi: 10.1016/j.ijpsycho.2013.04.019 Nelson, P. (1970). Information and consumer behavior. J. Polit. Econ. 78, 311–329. doi: 10.1086/259630 Nelson, P. (1974). Advertising as information. J. Polit. Econ. 82, 729–754. doi: 10.1086/260231 Copyright © 2018 Ma, Jin, Yu, Zhang, Xu and Ma. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Olofsson, J. K., Nordin, S., Sequeira, H., and Polich, J. (2008). Aﬀective picture processing: an integrative review of ERP ﬁndings. Biol. Psychol. 77, 247–265. doi: 10.1016/j.biopsycho.2007.11.006 Parasuraman, R., and Beatty, J. (1980). Brain events underlying detection and recognition of weak sensory signals. Science 210, 80–83. doi: 10.1126/science. 7414324 October 2018 \| Volume 12 \| Article 760 Frontiers in Neuroscience \| www.frontiersin.org 8
W2607842258.txt	https://periodicos.sbu.unicamp.br/ojs/index.php/riesup/article/download/8650576/16789	en		Condiciones laborales y socioeconómicas de estudiantes y egresados de una institución de educación superior en el Periurbano de la zona Metropolitana del Valle de México: un estudio exploratorio	Revista Internacional de Educação Superior	2,017	cc-by	10,247	DOI ARTIGO 10.22348/riesup.v0i0.7674 CONDICIONES LABORALES Y SOCIOECONÓMICAS DE ESTUDIANTES Y EGRESADOS DE UNA INSTITUCIÓN DE EDUCACIÓN SUPERIOR EN EL PERIURBANO DE LA ZONA METROPOLITANA DEL VALLE DE MÉXICO: UN ESTUDIO EXPLORATORIO SOCIO-ECONOMIC AND LABOR CONDITIONS OF STUDENTS AND ALUMNI OF A HIGHER EDUCATION INSTITUTION IN THE PERIURBAN AREA OF THE METROPOLITAN VALLEY OF MEXICO CITY: AN EXPLORATORY STUDY CONDIÇÕES SÓCIO-ECONÓMICAS E OBRA DE ESTUDANTES E GRADUADOS DE UMA INSTITUIÇÃO DO ENSINO SUPERIOR NA REGIÃO METROPOLITANO SUBURBAN VALLEY MÉXICO: UM ESTUDO EXPLORATÓRIO Gustavo Mejía Pérez1 Óscar Espinoza Ortega2 RESUMO O artigo apresenta a análise do trabalho e da situação económica dos diplomados de uma instituição de ensino superior localizada na Região Metropolitana do México. O estudo foi realizado com base em dados fornecidos pelos graduados Institucional Sistema de Acompanhamento da Universidade Autônoma do Estado do México no período 2000-2013. Os resultados da análise mostram uma entrada para o mercado de trabalho diferenciadas por sexo e grau, possivelmente, ele determina itinerários diferentes nos processos de independência família de estudantes e graduados. PALAVRAS-CHAVE: Educação superior. Independência famíliar. Mercado de trabalho. ABSTRACT The paper presents the analysis of the labor and economic situations of alumni from a higher education institution located in the Metropolitan Area of Mexico City. The study anazlyed data provided by the Institutional Alumni Tracking System of the Autonomous University of the State of Mexico from 2000 to 2013. The results of the analysis shows entries into the labor market differentiated by gender and degree, which possibly determine different itineraries in the students’ and alumni’s processes of family independence. KEYWORDS: Higher education. Family independence. Labor market. 1 Departamento de Investigaciones Educativas. Centro de Investigaciones y Estudios Avanzados del Instituto Politécnico Nacional (CINVESTAV), México. – E-mail: cabezahidra@gmail.com – ORCID: http://orcid.org/0000-0002-2370-4276 2 Centro Universitario UAEM Valle de Teotihuacán. Universidad Autónoma del Estado de México, México. Email: oeo_21@yahoo.com.mx - ORCID: http://orcid.org/0000-0002-2304-7587 Submetido em: 18-11-2016 – Aceito em: 13-12-2016 – Publicado em: 07-02-2017 © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 p.66-89 jan./abr. 2017 ISSN 2446-9424 [66] DOI ARTIGO 10.22348/riesup.v0i0.7674 RESUMEN El documento presenta el análisis de la situación laboral y económica de egresados de una institución de educación superior ubicada en el Zona Metropolitana del Valle de México. El estudio se realizó con base en los datos proporcionados por el Sistema Institucional de Seguimiento de Egresados de la Universidad Autónoma del Estado de México en el período 2000-2013. Los resultados del análisis muestran un ingreso al mercado laboral diferenciado por sexo y por carrera, que posiblemente determina diferentes itinerarios en los procesos de independencia familiar de estudiantes y egresados. PALABRAS CLAVE: Educación superior. Independencia familiar. Mercado laboral. INTRODUCCIÓN El objetivo de este documento es describir las condiciones socioeconómicas y laborales de los egresados del Centro Universitario UAEM Valle de Teotihuacán (CUVT), durante los estudios y al término de los mismos, con el fin de explorar algunas hipótesis acerca de los procesos de independencia económica que se dan en este periodo. Estos procesos son importantes pues como señalan De Vries y Navarro Rangel (2015), al terminar la escuela preparatoria, los y las jóvenes empiezan a convertirse en adultos, mediante un proceso que involucra las relaciones de pareja, una vivienda independiente de sus padres y la participación en el mercado laboral. La pregunta que surge aquí es si la universidad es un catalizador que dispara estos procesos o sólo un proceso que ocurre de forma paralela a los estudios superiores. El documento se compone de tres apartados. En el primero se muestra una síntesis de lo que dice la literatura sobre las condiciones laborales y socioeconómicas de quienes acuden y egresan a la universidad en México. En la siguiente sección se describe el caso de estudio, así como la metodología empleada para la investigación. La tercera parte muestra las características laborales y socioeconómicas del grupo estudiado en dos momentos: mientras estudiaban la licenciatura y al término de la universidad. Finalmente, en las conclusiones se discuten, con base en la literatura previa y los datos recopilados, los posibles cambios que la educación superior (ES) generó en los egresados respecto a sus condiciones laborales y su independencia de las familias de origen. SITUACIÓN LABORAL Y ECONÓMICA DE ESTUDIANTES Y EGRESADOS UNIVERSITARIOS MEXICANOS: UNA BREVE REVISIÓN DE LA LITERATURA Con el objetivo de saber qué es lo que dicen los estudios empíricos sobre la situación laboral y económica de estudiantes y egresados universitarios en México, se realizó una búsqueda de investigaciones que abordaran este tópico. Al final de esta búsqueda se encontraron 22 estudios publicados entre 2008 y 2016, en los cuales se utilizaron como marco de referencia las teorías del capital humano (Becker, 1964; Schultz, 1963), competencia por los puestos Thurow (1975), la asignación (Sattinger, 1993) y la reproducción social (Bourdieu y Passeron, 1979). Las metodologías y técnicas utilizadas incluyen encuestas, entrevistas, análisis de fuentes secundarias y revisión bibliográfica. De los 22 estudios sólo cuatro hacen referencia al contexto donde se realiza la investigación: tres se realizan en espacio urbanos y uno en un medio rural. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [67] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 La información que a continuación se presenta se organiza en dos grandes apartados: el trabajo de los estudiantes durante la licenciatura y condiciones laborales y económicas de los egresados. Las características demográficas y el trabajo de los estudiantes durante la licenciatura En un estudio realizado en el Centro Universitario de Los Altos, de la Universidad de Guadalajara (UDG), los egresados de contaduría (generación 2008-2012) son jóvenes que en su mayoría ronda los 24 y 25 años de edad y se inscribieron a la universidad después de haber terminado la educación media superior (GONZÁLEZ PÉREZ; GONZÁLEZ FRANCO, 2015). Según los resultados de esta investigación hay una diferencia notable en el estado civil de acuerdo al sexo: el 61% de las mujeres y el 11% de los hombres están casados. En el mismo campus de Los Altos, González Anaya y Carrillo Torres (2016) encontraron que en la carrera de psicología (generación 2008-2012) la mayoría de los estudiantes son originarios de municipios o localidades de la Región Altos Sur, las cuales son poblaciones de tipo rural en muchos de los casos. Además identificaron que la población analizada tiene entre 24 y 27 años de edad. El 28.57% de los entrevistados son varones, de los cuales solo 12.5% están casados, y el resto, es decir, 71.42% son mujeres, de las cuales 17.85 % están casadas. En otro estudio en la UDG, Enciso Ávila (2013) identificó que la distribución entre hombres y mujeres de los graduados persiste una ventaja de 8% a favor de las mujeres. En esta universidad los estudiantes de la UDG provienen mayoritariamente de familias no universitarias, es decir, son la primera generación de la familia que accede a la universidad y se gradúa. Además, casi la mitad de ellos, el 40%, proviene de familias cuyos padres tienen como máximo estudios primarios. Sin embargo, entre sus graduados están sobrerrepresentados aquellos cuyos padres tienen niveles educativos medios y superiores, mientras que están subrepresentados los que sólo disponen de estudios primarios. De acuerdo con un estudio realizado en la UDG, se pudo ver que la mayoría de los alumnos inician su itinerario laboral durante los estudios, y las situaciones mixtas de estudios y trabajo continúan presentes años después del egreso (PLANAS COLL, 2013). También en la UDG, Carrillo Regalado y Ríos Almodóvar (2014) realizaron un análisis comparativo de los factores que explican la condición laboral y la jornada de trabajo de los estudiantes, mediante el empleo de modelos de regresión probabilístico, soluciones de mínimos cuadrados ordinarios y datos de la Encuesta UDG y la Encuesta Nacional de Ocupación y Empleo (ENOE). En sus resultados encontraron una gran diferencia en las proporciones de participación laboral que presentan los alumnos de licenciatura de la UDG en comparación con el promedio de universidades de México: 43% de los alumnos de la UDG trabajan, mientras que hacen lo mismo 35% en promedio nacional. En cuanto a la explicación de la condición laboral del estudiante (desempeñarse o no en el mercado de trabajo), el factor más determinante es la carrera que cursan en la universidad, la cual es decisiva para que incursionen o no activamente en la economía. En el caso de la UDG las carreras que presentan mayor propensión al trabajo de los estudiantes son las de los centros universitarios de Ciencias Económico Administrativas, Ciencias Biológicas y Agropecuarias, Ciencias Sociales y Humanidades, y Ciencias Exactas e Ingenierías. Otros factor determinante de la condición laboral de los estudiantes es la transferencia de dinero de los padres a los alumnos, si estos ingresos no laborales se incrementan los alumnos podrían desistir de participar en el © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [68] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 mercado laboral. Finalmente, el sexo es una variable determinante para la amplitud de la jornada laboral, en el sentido de que ser hombre repercute en una mayor cantidad de horas trabajadas por jornada laboral. En un análisis previo realizado por Carrillo Regalado y Ríos Almodóvar (2009), a partir de la ENOE 2004, se mostró que la condición de trabajar o no es afectada por las transferencias del jefe del hogar a los estudiantes (o ingresos no laborales) y el sexo del estudiante (ser hombre), pues son las variables que determinan en mayor medida la condición laboral. Además, hay otros factores que también intervienen en esta condición, tales como la escolaridad del jefe del hogar, la edad del estudiante y disciplina de la licenciatura que cursan. En este estudio la población estudiantil estuvo compuesta de 51% de hombres y el resto fueron mujeres, aunque los estudiantes hombres que trabajan son más que las mujeres (26% frente a 17% de las mujeres) lo cual implica que poco más de tres quintos del total de puestos laborales ocupados por estudiantes sean desempeñados por hombres. A pesar de que en el imaginario se cree que el que un estudiante de licenciatura trabaje es un factor negativo para su desarrollo, la condición del estudiante que trabaja tiene implicaciones diversas entre las que se pueden identificar procesos de agobio y sacrificio, pero también de satisfacciones y orgullo personal. E incluso estos trabajos pueden significar la integración de saberes de distinto origen –escolar y laboral– para el desarrollo de competencias, que implica una conjunción de conocimientos, habilidades, actitudes y valores orientados al desempeño social y profesional, además de funcionar como un puente con los mercados de trabajo, las necesidades del entorno y el desarrollo laboral y personal. Estos fueron algunos de los resultados del estudio realizado por Cuevas y De Ibarrola (2013) con estudiantes del Instituto Tecnológico de una ciudad del centro del país que se encontraban cursando los últimos semestres en las áreas de Ingeniería, Administración e Informática. El ingreso al mercado de trabajo entre las personas que acuden a la universidad puede darse al término de los estudios o en el transcurso de los mismos, como lo muestran De Vries y Navarro Rangel (2015) en el análisis realizado con base en los datos obtenidos del proyecto El Profesional Flexible en la Sociedad del Conocimiento (Proflex), en el que participaron nueve universidades mexicanas (seis públicas y tres privadas). De acuerdo con De Vries y Navarro Rangel (2015), la mayoría de los egresados tuvo experiencias laborales relacionadas con sus estudios, aparte de las prácticas profesionales. Incluso un 24.9% de los estudiantes indica que los estudios no eran su actividad principal en los últimos dos años de su carrera. Los resultados presentados en el trabajo anterior coinciden con lo encontrado por González Pérez y González Franco (2015) en el Centro Universitario de Los Altos, en el que la gran mayoría de los egresados de la carrera de contaduría se sostuvo económicamente durante su formación universitaria. Las cifras son distintas en el caso de los egresados de psicología del mismo campus, de acuerdo con González Anaya y Carrillo Torres (2016), sólo el 16% fue autosuficiente económicamente durante la carrera y el resto dependía de la beca, los padres o el cónyuge. Esta investigación muestra que sólo el 21% de los padres y el 11% de las madres cuentan con estudios de licenciatura o superiores. Sin embargo, no todos los estudios apuntan a señalar las diferencias laborales entre hombres y mujeres, de acuerdo con un estudio sobre trayectorias laborales realizado en la UDG por Planas Coll (2013); en términos de calidad de la inserción para los egresados que están ocupados al final de su itinerario, no se observan diferencias significativas, ni por © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [69] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 sexos, ni por orígenes sociales. En cambio, sí se observan diferencias significativas relacionadas con el sexo de los egresados, pero no con su origen social, en el grado de actividad y de ocupación, de tal forma que se puede inferir que trabajar durante los estudios no es un comportamiento ni de pobres ni de hombres (PLANAS COLL; ENCISO ÁVILA, 2013). Condiciones laborales y económicas de los egresados El estudio de Burgos Flores y López Montes (2010), realizado con egresados de la Universidad de Sonora, encontró que el problema más grave no es el desempleo de profesionistas, sino las condiciones en las que se insertan en el mercado de trabajo: una proporción importante de ellos no obtiene remuneraciones adecuadas; se desempeña en puestos de carácter no profesional; ocupa un puesto para el cual no se requería de estudios de nivel superior; el empleo que desarrolla no tiene mucha coincidencia con la carrera estudiada; y no aplica plenamente los conocimientos y habilidades adquiridos en la universidad. Este mismo estudio mostró, a través de la construcción del indicador de pertinencia en el mercado laboral, diseñado a partir de seis variables que caracterizan a dicho mercado, que existe gran heterogeneidad en cuanto a la situación laboral para los egresados de cada una de las carreras. En términos generales, con buen mercado laboral destacan carreras del área de Ingeniería y Tecnología, así como algunas carreras del área de Ciencias Biológicas y de la Salud. En una situación intermedia se ubican las tradicionales carreras del área Económica-Administrativa y con inadecuado mercado laboral se encuentran carreras de las áreas de Ciencias Sociales y de Humanidades (BURGOS FLORES; LÓPEZ MONTES, 2010). En el estudio de González Pérez y González Franco (2015), la mayoría de los egresados de contaduría del Centro Universitario de Los Altos obtuvieron trabajo y ascendieron dentro de su área de formación profesional. En el mismo Centro Universitario se realizó un estudio con egresados de la carrera de psicología, y los resultados mostraron que las áreas de desempeño de los entrevistados se circunscriben en su mayoría a la Psicología clínica y la Psicología educativa, aunque un alto porcentaje de psicólogos laboraban en áreas distintas a su formación (GONZÁLEZ ANAYA; CARRILLO TORRES, 2016). Los mismos investigadores reportan que entre los egresados de psicología 4.54% manifestaron no estar trabajando, 40.9% laboran en actividades que no están relacionadas con la psicología, o donde no aplican los conocimientos adquiridos en su formación universitaria, mientras que el restante 54% sí lo hace (GONZÁLEZ ANAYA; CARRILLO TORRES, 2016). Resultados similares arrojó la investigación de Rubio Hernández y Salgado Vega (2014), realizada con egresados de la licenciatura de economía de la Universidad Autónoma del Estado de México (UAEMEX) del periodo 2000-2010. De acuerdo con este estudio los egresados presentan un comportamiento de buena aceptación respecto al proceso de transición al mercado laboral; sin embargo, son contratados en empleos que no corresponden a su formación y con baja remuneración. Un tercer estudio realizado en el Centro Universitario de Los Altos por Pérez Pulido, Moreno García y Caldera Montes (2011), con una muestra de 405 egresados de un total de 1,872 alumnos que terminaron sus estudios durante los ciclos 2001 y 2007 (de alguna de las 13 carreras que se imparten en el campus), encontró que en la mayoría de las carreras los hombres han encontrado un mejor posicionamiento laboral en comparación con las mujeres, pues ellos cuentan con mejores puestos de trabajo y son mejor remunerados. El promedio de © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [70] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 ingresos mensuales de los varones es de 8,731.00 pesos, mientras que las mujeres perciben 6,457.00 pesos. Esto representa una diferencia mensual de 2,274.00 pesos a favor de ellos. Sin embargo al comparar el ingreso promedio mensual por género y licenciatura, destaca que únicamente en tres carreras (Odontología, Veterinaria y Psicología) las mujeres están mejor remuneradas que los hombres. Estos resultados coinciden con lo observado por Jiménez Vázquez (2011) en su revisión de los estudios sobre trayectorias laborales. La autora señala que las condiciones laborales de quienes egresan de la licenciatura son diferentes para hombres y mujeres: las mujeres tienen que esforzarse más para desarrollar una trayectoria exitosa en comparación con los hombres, y esto no siempre es reconocido en el aspecto social, laboral y económico, pues el mercado de trabajo está segmentado y estratificado (JIMÉNEZ VÁSQUEZ, 2011, p. 10). Otro estudio que muestra las diferencias en las condiciones laborales entre hombres y mujeres es el realizado por Sánchez Olavarría (2012) con egresados de la carrera de comunicación en la Universidad del Altiplano (Tlaxcala). Los resultados de esta investigación permiten ver que la mayoría de los egresados de la carrera de ciencias de la comunicación han tenido escasa movilidad. No obstante, la movilidad ocupacional se presenta en mayor medida en los egresados con mayor tiempo en el mercado laboral, principalmente en las mujeres (SÁNCHEZ OLAVARRÍA, 2012). Resultados similares han sido reportados por De Vries, Cabrera y Ríos (2013a), donde a partir de datos de egresados de nueve universidades mexicanas, descubrieron que los factores que explican mejor la diferenciación salarial son aspectos del mercado de trabajo mismo, el tipo de institución, el área de conocimiento, la experiencia laboral, el puesto del padre y el género. También identificaron diferencias salariales entre áreas de conocimiento: los puestos de altos ingresos se encuentran básicamente en los sectores económicoadministrativo (administración, economía) y técnico (ingenierías). En ambas áreas, los sueldos promedio son más altos. Pero el ingreso al mercado de trabajo y las condiciones laborales no sólo tienen que ver con el origen socioeconómico y el sexo; hay factores estructurales, como el número de egresados y empleos en una ciudad o región, como lo indica el estudio de Segundo Ramírez (2009), que buscó encontrar de qué manera la política de diversificación que ha llevado a cabo la Universidad Autónoma de Ciudad Juárez (UACJ) ha repercutido en el tiempo que tardan sus estudiantes en encontrar empleo al término de sus estudios de nivel licenciatura, así como el ingreso que ellos perciben en su empleo después de dos años de egreso. Para ello utilizó la información del Seguimiento de Egresados que ha realizado año con año la UACJ a los egresados en el periodo 2001-2004 y que fueron encuestados en los periodos 2003-2006. Los resultados de este trabajo muestran que la ampliación de la matrícula por parte de la UACJ en carreras que por tradición cuentan con baja concentración de demanda estudiantil no ha afectado el tiempo que tardan sus egresados en encontrar empleo, lo que indica que la diversificación de la matrícula no ha generado un efecto negativo en la incorporación de los egresados de la UACJ a la vida laboral. Por el contrario, cuando los egresados pertenecen a una carrera de baja concentración, tiene una menor probabilidad de contar con un ingreso mensual de más de 6 salarios mínimos; esto genera un efecto negativo en el bienestar del © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [71] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 individuo, así como desigualdad entre individuos con los mismos años académicos (SEGUNDO RAMÍREZ, 2009). El estudio de Salas Durazo y Murillo García (2013) realizado a partir de los datos de la ENOE generada por el Instituto Nacional de Estadística y Geografía (INEGI) durante el intervalo comprendido entre 2005 y 2012, también descubrieron algunos patrones estructurales en el mercado de trabajo de los jóvenes. Por ejemplo, encontraron que la educación juega un papel importante en el posicionamiento y en el ingreso de los jóvenes al mercado de trabajo, pues en los segmentos medio bajo y bajo no existe una tendencia generalizada a llevar a cabo estudios medio superiores y superiores. También identificaron que los jóvenes ocupados que no cuentan con estudios de al menos nivel medio superior se encuentran casados; mientras que sus contrapartes con estudios superiores se encuentran en su totalidad solteros. De acuerdo con los autores, en el estrato bajo los costos de oportunidad de estudiar son tan altos que las personas deciden incorporarse al sector productivo con la restricción de que su salario en el tiempo nunca igualará al salario de quien decidió prepararse académicamente. Por su parte, en el estrato alto parecieran existir las condiciones para que los jóvenes se preparen en el ámbito universitario sin la presión de tener que emplearse (SALAS DURAZO; MURILLO GARCÍA, 2013). El mismo estudio reporta que el principal obstáculo que enfrentan las mujeres al incorporarse a la dinámica productiva es que en esencia el mercado laboral mexicano es masculino. De acuerdo con los autores el tipo de empleos a los que acceden las mujeres, es en los sectores de comercio y servicios personales, mientras que los hombres en actividades agrícolas, ganaderas, silvícolas y de caza y pesca; siendo el punto de convergencia los trabajadores industriales, artesanos y ayudantes. Sin embargo, en el sector de universitarios las brechas de salario o puestos se reducen sustancialmente (SALAS DURAZO; MURILLO GARCÍA, 2013). También de acuerdo con este análisis las mujeres que cuentan con estudios medio superiores o superiores disminuían drásticamente las brechas de género, ya que sus condiciones laborales (salario, puesto) estaban situadas por encima de los hombres. Sin embargo, es importante señalar que la mayoría de estas mujeres provenían de estratos económicos alto y medio alto. Además del sexo y los contextos socioeconómicos, también el tipo de carrera afecta las condiciones laborales de quienes egresan. A partir de un estudio de egresados realizado en la Benemérita Universidad Autónoma de Puebla por De Vries, Cabrera y Queen (2008), observaron que todos los indicadores para los egresados de carreras tradicionales (como derecho o contaduría) son mejores en comparación con carreras científicas (como física o bilogía) o novedosas (como administración turística o relaciones internacionales). Los autores explican: el mercado laboral ofrece más lugares para egresados de carreras tradicionales que para los demás, con puestos claramente definidos. Los egresados de carreras novedosas se pueden enfrentar con un mercado donde su especialidad no existe, lo que les obligue a ocupar puestos que no coinciden con sus estudios. En el caso de las carreras científicas, es posible que este mercado siempre haya sido muy limitado. Los resultados reportados por Sánchez Olavarría (2013) indican que las diferencias en el mercado laboral también se relacionan con el tipo de carrera estudiada y la región donde buscan trabajo los egresados. En su estudio con egresados de la licenciatura en comunicación de la Universidad del Altiplano (Tlaxcala), identificó que los comunicadores acceden al © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [72] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 empleo por la vía relacional, que la obtención del título no es prioritaria al momento de solicitar empleo y que los principales problemas de ingreso al mercado laboral son la falta de experiencia y la saturación del campo de trabajo del comunicador. De acuerdo con el autor, la falta de vinculación entre la universidad y el mercado de trabajo orilla a los jóvenes al subempleo en condiciones precarias y vulnerables. En cuanto a la participación de las mujeres como fuerza laboral se ha incrementado por medio de la obtención de mejores puestos de trabajo, sin embargo, estos no se han sostenido debido a que priorizan su participación en actividades propias de la familia como la manutención de los hijos (SÁNCHEZ OLAVARRÍA, 2013). Además de los factores antes descritos, el tiempo de egreso juega un papel importante en las condiciones laborales de los universitarios. De acuerdo con Sánchez Olavarría (2014), los profesionales de la comunicación de recién egreso tienden a buscar colocarse en espacios laborales acordes con su formación, pero conforme avanzan en su trayectoria esta tendencia decrece, debido a la dificultad de insertarse en el campo laboral, lo que los orilla a buscar alternativas que frecuentemente se ubican en los ámbitos indirectamente relacionados o no relacionados con su profesión. Pero el tiempo de experiencia no sólo afecta lo que hacen los egresados, sino también las condiciones en que trabajan. De Vries, Vázquez Cabrera y Ríos Treto (2013) identificaron que el ingreso es mejor conforme las personas cuentan con más experiencia. En concreto, los estudios previos nos indican que hay más hombres que mujeres estudiando y egresando de las licenciaturas, que una buena parte los estudiantes trabajó durante su estancia en la universidad y que las condiciones laborales de quienes egresaron dependen del lugar donde se estudió, de la carrera cursada, del sexo de la persona y de la experiencia laboral. DESCRIPCIÓN DEL CASO Y METODOLOGÍA El CUVT es una de las unidades desconcentradas de la Universidad Autónoma del Estado de México (UAEMEX) fundado en el año 2000 como parte del plan maestro de desconcentración. El CUVT está instalado en la comunidad de Santo Domingo Aztacameca, que pertenece al municipio de Axapusco, ubicado al norte del Estado de México y que forma parte de la Zona Metropolitana del Valle de México (ZMVM)3. El municipio se caracteriza por no tener asentamientos urbanos, las localidades de Axapusco son pueblos y rancherías. Santo Domingo Aztacameca cuenta con 3,012 habitantes y está a 6.9 kilómetros de la cabecera municipal. En la Figura 1 se puede apreciar como Axapusco, el municipio donde se ubica del CUVT pertenece a un espacio periurbano. 3 La ZMVM está conformada por las 16 delegaciones de la Ciudad de México (antes Distrito Federal), 59 municipios de la zona norte del Estado de México y un municipio del estado de Hidalgo. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [73] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 Figura 1. Ubicación del municipio de Axapusco en la Zona Metropolitana del Valle de México Fuente: Imagen tomada del sitio: http://goo.gl/0Htsh1 Simbología: Gris: Centros urbanos de la ZMVM. El término periurbano se utiliza para caracterizar a una extensión continua de la ciudad que absorbe paulatinamente los espacios rurales circundantes y que, de acuerdo con De Mattos (2006), tiene la forma de un archipiélago. Como se puede apreciar en la Figura 1 la concentración de centros urbanos se desvanece conforme los municipios se alejan de la Ciudad de México, los centros urbanos cada vez son menos, más pequeños y se encuentran más alejados unos de otros. Esta disminución de centros urbanos implica menos población, y donde hay menos población hay menos acceso a oportunidades y servicios: empleo, educación, transporte. La escuela inició con una matrícula de 140 alumnos y una planta docente de 18 profesores. Desde el 2012 el CUVT ofrece seis programas de licenciatura (Contaduría, Derecho, Informática Administrativa, Ingeniería en Computación, Psicología y Turismo), y contaba con una población estudiantil de 1,101 alumnos, de los cuales 738 recibían algún tipo de beca. Para ese mismo año el CUVT contaba con 83 profesores (14 de tiempo completo, 68 de asignatura y un técnico académico de tiempo completo) y 25 administrativos (un directivo, 10 de confianza y 14 sindicalizados) (UAEMEX, 2015). El área de influencia del CUVT abarca 21 municpios: 16 del Estado de México y cinco de Hidalgo, como se puede apreciar en la Figura 2. De los 3,848 estudiantes que habían ingresado al Centro hasta el ciclo 20142015, el 54.4% son mujeres y el 45.6% son hombres. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [74] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 Figura 2. Elaboración propia a partir de INEGI, Marco Geoestadístico Municipal, 2005. Fuente: Basado en datos del CUVT, Departamento de Servicios Escolares. La información que se presenta en este reporte se generó a partir de los datos ofrecidos por el Sistema Institucional de Seguimiento de Egresados (SISE) de la UAEMEX. Este sistema funcionó en su primera versión hasta el 2013. Actualmente funciona una nueva versión la cual sólo permite recuperar información de los ciclos 2014-2015 y 2015-2016, es por esta razón que se decidió trabajar con los datos que el sistema generó hasta 2013, pues la nueva versión genera información que en algunos casos no puede compararse con los datos previamente recopilados. El SISE aplicaba un cuestionario de 127 reactivos, de los cuales se recuperó la información que describe la situación socioeconómica de los encuestados. En particular se trabajó con los reactivos sobre la situación laboral y la dependencia económica en dos momentos: durante la licenciatura y al término de los estudios. El análisis de los datos generados por el SISE tiene un enfoque cercano a la encuesta longitudinal retrospectiva, un recurso técnico de análisis de datos que permite establecer nexos entre períodos, propuesta por Casal et al. (2006). Esta propuesta parte de la idea de que los jóvenes están ubicados en el interior de una estratificación social y económica derivada de sus padres y que ésta configura delimitaciones en los campos de maniobra o de construcción de expectativas de posición social. Estos autores conciben a la juventud como un tramo de biografía, que se inicia con la pubertad y que tiende a concluir en la emancipación familiar plena, lo que implica vivir fuera de la casa paterna y ser independiente económicamente (CASAL et al., 2006). La información que aquí se presenta es resultado de los reportes que el mismo sistema generaba con base en las respuestas al cuestionario que se aplica a los estudiantes cuando concluyen sus créditos y comienzan sus trámites de titulación. Este detalle es importante, pues el cuestionario difícilmente puede dar cuenta de qué pasa con las trayectorias laborales de los egresados, pues acaban de terminar sus estudios y muchos de ellos aún no trabajan. Otro detalle importante es que la versión anterior del SISE no permite consultar datos de © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [75] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 identificación de los estudiantes, como el grado de escolaridad de los padres o el municipio en el que residen. El número total de egresados registrados por el departamento de control escolar del CUVT hasta el 2013 fue de 2,106 de los cuáles 57.3% son mujeres. Del total de egresados sólo 1,252 contestaron la encuesta del SISE. La edad promedio de los hombres es de 24.15 y la de las mujeres 23.74. Aquí es importante señalar que la tasa de respuestas a los reactivos del SISE oscila entre 1,084 y 1,252, es por ello que al ver los totales, no siempre aparecen las mismas cantidades. Finalmente, dado que no se tiene acceso a los datos crudos, la información que en este apartado se muestra sólo tiene un carácter descriptivo. A continuación se presentan los resultados del análisis de los datos recuperados del SISE, comparando la distribución de los egresados en dos momentos: mientras estudiaban la licenciatura y al término de sus estudios. Las variables con las que se realizó el análisis fueron: el tipo de bachillerato de procedencia, la posesión de auto, con quién vivían, de quién dependían económicamente, número de egresados que trabajan, relación del empleo con la carrera estudiada e ingresos por salarios mínimos. RESULTADOS. PROCESOS DE INDEPENDENCIA ECONÓMICA E INGRESO AL MERCADO LABORAL EN ESTUDIANTES Y EGRESADOS UNIVERSITARIOS Una característica común en casi todos los egresados es el tipo de escuela de educación media superior de la que provienen: de acuerdo con el SISE el 93% provienen de escuela pública. Otra variable que puede arrojar luz sobre el nivel socioeconómico de los estudiantes y sus familias es la posesión de auto. Los datos del SISE indican que mientras estudiaban, sólo el 5% de los egresados tenían auto y sólo el 3% al egreso. Esta información sugiere que casi la totalidad de los egresados proviene de familias de clase media o baja. El primer aspecto a comparar es con quién vivían los egresados mientras estudiaban y con quién vivían al egresar. En la Tabla 1 se observa que el 90% vivía con sus padres y sólo un porcentaje pequeño vivía con la o el cónyuge o con familiares. ¿Con quién vivía? Padres Durante la carrera 1,145 Hermanos Familiares Cónyuge Solo Amigos Totales 14 39 48 5 1 1,252 Tabla 1. Con quién vivían % Al término de los estudios 91.4 961 1.1 3.1 3.8 0.3 0.0 99.7 18 38 144 22 0 1,183 % 81.2 1.5 3.2 12.6 1.8 0.0 100.3 Fuente: Elaboración propia a partir de datos tomados del “Sistema Institucional de Seguimiento de Egresados”, UAEMEX, 2013. En la Tabla 1 se puede se puede ver que la distribución cambió al término de los estudios, a pesar de que aún la mayoría vive con los padres, un 12% ya vive en pareja y un 2% vive solo. Estas diferencias pueden sugerir que el estudio de la carrera implica procesos © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [76] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 de independencia respecto a la familia, ya sea a través de la relación de pareja o de la vida individual. El siguiente aspecto analizado es la dependencia económica de los egresados mientras estudiaban y al término de los créditos. En este caso se muestran dos análisis, uno general y otro por carrera y género. Para la dependencia económica se hizo este análisis porque los datos muestran diferencias entre los grupos. En la Tabla 2 se puede ver que un 82% dependía económicamente de sus padres, 9% de sí mismo y 4% de la beca. Tabla 2. De quién dependían económicamente los egresados ¿De quién dependías económicamente? De mí mismo De mis padres De hermanos De familiares De mi cónyuge De beca Total Mientras estudiaban 113 1,025 15 17 26 51 1,247 % 9.0 82.1 1.2 1.3 2.0 4.0 99.6 Al término de los estudios 396 646 12 15 100 0 1,169 % 33.8 55.2 1.0 1.2 8.5 0.0 99.7 Fuente: Elaboración propia a partir de datos tomados del “Sistema Institucional de Seguimiento de Egresados”, UAEMEX, 2013. Al término de los estudios se notan algunos cambios, como lo muestra la Tabla 2 donde la dependencia de los padres disminuye un 26.9 %, el porcentaje de quienes se mantienen a sí mismos aumenta un 24.8 % y quienes depende de la o el cónyuge alcanza un 8.5%. Si sumamos un 33.8 % de quienes son autosuficientes más un 8.5 % de quienes dependen del o la cónyuge, se puede ver que un poco más del 40 % son independientes económicamente de los padres. Estos datos refuerzan la idea antes planteada, de que el tránsito por la universidad, en tres de cada 10 egresados, también implica un proceso de independencia económica de la familia, ya sea en pareja o de forma individual. En la Tabla 2 también destaca que algunos dependían económicamente de la beca: 51 casos, lo que representaba el 4% del total. Sin embargo, ¿hay diferencias entre carreras y entre género? En la Figura 3, Gráfico “A”, se puede ver que mientras estudiaban las diferencias por carrera y género casi son imperceptibles, casi todos dependen de los padres. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [77] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 Figura 3. Dependencia económica de los egresados por carrera y género Fuente: Elaboración propia a partir de datos del Sistema Institucional de Seguimiento de Egresados (UAEMEX, 2013). Carreras: ICO-Ingeniería en Computación; LD-Derecho; LCN-Contaduría; LIA-Informática Administrativa; LPS-Psicología. Sexo: H-Hombre; M-Mujer. Gráfico A, N=1,247; Gráfico B, N=1,169. Sin embargo, el panorama cambia al término de los estudios: en todos los casos aumenta el porcentaje de personas que dependen de sí mismas y en todas las carreras el porcentaje de hombres que ya no dependen de sus padres es mayor que en las mujeres, incluso en las carreras donde hay más mujeres que hombres, como en Psicología donde un 85% son mujeres. Otro dato que llama la atención es que el número de mujeres que dependen económicamente del cónyuge aumentó en todos los casos, pero a pesar de eso, en todas las carreras el porcentaje de mujeres que es autosuficiente aumentó, además las mujeres que son autosuficientes son más que las que dependen de su cónyuge (Figura 3, Gráfico B). Finalmente, en la Figura 1 Gráfico “B” se puede observar que al término de los estudios, en las carreras de Contaduría y Psicología, los hombres que son autosuficientes económicamente son más que aquellos que dependen de sus padres, y que en Ingeniería en Computación y Derecho, las cifras se distribuyen casi a la mitad. El único caso que muestra diferencias más marcadas entre quienes dependen de los padres y de sí mismos es Informática Administrativa, donde un 55% aún depende de los padres y un 37% depende de sí mismo. Estas cifras quizá indican que hay un ingreso diferenciado al mercado laboral por carrera y por sexo. El siguiente punto a analizar es el número de egresados que trabajaban mientras acudían a la universidad y al término de los estudios. Los datos del SISE indican que el porcentaje de egresados que trabajaba durante los estudios era de un 30% (480 personas) y un 41% cuando salió de la escuela, pero al igual que en la dependencia económica, las cifras muestran diferencias por carrera y género, como lo muestra la Figura 4. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [78] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 Figura 4. Distribución de quienes trabajaban por carrera y género Fuente: Elaboración propia a partir de datos tomados del “Sistema Institucional de Seguimiento de Egresados”, UAEMEX, 2013. Gráfico A, N=1,170; Gráfico B, N=1,247. Una de las primeras cosas que se puede ver son las diferencias por género, donde en todos los casos el número de hombres que trabajó mientras estudiaba es mayor que el porcentaje de mujeres. Estos datos coinciden con lo reportado por Solís y Blanco (2014) en jóvenes en la Ciudad de México, donde observaron un ingreso más temprano al mercado de trabajo por parte de los hombres en comparación con las mujeres. Otro dato que resalta son los casos de Contaduría e Informática Administrativa, donde casi la mitad de los hombres trabajó durante sus estudios universitarios. En la Figura 4, Gráfica “B”, se puede ver que las cifras cambian al término de los estudios, en todos los casos, tanto en hombres como mujeres, aumenta la frecuencia de quienes trabajaban. Estos datos pueden reforzar la idea de que los estudios universitarios contribuyen a la independencia económica de los hijos, aunque al igual que cuando estudiaban, es mayor el número de los hombres que trabajaban en relación con las mujeres. Otro dato que llama la atención es que a pesar de que en cuatro de las cinco carreras hay más egresadas que egresados, el número de hombres que reportó trabajar al término de los estudios es mayor al de las mujeres: 265 hombres y 210 mujeres. La pregunta obvia al mirar estos datos es: ¿dónde trabajan los egresados? Los datos arrojados por el SISE no permiten responder con precisión a esta pregunta, pero sí indican que de las 480 personas que respondieron haber estado trabajando al egreso, el 66% contestó que tiene un empleo que se relaciona con la carrera que estudió. Al igual que en los datos previos, estas cifras generales muestran diferencias por carrera y género como se puede ver en la Figura 5. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [79] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 Figura 5. Relación del empleo con los estudios por carrera y sexo Fuente: Elaboración propia a partir de datos tomados del “Sistema Institucional de Seguimiento de Egresados”, UAEMEX, 2013. N=480. Una de las primeras cifras que llaman la atención es que, a pesar de que hay más egresados que trabajan, la gran mayoría de las egresadas de Derecho, Contaduría e Informática Administrativa tiene un empleo relacionado con sus estudios universitarios. Quizá estos datos indican que las mujeres tienen más probabilidad de ejercer un empleo relacionado con su carrera que los hombres. En particular resaltan los casos de Contaduría y Derecho, donde a pesar de que hay menor número de mujeres egresadas que reportaron trabajar, el porcentaje de ellas que ejerce su carrera rebasa al de los hombres. De nuevo los datos muestran un ingreso al mercado laboral, en este caso al mercado laboral relacionado con los estudios cursados, diferenciado por género y carrera. En resumen, lo que estos datos sugieren es que en general los hombres inician su vida laboral antes que las mujeres, pero que las mujeres tienen más probabilidad de conseguir un empleo relacionado con su carrera al término de sus estudios. En lo que respecta a los ingresos, la mayoría de los egresados (39%) gana entre tres y cinco salarios mínimos; el 20% percibe entre cinco y diez salarios; el 10 % entre 10 y 15; el 18% menos de tres; el 7% entre 15 y 20; y el 6% entre gana más de 20 salarios mínimos. En general las diferencias entre los ingresos de hombres y mujeres no son muy marcadas, sin embargo los mejores ingresos para los varones muestran ligeras diferencias en los rangos de salarios más altos. Para cerrar este apartado se muestra en la Figura 3 la distribución de quienes egresaron del CUVT por el número de salarios mínimos, sexo y carrera. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [80] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 Figura 6. Distribución por número de salarios mínimos por sexo y carrera Fuente: Elaboración propia a partir de datos tomados del “Sistema Institucional de Seguimiento de Egresados”, UAEMEX, 2013. N=334. La Figura 6 muestra que en tres carreras hay más hombres que mujeres trabajando: ICO, LDE y LIA. En Contaduría el número de hombres y mujeres es el mismo, y en Psicología, dado que hay un mayor número de mujeres, el número de ellas que trabajan es mayor al de los hombres. En lo que respecta al número de salarios percibidos, sólo en los casos de Contaduría y Psicología hay más mujeres que hombres con menos de tres salarios; la mayoría, en todas las carreras, se encuentra en el rango de tres a cinco salarios. La distribución de egresados y egresadas en los rangos de mayores ingresos no presenta una tendencia clara. Por último, llama la atención el caso de Contaduría, en el que la distribución por sexo es muy similar, tanto en el número de egresados como en la manera como se distribuyen entre los distintos rangos de ingreso. Con esta información se cierra el apartado de descripción de los resultados, que en general nos muestra que quienes egresaron del CUVT provienen de familias de nivel socioeconómico medio y bajo y que a su paso por los estudios universitarios han conseguido cierta independencia económica de los padres. Además los datos sugieren un ingreso al mercado de trabajo diferenciado por carrera y sexo. CONCLUSIONES Y ALGUNAS HIPÓTESIS El principal objetivo de este trabajo fue describir las condiciones socioeconómicas y laborales de los egresados del CUVT, durante los estudios y al término de los mismos, con el fin de generar algunas hipótesis acerca de los procesos de independencia familiar que se dan en este periodo. Sin embargo, a lo largo del análisis se encontraron datos que sugieren ciertas tendencias diferenciadas por sexo y carrera, en lo que refiere al término de los estudios y las condiciones laborales. Es por ello que en este apartado se discuten en primer lugar estas tendencias encontradas, para después regresar a las hipótesis sobre los procesos de independencia económica. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [81] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 En lo que refiere al ingreso al mercado laboral, el análisis indicó que hubo más hombres que mujeres que trabajaban durante la carrera. Esta tendencia permanece al término de la licenciatura donde a pesar de que hay más mujeres que hombres (tanto al ingreso como al egreso), el número de hombres que reportó trabajar es mayor en todas las carreras (a excepción de Psicología). Lo cual sugiere que los hombres ingresan al mercado laboral antes que las mujeres y que tienen más probabilidades de encontrar empleo al término de la universidad. Además los datos indican que la mayoría de los encuestados fueron dependientes económicos de sus padres mientras estudiaban y continuaban siéndolo al término de la universidad. Algunas de estas tendencias contrastan con los resultados de las investigaciones revisadas, por ejemplo el estudio realizado por Planas Coll (2013) en el que la mayoría de los estudiantes trabajó mientras acudía a la universidad. Otro estudio que muestra resultados distintos es el realizado en el CU Los Altos donde la mayoría de los egresados de la carrera de Contaduría realizó un trabajo que les permitió ser autosuficientes económicamente mientras estudiaban (GONZÁLEZ PÉREZ; GONZÁLEZ FRANCO, 2015). Algo distinto ocurrió con quienes egresaron de psicología de la misma institución, donde sólo el 16% logró tener independencia económica durante la licenciatura (GONZÁLEZ ANAYA; CARRILLO TORRES, 2016). Las diferencias en las condiciones laborales también se ven en el salario. Aunque las diferencias entre hombres y mujeres encontradas en el análisis no son muy marcadas, la tendencia es que haya más hombres en los rangos de salarios altos y más mujeres en los rangos de salarios bajos. Las investigaciones revisadas indican que las condiciones laborales de quienes egresan de la licenciatura no sólo presentan diferencias entre carreras, también entre sexos, en promedio los hombres ganan más que las mujeres, pero no ocurre lo mismo en todas las carreras (PÉREZ PULIDO; MORENO GARCÍA; CALDERA MONTES, 2011). Además algunos estudios indican que el mayor tiempo de experiencia laboral puede mejorar las condiciones de trabajo de las mujeres (PLANAS COLL; ENCISO ÁVILA, 2013; SÁNCHEZ OLAVARRÍA, 2012). Una posible explicación a estas diferencias es la que ofrece Corbett (2005), que señala que la importancia de las credenciales educativas cambia entre hombres y mujeres, de tal forma que para las mujeres la educación formal es la única ocupación en un mercado diseñado para los hombres, y el pasaporte para salir de la región. De acuerdo con este autor hay un ingreso diferenciado al mercado laboral por género: los hombres que se quedan en la región suelen ocuparse en empleos que requieren baja capacitación; en cambio las mujeres que no salen de la región trabajan en puestos que exigen un mínimo de credenciales educativas (cajeras o empleadas). Pero además las condiciones laborales parecen también determinar los procesos de independencia económica y familiar. Es cierto que para cuando los estudiantes terminaron la carrera el número de quienes dependían económicamente de sus padres disminuyó, pero los procesos fueron distintos para hombres y mujeres. Mientras que sólo dos hombres dependían económicamente de su pareja, 98 mujeres estaban en la misma situación. Además si comparamos el número de quienes son autosuficientes financieramente, se puede identificar que mientras 5 de cada 10 hombres son autosuficientes, sólo 2 de cada 10 lo son en el grupo de mujeres. Sin embargo, las cifras también muestran que hay más egresadas que dependen © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [82] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 económicamente de sí mismas en comparación con aquellas que dependen de su cónyuge: el 67% de las mujeres que no depende económicamente de sus padres son autosuficientes. En lo que refiere a la dependencia económica de quienes egresaron del CUVT contrastan un poco con el dato de con quién viven, pues aunque el porcentaje de quien dependía financieramente de sus padres disminuyó, aun la gran mayoría sigue viviendo con ellos (el 81%). Lo que este dato puede indicar es que a lo largo de la licenciatura los estudiantes se hicieron más independientes económicamente de sus padres, pero que aún dependen de ellos, incluso aquellos que han formado una nueva familia. En la variable de quien vive solo, el cambio en términos porcentuales no parece muy grande, pues pasó del 1% al 2%. Aunque en general la mayoría de quienes reportaron estar trabajando al término de sus estudios lo hacían en alguna actividad relacionada con la licenciatura cursada, hay algunas diferencias por carrera. Por ejemplo, en la carrera de Contaduría casi el 90% trabaja en algo relacionado con los estudios. Algo similar pasa en Derecho e Informática Administrativa, donde alrededor del 80% ejerce su profesión. El único caso que no presenta esta tendencia es Psicología, pues sólo un poco más de la mitad trabaja en algo relacionado con su profesión. Estos resultados contrastan con los obtenidos en otros estudios donde los análisis indican que los egresados pueden conseguir empleos mal remunerados y no relacionados con sus estudios universitarios (GONZÁLEZ ANAYA; CARRILLO TORRES, 2016, RUBIO HERNÁNDEZ; SALGADO VEGA, 2014, BURGOS FLORES; LÓPEZ MONTES, 2010), aunque en el caso de los egresados de Contaduría del CU Los Altos, la mayoría encontró puestos relacionados con la licenciatura que estudiaron (GONZÁLEZ PÉREZ; GONZÁLEZ FRANCO, 2015), lo cual coincide con lo hallado en el presente estudio. Estas conclusiones parciales se ajustan a lo propuesto por De Vries y Navarro (2011), a partir de los datos del proyecto Proflex: hay problemas de desempleo, pero la inserción exitosa (medida en términos de empleo y de sueldo) es el resultado de una complicada interacción entre el desarrollo de la economía, el área de conocimiento, el tipo de institución (privada-pública), el género y el pasado familiar. Para finalizar y con base en lo que indican los estudios previos y el análisis realizado para este documento se sugieren algunas hipótesis:  Hay una forma diferenciada de ingreso al mercado laboral por sexo. Quienes más se benefician de esta diferenciación son los hombres, pues inician primero su experiencia laboral y suelen ser favorecidos con mejores salarios. Esta afirmación aplica tanto para estudiantes como egresados. En coincidencia con lo que propone Sendón (2013), se puede decir que la democratización educativa para las mujeres está mucho más lograda que su incorporación igualitaria al sistema productivo. La escuela, aunque a largo plazo puede tener un poder transformador determinante, y ser un espacio donde se transita aparentemente de manera igualitaria, siempre y cuando la igualdad no se traduzca en los ingresos o en las posiciones más importantes en el mundo laboral. El sistema ha cedido a la demanda de mayor acceso a la educación, siempre cuando esta © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [83] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674     acumulación de niveles educativos no surtan los mismos efectos para todos en el campo productivo (SENDÓN, 2013, p. 26). Hay una diferenciación en el ingreso al mercado laboral por carrera, quienes estudian carreras tradicionales, como Derecho y Contaduría (en el caso del CUVT), suelen tener mejores condiciones laborales. De Vries, Cabrera y Queen (2008) sugieren: los egresados de carreras tradicionales tienen más posibilidades de encontrar trabajo en el sector público y si se considera que el sector público consta básicamente de tres subsectores (educación, salud y la burocracia estatal), es de esperar que este mercado sea más accesible para el grupo tradicional que para el sector novedoso. Además, las carreras relacionadas con la tecnología están más orientadas al sector privado y más sujeto a los vaivenes del mercado. Finalmente, el grupo tradicional tiene mayores posibilidades de asociarse a una pequeña empresa o abrir un despacho o consultorio, lo que les convierte en dueño o socio, o en profesionista independiente y quizá también por ello muestran tasas más altas de congruencia entre estudios y empleo. Estas formas diferenciadas de ingreso al mercado laboral pueden sugerir cierta coincidencia con lo hallado por Jiménez Vásquez (2011), la existencia de un mercado de trabajo segmentado y estratificado, en este caso por sexo y carrera. El paso de los jóvenes por la universidad puede contribuir a su proceso de independencia económica, pero este tránsito es especialmente importante para las mujeres, pues la mayoría de las que tienen un empleo al término de los estudios trabaja en actividades relacionadas con su carrera, lo cual no sería posible de no haber cursado estudios superiores. Además, la mayoría de las mujeres que reportaron no depender económicamente de sus padres y ser autosuficientes es mayor al número de quienes dependen del cónyuge. Quizá estos datos sugieran un proceso de empoderamiento por parte de las mujeres a partir de sus estudios universitarios, lo cual concuerda con algunos estudios previos (KAUSHIK, KAUSHIK; KAUSHIK, 2006). El proceso de independencia económica no siempre se da de forma conjunta con la independencia familiar, pues aunque aumentó el número de jóvenes que expresaron no depender económicamente de sus padres al término de los estudios, la mayoría aún vive con ellos. Esto puede indicar una fase en el proceso de independencia de los jóvenes respecto de sus familias, pues no hay que perder de vista que quienes contestaron el cuestionario del SISE acaban de terminar los créditos de la carrera. Siguiendo este orden de ideas, se esperaría que al paso del tiempo habría más egresadas y egresados independientes económicamente y que vivan fuera de la casa paterna, pues como proponen Casal et al. (2006), la emancipación familiar plena tiene una concreción en el acceso a una nueva vivienda, aunque hay muchas formas de concreción y también formas de regresión o retorno: hay emancipaciones intermedias donde los padres ayudan mucho económicamente y a veces continúan ejerciendo cierto control sobre espacios y tiempos; hay también retornos al hogar de origen después de un período más o menos largo de ejercicio de la emancipación. Algunas © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [84] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 emancipaciones son muy diferidas y estos jóvenes prolongan el tiempo de residencia en casa de los padres. Una parte puede permanecer por mucho tiempo en casa de los padres (por ejemplo, bajo formas de soltería definitiva) e incluso no pasar por la emancipación domiciliar (CASAL et al., 2006, p.32-33). En conclusión se puede decir que las formas diferenciadas de ingreso al mercado laboral por sexo y carrera, determinan procesos de independencia familiar distintos, donde los hombres tienen más probabilidades de ser autosuficientes económicamente y vivir fuera de la casa de los padres, que las mujeres. En los itinerarios, en los procesos de transición de la juventud a la adultez, hay una diversidad social que es explicada por su proximidad a la estructura social y la segmentación. Los itinerarios varían sustancialmente según historia, territorio y culturas (CASAL et al., 2006). En un medio perirubano (con rasgos similares a los espacios rurales) donde el mercado de trabajo está diseñado para ser aprovechado prioritariamente por hombres con baja capacitación, los estudios superiores pueden funcionar como catalizador del proceso de independencia familiar, particularmente para las mujeres (CORBETT, 2005). El tener una licenciatura amplía las oportunidades de trabajo a las mujeres y con ello la probabilidad de recibir un ingreso suficiente para no depender de sus padres ni de algún hombre. Esta idea permite formular una última hipótesis: las personas que se pueden beneficiar más de una carrera universitaria son aquellas que tienen más desventajas sociales, en este caso concreto: las mujeres. REFERENCIAS BURGOS FLORES, Benjamín; LÓPEZ MONTES, Karla. La situación del mercado laboral de profesionistas. Revista de la Educación Superior, México, v. 39, n.4 [156], p.19-33, 2010. CARRILLO REGALADO, Salvador; RÍOS ALMODÓVAR, Jesús Gerardo. Oferta de trabajo de los estudiantes de la Universidad de Guadalajara y de México: un análisis comparativo. Perfiles Educativos, México, v.36, n.144, p.85-104, 2014. CARRILLO REGALADO, Salvador; RÍOS ALMODÓVAR, Jesús Gerardo. La oferta de trabajo en los estudiantes de licenciatura en México. Revista de la Educación Superior México, v.38, n.3 [151], p.39-56, jul./sep. 2009 CASAL, Joaquim et al. Aportaciones teóricas y metodológicas a la sociología de la juventud desde la perspectiva de la transición. Papers: Revista de Sociología, Barcelona, n.79, p.2148, 2006. CORBETT, Michael. Rural education and out-migration: the case of a coastal community. Canadian Journal of Education, v.28, n.1/2, p. 52-72, 2005. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [85] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 CUEVAS DE LA GARZA, José Fernand; DE IBARROLA NICOLÍN, María. Vidas cruzadas Los estudiantes que trabajan: un análisis de sus aprendizajes. Revista de la Educación Superior, México, ANUIES, Vv.42, n.1 [165], p.125–148, ene./mar. 2013. DE MATTOS, Carlos. Modernización capitalista y transformación metropolitana en América Latina: cinco tendencias constitutivas. In: GERAIGES DE LEMOS, A.; ARROYO, M.; SILVEIRA, M. Silveira, América Latina: cidade, campo e turismo. San Pablo: CLACSO, 2006. p. 7 41-773. Disponible en: <http://bibliotecavirtual.clacso.org.ar/ar/libros/edicion/ lemos/03mattos.pdf> Acceso: 26 de noviembre de 2015. DE VRIES, Wieste; NAVARRO RANGEL, Yadira. ¿Estudias o trabajas? La aportación de la experiencia laboral durante los estudios. In: CONGRESO NACIONAL DE INVESTIGACIÓN EDUCATIVA, 13.: 2015, Chihuahua, México. Anais… Chihuahua, México: Consejo Mexicano de Investigación Educativa, 2015. DE VRIES, Wieste, VÁZQUEZ CABRERA, Rebeca; RÍOS TRETO, David. Millonarios o malparados: ¿de qué depende el éxito de los egresados universitarios? Revista Iberoamericana de Educación Superior, México, v.4, n.9, p. 3-20, 2013. Disponible en: www.redalyc.org/articulo.oa?id=299126789001. Acceso: 26 nov. 2015. DE VRIES, Wietse; NAVARRO, Yadira. ¿Profesionistas del futuro o futuros taxistas? Los egresados universitarios y el mercado laboral en México. Revista Iberoamericana de Educación Superior, México, v.2, n.4, p. 3-27, 2011. Disponible en: <http://ries.universia.net/index.php/ries/article/view/71>. Acceso: 3 oct. 2016. DE VRIES, Wieste et al. Conclusiones a contrapelo. La aportación de distintas carreras universitarias a la satisfacción en el empleo. Revista de la Educación Superior, México, v.37, n.2 [146], p. 67-84, abr./jun. 2008. ENCISO ÁVILA, María Isabel. El origen social de los graduados y la equidad en el acceso a la universidad. Revista de la Educación Superior, México, v.42, n.1 [165], p. 11–29, ene./ mar. 2013. GONZÁLEZ PÉREZ, Cándido y GONZÁLEZ FRANCO, Margarita. Puestos laborales de los egresados de la carrera de Contaduría Pública del Centro Universitario de los Alto. In: CONGRESO NACIONAL DE INVESTIGACIÓN EDUCATIVA, 13.:2015, Chihuahua, México. Anais… Chihuahua, México: Consejo Mexicano de Investigación Educativa, 2015. GONZÁLEZ ANAYA, Ana Gabriela; CARRILLO TORRES, Paulo Alberto. Inserción laboral de los psicólogos egresados del Centro Universitario de Los Altos. Revista Iberoamericana para la Investigación y el Desarrollo Educativo, México, v.6, n.12, 2016. JIMÉNEZ VÁSQUEZ, Josefina. Tendencias y hallazgos en los estudios de trayectoria: una opción metodológica para clasificar el desarrollo laboral. Revista Electrónica de Investigación educativa. México, v.11, n.1, 2011. Disponible en: <http://redie.uabc.mx/vol11no1/contenido-jimenez.html>. Acceso: 6 jul. 2016. KAUSHIK, Surendra; KAUSHIK, Shorav; KAUSHIK, Shobha. How higher education in rural India helps human rights and entrepreneurship. Journal of Asian Economics, n.17, p. 29–34, 2006. Disponible en: doi:10.1016/j.asieco.2006.01.004. Acceso: 6 abr. 2015. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [86] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 PÉREZ PULIDO, Ignacio; MORENO GARCÍA, Hugo; CALDERA MONTES, Juan FRANCISCO. Análisis de la situación laboral de los egresados del Centro Universitario de Los Altos. In: CONGRESO NACIONAL DE INVESTIGACIÓN EDUCATIVA, 10.: 2011, Veracruz, México. Anais… Veracruz, Mexico: Consejo Mexicano de Investigación Educativa, 2011. PLANAS COLL, Jordi. Los itinerarios laborales de los universitarios y la calidad de su inserción profesional. Revista de la Educación Superior, México, v.1, n.16, p. 31–62, ene./ mar. 2013. PLANAS-COLL, Jordi; ENCISO ÁVILA, Isabel-María. Los estudiantes que trabajan: ¿tiene valor profesional el trabajo durante los estudios? Revista Iberoamericana de Educación Superior, México, v.5, n.12, p. 23-45, 2014. Disponible em: <http://ries.universia.net/index.php/ries/article/view/322>. Acceso: 4 de octubre de 2016. RUBIO-HERNÁNDEZ, Lisy; SALGADO VEGA, María Del Carmen. Características del proceso de transición al mercado laboral. Caso de los licenciados en economía del Estado de México, 2000-2010. CPU.e.: Revista de Investigación Educativa, México, n.19, p 28-59, 2014. SALAS DURAZO, Iván Alejandro; MURILLO GARCÍA, Favio. Los profesionistas universitarios y el mercado laboral mexicano: convergencias y asimetrías. Revista de la Educación Superior, México, v.42, n.1 [165], p. 63-81, ene./mar. 2013. SÁNCHEZ OLAVARRÍA, César. Los egresados de comunicación y el mercado laboral: un estudio de trayectorias profesionales. Revista Iberoamericana de Educación Superior, v.5, n.13, p. 40-54, 2014. Disponible en: http://ries.universia.net/index.php/ries/article/view/316. Acceso: 14 de octubre, 2016. SÁNCHEZ OLAVARRÍA, César. La inserción laboral de los comunicadores de la Universidad del Altiplano. Revista de la Educación Superior, México, v.42, n.1 [165], p. 105-123, ene./mar. 2013. SÁNCHEZ OLAVARRÍA, César. La movilidad profesional y generacional del comunicador de la Universidad del Altiplano. Revista Electrónica de Investigación educativa, México, v.14, n.2, p. 150-165, 2012. Disponible en: <http://redie.uabc.mx/vol14no2/contenidoschzolavarria2012.html>. Acceso: 14 ago. 2016. SEGUNDO RAMÍREZ, Mario Enrique. Inserción al mercado laboral de los profesionistas de la UACJ: desde el enfoque de la política de ampliación de cobertura. Tesis para obtener el grado de Maestría en Gobierno y Asuntos Públicos. FLACSO-México, 2009. SENDÓN, María Alejandra. Educación y trabajo: Consideraciones actuales en torno al debate del papel de la educación. Propuesta Educativa, Argentina, año 22, v.2, n.40, p.8-31, nov. 2013. SOLÍS, Patricio y BLANCO, EMILIO. ¿Relación duradera o divorcio? El vínculo entre la escolaridad y el logro ocupacional temprano en un contexto de deterioro laboral. In: BLANCO, E., SOLÍS, P.; ROBLES, H. Caminos desiguales: trayectorias educativas y © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [87] p.66-89 jan./abr. 2017 ISSN 2446-9424 DOI ARTIGO 10.22348/riesup.v0i0.7674 laborales de los jóvenes en la Ciudad de México. México: Colegio de México, 2014. p.107129. UNIVERSIDAD AUTÓNOMA DEL ESTADO DE MÉXICO. Agenda estadística 2015. Disponible en: 2015. En: <http://www.uaemex.mx/planeacion/Agenda.html>. Acceso: 25 mayo 2015. AGRADECIMIENTOS Al Consejo Nacional de Ciencia y Tecnología (Conacyt), por el apoyo para realización de este artículo. A Belem Carbajal Barroso, responsable del Departamento de Servicios Escolares del CUVT, por su apoyo y asesoría. © Rev. Inter. Educ. Sup. Campinas, SP v.3 n.1 [88] p.66-89 jan./abr. 2017 ISSN 2446-9424
https://openalex.org/W4230163523	https://peerj.com/articles/11814v0.2/submission	English	null	Peer Review #2 of "The importance of spawning behavior in understanding the vulnerability of exploited marine fishes in the U.S. Gulf of Mexico (v0.2)"	null	2,021	cc-by	32,922	The importance of spawning behavior in understanding the vulnerability of exploited marine ﬁshes in the U.S. Gulf of Mexico Christopher R. Biggs 1 , William D Heyman 2 , Nicholas A Farmer 3 , Shin'ichi Kobara 4 , Derek G Bolser 1 , Jan Robinson 5 , Susan K Lowerre-Barbieri 6 , Brad E Erisman Corresp. 1, 7 Christopher R. Biggs 1 , William D Heyman 2 , Nicholas A Farmer 3 , Shin'ichi Kobara 4 , Derek G Bolser 1 , Jan Robinson 5 , Susan K Lowerre-Barbieri 6 , Brad E Erisman Corresp. 1, 7 1 Marine Science Institute, The University of Texas at Austin, Port Aransas, Texas, United States 2 LGL Ecological Research Associates, Inc., Bryan, Texas, United States 3 Southeast Regional Oﬃce, NOAA National Marine Fisheries Service, St. Petersburg, Florida, United States 4 Department of Oceanography, Texas A&M University, College Station, Texas, United States 5 Australian Research Council (ARC) Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, Queensland, Australia 6 Fisheries and Aquatic Science Program, School of Forest Resources and Conservation, University of Florida, Gainesville, Florida, United States 7 Fisheries Resources Division, Southwest Fisheries Science Center, National Marine Fisheries Service, National Oceanic and Atmospheric Administration, La Jolla, California, United States 1 Marine Science Institute, The University of Texas at Austin, Port Aransas, Texas, United States 2 LGL Ecological Research Associates, Inc., Bryan, Texas, United States 3 Southeast Regional Oﬃce, NOAA National Marine Fisheries Service, St. Petersburg, Florida, United States 4 Department of Oceanography, Texas A&M University, College Station, Texas, United States 5 Australian Research Council (ARC) Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, Queensland, Australia 6 Fisheries and Aquatic Science Program, School of Forest Resources and Conservation, University of Florida, Gainesville, Florida, United States 7 Fisheries Resources Division, Southwest Fisheries Science Center, National Marine Fisheries Service, National Oceanic and Atmospheric Administration, Corresponding Author: Brad E Erisman Email address: berisman@utexas.edu The vulnerability of a ﬁsh stock to becoming overﬁshed is dependent upon biological traits that inﬂuence productivity and external factors that determine susceptibility or exposure to ﬁshing eﬀort. While a suite of life history traits are traditionally incorporated into management eﬀorts due to their direct association with vulnerability to overﬁshing, spawning behavioral traits are seldom considered. We synthesized the existing biological and ﬁsheries information of 28 ﬁsh stocks in the U.S. Gulf of Mexico to investigate relationships between life history traits, spawning behavioral traits, management regulations, and vulnerability to ﬁshing during the spawning season. Manuscript to be reviewed PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) The importance of spawning behavior in understanding the vulnerability of exploited marine ﬁshes in the U.S. Gulf of Mexico Our results showed that spawning behavioral traits were not correlated with life history traits but improved identiﬁcation of species that have been historically overﬁshed. Species varied widely in their intrinsic vulnerability to ﬁshing during spawning in association with a broad range of behavioral strategies. Extrinsic vulnerability was high for nearly all species due to exposure to ﬁshing during the spawning season and few management measures in place to protect spawning ﬁsh. Similarly, several species with the highest vulnerability scores were historically overﬁshed in association with spawning aggregations. The most vulnerable species included several stocks that have not been assessed and should be prioritized for further research and monitoring. Collectively, the results of this study illustrate that spawning behavior is a distinct aspect of ﬁsh ecology that is important to consider for predictions of vulnerability and resilience to ﬁsheries exploitation. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 1 The importance of spawning behavior in 2 understanding the vulnerability of exploited marine 3 fishes in the U.S. Gulf of Mexico 4 5 6 Christopher R. Biggs1, William D. Heyman2, Nicholas A. Farmer3, Shinichi Kobara4, Derek G. 7 Bolser1, Jan Robinson5, Susan K. Lowerre-Barbieri6, and Brad E. Erisman1,* 8 9 10 1Department of Marine Science, University of Texas at Austin, Port Aransas, TX, USA 11 2LGL Ecological Research Associates, Inc., Bryan, TX, USA 12 3 NOAA National Marine Fisheries Service, Southeast Regional Office, St. Petersburg, FL, USA 13 4Department of Oceanography, Texas A&M University, College Station, TX, USA 14 5Australian Research Council (ARC) Centre of Excellence for Coral Reef Studies, James Cook 15 University, Townsville, QLD, Australia 16 6Fisheries and Aquatic Science Program, School of Forest Resources and Conservation, 17 University of Florida, Gainesville, FL, USA 18 NOAA National Marine Fisheries Service, Southwest Fisheries Science Center, La Jolla, CA, 19 USA (present address) 20 21 Corresponding Author: 22 Brad E. Erisman1,7 23 NOAA National Marine Fisheries Service, Southwest Fisheries Science Center, La Jolla, CA, 24 USA 25 Email address: brad.erisman@noaa.gov Manuscript to be reviewed Manuscript to be reviewed PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) 51 Introduction 52 2005) and resilience 88 (Lowerre-Barbieri et al. 2017) in marine fishes such that the spatial and temporal components of 89 spawning may affect the relationship between stocks and recruitment, affecting how a species 90 responds to fishing pressure (Maunder and Deriso 2013; Donahue et al. 2015; Erisman et al. 91 2017a). For instance, the number of spawning sites and number of spawning opportunities are 92 positively correlated with increased reproductive resilience (Erisman et al. 2011; Lowerre- 93 Barbieri et al. 2015). The duration of the spawning season is inversely related to vulnerability, in 94 which species with predictable but brief spawning periods are associated with the most rapid and 95 severe population declines compared to those that spawn year-round or over protracted seasons 52 53 The vulnerability of a stock, population, or species of marine fish to become overfished or 54 experience overfishing is dependent upon both intrinsic aspects of its evolutionary history, 55 ecology, and population biology as well as extrinsic factors related to the fishery and its 56 management that determine the level of exposure to fishing pressure (Jennings et al. 1998; Dulvy 57 et al. 2004; Patrick et al. 2010). Intrinsic vulnerability is a function of various life history traits 58 (e.g., growth rate or longevity) and behavioral traits (e.g., spatiotemporal spawning patterns) that 59 influence stock productivity and resilience: the capacity of a fish population to recover once it 60 becomes depleted (Adams 1980; Reynolds et al. 2001; Stobutzki et al. 2001). Conversely, 61 extrinsic vulnerability factors are linked to the dynamics of the fishery (e.g., fishing effort or 62 catch efficiency), the effectiveness of management policies, and governance structure that 63 combine to determine stock susceptibility and the potential for the fishery to negatively impact 64 stock condition (Cinner et al. 2009; Hobday et al. 2011; Leslie et al. 2015). In situations where 65 insufficient information exists to perform quantitative assessments of biomass or modelling of 66 population dynamics, intrinsic and extrinsic attributes associated with productivity of the stock 67 and exposure to fishing pressure can be used to estimate the overall vulnerability of a stock 68 relative to others in the same region (Frisk et al. 2005; Patrick et al. 2010; Hobday et al. 2011). 69 Furthermore, vulnerability analyses and conservation evaluations have been useful in identifying 70 stocks that should be prioritized for additional research and monitoring (Morato et al. 51 Introduction 52 78 For example, fish with slow growth rates tend to have low natural mortalities and late onset of 79 sexual maturity, although there are exceptions (see Coulson 2019). Life history traits (e.g., 80 growth, reproduction, and death rates) are integral to data-limited stock assessments, but they are 81 also used within data-rich stock assessments (e.g., length or age-structured models) when 82 sufficient data are available (Hilborn and Walters 1992; Methot and Wetzel 2013). Likewise, 83 most vulnerability assessments are designed to account for vulnerability associated with life 84 history traits. However, certain types of spawning behaviors and reproductive patterns are also 85 associated with a high intrinsic vulnerability to fishing but are not typically incorporated into 86 assessment frameworks and thus do not account for this source of vulnerability. 87 Spawning behavior is associated with productivity (Cheung et al. 2005) and resilience 88 (Lowerre-Barbieri et al. 2017) in marine fishes such that the spatial and temporal components of 89 spawning may affect the relationship between stocks and recruitment, affecting how a species 90 responds to fishing pressure (Maunder and Deriso 2013; Donahue et al. 2015; Erisman et al. 91 2017a). For instance, the number of spawning sites and number of spawning opportunities are 92 positively correlated with increased reproductive resilience (Erisman et al. 2011; Lowerre- 93 Barbieri et al. 2015). The duration of the spawning season is inversely related to vulnerability, in 94 which species with predictable but brief spawning periods are associated with the most rapid and 95 severe population declines compared to those that spawn year-round or over protracted seasons 96 (Mullon et al. 2005; Claro et al. 2009; Sadovy De Mitcheson and Erisman 2012). Large changes 52 53 The vulnerability of a stock, population, or species of marine fish to become overfished or 54 experience overfishing is dependent upon both intrinsic aspects of its evolutionary history, 55 ecology, and population biology as well as extrinsic factors related to the fishery and its 56 management that determine the level of exposure to fishing pressure (Jennings et al. 1998; Dulvy 57 et al. 2004; Patrick et al. 2010). Intrinsic vulnerability is a function of various life history traits 58 (e.g., growth rate or longevity) and behavioral traits (e.g., spatiotemporal spawning patterns) that 59 influence stock productivity and resilience: the capacity of a fish population to recover once it 60 becomes depleted (Adams 1980; Reynolds et al. 51 Introduction 52 2001; Stobutzki et al. 2001). Conversely, 61 extrinsic vulnerability factors are linked to the dynamics of the fishery (e.g., fishing effort or 62 catch efficiency), the effectiveness of management policies, and governance structure that 63 combine to determine stock susceptibility and the potential for the fishery to negatively impact 64 stock condition (Cinner et al. 2009; Hobday et al. 2011; Leslie et al. 2015). In situations where 65 insufficient information exists to perform quantitative assessments of biomass or modelling of 66 population dynamics, intrinsic and extrinsic attributes associated with productivity of the stock 67 and exposure to fishing pressure can be used to estimate the overall vulnerability of a stock 68 relative to others in the same region (Frisk et al. 2005; Patrick et al. 2010; Hobday et al. 2011). 69 Furthermore, vulnerability analyses and conservation evaluations have been useful in identifying 70 stocks that should be prioritized for additional research and monitoring (Morato et al. 2006; 71 Davies and Baum 2012; Mamauag et al. 2013). 72 A suite of life history traits is associated with a high intrinsic vulnerability to becoming 73 overfished. Fish species that are slow growing, long-lived, late to mature, and experience low 74 natural mortality are consistently linked to reduced resilience and increased risk of population 75 collapse in response to fishing (Jennings et al. 1998; Musick 1999; King and McFarlane 2003; 76 Winemiller 2005). Moreover, certain life history traits correlate with each other as intrinsic 77 indicators of vulnerability or compensatory capacity (Dulvy et al. 2004; Kindsvater et al. 2016). 78 For example, fish with slow growth rates tend to have low natural mortalities and late onset of 79 sexual maturity, although there are exceptions (see Coulson 2019). Life history traits (e.g., 80 growth, reproduction, and death rates) are integral to data-limited stock assessments, but they are 81 also used within data-rich stock assessments (e.g., length or age-structured models) when 82 sufficient data are available (Hilborn and Walters 1992; Methot and Wetzel 2013). Likewise, 83 most vulnerability assessments are designed to account for vulnerability associated with life 84 history traits. However, certain types of spawning behaviors and reproductive patterns are also 85 associated with a high intrinsic vulnerability to fishing but are not typically incorporated into 86 assessment frameworks and thus do not account for this source of vulnerability. 87 Spawning behavior is associated with productivity (Cheung et al. 27 Abstract 28 The vulnerability of a fish stock to becoming overfished is dependent upon biological traits that 29 influence productivity and external factors that determine susceptibility or exposure to fishing 30 effort. While a suite of life history traits are traditionally incorporated into management efforts 31 due to their direct association with vulnerability to overfishing, spawning behavioral traits are 32 seldom considered. We synthesized the existing biological and fisheries information of 28 fish 33 stocks in the U.S. Gulf of Mexico to investigate relationships between life history traits, 34 spawning behavioral traits, management regulations, and vulnerability to fishing during the 35 spawning season. Our results showed that spawning behavioral traits were not correlated with 36 life history traits but improved identification of species that have been historically overfished. 37 Species varied widely in their intrinsic vulnerability to fishing during spawning in association 38 with a broad range of behavioral strategies. Extrinsic vulnerability was high for nearly all species 39 due to exposure to fishing during the spawning season and few management measures in place to 40 protect spawning fish. Similarly, several species with the highest vulnerability scores were 41 historically overfished in association with spawning aggregations. The most vulnerable species 42 included several stocks that have not been assessed and should be prioritized for further research 43 and monitoring. Collectively, the results of this study illustrate that spawning behavior is a 44 distinct aspect of fish ecology that is important to consider for predictions of vulnerability and 45 resilience to fisheries exploitation. 49 PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) 51 Introduction 52 52 53 The vulnerability of a stock, population, or species of marine fish to become overfished or 54 experience overfishing is dependent upon both intrinsic aspects of its evolutionary history, 55 ecology, and population biology as well as extrinsic factors related to the fishery and its 56 management that determine the level of exposure to fishing pressure (Jennings et al. 1998; Dulvy 57 et al. 2004; Patrick et al. 2010). Intrinsic vulnerability is a function of various life history traits 58 (e.g., growth rate or longevity) and behavioral traits (e.g., spatiotemporal spawning patterns) that 59 influence stock productivity and resilience: the capacity of a fish population to recover once it 60 becomes depleted (Adams 1980; Reynolds et al. 2001; Stobutzki et al. 2001). Conversely, 61 extrinsic vulnerability factors are linked to the dynamics of the fishery (e.g., fishing effort or 62 catch efficiency), the effectiveness of management policies, and governance structure that 63 combine to determine stock susceptibility and the potential for the fishery to negatively impact 64 stock condition (Cinner et al. 2009; Hobday et al. 2011; Leslie et al. 2015). In situations where 65 insufficient information exists to perform quantitative assessments of biomass or modelling of 66 population dynamics, intrinsic and extrinsic attributes associated with productivity of the stock 67 and exposure to fishing pressure can be used to estimate the overall vulnerability of a stock 68 relative to others in the same region (Frisk et al. 2005; Patrick et al. 2010; Hobday et al. 2011). 69 Furthermore, vulnerability analyses and conservation evaluations have been useful in identifying 70 stocks that should be prioritized for additional research and monitoring (Morato et al. 2006; 71 Davies and Baum 2012; Mamauag et al. 2013). 72 A suite of life history traits is associated with a high intrinsic vulnerability to becoming 73 overfished. Fish species that are slow growing, long-lived, late to mature, and experience low 74 natural mortality are consistently linked to reduced resilience and increased risk of population 75 collapse in response to fishing (Jennings et al. 1998; Musick 1999; King and McFarlane 2003; 76 Winemiller 2005). Moreover, certain life history traits correlate with each other as intrinsic 77 indicators of vulnerability or compensatory capacity (Dulvy et al. 2004; Kindsvater et al. 2016). Manuscript to be reviewed 100 Catchability is an important factor in fisheries assessments (Hilborn and Walters 1992; Arreguin- 101 Sanchez 1996) but can be difficult to estimate, as it is affected by a combination of extrinsic 102 factors (fishery and management related) and intrinsic factors such as aggregating behavior and 103 changes in relative abundance (Skjold et al. 1996; Solmundsson et al. 2003; Erisman et al. 2011). 103 changes in relative abundance (Skjold et al. 1996; Solmundsson et al. 2003; Erisman et al. 2011). 104 The continuum of adult behavioral dynamics that collectively determine the spatial and 105 temporal characteristics of spawning can be described based on the degree of aggregating 106 behavior, the duration of spawning season, and the change in relative abundance of fish during 107 spawning (Claydon et al. 2014; Lowerre-Barbieri et al. 2017; Erisman et al. 2017b). On one end 108 of the spectrum are transient aggregations, which include individuals that have migrated from 109 within a large catchment area to congregate in high densities at very specific locations during 110 predictable periods (Domeier 2012). On the other end of the spectrum are species that do not 111 aggregate to spawn or exhibit simple migratory behavior, in which an entire group or population 112 moves from a foraging ground to a spawning area without a change in relative abundance 113 (Domeier 2012). Also, within the spectrum of spawning behaviors are resident aggregations that 114 include fishes that form small spawning aggregations of a few to a few dozen individuals, which 115 often occur throughout the year and draw from small catchment areas. Lastly, some populations 116 or species of fish are mixed spawners that employ a combination of resident and transient 117 spawning behaviors in different areas and times (Lowerre-Barbieri et al. 2009; Tinhan et al. 118 2014). 119 Understanding interspecific variations in reproductive behaviors including reproductive 120 migration patterns, changes in relative abundance, and the timing, duration, and spatial 121 distribution of spawning activities may help scientists and managers better understand the 122 intrinsic vulnerability of a species to fishing in relation to spawning and manage for resilience. Manuscript to be reviewed 97 in fish densities or relative abundance in association with spawning are directly linked to marked 98 increases in catchability, which also increases susceptibility and overall vulnerability to fishing 99 (Wilberg et al. 2010; Erisman et al. 2011, 2014; Robinson and Samoilys 2013; Robinson 2015). 100 Catchability is an important factor in fisheries assessments (Hilborn and Walters 1992; Arreguin- 101 Sanchez 1996) but can be difficult to estimate, as it is affected by a combination of extrinsic 102 factors (fishery and management related) and intrinsic factors such as aggregating behavior and 103 changes in relative abundance (Skjold et al. 1996; Solmundsson et al. 2003; Erisman et al. 2011). 104 The continuum of adult behavioral dynamics that collectively determine the spatial and 105 temporal characteristics of spawning can be described based on the degree of aggregating 106 behavior, the duration of spawning season, and the change in relative abundance of fish during 107 spawning (Claydon et al. 2014; Lowerre-Barbieri et al. 2017; Erisman et al. 2017b). On one end 108 of the spectrum are transient aggregations, which include individuals that have migrated from 109 within a large catchment area to congregate in high densities at very specific locations during 110 predictable periods (Domeier 2012). On the other end of the spectrum are species that do not 111 aggregate to spawn or exhibit simple migratory behavior, in which an entire group or population 112 moves from a foraging ground to a spawning area without a change in relative abundance 113 (Domeier 2012). Also, within the spectrum of spawning behaviors are resident aggregations that 114 include fishes that form small spawning aggregations of a few to a few dozen individuals, which 115 often occur throughout the year and draw from small catchment areas. Lastly, some populations 116 or species of fish are mixed spawners that employ a combination of resident and transient 117 spawning behaviors in different areas and times (Lowerre-Barbieri et al. 2009; Tinhan et al. 118 2014). 119 Understanding interspecific variations in reproductive behaviors including reproductive 120 migration patterns, changes in relative abundance, and the timing, duration, and spatial 121 distribution of spawning activities may help scientists and managers better understand the 122 intrinsic vulnerability of a species to fishing in relation to spawning and manage for resilience. 51 Introduction 52 2006; 71 Davies and Baum 2012; Mamauag et al. 2013). ; g ) 72 A suite of life history traits is associated with a high intrinsic vulnerability to becoming 73 overfished. Fish species that are slow growing, long-lived, late to mature, and experience low 74 natural mortality are consistently linked to reduced resilience and increased risk of population 75 collapse in response to fishing (Jennings et al. 1998; Musick 1999; King and McFarlane 2003; 76 Winemiller 2005). Moreover, certain life history traits correlate with each other as intrinsic 77 indicators of vulnerability or compensatory capacity (Dulvy et al. 2004; Kindsvater et al. 2016). 78 For example, fish with slow growth rates tend to have low natural mortalities and late onset of 79 sexual maturity, although there are exceptions (see Coulson 2019). Life history traits (e.g., 80 growth, reproduction, and death rates) are integral to data-limited stock assessments, but they are 81 also used within data-rich stock assessments (e.g., length or age-structured models) when 82 sufficient data are available (Hilborn and Walters 1992; Methot and Wetzel 2013). Likewise, 83 most vulnerability assessments are designed to account for vulnerability associated with life 84 history traits. However, certain types of spawning behaviors and reproductive patterns are also 85 associated with a high intrinsic vulnerability to fishing but are not typically incorporated into 86 assessment frameworks and thus do not account for this source of vulnerability. y 87 Spawning behavior is associated with productivity (Cheung et al. 2005) and resilience 88 (Lowerre-Barbieri et al. 2017) in marine fishes such that the spatial and temporal components of 89 spawning may affect the relationship between stocks and recruitment, affecting how a species 90 responds to fishing pressure (Maunder and Deriso 2013; Donahue et al. 2015; Erisman et al. 91 2017a). For instance, the number of spawning sites and number of spawning opportunities are 92 positively correlated with increased reproductive resilience (Erisman et al. 2011; Lowerre- 93 Barbieri et al. 2015). The duration of the spawning season is inversely related to vulnerability, in 94 which species with predictable but brief spawning periods are associated with the most rapid and 95 severe population declines compared to those that spawn year-round or over protracted seasons 96 (Mullon et al. 2005; Claro et al. 2009; Sadovy De Mitcheson and Erisman 2012). Large changes PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 123 Yet, the reproductive potential of a stock is typically measured based on spawning stock biomass 124 or total egg production (i.e., fecundity), rather than traits affecting reproductive resilience 125 (Lowerre-Barbieri et al., 2017), including spatiotemporal behavioral traits, although spawning 126 season duration is used to estimate annual fecundity in indeterminate species (Cooper et al. 2013; 127 Maunder and Deriso 2013; Ganias et al. 2014). While spawning behavior traits have been 128 considered in some vulnerability analyses (Cheung et al. 2005; Erisman et al. 2014; Robinson 129 and Samoilys 2013; Robinson et al. 2015), this aspect of fish ecology remains poorly studied, 130 under-utilized for assessing stock health, and rarely emphasized in management frameworks (e.g. 131 Erisman et al. 2011; Sadovy De Mitcheson and Erisman 2012; Cheung et al. 2013). 132 Given the influence life history and spawning behavioral traits have on vulnerability, it is 133 important to consider whether these traits are correlated and how well they explain vulnerability 134 to fishing during spawning and exploitation status. If life history and spawning behavior traits 135 are not correlated, then for situations in which spawning behavior improves predictions of 136 vulnerability, increased research efforts to understand spawning behaviors would help identify 137 vulnerable species that may otherwise be overlooked and identify areas where targeted protection 138 of spawning fish may be needed to maintain sustainable harvest levels or rebuild overfished 139 stocks (Grüss and Robinson 2014; Grüss et al. 2018). Studies that have examined the relationship 140 between life history traits and spawning behavior were conducted in tropical regions and did not 141 focus explicitly on exploited species (e.g. Choat 2012; Nemeth 2012). Therefore, uncovering this 142 relationship and identifying species whose vulnerability to fishing is not explained by life history 97 in fish densities or relative abundance in association with spawning are directly linked to marked 98 increases in catchability, which also increases susceptibility and overall vulnerability to fishing 99 (Wilberg et al. 2010; Erisman et al. 2011, 2014; Robinson and Samoilys 2013; Robinson 2015). 100 Catchability is an important factor in fisheries assessments (Hilborn and Walters 1992; Arreguin- 101 Sanchez 1996) but can be difficult to estimate, as it is affected by a combination of extrinsic 102 factors (fishery and management related) and intrinsic factors such as aggregating behavior and 103 changes in relative abundance (Skjold et al. 1996; Solmundsson et al. 2003; Erisman et al. 2011). Manuscript to be reviewed 104 The continuum of adult behavioral dynamics that collectively determine the spatial and 105 temporal characteristics of spawning can be described based on the degree of aggregating 106 behavior, the duration of spawning season, and the change in relative abundance of fish during 107 spawning (Claydon et al. 2014; Lowerre-Barbieri et al. 2017; Erisman et al. 2017b). On one end 108 of the spectrum are transient aggregations, which include individuals that have migrated from 109 within a large catchment area to congregate in high densities at very specific locations during 110 predictable periods (Domeier 2012). On the other end of the spectrum are species that do not 111 aggregate to spawn or exhibit simple migratory behavior, in which an entire group or population 112 moves from a foraging ground to a spawning area without a change in relative abundance 113 (Domeier 2012). Also, within the spectrum of spawning behaviors are resident aggregations that 114 include fishes that form small spawning aggregations of a few to a few dozen individuals, which 115 often occur throughout the year and draw from small catchment areas. Lastly, some populations 116 or species of fish are mixed spawners that employ a combination of resident and transient 117 spawning behaviors in different areas and times (Lowerre-Barbieri et al. 2009; Tinhan et al. 118 2014). 97 in fish densities or relative abundance in association with spawning are directly linked to marked 98 increases in catchability, which also increases susceptibility and overall vulnerability to fishing 99 (Wilberg et al. 2010; Erisman et al. 2011, 2014; Robinson and Samoilys 2013; Robinson 2015). 100 Catchability is an important factor in fisheries assessments (Hilborn and Walters 1992; Arreguin- 101 Sanchez 1996) but can be difficult to estimate, as it is affected by a combination of extrinsic 102 factors (fishery and management related) and intrinsic factors such as aggregating behavior and 103 changes in relative abundance (Skjold et al. 1996; Solmundsson et al. 2003; Erisman et al. 2011). 97 in fish densities or relative abundance in association with spawning are directly linked to marked 98 increases in catchability, which also increases susceptibility and overall vulnerability to fishing 99 (Wilberg et al. 2010; Erisman et al. 2011, 2014; Robinson and Samoilys 2013; Robinson 2015). Manuscript to be reviewed Third, we conducted a vulnerability 166 analysis to identify species and stocks that are likely highly vulnerable during the spawning 167 season and should be prioritized for further research and monitoring. The results indicate 168 whether including spawning behavior characteristics can improve our ability to assess the 169 vulnerability and resilience of marine fishes to exploitation. Additionally, this information will 170 identify important data gaps in our understanding of the spawning behavior of exploited marine 171 fishes and provide the basis for further research on interactions between spawning behavior and 172 fishing activities. This information will be applicable to the management and monitoring of 173 exploited marine fishes in the GOM, and the approach should be transferable to regional 174 vulnerability assessments of fish stocks elsewhere. 175 143 traits alone would assist managers in prioritizing research, monitoring, and assessment efforts 144 accordingly. 43 traits alone would assist managers in prioritizing research, monitoring, and assessment efforts 44 accordingly. g y 145 The U.S. Gulf of Mexico (GOM) presents an ideal opportunity to answer important 146 questions about relationships between life history traits, reproductive behavior, and the 147 vulnerability of exploited marine fishes during the spawning season. The GOM is a highly 148 productive system that supports a diverse set of taxa (i.e., numerous families) of highly exploited 149 fish species that exhibit a wide range of life history strategies and reproductive patterns (Farmer 150 et al. 2016; Biggs et al. 2017). There is extensive information on life history characteristics for 151 most managed species, and the majority of fisheries in state and federal waters rely heavily upon 152 life history data as the basis for assessments of both data-limited and data-rich stocks (Sagarese 153 et al. 2015; SEDAR 2016a). There is also a growing recognition of the need to incorporate 154 reproductive behavior in the conservation and management of these and other exploited fishes in 155 the region (Lowerre-Barbieri et al. 2016; NOAA RESTORE Science Program, Kobara et al., 156 2017; Grüss et al. 2018; Erisman et al. 2018; Heyman et al., 2019). Therefore, it is a good system 157 to compare exploitation history to spawning behavior and evaluate the degree to which more 158 attention to spawning patterns in relation to fishing vulnerability is warranted. Manuscript to be reviewed Additionally, this information will 170 identify important data gaps in our understanding of the spawning behavior of exploited marine 171 fishes and provide the basis for further research on interactions between spawning behavior and 172 fishing activities. This information will be applicable to the management and monitoring of 173 exploited marine fishes in the GOM, and the approach should be transferable to regional 174 vulnerability assessments of fish stocks elsewhere. 143 traits alone would assist managers in prioritizing research, monitoring, and assessment efforts 144 accordingly. 145 The U.S. Gulf of Mexico (GOM) presents an ideal opportunity to answer important 146 questions about relationships between life history traits, reproductive behavior, and the 147 vulnerability of exploited marine fishes during the spawning season. The GOM is a highly 148 productive system that supports a diverse set of taxa (i.e., numerous families) of highly exploited 149 fish species that exhibit a wide range of life history strategies and reproductive patterns (Farmer 150 et al. 2016; Biggs et al. 2017). There is extensive information on life history characteristics for 151 most managed species, and the majority of fisheries in state and federal waters rely heavily upon 152 life history data as the basis for assessments of both data-limited and data-rich stocks (Sagarese 153 et al. 2015; SEDAR 2016a). There is also a growing recognition of the need to incorporate 154 reproductive behavior in the conservation and management of these and other exploited fishes in 155 the region (Lowerre-Barbieri et al. 2016; NOAA RESTORE Science Program, Kobara et al., 156 2017; Grüss et al. 2018; Erisman et al. 2018; Heyman et al., 2019). Therefore, it is a good system 157 to compare exploitation history to spawning behavior and evaluate the degree to which more 158 attention to spawning patterns in relation to fishing vulnerability is warranted. 159 The objective of this study was to investigate relationships between life history and 160 spawning behavioral traits and identify patterns among exploited species in the GOM that have 161 been historically overfished or may be particularly vulnerable to exploitation during the 162 spawning season. We employed multiple methods to evaluate these relationships. First, we used 163 a data synthesis approach to test whether life history and spawning behavioral traits were 164 correlated. Second, we used ordination to identify groups of traits (life history and spawning 165 behavior) that were common among overfished species. Manuscript to be reviewed 143 traits alone would assist managers in prioritizing research, monitoring, and assessment efforts 144 accordingly. 145 The U.S. Gulf of Mexico (GOM) presents an ideal opportunity to answer important 146 questions about relationships between life history traits, reproductive behavior, and the 147 vulnerability of exploited marine fishes during the spawning season. The GOM is a highly 148 productive system that supports a diverse set of taxa (i.e., numerous families) of highly exploited 149 fish species that exhibit a wide range of life history strategies and reproductive patterns (Farmer 150 et al. 2016; Biggs et al. 2017). There is extensive information on life history characteristics for 151 most managed species, and the majority of fisheries in state and federal waters rely heavily upon 152 life history data as the basis for assessments of both data-limited and data-rich stocks (Sagarese 153 et al. 2015; SEDAR 2016a). There is also a growing recognition of the need to incorporate 154 reproductive behavior in the conservation and management of these and other exploited fishes in 155 the region (Lowerre-Barbieri et al. 2016; NOAA RESTORE Science Program, Kobara et al., 156 2017; Grüss et al. 2018; Erisman et al. 2018; Heyman et al., 2019). Therefore, it is a good system 157 to compare exploitation history to spawning behavior and evaluate the degree to which more 158 attention to spawning patterns in relation to fishing vulnerability is warranted. 159 The objective of this study was to investigate relationships between life history and 160 spawning behavioral traits and identify patterns among exploited species in the GOM that have 161 been historically overfished or may be particularly vulnerable to exploitation during the 162 spawning season. We employed multiple methods to evaluate these relationships. First, we used 163 a data synthesis approach to test whether life history and spawning behavioral traits were 164 correlated. Second, we used ordination to identify groups of traits (life history and spawning 165 behavior) that were common among overfished species. Third, we conducted a vulnerability 166 analysis to identify species and stocks that are likely highly vulnerable during the spawning 167 season and should be prioritized for further research and monitoring. The results indicate 168 whether including spawning behavior characteristics can improve our ability to assess the 169 vulnerability and resilience of marine fishes to exploitation. Manuscript to be reviewed 123 Yet, the reproductive potential of a stock is typically measured based on spawning stock biomass 124 or total egg production (i.e., fecundity), rather than traits affecting reproductive resilience 125 (Lowerre-Barbieri et al., 2017), including spatiotemporal behavioral traits, although spawning 126 season duration is used to estimate annual fecundity in indeterminate species (Cooper et al. 2013; 127 Maunder and Deriso 2013; Ganias et al. 2014). While spawning behavior traits have been 128 considered in some vulnerability analyses (Cheung et al. 2005; Erisman et al. 2014; Robinson 129 and Samoilys 2013; Robinson et al. 2015), this aspect of fish ecology remains poorly studied, 130 under-utilized for assessing stock health, and rarely emphasized in management frameworks (e.g. 131 Erisman et al. 2011; Sadovy De Mitcheson and Erisman 2012; Cheung et al. 2013). 132 Given the influence life history and spawning behavioral traits have on vulnerability, it is 133 important to consider whether these traits are correlated and how well they explain vulnerability 134 to fishing during spawning and exploitation status. If life history and spawning behavior traits 135 are not correlated, then for situations in which spawning behavior improves predictions of 136 vulnerability, increased research efforts to understand spawning behaviors would help identify 137 vulnerable species that may otherwise be overlooked and identify areas where targeted protection 138 of spawning fish may be needed to maintain sustainable harvest levels or rebuild overfished 139 stocks (Grüss and Robinson 2014; Grüss et al. 2018). Studies that have examined the relationship 140 between life history traits and spawning behavior were conducted in tropical regions and did not 141 focus explicitly on exploited species (e.g. Choat 2012; Nemeth 2012). Therefore, uncovering this 142 relationship and identifying species whose vulnerability to fishing is not explained by life history PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 159 The objective of this study was to investigate relationships between life history and 160 spawning behavioral traits and identify patterns among exploited species in the GOM that have 161 been historically overfished or may be particularly vulnerable to exploitation during the 162 spawning season. We employed multiple methods to evaluate these relationships. First, we used 163 a data synthesis approach to test whether life history and spawning behavioral traits were 164 correlated. Second, we used ordination to identify groups of traits (life history and spawning 165 behavior) that were common among overfished species. Third, we conducted a vulnerability 166 analysis to identify species and stocks that are likely highly vulnerable during the spawning 167 season and should be prioritized for further research and monitoring. The results indicate 168 whether including spawning behavior characteristics can improve our ability to assess the 169 vulnerability and resilience of marine fishes to exploitation. Additionally, this information will 170 identify important data gaps in our understanding of the spawning behavior of exploited marine 171 fishes and provide the basis for further research on interactions between spawning behavior and 172 fishing activities. This information will be applicable to the management and monitoring of 173 exploited marine fishes in the GOM, and the approach should be transferable to regional 174 vulnerability assessments of fish stocks elsewhere. Manuscript to be reviewed These parameters were chosen, because they have 198 been shown to be directly associated with vulnerability to fishing pressure, and they are 199 commonly used in productivity-susceptibility analyses, stock assessments, and in defining 200 species stock complexes (Patrick et al. 2009; Robinson 2015; Farmer et al. 2016). The reported 201 values were specific to the GOM unless there were no data, in which case information from the 202 Atlantic or Caribbean was used. When multiple values were found, the average (± SE) was used. 203 Sexual pattern was not included in this study, because specific traits associated with sexual 204 pattern (e.g., diagnosis, sex ratios, timing of sex change) are unknown for most hermaphroditic 205 species in the GOM. Variations in such traits strongly influence the resilience of hermaphroditic 206 species to fishing, often in complex ways (Robinson et al. 2017; Schram and Steele 2020), and 207 should thus be the focus of separate study. 188 IUCN Red List (https://www.iucnredlist.org). A detailed description of the selection process is 189 available in supplemental material (S1). 190 208 Spawning behavior was characterized by the degree of aggregating behavior, spawning 209 season duration in months, and the estimated magnitude of change in relative abundance of fish 210 during peak spawning periods relative to non-reproductive periods (Table 1). The duration of the 211 spawning season was determined by the number of months that spawning was reported to occur 212 in the GOM based primarily on the sampling of mature females and assessments of their 213 reproductive condition (e.g. elevated gonadosomatic index levels or the presence of actively 214 spawning females) (Brown-Peterson et al. 2011; Lowerre-Barbieri et al. 2011). However, some 215 fish aggregate at spawning sites over a longer time period than active spawning occurs (Heyman 216 et al. 2005; Heyman et al. 2019). The phrase “during spawning” refers to the spawning season 217 and was used in this context throughout the paper. g p p 218 In all cases, the values for each category were specific to the GOM, as they may vary 219 among populations and regions along a species’ geographic range (Lowerre-Barbieri et al. 2009; 220 Heyman et al. 2019). In instances where the spawning season varied within the GOM, we 221 incorporated the entire range of spawning months. Manuscript to be reviewed 188 IUCN Red List (https://www.iucnredlist.org). A detailed description of the selection process is 189 available in supplemental material (S1). 190 191 Life history and spawning behavioral traits 192 We compiled information on the life history and spawning behavior for the selected species 193 through reviews of primary literature, technical reports, and stock assessments from NOAA’s 194 Southeast Data Assessment and Review (SEDAR; http://sedarweb.org/ ). Life history parameters 195 included maximum age (Amax), maximum weight (Wmax) and maximum length (Lmax), von 196 Bertalanffy growth coefficient (k), asymptotic length (Linf), age (Am) and length at maturity (Lm), 197 and rate of natural mortality (M) (Table 1). These parameters were chosen, because they have 198 been shown to be directly associated with vulnerability to fishing pressure, and they are 199 commonly used in productivity-susceptibility analyses, stock assessments, and in defining 200 species stock complexes (Patrick et al. 2009; Robinson 2015; Farmer et al. 2016). The reported 201 values were specific to the GOM unless there were no data, in which case information from the 202 Atlantic or Caribbean was used. When multiple values were found, the average (± SE) was used. 203 Sexual pattern was not included in this study, because specific traits associated with sexual 204 pattern (e.g., diagnosis, sex ratios, timing of sex change) are unknown for most hermaphroditic 205 species in the GOM. Variations in such traits strongly influence the resilience of hermaphroditic 206 species to fishing, often in complex ways (Robinson et al. 2017; Schram and Steele 2020), and 207 should thus be the focus of separate study. 208 Spawning behavior was characterized by the degree of aggregating behavior, spawning 209 season duration in months, and the estimated magnitude of change in relative abundance of fish 210 during peak spawning periods relative to non-reproductive periods (Table 1). The duration of the 211 spawning season was determined by the number of months that spawning was reported to occur 212 in the GOM based primarily on the sampling of mature females and assessments of their 213 reproductive condition (e.g. elevated gonadosomatic index levels or the presence of actively 214 spawning females) (Brown-Peterson et al. 2011; Lowerre-Barbieri et al. 2011). However, some 215 fish aggregate at spawning sites over a longer time period than active spawning occurs (Heyman 216 et al. 2005; Heyman et al. 2019). Manuscript to be reviewed The phrase “during spawning” refers to the spawning season 217 and was used in this context throughout the paper. 218 In all cases, the values for each category were specific to the GOM, as they may vary 219 among populations and regions along a species’ geographic range (Lowerre-Barbieri et al. 2009; 220 Heyman et al. 2019). In instances where the spawning season varied within the GOM, we 221 incorporated the entire range of spawning months. If data related to the degree to which a species 222 formed spawning aggregations was not available for the GOM (e.g., Epinephelus flavolimbatus), 223 we determined their behavior based on literature from the Southeast U.S. and Caribbean (see 224 Results and references in S3). Aggregation type was intended to reflect the distance traveled to a 225 spawning site, and the number/distribution of spawning sites, reflected as the degree to which the 226 species aggregates to spawn on a scale of 0-4. Species that do not aggregate to spawn were 227 scored 0, simple migratory spawners were scored 1, species that form resident aggregations were 228 scored 2, species that form resident and transient aggregations (i.e., mixed) were scored 3, and 229 transient aggregations were scored 4. The estimated change in relative abundance was based on 230 order of magnitude comparisons (e.g., 1X, 10X, 100X) between peak spawning times and 231 abundance during non-spawning periods. The scale (1-6) distinguished among species that are 232 solitary, grouping, or schooling for non-reproductive functions. The spawning behavior 233 categories were ordinal in this case, because they related to varying degrees of vulnerability to 188 IUCN Red List (https://www.iucnredlist.org). A detailed description of the selection process is 189 available in supplemental material (S1). 190 191 Life history and spawning behavioral traits 192 We compiled information on the life history and spawning behavior for the selected species 193 through reviews of primary literature, technical reports, and stock assessments from NOAA’s 194 Southeast Data Assessment and Review (SEDAR; http://sedarweb.org/ ). Life history parameters 195 included maximum age (Amax), maximum weight (Wmax) and maximum length (Lmax), von 196 Bertalanffy growth coefficient (k), asymptotic length (Linf), age (Am) and length at maturity (Lm), 197 and rate of natural mortality (M) (Table 1). 177 Species selection 178 A hierarchical ranking process was used to identify a manageable number of relevant species to 179 include in the analysis (Biggs et al. 2017). The preliminary list of species included commonly 180 occurring and commonly caught recreational or commercial species that inhabit either offshore, 181 coastal or estuarine waters of the GOM, including all species managed in United States federal 182 waters by the Gulf of Mexico Fisheries Management Council (GMFMC). Species were scored 183 based on their aggregating behavior associated with spawning and a fisheries index, which 184 included two aspects of management status (inclusion in GMFMC’s fisheries management plan 185 and NOAA’s fish stock sustainability index; https://www.fisheries.noaa.gov), importance to 186 commercial fisheries (based on total annual landings in kg.), importance to recreational fisheries 187 (based on total annual landing in number of fish), and their endangered status according to the PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed The results of the 250 PCA were represented with a biplot along with the stock status based on region-wide 251 assessments in the GOM and the designations of NOAA Fisheries 252 (https://www.fisheries.noaa.gov). “Not overfished” species were those that have never been 253 designated as overfished. “Overfished” species included those that are currently designated or 254 had previously been designated at some point. For example, Red Grouper were placed in the 255 category of overfished, because previous assessments had made that designation even while 256 newer criteria concluded that the stock has never been overfished during the time series (SEDAR 257 2019). “Unassessed” species were those that have not been assessed in the GOM. Harvest of 258 Nassau Grouper (Epinephelus striatus), Atlantic Goliath Grouper (Epinephelus itajara), and Red 259 Drum (Sciaenops ocellatus) is prohibited, because each of these stocks were historically 260 overfished, and thus each were classified as overfished. Gag (Mycteroperca microlepis) and Red 261 Grouper (Epinephelus morio) have also been previously designated as overfished and were 262 labelled as such, although their current stock status is not overfished. There are no region-wide 263 assessments for the coastal species, only limited state assessments, so those stocks were 264 considered unassessed. A one-tailed t-test was used to compare the average PC1 and PC2 score 265 between overfished and not overfished stocks to identify common traits among the different 266 stock statuses. A one-tailed test was used, because each PC has directionality that is related to the 267 theoretical vulnerability to overfishing. Lower PC1 scores and higher PC2 scores would indicate 268 higher vulnerability. 269 271 Figure 1 contains a flow chart illustrating the process of the vulnerability assessment. The 272 vulnerability analysis was based on previous studies that accounted for life history traits and 273 spawning behaviors associated with vulnerability to fishing during spawning (Cheung et al. 274 2005; Robinson and Samoilys 2013; Robinson 2015). The indicators used are split between two 275 axes and include intrinsic indicators (life history and behavior of the species) and extrinsic 276 indicators (behavior of the fishery and management) that measure the susceptibility and exposure 277 of spawning fish to fishing. Manuscript to be reviewed There are no region-wide 263 assessments for the coastal species, only limited state assessments, so those stocks were 264 considered unassessed. A one-tailed t-test was used to compare the average PC1 and PC2 score 265 between overfished and not overfished stocks to identify common traits among the different 266 stock statuses. A one-tailed test was used, because each PC has directionality that is related to the 267 theoretical vulnerability to overfishing. Lower PC1 scores and higher PC2 scores would indicate 268 higher vulnerability. 234 fishing pressure (Robinson and Samoilys 2013; Robinson, 2015). Spawning season duration and 235 the degree of aggregating behavior was obtained exclusively through a comprehensive literature 236 review. Relative abundance was based on expert judgement of the authors and collaborators 237 associated with this project. 234 fishing pressure (Robinson and Samoilys 2013; Robinson, 2015). Spawning season duration and 235 the degree of aggregating behavior was obtained exclusively through a comprehensive literature 236 review. Relative abundance was based on expert judgement of the authors and collaborators 237 associated with this project. 234 fishing pressure (Robinson and Samoilys 2013; Robinson, 2015). Spawning season duration and 235 the degree of aggregating behavior was obtained exclusively through a comprehensive literature 236 review. Relative abundance was based on expert judgement of the authors and collaborators 237 associated with this project. 238 239 Correlation analysis and PCA 240 Spearman’s Rank Correlation was used to explore relationships between life history traits and 241 spawning behavior parameters in R (R Core Team, 2016), tested at a significance level of 242 α=0.05. The parameters were also used to perform Principal Component Analysis (PCA) using a 243 correlation matrix with normalized data. We could not identify reliable estimates of M for the 244 range of species of interest and thus did not integrate this trait into the PCA. A PCA was chosen 245 rather than a Linear Discriminant Analysis (LDA) or cluster analysis, because we were interested 246 in the continuum of reproductive behaviors and life history traits. Although PCA and k-means 247 clustering are closely related, the PCA offers a continuous solution rather than clusters of 248 homogenous groups (Ding and He 2004). Likewise, PCA is unsupervised and finds the 249 directionality of maximum variance, where LDA maximizes class separability. Manuscript to be reviewed If data related to the degree to which a species 222 formed spawning aggregations was not available for the GOM (e.g., Epinephelus flavolimbatus), 223 we determined their behavior based on literature from the Southeast U.S. and Caribbean (see 224 Results and references in S3). Aggregation type was intended to reflect the distance traveled to a 225 spawning site, and the number/distribution of spawning sites, reflected as the degree to which the 226 species aggregates to spawn on a scale of 0-4. Species that do not aggregate to spawn were 227 scored 0, simple migratory spawners were scored 1, species that form resident aggregations were 228 scored 2, species that form resident and transient aggregations (i.e., mixed) were scored 3, and 229 transient aggregations were scored 4. The estimated change in relative abundance was based on 230 order of magnitude comparisons (e.g., 1X, 10X, 100X) between peak spawning times and 231 abundance during non-spawning periods. The scale (1-6) distinguished among species that are 232 solitary, grouping, or schooling for non-reproductive functions. The spawning behavior 233 categories were ordinal in this case, because they related to varying degrees of vulnerability to PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Manuscript to be reviewed 234 fishing pressure (Robinson and Samoilys 2013; Robinson, 2015). Spawning season duration and 235 the degree of aggregating behavior was obtained exclusively through a comprehensive literature 236 review. Relative abundance was based on expert judgement of the authors and collaborators 237 associated with this project. 238 239 Correlation analysis and PCA 240 Spearman’s Rank Correlation was used to explore relationships between life history traits and 241 spawning behavior parameters in R (R Core Team, 2016), tested at a significance level of 242 α=0.05. The parameters were also used to perform Principal Component Analysis (PCA) using a 243 correlation matrix with normalized data. We could not identify reliable estimates of M for the 244 range of species of interest and thus did not integrate this trait into the PCA. A PCA was chosen 245 rather than a Linear Discriminant Analysis (LDA) or cluster analysis, because we were interested 246 in the continuum of reproductive behaviors and life history traits. Although PCA and k-means 247 clustering are closely related, the PCA offers a continuous solution rather than clusters of 248 homogenous groups (Ding and He 2004). Likewise, PCA is unsupervised and finds the 249 directionality of maximum variance, where LDA maximizes class separability. The results of the 250 PCA were represented with a biplot along with the stock status based on region-wide 251 assessments in the GOM and the designations of NOAA Fisheries 252 (https://www.fisheries.noaa.gov). “Not overfished” species were those that have never been 253 designated as overfished. “Overfished” species included those that are currently designated or 254 had previously been designated at some point. For example, Red Grouper were placed in the 255 category of overfished, because previous assessments had made that designation even while 256 newer criteria concluded that the stock has never been overfished during the time series (SEDAR 257 2019). “Unassessed” species were those that have not been assessed in the GOM. Harvest of 258 Nassau Grouper (Epinephelus striatus), Atlantic Goliath Grouper (Epinephelus itajara), and Red 259 Drum (Sciaenops ocellatus) is prohibited, because each of these stocks were historically 260 overfished, and thus each were classified as overfished. Gag (Mycteroperca microlepis) and Red 261 Grouper (Epinephelus morio) have also been previously designated as overfished and were 262 labelled as such, although their current stock status is not overfished. Manuscript to be reviewed Scores indicated if spawning sites were closed all year 299 or involved a complete fishery closure (1), more than or equal to 10% of known spawning sites 300 were protected by complete or seasonal site closures (2), less than 10% of known spawning sites 301 were protected by complete or seasonal site closures (3), or if there were no spawning site 302 closures or the species did not reproduce in federal waters (i.e., coastal species) (4). State site 303 closures are few and spatially minimal. Therefore, all coastal species received a score of 4 (i.e., 304 no site-based regulations). 305 Reciprocal values were used for spawning season duration and k to preserve the direction 306 of their influence on, and association with, increasing vulnerability. The values for life history 307 traits (Amax, Linf, k, and Am) were combined into one category as a life history composite. The 308 values for each intrinsic and extrinsic indicator were scaled 0-1. Each parameter was then 309 weighted according to the impact and relative influence on vulnerability as determined in 310 Robinson (2015), which assigned the weights through an Analytic Hierarchy Process (Saaty 311 1987). The process involved ranking each parameter in terms of its influence on vulnerability 312 and performing pair-wise comparisons to develop a matrix of indicator weights with the final 313 value calculated as the average among the matrices. Then, the predictive ability of those 314 weighted parameters were validated against the status of global fisheries targeting spawning 315 aggregations (Robinson and Samoilys 2013). The validation of the vulnerability index showed 316 the correlation of the parameters with abundance trends in seven species of reef fishes, and 317 ultimately supported the use of the indicator-based framework (Robinson and Samoilys 2013). 318 The respective weights for the intrinsic and extrinsic indicators are in Tables 1 & 2. The values 319 were summed for each group of indicators to obtain an intrinsic and extrinsic vulnerability score. 320 The scores along the two axes were combined to get an overall index of vulnerability to fishing 321 during spawning, which was calculated as the Euclidian distance from the origin (position in 322 bivariate space). Equal weight was given to intrinsic and extrinsic indicators as those compound 323 factors were rescaled (0-1) to provide a relative comparison among the selected species. Manuscript to be reviewed 324 325 Results 281 state or federal waters of the GOM and included access to the fishery, catch limits, gear 282 measures, seasonal closures and site closures during the spawning season (Table 2). 283 The extrinsic indicators were each scored on a scale of 1-4 with a larger number denoting 284 a higher vulnerability. Federal regulations were considered for all species except for coastal 285 species that are primarily targeted in state waters. For those species, state regulations were 286 considered since they inhabit, spawn, and are fished in state waters. The assigned scores for 287 coastal species reflected the average score among the Gulf states (Texas, Louisiana, Mississippi, 288 Alabama, and Florida). Access to the fishery captured the extent to which access is restricted via 289 a regulated number of permits and the individual fishing quota (IFQ) program (1), or open, 290 requiring a basic commercial or recreational license (4). Catch limits included minimum and 291 maximum size limits as well as daily bag limits and quotas for the commercial and recreational 292 fishery. Scores ranged from 1 (for a total of 5 catch limits) to 4 (for no catch limits). Gear 293 measures indicated the restrictions on gear types used in the fishery and ranged from 0 to 2 294 allowable gear types (1), to 9 or more allowable gear types (4). Seasonal restrictions reflected the 295 level of spawning season prohibition of take, from prohibition during the entire spawning season 296 (1) scaling to no restrictions (4). The selected value was the least restrictive score of recreational 297 and commercial seasonal restrictions. Site closures ranged from total spatial closure of spawning 298 site (1) scaling to no spatial closures (4). Scores indicated if spawning sites were closed all year 299 or involved a complete fishery closure (1), more than or equal to 10% of known spawning sites 300 were protected by complete or seasonal site closures (2), less than 10% of known spawning sites 301 were protected by complete or seasonal site closures (3), or if there were no spawning site 302 closures or the species did not reproduce in federal waters (i.e., coastal species) (4). State site 303 closures are few and spatially minimal. Therefore, all coastal species received a score of 4 (i.e., 304 no site-based regulations). Manuscript to be reviewed The intrinsic indicators included the degree of aggregating behavior 278 associated with spawning, duration of the spawning season, change in relative abundance during 279 spawning, Amax, Linf, Am, and k, as their correlation to fishing vulnerability has been previously PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 280 illustrated. The extrinsic indicators reflect the degree to which spawning fish are protected within 281 state or federal waters of the GOM and included access to the fishery, catch limits, gear 282 measures, seasonal closures and site closures during the spawning season (Table 2). 280 illustrated. The extrinsic indicators reflect the degree to which spawning fish are protected within 281 state or federal waters of the GOM and included access to the fishery, catch limits, gear 282 measures, seasonal closures and site closures during the spawning season (Table 2). 283 The extrinsic indicators were each scored on a scale of 1-4 with a larger number denoting 284 a higher vulnerability. Federal regulations were considered for all species except for coastal 285 species that are primarily targeted in state waters. For those species, state regulations were 286 considered since they inhabit, spawn, and are fished in state waters. The assigned scores for 287 coastal species reflected the average score among the Gulf states (Texas, Louisiana, Mississippi, 288 Alabama, and Florida). Access to the fishery captured the extent to which access is restricted via 289 a regulated number of permits and the individual fishing quota (IFQ) program (1), or open, 290 requiring a basic commercial or recreational license (4). Catch limits included minimum and 291 maximum size limits as well as daily bag limits and quotas for the commercial and recreational 292 fishery. Scores ranged from 1 (for a total of 5 catch limits) to 4 (for no catch limits). Gear 293 measures indicated the restrictions on gear types used in the fishery and ranged from 0 to 2 294 allowable gear types (1), to 9 or more allowable gear types (4). Seasonal restrictions reflected the 295 level of spawning season prohibition of take, from prohibition during the entire spawning season 296 (1) scaling to no restrictions (4). The selected value was the least restrictive score of recreational 297 and commercial seasonal restrictions. Site closures ranged from total spatial closure of spawning 298 site (1) scaling to no spatial closures (4). Manuscript to be reviewed 327 Species selection 328 The species selection process identified 24 species to be included in the final assessment. Four 329 common coastal and estuarine species were added to the analysis post hoc, because they are 330 important to fisheries in state waters throughout the GOM: Spotted Seatrout (Cynoscion 331 nebulosus), Sheepshead (Archosargus probatocephalus), Southern Flounder (Paralichthys 332 lethostigma), and Black Drum (Pogonias cromis). Although the selected species are not a 333 random sample, they represent a wide range of exploited species from which to examine life 334 history traits and spawning behavior in relation to overfishing and vulnerability, which was our 335 objective. Although aggregating behavior in association with spawning was a main focus of the 336 analysis, six species that do not aggregate to spawn were included among the selected species. 337 Highly migratory, schooling, pelagic species (e.g., tunas) were excluded, because of the vastly 338 different reproductive habitats of open ocean pelagic species. As a group, they are well studied 339 and managed as a separate unit. This resulted in a final list of 28 species to be analyzed, which 340 are listed on Table 3 and described in greater detail in the supplementary material (S1). 341 342 Life history information and spawning behavior 343 A total of 801 documents including peer-reviewed literature, grey literature, and SEDAR reports 344 were reviewed for spawning and life history information on the 28 species. The values for each 345 parameter are presented in Table 3, and a full table with citations for each entry is available 346 online (http://geo.gcoos.org/restore/) and in supplementary material (S3). Values for M were not 347 found for Almaco Jack (Seriola rivoliana) or Warsaw Grouper (Hyporthodus nigritus). 348 The degree of aggregating behavior was determined directly from descriptions in the 349 literature for the GOM for all species except for Warsaw Grouper and Yellowedge Grouper 350 (Hyporthodus flavolimbatus), which were classified by the authors’ expert opinions based on 351 studies from the Southeast U.S. and Caribbean (Table 3, S3). Based on the literature review, 10 352 of 28 species were determined to form transient aggregations. Groupers (Epinephelidae; n=6;) 353 were the most common family listed in the transient group. Seven species were categorized as 354 forming mixed aggregations and included three grouper species, the two jacks (Carangidae) and 355 two sciaenid (Sciaenidae) species . Five species were determined to form resident aggregations,. Manuscript to be reviewed 356 The two mackerel (Scomberomorus) species were designated as simple migratory spawners. Red 357 Grouper, Snowy Grouper (Hyporthodus niveatus), Speckled Hind (Epinephelus dummondhayi), 358 and Tilefish were determined to not form spawning aggregations based on the lack of any 359 evidence of such behavior in the literature for the GOM or elsewhere. 360 The change in relative abundance of fish during spawning was greatest for Cubera 361 Snapper (Lutjanus cyanopterus) and Nassau Grouper, which corresponded to aggregations of 362 >10,000 fish (Table 3). Three additional species were classified as 5, corresponding to 363 aggregations of 1,000-10,000 fish. Most species (n=9) were scored 3, indicating that 364 aggregations were composed of small groups of 100-200 fish. Seven species were scored a 2, 365 corresponding to small groups or doubling of abundance relative to non-reproductive periods. 366 Spawning seasons ranged from 3 to 12 months, with grouper species having the largest 367 variation in spawning season (Table 4). Yellowedge Grouper had the most protracted spawning 368 seasons at 12 months followed by Yellowmouth Grouper (Mycteroperca interstitialis) at 10 369 months, while Nassau Grouper and Sheepshead only spawned three months out of the year. The 370 greatest number of species spawned in June (n=21), with an average of 18 species spawning per 371 month, April through August. The fewest number of species spawned in November (n=7) and 327 Species selection 328 The species selection process identified 24 species to be included in the final assessment. Four 329 common coastal and estuarine species were added to the analysis post hoc, because they are 330 important to fisheries in state waters throughout the GOM: Spotted Seatrout (Cynoscion 331 nebulosus), Sheepshead (Archosargus probatocephalus), Southern Flounder (Paralichthys 332 lethostigma), and Black Drum (Pogonias cromis). Although the selected species are not a 333 random sample, they represent a wide range of exploited species from which to examine life 334 history traits and spawning behavior in relation to overfishing and vulnerability, which was our 335 objective. Although aggregating behavior in association with spawning was a main focus of the 336 analysis, six species that do not aggregate to spawn were included among the selected species. 337 Highly migratory, schooling, pelagic species (e.g., tunas) were excluded, because of the vastly 338 different reproductive habitats of open ocean pelagic species. As a group, they are well studied 339 and managed as a separate unit. Manuscript to be reviewed 305 Reciprocal values were used for spawning season duration and k to preserve the direction 306 of their influence on, and association with, increasing vulnerability. The values for life history 307 traits (Amax, Linf, k, and Am) were combined into one category as a life history composite. The 308 values for each intrinsic and extrinsic indicator were scaled 0-1. Each parameter was then 309 weighted according to the impact and relative influence on vulnerability as determined in 310 Robinson (2015), which assigned the weights through an Analytic Hierarchy Process (Saaty 311 1987). The process involved ranking each parameter in terms of its influence on vulnerability 312 and performing pair-wise comparisons to develop a matrix of indicator weights with the final 313 value calculated as the average among the matrices. Then, the predictive ability of those 314 weighted parameters were validated against the status of global fisheries targeting spawning 315 aggregations (Robinson and Samoilys 2013). The validation of the vulnerability index showed 316 the correlation of the parameters with abundance trends in seven species of reef fishes, and 317 ultimately supported the use of the indicator-based framework (Robinson and Samoilys 2013). 318 The respective weights for the intrinsic and extrinsic indicators are in Tables 1 & 2. The values 319 were summed for each group of indicators to obtain an intrinsic and extrinsic vulnerability score. 320 The scores along the two axes were combined to get an overall index of vulnerability to fishing 321 during spawning, which was calculated as the Euclidian distance from the origin (position in 322 bivariate space). Equal weight was given to intrinsic and extrinsic indicators as those compound 323 factors were rescaled (0-1) to provide a relative comparison among the selected species. 324 325 Results PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 360 The change in relative abundance of fish during spawning was greatest for Cubera 361 Snapper (Lutjanus cyanopterus) and Nassau Grouper, which corresponded to aggregations of 362 >10 000 fish (Table 3) Three additional species were classified as 5 corresponding to Manuscript to be reviewed This resulted in a final list of 28 species to be analyzed, which 340 are listed on Table 3 and described in greater detail in the supplementary material (S1). 341 p y p y 330 important to fisheries in state waters throughout the GOM: Spotted Seatrout (Cynoscion 331 nebulosus), Sheepshead (Archosargus probatocephalus), Southern Flounder (Paralichthys 332 lethostigma), and Black Drum (Pogonias cromis). Although the selected species are not a 333 random sample, they represent a wide range of exploited species from which to examine life 334 history traits and spawning behavior in relation to overfishing and vulnerability, which was our 335 objective. Although aggregating behavior in association with spawning was a main focus of the 336 analysis, six species that do not aggregate to spawn were included among the selected species. 337 Highly migratory, schooling, pelagic species (e.g., tunas) were excluded, because of the vastly 338 different reproductive habitats of open ocean pelagic species. As a group, they are well studied 339 and managed as a separate unit. This resulted in a final list of 28 species to be analyzed, which 340 are listed on Table 3 and described in greater detail in the supplementary material (S1). 341 342 Life history information and spawning behavior 343 A total of 801 documents including peer-reviewed literature, grey literature, and SEDAR reports 344 were reviewed for spawning and life history information on the 28 species. The values for each 345 parameter are presented in Table 3, and a full table with citations for each entry is available 346 online (http://geo.gcoos.org/restore/) and in supplementary material (S3). Values for M were not 347 found for Almaco Jack (Seriola rivoliana) or Warsaw Grouper (Hyporthodus nigritus). 348 The degree of aggregating behavior was determined directly from descriptions in the 349 literature for the GOM for all species except for Warsaw Grouper and Yellowedge Grouper 350 (Hyporthodus flavolimbatus), which were classified by the authors’ expert opinions based on 351 studies from the Southeast U.S. and Caribbean (Table 3, S3). Based on the literature review, 10 352 of 28 species were determined to form transient aggregations. Groupers (Epinephelidae; n=6;) 353 were the most common family listed in the transient group. Seven species were categorized as 354 forming mixed aggregations and included three grouper species, the two jacks (Carangidae) and 355 two sciaenid (Sciaenidae) species . Five species were determined to form resident aggregations,. Manuscript to be reviewed 356 The two mackerel (Scomberomorus) species were designated as simple migratory spawners. Red 357 Grouper, Snowy Grouper (Hyporthodus niveatus), Speckled Hind (Epinephelus dummondhayi), 358 and Tilefish were determined to not form spawning aggregations based on the lack of any 359 evidence of such behavior in the literature for the GOM or elsewhere 336 analysis, six species that do not aggregate to spawn were included among the selected species. 337 Highly migratory, schooling, pelagic species (e.g., tunas) were excluded, because of the vastly 338 different reproductive habitats of open ocean pelagic species. As a group, they are well studied 339 and managed as a separate unit. This resulted in a final list of 28 species to be analyzed, which 340 are listed on Table 3 and described in greater detail in the supplementary material (S1). 341 342 Life history information and spawning behavior 343 A total of 801 documents including peer-reviewed literature, grey literature, and SEDAR reports 344 were reviewed for spawning and life history information on the 28 species. The values for each 345 parameter are presented in Table 3, and a full table with citations for each entry is available 346 online (http://geo.gcoos.org/restore/) and in supplementary material (S3). Values for M were not 347 found for Almaco Jack (Seriola rivoliana) or Warsaw Grouper (Hyporthodus nigritus). 348 The degree of aggregating behavior was determined directly from descriptions in the 349 literature for the GOM for all species except for Warsaw Grouper and Yellowedge Grouper 350 (Hyporthodus flavolimbatus), which were classified by the authors’ expert opinions based on 351 studies from the Southeast U.S. and Caribbean (Table 3, S3). Based on the literature review, 10 352 of 28 species were determined to form transient aggregations. Groupers (Epinephelidae; n=6;) 353 were the most common family listed in the transient group. Seven species were categorized as 354 forming mixed aggregations and included three grouper species, the two jacks (Carangidae) and 355 two sciaenid (Sciaenidae) species . Five species were determined to form resident aggregations,. 356 The two mackerel (Scomberomorus) species were designated as simple migratory spawners. Red 357 Grouper, Snowy Grouper (Hyporthodus niveatus), Speckled Hind (Epinephelus dummondhayi), 358 and Tilefish were determined to not form spawning aggregations based on the lack of any 359 evidence of such behavior in the literature for the GOM or elsewhere. Manuscript to be reviewed The mean PC score was significantly higher for stocks currently 399 or historically overfished than not overfished for PC2 (t = 2.03, df = 9.25, p = 0.04) but not for 400 PC1 (t = -0.844, df = 15, p = 0.79) (Figure 4). Overfished status was closely associated with 401 positive changes in relative abundance and aggregation type, and negatively related to spawning 402 season duration. Peak spawning months was included in an additional PCA, but it did not change 403 the output and had very low eigenvectors, so it was not considered further (Figure S2.1). 404 405 Vulnerability analysis 406 Scores for the intrinsic and extrinsic vulnerability analysis along with the extrinsic indicator 372 December (n=4). The snappers were the only family that showed consistency in spawning 373 season, with peak spawning occurring June through August. An annotated table with all 374 references is available online (http://geo.gcoos.org/restore/) and in supplementary material (S3). 375 376 Correlation analysis and PCA 377 Aggregation type was positively correlated with relative abundance (rs = 0.819, p < 0.01) (Figure 378 2, Table S2.1). Spawning season duration was negatively correlated with relative abundance (rs = 379 -0.538, p = 0.01) and k (rs = -0.526, p = 0.01). We found no other significant relationships 380 between spawning behavior parameters and life history traits. However, most of the life history 381 traits were significantly correlated with each other (Figure 2, Table S2.1). As expected, the 382 maximum growth parameters (age, weight, length, Linf,) were positively correlated with each 383 other and Am, and negatively correlated with k and M. 384 The first two axes of the PCA explained 64.4% of the variation in the data (Figure 3, 385 Table S2.2). Along PC2, the eigenvectors for spawning behavior characteristics had a greater 386 influence (as defined by the absolute value of the eigenvector) on the distribution of species than 387 life history traits. Relative abundance (0.634) was the greatest followed by aggregation type 388 (0.493) and spawning season duration (-0.481). Along PC1, the eigenvectors for life history traits 389 had a greater influence than those for spawning behaviors. Lm was the greatest (-0.440), followed 390 by Linf (-0.430), Lmax (-0.405), and Wmax (-0.399). The PCA biplot (Figure 3) illustrates the 391 separation of species by the reproductive and life history parameters along the first two PC axes 392 within the context of stock status. 342 Life history information and spawning behavior f y f p g 343 A total of 801 documents including peer-reviewed literature, grey literature, and SEDAR reports 344 were reviewed for spawning and life history information on the 28 species. The values for each 345 parameter are presented in Table 3, and a full table with citations for each entry is available 346 online (http://geo.gcoos.org/restore/) and in supplementary material (S3). Values for M were not 347 found for Almaco Jack (Seriola rivoliana) or Warsaw Grouper (Hyporthodus nigritus). 348 The degree of aggregating behavior was determined directly from descriptions in the 349 literature for the GOM for all species except for Warsaw Grouper and Yellowedge Grouper 350 (Hyporthodus flavolimbatus), which were classified by the authors’ expert opinions based on 351 studies from the Southeast U.S. and Caribbean (Table 3, S3). Based on the literature review, 10 352 of 28 species were determined to form transient aggregations. Groupers (Epinephelidae; n=6;) 353 were the most common family listed in the transient group. Seven species were categorized as 354 forming mixed aggregations and included three grouper species, the two jacks (Carangidae) and 355 two sciaenid (Sciaenidae) species . Five species were determined to form resident aggregations,. 356 The two mackerel (Scomberomorus) species were designated as simple migratory spawners. Red 357 Grouper, Snowy Grouper (Hyporthodus niveatus), Speckled Hind (Epinephelus dummondhayi), 358 and Tilefish were determined to not form spawning aggregations based on the lack of any 359 evidence of such behavior in the literature for the GOM or elsewhere. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Six of the species have been deemed not overfished. Ten 393 species have been overfished, including Atlantic Goliath Grouper and Nassau Grouper, which 394 are both closed to recreational and commercial fishing, as well as Red Drum, which is closed to 395 fishing in federal waters and to commercial fishing in most states along the GOM. Twelve 396 species have not been assessed, including three coastal species. The PC2 scores were positive for 397 7 out of 10 species that are currently or historically overfished whereas they were negative for 4 398 of the 6 not overfished species. The mean PC score was significantly higher for stocks currently 399 or historically overfished than not overfished for PC2 (t = 2.03, df = 9.25, p = 0.04) but not for 400 PC1 (t = -0.844, df = 15, p = 0.79) (Figure 4). Overfished status was closely associated with 401 positive changes in relative abundance and aggregation type, and negatively related to spawning 402 season duration. Peak spawning months was included in an additional PCA, but it did not change 403 the output and had very low eigenvectors, so it was not considered further (Figure S2.1). 404 405 Vulnerability analysis 372 December (n=4). The snappers were the only family that showed consistency in spawning 373 season, with peak spawning occurring June through August. An annotated table with all 374 references is available online (http://geo.gcoos.org/restore/) and in supplementary material (S3). 375 376 Correlation analysis and PCA Overfished status was closely associated with 401 positive changes in relative abundance and aggregation type, and negatively related to spawning 402 season duration. Peak spawning months was included in an additional PCA, but it did not change 403 the output and had very low eigenvectors, so it was not considered further (Figure S2.1). 404 405 Vulnerability analysis 406 Scores for the intrinsic and extrinsic vulnerability analysis along with the extrinsic indicator 407 scores for federal and state management regulations are available in supplemental material (S4). 408 Sheepshead and Southern Flounder had the two greatest overall vulnerability scores, but neither 409 are federally managed nor have they been assessed at the region-wide level (Figure 5). Cubera 410 Snapper had the greatest vulnerability scores among the federally managed species, but the stock 411 has not been assessed. All species scored high in extrinsic vulnerability (>0.5), except for Gray 412 Triggerfish (Balistes capriscus) and the three species with closed fisheries (Atlantic Goliath 413 Grouper, Nassau Grouper, Red Drum) that also had the lowest overall vulnerability scores. The 414 intrinsic vulnerability score was greatest for Nassau Grouper and least for Spanish Mackerel 415 (Scomberomorus maculatus). Cubera Snapper, Warsaw Grouper, and Yellowmouth Grouper had 416 the highest intrinsic vulnerability scores among the unassessed species. 417 y 377 Aggregation type was positively correlated with relative abundance (rs = 0.819, p < 0.01) (Figure 378 2, Table S2.1). Spawning season duration was negatively correlated with relative abundance (rs = 379 -0.538, p = 0.01) and k (rs = -0.526, p = 0.01). We found no other significant relationships 380 between spawning behavior parameters and life history traits. However, most of the life history 381 traits were significantly correlated with each other (Figure 2, Table S2.1). As expected, the 382 maximum growth parameters (age, weight, length, Linf,) were positively correlated with each 383 other and Am, and negatively correlated with k and M. 384 The first two axes of the PCA explained 64.4% of the variation in the data (Figure 3, 385 Table S2.2). Along PC2, the eigenvectors for spawning behavior characteristics had a greater 386 influence (as defined by the absolute value of the eigenvector) on the distribution of species than 387 life history traits. Relative abundance (0.634) was the greatest followed by aggregation type 388 (0.493) and spawning season duration (-0.481). Manuscript to be reviewed 372 December (n=4). The snappers were the only family that showed consistency in spawning 373 season, with peak spawning occurring June through August. An annotated table with all 374 references is available online (http://geo.gcoos.org/restore/) and in supplementary material (S3). 375 376 Correlation analysis and PCA 377 Aggregation type was positively correlated with relative abundance (rs = 0.819, p < 0.01) (Figure 378 2, Table S2.1). Spawning season duration was negatively correlated with relative abundance (rs = 379 -0.538, p = 0.01) and k (rs = -0.526, p = 0.01). We found no other significant relationships 380 between spawning behavior parameters and life history traits. However, most of the life history 381 traits were significantly correlated with each other (Figure 2, Table S2.1). As expected, the 382 maximum growth parameters (age, weight, length, Linf,) were positively correlated with each 383 other and Am, and negatively correlated with k and M. 384 The first two axes of the PCA explained 64.4% of the variation in the data (Figure 3, 385 Table S2.2). Along PC2, the eigenvectors for spawning behavior characteristics had a greater 386 influence (as defined by the absolute value of the eigenvector) on the distribution of species than 387 life history traits. Relative abundance (0.634) was the greatest followed by aggregation type 388 (0.493) and spawning season duration (-0.481). Along PC1, the eigenvectors for life history traits 389 had a greater influence than those for spawning behaviors. Lm was the greatest (-0.440), followed 390 by Linf (-0.430), Lmax (-0.405), and Wmax (-0.399). The PCA biplot (Figure 3) illustrates the 391 separation of species by the reproductive and life history parameters along the first two PC axes 392 within the context of stock status. Six of the species have been deemed not overfished. Ten 393 species have been overfished, including Atlantic Goliath Grouper and Nassau Grouper, which 394 are both closed to recreational and commercial fishing, as well as Red Drum, which is closed to 395 fishing in federal waters and to commercial fishing in most states along the GOM. Twelve 396 species have not been assessed, including three coastal species. The PC2 scores were positive for 397 7 out of 10 species that are currently or historically overfished whereas they were negative for 4 398 of the 6 not overfished species. 376 Correlation analysis and PCA y 377 Aggregation type was positively correlated with relative abundance (rs = 0.819, p < 0.01) (Figure 378 2, Table S2.1). Spawning season duration was negatively correlated with relative abundance (rs = 379 -0.538, p = 0.01) and k (rs = -0.526, p = 0.01). We found no other significant relationships 380 between spawning behavior parameters and life history traits. However, most of the life history 381 traits were significantly correlated with each other (Figure 2, Table S2.1). As expected, the 382 maximum growth parameters (age, weight, length, Linf,) were positively correlated with each 383 other and Am, and negatively correlated with k and M. 384 The first two axes of the PCA explained 64.4% of the variation in the data (Figure 3, 385 Table S2.2). Along PC2, the eigenvectors for spawning behavior characteristics had a greater 386 influence (as defined by the absolute value of the eigenvector) on the distribution of species than 387 life history traits. Relative abundance (0.634) was the greatest followed by aggregation type 388 (0.493) and spawning season duration (-0.481). Along PC1, the eigenvectors for life history traits 389 had a greater influence than those for spawning behaviors. Lm was the greatest (-0.440), followed 390 by Linf (-0.430), Lmax (-0.405), and Wmax (-0.399). The PCA biplot (Figure 3) illustrates the 391 separation of species by the reproductive and life history parameters along the first two PC axes 392 within the context of stock status. Six of the species have been deemed not overfished. Ten 393 species have been overfished, including Atlantic Goliath Grouper and Nassau Grouper, which 394 are both closed to recreational and commercial fishing, as well as Red Drum, which is closed to 395 fishing in federal waters and to commercial fishing in most states along the GOM. Twelve 396 species have not been assessed, including three coastal species. The PC2 scores were positive for 397 7 out of 10 species that are currently or historically overfished whereas they were negative for 4 398 of the 6 not overfished species. The mean PC score was significantly higher for stocks currently 399 or historically overfished than not overfished for PC2 (t = 2.03, df = 9.25, p = 0.04) but not for 400 PC1 (t = -0.844, df = 15, p = 0.79) (Figure 4). 418 Discussion GOM, stock assessments generate a probability 438 distribution function of overfishing limits that is converted to allowable biological catch based 439 on the Gulf Council’s risk tolerance for scientific uncertainty. Our findings imply a larger buffer 440 may be necessary to avoid overfishing for transient aggregating species, especially when this 441 spawning behavior is not explicitly considered in stock assessment. p g p y 442 There was a clear separation between life history traits and spawning behavior with 443 respect to loadings along the first two principal component axes. The component scores along 444 PC1 separated species based on life history traits, and PC2 was primarily loaded with spawning 445 behaviors. Spawning behavior provided a better distinction between overfished and not- 446 overfished species than the life history traits (i.e., more status information can be derived from 447 PC2 than PC1). Stock status of a fishery is influenced by multiple factors, which include 448 extrinsic components such as fishing practices (e.g., fishing effort; gear efficiency and 449 selectivity), management actions, stock assessment uncertainty stemming from data limitations 450 in the indicators (e.g., biomass), and changes in biological reference points (Rosenberg and 451 Restrepo 1994; Branch et al. 2011). We acknowledge that stock status may not completely 452 reflect the vulnerability and resilience of a species, but it is commonly assumed to be a function 453 of life history traits and historical exploitation (Costello et al. 2012). In addition, stock status 454 provides a metric to compare the relationships between exploitation, life history and reproductive 455 behavior, especially with respect to identifying species that might be at risk of overfishing. 456 Several unassessed stocks with high intrinsic and extrinsic vulnerability were identified, 457 including deep-water grouper stocks such as Speckled Hind and Warsaw Grouper. In the nearby 458 U.S. South Atlantic region, spatial protections for these stocks have been considered, to reduce 459 post-release mortality (Farmer & Karnauskas 2013) and then implemented by creating a 460 declaration of five new Spawning Special Management Zones (sSMZs) in which bottom fishing 461 is restricted (South Atlantic Fisheries Management Council 2017). 462 As has been shown in several review studies on spawning aggregations (e g Erisman et ( g ) 462 As has been shown in several review studies on spawning aggregations (e.g., Erisman et 463 al. Manuscript to be reviewed Spawning behavior provided a better distinction between overfished and not- 446 overfished species than the life history traits (i.e., more status information can be derived from 447 PC2 than PC1). Stock status of a fishery is influenced by multiple factors, which include 448 extrinsic components such as fishing practices (e.g., fishing effort; gear efficiency and 449 selectivity), management actions, stock assessment uncertainty stemming from data limitations 450 in the indicators (e.g., biomass), and changes in biological reference points (Rosenberg and 451 Restrepo 1994; Branch et al. 2011). We acknowledge that stock status may not completely 452 reflect the vulnerability and resilience of a species, but it is commonly assumed to be a function 453 of life history traits and historical exploitation (Costello et al. 2012). In addition, stock status 454 provides a metric to compare the relationships between exploitation, life history and reproductive 455 behavior, especially with respect to identifying species that might be at risk of overfishing. 456 Several unassessed stocks with high intrinsic and extrinsic vulnerability were identified, 457 including deep-water grouper stocks such as Speckled Hind and Warsaw Grouper. In the nearby 458 U.S. South Atlantic region, spatial protections for these stocks have been considered, to reduce 459 post-release mortality (Farmer & Karnauskas 2013) and then implemented by creating a 460 declaration of five new Spawning Special Management Zones (sSMZs) in which bottom fishing 461 is restricted (South Atlantic Fisheries Management Council 2017). 462 As has been shown in several review studies on spawning aggregations (e.g., Erisman et l d d i h d i ) i i h i b h i 418 Discussion 419 Spawning behavior represents a separate and distinct aspect of fish ecology that is important to 420 consider for accurate predictions of vulnerability and resilience in exploited stocks (Erisman et 421 al. 2017a,b; Lowerre-Barbieri et al. 2017). Our results show that characteristics of spawning 422 behavior known to be associated with vulnerability to fishing are not directly related to life 423 history traits that are typically associated with vulnerability in 28 species of exploited marine 424 fishes in the GOM. Using PCA analysis, we demonstrated that the characterization of spawning 425 behaviors improved the identification of vulnerable and overfished species more than 426 “traditional” life history traits alone. Manuscript to be reviewed We found that nearly all species showed susceptibility to 427 overfishing or becoming overfished during their spawning season due to very few state or federal 428 regulations to protect spawning fish in the GOM. However, there was a large range of intrinsic 429 vulnerability based on the diversity of spawning behavior and life history traits exhibited among 430 the species studied. Increased effort to understand patterns of spawning behavior and the 431 distribution of fishing effort and catch in relation to spawning would aid conservation and 432 management efforts in the GOM. Specifically, it would help identify species that are particularly 433 vulnerable to fishing during spawning and support the enactment of protection measures to 434 enhance resilience to fisheries exploitation (Erisman et al. 2018; Heyman et al. 2019). 435 Consideration of spawning dynamics in addition to more traditional life history traits, 436 management, and fisher behavior would help focus monitoring, research, and rebuilding efforts 437 on the most vulnerable species. In the U.S. GOM, stock assessments generate a probability 438 distribution function of overfishing limits that is converted to allowable biological catch based 439 on the Gulf Council’s risk tolerance for scientific uncertainty. Our findings imply a larger buffer 440 may be necessary to avoid overfishing for transient aggregating species, especially when this 441 spawning behavior is not explicitly considered in stock assessment. 442 There was a clear separation between life history traits and spawning behavior with 443 respect to loadings along the first two principal component axes. The component scores along 444 PC1 separated species based on life history traits, and PC2 was primarily loaded with spawning 445 behaviors. Spawning behavior provided a better distinction between overfished and not- 446 overfished species than the life history traits (i.e., more status information can be derived from 447 PC2 than PC1). Stock status of a fishery is influenced by multiple factors, which include 448 extrinsic components such as fishing practices (e.g., fishing effort; gear efficiency and 449 selectivity), management actions, stock assessment uncertainty stemming from data limitations 450 in the indicators (e.g., biomass), and changes in biological reference points (Rosenberg and 451 Restrepo 1994; Branch et al. 2011). We acknowledge that stock status may not completely 452 reflect the vulnerability and resilience of a species, but it is commonly assumed to be a function 453 of life history traits and historical exploitation (Costello et al. 2012). Manuscript to be reviewed In addition, stock status 454 provides a metric to compare the relationships between exploitation, life history and reproductive 455 behavior, especially with respect to identifying species that might be at risk of overfishing. 456 Several unassessed stocks with high intrinsic and extrinsic vulnerability were identified, 457 including deep-water grouper stocks such as Speckled Hind and Warsaw Grouper. In the nearby 458 U.S. South Atlantic region, spatial protections for these stocks have been considered, to reduce 459 post-release mortality (Farmer & Karnauskas 2013) and then implemented by creating a 460 declaration of five new Spawning Special Management Zones (sSMZs) in which bottom fishing 461 is restricted (South Atlantic Fisheries Management Council 2017). 462 As has been shown in several review studies on spawning aggregations (e.g., Erisman et 463 al. 2011; Sadovy de Mitcheson and Erisman 2012), species with spawning behaviors 376 Correlation analysis and PCA Along PC1, the eigenvectors for life history traits 389 had a greater influence than those for spawning behaviors. Lm was the greatest (-0.440), followed 390 by Linf (-0.430), Lmax (-0.405), and Wmax (-0.399). The PCA biplot (Figure 3) illustrates the 391 separation of species by the reproductive and life history parameters along the first two PC axes 392 within the context of stock status. Six of the species have been deemed not overfished. Ten 393 species have been overfished, including Atlantic Goliath Grouper and Nassau Grouper, which 394 are both closed to recreational and commercial fishing, as well as Red Drum, which is closed to 395 fishing in federal waters and to commercial fishing in most states along the GOM. Twelve 396 species have not been assessed, including three coastal species. The PC2 scores were positive for 397 7 out of 10 species that are currently or historically overfished whereas they were negative for 4 398 of the 6 not overfished species. The mean PC score was significantly higher for stocks currently 399 or historically overfished than not overfished for PC2 (t = 2.03, df = 9.25, p = 0.04) but not for 400 PC1 (t = -0.844, df = 15, p = 0.79) (Figure 4). Overfished status was closely associated with 401 positive changes in relative abundance and aggregation type, and negatively related to spawning 402 season duration. Peak spawning months was included in an additional PCA, but it did not change 403 the output and had very low eigenvectors, so it was not considered further (Figure S2.1). 404 PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 418 Discussion 419 Spawning behavior represents a separate and distinct aspect of fish ecology that is important to 420 consider for accurate predictions of vulnerability and resilience in exploited stocks (Erisman et 421 al. 2017a,b; Lowerre-Barbieri et al. 2017). Our results show that characteristics of spawning 422 behavior known to be associated with vulnerability to fishing are not directly related to life 423 history traits that are typically associated with vulnerability in 28 species of exploited marine 424 fishes in the GOM. Using PCA analysis, we demonstrated that the characterization of spawning 425 behaviors improved the identification of vulnerable and overfished species more than 426 “traditional” life history traits alone. We found that nearly all species showed susceptibility to 427 overfishing or becoming overfished during their spawning season due to very few state or federal 428 regulations to protect spawning fish in the GOM. However, there was a large range of intrinsic 429 vulnerability based on the diversity of spawning behavior and life history traits exhibited among 430 the species studied. Increased effort to understand patterns of spawning behavior and the 431 distribution of fishing effort and catch in relation to spawning would aid conservation and 432 management efforts in the GOM. Specifically, it would help identify species that are particularly 433 vulnerable to fishing during spawning and support the enactment of protection measures to 434 enhance resilience to fisheries exploitation (Erisman et al. 2018; Heyman et al. 2019). 435 Consideration of spawning dynamics in addition to more traditional life history traits, 436 management, and fisher behavior would help focus monitoring, research, and rebuilding efforts 437 on the most vulnerable species. In the U.S. GOM, stock assessments generate a probability 438 distribution function of overfishing limits that is converted to allowable biological catch based 439 on the Gulf Council’s risk tolerance for scientific uncertainty. Our findings imply a larger buffer 440 may be necessary to avoid overfishing for transient aggregating species, especially when this 441 spawning behavior is not explicitly considered in stock assessment. 442 There was a clear separation between life history traits and spawning behavior with 443 respect to loadings along the first two principal component axes. The component scores along 444 PC1 separated species based on life history traits, and PC2 was primarily loaded with spawning 445 behaviors. 418 Discussion 418 Discussion 419 Spawning behavior represents a separate and distinct aspect of fish ecology that is important to 420 consider for accurate predictions of vulnerability and resilience in exploited stocks (Erisman et 421 al. 2017a,b; Lowerre-Barbieri et al. 2017). Our results show that characteristics of spawning 422 behavior known to be associated with vulnerability to fishing are not directly related to life 423 history traits that are typically associated with vulnerability in 28 species of exploited marine 424 fishes in the GOM. Using PCA analysis, we demonstrated that the characterization of spawning 425 behaviors improved the identification of vulnerable and overfished species more than 426 “traditional” life history traits alone. We found that nearly all species showed susceptibility to 427 overfishing or becoming overfished during their spawning season due to very few state or federal 428 regulations to protect spawning fish in the GOM. However, there was a large range of intrinsic 429 vulnerability based on the diversity of spawning behavior and life history traits exhibited among 430 the species studied. Increased effort to understand patterns of spawning behavior and the 431 distribution of fishing effort and catch in relation to spawning would aid conservation and 432 management efforts in the GOM. Specifically, it would help identify species that are particularly 433 vulnerable to fishing during spawning and support the enactment of protection measures to 434 enhance resilience to fisheries exploitation (Erisman et al. 2018; Heyman et al. 2019). 435 Consideration of spawning dynamics in addition to more traditional life history traits, 436 management, and fisher behavior would help focus monitoring, research, and rebuilding efforts 437 on the most vulnerable species. In the U.S. GOM, stock assessments generate a probability 438 distribution function of overfishing limits that is converted to allowable biological catch based 439 on the Gulf Council’s risk tolerance for scientific uncertainty. Our findings imply a larger buffer 440 may be necessary to avoid overfishing for transient aggregating species, especially when this 441 i b h i i t li itl id d i t k t 435 Consideration of spawning dynamics in addition to more traditional life history traits, 436 management, and fisher behavior would help focus monitoring, research, and rebuilding efforts 437 on the most vulnerable species. In the U.S. Manuscript to be reviewed 2011; Maunder and Deriso 2013; Lowerre-Barbieri et al. 2015). 493 Large increases in relative abundance during spawning are linked to enhanced catchability for 494 most species that aggregate (Ellis and Wang 2007; Wilberg et al. 2009). For those species, the 495 increased catchability during spawning can also lead to hyperstability in which catch per unit 496 effort (CPUE) remains high even while the actual abundance of the stock decreases in response 497 to fishing mortality (Erisman et al. 2011). As a result, overfishing and stock declines may remain 498 undetected until after sudden, large decreases in catch or CPUE occur (Rose and Kulka 1999). 499 In contrast with a standard productivity susceptibility analysis (e.g., Patrick et al. 2010), 500 our vulnerability analysis followed the approach of Robinson and Samoilys (2013) and Robinson 501 (2015) by focusing specifically on identifying species that are highly sensitive to fishing during 502 spawning. Extrinsic vulnerability was high for 25 of the 28 species, which reflects the high level 503 of exposure to fishing during spawning throughout the GOM, and the existing threat of 504 overfishing via the targeted, pervasive exploitation of spawning fish by commercial and 505 recreational fishing activities (Grüss et al. 2018; Grüss et al. 2019; Heyman et al. 2019). The 506 level of compliance with any of the regulations may vary and can greatly influence the 507 effectiveness of the measure, but the regulations do serve as a metric of the perceived 508 vulnerability by managers. Future research focused on assessing fishing effort, catch, and 509 catchability (e.g., CPUE) patterns in relation to spawning as well as trends in market sales and 464 characterized by transient aggregations (that form over short durations and have large changes in 465 relative abundance) were more likely to be overfished than those that form resident aggregations, 466 mixed aggregations, or do not aggregate for spawning. Therefore, the consideration of spawning 467 behaviors offers a useful augmentation to the concept that slow-growing, long-lived and late 468 maturing species are always the most vulnerable to become overfished. Productivity 469 susceptibility analyses that rely heavily on life history traits have been used to identify at-risk 470 species (Hobday et al. 2011; Patrick et al. 2010), but they have also received criticism for their 471 inability to discriminate risk among species, except in the most extreme cases (Hordyk and 472 Carruthers 2018). Manuscript to be reviewed Since a relationship between spawning behaviors and vulnerability has been 473 observed (Erisman et al. 2011; Sadovy de Mitcheson and Erisman 2012), and spawning 474 behaviors are distinct from other life history traits, incorporating reproductive behaviors may 475 improve the discriminatory power of vulnerability analyses. Despite the issues and limitations 476 with discriminatory power, it is clear from the results of this study that the completion of a 477 vulnerability analysis that includes spawning behavior remains a valuable management exercise 478 to identify those species most vulnerable to fishing during spawning (i.e., extreme cases). The 479 results clearly define higher vulnerability and shows utility for prioritizing future research and 480 improving monitoring efforts. 481 Vulnerability scores displayed increasing trends with species that exhibit large changes in 482 relative abundance, have short spawning seasons and form transient aggregations, which is 483 expected because those components are included in the calculation of intrinsic vulnerability. 484 However, it is noteworthy that those trends persist because the overall vulnerability score 485 includes extrinsic vulnerability factors, and the life history composite was more heavily weighted 486 as an influence on vulnerability than any of the spawning behaviors. Species with short spawning 487 seasons that form transient spawning aggregations are prone to experience rapid depletion in 488 response to targeted fishing pressure of spawners and slower recovery rates due to impacts on 489 spatiotemporal egg production and stock recruitment relationships (Claro et al. 2009; Erisman et 490 al. 2012; Sadovy De Mitcheson and Erisman 2012). Yet, such declines can be difficult to detect 491 using traditional, catch-based methods of estimating abundance or when historical information is 492 unavailable (Erisman et al. 2011; Maunder and Deriso 2013; Lowerre-Barbieri et al. 2015). 493 Large increases in relative abundance during spawning are linked to enhanced catchability for 494 most species that aggregate (Ellis and Wang 2007; Wilberg et al. 2009). For those species, the 495 increased catchability during spawning can also lead to hyperstability in which catch per unit 496 effort (CPUE) remains high even while the actual abundance of the stock decreases in response 497 to fishing mortality (Erisman et al. 2011). As a result, overfishing and stock declines may remain 498 undetected until after sudden, large decreases in catch or CPUE occur (Rose and Kulka 1999). 499 In contrast with a standard productivity susceptibility analysis (e.g., Patrick et al. Manuscript to be reviewed 464 characterized by transient aggregations (that form over short durations and have large changes in 465 relative abundance) were more likely to be overfished than those that form resident aggregations, 466 mixed aggregations, or do not aggregate for spawning. Therefore, the consideration of spawning 467 behaviors offers a useful augmentation to the concept that slow-growing, long-lived and late 468 maturing species are always the most vulnerable to become overfished. Productivity 469 susceptibility analyses that rely heavily on life history traits have been used to identify at-risk 470 species (Hobday et al. 2011; Patrick et al. 2010), but they have also received criticism for their 471 inability to discriminate risk among species, except in the most extreme cases (Hordyk and 472 Carruthers 2018). Since a relationship between spawning behaviors and vulnerability has been 473 observed (Erisman et al. 2011; Sadovy de Mitcheson and Erisman 2012), and spawning 474 behaviors are distinct from other life history traits, incorporating reproductive behaviors may 475 improve the discriminatory power of vulnerability analyses. Despite the issues and limitations 476 with discriminatory power, it is clear from the results of this study that the completion of a 477 vulnerability analysis that includes spawning behavior remains a valuable management exercise 478 to identify those species most vulnerable to fishing during spawning (i.e., extreme cases). The 479 results clearly define higher vulnerability and shows utility for prioritizing future research and 480 improving monitoring efforts. 481 Vulnerability scores displayed increasing trends with species that exhibit large changes in 482 relative abundance, have short spawning seasons and form transient aggregations, which is 483 expected because those components are included in the calculation of intrinsic vulnerability. 484 However, it is noteworthy that those trends persist because the overall vulnerability score 485 includes extrinsic vulnerability factors, and the life history composite was more heavily weighted 486 as an influence on vulnerability than any of the spawning behaviors. Species with short spawning 487 seasons that form transient spawning aggregations are prone to experience rapid depletion in 488 response to targeted fishing pressure of spawners and slower recovery rates due to impacts on 489 spatiotemporal egg production and stock recruitment relationships (Claro et al. 2009; Erisman et 490 al. 2012; Sadovy De Mitcheson and Erisman 2012). Yet, such declines can be difficult to detect 491 using traditional, catch-based methods of estimating abundance or when historical information is 492 unavailable (Erisman et al. 418 Discussion 2011; Sadovy de Mitcheson and Erisman 2012), species with spawning behaviors 462 As has been shown in several review studies on spawning aggregations (e.g., Erisman et 463 al. 2011; Sadovy de Mitcheson and Erisman 2012), species with spawning behaviors PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 2010), 500 our vulnerability analysis followed the approach of Robinson and Samoilys (2013) and Robinson 501 (2015) by focusing specifically on identifying species that are highly sensitive to fishing during 502 spawning. Extrinsic vulnerability was high for 25 of the 28 species, which reflects the high level 503 of exposure to fishing during spawning throughout the GOM, and the existing threat of 504 overfishing via the targeted, pervasive exploitation of spawning fish by commercial and 505 recreational fishing activities (Grüss et al. 2018; Grüss et al. 2019; Heyman et al. 2019). The 506 level of compliance with any of the regulations may vary and can greatly influence the 507 effectiveness of the measure, but the regulations do serve as a metric of the perceived 508 vulnerability by managers. Future research focused on assessing fishing effort, catch, and 464 characterized by transient aggregations (that form over short durations and have large changes in 465 relative abundance) were more likely to be overfished than those that form resident aggregations, 466 mixed aggregations, or do not aggregate for spawning. Therefore, the consideration of spawning 467 behaviors offers a useful augmentation to the concept that slow-growing, long-lived and late 468 maturing species are always the most vulnerable to become overfished. Productivity 467 behaviors offers a useful augmentation to the concept that slow-growing, long-lived and late 468 maturing species are always the most vulnerable to become overfished. Productivity 469 susceptibility analyses that rely heavily on life history traits have been used to identify at-risk 470 species (Hobday et al. 2011; Patrick et al. 2010), but they have also received criticism for their 471 inability to discriminate risk among species, except in the most extreme cases (Hordyk and 472 Carruthers 2018). Since a relationship between spawning behaviors and vulnerability has been 473 observed (Erisman et al. 2011; Sadovy de Mitcheson and Erisman 2012), and spawning 474 behaviors are distinct from other life history traits, incorporating reproductive behaviors may 475 improve the discriminatory power of vulnerability analyses. Despite the issues and limitations 476 with discriminatory power, it is clear from the results of this study that the completion of a 477 vulnerability analysis that includes spawning behavior remains a valuable management exercise 478 to identify those species most vulnerable to fishing during spawning (i.e., extreme cases). The 479 results clearly define higher vulnerability and shows utility for prioritizing future research and 480 improving monitoring efforts. Manuscript to be reviewed 480 improving monitoring efforts. 481 Vulnerability scores displayed increasing trends with species that exhibit large changes in 482 relative abundance, have short spawning seasons and form transient aggregations, which is 483 expected because those components are included in the calculation of intrinsic vulnerability. 484 However, it is noteworthy that those trends persist because the overall vulnerability score 485 includes extrinsic vulnerability factors, and the life history composite was more heavily weighted 486 as an influence on vulnerability than any of the spawning behaviors. Species with short spawning 487 seasons that form transient spawning aggregations are prone to experience rapid depletion in 488 response to targeted fishing pressure of spawners and slower recovery rates due to impacts on 489 spatiotemporal egg production and stock recruitment relationships (Claro et al. 2009; Erisman et 490 al. 2012; Sadovy De Mitcheson and Erisman 2012). Yet, such declines can be difficult to detect 491 using traditional, catch-based methods of estimating abundance or when historical information is 492 unavailable (Erisman et al. 2011; Maunder and Deriso 2013; Lowerre-Barbieri et al. 2015). 493 Large increases in relative abundance during spawning are linked to enhanced catchability for 494 most species that aggregate (Ellis and Wang 2007; Wilberg et al. 2009). For those species, the 495 increased catchability during spawning can also lead to hyperstability in which catch per unit 496 effort (CPUE) remains high even while the actual abundance of the stock decreases in response 497 to fishing mortality (Erisman et al. 2011). As a result, overfishing and stock declines may remain 498 undetected until after sudden, large decreases in catch or CPUE occur (Rose and Kulka 1999). 499 In contrast with a standard productivity susceptibility analysis (e.g., Patrick et al. 2010), 500 our vulnerability analysis followed the approach of Robinson and Samoilys (2013) and Robinson 501 (2015) by focusing specifically on identifying species that are highly sensitive to fishing during 502 spawning. Extrinsic vulnerability was high for 25 of the 28 species, which reflects the high level 503 of exposure to fishing during spawning throughout the GOM, and the existing threat of 504 overfishing via the targeted, pervasive exploitation of spawning fish by commercial and 505 recreational fishing activities (Grüss et al. 2018; Grüss et al. 2019; Heyman et al. 2019). Manuscript to be reviewed Finally, Atlantic Goliath Grouper have been closed to 527 commercial and recreational harvest in state and federal waters of the GOM since 1990 due to 528 severe population declines associated with overfishing of their transient spawning aggregations 529 (Koenig et al. 2020). 530 Spotted Seatrout are classified as resident spawners, because they have a protracted 531 spawning season (April – September; Table S3), and spawning occurs on a daily basis within 532 small (i.e., 10s to a few hundred individuals) resident aggregations that persist at many locations 533 and habitat types along the coast and throughout estuaries (Saucier and Baltz 1993; Walters et al. 534 2009; Biggs et al. 2018). Although Spotted Seatrout are managed at the state level and not 535 assessed regionally, harvest levels have remained stable over the last 20 years in most areas 536 (NMFS 2018). Their resilience to persistent, intense fishing pressure is believed to be linked to 537 their high level of spawning productivity, both spatially and temporally (Biggs et al. 2018). For 538 this reason, traditional catch controls such as daily bag and minimum size limits appear to be 539 sufficient to maintain populations, and the targeted protection (e.g., seasonal restrictions or area 540 closures) of spawning fish is not a priority for management. ) p g p y g 541 Red snapper form schools of hundreds to thousands of fish throughout the year across a 542 broad array of coastal and offshore habitats (e.g. natural and artificial reefs, petroleum platforms) 543 in the GOM (Patterson et al. 2007; Erisman et al. 2020). At the population level, spawning 544 occurs continuously from May through September in small groups within these localized schools 545 rather than concentrating spawning at a few, highly populated sites (Porch et al. 2015; Farmer et 546 al. 2017; Glenn et al. 2017). While fishers have reported the previous existence of large 547 spawning aggregation sites of Red snapper (Lindeman et al. 2000), there is no evidence that 548 coordinated migrations to specific spawning sites occurs or that densities or relative abundance 549 at spawning sites is markedly higher than non-spawning sites (Patterson et al. 2007; Szedlmayer 550 and Bortone, 2019). Based on this information, we classified the species as resident aggregation 551 spawners, but it would also be reasonable to classify the species as non-aggregating based on the 552 specific criteria used (Domeier 2012; Claydon et al. 2014). Manuscript to be reviewed 2000), there is no evidence that 548 coordinated migrations to specific spawning sites occurs or that densities or relative abundance 549 at spawning sites is markedly higher than non-spawning sites (Patterson et al. 2007; Szedlmayer 550 and Bortone, 2019). Based on this information, we classified the species as resident aggregation 551 spawners, but it would also be reasonable to classify the species as non-aggregating based on the 552 specific criteria used (Domeier 2012; Claydon et al. 2014). Regardless, the more important point 553 is that the spatiotemporal dynamics of spawning behavior should confer reproductive resilience 554 to intense fishing effort during the spawning season (i.e. low overall vulnerability), which is 555 supported by the results of this study and the status of the regional stock. 510 values during spawning seasons would likely improve our understanding of vulnerability and the 511 impacts of harvesting spawning fish on stock resilience for exploited fishes in the GOM 512 (Erisman et al. 2018; Heyman et al. 2019). Understanding these behavioral impacts on resilience 513 allows managers to better target management measures on spawning for those species most 514 vulnerable to fishing due to such behaviors and discount those factors for species less vulnerable 515 to spawning behaviors. 514 vulnerable to fishing due to such behaviors and discount those factors for species less vulnerable 515 to spawning behaviors. 516 A detailed exploration of six important species further illustrates the management 517 implications of our results. These species include examples of both coastal and reef fishes whose 518 spawning behaviors fall along a continuum of intrinsic vulnerability scores (Table S4), ranging 519 between resident (Spotted Seatrout, Red Snapper), mixed (Red Drum and Greater Amberjack), 520 and transient aggregating species (Gag, Atlantic Goliath Grouper) (Table S3, Figure 5). Red 521 Drum and Spotted Seatrout are the two most important and valuable species for inshore 522 recreational fisheries throughout the GOM (Blanchet et al. 2001; NMFS 2018). Red Snapper 523 (Lutjanus campechanus) is arguably the most important and politically contentious commercial 524 and recreational fishery in the GOM (Farmer et al. 2020). Both Greater Amberjack and Gag are 525 also valuable, highly targeted species for both the recreational and commercial sectors in the 526 GOM that aggregate to spawn. Manuscript to be reviewed For 538 this reason, traditional catch controls such as daily bag and minimum size limits appear to be 539 sufficient to maintain populations, and the targeted protection (e.g., seasonal restrictions or area 540 closures) of spawning fish is not a priority for management. 541 Red snapper form schools of hundreds to thousands of fish throughout the year across a 542 broad array of coastal and offshore habitats (e.g. natural and artificial reefs, petroleum platforms) 543 in the GOM (Patterson et al. 2007; Erisman et al. 2020). At the population level, spawning 544 occurs continuously from May through September in small groups within these localized schools 545 rather than concentrating spawning at a few, highly populated sites (Porch et al. 2015; Farmer et 546 al. 2017; Glenn et al. 2017). While fishers have reported the previous existence of large 547 spawning aggregation sites of Red snapper (Lindeman et al. 2000), there is no evidence that 548 coordinated migrations to specific spawning sites occurs or that densities or relative abundance 549 at spawning sites is markedly higher than non-spawning sites (Patterson et al. 2007; Szedlmayer 550 and Bortone, 2019). Based on this information, we classified the species as resident aggregation 551 spawners, but it would also be reasonable to classify the species as non-aggregating based on the 552 specific criteria used (Domeier 2012; Claydon et al 2014) Regardless the more important point 510 values during spawning seasons would likely improve our understanding of vulnerability and the 511 impacts of harvesting spawning fish on stock resilience for exploited fishes in the GOM 512 (Erisman et al. 2018; Heyman et al. 2019). Understanding these behavioral impacts on resilience 513 allows managers to better target management measures on spawning for those species most 514 vulnerable to fishing due to such behaviors and discount those factors for species less vulnerable 515 to spawning behaviors. 516 A detailed exploration of six important species further illustrates the management 517 implications of our results. These species include examples of both coastal and reef fishes whose 518 spawning behaviors fall along a continuum of intrinsic vulnerability scores (Table S4), ranging 519 between resident (Spotted Seatrout, Red Snapper), mixed (Red Drum and Greater Amberjack), 520 and transient aggregating species (Gag, Atlantic Goliath Grouper) (Table S3, Figure 5). Manuscript to be reviewed Red 521 Drum and Spotted Seatrout are the two most important and valuable species for inshore 522 recreational fisheries throughout the GOM (Blanchet et al. 2001; NMFS 2018). Red Snapper 523 (Lutjanus campechanus) is arguably the most important and politically contentious commercial 524 and recreational fishery in the GOM (Farmer et al. 2020). Both Greater Amberjack and Gag are 525 also valuable, highly targeted species for both the recreational and commercial sectors in the 526 GOM that aggregate to spawn. Finally, Atlantic Goliath Grouper have been closed to 527 commercial and recreational harvest in state and federal waters of the GOM since 1990 due to 528 severe population declines associated with overfishing of their transient spawning aggregations 529 (Koenig et al. 2020). 530 Spotted Seatrout are classified as resident spawners, because they have a protracted 531 spawning season (April – September; Table S3), and spawning occurs on a daily basis within 532 small (i.e., 10s to a few hundred individuals) resident aggregations that persist at many locations 533 and habitat types along the coast and throughout estuaries (Saucier and Baltz 1993; Walters et al. 534 2009; Biggs et al. 2018). Although Spotted Seatrout are managed at the state level and not 535 assessed regionally, harvest levels have remained stable over the last 20 years in most areas 536 (NMFS 2018). Their resilience to persistent, intense fishing pressure is believed to be linked to 537 their high level of spawning productivity, both spatially and temporally (Biggs et al. 2018). For 538 this reason, traditional catch controls such as daily bag and minimum size limits appear to be 539 sufficient to maintain populations, and the targeted protection (e.g., seasonal restrictions or area 540 closures) of spawning fish is not a priority for management. 541 Red snapper form schools of hundreds to thousands of fish throughout the year across a 542 broad array of coastal and offshore habitats (e.g. natural and artificial reefs, petroleum platforms) 543 in the GOM (Patterson et al. 2007; Erisman et al. 2020). At the population level, spawning 544 occurs continuously from May through September in small groups within these localized schools 545 rather than concentrating spawning at a few, highly populated sites (Porch et al. 2015; Farmer et 546 al. 2017; Glenn et al. 2017). While fishers have reported the previous existence of large 547 spawning aggregation sites of Red snapper (Lindeman et al. Manuscript to be reviewed The 506 level of compliance with any of the regulations may vary and can greatly influence the 507 effectiveness of the measure, but the regulations do serve as a metric of the perceived 508 vulnerability by managers. Future research focused on assessing fishing effort, catch, and 509 catchability (e.g., CPUE) patterns in relation to spawning as well as trends in market sales and PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 510 values during spawning seasons would likely improve our understanding of vulnerability and the 511 impacts of harvesting spawning fish on stock resilience for exploited fishes in the GOM 512 (Erisman et al. 2018; Heyman et al. 2019). Understanding these behavioral impacts on resilience 513 allows managers to better target management measures on spawning for those species most 514 vulnerable to fishing due to such behaviors and discount those factors for species less vulnerable 515 to spawning behaviors. 516 A detailed exploration of six important species further illustrates the management 517 implications of our results. These species include examples of both coastal and reef fishes whose 518 spawning behaviors fall along a continuum of intrinsic vulnerability scores (Table S4), ranging 519 between resident (Spotted Seatrout, Red Snapper), mixed (Red Drum and Greater Amberjack), 520 and transient aggregating species (Gag, Atlantic Goliath Grouper) (Table S3, Figure 5). Red 521 Drum and Spotted Seatrout are the two most important and valuable species for inshore 522 recreational fisheries throughout the GOM (Blanchet et al. 2001; NMFS 2018). Red Snapper 523 (Lutjanus campechanus) is arguably the most important and politically contentious commercial 524 and recreational fishery in the GOM (Farmer et al. 2020). Both Greater Amberjack and Gag are 525 also valuable, highly targeted species for both the recreational and commercial sectors in the 526 GOM that aggregate to spawn. Finally, Atlantic Goliath Grouper have been closed to 527 commercial and recreational harvest in state and federal waters of the GOM since 1990 due to 528 severe population declines associated with overfishing of their transient spawning aggregations 529 (Koenig et al. 2020). 530 Spotted Seatrout are classified as resident spawners, because they have a protracted 531 spawning season (April – September; Table S3), and spawning occurs on a daily basis within 532 small (i.e., 10s to a few hundred individuals) resident aggregations that persist at many locations 533 and habitat types along the coast and throughout estuaries (Saucier and Baltz 1993; Walters et al. 534 2009; Biggs et al. 2018). Although Spotted Seatrout are managed at the state level and not 535 assessed regionally, harvest levels have remained stable over the last 20 years in most areas 536 (NMFS 2018). Their resilience to persistent, intense fishing pressure is believed to be linked to 537 their high level of spawning productivity, both spatially and temporally (Biggs et al. 2018). Manuscript to be reviewed Commercial landings in federal waters are 578 prohibited during the peak spawning season (March-May), while recreational landings for 579 Greater Amberjack are closed during the end and after the spawning season (June-July) (SEDAR 580 2014a). Recreational landings are consistently higher during the spawning season than the non- 581 spawning season (Kobara et al. 2017; Heyman et al. 2019), but the impacts of fishing during 582 spawning have not been evaluated. 583 Gag and Goliath Grouper are both transient spawners (Table S3) with a high intrinsic 584 vulnerability to fishing during spawning in the GOM (Table S4). Gag exhibit complex 585 reproductive ecology in which females form pre-spawning aggregations in coastal waters before 586 migrating offshore to deep-water spawning sites in the winter where males tend to occur year- 587 round (Koenig et al. 1996; Carruthers et al. 2015). The species is also protogynous, in which 588 female to male sex change occurs during both the pre-spawning aggregations and at the 589 spawning grounds (Lowerre-Barbieri et al. 2020). Unlike typical transient spawning aggregations 590 that involve group spawning and hundreds to thousands of fish, Gag aggregations are comprised 591 of tens to a few hundred fish and courtship occurs in pairs (Gilmore and Jones 1992; Coleman et 592 al. 1996; Lowerre-Barbieri et al. 2020). GOM Gag were overfished and undergoing overfishing 593 in the 1990s, with heavy fishing on Gag spawning aggregations that resulted in the severe 594 reduction of males, which contributed to stock declines (Coleman et al. 1996). Consequently, the 595 protection of spawning Gag has long been a focal point for management and the rebuilding of the 596 stock (SEDAR 2014b; SEDAR 2016b). Recent studies have demonstrated that Gag biomass and 597 fishing pressure are highest in nearshore waters, pre-spawning aggregations are a spatiotemporal 598 bottleneck to population productivity, and current regulations are not sufficient for the male 599 population to recover to historic levels (Carruthers et al. 2015; Lowerre-Barbieri et al. 2020). 600 Similar to Gag, Atlantic Goliath Grouper may migrate long distances (up to 500 km) to specific 601 sites to spawn within transient spawning aggregations that form from late July through October p g ( ; ) 561 Both Red Drum and Greater Amberjack (Seriola dumerili) are classified as mixed 562 spawners (Table S3) with an intermediate level of intrinsic vulnerability to fishing in relation to 563 spawning (Table S4). Manuscript to be reviewed 556 Fishing effort and landings for Red snapper in federal waters peak during the summer months in 557 association with the recreational fishing season, which coincides with the peak spawning season 558 for this species (Heyman et al. 2019). Nevertheless, rebuilding of the stock has occurred in 559 response to strict catch and effort controls implemented at state and federal levels that did not 560 focus on spawning (SEDAR 2018; Farmer et al. 2020). 557 association with the recreational fishing season, which coincides with the peak spawning season 558 for this species (Heyman et al. 2019). Nevertheless, rebuilding of the stock has occurred in 559 response to strict catch and effort controls implemented at state and federal levels that did not 560 focus on spawning (SEDAR 2018; Farmer et al. 2020). 561 Both Red Drum and Greater Amberjack (Seriola dumerili) are classified as mixed 562 spawners (Table S3) with an intermediate level of intrinsic vulnerability to fishing in relation to 563 spawning (Table S4). Red Drum are known to form large transient (hundreds to tens of 564 thousands of individuals) spawning aggregations that occur predictably at the mouths of estuaries 565 and tidal inlets throughout the GOM from August to November, but these aggregations are not 566 limited to discrete locations within such habitats, adults school year round, and resident 567 aggregations have been reported as well (Pearson 1928; Holt 2008; Lowerre-Barbieri et al. 2016; 568 Lowerre-Barbieri et al. 2019). Although the fishery is closed in federal waters, it is open to 569 recreational fishing in state waters throughout the GOM and commercially in state waters in 570 Mississippi (Heyman et al. 2019). A rapid increase in the targeted harvesting of adult fish from 571 their spawning aggregations by the commercial fishery during the mid-1980s led to overfishing 572 and the closure of the federal fishery (Porch 2000). However, recreational anglers continue to 573 target these well-known spawning aggregations during the late summer and fall months and this 574 activity peaks during the spawning season, but the impacts of this interaction on the population 575 or the fishery have not been assessed. Similar to Red Drum, Greater Amberjack spawning 576 behavior is classified as mixed with a spawning season from March to June in the GOM with 577 regional variations in seasonality (Table S3). Manuscript to be reviewed Regardless, the more important point 553 is that the spatiotemporal dynamics of spawning behavior should confer reproductive resilience 554 to intense fishing effort during the spawning season (i.e. low overall vulnerability), which is 555 supported by the results of this study and the status of the regional stock. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Red Drum are known to form large transient (hundreds to tens of 564 thousands of individuals) spawning aggregations that occur predictably at the mouths of estuaries 565 and tidal inlets throughout the GOM from August to November, but these aggregations are not 566 limited to discrete locations within such habitats, adults school year round, and resident 567 aggregations have been reported as well (Pearson 1928; Holt 2008; Lowerre-Barbieri et al. 2016; 568 Lowerre-Barbieri et al. 2019). Although the fishery is closed in federal waters, it is open to 569 recreational fishing in state waters throughout the GOM and commercially in state waters in 570 Mississippi (Heyman et al. 2019). A rapid increase in the targeted harvesting of adult fish from 571 their spawning aggregations by the commercial fishery during the mid-1980s led to overfishing 572 and the closure of the federal fishery (Porch 2000). However, recreational anglers continue to 573 target these well-known spawning aggregations during the late summer and fall months and this 574 activity peaks during the spawning season, but the impacts of this interaction on the population 575 or the fishery have not been assessed. Similar to Red Drum, Greater Amberjack spawning 576 behavior is classified as mixed with a spawning season from March to June in the GOM with 577 regional variations in seasonality (Table S3). Commercial landings in federal waters are 578 prohibited during the peak spawning season (March-May), while recreational landings for 579 Greater Amberjack are closed during the end and after the spawning season (June-July) (SEDAR 580 2014a). Recreational landings are consistently higher during the spawning season than the non- 581 spawning season (Kobara et al. 2017; Heyman et al. 2019), but the impacts of fishing during 582 spawning have not been evaluated. p g 583 Gag and Goliath Grouper are both transient spawners (Table S3) with a high intrinsic 584 vulnerability to fishing during spawning in the GOM (Table S4). Gag exhibit complex 585 reproductive ecology in which females form pre-spawning aggregations in coastal waters before 586 migrating offshore to deep-water spawning sites in the winter where males tend to occur year- 587 round (Koenig et al. 1996; Carruthers et al. 2015). The species is also protogynous, in which 588 female to male sex change occurs during both the pre-spawning aggregations and at the 589 spawning grounds (Lowerre-Barbieri et al. 2020). Manuscript to be reviewed This represents a critical data gap relevant to the 612 management of fish stocks in the GOM. Further, changes in relative abundance are directly tied 613 to changes in catchability, which is widely accepted as a crucial component of vulnerability 614 (Arreguín-Sánchez 1999; Salthaug and Aanes 2003; Wilberg et al. 2009) and as mentioned 615 above, hyperstability. Our results also demonstrate that relative abundance is an important factor 616 as it was not correlated to any life history traits. Therefore, it represents an independent aspect of 617 reproductive behavior relevant to assessing vulnerability and should be the focus of further 618 research for those species that form spawning aggregations that are targeted and heavily 619 exploited by recreational or commercial fishing. 620 In contrast to changes in the relative abundance of fish during spawning, spawning 621 seasons are well documented in the literature. However, spawning seasons are usually reported 622 on a monthly scale, which does not capture the finer variations in spawning periodicity, the 623 degree of aggregating behavior, or the distribution of spawning sites among species. For 624 example, Cubera Snapper spawn from June through September, but aggregations only form at a 625 few sites for 1-2 weeks each month, and actual spawning is restricted to just a few days within a 626 single aggregation period (Heyman et al. 2005; Biggs and Nemeth 2016). Nassau Grouper, 627 Mutton Snapper, and other GOM species in the grouper-snapper complex exhibit similar 628 behaviors in which spawning occurs over a few days within large aggregations that form at 629 specific sites in synchrony with lunar (or semilunar) rhythms (Heyman and Kjerve 2008). On the 630 other end of the behavioral spectrum, species such as Spotted Seatrout and Red Snapper can 631 spawn daily (population level) over the course of 5 to 6 months, and spawning sites are 632 numerous and widely distributed (Walters et al., 2009; Lowerre-Barbieri et al., 2015; Glenn et al. 633 2017; Biggs et al. 2018). Differences in the spatiotemporal dynamics of spawning affects their 634 respective vulnerability, but much of that variation is captured in the categorization of 635 aggregation type (e.g., Cubera Snapper form transient aggregations and Spotted Seatrout form 636 resident aggregations). 637 There is also some potential variation with respect to aggregation type across the region. Manuscript to be reviewed 638 For example studies on the reproductive behavior of Red Grouper conducted in the U S GOM 602 (Koenig et al. 2017). Severe population declines occurred in the GOM from the 1950s through 603 the 1980s due in part to overfishing of spawning aggregations, but the species has shown signs of 604 recovery following their complete protection from exploitation in state and federal waters in 605 1990 (Koenig et al. 2011). 606 One of the challenges to incorporating details of reproductive behavior within 607 assessments and management is the paucity of information available. For example, discrete 608 quantitative information on changes in the relative abundance (e.g., increases in fish density and 609 abundance during formation of spawning aggregations) of fish within a given area in relation to 610 spawning activity is rare and currently, has only published for red drum (Lowerre-Barbieri et al., 611 2019) and Gag (Lowerre-Barbieri et al., 2020). This represents a critical data gap relevant to the 612 management of fish stocks in the GOM. Further, changes in relative abundance are directly tied 613 to changes in catchability, which is widely accepted as a crucial component of vulnerability 614 (Arreguín-Sánchez 1999; Salthaug and Aanes 2003; Wilberg et al. 2009) and as mentioned 615 above, hyperstability. Our results also demonstrate that relative abundance is an important factor 616 as it was not correlated to any life history traits. Therefore, it represents an independent aspect of 617 reproductive behavior relevant to assessing vulnerability and should be the focus of further 618 research for those species that form spawning aggregations that are targeted and heavily 619 exploited by recreational or commercial fishing. 620 In contrast to changes in the relative abundance of fish during spawning, spawning 621 seasons are well documented in the literature. However, spawning seasons are usually reported 622 on a monthly scale, which does not capture the finer variations in spawning periodicity, the 623 degree of aggregating behavior, or the distribution of spawning sites among species. For 624 example, Cubera Snapper spawn from June through September, but aggregations only form at a 625 few sites for 1-2 weeks each month, and actual spawning is restricted to just a few days within a 626 single aggregation period (Heyman et al. 2005; Biggs and Nemeth 2016). Manuscript to be reviewed 602 (Koenig et al. 2017). Severe population declines occurred in the GOM from the 1950s through 603 the 1980s due in part to overfishing of spawning aggregations, but the species has shown signs of 604 recovery following their complete protection from exploitation in state and federal waters in 605 1990 (Koenig et al. 2011). 606 One of the challenges to incorporating details of reproductive behavior within 607 assessments and management is the paucity of information available. For example, discrete 608 quantitative information on changes in the relative abundance (e.g., increases in fish density and 609 abundance during formation of spawning aggregations) of fish within a given area in relation to 610 spawning activity is rare and currently, has only published for red drum (Lowerre-Barbieri et al., 611 2019) and Gag (Lowerre-Barbieri et al., 2020). This represents a critical data gap relevant to the 612 management of fish stocks in the GOM. Further, changes in relative abundance are directly tied 613 to changes in catchability, which is widely accepted as a crucial component of vulnerability 614 (Arreguín-Sánchez 1999; Salthaug and Aanes 2003; Wilberg et al. 2009) and as mentioned 615 above, hyperstability. Our results also demonstrate that relative abundance is an important factor 616 as it was not correlated to any life history traits. Therefore, it represents an independent aspect of 617 reproductive behavior relevant to assessing vulnerability and should be the focus of further 618 research for those species that form spawning aggregations that are targeted and heavily 619 exploited by recreational or commercial fishing. 620 In contrast to changes in the relative abundance of fish during spawning, spawning 621 seasons are well documented in the literature. However, spawning seasons are usually reported 622 on a monthly scale, which does not capture the finer variations in spawning periodicity, the 623 degree of aggregating behavior, or the distribution of spawning sites among species. For 624 example, Cubera Snapper spawn from June through September, but aggregations only form at a 625 few sites for 1-2 weeks each month, and actual spawning is restricted to just a few days within a 626 single aggregation period (Heyman et al. 2005; Biggs and Nemeth 2016). Manuscript to be reviewed Unlike typical transient spawning aggregations 590 that involve group spawning and hundreds to thousands of fish, Gag aggregations are comprised 591 of tens to a few hundred fish and courtship occurs in pairs (Gilmore and Jones 1992; Coleman et 592 al. 1996; Lowerre-Barbieri et al. 2020). GOM Gag were overfished and undergoing overfishing 593 in the 1990s, with heavy fishing on Gag spawning aggregations that resulted in the severe 594 reduction of males, which contributed to stock declines (Coleman et al. 1996). Consequently, the 595 protection of spawning Gag has long been a focal point for management and the rebuilding of the 596 stock (SEDAR 2014b; SEDAR 2016b). Recent studies have demonstrated that Gag biomass and 597 fishing pressure are highest in nearshore waters, pre-spawning aggregations are a spatiotemporal 598 bottleneck to population productivity, and current regulations are not sufficient for the male 599 population to recover to historic levels (Carruthers et al. 2015; Lowerre-Barbieri et al. 2020). 600 Similar to Gag, Atlantic Goliath Grouper may migrate long distances (up to 500 km) to specific 601 sites to spawn within transient spawning aggregations that form from late July through October PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Nassau Grouper, 627 Mutton Snapper, and other GOM species in the grouper-snapper complex exhibit similar 628 behaviors in which spawning occurs over a few days within large aggregations that form at 629 specific sites in synchrony with lunar (or semilunar) rhythms (Heyman and Kjerve 2008). On the 630 other end of the behavioral spectrum, species such as Spotted Seatrout and Red Snapper can 631 spawn daily (population level) over the course of 5 to 6 months, and spawning sites are 632 numerous and widely distributed (Walters et al., 2009; Lowerre-Barbieri et al., 2015; Glenn et al. 633 2017; Biggs et al. 2018). Differences in the spatiotemporal dynamics of spawning affects their 634 respective vulnerability, but much of that variation is captured in the categorization of 635 aggregation type (e.g., Cubera Snapper form transient aggregations and Spotted Seatrout form 636 resident aggregations). 637 There is also some potential variation with respect to aggregation type across the region. 638 For example, studies on the reproductive behavior of Red Grouper conducted in the U.S. GOM 639 have all concluded that this species does not migrate during spawning or form spawning 640 aggregations. Red grouper exhibit a haremic mating system in which resident males pair spawn 641 with individual females that reside in small groups within its home territory (Coleman et al. 642 2011; Nelson et al. 2011; Wall et al. 2011). Conversely, studies of Red Grouper fisheries and 643 populations off Campeche, Mexico in the southern GOM do refer to spawning aggregations 644 based on inferences drawn from increases in catchability and catch-per-unit effort during the 602 (Koenig et al. 2017). Severe population declines occurred in the GOM from the 1950s through 603 the 1980s due in part to overfishing of spawning aggregations, but the species has shown signs of 604 recovery following their complete protection from exploitation in state and federal waters in 605 1990 (Koenig et al. 2011). 606 One of the challenges to incorporating details of reproductive behavior within 607 assessments and management is the paucity of information available. For example, discrete 608 quantitative information on changes in the relative abundance (e.g., increases in fish density and 609 abundance during formation of spawning aggregations) of fish within a given area in relation to 610 spawning activity is rare and currently, has only published for red drum (Lowerre-Barbieri et al., 611 2019) and Gag (Lowerre-Barbieri et al., 2020). Manuscript to be reviewed 48 seasonal restrictions), including investigations of behavioral traits that may vary on regional 49 scales within the GOM. 649 scales within the GOM. 650 The benefit of the approach used in this study is that it can be used to inform state and 651 federal management groups by identifying and prioritizing which species should be targeted for 652 research, monitoring, and management actions in the absence of formal stock assessments. Of 653 the species in our analysis that have not been assessed through formal stock assessments, Cubera 654 Snapper, Warsaw Grouper, Sheepshead, Southern Flounder, and Black Drum are likely at risk of 655 overfishing due to strong interactions between fishing and spawning. These species all exhibit 656 reproductive behaviors that indicate they should be assessed in the GOM and that interactions 657 between fishing and spawning should be investigated further to determine if spawning fish 658 would benefit from additional protection (e.g., seasonal catch limits or inclusion with marine 659 protected areas). Additionally, this approach can be applied to other regions and fisheries. As one 660 example, the U.S. South Atlantic region contains many of the same harvested species as the 661 GOM, but the stocks are managed separately and have different regulations in both state and 662 federal waters. Additionally, the spawning seasons, locations, and environmental cues to 663 spawning are well described for many stocks in the U.S. South Atlantic (Farmer et al. 2017). 664 Therefore, our analysis can be transferred to that region with only minor adjustments to account 665 for the differences in extrinsic vulnerability parameters. Our analysis also provides the 666 framework to identify and prioritize management and monitoring of species that are vulnerable 667 to fishing during spawning, which is especially important and applicable in areas where 668 management resources, monitoring effort and fisheries data are limited (e.g., Caribbean, Tropical 669 Eastern Pacific, Indo-Pacific) (Salas et al. 2007; Gill et al. 2017). Moreover, quantitative 670 information on the periodicity and frequency of spawning and other spatiotemporal aspects of 671 spawning behavior can be directly incorporated into estimates of annual reproductive output and 672 spawning potential ratios (Cooper et al. 2013; Erisman et al. 2014, 2020; Lowerre-Barbieri et al. 673 2017), thus representing a clear pathway to incorporate such valuable information within formal 674 stock assessments. Manuscript to be reviewed Nassau Grouper, 627 Mutton Snapper, and other GOM species in the grouper-snapper complex exhibit similar 628 behaviors in which spawning occurs over a few days within large aggregations that form at 629 specific sites in synchrony with lunar (or semilunar) rhythms (Heyman and Kjerve 2008). On the 630 other end of the behavioral spectrum, species such as Spotted Seatrout and Red Snapper can 631 spawn daily (population level) over the course of 5 to 6 months, and spawning sites are 632 numerous and widely distributed (Walters et al., 2009; Lowerre-Barbieri et al., 2015; Glenn et al. 633 2017; Biggs et al. 2018). Differences in the spatiotemporal dynamics of spawning affects their 634 respective vulnerability, but much of that variation is captured in the categorization of 635 aggregation type (e.g., Cubera Snapper form transient aggregations and Spotted Seatrout form 636 resident aggregations). 637 There is also some potential variation with respect to aggregation type across the region. 638 For example, studies on the reproductive behavior of Red Grouper conducted in the U.S. GOM 639 have all concluded that this species does not migrate during spawning or form spawning 640 aggregations. Red grouper exhibit a haremic mating system in which resident males pair spawn 641 with individual females that reside in small groups within its home territory (Coleman et al. 642 2011; Nelson et al. 2011; Wall et al. 2011). Conversely, studies of Red Grouper fisheries and 643 populations off Campeche, Mexico in the southern GOM do refer to spawning aggregations 644 based on inferences drawn from increases in catchability and catch-per-unit effort during the 645 spawning season at spawning sites (López-Rocha et al. 2009; López-Rocha and Arreguín- 646 Sánchez 2013). These discrepancies further demonstrate the importance of fisheries-independent 647 research on spawning dynamics and behavior to inform fisheries management (e.g., areal or PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 729 692 prioritizing research and management around those species showing the greatest vulnerability to 693 fishing during spawning. 692 prioritizing research and management around those species showing the greatest vulnerability to 693 fishing during spawning. 694 692 prioritizing research and management around those species showing the greatest vulnerability to 693 fishing during spawning. 694 695 Acknowledgements 696 We are grateful to the many people that provided data, feedback, or other valuable support for 697 this study including M. Karnauskas, W. Stearns, F. Parker, J. Lartigue, C. Young, A. Grüss, D. 698 DeMaria, S. Hickman, K. Guindon, S. Cantrell, V. Ventura, W. Werner, D. Naar, S. Murawski, 699 M. Russell, C. Taylor, T. Kellison, R. Williams, R. Crabtree, L. Crabtree, E. Reed, C. Koenig, K 700 Boswell, S. Fulton, J. Locascio, J. Brenner, B. Kirkpatrick, B. Gallaway, K. McCain, T. 701 Wheatley, H. Binns, A. Acosta, A. Trotter, F. Giordano, F. Helies, R. Ellis, H. Staley, K. 702 Faherty-Walia, R. Devictor, G.P. Schmall, B. Gorst, T. Loughran, L. Yeager, L. Fuiman, and 703 many others. 704 705 Disclaimer: The scientific results and conclusions, as well as any views or opinions expressed 706 herein, are those of the author(s) and do not necessarily reflect those of NOAA or the 707 Department of Commerce. 708 709 710 Author Contributions 676 Conclusions 677 Exploited marine fish species displaying similar life history characteristics can be very different 678 with regards to their vulnerability and resilience to fishing when reproductive behavior is 679 considered. This distinction is meaningful, because spawning behaviors are underrepresented in 680 the conservation or management plans of most marine fishes species in the GOM and elsewhere. 681 Based on the results of this study, it is important to consider both aspects within management 682 frameworks, particularly for those species known to form large, transient spawning aggregations 683 that are targeted by commercial and recreational fishing activities. Assessing the vulnerability of 684 a marine fish species based on their size, longevity, and maturation rate alone may not capture 685 the true complexity of their biology and likewise their resilience or vulnerability to fishing 686 pressure, particularly when fishing efforts target spawning sites, seasons, or the actual spawning 687 period itself. As a result, incorporating spawning behavior within such analyses can significantly 688 improve our understanding of vulnerability. By extension, the lack of biological and fisheries 689 information on reproductive behavior hampers efforts to maintain healthy, productive stocks to 690 benefit fisheries and ecosystems. Moreover, these types of bivariate frameworks can be a 691 valuable tool for understanding the main factors underlying vulnerability of marine fishes and for PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) 695 Acknowledgements 705 Disclaimer: The scientific results and conclusions, as well as any views or opinions expressed 706 herein, are those of the author(s) and do not necessarily reflect those of NOAA or the 707 Department of Commerce. 708 Manuscript to be reviewed 692 prioritizing research and management around those species showing the greatest vulnerability to 693 fishing during spawning. 694 695 Acknowledgements 696 We are grateful to the many people that provided data, feedback, or other valuable support for 697 this study including M. Karnauskas, W. Stearns, F. Parker, J. Lartigue, C. Young, A. Grüss, D. 698 DeMaria, S. Hickman, K. Guindon, S. Cantrell, V. Ventura, W. Werner, D. Naar, S. Murawski, 699 M. Russell, C. Taylor, T. Kellison, R. Williams, R. Crabtree, L. Crabtree, E. Reed, C. Koenig, K. 700 Boswell, S. Fulton, J. Locascio, J. Brenner, B. Kirkpatrick, B. Gallaway, K. McCain, T. 701 Wheatley, H. Binns, A. Acosta, A. Trotter, F. Giordano, F. Helies, R. Ellis, H. Staley, K. 702 Faherty-Walia, R. Devictor, G.P. Schmall, B. Gorst, T. Loughran, L. Yeager, L. Fuiman, and 703 many others. 704 705 Disclaimer: The scientific results and conclusions, as well as any views or opinions expressed 706 herein, are those of the author(s) and do not necessarily reflect those of NOAA or the 707 Department of Commerce. 708 709 710 Author Contributions 711  Christopher R. Biggs conceived the study, compiled the data, designed the analytical 712 framework, analyzed the data, prepared the figures and tables, wrote the paper, reviewed and 713 edited drafts of the paper, and approved the final draft. 714  William Heyman compiled the data, wrote the paper, reviewed and edited drafts of the paper, 715 and approved the final draft. 716  Nicholas A. Farmer compiled the data, reviewed and edited drafts of the paper, and approved 717 the final draft. 718  Derek G. Bolser compiled the data, analyzed the data, reviewed and edited drafts of the 719 paper, and approved the final draft. 720  Shinichi Kobara compiled the data, reviewed and edited drafts of the paper, and approved the 721 final draft. 722  Jan Robinson designed the analytical framework, reviewed and edited drafts of the paper, 723 and approved the final draft. 724  Susan K. Lowerre-Barbieri reviewed and edited drafts of the paper and approved the final 725 draft. 726  Brad E. Erisman led the project associated with this study as senior investigator, conceived of 727 the study, compiled the data, designed the analytical framework, analyzed the data, wrote the 728 paper, reviewed and edited drafts of the paper, and approved the final draft. 731 References 732 Adams, P. B. 1980. Life history patterns in marine fishes and their consequences for fisheries 733 management. Fishery Bulletin 78:1–12. 733 management. Fishery Bulletin 78:1–12. 734 Arreguín-Sánchez, F. 1996. Catchability: a key parameter for fish stock assessment. Reviews in 735 Fish Biology and Fisheries 6:221-242. 736 Arreguín-Sánchez, F. 1999. Catchability estimates and their application to the red grouper 737 (Epinephelus morio) fishery of the Campeche Bank, Mexico. Fishery Bulletin, 1999 738 24(3):282. ( ) 739 Biggs, C. R., B. E. Erisman, W. D. Heyman, S. Kobara, N. Farmer, S. Lowerre-Barbieri, M. 740 Karnauskas, and J. Brenner. 2017. Species selection for evaluation with spawning 741 aggregation and fisheries management criteria. Cooperative monitoring program for 742 spawning aggregations in the Gulf of Mexico. Available at: 743 https://geo.gcoos.org/restore/selection_criteria. 744 Biggs, C. R., S. Lowerre-Barbieri, and B. Erisman. 2018. Reproductive resilience of an estuarine 745 fish in the eye of a hurricane. Biology Letters 14:10–14. 746 Biggs, C. R., and R. Nemeth. 2016. Spatial and temporal movement patterns of two snapper 747 species at a multi-species spawning aggregation. Marine Ecology Progress Series 558:129– 748 142. 749 Blanchet, H., M. Van Hoose, L. McEachron, B. Muller, J. Warren, J. Gill, T. Waldrop, J. Waller, 750 C. Adams, R. B. Ditton, D. Shively, and S. VanderKooy. 2001. The spotted seatrout fishery 751 of the Gulf of Mexico, United States: A regional management plan. Gulf States Marine 752 Fisheries Commission 87:1–204. 753 Branch, T. A., O. P. Jensen, D. Ricard, Y. Ye, and R. Hilborn. 2011. Contrasting global trends in 754 marine fishery status obtained from catches and from stock assessments. Conservation 755 Biology 25(4):777–786. 756 Brown-Peterson, N. J., D. M. Wyanski, F. Saborido-Rey, B. J. Macewicz, and S. K. Lowerre- 757 Barbieri. 2011. A standardized terminology for describing reproductive development in 758 fishes. Marine and Coastal Fisheries, 3(1):52-70. ( ) 759 Carruthers, T.R., Walter, J.F., McAllister, M.K., Bryan, M.D., and M. Wilberg. 2015. Modelling 760 age-dependent movement: an application to red and gag groupers in the Gulf of Mexico. 761 Canadian Journal of Fisheries and Aquatic Science 72: 1159−1176. 762 Cheung, W. W. L., T. J. Pitcher, and D. Pauly. 2005. A fuzzy logic expert system to estimate 763 intrinsic extinction vulnerabilities of marine fishes to fishing. Biological Conservation 764 124(1):97–111. 765 Cheung, W. W. L., Y. Sadovy de Mitcheson, M. T. Braynen, and L. G. Gittens. 2013. 710 Author Contributions 711  Christopher R. Biggs conceived the study, compiled the data, designed the analytical 712 framework, analyzed the data, prepared the figures and tables, wrote the paper, reviewed and 713 edited drafts of the paper, and approved the final draft. 714  William Heyman compiled the data, wrote the paper, reviewed and edited drafts of the paper, 715 and approved the final draft. 716  Nicholas A. Farmer compiled the data, reviewed and edited drafts of the paper, and approved 717 the final draft. 718  Derek G. Bolser compiled the data, analyzed the data, reviewed and edited drafts of the 719 paper, and approved the final draft. 720  Shinichi Kobara compiled the data, reviewed and edited drafts of the paper, and approved the 721 final draft. 722  Jan Robinson designed the analytical framework, reviewed and edited drafts of the paper, 723 and approved the final draft. 724  Susan K. Lowerre-Barbieri reviewed and edited drafts of the paper and approved the final 725 draft. 726  Brad E. Erisman led the project associated with this study as senior investigator, conceived of 727 the study, compiled the data, designed the analytical framework, analyzed the data, wrote the 728 paper, reviewed and edited drafts of the paper, and approved the final draft. 729 711  Christopher R. Biggs conceived the study, compiled the data, designed the analytical 712 framework, analyzed the data, prepared the figures and tables, wrote the paper, reviewed and 713 edited drafts of the paper, and approved the final draft. 714  William Heyman compiled the data, wrote the paper, reviewed and edited drafts of the paper, 715 d d th fi l d ft PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) 731 References Are the 766 last remaining Nassau grouper Epinephelus striatus fisheries sustainable? Status quo in the 767 Bahamas. Endangered Species Research 20(1):27–39. g 768 Choat, H. 2012. Spawning Aggregations in Reef Fishes; Ecological and Evolutionary Processes. 769 Pages 85–116 in Y. Sadovy and P. L. Colin, editors. Reef Fish Spawning Aggregations: 770 Biology, Research and Management, 1st edition. Springer. 768 Choat, H. 2012. Spawning Aggregations in Reef Fishes; Ecological and Evolutionary Processes. 769 Pages 85–116 in Y. Sadovy and P. L. Colin, editors. Reef Fish Spawning Aggregations: 771 Cinner, J. E., T. R. Mcclanahan, T. M. Daw, N. A. J. Graham, J. Maina, S. K. Wilson, and T. P. 772 Hughes. 2009. Report linking social and ecological systems to sustain coral reef fisheries. 773 Current Biology 19:206–212. 774 Claro, R., Y. Sadovy de Mitcheson, K. C. Lindeman, and A. R. García-Cagide. 2009. Historical 775 analysis of Cuban commercial fishing effort and the effects of management interventions on 776 important reef fishes from 1960-2005. Fisheries Research 99(1):7–16. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 777 Claydon, J., M. I. Mccormick, and G. P. Jones. 2014. Multispecies spawning sites for fishes on a 778 low-latitude coral reef: Spatial and temporal patterns. Journal of Fish Biology 84(4):1136– 779 1163. 777 Claydon, J., M. I. Mccormick, and G. P. Jones. 2014. Multispecies spawning sites for fishes on a 778 low-latitude coral reef: Spatial and temporal patterns. Journal of Fish Biology 84(4):1136– 779 1163. 780 Coleman, F.C., Koenig, C.C. and Collins, L.A. 1996. Reproductive styles of shallow-water 781 groupers (Pisces: Serranidae) in the eastern Gulf of Mexico and the consequences of fishing 782 spawning aggregations. Environmental Biology of Fishes 47(2):129-141. p g gg g gy ( ) 783 Coleman, F. C., K. M. Scanlon, and C. C. Koenig. 2011. Groupers on the edge: shelf edge 784 spawning habitat in and around marine reserves of the Northeastern Gulf of Mexico. 785 Professional Geographer 63(4):456–474. g p ( ) 786 Cooper, W. T., L. R. Barbieri, M. D. Murphy, and S. Lowerre-Barbieri. 2013. Assessing stock 787 reproductive potential in species with indeterminate fecundity: Effects of age truncation and 788 size-dependent reproductive timing. Fisheries Research 138:31–41. 789 Coulson, P. G. 2019. The life-history of Cheilodactylus rubrolabiatus from south-western 790 Australia and comparison of biological characteristics of the Cheilodactylidae and Latridae: 791 support for an amalgamation of families. Journal of Fish Biology 94(3):374–390. 792 Costello, C., D. Ovando, R. Hilborn, S. D. Gaines, O. Deschenes, and S. E. Lester. 2012. Status 793 and solutions for the world’s unassessed fisheries. Science (September):1–4. 794 Davies, T. D., and J. K. Baum. 2012. Extinction risk and overfishing: Reconciling conservation 795 and fisheries perspectives on the status of marine fishes. Scientific Reports 2(1):1–9. 796 Ding, C., and X. He. 2004. K-means clustering via principal component analysis. Proceedings of 797 the twenty-first international conference on Machine learning (21):29. 798 Domeier, M. L. 2012. Revisiting Spawning Aggregations: Definitions and Challenges. Page 20 799 in Y. Sadovy De Mitcheson and P. L. Colin, editors. Reef fish spawning aggregations: 800 Biology, Research and Management. Springer Netherlands. gy g p g 801 Donahue, M. J., M. Karnauskas, C. Toews, and C. B. Paris. 2015. Location isn’t everything: 802 Timing of spawning aggregations optimizes larval replenishment. PLoS ONE 10(6):1–14. 802 Timing of spawning aggregations optimizes larval replenishment. PLoS ONE 10(6):1–14. 803 Dulvy, N. K., J. R. Ellis, N. B. Goodwin, A. Grant, J. D. Reynolds, and S. Jennings. 2004. Manuscript to be reviewed 823 Erisman, B., W. Heyman, S. Kobara, C. Biggs, A. Grüss, M. Karnauskas, S 823 Erisman, B., W. Heyman, S. Kobara, C. Biggs, A. Grüss, M. Karnauskas, S. Lowerre-Barbieri, 824 D. Bolser, J. Brenner, N. Farmer. 2018. Cooperative Monitoring Program for Spawning 825 Aggregations in the Gulf of Mexico: An Assessment of Existing Information, Data Gaps, 826 and Research Priorities. NOAA RESTORE Science Program, Final Report, November 827 2018. 32 pp. Available at: https://geo.gcoos.org/restore/finalreport 828 Erisman, B. E., A. Grüss, I. Mascareñas-Osorio, H. Lícon-González, A. F. Johnson, and C. 829 López-Sagástegui. 2020. Balancing conservation and utilization in spawning aggregation 830 fisheries: a trade-off analysis of an overexploited marine fish. ICES Journal of Marine 831 Science 77(1): 148-161. 832 Farmer, N.A., Heyman, W.D., Karnauskas, M., Kobara, S., Smart, T., Ballenger, J., Reichert, M., 833 Wyanski, D., Tishler, M.S., Lindeman, K.C., Lowerre-Barbieri, S., Switzer, T., Solomon, J., 834 McCain, K., Marhefka, M., Sedberry, G.R. 2017. Timing and location of reef fish spawning 835 off the southeastern United States. PLoS ONE 12(3):e0172968. 836 Farmer, N.A., Karnauskas, M. 2013. Spatial distribution and conservation of speckled hind and 837 warsaw grouper in the Atlantic Ocean off the southeastern U.S. PLoS One, 8(11):e78682. 838 Farmer, N.A., J.T. Froeschke, and D.L. Records. 2020. Forecasting for recreational fisheries 839 management: a derby fishery case study with Gulf of Mexico Red Snapper, ICES Journal of 840 Marine Science 77(6): 2265–2284. 841 Farmer, N. A., R. P. Malinowski, M. F. McGovern, and P. J. Rubec. 2016. Stock complexes for 842 fisheries management in the Gulf of Mexico. Marine and Coastal Fisheries 8(1):177–201. 843 Frisk, M., T. J. Miller, and N. K. Dulvy. 2005. Life histories and vulnerability to exploitation of 844 elasmobranchs: inferences from elasticity, perturbation and phylogenetic analyses. Journal 845 of Northwest Atlantic Fishery Science 35:27–45. 846 Ganias, K., S. Somarakis, and C. Nunes. 2014. Reproductive potential. Pages 79–121 in K. 847 Ganias, editor. Biology and Ecolgoy of Sardines and Anchovies, 1st edition. CRC Press, 848 Boca Raton. 849 Gill, D. A., H. A. Oxenford, R. A. Turner, and P. W. Schuhmann. 2017. Making the most of 850 data-poor fisheries: Low cost mapping of small island fisheries to inform policy. Marine 851 Policy (April):0–1. 852 Gilmore, G.R. and Jones, R.S., 1992. Color variation and associated behavior in the epinepheline 853 groupers, Mycteroperca microlepis (Goode and Bean) and M. phenax Jordan and 854 Swain. Bulletin of Marine Science 51(1):83-103. Manuscript to be reviewed 804 Methods of assessing extinction risk in marine fishes. Fish and Fisheries 5(3):255–276. 805 Elli N d Y G W 2007 Eff t f fi h d it di t ib ti d ff t di t ib ti 803 Dulvy, N. K., J. R. Ellis, N. B. Goodwin, A. Grant, J. D. Reynolds, and S. Jennings. 2004. 804 Methods of assessing extinction risk in marine fishes. Fish and Fisheries 5(3):255–276. 805 Ellis, N., and Y.G. Wang. 2007. Effects of fish density distribution and effort distribution on 806 catchability, ICES Journal of Marine Science 64 (1): 178–191. 807 Erisman, B., L. G. Allen, J. T. Claisse, D. J. Pondella II, E. F. Miller, and J. H. Murray. 2011. 808 The illusion of plenty: hyperstability masks collapses in two recreational fisheries that target 809 fish spawning aggregations. Canadian Journal of Fisheries and Aquatic Sciences 68:1705– 810 1716. 811 Erisman, B., O. Aburto-Oropeza, C. Gonzalez-Abraham, I. Mascareñas-Osorio, M. Moreno- 812 Báez, and P.A. Hastings. 2012. Spatiotemporal dynamics of a fish spawning aggregation 813 and its fishery in the Gulf of California. Scientific Reports 2: 284. 814 Erisman, B. E., A. M. Apel, A. D. MacCall, M. J. Román, and R. Fujita. 2014. The influence of 815 gear selectivity and spawning behavior on a data-poor assessment of a spawning 816 aggregation fishery. Fisheries Research 159:75–87. Elsevier B.V. 817 Erisman, B., W. Heyman, S. Kobara, T. Ezer, S. Pittman, O. Aburto-Oropeza, and R. S. Nemeth. 818 2017a. Fish spawning aggregations: where well-placed management actions can yield big 819 benefits for fisheries and conservation. Fish and Fisheries 18(1):128–144. 820 Erisman, B. E., J. J. Cota-Nieto, M. Moreno-Báez, and O. Aburto-Oropeza. 2017b. Vulnerability 821 of spawning aggregations of a coastal marine fish to a small-scale fishery. Marine Biology 820 Erisman, B. E., J. J. Cota-Nieto, M. Moreno-Báez, and O. Aburto-Oropeza. 2017b. Vulnerability 821 of spawning aggregations of a coastal marine fish to a small-scale fishery. Marine Biology 822 164(5):1–18. Springer Verlag. p g gg g y gy 822 164(5):1–18. Springer Verlag. 822 164(5):1–18. Springer Verlag. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 869 period. Journal of Fish Biology 67(1):83–101. 870 Heyman, W.D. and Kjerfve, B., 2008. Characterization of transient multi-species reef fish 871 spawning aggregations at Gladden Spit, Belize. Bulletin of Marine Science, 83(3), pp.531- 872 551. 873 Heyman, W.D., Grüss, A., Biggs, C., Kobara, S., Farmer, N., Karnauskas, M., Lowerre- 874 Barbieri, S., and B. Erisman. 2019. Cooperative monitoring, assessment, and management 875 of fish spawning aggregations and associated fisheries in the U.S. Gulf of Mexico. Marine 876 Policy 109: 103689. 877 Hilborn, R., and C. J. Walters. 1992. Quantitative fisheries stock assessment : choice, dynamics 878 and uncertainty. Chapman and Hall, New York. 879 Hobday, A. J., A. D. M. Smith, I. C. Stobutzki, C. Bulman, R. Daley, J. M. Dambacher, R. A. 880 Deng, J. Dowdney, M. Fuller, D. Furlani, S. P. Griffiths, D. Johnson, R. Kenyon, I. A. 881 Knuckey, S. D. Ling, R. Pitcher, K. J. Sainsbury, M. Sporcic, T. Smith, C. Turnbull, T. I. 882 Walker, S. E. Wayte, H. Webb, A. Williams, B. S. Wise, and S. Zhou. 2011. Ecological risk 883 assessment for the effects of fishing. Fisheries Research 108:372–384. 881 Knuckey, S. D. Ling, R. Pitcher, K. J. Sainsbury, M. Sporcic, T. Smith, C. Turnbull, T. I. 882 Walker, S. E. Wayte, H. Webb, A. Williams, B. S. Wise, and S. Zhou. 2011. Ecological ris 883 t f th ff t f fi hi Fi h i R h 108 372 384 y g y p 882 Walker, S. E. Wayte, H. Webb, A. Williams, B. S. Wise, and S. Zhou. 2011. Ecological risk 883 assessment for the effects of fishing. Fisheries Research 108:372–384. 884 Holt, S.A., 2008. Distribution of red drum spawning sites identified by a towed hydrophone 885 array. Transactions of the American Fisheries Society, 137(2): 551-561. 886 Hordyk, A. R., and T. R. Carruthers. 2018. A quantitative evaluation of a qualitative risk 887 assessment framework: Examining the assumptions and predictions of the Productivity 888 Susceptibility Analysis (PSA). PLoS ONE 13(6):e0198298. y y 889 Jennings, S., J. D. Reynolds, and S. C. Mills. 1998. Life history correlates of responses to 890 fisheries exploitation. Proceedings of the Royal Society B-Biological Sciences 891 265(1393):333–339. 892 Kindsvater, H. K., M. Mangel, J. D. Reynolds, and N. K. Dulvy. 2016. Ten principles from 893 evolutionary ecology essential for effective marine conservation. Ecology and Evolution 894 6(7):2125–2138. 895 King, J. R., and G. McFarlane. 2003. Manuscript to be reviewed 855 Glenn, H.D., Cowan, J.H. Jr, and J.E. Powers. 2017. A comparison of red snapper reproductive 856 potential in the northwestern Gulf of Mexico: natural versus artificial habitats. Marine and 857 Coastal Fisheries 9:139–148 858 Grüss, A., C. Biggs, W. D. Heyman, and B. Erisman. 2018. Prioritizing monitoring and 859 conservation efforts for fish spawning aggregations in the U.S. Gulf of Mexico. Scientific 860 Reports 8:1–10. 861 Grüss, A., C. R. Biggs, W. D. Heyman, and B. Erisman. 2019. Protecting juveniles, spawners or 862 both: A practical statistical modelling approach for the design of marine protected areas. 863 Journal of Applied Ecology 56(10): 2328-2339. 864 Grüss, A., and J. Robinson. 2014. Fish populations forming transient spawning aggregations: 865 should spawners always be the targets of spatial protection efforts? ICES Journal of Marine 866 Science 72(2):480–497. 867 Heyman, W. D., B. Kjerfve, R. T. Graham, K. L. Rhodes, and L. Garbutt. 2005. Spawning 868 aggregations of Lutjanus cyanopterus (Cuvier) on the Belize Barrier Reef over a 6 year PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 915 Leslie, H. M., X. Basurto, M. Nenadovic, L. Sievanen, K. C. Cavanaugh, J. J. Cota-Nieto, B. E. 916 Erisman, E. Finkbeiner, G. Hinojosa-Arango, M. Moreno-Báez, S. Nagavarapu, S. M. W. 917 Reddy, A. Sánchez-Rodríguez, K. Siegel, J. J. Ulibarria-Valenzuela, A. Hudson Weaver, 918 and O. Aburto-Oropeza. 2015. Operationalizing the social-ecological systems framework to 919 assess sustainability. PNAS 112(19):5979–5984. y ( ) 920 Lindeman, K.C., Pugliese, R., Waugh, G.T. and Ault, J.S. 2000. Developmental patterns within a 921 multispecies reef fishery: management applications for essential fish habitats and protected 922 areas. Bulletin of Marine Science 66(3): 929-956. 923 López-Rocha, J. A., M. O. Albañez-Lucero, F. Arreguín-Sánchez, and J. A. de Anda-Montañez. 924 2009. Analysis of the spatial and seasonal variation in catchability of red grouper, 925 Epinephelus morio (Valenciennes, 1828), in the Campeche Bank before overfishing (1973- 926 1977). Revista de Biología Marina y Oceanografia 44(3):751–761. 927 López-Rocha, J. A., and F. Arreguín-Sánchez. 2013. Spatial dynamics of the red grouper 928 Epinephelus morio (Pisces: Serranidae) on the Campeche Bank, Gulf of Mexico. Scientia 929 Marina 77(2):313–322. 930 Lowerre-Barbieri, S., S. L. W. Burnsed, and J. W. Bickford. 2016. Assessing reproductive 931 behavior important to fisheries management: a case study with Red Drum, Sciaenops 932 ocellatus. Ecological Applications 26(4):979–995. 933 Lowerre-Barbieri, S., G. DeCelles, P. Pepin, I. A. Catalán, B. Muhling, B. Erisman, S. X. Cadrin, 934 J. Alós, A. Ospina-Alvarez, M. M. Stachura, M. D. Tringali, S. W. Burnsed, and C. B. 935 Paris. 2017. Reproductive resilience: a paradigm shift in understanding spawner-recruit 936 systems in exploited marine fish. Fish and Fisheries 18(2):285–312. 937 Lowerre-Barbieri, S., K. Ganias, F. Saborido-Rey, F., H. Murua, and J. Hunter. 2011. 938 Reproductive timing in marine fishes: variability, temporal scales, and methods. Marine and 939 Coastal Fisheries: Dynamics. Management, and Ecosystem Science, 3:71-91. 940 Lowerre-Barbieri, S., N. Henderson, J. Llopiz, S. Walters, J. Bickford, and R. Muller. 2009. 941 Defining a spawning population of spotted seatrout (Cynoscion nebulosus) over temporal, 942 spatial, and demographic scales. Marine Ecology Progress Series 394:231–245. 943 Lowerre-Barbieri, S., L. Crabtree, T. Switzer, S. Walters Burnsed, & C. Guenther. 2015. 944 Assessing reproductive resilience: an example with South Atlantic red snapper Lutjanus 945 campechanus. Marine Ecology Progress Series, 526, 125-141. doi:10.3354/meps11212 946 Lowerre-Barbieri, S., M. D. Tringali, C. P. Shea, S. Walters Burnsed, J. Bickford, M. Murphy, 947 and C. Porch. 2019. Manuscript to be reviewed Marine fish life history strategies: applications to fishery 896 management. Fisheries Management & Ecology 10:249–264. 897 Kobara S, Erisman B, Heyman W, Biggs C, Farmer N, Lowerre-Barbieri S, Karnauskas M, 898 Brenner J. 2017. Cooperative monitoring program for spawning aggregations in the Gulf of 899 Mexico: Data Portal Version 1.0. Available at: http://geo.gcoos.org:8010/restore/ p g g g 900 Koenig, C.C., Coleman, F.C., Collins, L.A., Sadovy, Y., and P.L. Colin. 1996. Reproduction in 901 gag (Mycteroperca microlepis) in the eastern Gulf of Mexico and the consequences of 902 fishing spawning aggregations. In: Arregu, F., Munro, J.L., Balgos, M.C., and D. Pauly 903 (eds) Biology, fisheries, and culture of tropical groupers and snappers: proceedings of an 904 EPOMEX/ ICLARM International workshop on tropical snappers and groupers. Campeche, 905 Mexico, 26–29 October 1993. ICLARM Conf Proc 48. WorldFish, 1996. 903 (eds) Biology, fisheries, and culture of tropical groupers and snappers: proceedings of an 904 EPOMEX/ ICLARM International workshop on tropical snappers and groupers. Campeche, 905 M i 26 29 O t b 1993 ICLARM C f P 48 W ldFi h 1996 05 Mexico, 26–29 October 1993. ICLARM Conf Proc 48. WorldFish, 1996 906 Koenig, C.C., Coleman, F.C. and Kingon, K., 2011. Pattern of recovery of the goliath grouper 907 Epinephelus itajara population in the southeastern US. Bulletin of Marine 908 Science, 87(4):891-911. 909 Koenig, C.C., Bueno, L.S., Coleman, F.C., Cusick, J.A., Ellis, R.D., Kingon, K., Locascio, J.V., 910 Malinowski, C., Murie, D.J. and C.D. Stallings. 2017. Diel, lunar, and seasonal spawning 911 patterns of the Atlantic goliath grouper, Epinephelus itajara, off Florida, United 912 States. Bulletin of Marine Science 93(2):391-406. 913 Koenig, C.C., Coleman, F.C. and C.R. Malinowski. 2020. Atlantic Goliath Grouper of Florida: 914 To Fish or Not to Fish. Fisheries 45(1): 20-32. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 961 Morato, T., W. W. L. Cheung, and T. J. Pitcher. 2006. Vulnerability of seamount fish to fishing: 962 fuzzy analysis of life-history attributes. Journal of Fish Biology 68(1):209–221. 963 Mullon, C., P. Fréon, and P. Cury. 2005. The dynamics of collapse in world fisheries. Fish and 964 Fisheries 6:111–120. 965 Musick, J. A. 1999. Criteria to Define Extinction Risk in Marine Fishes. The American Fisheries 966 Society Initiative, Fisheries 24(12):6–14. 967 National Marine Fisheries Service (NMFS) 2018. Fisheries of the United States, 2017. U.S. 968 Dept. of Commerce, NOAA Current Fishery Statistics No. 2017. U.S. Dept. of Commerce, 969 NOAA Current Fishery Statistics No. 2017. 969 NOAA Current Fishery Statistics No. 2017. 969 NOAA Current Fishery Statistics No. 2017. 970 Nelson, M. D., C. C. Koenig, F. C. Coleman, and D. A. Mann. 2011. Sound production of red 971 grouper Epinephelus morio on the west Florida shelf. Aquatic Biology 12(2):97–108. 972 Nemeth, R. S. 2012. Ecosystem aspects of species that aggregat to spawn. Pages 21–55 in Y. 973 Sadovy and P. Colin, editors. Reef Fish Spawning Aggregations: Biology, Research and 974 Management. Springer. 975 Patrick, W. S., P. Spencer, O. Ormseth, J. Cope, J. Field, D. Kobayashi, T. Gedamke, E. Cortés, 976 K. Bigelow, W. Overholtz, J. Link, and P. Lawson. 2009. Use of productivity and 977 susceptibility indices to determine the vulnerability of a stock: with example applications to 978 six U.S. fisheries. NOAA Techical Memo p90. p , P. Spencer, J. Link, J. Cope, J. Field, D. Kobayashi, P. Lawson, T. Gedamke, E. h i l h l i d i i d ibil 979 Patrick W.S, P. Spencer, J. Link, J. Cope, J. Field, D. Kobayashi, P. Lawson, T. Gedamke, E. 980 Cortés, O. Ormseth, K. Bigelow, W. Overholtz. 2010. Using productivity and susceptibility 981 indices to assess the vulnerability of united states fish stocks to overfishing. Fishery Bulletin 982 108:305–322. 983 Patterson, W.F. III, J.H. Cowan Jr., G.R. Fitzhugh, and D.L. Nieland. 2007. Red snapper ecology 984 and fisheries in the U.S. Gulf of Mexico. Am Fish Soc Symp 60 985 Pearson, J.C. 1928. Natural History and Conservation of Redfish and Other Commercial 986 Sciaenids on the Texas Coast, Government Printing Office, Washington, DC, 1928, pp. 987 129–214. 988 Porch, C.E. 2000. Status of the red drum stocks of the Gulf of Mexico. Southeast Fisheries 989 Science Center. Sustainable Fisheries Division Contributions: SFD-99/00-85. Manuscript to be reviewed Assessing red drum spawning aggregations and abundance in the 948 Eastern Gulf of Mexico: a multidisciplinary approach. ICES Journal of Marine Science 949 (75):516–529. 950 Lowerre-Barbieri, S., H. Menendez, J. Bickford, T. S. Switzer, L. Barbieri, and C. Koenig. 2020. 951 Testing assumptions about sex change and spatial management in the protogynous gag 952 grouper, Mycteroperca microlepis. Marine Ecology Progress Series 639:199-214. 953 Mamauag, S. S., P. M. Alino, R. J. S. Martinez, R. N. Muallil, M. V. A. Doctor, E. C. Dizon, R. 954 C. Geronimo, F. M. Panga, and R. B. Cabral. 2013. A framework for vulnerability 955 assessment of coastal fisheries ecosystems to climate change-Tool for understanding 956 resilience of fisheries (VA–TURF). Fisheries Research 147:381–393. 957 Maunder, M. N., and R. B. Deriso. 2013. A stock-recruitment model for highly fecund species 958 based on temporal and spatial extent of spawning. Fisheries Research 146:96–101. 959 Methot, R. D., and C. R. Wetzel. 2013. Stock synthesis: A biological and statistical framework 960 for fish stock assessment and fishery management. Fisheries Research 142:86–99. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 990 Porch, C.E, Fitzhugh, G.R., Lang, E.T., Lyon, H.M., and B.C. 2015. Estimating the dependence 991 of spawning frequency on size and age in Gulf of Mexico Red Snapper. Marine and 992 Coastal Fisheries 7:233–245 993 Reynolds, J. D., S. Jennings, and N. K. Dulvy. 2001. Life histories of fishes and population 994 responses to exploitation. Pages 147–169 in J. D. Reynolds, G. M. Mace, K. H. Redford, 995 and J. G. Robinson, editors. Conservation of Exploited Species. Cambridge University 996 Press, Cambridge University.Robinson, J. 2015. Understanding the Causes of Vulnerability 997 to Fishing in Reef Fishes that Aggregate. Australian Research Council Centre of Excellence 998 for Coral Reef Studies, James Cook University. 999 Robinson, J., and M. Samoilys. 2013. Evaluation of an indicator-based framework for assessing 1000 the vulnerability of reef fish populations to spawning aggregation fisheries. Pages 102-112 1001 in J. Robinson J and M. Samoilys, editors. Reef Fish Spawning Aggregations in the Western 1002 Indian Ocean: Research for Management. WIOMSA/SIDA/SFA/CORDIO. WIOMSA 1003 Book Series 13. 1004 Robinson, J. 2015. Understanding the Causes of Vulnerability to Fishing in Reef Fishes that 1005 Aggregate. Ph.D. Dissertation. Australian Research Council Centre of Excellence for Coral 1006 Reef Studies, James Cook University. 1004 Robinson, J. 2015. Understanding the Causes of Vulnerability to Fishing in Reef Fishes that 1005 Aggregate. Ph.D. Dissertation. Australian Research Council Centre of Excellence for Coral 1006 Reef Studies, James Cook University. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 1007 Robinson, J., N. A. J. Graham, J. E. Cinner, G. R. Almany, and P. Waldie. 2015. Fish and fisher 1008 behaviour influence the vulnerability of groupers (Epinephelidae) to fishing at a 1009 multispecies spawning aggregation site. Coral Reefs 34:371–382. 1010 Robinson, O. J., O. P. Jensen, M. M. Provost, S. Huang, N. H. Fefferman, A. Kebir, and J. L. 1011 Lockwood. 2017. Evaluating the impacts of fishing on sex-changing fish: a game-theoretic 1012 approach. ICES Journal of Marine Science 74(3):652-659. 1013 Rose, G. A., and D. W. Kulka. 1999. Hyperaggregation of fish and fisheries: how catch-per-unit- 1014 effort increased as the northern cod (Gadus morhua) declined. Canadian Journal of Fisheries 1015 and Aquatic Sciences 56(S1): 118-127. 1016 Rosenberg, A. A., and V. R. Restrepo. 1994. Uncertainty and risk evaluation in stock assessment 1017 advice for U.S. marine fisheries. Canadian Journal of Fisheries and Aquatic Sciences 1018 51(12):2715–2720. 1019 Saaty, R. W. 1987. The analytic heirarchy process - what it is and how it is used. Mathematical 1020 Modelling 9:3–5. 1021 Sadovy De Mitcheson, Y., and B. E. Erisman. 2012. Fishery and biological implications of 1022 fishing spawning aggregations, and the social and economic importance of aggregating 1023 fishes. Pages 405–565 in Y. Sadovy De Mitcheson and P. L. Colin, editors. Reef Fish 1024 Spawning Aggregations: Biology, Research and Management. Springer. 1025 Sagarese, S. R., J. F. Walter, J. J. Isely, M. D. Bryan, and N. Cummings. 2015. A comparison of 1026 data-rich versus data-limited methods in estimating overfishing limits. Page SEDAR 46- 1027 DW-01. SEDAR, North Charleston, SC. 1028 Salas, S., R. Chuenpagdee, J. C. Seijo, and A. Charles. 2007. Challenges in the assessment and 1029 management of small-scale fisheries in Latin America and the Caribbean. Fisheries 1030 Research 87(1):5–16. 1031 Salthaug, A., and S. Aanes. 2003. Catchability and the spatial distribution of fishing vessels. 1032 Canadian Journal of Fisheries and Aquatic Sciences 60:259–268. 1033 Saucier, M. H., and D. M. Baltz. 1993. Spawning site selection by spotted seatrout, Cynoscion 1034 nebulosus, and black drum, Pogonias cromis, in Louisiana. Environmental Biology of 1035 Fishes 36:257–272. 1036 Schram, M. J., and M. A. Steele. 2020. Experimental evidence of resilience to size-selective 1037 harvesting in a protogynous hermaphroditic reef fish, Rhinogobiops nicholsii. Journal of 1038 Experimental Marine Biology and Ecology 525:151320. p gy gy 1039 SEDAR. 2014a. SEDAR 33 – Gulf of Mexico Greater Amberjack Stock Assessment Report. Manuscript to be reviewed 1040 SEDAR N h Ch l SC 499 A il bl li h // d b / d 33 p gy gy 1039 SEDAR. 2014a. SEDAR 33 – Gulf of Mexico Greater Amberjack Stock Assessment Report. 1040 SEDAR, North Charleston SC. 499 pp. Available online at: http://sedarweb.org/sedar-33 1041 S A 2014b S A 33 G lf f i G S k A S A h 1040 SEDAR, North Charleston SC. 499 pp. Available online at: http://sedarweb.org/sedar-33 1041 SEDAR. 2014b. SEDAR 33 – Gulf of Mexico Gag Stock Assessment Report. SEDAR, North 1042 Charleston SC. 609 pp. Available online at: 41 SEDAR. 2014b. SEDAR 33 – Gulf of Mexico Gag Stock Assessment Report. SE 42 Charleston SC. 609 pp. Available online at: 43 http://www.sefsc.noaa.gov/sedar/Sedar_Workshops.jsp?WorkshopNum=33 1044 SEDAR. 2016a. SEDAR 49 Gulf of Mexico Data-limited Species: red drum, lane snapper, 1045 wenchman, yellowmouth grouper, speckled hind, snowy grouper, almaco jack, lesser 1046 amberjack. North Charleston, SC. j 1047 SEDAR. 2016b. SEDAR 33 Update Report Gulf of Mexico Gag Grouper. SEDAR. North 1048 Charleston, South Carolina. 123 pp. Available at: 49 http://sedarweb.org/docs/suar/GagUpdateAssessReport_Final_0.pdf 1050 SEDAR. 2018. SEDAR 52 – stock assessment report Gulf of Mexico red snapper. SEDAR, 1051 North Charleston, SC. 434 p. 1052 Skjold, F., A. Eide, and O. Flaaten. 1996. Production functions of the Norwegian bottom trawl PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 1053 fisheries of cod in the Barents Sea. ICES C. M. 1996/P(3):1–14. 1054 Solmundsson, J., H. Karlsson, and J. Palsson. 2003. Sexual differences in spawning behaviour 1055 and catchability of plaice (Pleuronectes platessa) west of Iceland. Fisheries Research 61(1– 1056 3):57–71. 1057 South Atlantic Fisheries Management Council. 2017. Amendment 36 to the fishery management 1058 plan for the Snapper Grouper fishery fo the South Atlantic Region; Actions to implement 1059 special management zones in the South Atlantic. Available at: Stobutzki, I., M. Miller, and 1060 D. Brewer. 2001. Sustainability of fishery bycatch: A process for assessing highly diverse 1061 and numerous bycatch. Environmental Conservation 28(2):167–181. 1062 Szedlmayer, S.T., and S.A. Bortone. 2019. Red Snapper Biology in a Changing World. New 1063 York, CRC Press. 306 pp. 1064 Tinhan, T., B. Erisman, O. Aburto-Oropeza, A. Weaver, D. Vázquez-Arce, and C. G. Lowe. 1065 2014. Residency and seasonal movements in Lutjanus argentiventris and Mycteroperca 1066 rosacea at Los Islotes Reserve, Gulf of California. Marine Ecology Progress Series 1067 501:191–206. 1068 Wall, C. C., B. T. Donahue, D. F. Naar, and D. A. Mann. 2011. Spatial and temporal variability 1069 of red grouper holes within Steamboat Lumps Marine Reserve, Gulf of Mexico. Marine 1070 Ecology Progress Series 431(1):243–254. 1071 Walters, S., S. Lowerre-Barbieri, J. Bickford, and D. Mann. 2009. Using a passive acoustic 1072 survey to identify spotted seatrout spawning sites and associated habitat in Tampa Bay, 1073 Florida. Transactions of the American Fisheries Society 138:88-98. 1074 Wilberg, M. J., J. T. Thorson, B. C. Linton, and J. Berkson. 2009. Incorporating time-varying 1075 catchability into population dynamic stock assessment models. Reviews in Fisheries 1076 Science 18(1):7–24. 1077 Winemiller, K. O. 2005. Life history strategies, population regulation, and implications for 1078 fisheries management. Canadian Journal of Fisheries and Aquatic Sciences 62(4):872–885. Manuscript to be reviewed 1083 Table 1 Definitions of life history and spawning behavior parameters included in the analysis. 1084 1085 Table 2 Definitions and categories of the extrinsic vulnerability indicators used in the analysis. 1086 1087 Table 3 Spawning behavior and life history characteristics for the 28 species analyzed. Values 1088 are specific to the GOM and reflect the average (± SE). Definitions of each characteristic are in 1089 Table 1. An annotated table with citations for each entry is available online 1090 (http://geo.gcoos.org/restore/) 1091 1092 Table 4 Spawning season of 28 species in the Gulf of Mexico, sorted by family. Grey indicates 1093 the extent of the spawning season; black indicates the peak spawning months. 1094 1095 1096 1097 1098 Figure 1 Flow chart illustrating the process of calculating intrinsic and extrinsic vulnerability 1099 scores. Definitions for each of the indicators, the scales used, and the weighted vulnerabilities are 1100 in Table 1 and Table 2. 1101 1102 Figure 2 Visualization of correlation matrix of spawning behavior and life history parameters. 1103 Blue indicates positive correlation and red reflects negative correlation. The size of the circle 1104 reflects the value of Spearman’s rank correlation coefficient and boxes without an “x” are 1105 significant (α=0.05). 1106 1107 Figure 3 Biplot of PCA showing the first two principal components. The arrows show the 1108 relative loadings of each PC axis and the color of the species indicates the stock status. 1109 Historically overfished species (Atlantic Goliath Grouper, Nassau Grouper, Red Drum, Gag 1110 Grouper, and Red Grouper) have been labelled overfished in addition to those species that are 1111 currently designated as overfished. 1112 1113 Figure 4 Box plot of Principal Component (PC) scores by stock status. The horizontal line 1114 within the box represents the mean value and the box outlines the interquartile range (IQR). 1115 Whiskers indicate the highest and lowest value within 1.5 IQR, values outside that range are 1116 outliers and are plotted as points. Overfished and not overfished on PC2 are significantly 1117 different (t=1.88, df=13.4, p=0.08). Statistical comparisons were only made between overfished 1118 and not overfished scores. 1119 1120 Figure 5 Distribution of the overall vulnerability index of the 28 species. The position of each 1121 species within each quadrant indicates their relative vulnerability from low to very high. 1120 Figure 5 Distribution of the overall vulnerability index of the 28 species. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Manuscript to be reviewed Manuscript to be reviewed Manuscript to be reviewed The position of each 1121 species within each quadrant indicates their relative vulnerability from low to very high. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Figure 1 Flow chart illustrating the process of calculating intrinsic and extrinsic vulnerability scores. Deﬁnitions for each of the indicators, the scales used, and the weighted vulnerabilities are in Table 1 and Table 2. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Manuscript to be reviewed PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Figure 2 Visualization of correlation matrix of spawning behavior and life history parameters. Visualization of correlation matrix of spawning behavior and life history parameters. Blue indicates positive correlation and red reﬂects negative correlation. The size of the circle reﬂects the value of Spearman’s rank correlation coeﬃcient and boxes without an “x” are signiﬁcant (α=0.05). PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Figure 3 Biplot of PCA showing the ﬁrst two principal components. Manuscript to be reviewed PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Figure 4 Box plot of Principal Component (PC) scores by stock status. Biplot of PCA showing the ﬁrst two principal components. The arrows show the relative loadings of each PC axis and the color of the species indicates the stock status. Historically overﬁshed species (Atlantic Goliath Grouper, Nassau Grouper, Red Drum, Gag Grouper, and Red Grouper) have been labelled overﬁshed in addition to those species that are currently designated as overﬁshed. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Manuscript to be reviewed Manuscript to be reviewed Manuscript to be reviewed Box plot of Principal Component (PC) scores by stock status. The horizontal line within the box represents the mean value and the box outlines the interquartile range (IQR). Whiskers indicate the highest and lowest value within 1.5 * IQR, values outside that range are outliers and are plotted as points. Overﬁshed and not overﬁshed on PC2 are signiﬁcantly diﬀerent (t=1.88, df=13.4, p=0.08). Statistical comparisons were only made between overﬁshed and not overﬁshed scores. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Manuscript to be reviewed PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Figure 5 Distribution of the overall vulnerability index of the 28 species. Distribution of the overall vulnerability index of the 28 species. The position of each species within each quadrant indicates their relative vulnerability from low to very high. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Manuscript to be reviewed Table 1(on next page) Table 1(on next page) Table 1(on next page) Deﬁnitions of life history and spawning behavior parameters included in the analysis. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Manuscript to be reviewed finitions of life history and spawning behavior parameters included in the analysis. Table 1 Definitions of life history and spawning behavior parameters included in th Table 1 Definitions of life history and spawning behavior parameters included in the analysis. Parameter Type Parameter Abbreviation Weighted influence on vulnerability Description Aggregating Behavior (0-4) Agg  The degree of aggregating behavior associated with spawning: does not aggregate=0; simple migratory spawner=1; resident spawning aggregation=2; mixed resident and transient aggregations=3; transient spawning aggregation=4. Spawning Season Duration (1-12) Duration  Number of months that the species spawns, with shorter spawning seasons conferring higher vulnerability to aggregation fishing. Spawning Behavior Relative Abundance (1-6) Rel. Ab.  Change in abundance of fish relative to non- reproductive periods. No change in abundance between spawning and non-spawning periods = 1; abundance doubles from solitary to few to ca. 10 fish (clustering of polygynous groups) = 2; abundance increases from small groups to 100-200 fish = 3; abundance increases from small groups to 500-1000 fish = 4; abundance increases from small groups to 1,000-10,000 fish =5; abundance increases from small groups to >10,000 fish = 6. Larger abundance changes confer higher vulnerability to aggregation fishing. Max Age (years) Amax Maximum age in years (longevity) Max Weight (kg) Wmax Maximum weight in kilograms Max Length (cm) Lmax Maximum reported length for the species in centimeters k (vB Growth Coefficient) k von Bertalanffy growth coefficient Linf (Asymptotic Length, cm) Linf Asymptotic length for von Bertalanffy growth equation, expressed in centimeters Age at Maturity (months) Am Age at 50% maturity in months Length at Maturity (cm) Lm Length at 50% maturity in centimeters Life History M (Natural Mortality) M  Death rate per year not associated with fishing 2 PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Manuscript to be reviewed Table 2 Definitions and categories of the extrinsic vulnerability indicators used in the analysis. Parameter Weighted influence on vulnerability Description Access to Fishery (1-4)  Extent to which access to the fishery is limited. 4= open (basic commercial/recreational license), 3= reef fish permit (comm. regulation) and or charter/headboat reef fish and coastal pelagics permit (rec. regulation), 2= reef fish permit & IFQ program, 1=closed Catch Limits (1-4)  Number of catch limits, 4= 0-1 regulation, 3= 2-3 regulations, 2= 4 regulations, 1= 5 regulations Gear Measures (1-4)  Amount of restrictions on gear used, 4= 9 or more allowable gear types, 3= 6-8 allowable gear types, 2= 3- 5 allowable gear types, 1= 0-2 allowable gear types Seasonal Restrictions (1-4)  Spawning season prohibition of take or trade. 4= none, 3=seasonal closure, not during spawning, 2= seasonal closure during peak spawning, 1= closed during entire spawning season Site Closures (1-4)  Spatial closure of spawning site. 4= no regulations, 3= restricted gear, 2= site closed part of the year, 1= site closed all year itions and categories of the extrinsic vulnerability indicators used in the analysis. 3 3 Manuscript to be reviewed Manuscript to be reviewed Table 2(on next page) Table 2(on next page) Table 2(on next page) Deﬁnitions and categories of the extrinsic vulnerability indicators used in the analysis Deﬁnitions and categories of the extrinsic vulnerability indicators used in the analysis PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Manuscript to be reviewed PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Table 3(on next page) Table 3(on next page) Spawning behavior and life history characteristics for the 28 species analyzed. Values are speciﬁc to the GOM and reﬂect the average (± SE). Deﬁnitions of each characteristic are in Table 1. An annotated table with citations for each entry is available online (http://geo gcoos org/restore/) Spawning behavior and life history characteristics for the 28 species analyzed. V l iﬁt th GOM d ﬂ t th (± SE) D ﬁiti f h Spawning behavior and life history characteristics for the 28 species analyzed. Values are speciﬁc to the GOM and reﬂect the average (± SE). Deﬁnitions of each characteristic are in Table 1. An annotated table with citations for each entry is available online (http://geo.gcoos.org/restore/) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed Table 3 Spawning behavior and life history characteristics for the 28 species analyzed. Values are specific to the GOM and reflect the average (± SE) when multiple values were reported. Definitions of each characteristic are in Table 1. An annotated table with citations for each entry is available online (http://geo.gcoos.org/restore/) 1 Table 3 Spawning behavior and life history characteristics for the 28 species analyzed. Values are specific to the GOM and reflect the 2 average (± SE) when multiple values were reported. Definitions of each characteristic are in Table 1. An annotated table with citations 3 for each entry is available online (http://geo.gcoos.org/restore/) y g g g Common Name Agg. (0-4) Duration (# mos) Rel. Ab. Manuscript to be reviewed Manuscript to be reviewed Table 4(on next page) Table 3(on next page) (1-6) Amax (yr) Wmax (kg) Lmax (cm) k Linf (cm) Am (mos) LM (cm) M Gray Triggerfish 2 4 4 15 ± 1 6 30 0.14 ± 0.06 59 ± 3 18 ± 2 17 ± 3 0.27 Almaco Jack 3 8 3 22 60 160 0.13 163 53 81 N/A Greater Amberjack 3 4 3 15 ± 1 81 190 0.14 ± 0.01 144 ± 7 27 ± 2 79 ± 4 0.25 ± 0.03 Black Grouper 4 5 4 33 ± 0 163 150 0.14 ± 0.01 133 ± 1 78 ± 4 86 ± 2 0.14 ± 0.02 Gag 4 4 3 31 ± 3 37 145 0.13 ± 0.01 128 ± 1 42 ± 2 54 ± 3 0.13 ± 0.01 Atlantic Goliath Grouper 4 5 3 37 ± 4 363 ± 68 250 0.09 ± 0.01 222 ± 6 72 ± 6 120 ± 3 0.12 ± 0.03 Nassau Grouper 4 3 6 29 ± 2 27 ± 0 100 0.13 ± 0.02 76 ± 3 60 ± 0 40 ± 2 0.18 ± 0.03 Red Grouper 0 5 2 29 ± 1 23 125 0.13 ± 0.02 83 ± 2 34 ± 6 29 ± 4 0.14 ± 0.02 Scamp 3 6 3 31 ± 1 13 107 0.09 ± 0.00 77 ± 9 24 ± 3 33 ± 1 0.15 ± 0.09 Warsaw Grouper 3 8 3 41 ± 0 198 ± 4 235 0.05 ± 0.00 239 ± 0 49 81 N/A Yellowedge Grouper 3 10 3 85 ± 10 20 ± 1 115 0.06 ± 0.01 100 ± 3 96 ± 0 55 ± 9 0.07 ± .11 Yellowfin Grouper 4 8 4 15 ± 1 19 ± 0 100 0.12 ± 0.02 89 ± 6 44 54 ±10 0.26 ± 0.05 Yellowmouth Grouper 4 12 3 28 ± 3 9 84 0.08 ± 0.01 83 ± 1 36 ± 11 43 ± 1 0.23 ± 0.02 Speckled Hind 0 3 1 45 ± 3 30 110 0.12 ± 0.01 89 ± 5 79 ± 9 53 ± 8 0.15 ± 0.01 Snowy Grouper 0 10 1 35 ± 2 30 122 0.09 ± 0.01 106 ± 5 60 ± 2 60 ± 4 0.19 ± 0.02 Hogfish 2 8 2 23 ± 1 10 ± 2 91 0.11 ± 0.08 85 ± 6 11 ± 2 15 ± 1 0.18 ± 0.11 Cubera Snapper 4 4 6 22 ± 1 57 ± 8 160 0.16 ± 0.01 120 ± 7 24 ± 9 62 ± 6 0.15 ± 0.08 Mutton Snapper 4 4 5 40 ± 6 16 ± 3 94 0.17 ± 0.01 86 ± 4 48 ± 3 50 ± 5 0.11 ± 0.03 Red Snapper 2 5 2 48 ± 4 23 ± 2 100 0.19 ± 0.05 86 ± 4 24 ± 3 23 ± 2 0.1 ± 0.02 Vermilion Snapper 2 6 2 26 ± 2 3 63 0.33 ± 0.04 34 ± 2 24 ± 4 14 ± 3 0.25 ± 0.03 Tilefish 0 6 2 40 ± 3 26 125 0.13 ± 0.01 83 ± 3 24 ± 2 34 ± 5 0.13 ± 0.01 Southern Flounder 4 4 5 8 ± 1 9 92 0.28 ± 0.08 65 ± 6 24 ± 5 40 0.36 PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Black Drum 3 7 4 58 ± 8 51 150 0.17 ± 0.02 114 ± 21 60 65 ± 2 0.06 ± 0.00 Red Drum 3 4 4 42 ± 2 45 160 0.32 ± 0.04 88 ± 7 48 ± 8 68 ± 6 0.16 ± 0.04 Spotted Seatrout 2 6 3 12 ± 1 8 70 0.32 ± 0.05 69 ± 3 12 ± 0 23 ± 1 0.3 Gray Triggerfish 1 6 2 24 ± 2 42 184 0.19 ± 0.02 115 ± 5 48 60 ± 12 0.17 ± 0.01 Almaco Jack 1 6 2 11 ± 2 6 101 0.61 ± 0.03 56 ± 3 8 ± 0.01 30 ± 2 0.3 Greater Amberjack 4 3 5 20 ± 2 10 92 0.36 ± 0.03 46 ± 1 24 ± 6 30 0.15 Manuscript to be reviewed Black Drum 3 7 4 58 ± 8 51 150 0.17 ± 0.02 114 ± 21 60 65 ± 2 0.06 ± 0.00 Red Drum 3 4 4 42 ± 2 45 160 0.32 ± 0.04 88 ± 7 48 ± 8 68 ± 6 0.16 ± 0.04 Spotted Seatrout 2 6 3 12 ± 1 8 70 0.32 ± 0.05 69 ± 3 12 ± 0 23 ± 1 0.3 Gray Triggerfish 1 6 2 24 ± 2 42 184 0.19 ± 0.02 115 ± 5 48 60 ± 12 0.17 ± 0.01 Almaco Jack 1 6 2 11 ± 2 6 101 0.61 ± 0.03 56 ± 3 8 ± 0.01 30 ± 2 0.3 Greater Amberjack 4 3 5 20 ± 2 10 92 0.36 ± 0.03 46 ± 1 24 ± 6 30 0.15 PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 5 5 PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Table 4(on next page) Table 4(on next page) Spawning season of 28 species in the Gulf of Mexico, sorted by family. Grey indicates the extent of the spawning season; black indicates the peak spawning months. PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021) Manuscript to be reviewed 1 Table 4 Spawning season of 28 species in the Gulf of Mexico, sorted by family. Grey indicates 2 the extent of the spawning season; black indicates the peak spawning months. 3 Manuscript to be reviewed Manuscript to be reviewed 1 Table 4 Spawning season of 28 species in the Gulf of Mexico, sorted by family. Grey indicates 2 the extent of the spawning season; black indicates the peak spawning months. 3 1 Table 4 Spawning season of 28 species in the Gulf of Mexico, sorted by family. Grey indicates 2 the extent of the spawning season; black indicates the peak spawning months. 3 Family Common Name JanFebMarAprMayJunJulAugSepOctNovDec Epinephelidae Black Grouper Epinephelidae Gag Epinephelidae Atlantic Goliath Grouper Epinephelidae Nassau Grouper Epinephelidae Red Grouper Epinephelidae Scamp Epinephelidae Snowy Grouper Epinephelidae Speckled Hind Epinephelidae Warsaw Grouper Epinephelidae Yellowedge Grouper Epinephelidae Yellowfin Grouper Epinephelidae Yellowmouth Grouper Lutjanidae Cubera Snapper Lutjanidae Mutton Snapper Lutjanidae Red Snapper Lutjanidae Vermilion Snapper Sciaenidae Black Drum Sciaenidae Red Drum Sciaenidae Spotted Seatrout Scombridae King Mackerel Scombridae Spanish Mackerel Carangidae Almaco Jack Carangidae Greater Amberjack Sparidae Sheepshead ParalichthyidaeSouthern Flounder Malacanthidae Tilefish Labridae Hogfish Balistidae Gray Triggerfish PeerJ reviewing PDF \| (2020:11:55081:1:0:NEW 8 Jun 2021)
https://openalex.org/W2900188526	https://europepmc.org/articles/pmc6218033?pdf=render	English	null	Development of a new 7BS.7HL winter wheat-winter barley Robertsonian translocation line conferring increased salt tolerance and (1,3;1,4)-β-D-glucan content	PloS one	2,018	cc-by	9,315	RESEARCH ARTICLE Edina Tu¨rko¨siID1, Eva Darko2, Marianna Rakszegi3, Istva´n Molna´r4, Ma´rta Molna´r-La´ng1, Andra´s CsehID5 1 Department of Plant Genetic Resources, Agricultural Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, Martonva´sa´r, Hungary, 2 Department of Plant Physiology, Agricultural Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, Martonva´sa´r, Hungary, 3 Cereal Breeding Department, Agricultural Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, Martonva´sa´r, Hungary, 4 Maize Breeding Department, Agricultural Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, Martonva´sa´r, Hungary, 5 Molecular Breeding Department, Agricultural Institute, Centre for Agricultural Research, Hungarian Academy of Sciences, Martonva´sa´r, Hungary a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * cseh.andras@agrar.mta.hu (AC); langne.molnar.marta@gmail.com (MML) Abstract Citation: Tu¨rko¨si E, Darko E, Rakszegi M, Molna´r I, Molna´r-La´ng M, Cseh A (2018) Development of a new 7BS.7HL winter wheat-winter barley Robertsonian translocation line conferring increased salt tolerance and (1,3;1,4)-β-D-glucan content. PLoS ONE 13(11): e0206248. https://doi. org/10.1371/journal.pone.0206248 Interspecific hybridization between bread wheat (Triticum aestivum, 2n = 42) and related species allows the transfer of agronomic and quality traits, whereby subsequent generations comprise an improved genetic background and can be directly applied in wheat breeding programmes. While wild relatives are frequently used as sources of agronomically favour- able traits, cultivated species can also improve wheat quality and stress resistance. A salt- tolerant ‘Asakaze’/‘Manas’ 7H disomic addition line (2n = 44) with elevated β-glucan content, but with low fertility and an unstable genetic background was developed in an earlier wheat- barley prebreeding programme. The aim of the present study was to take this hybridization programme further and transfer the favourable barley traits into a more stable genetic back- ground. Taking advantage of the breakage-fusion mechanism of univalent chromosomes, the ‘Rannaya’ winter wheat 7B monosomic line was used as female partner to the 7H addi- tion line male, leading to the development of a compensating wheat/barley Robertsonian translocation line (7BS.7HL centric fusion, 2n = 42) exhibiting higher salt tolerance and ele- vated grain β-glucan content. Throughout the crossing programme, comprising the F1-F4 generations, genomic in situ hybridization, fluorescence in situ hybridization and chromo- some-specific molecular markers were used to trace and identify the wheat and barley chro- matin. Investigations on salt tolerance during germination and on the (1,3;1,4)-β-D-glucan (mixed-linkage glucan [MLG]) content of the seeds confirmed the salt tolerance and ele- vated grain MLG content of the translocation line, which can be directly applied in current wheat breeding programmes. Editor: Aimin Zhang, Institute of Genetics and Developmental Biology Chinese Academy of Sciences, CHINA Copyright: © 2018 Tu¨rko¨si et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Development of a new 7BS.7HL winter wheat- winter barley Robertsonian translocation line conferring increased salt tolerance and (1,3;1,4)-β-D-glucan content Edina Tu¨rko¨siID1, Eva Darko2, Marianna Rakszegi3, Istva´n Molna´r4, Ma´rta Molna´r-La´ng1, Andra´s CsehID5 * cseh.andras@agrar.mta.hu (AC); langne.molnar.marta@gmail.com (MML) Introduction Research, Development and Innovation Office (K112226 and K129221). A.C. gratefully acknowledges a Marie Skłodowska-Curie Fellowship Grant (H2020-MSCA-IF-2016-752453- LANDRACES) from the European Union. I.M. gratefully acknowledges support a Marie Skłodowska-Curie Fellowship Grant ‘AEGILWHEAT’ [H2020-MSCA-IF-2016-746253] from the European Union. The exponential increase in the world population and the recent challenges imposed by the changing climate make it essential for crop breeding programmes to put the emphasis on develop- ing higher-yielding cultivars with better nutritional parameters. Bread wheat (Triticum aestivum, 2n = 6x = 42), one of the most important crop species, is significantly affected by environmental stresses that seriously reduce its productivity [1,2]. Among the enviromental factors, soil saliniza- tion, which has been worsened by climate change, is one of the most constraining abiotic stresses threatening wheat yields. The breeding of salinity-tolerant, high- yielding varieties is therefore a valuable solution to ensure food security. This can be achieved via prebreeding programmes that transfer genes responsible for salt tolerance from related species into wheat. Barley (Hordeum vul- gare, 2n = 2x = 14) is one of the most salt- tolerant crops, with significant variation among geno- types [3–5]. It can be cultivated on saline soil as it maintains growth despite accumulating Na+ in the leaves. Salt stress causes a complete metabolic rearrangement in the cells, affecting the sugar, amino acid and polyamine metabolism, and ion and redox homeostasis [6–8]. Similarly to other plant species, salt tolerance in barley is inherited as a polygenic trait with numerous quantitative trait loci (QTLs) mapped on several chromosomes including 7H [9,10]. Competing interests: The authors have declared that no competing interests exist. Competing interests: The authors have declared that no competing interests exist. Barley has a place in healthy diet, as the grain has a high content of (1,3;1,4)-β-D-glucan (MLG), a polysaccharide that helps to lower blood cholesterol and regulates blood sugar levels in humans [11,12]. The cellulose synthase-like F6 (HvCslF6) gene that encodes the putative MLG synthase has been mapped to barley chromosome 7H [13, 14]. Located within the cen- tromeric region of the long arm (7HL), the HvCslF6 gene has been proved to play a key role in controlling β-glucan biosynthesis, and its down-regulation results in decreased MLG within the endosperm [15,16]. Barley belongs to the tertiary genepool of wheat, so abiotic stress resistance and advanta- geous composition parameters can be transferred into wheat via wide hybridization. 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content Abstract Funding: The research leading to these results has received funding from the European Community’s Seventh Framework Programme (FP7/ 2007-2013) under grant agreement FP7- 613556, from the Whealbi project (http://www.whealbi.eu/project/ project) and from the Hungarian National 1 / 17 PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 Introduction Crossing wheat with barley is challenging to achieve and the F1 hybrids can only be raised in embryo culture, due to the lack of the endosperm [17]. The Ukrainian six-rowed winter barley ‘Manas’ may carry new allelic variation compared to the two-rowed spring cultivar ‘Betzes’ (generally used in wheat-barley crosses) as it has bet- ter agronomic traits (e.g. abiotic stress tolerance and yield ability), and it is well adapted to Central European conditions [18]. A hybrid previously developed between the winter wheat ‘Asakaze’ and the barley cultivar ‘Manas’ was subsequently screened to select a series of diso- mic and ditelosomic addition lines carrying the individual chromosomes/chromosome arms from barley in the wheat background [18,19]. Addition lines make it possible to study the effect of a specific barley chromosome added to wheat. However, as aneuploids are genetically unstable, this requires continuous cytogenetic checks. Wheat/barley translocation lines carry- ing 42 chromosomes are mostly stable and are directly applicable in prebreeding programmes [20,21]. Furthermore, compensating translocations are particularly valuable, as the alien homoeologous chromosomes are better able to compensate for the loss of a wheat chromo- some segment [22]. In a previous study the ‘Asakaze’/‘Manas’ 7H disomic addition (2n = 44) and 7HL ditelosomic addition lines (2n = 42+2 telocentrics) were characterized in detail for abiotic stress resistance and nutritional parameters. Germination tests performed under salt conditions and salt tolerance experiments in early development stages revealed higher salinity tolerance for both lines [19,23]. The addition of the barley chromosome 7H, carrying the gene responsible for MLG synthesis (HvCslF6 gene), to the hexaploid chromosome set of wheat has been found to increase the MLG content of the grain [24]. The aim of the present study was to develop a stable, fully fertile compensating winter wheat/winter barley translocation line (2n = 42) directly amenable to wheat breeding PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 2 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content programmes, which ultimately enhances salt tolerance and the MLG content of the grain. Plant material Plants with a monosomic Robert- sonian translocation were identified in the F3 generation using GISH, and disomic (homozy- gous) wheat-barley centric fusions were selected in the F4 generation (S1 Fig). Fertility (number of seeds/spike) was assigned based on the main spikes of plants grown in phytotron growth cabinets and in the Martonva´sa´r nursery in 2014–2015 and 2015–2016, respectively. h l d f k f h h h In phytotron experiments vernalisation was carried out at 4˚C for 6 weeks, after which the vernalised plants were grown in 2 L pots filled with a 2:1:1 mixture of garden soil, humus and sand. The plants were grown until tillering under an initial regime of 15˚C day: 10˚C night temperature, 12 h light: 12 h dark photoperiod. The temperature was raised by increments of 2˚C after tillering (day length 14 h), stem elongation (16 h illumination), flowering and 2 weeks after fertilization. Introduction The breakage-fusion mechanism of univalent chromosomes was exploited to produce a new 7BS.7HL Robertsonian translocation line (RobT) and the new line was characterized by means of molecular genetic and cytogenetic methods (genomic in situ hybridization: GISH and fluo- rescence in situ hybridization: FISH), which confirmed the high salinity tolerance during ger- mination and the higher MLG content of the grain, proving that 7BS.7HL RobT is a valuable genetic material readily applicable in wheat improvement. Plant material A crossing progamme was initiated in order to develop a compensating Robertsonian translo- cation in a wheat background, including the long arm of barley chromosome 7H, in order to obtain a stable, fully fertile genetic material carrying salt tolerance and elevated MLG content. A ‘Rannaya’ 7B winter wheat monosomic stock was pollinated with the previously generated ‘Asakaze’/‘Manas’ 7H disomic addition line. Double monosomics for wheat 7B and barley 7H chromosomes (carrying 42 chromosomes) were selected in the F1 generation using 7H-specific molecular microsatellite and STS markers and Feulgen staining. F1 plants were self-pollinated and descendants in the F2 generation were screened with specific markers for the 7HS and 7HL arms to detect individuals carrying the long arm of 7H. Plants with a monosomic Robert- sonian translocation were identified in the F3 generation using GISH, and disomic (homozy- gous) wheat-barley centric fusions were selected in the F4 generation (S1 Fig). Fertility (number of seeds/spike) was assigned based on the main spikes of plants grown in phytotron growth cabinets and in the Martonva´sa´r nursery in 2014–2015 and 2015–2016, respectively. In phytotron experiments vernalisation was carried out at 4˚C for 6 weeks, after which the vernalised plants were grown in 2 L pots filled with a 2:1:1 mixture of garden soil, humus and sand. The plants were grown until tillering under an initial regime of 15˚C day: 10˚C night temperature, 12 h light: 12 h dark photoperiod. The temperature was raised by increments of 2˚C after tillering (day length 14 h), stem elongation (16 h illumination), flowering and 2 weeks after fertilization. A crossing progamme was initiated in order to develop a compensating Robertsonian translo- cation in a wheat background, including the long arm of barley chromosome 7H, in order to obtain a stable, fully fertile genetic material carrying salt tolerance and elevated MLG content. A ‘Rannaya’ 7B winter wheat monosomic stock was pollinated with the previously generated ‘Asakaze’/‘Manas’ 7H disomic addition line. Double monosomics for wheat 7B and barley 7H chromosomes (carrying 42 chromosomes) were selected in the F1 generation using 7H-specific molecular microsatellite and STS markers and Feulgen staining. F1 plants were self-pollinated and descendants in the F2 generation were screened with specific markers for the 7HS and 7HL arms to detect individuals carrying the long arm of 7H. PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 Genomic in situ hybridization (GISH) and fluorescence in situ hybridization (FISH) Mitotic chromosome spreads were obtained from germinating root tips as described by Lukas- zewski [25]. GISH technique has been used to distinguish alien chromosomes from wheat chromosomes and wheat-alien translocated chromosomes in a wheat background [26, 27]. The GISH experiment was carried out using total genomic barley DNA labelled with a dig- nick-translation mix (Roche Diagnostics, Mannheim, Germany). Unlabelled wheat genomic DNA was used as a blocking agent at a ratio of 35:1. Probe detection was carried out with anti- digoxigenin-Rhodamine (Roche). Slides were mounted in Vectashield antifade solution (Vec- tor Laboratories, Burlingame, USA) containing 2 μgmL-1 4’-6-diamidino-2-phenylindole (DAPI). FISH was performed after rinsing the GISH hybridization signals off in 4 × saline- sodium citrate (4X SSC) Tween 20 (Sigma Aldrich, Darmstadt, Germany) at room tempera- ture. Wheat-specific DNA repetitive probes (pSc 119.2, Afa family and pTa 71) [28–30] were used to identify the wheat chromosome arm involved in the Robertsonian translocation and to confirm the presence of the entire wheat genome (except the 7BL chromosome arm). The Afa family and pSc119.2 repetitive probes were amplified and labelled with PCR using digoxi- genin-11-dUTP and biotin-16-dUTP, respectively, while the 45S rDNA clone pTa71 was PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 3 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content labelled with 50% biotin-16-dUTP and 50% digoxigenin-11-dUTP by nick translation. Digoxi- genin and biotin signals were detected with anti-digoxigenin-Rhodamine Fab fragments and streptavidin-FITC (Roche), respectively. The fluorescence signals were vizualized using a Zeiss Axioskop-2 fluorescence microscope (Zeiss, Oberkochen, Germany) equipped with filter sets appropriate for DAPI, FITC, Rhodamine and the simultaneous detection of FITC and Rhoda- mine (double filter set). Images were captured with a Spot CCD camera (Diagnostic Instru- ments, Sterling Heights, USA) and processed with Image Pro Plus software (Media Cybernetics, Silver Spring, USA). SSR and STS marker analysis Genomic DNA was extracted from fresh young leaves of wheat cultivar ‘Asakaze’, barley culti- var ‘Manas’ and the wheat-barley RobT using Quick Gene-Mini80 (FujiFilm, Japan) with a Quick-Gene DNA tissue kit (FujiFilm, Japan) according to the manufacturer’s instructions. Barley chromosome arm-specific SSR markers (Bmac0031-7HS, HvID-7HL) [31] and the gene-specific HvCslF6 STS marker [32] were used to reveal the presence of the 7HS and 7HL barley chromosome arms. The 7BL chromosome arm-specific SSR marker (Wmc311-7BL) [33] was used to reveal the absence of the 7BL wheat chromosome arm. The primer pairs were tested on DNA templates containing genomic DNA from wheat cultivar ‘Rannaya’, the ‘Asa- kaze’/‘Manas’ 7H disomic addition line (7H), and the ‘Asakaze’/‘Manas’ RobT. PCR amplifica- tion was carried out under the conditions described by Cseh et al. [34]. PCR products were separated with a Fragment Analyzer Automated CE System equipped with a 96-Capillary Array Cartridge (Advanced Analytical Technologies, USA). The results were interpreted using PROsize v2.0 software (Advanced Analytical Technologies, USA). Morphological characterization of the 7BS.7HL RobT The number of seeds per main spike was determined from 2 × 10 plants for each genotype (‘Rannaya’ wheat cultivar, 7H addition line and 7BS.7HL RobT) in experiments carried out in the Martonva´sa´r phytotron and nursery. Evaluation of the salinity tolerance of the 7BS.7HL RobT The salt stress response of the ‘Rannaya’ wheat genotype, the ‘Asakaze’/‘Manas’ 7H disomic addition line and 7BS.7HL RobT was evaluated after germination at the seedling stage. To determine the salt stress tolerance, 3×20 seeds from each genotype were surface-sterilized in 10% sodium hypochlorite for 15 minutes, rinsed twice in distilled water and germinated on wet filter paper in Petri dishes containing 0, 100, 150, 200 or 250 mM NaCl for 8 days (3 days in the dark and 5 days in the light) at a temperature of 25˚C. The percentage of germinated seeds was determined after 5 days and the length of roots and coleoptiles was measured on the 8th day after germination. Development of the wheat-barley 7BS.7HL RobT Earlier results revealed higher salt tolerance and MLG content in wheat-barley introgression lines carrying the 7H chromosome or 7HL arm [23,24]. In the present study the ‘Rannaya’ monosomic wheat line was crossed as female partner to the ‘Asakaze’/‘Manas’ 7H disomic addition line in order to induce rearrangements and develop a stable wheat/barley RobT carry- ing the favourable traits. Ten of the plants raised from 25 germinated F1 seeds were found to be monosomic for the 7B and 7H chromosomes when tested 7H with specific molecular mark- ers and Feulgen staining. The F1 plants produced a total of 68 seeds with an average yield of below 10 seeds/plant, except for two individuals which yielded 21 and 22 seeds, respectively. From the 68 F2 seeds 66 germinated and these plants were screened for the presence of the 7HS or 7HL chromosome arms with the 7HS-specific Bmac0031 SSR marker and the 7HL- specific HvCSlF6 STS marker. The analysis identified two plants carrying only the 7HS barley chromosome arm and three plants containing solely the 7HL arm. Plants carrying either 7HS or 7HL were analysed in F3 with GISH in order to select progenies with wheat-barley Robert- sonian translocations. Seventy-two F3 seeds were screened, of which 7 plants (9.72%) had the wheat-barley translocation (7BS.7HL). Ten other plants retained telocentrics either individu- ally or in pairs, while three plants carried one barley isochromosome (7HS), suggesting the fusion of a telocentric pair. The barley chromatin was entirely eliminated from 52 of the F3 plants analysed. Individuals carrying the Robertsonian translocations were selfed and 50 F4 plants were screened by GISH (S1 Fig) of which three were found to be disomic lines (6%), 22 (44%) monosomic lines, 21 (42%) nullisomic lines and 4 (8%) plants carried a barley telocen- tric chromosomes. During the examination of disomic RobT lines the GISH results comfirmed the presence of 40 wheat chromosomes and the recombinant chromosome pair (7BS.7HL) (Fig 1A). Following random selection 49 F5 seeds originated from disomic plants were germi- nated and analysed by GISH and these were 100% disomic 7BS.7HL translocations. The barley chromosome arms were identified by molecular marker analysis using the above specific molecular markers (Fig 2). The whole 7HL arm was detected by means of an STS marker located in the centromeric region of 7HL (HvCSLF6) and an SSR marker mapped to the telomeric region of the 7HL chromosome arm (HvID). Evaluation of (1,3;1,4)-β-D-glucan content of the seeds of 7BS.7HL RobT Four biological replicates of mature seeds (5 plants per replicate) were collected from the fol- lowing genotypes: ‘Rannaya’ wheat cultivar, ‘Manas’ barley cultivar, 7H disomic addition line and 7BS.7HL RobT, and milled in a Retch M400 ball laboratory mill to produce wholemeal. Samples of finely ground grains of the plants (100 mg) were used for MLG analysis in four par- allels according to AACC Method 32–23 [35] using a Megazyme kit (Megazyme, Bray, Ire- land). The samples were suspended and hydrated in a buffer solution (pH 6.5) and then incubated with purified lichenase enzyme and filtered. An aliquot of the filtrate was then PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 4 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content hydrolysed to completion with purified β-glucosidase. The quantity of produced D-glucose was determined using a glucose oxidase/peroxidase reagent. Statistical analysis The fertility of 7BS.7HL RobT, the wheat cultivar ‘Rannaya’ and the 7H disomic addition line was compared pair-wise and differences in fertility were evaluated by means of Tukey’s post hoc test (SPSS 16.0) at the 0.05 significance level. The data from the salinity tolerance experiment were also analysed using Tukey’s post hoc test (SPSS 16.0) at the 0.05 significance level. Chemical data from the MLG experiment were evaluated using single-factor analysis in the Microsoft Excel program and were also analysed using Tukey’s post hoc test (SPSS 16.0) Val- ues of P<0.05 were considered to be statistically significant. PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 Development of the wheat-barley 7BS.7HL RobT and descendants of the disomic plants were maintained and multiplied in phytotron growth cabinets (F5) and in the Martonva´sa´r nursery (F6-F7 generations). Development of the wheat-barley 7BS.7HL RobT After washing the GISH signals off, FISH was performed in order to verify the identity of the wheat chromosome arm taking part in the Robertsonian translocation. The translocated wheat chromosome arm exhibited only two subterminal pSc119.2 signals and was thus identified as 7BS (Fig 1B, Fig 3). The parental genotypes (‘Rannaya’ wheat, ‘Manas’ barley and the 7H disomic addition line used as control) PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 5 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content Fig 1. Genomic in situ hybridization and Fluorescence in situ hybridization on mitotic chromosomes of t 7BS.7HL Robertsonian translocation line. (a) Genomic in situ hybridization (GISH) on mitotic chromosom l 0206248 N b 5 2018 Fig 1. Genomic in situ hybridization and Fluorescence in situ hybridization on mitotic chromosomes of the 7BS.7HL Robertsonian translocation line. (a) Genomic in situ hybridization (GISH) on mitotic chromosomes of the Fig 1. Genomic in situ hybridization and Fluorescence in situ hybridization on mitotic chrom 7BS.7HL Robertsonian translocation line. (a) Genomic in situ hybridization (GISH) on mitoti Fig 1. Genomic in situ hybridization and Fluorescence in situ hybridization on mitotic chromosomes of the 7BS.7HL Robertsonian translocation line. (a) Genomic in situ hybridization (GISH) on mitotic chromosomes of the 6 / 17 PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content 7BS.7HL Robertsonian translocation line (RobT). Labelled total genomic DNA of barley was used as probe and barley chromosome arm 7HL is highlighted in magenta (arrows). The chromosomes were counterstained with DAPI (blue). (b) Fluorescence in situ hybridization on mitotic chromosomes of 7BS.7HL RobT. DNA repetitive probes: Afa family (red), pSc119.2 (green), pTa71 (orange). Unidentified A genomic chromosomes are labelled by an ‘A’ letter. Scale bar = 10μm. 7BS.7HL Robertsonian translocation line (RobT). Labelled total genomic DNA of barley was used as probe and barley chromosome arm 7HL is highlighted in magenta (arrows). The chromosomes were counterstained with DAPI (blue). (b) Fluorescence in situ hybridization on mitotic chromosomes of 7BS.7HL RobT. DNA repetitive probes: Afa family (red), pSc119.2 (green), pTa71 (orange). Unidentified A genomic chromosomes are labelled by an ‘A’ letter. Scale bar = 10μm and descendants of the disomic plants were maintained and multiplied in phytotron growth cabinets (F5) and in the Martonva´sa´r nursery (F6-F7 generations). Morphological characterization of the 7BS.7HL RobT The 7H addition line showed low fertility, especially in the apical part of the spike, under both phytotron and field conditions. On the other hand, the fertility of the RobT was similar to that Fig 2. Capillary gel electrophoresis pattern of the 7H chromosome specific molecular markers and the 7BL chromosome arm-specific marker. Capillary gel electrophoresis pattern of the Bmac0031 (7HS), HvCslF6 (7HL) and HvID (7HL) 7H chromosome arm-specific molecular markers and the Wmc311 7BL chromosome arm-specific marker on the following DNA templates: ‘Rannaya’ wheat cultivar (R), ‘Asakaze’/‘Manas’ 7H disomic addition line (7H), and 7BS.7HL disomic Robertsonian translocation lines (RobT1, RobT2, RobT3). Chromosome arm-specific bands are indicated by arrows. https://doi.org/10.1371/journal.pone.0206248.g002 Fig 2. Capillary gel electrophoresis pattern of the 7H chromosome specific molecular markers and the 7BL chromosome arm-specific marker. Capillary gel electrophoresis pattern of the Bmac0031 (7HS), HvCslF6 (7HL) and HvID (7HL) 7H chromosome arm-specific molecular markers and the Wmc311 7BL chromosome arm-specific marker on the following DNA templates: ‘Rannaya’ wheat cultivar (R), ‘Asakaze’/‘Manas’ 7H disomic addition line (7H), and 7BS.7HL disomic Robertsonian translocation lines (RobT1, RobT2, RobT3). Chromosome arm-specific bands are indicated by arrows. htt //d i /10 1371/j l 0206248 002 Fig 2. Capillary gel electrophoresis pattern of the 7H chromosome specific molecular markers and the 7BL chromosome arm-specific marker. Capillary gel electrophoresis pattern of the Bmac0031 (7HS), HvCslF6 (7HL) and HvID (7HL) 7H chromosome arm-specific molecular markers and the Wmc311 7BL chromosome arm-specific marker on the following DNA templates: ‘Rannaya’ wheat cultivar (R), ‘Asakaze’/‘Manas’ 7H disomic addition line (7H), and 7BS.7HL disomic Robertsonian translocation lines (RobT1, RobT2, RobT3). Chromosome arm-specific bands are indicated by arrows. 7 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content Fig 3. FISH pattern of chromosome 7B of the ‘Asakaze’ and ‘Rannaya’ wheat cultivars and the 7BS.7HL Robertsonian translocation chromosome. DNA repetitive probes: Afa family (red) and pSc119.2 (green). https://doi.org/10.1371/journal.pone.0206248.g003 Fig 3. FISH pattern of chromosome 7B of the ‘Asakaze’ and ‘Rannaya’ wheat cultivars and the 7BS.7HL Robertsonian translocation chromosome. DNA repetitive probes: Afa family (red) and pSc119.2 (green). https://doi.org/10.1371/journal.pone.0206248.g003 https://doi.org/10.1371/journal.pone.0206248.g003 of the ‘Rannaya’ winter wheat cultivar, demonstrating that the crossing strategy had sucessfully restored the fertility of the 7H addition (S2 Fig). Morphological characterization of the 7BS.7HL RobT The significant improvement in fertility was reflected by the fact that although the spikes of the RobT were shorter (Fig 4) the number of seeds/spike was twice as high for the RobT than for the 7H addition line. PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 Salt stress response of 7BS.7HL RobT The germination % and the root and shoot growth of the ‘Rannaya’ wheat cultivar, the 7H addition line and the 7BS.7HL RobT were determined after germination in NaCl solutions with various concentrations (0, 100, 150, 200 and 250mM). These results are presented in Table 1. All three genotypes germinated well without salt treatment and the germination percentage decreased slightly (86%) in ‘Rannaya’ under the mild salt stress induced by 100mM NaCl. Above this NaCl concentration, the germination percentage decreased, especially in ‘Rannaya’. When treated with 150, 200 mM and 250 mM NaCl, the germination capacity of the wheat genotype dropped considerably (to 73%, 46% and 30%, respectively). In contrast, both the 7H addition line and the RobT maintained a very high germination rate when subjected to 150, 200 and 250 mM NaCl (85%, 78% and 70% for the 7H addition line and 90%, 80% and 75% for RobT, respectively). The salt tolerance of the ‘Rannaya’ wheat cultivar, the ‘Asakaze’/‘Manas’ disomic addition line and 7BS.7HL RobT was also investigated by measuring the root and shoot lengths on the 8th day after germination. Germinating the seeds in 100, 150, 200 or 250 mM NaCl solution resulted in a decrease of root and shoot coleoptile length as a % of the control (Fig 5). The growth reduction was more pronounced in ‘Rannaya’ than in the addition and the RobT. For instance, the roots of the 7BS.7HL RobT were twice as long as those of the wheat parental line when treated with 200 mM NaCl (Fig 5). The differences between the genotypes were more pronounced for the roots than for the shoots. The differences in shoot lengths were not as sig- nificant due to the higher standard deviation (Fig 5). Salt stress reduced the growth vigour of PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 8 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content Fig 4. Phenotype of spikes of the parental genotypes and the 7BS.7HL centric fusion line. Phenotype of spikes of the parental wheat genotype ‘Rannaya’ (1), the 7H disomic addition (2) and the 7BS.7HL RobT line (3). Plants were grown in phytotron climate chambers. https://doi org/10 1371/journal pone 0206248 g004 Fig 4. Phenotype of spikes of the parental genotypes and the 7BS.7HL centric fusion line. Salt stress response of 7BS.7HL RobT Phenotype of spikes of the parental wheat genotype ‘Rannaya’ (1), the 7H disomic addition (2) and the 7BS.7HL RobT line (3). Plants were grown in phytotron climate chambers. https://doi.org/10.1371/journal.pone.0206248.g004 Fig 4. Phenotype of spikes of the parental genotypes and the 7BS.7HL centric fusion line. Phenotype of spikes of the parental wheat genotype ‘Rannaya’ (1), the 7H disomic addition (2) and the 7BS.7HL RobT line (3). Plants were grown in phytotron climate chambers. https://doi.org/10.1371/journal.pone.0206248.g004 all the genotypes during the first 4 days, compared to that of the control plants (without NaCl treatment), but the introgression lines showed better growth during the subsequent 4 days and exhibited better growth vigour, resulting in greater differences between the genotypes (Fig 5). Table 1. Germination percentage of the genotypes analysed. 0 Mm NaCl 100 Mm NaCl 150 Mm NaCl 200 Mm NaCl 250 Mm NaCl Wheat cv. ‘Rannaya’ 100a 86bc 73de 46f 30g 7H addition line 100a 100a 85bc 78d 70e 7BS.7HL RobT 100a 100a 90b 80cd 75de Values are means ±SD of 3 x 20 replicates per genotype. Different letters indicate significant differences at P < 0.05 using Tukey’s post hoc test. htt //d i /10 1371/j l 0206248 t001 Table 1. Germination percentage of the genotypes analysed. PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 9 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content Fig 5. Root and shoot length measured at five levels of NaCl treatments. Root and shoot length measured on the 8th day after germination for wheat cultivar ‘Rannaya’, the ‘Asakaze’/‘Manas’ 7H disomic addition line and 7BS.7HL RobT at five levels of NaCl treatments (0, 100, 150, 200 and 250 mM solution). Different letters indicate significant differences at P < 0.05 using Tukey’s post hoc test. The pictures are represented the salt tolerance of the wheat parent (‘Rannaya’) and 7BS.7HL RobT during germination at different NaCl concentrations (0–250 mM). Percentages show the inhibition of root and shoot growth compared to the controll (0 mM NaCl). The reduction of the germination capacity and the inhibition of root and shoot growth were greater in wheat parent ‘Rannaya’ than in the 7BS.7HL line. Scale bar = 1cm. Fig 5. Root and shoot length measured at five levels of NaCl treatments. Salt stress response of 7BS.7HL RobT Root and shoot length measured on the 8th day after germination for wheat cultivar ‘Rannaya’, the ‘Asakaze’/‘Manas’ 7H disomic addition line and 7BS.7HL RobT at five levels of NaCl treatments (0, 100, 150, 200 and 250 mM solution). Different letters indicate significant differences at P < 0.05 using Tukey’s post hoc test. The pictures are represented the salt tolerance of the wheat parent (‘Rannaya’) and 7BS.7HL RobT during germination at different NaCl concentrations (0–250 mM). Percentages show the inhibition of root and shoot growth compared to the controll (0 mM NaCl). The reduction of the germination capacity and the inhibition of root and shoot growth were greater in wheat parent ‘Rannaya’ than in the 7BS.7HL line. Scale bar = 1cm. https://doi.org/10.1371/journal.pone.0206248.g005 https://doi.org/10.1371/journal.pone.0206248.g005 Analysis of the MLG content of 7BS.7HL RobT The mean MLG levels in four biological replicates of the studied genotypes are presented in Fig 6, together with the LSD 5% values. The mean MLG level (mg/g dry matter) of the ‘Ran- naya’ wheat cultivar was 7.04 mg (0.7%), while that of the barley cultivar ‘Manas’ was seven times as high (48.89 mg/g, 4.89%). The β-glucan content of the seeds was 8.93 mg/g (0.9%) for 7BS.7HL RobT and 10.93 mg/g (1.1%) for the 7H addition line, so the MLG level of the 7BS.7HL RobT, carrying the HvCslF6 gene, exceeded the β-glucan content of wheat by 26.85%. PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 Discussion The ‘Asakaze’/’Manas’ winter wheat-winter barley 7BS.7HL compensating RobT was devel- oped with the aim of transferring agronomically favourable genes from cultivated barley to bread wheat. The present work demonstrated that a stably inherited Robertsonian transloca- tion comprising the 7HL barley chromosome arm in a wheat background could effectively enhance the salintity tolerance of the wheat plant and increase the MLG content of the grain. Wheat-alien amphiploids and addition lines cannot be directly used in wheat breeding pro- grammes because of their genetic instability. On the other hand, stably inherited translocation lines carry a subchromosomal segment of an alien chromosome integrated into the wheat genome and usually comprise 42 chromosomes. A number of wheat-ancestral Robertsonian translocations were reported involving wild relatives of wheat, such as Dasypirum villosum PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 10 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content Fig 6. (1,3;1,4)-β-D-glucan content in the grains. Mixed-linkage β-glucan content in the grains of wheat (‘Rannaya’), barley (‘Manas’), the 7H disomic addition line (7H) and 7BS.7HL RobT (7BS.7HL). Mixed-linkage β-glucan analysis was performed in four parallels according to AACC Method 32–23. Different letters indicate significant differences at P < 0.05 using Tukey’s post hoc test. Bars show the LSD 5% value: 0.74. https://doi.org/10.1371/journal.pone.0206248.g006 Fig 6. (1,3;1,4)-β-D-glucan content in the grains. Mixed-linkage β-glucan content in the grains of wheat (‘Rannaya’), barley (‘Manas’), the 7H disomic addition line (7H) and 7BS.7HL RobT (7BS.7HL). Mixed-linkage β-glucan analysis was performed in four parallels according to AACC Method 32–23. Different letters indicate significant differences at P < 0.05 using Tukey’s post hoc test. Bars show the LSD 5% value: 0.74. https://doi.org/10.1371/journal.pone.0206248.g006 Fig 6. (1,3;1,4)-β-D-glucan content in the grains. Mixed-linkage β-glucan content in the grains of wheat (‘Rannaya’), barley (‘Manas’), the 7H disomic addition line (7H) and 7BS.7HL RobT (7BS.7HL). Mixed-linkage β-glucan analysis was performed in four parallels according to AACC Method 32–23. Different letters indicate significant differences at P < 0.05 using Tukey’s post hoc test. Bars show the LSD 5% value: 0.74. Fig 6. (1,3;1,4)-β-D-glucan content in the grains. Mixed-linkage β-glucan content in the grains of wheat (‘Rannaya’), barley (‘Manas’), the 7H disomic addition line (7H) and 7BS.7HL RobT (7BS.7HL). Mixed-linkage β-glucan analysis was performed in four parallels according to AACC Method 32–23. PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 Discussion Different letters indicate significant differences at P < 0.05 using Tukey’s post hoc test. Bars show the LSD 5% value: 0.74. https://doi.org/10.1371/journal.pone.0206248.g006 https://doi.org/10.1371/journal.pone.0206248.g006 [36], Haynaldia villosa [37], Thinopyrum bessarabicum [38], Thinopyrum intermedium [39], Aegilops searsii [40] and Aegilops speltoides [41]. For instance, the stripe rust resistance gene Sr59 was introgressed into hexaploid wheat from rye through a compensating 2DS.2RL Robertsonian translocation [42]. One of the most well-known and widely applied ancestral introgressions was the transfer of the 1RS rye chromosome arm from the Russian cultivars ‘Aurora’ and ‘Kavkaz’ into the wheat background as a compensating 1BL.1RS translocation [43, 44]. Consequently, 1BL.1RS is present in more than 1000 wheat cultivars worldwide and contributed to improving the yield potential, adaptability and biotic stress resistance of wheat [45,46]. [36], Haynaldia villosa [37], Thinopyrum bessarabicum [38], Thinopyrum intermedium [39], Aegilops searsii [40] and Aegilops speltoides [41]. For instance, the stripe rust resistance gene Sr59 was introgressed into hexaploid wheat from rye through a compensating 2DS.2RL Robertsonian translocation [42]. One of the most well-known and widely applied ancestral introgressions was the transfer of the 1RS rye chromosome arm from the Russian cultivars ‘Aurora’ and ‘Kavkaz’ into the wheat background as a compensating 1BL.1RS translocation [43, 44]. Consequently, 1BL.1RS is present in more than 1000 wheat cultivars worldwide and contributed to improving the yield potential, adaptability and biotic stress resistance of wheat [45,46]. [36], Haynaldia villosa [37], Thinopyrum bessarabicum [38], Thinopyrum intermedium [39], Aegilops searsii [40] and Aegilops speltoides [41]. For instance, the stripe rust resistance gene Sr59 was introgressed into hexaploid wheat from rye through a compensating 2DS.2RL Robertsonian translocation [42]. One of the most well-known and widely applied ancestral introgressions was the transfer of the 1RS rye chromosome arm from the Russian cultivars ‘Aurora’ and ‘Kavkaz’ into the wheat background as a compensating 1BL.1RS translocation [43, 44]. Consequently, 1BL.1RS is present in more than 1000 wheat cultivars worldwide and contributed to improving the yield potential, adaptability and biotic stress resistance of wheat [45,46]. Only a few wheat-barley compensating translocation lines are so far available for wheat improvement. Koba et al. [47] reported the development of a 5HS.5BL RobT in addition to the 42 wheat chromosomes from a cross between the ‘Schinchunaga’ wheat and ‘Nyugoruden’ PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 11 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content barley cultivars. Danilova et al. Discussion [48] produced a complete set of homoeologous group 7 com- pensating translocations introducing chromosome arms from the ‘Betzes’ spring barley variety into the ‘Chinese Spring’ wheat background. The ‘Manas’ barley used in the present study is a six-rowed winter variety having good agronomic traits (high yield potential, winter hardiness, salt tolerance, high MLG content) [18]. Similarly, the ‘Rannaya’ wheat used as female parent in the cross is a cultivated winter wheat with high yielding potential, making it more suitable for breeding purposes. Depending on the wheat and alien chromosomes involved and the environmental condi- tions, the desired compensating wheat-alien Robertsonian translocations can be recovered at variable frequencies, ranging from low (1.77%) to fairly high (20%) [36,48–50]. In this study the efficiency of the breakage-fusion mechanism was 7.6% in the F2 generation. Besides the crossing procedure and the detailed genetic investigation, this study focused on the phenotyp- ing of the newly developed plant material. Wheat is not tolerant to high level of soil salt con- tent. King et al.[51] reported the possibility of transferring salt tolerance genes from a wild relative (Thinopyrum bessarabicum) into bread wheat. In a previous study, germination tests performed under salt conditions and salt tolerance tests during early development stages con- firmed the higher salinity tolerance of lines carrying the 7H barley chromosome or the 7HL arm compared to wheat and other wheat-barley addition lines (2H, 3H, 4H, 6H and 7HS) [19,23]. Detailed physiological, biochemical and molecular genetic investigations related to typical salt tolerance mechanisms revealed that Na+ uptake and transport are not responsible for the elevated salt tolerance of the 7H and 7HL addition lines. The higher salt tolerance was mainly due to the improved osmotic adjustment capacity of these lines [52]. Osmotic adjust- ment is one of the mechanisms contributing to the tolerance of salt-stressed plants [53]. QTLs associated with genes responsible for the relative water content (RWC) and water-soluble car- bohydrate concentration (WSC) of plants were identified and mapped using composite inter- val mapping to barley chromosome arm 7HL [54]. The bSS1B gene is associated with a QTL for RWC encodes barley sucrose synthase, which is a key enzyme in the carbohydrate metabo- lism, catalysing fructose and UDP-glucose synthesis [55,56], while the Acl3 gene encodes a co- factor protein associated with a QTL for RWC on the 7HL arm, having a role in membrane protection during stress [54]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 Discussion Earlier experiments revealed that the improved salt tolerance of the 7H addition line was due to osmotic adjustment achieved through proline accumulation and enhanced soluble car- bohydrate metabolism [52]. The similar experimental design used in the present experiments showed that both the ‘Asakaze’/‘Manas’ 7H disomic addition line and the 7BS.7HL RobT per- formed better under salt stress conditions than the wheat parental cultivar. Barley cv. ‘Manas’ and the 7HL ditelosomic addition thus gave results similar to those observed earlier for the 7H addition line, suggesting that the enhanced salt tolerance observed for the 7BS.7HL transloca- tion line also operates via osmotic adjustment [52]. The slow root and shoot growth during the first 4 days of salt treatment and the subsequent acceleration further support this assumption, as the reprogramming of the carbohydrate metabolism and the accumulation of large pools of soluble sugars require several days. However, further investigations are needed to confirm the role of sugars in the salt tolerance mechanisms operating in the 7BS.7HL RobT line. The seed MLG content of the ‘Manas’ barley cultivar is seven times as high as that of ‘Ran- naya’ wheat. The presence of the 7HL barley chromosome arm increased the MLG content of the seeds by 26.83%, demonstrating that the HvCslF6 gene responsible for its synthesis is func- tional in the wheat genetic background. Recent QTL and genome-wide association studies suggested that other proteins may also play a role in (1,3;1,4)-β-D-glucan biosynthesis, imply- ing that the introduction of other genes from barley would result in higher dietary fibre con- tent in wheat. Canditate genes were mapped to the 1HS, 2HL, 3HS, 3HL, 4HL, 5HS, 6HS and PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 12 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content 7HS chromosome arms [57–59]. The current model for (1,3;1,4)-β-D-glucan biosynthesis pro- poses that the HvCSLF6 enzyme is only responsible for creating the β-(1,4)-glucosidic linkages, which are then joined together to produce the mixed-linkage glucan through a β-(1,3)-gluco- sidic linkage catalysed by an as yet unidentified protein [60]. The higher MLG content of the 7H addition line compared to the RobT thus suggests that the QTLs mapped on 7HS and 7HL may have an additive effect and that the presence of both groups of loci results in a higher MLG level (Fig 6). Conclusions The introgression of barley chromatin segments into wheat, as in the stably inherited whole arm Robertsonian translocation developed in the present study, widens the genetic diversity of cultivated bread wheat. Evidence was provided that functional genes originating from the long arm of barley chromosome 7H improved the salinity tolerance and increased the (1,3;1,4)-β- D-glucan content of the seed in the newly developed ‘Asakaze’/‘Manas’ 7BS.7HL introgression line. The genome stability and fertility of the compensating translocation line make it a suitable genetic material for direct application in wheat breeding programmes. Discussion Besides the traits characterized in this study the 7HL chromosome arm carries other genes of breeding interest. For instance, a locus (Rps6) conferring stripe rust (Puccinia striiformis) resistance was likewise mapped to the distal region of 7HL in the barley cultivar ‘Tamalpais’ and the barley line ‘Y12’ [61]. The wheat-barley RobT produced in the present work thus has the potential to improve stripe rust resistance in wheat. The ‘Ranaya’/‘Manas’ 7BS.7HL compensating RobT carries 42 chromosomes, and genomic in situ hybridization revealed a high degree of chromosome stability, implying that the intro- gressed agronomic characters are stably inherited. Moreover, the fertility of the 7BS.7HL RobT was similar to that of wheat, indicating that the infertility of the parental 7H addition line had been successfully overcomed. The good yield potential and fertility of the RobT makes it possi- ble to use this line directly in wheat breeding programmes to improve biotic and abiotic stress resistance and to confer better nutritional parameters on cultivated bread wheat. S1 Fig. Crossing strategy applied for the development of 7BS.7HL RobT. (DOCX) S1 Fig. Crossing strategy applied for the development of 7BS.7HL RobT. (DOCX) S2 Fig. Number of kernels/main spike of the 7BS.7HL RobT, ‘Rannaya’ wheat cultivar and ‘Asakaze’/‘Manas’ 7H disomic addition line. Number of kernels/main spike of ten randomly selected plants (1–10) in Martonva´sa´r phytotron and field experiments for the 7BS.7HL RobT and for the ‘Rannaya’ wheat cultivar and ‘Asakaze’/‘Manas’ 7H disomic addition line used as control. Different letters indicate significant differences at P < 0.05 using Tukey’s post hoc test. (DOCX) References 1. Pareek A, Sopory SK, Bohnert HJ, Govindjee. Abiotic stress adaptation in plants. Physiological, Molec- ular and Genomic Foundation. Dordrecht: Springer; 2010. 2. Mantri N, Patade V, Penna S, Ford R, Pang E. Abiotic stress responses in plants: present and future. In: Ahmad P, Prasad MNV, editors. Abiotic Stress Responses in Plants. Metabolism, Productivity and Sus- tainability; New York: Springer; 2012. pp.1–19. 3. Slavich PG, Read BJ, Cullis BR. Yield response of barley germplasm to field variation in salinity quanti- fied using the EM-38. Aust J Exp Agr.1990; 30:551–556. 4. Jaradat AA, Shahid M, Al-Maskri A. Genetic diversity in the Batini barley landrace from Oman: I. Spike and seed quantitative and qualitative traits. Crop Sci. 2004; 44:304–315. 5. Munns R, Tester M. Mechanisms of salinity tolerance. Annu Rev Plant Biol. 2008; 59:651–681. https:// doi.org/10.1146/annurev.arplant.59.032607.092911 PMID: 18444910 6. Widodo, Patterson JH, Newbigin E, Tester M, Bacic A, Roessner U. Metabolic responses to salt stress of barley (Hordeum vulgare L.) cultivars, Sahara and Clipper, which differ in salinity tolerance. J Exp Bot. 2009; 60:4089–4103. https://doi.org/10.1093/jxb/erp243 PMID: 19666960 7. Wu H, Shabala L, Barry K, Zhou M, Shabala S. Ability of leaf mesophyll to retain potassium correlates with salinity tolerance in wheat and barley. Physiol Plantarum. 2013; 149:515–527. 8. Munns R, Gardner P, Tonnet ML, Rawson H. Growth and development in NaCl-treated plants. Do Na+ or Cl−concentrations in dividing or expanding tissues determine growth in barley? Aust J Plant Physiol. 1988; 15:529–540. 9. Zhou G, Johnson P, Ryan PR, Delhaize E, Zhou M. Quantitative trait loci for salinity tolerance in barley (Hordeum vulgare L.) Mol Breeding. 2012; 29:427–436. 10. Ma Y, Shabala S, Li C, Liu C, Zhang W, Zhou M. Quantitative trait loci for salinity tolerance identified under drained and waterlogged conditions and their association with flowering time in barley (Hordeum vulgare L.). PLoS ONE 2018; 10(8):e0134822. Available from: https://journals.plos.org/plosone/ article?id=10.1371/journal.pone.0134822 11. Kerckhoffs DA, Hornstra G, Mensink RP. Cholesterol-lowering effect of β-glucan from oat bran in mildly hypercholesterolemic subjects may decrease when β-glucan is incorporated into bread and cookies. Am J Clin Nutr. 2003; 78:221–227. https://doi.org/10.1093/ajcn/78.2.221 PMID: 12885701 12. Shelat KJ, Vilaplana F, Nicholson TM, Wong KH, Gidley MJ, Gilbert RG. Diffusion and viscosity in arabi- noxylan solutions: Implications for nutrition. Carbohyd Polym. 2011; 82:46–53. 13. Molina-Cano JL, Moralejo M, Elia M, Munoz P, Russell JR, Perez-Vendrell AM et al. Investigation: Edina Tu¨rko¨si, Eva Darko, Marianna Rakszegi, Andra´s Cseh. Supervision: Eva Darko, Istva´n Molna´r, Andra´s Cseh. Visualization: Edina Tu¨rko¨si, Marianna Rakszegi, Andra´s Cseh. Writing – original draft: Edina Tu¨rko¨si, Eva Darko, Andra´s Cseh. Writing – review & editing: Ma´rta Molna´r-La´ng, Andra´s Cseh. Writing – review & editing: Ma´rta Molna´r-La´ng, Andra´s Cseh. Author Contributions Conceptualization: Andra´s Cseh. Data curation: Andra´s Cseh. Formal analysis: Eva Darko, Marianna Rakszegi, Andra´s Cseh. Funding acquisition: Andra´s Cseh. Conceptualization: Andra´s Cseh. Data curation: Andra´s Cseh. Formal analysis: Eva Darko, Marianna Rakszegi, Andra´s Cseh. Funding acquisition: Andra´s Cseh. Conceptualization: Andra´s Cseh. Data curation: Andra´s Cseh. Formal analysis: Eva Darko, Marianna Rakszegi, Andra´s Cseh. Funding acquisition: Andra´s Cseh. PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 13 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content Investigation: Edina Tu¨rko¨si, Eva Darko, Marianna Rakszegi, Andra´s Cseh. Project administration: Edina Tu¨rko¨si, Andra´s Cseh. Resources: Eva Darko, Andra´s Cseh. Supervision: Eva Darko, Istva´n Molna´r, Andra´s Cseh. Validation: Eva Darko. Visualization: Edina Tu¨rko¨si, Marianna Rakszegi, Andra´s Cseh. Writing – original draft: Edina Tu¨rko¨si, Eva Darko, Andra´s Cseh. Writing – review & editing: Ma´rta Molna´r-La´ng, Andra´s Cseh. Investigation: Edina Tu¨rko¨si, Eva Darko, Marianna Rakszegi, Andra´s Cseh. Project administration: Edina Tu¨rko¨si, Andra´s Cseh. Resources: Eva Darko, Andra´s Cseh. Supervision: Eva Darko, Istva´n Molna´r, Andra´s Cseh. Validation: Eva Darko. Visualization: Edina Tu¨rko¨si, Marianna Rakszegi, Andra´s Cseh. Writing – original draft: Edina Tu¨rko¨si, Eva Darko, Andra´s Cseh. Writing – review & editing: Ma´rta Molna´r-La´ng, Andra´s Cseh. References 25. Lukaszewski AJ, Rybka K, Korzun V, Malyshev SV, Lapinski B, Whitkus R. Genetic and physical map- ping of homoeologous recombination points involving wheat chromosome 2B and rye chromosome 2R. Genome. 2004; 47:36–45. https://doi.org/10.1139/g03-089 PMID: 15060600 26. Schwarzacher T, Leitch AR, Bennett MD, Heslop-Harrison JS. In situ localization of parental genomes in a wide hybrid. Ann Bot.1989; 64:315–324. 27. Heslop-Harrison JS. The molecular cytogenetics of plants. J Cell Sci. 1991; 100:15–21. 28. Bedbrook JR, Jones J, O’Dell M, Thompson RD, Flavell RB. A molecular description of telomeric het- erochromatin in secale species. Cell. 1980; 19:545–560. PMID: 6244112 29. Nagaki K, Tsujimoto H, Isono K, Sasakuma T. Molecular characterization of a tandem repeat, Afa fam- ily, and its distribution among Triticeae. Genome. 1995; 38:479–486. PMID: 7557360 30. Gerlach WL, Bedbrook JR. Cloning and characterization of ribosomal RNA genes from wheat and bar- ley. Nucleic Acids Res. 1979; 7:1869–1885. PMID: 537913 31. Ramsay L, Macaulay M, degli Ivanissevich S, MacLean K, Cardle L, Fuller J, et al. A simple sequence repeat-based linkage map of barley. Genetics. 2000; 156:1997–2005. PMID: 11102390 32. Burton RA, Jobling SA, Harvey AJ, Shirley NJ, Mather DE, Bacic A, Fincher GB. The genetics and tran- scriptional profiles of the cellulose synthase-like HvCslF gene family in barley. Plant Physiol. 2008; 146:1821–1833. https://doi.org/10.1104/pp.107.114694 PMID: 18258691 33. Somers DJ, Isaac P, Edwards K. A high-density microsatellite consensus map for bread wheat (Triticum aestivum L.). Theor Appl Genet. 2004; 109:1105–1114. https://doi.org/10.1007/s00122-004-1740-7 PMID: 15490101 34. Cseh A, Kruppa K, Molna´r I, Rakszegi M, Dolezˇel J, Molna´r-La´ng M. Characterization of a new 4BS.7HL wheat–barley translocation line using GISH, FISH, and SSR markers and its effect on the β- glucan content of wheat. Genome. 2011; 54:795–804. https://doi.org/10.1139/g11-044 PMID: 21919737 35. McCleary BV, Codd R. Measurement of (1–3),(1–4)-β-D-glucan in barley and oats: a streamlined enzy- matic procedure. J Sci Food Agr. 1991; 55:303–312. 36. Zhao W, Qi L, Gao X, Zhang G, Dong J, Chen Q, Friebe B, Gill BS. Development and characterization of two new Triticum aestivum–Dasypyrum villosum Robertsonian translocation lines T1DS1V#3L and T1DL1V#3S and their effect on grain quality. Euphytica. 2010; 175:343–350. 37. Wilson, J. Production of wheat–Haynaldia villosa Robertsonian chromosomal translocations. M.Sc. Thesis, Kansas State University. 2008. Available from: http://krex.k-state.edu/dspace/bitstream/handle/ 2097/1085/JamieWilson2008.pdf?sequence=1. 38. Grewal S, Yang C, Hubbart Edwards S, Scholefield D, Ashling S, Burridge AJ, King IP, King J. Charac- terisation of Thinopyrum bessarabicum chromosomes through genome-wide introgressions into wheat. Theor Appl Genet. References QTL analysis of a cross between European and North American malting barleys reveals a putative candidate gene for beta-glucan content on chromosome 1H. Mol Breeding. 2007; 19: 275–284. 14. Burton RA, Wilson SM, Hrmova M, Harvey AJ, Shirley NJ, Medhurst A et al. Cellulose synthaselike CslF genes mediate the synthesis of cell wall (1,3;1,4)-beta-D-glucans. Science. 2006; 311: 1940– 1942. https://doi.org/10.1126/science.1122975 PMID: 16574868 15. Taketa S, Yuo T, Tonooka T, Tsumuraya Y, Inagaki Y, Haruyama N et al. Functional characterization of barley betaglucanless mutants demonstrates a unique role for CslF6 in (1,3;1,4)-β-D-glucan biosynthe- sis. J Exp Bot. 2012; 63:381–392. https://doi.org/10.1093/jxb/err285 PMID: 21940720 16. Nemeth C, Freeman J, Jones HD, Sparks C, Pellny TK, Wilkinson MD et al. Down-regulation of the CslF6 gene results in decreased (1,3;1,4)-β-D-glucan in endosperm of wheat. Plant Physiol. 2010; 152:1209–1218. https://doi.org/10.1104/pp.109.151712 PMID: 20089768 PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 14 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content 17. Islam AKMR, Shepherd KW. Incorporation of barley chromosomes into wheat. In: Bajaj YP, editor. Bio- technology in Agriculture and Forestry. Wheat, Vol. 13. Heidelberg: Springer; 1990. pp. 128–151. 18. Molna´r-La´ng M, Kruppa K, Cseh A, Bucsi J, Linc G. Identification and phenotypic description of new wheat–six-rowed winter barley disomic additions. Genome. 2012; 55:302–311. https://doi.org/10.1139/ G2012-013 PMID: 22439846 19. Tu¨rko¨si E, Cseh A, Darko E, Molna´r-La´ng M. Addition of Manas barley chromosome arms to the hexa- ploid wheat genome. BMC Genetics. 2016; 17:87. Available from: https://bmcgenet.biomedcentral. com/articles/10.1186/s12863-016-0393-2. PMID: 27328706 20. Ivanizs L, Farkas A, Linc G, Molna´r-La´ng M, Molna´r I. Molecular cytogenetic and morphological charac- terization of two wheat-barley translocation lines. PLoS ONE 2018; 13(6): e0198758. Available from: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0198758 PMID: 29889875 21. Ghazali S, Mirzaghaderi G, Majdi M. Production of a novel Robertsonian translocation from Thinopyrum bessarabicum into bread wheat. Cytol Genet. 2005; 49:378–381. 22. Friebe B, Jiang J, Raupp WJ, McIntosh RA, Gill BS. Characterization of wheat–alien translocations con- ferring resistance to diseases and pests: current status. Euphytica. 1996; 91: 59–87. 23. Darko E, Janda T, Majla´th I, Szopko´ D, Dulai S, Molna´r I et al. Salt stress response of wheat–barley addition lines carrying chromosomes from the winter barley “Manas”. Euphytica. 2015; 203:491–504. 24. Cseh A, Soo´s V, Rakszegi M, Tu¨rko¨si E, Bala´zs E, Molna´r-La´ng M. Expression of HvCslF9 and HvCslF6 barley genes in the genetic background of wheat and their influence on the wheat β-glucan content. Ann Appl Biol. 2013; 163:142–150. References 2018; 131:389–406. https://doi.org/10.1007/s00122-017-3009-y PMID: 29101420 39. Lang T, La S, Li B, Yu Z, Chen Q, Li J, Yang E, Li G, Yang Z. Precise identification of wheat–Thino- pyrum intermedium translocation chromosomes carrying resistance to wheat stripe rust in line Z4 and PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 15 / 17 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content its derived progenies. Genome. 2018; 61:177–185. https://doi.org/10.1139/gen-2017-0229 PMID: 29470932 40. Liu W, Jin Y, Rouse M, Friebe B, Gill B, Pumphrey MO. Development and characterization of wheat-Ae. searsii Robertsonian translocations and a recombinant chromosome conferring resistance to stem rust. Theor Appl Genet. 2011; 122:1537–1545. https://doi.org/10.1007/s00122-011-1553-4 PMID: 21347655 41. Liu W, Koo DH, Friebe B, Gill BS. A set of Triticum aestivum-Aegilops speltoides Robertsonian translo- cation lines. Theor Appl Genet. 2016; 129:2359–2368. https://doi.org/10.1007/s00122-016-2774-3 PMID: 27558595 42. Rahmatov M, Rouse MN, Nirmala J, Danilova T, Friebe B, Steffenson BJ, Johansson E. A new 2DS2RL Robertsonian translocation transfers stem rust resistance gene Sr59 into wheat. Theor Appl Genet. 2016; 129:1383–1392. https://doi.org/10.1007/s00122-016-2710-6 PMID: 27025509 43. Zeller FJ, Hsam SLK. Broadening the genetic variability of cultivated wheat by utilizing rye chromatin. In: Sakamoto S. editor. Proceedings of the 6th Intternational Wheat Genetics Symposium. Kyoto University;1983. pp. 161–173. 44. Braun HJ, Payne TS, Morgounov AI, Ginkel Mv, Rajaram S. The challenge: one billion tons of wheat by 2020. In: Slinkard AE. editor. Proceedings of the Ninth International Wheat Genetics Symposium. Uni- versity of Saskatchewan: University Extension Press;1998. pp 33–40. 45. Lukaszewski AJ. Frequency of 1RS/1AL and 1RS/1BL translocations in the United States wheats. Crop Sci. 1990; 30:1151–1153. 46. Schlegel R, Korzun V. About the origin of 1RS1BL wheat-rye chromosome translocations from Ger- many. Plant Breeding.1997; 116:537–540. 47. Koba T, Takumi S, Shimada T. Isolation, identification and characterization of disomic and translocated barley chromosome addition lines of common wheat. Euphytica. 1997; 96:289–296. 48. Danilova TV, Friebe B, Gill BS, Poland J, Jackson E. Development of a complete set of wheat–barley group 7 Robertsonian translocation chromosomes conferring an increased content of β-glucan. Theor Appl Genet. 2018; 131:377–388. https://doi.org/10.1007/s00122-017-3008-z PMID: 29124282 49. Lukaszewski AJ. Manipulation of the genome by chromosome breakage. In: Gill BS, Raupp WJ editors. Proceedings of the US-Japan symposium, classical and molecular cytogenetic analysis. Manhattan: 1994. pp 136–139. 50. Friebe B, Zhang P, Linc G, Gill BS. 60. Wilson SM, Ho YY, Lampugnani ER, Van de Meene AML, Bain MP, Bacic A, Doblina MS. Determining the subcellular location of synthesis and assembly of the cell wall polysaccharide (1,3;1,4)-β-D-Glucan in grasses. Plant Cell. 2015; Available from: www.plantcell.org/cgi/doi/10.1105/tpc.114.135970. 61. Li K, Hegarty J, Zhang C, Wan A, Wu J, Guedira GB et al. Fine mapping of barley locus Rps6 conferring resistance to wheat stripe rust. Theor Appl Genet. 2016; 129:845–859. https://doi.org/10.1007/s00122- 015-2663-1 PMID: 26875072 7BS.7HL wheat-barley Robertsonian translocation with increased salt tolerance and (1,3;1,4)-β-D-glucan content PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 References Robertsonian translocations in wheat arise by centric misdivision of univalents at anaphase I and rejoining of broken centromeres during interkinesis of meiosis II. Cyto- genet Genome Res. 2005; 109:293–297. https://doi.org/10.1159/000082412 PMID: 15753589 51. King IP, Forster BP, Law CC, Cant KA, Orford SE, Gorham J et al. Introgression of salt-tolerance genes from Thinopyrum bessarabicum into wheat. New Phytol. 1997; 137:75–81. 52. Darko E, Gierczik K, Huda´k O, Forgo´ P, Pa´l M, Tu¨rko¨si E et al. Differing metabolic responses to salt stress in wheat-barley addition lines containing different 7H chromosomal fragments. PLoS ONE. 2017; 123:20. Available from: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0174170. 53. Negrão S, Schmo¨ckel SM, Tester M. Evaluating physiological responses of plants to salinity stress. Ann Bot. 2017; 119:1–11. https://doi.org/10.1093/aob/mcw191 PMID: 27707746 54. Diab AA, Teulat-Merah B, This D, Ozturk NZ, Benscher D, Sorrells ME. Identification of drought-induc- ible genes and differentially expressed sequence tags in barley. Theor Appl Genet. 2004; 109:1417– 1425. https://doi.org/10.1007/s00122-004-1755-0 PMID: 15517148 55. Sa´nchez de la Hoz P, Vicente-Carbajosa J, Mena M, Carbonero P. Homologous sucrose synthase genes in barley (Hordeum vulgare) are located in chromosomes 7H (syn 1) and 2H. Evidence for a gene translocation? Febs Lett. 1992; 310:46–50. PMID: 1388123 56. Kleines M, Elster RC, Rodrigo MJ, Blervacq AS, Salamini F, Bartels D. Isolation and expression analy- sis of two stress-responsive sucrose-synthase genes from the resurrection plant Craterostigma planta- gineum (Hochst.). Planta. 1999; 209:13–24. https://doi.org/10.1007/s004250050602 PMID: 10467027 57. Islamovic E, Obert DE, Oliver RE, Harrison SA, Ibrahim A, Marshall JM et al. Genetic dissection of grain beta-glucan and amylose content in barley (Hordeum vulgare L.). Mol. Breeding. 2013; 31:15–25. 58. Shu X, Rasmussen SK. Quantification of amylose, amylopectin, and β-glucan in search for genes con- trolling the three major quality traits in barley by genome-wide association studies. Front. Plant Sci. 2014; 5: 197. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4030205/. https://doi.org/ 10.3389/fpls.2014.00197 PMID: 24860587 59. Houston K, Russell J, Schreiber M, Halpin C, Oakey H, Washington JM et al. A genome wide associa- tion scan for (1,3;1,4)-β-glucan content in the grain of contemporary 2-row spring and winter barleys. BMC Genomics. 2014; 15(1):907. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC4213503. PLOS ONE \| https://doi.org/10.1371/journal.pone.0206248 November 5, 2018 16 / 17 17 / 17
https://openalex.org/W4386754452	https://www.iieta.org/download/file/fid/106775	English	null	Influence of Cattle Grazing Methods on Changes in Vegetation Cover and Productivity of Pasture Lands in the Semi-Desert Zone of Western Kazakhstan	International journal of design & nature and ecodynamics	2,023	cc-by	6,799	Beybit Nasiyev1* , Amanbay Karynbayev2 , Madiyar Khiyasov1 , Askhat Bekkaliyev1 , Nurbolat Zhanatalapov1 , Mira Begeyeva1 , Aidyn Bekkaliyeva1 , Vladimir Shibaikin3 1 Zhangir khan West Kazakhstan Agrarian-Technical University, Uralsk 090000, Republic of Kazakhstan 2 International Taraz Innovative Institute Named after Sh. Murtaza, Taraz 080020, Republic of Kazakhstan 3 Vavilov Saratov State Agrarian University, Saratov 410012, Russia https://doi.org/10.18280/ijdne.180402 ABSTRACT Monitoring pasture vegetation indicators is critical as decreased productivity could jeopardize the stability of pasture lands. This study aimed to evaluate the status of vegetation cover across pastures utilized in diverse manners within the semi-desert zone of Western Kazakhstan. Specifically, the impacts of rotational, seasonal, and intensive grazing on the functionality, diversity, and productive potential of pasture vegetation were investigated. Systematic vegetation observations, including species diversity examination, projective coverage estimation, height measurements, and yield determination, were conducted in transects established across pastures with varying grazing methods at the Miras farm in Western Kazakhstan. The findings reveal that unregulated intensive grazing significantly altered vegetation indicators, suggesting potential degradation processes. Intensive grazing proved particularly detrimental, resulting in decreased species richness, projective coverage, height, and a lower productive vegetative mass. In contrast, rotational and seasonal grazing methods appeared more effective in Western Kazakhstan. Pastures where regulated grazing was practiced exhibited higher species richness and biometric and production indicators compared to those with unsystematic grazing. Understanding the alterations in biometric and productive vegetation indicators relative to grazing practices is crucial for quality assessment of pastures and determining appropriate pasture management. The absence of adequate and efficient grazing could trigger irreversible deterioration in pasture vegetation conditions. Therefore, to safeguard the biological resources and biodiversity of the region's pastures, it is strongly advocated to employ regulated grazing and entirely eliminate excessive intensive grazing. This approach will help maintain a healthy balance in the pasture ecosystem and promote sustainable farming practices. Received: 2 April 2023 Revised: 21 July 2023 Accepted: 8 August 2023 Available online: 31 August 2023 Keywords: pastures, grazing methods, vegetation condition, yield, feed value, species richness, projective cover, pasture yield, semi-arid grasslands 1. INTRODUCTION Additionally, overgrazing over extended periods alters the composition of dominant species [13, 14], the distribution of forage and poisonous species on pastures [15, 16], and negatively impacts vegetation productivity [17], greenhouse gas balance, and biodiversity [18-20]. In most countries, including Kazakhstan, due to climatic conditions and traditional systems of free (excessive) grazing, strong destruction of vegetation is observed [21-23]. The terrain of the farm is flat. The Miras farm is located on the Caspian lowland. The climate is continental, winters are frosty, and summers are moderately hot. Average temperatures in January reach 12-14℃ and in July 24-25℃. The average annual precipitation is 250-300mm. The Aralsor lake is located next to the farm. The soils are light chestnut and sandy. In research conducted in Extremadura, Spain, keeping animals at rates exceeding 1 astronomical unit (a.u.)/ha-1 showed signs of reduced productivity and pasture quality [24]. Schuman et al. [25] found that intensive grazing (IG) led to a decrease in the maximum yield of aboveground live phytomass, with plots becoming dominated by weeds. Grazing alters the reproduction, productivity, and composition of plant species [26], as well as the quantity and composition of root secretions, rhizosphere activity, and litter decomposition [27]. Experimental work was carried out according to the scheme indicated in Table 1. Table 1. Design of the field experiment on the study of methods of grazing farm animals on pastures of the semi- desert zone of western Kazakhstan Table 1. Design of the field experiment on the study of methods of grazing farm animals on pastures of the semi- desert zone of western Kazakhstan Table 1. Design of the field experiment on the study of methods of grazing farm animals on pastures of the semi- desert zone of western Kazakhstan Grazing Method Variants Pasture Area, Type, and Number of Livestock No grazing (NG) (control) NG RG 560 ha, cattle in the amount of 80 heads Seasonal grazing (SG) 560 ha, cattle in the amount of 80 heads IG 560 ha, cattle in the amount of 80 heads Notably, long-term exclusion of cattle grazing also negatively impacts grasslands, leading to an increase and spread of poisonous species and shrubs in the herbage [28]. Therefore, defining an effective grazing method is crucial for sustainable pasture management and efficient land use [29]. 2.2 Study of the vegetation cover of pastures To study the plant community on pasture transects measuring 100×50m were established, where all regime observations were carried out: the study of species composition, projective coverage, height, and yield of pasture grass. 1. INTRODUCTION The economic benefits of cattle grazing are primarily linked to a reduction in feed costs, making it essential to have a promising pasture management system for an almost continuous supply of high-quality pasture feed [30]. In the semi-desert zone of Western Kazakhstan, cattle grazing is a leading anthropogenic influence on the vegetation cover of pasture lands. Our study centers on effective grazing technologies that prevent the degradation of new territories. The research question involves evaluating the conditions and changes in the main parameters of vegetation cover, specifically the projective cover, height, and yield of pasture grass under the influence of grazing. Upon obtaining data from the chemical analysis of pasture grass quality, the productivity and energy, and protein value of pastures were assessed, depending on their usage methods. Experiment variants: RG, SG, and IG were studied on the territories of pasture lands of the Miras farm. The pasture plots were located next to each other within the same pasture land. In the RG variant, pastures are used by the rotation method, i.e., in the 1st year the pasture is used only in spring, in the 2nd year in summer, in the 3rd year in autumn, then in the 4th year the pasture rests. In the SG variant, pastures are used for grazing cattle only in a certain season without rest. In the IG variant, pastures are used in all seasons of the year without rest. This study aimed to: 1) Assess the impact of various cattle grazing methods on the vegetation cover and productivity of Western Kazakhstan's semi-desert pasture lands, 2) Ascertain the methods that foster greater species diversity and vegetative cover, and 3) Inform pasture management recommendations to promote productive and sustainable pasture use. The control variant implies that the untouched plot is located outside the farm territories. This plot is selected in a semi-desert zone as a reference for comparing pastures. 1. INTRODUCTION Managing the productivity and rational use of pastures necessitates an optimal grazing system. This includes parameters such as grazing methods, timing of grazing initiation and termination, livestock numbers and density, as well as the frequency and duration of individual grazing periods [6-8]. However, the impact of grazing methods on pasture quality remains a controversial topic. Some studies advocate the benefits of rotational grazing (RG) for sustaining pasture productivity and land condition [9], while others, such as the work by McDonald et al. [10], suggest that strategic rest in comparison to continuous livestock grazing could enhance vegetation condition and biodiversity, and maximize feed production per unit area. Cattle grazing is globally recognized as a primary contributor to the degradation of pasture lands, homogenization of spatial landscapes, functional alteration of pasture lands, and loss of species diversity [1]. The escalating demand for environmentally friendly food has amplified livestock numbers and expanded agricultural activities, even in challenging environmental conditions such as the semi-desert zone of Western Kazakhstan [2]. Thus, understanding the impact of animal husbandry on grassland vegetation quality becomes imperative for making informed management decisions [3]. Overgrazing, as Atamov et al. [11] noted, could lead to the transformation of the primary steppe into the secondary steppe, eventually causing the semi-desert to change into a desert vegetation zone, a particularly relevant issue for Western Kazakhstan [12]. Overgrazing is a significant concern as it stresses valuable pasture species, reduces the feed, energy, and protein value of pasture lands, and long-term productivity loss and vegetation cover reduction can cause community social losses due to malnutrition and poverty [11- Pasture lands, often found in plant biomes including meadows, shrubs, savannas, and deserts, are expansive natural vegetation areas supporting animal husbandry [4]. These regions are typically characterized by arid climates, low rainfall, and marked seasonal temperature fluctuations. The quality of these pasture lands is closely tied to animal productivity, rendering the maintenance of their sustainable forage resources without quality degradation as a critical concern [5]. 767 in the production of meat by raising cattle are mainly concentrated. The farm we have chosen is typical of a semi- desert zone with vegetation mainly consisting of Artemisia Lerchiana, where the productivity of pasture land strongly depends on management techniques, in particular on the methods of grazing livestock. 13]. 2.1 Description of the study plots The final indicator of accounting is the average height of plants on the plot [34]. Yield accounting was carried out by the cutting method. On pastures during cattle grazing, crop accounting was carried out when the pasture ripeness of the herbage was reached. To do this, along the diagonal of the plot, the grass was mowed with a hand sickle (Prompribor LLC, Yekaterinburg, Russia) on four accounting platforms of 2.5m2 each, the height of the cut reaching 4 to 5cm. The mowed mass was weighed and an average sample of 1kg was taken. The sample was dried under a canopy to a humidity of 17 to 18% to determine the yield of air-dry matter [34]. In the area of SG, where pastures were used annually only in the summer season with rest in the spring and autumn seasons, with 75% of the total projective coverage, cereals account for 44% of the total phytocenosis. In this area, wormwood occupies 13%, and the specific gravity of various types of grass was 14%. In the composition of various types of grass, 4% is used by weeds and poisonous plants. As the study data show, the worst qualitative composition of phytocenosis by economic and botanical groups of pasture plants is formed during the unsystematic grazing of farm animals. Thus, in an IG area, where pastures were used annually without rest in all seasons, i.e., haphazardly, the total projective coverage of pastures does not exceed 45%. As part of the phytocenosis, the specific weight of the most valuable forage plants is maximally reduced. The share of cereals in the total phytocenosis was at the level of 9%. As a result of trampling by cattle, an increase in the proportion of wormwood Artemisia lerchiana and Artemisia austriaca in the vegetation cover was noted to be 15%. To assess the quality of pasture vegetation, the content of crude nitrogen, crude fat, crude fiber, crude ash, carotene, Ca, P, and K. The content of crude nitrogen in vegetation was determined based on the Kjeldahl titrimetric method using the Kjeldahl installation (Labreaktiv LLP, Ust-Kamenogorsk, Kazakhstan) [35]. The content of crude fat was determined by extraction of crude fat with diethyl ether using a Vilitek ASV-6M Soxhlet apparatus (VILYTEK LLC, Moscow, Russia) [35]. The content of crude fiber was determined according to Genneberg and Shtoman using a Vilitek ASV-6M Soxhlet apparatus (VILYTEK LLC, Moscow, Russia). 2.1 Description of the study plots The quantitative ratio of species in pastures was characterized using the Drude scale: Soc (socialis): “abundant”, plants grow everywhere, connecting with their aboveground parts; Cop. 3 (copiosus): “copious”, plants are found in very large numbers; Cop. 2: “a lot”, plants are found in large numbers; Cop. 1: “many”, plants are found in considerable numbers; Sp (sparsus): “sparse”, the species is abundant, but does not form a continuous cover; Sol (solitarius): “single”, the species grows scattered; Un (unicum): the species occur in single instances [33]. The study was conducted in 2018-2022 at the Zhangir Khan West Kazakhstan Agrarian and Technical University (ZKATU, Republic of Kazakhstan) on the initiative of the Ministry of Agriculture on the pastures of the Miras farm in the semi-desert zone of Western Kazakhstan. Agrochemical analyses of soil samples were carried out in the accredited laboratory of the ZKATU. Currently, in the Republic of Kazakhstan, out of 188.9 million ha of pastures, 26.6 million ha have reached the state of extreme degradation (run-down) and as a result of pasture degradation, the loss of feed equals 4.3 million tons of fodder units [12, 31, 32]. The indicator projective coverage is the ratio of the area of the projection of vegetation cover to the area on which they are projected, expressed as a percentage. The projective coating was evaluated visually [34]. In the Western Kazakhstan region, the area of pasture lands is 10.2 million ha, while more than 65% of pastures are located in the semi-desert zone of the region, where farms specializing The height of the plants was measured before grazing using a measuring ruler. The end of the ruler was placed on the 768 composition of pastures was noted in the pastures of RG. Thus, in this area of pasture, with total projective coverage of 85%, the most valuable cereal plants, such as Agropyron desertorum, Stipa capillata, Festuca valesiaca, Leymus ramosus, Koeleria cristata, and Kochia prostrata account for 55% of the phytomass. In this area, wormwood occupies 12%, and mixed grasses with weeds occupy 16%. In the pasture phytocenosis, the specific weight of weeds and poisonous plants was 2%. surface of the soil. The sample size was 10 plants selected in different places along the diagonal of the accounting area. In this case, the stem is measured from the soil surface to the tip. 2.3 Data analysis Figure 1. Dynamics of the total projective cover of herbage and the projective cover of economic and botanical groups of pasture plants depending on grazing methods for 2018- 2021, %. No grazing: NG, Rotational grazing: RG, Seasonal grazing: SG, Intensive grazing: IG The productivity indicators of the plant community were statistically processed by single-factor analysis of variance [34]. The intergroup variance for three variants of the experiment (NG, SG, RG) was compared with statistical significance at the p-level<0.01 using the SAS OnDemand for Academics software. The chart of indicators of pasture lands by grazing variants was built using the MS Excel tabular editor. In comparison with other pasture areas, the specific gravity of various types of grass was minimal, at the level of 12%. It should be noted that haphazard grazing has affected the qualitative composition of the grass. In this grazing variant, weeds and poisonous plants Tanacetum achilleifolium, Lipidium perfoliatum, Anabasis aphylla, Datura, Xanthium strumarium, Alhagi pseudalhagi, and Euphorbia occupied up to 9% of the total grass. 2.1 Description of the study plots The content of raw ash was determined by dry ashing- burning of a sample of air-dry feed in a PM-8 muffle furnace (Stankopark LLP, Almaty, Kazakhstan) at a temperature of 500-525℃ [35]. Figure 1. Dynamics of the total projective cover of herbage and the projective cover of economic and botanical groups of pasture plants depending on grazing methods for 2018- 2021, %. No grazing: NG, Rotational grazing: RG, Seasonal grazing: SG, Intensive grazing: IG The content of carotene in the plant sample was determined by extraction with petroleum ether using a PE-5400VI spectrophotometer (Ekroskhim, St. Petersburg, Russia) [35]. The Ca content was determined based on the trilonometric method using an atomic absorption spectrophotometer (Iskroline LLP, St. Petersburg, Russia). The P content was determined based on the photometric method using a PE- 5400VI spectrophotometer (Ekroskhim, St. Petersburg, Russia). The K content was determined based on the flame photometric method using a Jenway PFP 7 flame photometer (Jenway, UK) [35]. Based on the obtained quality indicators, the fodder and energy-protein value of pasture vegetation was calculated depending on the grazing methods using the methodological recommendation [35]. 3.2 Influence of cattle grazing methods on indicators of vegetation cover According to the indicators of the collection of feed units, digestible protein, the productivity of pasture grass was high at the control plot, i.e., in the absence of grazing, reaching 0.28 and 0.032 t/ha, and when using pastures according to the methods of SG and RG, respectively, with 0.13-0.18 and 0.013-0.019 t/ha. Quantitative and qualitative indicators of pasture phytocenoses also depended on the way pastures were used. At the same time, the projective coverage of pastures, depending on the method of their use, ranged from 50 to 90% (Table 2). Table 2. Indicators of vegetation cover of pastures in the semi-desert zone of Western Kazakhstan, depending on grazing methods, the average for 2018-2022 Table 2. Indicators of vegetation cover of pastures in the semi-desert zone of Western Kazakhstan, depending on grazing methods, the average for 2018-2022 The yield of feed units and digestible protein from 1 ha was lower compared to the above-mentioned variants of SG on the variant of unsystematic grazing (0.06 and 0.003 t/ha). In block B of Figure 2, the indicator of the yield of feed units shows a tendency to decrease values from the maximum control values to the minimum values in the IG variant. Visually, one can note the difference between the average values and the experiment variant. The size of the blocks in the diagram shows the variation. All indicators for the experiment variants have an average variation. The intra-group deviation of the indicator values is average and the same is true for all variants of the experiment. Comparing the intergroup variance for three variants of the experiment (NG, SG, RG) using a one- factor analysis of variance, we determined that the average values of the trait for the variants of the experiment differed with statistical significance at the p-level<0.01. Consequently, the yield of feed units depends on the intensity of pasture use. Grazing Method Variants Projective Coverage, % Number of Species Height of the Herbage, cm NG (control) 95 22 60 RG 90 16 44 SG 80 15 35 IG 50 12 27 The largest projective coverage was determined in the control area (95%), and the RG pastures (90%). The smallest projective cover (50%) was observed on pastures with unsystematic farm animal grazing. On seasonally used pastures, the projective coverage was at the level of 80%. 3.1 Species composition and diversity The dynamics of the overall projective coverage of economic and botanical groups of plants on pastures depend on the ways they are used. In our studies, the most qualitative In studies on the reference plot with untouched vegetation and soil cover, located in a semi-desert zone outside the 769 The yield indicator (A) of Figure 2 shows the change in average values depending on the experiment variants. There is a tendency to decrease values from the maximum control values to the minimum values in the IG variant. All indicators for the experiment variants are stable. The intra-group deviation of the indicator values is slight. The smallest variation of the trait is observed for the SG experiment. Comparing the intergroup variance for three variants of the experiment (NG, SG, RG) using a one-factor analysis of variance, it turns out that the average values of the trait for the variants of the experiment differ with statistical significance at the p-level<0.01. Consequently, the yield indicator depends on the intensity of pasture use. territories of the farm, the total projective coverage was 95%. At this control plot, the share of valuable species in the feed ratio was 65%, and the specific weight of the wormwood group was minimal, at the level of 10%. In the rest plot, the proportion of various types of grass was also maximal and reached up to 19% of the total vegetation of the reference plot. Weeds and poisonous plants in the composition of various types of grass occupied only 1% of the pasture cenosis (Figure 1). 3.2 Influence of cattle grazing methods on indicators of vegetation cover The present study showed that the vegetation cover decreased with an increase in the intensity of grazing from 95 to 50%. Collection of digestible protein (Figure 2, block C). Visually, one can note the difference between the average values and the experiment variant. The size of the blocks in the diagram shows a slight variation. All indicators for the experiment variants have the same variation. Using a one- factor analysis of variance, we determined that the average values of the trait for the variants of the experiment differed with statistical significance at the p-level<0.01. In studies based on the results of botanical analysis, the largest number of plant species (22) was established in the control area with SG. The height of the herbage at the control was 60cm. On pastures, the number of pasture plant species depended on the way they were used. At the same time, the largest number of species (15-16) were identified on pastures of SG and RG. The height of pasture grass with SG and RG variants was 35-44cm. The output of the exchange energy in the experimental variants was at the level of 0.90-3.51 GJ/ha. The highest output of exchange energy was observed at the control plot, equaling 3.51 GJ/ha. In terms of energy value, the use of SG and RG pastures on chestnut sandy soils of the semi-desert zone has an advantageous position. On these pastures, the collection of exchange energy exceeds the variants for intensive use of pastures by 0.86-1.52 GJ/ha. With haphazard use of pastures, the collection of exchange energy was at a level of 0.90 GJ/ha. In studies, the herbage with the lowest growth of 27cm was recorded on pastures with unsystematic grazing. As a result of excessive load on pastures of unsystematic grazing, the species diversity of pasture plants decreases to 12, and degraded areas are formed, which is especially evident in the summer. 3.3 Nutritional quality and pasture productivity The indicator of the output of the exchange energy (Figure 2, block D) is stable for the IG variant of the experiment and has the smallest variation. The indicator of the average value “Output of exchange energy” decreases depending on the intensity of grazing. The average values of the indicator vary from the maximum values (in the control variant) to the minimum values in the IG variant. Differences in the average values of the trait in the experiment variants differ with statistical significance at the p-level<0.01. Consequently, the indicator of the output of exchange energy depends on the intensity of pasture use. According to research data, the productivity of pasture grass depends on the methods of farm animal grazing. In the 2018- 2022 studies, when using SG and RG, the yield of the dry mass of pasture grass was 0.68-0.87 t/ha. While at the control plot, the yield of the dry mass of pasture grass was 1.29 t/ha, with an increase in the load due to unsystematic grazing, the productivity of pasture cenosis decreased to the level of 0.40 t/ha of dry mass (Figure 2, block A) 770 Figure 2. Indicators of productivity, energy, and protein value of pastures, depending on the methods of their use the yield of green mass, t/ha; В: the output of feed units, t/ha; С: collection of digestible protein, t/ha; D: the output of the exchange energy, GJ/h; NG: no grazing (control); RG: rotational grazing; SG: seasonal grazing; IG: intensive grazing Figure 2. Indicators of productivity, energy, and protein value of pastures, depending on the methods of their use А: the yield of green mass, t/ha; В: the output of feed units, t/ha; С: collection of digestible protein, t/ha; D: the output of the exchange energy, GJ/h; NG no grazing (control); RG: rotational grazing; SG: seasonal grazing; IG: intensive grazing The conducted one-factor analysis of variance showed the statistical significance of differences in the average values of the indicator depending on the methods of pasture use. The most severe use case is IG (unsystematic grazing). indicators of vegetation were observed in studies on the reference plot with untouched vegetation. 4.2 Influence of IG on territory degradation The obtained results also confirm previous reports that the vegetation cover on pastures strongly depends on the intensity of grazing [41-43]. An increase in livestock grazing, as a rule, leads to an increase in plant mortality and, ultimately, to a decrease in species richness, especially in conditions of lack of water and nutrients [44-46]. In the case of IG, we have established the worst vegetation indicators, where the projective cover equals 50%, the number of species is 12, with a grass stand height of 27cm. A significant decrease (P<0.05) in grass height with an increase in degradation intensity corresponds to the data of Haider et al. [39] and Li et al. [45]. 3.3 Nutritional quality and pasture productivity In our study, the use of SG and RG methods for the livestock as a result of careful treatment and protection of pasture areas contributed to an increase in overall productivity and by increasing the most valuable species in the feed ratio in the total grassland, improving the quality and energy and protein value of pasture feed. Thus, for all three indicators: A-Yield of green mass, t/ha (0.40); B-Yield of feed units, t/ha (0.06); C-Collection of digestible protein, t/ha (0.003), the values for unsystematic grazing are the minimum of all four variants of the experiment (IG, NG, RG, SG). The difference in the average values of the indicators of the IG unsystematic grazing option compared to the control NG value is maximal: А-Yield of green mass, t/ha (0.89); В-Yield of feed units, t/ha (0.19); С-Collection of digestible protein, t/ha (0.029). The most optimal in terms of values of indicators is the use of pastures in the RG variant: А- Yield of green mass, t/ha (0.87); В-Yield of feed units, t/ha (0.18); С-Collection of digestible protein, t/ha (0.019). The combined effects of vegetation loss and pasture degradation affect primary production and, ultimately, animal productivity [40]. 4.1 Impact of lean grazing practices on study areas Our study confirmed the researchers’ assumption that moderate RG of livestock can be used as a useful management method to maintain species diversity and productivity of pasture lands [36]. Moreover, a low or moderate grazing level can increase production compared to SG [37]. In our studies, the most optimal species diversity, biometric indicators (projective coverage and height), and productivity of pasture vegetation were established on the variants of RG and SG, where grazing is carried out in a moderate mode. As noted by Hickman and Hartnett [40], overgrazing also reduces pasture productivity. In the studies, the lowest indicators of productivity and energy-protein value were found precisely on pastures of IG. Other studies show that although overgrazing can lead to pasture degradation [47], moderate grazing can promote plant growth and increase species diversity [40]. This is confirmed by the data of studies conducted in the semi-desert zone of Western Kazakhstan. Pasture rest significantly improves phytocenosis biodiversity [38, 39]. The best biometric and productive 771 However, the pastures of the semi-desert zone of Western Kazakhstan are not productive enough and they need proper protection, management, and rehabilitation using ecological and adaptive approaches. As the practice of pasture management shows, intensive grazing is increasingly used in Kazakhstan and is associated with an increase in the number of livestock due to an increase in demand for livestock products, as well as a shortage of pasture land [48]. In conclusion, such studies have the prospect of continuing in the area of interest, and the results can be used in countries with similar natural and climatic conditions and pasture ecosystem management strategies. Further research could be directed toward effective planning and management in arid and semi-arid grasslands. REFERENCES [1] Cutter, J., Hovick, T., McGranahan, D., Harmon, J., Limb, R., Spiess, J., Geaumont, B. (2022). Cattle grazing results in greater floral resources and pollinators than sheep grazing in low‐diversity grasslands. Ecology and Evolution, 12(1): e8396. https://doi.org/10.1002/ece3.8396 Our studies have established clear signs of the spread of weed and poisonous plants Tanacetum achilleifolium, Lipidium perfoliatum, Anabasis aphylla, Datura, Xanthium strumarium, Alhagi pseudalhagi, and Euphorbia. In this case, according to Chadaeva et al. [54], pastures need urgent restoration. Therefore, this study contributes to the understanding of the ecological impact of cattle grazing on the vegetation characteristics of pasture lands of the semi-desert zone of Western Kazakhstan, which have important theoretical and practical significance on a national scale. [2] Nasiyevich, N.B. (2013). The role of organic fertilizers in inceasing the fertility of West Kazakhstan soil. Polish Journal of Soil Science, 46(2): 115-146. https://doi.org/10.17951/pjss.2013.46.2.115 [3] Karaca, S., Dengiz, O., Turan, İ.D., Özkan, B., Dedeoğlu, M., Gülser, F., Sargin, B., Demirkaya, S., Ay, A. (2021). An assessment of pasture soils quality based on multi- indicator weighting approaches in semi-arid ecosystem. Ecological Indicators, 121: 107001. https://doi.org/10.1016/j.ecolind.2020.107001 Further unsystematic use of pasture resources poses a threat to the well-being of animal husbandry and destabilizes the habitat of the population and leads to environmental consequences due to the disappearance of valuable species of pasture plants and the increase in degradation and desertification of pasture lands. [4] Abdullah, M., Rafay, M., Sial, N., Rasheed, F., Nawaz, M.F., Nouman, W., Ahmad, I., Ruby, T., Khalil, S. (2017). Determination of forage productivity, carrying capacity and palatability of browse vegetation in arid rangelands of Cholistan desert (Pakistan). Applied Ecology & Environmental Research, 15(4): 623-637. https://doi.org/10.15666/aeer/1504_623637 Since the focus of this study was on grazing practices, we did not consider additional factors (technological factors: the use of different types of equipment, the influence of weather conditions, etc.) that other studies may take into account when interpreting the results. [5] Rubanza, C.D.K., Shem, M.N., Ichinohe, T., Fujihara, T. (2006). Biomass production and nutritive potential of conserved forages in silvopastoral traditional fodder banks (Ngitiri) of Meatu District of Tanzania. Asian- Australasian Journal of Animal Sciences, 19(7): 978-983. https://doi.org/10.5713/ajas.2006.978 ACKNOWLEDGMENT The work was carried out in the Zhangir khan West Kazakhstan Agrarian-Technology University within the framework of program-targeted financing of the Ministry of Agriculture of the Republic of Kazakhstan on the topic BR 10764915 “Development of new technologies for restoration and rational use of pastures (use of pasture resources)”. Muller et al. [52] noted that many acceptable and productive species, which are indicators of undisturbed pastures, have been greatly reduced as a result of long-term grazing, and many grazing-resistant species, which are indicators of disturbance, have spread over transformed pastures. Patra et al. [53] determined that the IG of livestock led to changes in the composition and structure of plant communities of pastures compared with light grazing of livestock. 4.3 Influence of grazing methods on the composition of herbage and ecological consequences for the study areas The data obtained on the species composition of grass stands are consistent with the conclusions of scientists that overgrazing contributes to the rapid penetration of less desirable invasive species, mainly species of shrubby plants [49-51]. As a result of IG, an increase in the specific weight in the herbage of invasive species of shrub plants such as Artemisia lerchiana and Artemisia austriaca was noted. 5. CONCLUSIONS https://doi.org/10.1111/1365-2664.13496 [23] Kamali, N., Sadeghipour, A., Souri, M., Mastinu, A. (2022). Variations in soil biological and biochemical indicators under different grazing intensities and seasonal changes. Land, 11(9): 1537. https://doi.org/10.3390/land11091537 [11] Atamov, V., Çetin, E.D., Cevheri, C. (2016). Seasonal and ecological aspects of plain pastures of upper mesopotamia from Sanliurfa (Turkey). Oxidation Communications, 39(1): 578-598. [24] Pulido, M., Schnabel, S., Lavado Contador, J.F., Lozano‐ Parra, J., Gonzalez, F. (2018). The impact of heavy grazing on soil quality and pasture production in rangelands of SW Spain. Land Degradation & Development, 29(2): 219-230. https://doi.org/10.1002/ldr.2501 [12] Nasiyev, B., Gabdulov, M., Zhanatalapov, N., Makanova, G., Izbasova, G. (2015). Study of the phenology, abundance and harmfulness of locusts in the semi-desert zone and the organization of locust control measures. Biosciences Biotechnology Research Asia, 12(2): 1759- 1766. https://doi.org/10.13005/bbra/1840 [13] Olofsson, J., Stark, S., Oksanen, L. (2004). Reindeer influence on ecosystem processes in the tundra. Oikos, 105(2): 386-396. https://doi.org/10.1111/j.0030- 1299.2004.13048.x [25] Schuman, G.E., Reeder, J.D., Manley, J.T., Hart, R.H., Manley, W.A. (1999). Impact of grazing management on the carbon and nitrogen balance of a mixed‐grass rangeland. Ecological Applications, 9(1): 65-71. https://doi.org/10.1890/1051- 0761(1999)009[0065:IOGMOT]2.0.CO;2 [14] Stark, S., Julkunen-Tiitto, R., Kumpula, J. (2007). Ecological role of reindeer summer browsing in the mountain birch (betula pubescens ssp. czerepanovii) forests: effects on plant defense, litter decomposition, and soil nutrient cycling. Oecologia, 151: 486-498. https://doi.org/10.1007/s00442-006-0593-y [26] Deng, L., Shangguan, Z.P., Wu, G.L., Chang, X.F. (2017). Effects of grazing exclusion on carbon sequestration in China’s grassland. Earth-Science Reviews, 173: 84-95. https://doi.org/10.1016/j.earscirev.2017.08.008 [15] Leu, S., Mussery, A.M., Budovsky, A. (2014). The effects of long time conservation of heavily grazed shrubland: A case study in the Northern Negev, Israel. Environmental Management, 54(2): 309-319. https://doi.org/10.1007/s00267-014-0286-y [27] de Vries, F.T., Williams, A., Stringer, F., Willcocks, R., McEwing, R., Langridge, H., Straathof, A.L. (2019). Changes in root‐exudate‐induced respiration reveal a novel mechanism through which drought affects ecosystem carbon cycling. New Phytologist, 224(1): 132-145. https://doi.org/10.1111/nph.16001 [16] Zhang, B., Beck, R., Pan, Q.M., Zhao, M.L., Hao, X.Y. (2019). Soil physical and chemical properties in response to long-term cattle grazing on sloped rough fescue grassland in the foothills of the Rocky Mountains, Alberta. Geoderma, 346: 75-83. https://doi.org/10.1016/j.geoderma.2019.03.029 [28] Navarro, L.M., Pereira, H.M. (2012). Rewilding abandoned landscapes in Europe. Ecosystems, 15: 900- 912. https://doi.org/10.1007/s10021-012-9558-7 [29] Zhang, J.P., Zhang, L.B., Liu, W.L., Qi, Y., Wo, X. (2014). Livestock-carrying capacity and overgrazing status of alpine grassland in the Three-River Headwaters region, China. 5. CONCLUSIONS Based on the results of the conducted study, we achieved the goal of assessing the conditions of the vegetation cover of pastures in the semi-desert zone of Western Kazakhstan, depending on the methods of their use. [6] Bork, E.W., Döbert, T.F., Grenke, J.S., Carlyle, C.N., Cahill Jr, J.F., Boyce, M.S. (2021). Comparative pasture management on Canadian cattle ranches with and without adaptive multipaddock grazing. Rangeland Ecology & Management, 78: 5-14. https://doi.org/10.1016/j.rama.2021.04.010 In addition, valuable theoretical data on the effects of various methods of cattle grazing on biometric and productive indicators of vegetation of pasture ecosystems of the semi- desert zone have been obtained. In practical terms, the use of the study results by introducing and implementing SG and RG methods in pasture management strategies will help relieve stress from some species and increase the productivity and feed value of pastures, as well as prevent degradation processes in pasture ecosystems. [7] Hunt, L.P., McIvor, J.G., Grice, A.C., Bray, S.G. (2014). Principles and guidelines for managing cattle grazing in the grazing lands of northern Australia: Stocking rates, pasture resting, prescribed fire, paddock size and water points-a review. The Rangeland Journal, 36(2): 105-119. 772 https://doi.org/10.1071/RJ13070 [20] Zhumanova, M., Mönnig, C., Hergarten, C., Darr, D., Wrage-Mönnig, N. (2018). Assessment of vegetation degradation in mountainous pastures of the Western Tien-Shan, Kyrgyzstan, using eMODIS NDVI. Ecological Indicators, 95: 527-543. https://doi.org/10.1016/j.ecolind.2018.07.060 [8] [8] Roche, J.R., Berry, D.P., Bryant, A.M., Burke, C.R., Butler, S.T., Dillon, P.G., Donaghy, D.J., Horan, B., Macdonald, K.A., Macmillan, K.L. (2017). A 100-year review: A century of change in temperate grazing dairy systems. Journal of Dairy Science, 100(12): 10189- 10233. https://doi.org/10.3168/jds.2017-13182 [21] Aghajanlou, F., Mirdavoudi, H., Shojaee, M., Mac Sweeney, E., Mastinu, A., Moradi, P. (2021). Rangeland management and ecological adaptation analysis model for astragalus curvirostris Boiss. Horticulturae, 7(4): 67. https://doi.org/10.3390/horticulturae7040067 [9] Briske, D.D., Derner, J.D., Brown, J.R., Fuhlendorf, S.D., Teague, W.R., Havstad, K.M., Gillen, R.L., Willms, W.D. (2008). Rotational grazing on rangelands: reconciliation of perception and experimental evidence. Rangeland Ecology & Management, 61(1): 3-17. https://doi.org/10.2111/06-159R.1 [22] Moradi, P., Aghajanloo, F., Moosavi, A., Monfared, H.H., Khalafi, J., Taghiloo, M., Khoshzaman, T., Shojaee, M., Mastinu, A. (2021). Anthropic effects on the biodiversity of the habitats of ferula gummosa. Sustainability, 13(14): 7874. https://doi.org/10.3390/su13147874 [10] McDonald, S.E., Lawrence, R., Kendall, L., Rader, R. (2019). Ecological, biophysical and production effects of incorporating rest into grazing regimes: a global meta‐ analysis. Journal of Applied Ecology, 56(12): 2723-2731. 5. CONCLUSIONS Journal of Arid Land, 6: 324-333. https://doi.org/10.1007/s40333-013-0202-y g j [48] Nasiyev, B., Bekkaliyev, A. (2019). The impact of pasturing technology on the current state of pastures. Annals of Agri Bio Research, 24(2): 246-254. [37] Patton, B.D., Dong, X.J., Nyren, P.E., Nyren, A. (2007). Effects of grazing intensity, precipitation, and temperature on forage production. Rangeland Ecology & Management, 60(6): 656-665. https://doi.org/10.2111/07-008R2.1 [49] Lin, Y., Hong, M., Han, G.D., Zhao, M.L., Bai, Y.F., Chang, S.X. (2010). Grazing intensity affected spatial patterns of vegetation and soil fertility in a desert steppe. Agriculture, Ecosystems & Environment, 138(3-4): 282- 292. https://doi.org/10.1016/j.agee.2010.05.013 [38] Gamoun, M., Patton, B., Hanchi, B. (2015). Assessment of vegetation response to grazing management in arid rangelands of southern Tunisia. International Journal of Biodiversity Science, Ecosystem Services & Management, 11(2): 106-113. https://doi.org/10.1080/21513732.2014.998284 [50] Chen, W.Q., Huang, D., Liu, N., Zhang, Y.J., Badgery, W.B., Wang, X.Y., Shen, Y. (2015). Improved grazing management may increase soil carbon sequestration in temperate steppe. Scientific Reports, 5(1): 10892. https://doi.org/10.1038/srep10892 [39] Haider, M.S., Maclaurin, A., Chaudhry, A.A., Mushtaque, M., Ullah, S. (2011). Effect of grazing systems on range condition in Pabbi Hills Reserve Forest, Kharian, Punjab, Pakistan. Chilean Journal of Agricultural Research, 71(4): 560-565. https://doi.org/10.4067/S0718-58392011000400010 [51] Terefe, H., Argaw, M., Tamene, L., Mekonnen, K., Recha, J., Solomon, D. (2020). Effects of sustainable land management interventions on selected soil properties in Geda watershed, central highlands of Ethiopia. Ecological Processes, 9: 1-11. https://doi.org/10.1186/s13717-020-0216-2 [40] Hickman, K.R., Hartnett, D.C. (2002). Effects of grazing intensity on growth, reproduction, and abundance of three palatable forbs in Kansas tallgrass prairie. Plant Ecology, 159: 23-33. https://doi.org/10.1023/A:1015534721939 [52] Muller, M., Siebert, S.J., Ntloko, B.R., Siebert, F. (2021). A floristic assessment of grassland diversity loss in South Africa. Bothalia-African Biodiversity & Conservation, 51(1): 1-9. https://hdl.handle.net/10520/ejc- bothalia_v51_n1_a11 [41] Kemp, D.R., Behrendt, K., Badgery, W.B., Han, G.D., Li, P., Zhang, Y.J., Wu, J.P., Hou, F.J. (2020). Chinese degraded grasslands-pathways for sustainability. The Rangeland Journal, 42(5): 339-346. https://doi.org/10.1071/RJ20033 [53] Patra, A.K., Abbadie, L., Clays-Josserand, A., Degrange, V., Grayston, S.J., Loiseau, P., Louault, F., Mahmood, S., Nazaret, S., Philippot, L., Poly, F., Prosser, J.I., Richaume, A., Le Roux, X. (2005). Effects of grazing on microbial functional groups involved in soil N dynamics. Ecological Monographs, 75(1): 65-80. https://doi.org/10.1890/03-0837 [42] Mayel, S., Jarrah, M., Kuka, K. (2021). How does grassland management affect physical and biochemical properties of temperate grassland soils? A review study. Grass and Forage Science, 76(2): 215-244. 5. CONCLUSIONS Journal of Geographical Sciences, 24: 303-312. https://doi.org/10.1007/s11442-014-1089-z [17] Lazareva, V.G., Bananova, V.A., Van Zung, N. (2020). Dynamics of modern vegetation for pasture use in the Northwestern Pre-Caspian Region. Arid Ecosystems, 10: 276-283. https://doi.org/10.1134/S2079096120040137 [18] Zamin, T.J., Grogan, P. (2013). Caribou exclusion during a population low increases deciduous and evergreen shrub species biomass and nitrogen pools in low a rctic tundra. Journal of Ecology, 101(3): 671-683. https://doi.org/10.1111/1365-2745.12082 [30] Brade, W. (2012). Advantages and disadvantages of grazing for high-yielding dairy cows. Berichte Über Landwirtschaft, 90(3): 447-466. [31] Nasiyev, B. (2016). The study of the processes, degradation factors and the selection of crops for the restoration of bioresourses capacity of the grassland of semi-desert zones. Research Journal of Pharmaceutical, Biological and Chemical Sciences, 7(3): 2637-2646. [19] Stark, S., Männistö, M.K., Ganzert, L., Tiirola, M., Häggblom, M.M. (2015). Grazing intensity in subarctic tundra affects the temperature adaptation of soil microbial communities. Soil Biology and Biochemistry, 84: 147-157. https://doi.org/10.1016/j.soilbio.2015.02.023 [32] Nasiyev, B., Zhanatalapov, N., Yessenguzhina, A., Yeleshev, R. (2019). The use of sudan grass for the production of green fodder, hay and haylage in Western 773 Kazakhstan. Ecology, Environment and Conservation, 25(2): 767-774. impact on plant species richness in nutrient‐poor vs. nutrient‐rich ecosystems. Ecology, 79(8): 2581-2592. https://doi.org/10.1890/0012- 9658(1998)079[2581:ROGIOP]2.0.CO;2 [33] Yesmagulova, B.Z., Assetova, A.Y., Tassanova, Z.B., Zhildikbaeva, A.N., Molzhigitova, D.K. (2023). Determination of the degradation degree of pasture lands in the West Kazakhstan region based on monitoring using geoinformation technologies. Journal of Ecological Engineering, 24(1), 179-187. https://doi.org/10.12911/22998993/155167 [45] Li, W., Huang, H.Z., Zhang, Z.N., Wu, G.L. (2011). Effects of grazing on the soil properties and C and N storage in relation to biomass allocation in an alpine meadow. Journal of Soil Science and Plant Nutrition, 11(4): 27-39. https://doi.org/10.4067/S0718- 95162011000400003 [34] Dospekhov, B.A. (2014). Methodology of Field Experience (with the basics of statistical processing of research results): Textbook for Higher Agricultural Educational Institutions. Alliance, Moscow. [46] Cooper, A., McCann, T., Ballard, E. (2005). The effects of livestock grazing and recreation on Irish machair grassland vegetation. Plant Ecology, 181: 255-267. https://doi.org/10.1007/s11258-005-7149-9 [35] Novoselov, Y.K., Kharkov, G.D., Shekhovtsova, N.S. (1983). Methodical Instructions for Conducting Field Experiments with Forage Crops. VNNIK, Moscow. [47] Sasaki, T., Okayasu, T., Takeuchi, K., Jamsran, U., Jadambaa, S. (2005). Patterns of floristic composition under different grazing intensities in Bulgan, South Gobi, Mongolia. Grassland Science, 51(3): 235-242. https://doi.org/10.1111/j.1744-697X.2005.00029.x [36] Gamoun, M. (2014). Grazing intensity effects on the vegetation in desert rangelands of Southern Tunisia. 5. CONCLUSIONS https://doi.org/10.1111/gfs.12512 [54] Chadaeva, V., Gorobtsova, O., Pshegusov, R., Tsepkova, N., Tembotov, R., Khanov, Z., Gedgafova, F., Zhashuev, A., Uligova, T., Khakunova, E., Stepanyan, E. (2021). Stages of grassland degradation in subalpine ecosystems of the Central Caucasus, Russia. Chilean Journal of Agricultural Research, 81(4): 630-642. https://doi.org/10.4067/S0718-58392021000400630 [43] Fynn, R.W.S., O’connor, T.G. (2000). Effect of stocking rate and rainfall on rangeland dynamics and cattle performance in a semi‐arid savanna, South Africa. Journal of Applied Ecology, 37(3): 491-507. https://doi.org/10.1046/j.1365-2664.2000.00513.x p g j [44] Proulx, M., Mazumder, A. (1998). Reversal of grazing 774
https://openalex.org/W4317813377	https://www.ijfmr.com/papers/2023/1/1398.pdf	English	null	Study of Effect of Temperature, Water, Light and Darkness on Seed Germination in Luffa Cylindrical L	International Journal For Multidisciplinary Research	2,023	cc-by-sa	3,189	Study of Effect of Temperature, Water, Light and Darkness on Seed Germination in Luffa aegyptiaca Mill. Jeetendra Sainkhediya1, Chhatarsingh Chouhan2 1Department of botany, Govt. P. G. College Sendhwa Dist. Barwani (M.P.), India 2School of Studies in Botany, Vikram University, Ujjain (M.P.), India Abstract Present observe describes the effect of temperature, water, light and darkness on seed germination in Luffa aegyptiaca Mill . (Cucurbitaceae). Seed germination is a Physico-chemical manner by using which the dormant embryo of the seed resumes active increase and forms seedling. Hormones are special organic materials related with increase associated occasions of flora ABA (abscisic acid.) is one among them. Seeds germination is adversely affected by the presence of abscisic acid. Auxins are control many features in flowers. Vernalization cuts the vegetative segment to result in flowering ensuing in elevated yield. Flowering is caused by florigen which synthesized with the aid of leaves. Stress is a temporary section which may be skilled by plant in any country of development. Water pressure can be as a result of lack or excess of water. On hydration, the seeds swell to mucilage formation. Excess or loss of salt results in salt stress. Commentary of plant growth is recorded after 15- 90 days and unearths that 1stpot offers first-class consequences. Due to this pot are installed environmental situation for statement of temperature effect, on seed germination & find that in early period of increase i.e. before flowering pots having most effective soil at 250C temperature showed better boom of root and better variety of leaves and large length of leaves. However later period pot having soil supplemented with bio-fertilizer and special temperature condition than circulate seed have germination but plant increase is lower. That result suggests bio-fertilizer and temperature is more effect at the boom of plants. International Journal for Multidisciplinary Research (IJFMR) E-ISSN : 2582-2160 ● Website: www.ijfmr.com● Email: editor@ijfmr.com International Journal for Multidisciplinary Research (IJFMR) E-ISSN : 2582-2160 ● Website: www.ijfmr.com● Email: editor@ijfmr.com International Journal for Multidisciplinary Research (IJFMR) E-ISSN : 2582-2160 ● Website: www.ijfmr.com● Email: editor@ijfmr.com photograph-blastic seeds. Seed germination depends on both internal and outside conditions (Bewley, 1997 b).The most important outside elements encompass temperature, water, oxygen, light or darkness ( Borthwick et.al.1954). Numerous plant lives require exceptional variables for successful seed germination (Singh 1953); regularly this depends at the individual seed variety and is closely connected to the ecological conditions of a plant's herbal habitat (Copeland, & McDonald, 1995). For a few seeds, their destiny germination response is suffering from environmental conditions all through seed formation; most often these responses are forms of seed dormancy (Baskin and Baskin, 2001, Fenner& Thompson 2005). Our study goal is to the observation of the impact of temperature, water, mild and darkness on seed germination in Luffa aegyptiaca Mill. photograph-blastic seeds. Seed germination depends on both internal and outside conditions (Bewley, 1997 b).The most important outside elements encompass temperature, water, oxygen, light or darkness ( Borthwick et.al.1954). Numerous plant lives require exceptional variables for successful seed germination (Singh 1953); regularly this depends at the individual seed variety and is closely connected to the ecological conditions of a plant's herbal habitat (Copeland, & McDonald, 1995). For a few seeds, their destiny germination response is suffering from environmental conditions all through seed formation; most often these responses are forms of seed dormancy (Baskin and Baskin, 2001, Fenner& Thompson 2005). Our study goal is to the observation of the impact of temperature, water, mild and darkness on seed germination in Luffa aegyptiaca Mill. Systematic position (According to Bentham & Hooker 1862) Division - Phanerogams Class - Dicotyledones Sub-class - Polypetalae Series - Calyciflorae Order - Passiflorales Family - Cucurbitaceae Genus - Luffa Species - aegyptiaca Systematic position (According to Bentham & Hooker 1862) Division - Phanerogams Class - Dicotyledones Sub-class - Polypetalae Series - Calyciflorae Order - Passiflorales Family - Cucurbitaceae Genus - Luffa Species - aegyptiaca Systematic position (According to Bentham & Hooker 1862) Materials and methods 1st pot is installed favorable environmental circumstance where to be had is sufficient temperature. Now observe of low or excessive temperature impact on seed germination. 2nd pot ware put in favorable circumstance however give the extra water amount and regular water are given by drip approach. Now on this pot we take a look at the impact of water on seed germination. Third pot is installed artificial light wherein temperature is low. Now we study the impact of light on seed germination. 4th pot is put in darkish room for germination wherein temperature is reduced & observes the darkness effect of seed germination. 10-10 seeds of Luffa aegyptiaca Mill had been sown in every pot & examine the impact of temperature, water, mild, and darkness on seed germination of Luffa aegyptiaca Mill. The pots have been watered as in line with need and outcomes have been recorded after 15-90 days. The parameters for recording consequences is root and shoot period, number, of leaves, flower and culmination according to flowers (Dadlani& Agrawal, 1987). Introduction Luffa aegyptiaca Mill. is broadly dispensed in the tropics and subtropics, as a cultivated and naturalized plant. Its cultivation is of historical starting place and its miles difficult to decide whether or not the local home is Africa or Asia. The plant takes place wild in West Africa. Luffa accommodates 7 species, four of these local to the old international tropics and 3 quite more distantly associated species indigenous to South the United States. Hybrids of Luffa aegyptiaca Mill. and Luffa acutangula are observed in cultivation. Those are bitter and inedible, however appropriate for the manufacturing of sponges. Luffa aegyptiaca Mill. is broadly dispensed in the tropics and subtropics, as a cultivated and naturalized plant. Its cultivation is of historical starting place and its miles difficult to decide whether or not the local home is Africa or Asia. The plant takes place wild in West Africa. Luffa accommodates 7 species, four of these local to the old international tropics and 3 quite more distantly associated species indigenous to South the United States. Hybrids of Luffa aegyptiaca Mill. and Luffa acutangula are observed in cultivation. Those are bitter and inedible, however appropriate for the manufacturing of sponges. Many elements are accountable for seed germination (Hilhort 1990).Water, temperature and air is essential for seed germination. Water is a basic necessity for seed germination as well as finest temperature & adequate quantity of air is also required for seed germination (Bradford &Bewley, 2002). Most of the seeds do not require light for his or her germination but light is critical for the germination of a few Many elements are accountable for seed germination (Hilhort 1990).Water, temperature and air is essential for seed germination. Water is a basic necessity for seed germination as well as finest temperature & adequate quantity of air is also required for seed germination (Bradford &Bewley, 2002). Most of the seeds do not require light for his or her germination but light is critical for the germination of a few Volume 5, Issue 1, January-February 2023 IJFMR23011398 1 1 International Journal for Multidisciplinary Research (IJFMR) E-ISSN : 2582-2160 ● Website: www.ijfmr.com● Email: editor@ijfmr.com (2) Water effect: This is the pot experiments extra water added to seed germination and at the time of seed sowing the environment temperature approximate 250C. accordingly within the pot after 30 days of seed sowing there are not germinate of some other seed in water logging situations due to the fact seeds are dormant degree inside the motive of difficult seed coat and water logging circumstance. as a result after 35 day the water quantity is less within the pot and it at the same time as after five days the seed germination is being. However after 40-50 days of seed germination tons flower and culmination are rise up at plant. However these reason of high temperature, now not right boom of leaves, fruit and flower. Consequently fruit are smaller in length. While temperature is long past up approximate 410C than leaves are dry and boom of plant is lower. (3) Light impact: For the germination of seed we took this pot in synthetic mild at room temperature. After 10 days of seed sowing 4 seed are germinated this time approximate 160C of room temperature as a consequence the cause of low temperature there aren't any extra seed germination. therefore in this pot at the time of seed germination plumule period is gradually will increase and leaf are small in length and the leaf shades is green yellowish and after 50 days plant boom is absolutely stopped and plant are dies on the low temperature situation. (4) Darkness impact: This kind of examine is performed by Anderson & Boardman, (1964) on the greening of darkish-grown bean plant. Bean vegetation were grown within the darkish for 10-14 days and then transferred to non-stop artificial mild for varying durations. After illumination, the plastids were purified by means of differential and density-gradient centrifugation. The structure of the plastids turned into studied by using section-assessment and fluorescence microscopy. In our experiments the pots are put on darkish room for the seed germination and to the study of darkness results of seed germination. On this situation room temperature had been between 14-170C due to the fact in keeping with time the temperature is changed. Out of 8 days 3 seeds are germinate due to the fact no other seed germination at low temperature and absence of light. Observation The parameters for the check of effect of Temperature, Water, light and Darkness on Seed Germination of Luffa aegyptiaca Mill has been categories on fallowing: (1) Temperature effect: Pots are putted in environmental situation & discovered the impact of temperature in seed germination. Seed sown the temperature is approximate at 250C and after 11days all sown seed are germinated. After 15 day of the seed sown we're gazing that pot no.1 the seeds are grown and locate that radical boom is better than plumule and increase the leaf size and leaf shade is plenty green. In line with time the temperature in regularly elevated whilst the temperature is 320C than regular affected to plant boom and the vegetation and fruits are gradually lower in size. Whilst the temperature is 400C whom lots flower are upward push at plant but these the cause of high temperature most flower and leaf are fall down at flowers and culmination are also small in length. Volume 5, Issue 1, January-February 2023 IJFMR23011398 2 International Journal for Multidisciplinary Research (IJFMR) E-ISSN : 2582-2160 ● Website: www.ijfmr.com● Email: editor@ijfmr.com Therefore in this pot on the time of seed germination plumule duration is gradually will increase and leaves are small in size and leaf coloration is greenish yellow. After 45 days plant growth is less at low temperature and absent of sunlight resultant the leaves are yellowish and plant died. All this four parameters result is presented in the table-1. Table – 1:Growth of Luffa aegyptiaca plants after 15-90 days in various pots after days 15 30 45 60 75 90 POT NO. 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 RL (cm) 3 - 6 5 6 - 7 7 1 2 5 9 1 0 14 7 1 0 D 17 10 D D 19 14 D D SL (cm) 2 - 9 1 0 1 9 - 2 0 1 9 6 0 4 3 5 3 0 10 0 2 5 4 5 - 13 0 60 - - 16 0 90 - - LN 4 - 2 2 1 4 - 6 5 5 0 4 9 8 13 0 1 2 1 4 - 20 0 55 - - 25 0 12 5 - - FN - - - - - - - - 1 - - - 7 - - - 12 3 - - 15 6 - - FrN - - - - - - - - - - - - 2 - - - 5 - - - 6 3 - abbreviation :RL=Root length, SL= Shoot length, LN=number of leaves, FN= Flower numbers, FrN= Fruit number ,D=Death Volume 5, Issue 1, January-February 2023 IJFMR23011398 3 International Journal for Multidisciplinary Research (IJFMR) E-ISSN : 2582-2160 ● Website: www.ijfmr.com● Email: editor@ijfmr.com International Journal for Multidisciplinary Research (IJFMR) E-ISSN : 2582-2160 ● Website: www.ijfmr.com● Email: editor@ijfmr.com Table-2: Rate of germination in Luffa aegyptiaca 160C in darkness. SN Days of germination Germination (%) Rate of germination 1 1 0 0 2 5 78 76 3 10 90 12 4 15 92 2 Table-2: Rate of germination in Luffa aegyptiaca 160C in darkness. Plant growths are recorded after 15, 30, 45, 60, 75 and 90 days and discover that 1st pot offers exceptional results. Because of this pot are installed environmental situation for commentary of temperature effect, on seed germination. Discussion For the seed germination is important of daylight, oxygen and mild because inside the absent of mild on extra chloroplast are manufacture or absolutely absent of chloroplast (Mayer and Mayer, 1982). While leaf sizes are small and yellow in shade consequently 25-300C temperature is critical for the seed germination. In the pot no.1, 250C temperature is occupant on the seed germination time that's important for germination and plant increase consequently well boom of plant. However temperature is boom at 400C than plant growth is quite effected and the leaves and flowers are decline from vegetation. For that reason greater water in pot no.2, after 30 days on seed germination whilst lower water quantity than after 35 days seed germination is begin. Accordingly in pot no. Three, seeds are germinate in presence of artificial light however plant increase is enormously effects at low temperature and occasional light circumstance. While pot no. Four, installed dark room where light and temperature could be very low for this region few seeds are germinated. International Journal for Multidisciplinary Research (IJFMR) E-ISSN : 2582-2160 ● Website: www.ijfmr.com● Email: editor@ijfmr.com It is obvious from effects of tables that in early period of growth i.e. earlier than flowering pots having best soil at 250C temperature confirmed better growth of root and better variety of leaves and huge size. But later length pot having soil supplemented with bio-fertilizer and distinctive temperature condition than move seed have germination but plant increase is decrease. That result shows bio-fertilizer and temperature is more impact on the growth of plant life (Hsiao &Vidaver 1989). Seeds had been imbibed in water for 15 days in darkness at 160C. Germination changed into recorded every day up to 15 days (table-2). The temperature effect on seed germination is given in fallowing figure. International Journal for Multidisciplinary Research (IJFMR) E-ISSN : 2582-2160 ● Website: www.ijfmr.com● Email: editor@ijfmr.com case of lotus seeds. Dormancy – in many plant life seed remains in dormant stage for a while (Hendrick&Taylorson 1974). case of lotus seeds. Dormancy – in many plant life seed remains in dormant stage for a while (Hendrick&Taylorson 1974). Economically the Luffa aegyptiaca Mill is cultivated in the course of the country. Fruits are used as vegetable & dried fruits used as a bathtub sponge. It’s far used for washing and scrubbing utensils in addition to the human frame & the manufacture of hats, insoles of shoes, automobile-wipers, pot-holders, table-mats, door and bathtub-mats, sandals and gloves. The fiber has also been used for its shock and sound soaking up homes, as an example in helmets and as a clear out in engines. Medicinally Luffa aegyptiaca Mill. Root used for the treatment of constipation and as a diuretic and leaves used to sell wound restoration and to maturate abscesses. The seed oil is used for the treatment of pores and skin issues. Unripe end results were used as fish poison. Acknowledgement:We specific deep indebtedness to Dr. R. C. Verma and Dr. S. Patel for his or her guiding principle and thankful to Dr. J. K. Sharma, Dr. C. L. Kadel. Grateful to Shri P.N. Sunhel for library helps. Acknowledgement:We specific deep indebtedness to Dr. R. C. Verma and Dr. S. Patel for his or her guiding principle and thankful to Dr. J. K. Sharma, Dr. C. L. Kadel. Grateful to Shri P.N. Sunhel for library helps. Conclusion Some inner & outside elements affected seed germination and the physiological strategies of seed germination. Some crucial ones are Water – Dry seeds do no longer germinate and water is essential for his or her germination. Suitable temperature – Seeds require a gold standard temperature for their germination. Most advantageous temperature range for exceptional germination of seed is 25 to 30⁰C. Light – most of the seeds both do now not require mild or do now not germinate within the presence of light. Some seeds need mild for germination. Seedlings grow greater vigorously throughout darkness alternatively in mild. Darkness – The seedling will become too lengthy and yellow in the absence of mild. Chlorophyll dose no longer from and stem becomes tender long and yellow in the plant life grown in the absence of mild. a few inner function like adulthood of embryo-Many seeds first of all does now not have hormones required for the improvement of the embryos. Such seeds take time for germination. Viability – normally the seeds remain possible for a specific period. Maximum viability has been stated inside the Volume 5, Issue 1, January-February 2023 IJFMR23011398 4 References 1. Anderson JA, & Boardman, NK (1964). Studies on the greening of dark-grown bean plant: IIEd. Development of photochemical activity. AJBS. 17: 93-101. 2. Baskin CC and Baskin, JM (2001). Seeds, Academic press San Diego San Francisco. 3. Bewley JD (1997 b). Seed germination and dormancy. Plant Cell 9:1055-1057. 4. Borthwick HA, Hendrick SB, Toole EH, Toole VK (1954). Action of light on Lettuce seed germination Bot. 13: 122. 4. Borthwick HA, Hendrick SB, Toole EH, Toole VK (1954). Action of light on Lettuce seed germination Bot. 13: 122. 5. Bradford KJ &Bewley, JD (2002). Seeds: Biology, Technology and Role in Agriculture. 5. Bradford KJ &Bewley, JD (2002). Seeds: Biology, Technology and Role in Agriculture. 6. Copeland, LO & McDonald, MB (1995). Principles of Seed Science and Technology, 2ndEd. Chapman & Hall New York. 6. Copeland, LO & McDonald, MB (1995). Principles of Seed Science and Technology, 2ndEd. Chapman & Hall New York. 7. Dadlani M & Agrawal, PK (1987). Techniques in Seed Science and Technology. South Asian Pub. New Delhi. 8. Fenner M, Thompson K (2005). The ecology of seeds. Cambridge Uni. Press. UK 9. Hendrick SB,&Taylorson RB (1974). Promotion of seed germination by nitrate hydroxylamine and ammonium salts. Plant Physiol 54, 304-306. 10. Hilhort HWM (1990). Does response analysis of factors involved in germination and secondary dormancy of seed of Sisymbriumofficinale II nitrate. Plant physiol. 94:1096-1098. 11. Hsiao AJ, Vidaver W (1989). Effect of temperature and various red or far-red irradiations on phytochrome and gibberellins A3- mediated germination control in partially hydrated Lettuce seed. J. Expt. Bot. 135, 1772-1776. 12. Mayer AM, and Mayer, AP (1982). The Germination of seeds, 3rdEdition press, Ox 13. Singh B (1953). Studies on the structure and development of seeds of Cucurbitaceae. Phyto morph 3:224-225. Volume 5, Issue 1, January-February 2023 5 IJFMR23011398 5
https://openalex.org/W2001060109	https://bmcvetres.biomedcentral.com/counter/pdf/10.1186/1746-6148-10-113	English	null	Leptin, adiponectin and serotonin levels in lean and obese dogs	BMC veterinary research	2,014	cc-by	7,035	Abstract Background: Serotonin (5-hydroytryptamine or 5HT) is associated with numerous behavioral and psychological factors and is a biochemical marker of mood. 5HT is involved in the hypothalamic regulation of energy consumption. 5HT t l tit i th t l t (CNS) d ti l t i t ti l bilit Th f t di l ki Background: Serotonin (5 hydroytryptamine or 5HT) is associated with numerous behavioral and psychological factors and is a biochemical marker of mood. 5HT is involved in the hypothalamic regulation of energy consumption. 5HT controls appetite in the central nerve system (CNS) and stimulates intestinal mobility. There are few studies looking at the role of 5HT and the relationship between peripheral circulating serotonin and obesity. The aim of this study was to find any differences in leptin, adiponectin, and 5HT between lean and obese dogs and to identify correlations among these factors. Results: Leptin, triglyceride (TG) and cholesterol levels were higher in the obese group (all p < 0.01). Adiponectin and 5HT levels were higher in the lean group compared to the obese group (p < 0.01). Leptin (r = 0.628, p < 0.01), TG (r = 0.491, p < 0.01) and cholesterol (r = 0.419, p < 0.01) were positively correlated with body condition score (BCS), and adiponectin (r = −0.446, p < 0.01) and 5HT (r = −0.490, p < 0.01) were negatively correlated with BCS. Leptin was negatively correlated with adiponectin (r = −0.294, p < 0.01) and 5HT (r = −0.343, p < 0.01). 5HT was negatively correlated with leptin (r = −0.343, p < 0.01), TG (r = −0.268, p < 0.05) and cholesterol (r = −0.357, p < 0.05). TG (r = 0.491, p < 0.01) and cholesterol (r = 0.419, p < 0.01) were positively correlated with body condition score (BCS), and adiponectin (r = −0.446, p < 0.01) and 5HT (r = −0.490, p < 0.01) were negatively correlated with BCS. Leptin was negatively correlated with adiponectin (r = −0.294, p < 0.01) and 5HT (r = −0.343, p < 0.01). 5HT was negatively correlated with leptin (r = −0.343, p < 0.01), TG (r = −0.268, p < 0.05) and cholesterol (r = −0.357, p < 0.05). Conclusions: 5HT is an important appetite control neurotransmitter, but there are limited studies for 5HT levels related to obesity in dogs. RESEARCH ARTICLE Open Access Open Access * Correspondence: songkh@cnu.ac.kr 1Laboratory of Veterinary Internal Medicine, College of Veterinary Medicine, Chungnam National University, Daejeon 305-764, South Korea Full list of author information is available at the end of the article Leptin, adiponectin and serotonin levels in lean and obese dogs Hyung-Jin Park1, Sang-Eun Lee2, Jung-Hyun Oh3, Kyoung-Won Seo1 and Kun-Ho Song1* Abstract To the best of our knowledge, this is the first study to evaluate peripheral 5HT levels in obese dogs. From this research, we can assume that 5HT may be correlated with canine obesity. Further studies will be needed to further elucidate the role of low serum 5HT levels in canine obesity. Keywords: Adiponectin, Leptin, Obese dog, Serotonin have a higher risk of obesity than other breeds. Canine obesity is associated with several health conditions such as insulin resistance, pancreatitis, cruciate ligament rupture, lower urinary tract disease, oral disease, neoplasia, osteo- arthritis and decreased longevity [9,10]. Additionally, obesity could be risk factors for brachycephalic syndrome and musculoskeletal problems and it may increase risks associated with anesthesia [7]. Leptin was the first identi- fied adipokine and has a role of regulating body fat mass through appetite control and increased energy metabolism [11]. Restriction of food intake results in a suppression of leptin levels, which is returned by refeeding [11]. Circulat- ing leptin levels correlates positively with adipose tissue mass and exogenous leptin replacement decreases fast- induced hyperphagia [11]. Circulating leptin interacts with peripheral 5HT and decreases appetite [3]. The majority of obese animals and humans have high serum leptin levels, which suggests leptin resistance [11]. Leptin resist- ance results from a signaling defect in leptin-responsive © 2014 Park et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Park et al. BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Park et al. BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Background Obesity is a disease characterized by excessive adipose tissue accumulation in the body. It is a common nutri- tional problem in small animal medicine [1,2]. Ideally, fat mass accounts for approximately 15% to 20% in dogs and cats [3,4]. Pets are considered overweight when body weight exceeds ideal body weight by 10% to 20%, and pets are considered obese when body weight exceeds ideal body weight by 20% to 30% [4]. Worldwide, approximately 25% to 35% of adult cats and 35% to 40% of adult dogs are overweight or obese [5,6]. Middle-aged neutered male cats and middle-aged spayed female dogs are at the highest risk for obesity [7]. According to Lund et al. [8], some purebred dogs including Shetland Sheepdogs, Golden Retrievers, Dachsunds, Cocker Spaniels, and Rottweilers * Correspondence: songkh@cnu.ac.kr 1Laboratory of Veterinary Internal Medicine, College of Veterinary Medicine, Chungnam National University, Daejeon 305-764, South Korea Full list of author information is available at the end of the article Park et al. BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Page 2 of 8 hypothalamic neurons and impaired transport into the brain [11]. Obesity can be a cause of leptin resistance, but a lack of sensitivity to circulating leptin may also induce obesity [11,12]. Adiponectin is believed to be produced by mature adipocytes and it contributes to increased insulin sensitivity, decreased blood glucose, and decreased tissue triglyceride (TG) content in the liver and muscle [12]. Related to obesity, adiponectin has a negative correlation with obesity [12]. The decreasing of adiponectin in obesity is more severe with visceral than subcutaneous adiposity in human and the composition of adiponectin is also changed. Increased production of pro-inflammatory cytokines such as TNF-α and IL-6 with increased fat mass may inhibit adiponectin gene expression [12]. Serotonin (5-hydroytryptamine or 5HT) is associated with numerous behavioral and psychological factors and is a biochemical marker of mood [13,14]. 5HT is involved in the hypothalamic regulation of energy consumption and 5HT levels in the central nervous system (CNS) are influenced by energy conditions [15]. In the CNS, 5HT has a hypophagic effect owing to an inhibitory interaction with the orexigenic system through hypocretins and neuropeptide Y (NPY). 5HT also likely has a stimulatory effect on the anorexigenic melanocortin system [16]. Sample analysis d k For adipokine analyses, we collected blood by venipunc- ture using minimal restraint from the jugular vein of dogs that were deemed healthy based on history and physical examination. A fraction of the collected blood was placed in an EDTA tube for a CBC. The remaining blood was allowed to clot at room temperature for 30 minutes and then centrifuged at 2,500 rpm for 15 minutes. Serum and plasma samples were frozen and stored at −80°C, and until analysis of adipokines, 5HT, tT4, cortisol, TG and cholesterol levels. Sandwich enzyme-linked immunosorbent assay and hormone assay Serum leptin and adiponectin levels were measured according to the manufacturer’s protocol using a canine leptin commercial sandwich enzyme-linked immuno- sorbent assay (ELISA) kit (Millipore, Billerica, MA, USA) and canine adiponectin commercial ELISA kit (Milipore, Billerica, MA, USA). All samples were tested in duplicate. Serum tT4 and cortisol levels were quantified using an Immulite 1000 immunoassay analyzer (Siemens Medical Diagnostics, Los Angeles, CA, USA). Plasma 5HT concen- trations were measured by a commercially available ELISA kit (Serotonin EIA Kit, Enzo Life Sciences Inc, MI, USA) according to the manufacturer’s instructions [18]. Background Concerning interactions among leptin and 5HT, a previous mouse model study reported that periph- eral leptin levels were reduced by 5HT, and 5HT exerted a direct effect on adipocytes and regulated leptin release from adipocytes [15]. Few studies have been published looking at the role of 5HT and the relationship between peripheral circulating serotonin and obesity [14,15]. There have been no studies performed to evaluate peripheral 5HT levels in obese dogs until now. The aim of this study was to identify differences and correlations among per- ipheral concentrations of leptin, adiponectin, and 5HT between lean and naturally obese dogs. care. Owners were asked to have their dogs fast for 12 hours prior to presentation for blood sampling. One investigator weighed and examined each dog to screen for potential underlying inflammatory conditions. The same investigator determined a body condition score (BCS) and assigned the dog to either the control group (BCS 4-5/9) or the obese group (BCS 7.5-9/9) utilizing a 9-point BCS system [17]. A screening test for overall health included a physical examination, along with history taking, complete blood count (CBC), serum biochemistry, electrolyte analysis, total thyroxine (tT4), and adreno- corticotropic hormone (ACTH) assay. After the screening test, 18 dogs were excluded in this study due to under- lying disease including liver failure, heart disease and hyperadrenocorticism. Also, dogs that were receiving sero- tonergic drugs, such as fluoxetine or being fed serotonin- containing food, such as bananas, tomatoes, pineapples, or walnuts were excluded from this study. Finally, 82 healthy dogs were enrolled for this research (Table 1). Ethics statement Blood samples of dogs owned by private individuals were used in the present study. The sample collection protocol was approved by Laboratory Animals of Chungnam National University committee (approved No. CNU 00370) and the investigators adhered to the Guide for the Care and Use of Laboratory Animals of Chungnam National University. Animals and sampling A total of 100 healthy dogs (age range: 1 to 15 years-old; mean age: 7 years) owned by clients of Chungnam National University Veterinary Medicine Teaching Hospital were recruited to be in the study from March 2012 to February 2013. Before examination, informed client consent was obtained and we adhered to a high standard of veterinary Statistical analyses were performed by a commercially available computer-based software program (IBM SPSS Statistics 20.0.0, SPSS Inc., Chicago, IL USA). Age, body weight, and BCS results are presented as means with a standard error. Serum leptin, adiponectin, 5HT, TG, cholesterol, tT4, and cortisol levels are presented as Page 3 of 8 Page 3 of 8 Park et al. BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Park et al. BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Table 1 Profile of dogs in lean and obese groups (Mean ± SE) Age Body weight BCS Gender Breed Lean group 4.12 ± 2.77 5.47 ± 0.23 4.32 ± 0.07 Male (4) Shih-Tzu (5) Male Poodle (3) Castrated (9) Maltese (3) Female (13) Mongrel (12) Female spayed (15) Schnauzer (3) Cocker spaniel (1) Yorkshire terrier (1) Beagle (13) Obese group 7.07 ± 3.20 10.73 ± 1.25 8.09 ± 0.09 Male (1) French bulldog (1) Male Shih-Tzu (13) Castrated (18) Cocker spaniel (5) Female (10) Poodle (3) Female spayed (12) Mongrel (2) Golden retriever (2) Yorkshire terrier (4) Beagle (2) Maltese (4) Dachshund (1) Jindo (1) Chihuahua (1) means with a standard error and 95% confidence intercal. A P value of < 0.05 was considered significant. Age, BCS, leptin, adiponectin, 5HT, total thyroxine (tT4), cortisol, platelets, TG and cholesterol were compared between the lean group and the obese group using the independent t-test. The correlation between variables including age, sex, platelets, leptin, adiponectin, 5HT, TG, cholesterol, tT4, cortisol, and BCS were assessed by Pearson’s cor- relation coefficients. levels were higher in the lean group than the obese group (p < 0.01). The distribution of plasma 5HT concen- tration is shown in Figure 1. No significant differences were found in tT4 or cortisol between the lean and obese groups. To evaluate the effect of sex or being castrated/ spayed, we divided animals intact and castrated or spayed within the lean and obese groups. In the obese group, spayed females had a lower level of 5HT than the intact female group (Table 3). Leptin, adiponectin and 5HT levels in the lean and obese groups In this study, dogs were separated by BCS into lean and obese groups. Each group had 41 dogs. In the lean group (male, n = 13; female, n = 28), the average age was 4.12 years, the average body weight was 5.47 kg, and the average BCS was 4.32. In the obese group (male, n = 19; female, n = 22), the average age was 7.07 years, the aver- age body weight was 10.73 kg, and the average BCS was 8.09. The distributions of BCS and gender in each group are presented in Table 1. Obesity-related parameters, including leptin, adiponectin and 5HT levels, are pre- sented in Table 2. The obese group showed significantly higher body weights and BCS (both p < 0.01). Leptin, TG, and cholesterol levels were higher in the obese group than in the lean group (p < 0.01). Adiponectin and 5HT Animals and sampling The general tendency of the obes- ity related parameter was not influenced by gender status (Table 3). To estimate the effect of aging on obesity related parameters, the dogs were divided into two groups (young: < 8 years old; old: > 8 years old) and within each group, we created two subgroups (lean vs. obese group). Regardless of age, the lean group showed a lower level of BCS and leptin, and a higher level of 5HT than the obese group (Table 4). means with a standard error and 95% confidence intercal. A P value of < 0.05 was considered significant. Age, BCS, leptin, adiponectin, 5HT, total thyroxine (tT4), cortisol, platelets, TG and cholesterol were compared between the lean group and the obese group using the independent t-test. The correlation between variables including age, sex, platelets, leptin, adiponectin, 5HT, TG, cholesterol, tT4, cortisol, and BCS were assessed by Pearson’s cor- relation coefficients. Discussion in canine obesity [19]. Circulating leptin levels are posi- tively correlated with adipose tissue mass, and exogenous leptin replacement decreases fast-induced hyperphagia [11]. However, the majority of obese animals and humans have high serum leptin levels, which suggests leptin resist- ance [11]. In this study, the serum leptin levels of dogs in the obese group were higher than those in the lean group, and adiponectin levels in the obese group were lower than those in the lean group (Table 2). Obesity can be a cause of leptin resistance, but a lack of sensitivity to circulating leptin may also induce obesity [11,20]. In the current research, we found that obese dogs had lower 5HT and adiponectin levels and higher leptin levels than lean dogs. TG and cholesterol levels were high in the obese group compared with the lean group. Body fat is no longer considered simply as energy storage. Adipocytes produce many cytokines and hormones referred to as fat-derived peptides [12]. Adipokines include adiponectin, leptin, resistin, amyloid A, transforming growth factor β, tumor necrosis factor and interleukin-6 [7,12,19]. Currently, there is much interest in the role of adipokines Middle-aged neutered male cats and middle-aged spayed female dogs have a high risk of being overweight or obese [8,21]. With aging, lean body mass gradually declines, resulting in a reduced basal metabolic rate (BMR) and reduced total daily energy needs. If accompanied by reduced exercise, the loss of lean body mass will be exacerbated. This study revealed no significant differ- ences in obesity-related parameters, with the exception of cholesterol, between young and old dogs. However, spayed female dogs had a significantly higher BCS than intact females and this is the same result as a previous study [8]. Neutering decreases the BMR by 25% to 33% [22]. Most dog owners are unaware of the altered metabolic activity after ovariohysterectomy and do not decrease the food supply, which leads to increased food intake in dogs. This feeding behavior is often accompanied by decreased physical activity and, consequently, spayed female dogs gain weight easily. Figure 1 Serotonin concentration between lean and obese groups. Data are presented as boxes and whiskers. Each box includes the interquartile range, whereas the line within a box represents the median, and the whiskers represent the range, extending to a maximum of 1.5 times the interquartile range. Serotonin levels were significantly lower in obese group compared with that of lean group. Correlation between variables Leptin (r = 0.628, p < 0.01), TG (r = 0.491, p < 0.01) and cholesterol (r = 0.419, p < 0.01) levels were positively correlated with BCS and adiponectin (r = −0.446, p < 0.01) and 5HT (r = −0.490; p < 0.01) levels were negatively correlated with BCS (Figure 2). Leptin was negatively correlated with adiponectin (r = −0.294; p < 0.01) and 5HT (r = −0.343; p < 0.01), but leptin was positively corre- lated with cholesterol (r = 0.516; p < 0.01) (Figure 3). Park et al. BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Page 4 of 8 Page 4 of 8 Table 2 Obesity parameter, adipokines (leptin and adiponectin) and serotonin levels between lean and obese groups (Mean ± SE, 95% confidence interval) Lean group Obese group P-value (N = 41) (N = 41) Mean ± SE 95% CI Mean ± SE 95% CI BCS 4.31 ± 0.07 4.16 4.45 8.07 ± 0.09 7.89 8.25 < 0.01 Leptin (ng/mL) 2.69 ± 1.08 1.91 3.46 10.27 ± 1.20 7.91 12.63 < 0.01 Adiponeptin (μg/mL) 12.18 ± 1.09 10.05 14.31 6.23 ± 0.79 4.66 7.78 < 0.01 Serotonin (ng/mL) 792.96 ± 102.45 749.49 839.47 560.43 ± 245.95 491.28 623.19 < 0.01 Triglyceride (mg/dL) 51.53 ± 29.03 41.78 61.28 138.64 ± 125.46 92.13 180.95 < 0.01 Cholesterol (mg/dL) 211.03 ± 73.80 186.23 235.82 289.97 ± 112.93 254.86 325.88 0.01 Total T4 (μg/dL) 1.74 ± 0.66 1.50 1.97 1.55 ± 0.55 1.36 1.726 0.19 Cortisol (μg/dL) 4.61 ± 2.36 3.78 5.44 5.39 ± 3.43 4.26 6.52 0.29 CI; confidence interval. Table 2 Obesity parameter, adipokines (leptin and adiponectin) and serotonin levels between lean and obese groups (Mean ± SE, 95% confidence interval) Discussion p < 0.01: significant difference were observed between lean and obese groups. In our study, the obese group had higher TG and chol- esterol levels than the lean group. However, in the obese group, the average TG level (138.64) did not exceed 1000 mg/dL and the average cholesterol level (289.97) was less than 750 mg/dL. The goal of hyperlipidemia treatment is to maintain plasma lipid concentrations under a level at which health problems are likely to occur. Dietary intervention is recommended in dogs that Figure 1 Serotonin concentration between lean and obese groups. Data are presented as boxes and whiskers. Each box includes the interquartile range, whereas the line within a box represents the median, and the whiskers represent the range, extending to a maximum of 1.5 times the interquartile range. Serotonin levels were significantly lower in obese group compared with that of lean group. p < 0.01: significant difference were observed between lean and obese groups. Figure 1 Serotonin concentration between lean and obese groups. Data are presented as boxes and whiskers. Each box includes the interquartile range, whereas the line within a box represents the median, and the whiskers represent the range, extending to a maximum of 1.5 times the interquartile range. Serotonin levels were significantly lower in obese group compared with that of lean group. *p < 0.01: significant difference were observed between lean and obese groups. Park et al. Discussion BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Page 5 of 8 Page 5 of 8 Table 3 Obesity parameter, adipokines, and serotonin levels according to sex (Mean ± SE) Males Females Intact Castrated Intact Spayed (N = 4) (N = 9) (N = 13) (N = 15) Lean group BCS 4.25 ± 0.25 4.44 ± 0.17 4.0 ± 0 4.53 ± 0.13 Leptin (ng/mL) 5.06 ± 2.85 2.60 ± 0.48 1.59 ± 0.18 2.88 ± 0.68 Adiponectin (μg/mL) 10.53 ± 1.29 11.55 ± 2.20 14.31 ± 2.61 11.14 ± 1.42 Serotonin (ng/mL) 871.31 ± 61.31 736.0 ± 43.98 739.49 ± 15.69 832.21 ± 22.45 Triglyceride (mg/dL) 64.75 ± 18.32 42.0 ± 9.54 38.57 ± 7.08 51.21 ± 5.44 Cholesterol (mg/dL) 282.0 ± 38.02 218.33 ± 27.36 163.14 ± 18.02 204.57 ± 17.45 Intact Castrated Intact Spayed (N = 1) (N = 18) (N = 10) (N = 12) Obese group BCS 8 8.16 ± 0.14★★ 8.10 ± 0.23★★ 7.67 ± 0.22★★ Leptin (ng/mL) 11.6 11.66 ± 1.70★★ 9.91 ± 3.20★★ 8.39 ± 1.91★★ Adiponectin (μg/mL) 5.2 5.92 ± 1.07★ 6.80 ± 1.38★★ 6.29 ± 1.99★ Serotonin (ng/mL) 622.45 533.26 ± 58.89★★ 682.14 ± 59.95 478.0 ± 74.06,★★ Triglyceride (mg/dL) 288 154.16 ± 35.13★ 148.57 ± 45.48★ 95.91 ± 14.08★ Cholesterol (mg/dL) 325 301.38 ± 23.22★ 240.20 ± 24.14★ 296.72 ± 49.91★ *p < 0.05 between intact and spayed groups. ★p <0.05 between lean and obese groups. ★★p <0.01 between lean and obese groups. Table 3 Obesity parameter, adipokines, and serotonin levels according to sex (Mean ± SE) have fasting TG levels greater than 500 mg/dL or choles- terol levels over 750 mg/dL [23]. In the present study, the obese group had high TG and cholesterol levels, but the likelihood of health problems induced by hyperlipid- emia in these animals is marginal. In this study, the per- ipheral 5HT levels in the obese group were significantly lower than in the lean group, which is similar to human study results [13]. Furthermore, a negative correlation between leptin and 5HT was observed in this study. In the obese group, spayed female showed lower 5HT levels compared with intact female. In human studies, natural postmenopausal and ovariectomised women had lower 5HT levels than regularly menstruating women [24]. Estrogen withdrawal alters serotonergic functioning and exogenous estrogen supplementation could increase 5HT levels in postmenopausal women [15,24]. Discussion As with human studies, we found a similar phenomenon in spayed obese dogs. 5HT throughout the body influences food consump- tion by controlling satiety [25]. Therefore, low levels of 5HT could be a risk factor for obesity due to increased appetite. 5HT has a hypophagic effect in the CNS, and 5HT concentrations in the peripheral nervous system may not necessarily equate to 5HT levels or availability in the brain, since 5HT cannot cross the blood–brain barrier [14]. L-tryptophan, an amino acid and precursor to serotonin is converted to 5-hydroxy-L-tryptophan (5-HTP) and then to serotonin in both the CNS and PNS [26]. Brain tryptophan and serotonin levels are determined by the ratio of plasma tryptophan to other large neutral amino acids (LNAAs), which compete with tryptophan for uptake Table 4 Obesity parameter, adipokines and serotonin levels according to age (Mean ± SE) Young group (< 8-year old) (N = 58) Old group (> 8-year old) (N = 24) Lean group Obese group P- value Lean group Obese group P- value (N = 36) (N = 22) (N = 5) (N = 19) BCS 4.31 ± 0.08 8.09 ± 0.15 < 0.01 4.40 ± 0.24 7.89 ± 0.16 < 0.01 Leptin (ng/mL) 2.64 ± 0.44 10.29 ± 1.87 < 0.01 3.02 ± 0.66 10.24 ± 1.47 0.023 Adiponeptin (μg/mL) 12.28 ± 1.19 5.83 ± 0.90 < 0.01 11.47 ± 2.66 6.68 ± 1.39 0.13 Serotonin (ng/mL) 794.70 ± 16.76 587.71 ± 54.94 < 0.01 780.41 ± 57.01 528.83 ± 53.80 0.032 Triglyceride (mg/dL) 48.21 ± 4.41 150.56 ± 34.27 < 0.01 70.80 ± 22.19 131.66 ± 26.08 0.251 Cholesterol (mg/dL) 205.97 ± 14.34 276.65 ± 19.72 < 0.01 240.40 ± 19.50 299.68 ± 31.85 0.363 Total T4 (μg/dL) 1.82 ± 0.12 1.50 ± 0.13 0.103 1.38 ± 0.17 1.58 ± 0.13 0.434 Cortisol (μg/dL) 4.42 ± 0.42 4.80 ± 0.83 0.652 5.54 ± 1.17 5.97 ± 0.78 0.793 4 Obesity parameter, adipokines and serotonin levels according to age (Mean ± SE) Park et al. BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Page 6 of 8 Figure 2 Correlation between BCS and obesity parameters. (A) BCS and adiponectin showed negative correlation (r = −0.446, p < 0.01). (B) BCS and Cholesterol showed positive correlation (r = 0.419, p < 0.01). (C) BCS and leptin showed positive correlation (r = 0.628, p < 0.01). Discussion Enterochromaffin (EC) cells in the intestinal epithelium release 5HT according to mechanical stimulation, to promote transit [28]. Experimentally a diet-induced obesity model showed decreased 5HT levels with a decreased number of EC cells [11]. The inflammation associated with changes in the GI microbiota is considered as the reason for decreased 5HT availability in obese status [28,29]. 5HT, as a neurotransmitter, controls food satiety, and, therefore, high 5HT concentrations decrease leptin and adiponectin concentrations. We found that 5HT is negatively correlated to leptin. Based on the previous result of a mouse model that 5HT could reduce the secretion of leptin, we can consider the possibility that lowered 5HT failed to properly suppress increasing secretion of leptin [15]. As seen in Figure 3C, however, there are more dogs presenting a high level of 5HT at a low level of leptin than ones presenting a low level of 5HT at a high level of leptin. Therefore, further studies would be needed with a larger number of dogs to evaluate the interaction between 5HT and leptin. The low level of adiponectin in the obese group was similar to the results of previous studies [30,31]. Adiponectin was negatively correlated with obesity [32]. The decreased adiponectin level in obesity is more significant in visceral than sub- cutaneous adiposity in humans, and the composition of adiponectin also changes with location in the body. 5HT suppresses adiponectin, and, therefore, a high peripheral adiponectin level can induce a low 5HT level. However, we observed low levels of both 5HT and adiponectin in the obese group in our study. Through this research, we found that the level of peripheral 5HT is low in the obese group. Since 5HT l d l b l h hypothesis, further research is needed to identify that 5HT treatment could actually activate intestinal mobility and how it might work to reduce obesity. It is possible that humans are becoming obese as a result of overeating in order to maintain serotonin levels and the resulting positive mood. Therefore, serotonin agonists may be a treatment option for obesity in humans [33]. From this research, we found that the 5HT level of obese dogs was lower than that of lean dogs, therefore, we could assume that a serotonin agonist might be helpful in increasing the 5HT level in obese dogs. Increased 5HT levels via a serotonin agonist may control appetite and prompt intestinal motility. Discussion (D) BCS and serotonin showed negative correlation (r = −0.490, p < 0.01). (E) BCS and tirglyceride showed positive correlation (r = 0.491, p < 0.01). Figure 2 Correlation between BCS and obesity parameters. (A) BCS and adiponectin showed negative correlation (r = −0.446, p < 0.01). (B) BCS and Cholesterol showed positive correlation (r = 0.419, p < 0.01). (C) BCS and leptin showed positive correlation (r = 0.628, p < 0.01). (D) BCS and serotonin showed negative correlation (r = −0.490, p < 0.01). (E) BCS and tirglyceride showed positive correlation (r = 0.491, p < 0.01). Figure 3 Correlation among leptin, adiponectin, cholesterol and serotonin. (A) Leptin and adiponectin showed negative correlation (r = −0.294, p < 0.01). (B) Leptin and Cholesterol showed positive correlation (r = 0.516, p < 0.01). (C) Leptin and serotonin showed negative correlation (r = −0.343, p < 0.01). Figure 3 Correlation among leptin, adiponectin, cholesterol and serotonin. (A) Leptin and adiponectin showed negative correlation (r = −0.294, p < 0.01). (B) Leptin and Cholesterol showed positive correlation (r = 0.516, p < 0.01). (C) Leptin and serotonin showed negative correlation (r = −0.343, p < 0.01). Figure 3 Correlation among leptin, adiponectin, cholesterol and serotonin. (A) Leptin and adiponectin showed negative correlation (r = −0.294, p < 0.01). (B) Leptin and Cholesterol showed positive correlation (r = 0.516, p < 0.01). (C) Leptin and serotonin showed negative correlation (r = −0.343, p < 0.01). Park et al. BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Page 7 of 8 into the brain [14]. After eating carbohydrates, insulin is released, which promotes uptake of the LNAAs, but not tryptophan, into skeletal muscles. This helps tryptophan pass more easily into the brain, which increases serotonin production in the brain [14,25]. Increased 5HT produc- tion in the brain from carbohydrate-rich diets can induce mood-enhancing post-ingestion effects that motivate intake of such foods and, consequently, promote weight gain [14]. Circulating leptin interacts with peripheral 5HT and decreases appetite [3]. One mouse model study reported that plasma leptin was reduced by 5HT, and 5HT exerted a direct effect on adipocytes and regulated leptin release from adipocytes [15]. Moreover, 5HT is able to down-regulate adiponectin in the mouse adipocyte cell line [27]. Peripheral 5HT concentration of the obese group is significantly lower than the lean group in this study, and this is similar with human study results. Conclusions 5HT is an important appetite control neurotransmitter, but there are limited studies regarding 5HT levels related to obesity in veterinary medicine. Overall, we found that peripheral 5HT levels were lower in obese dogs than lean dogs, and 5HT was negatively correlated with BCS and leptin levels. To the best of our knowledge, this is the first study to evaluate peripheral 5HT levels in obese dogs. Through this research, we found that the level of peripheral 5HT is low in the obese group. Since 5HT is related to intestinal mobility, we can expect that a lower level of 5HT could reduce intestinal mobility. The reduced mobility, in turn, could allow gut microorganism to undergo energy harvesting for a longer period, which consequentially could aggravate obesity. Based on this Competing interests The authors declare that they have no competing interests. Discussion In human medicine, high levels of 5HT may be dangerous, and is known as serotonin syndrome [34]. Unfortunately, there have been few studies on the relationship between obesity and 5HT in the veterinary field so far. As obesity is considered to cause several diseases, based on the results of our research, more studies are needed to determine the impact and the mechanism of 5HT on leptin and adiponectin. Furthermore, as a treatment option for obesity in veterinary medicine, clinical trials of a 5HT agonist and considerations of adverse effects should be pursued. into the brain [14]. After eating carbohydrates, insulin is released, which promotes uptake of the LNAAs, but not tryptophan, into skeletal muscles. This helps tryptophan pass more easily into the brain, which increases serotonin production in the brain [14,25]. Increased 5HT produc- tion in the brain from carbohydrate-rich diets can induce mood-enhancing post-ingestion effects that motivate intake of such foods and, consequently, promote weight gain [14]. Circulating leptin interacts with peripheral 5HT and decreases appetite [3]. One mouse model study reported that plasma leptin was reduced by 5HT, and 5HT exerted a direct effect on adipocytes and regulated leptin release from adipocytes [15]. Moreover, 5HT is able to down-regulate adiponectin in the mouse adipocyte cell line [27]. Peripheral 5HT concentration of the obese group is significantly lower than the lean group in this study, and this is similar with human study results. Enterochromaffin (EC) cells in the intestinal epithelium release 5HT according to mechanical stimulation, to promote transit [28]. Experimentally a diet-induced obesity model showed decreased 5HT levels with a decreased number of EC cells [11]. The inflammation associated with changes in the GI microbiota is considered as the reason for decreased 5HT availability in obese status [28,29]. 5HT, as a neurotransmitter, controls food satiety, and, therefore, high 5HT concentrations decrease leptin and adiponectin concentrations. We found that 5HT is negatively correlated to leptin. Based on the previous result of a mouse model that 5HT could reduce the secretion of leptin, we can consider the possibility that lowered 5HT failed to properly suppress increasing secretion of leptin [15]. As seen in Figure 3C, however, there are more dogs presenting a high level of 5HT at a low level of leptin than ones presenting a low level of 5HT at a high level of leptin. Discussion Therefore, further studies would be needed with a larger number of dogs to evaluate the interaction between 5HT and leptin. The low level of adiponectin in the obese group was similar to the results of previous studies [30,31]. Adiponectin was negatively correlated with obesity [32]. The decreased adiponectin level in obesity is more significant in visceral than sub- cutaneous adiposity in humans, and the composition of adiponectin also changes with location in the body. 5HT suppresses adiponectin, and, therefore, a high peripheral adiponectin level can induce a low 5HT level. However, we observed low levels of both 5HT and adiponectin in the obese group in our study. There are a few limitations to this study. First of all, concerning the selection of the dogs in this study, the distribution of dogs does not balance between intact and castrated males. For example, there was a relatively fewer number of intact males than castrated males in the obese group. In addition, even though we identified several corrections among adipokines, 5HT, and other obesity- related parameters, we did not go further to formulate the mechanism of interactions among them. Although we found a significant difference in 5HT levels between the lean and obese groups, there was also variation within the groups. We tried to control for factors that influence 5HT levels such as platelet, diarrhea and food. However, we could not control emotion, social situation or the rhythm activity of an enrolled subject. References 1. Lund EM, Armstrong PJ, Kirk CA, Klausner JS: Prevalence and risk factors for obesity in adult cats from private US veterinary practices. Int J Appl Res Vet M 2005, 3:88–96. 28. Bertrand RL, Senadheera S, Tanoto A, Tan KL, Howitt L, Chen H, Murphy TV, Sandow SL, Liu L, Bertrand PP: Serotonin availability in rat colon is reduced during a western diet model of obesity. Am J Physiol Gastrointest Liver Physiol 2012, 303:424–434. 2. Markwell PJ, Butterwick RF, Wills JM, Raiha M: Clinical studies in the management of obesity in dogs and cats. Int J Obes Relat Metab Disord 1994, 18:39–43. 29. Maslowski KM, Vieira AT, Ng A, Kranich J, Sierro F, Yu D, Schilter HC, Rolph MS, Mackay F, Artis D, Xavier RJ, Teixeira MM, Mackay CR: Regulation of inflammatory responses by gut microbiota and chemoattractant receptor GPR43. Nature 2009, 461:1282–1286. 3. Leibowitz SF, Alexander JT: Hypothalamic serotonin in control of eating behavior, meal size, and body weight. Biol Psychiatry 1998, 44:851–864. 3. Leibowitz SF, Alexander JT: Hypothalamic serotonin in control of eating behavior, meal size, and body weight. Biol Psychiatry 1998, 44:851–864. behavior, meal size, and body weight. Biol Psychiatry 1998, 44:851–864. 4. Hand MS, Thatcher CD, Remillard RL, Roudebush P, Novotny BJ: Small Animal Clinical Nutrition. 5th edition. Topeka: Mark Morris Institute; 2010. 4. Hand MS, Thatcher CD, Remillard RL, Roudebush P, Novotny BJ: Small Animal Clinical Nutrition. 5th edition. Topeka: Mark Morris Institute; 2010. p 30. Hu E, Liang P, Spiegelman BM: AdipoQ is a novel adipose-specific gene dysregulated in obesity. J Biol Chem 1996, 271:10697–10703. 5. Clarke SP, Mellor D, Clements DN, Gemmill T, Farrell M, Carmichael S, Bennett D: Prevalence of radiographic signs of degenerative joint disease in a hospital population of cats. Vet Rec 2005, 157:793–799. 31. Jeusette IC, Lhoest ET, Istasse LP, Diez MO: Influence of obesity on plasma lipid and lipoprotein concentrations in dogs. Am J Vet Res 2005, 66:81–86. 6. Lund EM, Armstrong PJ, Kirk CA, Klausner JS: Prevalence and risk factors for obesity in adult dogs from private US veterinary practices. Int J Appl Res Vet M 2006, 4:177–186. 32. Ishioka K, Omachi A, Sagawa M, Shibata H, Honjoh T, Kimura K, Saito M: Canine adiponectin: cDNA structure, mRNA expression in adipose tissues and reduced plasma levels in obesity. Res Vet Sci 2006, 80:127–132. 7. Towell LT: Practical Weight Management In Dogs And Cats. Blackwell: UK; 2011. 33. Received: 26 November 2013 Accepted: 8 May 2014 Published: 13 May 2014 27. Uchida-Kitajima S, Yamauchi T, Takashina Y, Okada-Iwabu M, Iwabu M, Ueki K, Kadowaki T: 5-Hydroxytryptamine 2A receptor signaling cascade modulates adiponectin and plasminogen activator inhibitor 1 expression in adipose tissue. FEBS Lett 2008, 582:3037–3044. Authors’ contributions HJP, KWS and KHS participated in the conception and design of this study. HJP and SEL participated in sample collection, laboratory testing and drafting of article. JHO, KWS and HJP participated in statistical analysis. KWS, KHS, and SEL participated in critical revisions. All authors read and approved the final manuscript. Page 8 of 8 Page 8 of 8 Park et al. BMC Veterinary Research 2014, 10:113 http://www.biomedcentral.com/1746-6148/10/113 Author details 1 Author details 1Laboratory of Veterinary Internal Medicine, College of Veterinary Medicine, Chungnam National University, Daejeon 305-764, South Korea. 2Division of Malaria & Parasitic Diseases, Korea National Institute of Health, Korea Centers for Disease Control and Prevention, Osong-eup, Cheongwon-gun, Chungbuk 363-951, South Korea. 3Department of Statistics, Chung-Ang University, Seoul 156-756, South Korea. 23. Ford RB: Idiopathic hyperchylomicronemia in miniature schnauzers. J Small Anim Pract 1993, 34:488–492. 24. Gonazles GF, Carrillo C: Blood serotonin levels in postmenopausal women: effects of age and serum oestradiol levels. Maturitas 1993, 17:23–29. 25. Fuemmeler BF, Agurs-Collins TD, McClernon FJ, Kollins SH, Kail ME, Bergen AW, Ashley-Koch AE: Genes implicated in serotonergic and dopaminergic functioning predict BMI categories. Obesity 2008, 16:348–355. 26. Zhang X, Beaulieu JM, Sotnikova TD, Gainetdinov RR, Caron MG: Tryptophan hydroxylase-2 controls brain serotonin synthesis. Science 2004, 305:217. References Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: Submit your next manuscript to BioMed Central and take full advantage of: 16. Aronne LJ, Thornton-Jones ZD: New targets for obesity pharmacotherapy. Clin Pharmacol Ther 2007, 81:748–752. 17. Laflamme D: Development and validation of a body condition score system for dogs. Canine Prac 1997, 22:10–15. 17. Laflamme D: Development and validation of a body condition score system for dogs. Canine Prac 1997, 22:10–15. • Convenient online submission • Thorough peer review 18. Lacerda CM, Maclea HB, Kisiday JD, Orton EC: Static and cyclic tensile strain induce myxomatous effector proteins and serotonin in canine mitral valves. J Vet Cardiol 2021, 14:223–230. 18. Lacerda CM, Maclea HB, Kisiday JD, Orton EC: Static and cyclic tensile strain induce myxomatous effector proteins and serotonin in canine mitral valves. J Vet Cardiol 2021, 14:223–230. 19. Eirmann LA, Freeman LM, Laflamme DP, Michel KE, Satyaraj E: Comparison of adipokine concentrations and markers of inflammation in obese versus lean dogs. Int J Appl Res Vet M 2009, 7:196–205. 19. Eirmann LA, Freeman LM, Laflamme DP, Michel KE, Satyaraj E: Comparison of adipokine concentrations and markers of inflammation in obese versus lean dogs. Int J Appl Res Vet M 2009, 7:196–205. 20. Pan W, Kastin AJ, Banks WA, Zadina JE: Effects of peptides: a cross-listing of peptides and their central actions published in the journal peptides from 1994 through 1998. Acknowledgments Thi k 21. Laflamme DP, Kuhlman G, Lawler DF: Evaluation of weight loss protocols for dogs. J Am Anim Hosp Assoc 1997, 33:253–259. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean Government (MOE) (NRF-2012R1A1A2041697). 22. Root MV, Johnston SD, Olson PN: Effect of prepuberal and postpuberal gonadectomy on heat production measured by indirect calorimetry in male and female domestic cats. Am J Vet Res 1996, 57:371–374. References Taylor JR, Dietrich E, Powell J: Lorcaserin for weight management. Diabetes Metab Syndr Obes 2013, 6:209–216. 8. Lund EM, Armstrong PJ, Kirk CA, Kolar LM, Klausner JS: Health status and population characteristics of dogs and cats examined at private veterinary practices in the United States. J Am Vet Med Assoc 1999, 214:1336–1341. 34. Dunkley EJ, Isbister GK, Sibbritt D, Dawson AH, Whyte IM: The hunter serotonin toxicity criteria : simple and accurate diagnostic decision rules for serotonin toxicity. QJM 2003, 96:635–642. 9. Perez Alenza MD, Peña L, del Castillo N, Nieto AI: Factors influencing the incidence and prognosis of canine mammary tumours. J Small Anim Pract 2000, 41:287–291. doi:10.1186/1746-6148-10-113 Cite this article as: Park et al.: Leptin, adiponectin and serotonin levels in lean and obese dogs. BMC Veterinary Research 2014 10:113. 10. Hodge S, Bunting BP, Carr E, Strain JJ, Stewart-Knox BJ: Obesity, whole blood serotonin and sex differences in healthy volunteers. Obes Facts 2012, 5:399–407. 11. Wynne K, Stanley S, McGowan B, Bloom S: Appetite control. J Endocrinol 2005, 184:291–318. 12. Radin MJ, Sharkey LC, Holycross BJ: Adipokines: a review of biological and analytical principles and an update in dogs, cats, and horses. Vet Clin Pathol 2009, 38:136–156. 13. Duffy ME, Stewart-Knox BJ, McConville C, Bradbury I, O'Connor J, Hel & Strain JJ: The relationship between whole-blood serotonin and subjective mood in apparently healthy postmenopausal women. Biol Psychol 2006, 73:165–168. 14. Horácek J, Kuzmiaková M, Höschl C, Andĕl M, Bahbonh R: The relationship between central serotonergic activity and insulin sensitivity in healthy volunteers. Psychoneuroendocrinology 1999, 24:785–794. 15. Stunes AK, Reseland JE, Hauso O, Kidd M, Tommeras K, Waldum HL, Gustafsson BI: Adipocytes express a functional system for serotonin synthesis, reuptake and receptor activation. Diabetes Obes Metab 2011, 13:551–558. References Peptides 1999, 20:1127–1138. 20. Pan W, Kastin AJ, Banks WA, Zadina JE: Effects of peptides: a cross-listing of peptides and their central actions published in the journal peptides from 1994 through 1998. Peptides 1999, 20:1127–1138.
https://openalex.org/W2511146801	https://www.ccsenet.org/journal/index.php/gjhs/article/download/62191/33422	English	null	Maternal Role Adaptation Scale in Neonatal Intensive Care Units (MRAS: NICU): Development, Validation and Psychometric Tests	Global journal of health science	2,016	cc-by	4,958	Abstract Background: Maternal role adaptation involves conceptualization and establishment of a responsible maternal role, which is characterized by a new identity and formation of mothering behaviors. Becoming a mother in intensive care unit is very different from becoming a mother with a term infant at home. The aim of the study was to develop a valid and reliable tool for assessment of maternal role adaptation of mothers with preterm neonates admitted to neonatal intensive care units. Methods: This was an exploratory study which was conducted in 2 phase of qualitative and quantitative. A 90-item scale was developed after semi-structured interviews with 25 mothers and review of literature. After merging the similar items, it reduced to 45-item scale. Validity was determined through assessment of face, content and constructs validities, and reliability was confirmed through internal consistency and test-retest. Results: Face validity led to elimination of 2 items, and further 8 items were eliminated through content validity index with cut-off point 0.79 and content validity ratio with cut-off point 0.42. Thus, the number of items reduced to 35-item. In exploratory factor analysis, 6 factors were identified that explained 54% of the variance. Construct validity led to elimination of 3 other items, and the final scale was developed with 32 items. Cronbach’s alpha and intra-class correlation coefficient were 0.77 and 0.81 respectively. Conclusion: The 32-item “Maternal role adaptation scale in mothers with preterm neonates admitted to neonatal intensive care units” (MRAS: NICU) is a valid and reliable tool. Keywords: maternal role adaptation, Neonatal Intensive Care Unit, preterm neonate, reliability, tool development, validity Maternal Role Adaptation Scale in Neonatal Intensive Care Units (MRAS: NICU): Development, Validation and Psychometric Tests Maternal Role Adaptation Scale in Neonatal Intensive Care Units (MRAS: NICU): Development, Validation and Psychometric Tests Sousan Heydarpour1, Zohreh Keshavarz2, Maryam Bakhtiari3 & Farid Zayeri4 Sousan Heydarpour1, Zohreh Keshavarz2, Maryam Bakhtiari3 & Farid Zayeri4 1 Student Research Office, School of Nursing & Midwifery, Shahid Beheshti University of Medical Sciences, Tehran, Iran 1 Student Research Office, School of Nursing & Midwifery, Shahid Beheshti University of Medical Sciences, Tehran, Iran 2 Department of Midwifery and Reproductive Health, School of Nursing & Midwifery, Shahid Beheshti University of Medical Sciences, Tehran, Iran 2 Department of Midwifery and Reproductive Health, School of Nursing & Midwifery, Shahid Behe of Medical Sciences, Tehran, Iran 3 Department of Clinical Psychology, Taleghani Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran ent of Clinical Psychology, Taleghani Hospital, Shahid Beheshti University of Medical Sciences n 4 Proteomics Research Center, Faculty of Paramedical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran Correspondence: Zohreh Keshavarz, Department of Midwifery and Reproductive Health, School of Nursing & Midwifery, Shahid Beheshti University of Medical Sciences, Tehran, Iran. Tel: 98-218-820-2512. E-mail: z.keshavarz@sbmu.ac.ir; keshavarzzohre@yahoo.com Received: June 28, 2016 Accepted: August 2, 2016 Online Published: August 15, 2016 doi:10.5539/gjhs.v9n4p115 URL: http://dx.doi.org/10.5539/gjhs.v9n4p115 Received: June 28, 2016 Accepted: August 2, 2016 Online Published: August 15, 2016 doi:10.5539/gjhs.v9n4p115 URL: http://dx.doi.org/10.5539/gjhs.v9n4p115 Global Journal of Health Science; Vol. 9, No. 4; 2017 ISSN 1916-9736 E-ISSN 1916-9744 Published by Canadian Center of Science and Education Global Journal of Health Science; Vol. 9, No. 4; 2017 ISSN 1916-9736 E-ISSN 1916-9744 Published by Canadian Center of Science and Education 2010). Becoming a mother in intensive care unit is very different from becoming a mother with a term infant at home. Admission of an infant to neonatal intensive care unit is a life-changing event for the mother and members of the family (Devellis, 2011). According to the theory of assuming maternal role or motherhood, a mother whose child has been admitted to NICU has little opportunity for looking after and interacting with her child, and these delays assuming her maternal role (Mercer, 2004). Today, a number of service providers use depression assessment tools such as Edinburgh Postnatal Depression Scale to screen mothers for postpartum depression. Many of these interventions are performed not just to assess depression, but also to evaluate women’s satisfaction with the role or experience of motherhood. Tools that measure mood alone do not provide any information about women’s experience of motherhood, and it is wrong to assume that poor mood as measured by Edinburgh Postnatal Depression Scale reflects dissatisfaction with maternal role. Hence, interventions aimed at increasing women’s satisfaction with motherhood need to use a tool that is sensitive to this area (Matthey, 2011; Javadifar, 2012). Development of a quantitative tool that can identify the need for help in certain circumstances and assess interventions performed in a particular situation is highly important, and can enhance our understanding of maternal-child relationship and maternal role and help improve the quality of postpartum primary cares. As we gain more knowledge about development, trend, role and experience of motherhood, it is more likely to recognize changed relationships and also improve health interventions (Javadifar, 2012). One of the important and challenging parts of health care is assessment of mother’s ability to take care of her child and also to properly and easily adapt to her maternal role. Identifying mothers who have problematic care behaviors in caring for their children, or have negative emotions and thoughts about motherhood, and their referral to specialists is essential and requires proper and valid assessment. Often, maternal and child healthcare providers are unable to carry out this assessment as they do not have the knowledge of mothering experience and there is no reliable and objective tool to measure components of motherhood. 2. Materials and Methods The present methodological study was conducted with the aim to develop a questionnaire (and examine its psychometric features) for the measurement of maternal role adaptation of mothers with preterm neonates admitted to neonatal intensive care units according to stages proposed by Schneider et al. (Schneider, Elliot, & LoBiondo-Wood, 2004). 2010). Hence, inclusion of such assessments in the set of postpartum care can have a major role in promotion of maternal health, appropriate development of infants, and improvement of quality of judgment of maternal and child service providers (Fowels & Horowitz, 2006). Questionnaires such as Parental Stressor Scale: Neonatal Intensive Care Unit (PSS: NICU) (Miles, Funk, & Carlson, 1993), The perceived maternal parenting self-efficacy questionnaire (PMP-S-E) (Barnes & Adamson-Macedo, 2007), Barkin Index of Maternal Functioning (Barkin et al., 2010), Cognitive Adaptation of stressful events scale (Affonso, Mayberry, Lovett, & Paul, 1994), and The Being A mother scale (Matthey, 2011) have already been developed. However, in the search conducted, no tool was found that directly measured maternal role adaptation of mothers with preterm neonates. Thus, in the absence of such a tool, it is essential to develop one. Hence, it was decided to extract the basic structure of such a tool by identifying components and factors affecting maternal role adaptation of Iranian women with preterm neonates admitted to NICU, and thus develop a suitable tool according to Iranian culture. 1. Introduction Mothering is the most important role for a woman (Haddadi, Chaldi, Sajjadi, & Lehi, 2011). Becoming a mother leads to adaptation of major changes in cognitive, emotional-social and behavioral functions (Shin & White-Traut, 2007) .Such psychological changes can be affected by the woman’s particular circumstances, beliefs and attitudes, socioeconomic status, readiness and knowledge, and also her social and mental conditions, and the more developed a person is in this area, the better her personal adaptation will be (Mercer, 2004). Maternal role adaptation involves conceptualization and establishment of a responsible maternal role, which is characterized by a new identity and formation of mothering behaviors (Mercer, 2004). Mothers who fail to suitably adapt to their maternal role will perform poorly in this role, especially in mother-child relationship (Mercer & Ferketich, 1995). In fact, a mother can take care of her child, only when she accepts her maternal role (Bailey, 115 Global Journal of Health Science Vol. 9, No. 4; 2017 gjhs.ccsenet.org 2.1 Stage One In this stage, the concept of maternal role adaptation of mothers with preterm neonates admitted to NICU was extracted according to their experiences using a qualitative conventional content analysis study and in-depth semi-structured interviews with mothers of preterm neonates who were born before the 36th gestational week and had been admitted to NICU for at least one week. These mothers were Iranian, married and lived with their spouses, and had no previous history of delivering a preterm neonate that was admitted to NICU, no physical or psychological problems, no neonates with survival-limiting diseases or congenital abnormalities, and also had the willing to participate in this study; were recruited. Participants were primiparus and multiparus mother who had preterm neonates hospitalized in neonatal intensive care units and selected according to purposive sampling method. Ultimately data saturation was obtained after 25 interviews. Study setting consisted of a hospital affiliated to Social Security Organization, and other hospitals affiliated to Kermanshah University of Medical Sciences in the city of Kermanshah. With the permission of participants, interviews were recorded, and then immediately transcribed. Data were analyzed according to content analysis. Codes were extracted and categorized, and main 116 Global Journal of Health Science Vol. 9, No. 4; 2017 gjhs.ccsenet.org themes were identified. In the present study, codes were the same as variables or items in each theme. 2.2 Stage Two hemes were identified. In the present study, codes were the same as variables or items in each theme. Items of the questionnaire were developed according to codes extracted from a qualitative study. Then, based on the extracted concept, main constructs of the questionnaire were determined, including interaction, self-efficacy, uncertainty, participation in care, and remote mothering. Review of literature was also used in extraction of themes. The questionnaire was designed as a self-reporting scale containing closed questions. Psychometric properties of the questionnaire were determined. Qualitative and quantitative methods were used in determining face validity. In qualitative method, 10 mothers were interviewed face-to-face, and issues such as levels of difficulty, suitability, and ambiguity were assessed. Next, in quantitative assessment of item impact, 10 mothers were asked to identify the importance of each item, and thus item-impact score was found for each item. Items scoring less than 1.5 were eliminated. In qualitative content validity, 10 experts assessed the questionnaire in terms of grammar, appropriate use of words, necessity, importance, appropriate placement of items, and scoring. 2.5 Ethical Considerations Participants were informed of study objectives, and were assured of confidentiality and withdrawal at any stage. Informed written consents were obtained from all participants. The present study was approved by the university ethics committee. 2.4 Stage Four Reliability of the questionnaire was determined using internal consistency and stability. Internal consistency was found through Cronbach’s alpha in a sample of 200 mothers. Stability was assessed through test-retest with 3 days interval in a sample of 20 mothers, and scores obtained in these tests were compared using ICC test. 2.5 Ethical Considerations 2.5 Ethical Considerations 2.5 Ethical Considerations 2.1 Stage One In quantitative content validity, Content Validity Ratio (CVR) and Content Validity Index (CVI) were used. To determine CVR, 20 experts reviewed each item. Based on Lawshe’s table, items with CVRs higher than 0.42 were retained (Polit & Beck, 2013). CVI was assessed according Waltz and Bausell’s Content Validity Index. To this end, the questionnaire was made available to 20 experts to use Waltz and Bausell’s Content Validity Index to determine relevance of each item using a 4-point Likert scale. Simplicity and clarity of each item were also assessed. According to Polit & Beck, CVI score higher than 0.79 is favorable, between 0.7 and 0.79 is debatable and needs modification, and less than 0.7 is a candidate for elimination (Polit & Beck, 2013). Construct validity was determined using exploratory factor analysis. Total five samples per each item of questionnaire (35 item) was considered which was about 160 mothers, and ultimately 200 mothers were selected. In determining construct validity through exploratory factor analysis, Kaiser-Meyer-Olkin (KMO) test and Bartlett’s Test (BT) were used. Items with factor loading less than 0.3 were eliminated, and elimination or preservation of items with factor loading between 0.3 and 0.5 was decided by research team. 2.4 Stage Four 3. Results In stage one, the concept of maternal role adaptation of mothers with preterm neonates admitted to NICU was defined by conducting a qualitative study. The results obtained showed that adaptation to maternal role in these mothers is a multidimensional concept that means growth, development, and self-efficacy, which generates less uncertainity confidence in mothers for participation in child care, proper interaction and distant mothering. In stage two, 90- items were developed through a qualitative study and review of literature. Then, overlapping items were integrated in meetings with research team, thereby reducing the number of items to 45. Items were arranged in 5 subthemes including distant mothering, self-efficacy, interaction, uncertainty, and participation in care. Scoring was based on a 5-point Likert scale from totally agree, agree, no comment, disagree, to totally disagree. In stage three, a total of 10 items were eliminated; 2 items due to impact score less than 1.5, 3 items due to CVR<0.42, and 5 items due to CVI<0.79. Content validity index for the whole questionnaire was 0.93. In the next stage, factor analysis was carried out on the remaining 35 items, and the results showed that the highest percentage of total variance (54%) is explained by the first 6 factors. KMO was 0.86, and BT was significant (P<0.001). Items with factor loading less than 0.3 were eliminated, and elimination or preservation of items with factor loading between 0.3 and 0.5 was decided by research team. Research team decided to eliminate items 9 and 10 with factor loading between 0.3 and 0.5, and item 21 was also eliminated because it could not be associated with 117 Global Journal of Health Science Vol. 9, No. 4; 2017 gjhs.ccsenet.org any of the factors. Accordingly, the first factor (participation in care) included 14 items (8, 19-20, 22-30 and 34-35), the second factor (self-efficacy) included 6 items (1-4, and 11-12), the third factor (distant mothering) included 3 items (31-33), the fourth (uncertainty) included 4 items (7, and 13-15), the fifth (interaction) included 3 items (16-18), and the sixth (growth and development) included 2 items (5-6) (Table 1). Finally, a 32-item questionnaire was developed. was developed. Table 1. Factor loading of each item according to varimax factor rotation of “MRAS: NICU” Factor/Item 1 2 3 4 5 6 1. I have matured since I became a mother. 0.539 2. I have become more patient since I became a mother. 0.629 3. 4. Discussion “Maternal Role Adaptation Scale in Neanatal Intensive Care Units (MRAS: NICU)” was developed according to the explained concept in qualitative study with 32 items in 6 dimensions. Face, content, and construct validities and internal consistency and stability were all confirmed. It is worth noting that application of the above questionnaire is relatively easy and can be completed by mothers of preterm neonates admitted to NICU in about 10 minutes. MRAS: NICU was developed based on qualitative part of this study and review of literature which contained 45 items in 5 dimensions, including distant mothering, self-efficacy, interaction, uncertainty, and participation in care. In face validity stage, 2 items were eliminated for having item impacts less than 1.5, thus reducing the number of items to 43. Face validity of the scale was therefore confirmed. Content validity was performed by 20 experts. According to Polit & Beck, 3 experts are sufficient in determination of content validity (Polit, Beck, & Owen, 2007), but others believe that between 15 and 20 experts are needed (Burns & Grove, 2008). In a study by Javadifar, a panel of 10 experts was used in determining content validity (Javadifar, 2012). Given the complex dimensions of Maternal role adaptation in mothers of preterm infants admitted to NICU, use of greater numbers of experts from different fields seems essential. In the present study, content validity was confirmed with acceptable content validity index and content validity ratio. Assessment of CVR and CVI led to elimination of 8 items, and CVI for the whole scale was found 0.93, which is within the acceptable range (Polit & Beck, 2006), and confirms content validity of the questionnaire. Javadifar reported content validity of 0.95 for “Adaptation to mothering role in primiparous mothers of term infants” questionnaire (Javadifar, 2012). To determine construct validity through exploratory factor analysis, KMO and BT tests were used. Javadifar reported KMO of 0.67. This index varies from 0 to 1, and higher values indicate better factor analysis. Values less than 0.5 are considered unacceptable, from 0.5 to 0.7 moderate, 0.7 to 0.8 balanced, 0.8 to 0.9 desirable, and higher than 0.9 excellent factor analysis (Munro, 2005). In the present study, KMO was 0.86 for all constructs, indicating adequacy of sample size and favorable factor analysis. Kruit-Bartlett test was also significant (P<0.001), which confirms adequacy of the model. 3. Results 0.615 Cronbach’s alpha was 0.77 for the whole questionnaire, and between 0.61 and 0.92 for different factors. In retest method, ICC was 0.81 for the whole questionnaire and between 0.54 and 0.88 for different factors (Table 2). Table 2. Cronbach’s alpha and test-retest scores of factors and the whole scale Factors Cronbach’s α Test-retest Reliability(ICC) participation in care 0.92 0.73 self-efficacy 0.74 0.54 distant mothering 0.63 0.64 uncertainty 0.66 0.84 interaction 0.61 0.88 growth and development 0.61 0.69 Whole Questionnaire 0.77 0.81 4 Discussion mothers to look after him/her. 34. I follow my child’s condition. 0.582 0.558 35. I search for information about child care. 0.615 Cronbach’s alpha was 0.77 for the whole questionnaire, and between 0.61 and 0.92 for different factors. In retest method, ICC was 0.81 for the whole questionnaire and between 0.54 and 0.88 for different factors (Table 2). pha was 0.77 for the whole questionnaire, and between 0.61 and 0.92 for different factors. In retest was 0.81 for the whole questionnaire and between 0.54 and 0.88 for different factors (Table 2). Table 2. Cronbach’s alpha and test-retest scores of factors and the whole scale Factors Cronbach’s α Test-retest Reliability(ICC) participation in care 0.92 0.73 self-efficacy 0.74 0.54 distant mothering 0.63 0.64 uncertainty 0.66 0.84 interaction 0.61 0.88 growth and development 0.61 0.69 Whole Questionnaire 0.77 0.81 Table 2. Cronbach’s alpha and test-retest scores of factors and the whole scale 3. Results I have become more hopeful since I became a mother. 0.587 4. I have become more independent since I became a mother. 0.696 5. I have become more active since I became a mother. 0.411 6. I have become more committed and responsible since I became a mother. 0.443 7. I regret having a baby. 0.782 8. I have no problem embracing my child. 0.453 9. Taking care of my child is easy. 0.442 10. Seeing other infants in the ward (NICU) makes me hopeful. 0.433 11. I become more adept at taking care of my child as time goes by. 0.556 12. Becoming a mother has enhanced my self-confidence. 0.592 13. I am not sure of my child’ health in future. 0.675 14. I am not sure if I can breastfeed my child. 0.442 15. I am not sure if I can take care of my child. 0.730 16. I have a good relationship with my child’s physician. 0.691 17. I have a good relationship with nurses about child care. 0.674 18. My relationship with my husband has improved since I became a mother. 0.441 19. I have a good relationship with my child. 0.733 20. I feel close to my child. 0.813 21. I change my child’s pampers 22. I massage my child. 0.702 23. I caress my child. 0.816 24. I cuddle my child. 0.796 25. I like to stay with my child in the ward. 0.758 26. I enjoy to breastfed my child. 0.657 27. I frequently visit my child. 0.657 28. I talk to my child. 0.722 29. I try to stay with my child for as long as possible. 0.655 0.417 30. When I’m away from my child, I cannot stop thinking about him/her. 0.691 31. When I’m away from my child, I talk to him/her from afar. 0.527 32. When I’m away from my child, I am concerned about his/her care. 0.724 33. When I’m away from my child, I ask nurses and other 0.542 r loading of each item according to varimax factor rotation of “MRAS: NICU” Table 1. Factor loading of each item according to varimax factor rotation of “MRAS: NICU” 118 Global Journal of Health Science Vol. 9, No. 4; 2017 gjhs.ccsenet.org mothers to look after him/her. 34. I follow my child’s condition. 0.582 0.558 35. I search for information about child care. 4. Discussion In initial design of the scale, items were categorized in 5 dimensions, including self-efficacy, uncertainty, interaction, participation in care, and distant mothering. The results from factor analysis showed a close match between participation in care and factor one, self-efficacy and factor two, distant mothering and factor three, uncertainty and factor four, and interaction and factor five. A number of items in self-efficacy dimension were placed in a sixth dimension, namely growth and development. Hence, construct validity of the scale was confirmed through the agreement found between items of factors resulting from factor analysis and definition and dimensions of adaptation to mothering role in mothers of preterm neonates admitted to NICU. To determine construct validity, exploratory factor analysis was used by Javadifar for “Adaptation to maternal role in primiparous mothers of term 119 Global Journal of Health Science Vol. 9, No. 4; 2017 gjhs.ccsenet.org infants” questionnaire, Matthey for “The being a mother” scale, and by Barnes-Adamson-Macedo for “The perceived maternal parenting self-efficacy questionnaire (PMP-S-E)” (Barnes & Adamson-Macedo, 2007; Matthey, 2011; Javadifar, 2012). infants” questionnaire, Matthey for “The being a mother” scale, and by Barnes-Adamson-Macedo for “The perceived maternal parenting self-efficacy questionnaire (PMP-S-E)” (Barnes & Adamson-Macedo, 2007; Matthey, 2011; Javadifar, 2012). In the present study, reliability was determined through internal consistency (Cronbach’s alpha) and stability (retest). Cronbach’s alpha was reported 0.76 in Javadifar study, 0.79 in Matthey study, 0.91 in Barnes-Adamson-Macedo study, and 0.87 in Barkin study (Barnes & Adamson-Macedo, 2007; Barkin et al. 2010; Matthey, 2011; Javadifar, 2012). Cronbach’s alpha of 0.7 and higher are the standard for reliability of a tool. However, alpha greater than 0.6 is acceptable in descriptive studies (Hajizadeh & Haji Asghari, 2011). In the present study, Cronbach’s alpha was found 0.77, which confirms reliability of the questionnaire. In the present study, reliability of the questionnaire was confirmed with ICC of 0.81. Javadifar reported ICC of 0.83 (Javadifar, 2012). Ma et al. believe that correlation higher than 0.75 shows good consistency, between 0.5 and 0.75 shows moderate consistency, and less than 0.5 shows poor consistency (Ma, Yam, Tsui, & Yau, 2006). According to Polit & Beck, 0.7 is acceptable correlation coefficient for the whole scale and higher than 0.4 is acceptable for subscales (Polit & Beck, 2008). Hence, reliability coefficients found for the questionnaire in the present study agrees with those in statistics literature and other studies. 6. Limitations The first limitation is that researchers used purposive sampling for the first stage of the study. Furthermore, in the present study, participants were from one particular geographical location, which limits generalizability of results to populations with different cultures. Study limitations included lack of concurrent assessment of validity. Thus, it is recommended that future studies consider this kind of validity. Another limitation is that we did not perform the confirmatory factor analysis for the last stage of validation because collecting the sample size for this part took a long time and researchers plan to perform it in future. Acknowledgements This article is part of a PhD thesis in Reproductive health from the University of Shahid Beheshti, Tehran, Iran. The study was approved by the Ethics Committee of the University of Shahid Beheshti under the code SBMU2.REC.1394.41. Our sincere appreciation goes to all the participants in this study. 5. Conclusion In the present study, maternal role adaptation scale in mothers with preterm neonates admitted in neonatal intensive care units (MRAS: NICU) was developed and adapted for Iranian culture. With features such as development based on understanding the concept of maternal role in mothers with preterm neonates through use of in-depth and qualitative research and review of literature, simple scoring, favorable reliability and validity and application by health service providers including nurses, the scale developed can be used for screening and assessment of maternal role adaptation in mothers with preterm neonates. 4. Discussion Burns and Grove consider two weeks to one month an appropriate interval between two tests (Burns & Grove, 2008). Barnes & Adamson-Macedo used retest method with 10 days interval to determine external consistency of “Perceive Maternal Parenting Self-Efficacy”. Spearman’s correlation coefficient showed a good correlation between these times (r=0.96, P<0.01) (Barnes & Adamson-Macedo, 2007). In a study by Javadifar, ICC was found 0.833 with two weeks interval. In Matthey study, reliability of the tool used in a test-retest on 47 mothers was reported 0.52 at 6 weeks interval and 0.74 at 2-3 weeks interval (Matthey, 2011). In the present study, a 3-day interval was considered to avoid sample loss. Competing Interests Statement The authors declare that there is no conflict of interests regarding the publication of this paper. Barkin, J. L., Wisner, K. L., Bromberger, J. T., Beach, S. R., Terry, M. A., & Wisniewski, S. R. (2010). References Affonso, D. D., Mayberry, I. J., Lovett, S. M., & Paul, S. (1994). Cognitive adaptation to stressful events during pregnancy and postpartum: development and testing of the case instrument. Nursing Research, 43(6), 338-343. http://dx.doi.org/10.1097/00006199-199411000-00004 Affonso, D. D., Mayberry, I. J., Lovett, S. M., & Paul, S. (1994). Cognitive adaptation to stressful events during pregnancy and postpartum: development and testing of the case instrument. Nursing Research, 43(6), 338-343. http://dx.doi.org/10.1097/00006199-199411000-00004 Affonso, D. D., Mayberry, I. J., Lovett, S. M., & Paul, S. (1994). Cognitive adaptation to stressful events during pregnancy and postpartum: development and testing of the case instrument. Nursing Research, 43(6), 338-343. http://dx.doi.org/10.1097/00006199-199411000-00004 Bailey, S. (2010). Postnatal care: Exploring the views of first-time mothers. Journal of Community Practice, 83(12), 26-9. Barkin, J. L., Wisner, K. L., Bromberger, J. T., Beach, S. R., Terry, M. A., & Wisniewski, S. R. (2010). 120 Vol. 9, No. 4; 2017 Global Journal of Health Science gjhs.ccsenet.org Development of the Barkin Index of maternal functioning. Journal of Women’s Health, 19(12), 2239-2246. http://dx.doi.org/ 10.1089/jwh.2009.1893 Development of the Barkin Index of maternal functioning. Journal of Women’s Health, 19(12), 2239-2246. http://dx.doi.org/ 10.1089/jwh.2009.1893 Barnes, C. R., & Adamson-Macedo, E. N. (2007). Perceived Maternal Parenting Self-Efficacy (PMP S-E) tool: development and validation with mothers of hospitalized preterm neonates. Journal of Advanced Nursing, Dec, 60(5), 550-60. http://dx.doi.org/10.1111/j.1365-2648.2007.04445.x Burns, N., & Grove, S. (2008). The practice of nursing research: Appraisal synthesis and generation of evidence (6th ed.). St Louis, NY: Saunders Elsevier. Devellis, R. F. (2011). Scale development: Theory and applications (3 ed.). Sage Publications. Fowels, E. R., & Horowitz, J. A. (2006). Clinical assessment of mothering during infancy. Journal of Obstetric, Gynecologic & Neonatal Nursing, 35(5), 662-670. http://dx.doi.org/10.1111/j.1552-6909.2006.00090.x Haddadi, M., Chaldi, A., Sajjadi, H., & Lehi, M. (2011). Relationship between occupational class and mental health in women. Journal of Social Welfare, 11(40), 107-27. Hajizadeh, A., & Haji- Asghari, M. (2011). Methods and statistical analysis looking at research method in biology and health sciences (1st ed.). Tehran, NY: Jahad Daneshgahi Publication. Javadifar, N. (2012). Development and psychometric evaluation of an instrument to assess the Iranian first-time mothers’ experiences of maternal role adaptation (Unpublished Phd thesis). Tehran University of Medical Sciences, Tehran, Iran. Matthey, S. (2011). Assessing the experience of motherhood: The being a mother scale (Bam-13). Journal of Affective Disorders, 128(1-2), 142-152. http://dx.doi.org/ 10.1016/j.jad.2010.06.032 Mercer, R. T., & Ferketich, S. L. (1995). Experienced and inexperienced mothers maternal competence during infancy. References Research in Nursing & Health, 18(4), 333-343. http://dx.doi.org/ 10.1002/nur.4770180407 Mercer, R. T. (2004). Becoming a mother versus role attainment. Journal of Nursing Scholarship, 36(3), 226-32. http://dx.doi.org/10.1111/j.1547-5069.2004.04042.x Miles, M. S., Funk, S. G., & Carlson, J. (1993). Parental stressor scale: Neonatal intensive care unit. Nursing Research, 42(3), 148-152. http://dx.doi.org/10.1097/00006199-199305000-00005 Ma, K. H., Yam, K. L., Tsui, K. W., & Yau, F. T. (2006). Internal consistency and test-retest reliability of the Chinese version of the selfreport health–related quality of life measure for children and adolescents with epilepsy. Epilepsy & Behavior, 9(1), 51-7. http://dx.doi.org/10.1016/j.yebeh.2006.04.003 Munro, B. H. (2005). Statistical methods for health care research (4th ed.). Philadelphia, Lippincott, NY: Williams & Wilkins. Polit, D. F., & Beck, C. T. (2006). Nursing research: Method, appraisal and utilization. Philadelphia, NY: Lippincott Williams & Wilkins. Polit, D. F., Beck, C. T., & Owen, S. V. (2007). Is the CVI an acceptable Indicator of content validity? Appraisal and recommendations. Research in Nursing & Health, 30(4), 459-67. http://dx.doi.org/10.1002/nur.20199 Polit, D. F., & Beck, C. T. (2008). Nursing research: Principles and methods. Philadelphia, NY: Lippincott Williams & Wilkins. Polit, D. F., & Beck, C. T. (2013). Essentials of nursing research: Appraising evidence for nursing practice (8th ed.). Philadelphia, NY: Lippincott Williams & Wilkins. Schneider, Z., Elliot, D., & LoBiondo-Wood, G. (2004). Nursing research methods: Critical, appraisal and utilization (2nd ed.). Sydney, NY: Mosby Co Shin, H., & White-Traut, R. (2007). The conceptual structure of transition to motherhood in the neonatal intensive care unit. Journal of Advanced Nursing, 58(1), 90-98. http://dx.doi.org/10.1111/j.1365-2648.2006.04194.x yright for this article is retained by the author(s), with first publication rights granted to the journal. Copyrights Copyright for this article is retained by the author(s), with first publication rights granted to the journal. Copyright for this article is retained by the author(s), with first publication rights granted to the journal. Copyright for this article is retained by the author(s), with first publication rights granted to the journal. This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/). This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/). This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/). 121
https://openalex.org/W3043034162	https://karolinum.cz/data/clanek/8172/PHS_50_2_0139.pdf	Czech	null	100 let od přijetí Ústavní listiny Československé republiky. Historické, politické a právní souvislosti	Právněhistorické studie	2,020	cc-by	915	ZPRÁVY ZPRÁVY ZPRÁVY 2020 (50/2) PRÁVNĚHISTORICKÉ STUDIE PAG. 139–145 100 let od přijetí Ústavní listiny Československé republiky. Historické, politické a právní souvislosti Marka Antoše, Ph.D., LL.M., (PF UK), který přítomným živým způsobem nastínil pozitiva i negativa, historický vývoj, současnou realitu a případné budoucí perspektivy volební povinnosti. Po diskuzi následovala sekce s názvem Československo jako mnohonárodnostní stát, kte- rou otevřel příspěvek doc. PhDr. Zlatice Zudové-Leškové, CSc., (HIÚ), který byl orientován na vztahy Čechů a Slováků a doplnil tak podrobně širší kontext přijímání Ústavní listiny 1920. Následoval příspěvek Mgr. Davida Hubeného, Ph.D., z Národního archivu pojedná- vající o vztahu Podkarpatské Rusi k Ústavě 1920, který byl přednesen v zastoupení. Posléze prezentoval JUDr. PhDr. René Petráš, Ph.D., (PF UK), který ve svém vystoupení rozebral postavení německé menšiny v novém státě. Blok zakončil navazující příspěvek PhDr. Evy Irmanové, CSc., (HIÚ) na téma vztahu maďarské menšiny k československé ústavě. Po skončení obsáhlé diskuze a krátkém občerstvení započal poslední blok pátečního pro- gramu. Ten tematicky doplňoval mezinárodní vlivy související s Ústavní listinou 1920. Prv- ní příspěvek přednesený doc. PhDr. Jiřím Friedlem, Ph.D., DSc., (HIÚ) byl soustředěn na vývoj v Polsku. Navazoval příspěvek PhDr. Miroslava Šeptáka, Ph.D., (Filozofická fakulta Jihočeské univerzity) zaměřený na historické souvislosti vzniku rakouské ústavy z roku 1920. Posledním příspěvkem bloku a zároveň celého prvního dne konference bylo vystou- pení prof. Alaina Soubigou (Université de la Sorbonne, Paříž), které komparativním způso- bem pojednalo o možných vlivech francouzské ústavy na Československou ústavní listinu. Druhý konferenční den, který z důvodu výročí připadl na sobotu dne 29. února, byl orga- nizován méně tradičně, formou kulatého stolu. Účastníky tohoto kulatého stolu byli prof. JUDr. Jan Kuklík, DrSc., (PF UK), prof. PhDr. Ivan Halász, PhD., (MTA TK Jogtudományi Intézete, Budapešť), prof. JUDr. Ladislav Vojáček, CSc., (Právnická fakulta Masarykovy uni- verzity), doc. Mgr. Jaroslav Šebek, Ph.D., (HIÚ) a doc. Dr. Ing. Ján Gronský, CSc., (PF UK). Po úvodní přednášce prof. Kuklíka byl program založen na diskuzních otázkách, které prof. Kuklík pokládal ostatním zúčastněným a tím moderoval vývoj debaty. Prof. Vojáček doplnil některé další aspekty přípravy ústavní listiny, prof. Halász rozvedl politický význam Ústavy 1920. Doc. Šebek pohovořil o vztazích církve ke státu a s tím souvisejících úvahách o ústavní listině. Doc. Grónský pak průřezově shrnul pozdější navazující ústavní vývoj. Druhý konferenční den byl výjimečný nejen z toho důvodu, že poněkud nezvykle při- padl na sobotu, ale také z důvodu avizované účasti řady studentů Právnické fakulty UK, kteří projevili o předmětnou problematiku zájem. Pro přítomné studenty byl organizač- ním týmem pod vedením prof. 100 let od přijetí Ústavní listiny Československé republiky. Historické, politické a právní souvislosti Ve dnech 28. a 29. února 2020 se uskutečnila konference s názvem „100 let od přijetí Ústavní listiny Československé republiky – Historické, politické a právní souvislosti“. Konference byla připomínkou výročí přijetí první československé ústavní listiny a konala se v prostorách Valdštejnského paláce Senátu Parlamentu České republiky ve spolupráci Univerzity Karlovy s Historickým ústavem AV ČR a s podporou Senátu PČR. Uskutečnila se pod záštitou prvního místopředsedy Senátu Jiřího Růžičky, rektora UK Tomáše Zimy a předsedkyně Akademie věd ČR Evy Zažímalové. Konferenci zahájila v pátek 28. února úvodní slova těch, kteří převzali nad akcí záštitu nebo jejich zástupců, a také proslovem jeho excelence Petera Weisse, velvyslance Sloven- ské republiky. Po úvodních projevech následovala sekce s názvem Právní a historické souvislosti ústavy. Sekci zahájil svým vystoupením děkan Právnické fakulty prof. JUDr. Jan Kuklík, DrSc. Barvitě přiblížil politická jednání, která předcházela přijetí Ústavy 1920, a nastínil s tím spojené některé vybrané problémy, názorové rozpory a dohodnuté kompromisy. Širší mezinárodní i vnitrostátní historické souvislosti připojil doc. PhDr. Jan Němeček, DrSc., z Historického ústavu AV ČR. Blok zakončil svým exposé na téma Inspirativní vliv Ústav- ní listiny z roku 1920 na ústavní systém České republiky prof. JUDr. Aleš Gerloch, CSc., z Právnické fakulty UK. Po zajímavé diskuzi k příspěvkům a krátké přestávce byl zahájen další blok s názvem „Vybrané části ústavy v historické retrospektivě do současnosti“. Zahájila jej doc. JUDr. Jana Reschová, CSc., z Právnické fakulty UK svým příspěvkem na téma dynamika konstitucionalismu, ve kterém otevřela řadu otázek týkajících se současného ústavního vývoje. Svými úvahami o fenoménu parlamentarismu navázal JUDr. Ondřej Preuss, Ph.D., (PF UK). Věnoval se srovnání současného ústavního systému s prvorepublikovým mode- lem a hodnotil odlišnosti obou systémů. Blok zakončil JUDr. PhDr. Petr Mlsna, Ph.D., z Právnické fakulty UK svou přednáškou o pojetí moci výkonné v Ústavní listině 1920 jako inspiračním zdroji pro současný systém. Zároveň připojil analýzu problémů a mož- ných nebezpečí, která s sebou toto propojení nese. Příspěvky opět rozvířily diskuzi nad některými přednesenými tématy. Po jejím zakon- čení a následné obědové pauze byla zahájena odpolední část pátečního programu konfe- rence. Jako první z odpoledního bloku nazvaného Vybrané právní a historické aspekty ústav do současnosti přednesl JUDr. Jiří Hřebejk, Ph.D., (PF UK) svůj příspěvek věno- vaný úpravě srovnání územní samosprávy v Ústavě 1920 se současnou ústavní listinou. Následovalo vystoupení Mgr. Elišky Klimentové (PF UK) s názvem Nezávislé úřady jako 139 trend současnosti. Sekci zakončil příspěvek doc. JUDr. PhDr. 100 let od přijetí Ústavní listiny Československé republiky. Historické, politické a právní souvislosti Kuklíka připraven vědomostní kvíz vycházející obsahově z témat rozebíraných u kulatého stolu. Po diskuzi přítomných konferujících se studenty byla sobotní část konference završena komentovaná prohlídkou Senátu. Celá akce byla nejen důstojným připomenutím významného výročí Ústavy 1920, ale také platformou pro sdílení poznatků, náhledů a závěrů z oblasti právní historie, obecné historie a ústavního práva. Jako taková bude (doufejme) impulsem pro další vědeckou práci, propojování různých přístupů k tématu a posouvání hranic odborného poznání. Lukáš Blažek Tereza Blažková doi: 10.14712/2464689X.2020.25 140
https://openalex.org/W2416483003	https://zenodo.org/record/1309089/files/4216ijcsity04.pdf	English	null	Mongodb Vs Mysql: A Comparative Study of Performance in Super Market Management System	International journal of computational science and information technology	2,016	cc-by	3,307	International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 KEYWORDS Relational database, MySQL, MongoDB, Super Market Management System ABSTRACT A database is information collection that is organized in tables so that it can easily be accessed, managed, and updated. It is the collection of tables, schemas, queries, reports, views and other objects. The data are typically organized to model in a way that supports processes requiring information, such as modelling to find a hotel with availability of rooms, thus the people can easily locate the hotels with vacancies. There are many databases commonly, relational and non relational databases. Relational databases usually work with structured data and non relational databases are work with semi structured data. In this paper, the performance evaluation of MySQL and MongoDB is performed where MySQL is an example of relational database and MongoDB is an example of non relational databases. A relational database is a data structure that allows you to connect information from different 'tables', or different types of data buckets. Non-relational database stores data without explicit and structured mechanisms to link data from different buckets to one another. This paper discuss about the performance of MongoDB and MySQL in the field of Super Market Management System. A supermarket is a large form of the traditional grocery store also a self-service shop offering a wide variety of food and household products, organized in systematic manner. It is larger and has a open selection than a traditional grocery store. Dipina Damodaran B, Shirin Salim and Surekha Marium Vargese Dipina Damodaran B, Shirin Salim and Surekha Marium Vargese Department of Computer Engineering,M A College of Engineering, Kothamangalam, Kerala, India Department of Computer Engineering,M A College of Engineering, Kothamangalam, Kerala, India 1. INTRODUCTION The relational database has been the foundation of enterprise applications for decades, and when MySQL is released in 1995 it has been popular and in expensive option. Due to the explosion of large volume and variety of data’s in recent years, non-relational database technologies like MongoDB become useful to address the problems faced by traditional databases. MongoDB is very useful for new applications as well as to augment or replace existing relational infrastructure. MySQL is a popular open-source relational database management system (RDBMS) that is distributed, developed, and supported by Oracle Corporation. The relational systems like, MySQL stores data in tabular form and uses structured query language (SQL) for accessing of data. In MySQL, the programmer should pre-define the schema based on requirements and set up rules to control the relationships between fields in the record. The related information may be stored in different tables, but they are associated by the use of joins. Thus, data duplication can be minimized. 31 DOI :10.5121/ijcsity.2016.4204 International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 MongoDB is an open-source database developed by MongoDB Inc. MongoDB stores data in JSON-like documents that can vary in structure. Related information can be stored together for fast query access through the MongoDB query language. MongoDB uses dynamic schemas, which helps to create records without first defining the structure, such as the attributes or the data types. It is possible to change the structure of records by simply adding new attributes or deleting existing fields. This model helps to represent hierarchical relationships, to store arrays, and other more complex structures very easily. Documents in a record need not have an identical set of fields. MongoDB is designed with high availability and scalability includes replication and auto- sharding. In this paper, we perform a comparison on both MySQL and MongoDB on the platform of supermarket. Figure 1.Example of Super Market Management System Home page Figure 1.Example of Super Market Management System Home page The “Supermarket Management System “which manages the sales activity in a supermarket, maintaining the records of stock details, maintaining the records of the sales done for a particular month/year etc. 1. INTRODUCTION Thus users will consume less time for calculation and the sales activity can be completed within a fraction of seconds whereas manual system will make the employ to write it down the cost of each item and perform calculation which is a long procedure and it also needs a lot of time. By using this system paper work can be reduced and the user can spend extra time for monitoring the supermarket. MongoDB is more applicable to large databases but for the simplicity we take supermarket data. 2. PROBLEM DEFINITION This section gives a brief definition on MySQL and MongoDB. Then evaluate the performance of both the databases on the application of hypermarket. When compared to MySQL it is observed that MongoDB is much better in query processing [9][12]. The MongoDB database consists of a set of databases in which each database contains multiple collections. Because MongoDB 32 International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 operates with dynamic schemas, every collection might contain different types of datas. Every object also called as documents is represented by a JSON structure: a list of key value pairs. The value can be of mainly three types: a primitive value, an array of documents or a list of key- value-pairs. For to query these objects, the client can set the collections expressed as a list of key value pairs. It is also possible to query nested fields. The queries are also JSON like structured; hence a complex query can take much more space than the same query for the relational databases. If the built-in queries are too limited, it is possible to send JavaScript logic to the server for more complex queries. MongoDB supports mainly two types of replication: master-slave and replica sets. In the master- slave replication, the master has control of full data access and which writes every change to its slaves. The slaves can only possible to read data. Replica sets works same as master-slave replications, but it is possible to select a new master if the original master become down. Another important feature that supported by MongoDB is automatic sharding. Using this feature data can be partitioned to different nodes. The administrator has to verify a sharding key for each collection which defines how to partition the documents. In such an environment, the clients connect to a special master node called mongos process which analyses and redirects the query to the appropriate node or nodes. To eliminate data losses, every logical node contain physical servers which act as a replica set. Using this infrastructure it is also possible to use Map/Reduce having a very good performance. 2.1 Architecture MongoDB supports standalone or single instance operations. The replica sets provide high performance of replication with automated failure handling, while sharded clusters make it possible to divide large data sets over different machines which are transparent to the users. MongoDB users combine replica sets and sharded clusters to provide high levels of redundancy of data sets which are transparent for applications [7] . MongoDB scales horizontally by using sharding. The key called shard key is chosen by user, which determines how the data in a collection will be distributed. The data is split into several ranges (based on the shard key) and distributed across different shards. A shard is a master with one or more slaves. Alternatively, the shard key can be hashed to map to a shard enabling an even data distribution. Figure 2.Deployment Architecture MongoDB supports sharding through the configuration of a sharded clusters. Figure 2.Deployment Architecture Figure 2.Deployment Architecture MongoDB supports sharding through the configuration of a sharded clusters. 33 International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 Figure 2.Sharding in Mongodb Figure 2.Sharding in Mongodb Figure 2.Sharding in Mongodb Figure 2.Sharding in Mongodb Sharded cluster has the following components: shards, query routers and config servers. • Sharded cluster has the following components: shards, query routers and config servers. g p q y g • Shards store the data. To provide high availability and data consistency, in a production sharded cluster, each shard is a replica set For more information on replica sets, see Replica Sets. • Shards store the data. To provide high availability and data consistency, in a production sharded cluster, each shard is a replica set For more information on replica sets, see Replica Sets. • Query Routers interface with client applications and direct operations to the appropriate shard or shards. The query router processes and targets operations to shards and then returns results to the clients. A sharded cluster can contain more than one query router to divide the client request load. A client sends requests to one query router. Most sharded clusters have many query routers. • Config servers store the cluster’s metadata. This data contains a mapping of the cluster’s data set to the shards. The query router uses this metadata to target operations to specific shards. Production sharded clusters have exactly 3 config servers. 3. METHODOLOGY Organizations of all sizes commonly adopting MongoDB because it enables them to build applications which are faster, handle highly diverse types of data’s, and manage applications more efficiently at scale. MongoDB documents map naturally to modern, object-oriented programming languages. MongoDB removes the complex object-relational mapping (ORM) layer which translates the objects in code to relational tables. MongoDB’s flexible data model helps the database schema can evolve with business requirements. For example to add a single new field to Craiglist’s MySQL database, it would take months to execute. The Craigslist team moved to MongoDB because it helps to accommodate changes to the data model without costly schema migrations. MongoDB can scale within and across multiple distributed data centers, providing new levels of scalability and availability which are unachievable with relational databases like MySQL. As your deployments grow in terms of data volume and throughput, MongoDB scales easily without much downtime, and without changing the application. But, to achieve scale with MySQL, it often requires significant, custom engineering work. While modern applications require a flexible and scalable system like MongoDB, there are use cases for which a relational database like 34 International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 MySQL are better suited. MongoDB is not a drop-in replacement for legacy applications built around the relational data model and SQL. MySQL are better suited. MongoDB is not a drop-in replacement for legacy applications built around the relational data model and SQL. A concrete example would be the booking of tickets behind a travel reservation system, which also involves complex transactions. While the core booking system might run on MySQL, those parts of the app that system with users – serving booking, integrating with social networks, managing sessions – would be better when placed in MongoDB. MongoDB came with the aim of giving the new way of data storage. Therefore database provide storage of document for the World Wide Web. Began in 2007, MongoDB is built to store data in a dynamic schema, instead of a tabular representation like SQL. The data in MongoDB is stored in the form of object notation based on the format of JSON (Java Script Object Notation). To data transfer over the network between the server and web application which use human readable format the standard method is using of JSON. Prior to JSON, the XML was used for that purpose. 3. METHODOLOGY MongoDB modified the JSON format into its own BSON, which stores the object as a binary format. BSON stands for Binary JSON. Due to its binary format provide more reliable and efficient in the area of storage space and speed. 4. EXPERIMENTAL RESULTS The results of experiments performed to test various aspects of the implementation employed in hypermarket are provided in this section i.e., using the insertion and search operations on databases for auditing purposes .The various operations are performed on the two databases and we obtain the below results. Operations No.of Records Execution Time (in ms) MongoDB MySQL INSERTION 100 0.01 0.01 1000 0.5 1.25 10000 1.2 2.2 25000 2.25 3 SEARCH 100 0.05 0.152 1000 0.12 1.52 10000 0.55 4.47 25000 1.25 5.21 Table 1 for insertion and searching operations The performance of MongoDB while comparing with SQL by performing two operations, Insertion and Searching. A large no of records were taken and performed the operations in both databases. The graph plotted based on the performance is shown below. 35 International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 Figure 3. Insertion operation Figure 4.Searching operation Figure 3. Insertion operation Figure 3. Insertion operation Figure 3. Insertion operation Figure 4.Searching operation Figure 4.Searching operation 36 International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 5. PERFORMANCE EVALUATION On analysing the performance of MySQL and MongoDB databases on hypermarket application, the performance of MongoDB is more when compared to that of MySQL. Organizations of all sizes are commonly adopting MongoDB because it enables them to build applications faster, handle highly diverse types of data, and manage applications more efficiently at large scale. Development is simplified because MongoDB documents map naturally to modern, object- oriented programming languages. Using MongoDB, it removes the complex object-relational mapping (ORM) layer that translates the objects in code to relational tables. MongoDB’s flexible data model helps that the database schema can evolve with business requirements. One of the most important drawbacks of relational databases is that each item can only contain single attribute. Consider a bank example; a customer’s relationship with a bank is stored as different row items in separate tables. So each customer’s master details are stored in one table, the account details of those customers are in another table, the loan details in yet another table, investment details are in a different table, and so on. But these tables are connected to each other by use of relations like primary keys and foreign keys. Non-relational databases, use key-value stores or key-value pairs, are different from this model. Key-value pairs provide possibility to store several related items in one “row” of data in the same table. For instance, in a non- relational table for the same bank example, each row can store the customer’s details as well as their account details, loan and investment details. All data relating to one customer can conveniently stored as one record. This implies an obviously superior method for storing of data, but it has a major limitation: key-value pairs, unlike relational databases, it cannot use relationships between data items. In key-value databases, the customer details like (name, social security, address, account number, etc.) are stored in one data record (instead of stored in several tables, as in the relational model). The customer’s transaction details (account withdrawals, account deposits, loan repayments, etc.) would also be stored as another single data record. 6. DISCUSSION MongoDB is widely used in the field of large databases. One of the most important advantages is its scalability. MongoDB follows BASE transaction, Basically Available Soft State and Eventual consistency .Another important feature is handling of failures. For the simplicity we conduct an analysis based on super market. But MongoDB is more suitable for other applications having large volume of data where data need high security. Since it is schema less, it supports different types of data. 7. CONCLUSION In this paper, we undergo performance evaluation between MySQL and MongoDB on hypermarket application. For evaluating its performance execution time is considered. We came to a conclusion that when number of records inserted or searched is smaller, there is no difference in the execution time taken for each of these operations to complete for both MongoDB and MySQL databases. However, when number of records is increased, MongoDB shows significant reduction in the time taken for execution compared to MySQL. Thus, when the number of records is higher, MongoDB takes less time compared to MySQL. MongoDB can be preferred for better performance. So in summary, RDBMS’s suffer from no horizontal scaling for high transaction loads (millions of read-writes), while NoSQL databases solve high transaction loads but at the cost of data integrity and joins. 37 International Journal of Computational Science and Information Technology (IJCSITY) Vol.4,No.2,May 2016 REFERENCES [1] K. Sanobar, M. Vanita, “SQL Support over MongoDB using Metadata”, International Journal of Scientific and Research Publications, Volume 3, Issue 10, October 2013 [2] https://www.mongodb.org/about/introduction/ [3] S. Hoberman, “Data Modeling for MongoDB”, Publisher by Technics Publications, LLC 2 Lindsley Road Basking Ridge, NJ 07920, USA, ISBN 978-1-935504-70-2, 2014. g g [4] http://dwhlaureate.blogspot.in/2013/10/features-of-mongo-db.html [5] R. P Padhy, M. R. Patra, S. C. Satapathy, “RDBMS to NoSQL: Reviewing Some Next-Generation Non-Relational Database’s”, International Journal of Advance Engineering Sciences and Technologies, Vol. 11, Issue No. 1, 015-030, 2011. [5] R. P Padhy, M. R. Patra, S. C. Satapathy, “RDBMS to NoSQL: Reviewing Some Next-Generation Non-Relational Database’s”, International Journal of Advance Engineering Sciences and Technologies, Vol. 11, Issue No. 1, 015-030, 2011. [6] https://en.wikipedia.org/wiki/MongoDB [6] https://en.wikipedia.org/wiki/MongoDB s://en.wikipedia.org/wiki/shard-(database-architectur [7] https://en.wikipedia.org/wiki/shard-(database-architecture) [8] Z. Wei-Ping, LI Ming-Xin, H. Chen, “Using MongoDB to Implement Textbook Management System instead of MySQL”, IEEE 3rd International Conference on Communication Software and Networks (ICCSN), ISSN 978-1-61284-486, 2011. [8] Z. Wei-Ping, LI Ming-Xin, H. Chen, “Using MongoDB to Implement Textbook Management System instead of MySQL”, IEEE 3rd International Conference on Communication Software and Networks (ICCSN), ISSN 978-1-61284-486, 2011. [8] Z. Wei-Ping, LI Ming-Xin, H. Chen, “Using MongoDB to Implement Textbook Management System instead of MySQL”, IEEE 3rd International Conference on Communication Software and Networks (ICCSN), ISSN 978-1-61284-486, 2011. AUTHORS Dipina Damodaran B completed her B.Tech degree from Malabar College of Engineering and Technology,Trissur in 2013 which is affiliated to Calicut University. She presented paper on National Level Conference. She is currently pursuing M.Tech in Computer Science in Computer Science and Engineering in Mar Athanasius College of Engineering. Her areas of research are Modern Databases, Data Structure and Data Mining. Shirin Salim currently pursuing M.Tech in Computer Science and Engineering in Mar Athanasius College of Engineering. She completed her B.Tech degree from Ilahia College of Engineering in 2014 which is affiliated to Mahatma Gandhi University. She presented paper in National Conference. Her areas of research are Modern Database, Data Mining and Machine Learning. Surekha Mariam Varghese is currently heading the Department of Computer Science and Engineering, M.A. College of Engineering, Kothamangalam, Kerala, India. She received her B-Tech Degree in Computer Science and Engineering in 1990 from CET affiliated to Kerala University and M-Tech in Computer and Information Sciences from CUSAT, Kochi in 1996. She obtained Ph.D in Computer Security from CUSAT, Kochi in 2009. Her research interests include Network Security, Database Management, Data Structures and Algorithms, Operating Systems and Distributed Computing. She has published 17 papers in international journals. Dipina Damodaran B completed her B.Tech degree from Malabar College of Engineering and Technology,Trissur in 2013 which is affiliated to Calicut University. She presented paper on National Level Conference. She is currently pursuing M.Tech in Computer Science in Computer Science and Engineering in Mar Athanasius College of Engineering. Her areas of research are Modern Databases, Data Structure and Data Mining. Shirin Salim currently pursuing M.Tech in Computer Science and Engineering in Mar Athanasius College of Engineering. She completed her B.Tech degree from Ilahia College of Engineering in 2014 which is affiliated to Mahatma Gandhi University. She presented paper in National Conference. Her areas of research are Modern Database, Data Mining and Machine Learning. Surekha Mariam Varghese is currently heading the Department of Computer Science and Engineering, M.A. College of Engineering, Kothamangalam, Kerala, India. She received her B-Tech Degree in Computer Science and Engineering in 1990 from CET affiliated to Kerala University and M-Tech in Computer and Information Sciences from CUSAT, Kochi in 1996. She obtained Ph.D in Computer Security from CUSAT, Kochi in 2009. Her research interests include Network Security, Database Management, Data Structures and Algorithms, Operating Systems and Distributed Computing. She has published 17 papers in international journals. 38 38
https://openalex.org/W2787911117	http://www.jeeng.net/pdf-83564-19729?filename=A Bioremediation Study of.pdf	English	null	A Bioremediation Study of Raw and Treated Crude Petroleum Oil Polluted Soil with Aspergillus niger and Pseudomonas aeruginosa	Journal of Ecological Engineering	2,018	cc-by	5,815	Journal of Ecological Engineering Volume 19, Issue 2, March 2018, pages 226–235 https://doi.org/10.12911/22998993/83564 Received: 2017.12.01 Accepted: 2018.01.18 Published: 2018.03.01 ABSTRACT This study was conducted to investigate the degree of bioremediation that would occur in the samples of soil polluted with raw and treated crude petroleum (oil) with the aid of Aspergillus niger (fungi) and Pseudomonas aeruginosa (bacteria). This was achieved by monitoring the Organic carbon content in the soil over 45 days and pH over 25 days. Four systems of 500 g soil were polluted with 40 g treated crude petroleum, while four systems were contaminated with 40 g raw crude petroleum. The eight systems were labeled accordingly. Two systems for raw crude control and treated crude control (RCC and TCC) were left as control, two systems for raw crude Aspergillus niger and treated crude Aspergillus niger (RCA and TCA) were treated with Aspergillus niger only, two systems for raw crude Pseudomonas aeruginosa and treated crude Pseudomonas aeruginosa (RCP and TCP) were treated with Pseudomonas aeruginosa only and the last two systems for raw crude Aspergillus niger and Pseudomonas aeruginosa (RCAP and TCAP) were treated with both Pseudomonas aeruginosa and Aspergillus niger. At the end of the bioremediation period, the results obtained showed that the pH was not particularly a solid parameter to estimate the degree of bioremediation. This is because a proper trend in the results obtained could not be deter mined. Additionally, it was observed that Aspergillus niger (fungi) and Pseudomonas aeruginosa (bacteria) alone and separate remediate raw crude polluted soil better than treated crude polluted soil. This was determined by the TOC (Total Organic Carbon) values on the 45th day of the experiment. 1 Chemical Engineering Department, Covenant University, P.M.B 1023, Canaan Land, Sango, Ogun State, 2 Microbiology Department, Covenant University, P.M.B 1023, Canaan Land, Sango, Ogun State, Nigeria 2 Microbiology Department, Covenant University, P.M.B 1023, Canaan Land, Sango, Ogun State, Nigeria 2 Microbiology Department, Covenant University, P.M.B 1023, Canaan Land, Sango, Ogun State, Nigeria 3 Department of Food Science and Technology, Federal University of Technology, P.M.B. 704, Akure, Ondo y y 3 Department of Food Science and Technology, Federal University of Technology, P.M.B. 704, Akure, Ondo State, Nigeria 3 Department of Food Science and Technology, Federal University of Technology, P.M.B. 704, Akure, Ondo State, Nigeria * Corresponding author’s e-mail: modupe.ojewumi@covenantuniversity.edu.ng Keyword: bioremediation, Aspergillus niger, Pseudomonas aeruginosa, pollution A Bioremediation Study of Raw and Treated Crude Petroleum Oil Polluted Soil with Aspergillus niger and Pseudomonas aeruginosa Modupe Elizabeth Ojewumi1, Ejemen Valentina Anenih1, Olugbenga Samson Taiwo2, Bosede Temitope Adekeye2, Olugbenga Olufemi Awolu3, Emmanuel Omotayo Ojewumi3 Modupe Elizabeth Ojewumi1, Ejemen Valentina Anenih1, Olugbenga Samson Taiwo2, Bosede Temitope Adekeye2, Olugbenga Olufemi Awolu3, Emmanuel Omotayo Ojewumi3 1 Chemical Engineering Department, Covenant University, P.M.B 1023, Canaan Land, Sango, Ogun State, Nigeria 1 Chemical Engineering Department, Covenant University, P.M.B 1023, Canaan Land, Sango, Ogun State, Nigeria 1 Chemical Engineering Department, Covenant University, P.M.B 1023, Canaan Land, Sango, Ogun State, Nigeria INTRODUCTION The remediation process in ques tion works by stimulating the microbes that occur naturally in the environment to degrade the or ganic wastes found in soil and groundwater. The microorganisms that degrade crude oil use the hydrocarbon breakdown as their source of chemi cal energy. Some microorganisms can naturally degrade petroleum hydrocarbons by utilizing the carbon within it to survive. The hydrocarbons that exist in crude oil serve as substrates for the microorganisms. Normally, without any external enhancement or intervention, as soon as an oil spill occurs there is a rise in the population of mi crobes that degrade the hydrocarbons within the ecosystem Pseudomonas aeruginosa as crude oil de grading bacteria is a gram-negative, rod-shaped bacterium that has an incredible nutritional versa tility. It is a rod about 1-5 µm long and 0.5-1.0 µm wide. Pseudomonas aeruginosa, as well as many other Pseudomonas, can degrade aromatic hydro carbons and can break down toluene, the simplest form of methylbenzene [Johson and Olsen, 1997]. Naturally, it is found in such environments as soil, water, humans, animals, plants, sewage, and hos pitals [Lederberg, 2000]. The factors that are nec essary for the microbial growth are temperature, oxygen, pH, as well as the content of nitrogen and phosphorus. The degree and rate of biodegrada tion are influenced by the type of soil in which the process occurs [Ritmann and McCarty, 2001]. The main microorganisms that degrade petroleum hy drocarbons are bacteria and fungi. Using microor ganisms for hydrocarbon degradation has proven to be the most environmentally-friendly oil spill clean-up method [Facundo et al., 2001; Chen et al., 2007]. Specifically, fungi offer a potential for biodegradation technology [Van et al., 2003]. The decomposition of dead plants and ani mals causes natural occurrence of carbon in soil. The range organic carbon found in soils cuts across freshly deposited litter like leaves, twigs, branches and highly decomposed forms such as humus. Apart from the carbon that occurs in soil naturally, the anthropogenic activities are also sources of carbon in soil, derived as a result of contamination through oil spills. These spills pol lute the environment and invariably increase the total carbon content present in the soil or sedi ment [Schuster et al., 2002]. The Total Organic Carbon is a measure of one of the chemical com ponents of organic matter that is used as a reliable indicator of its presence in a soil or sediment. INTRODUCTION Some of these compounds may include petroleum derivatives, aliphatic and aromatic hydrocarbons, industrial solvents, pesticides and metals [Jelena et al., 2008]. The remediation process in ques tion works by stimulating the microbes that occur naturally in the environment to degrade the or ganic wastes found in soil and groundwater. The microorganisms that degrade crude oil use the hydrocarbon breakdown as their source of chemi cal energy. Some microorganisms can naturally degrade petroleum hydrocarbons by utilizing the carbon within it to survive. The hydrocarbons that exist in crude oil serve as substrates for the microorganisms. Normally, without any external enhancement or intervention, as soon as an oil spill occurs there is a rise in the population of mi crobes that degrade the hydrocarbons within the ecosystem. Aspergillus niger has found wide application in waste management and bio transformations production and extensive hyphal growth in soil. Fungi offer a potential for biodegradation tech nology [Nilanjana and Preethy, 2011]. The fungi cause a disease called black mould on some fruits and vegetables like grapes, apricots and peanuts [Samson et al., 2001]. category include petroleum subsidiaries, aliphatic and sweet-smelling hydrocarbons, mechanical solvents, pesticides and metals [Che, 2002]. The microbial degradation is the major and ultimate natural mechanisms by which one can clean up the petroleum hydrocarbon pollutants from the environment [Atlas and Bragg, 2009]. Of all the technologies and methods for cleaning up oil spills on soils that have been researched in the recent past, bioremediation has come out as the most desirable approach due to its low cost and ability to hinder the formation and accumulation of contaminants [Margesin and Schinner, 2001]. Among the many techniques employed to decon taminate the affected sites, in situ bioremediation using indigenous microorganisms is by far the most widely used [Nwinyi, 2010; Ojewumi et al., 2017]. Bioremediation is a term used to describe a process that takes advantage of the natural abil ity of a microorganism to degrade toxic waste. This technique is very effective in cleaning up the petroleum hydrocarbon pollution. It is a modern technique, whereby the natural degrading ability of microorganisms is used to reduce the concen tration and/or toxicity of wide range chemical substances that are released into the environment. Some of these compounds may include petroleum derivatives, aliphatic and aromatic hydrocarbons, industrial solvents, pesticides and metals [Jelena et al., 2008]. INTRODUCTION tion on marine shore lines have been conducted [Efeovbokhan et al., 2001; Facundo et al., 2001; Jelena et al., 2008; Rittmann and McCarty, 2001; Pao-Wen et al., 2011]. One of the major concerns of the oil industry today is how to improve the recovery of large percentage of oil remaining un recovered in the old and new depleted producing fields [Ojewumi et al., 2017]. The famous biore mediation is an amazing spill clean-up technique in which the normal degrading ability of micro organisms is harnessed for the degradation and decrease of the harmful substances that pollute the environment. Some of the substances in this Since the nineteenth century, petroleum has been utilized for a very long time for power gen eration and lubrication. A greater interest in the natural resource and its by-products was created. The processes involved in obtaining the crude oil and converting it into more useful products usually causes contamination problems. These pollution problems can be minimized, yet not completely eliminated and hence bring on sev eral issues for the environment [Pala et al., 2006]. A number of studies of oil spill bioremedia 226 Journal of Ecological Engineering Vol. 19(2), 2018 category include petroleum subsidiaries, aliphatic and sweet-smelling hydrocarbons, mechanical solvents, pesticides and metals [Che, 2002]. The microbial degradation is the major and ultimate natural mechanisms by which one can clean up the petroleum hydrocarbon pollutants from the environment [Atlas and Bragg, 2009]. Of all the technologies and methods for cleaning up oil spills on soils that have been researched in the recent past, bioremediation has come out as the most desirable approach due to its low cost and ability to hinder the formation and accumulation of contaminants [Margesin and Schinner, 2001]. Among the many techniques employed to decon taminate the affected sites, in situ bioremediation using indigenous microorganisms is by far the most widely used [Nwinyi, 2010; Ojewumi et al., 2017]. Bioremediation is a term used to describe a process that takes advantage of the natural abil ity of a microorganism to degrade toxic waste. This technique is very effective in cleaning up the petroleum hydrocarbon pollution. It is a modern technique, whereby the natural degrading ability of microorganisms is used to reduce the concen tration and/or toxicity of wide range chemical substances that are released into the environment. Determination of pH trols many chemical processes that take place. It specifically affects the plant nutrient availability by controlling the chemical forms of the nutri ent. The optimum pH range for most plants is between 5.5 and 7.0. The closer to neutrality the pH of a soil sample is, the better the soil. An ef fective bioremediation would ensure that the pH of the polluted soil would return to its initial value before the pollution occurred. The changes in the pH level of the soil may be caused by oil spills. Hence, a close monitoring of the pH during a re mediation process would be a determinant to how effective the process is. The pH of the non-polluted soil sample was determined as well as the pH of the raw and treated crude petroleum (oil) samples of each system, one week after pollution just before the inoculation of the micro-organisms. The pH was subsequently determined at 5 day intervals, as the remediation process was underway, using a digi tal pH meter (Model: Jenway M50/Rev model CE 350 EU) in a ratio 1:1 soil sample solution in distilled water. The pH meter was calibrated us ing phthalate buffer solution with a pH of 4.0 and phosphate buffer solution with pH of 7.0. 5 g of the soil samples were mixed with 5 ml of distilled water and stirred very well, after which mixture was allowed to stand for 30 minutes. The elec trode of a pH meter was put into slurry of the soil- water mixture and the pH of the soil was read off. This research focuses on studying the degree of bioremediation in raw crude oil polluted soil and treated crude oil polluted soil using Asper gillus niger and Pseudomonas aeruginosa as po tential hydrocarbon degraders. This study shows the effect of the bioremediation of crude oil pol luted soil with each microorganism individually, as well as both microbes in a mixed microbial consortium. In order to achieve this, the organic content in the soil over the remediation period as well as the soil pH would be monitored. Soil sample The loamy soil samples used to simulate the crude oil onshore spill were obtained from Cov enant University Farms, Canaan land Ota, Nige ria. This soil was collected from the surface layer, about 20 to 30 cm below the land surface. The soil sample was air dried to get rid of the excess mois ture, as soil was collected during the heavy rains. The soil was homogenized and stored in a black plastic bucket at room temperature. The dried soil was characterized for the physicochemical and microbial parameters, according to the standard methods. The raw and treated crude oil samples (Escravos light) used for this study were obtained from Chevron Nigeria Limited, Delta state Nige ria in December 2015. 2 4 1 g of the soil sample from a system was weighed and put into a 250 ml conical flask. 10 ml of the 1.0M Potassium Dichromate solution was added to the conical flask. 20 ml of concentrated H2SO4 was also added to the flask under a fume hood and stirred gently. It was left under the fume hood for 5 minutes after which it was brought out. Distilled water was then added to the 125 mL mark, mixed vigorously by swirling and allowed to stand for 30 minutes. After the stipulated time, 3 drops of Ferroin indicator was added. This was then titrated with 0.5M ferrous sulphate solution. The titration was carried out twice for each sample and the aver age volume of the acid used was recorded. This test was carried out on the non-polluted soil sample, the soil samples from all the systems after pollu tion just before inoculation, and each soil sample at 5 day intervals during the remediation process. Determination of total organic carbon (TOC) This parameter was measured in the non- polluted soil, before the commencement of the remediation process. In order to determine the organic carbon, the Walkey-Black Wet Dichro mate method according was used. The required reagents were 0.5M Ferrous Sulphate solution, 1.0M Potassium Dichromate solution, concen trated H2SO4(98%)and Ferroin indicator. INTRODUCTION The determination of the soil organic carbon is based on the Walkley-Black chromic acid [Van et al., 2003] oxidation method. Estimating the Total Or ganic Carbon is used to determine the amount of hydrocarbons in the soil since they largely com prise carbon. Aspergillus niger has found wide application in waste management and bio-transformations. The fungus is most commonly found in meso philic environments, such as decaying vegeta tion or soil and plants [Schuster et al., 2002; Das and Mukherjee, 2007]. Filamentous fungi play an important role in degrading hydrocarbons by producing capable enzymes. This is as a re sult of their aggressive growth, greater biomass The term pH generally refers to the degree of acidity or alkalinity of a substance. Soil pH is a very important variable in soils, because it con 227 Journal of Ecological Engineering Vol. 19(2), 2018 SOM) contains 58% C ulated time, 3 drops o lphate solution The t hate solution. The titr omate solution was a Total organic carbon culated thus: Eq. (1) Eq. (2) Eq. p the acid used was recorded. This test was car l samples from all the systems after polluti 5 day intervals during the remediation process. ge TOC, the percentage easily oxidized car matter was also calculated once the total organ matter assumes that 77% of the organic carbon Matter (SOM) contains 58% C. ) is calculated thus: Eq. (1) Eq. (2) Eq. e acid used was recorded. This test was carrie samples from all the systems after pollution day intervals during the remediation process. TOC, the percentage easily oxidized carbo atter was also calculated once the total organic atter assumes that 77% of the organic carbon is ter (SOM) contains 58% C. calculated thus: Eq. (1) Eq. (2) Eq. o the flask under a fume hood and stirred gent es after which it was brought out. Distilled wa olution was mixed vigorously by swirling an ated time, 3 drops of Ferroin indicator was add hate solution. The titration was carried out tw he acid used was recorded. This test was carrie samples from all the systems after pollution day intervals during the remediation process. TOC, the percentage easily oxidized carbo atter was also calculated once the total organic atter assumes that 77% of the organic carbon is ter (SOM) contains 58% C. s calculated thus: Eq. (1) Eq. (2) Eq. Table 2 shows the values obtained for the eas ily oxidizable carbon, total organic carbon and organic matter in the soil. The most important parameter is the total organic carbon (TOC). At the start of the experiment (day 0), some of the samples had equal organic carbon content be cause no microbial degradation had occurred. For all the systems polluted with raw crude petroleum (RCA, RCP, RCAP, and RCC) a Total Organic Carbon of 2.057% was recorded. The systems polluted with treated crude petroleum (TCA, TCP, TCAP and TCC) had a Total Organic Carbon con tent of 2.8461% as shown in the table. This re sult showed that the treated crude oil systems had more carbon content on the average at the start of the experiment. Subsequently, the Total Or ganic Carbon for all the systems spiked on the 5th day. This was probably due to the fact that the carbon content originally in the soil was properly blended with the crude oil (petroleum), leading to an increase in the Total Organic Carbon. SOM) contains 58% C ulated time, 3 drops o lphate solution The t hate solution. The titr omate solution was a Total organic carbon The calculation of organi the method and that Soil Organic the method and that Soil Organic M non-polluted soil sample, the soi inoculation and each soil sample at M is the molarity of the ferrous sulphate solution = 0.5 M the method and that Soil Organic % E il O idi bl C b (EO % Easily Oxidizable Carbon (EOC) ocu at o a d eac so sa p e at To calculate the percenta 19 40+20 00+19 60 12/4000 = milliequivalent weight of car bon in grams. % Easily Oxidizable Carbon (EO (B S) M 12𝑚 %EOC = (B−S)∗M ∗12𝑚∗100% To calculate the percentag calculated. The amount of organic 19 40+20 00+19 60 12/4000 = milliequivalent weight of car bon in grams. % Easily Oxidizable Carbon (EO (B S) M 12𝑚 %EOC = (B−S)∗M ∗12𝑚∗100% To calculate the percentag calculated. The amount of organic B = 3 Percebt of TOC can be obtained from %EOC by applying a correction factor. %EOC = 𝑚𝑚∗4000 ∗100% %TOC %EOC known. The calculation of organic the method and that Soil Organic M %TOC = %EOC 0.77 %SOM = %TOC∗1.72 (2) %TOC = 0.77 %SOM = %TOC∗1.72 0.58 (3) the method and that Soil Organic M % Easily Oxidizable Carbon (EOC) (B−S)∗M ∗12𝑚 %TOC = %EOC 0.77 %SOM = %TOC∗1.72 (2) %TOC = 0.77 %SOM = %TOC∗1.72 0.58 (3) the method and that Soil Organic M % Easily Oxidizable Carbon (EOC) (B S)∗M ∗12𝑚 (2) Preparation of microorganisms The fungi (Aspergillus niger) and bacteria (Pseudomonas aeruginosa) used for this study were cultured in the Microbiology Department of Covenant University and prepared according to [Ojewumi et al., 2016; Ojewumi et al., 2018]. In order to calculate the percentage TOC, the percentage of easily oxidized carbon was cal culated first. The amount of organic matter was also calculated once the total organic carbon was known. The calculation of organic matter as 228 Journal of Ecological Engineering Vol. 19(2), 2018 s also calculated once the total organic car lution was mixed vigorously by swirling time, 3 drops of Ferroin indicator was add ( ) stem was weighed and put into a 250ml c sumes that 77% of the organic carbon is oxidized by the method and that soil organic matter (SOM) contains 58% C. known. The calculation of organic matt the method and that Soil Organic Matter stand for 30 minutes. After the st then titrated with 0.5M ferrous sample and the average volume then titrated with 0.5M ferrous sul sample and the average volume of 10ml of the 1.0M Potassium Dic concentrated H2SO4 was also adde SOM) contains 58% C ulated time, 3 drops o lphate solution The t hate solution. The titr omate solution was a Total organic carbon Between the 5th and 10th day, there was a commendable reduction in the Total Organic Carbon of all the systems. Some of the soil samples experienced a very drastic decrease in TOC within this period of time. A typical example can be seen in the system RCP in which the TOC decreased from 3.752% to 2.466%. In contrast, other systems experienced just a minute decrease within this period of time. Additionally, it was observed that some of the systems treated with Aspergillus niger developed a whitish substance (which appeared to be mould growth) on them. Approximately on the 20th day, it was observed that the systems began to produce a foul odour that resembled the smell of decom posing organic waste Percent of easily oxidizable carbon (EOC) is thus calculated in the following way: % Easily Oxidizable Carbon (EOC) is c sample and the average volume non-polluted soil sample, the s non-polluted soil sample, the soil inoculation and each soil sample at left under the fume hood for 5 min added to the 125mL mark and the %EOC = (B−S)∗M ∗12 𝑚𝑚∗4000 ∗100% (1) inoculation and each soil sample inoculation and each soil sample at T l l h added to the 125mL mark and the stand for 30 minutes. After the stipu (1) mple p e stip %TOC = %EOC where: B is the volume of ferrous solution used in the blank titration in ml; To calculate the percen To calculate the percentag calculated. The amount of organic then titrated with 0.5M ferrous su %TOC 0.77 S is the volume of ferrous solution used in the sample titration in ml; calculated. The amount of organ k Th l l ti f i g known. The calculation of organic m sample and the average volume of non polluted soil sample the soi %SOM = %TOC∗1.72 0.58 m is the mass of the sample in grams used in the analysis which is 1 g in this case; known. B = 19 Organic Carbon %TOC B = 3 3 Blank titration of the acidic dichromate with ferrous sulphate solution was performed at the beginning of the batch analysis using the same procedure with no soil added (Table 1). %SOM = %TOC∗1.72 0.58 Average Volume of Ferrous Sulphate used in Standardization, , B = 19.40+20.00+19.60 3 B = 19.66 ml. At the end of 45 days of experimenting, the TOC of majority of the systems had decreased to less than 1%, as shown in table. The system TCAP had residual TOC of less than 0% (-1.332%). These values came about as a result of the values obtained from titration. At this point, the volume of acid used to titrate each soil sample was great er than the volume of the blank titre, hence the The collated table of average titre values of ferrous sulphate solution used to oxidize the soil samples mixed with potassium dichromate solu tion over the period of experimentation as well as the %EOC, %TOC and %OM is shown (Table 2). The graphs 1-8 shows %EOC, %TOC and %SOM against time for each system over the course of 45 days. Table 1. Blank titre obtained from standardization Blank titre obtained from standardization Blank 1 Blank 2 Blank 3 Initial (ml) 0.00 19.40 0.00 Final (ml) 19.40 39.40 19.60 Volume used (ml) 19.40 20.00 19.60 229 Journal of Ecological Engineering Vol. 19(2), 2018 Table 2. B = 19 Organic Carbon %TOC Organic carbon data collation across the 45 days Day RCA TCA RCP TCP RCAP TCAP RCC TCC Day 0 Vol (ml) 9.1 5.05 9.1 5.05 9.1 5.05 9.1 5.05 % EOC 1.58 2.19 1.58 2.19 1.58 2.19 1.58 2.19 % TOC 2.06 2.85 2.06 2.85 2.06 2.85 2.06 2.85 % OM 6.10 8.44 6.10 8.44 6.10 8.44 6.10 8.44 Day 5 Vol (ml) 2.4 6.35 0.4 1.35 5.15 5.15 2.6 3.2 % EOC 2.59 1.99 2.89 2.75 2.18 2.18 2.56 2.47 % TOC 3.36 2.59 3.75 3.57 2.83 2.83 3.32 3.21 % OM 9.97 7.69 11.13 10.58 8.38 8.38 9.86 9.51 Day 10 Vol (ml) 5.2 6.9 7 8.55 8 9.2 5.65 7.3 % EOC 2.17 1.91 1.89 1.67 1.75 1.57 2.10 1.85 % TOC 2.82 2.49 2.47 2.16 2.27 2.04 2.73 2.41 % OM 8.35 7.37 7.31 6.41 6.74 6.04 8.09 7.14 Day 15 Vol (ml) 10.25 5.85 7.4 13 10.5 12 4.2 5.6 % EOC 1.41 2.07 1.84 0.99 1.37 1.15 2.32 2.11 % TOC 1.83 2.67 2.39 1.30 1.78 1.49 3.01 2.74 % OM 5.43 7.98 7.08 3.85 5.29 4.43 8.93 8.12 Day 20 Vol (ml) 12.3 7.8 11.5 12.7 11.7 13.6 5.4 8.3 % EOC 1.10 1.78 1.22 1.04 1.19 0.91 2.14 1.70 % TOC 1.43 2.31 1.59 1.36 1.55 1.18 2.78 2.21 % OM 4.25 6.85 4.71 4.02 4.60 3.50 8.24 6.56 Day 30 Vol (ml) 11.8 10.2 16.4 14.2 14.8 15.6 9.7 10.8 % EOC 1.18 1.42 0.49 0.82 0.73 0.61 1.49 1.34 % TOC 1.53 1.84 0.64 1.06 0.947 0.79 1.94 1.74 % OM 4.54 5.47 1.88 3.15 2.81 2.35 5.75 5.15 Day 45 Vol (ml) 18.4 15.7 19.4 17.05 16.9 26.5 16.3 18.7 % EOC 0.19 0.59 0.05 0.39 0.414 -1.03 0.51 0.14 % TOC 0.25 0.77 0.06 0.51 0.54 -1.33 0.66 0.19 % OM 0.73 2.29 0.18 1.51 1.59 -3.95 1.97 0.55 Key: TCC Treated Crude Control sample RCC Raw Crude Control sample RCA Raw Crude Treated with Aspergillus niger TCA Treated Crude Treated with Aspergillus niger RCP Raw Crude Treated with Pseudomonas aeruginosa TCP Treated Crude Treated with Pseudomonas aeruginosa RCAP Raw Crude Treated with Aspergillus niger and Pseudomonas aeruginosa TCAP Treated Crude Treated with Aspergillus niger and Pseudomonas aeruginosa EOC Easily Oxidizable Carbon TOC Total Organic Carbon OM Organic Matter negative TOC value. B = 19 Organic Carbon %TOC The plots shown in figures 1 to 8 clearly depict the trend in the EOC, TOC and OM. These three curves on each plot were observed to have the same rising and falling trend at each point for each of the systems. pH The pH of the unpolluted soil was measured was done at four different times within the first 25 days; the chart is shown in Figure 9. The val ues obtained were used to plot a simple column chart that shows the pH range for all 8 systems at four different times within the first 25 days. The pH of all the soil systems spiked within the first five days to the values above 7.0. The pH of some systems rose to as high as 7.9 (as seen in RCAP) within the first five days. On average, after the ii negative TOC value. The plots shown in figures 1 to 8 clearly depict the trend in the EOC, TOC and OM. These three curves on each plot were observed to have the same rising and falling trend at each point for each of the systems. was done at four different times within the first 25 days; the chart is shown in Figure 9. The val ues obtained were used to plot a simple column chart that shows the pH range for all 8 systems at four different times within the first 25 days. The pH of all the soil systems spiked within the first five days to the values above 7.0. The pH of some systems rose to as high as 7.9 (as seen in RCAP) within the first five days. On average, after the first five days till the 15th day, the pH of some of the systems decreased with the exception of TCA, pH pH The pH of the unpolluted soil was measured and estimated to be 6.81. The test for the degree of acidity or alkalinity in each of the systems 230 Journal of Ecological Engineering Vol. 19(2), 2018 Fig. 2. Carbon content variation for TCA Fig. 1. Carbon content variation for RCA Fig. 3. Carbon content variation for RCP Fig. 2. Carbon content variation for TCA Fig. 1. Carbon content variation for RCA Fig. 1. Carbon content variation for RCA Fig. 2. Carbon content variation for TCA Fig. 2. Carbon content variation for TCA Fig. 3. Carbon content variation for RCP 231 Journal of Ecological Engineering Vol. 19(2), 2018 Fig. 4. Carbon content variation for TCP Fig. 5. Carbon content variation for RCAP Fig 4 Carbon content variation for TCP Fig. 4. Carbon content variation for TCP Fig. 6. Carbon content variation for TCAP Fig. 5. Carbon content variation for RCAP Fig. 4. Carbon content variation for TCP Fig. 4. Carbon content variation for TCP Fi 5 C b i i f RCAP Fig. 5. Carbon content variation for RCAP Fig. 6. Carbon content variation for TCAP 232 Journal of Ecological Engineering Vol. 19(2), 2018 Fig. 8. Carbon content variation for TCC Fig. 7. Carbon content variation for RCC Fi 7 C b t t i ti f RCC Fig. 8. Carbon content variation for TCC Fig. 9. pH for all Systems Fig. 7. Carbon content variation for RCC Fig. 7. Carbon content variation for RCC Fi 8 C b t t i ti f TCC Fig. 8. Carbon content variation for TCC Fig. 9. pH for all Systems Fig. 9. pH for all Systems 233 Journal of Ecological Engineering Vol. 19(2), 2018 TCP, TCAP and TCC. Between the 15th and 20th day, all the pH values for all systems decreased but there was a fluctuation in the trend in some of the samples. The 25th day showed a major pH reduction in all the systems. The pH of all the systems decreased to values below 7.0. This shows that effective remediation had occurred in all the systems, proceeding towards return ing the soil samples to the original state of the soil. Around the 25th day, some pH values fell below the value of the pH of the unpolluted soil, as seen in RCP 2 (6.56). 2. Che M.D. 2002. pH Conservative cost estimaye in cluding investigation and monitoring expenses, soil and ground water remediation confenrence proceeding, Taipei. 3. Chen S.Y., Lu W.B., Wei Y.H., Chen W.M., Chang J.S. 2007. Improved production of biosufactant with newly isolated Pseudomonas aeruginosa S2. Bioresource Technology, 23, 661-666. 4. Das K., Mukherjee A.K. 2007. Crude petroleum- oil biodegradation eﬃciency of Bacillus subtilis and Pseudomonas aeruginosa strains isolated from a petroleum-oil contaminated soil from North-East India. Bioresource Technology, 98(7), 1339–1345. 5. Efeovbokhan V.E., Anawe P., Apeye L., Makinde F.A., Odunmbaku O. 2011. Comparison of the ef ficiency of sodium nitrate and superphosphate as nutrients in the bioremediation of petroleum hy drocarbon polluted water. American Journal of Scientific and Industrial Research, 2(2), 272-282. doi:10.5251/ajsir.2011.2.2.278.282. Acknowledgements The authors appreciate the partial sponsorship of Covenant University, Ota, Nigeria. 12. Ojewumi M.E., Obielue B.I., Emetere M.E., Awolu, O.O., Ojewumi, E.O. 2018. Alkaline Pre- Treatment and Enzymatic Hydrolysis of Waste Papers to Fermentable Sugar. Journal of Eco logical Engineering, 19(1), 211–217. https://doi. org/10.12911/22998993/79404. CONCLUSIONS Using Aspergillus niger alone remediates the raw crude oil polluted soil better than the treated crude oil polluted soil. Therefore, for a spill site polluted with raw crude oil, Aspergillus niger would be a better choice of biostimulant com pared to Pseudomonas aeruginosa. The use of Pseudomonas aeruginosa as a biostimulant is more effective in the remediation of the soil pol luted with treated crude oil. Therefore, for an oil spill site polluted with treated crude oil from a flow line or even the production platform, Pseu domonas aeruginosa is a better choice of biostim ulant compared to Aspergillus niger. Using both Aspergillus niger and Pseudomonas aeruginosa in combination is generally a better biostimulant compared to using either Aspergillus niger or Pseudomonas aeruginosa alone. This is observed from the biostimulant efficiency and percentage crude oil removal. Generally, the treated crude oil spills remediate faster and to a larger extent than the spills caused by raw crude oil. 6. Facundo J., Vanessa H., Teresa M. 2001. Biodeg radation of diesel oil in soil by a microbial consor tium. Water, Air and Soil Pollution, 313-320. 7. Jelena S.M., Beškoski V.P., Mila V.I., Samira A.M., Gordana D., Miroslaw M.V. 2008. Bioremediation of soil heavily contaminated with crude oil and its products: composition of the microbial con sortium. Journal of the Serbian Chemical Society, 74(4), 455-460. 8. Lederberg J. 2000. Pseudomonas. Encyclopedia of Microbiology, pp. 876-891. 9. Margesin R., Schinner F. 2001. Bioremediation (natual attenuation and biostimulation) of diesel- oil-contaminated soil in an alpine glacier skiing area. Applied and Environmental Miacrobiology, 67, 3127-3133. 10. Nilanjana D., Preethy C. 2001. Microbial Degra dation of Petroleum Hydrocarbon Contaminants: An Overview. Biotechnology Research Interna tional, Volume 2011, Article ID 941810, 13 pages. doi:10.4061/2011/941810 The pH test is not a very good determining factor for remediation level, because the result obtained did not exhibit a distinct trend as a result of fluctuations. 11. Nwinyi O. 2010. Degradation of Askarel (PCB Blend) by Indigenous Aerobic Bacteria Isolates from Dumpsites in Ore, Ondo State.Nigeria. Aus tralian Journal of Basic and Applied Sciences, 4 (8). 3938-3948. REFERENCES 1. Atlas R., Bragg J. 2009. Bioremediation of marine oil spills: when and when not—the Exxon Valdez experience. Microbial Biotechnology, 2(2), 213– 221. 1. Atlas R., Bragg J. 2009. Bioremediation of marine oil spills: when and when not—the Exxon Valdez experience. Microbial Biotechnology, 2(2), 213– 221. 13. Ojewumi M.E., Emetere M.E., Babatunde D.E., Okeniyi, J.O. 2017. In Situ Bioremediation of Crude Petroleum Oil Polluted Soil Using Math ematical Experimentation. International Journal of 234 Journal of Ecological Engineering Vol. 19(2), 2018 Chemical Engineering. Volume 2017, Article ID 5184760. 17. Robert M., Stephen J.A. 2003. Biodegradation of fuel oil under laboratory and arctic marine con ditions. Spill Science and Technology Bulletin, 297-302. 14. Pala M., De Carvalho D., Pinto J.C., Sant Anna Jr G.L. 2006. A suitable model to describe bioreme diation of a petroleum-contaminated soil. Journal of International Biodeterioration & Biodegrada tion, 58(6), 254-260. 18. Samson R., Houbraken J., Summerbell R., Flan nigan B. 2001. Common and important species of fungi and actinomycetes in indoor environments: Microogranisms in Home and Indoor Work Envi ronments. 15. Pao-Wen G.L., Tsung C.C., Liang-Ming W., Chun- Hsuan K., Po-Tseng P., Sheng-Shung C. 2011. Bioremediation of petroleum hydrocarbon con taminated soil: Effects of strategies and microbial community shift. International Biodeterioration and Biodegradation, 65(2011), 1119-1127. 19. Schuster E., Dunn-Coleman N., Frisvad J., Van Di jck P. 2001. On the safety of Aspergillus niger – a review. Applied Microbiology and Biotechnology, 59, 426-435. 16. Rittmann B.E., McCarty P.L. 2001. Enviromental Biotechnology: Principles and Applications. New York: McGraw-Hill. 20. Van Hamme J., Singh A., Owen W.P. 2003. Microbiology and Molecular Biology Review, 503-549. 235
https://openalex.org/W2108572261	http://pure-oai.bham.ac.uk/ws/files/20708603/Absoud_et_al_Paediatric_UK_demyelinating_disease_BMC_Pediatrics_2011.pdf	English	null	Paediatric UK demyelinating disease longitudinal study (PUDDLS)	BMC pediatrics	2,011	cc-by	8,545	Publisher Rights Statement: Publisher Rights Statement: Publisher Rights Statement: This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Publisher Rights Statement: This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproductio original work is properly cited. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any med original work is properly cited. Checked July 2015 Checked July 2015 Paediatric UK demyelinating disease longitudinal study (PUDDLS) Absoud, Michael; Cummins, Carole; Chong, WK; De Goede, C; Foster, K; Gunny, R; Hemingway, C; Jardine, P; Kneen, R; Likeman, M; Lim, MJ; Pike, M; Sibtain, N; Whitehouse, WP; Wassmer, E Citation for published version (Harvard): Absoud, M, Cummins, C, Chong, WK, De Goede, C, Foster, K, Gunny, R, Hemingway, C, Jardine, P, Kneen, R, Likeman, M, Lim, MJ, Pike, M, Sibtain, N, Whitehouse, WP & Wassmer, E 2011, 'Paediatric UK demyelinating disease longitudinal study (PUDDLS)', BMC Pediatrics, vol. 11, 68. https://doi.org/10.1186/1471-2431-11-68 Link to publication on Research at Birmingham portal Publisher Rights Statement: This is an Open Access article distributed under the terms of the Creative Commons Attribution License General rights U l li General rights Unless a licence is specified above, all rights (including copyright and moral rights) in this document are retained by the authors and/or the copyright holders. The express permission of the copyright holder must be obtained for any use of this material other than for purposes permitted by law. •Users may freely distribute the URL that is used to identify this publication. •Users may freely distribute the URL that is used to identify this publication. •Users may freely distribute the URL that is used to identify this publication. •Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private study or non-commercial research. Users may freely distribute the URL that is used to identify this publication. •Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private study or non-commercial research. U t t f th d t i li ith th t f ‘f i d li ’ d th C i ht D i d P t t A t 1988 (?) y •User may use extracts from the document in line with the concept of ‘fair dealing’ under the Copyright, Designs and Patents Act 1988 (?) •Users may not further distribute the material nor use it for the purposes of commercial gain. here a licence is displayed above, please note the terms and conditions of the licence govern your use of this document Where a licence is displayed above, please note the terms When citing, please reference the published version. * Correspondence: michaelabsoud@childdemyelination.org.uk 1School of Health & Population Sciences, University of Birmingham, Vincent Drive, Birmingham, B15 2TT, UK Full list of author information is available at the end of the article Abstract Background: There is evidence that at least 5% of Multiple sclerosis (MS) cases manifest in childhood. Children with MS present with a demyelinating episode involving single or multiple symptoms prior to developing a second event (usually within two years) to then meet criteria for diagnosis. There is evidence from adult cohorts that the incidence and sex ratios of MS are changing and that children of immigrants have a higher risk for developing MS. A paediatric population should reflect the vanguard of such changes and may reflect trends yet to be observed in adult cohorts. Studying a paediatric population from the first demyelinating event will allow us to test these hypotheses, and may offer further valuable insights into the genetic and environmental interactions in the pathogenesis of MS. Methods/Design: The Paediatric UK Demyelinating Disease Longitudinal Study (PUDDLS) is a prospective longitudinal observational study which aims to determine the natural history, predictors and outcomes of childhood CNS inflammatory demyelinating diseases. PUDDLS will involve centres in the UK, and will establish a cohort of children affected with a first CNS inflammatory demyelinating event for long-term follow up by recruiting for approximately 5 years. PUDDLS will also establish a biological sample archive (CSF, serum, and DNA), allowing future hypothesis driven research. For example, the future discovery of a biomarker will allow validation within this dataset for the evaluation of novel biomarkers. Patients will also be requested to consent to be contacted in the future. A secondary aim is to collaborate internationally with the International Paediatric Multiple Sclerosis Study Group when future collaborative studies are proposed, whilst sharing a minimal anonymised dataset. PUDDLS is the second of two jointly funded studies. The first (UCID-SS) is an epidemiological surveillance study that already received ethical approvals, and started on the 1st September 2009. There is no direct patient involvement, and UCID-SS aims to determine the UK and Ireland incidence of CNS inflammatory demyelinating disorders in children under 16 years. Discussion: A paediatric population should reflect the vanguard of MS epidemiological changes and may reflect trends yet to be observed in adult MS cohorts. The restricted window between clinical expression of disease and exposure to environmental factors in children offers a unique research opportunity. Abstract Studying a paediatric population from the first demyelinating event will allow us to investigate the changing epidemiology of MS, and may offer further valuable insights into the genetic and environmental interactions in the pathogenesis of MS. demyelinating disease of the CNS characterized by mye- lin loss, axonal degeneration and, often, progressive neu- rological dysfunction, that is usually relapsing remitting at onset. The incidence of childhood CIDD is unknown, and the UK and Ireland Childhood Inflammatory Demyelination Surveillance Study (UCID-SS) is already underway to determine this. At the first presentation of a CNS inflammatory demyelinating event, children are diagnosed with either an acute disseminated encephalo- myelitis (ADEM), optic neuritis, transverse myelitis, or © 2011 Absoud et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Take down policy Take down policy While the University of Birmingham exercises care and attention in making items available there are rare occasions when an item has been uploaded in error or has been deemed to be commercially or otherwise sensitive. If you believe that this is the case for this document, please contact UBIRA@lists.bham.ac.uk providing details and we will remove access to the work immediately and investigate. Download date: 24. Oct. 2024 Absoud et al. BMC Pediatrics 2011, 11:68 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Open Access Paediatric UK demyelinating disease longitudinal study (PUDDLS) Michael Absoud1,2*, Carole Cummins1, Wui K Chong3, Christian De Goede4, Katharine Foster5, Roxanna Gunny3, Cheryl Hemingway6, Philip Jardine7, Rachel Kneen8, Marcus Likeman9, Ming J Lim10, Mike Pike11, Naomi Sibtain12, William P Whitehouse13 and Evangeline Wassmer2 Early diagnosis allows for early treatment y g y For relapses or attacks of MS, corticosteroids are the mainstay of treatment. Early initiation of “disease modi- fying agents”, such as beta interferon and glatiramer acetate appears to be of benefit in adults. Thus, it is more critical than ever to accurately diagnose MS as early as possible. However, very little is published on the use of interferon in children [29-33]. The determination of the longitudinal outcome of the first CNS inflamma- tory demyelinating episode in children is an essential pre-requisite for the design of any prevention or treat- ment trials. Prediction of outcome at an early stage is critical to quantify the risk to benefit ratio of any intervention. KIDMUS [10] criteria: KIDMUS [10] criteria: All of: ≥1 lesion perpendicular to long axis of corpus callosum; sole presence of well defined lesions Background Introduction CNS i fl CNS inflammatory demyelinating diseases (CIDD) are rare childhood disorders but may culminate in physical and cognitive disability or ultimately be diagnosed as Multiple Sclerosis (MS). MS is a chronic inflammatory Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Page 2 of 10 definitions [17] for CIDD are operational and require evaluation in prospectively followed up cohorts. For iso- lated monophasic syndromes, normal MRI and cere- brospinal fluid (CSF) are thought to predict a very low risk for development of MS in adults [23]. However the same may not be true for children. In addition the dif- ferential diagnosis for childhood demyelinating disease and MS (documented in exclusion criteria) is much wider than that in adults. Furthermore, MRI criteria used in adults to diagnose MS e.g. MRI McDonald cri- teria may not be valid in children [24-26]. The utilisa- tion of revised McDonald criteria [27] and others including the KIDMUS MRI criteria [10] for paediatric MS need to be validated in new cohorts. another clinically isolated episode (CIS). Children who present with a CNS inflammatory demyelinating episode involving single or multiple symptoms may later present with a second event (majority usually within two years) to then meet criteria for MS diagnosis [1]. There is evi- dence that at least 5% of MS cases manifest in child- hood [2]. It is not clear at the first onset of symptoms which children will go on to develop MS. No available biomarker for MS In line with adult based studies [28], there is no cur- rently available biomarker that fulfils the criteria of a surrogate endpoint in MS in children. Current routine tests including CSF oligoclonal bands, lack sensitivity and specificity in individuals presenting with a first CNS inflammatory demyelinating event. A precondition for the identification of biomarkers is a well described and prospectively followed clinical cohort and the availability of stored tissue linked to relevant clinical information. Paediatric cohort may reveal new trends Various new trends in adult MS have been reported (reviewed in ref [20]), which include geographic rising rate of MS, and a rising female to male sex ratio by year of birth [21]. There is evidence that Indian and Pakistani immigrants who entered England younger than 15 had a higher risk of developing MS than those that entered after this age [22]. The vanguard of any change in trends in the epidemiology of MS is likely to be evident in a paediatric population. Genetic epidemiology to date supports the dual role of “nature and nurture” in MS pathobiology [20]. As such paediatric cohorts can also offer further valuable insights into the genetic and envir- onmental interactions in the pathogenesis of MS. The corner stone of cohorts that are likely to be able to address complex disorders such as MS, depend on the availability of biological sample with comprehensive databases with clinical information. Prospective studies are required The available literature on paediatric MS, is primarily limited to smaller case series and larger retrospective reviews of established adult MS populations [3-16]. The International Paediatric MS Study Group, (http://www. ipmssg.org) have recently published consensus defini- tions of paediatric CNS inflammatory demyelinating dis- orders and MS to help facilitate uniformity in future research [17]. Evidence for differences in natural history of childhood onset MS [18] are also primarily derived from retrospective cohorts, as are studies linking envir- onmental factors (EBV infections or Vitamin D defi- ciency) with possible MS development [19]. The risk of relapse in children and hence the risk of developing MS may be geographically different [1]. There is hence an urgent need for prospective population based studies to further understand clinical, radiological, and pathobiolo- gical features as well as outcome of childhood onset inflammatory demyelinating disease. McDonald et al. [27] criteria: Three of: 1) ≥9 white matter lesions or 1 gadolinium enhancing lesion, 2) ≥3 periventricular lesions, 3) 1 jux- tacortical lesion, 4) an infratentorial lesion. The need to know the extent of disease burden F h id if i h l hi The need to know the extent of disease burden Furthermore, identifying the natural history and out- comes of paediatric MS in the UK has major implica- tions when planning service provisions, where there is evolving evidence of lack of awareness, and patchy ser- vice provision. At the Paediatric MS meeting, London At first presentation of CIDD, children are diagnosed with optic neuritis, transverse myelitis, other clinically isolated episodes (CIS) or acute disseminated encephalo- myelitis (ADEM). The newly proposed consensus Page 3 of 10 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 November 2007 (attended by families and multidisci- plinary professionals) there was a call for a better under- standing of long-term outcomes including quality of life, cognitive, adaptive behavioural and functional skills, neuropsychiatric symptoms, and treatment effects for children presenting with CNS inflammatory demyelination. Case Definitions and Inclusion Criteria 1. Identify features (clinical, epidemiological, imaging, pathophysiolgical tests such as oligoclonal bands) that may predict which children are more likely to relapse and develop MS, whilst assessing the validity and utility of the new consensus definitions for childhood CNS inflammatory demyelination and of proposed MRI criteria. Children (younger than 16 years) experiencing first epi- sode of clinical neurological events consistent with site specific inflammatory CNS demyelination and except in Optic Neuritis confirmed with white matter changes on MRI (table 1). Exclusion criteria 2. Establish outcomes for children, for an initial period of up to five years depending on when the case was recruited (minimum 6 months follow up). 1. Leukodystrophies (e.g., metachromatic leukodystro- phy, adrenoleukodystrophy) or mitochondrial disease. 2. Proven CNS infection (e.g. bacterial meningitis, herpes simplex encephalitis, Lyme disease, HIV). 3. Describe disease modifying therapy usage, efficacy, and adverse effects. 3. Radiation/chemotherapy associated white matter damage. 4. Provide a platform for future a) identification of biological markers in MS, b) genetic studies and c) hypothesis driven biological studies; by storing blood and CSF samples already taken at the time of first demyelinating event (currently in 7 participating centres). 4. Condition fulfilling criteria for CNS connective tis- sue disease e.g. lupus, vasculitis. The sole presence of antibodies associated with CNS connective tissue dis- eases is not sufficient for exclusion. Study aims and objectives New hypotheses will be added if suggested by publica- tions by other centres prior to data analysis. PUDDLS aims to establish if prognostic factors for relapse of CNS inflammatory demyelination (and hence risk of progression to MS) can be identified for UK chil- dren (Figure 1). PUDDLS is a prospective UK based study with the following objectives to: Context of study in relation to current practice (iii) Children with multiphasic ADEM compared to children with recurrent ADEM are more likely to develop MS. (iii) Children with multiphasic ADEM compared to children with recurrent ADEM are more likely to develop MS. y p Children with CNS inflammatory demyelination are most often routinely seen by a Paediatric Neurologist as part of clinical care in order to provide expert advice and management. In line with guidelines by the Interna- tional Paediatric MS Group, demyelination clinics have been set up in several Tertiary Centres by paediatric neurologists with an interest in these conditions. This led to the UK and Ireland Paediatric CNS Inflammatory Demyelination Working Group being set up, providing a steering committee for future collaborative studies, net- working, future direction, writing clinical guidelines and future grant applications. As a result, the majority of these children are seen in these specialist centres where they will be recruited. p (iv) Children with recurrent ADEM are more likely to develop MS than children with monophasic ADEM. (iv) Children with recurrent ADEM are more likely to develop MS than children with monophasic ADEM. (v) MRI criteria (McDonald, KIDMUS) can be vali- dated to distinguish between ADEM, ADEM variants, CIS, and MS. (v) MRI criteria (McDonald, KIDMUS) can be vali- dated to distinguish between ADEM, ADEM variants, CIS, and MS. (vi) Identified future biomarkers with suggested prog- nostic significance, can be tested against PUDDLS outcomes. (vii) Early treatment with disease modifying therapy in relapsing CNS Inflammatory Demyelination, reduces relapse rate and slows neurological disability progress. (viii) Environmental risk factors can be identified for children with first episode CNS inflammatory demyeli- nation and MS. Specific Hypotheses We wish to test the following hypotheses (as suggested by the IPMSSG [17]): We wish to test the following hypotheses (as suggested by the IPMSSG [17]): (i) Children who present with CIS are more likely to develop MS than children who present with ADEM. (i) Children who present with CIS are more likely to develop MS than children who present with ADEM. (ii) ADEM Children with prolonged steroid- depen- dent events (≥3 months) are more likely to develop MS than children with rapid recovery ADEM. (ii) ADEM Children with prolonged steroid- depen- dent events (≥3 months) are more likely to develop MS than children with rapid recovery ADEM. (ii) ADEM Children with prolonged steroid- depen- dent events (≥3 months) are more likely to develop MS than children with rapid recovery ADEM. Relapses, MS, and Progression 5. Ensure that long term outcomes of childhood demyelinating disease are as complete as possible by flagging patients with their consent with the NHS Infor- mation Centre [Medical Research Information Service (MRIS)] which will allow them to be traced in future via their G.P.s (currently in 7 selected centres). A relapse is a new occurrence of neurological symptoms that lasted > 24 hours and stabilized or resolved either partially or completely. MS is defined in table 2 below. Progression of MS is a continuous worsening of symp- toms and signs for > 6 months, ± superimposed Absoud et al. BMC Pediatrics 2011, 11:68 Page 4 of 10 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 http://www.biomedcentral.com/1471-2431/11/68 Figure 1 PUDDLS Flowchart with link to epidemiological study. Figure 1 PUDDLS Flowchart with link to epidemiological study. up. At recruitment an initial Case Report Form (CRF) will be completed by the PI/Research Nurse/Research Fellow. MRI copies and biological samples will be archived. An environmenal questtionaire will be admi- nistered by telephone interview within three months of disease onset. Patient reported and proxy outcome ques- ttionaires will be completed online at onset and at speci- fied time intervals (Figure 3). At six months and one year from onset, and then at yearly intervals, a follow-up relapses, either primarily (at onset of disease) or secon- darily (after first relapse). Table 2 Multiple Sclerosis definition Table 2 Multiple Sclerosis definition Multiple Sclerosis (MS) - Two or more non ADEM episodes of CNS demyelination separated in time (4 weeks) and space. For children aged > 10 years: - Dissemination in space can be met if: MRI shows three of: 1) ≥9 white matter lesions or 1 gadolinium enhancing lesion, 2) ≥3 periventricular lesions, 3) One juxtacortical lesion, 4) an infratentorial lesion. OR abnormal CSF (oligoclonal bands or elevated IgG index) with 2 lesions on the MRI (one in the brain). - Dissemination in time can be met if: MRI shows new T2 or gadolinium enhancing lesions developing ≥3 months after initial event. - Two or more non ADEM episodes of CNS demyelination separated in time (4 weeks) and space. For children aged > years: Multiple Sclerosis (MS) - Dissemination in space can be met if: MRI shows three of: 1) ≥9 white matter lesions or 1 gadolinium enhancing les ≥3 periventricular lesions, 3) One juxtacortical lesion, 4) an infratentorial lesion. j OR abnormal CSF (oligoclonal bands or elevated IgG index) with 2 lesions on the MRI (one in the brain). - Dissemination in time can be met if: MRI shows new T2 or gadolinium enhancing lesions developing ≥ initial event. CRF will be completed in clinic. A telephone interview will also be conducted at the same frequency to ask if the patient has had a clinical relapse or hospital admis- sion and to provide an opportunity to ask questions. Any significant concerns will be referred back to the GP or clinician in charge. Patients will be ascertained via these available meth- ods, and given or sent information leaflets before being consented (minimum 24 hours) to take part in the study. A website (http://www.childdemyelination.org.uk) is set up to provide newsletters and information regard- ing study progress, and up to date information on man- agement directed at Paediatricians. If the patient or carer requests referral, normal clinical follow up pat- terns of care will be followed. Design PUDDLS ia a observational cohort study with sample biobanking which will characterize the spectrum of CNS demyelinating syndromes, their outcomes, and possible progression to MS (Figure 2). Children will be recruited over a 4 year and 6 month study period and followed Table 1 PUDDLS Inclusion Criteria Condition Inclusion Criteria Acute Disseminated Encephalo-myelitis (ADEM) A polysymptomatic clinical event with acute/subacute onset that must include encephalopathy (behavioural change or altered consciousness). (2) MRI Brain shows f multifocal lesions, predominantly involving white matter. • Relapsing ADEM: symptoms or signs within 3 months of initial onset of ADEM. IF a new event occurs ≥3 months later and ≥1 month after completing steroid treatment, it is defined as: • Recurrent ADEM: recurrence of initial symptoms without involvement of new clinical areas. • Multiphasic ADEM: New event, but involving new anatomical areas of the CNS. Clinically Isolated Syndrome (CIS) A first acute-clinical episode of CNS symptoms which may either be monofocal or multifocal, but does not include encephalopathy (except in brainstem syndromes). The MRI will show area of white matter demyelination. These include: Transverse myelitis: weakness and/or numbness of both legs +/- arms, usually with maximal deficits 1 week after symptom onset supported by demyelination on MRI spine. Brainstem, cerebellar, and/or hemisphericdysfunction, supported by demyelination on MRI. Optic Neuritis (ON) Acute or subacute loss of vision and ≥1 of: relative afferent pupillary defect (unilateral cases), visual field deficit or scotoma, impaired colour vision, optic disc oedema, or abnormal visual evoked potentials. MRI is not necessary for diagnosis. Neuro-myelitis Optica (NMO) Three of the following four criteria: i. Optic neuritis ii. Acute myelitis iii. Spinal MRI lesion extends over three or more segments iv. NMO antibody testing is positive. Table 1 PUDDLS Inclusion Criteria Table 1 PUDDLS Inclusion Criteria Page 5 of 10 Page 5 of 10 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Patient recruitment and consent Based on an expected incidence of 1-2/100,000 children, each Centre is expected to recruit approximately 10-15 cases/year over the 5 year length of the study, with a minimum of 6 months follow up. Children with CNS Inflammatory Demyelination are seen by Paediatric Neurologists with an interest in CNS Demyelination to provide expertise in diagnosis and management. Patients usually present to Neurologists’ attention via the follow- ing routes: Currently in 7 centres, PIs will obtain consent from families presenting to their centres. The Paediatric Research Nurse at each site (0.1 WTE at each site, except lead Centre [Birmingham] to have 0.3 WTE) or the Research Fellow (Dr M Absoud) may also approach families directly to obtain consent. Appropriate agree- ments and contracts will be in place for the Research Fel- low and Nurse as recommended by the National Institute for Health Research. Patients will be asked to consent to: - Presentation directly to Tertiary Centre. - Presentation directly to Tertiary Centre. - Taking part in the study and longitudinal follow up via The National Health Service Information Centre [Medical Research Information Service (MRIS)] - Referral from Paediatric DGH Consultant for admis- sion or an outpatients opinion. - MRI scans referred to weekly Tertiary Centre Neu- roradiology meeting for an opinion (permission to approach the clinician for referral will be obtained). - Storage of Samples (CSF, DNA, serum) - Storage of Samples (CSF, DNA, serum) - Obtaining MRI copies and information from Hospi- tal notes - Obtaining MRI copies and information from Hospi- tal notes - Shared care or outreach clinic. - Shared care or outreach clinic. Figure 2 The 5 major study themes are represented in flow the Figure below. igure 2 The 5 major study themes are represented in flow the Figure below. Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Page 6 of 10 Figure 3 Data and Outcome Collection Schedule Summary. Figure 3 Data and Outcome Collection Schedule Summary. - Sharing an anonymised dataset with the Interna- tional Paediatric Study MS Group. adaptive behavioural functioning skills, MRI criteria, and objective EDSS disability scores (ie. EDSS > 5.5, patient unable to walk without aid). Health related Quality of Life (QOL) - Generic QOL: Pediatric Quality of Life Inventory TM; Generates physical, psychological, and total score Generic QOL: Pediatric Quality of Life Inventory nerates physical, psychological, and total score ○Self report for ages 5-25 years. ○Parent report for ages 2-25 years. Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Page 7 of 10 (i) Conceptual skills: communication, functional aca- demics, and self-direction skills areas e. Information which might predict the subsequent development of recurrent ADEM, multiphasic ADEM, or MS, like details of the anatomic distribution (optic nerve or spinal cord, monofocal or multifocal distribu- tion), CSF findings, serological and neuroimaging results. (ii) Social skills: Social, and leisure skills areas Cognitive ○Two Healthtracker online games played by children aged 5-16 years that aim to measure memory, attention and concentration. These are embedded in between questionnaires. f. Treatment (eg, steroids, immunoglobulins, intensive care) given. g. Information as to whether the risk of future MS was given to the family. g. Information as to whether the risk of future MS was given to the family. ○Online D2 test game for participants aged 9 years- 60 years. The D2 Test measures processing speed and quality of performance of individuals, allowing for a neuropsychological estimation of individual attention and concentration performance. h. Neurological examination, Kurtze EDSS scales, and functional status. h. Neurological examination, Kurtze EDSS scales, and functional status. For 6 month and annual outcome data, follow up information will be obtained on neurological function and further demyelinating events (if these have occurred). We aim to collect 90% data within a window period of one month on either side of the designated follow up periods. Biobank sample collection Samples and Procedures - Disease specific QOL: Multiple Sclerosis Impact Scale-29 version 2 (MSIS-29 v2). Additional informed consent will be obtained to store sera, DNA, and CSF collected during routine clinical care at first onset of a CNS demyelinating event, and then at future hospital visits if clinical relapses occur. This will be done in accordance with the Human Tissue Act (HTA, 2004). Currently designated laboratories in each of the seven establishments hold appropriate HTA licences and would be responsible for storage: Generates physical, and psychological scores ○Self report for children 5 years -adulthood. MRI criteria Routine patient MRIs will be collected in order to apply proposed criteria (such as McDonald, KIDMUS) that the study team can validate against PUDDLS outcomes. Data will be collected using a standardised data collec- tion performa with agreed definitions designed by Neuro-radiology collaborators (available on request). MRIs will be randomly evaluated by one of four Neuror- adiologists blinded to presentation and clinical features. Inter-rater reliability will be assessed. Case Report Forms (CRFs) Data collection will take place at recruitment, 6 months after initial event, annually after initial event, and during clinical relapses (see Figure 3). Clinical follow ups for patients affected with CNS inflammatory demyelinating diseases will be part of clinical care, and no extra hospi- tal visits will be required. Children and parents/carers will be able to fill in questionnaires from a secure online programme (Healthtracker TM). Access to a computer with privacy will be made accessible if needed during clinic visits. The Research Nurse will make contact with the family at the designated follow up times, to ensure they understand the research, ask about any relapses or investigations the child might have had, and discuss further queries. The CRFs have been adapted with permission after consultation with the International Group. The initial CRF (first episode) will be completed by the PI, the Research Fellow, or the Research Nurse with assistance from the PI. A condition specific targeted neurological history (using a check box questionnaire) would take place at the time of first demyelinating event. The event would be classified as either ADEM, CIS, or MS (see case definition). Data collected can broadly be categorised to: a. Basic demographic data (ethnicity, place of birth, current residence). b. Family history of MS or autoimmune disease. The primary outcome measures will be (1) if the patient developed a clinical relapse, and (2) a diagnosis MS. Secondary outcome measures will include: Patient reported and proxy Quality of life, disease impact, neu- ropsychiatric & fatigue symptoms, conceptual & social c. Demyelinating symptoms and signs (using tick box table). d. Presence of behavioural disturbance or encephalo- pathy (to distinguish ADEM from CIS) at time of demyelinating event. Environmental Questionnaire One off telephone interview (30-45 minutes) with par- ent/carer to be conducted during study period within three months of disease onset if direct hospital contact whilst inpatient has not been achieved. Only once further funding is gained, case controls for administer- ing only the environmental questionnaire will be ascer- tained via a ‘buddy system’ to the child who is age and sex matched. No other involvement of controls will be required. Healthtracker®; patient self report and proxy outcomes PUDDLS will use Healthtracker® which is a sophisti- cated, user-friendly, UK, and web-based assessment sys- tem launched with Department of Health support in 2008. It comprises a suite of questionnaires and test games for children and their carers, and not for investi- gator use. These allow accurate measures of change across a wide range of symptoms, side-effects, psycholo- gical functions and quality of life. Assessments can be home-based for convenience and lower costs. This helps circumvent the many practical difficulties associated with long-term assessment of chronic diseases. Health related Quality of Life (QOL) Expected Numbers . Nottingham University Hospitals NHS Trust 7. Nottingham University Hospitals NHS Trust Based on an expected incidence of 1-2/100,00 children, the expected recruitment over the initial 5 year study period is 400-450 cases (12-13/year in every Centre). Sample collection, transport and storage will be undertaken using standard SOPs and be based at colla- borating centres as appropriate. There will be no additional venepuncture or lumbar puncture procedures for the child beyond those required for routine clinical care. Informed consent will be required, for extra volumes to be collected during these procedures for the study. Residues of CSF and blood already taken as part of the clinical management of the patient, will also be used for future research investiga- tions and to supplement the specimen archive. For each patient at least one CSF sample and two blood samples will be required. Clinicians will be encouraged to follow existing guidelines for the investigation of CNS demyeli- nating disorders (17,18) and as updated in the future. Longer term follow up To ensure that long term outcomes of childhood demyelinating disease can be identified by flagging patients with their consent via the NHS Information Centre (MRIS) and NHS Central Register which will allow them to be traced in the future via their G.P.s. We will consent patients for long term follow-up. Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Data Handling and Integrity • Each patient enrolled will be assigned a unique PUDDLS study identification number. This number will be used in all future correspondence between the study centre and principal investigators. Statistics and data analysis Descriptive statistics (with confidence intervals where appropriate) will be used to summarise the key compo- nents of the database. Univariate associations between potential prognostics factors and potential confounders and with MS diagnosis or other CNS demyelinating epi- sodes; outcomes. These associations will be further explored in logistic regression models that will allow for control of confounding factors. Factors associated with early occurrence of second episodes will also be explored and Cox models will be used if the propor- tional hazards assumption is met. Life table analysis methods will be used to calculate primary outcome time points. SPSS (v.17.0) statistical software will be used for data analysis. Confidentiality Samples will be stored pseudoanonymised in HTA approved labs, and linked with the clinical data by the unique PUDDLS number. Any information that is likely to identify participants (such as name, address, date of birth) will be removed from samples. Personal identifiers are kept separately under strict control with restricted access in secured cabinets, at the Institute of Child Health, Birmingham Children’s Hospital in a secure locked area. Access to data is limited to the Chief and Principal Investigators. The study adminis- trator at Birmingham Children’s Hospital will ensure pseudo-anonymous tracking of the samples, solely for the purposes of linkage once they are collected as informed by the Principal Investigator. The location, nature and date of samples will be recorded on the database. Research and development procedures Research & Development (R&D) approval from each participating NHS Trust will be obtained prior to recruitment. Principal investigators will follow rules of data protection ethics and confidentiality. Longer term follow up Research & Development (R&D) approval from each participating NHS Trust will be obtained prior to recruitment. Principal investigators will follow rules of data protection ethics and confidentiality. Fatigue - Pediatric Quality of Life Inventory TM; Multidimen- sional Fatigue Scale ○Self report for children 5-25 years ○Parent report; for children 2- 25 years Neuropsychiatric Symptoms (http://www.hta.gov.uk/licensingandinspections/listofli- censedestablishments.cfm) - Profile Of Neuropsychiatric Symptoms (PONS) ○Self report for children 5-16 years - Profile Of Neuropsychiatric Symptoms (PONS) ○Self report for children 5-16 years ○Parent report Functioning; Adaptive behavioural skills ○Adaptive Behaviour Assessment System II (ABAS II). 1. Alder Hey Children’s NHS Foundation Trust 2. Birmingham Children’s Hospital 3. Frenchay Hospital Functioning; Adaptive behavioural skills ○Adaptive Behaviour Assessment System II (ABAS II). 4. Great Ormond Street Hospital - Parent report for children 5 years -adulthood. Two out of the three domains will be used: 5. Guy’s & St Thomas’ Hospitals, Cellular Pathology 6. John Radcliffe Hospital Page 8 of 10 Page 8 of 10 Future research f d Separate funding will be sought for the biological/ genetic sample studies and a further five year longitudi- nal follow up, once sufficient samples and clinical data has been collected. The details of the studies are beyond the scope of this protocol but we have identified future potential studies of sufficient scientific validity to ethi- cally justify the storage of human samples. Collabora- tions with international experts and leaders in the field will maximise yield of investigations, and ensure a thor- ough review process: • Principal investigators will be responsible for the security of data at each local site according to their employing authorities’ rules and regulations. Investiga- tors will ensure that the study is conducted and data are generated, documented and reported in compliance with the protocol, and applicable regulatory requirements. • Patient identifiable information (name and address) will be sent to the study office on only one occasion (initial case report form) by secure fax. The original signed consent form to participate in the study will be kept in the patient’s medical records unless required otherwise by local regulations, and a copy kept securely in the site file. • Patient identifiable information (name and address) will be sent to the study office on only one occasion (initial case report form) by secure fax. The original signed consent form to participate in the study will be kept in the patient’s medical records unless required otherwise by local regulations, and a copy kept securely in the site file. The PUDDLS Biological Studies Steering Group will consider any future studies, and follow the agreed guidelines for study selection (available on request). Once studies are deemed of sufficient quality, and depending on the strand of study the group will colla- borate with the experts in their fields and with the International Paediatric Multiple Sclerosis Study Group (IPMSSG). • At the central study office (ICH, Birmingham Chil- dren’s Hospital), all patient identifiable study data will Page 9 of 10 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 SQL Server. Alternatively custom exports can be written. be stored in locked filing cabinets. Identifiable patient data will be separated from the clinical data and linked with a unique study code. Clinical data used for research may also include patient identifiable data, e.g. date of birth, sex, that are also important for the data analysis, linkage during follow up, and to answer the research questions. (ii)Healthtracker TM Healthtracker is hosted with Rackspace on a single server running the Microsoft Windows Server 2003 Standard Edition operating system. Rackspace ensures the server is continuously updated and has all the latest service packs and security patches. The server itself houses a Dual-Core AMD Opteron processor running at 2.20 GHz with 4.0 GB of RAM. It has three 73 GB SCSI hard drives in a RAID 5 configuration offering 140 GB of usable storage. The server acts as both the web and database server running Internet Information Services (IIS) and SQL Server 2005 Standard Edition. Future research f d These cannot be removed from the question- naire and will be stored for 30 years in secure archives (MRC guidelines). Only pseudo-anonymised clinical data only will be entered into the electronic study database for analysis. Case report forms will be kept in the site file in a secured locked cabinet, and a copy will be sent to the Central study office at the Institute of Child Health, Birmingham Children’s Hospital. Healthtracker (HT) Results are generated in real time and can reflect symptom trends/treatment efficacy/or changes over time. The program has been designed with the ability to analyse large amounts of longitudinal data with inbuilt real-time Bayesian/Neural Net algorithms. This system allows testing of hypothesis as new data comes in. The online presentation of HealthTracker has already meant that rapid acquisition of data for norm- ing, or comparing new measures with existing ones is possible. Non-linear mixed effects modelling will be used to generate estimates of expected recoveries. Data from HT will also be exported via an excel spreadsheet for the research team to analyse. Item analysis of ques- tionnaires will be via Rasch analysis, utilising the WIN- STEPS TM statistical software. Acknowledgements and funding This study is funded by a grant from the UK Multiple Sclerosis Society (893/ 08) and Action Medical Research (SP4472). The study is ethically approved by the South Birmingham Research Ethics Committee (09/H1207/160) in the UK. The protocol has also been developed and approved with the help of the UK Children’s Neurological Research Campaign. The study is supported by the Medicines for Children Research Network and the Birmingham Children’s Hospital Wellcome Trust Clinical Research Facility. Both the admin and non-admin websites use SSL 128 bit encryption to ensure data is securely encrypted between requests. The websites are written in C# and run under the Microsoft .Net 2.0 framework. HTML and cascading style sheets are used in both sites to apply common styling. All database transactions are per- formed via stored procedures (there is no in-line SQL). Flash is used to provide some questionnaire and game content to the child portion of the non-admin website. XML web services are used to send and receive infor- mation to and from these components. All data in the system is stored in a single SQL Server database and therefore can be exported in any format facilitated by List of abbreviations ADEM A t Di i ADEM: Acute Disseminated Encephalomyelitis; CNS: Central Nervous System; CIDD: CNS Inflammatory Demyelinating Disease; CRF: Case Report Form; CSF: Cerebrospinal Fluid; EDSS: Expanded Disability Status Scale; Health Tracker®®: A web-based assessment system which comprises a suite of questionnaires and test games for children. These allow accurate measures of change across a wide range of symptoms, side-effects, psychological functions and quality of life. Assessments can be home-based for convenience and lower costs; HTA: Human Tissue Act 2004 (http://www.opsi.gov.uk/acts/acts2004/ ukpga_20040030_en_1); IPMSSG: International Paediatric Multiple Sclerosis Study Group; Life-H: Assessment of Life Habits Questionnaire; MCRN: Medicines for Children Research Network; MRIS: Medical research Information Service; MS: Multiple Sclerosis; MSIS-29: The Multiple Sclerosis Impact Scale; Parent report version; NIHR: National Institute for Health Research; NMO: Neuro-myelitis Optica; ON: Optic Neuritis; PedsQL: Pediatric Quality of Life Inventory TM; PUDDLS: Paediatric UK Demyelinating Disease Longitudinal Study; SDQ: Strengths and Difficulties Questionnaire; SOP: Standard Operating Procedure; TM: Transverse Myelitis; UCID-SS: UK and Ireland Childhood Inflammatory Demyelination Surveillance Study; UKCRN: UK Clinical Research Network; UKCNRC: UK Children’s Neurological Research Campaign; Provides neurology network and study management support; WTCRF: Wellcome Trust Clinical Research Facility. • Database Information (i) Clinical Information The research nurse, data administrator, and/or the research fellow will enter and maintain the clinical information for all study participants from within the Institute of Child Health (ICH), Birmingham Children’s Hospital. The database will be anonymised. CRFs will be sent to the study team on paper. This has the added benefit of providing consistency of data entry, allowing for more complete and accurate set of data for long term analysis. This database will use the PUDDLS code only. The data base will be password protected, backed up by the ICH server backup daily, and encrypted for added protection. The database will be accessed accord- ing to Trust policies and procedures. • Data protection, security and confidentiality will be implemented according to the UK Data Protection Act 1998 and each Trust’s own Information Governance Policy. Authors’ contributions 20. Ebers GC: Environmental factors and multiple sclerosis. Lancet Neurol 2008, 7:268-277. All authors participated in the design of the study. MA, ML and EW drafted the manuscript. CH, MP, CD, PJ and EW edited the manuscript. All authors read and approved the manuscript. All authors participated in the design of the study. MA, ML and EW drafted the manuscript. CH, MP, CD, PJ and EW edited the manuscript. All authors read and approved the manuscript. 21. Orton SM, Herrera BM, Yee IM, Valdar W, Ramagopalan SV, Sadovnick AD, Ebers GC: Sex ratio of multiple sclerosis in Canada: a longitudinal study. Lancet Neurol 2006, 5:932-936. References Waubant E, Hietpas J, Stewart T, Dyme Z, Herbert J, Lacy J, Miller C, Rensel M, Schwid S, Goodkin D: Interferon beta-1a in children with multiple sclerosis is well tolerated. Neuropediatrics 2001, 32:211-213. 10. Mikaeloff Y, Adamsbaum C, Husson B, Vallee L, Ponsot G, Confavreux C, Tardieu M, Suissa S: MRI prognostic factors for relapse after acute CNS inflammatory demyelination in childhood. Brain 2004, 127:1942-1947. 33. Pohl D, Waubant E, Banwell B, Chabas D, Chitnis T, Weinstock-Guttman B, Tenembaum S: Treatment of pediatric multiple sclerosis and variants. Neurology 2007, 68:S54-65. 33. Pohl D, Waubant E, Banwell B, Chabas D, Chitnis T, Weinstock-Guttman B, Tenembaum S: Treatment of pediatric multiple sclerosis and variants. Neurology 2007, 68:S54-65. 11. Pohl D, Hennemuth I, von Kries R, Hanefeld F: Paediatric multiple sclerosis and acute disseminated encephalomyelitis in Germany: results of a nationwide survey. Eur J Pediatr 2007, 166:405-412. Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 Absoud et al. BMC Pediatrics 2011, 11:68 http://www.biomedcentral.com/1471-2431/11/68 16. Tenembaum S, Chamoles N, Fejerman N: Acute disseminate 16. Tenembaum S, Chamoles N, Fejerman N: Acute disseminated encephalomyelitis: a long-term follow-up study of 84 pediatric patients. Neurology 2002, 59:1224-1231. Street Hospital for Children, Great Ormond Street, London, WC1N 3JH, UK. 7Department of Neurology, Bristol Royal Hospital for Children, Upper Maudlin Street, Bristol, BS2 8BJ, UK. 8Department of Neurology, Alder Hey Children’s NHS Foundation Trust, Eaton Road, Liverpool, L12 2AP, UK. 9Department of Neuroradiology, Bristol Royal Hospital for Children, Upper Maudlin Street, Bristol, BS2 8BJ, UK. 10Department of Neurology, The Evelina Children’s Hospital at Guy’s & St Thomas’ NHS Trust, Great Maze Pond, London, SE1 9RT, UK. 11Department of Neurology, Oxford Children’s Hospital, Headley Way, Oxford, OX3 9DU, UK. 12Department of Neuroradiology, King’s College Hospital NHS Trust, Denmark Hill, London, SE5 9RS, UK. 13Academic Division of Child Health, Queen’s Medical Centre, Nottingham, NG7 2UH, UK. 17. Krupp LB, Banwell B, Tenembaum S: Consensus definitions proposed for pediatric multiple sclerosis and related disorders. Neurology 2007, 68: S7-12. 18. Banwell B, Ghezzi A, Bar-Or A, Mikaeloff Y, Tardieu M: Multiple sclerosis in children: clinical diagnosis, therapeutic strategies, and future directions. Lancet Neurol 2007, 6:887-902. 19. Pohl D, Krone B, Rostasy K, Kahler E, Brunner E, Lehnert M, Wagner HJ, Gartner J, Hanefeld F: High seroprevalence of Epstein-Barr virus in children with multiple sclerosis. Neurology 2006, 67:2063-2065. Received: 17 May 2011 Accepted: 28 July 2011 Published: 28 July 2011 Received: 17 May 2011 Accepted: 28 July 2011 Published: 28 July 2011 23. Tintore M, Rovira A, Martinez MJ, Rio J, Diaz-Villoslada P, Brieva L, Borras C, Grive E, Capellades J, Montalban X: Isolated demyelinating syndromes: comparison of different MR imaging criteria to predict conversion to clinically definite multiple sclerosis. AJNR Am J Neuroradiol 2000, 21:702-706. References Brass SD, Caramanos Z, Santos C, Dilenge ME, Lapierre Y, Rosenblatt B: Multiple sclerosis vs acute disseminated encephalomyelitis in childhood. Pediatr Neurol 2003, 29:227-231. 5. Brass SD, Caramanos Z, Santos C, Dilenge ME, Lapierre Y, Rosenblatt B: Multiple sclerosis vs acute disseminated encephalomyelitis in childhood. Pediatr Neurol 2003, 29:227-231. 28. Bielekova B, Martin R: Development of biomarkers in multiple sclerosis. Brain 2004, 127:1463-1478. 6. Brex PA, Miszkiel KA, O’Riordan JI, Plant GT, Moseley IF, Thompson AJ, Miller DH: Assessing the risk of early multiple sclerosis in patients with clinically isolated syndromes: the role of a follow up MRI. J Neurol Neurosurg Psychiatry 2001, 70:390-393. 29. Ghezzi A, Amato MP, Capobianco M, Gallo P, Marrosu MG, Martinelli V, Milanese C, Moiola L, Milani N, La Mantia L, et al: Treatment of early-onset multiple sclerosis with intramuscular interferonbeta-1a: long-term results. Neurol Sci 2007, 28:127-132. g y y 7. Dale RC, de Sousa C, Chong WK, Cox TC, Harding B, Neville BG: Acute disseminated encephalomyelitis, multiphasic disseminated encephalomyelitis and multiple sclerosis in children. Brain 2000, 123(Pt 12):2407-2422. 30. Mikaeloff Y, Moreau T, Debouverie M, Pelletier J, Lebrun C, Gout O, Pedespan JM, Van Hulle C, Vermersch P, Ponsot G: Interferon-beta treatment in patients with childhood-onset multiple sclerosis. J Pediatr 2001, 139:443-446. 30. Mikaeloff Y, Moreau T, Debouverie M, Pelletier J, Lebrun C, Gout O, Pedespan JM, Van Hulle C, Vermersch P, Ponsot G: Interferon-beta treatment in patients with childhood-onset multiple sclerosis. J Pediatr 2001, 139:443-446. 8. Hanefeld F, Bauer HJ, Christen HJ, Kruse B, Bruhn H, Frahm J: Multiple sclerosis in childhood: report of 15 cases. Brain Dev 1991, 13:410-416. 31. Pohl D, Rostasy K, Gartner J, Hanefeld F: Treatment of early onset multiple sclerosis with subcutaneous interferon beta-1a. Neurology 2005, 64:888-890. 31. Pohl D, Rostasy K, Gartner J, Hanefeld F: Treatment of early onset multiple sclerosis with subcutaneous interferon beta-1a. Neurology 2005, 64:888-890. 9. Hynson JL, Kornberg AJ, Coleman LT, Shield L, Harvey AS, Kean MJ: Clinical and neuroradiologic features of acute disseminated encephalomyelitis in children. Neurology 2001, 56:1308-1312. 32. Waubant E, Hietpas J, Stewart T, Dyme Z, Herbert J, Lacy J, Miller C, Rensel M, Schwid S, Goodkin D: Interferon beta-1a in children with multiple sclerosis is well tolerated. Neuropediatrics 2001, 32:211-213. 32. References 1. Ness JM, Chabas D, Sadovnick AD, Pohl D, Banwell B, Weinstock-Guttman B: Clinical features of children and adolescents with multiple sclerosis. Neurology 2007, 68:S37-45. 1. Ness JM, Chabas D, Sadovnick AD, Pohl D, Banwell B, Weinstock-Guttman B: Clinical features of children and adolescents with multiple sclerosis. Neurology 2007, 68:S37-45. 24. Banwell B, Shroff M, Ness JM, Jeffery D, Schwid S, Weinstock-Guttman B: MRI features of pediatric multiple sclerosis. Neurology 2007, 68:S46-53. 2. Duquette P, Murray TJ, Pleines J, Ebers GC, Sadovnick D, Weldon P, Warren S, Paty DW, Upton A, Hader W: Multiple sclerosis in childhood: clinical profile in 125 patients. J Pediatr 1987, 111:359-363. 2. Duquette P, Murray TJ, Pleines J, Ebers GC, Sadovnick D, Weldon P, Warren S, Paty DW, Upton A, Hader W: Multiple sclerosis in childhood: clinical profile in 125 patients. J Pediatr 1987, 111:359-363. p p gy 25. Hahn CD, Shroff MM, Blaser SI, Banwell BL: MRI criteria for multiple l i E l ti i di t i h t N l 2004 62 806 808 gy 25. Hahn CD, Shroff MM, Blaser SI, Banwell BL: MRI criteria for multiple sclerosis: Evaluation in a pediatric cohort. Neurology 2004, 62:806-808. sclerosis: Evaluation in a pediatric cohort. Neurology 2004, 62:806-80 26. Swanton JK, Rovira A, Tintore M, Altmann DR, Barkhof F, Filippi M, Huerga E, Miszkiel KA, Plant GT, Polman C, et al: MRI criteria for multiple sclerosis in patients presenting with clinically isolated syndromes: a multicentre retrospective study. Lancet Neurol 2007, 6:677-686. 3. Banwell B, Krupp L, Kennedy J, Tellier R, Tenembaum S, Ness J, Belman A, Boiko A, Bykova O, Waubant E, et al: Clinical features and viral serologies in children with multiple sclerosis: a multinational observational study. Lancet Neurol 2007, 6:773-781. 27. Polman CH, Reingold SC, Edan G, Filippi M, Hartung HP, Kappos L, Lublin FD, Metz LM, McFarland HF, O’Connor PW, et al: Diagnostic criteria for multiple sclerosis: 2005 revisions to the “McDonald Criteria”. Ann Neurol 2005, 58:840-846. 4. Boiko A, Vorobeychik G, Paty D, Devonshire V, Sadovnick D: Early onset multiple sclerosis: a longitudinal study. Neurology 2002, 59:1006-1010. 4. Boiko A, Vorobeychik G, Paty D, Devonshire V, Sadovnick D: Early onset multiple sclerosis: a longitudinal study. Neurology 2002, 59:1006-1010. 5. Brass SD, Caramanos Z, Santos C, Dilenge ME, Lapierre Y, Rosenblatt B: M lti l l i t di i t d h l liti i hildh d 5. Competing interests 22. Dean G, Elian M: Age at immigration to England of Asian and Caribbean immigrants and the risk of developing multiple sclerosis. J Neurol Neurosurg Psychiatry 1997, 63:565-568. The authors declare that they have no competing interests. The authors declare that they have no competing interests. Author details 1 1School of Health & Population Sciences, University of Birmingham, Vincent Drive, Birmingham, B15 2TT, UK. 2Department of Neurology, Birmingham Children’s Hospital, Steelhouse Lane, Birmingham, B4 6NH, UK. 3Department of Neuroradiology, Great Ormond Street Hospital for Children, Great Ormond Street, London, WC1N 3JH, UK. 4Department of Paediatric Neurology, Royal Preston Hospital, Sharoe Green Lane North, Lancashire PR2 9HT, UK. 5Department of Neuroradiology, Birmingham Children’s Hospital, Steelhouse Lane, Birmingham, B4 6NH, UK. 6Department of Neurology, Great Ormond Page 10 of 10 Pre-publication history 12. Renoux C, Vukusic S, Mikaeloff Y, Edan G, Clanet M, Dubois B, Debouverie M, Brochet B, Lebrun-Frenay C, Pelletier J, et al: Natural history of multiple sclerosis with childhood onset. N Engl J Med 2007, 356:2603-2613. The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1471-2431/11/68/prepub The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1471-2431/11/68/prepub The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1471-2431/11/68/prepub doi:10.1186/1471-2431-11-68 Cite this article as: Absoud et al.: Paediatric UK demyelinating disease longitudinal study (PUDDLS). BMC Pediatrics 2011 11:68. doi:10.1186/1471-2431-11-68 Cite this article as: Absoud et al.: Paediatric UK demyelinating disease longitudinal study (PUDDLS). BMC Pediatrics 2011 11:68. 13. Simone IL, Carrara D, Tortorella C, Liguori M, Lepore V, Pellegrini F, Bellacosa A, Ceccarelli A, Pavone I, Livrea P: Course and prognosis in early- onset MS: comparison with adult-onset forms. Neurology 2002, 59:1922-1928. 14. Stonehouse M, Gupte G, Wassmer E, Whitehouse WP: Acute disseminated encephalomyelitis: recognition in the hands of general paediatricians. Arch Dis Child 2003, 88:122-124. 15. Tardieu M, Mikaeloff Y: What is acute disseminated encephalomyelitis (ADEM)? Eur J Paediatr Neurol 2004, 8:239-242.
https://openalex.org/W4387183026	https://ijbms.net/assets/files/1694280950.pdf	English	null	Forecasting Exchange Rates Comparison In The Presence Of Instabilities	International journal of business & management studies	2,023	cc-by	10,200	Keywords Keywords Traditional Tests, Fluctuation Tests, Predictive Ability, Symmetric GARCH Models, Asymmetric G Models, South African Exchange Rate Azwifaneli I. Nemushungwa1 Azwifaneli I. Nemushungwa1 1Commerce and Law, Faculty of Management Sciences, University of Venda, South Africa 1Commerce and Law, Faculty of Management Sciences, University of Venda, South Africa IPRPD International Journal of Business & Management Studies ISSN 2694-1430 (Print), 2694-1449 (Online) Volume 04; Issue no 09: September, 2023 DOI: 10.56734/ijbms.v4n9a6 Abstract Exchange rate forecasting is an inherent approach in financial risk management, yet previous forecasting models were criticized for their poor predictive ability, mainly during periods of exceptional macroeconomic weaknesses. This is attributed to their failure to identify the importance and strength of key transmission and amplification channels, especially those linked to financial markets and uncertainty. Though there is no model that can be precise, especially during periods of crises, it is important to find a model that can yield near-accurate results. The present study therefore evaluates the different forecasting models, considering how each handles instabilities. The Rossi Sekhposyan forecast rationality test results reveal that the EGARCH model under general error distribution and APARCH under normal error distribution show the strongest evidence against rationality around the year 2009, identifying the concentration of instabilities during that time. This vindicates the need to control for instabilities in forecasting. This implies that, in the presence of instabilities, the fluctuation tests are more powerful than traditional tests. 1. Introduction 04 - Issue:9/September_2023 ©Institute for Promoting Research & Policy Development DOI Additionally, they claimed that "if these models had demonstrated strong predictive capability and could have provided accurate information on the current state of the economy, then this information could have been used to help shape the more medium-term outlook." This implies that, though no model can come up with 100% accurate results, especially during periods of crises, it is important to find a model that can yield more accurate results than other models. There is, therefore, a need for appropriate forecasting to determine models that can make appropriate forecasts. Only two decades ago have researchers become concerned about the consequences of the stationarity assumption in performing inference regarding predictive ability. In 2010, Giacomini and Rossi developed methods to perform inference on forecast comparisons when the forecasting ability may be affected by instabilities. In 2011, Rossi and Sekhposyan developed fluctuation optimality tests that are robust to the presence of instabilities. Unlike Giacomini and Rossi (2010), who compare models‟ relative forecasting performance, Rossi and Sekhposyan (2011) focus on individual models‟ absolute predictive ability and tests for forecast optimality. There is the scantiness of literature on the models that consider instabilities. Though studies by Giacomini and Rossi (2010), Rossi (2005), Muller and Petalas (2009), Elliott and Muller (2005), Andrews and Ploberger (1994), and Andrews (1993) amongst others, help to fill the gap in this topic, there is still a need to conduct a study that considers instabilities, particularly for South Africa where such topic seems missing. Exchange rate forecasting is an inherent approach in financial risk management, yet, previous studies that forecasted the South African exchange rate employed traditional techniques to compare forecasting models. Among those are the studies by Mokoma (2014). This paper will therefore serve to fill the gap in the literature on this topic by comparing both traditional and new techniques that are robust to instabilities to forecast the South African exchange rates. In each category, absolute as well as relative comparisons will be done. Accurately forecasting the movement of exchange rates is of interest in a variety of fields, such as international business, financial management, and monetary policy. Accurate exchange rate forecasting benefits stock market investors and traders by lowering risk and maximizing transaction rewards (Dahlquist & Hasseltoft, 2020). 1. Introduction From the monetary authorities‟ viewpoint, reliable exchange rate forecasting helps monetary authorities regulate exchange rates and carry out monetary policies (Shen et al., 2021). Exchange rates are permitted to fluctuate within an unidentified band under a managed exchange rate regime, and the government may intervene depending on their forecasts for the exchange rates in the future (Uz Akdogan, 2020). Additionally, when a government implements monetary measures to boost the economy, such as decreasing interest rates, this will raise revenue and demand for the country's imported goods, causing the currency to appreciate and ultimately harming the competitiveness of exported goods. As a result, an accurate exchange rate projection can assist a government in determining the appropriate level of interest rate reductions, which is related to assessing the effectiveness of monetary policies (Tillmann, 2016). Volatility tends to change over time. During periods of financial crisis, large volatility persistence occurs (Ning et al., 2015). There is therefore a need to use a data set that spans the period where the world has witnessed the greatest financial crisis of the time and also techniques that consider such instabilities. The present study, therefore, covers the period during and post-global financial crisis, using traditional models which have failed to consider instabilities that occurred during periods of crisis and compare them with the recent ones, that is, those which are robust to instabilities. Against this backdrop, this paper aims to evaluate the forecasting ability of models in the presence of instabilities by comparing traditional and new techniques in forecasting exchange rates using data that is non- stationary and symmetric and asymmetric GARCH models under different distribution errors, namely, normal distribution, student‟s t-distribution, and general error distributions. The models employed include simple GARCH, Exponential GARCH, Integrated GARCH, Threshold GARCH, and APARCH models. Furthermore, it evaluates the forecasting performance of a model either in isolation ("absolute" forecasting ability) or relative to other models ("relative" forecasting ability), the significance of economic instabilities in exchange rate forecasting and forecasting evaluations and verifies the presence of these stylized facts in the South African foreign exchange market. 2. Materials and Methods This section presents the research methodology. The section is structured as follows: Subsection 2.1 presents a brief discussion on symmetric and asymmetric GARCH models. Theoretical models underpinning the models used in the study are then presented In subsection 2.2, the GARCH models that are employed in this study are specified. The data employed in the study are described in subsection 2.3. Subsection 2.4 presents a discussion on the out-of- sample evaluation of the volatility forecasts. 1. Introduction When applying predictive ability tests to macroeconomic time series data researchers face important practical problems: economic time series data are prone to volatility (Rossi and Soupre, 2017). As stated by Rossi (2021), “instabilities are widespread in economic time series”. Odendahl et al,( 2022) also stressed that “despite various predictive comparison tests that have been proposed in the literature to help predictors in model selection, no single model is usually the best overall. Predictive performance is typically prone to volatility and is sample dependent”. Therefore, when researchers test a model's predictive ability, it can be important to allow its predictive ability to change over time. Indeed, traditional predictive assessment tests are unreliable in the presence of volatility and can lead to erroneous conclusions. This problem arises because traditional tests assume stationarity and the presence of instability destroys stationarity. Most time series techniques assume that the data are stationary. Stationary processes have a constant mean, variance, and autocorrelation structure (Songhao, 2021). It is often the case, however, that real-life data don't comply with the assumption of stationarity. Stock return data, for instance, tends to fluctuate over time, whereas economic time series are typically seasonal (Politis et al.,1999). For most time series models, the model assumptions are violated when nonstationary data is used. This leads to the estimators losing properties such as asymptotic normality and sometimes even consistency. Hence, applying a model that requires a stationary series to a nonstationary series could lead to poor estimates of the parameters and therefore poor forecasts. As stated by van Greunen, et al., (2014), the stationarity of a time series can have a significant influence on its properties and forecasting behavior. g A substantial body of research has been done on testing predicted optimality. It is, nonetheless, implicitly predicated on the notion of covariance stationarity. Because of this presumption, the models were unable to forecast events correctly when there was instability. Given the ubiquity of models that do not take instability into account, even after the financial crisis, forecasting models must take instability into account. Kenny and Morgan (2011) reported the poor predictive performance of all models that were employed in their study to capture the period of exceptional macroeconomic weakness in the fourth quarter of 2008 and the first quarter of 2009. 72 \| ©Institute for Promoting Research & Policy Development DOI: 10.56734/ijbms.v4n9a6 Vol. 04 - Issue:9/September_2023 l. y y Symmetric GARCH models are GARCH models that do not capture the asymmetry in financial returns data. Asymmetry implies that unexpected bad news (decrease in stock price or negative et) increases conditional 2.1.1 Symmetric models  ARCH (Autoregressive conditional heteroscedasticity) model ARCH model, developed by Engle in 1982, accounts for the difference between the unconditional and the conditional variance of a stochastic process. While conventional econometric models operate under the assumption of constant variance, the ARCH process allows the conditional variance to vary over time, leaving the unconditional variance constant (Ding, 2011). Although financial markets may experience excessive volatility from time to time, it appears that volatility will eventually settle down to a long-run level. In the case of exchange rates, this implies that exchange rate indices are mean reverting (they are stationary processes), then this implies that shocks to exchange rate indices will have a transitory effect, in that the exchange rate will return to its trend path over time (Goudarzi, 2013). According to Brooks (2008), under the ARCH model, the „autocorrelation in volatility‟ is modeled by allowing the conditional variance of the error term, , to depend on the immediately previous value of the squared error and ARCH (1) model takes the following form: = + ………………………………………………(1) = + ………………………………………………(1) Where represents the conditional variance of the error term, represents a lagged value of the squared error.  Generalized autoregressive conditional heteroscedasticity (GARCH) Bollerslev (1986) extended the ARCH model to the GARCH, which had the same key properties as the ARCH but required far fewer parameters to adequately model the volatility process. Th GARCH (1 1) d l i th f i b p q y y p The GARCH (1, 1) model is therefore given by: q p q y y p The GARCH (1, 1) model is therefore given by: = + ……………………………....(2) Where all the parameters must be positive, while the sum of quantifies the persistence of shocks to volatility. The GARCH (1, 1) model generates one-step-ahead forecasts of volatility as a weighted average of the constant long-run or average variance, the previous forecast variance, and previous volatility reflecting squared „news‟ about the return, . As volatility forecasts are increased following a large return of either sign, the GARCH specification captures the well-known volatility clustering effect. Though the GARCH model has greater applicability for easy computation, it has drawbacks in the application of asset pricing. Firstly, in the GARCH model, the impacts to conditional variance of the positive and negative sides are symmetrical. 2.1 Symmetric and asymmetric models Symmetric GARCH models are GARCH models that do not capture the asymmetry in financial returns data. Asymmetry implies that unexpected bad news (decrease in stock price or negative et) increases conditional 73 \| www.ijbms.net ISSN 2694-1430 (Print), 2694-1449 (Online) International Journal of Business & Management Studies volatility more than the unexpected good news of similar magnitude. The symmetric and asymmetric (theoretical) models which form the basis of the models used in the study are therefore discussed in detail below. volatility more than the unexpected good news of similar magnitude. The symmetric and asymmetric (theoretical) models which form the basis of the models used in the study are therefore discussed in detail below. 2.1.1 Symmetric models This means that the model cannot explain the negative correlation between the fluctuations in stock returns as it assumes that the conditional variance is a function of lagged squared residuals. 74 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa 2.1.2 Asymmetric GARCH Models To capture the asymmetry in return volatility (leverage effect), a new class of models was developed, termed the asymmetric GARCH models. The asymmetric GARCH models, unlike symmetric GARCH models, capture the asymmetry in financial returns data. These models are discussed in detail below. 2.1.4 Integrated general conditional heteroscedasticity (IGARCH) model Engle and Bollerslev (1986) also put forward the integrated GARCH (IGARCH), which is an extension to the GARCH model. To capture the characteristic of volatility persistence, the GARCH model features an exponential decay in the autocorrelation of conditional variances. It has been noted that squared and absolute returns of financial assets typically exhibit serial correlations that are slow to decay, like those of an integrated I (d) process. Shocks in the volatility series seem to have a long memory and lasting impact on future volatility over a long horizon. The IGARCH captures this effect but a shock in this model impacts upon future volatility over an infinite horizon and the conditional variances does not exist for this model (Granger and Poon, 2002). If the AR polynomial of the GARCH representation has a unit root, then we have an IGARCH model. Thus, IGARCH models are unit- root GARCH models. The integrated GARCH (IGARCH) is specified as: The integrated GARCH (IGARCH) is specified as: = ∑ 2 ∑ …………………….. (7) (7) The sum of coefficients is restricted to 1. The exogenous variable can be easily reflected in the various specifications of GARCH models just by the addition of and . Exponential General Autoregressive conditional heteroscedasticity (EGARCH)  Exponential General Autoregressive conditional heteroscedasticity (EGARCH) The EGARCH model proposed by Nelson (1991), specifically captures asymmetries in volatility. In this model, asymmetries are considered exponentially. This approach captures the skewness and allows the ARCH process to be asymmetrical. In the EGARCH model, the conditional variance is an asymmetric function of the lagged disturbances : The model is stated as follows: The model is stated as follows: = ; = 2 ∑ …………… (3) (3) The coefficient captures the asymmetric impact of news with negative shocks having a greater impact than positive shocks of equal magnitude if <0, while the volatility clustering effect is captured by a significant . Finally, the use of the logarithm form allows the parameters to be negative without the conditional variance becoming negative. 74 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa DOI: 10.56734/ijbms.v4n9a6 ©Institute for Promoting Research & Policy Development DOI: 10.56734/ijbms.v4n9 ©Institute for Promoting Research & Policy Development Vol. 04 - Issue:9/September_2023  Threshold general autoregressive conditional heteroscedasticity (TGARCH) model A h l ili d l l d h dl l ff i h h h ld GA  Threshold general autoregressive conditional heteroscedasticity (TGARCH) model Another volatility model commonly used to handle leverage effects is the threshold GARCH (TG introduced by Glosten, Jagannathan, and Runkle (1993) and Zakoian (1994). The TGARCH model of order can be written as: = ∑ ∑ ∑ )……… (4) In this model, good news, > 0, and bad news, < 0 have differential effects on the conditional variance. Here, has an impact of good news while + have an impact of bad news. If γ i > 0, bad news has a greater impact of conditional variance, whereas if ≠ 0, news impact is asymmetric. In this model, good news, > 0, and bad news, < 0 have differential effects on the conditional variance. Here, has an impact of good news while + have an impact of bad news. If γ i > 0, bad news has a greater impact of conditional variance, whereas if ≠ 0, news impact is asymmetric. 2.2 Specification of the models This section presents the empirical models that are employed in this study. 2.1.3 Asymmetric Power autoregressive conditional heteroscedasticity (APARCH) model Ding et al., (1993) concluded that there was no reason for the volatility to be a linear function of the squared residuals and introduce the APARCH (p, q) model, which allows the power δ of the heteroscedasticity equation to be estimated from the data (Pedro and Marcos et al, 2011).Though the APARCH model captures asymmetry in return volatility, in this model, volatility tends to increase more when returns are positive, as compared to negative returns of the same magnitude. The APARCH model also yields the long-memory property of returns. The power parameter on the standard deviation is estimated and not imposed: = + (\| \| ) + ……………………….. (5) (5) Parameter δ in the equation denotes exponent of conditional standard deviation, while parameter γ describes the asymmetry effect of good and bad news on conditional volatility. A positive value of γ means that negative shocks from the previous period have a higher impact on the current level of volatility, and otherwise (Miletic, 2015). An APARCH (p, q) model assumes that: An APARCH (p, q) model assumes that: = + (\| \| ) + …………………… (6) (6) A positive (resp. negative) value of the ‟s means that past negative (resp. positive) shocks have a deeper impact on current conditional volatility than past positive shocks (see Black, 1976, Laurent, 2004). 75 \| www.ijbms.net ARCH(1) ( ) A time-series {ϵ(t)} is given at each instance by: ϵ(t) = w(t)σ(t)………………………………… (8) (8) Where, w(t) is the white noise with zero mean and unit variance. white noise with zero mean and unit variance. Where, w(t) is the white noise with zero mean and unit variance. Var(x(t)) = σ²(t) = + σ²(t-1) Where , are parameters of the model and > 0, ≥ 0 to ensure that the conditional variance is positive. σ²(t-1) is lagged square error. gg q y that ϵ(t) is an autoregressive conditional heteroskedastic model of order unity, denoted by We say that ϵ(t) is an autoregressive conditional heteroskedastic model of order unity, denoted by ARCH(1). ϵ(t) = w(t)* σ(t) = w(t)* ⎷( + ϵ²(t-1)) ϵ(t) = w(t) σ(t) = w(t)* ⎷( + ϵ²(t-1)) Similarly, ARCH(p) is given by: ϵ(t) = w(t) ⎷( + ⍺(p) * ∑ ϵ²(t-i)……………. (9) (9) where: p is the number of lag squared residual errors to include in the ARCH model. i = (1,2,3,-,-,-, -, p) tells us the number of logged periods of the square error. Wh It follows that should be a positive parameter. However, in estimation, it might come out negative. Many programs allow the researcher to restrict to be positive and greater than zero. The mean equation estimates the conditional mean of the variable. It is important to get the mean equation correctly specified before estimating the ARCH/GARCH model. The mean equation typically can be modelled as an AR process, AR in combination with other explanatory variables, just as a function of other explanatory variables. It is important to test for no autocorrelation in the residuals before estimating the GARCH process. The variance equation estimates the variance processan as a type of autoregressive process. Both equations form a system that is estimated together using maximum likelihood. The mean equation is important because if this is not correctly specified the variance estimate will not be good either. The mean equation describes the expected value of the stochastic process * + The mean equation therefore can be an AR, an ARMAX or a structural econometric model, among others (Brooks, 2002). In the case of symmetric models, there are three coefficients in the Conditional variance equation. The constant term, ARCH term and GARCH term. The arch (alpha) term explains volatility clustering, GARCH (beta) term explains persistence (Bera and Higgins, 1993, Bonga, 2019). 76 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa 2.2.1 Empirical models The empirical models are stated as follows:  ARCH (1) The basic ARCH model consists of two equations, a conditional mean equation and a conditional variance equation. Both equations should be estimated simultaneously given that variance is a mean equation. The mean ISSN 2694-1430 (Print), 2694-1449 (Online) nternational Journal of Business & Management Studies equation estimates the conditional mean of the examined variable. The variance equation estimate this process as a typical autoregressive process. Both equations form a system that is estimated together with the maximum likelihood method (Dritsaki, 2019). equation estimates the conditional mean of the examined variable. The variance equation estimate this process as a typical autoregressive process. Both equations form a system that is estimated together with the maximum likelihood method (Dritsaki, 2019). ( ) ARCH (p) model is simply an AR(p) model applied to the variance of a time series. = 2 ∑ … (11) Where, is the log of the conditional variance. This implies that the leverage effect is exponential, rather than quadratic, and that forecasts of the conditional variance are guaranteed to be nonnegative. The presence of financial leverage result can be tested by the theory that . The impact is asymmetric if The coefficient captures the asymmetric impact of news with negative shocks having a greater impact than positive shocks of equal magnitude if <0, while the volatility clustering effect is captured by a significant . The forecast variance of Rand/US$ daily exchange rate is in logarithm form, implying that the model allows the parameters to be negative without the conditional variance becoming negative. A priori expectation is that the gamma ( ) parameter should be negative (implying that there is asymmetry, that is, negative news has more impact than positive news).  Threshold GARCH (TGARCH) model  Threshold GARCH (TGARCH) model Another volatility model commonly used to handle leverage effects is the threshold GARCH (TGARCH) model introduced by Glosten, Jagannathan, and Runkle (1993) and Zakoian (1994). Another volatility model commonly used to handle leverage effects is the threshold GARCH (TGARCH) model introduced by Glosten, Jagannathan, and Runkle (1993) and Zakoian (1994). The TGARCH model of order can be written as: The TGARCH model of order can be written as: model of order can be written as: = = ∑ ∑ ( )…… (12) if and if . (12)  Asymmetric ARCH (APARCH) model Following Ding, Granger, and Engle‟s (1993) model, the APARCH model can be stated as: ( ) Granger, and Engle‟s (1993) model, the APARCH model can be stated as: = + (\| \| ) + ………… (13) (13) Where is the forecast variance. Parameter δ in the equation denotes exponent of conditional standard deviation, while parameter γ describes the asymmetry effect of good and bad news on conditional volatility. A positive value of γ means that negative shocks from the previous period have a higher impact on the current level of volatility, and otherwise y ter γ is expected to be positive if there is asymmetric or leverage effect. Parameter γ is expected to be positive if there is asymmetric or leverage effect. A priori expectation is that and 0. Otherwise, the PARCH model will simply be a standard GARCH specification. ARCH(1) However, in asymmetric GARCH models, there are four coefficients in the conditional variance equation, namely, the constant term, ARCH term, GARCH term and the volatility asymmetry term. The fourth term or coefficient, gamma (leverage) term explains the leverage effect. The APARCH model adds the fifth term called ( ) delta (Kisinbay, 2003, Ding, 2011, Smolović, Lipovina- Božović and Vujošević, 2017). ARCH effect/spillover effect ( ) represents the impact of a magnitude of a shock (size). The existence of volatility spill-overs implies that one large shock increases the volatilities not only in its own asset or market but also in other assets or markets as well (Kharchenko and Tzvetkov, 2013). Therefore, negative ARCH means the spill-over is negative related (increase volatility in one country make the other country's volatility reduced) GARCH effect ( ) implies persistence of past volatility (past volatility explain current volatility). If GARCH term is positive, it shows that there is a positive relation between the past variance and the current variance in absolute value. A higher and positive GARCH coefficient indicates that volatility takes a longer period to decay, which is implies that there is long memory (Aluko et. al., 2017). y p g y Leverage/ volatility asymmetry term (represents the impact of a sign of a shock). If it is different from zero, this implies that there is an asymmetric effect (bad news and good news of the same size have different impacts). If the asymmetry term is negative, this implies that negative shocks have a greater impact on volatility rather than the positive shocks of the same magnitude. If the leverage coefficient (Gamma or ) is negative and significant, indicating the presence of an asymmetric behavior. The significance of negative shocks persistence or the volatility asymmetry indicates that investors are more prone to negative news in comparison to positive news. ©Institute for Promoting Research & Policy Development DOI: 10.56734/ijbms.v4n9a6 DOI: 10.56734/ijbms.v4n9a6 Vol. 04 - Issue:9/September_2023 This implies that the volatility spillover mechanism is asymmetric (Nelson, 1991). GARCH (1, 1) model  GARCH (1, 1) model The study follows Bollerslev (1986) model. A simple GARCH model can be stated as follows: ( , ) The study follows Bollerslev (1986) model. A simple GARCH model can be stated as follows: R/US$t = + + ……………………… (10) (10) Where R/US$t represents the forecast variance of Rand/US$ daily exchange rate. is the constant long- run or average variance, represents the previous forecast variance, and represents previous volatility reflecting squared „news‟ about the return all the parameters must be positive, while the sum of quantifies the persistence of shocks to volatility. Squared implies that volatility forecasts are increased following a large return of either sign (negative and positive news have same impact). A priori expectation is that all the parameters should be positive (implying that there is no asymmetry). A priori expectation is that all the parameters should be positive (implying  Exponential GARCH (EGARCH) model EGARCH (1, 1) model can be written as: p ( ) EGARCH (1, 1) model can be written as: = 2 ∑ … (11 = 2 ∑ … 2.5.1 Evaluation forecasting criteria To test the model‟s forecasting ability, two types of tests are used, namely, tests of relative forecast comparison and tests of absolute forecasting performance. Relative forecast comparison tests determine the best model between any two competing forecasting models in terms of predicting ability. Conversely, absolute forecasting performance tests are used to evaluate whether forecasts from one specific model fulfil some minimal requirements, like being unbiased or producing forecast errors that are unpredictable using any information available at the time the forecast is made. Whilst both tests are used to forecast the models‟ forecast ability, the former ones are used to compare two forecasting models, whilst the latter are used to evaluate one specific forecasting model (Rossi and Soupre, 2017). There are traditional methods that have for long been applied for both relative and absolute forecast evaluation. For relative assessment, examples are Diebold & Mariano (1995), West (1996) and Clark & McCracken (2001). Examples of absolute evaluation are Mincer & Zarnowitz (1969) and West & McCracken (1998). These traditional forecast evaluation tests assume stationarity, which is often violated when applying tests of forecasting ability to macroeconomic time series data as it is well known that economic time series data are prone to instabilities. g p g p There are traditional methods that have for long been applied for both relative and absolute forecast evaluation. For relative assessment, examples are Diebold & Mariano (1995), West (1996) and Clark & McCracken (2001). Examples of absolute evaluation are Mincer & Zarnowitz (1969) and West & McCracken (1998). These traditional forecast evaluation tests assume stationarity, which is often violated when applying tests of forecasting ability to macroeconomic time series data as it is well known that economic time series data are prone to instabilities. p Nominal exchange rates used in this study are macroeconomic time series; and there are reasons to assume instabilities. Notable examples of instabilities are the 2007-2009 global financial crisis. During the 2007-2009 global financial crisis, drastic changes on several macroeconomic relationships occurred. Ng and Wright (2013) found that credit spreads replaced interest rates in predicting output growth during that time. Likewise, Rossi (2013b) revealed the presence of severe instabilities in exchange rate forecasting models. 2.5. Out-of- sample forecasting In forecasting, it is not necessarily the model that provides the best in-sample fit that produces the best out-of- sample volatility forecast (Shamiri and Isa, 2009; Wennström, 2014). Hence it is common to use the out-of-sample forecast to aid the selection of model which is best suited for the series under study (Andersen and Bollerslev, 1998; Hansen and Lunde, 2001; Brandt and Jones, 2006; Cerqueira et al, 2020).Before the discussion on forecasting is presented, the different models that are used in comparing traditional and fluctuation tests are presented.  Integrated GARCH (IGARCH) model  Integrated GARCH (IGARCH) model The study borrows from Engle and Bollerslev‟s (1986) model and rewrites the IGARCH model as follows: The study borrows from Engle and Bollerslev‟s (1986) model and rewrites the IGARCH mod y borrows from Engle and Bollerslev‟s (1986) model and rewrites the IGARCH model as follows: = ∑ 2 ∑ ……………… (14) (14) iori assumption is that the ARCH term ( ) and the GARCH term ( ) sum up to one. A priori assumption is that the ARCH term ( ) and the GARCH term ( ) sum up to one 77 \| www.ijbms.net ISSN 2694-1430 (Print), 2694-1449 (Online) International Journal of Business & Management Studies 2.3. Description of the type of sample to be used A 5-day week Rand per US Dollar rates, sourced from the Federal Reserve Economic Data (FRED) online, is the sample data used for analysis with 3038 observations, covering the period 2007/01/01 to 2018/12/31. The period 2007 was chosen as the starting period to capture the 2007-2009 global financial era. Stata software package is employed as it brings the desired results wanted by the researcher. 2.5.1 Evaluation forecasting criteria When testing the model‟s forecasting ability, it is potentially important to allow their forecasting ability to change over time.As argued by Giacomini and Rossi (2010), in unstable environments, it is plausible that the relative forecasting performance of models may itself change over time. Regardless of the increasing empirical evidence suggesting instability in the forecasting performance of econometric models relative to the naïve benchmarks (for example, Stock and Watson, 2003a), existing literature for forecasting comparison did not account for this possibility. p y For example, Moran and Solomon (2017) apply traditional approaches like the Diebold and Mariano (1995) test to compare the predictive accuracy (for loss criteria MSE, MAE and MAPE), which are argued to have failed in the presence of instabilities, such as the 2007-2009 global financial crisis, despite Francq and Zakoian‟s (2010) caution that different GARCH models can lead to almost equivalent predictive formulas. The implication is that the forecasting success of a model relative to its competitor seems to be linked to some specific periods in time, and there are numerous situations in which there has been a reversal in the relative forecasting ability of two competing forecasting models (Giacomini and Rossi, 2010). In environments characterized by instabilities, it is important to compare different models to find out which models perform best in such instabilities. Traditional tests of forecast evaluation are not reliable in the presence of instabilities, which may lead to incorrect inference. To address this challenge, both relative forecast comparisons and absolute forecasting performance tests which are robust to instabilities were introduced (Rossi and Soupre (2017). To compare the relative out-of-sample forecasting performance of two competing models in the presence of possible instabilities, Giacomini and Rossi‟s (2010) fluctuation test will be used. Conversely, Rossi and Sekhposyan‟s (2016) fluctuation rationality test will be employed for testing absolute forecasting performance robust to instabilities. The reason for comparing these tests is to investigate and evaluate the forecasting claims that fluctuation tests are more powerful than traditional ones as stated by Giacomini and Rossi (2010), Rossi and Sekhposyan (2016) and Rossi and Soupre (2017) among others 3. Results The main objective of this paper is to compare the new and traditional tests in testing forecast unbiasedness/rationality and to test competing models‟ forecasting performance to find the model with the best out- 78 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa ©Institute for Promoting Research & Policy Development DOI: 10.56734/ijbms.v4n9a6 Vol. 04 - Issue:9/September_2023 of-sample fit. Against this backdrop, this section presents the results for both relative and absolute comparisons of fluctuation and traditional tests respectively. of-sample fit. Against this backdrop, this section presents the results for both relative and absolute comparisons of fluctuation and traditional tests respectively. 3.1 Testing for out of sample fit g f f p f The out-of-sample forecast tests are conducted using Stata commands that illustrate how to test forecast unbiasedness/rationality and how to test competing models‟ forecasting performance, in a way that is robust to the presence of instabilities. To test competing models‟ forecasting performance, Giacomini and Rossi (2010) test is employed. Conversely, Rossi and Sekhposyan (2016) test is used to test forecast unbiasedness/rationality. The results for the new and old approach (that is, the Giacomini and Rossi‟s (2010) fluctuation test and the Diebold and Mariano (1995) test) are compared. Tables 1 and 2, therefore present these results respectively. he new and old approach (that is, the Giacomini and Rossi‟s (2010) fluctuation test and the Diebold 95) test) are compared. Tables 1 and 2, therefore present these results respectively. 79 \| www.ijbms.net 3.2.1 Tests of Relative Forecasting Performance Robust to Instabilities The pairwise comparison of the model that accounts for instabilities in Table 1 show that t-distribution error assumption dominates all other error distribution assumptions, implying that all the models perform well under t- distribution assumption. Therefore, if forecasters are interested in using GARCH model to forecast series (at least exchange rate) considering inherent instabilities, the best assumption distribution to make is that of t-distribution. This is so, because it has fatter tails than the normal distribution, it can also be used as a model for financial returns exhibiting excessive kurtosis, enabling a more realistic calculation of the Value at Risk (VaR) in such cases. It can skew the accuracy concerning the normal distribution (Annapoorna, 2021). However, in the case of traditional test, the t- and general error distribution assumptions dominate, with t-distribution taking the lead. Normal -distribution RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant 0.001 0.002 0.440 0.659 (0.003) 0.005 L.arch 0.250 0.021 12.070 - 0.209 0.291 * L.garch 0.922 0.101 9.110 - 0.724 1.120 * Constant (0.002) 0.001 (1.650) 0.099 (0.004) - * Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square . Prob > chi2 . Akaike crit. (AIC) -3976.156 t-distribution RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant - 0.002 0.040 0.969 (0.004) 0.004 L.arch 0.234 0.029 8.150 - 0.177 0.290 * L.garch 0.821 0.119 6.920 - 0.588 1.053 * Constant (0.001) 0.001 (0.730) 0.464 (0.003) 0.002 Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square . Prob > chi2 . Akaike crit. (AIC) -4146.356 Ged distribution RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant (0.001) 0.002 (0.870) 0.382 (0.005) 0.002 L.arch 0.236 0.042 5.660 - 0.154 0.318 * L.garch 0.989 0.200 4.950 - 0.598 1.380 * Constant (0.003) 0.002 (1.170) 0.244 (0.007) 0.002 Constant 0.116 0.034 3.410 0.001 0.049 0.183 * Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square . Prob > chi2 . Akaike crit. (AIC) -4220.482 * p<.01, ** p<.05, * p<.1 Table 1: GARCH Results Table 1: GARCH Results Table 1: GARCH Results According to Table 2 , the pairwise comparison of the model that accounts for instabilities shows that GARCH under normal error distribution assumption dominates all other models, implying that the model has the best forecast during periods of instabilities. This is consistent with theory which postulates that, symmetric models perform better than asymmetric models under normal distribution. They are only outperformed by asymmetric models under distributions assumptions like t-distribution and general error assumption, due to their failure to ISSN 2694-1430 (Print), 2694-1449 (Online) International Journal of Business & Management Studies capture leverage effect (Hentschel, 1995, Islam, 2014). The results from the traditional reveals that there is no winner as three models (GARCHnor, IGARCHnor and TGARCHnor) beat APARCH model. capture leverage effect (Hentschel, 1995, Islam, 2014). The results from the traditional reveals that there is no winner as three models (GARCHnor, IGARCHnor and TGARCHnor) beat APARCH model. 80 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa Normal -distribution RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant 0.001 0.002 0.5 0.614 -0.003 0.005 L.arch 0.268 0.019 13.99 0 0.23 0.305 * L.garch 0.732 0.019 38.26 0 0.695 0.77 * Constant 0 0 1.73 0.084 0 0.001 * Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square . Prob > chi2 . Akaike crit. (AIC) -3975.881 t-distribution RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant 0 0.002 0.05 0.96 -0.004 0.004 L.arch 0.239 0.026 9.24 0 0.188 0.29 * L.garch 0.761 0.026 29.41 0 0.71 0.812 * Constant 0 0 -1.6 0.11 -0.001 0 Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square . Prob > chi2 . Akaike crit. (AIC) -4148.139 Ged distribution RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant -0.001 0.002 -0.82 0.41 -0.005 0.002 L.arch 0.258 0.039 6.64 0 0.182 0.334 * L.garch 0.742 0.039 19.08 0 0.666 0.818 * Constant 0 0 0.64 0.524 0 0.001 Constant 0.122 0.034 3.64 0 0.056 0.188 * Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square . Prob > chi2 . Akaike crit. (AIC) -4220.411 * p<.01, ** p<.05, * p<.1 Table 2: IGARCH Results APARCH normal RUS_D1 Coef. St.Err. Table 1: GARCH Results t-value p-value [95% Conf Interval] Sig Constant 0.002 0.002 0.77 0.44 -0.003 0.006 L.ar 0.025 0.027 0.93 0.354 -0.028 0.078 L.aparch 0.247 0.043 5.78 0 0.163 0.33 * L.aparch_e 0.035 0.031 1.14 0.253 -0.025 0.095 L.pgarch 0.732 0.107 6.88 0 0.524 0.941 * Constant 0 0 -0.2 0.842 0 0 power 3.829 0.712 5.38 0 2.433 5.225 * Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square 0.858 Prob > chi2 0.354 Akaike crit. (AIC) -3989.29 PARCH t-distribution RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant 0 0.002 0.19 0.848 -0.004 0.004 L.ar 0.022 0.028 0.78 0.436 -0.033 0.077 L.aparch 0.233 0.057 4.08 0 0.121 0.345 * L.aparch_e 0.026 0.039 0.65 0.517 -0.052 0.103 L.pgarch 0.647 0.109 5.92 0 0.433 0.861 * Constant 0 0 0.23 0.819 0 0 power 3.625 0.867 4.18 0 1.926 5.324 * Table 2: IGARCH Results 80 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa Vol. 04 - Issue:9/September_2023 ©Institute for Promoting Research & Policy Development DOI: 10.56734/ijbms.v4n9a6 Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square 0.606 Prob > chi2 0.436 Akaike crit. (AIC) -4157.56 APARCH GED RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant -0 0.002 -0.78 0.434 -0.005 0.002 L.ar 0.006 0.024 0.25 0.799 -0.041 0.053 L.aparch 0.25 0.063 3.97 0 0.127 0.374 * L.aparch_e 0.017 0.056 0.3 0.764 -0.092 0.126 L.pgarch 0.869 0.191 4.56 0 0.495 1.243 * Constant 0 0 -0.44 0.658 -0.001 0.001 power 3.016 0.898 3.36 0.001 1.256 4.776 * Constant 0.132 0.034 3.93 0 0.066 0.198 * Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square 0.065 Prob > chi2 0.799 Akaike crit. (AIC) -4222.3 * p<.01, p<.05, * p<.1 Table 3: APARCH normal EGARCH normal rus_d1 coef. st.err. t-value p-value [95% conf interval] Sig constant 0.002 0.002 1.16 0.247 -0.002 0.007 l.earch 0.053 0.018 2.99 0.003 0.018 0.089 * l.earch_a 0.448 0.027 16.31 0 0.394 0.502 * l.egarch 1.117 0.081 13.78 0 0.958 1.276 * constant 0.505 0.35 1.44 0.149 -0.181 1.192 mean dependent var 0.002 sd dependent var 0.116 number of obs 2504 chi-square . prob > chi2 . akaike crit. (aic) -3968.852 EGARCH t-distribution rus_d1 coef. st.err. Table 1: GARCH Results t-value p-value [95% conf interval] Sig constant 0.001 0.002 0.43 0.666 -0.003 0.005 l.earch 0.046 0.023 1.99 0.046 0.001 0.092 l.earch_a 0.442 0.038 11.65 0 0.368 0.516 * l.egarch 1.073 0.099 10.84 0 0.879 1.267 * constant 0.275 0.433 0.63 0.526 -0.574 1.124 mean dependent var 0.002 sd dependent var 0.116 number of obs 2504 chi-square . prob > chi2 . akaike crit. (aic) -4141.64 EGARCH GED rus_d1 coef. st.err. t-value p-value [95% conf interval] Sig constant -0.001 0.002 -0.7 0.486 -0.005 0.002 l.earch 0.039 0.031 1.26 0.209 -0.022 0.101 l.earch_a 0.434 0.054 8.1 0 0.329 0.539 * l.egarch 1.141 0.146 7.81 0 0.855 1.427 * constant 0.605 0.632 0.96 0.339 -0.634 1.844 constant 0.117 0.034 3.44 0.001 0.05 0.183 * mean dependent var 0.002 sd dependent var 0.116 number of obs 2504 chi-square . prob > chi2 . akaike crit. (aic) -4215.565 * p<.01, ** p<.05, * p<.1 Table 4: EGARCH Results ©Institute for Promoting Research & Policy Development DOI: 10.56734/ijbms.v4n9a6 Vol. 04 - Issue:9/September_2023 SD dependent var 0.116 Chi-square 0.606 Akaike crit. (AIC) -4157.56 Table 3: APARCH normal 81 \| www.ijbms.net ISSN 2694-1430 (Print), 2694-1449 (Online) International Journal of Business & Management Studies TGARCH normal RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant 0.001 0.002 0.49 0.626 -0.003 0.005 L.arch 0.242 0.025 9.75 0 0.193 0.29 * L.tarch 0.014 0.033 0.42 0.676 -0.05 0.078 L.garch 0.923 0.107 8.61 0 0.713 1.133 * Constant -0 0.001 -1.6 0.11 -0.004 0 Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square . Prob > chi2 . Akaike crit. (AIC) -3974.24 TGARCH t-distribution RUS_D1 Coef. St.Err. t-value p-value [95% Conf Interval] Sig Constant 0 0.002 -0.03 0.977 -0.004 0.004 L.arch 0.246 0.037 6.57 0 0.173 0.32 * L.tarch -0.02 0.043 -0.46 0.647 -0.105 0.065 L.garch 0.818 0.121 6.74 0 0.58 1.056 * Constant -0 0.001 -0.69 0.493 -0.003 0.002 Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square . Prob > chi2 . Akaike crit. (AIC) -4144.5 TGARCH GED RUS_D1 Coef. St.Err. 82 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa Table 1: GARCH Results t-value p-value [95% Conf Interval] Sig Constant -0 0.002 -0.87 0.384 -0.005 0.002 L.arch 0.241 0.052 4.59 0 0.138 0.344 * L.tarch -0.01 0.059 -0.14 0.89 -0.124 0.108 L.garch 0.988 0.203 4.86 0 0.589 1.386 * Constant -0 0.002 -1.14 0.254 -0.007 0.002 Constant 0.116 0.034 3.41 0.001 0.049 0.183 * Mean dependent var 0.002 SD dependent var 0.116 Number of obs 2504 Chi-square . Prob > chi2 . Akaike crit. (AIC) -4218.5 * p<.01, ** p<.05, * p<.1 Table 5: TGARCH Results Table 5: TGARCH Results In Table 3, GARCHt, IGARCHt and TGARCHt dominate EGARCHt and APARCHt respectively, in case where new techniques are applied. This is consistent with Alberga et al. (2008)‟s study that the EGARCH skewed Student-t model outperformed GARGH, GJR and APARCH models. Surprisingly, even the results from traditional tests show that, GARCHt, IGARCHt and TGARCHt dominate APARCHt. The results are consistent with those of Abdullah et al., (2017), who model the volatility dynamics of the taka–US dollar exchange rate return using GARCH, APARCH, EGARCH, TGARCH, and IGARCH models. Their findings suggest that GARCH and IGARCH models under the student-t error distribution outperform the other models. However, this is inconsistent with the findings by Schmidt (2021), which show that the symmetric GARCH (1,1) on average has the worst volatility forecasting performance when forecasting a crisis on Nordic indices, using GARCH, EGARCH, GJR and APARCH models. The superior forecasting models were found to be the GJR (1,1) and EGARCH (1,1). The results from Table 4, reveal that the pairwise comparison of the model that accounts for instabilities show that there is no model which is the overall winner under general error distribution. On the side of traditional tests, the IGARCH is the best performer in terms of forecast accuracy. The results from Table 4, reveal that the pairwise comparison of the model that accounts for instabilities show that there is no model which is the overall winner under general error distribution. On the side of traditional tests, the IGARCH is the best performer in terms of forecast accuracy. 82 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa DOI: 10.56734/ijbms.v4n9a6 ©Institute for Promoting Research & Policy Development DOI: 10.56734/ijbms.v ©Institute for Promoting Research & Policy Development Vol. 04 - Issue:9/September_2023 ©Institute for Promoting Research & Policy Development DOI: 10.56734/ijbms.v4n9a6 Vol. 04 - Issue:9/September_2023 ©Institute for Promoting Research & Policy Development Vol. 04 - Issue:9/September_2023 FIGURE 1: GARCHt vs GARCHnor comparison FIGURE 2: TGARCHt vs TGARCHged comparison FIGURE 2: TGARCHt vs TGARCHged comparison Figure 1 and 2 plot the sequence of overtime for pairwise TGARCHt vs TGARCHnor and TGARCHt vs TGARCHged (as shown by a continuous line) and shows that it is clearly outside the critical value lines, depicted by the dashed lines. The strongest evidence against the null hypothesis (implying the strongest empirical evidence in favour of the first model) appears to be around 2010 and 2011. 83 \| www.ijbms.net Table 5: TGARCH Results (Note the pairs, TGARCHt vs TGARCHnor and TGARCHt vs TGARCHged were chosen because they show the dominance of t-distribution and the strongest empirical evidence in favour of the first model as compared to other pairs under t-distribution). GARCH dominates all models; t dominates all 3 distribution assumptions. 3.2.2 Tests of Absolute Forecasting Performance Robust to Instabilities Traditional tests of forecast rationality (such as Mincer and Zarnowitz, 1969, and West and McCracken, 1998) assume stationarity and are thus invalid in the presence of instabilities (Rossi and Soupre, 2017). However, unlike traditional tests, the fluctuation rationality test is based on the idea of instability and therefore has a lower rejection of the null hypothesis of forecast rationality (de Prince et al., 2021). The results for Rossi - Sekhposyan test statistics reveals that the null assumption of forecast rationality is rejected by all models. So individually these models perform well, implying that they can make a good forecast. Hence they are used popularly. It is when one model has to be chosen from more than 1 pool where it matters. That is why the results of the relative comparison show some models as worse performers. 83 \| www.ijbms.net ISSN 2694-1430 (Print), 2694-1449 (Online) International Journal of Business & Management Studies Figure 3: Rossekk IGARCHt Figure 4: Rossekk APARCHt Figure 3 and 4 plot the sequence of overtime (depicted by a continuous line) of IGARCH model under t- distribution and APARCH under t-distribution, and shows that it is clearly outside the critical value line (depicted by the dashed line). The strongest evidence against the forecast rationality appears to be around 2009 for both models. This clearly supports the idea that, in the presence of instabilities, the fluctuation tests are more powerful than traditional tests and provide a visual illustration of when predictive ability appears or breaks down in the data (Rossi and Soupre 2017) Figure 3: Rossekk IGARCHt Figure 3: Rossekk IGARCHt Figure 4: Rossekk APARCHt Figure 3: Rossekk IGARCHt Figure 4: Rossekk APARCHt Figure 4: Rossekk APARCHt Figure 3 and 4 plot the sequence of overtime (depicted by a continuous line) of IGARCH model under t- distribution and APARCH under t-distribution, and shows that it is clearly outside the critical value line (depicted by the dashed line). The strongest evidence against the forecast rationality appears to be around 2009 for both models. This clearly supports the idea that, in the presence of instabilities, the fluctuation tests are more powerful than traditional tests and provide a visual illustration of when predictive ability appears or breaks down in the data (Rossi and Soupre, 2017). 84 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa References Alam, Md. Zahangir & Siddikee, Md & Masukujjaman, Mohammad. (2013). Forecasting Volatility of Stock Indices with ARCH Model. International Journal of Financial Research. 4. 10.5430/ijfr.v4n2p126. Alberg, D., Shalit, H. & Yosef, R. (2008). Estimating Stock Market Volatility Using Asymmetric GARCH Models. Applied Financial Economics. 1 (18), p.1201. Aluko, O.A, Olufemi, A.P and Oseko, M.S. (2017).Modelling Volatility Persistence and Asymmetry with Structural Break: Evidence from the Nigerian Stock Market. Journal of Economics and Behavioral Studies (ISSN: 2220-6140) Vol. 8, No. 6, pp. 153-160, December 2016. January 2017. DOI: 10 22610/jebs v8i6(J) 1489 j Andersen, T., and T. Bollerslev (1998) Answering the skeptics: Yes, standard volatility models do provide accurate forecasts, International Economic Review, 39, 4, 885-905. Andrews, D.W.K., 1993, Tests for Parameter Instability and Structural Change with Unknown Change Point. Econometrica 61(4), 821-856. Andrews, D.W.K., and W. Ploberger (1994). Optimal Tests When a Nuisance Parameter is Present only under the Alternative. Econometrica 62(6), 1383-1414. Bera, A. and M. Higgins (1993). ARCH models: Properties, estimation and testing. Journal of Economic Surveys 7, 305-362. Black, F., 1976, Studies of stock market volatility changes, Proceeding of the American Statistical Association, Business and Economic Statistics Section. Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics, 1986, vol. 31, issue 3, 307-327 Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of Econom vol. 31, issue 3, 307-327 Bonga, W. G. (2019), Stock Market Volatility Analysis using GARCH Family Models: Evidence from Zimbabwe Stock Exchange (May 30, 2019). MPRA Paper No. 94201, University Library of Munich, Germany., Available at SSRN: https://ssrn.com/abstract=3402342 p Brandt, M. and Jones, C. (2006). Volatility Forecasting with Range-Based EGARCH Models. Journal of Business and Economic Statistics, 24, 470-486. http://dx.doi.org/10.1198/073500106000000206. Brooks, C. (2002). Introductory Econometrics for Finance. Cambridge: Cambridge University Press Brooks, C. (2002). Introductory Econometrics for Finance. Cambri ooks, C. (2008) Introductory Econometrics for Finance. 2nd Edition, Cambridge University Press, Ne York.https://doi.org/10.1017/CBO9780511841644 Cerqueira, V., Torgo, L. & Mozetič, I. (2020). Evaluating time series forecasting models: an empirical study on performance estimation methods. Mach Learn 109, 1997–2028 (2020). https://doi.org/10.1007/s10994- 020-05910-7 Clark, T.E. and McCracken, M.W. (2001). Tests of Equal Forecast Accuracy and Encompassing for Models. Journal of Econometrics 105(1), 85. Dahlquist, M. & Hasseltoft, H. Economic momentum and currency returns. Journal of Financial Economics. 2020, 136, 152–167. de Prince, D., Marcal, E. and Valls Pereira, P.L. (2021). Are Professional Forecasters rational? What is the role of instability and what variables affect it? 4. Discussion The out- of sample forecast tests are conducted using Giacomini and Rossi‟s (2010) and Rossi and Sekhposyan‟s (2016) tests. The pairwise comparison of the model that accounts for instabilities show that t-distribution error assumption dominates all other error distribution assumptions, implying that all the models perform well under t- distribution assumption. Even, in case of traditional test, the t-distribution takes the lead. The results for Rossi - Sekhposyan test statistics reveals that, IGARCH model under t- distribution and APARCH under t-distribution show the strongest evidence against the forecast rationality, which appears to be around 2009 for all the models. This implies that the Rossi - Sekhposyan test can make accurate forecast using the IGARCH model and APARCH under t-distribution. We can therefore conclude that in the presence of instabilities, the fluctuation tests are more powerful than traditional tests and provide a visual illustration of when predictive ability ap in the presence of instabilities, it is not appropriate to test models‟ forecasts by using methods that are not robust to instabilities. In fact, as revealed in this study, traditional tests may be invalid in thepresence of forecast instabilities, and more powerful tests should be used. Policy recommendations that can be made from conclusion drawn is that, fluctuation tests should be used during the periods of instabilities as they can provide visual illustration of when predictive ability appears, especially for an emerging economy like South Africa,where exchange rates are subject to high levels of fluctuations. DOI: 10.56734/ijbms.v4n9a6 ©Institute for Promoting Research & Policy Development Vol. 04 - Issue:9/September_2023 Vol. 04 - Issue:9/September_2023 ©Institute for Promoting Research & Policy Development References Conference Paper. Brazilian Econometrics Meeting. December, 2021. Diebold, F. X., and R. S. Mariano. (1995). Comparing predictive accuracy. Journal of Business and Statistics 13: 253–263. D. (2011). Modelling of Market Volatility with APARCH Model. Department of Mathematics Upps University U.U.D.M. Project Report 2011:6. Maj 2011 Ding, Z., Granger, C.W.J. and Engle R. F. (1993), “A long memory property of stock market returns and a new model”, Journal of Empirical Finance, 1(1):83–106. , p ( ) Dritsaki, C. (2019), Modelling the Volatility of Exchange Rate Currency using GARCH Model, Economia Internazionale / International Economics, 72, (2), 209-230. Elliott,G and Müller, U.K (2006). Efficient Tests for General Persistent Time Variation in Regression Coefficients, The Review of Economic Studies, Volume 73, Issue 4, October 2006, Pages 907– 940, https://doi.org/10.1111/j.1467-937X.2006.00402.x p g j Engle R.F. (1982) Autoregressive conditional heteroscedasticity with estimates of the variance of UK inflation, Econometrica, 50, 987–1007. Engle, R. and Bollerslev, T. (1986). Modeling the Persistence of Conditional Variances, Econometric Rev., 5, pp. 5-50. Engle, R. and Bollerslev, T., 1986, Modelling the persistence of conditional variances, Econometric Reviews, Taylor and Francis Journals, vol. 5(1), pages 1-50. 85 \| www.ijbms.net ISSN 2694-1430 (Print), 2694-1449 (Online) International Journal of Business & Management Studies Francq, C. and Zakoian, J.M. (2010) GARCH Models: Structure, Statistical Inference and Financial Applications. John Wiley & Sons Ltd., Chichester. https://doi.org/10.1002/9780470670057 Glosten, L.R., Jagannathan, R and Runkle, D.E. (1993). On the Relation between the Expected Value and the Volatility of the Nominal Excess Return on Stocks, The Journal of Finance: Vol 48, pp. 1779-1801. Goudarzi, H. (2013). Volatility Mean Reversion And Stock Market Efficiency. Asian Economic and Financial Review, 2013, 3(12):1681-1692 1681 Hojatallah Granger, C., & Poon, S.H. and (2002). Forecasting volatility in financial markets: a review, Journal of economic literature, vol 41(2), 478-539 Hentschel, L. (1995) All in the Family Nesting Symmetric and Asymmetric GARCH Models. Journal of Financial Economics, 39, 71-104.https://doi.org/10.1016/0304-405X(94)00821-H , , p g ( ) Journal of Applied Econometrics 25(4), 595-620. Journal of Applied Econometrics 25(4), 595-620. Kenny, G. and J. Morgan (2011), „Some Lessons from the Financial Crisis for the Economic Analysis‟, unpublished manuscript, European Central Bank. p p p Kharchenko, I. and Tzvetkov, P. (2013), Estimation of Volatilities and Spill Over Effects between Developed and Emerging Market Economies, Master‟s Thesis, Department of Economics, Lund University, Sweden. Kişinbay, T. (2003). Predictive Ability of Asymmetric Volatility Models at Medium Term Horizon. IMF Working Paper Laurent, S. 86 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa References Analytical Derivates of the APARCH Model. Computational Economics 24, 51–57 (2004). https://doi.org/10.1023/B:CSEM.0000038851.72226.76 Ludger Hentschel, H. (1994). All in the Family: Nesting Symmetric and Asymmetric GARCH Models William E. Simon Graduate School of Business Administration University of Rochester Rochester, NY 14627 . Lunde A & Hansen P R 2001 "A Forecast Comparison of Volatility Models: Does Anything Beat a Ludger Hentschel, H. (1994). All in the Family: Nesting Symmetric and Asymmetric GARCH Models William E. Simon Graduate School of Business Administration University of Rochester Rochester, NY 14627 . Lunde, A. & Hansen, P.R , 2001. "A Forecast Comparison of Volatility Models: Does Anything Beat a GARCH(1 1)? " Working Papers 2001-04 Brown University Department of Economics Lunde, A. & Hansen, P.R , 2001. A Forecast Comparison of Volatility Models: Does Anything Be GARCH(1,1)?," Working Papers 2001-04, Brown University, Department of Economics. , , p y y g GARCH(1,1)?," Working Papers 2001-04, Brown University, Department of Economics. ( , ) , g p , y, p Miletić, S. (2015). Modeling and forecasting exchange rate volatility: comparison between EEC and Developed countries. Industrija, 43(1). Mincer, J.A and Zarnowitz, V. (1969). "The Evaluation of Economic Forecasts," NBER Chapters, in: Economic Forecasts and Expectations: Analysis of Forecasting Behavior and Performance, pages 3-46, National Bureau of Economic Research, Inc. eau of Economic Research, Inc. J. (2014). Exchange rate volatility in South Africa: A comparative analysis of the ARCH and GARCH els. Master‟s thesis.2014. Mokoma, T.J. (2014). Exchange rate volatility in South Africa: A comparative analysis of the ARCH models. Master‟s thesis.2014. Moran, J. and Solomon, P. (2017) Volatility in High-Frequency Intensive Care Mortality Time Series: Application of Univariate and Multivariate GARCH Models. Open Journal of Applied Sciences, 7, 385-411. doi: 10.4236/ojapps.2017.78030. j pp Muller, U. and Petalas, P. (2009), Efficient Estimation of the Parameter Path in Unstable Time Series Models. Review of Economic Studies (2010) 77, 1508–1539. © 2010 The Review of Economic Studies Limited. 0034-6527/10/00411011$02.00 doi: 10.1111/j.1467-937X.2010. 00603.x j Nelson, D.B. (1991). Conditional heteroscedasticity in asset returns: a new approach. Econometrica, 59, 347– 370.http://dx.doi.org/10.2307/2938260 Nelson, D.B. (1991). Conditional heteroscedasticity in asset returns: a new approach. Econometrica, 59, 347– 370.http://dx.doi.org/10.2307/2938260. Ng, S and Wright, J.H. (2013). "Facts and Challenges from the Great Recession for Foreca Macroeconomic Modelling." Journal of Economic Literature, 51(4):1120-54. Ng, S and Wright, J.H. (2013). "Facts and Challenges from the Great Recession for Forecasting and Macroeconomic Modelling." Journal of Economic Literature, 51(4):1120-54. References g g g g Macroeconomic Modelling." Journal of Economic Literature, 51(4):1120-54. Ning, C, Xu, D and Wirjanto,T.S. (2015). Is volatility clustering of asset returns asymmetric? Journal of Banking & Finance,Volume 52, 2015, Pages 62-76,ISSN 0378-4266, https://doi.org/10.1016/j.jbankfin.2014.11.016. Odendahl, F, Rossi, B & Sekhposyan, T..(2022). Evaluating forecast performance with state dependence, Journal of Econometrics, 2022,ISSN 0304-4076. Pedro, P.V and Marcos, W.J. (2011). Modeling and Forecasting Realized Volatility: Evidence Brazilian Review of Econometrics. 31. 315- 337. 10.12660/bre. v31n22011.4056. Politis, D.N., Romano, J.P., Wolf, M. (1999). Subsampling for Nonstationary Time Series. In: Subsa Springer Series in Statistics. Springer, New York, NY. https://doi.org/10.1007/978-1-4612- Politis, D.N., Romano, J.P., Wolf, M. (1999). Subsampling for Nonstationary Time Series. In: Subsampling. Springer Series in Statistics. Springer, New York, NY. https://doi.org/10.1007/978-1-4612-1554-7_4 R i B d S kh T (2011) O i li T i h P f I bili i W ki P 11 18 D k Rossi, B and Sekhposyan, T. (2011). Optimality Tests in the Presence of Instabilities. Working Pape University, Department of Economics. Rossi, B and Soupre, M. (2017). "Implementing tests for forecast evaluation in the presence of instabilities," Stata Journal, StataCorp LP, vol. 17(4), pages 850-865, December. p ( ) p g Rossi, B. (2005). Optimal Tests for Nested Model Selection with Underlying Parameter Instabilities, Econometric Theory 21(5), 962-990. Rossi, B. (2013). "Exchange Rate Predictability." Journal of Economic Literature, 51 (4): 1063-1119. Rossi, B. (2013). "Exchange Rate Predictability." Journal of Economic Literature, 51 (4): 1063-1119 86 \| Forecasting Exchange Rates Comparison in The Presence of Instabilities: Azwifaneli I. Nemushungwa DOI: 10.56734/ijbms.v4n9a6 Vol. 04 - Issue:9/September_2023 ©Institute for Promoting Research & Policy Development Rossi, B. (2021). Forecasting in the Presence of Instabilities: How We Know Whether Models Predict Well and How to Improve Them. Journal of Economic Literature 2021, 59(4), 1135–1190 https://doi.org/10.1257/jel.20201479 p g j Rossi, B. and Sekhposyan. T (2016). Forecast rationality tests in the presence of instabilities, with applications to federal reserve and survey forecasts. Journal of Applied Econometrics 31: 507–532. y pp Schmidt, L (2021). Volatility Forecasting Performance of GARCH Models: A Study on Nordic Indices During COVID-19. Master Thesis. Spring, 2021. Umeå University, Department of Economics, Sweden. COVID-19. Master Thesis. Spring, 2021. Umeå University, Department of Economics, Sweden. Shamiri, A. and Isa, Z. (2009) Modelling and Forecasting Volatility of the Malaysian Stock Markets. Journal of Mathematics and Statistics, 5, 234-240. Shamiri, A. and Isa, Z. References (2009) Modelling and Forecasting Volatility of the Malaysian Stock Markets. Journal of Mathematics and Statistics, 5, 234-240. Shen M.L, Lee C.F, Liu H.H, Chang P.Y & Yang C.H. (2021). An Effective Hybrid Approach for Forecasting Currency Exchange Rates. Sustainability. 2021; 13(5):2761. https://doi.org/10.3390/su13052761 Smolović, J.C, Lipovina-Božović, M and Vujošević, S.(2017). GARCH models in value at risk estimation: Shen M.L, Lee C.F, Liu H.H, Chang P.Y & Yang C.H. (2021). An Effective Hybrid Approach for Forecasting Currency Exchange Rates. Sustainability. 2021; 13(5):2761. https://doi.org/10.3390/su13052761 Smolović, J.C, Lipovina-Božović, M and Vujošević, S.(2017). GARCH models in value at risk estimation: empirical evidence from the Montenegrin stock exchange. Ekonomska Istraživanja / Economic Research. April 2017.30(1):1-22. p ( ) Songhao, W (2021). Stationarity Assumption in Time Series Data.Towards A Data Science. Juyl 4, 20 p Songhao, W (2021). Stationarity Assumption in Time Series Data.Towards A Data Science. Juyl 4, 2021. Tillmann, P. Unconventional monetary policy and the spillovers to emerging markets. Journal of International Songhao, W (2021). Stationarity Assumption in Time Series Data.Towards A Data Science. Juyl 4, 2021. Tillmann, P. Unconventional monetary policy and the spillovers to emerging markets. Journal of International Money Finance. 2016, 66, 136–156. Tillmann, P. Unconventional monetary policy and the spillovers to emerging markets. Journal of In Money Finance. 2016, 66, 136–156. Uz Akdogan, I. Understanding the dynamics of foreign reserve management: The central bank intervention policy and the exchange rate fundamentals. Int. Econ. 2020, 161, 41–55. Van Greunen, J. et al. 2014. The prominence of stationarity in time series forecasting. Studies in Economics and Econometrics, 38(1):1-16. Wennström, A. (2014). Volatility Forecasting Performance: Evaluation of GARCH type volatility models on Nordic equity indices. Master of Science Thesis, Spring 2014 Department of Mathematics, Royal Institute of Technology (KTH), Stockholm, Sweden. gy West, K. D., 1996, Asymptotic Inference about Predictive Ability. Econometrica 64(5), 1067-1084. y p y ( ) West, K.D. and McCracken, M.W. (1998). Regression-Based Tests of Predictive Ability. International Economic Review 39(4), 817-840. ( ), M. (1994) Threshold Heteroskedastic Models. Journal of Economic Dynamics and Control, 18, 931-95 87 \| www.ijbms.net
https://openalex.org/W1996538987	https://europepmc.org/articles/pmc4033200?pdf=render	English	null	Pharmacological or genetic orexin1 receptor inhibition attenuates MK-801 induced glutamate release in mouse cortex	Frontiers in neuroscience	2,014	cc-by	4,677	INTRODUCTION release from hypothalamic slices (van den Pol et al., 1998). In rats, intravenous administration of orexin A increased glutamate release in the amygdala, a brain region known to express OX receptors, but not in the cerebellum where OX receptors are not expressed (John et al., 2003). Intravenous injection of orexin A also increased glutamate release in the locus coeruleus (Kodama and Kimura, 2002) and it has been suggested that orexin neurons and NMDA receptors interact together in the control of the locus coeruleus noradrenergic activity (Tose et al., 2009). These actions might be mediated via the OX1 receptors since this OX receptor subtype is exclusively expressed in the locus coeruleus (Trivedi et al., 1998). In the present study, we tested the hypothesis that orexin modulates glutamate signaling in cortex via OX1 recep- tors by monitoring levels of glutamate in the prefrontal cortex of freely moving mice. In contrast to the locus coeruleus, in cortex both OX receptor subtypes are expressed (Marcus et al., 2001) and therefore the OX1 receptor might not be the only receptor involved. We used an enzyme coated biosensors to monitor real time change in glutamate release (Uslaner et al., 2012). MK-801, an NMDA receptor antagonist, was administered to indirectly disinhibit pyramidal neurons and increase glutamate release in Orexins, also known as hypocretins, are two peptides (orexin A and B) derived from a single precursor produced exclusively in the hypothalamus (de Lecea et al., 1998). Orexins partici- pate in neuronal regulation by activating their receptors (OX1 and OX2 receptors). The orexin system is emerging as a pow- erful integrator of multiple physiological functions including sleep/wakefulness states, energy balance, stress, reward, and emo- tion (for latest review see Li et al., 2014). Orexin neurons project broadly throughout the whole brain and can be modulated by multiple humoral signals and neuronal inputs. Several key areas from the limbic system (including the bed of the stria termi- nalis, the amygdala, and the medial septum) send axons to orexin cells (Sakurai et al., 2005). The majority of synapses on orexin somata and dendrites are asymmetric with small, round, clear vesicles that reﬂect excitatory neural transmission (Zhang et al., 2005). Orexins and glutamate interact at the synaptic level where orexins facilitate glutamatergic actions. In turn, glutamatergic neurons regulate orexinergic neuronal activity via presynaptic facilitation of glutamate release (Li et al., 2002). Edited by: Edited by: Eero Vasar, University of Tartu, Estonia Eero Vasar, University of Tartu, Estonia Reviewed by: Andrew Harkin, Trinity College Dublin, Ireland Jessica R. Barson, The Rockefeller University, USA Reviewed by: Andrew Harkin, Trinity College Dublin, Ireland Jessica R. Barson, The Rockefeller University, USA Reviewed by: Andrew Harkin, Trinity College Dublin, Ireland Jessica R. Barson, The Rockefeller University, USA Correspondence: Pascal Bonaventure, Janssen Research and Development, LLC, 3210 Merryﬁeld Row, San Diego, CA 92121, USA e-mail: pbonave1@its.jnj.com ORIGINAL RESEARCH ARTICLE published: 20 May 2014 published: 20 May 2014 doi: 10.3389/fnins.2014.00107 Pharmacological or genetic orexin1 receptor inhibition attenuates MK-801 induced glutamate release in mouse cortex Leah Aluisio, Ian Fraser, Tamara Berdyyeva, Volha Tryputsen , Brock T. Shireman , James Shoblock, Timothy Lovenberg, Christine Dugovic and Pascal Bonaventure Janssen Pharmaceutical Research and Development, LLC, San Diego, USA Janssen Pharmaceutical Research and Development, LLC, San Diego, USA The orexin/hypocretin neuropeptides are produced by a cluster of neurons within the lateral posterior hypothalamus and participate in neuronal regulation by activating their receptors (OX1 and OX2 receptors). The orexin system projects widely through the brain and functions as an interface between multiple regulatory systems including wakefulness, energy balance, stress, reward, and emotion. Recent studies have demonstrated that orexins and glutamate interact at the synaptic level and that orexins facilitate glutamate actions. We tested the hypothesis that orexins modulate glutamate signaling via OX1 receptors by monitoring levels of glutamate in frontal cortex of freely moving mice using enzyme coated biosensors under inhibited OX1 receptor conditions. MK-801, an NMDA receptor antagonist, was administered subcutaneously (0.178 mg/kg) to indirectly disinhibit pyramidal neurons and therefore increase cortical glutamate release. In wild-type mice, pretreatment with the OX1 receptor antagonist GSK-1059865 (10 mg/kg S.C.) which had no effect by itself, signiﬁcantly attenuated the cortical glutamate release elicited by MK-801. OX1 receptor knockout mice had a blunted glutamate release response to MK-801 and exhibited about half of the glutamate release observed in wild-type mice in agreement with the data obtained with transient blockade of OX1 receptors. These results indicate that pharmacological (transient) or genetic (permanent) inhibition of the OX1 receptor similarly interfere with glutamatergic function in the cortex. Selectively targeting the OX1 receptor with an antagonist may normalize hyperglutamatergic states and thus may represent a novel therapeutic strategy for the treatment of various psychiatric disorders associated with hyperactive states. METHODS All animal experimental procedures were performed in accor- dance with the Guide for the Care and Use of Laboratory Animals adopted by the US National Institutes of Health. All animal experimental procedures were performed in accor- dance with the Guide for the Care and Use of Laboratory Animals adopted by the US National Institutes of Health. The study was done in the set of three experiments. For each animal the area under the curve of the baseline change in glu- tamate release over 1 h post MK-801 stimulation was calculated. To assess whether the mean area under the curve differs for the two treatment groups (GSK-1059865 and vehicle treated mice) we used One-Way ANOVA, blocking for the experiment effect. Two- sided paired t-test was used to assess the effect of GSK-1059865 (vs. vehicle) on the area under the curve of glutamate release over 1 h time period. Male wild-type mice and fully backcrossed OX1 receptor knockout mice of the same age (8–12 weeks old) and strain (Jackson Labs C57/Bl6) were singly housed at the time of exper- imentation. Biosensor technology was used to measure real- time change of glutamate release in the cortex of freely moving mice as previously described (Uslaner et al., 2012). Each mouse was given a subcutaneous injection of Buprenex (0.1 mg/kg, buprenorphine hydrochloride; Reckitt Benckiser Pharmaceuticals Inc., Richmond, VA) 5 min prior to anesthesia. Animals were anesthetized with an isoﬂurane/air mixture and stereotaxically implanted with a guide cannula (BAS) in the prefrontal cor- tex (+1.54 mm anterior, 0.5 mm lateral, right hemisphere and 1.4 mm ventral to Bregma, Supplemental Data Figure S1). A grounding screw attached to the “headmount connector” was inserted slightly anterior to the interaural line. The guide can- nula and head mount connector were secured in place with dental cement. Animals were allowed at least 5 days to recover from surgery prior to experimentation. A second experiment using OX1 receptor knockout and wild- type mice was conducted as a within subject study design. All animals received a subcutaneous saline injection, and were treated with MK-801 (0.178 mg/kg S.C.) 90 min later. The area under the curve was calculated for each animal and the two-sample one- sided t-test was performed to compare the area under the curve of changes in glutamate from baseline over 1 h between MK-801 stimulated OX1 receptor knockout and wild-type mice. INTRODUCTION Early in vitro experimental data have shown that orexin A increased glutamate May 2014 \| Volume 8 \| Article 107 \| 1 www.frontiersin.org www.frontiersin.org Orexin1 receptor modulates glutamate release Aluisio et al. study subject design. At the end of each test session, naïve ani- mals to MK-801 received MK-801 (0.178 mg/kg S.C.) as a positive control test. cortex. In the ﬁrst experiment we investigated the effect of tran- sient inhibition of the OX1 receptor by systemic administration of the selective and brain penetrant OX1 receptor antagonist GSK-1059865 (Gozzi et al., 2011) on MK-801 induced glutamate release in the cortex of wild-type mice. In a second experiment, we investigated the effect of permanent inhibition of OX1 recep- tors by comparing the effect of MK-801 on glutamate release in the cortex of wild-type vs. OX1 receptor knockout mice. GSK-1059865 was synthetized at Janssen Research and Development LLC and was formulated in 30% sulfobutylether (7)-β-cyclodextrin (SBE) or 5% Pharmasolve, 20% Cremaphor and 75% D5W. MK-801 was purchased from Sigma and formu- lated in saline. All injection volumes were 10 ml/kg. The doses of GSK-1059865 (10 mg/kg) and MK-801 (0.178 mg/kg) were selected based on literature data where biological activity has been reported (Bonaventure et al., 2011; Gozzi et al., 2011; Uslaner et al., 2012). METHODS Changes in glutamate concentration were recorded as cur- rent (nA) every second for at least 60 min following the last drug administration. The data were averaged into 2-min bins and glutamate concentrations were calculated using post calibration values. The glutamate sensor (model #7004 Pinnacle Technology Inc., Lawrence, KS) speciﬁcation and hardware setup has been described in detail previously (Naylor et al., 2011). Brieﬂy, glu- tamate biosensors act through enzyme mediated processing. Glutamate is converted to hydrogen peroxide and detected by oxidation at the Pt-Ir electrode. A selectively passive membrane allows for exclusion of electroactive interferents. Prior to sen- sor insertion, each biosensor was calibrated in vitro to verify glutamate sensitivity and interference rejection. The biosen- sors extended beyond the guide cannulas by 1 mm. The after- noon prior to experimentation, under light isoﬂurane anesthesia, biosensors were inserted into the guide cannula of each animal which was then returned to its home cage for sensor equilibra- tion. Data acquisition started immediately and the animals were maintained connected to the acquisition hardware overnight. The sensor signal was processed by the 8401 Data Acquisition System and data acquisition, storage, and analysis were performed by the Pinnacle Acquisition Laboratory software suite. Experimentation was conducted the following day between 8:00 am and 3:00 pm (light cycle) in a controlled environment. At the completion of the experiment, each sensor was removed from the guide cannula and calibrated in vitro for glutamate sensitivity and interference rejection of ascorbate at 37◦C in a circulating water bath. At the end of the experimentation, the brains were coronally sectioned and sensor location was visually veriﬁed. Sensor placements which were outside the targeted area by ±0.2 mm were not included. Frontiers in Neuroscience \| Neuropharmacology RESULTS We ﬁrst investigated the effect of transient inhibition of the OX1 receptor with a selective brain penetrant OX1 receptor antagonist on MK-801 induced glutamate release in the cortex of wild-type animal (Figure 1A). In wild-type mice, GSK-1059865 (10 mg/kg, S.C.) signiﬁcantly attenuated cortical glutamate release elicited by MK-801 (0.178 mg/kg S.C.) (Figure 1A, treatment group: F = 5.537, df = 1, p = 0.037; block: F = 3.677, df = 2, p = 0.057). As compared to vehicle GSK-1059865 did not affect glutamate release per se (Figure 1B, t = 1.784, df = 7, p = 0.118). A group of wild-type mice was administered with saline (S.C.) and monitored for a 90- min baseline period. Each animal was then pretreated with the selective OX1 receptor antagonist (GSK- 1059865, 10 mg/kg S.C.) or vehicle 30 min prior to the injection of MK-801 (0.178 mg/kg, S.C.). The effect of GSK-1059865 on its own (in the absence of MK-801) was also assessed in an across We then investigated the effect of permanent inhibition of the OX1 receptor by comparing the effect of systemic administra- tion of MK-801 (0.178 mg/kg S.C.) on glutamate release in the cortex of wild-type and OX1 receptor knockout mice (Figure 2). Compared to wild-type mice, OX1 receptor knockout mice had May 2014 \| Volume 8 \| Article 107 \| 2 Frontiers in Neuroscience \| Neuropharmacology \| 2 Orexin1 receptor modulates glutamate release Aluisio et al. FIGURE 1 \| (A) Effect of GSK-1059865 (10 mg/kg S.C.) on MK-801 (0.178 mg/kg, S.C.) induced glutamate release in the cortex of freely moving wild-type mice (within subject study design). GSK-1059865 was administered at t = 0 and MK-801 was administered at t = 30 min. (B) Effect of GSK-1059865 (10 mg/kg S.C.) on glutamate release in the cortex of freely moving wild-type mice (across subject study design). GSK-1059865 was administered at t = 0. Real-time measurements of glutamate were conducted before and after compound administration. Results are expressed as change in glutamate concentrations as mean ± s.e.m. GSK-1059865 did not affect glutamate levels per se (p = 0.118) but attenuated MK-801-induced increase in glutamate release (p = 0.037). DISCUSSION In this study we used a pharmacological and genetic approach to investigate the functional interaction between the OX1 and NMDA receptors using MK-801 induced glutamate release in mouse frontal cortex. In wild-type mice, transient inhibition of the OX1 receptor with a selective OX1 receptor antagonist sig- niﬁcantly attenuated excessive cortical glutamate release elicited by MK-801. Systemic administration of the selective OX1 recep- tor antagonist had no effect on glutamate release on its own. Permanent inhibition of OX1 receptor in OX1 receptor knock- out mice resulted in a diminished glutamate release evoked by MK-801 comparable to the response obtained with the OX1 receptor antagonist. Therefore, these data ruled out the possibil- ity of a direct pharmacokinetic interaction between MK-801 and GSK-1059865 in wild-type mice. FIGURE 1 \| (A) Effect of GSK-1059865 (10 mg/kg S.C.) on MK-801 (0.178 mg/kg, S.C.) induced glutamate release in the cortex of freely moving wild-type mice (within subject study design). GSK-1059865 was administered at t = 0 and MK-801 was administered at t = 30 min. (B) Effect of GSK-1059865 (10 mg/kg S.C.) on glutamate release in the cortex of freely moving wild-type mice (across subject study design). GSK-1059865 was administered at t = 0. Real-time measurements of glutamate were conducted before and after compound administration. Results are expressed as change in glutamate concentrations as mean ± s.e.m. GSK-1059865 did not affect glutamate levels per se (p = 0.118) but attenuated MK-801-induced increase in glutamate release (p = 0.037). FIGURE 1 \| (A) Effect of GSK-1059865 (10 mg/kg S.C.) on MK-801 (0.178 mg/kg, S.C.) induced glutamate release in the cortex of freely moving wild-type mice (within subject study design). GSK-1059865 was administered at t = 0 and MK-801 was administered at t = 30 min. (B) Effect of GSK-1059865 (10 mg/kg S.C.) on glutamate release in the cortex of freely moving wild-type mice (across subject study design). GSK-1059865 was administered at t = 0. Real-time measurements of glutamate were conducted before and after compound administration. Results are expressed as change in glutamate concentrations as mean ± s.e.m. GSK-1059865 did not affect glutamate levels per se (p = 0.118) but attenuated MK-801-induced increase in glutamate release (p = 0.037). Brain penetration and biological activity of GSK-1059865 has been recently demonstrated (Gozzi et al., 2011) and later con- ﬁrmed in our lab in a similar dose range (Dugovic et al., 2014). RESULTS a blunted glutamate release response to MK-801 (t = 1.981, df = 9 p = 0 04) MK 801 still induced an increase in glutamate FIGURE 2 \| Effect of MK-801 (0.178 mg/kg, S.C.) on glutamate release in the cortex of freely moving OX1 receptor knockout mice and wild-type mice (within subject study design). Real-time measurements of glutamate were conducted before and after compound administration (MK-801 was administered at t = 0). Results are expressed as change in glutamate concentrations as mean ± s.e.m. MK-801 response in OX1 receptor knockout mice was attenuated compared to wild-type mice (p = 0.042). FIGURE 2 \| Effect of MK-801 (0.178 mg/kg, S.C.) on glutamate release in the cortex of freely moving OX1 receptor knockout mice and wild-type mice (within subject study design). Real-time measurements of glutamate were conducted before and after compound administration (MK-801 was administered at t = 0). Results are expressed as change in glutamate concentrations as mean ± s.e.m. MK-801 response in OX1 receptor knockout mice was attenuated compared to wild-type mice (p = 0.042). FIGURE 2 \| Effect of MK-801 (0.178 mg/kg, S.C.) on glutamate release f f O DISCUSSION The biosensor technology used to monitor glutamate release in this study is well established. Noteworthy, the effects of vehi- cle on glutamate release are substantially greater in the ﬁrst experiment (Figure 1A) when compared to the second exper- iment (Figure 1B). Inter-animal variability is common during animal handling and injections as glutamate release has been linked to an animal’s arousal state and individual stress response a blunted glutamate release response to MK-801 (t = 1.981, df = 9, p = 0.04). MK-801 still induced an increase in glutamate release in OX1 receptor knockout mice but the magnitude of this increase was in a range of about 30–50% of the release observed in wild-type animals after 20 min post treatment. May 2014 \| Volume 8 \| Article 107 \| 3 www.frontiersin.org www.frontiersin.org Orexin1 receptor modulates glutamate release Aluisio et al. (Westerink and Cremers, 2007). Previous studies have shown that MK-801 increased glutamate release in the rat cortex (Lopez-Gil et al., 2007; Bonaventure et al., 2011) and nucleus accumbens (Uslaner et al., 2012). The increase in cortical glutamatergic trans- mission elicited through blockade of an excitatory glutamate receptor (NMDA) results from an indirect effect of the NMDA receptor antagonist on GABAergic interneurons (Krystal et al., 2003). NMDA receptor antagonists attenuate the tonic activation of inhibitory neurons (GABA) most likely in the hippocampus and/or thalamus resulting in a disinhibition of glutamatergic input in the medial prefrontal cortex. Noteworthy, MK-801 also acts as a nicotinic acetylcholine receptor antagonist in addition to its NMDA receptor antagonistic properties (Amador and Dani, 1991). Ketamine, another NMDA receptor antagonist used in the clinic, has been reported to increase glutamate release indepen- dently from its effect on locomotor activity in mice (Schobel et al., 2013). Both MK-801 and ketamine have been used as animal models for psychosis at doses that produce substantial increase of glutamate release (Large, 2007). Interestingly, ketamine has been reported to have rapid antidepressant properties at low dosage in patients resistant to antidepressive treatment (Salvadore and Singh, 2013). The mechanism for this antidepressant action is not fully elucidated, in particular in reference to its effect on glutamate release. the anxiogenic properties of orexins are mediated through inter- action with the glutamatergic system and that engagement of the OX1 receptor is needed (Lungwitz et al., 2012). REFERENCES Amador, M., and Dani, J. A. (1991). MK-801 inhibition of nicotinic acetylcholine receptor channels. Synapse 7, 207–215. doi: 10.1002/syn.890070305 Evidences for glutamate and orexins co-localization and co- release from orexin terminals in the locus coeruleus have been presented (Henny et al., 2010). Orexins neurons originating from the hypothalamus and projecting to the cortex contain both orexins and glutamate. Interactions between orexinergic and glu- tamatergic neurons have been described in the locus coeruleus, amygdala and in the bed nucleus of the stria terminalis (John et al., 2003; Tose et al., 2009; Lungwitz et al., 2012). In locus coeruleus, the OX1 receptor is the only OX receptor present (Marcus et al., 2001). In amygdala, bed nucleus of the stria ter- minalis, and cortex both OX receptors are present (Trivedi et al., 1998; Marcus et al., 2001). Future studies should examine the potential contribution of the OX2 receptor in addition to the role of the OX1 receptor. Bonaventure, P., Aluisio, L., Shoblock, J., Boggs, J. D., Fraser, I. C., Lord, B., et al. (2011). Pharmacological blockade of serotonin 5-HT7 receptor reverses working memory deﬁcits in rats by normalizing cortical glutamate neurotrans- mission. PLoS ONE 6:e20210. doi: 10.1371/journal.pone.0020210 de Lecea, L., Kilduff, T. S., Peyron, C., Gao, X., Foye, P. E., Danielson, P. E., et al. (1998). The hypocretins: hypothalamus-speciﬁc peptides with neuroex- citatory activity. Proc. Natl. Acad. Sci. U.S.A. 95, 322–327. doi: 10.1073/pnas. 95.1.322 Dugovic, C., Shelton, J. E., Yun, S., Bonaventure, P., Shireman, B. T., and Lovenberg, T. W. (2014). Orexin1 receptor blockade dysregulates REM sleep in the presence of orexin2 receptor antagonism. Front. Neurosci. 8:28. doi: 10.3389/fnins.2014.00028 Gozzi, A., Turrini, G., Piccoli, L., Massagrande, M., Amantini, D., Antolini, M., et al. (2011). Functional magnetic resonance imaging reveals different neural substrates for the effects of orexin1 and orexin2 receptor antagonists. PLoS ONE 6:e16406. doi: 10.1371/journal.pone.0016406 Henny, P., Brischoux, F., Mainville, L., Stroh, T., and Jones, B. E. (2010). Immunohistochemical evidence for synaptic release of glutamate from orexin terminals in the locus coeruleus. Neuroscience 169, 1150–1157. doi: 10.1016/j.neuroscience.2010.06.003 Activity of orexin neurons is increased under stressful stimuli leading to release of more orexins in terminal ﬁelds located in the limbic system. It has been postulated that an abnormal persistence of excessive release of orexins could lead to pathological anxiety or panic attacks (Johnson et al., 2012). Emerging data indicate that Hoyer, D., and Jacobson, L. H. (2013). SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: http://www.frontiersin.org/journal/10.3389/fnins.2014. 00107/abstract AUTHOR CONTRIBUTIONS Pascal Bonaventure designed research, analyzed data, and wrote manuscript. Leah Aluisio analyzed data and edited manuscript. Ian Fraser and Tamara Berdyyeva conducted research and ana- lyzed results. Christine Dugovic and James Shoblock participated in research design and edited manuscript. Volha Tryputsen performed the statistical analysis. Brock T. Shireman provided compounds. Timothy Lovenberg participated in research design. In this study, we conducted the experiments during the light cycle when orexin neurons are less active (Lee et al., 2005). Interestingly, the OX1 receptor antagonist only partially blunted the effect of MK-801 on glutamate release. During the dark phase when orexin neurons are more active we postulate that the effect of the OX1 receptor antagonist on blunting MK-801 induced glutamate release might be more pronounced. However, other neurochemicals and/or receptors might be involved in this com- plex phenomenon as suggested by the data obtained in OX1 receptor knockout mice where MK-801 response was only par- tially diminished compared to the response obtained in wild-type animals. The partial response observed in knockout mice or after antagonist administration supports the view that the action of orexin is modulatory. ACKNOWLEDGMENT The assistance of Dr. Kevin Sharp and his staff at Janssen Research and Development, LLC is gratefully acknowledged. Frontiers in Neuroscience \| Neuropharmacology DISCUSSION The present study focused on the orexins/glutamate interaction in cortex, and this important interaction is likely to be involved with arousal during stress or attention to a stressor. Contribution of the OX2 recep- tor for the orexin/glutamate interaction still needs to be studied but from a therapeutic standpoint it is well established that block- ing the OX2 receptor will lead to an hypnotic effect (Hoyer and Jacobson, 2013; Dugovic et al., 2014) whereas in contrast selective OX1 receptor blockade does not alter spontaneous sleep. Selective targeting of the OX1 receptor with an antagonist may normalize hyperglutamatergic states without hypnotic effect and thus may represent a novel therapeutic strategy for the treatment of various psychiatric disorders associated with hyperactive states. REFERENCES Orexin in sleep, addiction and more: is the perfect insomnia drug at hand? Neuropeptides 47, 477–488. doi: 10.1016/j.npep.2013.10.009 Frontiers in Neuroscience \| Neuropharmacology May 2014 \| Volume 8 \| Article 107 \| 4 Orexin1 receptor modulates glutamate release Aluisio et al. Salvadore, G., and Singh, J. B. (2013). Ketamine as a fast acting antidepressant: current knowledge and open questions. CNS Neurosci. Ther. 19, 428–436. doi: 10.1111/cns.12103 John, J., Wu, M.-F., Kodama, T., and Siegel, J. M. (2003). Intravenously admin- istered hypocretin-1 alters brain amino acid release: an in vivo micro- dialysis study in rats. J. Physiol. 548, 557–562. doi: 10.1113/jphysiol.2002. 038729 Schobel, S. A., Chaudhury, N. H., Usman, K. A., Paniagua, B., Styner, M. A., Asllani, I., et al. (2013). Imaging patients with psychosis and a mouse model establishes a spreading pattern of hippocampal dysfunction and implicates glutamate as a driver. Neuron 78, 81–93. doi: 10.1016/j.neuron.2013.02.011 Johnson, P. L., Samuels, B. C., Fitz, S. D., Federici, L. M., Hammes, N., Early, M. C., et al. (2012). Orexin 1 receptors are a novel target to modulate panic responses and the panic brain network. Physiol. Behav. 107, 733–742. doi: 10.1016/j.physbeh.2012.04.016 Tose, R., Kushikata, T., Yoshida, H., Kudo, M., Furukawa, K., Ueno, S., et al. (2009). Interaction between orexinergic neurons and NMDA receptors in the control of locus coeruleus cerebrocortical noradrenergic activity of the rat. Brain Res. 1250, 81–87. doi: 10.1016/j.brainres.2008.10.041 Kodama, T., and Kimura, M. (2002). Arousal effects of orexin-A correlate with GLU release from the locus coeruleus in rats. Peptides 23, 1673–1681. doi: 10.1016/S0196-9781(02)00109-2 Krystal, J., D’Souza, D. C., Mathalon, D., Perry, E., Belger, A., and Hoffman, R. (2003). NMDA receptor antagonist effects, cortical glutamatergic func- tion, and schizophrenia: toward a paradigm shift in medication development. Psychopharmacology 169, 215–233. doi: 10.1007/s00213-003-1582-z Trivedi, P., Yu, H., MacNeil, D. J., Van der Ploeg, L. H. T., and Guan, X.-M. (1998). Distribution of orexin receptor mRNA in the rat brain. FEBS Lett. 438, 71–75. doi: 10.1016/S0014-5793(98)01266-6 Uslaner, J. M., Smith, S. M., Huszar, S. L., Pachmerhiwala, R., Hinchliffe, R. M., Vardigan, J. D., et al. (2012). T-type calcium channel antagonism produces antipsychotic-like effects and reduces stimulant-induced glutamate release in the nucleus accumbens of rats. Neuropharmacology 62, 1413–1421. doi: 10.1016/j.neuropharm.2010.11.015 Large, C. H. (2007). Do NMDA receptor antagonist models of schizophrenia pre- dict the clinical efﬁcacy of antipsychotic drugs? J. Psychopharmacol. 21, 283–301. doi: 10.1177/0269881107077712 Lee, M. G., Hassani, O. K., and Jones, B. E. REFERENCES (2005). Discharge of identi- ﬁed orexin/hypocretin neurons across the sleep-waking cycle. J. Neurosci. 25, 6716–6720. doi: 10.1523/JNEUROSCI.1887-05.2005 van den Pol, A. N., Gao, X. B., Obrietan, K., Kilduff, T. S., and Belousov, A. B. (1998). Presynaptic and postsynaptic actions and modulation of neuroen- docrine neurons by a new hypothalamic peptide, hypocretin/orexin. J. Neurosci. 18, 7962–7971. Li, J., Hu, Z., and de Lecea, L. (2014). The hypocretins/orexins: integrators of multiple physiological functions. Br. J. Pharmacol. 171, 332–350. doi: 10.1111/bph.12415 Westerink, B., and Cremers, T. (2007). The Handbook of Microdialysis Methods, Applications and Clinical Aspects. Oxford: Elsevier BV. Li, Y., Gao, X.-B., Sakurai, T., and van den Pol, A. N. (2002). Hypocretin/orexin excites hypocretin neurons via a local glutamate neuron a potential mechanism for orchestrating the hypothalamic arousal system. Neuron 36, 1169–1181. doi: 10.1016/S0896-6273(02)01132-7 Zhang, J.-H., Sampogna, S., Morales, F. R., and Chase, M. H. (2005). Age-related ultrastructural changes in hypocretinergic terminals in the brainstem and spinal cord of cats. Neurosci. Lett. 373, 171–174. doi: 10.1016/j.neulet.2003.08.085 Lopez-Gil, X., Babot, Z., Amargos-Bosch, M., Sunol, C., Artigas, F., and Adell, A. (2007). Clozapine and haloperidol differently suppress the MK-801-increased glutamatergic and serotonergic transmission in the medial prefrontal cortex of the rat. Neuropsychopharmacology 32, 2087–2097. doi: 10.1038/sj.npp.1301356 Conﬂict of Interest Statement: All the authors are full-time employee of Janssen Research and Development, L.L.C. Received: 31 March 2014; accepted: 24 April 2014; published online: 20 May 2014. Lungwitz, E. A., Molosh, A., Johnson, P. L., Harvey, B. P., Dirks, R. C., Dietrich, A., et al. (2012). Orexin-A induces anxiety-like behavior through interactions with glutamatergic receptors in the bed nucleus of the stria terminalis of rats. Physiol. Behav. 107, 726–732. doi: 10.1016/j.physbeh.2012.05.019 p p p y Citation: Aluisio L, Fraser I, Berdyyeva T, Tryputsen V, Shireman BT, Shoblock J, Lovenberg T, Dugovic C and Bonaventure P (2014) Pharmacological or genetic orexin1 receptor inhibition attenuates MK-801 induced glutamate release in mouse cortex. Front. Neurosci. 8:107. doi: 10.3389/fnins.2014.00107 Marcus, J. N., Aschkenasi, C. J., Lee, C. E., Chemelli, R. M., Saper, C. B., Yanagisawa, M., et al. (2001). Differential expression of orexin receptors 1 and 2 in the rat brain. J. Comp. Neurol. 435, 6–25. doi: 10.1002/cne.1190 This article was submitted to Neuropharmacology, a section of the journal Frontiers in Neuroscience. Naylor, E., Aillon, D. V., Gabbert, S., Harmon, H., Johnson, D. A., Wilson, G. S., et al. (2011). May 2014 \| Volume 8 \| Article 107 \| 5 REFERENCES Simultaneous real-time measurement of EEG/EMG and l-glutamate in mice: a biosensor study of neuronal activity during sleep. J. Electroanal. Chem. 656, 106–113. doi: 10.1016/j.jelechem.2010.12.031 Copyright © 2014 Aluisio, Fraser, Berdyyeva, Tryputsen, Shireman, Shoblock, Lovenberg, Dugovic and Bonaventure. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribu- tion or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Sakurai, T., Nagata, R., Yamanaka, A., Kawamura, H., Tsujino, N., Muraki, Y., et al. (2005). Input of orexin/hypocretin neurons revealed by a genetically encoded tracer in mice. Neuron 46, 297–308. doi: 10.1016/j.neuron.2005.03.010 May 2014 \| Volume 8 \| Article 107 \| 5 www.frontiersin.org www.frontiersin.org
https://openalex.org/W4382927844	https://link.springer.com/content/pdf/10.1007/s43253-023-00101-1.pdf	English	null	Growth drivers in emerging capitalist economies: building blocks for a post-Keynesian analysis and an empirical exploration of the years before and after the Global Financial Crisis	Review of evolutionary political economy	2,023	cc-by	20,536	ERROR: type should be string, got "https://doi.org/10.1007/s43253-023-00101-1\nReview of Evolutionary Political Economy (2023) 4:349–386 https://doi.org/10.1007/s43253-023-00101-1\nReview of Evolutionary Political Economy (2023) 4:349–386 ORIGINAL PAPER Growth drivers in emerging capitalist economies: building \nblocks for a post‑Keynesian analysis and an empirical \nexploration of the years before and after the Global \nFinancial Crisis Benjamin Jungmann1,2 Received: 17 February 2022 / Accepted: 22 May 2023 \n© The Author(s) 2023\n/ Published online: 3 July 2023 Extended author information available on the last page of the article \t Benjamin Jungmann \n\t\nBenjamin.jungmann@hwr-berlin.de \t Benjamin Jungmann \n\t\nBenjamin.jungmann@hwr-berlin.de\nExtended author information available on the last page of the article *\t Benjamin Jungmann Benjamin.jungmann@hwr-berlin.de Keywords Growth model · Growth driver · Financialisation · Emerging economies · \nPost-Keynesian economics · Comparative political economy Abstract This paper contributes to the growth models debate by expanding the concept of growth \ndrivers to emerging capitalist economies (ECEs). Conceptually, the paper synthesizes \ngrowth drivers with a growth model operationalization based on GDP growth contributions \nand financial sector balances. Drawing on post-Keynesian and structuralist economics, as \nwell as, empirical studies, seven growth drivers for ECEs are reviewed: income distribution, \nprice and non-price competitiveness, commodity prices, private debt, foreign direct \ninvestment (FDI) and fiscal policy. Descriptive data for these drivers are presented for 19 \nregionally grouped ECEs between 2000 and 2019. On average, Asian ECEs exhibit higher \ngrowth rates, stable real exchange rates, high and increasing non-price competitiveness and \nhigh private debt levels. Latin American countries show comparatively lower growth rates, \nhigh but decreasing income inequality, unstable exchange rates and relatively expansionary \nfiscal policy after the Global Financial Crisis (GFC). Central and Eastern European \ncountries generally display medium to high growth rates, lower income inequality, high \nnon-price competitiveness, a substantial FDI stock and, after the GFC, real depreciations \nand contractionary fiscal policy. The assessment of cross-country growth drivers via \nbivariate coefficients reveals limited robust results, except for non-price competitiveness, \nwhich emerges as a significant driver. Additionally, we find indications that private debt \nand expansionary fiscal policy became more important for growth in ECEs after the \nGFC. This is in line with the emergence of domestic demand- and private debt-led growth \nmodels in ECEs following the GFC in the course of private deleveraging and austerity \npolicies in developed capitalist economies. Keywords Growth model · Growth driver · Financialisation · Emerging economies · \nPost-Keynesian economics · Comparative political economy JEL classification E11 · E12 · E65 · F62 · F65 JEL classification E11 · E12 · E65 · F62 · F65 012341 456789)\n3 456789)\n3 350 B. Jungmann 1 Originally, post-Keynesian authors used the term “demand and growth regimes” or “macroeconomic \nregimes” while “growth models” originated within CPE. If not otherwise specified, we will use these \nterms interchangeably throughout this article. Irrespective of the term, the concept should not be con-\nfused with the distinction between wage-led and profit-led demand that is going to be introduced in 2.1 \nand works at a different level of analysis.\n2 Stockhammer (2022a) provides an overview on the post-Keynesian fundamentals of the growth models \napproach.\n3 ECEs have featured in some large-scale analysis by post-Keynesian authors without considering their \nspecific economic properties and role in the international growth model constellation (e.g. Dodig et al. \n2016; Hein and Mundt 2012). The term ECEs refers to economies with a capitalist mode of production \nthat feature some but not all of the characteristics of DCEs, e.g. in terms of financial and trade integra-\ntion into world markets or sectoral composition of the economy. f\n2 Stockhammer (2022a) provides an overview on the post-Keynesian fundamentals of the growth mode\napproach.\n3 3 ECEs have featured in some large-scale analysis by post-Keynesian authors without considering their \nspecific economic properties and role in the international growth model constellation (e.g. Dodig et al. \n2016; Hein and Mundt 2012). The term ECEs refers to economies with a capitalist mode of production \nthat feature some but not all of the characteristics of DCEs, e.g. in terms of financial and trade integra-\ntion into world markets or sectoral composition of the economy. 1 Originally, post-Keynesian authors used the term “demand and growth regimes” or “macroeconomic \nregimes” while “growth models” originated within CPE. If not otherwise specified, we will use these \nterms interchangeably throughout this article. Irrespective of the term, the concept should not be con-\nfused with the distinction between wage-led and profit-led demand that is going to be introduced in 2.1 \nand works at a different level of analysis.\n2 Stockhammer (2022a) provides an overview on the post Keynesian fundamentals of the growth models 1 Introduction (2023), on the other hand, look at commodity-driven growth models \nin general and highlight the pro-cyclical effects on these models exerted by capi-\ntal flows and commodity price swings. Mertens et al. (2022) investigate the growth \nmodels of eight ECEs and the social bloc dynamics of four. They add the category \nof an investment-led growth model and make the case that large ECEs may exhibit \nmultiple, regionally different growth models. This is echoed by Tan and Conran \n(2022) who argue that the rise of China was driven by a hybrid system that com-\nprised two growth models: an export-led and a state-led investment one. Akcay et al. (2022) conduct an analysis of growth models in eight large ECEs. In the post-GFC context when DCEs became overall more export-led, they find that \nECEs have not followed the trajectory of DCEs and instead switched to or continued \npursuing domestic demand-led models (India, Argentina and Brazil), private debt-\nled ones (Turkey and South Africa) or decreased their export-led stance (China). Hence, ECEs, together with domestic demand-led DCE like the US and UK, have \ncontributed to the necessary global counterpart to export-led mercantilist DCEs with \nhigh current account surpluses. Furthermore, Akcay et al. look into factors driving \nthese changes. They find that, in some ECEs, improved income distribution helped \nto stabilize income-financed domestic demand while, on the other hand, further \nincreases in financialisation boosted debt-financed private demand. In a similar way, \nCampana et al. (2024) complement their growth model analysis of Brazil, China, \nIndia and Russia by examining factors that drive export growth, income distribution \nand political conflict underlying fiscal policy.4 li\nIn this paper, we take up on the issue of what drives aggregate demand and \nultimately growth in ECEs. In this context, the contribution this paper attempts is \ntwofold. First, on a conceptual level, we address the criticism raised by Kohler and \nStockhammer (2022) that growth model operationalization, which relies solely on \nthe growth contributions of aggregate demand components and financial sector \nbalances, does not provide meaningful insights into the causal drivers of growth. Therefore, they argue for a growth model operationalization based on growth drivers \nwhich are “factors that are not themselves part of aggregate income but influence \nthe growth of its components” (Kohler and Stockhammer 2022, p. 1319). 1 Introduction In the post-Keynesian literature, growth models have been operationalized using the \nGDP growth contributions of aggregate demand – consumption, investment, government \nexpenditure and net exports – and the sectoral financial balances of the macroeconomic \nsectors – households, firms, the government and the external sector, found in national \nincome and financial accounts (Hein 2011, 2012). This decomposition gives insights into \nthe demand sources of growth and their financing at a macroeconomic level. Using this \nmethod to study the macroeconomics of financialisation, yielded the known constellation \nof export-led versus private debt-led growth models prior to the Global Financial Crisis \n(GFC) of 2007–09 (Stockhammer 2015). Meanwhile, this concept has been applied to \nthe post-GFC constellation (Akcay et al. 2022; Dodig et al. 2016; Hein 2019; Dünhaupt \nand Hein 2019; Hein and Martschin 2020; Hein et al. 2021). Furthermore, commenced \nby Baccaro and Pontusson (2016), the ‘growth models approach’ gained prominence \nwithin Comparative Political Economy (CPE).1 This approach marks a shift within \nComparative Capitalism from the Varieties of Capitalism (VoC) approach based on \nNew Consensus Macroeconomics with supply-side determined long-run equilibria \nto post-Keynesian based demand-focused approaches. By doing so, issues of demand \ngeneration, instability, economic policies and international interdependencies move \ncentre stage.2 While this approach has mostly been applied to developed capitalist economies \n(DCEs) (e.g. Hall 2018; Johnston and Regan 2018; Hein 2019; Hein and Martschin \n2020; Hein et al. 2021; Van Doorslaer and Vermeiren 2021; Hassel and Palier 2021; \nBaccaro et al. 2022 and the contributions therein), attempts have been made to \nextend it to emerging capitalist economies (ECEs).3 Schedelik et al. (2021) examine \nthe growth trajectories of India, Brazil and China. They recommend the growth \nmodels approach to study ECEs while emphasizing the importance of retaining \ninstitutionalist aspects from the VoC literature. Nölke et al. (2022) compare the \ndifferent growth trajectories of India and Brazil. Looking at VoC-categories and \napplying a growth models approach analysing demand formation, distribution and \nthe dominant social bloc, they conclude that the latter is better suited to explain the \ndivergence of the two countries. Morgan et al. (2021) examine the tension between \ninstitutional embeddedness and politics and its impact on growth model change by \nanalysing Brazil’s experience between 2002 and 2018. 1 3 Growth drivers in emerging capitalist economies: building… 351 Relatedly, Sierra (2022) investigates the underlying factors contributing to the \npersistence of commodity-driven growth models in Latin American countries. Sche-\ndelik et al. 4 Campana et al. (2024) conduct a growth decomposition based on the Sraffian supermultiplier, distin-\nguishing autonomous demand components and their contributions from that of induced components. \nWithin the growth models strand, this method was first revitalized by Morlin et al. (2022) examining the \nUnited States, Germany, Japan and Sweden. Passos and Morlin (2022) applied this method to growth \nmodels in Latin America. 1 Introduction However, \ninstead of arguing replacing one by the other, we make the case for a conceptual \nsynthesis and present a set of potential growth drivers in ECEs. Building on post-\nKeynesian and structuralist economics as well as empirical studies on ECEs, seven \ngrowth drivers for ECEs are reviewed: income distribution, price and non-price \ncompetitiveness, commodity prices, private debt, foreign direct investment (FDI) \nand fiscal policy. 1 3 352 B. Jungmann Second, we explore these growth drivers empirically for 19 ECEs.5 We pro-\nvide descriptive data on each growth driver for these ECEs in a regional grouping \nfrom 2000 and 2019. Furthermore, to investigate which growth drivers determined \ngrowth on a cross-country level in the years before and after the GFC, we establish \nbivariate coefficients. For the descriptive data, we find that the Asian ECEs exhibit \non average higher growth rates, stable real exchange rates (RERs), higher and ris-\ning non-price competitiveness and high private debt levels. In Latin America, we \nfind comparatively low growth rates, high levels of income inequality that are abat-\ning, unstable RERs and relatively expansionary fiscal stances after the GFC. The \ncountries of Central and Eastern Europe (CEE), often with the exception of Russia, \nexhibit medium to high growth rates, comparatively low levels of income inequal-\nity, high non-price competiveness and a high FDI stock. Moreover, their RERs were \nrather stable, particularly, after the GFC when they achieved overall depreciations \nwhile their fiscal policy stances became less expansionary. In terms of cross-country \ngrowth drivers derived from bivariate coefficients, our results remain sparse with \nthe exception of non-price competitiveness that drove cross-country growth. Finally, \nin line with the descriptive findings of overall higher private debt and expansionary \nfiscal policy stances after the GFC, we find indications of both factors becoming \nmore relevant as cross-country growth drivers during that time. This is in line with \nthe emergence, respectively, persistence of domestic demand- and private debt-led \ngrowth models in ECEs following the GFC in the course of private deleveraging and \nausterity policies in DCEs (Akcay et al. 2022). The remainder is structured as follows: Section 2 places the concept of growth \ndrivers within growth model operationalization and reviews seven possible growth \ndrivers for ECEs. 5 Our sample encompasses the Latin American ECEs of Argentina, Brazil, Chile, Colombia and Mex-\nico; the Asian ECEs of China, Indonesia, India, Korea, Malaysia and Thailand; the Central and Eastern \nEuropean ECEs of the Czech Republic, Hungary, Poland and Russia; the Middle Eastern ECEs of Israel, \nSaudi Arabia and Turkey; and South Africa. The sample was largely determined by data availability. 1 Introduction Section 3 presents, first, descriptive data of each growth driver \nfor 19 ECEs between 2000 and 2019 and, second, bivariate coefficients relat-\ning the development in growth drivers to GDP growth for the pre- and post-GFC \nperiod. Section 4 summarizes and discusses the empirical observations. Section 5 \nconcludes. 2 \u0007Building blocks for a post‑Keynesian analysis of growth drivers \nin emerging capitalist economies In their paper, Kohler and Stockhammer (2022) have questioned the usefulness of the \ncommon post-Keynesian method of growth model operationalization for the post-GFC \nperiod. This method of growth decomposition operationalizes growth models based \non the GDP growth contributions of aggregate demand and sectoral financial balances \n(e.g. Hein 2011, 2012, 2019). According to Kohler and Stockhammer (2022), the \nassessment of the formerly private debt-led Southern European peripheral economies 1 3 Growth drivers in emerging capitalist economies: building… 353 as export-led in the post-GFC years is misleading, as their export-led characteristics \nrather stem from private deleveraging and fiscal austerity depressing imports than \nfrom sustained export growth. Studies that identify these economies as export-led from \na growth decomposition standpoint also cite private deleveraging and fiscal austerity \nas driving this development (Dodig et al. 2016; Hein and Martschin 2020; Hein et al. 2021). Still, Kohler and Stockhammer (2022) argue growth models should instead be \nunderstood through their growth drivers, as they offer information on why or why not \ndemand components grew. The drivers considered by the authors are financial cycles, \nfiscal policy and international price as well as non-price competitiveness. In this \nperspective, the Southern European peripheral growth models underwent a debt-driven \ndepression following their pre-GFC debt-driven growth due to the cyclical nature of \nfinance-driven growth, exacerbated by contractionary fiscal policy. ii\nWe acknowledge the limitations of using GDP growth decompositions as the \nsole basis for growth model operationalization. While these decompositions \nidentify the most dynamic aggregate demand components, they do not explain \nwhy these components grew. However, one should not throw the baby out with \nthe bath water. One merit of growth model operationalization via national income \nand financial accounting is that they consider sectoral financial balances, which \nillustrate the financing of aggregate demand within and across economies, \nindicating the sustainability of and the interdependencies between growth models. This is particularly important as analysing national capitalisms in their totality \nand interdependence rather than in isolation sets the growth models approach \napart from VoC (Schedelik et al. 2021, p. 518). And although the export-led \nversus private debt-led dichotomy came to an end with the GFC, international \ninterdependencies between growth models persist. After the GFC, debt-led \nprivate demand growth models have ceased to exist among DCEs while export-led \nones prevailed and increased in number. 2 \u0007Building blocks for a post‑Keynesian analysis of growth drivers \nin emerging capitalist economies Domestic demand-led DCEs and ECEs \nwith high public deficits and debt-led private demand ECEs have become the \ncounterpart for export-led DCEs (Hein et al. 2021; Akcay et al. 2022). In general, \nnational income and financial accounts will always provide relevant information \non macroeconomic developments, even if they cannot in themselves provide causal \nexplanations for demand and growth. Applicants of this method are well aware \nof this and complement their growth model analyses therefore with indicators of \ndistribution and financialisation (e.g. Hein 2011, 2019; Hein and Mundt 2012; \nAkcay et al. 2022), link it to welfare models (Hein et al. 2021) or embed it in \na comprehensive analysis of macroeconomic policies (Hein and Martschin 2021) \n– factors that can be considered growth drivers. We advocate hence for the synthesis of growth decomposition and growth driver \nanalysis because both approaches have their merits and inform each other. For once, \ngrowth models rely on growth drivers to derive demand and growth. And seemingly \nsimilar growth models from a growth decomposition standpoint may differ \nsignificantly depending on their growth drivers, e.g. in terms of sustainability and \ncyclicality. On the other hand, different growth drivers may become more relevant \ndepending on the international growth model constellation, e.g. an economy may \nsee it easier to grow based on growth drivers that stimulate exports within an \ninternational growth model constellation that supplies sufficient external demand 1 3 354 B. Jungmann while in the face of depressed external demand growth drivers that stimulate \ndomestic demand become more relevant (Akcay et al. 2022). while in the face of depressed external demand growth drivers that stimulate \ndomestic demand become more relevant (Akcay et al. 2022). In what follows, we will set out the building blocks for an analysis of growth \ndrivers in ECEs.6 Building on post-Keynesian and structuralist economics as well \nas empirical studies on ECEs, seven growth drivers for ECEs are reviewed: income \ndistribution, price and non-price competitiveness, commodity prices, private debt, \nFDI and fiscal policy. Arguing for the synthesis of growth model operationalization \nvia national income and financial accounting with that via growth drivers, we point \nout how different growth drivers are expected to shape growth models. We will refer \nhere to three types of growth models common in the post-Keynesian literature (e.g. Hein 2011, 2019; Hein et al. 2021; Akcay et al. 6 For a more general account on how to employ post-Keynesian economics and economic structuralism \nto analyse growth models in ECEs and developing economies see Stockhammer (2022b).\n7 A fourth category found in the literature is that of a weakly export-led growth model. Such a model \nshares some but not all features of an export-led mercantilist model and features of the other two mod-\nels, e.g. it may exhibit positive growth contributions by net exports with a negative current account. We \nhence view it as an intermediate growth model which we for now exclude from the conceptual argument \nlinking growth models and drivers. Further growth models are possible, particularly as the presented \nones were derived from an analysis of DCEs. For example, Mertens et al. (2022) added an investment-led \nmodel, which for our purposes can be considered as a form of a domestic demand-led one. 2 \u0007Building blocks for a post‑Keynesian analysis of growth drivers \nin emerging capitalist economies 2022): 1) a domestic demand-led \ngrowth model is characterized by positive or balanced financial balances of the pri-\nvate sector and a balanced or negative current account. Correspondingly, net exports \ndo not contribute to growth which almost exclusively stems from domestic demand; \n2) an export-led mercantilist growth model exhibits positive financial balances of the \nprivate sector and a positive current account with positive net exports contributing \nto growth; 3) a debt-led private demand (boom) growth model is characterized by \nnegative or close to balance financial balances of the private sector, in particular, of \nthe household sector and a negative current account. Growth is driven by domes-\ntic demand, particularly, private demand and net exports contribute negatively to \ngrowth (Hein 2019, p. 980).7 2.1 \u0007Income distribution: rising wage share and reduced income inequality Based on post-Kaleckian distribution and growth models, we review changes in \nthe income distribution as a possible driver of growth (Bhaduri and Marglin 1990). In this framework, economies are either classified as wage-led if their demand \nand growth is positively affected by an increasing wage share or as profit-led in \nthe opposite case. Whether an economy exhibits a wage-led or profit-led demand \nregime rests on structural properties such as the different propensities to consume, \nthe responsiveness of investment to demand and profitability, the price sensitivity \nof net exports and the relevance of the respective aggregate demand component for \ntotal demand (Lavoie and Stockhammer 2013). Determining an economy’s demand \nregime is an empirical task: More often than not, domestic demand is found to be \nwage-led. This is due to the positive effect of an increased wage share on consump-\ntion because of higher propensities to consume out of wage income than out of profit 1 1 3 Growth drivers in emerging capitalist economies: building… 355 income. Meanwhile, the effect of changes in income distribution on investment is \noften found to be insignificant. For total demand, profit-led cases are more likely \ndue to a positive effect of the profit share on net exports. A total profit-led demand \nregime requires the rise in price competitiveness due to a lower wage share and the \nrespective increase in net exports to outweigh the depressing effects on consumption \nout of wages. In this context, ECEs are more likely to exhibit total profit-led demand \nregimes due to their relative openness and price sensitive exports, especially, if they \nare small (see Table 1 and Hein (2014, pp. 302–303)). In any case, as stressed by \nLavoie and Stockhammer (2013), the identified regime type neither implies that the \nfunctional income distribution developed accordingly nor that policies were applied \nto achieve such development; for example, a wage-led economy can exhibit a rising \nprofit share due to pro-capital policies. i\nParticularly due to rising wage inequality, the exclusive focus on the functional income \ndistribution has been questioned. Hein and Prante (2020) review the different Kaleckian \ngrowth models accounting for wage inequality: Some models distinguish direct from \nindirect/overhead labour, thereby, in the short run, the wage share becomes endogenous \nto economic activity in an inverse way, making demand appear profit-led when in fact \nthe causality is reversed (e.g. Lavoie 2009). 8 We focus here on increases in the wage share and reduced income inequality as profit-led regimes rest \non the expansionary effects of price competitiveness, which will be reviewed as a growth driver of its \nown in the next section. 2.1 \u0007Income distribution: rising wage share and reduced income inequality Alternatively, models split profits and wages \nbetween workers who own part of the capital stock and capitalists who receive wages \nin their function as managers. These models yield expansionary effects from increased \nworkers’ wage share irrespective of the demand regime due to workers’ lower propensity \nto save. Thus, higher workers’ wage shares increase the probability of wage-led demand as \nthe overall propensity to save of wage income falls (Palley 2017). Another type of model \nincorporates persistence in basic consumption needs and interdependent consumption \npatterns, where lower-income ranks mimic the consumption behaviour of higher ranks. This leads to increased consumption and private debt ratios in response to higher profit \nshares and income inequality (e.g. Kapeller and Schütz 2015). Hence, according to these \nmodels, growth models based on private debt would be unlikely to occur with increases in \nwage shares and decreases in income inequality. We conclude that growth driven by increasing wage shares and reduced income \ninequality are conducive to domestic demand-led regimes.8 Domestic demand-led \nregimes have most of their growth stemming from domestic demand components, \nprivate consumption being usually the biggest. Since domestic demand is over-\nwhelmingly found to be wage-led, it will be boosted by increases in the wage share \nand reduction in income inequality due to the higher propensities to consume out of \nwages and low income households. 2.2 \u0007Price competitiveness As outlined in the previous section, increases in price competitiveness, i.e. real \ndepreciations, increase demand if the rise in net exports it triggers outweighs the 8 We focus here on increases in the wage share and reduced income inequality as profit-led regimes rest \non the expansionary effects of price competitiveness, which will be reviewed as a growth driver of its \nown in the next section. 1 3 B. Jungmann 356 able 1 Overview of studies investigating ECEs demand regimes\nCountry\nDomestic demand\nTotal demand\nWage-led\nProfit-led\nWage-led\nProfit-led\nArgentina\nOnaran and Galanis (2012): 1970–\n2007; Alarco (2016): 1950–2012\nAlarco (2016): 1950–2012; Oyvat \net al. (2020): 1972–2007\nOnaran and Galanis (2012): \n1970–2007\nBrazil\nAlarco (2016): 1950–2012; Tomio \n(2020): 1956–2008; Araújo and \nGala (2012): 1960–2008\nAlarco (2016): 1950–2012; Tomio \n(2020): 1956–2008\nAraújo and Gala (2012): 1960–2008; \nde Jesus et al. (2018): 1970–2008\nChile\nAlarco (2016): 1950–2012\nAlarco (2016): 1950–2012; Oyvat \net al. (2020): 1967–1994\nChina\nOnaran and Galanis (2012): \n1978–2007; Jetin and Reyes Ortiz \n(2020): 1982–2016\nWang (2009, Chapter 3): 1993–\n2007; Molero-Simarro (2015): \n1978–2007\nJetin and Reyes Ortiz (2020): \n1982–2016\nOnaran and Galanis (2012): 1970–\n2007; Wang (2009, Chapter 3): \n1993–2007; Molero-Simarro (2015): \n1978–2007\nColombia\nAlarco (2016): 1950–2012\nAlarco (2016): 1950–2012; Loaiza \net al. (2017): 1970–2011\nOyvat et al. (2020): 1967–2011; \nCharpe et al. (2014): 1970–2010\nndia\nOnaran and Galanis (2012): \n1970–2007\nOnaran and Galanis (2012): \n1970–2007; Oyvat et al. (2020): \n1964–2011\nKohli (2018) finds it for 1981–2012 to be either wage- or profit-led depend-\ning on the source of the distributional change\nndonesia\nOyvat et al. (2020): 1971–2011\nKorea\nOnaran and Galanis (2012): 1970–\n2007; Kurt (2018): 1970–2011; \nJoo et al. (2020): 1982–2018\nOnaran and Galanis (2012): 1970–\n2007; Onaran and Stockhammer \n(2005): 1970–2000; Oyvat et al. (2020): 1964–2011; Joo et al. (2020): 1982–2018\nKurt (2018): 1970–2011\nMalaysia\nOyvat et al. (2020): 1972–2011 1 3\nTable 1 Overview of studies investigating ECEs demand regimes\nCountry\nDomestic demand\nTotal demand\nWage-led\nProfit-led\nWage-led\nProfit-led\nArgentina\nOnaran and Galanis (2012): 1970–\n2007; Alarco (2016): 1950–2012\nAlarco (2016): 1950–2012; Oyvat \net al. (2020): 1972–2007\nOnaran and Galanis (2012): \n1970–2007\nBrazil\nAlarco (2016): 1950–2012; Tomio \n(2020): 1956–2008; Araújo and \nGala (2012): 1960–2008\nAlarco (2016): 1950–2012; Tomio \n(2020): 1956–2008\nAraújo and Gala (2012): 1960–2008; \nde Jesus et al. (2018): 1970–2008\nChile\nAlarco (2016): 1950–2012\nAlarco (2016): 1950–2012; Oyvat \net al. 2.2 \u0007Price competitiveness (2020): 1967–1994\nChina\nOnaran and Galanis (2012): \n1978–2007; Jetin and Reyes Ortiz \n(2020): 1982–2016\nWang (2009, Chapter 3): 1993–\n2007; Molero-Simarro (2015): \n1978–2007\nJetin and Reyes Ortiz (2020): \n1982–2016\nOnaran and Galanis (2012): 1970–\n2007; Wang (2009, Chapter 3): \n1993–2007; Molero-Simarro (2015): \n1978–2007\nColombia\nAlarco (2016): 1950–2012\nAlarco (2016): 1950–2012; Loaiza \net al. (2017): 1970–2011\nOyvat et al. (2020): 1967–2011; \nCharpe et al. (2014): 1970–2010\nIndia\nOnaran and Galanis (2012): \n1970–2007\nOnaran and Galanis (2012): \n1970–2007; Oyvat et al. (2020): \n1964–2011\nKohli (2018) finds it for 1981–2012 to be either wage- or profit-led depend-\ning on the source of the distributional change\nIndonesia\nOyvat et al. (2020): 1971–2011\nKorea\nOnaran and Galanis (2012): 1970–\n2007; Kurt (2018): 1970–2011; \nJoo et al. (2020): 1982–2018\nOnaran and Galanis (2012): 1970–\n2007; Onaran and Stockhammer \n(2005): 1970–2000; Oyvat et al. (2020): 1964–2011; Joo et al. (2020): 1982–2018\nKurt (2018): 1970–2011\nMalaysia\nOyvat et al. (2020): 1972–2011 Growth drivers in emerging capitalist economies: building… 357 g g\np\ng\nSource: Jiménez (2020), Akcay et al. (2022) and own extension\nNo studies on the demand regimes of the Czech Republic, Hungary, Israel, Poland, Russia and Saudi Arabia\nTable 1 (continued)\nCountry\nDomestic demand\nTotal demand\nWage-led\nProfit-led\nWage-led\nProfit-led\nMexico\nOnaran and Galanis (2012): 1970–\n2007; Alarco (2016): 1950–2012\nAlarco (2016): 1950–2012\nOnaran and Galanis (2012): \n1970–2007; Oyvat et al. (2020): \n1972–2009; Charpe et al. (2014): \n1970–2011\nSouth Africa\nOnaran and Galanis (2012): \n1970–2007\nOyvat et al. (2020): 1972–2007; \nStrauss and Isaacs (2016): \n1970–2013\nOnaran and Galanis (2012): \n1970–2007\nThailand\nJetin and Kurt (2016): 1970–2011\nJetin and Kurt (2016): 1970–2011\nTurkey\nOnaran and Galanis (2012): 1970–\n2007; Yılmaz (2015): 1987–2006\nOnaran and Galanis (2012): 1970–\n2007; Onaran and Stockhammer \n(2005): 1963–1997; Oyvat et al. (2020): 1964–2009;\nYılmaz (2015): 1987–2006 1\nSource: Jiménez (2020), Akcay et al. (2022) and own extension\nNo studies on the demand regimes of the Czech Republic, Hungary, Israel, Poland, Russia and Saudi Arabia\nTable 1 (continued)\nCountry\nDomestic demand\nTotal demand\nWage-led\nProfit-led\nWage-led\nProfit-led\nMexico\nOnaran and Galanis (2012): 1970–\n2007; Alarco (2016): 1950–2012\nAlarco (2016): 1950–2012\nOnaran and Galanis (2012): \n1970–2007; Oyvat et al. (2020): \n1972–2009; Charpe et al. (2014): \n1970–2011\nSouth Africa\nOnaran and Galanis (2012): \n1970–2007\nOyvat et al. 2.2 \u0007Price competitiveness (2020): 1972–2007; \nStrauss and Isaacs (2016): \n1970–2013\nOnaran and Galanis (2012): \n1970–2007\nThailand\nJetin and Kurt (2016): 1970–2011\nJetin and Kurt (2016): 1970–2011\nTurkey\nOnaran and Galanis (2012): 1970–\n2007; Yılmaz (2015): 1987–2006\nOnaran and Galanis (2012): 1970–\n2007; Onaran and Stockhammer \n(2005): 1963–1997; Oyvat et al. (2020): 1964–2009;\nYılmaz (2015): 1987–2006 Source: Jiménez (2020), Akcay et al. (2022) and own extension\nNo studies on the demand regimes of the Czech Republic, Hungary, Israel, Poland, Russia and Saudi Arabia\nTable 1 (continued)\nCountry\nDomestic demand\nTotal demand\nWage-led\nProfit-led\nWage-led\nProfit-led\nMexico\nOnaran and Galanis (2012): 1970–\n2007; Alarco (2016): 1950–2012\nAlarco (2016): 1950–2012\nOnaran and Galanis (2012): \n1970–2007; Oyvat et al. (2020): \n1972–2009; Charpe et al. (2014): \n1970–2011\nSouth Africa\nOnaran and Galanis (2012): \n1970–2007\nOyvat et al. (2020): 1972–2007; \nStrauss and Isaacs (2016): \n1970–2013\nOnaran and Galanis (2012): \n1970–2007\nThailand\nJetin and Kurt (2016): 1970–2011\nJetin and Kurt (2016): 1970–2011\nTurkey\nOnaran and Galanis (2012): 1970–\n2007; Yılmaz (2015): 1987–2006\nOnaran and Galanis (2012): 1970–\n2007; Onaran and Stockhammer \n(2005): 1963–1997; Oyvat et al. (2020): 1964–2009;\nYılmaz (2015): 1987–2006 1 3 358 B. Jungmann domestic demand-depressing effects caused by the redistribution towards profits \n(Hein 2014, Chapter 7). Besides distributional issues, negative balance sheet effects \ncan also impede the expansionary effects of increased price competitiveness. This \nis because real depreciations increase the cost of external debt, which in turn can \nconstrain demand (e.g. Krugman 1999). Besides these potentially negative effects, \nthere exists an extensive body of literature stressing the positive effects of price \ncompetitiveness, proxied as a depreciated or low RER, particularly in developing \neconomies and ECEs (see Rapetti (2020) for an overview). Within this literature, \nprice competitiveness fosters growth via the ‘tradable-led growth channel’. This \nchannel stresses the role of ‘modern tradable activities’ and the associated structural \ntransformation towards higher productivity activities. The channel rests broadly \non three elements: higher level of productivity from modern tradable activities, \nan increase in overall productivity through structural change towards modern \ntradables and a low and stable RER that corrects for market failures and promotes \naccumulation in these sectors (Rapetti 2020, p. 36). Conversely, an appreciated RER \nmay avert such favourable structural transformation and depress growth.f In sum, increased price competitiveness may affect growth negatively through negative \nbalance sheet and distributional effects. On the other hand, it may boost growth via net \nexports and through the tradable-led growth-channel. 9 The indicator is derived through the export basket. Each export basket is classified according to the \nubiquity and diversity of its components—the more diverse and non-ubiquitous its export basket, the \nhigher a country’s ECI. 2.2 \u0007Price competitiveness Due to its positive effect on net \nexports and associated negative effects on consumption demand, price competitiveness as \na growth driver will be conducive for export-led growth models. 10 However, as we will argue further below, export-led growth models may also be based on commod-\nity prices. Thus export-led growth models from a growth decomposition perspective do not necessarily \nrequire non-price (or price) competitiveness. 2.3 \u0007Non‑price competitiveness The previously outlined tradable-led growth-channel bears resemblance with the \nliterature that stresses the importance of technological capabilities, i.e. non-price \ncompetitiveness. The importance of non-price factors can be derived from Thirlwall’s \n(1979) law according to which growth in an open economy is constrained by the ratio \nbetween the growth rate of exports and the income elasticity of demand for imports. The growth rate of exports can be decomposed into the rest of the world’s income \nelasticity of demand for the home country’s exports and the rate of growth of the rest \nof the world’s income. Thus, the growth rate depends on two income elasticities, both \ndetermined by technological capabilities (McCombie 1989). The more technological \ncapabilities a country has, the more complex and differentiated the products it can \nproduce. More sophisticated goods are characterized by higher income elasticities \nof demand for exports and lower income elasticity of demand for imports and hence \nassociated with higher growth rates (Gouvêa and Lima 2010). The importance of \nincome elasticities of exports and imports also has a long tradition in economic \nstructuralism (Ocampo and Parra 2006). Hidalgo and Hausmann (2009) have introduced the economic complexity index \n(ECI) to quantify the technological capabilities of an economy.9 Structuralist 1 3 Growth drivers in emerging capitalist economies: building… 359 scholars applied the ECI to vindicate their reasoning on the pivotal role of tech-\nnological capabilities in fostering economic growth and development via increas-\ning returns to scale and positive externalities (Gala et al. 2018). For DCEs, Gräbner \net al. (2020) show the crucial role of technological capabilities and their relation \nto growth models in explaining the divergences within the Eurozone. They argue \nthat export-led growth models require a certain degree of technological capabilities, \nwhile in the absence of these capabilities economies might tend to develop private \ndebt-led growth models.10 Likewise, the macroeconomic policy regime approach \n(Hein and Martschin 2021) and the analysis of growth drivers (Kohler and Stock-\nhammer 2022) also suggest the importance of non-price competitiveness for DCEs. As non-price competitiveness mainly boosts exports it is conducive to export-led \ngrowth models. But non-price competitiveness does not necessarily depress private \nconsumption as price competitiveness likely will. Non-price competitiveness can \nthus also feature in rather domestic demand-led growth models as current account \nsurpluses are not inevitable. 11 Schedelik et al. (2023) emphasize that commodity price-driven growth models are affected not only \nby the pro-cyclical behaviour of commodity prices but also, simultaneously, by that of capital flows. In \ntheir analysis, Campana et al. (2024) highlight the specific role of commodities in the growth models of \nBrazil and Russia. Similarly, Passos and Morlin (2022) and Sierra (2022) examine commodity exports as \na salient feature of Latin American growth models.\n12 During asset price booms, rising real estate prices increase household wealth, boost residential invest-\nment and allow real estate to be collateralized for credit-financed consumption (Stockhammer and Wil-\ndauer 2016). 2.4 \u0007Commodity prices Price and non-price competitiveness drive manufactured exports. However, ECEs’ \nexport basket are often characterized by a large share of commodities. UNCTAD \n(2021, p. 14) classifies around 55 per cent of ECEs as commodity dependent, \nmeaning their shareof commodities in total exports exceeds 60 per cent. For DCEs, \nthis value is only at 28 per cent, justifying the neglect of commodityprices as a \npotential growth driver for this country group by Kohler and Stockhammer (2022). Commodity prices, with the exception of oil, are essentially demand-driven and \nfollow a cyclical movement over 20–70 years, leading to the notion of ‘commodity \nsuper-cycles’ (Erten and Ocampo 2013). Furthermore, the financialisation of \ncommodities also contributes to commodity prices’ cyclicality (Pollin and Heintz \n2011). Investigating the effects of these cycles on economic activity, Fernández \net al. (2020) find that commodity super cycles play a determining role for aggregate \noutput in small and open DCEs and ECEs.f Besides the positive effects on external demand and income, rising commodity \nprices might also increase growth through the increased financial resilience of \nthe economy as the rising current account surplus allows for the accumulation of \ninternational reserves that can be used to cushion the detrimental effects of financial \nhavoc. Moreover, a favourable development of the terms of trade can allow for \nincreased imports for production purposes (Menezes and Souza 2019). However, \nif not counteracted, the increased export volume can lead to real appreciation with \ndetrimental effects on manufacturing industries’ price competitiveness leading to \npremature de-industrialization – as described by the concept of ‘Dutch disease’ \n(Bresser-Pereira 2008). Furthermore, as stressed by structuralist, reliance on 1 3 360 B. Jungmann commodity exports comes with deteriorating terms of trade and adverse effects on \nthe growth trajectory (Ocampo and Parra 2006). As commodity prices boost export demand, growth driven by commodity prices \nwill likely be export-led. An export-led commodity prices-driven growth model will \nexhibit different characteristics than export-led models driven by price or non-price \ncompetitiveness. If not counteracted rising commodity exports are mutually exclu-\nsive to price competitiveness in the short term with detrimental effects for non-price \ncompetitiveness in the longer term. Moreover, growth models based on commodity \nprices will exhibit the cyclical features of these prices and declining terms of trade \naffect the long-term prospects of growth negatively.11 2.5 \u0007Private debt: financialisation, financial development and cycles This stems from the depressing effects of \nfinancialisation on real corporate investment and an increased use of NFC debt for \nfinancial activities and payouts (Davis 2017). Likewise, for some ECEs, studies find \nrising NFC indebtedness to be associated with heightened involvement of NFCs in \nfinancial activities, increased holding of liquid assets and financial payouts at the \nexpense of real investment (Bonizzi 2013, p. 89, for an overview). At the same time, \ngrowth model analysis on ECEs indicates that investment demand plays a more \nprominent role there, leading Mertens et al. (2022) to add the category of an invest-\nment-led growth model. Hence, we take into account the possibility of NFC debt as the base of real invest-\nment, thereby driving growth. In general, finance occupies an important position \nwithin post-Keynesian economics as creditworthy firms can take out loans, irre-\nspective of prior savings, to finance investment and start production (Lavoie 2014, \nChapter 4). Correspondingly, post-Keynesian authors advocate for a financial sector \nsophisticated enough to provide necessary finance for NFC (Priewe and Herr 2005, \nChapter 4). However, this advocacy shall not be mistaken for full-blown financial \nliberalization which is deemed harmful for growth and development (e.g. Arestis \n2006). Increasing indebtedness in ECEs is hence not necessarily negatively connoted in \nthe context of growth. Rather, one has to distinguish favourable expansion of pri-\nvate debt associated with the built-up of productive capacities from unfavourable \nexpansions. Unfavourable expansions of private debt do not lead to the build-up of \nproductive capacities and include debt-financed household consumption, residential \ninvestment and debt-financed financial activities and financial payouts by NFCs. iii\nIn all cases, debt expansion follows cyclical dynamics as stressed by Kohler \nand Stockhammer (2022) by referring to the Financial Instability Hypothesis \n(Minsky 1977). During a boom, asset prices rise, credit expands and financial \npositions become increasingly risky. This built-up fragility eventually leads to \nbusts marked by falling asset prices, deleveraging and reduced spending. On top \nof theses domestic endogenous dynamics, ECEs’ cycles may be characterized \nby their subordinate integration into financial markets. This subordination stems \nfrom the inferior position their currencies occupy in the international currency \nhierarchy as international investors assign a lower liquidity premium to their \ncurrencies compared to key currencies. To attract capital inflows, ECEs must \ncompensate for the lower liquidity premium with higher interest rates, which \nin turn adversely affect corporate investment (Bortz and Kaltenbrunner 2018). 2.5 \u0007Private debt: financialisation, financial development and cycles In debt-led private demand growth models in DCEs, like in the US, UK and south-\nern Europe, wealth-based and debt-financed private consumption and residential \ninvestment drove demand and growth (Stockhammer 2015).12 AAlso ECEs followed \ndebt-led private demand growth models, namely Mexico and Hungary before the \nGFC, Turkey afterwards, and South Africa in both periods (Akcay et al. 2022, p. 83). But private debt also played a significant role in ECEs where no debt-led private \ndemand growth model was identified. Akcay et al. (2022, p. 89–91) argue that grow-\ning private debt contributed to the Chinese growth model becoming less export-led \nafter the GFC. In what follows, we will examine the role of private debt as a growth \ndriver in ECEs distinguishing between household debt, on the one hand, and debt of \nnon-financial corporations (NFCs) on the other. For NFC debt, we distinguish finan-\ncial sector development conducive to long-term growth from construction booms \nand debt used for financial activities. Finally, we emphasize the cyclicality of growth \ndriven by private debt in the context of EMEs’ subordinate integration into global \nfinancial markets. i\nFinancialisation in ECEs has come with increased indebtedness of households. Following the GFC, household debt growth accelerated in several ECEs (Karwowski \nand Stockhammer 2017, p. 76) Country- or region-specific accounts of this rise can \nbe found in Karwowski (2012) and Ashman et al. (2011) for South Africa; Chang \n(2016) for South Korea; Rethel (2010) for Malaysia; dos Santos (2013) for Bra-\nzil, Mexico, Poland and Turkey; Karacimen (2016) for Turkey; and Gabor (2010) \nfor Central and Eastern Europe. Notwithstanding the variegated manifestations of \nhousing financialisation among ECEs, e.g. in terms of mortgage debt to GDP and \nhomeownership rates (Fernandez and Aalbers 2020), the listed studies and reports 3 Growth drivers in emerging capitalist economies: building… 361 on China (Bird 2020) indicate that the increase in household debt in ECEs is largely \ndriven by housing-related financing needs. Household debt as a growth driver \nshould then come with increased residential investment and rising house prices. Despite their rise, household debt-to-GDP levels in ECEs are still relatively \nlow compared to those in private debt-led DCEs. The levels of NFC debt in ECEs, \nhowever, are much closer to those in DCEs (Karwowski and Stockhammer 2017, \npp. 75–76). The growth model literature on DCEs disregards NFC debt for real \ninvestment as a potential growth driver. 2.5 \u0007Private debt: financialisation, financial development and cycles Furthermore, ECEs that rely on capital inflows face the problem of their debt \nbeing of short maturity and denominated in foreign currency, making them \nadditionally vulnerable to exchange rate movements (Arestis and Glickman 1 3 3 3 362 B. Jungmann 2002). Concomitant balance sheet effects amplify the amplitude of the business \ncycle: during the boom, pro-cyclical capital flows to ECEs and exchange rate \nappreciations increase the net worth of firms with foreign currency denominated \ndebt boosting their investment. Conversely, during the bust, negative balance \nsheet effects amplify the contraction (Kohler 2019).13l f\nCapital flows to ECEs are not only pro-cyclical with respect to their domestic \neconomic activity. They are also partly determined by the decisions of institutional \ninvestors in DCEs. Liquidity considerations and global factors, such as the monetary \npolicy in DCEs, lead investors to withdraw or invest capital in ECEs (Bonizzi 2017). The respective cyclical and secular movement in capital flows, asset prices and \ncredit growth has become known as the global financial cycle (Rey 2015). In this \ncontext, ECEs’ capital inflows increase during expansionary and decrease during \nrestrictive US monetary policy stances (Bräuning and Ivashina 2020; Ahmed and \nZlate 2014). Otherwise put, financial cycles in financially globally integrated ECEs \nare not only pro-cyclical with respect to domestic endogenous factors but also to \ninternational ones.f The different characteristics of private debt as a growth driver described \nabove have correspondingly different implications for the growth model. Rising \nNFC debt in the context of financial sector development will boost investment \nand hence be conducive for a domestic demand-led model. Thereby productive \ncapacities are built up by which price and non-price competitiveness may rise \neventually and drive exports. While rising NFC debt for financing financial \nactivities and payouts is expected to have negative effects on demand and growth, \ndebt-financed household consumption and residential investment foster the \nemergence of a debt-led private demand growth model. If private debt expansion \nis based on capital inflows, appreciations are likely to occur. This will have \ndetrimental effects on price competitiveness and on non-price competitiveness \neventually, as explained in the respective sections. In all cases of private debt-\ndriven growth, cyclicality will prevail. The amplitude of the cycle and its trigger \nmay depend on the type and degree of international financial integration. The \nseverity of the growth downturn during the financial bust will depend on the \nexistence of alternative growth drivers. 13 Additionally to the effects on corporate balance sheets, capital flows to subordinated financially inte-\ngrated ECEs may fuel household indebtedness, as shown by Kaltenbrunner and Painceira (2018) for the \ncase of Brazil. 2.5 \u0007Private debt: financialisation, financial development and cycles While the post-GFC period saw growth \ndeceleration in all formerly private debt-driven countries, these were more \nsevere in peripheral Europe where fiscal policy was not utilized to cushion the \ndebt-driven stagnation like it was in the Anglo-Saxon countries (Kohler and \nStockhammer 2022; Hein et al. 2021). 2.6 \u0007Vertical foreign direct investment: tax havens and export platforms Vertical FDI already featured prominently in the second VoC generation. Nölke and \nVliegenthart (2009) coined the term ‘dependent market economies’ for the CEE 3 Growth drivers in emerging capitalist economies: building… 363 countries of Poland, Hungary, the Czech Republic and Slovakia. These economies are \nembedded into complex manufacturing value chains offering a skilled but relatively \ncheap labour force. The necessary capital is largely provided by multinational \ncorporations (MNCs) via FDI, making these economies dependent on decisions taken \nby hierarchically organised foreign corporations. In the growth models literature, CPE scholars invented the category of ‘FDI-led \ngrowth models’ for the CEE countries, the Baltic States and Ireland (Bohle 2018; \nBohle and Regan 2021; Regan and Brazys 2018). There, FDI-led growth models are \n“particular cases of export-oriented growth, because the major exporting firms are \nforeign-owned” (Bohle and Regan 2021, p. 82). Ownership is hence the key difference \nbetween FDI-led growth models and common export-led economies such as Germany. These FDI-led economies grow by attracting FDI that leads to the build-up of \nproductive capacities and consequently to rising external demand for domestically \nproduced goods. The main difference among FDI-led countries being then the sectors \nto which they receive their FDI (Bohle and Regan 2021). Applying a post-Keynesian \nperspective, Woodgate (20212023) questions this distinction. According to Woodgate, \nFDI-led economies not only differ in terms of ownership from export-led economies \nbut also in their national accounts. While both types have positive net exports, FDI-led \nones exhibit negative current account balances due to negative net income receipts, \nleading to GDP exceeding gross national income in these economies. Woodgate (2021, 2023) proposes two types of FDI-led growth models, tax \nhavens and export platforms, to understand the impact of FDI on demand, output, \nnational income, and growth in economies with a significant presence of MNCs. Both types are based on the ‘commercialisation of state sovereignty’ (Palan 2002), \nfeaturing low effective corporate tax rates, the establishment of special economic \nzones and investment promotion agencies to attract FDI.14 Tax havens are net \nrecipients of profits shifted within MNCs without receiving much tangible capital \nnor significant employment. Profit shifting takes place via transfer pricing or \nintellectual property royalties leading to growing net exports and output without \nactual value adding activities taking place in the tax haven. 14 Not all countries pursuing such a strategy are necessarily successful in attracting FDI. Those who are \nbenefit from some sort of first mover advantages (Woodgate 2021).\n15 Woodgate (2021, p.9) lists twelve tax havens that “have been found to be a net recipient of shifted profits \nin the literature” (Macao, British Virgin Islands, Luxembourg, Bermuda, Cayman Islands, Singapore, Ireland, \nPuerto Rico, Switzerland, Malta, the Netherlands and Hong Kong) and eight export platforms which have at \nleast 40% of their net exports coming from foreign-owned firms (China, Thailand, Malaysia, Slovakia, Estonia, \nthe Czech Republic, Hungary and Sweden). 14 Not all countries pursuing such a strategy are necessarily successful in attracting FDI. Those who a\nbenefit from some sort of first mover advantages (Woodgate 2021).i ii\n15 Woodgate (2021, p.9) lists twelve tax havens that “have been found to be a net recipient of shifted prof\nin the literature” (Macao, British Virgin Islands, Luxembourg, Bermuda, Cayman Islands, Singapore, Irelan\nPuerto Rico, Switzerland, Malta, the Netherlands and Hong Kong) and eight export platforms which have \nleast 40% of their net exports coming from foreign-owned firms (China, Thailand, Malaysia, Slovakia, Eston\nthe Czech Republic, Hungary and Sweden). 2.7 \u0007Fiscal policy In recent years, a consensus on the positive effects of expansionary fiscal policy on \nmacroeconomic performance emerged (Stockhammer et al. 2019, pp. 58–60). This \neffect is associated with the notion of fiscal multipliers which refer to the increase \nin output induced by the increase in public spending or decrease in taxation. Not \nonly is there a large amount of literature that finds fiscal multipliers to be larger \nthan one, they also seem to be higher in recessions compared to normal times and \nupswings (Gechert and Rannenberg 2018). Additionally, fiscal multipliers of public \nspending are found to be larger than those of taxation, while those of investment \ntend to exceed those of consumption. Furthermore, multipliers are larger in more \nclosed economies due to fewer demand leakages through imports (Gechert and \nRannenberg 2018). For ECEs, studies contend that fiscal multipliers are smaller \nthan for DCEs (Hory 2016; Ilzetzki et al. 2013). Moreover, fiscal consolidation may \ncome with negative hysteresis effects as the reductions in output can have long-term \nnegative effects on the economic performance while fiscal expansion could prevent \nsuch effects or even trigger positive ones (DeLong and Summers 2012; Gechert \net al. 2019). A more long-term perspective is taken in a strand of literature that introduces \nnon-capacity generating autonomous demand components into Kaleckian growth \nmodels. Allain (2015) uses autonomous government expenditure growth as such a \ndemand component in a basic Kaleckian framework with Harrodian instability. In \nthe medium (and under certain conditions, long) run, it is the growth rate of the \nautonomous demand component towards which the rate of accumulation converges. Therefore, higher growth rates of government expenditure imply higher rates of \ngrowth and accumulation. Amending the model by government debt, Hein (2018) \nand Hein and Woodgate (2021) show that constant growth rates of government \nexpenditure are under certain conditions compatible with stable public debt to GDP \nratios. Kohler and Stockhammer (2022) and Hein and Martschin (2021) have simi-\nlarly stressed the pivotal role of fiscal policy as a growth driver for DCEs.i i\nExpansionary fiscal policy will boost demand via government consumption and \ninvestment, as well as, private demand through social transfers and tax reductions. Hence, expansionary fiscal policy as a growth driver will be conducive to a domestic \ndemand-led regime. Furthermore, public expenditure can have positive supply-side \neffects. 2.6 \u0007Vertical foreign direct investment: tax havens and export platforms National income in these \neconomies grows through MNCs FDI only if part of the income is absorbed through \ntaxation of the shifted profits, which may be reinjected via government expenditure. Export platforms, on the other hand, “host foreign affiliates that are engaged in the \ngenuine production of goods and services” (Woodgate 2021, p.8) bringing tangible \ncapital and employment to the host country. Here output grows as foreign affiliates \nestablished via FDI produce to service external demand. This creates national \nincome in the form of wages paid to domestic households and taxes paid on the \nprofits of foreign affiliates. Wages and taxes in turn create demand and output as \nthey are consumed, respectively, reinjected into the economy.15 1 3 364 B. Jungmann Hence, FDI will boost net exports and be conducive to an export-led regime, \nhowever, the large presence of MNCs and net foreign income receipts may lead to \na negative current account. While positive net exports of tax havens are misleading \nas no true value adding activities take place, export platforms do service external \ndemand. The establishment of productive capacities in export platforms results in \ninvestment that may lead to productivity gains and a more diverse and sophisticated \nexport basket, resulting in increased price and non-price competitiveness. 2.7 \u0007Fiscal policy Not only may it prevent negative hysteresis effects but also promote productivity \nthrough public investment, leading to higher potential output, price competitiveness, \nand non-price competitiveness. 1 3 3 Growth drivers in emerging capitalist economies: building… 365 According to post-Keynesian and particularly Modern Monetary Theory, fiscal \nexpenditure in monetary sovereign countries is only constrained by potential \noutput (Wray 2019); government debt-to-GDP ratios can remain stable even with \nprimary deficits (Domar 1944; Hein and Martschin 2021). Hence, we contend \nthat the fiscal policy stance is foremost a policy variable that reflects political \nconflict (Kalecki 1943). However, the conditions of monetary sovereignty, such \nas issuing a domestic currency, having taxes, government expenditures and public \ndebt denominated in that currency and operating a free floating exchange rate \n(Wray 2019) do arguably not pertain to ECEs (Epstein 2019). Another possible \nconstraint on expansionary fiscal policy are balance-of-payments problems. These are especially conceivable for ECEs as the financial liabilities they incur \nassociated with current account deficits – e.g. triggered by expansionary fiscal \npolicy leading to import exceeding export demand – are denominated in foreign \ncurrency (Bonizzi et al. 2019). Table 2 summarizes the presented growth drivers, the main demand components \nthey drive, the growth model they are conducive to, further characteristics and which \ngrowth drivers are unlikely to occur simultaneously. 16 Table A2 in the supplementary material provides an overview on the periodic averages of GDP \ngrowth and current account balances for the pre- and post-GFC years. 3 \u0007Growth drivers before and after the Global Financial Crisis: \nan empirical exploration for 19 emerging capitalist economies This sections sets out an empirical exploration of growth drivers for a set of 19 \nECEs which size was constrained by data availability. First, descriptive data for \nthe growth drivers is displayed for the years between 2000 and 2019 to detect \nsalient features in their development over time and cross-regional and -country \ncharacteristics. Second, as done by Kohler and Stockhammer (2022), we check \nvia simple linear regressions for the pre- and post-GFC years for cross-country \ngrowth drivers. The trajectory of ECEs in terms of their growth models has been more \ndiverse than that of economies in the core (Akcay et al. 2022; Hein et al. 2021; \nKohler and Stockhammer 2022). To make sense of this heterogeneity but also \nfor visualization purposes, we investigate our sample of ECEs in subgroups. For \nthe time being we apply a regional grouping (Asia, Central and Eastern Europe, \nLatin America and Middle East/Africa) which, as we will see, captures the \nheterogeneity of some growth drivers better than others. As visible in Fig. 1, the \nGFC constituted a sharp decline in growth throughout the sample with varying \nmagnitude. Asian countries exhibited on average milder downturns and show the \nhighest growth levels throughout the years while the growth performance in Latin \nAmerica seems to be the most dismal. The trend of lower growth following the \nGFC across the sample, is the most pronounced in Latin America. Only Israel, \nIndia and Indonesia withstood that trend.16 Hence, although Fig. 1 displays 1 3 B. 3 \u0007Growth drivers before and after the Global Financial Crisis: \nan empirical exploration for 19 emerging capitalist economies Jungmann 368 −10\n−8\n−6\n−4\n−2\n0\n2\n4\n6\n8\n10\n12\n14\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHN\nIDN\nIND\nKOR\nMYS\nTHA\n−10\n−8\n−6\n−4\n−2\n0\n2\n4\n6\n8\n10\n12\n14\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCZE\nHUN\nPOL\nRUS\n−10\n−8\n−6\n−4\n−2\n0\n2\n4\n6\n8\n10\n12\n14\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nARG\nBRA\nCHL\nCOL\nMEX\n−10\n−8\n−6\n−4\n−2\n0\n2\n4\n6\n8\n10\n12\n14\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nISR\nSAU\nTUR\nZAF\nFig. 1 GDP growth of 19 ECEs between 2000 and 2019, regionally grouped. Sources: World Bank \n(2021), author’s calculations and presentation. Notes: Information on the empirical indicator and country \nabbreviations can be found in Table A1 Fig. 1 GDP growth of 19 ECEs between 2000 and 2019, regionally grouped. Sources: World Bank \n(2021), author’s calculations and presentation. Notes: Information on the empirical indicator and country \nabbreviations can be found in Table A1 serious heterogeneity in terms of growth deceleration and acceleration, we will \nuse the GFC in what follows as a structural breaking point for the 19 ECEs. Moreover, for several DCEs, the GFC constituted a structural breaking point in \ngrowth models: The debt-led private demand boom was followed by a debt-driven \nstagnation that was not in every country countered by sustained public demand. In consequence, GDP growth in the core slowed down in the years after the crisis \n(Kohler and Stockhammer 2022; Hein et al. 2021) constituting a drag on foreign \ndemand from the ECEs perspective (Akcay et al. 2022). That said, we do not \nrule out the possibility of alternative structural breaking points which may differ \nbetween countries and we will explicitly consider the end of commodity super \ncycle around 2014 as an alternative breaking point when assessing commodity \nprices. 3 \u0007Growth drivers before and after the Global Financial Crisis: \nan empirical exploration for 19 emerging capitalist economies Jungmann 366 Table 2 Overview of growth drivers in ECEs\nGrowth driver\nMain demand component\nGrowth model\nFurther characteristics\nUnlikely to occur with\nIncome distribution\nPrivate consumption\nDomestic demand-led\nNet effect on demand and growth \ndepends on the demand regime\nPrivate debt\nPrice competitiveness\nPrice competitiveness\nExports\nExport-led\nNegative balance sheet effects, \nredistribution towards profits\nIncome distribution\nCommodity prices\nNon-price competitiveness Exports\nExport-led\nNot necessarily as domestic demand \ndepressing as price competitive-\nness\nCommodity prices\nCommodity prices\nExports\nExport-led\nCyclical behaviour, risk of prema-\nture De-industrialization, declin-\ning terms of trade\nPrice competitiveness\nNon-price competitiveness\nPrivate debt\nConsumption and investment financial sector development: \nDomestic demand-led\ndebt-financed household consump-\ntion and residential investment: \nDebt-led private demand\nFinancial sector development is \nconducive for long-term growth; \nfinancial boom-bust cycle, which \nmay be marked by subordinated \nfinancialisation\nHousehold debt: rising wage \nshares and reduced income \ninequality\nAccompanied by capital flows: \nprice and non-price competi-\ntiveness\nVertical FDI\nExports and investment\nExport-led\nTax havens: rising exports due to \nprofit shifting of MNCs\nExport platforms: genuine value \nadding activities which may lead \nto increases in price and non-price \ncompetitiveness\nAdditional demand depends on \nincome absorption which is higher \nin export platforms Table 2 Overview of growth drivers in ECEs\nGrowth driver\nMain demand component\nGrowth model\nFurther characteristics\nUnlikely to occur with\nIncome distribution\nPrivate consumption\nDomestic demand-led\nNet effect on demand and growth \ndepends on the demand regime\nPrivate debt\nPrice competitiveness\nPrice competitiveness\nExports\nExport-led\nNegative balance sheet effects, \nredistribution towards profits\nIncome distribution\nCommodity prices\nNon-price competitiveness Exports\nExport-led\nNot necessarily as domestic demand \ndepressing as price competitive-\nness\nCommodity prices\nCommodity prices\nExports\nExport-led\nCyclical behaviour, risk of prema-\nture De-industrialization, declin-\ning terms of trade\nPrice competitiveness\nNon-price competitiveness\nPrivate debt\nConsumption and investment financial sector development: \nDomestic demand-led\ndebt-financed household consump-\ntion and residential investment: \nDebt-led private demand\nFinancial sector development is \nconducive for long-term growth; \nfinancial boom-bust cycle, which \nmay be marked by subordinated \nfinancialisation\nHousehold debt: rising wage \nshares and reduced income \ninequality\nAccompanied by capital flows: \nprice and non-price competi-\ntiveness\nVertical FDI\nExports and investment\nExport-led\nTax havens: rising exports due to \nprofit shifting of MNCs\nExport platforms: genuine value \nadding activities which may lead \nto increases in price and non-price \ncompetitiveness\nAdditional demand depends on \nincome absorption which is higher \nin export platforms Growth drivers in emerging capitalist economies: building… 367 Table 2 (continued)\nGrowth driver\nMain demand component\nGrowth model\nFurther characteristics\nUnlikely to occur with\nFiscal policy\nPublic consumption and \npublic investment\nDomestic demand-led\nTo a large extent a policy variable; \nmay also boost private demand \nthrough social transfers and tax \nreductions; potentially increasing \npotential output and measures of \ncompetiveness 1 3 B. 3.1 \u0007Income distribution: wage share and Gini coefficient of disposable income Income distribution as a growth driver is assessed through the wage share in \nnational income and the Gini coefficient of disposable income. Figure 2 a) reveals \na striking range, roughly between 20 and 60 percent of GDP, between which the \nwage share varies within the sample. Cross-country differences appear more \npronounced within than between regions. Regarding the development over time, 1 3 Growth drivers in emerging capitalist economies: building… 369 20\n30\n40\n50\n60\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHN\nIDN\nIND\nKOR\nMYS\nTHA\n20\n30\n40\n50\n60\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCZE\nHUN\nPOL\nRUS\n20\n30\n40\n50\n60\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nARG\nBRA\nCHL\nCOL\nMEX\n20\n30\n40\n50\n60\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nISR\nSAU\nTUR\nZAF\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHATUR\nZAF\n2\n4\n6\n8\n10\n−2.5\n−2.0\n−1.5\n−1.0\n−0.5\n0.0\n0.5\n1.0\n1.5\n2.0\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2000 − 2008\nslope: 0.014; p−val: 0.972; R^2: 0.00\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−0.50\n−0.25\n0.00\n0.25\n0.50\n0.75\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2009 − 2019\nslope: −0.635; p−val: 0.597, R^2: 0.02\n(b) real GDP growth in % (y−axis) and annual change in wage share in % of GDP (x−axis)\n(a) wage share in % of GDP (y−axis) between 2004 and 2019 (x−axis)\nFig. 2 Wage shares and GDP growth for 19 ECEs between 2004 and 2019. Sources: World Bank (2021) \nand ILO (2023); author’s calculations and presentation. Notes: Information on the empirical indicator \nand country abbreviations can be found in Table A1 Fig. 2 Wage shares and GDP growth for 19 ECEs between 2004 and 2019. Sources: World Bank (2021) \nand ILO (2023); author’s calculations and presentation. Notes: Information on the empirical indicator \nand country abbreviations can be found in Table A1 most ECEs do not exhibit remarkable level changes in their wage share. Notable \nis the pronounced increase of the wage share in Argentina until 2017 and the \nsteady increase witnessed in Malaysia. Looking at the bivariate coefficients in \npanel b) of Fig. 3.1 \u0007Income distribution: wage share and Gini coefficient of disposable income 2, we see no cross-country relation between changes in the wage \nshare and growth, as indicated by the flat line and insignificant coefficient. In the \npost-GFC period, the slope of the line becomes negative implying that countries \nwith falling wage shares tended to grow faster. However, the coefficient is highly \ninsignificant.fi i\nFigure 3 presents the development of the Gini coefficient for disposable \nincome, a measure for the personal distribution of income. The coefficient \ncan take values between zero and 100, with zero implying a perfectly equal \ndistribution and 100 implying that all income accrues to one person. With \nthe exception of the Middle Eastern/African subgroup, there appears to be \nsome homogeneity within the regions. The CEE countries exhibit on average \nthe lowest Gini coefficient, i.e. the most equal distribution. While the Latin \nAmerican countries started at comparatively high levels, there is a trend of \ndeclining inequality visible in all these countries until around 2015. In panel b) \nof Fig. 3, we see a positive relation between the Gini and growth for the pre-GFC \nperiod that is significant at the 5% level, implying that countries have grown \nfaster the more their income inequality increased. In the post-GFC period, the \ncoefficient stays positive without reaching the 10% significance level. As such a \nrelationship is not suggested by economic theory, we view this rather as a result \nof endogeneity or reversed causality that will not further be considered. 1 3 B. 3.1 \u0007Income distribution: wage share and Gini coefficient of disposable income Jungmann 370 30\n40\n50\n60\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHN\nIDN\nIND\nKOR\nMYS\nTHA\n30\n40\n50\n60\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCZE\nHUN\nPOL\nRUS\n30\n40\n50\n60\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nARG\nBRA\nCHL\nCOL\nMEX\n30\n40\n50\n60\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nISR\nSAU\nTUR\nZAF\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nTHATUR\nZAF\n2\n4\n6\n8\n10\n−0.50\n−0.25\n0.00\n0.25\n0.50\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2000 − 2008\nslope: 3.082; p−val: 0.02; R^2: 0.294\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−0.50\n−0.25\n0.00\n0.25\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2009 − 2019\nslope: 3.352; p−val: 0.126, R^2: 0.132\n(b) real GDP growth in % (y−axis) and annual change in Gini coeff. of disp. income (x−axis)\n(a) Gini coefficient of disposable income (y−axis) between 2000 and 2019 (x−axis)\nFig. 3 Gini coefficient of disposable income and GDP growth for 19 ECEs between 2000 and \n2019. Sources: World Bank (2021) and Solt (2020); author’s calculations and presentation. Notes: For \nthe pre-GFC simple linear regression Saudi Arabia was excluded due to insufficient data. Information on \nthe empirical indicator and country abbreviations can be found in Table A1 Fig. 3 Gini coefficient of disposable income and GDP growth for 19 ECEs between 2000 and \n2019. Sources: World Bank (2021) and Solt (2020); author’s calculations and presentation. Notes: For \nthe pre-GFC simple linear regression Saudi Arabia was excluded due to insufficient data. Information on \nthe empirical indicator and country abbreviations can be found in Table A1 3.2 \u0007Price and non‑price competitiveness Price competitiveness is proxied via the real effective exchange rate (REER), defined as \nthe price of a domestic consumption basket relative to a weighted consumption basket \nof trading partners. Following the Bank for International Settlements (BIS), an increase \n(decrease) in the REER refers to a real appreciation (depreciation) and is associated with \na decrease (increase) in international price competitiveness. Visible in Fig. 4 panel a), \nall ECEs experienced periods of real appreciations and depreciations, which differ in \ntheir extent. On average, the volatility of the REER seems less pronounced in Asia and \nthe CEE countries. Notably, the volatility in both regions decreased in the years after \nthe GFC. An exception to this is Russia, which experienced comparatively marked \nREER volatility throughout the years. Such volatile REERs can also be witnessed \nin the Latin American subgroup, in South Africa and to a lesser extent in the Middle \nEastern countries. Prior to the GFC, shown in panel b) of Fig. 4, real appreciations were \nmore prevalent and the relationship between the REER and growth was positive, i.e. appreciations and losses in price competitiveness were associated with higher growth. After the GFC, a slight majority experienced depreciations, particularly the Latin \nAmerican and CEE countries. Hence, the CEE countries with the exception of Russia, \nwere not only able to reduce the volatility of their REER but also to achieve on average \nincreases in price competitiveness throughout the post-GFC years. In Asia, all countries \nexcept Malaysia, experienced real appreciations. The positive relationship between \ngrowth and real appreciations prevailed after the GFC, becoming even more pronounced \nand significant. 3.2 \u0007Price and non‑price competitiveness 1 1 3 1 3 Growth drivers in emerging capitalist economies: building… 371 −50\n−25\n0\n25\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHN\nIDN\nIND\nKOR\nMYS\nTHA\n−50\n−25\n0\n25\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCZE\nHUN\nPOL\nRUS\n−50\n−25\n0\n25\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nARG\nBRA\nCHL\nCOL\nMEX\n−50\n−25\n0\n25\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nISR\nSAU\nTUR\nZAF\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−6\n−4\n−2\n0\n2\n4\n6\n8\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2000 − 2008\nslope: 0.104; p−val: 0.408; R^2: 0.041\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−6\n−5\n−4\n−3\n−2\n−1\n0\n1\n2\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2009 − 2019\nslope: 0.434; p−val: 0.066, R^2: 0.185\n(b) real GDP growth in % (y−axis) and REER growth rate in % (x−axis)\n(a) REER growth rate in % (y−axis) between 2000 and 2019 (x−axis)\nFig. 4 REER and GDP growth for 19 ECEs between 2000 and 2019. Sources: World Bank (2021) and \nBIS (2021b); author’s calculations and presentation. Notes: Information on the empirical indicator and \ncountry abbreviations can be found in Table A1 Fig. 4 REER and GDP growth for 19 ECEs between 2000 and 2019. Sources: World Bank (2021) and \nBIS (2021b); author’s calculations and presentation. Notes: Information on the empirical indicator and \ncountry abbreviations can be found in Table A1 Non-price competitiveness is assessed using the ECI – a higher ECI implies higher \nnon-price competitiveness. Comparing the ECI levels in panel a) of Fig. 3.2 \u0007Price and non‑price competitiveness 5, we see that \nthe CEE countries, with the exception of Russia, are found in the higher ranks with an 0.00\n0.25\n0.50\n0.75\n1.00\n1.25\n1.50\n1.75\n2.00\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHN\nIDN\nIND\nKOR\nMYS\nTHA\n0.00\n0.25\n0.50\n0.75\n1.00\n1.25\n1.50\n1.75\n2.00\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCZE\nHUN\nPOL\nRUS\n0.00\n0.25\n0.50\n0.75\n1.00\n1.25\n1.50\n1.75\n2.00\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nARG\nBRA\nCHL\nCOL\nMEX\n0.00\n0.25\n0.50\n0.75\n1.00\n1.25\n1.50\n1.75\n2.00\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nISR\nSAU\nTUR\nZAF\nARGBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−0.025\n0.000\n0.025\n0.050\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2000 − 2008\nslope: 22.22; p−val: 0.09; R^2: 0.16\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−0.025\n0.000\n0.025\n0.050\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2009 − 2019\nslope: 46.608; p−val: 0.012; R^2: 0.32\n(b) real GDP growth in % (y−axis) and annual change in ECI (x−axis)\n(a) ECI levels (y−axis) between 2000 and 2019 (x−axis)\nFig. 5 ECI and GDP growth for 19 ECEs between 2000 and 2019. Sources: World Bank (2021) and \nOEC (2021); author’s calculations and presentation. Notes: Information on the empirical indicator and \ncountry abbreviations can be found in Table A1 Fig. 5 ECI and GDP growth for 19 ECEs between 2000 and 2019. Sources: World Bank (2021) and \nOEC (2021); author’s calculations and presentation. Notes: Information on the empirical indicator and \ncountry abbreviations can be found in Table A1 3 3 372 B. Jungmann improving trend. The Asian and Latin America countries exhibit heterogeneity in terms \nof the ECI level but similar trends. Although to varying degrees, all Asian countries were \nable to increase their non-price competitiveness. In Latin America, only Mexico and to a \nlesser extent Colombia were able to do so. Notably, Saudi Arabia exhibited a significant \nincrease in its ECI in the years following the GFC, especially, as other countries marked \nby a high share of commodity exports, Brazil, Chile and Russia, experienced decreasing \nnon-price competitiveness. 3.3 \u0007Commodity export price index Figure 6 illustrates the recent commodity super cycle which started around 1999. Until the GFC, real energy and metal prices increased by over 100% and those of \nfood by 75%. After a short setback due to the GFC, commodity prices continued \ntheir upwards trajectory until the end of 2014. Traded on world markets, countries \nface the same commodity export and import prices. How commodity prices affect \na countries’ demand and growth will depend among other factors on the share of \ncommodity exports in a country’s GDP. Therefore, we will assess the role of \ncommodity prices in driving ECEs growth via a commodity export price index \n(CEPI) provided by the IMF that weighs a country’s export commodities’ prices \naccording to their share in the country’s GDP (Gruss and Kebhaj 2019) – i.e. the \nCEPI growth rate is higher for the country with a higher share of that commodity in \nits GDP. The pronounced variation in CEPI growth rates reflects the diverse degrees \nof commodity exports to GDP. Therefore, we present the CEPI growth rates with \nslight variation: The bottom right quadrant in panel a) of Fig. 7 groups those five \ncountries with significant higher CEPI growth rates, i.e. the most commodity export \nreliant countries in our sample: Chile, Indonesia, Malaysia, Russia and Saudi Arabia. In all countries, the development of the individual CEPI growth rates resembles the \ncommodity super cycle. With the structural break in this growth driver being so \napparent, we deviate from the GFC periodization for assessing the bivariate cross-\ncountry coefficient and instead choose 2000–2014 as our first period in line with \nthe commodity super cycle. Furthermore, we excluded India and China from the \nsample as it is suggested that their growth considerably drives commodity prices \n(Erten and Ocampo 2013). The bivariate coefficients are positive and insignificant in \nboth periods (panel b) Fig. 7). 3.2 \u0007Price and non‑price competitiveness Looking at panel b) which relates the annual changes in \nthe ECI to growth, we see that Asian and CEE countries exhibited more dynamism in \nimproving their ECI. In both periods, we observe a positive and significant bivariate \ncoefficient. The significance in the pre-GFC sample however hinges on China. In the \npost-GFC years the coefficient increases and becomes more significant. 3.4 \u0007Finance: corporate and household debt To assess the role of private debt as a growth driver, we look at the development \nof private sector debt, i.e. the sum of household and NFC debt, as a percentage 1 3 Growth drivers in emerging capitalist economies: building… 373 0\n1992\n1993\n1994\n1995\n1996\n1997\n1998\n1999\n2000\n2001\n2002\n2003\n2004\n2005\n2006\n2007\n2008\n2009\n2010\n2011\n2012\n2013\n2014\n2015\n2016\n2017\n2018\n2019\nAll Commodity Price Index\nAgriculture Price Index\nMetals Price Index\nEnergy Price Index\nFig. 6 Monthly commodity price indices (2016 = 100), 1992–2019. Sources: IMF (2021a); author’s presentation 0\n20\n40\n60\n80\n100\n120\n140\n160\n180\n200\n220\n240\n260\n280\n300\n320\n1992\n1993\n1994\n1995\n1996\n1997\n1998\n1999\n2000\n2001\n2002\n2003\n2004\n2005\n2006\n2007\n2008\n2009\n2010\n2011\n2012\n2013\n2014\n2015\n2016\n2017\n2018\n2019\nAll Commodity Price Index\nAgriculture Price Index\nMetals Price Index\nEnergy Price Index\nFig. 6 Monthly commodity price indices (2016 = 100), 1992–2019. Sources: IMF (2021a); author’s presentation 374 B. Jungmann −2.0\n−1.5\n−1.0\n−0.5\n0.0\n0.5\n1.0\n1.5\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHN\nIND\nKOR\nTHA\n−2.0\n−1.5\n−1.0\n−0.5\n0.0\n0.5\n1.0\n1.5\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCZE\nHUN\nISR\nPOL\nTUR\nZAF\n−2.0\n−1.5\n−1.0\n−0.5\n0.0\n0.5\n1.0\n1.5\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nARG\nBRA\nCOL\nMEX\n−20\n−15\n−10\n−5\n0\n5\n10\n15\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHL\nIDN\nMYS\nRUS\nSAU\nARG\nBRA\nCHL\nCOL\nCZE\nHUN\nIDN\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n1\n2\n3\n4\n5\n6\n0.0\n0.5\n1.0\n1.5\n2.0\n2.5\n3.0\n3.5\n4.0\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2000 − 2014\nslope: 0.327; p−val: 0.223; R^2: 0.1\nCHL\nCOL\nCZE\nHUN\nIDN\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n1\n2\n3\n4\n5\n6\n−2.0\n−1.5\n−1.0\n−0.5\n0.0\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2015 − 2019\nslope: 0.21; p−val: 0.76; R^2: 0.01\n(b) real GDP growth in % (y−axis) and cepi growth rate (x−axis)\n(a) cepi annual growth rate (y−axis) between 2000 and 2019 (x−axis)\nFig. 7 CEPI and GDP growth for 19 ECEs between 2000 and 2019. 3.4 \u0007Finance: corporate and household debt Sources: World Bank (2021) and \nIMF (2021a); author’s calculations and presentation. Notes: Information on the empirical indicator and \ncountry abbreviations can be found in Table A1 Fig. 7 CEPI and GDP growth for 19 ECEs between 2000 and 2019. Sources: World Bank (2021) and \nIMF (2021a); author’s calculations and presentation. Notes: Information on the empirical indicator and \ncountry abbreviations can be found in Table A1 of GDP. As elaborated in 2.5, this is a very broad measure that may be driven by \nvarious factors with diverse effects on growth, e.g. financialisation processes or \nfinancial development. Furthermore, the evolution of private debt is endogenous to \nthe development of growth. Hence, this examination is only a first step towards a \nmore sophisticated analysis of finance growth drivers in ECEs. Panel a) of Fig. 8 \nshows that the cyclicality varies among the countries in terms of the magnitude and \nbreaking points. In some countries the years around the GFC constitute a breaking \npoint. A case in point is China which saw a significant increase in its private debt \nlevel in the years following the GFC, rendering it the country with the highest \nprivate debt level in the sample. Despite the high degree of variety within the Asian \nregion, it has arguably the highest private debt levels. The CEE countries exhibit \na rather steady increase, with Hungary and its pronounced financial cycle, which \nreversed after the GFC, being the exception. For the Latin American countries, \nChile is that exception, exhibiting similarities to the Chinese trajectory albeit at a \nlower level. Looking at the bivariate coefficients, where we have separated NFC \ndebt from that of households, we see in panel b) that deleveraging was slightly \nmore prevalent in the pre-GFC years. After the GFC, however, a clear majority saw \nincreasing NFC debt. In the post-GFC period, the bivariate coefficient for NFC debt \nturns positive however not reaching the 10% level of significance. Panel c) relates \nthe annual change in household debt to growth. The bivariate coefficient becomes \nmore positive in the post-GFC years. While it also increases its significance, it does \nnot reach the 10% level either. of GDP. As elaborated in 2.5, this is a very broad measure that may be driven by \nvarious factors with diverse effects on growth, e.g. financialisation processes or \nfinancial development. Furthermore, the evolution of private debt is endogenous to \nthe development of growth. 3.4 \u0007Finance: corporate and household debt Hence, this examination is only a first step towards a \nmore sophisticated analysis of finance growth drivers in ECEs. Panel a) of Fig. 8 \nshows that the cyclicality varies among the countries in terms of the magnitude and \nbreaking points. In some countries the years around the GFC constitute a breaking \npoint. A case in point is China which saw a significant increase in its private debt \nlevel in the years following the GFC, rendering it the country with the highest \nprivate debt level in the sample. Despite the high degree of variety within the Asian \nregion, it has arguably the highest private debt levels. The CEE countries exhibit \na rather steady increase, with Hungary and its pronounced financial cycle, which \nreversed after the GFC, being the exception. For the Latin American countries, \nChile is that exception, exhibiting similarities to the Chinese trajectory albeit at a \nlower level. Looking at the bivariate coefficients, where we have separated NFC \ndebt from that of households, we see in panel b) that deleveraging was slightly \nmore prevalent in the pre-GFC years. After the GFC, however, a clear majority saw \nincreasing NFC debt. In the post-GFC period, the bivariate coefficient for NFC debt \nturns positive however not reaching the 10% level of significance. Panel c) relates \nthe annual change in household debt to growth. The bivariate coefficient becomes \nmore positive in the post-GFC years. While it also increases its significance, it does \nnot reach the 10% level either. 3.4 \u0007Finance: corporate and household debt Sources: World Bank (2021) \nand BIS (2021a); author’s calculations and presentation. Notes: In the pre-GFC period simple linear \nregression, South Africa was excluded due to insufficient data. Information on the empirical indicator \nand country abbreviations can be found in Table A1 3.4 \u0007Finance: corporate and household debt 1 3 Growth drivers in emerging capitalist economies: building… 375 50\n100\n150\n200\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHN\nIDN\nIND\nKOR\nMYS\nTHA\n50\n100\n150\n200\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCZE\nHUN\nPOL\nRUS\n50\n100\n150\n200\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nARG\nBRA\nCHL\nCOL\nMEX\n50\n100\n150\n200\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nISR\nSAU\nTUR\nZAF\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\n2\n4\n6\n8\n10\n−5\n−4\n−3\n−2\n−1\n0\n1\n2\n3\n4\n5\n6\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2000 − 2008\nslope: −0.155; p−val: 0.302; R^2: 0.07\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−3\n−2\n−1\n0\n1\n2\n3\n4\n5\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2009 − 2019\nslope: 0.386; p−val: 0.183; R^2: 0.1\n(b) real GDP growth in % (y−axis) and annual change in NFC debt to GDP in % (x−axis)\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\n2\n4\n6\n8\n10\n−2\n−1\n0\n1\n2\n3\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2000 − 2008\nslope: 0.362; p−val: 0.219; R^2: 0.093\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−1\n0\n1\n2\n3\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2009 − 2019\nslope: 0.613; p−val: 0.141; R^2: 0.12\n(c) real GDP growth in % (y−axis) and annual change in household debt to GDP in % (x−axis)\n(a) Private sector debt to GDP in % (y−axis) between 2000 and 2019 (x−axis)\nFig. 8 Private debt and GDP growth for 19 ECEs between 2000 and 2019. Sources: World Bank (2021) \nand BIS (2021a); author’s calculations and presentation. Notes: In the pre-GFC period simple linear \nregression, South Africa was excluded due to insufficient data. Information on the empirical indicator \nand country abbreviations can be found in Table A1 Fig. 8 Private debt and GDP growth for 19 ECEs between 2000 and 2019. 3.5 \u0007Foreign direct investment Looking at the development of the FDI stock since 2000 in panel a) of Fig. 9, we see \na rising trend with only few exceptions. The Asian countries are marked by com-\nparatively low to medium/high levels of FDI stock that has developed rather steady. The Middle Eastern and African countries are similar in terms of the level and trend \nhowever in a less stable way. Among the CEE countries, we find with the Czech \nRepublic but especially Hungary two countries with the highest level of FDI stock. Notably, the high level of FDI stock in Hungary was largely build up between 2005 \nand 2009 and plateaued more or less after the GFC. For the Latin American coun-\ntries we see a similar pattern as with private debt as Chile exhibits the largest stock \nof FDI, which growth significantly accelerated after the GFC. In all Latin American \ncountries with the exception of Argentina, FDI shows a continuous increasing trend. The bivariate coefficients in panel b) relating the FDI inflows to growth are negative \nand insignificant in both periods. 3.6 \u0007Fiscal policy: government structural balance The assessment of fiscal policy as a growth driver is associated with some issues. Important parameters of fiscal policy like government expenditures, tax revenue \nand the resulting government balance are highly endogenous with respect to growth 1 3 B. Jungmann 376 25\n50\n75\n100\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHN\nIDN\nIND\nKOR\nMYS\nTHA\n25\n50\n75\n100\n50\n100\n150\n200\n250\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCZE\nHUN (RHS)\nPOL\nRUS\n25\n50\n75\n100\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nARG\nBRA\nCHL\nCOL\nMEX\n25\n50\n75\n100\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nISR\nSAU\nTUR\nZAF\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n0\n1\n2\n3\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2000 − 2008\nslope: −0.39; p−val: 0.417; R^2: 0.04\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nSAU\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n0\n1\n2\n3\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2009 − 2019\nslope: −0.494; p−val: 0.232; R^2: 0.08\n(b) real GDP growth in % (y−axis) and net FDI inflows in % of GDP (x−axis)\n(a) FDI stock per GDP in % (y−axis) between 2000 and 2019 (x−axis)\nFig. 9 FDI and GDP growth for 19 ECEs between 2000 and 2019. Sources: World Bank (2021) and \nLane and Milesi-Ferretti (2018); author’s calculations and presentation. Notes: In the pre-GFC period \nsimple linear regression, Hungary was excluded as an extreme outlier. Information on the empirical indi-\ncator and country abbreviations can be found in Table A1 Fig. 9 FDI and GDP growth for 19 ECEs between 2000 and 2019. Sources: World Bank (2021) and \nLane and Milesi-Ferretti (2018); author’s calculations and presentation. Notes: In the pre-GFC period \nsimple linear regression, Hungary was excluded as an extreme outlier. Information on the empirical indi-\ncator and country abbreviations can be found in Table A1 and economic activity. To avoid this problem, the structural government balance is \nemployed which tries to filter out the cyclical and endogenous elements of the gov-\nernment balance. 17 The determination of the cyclical and structural elements of the government balance involves various \nestimations which have become a centre of discussion. Current estimation methods have been accused to \nunderestimate the potential output and output gap. This would then label fiscal policy more expansionary \nthan it actually is (see e.g. Heimberger and Kapeller 2017). By choosing the annual change in the struc-\ntural government balance we try to avoid at least the estimation issues in terms of the level (Hein and \nMartschin 2021). 3.6 \u0007Fiscal policy: government structural balance The more negative (positive) the structural government balance, \nthe more expansionary (contractionary) the fiscal policy stance.17 The four quad-\nrants in panel a) of Fig. 10 suggest that the GFC constitute some form of break-\ning point in terms of the fiscal policy stance. For the Asian countries this is not as \napparent as the levels and the trends are rather heterogeneous. The CEE countries \nexhibit with some variation a trend of improving structural government balances, \ni.e. a trend towards more contractionary fiscal policy. In the Latin American coun-\ntries, the GFC separates two distinctive periods, the former one exhibiting higher \nstructural balances and improving tendencies and the latter period a lower level and \ndownward trend – this development is most pronounced in Brazil and Argentina. For \nthe Middle East and African countries, the GFC was followed by lower structural \ngovernment balances. Notably, 2015 constituted a breaking point for Israel and Tur-\nkey after which the structural government balances fell considerably. 1 3 Growth drivers in emerging capitalist economies: building… 377 −10\n−8\n−6\n−4\n−2\n0\n2\n4\n6\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCHN\nIDN\nIND\nKOR\nMYS\nTHA\n−10\n−8\n−6\n−4\n−2\n0\n2\n4\n6\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nCZE\nHUN\nPOL\nRUS\n−10\n−8\n−6\n−4\n−2\n0\n2\n4\n6\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nARG\nBRA\nCHL\nCOL\nMEX\n−10\n−8\n−6\n−4\n−2\n0\n2\n4\n6\n2000\n2002\n2004\n2006\n2008\n2010\n2012\n2014\n2016\n2018\nISR\nTUR\nZAF\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−0.50\n−0.25\n0.00\n0.25\n0.50\n0.75\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2000 − 2008\nslope: 1.1; p−val: 0.362; R^2: 0.05\nARG\nBRA\nCHL\nCHN\nCOL\nCZE\nHUN\nIDN\nIND\nISR\nKOR\nMEX\nMYS\nPOL\nRUS\nTHA\nTUR\nZAF\n2\n4\n6\n8\n10\n−0.5\n−0.4\n−0.3\n−0.2\n−0.1\n0.0\n0.1\n0.2\n0.3\n0.4\nAfrica\nAsia\nCEE\nLatin America\nMiddle East\n2009 − 2019\nslope: −1.141; p−val: 0.59; R^2: 0.02\n(b) real GDP growth in % (y−axis) and government structural balance annual change in % of GDP (x−axis)\n(a) Government structural balance per GDP in % (y−axis) between 2000 and 2019 (x−axis)\nFig. 3.6 \u0007Fiscal policy: government structural balance 10 Government structural balance and GDP growth for 19 ECEs between 2000 and 2019. Sources: \nWorld Bank (2021) and IMF (2021b); author’s calculations and presentation. Notes: Information on the \nempirical indicator and country abbreviations can be found in Table A1 Fig. 10 Government structural balance and GDP growth for 19 ECEs between 2000 and 2019. Sources: \nWorld Bank (2021) and IMF (2021b); author’s calculations and presentation. Notes: Information on the \nempirical indicator and country abbreviations can be found in Table A1 The simple linear regressions in panel b) of Fig. 10 emphasize the descriptive \nfinding that fiscal policy became overall more expansionary after the GFC. Relating \nthe fiscal policy stance to growth, in the pre-GFC period, higher structural balances, \ni.e. more contractionary fiscal policy, were associated with higher growth. This rela-\ntionship reversed after the GFC. However, both cases lack significance. 4 \u0007ECEs growth drivers: heterogeneity, possible country clusters \nand their relation to international growth models Before further assessing our results, some caveats regarding the methodology, in \nparticular, the bivariate coefficients should be recalled. First, we do establish coef-\nficients that, at best, signify correlation. Correlations are not causalities, however, \nfor causalities to exist there have to be correlations. Furthermore, we derive cross-\ncountry coefficients. The absence of a cross-country correlation does not necessarily \nindicate that this indicator was irrelevant for each country. Instead, it may suggest \nheterogeneity among countries – e.g. in terms of financialisation (Karwowski 2020) \nor growth model (Akcay et al. 2022). Moreover, the very nature of some growth \ndrivers prohibits that they drive growth on a cross-country level. Commodity prices \nfor example rise in a secular way, their net effect on growth however does not only \ndepend on the importance of commodity exports for a country’s overall demand but \nalso on the negative demand effects that higher commodity prices might have. Simi-\nlarly, price competitiveness and the wage share can hardly become cross-country \ndrivers as the net effect of increased price competiveness, respectively, higher profit 1 3 3 3 378 B. Jungmann shares varies among countries (Table 1). A further problem concerning the estima-\ntion of growth drivers is that of endogeneity as GDP growth affects itself various \ngrowth drivers. Going forward, this should be addressed by choosing variables that \nare more exogenous to demand and growth or by employing more sophisticated \nmethods, e.g. via lagged or instrument variables. Albeit these problems, some conclusions can be drawn from our empirical explo-\nration. Our results indicate that non-price competitiveness drove growth across \nECEs, particularly in the post-GFC period (Fig. 5). This is in line with the liter-\nature on economic complexity (Hidalgo and Hausmann 2009), Thirlwall’s (1979) \nlaw, and economic structuralism (Ocampo and Parra 2006) and resembles the results \nfor DCEs (Gräbner et al. 2020; Hein and Martschin 2021; Kohler and Stockham-\nmer 2022). Hence, our results underscore the importance, as highlighted by Kohler \nand Stockhammer (2022), of expanding the notion of competitiveness beyond \nprice factors in the Comparative and International Political Economy literature, \neven for ECEs. Especially, as we likewise fail to find a cross-country role of price \ncompetitiveness.18ii Although lacking significance, the increased importance of expansionary fiscal \npolicy and private debt as growth drivers for the post-GFC years is noteworthy. This is accompanied by overall increased private debt levels and a more \nexpansionary fiscal policy stance (Fig. 8 and 10). 18 In contrast, we find a positive relation between real appreciations and growth. This may simply be \ndue to reversed causality, i.e. growth leading to real appreciations. On the other hand, it may signal the \nnegative effects associated with real depreciations, especially, in the light of progressed financialisation \nand foreign currency denominated debt (McCauley et al. 2015). Moreover, import dependencies may \nnarrow the possible gains from depreciations. 4 \u0007ECEs growth drivers: heterogeneity, possible country clusters \nand their relation to international growth models Notes: OECD member states \nas of June 2021 excluding the member countries that are part of our sample. Information on the empirical \nindicator and country abbreviations can be found in Table A1 least some of the heterogeneity in the descriptive data. The Asian group is marked \nby comparatively high growth rates, more stable REERs, higher and rising \nECI levels and high private debt levels. The Latin American countries exhibit \ncomparatively low growth rates, high levels of income inequality but with a \ndownward trend, rather unstable REERs and a relatively expansionary fiscal policy \nstance, especially after the GFC. The heterogeneity among the Middle Eastern \nand African countries are too pronounced. The CEE countries, often with the \nexception of Russia, exhibit medium to high growth rates and comparatively low \nlevels of income inequality. Moreover, their REERs are rather stable, particularly, \nafter the GFC when they achieved overall depreciations while their fiscal policy \nstance became less expansionary at the same time. Finally, their ECI levels and \nFDI stock is rather high. least some of the heterogeneity in the descriptive data. The Asian group is marked \nby comparatively high growth rates, more stable REERs, higher and rising \nECI levels and high private debt levels. The Latin American countries exhibit \ncomparatively low growth rates, high levels of income inequality but with a \ndownward trend, rather unstable REERs and a relatively expansionary fiscal policy \nstance, especially after the GFC. The heterogeneity among the Middle Eastern \nand African countries are too pronounced. The CEE countries, often with the \nexception of Russia, exhibit medium to high growth rates and comparatively low \nlevels of income inequality. Moreover, their REERs are rather stable, particularly, \nafter the GFC when they achieved overall depreciations while their fiscal policy \nstance became less expansionary at the same time. Finally, their ECI levels and \nFDI stock is rather high. It is noteworthy that, apart from Russia, the CEE countries not only share \nsimilarities in the descriptive statistics of growth drivers, but also resemble each \nother in the development of their growth models, following the path of many other \nEuropean DCEs to become or remain export-led after the GFC (Dodig et al. 2016; \nHein et al. 2021). At the same time, these countries were already subsumed in the \n‘dependent market economies’ category within the second VoC generation (Nölke \nand Vliegenthart 2009). 4 \u0007ECEs growth drivers: heterogeneity, possible country clusters \nand their relation to international growth models The indications of private debt \nand expansionary fiscal policy becoming more relevant in driving ECEs’ growth \nfollowing the GFC can be viewed as a consequence of developments in DCEs. ECEs are to a considerable extent subject to DCEs and their growth models, \nespecially, in terms of external demand that determines the feasibility of export-\nled growth models in ECEs. With DCEs reducing their imports and developing \nincreasingly export-led growth models due to austerity policies and private \ndeleveraging (Hein et al. 2021; Kohler and Stockhammer 2022) the increased \nimportance of private debt and fiscal policy constitutes a form of compensation \nfor this loss in external demand. In consequence, growth models in ECEs became \nrather domestic demand or even private debt-led (Akcay et al. 2022) accompanied \nby a rise in importance of the respective growth drivers (Table 2). This \ndevelopment is evident in the current accounts, with ECEs showing a tendency of \nworsening and DCEs one of improvement, as it is shown in Fig. 11. Not only did \nthe aggregated current account deficit of DCEs shrink after the GFC, it became \nroughly balanced since 2013 and turned into a surplus in 2016. Despite these tendencies, our results for robust cross-country growth drivers \nremain sparse. As mentioned, this can be attributed in part to the heterogeneity \namong ECEs, for example, in terms of their growth models and their respective \ndrivers (Table 2). We have seen that our regional grouping was able to sort at 1 3 Growth drivers in emerging capitalist economies: building… 379 -1000000\n-800000\n-600000\n-400000\n-200000\n0\n200000\n400000\n600000\n800000\n1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018\nARG\nBRA\nCHL\nCOL\nCHN\nCZE\nHUN\nIDN\nISR\nKOR\nMYS\nMEX\nPOL\nRUS\nSAU\nZAF\nTHA\nTUR\nIND\nOECD\nFig. 11 Current account balances in current million US$ of 19 ECEs and OECD member states, 1980–\n2019. Sources: World Bank (2021); author’s calculations and presentation. Notes: OECD member states \nas of June 2021 excluding the member countries that are part of our sample. Information on the empirical \nindicator and country abbreviations can be found in Table A1 Fig. 11 Current account balances in current million US$ of 19 ECEs and OECD member states, 1980–\n2019. Sources: World Bank (2021); author’s calculations and presentation. 5 \u0007Conclusions In this paper, we expanded the concept of growth drivers to ECEs. On a conceptual \nlevel, we made the case for a synthesis of growth drivers with a growth model \noperationalization based on GDP growth contributions and financial sector balances. Building on post-Keynesian and structuralist economics as well as empirical studies \non ECEs, seven growth drivers for ECEs were reviewed: income distribution, \nprice and non-price competitiveness, commodity prices, private debt, FDI and \nfiscal policy. In an empirical exploration we have presented descriptive data for \nseven growth drivers of 19 ECEs between 2000 and 2019. We find that the Asian \nECEs exhibit on average higher growth rates, stable REERs, higher and rising \nnon-price competitiveness and high private debt levels. In Latin America, we find \ncomparatively low growth rates, high levels of income inequality that are abating, \nunstable RERs and relatively expansionary fiscal stances after the GFC. The CEE \ncountries, often with the exception of Russia, exhibit medium to high growth rates, \ncomparatively low levels of income inequality, high non-price competiveness and a \nhigh FDI stock. Moreover, their RERs were rather stable, particularly, after the GFC \nwhen they achieved overall depreciations while their fiscal policy stance became \nless expansionary. Furthermore, we assessed which growth drivers determined growth on a cross-\ncountry level. To that end, we established bivariate coefficients. In that regard robust \nresults remained sparse. The only robust and significant exception concerns non-\nprice competitiveness which we found to be a cross-country growth driver for ECEs. Furthermore, we find indications of private debt and expansionary fiscal policy \nbecoming more important in driving growth in ECEs after the GFC. This increase in \nimportance is in line with the emergence and continuation of domestic demand- and \nprivate debt-led growth models in ECEs following the GFC (Akcay et al. 2022). In \nthis context, the increased importance of private debt and expansionary fiscal policy \nin driving ECEs’ growth can be viewed as a form of compensation for reduced \nexternal demand from DCEs in the course of private deleveraging and austerity \npolicies after the GFC (Hein et al. 2021; Kohler and Stockhammer 2022). Several issues remain regarding the empirical investigation of growth drivers, \nnamely, the problem of endogeneity with respect to growth and the heterogeneity \namong ECEs. 4 \u0007ECEs growth drivers: heterogeneity, possible country clusters \nand their relation to international growth models This suggest that growth drivers will not only be similar \nwithin the same growth models but should also be studied through the critically \nretention of institutionalist aspects from earlier VoC contributions as proposed by \nSchedelik et al. (2021) and implied by Stockhammer and Kohler (2022). 1 3 B. Jungmann 380 Supplementary Information The online version contains supplementary material available at https://doi.\norg/10.1007/s43253-023-00101-1. Declarations Conflict of interest The author reports no conflict of interest. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, \nwhich permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as \nyou give appropriate credit to the original author(s) and the source, provide a link to the Creative Com-\nmons licence, and indicate if changes were made. The images or other third party material in this article \nare included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the \nmaterial. If material is not included in the article’s Creative Commons licence and your intended use is \nnot permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission \ndirectly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licen\nses/by/4.0/. 5 \u0007Conclusions Going forward, building on our results, growth drivers should \nbe examined through more sophisticated methods dealing with endogeneity \nissues while the problem of heterogeneity could be addressed through adequate \ngrouping, e.g. along the lines of growth models (Akcay et al. 2022) or institutional \nconfigurations (Schedelik et al. 2021). Finally, while we have proposed that growth \ndriver analysis can inform growth model analysis using common growth model \ncategories, one may extend this synthesis to further categories and methodologies \n– for instance, the investment-led growth model (Mertens et al. 2022) and the \ngrowth decomposition based on the Sraffian supermultiplier (Campana et al. 2024; \nMorlin et al. 2022; Passos and Morlin 2022). Supplementary Information The online version contains supplementary material available at https://doi.\norg/10.1007/s43253-023-00101-1. 3 1 Growth drivers in emerging capitalist economies: building… 381 Acknowledgements This paper is a substantially revised version of my master thesis submitted to the \nBerlin School of Economics and Law and the University Paris Sorbonne Nord. I am most grateful for the \ncomments, help and advice I received from Eckhard Hein, Ümit Akcay and Jonathan Marie. Furthermore, \nI want to thank three anonymous referees and Engelbert Stockhammer for their comments, which helped \nto improve the paper substantially. All errors are mine. Funding Open Access funding enabled and organized by Projekt DEAL. Funding Open Access funding enabled and organized by Projekt DEAL. Data availability All data used in this study is derived from the public domain resources described in th\ntext and reference list and are also available from the author upon request. References Ahmed S, Zlate A (2014) Capital flows to emerging market economies: a brave new world? J Int Money \nFinanc 48(Part B):221–248\nÜ Akcay Ü, Hein E, Jungmann B (2022) Financialisation and macroeconomic regimes in emerging capital-\nist countries before and after the great recession. Int J Polit Econ 51(2):77–100. https://doi.org/10.\n1080/08911916.2022.2078009 Alarco G (2016) Distribución factorial del ingreso y regímenes de crecimiento en América Latina, 1950–\n2012. Rev Int Trab 135(1):79–103 Allain O (2015) Tackling the instability of growth: a Kaleckian-Harrodian model with an autonomous \nexpenditure component. Camb J Econ 39(5):1351–1371 Allain O (2015) Tackling the instability of growth: a Kaleckian-Ha\nexpenditure component. Camb J Econ 39(5):1351–1371 p\np\n( )\nAraújo E, Gala P (2012) Economic growth regimes in Brazil: empirical evidence and policy implications. Estudos Avançados 26(75):41–56i Estudos Avançados 26(75):41–56 Estudos Avançados 26(75):41–56 Bingley, vol 31, pp 109–134 Chang KS (2016) Financialization of poverty: Proletarianizing the Financial Crisis in post developmental \nKorea. In: Soederberg S (ed) Risking capitalism. Emerald Publishing. Bingley, vol 31, pp 109–134 Charpe M, Lee S, Arias D, Bridji S (2014) Does income distribution matter for development? Trends \nin labour share of income and their economic impacts in developing countries. World Work Rep \n2014(1):149–180 Davis LE (2017) Financialization and investment: a survey of the empirical literature. J Econ Su\n31(5):1332–1358. https://doi.org/10.1111/joes.12242 de Jesus CS, Araujo RA, Drumond CE (2018) An empirical test of the Post-Keynesian growth model \napplied to functional income distribution and the growth regime in Brazil. Int Rev Appl Econ \n32(4):428–449 DeLong JB, Summers LH (2012) Fiscal policy in a depressed economy. Brook Papers Econ Activity \nSpring 2012:233–297i Dodig N, Hein E, Detzer D (2016) Financialisation and the financial and economic crises: Theoretical \nframework and empirical analysis for 15 countries. In: Hein E, Detzer D, Dodig D (eds) Financiali-\nsation and the financial and economic crises. Edward Elgar, Cheltenham, pp 1–41 sation and the financial and economic crises. Edward Elgar, Cheltenham, pp 1–41 i\nDomar ED (1944) The “burden of the debt\" and the national income. Am Econ Rev 34(4):798– Dos Santos PL (2013) A cause for policy concern: the expansion of household credit in middle-income \neconomies. Int Rev Appl Econ 27(3):316–338 Dünhaupt P, Hein E (2019) Financialisation, distribution, and macroeconomic regimes before and aft\nthe crisis: a post-Keynesian view on Denmark, Estonia, and Latvia. J Baltic Stud 50(4):435–465 Epstein G (2019) Is MMT ‘America First’ Economics? INET blog. https://www.ineteconomics.org/per\nectives/blog/is-mmt-america-first-economics. Accessed 11 May 2023 i\nErten B, Ocampo JA (2013) Super cycles of commodity prices since the mid-nineteenth century. World \nDev 44:14–30 Fernández A, Schmitt-Grohé S, Uribe M (2020) Does the commodity super cycle matter? Working Paper \n27589. National Bureau of Economic Research, Cambridge, MAi Fernandez R, Aalbers MB (2020) Housing financialization in the Global South: in search for a compara-\ntive framework. Hous Policy Debate 30(4):680–701 Gabor D (2010) (De)Financialization and crisis in Eastern Europe. Compet Chang 14(3–4):248–270\nG l\nP R\nh\nI M\nh\nG (2018) Th\nli\ni\nl\ni\ni Gabor D (2010) (De)Financialization and crisis in Eastern Europe. Compet Chang 14(3–4):248–270\nGala P, Rocha I, Magacho G (2018) The structuralist revenge: economic complexity as an important \ndimension to evaluate growth and development. Estudos Avançados 26(75):41–56 Arestis P (2006) Financial liberalization and the relationship between finance and growth. In: Arestis P, \nSawyer M (eds) A handbook of alternative monetary economics. Edward Elgar, Cheltenham, pp \n246–364 Arestis P, Glickman M (2002) Financial crisis in Southeast Asia: dispelling illusion the Minskyan way. Camb J Econ 26(2):237–260i Ashman S, Fine B, Newman S (2011) The crisis in South Africa: neoliberalism, financialization and \nuneven and combined development. Socialist Register 47:174–195 Baccaro L, Blyth M, Pontusson J (eds) (2022) Diminishing Returns: The new Politics of Growth and \nStagnation. Oxford University Press, Oxford Baccaro L, Pontusson J (2016) Rethinking comparative political economy: the growth model perspective. Polit Soc 44(2):175–207 Bhaduri A, Marglin S (1990) Unemployment and the real wage: the economic basis for contesting polit\ncal ideologies. Camb J Econ 14(4):375–393 Bird M (2020) Chinese household debt surges through the pandemic. Retrieved 20 July 2021, from The \nWall Street Journal website: https://www.wsj.com/articles/chinese-household-debt-surges-through-\nthe-pandemic-11607417329 1 3 382 B. Jungmann BIS (2021b) Effective exchange rate indices. https://www.bis.org/statistics/eer.htm. Accessed 10 May \n2021 Bohle D (2018) European integration capitalist diversity and crises trajectories on Europe’s Eastern \nperiphery. New Political Economy 23(2):239–253. https://doi.org/10.1080/13563467.2017.13704\n48 Bohle D, Regan A (2021) The comparative political economy of growth models: Explaining the contin\nity of FDI-led growth in Ireland and Hungary. Polit Soc 49(1):75–106. Bonizzi B (2013) Financialization in developing and emerging countries: a survey. Int J Polit Eco\n42(4):83–107l Bonizzi B (2017) An alternative Post-Keynesian framework for understanding capital flows to emerging \nmarkets. J Econ Issues 51(1):137–162 Bonizzi B, Kaltenbrunner A, Michell J (2019) Monetary sovereignty is a spectrum: modern monetary \ntheory and developing countries. Real-World Econ Rev 89:46–61i Bortz PG, Kaltenbrunner A (2018) The international dimension of financialization in developing an\nemerging economies. Dev Chang 49(2):375–393 Bräuning F, Ivashina V (2020) U.S. monetary policy and emerging market credit cycles. J Monet Econ \n112:57–76 Bresser-Pereira LC (2008) The Dutch disease and its neutralization: a Ricardian approach. Braz J Polit \nEcon 28(1):47–71 Campana J, Embova Vaz J, Hein E, Jungmann B (2024) Demand and growth regimes of the BRICs coun-\ntries – the national income and financial accounting decomposition approach and an autonomous \ndemand-led growth perspective. Eur J Econ Econ Policies: Interv. forthcoming Chang KS (2016) Financialization of poverty: Proletarianizing the Financial Crisis in post-developmental \nKorea. In: Soederberg S (ed) Risking capitalism. Emerald Publishing. Estudos Avançados 26(75):41–56 Braz J Polit Econ 38(2):219–236fi Gala P, Rocha I, Magacho G (2018) The structuralist revenge: economic complexity as an impor\ndimension to evaluate growth and development. Braz J Polit Econ 38(2):219–236fi Gechert S, Horn G, Paetz C (2019) Long-term effects of fiscal stimulus and austerity in Europe. Oxford \nBull Econ Stat 81(3):647–666i Gechert S, Rannenberg A (2018) Which fiscal multipliers are regime-dependent? A meta-regression anal-\nysis. J Econ Surv 32(4):1160–1182 1 3 Growth drivers in emerging capitalist economies: building… 383 Gouvêa RR, Lima GT (2010) Structural change, balance-of-payments constraint, and economic growth: \nevidence from the multisectoral Thirlwall’s law. J Post Keynes Econ 33(1):169–204 Gräbner C, Heimberger P, Kapeller J, Schütz B (2020) Is the Eurozone disintegrating? Macroeconomic \ndivergence, structural polarisation, trade and fragility. Camb J Econ 44(3):647–669 Gruss B, Kebhaj S (2019) Commodity terms of trade a new database. Working Paper 19/21. IMF. https://\nwww.imf.org/en/Publications/WP/Issues/2019/01/24/Commodity-Terms-of-Trade-A-New-Datab\nase-46522. Accessed 24 June 2021 Heimberger P, Kapeller J (2017) The performativity of potential output: pro-cyclicality and path depen\nency in coordinating European fiscal policies. Rev Int Polit Econ 24(5):904–928 i\nHall PA (2018) Varieties of Capitalism in light of the euro crisis. J Eur Publ Policy 25(1):7–30 Hassel A, Palier B (2021) Growth and Welfare in Advanced Capitalist Economies: How H\nRegimes Evolved? Oxford University Press Oxford Hassel A, Palier B (2021) Growth and Welfare in Advanced Capitalist Economies: How Have Growth \nRegimes Evolved? Oxford University Press, Oxfordi Regimes Evolved? Oxford University Press, Oxfordi Hein E (2011) Redistribution, global imbalances and the financial and economic crisis – the case for a \nKeynesian New Deal. Int J Labour Res 3(1):51–73 Hein E (2011) Redistribution, global imbalances and the financial and economic crisis – the case for Hein E (2012) The Macroeconomics of Finance-dominated Capitalism – and Its Crisis. Edward Elg\nPublishing, Cheltenham Hein E (2014) Distribution and growth after Keynes: A Post-Keynesian guide. Edward Elgar, Cheltenhami (\n)\ng\ny\ny\ng\ng ,\nHein E (2018) Autonomous government expenditure growth, deficits, debt, and distribution in a neo-\nKaleckian growth model. J Post Keynes Econ 41(2):316–338 Hein E (2019) Financialisation and tendencies toward stagnation: the role of macroeconomic regime \nchanges in the course of and after the financial and economic crisis 2007–9. Camb J Econ \n43(4):975–999. https://doi.org/10.1093/cje/bez022 Hein E, Martschin J (2020) The Eurozone in crisis – a Kaleckian macroeconomic regime and policy pe\nspective. Estudos Avançados 26(75):41–56 Rev Polit Econ 32(4):563–588i Hein E, Martschin J (2021) Demand and growth regimes in finance-dominated capitalism and the role of \nthe macroeconomic policy regime: a post-Keynesian comparative study on France, Germany, Italy \nand Spain before and after the Great Financial Crisis and the Great Recession. Rev Evol Polit Econ \n2:493–527 Hein E, Meloni WP, Tridico P (2021) Welfare models and demand-led growth regimes before and aft\nthe financial and economic crisis. Rev Int Polit Econ 28(5):1196–1223 i\nHein E, Mundt M (2012) Financialisation and the requirements and potentials for wage-led recovery: a \nreview focussing on the G20. ILO, Geneva g\nHein E, Prante FJ (2020) Functional distribution and wage inequality in recent Kaleckian growth models. In: Bougrine H, Rochon LP (eds) Economic Growth and Macroeconomic Stabilization Policies in \nPost-Keynesian Economics. Edward Elgar, Cheltenham, pp 33–49 Hein E, Prante FJ (2020) Functional distribution and wage inequality in recent Kaleckian growth models. In: Bougrine H Rochon LP (eds) Economic Growth and Macroeconomic Stabilization Policies in Hein E, Woodgate R (2021) Stability issues in Kaleckian models driven by autonomous deman\ngrowth – Harrodian instability and debt dynamics. Metroeconomica 72(2):388–404 g\ny\ny\nHidalgo CA, Hausmann R (2009) The building blocks of economic complexity. Proc Natl Acad Sci \n106(26):10570–10575 Hory MP (2016) Fiscal multipliers in emerging market economies: can we learn something from \nadvanced economies? International Economics 146:59–84 ILO (2023) Labour Income Shares as a percent of GDP. ILOSTAT database. https://ilostat.ilo.org/dat\nAccessed 16 Jan 2023 IMF (2021a) Primary Commodity Prices. https://www.imf.org/en/Research/commodity-prices. Accessed \n30 April 2021 p\nMF (2021b) World Economic Outlook. https://www.imf.org/external/datamapper/GGXCNL_NGDP@ IMF (2021b) World Economic Outlook. https://www.imf.org/external/da\nWEO/OEMDC/ADVEC/WEOWORLD. Accessed 30 April 2021i IMF (2021b) World Economic Outlook. https://www.imf.org/external/datamapper/GGXCNL_NGDP@\nWEO/OEMDC/ADVEC/WEOWORLD. Accessed 30 April 2021i IMF (2021b) World Economic Outlook. https://www.imf.org/external/datamapper/GGXCNL_NGDP@\nWEO/OEMDC/ADVEC/WEOWORLD. Accessed 30 April 2021i (\n)\np\ng\nWEO/OEMDC/ADVEC/WEOWORLD. Accessed 30 April 2021i Ilzetzki E, Mendoza EG, Végh CA (2013) How big (small?) are fiscal multipliers? J Monet Eco\n60(2):239–254 tin B, Kurt OE (2016) Functional income distribution and growth in Thailand: a post Keynesian econo-\nmetric analysis. J Post Keynes Econ 39(3):334–360 Jetin B, Reyes Ortiz L (2020) Wage-led demand as a rebalancing strategy for economic growth in China. Estudos Avançados 26(75):41–56 J Post Keynes Econ 43(3):341–366 J Post Keynes Econ 43(3):341–366 y\nJiménez V (2020) Wage shares and demand regimes in Central America: an empirical analysis 1970–\n2016 Working Paper 151 Institute for International Political Economy (IPE) Berlin ménez V (2020) Wage shares and demand regimes in Central America: an empirical analysis 1970– ménez V (2020) Wage shares and demand regimes in Central America: an empirical analysis 1970–\n2016. Working Paper 151. Institute for International Political Economy (IPE), Berlin Johnston A, Regan A (2018) Introduction: is the European Union capable of integrating diverse models \nof Capitalism? New Polit Econ 23(2):145–159 1 3 384 B. Jungmann Joo S, Lee K, Nah WJ, Jeon SM, Joe DH (2020) The income-led growth in Korea: status, prospects and \nlessons for other countries. Research Paper Policy Analysis 20–01. Korea Institute for International \nEconomic Policy, Sejong-si Kalecki M (1943) Political aspects of full employment. Polit Q 14(4):322–330ii Kaltenbrunner A, Painceira JP (2018) Subordinated financial integration and financialisation in emerging Kaltenbrunner A, Painceira JP (2018) Subordinated financial integration and financialis\ni li\ni\nh\nili\ni\nli\n23(3) 290 313 Kaltenbrunner A, Painceira JP (2018) Subordinated financial integration and financial\ncapitalist economies: the Brazilian experience. New Polit Econ 23(3):290–313 (\n)i\ngi\ncapitalist economies: the Brazilian experience. New Polit Econ 23(3):290–313 p\np\nKapeller J, Schütz B (2015) Conspicuous consumption, inequality and debt: the nature of consumption \ndriven profit-led regimes. Metroeconomica 66(1):51–70 i\nKaracimen E (2016) Consumer credit as an aspect of everyday life of workers in developing countrie\nevidence from Turkey. Rev Radic Polit Econ 48(2):252–268 Karwowski E (2012) Financial Operations of South African Listed Firms: Growth and Financial Stability \nin an Emerging Market Setting. Paper presented at iii conferencia international do ieSe, Maputo, \nMozambique, 3–4 September. https://www.iese.ac.mz/~ieseacmz/lib/publication/III_Conf2012/\nIESE_IIIConf_Paper6.pdf. Accessed 16 July 2021i p\ny\nin an Emerging Market Setting. Paper presented at iii conferencia international do ieSe, Maputo, \nMozambique, 3–4 September. https://www.iese.ac.mz/~ieseacmz/lib/publication/III_Conf2012/\nIESE IIIC\nf P\n6 df A\nd 16 J l 2021 IESE_IIIConf_Paper6.pdf. Accessed 16 July 2021i Karwowski E (2020) Economic development and variegated financialization in emerging economies. In: \nMader P, Mertens D, van der Zwan N (eds) The Routledge International Handbook of Financializa-\ntion. Routledge, London, pp 162–176 i\nMader P, Mertens D, van der Zwan N (eds) The Routledge International Handbook of Financializa-\ntion. Estudos Avançados 26(75):41–56 Routledge, London, pp 162–176 g\npp\nKarwowski E, Stockhammer E (2017) Financialisation in emerging economies: a systematic overview \nand comparison with Anglo-Saxon economies. Econ Polit Stud 5(1):60–86fl Kohler K (2019) Exchange rate dynamics, balance sheet effects, and capital flows. A Minskyan model \nemerging market boom-bust cycles. Struct Chang Econ Dyn 51:270–283f Kohler K, Stockhammer E (2022) Growing differently? Financial cycles, austerity, and competitivene\nin growth models since the Global Financial Crisis. Rev Int Polit Econ 29(4):1314–1341f Kohli V (2018) Functional income distribution and effective demand in India: an applied post Keynesian \nmodel. J Post Keynes Econ 41(3):399–429i y\n( )\nKrugman P (1999) Balance sheets, the transfer problem, and financial crises. Int Tax Public Financ \n6:459–472 Kurt OE (2018) Functional income distribution and growth in South Korea: a post Keynesian econome\nric analysis. Marmara Üniv İktisadi İdari Bilimler Dergisi 39(2):553–576 Lane PR, Milesi-Ferretti GM (2018) External Wealth of Nations database based on Lane PR, Milesi-\nFerretti GM (2018) the external wealth of nations revisited: international financial integration in \nthe aftermath of the Global Financial Crisis. IMF Econ Rev 66:189–222 Lavoie M (2009) Cadrisme within a Kaleckian model of growth and distribution. Rev Polit Eco\n21(3):371–393 Lavoie M (2014) Post-Keynesian Economics: New Foundations. Edward Elgar, Cheltenham Lavoie M, Stockhammer E (2013) Wage-led growth: concept, theories and policies. In: Lavoie M, Stock-\nhammer E (eds) Wage-led growth: an equitable strategy for economic recovery. Palgrave Macmil-\nlan, Basingstoke, pp 13–39 (\n)\ng\ng\np\np\nhammer E (eds) Wage-led growth: an equitable strategy for economic recovery. Palgrave Macmil-\nlan, Basingstoke, pp 13–39 g\npp\nLoaiza O, Tobón A, Hincapié D (2017) Impacto de la distribución funcional del ingreso sobre el producto \ninterno bruto de Colombia, 1970–2011. Lecturas Econ 86:63–104 McCauley RN, McGuire P, Sushko V (2015) Dollar credit to emerging market economies. BIS Quarter\nReview December: 27–41 McCombie JSL (1989) ‘Thirlwall’s Law’ and balance of payments constrained growth – a comment o\nthe debate. Appl Econ 21(5):611–629 pp\nMenezes TC, de Souza LR (2019) The main determinants of economic growth in Brazil between 2003 \nand 2010: exports or domestic market? Contextualizaciones Latinoamericanas 11(20):1–14 Mertens D, Nölke A, May C, Schedelik M, ten Brink T, Gomes A (2022) Moving the center: adapting the \ntoolbox of growth model research to emerging capitalist economies. Working Paper 188. Estudos Avançados 26(75):41–56 https://doi.org/10.4000/regulation.21444 Pollin R, Heintz J (2011) How wall street speculation is driving up gasoline prices today. Research Brief \nJune 2011. Political Economy Research Institute University of Massachusetts, Amherst Pollin R, Heintz J (2011) How wall street speculation is driving up gasoline prices today. Research Br\nJune 2011. Political Economy Research Institute University of Massachusetts, Amherst June 2011. Political Economy Research Institute University of Massachusetts, Amherst Priewe J, Herr H (2005) The macroeconomics of development and p\nthe Washington Consensus. Nomos, Baden-Baden Priewe J, Herr H (2005) The macroeconomics of development and poverty reduction: strategies beyond\nthe Washington Consensus. Nomos, Baden-Baden Rapetti M (2020) The real exchange rate and economic growth: a survey. J Glob Dev 11(2):1–54 Regan A, Brays S (2018) Celtic phoenix or leprechaun economics? The politics of an FDI-led growth \nmodel in Europe. New Political Economy 23(2):223–238. https://doi.org/10.1080/13563467.2017.\n1370447 Rethel L (2010) Financialization and the Malaysian political economy. Globalizations 7(4):489–506i Rey H (2015) Dilemma not trilemma: the global financial cycle and monetary policy \nWorking Paper 21162. National Bureau of Economic Research, Cambridge, MA Rey H (2015) Dilemma not trilemma: the global financial cycle and monetary policy indepen\nWorking Paper 21162. National Bureau of Economic Research, Cambridge, MA Schedelik M, Nölke A, May C, Gomes A (2023) Dependency revisited: commodities, commodity-related \ncapital flows and growth models in emerging economies European Journal of Economics and Eco-\nnomic Policies: Intervention, forthcoming g\nSchedelik M, Nölke A, Mertens D, May C (2021) Comparative capitalism, growth models and emerging \nmarkets: the development of the field. New Polit Econ 26(4):514–526 i\nSierra J (2022) The politics of growth model switching: why Latin America tries, and fails, to abandon \ncommodity-driven growth. In: Baccaro L, Blyth M, Pontusson J (eds) Diminishing Returns: The i\nSierra J (2022) The politics of growth model switching: why Latin America tries, and fails, to abandon \ncommodity-driven growth. In: Baccaro L, Blyth M, Pontusson J (eds) Diminishing Returns: The \nNew Politics of Growth and Stagnation. Oxford University Press, Oxford, pp 167–188 Solt F (2020) Measuring income inequality across countries and over time: the standardized world \nincome inequality database. Soc Sci Q 101(3):1183–1199 Stockhammer E (2015) Rising inequality as a cause of the present crisis. Camb J Econ 39(3):935–958 Stockhammer E (2022a) Post-Keynesian macroeconomic foundations for Comparative Political Econ-\nomy. Polit Soc 50(1):156–187 Stockhammer E (2022b) Macroeconomic ingredients for a growth model analysis for peripheral econo-\nmies: a post-Keynesian-structuralist approach. Estudos Avançados 26(75):41–56 Institute \nfor International Political Economy (IPE), Berlini Minsky HP (1977) The financial instability hypothesis: an interpretation of Keynes and an alternative \n“standard” theory. Challenge 20(1):20–27 y\ng\n( )\nMolero-Simarro R (2015) Functional distribution of income, aggregate demand, and economic growth in \nthe Chinese economy, 1978–2007. Int Rev Appl Econ 29(4):435–454 Morgan G, Doering H, Gomes M (2021) Extending varieties of capitalism to emerging economies: what \ncan we learn from Brazil? New Polit Econ 26(4):540–553 Morgan G, Doering H, Gomes M (2021) Extending varieties of capitalism\ncan we learn from Brazil? New Polit Econ 26(4):540–553 Morlin GS, Passos N, Pariboni R (2022) Growth theory and the growth model perspective: insights from \nthe supermultiplier. Rev Polit Econ Adv Access. https://doi.org/10.1080/09538259.2022.2092998 3 Growth drivers in emerging capitalist economies: building… 385 Nölke A, May C, Mertens D, Schedelik M (2022) Elephant limps, but jaguar stumbles: unpacking the \ndivergence of state capitalism in Brazil and India through theories of capitalist diversity. Compet \nChang 26(3–4):311–333 g\nNölke A, Vliegenthart A (2009) Enlarging the varieties of capitalism: the emergence of dependent mar-\nket economies in East Central Europe. World Polit 61(4):670–702. https://doi.org/10.1017/S0043\n887109990098 OEC (2021) Economic Complexity Rankings. https://oec.world/en/rankings/eci/hs6/hs96. Accessed 20 \nMay 2021 Ocampo JA, Parra MA (2006) The commodity terms of trade and their strategic implications for devel-\nopment. In: Jomo KS (ed) Globalization under hegemony: the changing world economy. Oxford \nUniversity Press, New Delhi; New York, pp 164–194 y\npp\nOnaran Ö, Galanis G (2012) Is aggregate demand wage-led or profit-led? National and global effects. ILO, Genevaf Onaran Ö, Stockhammer E (2005) Two different export-oriented growth strategies: accumulation and dis-\ntribution in Turkey and South Korea. Emerg Mark Financ Trade 41(1):65–89i Oyvat C, Öztunalı O, Elgin C (2020) Wage-led versus profit-led demand: a comprehensive empiric\nanalysis. Metroeconomica 71(3):458–486 Palan R (2002) Tax havens and the commercialization of state sovereignty. Int Organ 56(1):151–176i Palley TI (2017) Wage- vs. profit-led growth: the role of the distribution of wages in determining regime \ncharacter Camb J Econ 41(1):49 61 y\n(\n)\ng\npi\ng\ng\ng\ng\ncharacter. Camb J Econ 41(1):49–61\nPassos N, Morlin G (2022) Growth models and comparative political economy in Latin America. Re\nRégulation 33(2). https://doi.org/10.4000/regulation.21444 character. Camb J Econ 41(1):49 61\nPassos N, Morlin G (2022) Growth models and comparative political economy in Latin America. Rev \nRégulation 33(2). https://doi.org/10.4000/regulation.21444 ,\n(\n)\np\np\ny\nRégulation 33(2). Estudos Avançados 26(75):41–56 New Polit Econ Adv Access. https://doi.org/10.\n1080/13563467.2022.2149723 Stockhammer E, Kohler K (2022) Learning from distant cousins? Post-Keynesian economics, compar\ntive political economy, and the growth models approach. Rev Keynes Econ 10(2):184–203fi Stockhammer E, Qazizada W, Gechert S (2019) Demand effects of fiscal policy since 2008. Rev Keynes \nEcon 7(1):57–74 Stockhammer E, Wildauer R (2016) Debt-drive\ncountries. Camb J Econ 40(6):1609–1634 Stockhammer E, Wildauer R (2016) Debt-driven growth? Wealth, distribution and demand in OECD \ncountries. Camb J Econ 40(6):1609–1634 Strauss I, Isaacs G (2016) Labour compensation growth in the South African economy: assessing its \nimpact through the labour share using the Global Policy Model. National Minimum Wage Research \nInitiative Working Paper Series No. 4. University of the Witwatersrand, Johannesburg 1 3 3 3 386 B. Jungmann Tan Y, Conran J (2022) China’s growth models in comparative and international perspectives. In: Baccaro \nL, Blyth M, Pontusson J (eds) Diminishing Returns: The New Politics of Growth and Stagnation. Oxford University Press, Oxford, pp 143–166 Thirlwall AP (1979) The balance of payments constraint as an explanation of the international growth \nrate differences. PSL Q Rev 32(128):45–53 f\nTomio BT (2020) Understanding the Brazilian demand regime: a Kaleckian approach. Rev Keynes Eco\n8(2):287–302 UNCTAD (2021) The State of Commodity Dependence 2021. United Nations, Geneva Van Doorslaer H, Vermeiren M (2021) Pushing on a String: Monetary Policy, Growth Models and the \nPersistence of Low Inflation in Advanced Capitalism. New Political Economy 26(5):797–816. https://doi.org/10.1080/13563467.2020.1858774 p\ng\nWang P (2009) Three essays on monetary policy and economic growth in China. Dissertation, University \nof Ottawa Woodgate R (2021) Multinational corporations and commercialised states: can state aid serve as the basis \nfor an FDI-driven growth strategy? Working Paper 161. Institute for International Political Econ-\nomy (IPE), Berlinfi y\nWoodgate R (2023) FDI-led growth models: Sraffian supermultiplier models of export platforms and tax \nhavens. Eur J Econ Econ Policies: Interv. forthcoming Woodgate R (2023) FDI-led growth models: Sraffian supermultiplier models of export platforms and ta fi\nhavens. Eur J Econ Econ Policies: Interv. forthcoming World Bank (2021) World Development Indicators. https://databank.worldbank.org/home.aspx. Accessed \n2 March 2021 Wray LR (2019) Alternative paths to modern money theory. Real-World Econ Rev 89:5–22i Yılmaz E (2015) Wage or profit-led growth? The case of Turkey. J Econ Issues 49(3):814–834 2\t\nCentre d’Economie de l’Université Paris Nord (CEPN), Université Sorbonne Paris Nord, \nVilletaneuse, France 1\t\nInstitute for International Political Economy (IPE), Berlin School of Economics and Law (HWR\nBerlin), Badensche Str. 52, 10825 Berlin, Germany Benjamin Jungmann1,2 1\t\nInstitute for International Political Economy (IPE), Berlin School of Economics and Law (HWR\nBerlin), Badensche Str. 52, 10825 Berlin, Germany 1 3 3 1 3"
https://openalex.org/W2140192046	http://mediccreview.org/wp-content/uploads/2018/04/mr_382-1.pdf	English	null	Cancer Screening: Global Debates and Cuban Experience	MEDICC review	2,014	cc-by	4,637	INTRODUCTION A national cancer control program is a public health program designed to reduce cancer incidence and mortality and improve cancer patients’ quality of life, through systematic and equitable implementation of evidence-based strategies for prevention, early detection, treatment and palliative care, making the best possible use of available resources.[1] Early detection is based on diagnosis early in a cancer’s develop- ment, in pre-symptomatic stages, when treatment is more likely to be effective. Early detection depends on two main components: education and screening.[1] An estimated 30% of cancers could be cured with early detection and appropriate treatment.[2] 6FUHHQLQJ LV WKH SUHVXPSWLYH LGHQWL¿FDWLRQ RI XQUHFRJQL]HG GLV- ease by means of tests that can be applied rapidly to separate the healthy from the possibly sick.[3] While appropriateness of some VSHFL¿FWHVWVLVVWLOOGHEDWHGFDQFHUVFUHHQLQJDVVXFKLVDFFHSWHG by the international medical community based on the evidence. This paper reviews screening strategies for the main cancer sites for which screening is recommended and assesses the criteria and results of studies by WHO and other international organiza- tions and institutions, as well as the recommendations of Cuba’s cancer control strategy. Screening program success depends on various factors related to the disease, the proposed test and the health system. The can- cers chosen for screening should be common and associated with high morbidity and mortality, and there should be treatment avail- able that can reduce these indicators. The screening test should EHDFFHSWDEOHWRWKHSRSXODWLRQIRUZKLFKLWLVGHVLJQHGVXI¿FLHQWO\ sensitive to detect early lesions, feasible for use in medical prac- tice, and relatively inexpensive.[1] It should be sensitive enough to GHWHFWFDQFHULQWKRVHZKRKDYHWKHGLVHDVHDQGVSHFL¿FHQRXJK to correctly rule out individuals without the disease. Two other LPSRUWDQWWHVWFKDUDFWHULVWLFVGHSHQGRQVHQVLWLYLW\DQGVSHFL¿FLW\ but also on disease prevalence in the population screened: positive predictive value, or extent to which those who test positive actually have the disease; and negative predictive value, or extent to which those who test negative are actually disease free.[1] Cervical cancer The purpose of cervical cancer screening is to detect precancerous lesions as well as in situ or very early invasive cancer, when treatment is most effective. Three different types of tests are available: conventional (Pap) and liquid-media cytol- ogy, visual inspection with acetic acid, and HPV testing for onco- genic HPV types. Screening with appropriate followup can attain VLJQL¿FDQWUHGXFWLRQVLQLQFLGHQFHDQGPRUWDOLW\IURPDGYDQFHG cervical cancer.[2] One risk of such screening is that detection of low-grade squamous intraepithelial lesions in young women may lead to followup tests and possibly unnecessary treatment. Cancer Screening: Global Debates and Cuban Experience Rebeca S. González MD MS &KDUDFWHULVWLFVRIQDWLRQDOKHDOWKV\VWHPVDOVRLQÀXHQFHZKLFK screening approaches are appropriate. The system must have WKHUHVRXUFHVWRGHOLYHUVFUHHQLQJWRDVXI¿FLHQWO\ODUJHSURSRU- WLRQRIWKHSRSXODWLRQWRREWDLQWKHH[SHFWHGEHQH¿WVDQGLWPXVW also have the capacity to provide subsequent diagnosis, treat- ment and followup. In developing such screening strategies, WHO recommends that national cancer control programs should not impose the advanced technologies of high-income countries on countries with more limited resources. Hence, policies will undoubtedly differ among countries.[1] ABSTRACT A review was conducted of screening strategies for detecting the main cancer sites for which screening has been recommended, assessing WHO and other international organizations’ positions, as well as the requirements of Cuba’s cancer control strategy. Universally, screening is recommended for cervical, breast and colorectal cancer, all included in the Cuban strategy. Addition- ally, in Cuba, PSA testing is indicated for men considered at risk (aged >45 years with family history) and those aged >50 years who request it; annual oral exams and teaching of oral self- examination are recommended for the entire population; and for DGXOWVDJHG!\HDUVDFWLYHDQQXDORUDOFDQFHUFDVH¿QGLQJ Screening for skin cancer is performed by physical examination RILQGLYLGXDOVDWULVN7RPD[LPL]HEHQH¿WVRIHDUO\FDQFHUGHWHF- tion, greater coverage is needed as well as studies of how well screening is performing under current Cuban conditions. Early detection is a strategic component of Cuba’s Comprehen- sive Cancer Control Program. This in turn comprises three dif- ferent approaches: population screening (of apparently healthy SHRSOHWDUJHWHGVFUHHQLQJRILQGLYLGXDOVDWULVNDQGFDVH¿QG- ing in symptomatic patients. The Program is considering orga- QL]LQJ SLORW DUHDV WR YDOLGDWH VWXG\ ¿QGLQJV LQ DV\PSWRPDWLF populations, to assess generalizability of screening trial results.[4] KEYWORDS Cancer, early detection, screening, secondary pre- vention, breast cancer, cervical cancer, colorectal cancer, skin cancer, oral cancer, Cuba In each of Cuba’s 168 municipalities, a Center for Comprehensive Active Screening has been set up, reporting to the municipal health GHSDUWPHQW7KHVHDUHODERUDWRULHVRXW¿WWHGZLWKXOWUDPLFURDQD- lytic system (SUMA, the Spanish acronym) technology, staffed by personnel trained in its use. Havana’s Immunoassay Center sup- ports laboratory operations through reagent distribution, technical assistance with equipment repair and maintenance, and analyti- cal assistance as requested by clients or in response to results of external audits. The laboratories were conceived to expand diag- nostic capabilities beyond traditional techniques and to conduct other activities for program development and quality assurance, as well as to enable early detection of cancers included in screen- ing programs, to conduct screening studies on population groups DFFRUGLQJWRULVNVWUDWL¿FDWLRQDQGSRSXODWLRQKHDOWKVWDWXVDQGWR ensure comprehensive services for improved diagnostic quality and health outcomes. MEDICC Review, July–October 2014, Vol 16, No 3–4 Perspective Perspective Perspective Furthermore, a mammogram every two years is recommended for women aged 50–64 years and suspected cases are referred directly to multidisciplinary clinics for diagnosis and hospital treat- ment (unpublished MINSAP document).[4] Contrary to international recommendations, Cuba’s breast cancer prevention strategy also targets women aged <50 years, because the epidemiology of breast cancer in Cuba demands it. In 2009, 26.5% breast cancer diagnoses (816/3078) were in women aged <50 years. In 2012, there were 482 breast cancer deaths in wom- en aged <60 years: 32 in women aged 20–39 years and 450 in women aged 40–59 years.[10] These same videocolposcopes have been installed in 98 munici- pal clinics for benign cervical pathology, reinforcing this strategy; all 168 municipalities in Cuba are expected to be equipped by 2015. Sets of instruments and disposable reagents are also being distributed to increase program coverage; all components are manufactured in Cuba, reducing costs. In 2013, 72,129 women aged >30 years were assessed in munici- pal breast health clinics and 35,398 in multidisciplinary clinics for diagnosis and hospital treatment. A total of 64,144 mammograms were performed, of which 25,420 were to screen women aged 50–64 years, only 2.6% coverage (data from Form 241-434-03, National Breast Cancer Control Program 2013, Statistical Infor- mation System, MINSAP). Much better program coverage is needed, since screening does not yet appear to impact reduction of breast cancer mortality, which continues to climb. In 2000, there were 1012 breast cancer deaths (18.2/100,000 women); in 2011, there were 1390 (24.8/100,000) and 2012, 1521 (27.1/100,000). >@$OWKRXJKWKHVHUDWHVDUHQRWDJHDGMXVWHGWKH\UHÀHFWDUHDO and substantial increase in disease burden. Since 1997, Pap tests are repeated every three years for the age group indicated. From 1997 to 1999, 2,074,806 women aged >25 years were tested, for an estimated coverage of 66.3%. From 2009 to 2011, the number increased slightly to 2,104,567, for an esti- mated coverage of 67%.[10] The proportion of cases diagnosed at clinical stage 1 rose from 12.2% in 2011 to 18.4% in 2012, while the proportion diagnosed at stage 2 rose from 5.7% to 9.2%. That same year, cancer became the leading cause of death in Cuba, but cervical cancer mortality fell, from 8.3 per 100,000 women (466 deaths) in 2011 to 7.9 per 100,000 (442 deaths) in 2012.[10] +RZHYHUWKHGHFUHDVHLQFHUYLFDOFDQFHUPRUWDOLW\WKH¿IWKOHDG- ing cause of cancer deaths in Cuban women) is less than hoped for. Perspective NCI and USPSTF recommend breast cancer screening with mam- mography every two years for women aged 50–74 years. They do not recommend either mammography or self-examination screening for women aged 40–49 years, noting that risks out- ZHLJKH[SHFWHGEHQH¿WVLQWKLVDJHJURXS>@:+2UHFRPPHQGV mammography screening for women aged 50–64 years who visit health institutions.[1,2] Unfortunately, mammography is expen- sive, so it is not a viable option for many countries, where priority is use for individual case diagnosis, particularly for women who detect an abnormality during self-examination.[1,3] KDYH D 3DSDQLFRODX F\WRORJ\ 3DS WHVW HYHU\ \HDUV LQ women aged 30–65 years, frequency may be reduced to every 5 years if combined with HPV testing.[6] These recommendations coincide with those of the American Cancer Society in association with the American Society for Colposcopy and Cervical Pathology and the American Society for Clinical Pathology.[7] WHO recom- mends, as a minimum, one-time screening for all women aged 30–49. WHO also recommends that screening intervals should EHDWOHDVW¿YH\HDUVRUDWOHDVWWHQ\HDUVLQWKHFDVHRI+39 testing.[8] NCI and USPSTF recommend breast cancer screening with mam- mography every two years for women aged 50–74 years. They do not recommend either mammography or self-examination screening for women aged 40–49 years, noting that risks out- ZHLJKH[SHFWHGEHQH¿WVLQWKLVDJHJURXS>@:+2UHFRPPHQGV mammography screening for women aged 50–64 years who visit health institutions.[1,2] Unfortunately, mammography is expen- sive, so it is not a viable option for many countries, where priority is use for individual case diagnosis, particularly for women who detect an abnormality during self-examination.[1,3] Since 1968, Cuba’s Ministry of Public Health (MINSAP, the Span- ish acronym), in coordination with civil-society organizations, has implemented cervical cancer control activities. The current strate- gy, applied in primary health care, is Pap testing every three years for women aged 25–64 years. If results are abnormal, the woman is referred to one of 45 secondary-level cervical pathology clinics that provide coverage across Cuba, equipped with the personnel and technologies needed for diagnosis, treatment and followup. [9] Each clinic has a SUMASCOPE videocolposcope and SUMA- CRAF electrocautery equipment coupled with an ASPIRAL smoke aspirator, all manufactured by Havana’s Immunoassay Center and distributed by TecnoSuma International. In Cuba, monthly breast self-examination and an annual checkup E\IDPLO\GRFWRUVDUHUHFRPPHQGHGIRUZRPHQDJHG\HDUV ,I ¿QGLQJV DUH VXVSLFLRXV SDWLHQWV DUH UHIHUUHG WR PXQLFLSDO breast health clinics. If suspicions persist, referrals are made to hospital-based multidisciplinary diagnosis and treatment clinics. INTRODUCTION There is also the chance of false positive results, causing anxiety, and of false negative results that can delay medical treatment.[5] The US National Cancer Institute (NCI) and the US Preventive Services Task Force (USPSTF) recommend that women aged 73 Peer Reviewed Perspective MEDICC Review, July–October 2014, Vol 16, No 3–4 Perspective In March 2014, the US National Comprehensive Cancer Network (NCCN) published updated guidelines for use of PSA testing in prostate cancer screening, recommending routine VFUHHQLQJRIPHQDJHG\HDUVZLWKIUHTXHQF\RIVXEVHTXHQW screening dependent on results of baseline screening with PSA and digital rectal examination.[24] Colorectal cancer screening does have potential negative effects. Sigmoidoscopy and colonoscopy can be uncomfortable, even painful. They can also, albeit infrequently, cause complications such as perforation, hemorrhaging or cardiovascular events. Fecal occult blood testing can produce false positives that can lead to further testing—with attendant iatrogenic risks—or false negatives (if the lesion bleeds only intermittently, or not at all), which can delay diagnosis.[17] :+2¶VSRVLWLRQLVWKDWGLJLWDOUHFWDOH[DPLQDWLRQLVQRWVXI¿FLHQWO\ sensitive to detect early stages of the disease, and it is not yet clear whether PSA reduces mortality. WHO insists on the need to establish PSA’s effectiveness with well-designed randomized trials, currently under way.[1,3] In 2007, the World Gastroenterology Organization published the International Digestive Cancer Alliance Practice Guidelines, based on review of the various screening options. The Guidelines involve a screening cascade, with recommendations for settings at six different resource levels, from minimum to high.[13] NCI and USPSTF recommend three options for screening individuals aged \HDUV IRUFRORUHFWDO FDQFHU DQQXDO IHFDO RFFXOW EORRG WHVW sigmoidoscopy every 5 years with fecal occult blood test every 3 years, or colonoscopy every 10 years.[18] WHO recognizes that screening with sigmoidoscopy and annual fecal occult blood test- ing reduces mortality from colorectal cancer and emphasizes the QHHGIRUKLJKHUSRSXODWLRQFRYHUDJHWRREWDLQJUHDWHUEHQH¿WVLQ proportion to costs generated.[1,2] In Cuba, PSA is recommended for symptomatic patients; men DJHG\HDUVZLWKDIDPLO\KLVWRU\RISURVWDWHFDQFHUDQGWKRVH aged >50 years who request it.[4] The test used is UMELISA PSA, an assay designed by the Immunoassay Center to quantify total and free PSA in human serum samples.[25] Prostate cancer is the second leading cause of cancer deaths in Cuban men, and crude mortality rates have risen in recent years, from 1844 deaths (33.1 per 100,000 men) in 2000 to 2703 deaths (48 per 100,000) in 2012.[10] In 2013, 238,799 PSA tests were performed in the national health system, in 14.5% of the male population aged >50 years, with 14.8% positivity (data from Form 241-509-01, SUMA Technology 2013, Statistical Information Sys- tem, MINSAP). Research is needed to assess the results of PSA testing in Cuba’s health system. An as-yet unpublished study in the province of Santiago de Cuba is under way, but focusing on PSA screening’s impact on incidence, not on mortality. Perspective ,Q&XEDLWLVUHFRPPHQGHGWKDWDV\PSWRPDWLFSDWLHQWVDJHG \HDUVKDYHDQDQQXDOIHFDORFFXOWEORRGWHVWDQGÀH[LEOHUHFWR- sigmoidoscopy or double-contrast barium enema every 5 years. Patients with a family history of colorectal cancer should have D FRORQRVFRS\ HYHU\ ¿YH \HDUV 6LQFH LQ DGGLWLRQ WR WKH traditional guaiac test for fecal occult blood, a rapid qualitative immunological test to detect occult human blood in fecal mat- ter has been available. Called SUMASOHF, it was designed by the Immunoassay Center and is distributed throughout Cuba by TecnoSuma International. It is a highly sensitive sandwich-type chromatographic immunoassay for early diagnosis and followup of gastrointestinal disorders that provoke bleeding. In 5 minutes, human hemoglobin levels as low as 0.2 μg/mL can be detected in fecal matter.[19] In 2013, SUMASOHF was introduced in the 116 Cuban municipalities with the highest colorectal cancer mortal- ity rates. A total of 50,756 individuals aged >50 were screened, coverage of only 2.7%; positivity was 10.4% (data from Form 241- 509-01 SUMA Technology 2013, Statistical Information System, MINSAP). Screening coverage must be increased as quickly as possible to reduce mortality from colorectal cancer, the third lead- ing cause of cancer deaths in Cuba, with 2089 deaths in 2012 (18.6 deaths per 100,000 population).[10] Oral cancer According to NCI and USPSTF, it has still not been demonstrated that oral cancer screening by direct visual exam- ination reduces mortality from the disease.[26] The Cochrane 2UDO+HDOWKURXSFRQFOXGHVWKDWWKHUHLVLQVXI¿FLHQWHYLGHQFH to determine whether screening by visual inspection reduces mortality rates, and there have been no trials of other screen- ing methods. However, there is some evidence that screening may help reduce mortality in patients who use tobacco and DOFRKRODOWKRXJKWKHRQO\VWXG\LQFOXGHGWKDWVKRZHGEHQH¿WV may have been affected by bias due to faulty randomization. [27] WHO agrees that detection of precancerous lesions or early-stage tumors has not yet been shown to reduce oral can- cer mortality.[1,2] Prostate cancer Expert opinion on the PSA test has varied over the years. Until recently, many physicians and organizations rec- ommended that men be tested annually beginning at age 50; and men at higher risk even earlier, at 40 or 45 years. Several random- ized trials of PSA screening have been conducted, with contra- dictory results. Perspective Indeed, 67% is below the coverage threshold for screening to SURGXFHSDOSDEOHEHQH¿WV>@7KLVEHJVUHYLHZRIKRZWKHVWUDW- egy is being applied, as well as of related organizational factors in the system that may be preventing attainment of screening’s IXOOEHQH¿W Colorectal cancer Screening for colorectal cancer is based on GHWHFWLQJDGHQRPDDVSHFL¿FSUHQHRSODVWLFOHVLRQWKDWFDQEH treated effectively. More than 90% of colorectal cancer cases are preceded by these premalignant changes. Lifetime risk of colorectal adenoma is 20%, with peak incidence at ages 55–65 years. Lifetime probability of colorectal cancer is 4%–6%, with maximum incidence at ages 65–75 years; it affects men and women equally. Some 75% of individuals who develop colorec- tal cancer have no personal or family history of the disease, so risk-group screening is not done, since it would detect only about 20% of cases. Early detection can lead to cure in 80%– 90% of cases.[13] Breast cancer The purpose of breast cancer screening is early detection. With appropriate facilities, screening by mammogra- phy, with or without physical breast examination, plus followup of LQGLYLGXDOVZLWKSRVLWLYHRUVXVSLFLRXV¿QGLQJVFDQUHGXFHEUHDVW cancer mortality by up to one third.[1] Such screening also has risks. It can lead to diagnosis of cancers that never would have caused symptoms or death and in so doing, expose patients to treatment risks and side effects. In addition, false positive results sometimes occur, engendering further diagnostic testing and cre- ating anxiety. On the other hand, false negative results can delay diagnosis and treatment. Radiation itself also increases breast cancer risk.[11] The tests most recommended for early detection of colorectal can- cer are the fecal occult blood test, sigmoidoscopy and colonosco- py. Double-contrast barium enema is another imaging procedure that has been recommended for screening. Other techniques, such as computerized tomographic colonoscopy or virtual colo- noscopy, have also been applied but have not yet been shown Peer Reviewed 74 MEDICC Review, July–October 2014, Vol 16, No 3–4 Perspective to reduce mortality rates. Trials are under way to assess tests for human DNA in fecal matter or blood samples.[14–16] early detection should be offered only to individuals well informed of the risk of detecting some prostate cancers that would never KDYHLQÀLFWHGKHDOWKSUREOHPVDQGWKDWFRXOGOHDGWRRYHUWUHDW- ment. Despite this, many insurance systems cover PSA testing, including Medicare in the United States.[23] The issue remains controversial. REFERENCES 8. World Health Organization. WHO guidance note: comprehensive cervical cancer pre- vention and control: a healthier future for girls and women. Switzerland: World Health Organization; 2013 [cited 2014 Apr 22]. 16 p. Available from: http://apps.who.int/iris/bitste am/10665/78128/3/9789241505147_engpdf?ua=1 9. Ministry of Public Health (CU). Programa Diag- nóstico Precoz del Cáncer de Cuello del Útero [Internet]. Havana: Ministry of Public Health (CU); 1999 [cited 2014 Apr 21]. 53 p. Avail- DEOH IURP KWWS¿OHVVOGFXVLGD¿OHV prog-detec-canc-cervicout.pdf. Spanish. from: http://www.cancer.org/espanol/cancer/ colonyrecto/guiadetallada/cancer-colorrectal -early-deteccion-recomendations. Spanish. 8. World Health Organization. WHO guidance note: comprehensive cervical cancer pre- vention and control: a healthier future for girls and women. Switzerland: World Health Organization; 2013 [cited 2014 Apr 22]. 16 p. Available from: http://apps.who.int/iris/bitste am/10665/78128/3/9789241505147_engpdf?ua=1 8. World Health Organization. WHO guidance note: comprehensive cervical cancer pre- vention and control: a healthier future for girls and women. Switzerland: World Health Organization; 2013 [cited 2014 Apr 22]. 16 p. Available from: http://apps.who.int/iris/bitste am/10665/78128/3/9789241505147_engpdf?ua=1 1. World Health Organization. National Cancer Control Programmes. Policies and manage- rial guidelines [Internet]. 2nd ed. Geneva: World Health Organization; c2008 [cited 2014 Mar 10]. 204 p. Available from: http://www.who.int/repro ductivehealth/publications/cancers/92415455 77/en/ 17. National Cancer Institute (US) [Internet]. Bethesda (US): National Cancer Institute (US); c2014. Exámenes de detección del cáncer colorrectal; 2014 [cited 2014 Apr 22]. Avail- able from: www.cancer.gov/espanol/pdq/dete ccion/colorrectal/HealthProfessional. Spanish. 9. 9. Ministry of Public Health (CU). Programa Diag- nóstico Precoz del Cáncer de Cuello del Útero [Internet]. Havana: Ministry of Public Health (CU); 1999 [cited 2014 Apr 21]. 53 p. Avail- DEOH IURP KWWS¿OHVVOGFXVLGD¿OHV prog-detec-canc-cervicout.pdf. Spanish. 2. World Health Organization. Detección temprana. Control del cáncer. Aplicación de los conocimien- tos. Guía de la OMS para desarrollar programas H¿FDFHV>,QWHUQHW@HQHYD:RUOG+HDOWK2UJD- nization; 2007 [cited 2014 Mar 10]. 48 p. Available from: http://www.who.int/cancer/publications/can cer_control_planning/es/. Spanish. 18. U.S. Preventive Services Task Force [Internet]. Maryland: U.S. Preventive Services Task Force; c2014. Screening for Colorectal Cancer: Rec- ommendation Statement; 2008 Oct [cited 2014 Mar 12]. Available from: http://www.uspreventive servicestaskforce.org/uspstf08/colocancer/co lors.htm 10. p g p p 10. National Health Statistics and Medical Records Division (CU). Anuario Estadístico de Salud 2012 [Internet]. Havana: Ministry of Public Health (CU); 2013 Apr [cited 2014 Apr 10]. 190 p. Avail- DEOH IURP KWWS¿OHVVOGFXGQH¿OHV anuario_2012.pdf. Spanish. 3. Fernández JA, Díaz J. Algunas consideraciones teóricas sobre la pesquisa activa Rev. Cubana Med Gen Integr [Internet]. 2009 [cited 2014 Apr 10];25(4):107–16. Available from: http://scielo .sld.cu/pdf/mgi/v25n4/mgi11409.pdf. Spanish. 19. 19. REFERENCES Herrera RC, Rodríguez R, Vivar JR, Lorenzo A, López L, Gato E, et al. SUMASOHF para el diagnóstico de sangre oculta: evaluación analíti- ca desarrollada en el Centro de Inmunoensayo [Internet]. Paper presented at: CUBA SALUD 2012. Proceedings of the Public Health Inter- national Convention CUBA SALUD 2012; 2012 Dec [cited 2014 Apr 23]; Havana, Cuba. Avail- able from: http://www.convencionsalud2012.sld .cu/index.php/convencionsalud/2012/paper/view Paper/1684. Spanish. 11. p g g p p 4. Ministry of Public Health (CU). Programa Integral para Control del Cáncer en Cuba. Estrategia Nacional para el Control del Cáncer. Havana: Ministry of Public Health (CU); 2012 [cited 2014 Mar 10]. 65 p. Available from: http://www .paho.org/cub/index.php%3Foption%3Dcom _docman%26task%3Ddoc_download% 2 6 gid%3D378%26Itemid%3D&sa=U&ei=3d6pU5a 5KIilyASy74GoBA&ved=0CAYQFjAA&client=int ernal-uds-cse&usg=AFQjCNGbj82tkF2Qv8Vdx YtRr-UC5j1XcQ. Spanish. 11. National Cancer Institute (US). Bethesda (US): National Cancer Institute (US); c2014. Exámenes de detección del cáncer de seno (mama); 2014 [updated 2014 Apr 18; cited 2014 Apr 22]. Avail- able from: www.cancer.gov/espanol/pdq/dete ccion/seno/HealthProfessional. Spanish. 12. U.S. Preventive Services Task Force. Screening for breast cancer: U.S. Preventive Services Task Force recommendations statement. Ann Intern Med. 2009 Nov 17;151(10):716–26. 20. 20. Schröder FH, Hugosson J, Roobol MJ, Tammela TL, Ciatto S, Nelen V, et al. Screening and pros- tate-cancer mortality in a randomized European study. N Engl J Med. 2009 Mar 26;360(13):1320–8. 13. 13. World Gastroenterology Organization; Interna- tional Digestive Cancer Alliance. Tamizaje del cáncer colorrectal [Internet]. Milwaukee (US): World Gastroenterology Organization; 2007 [cited 2014 Mar 12]. 19 p. Available from: http://www .worldgastroenterology.org/assets/downloads /es/pdf/guidelines/cancer_colorectal_tamiza je_screening_y_vigilancia.pdf. Spanish. 5. National Cancer Institute (US) [Internet]. Bethes- da (US): National Cancer Institute (US); 2014. Exámenes de detección del cáncer del cuello uterino; 2014 [updated 2014 Jun 13; cited 2014 Apr 22]; [about 3 screens]. Available from: www .cancer.gov/espanol/pdq/deteccion/cuelloute srino/HealthProfessional. Spanish. 21. 21. National Cancer Institute (US) [Internet]. Bethes- da (US): National Cancer Institute (US); c2014. Estudio de Exámenes de Detección de Cáncer de Próstata, Pulmón, Colorrectal y Ovarios (PLCO): preguntas y respuestas; 2014 [cited 2014 Mar 14]. Available from: http://www.cancer .gov/espanol/noticias/plcoscreeningQand ASpanish. Spanish. 6. U.S. Preventive Services Task Force [Internet]. Maryland: U.S Preventive Services Task Force; c2014. Screening for Cervical Cancer; 2012 [cited 2014 Apr 22]; [about 1 screen]. Available from: http://www.uspreventiveservicestaskforce .org/uspstf/uspscerv.htm 14. American Cancer Society [Internet]. Atlanta: American Cancer Society; c2014. Detailed Guide: Colon and Rectum Cancer; 2012 [cit- ed 2014 Mar 13]. Available from: http://www .cancer.org/Cancer/ColonandRectumCancer/ DetailedGuide/index 22. Eskersberger E, Finkelstein J, Sadri H, Margreiter M, Taneja SS, Lepor H, et al. Perspective (4.4/100,000).[10] Again, these are crude rates and apparent trends may be affected by demographic changes. must decide whether to include screening innovations in national SURJUDPVEDVHGRQHPHUJLQJVFLHQWL¿FHYLGHQFHDVZHOODVWKHLU own epidemiological and economic conditions. Skin cancer NCI, USPSTF and WHO do not consider that WKHUHLVVXI¿FLHQWHYLGHQFHWKDWYLVXDOLQVSHFWLRQVNLQFKHFNRI asymptomatic individuals leads to reduced skin cancer mortality. [1,2,29] Skin cancer NCI, USPSTF and WHO do not consider that WKHUHLVVXI¿FLHQWHYLGHQFHWKDWYLVXDOLQVSHFWLRQVNLQFKHFNRI asymptomatic individuals leads to reduced skin cancer mortality. [1,2,29] FINAL CONSIDERATIONS Through its Comprehensive Cancer Control Program, Cuba, despite its economic limitations, has established a screening strategy that enables use of advanced technology based on cur- UHQW VFLHQWL¿F HYLGHQFH DQG ZLWK D IDYRUDEOH FRVWEHQH¿W UDWLR the latter due to broad use of domestically manufactured prod- ucts. The general public and health personnel need to be more aware of the strategy, and coverage should be expanded, either through referrals by primary health care teams or as requested by SDWLHQWVDWIDPLO\GRFWRUDQGQXUVHRI¿FHV:HDOVRQHHGVFUHHQ- ing studies conducted in real-world Cuban conditions, studies that look not just at higher detection rates (diagnosis of premalignant lesions or early-stage cancer) but also at the strategy’s main goal: reduced mortality. In Cuba, screening by physical examination is recommended for individuals at risk for skin cancer.[4] Data needed to assess the success of this strategy have not been published. In 2000, there were 271 deaths from skin cancer (2.4/100,000 population), while in 2012 there were 386 (3.4 per 100,000). In both years, it was the 13th most frequent cause of cancer deaths.[10] Other cancer sites Today, screening tests for cancer at various sites are being evaluated through a range of trials that include so-called advanced technologies. Policymakers in each country Perspective A study by US NCI concluded that incidence was rising but mortality was not falling, while the European Random- L]HG6WXG\RI6FUHHQLQJIRU3URVWDWH&DQFHU(563&GLG¿QGD drop in mortality (a trend that seems to be supported by sustained lower rates over a longer period of followup).[20–22] Cuba’s strategy recommends an annual mouth-and-throat exam and teaching oral self-examination for the entire population, and DFWLYHDQQXDOFDVH¿QGLQJLQWKHFRPPXQLW\WRVHHNLQGLYLGXDOV DJHG\HDUVDWULVNZKRDUHWKHQUHIHUUHGIRUH[DPLQDWLRQ [4] Control actions for this cancer have been carried out in Cuba since 1982, as part of a national oral cancer detection strategy. [28] Data have not been published that could help assess extent and results of the strategy’s implementation. In 2012, cancer of the lip, oral cavity and pharynx was the ninth leading cause of cancer mortality in Cuba in both sexes, with 641 deaths (5.7/100,000 population), up from 492 in 2009 The majority of urological associations have concluded that gen- eralized population screening with PSA is not appropriate and Peer Reviewed 75 MEDICC Review, July–October 2014, Vol 16, No 3–4 MEDICC Review, July–October 2014, Vol 16, No 3–4 REFERENCES Screening for Pros- tate Cancer: A Review of the ERSPC and PCLO Trials. Rev Urol. 2009 Summer;11(3):127–33. 15. 23. g p p 7. Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam SL, Cain J, et al. American Can- cer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer. CA Cancer J Clin [Internet]. 2012 May– Jun [cited 2014 Apr 22];62(3):147–72. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3801360/ 15. Aune D, Chan DS, Lau R, Vieira R, Greenwood '& .DPSPDQ ( HW DO 'LHWDU\ ¿EUH ZKROH grains, and risk of colorectal cancer: systematic review and dose-response meta-analysis of pro- spective studies. BMJ. 2011 Nov 10;343:d6617. 15. Aune D, Chan DS, Lau R, Vieira R, Greenwood '& .DPSPDQ ( HW DO 'LHWDU\ ¿EUH ZKROH grains, and risk of colorectal cancer: systematic review and dose-response meta-analysis of pro- spective studies. BMJ. 2011 Nov 10;343:d6617. 16. American Cancer Society [Internet]. Atlanta: American Cancer Society; c2014. Recomen- daciones de la Sociedad Americana Contra el Cáncer para la detección temprana del cáncer colorrectal; 2014 [cited 2014 Mar 13]. Available 23. National Cancer Institute (US) [Internet]. Bethesda (US): National Cancer Institute (US); c2014. Análisis del antigen prostático espe- Ft¿FR 36$ >FLWHG $SU @ $YDLO- able from: www.cancer.gov/espanol/hojas-in formativas/deteccion-diagnostico/antigeno SURVWDWLFRHVSHFL¿FR6SDQLVK 16. American Cancer Society [Internet]. Atlanta: American Cancer Society; c2014. Recomen- daciones de la Sociedad Americana Contra el Cáncer para la detección temprana del cáncer colorrectal; 2014 [cited 2014 Mar 13]. Available S S S . Prostate Cancer Early Detection [Internet]. Pennsylvania: National Comprehensive Can- cer Network (US); c2014 [cited 2014 Apr S S S Prostate Cancer Early Detection [Internet]. Pennsylvania: National Comprehensive Can- cer Network (US); c2014 [cited 2014 Apr 76 MEDICC Review, July–October 2014, Vol 16, No 3–4 Peer Reviewed Perspective 21]. 35 p. Available from: http://www.tri-kobe .org/nccn/guideline/urological/english/prostate _detection.pdf THE AUTHOR Rebeca S. González Fernández (tspro- gnac1@cie.sld.cu), family physician with a master’s degree in comprehensive women’s health. Assistant professor, Medical University of Havana, and national programs coordinator, TecnoSuma International, Immunoassay Cen- ter, BioCubaFarma, Havana, Cuba. THE AUTHOR Rebeca S. González Fernández (tspro- gnac1@cie.sld.cu), family physician with a master’s degree in comprehensive women’s health. Assistant professor, Medical University of Havana, and national programs coordinator, TecnoSuma International, Immunoassay Cen- ter, BioCubaFarma, Havana, Cuba. and prevention of oral cancer. Cochrane Data- base of Systematic Reviews. MEDICC Review, July–October 2014, Vol 16, No 3–4 Submitted: April 7, 2014 Approved for publication: July 27, 2014 Disclosures: None 21]. 35 p. Available from: http://www.tri-kobe .org/nccn/guideline/urological/english/prostate _detection.pdf REFERENCES 2013 Nov 19 [cit- ed 2014 Jun 20];(11). doi: 10.1002/14651858. CD004150.pub4. Available from: http:// onlinelibrary.wiley.com/doi/10.1002/14651858 .CD004150.pub4/abstract;jsessionid=EEC9E40 7FD309514471C4BA5AE1BBE1F.f01t04 25. Immunoassay Center (CU). Prospecto UMELI- SA PSA. Edición No.2 Código UM 2036 [Inter- net]. Havana: Immunoassay Center (CU); 2010 Sep [cited 2014 Apr 21]. 11 p. Avail- able from: http://www.tecnosuma.com/Informa cion/Inserts%20(PDF)/UMELISA%20PSA.pdf. Spanish. 28. Ministry of Public Health (CU). Programa de Detección del Cáncer Bucal [Internet]. Havana: Ministry of Public Health (CU); 2001 Oct [cited $SU @ S $YDLODEOH IURP KWWS¿OHV .sld.cu/sida/files/2012/01/prog-detec-cancer -bucal2001.pdf. Spanish. Ministry of Public Health (CU). Programa de Detección del Cáncer Bucal [Internet]. Havana: Ministry of Public Health (CU); 2001 Oct [cited $SU @ S $YDLODEOH IURP KWWS¿OHV .sld.cu/sida/files/2012/01/prog-detec-cancer -bucal2001.pdf. Spanish. p 26. U.S. Preventive Services Task Force [Inter- net]. Maryland: U.S. Preventive Services Task Force; c2014. Screening for Oral Cancer: Recommendation Statement; 2004 [cited 2014 Mar 15]; [about 3 screens]. Avail- able from: http://www.uspreventiveservicestask IRUFHRUJXVSVWIRUDOFDQRUDOFDQ¿QDOUHFKWP 29. U.S. Preventive Services Task Force [Inter- net]. Maryland: U.S. Preventive Services Task Force; c2014. Screening for Skin Cancer: Recommendation Statement; 2009 [cited 2014 Mar 15]; [about 1 screen]. Avail- able from: http://www.uspreventiveservicestask force.org/uspstf/uspsskca.htm 29. U.S. Preventive Services Task Force [Inter- net]. Maryland: U.S. Preventive Services Task Force; c2014. Screening for Skin Cancer: Recommendation Statement; 2009 [cited 2014 Mar 15]; [about 1 screen]. Avail- able from: http://www.uspreventiveservicestask force.org/uspstf/uspsskca.htm Submitted: April 7, 2014 Approved for publication: July 27, 2014 Disclosures: None J S 27. The Cochrane Library [Internet]. Massachu- setts: John Wiley & Sons. Intervention Review. Screening programmes for the early detection 77 Peer Reviewed
https://openalex.org/W4312116083	http://www.thieme-connect.de/products/ejournals/pdf/10.1055/s-0042-1758535.pdf	English	null	Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN—An Observational Study from a Tertiary Care Center	Journal of digestive endoscopy	2,022	cc-by	3,772	Anudeep Katrevula1 Goutham Reddy Katukuri1 Aniruddha Pratap Singh1 Pradev Inavolu1 Hardik Rughwani1 Siddhartha Reddy Alla1 Mohan Ramchandani1 Nageshwar Reddy Duvvur1 Address for correspondence Mohan Ramchandani, MD, DM, Department of Gastroenterology, AIG Hospitals, Hyderabad, Telangana 500032, India (e-mail: ramchandanimohan@gmail.com). 1Department of Gastroenterology, AIG Hospitals, Hyderabad, Telangana, India J Digest Endosc 2023;14:3–7. J Digest Endosc 2023;14:3–7. © 2022. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution License, permitting unrestricted use, distribution, and reproduction so long as the original work is properly cited. (https://creativecommons.org/licenses/by/4.0/) Thieme Medical and Scientiﬁc Publishers Pvt. Ltd., A-12, 2nd Floor, Sector 2, Noida-201301 UP, India DOI https://doi.org/ 10.1055/s-0042-1758535. ISSN 0976-5042. Article published online: 2022-12-23 Article published online: 2022-12-23 Research Article 1Department of Gastroenterology, AIG Hospitals, Hyderabad, Telangana, India Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN—An Observational Study from a Tertiary Care Center Anudeep Katrevula1 Goutham Reddy Katukuri1 Aniruddha Pratap Singh1 Pradev Inavolu1 Hardik Rughwani1 Siddhartha Reddy Alla1 Mohan Ramchandani1 Nageshwar Reddy Duvvur1 article published online December 23, 2022 article published online December 23, 2022 DOI https://doi.org/ 10.1055/s-0042-1758535. ISSN 0976-5042. Abstract Background and Aims Precise optical diagnosis of colorectal polyps could improve the cost-effectiveness of colonoscopy and reduce polypectomy-related complications. We conducted this study to estimate the diagnostic performance of visual inspection alone (WLI þ NBI) and of EndoBRAIN (endocytoscopy-computer-aided diagnosis [EC- CAD]) in identifying a lesion as neoplastic or nonneoplastic using EC in real-world scenario. Methods In this observational, prospective, pilot study, a total of 55 polyps were studied in the patients aged more than or equal to 18 years. EndoBRAIN is an artiﬁcial intelligence (AI)-based system that analyzes cell nuclei, crypt structure, and vessel pattern in differentiating neoplastic and nonneoplastic lesion in real-time. Endoscopist assessed polyps using white light imaging (WLI), narrow band imaging (NBI) initially followed by assessment using EC with NBI and EC with methylene blue staining. The sensitivity, speciﬁcity, positive predictive value, negative predictive value, and accuracy of endoscopist and EndoBRAIN in identifying the neoplastic from nonneoplastic polyp was compared using histopathology as gold-standard. Results A total of 55 polyps were studied, in which most of them were diminutive (36/55) and located in rectum (21/55). The image acquisition rate was 78% (43/55) and histopathology of the majority was identiﬁed to be hyperplastic (20/43) and low-grade adenoma (16/43). EndoBRAIN identiﬁed colonic polyps with 100% sensitivity, 81.82% speciﬁcity (95% conﬁdence interval [CI], 59.7–94.8%), 90.7% accuracy (95% CI, 77.86– 97.41%), 84% positive predictive value (95% CI, 68.4–92.72%), and 100% negative predictive value. The sensitivity and negative predictive value were signiﬁcantly greater than visual inspection of endoscopist. The diagnostic accuracy seems to be superior; however, it did not reach statistical signiﬁcance. Speciﬁcity and positive predictive value were similar in both groups. Keywords ►EndoBRAIN ►endocytoscopy ►polyp Conclusion Optical diagnosis using EC and EC-CAD has a potential role in predicting the histopathological diagnosis. The diagnostic performance of CAD seems to be better than endoscopist using EC for predicting neoplastic lesions. Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN Katrevula et al. 4 from the study. Subjects who were scheduled for colonoscopy and found to have a polyp on white light endoscopy were included. Patients underwent colonoscopy with Olympus colonoscope (CF-HQ290, Olympus, Tokyo, Japan) equipped with EndoBRAIN technology and performed using EVIS LUCERA ELITE CV-290 processor (Olympus, Tokyo, Japan). Experienced endoscopist who performed more than 5,000 colonoscopies have performed the procedure. Endoscopic diagnosis of polyp was done under white light imaging and NBI using Japan NBI expert team classiﬁcation. Abstract EC-NBI and/or EC-stain images were acquired from the polyps. The acquired images were assessed by endoscopist in real-time and was asked to give a diagnosis (nonneoplastic/neoplastic) who was blinded to the EndoBRAIN diagnosis and histopathological diagnosis. The EndoBRAIN diagnosis of the polyp on EC NBI (nonneoplastic/neoplastic) and/or EC stain (nonneoplastic/ neoplastic) images was documented by the assistant. Resected polyps were sent for histopathological assessment who were blinded to the endoscopic diagnosis. Pathological assessment of polyps was performed by senior pathologist with experi- ence in the gastrointestinal histopathology. The number of polyps fromwhich good quality EC images could be acquired is calculated (image acquisition rate). The sensitivity, speciﬁcity, accuracy, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio, and negative likelihood ratio in identifying a neoplastic lesion are calculated. Polyps from which EC images could not be acquired were excluded from this analysis. Introduction Colorectal cancer (CRC) is a formidable health problem worldwide. In India, the annual incidence rates (AARs) for colon cancer and rectal cancer in men are 4.4 and 4.1 per 1,00,000, respectively. The AAR for colon cancer in women is 3.9 per 100000.1 Artiﬁcial intelligence (AI) known as computer vision in computer-aided diagnosis (CAD) and detection (CADe) helps in identifying health-related conditions based on medical imaging. Convolutional neural network (CNN) is a type of deep machine learning algorithm that uses convolutions of the input image to extract the most relevant information that helps to classify the image into different entities. Based on the accumulated data features, a deep CNN can diagnose newly acquired clinical images prospectively.2,3 Precise optical diagnosis of colorectal polyps could improve the cost-effectiveness of colonoscopy and reduce polypectomy-related complications. It is difﬁcult for com- munity-based nonexperts to obtain sufﬁcient diagnostic performance. CAD has potential for better accuracy and lower interobserver variability. Nonexpert endoscopists may more easily achieve accuracy levels sufﬁcient to meet the preservation and incorporation of valuable endoscopic innovations (PIVI) threshold.4 The EndoBRAIN technology has a potential in this regard with studies showing improved adenoma detection rate (ADR) and diagnostic accuracy reaching the PIVI thresholds. Removing precancerous polyps from the bowel during a colonoscopy is the cornerstone of CRC screening and prevents polyps developing into bowel cancer. Many polyps never grow into cancer and it can be difﬁcult for the clinicians performing the procedure (endoscopists) to tell which ones are precancer- ous. This means many polyps are removed unnecessarily, with a considerablewaste of resources. The EndoBRAIN system uses optical diagnostic technologies like endocytoscopy (EC) and narrow band imaging (NBI). EC enables in vivo observation of cells and nuclei at 520x ultramagniﬁcation using methylene blue staining, and combined with NBI, can observe micro- vessels in detail.5–8 EndoBRAIN may prove to be cost-effective by reducing biopsies and histopathology examinations. Usage of these technologies, especially in a high-volume center, may help us improve patient care, at the same time with cost- effectiveness.9 Statistical Analysis Baseline characteristics of the polys were described using descriptive statistics. Categorical data are described using percentages and frequencies and compared using Fisher’s exact test or chi-squared test. The normality of continuous data was assessed by Kolmogorov–Smirnov test and repre- sented as mean (standard deviation) or median (interquar- tile range). Comparison of the continuous data was done by independent Student’s t-test for parametric data and Mann– Whitney U-test for nonparametric data. Statistical analysis was performed at 5% level of signiﬁcance and p less than 0.05 was considered as statistically signiﬁcant. Results This is a pilot study conducted from January 2021 to June 2021. Institute review board and ethical board clear- ance was obtained prior to initiating the study (AHF/AIGH- IRB:02/46/2021). Study was conducted in accordance with ethical principles for human subjects as stated in the decla- ration of Helsinki. Informed consent was obtained from all the participants. Baseline characteristics showing number of polyps, image acquisition rate, and histopathological details were elucidated in ►Table 1. A total of 55 polyps were studied. Most of the polyps were diminutive and most of them were located in rectum. Good quality EC images using either EC-NBI or EC-stain mode were acquired from 43 out of 55 polyps (78.2%). However, the image acquisition rate was Journal of Digestive Endoscopy Vol. 14 No. 1/2023 © 2022. The Author(s). Methods This was a prospective, observational study conducted to estimate the accuracy of visual inspection alone and of Endo- BRAIN (EC-CAD) in identifying a lesion as neoplastic or non- neoplastic using EC.Thestudy population included individuals 18 years or older who were scheduled for screening, surveil- lance, diagnostic, or therapeutic colonoscopy. Patients with inﬂammatory bowel disease, polyposis syndrome (e.g., famil- ial adenomatous polyposis, serrated polyposis), history of chemotherapy or radiation therapy for colorectal lesions, and inability to undergo polypectomy (e.g., intake of anti- coagulants, comorbidities, or patient refusal) were excluded Journal of Digestive Endoscopy Vol. 14 No. 1/2023 © 2022. The Author(s). Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN Katrevula et al. eal-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN Katrevula et al. 5 5 5 Fig. 1 Nonneoplastic polyp. (A) White light imaging showing sessile polyp of size 5 mm (Paris 1s). (B) Narrow band imaging (NBI) showing sessile polyp with invisible vascular pattern with regular dots (Japan NBI expert team type 1). (C and D) Endocytoscopy with NBI and stain showing narrow serrated lumina and dense pattern of small roundish nodules (EC1b). lower in diminutive polyps at 66.7% (24 out of 36 polyps). Histopathological examination of the polyps from which EC images were acquired showed 22 (51%) nonneoplastic and 21 (49%) neoplastic polyps. EndoBRAIN (EC-CAD) detects a polyp as neoplastic or nonneoplastic using EC in real-time (►Supplementary Video S1). Nonneoplastic polyp on EC showed narrow serrated lumina and dense pattern of small roundish nodules (►Fig. 1). Neoplastic polyp showed slit like smooth lumina and regular pattern of fusiform or roundish nuclei (►Fig. 2). The sensitivity, speciﬁcity, PPV, NPV, and accuracy of endoscopist in identifying a neoplastic polyp based on EC were 90.48% (95% conﬁdence interval [CI], 69.2– 98.8), 81.81% (95% CI, 59.7–94.8), 82.61% (95% CI, 65.95– 92.1), 90% (95% CI, 70.36–97.15), and 86% (95% CI, 72.07– 94.70), respectively, with a positive likelihood ratio of 4.98 and negative likelihood ratio of 0.12. The sensitivity, speciﬁcity, PPV, NPV, and accuracy of EndoBRAIN in identifying a neo- plastic polyp based on EC were 100%, 81.82% (95% CI, 59.7– 94.8%), 84%(95% CI, 68.4–92.72%), 100%, and 90.7% (95% CI, 77.86–97.41%), respectively, with a positive likelihood ratio of 5.5and negativelikelihoodratioof0.Thesensitivityand NPVof EndoBRAIN were signiﬁcantly better than that of endoscopist (p < 0.05). Though diagnostic accuracy is more with Endo- BRAIN, it did not reach statistical signiﬁcance (p ¼ 0.5). Methods Speci- ﬁcity and PPV were similar in both groups (►Table 2). Fig. 1 Nonneoplastic polyp. (A) White light imaging showing sessile polyp of size 5 mm (Paris 1s). (B) Narrow band imaging (NBI) showing sessile polyp with invisible vascular pattern with regular dots (Japan NBI expert team type 1). (C and D) Endocytoscopy with NBI and stain showing narrow serrated lumina and dense pattern of small roundish nodules (EC1b). Fig. 2 Neoplastic polyp. (A) White light imaging showing sessile polyp of size 5 mm (Paris 1s). (B) Narrow band imaging (NBI) showing sessile polyp with regular vessel distribution and caliber with regular surface (Japan NBI expert team type 2A). (C and D) Endocytoscopy with NBI and stain showing slit-like smooth lumens with uniform fusiform or roundish nuclei (EC2). Journal of Digestive Endoscopy Vol. 14 No. 1/2023 © 2022. The Author(s). Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN Katrevula et al. 6 The image acquisition was difﬁcult in diminutive polyps at high magniﬁcation because minimal movement at the time of freezing the image can lead to false results. Images were also difﬁcult to obtain polyps located in traditionally difﬁcult locations like hepatic and splenic ﬂexure. Better bowel preparation, higher proce- dure volumes, and strategies such as examining the polyp at 12’0 clock position are few ways to improve the image acquisition in our experience. Whereas with the other parameters concerned such as EndoBRAIN sensitivity, speciﬁcity, and PPV values were also reported and they were observed to be better than values from endoscopists. Reports from our study were almost similar with the results published by the team who devel- oped the EndoBRAIN itself. However, the sensitivity, speci- ﬁcity, PPV values of our study and EndoBRAIN team were reported to be 100, 81.82, 84 and 96.9, 100, 100%, respec- tively.15 Reports from Shin et al also showed almost similar results as our study in both stained EC and EC-NBI.14 Overall, results from our study suggest that the sensitivity and NPV are statistically signiﬁcant and better in EndoBRAIN than that of endoscopists group suggesting the efﬁciency of the EndoBRAIN and its unlikely nature to miss a neoplastic polyp. In addition, EndoBRAIN is a good alternate to conventional methods in terms of cost-effectiveness, time-saving, and the trauma involved throughout the process. In future, these AI- based diagnostic systems like EndoBRAIN can be a game changer in reducing the unnecessary surgeries/resections because of their high accuracy, NPV, and speciﬁcity. In future, these AI systems also have a high potential to transform clinical endoscopic practice positively forever over the exist- ing conventional procedures. Studyby Mori etalisoneofthemostsigniﬁcantand theﬁrst benchmark study ever conducted using EndoBRAIN to clarify the value of an AI-assisted colonoscopy system in identifying cancerous lesions under the strictlycontrolled environment.17 Another important parameter to be considered as a bench- mark for optical diagnosis in adoption of AI systems as a clinical decision support device for diminutive polyp manage- ment is PIVI thresholds. Higher the accuracy of optical polyp diagnosis, higher will be the PIVI acceptance. Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN Katrevula et al. 6 6 Table 2 Comparison of evaluation of endocytoscopy between endoscopist and EndoBRAIN (n ¼ 43) Endoscopist (confidence interval) EndoBRAIN (EC-CAD) (confidence interval) p-Value Sensitivity 90.48% (69.2–98.8) 100% 0.03 Speciﬁcity 81.81% (59.7–94.8) 81.82% (59.7–94.8) 0.9 Positive predictive value 82.61% (65.95–92.1) 84% (68.4–92.72) 0.86 Negative predictive value 90% (70.36–97.15) 100% 0.03 Accuracy 86% (72.07–94.70) 90.7% (77.86–97.41) 0.5 Abbreviation: EC-CAD, endocytoscopy-computer-aided diagnosis. Table 2 Comparison of evaluation of endocytoscopy between endoscopist and EndoBRAIN (n ¼ 43) Abbreviation: EC-CAD, endocytoscopy-computer-aided diagnosis. study, an improvement in accuracy of EndoBRAIN (90.7%) over endoscopist (86.05%) was also observed to be same as Jin et al where the use of CADx has improved the overall accuracy of optical polyp diagnosis from 82.5 to 88.5% (p < 0.05).18 With advancement of technology like this, many nonexpert endo- scopists around the world can now easily achieve accuracy levels sufﬁcient to meet the PIVI threshold. nonhistological polyp evaluation with better accuracy and reduced intra- and interobserver variability.13 To overcome the limitations with the existing CADx systems, Kudo et al14 and Mori et al15 have collaborated, designed, and developed an advanced novel AI technology- based CADe (EndoBRAIN-EYE) and CADx tool—EndoBRAIN to help the surgeons in real-time (in-vivo) to differentiate the nonneoplastic lesions from neoplastic and help in avoiding unnecessary resection.16 To the best of our knowledge, this is the ﬁrst of its kind study in India to use EndoBRAIN (an AI software tool) to differentiate neoplastic versus nonneo- plastic polyp in real-time during a colonoscopy. Another major parameter to consider is NPV, where the NPV results (100%) from our study were observed to be better than multiple studies from the literature allowing diminu- tive hyperplastic polyps to be left in situ without a patholog- ical diagnosis. In studies conducted by Mori et al17 and Shin et al,14 the NPV in both the stain mode and NBI mode was observed to be only more than or equal to 93%. The ﬁrst step during the procedure is identiﬁcation and differentiation ofcancerouslesionsfrom noncancerous lesions invivo usinghigh-quality imagesacquired and analyzed inreal time. The image acquisition rate in our study (78.2%) was highlyefﬁcientand it wasinlinewithalarge multicenterstudy conducted by Mori et al (83.6%).17 This parameter is important in the context that unless a high-quality EC image is acquired, EndoBRAIN does not give us an output. Journal of Digestive Endoscopy Vol. 14 No. 1/2023 © 2022. The Author(s). Discussion In the past decade, development in AI and its applications in the medical ﬁeld were exponential. Being third-leading malignancy, technical, operator, and human dependent lim- itations are missing out on a signiﬁcant proportion of polyps during colonoscopy in CRC patients. These errors ultimately Table 1 Baseline characteristics Total no of polyps 55 Size Diminutive polyps 36 (65.4%) Polyps of size 5 mm–1cm 10 (18.1%) Polyps of size > 1 cm 9 (16.3%) Location Rectum 21 (38.1%) Left colon 15 (27.2%) Right colon 19 (34.5%) Image acquisition rate 43/55 (78.2%) Image acquisition rate from diminutive polyps 24/36 (66.7%) Histopathology of polyps from which endocytoscopy images acquired (n ¼ 43) Hyperplastic 20 Inﬂammatory 2 Low-grade adenoma 16 High-grade adenoma 5 Table 1 Baseline characteristics Fig. 2 Neoplastic polyp. (A) White light imaging showing sessile polyp of size 5 mm (Paris 1s). (B) Narrow band imaging (NBI) showing sessile polyp with regular vessel distribution and caliber with regular surface (Japan NBI expert team type 2A). (C and D) Endocytoscopy with NBI and stain showing slit-like smooth lumens with uniform fusiform or roundish nuclei (EC2). affect the patients and their overall CRC management. It was also reported that with each 1% increase in ADR, an equiva- lent 3% decrease in the subsequent risk of cancer was reported.10,11 In view of this, the highest level of accuracy is highly essential and much needed to deal with such unmet problems with minimal errors that can be feasible only with AI. AI has its own advantages in diagnosing the polyp characteristics easily, early, accurately, and economically than the existing conventional ex-vivo microscopic analysis methods.12 Considering the facts, the AI powered CADe and diagnosis (CADx) systems were developed to improve the Journal of Digestive Endoscopy Vol. 14 No. 1/2023 © 2022. The Author(s). Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN Katrevula et al. Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN Katrevula et al. 7 Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN Katrevula et al. 7 required to conclude on “do not resect” strategy. Third, objective assessment of additional time required to perform procedure and cost-effective analysis was not performed. assessment of the histology of diminutive colorectal polyps. Gastrointest Endosc 2011;73(03):419–422 5 Mori Y, Kudo SE, Wakamura K, et al. Novel computer-aided diagnostic system for colorectal lesions by using endocytoscopy (with videos). Gastrointest Endosc 2015;81(03):621–629 with videos). Gastrointest Endosc 2015;81(03):621–629 Conclusion 6 Misawa M, Kudo SE, Mori Y, et al. Characterization of colorectal lesions using a computer-aided diagnostic system for narrow- band imaging endocytoscopy. Gastroenterology 2016;150(07): 1531–1532.e3 Optical diagnosis using EC and EC-CAD has a potential role in predicting the histopathological diagnosis. The diagnostic performance of CAD seems to be better than endoscopist using EC for predicting neoplastic lesions. Large-scale data analysis in Indian population is needed prior to community practice. 7 Kudo SE, Misawa M, Wada Y, et al. Endocytoscopic microvascu- lature evaluation is a reliable new diagnostic method for colo- rectal lesions (with video). Gastrointest Endosc 2015;82(05): 912–923 8 Kudo SE, Wakamura K, Ikehara N, Mori Y, Inoue H, Hamatani S. Diagnosis of colorectal lesions with a novel endocytoscopic classiﬁcation - a pilot study. Endoscopy 2011;43(10):869–875 Conﬂict of Interest None declared. Conﬂict of Interest None declared. 14 Kudo SE, Misawa M, Mori Y, et al. Artiﬁcial intelligence-assisted system improves endoscopic identiﬁcation of colorectal neo- plasms. Clin Gastroenterol Hepatol 2020;18(08):1874–1881.e2 Supplementary Video S1 9 Mori Y, Kudo SE, East JE, et al. Cost savings in colonoscopy with artiﬁcial intelligence-aided polyp diagnosis: an add-on analysis of a clinical trial (with video). Gastrointest Endosc 2020;92(04): 905–911.e1 Title slide: Real-world experience of artificial intelli- gence-assisted endocytoscopy using EndoBRAIN—An observational study from a tertiary care center. Nonneoplastic polyp showing white-light imaging, narrow band imaging (NBI), endocytoscopy with NBI and methylene staining. Neoplastic polyp showing white-light imaging, NBI, endocytoscopy with NBI, and methylene staining. Online content including vid- eo sequences viewable at: https://www.thieme- connect.com/products/ejournals/html/10.1055/s- 0042-1758535. 10 Corley DA, Jensen CD, Marks AR, et al. Adenoma detection rate and risk of colorectal cancer and death. N Engl J Med 2014;370(14): 1298–1306 11 Nazarian S, Glover B, Ashraﬁan H, Darzi A, Teare J. Diagnostic accuracy of artiﬁcial intelligence and computer-aided diagnosis for the detection and characterization of colorectal polyps: systematic review and meta-analysis. J Med Internet Res 2021; 23(07):e27370 12 Choi SJ, Kim ES, Choi K. Prediction of the histology of colorectal neoplasm in white light colonoscopic images using deep learning algorithms. Sci Rep 2021;11(01):5311 algorithms. Sci Rep 2021;11(01):5311 13 Gross S, Trautwein C, Behrens A, et al. Computer-based classiﬁca- tion of small colorectal polyps by using narrow-band imaging with optical magniﬁcation. Gastrointest Endosc 2011;74(06): 1354–1359 Real-World Experience of AI-Assisted Endocytoscopy Using EndoBRAIN Katrevula et al. 6 Based on the PIVI threshold, anyone of the paradigm will be opted in the end clinically—“resect and discard” or “leave in situ.” In the present study, the accuracy rate of EndoBRAIN was 90.7%, exceeding the initiative threshold of more than 90% for the “resect and discard” strategy as proposed by the American Society for Gastrointestinal Endoscopy.4 Diagnostic accuracy reports of our study (90.7%) were observed to be similar to the reports obtained from Misawa et al6 (87.8%) and Shin et al.14 In our There are some limitations to our study. First, sample size is very small; hence, it is difﬁcult to generalize the ﬁndings to community. Second, sessile serrated adenoma (SSA) that appears similar to hyperplastic polyps on digital chromoen- doscopy was not studied in our study. As such the incidence of SSA is low and predominant distal location of polyps in our study may be the reason for not having SSA. Further polyp surveillance studies with EndoBRAIN involving SSA are Journal of Digestive Endoscopy Vol. 14 No. 1/2023 © 2022. The Author(s). Journal of Digestive Endoscopy Vol. 14 No. 1/2023 © 2022. The Author(s). References 15 Mori Y, Kudo SE, Misawa M, et al. Artiﬁcial intelligence-assisted colonic endocytoscopy for cancer recognition: a multicenter study. Endosc Int Open 2021;9(07):E1004–E1011 1 Three-Year Report of Population Based Cancer Registries Inci- dence and Distribution of Cancer (Report of 25 PBCRs in India). 2009 [cited 2022 Jan 10]; Accessed October 15, 2022, at: www. ncrpindia.org 16 Mori Y, Neumann H, Misawa M, Kudo SE, Bretthauer M. Artiﬁcial intelligence in colonoscopy - now on the market. What’s next? J Gastroenterol Hepatol 2021;36(01):7–11 2 Gao J, Jiang Q, Zhou B, Chen D. Convolutional neural networks for computer-aided detection or diagnosis in medical image analysis: an overview. Math Biosci Eng 2019;16(06):6536–6561 17 Mori Y, Kudo SE, Misawa M, et al. Real-time use of artiﬁcial intelligence in identiﬁcation of diminutive polyps during colo- noscopy: a prospective study. Ann Intern Med 2018;169(06): 357–366 3 Ebigbo A, Palm C, Probst A, et al. A technical review of artiﬁcial intelligence as applied to gastrointestinal endoscopy: clarifying the terminology. Endosc Int Open 2019;7(12):E1616–E1623 18 Jin EH, Lee D, Bae JH, et al. Improved accuracy in optical diagnosis of colorectal polyps using convolutional neural networks with visual explanations. Gastroenterology 2020;158(08):2169–2179.e8 18 Jin EH, Lee D, Bae JH, et al. Improved accuracy in optical diagnosis of colorectal polyps using convolutional neural networks with visual explanations. Gastroenterology 2020;158(08):2169–2179.e8 4 Rex DK, Kahi C, O’Brien M, et al. The American Society for Gastrointestinal Endoscopy PIVI (Preservation and Incorporation of Valuable Endoscopic Innovations) on real-time endoscopic
W2313096699.txt	https://zenodo.org/records/2108718/files/article.pdf	de		Ein Curs der medicinischen Psychologie mit Bezug auf Behandlung und Erziehung der angeboren Schwachsinnigen	Archiv für Psychiatrie und Nervenkrankheiten	1,906	public-domain	214	Ein Curs der medieinischen Psychologie mit Bezug auf Behandlung und Erziehung der angeboren Schwachsinnigen 1) wird yon Mont~g den 2. bis S a m s t a g den 7. April 1906 in Giessen (Klinik ftir psyehisehe und nerv6so Krankheiten) stattfinden. Der Curs wird folgende Themata umfassen: 1. Die versehiedenen Formen der Idiotic. 2. Ursachenforschung~ Prophylaxe und Therapie im Gebiet der Idiotio. 3. Untersuehung der Sch~idelabnormitiiten mit practisehen Uebungen. 4. Medicinische Psyehologie mit Bezug auf Behandlung und Erziehung der angeboren Schwaehsinnigen mit psyehophysisehen Uebungen. 5. Experimentelle Didactik mit Bezug auf die angeboren Sehwaehsinnigen. 6. Das Hilfsschulwesen. 7. Die Zwangserziehung. 8. Die str~freehtlichon Beziehungen des angeborenen Sehwachsinnes. 9. Jugendliehes Verbreeherthum. 10. Der angeborene Sehw~ehsinn im Militiirdienst. 11. Die Anstalten fiir Schwaehsinnige etc. mit Besichtigungen. Als Vortr~ge~de werden ausser dem Unterzeichneten und Herrn Privatdoeenten Dr. Dannemann-Giessen nech Herr Prof, Dr. Weyg~ndt-Wiirzburg und Herr Seminarlehrer L ay-Karlsruhe mitwirken. Das genauere Programm der Vortrgge und Uebungen sell Ende Februar 1906 versandt werden. Die Einsohreibegebiihr wird je n~ch den Kosten tier Vorbereitung etwa 10 his 20 Mark betragen. Zu dem Curs sind alle an der Behandlung und Erziehung der angeboren Schwaehsinnigen ernsthaft interessirten Personen, besonders Aerzte und L e h r e r eingeladen. Giessen, December 1905. Prof. Dr. Sommer, 1) Vergl. Psychiatr.-Neurol. Wochensehr. 1905, No. 20 und '23.
https://openalex.org/W2999884420	https://europepmc.org/articles/pmc6970486?pdf=render	English	null	Quantitative Evaluation of Articular Involvement of Posterior Malleolus Associated with Operative Indication: A Comparative Study of Six Methods Based on Radiography and CT	BioMed research international	2,020	cc-by	7,506	Kun Zhang,1 Xiaoyang Jia,1 Minfei Qiang,2 Song Chen,1 Shuguang Wang,1 Dongdong Wang,1 and Yanxi Chen 2 Kun Zhang, Xiaoyang Jia, Minfei Qiang, Song Chen, Shuguang Wang, Dongdong Wang,1 and Yanxi Chen 2 1Department of Orthopedic Trauma, East Hospital, Tongji University School of Medicine, 150 Jimo Rd, 200120 Shanghai, China 2Department of Orthopaedic Surgery, Zhongshan Hospital, Fudan University, 180 Fenglin Road, Shanghai 200032, China 1Department of Orthopedic Trauma, East Hospital, Tongji University School of Medicine, 150 Jimo Rd, 200120 Shanghai, China 2Department of Orthopaedic Surgery, Zhongshan Hospital, Fudan University, 180 Fenglin Road, Shanghai 200032, China Correspondence should be addressed to Yanxi Chen; cyxtongji@126.com Received 14 October 2019; Revised 9 December 2019; Accepted 13 December 2019; Published 8 January 2020 Academic Editor: Sae Hoon Kim Copyright © 2020 Kun Zhang et al. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Te aim of this study was to compare the values of six methods in measuring the involvement of posterior malleolus and to demonstrate the reliability and reproducibility of each method. Tree independent orthopaedic surgeons, retrospectively, measured 106 cases. Te diﬀerence between the six methods was analyzed using Bonferroni-corrected paired t-tests after one-way ANOVA. Te agreement between the six methods was analyzed using Bland–Altman analysis. Te intraclass correlation co- eﬃcient (ICC) was used to assess intraobserver reproducibility and interobserver reliability. Signiﬁcant diﬀerences were observed between values of any two of the six measurement methods (P < 0.0033), except between any two of the plane radiograph linear, axial CT linear, sagittal CT linear, and 3D CT linear. Te Bland–Altman plots demonstrated poor agreement between values of any two of the six methods. Te lowest intraobserver reproducibility was 0.46 (moderate) for resident surgeon using plain radiographs. Te intraobserver reproducibility for three surgeons using two-dimensional (2D) and 3D images was almost perfect (ICC, 0.82–0.96). Te lowest interobserver reliability was 0.41 (moderate) between chief and attending surgeon using plain radiographs, and it improved to almost perfect (ICC, 0.81–0.95) with the use of 3D CT images. Te standard error of measurement showed almost the same results as ICC values. Te existing operative indications which were determined based on plain radiography are neither reliable nor suitable for other measurement methods. Both 3D linear and 3D surface measurement methods are reliable and reproducible in measuring posterior fragment involvement, and experience is not so crucial. Hindawi BioMed Research International Volume 2020, Article ID 6745626, 9 pages https://doi.org/10.1155/2020/6745626 Hindawi BioMed Research International Volume 2020, Article ID 6745626, 9 pages https://doi.org/10.1155/2020/6745626 Hindawi BioMed Research International Volume 2020, Article ID 6745626, 9 pages https://doi.org/10.1155/2020/6745626 Kun Zhang,1 Xiaoyang Jia,1 Minfei Qiang,2 Song Chen,1 Shuguang Wang,1 Dongdong Wang,1 and Yanxi Chen 2 Operative indications for posterior malleolar fractures need to be redeﬁned based on the 3D measurement method. 1. Introduction A total of 235 patients met the inclusion criteria. Fifty-two patients with Haraguchi type II posterior malleolar fractures were excluded. Twenty-ﬁve patients were excluded for lack of CT data. Tirty-nine patients were excluded because posterior malleolar fractures could not be identiﬁed on lateral radiographs (Figure 1). Twenty-ﬁve patients were excluded because measurement could not be performed on 3D CT images. Measurement could not be performed because of severe comminution of articular surface or defect of the posterior malleolar frag- ments or very small shell-shaped Haraguchi type III frac- tures. Measurements could not be performed on both lateral radiographs and 3D CT images in 12 patients mentioned above. Te remaining 106 patients were ﬁnally analyzed (Table 1). All cases were classiﬁed based on CT scans according to AO/OTA (Arbeitsgemeinschaft f¨ur Osteo- synthesefragen/Orthopaedic Trauma Association) and Lauge-Hansen Classiﬁcation. Generally, the size of posterior malleolar fragments was assessed using plain radiographs. However, radiographs may be restricted by the orientation of the foot because of pain and swelling in the acute injury. Surgeons frequently failed to identify posterior malleolar fractures and precisely esti- mate the size of posterior malleolar fragments when interpreting plain radiographs [15–19]. Axial CT was also used to measure the percentage of posterior malleolar fractures [16, 20]. But two-dimensional (2D) CT still has limitations in assessment. Te selection of the observation plane was also aﬀected by the position of the ankle and the experience of observers. A study tried to use three-dimen- sional (3D) CT to assess the articular involvement of pos- terior malleolar fractures [21]. But the method adopted, which needs to integrate several sets of software, was complex and time-consuming. Te method could hardly be applied to measure large sample and be replicated by other scholars. So up to now, 3D CT has not been extensively applied in posterior malleolar fracture evaluation nor has it been reported in a study on large sample [22, 23]. With the upgrading of computer technology, an eﬃcient system for computer-assisted preoperative planning has been devel- oped [24–26]. Te use of computer technology enables multilevel and multiangle evaluation of fracture planes. Surgeons can also perform a virtual operation eﬃciently and conveniently, including reducing the fracture fragments and selecting a suitable internal ﬁxation device [24, 25]. With this technology, detailed evaluation and accurate measurement of posterior malleolar fracture may be promoted to a higher level. 1. Introduction or more, the normal dynamics of the joint were disrupted [10]. In the meanwhile, some scholars oﬀered diﬀerent operative indications. A biomechanical study involving 16 cadaveric specimens suggested operation when >33% of the joint is involved based on their ﬁndings that displaced posterior malleolar fractures produce a signiﬁcant decrease in contact area with 33% or greater involvement of the joint [11]. Other indications include >30% of the joint is involved with >2 mm displacement after closed reduction of the ankle [12], >20% of the joint is involved [13], >10% of the joint is involved [14], etc. Although the indications were diﬀerent, they were all based on a speciﬁc percentage. Terefore, precise estimate of the articular involvement of posterior Ankle fractures are among the most common lower limb fractures, accounting for about 9% of all fractures [1]. Posterior malleolar fractures comprise 14–42% of all ankle fractures [2]. Recently, many scholars recommended ana- tomical reduction and internal ﬁxation for treatment of ankle fractures [3–5]. But still no consensus was achieved on the operative indication for posterior malleolar fractures [6]. Several scholars suggested operative ﬁxation of posterior malleolar fractures when more than 25% of the tibial plafond is involved [7–9]. A biomechanical study also demonstrated that with fractures constituting 25% of the lateral joint line Ankle fractures are among the most common lower limb fractures, accounting for about 9% of all fractures [1]. Posterior malleolar fractures comprise 14–42% of all ankle fractures [2]. Recently, many scholars recommended ana- tomical reduction and internal ﬁxation for treatment of ankle fractures [3–5]. But still no consensus was achieved on the operative indication for posterior malleolar fractures [6]. Several scholars suggested operative ﬁxation of posterior malleolar fractures when more than 25% of the tibial plafond is involved [7–9]. A biomechanical study also demonstrated that with fractures constituting 25% of the lateral joint line BioMed Research International 2 malleolar fractures is crucial for orthopaedic surgeons to decide clinical treatment and judge prognosis [9]. fragment, which was conﬁrmed by CT or surgery, between the ages of 20 and 75. Te exclusion criteria were patients with pathologic fractures, Haraguchi type II or III posterior malleolar fractures or without standard lateral radiographs, or 16-row spiral CT examinations. Patients were also ex- cluded if posterior malleolar fractures could not be identiﬁed on lateral radiographs, or measurement could not be per- formed on 3D CT images. 1. Introduction If the measurement method is not reliable or repro- ducible, large sample and multicenter studies cannot be carried out, and the operative indications which were summarized from the unreliable results will have no ref- erence value. Terefore, the comparison of the results, re- liability, and reproducibility of various methods for measuring the articular involvement of posterior malleolar fractures is of great signiﬁcance. To our knowledge, little literature is available comparing diﬀerent kinds of methods in measuring the articular involvement of posterior mal- leolar fractures. 2.2. Image Evaluation. Te X-ray and CT scanning data (DICOM 3.0 format) of all patients were collected. Te data of all research subjects were ﬁrstly uploaded to picture ar- chiving and communication system (PACS) of the hospital and then imported into the computer-assisted orthopaedic research system (SuperImage orthopaedics edition 1.1, Cybermed Ltd, Shanghai, China) [24]. All the cases were evaluated by 3 independent ortho- paedic surgeons (one chief surgeon with 18 years of image reading and clinical experience, one attending surgeon with 9 years of experience, and one resident surgeon with 5 years of experience). Te examiners were asked to measure using diﬀerent methods in 6 phases. Tere was an interval of 2 weeks between each phase. All observers were blinded to the others’ analysis. Te measurements were repeated by three observers at an interval of 4 weeks. Te objective of the study was (1) to explore the dif- ference of 6 methods (plain radiograph linear, axial CT linear, sagittal CT linear, axial CT plane, 3D CT linear, and 3D CT surface) in measuring the involvement of Haraguchi type I posterior malleolar fragment based on a computer- assisted preoperative planning system and large sample and (2) to demonstrate the reliability and reproducibility of each method. In phase one, observers were asked to measure lateral radiographs of all the cases (Figure 2(a)) [14]. In phase two, axial CTimages were used. Te measurement was performed at the level of the tibial plafond (Figure 2(b)). In phase three, sagittal reconstruction images were used. Te measurement was performed on the section of the ﬁbular notch (Figure 2(c)). In the ﬁrst three phases, the size of posterior malleolar fragment was measured as the percentage of the involved distal tibial articular surface (Figure 1(a)) [14, 15, 27]. 2. Materials and Methods 2.1. Patient Demographic Data. Te research project was approved by the Ethics Committee of our hospital (Ethics number 2012–020), and informed consent was obtained. Trauma patients were retrospectively reviewed at our hos- pital between May 2009 and December 2015. Te inclusion criteria were ankle fracture with posterior malleolar In phase four, axial CT images were used. Te mea- surement was performed at the level of the tibial plafond. Te medial malleolus area should be revealed at the same level. Te posterior malleolar fragment area and the remaining cross-sectional area (avoid medial malleolus area) of the tibia were delineated and measured. Te ratio was then BioMed Research International 3 3 (a) (b) (c) (d) Figure 1: (a) Traditional measurement method on lateral plain radiographs is shown schematically. Te percentage of involved articular surface is ii/i + ii. (b) A case would be excluded if the fracture line could not be identiﬁed and therefore measurement could not be performed. Images showed the lateral plain radiograph of a 23-year-old male patient with Lauge-Hansen supination-external rotation type 3 ankle fracture. Te patient was excluded because the posterior malleolar fracture line was diﬃcult to be identiﬁed on the lateral radiograph. (c) On the sagittal multiplanar reconstruction CT image of the same patient, the posterior malleolar fracture line could be clearly identiﬁed. (d) A large diﬀerence in the articular involvement of posterior malleolar fractures is shown on the sagittal CT images (3D pseudocolor display mode) selected by diﬀerent observers. (a) (b) (c) (d) (b) (a) (c) (d) Figure 1: (a) Traditional measurement method on lateral plain radiographs is shown schematically. Te percentage of involved articular surface is ii/i + ii. (b) A case would be excluded if the fracture line could not be identiﬁed and therefore measurement could not be performed. Images showed the lateral plain radiograph of a 23-year-old male patient with Lauge-Hansen supination-external rotation type 3 ankle fracture. Te patient was excluded because the posterior malleolar fracture line was diﬃcult to be identiﬁed on the lateral radiograph. (c) On the sagittal multiplanar reconstruction CT image of the same patient, the posterior malleolar fracture line could be clearly identiﬁed. (d) A large diﬀerence in the articular involvement of posterior malleolar fractures is shown on the sagittal CT images (3D pseudocolor display mode) selected by diﬀerent observers. Table 1: Demographic data of the patients (n 106). 2. Materials and Methods Characteristic Value Age, y (range) 47.3 (21 to 75) Male, n (%) 59 (55.7) Left, n (%) 45 (42.5) AO/OTA classiﬁcation, n (%) 44-B 75 (70.8) 44-C 31 (29.2) Lauge-Hansen classiﬁcation, n (%) Supination-external rotation Type 3 31 (29.2) Type 4 43 (40.6) Pronation-external rotation type 4 27 (25.5) Pronation-abduction type 3 5 (4.7) Table 1: Demographic data of the patients (n 106). posterior malleolar fragment area to the total area of the tibial plafond was calculated (Figure 4(b)). 2.3. Statistical Analysis. Statistical analysis was performed using SPSS 18.0 (SPSS Inc, Chicago, IL, USA) and MedCalc 15.10.0 (MedCalc Software bvba, Ostend, Belgium). Te diﬀerence between the six measurement methods was an- alyzed using Bonferroni-corrected paired t-tests after one- way ANOVA. Only those P < 0.05/15 0.0033 were con- sidered statistically signiﬁcant. Te agreement between the six measurement methods was analyzed using Bland– Altman analysis. Te intraclass correlation coeﬃcient (ICC, two-way mixed, single consistency) was used to assess intraobserver reproducibility and interobserver reliability [28]. Te standard error of measurement (SEM) of the re- peated measurements was calculated to determine the size of the measurement error. Te SEM could be estimated as the square root of the mean square error term from the two-way random-eﬀect ANOVA [29]. calculated as the fragment area to the total cross-sectional area of the tibial plafond (Figure 2(d)). p g In phase ﬁve and six, 3D CT images were used. To perform 3D measurement, 3D images were ﬁrstly generated by surface shaded display (SSD) algorithm with a recon- struction interval of 0.625 mm. Secondly, all bones and fracture fragments were distinguished using the built-in interactive intelligent segmentation module (Figure 3). We hid the talus and turned over the distal tibial articular surface. In phase ﬁve, the size of posterior malleolar frag- ment was measured as the percentage of the involved distal tibial articular surface (Figure 4(a)). In phase six, the surface boundary of the posterior malleolar fragments and residual articular surface was delineated manually. Each surface area (along the curved plane of the distal tibia summing the surface areas from the separate transverse images) was calculated automatically by the software. Te ratio of the 3. Results 3.1. Articular Involvement Determined Using Six Measure- ment Methods. One-way ANOVA showed a signiﬁcant diﬀerence between ratios determined using six measurement methods (F 31.379, P < 0.001). Signiﬁcant diﬀerences were observed between values of any two of the six measurement methods (P < 0.0033), except between plane radiograph linear and axial CT linear (t 1.574, P 0.118), between plane radiograph linear and sagittal CT linear (t 1.471, P 0.144), between plane radiograph linear and 3D CT linear (t 2.339, P 0.021), between axial CT linear and sagittal CT linear (t 0.468, P 0.641), between axial CT BioMed Research International 4 (a) (b) (c) (d) Figure 2: Measurement of the articular involvement of posterior malleolar fractures on lateral radiographs and 2D CT images is shown. (a) Measurement on plain radiographs is shown practically. If we assume the horizontal distance between the front and back points of residual articular surface at the distal end of tibia is i and the horizontal distance between the posterior malleolar fracture line and the back point of articular surface is ii, then the percentage of involved articular surface is ii/i + ii. (b) When using axial CT linear measurement method, if we assume the distance between the front and back points of the lateral margin of residual articular surface at the level of tibial plafond is i and the distance between the posterior malleolar fracture line and the posterior point of lateral margin of the fragment is ii, then the percentage of involved articular surface is ii/i + ii. (c) When using sagittal CT linear measurement method, the process is similar to that on lateral radiographs. Te percentage of involved articular surface is ii/i + ii. (d) When using axial CT plane measurement method, the boundary of the medial malleolus should be revealed at the same level. Te boundary of the posterior malleolar fragment plane (ii) and residual plane (avoid medial malleolus area) (i) was delineated manually. Each area was calculated automatically by the software. Ten the ratio of the involved fragment area is ii/i + ii. (c) (a) (b) (d) (d) (b) (c) (a) Figure 2: Measurement of the articular involvement of posterior malleolar fractures on lateral radiographs and 2D CT images is shown. (a) Measurement on plain radiographs is shown practically. 3. Results Compared to 3D CT surface, the articular involvement measured using axial CTplane increased by 17.3% (Table 2). linear and 3D CT linear (t 2.567, P 0.012), and between sagittal CT linear and 3D CT linear (t 1.885, P 0.062). Te mean diﬀerence between plain radiograph linear and 3D CT surface was 8.58% (t 10.564, P < 0.0033). Te mean ratio determined using 3D CT surface was the smallest (16.0 ± 8.4%) and that of 3D CT linear was the largest (26.7 ± 8.3%) (Table 2). Compared to 3D CT linear, the articular involvement measured using plain radiographs and CT linear decreased by 8% and about 3%, respectively. Compared to 3D CT surface, the articular involvement measured using axial CTplane increased by 17.3% (Table 2). Te Bland–Altman plots demonstrated poor agreement between values of any two of the six measurement methods. Within the range of 95% limits of agreement (95% LoA), maximum diﬀerence of more than 10% could be observed on most plots except for axial CT linear-sagittal CT linear, axial CT linear-3D CT linear, and axial CT plane-3D CT surface plot. Te axial CT linear-sagittal CT linear plot showed the lowest mean diﬀerence: 0.21% (95% LoA −8.82 to 9.24%; linear and 3D CT linear (t 2.567, P 0.012), and between sagittal CT linear and 3D CT linear (t 1.885, P 0.062). Te mean diﬀerence between plain radiograph linear and 3D CT surface was 8.58% (t 10.564, P < 0.0033). Te mean ratio determined using 3D CT surface was the smallest (16.0 ± 8.4%) and that of 3D CT linear was the largest (26.7 ± 8.3%) (Table 2). Compared to 3D CT linear, the articular involvement measured using plain radiographs and CT linear decreased by 8% and about 3%, respectively. Compared to 3D CT surface, the articular involvement measured using axial CTplane increased by 17.3% (Table 2). Te Bland–Altman plots demonstrated poor agreement between values of any two of the six measurement methods. Within the range of 95% limits of agreement (95% LoA), maximum diﬀerence of more than 10% could be observed on most plots except for axial CT linear-sagittal CT linear, axial CT linear-3D CT linear, and axial CT plane-3D CT surface plot. Te axial CT linear-sagittal CT linear plot showed the lowest mean diﬀerence: 0.21% (95% LoA −8.82 to 9.24%; 3.2. Reliability and Reproducibility for Measurements. 3. Results If we assume the horizontal distance between the front and back points of residual articular surface at the distal end of tibia is i and the horizontal distance between the posterior malleolar fracture line and the back point of articular surface is ii, then the percentage of involved articular surface is ii/i + ii. (b) When using axial CT linear measurement method, if we assume the distance between the front and back points of the lateral margin of residual articular surface at the level of tibial plafond is i and the distance between the posterior malleolar fracture line and the posterior point of lateral margin of the fragment is ii, then the percentage of involved articular surface is ii/i + ii. (c) When using sagittal CT linear measurement method, the process is similar to that on lateral radiographs. Te percentage of involved articular surface is ii/i + ii. (d) When using axial CT plane measurement method, the boundary of the medial malleolus should be revealed at the same level. Te boundary of the posterior malleolar fragment plane (ii) and residual plane (avoid medial malleolus area) (i) was delineated manually. Each area was calculated automatically by the software. Ten the ratio of the involved fragment area is ii/i + ii. Figure 3: Tree-dimensional processing of thin-section axial CT images is shown. Tree-dimensional images (surface shaded display) were generated, and all component bones were distinguished by diﬀerent colors. Figure 3: Tree-dimensional processing of thin-section axial CT images is shown. Tree-dimensional images (surface shaded display) were generated, and all component bones were distinguished by diﬀerent colors. 95% conﬁdence interval for the bias −0.68 to 1.10%) (Ta- ble 3) (Figure 5(a)). Te axial CT plane-3D CT surface plot showed the smallest diﬀerence interval: 95% LoA was 3.02 to 8.55% (Figure 5(b)). linear and 3D CT linear (t 2.567, P 0.012), and between sagittal CT linear and 3D CT linear (t 1.885, P 0.062). Te mean diﬀerence between plain radiograph linear and 3D CT surface was 8.58% (t 10.564, P < 0.0033). Te mean ratio determined using 3D CT surface was the smallest (16.0 ± 8.4%) and that of 3D CT linear was the largest (26.7 ± 8.3%) (Table 2). Compared to 3D CT linear, the articular involvement measured using plain radiographs and CT linear decreased by 8% and about 3%, respectively. 3. Results For fracture size measurement, the lowest intraobserver re- producibility was 0.46 (moderate) for resident surgeon using plain radiographs, and it improved to substantial (ICC 0.77) for attending surgeon and to almost perfect (ICC 0.82) for chief surgeon using plain radiographs. Te intraobserver reproducibility for three surgeons using axial CT linear, sagittal CT linear, axial CT plane, 3D CT linear, and 3D CT surface was almost perfect (ICC, 0.82–0.96). For chief surgeon, the intraobserver repro- ducibility was almost perfect. And the ICC increased from 0.82 using plain radiographs to 0.94 using 3D CT surface (Table 4). Te Bland–Altman plots demonstrated poor agreement between values of any two of the six measurement methods. Within the range of 95% limits of agreement (95% LoA), maximum diﬀerence of more than 10% could be observed on most plots except for axial CT linear-sagittal CT linear, axial CT linear-3D CT linear, and axial CT plane-3D CT surface plot. Te axial CT linear-sagittal CT linear plot showed the lowest mean diﬀerence: 0.21% (95% LoA −8.82 to 9.24%; BioMed Research International 5 5 (a) (b) Figure 4: Measurement of the articular involvement of posterior malleolar fractures on 3D CTimage is shown. (a) When using 3D CT linear measurement method, on the lateral border of the tibial plafond, if we assume the distance between front and back points of the residual articular surface is i and the distance between posterior malleolar fracture line and back point is ii, then the percentage of involved articular surface is ii/i + ii. (b) When using 3D CT surface measurement method, the surface boundary of the posterior malleolar fragment (ii) and residual articular surface (i) was delineated manually. Each surface area was calculated automatically by the software. Ten, the ratio of the involved fragment area is ii/i + ii. (b) (a) (b) (a) Figure 4: Measurement of the articular involvement of posterior malleolar fractures on 3D CTimage is shown. (a) When using 3D CT linear measurement method, on the lateral border of the tibial plafond, if we assume the distance between front and back points of the residual articular surface is i and the distance between posterior malleolar fracture line and back point is ii, then the percentage of involved articular surface is ii/i + ii. (b) When using 3D CT surface measurement method, the surface boundary of the posterior malleolar fragment (ii) and residual articular surface (i) was delineated manually. 3. Results Each surface area was calculated automatically by the software. Ten, the ratio of the involved fragment area is ii/i + ii. Table 2: Fracture involvement determined using six measurement methods. Measurement method Fracture involvement in mean ± SD (%) 95% CI Plane radiograph linear 24.6 ± 7.9 23.1–26.1 Axial CT linear 26.0 ± 8.9 24.3–27.7 Sagittal CT linear 25.8 ± 7.8 24.3–27.3 Axial CT plane 18.8 ± 7.9 17.3–20.3 3D CT linear 26.7 ± 8.3 25.1–28.3 3D CT surface 16.0 ± 8.4 14.4–17.6 Table 2: Fracture involvement determined using six measurement methods. Table 3: Agreement between values of any two of the six measurement methods. Axial CT linear Sagittal CT linear Axial CT plane 3D CT linear 3D CT surface Plain radiograph linear 5.82, −10.75 to 22.38 −1.22, −17.89 to 15.46 −1.42, −19.69 to 16.84 −2.13, −20.52 to 16.25 8.58, −7.81 to 24.98 Axial CT linear 0.21, −8.82 to 9.24 7.24, 1.21 to 13.28 −0.71, −6.26 to 4.84 10.00, 3.88 to 16.13 Sagittal CT linear 7.03, −1.17 to 15.24 −0.92, −10.72 to 8.89 9.80, 2.87 to 16.73 Axial CT plane −7.95, −15.83 to −0.07 2.77, 3.02 to 8.55 3D CT linear 10.71, 2.72 to 18.71 Data are presented as mean diﬀerence (%), Bland–Altman 95% limits of agreement (mean diﬀerence ± 1.96 SD) (%). Data are presented as mean diﬀerence (%), Bland–Altman 95% limits of agreement (mean diﬀerence ± 1.96 SD) (%). Te lowest interobserver reliability was 0.41 (moderate) between chief and attending surgeon using plain radio- graphs, and it improved to substantial (ICC, 0.68–0.79) with the use of axial CT linear and to almost perfect (ICC, 0.81–0.95) with the use of 3D CT linear and 3D CT surface (Table 4). 4. Discussion Table 4: Inter- and intraobserver agreement for fracture involvement determined using six measurem Table 4: Inter- and intraobserver agreement for fracture involvement determined using six measurement methods. Examiner Plain radiographs Axial CT linear Sagittal CT linear Axial CT plane 3D CT linear 3D CT surface Chief-chief 0.82 (0.74–0.87) 0.89 (0.85–0.93) 0.88 (0.83–0.92) 0.92 (0.88–0.94) 0.91 (0.88–0.94) 0.94 (0.92–0.96) Attending-attending 0.77 (0.68–0.84) 0.92 (0.88–0.94) 0.85 (0.78–0.89) 0.96 (0.94–0.97) 0.91 (0.87–0.94) 0.95 (0.93–0.97) Resident-resident 0.46 (0.30–0.60) 0.90 (0.86–0.93) 0.82 (0.75–0.88) 0.94 (0.91–0.96) 0.88 (0.82–0.91) 0.91 (0.88–0.94) Chief-attending 0.41 (0.24–0.55) 0.79 (0.70–0.85) 0.73 (0.63–0.81) 0.66 (0.53–0.75) 0.84 (0.78–0.89) 0.95 (0.93–0.97) Chief-resident 0.43 (0.26–0.57) 0.68 (0.56–0.77) 0.52 (0.37–0.65) 0.70 (0.59–0.78) 0.81 (0.74–0.87) 0.95 (0.93–0.97) Attending-resident 0.67 (0.55–0.76) 0.70 (0.59–0.79) 0.59 (0.45–0.70) 0.60 (0.46–0.71) 0.83 (0.76–0.88) 0.90 (0.86–0.93) Values are presented as intraclass correlation coeﬃcient values with 95% conﬁdence interval; agreement was divided into 6 levels, including almost perfect (0.81 to 1.00), substantial (0.61–0.80), moderate (0.41 to 0.60), fair (0.21 to 0.40), slight (0.00 to 0.20), and poor (below 0.00). Values are presented as intraclass correlation coeﬃcient values with 95% conﬁdence interval; agreement was divided into 6 levels, including almost perfect (0.81 to 1.00), substantial (0.61–0.80), moderate (0.41 to 0.60), fair (0.21 to 0.40), slight (0.00 to 0.20), and poor (below 0.00). Values are presented as intraclass correlation coeﬃcient values with 95% conﬁdence interval; agreement was divided into (0.81 to 1.00), substantial (0.61–0.80), moderate (0.41 to 0.60), fair (0.21 to 0.40), slight (0.00 to 0.20), and poor (below er- and intraobserver reliability for fracture involvement determined using six measurement methods. Table 5: Inter- and intraobserver reliability for fracture involvement determined using six measurement methods. Examiner Plain radiographs Axial CT linear Sagittal CT linear Axial CT plane 3D CT linear 3D CT surface Chief-chief 4.66 3.01 3.25 2.37 2.59 2.01 Attending-attending 4.79 3.18 3.62 2.11 2.92 1.91 Resident-resident 7.49 3.13 3.99 2.25 3.11 2.68 Chief-attending 7.49 4.60 4.60 5.28 3.55 1.81 Chief-resident 7.51 5.35 6.28 4.80 3.69 1.98 Attending-resident 5.46 5.59 5.90 5.88 3.75 2.79 Values are presented as standard error of measurement; the larger the standard error of measurement, the lower the reliability. Table 5: Inter- and intraobserver reliability for fracture involvement determined using six measure linear, sagittal CT linear, and 3D CT linear. Poor agreement between values of any two of the six methods was observed. 4. Discussion Up to now, controversy still remains on the operative in- dication for posterior malleolar fractures [6], and various indications were proposed which were based on a speciﬁc percentage measured by plain radiographs [7–14]. However, the operative indications will have no reference value if the measurement method itself is not reliable or reproducible. Terefore, the comparison of the results, reliability, and reproducibility of various methods for measuring posterior malleolar articular involvement is of great signiﬁcance. In this study, signiﬁcant diﬀerences were observed between values of any two of the six measurement methods, except between any two of the plane radiograph linear, axial CT Te standard error of measurement showed almost the same results as ICC values. Te lowest intraobserver reli- ability was 7.49 for resident surgeon using plain radiographs, while the highest was 1.91 for attending surgeon using 3D CT surface. Te lowest interobserver reliability was 7.51 between chief and resident surgeon using plain radiographs, while the highest was 1.81 between chief and attending surgeon using 3D CT surface (Table 5). BioMed Research International 6 –15 –10 –5 0 5 10 15 20 10 20 30 40 50 60 Mean of axial CT linear and sagittal_CT linear Axial CT linear-sagittal_CT linear Mean 0.2 –1.96 SD –8.8 +1.96 SD 9.2 (a) –6 –4 –2 0 2 4 6 8 10 0 10 20 30 40 50 Mean of axial CT_plane and 3D_CT surface Axial CT_plane-3D_CT surface Mean 2.8 –1.96 SD –3.0 +1.96 SD 8.6 (b) Figure 5: (a) Bland–Altman plot of axial CT linear and sagittal CT linear shows the 95% limits of agreement. (b) Bland–Altman plot of axial CT plane and 3D CT surface shows the 95% limits of agreement. –15 –10 –5 0 5 10 15 20 10 20 30 40 50 60 Mean of axial CT linear and sagittal_CT linear Axial CT linear-sagittal_CT linear Mean 0.2 –1.96 SD –8.8 +1.96 SD 9.2 (a) –6 –4 –2 0 2 4 6 8 10 0 10 20 30 40 50 Mean of axial CT_plane and 3D_CT surface Axial CT_plane-3D_CT surface Mean 2.8 –1.96 SD –3.0 +1.96 SD 8.6 (b) (b) (a) Figure 5: (a) Bland–Altman plot of axial CT linear and sagittal CT linear shows the 95% limits of agreement. (b) Bland–Altman plot of axial CT plane and 3D CT surface shows the 95% limits of agreement. and intraobserver agreement for fracture involvement determined using six measurement methods. 4. Discussion However, linear measure- ment could only detect the injury degree of the ﬁbular notch and could not show the true injury area of the joint surface. Some studies suggested that morphology of the posterior malleolar fragment might be more important than size for clinical decision making [21, 31]. Haraguchi type II fracture was also considered to be posterior pilon fracture, which has more complex mechanism and more special morphological characteristics [32]. Terefore, patients with Haraguchi type II fractures were not enrolled in this study. However, the morphology of posterior malleolus is hard to be determined using plain radiographs. We can only account on unap- parent indications, such as a double-sign, to indicate the presence of Haraguchi type II fracture [31]. Tus, CT is useful not only in judging whether there is a fracture but also in identifying fracture morphology. Interestingly, the results revealed that the level of ex- perience was not so important to intraobserver reproduc- ibility when using 3D CT images. One may guess that more experience would result in better consistency in measuring fracture involvement. When based on plain radiographs, experience is exactly crucial. Te intraobserver reproduc- ibility of resident surgeon cannot be clinically accepted. However, using 3D CT images was not the case. Te level of intra- and interobserver agreement was similar among three surgeons. Te level of intraobserver reproducibility of res- ident surgeon was signiﬁcantly enhanced. In our study, the level of intraobserver reproducibility of resident surgeon and all interobserver reliability for size measurement when based on plain radiographs was not clinically accepted. Te previous study also concluded the impossibility to assess accurately the size of the posterior malleolar fragment on plain radiographs [17]. Meijer et al. [18] reported an interobserver agreement (ICC) of 0.61 on plain radiographs, which was similar to our results. A possible explanation of substantial bias is the obliqueness of the fracture line to the X-ray beam [10, 33]. Previous op- erative indications such as 25% and 33% were all set based on plain radiographs [7–9, 12–14]. Te clinical signiﬁcance of these indications is doubtful because the measurement method itself is unreliable and unrepeatable. Although the intraobserver reproducibility was almost perfect using 2D CT images (axial CT linear, sagittal CT linear, and axial CT plane), the interobserver reliability was substantial. Two main problems in measuring with CT images are important. 4. Discussion Tree-dimensional CT showed the highest intraobserver reproducibility and interobserver reliability among three imaging modalities, while plain radiography revealed the lowest. Haraguchi type I fractures are oblique, but Haraguchi type II fractures are almost parallel to the coronal plane [17]. Due to the diversity of fracture lines of posterior malleolus, we doubt if one lateral view can satisfy all situations. In addition, patients may not cooperate because of pain and swelling of the ankle in the acute injury. For diagnosing posterior malleolar fractures in our study, surgeons could not identify fracture line in 39 cases (18.6%) using plain radiographs. Te fracture line of pos- terior malleolus was hard to identify as the overlap of the distal tibia and ﬁbula. Terefore, for patients with conﬁrmed or suspected ankle fracture, the potential risk of missing diagnosis of posterior malleolar fracture makes further CT examination necessary [15]. Some researchers recom- mended to diagnose and measure posterior malleolar fractures with 50 degrees external rotation lateral view [30]. Plain radiographs are neither adequate in the diagnosis of posterior malleolar fractures nor reliable in assessing the posterior articular involvement. Ferries et al. [16] found a big diﬀerence between plain radiography and axial CT in the measurement of the articular involvement of posterior malleolar fractures. About 54% of the plain radiographic readings revealed >25% error. Meijer et al. [18] claimed a mean diﬀerence of 10.9% between plain radiographs and 3D CT images. In our study, similar ﬁndings were achieved. Te mean diﬀerence was 5.82% between plain radiograph linear 7 BioMed Research International and axial CT plane and 8.58% between plain radiograph linear and 3D CT surface. Te involvement of the fragment depends largely on which measurement method is used. Terefore, the existing operative indications determined based on plain radiography are not suitable for other measurement methods. total ﬁbular notch (plain radiograph, axial CT linear, sagittal CT linear, and 3D CT linear), which was in line with the traditional method [14]. Te other is to measure the ratio of the area of the fragment to the total tibial plafond (axial CT plane and 3D CT surface). Te 3D CT linear measurement method from the ﬁrst category and the 3D CT surface measurement method from the second category were proved to be reliable and reproducible. 4. Discussion Firstly, diﬀerent from plain radiographs and 3D CT images, the measurement level of 2D CT would be determined by each observer. Tis might reduce the reliability and re- producibility. Secondly, when measuring the total cross- sectional area at the level of the tibial plafond, some previous studies overestimated the denominator (the total area of the tibial plafond) as it contains the area of the medial malleolus [16, 20]. When 3D CT was used, the intra- and interobserver agreement was stable and acceptable. Distal tibial articular surface cannot be observed by traditional imaging technique under direct vision because of obstruction by talus. Te SSD technique, which had been proved to be superior in bony surface reconstructions, was used to distinguish bony structures in our study [24, 34]. Te tibial plafond was revealed and measured under direct vision after hiding the talus. Terefore, the operative indications need to be reevaluated with the use of a more reliable and reproducible measurement method such as 3D CT linear or 3D CTsurface methods. Te current study had some limitations. Firstly, we only included patients with striking posterior malleolar fracture line on lateral plain radiographs. Terefore, the ICC might be higher than those in published reports. Besides, a small part of the patients were excluded from the samples because of severe comminution or defect of posterior malleolar fragments on 3D CT images. Continuous further studies are required to solve these problems. 5. Conclusions Te existing operative indications which were determined based on plain radiography are neither reliable nor suitable for other measurement methods. Both 3D linear and 3D surface measurement method are reliable and reproducible in measuring posterior fragment involvement, and experi- ence is not so crucial. We call on further multicenter clinical research on large sample to demonstrate the relationship between prognosis and posterior malleolar fracture in- volvement on the basis of 3D CTimages, in order to redeﬁne the operative indications for posterior malleolar fractures. Data Availability Te data used to support the ﬁndings of this study are available from the corresponding author upon request. Conflicts of Interest Te authors declare that there are no conﬂicts of interest regarding the publication of this paper. References [17] N. Haraguchi, H. Haruyama, H. Toga, and F. Kato, “Patho- anatomy of posterior malleolar fractures of the ankle,” Te Journal of Bone and Joint Surgery, vol. 88, no. 5, pp. 1085– 1092, 2006. [1] C. M. Court-Brown and B. Caesar, “Epidemiology of adult fractures: a review,” Injury, vol. 37, no. 8, pp. 691–697, 2006. [2] K. J. Koval, J. Lurie, W. Zhou et al., “Ankle fractures in the elderly: what you get depends on where you live and who you see,” Journal of Orthopaedic Trauma, vol. 19, no. 9, pp. 635–639, 2005. [18] D. T. Meijer, J. N. Doornberg, I. N. Sierevelt et al., “Guess- timation of posterior malleolar fractures on lateral plain ra- diographs,” Injury, vol. 46, no. 10, pp. 2024–2029, 2015. [3] V. Ribeiro de ´Avila, T. Bento, W. Gomes, J. Leitão, and N. Fortuna De Sousa, “Functional outcomes and quality of life after ankle fracture surgically treated: a systematic review,” Journal of Sport Rehabilitation, vol. 27, no. 3, pp. 274–283, 2018. [19] O. Gonzalez, J. J. Fleming, and A. J. Meyr, “Radiographic assessment of posterior malleolar ankle fractures,” Te Journal of Foot and Ankle Surgery, vol. 54, no. 3, pp. 365–369, 2015. [20] Y. Yi, D. I. Chun, S. H. Won, S. Park, S. Lee, and J. Cho, “Morphological characteristics of the posterior malleolar fragment according to ankle fracture patterns: a computed tomography-based study,” BMC Musculoskeletal Disorders, vol. 19, no. 1, p. 51, 2018. [4] D. J. Keene, E. Williamson, J. Bruce, K. Willett, and S. E. Lamb, “Early ankle movement versus immobilization in the postoperative management of ankle fracture in adults: a systematic review and meta-analysis,” Journal of Orthopaedic & Sports Physical Terapy, vol. 44, no. 9, pp. 690–C7, 2014. [21] L. Mangnus, D. T. Meijer, S. A. Stufkens et al., “Posterior malleolar fracture patterns,” Journal of Orthopaedic Trauma, vol. 29, no. 9, pp. 428–435, 2015. [5] H. G. Jung, J. I. Kim, J. Y. Park et al., “Is hardware removal recommended after ankle fracture repair?,” BioMed Research International, vol. 2016, Article ID 5250672, 7 pages, 2016. [22] S. Zhong, L. Shen, J.-G. Zhao et al., “Comparison of post- eromedial versus posterolateral approach for posterior mal- leolus ﬁxation in trimalleolar ankle fractures,” Orthopaedic Surgery, vol. 9, no. 1, pp. 69–76, 2017. [6] J. Evers, M. Fischer, I. References Zderic et al., “Te role of a small posterior malleolar fragment in trimalleolar fractures: a biomechanical study,” Te Bone & Joint Journal, vol. 100-B, no. 1, pp. 95–100, 2018. [23] M. C. Solan and A. Sakellariou, “Posterior malleolus fractures: worth ﬁxing,” Te Bone & Joint Journal, vol. 99-B, no. 11, pp. 1413–1419, 2017. pp [7] J. Forberger, P. V. Sabandal, M. Dietrich et al., “Posterolateral approach to the displaced posterior malleolus: functional outcome and local morbidity,” Foot & Ankle International, vol. 30, no. 4, pp. 309–314, 2009. [24] Y. Chen, X. Jia, M. Qiang, K. Zhang, and S. Chen, “Computer- assisted virtual surgical technology versus three-dimensional printing technology in preoperative planning for displaced three and four-part fractures of the proximal end of the humerus,” Te Journal of Bone and Joint Surgery, vol. 100, no. 22, pp. 1960–1968, 2018. [8] T. L. Meyer Jr. and K. W. Kumler, “A.S.I.F. technique and ankle fractures,” Clinical Orthopaedics and Related Research, vol. 150, pp. 211–216, 1980. [9] N. C. Tejwani, B. Pahk, and K. A. Egol, “Eﬀect of posterior malleolus fracture on outcome after unstable ankle fracture,” Te Journal of Trauma: Injury, Infection, and Critical Care, vol. 69, no. 3, pp. 666–669, 2010. [25] Y. Chen, M. Qiang, K. Zhang, H. Li, and H. Dai, “Novel computer-assisted preoperative planning system for humeral shaft fractures: report of 43 cases,” Te International Journal of Medical Robotics and Computer Assisted Surgery, vol. 11, no. 2, pp. 109–119, 2015. [10] V. W. Macko, L. S. Matthews, P. Zwirkoski, and S. A. Goldstein, “Te joint-contact area of the ankle. Te contribution of the posterior malleolus,” Te Journal of Bone & Joint Surgery, vol. 73, no. 3, pp. 347–351, 1991. [26] Y. Yoshii, T. Kusakabe, K. Akita, W. L. Tung, and T. Ishii, “Reproducibility of three dimensional digital preoperative planning for the osteosynthesis of distal radius fractures,” Journal of Orthopaedic Research, vol. 35, no. 12, pp. 2646– 2651, 2017. & Joint Surgery, vol. 73, no. 3, pp. 347–351, 1991 [11] J. M. Hartford, J. T. Gorczyca, J. L. Mcnamara, and B. Mayor, “Tibiotalar contact area. Contribution of posterior malleolus and deltoid ligament,” Clinical Orthopaedics and Related Research, no. 320, pp. 182–187, 1995. [27] H.-L. Xu, X. Li, D.-Y. Zhang et al., “A retrospective study of posterior malleolus fractures,” International Orthopaedics, vol. 36, no. 9, pp. 1929–1936, 2012. [12] A. A. Abdelgawad, A. Kadous, and E. Acknowledgments Te authors thank the Radiology Department of our hospital for their assistance with original image collection. Tis re- search was supported by the National Natural Science Foundation of China (Grant no. 81370053) and Municipal Te above measurement methods can be summarized into two categories. One category is to measure the ratio of the lateral margin of the posterior malleolar fragment to the BioMed Research International 8 Human Resources Development Program for Outstanding Leaders in Medical Disciplines in Shanghai (2017BR059). Human Resources Development Program for Outstanding Leaders in Medical Disciplines in Shanghai (2017BR059). Trauma: Injury, Infection, and Critical Care, vol. 53, no. 1, pp. 55–60, 2002. [15] L. B¨uchler, M. Tannast, H. M. Bonel, and M. Weber, “Reli- ability of radiologic assessment of the fracture anatomy at the posterior tibial plafond in malleolar fractures,” Journal of Orthopaedic Trauma, vol. 23, no. 3, pp. 208–212, 2009. Supplementary Materials Figure S1: Bland–Altman plots and detailed comparison data of any two of the measurement methods (associated with the data of Table 3). (Supplementary Materials) Figure S1: Bland–Altman plots and detailed comparison data of any two of the measurement methods (associated with the data of Table 3). (Supplementary Materials) p pp [16] J. S. Ferries, T. A. Decoster, K. K. Firoozbakhsh, J. F. Garcia, and R. A. Miller, “Plain radiographic interpretation in tri- malleolar ankle fractures poorly assesses posterior fragment size,” Journal of Orthopaedic Trauma, vol. 8, no. 4, pp. 328– 331, 1994. References Kanlic, “Posterolateral approach for treatment of posterior malleolus fracture of the ankle,” Te Journal of Foot and Ankle Surgery, vol. 50, no. 5, pp. 607–611, 2011. pp [28] P. E. Shrout and J. L. Fleiss, “Intraclass correlations: uses in assessing rater reliability,” Psychological Bulletin, vol. 86, no. 2, pp. 420–428, 1979. [13] R. A. Jaskulka, G. Ittner, and R. Schedl, “Fractures of the posterior tibial margin: their role in the prognosis of malleolar fractures,” Te Journal of Trauma: Injury, Infection, and Critical Care, vol. 29, no. 11, pp. 1565–1570, 1989. [29] J. P. Weir, “Quantifying test-retest reliability using the intraclass correlation coeﬃcient and the SEM,” Journal of Strength and Conditioning Research, vol. 19, no. 1, pp. 231– 240, 2005. [14] J. F. Langenhuijsen, M. J. Heetveld, J. M. Ultee, E. P. Steller, and R. M. J. M. Butzelaar, “Results of ankle fractures with involvement of the posterior tibial margin,” Te Journal of [30] N. A. Ebraheim, A. O. Mekhail, and S. P. Haman, “External rotation-lateral view of the ankle in the assessment of the BioMed Research International 9 9 posterior malleolus,” Foot & Ankle International, vol. 20, no. 6, pp. 379–383, 1999. [31] T. A. Irwin, J. Lien, and A. R. Kadakia, “Posterior malleolus fracture,” Journal of the American Academy of Orthopaedic Surgeons, vol. 21, no. 1, pp. 32–40, 2013. [32] L. F. Amorosa, G. D. Brown, and J. Greisberg, “A surgical approach to posterior pilon fractures,” Journal of Orthopaedic Trauma, vol. 24, no. 3, pp. 188–193, 2010. pp [33] E. Geusens, W. Geyskens, P. Brys, and H. Janzing, “Te role of the reversed oblique radiograph in trauma of the foot and ankle,” European Radiology, vol. 10, no. 3, pp. 476–479, 2000. [34] B. S. Kuszyk, D. G. Heath, D. F. Bliss, and E. K. Fishman, “Skeletal 3-D CT: advantages of volume rendering over surface rendering,” Skeletal Radiology, vol. 25, no. 3, pp. 207–214, 1996.
https://openalex.org/W4362537510	https://figshare.com/articles/journal_contribution/Supplementary_Figure_from_Exploiting_Ligand-binding_Domain_Dimerization_for_Development_of_Novel_Androgen_Receptor_Inhibitors/22523610/1/files/39986403.pdf	English	null	Supplementary Figure from Exploiting Ligand-binding Domain Dimerization for Development of Novel Androgen Receptor Inhibitors	null	2,023	cc-by	415	1. Eerlings R, Barbakadze N, Nguyen T, Nadaraia N, Smeets E, Moris L, et al. Small-molecule profiling for steroid receptor activity using a universal steroid receptor reporter assay. J Steroid Biochem Mol Biol 2021;217:106043 doi 10.1016/j.jsbmb.2021.106043. Figure S5: Cross reactivity of DIMs on the steroid receptors, agonist setting Potential agonistic activity of DIM compounds (at 5, 10 and 20 µM) on the steroid recep and ER was tested. A universal steroid receptor luciferase reporter was co-transfected wi for the corresponding receptor in HEK 293T cells as described by Eerlings et al.(1). Luciferas to B-galactosidase activity to correct for transfection efficiency. As a control, known agonis included in the assay. As an agonist for AR, GR, PR, MR and ER; Dihydrotestosterone (DHT Progesterone (Prog), Aldosterone (Ald) and Estradiol (E2) were used, respectively. Dot pl points and their mean (black line) are given, with n=2 for AR and n=3 for the other recepto Figure S5: Cross reactivity of DIMs on the steroid receptors, agonist setting Potential agonistic activity of DIM compounds (at 5, 10 and 20 µM) on the steroid recep and ER was tested. A universal steroid receptor luciferase reporter was co-transfected wi for the corresponding receptor in HEK 293T cells as described by Eerlings et al.(1). Luciferas to B-galactosidase activity to correct for transfection efficiency. As a control, known agonis included in the assay. As an agonist for AR, GR, PR, MR and ER; Dihydrotestosterone (DHT Progesterone (Prog), Aldosterone (Ald) and Estradiol (E2) were used, respectively. Dot pl points and their mean (black line) are given, with n=2 for AR and n=3 for the other recepto Figure S5: Cross reactivity of DIMs on the steroid receptors, agonist setting Figure S5: Cross reactivity of DIMs on the steroid receptors, agonist setting Potential agonistic activity of DIM compounds (at 5, 10 and 20 µM) on the steroid receptors AR, GR, PR, MR, ER and ER was tested. A universal steroid receptor luciferase reporter was co-transfected with the expression plasmid for the corresponding receptor in HEK 293T cells as described by Eerlings et al.(1). Luciferase values were normalized to B-galactosidase activity to correct for transfection efficiency. As a control, known agonists for each receptor were included in the assay. As an agonist for AR, GR, PR, MR and ER; Dihydrotestosterone (DHT), Dexamethasone (Dex), Progesterone (Prog), Aldosterone (Ald) and Estradiol (E2) were used, respectively. Dot plots of the individual data points and their mean (black line) are given, with n=2 for AR and n=3 for the other receptors. 1.
https://openalex.org/W2161731621	https://europepmc.org/articles/pmc2908116?pdf=render	English	null	Climatic Variability Leads to Later Seasonal Flowering of Floridian Plants	PloS one	2,010	cc-by	9,044	Abstract Understanding species responses to global change will help predict shifts in species distributions as well as aid in conservation. Changes in the timing of seasonal activities of organisms over time may be the most responsive and easily observable indicator of environmental changes associated with global climate change. It is unknown how global climate change will affect species distributions and developmental events in subtropical ecosystems or if climate change will differentially favor nonnative species. Contrary to previously observed trends for earlier flowering onset of plant species with increasing spring temperatures from mid and higher latitudes, we document a trend for delayed seasonal flowering among plants in Florida. Additionally, there were few differences in reproductive responses by native and nonnative species to climatic changes. We argue that plants in Florida have different reproductive cues than those from more northern climates. With global change, minimum temperatures have become more variable within the temperate-subtropical zone that occurs across the peninsula and this variation is strongly associated with delayed flowering among Florida plants. Our data suggest that climate change varies by region and season and is not a simple case of species responding to consistently increasing temperatures across the region. Research on climate change impacts need to be extended outside of the heavily studied higher latitudes to include subtropical and tropical systems in order to properly understand the complexity of regional and seasonal differences of climate change on species responses. ei Y, Nickerson D (2010) Climatic Variability Leads to Later Seasonal Flowering of Floridian Plants. PLoS ONE 5(7): e11500. doi:10.1371/ n: Von Holle B, Wei Y, Nickerson D (2010) Climatic Variability Leads to Later Seasonal Flowering of Floridian Plants. PLoS ONE 5(7): e1 pone.0011500 Received March 9, 2010; Accepted June 11, 2010; Published July 21, 2010 Copyright: 2010 Von Holle et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: 2010 Von Holle et al. Betsy Von Holle1, Yun Wei2, David Nickerson2 ity of Central Florida, Orlando, Florida, United States of America, 2 Department of Statistics, University of Central Florida, Orlando, Florida 1 Department of Biology, University of Central Florida, Orlando, Florida, United States of America, 2 Department of Statistics, Universi United States of America 1 Department of Biology, University of Central Florida, Orlando, Florida, United States of America, 2 Department of Statistics, University of Central Florida, Orlando, Florida, United States of America Abstract This is an open-access article distributed under the terms of the Creative C unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cred Funding: Funding for this research was provided to BVH by the Invasive Plant Management Section of the Florida Fish and Wildlife Conservation Commission, the Office of Research of the University of Central Florida, and research start-up dollars provided by the UCF Department of Biology. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. E-mail: vonholle@mail.ucf.edu * E-mail: vonholle@mail.ucf.edu nonnative species to expand their ranges into areas where they previously could not survive and reproduce [6]. Florida has the second highest number of nonnative plant species in the US comprising 27% of the flora [8] and state, federal, and local agencies devoted approximately $250 million dollars for the control and eradication of invasive nonnative species in Florida from 1980 to 2007 [9]. Under conditions of climatic change, plants in Florida have the potential to expand or contract their ranges to areas where they are better suited to the environmental conditions. While it is not possible to conclusively separate range expansion of nonnative species from climate-induced range expansion, a response to climate change would be implicated by changes in the flowering phenologies of plant species correspond- ing to specific changes in climate. Climatic Variability Leads to Later Seasonal Flowering of Floridian Plants Betsy Von Holle1, Yun Wei2, David Nickerson2 July 2010 \| Volume 5 \| Issue 7 \| e11500 Introduction During the past century, average annual global temperatures for land and ocean surfaces have increased at a rate near 0.6uC/ century (1.1uF/century), however the trend has been three times larger since 1976, with some of the greatest increases in temperature occurring in the high latitudes [1]. Florida has been getting increasingly warmer, with the average annual temperature increasing by 0.02uC per decade, with temperatures in 2007 0.2uC warmer than in 1895 [2]. Spring events such as leaf unfolding and flowering are associated with changes in air temperature [3]. Changes in phenology (the timing of seasonal activities of animals and plants) over time may be the most responsive and easily observable indicator of environmental changes associated with global climate change [3,4]. Global climate change has had pronounced effects on the developmental events of species; the majority of observed changes in phenology have occurred in the direction that would be expected under warming, occurring earlier in the season[5]. Climate change and species invasions are two of the biggest contributors to global change, yet their effects have typically been considered separately [6]. It is expected that most aspects of global climate change (e.g. increasing CO2, nitrogen deposition, etc.) will favor nonindigenous species because invasive species share traits that allow them to capitalize on these perturbations [7]. In fact, warming temperatures have allowed We tested the hypothesis that global climate change has altered the reproductive phenologies of populations of high-impact nonnative plant species and their closely related native congeners, predicting that nonnative species will have greater responses to climatic change than natives. To investigate this, we focused on the change in reproductive status of 29 high impact invasive plant species and 41 closely related native species in Florida over historical time, using herbarium specimens (Supporting material available online, Table S1). Our first objective was to determine if temperature and precipitation levels have changed in Florida counties over time by season (spring, summer, fall, winter). For PLoS ONE \| www.plosone.org July 2010 \| Volume 5 \| Issue 7 \| e11500 1 July 2010 \| Volume 5 \| Issue 7 \| e11500 Delayed Subtropical Flowering this, we analyzed historical climatic records for the entire record of climate data available for each county. Second, we determined if the flowering phenologies of highly invasive nonnative plant species as well as closely related native species have changed over time, using flowering records from herbarium specimens. Statistical Analyses Climatic trends by county. We performed regression analyses of monthly temperature and precipitation data for 57 Florida counties to determine if these climatic factors have changed over historical time. We used the entire record of climate data collection for each county, which ranged from 35 to 108 years (Table S2). To ascertain if the stations with more recent data would capture greater warming trends than the stations which covered longer time periods, we ran a Fishers Exact Test. In this test, significance (significantly positive, significantly negative, nonsignificant) and year category (35–49, 50–99, 100+ years) were the variables. Methods To determine shifts in flowering time for each species, we performed t-tests on the difference between average flowering dates that occurred in an earlier time period (1890–1969) to those collected from 1970 to 2008. Only those species with more than ten specimens in both the earlier and later time periods were used for the analyses. Owing to low sample sizes when the data were separated by species, biogeographic region and the two time periods, we summed the data for each species across all biogeographic regions of the state. If the data were not normally distributed, we performed Mann-Whitney tests. We recorded 6,218 herbarium specimens for the 70 study species which were collected from 1819 to 2008, the majority of the accessions collected between 1929 and 2007. We used 5,019 of these specimens for this analysis, less than that collected, either because there were no matching climatic data for that county at the time of flowering or the specimen was not flowering. Flowering data We used the average monthly minimum temperature, maximum temperature, and precipitation data for each county (an average of 36 data points per year) from 1973–2007 as the attributes to group the most similar counties. We used average linkage clustering to compute the distance between clusters and Euclidian distance as the measure of similarity. We checked for variability of each factor and standardized the temperature and precipitation variables so that they had a mean of 0 and a variance of 1. We determined the Methods Twenty-nine species were chosen from the 133 Category I and II Florida Exotic Pest Plant Council (EPCC) species for their distinct, short reproductive phenologies (as in Primack et al. 2004) and because they occur in more than one region of Florida. Category I species are high impact invaders that alter native plant communities by displacing native species, change ecological functions or community structures, or hybridize with natives, based on documented ecological damage [10]. Category II species are invasive exotics that have increased in abundance or frequency but have not yet altered Florida plant communities to the extent shown by Category I species (FLEPPC 2007). These species may be ranked Category I, if ecological damage is demonstrated. We identified 41 native species which were the most closely related to these 29 nonnative species below the family level to compare climate change impacts and to compare differences between the flowering times of native and nonnative species with climate change [11]. We chose taxonomically related native species commonly found in Florida that had distinct, short reproductive phenologies in order to compare the nonnative species with their most similar native species. If there were no native congeners to the focal nonnative species, we chose the native species in the most taxonomically similar genera to the focal genus, below the family level [11,12]. There were five nonnative species (Casuarina cunninghamiana, C. equisitifolia, C. glauca, Eleagnus pungens, and Melia azeradach) which were used in the analysis but which had no confamilial native species in Florida. Low temperatures in the winter, high temperatures in the summer, high and low temperature extremes, as well as water availability are among the factors that limit plant distributions [15]. Thus, we analyzed the average minimum winter (December, January, February) and spring (March, April, May) temperatures, the average maximum summer (June, July, August) and fall (September, October, November) temperatures and mean precip- itation for each county and season over time to understand how these climatic variables, which may influence plant reproduction, vary in Florida by season and county. To determine if autocorrelation occurred between years, we tested the residuals from the ordinary least squares regression for serial correlation using the generalized Durbin-Watson statistic [16]. If autocorre- lation was detected, autoregressive error terms of the appropriate order were added to the model until the generalized Durbin- Watson statistic indicated there was no autocorrelation. Methods Then the model was refitted, using the method of maximum likelihood with the autoregressive error terms of the required order. g q Patterns of flowering time. To determine shifts in flowering time for each species, we performed t-tests on the difference between average flowering dates that occurred in an earlier time period (1890–1969) to those collected from 1970 to 2008. Only those species with more than ten specimens in both the earlier and later time periods were used for the analyses. Owing to low sample sizes when the data were separated by species, biogeographic region and the two time periods, we summed the data for each species across all biogeographic regions of the state. If the data were not normally distributed, we performed Mann-Whitney tests. Biogeographic regions. To reduce spatial autocorrelation and climatic heterogeneity for the 70 species analyzed, we clustered the 67 Florida counties by similar historic climatic trends. We did this by performing a hierarchical cluster analysis, which groups counties based on a measure of similarity among county attributes, for the 57 counties with weather stations (see, e.g., [17]). To maximize the number of counties with climate data and to capture the period of greatest climatic change, we used records from 1973–2007 for the cluster analysis. We used the average monthly minimum temperature, maximum temperature, and precipitation data for each county (an average of 36 data points per year) from 1973–2007 as the attributes to group the most similar counties. We used average linkage clustering to compute the distance between clusters and Euclidian distance as the measure of similarity. We checked for variability of each factor and standardized the temperature and precipitation variables so that they had a mean of 0 and a variance of 1. We determined the Patterns of flowering time. To determine shifts in flowering time for each species, we performed t-tests on the difference between average flowering dates that occurred in an earlier time period (1890–1969) to those collected from 1970 to 2008. Only those species with more than ten specimens in both the earlier and later time periods were used for the analyses. Owing to low sample sizes when the data were separated by species, biogeographic region and the two time periods, we summed the data for each species across all biogeographic regions of the state. If the data were not normally distributed, we performed Mann-Whitney tests. Patterns of flowering time. Flowering data We monitored the length of time that each of these species was historically in flower and fruit by using specimens from six herbaria in Florida (Archbold Biological Station, Fairchild Tropical Botanical Garden, Florida State University, University of Central Florida, University of Florida, and the University of South Florida). We utilized accessions from these herbaria to determine the date and season that these species flowered over historical time. Using online or physical herbarium specimens, we recorded reproductive status (budding, flowering, or fruiting), the county, and date of collection of each specimen [13]. The use of herbarium specimens has been demonstrated to be successful in documenting phenological responses to global change and have been found to be comparable to field observations [14]. were not normally distributed, we performed Mann Whitney tests. Biogeographic regions. To reduce spatial autocorrelation and climatic heterogeneity for the 70 species analyzed, we clustered the 67 Florida counties by similar historic climatic trends. We did this by performing a hierarchical cluster analysis, which groups counties based on a measure of similarity among county attributes, for the 57 counties with weather stations (see, e.g., [17]). To maximize the number of counties with climate data and to capture the period of greatest climatic change, we used records from 1973–2007 for the cluster analysis. We used the average monthly minimum temperature, maximum temperature, and precipitation data for each county (an average of 36 data points per year) from 1973–2007 as the attributes to group the most similar counties. We used average linkage clustering to compute the distance between clusters and Euclidian distance as the measure of similarity. We checked for variability of each factor and standardized the temperature and precipitation variables so that they had a mean of 0 and a variance of 1. We determined the y p y Biogeographic regions. To reduce spatial autocorrelation and climatic heterogeneity for the 70 species analyzed, we clustered the 67 Florida counties by similar historic climatic trends. We did this by performing a hierarchical cluster analysis, which groups counties based on a measure of similarity among county attributes, for the 57 counties with weather stations (see, e.g., [17]). To maximize the number of counties with climate data and to capture the period of greatest climatic change, we used records from 1973–2007 for the cluster analysis. Introduction We also evaluated the environmental drivers that affected changes in reproductive phenologies with statistical models that matched individual flowering times with climatic variables for all biogeo- graphic regions of Florida. were established as early as 1900 (12 counties) to as late as 1973, with the majority installed by 1960 (Florida Climate Center, Center for Ocean-Atmospheric Prediction Studies). Climate data were obtained through July 2007. If there were more than one weather station per county, we averaged the monthly climate data for those stations for each climate variable. PLoS ONE \| www.plosone.org Climate data Monthly climate data (minimum temperature, maximum temperature, precipitation) were obtained from each of the 57 Florida counties with weather stations. These weather stations PLoS ONE \| www.plosone.org July 2010 \| Volume 5 \| Issue 7 \| e11500 2 Delayed Subtropical Flowering optimal number of clusters for the 57 Florida counties by applying Mojena’s stopping rule, which identifies the first stage in the dendrogram at which there is a large change in the distance between clusters [18]. We grouped the ten Florida counties without a weather station with the closest county with a weather station, as determined by the Florida Climate Center, Center for Ocean-Atmospheric Prediction Studies. This resulted in seven clusters of counties (Fig. 1, Table S3A) which had similar historic climatic trends that we treated as separate biogeographic regions for the following analyses. of flowering [19]. To account for lagged flowering responses to environmental cues, we selected the lowest (min_T) and highest (max_T) average monthly temperatures in the months of the calendar year prior to flowering for each individual in the analysis. Thus, we selected the min_T and max_T for each specimen from January to the month of flowering of the same year. The independent variables used for the multiple regression model (hereafter described as ‘‘model 1’’) were year, average precipitation for the month of flowering (precip), the lowest average monthly minimum temperature in the calendar year prior to flowering (min_T), the highest average monthly maximum temperature in the calendar year prior to flowering (max_T), range of minimum temperatures for the months previous to flowering of that calendar year (range_minT), range of maximum temperatures for the months previous to flowering of that calendar year (range_maxT), plant origin (native or nonnative), and two-way interactions (yearprecip, yearmin_T, yearmax_T, precipmin_T, precipmaxT, preciporigin, year origin, min_Torigin, max_Torigin, rangeminTorigin, range_ maxTorigin) between these variables. The range in minimum temperatures were calculated in the following way: greatest ‘min_T’ minus the lowest ‘min_T’, which occurred from January Environmental drivers of flowering time. To understand if the flowering phenologies of nonnative and 41 closely-related native plant species have changed over time as well as the environmental drivers of these changes, we performed multiple regression analyses for all species within each biogeographic region using the Julian date of specimens that were in flower as the dependent variable. We matched flowering and county climate data for each specimen, using all available flowering and climatic data. Climate data There were no significant differences in warming, cooling, or lack of trend by season for the weather stations with different durations of climatic data, suggesting that climatic trends did not differ by the length of time the weather station was in operation. Precipitation did not appear to change across most of Florida over time, with the majority of counties registering no significant seasonal changes in precipitation in the last 35 to 108 years (Table S5). However, significant increases in average monthly precipita- tion occurred in 11% of the counties in winter and summer months and significant decreases in monthly precipitation in 4% of the counties for the spring and summer months (Table S5). to the flowering month of the year of flowering. The range for the maximum temperatures were calculated similarly, but using the ‘max’_T in place of the ‘min_T’. For example, the native Ardisia escallonioides was found in Brevard county in 2006 and this specimen was in flower on June 18th, or Julian date 170. The average monthly minimum temperatures that occurred prior to this were 1 = 11.6, 2 = 10.2, 3 = 13.9, 4 = 18.3,5 = 19.2,6 = 22.3, with January = 1, Februar- y = 2,etc. Thus the lowest mean minimum temperature (min_T) selected for this specimen was 10.2, for February of 2006. The greatest mean minimum temperature in 2006 prior to this specimen flowering was 22.3, which occurred in June of that year. Thus, the range in minimum temperature for this species was (22.3210.2 = 12.1). We adjusted for multiple significance tests by applying a sequential Bonferroni adjustment within each of the clusters [20]. Additionally, we evaluated a second model to explore other environmental cues on flowering, specifically the effect of temperature cues in the form of heat accumulation as well as the number of freezing days prior to flowering, hereafter described as ‘‘model 2’’. We performed multiple regression analyses as described above, using the Julian date of specimens that were in flower as the dependent variable and the following independent variables: year, total precipitation the day of flowering (precip), the cumulative growing degree days in the calendar year prior to flowering (CGDD), the number of days below freezing (,0uC) in the calendar year prior to flowering (Freeze), plant origin and two- way interactions between these variables (YrPrecip, YrCGDD, YrFreeze, Precip&CGDD, PrecipFreeze, CGDDFreeze, Yr- Origin, CGDDOrigin, FreezeOrigin, PrecipOrigin). Climate data If a plant flowered in a county without a weather station at that time, we did not use those data. Environmental cues for flowering time, such as temperature or precipitation, may occur several months prior to the conditions measured at the time Figure 1. Florida counties grouped by similar climatic conditions. The climatic factors included monthly averages for minimum temperature, maximum temperature, and the precipitation for each county from 1973–2007. These seven clusters of counties had similar historic climatic trends Figure 1. Florida counties grouped by similar climatic conditions. The climatic factors included monthly averages for minimum temperature, maximum temperature, and the precipitation for each county from 1973–2007. These seven clusters of counties had similar historic climatic trends that we treated as separate biogeographic regions for these analyses. See Table S3A for county identity in each biogeographic region. doi:10.1371/journal.pone.0011500.g001 Figure 1. Florida counties grouped by similar climatic conditions. The climatic factors included monthly averages for minimum temperature, maximum temperature, and the precipitation for each county from 1973–2007. These seven clusters of counties had similar historic climatic trends that we treated as separate biogeographic regions for these analyses. See Table S3A for county identity in each biogeographic region. doi:10.1371/journal.pone.0011500.g001 PLoS ONE \| www.plosone.org July 2010 \| Volume 5 \| Issue 7 \| e11500 3 Delayed Subtropical Flowering and fall months, with 26% of counties experiencing significantly increased average maximum temperatures in the summer months (June, July, August) and 35% of the counties with significantly increased mean maximum temperatures in the fall months (September, October, November). These increases were within a range of 0.02 to 0.09uC per year for the last 35 to 108 years (Fig. 2, A and B, Table S4). The counties that experienced warmer fall maximum temperatures trends were clustered in southern Florida (Fig. 2B). Surprisingly, the majority of counties with significant changes in average monthly winter and spring temperatures had decreased minimum temperatures over the last century, with 16% of Florida counties recording significantly lower mean minimum temperatures in the winter months (December, January, February) and 26% of the counties documenting significantly lower average temperatures in the spring months (March, April, May). The counties with significantly lower winter and spring minimum temperatures tended to cluster in northern Florida (Fig. 2, C and D). The average monthly minimum temperature decreases ranged from -0.04 to 20.18uC per year for the last 34 to 108 years (Table S4). Climate data To calcu- late growing degree days, or a measure of heat accumulation, we used the GDD equation described in Otto et al. (2007, [21]), with a base temperature of 10uC, and no ceiling temperature requirement. We then summed the GDD numbers between the first day of the year of flowering to the date that the specimen was flowering to calculate cumulative growing degree days (CGDD). Thus, for the Ardisia escallonioides example above, GDD numbers were summed from January 1, 2006 to the flowering date, June 18, 2006, to calculate the cumulative growing degree days (CGDD) for this specimen. Methods used to predict flowering responses to climatic conditions, such as cumulative growing degree days [21] and the range in minimum and maximum temperatures, rely on a greater number of data points as the calendar year progresses and may affect the results of the model. Patterns of flowering time Across the state, two species flowered significantly later in the year (nonnative Albizia lebbeck, native Sassafras albidum) and one species flowered significantly earlier in the year (native Morus rubra) (Table S6). Biogeographic regions In order to reduce spatial autocorrelation and climatic heterogeneity for the 70 species analyzed, we clustered the 67 Florida counties by similar historic climatic trends. This resulted in seven clusters of counties (Fig. 1, Table S3 A and B) which had similar historic climatic trends that we treated as separate biogeographic regions for the following analyses. In order to understand the effects of seasonal and regional climatic changes on plant reproduction in Florida, we analyzed flowering data for each of the Florida county clusters. We performed follow-up analyses to the multiple regression analyses described above with simple linear regressions between the flowering dates of all specimens in each region as well as flowering dates for each species within each region by the range in minimum temperatures, calculated for each specimen. Flowering data were log-transformed to account for non-linear responses of flowering date to the range in minimum temperatures as well as an increased range in the variation of flowering date with increasing minimum temperatures. Environmental drivers of flowering time Plant flowering time was strongly delayed by variable minimum temperatures over historical time, with a range of approximately four to nineteen days later in the year, opposite the pattern observed for most phenological studies conducted worldwide [5,22,23]. Later flowering time was significantly correlated with the within-year variability in minimum temperatures, or the range of mean monthly minimum temperatures that occurred in the months of the calendar year prior to flowering, in all seven Florida biogeographic clusters (Fig. 3A, Table 1). Additionally, flowering time was delayed at the species level by greater variability in minimum temperatures (Table S7). Twenty-one of the seventy- nine simple linear regressions between flowering date and the range in minimum temperature for each species were statistically significant, after correcting for multiple comparisons. The regression coefficients for the range of minimum temperatures of PLoS ONE \| www.plosone.org Climatic trends by county Historically, there has been a trend in Florida of warmer and wetter climate, with an average decadal temperature increase of 0.02uC (R2 = 0.025, p = 0.09) and precipitation increase of 0.75 cm (R2 = 0.022, p = 0.12) between 1895 and 2008 [2]. However, this overall warming trend for the state is complicated by warming and cooling trends that differ seasonally and by region. Over time Florida has been getting warmer in the summer PLoS ONE \| www.plosone.org July 2010 \| Volume 5 \| Issue 7 \| e11500 July 2010 \| Volume 5 \| Issue 7 \| e11500 4 Delayed Subtropical Flowering Figure 2. Historic temperature trends for Florida counties. Average maximum temperature trends for each Florida county in the (A) summer (June, July, August) and (B) fall (September, October, November) months. Average minimum temperature trends for each Florida county in the (C) winter (December, January, February) and (D) spring (March, April, May) months. Counties are colored for the change in average temperature over the monitored period, which ranged from 35 to 108 years. Orange indicates a significant increase, aqua denotes a significant decrease, and tan indicates no change in historical temperature for that county and season. doi:10.1371/journal.pone.0011500.g002 Figure 2. Historic temperature trends for Florida counties. Average maximum temperature trends for each Florida county in the (A) summer (June, July, August) and (B) fall (September, October, November) months. Average minimum temperature trends for each Florida county in the (C) winter (December, January, February) and (D) spring (March, April, May) months. Counties are colored for the change in average temperature over the monitored period, which ranged from 35 to 108 years. Orange indicates a significant increase, aqua denotes a significant decrease, and tan indicates no change in historical temperature for that county and season. doi:10 1371/journal pone 0011500 g002 Figure 2. Historic temperature trends for Florida counties. Average maximum temperature trends for each Florida county in the (A) summer (June, July, August) and (B) fall (September, October, November) months. Average minimum temperature trends for each Florida county in the (C) winter (December, January, February) and (D) spring (March, April, May) months. Counties are colored for the change in average temperature over the monitored period, which ranged from 35 to 108 years. Orange indicates a significant increase, aqua denotes a significant decrease, and tan indicates no change in historical temperature for that county and season. doi:10.1371/journal.pone.0011500.g002 Figure 2. PLoS ONE \| www.plosone.org Climatic trends by county doi:10.1371/journal.pone.0011500.g003 Figure 3. Phenological and climatic data, from region one. Region one was chosen as a typical representive of Florida regions, as all seven biogeographic regions displayed similar trends. A. Flowering date (log) increases linearly with the range in minimum temperature (p,0.0001, R2 = 0.69, n = 1,274). Each dot represents a single specimen, with the Julian date of flowering matched to the range in minimum temperature calculated for that specimen from climatic data from that county. B. The average range in minimum temperatures in region one increases over time (p = 0.009, R2 = 0.08, n = 86). Each dot represents a year of climatic data in region one, where the range in minimum temperatures in that region are matched with the year. doi:10.1371/journal.pone.0011500.g003 did not provide as clear of a signal for flowering time as the previous model which incorporated variability in minimum temperatures prior to flowering (Table S8). However the results of the second model supported the previous model (Table 1), which we highlight below. In regions two, four, and five, increasing numbers of freezing days prior to flowering delays flowering by plants to later in the year. In these three regions, this is tempered over time, as the effect of freeze dates on flowering declined over time, as indicated by the negative value of the interaction between the variables ‘Year’ and ‘Freeze’. Addition- ally, in region five, as the precipitation increases, this causes the effect of freeze dates to delay flowering even further, as indicated by the positive value of the interaction between the variables for precipitation and number of freezing days prior to flowering. The effect of cumulative growing degree days (CGDD) is greater for nonnative species in region six, as indicated by the positive term of the interaction between ‘CGDD’ and ‘Origin’. nonnative species flowering approximately 7 days later than native species. However, it should be noted that plant origin was not a significant factor explaining differences in flowering time in five of the seven biogeographic regions (2, 3, 4, 5, and 7) for model 1, suggesting that native and nonnative species did not differ in their flowering responses to these environmental factors in these regions during this time period. In the statistical model (model 1), there were some environ- mental factors with negative correlation coefficients, implying that these factors would be associated with earlier flowering times (Table 1). Climatic trends by county Historic temperature trends for Florida counties. Average maximum temperature trends for each Florida county in the (A) summer (June, July, August) and (B) fall (September, October, November) months. Average minimum temperature trends for each Florida county in the (C) winter (December, January, February) and (D) spring (March, April, May) months. Counties are colored for the change in average temperature over the monitored period, which ranged from 35 to 108 years. Orange indicates a significant increase, aqua denotes a significant decrease, and tan indicates no change in historical temperature for that county and season. doi:10.1371/journal.pone.0011500.g002 flowering later in the year than expected from simply adding the effects of these two independent variables in these regions. In region 7, plant species flowered later in the year over historical time outside of the influence of the increased variability in minimum temperatures, as indicated by the statistical significance of the variable ‘year’. Thus, there may be an additional environmental factor that we did not measure that accounts for the later seasonal flowering in this region. The effect of the range of minimum temperatures differed by plant origin in region 1, with these twenty-one significant regressions were positive, suggesting that as the range in minimum temperature increases in a given year, the flowering date for these species is significantly delayed (Table S7).Variability of minimum temperatures has increased in Florida over historical time (Fig. 3B). The two-way interaction of precipitation and minimum temperatures on flowering date was significant and positive in regions 1 and 7, suggesting that increased precipitation and average monthly minimum temperatures is correlated with plants July 2010 \| Volume 5 \| Issue 7 \| e11500 5 Delayed Subtropical Flowering Figure 3. Phenological and climatic data, from region one. Region one was chosen as a typical representive of Florida regions, as all seven biogeographic regions displayed similar trends. A. Flowering date (log) increases linearly with the range in minimum temperature (p,0.0001, R2 = 0.69, n = 1,274). Each dot represents a single specimen, with the Julian date of flowering matched to the range in minimum temperature calculated for that specimen from climatic data from that county. B. The average range in minimum temperatures in region one increases over time (p = 0.009, R2 = 0.08, n = 86). Each dot represents a year of climatic data in region one, where the range in minimum temperatures in that region are matched with the year. Climatic trends by county The environmental factors responsible for trends to earlier flowering times were the interaction between average monthly precipitation levels and maximum temperatures (regions 1, 6, and 7), and the variability in maximum temperatures (regions 1 and 7). Additionally, the effect of precipitation differed by species origin in region 6, with nonnative plants flowering approximately one day earlier per 1 centimeter increase in precipitation than native plants. Last, the change in the variability of maximum temperatures over historical time was a significant factor that played a role for slightly earlier flowering times in regions 1 and 7. While these factors contributed to earlier flowering times, the primary driver of flowering time appears to be the range of minimum temperatures, whose effect estimate is much greater than range of maximum temperatures, and causes plants to flower later in the year. Among the variables considered, the range in minimum temperatures was the only environmental factor responsible for the change in plant flowering times in four regions (2, 3, 4, and 5), accounting for 78 to 81% of the variation in the flowering times of native and nonnative species. In sum, the environmental factor with the greatest correlation with plant flowering times in all seven regions was the variability in minimum temperatures, and this factor was strongly associated with later seasonal flowering times. To follow up on these results, we conducted simple linear regressions between the number of freezing days in each biogeographic region per calendar year over time. The number of freezing days have significantly increased over time in regions one, two, four, five, and seven [sample size, R2, regression coeffient (r.c.), and p-values presented by region: One, n = 964, R2 = 0.03, r.c. = 0.04, p = ,0.0001; Two, n = 356, R2 = 0.07, r.c. = 0.13, p = ,0.0001, Four, n = 75, R2 = 0.06, r.c. = 0.15, p = 0.04; Five, n = 138, R2 = 0.08, r.c. = 0.10, p = 0.0007, Seven, n = 1724, R2 = 0.005, r.c. = 0.009, p = 0.003], suggesting that the trend for increasing days below freezing per calendar year is common in Florida. PLoS ONE \| www.plosone.org Discussion doi:10.1371/journal.pone.0011500.t001 Response variables were year (yr), precipitation (precip), the lowest average monthly minimum temperature in the calendar year prior to flowering (min_T), the highest average monthly maximum temperature in the calendar year prior to flowering (max_T), range of minimum temperatures for the months previous to flowering of that calendar year (range_minT), range of maximum temperatures for the months previous to flowering of that calendar year (range_maxT), plant origin (native or nonnative; ‘‘origin’’). Two-way interactions between these variables are listed by variable names separated by an ‘’. Precipitation is in centimeters and temperature is in degrees Celsius. We applied a sequential Bonferroni adjustment within each cluster. P,0.05, P,0.001, P,0.001, P,0.0001, for tests of significant difference of parameter values from 0. doi:10.1371/journal.pone.0011500.t001 plants [15]. In this study, we found that greater variability in minimum temperatures was associated with strongly delayed flowering time in all seven biogeographic regions and that increasing freezing temperatures occurring prior to flowering were associated with later flowering times in three biogeographic regions. Thus, delays in flowering time associated with fluctuating minimum temperatures might have been caused by the lowest temperatures in the range, in other words, increased freezing temperatures might have been primarily responsible for delayed flowering. At the species level, both native and nonnative species experienced delayed flowering in years with large variability in minimum temperatures, suggesting that fluctuating minimum temperatures significantly delayed reproduction by plants. It may be difficult for plants to respond physiologically to large temperature fluctuations and so plants may cue their flowering times on the variation in minimum temperatures, rather than the lowest minimum temperatures, which could lead to bud dormancy extending later in the year. Whether plants cue their reproduction on extreme low temperatures or variability in low temperatures is an issue that needs to be explored further. The variation in minimum temperatures has increased over historical time in all regions of Florida. Additionally, the number of days occurring per calendar year below freezing has significantly increased over time in five of the seven biogeographic regions. If this trend continues, reproductive events by native and nonnative species may continue to be delayed to later in the season in Florida. time. Florida counties experienced significantly colder winter and springs while simultaneously experiencing significantly warmer summer fall temperatures over historical time. PLoS ONE \| www.plosone.org Discussion Contrary to expectation, the majority of counties with significant changes in average monthly winter and spring temperatures had lower minimum temperatures over historical The second model, which included the effects of cumulative growing degree days and the number of freezes on flowering time, July 2010 \| Volume 5 \| Issue 7 \| e11500 July 2010 \| Volume 5 \| Issue 7 \| e11500 6 Delayed Subtropical Flowering Table 1. Analyses of the flowering time with climatic variables for each county cluster (model 1). Regression Parameters, by Cluster Response Variable 1 2 3 4 5 6 7 Year 1.17 2.04 23.56 21.56 1.81 1.63 2.76 Precip 9.95 1.38 2209.70 46.94 2128.31 214.32 216.64 min_T 52.36 149.20 21043.0 2251.34 2283.31 11.08 266.16 Max_T 61.90 106.25 38.07 2106.22 133.94 156.55 238.26 Range_minT 15.85 4.07 17.6* 13.49** 19.57* 12.38** 18.99 Range_maxT 23.39** 21.17 211.0 25.37 0.73 1.57 23.32* Origin 414.63 2114.03 2767.06 12071.86 6524.23 18.26 2591.29 YrPrecip 20.004 0.001 0.11 20.03 0.06 0.01 0.01 Yrmin_T 20.02 20.07 0.53 0.13 0.15 20.002 0.04 Yrmax_T 20.03 20.05 20.01 0.06 20.07 20.07 20.12*** Precipmin_T 0.17 20.02 0.19 0.31 0.18 0.09 0.18**** Precipmax_T 20.14 20.11 20.08 0.21 0.18 20.34** 20.40** YrOrigin 20.17 0.08 21.31 25.80 23.42 20.05 0.28 min_TOrigin 3.24 5.68 212.10 16.24 214.63 2.61 21.72 max_TOrigin 24.47 24.89 20.81 240.71 17.83 2.08 2.45 Range_minTOrigin 6.69**** 6.40 26.87 46.06 228.57 5.15 20.09 Range_maxTOrigin 23.31 1.37 4.50 0 12.95 26.72 21.06 PrecipOrigin 20.69 21.48 22.55 0 3.15 21.05* 20.05 R2 0.796 0.784 0.815 0.808 0.793 0.798 0.785 p-value ,.0001 ,.0001 ,.0001 ,.0001 ,.0001 ,.0001 ,.0001 N 1306 400 79 103 154 1055 1922 Response variables were year (yr), precipitation (precip), the lowest average monthly minimum temperature in the calendar year prior to flowering (min_T), the highest average monthly maximum temperature in the calendar year prior to flowering (max_T), range of minimum temperatures for the months previous to flowering of that calendar year (range_minT), range of maximum temperatures for the months previous to flowering of that calendar year (range_maxT), plant origin (native or nonnative; ‘‘origin’’). Two-way interactions between these variables are listed by variable names separated by an ‘’. Precipitation is in centimeters and temperature is in degrees Celsius. We applied a sequential Bonferroni adjustment within each cluster. P,0.05, P,0.001, * P,0.001, **** P,0.0001, for tests of significant difference of parameter values from 0. Discussion This would suggest that nonnative plants in Florida do not have an overwhelmingly greater phenological response to climatic change in the form of increased freezing temperatures than closely related native species do, and may not be differentially favored by climate change in regions experiencing colder winter and spring conditions. However, nonnative species may be able to capitalize on warming conditions, as suggested by an empirical study of marine invertebrate species in the northeastern United States where nonnative recruitment was stronger than native species under conditions of warming spring water temperatures [26]. Likewise, non-native invasive species tracked seasonal temperature variation better than natives did in Massachusetts, flowering significantly earlier than natives with warming spring temperatures [25]. There are many examples from around the world where warming temperatures have resulted in extended growing and reproductive periods which have provided nonnative species from warmer climates opportunities to expand and invade into new ranges [6]. However, it is unknown what types of environmental conditions associated with climatic change that nonnatives will be able to capitalize on, as compared to native species [7]. Further research should be conducted to understand nonnative species responses to the specific environmental conditions (e.g. changes in temperature, precipitation, and nutrients) of climatic change in order to predict community composition under changing climatic conditions. phases have occurred in several cases. In the Balkans, leaf unfolding and flowering has been retarded in the time period from 1959 to 1993 [3]. In a phenological study of twelve plant and animal taxa in Japan and South Korea, first observations of a five of these species (frog, butterfly, wasp, and two bird species) were delayed over the time period of 1953–2005 at the majority of sites [29]. Interestingly, the sites from this study spanned a wide latitudinal gradient (24u20.2N to 45u24.99N) including boreal as well as subtropical climates, rarely the focus of phenological studies. Seasonal and regional differences in climatic changes strongly affect species reproductive phenologies and likely have cascading effects on the populations, communities and ecosystems of these regions [30]. While the greatest levels of warming of land and ocean surfaces are expected to occur in high latitudes[1], the complexity of air temperature changes in the subtropics found in this study warrants further attention. Discussion The warming trend in the fall was concentrated in southern counties while the cooling trends in the winter and spring were clustered in northern Florida. These results suggest that climate change varies by region and season. The strongest driver of these later flowering times in our study was the variability in the minimum temperatures, rather than simply the lowering of the minimum temperatures that occurred in 16% of the Florida counties in winter and 26% of the counties in spring months. Only two species had significantly later flowering times (Albizia lebbeck, Sassafras albidum), and one species flowered significantly earlier in the season (Morus rubra) across the state. The low number of species that demonstrated significant changes in flowering time across the state in this analysis may owe to the different climatic conditions occurring in each individual biogeographic region. Each biogeographic region experienced distinct climatic condi- tions during the study period, and these environmental conditions had strong effects on plant reproduction. Thus, varying environ- mental influences of each biogeographic region on flowering time may be obscured when analyzed together across a large region. Temperature is a limiting factor for reproduction, survival and growth in plants [24]. Low temperature extremes limit plants by frost or cold damage to leaves and buds or the freezing death of whole July 2010 \| Volume 5 \| Issue 7 \| e11500 7 Delayed Subtropical Flowering Worldwide, there are numerous examples where temperature changes have resulted in extended growing and reproductive periods which have provided nonnative species from warmer climates opportunities to expand and invade into new ranges [6]. In our study, the effect of the range of minimum temperatures differed by plant origin in region 1, with nonnative species flowering approximately 7 days later than native species. However, it should be noted that plant origin was not a significant factor explaining differences in flowering time in five of the seven biogeographic regions (2, 3, 4, 5, and 7) in model 1, implying that native and nonnative species did not differ in their flowering times in these regions during this time period. Likewise, in our second model, there were no differences between plant origins to freezing temperatures but there were differences in origin for heat accumulation, suggesting that nonnatives may be able to track warming temperatures more quickly than natives do [25]. Supporting Information Table S1 The 29 high impact nonnative species and their 41 most closely related native species in Florida. Table S1 The 29 high impact nonnative species and their 41 most closely related native species in Florida. Found at: doi:10.1371/journal.pone.0011500.s001 (0.05 MB DOC) Found at: doi:10.1371/journal.pone.0011500.s001 (0.05 MB DOC) Table S2 The 57 Florida counties with weather stations, listed alphabetically. Table S2 The 57 Florida counties with weather stations, listed alphabetically. Found at: doi:10.1371/journal.pone.0011500.s002 (0.05 MB DOC) Found at: doi:10.1371/journal.pone.0011500.s002 (0.05 MB DOC) Table S3 A. Florida counties clustered by similar climatic conditions, from 1973-2007. B. Average annual climatic variables for each biogeographic region, from 1973-2007. Found at: doi:10.1371/journal.pone.0011500.s003 (0.04 MB DOC) Table S3 A. Florida counties clustered by similar climatic conditions, from 1973-2007. B. Average annual climatic variables for each biogeographic region, from 1973-2007. Found at: doi:10.1371/journal.pone.0011500.s003 (0.04 MB DOC) Table S4 Historic trends in temperature by Florida county. Found at: doi:10.1371/journal.pone.0011500.s004 (0.03 MB DOC) Climate change varies around the world and concomitant ecological responses are likely to differ by region [4]. However, the majority of research on global climate change and species phenology has been conducted in northern latitudes, ranging from 31.9 to 71.2u [27], with mean latitudes of 49.8 and 51.7 degrees reported by two of the most comprehensive meta-analyses of phenological responses of species to climate change ([5] and [12], respectively). A more recent meta-analysis of the phenolog- ical shifts due to climate change included 125,000 observational series of 542 plant and 19 animal species in 21 European countries[22] which ranged in latitude from 37.35u N to 69.75u [28]. In this study, 30% of the leafing, flowering, and fruiting records were significantly earlier while 3% were significantly delayed[22]. Phenological studies conducted in a temperate- subtropical climate are extremely rare. Florida’s latitude ranges from 24u 309 N to 31u N and it is possible that the delays in onset of species reproduction may be associated with environmental conditions of lower latitudes. Table S5 Historic trends in average monthly precipitation levels by Florida county. Found at: doi:10.1371/journal.pone.0011500.s005 (0.03 MB DOC) Table S6 Patterns in flowering time for nonnative and native species. Found at: doi:10.1371/journal.pone.0011500.s006 (0.07 MB DOC) Table S7 Simple linear regression of the log of flowering date with range in minimum temperature for each species in each region. Discussion Furthermore, the velocity at which low-elevation regions with low topographic relief, such as Florida, will experience climate change is expected to be higher than in areas with greater topographic relief [31]. Species in regions with low topographic relief will require faster response times to climate change [31] and therefore these regions should be a high priority for research on species adaptations to climate change. PLoS ONE \| www.plosone.org Acknowledgments We are grateful to A. Von Holle for SAS programming assistance. We express our sincere gratitude to the following people for their help in obtaining herbarium specimen data; K. Perkins (UF), L. Woodbury (FBC), C. Weekley (ABS), C. Knickerbocker (UCF); and B. Hansen, R. Wunderlin, and J. Kunzer (USF). We appreciate the help of A. Landi and A. Weeks in compiling the data. D.D. Ackerly, P. Quintana-Ascencio, We are grateful to A. Von Holle for SAS programming assistance. We express our sincere gratitude to the following people for their help in obtaining herbarium specimen data; K. Perkins (UF), L. Woodbury (FBC), C. Weekley (ABS), C. Knickerbocker (UCF); and B. Hansen, R. Wunderlin, and J. Kunzer (USF). We appreciate the help of A. Landi and A. Weeks in compiling the data. D.D. Ackerly, P. Quintana-Ascencio, B Ek l D J F i d C R Hi kl J Sil d d We are grateful to A. Von Holle for SAS programming assistance. We express our sincere gratitude to the following people for their help in obtaining herbarium specimen data; K. Perkins (UF), L. Woodbury (FBC), We are grateful to A. Von Holle for SAS programming assistance. We express our sincere gratitude to the following people for their help in obtaining herbarium specimen data; K. Perkins (UF), L. Woodbury (FBC), Supporting Information Found at: doi:10.1371/journal.pone.0011500.s007 (0.16 MB DOC) Table S8 Analyses of the flowering time with cumulative growing degree days, freezing days, and precipitation climatic variables for each county cluster (model 2). Found at: doi:10.1371/journal.pone.0011500.s008 (0.05 MB DOC) While the vast majority of spring events in mid- to high- latitudes have occurred earlier in the season and are associated with warming spring temperatures [22], delayed onset of spring July 2010 \| Volume 5 \| Issue 7 \| e11500 8 Delayed Subtropical Flowering reviewer made comments on a previous version of this manuscript. L. McCauley assisted with the maps. Author Contributions Conceived and designed the experiments: BVH. Performed the experi- ments: BVH. Analyzed the data: BVH YW DMN. Contributed reagents/ materials/analysis tools: BVH YW DMN. Wrote the paper: BVH. Wunderlin, and J. Kunzer (USF). We appreciate the help of A. Landi and A. Weeks in compiling the data. D.D. Ackerly, P. Quintana-Ascencio, B. Ekwurzel, D.J. Friedman, C. R. Hinkle, J. Silander, and an anonymous References 16. Vinod HD (1973) Generalization of the Durbin-Watson statistic for higher order autoregressive process. Communications in Statistics 2: 115–144. 1. NOAA Satellite and Information Service N (2009) State of the Climate Global Analysis. Annual 2009. In: National Climatic Data Center USDoC, editor Asheville, NC, USA: National Climatic Data Center, U.S. Department of Commerce. 17. Rencher AC (2002) Methods of Multivariate Analysis. New York: Wiley. 18. Mojena R (1977) Hierarchical grouping methods and stopping rules: An evaluation. Computer Journal 20: 359–363. 2. Center FC (2008) Monthly temperature and precipitation data. Center for Ocean-Atmospheric Prediction Studies. p J 19. Miller-Rushing AJ, Primack RB (2008) Global warming and flowering times in Thoreau’s Concord: A community perspective. Ecology 89: 332–341. Thoreau’s Concord: A community perspective. Ecology 89: 332– 3. Menzel A, Fabian P (1999) Growing season extended in Europe. Nature 397: 659–659. 20. Rice WR (1989) Analyzing tables of statistical tests. Evolution 43: 223–225 21. Otto S, Masin R, Chiste G, Zanin G (2007) Modelling the correlation between plant phenology and weed emergence for improving weed control. Weed Research 47: 488–498. 4. Walther GR, Post E, Convey P, Menzel A, Parmesan C, et al. (2002) Ecological responses to recent climate change. Nature 416: 389–395. responses to recent climate change. Nature 416: 389–395. 5. Parmesan C, Yohe G (2003) A globally coherent fingerprint of climate change impacts across natural systems. Nature 421: 37–42. 22. Menzel A, Sparks TH, Estrella N, Koch E, Aasa A, et al. (2006) European phenological response to climate change matches the warming pattern. Global Change Biology 12: 1969–1976. 6. Walther GR, Roques A, Hulme PE, Sykes MT, Pysek P, et al. (2009) Alien species in a warmer world: risks and opportunities. Trends in Ecology & Evolution 24: 686–693. g gy 23. Root TL, Price JT, Hall KR, Schneider SH, Rosenzweig C, et al. (2003) Fingerprints of global warming on wild animals and plants. Nature 421: 57–60. 7. Dukes JS, Mooney HA (1999) Does global change increase the success of biological invaders? Trends in Ecology & Evolution 14: 135–139. 24. Woodward FI (1987) Climate and Plant Distribution. Cambridge, UK: Cambridge University Press. 8. Ward DB (1990) How many plant species are native to Florida? Palmetto 89/90: 3–5. 25. Willis CG, Ruhfel BR, Primack RB, Miller-Rushing AJ, Losos JB, et al. (2010) Favorable climate change response explains non-native species’ success in Thoreau’s Woods. Plos One 5: e8878. 9. References Schmitz DC (2007) Florida’s invasive plant research: Historical p 9. Schmitz DC (2007) Florida’s invasive plant research: Historical perspective and the present research program. Natural Areas Journal 27: 251–253. 26. Stachowicz JJ, Terwin JR, Whitlatch RB, Osman RW (2002) Linking climate change and biological invasions: Ocean warming facilitates nonindigenous species invasions. Proceedings of the National Academy of Sciences of the United States of America 99: 15497–15500. 10. FLEPPC (2007) List of Florida’s Invasive Plant Species. Wildland Weeds 10: http://www.fleppc.org/07list.htm. 11. Wunderlin RP, Hansen BF (2003) Guide to the Vascular Plants of Florida. Gainesville: University Press of Florida. 787 p. 27. Parmesan C (2007) Influences of species, latitudes and methodologies on estimates of phenological response to global warming. Global Change Biology 13: 1860–1872. 12. Burns JH (2004) A comparison of invasive and non-invasive dayflowers (Commelinaceae) across experimental nutrient and water gradients. Diversity and Distributions 10: 387–397. 28. Estrella N, Menzel A (2009) Personal communication. 13. Primack D, Imbres C, Primack RB, Miller-Rushing AJ, Del Tredici P (2004) Herbarium specimens demonstrate earlier flowering times in response to warming in Boston. American Journal of Botany 91: 1260–1264. 29. Primack RB, Ibanez I, Higuchi H, Lee SD, Miller-Rushing AJ, et al. (2009) Spatial and interspecific variability in phenological responses to warming temperatures. Biological Conservation 142: 2569–2577. warming in Boston. American Journal of Botany 91: 1260–126 14. Miller-Rushing AJ, Primack RB, Primack D, Mukunda S (2006) Photographs and herbarium specimens as tools to document phenological changes in response to global warming. American Journal of Botany 93: 1667–1674. p g 30. Walther GR (2002) Weakening of climatic constraints with global warming and its consequences for evergreen broad-leaved species. Folia Geobotanica 37: 129–139. 31. Loarie SR, Duffy PB, Hamilton H, Asner GP, Field CB, et al. (2009) The velocity of climate change. Nature 462: 1052–U1111. 15. Box EO (1995) Factors determining distributions of tree species and plant functional types. Vegetatio 121: 101–116. PLoS ONE \| www.plosone.org July 2010 \| Volume 5 \| Issue 7 \| e11500 9
https://openalex.org/W4249460072	https://rjmp.com.ro/articles/2017.2/RJMP_2017_2_Art-01.pdf	Romanian, Moldavan	null	null	Romanian Journal of Medical Practice	2,017	cc-by	4,275	INTRODUCERE sunt necesare și anumite informaţii primite din mediul extern. De aceea creierul a devenit de fapt un organ informaţional (2). În ultimul timp s-a constatat însă că, pe lângă informaţia geneti- că adusă de cele 20.000 și de gene pe care le-am moștenit de la părinţii noștri și pe lângă informaţia primită din mediul înconjurător, în dezvoltarea și în funcţionarea creierului mai in- tervin și o serie întreagă de mecanisme epigene- tice (3), capabile să regleze activitatea genelor în funcţie de informaţiile primite din mediul în- conjurător. În felul acesta mecanismele epige- netice se interpun între informaţia genetică sto- cată în molecula de ADN și informaţia primită din mediul extern, devenind de fapt un fel de software care reglează hardware-ul genetic al creierului (4). Creierul este probabil unul dintre cele mai complicate sisteme din univers, deoarece el este format din peste 100 de miliarde de neu- roni, care se leagă între ei prin intermediul a peste 100 de trilioane de sinapse, care dau naș- tere la o reţea formată din peste 1 milion de mi- liarde de conexiuni posibile, care depășesc cu mult numărul total al atomilor din univers. Și toate acestea se realizează plecând de la o sin- gură celulă stem, adică de la aceeași informaţie genetică, care dă naștere nu numai la neuroni, ci și la celulele musculare, la celulele hepatice, la celulele epiteliale și la toate celelalte 200 de tipuri de celule din care este format organismul uman. Dar încă de multă vreme s-a constatat că pentru formarea creierului nu este suficientă in- formaţia genetică pe care am moștenit-o de la părinţii noștri, ci sunt necesare și anumite infor- maţii primite din mediul exterior. În acest sens, D.H. Hubel și T.H. Wiesel (1) au arătat că dacă copilul nu primește informaţiile optice necesa- re, într-o anumită perioadă de dezvoltare, scoar- ţa cerebrală occipitală nu se va putea structura în mod corespunzător și copilul nu va mai putea vedea niciodată. Aceasta înseamnă că creierul are nevoie de informaţie nu numai pentru regla- rea comportamentului în funcţie de variaţiile din mediul extern, ci și pentru construirea struc- turilor sale informaţionale. Adică în timp ce pentru formarea ficatului, a rinichilor și a plă- mânilor este suficientă informaţia genetică pri- mită de la părinţi, pentru formarea și funcţiona- rea creierului, pe lângă informaţia genetică, mai Adresă de corespondenţă: Prof. As. Dr. Adrian Restian, Membru titular al Academiei de Ştiinţe Medicale e-mail: restian2003@yahoo.com PRACTICA MEDICALÅ PRACTICA MEDICALÅ EDITORIAL MUTAŢIA CARE A DUS LA CREŞTEREA CREIERULUI Deși prin prelucrarea superioară a informaţi- ilor creierul s-a dovedit foarte util pentru păs- trarea homeostaziei într-un mediu foarte varia- bil și de multe ori chiar foarte ostil, creierul s-a dezvoltat foarte încet. I-au trebuit peste 450 de milioane de ani pentru a ajunge de la ganglionul primitiv al trilobiţilor la greutatea de 400 de gra- me a creierului australopitecului de acum 3 mi- lioane de ani. În tot acest timp creierul animale- lor a rămas foarte mic. De aceea ele s-au bazat pe celelalte organe, adică pe sistemul lor os- teoarticular, pe dinţii lor, pe forţa lor, uneori și pe văzul și mirosul lor și pe instinctele lor de apărare. Abia după 450 de milioane de ani a PRACTICA MEDICALÅ – VOL. 12, NR. 2(50), AN 2017 63 PRACTICA MEDICALÅ – VOL. 12, NR. 2(50), AN 2017 mecanisme, așa cum sunt mecanismele enzima- tice care intervin în funcţionarea genomului. Iar aceste mecanisme genetice, care au rolul de a păstra stabilitatea organismului în pofida nume- roaselor variaţii ale mediului înconjurător, sunt și ele declanșate la rândul lor de niște factori ex- terni. De aceea genele, care nu sunt decât niște tipare, nu funcţionează decât atunci când sunt stimulate de niște factori externi, de niște fac- tori de creștere, de niște hormoni sau de niște informaţii primite din mediul extern, care ajung prin intermediul receptorilor celulari și al căilor de semnalizare intracelulară până la nivelul unor mecanisme epigenetice, care reglează ac- tivitatea genelor în funcţie de variaţiile mediului înconjurător. De aceea, în timp ce majoritatea cercetătorilor studiau genomul uman, unii cer- cetători s-au orientat spre studierea mecanis- melor epigenetice care se interpun între genom și factorii de mediu. apărut la un moment dat mutaţia unei gene, care a stimulat brusc dezvoltarea creierului uman. Din acel moment omul a început să folo- sească tot mai mult capacităţile extraordinare ale creierului, care a devenit organul capabil să regleze în timp util comportamentul organismu- lui pentru a face faţă condiţiilor foarte variabile ale mediului înconjurător (5). Unii cercetători susţin, deci, că creierul nos- tru s-a dezvoltat până la stadiul în care îl avem noi astăzi, datorită mutaţiei unei gene denumite ARHGAP11B. DE LA HARDUL GENETIC LA SOFTUL EPIGENETIC Deși, la un moment dat, a devenit evident că genomul nostru nu reprezintă de fapt decât un depozitar al informaţiei genetice, asemănător cu hardul unui disc de calculator și că acest hard nu poate funcţiona fără un soft, adică fără un program de funcţionare, care să adapteze activi- tatea genelor la condiţiile foarte variabile și de multe ori chiar foarte ostile ale mediului încon- jurător, totuși majoritatea cercetătorilor au con- tinuat să studieze hardul, adică structura mate- rială a genomului, ignorând softul epigenetic, care ar trebui să regleze funcţionarea hardului genetic. Abia după ce genetica a ajuns la apoge- ul său și s-a constatat că genomul, care este de- pozitarul informaţiei genetice, nu poate funcţio- na fără un program care să gestioneze informaţia respectivă, cercetătorii s-au orientat asupra me- canismelor epigenetice care reglează activitatea genelor în funcţie de nevoi. De aceea cea mai lungă perioadă privind studiul eredităţii a fost consacrată structurii genomului, adică a hardu- lui genetic, și a culminat cu descifrarea genomu- lui uman (9). Dar genele nu lucrează nicidată singure și de aceea nici gena ARHGAP11B nu putea realiza singură dezvoltarea unui sistem atât de compli- cat cum este creierul uman. În acest sens, P.D. Evans, J.R. Anderson și E.J. Vallender (7) au ară- tat că în dezvotarea și funcţionarea creierului intervin cel puţin 200 de gene. Printre acestea se numără și gena denumită microcefalina, ge- nele FOXG1 și ASPM (Abnormal Spindle-like Mi- crocephaly), a căror mutaţie determină micro- cefalia și care în mod normal joacă deci un rol foarte important în dezvoltarea creierului. În cele din urmă se admite că în dezvoltarea și funcţionarea creierului intervin câteva mii de gene, care contribuie la diferenţierea neuroni- lor, la migrarea lor, la secreţia factorilor de creș- tere neuronală, la apariţia sinapselor, la sinteza mediatorilor sinaptici și așa mai departe (8). Așa spre exemplu, gena USP9X este implicată în for- marea neuronilor și a sinapselor, gena DLG3 este implicată în memorie și învăţare. Gena FOXP2 este implicată în limbaj. Genele DYX1C1, DCDC2, KIA0319, ROBO1 sunt implicate în disle- xie și așa mai departe. Dar majoritatea genelor de care dispune genomul nostru nu reprezintă decât tiparul după care sunt sintetizate sau ar putea fi sintetizate proteinele, enzimele, hor- monii și neuromediatorii. MUTAŢIA CARE A DUS LA CREŞTEREA CREIERULUI Înlocuirea unei molecule de citozi- nă cu molecula de guanină i-a conferit genei AR- HGAO11B, care nu se găsește la nici o altă spe- cie, capacitatea de a stimula diviziunea neuro- nilor și de a face astfel posibilă creșterea extrem de rapidă a creierului uman (6). De aceea unii cercetători susţin că gena ARHGAP11B este cea care i-a conferit omului posibilitatea de a se de- tașa de toate celelalte rude ale sale. Ei cred că prin intermediul genei ARHGAP11B mutante s-a ajuns de la creierul de 700 de grame al lui Homo habilis, de acum 2 milioane de ani, la creierul de 1.000 de grame al lui Homo erectus, de acum un milion de ani, și la creierul de 1.500 de grame al lui Homo sapiens (5). SUBSTRATUL SOFTULUI EPIGENETIC Primele observaţii privind influenţa factori- lor epigenetici asupra funcţionării genomului au fost făcute încă din 1913, când A.H. Sturtevant a constatat că rearanjarea unor porţiuni din cro- mosomi poate influenţa fenotipul la urmași. În 1950, E. Stedman (16) a constatat că acetilarea histonelor inhibă activitatea genelor adiacen- te. Iar în 1962, J. Doskocil și F. Sorm au confirmat că prin metilarea ADN se reduce sinteza enzima- tică. În 1964, V.G. Allfrey și A.E. Mirsky (17) au arătat că nu numai metilarea ADN, ci și metila- rea și acetilarea histonelor influenţează activita- tea genelor. În 1985, B.M. Cattanach și M. Kirk au arătat că metilarea ADN are rolul de a inhiba secvenţele repetitive ale ADN, care re- prezintă peste 40% din genom. În 1994, M.J.Pazin, R.T. Kamakaka și J.T. Kadonaga au ară- tat că și modelarea și remodelarea cromatinei poate influenţa activitatea genelor. În 1996, J.E. Bronwenil, J. Zhou și T. Ranalli (18) au arătat că histon-acetil-transferaza crește expresia genelor respective. Iar J. Tauton, C.A. Hassing și S.L. Schreiber au arătat că histon-deacetilaza inhibă activitatea genelor respective. Aceste modificări pot să fie reversibile, dar ele pot să fie și stabile. Apoi s-a constatat că nu numai metilarea și de- metilarea AND, ci și metilarea și acetilarea histo- nelor, modelarea cromatinei și genele săltăreţe, ci și numeroasele specii de ARN noncodant au o activitate foarte importantă în reglarea epigen- tică. De aceea, R. Holliday (19) a introdus, în 1987, termenul de ereditate epigenetică. Pentru a sublinia importanţa informaţiei în transmiterea caracterelor ereditare, precum și a unor boli genetice, ar fi suficient să remarcăm faptul că viitorul copil primește de la părinţii săi într-o infimă cantitate de substanţă o foarte mare cantitate de informaţie. Adică el primește într-o infimă cantitate de 7 pg de ADN, adică în 7 milionimi de gram de ADN, o foarte mare can- titate de informaţie, apreciată la 1-1,5 GB (12). Pentru a putea aprecia discrepanţa dintre canti- tatea infinitezimală de substanţă din care este constituită molecula de ADN și cantitatea foarte mare de informaţie genetică pe care o conţine ar fi suficient să arătăm că pentru a scrie această informaţie pe hârtie ne-ar trebui 400 de volume de câte 1.000 de pagini fiecare. Iar pentru a citi fără întrerupere această informaţie ne-ar trebui, după cum arată geneticianul Francis Collins (13), conducătorul uneia dintre cele două echipe care au descifrat genomul uman, peste 31 de ani. DE LA HARDUL GENETIC LA SOFTUL EPIGENETIC Pentru a putea intra în acţiune tiparul genetic mai are nevoie și de niște Dar deși genetica a făcut o serie întreagă de descoperiri extraordinare, până când s-a ajuns la descifrarea genomului uman, probabil că una dintre cele mai importante descoperiri din ge- netică a fost făcută, în 1960, de Marchal Ni- renberg (10), care a descoperit codul genetic. Descoperirea codului genetic a arătat că nu sub- stanţa, din care este formată molecula filiformă 64 PRACTICA MEDICALÅ – VOL. 12, NR. 2(50), AN 2017 de ADN, ci informaţia genetică pe care ea o con- ţine este cea prin intermediul căreia se transmit caracterele ereditare, precum și eventualele boli, de la părinţi la copii. Ar fi suficient să ară- tăm că noţiunea de codificare se referă numai la informaţie, ceea ce înseamnă că nu substanţa, ci informaţia genetică este cea care se transmite, de fapt, de la părinţi la copii. Iar informaţia re- prezintă un alt aspect al realităţii, care are alte legi de conservare și de transformare (11). suntem confruntaţi reprezintă în cele din urmă rezultatul modului în care epigenomul nostru a prelucrat, sau mai bine zis a compatibilizat, in- formaţia primită de la părinţii săi cu informaţia primită din mediul înconjurător (15). de ADN, ci informaţia genetică pe care ea o con- ţine este cea prin intermediul căreia se transmit caracterele ereditare, precum și eventualele boli, de la părinţi la copii. Ar fi suficient să ară- tăm că noţiunea de codificare se referă numai la informaţie, ceea ce înseamnă că nu substanţa, ci informaţia genetică este cea care se transmite, de fapt, de la părinţi la copii. Iar informaţia re- prezintă un alt aspect al realităţii, care are alte legi de conservare și de transformare (11). SUBSTRATUL SOFTULUI EPIGENETIC Iar după cum arată unii autori, o cantitate de 6 gra- me de ADN ar conţine o cantitate uluitoare de informaţie, de 3.072 de exabites, un exabite fi- ind egal cu un miliard de GB. De aceea matema- ticienii caută să construiască niște ordinatoare bazate pe ADN (14). De asemenea, este important de remarcat faptul că nici în procesul de transcripţie și de translaţie nu se transmite substanţă de la ADN la proteine. Nu substanţa, ci informaţia genetică este cea care trece de pe ADN pe ARN și apoi de pe ARN pe proteinele din care este format orga- nismul nostru. ADN rămâne în nucleu, iar ARN care transmite informaţia genetică de la ADN la ribosomii din citoplasmă este reciclat pentru a fi utilizat la o altă transmitere. De aceea genomul este nu numai o mașinărie chimică foarte com- plicată, ci și un calculator biologic foarte compli- cat. Simplificând oarecum fenomenele, am pu- tea compara genomul cu hardul unui disc de calculator, în care este depozitată informaţia genetică primită de la părinţi. Epigenomul ar re- prezenta softul, adică programul care caută să regleze activitatea genelor în funcţie de provo- cările mediului înconjurător. Iar bolile cu care Dar termenul de epigenetică a fost introdus de Conrad Waddington (20) încă din 1942. Prin epigenetică el se referea la influenţele pe care mediul înconjurător le poate avea asupra fenoti- pului nostru, fără a modifica structura genelor. Adică în timp ce genele codifică structura chimi- că a proteinelor, care vor avea rolul de enzime, de mesageri chimici, de receptori celulari sau de anticorpi, epigenetica stabilește care gene tre- buie să acţioneze pentru a sintetiza proteinele respective, în funcţie de solicitările mediului ex- tern. În felul acesta epigenetica reprezintă un fel de interfaţă dintre genom și mediu, care nu nu- mai că nu se opune programului genetic, ci cau- tă să pună în valoare potenţialul genelor pe care 65 PRACTICA MEDICALÅ – VOL. 12, NR. 2(50), AN 2017 (23) arată că și genele săltăreţe, capabile să re- modeleze arhitectura genomului, sunt mult mai active în creier decât în celelalte organe. le-am moștenit, în funcţie de condiţiile foarte va- riabile ale mediului înconjurător. După cum au arătat M. Kaneda, M. Okano și K.M. Hata (24), în timpul embriogenezei, când se construiește creierul, sunt mai active metil- transferazele de novo, așa cum ar fi DNMT3a și DNMT3b, care iniţiază diferenţierea neuronilor. Iar după cum arată A.A. SUBSTRATUL SOFTULUI EPIGENETIC Brooks, M.R. Johnson și P.J. Steer (25), în neuronii adulţi este mai activă DNMT1, care este o metiltransferază de mente- nanţă. G. Fan și L. Hutnick (26) au arătat că scă- derea metilării grupelor CpG din gena care codi- fică sinteza factorului de creștere neuronală determină o creștere a sintezei factorului de creștere care are un rol deosebit în dezvoltarea creierului. J. Hsieh, K. Nakashima și F.H. Gage (27) au arătat că injectarea în hipocamp de acid valproic, care este un inhibitor al histon-deaceti- lazei, determină o scădere a proliferărilor neu- ronale. R.D. Smirt și X. Zha (28) au arătat că me- tiltransferaza influenţează procesul de formare a sinapselor. C.A. Miller, S.L. Campbell și J.D. Sweat (29) au arătat că metilarea ADN intervine în procesul de memorare. Iar J.S. Guan, J. Haaggarty și E. Giacometti (30) au arătat că acetilarea histonelor, determinată de inhibarea histondeacetilazei, facilitează memoria și învă- ţarea. Toate acestea demonstrează că pe lângă cele câteva mii de gene și pe lângă stimulările din mediu extern, în dezvoltarea și în funcţiona- rea creierului mai intervin și niște mecanisme epigenetice care reglează relaţiile dintre genom și mediu, sau mai bine zis dintre informaţia ge- netică și informaţia primită din mediu. ROLUL EPIGENETICII ÎN DEZVOLTAREA ŞI FUNCŢIONAREA CREIERULUI Dar deși ea are un rol foarte important în re- glarea relaţiilor dintre genom și mediu, rolul cel mai important al epigeneticii în organismele pluricelulare este acela de a realiza diferenţie- rea celulară. După cum se știe, în timpul fecun- dării majoritatea genelor sunt metilate, adică inhibate. Dar imediat după fecundare începe o demetilare, până la 20%, a genelor. Iar apoi se produce o nouă remetilare a genelor, până la proporţia de 60% din genom, realizând-se astfel o reprogramare genetică a viitorului organism. De aceea, după cum au arătat J.H. Lee și S. Hart și D. Skalnik (21), acetilarea și deacetilarea his- tonelor pot influenţa dezvoltarea embrionului. Dar realizând o demetilare și o remetilare se- lectivă, adică activarea unor gene și inhibarea altor gene din zestrea genetică a zigotului, me- canismele epigenetice pot realiza o diferenţiere a celulelor pentru a da naștere la diferitele ţesu- turi și organe înalt specializate. Mecanismele epigenetice sunt cele care, plecând de la celula ou, adică de la o anumită informaţie genetică, reușește, prin activarea și inactivarea corespun- zătoare a celor aproximativ 20.000 de gene, să realizeze diferenţierea celulelor care vor forma organismul nostru. Epigenetica este, deci, cea prin intermediul căreia îi putem răspunde astăzi lui Thomas Morgan, care se întreba, încă prin 1910, cum se face diferenţierea celulelor din or- gansimul nostru. Dar de unde știu mecanismele epigenetice ce gene să activeze și ce gene să in- activeze? Este o problemă foarte complicată, care mută discuţia de la programearea genetică la programarea epigenetică. ROLUL INFORMAŢIEI ÎN PLASTICITATEA CREIERULUI Încă de pe vremea lui Rene Descartes, s-a consatatat că stimularea creierului poate să de- termine niște răspunsuri reflexe. Concepţia be- havioristă s-a bazat tocmai pe capacitatea creie- rului de a răspunde reflex la solicitările din me- diu. Mai recent s-a constatat că solicitările infor- maţionale pot influenţa secreţia de neurohor- moni și de mediatori sinaptici, așa cum se în- tâmplă în cazul stresului informaţional (31). Dar încă din 1793, Michele Vincenzo Malacarne a arătat că solicitările informaţionale pot influen- ţa nu numai funcţionarea, ci și structura creieru- lui, deoarece păsările crescute într-un mediu mai bogat, adică într-un mediu care emite mai multe informaţii, au un creier mai mare. În 1962, M.R. Rosenzwig (32) a constatat că șobolanii După cum se știe, dezvoltarea crieierului în- cepe încă din primele zile de viaţă a embrionului și probabil că nu se termină niciodată, deoarece creierul este singurul organ care este mereu in- fluenţat de informaţiile pe care le primește din afară și care reușesc să îl modeleze și să îl remo- deleze în permanenţă. Adică modelarea și re- modelarea creierului depind pe de o parte de informaţia genetică a organismului iar pe de altă parte de informaţiile primite din mediu și de mecanismele epigenetice care se interpun între informaţia genetică și informaţia oferită de me- diu. De aceea Feng, S. Fouse și G. Fan (22) arată că procesele de metilare a ADN din creier sunt mult mai active decât în alte organe. Iar F. Gage 66 66 PRACTICA MEDICALÅ – VOL. 12, NR. 2(50), AN 2017 măr mai mic de receptori pentru corticoizi în hi- pocamp și vor avea o reacţie exagerată la stres. Iar J. Sing și J.M. Meaney (37) au arătat că influ- enţa informaţiilor asupra structurii creierului se face prin intermediul mecanismelor epigenetice. Adică informaţiile acţionează mai întâi asupra mecanismelor epigenetice, care vor acţiona la rândul lor asupra genomului, care reprezintă hardul genetic al creierului. Unele modificări epigenetice ar putea fi dinamice, foarte utile în adaptarea genomului la modificările curente ale mediului înconjurător. Iar altele vor putea fi re- manente pentru a folosi la modificările de dura- tă ale mediului înconjurător. De aceea unii au- tori vorbesc despre o memorie epigenetică, iar alţii chiar despre o ereditate fără gene (38). BIBLIOGRAFIE neurobiology and genome sciences, Nat. Neurosci. 7, 2004, 429-33 1. Hubel, D.H. Wiesel, T.H., Functional architecture of macaque monkey visual cortex, Procedings of Royal Society, 198, 1977, 1-59 17. Allfrey A.G., Faulkner R., Mirsky A.E., Acetylation and methylation of hystones and their possible role in the regulation of RNA synthesis, Proc Natl Acad Sci USA. 5, 1964, 786–794 9. Restian A., Bolile softului epigenetic, Congresul Asociaţiei Medicale Române, Aprilie, 2017 2. Restian A., Creierul informaţional, Conferinţa Naţională de Medicină Integrativă, Bucureşti, 2015 18. Browenil, J.E., Zhou J., Ranalli T., Kobayshi, R., Tetrahymena histone acetyltransferase A: A homolog to yeast Gcn5p linking histone acetylation to gene activation. Cell, 84, 1996, 843–851. 10. Nirenberg M., Leader P., Bernfield R., DNA code wards and protein synthesis, Proceedings of Natural Academy Science, 5, 1965, 1161-1168 3. Holiday R., Epigenetics: A historical overview, Epigenetics, 2, 2006,76-80 4. Restian A., Epigenomul ca maşină Turing, Practica Medicală, 2, 2016, 107-114 11. Restian, A., Principiile de transformare şi de conservare a informaţiei, Studii şi Cercetări de Biotehnologie, 9, 1980, 51-61 19. Holliday R., The inheritance of epigenetic defects. Science, 238, 1987, 163-70 5. Dorus S., Vallendr E.J., Evans, P.D., Accelerated evolution of nervous system gene in the origin of Homo sapiens, Cell. 119, 2004, 1027-1035 20. Waddington C., The epigenotype, Endeavour, 1, 1942, 18 12. Orcutt M., Bases to Bytes, MIT Technology Review, April 25, 2012 21. Lee J.H., Hart S.R., Skalnik D.G., Histone deacetylase activity is required for embryonic stem cell differentiation, 38, 2004, 32-8. 13. Collins F.S., Language of God, London, Toronto, Sydney, 2006 6. Florio M., Albert M., Taverna E., Human-specific gene ARHGAP11B promotes basal progenitor amplification and neocortex expansion, Science, 27, 2015, 1465-70 14. Boneh D., Dunworth C., Lipton R.J. Segal, J.Í., On the computational power of DNA, Discrete Applied Mathematics, 71, 1996, 79–94 22. Feng, J., Fouse S., Fan G., Epigenetic regulation of neural gene expression and neuronal function, 61, 2007, 58-63 7. Evans P.D., Anderson J.R., Vallender E.J., Adaptive evolution of ASPM, a major determinant of cerebral cortical size in humans, Hum Mol Genet 5, 2004, 489-494 15. Restian A., Bolile softului epigenetic, Congresul Asociaţiei Medicale Române, Aprilie, 2017 23. Gage F., Moutri, A., Des genes sauteurs dans le cerveau, Dossier pour la Science, 81, 2013, 18-30 16. Stedman E., Cell specificity of histones, Nature, 166,1950, 780–781 8. Boguski M.S., Jones A.R., Neurogenomics: at the intersection of 24. ROLUL INFORMAŢIEI ÎN PLASTICITATEA CREIERULUI Dar probabil că cea mai frecventă influenţă pe care o au informaţiile asupra creierului este repre- zentată de selecţionarea, sub presiunea solicită- rilor informaţionale ale mediului înconjurător, a unor circuite preferenţiale de prelucrare a infor- maţiilor, a unui așa numit conectom (39). Iar noi am arătat că dacă Sinele este reprezentat mai ales de hardul genetic al creierului, în care sunt înscrise toate instinctele, reflexele și dorinţele noastre, atunci Eul este reprezentat de conecto- mul construit de mecanismele epigenetice sub influenţa solicitărilor informaţionale ale mediu- lui înconjurător (40). crescuţi într-un mediu mai bogat, care generea- ză mai multe informaţii, determină o creștere greutăţii creierului cu 10%. W.T. Greenough (33) a constatat că stimularea creierului determină creșterea cu 25% a numărului de sinapse. A. Pa- scuale-Leone (34) a constatat că zonele de pro- iecţie a degetelor care citesc alfabetul Braille al orbilor au o reprezentare mai mare. La fel, și re- prezentarea degetelor pianiștilor este mai mare decât a celor care nu cântă la pian. Dar A. Pas- cuale-Leone a constatat că nu numai informaţii- le primite din afară, ci și gândurile, adică infor- maţiile generate de el însuși, pot influenţa structura creierului. Dacă un subiect își imagi- nează că efectuează o activitate, așa cum ar fi cântatul la pian, imaginaţia lui va putea influen- ţa de asemenea dimensiunea reprezentării cor- ticale a degetelor folosite de pianist. Acest lucru a fost sugerat cu mult timp în urmă chiar de că- tre Sigmund Freud. Iar apoi E.R. Kandel (35), care a confirmat că psihoterapia poate influenţa structura creierului, ceea ce s-a constatat în ca- zul psihoterapiei comportamentale în tulburări- le obsesiv-compulsive. Pe de altă parte, D. Liu, J. Diorio și M.J. Mea- ney (36) au constatat că privarea de afectivitate a șoriceilor de către mamele lor poate duce la modificări ale structurii creierului. Șoarecii privaţi de informaţiile afective vor avea un nu- BIBLIOGRAFIE Kaneda M., Okano M., Hata K., Essential role for de novo DNA methyltransferase 67 PRACTICA MEDICALÅ – VOL. 12, NR. 2(50), AN 2017 35. Kandel E.R., A new intellectual framework for psychiatry, Am J of Psychiatry, 115, 1998, 457-469 30. Guan J.S., J. Haaggarty S.J., Giacometti E.,HDAC2 negatively regulates memory formation and synaptic plasticity, Nature 459, 2009, 55-60 Dnmt3a in paternal and maternal imprinting, Nature 429, 2004, 900-903 25. Brooks A.A., Johnson M.R., Steer P.J., Birth weight: nature or nurture? Early Human Development, 42, 1995, 29-35 36. Liu D., Diorio J., M.J. Meaney M.J., Maternal care, hippocampal glucocorticoid receptors and hipothalamic-pituitary-adrenal response to stress, Science, 277, 1997, 1659-62 31. Restian A., Informational stress, Journal of Royal Society of Medicine, 6, 1990, 280-283 26. Fan G., Hutnick, L., Methyl-CpG binding proteins in the nervous system, Cell Research 15, 2005, 255–261 32. Rosenzweig M.R., Krech D., Bennet E.L., Diamond M.C., Effects of environmental complexity and training on brain chemistry and anatomy, Journal of Comparative and Physiological Psychology, 55, 1962, 429-437 37. Sing J., Meaney M.J., Environmental regulation of the neural epigenome, Epigenomics, 1, 2009, 131-151 27. Hsieh J., Nakashima K., Gage, F.H. Histone deacetylase inhibition-mediated neuronal differentiation of multipotent adult neural progenitor cells, PNAS, 6, 2017 38. Paldi A., L’hérédité sans gènes, Le Pommiere, Paris, 2009 33. Greenough W.T., Volkmar F.R., Pattern of dendritic branching in occipital cortex of rats reared in complex environments, Experimental Neurology, 40, 1973, 491-504 28. Smirt R.D., Zha, X., Epigenetic regulation of neuronal dendrite and dendritic spine development, Frontiers in Biology, August 2010, Volume 5, Issue 4, pp 304–323 39. Lewis T., A new map of human cortex combines data from multiple imaging modalities and comprises 180 distinct regions, Scientist, July20, 2016 29. Miller C.A., Campbell S.L., Sweat J.D., DNA methylation and histone acetylation work in concert to regulate memory formation and synaptic plasticity 34. Pascuale-Leone, A., Tores, F., Plasticity of sensorimotor cortex representation of the reading finger in Braille readers, Brain, 116, 1993, 39-52 40. Restian A., Personalitatea umană de la Sinele genetic la Eul epigenetic, Conferinţa de Personologie, Sighişoara, aprilie 2016 68
https://openalex.org/W4207078533	https://periodicorease.pro.br/rease/article/download/3811/1481	Portuguese	null	APLICABILIDADE DA TECNOLOGIA NO ENSINO DE LÍNGUA PORTUGUESA COMO RECURSO DE APRENDIZAGEM NA LEITURA	Revista Ibero-Americana de Humanidades, Ciências e Educação	2,021	cc-by	2,808	Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE doi.org/ 10.51891/rease.v7i10.3811 doi.org/ 10.51891/rease.v7i10.3811 1 Formado em Direito pela Universidade Estadual do Piauí e em Letras Português pela Universidade Federal do Piaui. E- mail: adv.gerson38@gmail.com. 2 Formada em Enfermagem pela Universidade Federal de Alfenas. E-mail: aretuzasilvadefigueiredo@gmail.com. 3 Formada em Licenciatura Plena em Matemática pela Universidade Estadual do Piauí. E-mail: Djimara_rocha@hotmail.com. 4 Formada em Licenciatura Plena em Pedagogia pela Universidade Estácio de Sá. E-mail: lucienepb@gmail.com. 5 Formado em Licenciatura Plena em Pedagogia pela Universidade Estadual do Piaui. E-mail:gomesfsil@gmail.com. 6 Formado em Análise e Desenvolvimento de Sistemas pelo Instituto Federal de Rondônia. E-mail: Tsnobre90@gmail.com. APLICABILIDADE DA TECNOLOGIA NO ENSINO DE LÍNGUA PORTUGUESA COMO RECURSO DE APRENDIZAGEM NA LEITURA Gerson de Sousa Batista1 Aretuza Silva de Figueredo2 Djimara de Assis Rocha de Figueiredo3 Luciene pereira Braga4 Francisco Silva Gomes5 Thiago dos Santos Nobre6 RESUMO: O presente artigo procurou investigar sobre a utilização das tecnologias no ensino de Língua Portuguesa, uma vez que a leitura e a escrita praticadas na tela dos aparelhos tecnológicos têm sido observadas como método inevitável, frente a realidade atual. Tendo em vista uma nova formação de alunos com pouco interesse pela leitura, os educadores viram a necessidade de criar estímulos de maneira atrativa e interativa com novas formas de ler e escrever. Em razão disso, as Tecnologias de Comunicação e Informação, conhecidas como tecnologia da informação, tiveram seus recursos explorados em todo campo onde a humanidade atua, porém é no campo da educação que essa pesquisa manteve seu foco. As tecnologias são um conjunto de ferramentas de informação: como o computador, o celular, [o tablete], a TV e suas vantagens, como e-mail, bate-papo, aprendizagem de línguas, blogs e hipertextos (BEATO, 2009), que visa facilitar várias ações na vida das pessoas, utilizando-se de aparelhos de última geração. 3502 Palavras-chaves: Tecnologia. Ensino. Língua Portuguesa. Revista Ibero-Americana de Humanidades, Ciências e Educação. São Paulo, v.7.n.10. out. 2021. ISSN- 2675 – 3375 g g p g @g 6 Formado em Análise e Desenvolvimento de Sistemas pelo Instituto Federal de Rondônia. E-mail: Tsnobre90@gmail.com. Revista Ibero-Americana de Humanidades, Ciências e Educação. São Paulo, v.7.n.10. out. 2021. ISSN- 2675 – 3375 Revista Ibero-Americana de Humanidades, Ciências e Educação. São Paulo, v.7.n.10. out. 2021. INTRODUÇÃO Diante do cenário do ensino atual, as tecnologias mostram na prática um processo importante que é tirar o aluno da rotina de suas aulas tradicionais. No entanto, é fácil perceber que “as tecnologias sozinhas não mudam a escola, mas trazem mil Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE possibilidades de apoio ao professor e de interação com [...] os alunos” (MORAN; MASETTO; BEHRENS, 2003), principalmente, na prática de leitura e escrita. As novas tecnologias se expandiram no contexto escolar, tornando-se um apoio pedagógico, despertando o interesse dos alunos e professores pelas novas ferramentas tecnologias que tem favorecido no processo de leitura e escrita no ensino de Língua Portuguesa. Os procedimentos técnicos desenvolveram-se bastante nos últimos anos sobretudo motivados pelo ensino remoto ocasionado pela pandemia e ancorado na necessidade de uma nova forma de ensino e adaptação aos tempos da tecnologia pedagógica. Nos tempos atuais, o ser humano, o cenário e suas ações têm passado por rápidas e grandes transformações, atualizando-se em várias áreas do conhecimento. Para Machado (2017, p.1) essas mudanças vêm motivando as pessoas a utilizarem novas práticas de ensinar e de aprender, tendo em vista o progresso na sociedade atual. 3503 Com o passar dos tempos, a sociedade testemunhou que “os avanços tecnológicos permitiram que a leitura e a escrita fossem desenvolvidas em outros suportes, como a tela” dos aparelhos tecnológicos (MACEDO; JUNIOR; VELOSO, 2013, p.212). Conhecer e entender como utilizar as ferramentas tecnológicas tem sido visto como ações indispensáveis, tanto quanto foi à necessidade de ler e de escrever (OLIVEIRA, 2019, p.7). A utilização do computador progrediu significativamente com relação às cansativas atividades realizadas nos livros tradicionais, uma vez que aquela motiva e atrai mais seus usuários (GARCIA, 2012, p.22) a realizarem atividades pedagógicas, considerando as transformações oriundas das tecnologias e das práticas sociais. Com as mudanças nos modos de interação da sociedade, “a tecnologia da leitura e da escrita, como mediadoras do ser e agir do homem sobre o mundo, não estão imunes aos efeitos do surgimento das tecnologias digitais”. (FERRAZ, 2019, p,150). Tal Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE reconhecimento faz referência às inúmeras funções e facilidades disponíveis nos dispositivos eletrônicos na hora de escrever e ler na tela dos dispositivos tecnológicos ISSN- 2675 – 3375 De acordo com Dariva (2010, p.7): De acordo com Dariva (2010, p.7): [...] Os recursos da tecnologia nos oferecem certa [...] possibilidade de escrever e reescrever um texto na tela do computador, alterar o tamanho e o estilo das letras, sublinhar, colorir trechos, podendo ter uma qualidade de layout quase perfeita dependendo da habilidade de cada um. A autora destaca, na informação supradita, umas das contribuições que a escrita digital possibilita aos usuários: a formatação e edição, uma alternativa que visa organizar e melhorar o aspecto dos textos eletrônicos de acordo com os conhecimentos da pessoa. A utilização das tecnologias no meio educativo, também, oferece aos estudantes conteúdos ilimitados de textos quando estão conectados a uma rede de internet. Alinhados a essa informação, vários autores relatam que “a comunicação através da internet criou novos horizontes de utilização do código escrito, o que acaba exigindo do autor-leitor novas habilidades da escrita, como também da leitura” (COSTA; MARTINS; QUEIROZ, 2015, p.2) na tela dos dispositivos inteligentes. 3504 As contribuições advindas das tecnologias, no processo de leitura são praticadas e pautadas nas transformações que o mundo atual vem apresentando nas suas práticas sociais. E à medida que o tempo vai passando, cada dia o estudante, o professor e o cidadão são – praticamente - forçados a evoluir, devido às influências modernas que as tecnologias proporcionam para a quebra de novos paradigmas na educação. Porém, de acordo com Rusto e Rubio (2013, p.6) “o processo de ensino- aprendizagem de leitura e de escrita na escola não pode ser configurado como um mundo à parte e não ter a finalidade de preparar o sujeito para a realidade na qual se insere.”. Ou seja, o ensino, ao fazer uso das tecnologias e das ferramentas adequadas, permite ao professor conduzir o aluno a experimentar vários recursos na hora de ensinar, além de retirá-lo da uniformidade. Para Santos (2010, p.1), essa “inclusão no mundo digital oportuniza ao sujeito experimentações, desafios e novas possibilidades de usos sociais da leitura e escrita.”. A ISSN- 2675 – 3375 Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE exemplo da leitura e escrita digital como práticas de letramento, que proporcionam evoluções na interação social e representam ações significativas no desenvolvimento das pessoas, dentro de uma sociedade que integre comunicações reais e virtuais através das TICs (CHAGAS, 2014, p.331). Revista Ibero-Americana de Humanidades, Ciências e Educação. São Paulo, v.7.n.10. out. 2021. ISSN- 2675 – 3375 De acordo com Dariva (2010, p.7): Diante do exposto, entende-se que as TICs provocaram progressões reais na aquisição do conhecimento, sendo que a internet, bem como sua facilidade de acesso, proporciona para os jovens diferentes formas de ler e de escrever (LEAL E LIMA, 2015, p. 36890). Para uma melhor compreensão sobre as informações supraditas, Oliveira (2019, p.4) cita que “vivemos em uma época marcada pela cultura digital e, muitas vezes, não nos damos conta ou paramos para refletir sobre como as influências e implicações das novas tecnologias de informação e comunicação estão conectadas aos nossos hábitos cotidianos.” 3505 Seguindo essa visão, percebe-se que a internet além de ser um “recurso dinâmico, atraente, atualizadíssimo, [...] que possibilita o ingresso a um número ilimitado de informações e dá a oportunidade de contatar todas as grandes bibliotecas do mundo inteiro.” (MORAN, 2006, p.160). Além de vir sendo atribuída como um recurso capaz de potencializa e incentivar o processo de leitura e escrita digital, tendo em vista novos patamares e acesso a novos paradigmas na educação. Mesmo com essa nova realidade de ensino e aprendizagem ajustada às novas tecnologias nas escolas, ainda assim, é uma temática que gera uma polêmica acompanhada de vertentes: ler e escrever na tela do computador são vistos como novas práticas que ajudam o aprendizado. De acordo com a Unesco (2014, p.22), “no mundo desenvolvido, a transição para livros didáticos digitais em ambientes de educação formal é uma das tendências mais bem estabelecidas da aprendizagem móvel.”. E mesmo que a leitura digital ainda possua algumas concepções contra, ainda vale a pena investir nesse tipo de prática, com o objetivo de aumentar o grau de interatividade entre os usuários. ISSN- 2675 – 3375 Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE Utilizar a ferramenta certa para educar pode ser um fator essencial para o professor desempenhar melhor seu trabalho e manter-se ativo, perante, as constantes mudanças da contemporaneidade. É preciso entender a relação entre a tecnologia e seus alunos, para conseguir melhor resultados nos processos pedagógicos, além de ensinar - de forma crítica - os meios necessários para o aluno absorve o assunto. Na busca de melhorar os métodos das instituições de ensino, a temática da leitura e escrita nos aparelhos tecnológicos vem sempre acompanhada de reflexões e questionamentos sobre o seu verdadeiro potencial. Revista Ibero-Americana de Humanidades, Ciências e Educação. São Paulo, v.7.n.10. out. 2021. ISSN- 2675 – 3375 De acordo com Dariva (2010, p.7): E com as seletas ideias reunidas neste trabalho, foi possível constatar que as TICs aliadas à leitura e a escrita, permitem proporcionar novas experiências e facilitar a vida do ser humano, tanto quanto cidadão, como estudante. Os professores de Língua Portuguesa são os principais autores quando se trata de ensinar os alunos a ler e a escrever. Esses ensinamentos exigem, desses profissionais, estarem capacitados para ensinar, uma vez que, nesse contexto, estão inseridos outros tipos de uso social da língua – como os textos digitais (MAJORA, 2018). 3506 Essas considerações visam situar os educadores, tendo em vista as constantes transformações ocorridas em todas as esferas sociais. Sendo que, no âmbito educacional, não é diferente. As instituições de ensino, os professores e todos que pensam em melhorar a educação, precisam sempre estar passando por um processo de atualização nos métodos e nos materiais didáticos. E uma das melhores alternativas de potencializar o ensino-aprendizagem, inserindo as TICs como apoio pedagógico, principalmente explorando a leitura e escrita dos alunos. Assim, com essa evolução dos meios de comunicação digitais, tem surgido uma série de recursos linguísticos que, consequentemente, ocasionou outros tipos de produção textual, determinados pelas transformações da própria escrita (RODRIGUES, 2010). Essas mudanças vêm ocorrendo paulatinamente e são marcadas pela presença em massa dos aparelhos tecnológicos, em praticamente, todos os ambientes e seguimentos , ISSN- 2675 – 3375 Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE da vida humana. Por esse motivo, o novo hábito de leitura e de escrita digital dos estudantes de Língua Portuguesa vem se mantendo, modificando e se adaptando às novas estruturas de leitura e escrita. Sendo assim, vale destacar que os dispositivos levam os fatos até as pessoas, deixando-as informadas de forma automática. E um processo parecido com esse é utilizado na educação à distância, e acontece quando a leitura e atividades podem ser realizadas de outros lugares e de outros aparelhos com acesso a internet (DOCTRINA, 2014). Analisando a ideia da autora, fica fácil concordar com seus pensamentos, quando comparada a atual situação que se encontra o ensino-aprendizagem. Nesse caso, é importante lembrar que as instituições de ensino estão trabalhando de forma parcial, pois passaram por várias e rápidas transformações e adaptações no sistema de ensino tradicional para o on-line. Revista Ibero-Americana de Humanidades, Ciências e Educação. São Paulo, v.7.n.10. out. 2021. ISSN- 2675 – 3375 De acordo com Dariva (2010, p.7): E isso só foi permitido, porque os estudos e as leituras dos alunos passaram a ser feitos na tela dos aparelhos tecnológicos, e as TICs passaram a ter maior notoriedade por mostrar, ao mundo, as enumeras vantagens dos aparelhos tecnológicos. 3507 Vale lembrar, que as variadas formas de leitura e escrita existentes no contexto social, recebem apoio de vários meios de última geração, como a internet, o celular e o computador para serem utilizados na comunicação (RODRIGUES, 2010). Todos esses recursos, citados acima, são muito valorosos porque oferecem para os alunos e para a sociedade geral as vantagens de ler, escrever, estudar e, de forma imediata, enviar para qualquer pessoa o conteúdo através de e-mails ou até mesmo por mensagens instantâneas. E isso tem impactado diretamente na educação. Outra vantagem dessas TICS está na escrita digital, pois ao digitar um texto, a pessoa, dependendo de sua habilidade, poderá fazer varias alterações como: a escolha, o tamanho, a cor da fonte e a chance de reescrita para melhorar a aparência do documento. (DARIVA, 2010). , ISSN- 2675 – 3375 Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE A inclusão das TICs no ensino aprendizagem também permite ao aluno receber um aprendizado com conteúdo mais atualizados, uma vez que a leitura através de sites educativos traz uma maior variedade de tipos de textos, que permite otimizar a interação entre os alunos e o professor de forma mais dinâmica e prazerosa. A utilização do computador progrediu significativamente com relação às cansativas atividades realizadas nos livros tradicionais, uma vez que aquela motiva e atrai mais seus usuários (GARCIA, 2012, p.22) a realizarem atividades pedagógicas, considerando as transformações oriundas das tecnologias e das práticas sociais. Conforme já mencionado, a necessidade de novas formas de aprendizagem não surgiu à toa. O ser humano, em busca de ações que visam facilitar os afazeres do dia a dia, viu nas tecnologias excelente oportunidade que favorecesse a realização de várias tarefas ao mesmo tempo. Além disso, o uso da tecnologia permite levar a seus usuários, vantagens no desenvolvimento de sua formação, tanto no âmbito social, como no âmbito educativo. Uma dessas vantagens seria a leitura digital, que utiliza aparelhos tecnológicos como suporte para oferecer aos leitores a oportunidade de comportar vários livros digitais para ler onde, como e quando quiser. De acordo com Dariva (2010, p.7): Essa nova modalidade de leitura, traz maior dinamicidade e proporciona aos seus leitores outras possibilidades de ler na tela dos aparelhos tecnológicos, denominada de leitura não linear. 3508 Revista Ibero-Americana de Humanidades, Ciências e Educação. São Paulo, v.7.n.10. out. 2021. ISSN- 2675 – 3375 Revista Ibero-Americana de Humanidades, Ciências e Educação. São Paulo, v.7.n.10. out. 2021. ISSN- 2675 – 3375 REFERÊNCIAS BIBLIOGRÁFICAS ASSUNÇÃO, M.S.H. Narrativas digitais: uma proposta para leitura e produção de textos multimodais. 2018. 176 f. Dissertação (Mestrado Profissional em Letras em Rede Nacional - PROFLETRAS) - Faculdade de Letras, Programa de Pós Graduação em Mestrado Profissional em Letras e em Rede Nacional, Universidade Federal de Alagoas, Maceió, 2018. Disponível em:http://www.repositorio.ufal.br/handle/riufal/3186 . Acesso em: 06 de jul. de 2020. AZEVEDO, E.N.; STEYER, F.A. Os Desafios da Escola Pública Paranaense na Perspectiva do Professor PDE. O Uso da Tecnologia no Ensino e Aprendizagem da Leitura. Versão online ISBN 978-85-8015-080-3, 2014. AZEVEDO, E.N.; STEYER, F.A. Os Desafios da Escola Pública Paranaense na Perspectiva do Professor PDE. O Uso da Tecnologia no Ensino e Aprendizagem da Leitura. Versão online ISBN 978-85-8015-080-3, 2014. Revista Ibero- Americana de Humanidades, Ciências e Educação- REASE BOTTENTUIT; JUNIOR, J.B. Do Computador ao Tablet: Vantagens Pedagógicas na Utilização de Dispositivos Móveis na Educação. Revista Educaonline, v. 6, p. 125-149, 2012.CHAGAS, M.F.L. et al. Atuação Docente na Inter-relação dos Letramentos Alfabético e Digital no Ciberespaço. Holos, Natal, v. 6, p.1, dez. 2014. DARIVA. A Leitura na Era Digital: Uma Proposta de Trabalho Pedagógico com o Gênero Discursivo HD. Caderno O Professor PDE e os Desafios da Escola Pública Paranaense. Secretaria de Educação do Estado do Paraná, Volume 1, 2010. Disponível em: http://www.diaadiaeducacao.pr.gov.br/portals/cadernospde/pdebusca/producoes_pde/ 2010/2010_unicentro_port_artigo_seila_maria_de_oliveira.pdf. Acesso em: 14 de jul. de 2020. evista Eletrônica Saberes da Educação, volume 4, n 1, 2013. Disponível em: http://docs.uninove.br/arte/fac/publicações/pdf/v4-n1-2013/Marcia.pdf. Acesso em: 14 de Ago. de 2020. 3509 PEDRÓ, F. A Tecnologia e as Transformações da Educação. Madrid: Fundación Santillana, 2016. PEREIRA, L. Leitura, Gêneros Textuais e Novas Tecnologias. Revista de Educação Ciência e Tecnologia, Canoas, v.1, n.1, 2012. SANTOS, F.M.A. Interferências da Linguagem Digital na Aquisição do Português Escrito. In: Simpósio Hipertexto e Tecnologias na Educação, 3. 2010, Recife. Anais. Disponível em: http:// https://docplayer.com.br/8415096-Interferencias-da-linguagem- digital-na-aquisicao-do-portugues-escrito.html . Acesso em: 13 de jul. de 2020 SILVA, C.G. A Importância do Uso das TICS Na Educação. Revista Científica Multidisciplinar Núcleo do Conhecimento. Ano 03, Vol. 16, pp. 49-59, Ago. de 2018. ISSN:2448-0959 SILVA, C.G. A Importância do Uso das TICS Na Educação. Revista Científica Multidisciplinar Núcleo do Conhecimento. v.16, ed. 08, p.00 – 00, ago. de 2018. SILVA, C.G. A Importância do Uso das TICS Na Educação. Revista Científica Multidisciplinar Núcleo do Conhecimento. v.16, ed. 08, p.00 – 00, ago. de 2018.
https://openalex.org/W4230710683	http://pure.iiasa.ac.at/id/eprint/8473/1/acp-8-7101-2008.pdf	English	null	Climate forcing and air quality change due to regional emissions reductions by economic sector	null	2,008	cc-by	12,392	Climate forcing and air quality change due to regional emissions reductions by economic sector D. Shindell1, J.-F. Lamarque2, N. Unger1, D. Koch1, G. Faluvegi1, S. Bauer1, M. Ammann3, J. Cofala3, and H. Teich1 1NASA Goddard Institute for Space Studies, New York, NY, USA 2National Center for Atmospheric Research, Boulder, CO, USA 3International Institute for Applied Systems Analysis, Laxenburg, Austria D. Shindell1, J.-F. Lamarque2, N. Unger1, D. Koch1, G. Faluvegi1, S. Bauer1, M. Ammann3, J. Cofala3, and H. Teich1 1NASA Goddard Institute for Space Studies, New York, NY, USA 2National Center for Atmospheric Research, Boulder, CO, USA 3International Institute for Applied Systems Analysis, Laxenburg, Austria Received: 16 May 2008 – Published in Atmos. Chem. Phys. Discuss.: 12 June 2008 Revised: 30 September 2008 – Accepted: 26 October 2008 – Published: 8 December 2008 Abstract. We examine the air quality (AQ) and radiative forcing (RF) response to emissions reductions by economic sector for North America and developing Asia in the CAM and GISS composition/climate models. Decreases in an- nual average surface particulate are relatively robust, with intermodel variations in magnitude typically <30% and very similar spatial structure. Surface ozone responses are small and highly model dependent. The largest net RF results from reductions in emissions from the North America indus- trial/power and developing Asia domestic fuel burning sec- tors. Sulfate reductions dominate the ﬁrst case, for which intermodel variations in the sulfate (or total) aerosol opti- cal depth (AOD) responses are ∼30% and the modeled spa- tial patterns of the AOD reductions are highly correlated (R=0.9). Decreases in BC dominate the developing Asia domestic fuel burning case, and show substantially greater model-to-model differences. Intermodel variations in tropo- spheric ozone burden changes are also large, though aerosol changes dominate those cases with substantial net climate forcing. The results indicate that across-the-board emissions reductions in domestic fuel burning in developing Asia and in surface transportation in North America are likely to offer the greatest potential for substantial, simultaneous improve- ment in local air quality and near-term mitigation of global climate change via short-lived species. Conversely, reduc- tions in industrial/power emissions have the potential to ac- celerate near-term warming, though they would improve AQ and have a long-term cooling effect on climate. These broad conclusions appear robust to intermodel differences. Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ © Author(s) 2008. This work is distributed under the Creative Commons Attribution 3.0 License. Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ © Author(s) 2008. This work is distributed under the Creative Commons Attribution 3.0 License. Atmospheric Chemistry and Physics 1 Introduction Changes in AQ and climate typically occur on different tem- poral and spatial scales. As climate change is by deﬁnition a long-term phenomenon, climate policy typically focuses on well-mixed greenhouse gases, as these have strong radiative forcing and are long-lived (residence times of ∼10 years or longer) in the atmosphere. AQ, in contrast, is usually con- cerned with short-lived pollutants (residence times of months or less) which do not currently play as prominent a role in climate policy. The two issues are linked in several ways, however. Tropospheric ozone and aerosols, among the pri- mary short-lived pollutants regulated for AQ purposes, inﬂu- ence global climate change, and conversely are affected by the impact of climate change on factors such as temperature and humidity. Though emissions of long-lived greenhouse gases, especially carbon dioxide, are the main driver of pro- jected future climate change, AQ-related emissions have the potential to substantially affect 21st Century climate as well (Shindell et al., 2008). In addition, AQ-related emissions re- ductions may lead to substantial economic and health bene- ﬁts (West et al., 2006; West et al., 2007). AQ regulations are typically made without consideration of their effect on climate. Different regulatory choices poten- tially affect climate in opposite ways, however, as some AQ- related emissions lead to warming while others lead to cool- ing. Precipitation may also be either enhanced or suppressed by emissions changes resulting from AQ policy. Hence a carefully designed regulatory approach for short-lived pollu- tants could both mitigate some of the effects of global warm- ing and improve AQ, while a poorly designed approach could improve AQ but accelerate warming. Correspondence to: D. Shindell (drew.t.shindell@nasa.gov) ing and improve AQ, w improve AQ but accel Published by Copernicus Publications on behalf of the European Geosciences Union. Correspondence to: D. Shindell (drew.t.shindell@nasa.gov) Correspondence to: D. Shindell (drew.t.shindell@nasa.gov) Published by Copernicus Publications on behalf of the European Geosciences Union. 7102 D. Shindell et al.: Climate and air quality by emission region and sector D. 2 Composition and climate models The components of the GISS model for Physical Understand- ing of Composition-Climate INteractions and Impacts (G- PUCCINI) version used here include photochemistry from the surface to the mesosphere (Shindell et al., 2006), and sul- fate, carbonaceous, sea salt, nitrate, and mineral dust aerosols (Koch et al., 2007; Bauer et al., 2006; Miller et al., 2006). Dust and sea-salt are segregated into size bins, nitrate and sulfate aerosols have ﬁne and coarse modes, while only total mass is simulated for carbonaceous aerosols. The compo- nents interact with one another, with linkages including ox- idants affecting sulfate, gas-phase nitrogen species affecting nitrate, sulfate affecting nitrogen heterogeneous chemistry, thermodynamic ammonium nitrate formation being affected by sulfate, sea salt and mineral dust (Metzger et al., 2006), and sulfate and nitrate being absorbed onto mineral dust sur- faces (Bauer et al., 2006; Bauer et al., 2007). As in our prior CCSP simulations (Shindell et al., 2007), the enhanced con- vective scavenging of aerosols in Koch et al. (2007) was not included, leading to a higher burden of these species relative to that study but values still roughly consistent with obser- vations. The simulations described here were run using a 4 by 5 degree horizontal resolution version of the ModelE climate model with 23-layers extending from the surface to ∼0.01 hPa (Schmidt et al., 2006). Chemistry and dynam- ics are fully coupled in the model (i.e. composition is simu- lated online rather than using previously saved meteorologi- cal ﬁelds). ) These experiments were done in support of the US Cli- mate Change Science Program (CCSP) Synthesis and As- sessment Product 3.2: “Climate Projections Based on Emis- sion Scenarios for Long-lived and Short-lived Radiatively Active Gases and Aerosols”. They were designed to exam- ine the climate forcing due to short-lived species in a policy- relevant way by isolating the inﬂuence of economic activities subject to regulation and/or reduction in usage (e.g. by im- proved efﬁciency). We look at 30% reductions in total emis- sions from domestic fuel usage (residential and commercial fossil and biofuel burning), surface transportation, and from a combined industry and power generation sector. These three are the main anthropogenic sources for emission of short- lived species and their precursors. 1 Introduction Shindell et al.: Climate and air quality by emission region and sector To aid in formulation of policies to address both AQ and climate change, we performed a series of simulations ex- amining the impact of speciﬁc emission sectors on radia- tively active short-lived species using composition and cli- mate models from the Goddard Institute for Space Stud- ies (GISS) and the Community Atmosphere Model (CAM) model, developed in collaboration with the National Center for Atmospheric Research (NCAR). The use of two mod- els allows us to characterize the robustness of the results to different plausible representations of chemical and physical processes, at least to a limited extent. In addition to calcu- lating RF from short-lived species (in the GISS model), we also estimate the RF from long-lived species for comparison. These species were not simulated by the composition mod- els, so their forcing estimates were instead based on the car- bon dioxide (CO2) and methane (CH4) emissions from each sector. We concentrate here on the near-term forcing over the next two decades from potential present-day emissions perturbations. This timescale is chosen to be long enough that the climate system would have sufﬁcient time to re- spond to a substantial portion of the forcing from short-lived species, which occurs virtually right away following emis- sions changes, but not to be so long that AQ-related emis- sions would likely diverge enormously from their current lev- els (the inﬂuence of the choice of time horizon is discussed further in Sect. 5). North America (60–130 W, 25–60 N) and developing Asia (60–130 E, 0–50 N) are studied. These two regions were cho- sen based on their being the focus of CCSP interest and being the largest current emitter of many pollutants, respectively. This work complements prior regional and sectoral pertur- bations studies examining the response of gas-phase species only (Berntsen et al., 2005; Fuglestvedt et al., 1999; Derwent et al., 2001), aerosol species only (Koch et al., 2007), or both but with a subset of the species included here (Unger et al., 2008) or for a single sector for the entire globe (Fuglestvedt et al., 2008). 2 Composition and climate models Note that the combined industry/power sector can be largely separated into its two components for aerosols, as sulfur emissions are largely from power generation while carbonaceous aerosol emissions are mostly from industrial processes (use of the combined sector saves computational resources). Since we consider across- the-board reductions in emissions from a given sector, our results are useful for assessing the potential impacts of re- ductions in total power/fuel usage rather than changes in the mix of power generation/transportation types or in emissions control technologies targeted at speciﬁc pollutants. How- ever, we examine both the net forcing from each sector and the contribution from individual species, so that both the sectors and individual pollutants within sectors that are the most attractive climate mitigation targets can be determined. The responses to separate emissions perturbations in both Simulations were also performed using the Community Atmosphere Model CAM3 (Collins et al., 2006) coupled to the Model for Ozone and Related Tracers (MOZART) chem- istry (Horowitz et al., 2003). MOZART tropospheric chem- istry is an extension of the mechanism presented in Horowitz et al. (2003); changes include an updated terpene oxidation scheme and a better treatment of anthropogenic and natu- ral non-methane hydrocarbons (NMHCs) treated up to iso- prene, toluene and monoterpenes. Aerosols have been ex- tended from the work by Lamarque et al. (2005) to include actively simulated sea-salt and dust aerosols (each segregated into 4 bin sizes, from ﬁne mode to coarse mode) and a repre- sentation of ammonium nitrate that is dependent on sulfate Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ 7103 D. Shindell et al.: Climate and air quality by emission region and sector following the parameterization of gas/aerosol partitioning by Metzger et al. (2002). The chemical and physical processes included are similar to those in the GISS model, with the exception that there is no uptake of sulfate and nitrate onto mineral dust. CAM includes secondary organic aerosols, however, which GISS does not, and these are linked to the gas-phase chemistry through the oxidation of atmospheric NMHCs as in Chung and Seinfeld (2002). Only the bulk mass is calculated for aerosols other than sea-salt and dust, and all aerosols are radiatively active. The horizontal reso- lution is 2 by 2.5 degrees, with 26 levels from the surface to ∼4 hPa, and this model also includes coupled chemistry and dynamics. O3 (ppbv), Dev. 3 Experimental setup Six simulations were performed, each reducing present-day emissions by 30% in one sector for one region. Reductions are uniform for all species in a given sector, and we maintain the same temporal and spatial emission patterns. By using present-day emissions, the results are not tied to any partic- ular future scenario. We use IIASA 2000 sectoral emissions (Table 1) in both models, based on the 1995 EDGAR3.2 in- ventory extrapolated to 2000 using national and sector eco- nomic development data (Dentener et al., 2005). The excep- tion is biomass burning emissions, which are averages over 1997–2002 from (Van der Werf et al., 2003) with emission factors from (Andreae and Merlet, 2001) for aerosols (again in both models). The CAM model did not perform the trans- portation sector simulations. We ﬁrst compare changes in the surface concentrations rel- evant for AQ. We then examine burdens of radiatively active species, the optical depth for aerosols, and ﬁnally RF (though CAM did not calculate RF). To average out meteorological variability, we analyze the last 10 years of 11-year runs (use of the last 8 gives values with negligible differences to using the last 10) relative to an analogous control run. As these runs had constant present-day greenhouse gases and constant year 2000 emissions of short-lived species, they are representative of general present-day conditions rather than any particular years. All simulations kept long-lived greenhouse gases ﬁxed at present day values as these were (generally) not part of the composition models that were used here, which instead focused on shorter-lived pollutants. Changes in CO2 and CH4 in response to changes in anthropogenic emissions of those gases are relatively straightforward to estimate to ﬁrst- order, and do not require full 3-D chemical-transport models (though such estimate do not include all relevant feedbacks, such as the response of the carbon cycle to climate change). RF from changes in emissions of long-lived species based on such estimates is discussed in Sect. 5. Our primary goal, 2 Composition and climate models Asia, Domestic, CAM -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 O3 (ppbv), Dev. Asia, Domestic, GISS -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 BC (pptv), Dev. Asia, Domestic, GISS -1800 -1000 1000 1800 -200 200 BC (pptv), Dev. Asia, Domestic, CAM -1800 -1000 1000 1800 -200 200 SO4 (pptv), Dev. Asia, Domestic, GISS -225 -125 125 225 -25 25 SO4 (pptv), Dev. Asia, Domestic, CAM -225 -125 125 225 -25 25 Fig. 1. Annual average surface pollutant response to 30% reduction in Developing Asian domestic fuel burning sector emissions in the GISS (left) and CAM (right) models for the indicated pollutants. SO4 (pptv), Dev. Asia, Domestic, GISS SO4 (pptv), Dev. Asia, Domestic, CAM SO4 (pptv), Dev. Asia, Domestic, GISS SO4 (pptv), Dev. Asia, Domestic, CAM O3 (ppbv), Dev. Asia, Domestic, GISS -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 -1800 -1000 1000 1800 -200 200 Present-day simulations of trace gases and aerosols in both models agree fairly well with observations over North Amer- ica for all species, while sulfate, BC, OC and CO are all bi- ased low over developing Asia (Shindell et al., 2006; Koch et al., 2007; Horowitz et al., 2003; Lamarque et al., 2005). Such biases are common among models, and have been linked to underestimates of developing Asian emissions in current inventories (e.g. Arellano et al., 2004; P´etron et al., 2004; Bergamaschi et al., 2000). Ratios of positive and negative forcing agents, e.g. BC/OC or SO4/BC, are fairly reasonable in both regions, however. Neither model included indirect aerosol effects or radiative forcing from changes in black car- bon deposition onto snow and ice surfaces. O3 (ppbv), Dev. Asia, Domestic, CAM -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 -1800 -1000 1000 1800 -200 200 -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 Fig. 1. Annual average surface pollutant response to 30% reduction in Developing Asian domestic fuel burning sector emissions in the GISS (left) and CAM (right) models for the indicated pollutants. however, is to use the composition-climate models to cal- culate the RF from all the short-lived species and hence to identify the relative contributions of the sectors and regions. As the forcings were expected to be small, we concentrate on the RF metric, which has little interannual variability, rather than the climate response (which would have a much lower signal-to-noise ratio). Table 1. Regional annual average emissions by sector (Tg/yr). America, Ind/Pow, CAM -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 SO4 (pptv), N. America, Ind/Pow, CAM -225 -125 125 225 -25 25 -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 Fig. 2. As Fig. 1 but for 30% reduction in North American industrial and power generation emissions (decreases in BC are all less than 20 pptv, and hence are not shown). Decreases in North American industrial/power emissions likewise lead to decreases in local pollutant levels (Fig. 2), with the exception of BC for which no signiﬁcant changes are seen. Sulfate decreases exhibit very similar spatial struc- tures in the two models, with maxima over the Ohio River valley and reductions of more than 125 pptv extending over most of the Eastern US. In contrast to the response to reduced developing Asia domestic emissions, in this case the GISS sulfate response is ∼20% larger than its CAM counterpart. Ozone reductions are stronger, and extend over a larger area, in the GISS model as well, as they were in the developing Asia domestic case, with both models showing the greatest response near the western US Gulf Coast. Again the ozone reductions increase to 2–3.5 ppbv during summer over most of the eastern US in the GISS model, but not in the CAM results. Variations in the ozone response in the two models may be related to differences in incorporation of NMHCs and in gas-aerosol coupling (Sect. 2). There is also a signiﬁcant remote response over Greenland in the GISS model (1.5– the two models for both BC and sulfate, but the sulfate re- ductions are much larger in the CAM model than in the GISS model. The total change in particulate is dominated by BC, however, which shows decreases of similar magni- tude (∼20% greater in the CAM model) and a nearly iden- tical spatial structure in the two models. If we focus in on the ∼15 000 km2 with the largest reductions (a scale more suitable for pollution studies, yet not so small as to over- state the abilities of comparatively coarse-grid global mod- els), we ﬁnd a similar result, with BC concentration reduc- tions roughly an order of magnitude greater than those for sulfate, and an approximately 20% larger response in the CAM model (Table 2). Table 1. Regional annual average emissions by sector (Tg/yr). Table 1. Regional annual average emissions by sector (Tg/yr). Sector NOx CO SO2 BC OC CO2 CH4 Industrial/Power: North America 2.7 7 8.8 0.003 0.003 3890 5.8 Developing Asia 3.4 4 16.6 0.17 0.22 5285 6.0 Domestic: North America 0.3 6 0.2 0.06 0.02 745 0 Developing Asia 0.5 126 2.8 2.09 6.80 920 0 Surface North America 3.5 70 0.5 0.18 0.24 1740 0 Transportation: Developing Asia 3.4 38 1.1 0.43 0.60 693 0 Emissions are in Tg N/yr for NOx, Tg S/yr for SO2, in Tg C/yr for BC and OC, and Tg/yr of the species for others. CO2 emissions are from the EDGAR 2000 inventory (updated from Olivier and Berdowski, 2001). The industry and power generation sector includes the EDGAR categories industrial, power generation, other transformation sector (reﬁneries, coke ovens, gas works, etc.), non-energy use and chemical feedstocks, oil production, transmission and handling, and cement manufacturing. Surface transportation includes on and off road transport and international shipping. Domestic includes residential and commercial fossil fuel usage and residential biofuel usage. Methane emissions are based on the coal burning source from (Fung et al., 1991). Sources for other emissions are given in Sect. 3. Sector NOx CO SO2 BC OC CO2 CH4 Industrial/Power: North America 2.7 7 8.8 0.003 0.003 3890 5.8 Developing Asia 3.4 4 16.6 0.17 0.22 5285 6.0 Domestic: North America 0.3 6 0.2 0.06 0.02 745 0 Developing Asia 0.5 126 2.8 2.09 6.80 920 0 Surface North America 3.5 70 0.5 0.18 0.24 1740 0 Transportation: Developing Asia 3.4 38 1.1 0.43 0.60 693 0 Emissions are in Tg N/yr for NOx, Tg S/yr for SO2, in Tg C/yr for BC and OC, and Tg/yr of the species for others. CO2 emissions are from the EDGAR 2000 inventory (updated from Olivier and Berdowski, 2001). The industry and power generation sector includes the EDGAR categories industrial, power generation, other transformation sector (reﬁneries, coke ovens, gas works, etc.), non-energy use and chemical feedstocks, oil production, transmission and handling, and cement manufacturing. Surface transportation includes on and off road transport and international shipping. Domestic includes residential and commercial fossil fuel usage and residential biofuel usage. Methane emissions are based on the coal burning source from (Fung et al., 1991). Sources for other emissions are given in Sect. 3. D. Shindell et al.: Climate and air quality by emission region and sector 7104 Table 1. Regional annual average emissions by sector (Tg/yr). Table 1. Regional annual average emissions by sector (Tg/yr). Emissions are in Tg N/yr for NOx, Tg S/yr for SO2, in Tg C/yr for BC and OC, and Tg/yr of the species for others. CO2 emissions are from the EDGAR 2000 inventory (updated from Olivier and Berdowski, 2001). The industry and power generation sector includes the EDGAR categories industrial, power generation, other transformation sector (reﬁneries, coke ovens, gas works, etc.), non-energy use and chemical feedstocks, oil production, transmission and handling, and cement manufacturing. Surface transportation includes on and off road transport and international shipping. Domestic includes residential and commercial fossil fuel usage and residential biofuel usage. Methane emissions are based on the coal burning source from (Fung et al., 1991). Sources for other emissions are given in Sect. 3. heights, or the transformation of BC from hydrophobic to hy- drophilic. Reductions in surface ozone are generally small, with maxima of about 1–1.5 ppbv in both models but a re- sponse that extends over a larger, more spatially coherent area in the GISS model (Fig. 1). The surface ozone reduc- tion in the GISS model also shows a distinct local maximum over the Yellow Sea co-located with the largest decreases in surface aerosols, while the CAM model does not. Local ozone reductions increase to levels of 1.5–3.5 ppbv during boreal summer over parts of India, China and Korea in the GISS model, but not in the CAM model. Both these annual and summer increases are statistically signiﬁcant in the GISS model (all signiﬁcance is based on the 95% conﬁdence level derived from the interannual variability in the simulations). Reductions in surface pollution in areas far from the source region are not generally statistically signiﬁcant in these sim- ulations. -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 O3 (pptv), N. America, Ind/Pow, GISS O3 (pptv), N. America, Ind/Pow, CAM -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 SO4 (pptv), N. America, Ind/Pow, GISS -225 -125 125 225 -25 25 SO4 (pptv), N. America, Ind/Pow, CAM -225 -125 125 225 -25 25 Fig. 2. As Fig. 1 but for 30% reduction in North American industrial and power generation emissions (decreases in BC are all less than 20 pptv, and hence are not shown). -1.8 -1 .6 1 1.4 1.8 -.2 .2 -.6 O3 (pptv), N. America, Ind/Pow, GISS SO4 (pptv), N. America, Ind/Pow, GISS -225 -125 125 225 -25 25 O3 (pptv), N. 4.1 Surface pollutants Annual average regional reductions take place for all the major pollutants in both models in response to reductions in emissions from developing Asia domestic fuel burning (Fig. 1). The reductions have a similar spatial pattern in Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ D. Shindell et al.: Climate and air quality by emission region and sector Table 1. Regional annual average emissions by sector (Tg/yr). Changes in a particular oxidant can differ by a factor of two or more between the two models, but they are consistent in being small in both cases, with decreases typically about 2– 3% for ozone and less than 1% for OH and hydrogen per- oxide. These reductions are clearly smaller than the sulfate burden decreases in the industrial/power sector experiments. Interestingly, the only case in which an oxidant increases is in the Developing Asia Domestic sector experiment, in which OH increases slightly in both models (0.3% in GISS, 0.5% in CAM). Nevertheless, the sulfate burdens decreased, indi- cating that as in the industrial/power sector runs, changes in sulfate precursor emissions were the primary driver of the burden changes. Changes in BC and ozone burdens are less consistent than those in sulfate between the two models. Re- sponses are generally larger in the GISS model. This is espe- cially so for BC burden changes in response to reductions in developing Asia emissions, in which cases the GISS model responses are about a factor of 2.5–3 times greater than the CAM model responses. 3.5 ppbv summertime decreases), consistent with the strong inﬂuence of North American emissions on surface pollution levels there (Stohl, 2006). In all cases other than North America domestic fuel burn- ing, substantial local reduction in both annual and summer surface particulate concentrations result from the regional sector emissions reductions (Table 2). They are especially large for the developing Asia domestic case, but are also clear in the response to industrial/power and transportation emissions reductions in both regions. The annual average re- ductions are generally quite similar in the two models, with differences of <30% for all but one of the statistically signif- icant local responses (Table 2) and in the large-scale aerosol responses (Figs. 1 and 2). Differences between the two mod- els are much greater for seasonal pollutant responses (Ta- ble 2). Shifts in oxidants in response to ozone precursor emission decreases can sometimes outweigh effects of re- ductions in sulfur-dioxide emissions, leading to increases in surface sulfate in these emission reduction experiments (Ta- ble 2). Increases in surface aerosols are not statistically sig- niﬁcant, however. Table 1. Regional annual average emissions by sector (Tg/yr). Interestingly, the summer surface BC response is larger than the annual mean response in the GISS model, but smaller in the CAM model. Differences in seasonality may be related to variations in how the mod- els simulate convective transport, planetary boundary layer Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ 7105 D. Shindell et al.: Climate and air quality by emission region and sector D. Shindell et al.: Climate and air quality by emission region and sector Table 2. Local change in surface pollutants (pptv) due to 30% emissions reductions in the given region and sector. Region and sector summer SO4 summer BC annual SO4 annual BC North America domestic [105] [−62] [18] [−53] −198 [−70] [−44] [−14] North America surface transportation [75] −235 [36] −169 NA NA NA NA North America industry/power −383 [−47] −268 [−14] −285 [21] −164 [12] Developing Asia domestic −127 −2430 [−64] −1630 −264 −1815 −228 −2049 Developing Asia surface transportation [−119] −791 [−41] −436 NA NA NA NA Developing Asia industry/power −558 −612 −328 −280 −264 −178 −242 −309 Values are averages over comparable areas (∼15 000 km2) containing local maxima with the ﬁrst row giving results from the GISS PUCCINI model and the second from the CAM MOZART model for each sector. Values in brackets are not statistically signiﬁcant (signiﬁcance is based on the 95% conﬁdence level derived from the interannual variability). hange in surface pollutants (pptv) due to 30% emissions reductions in the given region and sector. Values are averages over comparable areas (∼15 000 km2) containing local maxima with the ﬁrst row giving results from the GISS PUCCINI model and the second from the CAM MOZART model for each sector. Values in brackets are not statistically signiﬁcant (signiﬁcance is based on the 95% conﬁdence level derived from the interannual variability). largest changes are decreases of the global sulfate burden by ∼5.5% for reductions in developing Asia industrial/power emissions, by ∼3–4% for reductions in North American in- dustrial/power emissions, and by 0.6–0.8% for reductions in developing Asia domestic fuel burning emissions. The two models’ changes in the sulfate burden in Tg agree within 5–12% (Table 3). Both models produce small decreases in oxidants which contribute to the decreased sulfate burdens. Table 1. Regional annual average emissions by sector (Tg/yr). Nearly all the aerosols at the surface are in ﬁne mode, so that the PM2.5 (particulate matter with a mean radius of <2.5 microns) concentration changes are approxi- mately the sum of the BC and sulfate concentration changes given in Table 2 (additional contributions from OC and ni- trate are comparatively small). D. Shindell et al.: Climate and air quality by emission region and sector 7106 Table 3. Tropospheric burden changes due to 30% emissions reductions in the given region and sector (Tg and %). Region and sector SO4 SO4 BC BC O3 O3 GISS CAM GISS CAM GISS CAM North America domestic [−0.0020] [−0.0006] −0.0029 −0.0017 [0.12] −0.65 (−0.1%) (−0.1%) (−1.1%) (−0.8%) (−0.4%) (−2.3%) North America surface transportation [−0.0029] NA −0.0015 NA −0.69 NA (−0.2%) (−0.6%) (−2.2%) North America industry/power −0.0460 −0.0484 [−0.0006] [0.0004] −1.20 −0.55 (−2.9%) (−3.7%) (−0.2%) (0.2%) (−3.8%) (−2.0%) Developing Asia domestic −0.0091 −0.0101 −0.0302 −0.0093 −1.14 −0.65 (−0.6%) (−0.8%) (−11.6%) (−4.4%) (−3.6%) (−2.3%) Developing Asia surface transportation −0.0068 NA −0.0056 NA −1.31 NA (−0.4%) (−2.2%) (−4.1%) Developing Asia industry/power −0.0834 −0.0733 −0.0027 −0.0009 −0.87 −0.55 (−5.3%) (−5.6%) (−1.0%) (−0.4%) (−2.8%) (−2.0%) Values in brackets are not statistically signiﬁcant (as in Table 2). Values in parentheses are the percent change relative to each model’s own burden in its control run. Tropospheric burden changes due to 30% emissions reductions in the given region and sector (Tg and %). signiﬁcant (as in Table 2). Values in parentheses are the percent change relative to each model’s own Values in brackets are not statistically signiﬁcant (as in Table 2). Values in parentheses are the percent change relative to each model’s own burden in its control run. are even consistent in showing a reduction in sulfate AOD over East Asia due to decreases in oxidants stemming from reduced emissions of ozone precursors in North America. Note that the ∼30% difference in the AOD change between the models is larger than the ∼10% difference in the burden change. x 10-2 -3 -2 -1.2 .8 1.2 2 3 -.4 .4 -.8 Developing Asia, Domestic, GISS -.15 Developing Asia, Domestic, CAM -.12 North America, Industrial/Power, GISS -.09 North America, Industrial/Power, CAM -.12 Fig. 3. Annual average total aerosol optical depth change due to 30% reductions in emissions from the given region and economic sector in the GISS (top) and CAM (bottom) models. Changes greater than ∼0.012 are statistically signiﬁcant (95%). Values in the upper right give the global annual mean. x 10-2 -3 -2 -1.2 .8 1.2 2 3 -.4 .4 -.8 Developing Asia, Domestic, GISS -.15 Developing Asia, Domestic, CAM -.12 North America, Industrial/Power, GISS -.09 North America, Industrial/Power, CAM -.12 For developing Asia domestic sector emission reductions, BC is the largest aerosol change (in absolute or percentage terms). D. Shindell et al.: Climate and air quality by emission region and sector As noted, the total AOD changes in the two models are similar (Fig. 3). In this case, the contribution to AOD from BC was not available for the CAM model, though to- tal carbonaceous aerosol AOD changes were recorded. The AOD change from BC+OC is −0.0013 in the GISS model and −0.0008 in the CAM model. Much of the 47% greater response in the GISS model appears to be due to a larger ef- fect of OC on the AOD (the GISS AOD change due to OC is −0.0008, as large as the total AOD change from BC+OC in the CAM model). However, there may be important dif- ferences in the AOD change due to BC as well, to which the greater BC burden change in the GISS model certainly contributes. Also in the developing Asia domestic case, the AOD change from sulfate is twice as large in the CAM model as in the GISS model (−0.0004 and −0.0002, respectively). This is consistent with the sulfate burden changes (Table 3), though AOD shows a larger relative difference than burden. Conversely, the BC burden changes differ much more than the AOD changes due to carbonaceous aerosols. This high- lights how AOD (and RF) depend on several factors, includ- ing aerosol optical properties, water uptake, and vertical dis- tribution, as well as on aerosol mass. x -3 -2 -1.2 .8 1.2 2 3 -.4 .4 -.8 Fig. 3. Annual average total aerosol optical depth change due to 30% reductions in emissions from the given region and economic sector in the GISS (top) and CAM (bottom) models. Changes greater than ∼0.012 are statistically signiﬁcant (95%). Values in the upper right give the global annual mean. (Lamarque et al., 2005; Koch et al., 2007). The greatest re- duction in AOD for the developing Asia perturbations is in the domestic case, while for North American it is in the in- dustrial/power case. The GISS and CAM models show sim- ilar AOD decreases for these two perturbation experiments (Fig. 3). The two models’ responses are highly correlated spatially (R=0.8) and the global mean values differ by only 20–30%. For North American industrial/power emissions, the main aerosol change is sulfate. 4.2 Burdens and aerosol optical depths The climate inﬂuence of aerosols is not determined solely by the change in concentration or mass, but by the resulting change in optical properties. To examine this, we compare the AOD changes in the two models. AOD calculations are based upon Mie scattering theory. Dry particle sizes are pre- scribed for each aerosol type, with actual sizes and scattering efﬁciencies dependent upon uptake of water in both models We now turn to global scale changes and their potential inﬂu- ence on climate. We begin by examining the global annual average change in the tropospheric burdens of radiatively ac- tive species in the two models (Table 3). The models’ sul- fate responses are quite consistent with one another. The Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ D. Shindell et al.: Climate and air quality by emission region and sector D. Shindell et al.: Climate and air quality by emission region and sector D. Shindell et al.: Climate and air quality by emission region and sector Area weighted averages over 60–90 N, 28–60 N, and 28 S to 28 N are given on the edges. Signiﬁcance thresholds (95%) are roughly 100 mW/m2, so that local responses are generally signiﬁcant while of the “remote” responses only the negative forcing covering much of the NH in the developing Asia domestic sector simulation is signiﬁcant. and spatial pattern of AOD changes. For the domestic fuel burning sector emissions reductions, where BC has the great- est impact on climate, the aerosol changes are less robust be- tween the two models for either the decrease in atmospheric mass of BC or the AOD reductions due to individual aerosol species. Nevertheless, the total AOD change due to reduc- tions in emissions from this sector is fairly similar in both magnitude and spatial pattern (Fig. 3). Although decreases in absorbing and reﬂective aerosols will clearly have opposing inﬂuences on climate, both models ﬁnd decreases in sulfate burdens of <1% and substantially greater percentage reduc- tions in BC burdens. Combined with the larger forcing per Tg BC than sulfate, this indicates that RF will be dominated by BC changes and hence that the aerosol forcing is at least qualitatively consistent in these models, especially in its spa- tial structure (see also Sect. 4.3). cases where the net RF from short-lived species emissions is greatest: developing Asia domestic and North American industrial/power (Fig. 3 versus Fig. 4). Correlation values either with total RF, as in Fig. 4, or with aerosol RF only are within 0.1 of each other, emphasizing the dominance of aerosol forcing over ozone forcing in the largest local RF changes. Hence the AOD serves as a useful rough indica- tor of RF, especially for regions where the forcing is great- est. In the industrial/power cases, where AOD changes are largely from sulfate and are in reasonable agreement in the models, both the magnitude and spatial pattern of the RF are also likely to be quite consistent. The standard deviation in global mean RF per unit global mean AOD was 32% in a re- cent model intercomparison (Schulz et al., 2006), however, so that the total GISS/CAM intermodel variation in RF from aerosols may be slightly greater than that for AOD. D. Shindell et al.: Climate and air quality by emission region and sector The values of the global mean change in sulfate AOD in the two models are within 30%, and the spatial correlation of the AOD decreases is R=0.9 (the total AOD change is so strongly dominated by sulfate in this case that the response pattern and magnitude are virtually identical to those shown in Fig. 3). The models Hence for industrial/power sector emissions reductions, where sulfate is the dominant aerosol change, the total aerosol response appears to be reasonably robust across these two models in terms of the burden change and the magnitude Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ www.atmos-chem-phys.net/8/7101/2008/ D. Shindell et al.: Climate and air quality by emission region and sector 7107 Developing Asia, Domestic North America, Industrial/Power Developing Asia, Industrial/Power Developing Asia, Transportation North America, Transportation North America, Domestic -49 38 6 } } -45 -33 -11 } } } -1 -3 -2 } } } } -94 -108 -34 } } } -27 3 5 } } } -25 -28 -14 } } } Fig. 4. Annual average instantaneous tropopause radiative forcing (mW/m2) from short-lived species in the GISS model due to 30% reduc- tions in emissions from the given region and economic sector. Values include the direct forcing from ozone and aerosols, but not indirect aerosol or chemical effects (O3 LT, methane indirect, and sulfate indirect in Table 4). Area weighted averages over 60–90 N, 28–60 N, and 28 S to 28 N are given on the edges. Signiﬁcance thresholds (95%) are roughly 100 mW/m2, so that local responses are generally signiﬁcant while of the “remote” responses only the negative forcing covering much of the NH in the developing Asia domestic sector simulation is signiﬁcant. Developing Asia, Domestic North America, Industrial/Power Developing Asia, Industrial/Power Developing Asia, Transportation North America, Transportation North America, Domestic -49 38 6 } } -45 -33 -11 } } } -1 -3 -2 } } } } -94 -108 -34 } } } -27 3 5 } } } -25 -28 -14 } } } Developing Asia, Domestic North America, Domestic Fig. 4. Annual average instantaneous tropopause radiative forcing (mW/m2) from short-lived species in the GISS model due to 30% reduc- tions in emissions from the given region and economic sector. Values include the direct forcing from ozone and aerosols, but not indirect aerosol or chemical effects (O3 LT, methane indirect, and sulfate indirect in Table 4). D. Shindell et al.: Climate and air quality by emission region and sector Part of the enhanced variability in RF versus AOD (and the discrep- ancy between contributions of individual species to AOD and RF) likely comes from the dependence of forcing on the ver- tical position of aerosols relative to clouds. As global AOD observations are largely clear-sky, it is difﬁcult to evaluate this aspect of the simulations. Note that the overall aerosol RF uncertainty in both models would be substantially larger including aerosol indirect effects. 4.3 Radiative forcing Values are instantaneous tropopause forcings annually averaged and are grouped by the timescales for the emissions causing the indicated forcings. For methane, indirect forcing via changes in oxidants is included with ozone and aerosols in “all short-lived”, while direct forcing from reductions in methane emissions is given under long-lived. For ozone, short-term (ST) response is that due to changes in ozone precursor and aerosol emissions, while long-term (LT) is that due to changes in methane (see text for additional details). Signiﬁcance thresholds (90%) for individual forcings are ∼4 mW/m2. Values are instantaneous tropopause forcings annually averaged and are grouped by the timescales for the emissions causing the indicated forcings. For methane, indirect forcing via changes in oxidants is included with ozone and aerosols in “all short-lived”, while direct forcing from reductions in methane emissions is given under long-lived. For ozone, short-term (ST) response is that due to changes in ozone precursor and aerosol emissions, while long-term (LT) is that due to changes in methane (see text for additional details). Signiﬁcance thresholds (90%) for individual forcings are ∼4 mW/m2. We also perform an ofﬂine calculation of the steady-state methane response to changes in modeled oxidants for all six experiments. With methane prescribed at present-day val- ues, changes in modeled methane oxidation are due solely to changes in oxidizing agents. Note that the model’s oxida- tion rate fully captures spatial and seasonal variations. We also include the feedback of methane on its own lifetime (using δln(OH)=−0.32δln(CH4) as recommended in Prather et al., 2001). Finally, we calculate the RF from these indi- rect methane changes as in Ramaswamy et al. (2001). We consider the direct response of methane to methane emis- sions changes to be part of the response to emissions of long-lived species (Sect. 5), but include the indirect re- sponse of methane to oxidant changes as part of the RF due to short-lived species emissions (as it results from change in emissions of short-lived species and their precursors). Tropospheric ozone changes take place with two distinct timescales. A nearly instantaneous response, which we term O3 short-term (ST) occurs as a result of changes in emissions of NOx, CO, VOCs and SO2, while a slower response, O3 long-term (LT) occurs owing to the decadal timescale change in methane due to these same emission changes. 4.3 Radiative forcing RF was calculated in the GISS model (Fig. 4), and is given by the difference between instantaneous tropopause radiative ﬂuxes calculated with and without the presence of each indi- vidual species, compared to the same calculation in the con- trol run. Unsurprisingly, the results are closely related to the AOD changes. The spatial pattern of the AOD changes is in fact highly correlated (R=0.7) with the areas of substan- tial (greater than ±100 mW/m2) RF in the GISS model in the Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ 7108 D. Shindell et al.: Climate and air quality by emission region and sector D. Shindell et al.: Climate and air quality by emission region and sector Table 4. Global mean radiative forcing due to 30% emissions reductions in the given region and sector in the GISS model (mW/m2). North America North America North America Developing Developing Developing domestic surface industry and Asia domestic Asia surface Asia industry transportation power transportation and power Short-lived Emissions Sulfate, direct 0 −3 14 0 2 13 Sufate, indirect 0 −5 24 0 3 21 Black carbon −3 −5 −2 −42 −8 −4 Organic carbon 2 0 −1 13 1 0 Nitrate 1 1 0 1 2 −1 All aerosols 0 −12 35 −28 0 29 Ozone, ST 2 −5 5 −12 −5 −1 Ozone, LT 0 1 −2 −1 3 −1 CH4, indirect 1 4 2 −2 8 6 All short-lived 4 −14 42 −41 1 34 Long-lived Emissions CO2, 20 year −8 −21 −48 −9 −8 −60 CH4, direct 0 0 −8 0 0 −9 Total −4 −35 −14 −50 −6 −35 Values are instantaneous tropopause forcings annually averaged and are grouped by the timescales for the emissions causing the indicated forcings. For methane, indirect forcing via changes in oxidants is included with ozone and aerosols in “all short-lived”, while direct forcing from reductions in methane emissions is given under long-lived. For ozone, short-term (ST) response is that due to changes in ozone precursor and aerosol emissions, while long-term (LT) is that due to changes in methane (see text for additional details). Signiﬁcance thresholds (90%) for individual forcings are ∼4 mW/m2. Table 4. Global mean radiative forcing due to 30% emissions reductions in the given region and sector in the GISS model (mW/m2). diative forcing due to 30% emissions reductions in the given region and sector in the GISS model (mW/m2). Atmos. Chem. Phys., 8, 7101–7113, 2008 5 Relative importance of long-lived and short-lived emissions To explore the total forcing in response to both short- and long-lived species emissions reductions from our three sec- tors, we calculate forcing based on the estimated atmospheric carbon dioxide response to 30% reductions in current emis- sions from each sector (Table 4). For these calculations, we use sectoral CO2 emissions from a year 2000 inventory (updated from Olivier and Berdowski, 2001). Atmospheric concentration changes and the resulting forcing are calcu- lated for the 20-year time horizon, as in the Intergovernmen- tal Panel on Climate Change Third Assessment Report (Ra- maswamy et al., 2001). We believe that using a longer time horizon, such as the 50 or 100 year values also used in that re- port, is not as useful for comparison with forcing from short- lived emissions. As short-lived species responses times are less than a year, use of a many decades long timescale would in a sense be assuming that changes in short-lived species precursor emissions took place and then were maintained without further modiﬁcation for another 50 or 100 years, an implausible scenario for most emission changes. For exam- ple, mandating a shift from incandescent to more efﬁcient compact ﬂuorescent lighting (CFL) during the next 10 years would lead to a reduction in emissions from power genera- tion relative to continued use of the less efﬁcient technology. While the resulting decrease in CO2 emissions would indeed still affect atmospheric concentrations a century later, short- lived species emissions resulting from power generation to provide lightning at that time will depend upon the power consumption of 22nd Century lighting equipment, which is unlikely to have any clear relationship with a present-day mandate for CFL technology. Hence the 50 and 100 year time horizons are not well-suited to short-lived species emis- sions. Conversely, using a very short timescale of only a Coupling between aerosols and oxidants is clear in the re- sponses. Ozone increases in response to reductions in North American emissions from the domestic or industrial/power sectors despite reduced emissions of ozone precursors, es- pecially NOx for which the industrial/power sector in par- ticular is a large source. This response is due to reductions in SO2 emissions, which lead to altered radiative ﬂuxes and to reduced heterogeneous nitrogen chemistry on sulfate sur- faces (Bell et al., 2005; Liao and Seinfeld, 2005). 4.3 Radiative forcing The former is calculated by the composition model, while the latter is es- timated based on the methane response to oxidant changes. As both effects result from change in emissions of short-lived species, both are included in the short-lived portion of Ta- ble 4. In the developing Asia domestic case, the total AOD changes are again comparable in magnitude and distribution. However, the contribution from sulfate (a cooling agent) differes substantially, and given the large difference in BC burden changes and in carbonaceous aerosol optical depths (from warming BC and cooling OC), the net RF in the CAM model may have a substantial different magnitude than that in the GISS model. Nonetheless, the relatively consistent re- sponse of total AOD in the GISS and CAM models (Fig. 3) suggests that the spatial pattern of RF will be fairly consis- tent. We have evaluated the RF due to reductions in the emis- sions of short-lived species and their precursors in the GISS model for each individual forcing agent (Table 4). In addition to the direct forcing due to ozone and aerosol changes, we include within the response to short-lived emissions several indirect effects. The indirect effects of aerosols on clouds are thought to be important, but the RF from these effects is quite uncertain, with recent model estimates differing by roughly a factor of ﬁve (Penner et al., 2006). Hence rather than calculate these forcings directly in the models (which is computationally expensive as well as uncertain), we estimate their magnitude as in Fuglestvedt et al. (2008) in which the indirect forcing is simply 1.5 times the direct sulfate forcing over land and 2 times the direct sulfate forcing over oceans, based on (Kvalev˚ag and Myhre, 2007). Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ 7109 D. Shindell et al.: Climate and air quality by emission region and sector D. Shindell et al.: Climate and air quality by emission region and sector (Fig. 4). This “remote” forcing is in addition to a strongly localized maximum near the source region. We can now examine the contribution of individual species to the RF from emissions perturbations in the GISS model and compare the totals across sectors and regions. The effect of the perturbations is generally larger for develop- ing Asian than North American emissions. 4.3 Radiative forcing The only RF from an individual species to reach 10 mW/m2 from a North American perturbation is the sulfate forcing from indus- trial/power emissions. In contrast, forcings from sulfate, BC, OC and ozone exceed 10 mW/m2 in response to perturba- tions in developing Asia, with the largest response due to decreased BC when domestic fuel burning emissions are re- duced (−42 mW/m2). The total RF due to reduced short- lived species emissions does not show such a simple rela- tionship between the two regions. The net RF from short- lived species shows comparably large values for both regions for the industry/power generation sector (∼30–40 mW/m2), a larger response to surface transportation for North Amer- ica, and larger response to domestic fuel burning emissions reductions from developing Asia. The largest positive forc- ing comes from the industrial/power sector, while the great- est negative forcing comes from developing Asia domestic fuel burning, whose net RF is almost exactly equal to the RF from BC alone. For this sector, the net short-lived emission RF for developing Asia is an order of magnitude larger than for North America, primarily owing to the much larger RF from BC decreases. Conversely, the net RF from short-lived emissions from the surface transportation sector is an order of magnitude larger for North America than for developing Asia, largely due to the difference in the sulfur content of fuels. Though forcings were calculated for the GISS model only, we have shown that values for RF from sulfate are likely to be fairly consistent in the two models given the similarity in modeled AOD changes, indicating that the response to in- dustrial/power sector perturbations is relatively robust. Un- certainties in the response of BC are larger, and could have a substantial effect on RF. Given the dominance of BC in the RF from the domestic fuel burning sector, the net RF may be different in magnitude than that shown here but is likely to be a substantial negative value despite uncertainties. The RF from tropospheric ozone is also uncertain given the large dif- ferences in the burden changes in the two models. This will lead to additional uncertainty in the net RF, especially for the surface transportation sector for which the ozone response plays a substantial role, albeit still a small one compared to the inﬂuence of aerosols. 5 Relative importance of long-lived and short-lived emissions Analo- gous results were seen previously for the response of ozone to emissions changes in these sectors in 2030 (Unger et al., 2008). Similarly, sulfate abundances are enhanced in re- sponse to reductions in North American surface transporta- tion emissions. This occurs because reductions in CO and VOC emissions shift the OH/HO2 ratio toward OH, enhanc- ing gas-phase sulfate formation. This indirect inﬂuence of oxidant changes outweighs the direct impact of reduced SO2 emissions for North American surface transportation, with low-sulfur fuels, but the opposite is true for developing Asia surface transportation emissions (Table 4). The spatial pattern of the RF reveals that aside from devel- oping Asia surface transportation and North America domes- tic, the RF in response to emissions reduction from the other four region/sector combinations all show substantial forcing extending over large parts of the Northern Hemisphere (NH) Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ 7110 D. Shindell et al.: Climate and air quality by emission region and sector few years or less would overemphasize the response to short- lived species emissions. This response is much faster than the response to long-lived species emissions and can give opposite forcing to the longer term response when looking at all effects (Table 4) or even when looking only at the ef- fect of short-lived precursor emissions (Wild et al., 2001; Fu- glestvedt et al., 1999). While the RF response to short-lived species emissions changes is nearly instantaneous (typically days to months for all but those forcing that take place via methane changes), we believe the 20-year time horizon pro- vides a reasonable balance between the effects of long- and short-lived species, is a widely used time horizon in the liter- ature, and is consistent with timescales for air quality policy. Using a longer time horizon would of course give larger rel- ative importance to forcing from long-lived emissions, and vice-versa, and we stress that there is no “best” timescale. term total forcings can be substantially different from one another. This is especially important in the case of indus- trial/power generation emissions reductions, for which the short and long term forcings are of the opposite sign. They would hence lead to near-term warming but long-term cool- ing. 6 Discussion and conclusion The key results for radiative forcing can be summarized as follows. Reductions in surface transportation emissions in North America have a net cooling effect (negative forcing). Negative forcing from long-lived emissions is compounded by negative forcing from short-lived emissions, which results from the net effect of several factors of comparable magni- tude: (1) reductions in BC (which is largely from diesel in this sector), (2) reduction in ozone precursors, which leads to direct reductions in ozone and indirect increases in methane (via reduced oxidation capacity), and (3) decreases in CO and VOC emissions which shift the OH/HO2 ratio toward OH, thereby indirectly leading to increased sulfate (Table 4), as noted previously. Note that the contrasting inﬂuence of OH changes on sulfate and methane reﬂects the difference between the more global methane oxidation and the more regional sulfate oxidation, and implies that these indirect ef- fects are quite sensitive to the NOx/(CO+VOCs) emission ra- tio. Direct impacts on sulfate via SO2 emission changes are small owing to the low-sulfur fuel used in North America. Reductions in developing Asia domestic emissions also have a substantial cooling effect via short-lived species emissions which results primarily from reductions in BC and ozone. There is a net decrease in surface particulate and ozone in both cases (Table 2), and hence reducing emissions from these regional sectors offers a way to simultaneously improve human health and mitigate climate warming. In contrast, industrial/power generation emissions have their largest ef- fect on climate via short-lived emissions through sulfate, and hence the net near-term effect of reductions in short-lived species emissions is a positive forcing. We also calculate the methane response to 30% reductions in regional fossil fuel methane emissions. Emissions are taken from (Fung et al., 1991), with the coal burning source assigned to the power generation sector. As noted previously, simulations did not include interactive methane. We instead calculate the response of methane to the reduced emissions ofﬂine, again including the feedback of methane on its own lifetime. We also include the longer-term response of ozone to the methane changes, the O3 LT response discussed pre- viously. While strictly speaking this portion of the O3 LT response should be included under long-term emissions, the values are so small that we group all the O3 responses to- gether in Table 4. 5 Relative importance of long-lived and short-lived emissions In the cases where emissions reductions lead to sub- stantial negative forcing via short-lived emissions, develop- ing Asia domestic fuel burning and North America surface transportation, inclusion of the short-lived pollutants dramat- ically enhances near-term climate change mitigation poten- tials of these sectors relative to consideration of the effects of long-lived emissions alone. 6 Discussion and conclusion As with the indirect methane response to short-lived precursor emissions, RF is calculated as in Ra- maswamy et al. (2001). Looking at the response to long-lived emissions alone, re- ductions in North American industrial/power generation sec- tor emissions are more than twice as effective in mitigat- ing RF as reductions in the surface transportation sector. The inclusion of the response to emissions of short-lived species alters the comparison markedly, with net total RF from North American surface transportation emissions re- ductions now more than double that for the North American industrial/power generation sector. For developing Asia, the industrial/power generation sector is by far the most effective for CO2, and long-lived emissions in general, but the inclu- sion of the response to short-lived species emissions makes the total net RF from the domestic fuel burning sector larger. Hence the inclusion of RF due to short-lived species emis- sions reductions strongly shifts the relative importance of emissions from the different economic sectors when look- ing at the 20-year time horizon. Relative to the values due to CO2 and CH4 (direct), adding the effects of short-lived pol- lutants increases the forcing from the developing Asia do- mestic fuel burning sector several times over, enhances the forcing from North American transportation sector emissions by ∼2/3, and reduces the net forcing from industrial/power emissions by ∼50–75%. We reiterate that the timescales for these species differ dramatically, so that short-term and long- Overall, developing Asia domestic emissions offer the strongest leverage on near-term climate forcing via short- lived species emissions. This is partially a result of their magnitude, and also because they yield a much greater ef- fect via BC changes than via sulfate changes. It also stems from their occurrence at lower latitudes than North American (or European) emissions, which enhances their photochemi- cal and radiative impacts as incoming radiation is more abun- dant at lower latitudes. The GISS and CAM results differ in Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ 7111 D. Shindell et al.: Climate and air quality by emission region and sector photochemistry involving an assortment of NMHCs, so that differences in the models’ chemical mechanisms may play a role in the greater uncertainties in surface ozone than in sur- face aerosols. 6 Discussion and conclusion Our analyses suggest that changes in aerosol emissions of- ten have the largest potential leverage on near-term climate forcing via short-lived species emissions. Due to large inter- model differences, the role of ozone changes cannot yet be as clearly assessed, although it is likely that these will have a smaller but still important effect both on air quality and cli- mate. Further work is required to more thoroughly character- ize the robustness of these conclusions across a larger num- ber of models and emissions inventories, to explore the im- pact of aerosol mixing and aerosol indirect effects on clouds more fully, to explore additional sectors such as agriculture or sub-sectors such as diesel transportation, and to exam- ine alternative emissions scenarios considering changes in the mix of sources constituting a given sector and the inﬂu- ence of potential technological changes. The latter could be designed to reduce emissions of particular pollutants, while not affecting others. Our results for the RF from individual species provide a guide to the potential impact of such tech- nologies. However, we note that these technologies could also have effects on overall fuel consumption by altering the efﬁciency of a particular process. The spatial pattern of RF varies substantially between the regional sectors. The RF reduction from decreases in de- veloping Asia domestic emissions extends over much of the NH (Fig. 4). Hence emission changes from this region could in some cases have a larger near-term effect on the RF in other parts of the NH via short-lived species emissions than changes in those areas’ own local emissions. Additionally, reductions in transportation or domestic sector emissions cause zonal mean RF due to short-lived species emissions of the same sign throughout the NH, but reductions in in- dustrial/power emissions induce a substantial negative RF in the Arctic while yielding positive RF at lower latitudes in the GISS model (Fig. 4). In this instance, sulfate reduc- tions dominate the forcing closer to the emissions at lower latitudes, while BC and ozone are the most important (and roughly comparable) in the Arctic. y p p While there are many difﬁculties associated with inclu- sion of short-lived species emissions in international treaties (Rypdal et al., 2005; Bond and Sun, 2005), there are also many tangible near-term beneﬁts in human health, agricul- ture and forestry that result from improved AQ. 6 Discussion and conclusion Additionally, surface pollution levels are par- ticularly important to human health in highly polluted areas, which often cover small areas and are difﬁcult to simulate correctly in relatively low-resolution global models. Thus although the responses are similar in these two models, they may suffer from biases common to both. Hence our ability to predict both the AQ and RF responses to emissions pertur- bations has important limitations due to the relatively coarse resolution and simpliﬁed aerosol and cloud physics used in current global models. the magnitude of the total AOD change resulting from the developing Asia domestic sector perturbations by 22%, and this sector/region has the largest inﬂuence in both models for both AOD and surface pollution. However, substantial differ- ences are seen in the two models’ BC change, which domi- nates the RF in this case, and further work to understand the differences in the aerosol simulations in these models, which are substantial for present-day climatologies as well as for many aspects of the response to perturbations (Shindell et al., 2008), would clearly be useful. ) y Identical emissions inventories were used in the two mod- els. Emissions can be difﬁcult to quantify, however, espe- cially for carbonaceous aerosols. The inventory of Bond et al. (2004), for example, has anthropogenic non-biomass burning emissions that are 46% greater for BC and 57% greater for OC for North America than those used here. Sim- ilarly, for developing Asia it has 17% less BC and 19% less OC. The 1984 emissions inventory of Cooke et al. (1999) has much larger (26–77%) carbonaceous aerosol emissions for both regions from fossil fuels than Bond et al. (2004). Thus the magnitude of the BC and OC forcings calculated here might be fairly different using one of these other inven- tories. The sign of the net BC+OC effect appears robust, however, as the ratio of emissions of warming BC to cooling OC is within 10% in the three inventories. The lone excep- tion is that the BC/OC ratio is 40% more for North America in Cooke et al. (1999) compared with Bond et al. (2004). Even this would simply enhance the net cooling effect seen here in response to reduced emissions from the surface trans- portation sector there, however, and would not dramatically alter our conclusions. Uncertainties in emissions of sulfate and ozone precursors are generally smaller. References Res., 105, 1909–1928, 2000. Lamarque, J.-F., Kiehl, J. T., Hess, P. G., Collins, W. D., Em- mons, L. K., Ginoux, P., Luo, C., and X., T. X.: Response of a coupled chemistry-climate model to changes in aerosol emis- sions: Global impact on the hydrological cycle and the tropo- spheric burdens of OH, ozone and NOx, Geophys. Res. Lett., 16, L16809, doi:10.1029/2005GL023419, 2005. Berntsen, T. K., Fuglestvedt, J. S., Joshi, M. M., Shine, K. P., Stu- ber, N., Ponater, M., Sausen, R., Hauglustaine, D. A., and Li, L.: Response of climate to regional emissions of ozone precur- sors: Sensitivities and warming potentials, Tellus B, 57, 283– 304, 2005. Liao, H. and Seinfeld, J. H.: Global impacts of gas-phase chemistry-aerosol interactions on direct radiative forcing by an- thropogenic aerosols and ozone, J. Geophys. Res., 110, D18208, doi:10.1029/2005JD005907, 2005. Bond, T., Streets, D., Yarber, K. F., Nelson, S. M., Woo, J.-H., and Klimont, Z.: A technology-based global inventory of black and organic carbon emissions from combustion, J. Geophys. Res., 109, D14203, doi:10.1029/2003JD003697, 2004. Metzger, S., Dentener, F., Pandis, S., and Lelieveld, J.: Gas/aerosol partitioning: 1. A computationally efﬁcient model, J. Geophys. Res., 107, 4312, doi:10.1029/2001JD001102, 2002. Bond, T. C. and Sun, H.: Can reducing black carbon emissions counteract global warming?, Environ. Sci. Technol., 39, 5921– 5926, 2005. Metzger, S., Mihalopoulos, N., and Lelieveld, J.: Importance of mineral cations and organics in gas-aerosol partitioning of reac- tive nitrogen compounds: Case study based on MINOS results, Atmos. Chem. Phys., 6, 2549–2567, 2006, http://www.atmos-chem-phys.net/6/2549/2006/. Chung, S. H. and Seinfeld, J. H.: Global distribution and climate forcing of carbonaceous aerosols, J. Geophys. Res., 107, 4407, doi:10.1029/2001JD001397, 2002. Collins, W. D., Rasch, P. J., Boville, B. A., Hack, J. J., McCaa, J. R., Williamson, D. L., Briegleb, B. P., Bitz, C. M., Lin, S.-J., and Zhang, M.: The formulation and atmospheric simulation of the Community Atmosphere Model: CAM3, J. Climate, 19, 2144– 2161, 2006. Miller, R. L., Cakmur, R. V., Perlwitz, J., Geogdzhayev, I. V., Ginoux, P., Koch, D., Kohfeld, K. E., Prigent, C., Ruedy, R., Schmidt, G. A., and Tegen, I.: Mineral dust aerosols in the NASA Goddard Institute for Space Studies modelE AGCM, J. Geophys. Res., 111, D0208, doi:10.1029/2005JD005796, 2006. Cooke, W. F., Liousse, C., Cachier, H., and Feichter, J.: Con- struction of a 1 x 1 fossil fuel emission data set for carbona- ceous aerosol and implementation and radiative impact in the ECHAM4 model, J. References Fuglestvedt, J., Berntsen, T., Myhre, G., Rypdal, K., and Skeie, R. B.: Climate forcing from the transport sectors, P. Natl. Acad. Sci., 105, 454–458, 2008. Andreae, M. O. and Merlet, P.: Emission of trace gases and aerosols from biomass burning, Global Biogeochem. Cy., 15, 955–966, 2001. Fuglestvedt, J. S., Berntsen, T. K., Isaksen, I. S. A., Mao, H., Liang, X.-Z., and Wang, W.-C.: Climatic forcing of nitrogen oxides through changes in tropospheric ozone and methane; global 3- D model studies, Atmos. Environ., 33, 961–977, 1999. Arellano, A. F. J., Kasibhatla, P. S., Giglio, L., van der Werf, G. R., and Randerson, J. T.: Top-down estimates of global CO sources using mopitt measurements, Geophys. Res. Lett., 31, L01104, doi:10.1029/2003GL018609, 2004. Fung, I., John, J., Lerner, J., Matthews, E., Prather, M., Steele, L. P., and Fraser, P. J.: Three-dimensional model synthesis of the global methane cycle, J. Geophys. Res., 96, 13 033–13 065, 1991. Bauer, S. E., Mishchenko, M. I., Lacis, A., Zhang, S., Perl- witz, J., and Metzger, S. M.: Do sulfate and nitrate coat- ings on mineral dust have important effects on radiative prop- erties and climate modeling?, J. Geophys. Res., 112, D06307, doi:10.1029/2005JD006977, 2006. Horowitz, L. W., Walters, S., Mauzerall, D. L., Emmons, L. K., Rasch, P. J., Granier, C., Tie, X. X., Lamarque, J.-F., Schultz, M. G., Tyndall, G. S., Orlando, J. J., and Brasseur, G. P.: A global simulation of tropospheric ozone and related tracers: Description and evaluation of mozart, version 2, J. Geophys. Res., 108(D24), 4784, doi:10.1029/2002JD002853, 2003. Bauer, S. E., Koch, D., Unger, N., Metzger, S. M., Shindell, D. T., and Streets, D.: Nitrate aerosols today and in 2030: Importance relative to other aerosol species and tropospheric ozone, Atmos. Chem. Phys., 7, 5043–5059, 2007, http://www.atmos-chem-phys.net/7/5043/2007/. Koch, D., Bond, T., Streets, D., Bell, N., and van der Werf, G. R.: Global impacts of aerosols from particular source regions and sectors, J. Geophys. Res., 112, D02205, doi:10.1029/2005JD007024, 2007. Bell, N., Koch, D., and Shindell, D.: Impacts of chemistry- aerosol coupling on tropospheric ozone and sulfate simulations in a general circulation model, J. Geophys. Res., 110, D14305, doi:10.1029/2004JD005538, 2005. Kvalev˚ag, M. M. and Myhre, G.: Human impact on direct and diffuse solar radiation during the industrial era, J. Climate, 20, 4874–4883, 2007. Bergamaschi, P., Hein, R., Heimann, M., and Crutzen, P. J.: Inverse modeling of the global CO cycle, 1. Inversion of co mixing ratios, J. Geophys. 6 Discussion and conclusion Hence a strat- egy with a dual focus on both AQ and climate change mitiga- tion may present a unique opportunity to engage parties and nations not yet fully commited to climate change mitigation for its own sake. Although this is only an initial study, this work suggests that reducing emissions from the developing Asia domestic fuel burning and North America surface trans- portation sectors may be among the best options for pursuing such a strategy, providing substantial beneﬁts for air quality while simultaneously contributing to climate change mitiga- tion. The response of AQ, especially particulate, appears to be more robust than the climate forcing across models. This is perhaps not surprising given that RF depends not only on concentrations, but also on such things as water uptake of aerosols, the vertical distribution of aerosols and ozone, and the position of ozone and aerosols relative to clouds. Sur- face ozone amounts are determined by complex, non-linear Acknowledgements. We thank the NASA Atmospheric Chemistry Modeling and Analysis Program (ACMAP). JFL was supported by the SciDAC project from the Department of Energy. This research used resources of the National Energy Research Scientiﬁc Computing Center, which is supported by the Ofﬁce of Science of the US Department of Energy under Contract No. DE-AC03- Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ 7112 D. Shindell et al.: Climate and air quality by emission region and sector 76SF00098. The National Center for Atmospheric Research is operated by the University Corporation for Atmospheric Research under sponsorship of the National Science Foundation. ozone and radiative forcing: CTM calculations for the period 1990–2030, Atmos. Chem. Phys., 5, 1731–1755, 2005, h // h h /5/1731/2005/ ozone and radiative forcing: CTM calculations for the period 1990–2030, Atmos. Chem. Phys., 5, 1731–1755, 2005, 76SF00098. The National Center for Atmospheric Research is operated by the University Corporation for Atmospheric Research under sponsorship of the National Science Foundation. http://www.atmos-chem-phys.net/5/1731/2005/. Derwent, R. G., Collins, W. J., Johnson, C. E., and Stevenson, D. S.: Transient behaviour of tropospheric ozone precursors in a global 3-D CTM and their indirect greenhouse effects, Climatic Change, 49, 463–487, 2001. Edited by: R. Vautard D. Shindell et al.: Climate and air quality by emission region and sector Guo, H., Kirkevag, A., Kristjansson, J. E., and Seland, Ø.: Model intercomparison of indirect aerosol effects, Atmos. Chem. Phys., 6, 3391–3405, 2006, http://www.atmos-chem-phys.net/6/3391/2006/. Guo, H., Kirkevag, A., Kristjansson, J. E., and Seland, Ø.: Model intercomparison of indirect aerosol effects, Atmos. Chem. Phys., 6, 3391–3405, 2006, h // h h /6/3391/2006/ Shindell, D. T., Faluvegi, G., Unger, N., Aguilar, E., Schmidt, G. A., Koch, D., Bauer, S. E., and Miller, R. L.: Simula- tions of preindustrial, present-day, and 2100 conditions in the NASA GISS composition and climate model G-PUCCINI, At- mos. Chem. Phys., 6, 4427–4459, 2006, http://www.atmos-chem-phys.net/6/4427/2006/. http://www.atmos-chem-phys.net/6/3391/2006/. P´etron, G., Granier, C., Khattatov, B., Yudin, V., Lamar- que, J.-F., Emmons, L., Gille, J., and Edwards, D. P.: Monthly CO surface sources inventory based on the 2000– 2001 MOPITT satellite data, Geophys. Res. Lett., 31, L21107, doi:10.1029/2004GL020560, 2004. Shindell, D. T., Faluvegi, G., Bauer, S. E., Koch, D. M., Unger, N., Menon, S., A., Miller, R. L., Schmidt, G. A., and Streets, D. G.: Climate response to projected changes in short-lived species un- der an A1B scenario from 2000–2050 in the GISS climate model, J. Geophys. Res., 112, D20103, doi:10.1029/2007JD008753, 2007. Prather, M. J., Ehhalt, D., and Dentener, F., et al.: Atmospheric chemistry and greenhouse gases, in: Climate change 2001, edited by: Houghton, J. T., Cambridge Univ. Press, Cambridge, 239– 287, 2001. Shindell, D. T., Levy II, H., Schwarzkopf, M. D., Horowitz, L. W., Lamarque, J.-F., and Faluvegi, G.: Multi-model projections of climate change from short-lived emissions due to human activities, J. Geophys. Res., 113, D11109, doi:10.1029/2007JD009152, 2008. Ramaswamy, V., Boucher, O., Haigh, J., et al.: Radiative forcing of climate change, in: Climate change 2001, edited by: Houghton, J. T., Cambridge Univ. Press, Cambridge, 349–416, 2001. Rypdal, K., Berntse, T., Fuglestvedt, J. S., Aunan, K., Torvanger, A., Stordal, F., Pacyna, J. M., and Nygaard, L. P.: Tropospheric ozone and aerosols in climate agreements: Scientiﬁc and politi- cal challenges, Environ. Sci. Policy, 8, 29–43, 2005. Stohl, A.: Characteristics of atmospheric transport into the Arctic troposphere, J. Geophys. Res., 111, D11306, doi:10.1029/2005JD006888, 2006. Schmidt, G. A., Ruedy, R., Hansen, J. E., Aleinov, I., Bell, N., Bauer, M., Bauer, S., Cairns, B., Canuto, V., Cheng, Y., Del Ge- nio, A., Faluvegi, G., Friend, A. D., Hall, T. M., Hu, Y., Kelley, M., Kiang, N. Y., Koch, D., Lacis, A. A., Lerner, J., Lo, K. K., Miller, R. References Geophys. Res., 104, 22 137–22 162, 1999. Olivier, J. G. J. and Berdowski, J. J. M.: Global emissions sources and sinks, in: The climate system, edited by: Berdowski, J., Guicherit, R., and Heij, B. J., A. A. Balkema Publishers/Swets & Zeitlinger Publishers, Lisse, The Netherlands, 33–78, 2001. p y Dentener, F. D., Stevenson, D. S., Cofala, J., Mechler, R., Amann, M., Bergamaschi, P., Raes, F., and Derwent, R. G.: The impact of air pollutant and methane emissions controls on tropospheric Penner, J. E., Quaas, J., Storelvmo, T., Takemura, T., Boucher, O., Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/ 7113 D. Shindell et al.: Climate and air quality by emission region and sector D. Shindell et al.: Climate and air quality by emission region and sector L., Nazarenko, L., Oinas, V., Perlwitz, J., Perlwitz, J., Rind, D., Romanou, A., Russell, G. L., Sato, M., Shindell, D. T., Stone, P. H., Sun, S., Tausnev, N., Thresher, D., and Yao, M.-S.: Present day atmospheric simulations using GISS modelE: Com- parison to in-situ, satellite and reanalysis data, J. Climate, 19, 153–192, 2006. Unger, N., Shindell, D. T., Koch, D. M., and Streets, D. G.: Air pol- lution radiative forcing from speciﬁc emissions sectors at 2030, J. Geophys. Res., 113, D02306, doi:10.1029/2007JD008683, 2008. Van der Werf, G. R., Randerson, J. T., Collatz, G. J., and Giglio, L.: Carbon emissions from ﬁres in tropical and subtropical ecosys- tems, Glob. Change Biol., 9, 547–562, 2003. D., Romanou, A., Russell, G. L., Sato, M., Shindell, D. West, J. J., Fiore, A. M., Horowitz, L. W., and Mauzerall, D. L.: Global health beneﬁts of mitigating ozone pollution with methane emission controls, P. Natl. Acad. Sci., 103, 3988–3993, 2006. Schulz, M., Textor, C., Kinne, S., Balkanski, Y., Bauer, S. E., Berntsen, T., Berglen, T., Boucher, O., Dentener, F., Grini, A., Guibert, S., Iversen, T., Koch, D., Kirkev˚ag, A., Liu, X., Mon- tanaro, V., Myhre, G., Penner, J., Pitari, G., Reddy, S., Seland, Ø., Stier, P., and Takemura, T.: Radiative forcing by aerosols as derived from the AeroCom present-day and pre-industrial simu- lations, Atmos. Chem. Phys., 6, 5225–5246, 2006, http://www.atmos-chem-phys.net/6/5225/2006/. West, J. J., Szopa, S., and Hauglustaine, D. A.: Human mortal- ity effects of future concentrations of tropospheric ozone, C. R. Geosci., 339, 775–783, doi:10.1016/j.crte.2007.08.005, 2007. Wild, O., Prather, M., and Akimoto, H.: Indirect long-term global radiative cooling from NOx emissions, Geophys. Res. Lett., 28, 1719–1722, 2001. Atmos. Chem. Phys., 8, 7101–7113, 2008 www.atmos-chem-phys.net/8/7101/2008/
https://openalex.org/W3040792891	https://hispaniasacra.revistas.csic.es/index.php/hispaniasacra/article/download/828/828	es		Los miembros femeninos de la Tercera Orden Franciscana en Andalucía a finales de la Edad Media	Hispania sacra	2,020	cc-by	13,897	Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA A FINALES DE LA EDAD MEDIA POR Silvia María Pérez González1 Universidad Pablo de Olavide de Sevilla Y José Sánchez Herrero2 Universidad de Sevilla Resumen El presente trabajo tiene como objeto de estudio a los miembros femeninos de la Tercera Orden franciscana que vivieron en Andalucía a finales de la Edad Media. En la primera parte tratamos de precisar el título de «Tercera Orden» y su origen. En la segunda exponemos los beaterios existentes, muchos de los cuales fueron obligados a convertirse o a fusionarse con un convento de monjas. Finalmente, abordamos el estudio de terceras en particular profundizando en los distintos aspectos que conformaban su vivir diario. Palabras clave: Tercera Orden Franciscana; mujeres religiosas; terceras; beaterio; actividades económicas FEMALE MEMBERS OF THE THIRD FRANCISCAN ORDER IN ANDALUSIA AT THE END OF THE MIDDLE AGES Abstract This paper is focused on the study of the female members of the Third Franciscan Order who lived in Andalusia in the late Middle Ages. Firstly we try to specify the title of «Third Order» and its origin. In the second part we expose the existing communities, many of whom were forced to convert or to merge with a convent of nuns. Finally, we approach the study of tertiaries in particular, by deepening the different aspects of their daily living. Key words: Third Franciscan Order, religious women; tertiaries; beguinaire; economic activities Cómo citar este artículo / Citation: Pérez González, Silvia María y José Sánchez Herrero. 2020. «Los miembros femeninos de la Tercera Orden Franciscana en Andalucía a finales de la Edad Media». Hispania Sacra LXXII, 145: 25-38. https://doi.org/10.3989/ hs.2020.002 Recibido/Received 19-05-2017 Aceptado/Accepted 25-10-2018 El título de «Tercera Orden» 12 La expresión «Tercera Orden» se ha entendido frecuentemente como «tercera comunidad religiosa», «tercer or1 spergon@upo.es/ ORCID iD: https://orcid.org/0000-0002-3535-1826 2 josanhe1935@gmail.com / ORCID iD: https://orcid.org/0000-0003-4926-6777 den religioso», en sentido estricto, a partir de cuanto afirmaban las leyendas medievales que atribuían a san Francisco la fundación de tres órdenes: la primera los Frailes Menores, la segunda las Clarisas y la tercera la Tercera Orden, lo que ha sido objeto de larga discusión. Según las investigaciones hasta el presente realizadas parece que la expresión Tercera Orden (TO), que aparece por primera vez en un testamento de 1292, se refiere a Copyright: © 2020 CSIC. Este es un artículo de acceso abierto distribuido bajo los términos de la licencia de uso y distribución Creative Commons Reconocimiento 4.0 Internacional (CC BY 4.0). 26 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… la Orden Franciscana,3 mientras que en las otras Órdenes Mendicantes este término no aparece sino más tardíamente, en los Dominicos hacia la mitad del siglo XV, en los Carmelitas después de 1450 y regularmente en 1660. Pero en el testamento de 1292, la expresión «tertii ordinis» no se refiere a la TO de san Francisco, como hoy la entendemos, sino a una orden de los penitentes de san Francisco, de la que el testador era miembro, es decir, a uno de aquellos «poenitentium collegia» de los que habla Gregorio IX, en este caso de san Francisco. Es decir, la expresión TO no se refería a los laicos en general que seguían a san Francisco, sino a un grupo especial de penitentes franciscanos. Si la expresión TO es tardía, mucho más parece serlo el adjetivo terciario, terciaria, signo claro de que la realidad de la TO entendida en el sentido «de orden que viene después de otras dos» no estaba aún madura, mucho más cuando ni siquiera se usaba el termino Segunda Orden, sino que los documentos hablan de «sororis ordinis nostri». En conclusión, la expresión TO recibió en sus comienzos diferentes significados: simples laicos, seglares, que tiene un puesto dentro de la estructura de la Iglesia, que vienen después del clero; pero también laicos seculares que deseaban vivir de manera más estricta su vocación cristiana y diferenciándose en esto del común de los fieles, grupos que se unieron a cualquier orden mendicante colocándose después del Primer Orden (de los hermanos) y del Segundo Orden (de las monjas). Más tardíamente indicará una institución más cercana a una auténtica Orden religiosa en sentido estricto y entonces se dividirá en TO regular («los Terceros») y la TO secular (la que hoy conocemos como VTO. Ahora bien, si la terminología de TO, terciario, terciaria es tardía, la realidad que nos interesa está claramente presente desde los orígenes de varias Órdenes mendicantes, pero con otra terminología. Las fuentes antiguas hablan, en efecto, de «hermanos de penitencia», «hermanas de penitencia», «poenitentes», «continentes» (refiriéndose a los casados que habían aceptado la vida penitencial), encarcelados, reclusos, palabras con las que se quiere destacar el sentido penitencial. «Órdenes de penitencia» Los «penitentes», en sus orígenes, pudieron comprender también a los clérigos, fueron un movimiento independiente de las Órdenes religiosas mendicantes, que reunían simplemente a laicos que deseaban vivir una vida ascética o de penitencia. En Italia aparecieron unas «Ordines de Penitentia» ya presentes en Florencia en el hospital de San Galo en 1218 y en el de San Pancracio, 1219.4 Se trataba de fraternidades de penitentes voluntarios que se entregan a prácticas devocionales y penitentes y a la caridad pública. Abrían fuera de las puertas de la ciudad sus hospitales, donde recibían a viajeros y caminantes, y donde escuchaban la predicación de los dominicos o de los franciscanos sin ponerse bajo su dependencia. 3 4 Meersseman 1961, 209, nº 23 y 1977, 698 y ss. Meersseman 1977, 365 y ss. El origen de la TO Raíces históricas Los estudios realizados en los últimos años han confirmado que la obra de san Francisco y su misma conversión se insertan en aquel vasto movimiento penitencial (penitentes) que lo precedió desde varios siglos atrás y que había movido no solo a los simples e individuales fieles laicos, sino también a los clérigos y a familias enteras con la adopción de un estilo de vida comunitario. En otras palabras la TO no es otra cosa que una ulterior expresión de la espiritualidad de los laicos quienes, desde san Benito, habían preferido unirse a los monasterios en formas muy diferentes: «familiar», «oblato», «oblación» y que, a partir del siglo XIII y con la aparición de las Órdenes mendicantes, encuentran otro modo de expresarse, sin renunciar a su trabajo y al estado conyugal. Son laicos que intentan superar su vida espiritual cristiana ordinaria y que vendrán a ocupar un puesto especial en la estructura de la Iglesia, entre los clérigos y los simples fieles que no adoptan este estilo de vida penitencial. La obra de san Francisco se inscribe en este contexto, llama también a los laicos, casados y no casados, a constituir una orden de penitencia y no una T.O. Aun aceptando que Francisco diera una propia configuración franciscana a sus penitentes, parece difícil exponer cuáles fueron estas características, ya porque el estatuto penitencial, como hemos afirmado, era anterior a las Órdenes mendicantes, pues el Memoriale propositi de 12215 no contiene nada específicamente franciscano. No cita particulares actuaciones pauperísticas, ya porque los Hermanos Menores no se preocuparon de los hermanos penitentes hasta que san Buenaventura (1221-1274) se interesó por ellos, más por justificar la despreocupación de la Orden que por darles estas características, cosa que no ocurrió sino más tarde, cuando de hecho los Hermanos Menores se preocuparon de los laicos penitentes. No será sino la bula Supra montem del papa Nicolás IV de 1289 quien afirmará la calidad de san Francisco como «huius ordinis institutor» refiriéndose a la TO franciscana.6 A imitación de san Francisco todas las órdenes mendicantes tuvieron su fraternidad de penitentes. Una diferencia comienza a manifestarse en la Regla que el maestro general de los dominicos, fray Munio de Zamora, preparó en 1285,7 cuando quiso que los penitentes, asistidos espiritualmente por sacerdotes dominicos, estuviesen sujetos a la dirección de la Orden. Tipología de la TO regular y TO secular En línea general las TO asumen una fisonomía casi idéntica, pues todas practicaron: la vida común, la vida aislada en la propia casa, la vida eremítica (especialmente como custodios de santuarios), la vida en reclusión, la vida célibe, etc. Con el tiempo y dentro de los Franciscanos se produjo una distinción entre TO regular y TO secular. Encontramos 5 En el nombre del Padre, el Hijo y del Espíritu Santo. MEMORIAL DEL PROPÓSITO de los hermanos y hermanas de la Penitencia que viven en sus propias casas. Inicios del año del Señor 1221. Aprobado este propósito de vida el 20 de mayo de 1228 por el papa Gregorio IX. 6 Sobre esta cuestión véase más ampliamente tratado en Meersseman 1977, 361. 7 Meersseman 1977, 377-380. Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… esta distinción en la constitución «Dum intra» del Concilio Lateranense V, del 19 de diciembre de 1516, aunque sin la calificación aún de «regular» y «secular». Se entiende que la TO regular es posterior a la TO secular, pero con anterioridad a la TO regular existieron en Europa grupos similares: beguinas, begardos, beatas, comunidades religiosas de hombres y de mujeres entregados al cuidado de los hospitales que seguía normalmente una de las Reglas aprobadas, normalmente la de san Agustín, o simplemente costumbres locales. Dentro de la TO regular han existido y, en parte subsisten hoy, diversas tipologías: terciarios religiosos de votos simples, terciarios que admitían los tres votos religiosos o solo alguno de ellos (el de castidad), asociaciones de terciarios sin votos públicos pero con vida común, terciarios eremitas y terciarios viviendo en conventos juntamente con religiosos de votos solemnes. Pero la característica más importante fue y es que la TO regular se configura de manera análoga a un monasterio o a una Orden o un instituto centralizado, con su propio ministro o superior general. En cambio la TO secular o tercer estado de los casados, estaba y está lógicamente compuesto de gente casada, que vive en un mismo lugar, ciudad o pueblo, relacionada o unida a una de las diferentes Órdenes mendicantes. En sus orígenes eran mixtas, pero la mujer casada tenía que tener el consentimiento de su marido para ingresar en una TO y en la fraternidad no gozaba de autoridad, que permanecía en manos de los hombres, según el uso del tiempo. Se crearon también fraternidades únicamente femeninas, que después de la crisis de los siglos XIV-XV, cuando los terciarios (hombres) perdieron los privilegios de que gozaban en el siglo XIII, quedaron solo las fraternidades femeninas. Este régimen permaneció hasta el siglo XX, cuando las fraternidades volvieron a ser mixtas. Algunas cuestiones jurídicas De lo dicho se deduce que la Orden Tercera secular no era ni es una «Orden religiosa» en sentido estricto, sino solo un «estado», integrado por personas: viudas, casadas, célibes. Los componentes de esta TO secular no eran ni son personas eclesiásticas, «viri religiosi», por lo que a partir del siglo XIV las autoridades comunales dejaron de concederles los privilegios que tenían los «religiosos». Cuando los penitentes se acercaron a las Órdenes mendicantes aceptaron también el hábito, que los terciarios usaron durante bastante tiempo (un manto, «signum habitus») gris los franciscanos y negro los dominicos. Posteriormente se establece una distinción: los terciarios y las terciarias que habían aceptado una «regularidad» de vida continuaron llevando un hábito siempre similar o cercano al de los religiosos de la Orden mendicante a la que estaban ligados, mientras que los terciarios seculares, en cambio, puesto que el hábito entero resultaba incómodo para su vida en el mundo, adoptaron un escapulario (llamado pequeño hábito o grande escapulario) para llevarlo bajo sus vestidos, también de noche o escapulario nocturno. La Tercera Orden regular se convirtió en el siglo XVI en primera orden, como los Hermanos Menores, los Conventuales y los Capuchinos y tuvo también su Tercera Orden secular.8 8 Pellicia y Rocca 1992. 27 Grupos de terceras en Andalucía a finales de la edad media A finales de la Edad Media se documentan los siguientes grupos de terceras franciscanas dentro del ámbito andaluz que analizamos individualmente. San Juan de la Penitencia de Cazorla (Jaén):9 Una casa de terciarias adoptó la Regla de clarisas hacia el año 151315. Intervino en la fundación don García de Villarroel, sobrino del Cardenal Cisneros. Gracias a la intervención de don García de Villarroel, Adelantado de Cazorla, Comendador de Carrizosa de la Orden de Santiago, Capitán de la ciudad de Almería y Comisario General y sobre todo, sobrino del Cardenal Arzobispo de Toledo, fray Francisco Ximénez de Cisneros, se permitió el traslado del monasterio al interior de la villa. Esta concesión, además de ser el principal mecenas de la fábrica conventual, le reportó el patronato del convento. Santa Clara de Andújar (Jaén):10 La tradición afirma que habría sido fundado en 1225 por discípulas de Santa Clara. De ser así, significaría que se trató del primer establecimiento religioso de Andalucía. Sin embargo, las noticias documentales sobre vida religiosa femenina en Andújar se hacen esperar a 1450, fecha en que una comunidad de terciarias creada en 1430 recibió licencia papal para abrazar la regla clarisa. A mediados del siglo XV, consideraba insuficientes sus instalaciones y deseaba expandir su fábrica con unas viviendas que la comunidad había adquirido frente al cenobio primitivo. Santa Inés de Andújar (Jaén):11 En 1451 se autoriza la erección como clarisas de Santa Inés antes perteneciente a tereceras franciscanas a petición de Marina López. Concepción de Beas de Segura (Jaén):12 Su fundación tuvo lugar en 1507, siendo uno de los primeros monasterios dedicados a la Concepción. Algunas fuentes lo sitúan como monasterio de terceras que tras una serie de problemas se convirtieron a la regla de Santa Clara e incluso intentaron abandonar la dependencia de la Orden por la del diocesano. Santa Isabel de los Ángeles de Córdoba:13 Fue fundado en la segunda mitad del siglo XV en la collación de San Pedro, donde hoy se encuentra parte del Convento de Santa Cruz, es decir en la calle Valderrama, haciendo esquina con la calle del Sol. A partir de 1489 las religiosas pasaron a la que definitivamente sería su casa en la collación de Santa Marina, en la plazuela de Santa Isabel. Entre sus miembros destacó Magdalena de la Cruz,14 reconocida como santa viva, que fue abadesa entre 1533 y 1542. Acabó siendo condenada por la Inquisición por pacto diabólico. Santa Inés del Valle de Écija (Sevilla):15 Se data entre 1482-1487. En 1482, doña Inés Chirino, viuda del capitán y alcalde de Osuna don Luis de Pernía, fundó este monasterio para 70 monjas extramuros de la ciudad. Después se trasladó a los arrabales del norte de la misma, junto al camino que Almansa Segura 2003. Serrano Estrella 2008, 773. Graña Cid. CLAUSTRA. Atlas de espiritualidad femenina en los Reinos Peninsulares. Barcelona: Institut de Recerca en Cultures (http:// www.ub.edu/claustra/Monestirs/view/523). Consultado el 18 de febrero de 2017. Serrano Estrella 2008, 210. 11 Serrano Estrella 2008, 102. 12 Serrano Estrella 2008, 790. 13 Graña Cid 2002a y 2011, 187-221. 14 Graña Cid 2002b. 15 Miura Andrades 1992, 51-53 y 1994; Rubio 1953. 9 10 Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 28 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… iba al santuario de Nuestra Señora del Valle. La fundación definitiva parece datar de 1487. San Blas de Sevilla:16 En 1485 tomó posesión de la ermita de la ermita de San Blas Leonor Díaz, beata. En 1491 los Reyes Católicos ordenaron que se ejecutasen las sentencias dadas a favor de las religiosas de San Blas sobre ciertas huertas y casas, de donde se deduce que ya eran comunidad constituida. Más tarde el beaterio de San Blas se unió a la comunidad del Convento de Santa Inés. Nuestra Señora de la Concepción de San Juan de la Palma de Sevilla:17 Debe su origen a doña Leonor de Ribera en 1474 como emparedamiento bajo la tercera regla de San Francisco. En 1511 se transformó en convento de concepcionistas. Concepción o Santa Isabel de los Ángeles de Carmona (Sevilla):18 El convento fue fundado por Lucía Sánchez de Baeza en 1510 para monjas clarisas, aunque finalmente esta intención se modificó en 1513 al recibir el hábito concepcionista pero de la tercera regla. Nuestra Señora de la Antigua de Utrera (Sevilla):19 Este Beaterio-convento de religiosas de la tercera regla de Santa Clara se situó en la Ermita de Nuestra Señora de la Antigua en 1506. Lo fundaron las emparedadas sevillanas, procedentes de la iglesia de San Miguel de Sevilla, Mencía de Santa María, Leonor de los Ángeles y Constancia de San Francisco. Las llevó desde Sevilla en 1505 Alonso Álvarez Chamorro con bula de Julio II. Madre de Dios de Jerez de la Frontera (Cádiz):20 El convento de clarisas de Madre de Dios es el resultado de la concentración de las beatas franciscanas existentes en la ciudad de Jerez que, tras el abandono, en 1495, de los observantes del convento de Madre de Dios, para trasladarse al de San Francisco el Real (desalojado por los conventuales) se instalaron en él. La primera noticia documentada que tenemos es la 1504, fecha en la que Catalina Espíndola, mujer de Pedro Martínez de Hinojosa, entrega a Isabel de Melgarejo, su hija, la legítima con el fin de que profese en el convento de Madre de Dios. En igual fecha su hermano Álvaro López hacía donación a María Carrillo, abadesa, Isabel de Cisneros, provisora, Ana González, María de Herrera, Leonor Jiménez, Eufrasia de la Cruz, Francisca Marroquí y Catalina Galindo, monjas profesas en el monasterio de Santa María Madre de Dios de Jerez, de los bienes que había heredado de su padre.21 De la cita de las religiosas hemos de llamar la atención sobre el hecho que dos de ellas, Ana González y María de Herrera, eran religiosas de Santa Inés del Valle de Écija en 1499,22 y continúa siéndolo Ana González en febrero de 1504, cuando las monjas astigitanas otorgaron su carta de poder a fray Cristóbal de Villanueva,23 para volver a serlo María de Herrera en 1509, cuando venden Miura Andrades 1998, 256. Miura Andrades 1998, 254. 18 Miura Andrades 1998, 255. 19 Miura Andrades 1998, 255. 20 Mesa Xinete 1888, 463. Miura Andrades, J. M. 1994. Sancho de Sopranis y Lastra y Terry 1964. 21 1504, enero, 12. Citado por Mesa Xinete, F. 1888, 463. Citado por Miura Andrades 2014, 568-570. 22 1499, mayo, 10. Écija. AGS = Archivo General de Simancas, Casa y Descargo, leg. 1, fol. 285. Citado por Miura Andrades 2014, 568. 23 1504, febrero, 14. Écija. AGS, Casa y Descargo, leg. 4, fol. 72. Citado por Miura Andrades 2014, 568. unas casas de la dote de Francisca de Valderrama.24 Además otras de ellas, Francisca Marroquí y Catalina Galindo tienen en sus apellidos amplias resonancias de la ciudad astigitana. Pensamos que tal comunidad puede considerarse, por los múltiples y frecuentes contactos entre ambas poblaciones fronterizas, filial de Santa Inés del Valle de Écija desde donde, probablemente, llegaran algunas religiosas para estructurar la comunidad. Reafirma esta opinión el hecho de que, en 1512, nos consta que eran «monjas de Santa Clara de la Tercera Regla» a las que el provisor del arzobispado de Sevilla da copia del acuerdo entre Diego de Deza, arzobispo de Sevilla, y el cabildo de Sevilla, por una parte, y fray Cristóbal de Aguilar, Ministro Provincial de los frailes de la Observancia de San Francisco de la Provincia de Andalucía y de la monjas de Santa Clara de la Tercera Regla de la dicha Provincia, en nombre de la abadesa y monjas del monasterio de Santa Inés del Valle de Écija y de todas las otras religiosas, casas y monasterios de monjas de Santa Clara de la Tercera Regla del arzobispado de Sevilla, sobre el pago del diezmo de sus bienes.25 Nuestra Señora de Gracia de Vélez-Málaga (Málaga):26 La fundación como convento tiene que ver con Beatriz de Arellano. En su testamento, otorgado el 18 de diciembre de 1540 mandaba fundar un convento de clarisas con la advocación de Nuestra Señora de la Concepción. Esta fundación nada tenía que ver con el preexistente convento de terciarias de Nuestra Señora de Gracia, a quien Beatriz de Arellano había legado cinco arrobas de aceite al año y con el que, incluso, había llegado a pleitear. Sin embargo, tras su fallecimiento, sus albaceas testamentarios se pusieron de acuerdo con los regidores del municipio en reunificar ambas comunidades, con lo que se ahorrarían costos. Las cláusulas fundacionales estipulaban para el nuevo convento que no sobrepasase el número de 12 monjas de Vélez-Málaga sin dote. Otras 10 fueron las terciarias de Nuestra Señora de Gracia, que para facilitar la refundición adoptaron la orden segunda, pasando a convertirse en monjas clarisas, sujetas a la observancia de monasterio de San Francisco de la ciudad de Vélez-Málaga. Concepción de Vélez-Málaga (Málaga): 1512-1540. Sobre la fundación del convento de Granada tenemos dos crónicas. La de Francisco Bermúdez de Pedraza es concisa: «El séptimo es el Monasterio de la Concepción de nuestra Señora, del hábito y regla tercera de san Francisco, sujeto al Ordinario, fundado por Juliana de San Francisco, de nación romana, el año de mil quinientos y treinta. Son sesenta monjas de hábito pardo y escapulario azul, tienen de renta tres mil ducados».27 La segunda crónica de fray Alonso de Torres28 es mucho más completa. Comienza por afirmar que al Cabildo del templo de El Salvador de Roma, junto al Palacio Lateranense (Ba- 16 17 24 1509, mayo, 16. Écija. APNEC=Archivo de Protocolos Notariales de Écija, Leg. 71. Citado por Miura Andrades 2014, 568. 25 1512, mayo, 21. El acuerdo se conserva en ACS=Archivo de la Catedral de Sevilla, Fondo Histórico General, Leg. 94, doc. n.20/6. Citado por Miura Andrades 2014, 568. 26 Graña Cid. CLAUSTRA. Atlas de espiritualidad femenina en los Reinos Peninsulares. Barcelona: Institut de Recerca en Cultures (http://www.ub.edu/claustra/Monestirs/view/523). Consultado el 18 de febrero de 2017. 27 Bermúdez de Pedraza 1608, 118v. 28 Torres 1683, 863-864. Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… sílica de San Juan de Letrán), los Pontífices le concedieron la facultad de erigir iglesias, monasterios y hospitales, que fuesen «miembros de su cuerpo» y estuviesen a su protección. Según el testimonio del Provisor del Arzobispado de Granada, antes de la fundación canónica de este monasterio en 1618, «había dentro de la casa trece religiosas, que vivían según la Regla de la Tercera Orden de nuestro seráfico padre san Francisco y traían el hábito, según y conforme a la dicha Orden, que sin duda es el que hoy conservan, de color negro y escapulario azul, por la Concepción Purísima, que es la advocación y título», monjas de la Tercera Orden regular de san Francisco o beatas que se habían reunido formando una de los muchos grupos que existieron de órdenes de penitencia. La primera fundación de este monasterio parece se debió a Juliana de san Francisco, de nación romana, siguiendo, sin duda, los privilegios del Cabildo de la Basílica del Salvador de Roma y colocándose bajo la jurisdicción del Ordinario de Granda y no sometidos a alguna de las órdenes existentes. Pero la fundación definitiva tuvo lugar el 23 de noviembre de 1518, a partir de las Letras dadas por el Cabildo de la Basílica de El Salvador de Roma, pero precediendo una bula apostólica del Papa León X del 15 de marzo de 1518, «mandando en una y otra guardasen la regla del Orden Tercero de Penitencia de nuestro seráfico padre san Francisco, con título de la Inmaculada Concepción». Sin embargo el cumplimiento de dichas bulas tuvo que esperar hasta que doña Leonor Ramírez, vecina de la ciudad de Granada, las presentó al arzobispo don Antonio de Roxas, quien mandó su ejecución el año de mil quinientos veinte y tres. Beaterio de Vélez-Málaga (Málaga):29 Francisco Enríquez y Quiñones había dejado en su testamento una casa en Vélez-Málaga para fundar un hospital. Su única hija, Juana Enríquez, fue su heredera y falleció muy joven. Puesto que Vélez-Málaga contaba ya con hospital, sus albaceas acordaron destinar el legado a un monasterio de monjas, puesto que Vélez-Málaga no contaba con ninguno. El cabildo de la ciudad debió aprobar esta decisión en 1503, fecha que se acepta como fundacional. La casa, hoy desaparecida, se ubicaba en la actual plaza de Rojas. Esta casa la tenían ocupada los frailes franciscanos hasta que en 1508 se trasladaron a su convento, por lo que las mujeres, al hallarse bajo la protección del cabildo civil, recibieron de este alojamiento en una casa del barrio alto, delimitada por las calles Real de la Villa, Subida a la Fortaleza y San Antonio. Esta edificación, hoy no conservada, fue habilitada para la vida en clausura. La vida espiritual de estas mujeres, en realidad beatas, giraba en torno a la imagen de la Virgen de Gracia, que los Reyes Católicos habían traído consigo al tomar la ciudad, donde la dejaron. Sin embargo la fundación tenía que ser aprobada por la autoridad eclesiástica, así como el cambio de uso del legado de Juana Enríquez, lo que aconteció en el año 1512, cuando el papa emitió una bula dando a estas mujeres la orden tercera de San Francisco, pasando pues, a detentar el carácter de terciarias franciscanas. Nuestra Señora de los Ángeles de Granada:30 Don Rodrigo Ponce de Ocampo, caballero de Santiago, y su esposa 29 Graña Cid. CLAUSTRA. Atlas de espiritualidad femenina en los Reinos Peninsulares. Barcelona: Institut de Recerca en Cultures (http:// www.ub.edu/claustra/Monestirs/view/523). Consultado el 18 de febrero de 2017. 30 Bermúdez de Pedraza 1608. Torres 1683. 29 doña Leonor de Cáceres determinaron fundar en 1538 un convento de terciarias regulares franciscanas. Al morir el fundador sin terminarlo, prosiguieron la obra dos terciarias seglares de Granada, doña Leonor de Saavedra y doña Inés de Jesús. Ambas lo finalizaron en 1540 y profesaron en él, bajo la jurisdicción de los terciarios de Granada durante 30 años. En 1567, con bula de Pío V, profesaron en la Orden de Santa Clara y se sujetaron a la Observancia. Concepción de Granada: Pocas son las noticias que tenemos de este monasterio. Para situarnos debemos de comenzar recordando que fueron los Reyes Católicos quienes planificaron la fundación de la Orden femenina franciscana y que su preocupación recayó, como era lógico, sobre la Orden de Santa Clara. Doña Isabel, la católica, fundó en Granada el monasterio de Santa Isabel la Real al que concedió copiosas rentas.31 Monjas de Santa Isabel la Real de Granada poblaron el de la Concepción de Málaga.32 María del Mar Graña Cid nos informa de «agrupaciones de beatas o mulieres religiosas que vivían en comunidad sin sujetarse a ninguna de las reglas oficialmente reconocidas, quizá como respuesta al previo influjo franciscano, que habría tenido tiempo para extenderse y consolidarse, y quizá por los márgenes más amplios que los largos periodos de sede vacante que caracterizan estos años propiciaban».33 Debemos, también, tener en cuenta que entre 1454 y 1492 doña Beatriz de Silva fundaba en Toledo una congregación de monjas de estricta clausura en honor de la Inmaculada Concepción o Concepcionistas, de hábito blanco, que primero recibieron la Regla de Cister, posteriormente los dominicos trataron de agregárselas y, finalmente, frey Juan de Tolosa, franciscano, consiguió que la Orden no desapareciera y adoptara la regla de las clarisas y quedara bajo la dirección de los frailes menores. Con estos preámbulos podemos concluir que la fundación de la Concepción de Vélez Málaga se debió a un grupo de mujeres beatas o autotituladas como terciarias franciscanas que se constituyeron como orden religiosa bajo la obediencia episcopal. Las que en 1512 «obtenían facultad de Julio II para constituirse en convento de terciarias regulares bajo la advocación de Santa María de Gracia y la sujeción episcopal; en 1518, a instancias de doña Leonor Ramírez, concedía León X que su comunidad de trece mujeres beatas de Granada guardase la tercera orden con la advocación de la Inmaculada Concepción, aunque oficialmente no se reconoce esa fundación hasta 1523».34 Nada tienen que ver estos conventos con el título de Concepción con las Concepcionistas fundadas por doña Beatriz de Silva. Las terceras y sus vivir cotidiano Cuestiones terminológicas De la documentación consultada podemos extraer una primera conclusión: en la sociedad andaluza bajomedieval, al menos en un grupo con una formación intelectual especial como eran los escribanos públicos, no existía una idea clara de 31 32 33 34 Torres 1683, 395. Graña Cid 1997, 115. Ibídem, 116. Ibídem, 117. Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 30 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… lo que era una tercera. Es más, jamás utilizan este concepto para referirse a ellas, sino que lo aplicamos nosotros de acuerdo con las bendiciones o consagraciones por ellas recibidas, puesto que la variedad terminológica empleada es enormemente rica. Ello no hace sino subrayar la indefinición existente en relación con esta forma de religiosidad femenina laica. El término utilizado en un mayor número de ocasiones es el de freila: freila de la Tercera Regla de san Francisco.35 La denominación más empleada en segundo lugar es la de honesta: honesta de la Tercera Regla de San Francisco.36 En tercer lugar figura el término beata: beata de la Orden de san Francisco.37 En un menor número de casos se hace referencia a las terceras como religiosas: religiosa de la Tercera Regla de san Francisco;38 o aplicándoles la denominación de hermana: hermana de la Orden de san Francisco.39 Los escribanos públicos y, suponemos, el resto de la sociedad andaluza bajomedieval, eran conscientes de que estas mujeres habían optado por una forma peculiar de religiosidad, distinguían perfectamente la Orden a la que pertenecían, sin duda ayudados por los hábitos que vestían, pero a la hora de otorgarles una denominación precisa y definitoria de su situación carecían del término exacto que debían aplicar. El mismo escribano emplea indistintamente varios términos, lo que no es más que el reflejo de un fenómeno que la sociedad contemporánea siente como propio y cotidiano, pero a la vez oscuro e indeterminado. Quizás esta indefinición pueda hacerse extensiva a sus protagonistas: ellas mismas se autodenominaban bajo distintas formas por lo que el escribano carecía de una denominación precisa y definitoria de su situación y de un término exacto que debían aplicar. Esta caracterización que se acaba de presentar no está exenta de problemas porque, por ejemplo, no siempre es fácil distinguir cuándo estamos ante un rasgo lingüístico específico y cuándo ante un rasgo general. Como han señalado diversos autores en este campo los grados de especificidad posibles son diversos porque, junto a las formas que solo pertenecen a la competencia de los especialistas en un ámbito específico, habría que incluir aquellas que pertenecen a la competencia de los especialistas y a la competencia pasiva de los demás hablantes, aquellas que son utilizadas en la lengua general y en la de especialidad, pero con acepciones funcionalmente bien diferenciadas, y aquellas que forman parte de la competencia de los especialistas y de los no especialistas, aunque su uso es mucho más frecuente en la esfera de los primeros.40 Ingreso en la orden La documentación estudiada nos permite conocer detalladamente el ritual de ordenación de dos terceras franciscanas, Constanza de la Cruz, beata romana estante en Jerez, 35 AHPS=Archivo Histórico Provincial Sevilla. Sección Protocolos. Leg.17422. Año 1496. Fol.536r., Leg.17424. Año 1498. Fols.97r., 215v., Leg.1500. Año 1502. Fol.750r., Leg.17426. Año 1500. Fol.167v., Leg.17427. Año 1501. Sin foliar. 36 AHPS. Sección Protocolos. Leg.2158. Año 1500. Fol.107v., Leg.3212. Año 1485. Fol.2r., Leg. 2160. Año 1501. Fol. 196r, Leg.2161. Año 1502. Fol.43r. 37 Leg. 9102. Año 1504. Fol. 170v 38 AHPS. Sección Protocolos. Leg.2163. Año 1503. Fol.5r. 39 AHPS. Sección Protocolos. Leg.9099. Año 1495. Fol.130v. 40 Cedergre 1983, 150. Sánchez García 2009, 157. Aldonza Rodríguez y Leonor de Ortega. En el caso de Constanza41 la ceremonia, que se desarrolló a hora de vísperas en la capilla mayor de la iglesia del Convento de San Francisco de Jerez de la Frontera, fue oficiada por el padre guardián de la Casa, fray Jorge, en presencia de los frailes pertenecientes a la misma. Constanza declaró, tras ser interrogada por el guardián, que quería ser freila beata de la Tercera Orden y Regla de san Francisco, lo que viene a ratificar lo anteriormente expuesto sobre la indefinición de este colectivo religioso femenino. A continuación le informó sobre lo que estaba obligada a cumplir y obedecer, tras lo cual se arrodilló delante del guardián para hacer profesión, este tomó sus manos entre las suyas y Constanza pronunció en voz alta su voto y promesa por la Virgen, san Francisco y todos los santos de guardar en todo tiempo los mandamientos de Dios y de la Tercera Orden de san Francisco así como de no partir ni salir de la Orden. El testimonio del ingreso en la Orden Tercera de san Francisco de Aldonza Rodríguez, que tuvo lugar el 25 de octubre de 1496, es más completo. La ceremonia, que tuvo lugar en el Convento de San Francisco de Sevilla, fue oficiada por el custodio de la Casa, el doctor fray Antonio de los Ríos, vicario y ministro por el reverendo maestro fray Sancho de Hontañón, ministro de los frailes menores, monjas de Santa Clara, frailes y freilas de la Tercera Regla de san Francisco de la provincia de Sevilla. En primer lugar le preguntó si quería ser freila de la Tercera Regla de san Francisco, a lo que la citada Aldonza respondió afirmativamente. A continuación fray Antonio la conminó a hacer profesión en sus manos y ella, en la mejor forma y manera que podía, así lo hizo. Luego a instancia de fray Antonio hubo de declarar públicamente que quería ser freila de la Tercera Regla de la Orden de san Francisco y llevar su hábito. Tras pronunciar su voluntad fray Antonio cogió sus manos entre las suyas y ella tomó una candelilla y repitió tres palabras que el fraile le dijo, seguidas de la promesa a Dios, santa María, san Pedro, san Pablo y el santo padre san Francisco, de que guardaría la Tercera Regla de la Orden de san Francisco, así como observar obediencia y castidad. Finalmente, fray Antonio la bendijo. Como puede observarse, el parecido entre las fórmulas empleadas en el ingreso en una Orden Tercera y las características del contrato de vasallaje son evidentes: volo, inmixtio manuum, juramento de fidelidad.42 Desconocemos si en el caso de las terceras se prescindía del ósculo. Los detalles de la ceremonia se muestran menos desarrollados en el testimonio prestado por el reverendo padre fray Sancho de Hontañón, ministro de los frailes menores, monjas de Santa Clara, frailes y freilas de la Tercera Regla de san Francisco de la provincia de Sevilla, en relación con el ingreso en esta Orden de Leonor de Ortega.43 Simplemente se limitó a testimoniar que Leonor hizo profesión en sus manos, segund lo acostumbran a faser las otras freylas, en el Convento de San Francisco y ante varios testigos. De estos ejemplos podemos sacar algunas conclusiones: los ministros encargados de oficiar la ceremonia eran personas relevantes en la Orden, el ministro general de la provincia de Sevilla o el custodio del Convento de San 41 AMJFPN=Archivo Municipal Jerez Frontera Protocolos Notariales. Rodrigo de Rus. 7 de octubre de 1537. Fol.995v. 42 Alvarado Planas y Pérez Marcos 2010. Bisson 1989, 170. 43 AHPS. Sección Protocolos. Leg.3215. Año 1495. Fol.17v. Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… Francisco. Y son precisamente estos institutos, la Casa Grande hispalense y el cenobio jerezano, el escenario donde había de desarrollarse. Tras el ingreso las terceras, no sabemos si todas o solo algunas de ellas, llevaban un hábito distintivo de la Orden en la que había hecho la profesión. La documentación consultada es enormemente parca en lo que a esta cuestión se refiere. Solo contamos con una única referencia de una tercera franciscana que vestía el hábito correspondiente: Ana Rodríguez la farfana.44 Precisamente el hecho de vestir honesto áuito fue uno de los elementos considerados por la Cofradía de San Bernardo de la collación de San Juan, heredera de los bienes del racionero Francisco Fernández, tío de la citada Ana, para entregarle unas casas que el clérigo le había dejado en su testamento con la condición de que se casase o ingresase en alguna orden. Ana Rodríguez optó por una forma de vida religiosa fuera de un recinto monástico, superando el condicionante impuesto por su tío para heredar las casas al vivir de acuerdo con la Tercera Regla franciscana tanto en las actitudes internas como en las manifestaciones externas que la distinguían. La vinculación con el convento franciscano masculino de su ciudad se va a mantener durante toda la existencia de estas mujeres. Juana Sánchez, donada o freila de la Orden Tercera en 1417 y madre de fray Juan de Todos los Santos, Custodio de la Custodia de Sevilla,45 y otras terciarias franciscanas se encargaban de atender la Iglesia de San Francisco.46 Leonor Rodríguez, la beata,47 se autodenominó «freila» del monasterio de San Francisco.48 Otra prueba de esta vinculación son las últimas voluntades de las terceras, que no son muy abundantes. Desconocemos las razones que justifican el bajo número de testamentos49 a ellas pertenecientes: carecían de bienes que dejar en herencia, no necesitaban dejar establecidos los oficios de difuntos que debían celebrarles, pues de ello se ocupaban sus hermanos y hermanas en religión, celebrando, con una diferenciación litúrgica por sexos, los ofiçios que los frailes acostumbran faser a otras personas freilas de su regla,50 etc. El 28 de febrero de 1477 dictó su testamento Inés Gómez, freila de la Tercera Regla de san Francisco.51 Su vinculación con la Casa franciscana de Sevilla era notoria: se mandó enterrar en la iglesia del monasterio52 sin ningún tipo de boato ni ornamentación en las ceremonias de inhumación ni en la sepultura, que tenía en propiedad. Los frailes debían celebrarle los oficios propios de su condición como acostumbraban a realizar con otros miembros de la Orden, y nombró albacea de su testamento a fray Francisco, fraile profeso del convento. La única propiedad que dejó en herencia fue su sepultura, siendo 31 su receptora su sobrina Inés González. Por tanto, el espíritu de pobreza y austeridad que imponía la Orden franciscana preside el contenido testamentario. Los ejemplos reseñados evidencian que las terceras se sentían vinculadas a la casa masculina de la Orden franciscana y en ningún momento con un convento femenino. Precisamente no quieren vincularse con el instituto donde se desarrollaba la vida de clausura de las monjas, a las que ellas de forma expresa habían renunciado y de la que intencionadamente querían desvincularse reivindicando su pertenencia a la Orden a partir de su vinculación espiritual e institucional las casas de los frailes. Por su parte los franciscanos actuaron como impulsores de esta forma de religiosidad femenina laica, así como de instrumentos de institucionalización de la misma.53 Vida en sociedad Las terceras se encontraban perfectamente incardinadas en la sociedad a la que pertenecían. Estas mujeres, pese a la vinculación con el convento de la Orden en la que profesaban, vivían solas en sus casas como un vecino más, lo que refuerza su carácter laical.54 Por otro lado, esa estrecha relación con la sociedad a la que pertenecían las llevó a cumplir para ella funciones derivadas también de su peculiar forma de vivir la religión, de lo que hemos denominado el prestigio socio-religioso de las terceras. De esta manera el hecho de haber optado por una forma personal de religiosidad presente en su vivir diario, las revestía de una consideración especial y las hacía portadoras de unos valores de especial reconocimiento social. Por ello fueron designadas como albaceas testamentarios y se les encomendó el cumplimiento de mandas pro remedio animae.55 Las actividades económicas Algunos miembros femeninos de la Orden tercera franciscana participaron en actividades económicamente muy lucrativas como el mercado inmobiliario, el mundo de las finanzas y el comercial, al tiempo que se insertaron en el mercado laboral. No obstante queremos subrayar que la situación de prosperidad económica no es general para todo el colectivo, pues se trata de las terceras que disponían de los medios suficientes como para negociar y acudir ante el escribano público para ratificar sus actuaciones económicas.56 En el extremo opuesto, Graña Cid 2006, 285. Doña Brianda de Villavicencio, viuda del veinticuatro Fernán Ruiz Cabeza de Vaca, en la collación de San Marcos (AMJFPN. Rodrigo de Rus. 2 de mayo de 1536. Fol.370r.); Isabel de Sierra, en la collación de San Miguel (AMJFPN. Alonso Sarmiento. 30 de junio de 1535. Fol.453v.); Luisa de Santana, en la collación de Santiago (AMJFPN. Alonso Sarmiento. 19 de febrero de 1532. Fol.138r.); y María de la Cruz, en la collación de San Miguel (AMJFPN. Baltasar de Lueña. 4 de enero de 1537. Fol.38r.); María Rodríguez, viuda, en la collación de San Miguel (AMJFPN. Rodrigo de Rus. 14 de mayo de 1536. Fol.421r.). 55 Pérez González 2005, 113-114. 56 Un ejemplo de tercera muy poderosa económicamente es el de Juana de Leiva, hija de Alonso Ortiz y Mencía de Zúñiga, quien siguió la vocación de su tía‑abuela Leonor de Zúñiga «La Buena», y también fue beata de la Orden Tercera de San Francisco, aunque la documentación nos la presenta como honesta. Ello no le impidió fundar, en 1504, un 53 54 AHPS. Sección Protocolos. Leg.17422. Año 1496. Fol.536r. 1417, octubre, 18. Juana Sánchez, donada o freila de la Orden Tercera, dona al convento de San Francisco de Sevilla dos pares de casas. Rubio 1953, documentos 244, 246, 719‑720. 46 Graña Cid 1993, 701. 47 1431, octubre, 1. AMS, Sec. 16, carpeta 5ª, nº 128. Citado en Miura Andrades 1998, 238. 48 1480, agosto, 7. Sevilla. ADPS=Archivo Diputación Provincial Sevilla, Espíritu Santo, Leg. 37, inserto en un documento de 1524. Citado en: Miura Andrades 1998, 239. 49 Bazán Díaz y González Mínguez 2005. 50 AHPS. Sección Protocolos. Leg. 17415. Año 1477. Fol. 108v 51 AHPS. Sección Protocolos. Leg.17415. Fol.108v. 52 Bejarano Rubio 1990, 51-53. 44 45 Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 32 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… existieron numerosos casos de situación de pobreza57 tan evidentes que los concejos incluyeron a estas mujeres entre los beneficiarios de las limosnas repartidas entre diversos colectivos.58 El mercado inmobiliario El mercado inmobiliario representa unos de los ámbitos económicos con mayor actividad a finales de la Edad Media.59 La documentación pone de manifiesto los diferentes negocios a los que son sometidos los inmuebles, aunque la descripción de los mismos es muy poco ilustrativa. En ella, bajo la denominación de casas, en plural, se hace referencia a un conjunto formado por unos palacios, soberados y corral, elementos que pueden estar presentes en su totalidad o en cantidad variable.60 No podemos dilucidar el tamaño de estos inmuebles y tampoco podemos averiguarlo de acuerdo con los alquileres establecidos en los contratos, pues la cantidad estipulada debió estar en proporción directa con factores61 como su localización dentro del espacio urbano, las condiciones de conservación, los materiales constructivos y el contar con alguna infraestructura especial como pozos, pilas, cocinas, mobiliario, etc. Dentro del patrimonio de las terceras las casas tienen un predominio absoluto. No podemos olvidar que su existencia se desarrollaba en un ámbito eminentemente urbano, en este caso la ciudad de Sevilla, donde realizaban todas las actividades vinculadas a su vivir cotidiano. Por tanto, debió existir una clara preferencia por unos bienes localizados en esa ciudad en la que las terceras están insertas y donde podían ejercer un control más eficiente de los mismos. La supervisión del cumplimiento de los contratos y de la conservación de su patrimonio inmobiliario podían realizarla sin tener que desplazarse fuera de la ciudad. Las cuatro casas propiedad de terceras que documentamos estaban distribuidas por la ciudad de Sevilla, con la excepción de una de ellas, que se localizaba en Jaén.62 Las collaciones hispalenses donde se encuentran dichas casas son San Vicente,63 San Miguel64 y San Marcos.65 En la mayoría de los casos desconocemos los procesos previos y necesarios por los que estas casas que formaban parte del patrimonio de las terceras se incorporaron al mismo, pues la documentación guarda un absoluto silencio al respecto y no contamos con ningún contrato de compra. Un ejemplo consignado se refiere a la herencia de unas mayorazgo con sus bienes en la persona de su sobrino Alonso, hijo de su hermano Fernando, y con cargo del entierro y altar de su tía‑abuela en la capilla mayor del convento de San Francisco (Miura Andrades 1998, 241). 57 Miura Andrades (1998, 239) documenta numerosos casos de terceras pobres como Catalina Muñoz y Juana Sánchez, vecinas de Santa Marina, en 1438 y en 1444 vivía en la collación San Juan de la Palma Juana Fernández, beata pobre. 58 Ibídem, 238-239. 59 Collantes de Terán Sánchez 1988 y 2007b. 60 Carlé 1982. 61 Collantes de Terán Sánchez 2007a. Álvarez Fernández y Suárez Beltrán 2015, 80. 62 AHPS. Sección Protocolos. Leg.2161. Año 1502. Fol.374v 63 AHPS. Sección Protocolos. Leg.17422. Año 1496. Fol.536r. 64 AHPS. Sección Protocolos. Leg.17424. Año 1498. Fol.97r. 65 AHPS. Sección Protocolos. Leg.17427. Año 1501. Sin foliar. casas en la collación de San Vicente por Ana Rodríguez la farfana.66 El tío de la citada Ana, el racionero de la Catedral Francisco Fernández, había mandado en su testamento 15.000 maravedís como dote para que su sobrina se casase o ingresase en algún convento o monasterio. La heredera de los bienes del racionero, la Cofradía de San Bernardo de la collación de San Juan, teniendo en cuenta que la citada Ana non era persona para casarse nin entrar en religión por çiertas ocupaçiones e enfermedades, efectuó el pago de los 15.000 maravedís entregándole las casas de San Vicente. Las razones aducidas por la Cofradía para ello fueron el deseo de practicar con ella la caridad, descargar el alma del racionero y que, tras profesar en la Tercera Regla de san Francisco, viue en áuito onesto e en seruiçio de Dios Nuestro Señor. Por su parte Leonor Díaz tomó posesión en 1485 de la casa y ermita de San Blas, donde se había retirado la fundadora del convento de Santa Inés,67 doña María Coronel, antes de la erección del mismo.68 Los contratos suscritos por las terceras en relación con sus casas consisten en una venta (Catalina Díaz vendió unas casas por 8.400 maravedís)69 y un arrendamiento: Leonor Fernández de Varea alquiló unas casas por un periodo de un año y una renta de 1.400 maravedís que debían pagarse trimestralmente y cuatro gallinas pagaderas ocho días antes de Navidad.70 La escasez de contratos de arrendamientos nos impide determinar si las terceras ejercían un control exhaustivo de sus propiedades, tal como lo indican el periodo corto de alquiler o si, por el contrario, preferían garantizarse unas rentas fijas durante un periodo prolongado de tiempo (contratos de tres vidas), sin necesidad de estar atentas a la renovación de los contratos cada cierto tiempo. Lo cierto es que en la época estudiada existía entre los particulares, a diferencia de las instituciones,71 una predilección por el contrato de corta duración, preferentemente anual. Este predominio de contratos de arrendamiento de tiempo limitado responde a un control exhaustivo de la economía de los particulares, al tiempo que evidencia una mentalidad mercantilista, un afán de lucro, pues la renovación anual de los contratos permitiría elevar las rentas cuando unas circunstancias de prosperidad económica lo hiciesen posible y así adecuarlos a los incrementos en el coste de la vida.72 Por tanto, nos encontramos con un mercado inmobiliario muy dinámico, donde la suscripción de contratos de arrendamiento era constante, con cambios continuos en los inquilinos. Entre las terceras no solo documentamos propietarias de casas, sino que también algunos de sus miembros alquilaron ciertas propiedades al carecer, suponemos, de las suyas propias. En el ejemplo consignado se trata de contratos de larga duración, por la vida de la tercera, fijándose la renta en 700 maravedís pagados por tercios y dos gallinas, entregadas ocho días antes de Navidad.73 AHPS. Sección Protocolos. Leg.17422. Año 1496. Fol.536r. Valdivieso González y Morales Martínez 1991, 77-102. 68 1485, s.m., s.d. Archivo del Convento de Santa Inés de Sevilla, Leg. 1, n. 1, «Libro Registro de Escrituras», apunte 72, fo 145ro. Citado por: Miura Andrades 1998, 237. 69 AHPS. Sección Protocolos. Leg.17424. Año 1498. Fol.97r. 70 AHPS. Sección Protocolos. Leg.17427. Año 1501. Sin foliar 71 Pérez González 2016, 510. 72 Cabrera Sánchez 1993, 115. 73 AHPS. Sección Protocolos. Leg.17425. Año 1499. Fol.458v. 66 67 Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… Actividades financieras Las terceras, al igual que otros miembros de colectivos femeninos,74 se dedicaron al préstamo de dinero. Esta actividad desempeñó un papel muy importante dentro de una estructura económica donde no existían los bancos75 ni una regulación en torno a los prestamistas. Esta escasez de una infraestructura favoreció el que numerosas personas pudieran participar en diferentes actividades que podían ofrecerles una alta rentabilidad a su inversión.76 Por otro lado en periodos de inflación el dinero atesorado perdía su valor provocando una bajada del poder adquisitivo. De esta forma quienes contaban con dinero atesorado tenían a su disposición diferentes maneras de invertirlo de forma conveniente y segura. Durante el periodo medieval la práctica del préstamo de dinero experimentó diversas transformaciones en lo que se refiere tanto al significado social como económico del crédito.77 Existían elementos de importancia capital tales como la reputación del prestamista y su honestidad, al mismo tiempo que era fundamental elegir a personas que ofreciesen las suficientes garantías para devolver el dinero prestado de acuerdo con las condiciones establecidas entre ambas partes.78 Las actividades de préstamo desarrolladas durante la Edad Media se caracterizaron por su marcado carácter local, pues se daban entre personas que tenían un mutuo conocimiento. En un primer momento casi todos los préstamos carecieron de un carácter profesional y se basaron en contratos no siempre ratificados legalmente que unían a vecinos y conocidos. Pero a partir del siglo XV el desarrollo de un mercado claramente expansionista condicionó que las bases sobre las que descansaba el préstamo de dinero tuvieran un carácter más legal. De esta forma las relaciones entre los distintos agentes que intervenían en estas operaciones experimentaron una redefinición en el sentido de que la confianza personal fue reemplazada por términos de igualdad contractual.79 Al hacerse más complejos los mecanismos del crédito, este ya no podía fundamentarse en un elemento tan inestable como la confianza hacia el peticionario del préstamo. Por el contrario el prestamista exigió una rigurosa puesta por escrito del préstamo y su ratificación por el escribano público, como prueba legal del acto y como garantía de su restitución bajo el cumplimiento de distintas clausulas condenatorias en caso de no devolución de la cantidad prestada con sus correspondientes intereses. En la obtención del beneficio a partir del dinero que era objeto de préstamo, el interés resultaba fundamental en el marco de una economía inflacionista pues de esta forma el poder adquisitivo del dinero se mantenía en un nivel constante. Para obtener una ganancia de la inversión realizada, esta debía ser cargada con una cantidad superior a la tasa de inflación. El valor del interés no quedaba registrado debido a la prohibición por parte de la Iglesia de Lemire 2001, 18-23. Monguio Becher 1978, 50. 76 Colombo 2016, 257. 77 Bourin 2007, 108. 78 Tittler 1994, 255. 79 Así sucede en el ámbito andaluz, mientras que en el mundo británico este se dio con posterioridad, a partir de los siglos XVI y XVII (Muldrew 1998, 7). 33 la usura.80 En su lugar el interés se ocultaba dentro de la cantidad total que debe ser pagada por el deudor, bajo la fórmula que los juristas denominan mutuo oneroso. Dentro de las actividades vinculadas al préstamo documentamos terceras que pusieron en práctica la fórmula de la venta de rentas. En la época estudiada tanto Sevilla como Jerez de la Frontera eran importantes ciudades comerciales con un destacado desarrollo de la economía monetaria. Sin embargo a lo largo del periodo estudiado las dificultades monetarias eran evidentes así como la falta de liquidez.81 Además el desarrollo de la economía monetaria no benefició de igual forma a todos los vecinos, que sí se vieron afectados por los resultados de la misma puesto que los precios subían y su poder adquisitivo se reducía. La necesidad de dinero se intentó paliar a través de distintas fórmulas. Una de ellas fue los sistemas de crédito. Desde fines del siglo XV existían dos supuestos básicos para el desarrollo de los sistemas de crédito: una persona necesitada de dinero y el capitalista que deseaba hacer fructificar los ahorros monetarios de que disponía. En estos momentos la normativa de la Iglesia sobre la usura había sobrepasado ampliamente los límites establecidos por el Derecho canónico. El freno a cualquier operación de crédito era evidente, aunque ello no impidió el desarrollo de fórmulas financieras. En la época estudiada se van a poner en práctica medios indirectos para prestar dinero y recibir lo prestado en cantidad superior a través de la entrega diferida.82 Una de estas fórmulas fue la compra de rentas. Se trataba de una operación según la cual un propietario vendía un censo o tributo perpetuo sobre sus bienes por una cierta cantidad de dinero. El proceso consistía en que un propietario necesitado de dinero suscribía un contrato similar a una compra-venta. El bien vendido era una parte de lo que rentaba la propiedad, que quedaba sujeta a perpetuidad a un canon anual que debía entregar al comprador. En la operación el propietario conseguía la cantidad de dinero líquido que precisaba, mientras el comprador efectuaba una inversión de capital de la que iba a percibir anualmente unos ingresos, pero no la devolución del principal. La fórmula en sí misma no se puede considerar como un auténtico préstamo de capital, puesto que en este se exigía el reembolso del capital en un tiempo determinado y en el caso de la compra de rentas el vendedor recibía el capital para siempre sin la obligación de devolverlo en un plazo concreto.83 Hubo terceras que estuvieron perfectamente informadas y preparadas para acudir a esta forma de préstamo disfrazado bajo la fórmula de la compra de rentas. Es el caso de María de la Cruz, vecina de Jerez, quien vendió 5.000 maravedís por 500 maravedís situados sobre unas casas en la collación de San Miguel. La cantidad prestada son los 5.000 maravedís que la tercera cobraría a perpetuidad a razón de 500 maravedís anuales. La garantía del pago eran las casas.84 En el caso de Sevilla contamos con el ejemplo de Ana Rodríguez, quien prestó 3.000 maravedís a cambio 74 75 80 Benito Ruano 1970. Clavero 1979. García de Valdeavellano 1973. Nelson 1969. 81 Borrero Fernández 1986, 233. 82 Se trata de medios simulados in fraudem usurae. Clavero Salvador 1977. 83 De Almeida Costa 1961, 78. 84 AMJFPN. Baltasar de Lueña. 4 de enero de 1537. Fol.38r. Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 34 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… de 300 maravedís anuales pagados a perpetuidad, siendo el bien garante del préstamo unas casas en la collación de San Vicente.85 Actividades comerciales Además de sus obligaciones religiosas, hubo terceras que se dedicaron a actividades con un carácter eminentemente comercial. Desde finales del siglo XV el Reino de Sevilla se convirtió en el centro fundamental de la economía castellana. Esta zona constituía una encrucijada comercial, que actuaba como zona de contacto de los principales núcleos mercantiles de la época.86 De importancia muy destacada fueron las relaciones con el Atlántico septentrional, que no se limitaban al intercambio bilateral de productos. Las mercancías que desde el Norte llegaban a las costas atlánticas andaluzas abastecían tanto al mercado local como a los del interior de Castilla, o seguían hasta lugares más alejados en el ámbito mediterráneo o atlántico.87 En el sentido contrario, los productos recogidos en la Baja Andalucía para su exportación eran enormemente variados, ya que procedían de diversas zonas de producción. A este complejo comercio contribuyó la diversidad de sus agentes. Por un lado, un amplio abanico de comerciantes castellanos y extranjeros, que controlaban diversos mercados. Por otro lado, la colaboración entre mercaderes y transportistas de orígenes diversos. Finalmente, la convergencia de relaciones puramente comerciales con otras de carácter pirático.88 Entre esos comerciantes documentamos a las terceras que son objeto de nuestro estudio. Actuaron como intermediarias en negocios que, en principio, requerían una cierta especialización en un ámbito de predominio masculino en cuanto a sus intervinientes. Se evidencia al mismo tiempo un profundo conocimiento del mercado de la ciudad y de los productos que mejor salida tenían y mayores beneficios reportaban, donde la intervención femenina era muy minoritaria. Adquirieron los conocimientos que exigían y contactaron con los demandantes de tales productos, negociando para obtener los más pingües beneficios. Es el caso de la tercera franciscana Catalina de la Cruz,89 que estaba especializada en el tráfico del azúcar blanco de pilón.90 Las cantidades con las que comercializaba eran considerables: 75 arrobas (862,5 kilos, a 11 kilos y 502 gramos la arroba), apreciada la arroba en 406.66 maravedís, y una caja valorada en 4.800 maravedís, que debía contener, atendiendo a las cifras indicadas, unas 11.8 arrobas (135,72 kilos). Las cantidades consignadas son apreciables, por lo que cabe pensar que la citada Catalina de la Cruz era un importante agente comercial en el tráfico del azúcar, situación que se sale del común denominador de las terceras franciscanas. En la época estudiada el azúcar no tenía un consumo tan extendido como en la actualidad. Por este motivo su precio AHPS. Sección Protocolos. Leg.2158. Año 1500. Fol.107v. Aznar Vallejo 2003, 103. 87 Sevillano Colom 1970, 364. 88 Aznar Vallejo 1997, 419. 89 AHPS. Sección Protocolos. Leg.2158. Fol.107v. 90 Se trata de un azúcar refinado, obtenido en panes de forma cónica. Para la tipología de azúcares en este periodo véase: Ashtor 1981, 233-236. era elevado y solo podían acceder a ella quienes poseían los suficientes recursos económicos, mientras que los grupos menos pudientes empleaban la miel como edulcorante.91 Los fines fundamentales del azúcar eran la medicina y la farmacia, y al tratarse de un producto de lujo las redes comerciales por las que circulaba tenían un alcance corto.92 Actividades profesionales La mayoría de las terceras no parecen desarrollar una actividad profesional, cumpliendo así con el espíritu franciscano de prescindir del trabajo como fuente de ingresos y realizar tan solo el necesario para subsistir, a excepción de la citada Catalina de la Cruz. No era este el caso de Teresa López,93 tercera franciscana, que trabajaba como criada en casa de Alfonso López, curtidor. Los servicios que le prestó durante cuatro años y cuatro meses le reportaron unos ingresos de 6.800 maravedís, una cantidad bastante superior a lo necesario para garantizar la subsistencia. Para el periodo analizado el servicio doméstico ya no se fundamentaba, como sucedía anteriormente, en una red de muy diversas interacciones que incluía conceptos como la lealtad, la protección e, incluso, derechos de propiedad.94 Una de las bases de este sistema era una fuerte jerarquización que hacía completamente imposible el ascenso social y limitaba la obtención de cualquier tipo de objetivo personal.95 Pero a fines de la Edad Media el carácter del trabajo doméstico comenzó a cambiar hacia un sistema de trabajo remunerado, que permitía a los criados negociar las condiciones laborales con quienes los empleaban. Con ello, y de forma progresiva, su remuneración pasó a estar sujeta a las condiciones del mercado para cambiar su posición o ascender socialmente. Dentro del desarrollo de la cultura del consumismo fue surgiendo un concepto del servicio doméstico determinado por las ideas de producción y beneficio.96 Esta nueva concepción fue especialmente importante para el sector femenino de la sociedad que fue paulatinamente reemplazando al masculino como servicio doméstico en las casas de importante nivel económico. La naturaleza temporal del trabajo basado en un sueldo conllevó distintas consecuencias para las mujeres que para los hombres, pues su trabajo como criadas se consideraba no solo una forma de ocupación laboral per se (como en el caso de los hombres), sino también un medio para formarse en los conocimientos relativos a la casa que habrían de necesitar como esposas.97 Las criadas también se vieron inmersas en la economía de mercado. Dado que con la nueva concepción del trabajo doméstico percibían un sueldo en metálico a cambio de sus servicios, tuvieron que decidir cómo emplear ese dinero, bien en un gasto inmediato o en alguna forma de inversión que protegiera e incrementara los ahorros. Una criada podía 85 86 91 92 93 94 95 96 97 Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 Fábregas García 2013. Coulon 2001, 729. AHPS. Sección Protocolos. Leg.2158. Fol.107v. Wall 2002, 201-203. Neill 2000, 22. Kussmaul 1981, 8-10. Hanawalt 1986, 145. Barron 1996, 147. S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… ser capaz de recabar a lo largo de su vida una importante suma de dinero o bienes para su uso personal en el futuro.98 Desconocemos el fin de los 6.800 maravedís ganados por Teresa López. Relaciones familiares y sociales Las terceras mantuvieron estrechas relaciones con sus familiares,99 lo cual hubiera sido imposible, en principio, en el caso de profesar en alguna Orden como monjas. Esta vinculación presenta distintas formas de manifestación. Relaciones de carácter económico, como los casos de Isabel Sierra que vendió a su hermano su olivar en el Pago de Capirote100 por 4.500 maravedís,101 y el de doña Brianda de Villavicencio,102 viuda103 de don Fernán Ruiz Cabeza de Vaca, veinticuatro, que arrendó a un miembro de su familia, Benito Núñez de Villavicencio, también veinticuatro, un donadío de tierras en la Fuente de los Ballesteros.104 Los datos biográficos que aporta el documento resultan de un gran interés. Brianda había enviudado y, posiblemente, fue tras la muerte de su marido105 cuando ingresó en la Orden Tercera franciscana. En ese mismo año, 1536, documentamos a Brianda de Villavicencio en el Convento de Madre de Dios, pero recibe el calificativo de monja profesa.106 En el apartado anterior hemos indicado que este Instituto, como convento femenino, fue una fundación autogenética resultado del ingreso en el mismo de un grupo de beatas franciscanas. Estamos ante uno de esos ejemplos en los que a las beatas se las denomina monjas cuando el acto documental tenía lugar en el edificio conventual, pero realmente pertenecían a la Orden tercera y así se las califica en el desenvolvimiento de actividades personales e individuales.107 Al mismo tiempo el documento pone de manifiesto la convivencia entre terceras y monjas, pues el Convento contaba por esa fecha con una abadesa y una vicaria, y junto a ellas estuvo Brianda de Villavicencio en el acto formal registrado ante notario de la entrega de una dote108 de 70.000 maravedís por parte de Ángela de Orellana. Se trataba de una niña de 11 años que a la muerte de su padre quiso ingresar en el Convento de Madre de Dios. Muy especial debió ser el vínculo entre María Rodríguez y la hija de su sobrino, a quien mandó para su dote,109 y por ser commo soys doncella onesta e tenéys pocos bienes, 5.000 maravedís que obtendría del negocio a Mcintosh 2005, 49. Pastor de Togneri 2004. 100 Martín Gutiérrez 2004, 140. 101 AMJFPN Alonso Sarmiento. 30 de junio de 1535. Fol.453v. 102 Ruiz Pilares 2012. 103 No es el primer caso de viuda que tras el fallecimiento del marido decidió ingresar en la Orden Tercera Franciscana. En Córdoba contamos con el ejemplo de doña Marina de Aguayo. Véase: Graña Cid 2006, 288-289. 104 AMJFPN Rodrigo de Rus. 2 de mayo de 1536. Fol.370r. 105 La condición de viuda permitía a la mujer medieval gozar de un amplio margen de actuación en muy diversos ámbitos personales y sociales: Pérez González 2010, 35-46. 106 AMJFPN. Luis de Llanos. 1536. Fol.600r. 107 En ocasiones se las denomina Clarisas de la Tercera Regla: Miura Andrades 1998, 251. 108 Pérez Carazo 2008, 105. 109 Sánchez Collada 2016, 717-718. 98 99 35 que fuese sometida la casa de la tercera tras su muerte.110 Este ejemplo resulta sumamente interesante, puesto que una mujer que había optado por una forma de vida muy distinta de la que la sociedad esperaba de ella (un marido, los hijos, el hogar) contribuyó con su donación a que una sobrina cumpliese con el principal rol asignado al colectivo femenino.111 Una categoría diplomática a partir de la cual podemos tener un mayor conocimiento de la implicación social de las terceras son los poderes. En ellos se comisiona a una persona para llevar a cabo una determinada actuación en nombre de quien otorga dicho documento. Pero, dado que el acto principal de este tipo de contrato es precisamente esa comisión autorizada, se omiten otros datos en relación con la misión que se va a cumplir, datos que nos permitirían conocer algunos patrimonios de miembros de las Órdenes Terceras, pero que resulta imposible al prescindirse de detalles que se consideran innecesarios para la concesión del poder. El ejemplo más claro de cuanto hemos afirmado es el de María Castillo,112 que otorgó un poder a Martín Castillo, clérigo, vecino de Alcalá de Guadaira, para que en su nombre cobrase lo que le debían de las rentas de ciertos bienes que ella tenía en dicha localidad. Fuera de toda duda queda el cometido que fray Pedro, fraile profeso del convento de Santa María de la Merced, debía llevar a cabo en nombre de Ana Fernández de Fonseca:113 cobrar a Francisco de Ávila 1.000 maravedís que debía a Beatriz de Fonseca, difunta, hija de la citada Ana. Llamamos la atención sobre el hecho de que siendo tercera franciscana no encomiende el cobro a un fraile de su Orden, sino a un mercedario. Quizás la confianza que Ana tenía depositada en fray Pedro era mayor que en otro cualquier hermano franciscano. El prestigio socio-religioso Los miembros de las Órdenes Terceras gozaron de un prestigio importante dentro de la sociedad a la que pertenecían. Estas mujeres optaron por una forma especial de religiosidad con una doble inspiración activa-contemplativa, lo que se traducía en una combinación entre oración y desempeño de funciones englobadas en la cura animorum. Las labores asistenciales, caritativas, hospitalarias y solidarias que desarrollaron fuera de los claustros las revistieron de una consideración especial y las hacían portadoras de unos valores de especial reconocimiento social.114 De hecho es una dimensión pastoral, de caridad y de acompañamiento en la acción interna y externa lo que constituye uno de los rasgos más distintivos de la forma de vida que representan las terceras. Entre las actuaciones más vinculadas a cuestiones de índole religiosa observamos que es en los testamentos y, más en concretamente, en el nombramiento de las terceras como albaceas, donde estas tuvieron un protagonismo más destacado. Su persona, su carácter cuasi sagrado derivado de las bendiciones y consagraciones que habían recibido y 110 111 112 113 114 AMJFPN Rodrigo de Rus. 14 de mayo de 1536. Fol.421r. Leva Cuevas 2008. AHPS Sección Protocolos. Año 1502. Leg.2161. Fol.43r AHPS Sección Protocolos. Leg.1500. Fol.750r. Graña Cid 2006, 291. Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 36 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… realizado al ingresar en los grupos terceros y, creemos que especialmente, una estrecha relación con quien expresaba su última voluntad, convirtieron a algunas terceras en garantes del cumplimiento de una serie de mandas cuyo fin las hacía merecedoras de una atención especial: las honras fúnebres y celebraciones pro remedio animae. Entre los nombramientos como albaceas testamentarios115 documentamos a Ana Sánchez,116 que fue designada para el cumplimiento de sus mandas testamentarias por Juana de Casas, honesta. La citada Ana Rodríguez acabó rechazando el cometido por hallarse enferma. Podemos pensar en una vinculación derivada de unas opciones similares de formas de vida cristiana. Conclusiones El estudio de las mujeres pertenecientes a la Orden tercera franciscana no es sencillo debido a su carácter complejo y heterogéneo. Las terceras tenían una condición canónica difícil de precisar, situada entre lo laico y lo eclesiástico, con una adaptación permanente a las diferentes coyunturas. Ello dificulta en extremo la localización de evidencias documentales que permitan establecer con rotundidad tiempos y tipologías. Las terceras franciscanas representan una de las novedades religiosas más destacadas del periodo bajomedieval. Su vida religiosa se basaba en los ideales que hicieron posibles la promoción espiritual del laicado desde la Baja Edad Media tales como la penitencia, la comunicación directa con Dios y la pobreza. Al mismo tiempo gozaron de un carisma activocontemplativo que las asemejaba a los beaterios en cuanto a un modelo existencial apostólico fuera de la rigidez claustral. Finalmente, su estilo de vida recibió reconocimiento público hasta casi caracterizarse como verdadera profesión, y fueron capaces de ser identificadas como colectivo femenino a nivel social y dentro de su Orden lo que supuso la institucionalización de las terceras. Sin embargo esa identificación institucional y profesional, favorecida sin duda por la vinculación con el convento franciscano y una clara sintonía institucional con los frailes, no estaba acompañada de una definición terminológica precisa. Los escribanos públicos no tenían una denominación exacta para aplicarla a los miembros femeninos de la Orden tercera. Ni siquiera ellas reclamaron un único nombre que definiera su religiosidad laical, por lo que creemos que ni la sociedad ni las terceras consideraron necesario implementar un proceso de concretización terminológica. A nivel social las terceras tuvieron un escaso protagonismo en la ciudad como propietarias de distintos bienes, que se limitan a casas y rentas perpetuas. La escasez documental que existe al respecto no creemos que sea resultado de una pérdida de los asientos notariales, sino más bien el resultado del acatamiento de los principios regidores de las Órdenes Mendicantes, que obligaba a una renuncia casi absoluta de todo lo que fuese contrario a una existencia dominada por la austeridad y la pobreza. Por el contrario, ciertas actividades comerciales y financieras visibilizaron a algunas de las terceras como protagonistas importantes e, incluso, poderosos en las redes que articulaban ambos mercados. El comercio de 115 116 García Guzmán y Abellán Pérez 1997, 54. AHPS Sección Protocolos. Año 1504. Leg.17422. Fol.536r. un bien de lujo como el azúcar y los contratos de compras de rentas posibilitaron el acceso de estas mujeres a una esfera eminentemente masculina, que exigía unos certeros conocimientos de sus mecanismos así como de un capital que prestar o invertir según los casos. Esta forma de vida religiosa fuera del marco institucional tradicional que representaba el convento hizo posible el mantenimiento y alimento de los vínculos familiares y amistosos. Estas relaciones con los miembros de la propia familia discurrieron por cauces tan diversos como los vinculados al mundo de los negocios o a las actividades solidarias ejercidas con otras mujeres del núcleo familiar. En ocasiones los lazos personales se estrecharon sobre la base de la mutua confianza. Por tanto, las terceras representan una opción personal de vivir la religiosidad cristiana por parte de un grupo de mujeres que decidieron ser dueñas de sus destinos, por lo que renunciaron a toda forma de institucionalización que las pudiese constreñir en su individualizada toma de decisiones. Perfectamente integradas en buena parte de los mecanismos articuladores de la sociedad a la que pertenecían, decidieron vivir la religiosidad cristiana de forma no original, pues estaban impregnadas de todo un movimiento espiritual vigente en la época, pero sí de forma personal y adaptada a su vivir cotidiano en un entorno social al que no estaban dispuestas a renunciar y en el que ejercieron su influencia a través de distintos cauces de implicación. Cuando los valores que representaban dejaron de ser considerados válidos, pues la jerarquía eclesiástica comenzó tomar medidas en relación a los colectivos de mujeres «sueltas», las terceras, al igual que otros grupos similares, hubieron de elegir dos caminos: la integración en el marco institucional a través de la vía de la clausura en el convento de la Orden segunda franciscana; o continuar con las pautas existenciales por las que había optado, habiendo de sufrir todo tipo de presiones por parte de la jerarquía eclesiástica. La falta de estudios sobre estos grupos durante el Barroco nos impide dilucidar en qué acabaron convirtiéndose o cómo se adaptaron a las nuevas y poco favorables circunstancias para su vivir diario. Bibliografía Alvarado Planas, Javier y Regina María Pérez Marcos. 2010. «Feudalismo y Derecho feudal». En Historia del Derecho Español, coord. Javier Alvarado Planas, 324-335. Madrid: Universidad Nacional de Educación a Distancia. Almansa Segura, Rufino. 2003. «El monasterio de San Juan de la Penitencia de Cazorla, de monjas clarisas». En El Franciscanismo en Andalucía, ed. Manuel Peláez del Rosal, 299-308. Córdoba: Cajasur. Álvarez Fernández, María y Soledad Beltrán Suárez. 2015. Vivienda, gestión y mercado inmobiliario en Oviedo en el tránsito de la Edad Media a la Modernidad. Vitoria: Universidad del País Vasco. Ashtor, Eliyahu. 1981. «Levantine Sugar Industry in the Late Middle Ages: A Case of Technological Decline». En The Islamic Middle East, 700-1900: Studies in Economic and Social History, ed. Abraham Udovitch, 220-260. Princeton: Darwin Press. Aznar Vallejo, Eduardo. 1997. «Corso y piratería en las relaciones entre Castilla y Marruecos en la baja Edad Media». En la España Medieval 20: 407-419. Aznar Vallejo, Eduardo. 2003. «Andalucía y el Atlántico norte a finales de la Edad Media». Historia, Instituciones y Documentos 30: 103-120. Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… Barron, Caroline. 1996. «The Education and Training of Girls in FifteenCentury London». En Courts, Counties and the Capital in the Later Middle Ages, 139-153. New York: Sutton Publishing Company. Bazán Díaz, Iñaki y Cesar González Mínguez, coord. 2005. El discurso legal ante la muerte durante la Edad Media en el nordeste peninsular. Vitoria: Universidad del País Vasco. Bejarano Rubio, Amparo. 1990. El hombre y la muerte. Los testamentos murcianos bajomedievales. Cartagena: Ayuntamiento de Cartagena. Benito Ruano, Eloy. 1970. «Usura y cambios en el León medieval». Archivos Leoneses 47-48: 203-208. Bermúdez de Pedraza, Francisco. 1608. Antigüedad y excelencias de Granada. Madrid: Por Luis Sánchez, impresor del rey. Borrero Fernández, Mercedes. 1986. «Efectos del cambio económico en el ámbito rural. Los sistemas de crédito en el campo sevillano (fines del siglo XV y principios del XVI». En la España medieval V: 219-244. Bisson, Thomas N. 1989. Feudalism in Twelfth Century Catalonia, en Medieval France and her Pyrenean Neighbours. Studies in Early Institutional History. London-Ronceverte: The Hambledon Press. Bourin, Michel. 2007. «Peasant Elites and Village Communities in the South of France, 1200-1350». Past And Present 2: 101-114. Cabrera Sánchez, Margarita. 1993. «Oligarquía urbana y negocio inmobiliario en Córdoba en la segunda mitad del siglo XV». Historia, Instituciones y Documentos 20: 107-126. Carlé, María del Carmen. 1982. «La casa en la Edad media castellana». Cuadernos de Historia de España 67-68: 165-229. Cedergre, Harry R. 1983. «Sociolingüística». En Introducción a la lingüística actual, coord. Humberto López Morales. Madrid: Playor. Clavero Salvador, Bartolomé. 1977. «Prohibición de la usura y constitución de rentas». Moneda y crédito: 107-131. Clavero Salvador, Bartolomé. 1979. «Interesse: traducción e incidencia de un concepto en la Castilla del siglo XVI». Anuario de historia del derecho español 49: 39-98. Collantes de Terán Sánchez, Antonio. 1988. «Propiedad y mercado inmobiliario en la Edad Media: Sevilla, siglos XIII-XVI». Hispania 169: 493-528. Collantes de Terán Sánchez, Antonio. 2007a. «El modelo meridional, Sevilla». En Mercado inmobiliario y paisajes urbanos en el Occidente europeo (siglos XI-XV), 591-630. Pamplona: Gobierno de Navarra. Collantes de Terán Sánchez, Antonio. 2007b. «Los centros urbanos andaluces de la frontera con Granada». En Jornadas de Historia de Lucena, 41-66. Lucena: Excelentísimo Ayuntamiento. Colombo, Octavio. 2016. «Los dueños del dinero. Prestamistas abulenses a mediados del siglo XV». Espacio, tiempo y Forma. Historia Medieval 29: 249-277. http://doi.org/10.5944/etfiii.29.2016 Coulon, Damien. 2001. «El comercio catalán del azúcar en el siglo XIV». Anuario de Estudios Medievales 31 (2): 729-756. https://doi. org/10.3989/aem.2001.v31.i2.267 De Almeida Costa, Mario Joao. 1961. Raizes do censo consignativo. Para la historia do crédito medieval portugués. Coimbra: Atlántica. Fábregas García, Adela. 2013. «El azúcar en los mercados europeos a fines de la Edad Media. El ejemplo del azúcar nazarí». En El sucre en la historia: alimentació, quotidianitat i economía, coord. Flocel Sabaté y Currull, 147-160. Lleida: Ajuntament de Menàrguens. García de Valdeavellano, Luis. 1973. «El “Renovo”: notas y documentos sobre los préstamos usurarios en el reino astur-leonés (siglos X-XI)». Cuadernos de Historia de España 57-58: 408-448. García Guzmán, María del Mar y Juan Abellán Pérez. 1997. La religiosidad de los jerezanos según sus testamentos (siglo XV). Cádiz: Agrija Ediciones. Graña Cid, María del Mar. 1993. «Las primeras clarisas andaluzas: franciscanismo femenino y reconquista en el siglo XIII». Archivo Ibero-Americanos 215-216: 661-704. Graña Cid, María del Mar. 1997. «Franciscanos y franciscanas en el reino de Granada. Panorama fundacional (ca. 1485-1550)». En El Franciscanismo en Andalucía, ed. Manuel Peláez del Rosal, 105119. Córdoba: Cajasur. 37 Graña Cid, María del Mar. 2002a. «En torno a la fenomenología de las santas vivas (Algunos ejemplos andaluces, siglos XV-XVI)». En Responsabilidad y diálogo. Homenaje a José Joaquín Alemany Briz, SJ (1937-2001), ed. Xavier Quinzá Lleó y Gabino Uríbarri Bilbao, 415-454. Madrid: Universidad Pontificia Comillas. Graña Cid, María del Mar. 2002b. «La santa/bruja Magdalena de la Cruz. Identidades religiosas y poder femenino en la Andalucía pretridentina». En Actas del III Congreso de Historia de Andalucía, 103-120. Córdoba: Universidad de Córdoba. Graña Cid, María del Mar. 2006. «Las terciarias franciscanas en la pugna conventuales /observantes. Jurisdicción religiosa y problemas de género». En Los Franciscanos Conventuales en España. Actas del III Congreso Internacional sobre el Franciscanismo en la Península Ibérica, coord. Gonzalo Fernández-Gallardo Jiménez, 283-294. Priego de Córdoba: Asociación Hispánica de Estudios Franciscanos. Graña Cid, María del Mar. 2011. «Autoría femenina y carisma religioso. Orígenes de las clarisas descalzas en Andalucía». En Historia magistra vitae. Miscelánea de estudios en homenaje a Tarsicio de Azcona, ed. José Ángel Echeverría. Pamplona: Hermanos Menores Capuchinos. Hanawalt, Barbara. 1986. Women and Work in Preindustrial Europe. Bloomington: Indiana University. Kussmaul, Adolf. 1981. Servants in Husbandry in Early Modern England. Cambridge: University Press. Lemire, Beverly. 2001. The business of everyday life. Gender, practice and social politics in England. Manchester: University Press. Leva Cuevas, Josefa. 2008. «El papel de la mujer en la Edad Media. La dote ¿impulsora del nuevo hogar o yugo para las mujeres». Ámbitos: Revista de Estudios de Ciencias Sociales y Humanidades 19: 69-90. Martín Gutiérrez, Emilio. 2004. La organización del paisaje rural durante la Baja Edad Media. El ejemplo de Jerez de la Frontera. Sevilla: Universidad de Sevilla. Mcintosh, Majori Keniston. 2005. Working Women in English society, 1300-1620. Cambridge: University Press. Meersseman, Gilles Gerard. 1961. Dossier de l´ordre de la penitence au XIIIe siècle. Spicilegium Fruburgense, 7. Friburg: Editions Universitaires. Meersseman, Gilles Gerard. 1977. Ordo Fraternitatis. Confraternite e pietá dei laici nel Medioevo. II. Roma: Herder Editrice e Librería. Mesa Xinete, Francisco. 1888. Historia sagrada y política de la muy noble y muy leal ciudad de Jerez de la Frontera. Jerez: Imprenta de Melchor García Ruiz. Miura Andrades, José María. 1992. Fundaciones religiosas y milagros en Écija a fines de la Edad Media. Écija: Gráficas Sol. Miura Andrades, José María. 1994. «Las fundaciones de clarisas en Andalucía del siglo XIII a 1525». En Las clarisas en España y Portugal, Actas del Congreso Internacional, coord. José Martí Mayor y María del Mar Graña Cid, 705-721. T. II, vol. II. Salamanca: Junta de Castilla y León. Miura Andrades, José María. 1998. Frailes, monjas y conventos. Las Órdenes mendicantes y la sociedad sevillana bajomedieval. Sevilla: Diputación Provincial. Miura Andrades, José María. 2014. Monjes, frailes, monjas y otras formas de vida religiosa en Jerez de la Frontera a fines de la Edad Media». En 750 aniversario de la incorporación de Jerez a la Corona de Castilla: 1264-2014, 559-575. Jerez de la Frontera: Ayuntamiento. Monguio Becher, Fernando. 1978. Antecedentes de la banca en Jerez de la Frontera. Cádiz: CSIC. Muldrew, Craig. 1998. The Economy of Obligation: The Culture of Credit and Social. New York: St. Martin’s Press. Neill, Michael. 2000 Putting History to the question: Power, Politics, and Society in English Renaissance Drama. New York: Columbia University. Nelson, Benjamin. 1969. The Idea of Usury. From Tribal Brotherhood to Universal Otherhood. Chicago: University of Chicago. Pastor de Togneri, Reina. 2004. «Mujeres en los linajes y en las familias: las madres, las nodrizas, mujeres estériles. Funciones, espacios, representaciones». En Mujeres, familia y linaje en la Edad Media, coord. Carmen Trillo San José, 31-68. Granada: Universidad de Granada. Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002 38 S. M.ª PÉREZ GONZÁLEZ Y J. SÁNCHEZ HERRERO, LOS MIEMBROS FEMENINOS DE LA TERCERA ORDEN FRANCISCANA EN ANDALUCÍA… Pellicia, Guerrino y Giancarlo Rocca, eds. 1992. Dizionario degli Instituti di Perfezione. cc. 1042-1050. T. 6. Roma: Edizioni Paoline. Pérez Carazo, Pedro. 2008. Santa María de Herce y su abadengo en la Edad Media. Logroño: Instituto de Estudios Riojanos. Pérez González, Silvia María. 2005. Las mujeres en la Sevilla de finales de la Edad Media. Solteras, casadas y vírgenes consagradas. Sevilla: Servicio de Publicaciones de la Universidad de Sevilla. Pérez González, Silvia María. 2010. «Mujeres liberadas de tutela masculina: de solteras y viudas a fines de la Edad Media». Cuadernos Kóre 2: 31-54. Pérez González, Silvia María. 2016. «Nuevas aportaciones al estudio de la cofradías y hermandades en la Castilla bajomedieval: El ejemplo de Jerez de la Frontera». Hispania Sacra 138: 502-520. https://doi. org/10.3989/hs.2016.032 Rubio, Germán. 1953. La custodia franciscana de Sevilla. Ensayo histórico sobre sus orígenes, progresos y vicisitudes (1220-1499). Sevilla: San Antonio. Ruiz Pilares, Enrique. 2012. «El mayorazgo del veinticuatro Pedro Camacho de Villavicencio ‘el rico’ (1507). El patrimonio del caballero jerezano más acaudalado de su tiempo». En la España Medieval 35: 317-347. Sánchez Collada, Teresa. 2016. «La dote matrimonial en el Derecho castellano de la Baja Edad Media. Los protocolos notariales del Archivo Histórico Provincial de Cuenca (1504-1507)». Espacio, Tiempo y Forma. Serie III. Historia Medieval 29: 699-734. Sánchez García, Francisco José. 2009. El léxico religioso en los diccionarios de la Real Academia Española: estudio diacrónico. Barcelona: Publicacions i Edicions Universitat de Barcelona. Sancho de Sopranis, Hipólito y Juan Lastra y Terry. 1964. Historia de Jerez de la Frontera. T. II. Jerez: Editorial Jerez Industrial. Serrano Estrella, Felipe. 2008. Órdenes mendicantes y ciudad. El patrimonio conventual de Jaén en la Edad Moderna. Granada: Universidad de Granada. Sevillano Colom, Francesc. 1970. «Mallorca y Castilla (1276-1343)». Boletín de la Sociedad Castellonense de Cultura 46: 321-366. Tittler, Robert. 1994. «Money-Lending in the West Midlands». Historical Research 164: 249-263. Torres, Alonso. 1683. Chrónica de la Santa Provincia de Granada, de su regular observancia de nuestro seráfico Padre San Francisco. Madrid: Por Juan García Infanzón, impresor de libros. Valdivieso González, Enrique y Alfredo José Morales Martínez. 1991. Sevilla oculta: Monasterios y conventos de clausura. Sevilla: Francisco Arenas Peñuela. Wall, Wendy. 2002. Staging, Domesticity: Household, Work and English Identity in Early Modern Drama. Cambridge: University Press. Hispania Sacra, LXXII 145, enero-junio 2020, 25-38, ISSN: 0018-215X, https://doi.org/10.3989/hs.2020.002
https://openalex.org/W4246077980	https://figshare.com/articles/thesis/_With_an_Ear_to_the_Line_An_Auditory_Reading_of_Seamus_Heaney_s_Wintering_Out_1972_/14672274/1/files/28169997.pdf	English	null	“With an Ear to the Line”: An Auditory Reading of Seamus Heaney’s Wintering Out (1972)	null	2,021	cc-by	121,875	i i By ii Abstract The Irish Nobel laureate Seamus Heaney (1939-2013) is arguably one of the most widely read, studied and translated poets of our time. Sound − in all its manifestations, literal and metaphorical, including silence − is referenced throughout his poetry and prose. From the rasping sound of his father’s spade to the shunting of trains; from the gurgling of the local river to the silence of bogs and bog-bodies; from the noise of tractors and airplanes to the quietude of lakes and canals, moments of sound and silence inspire and define not only Heaney’s poems but also our experience of them. As Heaney himself indicates, listening was a way for him not only of perceiving the surrounding world but also reaching out and staying in touch with the wideness of the world. In his Nobel Lecture (1995), Heaney recalls how as a child he would take in “everything that was going on” beyond the walls, from “the sounds of the horse in the stable at night, the voices of adults conversation from the kitchen”, “a steam train rumbling along the railway line one field back from his house” to the “bursts of foreign languages” coming from the radio. This study combines the linguistic and literary practice of close reading with ecological theories of auditory perception and soundscape interpretation to map and analyse references to sounds − and their absence − in Wintering Out (1972). This collection has been chosen because it was published at a transitional stage in the poet’s personal and professional life. Heaney’s third collection is born out of everyday childhood memories and his concerns about identity, territory, language, religion and history. It documents the poet’s standpoint in relation to the Troubles, his anxieties as a young parent, his hopes for the appreciation of the common ground and his confidence about his vocation as a poet. Wintering Out echoes the poet’s thoughts and concerns through moments of sound and silence. Historically, studies of sounds and audition have been informed by a concern with the understanding of music and the physical attributes of sound waves − e.g. amplitude, iii frequency, timbre. Studies of sounds in poetry have focused primarily on understanding prosody and the relationship between poetry and music. Abstract This acoustic study of Heaney’s Wintering Out sets out to demonstrate that references to sounds in poetry are not only guided by a feel for the sounds of words but also by a strong sense of places and times they evoke, and thus, can be socially, culturally, and personally charged and meaningful. iv Acknowledgements My depth of gratitude extends to a wide range of people who have informed and assisted me in this project. I want to express my heartfelt thanks to my supervisors, Associate Professor Marco Sonzogni and Professor Peter Whiteford at Victoria University of Wellington, New Zealand. Their professional advice and support have been invaluable in guiding me through the ups and downs of my research journey and in managing what seemed to be an unruly amount of ideas. My heartfelt thanks also goes to Associate Professor Sydney Shep from Wai-te-ata Press and Associate Professor Dugal McKinnon from the School of Music, at Victoria University of Wellington, who kindly and generously offered suggestions and supported my research from a historical and sound studies point of view. I would also like to take this opportunity to express my sincere appreciation to my viva examiners, Dr Eugene O’Brien from the University of Limerick, Dr Maebh Long from the University of Waikato, and Professor Harry Ricketts from Victoria University of Wellington for their invaluable insights and suggestions. Before and during the course of my research, as I travelled, I met a lot of people who in one way or another inspired me, I am grateful to them all. In particular, I would like to thank Mbah Siwang, the chief of Temuan tribe, for sharing with me her stories of silent meditation in Nature; Bakri Razak, who went out of his way to make this meeting possible; and Karen Adgo, who over the course of drowning in this thesis gave me reality checks and helped me trust my intuition. I shall not forget my equine and furry companions. Toofan (the old man), Wireina (my muse) and most recently Delboy (the little man) have helped give insight into the complexities of the human and non-human relationship and have kept my spirit up. I also wish to acknowledge the moral and emotional support I have received from my family and friends. My great thanks goes to my parents, Mahin and Mo, who instilled in v me the love for poetry. I am grateful to Ali, Reza, Sara, Farhad, Raazi, Claudio, Manu, Erica, Lita, Nina, Amir, and Shahram for their continuous support and encouragement. me the love for poetry. I am grateful to Ali, Reza, Sara, Farhad, Raazi, Claudio, Manu, Erica, Lita, Nina, Amir, and Shahram for their continuous support and encouragement. My final thanks are saved for Dr. Acknowledgements Diego Bonelli, who generously supported me in every step of this path. Without his professional and emotional support I could not have been able to manage this project. vi Table of Abbreviations Titles of books frequently referred to in this thesis, especially Seamus Heaney’s volumes of poetry and prose. DN Death of a Naturalist DD Door into the Dark WO Wintering Out N North FW Field Work S Stations SI Station Island HL The Haw Lantern ST Seeing Things SL The Spirit Level EL Electric Light DC District and Circle HC Human Chain SS Stepping Stones OG Opened Ground: Poems 1966-1996 PO Preoccupations: Selected Prose 1968-1978 GT The Government of the Tongue: Selected Prose 1978-1987 RP The Redress of Poetry: Oxford Lectures FK Finders Keepers: Selected Prose 1971-2001 PW The Place of Writing CT The Cure at Troy SA Sweeney Astray vii vii viii Table of Contents Abstract ..................................................................................................................................... ii Acknowledgements .................................................................................................................. iv Table of Abbreviations ............................................................................................................. vi Stand Still You Can Hear: Introduction .................................................................................... 1 Summary of the Thesis .......................................................................................................... 1 From the Landscape I Was Born into: Seamus Heaney’s Biography ................................... 5 Feeling into Words: Review of the Works .......................................................................... 12 Summary of the Chapters .................................................................................................... 18 Chapter 1: Uncoding All Landscapes: Theoretical Framework and Methodology ................ 21 Introduction ......................................................................................................................... 21 1.1 Sounds in Poetry ............................................................................................................ 23 1.2 A Historical and Anthropological Perspective .............................................................. 30 1.3 Towards an Ecological Approach to Sounds and Listening ......................................... 38 1.4 Methodology and Theoretical Framework .................................................................... 43 1.4.1 An Ecological Approach to Listening: A Descriptive Framework ........................ 45 1.4.2 Acoustic Ecology and Soundscape Ecology: Terms and Classifications .............. 50 1.4.3 Wintering Out: A Telling Case Study .................................................................... 56 Chapter 2: On Seamus Heaney: The Critical Reception ......................................................... 59 Introduction ......................................................................................................................... 59 2.1 The Persuasive Voices Behind: Poets and Poetics ........................................................ 60 ix 2.2 The Republic of Conscience: Landscapes, Places and Spaces ...................................... 67 2.3 The Music of What Happens: Music, Sound, Voice and Silence ................................ 79 Conclusion ........................................................................................................................... 88 Chapter 3: Sounding, Soundings: ............................................................................................ 91 A Study of Sound in Heaney’s Wintering Out ........................................................................ 91 Introduction ......................................................................................................................... 91 3.1 Through the Ear of a Raindrop: Geophonic Sounds ..................................................... 95 3.2 My Serenades: Biophonic Sounds ............................................................................... 112 3.3 The Epic of the Noise: Anthrophonic Sounds ............................................................. 134 3.4 The Gallery of Tongue: vocal sounds ......................................................................... 167 Conclusion ......................................................................................................................... 189 Chapter 4: The Silence Listened For: Silence ...................................................................... 195 Introduction ....................................................................................................................... 195 4.1 A Contested Zone: Silence and Place ......................................................................... 198 4.2 Whatever You Say, Say Nothing: Silence and People ................................................ 207 4.3 I Might Tarry: Silence and Time ................................................................................. 229 4.4 The Vanished Music: Silence and Absence ................................................................ 235 Conclusion ......................................................................................................................... 240 Listen Now Again: Conclusions ........................................................................................... 246 Works Cited ........................................................................................................................... 261 Primary Print and Non-print Sources ................................................................................ 261 x Secondary Print Sources .................................................................................................... 264 Secondary Audio-visual Sources ...................................................................................... 284 Secondary Sitography ....................................................................................................... 285 1 1 1 The title of this chapter is inspired by Heaney’s ‘The Loaning’ (SI 51-52). 1 The title of this chapter is inspired by Heaney’s ‘The Loaning’ (SI 51-52). Summary of the Thesis This thesis is interested in listening as an important medium of experiencing, understanding, communicating and remembering using the poetry of Seamus Heaney as a particularly illuminating case study. I propose that poetic worlds have the potential to be impregnated with references to sounds and silence and that the identification and interpretation of such references can enhance our experience and understanding not only of the poems we read, but also the world experienced by their author. The Irish Nobel laureate Seamus Heaney (1939-2013) − arguably one of the most widely read, studied and translated poets of our time − was not one to neglect this very important aspect of life. Listening for Heaney was a way not only of perceiving the surrounding world but also of reaching out and staying in touch with the world beyond his sight and reach. Heaney’s poetry is his record of making sense of diverse local, national and international experiences. Sound – in all its manifestations, literal and metaphorical, including silence – is referenced throughout his poetry and prose. More specifically, this thesis aims to map and analyse references to sounds in Heaney’s third collection of poetry, Wintering Out (1972), in particular, and more broadly to Heaney’s other collections. This collection has been chosen because it was published at a transitional stage in the poet’s personal and professional life and because it presents a complex and nuanced engagement with his acoustic experiences. The collection is placed at the brink of his poetic and personal maturity. It came out a year after his academic experience at Berkeley and right before he resigned from the Queen’s University of Belfast to retire to a 2 cottage in Glanmore, Co. Wicklow, in order to dedicate his time and energy just to writing only. In 1972, the year of the publication of his third collection, Heaney witnessed the Bloody Sunday shootings in Co. Derry and Bloody Friday bombings in Belfast. The collection is born out of the poet’s everyday childhood memories, his concerns about identity, territory, language, religion and politics, as well as human history in a more global sense. It documents the poet’s standpoint in relation to the Troubles, his anxieties as a husband and a young parent, his hopes for the appreciation of the common ground, as well as his confidence about his vocation as a poet. Summary of the Thesis Wintering Out captures and echoes the poet’s thoughts and concerns through moments of sound and silence. Throughout the thesis I wish to answer the following research questions: What sounds have been evoked in the poems? What are the socio-political norms governing the presence and interpretation of sounds and what is the poet’s attitude towards them? How does Heaney use sounds – or their absence − to re-imagine places and his relationship with them? Historically, studies of sounds and audition have been informed by a concern with the understanding of music and the basic attributes of soundwaves − e.g. amplitude, frequency, timbre. Traditionally, the studies of sounds in poetry have also focused primarily on understanding the relationship between poetry and music, exploring the prosodic features of the language and other aural effects, such as rhyme, half-rhyme, consonance, assonance, alliteration. More recently, however, literary scholars have begun to treat literary texts as an archive of soundscapes − including everyday sounds, musical sounds, vocal sounds, noises and silence. These studies offer interdisciplinary approaches to the study of sounds and listening in literature and place literary texts at the intersection of science, sense-making, remembrance, and the surrounding environment. This auditory study of Heaney’s Wintering Out combines the linguistic and literary practice of close reading with ecological theories of auditory perception and soundscape 3 interpretation, as well as with historical and anthropological discussions on hearing, sound and silence. While close reading is an important element of the reading of individual poems, the theoretical framework applied in the thesis – based primarily on the ecological approach to auditory perception (Gaver 1992), acoustic ecology (Schafer 1977) and soundscape ecology (Pijanowski et. al. 2011) – offers an approach that has not been attempted before with regard to Heaney. This approach – while it is not intended to suggest a radical re- interpretation of Heaney’s work – nevertheless provides a number of new readings. In the first place, the descriptive framework offered by the ecological approach to aural perception provides a suitable lens for identifying auditory references in the poems and thus makes clear just how pervasive the references to sound and silence are in his poetry – both as subjects he writes about directly or indirectly and as materials with which he writes. Secondly, this approach offers a way in which that pervasiveness can be appreciated as part of a whole. Summary of the Thesis Thirdly, the taxonomy offered by acoustic ecology and soundscape ecology provides an effective lens for a more complete appreciation of Heaney’s poetic soundscapes – one that includes all aspects of sound, including its associate, silence – and can give us the resources for improving our understanding of his world and poetry. In summary, this study makes sound centre-stage and centre-poetry, in a holistic way, whereas other approaches get to sounds more peripherally: as one of the aspects to meaning and interpretation rather than as independent poetic agent that has shaped the poet’s life as well as his imagination. The identification and decoding of auditory references in Heaney’s poetry can be advantageous in many ways. To begin with, by exploring this sensory aspect of the poems, we become able to experience them more readily and thoroughly. As the sensory historian Peter Hoffer suggests, an auditory reading of poetry stimulates our sensory imagination to the fullest, encouraging us to “live” the poems (253). This experience, in turn, lifts the references to now-rare auditory properties of the past – e.g. the grating sound of the spade, the strike of 4 the blacksmith’s hammer, and the rumbling of a steam train, as well as the silence of bogs and bog-bodies in Heaney’s poetry – from the pages of the printed evidence and rescues them from being forgotten or treated as merely literary metaphors. Secondly and more importantly, studying the poet’s references to the auditory events and his stance towards them enriches our interpretation of the poems. This approach enables us to add to our understanding of the poems, the poet himself and his surrounding world. Heaney’s poetry and commentary enlighten us about the way contemporary listeners and the poet himself perceived their acoustic environment. Moreover, as I explain in the literature review of his work, Heaney’s concerns are also globally oriented. His poetic soundscapes are a testimony to the direct relationship between the social, cultural, and political dynamics, and the changes not only in the Irish soundscape but also in the world soundscape. To use Mark M. Smith’s words, an auditory approach to Heaney’s poetry would ultimately enable us to also “make sense” of the world we inhabit (847). Lastly, an auditory approach to interpreting poetry will challenge what is sometimes referred to as the hierarchy of the senses, placing confidence in the ear as a way of experiencing and meaning-making. 2 The title of this section is inspired by one of Heaney’s early essays, ‘Belfast’ (PO 37). Summary of the Thesis This approach is not only beneficial to literary critics and students of literature, but also to literary translators and writers, as well as to readers of poetry in general. To conclude, this study is based on the proposition that the experience of everyday sounds and listening is fundamentally a perceptual experience as well as the fact that listening, as a way of perceiving the world and remembering, occupies a central position for Heaney, the author of the poems, and for prospective readers, scholars and translators. It provides a different lens through which the poetry may be viewed. It aims to demonstrate that references to sound in poetry are not only guided by a feel for the sound of words but also by a strong sense of places and times they evoke, and thus can be politically, socially, culturally, and personally charged and meaningful. 5 From the Landscape I Was Born into: Seamus Heaney’s Biography2 Heaney has been known as the “laureate of Mossbawn” (Allen 173). The northern Irish countryside that often nourished or troubled his imagination is almost inseparable from his poetry. He argues “poetry’s existence as a form of art relates to our existence as citizens of society” (RP 1). But poetry, for him, was also “a journey into the wideness of the world” (OG 449). His poetic soundscapes open the space for a more universal sense of place than merely Northern Ireland. His poetic imagination was also profoundly affected by his extensive readings, translations, travels and life experiences. Heaney’s intellectual response to his acoustic experiences is reflected in his poetry, literary translation, critical prose and interviews as well as in the artistic and critical reception of his work. He writes of life in Northern Ireland, of a livelihood earned through manual labour and agriculture, of the way in which the political concerns of the Troubles were embedded in the very archaeology and topography of his place, and of his more global perspectives. The following pages provide an introduction to Heaney’s life, education and career. The purpose of this section is to offer an overview of the soundscape Heaney was exposed to during his early life and as he travelled. I also provide a review of his education and literary influences. I elaborate on these topics in the third chapter of the thesis. Seamus Justin Heaney was born on 13 April 1939 into a Catholic family and, until 1953, lived “a kind of den-life” in a traditional Irish thatched house on Mossbawn – the family farm near the village of Castledawson located in the mainly Protestant Co. Derry in Northern Ireland (OG 447). Heaney describes his home place as “a small, ordinary, nose-to- the-grindstoney place” and as “a subsistence-level life” (SS 8). The intimate and private household Heaney grew up in was “emotionally and intellectually proofed against” (OG 447) the ever-growing bustle of the outside world, but, instead, granted “the writer-in-waiting” (SS 6 35) the chance to grow closer to his “physical” and “creaturely” existence, and, as such, to grow as “susceptible” and “impressionable” to everyday sounds as the delicate but silent surface of water in a bucket can be to every movement of the earth beneath it (OG 447). In Mossbawn, Heaney was exposed to a range of sensory experiences. 3 I explain the terminology applied in this thesis in the first chapter. From the Landscape I Was Born into: Seamus Heaney’s Biography2 In the foreground of the soundscape, he could hear discrete geophonic and biophonic sounds:3 the howling of wind in young alder trees by the river bank or in beech trees on either side of the lane across the front garden, neighing of horses from the stable behind the bedroom, the sound of rain hitting the byre’s iron roof and on the dwelling house’s thatch, the pumping of an old hand-operated water pump immediately outside the back door, and the crackling of the turf fire inside the house. But he could also catch a trace of wood, zinc, fertiliser and old meal from inside the shed or recognise beast smells, manure smells, and a whiff of hay, straw and fodder in the farmyard (SS 3-10). The quiet ambience of the countryside also allowed the poet to hear far beyond the boundaries of home into the distance. He explains that if you “stand still” for a minute in the front yard of the house, every now and then, you may hear “the factory horn” of Clarke’s Mill or “a train shunting” at the station in Castledawson about “a hundred yards away at the end of the field behind the byre” (SS 8). One of the most heart- wrenching sounds he vividly remembers is the Tuesday morning pig cries coming from “a quarter of a mile away” in the Gribbin slaughterhouse in Anahorish, or the occasional “thud of explosion from a quarry somewhere or other on the horizon, maybe at Lavey, maybe farther away at Glenshane Pass” (SS 8). Therefore, while city life could have abbreviated this chance for “distant hearing” (Schafer 43), the hi-fi soundscape Heaney grew up in broadened his “auditory horizon” (Ihde 49) and added “perspective” − i.e. foreground and background − to his perception of the environment (Schafer 43). 7 The quietude of farm life amplifies the slightest disturbance in the regular soundscape, encouraging the listener to decode its message. In a hi-fi soundscape a so-called noise has the potential to be a signal communicating “vital or interesting information” about the sound-producing source (Schafer 43). From among the sound of cars, buses, lorries, vans, regular pedallers, occasional travellers, horses and carts frequenting the road in front of their Mossbawn house during the forties, Heaney recalls the spectacular “pelting” of a special cyclist, Master Pollock, for creating “a bit of a stir”. From the Landscape I Was Born into: Seamus Heaney’s Biography2 Another sound that strikes him as “special” is the “rat-a-tat of a pony and a cart”, which has been carved in his memory ever since for bringing “a kind of storybook glamour to the place” (SS 4-8). His attentive ears take in and reflect on such sounds, their sources and their implications. Next, what would render as noise to one, triggers the imagination of the poet-in-the making. Referring to Cushley’s driving manner, Heaney explains, “every other horse and cart lumbered and lock-stepped along, but Bob seemed to have some kind of Phaethon complex and to be always trying for lift-off” (SS 8). Mossbawn was only half a mile from the river Moyola and had a field that ran right to the riverbank. Therefore, Heaney familiarised himself with fishing and the sense experiences of the fishermen – “[t]he nibble on the worm, the tugs, the arc and strum of the line in the water, the moods of the water and the moods of the weather”, many of which reappear in his poetry (SS 94). The depth of those sensory experiences and memories is so profound that after decades the sixty-eight-year-old Heaney is able to relive his twelve-year-old feelings and senses. Heaney refers to the moments near the riverbank as his Wordsworthian “[s]pots of time”, as the memories that would be “enough for a lifetime of poems”: “Fleetness of water, stillness of air, and stealthiness of action” (SS 95). Later in 1954, when the family moves to The Wood, another family farmhouse just on the other side of the parish, Heaney leaves behind what he later identifies as “an ingredient of at-homeness” (SS 25). With the 8 new life on a more landlocked place and then the boarding school, habits of fishing at the riverside slip away. Instead, the silence of The Wood farm and its surrounding bogs and moss captures the poet’s ears to eventually find their voice in many of his poems.4 The centrality of home and the issue of rootedness in the Irish soil is known to be one of the major characteristics of Heaney’s poetry and prose, as well as the topic of much Heaney criticism. Although Heaney grounded much of his poetry and identity in Ireland, his poetics transcends the political borders of his country as well as of his time. 4 As in ‘Digging’ (DN 1), ‘The Other Side’ (WO 24) and ‘The Ash Plant’ (ST 19). From the Landscape I Was Born into: Seamus Heaney’s Biography2 Heaney’s cosmopolitan disposition, in part, originates from his education, career, travels and extensive reading of poetry of the Irish and non-Irish poets and critics. It was during his school days at Anahorish School (1945-51) that he learnt to read. Later, as a boarder at St Columb’s College (1951-57) he was exposed to Gaelic literature of Ireland, Latin and English literature (SS 19). He recalls the first day at St Columb’s College as a touchstone of his life marking his separation from the known world of the home. The Catholic education had structured his way of thinking and feeling in such a way as to see the outside reality as “a triad of danger that involved ‘the world, the flesh and the devil’” (SS 39). Instead, going to Queen’s University (1957-61), marked the entry to the world putting “life” in touch with “literature” (SS 39). Heaney completed an Honours degree in English Language and Literature at Queen’s, where he published his first poems in student magazines and where he continued his education as a postgraduate student. He began his teaching career at St Thomas’s Intermediate School, where he was introduced to the poetry of Patrick Kavanagh, who became one of his major literary influences. In 1966, the year of the publication of his first full-length collection of poems, Death of a Naturalist, he was appointed as lecturer in Modern English Literature at Queen’s. Yet, after returning from his sabbatical year (1970-71) and experiencing the “freer” and “opener” 9 atmosphere at Berkeley, Heaney felt the need for a move from the restricting taboos and reticence of life at Queen’s. He commented: “You narrowed and tightened yourself a bit in the Queen’s […] I didn’t realise just how much this was the case until I came back from my year in Berkeley” (SS 103). Heaney resigned from Queen’s to begin his work as a full-time writer in Co. Wicklow, in the Republic of Ireland (SS 162-163). Later in 1975, the year of the publication of North, Heaney took up a teaching position at Carysfort College of Education in Dublin, where he settled permanently (SS XXVI). Heaney left the sectarian violence of Northern Ireland to safeguard the independence of his writing, culminating in the individual triumph of the Nobel Prize in Literature. Heaney’s life is full of incidents that reveal his sensitivity towards known and unknown soundscapes. From the Landscape I Was Born into: Seamus Heaney’s Biography2 The sabbatical year in Berkeley was the first time Heaney lived outside Northern Ireland. For him the main distinction between Queen’s and Berkeley was the liberal and vivid environment of the university resonating in “Californian students’ chorus of ‘Hi-i-s’ and ‘Wow-w-w-s’ and ‘He-e-eys!’” (SS 138). But “[t]he empty amphitheatre / Of the west” (WO 67) only directed the poet back to his Irish memory banks. Conversely, in his 1973 visit to Denmark, it is the silence of Tollund man and the bogs that strikes familiar and grants him a new point in history from which he can look at the soundscape of violence in Ireland. As early as 1944, he experienced the acoustics of war and the presence of American forces near his home. The death of his younger brother, Christopher, in a road accident in 1953, left an indelible impact on his psyche. Life in Northern Ireland inevitably introduced him to the violence of the Troubles that erupted in the late 1960s. Heaney settled permanently in Dublin in 1976 and resumed his teaching career three years later at Harvard University, where he organised poetry workshops. He divided his time between Dublin and America and continued to publish his poetry, prose and translations. He was appointed as Professor of Poetry at Oxford (1989-94) and awarded the Nobel Prize for 10 Literature in 1995. He continued to write poetry for almost two decades until his death in 2013, in Dublin. Heaney’s intellectual and literary world also expanded through his wide-ranging readings, ranging from ancient Greece and Rome to twentieth-century Eastern Europe, continental Europe and to medieval England and Ireland. These intellectual experiences expanded his knowledge of literary genres and canons. Heaney believed rhymes learnt early in life and their strong imaginative stimuli can “stay with you for a lifetime” and “end up being sounding lines out to the world and into yourself” (SS 35). As such, one of the very first literary influences on him came from his mother rhyming off Latin prefixes and suffixes and reciting verses of Longfellow, Keats, and anonymous nursery rhymes, which “animated and instigated” something in the depth of his spirit (SS 35). The “down-to-earthness” of her attitude too shaped much of his literary taste. Years later, Frost’s “blunt” treatment of the tragic incident in ‘Out, Out’ (1916) fascinated him, providing an important input for the writing of ‘Mid-Term Break’. p p g , g y he desired, Heaney wrote under the pseudonym Incertus (SS 37). 5 An example of such poems is ‘October Thought’. For a while, before he gained the “writerly self-con An example of such poems is October Thought . For a while, before he gained the writerly self-confidence he desired, Heaney wrote under the pseudonym Incertus (SS 37). 5 An example of such poems is ‘October Thought’. For a while, before he ga he desired, Heaney wrote under the pseudonym Incertus (SS 37). 5 An example of such poems is ‘October Thought’. For a while, before he gained the “writerly self-confidence” he desired, Heaney wrote under the pseudonym Incertus (SS 37). 5 An example of such poems is ‘October Thought’. For a while, before he gained the “writerly self-confidence” p p g , he desired, Heaney wrote under the pseudonym Incertus (SS 37). From the Landscape I Was Born into: Seamus Heaney’s Biography2 Through them, Heaney realised reality and actuality can be far more intriguing and unforgettable and, in his own writing, he remains devoted to “what happened” (SS 36). Heaney’s interest in “the concrete” and “the condensed” was reinforced and guided by the poetry of Hopkins and the prose of Hemingway. He was especially attracted to the evocation of regional speech in Hopkins’ alliterating music and found a connection between his “heavily accented consonantal noise” and the Northern Irish accent (SS 40). His early Incertus poems were written in conscious homage to Hopkins and reflected Hemingway’s “clean” and “quick diction” (SS 38).5 In fact, Heaney owes much of his popularity to his ability to evoke a sense of “immediacy” and “intimacy” in his readers (Crotty 44-45). In one 11 of his most quoted essays, ‘Feeling into Words’, he quotes his early poem ‘Line to myself’ as a reminder: of his most quoted essays, ‘Feeling into Words’, he quotes his early poem ‘Line to myself’ as a reminder: Without forcing, without violence. Whose music is strong and clear and good Like a saw zooming in seasoned wood. You should attempt concrete compression half guessing, half expression. (PO 46) Without forcing, without violence. Whose music is strong and clear and good Like a saw zooming in seasoned wood. You should attempt concrete compression half guessing, half expression. (PO 46) During his years at Queen’s, Heaney also developed an interest in Old English poetry. Through his readings of ‘The Wanderer’, ‘The Seafarer’ and ‘The Battle of Maldon’, the diction of Anglo-Saxon poetry established some sort of a register for him (SS 40). The degree in English Literature also grounded him in iambic pentameter, free verse and Hopkins’ sprung rhythm. In ‘The Bookcase’, he refers to MacDiarmid, Bishop, Macmillan, Hardy, Stevens, Caedmon and Thomas as his early English-speaking influences (EL 51-52). Later, his readings of Lawrence’s and Eliot’s poetry and literary criticism further establish in him the orthodox idea that “the age demanded a roughening up of the utterance, an avoidance of smooth numbers – you were meant to hit the stride of living speech” (SS 40). Heaney situated himself in the lyric tradition of Wordsworth and referred to Yeats and Kavanagh as his major Irish influences. p y y g ( ) 7 The poetry collections: Death of a Naturalist (1966), Door into the Dark (1969), Wintering Out (1972), North (1975), Stations (1975), Field Work (1979), Station Island (1984), The Haw Lantern (1987), Seeing Things (1991), The Spirit Level (1996), Electric light (2001), District and Circle (2006) and Human Chain (2010). The selected editions: Selected Poems 1965–1975 (1980), New Selected Poems 1966–1987 (1990), Opened Ground: Poems 1966–1996 (1998), and the posthumously published New Selected Poems 1988–2013 (2014), 100 Poems (2018), and a spoken-word recording of Heaney reading his poems produced to mark his 70th birthday. 6 The title of this section is inspired by Heaney’s article ‘Feelings into Word’ (PO 41). From the Landscape I Was Born into: Seamus Heaney’s Biography2 He was particularly fascinated by “the spoken force” in Kavanagh’s Great Hunger (1944) and Hughes’s ‘View of a Pig’ (1959), influences of which can be traced in his early poems (Crotty 44-45). Despite the limited access to resources of Gaelic literature in libraries during the 1950s and 60s, he maintained a strong sense of preserving his Irish heritage and used every opportunity to look for books about Irish poetry, among which are Farren’s The Course of Irish Verse (1947). Heaney attempted to apply and adjust his knowledge of English language and literature with the cultural and political pieties he grew up with so that his poetry captures his “whole” experience of living and reading: Certainly the secret of being a poet, Irish or otherwise, lies in the summoning of the energies of words. But my quest for definition, while it may lead backward, is conducted in the living speech of the landscape I was born into. If you like, I began as a poet when my roots were crossed with my reading. I think of the personal and Irish 12 pieties as vowels, and the literary awarenesses nourished on English as consonants. My hope is that the poems will be vocables adequate to my whole experience. (PO 36-37) 8 The translations include: Sweeney Astray (1983), The Cure at Troy (1990), The Midnight Verdict (1993), Laments (1995), Beowulf (1999), Diary of One Who Vanished (1999), Arion (2002), The Burial at Thebes (2004), The Testament of Cresseid & Seven Fables (2009), and two posthumously published works: The Last Walk (2013) and Aeneid Book VI (2016). 9 The essays are collected in three volumes: Preoccupations (1980), The Government of the Tongue (1988), and The Redress of Poetry (1995), as well as the selected essays Finders, Keepers (2002), Stepping Stones (2008) and the more recent official bibliography. 10 The Incertus poems appeared in the student magazines of Q and Gorgon at Queen’s in 1959 (SS XXIII). His first article ‘Shall We Jive This Jig’ was published in the Irish Digest, in 1961 (XXIV). p y y g ollections: Death of a Naturalist (1966), Door into th Feeling into Words: Review of the Works6 One only needs to read the opening of Heaney’s Nobel Lecture to discover that hearing is a way of “feeling”, “knowing” and remembering the world around him. The poet recalls how his childhood environment allowed him to cultivate and practise his aural perception and imagination freely. As a child, he attuned his ears to and “took in” the sound of everything that was going on around, from neighing of horses from beyond the walls, scampering of mice on old ceilings, howling of wind in beech trees, rumbling of the train, and to the English accent of a BBC newsreader and familiar voices of adults in the kitchen, many of which appear in his poetry, prose and literary translations (OG 447). During his life, Heaney published thirteen poetry collections,7 several translations8 and critical essays.9 His first poems were published in student magazines during his undergraduate years.10 There is a thematic thread connecting his poetry, prose and translation with their real-life context embodied in his references to sound, silence and listening. In the following pages, I shall provide a brief chronological overview of his poetry and highlight the connections outlined above giving examples from acoustic references. 13 From the publication of his first poetry collection, Heaney established his reputation as a major new talent of brilliant linguistic “fidelity”, “precision” and “evocativeness” (O’Donoghue 2). Death of a Naturalist reflects Heaney’s skilful application of onomatopoeic language, vivid descriptions and metonymies to re-create rural images from his childhood landscape. It displays multiple references to sounds and silences the poet experienced early in life: his father’s familiar silence and the “clean rasping sound” of his spade (DN 1), “the squelch and slap” in soggy peats (DN 2), the “heavy rain” of August (DN 8), the “rhythms” his mother sets up on churning days, her silence and the “plash and gargle” of the milk (DN 9-10), the wordless diviner (DN 23), the “helter-skelter” of the waterfall (DN 27), and the echo of the poet’s “own call” in the well (DN 44). The publication of his second poetry collection enhanced Heaney’s poetic reputation and confirmed the revelatory feature of his descriptions. In Door into the Dark, he deliberately turns his back on the brightness of the familiar to open a door into the dark corners of the earth and the “buried life” of his inner self (Corcoran 54). Feeling into Words: Review of the Works6 Therefore, the guttural grunts of Old Kelly, the shunting of the steam engine (DD 6), the “short-pitched ring” of the hammer in the anvil or the hissing sound of a hot shoe toughening in water, the blacksmith’s serene pose (DD 9), the voices of the fishermen at Lough Neagh (DD 28), “the hush and the mush / Of its whispering treadmill” or “the hum of the traffic” (DD 38-39), and the kind, deep and noiseless presence of “the waterlogged trunks” (DD 44) arise from the most hidden corners of his memory. In contrast to the private setting of the previous collections, Wintering Out is set in a wider public context. Hereafter, Heaney’s characteristic strengths − precision and evocation of description − are accompanied by more political consciousness (O’Donoghue 3). Heaney gestures towards the political context from which his inner distress resulted. In ‘Fodder’, ‘Anahorish’, ‘Toome’ and ‘Broagh’ and ‘Traditions’, he demonstrates his concerns for Irish 14 language and culture, while in ‘Land’, ‘Bog Oak’, ‘Gifts of Rain’, ‘Oracle’, ‘Carin-maker’, he pitches our ears to the keynote sounds of his places, and in ‘Navvy’, ‘The Servant Boy’, ‘The Backward Look’, ‘Mother of the Groom’, ‘The Shore Woman’, he addresses various types of silences. These complexities of the socio-political context and the inner world of the poet are also embodied through the juxtaposition of various places and time periods and through the transposition of their soundscapes to the landscape of his poems. Published about three years after Heaney had left Ulster for Wicklow, North became at once his most admired and controversial book of poems (O’Donoghue 4). A sense of urgency to respond to the crisis in Northern Ireland dominates the entire collection and its title (Corcoran 71). Heaney himself does not deny the role of the socio-political context in shaping his sub-consciousness but refuses any direct interaction (SS 66). His dilemma is readily evident in the “space” between “the tick of two clocks” (N x). Feeling into Words: Review of the Works6 He attempts to capture the violence and its influence on individuals through various references to sounds and silence, from the heavy silence hovering over coffins, and “customary rhythms” of the funeral ceremony (N 7), to “ocean-deafened voices” (N 10), Jimmy Farrell’s monologue about the various “skulls they have / in the city of Dublin” (N 15), the silence and indifference of the man to unjust punishments and the “gargling” of blood (N 38), and to Heaney’s own silence and doubt about “void”, “death” and “eternal” rest in peace (N 18). Compared to his previous collection, Field Work demonstrates a renewed trust in artfulness. The volume warmly weaves varying styles and subjects. Elegies for the violence in Northern Ireland are juxtaposed with pastoral poems, love lyrics and translations. Early in the volume there are references to the movement of “tongue” and the willingness to enquire about the future. Heaney encourages his fellow country people to keep their “ear” to the ground for a long time in order to pick up all the “comfortless noises” (FW 5). He recalls the pumping of the “omphalos” (FW 7), “the bucket’s clatter” and its “slow diminuendo as it 15 filled” (FW 8). He speaks of “old decency”, the “[l]eading” voice at the pulpit and its “melody” (FW 12). He remembers the “thick and comforting” voices of his guttural muse (FW 22) and looks for “echo”, “music”, “song”, “singing” and reminds us “[w]e still believe in what we hear” (FW 20-21). If up to North the poems grow together, scrutinising the self and the cultural context that nourished it, Field Work demonstrates a desire to look outward. He says: “I no longer wanted a door into the dark, I wanted a door into the light” (Corcoran 127-128).11 Heaney’s poetic career takes a distinctively different course from 1980. With the hunger strikes, Northern Ireland experienced one of the most difficult periods of the Troubles. A strong sense of social awareness and poetic vocation is reflected through his first published translation, Sweeney Astray and its twin volume, Station Island. In a 1979 interview with James Randall, Heaney said: I very deliberately set out to lengthen the line again because the narrow line was becoming habit […] I wanted to turn out, to go out, and I wanted to pitch the voice out […] a return to an opener voice and to a more – I don’t want to say public – but a more social voice. (Heaney 182) y p ( y ) 12 In an interview with Dennis O’Driscoll in the Irish periodical Hibernia in 1983, Heaney said that he immediately felt “there was something here for me” (8). 11 In a 1979 interview with James Randall, Heaney said: a 1979 interview with James Randall, Heaney said: 13 His contemporary critical prose, The Government of the Tongue, reflects a similar, but politically concentrated attitude. The book “offered a sustained exploration of the rights and obligations of the writer, whether in the East or the West” (O'Donoghue 10). 13 His contemporary critical prose, The Government of the Tongue, reflects a similar, but politically con East or the West” (O'Donoghue 10). Feeling into Words: Review of the Works6 Heaney’s continued awareness of the conflict between the pull of the free imagination and the claims of social responsibility is one of the things that prompted a special interest in the story of Sweeney, who appears in both volumes.12 The character of Sweeney provides the poet with an epic framework and a liberating force for his own psychic material (Corcoran 108; Andrews 146). In Station Island, the poet-narrator encounters literary figures and people from his past in Lough Derg, Co. Donegal. He writes of the dying “echoes” (SI 13), the haunting sounds of “‘convicts’ chains” (SI 18), the “creaks and the click / of stones” (SI 23), the loud neighing of the wind and sea (SI 30). There are also references to the silence during writing when you can imagine all the sounds (SI 24), silence of “the empty city” (SI 37), and finally, the silence that enhances hearing: “Stand Still. You can hear / everything going on” (SI 52). 16 His shortest poetry collection, The Haw Lantern, was published three years after Heaney enjoyed a wider literary community in Harvard. Compared to the previous poetry collections, the volume reflects a more deliberate sense of bringing the literary and the public imperatives together.13 The poems are in the form of parables and often transcend time and place. The main concern of the book is to display, as Heaney puts it, “loss of faith − or rather loss of faiths, of all kinds” (SS 287). Among other references to voice and listening, the book refers to absence in a great many ways: the “noiseless” landscape of “the republic of conscience” (HL 14), his father’s “speechlessness” (HL 17), the “disappeared people” and “unspoken assumptions” (HL 18), and “loss” and “death” (HL 20). Heaney’s next translation, The Cure at Troy, is the verse adaptation of Sophocles’ tragedy Philoctetes about the Trojan War whose narrative attracted Heaney’s attention for the resonance with the protracted sectarian conflicts in Northern Ireland, as well as with the end of the Apartheid regime in South Africa. In the choral prologue of the play, Heaney expresses his opinion about the key role of poetry as “the voice of reality and justice” (Corcoran 700); thus, his well-known lines: History says, Don’t hope On this side of the grave But then, once in a lifetime The longed-for tidal wave Of justice can rise up And hope and history rhyme. Feeling into Words: Review of the Works6 (CT 2) Of justice can rise up In Seeing Things, Heaney draws inspiration from the visions of the afterlife in Virgil and Dante, bringing together persons and objects from a visionary realm. The poet’s declaration of artistic metamorphosis can be found in ‘Fosterling’ where he admits to having waited until he was “nearly fifty / To credit marvels” (ST 50). The references to sounds and voice are scattered in the volume: the chanting of “equivocal words” and their “echo” in the 17 cave (ST 1), the plucking of the “Thracian lyre” and the peach of the “prophetess” (ST 2), the call of “fate” (ST 3), and the “whispering through the air, saying hush / And lush [emphasis in the original]” (ST 13). Compared to the previous collections, The Spirit Level reflects more freedom for expressing pessimism and resentment towards politics. This tone and attitude are also present in Electric Light and District and Circle. Heaney opens The Spirit Level by inviting us to “listen for” the music in the “downpour”, “spillage” and “flowing” of rain water (SL 1), and to beware of the “Chooka-chook” of the sofa leg or the Yippee-i-ay in the age of “wireless” and “ignorance” [emphasis in the original] (SL 7-8), and “polluted” words flowing with “the dirt / Of blurbs and the front pages” (SL 38), and, instead, to take nourishment from “what the birds and the grass and the stones drink” (SL 38). District and Circle encapsulates Heaney’s powerful descriptive techniques and astonishing themes. It is possible to relate the poems to such political events as the period of the Troubles in the early 1970s, the ceasefire in Northern Ireland in 1994, and the New York attacks of 9/11. In this collection, the matter of the public wins over the matter of the private and tragedy keeps haunting the poet − a tension that is central to The Burial at Thebes. Much of the collection takes place in the ‘Underworld’ or afterlife, as well as by recalling visionary and dead figures from the ancient Classic or the Northern Past (O’Donoghue 13-15). This poignancy increases as Heaney reflects on the gloom of the world in the early twenty-first century, as reinforced by the rudderless dominancy of the imperial West. Feeling into Words: Review of the Works6 In his last collection, Human Chain, he celebrates the beginning of life – the birth of the first two granddaughters, but the memories of the past and ghosts of the dead keep haunting the elegies and translations. In the opening poem, he reminds us of the transient existence of sounds and that if we are not “awake” and alert we might miss them and their 18 message (HC 3) and then explains that sounds are more “[t]han any allegory” (HC 33), they delight and inform the ear and the spirit. message (HC 3) and then explains that sounds are more “[t]han any allegory” (HC 33), they delight and inform the ear and the spirit. Summary of the Chapters Following this introduction, the first chapter describes how I use primary and secondary sources in order to answer my research questions. This chapter is divided into four sections. I first provide a brief overview of the approaches to sound in poetry. I then offer a historical overview showing the persistent role of sound and listening in human experiences. In the third section, I review the ecological approaches to auditory perception and soundscape interpretation as the main theoretical frameworks applied in the analysis of the poems. The final section provides details about the research methodology, the conceptual frameworks and the scope of this thesis. In this section, I define the key terms I refer to and use throughout the thesis and provide a more detailed rationale for my choice of Wintering Out as the main focus of my analysis. Heaney’s talent and popularity has generated a large volume of critical responses from literary scholars. The second chapter of the thesis provides a review of the relevant body of academic literature on Heaney’s work starting with studies on the literary influences − from the classics to modern European literatures − on the poet’s conceptualisation of place and his relationship with it in his poetry. The second section is dedicated to the studies that examine Heaney’s position in relation to various places and landscapes. This section prepares the ground for the analytical chapters of the thesis in which I examine sound and silence as the mediator between the poet and the places he re-creates in his poems. In the third and last section of this chapter, I explore the studies that examine the concepts of listening, music, sounds, voice and silences in Heaney’s work. 19 In the third chapter of the thesis, I examine the concept of sound in Heaney’s work in great detail. This chapter is my main analytical chapter. This chapter is divided into the four categories of geophonic sounds, biophonic sounds, anthrophonic sounds and human vocal sounds and aims at identifying and interpreting such references in Wintering Out. In the first section, I focus on Heaney’s representation of non-sentient ambient sounds of the natural phenomena. Summary of the Chapters I argue that the manifestation of the sounds of natural phenomena has, on the one hand, allowed the poet to evoke the underlying political, cultural, social and historical associations, and on the other hand, it has enabled him to evolve and enrich the world of his poetry. In the second section of this chapter, I examine the references to the sounds produced by non-human living organisms. I first identify Heaney’s manifestation of the human/non- human animal and then examine how Heaney makes effective use of animals and their sounds to tell us about himself and the world around him. In the third section of this chapter, I focus on the representation of the sounds produced by human-made objects − e.g. tools, crafts and vehicles. My purpose is to identify and interpret the poet’s attitude towards such sounds and how he uses them to tell us not only about his own society but also about human civilisation in general. In the final section of this chapter, I focus on the representation of human vocal sounds. Through this analysis, I stress the importance of voice as not only the medium but also the material and topic of his poetry. In the fourth and last chapter of the thesis, I examine silence in Wintering Out. This chapter is divided into four sections: silent places, silent people, silence as a pause and silence as absence. After providing a definition of a silent place, in the first section, I examine the way in which silence can be perceived differently in various places. The purpose of this section is to show how the structural or topographical features of a place can influence the perception of silence in that place and how Heaney uses these differences to develop his themes. In the second section, I focus on Heaney’s representation of silence among people. I 20 show how identifying this type of silence can uncode the social, cultural, political and personal implications involved. The third section is based on the relationship between silence and time. I trace and interpret the instances in which the poet represents silence as a short pause or stoppage as well as what his purposes are. The chapter ends by pointing out silence as indicative of absence and nothingness. At the same time, it argues that, even as absence, silence in Heaney’s poetry is filled by various hints and connotations. circumstances of Ireland, saying: Chronologically, I grew up in the 20th century that had roots really in the medieval. I was in countryside where we still ploughed with horses, lit the fire in the morning, you know, carried water from wells and so on, so in a very quick time all that changed in common life in Northern Ireland and in my own personal life. So what the poetry is trying to do is to make sense of our life in that time and in that time included, of course, not only my anthropological experience, but included the political history of Northern Ireland. (2:38 – 3:25) 14 In a 2010 interview with Tiago Moura, Heaney refers to the inseparability of his poetry and the political circumstances of Ireland saying: y ( ) www.youtube.com/watch?v=s7sskc1pi_k&fbclid=IwAR1iQEVlkXvYvzSNidaASXoD3KLcOHqjrRzLSaJGFF o08JdzLPj7vupc39k. 14 In a 2010 interview with Tiago Moura, Heaney refers to the inseparability of his poetry and the political y ( ) utube.com/watch?v=s7sskc1pi_k&fbclid=IwAR1iQEVlkXvYvzSNidaASXoD3KLcOHqjrRzLSaJGFF g circumstances of Ireland, saying: y 17 See brickmag.com/an-interview-with-seamus-heaney/. p p y y ( ) 16 The quotation is taken from Eleanor Wachtel’s interview with Heaney. See brickmag.com/an-interview-with- seamus-heaney/. Summary of the Chapters Just like his contemporary Northern Irish poets, much of Heaney’s poetry hinges on the predicaments of colonisation and conflict on the soundscape of his homeland (Obert 2- 3).14 The troubled North constantly posed unique challenges to Heaney’s deep-seated senses of rootedness, place-consciousness and belonging. Heaney was not a silent bystander of the Troubles. For him there was “[n]o such thing / as innocent / bystanding” (SL 30). As this thesis shows, Heaney’s auditory engagement with the politics of his country is represented through his attention to the transformation of the local soundscapes, to the loss of local keynotes and soundmarks, and to their replacement with the acoustics of violence, conflict and sectarianism. Throughout this thesis, I benefit from tools, observations and insights drawn from sound studies to identify and interpret the ways in which Heaney gives rise to his not only political, but also social, cultural, environmental and personal concerns about his surroundings through particular moments of sound and silence. The contextual information provided in the analytical chapters is intended to link real and poetic soundscapes. By adopting an interdisciplinary approach, I aim at adding an original contribution to academic debates on Heaney’s work in particular and to literary studies in general. 21 Chapter 1: y 17 See brickmag.com/an-interview-with-seamus-heaney/. 15 The title of this chapter is inspired by a line from Heaney’s ‘The Peninsula’ (OG 21). y 17 See brickmag.com/an-interview-with-seamus-heaney/. e title of this chapter is inspired by a line from Heaney’s ‘The Peninsula’ (OG 21). e quotation is taken from Eleanor Wachtel’s interview with Heaney. See brickmag.com/an-interview-with Introduction For me it [the physical environment that nurtured my childhood] was all-important. When I think back, it’s sensation, really, rather than intellection that returns to me. A feel for places. I mean, the body stores so much. […] I still remember that, but I think that’s not uncommon, is it? What is stored bodily is very important for memory, and I think that other bodily sensations later on can bring it all back.16 The centrality of sensory and bodily experiences as a way of remembering and making sense of the surrounding world underpins Seamus Heaney’s poetic imagination. In a flashback, he notes that “my earliest memory is of my foot touching the ground of Mossbawn, […]” and then continues “and I can still feel my little foot inside my old foot here”.17 Elsewhere, he points out how “[t]he amount of sensory material stored up or stored down in the brain’s and the body’s systems is inestimable. It’s like a culture at the bottom of a jar” and then accents the active effort it takes to locate and dig out the underlying life force in order to enrich his mind and imagination: “it doesn’t grow, I think, or help anything else to grow unless you find a way to reach it and touch it. But once you do, it’s like putting your hand into a nest and finding something beginning to hatch out in your head” (SS 58). But for Heaney, the centrality of sensory experiences, in general, and auditory experience, in particular, is not limited to his relationship with his birthplace only. In Crediting Poetry, his 1995 Nobel lecture, Heaney describes listening as a salient way of not only perceiving the world around him but also undertaking “a journey into the wideness of the world beyond” (OG 449). 22 As suggested in the introduction to the thesis, Heaney’s poetry is a repository of his acoustic experiences and memories. Making use of the ecological approach to auditory perception, and the analytical lens of acoustic ecology and soundscape ecology, I attempt to identify and interpret references to sounds – and their absence – in Heaney’s third poetry collection, Wintering Out (1972). The purpose of this study is to show that such acoustic references in Heaney’s poems are guided by the time and place they evoke and thus can be personally, socially, culturally and politically charged and meaningful. 18 See Clarke’s Ways of Listening (2005) for a schematic representation of approaches to listening (13-15). Introduction Questions relating to aural perception, sound, music, voice and silence, as well as their actual and metaphorical implications have been considered at great length by a variety of thinkers and scholars. Their approaches vary in the definition of sound and in their methods for measuring and analysing it.18 From acoustics and ecological acoustics we learn about the physical properties of sound-producing sources and the relationship between human behaviour and the changes in the acoustic environment throughout time. Psychoacoustic researchers indicate the way the human brain interprets the basic attributes of soundwaves, such as frequency, amplitude, phase, duration, pitch, timbre and rhythm. Scholars using a cognitive approach are interested in the mental associations of the form, tonality, rhythm and melodic organisation of the sounds experienced by the listener. Humanities and social sciences explore the social, cultural, political, and aesthetic associations of the sounds, music, and silence. And, finally, literature and art reflect man’s perception, interpretation and imagination about his auditory experiences. This thesis does not engage with psychological and cognitive approaches that investigate the attributes of the soundwaves and their mental associations; nor is it concerned with the applied sciences of acoustics, such as architectural acoustics, noise abatement practices and electroacoustics. Likewise, in this thesis, I do not engage with the physiology of 23 hearing, otology or neurolinguistics and, while some of the discussions might touch on phenomenological questions about auditory perception, I am not concerned with metaphysical and epistemological matters. The theoretical framework applied in this thesis relies on an ecological approach to sound and listening, which reaches beyond the traditional sensation-based theories of elemental stimuli – e.g. frequency, amplitude, phase and duration – and focuses on the physical properties of sound-producing sources and their surrounding environment.19 It also incorporates studies in humanities and social sciences to explore the social, cultural, political, or aesthetic associations of the sounds, music, and silence illustrated in Heaney’s poetry. In order to establish the groundwork for this analysis, in this chapter I (I) provide a brief overview of traditional approaches to the study of sound and listening in poetry; (II) provide a selective review of the literature relevant to the cross-disciplinary field of hearing history within the broad study of sensory history; (III) introduce the theoretical basis of my study – the ecological approach to auditory perception and soundscape interpretation; and (IV) explain the terminology associated with that approach. 19 “Ecology is the study of the relationship between living organisms and their environment. Acoustic ecology is therefore the study of sounds in relationship to life and society” (Schafer 205). 19 “Ecology is the study of the relationship between living organisms and their environment. Acoustic e gy y p g g therefore the study of sounds in relationship to life and society” (Schafer 205). 19 “Ecology is the study of the relationship between living organisms and their environment. Acoustic ecology is th f th t d f d i l ti hi t lif d i t ” (S h f 205) 1.1 Sounds in Poetry Although traditional literary criticism has made seminal contributions to the study of sound in poetry, the importance of sound and listening remains one of the more challenging aspects for critical analysis. As this review suggests, whereas the majority of traditional approaches deal with the major elements that contribute to the aural effect of the poetic language, more recently, literary scholars offer interdisciplinary approaches to the study of sound – whether understood as noise, music, or voice – and treat literary texts as an archive of soundscapes. 24 In The Sound Shape of Language (1979/2002), a study focused on uncovering the function and structure of sound in language and its relation to meaning, the linguists Roman Jakobson and Linda R. Waugh maintain that “nowhere is the direct interplay of sound and meaning more salient than in poetry” (4-5): Poetry, whether written or oral, whether the production of experienced professionals or of children, or whether oriented towards or against ordinary language, displays its own peculiar sound shape and grammatical structuration. […] The sounds of poetry indispensably carry a distinctively more autonomous task, and their bonds with poetic semantics are not reducible to the ordinary role required for them within these conventional units by the humdrum use of language. In poetry speech sounds spontaneously and immediately display their proper semantic function. (225) Poetry, whether written or oral, whether the production of experienced professionals or of children, or whether oriented towards or against ordinary language, displays its own peculiar sound shape and grammatical structuration. […] The sounds of poetry indispensably carry a distinctively more autonomous task, and their bonds with poetic semantics are not reducible to the ordinary role required for them within these conventional units by the humdrum use of language. In poetry speech sounds spontaneously and immediately display their proper semantic function. (225) Jakobson and Waugh’s study sheds light on the way sounds function differently in ordinary and poetic language, and, although it does not offer specific ways for analysing sound in poetry, it provides a linguistic basis for the long-standing attention literary critics have historically paid to the analysis of sound. Eagleton, however, critiques the tendency to treat a poem as merely ‘language’ and to reduce it to its constituting devices. 1.1 Sounds in Poetry In How to Read a Poem (2007), he maintains that poetry should, instead, be approached holistically and read as a “discourse” by attending to language in all of its material density, arguing that although there is a “conceptual distinction” between poetic form – i.e. tone, pitch, rhythm, diction, voice, syntax, register, point of view and punctuation – and its content – i.e. idea, moral vision, argument, the two are “inseparable in experience” (2-65). Jakobson and Waugh’s study sheds light on the way sounds function differently in ordinary Jakobson and Waugh’s study sheds light on the way sounds function differently in ordinary There exists a considerable body of literature on the formal analysis of the aural features of poetic language that precedes and follows Eagleton’s caveat. George Saintsbury’s A History of English Prosody (1906) is a classic example that provides a comprehensive - albeit now dated and controversial - historical account of English prosody from the twelfth to the end of the nineteenth century, including an overview of the Anglo-Saxon accentual verse and the impact of the Latin, Celtic, Scandinavian, as well as Italian and French traditions in the 14th century with the introduction of a stricter alternation of stressed and unstressed 25 syllables, which dominated English poetry until the modern period. Saintsbury takes as his starting point the notion of “long” and “short” syllables and defines prosody as “the laws and variations observable in the rhythmical and metrical grouping of sets of the two […] (5). It has been a century since Saintsbury produced his three-volume book and in that period, there have been many changes in the ways poets versify and even more in the methods critics use to analyse poetry. Saintsbury’s theory of a strictly temporal metre is not seen as applicable to English, which functions instead by an alternation of stressed and unstressed syllables. His work nevertheless provides evidence of a long-standing interest in the aural features of poetry and an attempt to understand and account for those features. In Poetic Meter and Poetic Form (1965), Paul Fussell highlights metre as “the most fundamental technique of order available to the poet” while other poetic techniques of order – i.e. rhyme, line division, stanzaic form, and structure – are all projections of the kind of formalizing repetition that metre embodies (4-5). 1.1 Sounds in Poetry His definition of metre evinces its importance and complexity: “Meter is what results when the natural rhythmical movements of colloquial speech are heightened, organized, and regulated so that pattern – which means repetition – emerges from the relative phonetic haphazard of ordinary utterance” (4-5). According to him, the impulse toward the metrical organisation of assertions stems from the more primal human impulse towards order, one that was originally used as a mere mnemonic device to help individuals remember epics, prescriptions, legal codes and recipes, before the days of the printed book (17-30). However, he stresses that the importance of metre goes beyond an aid to the memory and states “[t]he empirical study of poetry [reveals] meter is a prime physical and emotional constituent of poetic meaning” (3). He points out that metre echoes rhythmical human phenomena and its effect is – like breathing, walking, love making, foot tapping, and head nodding – essentially physiological (4). 26 This emphasis on poetic metre and rhythm is also present in Harvey Gross’s Sound and Form in Modern Poetry (1964/2014). In the 2014 expanded version, Robert McDowell maintains that a poem is not the rendering of an idea, an experience or an attitude, but their transmutation into feelings. It is, then, prosody and its structure that articulate and transmit the arrangement of feelings in a poem and enhance our understanding of its meaning. In his opinion, the various elements of prosody – i.e. rhythmic form, onomatopoeia, the patterning of syntax and stress, vowels and alliterations – have a more crucial role than exclusively phonetically surfacing the meaning. He maintains: “Prosody enables the poet to communicate states of awareness, tensions, emotions, all of humanity’s inner life that the helter-skelter of ordinary propositional language cannot express” (8-9). John Hollander’s Rhyme’s Reason (1981/2014) is another classic text that surveys aspects of poetic form including those that contribute to the aural features of a poem – i.e. the patterns of rhyme, alliteration, assonance, and refrains, among others. Echoing Walter Pater’s well-known statement, “all art constantly aspires towards the condition of music” (86), Hollander argues that “poetry is not speech raised to the level of music, but music lowered to the level of speech” (x) and, therefore, poetic form is “an abstraction from or a residue of musical form, from which it came to be divorced when writing replaced memory as a way of preserving poetic utterance […]” (4). 1.1 Sounds in Poetry In a poem, details are fuzzier, but the structure of sounds – built of repetitions and variations, of expectations fulfilled or thwarted – has a richness to it. Poetic form is the very underpinning of poetry, which echoes and reinforces the poetic meaning, giving the poem “a tone of voice, an emotional intensity, and a moral force quite unlike anything else” (X-XI). In The Sound of Poetry (1998), Robert Pinsky echoes Fussell’s views of the physiological effect of poetic metre and rhythm. He gives an account of poetry as a bodily and aural experience when he says: “Poetry is a vocal, which is to say a bodily, art” (8). 27 Pinsky provides a guide for enhancing the readers’ ability to both perform and hear the sounds of a poem, which, according to him, is, nonetheless, an intuitive skill. He brings examples from the work of fifty different poets – including Shakespeare, Frost, Bishop and Bidart – to demonstrate the various terms and principles that shape the overall sound of a poem – accent and duration of a syllable, syntax and line, like and unlike sounds, as well as blank verse and free verse. Pinsky encourages his readers to be aware of the distinction between these pairs in order to gain more enjoyment and understanding of a poem. His emphasis is that the knowledge to recognise and deal with sounds in a line of poetry stems organically from the knowledge of patterns in everyday speech, which we begin to absorb from the cradle. Therefore, readers of poetry should not rely on absolutist or rigid principles of prosody and instead adopt a more flexible approach to reading a poem, one that could open up new working principle options for poetry (4-5). The brief sample of the literary works above indicates how traditional literary criticism deals with specific aural aspects of a poem. In the same way, a considerable number of works of literary criticism provide broad introductions to the art of poetry typically including extended and detailed accounts of how language and sound function in a poem. Shira Wolosky’s The Art of Poetry (2008) is one such example. Her analysis moves progressively from smaller units such as the word and line to larger features such as verse forms in the English lyric tradition discussing in great detail the concepts of metre, rhythm, syntax, metaphor, personification, poetic voice and diction. 1.1 Sounds in Poetry Wolosky’s emphasis is that what makes poetry distinctive from other literary forms is its language: A definition that underscores what makes poetry distinctive is: poetry is language in which every component element – word and word order, sound and pause, image and echo – is significant, significant in that every element points toward or stands for further relationships among and beyond themselves. (3) According to her, poetic language comprises a highly organised pattern in which each word has a specific place and purpose. These words are chosen partly for the purpose of a According to her, poetic language comprises a highly organised pattern in which each word has a specific place and purpose. These words are chosen partly for the purpose of a 28 particular sound pattern – as through consonants, vowels, and the sound repetitions of rhyme, half-rhymes and full-rhymes, with variations of unrhymed words – or a particular metrical pattern – as through the rhythm of the words – so that the poem has a melody or rhythm, like music. Therefore, a serious consideration of form in poetry can lead to a full understanding and enjoyment of poetry (Wolosky 3-4). Sam Halliday’s Sonic Modernity (2013) brings to light what he identifies as a ‘modern’ approach to sound in literature and other forms of art, such as painting, sculpture and cinema, where the act of listening turns into an object of visual representation. His study is a significant addition to the surge of interdisciplinary approaches to sound studies I discuss further in the following sections of this chapter. He explores the relationship between music and poetry in the light of Eliot’s theory of the ‘auditory imagination’ – i.e. in relation to ‘the feeling for syllable and rhythm’ – but also explores how the themes of sound, voice and silence in the work of modernist writers reflect the auditory dimensions of their time. According to Halliday, in modern art and culture, the concepts of sound and listening are presented in a trans-sensory and trans-disciplinary matrix of concerns, which are in dialogue presented in a trans-sensory and trans-disciplinary matrix of concerns, which are in dialogue presented in a trans-sensory and trans-disciplinary matrix of concerns, which are in dialogue with social, cultural, technological associations (3). He maintains: with social, cultural, technological associations (3). He maintains: with social, cultural, technological associations (3). 1.1 Sounds in Poetry He maintains: Sound in modernism, in whatever art form, is irreducible to sound alone. Sound, instead, is best conceived as a configuration, with real sound at its centre, to be sure, but other sense phenomena, such as touch and vision, rarely at more than one or two removes at its periphery. To fully grasp the significance of sound in modern culture, it follows, we must consider visual cultures of sound and verbal cultures of sound, and see all of these in dialogue with ‘sounded’ cultures of sound, more self-evidently made out of sound itself. (3) Julia C. Obert’s Postcolonial Overtures (2015) is a testimony to the proposition that an auditory study of the work of modernist writers reflects some of the contemporary auditory dimensions of their time. According to her, contemporary Northern Irish writers, particularly those who came of age during the Troubles, take the concepts of “discomfort and dislocation as their driving themes” (2). Her study brings together the postcolonial theories of, among 29 others, Edward Said and Homi K Bhabha, as well as the major theoretical concepts by philosophers and sound scholars, including McLuhan, Idhe and Schafer to examine the work of the three contemporary Northern Irish poets – Ciaran Carson, Derek Mahon, and Paul Muldoon – for the role of sound in the ongoing work of reconciliation with their troubled land. She maintains: “[t]his strategy suggests sound’s political and affective potential: music, accent, and even familiar white noise can help otherwise unmoored subjects feel at home” (Obert 3). Yet although Obert hopes to offer an exemplary work for a systematic study of sound in contemporary Northern Irish poetry, her concerns remain solely within Northern Irish politics. Angela Leighton’s Hearing Things (2018) bridges the traditional studies of sounds in poetry and the more recent approaches to sound and listening in humanities. Leighton remarks that “the fact that so many writers have stressed the importance of hearing, as the mainspring of both composition and interpretation, might be a reminder of something in the literary text which challenges our critical commonplaces, as well as perhaps our very styles of critical writing” (3). Drawing on the writings of critics and philosophers, including Idhe and Schafer, as well as on the commentaries of many poets and novelists from the nineteenth to the twenty- first centuries, Leighton offers a re-consideration of the role of the ear in literature. 1.1 Sounds in Poetry For her the concept of sound in literature could encompass a wide range of referents from the immediate sounds of the words on the page, to the echo of other texts or the acoustic memory of physical surroundings in the writer’s head while writing, to the voices of the characters or sounds experienced by them evoked in the text, as well as to the sounds readers pick up in the process of reading from both their memory and the surrounding world. Her main thesis is that listening is a form of cognitive perception that has often been overlooked and that all engagement with literature − i.e. writing, reading and interpreting − by the very nature requires listening. Poetry, she maintains, is: 30 […] the art which speaks the act of listening, not only as a check on composition − the poet submitting his work to a later painstaking listening which might revise the original − but also as a kind of writing which contains the space of listening within itself. (11) This thesis is aligned with the work of more recent literary scholars in the sense it bridges the traditional approaches in literary criticism and the more recent approaches in sound studies for interpretive purposes. But unlike any of the previous studies, it also offers a framework for identifying potentially auditory references in writing, one that also includes silence. In other words, as one of its primary goals, this study sets out to offer a way to read poetry more acoustically, but a way that goes beyond the traditional studies of the word- sound per se. Moreover, this thesis argues for the Northern Irish poet’s acoustic attunement with his surroundings in a more global sense. Thus while it testifies to the significance of sound in understanding Heaney’s engagement with the political circumstances of Ireland, this thesis relates to the concept of sound in Heaney’s work in a much broader context, one that includes his aesthetic, personal, social, cultural and environmental concerns as well. 20 While, in their publications, some historians refer to this approach as ‘the history of the senses’ or ‘history of hearing’, others prefer the term ‘sensory history’ or ‘hearing history’. These scholars refer to relatively similar concerns and can be used interchangeably (M. M. Smith 842). 1.2 A Historical and Anthropological Perspective A well-established approach to the study of sounds has been developed by scholars in hearing history and anthropology.20 These scholars have attempted to identify the characteristics and potentials of auditory experience, as a sensory experience, in constructing and reflecting the culture − i.e. “the habits, systems of belief, knowledge and action” in a given time and place (Bull & Back 223). This section provides a brief overview of historical and anthropological approaches to hearing and sounds in order to prepare an intellectual and conceptual context for the ecological approach applied in the thesis. 31 In many respects, the scholarly attention to sound that has burgeoned in the late twentieth and twenty-first centuries can be connected to the notion of ‘orality’ and of oral cultures explored at length in the work of Walter Jackson Ong. In The Presence of the Word (1967), Ong discusses the three phases of human civilisation: the primary preliterate or the oral/aural period of spoken communication; the scribal or chirographic/typographic phase; and, finally, the electronic or secondary orality/aural phase (72). Ong’s theory highlights the difference he sees between orality and writing, privileging the latter as a more advanced phase in human civilisation. In Orality and Literacy (1982), he notes: Without writing, words as such have no visual presence, even when the objects they represent are visual. They are sounds. You might ‘call’ them back – ‘recall’ them. But there is nowhere to ‘look’ for them. They have no focus and no trace (a visual metaphor, showing dependency on writing), not even a trajectory. They are occurrences, events. (31) Ong is certainly not the only one to write on oral cultures, and in Orality and Literacy, he refers to earlier scholars such as Milman Parry, Alfred Lord and Eric Lovelock, together with his contemporary and colleague, Marshall McLuhan, who also wrote at length on the significance of the historical emergence and gradual dominance of print cultures. McLuhan and Ong agree on the fundamental bracketing of the pre-modern age as non-visual, or primarily aural, and the modern age as visual, generally cited as “the orality theory” or “the great divide theory” (McLuhan 20-21). In The Gutenberg Galaxy (1962), McLuhan also introduces the concept of “the ratio of the senses” or “sensory ratios” − i.e. 1.2 A Historical and Anthropological Perspective the ranking of the senses − in order to explain the transformations of language and technology, as well as the differences in social structures throughout history (20-21). According to McLuhan’s orality theory, before the invention of writing humans lived in a “boundless, directionless, horizonless” acoustic space in which speech and hearing − and thus tactility and olfaction − constituted the basis of communication and the nature of collective behaviour (207). With the invention of the phonetic alphabet humans abandoned 32 this highly social and hyperesthetic primal space for the sake of the more detached, homogeneous and static domain of vision (54). Yet it was not until the invention of movable type that visual culture became prominent. The manuscript or scribal culture was still intensely audile-tactile. As in antiquity, people read principally with their lips pronouncing what they saw and listening to the words pronounced. With the advent of print technology in the fifteenth century, texts became eventually standardised, affordable, portable and reproduceable on a large scale, pushing auditory and other senses to the background and contributing to the new cult of individualism, detachment and non-involvement (125-206). It was this visual hegemony that, according to McLuhan, characterised Western society until fairly lately. In Understanding Media (1994), he maintains that, thanks to electronic technology and the advent of telephone, gramophone, radio and even television, the modern age witnessed a resurrection of hearing and sounds as a major medium of perceiving and communicating. In post-literate acoustic spaces, the eye no longer reigns. Time has ceased. Space has vanished. We now live in a “global village” (263-269). The study of sounds and hearing was affected by McLuhan’s and Ong’s ear/eye, premodern/postmodern binarism in many ways. This binarism has, on the one hand, pushed aural perception to the secondary position in the traditional hierarchy of the senses. As a result, in Western sensory history, hearing has often been regarded as the second sense and as the bridge between what is commonly referred to as the “highest” sense of sight and the “lower” senses of smell, taste and touch or kinesthetic. The emphasis placed on the transition from oral to printed word has, on the other hand, privileged orality and linguistic sounds over non-linguistic sounds (M. M. Smith 850). 21 See, for instance, Edward J. Chamberlin’s chapter “A New History of Reading: Hunting, Tracking, and Reading” in Geography of a Soul (2001) and Bruce Rosenberg’s “The Complexity of Oral Tradition” (1987). , , p y g g, Reading” in Geography of a Soul (2001) and Bruce Rosenberg’s “The Complexity of Oral Tradition” (1 See, for instance, Edward J. Chamberlin’s chapter “A New History of Reading: Hunting, Tracking See, for instance, Mark M. Smith s Sensing the Past (2007) and Alain Corbin s Time, Desire and Horror: Toward a History of the Senses (1995), for an introduction to sensory history; and David Howes’ The Varieties of Sensory Experience (1991) for a typology of various cultures based on variations in the organisation of the sensorium. , , g ( ) , Toward a History of the Senses (1995), for an introduction to sensory history; and David Howes’ The Varieties of Sensory Experience (1991) for a typology of various cultures based on variations in the organisation of the sensorium See, for instance, Mark M. Smith’s Sensing the Past (2007) and Alain Corbin’s Time, Desire and Ho Toward a History of the Senses (1995), for an introduction to sensory history; and David Howes The of Sensory Experience (1991) for a typology of various cultures based on variations in the organisati sensorium. 1.2 A Historical and Anthropological Perspective McLuhan’s and Ong’s treatment of orality has attracted much criticism,21 informing much of the later scholarly discussions on sensory studies in general, and on hearing studies in particular. These historians refrain from leaning 33 too heavily on such meta-historical frameworks, whose application is limited by new findings. There have been numerous studies to investigate the role of the senses in various cultures contributing to the broad field of sensory studies.22 Sound and hearing scholars aim mainly at disrupting the privileging of sight by giving voice and place, in their studies, to the role of listening, sound and silence in shaping and reflecting different cultures. As I discuss in the following paragraphs, their studies show that different modes of perception may overlap one another and that aural perception has remained an important aspect of human life throughout history. Aurality and orality had a prominent status in many ancient cultures, when the word of God, the knowledge and history of the tribe were transferred orally. Denys Thompson’s study of the role of poetry in The Uses of Poetry (1978) points out the cultural, social, medical and legal prominence of orality in ancient Persia, Greece, China, Pre-Islamic Arabia, Scotland and Ireland. His study shows that vocable and non-vocable sounds were integral parts of many ancient and medieval societies not only for communication but also for mediating various forms of social engagements, economic arrangements, cultural and religious rituals, as well as legal and political expressions (117-212). In Coming to Our Senses (1992), Bernard Hibbitts introduces the concept of “performance culture” to refer to societies that have maintained their pre-literate expressive and intellectual habits (882). Among these societies are pre-Hellenistic Greece, biblical Israel, early republican Rome, tenth-century England and medieval Europe, in which the “saying mattered more than the writing” (897). The term “performance culture” makes a general reference to theatrical connotations of interaction and the fact that in oral cultures significant information was delivered through a combination of all sensory media and not just speech. Hibbitts remarks: 34 “Individuals in these societies are performers in the sense of being culturally fluent in speech, gesture, touch, smell, and taste” (883). 23 The technical terms are formally defined at the end of this chapter. 1.2 A Historical and Anthropological Perspective Peals solemnized an occasion and gave rise to or expressed rejoicing. They were far more effective in this regard than were rough music or the charivari. Any collective emotion that ran deep involved use of a bell, be it the threat of fire or bloodshed announced by an alarm or the terror aroused by the passing bell tolled during epidemics. (288) Finally, we have come to realize just what emotional power bells possessed. Peals solemnized an occasion and gave rise to or expressed rejoicing. They were far more effective in this regard than were rough music or the charivari. Any collective emotion that ran deep involved use of a bell, be it the threat of fire or bloodshed announced by an alarm or the terror aroused by the passing bell tolled during epidemics. (288) Corbin’s study also identifies how the conceptualisation of bells − or any sound, keynote or otherwise − can be class-based and clash as we pass from rural to urban life. He maintains that “complaining of discomfort caused by the din of bells was a venerable urban tradition […]” (299). The interpretation of bell sounds − and street music − also “formed part of a struggle of the elites, who were intent on imposing their fastidious taste” (299). Corbin’s study also identifies how the conceptualisation of bells − or any sound, keynote or otherwise − can be class-based and clash as we pass from rural to urban life. He maintains that “complaining of discomfort caused by the din of bells was a venerable urban tradition […]” (299). The interpretation of bell sounds − and street music − also “formed part of a struggle of the elites, who were intent on imposing their fastidious taste” (299). With colonisation, the acoustic markers of place and identity, and the associations of silence, loudness and noise with social class expanded to new territories. Colonisers carefully structured the new soundscapes in such a way as to create and regulate social hierarchies and to define their own cultural superiority (Rath 104-175). A clash between cultures and languages is evident in the course of English colonialism beginning in the sixteen and seventeenth centuries. 1.2 A Historical and Anthropological Perspective Richard Cullen Rath maintains that, just as in Greek poleis, Medieval European geographical territories were sometimes established by the range of a crier’s voice and according to the hearing distance, a practice that endured until the time Christian parishes were defined by the distance at which church bells could be heard (56-57). As the printing press became an agent of cultural transformation in Western societies, the more traditional oral-aural channels of communication lost a crucial measure of their legitimacy. Nonetheless, with industrialisation, sound and aural communication turned into a critical matter to everyday urban life, particularly in early modern Europe, serving to coordinate civic, political, economic and social life. Davis Garrioch notes that for the inhabitants of seventeenth, eighteenth and even nineteenth-century Europe, city noise had a different function from today. Whereas today many may try to escape city noise, the acoustic space of early modern European cities functioned as a crucial source of information: “it formed a semiotic system, conveying news, helping people to locate themselves in time and in space, and making them part of an auditory community” (5). From market voices and shouts, to the rattling of bicycles, the hissing of hot metal in a blacksmith's workshop, and to the noises of local industries, each city developed the ‘keynotes’ of what Mark M. Smith calls “a distinctive aural signature” (45).23 However, the sounds of economic development and productivity during the modern ages became so loud and blended with one another that formerly ‘soundmarks’ − sounds of a particular trade in a particular geographic area − collapsed. A mixing of sounds, from varied trades and businesses, marked the soundscape of the modern era and formerly discrete soundmarks either disappeared or spilled their sounds to adjacent regions (45). 35 Among others, the church bells of early modern and modern years have attracted much attention from sound historians and urban sound scholars. According to Garrioch, the sound of bells in early modern cities functioned as alarm clocks, factory horns and school bells of today (5). In Village Bells (1998), Alain Corbin explores the history and meaning of bells in nineteenth-century France as a broad proxy for the Western experience in general. Corbin summarises the important role of bells as signals, markers and symbols of local and communal identity and resistance: Finally, we have come to realize just what emotional power bells possessed. 1.2 A Historical and Anthropological Perspective In Shakespeare, Spenser, and the Crisis in Ireland (1997), Christopher Highley calls attention to the English impulse to merge all its borderlands − Wales, Scotland, and Ireland − “into a single territorial and ethnographic zone, with common linguistic and cultural ties, and with a shared hostility to the English” (68). Bruce R. Smith’s The Acoustic World of Early Modern England (1999) maintains that for the ears of the imperial-minded 36 English, Gaelic was treated as not only an “aural assault” but also a mode of communication that was only “one step removed from noise” (306). One reason for the English hostility may have been the fact that Gaelic was “notoriously difficult for a non-native to learn” (306). More importantly, however, it was encouraged by the imperialistic attitudes towards the Irish themselves. Such attitudes had roots in pre- and early modern theories of the racial origins of the Irish as “mixed of Scithian and Spanish blood” by such figures as William Harrison and Edmund Spenser, among others (Highley 68). During the eighteenth and nineteenth century, in which the idea of noise and silence was closely dependent on social identity and class, some commentators continued to describe the Irish acoustically as rowdies and noisy. In his 1839 Diary, for instance, the British Royal Navy officer Frederick Marryat spoke of “the turbulence of the Irish lower classes … [a] disposition that appears to follow them everywhere” (140-141). Although the majority of hearing scholars explore historical agencies of sound, silence too played a significant role in shaping and reflecting social and political dynamics. According to M. M. Smith, in Hearing History, the injunctions for silence, as the marker of class and taste, were rooted in an older, more hierarchical use of sounds and listening in which ruling classes demanded quietude and obedience in an effort to reaffirm the contours of power in their society (93-260). As I explain in the final chapter of the thesis, this authoritative use of silence − more apparent in the relationship between slaves, servants and masters as well as women and children, as inferior and dependent categories of society − is also explored in Heaney’s poetry. It is in listening for moments of silence and in attempting to redefine the concepts of noise and music that one can unfold pivotal turning points in political, social and cultural history (Parr 740). 1.2 A Historical and Anthropological Perspective Raymond Smilor has shown that the period 1893-1932 witnessed shifts in what was deemed the ‘sound’ of modernity to the ‘noise’ of modernity. Whereas an earlier 37 generation applauded the sound of industries as inseparable from modernity and capitalist progress, the reformers of the early twentieth century Progressive Era redefined it as the excessive, frustrating and damaging noise of modernity, giving rise to multiple anti-noise campaigns and legislative attacks (24-25). Schafer suggests that the very emergence of the concept of noise pollution as a public concern testifies to the fact that the modern world is not unaware of the role of sounds and hearing on human society and the natural environment (4). However, modern efforts to combat the noise of modernity reside in the very subjectivity of the distinction between noise, sound and music. The First and Second World War, for instance, witnessed a deliberate use of noise as a terror tactic (Priestley 100). White noise − a high-pitch disorienting electronic sound, was used by British security forces in Northern Ireland during the early 1970s for interrogation purposes (Bailey 25). This insight becomes particularly relevant when interpreting the political connotations of Orange drumming, British helicopters, paramilitary shootings and bombings in Heaney’s poetry. generation applauded the sound of industries as inseparable from modernity and capitalist progress, the reformers of the early twentieth century Progressive Era redefined it as the excessive, frustrating and damaging noise of modernity, giving rise to multiple anti-noise campaigns and legislative attacks (24-25). Schafer suggests that the very emergence of the concept of noise pollution as a public concern testifies to the fact that the modern world is not unaware of the role of sounds and hearing on human society and the natural environment (4). However, modern efforts to combat the noise of modernity reside in the very subjectivity of the distinction between noise, sound and music. The First and Second World War, for instance, witnessed a deliberate use of noise as a terror tactic (Priestley 100). White noise − a high-pitch disorienting electronic sound, was used by British security forces in Northern Ireland during the early 1970s for interrogation purposes (Bailey 25). This insight becomes particularly relevant when interpreting the political connotations of Orange drumming, British helicopters, paramilitary shootings and bombings in Heaney’s poetry. The persistent presence and development of auditory experience can be traced in the advent of recording and electronic technology. 1.2 A Historical and Anthropological Perspective Jonathan Sterne’s The Audible Past (2003) shows that while early wax cylinder recordings and even metal or shellac disks were fragile, the advent of sound recording and reproduction technology at the end of the nineteenth century re-established the position of sound and listening. Audible records gave the appearance of permanence to the hitherto ephemeral sounds, music and voices, that could till then be captured by silent script and print only (288). Don Ihde also notes that the birth of electronic instruments has revolutionised our ability to experience sounds. Thanks to telephones, radios, sound amplifiers and radio telescopes, to name a few, we have learnt to extend the range and scope of our aural experiences. We are now aware that once silent oceans resound with the songs of sharks and whales and the Galilean macrocosm is not thoroughly quiet (14). 38 In sum, despite the growing dominance of print culture, as well as the emphasis Renaissance perspective and Enlightenment science and rationalism placed on sight as a conveyor of truth and knowledge, sound continued to function as a marker of time and place, as well as the concepts of identity, nationalism, and social and economic class. The persistent role of sounds and aural perception during the nineteenth and twentieth centuries, too, suggests that seeing and hearing continue to operate in conjunction with one another. 25 According to Gibson, the anatomical definition of sensory system – as merely receptor elements, cerebral cortex and neural pathways – fails to accommodate factors like adaptation, purposefulness and anticipation that define the relationship between actions and behaviours of a living being and the surrounding environment. From this perspective, there are several kinds of knowing, of which sensory perceptions is the “simplest and the best kind”. Knowing can also be obtained by means of instruments, such as movies and books, which extend perceiving into the realm of the very “distant” and the very “small” (246-251). 1.3 Towards an Ecological Approach to Sounds and Listening As sonic worlds have changed, so too have the conceptual infrastructures and scholarly approaches to examine and interpret them. Scholars who champion an ecological approach argue that the questions concerning sounds and listening have to deal with higher levels of cognition than the elemental stimuli − e.g. frequency, amplitude, phase and duration − discussed primarily in traditional psychoacoustic and cognitive approaches. An ecological approach is more concerned with the content of the cognition and explores auditory perception along the dimensions of sound-producing source, the location and the environment rather than soundwaves per se (Clarke 7). The ecological approach to sounds and listening is intertwined with complementary fields of acoustic ecology, bioacoustics, urban soundscape and soundscape ecology which similarly explore the potentials of an ecological approach in relation to aural perception, natural sounds, speech sounds and even musical sound. The ecological approach to auditory perception − also sometimes referred to as ecological acoustics − stems from James Jerome Gibson’s study of the senses as perceptual systems and his accounts of the ecological approach to visual perception. According to Gibson’s 1966 theory, the surrounding environment consists of enormously complex but lawful and legitimate “affordances” essentially perceived through touch, sounds, odour, taste 39 and the ambient light (127).24 The human sensory system is a perceptual system that can facilitate knowing and understanding at the ecological level (251).25 Gibson maintains that a person’s understanding of things in life depends on how he perceives the surrounding environment and interacts with it: “to perceive is to be aware of the surfaces of the environment and of oneself in it” (XXIII) and “knowing is an extension of perceiving” (246).26 Thus, according to an ecological approach, people’s knowledge of themselves and their surroundings is acquired through looking, listening, feeling, smelling, and tasting, which is, of course, influenced by the knowledge coming from society in general. As Eric Clarke points out, [i]n ecological theory, perception and meaning are closely related. When people perceive what is happening around them, they are trying to understand and adapt to what is going on. In this sense, they are engaged with the meaning of the events in their environment. (7) [i]n ecological theory, perception and meaning are closely related. When people perceive what is happening around them, they are trying to understand and adapt to what is going on. p g y y ( ) 26 See Claire Michaels and Claudia Carello’s Direct Perception (1981), Edward Reed and Rebecca Jones’s “Perception and Cognition” (1982) for more complete reviews of Gibson’s perceptual theory. Also see Robert M. Boynton’s Visual and Auditory Perception (1975) for counter arguments. y y p ( ) g 27 See, for instance, Nancy Jean Vanderveer’s “Ecological Acoustics” (1979); Stephen Handel’s Listening: An Introduction to the Perception of Auditory Events (1989); Albert Bregman’s Auditory Scene Analysis (1990) and Rainer Guski’s “Auditory Localization: Effects of Reflecting Surfaces” (1990). 24 Gibson maintains: “The affordances of the environment are what it offers the animal, what it provides or furnishes, either for good or ill […]. It implies the complementarity of the animal and the environment” (127). 25 di ib h i l d fi i i f l l b l perceiving into the realm of the very distant and the very small (246 251). 26 See Claire Michaels and Claudia Carello’s Direct Perception (1981), Edward Reed and Rebecca Jones’s “Perception and Cognition” (1982) for more complete reviews of Gibson’s perceptual theory. Also see Robert M. Boynton’s Visual and Auditory Perception (1975) for counter arguments. 27 See, for instance, Nancy Jean Vanderveer’s “Ecological Acoustics” (1979); Stephen Handel’s Listening: An I t d ti t th P ti f A dit E t (1989) Alb t B ’ A dit S A l i (1990) d 1.3 Towards an Ecological Approach to Sounds and Listening In this sense, they are engaged with the meaning of the events in their environment. (7) Since then, sound scholars have attempted to apply the ecological approach to a variety of research in the study of listening in order to gain a different and deeper insight into the relationship between listening, meaning, culture and nature.27 From an ecological perspective, the soundscape of a place can be seen as the way the surroundings present themselves acoustically to a listener. Thus, everything audible in the setting can be treated as conveying meaning (Clarke 6-7). In other words, human aural perception, the surrounding soundscape and the process of meaning-making are deeply intertwined: 40 […] to hear a sound and recognize what it is […] is to understand its perceptual meaning, which will result in corresponding actions. By contrast, to hear a sound and not recognize what it is, is to fail to understand its meanings and thus to act appropriately. (Clarke 6-7) An ecological study of aural perception aims at “uncovering ecologically relevant dimensions of perception and the invariant perceptual information for them” (Clarke 7). In his 1990s ground-breaking twin articles − “What in the World Do We Hear?” and “How in the World Do We Hear?” – William Gaver endeavours to organise a descriptive framework for identifying ecologically relevant perceptual attributes of the sound-producing sources. The upshot of this perspective is the premise that rebukes the traditional assumption of perception in many ways. An ecological approach to aural perception stretches beyond the accounts of elemental sensation by exploring the perceptual parameters of the sound source; it focuses on all sounds, including musical sounds and the everyday sounds of the mundane; it explains that the perception of complex sound sources does not rely on a complex integration of sensations, but may instead be stimulated by the complex perceptual information emitted by a combination of simple physical features; finally, it is direct and first-hand and far from any biased interventions by memory or subconscious (Gaver, “What” There is rich and varied information in the world, both because our descriptions are no longer limited to primitive physical dimensions and because exploration of the world over time − as opposed to passive exposure to discrete stimuli − becomes an important component of perception. 28 The project was inspired by Schafer's interest in environmental acoustics in the late 1960s. Publications which emerged from the World Soundscape Project include Schafer’s The Book of Noise (1970) and The Tuning of the World (1977), Truax’s Handbook of Acoustic Ecology (1978), and Järviluoma and Wagstaff’s Soundscape Studies and Their Methods (2002). Institutes of sound studies are now aggregated by the World Forum for Acoustic Ecology (WFAE) and unified by the field’s seminal journal, Soundscape. See www.wfae.net for more details. 29 See www.wikiwand.com/en/World_Soundscape_Project. 1.3 Towards an Ecological Approach to Sounds and Listening (“What” 4-5) The social, ethical and aesthetic dimensions of auditory perception are pervasive themes in acoustic ecology, a relatively new field of study initiated by the Canadian musicologist and composer Raymond Murray Schafer in the early 1960s.28 The ecological 41 study of soundscape, as envisioned by Schafer, Barry Truax, Helmy Järviluoma and Gregg Wagstaff is an interdisciplinary approach that studies relationships and interactions between human beings and their sonic environment. Schafer’s foundational work was conducted with the World Soundscape Project which inaugurated the field of acoustic ecology, a discipline that quickly developed an international profile among academics and activists. Its ultimate purpose, in addition to recording and cataloguing international soundscapes with a focus on dying soundmarks, was to investigate the impact of soundscape on our sense of place and to raise consciousness about the impact of increasing noise pollution on the sonic environment.29 Schafer’s benchmark 1977 book includes essays on the history of soundscape and introduces a rich terminology and taxonomy of sounds − e.g. keynote, soundmark, sound signal, and archetypal sounds. For Schafer touch is the most personal and intimate of the senses and hearing is “a special sense” because its vibrational spectrum includes frequencies that are felt as well as heard − i.e. about 20 hertz: “Hearing is a way of touching at a distance and the intimacy of the first sense is fused with sociability whenever people gather together to hear something special” (11). Like McLuhan, Schafer too believes that the invention of print technology and perspective painting shifted the ratio of the senses towards the eye, but also that the advent of electric technology and the emergence of noise pollution as a public concern testifies to the fact that aurality is regaining its relevance once again. He reminds us of the subjectivity of the terms noise, sounds and music throughout history and writes about the “positive” role of silence in our lives. He treats the world as “a macrocosmic musical composition” and calls for a culture of “ear cleaning” or “clairaudience” and, ultimately, for an improvement in the “acoustic design” of the sonic world (3-12). 42 Truax’s 1978 publication introduces the term ‘soundscape ecology’ almost interchangeably with acoustic ecology to refer to the “study of the effects of the acoustic environment on the physical responses or behaviour of those living in it” (127). In 2011, Pijanowski et. al. 1.3 Towards an Ecological Approach to Sounds and Listening published the foundational elements of soundscape ecology as an independent field of study defined as “the science of sounds in the landscape” (203). According to them, the arguments in acoustic ecology are largely human-centred, while the field of soundscape ecology focuses on the larger ecological systems (204). However, there are strong parallels and shared terminology between the two fields. Soundscape ecologists treat sounds as perpetual and dynamic properties of all landscapes and, unlike the previous eco-acoustic, urban acoustic and bioacoustic approaches, are interested in the entirety of sounds. Soundscape ecologists identify the three fundamental sources comprising the soundscape − geophony, biophony and anthrophony − to examine relationships between living organisms, including human and other terrestrial or marine organisms, and their environment across different spatial and temporal scales. The aim of this approach is to describe how various factors − e.g. climate, land transformations, biodiversity patterns, timing of life history events and human activities − control and shape the soundscape. Soundscape ecologists argue that natural sounds engage our aural sense and provide us with a lot of information about our surroundings, and therefore they should be carefully listened to and valued: “as with other natural resources, natural and unique soundscapes have many associated human ideals, such as cultural, sense of place, recreational, therapeutic, educational, research, artistic, and aesthetic values” (Pijanowski, et. al 204). Finally, the 1976 ground-breaking publication of Ihde has provided the philosophical Finally, the 1976 ground-breaking publication of Ihde has provided the philosophical and epistemological underpinnings for sound studies. In its 2007 expanded edition, Listening and Voice, Ihde offers a phenomenological interpretation of auditory perception, ranging from the experience of sound through language, music, religion and silence. He maintains 43 that a mute and stable object is given “voice” only in the encounter with some other object and, thus, in listening to the voice of an object we often hear interactions, configurations, materials, surfaces, and shapes related to more than just one object. Voices of objects often also reveal information about the specific time, place and environment from which they are emitted (67-68). Ihde also speaks about the horizons or limitations of auditory experience and the fact that there is often some degree of overlapping of seeing and listening domains. 1.3 Towards an Ecological Approach to Sounds and Listening According to him, there remains the “excess” of sight over sound in the realm of the mute object, while listening “exceeds” seeing in the realm of the invisible, or where sight may not matter as much. In other words, silence is the horizon of sounds, and the invisible offers a horizon to vision (50-55). The study of sound and listening is not exclusive to sound scholars, but as shown in the previous sections of this chapter belongs to multiple scholarly traditions. Over the last few decades, there has been a growth in writings by humanities and social science scholars, whose work is distinguished by an interdisciplinary discussion that involves any sonic phenomena. Scholars from different fields such as history, anthropology, media studies, postcolonial studies, and many others, began to use sounds to investigate major cultural and political moments. These scholars are not necessarily sound scholars, but, as Sterne phrases it in Sound Studies Reader, they “think sonically” (1-2). They reach across disciplines and traditions in sound studies in order to “re-describe what sound does in the human world, and what humans do in the sonic world” (1-2). It is this multi and transdisciplinary curiosity that informs the research questions of this thesis. 1.4.1 An Ecological Approach to Listening: A Descriptive Framework As discussed above, a descriptive framework based on the ecological approach to listening reaches beyond the traditional sensation-based theories of elemental stimuli. Instead, it focuses on the range of the perceptual attributes and dimensions that characterise the auditory perception of the sound-producing source, its specific time and location (Gaver 5-8). This section provides an elaborate discussion of the concepts of source, place and time as the three main dimensions of a sound from an ecological point of view. 1.4 Methodology and Theoretical Framework This thesis adopts a sonic approach to the study of the poetry of Seamus Heaney, for whom listening was a prominent way of perceiving the surrounding world and whose poetry 44 is an archive of culturally, politically, socially and personally loaded references to sounds. In attempting to answer the research questions of this thesis, I rely on close reading as an effective lens to identify and interpret the numerous references to sound – and their absence − that are found in Heaney’s poetry. The theoretical framework applied in the thesis offers a new strategy for both identifying and interpreting auditory references in his work, one that unlike the previous studies on Heaney’s auditory imagination looks specifically and holistically at all the various aspects of the soundscape in his writing. The focus of my analysis is on individual poems, even individual stanzas or lines in Wintering Out. I contextualise this close reading within wider discourses of literary criticism on Heaney. The theoretical framework applied in this thesis is primarily based on the descriptive framework in the ecological approach to auditory perception (Gaver 1992) in order to facilitate the identification of the references to sounds − and their absence – in the poems. It also benefits from the vocabulary related to the taxonomy of sounds in acoustic ecology (Schafer 1977) and soundscape ecology (Pijanowski et. al. 2011) as well as a variety of aspects in hearing history and anthropology so as to explain the personal-social-cultural associations of sounds and their absence. In addition, where appropriate, I draw from acoustic phenomenology (Ihde 2007). I propose that the integration of these approaches provides the framework for identifying and interpreting the references to sounds woven into the soundscape of the poems. This approach provides readers, scholars and translators alike with a perspective for reading and experiencing poetry in a way that is more evocative, tangible and interpretative. In the remainder of this chapter, I elaborate on the concepts and terminology that are used in the ecological approach to auditory perception and soundscape interpretation and I consider the implications of using them in the analysis of the poems. I complete this chapter with an examination of Wintering Out as the specific focus of this study. 45 y ( ), ‘sound object’ which is an acoustic object independent of the original source. 30 Other scholars alternatively use the terms “auditory source”, “sonic source” and “acoustic source” to refer to sound-producing events in an environment. See William Whitmer, et. al. (2014), Brian Shaw, et. al (1991). and y ( ) p ‘sound object’ which is an acoustic object independent of the original source. ound producing events in an environment. See William Whitmer, et. al. (2014), Brian Shaw, et. al (1991 abriele Proy (2002), for instance. These terms should not be confused with Pierre Schaeffer’s conce Gabriele Proy (2002), for instance. These terms should not be confused with Pierre Schaeffer’s concept of ‘sound object’ which is an acoustic object independent of the original source. roducing events in an environment. See William Whitmer, et. al. (2014), Brian Shaw, et. al (1991). and Proy (2002), for instance. These terms should not be confused with Pierre Schaeffer’s concept of bject’ which is an acoustic object independent of the original source 30 Other scholars alternatively use the terms “auditory source”, “sonic source” and “acoustic source” t Auditory Source One of the first things we identify when hearing a sound is its source. Sound sources share several common features, the most common of which is the “interaction of materials” (Gaver 8). The terms “sound-producing event” (Gaver 9) or “sound event” (Schafer 274) are often used to refer to the smallest self-contained sound-producing source in a soundscape. In this thesis, I use these terms interchangeably to refer to a sound-producing source.30 Gaver’s study indicates that, in each auditory experience, the sounds emitted reveal the type of the material (e.g. elasticity, viscosity, tension, texture), their configuration (e.g. size, dimension, shape) and the type of the interaction (e.g. vibration, escarping) involved in producing the sound. Sound-producing sources fall into the three “basic-level” events caused by (a) vibrating solids; (b) changes in liquids; and (c) changes in aerodynamic materials. The combination of more than one type of interaction and/or material, and the involvement of an initial force can add more “complexity” to the energy patterns emitted by the sound- producing event and, thus, more information to the sounds perceived by the listener (Gaver 9- 17). The identification of potentially audible physical properties and interactions, by Gaver, 46 can be beneficial when tracing the references to sound − and their absence − in Heaney’s poetry. However, this is not to claim that every sound provides the listener with all the physical attributes of their source, but, as Ihde points out, neither does every sighting give the observer a full outline of objects seen. He maintains that our inability to discriminate the perceptual characteristics of sounds often stems from the “inadequacy” of our prior observations and knowledge and encourages us to sharpen our listening abilities and learn to listen to the information transmitted through sounds (61-64). The definition of the sound source as an “event” indicates that a sound cannot be abstracted from the spatial and temporal continuum that defines it. Moreover, a sound event is “a symbolic, semantic or structural object for study” further bound to the socio-cultural context in which it occurs (Schafer 274). Auditory Space Listening is also a spatial experience. A sound-producing event always occurs in a specific “location” and within a specific “environment” (Gaver 6). Once produced, the energy waves emitted by the source event are exposed to several landmarks, whose specific characteristic modifies the original energy patterns through “reflection”, “absorption”, “refraction” and “diffraction” (Schafer 217). These sound waves converge at the listening point at different times enabling the listener to distinguish the information about the source and its location from the ones reflected from the surroundings. Therefore, listening to a sound enables the listener to identify the spatial properties of the sound source − i.e. position, proximity, force, mobility and direction − and the sonic environment − i.e. structure, dimensions, density and changes (Gaver 6-17). 47 The energy waves emitted by a sound source can convey information about the proximity of a source event and its changes with respect to a certain point. Directional information can also be identified as a result of the sensitivity of our auditory system to the difference in the arrival time between the ears. However, since the auditory system is mobile the listener can turn his head to improve localisation, orienting towards what he may consider as the source event (Gaver 6-8). The experience of echo and reverberation conveys the sense of distance, depth, surface, size and continuity. Echo is produced when sound reflections − caused, for instance, by the bottom of a well or the walls of an enclosed and empty room − return to the listener with a delay. It is distinguishable as “repetition or partial repetition” of the original soundwave. Reverberation is created when the reflections persist for a while and then decay and die as the soundwaves are absorbed by surfaces they hit. It implies an enormous space. Therefore, both reverberation and echo re-sound the original sound and, because of their continuity, give the illusion of permanence to it (Schafer 130; 219). The last two spatial terms that I use throughout this thesis and would like to define in this section are acoustic space and auditory field. Schafer suggests the term “acoustic space” to refer to “the profile of a sound over the landscape” (271) and explains how technology − e.g. radios, microphone and amplifiers − have enabled individuals to expand their acoustic space (91-92). Auditory Space He also introduces the concept of “the sonic intruder” to refer to the unwanted sounds that invade other acoustic territories (214) and the term “imperialistic” sounds to refer to the sounds that have “the power to interrupt and dominate other acoustic activities in the vicinity” (77). The term “auditory field” is offered by Ihde to refer to the implicit context “that situates and surrounds” a more “explicit and focal” sounding source. However, he also explains that the distinction between “a fringe phenomenon” and “a focal phenomenon” constantly shifts and is keyed to “personal-social structures” (73-75). As I explain in the 48 following paragraphs, such elements as personal interests, habits, social occasions, as well as the geographical and climatic characteristics of a place are the determining factors in distinguishing foreground and background sounds. , gy f f Søren Kierkegaard’s Either/or (1959), and Ong’s Orality and Literacy (1982). 31 See for instance, Edmund Husserl’s On the Phenomenology of the Consciousness of Internal Tim See for instance, Edmund Husserl s On the Phenomenology of the Consciousness of Internal Time (1893), Søren Kierkegaard’s Either/or (1959), and Ong’s Orality and Literacy (1982). 31 See for instance, Edmund Husserl’s On the Phenomenology of the Consciousness of Internal Time (1893), Søren Kierkegaard’s Either/or (1959), and Ong’s Orality and Literacy (1982). 31 See for instance, Edmund Husserl’s On the Phenomenology of the Consciousness of Internal Time (1893), Auditory Time There is an established debate about the temporality of auditory experience amongst philosophers and sound scholars.31 From an ecological perspective, the definition of a sound source as an “event” involving an action or movement suggests that a sound “is not abstractable from the time-and-space continuum” (Schafer 274). The temporal dimension of sound and auditory experience can be revealed in many ways, including its instantaneousness, sequentiality, pause, or rhythm and repetitive patterns, which I illustrate throughout my analysis. One of the most obvious time-related characteristics of sounds is its instantaneousness. Sounds are temporal. They exist now and then they are gone. This relationship between time and sound reveals itself as soon as we begin to “listen reflectively” − i.e. with conscious awareness (Ihde 57-59). Reflective listening, therefore, unfolds the transitory nature of sounds as temporal events and highlights the essentiality of mindfulness in order to capture them. Another time-related feature of auditory experience becomes apparent when sounds occur sequentially representing a succession of acoustic events (Ihde 59). This characteristic is more evident when metaphorical or descriptive language adopts an auditory tone, as in the first stanza of the poem ‘Land’ where Heaney describes the Ulster farmer − or himself − stepping through the land and doing the daily chores one after another. 49 The auditory system is sensitive to the patterns and repetitions in the flux and flow of the sound events that occur. The rhythmic patterns of sound events register a sense of motion and continuity, as well as a sense of stability and orderliness in the listener (Ihde 87-89). The rhythms of the universe are infinitely varied in scale and tempo, ranging from our own heartbeats, breathing, footsteps to circadian and seasonal rhythms in nature and human life, and to the celestial motions (Schafer 226-229). The appreciation of rhythm in poetry has a long tradition and is illustrated through not only metre or foot, but also alliteration, assonance, repetitions, descriptions, and metaphorical images. The relationship between sound and time is also revealed in the experience of pause. In musical notation, the term ‘rest’ is applied to define a measurable stoppage − i.e. half-beat or four beats, the term ‘pausa’ is used to define a long rest, and the term ‘pause’ is applied when the stoppage lasts for an imprecise amount of time − i.e. at the performer’s discretion (Kennedy and Bourne 551-602). Auditory Time OED defines ‘pause’ as “a break or rest made for effect, according to the sense, in speaking, reading, […]”. In the general use, it is used to refer to “an act of stopping or ceasing for a short time in a course of action; a short interval of silence or inaction, especially one arising from uncertainty, doubt, or reflection; an intermission; a delay, a hesitation”. Likewise, throughout the thesis, I will use the term ‘pause’ to refer to a stoppage that occurs over the course of an acoustic event or in between the succession of acoustic events. Ear and the Others: Overlaps and Horizons Senses interact with one another and might be transformed by particular circumstances. We might then question to what extent these senses overlap or in what ways they diverge in their respective domain and also if there are any boundaries. The definition of sound-event as an interaction of materials that occurs at a certain time and in a certain place 50 implies that while a stable and still material object belongs to the realm of silence and vision, a moving and active object could potentially belong to both the visual and the auditory realm, where what is seen and what is heard might overlap (Ihde 50). Ihde remarks: The mute object stands beyond the horizon of sound. Silence is the horizon of sound, yet the mute object is silently present. […] Of both animate and inanimate beings, motion and sound, when paired, belong together. Visualistically, sound overlaps with moving being. (50) My analysis of references to sounds and their absence in Heaney’s poetry occasionally overlaps with visual references as well. However, there is an area beyond the overlapping noted above, where listening exceeds seeing or where sight may not enter. Darkness amplifies the presence of sounds. In darkness the ears become more aware of the sounds in the surroundings. However, it is in the realm of the invisible − i.e. the horizon of sight − when through listening we can perceive the presence of the invisible (Ihde 51). 1.4.2 Acoustic Ecology and Soundscape Ecology: Terms and Classifications Scholars of acoustic ecology and soundscape ecology offer useful terms and taxonomy to identify significant features of sounds. In this section, I provide the definition of the technical terms I apply in my analysis of the poems. Soundscape The term soundscape has been used by scholars from a variety of disciplines, from scientific, environmental, social, political and cultural to literary and artistic domains, where soundscape studies can lead to better insight into the research. The first mention of the term appears in Michael Southworth’s 1969 article “The Sonic Environment of Cities” and his exploration of the relationship between the urban soundscape and people’s perception. The term reappears in Schafer’s 1977 book and has remained relatively unchanged throughout the past decades being adopted in more contemporary studies by Bryan Pijanowski, et. al., Jøran 51 Rudi and Almo Farina, for instance, who continue to use the term soundscape to refer to the entirety of sounds being audible in a certain region and at a certain time. The term soundscape may be applied to “any portion of the sonic environment regarded as a field for study”. It can refer to an “actual environment” or an “abstract construction” of sounds, such as a musical composition, a radio or television program, or a stanza of a poem (Schafer 274). Relying on Schafer’s definition, in this study I use the term soundscape to refer to the sounds − and their absence − woven into the world of each poem. Geophony, Biophony and Anthrophony According to soundscape ecologists the soundscape of any given environment is composed of a complex arrangement of sounds produced by geophysical, biological, and anthropogenic sources. This taxonomy of sound-producing sources can help illustrate the dynamics between human beings and the soundscape of their rural and urban environment. In 1987, Bernie Krause introduced the term biophony to describe sounds emanating from non- human sentient organisms and the term geophony to describe sounds of non-sentient natural phenomena, such as wind, thunder and rain. The term “anthrophony” refers to all sounds produced by humans, as well as to sounds produced by stationary and moving human-made objects (qtd in Pijanowski et al. 1214). This classification informs my categorisation of sounds in Heaney’s poetry in this thesis. Keynotes, Signals, Soundmarks and Archetypal Sounds Acoustic ecology offers the classification of keynote sounds, sound signals and soundmarks to distinguish between the sounds important for their “individuality”, “numerousness” or “domination”, respectively (Schafer 9; Rudi 187). This distinction is 52 conducive to identifying values inherent in each sound as shaped by the social and cultural structures. In musical composition, the term keynote refers to the note that identifies the tonality or the arrangement of pitches and/or chords of a particular composition. Although the material modulated around the keynote often obscures its centrality, keynote remains the anchor or fundamental tone according to which everything else takes on its special meaning (Schafer 9). In soundscape studies, keynote sounds are created by the characteristic geological, geographical, topographical and climatic features of a region, such as wind, water, vegetation and the distribution of living organisms (Schafer 272). There is a reciprocal relationship between the type of landscape and its “basic sonic character” (Rudi 187). Keynote sounds “are heard by a particular society continuously or frequently enough to form a background against which other sounds are perceived” (Schafer 272). Keynote sounds of a landscape become “listening habits” and leave a deep and pervasive influence on the moods, character, habits, and as such, lifestyle and culture of a particular community (Schafer 9). They are then the familiar white noises that make individuals feel at home. In fact, as Schafer puts it, “they may have imprinted themselves so deeply on the people hearing them that life without them would be sensed as a distinct impoverishment” (10). Many of the keynote sounds may also possess symbolic connotation for people inhabiting a particular place and might be remembered nostalgically. Since keynote sounds have a direct impact on the mood and lifestyle of the inhabitants of a landscape, their transformation and disappearance have been the concern of ecomusicologists, anthologists and literary writers. Likewise, this thesis defamiliarises the keynotes evoked by Heaney and provides a lens for unfolding their political, cultural and environmental implications. 53 Keynote sounds provide the context for the more foreground signal sounds, which are emitted “with the purpose of catching someone’s attention and being listened to” attentively (Rudi 187). Signals are consciously listened to because they carry a specific message for the listener (Schafer 10). Keynotes, Signals, Soundmarks and Archetypal Sounds Examples could include a bird’s shriek, a dog’s bark, or the rush of a flood, as well as clock alarms, electronic signals from cell phones, crosswalks alarms, church bells, car horns and sirens, which may carry specific semantic meanings or acoustic warnings. In urban soundscapes, such electronic devices are often organised by the controlling institutions into elaborated codes so as to transmit specific messages to the individuals who are able to interpret them correctly. Therefore, as a member of a community one needs to acquire the knowledge to interpret and engage with the shared community signals (Schafer 10). Location-specific sounds, such as the chorus of native birds, the church bell in a small town, or a specific mechanical noise, are referred to as soundmarks. The term is the sonic equivalent of landmark and can be defined as a “sound unique to a specific area or location” (Rudi 187). Soundmarks designate the unique acoustic experience of the community living in a specific place and, thus, establish a unique sense of place often only experienced by the locals. Schafer defines a soundmark as “a community sound which is unique or possesses qualities which make it specially regarded or noticed by people in that community” (10). Both keynote sounds and soundmarks can also fall into the category of signal sounds by marking a specific time or bringing a specific message (Rudi 187). Schafer extends this taxonomy to include archetypal sounds and defines them as “those mysterious ancient sounds, often possessing felicitous symbolism, which we have inherited from remote antiquity or prehistory” (9). The associations related to these sounds often surpass the boundaries of time and place to create sound symbolism. The experience of 54 archetypal sounds can often stir universal responses and connect listeners “with ancient ancestral heritages, providing continuity at the deepest levels of consciousness” (47). archetypal sounds can often stir universal responses and connect listeners “with ancient ancestral heritages, providing continuity at the deepest levels of consciousness” (47). Sound, Noise or Music In 1966, Schaeffer coined the term ‘sound object’ to refer to “an acoustical object for human perception and not a mathematical or electro-acoustical object for synthesis” (127). The sound object is “defined by the human ear as the smallest self-contained particle of a SOUNDSCAPE [emphasis in the original]” and may be referential − i.e., a car, a bell, or a drum. In sound studies, it is considered primarily as a phenomenological sound formation independently of the referential qualities of the sound event (Schafer 274). Throughout this thesis, I use the term ‘sound’ to refer to anything audible by the human auditory system and the term ‘silence’ to the absence of any such sound. However, as I explain in the following chapter, our interpretation of sounds and silences can be mediated through culture, religion, politics as well as personal preferences. A sound can be interpreted as noisy or musical. The arbitrariness in the interpretation of sounds, noise and music − and even silence − is best reflected when we explore the history of hearing (see 1.2). Noise, in its contemporary sense as a technical phenomenon, varies greatly from its initial definition and application. From an ecological point of view, sound and noise constitute the same phenomenon; however, the differentiation is greatly subjective. As Jacques Attali notes, “noise, then, does not exist in itself, but only in relation to the system within which it is inscribed: emitter, transmitter, receiver” (26-27). In music studies, psychology, noise abatement practices, media theory and electronics, in particular, the term ‘noise’ refers to any random, unpredictable, disagreeable, unwanted and undesirable sound that may interfere with the detection of a target or signal sound or have adverse effects on 55 health (Seidman and Standring 3731-3732). Throughout my analysis, I apply the standard definition of the term noise as “any sound that is undesired by the recipient”.32 Music − especially in ancient societies − originated in a collective longing for order and harmony (Attali 20). A practical and conventional definition of music includes sounds that are, among other things: (1) “humanly organised”, (2) “organised with a goal”, and (3) can be identified as “a recognisable aesthetic entity” (Godt 84). From an ecological perspective, however, music is not merely the sound of musical instruments but could potentially be that of other human-made objects and ultimately the natural environment. Sound, Noise or Music As John Cage has declared: “Music is sounds, sounds around us whether we’re in or out of concert halls” (30). In this thesis, too, I use the terms music and musical more broadly and to refer to any sound that is perceived as harmonic and pleasant. 32 Most definitions of acoustic terms have been internationally standardised. See IEC 801-21-08 at y www.acoustic-glossary.co.uk. 32 Most definitions of acoustic terms have been internationally standardised. See IEC 801-21-08 at www.acoustic-glossary.co.uk. www.acoustic-glossary.co.uk. Hi-fi and Low-fi Independently from the perspective with which one enters the domain of soundscape analysis, key terms of hi-fi and low-fi are applied to describe its density or transparency. A hi-fi or high-fidelity soundscape has a favourable signal-to-noise ratio. It is one in which the ambient noise level is low, sounds overlap less frequently and discrete sounds can be heard clearly (Schafer 272). In a hi-fi soundscape, the listeners can experience perspective – foreground and background sounds − and expand their auditory horizon and practise “distant hearing” (43). Soundscapes with an unfavourable signal-to-noise ratio are known as low-fi or low- fidelity soundscapes. In such environments, signals compete for their space in the same niche, “resulting in masking or lack of clarity” (Schafer 272). In a low-fi soundscape, “[p]erspective is lost. […] there is no distance; there is only presence. There is cross-talk on all the channels, 56 and in order for the most ordinary sounds to be heard they have to be increasingly amplified” (Schafer 43). In a busy city centre, for instance, one frequency spectrum dominates and, therefore, distant and low-frequency sounds cannot be discerned (Rudi 187). 1.4.3 Wintering Out: A Telling Case Study Wintering Out presents a variety of cultural, political, societal, linguistic and personal soundscapes. The poems in the collection were largely composed during his sabbatical year in Berkeley and in the months following his return to Belfast. Wintering Out appeared when the poet was in his early thirties facing the financial and domestic demands of life as a young parent as well as the daily professional routines as a lecturer in Queen’s. He had also recently returned to Northern Ireland from a sabbatical year (1970-71) in Berkeley. The year had brought him into contact with a new cultural, intellectual, geographical and climatic environment. Upon his return, Heaney witnessed the introduction of internment without trial, the Bloody Sunday shootings in Co. Derry and Bloody Friday bombings in Belfast, and the constant deterioration of political and sectarian relations in Northern Ireland, the Republic of Ireland and the UK. Wintering Out was published in 1972 by Faber and Faber. With the publication of his third collection, Heaney felt confident in his vocation as a poet and decided to leave the social and professional pressures of life in Northern Ireland behind him and move to Co. Wicklow to fulfil his life-long intention of being a full-time writer, as he commented: “I had passed the stage where just ‘writing poems’ was enough. I had passed the stage of probation and felt confident of vocation […], it was time to lose the nine-to-five life and try to find a poetic life” (SS 127). With the experience of new perspectives, new soundscapes and a sharpened sense of perception, the poet was ready to make changes in the direction of his life, to winter out. The 57 title comes from the poet’s childhood memories of the old farming customs of wintering cattle out in pastures around Co. Derry, but also alludes to the deteriorating social and political condition of Northern Ireland in 1972. As Heaney noted, the title does not guarantee a forthcoming greenness and prosperity, but still hints at hope, survival, continuance and a change for the better: “No spring was being promised, but I still didn’t think of the title as despairing” (SS 121). 34 See www.youtube.com/watch?v=WT-dub5v4YA (17:40-18:35). 33 See jrnl.ie/5065553. 33 See jrnl.ie/5065553. 34 See www.youtube.com/watch?v=WT-dub5v4YA (17:40-18:35). 1.4.3 Wintering Out: A Telling Case Study In an interview published in November 1972 in the Cork Examiner, he commented: “If we can winter this one out, we can summer anywhere”.33 To winter out, for him, is to get to the end of a tough situation, to reach the other side, and be better for having been through it. The line was posted again in 2020, on the Twitter account of the Listen Now Again exhibition, and has since become a well-known refrain in Ireland both on social media and in the real world appearing on walls and banners to offer messages of optimism and hope during the course of the Covid-19 pandemic. In sum, the collection charts a transitional stage in the poet’s personal life and career bringing together a wide range of new challenges, perspectives, expectations, worries and tensions that he undergoes. It is placed at a turning point of his poetic development when one of the most crucial aspects of his early poems, i.e. the uncertain, more tentative cast of his mind gradually slipped away, to be replaced by a growing sense of certainty about his poetic voice. This drive toward a more certain poetic voice, amidst the uncertain years of the Troubles, must have given the poet the urgency to move to Wicklow. Looking back at his career, in a 1996 interview, he noted: The move I made in 1972, […] had, I noticed clearly, nothing to do with the political situation. It had to do with an inner development, an inner necessity in myself as a writer, as a poet to change my life […]. It was the following of an inner compass which had to do with my own imaginative and psychic needs.34 The acoustic images of Wintering Out are born out of the everyday memories of the cattle in winter fields during the 40s and 50s, the warmth and tolerance of his family towards 33 See jrnl ie/5065553 cattle in winter fields during the 40s and 50s, the warmth and tolerance of his family t 58 their Protestant neighbours, the poet’s anxieties as a new parent, his frustrations with Irish predicaments, the longing for his cultural and linguistic roots, and his transition to a new social and personal consciousness as a Northern Irish poet. 1.4.3 Wintering Out: A Telling Case Study Wintering Out is one of Heaney’s most compelling volumes for reaching back to the inner world of the poet himself, as well as for reaching out to other soundscapes in time and space and transposing them into his poems. Neil Corcoran remarks that Wintering Out is the “seminal single volume of the post- 1970 period of English poetry” (182). He argues that Wintering Out − and North − helped introduce “a lexicon and a register of pronunciation distinct from received or standard English” and continues: “in taking etymology itself as theme and preoccupation, these volumes may also be read as paradigms of the decisive shift in cultural consciousness after the 1960s” (196-197). Richard Rankin Russell notes that Wintering Out is one of Heaney’s most significant volumes for containing the Northern Irish poet’s reclamation of his “linguistic landscape” in Co. Derry (56). Heaney’s place-name poems probe particular locales within the landscape and history of Ireland. In the “languagy poems”, he tries to balance between recovering the lost landscape of the “hidden Ulster, the Uladh of Doire Cholmcille” and his debt to the English language and literature. Heaney is determined to subvert the Protestant myth of belonging in Ulster that would exclude Catholics and Irish culture from their province (56-57). Thomas C. Foster notes that “scattered throughout Wintering Out is evidence of a new approach to poetry, a heightened sensitivity to the historical and political implications of many of Heaney’s interests and preoccupations” (35). Heaney himself said that with the escalation of the Troubles in the late 1960s, “the problems of poetry moved from being simply a matter of achieving the satisfactory verbal icon to being a search for images and symbols adequate to our predicament” (PO 56). A study of the references to sound and silence is important in showing this new approach in his poetry. 59 Chapter 2: Introduction In 1992, three years before Heaney was awarded the Nobel Prize, a Guardian International article revealed that “Shakespeare takes the second place to Seamus Heaney as the writer whose name appears most in English Literature courses at polytechnics and colleges of higher education [in Britain]” (Meikle 1). Since the publication of his first major poetry collection, Death of a Naturalist, more critical and public attention has been given to Heaney and his work than any other contemporary Irish poet − perhaps more than any other poet in the English-speaking world (O’Donoghue 2). Heaney’s talent and popularity have generated a large volume of critical materials − reviews, interviews, readings, conferences, articles, bibliographies, monographs and collections of essays together with a wealth of artistic responses − paintings, musical settings, sculptures, audio and video recordings − from all over the world. Heaney’s poetry, translations and prose have been examined by numerous scholars for traces of the voices and perspectives of other poets, manifestations of such genres as pastoral and elegy, as well as themes of liminality, memory, archaeology, mythology and music. To attempt a comprehensive review of the critical publications on Heaney’s work would be a nearly impossible task and goes beyond the scope of the present study. The essays reviewed in this chapter are selective but representative of the scholarly writings whose focus is of interest and significance to this thesis. I narrow this scope further to the studies on his poetry in general, and on Wintering Out in particular. Where appropriate, I benefit from earlier studies of Heaney’s work, but to avoid repetition, in this literature review, I mainly focus on a number of more recent publications. 60 Since the beginning of his career, Heaney’s reputation has emerged both as one of the most gifted poets of the English-speaking world and as an intelligent reader of poetry. As Burris has noted, Heaney undertook the task of reading with a “native sensibility” and “indigenous feeling” for words (59-60). Introduction In order to set the ground for the analytical chapters of the thesis, in this chapter, I (i) explore studies on the voices behind Heaney’s own, literary figures from the classics to broader European and anglophone world, for their influence on Heaney’s conceptualisation of place and his relationship with it, as well as on the way they shaped or steered his poetic imagination; (ii) review the studies that examine Heaney’s position in relation to various landscapes, from rural to urban. I attend to the themes of place, identity, liminality, as well as more recent works on the concepts of landscape, environment and ecology; and (iii) explore the studies that examine the concepts of listening, music, sounds, voice and silence in Heaney’s work. 35 The title of this section is inspired by a quote from Heaney’s Nobel Lecture (1995). 2.1 The Persuasive Voices Behind: Poets and Poetics35 Heaney’s poetic language is widely recognised for its vibrant diction, intense imagery and application of powerful symbols, metaphors and allusions that open a gate to his cultural identity and global history. Hardy’s 2007 study shows a detailed list of literary allusions in Heaney’s work. Yet Heaney’s relationship with his literary precursors and contemporaries goes beyond allusions and has been the concern of many scholars. If one looks back over the trajectory of Heaney’s poetic career, it is clear that Virgil, Dante, along with Mandelstam, Yeats, Miłosz, and others have “completely exemplary force” for Heaney (FK 172). Moreover, Heaney’s extensive use of the classics has encouraged many scholars to explore the affinities between his work and the work of Greek, Roman, Gaelic and Anglo-Saxon ancestors. His oeuvre has also been studied for traces of his American and European contemporaries. The following pages of this critical review focus on a selection of studies 35 The title of this section is inspired by a quote from Heaney’s Nobel Lecture (1995). 61 that have examined the literary influences on his perception and conceptualisation of places and, thus, their soundscapes. Heaney encountered Virgil’s Aeneid Book VI and IX in school, but his engagement with Virgil began more earnestly and consciously later (SS 296). The contextualisation of pastoral and epic in his poetry has attracted the attention of scholars. The presence of Virgilian – and also Theocritean – pastoral in Heaney’s early work presents itself as a natural process: Having grown up on the family farm, in Death of a Naturalist and Door into the Dark, Heaney often returns to his Irish bucolic landscape with the sadness of a child who comes to new realisations about life. Yet even those early pastorals are by no means idyllic and overlap with the elegiac mode: in ‘Digging’, for instance, he declares he cannot follow the tradition of his fathers; in ‘Blackberry Picking’, he illustrates rotting and death; and in ‘Death of a Naturalist’ he expresses the loss of innocence. Henry Hart remarks, “the celebration of rural securities collapses as he grows up and realises the antipastoral nature of that landscape and the society that lives within it” (9-13). Likewise, Bernard O’Donoghue states that although eclogue appears as the central form in his later poetry, as in Electric Light, it strikes dark notes about the years to come even more explicitly (117). 2.1 The Persuasive Voices Behind: Poets and Poetics35 Ruben Moi notes how Heaney revisits classical models in order to reflect on his own position in literary history and theory. In his opinion, Electric Light in particular reflects Heaney’s “canonical self-awareness” and his attempt at transcending his measured time and place (174). This concern is echoed through the volume with the poems juxtaposing the wonders of births and the commemorations of losses, as well as through Heaney’s contemplation of the contemporary socio-political conditions of Ireland within the frames offered by Virgil’s eclogues, Shakespeare’s plays and the many allusions to Macbeth. Moi views this attempt by Heaney as “a canonical recontextualisation” and writes: “Electric Light appears as a canon-building project, in which Heaney both revisits and renews his own 62 writing” (Moi 174). Throughout his poetic career, Heaney has been in search of what poetry can and should do in response to the harsh realities of life. Rachel Falconer traces the constant presence of the Virgilian katabatic journey in Heaney’s artistic development. From digging in Death of a Naturalist through to exhuming bodies from peat bogs in Wintering Out and North and probing the ground and listening to the ghosts in the subsequent volumes, as Falconer notes, the poet is in search of “inspiration within the recesses of his memory and a home ground” (431-432). A recent study by Ian Hickey draws similar conclusions by adopting another theoretical lens. Hickey, too, examines the haunting presence of Virgil in the later poetry of Heaney, but his starting point is Derrida’s Spectres of Marx (1994), in which the repetitive nature of history is explained through its haunting spectre (48). The presence of Virgil – and classical literature and history, in general – not only adds a sense of universality to Heaney’s work, but also revisits the pastoral and politics of contemporary Northern Ireland from a more stable standpoint, one whose spectre, like that of Virgil’s, lives on for eternity (26-28). There has been similar critical interest in Dante’s influence on Heaney’s oeuvre and his representation of the contemporary political context. According to Joseph Heininger, Heaney’s work, in the last twenty-five years of his poetic career, shows strong Dantean influence. Field Work and Station Island, in particular, indicate the poetics and habits of mind cultivated by Dante. In an interview with Robert Hass, Heaney clarifies that his approach to translation relies on different motives, of which ‘raid’ best describes his approach to Dante and ‘settlement’ describes his approach to Beowulf. Raid, he explains, is slightly predatory, “the translator hears something in the other language and says, ‘I would like that, that sounds right, I need that’”. In settlement, “the translator enters an oeuvre and colonises it but also stays with it to interact and make mutual adaptations” (1-2) , g , , y with it to interact and make mutual adaptations” (1-2). 36 In an interview with Robert Hass, Heaney clarifies that his approach to translation relies on different motives, 36 In an interview with Robert Hass, Heaney clarifies that his approach to translation relies on different motives, of which ‘raid’ best describes his approach to Dante and ‘settlement’ describes his approach to Beowulf. Raid, he explains, is slightly predatory, “the translator hears something in the other language and says, ‘I would like that, that sounds right, I need that’”. In settlement, “the translator enters an oeuvre and colonises it but also stays with it to interact and make mutual adaptations” (1-2). 2.1 The Persuasive Voices Behind: Poets and Poetics35 While the episode of ‘Ugolino’ can be found in Inferno, the model for the pilgrimage sequence and encounters with familiar ghosts in ‘Station Island’ is Purgatorio (Heininger 50). Heaney translated Dante in the intense socio-political atmosphere of Northern Ireland and amidst “Dirty” protests in the H-blocks in Long Kesh. For him, the harrowing experience of Republicans hunger striking in prison is a reminder of the harrowing starvations of Ugolino and his children in Inferno. Heininger maintains that Heaney's adaptations of Dante’s work as a series of aesthetic choices provide him with “a more open 63 and self-scrutinizing poetic practice” and “a political stance in which self-acceptance triumphs over rage, timorousness, and guilt” (63). Daniela Panzera’s more recent study focuses on Heaney’s fascination with Dante’s language and concern for themes such as land and politics. Heaney admires Dante’s profound devotion to his local culture and vernacular, but is principally fascinated by the medieval Italian poet’s ability to transcend “ethnic boundaries and create an increasingly cosmopolitan poetry” (Panzera 200). It is Dante’s creation of “a unitary poetic language” that becomes fundamental to the Irish poet and reappears, in particular, in his place-name poems in Wintering Out. Both ‘Broagh and ‘Anahorish’ are the English transliterations of the original Irish ‘bruach’ and ‘anach fhíor uisce’ (O’Brien 16). Referring to these poems, Panzera comments that Heaney displays the willingness to reconcile both the English and the Irish traditions by meditating on the semantic elements of both cultures, in the same way that Dante believed the creation of a unitary poetic language would promote national solidarity (200). If Heaney raided Dante, he settled with Beowulf for 15 years.36 Heaney’s knowledge of the Anglo-Saxon poem can be traced back to his undergraduate days. One aspect of Old English poetry that influenced Heaney from the beginning is its highly alliterative language and diction. Critics have been meticulously examining the Irish poet’s choice of word in his translation of it. Howell Chickering condemns what he calls the appropriation of the text, calling it pejoratively “Heaneywulf” (160-178). Yet Seth Lerer sees the “Irishness” of Heaney’s translation as a way to transcend his postcolonial relationship to England and to let philology reverberate away from empire and colony (13-14). Alison Finlay celebrates the marriage of Anglo-Saxon culture and modern Irish concerns in Beowulf (136-154). 2.1 The Persuasive Voices Behind: Poets and Poetics35 Karlo Megec considers Heaney’s attempts at vernacularizing and localizing the language as a very 64 “genuine way of translation” (25-26). Heaney’s long settlement in the text, in fact, allowed the Anglo-Saxon poem the chance to enter his poetic language and broaden his sense of historical heritage. Heaney’s Hiberno-English version of Philoctetes’ has also received contradictory reviews from the critics. For instance, while Paul Turner scrutinises Heaney’s The Cure at Troy for the general “ungreekness” (121-122), Marianne McDonald and Michael Walton admire Heaney’s contextualisation of the classic text (70-71). Alan Peacock discusses The Cure at Troy as a major point in Heaney’s development of a public poetic voice (233-255). Marylinn Richtarik pursues the question of its relevance to the contemporary Irish context. She examines the Chorus speech, in particular, as Heaney’s attempt to encourage the audience to believe in what seemed utopian in 1990, but what would come to be referred to as the Northern Ireland peace process (98-112). The influence of Eastern European poetry on anglophone poetry during the Cold War was deep and wide-ranging. Heaney wrote essays on the Polish poets Czesław Miłosz and Zbigniew Herbert, the Russian poets Osip Mandelstam and Joseph Brodsky, and the Czech poet Miroslav Holub, among others, paying particular attention to Mandelstam and Miłosz in his poetry. The fact that the Northern Irish poet has been influenced by Eastern European poetry is beyond dispute. The way in which he has been influenced and the degree to which he has responded to them have been the focus of many critics. According to Clare Cavanagh, in his later poetry Heaney turns repeatedly to the voices of poets from Eastern Europe to reframe the relationship between his poetry and the social and political obligation (105). p g y ( ) 38 For Heaney’s views, see ‘The Impact of Translation’ (GT 36-44). 37 Heaney explains the reason for finding their voices compelling: “because there is something in their situation that makes them attractive to a reader whose formative experiences have been largely Irish” (GT XX). p 38 For Heaney’s views, see ‘The Impact of Translation’ (GT 36-44). 2.1 The Persuasive Voices Behind: Poets and Poetics35 He remarks that Heaney is united with Russian Mandelstam and Polish Miłosz, who have likewise witnessed victimisation and deprivation of people in the history of their nations and who take as their mission “not merely to bear witness to history, but to participate in its 65 creative reshaping as national myth and legend” (105-106).37 Justin Quinn explores the depth and dimensions of Eastern European poetry influence on Heaney’s oeuvre, suggesting Heaney’s engagement with Eastern European poets was, paradoxically, at once superficial and profound: superficial in the sense that Heaney’s access to these texts was confined to translated versions, and profound in the sense that “these poets provided him with new ways to respond to the pressure of politics on poetry” (93).38 Magdalena Kay’s more recent study traces the subtle manner in which Heaney acknowledges the value of Eastern European poetry. According to Kay, for Heaney, the power of Eastern European poetry does not inhere in political utility alone. Instead, it relies greatly on the fierce manner in which these poets use such abstract words as faith, spirit, justification and love. In doing so, she also indicates the ways Eastern European poets have encouraged Heaney to recalibrate his relationship with words in a broader sense: “Heaney has always insisted on the word’s ultimate ability to release energies that act within and upon both the individual and the community. The word does not simply mediate but creates, performs, makes present (18-19). Heaney’s poetry also reflects his awareness of the broader international context than just Europe. In Poetry and Posterity (2000), Edna Longley suggests Heaney and several of his contemporaries show a strong pull towards the American dream drawing them eventually to prestigious US universities (253). Daniel Ross argues that the years of teaching at Harvard have left a much more elemental American mark on Heaney’s poetic career than merely broadening his circle of literary figures. Having lived through the years of the Troubles in Northern Ireland, Heaney had become able to see America in a special light: “a former British colony that has distinctively found its own identity and literary tradition within the confines of the English language” (Ross 253). 39 In ‘Belfast’, Heaney discusses his relationship with English language and literary tradition, saying: “I speak and write in English, but do not altogether share the preoccupations and perspective of an Englishman. I teach English literature, I publish in London, but the English tradition is not ultimately home” (PO 34). 40 y y p y y , , y writings as “the best Irish literary criticism since Yeats” (2). In 2008, O'Donoghue comments on Heaney’s lifestyle who chose to be a busy “career-teacher of literature as well as a writer” rather than “a man of letters” (2). Patrick Crotty compares him to Yeats and highlights Heaney’s “warm and accessible demeanour”, identifying it as a major factor in his reception (38). 40 For instance, Robert Lowell famously calls Heaney “the most important Irish poet since Yeats” (19). Jon Stallworthy, Steven Matthews, and Corcoran examine Heaney’s poetry for elements of Yeatsian influence. David Wheatley’s 2008 study pays tribute to Heaney as a critic and, in 1995, Foster refers to Heaney’s critical g , p , g y ( ) For instance, Robert Lowell famously calls Heaney “the most important Irish poet since Yeats” (19) , y y p p ( ) worthy, Steven Matthews, and Corcoran examine Heaney’s poetry for elements of Yeatsian influenc d Wheatley’s 2008 study pays tribute to Heaney as a critic and, in 1995, Foster refers to Heaney’s critic 41 The title of this section is inspired by Heaney’s ‘From the Republic of Conscience’ (HL 14). 42 p y y p ( ) 42 Heaney comments about the concept of place in his essay ‘The Sense of Place’(PO). 2.1 The Persuasive Voices Behind: Poets and Poetics35 Likewise, Suwa presents the counter argument for Bloom’s theory by highlighting the conflicts and complexities in the poet’s religious, cultural and historical contexts (49). Suwa identifies three distinct phases in Heaney’s work during which he responds differently to Yeats’s influence. The period between the 1960s and the 1980s marks a phase of “ambivalent apprenticeship” when Heaney’s relationship with Yeats was more polemical than is generally known. In this early period, Heaney often refers to Kavanagh and Joyce as his literary influences. Suwa recognises a turning point from the mid-1980s to the early 1990s, when Heaney develops a broader sense of literary borders and, thus, when a discreet re-appraisal of Yeats begins. The years after the mid-1990s designate a movement to the next phase of world history, the era of globalisation and Heaney’s reconciliation towards “the other” in Ireland (50-57). 2.1 The Persuasive Voices Behind: Poets and Poetics35 American writers have given the Northern Irish poet a model for making the language of his oppressors his own, an element that, as 66 noted before, is evident as early as 1970s in such poems as ‘Fodder’, ‘Broagh’ and ‘Anahorish’ in Wintering Out, when Heaney includes some Irish words and place names.39 Ross also believes that the influence of Eliot and Frost, in particular, is more evident in Heaney’s later publications. He identifies the two great hungers undergirding Heaney’s poetic career: the hunger for origins, which has enabled Heaney to recapture an original sense of encountering with an ancient language and, thus, to escape Englishness. The hunger for transcendence began in the 1980s in North and Field Work, with Dante as the key figure. But, in turning to Dante, as Ross argues, Heaney was also following in the footsteps of Eliot, hoping “to grow up to the end of life” (26). Frost, too, offered him a new perspective to re- consider the values of balance in life and in poetry. Beginning with Seeing Things, Heaney described the political Troubles of Northern Ireland as “a too-consuming passion” (Ross 92- 95). Ross believes that the American influence encouraged Heaney to broaden his vision and, thus, to transcend the condition of “boundedness” (95). Yeats is regarded as the father of modern Irish poetry who elevated the state of Irish literature to the international level. Since the publication of his first poetry collection, Heaney has been frequently compared to Yeats.40 Patrick Ryan compares Yeats’ and Heaney’s political engagement (21), while Longley and Tomoaki Suwa address Bloom’s theory of the anxiety of influence in Heaney’s relationship with Yeats (194; 49). According to Ryan, Yeats and Heaney are the right examples for proving that “there are no political neutrals” (21). While Yeats preferred to avoid a deliberate and direct political theme in literature, Heaney was more consistent than Yeats in facing up to the political realities around him (21-24). 67 Longley’s 2013 study touches on a similar topic by refuting Bloom’s theory and bringing up the Irish context as a case in which religious and political factors interfere with such influences. The relationship between Heaney and Yeats is complicated by Heaney’s admiration of Yeats’ use of form and the post-colonial consideration of Yeats as a “taboo model” for a Northern Irish poet of Catholic background (209). of the region, but he also emphasises the desire to move beyond them: [T]he whole population are adepts in the mystery of living at two places at one time. Like all human beings, of course, they would prefer to live in one, but in the meantime they make do with a constructed destination, an interim place whose foundations straddle the areas of self-division, a place of resolved contradiction, beyond confusion. (RP 190) 43 The divisions in the landscape of the region can be traced back to the contacts with largely Protestant Scottish and English settlers from 1608, who wrestled economic, political and religious power from the Catholic minority. Up to the present day, the region has suffered from the effects of sectarian violence, culminating in the years that led up to the publication of Wintering Out and North (Whyte 342–343). y p p g ( y ) 44 In ‘Frontiers of Writing’, Heaney describes the impact of the riven landscape on himself and all the of the region but he also emphasises the desire to move beyond them: 44 In ‘Frontiers of Writing’, Heaney describes the impact of the riven landscape on himself and all the residents of the region, but he also emphasises the desire to move beyond them: y p p g ( y ) 44 In ‘Frontiers of Writing’, Heaney describes the impact of the riven landscape on hi of the region, but he also emphasises the desire to move beyond them: Heaney maintains: I like to remember that Dante was very much a man of a particular place, that his great poem is full of intimate placings and place-names, and that as he moves round the murky circles of hell, often heard rather than seen by his damned friends and enemies, he is recognised by his local speech or so he recognises them. (PO 136-137) 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 Heaney’s relationship with the concept of place has been examined in relation to the themes of exile, liminality, displacement and identity, as well as through the theoretical lenses of landscape theories and ecocritical theories.42 These studies highlight the centrality of landscape and place, as a leitmotif of Heaney’s work, and indicate the various ways in which he perceives his relationship with his surroundings. I start this section with the review of the studies on exile and liminality, as early concepts of Heaney scholarship, and proceed to more recent studies with ecocritical concerns, which explore the implications attached to the 68 concept of place and landscape. I end this section with a review of the discussions on major biophonic and geophonic elements of the Irish soundscape − i.e. the bogland, the wood, stones, birds and some of the other animals. The understanding of these concepts underlies Heaney’s illustration of his sense of place and his relationship with the surrounding landscape through sounds. The political situation in the late 1960s and early 1970s, when Heaney was writing the poems in Wintering Out, was an inescapable barbaric horror (Mahon 115), leaving people divided between “Britain’s Ireland” and “Ireland’s Ireland” (RP 188).43 Early plantation settlement in the seventeenth century and the sectarian divisions and tensions that followed it have left a deep impact on the landscape of Ireland, both literally and ideologically. The experience of the divisions and the quest for defining the concepts of identity and belonging have impacted the literary enterprises of the territory, including Heaney’s. In Writing Home (2008), Elmer Andrews notes that there is a particular dimension to Irish writers’ preoccupation with their land: “Yeats and Lady Gregory’s Co. Galway, J. M. Synge’s Aran Islands, Patrick Kavanagh’s Co. Monaghan, John Hewitt’s Glens of Antrim, Seamus Heaney’s Co. Derry, John Montague’s Co. Tyrone, Michael Longley’s Co. Mayo” (1).44 Exile, displacement, liminality and in-betweenness are some of the early concepts with which scholars have approached and interpreted Heaney’s relationship with his homeland. Such debates are often contradictory in their interpretation of liminality. While some consider it as an empowering element, others view in-betweenness as a restricting position. Jonathan Allison refers to liminality as a frequent topic of Heaney criticism, and 69 comments: “To regard Heaney as a poet of self-division has become a critical commonplace. 45 Heaney maintains: 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 His poetry is said to mediate between, oscillate between, chart a course between, struggle between, and voice the conflict between certain opposing choices” (187). Considering Heaney’s Catholic upbringing in an officially Protestant country, as well as his persistency in placing himself in between the binaries of Irish/Englishness, North/South and classics/contemporary, Heaney’s experience of in-betweenness is undeniable: “Two buckets were easier carried than one. / I grew up in between” (HL 5). Dominic Manganiello counters Allison’s interpretation of displacement as an uncertain and struggling state by arguing that liminality is, in fact, a fundamental element of any creating process (101-103). Heaney himself notes, “a writer cannot dwell completely in origin − Origin is almost Eden, you know. You have to leave Eden and get the division; the loss of Eden, the memory, is one of the ways writing occurs” (qtd in Foster 139.). Manganiello highlights the sense of self-exile that the “Northern Irish Dante” shares with his Italian precursor and exemplar. An early point of contact between the two poets occurs in the rootedness of their literature in the local and the vernacular (see 2.1). Heaney finds the origin of all writing in the distance the “inner émigré” (N 68) travels to retrieve his cultural memories.45 In his place-name poems, in particular, it is in this journey back “[i]nto the heartland of the ordinary” (ST 7) − to rocks, bogs, the wool trade, buses and bicycles from his homeland − that Heaney reaches the embedded meanings and rooted memories (Manganiello 101-103). Likewise, in “Poems without Frontiers” (2007), Rankin Russell argues that, for Heaney, in-betweenness was an enabling and empowering meditative state (26). In ‘Digging’, for instance, the liminal positions between the domestic space and exterior world as well as 70 between the present and the memory of his grandfather enable Heaney to project his future as a writer. The composition of the poems in Station Island is another evident illustration of Heaney’s commitment to this poetic approach. The tripartite structure of the volume reflects a correspondence to Dante’s Divine Comedy: opening with ‘the Underground’; entering the middle passage of ‘Station Island’; and, finally, encountering the poetic muse, ‘Sweeney Redivivus’. The titular poems are also carefully and purposely placed in the middle section of the entire collection dramatizing a purgatorial phase in which the poet-narrator encounters various dead personages (Rankin Russell 30-33). 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 Jonathan Hufstader refers to this tripartite structure as Heaney’s “ritual procedure” (61–62). In his view, this structure is followed in many of his other poems where “[there is an] entrance rite, [a] central action, and [then] the subject’s emergence from the ritual in a new state of mind” (61–62). Rankin Russell concludes the Janus-faced poet “is straddling the threshold between what is and what can be. His poetic invitation to us is to join him at the frontier of writing, where hope, tempered by reality, awaits us” (39). A more novel branch of Heaney scholarship has been fashioned by studies that approach the concept of place and landscape from a broad range of perspectives, from natural and social sciences to humanities and arts. Landscapes are composed of complex physical, chemical and biological systems, which change over time and whose change determines the transformations of the soundscape. They are shaped by human societies who inhabit them and base their existence and habits on the natural features of their landscapes. “We are dwellers, we are namers, we are lovers”, says Heaney, “we make homes and search for our histories” (PO 148-149). In the next paragraphs, I exemplify how Heaney’s way of redressing the fractures and divisions in the socio-political landscapes of Ireland has been examined by Heaney scholars. 71 Drawing on the key elements in landscape theory, Sukanya Basu presents the concept of “poetic landscape” and defines it as a creative domain in which the Irish poet redresses this liminal state (29).46 He maintains that poetic landscapes are “verbal entities because their visual reality is truly realised only through the power of the poetic utterance” (29). According to Michael Harkin, Heaney is aware of the impacts of colonisation and divisions on the landscape of Ireland and views it as not merely a “passive recipient […] of human agency, but” a “subject” that needs to be given voice (50). Heaney’s poetic landscapes are the space where he fuses both “material places” and “imagined spaces” in order to uncover and vocalise the hidden energies of landscape through the power of his poetry (Harkin 50). p 47 In landscape theories, ‘place’ is applied to refer to a space invested with meaning in the context of power relations (Cresswell 12; Basu 28) and the term ‘landscape’ to refer to “a set of relational places” (Tilley 44) or “a wide range of different interlocking places which interact with one other to generate a complex web of meaning” (Basu 29). 46 See Christopher Tilley’s A Phenomenology of Landscape (1994), John Wylie’s Landscape (2007) and Tim Cresswell’s Place (2004), for a detailed discussion on the theories regarding the concepts of ‘place’, ‘space’ and ‘landscape’. 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 Therefore, whereas in essays − like ‘Mossbawn’, ‘Belfast’, ‘The Sense of Place’, ‘Place’ and ‘Displacement’, ‘The Place of Writing’ and ‘Frontiers of Writing’ − Heaney unravels the social and cultural implications attached to his places and landscapes,47 in his place-name poems, language is intimately tied to the landscape in order to create a geographical myth of Irish identity (50). Heaney explains that, as an Irish poet, his “quest for definition” is to be “conducted in the living speech of the landscape” he was born into (PO 37). Thus, the soundscapes of Heaney’s childhood are placed at the centre of his Irishness and poetic agenda. The river Moyola, the bogs, trees, animals, farms, and fields become personal and collective touchstones, to which Heaney returns to, as Basu puts it, “renew and sustain his poetic imagination” (20). Walt Hunter too highlights the centrality of landscape and place as a leitmotif of Irish poetic creation, but one that is intensely politicised and globally oriented. From Wilde’s ‘The Famine Year’ (1848), Yeats’s ‘The Stolen Child’ (1889), Kavanagh’s ‘The Great Hunger’ 72 (1942), through to French’s Broken Harbour (2012) and one of Heaney’s final poems, ‘Banks of a Canal’ (2013), Hunter argues the landscape of Irish poetry is haunted by violence of dispossession and uprooting (19-26). Drawing on Heaney’s essay, ‘Place and Displacement’, he comments: “What he names as a ‘displaced perspective’ amplifies rather than mutes the poem’s effects” (27).48 This sense of displacement is dramatised in Heaney’s landscape poems, when he attempts to open a liminal space for poetry in a politicised context, by preventing its “pigeon-holing” (Hunter 27) in one of two alternatives, to use the poet’s terms, of “deliberately provocative” or “culpably detached” (Heaney 29). The contemporary Irish landscape poetry’s obsession with local places is not for the purpose of commodifying and marketing their Irishness, but in order to respond to the process of globalisation from an exilic and resistant position (Hunter 28). Ross Moore examines the representation of natural landscapes in some of the poems that feature Heaney driving in the countryside (71-72). ‘Postscript’, for instance, begins with a sense of motion urging the reader to make some time and drive along the coastal line in Co. Clare and then continues, “useless to think you’ll park and capture it” (SL 70). In Place and Displacement , Heaney speaks of the liminal position as enabling and empowering: The poet is stretched between politics and transcendence, and is often displaced from a confidence in a single position by his disposition to be affected by all positions, negatively rather than positively capable. This, and the complexity of the present conditions, may go some way to explain the large number of poems in which the Northern Irish writer views the world from a great spatial or temporal distance, the number of poems imagined from beyond the grave, from the perspective of mythological or historically remote characters. (130-131) 48 In ‘Place and Displacement’, Heaney speaks of the liminal position as enabling and empowering: p , y p p g p g The poet is stretched between politics and transcendence, and is often displaced from a confidence in 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 Looking closely at the language and the imagery, Moore suggests the effect is to create a sense of transience and liminality. The poem signals a fundamental shift in Heaney’s approach to the natural environment and this shift is reliant on poetic procedures which Heaney had come to trust early in his career. ‘Postscript’ arches back to the opening of ‘The Peninsula’ in the way it maintains the perspective of a driver on a solitary trip through the landscape of Ireland: “When you have nothing more to say, just drive / For a day all around the peninsula” (DD 11). Moore comments that both poems are geographically and temporally grounded, giving a 73 sense of direction to the reader, but for all its determination and forward look, the drive in ‘Postscript’ retains something of the melancholic freedom of ‘The Tollund Man’ (WO 36- 37). Both poems share some of the unstable attributes of an in-between protagonist but reveal a significant register in the tone of the poet. ‘The Peninsula’, he writes, “hopes to “uncode” truths through attention to physical description and material actualities. In taking the landscape on its own terms and remaining open to place and experience, ‘Postscript’ does not grant the narrator any defining comprehension over it. Instead, it encourages the reader to open their heart to the marvellous and feel “the insignificance of the self when placed against the natural landscape” (Moore 71-72). Ecologically, the poem ‘Postscript’ associates itself with the West of Ireland, a source of inspiration for many of Heaney’s contemporary poets who, as Tom Herron notes, contribute to its mythic status by depopulating their poems in favour of “pristine ecological experiences” (81). In that sense, Heaney’s “driving” poems too contribute to the literary mythologizing of the West of Ireland by eliding the narrator along with any other potential human characters. Yet, unlike his contemporaries,49 Heaney attempts to portray the place on its own terms, by keeping the physical description of the landscape at the heart of his poem. Heaney’s main emphasis on elemental physical attributes of the West − i.e. Burren rock, lake and sea − “achieves both a unique presence for itself as poem, and for the place itself, beyond, or in spite of, all the ‘scripts’ associated with the West of Ireland” (Moore 75-76). The features of any landscape play significant roles in the development of its culture and society. g y p ( ) 50 Recent environmentalist concerns and the question of their relevance in the Irish literary enterprise have led to the publication of various notable books, including Christine Cusick’s Out of the Earth (2010), Robert Brazeau and Derek Gladwin’s Eco-Joyce (2014) and Gladwin’s Contentious Terrains (2016). 49 See, for instance, Longley’s collection An Exploded View (1972). e, for instance, Longley s collection An Exploded View (1972). cent environmentalist concerns and the question of their relevance in the Irish literary enterprise have led q y p the publication of various notable books, including Christine Cusick’s Out of the Earth (2010), Rober and Derek Gladwin’s Eco-Joyce (2014) and Gladwin’s Contentious Terrains (2016). g y p ( ) 50 Recent environmentalist concerns and the question of their relevance in the Irish literary enterprise have led to the publication of various notable books including Christine Cusick’s Out of the Earth (2010) Robert Brazeau 52 Heaney’s bog poems include the following: ‘Digging’ (DN); ‘Bogland’ (DD); ‘The Tollund Man’, ‘Bog Oak’ and ‘Nerthus’ (WO); ‘Come to the Bower’, ‘Belderg’, ‘Bog Queen’, ‘The Grauballe Man’, ‘Punishment’, ‘Kinship’, ‘Strange Fruit’ and ‘Act of Union’ (N). The bog poems based on Tollund man’s body are: ‘The Tollund man’ (WO); ‘Tollund’ (SL) and ‘The Tollund Man in Springtime’ (DC). ( ); ( ) p g ( ) 53 The book was translated into English in 1969, at the height of the Troubles. According to Heaney, Glob argues that the bog bodies are related to a number of “ritual sacrifices to the Mother Goddess […], who needed new bridegrooms each winter to bed with her in her sacred place, in the bog, to ensure the renewal and fertility of the territory in the spring”. Heaney quickly associated these sacrificial rites with “the tradition of Irish political martyrdom for that cause whose icon is Kathleen Ni Houlihan” and realised how they constituted “an archetypal pattern” (PO 57). erard Boate’s Ireland's Natural History (1652) for more information of how colonisers compared the of Irish people to bogs. 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 With the rise of ecocriticism, the question of how geography and ecology of the Irish landscape influence the imagination of its writers has turned into a sustained critical attention.50 In the remainder of this section, I turn to the representation of the major elements 74 of Irish landscape, bogs, water, stones, birds. Donna Potts, for example, focuses on the cultural, social, historical and ecological implications of bog and Burren in Irish poetry. Regarded as waste land by the British, both the Burren and bog have come to epitomise Irish character, resisting both colonisation and settlement and thus, any alteration to their landscape.51 Irish people themselves came to associate bogs with supernatural stories. In pre- colonial Ireland, bogs were repositories for handcrafts and human bodies since the Iron Age, or even the Stone Age. During the industrial period, peat bogs became a source of economic value. Today bogs have emerged as the focus of conservation efforts in Ireland, celebrated by writers and artists for their cultural and aesthetic values (Potts 69-72). Potts’ draws attention to Heaney’s bog poems as illustrative examples.52 Heaney’s Mossbawn house was surrounded by bogs (see my Introduction). He encountered Danish bog bodies and the Jutland landscapes in the work of the Danish archaeologist Peter Glob, The Bog People (1965), when he visited the Danish museum in 1973.53 Potts explains Heaney’s personification of the bogland as feminised earth, in ‘The Tollund Man’, recalls the ancient earth’s role as a “deity”, “goddess” and “holy ground” into which victims were ritually sacrificed and the juxtaposition of human bodies and grain is the reminder of “the ultimate inseparability of human and non-human nature” (76-77). The presence of the bog as one of the most familiar and characteristic features of Irish landscape has attracted the attention of many critics. For Jahan Ramazani, the structure of peat bogs is indicative of multiple temporalities. He argues that peat bogs are particularly 75 significant to the cultural memory of Ireland for being a reminder of the hidden layers of history: “the imaginative topography of Heaney’s poetry is an intercultural space, a layered geography” (346). Karin Sanders discusses bogs’ capacity to preserve human bodies and artefacts. In Heaney’s work the silent peat bogs serve as useful loci for preserving history and revisiting memories of the human past with such startling details that only the Tollund man’s body would preserve under anaerobic conditions. f ( ) p ( ) 55 The exilic quality of Heaney’s bog poetry is connected to the continuing legacy of British colonialism in Northern Ireland. David Gleeson suggests it was Britain that “dragged [Ireland], willing or not, into the nascent English/British Atlantic” in the early modern period and that we can read “Ireland, in some ways, [as] the prototype for the whole transatlantic English colonial enterprise of the 1600s” (2). 54 See Cities of the Dead (1996) for Roach’s concept of a “circum-Atlantic” (4). ties of the Dead (1996) for Roach’s concept of a “circum-Atlantic” (4). f ( ) p quality of Heaney’s bog poetry is connected to the 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 Sanders maintains bog bodies “negotiate the liminality that comes with having to travel between their material reality as archaeological artefacts and the temporality that comes with their humanness” (9). Gladwin’s 2014 study spans a broad period in Irish writing, examining the work of Marina Carr, Frank O’Connor, Bram Stoker and Seamus Heaney, among others, to probe the connections between Ireland’s peat bogs and the postcolonial and Gothic literary traditions. Adopting Joseph Roach’s notion of the circum-Atlantic, Alison Garden traces the association between Heaney’s waterlogged poems and the familiar themes of exile and diaspora in his poetry.54 For Ireland, the Atlantic is marked by its association with decades of mostly traumatic emigration as a result of the Famine years (1845-1852) and the sectarian violence and unrest as a continuing legacy of colonisation until recent years. The archipelagic nature of Ireland and the transatlantic associations of its history have encouraged Heaney – as well as the American poet Natasha Trethewey and the Caribbean poet Kwame Senu Neville Dawes – to fashion watery interstices as a site for evoking the shared memory of the “the chorus of water-lost” voices (Garden 91).55 Heaney’s boglands present an uncanny type of landscape that brings together land and water as the locus of an intercultural heritage Ireland shares with nations across the ocean. The fluidity of bogs, particularly in Door into the Dark, Wintering Out and North, not only embodies the unsteadiness of Ireland, in terms of its 76 political, cultural and social condition but it also refers to the sheer expanse of the lives lost throughout Irish history. Bogs are the sites where an “unfenced Ireland seeps into the Atlantic” (Garden 95). Garden particularly invites us to listen to what the Atlantic waves echoed in Heaney’s bogland and hear the long history of outmigration across the Atlantic (95). Nicholas Allen too is interested in the representation of water and liquidity in Heaney’s poetry. From the dripping water in the backyard of his Mossbawn house to the ink at the tip of his pen in Death of Naturalist, from the river Moyola flowing in Wintering Out to the feel of a greasy eel in Human Chain, according to Allen, Heaney’s poetic and intellectual landscape is rich with symbolic representations of water in all its manifestations (173-174). 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 Water – in the form of drips, juices, ink, riverbanks, boggy stretches, lakes and oceans – can be seen as a key medium for constructing cultural and historical associations (173). He views this liquidity as a mode of transition and therefore maintains that the profusion of the many varieties of water in Heaney’s work is a metaphor for changing states and flexibility required to confront the sharp realities in Northern Ireland (174). In addition to land and water, Heaney scholars have been drawn to the presence and meaning of trees, birds and other animals in Heaney’s poetry. In one of his early essays, Heaney recalls crouching in one of his secret nests, “the fork of a beech tree”, and comments “once you squeezed in through it, you were at the heart of a different life” (PO 18). Heaney also refers to the “mysteries of the grove” and “the powers of the Celtic otherworld” to describe the symbolism of the wood in early Irish nature poetry (PO 186). From Sweeney’s praise of the trees, to Frank O’Connor and James Patrick Carney and to Heaney himself, Celtic imagination has been “beautifully entangled with the vegetation and the weathers and animals of the countryside […] attesting to the god in the tree as a source of poetic 77 inspiration” (PO 188).56 This engagement with nature, and the wood in particular, has been examined by a score of literary critics and environmental activists. Terry Gifford includes Heaney in Green Voices (1995) and Sidney Burris meticulously examines Heaney’s handling of pastoral and anti-pastoral traditions in The Poetry of Resistance (1990). Meg Tyler too pays specific attention to Heaney’s re-writings of pastoral texts (4-28). In the following paragraphs, I discuss such elements of the Irish environment as the wood and native animals. Colleen McKenna explores the literary, mythical and cultural connotations of trees in Heaney’s poetry and translations. For him, ‘Sweeney Astray’ is a rich source of the liminal image of the tree: set in the Irish forest beyond the walls of the city, with named and venerated trees, the wood serves as a “passing phase” for Mad Sweeney to transform and achieve redemption (42-43). y There are some poems which are clearly autobiographical. […] Now, the Sweeney figure spends a lot of his time roosting in trees – so in a sense, that gives permission for this voice to speak and it remembers a moment which I think all children have. […] And in my case, I remember this particular tree I used to climb; […] Sweeney in the tree, and the child Heaney in the tree are merged together there and I use them to re-collect and to re-member [emphasis in the original]. (Heaney qted in McKenna) p p ( ) 57 Heaney’s remark on this famously known personal allegory further confirms the Heaney-Sweeney conflation. In his 1984 interview with Paul Vaughan, he explains that the identity of Sweeney blends with “the child Heaney in the tree”: 56 In his essay ‘The Placeless heaven’, he talks about how a chestnut tree in Mossbawn – planted by his aunt at the time of his birth – once cut down, became a symbol of “being rooted in the home ground” and how he envisioned the place where it had been as “a kind of luminous emptiness” (181-182). envisioned the place where it had been as a kind of luminous emptiness (181 182). 57 Heaney’s remark on this famously known personal allegory further confirms the Heaney-Sweeney conflation. In his 1984 interview with Paul Vaughan, he explains that the identity of Sweeney blends with “the child Heaney in the tree”: There are some poems which are clearly autobiographical. […] Now, the Sweeney figure spends a lot of his time roosting in trees – so in a sense, that gives permission for this voice to speak and it remembers a moment which I think all children have. […] And in my case, I remember this particular 56 In his essay ‘The Placeless heaven’, he talks about how a chestnut tree in Mossbawn – planted by his aunt at the time of his birth – once cut down, became a symbol of “being rooted in the home ground” and how he envisioned the place where it had been as “a kind of luminous emptiness” (181-182). 57 Heaney’s remark on this famously known personal allegory further confirms the Heaney-Sweeney conflation. In his 1984 interview with Paul Vaughan, he explains that the identity of Sweeney blends with “the child Heaney in the tree”: There are some poems which are clearly autobiographical. […] Now, the Sweeney figure spends a lot of his time roosting in trees – so in a sense, that gives permission for this voice to speak and it remembers a moment which I think all children have. […] And in my case, I remember this particular tree I used to climb; […] Sweeney in the tree, and the child Heaney in the tree are merged together there and I use them to re-collect and to re-member [emphasis in the original]. (Heaney qted in McKenna) 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 The voice of Heaney-Sweeney continues to be heard in a number of poems in Wintering Out, North and especially Station Island, where Heaney identifies himself with Sweeney as the “small dreamself in the branches” (McKenna 43).57 In ‘Oracle’, ‘Exposure’ and ‘Glanmore Sonnets V’, for instance, the tree is seen as a sanctuary for the poet, while ‘In the Beech’ probes the contradictory nature of the tree: “it is solid and soft, strange and comfortable, timeless and time-specific, personal and universal” (44-45). Heaney, as it were, “unwrites” the symbol of the tree in his poetry and translation. This conceptualisation of the tree as a liminal space is conceptualisation of the tree as a liminal space is indicative of an epistemological shift and poetic crossing, and the event itself can be read as an analogue for an artistic passing through, a paradigm for the general indicative of an epistemological shift and poetic crossing, and the event itself can be read as an analogue for an artistic passing through, a paradigm for the general 78 lightening of Heaney’s verse, which has been increasingly engaged with issues of space and the ephemeral. (McKenna 55-56) Krishnendu Bera adds the ecological concerns of the last few decades to the large body of Heaney scholarship to examine Heaney’s “eco-consciousness” – i.e. the way he establishes his individual and communal identity in relation to the natural environment − arguing that, in his poems and translations, Heaney subverts the stereotypical binaries of human/non-human encouraged by anthropomorphism throughout the centuries of human civilisation (1-3).58 Bera suggests that Heaney’s eco-consciousness is particularly evident in his presentation of the peat bogs as the archetypal memory of the nation’s past and as a symbol of Irish identity. In his bog poems, he not only identifies his communal identity with the land but also indicates that violence has always been an integral part of human culture. The sensibility that “the whole world is a vast eco-system and we are merely part of it like any other animate or inanimate element” is also evident in Station Island, where the central voice belongs to Sweeney, the king transformed into a bird, which serves as the poet’s own self (Bera 1-3). Just like bogs, the trees are not merely an ornamental part of a pastoral landscape or a material for construction in Heaney’s poetry. Irish vegetation forms an integral part of his personal and collective identity. 58 Bera bases his ecocritical perspectives on Glotfelty’s 1996 definition of ecocriticism as “the study of the relationship between literature and the physical environment” in order to look at the psychological significance of environment in identity formation (XIX). p p y of environment in identity formation (XIX). 2.2 The Republic of Conscience: Landscapes, Places and Spaces41 In ‘Oracle’ and ‘Sweeney Praises the Trees’, the poet’s treatment of the trees contributes to their status as “mythic” and “spiritual” (Bera 1-3). In ‘Land’ the greenery serves as a symbol of the spirit of the country and memory. In ‘Sibyl’, the partitioned Ireland is described in the image of a “helmeted and bleeding tree” which requires “green and open buds” to recover (FW 5). Heaney’s poetry is reflective of the “ecosophical outlook” he has developed through a deep identification of his identity with his natural environment, one that is not delimited by the personal ego and instead urges him to experience his self as “a genuine part of all life” (Bera 174). 58 Bera bases his ecocritical perspectives on Glotfelty’s 1996 definition of ecocriticism as “the study of the 79 Indigenous Irish animals too carry a symbolic weight in Irish literature, culture and politics.59 Potts’ sharp ecocritical approach traces Heaney’s concerns for the Irish ecology at the heart of his bog poems, as in ‘Bogland’ where Heaney seeks his Irish identity by corkscrewing downwards into the repository of Irish history only to find the skeleton of the long-extinct Irish Elk. “The extinction of animals”, she argues, “also provides a means for talking about the consequences of colonisation” (144-145). In ‘Midnight’, similarly, Heaney compares the disappearance of the Irish language to the extinction of the wolf in Ireland after the wars of the seventeenth century and the colonial settlement, but also laments the extinction of the Irish wolf as the consequence of human greed and deforestation. In No. III from his sequence ‘Glanmore Sonnets’, Heaney writes of the cry of the cuckoo and corncrake, which, in addition to its pastoral implications, reveals the poet’s concern for the extinction of avian life due to the mechanisation of agriculture (Potts 165-167). y 60 The title of this section is inspired by Heaney’s ‘Song’ (FW 53). 59 See Kathryn Kirkpatrick and Borbála Faragó’s Animals in Irish Literature and Culture (2015), for a symbolic study of animals in Irish literature. See Kathryn Kirkpatrick and Borb study of animals in Irish literature. y p q g g g 62 For instance, Michael Molino views Heaney’s language as “an innovative way” of addressing centuries-old antagonism in Ireland (181). Crotty identifies a trajectory from a first-person, private point of view to more public verses and varying viewpoints in Heaney’s early poetry and interprets this transition as reflective of the poet’s response to his literary exemplars − i.e. Kinsella, Murphy and Montague (41). Rankin Russell comments on how Heaney’s play with various etymologies contributes to the reconciliation of his divided landscape: “Heaney clearly draws upon his Scots, Irish, and English linguistic heritage in his poetic vocabulary, implicitly suggesting that so should the inhabitants of Northern Ireland in order to form an imagined community” (Poetry and Peace 203). Tyler discusses Heaney’s relationship to the genres of elegy and pastoral elegy at the level of language and examines Heaney’s word choice as influenced by a knowledge of etymology and prosody: “the diction he uses, ranging from Ulster idioms to Latinate phrasing, illustrates his intentional reach towards languages and cultures other than his own” (4-28). Wheatley’s 2016 study examines the dialogue between English and Irish languages in Heaney’s work as markers of cultural identity and the complexity of his relationship to British colonialism. 61 For instance, David Lloyd examines Heaney’s language use in relation to identity. His work has also been studied from a cognitive stylistic perspective by many critics. Elena Semino uses Schema Theory to examine Heaney’s linguistic description in ‘The Pillowed Head’. Maria Teresa Calderón Quindós applies Blending Theory to explain the integration of the various pieces of the poem ‘Oracle’ in relation to the aesthetic principle of unity-in-variety. Nigel McLoughlin applies Text World Theory to study interactions of the reader with the different worlds Heaney has constructed in his poem ‘Squarings: Lightenings VIII’. 2.3 The Music of What Happens: Music, Sound, Voice and Silence 60 In this section, I shall review and critique a number of scholarly works where the concepts of language and voice, sound and music, as well as silence in the poetry of Heaney are examined. My review begins by acknowledging a score of critical works that have analysed Heaney’s poetic language from various critical lenses – e.g. cognitive theories, New Criticism, as well as etymological and structural analysis, prosody and phonetics. I direct the discussion towards studies that examine the concepts of music, sound and silence in his work. This literature review particularly indicates the original contribution of this thesis to what is becoming the field of Heaney Studies. It shows that despite the scholarly attention Heaney’s auditory imagination has received to date, there is none that offers a systematic approach for both identifying and interpreting auditory references in his work; nor does any of the previous 80 studies look specifically and thoroughly at all the various aspects of the soundscape in his writing. Heaney’s language has been studied from a variety of theoretical perspectives.61 Since the publication of Wintering Out, with emphatic reflections on place-names – and their etymology, pronunciation and societal relevance – a considerable body of literature has been produced with a focus on Heaney’s use of language for its syntactic structure and etymological depth as symptoms of human history, culture, memory and identity, and as the poet’s attempt to unify his culturally diverse roots.62 Since Heaney’s acknowledgement of his debt to Eliot’s concept of “auditory imagination”, many critical debates have centred on the sonic and musical aspects in Heaney’s language.63 In his All the Realms of Whisper (1988), Andrews highlights the cognitive import of auditory perceptions and responses in Heaney’s poetry (41). He also shows a linguistic bent in his analysis of Heaney’s poetry focusing mostly on the composition of his vowels and consonants. In Seamus Heaney (1989), Thomas C. Foster too presents an enlightening study of the prosodic features of Heaney’s poetry 63 Eliot defines “auditory imagination” as: the feeling for syllable and rhythm, penetrating far below the conscious levels of thought and feeling, invigorating every word; sinking to the most primitive and forgotten, returning to the origin and bringing something back, seeking the beginning and the end. 63 Eliot defines “auditory imagination” as: the feeling for syllable and rhythm, penetrating far below the conscious levels of thought and feeling, invigorating every word; sinking to the most primitive and forgotten, returning to the origin and bringing something back, seeking the beginning and the end. It works through meanings, certainly, or not without meanings in the ordinary sense, and fuses the old and obliterated and the trite, the current, and the new and surprising, the most ancient and the most civilised mentality. (118-119) g , p y , y y It is one thing to find lexical meanings for the words and to have some feel for how the metre might go, but it is quite another thing to find the tuning fork that will give you the note and pitch for the overall music of the work. Without some melody sensed or promised, it is simply impossible for a poet to establish the translator’s right of way into and through a text. (Beowulf XXVI) 64 Referring to ‘music’, in the context of poetry translation, Heaney says: ferring to ‘music’, in the context of poetry translation, Heaney says: 2.3 The Music of What Happens: Music, Sound, Voice and Silence 60 It works through meanings, certainly, or not without meanings in the ordinary sense, and fuses the old and obliterated and the trite, the current, and the new and surprising, the most ancient and the most civilised mentality. (118-119) 81 focusing mostly on its syllables and metre. Michael Parker’s 1993 book, instead, presents an analysis of phonemes and sounds in Heaney’s poetry. Other scholars, like Alan Robinson and Hart, have emphasised the cultural and political relevance of the phonetic features of ‘Anahorish’ and ‘Broagh’, a significant aspect of vocal sounds in Heaney’s poetry that I will elaborate in the analytical chapters of this thesis (see 3.1 and 3.4). In The Poetry of Seamus Heaney (1998), Corcoran admires Heaney’s onomatopoeia, alliteration and synaesthesia, which allow him to convey the “observed and recollected facts of his early rural experience [...] in a language of great sensuous richness and directness” (i). In Seamus Heaney and the Language of Poetry (1994), O’Donoghue highlights the centrality of the sound of the words in Heaney’s place-name poems and applies the traditional linguistic categories of phonology, word class and vocabulary to present a formal and comprehensive analysis of both Heaney’s poetry and his commentaries. For David Perkins, Heaney is a born poet, a person of truly poetic gifts and that impression is made readily apparent when listening to the music of his language. He illustrates Heaney’s masterful use of “poetic sound” as a way of achieving “multeity in unity”, explaining that Heaney’s language weaves together a variety of sounds that − as a pattern or not − reinforces the semantic meaning of words and profoundly affects the reader, reminding one of what, in Biographia Literaria (1817), Coleridge described as “the sense of musical delight” (qtd in Perkins 63). In 2017, Falconer studies Heaney’s Aeneid Book VI highlighting that Heaney matches the music of the original work − i.e. rhythm, metre, lineation, the voice, tone, register, diction, pacing and pitch − enough to be true to Virgil but not so antique as to stand out of tune with contemporary taste (430-439).64 Other contributions worth noting are by Anthony Cuda, who focuses on the sonic features of 82 Heaney’s language and its relationship to his sense of “historical and racial continuity” (171), and Jeffrey Bilbro, who highlights the “linguistic delights” or “healing delights” that come from the poet’s “delicate feel for the aural qualities of language” (322). On January 5th 1976, near the village of Kingsmill in Northern Ireland, a bus carried sixteen textile workers five Catholics and eleven Protestants – home from work in Glenanne. Four of the Catholics got off the bus on the way, while one continued on the road to Bessbrook. Somewhere on the road, the bus was stopped by what they first assumed to be a British Army or RUC checkpoint. However, the gunmen identified and released the only Catholic man, but shot the remaining eleven men, for the gunmen were members of the Provisional IRA (O’Brien 13-14). ( ) 66 Heaney took the line from a translation by James Stephens of an Irish folktale: “‘The music of what happens’, said great Fionn, ‘that is the greatest music in the world’”: www.youtube.com/watch?v=e87nqv9PiLA (22:30- 22:50). He re-uses this phrase in ‘Song’: “that moment when the bird sings very close / To the music of what happens” (FW 53). 2.3 The Music of What Happens: Music, Sound, Voice and Silence 60 For Bilbro, what is pivotal is “the way in which Heaney embeds the soothing effect of the sounds into his poem instead of attempting to didactically explicate their function” (322). The association between music and poetry has often been overshadowed by the nineteenth and twentieth-century insistence on the aesthetics of poetic language. However, not all Heaney scholars restrict commentary to the musicality and sonic properties of his language. More recent studies have examined the concepts of sound and music from ecological and eco-musicological approaches. In her 2014 study, Leighton explores the concept of music in Heaney’s “the music of what happens” (FW 56). Her argument revolves around the arbitrariness of this familiar concept, its particular relation to the definitive article that precedes it − i.e. ‘the’ − and the preposition that follows it − i.e. ‘of’ − in Heaney’s quotation. Leighton argues that Heaney’s music is not exactly music in the traditional Romantic-aesthetic sense and the art of sounds indifferent to the cruelties of history, nor is it merely a masterful composition of sounds intended to lift the semantic meaning of the words, it is rather “a poetic song, which comes close to the reality of the world” (20). Heaney’s music records the actual presence of things that happen: someone digging with a spade, a singing bird, a leaking pump, or a gunshot. It is not composed, it happens. She then quotes a well-known poem from The Spirit Level, where Heaney encourages his readers to listen ‘for’ − and not ‘to’ − this music: “what happens next / Is a music that you never would have known / To listen for” (SL 1). According to Leighton, the music Heaney refers to in these lines happens where there is a chance to go beyond the apparent meaning of what is happening, offering a better chance to capture the message beyond (20). As Heaney says: “Who cares if all the music that transpires // Is the fall of grit or dry seeds through a cactus?” 83 (SL 1). Indeed, in Crediting Poetry, referring to the poignant story of the Catholic who survived what first appeared to be a Protestant paramilitary plot, but turned out to be an IRA’s ambush,65 Heaney says: “the gunfire that followed, so absolute and so desolate, if also so much part of the music of what happens” (OG 457). 65 On January 5th 1976, near the village of Kingsmill in Northern Ireland, a bus carried sixteen textile workers – 2.3 The Music of What Happens: Music, Sound, Voice and Silence 60 The music in Heaney’s poetry is “beyond what is told, beyond the register of accessible plain-speaking and factual recall” (Leighton 30). This music can be the sound of whatever happens as long as we train our ears not to ignore it. As the Irish mythology sums it, “the music of what happens […] that is the greatest music in the world”.66 The relationship between poetry and its surrounding natural soundscape has been the focus of the poet and his critics. Heaney celebrates the natural soundscape for nourishing poetry as much as he celebrates poetry in the acoustic ways of reflecting it. In his review of The Penguin Book of Pastoral Verse (1974), he criticises the short-sightedness of Marxist interpretation of pastoral poetry as a social criticism, which “sweeps the poetic enterprise clean of those somewhat hedonistic impulses towards the satisfactions of aural and formal play out of which poems arise, whether they aspire to delineate or to obfuscate ‘things as they are’” (PO 174). Instead, he endeavours to capture the sounds of the natural environment as much for educating the ear as for delighting it. Fanny Quément’s 2016 study uses the ecological lens to investigate the auditory space of Heaney’s pastoral poems for the various ways in which the poet expresses his mission to enlighten his readers about the endangered sounds. From the “strong gauze of sound” around the “flax-dam” in ‘Death of a Naturalist’ to the song of the curlew in ‘From the Republic of Conscience’, Heaney attempts to evoke and 84 archive the slightest details in the disappearing soundscape of Ireland (38). Yet in “The Tollund Man in the Springtime’, he stresses the more global dimension of sound pollution in a space where he can “smell the air, exhaust fumes, silage reek” (DC 56). In the meantime, however, he hints at the limitations of language for translating and preserving the soundscape encouraging his readers to go back to and appreciate the “extratextual landscape” (Buell 33). The examination of the concept of soundscape and music in relation to the ecological environment has also been the critical concern of Potts, who examines how the Irish environmental movement − which began gaining momentum in the 1970s − has influenced and been addressed by contemporary Irish writers and artists (Potts XIX). 2.3 The Music of What Happens: Music, Sound, Voice and Silence 60 As a subdiscipline of environmentalism, ecomusicology is essentially concerned with recognizing the aesthetic values of nature and resisting the relentless quests for its alteration. Within this field, the concept of biosemiotics interprets both plants and animals as communicating agents in the natural soundscape. Potts’ chapter ‘Music in Stone’, in particular, explores the Burren region of Southwestern Ireland as a source of imaginative inspiration for modern Irish writers such as Cora Harrison, Ré Ó Laighléis, Moya Cannon, Michael Longley and Seamus Heaney. The Burren − from the Irish word Boireann meaning ‘great rock’ − is a unique lunar-like landscape of bedrock resistant to cultivation and commodification (Potts 39-41). Potts explains how the Burren has been affected by British colonials, nationalists or tourists throughout time focusing on ‘An Aisling in the Burren’ as the example in which the poet, as an environmental activist, seeks to convey the intrinsic values of this so-called barren, wild and untamed region of Ireland. In the poem, Heaney’s Aisling − meaning dream or vision in Irish − is embodied in the figure of a woman who arrives “licked with the wet cold fires of St Elmo, / angel of the last chance” (SI 47). The Burren is thus a symbol of hope amidst the Troubles. 85 In addition to sounds, in both his poetry and prose, Heaney voices his awareness of silence in all its forms, meanings and implications. However, most early studies as well as current work focus on silence in the context of Irish studies in general and more specifically in relation to the Troubles. In “Speaking of Silence” (2012), Maria Beville and Sara Dybris McQuaid discuss silence as “a forbearing presence in literary, historical, cultural and political discourse in Ireland”, drawing on Heaney’s ‘Whatever You Say, Say Nothing’ as an example in which silence defines the cultural encounter between the communities (7). In “The Stones of Silence” (2016), Petar Penda focuses on Heaney’s political views regarding the Troubles in his poetry. 67 For instance, Lloyd accuses Heaney for being subservient to nationalist politics; Robert McLiam Wilson and Maurice Harmon accuse him of not confronting the violence directly and not having a firm standpoint about British discrimination against the Irish; while Ciaran Carson accuses him of aestheticizing and mythicizing violence, Eugene O’Brien and Longley disapprove of him for politicizing poetry. Non-scholarly works dealing with the concept of silence in Heaney’s poetry include the 2001 article in The Irish Times, which indicates Heaney’s preoccupation with silenced and forgotten voices of the survivors of the Holocaust: www.irishtimes.com/culture/remembering-the-unspeakable-1.273250 and the 2013 article in The New York Times by Francis X. Clines who argues that for Heaney, poetry was an escape from the fear of being silent: www.nytimes.com/2013/08/31/opinion/seamus-heaney-poet-of-the-silent-things.html). See Heaney’s own views on the topic in his 1996 interview with Charlie Rose: charlierose.com/videos/12382. y p y p g ) on the topic in his 1996 interview with Charlie Rose: charlierose.com/videos/12382. g p Times by Francis X. Clines who argues that for Heaney, poetry was an escape from the fear of being silent: www.nytimes.com/2013/08/31/opinion/seamus-heaney-poet-of-the-silent-things.html). See Heaney’s own views on the topic in his 1996 interview with Charlie Rose: charlierose.com/videos/12382. 67 For instance, Lloyd accuses Heaney for being subservient to nationalist politics; Robert McLiam Wilson and Maurice Harmon accuse him of not confronting the violence directly and not having a firm standpoint about British discrimination against the Irish; while Ciaran Carson accuses him of aestheticizing and mythicizing violence, Eugene O’Brien and Longley disapprove of him for politicizing poetry. Non-scholarly works dealing with the concept of silence in Heaney’s poetry include the 2001 article in The Irish Times which indicates 67 For instance, Lloyd accuses Heaney for being subservient to nationalist politics; Robert McLiam Wilson and Maurice Harmon accuse him of not confronting the violence directly and not having a firm standpoint about British discrimination against the Irish; while Ciaran Carson accuses him of aestheticizing and mythicizing violence, Eugene O’Brien and Longley disapprove of him for politicizing poetry. Non-scholarly works dealing with the concept of silence in Heaney’s poetry include the 2001 article in The Irish Times, which indicates Heaney’s preoccupation with silenced and forgotten voices of the survivors of the Holocaust: www irishtimes com/culture/remembering the unspeakable 1 273250 and the 2013 article in The New York www.irishtimes.com/culture/remembering the unspeakable 1.273250 and the 2013 article in The N Times by Francis X. Clines who argues that for Heaney, poetry was an escape from the fear of bei www.nytimes.com/2013/08/31/opinion/seamus-heaney-poet-of-the-silent-things.html). See Heaney’s o on the topic in his 1996 interview with Charlie Rose: charlierose com/videos/12382 www.nytimes.com/2013/08/31/opinion/seamus-heaney-poet-of-the-silent-things.html). See on the topic in his 1996 interview with Charlie Rose: charlierose.com/videos/12382. 68 The poem refers to the 1951 discovery of the body of a girl of about fourteen years old from the Iron Age weighed down in a bog by trees and a stone. Heaney finds the parallel in contemporary Ireland, when women who entertained intimate relationship with British soldiers were shaved, stripped of their clothes and handcuffed to rails by IRA members as punishment for adultery (Penda 270). 2.3 The Music of What Happens: Music, Sound, Voice and Silence 60 To begin with, he identifies two major types of critical response to Heaney’s work: on the one hand, critics who either accuse Heaney of taking sides with the nationalist ideology or those who criticise him for silence and appropriation towards the British and, on the other, the critics who denounce politics as the subject matter of poetry.67 While he condemns the first category of Heaney critics for imposing their own personal political inclination and neglecting “what literary criticism is about”, Penda argues that separating poetry from politics “would denote separation of poetry from life itself” (269). “Belonging to a society”, he argues, “the poet is a political being − homo politicus. Hence, a work of art is, to a certain extent, a product of a politics or its ideology; it produces an ideology as well” (269). As already noted, Heaney himself maintains that “poetry’s existence as a form of art relates to our existence as citizens of society” (RP 1). Heaney condemns what Vendler calls “a generalised cultural approval of violence” (51). According to Penda, his disapproval of silence, in this context, is illustrated in the poem ‘Punishment’ from North, where he 67 For instance Lloyd accuses Heaney for being subservient to nationalist politics; Robert McLiam Wilson and 86 juxtaposes similar but distant events in the history of human violence.68 The poem stretches between the stoning of a fourteen-year-old girl by Germanic tribes and the punishment of some Irish women in contemporary Belfast by members of the IRA, condemning the poet- narrator for remaining uncertain, passive and “dumb” casting again “the stones of silence” (N p , pp to rails by IRA members as punishment for adultery (Penda 270). 31). He ends the poem saying: Who would connive in civilized outrage yet understand the exact and tribal, intimate revenge. (N 31) In 1990, Hart wrote about the sense of “silence”, “absence” and “emptiness” that disappointed several Heaney critics after reading his then newly published collection, The Haw Lantern, arguing that Heaney’s illustration of absence is in fact a sign to be identified and interpreted. The emptiness that haunts his volume unfolds the maturing poet’s mistrust of speech and therefore hints at the underlying political impulses behind speech and writing (Hart 461-492). This perspective is also echoed in Wheatley’s 2001 study of the various ways in which, in a violently divided society, Heaney and other Northern Irish poets “deploy strategies of silence, secrecy, private reference, and tribal shibboleth rather than “blabb[ing] out’” (HL 19). He maintains that these strategies may superficially appear to work against self-expression, but in reality they unfold layers of meaning in the most unusual ways (Wheatley 1-2). Similarly, Eileen Cahill writes about Heaney’s poetry as the space wherein the poet fuses speech and silence in “introspective rather than expressive” manner (29). Heaney, like his father, was a naturally quiet person but his poetry discloses his feelings about the political situation in Northern Ireland. Heaney’s poetry is, therefore, “tactically political” without being explicitly and explosively so (Cahill 29). Elizabeth Lunday argues that Heaney is not a 87 silent and indifferent bystander of the Troubles that we could accuse of aestheticizing the violence (111). She gives the example of ‘Mycenae Lookout’, in which the poet declares that there is “No such thing / as innocent / bystanding” (SL 30). Heaney notes that he was in search of “an adequate response to conditions in the world at a moment when the world was in crisis” (RP 191). In ‘Whatever You Say Say Nothing’, he shows his readers how violence in a society can affect each individual living in it and result in “a conspiracy of silence” (Lunday 111). But Heaney’s silence is, in fact, the silence of contemplation. It is a type of silence that often breaks into a distinctive voice. Heaney read Aeschylus’s The Oresteia after the 1997 ceasefire, finding a parallel between the end of the Trojan War and the end of The Troubles in Ireland. 31). He ends the poem saying: He identified with the silence of the character of the Watchman in the first of the three plays in the Oresteia trilogy: The Watchman […] began to keep coming back to me with his in-between situation and responsibilities and inner conflicts, his silence and his knowledge, and all this kept building until I very deliberately began a monologue for him using a rhythmed couplet like a pneumatic drill, just trying to bite and shudder inward whatever was there. (“Art of Poetry” 137) The Watchman […] began to keep coming back to me with his in-between situation and responsibilities and inner conflicts, his silence and his knowledge, and all this kept building until I very deliberately began a monologue for him using a rhythmed couplet like a pneumatic drill, just trying to bite and shudder inward whatever was there. (“Art of Poetry” 137) The Watchman’s character and his silence have been interpreted in various ways by various translators, primarily as a way to set the scene and start the narration of the plot. For Heaney, however, the Watchman functions as more than a literary device. He has a story to tell. Heaney offers the Watchman “a voice that breaks the silence” (Lunday 113-114). Likewise, in his poetry, Heaney grants himself the right to break free from “the pincer jaws of an emotional and artistic dilemma” (SS 392). He does not speak for either side, but opens in- between a free space to voice his thoughts, and in doing so he expresses the responsibility to make himself heard. His poetry, therefore, comes, as the poet himself quotes from Yeats, like “[s]peech after long silence. It is right” (SS 203). Silence in these studies has been examined more generally and as a response to the contemporary socio-political situation. To my knowledge, very little scholarly work has been 88 devoted to the other aspects of this significant and empowering concept in Heaney’s work. More recently, Jae Joon Kim has focused on silence as an aesthetic device in Heaney’s poetry. Kim traces two types of silence in Heaney’s ‘Station Island’ sequence. His paper begins with a brief reference to a type of silence he compares to the intervals of rest in or the end of a musical piece. 31). He ends the poem saying: This type of silence, he comments, is prevalent throughout the poem: The opening poem in the sequence opens with: “[…] an escaped ringing / that stopped as quickly // as it started” and towards the end of the first part he writes: “[t]he quick bell rang again” (SI 61-63). The second type he identifies is the silence that prompts a discourse (Kim 5). In psychoanalysis, the psychiatrist’s absolute silence is known as the Freudian methodological principle encouraging an uninterrupted monologue by the patient (Reik 122). According to Kim, this type of silence is often performed by one of the characters in Heaney’s poem as in the first part of ‘Station Island’, prompting the persona to develop a poetic discourse in the form of a monologue. In this case, the persona performs one part of the monologue while the other side is performed by the silent auditor. In these instances, Heaney attempts to “bring out the oppressed, silenced voices from his consciousness” (5). Drawing on Bakhtin, Kim claims that, during the monologue, silent readers “should try to participate actively in developing the monologue into a dialogue” (5), but a further question that remains unanswered is who are the oppressed voices in the poet’s mind and memory, which I address in the final section of the thesis. According to Kim, this type of silence is often performed by one of the characters in Heaney’s poem as in the first part of ‘Station Island’, prompting the persona to develop a poetic discourse in the form of a monologue. In this case, the persona performs one part of the monologue while the other side is performed by the silent auditor. In these instances, Heaney attempts to “bring out the oppressed, silenced voices from his consciousness” (5). Drawing on Bakhtin, Kim claims that, during the monologue, silent readers “should try to participate actively in developing the monologue into a dialogue” (5), but a further question that remains unanswered is who are the oppressed voices in the poet’s mind and memory, which I address in the final section of the thesis. Conclusion The association between Heaney’s poetry, sound and auditory perception has often been overshadowed by the traditional literary insistence on the aesthetics of poetic language. Since the publication of his early poetry collections, several scholars have addressed the aural aspects of Heaney’s poetry in relation to Eliot’s auditory imagination, restricting their 89 commentary largely to the aural functions of his language in poetry. More recently, scholars have examined the concepts of sound, listening and music from ecological approaches, in relation to the Irish political climate. However, their exploration of Heaney’s poetry often approaches the concept of sound in a more peripheral manner rather than directly examining how it may contribute to the reading and interpretation of poetry. This thesis is interested in listening as an important medium of experiencing, understanding, communicating and remembering, and approaches the study of sound in Heaney’s poetry in a systematic and holistic way. It offers a framework that helps identify and interpret auditory references in Heaney’s writing in general, one that includes not only the aural features of his poetic language, but also his mere descriptions and commentaries. Moreover, this thesis relates to the concepts of sound and listening in Heaney’s work in a much broader context than just political to include his aesthetic, personal, social, cultural and environmental concerns as well. Heaney scholars have demonstrated how reading and translating other poets enabled him to enter and document other times and places in history, leading to a better understanding of his own time and place and to tilt his scale towards a more global reality. The concern for the idea of balance has in turn expanded his audience beyond the limits of time and place. In his Nobel Lecture, Heaney states he regards poetry as “a journey into the wideness of the world” and credits poetry “for making this space-walk possible” (OG 449). In The Redress of Poetry, Heaney notes the “tendency to place a counter-reality in the scales – a reality which may be only imagined but which nevertheless has weight because it is imagined within the gravitational pull of the actual and can therefore hold its own and balance out against the historical situation” (RP 3-4). Heaney’s work bears witness to his talent as a reader while paying homage to his private experiences. p p y ( ) 70 Heaney talks about the poem ‘Changes’ in Stepping Stones, answering a question concerning the family’s Chapter 3: Chapter 3: The title of this chapter is inspired by a line from ‘Gifts of Rain’ (WO 13). y p g pp g , g q g y move from Co. Wicklow to Dublin and the urban upbringing to which this would expose his children. (See 55) y p g pp g g q g y move from Co. Wicklow to Dublin and the urban upbringing to which this would expose his children. (See 55) Conclusion In the analytical chapters of this thesis, I explore 90 how his imaginary and real soundscapes further our understanding of not only his own time and place, but also the soundscape of the world throughout history. The fact that the surrounding landscapes − rural as well as urban − nourished Heaney’s imagination and conceptualisation of the world hardly needs saying. Heaney is known as an author who, in his poetry and prose, writes intimately of a locality. But, as Allen, Basu and other scholars have reminded us, Heaney’s ‘given landscapes’ and ‘given notes’ ought not to be misinterpreted as mere topographical descriptions. Heaney’s poetry is a record of his memory and experiences of life and a medium to probe depths of meaning associated with them. His acoustic representation of the locale is one of his major poetic tropes and a metaphor for his ideas of human relationship, sense of community, and by extension, identity. In an early translation from the Irish, Heaney reminds us that “[p]oetry of any power is always deeper than its declared meaning. The secret between the words, the binding element, is often a psychic force that is elusive, archaic and only half-apprehended by maker and audience” (PO 186). In this thesis, I show how in his re-construction of a place, Heaney brings together the various elements of the soundscape into a tapestry of personal, cultural, social and historical implications that he wishes us to feel, to ponder upon and to remember. 91 91 move from Co. Wicklow to Dublin and the urban upbringing to which this would expose his children. (See SS 255) Introduction A close reading of the lines above indicates the way Heaney introduces silence only to immediately suggest that in fact silence is filled with many sounds – sounds that growing up in a city his young companion could no longer hear: the bite of the spade as it sinks into the ground, the mixing of mortar, the footsteps of women approaching a water pump and the rattling of empty buckets. Poetically, if we read the lines aloud, we realise how Heaney recreates these sounds through his extensive alliterations, linked velar and bilabial plosives, sibilants and even onomatopoeia. The third line of the poem becomes a key indicator of Heaney’s alertness to sounds, of the importance of listening, of experiencing the world through sounds and of remembering the sounds that could no longer be heard. Heaney has a lesson to impart to everyone: “Remember this. / It will be good for you to retrace this path” (SI 37).70 The purpose of this chapter is to highlight the significance of sounds in Heaney’s poetry and to explore the relationship between human beings and their acoustic environment 92 as illustrated by the poet. The interpretation of sounds in Heaney’s poetry is integral to an understanding of the poet’s thoughts and experiences. The fact that Heaney writes intimately of places hardly needs saying now. From Mossbawn and its neighbouring regions, to the North and South of Ireland, Denmark, Greece, France, Italy, Poland and America, the concept of place has been a consistent source of inspiration for Heaney’s work. In his poetic re-construction of a sense of place, Heaney brings together the various components of the soundscape − i.e. geophony, biophony and anthrophony − to weave together the underlying social, political, cultural and personal associations. This chapter − divided into the four sections of geophonic sounds, biophonic sounds, anthrophonic sounds and human vocal sounds − aims at identifying and interpreting such references in Wintering Out.71 In the first section, I focus on Heaney’s representation of geophonic sounds − i.e. non- sentient ambient sounds of the natural phenomena (Pijanowski et al. 1214). Examples of literal and metaphorical references to geophonic sounds in Heaney s poems include: water goes over / like villains dropped screaming to justice” (DN 40); “the wind heavy / With spits of rain” (DD 19); “mizzling rain” (WO 4); “[b]ehind a windbreak wind is breaking through” (N XI); “rustling and twig-combing breeze” (FW 30); “wind blowing round the car” (SI 64); “From a hot spring, I could hear nothing / But the whole mud-slick muttering and boiling” (HL 39); “I hear an old sombre tide awash in the headboard” (ST 28); “Listening to the rain drip off the trees” (SL 59); “[t]he sea hushed and glittered” (EL 25); “at the back of a garden, in earshot of river water” (DC 72); “[a] white wing beating high against the breeze” (HC 85 ). 71 Since ‘voice’ is an important concept for Heaney and in Heaney scholarship, in this chapter, I have separated my study of human voice from other anthrophonic sounds. 72 In this chapter, I am using the term ‘element’ in the general sense of the term in OED as “a component part of a complex whole” − i.e. major components of the soundscape. For an elemental approach − the four elements of earth, water, air and fire − to Heaney’s work see Tobin, Garden and Allen, for instance. See also 2.2 and 2.3. 73 Examples of literal and metaphorical references to geophonic sounds in Heaney’s poems include: “water goes over / like villains dropped screaming to justice” (DN 40); “the wind heavy / With spits of rain” (DD 19); “mizzling rain” (WO 4); “[b]ehind a windbreak wind is breaking through” (N XI); “rustling and twig-combing breeze” (FW 30); “wind blowing round the car” (SI 64); “From a hot spring, I could hear nothing / But the whole mud-slick muttering and boiling” (HL 39); “I hear an old sombre tide awash in the headboard” (ST 28); “Listening to the rain drip off the trees” (SL 59); “[t]he sea hushed and glittered” (EL 25); “at the back of a garden, in earshot of river water” (DC 72); “[a] white wing beating high against the breeze” (HC 85 ). 71 Since ‘voice’ is an important concept for Heaney and in Heaney scholarship, in this chapter, I have my study of human voice from other anthrophonic sounds. 72 In this chapter, I am using the term ‘element’ in the general sense of the term in OED as “a compone a complex whole” − i.e. major components of the soundscape. For an elemental approach − the four ele earth, water, air and fire − to Heaney’s work see Tobin, Garden and Allen, for instance. See also 2.2 and 73 Examples of literal and metaphorical references to geophonic sounds in Heaney’s poems include: “w y y p 72 In this chapter, I am using the term ‘element’ in the general sense of the term in OED as “a compone a complex whole” − i.e. major components of the soundscape. For an elemental approach − the four ele ( ) 75 For instance, “angry frogs” and “gross-bellied frogs” (DN 3); “an unlicensed bull” (DD 6); “the wolf has died out” (WO 35); “like a dog turning / its memories of wilderness” (N 33); the “[a]live and violated” oysters (FW 3); the “fortified and bewildered” lobsters (SI 16); “the peacock’s feather in the grass” (HL 41); “the roosters in the farm” (ST 32); “we were killing pigs” (DC 7); “gulls in excelsis / Bobbed and flashed on air” (EL 25); “it’s 3); the “fortified and bewildered” lobsters (SI 16); “the peacock’s feather in the grass” (HL 41); “the roosters in the farm” (ST 32); “we were killing pigs” (DC 7); “gulls in excelsis / Bobbed and flashed on air” (EL 25); “it’s you, blackbird, I love” (DC 75); “the cattle” (HC 11). y , , ( ); ( ) 76 For an environmental approach to Irish animals, see Cusick’s Out of the Earth (2010) and “A Capacity for Sustained Flight” (2015), as well as Potts’s Contemporary Irish Writing and Environmentalism (2018). 74 For animal symbolism in Irish literature see Kirkpatrick and Faragò’s Animals in Irish Literature and Culture (2015). 3); e o ed d bew de ed obs e s (S 6); e pe coc s e e e g ss ( ); e oos e s the farm” (ST 32); “we were killing pigs” (DC 7); “gulls in excelsis / Bobbed and flashed on air” (EL 25); “it’s you, blackbird, I love” (DC 75); “the cattle” (HC 11). Introduction In soundscape studies, the major elements of the natural environment are ‘given voice’ in the form of wind in trees, rocks and grass, water flowing in a stream or as raindrops and ocean waves, as well as movement of sand dunes and rocks, among others.72 Geophonic sounds have countless manifestations throughout Heaney’s poetry.73 The transposition of the sounds of natural phenomena into his poetry has, on the one hand, allowed the poet to evoke the cultural, social and historical associations underlying the apparent topography and geography, and thus to bridge the “lived, illiterate and unconscious” and the “learned, literate and conscious” (PO 93 131). On the other hand, it has enabled him to expand and enrich the world of his mind and poetry. Just as natural phenomena engender and convey various cultural, social and historical associations, the complexities and transformations in the distribution of biophony − i.e. sounds produced by non-human sentient organisms − can reflect certain characteristics of landscapes and societies living within them. Irish folklore and poetry have a long-standing tradition of featuring animals.74 Heaney’s poetry is populated with a wide variety of river and sea animals, such as the eel, salmon, trout, lobster, oyster; farm animals such as horse, pig, cow, sheep, dog and cat; other land animals including wolf, deer, rabbit, rodents, rat, ferret; and a variety of birds such as the snipe, blackbird, swan, nightingale, cuckoo, corncrake and magpie.75 His poetry has been approached by several scholars for the symbolic significance of native animals as well as to illustrate the poet’s eco-awareness.76 Heaney makes effective use of animals and their sounds to tell us about himself and the world around him. My examination of the references to animals in Wintering Out, in the second section of this chapter, builds upon previous studies of animals in Heaney’s work but introduces the analytical lens of soundscape ecology and acoustic ecology. In the third section of this chapter, I focus on the type of sounds our ears might interpret as unnecessary human ‘noises’. Over history, transformations in the quality, intensity and spread of anthrophonic sounds − i.e. all sounds produced by humans and human-made objects (Pijanowski et al. 1214) − have led to a change in our perception of sound and noise (See 1.2). Yet, as Schafer has suggested, it is our definition of ‘noise’ that 94 underlies such discussions and interpretations (4). 77 Examples of vocal sounds include: “Bloody pups” (DN 11 ); “[t]hen grunts and goes in” (DD 9); “my children weep out” (WO 49); “‘Go Back’ one said ‘try to touch the people” (N 64); “‘Oh, Sir Jasper, do not touch me!’ / You roared across at me” (FW 13); “[a]ll the time they were shouting, ‘Shop! // Shop!’” (SI 78); “Then voices over, in different Irishes, / Discussing translation job and rates per line” (HL 47); “Youngsters shouting their heads off in a field” (ST 8); “I coughed and coughed and coughed” (SL 44); “‘Light came from the east’, he sang” (EL 57); “‘Well, for Jesus sake’ cried Duffy coming at me” (DC 29); “[w]aving and calling something I cannot hear” (HC 11). Introduction Interestingly, Heaney’s poetry is not only a medium for his own voice, but also a platform for a polyphony of voices − in the form of crying, laughing, humming, grunting, calling, naming, whispering, 95 screaming, yelling, singing or speaking.77 Since the beginning of his career, Heaney has been aware of his poetry as an enabling and empowering medium for those who unlike him did not have this platform. When asked about the transition from ‘I’ in the first collection to ‘we’ in the latter, he remarks that Not many poets had come to the fore in that particular group [Northern Irish Catholic with a nationalist background] […]. From the beginning, I was conscious of a need to voice something that hadn’t got voiced, to tune the medium in order to do that particular job. (SS 90) My analysis of the representation of human vocal sounds in Wintering Out aims at identifying and interpreting the human voices evoked by the poet. An ecological study of vocal sounds takes into consideration the relationship between human voice and its ‘ecology’. It is the study of human voice in relation to the physical environment it inhabits (B. R. Smith 16-17). Therefore, in this final section, I explore the social and political norms governing voices of the individuals and communities inscribed in the poems. This study, in summary, leads to a more complete appreciation of the sounds evoked by the poet and can give us the resources for improving our understanding of his poetry.78 p p g 79 The title of this section is inspired by a line from ‘The Rain Stick’ (SL 1). Introduction On many occasions, in his poetry, prose, as well as interviews, as Vendler has remarked, Heaney makes himself into an “anthropologist” of not only his own culture but also human civilisation through auditory experiences (18). It is hard to read Heaney’s work and not be struck by the sounds of human activities and human artefacts: the squeak of a hand-operated water “plunger slugging up and down” (PO 17); the shunting of the train at the station in Castledawson (SS 8); the “clean rasping sound / When the spade sinks into gravelly ground” (DN 1); “the hammered anvil’s short pitched ring” from inside the forge (DD 9); the clatter and clang of the wool factory (WO 27); “the tick of two clocks” in the kitchen (N x); the gargling of a tractor (FW 41); the “hurry of bell-notes” during the “morning hush” (SI 61); the “shunting” of the car engine (HL 4); the whirring of a bicycle’s tyre (ST 46-47); the rifle’s “bullet’s song” (ST 77); the groaning of a lorry’s engine (SL 13); “the throttle and articulated whops / Of a helicopter crossing” (DC 42); or the kettling sound of the boiler as it “comes to life / [a]bruptly, drowsily” (HC 4). Some of these sounds are discrete, others are continuous and monotonous. Some carry a cultural value, others are perceived as unwanted and disturbing. The purpose of this section is to identify and interpret Heaney’s attitude towards such human-produced sounds in Wintering Out. In the final section of this chapter, I focus on the representation of human voice. Nowell Smith remarks that any poem is “constructed out of voice as material or medium” and comments: “poems display, or stage, or generate, a speaking voice, or speaking voices, and these effects are registered as we readers, silently or aloud, are invited to ‘voice’ a poem” (1). The unique and distinctive qualities of Heaney’s poetic voice have attracted the attention of many scholars. Much work has been done to examine its vibrant diction, intense imagery and application of powerful symbols, metaphors and allusions (see 2.3). something I cannot hear (HC 11). 78 At no point does this study attempt to exhaust acoustic references in Heaney’s poetry. Rather, it desires to use examples that illustrate the importance of sounds and listening. the east , he sang (EL 57); Well, for something I cannot hear” (HC 11). 3.1 Through the Ear of a Raindrop: Geophonic Sounds79 The aquatic resonance of the amniotic fluid, the heartbeat, blood running through the veins and then the laps, splash and gurgle of the waters from the oceans, rain and rivers are amongst the first sounds perceived by the human ear (Schafer 15). In this section, I trace and examine geophonic sounds evoked in Heaney’s poetry focusing mostly on water as the major 96 geophonic sound-source in Wintering Out. The sound of water is a fundamental element of the natural soundscapes and the sound that often gives the listener delight in its varied transformations. As a dominant feature of the Irish climate and landscape, water, in its various forms, has numerous manifestations throughout Heaney’s poetry. From ‘Death of a Naturalist’ and ‘Personal Helicon’, in his first collection, to the water-logged poems of Wintering Out, to the “misty” atmosphere of ‘The Strand at Lough Beg’,80 to ‘The Sound of Rain’ in Seeing Things and the waves in ‘Postscript’ from The Spirit Level, among others, the sound of water is recalled not only as a soundmark and a keynote of the Irish landscape, but also as an acoustic symbol to speak of cleansing, purification, change, renewal, healing, eternity, serenity and perseverance. In the following paragraphs, I trace these implications in the references to the sounds of sea, river and rain, concluding this section with a brief consideration of some of the poems in which the sound of wind is to the fore. The sea is one of Heaney’s primary sources of inspiration. The patterns of the sea are many: from the scarcely perceptible rippling of the wavelets, to rolling breakers, enormous tidal waves, and to the daily ebb and flow, each drop rings at a different pitch and each wave has a different velocity. Seascapes affect and reflect the life, mood and language of the people within earshot. When calm, the sea is a “sound romance” that conjures up nostalgic feelings and when worked into anger, it is a brutal force and “trembling presence” that cares nothing about human hopes and fears (Schafer 170-171). The relationship between the sea and the land is made audible at the shore. As Heaney puts it, “each drew new meaning from the waves’ collision” (DN 34). 80 Heaney recalls it as a poem with poignant watery moments at The John Adams Institute, in Amsterdam, in 1993. See www.youtube.com/watch?v=2bEmyan5cMA (30:40-35:20). Heaney recalls it as a poem with poignant watery moments at The John Adams Institute, in Amsterdam, in 1993. See www.youtube.com/watch?v=2bEmyan5cMA (30:40-35:20). Heaney recalls it as a poem with poignant watery moments at The John Adams Institute, in Amsterda y p p g y 1993. See www.youtube.com/watch?v=2bEmyan5cMA (30:40-35:20). 3.1 Through the Ear of a Raindrop: Geophonic Sounds79 The Flaggy Shore, the long stretches of Donegal coast and Kerry coast, and the West coast of Ireland often appear in Heaney’s poetry to recall the traditional rural Irish lifestyle and to transpose the poet to his memories of the coastal Irish 97 landscape, but also to convey the concepts of life, hope, transformation, separation, adventure, danger and even death. landscape, but also to convey the concepts of life, hope, transformation, separation, adventure, danger and even death. On several occasions, Heaney’s vivid descriptions capture and manifest the orchestrations of the sea to reflect the dynamics and variations of his themes. In ‘Valediction’, from his first collection, Heaney uses marine imagery to depict the sense of stability and liveliness that Marie Devlin’s “presence” brings to him: In your presence time rode easy, anchored on a smile: but absence rocked love’s balance, unmoored the days. (DN 33) In your presence time rode easy, anchored on a smile: but absence rocked love’s balance, unmoored the days. (DN 33) In your presence time rode easy, anchored on a smile: but absence rocked love’s balance, unmoored the days. (DN 33) In Marie’s presence, all is well in the poet’s inner and outer world, but when she is gone, he loses the sense of time and the command of his feelings and thoughts. Heaney reinforces the concept by unveiling the association between her “flower-tender / voice” and the acoustics of the sea pleading her to return and retune his days with her soft and melodious voice (DN 33). In ‘Lovers on Aran’, the eye and the ear work side by side to transfer a full-fledged sense of the place: “The timeless waves, bright, sifting, broken glass, / came dazzling around, into the rocks” (DN 34). The poem refers to the shores of the Aran Islands at Galway Bay, on the West Coast of Ireland, where the couple spent a holiday. Heaney continues to use marine imagery of ‘Valediction’ to celebrate “the unity and exhilaration of love and marriage” (Parker 72). He is at the sea, where the sight sets a vibrant scene before his eyes and allows him to envision his marriage in the embrace of tides and the rocks, and where the comforting patterns of the waves can “penetrate” his awareness (Ihde 81). 3.1 Through the Ear of a Raindrop: Geophonic Sounds79 Similarly, in ‘Postscript’, Heaney presents a blend of his auditory and visual experiences of “County Clare, along the Flaggy Shore” (SL 70) to evoke a profound sense of the place for himself as well as for his readers. He describes the poem as having the ability to seize a fleeting moment of poetic In Marie’s presence, all is well in the poet’s inner and outer world, but when she is gone, he loses the sense of time and the command of his feelings and thoughts. Heaney reinforces the concept by unveiling the association between her “flower-tender / voice” and the acoustics of the sea pleading her to return and retune his days with her soft and melodious voice (DN 33). 98 revelation and a glimpse of the past memories:81 “They [some poems] leave you with a sensation of having been visited, and this was one of them. It excited me, and yet publishing it in The Irish Times was, as much as anything else, a way of sending a holiday postcard – a PS of sorts – to the Friels” (SS 366). In the poem, he encourages his readers to let their imagination travel with him from scene to scene. By experiencing places through movement, a person enters into a “dialogue” with the environment (Järviluoma et. al. 31). Aware of the rush of modern man and his affinity with machinery, Heaney invites his readers to do so even while driving, perhaps with the car windows open, so the wind breathes in the voice of the foaming “ocean” from one direction and the silence of the glittering “lake” from the other: “Useless to think you’ll park and capture it / More thoroughly. You are neither here nor there, / A hurry through which known and strange things pass […]” (SL 70). ‘Ballynahinch Lake’ captures and recalls a memory of a Sunday morning in Connemara when Heaney and his family drove through Co. Galway. This time, however, he invites his readers to stop and indulge in the sounds and the sights of the landscape “more thoroughly” (SL 70): “[…] that this time, yes, it had indeed / Been useful to stop […] (EL 26). In Wintering Out, the poems ‘Shore Woman’ and ‘Maighdean Mara’ depict other contrasting connotations of the sea and its keynote to depict some of the predicaments surrounding women. coast of Galway Bay while the poet was teaching at Oxford: It came from remembering a windy Saturday afternoon when Marie and I drove with Brian and Anne Friel along the south coast of Galway Bay. We had stopped to look at Mount Vernon, Lady Gregory’s summer house – still there, facing the waters and the wild; then we drove on into this glorious exultation of air and sea and swans. (SS 366) 81 Heaney refers to the poem as a “surge of utterance” that represents a glimpse of past memory of 81 Heaney refers to the poem as a “surge of utterance” that represents a glimpse of past memory of the south coast of Galway Bay while the poet was teaching at Oxford: 81 Heaney refers to the poem as a surge of utterance that represents a glimpse of past memory of the south coast of Galway Bay while the poet was teaching at Oxford: 3.1 Through the Ear of a Raindrop: Geophonic Sounds79 ‘Maighdean Mara’ − literally meaning the ‘maiden/virgin of the sea’ − is based on a version of the mermaid legend in Northern Ireland and depicts the failed attempt of a female selkie to return to her original home, the sea, after enduring several years of enslavement on land. For Heaney, the story is an allegory for a real-life suicide of an Irish woman by drowning. During the 1970s, in Irish society, single mothers were still treated as 99 outcasts and considered sinful, often bullied by the family and the Church to give up their children (Nugent et al. 372). Heaney offers his perspective on the matter in ‘Maighdean Mara’, as well as in poems like ‘Limbo’ when he writes “A small one thrown back / To the waters” (WO 58). In these poems, the presence of water helps broaden the vision and emotions of his readers to perceive the entirety of the circumstances concerning the event. Rather than interpreting the incident as sinful and unfortunate, he invites us to see the woman’s suicide as a fearless attempt to break free from the confining laws and taboos of Irish society and return to the purity and comfort of the original life element to which she truly belongs. His tripartite poem, ‘Maighdean Mara’, opens with the relaxing acoustics of the sea. Undercurrents are drifting towards the coast where the woman’s cold and naked body is laid, embracing it, and then retreating back to the sea. The opening lines, which also provide the final refrain, engulf the poem in what Schafer categorises as “biological” wave rhymes (16): She sleeps now, her cold breasts Dandled by undertow, Her hair lifted and laid. Undulant slow seawracks Cast about shin and thigh, Bangles of wort, drifting Liens catch, dislodge gently. (WO 56) She sleeps now, her cold breasts Dandled by undertow, Her hair lifted and laid. Undulant slow seawracks Cast about shin and thigh, Bangles of wort, drifting Liens catch, dislodge gently. (WO 56) Dandled by undertow, Her hair lifted and laid. 3.1 Through the Ear of a Raindrop: Geophonic Sounds79 Undulant slow seawracks Cast about shin and thigh, Bangles of wort, drifting Without the single discordant word “cold”, almost the entire passage is an evocative and beautiful rendition of life, lyrical in its language and acoustic image: the present tense of the verbs conveys the concept of life; the clusters of soft sibilants introduce a soothing effect; and, the balance of enjambment and punctuation regulates the reader’s breath with the rhythm of ebb and flow. Heaney carefully portrays the scene: the woman’s dandling breasts, the rising of her hair in each gentle and steady coming of the waves, and the undulating bangles of marine plants present her lifeless body “with the patterns of heart and lung” (Schafer 16), giving the impression that the reunion has granted the woman an eternal life and 100 peace. In each wave, the sea brings forth “seawracks” and “bangles of wort” to bedeck her daughter from shin to chest: “This is the great first sleep / Of homecoming […]” (WO 56). A similar sea wave pattern can be perceived at the opening and closing stanzas of ‘Shore Woman’. The poem is “a sophisticated blend of observation and reminiscence, realism and fancy” moving from “mood to mood, image to image, [and] thought to thought” to depict the dynamics of the marital relationship between the silent fisherman and his unhappy wife (Andrews 69). Although the opening epigraph of the poem links men to “the hills” and women to “the shore” and despite the fact that, in myth and literature, water has often been a primary symbol of rebirth, fertility and purity, the woman-narrator herself chooses to remain in the liminal space between sea and land, to be a shore-walker: “I have rights on this fallow avenue” (WO 54-55). For her, the sea is the state of terror, brute force, chaos and masculine sexuality, while the shoreline is the place she can experience “the taste of safety” and exult to the “moonlight”. Although at some point, the woman-narrator says she is “conscious” of what is happening around, it is the sea symphony that gives voice to the hidden depths of her psyche (WO 54-55). The opening stanza of ‘Shore Woman’ is an audio-tactile presentation of the shoreline as mediated through the wind and the waves. f ( ) 83 Other examples of references to seascape include: “[a]t dusk, horizons drink down sea and hill” (DD 11); “You might think that the sea is company, / Exploding comfortably down on the cliffs, / But no: when it begins, the flung spray hits / The very windows, spits like a tame cat / Turned savage” (DN 38); “the longed-for tidal wave / of justice” and “a great sea-change” (CT 2). 82 See Auden’s The Enchafed Flood (1950). 82 See Auden’s The Enchafed Flood (1950). 3.1 Through the Ear of a Raindrop: Geophonic Sounds79 Here, the regular, continuous and indolent coming and going of tidal froth registers a pleasant tempo that corresponds with “relaxed breathing pattern” and “the sense of well-being” (Schafer 227). The ubiquitous nature of the wind, on the other hand, brings the woman into touch with the “riddling” sand and “whistling” grass from far away assuring her that no one except her walks along the shoreline for miles. The woman’s calm strides on “the firm margin” further reinforce the sense of security and solidarity for which she returns to the shore after accompanying her husband on a fishing journey (WO 54-55). As we move further into the sea, in the next section of the poem, rhythm gives way to disorder, silence to white noise, and the sense of security to 101 uncertainty and threat. The tension between the woman and her husband is made audible in the slapping of “the mackerel” on the boat, “the close irruption” of the porpoises from the water, “the flywheels of the tide” and the “rocking boat” (WO 54-55). Here, the sea brings the woman into touch with “suprabiological rhythms” giving the impression of an “immense and oppressive power expressed as a continuous flow of acoustic energy” (Schafer 170-171), which she associates with the man. For her, the sea is the zone of the unknown and the unknowable, and the land, the zone of safety and comfort. After experiencing what Auden refers to as “the state of barbaric vagueness and disorder” of the sea (6-7),82 the woman feels the need to return to the shoreline, where rhythm replaces chaos and the sea becomes benign: “I sometimes walk this strand for thanksgiving / Or maybe it’s to get away from him” (WO 55). 55). For the woman, the shore is the place where she can seek the muse of the moon, let her imagination wander freely in the equivocal world of the wind, and rhyme her breaths with the gentle coming and going of the waves. Yet, like the poet, she is far from forgetting the tales of sufferings and gasping breaths lying out “[i]n darker fathoms” of her memories (WO 55). After returning from the sabbatical year in Berkeley and witnessing the deteriorating sectarian violence in Northern Ireland, Heaney decided to move to Co. Wicklow to fulfil his need for solitude and commit himself totally to writing, getting away from the duties of the academic profession and societal obligations. In the poem ‘North’, from the collection of the same title, which was published about three years after he had left Ulster, Heaney writes: “I returned to a long strand” (N 10), not to escape the violence, but to find inspiration and to reconcile with himself.83 Rain, another determining factor and a recurrent feature of the Irish landscape, 102 appears frequently in Heaney’s poetry.84 In ‘Broagh’, the poet rejoices in the correspondence between the shape and sound of the place-name and its vernacular geophony. But before bringing up the analogy, he invites his companion to undertake what Schafer calls “ear cleaning” (272) and train his ears to perceive rain and wind, the two major elements of the ambient sound in rural Ireland: […] the shower gathering in your heelmark was the black O ended almost suddenly, like that last gh […] [emphasis in the original]. (WO 17) Keynotes inform a community’s listening habits and lifestyle throughout time. They are not listened to consciously, but heard so frequently that they form an imprinted connection to the sense of time and place (see 1.4.2). The poem ‘Broagh’ is the creation of just another “soundwalk” (Schafer 212) on the familiar hills of Ulster and the poet’s attempt to bring into consciousness the sense of territory, history, and community by underpinning the everyday acoustics of the land: “its low tattoo” (WO 17). The sound of rain on the windswept vegetation, which Heaney hears in Broagh, is unlike the sound of raindrops that “hit the roof with smacking little clicks, uneven and stabbing [...]. It is more like a continuous sigh, a breath always spending with no fresh intake” (Carr 406). g g “It’s raining on black coal and warm wet ashes” (SL 13). 84 Other positive references to rain include: “[r]ain and hay and woods on the air / Made warm droughts in the open car” (DD 24); “the road, the mountain top, // and the air, softened by a shower of rain” (SI 65); but the monotonous drumming of rain in ‘Two Lorries’ connotes to his bleak memories of the IRA during the nineties: “It’s raining on black coal and warm wet ashes” (SL 13). 84 Other positive references to rain include: “[r]ain and hay and woods on the air / Made warm droughts in the open car” (DD 24); “the road, the mountain top, // and the air, softened by a shower of rain” (SI 65); but the monotonous drumming of rain in ‘Two Lorries’ connotes to his bleak memories of the IRA during the nineties: open car (DD 24); the road, the mountain top, // and the air, softened by a shower of rain (SI 65) monotonous drumming of rain in ‘Two Lorries’ connotes to his bleak memories of the IRA during the and identity in my life all coinciding with, as you say, love I do regard it as a real benediction. And, of course, there’s the whole matter of friendships and family solidarity and the trust of cherished ones. , p y y See www.ricorso.net/rx/library/criticism/revue/Zundry_AZ/Cole_H.htm. his 1997 conversation with Henri Cole, Heaney confirms the extension of the metaphor to himself: I still think I have been inordinately lucky. I regard first of all the discovery of a path into the writing poems as luck. And the salute that my early poems received and the consequent steadying of directio and identity in my life all coinciding with as you say love – I do regard it as a real benediction An , y p I still think I have been inordinately lucky. I regard first of all the discovery of a path into the writing of In his 1997 conversation with Henri Cole, Heaney confirms the extension of the metaphor to himself: I still think I have been inordinately lucky. I regard first of all the discovery of a path into the writing of poems as luck. And the salute that my early poems received and the consequent steadying of direction and identity in my life all coinciding with, as you say, love – I do regard it as a real benediction. And, of course, there’s the whole matter of friendships and family solidarity and the trust of cherished ones. 85 In his 1997 conversation with Henri Cole, Heaney confirms the extension of the metaphor to himself: I still think I have been inordinately lucky. I regard first of all the discovery of a path into the writing of poems as luck. And the salute that my early poems received and the consequent steadying of direction and identity in my life all coinciding with, as you say, love – I do regard it as a real benediction. And, of course, there’s the whole matter of friendships and family solidarity and the trust of cherished ones. See www.ricorso.net/rx/library/criticism/revue/Zundry_AZ/Cole_H.htm. 55). For Heaney, the continuous and gentle shower that registers as the keynote of Broagh is like the extended exhalation that breathes out the sound of the letter O in the place-name; one that could cease almost as abruptly as the final throaty gh brings the word to an end. There is a linguistic and phonetic 103 politics at play here. Despite the fact that ‘Broagh’ is an anglicisation of the original Gaelic ‘bruach’ (O’Brien 16), Heaney’s description details the affinity between the place-name and its soundscape, and highlights its unique Irishness from the British Empire of which it is now a part (see also 3.4). The ceaseless rainfall in ‘The Sounds of Rain’ and rainstorm of ‘Gifts of Rain’ are nothing like the breath-like precipitation readers perceive in ‘Broagh’. In ‘The Sounds of Rain’, rain drums “[o]n the veranda” (ST 48), tapping at the poet’s inner ear persuading him to listen to the assuring voices of the fellow authors lying “in the long moil” of his subconscious. In the middle and longest part of the three-fold poem, Heaney unfolds reminiscences of meeting the Russian Nobel Laureate Boris Pasternak and William Alfred, colleagues at Harvard. It is as if the poet awakens from a long sleep or returns from a long state of trance to the recognition of his vocation as a poet: state of trance to the recognition of his vocation as a poet: state of trance to the recognition of his vocation as a poet: The eaves a water-fringe and steady lash Of summer downpour: You are steeped in luck, I hear them say, Steeped, steeped in luck. And hear the flood too, gathering from under, Biding and boding like a masterwork Or a named name that overbrims itself [emphasis in the original]. (ST 49) Or a named name that overbrims itself [emphasis in the original]. (ST 49) The “steady” lashing of the rain on the roof that provokes the poet’s thoughts and imagination is the mantra of blessing and bounty that promises the ultimate overflow of “a masterwork” (ST 49).85 Very often, especially in his later years, Heaney acknowledges having been “lucky” to have discovered a path into the writing of poems and to have experienced a steady personal and professional growth in this path, an element that was accompanied with support and love from family and friends. See www.ricorso.net/rx/library/criticism/revue/Zundry_AZ/Cole_H.htm. 86 See www.youtube.com/watch?v=AdTQPZGW0Bs (0:05-0:30). 86 See www.youtube.com/watch?v=AdTQPZGW0Bs (0:05-0:30). 87 The townland takes the title of some of Heaney's best-known poems − ‘Anahorish’ and ‘Anahorish 1944’ − in which Heaney gives voice to the locals Owen and Roddy Gribbin. For disputes over the name see: l i / l d il h ? 3 2 d i i h / /2013/09/09/ /h 87 The townland takes the title of some of Heaney's best-known poems − ‘Anahorish’ and ‘Anahorish 1944’ − in which Heaney gives voice to the locals Owen and Roddy Gribbin. For disputes over the name see: www.placenamesni.org/resultdetails.php?entry=5352 and www.irishnews.com/news/2013/09/09/news/heaney- townland-doesn-t-exist-69745/. 55). In his 2012 American Ireland Fund AWB Vincent Literary Award acceptance speech, he says “a woman from Mayo […] The “steady” lashing of the rain on the roof that provokes the poet’s thoughts and imagination is the mantra of blessing and bounty that promises the ultimate overflow of “a masterwork” (ST 49).85 Very often, especially in his later years, Heaney acknowledges having been “lucky” to have discovered a path into the writing of poems and to have experienced a steady personal and professional growth in this path, an element that was accompanied with support and love from family and friends. In his 2012 American Ireland Fund AWB Vincent Literary Award acceptance speech, he says “a woman from Mayo […] 104 once said to me, you are steeped in luck, and after this evening, after this award I feel myself deeper still in that happy, lucky element”. Here the sound of rain too seems to remind him of what he perceives as “real benediction[s]”. 86 In ‘Gifts of Rain’, however, the rainfall has a different implication. The title of the poem may connect with rain as a heavenly gift suggesting a mythical and healing atmosphere to follow. Yet the auditory images of the opening line and the tactility of the experience disclose the poignant irony of the title: A nimble snout of flood licks over stepping stones and goes uprooting. He fords his life by sounding. Soundings. (WO 13) The “cloudburst and steady downpour” with which Heaney opens the poem are the archetypal acoustics of bleak and threatening conditions that “he begins to sense […] / by his skin” (WO 13). This time the gifts of rain are not bounty, purity, joy and life. They are life- threatening. Written while in Berkeley, the poem borrows its acoustic images from the poet’s memory bank and his contemplation on the effect of the persistent socio-political upheavals on the life and landscape of Ulster. Heaney’s reference to the noisy turbulence of the flood, as the immediate aftermath of rain, and the tense silence of the vulnerable land dwellers suggest the pervasiveness of the conflicts and the depth of their impact on the life and well-being of people in Northern Ireland. y g y p www.placenamesni.org/resultdetails.php?entry=5352 and www.irishnews.com/news/2013/09/09/news/heaney- townland-doesn-t-exist-69745/. 55). Yet, at the end of the passage, he adds that although the flooding river forges its way by “uprooting” the “stepping stones” and despite the fact that its mighty 105 voice takes over the quiet voices of the inhabitants, the Ulster men manage to survive and remain a “sounding” entity among the “[s]oundings” (WO 13). As Parker comments: To survive, he must acknowledge his kinship with and dependence on the rest of the natural world, for he is part of the pattern not its center. Heaney contrasts the figure’s hesitant progress with the assured movement of the water, which seems to possess a sense of purpose. (100) Heaney’s childhood memories are shaped by his experiences within Anahorish, a flat townland in Co. Derry that stretches from the river Moyola to Lough Beg. Anahorish, officially known as Creagh, features in Heaney’s poetry very often and is highly regarded by the locals.87 Geography and climate determine the vernacular keynotes that form an imprinted connection to our sense of place and time. Soundmarks, on the other hand, establish a unique sense of place often only experienced by the locals (see 1.4.2). In the opening line of ‘Anahorish’, the poet offers the translation of anach fhíor uisce – the English transliteration of the Gaelic name − and in doing so, introduces the sounds of water as both the keynote and soundmark of the hills on which his earliest memories are grounded: “My ‘place of clear water’” (WO 6). Water is a primary life element. It belongs to the “first soundscapes” of the world (Schafer 13). For the poet, Anahorish is, in name and essence, the source of purity, hope and life, a place where his mind is blessed by the unique acoustics of the river Moyola: the first hill in the world where springs washed into the shiny grass y p which Heaney gives voice to the locals Owen and Roddy Gribbin. For disputes over the name see: www.placenamesni.org/resultdetails.php?entry=5352 and www.irishnews.com/news/2013/09/09/news/heaney- townland-doesn-t-exist-69745/. y ( ) 89 Other references include: “We crossed the quiet river” (DN 31); “Slubbed with eddies, / the laden silent river / ran mud and olive into summer” (FW 41); “River gravel. In the beginning, that” (SL 39). Schönach’s “River histories” (2017); Cioc’s The Rhine (2009); Simmons and Scott’s “The River has Recorded the Story” (2006). 88 For more information on the sound of rivers from the perspective of environmental history see, for 88 For more information on the sound of rivers from the perspective of environmental history see, for instance: Schönach’s “River histories” (2017); Cioc’s The Rhine (2009); Simmons and Scott’s “The River has Recorded 88 For more information on the sound of rivers from the perspective of environmental history see, for instance: and darkened cobbles […]. (WO 6) Even the echoes of the first land-dwellers in Anahorish are characterised by water associations, but not with the vibrant acoustics of water in spring. In spring, there is a sense of urge, liveliness, and hope in the air. Heaney evokes the memories of his ancestors with the 106 acoustics of frozen water so as to acknowledge their hard work and determination during the bleak winter days: acoustics of frozen water so as to acknowledge their hard work and determination during the bleak winter days: those mound-dwellers go waist-deep in mist to break the light ice at wells and dunghills. (WO 6) A similar impression is evoked in the second part of ‘Gifts of Rain’, when a resolute Ulster country man appears squelching through the flooded land: “A man wading lost fields / breaks the pane of flood […]” (WO 13). The man has survived the flooding but his “reflection” in water, his hopes for a brighter and more settled future, is immediately distorted as “a flower of mud- / water blooms up” and ripples blood-like over the surface of the water (WO 13). “Lost” in the flood or to the land seizers, the kinship between the Ulster man and his land is unshakable: he “depends on” it, he is “hooped to” it, he embraces it and the universe rotates between them “naturally” (WO 14). For Heaney, in Ulster, the man is an integral part of the harmonious cycle of natural landscape and, thus, the universal symphony. Water never dies; rather it changes its state and location as rain, bubbling brooks, swirling rivers, waterfalls, fountains, bogs, lakes, the sea, oceans, and their tides. Musicians, anthropologists, historians and environmentalists have utilised local music, songs or lyrics to discuss the relationships between the sound of rivers, cultures and concept of identity.88 These studies suggest that rivers of the world have unique acoustics that only locals can fully identify and decipher. For Heaney too rivers are the repository of local history and the failure to listen to their voice and music would impact our understanding of a place and its culture.89 In ‘Gifts of Rain’, Heaney notes how the living being has been able to be a sound-producing member of the soundscape and also develop an intuitive understanding of water depth 107 through its acoustics: “sounding. / Soundings” (WO 13). 90 Heaney commented: Moyola had overflowed and flooded some of the houses in Broagh. Older people would occasionally remember events and date them by asking, ‘Was that before or after the flood? So ‘Gifts of Rain’ came out of this sense of belonging to an antediluvian world […]”. (SS 125) 91 Traditionally, rain sticks are made from dried out cactus tubes and used to summon supernatural powers to change the weather. When the cacti die, the waxy green pulp dries out and the cacti spines are pressed inward. The hollow tube is gradually filled with grit before the ends are sealed. If the stick is upended, the grit falling and darkened cobbles […]. (WO 6) As mentioned earlier, an ecological approach to listening explores auditory perception along the dimensions of a sound producing source and its location, proposing that through listening the human ear can identify such information as the material, texture, and configuration of the objects, the force and the type of the interaction involved, as well as the proximity, time and direction of the sound event (see 1.3). In the third section of the poem, there is a sense of the flood getting stronger and more aggressive during the night but the poet assures his readers that the well-trained ears of the locals were able to discern every note amidst the chord of many notes that Moyola persistently played on “its gravel beds” (WO 14). The changes in level of the water (“roaring off the ford”), direction, proximity and velocity (“the race / slabbering past the gable”) were the information that “[t]heir world-schooled ear // could monitor” in the absence of daylight (WO 14). ‘Gifts of Rain’ originates from the poet’s memories of the conversations of the locals he had overheard during his teenage years.90 Returning to the acoustic images of Moyola’s flooding recalls the poet’s own celebration of the river Moyola as “the source of his initiation into music, [and] his baptism in sound” (Parker 101). Based on an ecological approach to listening, the distinction between “everyday listening” and “musical listening” relies on the kind of the auditory experience and not on the sounds per se. It is possible, at any time, to listen to the perceptual attributes of the sound − e.g. harmony and sensuousness − or the attributes of the sound-producing event − e.g. intensity of rain (Gaver 1-2). In ‘Gifts of Rain’, Heaney’s attentive ears also celebrate the transpiring music in Moyola’s everyday cocktail of sounds: “The Moyola harping on // its gravel beds […]” (WO 14). His interpretation of Moyola’s “usual / confabulations” (WO 14) may remind readers of Hardy’s description of 108 rivers at night as “a lampless orchestra, all playing in their sundry tones from near and far parts of the moor”, which he encourages the wanderer to “stand still” and attentively listen to (Hardy 341). Traditionally, rain sticks are made from dried out cactus tubes and used to summon supernatural pow y p p change the weather. When the cacti die, the waxy green pulp dries out and the cacti spines are pressed The hollow tube is gradually filled with grit before the ends are sealed. If the stick is upended, the gr and darkened cobbles […]. (WO 6) Heaney invites his readers to school their ears and learn through listening but also to appreciate the sheer poetry and music inherent in the acoustic environment: rivers at night as “a lampless orchestra, all playing in their sundry tones from near and far parts of the moor”, which he encourages the wanderer to “stand still” and attentively listen to (Hardy 341). Heaney invites his readers to school their ears and learn through listening but also to appreciate the sheer poetry and music inherent in the acoustic environment: […] Moyola […] Moyola is its own score and consort, is its own score and consort, bedding the locale in the utterance, reed music, an old chanter reed music, an old chanter breathing its mists through vowels and history. (WO 15) For the poet, Moyola is wedded to the local land and impregnated with the shared history. As Tobin has noted, “spelling itself, the river becomes a transcendental signifier” (82). It is a poem in which language and land have reached a perfect sexual union. It is also a self- contained orchestra. Heaney avers that Moyola’s “reed music” breaks from the chest of the native land speaking with the familiar “guttural” voice and singing the “old chanter” of Ulster (WO 15). Heaney elaborates further on his perception of the ‘music’ of everyday sounds in ‘The Rain Stick’. Earlier, I explained how the differentiation between ‘noise’ and ‘music’ can be subjective and reliant on social and personal preferences, as well as how acoustic ecology calls for a culture of ‘ear cleaning’, a culture in which we train our ears to listen more “discriminatingly” to all sounds (Schafer 272) (see 1.2 and 1.4.2). In the poem ‘The Rain Stick’, Heaney encourages his readers to take a moment and recognise the never-aging music of “what happens”: “Upend the rain stick and what happens next / Is a music that you never would have known / To listen for” (SL 1).91 The kind of music that Heaney invites his 109 readers to recognise is the music of every sound and everyday sounds. A music that “needs no preamble” and has in it “the sense of accident, chance, and sheer provisionality” (Leighton 21). It is a music that could be dismissed as the sound of rain or noise of pebbles if not “listened for” (SL 1). j y 92 The poem is dedicated to the literary critic and biographer, Rand Brandes, who gifted the poet a rain stick. See www.irishtimes.com/culture/books/sorting-seamus-heaney-s-study-i-could-hear-archivists-groan-as-we-threw- things-away-1.3087628. hings away 1.3087628. See: www.nobelprize.org/nobel_prizes/literature/laureates/1995/heaney-facts.html. over the nodes produces the sound of a gentle rain, which Heaney describes in this poem. See johnseandoyle.com/the-rain-stick/. and darkened cobbles […]. (WO 6) It is a music that in each occurrence brings about a new insight about life and is “undiminished for having happened once, / twice, ten, a thousand times before” (SL 1). ‘The Rain Stick’ is the first poem of Heaney’s post-Nobel Prize volume. When we first read it, the poem seems to present ample reference to elements: “Grit”, “Rain”, “breaths of air”, “grass and daisies” and “glitter”, but Heaney re-directs the attention to the transpiring music occurring between them. Heaney writes that he had occasionally heard the sound of the instrument in ethnic shops but that on one occasion, when he carefully listened to it, the experience was different and unforgettable:92 “It was so lush and I was so entranced (SS 345). The Spirit Level was published in 1996 when the poet was in his late fifties and at the height of his career. Heaney shares the harmonising and elevating effect of his experience, only to continue to acknowledge the “everyday miracles” which a year earlier the Swedish Academy had recognised in its Nobel citation.93 Towards the end of the poem, Heaney asks: Who cares if all the music that transpires Who cares if all the music that transpires Is the fall of grit or dry seeds through a cactus? Is the fall of grit or dry seeds through a cactus? Is the fall of grit or dry seeds through a cactus? You are like a rich man entering heaven You are like a rich man entering heaven Through the ear of a raindrop. (SL 1). In Stepping Stones, he comments on how the poem serves as a “reminder” for himself as well as the reader: “That first poem is as much about middle age as about a rain stick. The instruction to listen was directed more to myself than to the reader, a reminder to keep the 110 lyric faith, […]. ‘The Rain Stick’ is about being irrigated by delicious sound, about water music being created by the driest of elements – desiccated seeds falling through a cactus stalk” (SS 345). Like water, the wind possesses splendid voice variation. g ( ) 96 The album is made up of a mix of instrumental tracks and spoken poetry, and features traditional and contemporary music and lyrics. The story originates in the Blasket Islands, off the west coast of Ireland. In the story, a local fiddler hears a mysterious music floating across the night air and recreates the song onto his fiddle. Inspired by both Heaney’s poem and the original composition, in 2013, David Bruce produced his interpretation of the story. See www.davidbruce.net/works/the-given-note.asp. 94 The sound of wind depends on various factors such as the speed, direction, distance of the wind, the shape of the landscape and the objects on its way. The source of sound from wind can be solely aerodynamic − e.g. changes in the pressure − or a result of interaction between wind and other materials − e.g. when the wind passes over the rotating blades of a wind turbine (Gaver, “what” 9-17). p g ( ) 95 Other poems with reference to wind include: “Air from another life and time and place, / Pale blue heavenly air is supporting / A white wing beating high against the breeze” (HC 85); “‘See me?’ it says, / ‘The wind // Has me well-rehearsed / In the ways of the world. // Unstable is good. / Permission granted! // Go then, citizen / Of the wind. / Go with the flow’” (HC 36-37); “It’s not that I can’t imagine still / That slight untoward rupture and world-tilt / As a wind freshened and the anchor weighed” (HC 84). changes in the pressure or a result of interaction between wind and other materials e.g. when the wind passes over the rotating blades of a wind turbine (Gaver, “what” 9-17). 95 Other poems with reference to wind include: “Air from another life and time and place, / Pale blue heavenly air is supporting / A white wing beating high against the breeze” (HC 85); “‘See me?’ it says, / ‘The wind // Has me well-rehearsed / In the ways of the world. // Unstable is good. / Permission granted! // Go then, citizen / Of the wind. / Go with the flow’” (HC 36-37); “It’s not that I can’t imagine still / That slight untoward rupture and world-tilt / As a wind freshened and the anchor weighed” (HC 84). 96 The album is made up of a mix of instrumental tracks and spoken poetry and features traditional and he wind. / Go with the flow (HC 36 37); It s not that I can t imagine still / That slight untoward ruptu orld-tilt / As a wind freshened and the anchor weighed” (HC 84). 97 See www.dailymotion.com/video/x2yu7h2 (0:00-0:37). 97 See www.dailymotion.com/video/x2yu7h2 (0:00-0:37). and darkened cobbles […]. (WO 6) The sound of wind is broadband and within the wide range of its frequencies gives voice to various objects on its way, from amongst rotating blades of wind turbines, tree branches, rocks and grass from far and near.94 Like references to the sounds of water, Heaney’s poetry is characterised by a remarkably large vocabulary for wind features.95 To close the section of geophonic sounds, I examine ‘The Given Note’, from Door into the Dark, which was read at Heaney’s funeral, and ‘Had I Not Been Awake’, from his final collection, Human Chain, where the fleeting nature of the wind and its energy and ability to give life and voice to otherwise mute objects turn it into a symbol of artistic inspiration. In ‘The Given Note’, as Tobin has stressed, the wind is the presiding element and it is through wind that the source of creation is disclosed (53). The poem is inspired by a story about the traditional Irish folk song Port na bPúcaí − meaning ‘song of the fairies’ − which the poet had heard from the Irish folk composer, Sean O’Riada. In The Poet and the Piper, a 2003 studio album by Heaney and piper Liam O’Flynn, Heaney explains the origin of his poem and how Sean O’Riada’s piece, ‘The Music of the Spirits’ − also based on the Irish folk song Port na bPúcaí − inspired him too:96 111 This one’s called The Given Note and it’s really just the retelling of a story that I heard Sean O’Riada tell when he was in Belfast a number of years ago as composer at the Belfast Festival. He played a piece of music which he called the Music of the Spirits and told a story about a fiddler getting it over the air at the Blasket Island and it seemed to me an image of inspiration, a mighty wind blowing the music to you. So, I wrote it down just as a figure of craft and inspiration.97 This one’s called The Given Note and it’s really just the retelling of a story that I heard Sean O’Riada tell when he was in Belfast a number of years ago as composer at the Belfast Festival. and darkened cobbles […]. (WO 6) 112 leaves onto the roof making a tapping sound (“pattered”), absorbs the spark, energy and message of it (“the whole of me a-patter”) and captures this fleeting moment of inspiration in his poem forever: Had I not been awake I would have missed it, A wind that rose and whirled until the roof Pattered with quick leaves off the sycamore And got me up, the whole of me a-patter, Alive and ticking like an electric fence […]. (HC 3) Had I not been awake I would have missed it, A wind that rose and whirled until the roof Pattered with quick leaves off the sycamore And got me up, the whole of me a-patter, Alive and ticking like an electric fence […]. (HC 3) Had I not been awake I would have missed it, A wind that rose and whirled until the roof Pattered with quick leaves off the sycamore Had I not been awake I would have missed it, A wind that rose and whirled until the roof Pattered with quick leaves off the sycamore And got me up, the whole of me a-patter, And got me up, the whole of me a-patter, Alive and ticking like an electric fence […]. (HC 3) The wind, here, re-awakens Heaney’s internal voice and retains its marvel through his poetic imagination. In his 2010 poetry reading, Heaney explains the reasons for choosing this “windy” poem as the opening to the session, bringing up the two sides of artistic revelation, the ability to surprise and inspire, as well as the technique to be able to capture it, work it into art, and thus keep it forever: […] I’ll read it because it came very quickly to me, and it’s the way poems tend to come, I think, to poets, suddenly and, if you are lucky, joyfully and, then, it’s over […] if they’re successful, the poem doesn’t slip back into the ordinary; something of the surprise and the gift of the thing stays […].98 and darkened cobbles […]. (WO 6) He played a piece of music which he called the Music of the Spirits and told a story about a fiddler getting it over the air at the Blasket Island and it seemed to me an image of inspiration, a mighty wind blowing the music to you. So, I wrote it down just as a figure of craft and inspiration.97 In the poem, the local fiddler grasps a melody across the night air and deftly reproduces it on his fiddle; the attempts of the other villagers, however, are not as remarkable: He got this air out of the night. Strange noises were heard By others who followed, bits of a tune Coming in on loud weather Though nothing like melody. He blamed their fingers and ear As unpractised, their fiddling easy, […]. (DD 36) He got this air out of the night. Strange noises were heard By others who followed, bits of a tune Coming in on loud weather Though nothing like melody. He blamed their fingers and ear As unpractised, their fiddling easy, […]. (DD 36) While the villagers wonder about the source of the melody, the fiddler blames their “unpractised” fingers and ears for not being able to capture it. Heaney reads ‘The Given Note’ as part of the Poetry Ireland lunchtime reading series, placing emphasis on wind as a source of inspiration, on the one hand, and on the poet’s capacity and readiness to seize it, on the other. The poem can, in fact, be read as Heaney’s meditation on the origins of all artistic inspiration “given” (DD 36) to the artists from an unknown origin, on the wind as the carrier of the message and on the artists as the ones who capture it and eternalise it through their art: So whether he calls it spirit music Or not, I don’t care. He took it Out of wind off mid-Atlantic. Still he maintains, from nowhere. t comes off the bow gravely, Rephrases itself into the air. (DD 36). The poem ‘Had I not been awake’ captures one of these moments of consciousness and poetic revelation inspired by the sound of wind. Heaney recalls a moment during his recovery from a stroke in 2006. In the poem, he is woken up by a gust of wind that blows 97 See www.dailymotion.com/video/x2yu7h2 (0:00-0:37). y ( ) 99 The title of this section is inspired by the poem ‘Serenades’ (WO 50). y ( ) The title of this section is inspired by the poem ‘Serenades’ (WO 50). See www.youtube.com/watch?v=wOTrSdlsPVk (0:28-0:52). g p g See also www.poetryfoundation.org/poetrymagazine/articles/70148/omphalos. g p g , , p , g , sphere of activity”. The term was used to refer to the hand-operated water pump near the poet’s house, in Co. Derry. In the following section on anthrophonic sounds, I return to this term and how its sound and meaning have forged the poet’s imagination. 3.2 My Serenades: Biophonic Sounds99 An ecological study of sounds in any given soundscape is based on an understanding of how sounds from various geological, biological and anthropogenic sources can be used to illustrate the relationship between human beings and their sonic environment. In this sense, soundscape ecology shares considerable parallels with landscape ecology, which, as one of its main focuses, emphasises the interaction between biological and anthropogenic factors (see 1.3). Since the coupled human-animal relationship is a central focus of soundscape studies, in this section I focus on Heaney’s illustration of the interactions between humans and non- human animals starting from the rural environment he grew up in and to which he returns in his poetry. This analysis is followed by a study of biophonic sounds from a broader range of 113 soundscapes Heaney evokes. I explore how attention to sounds of particular animals and their presence or absence as major components of the soundscape can unfold the underlying political, historical, personal and environmental dimensions and thus enrich our understanding of Heaney’s poetry. In contrast with the urban landscapes that are dominated by human-produced sounds, such as cries of street vendors, friction of tyres on the roads, car horns and sirens, rural landscapes are characterised by an abundance of vocal and mechanical sounds of animals, such as barking of dogs and stridulation of crickets, and of geophonic sounds of running water and rustling wind (Pijanowski et. al. 203). Heaney grew up on a farm, where humans share their habitation with animals and where they live in close relationship with the natural world. The overlapping of human-animal acoustic spaces in rural environments is reflected in hi d i ti f th b k d f th i M b h h h ld h i f his description of the backyard of their Mossbawn house, where he could hear voices of men and women, their footsteps, rattling buckets, along with the clip-clop of horses, their drinking and snorting, and the ceaseless pumping of the water: Women came and went, came rattling, between empty enamel buckets, went evenly away, weighed down by silent water. The horses came home to it in those first lengthening evenings of spring, and in a single draught emptied one bucket and then another as the man pumped and pumped, the plunger slugging up and down, omphalos, omphalos, omphalos. 100 According to OED ‘Omphalos’ is a Greek term meaning “centre, heart, or hub of a place, organisation, p y p p p p , Derry. In the following section on anthrophonic sounds, I return to this term and how its sound and meaning have forged the poet’s imagination. 0 According to OED ‘Omphalos’ is a Greek term meaning “centre, heart, or hub of a place, organisation, 3.2 My Serenades: Biophonic Sounds99 (PO 17)100 Women came and went, came rattling, between empty enamel buckets, went evenly away, weighed down by silent water. The horses came home to it in those first lengthening evenings of spring, and in a single draught emptied one bucket and then another as the man pumped and pumped, the plunger slugging up and down, omphalos, omphalos, omphalos. (PO 17)100 In rural soundscapes, the sounds of natural elements co-exist with the sounds of animals to provide a vibrant blend of geophony and biophony. Reminiscing about his childhood, Heaney speaks about the impact of the land, the vegetation and the animals on his life and mind: childhood, Heaney speaks about the impact of the land, the vegetation and the animals on his life and mind: The world grew. Mossbawn, the first place, widened. There was what we called the Sandy Loaning, a sanded pathway between old hedges leading in off the road, first among fields and then through a small bog, to a remote farmhouse. It was a silky, 114 fragrant world there, and for the first few hundred yards you were safe enough. The sides of the lane were banks of earth topped with broom and ferns, quilted with moss and primroses. Behind the broom, in the rich grass, cattle munched reassuringly. Rabbits occasionally broke cover and ran ahead of you in a flurry of dry sand. There were wrens and goldfinches. (PO 18) Keynote sounds are the background sounds that imprint the sense of time and place in memory. They can almost instantly align themselves with our memories of a particular place and time without the need for a visual clue (see 1.4.2). The reassuring munching of the cattle, the swift flurry of rabbits running across fresh grass and dry sand, the songs of wrens and the undulating flight of goldfinches, which he refers to in the lines quoted above (PO 18), as well as the constant “cackles and clucks” (SS 8) of hens, occasional roaring of cows, grunting pigs, scampering of mice on the old ceilings of their house, comforting neighing of horses, their big body rolls and foot stamps in the stable and even the cacophony of turkeys or geese every Christmas (SS 8-15) are among the biophonic keynotes of farm life that Heaney’s attentive ears took in early in life. 3.2 My Serenades: Biophonic Sounds99 These acoustic memories re-surface every now and again in his poetry so as to re-construct the soundscape of his childhood and re-orient the poet towards the earliest bonds. Keynote sounds are the background sounds that imprint the sense of time and place in memory. They can almost instantly align themselves with our memories of a particular place and time without the need for a visual clue (see 1.4.2). The reassuring munching of the cattle, the swift flurry of rabbits running across fresh grass and dry sand, the songs of wrens and the undulating flight of goldfinches, which he refers to in the lines quoted above (PO 18), as well as the constant “cackles and clucks” (SS 8) of hens, occasional roaring of cows, grunting pigs, scampering of mice on the old ceilings of their house, comforting neighing of horses, their big body rolls and foot stamps in the stable and even the cacophony of turkeys or geese every Christmas (SS 8-15) are among the biophonic keynotes of farm life that Heaney’s attentive ears took in early in life. These acoustic memories re-surface every now and again in his poetry so as to re-construct the soundscape of his childhood and re-orient the poet towards the earliest bonds. References to farm animals and their associated sounds are scattered through Heaney’s poetry. ‘Follower’ is an early example that brings together biophonic and anthrophonic sounds related to ploughing to create a familiar acoustic image from the poet’s childhood, which he fondly associates with his father and the family tradition of farming. Although by choosing to be a poet Heaney does not “follow” in the footsteps of his father, his implicit reference to the heavy pounding of draught horses’ hooves as they “strained” at each “clicking” of his father’s tongue registers the poet’s admiration of his father’s skill and his own desire to bridge the gap between his career as a poet and his agrarian background (DN 12). In ‘Cow in Calf’, he shows his own acoustic experiences and skills demonstrating what Ihde calls “hearing interiors” through echoes and resounds (Ihde 70). Heaney detects the signs 115 of pregnancy while “slapping” the cows’ belly which feels and sounds “like slapping / a great bag of seed” and echoes “like a depth-charge / far in her gut” (DN 25). ), g ( ) 102 See Lambin and Geist’s Land-use and Land-cover Change (2006) for how human beings can transform the composition of a soundscape, and Barber et al. “The costs of chronic noise exposure for terrestrial organisms” (2009) for the impact of human noise on the distribution of animals. 101 Heaney refers to his father’s cattle trade in other poems, as in ‘Ancestral Photograph’ (DN 13-14), his memories of bulls and cows in ‘The Outlaw’ (DD 6) and ‘Mother’ (DD 19), “barking dogs” in ‘The Loaning’ (SI 52), and the sound of a “trotting” horse in ‘Terminus’ (HL 4). 3.2 My Serenades: Biophonic Sounds99 Furthermore, to him, the growing udder resembles the “windbag” of a bagpipe “dron[ing] to her lowing” and singing the unchanging rhythm of bovine life: “Her cuts and her milk, her heats and her calves / keep coming and going” (25). In ‘First Calf’ from Wintering Out, Heaney does not refer to the sounds of the cattle directly; instead, he focuses on a deep sense of silence and absence.101 Nonetheless, there is no doubt the sight of a hanging placenta and a calf sucking “hard” at the cow’s full udder, which he encounters again after “a long time” and interprets as “semaphores of hurt” (WO 62), evoke the same continuous and melancholic melody of the bagpipe in ‘Cow in Calf’. The co-habitation of human and animal is not always harmonious. The increasing presence of “indiscriminate and imperialistic” human-produced sounds (Schafer 3) and the “modification of the land” by human beings (Pijanowski et. al. 204) affect the natural patterns of biophony. In the opposite direction, the transformation of the natural and unique soundscapes inflicts a reciprocal impact on the well-being of human beings and many associated human concepts, such as culture, identity, sense of place, as well as educational, artistic, and aesthetic values (Pijanowski et. al. 203-205).102 Heaney illustrates the tensions underpinning human-animal relationship in some of his best-known poems, such as ‘Death of a Naturalist’, ‘The Early Purges’ and ‘Dawn Shoot’ from his first collection, and ‘Shore Woman’ from Wintering Out. Looking at it through the lens of acoustic ecology and soundscape ecology indicates how Heaney uses biophonic sounds to show a range of tangible instances where human beings have not been able to come to terms with the idea of sharing 116 their acoustic territory with non-human animals. This theme is further developed in ‘Bogland’ from his second collection and ‘Midnight’ from Wintering Out, which depict the impact of the confiscation and plantation of the Irish lands and the extinction of the native animals. According to Vendler, ‘Death of a Naturalist’ demonstrates the death of the “naive” and “dutiful” child and his transition to a new intellectual level. She sees Heaney’s “most virtuosic moment of sound” in using the frogs’ sexual noises to “awaken self-lacerating shame” in the boy whose pastoral innocence is now contaminated by the smear of the freshly laid spawn (29-30). 103 See animals.mom.me/indigenous-animals-ireland-5197.html. 3.2 My Serenades: Biophonic Sounds99 Frogs are among the native amphibians of Ireland.103 Parker raises a similar point, noting that the poem indicates the young boy’s “fall from innocence into experience” (65). In the following paragraphs, I will indicate how Heaney’s representation of biophonic sounds hints at a broader tension in human/non-human relationship. In the early spring, marshes and flax-dams are filled with frogs, toads and their chorus of “mating calls” and “territorial calls” as auditory signals to attract mates and indicate their readiness for territorial defence (Capranica 9). In ‘Death of a Naturalist’, Heaney refers to sounds of frogs, as a familiar Irish keynote, to evoke the soundscape of his childhood and re-live the local townland, but also to indicate a transitional phase in his relationship with the natural environment. The first stanza of the poem presents a rather pleasant and delightful experience of the poet’s local flax-hole, where all he can hear is the delicate gargling sound of bubbles and flies buzzing around the flax-dams. The acoustics are powerfully recreated through Heaney’s rich onomatopoetic effect of “bubbles gargled” and the notable clusters of sibilant sounds (DN 3): Bubbles gargled delicately, bluebottles Wove a strong gauze of sound around the smell. There were dragonflies, spotted butterflies […]. (DN 3) 117 Heaney recalls his fascination when watching the growth of frogspawns in a jar he often placed on the window-sills at his house and his satisfaction with his biology teacher’s age- appropriate explanation of mating process between “the daddy frog” and “the mammy frog” (DN 3). Yet, on one particular day, Heaney’s personal encounter with the croaking bullfrogs proves to be different to the extent that he believes he had not experienced anything like that before: “a coarse croaking that I had not heard / before” (DN 3). This statement on Heaney’s own part signals the child’s awakening to something new. In one sense, of course, the pre- adolescent Heaney had “heard” the sound of the bullfrogs before but had not noticed it. The internal development − from childhood to adolescence − is also illustrated through the transition, in the poem, from the quiet ambience of “the flax-dams” in the first stanza to the heavy din of bullfrogs in the second: “the air was thick with a bass chorus” (DN 3). 3.2 My Serenades: Biophonic Sounds99 Phrases like “angry frogs”, “a coarse croaking”, “gross-bellied frogs”, “mud grenades” and “obscene threats” (DN 3) show a direct contrast to his initial fascination and sense of wonder and innocence. Here too Heaney manipulates the sounds of the words to serve his purpose. The proliferation of plosive and harsh consonants [p], [d], [k] and [g] indicates feelings of uneasiness. The growing Heaney experiences feelings of disgust and fear at the sight and sound of the frogs and begins to feel unsafe in what he now believes to be the territory of the animal and runs away: […] I ducked through hedges To a coarse croaking that I had not heard Before. The air was thick with a bass chorus. Right down the dam gross bellied frogs were cocked On sods; their loose necks pulsed like sails. Some hopped: The slap and plop were obscene threats. Some sat Poised like mud grenades, their blunt heads farting. I sickened, turned, and ran. (DN 3-4) Heaney’s initial wonder toward animals as a child and his later uneasiness in the acoustic territory of the frogs reflect the complex dynamics of the human-animal relationship. ‘The Early Purges’ shows another episode of this tension, where the young Heaney watches the 118 drowning of kittens by another boy. Such incidents initially arouse mixed feelings of sympathy and fright in him but over time he begins to accept it as a matter of fact as he watches more and more animals dying at the hands of man. This tension is particularly evident when Heaney juxtaposes Dan’s position of superiority and power − when he pitches the kittens into the bucket saying: “the scraggy wee shits” (DN 11) − with the “frail” sounds as the kittens desperately scrape the bucket, as well as when “their tiny din” dies but the noise of the omphalos ceaselessly pumping out water continues. The animals are dead, but life goes on for the man, undisturbed. The disturbing transition from a child’s innocent compassion for animals to an adult’s recognition of the need to rule over and manipulate nature is a common aspect of human- animal relationships in soundscape ecology that Heaney wants to de-familiarise and highlight.104 The grown-up Heaney has become desensitised toward animal cruelty. 104 I am using the term ‘de-familiarise’ as in its general sense: “to cause (a person, the mind) to take a fresh view of something that is familiar” in OED. Defamiliarisation, as a literary technique, introduced by the Russian Formalist Šklovskij's is not my intention here and throughout the thesis. 3.2 My Serenades: Biophonic Sounds99 Now, he turns a deaf ear to their cries of agony and adds insults to ease his conscience: “And now, when shrill pups are prodded to drown, / I just shrug, ‘Bloody pups’” (DN 11). He even sees it as a sensible formula for orderly management of the farm: […] It makes sense: […] It makes sense: […] It makes sense: ‘Prevention of cruelty’ talk cuts ice in town ‘Prevention of cruelty’ talk cuts ice in town Where they consider death unnatural, But on well-run farms pests have to be kept down. (DN 11) The rats, rabbits, crows, hens, kittens and pups which set the vibrant rhythm of the farmland could be seen as merely “pests” and thus unwanted noises that humans feel free to eliminate (DN 11). Still more disturbing is the thrill with which the poet describes a hunting day in The rats, rabbits, crows, hens, kittens and pups which set the vibrant rhythm of the farmland could be seen as merely “pests” and thus unwanted noises that humans feel free to eliminate (DN 11). Still more disturbing is the thrill with which the poet describes a hunting day in ‘Dawn Shoot’. The adult Heaney joins a friend, Donnelly, on a pseudo-military mission, a ‘Dawn Shoot’. The adult Heaney joins a friend, Donnelly, on a pseudo-military mission, a 119 bounty hunting that turns to a pleasure hunting.105 The pair set out at dawn, “rubber-booted, belted, [and] tense as two parachutists”, walking stealthily, crawling, hiding behind the shrubs “with ravenous eyes” (16). Nature detects the intruders and sets off alarming signals: a corncrake’s cry challenges them like “a hoarse sentry”, a snipe darts away “on reconnaissance”, the cock is “sounding reveille”, a mare whines, another snipe flies away, yet Heaney and his companion dismiss them all and compete for the thrill of killing: “‘For Christ sake’, I spat, ‘Take your time, there will be more’” (17). In Wintering Out, the poem ‘Shore Woman’ describes a similarly tense relationship between human beings and sea animals, particularly mackerel and porpoises. The poem is about a wife who accompanies her husband on a fishing journey (see 3.1). 105 The poem does not overtly deal with politics and history. However, Heaney’s use of what Corcoran interprets as an “almost absurd range of military metaphors” (43) suggests that Heaney and his companion sense the violence inherent in the land. 106 See irelandswildlife.com/atlantic-mackerel-scomber-scombrus/. 106 See irelandswildlife.com/atlantic-mackerel-scomber-scombrus/. 3.2 My Serenades: Biophonic Sounds99 While she finds the man’s catching of the mackerel to be brutal and frightening, the woman herself fails to empathise with porpoises and whales: “Under the boards the mackerel slapped to death / Yet still we took them in at every cast, / Stiff flails of cold convulsed with their first breath” (WO 54). Mackerel is a common and widespread fish around Irish coasts often caught in great numbers during the summer.106 In these lines, Heaney shows how the cold-blooded creature is being caught so cold-bloodedly by both the fisherman and his wife. Earlier I discussed how sounds can inform the listener about the type, force and proximity of the sound-producing event (see 1.4.1). The soundscape of the fisher’s boat speaks of the desperation and suffering of the animal in the face of the indifference, brutality and endless greed of human beings. This effect is made more notable in the backdrop of the acoustic properties of the lines – the strict iambic of the monosyllables in line 9 combined with a Hopkinsian overloading of consonants in line 10. Moreover, the use of the plural pronoun (“we”) and the woman’s cry of 120 joy and excitement (“this is so easy that it’s hardly right”) show that, in fact, she shares similar feelings towards catching fish to the man. Porpoises’ practice of breaking the water’s surface and then returning to it, and the splashing and pressure waves they may cause can, according to the ecological approach to listening, be acoustic (see Gaver 9-10). But porpoises are less “showy” than dolphins and can often be difficult to spot or hear in the water. They typically surface very briefly in a sequence just to breathe and then disappear beneath the waves again until the next sequence of breathing. In fact, they may only produce a “sneeze-like” sound without being seen.107 Yet the woman renders a very dramatic acoustic − and visual − description of the experience: Cartwheeling like the flywheels of the tide, Soapy and shining. To have seen a hill Splitting the water could not have numbed me more Than the close irruption of that school, Tight viscous muscle, hooped from tail to snout, Each one revealed complete as it bowled out And under. They will attack a boat. 107 It is a trait that has earned them the old English name of “Puffing Pig” and the Irish name of “Muc Mhara” meaning sea pig. See www.irelandswildlife.com/harbour-porpoise-phocoena-phocoena/. g p g p p p p 8 Other examples of human-animal tension can often be seen in his early poetry, including ‘Turkeys Obse p y p y, g y (DN 24), ‘For the Commander of the ‘Eliza’’ (DN 21-22) and ‘The Outlaw’ (DD 6). 108 Other examples of human-animal tension can often be seen in his early poetry, including ‘Turkeys Observed’ (DN 24), ‘For the Commander of the ‘Eliza’’ (DN 21-22) and ‘The Outlaw’ (DD 6). p g p p p p examples of human-animal tension can often be seen in his early poetry, including ‘Turkeys Observed’ ‘For the Commander of the ‘Eliza’’ (DN 21-22) and ‘The Outlaw’ (DD 6). 109 Around 10,000 years ago, Ireland became isolated from mainland Europe limiting its indigenous animal population. See animals.mom.me/indigenous-animals-ireland-5197.html. Ireland’s extinct indigenous mammals are the giant deer (the great Irish elk), the grey wolf and the great auk. In both Irish literature and environmental discussions, the extinction of the native Irish animals is seen as an emblem of the concurrent brutalisation of Irish people. See animals.mom.me/indigenous-animals-ireland-5197.html. 3.2 My Serenades: Biophonic Sounds99 (WO 54-55) She describes the porpoise’s sequential surfacing as an act of “irruption” or “cartwheeling like the flywheels of the tide” and interprets it as an attempt to attack the boat. She is completely repulsed by their vigorous bodies, which she associates with masculinity (“Tight viscous muscle”) to rationalise her fear and dislike (“Sick at their huge pleasures in the water”) (WO 54-55). The poem is, therefore, an example of misunderstanding, misinterpreting and disregarding the world of non-human animals.108 Just as geography and climate determine vernacular keynotes, the sounds of native animals provide soundmarks that establish a unique sense of place often only experienced by the locals. Therefore, the loss of unique biophonic sounds can often impose a profound impoverishment on the sense of place and identity (see 1.4.2). Human beings transform 121 landscapes mainly through “land-use” and “land-cover change”. These human “modifications” of the landscape influence the distribution and abundance of biodiversity and at times lead to the complete extinction of some species (Pijanowski et. al. 204-205). Heaney’s reference to the absence of the Great Irish elk and the grey Irish wolf, which I discuss next, not only highlights the human/non-human tension at the ecological level but also explains their symbolic weight in Irish literature, culture and politics.109 In these instances, Heaney has not personally heard the elk or the wolf, but he acknowledges that they have nevertheless been members of the past soundscapes. What he highlights in these instances are the changes in the soundscape throughout time – symbolically or literally – because of the absence or disappearance of some biophonic sound sources. Heaney inscribes their presence in his poetry and in the memory of his readers (see also 4.4). In ‘Bogland’, Heaney brings up the striking image of the unearthed skeleton of the long-extinct “Great Irish Elk” (DD 43), highlighting its absence from the contemporary Irish soundscape. The giant deer became extinct around 10,000 years ago, even before the arrival of the first Mesolithic people in Ireland. p p g 110 Although the Irish Elk “is neither exclusively Irish nor an elk” − it is a Pleistocene deer also widely spread through Europe, northern Asia and Africa − it has a symbolic significance in the Irish culture (Gould 191). However, deer and elk have very complex vocal mechanisms producing high-pitched calls, which they engage in elaborate manoeuvers for courtship and territorial purposes (Roland and Riede 307). 111 See www.thelearnedpig.org/wild-alphabet-wolf-irish-poetry/4386. 111 See www.thelearnedpig.org/wild-alphabet-wolf-irish-poetry/4386. 3.2 My Serenades: Biophonic Sounds99 Nonetheless, it has become a key icon in Irish history and mythology (Gould 191).110 Heaney’s recognition of the Irish Elk as an archetypal element of the ancient Irish history can be perceived as an attempt to emphasise the need to return to a mythologised Irish past, before the disappearance of the Irish wood and wildlife, in order to summon and re-establish the sense of national identity in the present day of Ireland. In ‘Midnight’, from Wintering Out, similarly, Heaney links the disappearance of the Irish language and identity to the extinction of the grey wolf after the wars of the seventeenth 122 century and the colonial settlement, but also laments the extinction of the Irish wolf as the consequence of colonisation, agricultural expansion, deforestation and hunting. Before 1600 and the arrival of the first English settlers a number of wolves were living in the woodland of Ireland. Although often associated with evil and threat, the wolf was regarded as a keynote component of the Irish landscape, leading to Ireland being called “Wolf-Land” by newly arrived English settlers (Hickey 194). The systematic eradication of the wolf by the colonisers was in many ways parallel to the restrictive measures imposed on the Gaelic language and culture. In Irish poetry, the extinction of the Irish wolf has been transformed into a recurring metaphor for the loss of a linguistic and national identity, as well as a symbol for Ireland.111 In “Midnight”, Heaney writes: The wolf has died out In some scraggy waste of Kildare. The wolfhound was crossed With inferior strains, Forests coopered to wine casks. (WO 35) According to Hickey, there are two primary causes for the decline and extermination of the wolf in Ireland: habitat change and professional hunting. Between 1600 and 1700, the forested area of Ireland, the primary habitat of the wolves, declined from approximately 12.5 percent to 2 percent. This deforestation came about as a result of a number of factors, including the settlement of the English colonisers, who wanted to replicate an English landscape in Ireland, the plantations of Ulster in order to meet the growing demands for arable land, and the deforestation of the Irish woodland for industrial and construction purposes in Britain (Hickey 193-195). Heaney refers to this factor in the last line of the 123 stanza: “Forests coopered to wine casks” (WO 35). 112 In 1610, Lord Blennerhasset, Chief Baron of the Irish Exchequer, described the woodkernes as “human wolves” and “recommended periodic manhunts to track down the human wolves to their lairs”. The Irish were thereby bestialised and political resistance was coded as “wolf-like” (Neeson 140-141). f ( ), p y g , g , even transforming into them. See celt.ucc.ie//published/E500000-001/. In As You Like It (1562), Shakespeare too writes disparagingly of the association, describing Rosalind’s lovers’ complaints “like the howling of Irish wolves” (145). In a testimony, a military captain and his forty soldiers, who witnessed the bodies of an Irish garrison slaughtered at Cashel in 1647, asserted the dead “had tails near a quarter of a yard long” (Wiseman 71). 113 Cunningham’s 2017 research indicates the positive associations between the Irish people and wolves in Irish mythology and hagiography. See www.thelearnedpig.org/wild-alphabet-wolf-irish-poetry/4386. However, the English equated the Irish with wolves as a means of justifying their brutalisation. In A View of the Present State of Ireland (1596), Spenser viciously states the Irish were in the habit of eating wolves, befriending them, and y p ( ) 113 Cunningham’s 2017 research indicates the positive associations between the Irish people and wolves in Irish mythology and hagiography. See www.thelearnedpig.org/wild-alphabet-wolf-irish-poetry/4386. However, the English equated the Irish with wolves as a means of justifying their brutalisation. In A View of the Present State of Ireland (1596), Spenser viciously states the Irish were in the habit of eating wolves, befriending them, and even transforming into them. See celt.ucc.ie//published/E500000-001/. In As You Like It (1562), Shakespeare too writes disparagingly of the association, describing Rosalind’s lovers’ complaints “like the howling of Irish English equated the Irish with wolves as a means of justifying their brutalisation. In A View of the Present State of Ireland (1596), Spenser viciously states the Irish were in the habit of eating wolves, befriending them, and even transforming into them. See celt.ucc.ie//published/E500000-001/. In As You Like It (1562), Shakespeare too writes disparagingly of the association, describing Rosalind’s lovers’ complaints “like the howling of Irish wolves” (145). In a testimony, a military captain and his forty soldiers, who witnessed the bodies of an Irish garrison slaughtered at Cashel in 1647, asserted the dead “had tails near a quarter of a yard long” (Wiseman 71). ythology and hagiography. See www.thelearnedpig.org/wild-alphabet-wolf-irish-poetry/4386. Howeve nglish equated the Irish with wolves as a means of justifying their brutalisation. In A View of the Present Ireland (1596), Spenser viciously states the Irish were in the habit of eating wolves, befriending them 3.2 My Serenades: Biophonic Sounds99 The second factor that led to the extinction of the Irish wolf was the legislation and bounties, starting as early as 1584, to encourage the professional hunting of wolves that were now seen as a threat to the agricultural economy (Hickey 193-195). As Heaney too mentions the last wolf was killed by a professional hunter equipped with two wolf dogs: “[t]ill a Quaker buck and his dogs / Killed the last one” (WO 35). By 1786, wolves were completely exterminated from the Irish landscape leading to the loss of one of Ireland's most formidable post-glacial species and the only major carnivore to have survived into the historic period (Hickey 197). Today, Heaney feels the absence of the Irish wolf in its native habitat: “[t]he old dens are soaking”, “pads are lost”, “[n]othing is panting, lolling, / Vapouring” (WO 35). Politically and symbolically, the rampant deforestation of the Irish land and eradication of the wolf were also seen as ways of leashing the “wolf-like” Irish rebels − known by the English as “woodkerne” − and depriving them of a place of refuge (Neeson 140-141).112 Given that the wolf had disappeared in both England and Scotland in the mid 18th century, the new settlers associated the wild animal with the language and the behaviour of the natives, which they were determined to suppress. The fact that the Irish wolf was a keynote element of the Irish soundscape and had a great impact on the Irish culture is indisputable.113 But the association between Irish identity and the Irish wolf can be viewed differently from the perspective of sound studies. In the haunting howling of wolves we encounter “a vocal ritual” (Schafer 39) which defines the territorial claim of the pack to their 124 acoustic space. Perhaps it was this territorial call of the grey wolf, and not the assumptions about its wildness, that rang as a threat to the ears of English colonisers. Heaney ends his poem saying the ruthless suppression of his language, culture and territory has left him speechless: “The tongue’s / Leashed in my throat” (WO 35). Heaney also makes an effective use of the vocal and mechanical sounds of native birds to communicate themes regarding the environment, society and even personal memories. p , g j p g 115 After observing and analyzing his recordings of soundscapes in multiple habitat types, Krause postulated that the competition for auditory space prompts organisms to adjust their signal calls so as to exploit vacant niches in the auditory spectrum and minimise spectral or temporal overlaps (qtd in Pijanowski et.al. 14-18). 114 Almost 500 varieties of bird species can be found in Ireland. Swans, ducks, waders and curlew are the vulnerable species to which Heaney refers in his poetry. For environmental approaches of Irish writers see Potts. p , , ulnerable species to which Heaney refers in his poetry. For environmental approaches of Irish writers see 114 Almost 500 varieties of bird species can be found in Ireland. Swans, ducks, waders and curlew Almost 500 varieties of bird species can be found in Ireland. Swans, ducks, waders and curlew are the vulnerable species to which Heaney refers in his poetry. For environmental approaches of Irish writers see Potts. 116 See www.youtube.com/watch?v=bAuucSly7t8 (3:50-3:59). y y ( ) nd, more extensively, here, where the poet reads several translations from the Irish of this particular trad , y, , p p www.youtube.com/watch?v=emuYwWT7s4A (1:06:02-1:06:04). , y, , p www.youtube.com/watch?v=emuYwWT7s4A (1:06:02-1:06:04). 3.2 My Serenades: Biophonic Sounds99 The diversity, abundance and richness of bird species has been the focus of scholars including bio-acousticians, ornithologists and soundscape scholars.114 Like birds themselves, bird vocalisations are of different types, some are penetratingly loud and intense − e.g. the scrubbird and the corncrake, some have a bell-like sound − e.g. the sparrow, some others are two-tone − e.g. the male cuckoo, and some are superb mimics − e.g. the lyrebird. Birds can often be distinguished by the sounds of their flights too. The slow flapping of an eagle is different from the startled slapping of a flock of geese on the water or the shaking of the sparrow’s wings against the air. We may not see the birds but hear the fast whirl of their wings or tails and distinguish the type and direction of their flight (Schafer 31-33). The song of birds can register as a vernacular keynote, given the uniqueness of some species, and can even dominate the soundscape because of their abundance. Nonetheless, as Krause’s soundscape archive shows, birds’ vocalisations have in fact little spectral or temporal overlap (14-18).115 As Schafer notes, birdsongs are rich and varied without “imperialistically dominating” the soundscape leading to listeners identifying them as one of the most “pleasant” elements of the natural environment (29-31). In literature and mythology of the East and the West, birds have often been attached to the human imagination as the 125 archetypes of gentleness, delightfulness, spirituality and peace. The mythological bird Sīmurğ − also known as Humā or the Persian phoenix − is the iconic figure of poetry and painting within the realm of the Persian and Byzantine empires and appears along with nightingale, cuckoo and other birds in the Sufi poetry of Hafez, Saadi and Ferideddin Attar. Birds, particularly the nightingale, appear in the epic poetry of Homer and the bucolic poetry of Theocritus, in the poetry of Virgil and Ovid, and in the writings of Chaucer, Eliot, Hardy, Hopkins, Hemingway and Wordsworth, some of whom are among Heaney’s inspirations. In ‘The God in the Tree’ Heaney quotes his translation of the 9th century Irish poem ‘The Blackbird of Belfast Lough’ to refer to the “Little jabs of delight in the elemental” communicated through the presence of “woods”, “water” and “birdsong” in the ancient Irish nature poems (PO 181): The small bird let a chirp from its beak: I heard woodnotes, whin- gold, sudden. The Lagan blackbird! (PO 181) The influence of early Irish nature poetry and the presence of the archetypes of idyllic nature continue to echo throughout Heaney’s work and appear as a focus of the scholarship on Heaney (see 2.1 and 2.2). In his 1989 Desert Island Disc interview with Sue Lawley, Heaney confirmed this while expressing his feelings regarding the idyllic countryside of Ireland and the early Irish nature poetry: “I’m in the tradition of the hermit poets of the early Celtic nature tradition”.116 As I argue, while appreciating their overtly bucolic associations, Heaney’s poetry goes beyond the archetypal representation of birds and birdsong to deal with the underlying personal, social and environmental associations. 126 One of the birds that features in Heaney’s poetry frequently is the blackbird. The blackbird is a keynote feature of both the urban and agricultural Irish landscape and fondly known for its rich, warbling song and restless hopping flights (Jones).117 In a 1978 interview, Heaney explains that he had to leave the North in order to gain a greater perspective on his home place and on his own status as a poet during the Troubles.118 The soundscape of Glanmore, which features the song of blackbird as one of its main keynotes, evokes the memories of Mossbawn and becomes a place of contemplation and inspiration for much of his poetry. ‘The Blackbird of Glanmore’ opens as, on his way to visit the cottage, the aging poet reflects upon his views on the transitory nature of life, the memory of his father and younger brother Christopher, all inspired by a simple encounter with a blackbird which immediately calls to mind the soundscape of his childhood. Keynotes are often heard but not carefully listened to (see 1.4.2). Yet Heaney is not one to overlook a keynote. He decides to “park, pause, [and] take heed. / Breathe. Just breathe and sit” to observe and listen to the bird (DC 75). 118 In a 1979 interview with Randall, Heaney comments on his reasons for moving to Wicklow, saying: I left […] not really out of any rejection of Belfast but because […] well, I had written three books, had published two, and one was due to come out. I had the name for being a poet but I was also discovering myself being interviewed as, more or less, a spokesman for the Catholic minority during this early stage of the Troubles. I found the whole question of what was the status of art within my own life and the question of what is an artist to do in a political situation very urgent matters […] I wanted to step out of the rhythms I had established; I wanted to be alone with myself. See www.pshares.org/authors/james-randall. 119 Heaney here is also referring to the superstition surrounding the blackbird as an evil omen, which according to a neighbour had been sitting on their roof prior to Christopher’s death. In a 2006 interview with Sam Leith, in the Daily Telegraph, he reveals that “an old woman – a weird sister figure, she lived down the fields. […] She read the world in terms of signs, omens”. See www.theage.com.au/entertainment/books/back-on-home-turf- 117 See www.irelandswildlife.com/blackbird-turdus-merula/. 118 y ; See www.pshares.org/authors/james-randall. 117 See www.irelandswildlife.com/blackbird-turdus-merula/. 118 , p y 120 References to the snipe recur in Heaney’s work. For instance, in ‘A Kite for Michael and Christopher’, it is a metaphor for the human soul: “the human soul / is about the weight of a snipe” (SI 44). In ‘At the Water’s Edge’, he hears the sound of the snipe among the monastic recitals: “On Devenish I heard a snipe” (FW 6). 121 S bi d t hi l d i /bi d / k l k/ y 122 In the poem, Heaney also refers to the welcoming song of the cuckoo: “[e]arly summer, cuckoo, cuckoos, / Welcome, summer is what he sings” (EL 37), alluding to the 13th century English poem about the approach of summer, ‘The Cuckoo Song’; In ‘A Kite for Michael and Christopher’, Heaney likens the kite to a “small black lark” flying high up in the sky and uses both as a symbol of the human soul (SI 44). 20060527-ge2e94.html. Nevertheless, Heaney himself recalls the bird affectionately: “It’s you, blackbird, I love” (DC 75). In ‘The Loaning’ Heaney makes a passing reference to its “startled volubility” (SI 52), and in ‘St. Kevin and the Blackbird’, he associates the blackbird with spirituality. nature poems (PO 181): The sudden flight of the blackbird startled by the clunk of the car lock (“the automatic lock / clunks shut”) and its restless flitting from one hedge to another (“hedge- hop”) is thought provoking for Heaney, encouraging him to reflect on the passing of life, to commute between the “house of life” in Glanmore where he lived and raised his children to the “house of death” where the memories of his father and brother lie, and thus, to reflect on his own longevity from “a bird’s eye view” (DC 76).119 See www.pshares.org/authors/james randall. 119 Heaney here is also referring to the superstition surrounding the blackbird as an evil omen, which according to a neighbour had been sitting on their roof prior to Christopher’s death. In a 2006 interview with Sam Leith, in the Daily Telegraph, he reveals that “an old woman – a weird sister figure, she lived down the fields. […] She read the world in terms of signs, omens”. See www.theage.com.au/entertainment/books/back-on-home-turf- 127 Along with blackbirds, the lark, corncrakes, and snipe are known as familiar elements of the Irish landscape and major contributors to local biodiversity. The snipe is a mostly ground bird and prefers bogs, marshy ground and fields filled with rushes and grass as its habitat. It is a mostly nocturnal or crepuscular unobtrusive wading bird that depends on silence and camouflage to avoid detection. When flushed, it utters a unique alarming cry and erupts suddenly from the ground (Campbell and Lack 202). Heaney describes this characteristic behaviour of the snipe in ‘Dawn Shoot’, when he writes: “a snipe rocketed away on reconnaissance” or when, towards the end of the shooting day, “another snipe catapulted into the light” (DN 16-17).120 Like snipes, larks are one of the most common residents of Ireland, but unlike them, they are known as vocal birds whose distinctive continuous stream of warbling songs, mostly heard during spring and summer, can last up to half an hour and while the bird is flying 50 to 100 metres overhead.121 In ‘Glanmore Eclogue’, the “poet” refers to this fact and writes that the song of the lark makes the idyllic soundscape of Wicklow even more “perfect” during summer days (EL 37):122 A little nippy chirpy fellow Hits the highest note there is; The lark sings out his clear tidings. Summer, shimmer, perfect days. 20060527-ge2e94.html. Nevertheless, Heaney himself recalls the bird affectionately: “It’s you, blackbird, I love” (DC 75). In ‘The Loaning’ Heaney makes a passing reference to its “startled volubility” (SI 52), and in ‘St. Kevin and the Blackbird’, he associates the blackbird with spirituality. 120 References to the snipe recur in Heaney’s work. For instance, in ‘A Kite for Michael and Christopher’, it is a metaphor for the human soul: “the human soul / is about the weight of a snipe” (SI 44). In ‘At the Water’s Edge’, he hears the sound of the snipe among the monastic recitals: “On Devenish I heard a snipe” (FW 6). 121 See birdwatchireland.ie/birds/skylark/. 122 In the poem, Heaney also refers to the welcoming song of the cuckoo: “[e]arly summer, cuckoo, cuckoos, / Welcome summer is what he sings” (EL 37) alluding to the 13th century English poem about the approach of nature poems (PO 181): (EL 37) A little nippy chirpy fellow Hits the highest note there is; The lark sings out his clear tidings. Summer, shimmer, perfect days. (EL 37) Corncrakes are relatively shy and hard to see but have a very distinctive rasping call which they use repeatedly to indicate their presence (Dempsey and O’Clery 82). In ‘Dawn Corncrakes are relatively shy and hard to see but have a very distinctive rasping call which they use repeatedly to indicate their presence (Dempsey and O’Clery 82) In ‘Dawn 128 Shoot’, discussed earlier in this section, the unexpected cry of the corncrake signals the presence of an intruder in their habitat: “[…] A corncrake challenged / Unexpectedly like a hoarse sentry” (DN 16). In a later poem, ‘Castling and Gathering, the call of the corncrake reaffirms its status as a prominent keynote of the soundscape inscribed in the memory and mindscape of the poet. The poem captures the sharp contrast between the views of two fishermen − Ted Hughes, a poet to whom the poem is dedicated, and Barrie Cook, an artist − through the sounds they are producing on either side of the river: “these sounds took sides” (ST 13). On one side, the fishing fly goes “whispering through the air” and, on the other, the repetitive, intense and grating cry of the corncrake becomes a metaphor for “the sharp ratcheting” sound of the fisher’s reel that goes “on and on” disturbing the “stillness” of the riverbank (ST 13). Heaney states that, although he liked fishing with a fly, he was never patient enough to cast one, but also that he could never forget the feelings − and sounds − related to fishing: “I wanted to fish with a fly and just never stayed at it long enough to learn to cast, so that effort fizzled out […]. But inside my sixty-eight-year-old arm there’s a totally enlivened twelve-year-old one, feeling the bite” (SS 95). The poem gives voice to “the music of fishing, its swishing and whispering and sharp ratcheting” (Parker 220), and in doing so it indicates the grating call of the corncrake as a central keynote of the poet’s mind and memory. In addition to their personal and idyllic associations, birds appear in Heaney’s poetry to illuminate his social and environmental awareness. nature poems (PO 181): In poem xlii of ‘Squarings’, Heaney commemorates the farmers of his father’s generation, the “apparitions” who are still “active”, “territorial”, “sure of their ground” and “interested” (ST 102). Yet he laments over their lack of awareness that their intensive methods of agriculture have led to the gradual disappearance of their natural landscape and its biophony. He notes heather, kesh, turf stacks and grasshoppers reappear each summer, but they are becoming increasingly rare: “Heather and 129 kesh and turf stacks reappear / Summer by summer still, grasshoppers and all, / The same yet rarer […]” (ST 102). The transformation of farming methods and presence of noise from agricultural machinery can also heavily impact the population of the larks, who favour uplands and areas of farmland as their breeding places. Heaney also laments the general local assumption that the spread of larks is still as “unstoppable” as the sudden spread of “sunshine” on hills, and notes: “How long the lark has stopped outside these fields / And only seems unstoppable to them / Caught like a far hill in a freak of sunshine” (ST 102). Although the snipe is a well-established breeding species in Ireland, its population has undergone a decline due to seasonal shooting and the loss of its habitat through afforestation of moorlands and bogs.123 In ‘The Backward Look’, Heaney attempts, on the one hand, to celebrate the native bird as a keynote of the Irish soundscape and culture and, on the other hand, to bring to light the threat of its gradual decline. Earlier I discussed how the call of native birds can provide a vernacular keynote as characteristic as the language of the people who live there (see 1.4.2). In ‘The Backward Look’, Heaney links Gaelic language and the snipe, which as Foster has also noted, is a “particularly Irish bird” (38), but also indicates the call of the snipe we hear in the poem is not what could be interpreted as an aggressive “territorial” call, a cheerful “pleasure” call or an intense “mating” call. Heaney implies that the snipe’s cry is what Schafer identifies as a feeble “distress” call (33-40) and likens it to the status of the Irish language in its original land. Today, Northern Ireland has no official language. English is the de facto first official language, and Irish and Ulster-Scots are recognised regional languages. nature poems (PO 181): Like the native language, as Heaney states, the snipe gives out a mournful cry as distancing its natural habitat and original nesting ground into less suitable realms: 123 See Barry’s Shooting: Moor and Marsh (2013) for details about snipe shooting in Ireland. The population of the local snipe in Ireland has undergone a decline in recent decades. See snipeconservationalliance.org/snipe-in- ireland/. A snipe’s bleat is fleeing A snipe’s bleat is fleeing 123 See Barry’s Shooting: Moor and Marsh (2013) for details about snipe shooting in Ireland. The population of the local snipe in Ireland has undergone a decline in recent decades. See snipeconservationalliance.org/snipe-in- ireland/. 130 into variants […]. (WO 19) In ‘The Backward Look’, even the flight of the snipe speaks of a mournful situation and a forthcoming threat. Heaney compares the decline of the Irish language in its native land to an ominous sound, a change in the familiar flapping and throbbing sound of the snipe’s wing as it struggles to maintain its balance in the air: A stagger in air as if a language failed, a sleight of wing. (WO 19) A stagger in air as if a language failed, a sleight of wing. (WO 19) Garrard draws attention to the intertwined relationship between ecology and language in ‘The Backward Look’, but rather than drawing a metaphorical comparison, he suggests that the fate of the bird itself has a direct impact on the language of its native place (193). Lidström also comments that Heaney’s use of the term “sleight” to describe the snipe gives it a kind of “personhood” that further reinforces the relationship between nature and culture (Lidström 88). In fact, the term “dialect” − as a form of a language particular to a specific region or social group − directly links the snipe to its acoustics and the language and culture of its native place. Heaney compares the unfamiliar sound of the bird’s flight to the Norse “transliterations whirr[ing] / on the nature reserves” (WO 19) and then offers the more colourful and Irish folk alternative in italics: “little goat of the air, / of the evening, // little goat of the frost” (WO 19), which becomes a faded memory if the local snipe itself becomes extinct. The snipe is widely recognised for the beating sound it can produce in the air through vibrating its tail-feathers as it dives. Thus, the bird being audible is only a sign of descent (O’Brien Johnson 137). In the poem, the bird’s tail-feathers are “drumming”, not to celebrate, but to lament. The “drumming elegies” of the snipe reinforce the theme of the extinction of the bird and the disappearance of the language while travelling “in the slipstream // of [the less common] wild goose / and yellow bittern” (WO 19). A snipe’s bleat is fleeing The possibility that the snipe 131 becomes a forgotten entry is intensified by the vocabulary of decline (“stagger”, “failed”, “bleat”, “fleeing”), closure (“gleaning”, “leavings” and “archive”) and threat (“sniper”). The term sniper was originally coined by British soldiers in India in the early nineteenth century and refers to a skilled gunman who could shoot a snipe (Lidström 89). Therefore, while associating with the contemporary Troubles and the legacy of colonisation, the term “sniper” too highlights the impact of seasonal hunting on the population of the local snipe. Heaney is worried that, like Irish language and culture, the local snipe becomes a forgotten entry in the “archive” of history (WO 20). The bird’s downward flight and disappearance is made almost inevitable by Heaney’s use of short-lined stanzas, enjambment, and the assonance woven in the final lines.: disappearing among gleanings and leavings in the combs of a fieldworker’s archive. (WO 20) disappearing among gleanings and leavings in the combs of a fieldworker’s archive. (WO 20) disappearing among gleanings and leavings in the combs of a fieldworker’s archive. (WO 20) In ‘Glanmore Sonnets’ the presence of the corncrake exceeds its idyllic associations and contributes to the thematic development of the poem. ‘Glanmore Sonnets’ are among the first poems Heaney wrote after moving to Wicklow, a countryside environment that in Preoccupations he defines “as being pastoral rather than rural, trying to impose notions of a beautified landscape on the word, in order to keep ‘rural’ for the unselfconscious face of raggle-taggle farmland” (PO 173). By comparison with the “turmoil of activities” on the farmland where human beings and animals cohabit, the pasture feels “generally quieter” characterised by pleasing geophonic and biophonic sounds (Schafer 44-48). Heaney explains that it was “the too literary nature of the reality of Wicklow” in an evening in the month of May that inspired the first sonnet featuring the cuckoo and the corncrake (SS 162-163). The sonnet opens: “This evening the cuckoo and the corncrake / (So much, too much) consorted at twilight. / It was all crepuscular and iambic” (FW 30). Ironically, these lines are not iambic pentameter and as Gifford notes, “the call of the cuckoo is a trochee, and that of the 132 corncrake is a spondee” (106). However, Heaney uses the term “iambic” to suggest the rhythmic call of the birds and the idyllic soundscape of Wicklow. 124 Heaney comments on Sonnet III at a poetry reading session in New York, in September 2011. See www.youtube.com/watch?v=tj8dATdoES0 (12:17-14:52). In Stepping Stones, he explains: “I remember wondering, what the hell is all this iambic pentameter doing in my life? There’s even a line mocking the too literary nature of the reality of Wicklow that evening” (SS 162-163). wo de g, w at t e e s a t s a b c pe ta ete do g y e? e e s eve a e oc literary nature of the reality of Wicklow that evening” (SS 162-163). A snipe’s bleat is fleeing In a 2011 poetry reading session, he describes the organic continuum between the rhythmic singing of the birds and the sonnet: “one evening the iambic pentameter flew across and I let it happen”.124 Yet the birdsongs Heaney refers to in ‘Sonnet III’ are not “beautified” in the escapist sense (Gifford 104-105). The consorting call of the cuckoo and nocturnal corncrake − together with the sight of the deer and the rabbit, and the sound of “a rustling and twig-combing breeze” (FW 30) to which Heaney refers in the final line of the poem − are the natural “cadences” (FW 30) that recall the now-vanished soundscape of the pre-mechanised farmlands of Ulster. In addition to illustrating the poet’s ecological awareness, the poem also can be read in the context of postcolonial literature for recalling the legacy of English colonisation (Allison 79). In 1988 Cheltenham Literature Festival, Heaney explained his mixed feelings towards these lines, which come across to him as “too comfortable […] English even” (qtd in Gifford 106). In an interview with Frank Kinahan he explained: “Back then I thought that music, the melodious grace of the English iambic line, was some kind of affront, that it needed to be wrecked; and while I loved the poem, I felt at the time that its sweetness disabled it somehow” (412). The reference to colonialism is made earlier in ‘Sonnet II’, where Heaney speaks of landing in “the hedge-school of Glanmore” (FW 29). Hedge-schools were the only way the native Irish, specifically those who were not Anglican, could have education during the period of the Penal Laws of the 1830s and even earlier. These schools were illegally operated by and for Irish Catholic peasants in barns and cowsheds (Corcoran 147). In ‘Sonnet III’, the phrase “too much” (FW 30), in the second line, establishes a tension which prevents the temptation to pastoralise the call of the cuckoo and the corncrake at 133 twilight. The hoarse cry of the corncrake which registers as a keynote of the rural Irish soundscape becomes, as Allison suggests, an apt symbol for the guttural sounds of the Irish language (74). 125 In OED, ‘Ack-ack’ or ‘a. a.’ is also a British colloquial for “an anti-aircraft gun”. In these poems, however, the poet is remembering the sound fondly and as a keynote element of the Irish soundscape. A snipe’s bleat is fleeing Heaney’s reference to “the ack-ack” (WO 50) of the corncrake in ‘Serenades’ can also be read in the same colonial/ecological context.125 In the poem, he encourages his readers to join him and listen to his “serenades”, to the poetry which as much celebrates the scolding chattering of the “sedge-warbler”, the throaty kraa of “a crow”, the hoarse call of the “corncrake” and the barely audible “wheeze of bats” as the melodic song of a “nightingale” (WO 50). In his blend of discordant word sounds too Heaney breaks the conventional idea of a serenade: My serenades have been The broken voice of a crow In a draught or a dream, The wheeze of bats Or the ack-ack Of the tramp corncrake […]. (WO 50) My serenades have been The broken voice of a crow In a draught or a dream, The wheeze of bats Or the ack-ack Of the tramp corncrake […]. (WO 50) Heaney likens his own voice and poetry to the grating call of the corncrake contrasting it with the melodic song of the nightingale that occupies a special place in English literature for the mellifluous quality of its song. To make his point clearer, he goes as far as to instead suggest the sedge-warbler, a very common bird with a high-pitched rattling call, as the “Irish nightingale”: The Irish nightingale Is a sedge-warbler, A little bird with a big voice Kicking up a racket all night. (WO 50) A little bird with a big voice He demonstrates his ecological concerns when stating that the species’ survival is threatened by the “combines and chemicals” (WO 50). Allison also argues that “the crake might also be considered as a symbol of Ulster’s Catholics, marginalised and intimidated by Unionist 125 In OED, ‘Ack-ack’ or ‘a. a.’ is also a British colloquial for “an anti-aircraft gun”. In these poems, however, 134 hegemony” (77). This statement too can be better understood in relation to the “imperialistic” (Schafer 3) noises of modern agricultural machinery and techniques in the habitat of the corncrake. In Heaney’s poetry, the songs of birds are not just a metaphor for happiness and an idyllic world, or a symbol of cultural identity and childhood memory. Vehicle and tenor are so intertwined with each other that an exclusive prioritisation of the latter would be an obvious example of “eco-blindness” (Garrard 193). A snipe’s bleat is fleeing The lark and the blackbird sing out to make the soundscape of the poems we read “perfect” (EL 37) but also to alert us to the fact that their absence will be felt as a deep impoverishment in the soundscape of the world we live in. Heaney’s poetry makes a close association between the vocalisation of the native birds and the local language encouraging his readers to appreciate them as equally significant elements of cultural identity that need to be preserved and protected. Heaney writes about non-human animals as a major component of the world soundscape to tap into the consciousness of his readers and awake their compassion for the living beings with whom they share the planet. Yet he also writes about humans, their sounds, noises and voices which I embark on in the following section. 126 The title of this section is inspired by Heaney’s description of Barney Devlin’s hammer strikes (SS 9 3.3 The Epic of the Noise: Anthrophonic Sounds126 In this section, I identify and interpret the references to a selection of man-made sounds. This analysis begins with anthrophonic sounds of farm life and early crafts evoked by the poet and, then, attends to sounds of various means of transport. I close this section with the symbolism of the sounds of bells and time. Throughout his poetry, Heaney gives us a precise acoustic picture of Irish farm life. Earlier in this chapter, I wrote about the biophonic sounds of farm life − e.g. the neighing and snorting of horses, the munching and moaning of cows, the grunting of pigs, the constant 126 The title of this section is inspired by Heaney’s description of Barney Devlin’s hammer strikes (SS 91). 135 cackles of hens and the scurry of rats on the ceiling. The keynotes of the rural Irish life that Heaney recalls and awakens in his work include a diverse variety of anthrophonic sounds as well. I have noted that keynotes are the familiar background sounds of any soundscape, and that although the material modulated around keynotes often obscures their centrality, keynotes remain the fundamental sounds that influence and reflect the mood and lifestyle of the people living within their earshot (See 1.4.2). Heaney captures the sounds and rhythms of early trades and manual farming practices, sounds that were “sealed down” in his memory and re-appear in his poetry to allow the poet to reach into the world of his father and ancestors. In his 1980 interview with John Edwards, Heaney remarks: When I began to write for some reason or other, it was that first deposit of experience that was almost hermetically sealed down came up. It was like reaching into another world even though it was a world that I had experienced 15 or 20 years ago.127 The soundscape of farm life has changed over time. 127 See www.youtube.com/watch?v=3yt4m2Z4Pmw (1:53-2:14). y y ( ) 129 In ‘The Toome Road’, he presents the image of “[t]he invisible, untoppled omphalos” to give a glimpse of hope (FW 7). In ‘A Drink of Water’, he uses the “whooping cough” of the pump and “clatter” of the empty bucket as the acoustic signs that announce the presence of an elderly woman in the yard every morning (FW 8). 128 See www.youtube.com/watch?v=3yt4m2Z4Pmw (0:50-1:38). 3.3 The Epic of the Noise: Anthrophonic Sounds126 The reassuring stamping of the farmer’s boots and the horse’s hooves during ploughing, the earthy grating of the spade, the rhythmic swish and slash of the mowing scythes, the squeak of old rusty hinges and the hand- operated water pump, as well as the acoustics of such crafts as blacksmithing, turf-cutting, water divining, dealing cattle, retting flax in a flax dam, thatching, churning, and even purging unwanted farm animals – all of which Heaney evokes in his poetry – are the vernaculars of the old ways of agriculture and early crafts that have been replaced by the whirring of combines, threshers, mowers, tractors and the din of factories. In this interview, Heaney continues to comment on the rhythm of farm life during the forties comparing it with the more mechanised pattern during the fifties and afterwards: Well it’s what they called a small farm […]. This was actually about fifty acres. I say ‘was’, the land is still the same, but things have changed on the land, I mean, when I was growing up in the forties, you could still see people ploughing with horses, you could still see people mowing with scythes, you could still see people using forks to build hay and the rhythm of life was really a medieval rhythm almost, you know. When the fifties, with mechanisation and so on, that all changed. And now, it’s the 136 same land, the same ground, but there are milking parlours and tractors and mechanisations and the rhythms have changed considerably.128 Listening to such background sounds provides us with the opportunity to measure social and historical changes in the soundscape of the poet’s world and reflect on their social and cultural associations. One of the most familiar anthrophonic sounds, for Heaney and Heaney scholars, which I have frequently referred to in the previous sections, is the familiar sound of a hand-operated water pump referred to as omphalos, positioned outside the back door of Heaney’s house. For Heaney the sound of the omphalos is what Schafer calls “a soundmark” and “a centripetal sound” (10; 56). Heaney describes the sound of the omphalos as a familiar h h i k h ifi d h h h f lif i h b k d Women came and went, came rattling between empty enamel buckets, went evenly away, weighed down by silent water. y y ( ) he Toome Road’, he presents the image of “[t]he invisible, untoppled omphalos” to give a glimpse of 130 The position of ‘Digging’ not only on the opening page of his first volume, but also his Selected Poems, New Selected Poems and Opened Ground indicates the significance of the poem for Heaney (Murphy 8-28). Heaney affirms the essential role of this poem in ‘Mossbawn’: “I now believe that the ‘Digging’ poem had for me the force of an initiation: […] having experienced the excitement and release of it once, I was doomed to look for it again and again” (PO 42-43). He explains: 3.3 The Epic of the Noise: Anthrophonic Sounds126 The horses came home to it in those first lengthening evenings of spring, and in a single draught emptied one bucket and then another as the man pumped and pumped, the plunger slugging up and down, omphalos, omphalos, omphalos. (PO 17) The central role of the water pump is so ingrained in the mind and life of the people that even the manoeuvre of the American troops in the neighbouring fields would not disturb it. Heaney captures this in his description of the pump’s verticality, armoured and ironed appearance and position in the yard: “there the pump stands, a slender, iron idol, snouted, helmeted […] marking the centre of another world” (PO 17). For the poet, however, the rhythmic squeak of the water pump was as much a central element of life as a source of the poetic imagination. Until years later, it is the sound of the word and its repetition that taps into the memory of the poet and conjures up the soundscape of his childhood.129 Equally important to him is digging and “the ring of it” (DC 25). As many scholars have noted, the rhythmic grating sound of the spade is the sound with which he opens not 137 only his first collection of poetry, but also his poetic vocation.130 In ‘Digging’, Heaney celebrates the family tradition of farming. The poem was written in the summer of 1964, when, amidst the bustle of life in Belfast, Heaney recalls the most familiar and comforting sound he associates with his father and forefathers. The first stanza portrays the poet-narrator sitting at a window, holding a pen between his fingers and listening to his father digging the flowerbeds in the backyard of their house: “Under my window, a clean rasping sound / When the spade sinks into gravelly ground: / My father, digging” (DN 1). The rhythmic grating sound of the spade as it cuts through the gravelly ground immediately transposes Heaney to twenty years before and the memories of his father “stooping in rhythm through potato drills” (DN 1). It is not just the “sound” of digging but also its distinctive “rhythm” that registers the familiar keynote of the family tradition in his memory perpetually (see 1.4.1). Heaney declares that he will continue the family tradition, not with a spade, but with the pen in his grasp. Digging is emblematic. g g ( ) p ‘Digging’, in fact, was the name of the first poem I wrote where I thought my feelings had got into words, or to put it more accurately, where I thought my feel had got into words. Its rhythms and noises still please me, [...]. I wrote it in the summer of 1964, almost two years after I had begun to ‘dabble in verses’. This was the first place where I felt I had done more than make an arrangement of words: I felt that I had let down a shaft into real life. (PO 41) 131 g ( ) 1 The metaphor of digging returns in a number of other poems, as in ‘Poem’ (DN 35) and ‘Undine’ (DD 1 130 The position of ‘Digging’ not only on the opening page of his first volume, but also his Selected Poe 3.3 The Epic of the Noise: Anthrophonic Sounds126 It establishes the centripetal and centrifugal rhythm Heaney would maintain throughout his entire poetic vocation. From the beginning, the young poet is determined to bring the world of farming and writing together, to excavate his cultural identity from the hidden corners of memory and to restore “culture to itself” (PO 41).131 Digging takes us to the memories of his fathers and ancestors in the potato fields and turfs, to their silence, rough hands and determined steps which he greatly admires. In Heaney’s poetry, even the farmer’s footsteps and silence are amplified and celebrated. In ‘Digging’, he recalls “the squelch and slap / of soggy peat” as his grandfather waded through 138 peat bogs (DN 1). In ‘Follower’, Heaney expresses his respect for his father, who unlike the poet, is “an expert”. He is a man of few words, strong, tenacious and sure-footed, mapping and digging the furrows with precision, deftly handling the plough, the shafts and the sod. The young Heaney, by comparison, is “tripping, falling, / yapping always” (DN 12). This theme is subtly brought up in ‘Poem’, a poem dedicated and addressed to his wife. Heaney is determined to renew his wedding vows with Marie promising her to mature the child of his memories into a perfect man (“love I shall perfect for you the child”) (DN 35). The child of Heaney’s memories is comparable to the “stumbling” child of ‘Follower’ (DN 12). He too is “puddling through muck” and “splashing delightedly” (DN 35). However, Heaney hopes that, in this new world of companionship that their union has brought into being, he will overcome his own “imperfect limits” and grow into the ideal man of his mind. The “perfect” he aspires to become is, like his father, a skilled farmer capable of handling “heavy” spades, wading through the mushy land, sowing seeds, and building dams (DN 35). In ‘Anahorish’, Heaney evokes the memories of “those mound-dwellers”, whose way of life “differed little” from that of his father and grandfather (Vendler 18). They also worked “[w]ith pails and barrows” (WO 6). They were such resilient men who would “go waist-deep in mist / to break the light ice / at wells and dunghills” (WO 6). 3.3 The Epic of the Noise: Anthrophonic Sounds126 In ‘Land’, the poem in which Heaney expresses his readiness to leave his homeland and start a new lifestyle, the poet’s own determined footsteps immediately reverberate in the ear of the reader through Heaney’s use of punctuation, linked plosives and word repetition in the first line. He is not stumbling, yapping, splashing, or puddling. He is not following in the “broad shadow” of his father (DN 12). Here, he is the resolute farmer himself: 139 It is impractical to list and describe all the diverse anthropogenic sounds related to agricultural lifestyle. The vernaculars of manual farming methods are often discrete, slow, and heavyweight. Heaney shows that through a careful balance between enjambment and punctuation in these lines. The farmer’s calendar and clock are synchronised by the circadian and seasonal rhythms, and the rhythms of his labour are synchronised with the breath cycle and the habits of hands and feet (Schafer 63). In ‘Land’, Heaney performs the family tradition for one last time. His use of first-person voice reveals his affinity and familiarity with the practice. He carefully recreates the acoustics of farming routine through his action verbs and metonymic references. He is awake at dawn to make sure he has enough time to finish the routine before the sun sets, marches through his land in his boots, stops, stoops, straightens up, and then falls in again, digging, ploughing, weeding, cutting, plucking, irrigating. Having performed and instilled the rhythms of the farm life in his mind, Heaney is now ready to begin a new lifestyle: I composed habits for those acres so that my last look would be neither gluttonous nor starved. I was ready to go anywhere. (WO 11) I composed habits for those acres so that my last look would be neither gluttonous nor starved. g I was ready to go anywhere. (WO 11) Among the other sounds and rhythms absent from the contemporary soundscape of Irish agriculture is the traditional way of making butter from milk. The mechanical separators were introduced into the country towards the end of the nineteenth century to gradually replace the older methods, which remained in use on small farms until the 1940s (Keenan 47). Like the tasks of traditional agriculture, the process of churning manually is laborious and requires much patience, accuracy and skill. 3.3 The Epic of the Noise: Anthrophonic Sounds126 In ‘Churning Day’, Heaney uses a wide a range of sensory images and metaphors to render his experience of the process in the family farm − visual (“a yellow curd” and “white insides”), olfactory (“acrid as a sulphur mine” and 140 “the house would stink long after”), and tactile (“arms ached. Hands blistered”). Heaney uses onomatopoeia and precise auditory images to also render his memory of the curdling milk as it “slugged and thumped” and “spattered”, of “the plash and gurgle of the sour-breathed milk”, and of “the pat and slap of small spades on wet lumps” (DN 9-10). He comments that his use of such an “exact” and “evocative” way of describing the experience was “more a case of personal securing […], an entirely intuitive move to restore something to yourself” (SS 16). Yet there is no doubt that his private motives have acquired more general and social dimensions. More importantly churning had a regular and reassuring rhythm that was set by his mother and continues to stay with him and reverberate in the ears of his readers: “My mother took first turn, set up rhythms / that slugged and thumped for hours” (DN 9-10). Another acoustic experience inscribed in the mind of the young Heaney is the smithy’s strikes on the anvil, a resonating tintinnabulation that Heaney describes as “the noise of myth” (SS 91). Although the first strikes on iron could be heard as early as the Iron Age and human discovery of the basic properties of iron, it was not until the medieval period that blacksmithing became an indispensable part of everyday life in almost every town and village. Farmers, in particular, needed the services of the blacksmith to shoe their horses, repair their ploughs and iron gates, and constantly sharpen their scythes during harvest seasons. Artisans, armies and households needed something made with iron (Westfahl 589- 592). In fact, until the mid-nineteenth century, the strikes of the blacksmith's hammer produced one of the most strident sounds that could reach to over 100 decibels and be heard by most residents every day from dawn to dusk (Schafer 57-58). With the introduction of machinery and mass production during the Industrial Revolution, blacksmithing became a rather redundant craft limited to farrier work. o ge : Why did he say all I know is a door into the dark? The reason was he never had been in. He was a lad going to school and he’d go past. He refers to the man standing outside with the hair under his lip, hair in his nose, leaning against the jamb. That was my father. The description of the anvil crowned like a unicorn at the centre of the floor, he describes everything which is directly here. y q g See also Sean McCormack’s 2014 (www.youtube.com/watch?v=CDWLUL5s7NI) and Roger Lennox’s 2015 (www.youtube.com/watch?v=eGw0NhUXG00) interviews with Barney, where the old man talks about his long- term relationship with the poet and shows around his forge. , See www.youtube.com/watch?v=IqWGTeBcxYg. 132 In a 2014 interview by Countryfile, Barney Devlin comments about the child Heaney and the poem ‘The Forge’: 3.3 The Epic of the Noise: Anthrophonic Sounds126 For Heaney, the tintinnabulation that came from the local blacksmith’s forge belonged to the age of epics and myths, while car horns and rolling tyres produced contemporary keynotes. Referring to Barney Devlin’s forge, situated at 141 Hillhead roadside, in Bellaghy, Co. Derry, he remarks: “you had the noise of myth in the anvil and the noise of the 1940s in the passing cars. As ordinary or archetypal as you cared to make it. Barney’s in his late eighties now, but still capable of striking the epic out of the usual” (SS 91). 132 In the poet’s mind, Barney Devlin remains the “master blacksmith” of all times (Westfahl 590) inspiring his early poem ‘The Forge’, as well as ‘Poem to Blacksmith’ and ‘Midnight Anvil’ in his later collections. ‘The Forge’ can be read as an elegy to the dying craft. Heaney’s second collection takes its title from the first line of the sonnet: “All I know is a door into the dark” (DD 9). To begin with, the term “dark” adds a sense of mysteriousness to the interior of the forge. The fact that the profession belongs to the past getting rusty and dusty in the dark corners of the archives becomes more evident when placed in contrast with the brightness of the outside world. The idea is reinforced through the images of the “old” and “rusting” tools visible from the outside and through the “short-pitched ring” coming from inside of the forge: “Outside, old axles and iron hoops rusting; / Inside, the hammered anvil’s short-pitched ring” (DD 9). Heaney tells us that even Barney is saddened when reminiscing about the forge’s thriving past in the days of horses and carts: “He leans out the jamb, recalls a clatter / Of hoofs where traffic is flashing in rows; / Then grunts and goes in, […]” (DD 9). Nonetheless, the poem opens the door into the darkness of the forge’s interior to see “the unpredictable fantail of sparks” of creation as they enlighten the place and hear the “hiss when a new shoe toughens in water” (DD 9). The precision of his visual and acoustic images and the first-person point of view with which Heaney narrates his recollections transfer a 142 strong impression of immediacy and intimacy to the readers, placing them in direct sensory relation with the nature of the profession. 3.3 The Epic of the Noise: Anthrophonic Sounds126 Heaney glorifies the craft by also adding an air of otherworldliness and spirituality to its seemingly ordinary tools and space. During the Middle Ages, blacksmiths were always respected, or even feared, for their seemingly magical ability to start and maintain huge fires and crafting polished items out of black and heavy irons (Westfahl 589-590). To forge iron, blacksmiths required several tools: a furnace, a bellow, tongs, several types of hammers, axes, chisels, moulds, and an anvil. For Heaney, it is the anvil that has a central role in the bl k ith’ i l biliti blacksmith’s magical capabilities: Heaney recalls the occasion in Stepping Stones: He’s still going strong; the last time I was with him, he showed me two different anvils and played them for their two different musics: a sweet and carrying note from the one that had belonged to his grandfather – which is the one I would have heard a mile away when I was a youngster – and an abrupt unmelodious dint from a later industrial ingot, definitely not the one that rang in the year 2000. (SS 91) g , y g y ( ) 134 The poem was written in the 15th century. Besides the din of the forge, the writer conjures the noisiness of the smiths themselves as they yell for more coal, spit, or groan. The poem appears in Davies’s Medieval English Lyrics (1991). See also: www.theguardian.com/books/2012/feb/20/poem-of-the-week-the-blacksmiths. 133 Heaney recalls the occasion in Stepping Stones: He’s still going strong; the last time I was with him, he showed me two different anvils and played them for their two different musics: a sweet and carrying note from the one that had belonged to his grandfather – which is the one I would have heard a mile away when I was a youngster – and an abrupt unmelodious dint from a later industrial ingot, definitely not the one that rang in the year 2000. (SS 91) 134 The poem was written in the 15th century. Besides the din of the forge, the writer conjures the noisiness of the smiths themselves as they yell for more coal, spit, or groan. The poem appears in Davies’s Medieval English Lyrics (1991). See also: www.theguardian.com/books/2012/feb/20/poem-of-the-week-the-blacksmiths. 133 Heaney recalls the occasion in Stepping Stones: blacksmith’s magical capabilities: The anvil must be somewhere in the centre, The anvil must be somewhere in the centre, Horned as a unicorn, at one end square, Horned as a unicorn, at one end square, Set there immovable: an altar Where he expends himself in shape and music. (DD 9) The word “unicorn” with which he describes the anvil’s front horn gives it a fantastical quality linking it, as Andrews and Tobin note, to the realm of legends and medieval romantic tales (Andrews 26; Tobin 50). The magical anvil is positioned “somewhere” in the middle of Barney’s workshop; but more importantly, it is at the “centre” of his world and the core of his art. The anvil is an “altar”, his workshop a sacred place and the blacksmith himself a priest performing a ritual. But he is also, like Heaney himself, a maker expending his life in “shape and music” (DD 9). To Heaney, the anvil is not an ordinary tool but an archetypal one. Likewise, the sound it produces is not a noise. It is, like its position and shape, irremovable from memory. The archetypal status of the anvil and its melody is highlighted in ‘Midnight Anvil’ as well. The poem is for the millennium New Year’s Eve in Co. Derry, when Barney, already in his 80s, strikes the anvil with his hammer twelve times. Although Heaney was not personally present at the forge, the “sweet and carrying note” of the strikes continues to reverberate in 143 his ear. Thus, he opens the poem with: “I can still hear it” (DC 26).133 Here too Heaney highlights the otherworldliness of the clang by putting it side by side with church bells alluding to Herbert’s ‘Prayer 1’ in which a prayer is described in terms of several images including “a kind of tune”: “Afterwards, I thought / Church bels beyond the starres heard […]” (DC 26). In the final lines, he quotes from ‘Poet to Blacksmith’ to capture the perfection of the outcome: the forged spade “ringing sweet as a bell” (DC 27). In fact, for Heaney, any sound coming from the smithy’s workshop is not only a source of inspiration, but also the material of poetry itself: his ear. 133 Heaney recalls the occasion in Stepping Stones: grandfather – which is the one I would have heard a mile away when I was a youngster – and an abrupt unmelodious dint from a later industrial ingot, definitely not the one that rang in the year 2000. (SS 91) 134 The poem was written in the 15th century. Besides the din of the forge, the writer conjures the noisiness of the smiths themselves as they yell for more coal, spit, or groan. The poem appears in Davies’s Medieval English Lyrics (1991). See also: www.theguardian.com/books/2012/feb/20/poem-of-the-week-the-blacksmiths. 133 Heaney recalls the occasion in Stepping Stones: He’s still going strong; the last time I was with him he showed me two different anvils and played 133 Heaney recalls the occasion in Stepping Stones: He’s still going strong; the last time I was with him, he showed me two different anvils and played blacksmith’s magical capabilities: Thus, he opens the poem with: “I can still hear it” (DC 26).133 Here too Heaney highlights the otherworldliness of the clang by putting it side by side with church bells alluding to Herbert’s ‘Prayer 1’ in which a prayer is described in terms of several images including “a kind of tune”: “Afterwards, I thought / Church bels beyond the starres heard […]” (DC 26). In the final lines, he quotes from ‘Poet to Blacksmith’ to capture the perfection of the outcome: the forged spade “ringing sweet as a bell” (DC 27). In fact, for Heaney, any sound coming from the smithy’s workshop is not only a source of inspiration, but also the material of poetry itself: What I’ll do instead Is quote those waterburning Medieval smiths: ‘Huf, puf! Lus, bus! Col!’ Such noise On nights heard no one never. (DC 26) Heaney also quotes from an anonymous medieval poem, usually known as ‘The Blacksmiths’, that captures and criticises the clamour of a busy medieval forge through skilled alliteration and precise auditory and visual images.134 The “huf! puf!” of the bellow, as it pipes the air to the base of the fire, and the “lus, bus” of the hot metal, when immersed in water, are not ordinary noises, but archetypal sounds and lines from the poetry of the smithy that Heaney would like to “quote” in his own poem (DC 26). Heaney dedicates his poem not only to Barney, but to all the blacksmiths: “those waterburning / Medieval smiths” (DC 26). After all, to him, “any one forge is all the forges” (SS 91). 144 Heaney’s treatment of the thatcher’s acoustics is no different. He makes some passing references to the thatcher in Wintering Out. In ‘Land’, he is the creator of the “plaited and branchy” effigy that transposes the poet to the Irish past (WO 11). In ‘Maighdean Mara’, he is the tradesman who hides the selkie’s magic garment in his stack and unwittingly brings about her departure (WO 57). In ‘Thatcher’, from the previous collection, he captures the thatcher’s acoustics accurately. Like the blacksmith, the figure of the thatcher comes from the countryside where he grew up. Unlike the blacksmith’s, however, the nature of his work is quieter. The keynotes of the thatcher are bound to his tools. The man deals with reeds, rushes, “wheat-straw”, “rods”, and “hazel and willow” timbre (DD 10). 135 For making eel traps, harvesting baskets, roofs, huts, and other wickerwork and basketwork, some of which are mentioned in ‘At a Potato Digging’ (DN 18-20) and ‘Land’ (WO 11-12) y, y “unfussed” but works “professionally” (DN 23). gg g ( ) ( ) 136 Much in the same way, Heaney imbues work of the water diviner with a sense of otherworldliness. He too is “unfussed” but works “professionally” (DN 23). re mentioned in At a Potato Digging (DN 18 20) and Land (WO 11 12) 6 Much in the same way, Heaney imbues work of the water diviner with a sense of otherworldliness. He unfussed” but works “professionally” (DN 23) or making eel traps, harvesting baskets, roofs, huts, and other wickerwork and basketwork, some of whi ti d i ‘At P t t Di i ’ (DN 18 20) d ‘L d’ (WO 11 12) gg g ( ) ( ) 136 Much in the same way, Heaney imbues work of the water diviner with a sen “unfussed” but works “professionally” (DN 23). 137 Heaney is not alone in this influence from the technological transformations. The representation of machinery in the rural environment is a recurrent feature of some poetry from the period. For instance, in ‘Cynddylan on a tractor’, from An Acre of Land (1952), the Welsh poet R. S. Thomas, similarly recounts the role of the tractor in rural Wales. 138 Similarly, in ‘High Summer’, the children soothed into sleep by the “familiar, [yet] ignorant and hard” gargling of the neighbour’s tractor (FW 41). Coming from an agricultural background, the fisher men of ‘Eelworks’ adopt tractor engines for their boats (HC 28-32). In ‘Keeping Going’, the tractor symbolises the poet’s brother’s “stamina” who is driving it (SL 12). 139 p g ( ) 139 See www.belfasttelegraph.co.uk/news/northern-ireland/how-belfast-telegraph-gave-seamus-heaney-an-early- break-29546167.html. blacksmith’s magical capabilities: Due to the geography of Ireland and the abundance of vegetation, these materials have been familiar to the Irish since the earliest period of Celtic civilisation and possess archetypal significance.135 The movements of the thatcher’s hands and the rhythm of his work too are also unique to him. He works gradually and quietly but steadily and deftly. Heaney reflects this through his assonance and rhyme (“flicked” and “twisted”, “eaves” and “sheaves”) and meticulous descriptions: he flicks and twists the rods to test their strength, sharpens their ends, carries up sheaves “handful by handful” and lays them in regular breadths, and stitches the butts together with ropes to build the summit. But, like the blacksmith, the thatcher too is an inspirational figure. He is standing upon his ladder and the young Heaney watches him in awe from the ground, as he handles everything with “his Midas touch” (DD 10).136 The thatcher’s mastery of his work and his dedication become a metaphor for the practice of writing, which involves technique and craftsmanship as much as inspiration and intuition. As the poet’s world grows, the rhythm of life begins to change. Until the arrival of the As the poet’s world grows, the rhythm of life begins to change. Until the arrival of the diesel engine tractor, in the 1920s, people, bulls and horses played an essential role in Irish 145 agriculture. With the introduction of mechanisation and the advancement of technology, the internal combustion engine became the fundamental keynote of farm life. Soon the motor of newly introduced tractors became increasingly light and versatile, able to manage a wide range of farm tasks from ploughing, threshing, hay making, fertilizing, and pulping with the speed of almost six horses (Keenan 139). Heaney recalls this transitional phase vividly: The horses were used for carting stuff from field to barn, or from shop to farm – grain, potatoes, provisions; or for work in the fields – ploughing, harrowing, drilling, grubbing, rolling, mowing, potato digging, drawing in hay or corn or turf. And yes, of course, the tractor gradually replaced them. […] . The pace of farm work speeded up greatly with the tractor: on those small-holdings, a man who owned a tractor could do his ploughing in less than half the time it had taken previously; […]. blacksmith’s magical capabilities: (SS 10) The fact that agricultural machinery had begun to turn into an unquestionable part of farming soundscape is reflected in the considerable numbers of poems in which Heaney evokes them.137 In ‘Tractors’, ‘A Potato Digging’, ‘The Baler’, ‘Glanmore Sonnets I’ and ‘In a Field’, Heaney revisits the soundscape of rural Ireland through the sounds of the mechanical digger, tractor, combine harvester and hay baler.138 The fact that agricultural machinery had begun to turn into an unquestionable part of farming soundscape is reflected in the considerable numbers of poems in which Heaney evokes them.137 In ‘Tractors’, ‘A Potato Digging’, ‘The Baler’, ‘Glanmore Sonnets I’ and ‘In a Field’, Heaney revisits the soundscape of rural Ireland through the sounds of the mechanical digger, tractor, combine harvester and hay baler.138 For Heaney, the gargling of tractors is the acoustic symbol of empowering and enduring power and almost synonymous with farm life and prosperity. Published in The Belfast Telegraph in November 1962, the poem ‘Tractors’ captures the fascination of the young Derry poet with tractors lumbering ruthlessly “on roadways” and “on land”, in both hot summer days and wet winters with “no fear”.139 Tractors return frequently in his poetry. Spring is the time when many farmers prepare the soil for sowing. In ‘Glanmore Sonnets I’, the “gargling tractors”, the lathe, and the ploughing machine permeate the poem to create the 146 sense of rebirth and fertility (FW 28). Heaney thinks of his “personal and Irish pieties” as vowels (PO 37). In these poems, it is the presence of the agricultural machinery that breathes “good life” and “vowels” to the ground of the fields as to his poetry (FW 28). In ‘The Baler’, the familiar and continuous “clunk” of a hay baler triggers the memories of haymaking in his childhood home. Heaney wrote the poem after experiencing a stroke in 2006. The “all day” and “ongoing” sound of the baler not only affirms its status as a keynote of farm life, but also turns it into a metaphor for life itself and the dull and continuous pounding of the heart, which we take for granted (“All day the clunck of a baler / Ongoing, cardiac-dull, / So taken for granted”). 140 The poem was written at the request of the poet laureate Carol Ann Duffy, who asked poets to respond to poetry, letters and diary entries from the time and Heaney chose Thomas’ poem. The poem was completed two months before Heaney’s death and published by the Guardian before it appeared in the anthology. See www.theguardian.com/books/2013/oct/25/seamus-heaney-last-poem-published. 0 The poem was written at the request of the poet laureate Carol Ann Duffy, who asked poets to respo months before Heaney s death and published by the Guardian before it appeared in the www.theguardian.com/books/2013/oct/25/seamus-heaney-last-poem-published. blacksmith’s magical capabilities: He suggests that the loud, continuous clunk- sound of the hay baler’s blades as well as the accelerating engine of the tractor (“the giddied- up race of a tractor”) enriched the summer evenings of the hayfields (HC 24). Heaney wrote his final poem, ‘In a Field’, as a response to Edward Thomas’ poem, ‘As the Team’s Head Brass’, in order to contribute to a memorial anthology marking the centenary of the outbreak of the first world war. Thomas’ poem is set in the wood and features a farmer ploughing with a horse. Heaney’s poem features a tractor from the familiar fields of Ulster racing to finish the farm tasks before sunset: “The tractor with its hoisted plough just gone / Snarling at an unexpected speed / Out on the road. Last of the jobs, […]”. 140 In the present day, the constant engine noises, from water pump generators, tractors and power tools, have replaced the slower and quieter stamping of the horses’ hooves and the farmers’ boots, as well as the regular rhythm of scything, digging and tilling in the once quiet ambience of the countryside. Heaney is very well aware of the downsides of machinery, of the disappearance of rhythms, songs and collective practices in the fields, and of the presence and dominance of noise. His awareness is reflected in ‘At a Potato Digging’ which features 147 “a mechanical digger” (DN 19). The poem depicts the relationship between the Irish and the earth, as well as their dependence on potato crops, hinting at the Irish potato famine of the 1840s, beginning with the following lines: “A mechanical digger wrecks the drill, / Spins up a dark shower of roots and mould” (DN 19). The presence of the harsh and mechanised harvesting process indicates the introduction of the modern harvesting methods and a more contemporary period but it is juxtaposed with the demonstrations of a menacing condition suggested by such terms as “wrecks” and “a dark shower of roots and mould” (DN 19). The workers, on the other hand, are involved in a traditional and intense manual labour. They are being cooperative with the machine, stooping to collect the potatoes and fill their creels. Yet they are lacking harmony, lyric and rhythm. blacksmith’s magical capabilities: To begin with, they “swarm in behind” the machine, like a flood of insects, and then: “Like crows attacking crow-black fields, they stretch / A higgledy line from hedge to headland […]” (DN 19). Their disorganised and rushed movements across the field are compared to the flight of a flock of crows. All of the labourers’ efforts are put towards the operation of the machine but their circumstances, as Heaney says, have not improved: “Fingers go dead in the cold” (DN 19). Once again, the harvesting process is seen as harsh and intense for the workers, just as it was in the past. In Wintering Out, ‘Serenades’ hints at this contemporary issue of noise pollution by inviting his wife to delight in the sounds of birds before they disappear “[b]etween combines and chemicals” (WO 50). Heaney sees the increasing noise of machinery as a threat not only to the natural environment but also the age-old traditions and crafts of Ireland, urging his readers to beware of it. Like the sounds of crafts and industries, the sounds emitted by the various means of Like the sounds of crafts and industries, the sounds emitted by the various means of transport signal the presence of human beings and fall into the category of anthrophonic sounds. Heaney evokes the sounds of various public, private and military, as well as ancient and contemporary modes of transportation, such as horses, carts, wagons, bicycles, taxis, 148 lorries, ambulances, hearses, trains, planes, helicopters, boats and ferries. In 2019, the National Library of Ireland (NLI) launched a special campaign running from the poet’s 80th birthday on 13 April, during which five of Heaney's poems − ‘The Clothes Shrine’, ‘Postscript’, ‘The Railway Children’, ‘Route 110’ and ‘Squarings’ − were displayed on buses, trams and trains throughout the country. The poems were chosen by Heaney’s family and share common themes related to journeys, transport and daily life. Referring to this occasion, the Director of the NLI, Sandra Collins echoes the 1995 Nobel Prize citation141 commenting on Heaney’s ability to bring “the wonderful” out of “the ordinary”. 142 My study of the sounds of transport in Heaney’s Wintering Out highlights the same aspect of Heaney’s poetry. I apply the theoretical perspective of acoustic ecology to show how the various means of transport and their noises are charged with multiple layers of personal and collective implications. p , y y g p See www.nobelprize.org/prizes/literature/1995/press-release/. p g p p 142 The intention was to not only celebrate the poet’s 80th birthday, but also bring his thoughts and words into the lives of “harried commuters” caught up in the daily grind. See www.lonelyplanet.com/articles/seamus-heaney- public-transport. 141 In 1995, the Nobel Prize citation paid tribute to Heaney’s work for its combination of “lyrical be ethical depth, which exalt everyday miracles and the living past”. p y y g p See www.nobelprize.org/prizes/literature/1995/press-release/. p g p p 142 The intention was to not only celebrate the poet’s 80th birthday, but also bring his thoughts and word li f “h i d ” h i h d il i d S l l l / i l / , p y ethical depth, which exalt everyday miracles and the living past”. blacksmith’s magical capabilities: The sounds related to various types of transport in more ancient times, such as the stamping of hooves, creaking of cart wheels, cracking of whips, and the more recent clang of bicycles, hum of engines, rumbling of trains, and the high-frequency noises of airplanes and helicopters were each keynotes of specific periods of time in history. Prior to the advent of motor cars, walking, horseback, jaunting cars, were the major modes of transportation across Ireland until well into the 1960s and the expansion of road construction (Biagini and Daly 206). Heaney refers to the sounds of passing cars as “the noise of the 1940s”, when he lived in Mossbawn, Co. Derry (SS 91). The family house was on the main road − between Belfast and Magherafelt, Belfast and Cookstown, Belfast and Derry − where he could hear the sounds of passing cars, buses, bicycles, lorries and vans. But the clip-clopping of horses and 149 donkeys, and rattling of carts and wagons were the sounds that had already started to become “rare enough” (SS 4). In ‘Bog Oak’, ‘The Tollund Man’, and ‘Linen Town’ from Wintering Out, Heaney evokes the sounds of cart wheels and hoofbeats to distance his readers from the contemporary urban soundscape and transpose them to an earlier period of time in history. For the poet, peat bogs are the archives and witnesses of history and sites of memory banks. In ‘Bog Oak’, Heaney makes use of an approaching cart to evoke a sense of history and the need to retrieve it in the present time. A crucial figure in this process is the carter. The approaching cart is moving in triumph, pounding fervently and unweariedly along “the cart track” bringing its “trophy” into the present and future of his Ireland (WO 4). In ‘The Tollund Man’, Heaney compares his “freedom” − while driving to Aarhus − with the ironic “freedom” of the Iron Age man − as he was being driven to the site of his slaughter (WO 36-37) (see also 3.4 and 4.2). Heaney uses this acoustic feature in ‘Linen Town’ to transport his readers to Belfast in the early Industrial period and before the arrival of cars, when the clip-clop of horse hooves on the pavements was the keynote sound of urban streets. 143 Also, in ‘Eelworks’, he refers to the fisherman as a “[h]orse-and-cart men” to create a rural atmosphere (HC 28); in ‘Known World’, he uses tractors and carts to create images of countryside roads in Mossbawn and empathise with the refugees from Eastern Europe during the 1990s Balkan Wars (EL 19-25); in ‘Station Island’, the cart recalls the memories of his childhood and the tinkers (SI 62); in ‘Birthplace’, he likens the “slow” pace of the wind in Hardy’s hometown, Upper Bockhampton, to the pace of “Victorian rural life” which he associates with the “slow” pace of “a cart / coming late from market” (SI 34); in The Last Walk, Heaney’s version of Pascoli’s L’Ultima Passeggiata, there is reference to a cart too. In fact, one of the reasons why Heaney was drawn to Pascoli was precisely the shared interest in rural activities and sounds (Rankin Russell 252-256). blacksmith’s magical capabilities: In ‘Linen Town’ the sound of time is very prominent, yet it is the rhythmic clopping of the cloaked rider’s horse that re-creates the acoustic space of Belfast’s Linen Square in 1786, where the young Henry Joy McCracken was hanged on lands once donated to the city by his family.143 Like the rattling of carts and carriages, the clicks and clunks of bicycles mark the poet’s early years. Heaney’s earliest experiences of such sounds date back to the forties and 150 the memories of Mossbawn, when recounting the sounds of transport coming from the road in front of the family house: the memories of Mossbawn, when recounting the sounds of transport coming from the road in front of the family house: mories of Mossbawn, when recounting the sounds of transport coming from the road f th f il h There were bicycles too, a few regular pedallers and the occasional traveller from farther up or down the country. Kathleen Garvin going to her job in Castledawson. Paddy McNicholl going to do yardwork in Gribbin’s of Anahorish. RUC men now and again. But the most spectacular was definitely Master Pollock, a schoolmaster, a recreational cyclist in shorts, pelting along on his ‘racer’ – he always created a bit of a stir. (SS 4) As a child, Heaney was fascinated by bicycles. In ‘Wheels within Wheels’, he reveals that “the first real grip [he] ever got on things” involved a bicycle (ST 46). The poem lists the discoveries of the inquisitive youngster through his bicycle, recounting any experiment from turning the bicycle upside down, pedalling it with his hand really fast until the spokes became indistinguishable, throwing a potato and chucking a straw into the spinning wheels, riding in the mud and dung and spinning the wheels in the water (ST 46-47). In fact, such descriptions and their visual, aural, kinaesthetic and tactile associations could trigger the reader back to their own childhood and the period of life filled with innocence, curiosity and fascination. The young Heaney associates bicycles with creativity, speed, aspiration and power (“the art of pedalling”, “an access of free power” and “a new momentum”, “an orbit coterminous with longing”) (ST 46-47). After all, each time, the schoolmaster’s fast pedalling along the road “created a bit of a stir” (SS 4). Aunt Mary, with whom Heaney was very close was a “bicycle woman” (SS 5). 144 In ‘An Ulster Twilight’, he associates the sound of Eric Dawson’s pedalling with his childhood memories of Christmas (SI 38); in ‘Station Island IV’, he associates his childhood memories of Terry Keenan with the man’s cycling (SI 70); in ‘The Summer of Lost Rachel’, he sets the innocence of the young Rachel and her bike against the merciless power of the car that caused the fatal accident (HL 36-37); yet in ‘Senior Infants’, the young Heaney and his friend, Tommy, shoot the bicycle’s bell with an airgun (DC 29-30). 144 In ‘An Ulster Twilight’, he associates the sound of Eric Dawson’s pedalling with his childhood mem Christmas (SI 38); in ‘Station Island IV’, he associates his childhood memories of Terry Keenan with the man’s cycling (SI 70); in ‘The Summer of Lost Rachel’, he sets the innocence of the young Rachel and her bike against the merciless power of the car that caused the fatal accident (HL 36-37); yet in ‘Senior Infants’, the young Heaney and his friend, Tommy, shoot the bicycle’s bell with an airgun (DC 29-30). p ( ); y Heaney and his friend, Tommy, shoot the bicycle’s bell with an airgun (DC 29-30). the merciless power of the car that caused the fatal accident (HL 36-37); yet in ‘Senior Infants’, the young Heaney and his friend, Tommy, shoot the bicycle’s bell with an airgun (DC 29-30). 145 The first pedal-driven bicycle arrived in Ireland in the 1860s. Yet it is interesting to know that, the first “safety bicycles” with a pneumatic tyres were invented by the Belfast-based John Dunlop in 1888, turning it into an important part of the social and cultural life in Ireland. Perhaps this cultural “craze” for the bicycle could partially explain Heaney’s fascination with the vehicle. p y p y See www.museum.ie/Country-Life/Exhibitions/Current-Exhibitions/Cycling-the-Country/The-History-of-the- Bicycle. 145 The first pedal-driven bicycle arrived in Ireland in the 1860s. Yet it is interesting to know that, y p y www.museum.ie/Country-Life/Exhibitions/Current-Exhibitions/Cycling-the-Country/The-History-of-the- cle blacksmith’s magical capabilities: 152 Heaney captures this striking characteristic of trains in some of the lines from his Nobel lecture, when describing his and his siblings’ intuitive receptiveness to the ambient sounds which he concretises through an aquatic analogy: lecture, when describing his and his siblings’ intuitive receptiveness to the ambient sounds It [the Mossbawn house] was an intimate, physical, creaturely existence […]. We took in everything that was going on, of course – rain in the trees, mice on the ceiling, a steam train rumbling along the railway line one field back from the house – but we took it in as if we were in the doze of hibernation. Ahistorical, pre-sexual, in suspension between the archaic and the modern, we were as susceptible and impressionable as the drinking water that stood in a bucket in our scullery: every time a passing train made the earth shake, the surface of that water used to ripple delicately, concentrically, and in utter silence. (OG 447) Like the noises of factories, the rumbling of the locomotives was the by-product of the Industrial Revolution. The first railway was set in England in 1825 to carry coal from the mines across Britain. Yet within just a few years, it conquered the world as one of the most familiar and exciting keynotes of travelling unlike any other previous means of transportation. Compared to the previous horse-drawn carts, carriages and wagons, trains were faster, more comfortable, smoother, and more reliable, particularly for merchants and manufacturers (Herbst 29-30). Colonisation facilitated the introduction of railways to Ireland. In the 1940s, when Heaney was growing up in Mossbawn, the sound of the steam train coming from the railway behind the fields had turned into one of the most familiar keynotes of Co. Derry as surely as the sounds of passing cars, buses, vans, lorries driving through the road. He writes: “[m]aybe a train shunting at the station in Castledawson – because we had the railway, running parallel to the road, about a hundred yards away at the end of the field behind the byre” (SS 8). blacksmith’s magical capabilities: The thatcher, whom he admired, was also a pedaller (DD 10). Later, during Queen’s summer vacations, he would go on romantic bicycle rides along the Bann’s riverbank (SS 406). But growing up in a country that had experienced centuries-long colonialism, sectarian violence and religious segregation, the sounds of bicycle could stir up contradictory emotions.144 151 Bicycles − in their contemporary form with chain and rubber tyres − began to be mass-produced from the beginning of the 1890s, turning this vehicle and its whirring and clicking sounds into the familiar keynote of both urban and rural life in Ireland.145 During the forties, as Heaney recalls it, Northern Ireland’s armed police force − the Royal Ulster Constabulary men (RUC), whom Heaney mentions in the lines quoted above − would patrol the countryside by bicycle. In ‘A Constable Calls’, the sight and sounds of the bicycle trigger feelings of fear and distrust his Catholic family had towards the mostly protestant police force. The poem begins with a close-up of the policeman’s bicycle standing by the window- sill, portrayed from the gaze of the terrified young Heaney, who is witnessing the interrogation of his father. When put alongside the policeman’s boots, cap and gun, the “rubber cowl”, “mud-splasher”, “front mudguard”, “handlegrips”, “dynamo” and “pedal treads” are the signs of threat (N 61-62). The freedom and power machine now belongs to the antagonist. The poem ends with the rhythmic clicking sound that resonates in the mind of the reader ceaselessly: “[…] His boot pushed off / And the bicycle ticked, ticked, ticked” (N 62). Heaney’s onomatopoeic description of the bicycle’s rhythmic clicking sound, as the policeman pedals away, reveals its deep and enduring impact on not only his own mind but also the soundscape and atmosphere of the neighbourhood he lived in. Moreover, the bicycle’s noise recalls that of a clock, as if foreseeing and counting towards the Troubles. Although the sound of the bicycle marks Heaney’s early years, as I noted above, the vehicle had arrived in Ireland much earlier and about the same time as railways began to transform urban and rural landscapes. Unlike the subtle clicking and whirring of bicycles, the rumbling of the locomotives could make the earth quake and walls tremble from miles away. 146 In 2008, Heaney read an excerpt from the poem on Holocaust Memory Day. In the poem, he pays tribute to the victims and makes a reference to the Nazis using trains to move Jews to ‘the place of birth trees’, Auschwitz, during WWII. He talks about “the abstract, lonely curve of distant trains / As we entered history and ignorance” (SL 8). The Shoah of course changed the perceptions of trains. But the train, here, can be a metaphor for transporting everyone from ignorance to the awareness of atrocities occurring in history. 147 James is commenting in reference to an in a 1994 personal interview with Heaney where he comments on the word ‘sway’: “an image of command […] and then there is one’s swaying in sympathy or of necessity. […] It has that double sense, that double possibility, of active or passive engagement” (qtd in James 273). blacksmith’s magical capabilities: The train introduced a rich variety of sounds − the whistle of an approaching train, the slowing of the engine as the train shunts into the station, the chattering of passengers, hurried footsteps, the loud cries of the ticket man, the collecting and punching of tickets, the gradual acceleration of the wheels, the rattling of carriages, and the sudden explosions of escaping steam − into the soundscapes, which Heaney carefully evokes in a childhood make-believe game in ‘A Sofa in the Forties’: 153 All of us on the sofa in a line, kneeling Behind each other, eldest down to youngest, Elbows going like pistons, for this was a train And between the jamb-wall and the bedroom door Our speed and distance were inestimable. First we shunted, then we whistled, then Somebody collected the invisible For tickets and very gravely punched it As carriage after carriage under us Moved faster, chooka-chook, the sofa legs Went giddy and the unreachable ones Far out on the kitchen floor began to wave. (SL 7) As Heaney suggests in these lines, the train and its memorable and inescapable noises became almost synonymous with “inestimable” distances, and as such with the concepts of travelling, migrating, the unknown (“Ghost-train”) and even death (“Death-gondola”) (SL 7).146 The idea of the train as a means of transporting passengers into a new sound space and as an acoustic signal of the colonisation is reinforced when Heaney juxtaposes it with the voices of the British speaker coming from the radio: Swept and swayed in us like nets in water Or the abstract, lonely curve of distant trains As we entered history and ignorance. (SL 8) 146 In 2008, Heaney read an excerpt from the poem on Holocaust Memory Day. In the poem, he pays tribute to the victims and makes a reference to the Nazis using trains to move Jews to ‘the place of birth trees’, Auschwitz, during WWII. He talks about “the abstract, lonely curve of distant trains / As we entered history and ignorance” (SL 8). The Shoah of course changed the perceptions of trains. But the train, here, can be a metaphor for transporting everyone from ignorance to the awareness of atrocities occurring in history. blacksmith’s magical capabilities: 154 The invention of radio separated sounds from sound-sources and transformed the idea of acoustic space by extending the outreach of sounds and connecting various interrupted spaces: “[n]ever before had sound disappeared across space to reappear again at a distance” (Schafer 91-92). This idea is captured in the lines from the poem where Heaney says “Between him and us / A great gulf was fixed” (SL 8). The radio crossed both the geographical and metaphorical “gulf” between Ireland and Britain, enabling the British accent to reach, imperialistically, the ears of the inhabitants of the lands where Irish accents and dialects belonged. Heaney comments that “The sway of language and its furtherings // Swept and swayed in us like nets in water (SL 8). Just like the noises of the train, the “aerial” voices streaming through the radio became a familiar and inescapable background sound for the young Heaney, exposing him to a strong sense of linguistic and cultural divide (SL 8). As Stephen James has noted, Heaney exploits the ambiguity of the term ‘sway’ to suggest how the imperial influence of “the absolute speaker” was nonetheless resisted in the Irish context. Heaney’s “the sway of language” (SL 8) reveals the dynamics between an externally imposed rhetoric and an internally generated drive According to James: Heaney’s “the sway of language” (SL 8) reveals the dynamics between an externally imposed hetoric and an internally generated drive. According to James: The ‘command’ of the newscaster’s voice, whose ‘pronunciation’ is presumably that of BBC ‘Queen’s English’, is presented in the poem as a form of cultural imperialism in a 1940’s County Derry Catholic home. Broadcast across ‘a great gulf’ from the supposed centre to the provinces, this voice communicates an ostensibly definitive version (‘THE NEWS’) of momentous events going on elsewhere; its tone and concerns are so at odds with the dialect and day-to-day experiences of the rural household as to seem a tyrannical imposition. (137)147 ‘A Sofa in the Forties’ was published long after Northern Ireland was annexed to the United Kingdom. 149 McGuckian’s study brings to light Heaney’s fascination with cars and its role as not only an occasionally suitable literary vehicle but also a “fifth element” and a symbol of “a gradual mobilisation of the spirit” (71-72). Examples of Irish poets using the car symbol include John Montague, Paul Muldoon, Paul Durcan and Nuala Ní Dhomhnaill. 148 In ‘Dawn Shoot’, the railway is part of the rural landscape (DN 16). In ‘The Railway Children’ too, the railway and the telegraph’s cable are juxtaposed (SI 45). blacksmith’s magical capabilities: It appears that it was raining heavily and Durcan decided to join Heaney as he drove across the hills of Donegal to drop off friends at a bus stop. Here is Durcan’s impression of the experience: attitude towards the modern-day vehicle is reflected in Paul Durcan’s poem ‘A Spin in the Rain with Seamus Heaney’. It appears that it was raining heavily and Durcan decided to join Heaney as he drove across the hills of Donegal to drop off friends at a bus stop. Here is Durcan’s impression of the experience: Bales of rain But you did not alter your method of driving, Bales of rain Which is to sit right down under the steering wheel And to maintain an upward-peering posture Treating the road as part of the sky, Which is to sit right down under the steering wheel And to maintain an upward-peering posture Treating the road as part of the sky, Which is to sit right down under the steering wheel And to maintain an upward-peering posture p p g p Treating the road as part of the sky, A method which motoring correspondents call Horizontal-to-the-vertical. (Durcan 308) The fast, smooth and quiet operation of the motor engine, conveyed by the poet through the persona of a driver, transfers a sense of liberty, power and autonomy, as we see reflected in a number of Heaney’s own car poems such as ‘The Tollund Man’ and ‘Westering’ from Wintering Out.150 In ‘The Tollund Man’, Heaney directly associates “freedom” with driving. At the beginning of the poem, Heaney states his intention and destination: “Some day I will go to Aarhus” (WO 37).151 In fact, the entire poem takes place as the poet is on an imaginary drive to the sites of killings during the Iron Age, where the body of the Tollund man − now displayed in Silkeborg, Denmark − was first discovered (see 2.2). It is while driving that the poet’s thoughts are drawn to the Tollund man’s last day as he was being carried on “the tumbril” to the place of his slaughter (WO 37). Both men are in motion, free and empowered by the manmade vehicle prevalent in their times. blacksmith’s magical capabilities: In these lines, Heaney showcases the strangely divided sense of identity − as both British and Irish − that has since continued to haunt Northern Ireland, but also acknowledges the broadening of his “den-life” and the chance to reach out of the “emotionally and 155 intellectually proofed” (OG 447) acoustic space of his household and into the wideness of the world.148 It was not until they had moved to ‘The Wood’ farm in the 1950s that Heaney’s family bought their own first car. Yet the fact that Heaney’s Mossbawn house was near the main road exposed him to the sound of cars, buses, lorries and vans early on. The motor car opened access to areas without railway and made medium-distance travel more convenient and affordable for individuals (Biagini and Daly 206). Heaney bought his own first car early in the 1960s after starting his career as a lecturer. The liberty and freedom of movement that the motor car introduced to society was noticeable for him (SS 4). Recollecting his days as a lecturer at St Joseph’s, he writes that he bought his first car so that he “could drive round the country to inspect students on teaching practice” (SS 45). Since the 1960s, the internal combustion engine has turned into a fundamental keynote of Irish roadways. Not only cars, but also motorcycles, trucks, tractors, power lawnmowers, generators, and other power tools were operated with internal combustion engines, turning the sound of the engine into a fundamental soundmark of contemporary civilisation in both urban and rural soundscapes. The internal combustion engine is light and easy to operate transferring a sense of authority to the individual in charge. The car facilitates freedom of movement and the movement, in turn, invites the driver into “dialogue” with various soundscapes (Järviluoma et. al., 31). Moreover, driving facilitates access to unknown worlds, new sounds, and fresh inspirations. Not surprisingly, as McGuckian writes, during the twentieth century, the car became a potent symbol of power and freedom for not only several Irish poets but indeed for contemporary poetry in general (70).149 Heaney was a poet with a passion for cars. This 156 attitude towards the modern-day vehicle is reflected in Paul Durcan’s poem ‘A Spin in the Rain with Seamus Heaney’. ( ) p ( ) p , y the gear and the breake in an excursion to Glenshane Pass (ST 42-44) and in ‘On the Road’ (SI 119), he even enters a state of trance through such car noises as the “reeling” wheels and the “steady” sound of the engine. 150 Heaney refers to the car and driving positively as an empowering experience in ‘Oyster’ (FW 3), ‘A Pillowed Head’ (ST 38-39) and ‘Postscript’ (SL 70). In ‘A Retrospect’, he affectionately recalls various sounds related to the gear and the breake in an excursion to Glenshane Pass (ST 42-44) and in ‘On the Road’ (SI 119), he even enters a state of trance through such car noises as the “reeling” wheels and the “steady” sound of the engine. e s s e o ce oug suc c o ses s e ee g w ee s d e s e dy sou d o e e g e 1 The poem calls to mind one of Yeats’s best-known poems, ‘The Lake Isle of Innisfree’, particularly the 151 The poem calls to mind one of Yeats’s best-known poems, ‘The Lake Isle of Innisfree’, particularly “I will arise and go now, and go to Innisfree” (10). 150 Heaney refers to the car and driving positively as an empowering experience in ‘Oyster’ (FW 3), ‘A Pillowed Head’ (ST 38-39) and ‘Postscript’ (SL 70). In ‘A Retrospect’, he affectionately recalls various sounds related to the gear and the breake in an excursion to Glenshane Pass (ST 42 44) and in ‘On the Road’ (SI 119) he even The poem calls to mind one of Yeats s best known poems, The Lake Isle of Innisfree , particularly the lines: “I will arise and go now, and go to Innisfree” (10). 150 Heaney refers to the car and driving positively as an empowering experience in ‘Oyster’ (FW 3), ‘A Pillowed Head’ (ST 38-39) and ‘Postscript’ (SL 70). In ‘A Retrospect’, he affectionately recalls various sounds related to 150 Heaney refers to the car and driving positively as an empowering experience in ‘Oyster’ (FW 3), ‘A Head’ (ST 38-39) and ‘Postscript’ (SL 70) In ‘A Retrospect’ he affectionately recalls various sounds r blacksmith’s magical capabilities: Yet they vary in the fact that the poet is in charge of the engine power roaring along the road, while the Tollund man is being carried perhaps against his will, and the poet is aware of this: 157 Something of his sad freedom As he rode the tumbril Should come to me, driving, […] Out there in Jutland […]. (WO 37) The last three poems of Wintering Out reveal the poet’s desire to make a change in his life. ‘Dawn’ already hints at a turning point in its title. Heaney is driving through a town with a group of academics. The progress seems tedious and slow-paced: “We went at five miles an hour” and the poet feels the need to get away by himself (WO 65). Heaney’s desire to get away is revealed in the next poem, ‘Travel’, to be “a quest for psychic release, an effort to transcend routine responsibility, to discover the means to cope with violence and aftermath” (Andrews 78). In ‘Westering’, he reveals the mixed feelings of freedom and sadness while driving. The poem makes a direct reference to California in its epigraph – “In California” (WO 67) − as if it intends to prepare the reader for the concluding episode of what would be a diary of an “inner émigré” embracing the West-land (Longley 63). The first lines, however, depict the “homesick” poet (Parker 113) sitting under the “Official Map of the Moon” (WO 67). A glance over the “frogskin” colour of the lunar surface is enough for his imagination to return to the familiar shores of Donegal. Heaney reminisces about his departure day as he was driving westward to prepare for his journey across the ocean. His excitement at the thought of starting a new phase of his life in California coincides with the expectation of silence and stillness on Good Friday:152 […] Good Friday We had started out Past shopblinds drawn on the afternoon. Cars stilled outside still churches, Bikes tilting to a wall; We drove by, A dwindling interruption […]. (WO 67) […] Good Friday We had started out Past shopblinds drawn on the afternoon. Cars stilled outside still churches, Bikes tilting to a wall; We drove by, A dwindling interruption […]. (WO 67) We drove by, A dwindling interruption […]. (WO 67) 152 The poem makes an allusion to Donne’s ‘Good Friday, 1613. 152 The poem makes an allusion to Donne’s ‘Good Friday, 1613. Riding Westward’, which was composed while the poet was travelling from Warwickshire to Wales, on Good Friday. In the poem Donne expresses “his doubts, his prayers for God’s acceptance, and his hesitant step towards priesthood” (Barfoot and Todd 46-47). 153 In ‘Casualty’, similarly, Heaney associates the “[p]urring of the hearse” and the slow muffing engines to death and funeral (FW 17). The poem refers to the casualties of Bloody Sunday, in Londonderry, in 1972, during which 13 civil rights protesters were killed by the British army. ‘The Toome Road’ opens with an intrusive presence of army cars near his house. For him, the “warbling” of the armoured cars is “the bringer of bad news”, signalling the presence of the British army in Ulster (FW 7). blacksmith’s magical capabilities: Riding Westward’, which was composed while the poet was travelling from Warwickshire to Wales, on Good Friday. In the poem Donne expresses “his doubts, his prayers for God’s acceptance, and his hesitant step towards priesthood” (Barfoot and Todd 46-47). 158 Heaney’s reference to the dominating silence of the day draws on the major issues of sectarian burdens and the mainstream Irish submission to the dominion of religion: Lifeless streets and silenced believers (“shopblinds drawn”, “cars stilled”, “bikes tilted to a wall”, “still churches” with their “congregations bent / To the studded crucifix”) (WO 67-68). He bitterly remembers that his driving was perceived as an unwanted distraction to the devotees, while the vigorous sound of the church bells was the “sacred noise” of religion (Schafer 51). In Heaney’s opinion, the atrocities and estrangement in the name of Christ won’t lead to any healing or redemption. Living in Northern Ireland, the sense of freedom, power, and adventure-seeking fuelled by the roaring of the gas pedal is overshadowed by the spectres of politics, violence, and death.153 In ‘Funeral Rites’ the recurring history of sectarian violence in Ireland is echoed through its “customary” noises, therefore, the presence of the engines. The poem juxtaposes three scenarios from the rites of traditional Irish Catholic family funerals, the burial of individuals killed during the Troubles and the burials of a Viking hero. In the first section, Heaney describes and embraces the sombre silence of the traditional Irish Catholic ceremonies by highlighting the sense of courtesy and admiration he has for the deceased (“I shouldered a kind of manhood”, “I knelt courteously, / admiring it all”) and the sense of orderliness, compliance and spirituality (“their dough-white hands / shackled in rosary beads”, “the wrists / obediently sloped”) that is dominating the ceremony (N 6-7). In the final section too, he speaks about “chanting”, “honour” and “lights” (N 9). By contrast, in the second section, he resonates the rumbling and roaring of the cars every time the “news […] / of each neighbourly murder” is made public (N 7): Out of side-streets and bye-roads 159 purring family cars nose into line, the whole country tunes to the muffled drumming p y y y g , [ ] g y p the eructation of Orange drums” (N 53). In ‘Orange Drums, Tyrone, 1966’, he likens the drums to “giant tumours” (N 63). In ‘July’, Heaney concludes: “and so my ear was winnowed annually” suggesting how his ears have become attuned to the sectarian nature of drumming (S 15). In ‘Ocean’s Love to Ireland’, he links iambic pentameter of English poetry with drumming (N 40-41). y , g y, p , Scotland and England to commemorate the triumph of Protestant king William of Orange over King James II, the last English Catholic monarch, in the 1690 battle of the Boyne. The drumming resonates through Heaney’s poetry. In ‘Whatever You Say Say Nothing’, he notes that “[l]ast night you didn’t need a stethoscope / To hear 154 Each year, since the late eighteenth century, the Protestant parade is held in the streets of Northern Ireland, S l d d E l d h i h f P ki Willi f O Ki J II 154 Each year, since the late eighteenth century, the Protestant parade is held in the streets of Northern Ireland, Scotland and England to commemorate the triumph of Protestant king William of Orange over King James II, of ten thousand engines. (N 7-8) The poem makes a reference to the sectarian violence that dominated Northern Irish politics during the mid-twentieth century. Upon witnessing the disorder of the mourners (“purring family cars” and “somnambulant women, / [who are] left behind”), Heaney longs for the “customary rhythms” and “temperate footsteps” that coordinate the feelings of the mourners in traditional Catholic funerals. He criticises the normality and prevalence of the “neighbourly” killings, which he associates with broadband noises of the car engines. Moreover, the roaring car engines echo the boisterous drumming of Orangemen, which have been showcasing their triumph and filling the air with the ding of sectarianism since the late eighteenth century. Every year, as Paul Moore notes, more that the change in climate conditions, it is the drumming period that defines the construction of the seasons in Northern Ireland: While time might be measured by the natural evolution of spring, summer and autumn, cultural memory is measured by the start and finish of the ‘marching season’, a source of pride and cultural reinforcement for one community, a period of fear and perceived triumphalism for the other. A central feature of this season is the insistent vibration of the drum. (254) For Heaney, Orangemen’s drumming is the acoustic representation of sectarianism and violence. He wishes for the end of what he perceives as an unwanted noise. He wishes for silence, serenity and peace in the soundscape of Ireland. The sound of drumming is evoked in ‘Land’ as well. Heaney links “drumming” as a symbol of sectarianism with its lasting social and cultural consequences (WO 19). As 160 explained earlier in this section, in the poem, he reveals his decision to leave his birthplace and make changes in his lifestyle. Residing on the foreign “ground”, in the third section of the poem, the poet has no other way to learn about what is happening in his homeland than to keep an ear to the ground so he can detect the vibrations made by every movement. To his disappointment, however, there is no sign of improvement. 154 Each year, since the late eighteenth century, the Protestant parade is held in the streets of Northern Ireland, Scotland and England to commemorate the triumph of Protestant king William of Orange over King James II, the last English Catholic monarch, in the 1690 battle of the Boyne. The drumming resonates through Heaney’s poetry. In ‘Whatever You Say Say Nothing’, he notes that “[l]ast night you didn’t need a stethoscope / To hear of ten thousand engines. (N 7-8) The drumming of the Orangemen continues to reverberate in “this loop of silence” (WO 12):154 if I lie with my ear in this loop of silence if I lie with my ear in this loop of silence long enough, thigh-bone long enough, thigh-bone and shoulder against the phantom ground, I expect to pick up a small drumming (W 12) I expect to pick up a small drumming (W 12) In the final lines of the poem, Heaney refers to yet another familiar sound of violence: the blast of bombs going off. Heaney had experienced the groaning of the American bombers as a young boy: “The American bombers groan towards the aerodrome at Toomebridge, the American troops manoeuvre in the fields along the road […] (PO 17). In the frontispiece of Wintering Out, a poem dedicated to David Hammond and Michael Longley, the poet’s Protestant friends, the sight of a bombing site becomes the symbol of sectarian violence and the depth and immediacy of its impact: “A bomb had left a crater of fresh clay / In the roadside” (N 55). High-intensity sounds, such as bombing or supersonic aircrafts, create intense and sudden release of energy wavesthat can actually inflict damage to the surrounding and deeply impact listeners (Schafer 86). In the third section of the poem ‘Land’, his 161 imagination and memories enable him to identify the signs of a deteriorating Ulster and the scale and depth of this damage on his mind and life: imagination and memories enable him to identify the signs of a deteriorating Ulster and the scale and depth of this damage on his mind and life: scale and depth of this damage on his mind and life: and must not be surprised in bursting air to find myself snared, swinging The sombre footsteps of people in a cortege, huffing of many engines, sirens, and bombings are not the only sounds that Heaney associates with funeral, death and therefore the long history of violence in Ireland. The rhythmic throbbing of the helicopters was one of the most ordinary aspects of the everyday life of the poet and his fellow men until the mid-1980s, turning it into a recurring motif in many audio-visual works about Northern Ireland.155 ‘The Lift’ captures the solemn silence of the funeral of Heaney’s aunt Mary, who passed away in the mid-1970s. y g p y Northern Ireland, particularly in Newry and Derry (FW 6). The sound of airplane engines is also evoked in Heaney’s poetry. By contrast to the helicopter, the airplane is evoked to represent a sense of advancement, progress and the unknown. This association is raised in ‘Honeymoon Flight’, where “[t]he engine noises” stimulate mixed feelings of progression and unpredictability (DN 36). In ‘Dream of Solstice’, however, the sounds of the flight’s engine, as it prepares for landing, awakens the dreaming poet. In ‘From the Republic of 155 In Orla Walsh’s The Visit (1992), Neil Jordan’s The Crying Game (1992) and Terry George’s Some Mother’s Son (1997), for instance, helicopters are the symbol of terror and aerial surveillance. ( ), , p y 156 In ‘At the Water’s Edge’, Heaney recalls “listen[ing] to the thick rotations / Of an army helicopter p in the silence of Horse Island and how they were “shadowing” the People's Democracy March of Northern Ireland particularly in Newry and Derry (FW 6) The sound of airplane engines is also e 155 In Orla Walsh’s The Visit (1992), Neil Jordan’s The Crying Game (1992) and Terry George’s Some Mother’s Son (1997), for instance, helicopters are the symbol of terror and aerial surveillance. 156 In ‘At the Water’s Edge’, Heaney recalls “listen[ing] to the thick rotations / Of an army helicopter patrolling” in the silence of Horse Island and how they were “shadowing” the People's Democracy March of 1969, in Northern Ireland, particularly in Newry and Derry (FW 6). The sound of airplane engines is also evoked in Heaney’s poetry. By contrast to the helicopter, the airplane is evoked to represent a sense of advancement, progress and the unknown. This association is raised in ‘Honeymoon Flight’, where “[t]he engine noises” stimulate mixed feelings of progression and unpredictability (DN 36). In ‘Dream of Solstice’, however, the sounds of the flight’s engine, as it prepares for landing, awakens the dreaming poet. In ‘From the Republic of 5 In Orla Walsh’s The Visit (1992), Neil Jordan’s The Crying Game (1992) and Terry George’s Some Mo on (1997) for instance helicopters are the symbol of terror and aerial surveillance on ( 997), o sta ce, e copte s a e t e sy bo o te o a d ae a su ve a ce. 6 In ‘At the Water’s Edge’, Heaney recalls “listen[ing] to the thick rotations / Of an army helicopter patro n the silence of Horse Island and how they were “shadowing” the People's Democracy March of 19 In At the Water s Edge , Heaney recalls listen[ing] to the thick rotations / Of an army helicopter patro the silence of Horse Island and how they were “shadowing” the People's Democracy March of 19 orthern Ireland, particularly in Newry and Derry (FW 6). The sound of airplane engines is also evok h h li h i l i k d f d of ten thousand engines. (N 7-8) Yet, as Heaney documents, the silence of this private cortege is disrupted by “the throttle and articulated whops / Of a helicopter crossing” (DC 42). During the Troubles helicopters were plentifully available to the British Army, “to patrol, survey and photograph, as well as to drop off and evacuate soldiers” (Dalsimer and Kreilkamp 226). The presence of the helicopters hovering in the “open air” above the heads of the mourners, in ‘The Lift’, signals the ceaseless and intensive surveillance of the British army during the Troubles. Heaney juxtaposes the aerial throbbing of the helicopter with sounds of their “own footsteps” (DC 42) which become audible only after the helicopter has passed and silence is restored, to indicate their “imperialistic” presence in the private acoustic space (Schafer 77). The image exemplifies the smothering situation of Ireland.156 162 In addition to the rich and varied references to the means of transport, Heaney pays great attention to church bells and clocks. In the remainder of this section, I attend to the sounds of time and of bells, both of which feature frequently in Heaney’s poetry, stretching from being a signal sound that transmits a specific message to the listeners, to being a sound symbol, stirring feelings and reactions beyond the limits of time and space. One of the most salient signal sounds in a Christian community is the church bell (see 1.2). Signal sounds are sounds that must be listened to as they carry a specific meaning (see 1.4.2). Church bells are signal sounds as, in their various ways of ringing, they mediate various forms of social messages. Church bells are community sounds as they define the acoustic space of the community and the territory of each parish. They are also centripetal sounds as they can “unify and regulate the community”. Church bells were widespread in Europe by the eighth century. With the spread of Christianity, church bells were introduced to new regions “acoustically demarking the civilisation of the parish” from what, in comparison, would be interpreted as the chaos of the wilderness before their presence and beyond their earshot (Schafer 53-56). The idea that church bells define the acoustic territory of the Christian community and control the rhythm of life underpins the Irish bardic work Buile Shuibhne and Heaney’s translation, Sweeney Astray. Conscience’, when the plane’s engine stops, the poet is able to hear the slightest sounds: “a curlew high above the runway” (HL 14). − It is Ronan Finn, the son of Bearach, they said. He is marking out a church in your territory and what you hear is the ringing of his bell. (SA 3) of ten thousand engines. (N 7-8) The narrative focuses on a time of religious change in Ireland and revolves around the tensions between Sweeney, the pagan king of Dal-Arie, in what is now Ulster, and St. Ronan Finn, a Christian missionary, who is attempting to establish a church in Sweeney’s kingdom. In this extract from the opening of the work, Sweeney becomes aware of Ronan’s presence through the sound of his bells: One time when Sweeney was king of Dal-Arie, Ronan was there marking out a church called Killaney. Sweeney was in a place where he heard the clink of Ronan’s bell as he was marking out the site, so he asked his people what the sound was. − It is Ronan Finn, the son of Bearach, they said. He is marking out a church in your territory and what you hear is the ringing of his bell. (SA 3) 163 Sweeney feels the threat, is infuriated and rushes to prevent the construction and banish the priest from his territory. In the first encounter, Sweeney throws Ronan’s psalter into a nearby lake and, then, in the battle of Magh Rath, he eventually pierces a hole in Ronan’s bell with his spear, which causes the priest to discharge a curse that transforms Sweeney into a bird- man. Sweeney’s dismay over the sound of the bell is stressed throughout the book. On one occasion, Ronan recalls how “Sweeney was […]. Screaming against my bell, he said he preferred the sound of birds in the trees and the rushing waters of the glens” (SA 5). Heaney resurrects Sweeney in a number of poems in Station Island. In the opening poem of the title sequence, Heaney is poised between the call of church bells urging him to join the pilgrimage and the dissenting voice of Sweeney that pressures him to break free from the orthodoxy of his Catholic upbringing. Church bells function as “an acoustic calendar”, celebrating cultural festivals and religious rituals with the members of the community and, in doing so, they regulate the rhythm of life in the parish (Schafer 53-55). The poem describes an early Donegal Sunday morning when the church bells of Station Island − a thousand-year- old site of pilgrimage in Lough Derg, more commonly known as St Patrick’s Purgatory − urge people to cease their daily chores and join the procession. of ten thousand engines. (N 7-8) As the faithful cease their daily routine, silence takes over the soundscape, but is almost immediately disturbed by the presence of the pagan figure from the pre-Christian period. Simon Sweeney is the head of a family of tinkers who camped in the neighbourhood of Heaney’s Co. Derry home, but in the poem, his memory blends with the figure of the anarchic pagan king Sweeney, “an old Sabbath-breaker”, who doesn’t abide by the aural implications of the church bell and encourages Heaney to “stay clear of all processions” − of religion, politics, and literary and cultural conformity: 164 as it started. Sunday, the silence breathed and could not settle back for a man had appeared at the side of the field with a bow-saw, held stiffly up like a lyre. (SI 61) Yet, just as the lines of the verse hurry forward, Heaney is swept along by the crowd of pilgrims onto the trail his upbringing has set him up for: “a drugged path // I was set upon” (SI 63). Yet, just as the lines of the verse hurry forward, Heaney is swept along by the crowd of pilgrims onto the trail his upbringing has set him up for: “a drugged path // I was set upon” (SI 63). A signal sound becomes an acoustic symbol “when it stirs in us emotions or thoughts beyond its mechanical sensations or signalling function, when it has a numinosity or reverberation that rings through the deeper recesses of the psyche” (Schafer 169). In ‘Mid- term Break’, ‘Midnight Anvil’ and ‘Poet to Blacksmith’ (DC 25) Heaney uses the symbolic association between church bells and spirituality.157 In ‘Mid-term Break’, he tells us that he “sat all morning in the college sick bay / counting bells knelling classes to a close” (DN 15). The “knelling” he is waiting for in the college is meant to signal the end of the class but, as Heaney reveals later in the poem, it also resonates with the sound of funeral bells. Heaney uses the social and religious connotations of bell sounds to prepare readers with the grieving atmosphere of his household and the death of his four-year-old brother, Christopher. , y g [ ] The grass is all a-tremble. // But only then. / Not every time any old bell // Rings” (HC 38). 7 In Station Island IX, the bell calls for a pilgrimage (SI 85). In ‘Parable Island’, the bell tower and the 157 In Station Island IX, the bell calls for a pilgrimage (SI 85). In ‘Parable Island’, the bell tower and the sound of the bell are directly associated with the Christianisation of Ireland (HL 11-13). In ‘Out of Shot’, he describes the sunny Sunday weather as “bell-clear” (DC 15). In ‘Chanson d’Aventure’, Heaney likens the word “[a]part” 157 In Station Island IX, the bell calls for a pilgrimage (SI 85). In ‘Parable Island’, the bell tower and the sound of the bell are directly associated with the Christianisation of Ireland (HL 11-13). In ‘Out of Shot’, he describes the sunny Sunday weather as “bell-clear” (DC 15). In ‘Chanson d’Aventure’, Heaney likens the word “[a]part” (HC 15) to the knelling of the bell, which similar to the poem ‘Mid-Term Break’, recalls the announcement of a of the bell are directly associated with the Christianisation of Ireland (HL 11-13). In ‘Out of Shot’, he the sunny Sunday weather as “bell-clear” (DC 15). In ‘Chanson d’Aventure’, Heaney likens the word 157 In Station Island IX, the bell calls for a pilgrimage (SI 85). In ‘Parable Island’, the bell tower and the sound of the bell are directly associated with the Christianisation of Ireland (HL 11-13). In ‘Out of Shot’, he describes the sunny Sunday weather as “bell-clear” (DC 15). In ‘Chanson d’Aventure’, Heaney likens the word “[a]part” (HC 15) to the knelling of the bell, which similar to the poem ‘Mid-Term Break’, recalls the announcement of a funeral by the church (DN 15). The poem was written after Heaney was hospitalised for a stroke. However, in ‘A Herbal’, he reminds us that not every time that bells ring is there a funeral: “[w]hen the funeral bell tolls / The grass is all a-tremble // But only then / Not every time any old bell // Rings” (HC 38) (HC 15) to the knelling of the bell, which similar to the poem ‘Mid-Term Break’, recalls the announcement of a funeral by the church (DN 15). The poem was written after Heaney was hospitalised for a stroke. However, in y y ( ) y [ HC 15) to the knelling of the bell, which similar to the poem ‘Mid-Term Break’, recalls the announcemen uneral by the church (DN 15). The poem was written after Heaney was hospitalised for a stroke. Howev A Herbal’ he reminds us that not every time that bells ring is there a funeral: “[w]hen the funeral bell funeral by the church (DN 15). The poem was written after Heaney was hospitalised for a stroke ‘A Herbal’, he reminds us that not every time that bells ring is there a funeral: “[w]hen the funer ) g p uneral by the church (DN 15). The poem was written after Heaney was hospitalised for a stroke. Howe A Herbal’, he reminds us that not every time that bells ring is there a funeral: “[w]hen the funeral bell the sunny Sunday weather as “bell-clear” (DC 15). In ‘Chanson d’Aventure’, Heaney likens the word “[a]part” (HC 15) to the knelling of the bell, which similar to the poem ‘Mid-Term Break’, recalls the announcement of a funeral by the church (DN 15). The poem was written after Heaney was hospitalised for a stroke. However, in ‘A Herbal’, he reminds us that not every time that bells ring is there a funeral: “[w]hen the funeral bell tolls / p gg p , y p region of France to fulfill a commitment for the Somerset Maugham Award in the previous year. 158 The poem suggests a parallel with the summer of 1969, when Heaney spent a week in the Bas- 158 The poem suggests a parallel with the summer of 1969, when Heaney spent a week in the Bas-Pyrénées region of France to fulfill a commitment for the Somerset Maugham Award in the previous year. 159 In ‘Mossbawn: Sunlight’, he recalls a scene when his aunt, Mary is baking in the kitchen. In the poem Heaney refers to “the tick of two clocks”, which while it can be read symbolically to mean both his mother and aunt, as the sources of order and care, can also refer to the abundance of “space” and, therefore, a calm and homely atmosphere in the family home. In ‘Milk Factory’, the concept of time (“round the clock”) is sued to refer to the routine of shift-workers and their destiny (HL 35). of ten thousand engines. (N 7-8) In ‘Poet to Blacksmith’, he draws an analogy between the resonating clang of a sharpened spade, as it hits the soil, to the chime of the bell (“sweet as a bell”) (DC 25), and in ‘Midnight Anvil’, he likens the sound of the blacksmith’s strikes to church bells to highlight the social and cultural 165 values of this disappearing keynote (DC 26). In ‘Summer Home’, from Wintering Out, it is Marie Devlin’s voice that “bells” in the poet’s ear. Although he does not mention the church bell directly, his reference to the offering of “wild cherry and rhododendron”, the flowers and blooms composed on the “May altar” and “a sweet chrism” evokes the atmosphere of a Christian liturgy. The unbearable summer heat (“the hot foreign night”), unpleasant accommodation (“a fouled nest incubating somewhere?”), marital complications (“The loosened flowers between us”) and the demands of domestic life (“My children weep out”) spoil the summer plans (“the summer gone sour”). The resonating and inescapable toll of the church bell makes it an apt metaphor for Marie’s soundless weeping (WO 47-49). The remorseful poet finds it impossible to escape the sense of guilt and responsibility: “I hear her small lost weeping / through the hall, that bells and hoarsens / on my name, my name” (WO 47). 158 It was during the fourteenth century that the mechanical clock was wedded to the inescapable chimes of church bells to regulate the rhythm of life with even more precision (Schafer 55). The historian, Spengler, believes that the chimes of countless clock towers, which echoed day and night over the auditory space of European cities, were perhaps “the most wonderful expression of which a historical world-feeling is capable” (8). The high- pitched metallic sound of some bells imposed a presence over the noises of everyday life, evoking a sphere of order and serenity. The clock-regulated bells had the great convenience of announcing the passing of time far and wide. In ‘Mid-term Break’, as I explained above, the “knelling” of the school bells suggests the solemn atmosphere of a church funeral. of ten thousand engines. (N 7-8) Even more importantly, it is the integration of time, through such phrases as “counting” and “at two o’clock”, to the sound of bells that recalls the passage of time, preparing the readers to 166 hear about Christopher’s call of destiny: “I sat all morning in the college sick bay, / Counting bells knelling classes to a close. / At two o’clock our neighbours drove me home” (DN 15). Unlike the previously silent means of measuring time − i.e. water clocks, sand clocks, candle clocks, time sticks and sundials − the mechanical clock measured “the passing of time audibly” (Schafer 55). Their tick-tock sounds fly forth even more quickly than bells’ notes, reminding the listeners that the passing of life is inexorable. It was the discovery of the mechanical clock that gave European soundscape “the dread symbol of the flow of time” (Spengler 8). In ‘A Call’ and ‘Linen Town’, the ticking of the clock and the strike of its pendulums signal the passage of time and evoke a sense of destiny and the transient nature of life.159 ‘A Call’ is paradigmatic. The poet recalls a time before his parents passed away. Heaney has left the family home to go to university, work and build his own family, but keeps in telephone contact with his parents. While waiting for his mother to go to the garden and ask his father to get the phone, Heaney is left to his thoughts and memories, using his imagination to re-create the household in his mind: So I saw him Down on his hands and knees beside the leek rig, Touching, inspecting, separating one Stalk from the other […]. (SL 53) Heaney’s attention is then drawn to a ticking echoing through the receiver: Then found myself listening to The amplified grave ticking of hall clocks Where the phone lay unattended in a calm Of mirror glass and sunstruck pendulums …. (SL 53) The recurring present participles (“touching”, “inspecting”, “separating”, “listening”, “ticking”) create a momentary sense of stasis in the poem, but the ticking of clock, almost 167 echoed through the internal rhyme, takes the poet’s imagination to the time when he lived in that house and his parents were young, and then back to the present. He waits on. g g y ( 161 The title of this section is inspired by a line from ‘The Wool Trade’ (WO 27). of ten thousand engines. (N 7-8) The inescapable sound of time carries him forward to the future and the thoughts of death run through his mind: “And found myself then thinking: if it were nowadays, / This is how Death would summon Everyman” (SL 53). Similarly, ‘Linen Town’ illustrates this association between time and the notions of life, fate and death. The clock fastens time onto a still image taking the reader to High Street, Belfast, in 1786, twelve years before the Irish Rebellion of 1798: “It’s twenty to four / By the public clock” (WO 28). The imagined ticking clock prepares the reader for an ominous announcement. The print unfreezes (“the civic print unfrozen”) moving twelve years into the future when the young Henry Joy McCracken, a member of the United Irishmen, is hanged and his voice lost in the bustle of a fast thriving town (“This lownecked belle and tricorned fop’s // Still flourish undisturbed / By the swinging tongue of his body”) (WO 28).160 The passing of time indicates what seems like a fated violence. As Tobin comments, “Heaney warns us of the irreversibility of history and cautions us against the historical recurrence of brutality” (84). Fast forward to the present, in the final stanza, the clock is sitting at twenty to four again. Heaney encourages his fellow compatriots to “Take a last turn / In the tang of possibility” and avoid yet another recurrence of violence in the history of Ireland (WO 28). 160 Henry Joy McCracken was born in High Street, Belfast, into one of the city’s most prominent Protestant families. His family led linen manufacture and founded the Belfast News Letter. He established the Society of the United Irishmen in 1795 and led the United Irishmen Rebellion in 1798. He was eventually defeated and hanged at Corn Market on land his grandfather had donated to the city (Hamilton 23). families. His family led linen manufacture and founded the Belfast News Letter. He established the United Irishmen in 1795 and led the United Irishmen Rebellion in 1798. He was eventual hanged at Corn Market on land his grandfather had donated to the city (Hamilton 23). 160 Henry Joy McCracken was born in High Street, Belfast, into one of the city’s most prominent Protestant 160 Henry Joy McCracken was born in High Street, Belfast, into one of the city’s most prominent Protestant families. His family led linen manufacture and founded the Belfast News Letter. He established the Society of h i d i h i d l d h i d i h b lli i ll d f d d 162 Dolar introduces the two categories of pre-linguistic and post-linguistic phenomena to refer to the non- structured, pre-cultural, and pre-signifying voices − such as a child’s babbling and scream or a person coughing and having hiccups − and the voices that exceed structure of speech and bring it to the forefront − such as singing − respectively (12-33). g g p y ( ) 163 See www.youtube.com/watch?v=zjV7APxLa8c (0:40-0:50). g g p y ( ) 163 See www.youtube.com/watch?v=zjV7APxLa8c (0:40-0:50). 3.4 The Gallery of Tongue: Vocal Sounds161 In any soundscape, human vocal sounds take their place in a larger structure of biophonic, geophonic and anthrophonic sounds. Earlier I discussed how we might be led to believe that our modes of perception are dominated by vision (see 1.2). Yet the musician and 168 audio-visual theorist, Chion, urges us to remember our “vococentrism” as one of the elements countering this assumption: “[In every sonic environment] there are not all the sounds including the human voice. There are voices, and then everything else. In other words, […], the presence of a human voice instantly sets up a hierarchy of perception [emphasis in original]” (5). According to the philosopher and psychoanalyst, Dolar, what distinguishes voice from other sounds is its inner relationship with meaning: “the voice is the instrument, the vehicle, the medium, and the meaning is the goal” (14-15). However, the function of voice exceeds its immediate and common use as merely the bearer of a linguistic meaning. In the following paragraphs, I explore the references to human vocal sounds − including pre- linguistic, linguistic and post-linguistic (23)162 − in Wintering Out, considering such questions as: Whose voices have been evoked in the poems? What are the socio-political norms governing them, and what is the poet’s attitude towards them? In order to answer these questions, I also include Heaney’s evocations of the voices of the dead. I close this section with the examination of Heaney’s perspective towards the dynamics between English and Gaelic in the Irish soundscape and society. In his Lannan Foundation poetry reading, Heaney comments on his depiction of the kitchen in ‘Mossbawn: Sunlight’ from the point of view of an infant in the cradle, defining infancy as a pre-speaking and unspeaking phase of life: “Infans, the Latin word, means unspeaking, and the infant is the unspeaking one”.163 Pre-linguistic sounds are primal in essence. From the first cry, a child uses sounds as a way of communicating its basic needs and a way of projecting its inner feelings into the surrounding world (B. R. Smith 13). Heaney refers to this basic yet highly communicative signal in ‘Summer Home’ and ‘Serenades’, and highlights it even more explicitly in ‘High Summer’, for instance. In 169 ‘Serenades’ the reference is rather implicit. 3.4 The Gallery of Tongue: Vocal Sounds161 Heaney invites his wife to join him and listen to the quiet ambience of rural Wicklow, but he recommends her to first “fill the bottles” with milk and put the babies “inside their cots” so they won’t cry from hunger or the need to sleep (WO 50). In ‘Summer Home’, he gives an account of a domestic tribulation − during a 1969 holiday in France, when the damp and hot nights and the foul odour of the rubbish and the maggots in the cottage are aggravating the family’s discontent: “My children weep out the hot foreign night” (WO 49). In ‘High Summer’, he takes the reader back to the same holiday atmosphere, in the Basque country, when “Christopher is teething and cries at night” (FW 41). The crying and screaming of a child represent “the pure process of enunciation before the infant is capable of any statement” (Dolar 28). In the examples above, the baby’s cries are not irritating for the father who is capable of interpreting them justly. The resonance of pre-linguistic sounds continues to signal the inner feelings and psychological states of human beings even after they learn to speak, but often displays more sophisticated social and cultural traits. For this reason, Smith categorises them within ‘non- verbal’ and not necessarily ‘pre-verbal’ sounds: “In crying, screaming, moaning, wailing, ululating the human voice emits sounds that are non-verbal if not pre-verbal − sounds that, according to Aristotle, ally the human voice with the voices of all breathing creatures in the soundscape of the world” (B. R. Smith 45). The sounds of sobbing, laughing, humming, grunting, evoked in Heaney’s poetry, are the non-verbal vocal signals executed to communicate inner feelings beyond words, sounds that the poet does not want us to disregard or forget. Heaney’s father’s quiet crying in ‘Mid-Term Break’ indicates his feelings of grief and despair over the loss of his young child: “in the porch I met my father crying” (DN 15). In ‘The Forge’, he points out the grunting sound that Barney makes while watching the traffic in the street and reminiscing about the forge’s thriving past in the days of horses and carts: “Then grunts and goes in […]” (DD 9). 165 See fawbie.info/wintering-out/wedding-day/. 164 The term laughter, of course, includes a wide array, from a mild smile to uncontrollable laughter. According t D l l ht i “ lt l t it f hi h l h ki d i bl ” (29) The term laughter, of course, includes a wide array, from a mild smile to uncontrollable laughter. A to Dolar, laughter is “a cultural trait of which only humankind is capable” (29) 165 164 The term laughter, of course, includes a wide array, from a mild smile to uncontrollable laughter. According to Dolar, laughter is “a cultural trait of which only humankind is capable” (29) g 165 See fawbie.info/wintering-out/wedding-day/. 3.4 The Gallery of Tongue: Vocal Sounds161 In ‘Keeping Going’, Heaney pays tribute to his 170 brother Hugh who expresses his confidence and sociable nature through his attitude and laughter: “[…] you wave at people, / You shout and laugh above the revs” (SL 12).164 Heaney’s alertness to these sounds is demonstrated particularly in the second part of Wintering Out, in ‘Wedding Day’, ‘Summer Home’ and ‘Winter’s Tale’. For most people, the wedding day is a prominent and life-changing occasion. Yet anyone who has been to a wedding has experienced the co-existence of cries and laughter on this special day. Heaney’s wedding day, as depicted in ‘Wedding Day’, is filled with mixed feelings of celebration, anxiety and sorrow for both the couple and their families. As the moment of departure approaches and the couple prepare for their honeymoon “the sap / Of mourning rises” (WO 45). The groom “is baffled by the emotionality of wedding guests surrounding the honeymoon taxi”165 and reflects on the reasons behind “all those tears” (WO 45). The wedding day is not only a transitional moment in the life of the couple but also their families. The feeling of separating from a daughter or a son, on a wedding day, underlies the happiness for their union and new start. Therefore, the moment of departure is symbolic in the way it dramatises the couple’s separation from their parental home. Heaney’s recognition of the tears, on the day of celebrating love, is a way of appreciating the nuances of the feelings involved and of foregrounding the parent-child relationship. Crying is a way of projecting oneself into the surrounding environment. In crying, humans seek communication. They seek a listener (B. R. Smith 13). In ‘Winter’s Tale’, Heaney lets us hear the faint “weeping” of the village girl to bring to light her inner distress and isolation (WO 52). The poem is based on the story of a young girl from Ulster but also alludes to Shakespeare’s play, The Winter’s Tale, and the figure of Perdita, who was affected by complex circumstances and psychological states. In Shakespeare’s play, Perdita − meaning ‘the lost one’, ‘abandoned’ or ‘desperate’, in Latin − is unaware of her royal lineage 171 as the daughter of the King of Sicilia. The young woman lives the life of a simple shepherd in Bohemia for more than sixteen years until, after a set of complex events, she is reunited with her parents. 166 OED offers the multiple meanings of the verb ‘attend’, of which ‘to listen to’ is amongst the firsts. 3.4 The Gallery of Tongue: Vocal Sounds161 Heaney’s anonymous character stands for the men and women who cannot fit into the society they live in and are in need of an empathetic listener to communicate their stories. Likewise, in ‘Summer Home’, Marie’s sobbing is a call for communication with her husband. Heaney’s juxtaposition of his own “unmusical drive” and “foul mouth” with Maria’s “small lost weeping” (WO 47-49) further evokes her feelings of “isolation”, “repression” and “disenchantment” (Corcoran 30). However, a profound listening enables the fair-minded poet to hear his name as the only one to blame. The polysemous word “attend” indicates his intention to turn both his ears and mind to his wife and attempt to heal the growing “wound” in their marital relationships (WO 47-49):166 I hear her small lost weeping th h th h ll th t b ll d h I hear her small lost weeping I hear her small lost weeping through the hall, that bells and hoarsens through the hall, that bells and hoarsens through the hall, that bells and hoarsens on my name, my name. on my name, my name. [ ] […] Attend. Anoint the wound. (WO 48-51). Attend. Anoint the wound. (WO 48-51). In ‘Shore Woman’, the woman’s cry is a projection of her fears and frustration. Earlier I highlighted how sounds can show the greed and insensitivity of the fisherman and his wife towards sea animals (see 3.2). The fisher’s wife is, on the one hand, alarmed by propelling porpoises; on the other, she is frustrated over her husband’s unsympathetic behaviour towards herself (see also 3.1 and 4.2). On the fisher’s boat, the woman’s peace is violated, but more importantly, her voice is ignored and her feelings are disregarded. Disappointed in her efforts to encourage her husband to retire, she lies helpless on the boat and screams. But her cry is not a “pleasure call”. It is what Schafer describes as an “alarm call” and a “distress call” (33): […] I lay and screamed Under splashed brine in an open rocking boat Feeling each dunt and slither through the timber, Sick at their huge pleasures in the water. (WO 55) 172 Yet ‘Shore Woman’ points out the woman’s voice in more ways than just one. The entirety of the poem is narrated in first person by the woman herself. 167 In her autobiographical book, All of Us There (1983), Polly Devlin refers to “the thin membrane […] between grief and joy in Irish celebrations” as illustrated on Marie and Seamus’s wedding day, and explains [H’s comment not clear] Heaney noticed the impact of Marie’s poignant rendering of the traditional ballad ‘Slieve Gallion Brae’ on the listeners (84). ( ) 168 According to B. R. Smith, the other three main models put forward to explain the relationship between language and society include: the long-standing view that linguistic and social structures are analogous (Saussure, Levi-Strauss, Derrida); social structures shape linguistic structures (Durkheim); linguistic structures shape social structures (Sapir, Whorf, Ong). (17-18). 3.4 The Gallery of Tongue: Vocal Sounds161 The epigraph arches back to a passage in Joyce’s Portrait of the Artist as a Young Man: “How different are the words ‘home’, ‘Christ’, ‘ale’, ‘master’, on his lips and on mine” (WO 27). The passage, uttered by Joyce’s character Stephen Dedalus, indicates the unresolved discomfort of the young Irish man towards the English language, as a major legacy of British colonisation and indicates how the connotations of a linguistic utterance can exceed its semantic meaning. The terms “ale” and “home” show how the use of English language can suddenly make even the most familiar concepts in Irish life feel foreign, and the term “master” directly refers to the bitterness with which the Irish recall colonisation. In the poem, Heaney depicts a similar situation. The unnamed speaker appears to be a Protestant merchant who speaks to the Catholic Heaney nostalgically recalling memories of: “[…] square-set men in tunics / Who plied soft names like Bruges // In their talk, […]” (WO 27). For the speaker, the name of the Flemish city, Bruges, which was a very significant commercial hub renowned for the thriving textile industry during the medieval period, recalls the memories of prosperity of the Protestant urban middle class who would go to Flanders for trade. For the Catholic poet of rural working-class background, instead, it is a reminder of the voices of men who raised the sheep and spun the wool and wove the cloth, and as Mark M. Smith notes, of the rural The poem ‘The Wool Trade’ laments the loss of the Irish wool industry and its associated sounds after the restrictions placed by English commercial policies on the Irish wool industry during the 17th century.169 But the poem is in fact constructed on the memory of the vanished trade as it resonates richly and warmly from the mouth of the unnamed Irish interlocutor: “‘The wool trade’ − the phrase / Rambled warm as a fleece // Out of his hoard” (WO 27). Listening to the man’s voice opens up much knowledge about the long gone keynotes of Irish prosperity and the soundmarks of national identity − “to shear, to bale and bleach and card” − but it also reveals the social divide it generated. 169 See celt.ucc.ie/published/E900040/text007.html. 3.4 The Gallery of Tongue: Vocal Sounds161 Similarly, in ‘Maighdean Mara’, despite her isolation and suffering on earth, the mermaid can voice her feelings through ballad. Tired of judgemental voices and the gazes of other village women, exacerbated by greedy men luring her to their land lives, and unable to retrieve her garment, the mermaid has no other choice but to relive the “Patterns of home” through her voice. Her singing is an act of tuning her feelings with the rhythm of tidesongs and projecting her inner world into the surrounding world: “She had no choice − conjured / Patterns of home and drained / The tidesong from her voice” (WO 56-57). Likewise, in ‘Wedding Day’, the bride is the only individual who has the power to bear the strain and restore balance between “grief and joy” during the ceremony performing a solo ballad that according to her sister Polly Delvin, “shivered the day” (84).167 In these poems, Heaney gives women not only a voice but also the medium of expressing and projecting it, one that emerges from their inner feelings and is projected into the outside world. References to linguistic utterances in Heaney’s poetry can be viewed in a similar manner as pre-linguistic and post-linguistic vocal sounds. From an ecological point of view, there is a complicated and mutually dynamic relationship between social structures and linguistic structures (B. R. Smith 17-18).168 Here, I explore how the way in which we interpret linguistic utterances in Wintering Out depends on the relationship between language and its social, political, academic, natural and cultural environment as a set of protocols for perceiving, interpreting, shaping and issuing information. 173 The poem ‘The Wool Trade’ laments the loss of the Irish wool industry and its associated sounds after the restrictions placed by English commercial policies on the Irish wool industry during the 17th century.169 But the poem is in fact constructed on the memory of the vanished trade as it resonates richly and warmly from the mouth of the unnamed Irish interlocutor: “‘The wool trade’ − the phrase / Rambled warm as a fleece // Out of his hoard” (WO 27). Listening to the man’s voice opens up much knowledge about the long gone keynotes of Irish prosperity and the soundmarks of national identity − “to shear, to bale and bleach and card” − but it also reveals the social divide it generated. 169 See celt.ucc.ie/published/E900040/text007.html. 3.4 The Gallery of Tongue: Vocal Sounds161 The epigraph arches back to a passage in Joyce’s Portrait of the Artist as a Young Man: “How different are the words ‘home’, ‘Christ’, ‘ale’, ‘master’, on his lips and on mine” (WO 27). The passage, uttered by Joyce’s character Stephen Dedalus, indicates the unresolved discomfort of the young Irish man towards the English language, as a major legacy of British colonisation and indicates how the connotations of a linguistic utterance can exceed its semantic meaning. The terms “ale” and “home” show how the use of English language can suddenly make even the most familiar concepts in Irish life feel foreign, and the term “master” directly refers to the bitterness with which the Irish recall colonisation. In the poem, Heaney depicts a similar situation. The unnamed speaker appears to be a Protestant merchant who speaks to the Catholic Heaney nostalgically recalling memories of: “[…] square-set men in tunics / Who plied soft names like Bruges // In their talk, […]” (WO 27). For the speaker, the name of the Flemish city, Bruges, which was a very significant commercial hub renowned for the thriving textile industry during the medieval period, recalls the memories of prosperity of the Protestant urban middle class who would go to Flanders for trade. For the Catholic poet of rural working-class background, instead, it is a reminder of the voices of men who raised the sheep and spun the wool and wove the cloth, and as Mark M. Smith notes, of the rural 174 hamlets “locked in stagnation and marginality” after the British embargo on the Irish wool industry left the accumulated resources in the hands of the Protestant middle class (62). Moreover, the speaker’s speech, articulated in the language of the oppressor reminds the poet of the disappearance of the native language of those hamlets: O all the hamlets where Hills and flocks and streams conspired To a language of waterwheels, A lost syntax of looms and spindles, How they hang Fading, in the gallery of the tongue! (WO 27) Heaney indicates that the language of the oppressor can still be heard today, but the native tongue of the Irish is disappearing from memory in their native land. It is this absence that he wants his readers to remember and listen to again (see also 4.4). 171 The relationship between voice, existence and presence has been theorised by philosophers, anthropologist, and phenomenologists. See Ong’s The Presence of the Word (1967), for the relationship between voice and presence in religious studies (114); Derrida’s Voice and Phenomenon (1967), for arguments against the deep- rooted Western philosophy of voice as present-in-itself (56-67); and Ihde’s Listening and Voice (2007), for an ecological perspective (67-68). 170 The examples of these poems are scattered in his poetry: ‘A Postcard from North Antrim’ was written in memory of an Ulster writer, Sean Armstrong, killed during the Troubles. Heaney writes: “It [his voice] was independent, rattling, non-transcendent / Ulster-old-decency” (FW 12). ‘In Memoriam Sean O’Riada’ pays tribute to the Irish composer who died in 1971 transcribing a few lines of their dialogue: “‘How do you work? / Sometimes I just lie out / like ballast in the bottom of the boat / listening to the cuckoo’” (FW 23). In ‘The violence (N 17 18). In Ocean s Love to Ireland , Heaney evokes the voice of the girl raped by S Raleigh during the Elizabethan conquer of Ireland (N 40-41). j g ( ) Digging Skeleton’ – translation of ‘After Baudelaire’, the skeletons come to life to talk about brutalities and violence (N 17-18). In ‘Ocean’s Love to Ireland’, Heaney evokes the voice of the girl raped by Sir Walter p g g y Sometimes I just lie out / like ballast in the bottom of the boat / listening to the cuckoo’” (FW 23). In ‘The Digging Skeleton’ – translation of ‘After Baudelaire’, the skeletons come to life to talk about brutalities and violence (N 17-18). In ‘Ocean’s Love to Ireland’, Heaney evokes the voice of the girl raped by Sir Walter Raleigh during the Elizabethan conquer of Ireland (N 40-41). 3.4 The Gallery of Tongue: Vocal Sounds161 Among the other voices Heaney listens to and brings up in his poetry are the voices of the dead.170 In such poems, Heaney uses the existential dimension of listening and the direct relationship between sound and sound-producing source to evoke the presence of the dead and ensure their continuity in his poetry (see 1.4.1).171 Just like any other sound, vocal sounds convey information about their sources − e.g. the vocal organs involved, how certain phonemes or words are pronounced, speaker’s attitude, accent, force, loudness, pitch, tone, intonation, timbre as well as other associated elements, such as the time and place in which they occur (B. R. Smith 3-10). For Heaney, evoking the voices of the characters is a way of 175 recalling their characteristic features and, as such, evoking their presence in his mind and poetry. ‘A Drink of Water’ is an example of how the speaker brings back his memories of an old woman who would come every morning to the well to fill her bucket with water. Her acoustics and visuals are evoked in the elegy to recall his memories: “[h]er grey apron”, “[c]reak of her voice” and the “slow diminuendo as it filled” (FW 8). In ‘Casualty’, Heaney remembers a friend of his, Louis O'Neill, a Northern Irish fisherman and fellow, through his voice, which he recalls from a conversation in a pub: recalling their characteristic features and, as such, evoking their presence in his mind and poetry. ‘A Drink of Water’ is an example of how the speaker brings back his memories of an old woman who would come every morning to the well to fill her bucket with water. Her acoustics and visuals are evoked in the elegy to recall his memories: “[h]er grey apron”, “[c]reak of her voice” and the “slow diminuendo as it filled” (FW 8). In ‘Casualty’, Heaney remembers a friend of his, Louis O'Neill, a Northern Irish fisherman and fellow, through his voice, which he recalls from a conversation in a pub: ‘Now you’re supposed to be An educated man,’ I hear him say. ‘Puzzle me The right answer to that one’. (FW 16) ‘Now you’re supposed to be An educated man,’ I hear him say. ‘Puzzle me The right answer to that one’. 172 See www.theguardian.com/books/2001/jun/16/poetry.features. 172 See www.theguardian.com/books/2001/jun/16/poetry.features. 173 See www.theguardian.com/books/2001/jun/16/poetry.features. 173 See www.theguardian.com/books/2001/jun/16/poetry.features. 3.4 The Gallery of Tongue: Vocal Sounds161 (FW 16) O’Neill ignored the IRA curfew on the day of the funeral of the thirteen victims of 1972 Bloody Sunday to go for a drink and was killed when a bomb went off. Heaney’s personal memory turns into a collective memory by picturing the doomed life of the innocent people involved in the sectarian violence. The final and title poem of Electric Light is another example in which Heaney evokes and perpetuates the memory of his grandmother through her whispery and archetypal voice: She sat […], […] and whispered In a voice that at its loudest did nothing else But whisper. (EL 80) Although the poem does not explicitly point it out, in Stepping Stones Heaney reveals that the old woman of the poem is his grandmother: “[t]hat’s her, with her mangled thumb and grey overall and whispery voice and unzipped slippers” (SS 27). In a 2001 interview for the Guardian, he explains that “there are cues to show that she is ancient, archetypal and central to the family”.172 As Rankin Russell notes, an early clue is “not so much in her prophecies 176 […], but in her way of speaking, which was an early model for his regionally grounded poetry” (186): […], but in her way of speaking, which was an early model for his regionally grounded poetry” (186): […] ‘What ails you, child, What ails you, for God’s sake?’ Urgent, sibilant Ails, far off and old [his emphasis]. […] (EL 80) Heaney further commented on his grandmother’s use of “the strange and slightly literary word ‘ails’” and his own “sense of historical and literary England”, merging the voice of the sibylline old woman with the voices of literary tradition, the voices of his poetic ancestors.173 In ‘Gifts of Rain’ and ‘Bog Oak’ Heaney goes as far as to evoke the voices of his pre- colonial ancestors. Listening to their whisperings and conjuring up their presence is a way of ensuring the continuity not only of their memory but also their wisdom. In ‘Bog Oak’, Heaney revisits the archive of history stored in the peat bog. The imaginary visit brings him into touch with the acoustic world of his Irish ancestors from the pre-Christian time, the first farmers and the turf cutters, as well as the peasants of the Elizabethan period. 3.4 The Gallery of Tongue: Vocal Sounds161 The poet evokes their subjugated voice and unacknowledged “wisdom”: I might tarry with the moustached dead, the creel-fillers, or eavesdrop on their hopeless wisdom […]. (WO 4) Likewise, amidst Moyola’s familiar orchestration, in ‘Gifts of Rain’ the poet “cock[s] his ear / at an absence” (WO 14), evoking the guttural voices of the first land dwellers. Listening to their faint and ghostly “whispering” is, for him, a way of invoking their presence. Heaney listens to the “soft voices of the dead” to feel their presence “by the shore” (15). The absence of Gaelic, as a keynote of the Irish landscape, is a cultural wound and its remembrance an attempt towards a healing. Yet the poet’s intention is more than simply to honour his 177 deceased ancestors. ‘The Tollund Man’, ‘Bogland’, ‘A Sofa in the Forties’, ‘England's Difficulty’, and ‘Act of Union’ are some of the poems in which Heaney expresses his stance toward history and the role of literature in representing historical events, especially those that caused much horror, death and trauma in survivors. According to Horkheimer and Adorno, “only when the horror of annihilation is raised fully into consciousness are we placed in the proper relationship to the dead: that of unity with them, since we, like them, are victims of the same conditions and of the same disappointed hope” (178). For Heaney, Moyola’s constant roaring, the rotting crops and the muddy riverbed are the ominous signs of yet another flooding of hardship in the future of Ireland, one that could be similar to “the flood of colonisation” that washed away much of the native tradition (Corcoran 41). Heaney is concerned that history may repeat itself. As Tobin has suggested, he is concerned that the violence involved in sectarian conflicts may jeopardise particularly his own Catholic community (81). He feels the “need / for antediluvian lore” (WO 15). He listens to the voices of his pre-colonial Gaelic ancestors to learn about the stories of perseverance and kinship. As an Ulster poet and as a father, Heaney feels the responsibility to retrieve and pass on the “shared” wisdom to the future generations (14-15). In many of the instances listed above, in his giving space to the voices of others and in evoking the whispering of the dead in his poetry, Heaney also hints at the social dimension inherent in any oral-aural communication. Oral communication is the foundation of a society. 3.4 The Gallery of Tongue: Vocal Sounds161 The conversation between two or more individuals constructs an acoustical community in the auditory field: “When he speaks the interlocutor’s voice fills the auditory field. My bones reverberate to his [O]. In my reply, I interject myself in that auditory field. His bones reverberate to my voice. We create an acoustical community” (B. R. Smith 21). Oral communication keeps the body at the centre of attention separating it from the linguistic component. The understanding we have within any such communication could essentially be 178 a bodily experience, one beyond and above the linguistic meaning of the utterances, one that is grounded in both the speaker’s body and the listener’s body.174 Heaney subtly refers to it in the poem ‘M.’, when describing a deaf phonetician’s embodied perception of the soundwaves: “He could tell which diphthong and which vowel / By the bone vibrating to the sound” (SL 57). Here, of course, the phonetician is not able to hear the sound, but perceives the vibration of the speaker’s body fully by reaching out to his skull. Heaney’s emphasis on the social and embodied dimension of oral communication is even more evident in such poems as ‘Twice Shy’ in which there is less emphasis on the semantic meaning of the utterance. In the poem, Heaney celebrates the earliest stages of his relationship with Marie Devlin, when irrespective of the content, the sheer fact of having a conversation becomes a key bonding element in their companionship. The speaker describes an evening stroll along the quiet riverside with a young woman, whose good taste and sensibility attract his attention at the onset: “Her scarf à la Bardot, / In suede flats for the walk” (DN 31). He shows the initial thoughts behind this walk: “for air and friendly talk” (31). Yet soon the couple find it challenging to overcome the indecisive flux of emotions mounting in their hearts. However, since their virtuous nurturing has taught them “not to publish feeling” so warily, they decide not to vocalise their emotion: “preserved classic decorum, / deployed our talk with art” (31). In this example, speaking is still a bonding element, while in ‘Summer Home’ the absence of connection between the couple is demonstrated through their silence (See also 4.2). In ‘Mid-Term Break’, the social expression of sympathy requires both the words and the physical contact. 174 For more information regarding the proprioceptive effects of voice see Appelbaum’s Voice (1990). 3.4 The Gallery of Tongue: Vocal Sounds161 Heaney remembers the voices of the neighbour men who expressed their condolences for the passing of his younger brother, Christopher (see 3.3). The neighbours complete this social aspect by eliminating the physical distance and shaking the young boy’s 179 hands: “old men standing up to shake my hand // And tell me they were ‘sorry for my trouble’” (DN 15). A similar but more intricate situation is described in the final part of ‘The Other Side’. The poem looks back at the relationship between Heaney’s family and their Protestant neighbour, Johnny Junkin, in the 1940s (SS 130). Despite his occasional outbursts, described in the first part of the poem, the man also has a benign nature and inner bond with his Catholic neighbour, reflected in the final section of the poem. Hoping to restore the bond with his neighbour, the white-haired man approaches their household for a friendly chatter. His willingness to connect is reflected through the manner and tone as he “courteously” and “embarrassedly” lingers outside the Heaney household and remains silent until they have completed the evening prayers, and only then approaches to knock on the door (Vendler 81): “he might say, ‘I was dandering by / and says I, I might as well call’” (WO 25). The young Heaney too who welcomes the man, thinks of touching the man’s shoulder and having a conversation as the options he has for socialising and connecting with him: But now I stand behind him […] Should I slip away, I wonder, Should I slip away, I wonder, p y or go up and touch his shoulder and talk about the weather or go up and touch his shoulder and talk about the weather 176 In Stepping Stones, he says: Several of them appeared in Wintering Out: ‘Limbo’, ‘Serenades’, ‘Veteran’s Dream’, ‘Midnight’, ‘Navvy’, ‘Dawn’. But a lot more saw the light of day just once, […]. I don’t know how to explain the dam-burst. It began with a hangover and continued with late nights and free days – this was in the month of May: Queen’s University classes stopped around that time of year, [..]. It was a visitation, an onset, and as such, powerfully confirming. This, you felt, was ‘it’. You had been initiated into the order of the inspired. […], the experience itself was crucial. From that point on, I felt different in myself as a writer. (SS 147) 175 Smith comments that, especially after Descartes, western philosophy has granted “superior power” to speech and the ability to construct rational arguments using language (22). or the price of grass-seed? (WO 26) or the price of grass-seed? (WO 26) There is another type of speech − academic discourse − that according to Heaney does not create a sense of community. An example is the poem ‘Dawn’ where the poet describes his experience of a car trip with a group of academic friends. While in the examples above, both the speaker and the listener keep the body at the centre of attention, the academic discourse − often operating through books, articles, conferences as opposed to the discussion that might follow up − points towards something incorporeal. In this context, the emphasis is often more on what is considered to be the superior power of speaking than the humble task 180 of listening and connecting (B. R. Smith 22).175 Heaney shows his frustration over the lack of embodied communication among some of the academics and the need for listening to one another and the surrounding world in ‘Dawn’, one of the final poems of Wintering Out. ‘Dawn’ emerged from a week of creative activity in May 1969. Heaney had been involved in the everyday routines of academic life as a lecturer in Queen’s and the end of the classes provided an opportunity for him to pause and reflect over his academic career. He explains how crucial this experience was for him, perhaps an initiation to a new intellectual phase that eventually urged him to leave the prissiness of Queen’s behind and begin his life as a full-time writer in Wicklow. “From that point on”, he writes, “I felt different in myself as a writer” (SS 147).176 The title of the poem too refers to a turning point, a moment of epiphany introduced by the sunlight: “Somebody lets up a blind” (WO 65). Heaney has joined a group of scholars for a car trip in a quiet Mediterranean city. The scholars have been engaged in heated discussions throughout the trip. For the poet, however, it has proved to be a laborious marathon and an irritating companionship: We went at five miles an hour. We went at five miles an hour. A tut-tutting colloquy Was in session, scholars Arguing through until morning A tut-tutting colloquy Was in session, scholars Arguing through until morning , p y , p p y and the ability to construct rational arguments using language (22). 176 I S i S h 177 See fawbie.info/wintering-out/dawn/. In a Pompeian silence. (WO 65) Despite being caught amidst the endless conversation of his companions, the surrounding feels “dead and empty” to the poet (Andrews 78). In this context, there is more emphasis on speaking than listening. As Fawbert puts it, “his companions are by nature censorious […] 181 everything is formality”.177 Heaney deliberately places the scholar’s “colloquy” in the informal setting of a private “car” and juxtaposes it with the pejorative qualifier “tut-tutting” to undercut their pretentiousness. For his academic friends, who have achieved a mastery of rhetoric and internalised its superior position, listening seems a humble and exacting effort. Gemma Corradi-Fiumara, who specialises in the philosophy of language and listening, writes: [A]n élitist power which has become internalised seems to ask rhetorically why there should be any need to listen, when one has not only achieved a mastery of language but also of the metalanguages whereby one can soar to the level of the relations that exist between discourse and reality, or among different types of discourse. (58-59) [A]n élitist power which has become internalised seems to ask rhetorically why there should be any need to listen, when one has not only achieved a mastery of language but also of the metalanguages whereby one can soar to the level of the relations that exist between discourse and reality, or among different types of discourse. (58-59) [A]n élitist power which has become internalised seems to ask rhetorically why there should be any need to listen, when one has not only achieved a mastery of language but also of the metalanguages whereby one can soar to the level of the relations that exist between discourse and reality, or among different types of discourse. (58-59) Heaney is skeptical about any resulting enlightenment from such scholarly arguments. Their academic discussion lacks embodied understanding and connection that the poet expects from oral communication. There isn’t even enough chance for him to listen to the surrounding environment while driving, which − as he showed us in the poem ‘Postscript’ − he has been fond of (see 3.1). Joining the scholarly mentality holds him back from the practice of listening, which for him is on the same level of significance as speaking. He feels the need to distance himself from the academic rituals based on a disembodied discourse and retrieve his ability to reconnect with the surrounding world through sounds. In a Pompeian silence. (WO 65) In fact, as soon as he does so, Heaney realises that even a Pompeii-like silence is full of sounds (see 4.4). The shells crunching beneath his feet enable the bewildered poet and thinker to once again realise his own position in relation to the surrounding reality: 182 In addition to highlighting the social, political, cultural and personal aspects of voice and speech, Heaney also pays close attention to the dynamics between English and Gaelic language in the Irish soundscape. From an ecological perspective, each culture has distinctive ways of understanding the world through various signals, keynotes and soundmarks from among geophonic, biophonic, and anthrophonic sounds. Among the other varied shared keynotes − such as rain, wind, local crafts and transportation − the sounds that give any community its coherence and demarcate its acoustic space are the distinctive phonemes of its language. The imperialistic presence of the English language in the Irish soundscape, for instance, can be interpreted as an attempt to assimilate the newly-colonised culture and to establish an “aural empire” in Ireland (B. R. Smith 288-289) (see 1.2). The topic of the linguistic dispossession of the Irish as a major legacy of colonialism has been examined from various angles in a succession of poems in Wintering Out, including ‘Traditions’, ‘A New Song’, ‘The Backward Look’, ‘The Last Mummer’, and ‘Gifts of Rain’, as well as in his place-name poems, ‘Anahorish’, ‘Broagh’ and ‘Toome’. In ‘Traditions’, Heaney declares his opposition to the imperialistic presence of the English language in Ireland through a “linguistic-sexual analogy” (Corcoran 40) wherein the muse-like “guttural” sounds of Gaelic are ravished and replaced by the brutal and rapacious “alliterative tradition” of Old English and early Middle English verse: “Our guttural muse / was bulled long ago / by the alliterative tradition […]” (WO 21). This linguistic imposition, as Heaney emphasises, initiated “long ago” at the same time as the confiscation of Irish lands by the English Crown from Henry VIII through to Elizabeth I and the arrival of the first settlers in the 16th and 17th centuries. B. R. Smith points out that “if each culture has its own distinctive ways of understanding the world through sound, the borders between cultures become, potentially at least, sites of noise, confusion, pandemonium” (289). In a Pompeian silence. (WO 65) While some may take delight in this merging of diverse linguistic sounds and open their ears to new 183 opportunities, the political factors at work often turn this encounter into a clash of cultures. Needless to say, the relationship between English and Gaelic was an uneasy one. As noted earlier, the writings of early modern English commentators show that for the ears of the imperial-minded English, Gaelic was classified as not only an “aural assault” but also a mode of communication that was only “one step removed from noise” (R. B. Smith 306) (see 1.2). This attitude can be inferred from the following passage on the title page on A Consolation for Our Grammar Schooles (1622), a pedagogical book by the Puritan minister and English schoolmaster, John Brinsley the Elder, which proposes educational methods for English colonies: More specially, for all those in the inferior sort, and all ruder countries and places namely, for Ireland, Wales, Virginia, with the Sommer Ilands, and for their more speedie attaining of our English tongue by the same labour, that all may speake one and the same Language. (1) More specially, for all those in the inferior sort, and all ruder countries and places namely, for Ireland, Wales, Virginia, with the Sommer Ilands, and for their more speedie attaining of our English tongue by the same labour, that all may speake one and the same Language. (1) During the 16th and 17th centuries the dominance of the English language in Ireland was still not well-established and the sounds of Gaelic could still be identified as one of the main keynotes of the Irish landscape. In contemporary Ireland, as Heaney remarks, “her uvula grows // vestigial, forgotten” (WO 21). Heaney maintains that today the sounds of Gaelic have disappeared from the landscape of Ireland and the mindscape of the Irish leading, in turn, to the atrophy of the related vocal organs (see 4.4). In a sarcastic tone, in the second section of the poem, Heaney reveals what the present soundscape of Ireland sounds like: During the 16th and 17th centuries the dominance of the English language in Ireland was still not well-established and the sounds of Gaelic could still be identified as one of the main keynotes of the Irish landscape. In contemporary Ireland, as Heaney remarks, “her uvula grows // vestigial, forgotten” (WO 21). 178 Here the reference gains impact from Heaney’s subsequent career, for in the 1983 pamphlet An Open Letter, he energetically, if humorously, objected to being included in an anthology of British verse, noting, among other points, “the name is not right” (4). Although he speaks and writes in the language of the coloniser, Heaney feels related to Gaelic Ireland. In a Pompeian silence. (WO 65) Heaney maintains that today the sounds of Gaelic have disappeared from the landscape of Ireland and the mindscape of the Irish leading, in turn, to the atrophy of the related vocal organs (see 4.4). In a sarcastic tone, in the second section of the poem, Heaney reveals what the present soundscape of Ireland sounds like: We are to be proud of our Elizabethan English: ‘varsity’, for example, is grass-roots stuff with us; we ‘deem’ or we ‘allow’ when we suppose and some cherished archaisms are correct Shakespearean. (WO 21) Ironically, “Elizabethan English” is now the language of the élite and educated Irish who are supposed to take pride and pleasure in “correct Shakespearean” terms as their archaic Irish 184 words.178 Such English words as “varsity” − shortened form of ‘university’ − have become part of the everyday conversation and the terms “deem” or “allow” have preserved their archaic meaning. Not to mention the foreign footprints on the name of his well-loved home Mossbawn: ‘moss-’ from the Scottish word for ‘bogland’ and ‘-bawn’ drawn from the Irish ‘bábhún’ meaning ‘walled enclosure’, a term used to identify the English fortified farm house (Corcoran 40). As Parker has observed, Ulster dialect retains ‘strikingly Elizabethan’ words and turns-of-phrase, but Heaney again shows his ambivalent attitude towards these ‘cherished archaisms’, his English inheritance. ‘Correct Shakespearean’ they may be, but they remind him of defeat, like the name of his birthplace, which brings together ‘Moss’ from the Scots of the Planters, and ‘bawn’, an English colonialist’s fortified farmstead. (98) Ulster dialect retains ‘strikingly Elizabethan’ words and turns-of-phrase, but Heaney again shows his ambivalent attitude towards these ‘cherished archaisms’, his English inheritance. ‘Correct Shakespearean’ they may be, but they remind him of defeat, like the name of his birthplace, which brings together ‘Moss’ from the Scots of the Planters, and ‘bawn’, an English colonialist’s fortified farmstead. (98) Heaney brings up the topic again in the opening epigraph of ‘The Wool Trade’, discussed earlier in this section. He makes a joint reference to the bitterness of the experience for him and for Stephen Dedalus, who declares: “The language in which we are speaking is his before it is mine. How different are the words ‘home’, ‘Christ’, ‘ale’, ‘master’, on his lips and on mine! I cannot speak or write these words without unrest of spirit. His language, so familiar and so foreign, will always be for me an acquired speech” (Joyce 221). As common as they may sound in the contemporary soundscape of Ireland, the terms “home”, “Christ”, “ale” and “master” stir up feelings of uneasiness in every Irish. Perhaps these mixed feelings for English words partly explain the reason why in ‘Fodder’, the first poem of the collection, Heaney replaces the title-word with “fother” – as spoken in the Ulster dialect: – in the first line: “Or, as we said, fother” (WO 3). The spoken word “fother” is a linguistic soundmark that relocates the poet, all the way from California, to the soundscape of Mossbawn and 185 brings him into touch with the voices of the Ulster farm men.179 It is also a disappearing keynote − hence the past tense, “said” (WO 3) − whose very ordinariness turns it into what Schafer calls a “sound souvenir” from the Irish past to the future reader (239). The disappearance of the Irish language is synonymous with the absence of cultural identity. In ‘Traditions’, Heaney remarks that today Ireland is just one of “the British isles” (WO 21). y 181 See celt.ucc.ie/published/E500000-001/. p g p p p ( ) 180 The poem ‘Traditions’ is dedicated to Tom Flanagan to whom Heaney owed much of his Ireland standpoint: standpoint: It was Tom’s poem because I lifted the conclusion of it from his book on the Irish novelists [i.e. The Irish Novelists 1800-1850]. The epigraph to that book juxtaposes MacMorris’s question in Henry V […] with Bloom’s answer in Ulysses […]. That seemed to cut through a lot of the identity crisis stuff that surrounded us in the early seventies, so I stole it for the end of the poem. (SS 143) 179 Similarly, in ‘The Guttural Muse’, the “guttural” voices of a crowd of youngsters, which Heaney hears from the window of his hotel room and describes as “thick and comforting”, are the acoustic reminders of the Irish soundscape and thus of healing properties for the Irish poet (FW 22). 179 Similarly, in ‘The Guttural Muse’, the “guttural” voices of a crowd of youngsters, which Heaney hears from the window of his hotel room and describes as “thick and comforting”, are the acoustic reminders of the Irish d d h f h li i f h I i h (FW 22) y g y g y the window of his hotel room and describes as “thick and comforting”, are the acoustic reminders of soundscape and thus of healing properties for the Irish poet (FW 22). […] with Bloom s answer in Ulysses […]. That seemed to cut through a lot of the identity that surrounded us in the early seventies, so I stole it for the end of the poem. (SS 143) 81 181 See celt.ucc.ie/published/E500000-001/. 179 Similarly, in ‘The Guttural Muse’, the “guttural” voices of a crowd of youngsters, which Heaney he (Corcoran 40). As Parker has observed, (38) Bloom speaks in English, as does everyone else in the novel, as do Heaney and many others However much one may feel the ignominy of speaking the conqueror’s language, English is not merely the mother tongue but the native tongue of modern Ireland, just as English literary tradition also forms, like it or not, a major part of the Irish literary landscape. (38) However much one may feel the ignominy of speaking the conqueror’s language, English is not merely the mother tongue but the native tongue of modern Ireland, just as English literary tradition also forms, like it or not, a major part of the Irish literary landscape. (38) However much one may feel the ignominy of speaking the conqueror’s language, English is not merely the mother tongue but the native tongue of modern Ireland, just as English literary tradition also forms, like it or not, a major part of the Irish literary landscape. (38) Bloom speaks in English, as does everyone else in the novel, as do Heaney and many others in modern Ireland. By having Bloom − a Jew of Hungarian origins − respond to the question, Heaney also challenges the notion of linguistic imperialism he opened up earlier, and as such offers “a criticism of the national stereotypes exhibited in the poem and therefore, implicitly, of all such stereotyping” (Corcoran 40). Like Bloom and like Joyce, Heaney is now owning the medium of the invader to project his personal and collective selfhood. In his place-name poems, ‘Broagh’ and ‘Anahorish’, Heaney treats the matter of linguistic dispossession and English as the medium of his poetry with dexterity unique to him (see 4.1). In his essay ‘Belfast’, he speaks of Broagh and Anahorish as “townlands that are forgotten Gaelic music in the throat, bruach and anach fhíor uisce, the riverbank and the place of clear water” (PO 36). The poems renew an ancient genre of Irish poetry called “dinnseanchas, poems and tales which relate the original meanings of place names and constitute a form of mythological etymology” (PO 131). In his place name poems ‘Broagh’ and ‘Anahorish’, the dialectical linguistic evolution from the native Irish words ‘anach fhíor uisce’ and ‘bruach’ is traced through their eponymous transliterations into a form of language that is shot through with both the native Irish and the colonising English language. (Corcoran 40). As Parker has observed, In the third and final section of the poem, he highlights the distinction between the imperialistic attitude of English speakers towards the sounds of Irish language by juxtaposing an excerpt from the work of the Elizabethan playwright Shakespeare, with one from the twentieth century Irish novelist, Joyce.180 Heaney’s choice of Shakespeare is crucial for re- creating a historical moment – the Elizabethan conquest and Plantation of Ulster – occurring during Shakespeare’s lifetime (Corcoran 40). To the ears of the English coloniser, Irish sounded “whiny” “brutish” and “uncivilised” (B. R. Smith 288-309). Earlier in 1596, in his infamous passage from A View of the Present State of Ireland, Spenser too had described the Irish victims of famine during the sixteenth century Desmond Rebellion “as anatomies of death”.181 In Henry V, Shakespeare fashions the character of MacMorris to render this tell- tale version of Irish as: aimless (“gallivanting / around the Globe”), having a whining tone (“whinged / to courtier”), uneducated (“as going very bare / of learning”), and uncivilised (“as wild hares”), all summed up in his utterance (“What ish my nation?”). Almost three centuries later, Joyce’s character, Leopold Bloom, replies to this question in Ulysses, in a tone and with a determination and consciousness that truly represents the Irish: And sensibly, though so much 186 Bloom’s reply challenges the stereotypical notions of Irishness established and echoed by the English commentators. Yet there are still more implications in his reply. Firstly, as Foster notes, both the English language and English literary tradition are undeniable elements of modern Ireland: Bloom’s reply challenges the stereotypical notions of Irishness established and echoed by the English commentators. Yet there are still more implications in his reply. Firstly, as Foster notes, both the English language and English literary tradition are undeniable elements of modern Ireland: Bloom’s reply challenges the stereotypical notions of Irishness established and echoed by the English commentators. Yet there are still more implications in his reply. Firstly, as Foster notes, both the English language and English literary tradition are undeniable elements of modern Ireland: modern Ireland: However much one may feel the ignominy of speaking the conqueror’s language, English is not merely the mother tongue but the native tongue of modern Ireland, just as English literary tradition also forms, like it or not, a major part of the Irish literary landscape. Festival, Heaney says: […] in fact the place was called Moss bann, and then it became, when you were writing down the address on your letters, it was Mossbawn. And it was only when, I think 1972, when I came back from Berkeley, […] that came to me as an idea for the divided nature of the world we lived in. […] The names of many [places], all over Northern Ireland, lead you straight back to pre-Norman, pre- foundation, pre-Tudor Ireland to an option of homeland in the first Celtic place or Gaelic place. So that was Moss bann, Mossbawn b-a-w-n led you to Plantation Ulster, where history begins in the 1620s and everything before that is not there, the land was theirs before the land was ours, so to speak. So, [..] Mossbawn and Moss bann, it seems to be that in the very name of the first house, the division, the problem was implicit ethimologically. (10:40 -12:33) (Corcoran 40). As Parker has observed, In these poems, the dialectical movement between different cultural significations is instantiated in 187 the poet’s desire to see the place name ‘Anahorish’ as translated into: “My ‘place of clear water”’ (WO 6), and in a similar movement, in the first line of ‘Broagh’ which translates the title as “Riverbank” (WO 17). The terms, however, are anglicisations and transliterations of the original spoken Irish words as opposed to translations; hence they cannot be validated by any English or Irish dictionary. What they signify is a dialectic, a movement between languages which is creative of a new sense of English with an Irish touch. Herman Rapaport makes this point in connection with Derrida, noting that, for Derrida, “language isn’t ever entirely masterable to perfection; therefore, one always senses that one is, however slightly, a stranger or foreigner to it” (31). Heaney, in these poems, dealing specifically with language, takes this position fully on board. Early modern colonial interventions in the toponomy of Ireland began as part of the strategy of extending British authority in Ireland “through plantation, military control and cultural repression” (Nash 461). As part of the mid-nineteenth century colonial project, Gaelic place-names were systematically collected and converted to English language versions. These names became the official versions in colonial cartography and administration in Ireland (Nash 462). Given the difficulty of the Gaelic phonetic system, British cartographers transliterated many such names into the British phonetic and graphological system. This approach to Irish place-names, as Catherine Nash suggests, is a paradigmatic case of colonial cultural erosion: Conventionally, the existence of a shared language has been one of the fundamental criteria for nationhood […]. Placenames thus combine the two elements – land and language – that have been central to the cultural projects of romantic nationalism, not least in the late nineteenth-century Irish cultural nationalism in which land and language were twined in an imaginative, political and economic geography of rural life and native landownership. The Gaelic language and the soil of Ireland were, it was argued, deep repositories of native spirit. To lose the Gaelic language would be to lose the soul of the nation and, crucially, the ‘natural’ connection to the land that could only be experienced and articulated through the native tongue. (461) 188 Earlier in this chapter, I examined how Heaney’s description, in ‘Broagh’, fuses the shape and sound of the place name (i.e. p p g y ( ) See https://www.youtube.com/watch?v=8HWurkQ1ao4&t=2094s. 182 Speaking about the etymology of Mossbawn, in a 2008 interview with Paul Muldoon at the New Yorker king about the etymology of Mossbawn, in a 2008 interview with Paul Muldoon at the New Yorker Heaney says: [ ] i f t th l ll d M b d th it b h iti d th (Corcoran 40). As Parker has observed, the sound gh and the letter O) and its vernacular geophony (i.e. the wind and the rain), and how in doing so he highlights their unique Irishness (see 3.1). ‘Broagh’, like ‘Anahorish’ and ‘Mossbawn’,182 is not a Gaelic word, but rather it is a transliteration of the Gaelic word ‘bruach’ by the very “strangers” who, as Heaney indicates in the poem, are deemed to find it too “difficult to manage” (WO 17). The “strangers” who found the original pronunciation “difficult” have altered it, changing the phonetic area of difficulty into another sound that they would be able to manage. Yet the italicised “gh” at the end of ‘Broagh’ continues to operate as what Auden calls “a verbal contraption” (50). The guttural consonant is what the “strangers” still find “difficult to manage” (WO 17). While the term “strangers” could refer to any non-native of Co. Derry, Heaney is particularly referring to the experience of English settlers. The Gaelic velar fricative gh − pronounced ch [x] − as Foster points out, “was also an English sound that disappeared early in the Modern English Period” (48). Heaney’s highlighting of the sound is a way of cherishing the demarcating phoneme and poking the Anglo-Scottish settlers: “the English can’t altogether manage that last gh sound, that guttural slither […]. If I can write the right poem about Broagh it might be minuscule definition, getting in affection, elegy, exclusiveness [...] I mean it might touch immediately, though not spectacularly, the nerve of history and culture” (qtd in Foster 10). Heaney turns to the experience of guttural sounds also to indicate the fate of Protestant and Catholic divisions in Northern Ireland (Hart 65). In a 182 Speaking about the etymology of Mossbawn in a 2008 interview with Paul Muldoon at the New Yorker 189 much quoted 1972 interview with Elgy Gillespie in The Irish Times, Heaney says: “Protestant and Catholic can say it perfectly in this part of the world. Yet the Protestant won’t be entranced by its Gaelic music. He’ll think of the gh sound perhaps, as in Scottish” (10). On the other hand, however, ‘Broagh’ can be seen as the poet’s finest attempt to bridge the gap between his professional language (English) and the home-grown Ulster tongue (Gaelic). 183 See www.ricorso.net/rx/library/criticism/classic/Anglo_I/Heaney_S/Unhappy.htm. (Corcoran 40). As Parker has observed, As previously noted, this level of contextual cultural and linguistic complication is also to be found in ‘Anahorish’, where Heaney translates the anglicised transliteration of the original Irish version into: “My ‘place of clear water”’ (WO 6). The hills and meadows of Anahorish are echoed in the name: the “soft gradient / of consonant, vowel meadow” (WO 6). Earlier I noted that Heaney thinks of “the personal and Irish pieties as vowels, and the literary awareness nourished on English as consonants” (PO 37). Therefore, in shape and in name, Anahorish echoes the poet’s life experience as a whole. In other words, although on one level his language and place-name poems mourn the linguistic dispossession of the Irish people as a major legacy of colonialism, on the other, as Parker has identified, “they deny silence and loss” in their utterance (97). In a 1977 interview with Seamus Dean, the poet speaks about his place-name poems in Wintering Out:183 I had a great sense of relief as they were being written, a joy and devil-may-carenes I had a great sense of relief as they were being written, a joy and devil-may-careness, I had a great sense of relief as they were being written, a joy and devil-may-careness, and that convinced me that one could be faithful to the nature of the English language and that convinced me that one could be faithful to the nature of the English lang – for in some sense these poems are erotic mouth-music by and out of the Anglo- Saxon tongue − and, at the same time be faithful to one’s own non-English origin, for me that is County Derry. Conclusion The four sections in this chapter have demonstrated the poet’s attention to geophonic, biophonic and anthrophonic sounds. Listening to the natural soundscape is very rewarding. Nature is both a teacher and a joy-bearer. It teaches us that there is much to learn about the habits of the land and its inhabitants. For the attentive listener, water never dies. It lives in the 190 continuous rhythm of waves, chords of streams, patterns of calm oceans and hasty rains. The wind is an element that grasps the ear forcefully when it is heard in the distance. Close by, the sensation is both tactile and aural. Heaney listens to the natural environment like a soundscape analyst to discover and record its significant features: familiar everyday sounds, sounds important to a culture, sounds that are unique to a place or sounds that carry a specific message. Listening to the natural ambient teaches him about the fabulous diversity and rich musicality of the apparently identical and familiar sounds. Like the mad Sweeney, Heaney’s imagination is another testimony to nature’s abundance and beauty. Sweeney says: “The alder is my darling” (PO 187). Like his Celtic ancestors, Heaney is just “another wood-lover and tree-hugger” (PO 186). In listening to natural sounds, he also practises musical listening. Heaney listens in order to tune in his senses and emotions to the patterns of nature and perceive the “little jabs of delight in the elemental” (PO 181). In doing so, however, he also reminds his readers of the transient existence of pastoral sounds (Quément 38). He encourages the reader to be in the now, in the present tense. In ‘Had I not been awake’, he signals the need to be alert and mindful of the present moment, when he writes: Had I not been awake I would have missed it, […] It came and went so unexpectedly [ ] […] […] But not ever After. And not now. (HC 3) […] But not ever After. And not now. (HC 3) He teaches us to listen attentively for what might happen, and then listen to it carefully, because what might happen now, might not happen again. You capture it now or it is gone forever. Just as with the sounds of the natural phenomena, Heaney’s references to the biophonic sounds of the natural world open a wealth of knowledge about the relationship 191 between humans and non-human animals. Conclusion Heaney writes not only about birds and other animals in general. He meticulously names a wide variety of species. In a world dominated by the voice of man and the din of his machinery, when the name of a species fades, its very existence and memory can be imperiled. In his poetry, Heaney writes as much about the nightingale and the blackbird, as he does about the sparrow, the corncrake, the cuckoo, the turkey, the snipe, the sand martin, the albatross and the lark. Heaney writes about such farm animals as bulls, sheep, dogs, pigs and horses to recall the soundscape of his childhood and the way of life that his father and grandfather pursued, but also writes about the otter, the ferret, the badger, the deer, the skunk, the fox, the hare and the rodent of which even many contemporary countryside men may barely be aware. He writes not only about the whale, the seals and the porpoise, which might appeal to environmentalists as “charismatic megafauna” − i.e. large species with popular appeal and symbolic values (Bekoff 87). Heaney also writes about a wide variety of less celebrated sea animals, including the oyster, the cod, the minnow, the eel, the spike, the mackerel and the salmon. As Schafer suggests, this “[l]inguistic accuracy is not merely a matter of lexicography” (34). Naming is caring. It is remembering. Naming is a prelude to listening to “a music that you never would have known” (SL 1). In his poetry, Heaney encourages his readers to listen to the frightened cry of birds, the agony of pigs, and the silent suffering of farm animals. He writes about the disappearing keynotes and identities, about the biophonic sounds that were once part of the natural soundscape, but are now either masked by the noises of “combines and chemicals” (WO 50) or lost in “a fieldworker’s archive” (20). Heaney’s poetry can be seen as a call for his fellow human beings to share the acoustic territory of the world with non-humans. O’Driscoll asks Heaney whether he thinks “poetry can play any practical or meaningful role in changing minds, and hearts on environmental issues” − gently reminding him that he had once, in 192 O’Driscoll presses him, asking if a poem stops a SUV? Heaney responds: I think that one answers itself. What has happened, however, is that environmental issues have to a large extent changed the mind of poetry. Again, it’s a question of the level of awareness, the horizon of consciousness within which poet and audience operate. (SS 407) I think that one answers itself. What has happened, however, is that environmental issues have to a large extent changed the mind of poetry. Again, it’s a question of the level of awareness, the horizon of consciousness within which poet and audience operate. (SS 407) Just as the sounds of the natural environment and non-human living organisms contribute to and reflect various cultural, social and historical associations, the sounds produced by human beings and man-made tools can reveal the social, cultural and political changes in the history of human civilisation. Therefore, blocking them as noise will deprive us of one of the greatest resources to understand the relationship between man and his acoustic environment. Commenting on the contemporary issue of noise pollution, Schafer writes, “[n]oise pollution results when man does not listen carefully. Noises are the sounds we have learned to ignore” (4). Heaney, however, is a careful and attentive listener to all sounds. His poetry brings up both the constructive and destructive noises of modernity. He is fascinated by the empowering presence of tractors and generators in farm life and acknowledges the freedom of movement brought forth by bicycles, cars, trains and planes. Yet he also insists that armoured cars and infantries are the unwelcome noises of human civilisation. He also reminds us that the acoustic space belongs to all living beings and that the noise of the plane engines and car engines can disturb the quietude of the natural environment and transform the habitat of animals. Heaney is aware of the need to preserve disappearing keynotes of human civilisation. Until not long ago, seed cutting, turf cutting, blacksmithing, thatching, water divining and churning were essential parts of not only Irish rural life, but also many other cultures. Yet, the vernaculars of early trades and manual farming practices which Heaney evokes are on the decline. In the present day, many of these crafts have already become rare or redundant. In his poetry, Heaney evokes the sounds of traditional rural crafts to re-orient himself towards 193 his origin and bridge the gap between his own profession and those of his ancestors. 184 See, for instance, Corcoran, Morrison, Foster. 184 See, for instance, Corcoran, Morrison, Foster. O’Driscoll presses him, asking if a poem stops a SUV? Heaney responds: As Vendler (1998) explains, “he makes himself into an anthropologist of his own culture, and testifies, in each poem, to his profound attachment to the practice described while not concealing his present detachment from rural life” (19). Yet Heaney evokes them not only to recollect the keynotes of his childhood, but also to collect the disappearing keynotes and rhythms of life and to seal their resonance on silent pages, making them a permanent and audible property of the world to come. Heaney has sometimes considered his first four poetry collections as forming “a single movement” or phase (SS 124). Yet the very notion of linguistic sounds as both the means and subject matter of poetry is the indication of a landmark in the poet’s intellectual development, leading Corcoran to stamp Heaney’s Wintering Out as the “seminal single volume of the post-1970 period of English poetry” (182). Both the word ear and tongue become prominent in Wintering Out. Scholars have not failed to identify the numerous images of tongue in his third collection: the “river tongues” in ‘A New Song’, the “swinging tongue” of Henry Joy McCracken’s boy in ‘Linen Town’, “the slab of the tongue” in ‘Toome’, ‘the civil tongues” of ‘The Last Mummer’ and “the gallery of the tongue” in ‘The Wool Trade’.184 Heaney’s sensitivity and curiosity towards human voice can also be traced in the polyphony of voices evoked through his own voice and poetry. Sounds of crying, sobbing, laughing, grunting, sighing and humming, as well as the various utterances made to express thoughts and feelings are the integral parts of everyday life. Heaney is aware that the world around us is full of voices, full of people with something to say, people who deserve to be remembered and listened to. “When, […], environmental sound reaches such proportions that human vocal sounds are masked or overwhelmed, we have produced an inhuman environment” (Schafer 207). Likewise, when louder voices 194 obscure others by the din of modern life, we have produced an inhuman environment. However, for Heaney, listening to other voices is vital to his attempt at creating a poetry that is not only personally satisfying but also socially aware. In a 1983 interview with Francis X. 185 See https://www.nytimes.com/1983/03/13/magazine/poet-of-the-bogs.html. ) p g p p p g p 188 See Cage’s Silence (1961), for the relationship between sound, music, noise and silence; Schafer’s The Tuning of the World (1977), for the definitions of negative and positive silence; Bruneau’s “Communicative Silences” (1973) and Jensen’s “Communicative Functions of Silence” (1973), for the psychological, interactive and sociocultural functions of silence. p p y ( ) 7 See Heaney’s 2003 Lannan Foundation Poetry Reading: www.youtube.com/watch?v=zjV7APxLa8c ( 05). For him the poems are the gifts of inspiration and reflection that occur in a pre-speaking period. 186 The title of this chapter is inspired by the final line of ‘Clearances’ (HL 26-34). O’Driscoll presses him, asking if a poem stops a SUV? Heaney responds: Clines, Heaney maintains he wants his poem to be “full of voices, full of people”.185 In The Redress of Poetry, he said: “[p]oetry […] has to be a working model of inclusive consciousness” (RP 7-8). Heaney’s attention to the voices of others and their evocation in his poetry brings us back to one of the most compelling aspects of acoustic ecology: its emphasis on listening to the other. Heaney’s poetry encourages a sensitivity towards surrounding voices. It highlights the need for opening our ears to not only louder voices but competing voices, silenced voices and forgotten voices. Heaney’s poetry is an encouragement for what Schafer phrases as “clairaudience” or “a total appreciation of the acoustic environment” (4). On numerous occasions, in his poetry and prose, through his onomatopoeic effects, internal rhymes, alliterations, enjambments, careful word choices and punctuations, metaphorical language, vivid descriptions and deep insights, Heaney invites his readers and himself to listen to ‘the music of what happens’ (FW 53). This music, for him, can be the sound of what happens and what happens next, waves and wind, a singing bird, a digging spade, the bells, the train, the familiar voices or the voices of foreign languages coming from the radio. Heaney draws inspiration from the world around him, where he finds the extraordinary in the ordinary. 195 Chapter 4: See Heaney s 2003 Lannan Foundation Poetry Reading: www.youtube.com/watch?v=zjV7APxLa8c (0:38- 1:05). For him the poems are the gifts of inspiration and reflection that occur in a pre-speaking period. p p y ( ) 187 See Heaney’s 2003 Lannan Foundation Poetry Reading: www.youtube.com/watch?v See Heaney s 2003 Lannan Foundation Poetry Reading: www.youtube.com/watch?v zjV7 1:05). For him the poems are the gifts of inspiration and reflection that occur in a pre-speaking p p p y ey’s 2003 Lannan Foundation Poetry Reading: www Introduction All poems, in a sense, are born out of infancy. Infans, the Latin word, means unspeaking, and the infant is the unspeaking one. So all good poems have been gathered in silence and have moved from the unspoken need to the luck of getting spoken right.187 According to John Cage, “there is no such thing as silence. Something is always happening that makes a sound” (191). Although the experience of absolute silence does not exist, as Heaney’s poetry also shows, we can experience silence in several ways. There is a quiet called silence, a stillness called silence, a loneliness called silence, an absence called silence, or a presence called silence. There have been several attempts by scholars from among psychologists, sociologists and ecologists to enumerate the various meanings of silence.188 The interpretation of silence in Heaney’s poetry is integral to the understanding of the poet’s thoughts and experiences. The more we dig down, the more we become aware that there are dense strands of silence interwoven into the soundscape of each poem. My interpretation of silence in Heaney’s Wintering Out is not encyclopaedic but presents a selection of examples − divided into the four categories of silent places, silent people, silence as a pause and silence as absence − where silence contributes to our understanding of Heaney’s poetry. This chapter begins with a focus on the relationship between silence and place. A soundscape is often called silent or quiet if it is possible for discrete sounds to be heard 196 clearly (Schafer 43). Moreover, in silent places − where seeing dominates listening − the structure and morphology of the place or the landscape, as well as other sensory, emotional or mental experiences affect the perception of silence (See 1.4.1). The experience of silence in vast prairies is unlike the experience of silence in deep valleys, silence in dense forests, silence on top of a mountain, silence in a farming field, silence on the beach, or silence in a library or a prayer room. The silence of a city at night is different from the silence of countryside at dawn, and silence in winter is different from silence in a summer day. The experience of silent landscapes and places can also be closely tied up with cultural, social, political associations. Silence can also carry different emotional, intellectual and psychological implications for different individuals. 189 ‘Sunlight’ provides a powerful example of a quiet and comforting silent place (N ix-x). But the experience of silence in ‘Summer Home’ − with the presence of maggots and decomposing rubbish − is negative and ‘Winter’s Tale’ opens into the bleak silence of a winter night. For Heaney, the silence of the Flaggy Shore, in ‘Postscript’, is delightful, yet the silence of the seashore as experienced by the women figures in ‘Maighdean M ’ d ‘Sh W ’ i l d i l ti I ‘B l d’ th il f th A i i i y p y p g p p gg g , work and the poet himself is contemplating (DN 1). In ‘Dawn Shoot’, the boys are intentionally silent so as not to wake up the prey: “[s]ilent we headed up the railway” (DN 16). In ‘Poor Women in a City Church’, the devotees are kneeling and praying silently (DN 29). In ‘The Play Way’, students are mesmerised by the music: “now / [t]he big sound has silenced them. […] A silence charged with sweetness” (DN 43). In ‘The Peninsula’, silence is caused as a result of feeling inarticulate, uncertain or absorbed in thoughts: “When you have nothing more to say, just drive […] (DD 11). In part two of ‘Funeral Rites’ and in ‘Casualty’, the funeral attendants are fected by its sights, is liberating, while the profound silence of the Irish bogs encourages a brooding moo 0 Heaney’s poetry encompasses a diverse range of silent people: In ‘Digging’, his father is immersed by its sights, is liberating, while the profound silence of the Irish bogs encourages a brooding mood. ey’s poetry encompasses a diverse range of silent people: In ‘Digging’, his father is immersed in his Heaney s poetry encompasses a diverse range of silent people: In Digging , his father is immersed ork and the poet himself is contemplating (DN 1). In ‘Dawn Shoot’, the boys are intentionally silent so a p , g , y p y g g Mara’ and ‘Shore Woman’, is gloomy and isolating. In ‘Bogland’, the silence of the American prairies, as affected by its sights, is liberating, while the profound silence of the Irish bogs encourages a brooding mood. 190 Heaney’s poetry encompasses a diverse range of silent people: In ‘Digging’ his father is immersed in his silent: “[t]hose quiet walkers” (FW 17). In poem 3 of ‘Clearances’, the silence between the young Heaney and his mother is unifying (HL 29). In ‘Wedding Day’ and ‘Mother of the Groom’, his mother is emotionally overwhelmed. The silence of his aunt Mary in ‘Chairing Mary’ is, perhaps, out of boredom, old-age, lack of topic or caused by daydreaming: “She sat in all day as the sundialled / Window-splays across the quiet floor” (DC 67). ( ) 191 For instance, like many rural communities, Ulster people were renowned for their “inwardness and reserve” (Morrison 23), while the silence of women and children in many societies is due to social-cultural expectations. For the working-class Irish community, for instance, the ideal woman was “dutifully subordinate, submissive, and obedient” (Isenberg 47). 192 Bruneau explains that it is often difficult to distinguish psychological silences from the interactive ones and that in many cases sociocultural silence underlies other forms of silence (36). Berger and Kurzon too note that there may be transitional cases between the two extremes. For instance, “not finding the right words” or “memory failure” might be the initial reasons behind someone’s silence, which are unintentional, but this might be followed by “a decision not to speak”, which is intentional (Kurzon 1677). Introduction One can interpret silence in the mountains as delightful, another as spooky, melancholic or empowering.189 In the first section, I discuss the different ways in which Heaney’s silent places enable us to gain a deeper understanding of his themes. In the second section, I focus my attention on Heaney’s representation of silence among people. According to Johannesen the silence of individuals can have several implications: people can be silent because they are, among other things, listening, daydreaming, thinking, attentive, bored, afraid, doubtful, unhappy, surprised, angry, stressed, impolite or surprised. They may use silence as a response, a way of showing empathy, distancing oneself or isolating others (29).190 However, their choice of silence may or may 197 not be a conscious and strategic decision. The silence of a vigilant hunter is different from the contemplative silence of a poet, and their silence is different from the silence of their audience or readers. The distinction should also be made in terms of the sources of silence − i.e. internal or external force (Kurzon 1675-1676). Silence can occur in dialogical, educational, legal, political, pathological, psychological or religious settings (Johannesen 29). However, in many situations, the sociocultural contexts can “manipulate” a person’s choice of silence (Bruneau 36).191 Finally, when exploring the communicative or performing implications of silence in Heaney’s poetry, we need to consider that in many instances there might be very little difference between different types of silence and that there might be several overlapping transitional phases.192 In the second section, I trace and interpret the silence of people in Heaney’s Wintering Out relying mainly − but not solely − on Kurzon’s typology of silence in sociocultural contexts. The length of silence, in many instances, may be hard to determine. But the perception of time in silence may still be relevant.193 Silences are differentiated according to their length. There can be a short pause, a long waiting or the eternal silence of death. The third section focuses on the interpretation of silence as a pause (see 1.4.1).194 In everyday life, y p , ( ) 193 The concept of time in silence is often measurable. For instance, the silence of the boys in ‘Dawn Shoot’ lasts as long as they can resist speaking and the period of silence of the students in ‘The Play Way’ is perhaps the same as the length of the music they listen to. 195 Examples of silence as a pause in Heaney’s poetry include: ‘Digging’, when Heaney’s grandfather stops for a short while to drink a bottle of milk; ‘A Call’ by implying the poet’s silent wait at the other end of the line; ‘The Blackbird of Glanmore’, when the poet stops the car to listen to the blackbird’s song; ‘The Makings of Music’ when he advises his readers to stop and listen to the surroundings; as well as poems such as ‘The Tollund Man’ and ‘Land’. 196 For instance, ‘Traditions’, ‘The Last Mummer’ and ‘New Song’ mourn the absence of the Irish language and tradition; ‘Midnight’ recalls the extinction of the Irish wolf; and ‘The Wool Trade’ mourns the eradication of the Irish wool trade. 197 The title of this section is inspired by Heaney’s description of Glanmore Cottage (SS 325). Introduction But, as Kurzon notes, silence about a subject matter cannot be timed as the speaker does not stop speaking and the listener is not often aware of the topic left out (1679). p p p g p ( ) 194 The term ‘pause’ in the prosodic sense is obviously not the concern of this thesis. In the third section, I also exclude verbal pauses discussed in the previous section. A verbal pause is often considered as an instance of psychological/psycholinguistic silence (e.g. a hesitation or self-correction), which is often very short, or in an 198 there are several short periods when silence creates a little space for reflecting upon what happened before or what might come after. A pause or stoppage may occur consciously or unconsciously because we are hesitant, tired, or just because we need time to listen, reflect, unwind and retune, and then to re-start. Heaney’s life and poetry demonstrate several such periods of silence, short or long, when a pause creates the space for deep listening and pondering.195 In one of its most conventional meanings, silence indicates the absence of a sound (Schafer 256). The interpretation of silence as ‘absence’ is also listed as the second definition in OED: “The state or condition when nothing is audible; absence of all sound or noise; complete quietness or stillness; […]”. The final section is dedicated to the interpretation of silence as absence. By ‘absence’, in this section, I mean the eradication of a sound-producing source − ranging from natural elements, to animals, and to people and their traditions − from a soundscape.196 interactive silence (e.g. take turning) that is often longer (Bruneau 36). y j ( ) 199 Heaney compares his poem to a Vermeer painting saying that there is an almost “ekphrastic” quality to this poem. See www.youtube.com/watch?v=ywMY7bkSn9g&t=20s. p y y g 200 The poem was recognised as Ireland’s best-loved poem in 2015: www.rte.ie/news/2015/0311/686370-poem/. 198 See www.youtube.com/watch?v=zjV7APxLa8c (1:08-1:12). y y g he poem was recognised as Ireland’s best-loved poem in 2015: www.rte.ie/news/2015/0311/686370-poem 4.1 A Contested Zone: Silence and Place197 In this section, I trace and interpret Heaney’s silent places. In quiet soundscapes, there is a low ambient noise level and discrete sounds − of wind, birds and human voice, for instance − can be heard with clarity communicating vital information about their sound- producing sources and their position (Schafer 43). Yet silent places not only amplify sounds encouraging people to listen more attentively, but also intensify sensing as a whole (Sardello 89). Heaney’s opening remarks at his Lannan Foundation Poetry Reading, quoted in the 199 epigraphs to this chapter, are followed by a reference to one of the most memorable silent places he experienced as a child: “the silence in the kitchen where I was infans”.198 In the absence of sounds, everything in the surroundings, every movement, even the empty space between things is bathed in “the reverence of silence” (Sardello 89). Heaney’s early poem, ‘Sunlight’, depicts a Vermeerian account of the reassuring atmosphere of the kitchen in the family house in 1940s, from the perspective of the young poet:199 “There was a sunlit absence” (N IX). The growing feelings of warmth and security are further reinforced by and reflected through his aunt’s unhurried movements around the kitchen, the hot stove, the rising scone and the vast space that Heaney realises between the tick of the two clocks in the kitchen: here is space again, the scone rising to the tick of two clocks. here is space again, the scone rising to the tick of two clocks. And here is love […]. (N X) This sensory awareness and coming together of all the senses in a silent place are also reflected in the third poem of the sequence ‘Clearances’ that opens into the similar intimate and familiar atmosphere of the Mossbawn kitchen in the presence of his mother. Heaney uses the sonnet form to romanticise the memory of the pair existing and working together:200 When all the others were away at Mass I was all hers as we peeled potatoes. They broke the silence, let fall one by one Like solder weeping off the soldering iron: Cold comforts set between us, things to share Gleaming in a bucket of clean water. And again let fall. Little pleasant splashes From each other’s work would bring us to our senses. (HL 29) When all the others were away at Mass I was all hers as we peeled potatoes. 4.1 A Contested Zone: Silence and Place197 They broke the silence, let fall one by one Like solder weeping off the soldering iron: Cold comforts set between us, things to share Gleaming in a bucket of clean water. And again let fall. Little pleasant splashes From each other’s work would bring us to our senses. (HL 29) The first lines create the image of an exclusive moment of togetherness (“When all the others The first lines create the image of an exclusive moment of togetherness (“When all the others were away at Mass / I was all hers”). Heaney’s use of the mundane language of everyday e away at Mass / I was all hers”). Heaney’s use of the mundane language of everyda y j ( ) 199 Heaney compares his poem to a Vermeer painting saying that there is an almost “ekphrastic” quality to this poem. See www.youtube.com/watch?v=ywMY7bkSn9g&t=20s. y j ( ) 199 Heaney compares his poem to a Vermeer painting saying that there is an almost “ekphrastic” quality to this poem. See www.youtube.com/watch?v=ywMY7bkSn9g&t=20s. 200 household chores lends the poem a touch of domestic authenticity (“peeled potatoes”). The poem is delicately balanced with internal rhymes (others/hers, weeping/soldering/gleaming, splashes/senses; one/one; all/fall) and assonances (all/ all/ water, peeled/weeping/between/clean/each, silence/like/beside) to link the words and images, creating an even greater sense of intimacy (HL 29). Heaney wrote the poem in memory of his mother who had passed away in 1984. His mother’s final moments trigger his memories of the days in the past when the silence dominant in the kitchen allowed them to share the deepest and purest feelings of love, comfort and support. Just as it intensifies the slightest sound from the surroundings, a quiet place also enables the inner voice of the poet to resound. The quietude at Glanmore cottage was similar to that at Mossbawn kitchen in the sense that it enabled the poet to listen to his senses and deepen his understanding of the surroundings but differed in the sense that it enabled him to find his poetic voice. For him, Glanmore was both a country retreat and a workspace, a zone of contest between poetic imagination, thoughts and reality. When asked about the status of the cottage in his life and poetry, he remarks: Workspace, first and foremost. Which meant that it was also a retreat. It remained what it always had been, the poetry house. 201 He refers to Glanmore cottage as “a completely silent place of writing” on a number of other occasions: “Meanwhile, the formal purchase we’d arranged with Ann Saddlemyer restored us to the ‘beloved vale’ in Wicklow. Glanmore Cottage was available from then on as a completely silent place of writing, close to Dublin, no phone, no interruptions whatsoever” (SS 322). p g p 903757?fbclid=IwAR2VU7aYJph2dNsB8UVnU3cY0uHfX8lNjPQ08YGZnVrC5xLP9eOuk0nevVE. 203 The writings collected by Amnesty International Irish Section and published by the Irish Times. See: https://www.irishtimes.com/news/human-rights-poetic-redress- 4.1 A Contested Zone: Silence and Place197 It’s a silence bunker, a listening post, a holding, in every sense of that word. It holds meaning and things, and even adds meaning. In my life, Glanmore Cottage stands for what Wallace Stevens said poetry stands for, the imagination pressing back against the pressures of reality. Glanmore is a contested zone – […], I think it’s a studio for writing in, […]. I always found the place conducive to writing and it saved my writing life, because I was able to disappear from home in Dublin – which was becoming like a cross between a travel agency and a telephone exchange – and bury myself down there. (SS 325-326)201 Listening to the inner voice, in turn, enables the poet to access his creative power. In order for 201 silent place opens for him a field of artistic possibilities and enables him to bridge between his ordinary bodily experiences and imagination, directing them towards a poem.202 Heaney elaborates on the connection between the silence of Orkney, the poem and its title, saying:203 I took it that Conscience would be a republic, a silent, solitary place where a person would find it hard to avoid self-awareness and self-examination; and this made me think of Orkney. I remembered the silence the first time I landed there. When I got off the small propeller plane and started walking across the grass to a little arrivals hut, I heard the cry of a curlew. And as soon as that image came to me, I was up and away, able to proceed with a fiction that felt workable yet unconstrained, a made-up thing that might be hung in the scale as a counterweight to the given actuality of the world. Natural landscapes are generally imagined to be silent or quiet because they are away from the clamour of the city. They are high-fidelity environments where “sounds overlap less frequently; there is perspective − foreground and background” (Schafer 43). The quiet ambience of Co. Derry’s countryside that Heaney grew up in was an early example of the hi- fi soundscape that allowed the poet to exercise this distant listening and expand his sense of place (see my Introduction). The quietude of the countryside deepens every visitor’s engagement with the topography and ecology of the place. 202 The poem was written in commemoration of the 25th anniversary of Amnesty International. 203 4.1 A Contested Zone: Silence and Place197 In Heaney’s poetry, there are several other hi-fi soundscapes − ranging from the quiet soundscape of peat bogs, hills and flax dams, to coastlines and lakes, where the experience of silence is also closely bound to the topography of the landscape, as well as the various personal, cultural and social associations. ‘Oracle’ is set in the quietude of the woodland surrounding Heaney’s childhood house. The silence of the wood is full of secrets. There can be a gentle humming through the grass, a whistling through the branches, a gusty wind, a creaking from afar, a lively chattering of birds and then suddenly a rest. In the wood, Nature is not silent, it is not still, but one needs to be both silent and still to be able to see and hear what is happening. There is, therefore, a sense of waiting in this silence and stillness, one that invites the listener to listen attentively 202 for something, anything (Maitland 160). As Vendler has noted, ‘Oracle’, is Heaney’s announcement of his poetic vocation, for the non-generic way in which the poet thinks of himself (28), as the “lobe and larynx / of the mossy places” (WO 18). In ‘Oracle’, listening becomes the only way of sense-making. The poet-in-the-making becomes a silence seeker and an intent listener, so much so that he proceeds to “[h]ide in the hollow trunk / of the willow tree” (WO 18), allowing the voices of the surroundings to be one with his inner voice. The stillness of open hills and valleys blurs the sense of time and space (see 1.4.1). In Anahorish − “the first hill in the world” (WO 6) − nature is best listened to. The sounds of the place-name become almost synonymous with the song of the local river in every spring (see 3.1 and 3.4): “[m]y ‘place of clear water’” (WO 6). But in the silence of Anahorish, the poet is also able to travel backward in time to visit men “[w]ith pails and barrows // those mound- dwellers” (WO 6). In the silence of Toome too he embarks on a time journey. Toome Bridge, located in Bann Valley, was the site of archaeological finds and the site of the Irish rebellion in 1798 when the Royal Ulster Constabulary with the king’s soldiers massacred thousands of individuals from the United Irish Army (Molino 183). 4.1 A Contested Zone: Silence and Place197 In the silence of Toome, the wordsound continues to reverberate in the poet’s inner ear “Toome, Toome” like a “souterrain” sending the poet back to the “blastings” of British cannon fire and remains of past civilisations: “fragmented ware” and “torcs and fish-bones” (WO 16). But it also sends the poet “a hundred centuries” into the future of Ireland to seek “what [is] new” (WO 16). For Heaney, the silence of his familiar hills and valleys are the sites for travelling through time and space. The poem ‘Land’ evokes another quiet territory from Heaney’s childhood, but one combined with the view of acres of “outlying fields” (WO 11). Like the wood and hills, the silence of farming fields is nothing like absence. There is perhaps the sound of water running through the stone and grass in narrow streams, the sound of wind whooshing across the fields 203 and squeezing through the cairns. But more importantly, the silence of the fields amplifies the vernaculars of farming lifestyle (see 3.3). Both in ‘Oracle’ and ‘Land’, as well as in ‘Personal Helicon’ as Tobin has noted, Heaney discloses his poetic vocation (79). While in ‘Oracle’, Heaney is all ears, silent and still waiting in the hollowed trunk of a tree, in ‘Land’ he participates in the making of the soundscape to not only stock up his memories of the family farmland, but also internalise them. Heaney proves his profound attachment to his land by stepping in through the persona of an anonymous and silent Ulster farmer, participating in the composition of the soundscape and making it the “forever part of his inner landscape” (Vendler 21): “I stepped it, perch by perch. / […] I composed habits for those acres” (WO 11). In the final section of the poem, the poet traveller resorts to listening as a way of reconnecting with his homeland and reliving the sensation of farming life but, having experienced the farming soundscape, it suffices if he lends his ears to silence: I sense the pads unfurling under grass and clover: I sense the pads unfurling under grass and clover: if I lie with my ear in this loop of silence. (WO 12) As his first place and a legacy of his Irish memories, the soundscape of Mossbawn remains an integral part of Heaney’s “Irish memory bank” (SS 142). 4.1 A Contested Zone: Silence and Place197 In the empty silence of California nights, in ‘Fodder’ − the poem he wrote while in Berkeley − he continues to search for the familiar rural and linguistic vernaculars of his first place. The distance from home only heightens and intensifies the Ulster poet’s mixed feelings of homesickness and guilt, and the “chorus of ‘Hi-i-s’ and ‘Wow-w-w-s’ and ‘He-e-eys!’” of Californian students (SS 138) only makes him long for the most homely word “fother”, a word whose articulations spills out of his sensory memories of the family home in minute details (WO 3) (see also 3.4). The acoustic clarity with which Heaney experienced the soundscape of Mossbawn now enables him to recall the slightest sound, sight and other sensory experiences. Hence, the weathered 204 eaves gently “falling” on his feet, a bundle of fother being “tossed” over the half-doors onto the wet floor stables and a whiff of old “tumbled” grass and a scent of “meadowsweet” bring the poet to being actually present in the backyard of his Mossbawn house (WO 3). For him “these” long nights of separation feel too quiet, empty and disheartening, and as Parker has remarked: “Heaney summons its memory to sustain him during the cold bleak present” (94). But Heaney’s acoustic memories of his childhood home are not limited to fields and hills. The profound silence of the surrounding bogs or the kind of rare silence he experienced in the rocky lands of Co. Clare have also forged the poet’s perception of life and ultimately the soundscape of his poetry. In ‘Bogland’, Heaney associates Irish national identity with Irish landscape distinguishing it from that of the US characterised by vast prairies. In the first lines, he juxtaposes the short word ‘bog’, characterised by a short vowel enclosed with hard consonants, with the open expansive word ‘prairies’: We have no prairies To slice a big sun at evening – [...] […] Our unfenced country Is bog that keeps crusting Between the sights of the sun. (DD 43) In a 2008 poetry reading session at the New Yorker Festival, Heaney explains the genesis of the poem hinting at the main distinction between “outward, direction-finding, [and] pushing- out” American prairies and the “going-down” Irish bogs (DD 43).204 The term soundscape is the sonic equivalent of its visual counterpart, landscape (see 1.4.2). g, p g , g g g See www.youtube.com/watch?v=8HWurkQ1ao4&t=1913s (34:33-34:53). ea ey e p a s t s topog ap ca d st ct o betwee e ca a d e a d a d ts sy bo c po ta ce a reading-conversation with Muldoon, when he says: “In America, the prairie is very important, the frontier is very important and that is their myth. So, after I taught our class, I thought what’s ours? I mean, Americans have this terrific outward direction-finding, pushing-out, we have a bog going down” ery important and that is their myth. So, after I taught our class, I thought what s ours? I mean, America his terrific outward direction-finding, pushing-out, we have a bog going down” 4 Heaney explains this topographical distinction between America and Ireland and its symbolic importanc very important and that is their myth. So, after I taught our class, I thought wha this terrific outward direction-finding, pushing-out, we have a bog going down y p p g p y p reading-conversation with Muldoon, when he says: “In America, the prairie is very important, the frontier is very important and that is their myth. So, after I taught our class, I thought what’s ours? I mean, Americans have g, p g , g g g See www.youtube.com/watch?v=8HWurkQ1ao4&t=1913s (34:33-34:53). reading-conversation with Muldoon, when he says: “In America, the prairie is very important and that is their myth. So, after I taught our class, I thought what’s ody, the rope around his neck, the calm and solemn look on his face, led to questions about his life and de year after the publication of Wintering Out, he wrote an extract of the poem in the guest book of the mus p p g j been living during the first phase of the Iron Age (300-400 B.C.). The Tollund man’s incredibly well-preserved body, the rope around his neck, the calm and solemn look on his face, led to questions about his life and death. 5 The ‘pre-historic face of Denmark’ was discovered in 1950, in the peat bogs of Bjældskovdal. The man See www.tollundman.dk/. y p g , p g See www.tollundman.dk/. p , , q fter the publication of Wintering Out, he wrote an extract of the poem in the guest book of the museum. w.tollundman.dk/. 4.1 A Contested Zone: Silence and Place197 The contrast between the landscapes of the two countries, in the lines quoted above, also highlights the distinction between the silence of vast American prairies and the role of Irish bogs as the echo-chamber of history and culture. Like a historical archive, bogs store and preserve what has been 205 deposited in them over millions of years. The kind of silence experienced in a bogland, for Heaney, is earthbound. It is layered. You probe it “inwards and downwards” (DD 43). The layered structure of bogs allows Heaney to transcend temporality and perceive the echoes of the past sounds and voices. The Irish poet is the most scrupulous fieldworker of silent peat bogs. For him every seam of coal underground represents a buried layer of human history. In ‘Bog Oak’, the poet sets out to retrieve the hidden treasures from the deepest layers of the peat bog and bring a “long-seasoned rib” (WO 4) to the troubled days of Ireland. For him, each rift and bruise relate a story of Irish strength and endurance. Schafer says: “[m]an is an anti-entropic creature; he is a random-to-orderly arranger and tries to perceive patterns in all things” (226). The preserving capacity of bogs allows Heaney to not only unearth national memories, but also witnesses the patterns with which the history of mankind repeats itself. In ‘The Tollund Man’, Heaney alludes to this aspect of history by associating the atrocities of the Iron Age in the Jutland Peninsula in Denmark with the modern version of human barbarities, Northern Irish political killings in particular.205 Referring to the Iron Age rituals he says: “[…] this is more than an archaic barbarous rite: it is an archetypal pattern” (PO 57- 58). 58). In the lines quoted from ‘Bogland’ Heaney refers to yet another attribute of the silence he experienced in the Irish bogs, which corresponds with Garden’s references to the transatlantic associations of Heaney’s waterlogged poems (see 2.2). For him, the silence of the Irish bogs is “unfenced” (DD 43). In ‘Bogland’, Heaney cherishes bogs as a “soft” and “waterlogged” ground. The silence that accumulates at the “wet centre” of a bogland is limitless, in the sense that it is, as Heaney puts it, “bottomless” connecting it to the Atlantic waters (DD 43). According to Garden, the shifting and soft soil of the bog has a crucial role 206 as the shared site of “haunting memories” between Ireland and her neighbouring nations (91). The geological similarities of Jutland are another reason why Heaney feels “at home” in his imaginary trip to Denmark. The setting of ‘The Tollund Man’ crosses the boundaries of the Ulster landscape and stretches over the boglands of Denmark. In his 1996 talk at the Silkeborg Museum, where the body is currently preserved, Heaney reveals his fascination with the mystery and silence that surrounded the peat bogs of his Mossbawn home and Jutland, as well as with their layered structure and their capacity to preserve the treasures of time and place. In his 1996 speech at Silkeborg Museum, he says: When I was a child and an adolescent I lived among peat-diggers and I also worked in the peat bog myself. I loved the structure the peat bank revealed after the spade had worked its way through the surface of the peat. I loved the mystery and silence of the place when the work was done at the end of the day and I would stand there alone while the larks became quiet and the lapwings started calling, while a snipe would suddenly take off and disappear [...].206 Both bogs and the Burren have come to epitomise Irishness for resisting the colonial settlement and thus, any alteration to the landscape (see 2.2).207 The karst landscape houses over 120 cairns of varying sizes that once served as a landmark for boundaries and routes or memorials and burial chambers from the pre-Christian period.208 Today the resonating silence of the Burren has remained as a symbol of resistance and source of imaginative inspiration for many modern Irish writers and artists including Heaney (see 2.2). g 208 The Burren − from the Irish word ‘Boireann’ meaning the ‘rocky place’ − stretches over 250 square kilometres in the southwest of the Republic of Ireland in Co. Clare. The landscape has over 6000 years of agricultural history and today is especially recognised for its distinct flora and fauna as well as for the high density of the preserved monuments. See www.burrennationalpark.ie/. 207 For this reason, both bogs and Burren were regarded by the British as wasteland. See Gerard Boate’s Ireland's natural history (1652) for more information on how colonisers compared the character of Irish people in relation to bogs and their formation. 206 See archive.archaeology.org/blog/the-poet-and-the-bog-body/. poet (SS 148). 210 The title of this section is inspired by Heaney’s poem ‘Whatever You Say Say Nothing’ (N 52-55). 211 This structure follows Bruneau’s categories of silence in “Communicative Silences” (1973) − i.e. psychological, interactive and sociocultural − in order to encompass the range of silences presented by Heaney. Bruneau, however, explains that it is often difficult to distinguish psychological (i.e. psycholinguistic) silences from the interactive ones and that in many cases the third type − sociocultural silence − underlies the first two forms (36). 58). Cairn-making for Heaney is an art that associates him closely with his cultural heritage. In fact, before leaving the farming lifestyle behind, in ‘Land’, the poet makes sure to perform cairn-making for one last time: “[…] gathered stones off the ploughing / to raise a small cairn” (WO 11). Heaney’s cairn-maker, in the poem ‘Cairn-maker’, is a painter, who attempts to encode and imprint the 207 music of the Burren on the canvas.209 Like the poet himself, he is a meticulous observer, a silent listener and a devoted keeper of his cultural heritage. He spends the “whole day” patiently undoing the old cairns and then clamping, balancing, chambering “small cairn after cairn” (WO 39). The so-called barren land continues to resonate what Potts calls the “music in stone” (39), cradling the ears of the attentive observers and listeners. The cairn-maker is, in fact, any contemporary Irish artist with a unique way of restoring “the culture to itself” (PO 41). ( ) 212 See www.youtube.com/watch?v=AdTQPZGW0Bs (1:20-1:52). 209 The poem is dedicated to Barrie Cooke, the Ireland-based British born painter who painted a portrait of the poet (SS 148). p y y p y y g ( ) 211 This structure follows Bruneau’s categories of silence in “Communicative Silences” (1973 h l i l i i d i l l i d h f il d b 209 The poem is dedicated to Barrie Cooke, the Ireland-based British born painter who painted a portrait of the poet (SS 148). 210 The title of this section is inspired b Heane ’s poem ‘Whate er Yo Sa Sa Nothing’ (N 52 55) p p p p poet (SS 148). 210 The title of this section is inspired by Heaney’s poem ‘Whatever You Say Say Nothing’ (N 52-55) o s (36). 212 See www.youtube.com/watch?v=AdTQPZGW0Bs (1:20-1:52). ( ) 214 See www.telegraph.co.uk/culture/4722682/Seamus-Famous.html. 4.2 Whatever You Say, Say Nothing: Silence and People210 Just as there are various kinds of silent places, the silence of people can have different or even contrasting functions and implications. In this section, my study of the silence of individuals in Wintering Out focuses on silence as a psychological state and personal attribute (e.g. silence of an introverted person), silence in everyday interpersonal communications (i.e. silence in a dialogical context between two or more individuals), and the silence of individuals and communities in the broader social, cultural and political context (i.e. silence of minorities, subordinates, women and children).211 One of the first silent figures in the poet’s life was his father, Patrick Heaney. In his 2012 AWB Vincent Literary Award acceptance speech, Heaney thinks of his father as a man who considered speech as “affectation”.212 Patrick Heaney’s silence does not stem from inner conflict, indecisiveness or forgetfulness, but from his lack of faith in and mistrust of the efficiency of language. As Heaney himself puts it, the old man’s silence was “a kind of 208 counterweight to all speechifying and theory-speak” (SS 287). This impression is given from Heaney’s first poem, ‘Digging’, where his father is described as being very absorbed in the activity.213 In a 2001 interview with Nigel Farndale, Heaney recounts memories of his father describing him on many occasions as a silent person. One of his earliest and most unforgettable memories is when his father returns home in the afternoon “without his hat and going to bed”. The man had nearly drowned after his horse reared up near a riverbank, but it was the shock of seeing this unusual behaviour − not wearing a hat and not working for the rest of the day − weighed with his silence that turns the incident into “an eternal image” for the poet.214 Patrick Heaney left school at the age of 14 and started his career as a cattle dealer and farmer. While Heaney admired his father’s skill and accompanied him to cattle fairs and markets, he chose to follow his education in literature and writing poetry. Yet Heaney’s father’s reaction to his son’s career choice was merely silence. Reflecting on his father’s silence in response to his education and career, Heaney remarks: I mean, he didn't devalue it, he wasn't afraid of it, or against it, he watched it happen - as he did the oddity of me publishing a book and himself being in it. 213 This kind of silence is internal and intentional − as opposed to the imposed silence, which I explain later in this section − and can be considered as a personal attribute. Several other figures in his poetry seem to have a similar attribute: In ‘Digging’, Heaney gives a similar impression about his grandfather/uncle, whom he remembers as a child: “Once I carried him milk in a bottle […] He straightened up / To drink it, then fell to right away” (DN 1). The thatcher too was a man of few words. In the poem, he almost explicitly states that the man does not engage in any conversation when he arrives at Heaney’s home for work and nor does he talk to the child Heaney who spends the entire day observing him: “[…] he turned up some morning […]. He eyed the old rigging, [...]. Next, the bundled rods […]” (DD 10). The tailor, in ‘At Banagher’, is described as being tight- mouthed while working: “His lips tight back, […], / Keeping his counsel always, giving none” (SL 67). This kind of silence could also be interpreted as cultural. In fact, as Morrison has noted, Ulster was “renowned like all rural communities for its inwardness and reserve” (23). g p 215 See www.telegraph.co.uk/culture/4722682/Seamus-Famous.html. y p y g [ ] p g [ ] y rigging, [...]. Next, the bundled rods […]” (DD 10). The tailor, in ‘At Banagher’, is described as being tight- mouthed while working: “His lips tight back, […], / Keeping his counsel always, giving none” (SL 67). This kind of silence could also be interpreted as cultural. In fact, as Morrison has noted, Ulster was “renowned like all rural communities for its inwardness and reserve” (23). mouthed while working: “His lips tight back, […], / Keeping his counsel always, giv kind of silence could also be interpreted as cultural. In fact, as Morrison has noted, Ul all rural communities for its inwardness and reserve” (23). 216 See www.telegraph.co.uk/culture/4722682/Seamus-Famous.html. g p 217 See www.telegraph.co.uk/culture/4722682/Seamus-Famous.html. 4.2 Whatever You Say, Say Nothing: Silence and People210 That must have been a curiosity. We didn't quite deal with that, we didn't even discuss it. No way of discussing it.215 Even on his son’s wedding day, Patrick Heaney refused to give the expected traditional wedding speech. Yet on this occasion too his silence was internal and intentional and a way of expressing his presence and support. The poet later commented: “[On the wedding day,] [t]he families being themselves, even more themselves than usual, [...]. My 209 father refused to get into the hired gear; nor would he make a speech, since speech was never his thing anyhow” (SS 253-254). Although we cannot know what the person would have exactly said if he or she had spoken, as Kurzon has noted, this type of silence is equivalent to a “speech act” (1676). Heaney says: “There wasn't that much substantial exchange but there was a good bit of silent assignation between us”.216 Patrick Heaney’s silence was in fact a way of communicating his thoughts and feelings to his son. In ‘The Harvest Bow’, he gifts his son a plaited bow without speaking a word, yet Heaney is able to correctly interpret his father’s silence as the indication of his love: As you plaited the harvest bow countryside, which he drew from Guthrie s The Greeks and Their Gods (1950): Hermes then is an ancient god of the countryside, named by the Greeks from the herma, also called hermaion, which was a cairn or heap of stones. These cairns served as landmarks […] to explain the connexion of Hermes with the cairns, the Greeks characteristically invented an aetiological myth. When Hermes killed Argos, he was brought to trial by the gods. They acquitted him, and in doing so each threw his voting-pebbles (phethos) at his feet. Thus, a heap of stones grew up around him. (qtd in Vendler 75) 218 Heaney highlights this in ‘The Other Side’ when describing the chanting of the liturgy in their kitchen, juxtaposing it with the kitchen of the Protestant neighbor “hung with texts” (WO 25). p g g g ( ) ere, Heaney foregrounds the association between Hermes − the tutelary god of speech and oratory − and t tryside, which he drew from Guthrie’s The Greeks and Their Gods (1950): p g g g ( ) ere, Heaney foregrounds the association between Hermes − the tutelary god of speech and oratory − and t id hi h h d f G h i ’ h G k d h G d (19 0) As you plaited the harvest bow As you plaited the harvest bow y p You implicated the mellowed silence in you […] A throwaway love-knot of straw. (FW 55) In another instance, in ‘Mid-Term Break’, the poem that recounts the tragic death of Heaney’s younger brother Christopher (see 3.4), the poet’s father is portrayed crying silently in front of the entrance of the house − whereas one of the neighbours, big Jim Evans, sympathises with the poet about Christopher’s accident describing it as “a hard blow” (DN 15). When asked by Farndale if there was any unfinished business left between him and his father when he passed away, Heaney replies: “plenty left unsaid, but nothing left un- understood” – which is a wonderful way of describing his silence as a way of communication between himself and his father.217 Knowing his father’s mistrust of speech, in ‘The Stone Verdict’, the poem he wrote one year after his father passed away, Heaney anticipates that even on the day of judgement his father will hold on to his silence and pride: When he stands in the judgment place With his stick in his hand and the broad hat Still on his head, maimed by self-doubt And an old disdain of sweet talk and excuses […]. (HL 19) 210 On the other hand, Heaney is aware of the significance of orality in Catholicism – by contrast with Protestantism that mainly relies on the interpretation of the written script (Ong 284-285).218 Moreover, according to Christian teachings, “'Word” is almost synonymous with Jesus Christ: “In the beginning was the Word, and the Word was with God, and the Word was God” (Gospel of John 1:1). It is also believed that immediately after death, Christ will render judgement to the departed person (Gospel of Mathew 25:31-46). Knowing that his father was a man of few words, Heaney suggests that it will not be fair if the old man’s verdict “is blabbed out” in front of him as “he will expect more than [just] words”. Instead, just like the pre-Christian god Hermes, his father should be judged surrounded by silent stones and venerated on the earth by “the cairn / Of his apotheosis” (HL 19): It will be no justice if the sentence is blabbed out. He will expect more than words in the ultimate court He relied on through a lifetime's speechlessness. He relied on through a lifetime's speechlessness. 218 Heaney highlights this in ‘The Other Side’ when describing the chanting of the liturgy in their kitchen, j t i it ith th kit h f th P t t t i hb “h ith t t ” (WO 25) 219 Here, Heaney foregrounds the association between Hermes − the tutelary god of speech a countryside, which he drew from Guthrie’s The Greeks and Their Gods (1950): As you plaited the harvest bow Let it be like the judgment of Hermes, God of the stone heap, where the stones were verdicts […]. (HL 19) As Vendler notes, many of Heaney’s elegiac poems could be read as “part of the distinctive (and often successful) modernist effort to rewrite, in more believable terms, the heroic, sublime and religious conventions of the classical elegy” (74). However, when it comes to elegizing his father, who is mistrustful of speech, the poet is driven to the ancient Greek myth of Hermes for what he thinks would be the appropriate judgement process for his father, who might have likewise stood by “a gate-pillar / Or a tumbled wallstead where hogweed earths the silence […] (HL 19).219 211 Heaney also speaks about the silence of his mother on his wedding day. But the silence of Heaney’s mother on that day was of a different sort. The poems ‘Wedding Day’ and ‘Mother of the Groom’ in Wintering Out flash back to two different scenes from the poet’s wedding to Marie Devlin in August 1965. For most people, a wedding is a prominent and life-changing occasion. Yet anyone who has been to a wedding has experienced the co- existence of cries and laughters on this special day. Recalling his own wedding party, the poet says: “[e]verybody [was] in good form and in full cry” (SS 253). In ‘Mother of the Groom’, he captures some of his mother’s contrasting feelings through what seems to be silence of a person who “is emotionally overcome” (Johannesen 29). While everyone else is clapping and rejoicing, Margaret Kathleen McCann Heaney is portrayed sitting quietly with her hands on “her [now] voided lap” remembering the time when today’s groom was a small toddler (WO 46). As the bride and groom enter the room, she experiences a surge of guilt and grief, as if the child “slipped her soapy hold” (46). For her, the union of the young couple is the indication of an ending to the parent-child circle she and her son have been holding for years. As Heaney puts it: “a wedding always has its moments of strangeness, sudden lancings or fissures in the fun when parent and child have these intense intimations that the first circle is broken. 220 OED defines the term ‘unheimlich’ as ‘uncanny and weird’. In Das Unheimliche (1919), Sigmund Freud used the concept to refer to the experience of strangeness in the familiar. The term has a considerable use in the literary treatment of gothic literature. As you plaited the harvest bow It is in the literal sense unheimlich, an unhoming [emphasis in the original]” (SS 254).220 Yet the wedding ring “bedded” in her finger is the proof that the mother of the groom is not foreign to the “legend of love” herself (45). The woman overcomes her worries and breaks her silence by joining the clapping guests. Heaney’s own silence on the day of his wedding was a rather strange and unforgettable experience for him. It was similar to that of his mother in the sense that it was perhaps unintentional and caused by an external situation. ‘Wedding Day’ introduces a state of confusion and anxiety experienced by unreeling silent images from the groom’s memories. 212 It’s all vision, “[s]ound has stopped in the day” (WO 45). Unlike sound waves, which “penetrate the body of the listener” inevitably and evoke a sense of connection, visualised objects and images “stay out there” giving the viewer a sense of perspective and otherness (B. R. Smith 7). Heaney’s inability to speak and hear the sounds and voices in the wedding party indicate his feelings of being distant, overwhelmed, uncertain and “afraid” (WO 45). Heaney remarks that it was, after all, a party everybody enjoyed, but the grieving faces, cries and waving hands when he and his bride were driving away made it “hallucinatory” and thus, haunting (SS 254). In the poem, Heaney juxtaposes his own speechlessness with Marie’s singing. The “deserted bride” (WO 45) is the only individual with the courage and strength to bear the strain, restore balance between “grief and joy” (Devlin 84) during the ceremony and voice her feelings (see also 3.4). The functions of silence in more overtly dialogical and interpersonal contexts are also brought up in such poems as ‘Mid-Term Break’, ‘Funeral Rites’, ‘Docker’, and ‘The Sharping Stone’ as well as in ‘Summer Home’ and ‘The Other Side’ from Wintering Out. As mentioned at the beginning of this section, silence between two or more individuals can have several communicative functions. 222 The scale and impact of sectarian discrimination in the shipbuilding industry of Ulster has been the subject of many debates. See www.independent.co.uk/voices/commentators/david-mckittrick-not-everyone-is-weeping-as- the last ship leaves titanic town 221 Jensen identifies the five functions of silence in an interpersonal context: connecting-distancing 221 Jensen identifies the five functions of silence in an interpersonal context: connecting-distancing function, perlocutionary effect, revelational function, judgemental function and activating function (Jensen 249-257). the last ship leaves titanic town 5352380.html?fbclid=IwAR10rF8YnK6WNKW4issZ2ZdXZ8OYcKDGFGIaAUVIYBa8oYTf0PGOrN-4ZPc. 221 Jensen identifies the five functions of silence in an interpersonal context: connecting-distancing function, perlocutionary effect, revelational function, judgemental function and activating function (Jensen 249-257). y debates. See www.independent.co.uk/voices/commentators/david-mckittrick-not-everyone-is-weeping-a ast-ship-leaves-titanic-town- p g g perlocutionary effect, revelational function, judgemental function and activating function (Jensen 249-2 many debates. See www.independ the-last-ship-leaves-titanic-town- perlocutionary effect, revelational function, judgemental function and activating function (Jensen 249 2 222 The scale and impact of sectarian discrimination in the shipbuilding industry of Ulster has been the s many debates. See www.independent.co.uk/voices/commentators/david-mckittrick-not-everyone-is-we p y y f ( ) g for instance, which are the canonical examples of the genre of confessional poetry (SS 147). 223 Heaney later commented that “[t]hose little poems in ‘Summer Home’ come more from pressure of personal experience than any literary influence” of Lowell ‘s Life Studies (1959) or Snodgrass’s Heart’s Needle (1959), As you plaited the harvest bow It can heal past wounds and bind people or just hurt more and sever relationships (Jensen 149-52).221 ‘Docker’− amongst Heaney’s first “achieved” poems as a student, reprinted in Death of a Naturalist − uses silence to illustrate the uncompromising sectarian prejudice dominating the recruitment policies of the mid-twenties in Belfast (SS 67).222 The poem describes a Protestant man sitting alone and silent at the corner of a pub, refusing to interact with the rest of the − Catholic − dockers who are involved in a conversation. His tight-shut mouth and uncommunicative attitude, as well as his 213 irritated facial expression and body language, speak of his intentional silence and attempt to isolate himself from the rest whom he despises. Heaney’s chooses the vocabulary of heavy and rough equipment (i.e. “gantry”, “jaw”, “vice” and “hammer”) to convey a sense of hostility and resentment. Even the lines are clotted with thick and hard sounds (i.e. plosives, fricatives, sibilants and velars), making it difficult to manage all the consonants in sequence. It’s all hard-edges and metallic: 223 Heaney later commented that “[t]hose little poems in ‘Summer Home’ come more from pressure of 223 Heaney later commented that “[t]hose little poems in ‘Summer Home’ come more from pressure of personal experience than any literary influence” of Lowell ‘s Life Studies (1959) or Snodgrass’s Heart’s Needle (1959), for instance, which are the canonical examples of the genre of confessional poetry (SS 147). 3 Heaney later commented that “[t]hose little poems in ‘Summer Home’ come more from pressure of per xperience than any literary influence” of Lowell ‘s Life Studies (1959) or Snodgrass’s Heart’s Needle (1 It’s all hard-edges and metallic: There, in the corner, staring at his drink. There, in the corner, staring at his drink. The cap juts like a gantry’s crossbeam, Cowling plated forehead and sledgehead jaw. Speech is clamped in the lips’ vice. Speech is clamped in the lips’ vice. That fist would drop a hammer on a Catholic […]. (DN 28) That fist would drop a hammer on a Catholic […]. (DN 28) In a sequence of five short poems in ‘Summer Home’, from Wintering Out, the remorseful poet narrates what can be described as an “unmusical” period in their marriage, when silence between the poet and his wife has “loosened” the tie between them (WO 48).223 The poem is based on an account of a domestic tribulation − during a 1969 holiday in France, when the damp weather, hot nights and the foul odour of the rubbish and the maggots in the cottage have been aggravating the family’s discontent (see also 3.4). The poem begins by trying to locate the source of the bitter silence that prevails at the summer home rejecting the promises of the title in the opening line: “summer [has] gone sour” (47). Like “something in heat” or “a fouled nest incubating” in a hidden place (47), the void between the couple is quietly giving way to more and more agony and unhappiness. The rest of the poem shows the couple’s attempts to communicate their feelings and “anoint the wound” (48). But the persistence of silence indicates that the gap between the couple “begin[s] / to open and split // ahead” (48). In the first attempt, the husband and wife lie in the bed, under a warm blanket hoping that the physical proximity will bring their spirits closer. But they are paralyzed by the 214 “cold” and sharp “blade” of estrangement that keeps them apart (48). In a second futile attempt, in the final part of the poem, the poet and his wife “lie stiff till dawn” on their bed, unable to talk their way to reunion and reconciliation (48). In The Magic Mountain (1927), Thomas Mann famously notes that in a society where “speech is civilisation itself […] it is silence which isolates” (Mann 518). But in the same context, silence, this instrument of isolation, can also create a sense of connection and community. The deliberate observation of silence in memorials, funerals and during worship are the examples of silence as a positive and connecting element (Jensen, 1973; Johannesen, 1974; Schafer, 1977). Heaney brings up these unifying and positive aspects of silence in several instances in his poetry. The silence between Heaney and his mother in “When all the others were away at Mass” has a connecting and unifying function (see also 4.1). That fist would drop a hammer on a Catholic […]. (DN 28) Later, at his dying mother’s bedside, while the parish priest is vehemently saying the prayers and some of the other people are repeating and crying, the poet decides to remain silent. At these last minutes, he is reminded of the day when he shared a moment of emotional closeness with his mother in the kitchen while peeling the potatoes. The mother and son once again resort to silence to unite for one last time: So while the parish priest at her bedside Went hammer and tongs at the prayers for the dying And some were responding and some crying I remembered her head bent towards my head, Her breath in mine, our fluent dipping knives − Never closer the whole rest of our lives. (HL 29) Her breath in mine, our fluent dipping knives − Never closer the whole rest of our lives. (HL 29) The silence of the man in ‘Mid-Term Break’ who only speak in whispers (“Whispers informed strangers I was the eldest”) and limit their verbal communication to brief expressions of condolences (“tell me they were ‘sorry for my trouble’”) is an attempt to empathise with the Heaney family who had just lost the young Christopher in a car accident (DN 15). In ‘Funeral Rites’, similarly, Heaney juxtaposes the orderliness and sombre silence of the traditional Irish Catholic ceremonies and the sense of courtesy he has for the deceased 215 (“I shouldered a kind of manhood”, “I knelt courteously / admiring it all”) comparing it with the rumbling and roaring of the cars in the contemporary funerals in the second section of the poem (N 6-9) (see also 3.3). In ‘Casualty’ too, Heaney elegises the death of the fisherman killed during Bloody Sunday and describes the funeral attendants as being quiet: “[t]hose quiet walkers” (FW 17). In Wintering Out, the positive and connecting function of silence in an interpersonal context is illustrated in the first and final part of ‘The Other Side’. Heaney describes his family and their Catholic neighbours as being “both beside us and on the other side” of the dividing social lines (SS 132). While sectarian divisions of Northern Ireland were reaching their peak during the 1970s, in ‘The Other Side’ the poet looks back at the more stable bond connecting his Catholic family to their Protestant neighbours in the 1940s and presents silence as a unifying element. That fist would drop a hammer on a Catholic […]. (DN 28) The first section of the poem opens to Mr. Johnny Junkin’s fast and fierce footsteps accompanied by his unkind critique of Heaney family’s management of the property, which the young poet shrugs off: “my ear swallowing / his fabulous, biblical dismissal, / that tongue of chosen people” (WO 24). While the young Heaney’s disagreement with him is evident from his ironic language, his tolerance and silence in response can be seen as an attempt to bear opposing views and to encourage a sense of community. In the final section, this attempt is made by both sides. This time the white-haired man appears very considerate of his neighbour’s private space. He approaches the Heaney household quietly, waits for them to complete their prayer and only then places a soft knock on the door. The gentle distraction is warmly welcomed by the bored youngster, who mutually desires “rapprochement” (Parker 101). Heaney stands behind the old man, in the dark, but is unable to articulate a single word or make another move. The old man, on the other hand, has reached a moment of self-awareness in which he has a feeling of closeness, empathy and unity with the person from who he was once distant. Heaney faces a dilemma. He wonders 216 whether he should escape the situation (“slip away”) or approach the man touching him on the shoulder (“go up and touch his shoulder”) and moderate the communication with unrelated trivial chatter (“talk about the weather // or the price of grass-seed?”) (WO 26): whether he should escape the situation (“slip away”) or approach the man touching him on the shoulder (“go up and touch his shoulder”) and moderate the communication with unrelated trivial chatter (“talk about the weather // or the price of grass-seed?”) (WO 26): But now I stand behind him in the dark yard, in the moan of prayers. He puts a hand in a pocket or taps a little tune with the blackthorn shyly, as if he were party to lovemaking or a stranger's weeping. Should I slip away, I wonder, or go up and touch his shoulder and talk about the weather or the price of grass-seed? (WO 26) As Parker has put it, the possibility of rapprochement, as the subject of the poem, is illustrated through the encounters between Heaney and a Protestant neighbour. 224 Other examples, in Heaney’s poems, when silence is presented as a way of bonding include: ‘The Sharping Stone’, where Heaney and his wife are described lying down on their backs, “Listening to the rain drip off the trees / And saying nothing” (SL 59). Silence, here, can be interpreted as a way of pondering, daydreaming, listening and a means of bringing the couple closer together. y g g ( ) , , p y p g, y listening and a means of bringing the couple closer together. That fist would drop a hammer on a Catholic […]. (DN 28) ‘The Other Side’, indicates “how centuries of conflict and distrust cannot be easily brushed aside” (101- 103). In the poem, Heaney ponders over the ways of bringing the two sides together. Silence, in this case, emerges as “an effective instrument for inducing profound experiences, and for lowering the barriers between the self and the Other” (Maitland 203). Robert Oliver, who was an authority on public speaking and spent several years researching communication and rhetoric, states that “it is not through speech or acts but through silence that the deepest bonds are cemented” (153). This type of silence is evoked when language fails. It is the silence of ineffability, when one is able to hear the inner voices of the other. Standing on one side of the land, the two neighbours have now crossed the borders of otherness and divisions.224 One of the other main types of silence Heaney manifests in Wintering Out is the silence of subordinates and minorities. A person may be silent in a social interaction, but the 217 source of his silence may not always be found within the silent person himself. Silence can be imposed upon individuals and communities by sociocultural codes, a dominating voice or a controlling presence (Kurzon 1676). An individual is silenced when he or she cannot speak or cannot be heard. The history of colonisation, slavery and patriarchy shows numerous instances where individuals and societies have been silenced, minority groups have been marginalised, slaves have been silenced by their masters, women have been silenced by men in patriarchal societies, children have been silenced by their parents and other adults (see also 1.2). Silence, in these contexts, is used as “a murder weapon” (Maitland 216). It is used as a way to not only suppress individual’s voices and opinions, but also annihilate their identities and personhoods. In his poetry, Heaney explores situations in which silence has been employed as a means of imposing prejudice and power. The poems ‘Servant Boy’, ‘The Last Mummer’, ‘The Tollund Man’, in the first half of the collection, as well as ‘Bye-child’, ‘A Winter’s Tale’, ‘Shore Woman’ and ‘Maighdean Mara’ in the second, explore the many subtle ways in which so-called subordinates responded to prevailing voices and noises. That fist would drop a hammer on a Catholic […]. (DN 28) According to O’Driscoll, ‘Servant Boy’ and ‘The Last Mummer’ are situated in the “shadowy area between folklore and politics”, giving the poet and his readers a glimpse of the remote past while also connecting them to the contemporary socio-political context (SS 130). Both poems critique the absence of voices, names and identities. Yet silence has several other implications in both poems. ‘Servant Boy’ draws on the tales of Ned Thompson, a small, stooped and moustached neighbour of the Heaneys, who visited the poet’s house often and talked about his life as a servant boy. Heaney writes: “[…] And from him I did hear talk about different masters, about sleeping in the loft, about the ones who fed him well and the ones who didn’t, about having to walk for miles home and back on a Sunday, and all that [emphasis in the original]” (SS 130). Heaney mentions that he wrote ‘Servant Boy’ after he returned from Berkeley witnessing the deteriorating circumstances of the Catholic minority in 218 Northern Ireland (SS 121). Vendler comments that through the figure of the servant boy, Heaney reflects the “inferior social status of Catholics in the North” (86). According to Rankin Russell, however, Heaney’s use of the term ‘boy’ in the title hints at slave societies in a broader geographical scope. In the antebellum American South, for instance, the white men commonly called the grown black men ‘boy’ in order to “emasculate them and keep them subjugated” (130). In addition to depriving the slaves of their names and identities, the ruling classes required quiet and obedience to reaffirm the contours of power in the society and feed their dreams of efficiency (M. M. Smith 23-24). In the poem, Heaney refers to this authoritative use of silence in colonial Ireland, by highlighting the young boy’s hushed manners and monotonous routines, as well as by remaining indeterminate about his name and location:225 a jobber among shadows. Old work-whore, slave- blood, who stepped fair-hills under each bidder’s eye […]. (WO 7) In the second stanza from the lines quoted above, Heaney takes the reader to the bustle of the slave market, introducing what Schafer calls the “enigma” of noise (74). In the earlier chapters, I elaborated on the concept of ‘noise’ as unfavourable sounds. 225 See “boy, n.1 and int.” in OED Online. See also Racism Matters (1998) by Wright and Amistad's Orphans (2014) by Lawrance for the application of the terms ‘boy’ and ‘man’ in slave societies. The close relationship between name and identity becomes particularly relevant in the context of sectarian divisions in Ireland. In his play Translations (1980), the Irish playwright Friel deals with the theme of language, name and identity in 19th century rural Ireland. The play is set during the period when Gaelic names were suppressed by the Protestant bureaucracy. The playwright’s first name, Brian, for instance, was registered as Bernard (Clarkson Holstein 1) y p y g p pp y aucracy. The playwright’s first name, Brian, for instance, was registered as Bernard (Clarkson Holstein 1) 5 See “boy, n.1 and int.” in OED Online. See also Racism Matters (1998) by Wright and Amistad's Orp See boy, n.1 and int. in OED Online. See also Racism Matters (1998) by Wright and Amistad s (2014) by Lawrance for the application of the terms ‘boy’ and ‘man’ in slave societies. The close rel That fist would drop a hammer on a Catholic […]. (DN 28) In the section on biophonic sounds, for instance, we saw how Heaney alerts his readers to the impact of noise from agricultural machinery on the distribution and diversity of the bird species (3.2). In the section on vocal sounds, we saw how to the ears of the colonial British, Irish sounded loud 219 and noisy (See 3.4). In these lines, however, the poem suggests that “noise equals power” (Schafer 74). The servant boy is silently and obediently presenting his service so as to catch the eye of bidders and masters, for whom the servant boy is merely a slave, a “jobber” of no identity and right to speak (WO 7). His voice is silenced and his identity is obliterated. Maitland likewise comments on the silencing of slaves and minorities during colonisation and writes: “slaves and toys are both possessions, have no rights, no independence, no instrumentality, no true personhood” (216). The terms − “[o]ld work-whore” and “slave- blooded” − indicate the archetypal view of the Irish revealing that the dispossession of Irish people of their voice, identity and dignity is not a recent issue. The silence of slaves, labourers and minorities discussed above are instances of the type of silence imposed by an external force. In some other occasions, silence can be “internal to the silent person, that is to say, the decision not to speak originates within the silent person”. This type includes instances when silence is used as a way of expressing disapproval, contempt or disagreement and can be regarded as “equivalent to a speech act” (Kurzon 1676-1677). The poem ‘Servant Boy’ also manifests this − or the transition to this − type of silence by highlighting the unnamed boy’s wintering-out attitude at the onset of the poem: “He is wintering out / the back-end of a bad year” (WO 7). His hushed manner can also be indicative of the feelings of resentment and disapproval about his subservient position. Berger points out that there are transitional and overlapping phases between different types of silence (173). In the poem, the silenced boy is also intentionally using silence as a way to survive the difficult situation he is living in. Maitland writes: “in the face of oppressive power, silence is often a sound strategy, at least in the short term” (181). That fist would drop a hammer on a Catholic […]. (DN 28) After describing the repressive socio-political environment in which the servant boy lives and works, Heaney brings our attention to the boy’s “patience” and “counsel” (WO 7). If voice and sound are a superior’s instruments of command, silence is the instrument of the inferior. 220 The servant boy is unapologetically resentful about his unfair situation, but he knows the In the poem, the servant boy ‘kept his patience and his counsel’ while he was wintering out: whatever he said, he said nothing; he knew the score, bore the brunt and bided his time. But by the time of the Civil Rights marches his stoop had begun to be straightened and his walk, I would like to think, was being braced by the poem in which he appeared. (SS 130) In his interpretation of the final lines of the poem, Tobin refers to the “implicit connection” (85) Heaney draws between his own poetic work and that of the servant boy portrayed as “[…] resentful / and impenitent, / carrying the warm eggs” (WO 7). According to Tobin, there is certainly a sense of impenitence in the poem too, “an elusiveness that neither ignores the burden of history nor succumbs to mere ideology” (85). The servant boy’s integrity and resilience are precisely the stance the poet himself wishes to maintain and encourage. Heaney is drawn into the boy’s trail and determined to follow in his footsteps: “how / you draw me into / your trail” (WO 7). Despite the wintry condition of his homeland, the poet, like the servant boy of the poem, would prefer to maintain his patience and hope for a better future. At the end of the poem, there is a faint squeak of hinges as the “back” door opens. The servant boy appears in the threshold. He is “resentful”, “impenitent” and tight-lipped but “carrying the warm eggs” (7) into a warm spring, into a new and better year. The servant boy of the poem, as Heaney indicates, is also “a portrait of a minority consciousness” (SS 130) in the socio-political situation of Ireland: ‘carrying the warm eggs’ is what a servant boy would have had to do in the morning, check the nests and bring in what the laying hens had laid; but it’s also an emblem of the human call to be more than just ‘resentful and impenitent’, even while injustices are being endured. That fist would drop a hammer on a Catholic […]. (DN 28) (SS 130) ‘carrying the warm eggs’ is what a servant boy would have had to do in the morning, check the nests and bring in what the laying hens had laid; but it’s also an emblem of the human call to be more than just ‘resentful and impenitent’, even while injustices are being endured. (SS 130) ‘The Last Mummer’ is seen as “the more complex companion piece” to ‘Servant Boy’ (Parker 97). The last mummer is silent, but we are not told yet if his silence is out of resentment and resistance or compliance and conformity. In a flashback, the poet unwraps the mummer’s years of silence: “trammelled / in the taboos” of his community, the mummer has 221 learned to use a “whoring” tongue and “don manners” (WO 8-9). Like the servant boy, he too has lived the life of an invisible shadow “picking a nice way” among masters (8-9). Heaney himself has stated the analogy between the two characters: Like the servant boy, the last mummer too is “resentful and impenitent” (SS 130). Like the servant boy, the last mummer appears as the alter ego of the poet himself, walking through “the long toils of blood // and feuding” (WO 8-9) but avoiding to accept the situation.226 As Tobin has justly noted, the mummer can be viewed as “an exemplary figure not only for the poet, but for anyone who retains their humanity as well as their imagination in oppressive circumstances” (85). Earlier in this section, I noted that silence can be imposed by external social norms and political rules. Yet, as Kurzon notes, silence can also be caused by “internalised” codes and norms (1678). This type of silence − or silencing − is illustrated in ‘The Tollund Man’. In his book, Glob suggests that the Tollund man was a priest who volunteered to be offered as a sacrifice to the goddess of fertility during a winter festival. He writes that the man was found with eyes “slightly closed, lips softly pursed, as if in silent prayer” (1).227 In other words, he implies that the Tollund man’s silence was internal and intentional. Even so the man’s voice was silenced. The voluntary silence and death of the Tollund man was perhaps a result of internalising religious rituals and social norms. He carries a stone in his pocket. He beats the bars of a gate with his stick. He fires a stone up at the roof of ‘the little barons’. But he also controls his anger, fits in, his tongue goes ‘whoring / among the civil tongues’. In a sense, he’s the kind of guy who can be a spokesman, can go on a BBC panel without wrecking the decorum of the studio […]. (SS 130) g [ ] ( ) 227 Glob writes “It was on just such occasions that bloody human sacrifices reached a peak in the Iron Age” (57). However, the distinction between other discovered Iron Age bodies and that of the Tollund man was that the other victims carried signs of extremely violent deaths: “This one bore no sign of violence other than the mark of the noose” (57). 226 In ‘Servant Boy’ and ‘The Last Mummer’, Heaney employs a persona to examine his position as a poet. He writes: 228 The poem refers to the recruitment policies of the 50s and 60s in Belfast, according to which Catholics would be casual laborers with the exhausting, ill-paid jobs, whereas the Protestants would hold the well-paying, high- skilled jobs (Crowder 117-118). See Wilson, Vendler and Crowder (2005), for instance, for discussing sectarianism as the main theme of the poem. That fist would drop a hammer on a Catholic […]. (DN 28) His household is, however, prepared for and used to his demands of silence and compliance: He sits, strong and blunt as a Celtic cross, Clearly used to silence and an armchair: 228 Earlier I brought up examples from the history of slavery, when silence was used as a means of imposing authority and exploiting individuals. This authoritative use of silence has also been apparent in relation to women and children, mainly for the purpose of reassuring the superior status of men (see 1.2). Isenberg remarks that, within the working class, for instance, “Irish males saw the ideal women as dutifully subordinate, submissive, and obedient to the natural authority of men. Indeed, Irish women never challenged male public hegemony” (47). In her overview of women’s condition in nineteenth century Ireland, Luddy maintains “women were also expected to acquire those virtues of gentility, sobriety, passivity, and humility” (3). Heaney’s poem ‘Docker’ provides an early example of the use of silence in power relations in domestic contexts. Critics have often viewed the poem in the light of the Troubles in Northern Ireland, leaving out Heaney’s equally prominent illustration of the oppressive masculine power in the final stanza.228 According to Crowder, for instance, the poem begins in the form of a dramatic monologue in which the speaker − a Catholic docker, an unskilled labourer, sitting in a pub and drinking with a mate − directs the attention to a Protestant shipbuilder, who, as I discussed earlier in this section, uses silence as a way of distancing himself from Catholic men (117-118). In the remaining lines of the poem, Heaney illustrates the outcome of sectarianism portraying the privileged Protestant man from the Catholic docker’s view as an uncommunicative man (“speech is clamped in the lips’ vice”), who is filled with hatred towards his Catholic fellow workers (“that fist would drop a hammer on a Catholic”), as well as a person who abuses and intimidates his wife and children when he returns home. That fist would drop a hammer on a Catholic […]. (DN 28) Heaney imagines that while driving on the roads of Aarhus, “[s]omething of his [Tollund man’s] sad freedom” will reach him as familiar (WO 37). The thoughts of Tollund man remind the poet of the voices of his fellow Irish men silenced by religious and political prejudice of their time. “In the old man-killing parishes” of Jutland, the Ulster poet will feel “lost, / Unhappy and at home” (37). 222 Earlier I brought up examples from the history of slavery, when silence was used as a means of imposing authority and exploiting individuals. This authoritative use of silence has also been apparent in relation to women and children, mainly for the purpose of reassuring the superior status of men (see 1.2). Isenberg remarks that, within the working class, for instance, “Irish males saw the ideal women as dutifully subordinate, submissive, and obedient to the natural authority of men. Indeed, Irish women never challenged male public hegemony” (47). In her overview of women’s condition in nineteenth century Ireland, Luddy maintains “women were also expected to acquire those virtues of gentility, sobriety, passivity, and humility” (3). Heaney’s poem ‘Docker’ provides an early example of the use of silence in power relations in domestic contexts. Critics have often viewed the poem in the light of the Troubles in Northern Ireland, leaving out Heaney’s equally prominent illustration of the oppressive masculine power in the final stanza.228 According to Crowder, for instance, the poem begins in the form of a dramatic monologue in which the speaker − a Catholic docker, an unskilled labourer, sitting in a pub and drinking with a mate − directs the attention to a Protestant shipbuilder, who, as I discussed earlier in this section, uses silence as a way of distancing himself from Catholic men (117-118). In the remaining lines of the poem, Heaney illustrates the outcome of sectarianism portraying the privileged Protestant man from the Catholic docker’s view as an uncommunicative man (“speech is clamped in the lips’ vice”), who is filled with hatred towards his Catholic fellow workers (“that fist would drop a hammer on a Catholic”), as well as a person who abuses and intimidates his wife and children when he returns home. That fist would drop a hammer on a Catholic […]. (DN 28) His household is, however, prepared for and used to his demands of silence and compliance: He sits, strong and blunt as a Celtic cross, Clearly used to silence and an armchair: He sits, strong and blunt as a Celtic cross, Clearly used to silence and an armchair: He sits, strong and blunt as a Celtic cross, Clearly used to silence and an armchair: 228 The poem refers to the recruitment policies of the 50s and 60s in Belfast, according to which Catholics would be casual laborers with the exhausting, ill-paid jobs, whereas the Protestants would hold the well-paying, high- skilled jobs (Crowder 117-118). See Wilson, Vendler and Crowder (2005), for instance, for discussing sectarianism as the main theme of the poem. 223 Tonight the wife and children will be quiet At slammed door and smoker's cough in the hall. (DN 28) Despite its more “personal” and “immediate” concerns (Foster 41), part II of Wintering Out continues to display “history’s plague on the poet’s imagination” (Tobin 96). Heaney explains that in 1969 and 1970, before heading to California in 1971, he had composed a sequence of poems about “women in distress” (SS 124). These poems, including ‘Shore Woman’, ‘Maighdean Mara’, ‘A Winter’s Tale’, ‘Limbo’ and ‘Bye-child’ among others, portray women driven by circumstances into peripheral positions, silence, madness and ultimately suicide. ‘A Winter’s Tale’ is based on the story of a young mad girl from Ulster who would often run away unclothed from her family and village people into the wood (see 3.4). Initially, the young girl’s daily escapes into the countryside call to mind the free spirit of the little boy in ‘Oracle’. But, instead of a tranquil and delightful soundscape, the poem soon introduces the silence and darkness of a winter night. The silence and stillness of wintertime is “oppressive” (Schafer 20). It manifests the absence of sound, life and liveliness. In the stillness of winter, it is normally a relief to hear a sound, see a track of footprints and feel that there is still life around. But the girl flees further and further away precisely from the approaching sounds of men searching for her. Heaney shows that the girl is “[a] pallor” anguished by the strong light of men’s “lanterns”, “torches” and “headlights” (WO 52). She is only a faint “weeping” masked by “the searchers’ gay babble” (WO 52). Tonight the wife and children will be quiet That fist would drop a hammer on a Catholic […]. (DN 28) Like the servant boy of the previous poem, the girl is merely a wavering existence in her village. The young girl of the poem could, in fact, stand for every man and woman who has experienced otherness but, given the space, could promote the sense of community and solidarity in a society. In the context of the political, religious and social segregation, as Heaney suggests, the mad girl proves to be the best example to learn from. Some cold winter nights, she enters empty neighbour houses to rest by chimneys. As Tobin points out, “her 224 madness, like Sweeney’s, enables her to transgress sectarian boundaries” without hesitation (96). The girl’s “unmalicious” companionship was considered intruding or invasive by house owners, whom she warmly greeted. “After all”, Heaney says, “they were neighbours” (WO, 52). Perhaps Heaney wished that one day, all Ulster would be able to steer away from estrangement and segregation as swiftly, to appear “no haunter” to one another, to cross walls of divisions, to make “all comers guests” and to be the “first” one to offer peace to their “own people” (WO 52-53). madness, like Sweeney’s, enables her to transgress sectarian boundaries” without hesitation (96). The girl’s “unmalicious” companionship was considered intruding or invasive by house owners, whom she warmly greeted. “After all”, Heaney says, “they were neighbours” (WO, 52). Perhaps Heaney wished that one day, all Ulster would be able to steer away from estrangement and segregation as swiftly, to appear “no haunter” to one another, to cross walls of divisions, to make “all comers guests” and to be the “first” one to offer peace to their “own people” (WO 52-53). ‘Shore Woman’ is based on a series of incidents in Kerry, either experienced by Heaney personally or narrated by a fisherman in Dingle (SS 124).229 Earlier, I elaborated on how sounds − geophonic, biophonic and anthrophonic − can reveal the relationship between the fisherman, his wife and the natural environment (see 3.1 and 3.2). Here, I consider the way the poem foregrounds the theme of women’s silence over centuries. On the fisherman’s boat the woman is silent, in the sense that her voice is ignored, her feelings of joy and fear are disregarded, and she is looked down upon as careless, idle and ignorant. In the fisherman’s world, the woman is silenced. 229 Heaney explains that these poems were written in 1969 and 1970, about the same time as ‘The Tollund Man’. The poet and his wife had frequently visited the place much earlier during 1960. Therefore “the fishing scenario in ‘Shore Woman’ is in fact an amalgam of two Kerry occasions – one when we went out on Kenmare Bay with Sean O Riada and actually caught a heap of mackerel; another, reported to us by a man in Dingle, who told how his wife panicked when their boat was surrounded by porpoises. Anyhow, those narratives and monologues formed one segment of the contents” (SS 124). (Almqvist 20). 231 See www.theguardian.com/news/2002/feb/04/guardianobituaries.books1. 230 The Northern Irish folklorist’ name is, as Almqvist (1990) has put it, more or less synonymous with Donegal in Northern Ireland. Seán Ó hEochaidh was born and bred in Teelin and did most of his collecting in the adjoining areas of Donegal. Heaney met him when he was an undergraduate at Queen’s University in Belfast. The folklorist impressed the young poet with his “stylised and formal manner” of narration, the “unusual” and “slightly archaic” vocabulary, and the link between the “mermaid legend” and his family intrigued him (Almqvist 20). That fist would drop a hammer on a Catholic […]. (DN 28) After returning to the shore, she resentfully remembers this “failure” of love, respect and communication between her and the husband (Foster 42): […] I called ‘This [catching fish] is so easy that it’s hardly right,’ But he unhooked and coped with frantic fish Without speaking (WO 54) […] I called ‘This [catching fish] is so easy that it’s hardly right,’ B t h h k d d d ith f ti fi h […] I called ‘This [catching fish] is so easy that it’s hardly right,’ [ ] ‘This [catching fish] is so easy that it’s hardly right,’ [ g ] y y g But he unhooked and coped with frantic fish g y y g But he unhooked and coped with frantic fish Without speaking. (WO 54) As in ‘The Wife’s Tale’, “the poem’s central incident belongs to the male domain” (Andrews 70). The fisherman “enjoys an attitude of superiority” over his wife as over nature. As in ‘The Wife’s Tale’, “the poem’s central incident belongs to the male domain” (Andrews 70). The fisherman “enjoys an attitude of superiority” over his wife as over nature. Later in the journey, when she cries expressing her apprehension about the porpoises’ attack Later in the journey, when she cries expressing her apprehension about the porpoises’ attack and requests her husband to call it a day, the man “scathingly dismisses her pronouncement” 225 (Andrews 70): “I knew it and I asked him to put in / But he would not, declared it was a yarn […]” (WO 55). The fisherman brutally ignores her safety and comfort. Heaney also refers to the wider scale and more ancient roots of the split. The boatman’s declaration does not address his wife only. He thinks the woman’s opinion is “a yarn” that has fooled her and her “people” for “far too long” (WO 55). For her, the “huge” sea creature that is now slithering its “oily” back against the boat is the embodiment of the “huge” masculine egoism, pleasure- seeking and intrusive attitude of the man, which she now recalls with resentment: “Sick at their huge pleasures in the water” (WO 55). The acoustics of the marine world weighs too much on her inner freedom, safety and solitude. The woman belongs to the silence and privacy of the shore: “I have rights on this fallow avenue” (WO 55). 230 The Northern Irish folklorist’ name is, as Almqvist (1990) has put it, more or less synonymous with The Northern Irish folklorist name is, as Almqvist (1990) has put it, more or less synonymous with in Northern Ireland. Seán Ó hEochaidh was born and bred in Teelin and did most of his collectin 0 The Northern Irish folklorist’ name is, as Almqvist (1990) has put it, more or less synonymous with Do n Northern Ireland. Seán Ó hEochaidh was born and bred in Teelin and did most of his collecting That fist would drop a hammer on a Catholic […]. (DN 28) (Andrews 70): “I knew it and I asked him to put in / But he would not, declared it was a yarn […]” (WO 55). The fisherman brutally ignores her safety and comfort. Heaney also refers to the wider scale and more ancient roots of the split. The boatman’s declaration does not address his wife only. He thinks the woman’s opinion is “a yarn” that has fooled her and her “people” for “far too long” (WO 55). For her, the “huge” sea creature that is now slithering its “oily” back against the boat is the embodiment of the “huge” masculine egoism, pleasure- seeking and intrusive attitude of the man, which she now recalls with resentment: “Sick at their huge pleasures in the water” (WO 55). The acoustics of the marine world weighs too much on her inner freedom, safety and solitude. The woman belongs to the silence and privacy of the shore: “I have rights on this fallow avenue” (WO 55). Like ‘Shore Woman’, ‘Maighdean Mara’ is rooted in Irish folklore and real-life predicaments surrounding women characters. The poem is dedicated to the well-known folklorist of Northern Ireland, Seán Ó hEochaidh, who was himself known to be a “committed and good listener” to the local storytellers230 and who had shared some of his stories with the poet on a number of occasions.231 The reports of a real-life drowning incident in Northern Ireland prompts the Ulster poet to turn himself into a patient listener of untold stories of woman suffering. Heaney was aware of − particularly north western − Irish mythologies wherein the female selkie’s garment is stolen while she is outside the sea and hidden by a human man who takes her as a wife and mother of his children (Almqvist 16-20). The folktales say that mermaids make very good wives and mothers, but Heaney’s profound listening enables him to see the dark side of the story and give voice to silent stories and silent individuals in them. Heaney turns to the pre-historic woman’s silence to narrate 226 unending feminine “repression”, “isolation”, “exploitation” and “disenchantment” (Corcoran 30). The silence of the woman in ‘Maighdean Mara’ closely resembles that of the fisher’s wife in the sense that the men she encounters in her earthly life expect her to be an obedient follower. In her “land” life, she lived the life of a prisoner. That fist would drop a hammer on a Catholic […]. (DN 28) Deprived of her identity and unable to return to the sea, the mermaid “had no choice” but to suppress her voice and bury it in a dark silence. After this incident, Heaney says, “follow / Was all she could do” (WO 56- 57). The sea creature lived “the dead hold of [men’s] bedrooms” and “suffered” labour pain and suckling. She was a stranger among other women whose “forever and uncharmed” judgemental voices further buried her in silence (57). Maighdean Mara does not belong to the land, but without her selkie skin, she cannot return to the sea either. Maitland says that, in fairy tales, “there is also the silence of renunciation, often of penance” (180), one that leads to new knowledge and more wisdom. The woman of the story is consigned to the silence of death as a consequence of masculine lust and social prejudice. She sleeps now in a cradle of sea waves. A type of silence closely related to speechlessness is what Kurzon calls “thematic silence” (1677). While the last mummer, the servant boy, the Protestant ship-maker and his wife and children, in the examples discussed above, did not participate in a conversational exchange − voluntarily or involuntarily − in thematic silence “a person when speaking does not relate to a topic” (1677). This type of silence can occur in a dialogical context, as well as in a broader social and political context or even written texts in which the speaker/writer deliberately avoids a topic. The poems ‘A Winter’s Tale’ and ‘Maighdean Mara’ discussed previously are unequivocal examples of thematic silence in gender relations. The opening epigraph of ‘Shore Woman’, which is the English translation of a Gaelic proverb, grounds the poem in the dominant local mentality: “man to the hills” and, as the 227 poem later indicates, man to the sea too, but woman to the margins, “woman to the shore” (WO 54). The man has internalised the cultural norm. The saying is, therefore, an extract from the fisherman’s “stock of proverbial wisdom” (Andrews 69). The fisherman is fearless, cold-blooded, self-absorbed and “all business in the stern” (WO 54). Like Heaney’s father, his silence seems to stem from his distrust of speech and language. But the saying suggests the presence of another type of silence. On the fisherman’s boat, there is no room for such topics as peace, delight or romance. 232 In southern versions, particularly from Kerry and Limerick, it is the husband himself who, in his frantic search for a farming tool, unwittingly provides his wife with an opportunity to find her hidden garment and return to the ocean (Almqvist 1-43). 233 Commenting on the relationship between politics and poetry with Seamus Deane, Heaney asserted: Poetry is born out of the watermarks and colourings of the self. But that self in some ways takes its spiritual pulse from the inward spiritual structure of the community to which it belongs; and the community to which I belong is Catholic and nationalist. I believe that the poet's force now, and hopefully in the future, is to maintain the efficacy of his own "mythos", his own cultural and political colourings, rather than to serve any particular momentary strategy that his political leaders, his paramilitary organisation or his liberal self might want him to serve. I think that poetry and politics are, in different ways, an articulation, an ordering, a giving of form to inchoate pieties, prejudices, world- views, or whatever. And I think that my own poetry is a kind of slow, obstinate, papish burn, emanating from the ground I was brought up on. (62) y p 235 Referring to an iconic 1970s Republican poster titled “Loose Talk”, illustrated with the photo of a gunman. The text began with “Loose talk costs lives” and concluded with a line written in red: “whatever you say, say nothing!”. The poster discouraged open political-religious speech particularly in Republican areas and warned about the consequences of ignoring the advice. See cain.ulster.ac.uk/images/posters/ira/index.html. g g g p ( ) 234 For other discussions on Heaney’s involvement with politics see Fennell and Vendler. 233 Commenting on the relationship between politics and poetry with Seamus Deane, Heaney asserted: That fist would drop a hammer on a Catholic […]. (DN 28) There, the cold-blooded creatures have a doomed fate: their “[s]tiff flails of cold [are] convulsed with their first breath” (WO 54). On his boat, the woman’s calls of joy and fear are always ignored. The fisherman does not talk with her calmly and lovingly, but he does yell: ‘“Count them up at your end,’ was all he said” (WO 54). The silence of the men in ‘Maighdean Mara’ is similar to the fisher’s in the sense that it is thematic, but different in the topic: men in ‘Maighdean Mara’ are hiding a “secret” − i.e. the woman’s true identity (WO 57). In most versions of the story coming out of the North West of Ireland, the mermaid’s tragic separation from her husband and return to the sea is brought about unwittingly by one of their small children.232 Driven by their “man-love” and lust, the two men the mermaid encounters in her earthly life continue to deprive her of her independence and happiness, by hiding her “magic garment” and preventing her from returning to her original home and identity (WO 56). The last mummer’s silence, who was “trammelled / in the taboos of the country // picking a nice way through / the long toils of blood” (8) is also an instance of thematic silence, on which Heaney further elaborates in ‘Whatever You Say Say Nothing’. Ulster has been officially a British and predominantly Protestant province since the early 17th century. The division between the Catholic nationalist minority and the Protestant unionist majority had widened − even after the civil rights movement and towards the end of the Troubles − to 228 the extent that pronouncing one’s identity and political beliefs could lead to verbal or even physical violence. In this context, as Rankin Russell has noted, “the presence of a significant minority in Northern Ireland that suffered widespread discrimination over the course of the Stormont regime from 1921 to 1971 was occluded and often rendered silent” (165). Coming from a religiously and politically minority background, Heaney was aware of the Northern Irish reticence.233 In fact, his preoccupation with historically known taciturnity has been known to scholars. In his early influential book, Morrison argues that the community Heaney came from, and which he wanted his poetry to express solidarity with, was one on which the pressure of silence weighed heavily. p ( ) 237 The title of this section is inspired by a line from ‘The Tollund Man’ (WO 36-37). ( ), y g p his feelings towards the constable, is another instance of thematic silence. ‘To George Seferis in the Underworld’ presents an instance when the poet ultimately overcomes his “contested silence” and make a public A Constable Calls’ (N 61-62), in which Heaney’s father is not telling the full truth about the crops and abo Underworld presents an instance when the poet ultimately overcomes his contested silence statement about the political Greek situation (DC 21). That fist would drop a hammer on a Catholic […]. (DN 28) (N 54) In his poignant ode to Northern Ireland, he describes it as a “land of password”, a place in which the inhabitants communicate through “smoke-signals” and “whispering morse” (N 54- 55). In the poem, as Hart notes in Poet of Contrary Progressions (1993), Heaney celebrates Ulster’s silence to “underscore solidarity with his Irish Catholic ancestors and peers” (38) and, in doing so, he awakens their voice and presence: Oh, land of password, handgrip, wink and nod, Of open minds as open as a trap, Where tongues lie coiled, as under flames lie wicks, Where half of us, as in a wooden horse Were cabin’d and confined like wily Greeks, Besieged within the siege, whispering morse. (N 55) 236 4.3 I Might Tarry: Silence and Time237 In his poignant ode to Northern Ireland, he describes it as a “land of password”, a place in which the inhabitants communicate through “smoke-signals” and “whispering morse” (N 54- 55). In the poem, as Hart notes in Poet of Contrary Progressions (1993), Heaney celebrates Ulster’s silence to “underscore solidarity with his Irish Catholic ancestors and peers” (38) and, in doing so, he awakens their voice and presence: Underworld presents an instance when the poet ultimately ov statement about the political Greek situation (DC 21). That fist would drop a hammer on a Catholic […]. (DN 28) It was not only rural, renowned like all rural communities for its inwardness and reserve, but also Northern Catholic, with additional reasons for clamming up. (23)234 the community Heaney came from, and which he wanted his poetry to express solidarity with, was one on which the pressure of silence weighed heavily. It was not only rural, renowned like all rural communities for its inwardness and reserve, but also Northern Catholic, with additional reasons for clamming up. (23)234 In ‘Whatever You Say Say Nothing’, Heaney attempts to “speak truth into the tired cliché” that had governed everyday speech in Northern Ireland (Rankin Russell 165).235 The poem is set in the political reality of the 1970s, when “bad news is no longer news” and when at the street level people exclude religious and political topics or avoid favouring a side: “‘Religion’s never mentioned here,’ of course. / ‘You know them by their eyes,’ and hold your tongue. / ‘One side’s as bad as the other,’ never worse” (N 54). Heaney condemns “[t]he famous // Northern reticence” and “the tight gag” placed on the mouth of the minority group (N 54). He believes it is time that the minority group − to which he himself belongs − speaks out its thoughts, raises its voice and resists the oppression imposed on it by the Protestant In ‘Whatever You Say Say Nothing’, Heaney attempts to “speak truth into the tired cliché” 229 majority, yet almost immediately realises that despite possessing the powers of imagination and expression as a poet, he himself is “incapable” of doing so (N 54). Instead, like the last majority, yet almost immediately realises that despite possessing the powers of imagination and expression as a poet, he himself is “incapable” of doing so (N 54). Instead, like the last mummer, he must hold his tongue and save his face: mmer, he must hold his tongue and save his face: Yet for all this art and sedentary trade I am incapable. The famous Yet for all this art and sedentary trade I am incapable. The famous Yet for all this art and sedentary trade I am incapable. The famous Northern reticence, the tight gag of place And times: yes, yes. Of the “wee six” I sing Where to be saved you only must save face And whatever you say, you say nothing. 4.3 I Might Tarry: Silence and Time237 The silences experienced by individuals can also be varied according to their length. In this section, I focus on the interpretation of silence as a pause (see 1.4). In everyday life, there are several short periods when a pause or a stoppage in an activity creates a little space for, among others, resting, reflecting, listening carefully or preparing to re-start. As Maitland has remarked, “language has lots of vocabulary for different small silences […]” (185). 230 Pause, stop, stand, rest, halt, caesura, hesitation, lacuna, delay, break, lull, interval, wait and stay are some of the terms we might use to describe our experience of small silences, moments of pause over the course of an acoustic event or in between the succession of acoustic events (see 1.4.1). In this section, I trace and interpret Heaney’s manifestations of silence as a pause, and explain how it enhances our understanding of his themes. In ‘Digging’, Heaney provides an early example of a very short pause when recalling the day he took a bottle of milk for his grandfather, who was working in the peat bogs and stopped for only a few minutes of break: “[…] He straightened up / To drink it, then fell to right away” (DN 1). In fact, perhaps the entire poem portrays one protracted pause: the poet now sitting at the window, probably at a desk, writing, stops suddenly − to listen to the rhythmical digging sound coming from outside, to reflect on his father aging, to dig out the memories of his fathers, to re-assess his own career choice − and then, resumes writing with even more determination and motivation. As the opening lines of the poem imply: “Between my finger and my thumb / The squat pen rests; snug as a gun” (DN 1). In ‘A Call’ too, Heaney places the entire poem in a lengthy pause. Heaney’s silent wait, though not mentioned directly, is evoked by the image of the unattended phone, his envisioning of the pendulums, and the ticking of clocks, and, as noted earlier, through the recurring –ing words (i.e. “touching”, “inspecting”, “separating”, “listening”, “ticking”), which create the effect of a momentary stasis. 4.3 I Might Tarry: Silence and Time237 As Kay has remarked “this eerily liminal moment” (59) provides him with the space wherein he could not only reflect on his childhood memories and the aging of his father, but also take an anachronistic leap to the medieval morality play and reflect on the shared fate of mankind: morality play and reflect on the shared fate of mankind: 231 Then found myself listening to The amplified grave ticking of hall clocks Where the phone lay unattended in a calm Of mirror glass and sunstruck pendulums ... And found myself then thinking: if it were nowadays, This is how Death would summon Everyman. (SL 53) In defining pause, the OED makes it clear that a pause − whether as a break or rest − is made “for effect”. The poem ‘The Blackbird of Glanmore’ opens with the scene of the aging poet driving to the Glanmore cottage (see 3.2). The sight of a much-loved blackbird calls to his mind the soundscape of his childhood and prompts the poet to stop the car: And found myself then thinking: if it were nowadays, This is how Death would summon Everyman. (SL 53) In defining pause, the OED makes it clear that a pause − whether as a break or rest − is made “for effect”. The poem ‘The Blackbird of Glanmore’ opens with the scene of the aging poet driving to the Glanmore cottage (see 3.2). The sight of a much-loved blackbird calls to his mind the soundscape of his childhood and prompts the poet to stop the car: I park, pause, take heed. Breathe. Just breathe and sit And lines I once translated Come back: “I want away To the house of death, to my father Under the low clay roof. (DC 75) Heaney pauses as a deliberate attempt to use silence to enhance listening, to re-imagine the soundscape of his childhood where he can see his home, his father, his young brother Christopher, but more importantly as he explicitly mentions in the lines quoted above, to reflect over his own mortality. In ‘The Makings of Music’, he advises himself and his readers to stop and stand still to celebrate the surrounding soundscape: “Stand still. You can hear / everything going on” (SI 52). 4.3 I Might Tarry: Silence and Time237 In the second part of ‘The Other Side’, after approaching Heaney’s house, the old neighbour pauses out of respect to the praying voices coming from inside the kitchen and knocks on the door only after they have finished (see also 4.2). Towards the end of the poem too, Heaney intends to approach the man to welcome him in and perhaps engage in a conversation, but immediately stops because he is hesitant. In the remainder of this section, I discuss how in ‘Bog Oak’, ‘Land’, ‘The Last Mummer’ and ‘The Tollund Man’, in Wintering Out, Heaney uses pause as a way to create a space of silence for listening and reflecting. 232 The strong and rough footsteps of the farmer in ‘Land’ introduce a reassuring rhythm to the quiet soundscape. Not long after describing the everyday routine of farming lifestyle, the poet notifies us that the man is preparing to depart the land and start a new lifestyle (See 3.3). Yet before doing so, he seeks a moment of stillness and silence. After completing his tasks for one last time, the man stands on top of the hill to cast a “last look” at “those acres” (WO 11). Earlier, I explained that in the realm of the visible, the sound and sight could very well overlap (see 1.4.1). While, as readers, we are wondering about the poet’s new destination, new land, and perhaps his new set of “habits” (WO 11), Heaney re-directs our attention back to the sights and acoustics of his “first place” (PO 18). This pause for him is not a timely break, a chance to unwind, a fumbling moment of hesitation, or a pre-arranged platform to build-up readers’ curiosity. This hiatus provides the poet-traveller with an opportunity for stocking up and sealing down his memories of the land.238 Before setting out on his new journey in life, he attempts to instil his farmer’s sensibility in his mind and heart. What started as a brief caesura turns into a prolonged span of silence that stretches over the entire second section of the poem. Heaney has fixed his eyes on a vegetal effigy in front of him. This silence doesn’t represent absence. This is an audible silence that the Irish poet has trained his ears to listen. 238 Referring to Heaney’s 1980 interview with John Edwards, in which Heaney asserts that when he began to write, “it was that first deposit of experience [in Mossbawn farm] that was almost hermetically sealed down came up”. www.youtube.com/watch?v=3yt4m2Z4Pmw (1.53 - 2.14). 4.3 I Might Tarry: Silence and Time237 The sight of his muse, “a woman of old wet leaves” (WO 11), unreels the memories of the people and the materials he knows best. The muse’s tightly plaited strands speak of the old thatcher’s mastery, her ribbons and bows echo the bustle of past harvest feasts, and the “open-work” of her breasts herald the arrival of new harvests and new generations. Schafer writes: “[w]hen it[silence] interrupts or follows sound, it reverberates with the tissue of that which sounded, and this state continues as long as memory holds it” (256). With the vision of a timeless effigy sculpted in his mind, the Ulster poet is ripe for any journey. Sure-footed, he says, “I was ready to go anywhere” (WO 11). 233 ‘The Tollund Man’, introduces a similar sense of suspension that briefly touches the reader in ‘Land’. Both poems follow an analogous pattern of beginning with motion and then a long pause at the sight of a stationary and silent figure. In both poems Heaney brings all the activities to an extended halt in order to provide ample space for memory digging and pondering. Heaney’s encounter with the images of Iron Age people in Glob’s The Bog People (1965) urges the poet to plan a journey “to see” the Iron man’s body in person: “Someday I will go to Aarhus / To see his peat-brown head […]” (WO 36) (see also 2.2 and 4.2). But for a poet who wants to get a profound sense of history a short visit will not suffice. Heaney does not want to merely ‘see’ the man’s body, he wants to ‘listen’ to his silence. The silent and still body of the Iron man now reposing back at Aarhus has a hoard of tales in his chest (WO 36). But to make a profound listening possible he needs to be quiet himself. Heaney makes sure to stay still long enough to hear the resonance of the voices and sounds experienced by the Iron-age man: “I will stand a long time” (WO 36). Death itself is silence, but it is merely a stoppage to the sounds of cries, laughter and weeping, spoken words and soft breathings that continue to reverberate in the atmosphere. Heaney’s profound listening allows him to broaden the horizons of his auditory perception. 4.3 I Might Tarry: Silence and Time237 It allows him to make sense-making possible in a space that includes “the flat country nearby” and a time passed (WO 36). But Heaney’s imagined silence and stillness at the sight of the Tollund man’s corpse has another reason too. In the second part of the poem, Heaney continues to allow himself enough time to be in the silence of death and honour it. This time, in memory of the victims of Irish political and religious rites. Heaney reveals the association with another photograph he had encountered in Barry’s book Guerilla Days in Ireland (1946): “It was of a farmer’s family who had been shot in reprisals by the Black and Tans, left lying on their backs beside their open door” (SS 135). The profound silence of the Tollund man’s dead body carries the observing poet into a deep and prolonged state of contemplation. In Jutland, as in Ulster, 234 history repeats itself. The ritual killings in Jutland during the Iron Age provide a historical simile, a precedent, and a ground to speak about the killings in Northern Ireland: history repeats itself. The ritual killings in Jutland during the Iron Age provide a historical simile, a precedent, and a ground to speak about the killings in Northern Ireland: Out here in Jutland In the old man-killing parishes I will feel lost, Unhappy and at home. (WO 37) Out here in Jutland In the old man-killing parishes I will feel lost, Unhappy and at home. (WO 37) Out here in Jutland In the old man-killing parishes I will feel lost, Unhappy and at home. (WO 37) The poem ‘Bog Oak’ manifests another example of silence as a pause, one that allows him to take a break from the English education system that had contributed to the shaping of the soundscape of his mind and listen to the voices of his Irish ancestors and predecessors. As a poet speaking and writing in the language of the conqueror, Heaney is inevitably part of the English literary tradition − containing Edmund Spenser’s The Faerie Queen (1590) and A View on the Present State of Ireland (1598). In Berkeley, Heaney comes into contact with Thomas Flanagan’s The Irish Novelists 1800-1850 (1959), which includes a quotation from Spenser, who, with no sympathy, describes the survivors of Irish famine as “Anatomies [of] death” and “ghosts” (1959 8). The acquaintance with Flanagan’s Irish-centred thinking retunes his poetic voice. 239 As Corcoran has observed, “As a poet writing in the English language, Heaney is inevitably part of the poetic tradition which contains The Faerie Queen; but ‘Bog Oak’ suggests how tangentially and suspiciously related to it he is when it reminds us that such literary perfections as that great Renaissance poem − written by Spencer ‘dreaming sunlight’ in Kilcolman Castle, his planter’s state in Co. Cork – were the flower of a culture whose roots lay in the brutal political realities described in the State of Ireland (31-32). y p ( ) 240 The title of this section is inspired by Heaney’s poem ‘A New Song’ (WO 23). 4.3 I Might Tarry: Silence and Time237 This moment of pause and reflection is illustrated in ‘Bog Oak’. In his poem, Heaney digs down into the silence of the peat bogs surrounding his childhood home waiting long enough in the Irish past until his ears pick up the faint reverberations left from his subjugated and dispossessed ancestors: or eavesdrop on their hopeless wisdom […]. (WO 4) This revelation is an opportunity he would have missed, had he not sought a pause to meditate and retune his voice. Ironically, Spenser’s trophy – the English language – is now in the possession of the Irish poet who encroaches upon Spenser’s tranquil time at his 235 22):239 geniuses who creep ‘out of every corner of the woodes and glennes’ towards watercress and carrion. (WO 4-5) 4.4 The Vanished Music: Silence and Absence240 In its most conventional sense, silence is often associated with absence (Schafer 256). This is the type of silence that Heaney relates to the ruins of the ancient Roman city in ‘Dawn’ and uses to describe the void and emptiness he feels in the company of his academic friends in a Mediterranean town: “scholars / Arguing through until morning // In a Pompeian silence” (WO 65) (see also 3.4). It is also the type of silence that he associates with the extinction of the native Irish animals and the disappearance of the Irish language and identity in ‘Midnight’ when he says: The wolf has died out // In Ireland” (WO 35), or in ‘Bogland’, when he describes the unearthing of “the skeleton / Of the Great Irish Elk” (DD 43) from the peat bogs (see also 3.2). The poems ‘Traditions’, ‘The Last Mummer’, ‘The Wool Trade’, and ‘New Song’ from Wintering Out are some of the other poems in which Heaney uses silence to mourn the absence of a sound source and its eradication from the soundscape. Earlier, I elaborated on how the poem ‘Traditions’ manifests the poet’s feelings towards the imperialistic presence of English language and culture in Ireland (see 3.4). While doing so, Heaney mourns the disappearance of the “guttural” sounds of Gaelic from its native 236 land and culture by indicating the silence, absence or non-functionality of the vocal organism that forge them: land and culture by indicating the silence, absence or non-functionality of the vocal organism that forge them: Our guttural muse was bulled long ago by the alliterative tradition, her uvula grows vestigial, forgotten like the coccyx or a Brigid's Cross yellowing in some outhouse […]. (WO 21) During the 16th and 17th centuries the dominion of the English language in Ireland was still not certain and the sounds of Gaelic could still be identified as one of the main keynotes of the Irish landscape. Yet, as Heaney notes, the most idiosyncratic Irish vocal sounds – the throaty sounds of Gaelic – have disappeared from the landscape of Ireland and their memories are fading from the collective memory leading, in turn, to the atrophy of the related vocal organs. In the lines quoted above Heaney remarks that Gaelic has been brutally and savagely eradicated from its nesting ground and replaced by the “alliterative traditions” of the Anglo-Saxon and Middle English poetry. 4.4 The Vanished Music: Silence and Absence240 And now, her “uvula” is perishing in a forgotten outhouse of the memory. As Corcoran has commented, in ‘Traditions’, “the fate of a virtually disappeared tongue [is] commemorated in [a] poignant, glancing, allegorical elegy” (40). The absence or scarcity of Gaelic sounds in the soundscape of their native land, for Heaney, indicates an emptiness that the poet wants to inscribe in the present history so as to avoid its total disappearance from memory. The disappearance of the Gaelic tongue is a synecdoche for the vast silence of Gaelic traditions and communities. Heaney announces that his native land has lost its name and identity, and has been reduced to an insignificant component of “the British isles” (WO 21). The sense of absence is, as the poet indicates, empty and nostalgic, because it is created by the wanton destruction of the Gaelic culture, and because it is created by the silence of the 237 vulnerable people who were forced to leave their homelands or smother their customs under the reign of “that ‘most / sovereign mistress’” (21). This silence indicates an absence that the poet wants the society and history to recognise and remember when, for example, hearing the terms “varsity”, “deem” or “allow” in the everyday life of the Irish. The presence of “Elizabethan English” and Shakespearean archaic words, as he explains, is the evidence of “Ireland’s traumatic colonial history, a history whose crucial moment – the ‘Elizabethan conquest and Plantation of Ulster – occurred during Shakespeare’s lifetime” (Corcoran 40). As Maitland notes, this silence “exists under the shadow of the people silenced in order to create it. The silence of oppression, the silence that does ‘wait to be broken’ and needs to be broken in the name of freedom” (257). It is the silence that the Irish poet is determined to break within the freedom of his poetic space. The silence of Irish traditions and culture is an issue that the poet addresses in the poem ‘The Last Mummer’ as well. As the title suggests, the poem features the last survivor of the Irish tradition of mummery, who tries in vain to awaken his people and warn them about its extinction (see also 4.2).241 The tradition was already “in its last gasp” during the poet’s young years (SS 131). armature actors, known as mummers or Christmas rhymers, in Northern Ireland. John J. Marshall says: In the north of Ireland as Christmas drew near it was customary to get up a company of Rhymers who went round the shops and private dwellings reciting their rhymes and collecting money. These were the latter day descendants of the mummers of olden times, who at times of festivity played their pranks for the amusement of their fellows as well as their own. They dressed up in such fantastic costumes as they could manage, and represented various characters, such as St. George, Oliver Cromwell. Beelzebub, Devil Doubt, the Doctor, etc. Each had a rhyme to recite about himself, which he did with such intonation and gestures as he considered appropriate. g pp p See Popular Rhymes and Sayings of Ireland (1924), for a description of the scenarios and records of the rhymes at www.libraryireland.com/social-history/popular-rhymes/christmas-rhymers.php. mmers plays date back to medieval times and were traditionally performed by all-male troupes of actors known as mummers or Christmas rhymers in Northern Ireland John J Marshall says: 241 Mummers plays date back to medieval times and were traditionally performed by all-male troupes of armature actors, known as mummers or Christmas rhymers, in Northern Ireland. John J. Marshall says: 241 Mummers plays date back to medieval times and were traditionally performed by all-male tr 242 Mummering was often carried out as a kind of performative anarchy. It eventually led to a growing public concern and it was being regarded as “a potential threat to social order”. By the 1950s, mummers in Northern Ireland had to apply to the Royal Ulster Constabulary for a permit (Rankin Russell 167-169). Heaney noted: Cockfighting and mumming, by the way, cut across the sectarian divide. All sides attended the fights. And because the fights were illegal, a special bond was created among the aficionados. My last mummer could have been a cockfighter too, ‘picking a nice way through / the long toils of blood / and feuding’. (SS 131) 4.4 The Vanished Music: Silence and Absence240 In the poem, the mummer creeps out of foggy spaces of memory, for one last time, and lurks behind the residents of the electric age now mesmerised by the “luminous screen” (WO 8). His hopes turn into frustration when he realises that his people have chosen, what he considers as a superfluous entertainment in comparison with the authentic Irish tradition of mummery. Heaney brings our attention to some of the major 238 drawbacks of the age of electronic communication: just as the advent of print technology in the fifteenth century prepared the ground for the standardisation of most texts − “eliminating the vagaries of human handwriting” (M. M. Smith 10) - the spread of mass media throughout the twentieth century and later brought forth reproduction, uniformity and repetition, but led to the lack of authenticity of voice and auditory experience (see 1.2). Discussing the consequences of recording and stockpiling musical productions in mid-twentieth century Attali writes: “Reproduction, in a certain sense, is the death of the original, the triumph of the copy, and the forgetting of the represented foundation […]” (Attali 89). In the poem, Heaney avers: “St. George, Beelzebub and Jack Straw // can’t be conjured from mist” (WO 8). According to the poet, modern age entertainment will not replace and transfer the sense of authenticity and presence brought about in the live performances of St. George, Beelzebub and Jack Straw. The culture of television has, instead, rendered the mummer and other actors “redundant” (Corcoran 30) − or what in the context of sound studies could be identified as unwanted sounds, i.e. noises − for the contemporary audience.242 Shrouded by anger and resolved to be heard, the quiet mummer beats the bars of the gate with a stick and hits the road with his fierce steps, but all in vain. At the end, like the servant boy, silent but “resentful / and impenitent” (WO 7) the mummer disappears as quietly as he once emerged (see also 4.2). Yet just as he does with the dying rural crafts of the 1960s, Heaney seals their memories on the pages of his poetry forever: a line of mummers marching out the door as the lamp flares in the draught. as the lamp flares in the draught. Melted snow off their feet leaves you in peace. y , Brass (1975), Henry Glassie remarks that: Generally, all of the mummers were Catholics, though a few Protestants might travel along with them, but they went to Protestant homes as well as Catholic ones and acted with particular politeness at houses where they had heard they were not welcome. (127) 243 Mumming in Ireland has more that 2500 years of history and was a non-sectarian practice. Althou g y y p often carried out by Catholics, the mummers visited both Catholic and Protestant homes. In All S Mumming in Ireland has more that 2500 years of history and was a non sectarian practice. Although it was often carried out by Catholics, the mummers visited both Catholic and Protestant homes. In All Silver and No 4.4 The Vanished Music: Silence and Absence240 (WO 9) 239 However, the silence of the mummer and decline of the Irish mummery tradition is a synecdoche for the vast silence of the Irish identity and tongue as a major legacy of colonialism, which as Parker has rightly noted, are explored in a succession of poems − ‘Traditions’, ‘Anahorish’, ‘Broagh’ and ‘Toome’, ‘A New Song’ and ‘Gifts of Rain’ (97) − discussed in great detail in the analytical chapters of the thesis. The last survivor of the Irish mummery cannot awaken the people deafened by the clamour and glamour of modern entertainment: “The luminous screen in the corner / has them charmed in a ring” (WO 8). Even his rattling and hammering is shrugged off. Here, Heaney is referring to the silence concealed by so much noise to hint at a kind of silence prevalent in the midst of the Troubles. In the poem, the twentieth century’s invention has turned into a means of silencing the audience or, what Attali calls, “death in the heart of life” (120). Written against the backdrop of the Troubles that Northern Ireland was experiencing during the 1970s, in ‘The Last Mummer’, Heaney fears the indifference to the sectarian violence and the loss of hope for a better future. He remarks that mummery, like cockfighting, is a tradition that transcended the sectarian divide (SS 131),243 an element that Rankin Russell highlights in order to indicate mummying’s ability to bring unity amongst the Protestants and Catholics (167-169). In fact, the mummer’s insistence on being heard connotes the urge he feels for its presence in contemporary Ulster. Heaney’s poem as much projects his concerns about any insensitivity to violence as it reveals his hopes for “peace” in Ulster (WO 9): “The more I talk about those poems, the more I see that they were about the 240 need to break out of the consensus that Ulster was ‘a good wee place’, the need to get on the road to Aarhus, to acknowledge the lostness and unhappiness of ‘home’” (SS 130). need to break out of the consensus that Ulster was ‘a good wee place’, the need to get on the road to Aarhus, to acknowledge the lostness and unhappiness of ‘home’” (SS 130). 244 See celt.ucc.ie/published/E900040/text007.html. 4.4 The Vanished Music: Silence and Absence240 Heaney brings up the topic of silence and absence in ‘The Wool Trade’, where he laments the loss of the Irish woollen industry after the restrictions placed by English commercial policies on Irish woollen manufacture in the 17th century.244 Therefore, the wool trade is not just a name of an outdated industry but a keynote and a soundmark of Irish prosperity eradicated by the Anglo-Scottish invaders of 1700. The poem is constructed on the memories of a vanished trade as recalled by the unknown and unnamed Irish interlocutor (see 3.4). The sound of shearing, washing, baling, bleaching and sorting the wool, carding, spinning and weaving the yarn, or the clang and clatter of looms and spindles − that the man recalls − are audible fragments from Heaney’s “personal and Irish pieties” (PO 37). In present day Ireland, the memories of the Irish spinning machines are “fading” from their native land while the Irish “must” instead recite the vocabulary of the Scottish tweed industry. The vanished industry is also a metonymy for the loss of the Irish language and identity as a consequence of colonialism. The empty silence of Irish villages and hills that Heaney can hear today recalls the absence of the native culture and vernacular that he would like to conjure up and resonate throughout his poetry: O all the hamlets where Hills and flocks and streams conspired O all the hamlets where Hills and flocks and streams conspired To a language of waterwheels, A lost syntax of looms and spindles, How they hang Fading, in the gallery of the tongue!” (WO 27). Conclusion Heaney’s stance as the earwitness of the Irish landscape can be described as that of a silent listener. As a listener and observer, as a thinker and meditator and as a poet, silence has 241 been integral to Heaney’s life and career. Yet no two states of silence are the same for him. In ‘Anahorish’, ‘Toome’, ‘Broagh’, ‘May’ and ‘A New Song’, the poet seeks silence to hear and internalise the familiar symphony of the local river. In ‘Land’, he seeks the acoustic memories of the rural life his father sought. In ‘Postscript’, ‘Valediction’, ‘Lovers on Aran’, ‘Shore Woman’ and “Maighdean Mara’ he tunes his ears with the voices of the sea. In ‘The Sound of Rain’ and ‘Gifts of Rain’ he listens to the sounds of the rain. In the Burren and bogland, he searches for the echoes of past voices and sounds. In ‘Oracle’, he returns to his days as an innocent little boy who escapes everyday noises and family expectations to find comfort and pleasure in the soundscape of “a pristine world full of woods and water and birdsong” (PO 181). He seeks the spaces of silence to practise listening and tune in his senses and emotions with the patterns of life. Heaney seeks silent places to empty his mind, broaden his view, expand the limits of his identity and merge it with, what he calls, “the god in the tree” and defines as “a psychic force that is elusive [and] archaic” (PO 186). God, in traditional Christian thinking, would be the creator, the life-giver and redeemer that the practitioner of silence wants to connect with. In a more archaic sense, and for the poet, this elusive force would be a deity “shrouded in the living matrices of stones and trees, immanent in the natural world” (PO 186). Heaney’s silence in much of his nature poetry is the state of mind that Maitland associates with hermits, when “one is trying to empty oneself of ego; pour oneself out, become permeable, translucent, empty, open to the transcendent” (193). Heaney states this association himself: “I’m in the tradition of the hermit poets of the early Celtic nature tradition” (see also 3.2).245 To listen attentively, one must be silent oneself stilling the chatter of the voices in the mind. Heaney’s silence also teaches his readers not just to “listen to” but also to “listen for” (SL 1). 245 See Heaney’s 1989 interview with the Desert Island Disc host at www.youtube.com/watch?v=bAuucSly7t8 (3:50-3.59). And, more extensively, here, where he reads his own blackbird poems and his translations from the Irish blackbird poems at www.youtube.com/watch?v=emuYwWT7s4A (1:06:02-1:06:04). (3:50 3.59). And, more extensively, here, where he reads his own blackbird poems and his translatio Irish blackbird poems at www.youtube.com/watch?v=emuYwWT7s4A (1:06:02-1:06:04). p j Q 7 See rightsni.org/2013/08/seamus-heaney-ambassador-of-conscience/. 246 The writings collected by Amnesty International Irish Section and published by the Irish Times. See: https://www.irishtimes.com/news/human-rights-poetic-redress- ttps://www.irishtimes.com/news/human rights poetic redress 903757?fbclid=IwAR2VU7aYJph2dNsB8UVnU3cY0uHfX8lNjPQ08YGZnVrC5xLP9eOuk0nevVE. p j Q 247 See rightsni.org/2013/08/seamus-heaney-ambassador-of-conscience/. p g p 1.903757?fbclid=IwAR2VU7aYJph2dNsB8UVnU3cY0uHfX8lNjPQ08YGZn Conclusion His silence and the sense of waiting that his listening stimulates, encourages the reader too to 242 be in the now, in the present tense, and to adopt the stance that Buddhists call “mindfulness” (Maitland 159). “The republic of conscience” (HL 14), the liminal space from which he comes back with the urge to tell, is a noiseless space. The landscape of Heaney’s imagination is endowed with the kind of silence in which his senses are most free and alert. It is a space wherein, according to the poet himself, one feels the need to become more conscious of oneself and one’s surroundings. It is a “place of silence and solitude where a person would find it hard to avoid self-awareness and self-examination” (SS 292). That statement could partly explain the reason, after returning from California, he chooses to retire to Glanmore Cottage, which he defines as “a completely silent place of writing” (SS 322). Heaney seeks silence to reach the deep well of sensory memories and to reflect upon them, but this initial movement of extraction is completed when the poet returns to the white surface of the page with a hoard of stories to tell. As Maitland writes, “[s]ilence firstly puts one in that ‘other’ place and secondly gives one an opportunity, without interruptions or comments, to retrieve and shape those memories” (240). In his introduction to the series of writings that marks the 60th anniversary of the Universal Declaration of Human Rights, which I quoted above (see 4.1), Heaney remembers that in the silence of Orkney he was able to reach a kind of “self-awareness and self-examination” that almost immediately enabled him to weave his sensory experiences and imagination, writing the poem ‘The Republic of Conscience’ (HL 14-15).246 Heaney, as the ambassador of “the republic of conscience” (HL 14), would necessarily need to have a voice.247 One cannot be a silent listener merely and at the same time be creating new words and new worlds. In ‘Personal Helicon’, he listens to the echoes of his own voice. In ‘Oracle’, he links ear to mouth, lobe to larynx, listening to speech, and silence to voice, in 243 his own name and existence. He recalls his name being cuckooed across the fields as the “small mouth and ear” (WO 18). y p , Berkeley, saw the right to speak freely and to be heard, as an absolutely crucial human right. 248 In many occasions, Heaney refered to his poetry as “an escape from a terrible fear of silence that always haunted him”. See www.nytimes.com/2013/08/31/opinion/seamus-heaney-poet-of-the-silent-things.html. 249 In the second half of the twentieth century the political freedom movement, that started at the University of Berkeley, saw the right to speak freely and to be heard, as an absolutely crucial human right. haunted him . See www.nytimes.com/2013/08/31/opinion/seamus heaney poet of the silent things.htm 249 In the second half of the twentieth century the political freedom movement, that started at the Univ 8 In many occasions, Heaney refered to his poetry as “an escape from a terrible fear of silence that a y p y p g e second half of the twentieth century the political freedom movement, that started at the University of y saw the right to speak freely and to be heard as an absolutely crucial human right Conclusion ‘Shore Woman’, Maighdean Mara’, ‘Summer Home’, ‘Bye-Child’, ‘A Winter’s Tale’ and even ‘Wedding Day’ − Heaney creates a space wherein the female characters’ monologues, songs and weeping can be heard. In many of these poems, he holds his silence, almost like a psychoanalyst, so as to create a space for his characters to freely voice their feelings and existence. In ‘Servant Boy’, ‘Veteran’, ‘Navvy’, ‘The Wool Trade’, and ‘Bog Oak’ he listens to the silence of the marginalised and the overlooked. As Corcoran has remarked, in his poetry, Heaney “registers a sympathy for these historically dispossessed and maltreated, and their successors” (31-32). Heaney’s silence and the platform of his poetry create a warm and comfortable atmosphere for silent men and women to unwind the hoard of their untold stories and be finally heard not only by the poet, but also by the future generations. Listening precedes speech. It prepares speech (B. R. Smith 13). In the series of poems with distressed women as their subject matter − i.e. ‘Shore Woman’, Maighdean Mara’, ‘Summer Home’, ‘Bye-Child’, ‘A Winter’s Tale’ and even ‘Wedding Day’ − Heaney creates a space wherein the female characters’ monologues, songs and weeping can be heard. Listening precedes speech. It prepares speech (B. R. Smith 13). In the series of poems with distressed women as their subject matter − i.e. ‘Shore Woman’, Maighdean Mara’, ‘Summer Home’, ‘Bye-Child’, ‘A Winter’s Tale’ and even ‘Wedding Day’ − Heaney creates a space wherein the female characters’ monologues, songs and weeping can be heard. In many of these poems, he holds his silence, almost like a psychoanalyst, so as to create a space for his characters to freely voice their feelings and existence. In ‘Servant Boy’, In many of these poems, he holds his silence, almost like a psychoanalyst, so as to create a space for his characters to freely voice their feelings and existence. In ‘Servant Boy’, ‘Veteran’, ‘Navvy’, ‘The Wool Trade’, and ‘Bog Oak’ he listens to the silence of the marginalised and the overlooked. As Corcoran has remarked, in his poetry, Heaney “registers a sympathy for these historically dispossessed and maltreated, and their successors” (31-32). Heaney’s silence and the platform of his poetry create a warm and comfortable atmosphere for silent men and women to unwind the hoard of their untold stories and be finally heard not only by the poet, but also by the future generations. Conclusion The far-reaching and penetrating resonance foretells that the little innocent boy lurking up in “a woody cleft” is also the voice of his pastoral places. Heaney indicates that he seeks silence to practise concentration, meditation and mindful listening, as well as to search for his own voice and identity as a poet of mossy lands. “What is the source of our first suffering?” Heaney asked, quoting the French philosopher Gaston Bachelard. “It lies in the fact that we hesitated to speak”.248 As a poet, he sees silence as a means to find a balanced position and a way to strengthen this stance by protecting it from social and political pressures and expectations. Heaney seeks silence to establish an authentic voice in which to write - a voice that, like that of the cairn-maker, has “his own mark” on it (WO 39). Heaney’s concerns are not merely environmental but also humanistic. Everyone can have an inner self and a voice, but not everyone is ready, willing or allowed to express it. In his poetry and prose, Heaney provides a space for listening to the silence of the people whose thoughts, feelings and presence would otherwise have gone unheard. Sometimes people are silent because they are not free to speak at all − or at least not audibly. This situation often applies to individuals and groups who have been oppressed or marginalised by society. Throughout human history, in general, and the history of the colonisation of the Irish, in particular, the experience of the oppressed is that they have been silenced, rather than freely choosing silence. Moreover, Heaney’s references to “[t]he famous // Northern reticence” hints at silence as an equally significant element as speech in shaping Irish history (N 54). There is, however, another kind of silence that allows this silence to break: the silence of a listener and a thinker.249 The experience of silence early on in his life − through such silent 244 figures as his father, mother, aunt Mary, the thatcher, the navvy, the blacksmith and neighbours − fostered and shaped the poet’s auditory and perceptual abilities allowing him to recognise silence as a means of not only communicating and connecting with other individuals, but also uncoding the socio-political context of his homeland. Listening precedes speech. It prepares speech (B. R. Smith 13). In the series of poems with distressed women as their subject matter − i.e. Listen Now Again: Listen Now Again: 250 The title of this section is inspired by the last line of ‘The Rain Stick’ (SL 1). Conclusion It can be very tempting to surrender completely to the hurry and scurry of everyday life. But the poet is aware of the importance of slowing down and taking a moment to pause. Heaney’s pauses are his deliberate attempts to create silence, a space for listening, memory digging, searching for the voices lost in the archive of history, reflecting over life and finding his voice. Heaney’s poetry epitomises moments of resounding quietness, when the poet travels internally to enter the soundscape of his subconscious mind and to get in touch with the origin of the sounds and voices knocking at his most “buried ear” (DC 55). Heaney writes that poetry begins with “the thump in the ear” (PO 53). Digging, therefore, is not just the title of Heaney’s first poem and a metaphor for poetry writing but also the indication of the poet’s 245 solitude and meditative state, the space he seeks for retrieving the buried shards and places he would like to retrieve and bring back to life. In his ‘Feeling into Words’, he refers to a view of poetry “as elements of continuity, with the aura and authenticity of archaeological finds, where the buried shard has an importance that is not diminished by the importance of the buried city; poetry as a dig, a dig for finds that end up being plants” (PO 41). Moreover, his poetry encourages readers to take a pause. Heaney highlights that literature, especially poetry, is not a way of solving social and political problems but can play a role in foregrounding them: “it [poetry] creates a pause in the action, a freeze-frame moment of concentration, a focus where our power to concentrate is concentrated back upon ourselves” (SS 383). Heaney’s pause in ‘The Blackbird of Glanmore’ and ‘The Makings of Music’ is not only a moment for retrieving personal memories but also an attempt to nudge his readers to take a moment and listen to the otherwise lost music of every sound. His poetry records the memory and history of the disappeared. The poems ‘Bog Oak’, ‘Anahorish’, ‘Midnight’, ‘The Tollund Man’ ‘Traditions’, ‘The Last Mummer’, ‘Midnight’ and ‘The Wool Trade’ are a reminder that the apparently empty silence exists only under the shadow of the people, languages, cultures silenced to create it. 246 Conclusions250 In this thesis, I focused on listening as an important medium of experiencing, understanding, communicating and remembering as illustrated in the poetry of Seamus Heaney. I proposed that Heaney’s poetry is saturated with references to sounds and references to the absence of sounds, and that the identification and interpretation of such references can enhance our experience and understanding of not only the poems, but also the world experienced by the poet himself. It is widely acknowledged that poetry is an art form that foregrounds language and its auditory elements (see 1.1). Traditionally, literary scholars and critics try to account for the aural effects of poetry in a variety of ways. The aural features of a poetic language are part of its aesthetic value but can also contribute significantly to the way in which a poem communicates ideas and feelings, as well as the way it helps retain them in and retrieve them from memory. In spoken language, as Don Paterson notes, the variations in the patterning of sounds, rhythm, pitch, intonation and timbre are responsible for conveying the nuances of feelings that the mere word-sense cannot carry. Poetry, then, may introduce music into language through the careful arrangement of word-sounds, reinforcing the communication of ideas and feelings (5-8). The effect of this is twofold: Our long-term memory encodes information semantically, our short-term memory acoustically, and these rhythmic, parallel and repetitive sound-tricks simply give the line a better chance of hooking on a single hearing; this way they can be later recalled, and their meaning more carefully dwelt upon. (Paterson 5-8) The work of some recent critics, however, is distinguished by more novel approaches to the study of sound, noise, silence and listening in literature. In this thesis, I have benefited from the various disciplines in sound studies and literary criticism, applying a new 247 interdisciplinary approach to the study of sounds in Heaney’s Wintering Out. The descriptive framework offered by the ecological approach to aural perception, and the taxonomy offered by acoustic ecology and soundscape ecology have helped develop the conceptual framework of this thesis. Alongside the linguistic and literary practice of close reading, these approaches have facilitated the identification and interpretation of auditory references in his poetry (see 1.4). Conclusions250 An ecological approach to listening explores auditory perception along the dimensions of a sound producing source, the location and the environment rather than the basic attributes of the soundwaves per se. Studies of listening from an ecological approach have shown that, through listening to sounds, the human ear can identify such information as the material, texture, and configuration of the objects, the force and the type of the interaction involved, as well as the proximity, time and direction of the sound event (see 1.3). This descriptive approach to audible source attributes provides a suitable framework for identifying audible and non-audible references in the poems, the references that otherwise would go unnoticed or interpreted as merely visual. On many occasions, these references unravel the poet’s own acoustic experiences and, ultimately, the aural responses they could trigger in the readers. The analytical lens offered by acoustic ecology is an interdisciplinary approach that places the home territory of soundscape studies in the middle ground between science, society and the arts to study the interrelations between human beings and the changes in their acoustic environment (see 1.3). The aesthetic, social, cultural, political and ethical dimensions of auditory perception are pervasive themes in acoustic ecology, which offers a rich taxonomy of sounds − e.g. ‘keynote sounds’ to refer to background sounds, ‘soundmarks’ to refer to locally unique sounds, ‘signal sounds’ to refer to foreground sounds, ‘archetypal sounds’ to refer to symbolic sounds, and ‘imperialistic sounds’ to refer to sounds 248 that can create a large acoustic profile − to show how the acoustic environment shapes human beings’ experiences and understanding of the world. The closely related field of soundscape ecology broadens the humanistic scope of acoustic ecology to study the inter-relations between the acoustic environment and the behaviour of all the living beings in it (see 1.3). Soundscape ecologists identify the three basic sources of sounds − ‘geophony’ to describe sounds of non-sentient natural phenomena, ‘biophony’ to describe sounds emanating from non-human sentient organisms and ‘anthrophony’ to describe all sounds produced by humans, and stationary and moving human-made objects − to examine relationships between living organisms, including humans and other terrestrial or marine organisms, and their environment across different spatial and temporal scales. Conclusions250 Soundscape ecologists argue that the surrounding soundscape engages our aural sense, providing us with much information about many associated human ideals, such as identity and sense of place, as well as recreational, therapeutic, educational, and aesthetic values. Reading Heaney through the lens of an ecological approach to auditory perception, acoustic ecology and soundscape ecology allows us to better understand and appreciate his poetry. Heaney spent significant portions of his life in Ulster and travelled extensively to other countries such as the Republic of Ireland, Denmark, Greece, France, Italy, Spain, Poland, England and the United States. The concept of place has been a consistent source of inspiration for his poetry. As I have shown in this thesis, place is significantly mediated by the ear in his work. Heaney concluded his essay ‘The Sense of Place’ by asserting that the sense of a place is not determined through its political and geographical territory, but through “the way the surface of the earth can be accepted into and be a steadying influence upon the quiet depth of the mind” (PO 145). Heaney’s interest in soundscapes and the role that sounds might play in shaping and reflecting the socio-cultural dynamics is evident through his prose 249 and poetry. In his poetic re-construction of a sense of place, Heaney brings together the various components of the soundscape − i.e. geophony, biophony and anthrophony − as well as silence. He writes of the sounds directly experienced and intimately known, providing his readers with what Schafer refers to as a “trustworthy” source of information about how various soundscapes mediated the underlying social, political, cultural and personal associations (8).251 Heaney writes out of his own sensory experiences or as inspired by his in- depth knowledge of history and mythology. He brings forth a mixture of both personal and second-hand information. Therefore, we can trust Heaney as much for the literal accounts of auditory references rendered in his poetry and prose, as for his accounts of auditory illusions and imagined soundscapes. Throughout my thesis I did apply some degree of selectivity assembling and interpreting the important features of Heaney’s poetic soundscapes, noting differences and parallels, in order to identify the contribution of sounds − and their absence − to his themes. Heaney’s imagination is a testimony to nature’s abundance and beauty. Listening to the sounds of the natural environment is for him a source of personal joy and delight. 251 Schafer notes, “[w]riting about other places and times usually results in counterfeit descriptions. […]. In such ways the authenticity [and non-authenticity] of the earwitness established”. Therefore, Swift’s claim that Niagara Falls made a sudden “terrible squash” cannot be trusted because we know he never visited the place. However, we can rely on Chateaubriand’s description of the roaring fall “from eight to ten miles away”. In other words, Chateaubriand’s description of his first-hand aural experience provides us with reliable information about the ambient sound level, one that can be used as a source of reference for interpreting the ambient sound level of Niagara Falls today. However, once the authenticity of the earwitness-author is established, we can trust the author for what might even seem like an “aural illusion” or the accounts of “unusual sound events”. For instance, we can trust Remarque’s descriptions of the hissing and belching dead bodies in All Quiet on the Western Front (1929) and Faulkner’s accounts of noise from corpses in As I Lay Dying (1930) (8-9). 251 Schafer notes, “[w]riting about other places and times usually results in counterfeit descriptions. […]. In such ways the authenticity [and non-authenticity] of the earwitness established”. Therefore, Swift’s claim that Niagara Falls made a sudden “terrible squash” cannot be trusted because we know he never visited the place. However, we can rely on Chateaubriand’s description of the roaring fall “from eight to ten miles away”. In other words, Chateaubriand’s description of his first-hand aural experience provides us with reliable information about the ambient sound level one that can be used as a source of reference for interpreting the ambient sound p g level of Niagara Falls today. However, once the authenticity of the earwitness-author is established, we can trust the author for what might even seem like an “aural illusion” or the accounts of “unusual sound events”. For instance, we can trust Remarque’s descriptions of the hissing and belching dead bodies in All Quiet on the Western Front (1929) and Faulkner’s accounts of noise from corpses in As I Lay Dying (1930) (8-9). 251 Schafer notes, “[w]riting about other places and times usually results in counterfeit descriptions. […]. In such ways the authenticity [and non-authenticity] of the earwitness established”. Therefore, Swift’s claim that Niagara Falls made a sudden “terrible squash” cannot be trusted because we know he never visited the place. H l Ch t b i d’ d i ti f th i f ll “f i ht t t il ” I th However, we can rely on Chateaubriand’s description of the roaring fall “from eight to ten miles away” words, Chateaubriand’s description of his first-hand aural experience provides us with reliable inf about the ambient sound level, one that can be used as a source of reference for interpreting the ambie Conclusions250 He seeks the spaces of silence in nature to tune in his imagination and feelings with the patterns of nature, the ceaseless rhythms of waves and streams, and the symphonies of birdsong at dawn. Heaney is a silent and attentive listener. In order to listen attentively, one must be silent, stilling the chatter of the mind in order to allow other voices to speak (Scharper and Leman-Stefanovic 166). His silence teaches his readers to listen patiently and attentively for what might happen and might become a sound, for that moment when suddenly, out of the 250 commotion, a sound jumps forward. He encourages his readers to listen to the surrounding world carefully, because what might happen now, might not happen again. Heaney’s emphasis on the transient nature of sounds and on inner silence encourages the development of greater sensitivity to the surrounding soundscape. This type of silence and attentiveness, in fact, disciplines all of our senses. Once we become aware of the presence of something, we automatically and subconsciously look for it and listen for it in other places in life (Maitland 166). Heaney opens the space for a more universal sense of place than just Ireland, disciplines not only our ears, but all of our senses and shapes our relationship with our own natural environment. In addition to ear awakening, acoustic ecology also aims to foster a deeper awareness of the social (including political and cultural) and the personal (including psychological) significance of surrounding soundscapes. For acoustic ecologists, mindful listening can be an act of engagement with the world. Listening becomes a metaphor for reflecting a more inclusive, focused and respectful awareness of the surroundings and of one’s position in them. Heaney listens to natural soundscapes, like a soundscape analyst, to discover and record their significant features, familiar sounds, sounds important to a culture, sounds that are unique to a place or sounds that carry a specific message. Nature for him is also a source of knowledge. He is aware that background keynotes establish an imprinted connection to the sense of place and time. He identifies with his familiar locations by recalling their associated keynotes and soundmarks − from amongst the natural sounds of the ambient and biophonic sounds, to sounds of local crafts, voices and accents of local people. Conclusions250 These are the everyday sounds that have been inscribed in his shared memory and turned into aural treasures of identity. Recalling these sounds becomes a way the poet re-connects with his land, cherishes the collective heritage and thus reinforces the sense of belonging. 251 Heaney’s references to the biophonic sounds of the natural world open a wealth of knowledge about the relationship between humans and non-human animals. Listening to the sounds of farm animals for him is a way of returning to the soundscape of his childhood and the way of life that his father and ancestors lived. But Heaney also writes about the animals contemporary man may barely be aware of. In a man-dominated world, Heaney writes meticulously about the disappearing keynotes of the natural world, the songs of birds and the sounds of a wide variety of land and sea species. His poetry inspires his readers to celebrate the existence of animals as keynote elements of the natural soundscape without whom human life experience would be incomplete. Heaney’s poetry brings to light the empowering presence of the cars, tractors and generators, trains and planes but also reminds us that the heedless spread of anthrophonic sounds can impose constraints on the acoustic environment of non-human animals. Heaney is a careful listener to all sounds. What an unaware ear interprets as noise, for Heaney becomes a great resource for understanding the social, cultural and political dynamics in the history of human civilisation, an attitude that is aligned with the perspective of many sound scholars, who maintain that the scant attention that so-called noise has received has led to a dimmed knowledge about history and society (Attali 3). His poetry encourages his readers to listen to everyday noise attentively in order to understand not only the present time, but also the past our ancestors lived in and the future we can hope for. The reassuring footsteps of the farmer, his serene silence, the earthy grating sound of the spade, the rhythmic swish and slash of the mowing scythes, are amongst the keynotes of manual farming methods that Heaney evokes to re-orient himself towards his rural origins and reconcile with his rural roots, bridging the gap between his own profession as a poet and those of his farmer ancestors. Conclusions250 Heaney also writes of the throaty sounds of the Gaelic language, the click and clang of the Irish wool factory, the churning of butter or the high- 252 pitch sound of the blacksmith’s hammer, the anthrophonic keynotes and soundmarks that have long disappeared or are becoming rare. For him, writing of these sounds is a way of not only re-collecting his personal memories and returning to the soundscape of his childhood, but also collecting disappearing soundmarks and keynotes. Heaney shoulders the responsibility of recording and preserving the human sonic heritage in his poetry and of carving it in the ongoing story of human history. In his poetry, he also lends himself to ecocritical politics of Ireland, presenting the ways in which the white noise of sectarianism has transformed the auditory texture of his homeland, offering, at times, a feeling of at-homeness. He represents the recurring history of sectarian violence in Northern Ireland through its customary noises, suggesting that violence has become an integral part of the society. Living in Northern Ireland, he can hardly avoid the choppy drone of helicopter rotors hovering over the city, the broken rhythms of machine-gun fire, the blasting of bombs, the boisterous drumming of Orangemen, the cadences of pub talks, the huffing of many car engines in each funeral and sirens, which for him have turned into the familiar keynotes of incessant army surveillance, division, insecurity and death, leaving a lasting imprint on the mind and memory of the individuals. Heaney also encourages his readers to listen to the silence of individuals as much as to their voices. In his poetry, he shows how violence can affect people in a society, resulting often in what Lundy has identified as “a conspiracy of silence” (111). But he also reminds his readers that, as Maitland has suggested, in any repressive socio-political environment, silence can be “a sound strategy” (181). In much the same way, Heaney’s illustrations of silence – as absence and emptiness – disclose the deep impact of the political situation on the social and cultural soundscape of Ireland. He directly links the Irish wood, wildlife and local industries to its language and culture. His reference to the disappearance of their acoustics provides a 253 means for him for talking about the legacy of colonisation and the suppression of the Irish identity. Conclusions250 Heaney’s involvement in the socio-political climate of his homeland is also represented through his lament for the suppression of the Gaelic language and his ambivalence about the place of the English language and the English literary canon. Heaney stages language as a show of imperial force and brings our attention to the gradual decline of the sounds of Gaelic, which up until the 16th and 17th centuries could still be identified as one of the main keynotes of the Irish soundscape. He sees the Irish language as a potent symbol of national identity and independence, speculating that the feelings of absence and in- betweenness experienced by many Irish are peculiarly postcolonial predicaments. He links the Irish language to national resistance ever since the efforts to de-Gaelicise the Isles and declares his opposition to the imperialistic presence of the English language in Ireland. He indicates that even the simplest terms spoken by Irish are often freighted with political baggage. His etymological excavations of Irish place names expose the ways in which words retain cross-cultural affinities. On the other hand, however, his poetry in and by itself critiques the Irish-English dichotomy. As Foster has noted, English for him is not merely the conqueror’s language, but also the native tongue of modern Ireland and a major part of the Irish literary landscape today (38). Finally, he turns his poetry into a platform for hearing not only his own voice, but also voices obscured by louder voices in history, voices of objection, forgotten voices, silenced voices, voices that otherwise would have gone unheard. Listening is accepting the presence of a sound source. Heaney’s attention to the voices of others and their evocation in his poetry brings us back to one of the most compelling aspects of acoustic ecology: its emphasis on appreciating the voice and existence of the other. For Heaney, listening to other voices is vital to his attempt to create a poetry that is not only personally satisfying but also socially and 254 ethically aware. His attention to listening to other voices as a way of perceiving and communicating is mediated through and placed at the heart of his life stance and poetic technique. In ‘Feeling into Words’, Heaney differentiates between craft and technique in poetry. display […] all voice and nothing else – but not voice as in ‘finding a voice’. (PO 47). Technique, however, involves not only a poet’s way with words, his management of metre, rhythm and verbal texture; it involves also a definition of his stance towards life, a definition of his own reality. It involves the discovery of ways to go out of his normal cognitive bounds and raid the inarticulate: a dynamic alertness that mediates between the origins of feeling in memory and experience and the formal ploys that express these in a work of art. (PO 47) g p be deployed without reference to the feelings or the self. It knows how to keep up a capable verba 252 He defines craft as “the skill of making. It wins competitions in the Irish Times or The New Statesman. It can be deployed without reference to the feelings or the self. It knows how to keep up a capable verbal athletic 252 He defines craft as “the skill of making. It wins competitions in the Irish Times or The New Statesman. It can 252 He defines craft as “the skill of making. It wins competitions in the Irish Times or The New Statesman. It can be deployed without reference to the feelings or the self. It knows how to keep up a capable verbal athletic display [ ] all voice and nothing else – but not voice as in ‘finding a voice’ (PO 47) Technique however efines craft as “the skill of making. It wins competitions in the Irish Times or The New Statesman. It can yed without reference to the feelings or the self. It knows how to keep up a capable verbal athletic yed without reference to the feelings or the self. It knows how to keep up a capable verbal athletic …] all voice and nothing else – but not voice as in ‘finding a voice’. (PO 47). Technique, however, Conclusions250 This is particularly reflected in his nown lines from the opening of his first essay, Mossbawn: I would begin with the Greek word, omphalos, meaning the navel, and hence the stone that marked the centre of the world, and repeat it, omphalos, omphalos, omphalos, until its blunt and falling music becomes the music of somebody pumping water outside our back door. (PO 17) In this context, one major element that Heaney focuses on is the personal, social and historical aspects of the auditory experience and imagination, on the fact that omphalos conjures up the momentum of rural Irish life, the footsteps of men and women coming and going in the backyard of his childhood house, the rattling of their buckets and the clip- clopping of horses. The word places the poet in Co. Derry in the early 1940s, to also hear the hovering helicopters and groaning of bombers above their heads. In this context, one major element that Heaney focuses on is the personal, social and historical aspects of the auditory experience and imagination, on the fact that omphalos conjures up the momentum of rural Irish life, the footsteps of men and women coming and going in the backyard of his childhood house, the rattling of their buckets and the clip- clopping of horses. The word places the poet in Co. Derry in the early 1940s, to also hear the hovering helicopters and groaning of bombers above their heads. In sound studies, the concept of auditory imagination is applied in a broad sense that includes all manners of sounding events. It refers to an openness to and awareness of sounds as a crucial component of human life, culture, history and society. It is about sounds but is more concerned with the contemplative space that surrounds them. Auditory imaginations are essentially reliant on the capacity of the mind to critique and create new sonic narratives across a wide range of registers. As Stern sums it up: “sonic imaginations are necessarily plural, recursive, reflexive, driven to represent, refigure and redescribe. They are fascinated by sound but driven to fashion some new intellectual facility to make sense of some part of the sonic world” (5). Heaney’s auditory imagination is aligned with the concerns of sound studies. It is both the anchor and the ship of his thoughts and feelings. Conclusions250 The early stages of his poetic life – particularly the university years when he mostly wrote as Incertus for literary magazines − were demonstrative of not so much technique. Instead, he was drawn towards craft, which he describes as “the skill of making” (PO 47). However, Heaney links a poet’s technique to their “stance towards life” and the need for probing “the origins of feeling in memory and experience” (PO 47).252 Wintering Out was born out of this poetic awareness and intellectual maturity. To achieve his technique, Heaney follows Eliot in using the concept of the “auditory imagination”, which facilitates a trip deep into sub- consciousness and memory, into the origin of the long forgotten and the ordinary. He explains how Eliot’s auditory intelligence has been encouraging him to exercise deep listening and to search for the music of the world within the linguistic soundscape: Eliot’s revelation of his susceptibility to such lines, the physicality of his ear, […] confirmed a natural inclination to make myself an echo-chamber for the poem’s sounds. I was encouraged to seek for the contour of meaning within the pattern of rhythm. (FK 36-37) Heaney declares that his stance in life and the inclination at the heart of his poetic technique Heaney declares that his stance in life and the inclination at the heart of his poetic technique are but to make himself “an echo-chamber” of the sounds, which is a meditative space where are but to make himself “an echo-chamber” of the sounds, which is a meditative space where the sounds of the words meet the sounds of the world: “Words themselves are doors; Janus is the sounds of the words meet the sounds of the world: “Words themselves are doors; Janus is to a certain extent their deity, looking back to a ramification of roots and associations and to a certain extent their deity, looking back to a ramification of roots and associations and 255 forward to a clarification of sense and meaning” (PO 52). This is particularly reflected in his forward to a clarification of sense and meaning” (PO 52). 253 Commenting about the relationship between poetic voice and a poet’s real-life, in Preoccupations perspective he notes: finding a voice means that you can get your own feeling into your own words and that your words have the feel of you about them; and I believe that it may not even be a metaphor, for a poetic voice is probably very intimately connected with the poet’s natural voice, the voice that he hears as the ideal speaker of the lines he is making up. (PO 43) p g p ( ) 254 www.nobelprize.org/nobel_prizes/literature/laureates/1995/heaney-facts. Conclusions250 It is as concerned with delight in the sounds of the senses as it is with the senses arising from sounds. It brings us to ordinary and familiar sounds and sonic experiences from a completely new and unexpected perspective. It ranges from the private experiences of a variety of individuals to communities from larger historical and geographical landscapes, and at every turn reproduces cultural understanding. 256 Heaney’s poetic voice transcends spatiality and temporality. He identifies with voices from near and distant places in history and from beyond Irish borders in order to find a voice that articulates his own intentions and purposes from a new and more balanced stance. He writes in a voice that belongs to him, one that puts his own feelings and thoughts into his own words, but also one that akin to his real-life stance is inclusive and socially, culturally, historically, and environmentally aware.253 The ever-growing reputation of Heaney’s poetry among the general public and the academic community derives from the intrinsic qualities of his poetry. In 1995, the Swedish Academy awarded Seamus Heaney the Nobel Prize in Literature “for works of lyrical beauty and ethical depth which exalt everyday miracles and the living past”.254 Heaney recognised the borders and limits, only to push them further. In the preface to Finders Keepers, he reprints words from the foreword to his former critical prose, Preoccupations, assuring his readers that his major poetic concerns have remained unchanged: “How should a poet properly live and write? What is his relationship to be to his own voice, his own place, his literary heritage and his contemporary world?” (PO 13). Heaney’s career as a poet was an intellectual response to his making sense of the landscape and soundscapes he lived in, and of continuously having to find the balance between the pulls of roots and the drive for transcendence. Throughout his poetry and criticism, Heaney strove to safeguard the principle of balance; the balance between artistic freedom and ethical responsibility, between the ordinary and the marvellous, between locality and universality, the north and the south, and between future and past. 257 The other main reason behind the wide recognition of Heaney’s poetry is the immediacy with which he captures the attention of his readers. Conclusions250 This capacity is partly a result of “Heaney’s ability to evoke an action or sensation in a phrase constructed out of some of the most commonly used lexical resources [which] is a very rare phenomenon in literature” (Crotty 44-45). Amongst other contributing factors to Heaney’s success is his attention to the empirical and the sensory rather than the philosophical aspects of experience. Heaney’s instinct delights in an evocatively sensory language with which he arrests the attention of non-specialist readers and to which he remained dedicated throughout his poetic career. Heaney’s interest in the sensory and the concrete granted him what Crotty identifies as “the capacity for showing rather than merely stating” that characterises Heaney’s artistic intelligence and distinguishes his poetry from the work of other twentieth century poets (44- 45). Heaney’s voice has something essential to tell his readers, something that they can recognise instantly and instinctively if they lend their ears to the sounds of the words, something as “a true sounding aspect” (PO 43) of the poet’s perspective and experiences. His voice takes us beyond the linguistic space and into its origin in the physical world. Heaney’s poetic voice is composed of “the tangible aural sensations that create meaning almost independent from the semantics of the language, scraping down even further into the unconscious” (Morris 128). In his voice we can find a “rightness” at all levels. “Heaney’s voice is true, it is readily apparent. That is enough. […] (128). With his onomatopoeic words, metaphors, metonymies, rhymes, rhythms and descriptions, Heaney uses his auditory imagination to locate himself in particular parts of soundscapes in the actual world and in the depths of his mind and body, in order to retrieve “the most primitive and forgotten” senses and meanings (PO 150). 258 Throughout his poetic career, Heaney developed a technique that entailed the essential patterns of his own perception and voice, a perception that re-imagined the sounds and silences he experienced and a voice that wove them into the texture of his lines. Heaney owed it, as he says, to the “creative effort of mind’s and body’s resources to bring the meaning of [sensory] experience within the jurisdiction of [poetic] form” (PO 47). These auditory experiences include sounds personally heard or actively engaged in the making, sounds remembered, imagined or only heard about. Conclusions250 The concept of auditory imagination reveals the many ways in which the Irish poet turns to listening as an effective method for memory digging, discovering, sense-making and creating. Heaney uses the concept of the auditory imagination to plumb the depths of subconscious and the hidden corners of the memory, of himself and of his readers, to discover the origins and the marvels of meaning and feeling, and to come back with the power of a muse, an inspiring poetry that amidst the most darkening and silencing threats, continues to resonate. Heaney’s poetry calls for an active practice of listening. The study of sounds in his poetry has led us to an understanding of some of the features and trends of the sounds perceived by the poet: sounds and silences that carry personal and communal implications, sounds threatened with extinction, sounds that have been indiscriminately released into the environment, sounds that have been imperialistically muffled by other sounds or rendered as noises, and sounds and voices that have been unheard, ignored or silenced. This insight has led us to a better understanding of Heaney’s poetry and his relationship with the sonic environment. It has also helped us figure out the poet’s aural sensitivity, have access to his re- construction of the past soundscapes and imagine the ideal soundscapes he longed for. Although the focus of this thesis has been to highlight the nuances of aural perception, I did not wish to place the ear at the top of a hierarchy of human sense receptors, nor did I want to treat hearing as an abstractable discipline. As readers, writers, critics or translators of 259 poetry, we are constantly engaged in the process of meaning-making and transposing our sensory experiences into language. In this thesis, I have tried to foreground the underlying relationship between what is generally known to be the abstract domain of poetry and the more practical aspect of human life, that of sounds and listening, hoping, ultimately, to broaden the readership of poetry among both academics and non-academics. The approach offered in this thesis is potentially applicable not only to Wintering Out, but also to other works by Heaney and to a variety of works by other poets − and also prose writers − whose work is, likewise, reflective of their interest in the empirical and the sensory. This thesis offers a new way to reading poetry. Conclusions250 Reading with the ears instigates a trip into the auditory imagination and puts us in touch with the poet and the soundscapes he not only heard but also attentively and consciously listened to. With the heightened attention that we bring to the reading of poetry, we can identify such auditory references. These references are often made through the onomatopoetic mirroring of sounds, sound patterns in the language, action verbs, metonymies, literal and metaphorical descriptions of sound events and silences, brief insights, passing comments and suddenly ringing phrases that slip elusively between the perceptual and the imaginary. This thesis has served as a provocation to deep listening in poetry reading, hoping to also draw the attention of literary critics and translators to the relationship between poetry and the dynamics of history, society, culture and nature. We have often let our eyes quickly scan the words to comprehend the meaning. But perhaps it is about time we adjusted our senses and let our ears be engaged too. This thesis is an invitation to pay attention to not only sound devices in the poetic language, but also the everyday sounds and silences evoked through it; to read poetry, as Heaney phrases it, “with an ear to the line” (FW 31). I As I went down the loaning the wind shifting in the hedge was like an old one’s whistling speech. And I knew 260 […] III Stand still. You can hear everything going on. High-tension cables singing above cattle, tractors, barking dogs, juggernauts changing gear a mile away. And always the surface noise of the earth you didn’t know you’d heard till a twig snapped and a blackbird’s startled volubility stopped short. When you are tired or terrified your voice slips back into its old first place and makes the sound your shades make there …. (SI 51-52) 261 Heaney, Seamus. An Open Letter. No. 2. Field Day Theatre Company, 1983. - - - . “Place and Displacement: Reflections on Some Recent Poetry from Northern Ireland.” Contemporary Irish Poetry. A Collection of Critical Essays, edited by Elmer Andrews, The MacMillan Press, 1992, pp. 124-144. - - - . 100 Poems. Faber & Faber, 2018. - - - . Crediting Poetry. Gallery Press, 1995. - - - . Death of a Naturalist. Faber & Faber, 2006. - - - . District and Circle. Faber & Faber, 2006. - - - . Door into the Dark. Faber & Faber, 1972. - - - . Electric Light. Faber & Faber, 2001. - - - . Field Work. Faber & Faber, 2001. - - - . Field Work. Faber & Faber, 2001. - - - . Finders Keepers. Selected Prose 1971-2001. Faber & Faber, 2002. - - - . Human Chain. Faber & Faber, 2012. - - - . Human Chain. Faber & Faber, 2012. - - - . New Selected Poems 1966-1987. Faber & Faber, 1990. - - - . New Selected Poems 1988-2013. Faber & Faber, 2014. - - - . North. Faber & Faber, 2001. - - - . North. Faber & Faber, 2001. - - - . Opened Ground. Selected Poems. Faber & Faber, 1998. - - - . Preoccupations. Selected Prose 1968-1978. Faber & Faber, 1980. - - - . Seeing Things. Faber & Faber, 2001. - - - . Seeing Things. Faber & Faber, 2001. - - - . Selected Poems 1965-1975. Faber & Faber, 1980. - - - . Station Island. Faber & Faber, 1984. - - - . Stations. Faber & Faber, 1975. 262 - - - . Sweeney Astray. Faber & Faber, 2001. - - - . Sweeney Astray. Faber & Faber, 2001. - - - . The Government of the Tongue. Faber & Faber, 1989. - - - . The Haw Lantern. Faber & Faber, 1987. - - - . The Haw Lantern. Faber & Faber, 1987. - - - . The Place of Writing. Scholars Print, 1989. - - - . The Redress of Poetry: Oxford Lectures. Faber & Faber, 1996. - - - . The Spirit Level. Faber & Faber, 1996. - - - . Wintering Out. Faber & Faber, 1972. “Radio Interview with Paul Vaughan.” Kaleidoscope, broadcast by BBC Radio 4, 11 October 1984, National Sound Archive Tape #T10618w. Heaney, Seamus. An Open Letter. No. 2. Field Day Theatre Company, 1983. Interview with Stephen James, 24 May 1994. Heaney, Seamus. An Open Letter. No. 2. Field Day Theatre Company, 1983. “Robert Lowell Memorial Lecture: Seamus Heaney.” YouTube, uploaded by Boston University, April 8 2010, www.youtube.com/watch?v=emuYwWT7s4A. Accessed 15 June 2019. “Seamus Heaney – Desert Island Discs – Sun 19 November 1989.” YouTube, uploaded by calumisdead, June 19 2014, www.youtube.com/watch?v=bAuucSly7t8. Accessed 25 January 2020. “Seamus Heaney ‘The Given Note’.” DailyMotion, uploaded by Sanders Aaden, 2015, www.dailymotion.com/video/x2yu7h2. Accessed 13 May 2019. “Seamus Heaney Interview \| Poetry \| Afternoon Plus \| 1980.” YouTube, uploaded by Thames TV, 26 January 2018, www.youtube.com/watch?v=3yt4m2Z4Pmw. Accessed 15 January 2020. “Seamus Heaney Interview on his Life and Career + Poetry Reading.” YouTube, uploaded by Manufacturing Intellect, 20 August 2018, www.youtube.com/watch?v=e87nqv9PiLA. Accessed 18 May 2019. “Seamus Heaney on an Evening with Irish Poet – The John Adams Institute.” YouTube, uploaded by The John Adams Institute, February 27 2019, www.youtube.com/watch?v=2bEmyan5cMA. Accessed 11 July 2019. 263 “Seamus Heaney On Poetry: The New Worker Festival.” YouTube, uploaded by The New Yorker, July 22 2014, www.youtube.com/watch?v=8HWurkQ1ao4. Accessed 15 June 2019. “Seamus Heaney Reads from His Work.” YouTube, uploaded by 92Y Plus, 13 April 2015, www.youtube.com/watch?v=tj8dATdoES0. Accessed 20 January 2019. “Seamus Heaney Recipient of The American Ireland Fund AWB Vincent Literary Award 2012.” YouTube, uploaded by irifunds, 27 June 2012, www.youtube.com/watch?v=AdTQPZGW0Bs. Accessed 20 June 2019. “Seamus Heaney with Dennis O’Driscoll.” YouTube, uploaded by Lannan Foundation, 14 February 2010, www.youtube.com/watch?v=zjV7APxLa8c. Accessed 15 January 2020. “Seamus Heaney.” YouTube, uploaded by Poetry Ireland, 26 September 2013, www.youtube.com/watch?v=wOTrSdlsPVk. Accessed 10 January 2019. “Seamus Heaney.” Charlie Rose, uploaded by Charlie Rose, charlierose.com/videos/12382. Accessed 15 January 2019. “Nobel Lecture by Seamus Heaney.” YouTube, uploaded by Nobel Prize, 30 August 2013, www.youtube.com/watch?v=P7KzfqtL5qY. Accessed 20 February 2019. Seamus Heaney. “Facts.” The Nobel Prize, “Nobel Lecture by Seamus Heaney.” YouTube, uploaded by Nobel Prize, 30 August 2013, www.youtube.com/watch?v=P7KzfqtL5qY. Accessed 20 February 2019. Seamus Heaney. “Facts.” The Nobel Prize, www.nobelprize.org/prizes/literature/1995/heaney/facts/. Accessed 10 June 2018. - - - . “Lux Perpetua. Seamus Heaney on the Making of his Recent Collection.” Our Daily Read, 16 June 2001, www.ourdailyread.com/2001/06/lux-perpetua/. Accessed 5 January 2019. - - - . “Nobel Prize Lecture. Crediting Poetry.” The Nobel Prize, www.nobelprize.org/prizes/literature/1995/heaney/lecture/. Accessed 10 June 2018. - - - . “A Political Stance.” Interview by Elgy Gillespie, The Irish Times. 19 May 1972. Accessed 10 January 2021. 264 Heaney, Seamus, and James Randall. “James Randall: From the Archive: An Interview with Seamus Heaney.” Ploughshares, vol. 37, no. 1, 2011, pp. 173–188. JSTOR, www.jstor.org/stable/41104192. Accessed 18 Mar. 2021. Heaney, Seamus. Secondary Print Sources Allen, Nicholas. “Seamus Heaney and Water.” The Irish Review, no. 49/50, 2014, pp. 173-82, JSTOR, www.jstor.org/stable/44473891. Accessed 10 June 2019. Allison, Jonathan. “Seamus Heaney’s Anti-Transcendental Corncrake.” Seamus Heaney: The Shaping Spirit, edited by Catharine Malloy and Phyllis Carey, University of Delaware Press, 1996, pp. 71-81. - - - . “Seamus Heaney and the Romantic Image.” The Sewanee Review, vol 106, no. 2, 1998, pp. 184-201, JSTOR, www.jstor.org/stable/27548503. Accessed 18 June 2020. Almqvist, Bo. “Irish Migratory Legends on the Supernatural: Sources, Studies and Problems.” Béaloideas, vol. 59, 1991, pp. 1–43, JSTOR, www.jstor.org/stable/20522374. Accessed 18 June 2019. Andrews, Elmer. Poetry of Seamus Heaney: All the Realms of Whisper. Springer, 1988. - - - . Writing Home: Poetry and Place in Northern Ireland, 1968-2008 (Vol. 2). Boydell & Brewer Ltd, 2008. Appelbaum, David. Voice. SUNY Press, 1990. Attali, Jacques. Noise: The Political Economy of Music. Manchester University Press, 1985. Auden, Wystan Hugh. The Enchafed Flood. Random House, 1950. Back, Les, and Michael Bull, editors. The Auditory Culture Reader. Beck, 2003. Bailey, Peter. Popular Culture and Performance in the Victorian City. Cambridge University Press, 1998. 265 Barber, Jesse, et al. “The Costs of Chronic Noise Exposure for Terrestrial Organisms.” Trends in Ecology and Evolution, vol. 25, 2010, pp. 180–89, doi.org/10.1016/j.tree.2009.08.002. Accessed 10 May 2019. Barfoot, Charles, and Richard Todd, editors. The Great Emporium. Rodopi, 1992. Basu, Sukanya. “Rupture and Redress—Heaney’s Poetic Landscapes.” Ruptured Landscapes, edited by Helen Sooväli-Sepping, Hugo Reinert, and Jonathan Miles-Watson, Springer, 2015, pp. 25-40. Bera, Krishnendu. “An Ecocritical Study of the Poetry of Seamus Heaney and Jibanananda Das.” Visva Bharati, 2018, pp. 1-7, shodhgangotri.inflibnet.ac.in/bitstream/123456789/6161/1/195_synopsis.pdf. Accessed 10 March 2019. Berger, Charles. “Speechlessness: Causal Attributions, Emotional Features and Social Consequences.” Journal of Language and Social Psychology, vol. 23, no. 2, 2004, pp. 147–179, doi.org/10.1177/0261927X04263821. Accessed 18 October 2019. Beville, Maria, and Sara Dybris McQuaid. “Speaking of Silence: Comments from an Irish Studies Perspective.” Nordic Irish Studies, vol. 11, no. 2, 2012, pp. 1–20, JSTOR, www.jstor.org/stable/41702633. Accessed 18 June 2020. Biagini, Eugenio, and Mary Daly, editors. The Cambridge Social History of Modern Ireland. Cambridge University Press, 2017. Bilbro, Jeffrey. “Sounding the Darkness and Discovering the Marvellous: Hearing ‘A Lough Neagh Sequence’ with Seamus Heaney’s Auditory Imagination.” Irish Studies Review, vol. 19, no. 3, 2011, pp. 321-340, doi.org/10.1080/09670882.2011.604967. Accessed 12 January 2020. Boate, Gerard. Ireland’s Natural History. Samuel Hartlib, 1652, Boate, Gerard. Ireland s Natural History. Samuel Hartlib, 1652, quod.lib.umich.edu/e/eebo/A28496.0001.001?view=toc. Accessed 25 April 2020. quod.lib.umich.edu/e/eebo/A28496.0001.001?view=toc. Secondary Print Sources Accessed 25 April 202 Boly, John R. Reading Auden: The Return of Caliban. Cornell University Press, 1991. Boynton, Robert M. “Visual and Auditory Perception.” The American Journal of Psychology, vol. 88, no. 1, 1975, pp. 155–158, JSTOR, www.jstor.org/stable/1421678. Accessed 4 October 2019. 266 Brazeau Robert, and Derek Gladwin, editors. Eco-Joyce: The Environmental Imagination of James Joyce. Cork University Press, 2014. Brazeau Robert, and Derek Gladwin, editors. Eco-Joyce: The Environmental Imagination of James Joyce. Cork University Press, 2014. Bregman, Albert. Auditory Scene Analysis: The Perceptual Organisation of Sound. MIT Press, 1990. Brinsley, John, the Elder. A Consolation for Our Grammar Schooles. Richard Field, 1622, www.questia.com/read/99077108/a-consolation-for-our-grammar-schooles. Bruneau, Thomas. “Communicative Silences: Forms and Functions.” Journal of Communication, vol. 23 no. 1, 1973, pp. 17–46, doi.org/10.1111/j.1460- 2466.1973.tb00929.x. Accessed 21 September 2019. Buell, Lawrence. The Future of Environmental Criticism: Environmental Crisis and Literary Imagination. Blackwell, 2005. Burris, Sidney. The Poetry of Resistance: Seamus Heaney and the Pastoral Tradition. Ohio University Press, 1990. - - - . “Reading Heaney.” Seamus Heaney, Poet, Critic, Translator, edited by Ashby Bland Crowder and Jason David Hall, Palgrave Macmillan, 2007, pp. 59-73. Cage, John. Silence: Lectures and Writings by John Cage. Wesleyan University Press, 1976. Cahill, Eileen. “A Silent Voice: Seamus Heaney and Ulster Politics.” Critical Quarterly, vol. 29, no. 3, 1987, pp. 55–9, doi.org/10.1111/j.1467-8705.1987.tb00091.x. Accessed 25 January 2020. Calderón Quindós, Maria Teresa. “Blending as a Theoretical Tool for Poetic Analysis: Presenting an Integrational Methodology.” Annual Review of Cognitive Linguistics, vol. 3, no. 1, 2005, pp. 269–99, doi.org/10.1075/arcl.3.14cal. Accessed 14 November 2019. Campbell, Bruce, and Elizabeth Lack. A Dictionary of Birds. T & AD Poyser, 2013. Capranica, Robert. “The Vocal Repertoir of the Bullfrog (Rana Catesbeiana).” Behaviour, vol. 31, pp. 302-25, JSTOR, www.jstor.org/stable/4533232. Accessed 18 June 2019. 267 Carr, Emily. Hundreds and Thousands. The Journals of Emily Carr. Douglas & McIntyre Ltd, 2007. Carr, Emily. Hundreds and Thousands. The Journals of Emily Carr. Douglas & McIntyre Ltd, 2007. Carson, Ciaran. “Escaped from the Massacre?” The Honest Ulsterman, no. 50, 1975, pp. 183-6. Cavanagh, Clare. “From the Republic of Conscience: Seamus Heaney and Eastern European Poetry.” Harvard Review, vol. 6, 1994, pp. 105-12, JSTOR, www.jstor.org/stable/27560061. Accessed 25 June 2019. Chamberlin, Edward J. “A New History of Reading: Hunting, Tracking and Reading.” Geography of a Soul, For the Geography of a Soul: Emerging Perspectives on Kamau Brathwaite, edited by Timothy J. Reiss, Africa World Press, 2002, pp.145-64. Chickering, Howell. Secondary Print Sources “Beowulf and ‘Heaneywulf’.” The Kenyon Review, vol. 24, no. 1, 2002, pp. 160-178, JSTOR, www.jstor.org/stable/4338314. Accessed 10 March 2019. Cioc, Mark. The Rhine: An Eco-Biography. University of Washington Press, 2006. Clarke, Eric. Ways of Listening: An Ecological Approach to the Perception of Musical Meaning. Oxford University Press, 2005. Clarkson Holstein, Suzy. “Carrying Across Into Silence: Brian Friel's Translations.” The Journal of the Midwest Modern Language Association, vol. 37, no. 2, 2004, pp. 1–10, JSTOR, www.jstor.org/stable/4144692. Accessed 10 February 2018. Corbin, Alain. Village Bells. Sound and Meaning in the 19th Century French Countryside. Columbia University Press, 1994. - - - . Time, Desire and Horror: Toward a History of the Senses. Polity Press, 1995. Corcoran, Neil. Seamus Heaney. Faber and Faber,1986. Accessed 20 February 2020. - - - . The Poetry of Seamus Heaney: A Critical Study. Faber & Faber, 1998. - - - . Poets of Modern Ireland. University of Wales Press, 1999. - - - . “Heaney and Yeats.” The Cambridge Companion to Seamus Heaney, edited by Bernard O’Donoghue, Cambridge University Press, 2009, pp. 165-177. 268 Corradi Fiumara, Gemma. The Other Side of Language. A Philosophy of Listening. Routledge, 2013. Corradi Fiumara, Gemma. The Other Side of Language. A Philosophy of Listening. Routledge, 2013. Cresswell, Tim. Place: An Introduction. John Wiley & Sons, 2014. Crotty, Patrick. “The Context of Heaney’s Reception.” The Cambridge Companion to Seamus Heaney, Cambridge University Press, 2008, pp. 37-55. Crowder, Ashby Bland. 2015. “Seamus Heaney’s Revisions for Death of a Naturalist.” New Hibernia Review / Iris Éireannach Nua, vol. 19, no. 2, 2015, pp. 94–112, www.jstor.org/stable/24625096. Accessed 23 November 2019. Cuda, Anthony. “The Use of Memory: Seamus Heaney, T.S. Eliot, and the Unpublished Epigraph to ‘North’.” Journal of Modern Literature, vol 28, no. 4, 2005, pp. 152- 75, JSTOR, www.jstor.org/stable/25167546. Accessed 20 August 2019. Cusick, Christine. Out of the Earth: Ecocritical Readings of Irish Texts. Cork University Press, 2010. - - - . “A Capacity for Sustained Flight: Contemporary Irish Poetry and the Ecology of Avian Encounter.” Animals in Irish Literature and Culture, edited by Kathryn Kirkpatrick and Borbála Faragó, Palgrave Macmillan, 2016, pp. 182-96. Dalsimer, Adele, and Adele Kreilkamp. “Reading Rita Duffy’s ‘Adele’.” Back to the Present, Forward to the Past, Volume 1. International Association for the Study of Irish Literatures. Conference, pp. 219-227. Davies, Reginald Thorne. Medieval English: A Critical Anthology. Northwestern University Press, 1991. Secondary Print Sources Dempsey, Eric, and Michael O’Clery, The Complete Guide to Ireland’s Birds. Gill & Macmillan, 2002. Derrida, Jacques. Voice and Phenomenon. Presses Universitaires de France, 1967. - - - . Spectres of Marx. Psychology Press, 1994. - - - . Spectres of Marx. Psychology Press, 1994. Devlin, Polly. All of Us There. Weidenfeld and Nicolson, 1983. Dolar, Mladen. A Voice and Nothing More. MIT Press, 2006. Eagleton, Terry. How To Read a Poem. John Wiley & Sons, 2006. 269 Eliot, Thomas Stearns. The Use of Poetry and the Use of Criticism: Studies in the Relation of Criticism to Poetry in England. Vol. 39. Harvard University Press, 1986. Falconer, Rachel. “The Music of Heaney’s Aeneid VI.” Comparative Literature, vol. 69, no. 4, 2017, pp. 430-48, Accessed 27 February 2019. Farina, Almo. Soundscape Ecology: Principles, Patterns, Methods and Applications. Springer Science & Business Media, 2013. Farren, Robert. The Course of Irish Verse in English. Sheed and Ward, 1947. Fennell, Desmond. Whatever You Say, Say Nothing: Why Seamus Heaney Is No 1. ELO Publications, 1991. Finlay, Alison. “Putting a Bawn into Beowulf.” Seamus Heaney, edited by Ashby Bland Crowder and Jason David Hall, Palgrave Macmillan, 2007, pp. 136- 154. Ashby Bland Crowder and Jason David Hall, Palgrave Macmillan, 2007, pp. 136- 154. Foster Wilson, John. The Achievement of Seamus Heaney. The Lilliput Press, 1995. Foster, Thomas C. Seamus Heaney. O’Brien Press, 1989. Freud, Sigmund. The Uncanny. Penguin Classics, 2003. Frey, Roland, and Tobias Riede. “The Anatomy of Vocal Divergence in North American Elk and European Red Deer.” Journal of Morphology, vol. 274, no. 4, pp. 274- 307, doi.org/10.1002/jmor.20092. Accessed 19 November 2019. Friel, Brian. Translations. Faber & Faber, 2000. Fussell, Paul. Poetic Meter and Poetic Form. Random House Inc, 1979. Gallwey, Sir Ralph Payne. Shooting. The Badminton Library (Moor and Marsh). Longman Greens, 1886 Garden, Alison. “We Listen for What the Waves Intone: Intertextuality and the Circum- Atlantic Poetics of Seamus Heaney, Natasha Trethewey and Kwame Dawes.” Comparative American Studies An International Journal, vol. 14, no. 2, 2016, pp. 91-108, doi.org/10.1080/14775700.2016.1242576. Accessed 15 September 2019. Garrard, Greg. Ecocriticism. Routledge, 2011. 270 Garrioch, Davis. “Sounds of the City: The Soundscape of Early Modern European Towns.” Urban History, vol. 30, no. 1, 2003, pp. 5-25, doi.org/10.1017/S0963926803001019. Accessed 30 August 2019. Gaver, William. “How Do We Hear in the World?” Explorations in Ecological Acoustic. Vol. 5, no. 4, 1993, pp. 285-313. - - - . Secondary Print Sources “What in the World Do We Hear?” Journal of Ecologic Psychology, vol. 5, no. 1, 1993, pp. 1-19. Gibson, James Jerome. The Senses Considered As Perceptual Systems. Houghton Mifflin, 1966. Gifford, Terry. Green Voices: Understanding Contemporary Nature Poetry. Manchester University Press, 1995. Gladwin, Derek. Contentious Terrains: Boglands in the Irish Postcolonial Gothic. Cork University Press, 2016. Glassie, Henry. All Silver and No Brass: An Irish Christmas Mumming. Indiana University Press, 1975. Gleeson, David. The Irish Atlantic? The Irish in the Atlantic World. University of South Carolina Press, 2010. Glob, Peter. The Bog People. Faber & Faber, 1965. Godt, Irving. “Music: A Practical Definition.” Musical Times, vol. 146, no. 1890, 2005, p. 84, JSTOR, www.jstor.org/stable/30044071. Accessed 10 December 2019. Gould, Stephen Jay. “The Origin and Function of ‘Bizarre’ Structures: Antler Size and Skull Size in the ‘Irish Elk,’ Megaloceros Giganteus.” Evolution, vol. 28, no. 2, 1974, pp. 191–220, JSTOR, www.jstor.org/stable/2407322. Accessed 10 October 2019. Nugent, Kevin. “Contraceptive Practices of Irish Married and Single First-Time Mothers.” The Journal of Biosocial Science, vol. 21, 1989, pp. 379-385, doi.org/10.1017/S0021932000018113. Accessed 11 September 2019. Gross, Harvey. Sound and Form in Modern Poetry. The University of Michigan Press, 1964. 271 Halliday, Sam. Sonic Modernity: Representing Sound in Literature, Culture and the Arts. Edinburgh University Press, 2013. Halliday, Sam. Sonic Modernity: Representing Sound in Literature, Culture and the Arts. Edinburgh University Press, 2013. Guski, Rainer. “Auditory Localization: Effects of Reflecting Surfaces.” Perception, vol. 19, no. 6, 1990, pp. 819–830, doi.org/10.1068/p190819. Accessed 8 November 2019. Hamilton, Thomas. “McCracken, Henry Joy.” In Dictionary of National Biography, edited by Sidney Lee, Smith, Elder & Co., p. 35, 1893. Handel, Stephen. Listening: An Introduction to the Perception of. Auditory Events. The MIT Press, 1989. Hardy, Thomas. The Mayor of Casterbridge, Wessex Edition, Macmillan, 1912. Accessed 15 February 2021. Harkin, Michael. “Sacred Places, Scarred Spaces.” Wicazo Sa Review, vol. 15, no.1, 2000, pp. 49-70, JSTOR, doi.org/10.1215/00104124-4260445. Accessed 21 October 2019. Harmon, Maurice. “‘We Pine for Ceremony’: Ritual and Reality in the Poetry of Seamus Heaney, 1965–75.” Seamus Heaney: A Collection of Critical Essays, edited by Elmer Andrews, Palgrave MacMillan, 1992, pp. 67-86. Hart, Henry. Seamus Heaney: Poet of Contrary Progressions. Syracuse University Press, 1993. - - - . “Heaney among the Deconstructionists.” Journal of Modern Literature, vol. 16, no. 4, 1990, pp. 461–492, JSTOR, www.jstor.org/stable/3831415. Accessed 25 August 2019. Hass, Robert, and Seamus Heaney. Sounding Lines. The Art of Translating Poetry. Secondary Print Sources Doreen B. Townsend Centre for Humanities, 2000. Heininger, Joseph. “Making a Dantean Poetic: Seamus Heaney’s ‘Ugolino’.” New Hibernia Review / Iris Éireannach Nua, vol. 9, no. 2, 2005, pp. 50-64, JSTOR, www.jstor.org/stable/20646497. Accessed 15 July 2019. Herbst, Judith. The History of Transportation. Twenty-First Century Books, 2006. 272 Herron, Tom. “Mayo - Littoral, Michael Longley's Eco Elegies.” New Hibernia Review, vol. 14, no. 4, 2010, pp. 74-89, JSTOR, www.jstor.org/stable/25801029. Accessed 15 June 2019. Hibbitts, Bernard. Coming to Our Senses: Communication and Legal Expression in Performance Cultures. University of Pittsburgh, School of Law, 1992. Hickey, Ian. “Virgilian Hauntings in the Later Poetry of Seamus Heaney.” Estudios Irlandeses, vol. 13, no. 13, 2018, pp. 27-40, www.estudiosirlandeses.org/wp- content/uploads/2018/03/DEF.HICKEY.pdf. Accessed 11 April 2019. Hickey, Kieran. Wolves in Ireland. A Natural and Cultural History. Four Courts Press, 2011. Highley, Christopher. Shakespeare, Spenser, and the Crisis in Ireland. Cambridge University Press, 2009. Hoffer, Peter Charles. Sensory Worlds in Early America. JHU Press, 2003. Hollander, John. Rhymes Reason: A Guide to English Verse. Yale University Press, 2001. Horkheimer Max, and Theodor Adorno. Dialectic of Enlightenment: Philosophical Fragments. Stanford University Press, 2002. Howes, David, editor. The Varieties of Sensory Experience: A Sourcebook in the Anthropology of the Senses. Toronto University Press, 1991. Hufstader, Jonathan. Tongue of Water, Teeth of Stones: Northern Irish Poetry and Social Violence. University Press of Kentucky, 1999. Hunter, Walt. Forms of a World: Contemporary Poetry and the Making of Globalization. Fordham University Press, 2019. Husserl, Edmund, and Martin Heidegger. The Phenomenology of Internal Time Consciousness. Indiana University Press, 1964. Ihde, Don. Listening and Voice: Phenomenologies of Sound (2nd ed). SUNY Press, 2007. Isenberg, Michael T. John L. Sullivan and His America. University of Illinois Press, 1994. 273 Jakobson, Roman and Linda Waugh. The Sound Shape of Language. Indiana University Press, 1979. Accessed 10 June 2020. Jakobson, Roman and Linda Waugh. The Sound Shape of Language. Indiana University Press, 1979. Accessed 10 June 2020. James, Stephen. “Seamus Heaney’s Sway.” Twentieth Century Literature, vol. 51, no. 3, 2005, pp. 263–284. JSTOR, www.jstor.org/stable/20058769. Accessed 27 Mar. 2021. James, Stephen. Shades of Authority: The Poetry of Lowell, Hill and Heaney. Liverpool University Press, 2007. Järviluoma, Helmi, et al, editors. Acoustic Environments in Change. University of Joensuu, 2009. Järviluoma, Helmi, and Gregg Wagstaff, editors. Soundscape Studies and Methods. Univerity of Turku, 2002. Johannesen, Richard. “The Functions of Silence: A Plea for Communication Research.” Western Speech. vol. 31, no. Secondary Print Sources 1, 1974, pp. 25-35, DOI: 10.1080/10570317409373806. Accessed 1 December 2019. Joyce, James. A Portrait of the Artist as a Young Man. Huebsch, 1916. Accessed 1 December 2019. Kay, Magdalena. Looking Eastwards. Gratitude for All the Gifts: Seamus Heaney and Eastern Europe. University of Toronto Press, 2012. Keenan, Desmond. Post-Famine Ireland: Social Structure: Ireland As It Really Was. Desmond Keenan, 2006. Kennedy, Michael, and Joyce Bourne. The Concise Oxford Dictionary of Music. Oxford University Press, 2004. Kierkegaard, Søren. Either/Or. Anchor Books, 1959 Kierkegaard, Søren. Either/Or. Anchor Books, 1959 Kierkegaard, Søren. Either/Or. Anchor Books, 1959 Kim, Jae Joon. “Silence in Seamus Heaney’s Station Island.” English 21, vol. 30, no. 4, 2017, pp. 5-27, www.earticle.net/Article/A320930. Accessed 2 November 2019. Kirkpatrick, Kathryn, and Borbála Faragó. Animals in Irish Literature and Culture. Palgrave Macmillan, 2016. 274 Kurzon, Dennis. “Towards a Typology of Silence.” Journal of Pragmatics, vol. 39, no. 10, pp. 2007, pp. 1673–88, doi.org/10.1016/j.pragma.2007.07.003. Accessed 15 January 2020. Lambin, Eric, and Helmut Geist. Land-Use And Land-Cover Change, Global Change— The IGBP Series. Springer, 2006. Lawrance, Benjamin. Amistad’s Orphans. An Atlantic Story of Slavery, Children and Smuggling. Yale University Press, 2015. Leighton, Angela. “Heaney and the Music.” The Irish Review, vol. 49/50, 2014, pp. 19- 32, JSTOR, www.jstor.org/stable/44473877. Accessed 1 September 2019. g , g y , , , pp 32, JSTOR, www.jstor.org/stable/44473877. Accessed 1 September 2019. - - - . Hearing Things: The Work of Sound in Literature. Belknap Press of Harvard University Press, 2018. Leman Stephanovic, Ingrid, and Stephen Bede Scharper, editors. The Natural City: Re- envisioning the Built Environment. University of Toronto Press, 2011. Lerer, Seth. “‘On Fagne Flor’: The Postcolonial Beowulf, from Heorot to Heaney.” Postcolonial Approaches to the European Middle Ages: Translating Cultures edited by Ananya Kabir and Deanne Williams, Cambridge University Press, 2005, pp. 54-77. Leslie, Stephen, editor. Dictionary of National Biography. Smith, Elder & Co., 1885. Lidström, Susanna. Nature, Environment and Poetry: Ecocriticism and the Poetics of Seamus Heaney and Ted Hughes. Routledge, 2015. Lloyd, David. “‘Pap for the Dispossessed’: Seamus Heaney and the Poetics of Identity.” Seamus Heaney, edited by Ashby Bland Crowder and Jason Davis Hall, Palgrave Macmillan, 1997, pp. 155-184. Longley, Edna. “Poetry and Politics in Northern Ireland.” The Crane Bag, vol. 9, no. 1, 1985, pp. 26–40, JSTOR, www.jstor.org/stable/30059561. Accessed 18 June 2019. - - - . Poetry & Posterity. Bloodaxe, 2000. - - - . Yeats and Modern Poetry. Cambridge University Press, 2013. Lowell, Robert. Life Studies. Farrar, Straus and Giroux, 1959 275 - - - . “Books of the Year.” The Observer, 14 December 1975, p. 19, https://www.nytimes.com/1982/12/05/books/notable-books-of-the-year.html. Luddy, Maria. Women in Ireland, 1800-1918. Cork University Press, 1995. Lunday, Elizabeth. “Violence and Silence in Seamus Heaney’s ‘Mycenae Lookout’.” New Hibernia Review / Iris Éireannach Nua, vol. 12, no. 1, 2008, pp. 111-27, JSTOR, www.jstor.org/stable/25660758. Accessed 18 April 2019. Mahon, Derek. Kierkegaard, Søren. Either/Or. Anchor Books, 1959 “Interview.” In the Chair: Interviews with Poets from the North of Ireland, edited by John Brown, Salmon Publishing, 2002, pp. 109–124. Maitland, Sara. A Book of Silence. Counterpoint, 2009. Maitland, Sara. A Book of Silence. Counterpoint, 2009. Manganiello, Dominic. “The Language of Exile: Heaney and Dante.” The Canadian Journal of Irish Studies, vol. 26, no. 1, 2000, pp. 100-113, www.jstor.org/stable/25515310. Accessed 10 May 2019. Mann, Thomas. The Magic Mountain. Secher and Warburg, 1927. Marryat, Frederick. A Diary in America: With Remarks on Its Institutions. Vol. 2. Green & Longmans, 1939. Marshall, John Johnson. Popular Rhymes and Sayings of Ireland. Books Ulster, 2015. Matthews, Steven. “The Object Lessons of Heaney, Carson, Muldoon and Boland.” Critical Survey, vol. 15, no.18, 2003, pp. 19-33, doi.org/10.3167/001115703782351925. Accessed 20 August 2019. doi.org/10.3167/001115703782351925. Accessed 20 August 2019. McDonald, Marianne, and Michael Walton. Amid Our Troubles: Irish Versions of Greek Tragedy. Methuen, 2002. McGuckian, Medbh. Horsepower Pass By!: A Study of the Car in the Poetry of Seamus Heaney. Cranagh Press, 1999. McKenna, Colleen. “‘A Meaning Made of Trees’: The Unwriting of a Symbol”. Seamus Heaney, edited by Jason David Hall and Ashby Bland Crowder, Palgrave Macmillan, 2007, pp. 42-58. McLiam Wilson, Robert. “The Glittering Prize.” Fortnight Review, vol. 344, 1995, pp. 23-24, JSTOR, www.jstor.org/stable/25558636. Accessed 23 February 2019. 276 McLoughlin, Nigel. “The Marvellous As We Know It: A Text World Analysis of Seamus Heaney's Squarings: Lightenings VIII.” New Writing, vol. 11, no. 2, 2013, pp. 1-12, doi.org/10.1080/14790726.2013.837934. Accessed 15 January 2019. McLuhan, Marshall. The Gutenberg Galaxy. University of Toronto, 1963. McLuhan, Marshall. The Gutenberg Galaxy. University of Toronto, 1963. - - - . Understanding Media: The Extensions of Man. MIT Press, 1994. Megec, Karlo. Old English Verse in Modern Translation: Beowulf by Seamus Heaney, JRR Tolkien and FB Gummere. University of Rjieka, 2018, PhD dissertation, repository.ffri.uniri.hr/islandora/object/ffri%3A1406/datastream/PDF/view. Accessed 11 June 2019. Meikle, James. “To Be, Or Not to Be on Syllabus.” The Guardian. Friday, April 10, 1992, p. 1, theguardian.newspapers.com/search/#query=to+be+or+not+to+be+on+syllabus& dr_year=1992-1992&offset=114. Accessed 11 June 2019. Michaels, Claire, and Claudia Carello. Direct Perception. Prentice-Hall, 1981. Moi, Ruben. “‘The Cure by Poetry that Cannot Be Coerced’: Text, Canon and Context in Seamus Heaney’s Electric Light.” Seamus Heaney, edited by Jason David Hall and Ashby Bland Crowder, Palgrave Macmillan, 2007, pp. 172-188. Molino, Michael. “Flying by the Nets of Language and Nationality: Seamus Heaney, the ‘English’ Language, and Ulster's Troubles.” Modern Philology, vol. 91, no. 2, 1993, pp. 180-201, JSTOR, www.jstor.org/stable/438827. Accessed 18 April 2019. Moore, Paul. “Sectarian Sound and Cultural Identity in Northern Ireland”. Auditory Culture Reader, edited by Michael Bull and Les Back, Routledge, 2016, pp. 251- 260. Moore, Ross. “Precursors, the Environment and the West in Seamus Heaney’s ‘Postscript’.” Estudios Irlandeses: Journal of Irish Studies, vol. 13, no. 13, 2018, pp. 69-81, doi.org/10.24162/EI2018-8092. Accessed 17 May 2019. 277 Morris, Trent. “Clean New Music: On Seamus Heaney: Trust In Rightness.” Irish Pages, vol. 8, no. 2, 2014, pp. 126–134, www.jstor.org/stable/43776041. Accessed 10 April 2019. Morrison, Blake. Seamus Heaney. Methuen, 1982. Murphy, Andrew. “Living Roots Awaken in my Head: Place and Displacement.” Seamus Heaney edited by Andrew Murphy, Liverpool University Press, 2009, pp. 8-28. Nash, Catherine. “Irish Placenames: Post‐Colonial Locations.” Transactions of the Institute of British Geographers 24.4 (1999): 457-480. Accessed 18 February 2021. Neeson, Eoin. A History of Irish Forestry. The Lilliput Press, 1990. Neeson, Eoin. A History of Irish Forestry. The Lilliput Press, 1990. Obert, Julia C. Postcolonial Overtures: The Politics of Sound in Contemporary Northern Irish Poetry. Syracuse University Press, 2015. JSTOR, www.jstor.org/stable/j.ctt1j2n7tg. Accessed 18 Mar. 2021. O'Brien, Eugene. “‘Any Catholics Among You...?’: Seamus Heaney and the real of Catholicism.” Pre-published version, 2011. Accessed 24 February 2021. O’Brien, Eugene. Seamus Heaney. Crediting Irelands of the Mind. Liffey Press, 2002. Accessed 25 February 2021. - - - . Seamus Heaney: Searches For Answers. Pluto Press, 2003. Accessed 25 February 2021. O’Brien Johnson, Toni. “Making Strange to See Afresh: Re-visionary Techniques in Some Contemporary Irish Poerry.” On Strangeness, edited by Margaret Bridges, G. Narr Verlag, 1990, pp. 133-150. Accessed 15 January 2021. O’Connor, Joseph, et al. From the Republic of Conscience: Stories Inspired by the UniversalDeclaration of Human Rights. Liberties Press, 2009. O'Donoghue, Bernard. Seamus Heaney and the Language of Poetry. Routledge, 1994. O'Donoghue, Bernard. Seamus Heaney and the Language of Poetry. Routledge, 1994. - - - . “Introduction.” The Cambridge Companion to Seamus Heaney, edited by Bernard O'Donoghue, Cambridge University Press, 2008, pp. 1-19. O'Driscoll, Dennis. “In the Mid-Course of His Life.” Hibernia, vol. 8, no. 11, 1979. 278 O’Brien, Eugene. Seamus Heaney. Crediting Irelands of the Mind. Liffey Press, 2002. - - - . Seamus Heaney: Searches For Answers. Pluto Press, 2003. Oliver, Robert T., and Dominick Barbara. The Healthy Mind in Communion and Communication. Literary Licensing, 2013. Ong, Walter Jackson. The Presence of the Word. Yale University Press, 1967. - - - . Orality and Literacy: The Technologizing of the Word. Methuen, 1982. Panzera, Daniela. “Dante’s Legacy: Kinship between Languages in Seamus Heaney’s Poetics.” Estudios Irlandeses, vol. 11, no. 11, 2016, pp. 199-212, www.estudiosirlandeses.org/wp-content/uploads/2016/03/Daniela- Panzera_EI11.pdf. Accessed 19 September 2019. Parr, Joy. “Notes for a More Sensuous History of Twentieth-century Canada: The Timely, the Tacit, and the Material Body.” Canadian Historical Review, vol. 82, no. 4, 2001, pp. 720-745, doi.org/10.3138/chr.82.4.720. Accessed, 12 June 2019. Parker, Michael. Seamus Heaney. The Making of the Poet. Palgrave Macmillan, 1993. Pater, Walter. The Renaissance. Edited and with an Introduction by Adam Philips. Oxford University Press, 1987. Paterson, Don. The Poem: Lyric, Sign, Metre. Faber & Faber, 2018. Peacock, Alan. “Mediations: Poet as Translator, Poet as Seer.” Seamus Heaney: A Collection of Critical Essays, edited by Elmer Andrews, Palgrave Macmillan, 1992, pp. 233-255. Penda, Petar. “‘The Stones of Silence’: Aesthetics of the Political in Seamus Heaney's Poetry.” ANQ, vol. 29, no. 4, 2016, pp. 268-272, doi.org/10.1080/0895769X.2016.1233804. Accessed 15 May 2019. Perkins, David. “Sounds of Seamus Heaney.” Harvard Review, vol. 10, 1996, pp. 63-69, JSTOR, www.jstor.org/stable/27560612. Accessed 18 October 2019. 279 Pijanowski, Bryan, et al. “Soundscape Ecology: The Science of Sound in the Landscape.” BioScience, vol. 61, no. 3, 2011, pp. 203-216, doi.org/10.1525/bio.2011.61.3.6. Accessed 13 October 2019, - - - . “What is Soundscape Ecology? An Introduction and Overview of an Emerging New Science.” Landscape Ecology, vol. 26. 1213-32, 2011, doi.org/10.1007/s10980-011-9600-8. Accessed 6 November 2019. Pinsky, Robert. The Sound of Poetry. Farrar, Straus and Giroux, 1998. Potts, Donna. Contemporary Irish Writing and Environmentalism. Palgrave Macmillan, 2018. Proy, Gabriele. “Sound and Sign.” Organised Sound. 2002, pp. 15-19, doi.org/10.1017/S1355771802001036. Accessed 10 February 2020. Quément, Fanny. “‘At My Buried Ear’: Seamus Heaney’s Pastoral Soundings.” Green Letters, vol. 20, no. 1, 2015, pp. 1-13, doi.org/10.1080/14688417.2015.1127175. Accessed 17 March 2019. Quinn, Justin. O'Donoghue, Bernard. Seamus Heaney and the Language of Poetry. Routledge, 1994. “Heaney and Eastern Europe.” The Cambridge Companion to Seamus Heaney, edited by Bernard O’Donoghue, Cambridge University Press, 2009, pp. 92-105. - - - . Between Two Fires: Transnationalism and Cold War Poetry. Oxford University Press, 2015. Ramazani, Jahan. “A Transnational Poetics.” American Literary History, vol. 18, no. 2, 2006, pp. 332–359, doi.org/10.1093/alh/ajj020. Accessed 14 October 2019. Rankin Russell, Richard. “Poems Without Frontiers: Poetic Reception and Political Possibility in the Work of Seamus Heaney.” Seamus Heaney, edited by Ashby Bland Crowder and Jason David Hall, Palgrave Macmillan, 2007, pp. 26-41. - - - . Poetry & Peace: Michael Longley, Seamus Heaney, and Northern Ireland. University of Notre Dame Press, 2010. - - - . Seamus Heaney’s Regions. University of Notre Dame Press, 2014. Rapaport, Herman. Later Derrida: Reading the Recent Work. Routledge, 2003. Rath, Richard Cullen. How Early America Sounded. Cornell University Press, 2003. 280 Reik, Theodor. Listening with the Third Ear. Grove Press, 1949. Reik, Theodor. Listening with the Third Ear. Grove Press, 1949. Reed, Edward, and Rebecca Jones. “Perception and Cognition: A Final Reply to Heil.” The Journal for The Theory of Social Behaviour, pp. 223-24, 1982, doi.org/10.1111/j.1468-5914.1982.tb00448.x. Accessed 2 September 2019. Richtarik, Marylinn. “Reality and Justice: Seamus Heaney’s The Cure at Troy.” Estudios Irlandeses, vol. 13, no. 13, 2018, pp. 98-112, doi.org/10.24162/EI2018-8103. Accessed 10 October 2019. Roach, Joseph. Cities of the Dead: Circum-Atlantic Performance. Columbia University Press, 1996. Robinson, Alan. “Seamus Heaney: the Free State of Image and Allusion.” Instabilities in Contemporary British Poetry, Palgrave Macmillan, 1988, pp. 123-160. Rosenberg, Bruce. “The Complexity of Oral Tradition.” Oral Tradition, vol 2., no. 2, 1987, pp. 73-90, mospace.umsystem.edu/xmlui/handle/10355/64049. Accessed 23 April 2019. Ross, Daniel. “The ‘Upward Waft’: The Influence of Frost and Eliot on Heaney’s Later Phase.” Seamus Heaney, edited by Ashby Bland Crowder and Jason David Hall, Palgrave Macmillan, 2007, pp. 92-102. Rudi, Jøran. “Soundscape and Listening.” Soundscape in the Arts, edited by Jøran Rudi, Notam, 2011, pp. 185 – 194. Ryan, Patrick. “‘When a Deeper Need Enters’: Politics in the Poetry of Yeats & Heaney.” Commonweal Magazine, vol. 141, no. 13, 2014, p. 21, www.commonwealmagazine.org/%E2%80%98when-deeper-need- enters%E2%80%99. Accessed 12 December 2019. Saintsbury, George. A History of English Prosody from the 12th Century to the Present Day. Macmillan, 1906. Sanders, Karin. Bodies in the Bog and the Archeological Imagination. University of Chicago Press, 2009.Sardello, Michael. Silence: The Mystery of Wholeness. North Atlantic Books, 2006. 281 Schaeffer, Pierre. O'Donoghue, Bernard. Seamus Heaney and the Language of Poetry. Routledge, 1994. Treatise on Musical Objects: An Essay Across Disciplines. Translated by Christine North and John Dack. University of California Press, 2017. Schafer, Raymond Murray. The Book of Noise. Price Milburn, 1970 - - - . The Soundscape. Our Sonic Environment and the Tuning of the World. Destiny Books, 1977. Schönach, Paula. “River Histories: A Thematic Review.” Water History, vol. 9, no. 3, 2017, pp. 233–57, doi.org/10.1007/s12685-016-0188-4. Accessed 22 October 2019. Seidman, Michael, and Robert Standring. “Noise and Quality of Life.” International Journal of Environmental Research and Public Health, 2010, pp. 3730-38, doi.org/10.3390/ijerph7103730. Accessed 13 March 2019. Semino, Elena. “Schema Theory and the Analysis of Text Worlds in Poetry.” Language and Literature: International Journal of Stylistics, vol. 4, no. 2, 1995, pp. 79-108, doi.org/10.1177/096394709500400201. Accessed 7 June 2019. Shaw, Brian, et al. “An Acoustic Variable Specifying Time-to-Contact.” Ecological Psychology, vol. 3, no. 3, 1991, pp. 253-61, doi.org/10.1207/s15326969eco0303_4. Accessed 10 November 2019. Simmons, Bruce, and Jennifer Scott. “The River Has Recorded the Story: Living with the Hawkesbury River, Sydney, NSW, Australia.” A History of Water. Vol. 1. Water Control and River Biographies, edited by Terje Tvedt and Eva Jacobsson, I.B. Tauris, 2006, pp. 253-76. Smilor, Raymond. “Personal Boundaries in the Urban Environment: The Legal Attack on Noise 1865-1930.” Environmental Review, vol. 3, 1979, pp. 24-36, JSTOR, www.jstor.org/stable/3984041. Accessed 9 June 2019. Smith, Bruce R. The Acoustic World of Early Modern England. Attending to the O- Factor. University of Chicago Press, 1999. Smith, Mark M. Hearing History: A Reader. University of Georgia Press, 2004. - - - . Sensing the Past: Seeing, Hearing, Smelling, Tasting, and Touching in History. University of California Press, 2007. 282 Southworth, Michael. “The Sonic Environment of Cities.” Environment and Behaviour, vol. 1, no. 1, p. 67, 1967, dspace.mit.edu/bitstream/handle/1721.1/102214/22214967-MIT.pdf?sequence=1. Accessed 7 October 2019. Snodgrass, William de Witt. Heart’s Needle. Knopf, 1959. Snodgrass, William de Witt. Heart’s Needle. Knopf, 1959. Spengler, Oswald. The Decline of the West. Knopf, 1918. Spengler, Oswald. The Decline of the West. Knopf, 1918. Stallworthy, Jon. “The Poet as Archaeologist: W. B. Yeats and Seamus Heaney.” The Review of English Studies, vol. 33, no. 130, 1982, pp. 158–174, JSTOR, www.jstor.org/stable/517203. Accessed 5 June 2019. Sterne, Jonathan. The Audible Past. Duke University Press, 2003. - - - . The Sound Studies Reader. Routledge, 2012. Suwa, Tomoaki. “An Initiation into the Other: Seamus Heaney’s Readings of W.B. Yeats Reconsidered.” Journal of Irish Studies, vol. 30, 2015, pp. - - - . The Achievement of Seamus Heaney. Lilliput Press, 1995. O'Donoghue, Bernard. Seamus Heaney and the Language of Poetry. Routledge, 1994. 49-59, JSTOR, www.jstor.org/stable/43737509. Accessed 2 March 2019. www.jstor.org/stable/43737509. Accessed 2 March 2019. The Gospel of John. General Editor, William Barclay. Westminster John Knox Press, 1975. Thomas, Ronald Stuart. An Acre of Land. Montgomeryshire Printing & Co., 1952. Thompson, Denys. The Uses Of Poetry. Cambridge University Press, 1978. Tilley, Christopher. A Phenomenology of Landscape: Places, Paths and Monuments. Berg, 1994. Tobin, Daniel. Passage to the Center: Imagination and the Sacred in the Poetry of Seamus Heaney. University Press of Kentucky, 1991. Truax, Barry, editor. Handbook for Acoustic Ecology. A.R.C. Publications, 1978. Turner, Paul. “The Cure at Troy: Sophocles or Heaney?” Seamus Heaney, edited by Ashby Bland Crowder and Jason David Lee, Palgrave Macmillan, 2007, pp. 121- 135. Tyler, Meg. “What We Leave Behind: Poetry, Music, and Seamus Heaney.” Five Points: A Journal of Literature and Art, vol. 18, no. 1, 2017, pp. 159-167, 283 go.gale.com/ps/anonymous?id=GALE%7CA496645720&sid=googleScholar&v= 2.1&it=r&linkaccess=abs&issn=10888500&p=AONE&sw=w. Accessed 8 May 2019. Vanderveer, Nancy Jean. Ecological Acoustics: Human Perception Of Environmental Sounds. UMI Dissertation Services, 1979. Vendler, Helen. Seamus Heaney. Harvard University Press, 1998. Westfahl, Gary. A Day in a Working Life: 300 Trades and Professions through History. ABC CLIO, 2015. Wheatley, David. “‘That Blank Mouth’: Secrecy, Shibboleths, and Silence in Northern Irish Poetry.” Journal of Modern Literature, vol. 25, no. 1, 2001, pp. 1-16, JSTOR, www.jstor.org/stable/3831863. Accessed 19 June 2019. - - - . “Professing Poetry: Seamus Heaney as Critic.” The Cambridge Companion to Seamus Heaney, edited by Bernard O’Donoghue, Cambridge University Press, 2008 pp. 122-135. - - - . “The Bilingual Race / And Truth of That Water: Seamus Heaney and the Irish Language.” Journal of European Studies, vol. 46, no. 1, 2016, pp. 10-23, doi.org/10.1177/0047244115617716. Accessed 13 October 2019. Whitmer, William, Seaber, Bernhard & Akeroy, Michael. “Measuring the Apparent Width of Auditory Sources in Normal and Impaired Hearing.” Basic Aspects of Hearing. Physiology and Perception, edited by Brian C. J. Moore, Robert P. Carlyon, Hedwig E. Gockel, and Roy D. Patterson, Springer, 2014, pp. 303-310. Whyte, John. “How Much Discrimination Was There Under the Unionist Regime, 1921- 1968?” Contemporary Irish Studies, edited by Tom Gallagher and James O’Connell, Manchester University Press, 1983, pp.1-36 Wilson Foster, John. “The Poetry of Seamus Heaney.” Critical Quarterly, vol. 16, no. 1, 1974, pp. 35-48, doi.org/10.1111/j.1467-8705.1974.tb01501.x. Accessed 3 August 2019. - - - . The Achievement of Seamus Heaney. Lilliput Press, 1995. 284 Wiseman, Nicholas Patrick, editor. The Dublin Review. Volume 21. O'Donoghue, Bernard. Seamus Heaney and the Language of Poetry. Routledge, 1994. Thomas Richardson and Son, 1846. books.google.co.nz/books?id=wNwsAQAAIAAJ&lpg=PA71&ots=EktDbA_azB &dq=had%20tails%20near%20a%20quarter%20of%20a%20yard%20long&hl=it &pg=PA68#v=onepage&q=had%20tails%20near%20a%20quarter%20of%20a% 20yard%20long&f=true. Accessed 12 March 2019. Wolosky, Shira. The Art of Poetry. How To Read a Poem. Oxford University Press, 2001. Wright, William D. Racism Matters. Praeger, 1998. Wylie, John. Landscape (Key Ideas in Geography). Routledge, 2007. Yeats, William Butler. Yeats Poetry, Drama and Prose. Norton Critical Editions. W.W. Norton & Company, 2000. Secondary Audio-visual Sources “Barney Devlin -The Forge.” YouTube, uploaded by Sean McCormack, February 19 2016, www.youtube.com/watch?v=eGw0NhUXG00. “Barney Devlin’s Thick Irish Accent Is….” YouTube, uploaded by Lyudmila Stashak, 19 November 2014, www.youtube.com/watch?v=IqWGTeBcxYg. “Regular Folks, Episode 1 (The Blacksmith, Barney Devlin).” YouTube, uploaded by Sessions in the House, 13 November 2015, www.youtube.com/watch?v=CDWLUL5s7NI. Some Mother’s Son. Directed by Terry George, performance by Helen Mirren, Castel Rock Entertainment, 25 December 1996. The Crying Game. Directed by Neil Jordan, performance by Stephen Rea, Palace Pictures, 20 October 1992. O’Flynn, Liam & Heaney, Seamus. The Poet and the Piper. Claddagh Records, 2003. The Visit. Directed by Orla Walsh, performance by Marcus Alastair, Grand Pictures, 1992. 285 Secondary Sitography “Ack-Ack.” Oxford English Dictionary, www-oed- com.helicon.vuw.ac.nz/view/Entry/1600?redirectedFrom=ack-ack#eid. Accessed 11 June 2019. “Acoustic Glossary.” Acoustic Glossary, www.acoustic-glossary.co.uk/. Accessed 10 July 2019. “Avibase - Bird Checklists of the World Ireland.” Avibase -The World Bird Database, 2020, avibase.bsc-eoc.org/checklist.jsp?region=IE&list=howardmoore. Accessed 11 June 2019. “Common Snipe.” Snipe Conservation Alliance, 2020, snipeconservationalliance.org/snipe-in-ireland/. Accessed 10 June 2019. “Creagh. The County Derry.” Placenamesni.org, www.placenamesni.org/resultdetails.php?entry=5352. Accessed 10 July 2019. “Cycling the Country.” National Museum of Ireland, www.museum.ie/Country- Life/Exhibitions/Current-Exhibitions/Cycling-the-Country/The-History-of-the- Bicycle. Accessed 10 December 2019. “Human Rights, Poetic Redress.” The Irish Times, 15 March 2008, www.irishtimes.com/news/human-rights-poetic-redress- 1.903757?fbclid=IwAR2VU7aYJph2dNsB8UVnU3cY0uHfX8lNjPQ08YGZnVr C5xLP9eOuk0nevVE. Accessed 25 March 2021. “Human Rights, Poetic Redress.” The Irish Times, 15 March 2008, www.irishtimes.com/news/human-rights-poetic-redress- 1.903757?fbclid=IwAR2VU7aYJph2dNsB8UVnU3cY0uHfX8lNjPQ08YGZnVr C5xLP9eOuk0nevVE. Accessed 25 March 2021. “James Randall.” Ploughshares at Emerson College, www.pshares.org/authors/james- randall. Accessed 20 December 2019. “Remembering the Unspeakable.” The Irish Times, 24 January 2001, www.irishtimes.com/culture/remembering-the-unspeakable-1.273250. Accessed 20 December 2020. “Seamus Heaney Poem Chosen as Ireland’s Best-Loved.” RTE, 11 March 2015, www.rte.ie/news/2015/0311/686370-poem/. Accessed 15 May 2019. 286 “Silence.” Oxford English Dictionary, www-oed- “Silence.” Oxford English Dictionary, www-oed- com.helicon.vuw.ac.nz/view/Entry/179646?rskey=zOuh2H&result=1#eid. Accessed 9 May 2019. “Skylark.” BirdWatchIreland, 2020, birdwatchireland.ie/birds/skylark/. “Unheimlich.” Oxford English Dictionary, www-oed- com.helicon.vuw.ac.nz/view/Entry/214140?redirectedFrom=unheimlich#eid. Accessed 9 May 2019. “Welcome to Burren National Park.” Burren National Park, 2017, www.burrennationalpark.ie/. Accessed 10 May 2019. “Welcome to the Story About Tollund Man.” Museum Silkeborg, www.tollundman.dk/. Accessed 9 May 2019. “Welcome to the World Forum for Acoustic Ecology.” World Forum of Acoustic Ecology, 2018, www.wfae.net/. Accessed 18 August 2020. “World Soundscape Project.” Wikiwand, www.wikiwand.com/en/World_Soundscape_Project. Accessed 8 January 2019. Bloom, Harold, and Pamela Loos. “Bloom’s Shakespeare through the Ages. As You Like It.” EPDF, 2020, epdf.pub/as-you-like-it-blooms-shakespeare-through-the- ages.html. Accessed 18 May 2020. Brady, Sasha. “Seamus Heaney’s Poems To Be Displayed on Public Transport in Ireland.” Lonely Planet, 11 April 2019, www.lonelyplanet.com/articles/seamus- heaney-public-transport. Accessed 20 July 2020. Brandes, Rand. “Sorting Seamus Heaney’s Study: I Could Hear Archivists Groan As We Threw Things Away.” The Irish Times, 18 May 2017, www.irishtimes.com/culture/books/sorting-seamus-heaney-s-study-i-could-hear- archivists-groan-as-we-threw-things-away-1.3087628. Accessed 25 June 2020. Brice, Colette. “Omphalos. Returning to the Troubles of a Northern Irish Childhood.” Poetry Foundation, 1 October 2014, www.poetryfoundation.org/poetrymagazine/articles/70148/omphalos. Accessed 25 June 2020. 287 Brown, Mark. “New Seamus Heaney Poem Published.” The Guardian, 25 October 2013, www.theguardian.com/books/2013/oct/25/seamus-heaney-last-poem-published. Accessed 22 May 2019. Bruce, David. “The Given Note.” Davidbruce, 2020, www.davidbruce.net/works/the- given-note.asp. Accessed 20 June 2020. Clines, Francis X. “Poet of the Bogs.” The New York Times, 13 March 1983, www.nytimes.com/1983/03/13/magazine/poet-of-the-bogs.html. Accessed 15 June 2020. - - - . “Poet of Silent Things.” The New York Times, 30 August 2013, www.nytimes.com/2013/08/31/opinion/seamus-heaney-poet-of-the-silent- things.html. Accessed 15 June 2020. Cole, Henry. “Seamus Heaney. The Art of Poetry.” Ricorso, 4 March 2014, www.ricorso.net/rx/library/criticism/revue/Zundry_AZ/Cole_H.htm. Accessed 10 June 2019. Cunningham, Erin. “The Wolf in Irish Poetry.” The Learned Pig, 28 March 2017, http://www.thelearnedpig.org/wild-alphabet-wolf-irish-poetry/4386. Accessed 15 July 2020. Deane, Seamus. “‘Unhappy and at Home’, Seamus Deane Interviews Seamus Heaney.” Deane, Seamus. “‘Unhappy and at Home’, Seamus Deane Interviews Seamus Heaney.” Ricorso, www.ricorso.net/rx/library/criticism/classic/Anglo_I/Heaney_S/Unhappy.htm. Accessed 9 May 2019. Doyle, Sean. “The Rain Stick.” John Sean Doyle, 18 January 2014, johnseandoyle.com/the-rain-stick/. Accessed 15 June 2019. Farndale, Nigel. “Seamus Famous.” The Telegraph, 5 April 2001, l h k/ l /4722682/S F h l A d 10 Farndale, Nigel. “Seamus Famous.” The Telegraph, 5 April 2001, www.telegraph.co.uk/culture/4722682/Seamus-Famous.html. Accessed 10 August 2019. www.telegraph.co.uk/culture/4722682/Seamus-Famous.html. Accessed 10 August 2019. Fawbert, David. “Dawn.” Fawbie, fawbie.info/wintering-out/dawn/. Accessed 10 June 2019. - - - . “Summer Home.” Fawbie, fawbie.com/2016/05/31/summer-home/. Accessed 9 May 2019. 288 - - - . “Wedding Day.” Fawbie, fawbie.info/wintering-out/wedding-day/. Accessed 22 May 2019. Gillan, PJ. “Seán Ó hEochaidh. “Silence.” Oxford English Dictionary, www-oed- Collector of Ireland's Rich Folkore Heritage.” The Guardian, 4 February 2002, www.theguardian.com/news/2002/feb/04/guardianobituaries.books1. Accessed 11 June 2019. Harrison, Claire. “How Belfast Telegraph Gave Seamus Heaney an Early Break.” Belfast Telegraph, 3 September 2013, www.belfasttelegraph.co.uk/news/northern- ireland/how-belfast-telegraph-gave-seamus-heaney-an-early-break- 29546167.html. Accessed 10 February 2019. Jones, Calvin. “Turdus Merula.” Ireland’s Wildlife, 13 April 2011, www.irelandswildlife.com/blackbird-turdus-merula/. Accessed 10 February 2019. - - - . “Atlantic Mackerel. (Scomber Scombrus).” Ireland’s Wildlife, 7 July 2011, irelandswildlife.com/atlantic-mackerel-scomber-scombrus/. Accessed 11 February 2019. - - - . “Harbour Porpoise (Phocoena phocoena).” Ireland’s Wildlife, 21 April 2014, www.irelandswildlife.com/harbour-porpoise-phocoena-phocoena/. Accessed 10 February 2019. Leith, Sam. “Return of the Naturalist.” The Telegraph, 2 April 2006, www.telegraph.co.uk/culture/books/3651313/Return-of-the-naturalist.html. Accessed 13 October 2019. McGrath, Dominic. “The Story Behind the Seamus Heaney Quote Guiding People Through the Covid-19 Crisis.” The Journal.ie, 5 April 2020, www.thejournal.ie/seamus-heaney-quote-winter-this-one-out-summer-5065553- Apr2020/?utm_source=shortlink. Accessed 15 October 2019. McKittrick, David. “Not Everyone is Weeping as the Last Ship Leaves ‘Titanic Town’. The End of Belfast’s Shipbuilding Industry.” The Independent, 16 March 2003, www.independent.co.uk/voices/commentators/david-mckittrick-not-everyone-is- weeping-as-the-last-ship-leaves-titanic-town-5352380.html. Accessed 16 November 2019. 289 Melaugh, Martin. “Examples of IRA Posters.” CAIN Web Service, 2 July 2019, cain.ulster.ac.uk/images/posters/ira/index.html. Accessed 16 January 2019. cain.ulster.ac.uk/images/posters/ira/index.html. Accessed 16 January 2019. Murray, Alice Effie. “The Irish Woollen Industry.” CELT – Corpus of Electronic Texts Edition, celt.ucc.ie/published/E900040/text007.html. Accessed 9 May 2019. Paretts, Susan. “Indigenous Animals in Ireland.” PetsOnMom, 2020, Paretts, Susan. “Indigenous Animals in Ireland.” PetsOnMom, 2020, animals.mom.me/indigenous-animals-ireland-5197.html. Accessed 19 January Paretts, Susan. “Indigenous Animals in Ireland.” PetsOnMom, 2020, animals.mom.me/indigenous-animals-ireland-5197.html. Accessed 19 January 2019. Rumens, Carol. “Poem of the week: The Blacksmiths.” The Guardian, 20 February 2012, www.theguardian.com/books/2012/feb/20/poem-of-the-week-the- blacksmiths. Accessed 16 March 2019. Schine, Jennifer. “Memory, Movement & the Senses in Soundscape Studies.” Sensory Studies, www.sensorystudies.org/sensorial-investigations/movement-memory- the-senses-in-soundscape-studies/. Accessed 9 May 2019. Spenser, Edmund. “A View on the Present State of Ireland.” CELT: The Corpus of Electronic Texts, 2019, celt.ucc.ie//published/E500000-001/. Accessed 16 May 2019. Watchel, Eleanor. “An Interview with Seamus Heaney.” Brick. A Literary Journal, 30 September 2013, brickmag.com/an-interview-with-seamus-heaney/. Accessed 18 May 2019. Young, Connla.“Heaney’s Townland Doesn’t Exist.” The Irish News, 9 September 2013, www.irishnews.com/news/2013/09/09/news/heaney-townland-doesn-t-exist- 69745/. Accessed 21 March 2019.
https://openalex.org/W2744997356	https://hess.copernicus.org/articles/22/819/2018/hess-22-819-2018.pdf	English	null	Technical note: Using distributed temperature sensing for Bowen ratio evaporation measurements	Hydrology and earth system sciences	2,018	cc-by	10,238	Correspondence: Bart Schilperoort (b.schilperoort@tudelft.nl) Correspondence: Bart Schilperoort (b.schilperoort@tudelft.nl) Received: 21 July 2017 – Discussion started: 4 August 2017 Revised: 27 November 2017 – Accepted: 20 December 2017 – Published: 30 January 2018 Received: 21 July 2017 – Discussion started: 4 August 2017 Revised: 27 November 2017 – Accepted: 20 December 2017 – Published: 30 January 2018 Abstract. Rapid improvements in the precision and spatial resolution of distributed temperature sensing (DTS) technol- ogy now allow its use in hydrological and atmospheric sci- ences. Introduced by Euser et al. (2014) is the use of DTS for measuring the Bowen ratio (BR-DTS), to estimate the sensible and latent heat ﬂux. The Bowen ratio is derived from DTS-measured vertical proﬁles of the air temperature and wet-bulb temperature. However, in previous research the measured temperatures were not validated, and the cables were not shielded from solar radiation. Additionally, the BR- DTS method has not been tested above a forest before, where temperature gradients are small and energy storage in the air column becomes important. (EC) estimates (r2 = 0.59). The average energy balance clo- sure between BR-DTS and EC is good, with a mean under- estimation of 3.4 W m−2 by the BR-DTS method. However, during daytime the BR-DTS method overestimates the avail- able energy, and during night-time the BR-DTS method esti- mates the available energy to be more negative. This differ- ence could be related to the biomass heat storage, which is neglected in this study. The BR-DTS method overestimates the latent heat ﬂux on average by 18.7 W m−2, with RMSE = 90 W m−2. The sen- sible heat ﬂux is underestimated on average by 10.6 W m−2, with RMSE = 76 W m−2. Estimates of the BR-DTS can be improved once the uncertainties in the energy balance are reduced. However, applying, for example, Monin–Obukhov similarity theory could provide independent estimates for the sensible heat ﬂux. This would make the determination of the highly uncertain and difﬁcult to determine net available en- ergy redundant. In this paper the accuracy of the wet-bulb and air tempera- ture measurements of the DTS are veriﬁed, and the resulting Bowen ratio and heat ﬂuxes are compared to eddy covariance data. The performance of BR-DTS was tested on a 46 m high tower in a mixed forest in the centre of the Netherlands in August 2016. Correspondence: Bart Schilperoort (b.schilperoort@tudelft.nl) The average tree height is 26 to 30 m, and the temperatures are measured below, in, and above the canopy. Using the vertical temperature proﬁles the storage of latent and sensible heat in the air column was calculated. 1 Introduction We found a signiﬁcant effect of solar radiation on the tem- perature measurements, leading to a deviation of up to 3 K. By installing screens, the error caused by sunlight is reduced to under 1 K. Wind speed seems to have a minimal effect on the measured wet-bulb temperature, both below and above the canopy. After a simple quality control, the Bowen ra- tio measured by DTS correlates well with eddy covariance In recent years distributed temperature sensing (DTS) tech- nology has quickly improved (Bao and Chen, 2012). The precision and spatial resolution now allow its widespread use in hydrological and atmospheric sciences (Selker et al., 2006; Thomas et al., 2012), from measuring groundwater ﬂow (Blume et al., 2013) and seepage into streams (West- hoff et al., 2007) to soil moisture (Steele-Dunne et al., 2010), Hydrol. Earth Syst. Sci., 22, 819–830, 2018 https://doi.org/10.5194/hess-22-819-2018 © Author(s) 2018. This work is distributed under the Creative Commons Attribution 4.0 License. Hydrol. Earth Syst. Sci., 22, 819–830, 2018 https://doi.org/10.5194/hess-22-819-2018 © Author(s) 2018. This work is distributed under the Creative Commons Attribution 4.0 License. Technical note: Using distributed temperature sensing for Bowen ratio evaporation measurements Bart Schilperoort1, Miriam Coenders-Gerrits1, Willem Luxemburg1, César Jiménez Rodríguez1,3, César Cisneros Vaca2, and Hubert Savenije1 1Delft University of Technology, Water Resources Section, Stevinweg 1, 2628 CN Delft, the Netherlands 2University of Twente, Faculty of Geo-Information Science and Earth Observation (ITC), Hengelosestraat 99, 7514 AE, Enschede, the Netherlands 3Tecnológico de Costa Rica, Escuela de Ingeniería Forestal. 159-7050, Cartago, Costa Rica Bart Schilperoort1, Miriam Coenders-Gerrits1, Willem Luxemburg1, César Jiménez Rodríguez1,3, César Cisneros Vaca2, and Hubert Savenije1 1Delft University of Technology, Water Resources Section, Stevinweg 1, 2628 CN Delft, the Netherlands 2University of Twente, Faculty of Geo-Information Science and Earth Observation (ITC), Hengelosestraat 99, 7514 AE, Enschede, the Netherlands 3Tecnológico de Costa Rica, Escuela de Ingeniería Forestal. 159-7050, Cartago, Costa Rica Published by Copernicus Publications on behalf of the European Geosciences Union. The Bowen ratio (β) is the ratio of the sensible heat ﬂux to the latent heat ﬂux and can be approximated using the air temperature gradient and the vapour pressure difference over the height (Bowen, 1926) β = H ρλE ≈γ 1Ta 1ea , (2) (2) where γ is the psychrometric constant (kPa K−1) (see Eq. 10), 1Ta the difference in air temperature between two heights (K) and 1ea the difference in actual vapour pressure between the two heights (kPa). However, when gradients are very small, the adiabatic lapse rate cannot be neglected (Barr et al., 1994). Therefore the potential temperature should be used instead: where γ is the psychrometric constant (kPa K−1) (see Eq. 10), 1Ta the difference in air temperature between two heights (K) and 1ea the difference in actual vapour pressure between the two heights (kPa). However, when gradients are very small, the adiabatic lapse rate cannot be neglected (Barr et al., 1994). Therefore the potential temperature should be used instead: In addition to estimating the latent and sensible heat ﬂux, the measurements can also be used to get a better understand- ing of the processes taking place in complex ecosystems, such as forests. A vertical temperature and humidity proﬁle is available in high resolution and precision, both above, in- side, and under the canopy. DTS can also estimate different components of the energy balance, such as the heat storage in the air column, and the soil heat ﬂux (Jansen et al., 2011). Finally, it can be used to increase our understanding of the energy exchange between the canopy and undergrowth layers by looking at the air temperature gradient under the canopy. β = H ρλE = cp λ ∂2/∂z ∂q/∂z = γ ∂2/∂z ∂ea/∂z, (3) (3) where cp is the speciﬁc heat of air (MJ kg−1) (see Eq. 6), λ the latent heat of vaporization (2.45 MJ kg−1 K−1), 2 the potential temperature (K), q the speciﬁc humidity (kg kg−1) (see Eq. 7) and z the height above the ground (m). The poten- tial temperature gradient can be approximated by the right- hand side of Eq. (4), as the ratio 2 Ta is nearly 1 (Pal Arya, 1988): This paper elaborates on the method of Euser et al. Published by Copernicus Publications on behalf of the European Geosciences Union. Published by Copernicus Publications on behalf of the European Geosciences Union. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements 820 soil heat ﬂux (Bense et al., 2016), and wind speed (Sayde et al., 2015). First introduced by Euser et al. (2014), DTS can also be used for measuring the Bowen ratio, to estimate the evaporation ﬂux. A dry and wet stretch of the same ﬁbre optic cable are installed vertically to obtain the so-called dry- and wet-bulb temperature gradient, respectively. This method mitigates some problems of the conventional Bowen ratio, since usually at least two different sensors are used to mea- sure the temperature and vapour pressure gradients, of which each has its own independent error (Angus and Watts, 1984; Fuchs and Tanner, 1970). The DTS-based Bowen ratio does not suffer from this drawback, by having a large amount of data points over the height (up to 8 per metre) with only a single sensor. It also has a resolution of 0.06 K for 1 min av- erages (Silixa machine calibration), and will be more accu- rate when measuring over a longer time period, allowing for very small temperature gradients to be measured. heat ﬂux (W m−2), and dQ dt is the change of energy storage in the system (W m−2). A represents a net advection of energy into the system (W m−2), but is assumed to be 0. The energy ﬂux associated with photosynthesis (GP) was not measured, and is therefore not included in the equation. The Bowen ratio (β) is the ratio of the sensible heat ﬂux to the latent heat ﬂux and can be approximated using the air temperature gradient and the vapour pressure difference over the height (Bowen, 1926) heat ﬂux (W m−2), and dQ dt is the change of energy storage in the system (W m−2). A represents a net advection of energy into the system (W m−2), but is assumed to be 0. The energy ﬂux associated with photosynthesis (GP) was not measured, and is therefore not included in the equation. 2 Materials and methods cp = 1.004 + 1.84q (6) (6) Published by Copernicus Publications on behalf of the European Geosciences Union. (2014), by considering more energy balance components like the la- tent and sensible heat storage in the air column, including a data-quality system, and using the potential air temperature. The performance of the method is tested in a mixed forest in the Netherlands by looking at the accuracy of the DTS- measured air temperature and wet-bulb temperature, com- pared to reference temperature and humidity sensors. It ap- pears that solar radiation can have a signiﬁcant inﬂuence on the cable temperature, which can be mitigated by providing artiﬁcial shadow. Lastly, the ﬂuxes resulting from the method are compared to an eddy covariance (EC) system, and the sources of differences between the methods are shown. ∂2 ∂z = 2 Ta ∂Ta ∂z + 0 ≈∂Ta ∂z + 0, (4) (4) where Ta is the air temperature (K), and 0 is the adiabatic lapse rate (typically around 0.01 K m−1). The numerical im- plementations of Eqs. (3) and (4) are explained in Sect. 3.2. Under dry and unsaturated conditions the lapse rate is equal to (Pal Arya, 1988) where Ta is the air temperature (K), and 0 is the adiabatic lapse rate (typically around 0.01 K m−1). The numerical im- plementations of Eqs. (3) and (4) are explained in Sect. 3.2. Under dry and unsaturated conditions the lapse rate is equal to (Pal Arya, 1988) 0 = g cp , (5) 0 = g cp , (5) where g is the gravitational acceleration (9.81 m s−2). The speciﬁc heat capacity of air is determined by (Stull, 2015) 2.1 Theory and the speciﬁc humidity by (Pal Arya, 1988) and the speciﬁc humidity by (Pal Arya, 1988) The Bowen ratio energy balance method (BREB) combines the energy balance with the Bowen ratio (Oliphant et al., 2004). The energy balance can be described by q = εea P , (7) q = εea P , (7) RN + A = ρλE + H + GS + dQ dt , (1) where ε is the ratio of molecular mass of water vapour to dry air (0.622), and P the atmospheric pressure (kPa). The actual vapour pressure is determined by (Allen et al., 1998) where ε is the ratio of molecular mass of water vapour to dry air (0.622), and P the atmospheric pressure (kPa). The actual vapour pressure is determined by (Allen et al., 1998) (1) where RN is the net radiation (W m−2), ρλE the latent heat ﬂux (W m−2), H the sensible heat ﬂux (W m−2), GS the soil ea(Ta) = es(Tw) −γ (Ta −Tw), (8) (8) www.hydrol-earth-syst-sci.net/22/819/2018/ www.hydrol-earth-syst-sci.net/22/819/2018/ Hydrol. Earth Syst. Sci., 22, 819–830, 2018 Hydrol. Earth Syst. Sci., 22, 819–830, 2018 oort et al.: Using DTS for Bowen ratio evaporation measurements 821 B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements 5°41'30" E 5°41'0" E 52°15'15" N 52°15'0" N 0 0.25 0.5 0.75 1 0.125 km Flux tower 500 m radius Forest type Broadleaf Coniferous Forest gap Mixed forest Germany France North Sea Speulderbos North Sea Germany Belgium Figure 1. Forest type distribution within 500 m of the tower site at Speulderbos Forest, the Netherlands. 5°41'30" E 5°41'0" E 52°15'15" N 52°15'0" N 0 0.25 0.5 0.75 1 0.125 km Figure 1. Forest type distribution within 500 m of the tower site at Speulderbos Forest, the Netherlands. age of heat in the form of water vapour in the air column: where Tw is the wet-bulb temperature (K), and es the satura- tion vapour pressure (kPa) given by (Koutsoyiannis, 2012) dQ dt = dQH dt + dQE dt . (14) (14) es(Tw) = 0.61 · exp 19.9 · Tw 273 + Tw . (9) (9) The changes in storage of heat and water vapour in the air column below the height at which the energy ﬂuxes (RN, H and ρλE) are measured are represented by dQH dt and dQE dt respectively (W m−2). 2.1 Theory The change in biomass heat storage ( dQB dt ) was not measured, and is therefore not included in this equation. dQH dt and dQE dt are deﬁned as (Barr et al., 1994) The psychrometer constant is related to the air pressure and ventilation of the psychrometer (Harrison and Wood, 2012; Allen et al., 1998). If sufﬁciently ventilated, the psychromet- ric constant is deﬁned by (Allen et al., 1998) γ = cpP ελ = 0.665 × 10−3 · P. (10) (10) dQH dt = z Z 0 ρacp dTa dt dz, (15) dQE dt = z Z 0 ρaλdq dt dz. (16) (15) As the air pressure also varies over height, the measurements have to be corrected for elevation using the following approx- imation (Stull, 2015, p. 8): (16) P(z) = P0 · exp(−z/7290) (11) (11) with P0 being the pressure at sea level (kPa). By combining the Bowen ratio (Eq. 3) with the energy balance (Eq. 1), the latent heat ﬂux and sensible heat ﬂux can be determined: 3 Study site H = RN −GS −dQ dt 1 + 1 β , (12) ρλE = RN −GS −dQ dt 1 + β . (13) The measurements were carried out at the Speulderbos mixed forest (52◦15′4′′N, 5◦41′24′′E), on a tower located within a patch of Douglas ﬁr trees (Pseudotsuga menziesii (Mirb.) Franco) of 2.5 ha in Garderen, the Netherlands (Fig. 1). The surrounding area is characterized by the presence of broadleaved and coniferous tree species, distributed in blocks around the tower site (Bosveld and Bouten, 2001). Within a (12) (13) The storage component in the energy balance has multiple parts, ranging from the storage of heat in the soil, to the stor- The storage component in the energy balance has multiple parts, ranging from the storage of heat in the soil, to the stor- www.hydrol-earth-syst-sci.net/22/819/2018/ B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements Some coniferous canopies like the Douglas ﬁr have a canopy height between 26 and 30 m, while the broadleaved stands can reach up to 30 m height for old-growth beech trees (Weligepolage et al., 2012; Wilkes et al., 2017), or heights under 10 m for smaller pedunculate oak trees. 500 m radius it is possible to ﬁnd native tree species such as beech (Fagus sylvatica L.), pedunculate oak (Quercus robur L.) and Scots pine (Pinus sylvestris L.), as well as the in- troduced species hemlock (Tsuga heterophylla (Raﬁnesque) Sargent) and Japanese larch (Larix kaempferi (Lambert) Car- riére) (Erisman et al., 1998; Raj et al., 2014; Su et al., 2009; Bosveld and Bouten, 2001; Tietema et al., 2002; Van Wijk et al., 2000; Weligepolage et al., 2013). Canopy heights dif- fer between cover types depending on species and growing stage. Some coniferous canopies like the Douglas ﬁr have a canopy height between 26 and 30 m, while the broadleaved stands can reach up to 30 m height for old-growth beech trees (Weligepolage et al., 2012; Wilkes et al., 2017), or heights under 10 m for smaller pedunculate oak trees. The DTS machine used was the Silixa Ultima (Silixa Ltd, 2017), which has a sampling resolution of 12.5 cm, measure- ment resolution of 35 cm, and a measurement standard devi- ation of 0.06 K at a 1 min time resolution. The study site has an oceanic climate (Cfb) under the Köpen classiﬁcation system, with a yearly average temper- ature of 9.8 ◦C and an average precipitation of 910 mm yr−1 (Sluijter, 2011). The topography is slightly undulating with smooth height differences (Raj et al., 2014), a well-drained soil, and a groundwater table below 40 m depth (Tiktak and Bouten, 1994). The soil texture ranges from ﬁne sand to sandy loam (Weligepolage et al., 2012; Tietema et al., 2002; Van Wijk et al., 2000). The ﬁbre optic cable with a diameter of 6 mm was secured at the top of the tower, with the dry stretch hanging 1.2 m away from the tower, and the wet stretch 0.25 m away. The cable with a diameter of 3 mm was secured next to the dry 6 mm cable. The response times of the cables are of the order of 2–3 min for the 6 mm cable, and 20 to 40 s for the 3 mm ca- ble. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements The cables were secured at multiple locations distributed over the height (in and above the canopy; see Fig. 2), us- ing loops (with a diameter of 5 cm) to prevent direct contact with the support structure. For both cables a stretch of 10 m at both the start and end was placed in a calibration bath, an enclosed styrofoam box ﬁlled with water, along with two B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements Ultima 46 m 42 m 38 m 36 m 32 m 24 m 26 m 20 m 16 m 4 m 36 m Temperature and humidity sensor Eddy covariance setup Cup anemometer Net radiometer Wetted fibre optic cable (6 mm) Dry fibre optic cables (6 mm, 3 mm) Pump and water supply Calibration bath Soil heat flux setup 38 m 46 m 48 m 44 m Water recapture bucket Shade shielding Maximum canopy height Average canopy height Start canopy Figure 2. Schematic overview of the measurement setup at the tower. Ultima 46 m 42 m 38 m 36 m 32 m 24 m 26 m 20 m 16 m 4 m Maximum canopy height Average canopy height Start canopy Figure 2. Schematic overview of the measurement setup at the tower. Shade shielding Figure 2. Schematic overview of the measurement setup at the tower. ameter of 6 mm and has both a dry and a wetted stretch. To wet the cable it was wrapped in cloth, and water was supplied to it continuously. A second cable with a diameter of 3 mm was used to study the effects of solar radiation, as a thinner cable will warm up less (De Jong et al., 2015). However, this method added additional uncertainties due to the required ex- trapolation and the 3 mm cable was not used in this study. (While correlation with reference sensors improved, the un- certainty of extrapolation caused extra noise in the Bowen ratio calculations). Both cables were connected to the same DTS machine (in single-ended mode) and calibrated in a cal- ibration bath (see Fig. 2) 500 m radius it is possible to ﬁnd native tree species such as beech (Fagus sylvatica L.), pedunculate oak (Quercus robur L.) and Scots pine (Pinus sylvestris L.), as well as the in- troduced species hemlock (Tsuga heterophylla (Raﬁnesque) Sargent) and Japanese larch (Larix kaempferi (Lambert) Car- riére) (Erisman et al., 1998; Raj et al., 2014; Su et al., 2009; Bosveld and Bouten, 2001; Tietema et al., 2002; Van Wijk et al., 2000; Weligepolage et al., 2013). Canopy heights dif- fer between cover types depending on species and growing stage. www.hydrol-earth-syst-sci.net/22/819/2018/ Hydrol. Earth Syst. Sci., 22, 819–830, 2018 822 B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements Figure 3. Schematic of one 2 m segment of the solar screen con- struction. Pt100 temperature probes that were connected to the DTS machine. An air bubbler was installed in the styrofoam box to ensure a homogeneous temperature distribution. The ca- bles were shielded from direct solar radiation using screen gauze secured onto PVC rings; see Fig. 3. Only the south- ern 180◦of the cables was shielded, to allow for sufﬁcient ventilation. The screen gauze had holes 1.5 mm wide, and the mesh material had a diameter of 0.3 mm. Two layers of the gauze were used. Each segment of shield was 2 m long, and was secured to the tower with a horizontal beam. Due to the angle of the incident sunlight the gauze was able to block most direct sunlight, except during the early morning. To supply the wet cable with water, a reservoir was installed near the top of the tower, along with a pump. The pump speed was set to 1500 mL h−1 during sunny days without rainfall, and to 800 mL h−1 on other days, which was enough to keep the cable wet over the entire height, while keeping the in- ﬂuence of relatively warm water at the top of the cable at a minimum. As water supplied at the top has a higher temper- ature than the wet-bulb temperature, the top 2 m of wet cable data was excluded from the data analysis to allow the slowly ﬂowing water to reach the wet-bulb temperature. Figure 3. Schematic of one 2 m segment of the solar screen con- struction. Figure 3. Schematic of one 2 m segment of the solar screen con- struction. measurements done at 8 cm. The soil heat ﬂux was then de- termined using the harmonics method (van der Tol, 2012). A net radiometer (Kipp & Zonen CNR4) was located on the top of the tower (48 m), measuring both incoming and outgoing short- and longwave radiation. One-minute aver- ages were logged. On the tower six humidity and tempera- ture sensors were located over the height, at 4, 16, 24, 32, 36 and 46 m a.g.l. (above ground level). The lower four were Rotronic HC2-S3C03 sensors (with active ventilation), and the top two were Campbell CS216 sensors with passive ven- tilation. The sensors were inter-calibrated to the sensor at 24 m. 3.2 Data processing The DTS machine was set to measure the cable temperature at 1 min averaging intervals. For the comparison with ref- erence temperature sensors, these 1 min resolution data are used. To compare the wet-bulb temperature measured by the ﬁbre optic cable to the reference sensors, the reference wet- bulb temperature is iteratively derived from the reference air temperature and relative humidity. For the purpose of cal- culating the Bowen ratio, the temperature and actual vapour pressure are averaged over time for 15 min time periods. For DTS Bowen ratio calculations, the temperatures between 38.5 and 44 m are used. This area is shaded from the sun by the screen gauze, and at the top of the stretch the new water on the wet cable has reached the wet-bulb temperature. At the top of the tower an EC system was installed to measure the sensible and latent heat ﬂuxes. It consisted of a Campbell CSAT3 sonic anemometer and a LI-COR Bio- sciences LI7500 gas analyser connected to a CR5000 Camp- bell data logger, to which the data were logged at 20 Hz. Two cup anemometers (Onset S-WSB-M003) were used to measure the wind speed, one at the top of the tower (48 m), and one below the canopy (4 m). The data from the lower anemometer lacks the resolution to properly measure the low wind speeds below the canopy, which are at times too low to be registered. One-minute average wind speeds, along with the maximum gust speeds, were logged. When calculating the gradients for the Bowen ratio, the 15 min average temperature and vapour pressure are ﬁt to the natural logarithm of the height, in the following form: Ta, ﬁt = a · ln(z) + b. (17) (17) The biomass heat storage change and the photosynthesis energy ﬂux were not measured. The biomass heat storage change is estimated to have a maximum of 45 W m−2, and the photosynthesis energy ﬂux is estimated to be of the order of 5 W m−2 (Barr et al., 1994; Michiles and Gielow, 2008). For the soil heat ﬂux, the soil temperature was measured at different depths (1, 3, 4, 8, 20, 50 cm). Soil moisture was measured using Campbell Sci. Inc. CS616 water content re- ﬂectometers. Thermal conductivity was ﬁtted to soil heat ﬂux A logarithmic shape of the proﬁles was assumed based on Monin–Obukhov similarity theory. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements The temperature and humidity was logged at 1 min av- erages. 3.1 Setup The temperature of ﬁbre optic cables is measured using the DTS technique (Selker et al., 2006). In the setup, two cables with different diameters were used. The ﬁrst cable has a di- www.hydrol-earth-syst-sci.net/22/819/2018/ Hydrol. Earth Syst. Sci., 22, 819–830, 2018 823 www.hydrol-earth-syst-sci.net/22/819/2018/ 4.2 Temperature validation In Fig. 5 the comparison between the 6 mm DTS cable and the reference sensor is shown. For the above-canopy compar- ison, the 46 m reference sensor is compared to the cable tem- peratures at 44 m height, as the temperatures at the top are unreliable due to inﬂuence from the sun and the warm wa- ter from the reservoir. Below the canopy the dry cable tem- perature correlates perfectly with the reference sensor tem- perature (Fig. 5e). In and above the canopy incoming solar radiation warms up the ﬁbre optic cable (Fig. 5a, c), which causes an error at 34 m, where no screen was installed. This error is a deviation of up to 3 K from the reference sensor temperature (for 1 min temperature averages). The compar- ison at 34 m also has an offset – this is a constant error of about 1 K, due to the reference temperature sensor drift and inter-calibration problems. The addition of screens above the canopy largely reduces the error from solar radiation to un- der 1 K, leading to a very good agreement between the two sensor types (Fig. 5a). Flag 1 : r2 ea,z > 0.20, (20) Flag 2 : β < −1.1 or β > −0.9. (21) (20) (21) (20) (21) If ﬂag 1 is true, the outcome of the Bowen ratio calcula- tion is considered reliable. The other data points are removed from further analysis. If ﬂag 2 is also true, then the Bowen ratio can be used for calculating the atmospheric heat ﬂuxes. If ﬂag 1 is true, the outcome of the Bowen ratio calcula- tion is considered reliable. The other data points are removed from further analysis. If ﬂag 2 is also true, then the Bowen ratio can be used for calculating the atmospheric heat ﬂuxes. After processing the EC data using LI-COR’s EddyPro® software (LI-COR Inc., 2016), several quality ﬂags are avail- able. The quality ﬂag system used is from Mauder and Foken (2006), ranging from 0 (best) to 2 (worst). The EC ﬂuxes with a quality ﬂag of 0 or 1 are used in this research. After processing the EC data using LI-COR’s EddyPro® software (LI-COR Inc., 2016), several quality ﬂags are avail- able. The quality ﬂag system used is from Mauder and Foken (2006), ranging from 0 (best) to 2 (worst). 4 Results and discussion ∂z 4.2 Temperature validation The EC ﬂuxes with a quality ﬂag of 0 or 1 are used in this research. Below the canopy the wet cable temperature is in good agreement with the reference wet-bulb temperature (Fig. 5f), even though wind speeds were often low. This shows that the wet cable gives a good estimate of the wet-bulb tempera- ture. At 34 m, where no screens were placed, the error in the wet-bulb temperature is larger than the error in the air tem- perature. Deviations of up to 4 K occur in the measurement period. The shielded top part of the wet cable performs much better (Fig. 5b), and errors are small (under 1 K). To summarize, the method of this paper differs in a few points from Euser et al. (2014). The ﬁt of the Bowen ratio temperature and vapour pressure proﬁles is done separately, to get the correct ratio, as ∂T ∂z / ∂ea ∂z ̸= ∂T ∂ea . More energy bal- ance storage terms are taken into account, namely the latent and speciﬁc heat storage in the air column. The potential tem- perature is used instead of the air temperature, to correct for the lapse rate. The local air pressure is taken into account in the calculations, as it has an inﬂuence on the psychrometric constant, speciﬁc heat capacity and speciﬁc humidity. Lastly, a system for simple quality ﬂags is introduced to allow for simple objective quality control. 4.1 Meteorological conditions The temperature and speciﬁc humidity are integrated over the height from 0 to 41 m, up to the height of the Bowen ratio measurements. As quality control scheme for the DTS-measured Bowen ratio, two ﬂags are used. The ﬁrst ﬂag tests the correlation co- efﬁcient of the actual vapour pressure over height, for which we chose a lower limit of 0.20 (Eq. 20). We do not consider rTa,z of the air temperature gradient as it is always higher than rea,z (as the uncertainty in ea is higher due to the prop- agation of errors in Ta and Tw). The second ﬂag is for the case where the Bowen ratio approaches −1, which causes the uncertainty in the BREB ﬂuxes to be very high, as the de- nominator of Eqs. (12) and (13) approaches 0 (Payero et al., 2003): (19) (19) 4.1 Meteorological conditions where 1Ta, ﬁt is the difference in air temperature (K) of the ﬁtted temperature curve, between the top and bottom of the height range used for the Bowen ratio. 1ea, ﬁt is the differ- ence in vapour pressure (kPa) of the ﬁtted vapour pressure curve between those heights. 1z is the difference in height (m). The coefﬁcients of determination of the regressions of the temperature and vapour pressure, rTa,z and rea,z, can be used for determining the goodness of ﬁt. A high (positive or negative) regression means that the logarithmic slope (of the 15 min average) is very well deﬁned. d d where 1Ta, ﬁt is the difference in air temperature (K) of the ﬁtted temperature curve, between the top and bottom of the height range used for the Bowen ratio. 1ea, ﬁt is the differ- ence in vapour pressure (kPa) of the ﬁtted vapour pressure curve between those heights. 1z is the difference in height (m). The coefﬁcients of determination of the regressions of the temperature and vapour pressure, rTa,z and rea,z, can be used for determining the goodness of ﬁt. A high (positive or negative) regression means that the logarithmic slope (of the 15 min average) is very well deﬁned. dQ dQ For the comparison of the DTS temperature with the ref- erence temperature data (Sect. 4.2), the days 10–22 Au- gust 2016 are used. For a good comparison between DTS and EC, both devices should work properly. Due to several technical problems with data collection, only 11 days within the measurement cam- paign have both EC and DTS data available, namely 10, 12– 14, 19–22, and 28–30 August 2016. On the other days data are missing in either the EC or the DTS. The meteorological conditions of these days are shown in Fig. 4. All days were partially clouded, or completely clouded. The wind direction was mainly west and northeast. Above the canopy the wind speed varied between 2 and 6 m s−1, while under the canopy the wind speed was often too low to be measured with the cup anemometer (under 0.4 m s−1). To calculate the air column storage terms dQH dt and dQE dt (Eqs. 15 and 16), the DTS-measured temperature and vapour pressure are used, except for the centre of the canopy where DTS data are not accurate due to the sunlight and lack of screens in the canopy. 3.2 Data processing A linear ﬁt was also looked at, but it resulted in a minimal difference in the result- ing ﬁt. From the ﬁts the temperature difference over height is then calculated: ∂2 ∂z ≈∂Ta ∂z + 0(z) ≈1Ta,ﬁt 1z + 0(z) ≈Ta, ﬁt(z = 44) −Ta, ﬁt(z = 38.5) 44 −38.5 + 0(¯z = 41.25), (18) (18) Hydrol. Earth Syst. Sci., 22, 819–830, 2018 www.hydrol-earth-syst-sci.net/22/819/2018/ 824 4.4 Energy balance closure around the 1 : 1 line, and a good correlation (r2 = 0.59). The EC Bowen ratio was only calculated for ﬂuxes with an ab- solute value larger than 10 W m−2, as the uncertainty of the EC Bowen ratio is very high when the ﬂuxes are small. Even the negative (night-time) values seemed to be accurate, since they passed the quality control ﬂags. However, both EC and BR-DTS have problems measuring the night-time Bowen ra- tio. For EC this is due to the lower friction velocity at night (Wilson, 2002), while for the BR-DTS method the gradients are very small due to the small ﬂuxes. A known problem in measuring ﬂuxes is that the energy bal- ance often does not close well. This is caused by differences in fetch between the used devices, device inaccuracies, and possibly problems with the EC method (Wilson, 2002). Part of the difference between the BR-DTS method and the EC method may be explained by this energy balance closure problem. EC measurements have a fetch which does not in- clude the area close to the ﬂux tower. The available energy in the BR-DTS method depends on measurements of net radia- tion, ground heat ﬂux and heat storage change ( dQ dt ) close to the tower. Heterogeneity in the fetch may cause differences between the two methods. In addition, the biomass heat stor- age change ( dQB dt ) was not measured for the BR-DTS method, and was assumed to be 0 W m−2. The photosynthesis energy ﬂux (GP) was also assumed to be 0 W m−2. One drawback of the DTS-based Bowen ratio is the as- sumption that the eddy diffusivity of heat and water vapour are the same. In reality these eddy diffusivities can be dissim- ilar (Irmak et al., 2014). This can cause an error (both a bias and extra noise) in the Bowen ratio as measured by the tem- perature and vapour pressure gradients compared to the EC Bowen ratio. Another source of differences between βDTS and βEC is that the two are measured at different heights. To investigate the energy balance closure for the two meth- ods, we summed up the available ﬂuxes in the following equations, where dQ dt is the storage term from Eq. (14): During the measurement period the 80 % fetch of the EC system was between 200 and 300 m. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements 825 0 6 Wind speed (m s−1 ) z = 46 m 08/10 08/12 08/13 08/14 08/19 08/20 08/21 08/22 08/28 08/29 08/30 0 1 Wind speed (m s−1 ) z = 4 m N E S W N Wind direction z = 46 m 6 h 18 h 0 200 400 600 Energy flux (W m−2 ) 6 h 18 h 6 h 18 h 6 h 18 h 6 h 18 h 6 h 18 h 6 h 18 h 6 h 18 h 6 h 18 h 6 h 18 h 6 h 18 h Figure 4. Meteorological conditions during the days that both DTS and EC data were available. From top to bottom: wind speed at the top of the tower, wind speed at the bottom of the tower, wind direction at the top of the tower, and the measured energy ﬂuxes (green: net radiation; red: soil heat ﬂux; black: energy storage change dQ dt ). Figure 4. Meteorological conditions during the days that both DTS and EC data were available. From top to bottom: wind speed at the top of the tower, wind speed at the bottom of the tower, wind direction at the top of the tower, and the measured energy ﬂuxes (green: net radiation; red: soil heat ﬂux; black: energy storage change dQ dt ). 4.3 Bowen ratio veriﬁcation The Bowen ratio resulting from the BR-DTS method (βDTS) is compared to the EC Bowen ratio (βEC), at a 15 min averag- ing interval. In Fig. 6 the correlation between the EC Bowen ratio estimate and the BR-DTS is shown. It shows a grouping Hydrol. Earth Syst. Sci., 22, 819–830, 2018 www.hydrol-earth-syst-sci.net/22/819/2018/ Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements www.hydrol-earth-syst-sci.net/22/819/2018/ B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements 826 gure 5. Comparison between the 6 mm DTS cable and reference temperatures. Grey line shows 1:1 correlation. Da st 2016. (a) Dry cable at 44 m and reference air temperature at 46 m; the cable is shielded by the screen. (b) Wet ference wet-bulb temperature at 46 m; the cable is shielded by the screen. (c) Dry cable and reference air temperature a posed to direct sunlight. (d) Wet cable and reference wet-bulb temperature at 34 m; the ﬁbre optic cable is exposed to dire ble and reference air temperature at 16 m, under the canopy so less direct sunlight hits the ﬁbre optic cable. (f) Wet c et-bulb temperature at 16 m, under the canopy so less direct sunlight hits the ﬁbre optic cable. Shown are the linear corr e coefﬁcient of determination (r2), the slope (s) and the intercept (i). Figure 5. Comparison between the 6 mm DTS cable and reference temperatures. Grey line shows 1:1 correlation. Data from 10–23 Au- gust 2016. (a) Dry cable at 44 m and reference air temperature at 46 m; the cable is shielded by the screen. (b) Wet cable at 44 m and reference wet-bulb temperature at 46 m; the cable is shielded by the screen. (c) Dry cable and reference air temperature at 34 m; the cable is exposed to direct sunlight. (d) Wet cable and reference wet-bulb temperature at 34 m; the ﬁbre optic cable is exposed to direct sunlight. (e) Dry cable and reference air temperature at 16 m, under the canopy so less direct sunlight hits the ﬁbre optic cable. (f) Wet cable and reference wet-bulb temperature at 16 m, under the canopy so less direct sunlight hits the ﬁbre optic cable. Shown are the linear correlation coefﬁcients: the coefﬁcient of determination (r2), the slope (s) and the intercept (i). Figure 5. Comparison between the 6 mm DTS cable and reference temperatures. Grey line shows 1:1 correlation. Data from 10–23 Au- gust 2016. (a) Dry cable at 44 m and reference air temperature at 46 m; the cable is shielded by the screen. (b) Wet cable at 44 m and reference wet-bulb temperature at 46 m; the cable is shielded by the screen. (c) Dry cable and reference air temperature at 34 m; the cable is exposed to direct sunlight. 4.4 Energy balance closure By applying the ﬁnd- ings of Stannard (1997), the Bowen ratio 80% equilibrium ratio would be reached at a fetch-to-height ratio of 20 to 40. This corresponds to a distance of 350 to 700 m. The fetch of the Bowen ratio will therefore not be equal to the EC fetch, which could cause some differences in measured ﬂuxes. BDTS = RN −GS −dQ dt , (22) BEC = HEC + ρλEEC, (23) (22) (22) (23) (23) where BDTS is the energy available for heat ﬂuxes in the BR-DTS method (W m−2) and BEC is the sum of the EC- measured heat ﬂuxes (W m−2). where BDTS is the energy available for heat ﬂuxes in the BR-DTS method (W m−2) and BEC is the sum of the EC- measured heat ﬂuxes (W m−2). www.hydrol-earth-syst-sci.net/22/819/2018/ Hydrol. Earth Syst. Sci., 22, 819–830, 2018 B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements (d) Wet cable and reference wet-bulb temperature at 34 m; the ﬁbre optic cable is exposed to direct sunlight. (e) Dry cable and reference air temperature at 16 m, under the canopy so less direct sunlight hits the ﬁbre optic cable. (f) Wet cable and reference wet-bulb temperature at 16 m, under the canopy so less direct sunlight hits the ﬁbre optic cable. Shown are the linear correlation coefﬁcients: the coefﬁcient of determination (r2), the slope (s) and the intercept (i). www.hydrol-earth-syst-sci.net/22/819/2018/ Hydrol. Earth Syst. Sci., 22, 819–830, 2018 peroort et al.: Using DTS for Bowen ratio evaporation measurements B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements 827 B. Schilperoort et al.: Using DTS for Bowen ratio evapora 5 4 3 2 1 0 1 2 3 βEC 5 4 3 2 1 0 1 2 3 βDTS 1:1 line Day Night Figure 6. Correlation between the DTS-measured (βDTS) and EC- measured (βEC) Bowen ratios. Daytime data are between 07:00 and 18:00. Data from 10, 12–14, 19–22, and 28–30 August. R2 = 0.59. RMSE = 0.81. n = 319 data points. 100 0 100 200 300 400 500 ρλE DTS + ρλE EC 2 (W m−2 ) 400 300 200 100 0 100 200 300 400 ρλE DTS−ρλE EC (W m−2 ) µ (18.7) Figure 8. Tukey mean-difference plot comparing ρλEEC and ρλEDTS. With µ = 18.7 W m−2, RMSE = 90 W m−2 (15 min av- erages). Data from 10, 12–14, 19–22, and 28–30 August 2016. 100 0 100 200 300 400 500 ρλE DTS + ρλE EC 2 (W m−2 ) 400 300 200 100 0 100 200 300 400 ρλE DTS−ρλE EC (W m−2 ) µ (18.7) Figure 8. Tukey mean-difference plot comparing ρλEEC and ρλEDTS. With µ = 18.7 W m−2, RMSE = 90 W m−2 (15 min av- erages). Data from 10, 12–14, 19–22, and 28–30 August 2016. 5 4 3 2 1 0 1 2 3 βEC 5 4 3 2 1 0 1 2 3 βDTS 1:1 line Day Night Figure 8. Tukey mean-difference plot comparing ρλEEC and ρλEDTS. With µ = 18.7 W m−2, RMSE = 90 W m−2 (15 min av- erages). Data from 10, 12–14, 19–22, and 28–30 August 2016. βEC Figure 6. Correlation between the DTS-measured (βDTS) and EC- measured (βEC) Bowen ratios. Daytime data are between 07:00 and 18:00. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements Data from 10, 12–14, 19–22, and 28–30 August. R2 = 0.59. RMSE = 0.81. n = 319 data points. 100 0 100 200 300 400 H DTS + H EC 2 (W m−2 ) 400 300 200 100 0 100 200 300 400 H DTS−H EC (W m−2 ) µ (-10.6) Figure 9. Tukey mean-difference plot comparing HEC and HDTS. With µ = −10.6 W m−2, RMSE = 82 W m−2. (15 min averages). Data from 10, 12–14, 19–22, and 28–30 August 2016. 100 0 100 200 300 400 H DTS + H EC 2 (W m−2 ) 400 300 200 100 0 100 200 300 400 H DTS−H EC (W m−2 ) µ (-10.6) 100 0 100 200 300 400 500 600 700 800 B DTS + B EC 2 (W m−2 ) 300 200 100 0 100 200 300 B DTS−B EC (W m−2 ) µ (-3.4) Figure 7. Tukey mean-difference plot comparing BDTS and BEC. With µ = −3.4 W m−2, RMSE = 76 W m−2, n = 741 data points (15 min averages). Data from 10, 12–14, 19–22, and 28–30 August 2016. 100 0 100 200 300 400 500 600 700 800 B DTS + B EC 2 (W m−2 ) 300 200 100 0 100 200 300 B DTS−B EC (W m−2 ) µ (-3.4) Figure 9. Tukey mean-difference plot comparing HEC and HDTS. With µ = −10.6 W m−2, RMSE = 82 W m−2. (15 min averages). Data from 10, 12–14, 19–22, and 28–30 August 2016. Figure 7. Tukey mean-difference plot comparing BDTS and BEC. With µ = −3.4 W m−2, RMSE = 76 W m−2, n = 741 data points (15 min averages). Data from 10, 12–14, 19–22, and 28–30 August 2016. which causes an underestimation of the available energy in BDTS during the night, and an overestimation during the day. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements 828 100 0 650 ρλE (W m−2 ) 08/10 08/12 08/13 08/14 08/19 08/20 08/21 08/22 08/28 08/29 08/30 EC DTS 6 h 12 h 18 h 250 0 600 H (W m−2 ) 6 h 12 h 18 h 6 h 12 h 18 h 6 h 12 h 18 h 6 h 12 h 18 h 6 h 12 h 18 h 6 h 12 h 18 h 6 h 12 h 18 h 6 h 12 h 18 h 6 h 12 h 18 h 6 h12 h18 h Figure 10. Plot comparing the BR-DTS and EC measured sensible (H) and latent (ρλE) heat ﬂuxes over time (15 min averages). Data from 10, 12–14, 19–22, and 28–30 August 2016. Figure 10. Plot comparing the BR-DTS and EC measured sensible (H) and latent (ρλE) heat ﬂuxes over time (15 min averages). Data from 10, 12–14, 19–22, and 28–30 August 2016. ror, up to 3 K. This error can be largely mitigated by placing screens to block the sunlight, reducing the error to less than 1 K. Hence screens are effective and should also be placed in the canopy. ﬂuxes, resulting from the negative bias in the energy bal- ance comparison (Fig. 7). At positive ﬂuxes there seems to be a positive bias (HDTS > HEC). The mean difference is small, being a 10.6 W m−2 underestimation by the BR-DTS method. The Bowen ratio measured by DTS correlates well with EC estimates (r2 = 0.59). A simple quality control method, using the goodness of ﬁt of the vapour pressure gradient, also works well, and ﬁlters out most outliers and errors. The small gradients above the forest canopy are hard to measure ac- curately, which increases the uncertainty during days where ﬂuxes (and thus gradients) are small. The Bowen ratio as- sumption that the eddy diffusivities of heat and vapour are equal was not studied, but can be a source of differences be- tween the BR-DTS and EC methods. The difference in fetch for the two methods can also be a cause for differences. Figure 10 shows the time series of the BR-DTS and EC measured heat ﬂuxes. The daytime ﬂux estimates correspond well, and follow the same trends. The night-time BR-DTS estimates of the sensible heat ﬂux are more negative than the EC estimates, one possible reason being the energy balance differences discussed before. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements On many days, during the early morning and start of the evening, the BR-DTS has missing values, which is mainly due to the inversion of the gradient, as the temperature gradients changes from negative (stable conditions) to positive (unstable conditions) and vice versa. This inversion causes uncertainty, which is ﬁltered out by the quality control ﬂags. The energy balance closure between the BR-DTS method and EC is in good agreement, with the mean difference be- ing a 3.4 W m−2 underestimation by the BR-DTS method, and an uncertainty of RMSE = 76 W m−2. However, the BR- DTS method estimates a more negative amount of available energy during night-time, and a more positive amount dur- ing daytime compared to EC. One cause could be the lack of biomass heat storage change measurements, which is of the order of 45 W m−2. Another source for the difference is that the energy balance components of the BR-DTS method are generally point measurements, while EC and the Bowen ratio both have a large fetch. As a result, heterogeneity can cause large differences in the available energy for latent and sensible heat ﬂuxes. 4.5 Energy ﬂuxes To compare the two measurement methods, a Tukey mean- difference (or Bland–Altman) plot was made (Fig. 7) (Alt- man and Bland, 1983). The mean of the two measurement methods is plotted against the difference between them. The mean difference (µ) between BDTS and BEC is a 3.4 W m−2 underestimation by the BR-DTS method. At low ﬂuxes (be- low 100 W m−2), the BR-DTS method measures less energy available for ﬂuxes compared to EC. At high ﬂuxes (over 400 W m−2) the opposite is visible. One possible reason for this is that the biomass heat ﬂux ( dQB dt ) was not measured, Figures 8 and 9 show the mean difference plots comparing the latent and sensible heat ﬂuxes of the EC method to the BR-DTS method. The BR-DTS ﬂuxes are calculated above the canopy, using only temperature data from the shielded ca- bles. The Tukey mean-difference plot for the latent heat ﬂux shows no large bias when comparing the BR-DTS method to EC, with the mean difference being a 18.7 W m−2 overesti- mation by the BR-DTS method (Fig. 8). The Tukey mean-difference plot comparing the sensible heat ﬂux (Fig. 9) shows a strong negative bias for negative www.hydrol-earth-syst-sci.net/22/819/2018/ Hydrol. Earth Syst. Sci., 22, 819–830, 2018 Competing interests. The authors declare that they have no conﬂict of interest. Competing interests. The authors declare that they have no conﬂict of interest. Euser, T., Luxemburg, W. M. J., Everson, C. S., Mengistu, M. G., Clulow, A. D., and Bastiaanssen, W. G. M.: A new method to measure Bowen ratios using high-resolution vertical dry and wet bulb temperature proﬁles, Hydrol. Earth Syst. Sci., 18, 2021– 2032, https://doi.org/10.5194/hess-18-2021-2014, 2014. Acknowledgements. This research was funded by NWO Earth and Life Sciences (ALW), veni-project 863.15.022, the Netherlands. We would like to thank Murat Ucer and Christiaan van der Tol (University of Twente) for providing us with access to their measurement location and reference data, and the students Tara van Iersel and Tom Oostdijk for assisting with setting up the DTS measurements. Fuchs, M. and Tanner, C.: Error analysis of bowen ratios mea- sured by differential psychrometry, Agr. Meteorol., 7, 329–334, https://doi.org/10.1016/0002-1571(70)90027-0, 1970. Harrison, R. G. and Wood, C. R.: Ventilation effects on humidity measurements in thermometer screens, Q. J. Roy. Meteor. Soc., 138, 1114–1120, https://doi.org/10.1002/qj.985, 2012. Edited by: Pierre Gentine Edited by: Pierre Gentine Reviewed by: two anonymous referees Reviewed by: two anonymous referees Irmak, S., Kilic, A., and Chatterjee, S.: On the Equality Assump- tion of Latent and Sensible Heat Energy Transfer Coefﬁcients of the Bowen Ratio Theory for Evapotranspiration Estimations: Another Look at the Potential Causes of Inequalities, Climate, 2, 181–205, https://doi.org/10.3390/cli2030181, 2014. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements Bao, X. and Chen, L.: Recent progress in distributed ﬁber optic sensors, Sensors (Basel), 12, 8601–8639, https://doi.org/10.3390/s120708601, 2012. ilar, with an uncertainty of RMSE = 82 W m−2, and the BR-DTS method underestimating the sensible heat ﬂux by 10.6 W m−2. However, the underestimation mainly takes place during night-time, which can be caused by differences in available energy. Barr, A. G., King, K. M., Gillespie, T. J., Den Hartog, G., and Neumann, H. H.: A comparison of bowen ratio and eddy correlation sensible and latent heat ﬂux measurements above deciduous forest, Bound.-Lay. Meteorol., 71, 21–41, https://doi.org/10.1007/BF00709218, 1994. While the average proﬁles can be useful and valuable, ex- tra information could be gained by opting for a smaller di- ameter ﬁbre optic cable, and measuring at a high frequency (1 Hz). This could give new insights into surface interactions and could show convective cells transporting heat upwards. Bense, V., Read, T., and Verhoef, A.: Using distributed temperature sensing to monitor ﬁeld scale dynamics of ground surface tem- perature and related substrate heat ﬂux, Agr. Forest Meteorol., 220, 207–215, https://doi.org/10.1016/j.agrformet.2016.01.138, 2016. A way to improve the performance of the BR-DTS method is to ﬁnd an independent estimate for the sensible heat ﬂux (H), to avoid the uncertainties in the energy balance com- ponents (RN, dQ dt ). Through the universal functions of the Monin–Obukhov similarity theory estimates of the sensible heat ﬂux can be made. This could be done either by mea- suring the wind speed over height (Stricker and Brutsaert, 1978) using DTS (Sayde et al., 2015) or by applying the ﬂux-variance method (Katul et al., 1995). The Bowen ratio can then be used to calculate the latent heat ﬂux. Blume, T., Krause, S., Meinikmann, K., and Lewandowski, J.: Upscaling lacustrine groundwater discharge rates by ﬁber-optic distributed temperature sensing, Water Resour. Res., 49, 7929– 7944, https://doi.org/10.1002/2012WR013215, 2013. Bosveld, F. and Bouten, W.: Evaluation of transpiration models with observations over a Douglas-ﬁr forest, Agr. Forest Meteorol., 108, 247–264, https://doi.org/10.1016/S0168-1923(01)00251-9, 2001. Bowen, I. S.: The ratio of heat losses by conduction and by evaporation from any water surface, Phys. Rev., 27, 779–787, https://doi.org/10.1103/PhysRev.27.779, 1926. Data availability. The data used in this study are available online from the 4TU data repository (https://doi.org/10.4121/uuid:5c81f10a-1249-4b85-8dec- 2b029dd88b99; Schilperoort et al., 2017). Data availability. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements The data used in this study are available online from the 4TU data repository (https://doi.org/10.4121/uuid:5c81f10a-1249-4b85-8dec- 2b029dd88b99; Schilperoort et al., 2017). de Jong, S. A. P., Slingerland, J. D., and van de Giesen, N. C.: Fiber optic distributed temperature sensing for the deter- mination of air temperature, Atmos. Meas. Tech., 8, 335–339, https://doi.org/10.5194/amt-8-335-2015, 2015. Erisman, J., Draaijers, G., Steingröver, E., Van Dijk, H., Boxman, A., and De Vries, W.: Assessment of the exposure and loads of acidifying and eutrophying pollutants and ozone, as well as their harmful inﬂuence on the vitality of the trees and the Speulder forest ecosystem as a whole, Water Air Soil Poll., 105, 539–571, 1998. References Allen, R. G., Pereira, L. S., Raes, D., and Smith, M.: FAO Irrig. Drain. Paper No. 56, Irrig. Drain., 300, 300 pp., http://www. kimberly.uidaho.edu/water/fao56/fao56.pdf, 1998. Jansen, J., Stive, P. M., van de Giesen, N., Tyler, S., Steele- Dunne, S. C., and Williamson, L.: Estimating soil heat ﬂux us- ing Distributed Temperature Sensing, GRACE, Remote Sensing and Ground-based Methods in Multi-Scale Hydrology, 140–144, 2011. Altman, D. G. and Bland, J. M.: Measurement in Medicine: The Analysis of Method Comparison Studies, J. R. Stat. Soc. D-Sta., 32, 307–317, https://doi.org/10.2307/2987937, 1983. Katul, G., Goltz, S. M., Hsieh, C. I., Cheng, Y., Mowry, F., and Sigmon, J.: Estimation of surface heat and momen- tum ﬂuxes using the ﬂux-variance method above uniform Angus, D. E. and Watts, P. J.: Evapotranspiration – How good is the Bowen ratio method?, Agr. Water Manage., 8, 133–150, https://doi.org/10.1016/0378-3774(84)90050-7, 1984. Hydrol. Earth Syst. Sci., 22, 819–830, 2018 5 Conclusions and recommendations This technical note investigates the use of the BR-DTS method above a forest canopy, and introduces a number of improvements on the method as presented by Euser et al. (2014). The performance is investigated by comparing the measured DTS cable temperatures to reference sensors, look- ing at energy balance closure, and comparing the measured Bowen ratio, sensible heat ﬂux and latent heat ﬂux to EC measurements. When comparing the ﬁbre optic cable temperature to ref- erence sensors, it can be seen that the wet-bulb and air tem- peratures can be well represented. Under the canopy, where the cables are shaded from direct sunlight, the DTS cable and reference sensors are in near-perfect agreement. How- ever, above the canopy direct sunlight may cause a large er- When comparing the latent heat ﬂux of the two meth- ods, they are in agreement, although the uncertainty is high (RMSE = 90 W m−2). The BR-DTS method slightly over- estimates the latent heat ﬂux, with a mean difference of 18.7 W m−2. The results for the sensible heat ﬂux are sim- www.hydrol-earth-syst-sci.net/22/819/2018/ Hydrol. Earth Syst. Sci., 22, 819–830, 2018 829 B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements Tiktak, A. and Bouten, W.: Soil water dynamics and long-term wa- ter balances of a Douglas ﬁr stand in the Netherlands, J. Hydrol., 156, 265–283, 1994. Raj, R., Hamm, N. A., van der Tol, C., and Stein, A.: Variance-based sensitivity analysis of BIOME-BGC for gross and net primary production, Ecol. Model., 292, 26–36, 2014. van der Tol, C.: Validation of remote sensing of bare soil ground heat ﬂux, Remote Sens. Environ., 121, 275–286, https://doi.org/10.1016/j.rse.2012.02.009, 2012. Sayde, C., Thomas, C. K., Wagner, J., and Selker, J.: High-resolution wind speed measurements using actively heated ﬁber optics, Geophys. Res. Lett., 42, 10064–10073, https://doi.org/10.1002/2015GL066729, 2015. Van Wijk, M., Dekker, S., Bouten, W., Bosveld, F., Kohsiek, W., Kramer, K., and Mohren, G.: Modeling daily gas exchange of a Douglas-ﬁr forest: comparison of three stomatal conductance models with and without a soil water stress function, Tree Phys- iol., 20, 115–122, 2000. Schilperoort, B., Cisneros Vaca, C. R., and Jiménez Ro- dríguez, C. D.: Speulderbos ﬂux, storage, and temperature pro- ﬁle measurements, https://doi.org/10.4121/uuid:5c81f10a-1249- 4b85-8dec-2b029dd88b99, 2017. Weligepolage, K., Gieske, A., and Su, Z.: Surface roughness anal- ysis of a conifer forest canopy with airborne and terrestrial laser scanning techniques, Int. J. Appl. Earth Obs., 14, 192–203, https://doi.org/10.1016/j.jag.2011.08.014, 2012. Selker, J. S., Thévenaz, L., Huwald, H., Mallet, A., Luxemburg, W., Van De Giesen, N., Stejskal, M., Zeman, J., Westhoff, M., and Parlange, M. B.: Distributed ﬁber-optic temperature sensing for hydrologic systems, Water Resour. Res., 42, 1–8, https://doi.org/10.1029/2006WR005326, 2006. ® Weligepolage, K., Gieske, A., and Su, Z.: Effect of spatial resolution on estimating surface albedo: A case study in Speulderbos forest in The Netherlands, Int. J. Appl. Earth Obs., 23, 18–28, 2013. Silixa Ltd: ULTIMA® DTS, available at: http://silixa.com/ technology/ultima-dts/, last access: 26 June 2017. Westhoff, M. C., Savenije, H. H. G., Luxemburg, W. M. J., Stelling, G. S., van de Giesen, N. C., Selker, J. S., Pﬁster, L., and Uh- lenbrook, S.: A distributed stream temperature model using high resolution temperature observations, Hydrol. Earth Syst. Sci., 11, 1469–1480, https://doi.org/10.5194/hess-11-1469-2007, 2007. Sluijter, R.: De Bosatlas van het klimaat, Noordhoff Uitgevers, Groningen, 2011. Stannard, D. I.: A Theoretically Based Determination of Bowen- ratio Fetch Requirements, Bound.-Lay. Meteorol., 83, 375–406, https://doi.org/10.1023/A:1000286829849, 1997. Wilkes, P., Lau, A., Disney, M., Calders, K., Burt, A., de Tanago, J. G., Bartholomeus, H., Brede, B., and Herold, M.: Data acquisition considerations for Terrestrial Laser Scan- ning of forest plots, Remote Sens. Environ., 196, 140–153, https://doi.org/10.1016/j.rse.2017.04.030, 2017. B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements 830 and non-uniform terrain, Bound.-Lay. Meteorol., 74, 237–260, https://doi.org/10.1007/BF00712120, 1995. and non-uniform terrain, Bound.-Lay. Meteorol., 74, 237–260, https://doi.org/10.1007/BF00712120, 1995. Stull, R.: Practical Meteorology – An Algebra-based Survey of Atmospheric Science, Department of Earth, Ocean & Atmo- spheric Sciences, University of British Columbia, Vancouver, BC, https://doi.org/10.14288/1.0300441, 2015. Koutsoyiannis, D.: Clausius-Clapeyron equation and satura- tion vapour pressure: simple theory reconciled with prac- tice, Eur. J. Phys., 33, 295–305, https://doi.org/10.1088/0143- 0807/33/2/295, 2012. p g Su, Z., Timmermans, W. J., van der Tol, C., Dost, R., Bianchi, R., Gómez, J. A., House, A., Hajnsek, I., Menenti, M., Magliulo, V., Esposito, M., Haarbrink, R., Bosveld, F., Rothe, R., Baltink, H. K., Vekerdy, Z., Sobrino, J. A., Timmermans, J., van Laake, P., Salama, S., van der Kwast, H., Claassen, E., Stolk, A., Jia, L., Moors, E., Hartogensis, O., and Gillespie, A.: EAGLE 2006 – Multi-purpose, multi-angle and multi-sensor in-situ and airborne campaigns over grassland and forest, Hydrol. Earth Syst. Sci., 13, 833–845, https://doi.org/10.5194/hess-13-833-2009, 2009. LI-COR Inc.: EddyPro® Version 6.2 [Computer Software], 2016. Mauder, M. and Foken, T.: Impact of post-ﬁeld data process- ing on eddy covariance ﬂux estimates and energy balance clo- sure, Meteorol. Z., 15, 597–609, https://doi.org/10.1127/0941- 2948/2006/0167, 2006. Michiles, A. A. and Gielow, R.: Above-ground thermal energy stor- age rates, trunk heat ﬂuxes and surface energy balance in a cen- tral Amazonian rainforest, Agr. Forest Meteorol., 148, 917–930, https://doi.org/10.1016/j.agrformet.2008.01.001, 2008. Thomas, C. K., Kennedy, A. M., Selker, J. S., Moretti, A., Schroth, M. H., Smoot, A. R., Tuﬁllaro, N. B., and Zeeman, M. J.: High-Resolution Fibre-Optic Temperature Sensing: A New Tool to Study the Two-Dimensional Structure of Atmo- spheric Surface-Layer Flow, Bound.-Lay. Meteorol., 142, 177– 192, https://doi.org/10.1007/s10546-011-9672-7, 2012. Oliphant, A. J., Grimmond, C. S. B., Zutter, H. N., Schmid, H. P., Su, H. B., Scott, S. L., Offerle, B., Randolph, J. C., and Ehman, J.: Heat storage and energy balance ﬂuxes for a tem- perate deciduous forest, Agr. Forest Meteorol., 126, 185–201, https://doi.org/10.1016/j.agrformet.2004.07.003, 2004. Tietema, A., Mol-Dijkstra, J. P., Kros, J., and De Vries, W.: Dy- namic nitrogen deposition thresholds during forest stand devel- opment in a Douglas ﬁr forest analysed with two nitrogen models SMART2 and MERLIN, Hydrol. Earth Syst. Sci., 6, 375–382, https://doi.org/10.5194/hess-6-375-2002, 2002. Pal Arya, S.: Introduction to Micrometeorology, Academic Press, San Diego, California, 1988. Payero, J. O., Neale, C. M. U., Wright, J. L., and Allen, R. G.: Guidelines for validating Bowen ratio data, T. ASAE, 46, 1051, https://doi.org/10.13031/2013.13967, 2003. www.hydrol-earth-syst-sci.net/22/819/2018/ www.hydrol-earth-syst-sci.net/22/819/2018/ B. Schilperoort et al.: Using DTS for Bowen ratio evaporation measurements Steele-Dunne, S. C., Rutten, M. M., Krzeminska, D. M., Haus- ner, M., Tyler, S. W., Selker, J., Bogaard, T. A., and Van De Giesen, N. C.: Feasibility of soil moisture estimation using pas- sive distributed temperature sensing, Water Resour. Res., 46, 1– 12, https://doi.org/10.1029/2009WR008272, 2010. Wilson, K.: Energy balance closure at FLUXNET sites, Agr. Forest Meteorol., 113, 223–243, https://doi.org/10.1016/S0168- 1923(02)00109-0, 2002. Stricker, H. and Brutsaert, W.: Actual evapotranspiration over a summer period in the “Hupsel catchment”, J. Hydrol., 39, 139– 157, https://doi.org/10.1016/0022-1694(78)90119-1, 1978. Hydrol. Earth Syst. Sci., 22, 819–830, 2018 www.hydrol-earth-syst-sci.net/22/819/2018/
https://openalex.org/W2906392928	https://figshare.com/articles/thesis/The_classification_of_gesture_interactions_and_the_study_of_their_ergonomic_effect_on_hands/14657838/1/files/28139841.pdf	English	null	The classification of gesture interactions and the study of their ergonomic effect on hands	null	2,021	cc-by	28,054	by I hereby declare that I am the sole author of this thesis. I hereby declare that I am the sole author of this thesis. I authorize Ryerson University to lend this thesis to other institutions or individuals for the purpose of scholarly research. I further authorize Ryerson University to reproduce this thesis by photocopying or by other means, in total or in part, at the request of other institutions or individuals for the purpose of scholarly research. I understand that my thesis may be made electronically available to the public. I understand that my thesis may be made electronically available to the public. ii ii Abstract In order to investigate gestural behavior during human-computer interactions, an investigation into the designs of current interaction methods is conducted. This information is then compared to current emerging databases to observe if the gesture designs follow guidelines discovered in the above investigation. The comparison will also observe common trends in the currently developed gesture databases such as similar gesture for specific commands. In order to investigate gestural behavior during interactions with computer interfaces, an experiment has been devised to observe and record gestures in use for gesture databases through the use of a hardware sensor device. It was discovered that factors such as opposing adjacent fingers and gestures that simulated object manipulation are factors in user comfort. The results of this study will create guidelines for creating new gestures for hand gesture interfaces. iii Acknowledgments iv iv v Table of Contents Declaration ii Abstract iii Acknowledgements iv List of Tables vii List of Figures viii List of Appendices ix 1 Introduction 1 1.1 Background 1 1.2 Purpose and Scope 2 1.3 Contributions 3 1.4 Overview of Thesis 4 2 Gesture Based Computer Interfaces 5 2.1 Gesture Recognition Methods 5 2.1.1 Image Recognition Techniques 6 2.1.2 Data Management 8 2.1.3 Specialty Hardware 9 2.2 Gesture Classification 11 2.3 Gesture and Language 13 2.4 Designs of Gestures 15 2.5 Summary 18 3 Gesture Design Investigative Methodology 19 3.1 Gesture Types 21 3.1.1 Static Gestures 23 3.1.2 Dynamic Gestures 24 3.2 Gesture Designs 25 3.2.1 Exertion 25 3.2.2 Position 30 Table of Contents Declaration Abstract Acknowledgements List of Tables List of Figures List of Appendices 1 Introduction 1.1 Background 1.2 Purpose and Scope 1.3 Contributions 1.4 Overview of Thesis 2 Gesture Based Computer Interfaces 2.1 Gesture Recognition Methods 2.1.1 Image Recognition Techniques 2.1.2 Data Management 2.1.3 Specialty Hardware 2.2 Gesture Classification 2.3 Gesture and Language 2.4 Designs of Gestures 2.5 Summary 3 Gesture Design Investigative Methodology 3.1 Gesture Types 3.1.1 Static Gestures 3.1.2 Dynamic Gestures 3.2 Gesture Designs 3.2.1 Exertion 3.2.2 Position Table of Contents Declaration ii Abstract iii Acknowledgements iv List of Tables vii List of Figures viii List of Appendices ix 1 Introduction 1 1.1 Background 1 1.2 Purpose and Scope 2 1.3 Contributions 3 1.4 Overview of Thesis 4 2 Gesture Based Computer Interfaces 5 2.1 Gesture Recognition Methods 5 2.1.1 Image Recognition Techniques 6 2.1.2 Data Management 8 2.1.3 Specialty Hardware 9 2.2 Gesture Classification 11 2.3 Gesture and Language 13 2.4 Designs of Gestures 15 2.5 Summary 18 3 Gesture Design Investigative Methodology 19 3.1 Gesture Types 21 3.1.1 Static Gestures 23 3.1.2 Dynamic Gestures 24 3.2 Gesture Designs 25 3.2.1 Exertion 25 3.2.2 Position 30 Table of Contents Declaration Abstract Acknowledgements List of Tables List of Figures List of Appendices 1 Introduction 1.1 Background 1.2 Purpose and Scope 1.3 Contributions 1.4 Overview of Thesis 2 Gesture Based Computer Interfaces 2.1 Gesture Recognition Methods 2.1.1 Image Recognition Techniques 2.1.2 Data Management 2.1.3 Specialty Hardware 2.2 Gesture Classification 2.3 Gesture and Language 2.4 Designs of Gestures 2.5 Summary 3 Gesture Design Investigative Methodology 3.1 Gesture Types 3.1.1 Static Gestures 3.1.2 Dynamic Gestures 3.2 Gesture Designs 3.2.1 Exertion 3.2.2 Position Acknowledgments I would like to give thanks to my supervisor Dr. Kristina Mai for supporting and guiding me during my studies in the Master of Applied Sciences in Electrical and Computer Engineering program, who provided me with valuable ideas and resources during the creation of this thesis. I would like to give my loving thanks to my mother and sister, without whom this work would not be possible, for their support, care, and love during my studies. Their belief in me was inspirational. Table of Contents 3.3 Summary 33 Summary 4 Experiments and Results 4.1 Gesture Categorization 4.2 Gesture Exertion on User 4.2.1 High Exertion 4.2.2 Low Exertion 4.3 Summary 5 Discussions and Conclusions 5.1 Design Guidelines 5.2 Possible Future Work 5.3 Conclusion References 4 Experiments and Results 98 References References vi vi List of Tables Table 3.1: Rest and Exertion point reference table (Activation Bits) Table 3.2: Rest and Exertion point reference table (Normalized) Table 3.3: Rest, flex and clenched reference Table 3.4: Range assigned to each value by degrees (X0) Table 4.1: Categorization of compiled database gestures Table 4.2: High Exertion Gestures (Activity Levels) Table 4.3: High Exertion Gestures Movement Range Table 4.4: Low Exertion Gestures (Activity Levels) Table 4.5: Low Exertion Gesture Movement Range Table 4.1: Compiled Database Gestures Table 4.2: American Sign Language A-Z Table 3.1: Rest and Exertion Points Reference Table Table 4.1: Myo Armband sensor Readings Static (Activity Levels) Table 4.2: Static Gesture Myo Armband Readings Normalized Table 4.3: Myo Armband sensor Readings Dynamic Table 4.4: Dynamic Gesture Myo Armband Readings Normalized Table 4.5: Myo Armband EMG Plots Table 3.2: Rest, flex and clenched reference table Table 4.6: Movement Range Table Static Table 4.7: Movement Range Table Dynamic Table 3.1: Rest and Exertion point reference table (Activation Bits) Table 3.2: Rest and Exertion point reference table (Normalized) Table 3.3: Rest, flex and clenched reference Table 3.4: Range assigned to each value by degrees (X0) Table 4.1: Categorization of compiled database gestures Table 4.2: High Exertion Gestures (Activity Levels) Table 4.3: High Exertion Gestures Movement Range Table 4.4: Low Exertion Gestures (Activity Levels) Table 4.5: Low Exertion Gesture Movement Range Table 3.1: Rest and Exertion Points Reference Table Table 4.1: Myo Armband sensor Readings Static (Activity Levels) Table 4.2: Static Gesture Myo Armband Readings Normalized Table 4.3: Myo Armband sensor Readings Dynamic Table 4.4: Dynamic Gesture Myo Armband Readings Normalized Table 4.5: Myo Armband EMG Plots vii Figure 3.1: Example of iconic gestures, American Sign Language letters ASL 21 Figure 3.2: Example of Metaphoric gesture, the spreading of the hand 22 Figure 3.3: Example of deictic gesture, the pointing gesture 22 Figure 3.4: An example of a static gesture 23 Figure 3.5: Example of dynamic gesture, wave out 24 Figure 3.6: Myo Armband pod numbering 26 Figure 3.7: Myo EMG reading of a closed fist 28 Figure 3.8: Sensor Data Rectifying, Rounding, Averaging Algorithm 30 Figure 3.9: Relaxed hand used in experiments (Reference 0, 900) 31 Figure 3.10: Closed fist used in experiments (Reference -10, 1800) 31 Figure 3.11: Fully extended open hand used in experiments (Reference 10, 00) 32 Figure 3.12: Position Measuring Block Diagram; (A) Static, (B) Dynamic 33 Figure 4.1: Myo Armband EMG reading for full exertion 48 Figure 4.2: Myo Armband EMG readings for: ASL B and ASL C 49 Figure 4.3: Myo Armband EMG readings for: Tilt, Fork Tilt, Palm and Moving Point 50 Figure 4.4: Finger Difference vs Exertion (Dynamic) 56 Figure 3.1: Example of iconic gestures, American Sign Language letters ASL 21 Figure 3.2: Example of Metaphoric gesture, the spreading of the hand 22 Figure 3.3: Example of deictic gesture, the pointing gesture 22 Figure 3.4: An example of a static gesture 23 Figure 3.5: Example of dynamic gesture, wave out 24 Figure 3.6: Myo Armband pod numbering 26 Figure 3.7: Myo EMG reading of a closed fist 28 Figure 3.8: Sensor Data Rectifying, Rounding, Averaging Algorithm 30 Figure 3.9: Relaxed hand used in experiments (Reference 0, 900) 31 Figure 3.10: Closed fist used in experiments (Reference -10, 1800) 31 Figure 3.11: Fully extended open hand used in experiments (Reference 10, 00) 32 Figure 3.12: Position Measuring Block Diagram; (A) Static, (B) Dynamic 33 Figure 4.1: Myo Armband EMG reading for full exertion 48 Figure 4.2: Myo Armband EMG readings for: ASL B and ASL C 49 Figure 4.3: Myo Armband EMG readings for: Tilt, Fork Tilt, Palm and Moving Point 50 Figure 4.4: Finger Difference vs Exertion (Dynamic) 56 viii List of Appendices 63 Compiled Gesture Database 79 Myo Sensor Data 95 Gesture Movement Rating ix ix 1. 1 Background With the advancements of new image processing algorithms and powerful processors, the ability to recognize shapes and interpret them in real time is a possibility. This has given rise to the growing field of gesture recognition for human-computer interactions. Due to the complexity of human-human communications, computers must rely on a rigid standard for communication that is both simple and robust. However this effectively limits the amount of information that can transfer to the computer as humans can naturally output more information through speech, body language and hand gestures. With the addition of gesturing in human-computer interaction, more information can be transferred and due to the spatial nature of gestures 3D dimensional concepts can be more intuitively conveyed by the user. However one obstacle in implementation of these systems is that they are still in their infancy with new methods being developed to eliminate or minimize current weaknesses. As the technology is still in its infancy, its rarity has given rise mostly to novelty in its applications. As a rare element that is being developed, little studies have been performed into how the interaction should be developed as the main focus of the technology currently is in the computer systems. 1 1 The use of gesture-recognition in human-computer interactions offers distinct and decisive advantages to current methods. The main advantage offered is a physical quantity can be more quickly and effectively conveyed by quantities such as shape, movement, and physical behaviors. Studies in human cognitive functions have shown that speech can be supplemented with gestures to convey a clearer meaning to the recipient which would in turn help in speech recognition in machines and computing systems. Gesture Recognition enjoys two recognition methods that are currently in development; those of hardware sensor devices and image processing techniques. Image processing techniques make use of simple color cameras that are widely distributed while hardware sensors make use of a suite of sensory devices to obtain gesture information. An example of such a device is the Myo Armband which makes use of Electromyography (EMG) sensors, accelerometers, and gyroscopes to get position and hand shape to facilitate in its gesture recognition. However while hardware sensors offer an advantage in accuracy, the widespread distribution of color cameras make image processing have a distinct advantage over hardware sensors. 1. 1 Background Currently the technology is developed per application as each application of gesture- recognition is developed to communicate a specific purpose. Much like text recognition and speech recognition, a fixed set of text or words is used to communicate in clear rigid understanding of the intents and directed actions of the user. In most cases gesture-recognition is typically designed for gaming application with some parties interested in design applications. This would limit the growth and potential of this technology. As one of the primary means that humans use to communicate with each other, limiting this technology to such specific applications will also limit our ability to communicate with computers more effectively. Some focus should be shifted into the development of how to clearly communicate with computers intuitively. 1.2 Objective and Scope The goal of this paper is to observe the current development of gestures used when creating gesture databases for recognition and their interpretations and to provide a framework for future endeavors to develop proper general purpose recognition databases. The study will include 2 current public sources and their study samples. This paper will categorize gestures based on similarities in shape and intents to create general cases for specific and basic commands that a computer should recognize to have meaningful communication. Additionally, the exertion on the user will be investigated to provide a framework for ergonomic designs and considerations. Though this investigation, the goal of this thesis is to improve the design process of gesture recognition systems as well as help achieve better public support through better design goals and improved alternative computing system inputs. The intentions are not to simply replace current existing methods, but to enhance, substitute and supplement current communication methods. The long-term goal of this paper is to provide a compare and contrast the current interaction methods to provide a base to which new design considerations can be drawn from. To accomplish this, this thesis will focus on gesture categorization in an effort to break down and separate gestures to facilitate better understanding of them individually and to help organize results from other work this thesis will conduct. Gestures will then be investigated though their ergonomic effects of exertion and positions. Through this study traits and features of gestures can be recognized as harmful or injuring of gesture practitioners of the emerging technology. By devising a method to view exertion by the user on gesture making, it is possible to demonstrate the effect of gestures on a user relative to their abilities and physical characteristics. 1.3 Contributions This work contributes in the following: Through investigation into gesture category with respect to usage, design, meaning, and interpretations, a better understanding of gestures and the expectations of the designs can be observed and used to match with user expectations. This allows studies to be conducted to find if a target gesture, either static or dynamic is suitable for the command or if an alternative is available. In the study of exertion with respect to finger position and movement ranges a system to observe and quantize exertion on users was created; through this system studies can observe and 3 3 discover characteristics for both harmful gestures and sustainable gestures to facilitate better guidelines with designing gesture databases. Using the system above and a compilation of various in research databases, several characteristics and guidelines were produce to demonstrate the success of the research. The research was able to identify key physical characteristics of gesture making for both exertive and restive gestures and produce guidelines to minimize exertion experienced by users. More extensive work and variety in this study could produce a better understanding and design rules when creating a system that utilizes human gesture inputs. Gesture Based Computer Interfaces As interactions with technology reaches beyond the bounds of Graphical User Interfaces, the need for interaction between Human and Computer to become more natural is growing [12]. While physical sensor on the user will produce accurate measurements and eliminate errors in reading hand positions, vision based system provide a closer approximation to human to human interaction [9]. In order to ensure computer systems have the resources to ensure fast or real time processing speeds, a general framework is common among most recognition software. The main components of which are segmentation, feature extraction and then recognition [41]. Some of the design considerations when creating a recognition system are the specific use; this may involve different peripherals such as infrared sensors, lighting fixtures, and Z-axis considerations [55]. 1.4 Overview of thesis In Chapter 2 of this paper, current developments in the field will be examined; recognition methods, gesture types, gesture categories, and gesture ergonomics will be studied. Chapter 3 will outline the method of analysis for static and dynamic gestures as well as the exertion with respect to the degree of flex the user experiences. Chapter 4 will discuss the results and examine the characteristics and discoveries found using the method listed above. Chapter 5 will summarize the analysis and propose future work and methods that can be undertaken. This work will look at recognition methods primarily taken from single camera based systems that implement color and pattern recognition. This is done because the hardware for this application has already seen widespread distribution and thus has the largest potential to reach the highest number of users. Perhaps the most investigated method of recognition is shape and contour classification, given that this method ignores skin tones, background texture, and color contrast in the image. It has also been shown to have the highest success rate given the diversity of users in size, skin tone, and environment and variety applications. This work will study and observe current gesture databases in order to investigate the nature of the gestures as well as observe its effects on the user. The primary focus of this work is to observe the function of the gesture, the type and its ergonomic impact on the user. This will set a guideline on how to approach gesture design. To accomplish this, a Myo Armband will be employed to obtain EMG readings. 4 2.1 Gesture Recognition Methods Typical vision recognition systems use variety of techniques to recognize gestures. The following are state of the art techniques employed by current research. The most basic these techniques employs the use of RBG color cameras using only color and light spectrum to process the shapes and hand gestures. However, some techniques proposed to employ the use of specialty hardware in conjunction with the RBG color camera to enhance the image processing. 5 5 The Microsoft Kinect is often cited as a prime example of improved recognition techniques at a reasonable cost [44]. Below are recognition algorithms and systems organized by traits; in section 2.1.1 recognition methods that use image characteristics are examined, in 2.1.2, data comparison and retrieval will be examined and in 2.1.3 specialty hardware proposed for improved performance will be investigated. 2.1.1 Image Recognition Techniques Discriminating Features: The method in which the information will be extracted is as follows: First the RGB image and depth will be extracted, and then the histogram of oriented gradients and local binary patterns features are extracted. To effectively extract information, only dynamic regions will be examined [9]. Some algorithms classify and look for details such as finger tips and joints to better recognize gestures. However, this method is extremely difficult as the features may experience collisions and/occlusions. Other features that algorithms may track are 3D features which track a 3D model of the hand relative to specific key features the algorithm looks for [29] [39] [41]. Urdu Alphabets using Geometrical Classification: This system looks to extract features from the hand by color coordination, by coloring different features of the hand in color; the algorithm can track and recognize features more effectively. As demonstrated in the study, the algorithm is able to distinguish the orientation of the fingers as the fronts of the fingers are colored blue while the back is colored red [6]. Shape Parameters: In this method, the followings steps are taken when processing the gesture images, Image Segmentation, Orientation Detection, Feature Extraction, Feature Classification, Bits Generation and finally Gesture-Interpretation. The main feature in this method is the shape the hand is making towards the camera; as such features such as thumb and fingers are more thoroughly examined for interpretation [5]. This method is also called low-level feature as the algorithms observe the contours, shape, or image and compares it to a sample in the database. Though it offers robust and noise resistance though its major drawback is the restriction of backgrounds the system can be used in [29]. Another use of the shape parameter is to supplement the use of dynamic gesture recognition as they can help the system recognize the 6 start and the end point of the gesture [53]. start and the end point of the gesture [53]. Sub Gesture Modeling: In this method the main feature is the flow of the movements, that is, that gestures, much like sentences, are comprised of smaller sub-gestures. As a measure to ensure high recognition chances, the method discusses “start” and “end” phases to gestures [4]. Another description of this type of recognition is a temporally ordered set of states representing spatial information at the scene [44]. 2.1.1 Image Recognition Techniques Dynamic Time Warping: This technique shares similarities with speech recognition in that the process is a similarity measure of the input signal to that of a sample in the database [7]. Local Motion Signatures of histogram of oriented gradients descriptors: The basic frame work for the system is as follows: the first step is to compute the features in the scene, then for each feature, associate a 2D descriptor which is then tracked over time to create a local motion signature. The signatures are then clustered into local motion patterns, the system then extracts the local motion signatures and then voting on what is the most likely gesture from the patterns [11]. Contour Classification: This approach identifies each static gesture as a unique set of contours that can be represented by a series of equations. By filtering each image until only the basic shape of the gesture is obtained and then finding all the contours of the image will create a robust and simple approach to recognition. Furthermore, the use of this technique allows for a system that uses minimal processing to find the related gesture within the database [3]. Another method of using Contour classification is the use of finger counting algorithms based on the shape of the hand. Due to the simple nature of this system, classification and retrieval of the gesture is relatively quick and robust. However, in its current form, the system is only able to recognize a small sample with large distinctions between each gesture [30]. Depth Camera Recognition: Using a metric called the Finger-Earth Mover's Distance to measure hand shape dissimilarities, the accuracy of the recognition system was vastly improved. It has been shown that the metric allows for better shape recognition in terms of speed and accuracy. This allows the system to have better performance in real time [29]. 7 7 2.1.2 Data Management Gesture Vocabulary: One such method of recognition is to use the vocabulary of the system’s gesture language design as a method for recognition and data retrieval. The use of logical gesture sequencing can reduce error and computing power. These are characteristics noted in Gesture Vocabulary studies for both robotic and human vocabularies. 1. Human: Due to the nature of this type of gesture being innate to human nature, there are no predefined limits on designs of this type of gesture [54]. 2. Robot: Participants of the study found that gestures most associated with a specific task is generally the best gesture for that response, however, the study notes that cultural differences may differ the meaning of the gesture [54]. Database Query: As gestures are a series of body silhouettes one proposed method of gesture recognition is to extract global feature from a continuous input from the user then adopting a set of descriptors for matching [28]. Neural network: In order to make gesture recognition more intuitive and ease of use, a proposed neural network has been proposed to enable the system to learn and adjust to new gestures, changes, and minor variations. The problem this approach brings forth is the issue of weights, in the network, current methods generate local minimums and new inputs learned causes the system to forget older ones. The experiments in the article sought to eliminate and minimize these issues [59]. One application of gesture recognition is for security purposes. In the study performed in [34], the use of gesture recognition for electronic examination is considered. To improve security and uphold academic integrity, the study demonstrates that each person has a unique habit during examination. By using gesture recognition to identify the user in question and to monitor action taken by the user, the system is able to identify if the current user is the initial user. The study also proposes that the system, if developed further, can be used to ensure the user does not cheat but looking through a book or browsing through online sources [34]. Another approach to 8 use gesture recognition as security is its use in place of pass codes. A study was conducted using different user bio-metrics to identify a user. By having the user perform a specific gesture, the system can observe and identify similarities to the sample taken previously. 2.1.2 Data Management Such a measure can be used to ensure better security as another user will have to obtain similar bio-metrics to another user in the database [36]. 2.1.3 Specialty Hardware Gaze and Touch Gestures, are among some of the new methods of improving current gesture recognition. It has been proposed the focus of the gaze of the user can help the system recognize and decide the intentions of the user, as manipulations of objects on screen has shown to confuse and even mislead recognition systems. In addition to gaze tracking, the rising popularity of tablet computers has seen gesture recognition software implementation. Some leading research proposes in addition to the touch screen interaction, touch gestures can be applied as well to enrich the experience. These touch gestures make use of the imagery supplied by a gesture and the physical interaction of an on screen object to achieve a meaningful command. One proposed method of control is through the use of gaze gestures, in this study a simple program was created to perform specific tasks based on actions observed in the user's eyes. These tasks were simple web browsing tasks. One idea proposed was the use of gaze to help the system identify and predict user intentions [35]. A similar idea to the study conducted in [35] is to use gaze gesture enhance touch gesture input of phones. One method of implementation of gaze with gestures is the system tracking the user's gaze and focus on an object on screen as a means of selection [47]. Gesture Interfaces are seeing uses in tablet computers, as such, the use of the “touch gesture” has been proposed for use with them. Taking advantage of the typical built in camera or additional peripherals that can be attached, the tablet can create shortcuts or separate commands based on the interaction performed by the user. An example is the basic touch which is the hand in the pointing posture touching an onscreen object; this simulates the basic click, or selector command. With gesture recognition, the camera can read what shape the hand is in when it 9 interacts with the object and use an alternate command, such as copy, edit, etc. (common commands in tablet applications) [47]. Some studies that primarily involved camera based systems have proposed additional hardware to their systems. In many systems, a proposed addition to the camera would alert the system to broad range strokes of the hand or to activate the system when a user is in range for use. 2.1.3 Specialty Hardware Some studies have also proposed the use of these additional systems to identify the start and end of a gesture to further increase the recognition success of these systems. Pyro-Electric Infrared (PIR): This approach uses a 4x4 PIR sensor array to recognize hand gestures. With the use of the PIR sensor is it possible to detect motion from the user rather than motion of the camera, with this system in place, real time processing and recognition is possible. The PIR system is also able to distinguish circular movement rather than the traditional horizontal and vertical movements [10]. Motion Sensor Array: It has been noted that some design applications of gesture input require extreme precision in its design that a simple camera based system may not provide, hence the study proposes a motion capture which allows increased precision when working with the system [53]. Tool/Prop: Some hardware may be required for gesture intensive use in order to enhance gesture recognition; this includes colored gloves or labelling methods to highlight different sections of the hand. This is a novel approach to increase recognition rate at a small increase to the cost as alternate labelling do not require expensive hardware like sensors and microprocessors [24]. Some alternative approaches have been tried to minimize the cost of glove and sensors, and it has demonstrated significant improvement and recognition rates. However, when deciding an application for this technology, the cost versus performance must be considered [57]. One such example of a tool is that of a simple pencil, given that color based camera systems can track uniform objects easily, the use of an object to simulate an actual counterpart will help improve the accuracy of the system. A study was conducted to improve the means of electronic equation entry using gestures and handwriting. The main advantage of this system is that current devices such as phones, tablets and netbooks can make use of the built in camera and software rather than expensive peripherals such as dedicated pens and touch screen surfaces. This also allows the user a larger work space as they are now limited only by the 10 camera's view range rather than the comparatively smaller touch screen [45]. Another approach proposed is the user using gestures to manipulate virtual objects as if they would if they had the object in hand [56]. 2.1.3 Specialty Hardware Another utilization of smart phone technology is to use the phone itself as a tool in the recognition. As demonstrated by the large array of sensors built into the phone such as accelerometers, gyroscopes, and compass, the array of motion-sensors provides accurate and extensive data on the movement of the phone. In this article, the study was conducted using the phone as the gesturing device where in the user moves the phone in patterns that would be recognized by the system so that it may perform predetermined actions. The purpose of this approach is to minimize power overhead of other methods as the power consumption of the phone in an active state is two magnitudes higher than its sleep state. The use of efficient processing algorithms allows the sensors to use lower sampling cycles to further reduce power consumption [58]. Additional Camera: The use of the system allows for the system to recognize depth in gestures. Depth will provide the system with additional features to recognize, improving the recognition success as well as offering redundancy or in some cases, a differential error checking [44]. Sensor Device: Devices that uses the physical quantities such as position and muscle exertion allow for systems improve accuracy as data obtained from these sensors can minimize errors during gesture recognition. Devices such as the Myo Armband employ sensors such as EMG sensors, gyroscopes and accelerometers provide information such as finger positions, orientation and arm movement. This information must be calculated in image recognition and is not as reliable as sensor information [68]. 2.2 Gesture Classifications Gestures are typically placed into two categories, Static and Dynamic Gestures [13] [9] [1] [7]. Static Gestures are observed in an instance of a period of time, and Dynamic Gestures are observed over a period of time [7]. However, this is a simplistic approach that observes directed movement. A more in-depth approach looks at the intent in which the gestures are presented with as defined by these categories: Para-language, Body-language, and Sound-Language. Para- 11 language refers to non-verbal cues the body produces such as intensity and speed of emission. Body-Language is observed as a distinct characteristic of the body being used; this can also refer to such things as pose/form and proximity. Finally, Sound-language refers to features with sound quality, an example is silence, such in the absence of sound or action can be interpreted as having meaning [12]. Typically recognition software is designed for Static or Dynamic Gestures. In the case of Static gestures, algorithms isolate the hand shape or define finger differentials [1] [5] [6]. Static Gestures are usually compared to databases for successful recognition. A sub-set of shapes that is most designed for are sign-language [3] [9] [7]. Dynamic Gestures use motion tracking to compare to a predefined database entry, however, some track the simple transition of one gesture to another [11]. Another dynamic tracking is to track motion commands such as swipes, hand rotation, etc. [4] [10]. Proposed Sign-language data models are designed similar to speak recognition software using offline and online models. Similarly, the approach to this type of recognition is based on speech recognition algorithms. As the simpler of the two types of gestures presented, the basic operation for processing these are fairly simple. The process proceeds as follows: image is identified as valid command by temporal definition or by state definition, a specific feature is extracted. The extracted feature is compared to database of predefined which will then decide what interpretation will fit the data most appropriately [3] [9] [7]. Some processing techniques commonly used for this type of recognition: K-mean [3], Markov Models [9], and Principle components analysis [7]. Some models propose a continuous active model rather than requiring an active state like the above mentioned speech recognition models. This type of model has introduced new issues into the field such as Real-Time processing, Machine Learning, and Gesture Classification [10]. 2.2 Gesture Classifications This process in which this approach is bare much similarity to the process above, however, a necessary step of hand tracking is a fixed requirement. Due to the nature of the static gesture, hand tracking may not be necessary in all cases, however with dynamic gesturing, the algorithm must always be aware of the relative location of the gesture [11]. Some techniques used in this process: Hough-transform-based voting [11], K-Means clustering [11]. Touch Gestures: Using touch screen hardware installed on smart systems (phones, 12 tablets, etc.) touch gestures simulates gesture interactions with objects on screen. Actions such as object selection and move are some of the most popular actions [36]. While dynamic gestures are based heavily and use similar techniques as static gestures, a distinction must be made between the two types of gestures as the uses for each type carry in of itself a meaning. In the case of static gesture, the use of this type denotes a symbol, a representation of an object in the context of the current mode of communication. The use of a static gesture is similar to pressing a key on a keyboard in that the input is of a fixed nature. Dynamic gestures are primarily used as command or action inputs. Using the same analogy as above, this is similar to the moving of a mouse peripheral. Dynamic gestures carry the notation of interaction in that the action is directed to control an object of interest presented to the user. Such an example can be seen when a child is interacting with a toy such as grabbing the object, moving it around, etc. [13]. One method examined to recognize dynamic gestures is to use hand specific gestures to establish a start and an end point for specific gestures in a continuous input. In [53], the study aimed towards building a design system for Computer-Aided Design (CAD) project using human gestures to increase the amount of information that can be transmitted to the computer. The designers decided to use static gestures to supplement the dynamic gestures for simpler recognition. The system has the user perform a static gesture with their left hand, while performing a dynamic gesture with their left. 2.2 Gesture Classifications The static gesture of the left hand tells the system to analyze the actions of the right hand in order to perform the command, when the gesture of the left hand is released; the system understands that the end point of the right hand's dynamic gesture has ended [53]. One method of recognition is to instead treat all user actions as one continuous gesture where each action will be process and mapped to a related action to be performed by the computer. In the method proposed the gesture interface would not simply augment mouse and keyboard interface, but rather replace them. In this instance, the recognition method continuously reads the input and behaves in accordance to the user's actions [60]. 2.3 Gesture and Language Gestures are a type of language that humans create much like speech language. When using gesture language, the complexity and structure are similar to spoken language, complete with 13 grammar, verbs and other such characteristics. However, much like spoken language, gestural language is learned through experience and interactions with parents or other such guardian-like figures. An example of this is the Anna experiment (1932) in which an infant was isolated from birth with minimal to no contact with another living human being. Anna was described as having “no sense of gesture” or indeed little understanding of language. However, through contact and interaction with parental figures, Anna was able to acquire language skills and develop normally much later due to rescue and care [14]. Demonstratively so, humans are known to be hardwired to develop language as even should sense such as hearing are disabled, the area of the brain that is identified as the “seat for language” remains active. Further studies find that all language learned and developed are highly structured and complex. Children of immigrants to a new country with a different language have been documented to know language never heard before; proof that humans do not simply learn language, but part of the process of learning language is to also develop and create language [13]. In a study on aphasia it was discovered that suffers of the condition are able to retrieve words and phrases with accompanied by gestures, as demonstrated in the study, it is shown that gestures share many characteristics with speech and stimulate activities in similar areas of the brain as well [52]. In a study conducted on teacher habits in relation to gesture, it is found that culturally different background affect the use of gestures during instructional sessions. The study observed instructional sessions of mathematics to remove bias from the subject material and observed habits from multiple levels. In the early levels of instructions, teachers used similar gestural techniques to deconstruct problems, however when the sessions of advance and abstract courses, culturally different teachers used wildly different gestural techniques. Western teachers were noted to use fewer gestures and utilized less spatial gestures than eastern teachers [51]. 2.3 Gesture and Language The focus of the user's sight demonstrates the target to which their actions want to control, while this technology is mostly used as an alternative for handicapped users, it has found applications for tablet use in conjunction with other interfacing techniques. Participants performed better with Touch Based Interaction based on the study conducted in [31], which kindergarten children were examined between Gesture Based Interaction and Touch Based Interaction. It was concluded in relation to temperament, touch based interaction yielded a better 14 result. In a study performed in [33] found that touch screen interaction required less focus from the user on the peripheral as well as a preference for directly interacting with an on screen object rather than through the traditional peripheral of button presses [33]. In an experiment conducted regarding touch screen task performance, it was discovered that the texture of the surface yielded different results for the task performed. The experiment concluded that different textures have specific design advantages such as a slippery surface allowing the user to perform faster movements, but at the cost of lower accuracy [46]. In a study conducted on game experiences with pens and touch surface devices, it was found that during the trials, participants did not find visual trace feedback to have importance. [24] It has been discussed that in addition to gestures being based on the internal experience of the user, gestures are also context specific unlike speech, hence its importance and close relationship to speech. It has been noted that gestures do affect speech preferences however; it is shown that gestures mainly provide demonstrative properties to the information being conveyed. On their own, gestures are more difficult to build context in relation to speech. Participants of this study had their gestures habits observed, it was found that gestures that required actions were preformed almost exclusively on the right hand, regardless of the participant's dominant hand. The study notes that this suggests that this is controlled by the left cerebral hemisphere [25]. In another study conducted on infants, similar results have been noted to have occurred; however in infants the gesture of pointing had been used instead [26]. 2.4 Designs of Gestures A study into hand gestures in relation to sign language has been conducted. The focus of this study was to find discomfort in the postures the language invokes. The study proposes that as this technology gains use in everyday applications, users will be in a position that requires 20 to 50 hours of use with the interface and health issues that arise with will share similarities with sign language interpreters [21]. Much like the previously discussed static gestures, the dynamic counterpart to this system is the use of simulated interactions. This system observes and interprets all movements as commands and must decide which commands will be acted upon and when. Designs for this type of system looks to minimize movement from the user to 15 increase comfort when using it. As such, objects of interest are usually kept small and the range of movement minimized to reduce strain on the user [48]. In the study conducted in [55], users were asks to perform a gesture that they believe would be best suited for a specific command, such as selecting. They found that when giving commands such as positioning, selecting and activating, one finger is the norm while commands for orientation, zoom factor and deactivation, participants used the entire hand. The study also describes when manipulating the observational properties of the virtual object, participants use gestures reserved for interacting with physical objects, such as moving the volume closer to their persons for the command “zoom in” [55]. Symbol Gestures: Similar to keyboard inputs, systems like this rely on imagery invoked by gestures to issue commands and inputs. As with the issues presented before, some design considerations must be taken when creating meaningful symbols. Here are some guidelines proposed: Common/extensive use commands assigned to forms comfortable and easy to repeat, hands should be close to the body and below the shoulders, forceful impact should be avoided, wrist and forearm movement should maintain neutral forms, and fingers should avoid prolonged full extension [21] [37]. Simulated Interaction: An alternate design scheme for gestures is to simulate the user would experience if they were to interacting with a real world object. In a robotic control scheme proposed, the user would interact with an object in question while the robotic arm performs actions as controlled by the user [22]. 2.4 Designs of Gestures Hand Preference for Gesture Types: In a study conducted on infants, it is found that participants preferentially use their right hands for communicative gestures [26]. The design of gesture interfaces sees a difficult issue in the balance between recognition and intuitiveness and comfort. It has been demonstrated that rigid gesture structures are more easily recognized and see a much higher success with current recognition systems, however ergonomically, these gestures are uncomfortable given long periods of continuous use; however the intuitiveness of gestures is affected by many factors both in interpretation and in origin [48] [49]. In a study performed on gestural habits of designs in a team design environment it was discovered designers use different gesture types in different roles to effectively demonstrate the theoretical design. However, a large limiting factor of a universal system of design is that gestures, by nature, draw upon the person's own personal experience, as such, ideas and physical 16 meaning is lost when conveyed to another person [50]. Below are factors that influence gesture vocabulary. meaning is lost when conveyed to another person [50]. Below are factors that influence gesture vocabulary. 1. By product of human learning process: Gestures and gesture interactions are a by- product of the human learning process as children learn through physical interaction with surrounding objects [13]. 2. Influenced Socially: As is the case with language, gestures and body language are influenced by social interactions. The effect is that different cultures and different social groups can exhibit many different changes to certain body language [14]. 3. Children demonstrate innate ability of the deep structure of language and the units measuring it [13]. 3. Children demonstrate innate ability of the deep structure of language and the units measuring it [13]. 4. Gesture Types 1. Iconic: Pictorial gestures that represent the shape or style of the object being referred to [25]. 2. Metaphoric: Pictorial gestures that represent an abstract concept being referred to [25]. 3. Deictic: Pointing gestures that refer to a referent of the environment [25]. 4. Beat: Rhythmic movement of the hand(s) that demonstrate the discourse that occurs [25]. 5. Speech difference between gestural or non-gestural: In a study conducted in relation to speech patterns and gestures, it was found that participants that were not allowed to use hand gestures used much more descriptive words and metaphors when asked to describe line drawings. 2.4 Designs of Gestures However the study also notes that with the elimination of gestures, there was an increase in spatial description while there was a decrease in demonstrative phrases [27]. 6. Touch: An experiment conducted on children of advanced age concluded that Touch- Based interactions showed higher learning performances to pure Gesture-Based interactions. It was observed that the Touch Based Interface (TBI) offered more stimuli to the participants resulting in better results than the Gesture Based Interface (GBI) group [31]. 6. Touch: An experiment conducted on children of advanced age concluded that Touch- Based interactions showed higher learning performances to pure Gesture-Based interactions. It was observed that the Touch Based Interface (TBI) offered more stimuli to the participants resulting in better results than the Gesture Based Interface (GBI) group [31]. 17 While the user of spoken language can be unique and completely foreign to different cultures, there is cross-cultural vocabulary in gestures. The most basic and simple of these is the deictic gesture, it was shown that users of different languages can use deictic gestures to establish links in another culture's language; this enhances the learning experiences as this allows a culture to learn the similarities between the two languages. Deictic gestures, while simple and basic, conveys a large amount of information, and it is because of this simple nature that it is an effective tool in gesture-learning systems [51]. 2.5 Summary In the background study various recognition methods were investigated and examined. In the various algorithms and technology, there is very little to no consideration in the design of gesture characteristics, ergonomics and recognisability. Interestingly, the approach only considers varied and shapes that can detail and demonstrate recognition success of the algorithm. Furthermore, in the study conducted in, it was found that sign language interpreters suffer from hand injuries and disorders from the practice, indicating that design foresight into the prolonged usability of the language may not have been considered, or that greater emphasis was placed on recognisability over the comfort of the user [21]. It is therefore important that a method of identifying and observing characteristics of good gesture design and poor gesture design should be created and implemented to accomplish this. As identified in [21], exertion by the hands in unhealthy positions and muscle strain is the cause of such injuries, some focus and investigation into this will illuminate characteristics of gestures and hand positions that would cause such injuries. Some technologies identified to aid in this study are sensor devices, specifically the Myo Armband. Given its simple nature of operation and accessible data registers, this device can be employed to investigate exertion on the user [68]. 18 Gesture Design Investigation Methodology This chapter will detail examination of current gesture databases through different aspects. The approach will look at the following: the types of gestures used in these databases, what kind of static gestures are used, what kinds of dynamic gestures are used, and the ergonomic design of the gestures. From this examination, a set of guidelines should be produced to ensure new gesture designs meet the requirements of being easy to recognize, intuitive, and comfortable. Gesture types are important in this study as these are often defined by the user's culture, background, and experiences [51]. Current studies use touch interfaces as a point of reference for the types of gestures developed, as such, most databases are intuitively adjusted with users that are already familiar with touch technology or have some level of experience with the gestures and actions associated with its use. Databases from different countries and cultures will also demonstrate similarities in some gesture types, as well as highlight key differences that will be useful in gestural designs. To effectively achieve this, all gestures will be placed into three categories: Iconic, Metaphoric, and Deictic [50] [51] [55]. Static and Dynamic gestures can carry different meaning and have different recognition methods, as well as varying challenges associated with each. Static gestures are normally shown 19 to be iconic in nature and are meant as a substitute as a representation of an object or concept. The commands that derive from this set of gestures closely resemble that of button presses. Static gestures invoke a fixed command each time it is presented. Dynamic gestures are gestures that are normally shown in metaphoric or iconic in nature. These gestures are closely related to actions and direct interactions. This type of gesture usually produces commands that closely resemble mouse movements. Dynamic gestures invoke involvement and actions within the environment the user interactions with the system is in. Systems typically recognize dynamic gestures as static gestures with respect to the time-domain, as distinction must be made between the two gestures [50]. Another consideration is the ergonomic design of the gesture. A failure of systems that use touch gestures is that the design does not enhance the experience of using the system, a replacement of current user methods should be for the simple purpose of novelty but rather it fulfills a purpose of either increasing efficiency, or that it is more comfortable for the user. Gesture Design Investigation Methodology Gestures that strain the user will limit the active time the user is able to use the system, as such gestures should be relaxed and provide the least amount of strain to the user rather than the distinctiveness of the gesture. To accomplish this, the use of the Myo armband will be employed. The Myo armband is a device that uses EMG sensors to measure activity in the forearm to determine the shape and form of the hand. This allows the device to recognize what gesture the user is making. Using this tool, it is possible to record and display the effort and strain the user undergoes for each gesture [49] [51]. As a new technology that is rapidly gaining popularity, there are little considerations in the design of the gestures to be used. The best references for these designs are touch surfaces that are still relatively young as well. As such, the failings and shortcomings of these designs are affecting the willingness of users to experience the upcoming technology. It is also a fault of developers that do not put thought or importance of interface design when designing the interactive system as the experience of the interaction affects a user's willingness to continue using the interface. However, as the technology continues to advance, new status quo, cross cultural elements, and customs emerge, the technology and the designers are incorporating these new trends into their databases, however a guideline should still be set on how this approach should be taken [49]. 20 3.1 Gesture Types Recognition of gestures can be categorized into three types: Iconic, Metaphoric and Deictic. Gestures taken from databases will be categorized in to these three categories. Each type of gesture serves a specific purpose. Iconic gestures are the most simple and most used gestures for recognition, these are primarily static gestures that rely on the image they invoke to convey their meanings, this plays favorably to recognition software as the images they invoke are either commonly experienced symbols, imagery, or objects, the user may use gestures to simulate or represent the functionality of an object that may or may not be on screen. American Sign Language is usually used as iconic gestures; Figure 3.1 demonstrates the American Sign Language (ASL) equivalent of the letters “A”, “S”, “L”. Dynamic gestures can also fall into this category as they invoke an image or symbol. To fall into this category, the gesture must display these characteristics: they are symbolic, either using an abstract shape that is corresponding to a fixed meaning. They are in the general shape of an object in question. They are a simulation or representation of an action that an object in question can perform [21] [25] [37]. Figure 3.1: Example of iconic gestures, American Sign Language letters ASL Figure 3.1: Example of iconic gestures, American Sign Language letters ASL Metaphoric gestures are primarily dynamic gestures that invoke abstract functions or concepts. Gestures that typically fall into this category are movement gestures, such as the act of picking up a virtual object and moving it around the interactive environment. Figure 3.2 shows an example of a metaphoric gesture. Other abstract interactions will also fall into this category such as describing the shape of action of an object. To fall into this category the gesture must 21 display the following characteristics: They do not correspond with any fixed meaning. There is no real world object that corresponds with the actions performed in the gesture [25] [50]. display the following characteristics: They do not correspond with any fixed meaning. There is no real world object that corresponds with the actions performed in the gesture [25] [50]. 3.1 Gesture Types Figure 3.2: Example of Metaphoric gesture, the spreading of the hand Figure 3.2: Example of Metaphoric gesture, the spreading of the hand Deictic gestures as shown in Figure 3.3 are gestures that point to a direction, these types of gestures of closely related to iconic gestures; however, deictic gestures are spatial in nature and require awareness of the space they are used in to be effectively utilized. Both dynamic and static gestures make use of this type. The main gesture that falls into this category is the pointing gesture. To fall into this category, the gesture must display the following characteristics. Any gesture that has a directional vector as part of its information is considered a deictic gesture. This type of gesture draws attention to a specific location for the purpose of conveying information. Directional movement as part of a simulation or representation is not included in this category [25] [50]. Figure 3.3: Example of deictic gesture, the pointing gesture Figure 3.3: Example of deictic gesture, the pointing gesture 22 3.1.1 Static Gestures Static gestures as shown in Figure 3.4 are gestures that issue a command specific to the shape that the system interprets. Sign language, key prompts, and key commands are commands that make the most use out of static gestures. Given their simple nature, static gestures have a higher recognition success rate given that highly unique gestures can be used to issue commands to a system that uses few commands. Databases that tracks these gestures are mostly image based processing, as such; image pattern recognition is the most widely studied method for tracking and recognition. However the limitations of this approach are apparent as the system only achieves success when the hand shapes are highly varied from one another. It is hence proposed that modeling techniques also be applied to these system to improve upon the success and to allow the system to recognize more subtle changes and variations in the user's hand shapes. This allows for a larger gesture set that the system will be capable of recognizing [1] [3] [7]. Figure 3.4: An example of a static gesture Figure 3.4: An example of a static gesture Currently the most effective use for static gestures is the emulation of a button press, or in this instance, an iconic use. Databases that use static gestures map the shape and profile to a set command, most often a key press, like in the American Sign Language, the shape of a specific gesture corresponds with a specific letter. Other examples include making a fist to command the system that a “right-click” command is being issued. In design system aimed to making gesture interaction a design tool similar to auto-cad, static gestures are used to select tools, often by imitating their shape to switch use to these tools. This allows the system to obtain 23 information from the user based on experience rather than memorization of names and symbols as the user can convey the information visually using shapes that are much more memorable to the user than names. 3.1.2 Dynamic Gestures Dynamic gestures, shown in Figure 3.5, are gestures that generate a motion corresponding to a command, the gesture, from its direction, hand shape, and overall speed can convey a large amount of information for the system to interpret. They are often used with commands such as object moving, control, and interaction. There are currently no standard method of recognizing dynamic gestures, however some issues that are still currently with in studies is how to define the start and end of a gesture as inaction or stillness of a dynamic gesture can have meaning as well. Databases that use this system often store the information in frames, each frame is treated as an individual image and they are transformed into simplified representations and compared to a stored value in the database, if the system finds a match, then the system will use the corresponding movement to match with the appropriate command. However, newer algorithms use sub-modeling techniques to gesture movement, allowing the system an alternate method to recognize dynamic gestures [7] [11]. Figure 3.5: Example of dynamic gesture, wave out Figure 3.5: Example of dynamic gesture, wave out Dynamic gestures see many use in object focused applications. Currently databases that have their primary focus on dynamic gestures often use gestures in a simulated environment. 24 One such application is gaming where the user will interact with objects on screen as they would if the object in question were real. The use of this can be both Iconic and Metaphoric; the object in question can be a representation of a real object, demonstrating Iconic use. The user is showing to the system the actions associated with the object. However, metaphoric approaches see the dynamic gestures controlling more abstract concepts, like interacting with a representation of a concept with no physical equivalent [12]. 3.2.1 Exertion The exertion of a gesture impacts the usefulness of the system; exertion in this context is defined as the user exerting their bodies beyond a reasonable and comfortable degree. An example of an exertion gesture is a pointing gesture with the finger fully extended; repeated use of this gesture is known to incite pain in the subject. Exertion and the effort required to make the gesture are factors in the comfort and usability of the system. Current databases will be analyzed and observations will be made to how much exertion the gesture places on the user. Using the gestures extracted from databases, EMG readings will be recorded in reference to the signal levels required. The measurement of the muscle group in the forearms will give an indication to the level of exertion the user experiences when making a gesture. To accomplish this, a Myo armband was used to measure the muscle activity when the user makes a specific gesture. The subject used in this measurement is a male in their mid-twenties. The Myo armband is a device that uses EMG readings to map gestures, by matching the readings to a predefined record, a gesture can be recognized. Each armband has 8 sensor modules called pods that attach to position in the arm in order to get readings. The pod numbering in accordance at the time of experimentation is shown below [68]. 25 Figure 3.6: Myo Armband pod numbering Figure 3.6: Myo Armband pod numbering As the measurement device is an armband with sensors, below is a description of the pods and their relative locations to the muscles of the arms [69].  Pod 1: Flexor Carpi Ulnaris  Pod 2: Brachioradialis  Pod 3: Extensor Carpi Radialis Longus Brevis  Pod 4: Extensor Carpi Ulnaris  Pod 5: Extensor Digitorum  Pod 6: Extensor Carpi Radialis Brevis  Pod 7: Flexor Carpi Radialis  Pod 8: Flexor Carpi Ulnaris  Pod 1: Flexor Carpi Ulnaris  Pod 2: Brachioradialis  Pod 3: Extensor Carpi Radialis Longus Brevis The Myo was selected as the device to measure exertion due to its ease of use and simple design. The Myo device offers easy access to the EMG signals of different subjects. It should be noted that due to the varying characteristics of different people, EMG signals, even with precision instruments will still vary from person to person. 3.2.1 Exertion Thus, the Myo Armband is not in an immediate disadvantaged to more precise instruments. Additionally, the Myo Armband was 26 designed with various tools to access the data registry of the device to create custom software to make use of the data obtained by the device. This made the Myo Armband an excellent candidate to measure exertion levels once the appropriate method had been identified. Once the Myo Armband has been secured on the arm, the subject entered a rest posture and waited until the Myo Sensors had reached a steady state. Once a steady state had been achieved, a button was pressed to start the averaging program; the gesture was then performed during this time. For static gestures the position is held until a steady state was achieved, for static gestures a pilot study was conducted to determine the repeatability of the measures. This pilot study was performed on ASL gestures A, B and C (Appendix 1, Table 4.2). In this pilot study each of the three gestures was held for an average of 2 seconds each and the steady state values were recorded from the Myo output. This process was repeated 3 times with a rest period in between each trial. The 3 measures were averaged; this process was repeated for the other 2 gestures. It was found that the 3 values obtained for each gesture within +/-10 activities levels. Therefore it was decided to conduct one trial per gesture, the static gestures were held for 2 seconds each. Dynamic gestures were performed 3 times in succession, and the 3 values obtained from the Myo device was averages for each gesture. Once the actions previously have been performed, a button is pressed to stop the program and begin its calculations. The values are then recorded into a data sheet. The subject enters a period of rest and prepares the next gesture. This process will be repeated until all database gestures have been recorded. As an initial test of the Myo armband, the arm was made into a tight fist (Figure 3.10) in order to get large sensor readings, as the graph indicates the sensors have a range of -127 to 127, which will be used in this study to simplify the examination of the exertion on the user. Figure 3.8 is a graphical representation of the data obtained by the Myo device. 3.2.1 Exertion Many samples were taken using a tight fist and it was found that the highest appropriate exertion level the user experiences is 110 to -110. Currently the developer of the Myo Armband has not released full data specification on the sensors of the device, and has described the values as activation values of the sensors [68]. During a test with the rest position it was found that the rest position (Figure 3.9) when the hand is fully rested for a period of time, that value read on the graph was close to 0, though features of signals were still apparent. For the purposes of this study, the rest position and all values close to that feature will be considered 0. Figure 3.7 demonstrates the graphs 27 produced using measuring software provided by the developer; the gesture used was a tight fist as demonstration of the possible exertion a user can experience. Figure 3.7: Myo EMG reading of a closed fist Figure 3.7: Myo EMG reading of a closed fist First a neutral stage will be established as a reference point for the rest state, any activity that deviates from this state will be considered exertion on the user. The user's arm will be relaxed and held idle, facing a computer camera as the standby phase of the system. Following a list of gestures found previously, the measurements of the exertion on the muscles will be recorded and compared to the rest state and a modified relaxed gesture to demonstrate reduced exertion. Gestures used for the same or similar commands will also be compared to find the gesture with the least exertion. In the context of this study, the activation levels (bits) of the sensor data will be referred to as the Activity Level as greater exertion is indicated by higher activity in the muscle groups. Common features will be observed between the gestures with the least exertion to establish guidelines into the design of gestures for recognition. Table 3.1 is a chart representing the hand in its neutral or rest state. The chart will be simplified as a range from 0 to 110, where 0 is the rest position and the user fully exerting their hand will be 110. These values were obtained by 28 averaging rectified signal. The values were then filtered to the nearest tenth to remove noise. 3.2.1 Exertion Equation 3.1 was implemented into the algorithm in Figure 3.8 which demonstrates the simple algorithm used to accomplish this. In Equation 3.1, A is the average Activity Level, D is the data from the pod, P is the pod number, and N is the number of samples the device took during the gesture recording process. Even in the rest position, the subject still gives off minor readings, however these readings are still the lowest given any activity will produce a larger signal. During a trial with the subject, it was found that the highest value produced from full exertion was 110; hence this value will be used as the upper limit of the readings. Table 3.2 has the Activity Level normalized to 110, which was the highest value obtained during the recording process. Gesture Pod 1 Pod 2 Pod 3 Pod 4 Pod 5 Pod 6 Pod 7 Pod 8 Rest 0 0 0 0 0 0 0 0 Fist 110 70 40 80 40 20 40 60 Table 3.1: Rest and Exertion point reference table (Activation Bits) Table 3.1: Rest and Exertion point reference table (Activation Bits) To create a better comparison for other subjects, Table 3.2 is the normalized activation level of Table 3.1. Gesture Pod 1 Pod 2 Pod 3 Pod 4 Pod 5 Pod 6 Pod 7 Pod 8 Rest 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Fist 1.00 0.64 0.36 0.73 0.36 0.18 0.36 0.55 Table 3.2: Rest and Exertion point reference table (Normalized) Table 3.2: Rest and Exertion point reference table (Normalized) As these are the muscles responsible for finger movement, the usage of these muscles has a direct correlation with how much exertion will be placed on the user. Hence the higher the uses, as measured as an EMG reading, the more exertion it will be on the user. In trials with the device balling the hand into as tight fist as produced the highest reading of any of the identified gestures, this is shown in Figure 3.10. 29 (3.1) (3 1) (3.1) The algorithm in Figure 3.8 was implemented as a driver for the Myo device. Figure 3.8: Sensor Data Rectifying, Rounding, Averaging Algorithm Figure 3.8: Sensor Data Rectifying, Rounding, Averaging Algorithm Figure 3.8: Sensor Data Rectifying, Rounding, Averaging Algorithm 3.2.2 Position Another factor in the ergonomic design of a gesture is the range of movement that the user must make to produce an approved gesture. Gestures that use the upper limits of this range causes discomfort and pain in some instances, even without prolonged use. An example of an extreme form of gesture is a “stretch” gesture, illustrated in Figure 3.11, where the natural movement of the wrist or finger is flexed to the absolute limit that the user can tolerate without injury. Extreme gestures such as these will cause injury and damage over time and repeated use. The range of movements possible could go beyond that the natural unassisted range, an example is the use of the other hand to flex a finger past what the muscles and tendons are naturally capable of [49]. 30 Figure 3.9: Relaxed hand used in experiments (Reference 0, 900) Figure 3.9: Relaxed hand used in experiments (Reference 0, 900) To establish the range of movement of a gesture the reference point of a hand in rest position is taken. The two extremes from this rest position are the full extension and spread of the fingers to a tightly clenched fist. The reference point will be considered 0 and the two extremes will be assigned a value of 10 and -10, where the value of 10 will be the fingers fully extended and spread out and -10 will be the hand in a clenched fist. The values will be assigned to each finger and the thumb in order to observe the movement range of the gestures and its effect on the user. The fingers were measured where a fully extended hand measured 0 degrees and a fully clenched hand will have the fingers pointing the opposite direction, or at the 180 degree mark with a relaxed hand approaching approximately 90 degrees. Due to the inconsistent nature of gesture making and for comparison using the user’s limits, the system described above was implemented instead. In a test with using measurements it was found that for more difficult gestures, measuring gestures became inconsistent due to fatigue, difficulty maintaining the gesture or involuntary muscle twitches. The scale was introduced to reduce the time required to sustain the gesture. 3.2.2 Position With degree measurements it was found that the subject experienced an error of +/- 10 degrees roughly, however with the scale there was little to no error, and gestures fell within their expected range 95% of the time. Figure 3.10: Closed fist used in experiments (Reference -10, 1800) Figure 3.10: Closed fist used in experiments (Reference -10, 1800) 31 Due to the extra movement of the thumb in reference to the fingers, additional values have been assigned to observe and measure gesture exertion. The thumb in addition to flex and extend movement is also capable of vertical and horizontal movements, this movement will be also assigned a value of -10 to 10. Figure 3.11: Fully extended open hand used in experiments (Reference 10, 00) Figure 3.11: Fully extended open hand used in experiments (Reference 10, 00) Below is a chart demonstrating the breakdown of how movement range is represented. Each gesture will be identified and have its corresponding movement range recorded. Gesture Thumb Axis Thumb Flex Index Middle Ring Little Wrist Rest 0 0 0 0 0 0 0 Fist -10 -10 -10 -10 -10 -10 0 Extended 10 10 10 10 10 10 0 Table 3.3: Rest, flex and clenched reference Table 3.3: Rest, flex and clenched reference The choice using a scaling system of -10 to 10 is due to different persons has different physical characteristics. For example the subject in the experiments has a range of movement of 0 degrees with respect with the palm; however, another person had measured to have a maximum flex of 5 degrees. The scaling was implemented using the subject's maximum limits, and the use of this system will enable other studies into this interaction to use a similar system for comparison. Table 3.4 demonstrates the flex indicator as degrees where the top row values are the range as defined in this study, and the bottom row values are the corresponding degrees. 3.2.2 Position 32 Scale 10 9 8 7 6 5 4 3 2 1 0 -1 -2 -3 -4 -5 -6 -7 -8 -9 -10 Degree(0) 0 9 18 27 36 45 56 63 72 81 90 99 108 117 126 135 144 153 162 171 180 Table 3.4: Range assigned to each value by degrees (X0) Table 3.4: Range assigned to each value by degrees (X0) Figure 3.13 demonstrates the process in which the hand positions will be recorded for static (A) and dynamic gestures (B). The subject will first perform the gesture, if the gesture was static, the subject will hold the gesture while each individual finger’s position is recorded using the scale, once the fingers have been recorded, the thumb’s bend and angle are recorded. Once the thumb has been recorded, the wrist bend is recorded. For dynamic gestures, the entire gesture is performed, in this process the maximum range that the gesture has achieved is measured as the hand is held in a position for measuring. Once again, the fingers will be measured first, then the thumb, and then the wrist. The data is then recorded and the subject prepares the next gesture. Table 3.4 shows the angle range for each reading of the scale. Figure 3.12: Position Measuring Block Diagram; (A) Static, (B) Dynamic Figure 3.12: Position Measuring Block Diagram; (A) Static, (B) Dynamic 33 3.3 Summary With this method of classifying gestures into sub categories and investigating their effects on the user, a better understanding of how gesture interaction with machines can be achieved effectively. The consideration of the type of gesture to be placed into the system will help design its look. Iconic, metaphoric, and deictic gestures each have separate meanings and frequency of interaction with the user on a regular basis, the categorization of gestures in this style enable the designer to select a gesture that maximizes recognition and minimizes exertion on the user. Furthermore, the separation of static and dynamic gestures allows designs to design their focus on how their system will respond to these two types of commands. It has been shown that the algorithms for both dynamic and static gestures can vary greatly, and if the system is able to recognize which type the user is attempting to make, can select the more appropriate recognition algorithm. Another focus of gesture interaction with a computing system is the exertion on the user, if the system causes more exertion on the user than traditional systems, then the use of gesture recognition in the system would be useless. Consideration must be made such that the minimization of exertion and effort for the system to be effective. The assessment of currently developed gestures will help in reducing exertion on the user given that modification or alternate gestures can be used in place to reduce exertion. This focus needs to be applied to movement range of the hand as well. This approach to categorizing and observing will allow for a better understanding of how gesture interaction with machines should be approached. The use of gesture types allows the system to place more emphasis on a specific type of gesture given its specific and unique application. Furthermore, while the use of static gesture is a more simple approach, the nature of how humans communicate using a combination of static and dynamic gesture should be reflected in how machines are developed to recognize this. As the relation between human and machine interaction is further deepened and explored, it is also important to observe the effects the interaction has on the user, in this instance, exertion and comfort will define the success of the system to a growing user base. 34 Experiments and Results As outlined previously gestures were obtained from databases currently in use for gesture recognition development, these were categorized into different gesture organizations and gesture types. They were then examined in the aspects of exertion and movement range. For the gesture classification, gestures were individually looked at and counted in databases and compiled. Gestures that were the same or similar were disregarded for subsequent inclusions, some factors into separating gestures with similar shapes and movements are the meaning and purpose implied in the original context where the gesture is used. For example it was found that the fingers fully extended and closed was a common gesture in all databases; however its orientation altered the meaning of the gesture or the information it conveyed. These modifications were considered different gestures and classified accordingly. The classification noted that gestures used tend to be approximations to hand gestures used when interacting with real world objects, such as dial and lever controls. The exertion of the gestures were also measured using the Myo Armband, as shown, the greater the exertion by the user, the higher the EMG readings are on picked up by the armband. Gestures compiled in the classifications previously were all subject to stress measurement using the Armband, with the EMG levels recorded into a table. The records indicate what the readings were of each pod, or sensory section of the device. It should be noted that maximum flex or a tight fist do not necessarily produce the highest readings as the user may not be under full 35 exertion when making the gesture. As the figures demonstrates that pod 3 and 4 experience the most exertion, in reference to the placement of the pods on the subject these two sensors correspond with the Extensor Carpi Radialis Longus Brevis muscle and the Extensor Carpi Ulnaris muscles with the values 44.7 and 39.5. The movement ranges of the gestures were also considered with respect to the fingers, thumb and wrist. As with the previous experiment, the categorized gestures were measured in accordance to the tolerances and limits of the subject. Experiments and Results As shown in the previous experiment, the maximum extension of the fingers and thumb as well as making a fist both require exertion and the resting point is in the approximate middle location between the two extremes, the scaling was decided to be -10 to 10 where -10 represented the fingers positions are touching the palm, and 10 representing the fingers fully extended. The thumb demonstrates a wider range of movements and thus was given two measurement parameters in the thumb axis and flex. These were also assigned a value of -10 to 10. The wrist, as the final component in this study is assigned a value between 0 and 10, where 10 is the maximum flex it can make in any direction. In this study it was found that gestures used in current databases have a tendency to induce maximum flex or maximum retraction in combinations to achieve their intended gestures. As such, many of the gestures have a rating of 10 or -10 for their respective fingers and wrists. The trial was conducted in computer lab with the participant sitting down in front of a desktop computer with data capture software to capture and process the data stream from the device. The participant has the Myo armband secured on their right forearm. To establish reference points for the purpose of this study, the subject made an effort to obtain full exertion by as many muscle groups as possible, once full exertion as recorded, the subject entered a period of rest state to obtain the rest readings. The process begins again with the subject performing gestures from the compiled database in trials of three; the data is averaged between the three trials for each gesture and recorded. This process is repeated for all gestures listed in the database. 4.1 Gesture Categorization As previously discussed, gestures were classified into Static and Dynamic gestures and then 36 further categorized into Iconic, Metaphoric, and Deictic types. In the databases used in this study, a total of 66 gestures were identified and categorized. Table 4.1 demonstrates this categorization. This table was compiled using databases taken from various recognition studies as well as independent research databases obtained on-line; research databases comprised of multiple participants with both genders and a variety of skin tones. Gesture Description Static: Iconic ASL American Sign Language, each symbol used in this set is a representation of a letter in the English alphabet. These are usually used in place of a keyboard to type or enter a character into an interactive system. (For all letter gestures, please refer to Table 3.2 in Appendix 1) Stop Gesture A forward facing palm that is normally associated with the command or request to stop. Okay Gesture Resembling the letters O and K, this is generally known to be related to a positive response to an inquiry. Description American Sign Language, each symbol used in this set is a representation of a letter in the English alphabet. These are usually used in place of a keyboard to type or enter a character into an interactive system. (For all letter gestures, please refer to Table 3.2 in Appendix 1) ASL ASL A forward facing palm that is normally associated with the command or request to stop. Stop Gesture Stop Gesture Okay Gesture Okay Gesture Okay Gesture Okay Gesture 37 37 Fingers Closed The fingers are closed together simulating a flat surface or a floor. This gesture is sometimes accompanied by another gesture made by the other hand representing another object on this surface. Thumbs up The Fingers are balled up into a fist with the thumb facing up, this is associated with the concept of “Yes” in an inquiry Thumbs Down The Fingers are balled up into a fist with the thumb facing down, this is associated with the concept of “No” in an inquiry Deictic Point A single finger point into the direction of interest by the user, this is the most common deictic gesture and is used by the majority of databases that require spatial awareness as part of its interaction. The fingers are closed together simulating a flat surface or a floor. 4.1 Gesture Categorization This gesture is sometimes accompanied by another gesture made by the other hand representing another object on this surface. Fingers Closed Fingers Closed Fingers Closed The Fingers are balled up into a fist with the thumb facing up, this is associated with the concept of “Yes” in an inquiry Thumbs up Thumbs up Thumbs up Thumbs Down Point Point Point 38 Hand point Using all fingers and thumbs to point to specific direction, this is a highly uncommon pointing gestures but it is sometimes used when confirming four or more objects are related to the direction identified. Two-Finger Point Using two fingers to point in specific direction, this deictic gesture is normally interchangeable with the single point gesture; however, sometimes it is used to declare two objects in relation to the identified direction. Dynamic: Iconic Circular Right The hand is in a neutral position and makes a circular motion right at the wrist. This simulates an object rotating right. Circular Left The hand is in a neutral position and makes a circular motion left at the wrist. This simulates an object rotating left. Using all fingers and thumbs to point to specific direction, this is a highly uncommon pointing gestures but it is sometimes used when confirming four or more objects are related to the direction identified. Hand point Hand point Hand point Using two fingers to point in specific direction, this deictic gesture is normally interchangeable with the single point gesture; however, sometimes it is used to declare two objects in relation to the identified direction. Two-Finger Point Circular Right Circular Right Circular Left Circular Left Circular Left 39 Dial Control The index and middle finger at pinched forward with a space in between representing an object between them; the hand is then moved in a twisting motion with the space in between the index and middle finger at the axis. Lever Control The fingers are curled up as if there was a bar between the palms in fingers. The entire arm is then moved up or down to simulate lever operation. Wave Right The hand is held with the fingers facing forward vertically and then the hand is moved to the right at the wrist. A common use for this gesture is to signify the user brushing an object to the side. 4.1 Gesture Categorization Wave Left The hand is held with the fingers facing forward vertically and then the hand is moved to the left at the wrist. A common use for this gesture is to signify the user brushing an object to the side. The index and middle finger at pinched forward with a space in between representing an object between them; the hand is then moved in a twisting motion with the space in between the index and middle finger at the axis. Dial Control Dial Control Lever Control The fingers are curled up as if there was a bar between the palms in fingers. The entire arm is then moved up or down to simulate lever operation. Lever Control The hand is held with the fingers facing forward vertically and then the hand is moved to the right at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Right Wave Right The hand is held with the fingers facing forward vertically and then the hand is moved to the left at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Left Wave Left Wave Left 40 Wave Up The hand is held with the fingers facing forward horizontally and then the hand is moved up at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Down The hand is held with the fingers facing forward horizontally and then the hand is moved down at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Lever Spin The fingers are curled up as if there was a bar between the palm and fingers. The hand is then rotated left or right, similar to dial control. Lever Pull The fingers are curled up as if there was a bar between the palms in fingers; the hand is then pulled towards the user in a puling motion. The hand is held with the fingers facing forward horizontally and then the hand is moved up at the wrist. A common use for this gesture is to signify the user brushing an object to the side. 4.1 Gesture Categorization Wave Up Wave Up The hand is held with the fingers facing forward horizontally and then the hand is moved down at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Down Wave Down The fingers are curled up as if there was a bar between the palm and fingers. The hand is then rotated left or right, similar to dial control. Lever Spin Lever Spin The fingers are curled up as if there was a bar between the palms in fingers; the hand is then pulled towards the user in a puling motion. Lever Pull Lever Pull 41 Switch The index finger is held forward and then either moved up to down to simulate the look of flipping a switch. Pinch The index finger and thumb move closer together until they touch. In touch interfaces this gesture is used to zoom out but in vision gesture interactions is used as an object select or grabs function. Spread Pinch The index finger and thumb move further apart from a touching position. In touch interfaces this gesture is often used with the zoom in feature, however, in vision gesture interaction systems; this is used to release an object under select. Grab The fingers are open and then are closed into a fist. This gesture is commonly use with the select command and used to select an object in virtual 3D space. The index finger is held forward and then either moved up to down to simulate the look of flipping a switch. Switch Switch The index finger and thumb move closer together until they touch. In touch interfaces this gesture is used to zoom out but in vision gesture interactions is used as an object select or grabs function. The index finger and thumb move closer together until they touch. In touch interfaces this gesture is used to zoom out but in vision gesture interactions is used as an object select or grabs function. Pinch Pinch The index finger and thumb move further apart from a touching position. In touch interfaces this gesture is often used with the zoom in feature, however, in vision gesture interaction systems; this is used to release an object under select. Spread Pinch Spread Pinch The fingers are open and then are closed into a fist. 4.1 Gesture Categorization This gesture is commonly use with the select command and used to select an object in virtual 3D space. Grab 42 Press down The palm is faced down and the entire hand is moved down. Used for pushing an object down or to signify the lowering of an object. Lift Up The palm is faced up and the entire hand is moved up. Used for raising an object up or to signify the raising of an object. Arm Rotate The entire arm is rotated back and forth. This gesture is commonly associated with object rotation. Knocking A fist is made with the palm facing forward; the fist is then moved back and forth at the wrist. This gesture is used to simulate the knocking of a door. The palm is faced down and the entire hand is moved down. Used for pushing an object down or to signify the lowering of an object. Press down Press down Lift Up Lift Up Arm Rotate Arm Rotate Knocking Knocking Knocking 43 44 Two-finger pinch The index finger and middle finger are moved until they touch the thumb. Similar to the pinch gesture before, however this has some use as an alternate select. Index Finger Pull The palm is facing up with the index finger fully extended, it is then retracted. This gesture is usually associated with the “come here” command. Metaphoric Good-bye wave The palm is facing forward while the hand is moved side to side at the wrist. This can be associated with the concept of “good bye”, and ending sequence. Index-middle close The index and middle fingers are in a V-shape and then are closed together. This can be used to demonstrate two objects coming together. The index finger and middle finger are moved until they touch the thumb. Similar to the pinch gesture before, however this has some use as an alternate select. Two-finger pinch Two-finger pinch The palm is facing up with the index finger fully extended, it is then retracted. This gesture is usually associated with the “come here” command. Index Finger Pull Index Finger Pull The palm is facing forward while the hand is moved side to side at the wrist. This can be associated with the concept of “good bye”, and ending sequence. Good-bye wave Good-bye wave The index and middle fingers are in a V-shape and then are closed together. 4.1 Gesture Categorization This can be used to demonstrate two objects coming together. Index-middle close Index-middle close 44 Right Grab The hand is tilted right and then moved into a fist centered in the middle. This was used to signify moving from the right to the center. Left Grab The hand is tilted left and then moved into a fist centered in the middle. This was used to signify moving from the left to the center. Open to close fingers The fingers are spread out then closed together, but still extended. This gesture was used to signify the ending or closing of an object. Spread Hand The fingers are closed together, extended and are then spread out. This gesture was used to signify the compression. The hand is tilted right and then moved into a fist centered in the middle. This was used to signify moving from the right to the center. Right Grab Right Grab Left Grab Left Grab Open to close fingers Open to close fingers Spread Hand Spread Hand 45 Arm side to side The wrist and hand are in fixed position while the arm moves them side to side. This gesture used to signify the shaking or movement of an object. Wrist Tilt The hand is in neutral position and is tilted left and right at the wrist. This object was used to demonstrate the rotation or axial movement of an object. Deictic Tilt The hand tilted in the desired direction. This gesture is used as a command to signify when to move to the desired direction. The object or concept is instructed to stop, when the hand tilts to a direction, it is instructional for the object to move in that direction. Fork Tilt The index and middle fingers are in a V-shape and then pointed at a desired direction. The wrist and hand are in fixed position while the arm moves them side to side. This gesture used to signify the shaking or movement of an object. Arm side to side Arm side to side Arm side to side Wrist Tilt Wrist Tilt The hand tilted in the desired direction. This gesture is used as a command to signify when to move to the desired direction. The object or concept is instructed to stop, when the hand tilts to a direction, it is instructional for the object to move in that direction. 4.1 Gesture Categorization Tilt Tilt Fork Tilt Fork Tilt 46 Palm With the fingers extended, the palm is pushed into the desired direction. Used as a command to push an object in the desired direction. Moving Point With the index finger extended, the entire hand is moved forward in the direction that the finger is pointing in. Table 4.1: Categorization of compiled database gestures With the fingers extended, the palm is pushed into the desired direction. Used as a command to push an object in the desired direction. With the fingers extended, the palm is pushed into the desired direction. Used as a command to push an object in the desired direction. Palm Palm Table 4.1: Categorization of compiled database gestures In Table 4.1 ASL gestures were grouped into the single grouping as the gestures carried similar meaning but for different symbols. The full description of all ASL gestures used in this study can be referred to in Appendix 1, Table 4.2. For the compilation of this database, gestures of similar shape and style were omitted when found in later databases. As a study into the exertion of gestures, different hand positions and styles were considered to be more preferable to minor variants of similar gesture groups. The compilation of the databases gestures demonstrate that simulated object manipulation is the preferred of interaction for database systems. As seen in previous research, the manipulation of virtual objects via gesture recognition offers context and background for the system to be more effectively utilized. For systems that operate on static gestures map them to predefined commands. Static gestures are usually universal in their meanings, such as stop, positive/negative response, and directional. As such there were no metaphoric gestures found in the databases that utilized static gestures. Dynamic gestures were found to have more variations and more complex commands are assigned to them. Given their nature, it was found that context is very important for these gestures as the information given by 47 one dynamic gesture can offer multiple interpretations. Some gestures are universal in their roles such as simulated object manipulations, or conventional knowledge. Dynamic gestures also offer several variations that had no specific meaning nor could be interpreted without context provided. These gestures were not commonly found in databases they were mapped to commands that were not reflective of the shape or action of the gesture. 4.2 Gesture Exertion on User In chapter 3, the use of the Myo Armband was introduced in measuring exertion by the subject. To accomplish this, the subject first established two reference points, a relaxed rest posture and then made full exertion in a tight fist. Both values were recorded and assigned values based on readings measured. Initial values were ignored and only the steady state was measure and recorded. Due to involuntary muscle twitch and other factors, the exclusive inclusion of the steady state was ideal. Figure 4.1 demonstrates the graph obtained from full exertion by the subject. Figure 4.1: Myo Armband EMG reading for full exertion Figure 4.1: Myo Armband EMG reading for full exertion The recording would begin when the subject has made a gesture and held the posture for a one to two seconds, after holding the posture until a steady state has been recorded, the subject 48 will then adopt the rest posture and wait for the readings to reach their rest states, then the subject will perform another gesture and hold the posture until a steady state has been reached. This process is repeated for all static gestures and their values recorded, Table 4.2 is a graphical representation of the data obtained during trials for ASL B and ASL C. Figure 4.2: Myo Armband EMG readings for: ASL B and ASL C Figure 4.2: Myo Armband EMG readings for: ASL B and ASL C For dynamic gestures, due to their nature do not have a steady state, however the gestures were performed repeatedly for several samples, the average peak of these values were then recorded. As with static gestures, the subject then adopted a rest position for several seconds before making the next gesture. This was repeated until all dynamic gestures were recorded, Figure 4.3 demonstrates the data obtained during dynamic gestures. 49 Figure 4.3: Myo Armband EMG readings for: Tilt, Fork Tilt, Palm and Moving Point Figure 4.3: Myo Armband EMG readings for: Tilt, Fork Tilt, Palm and Moving Point Tilt, Fork Tilt, Palm and Moving Point In Appendix 2, Table 4.5, graphical data regarding the Activity Levels of all gestures are recorded. They are similar to the data obtained by the averaging software, however not all gestures recorded in the table will match with the data in Table 4.2 and Table 4.4 due to the inability of the human body to perfectly replicate a specific action. 4.2 Gesture Exertion on User The data obtained from the Myo Device used in Table 4.2 and Table 4.4 were averages of several iterations of gesture making while the graphs in Table 4.5 are single snap shots of a single iteration during the trials. Data should be obtained by the average of several iterations as it serves as a better comparison range. A plot of the finger position difference with respect to exertion was constructed with Equation 4.1 where AFD is the Average Finger position Difference, I is the index finger position, M is the middle finger position, R is the ring finger position, L is the little finger position, TF is the thumb flex, TA is the thumb axis and W is the wrist movement. This equation takes into consideration the distance between adjacent fingers, since the thumb and wrist are independent agents the position they are away from the neutral position can attribute to exertion, they are separately considered in the equation. The difference will be plotted with respect to exertion. AFD=∣(I−M )∣+∣(M−R)∣+∣(R−L)∣+∣TF∣+∣TA∣+∣W∣ AFD=∣(I−M )∣+∣(M−R)∣+∣(R−L)∣+∣TF∣+∣TA∣+∣W∣ (4.1) 50 4.2.1 High Exertion In the study conducted on exertion, a tight fist was made with the effort of maximum exertion by the subject; the value obtained in this trial was a total of 460 as demonstrated by the readings while the rest position given its value is declared to be 0. However due to the nature of the exertion, a stable reading at the highest possible exertion level could not be read, instead the highest exertion for steady state was obtained, that is the subject was able to sustain the gesture for an extended period of time. It should be noted that above the total value of 300, the subject felt uncomfortable and thus this was used to define a high exertion gesture. Table 4.2 contains all gestures from the trial that were identified as high exertion. Table 4.2: High Exertion Gestures (Activity Levels) Gesture Pod 1 Pod 2 Pod 3 Pod 4 Pod 5 Pod 6 Pod 7 Pod 8 Total ASL B 20 40 80 70 30 20 40 30 330 ASL K 30 90 90 90 70 30 30 40 470 ASL M 50 80 80 60 20 20 10 30 350 ASL W 50 50 70 80 40 30 30 40 390 Stop Gesture 40 50 100 80 40 30 30 20 390 Wave Right 40 70 100 100 50 40 30 30 460 Wave Left 60 20 40 20 20 20 80 40 300 Grab 90 80 80 80 30 20 30 50 460 Knocking 60 60 50 50 20 20 40 40 340 Arm side to side 60 60 70 40 20 20 10 20 300 High exertion gestures demonstrate three key configurations, the first of which is the fingers fully extended with large movements of the wrist, the second is a fist gesture that uses large movement of the wrist, and the third are adjacent fingers in opposite direction to one another. Other gestures that do not demonstrate these features, the exertion values are lower and thus considered more comfortable. Figure 4.4 shows dynamic gestures the exertion values vs the finger difference. Simulated object manipulation can be considered a special case, and observing non-object oriented gestures show that generally, as exertion values increase so do average finger difference. 4.2.1 High Exertion It should be noted that some of these gestures are not able to be made quickly 51 and in some cases require assistance from the other hand to aid in forming the proper shape. As shown in EMG graphs taken during the experiment these gestures also require a significant effort to maintain and sustained used of can possibly injure the user. Gestures that invoke this level should be avoided or used sparingly. Table 4.3: High Exertion Gestures Movement Range Gesture Thumb Axis Thumb Flex Index Middle Ring Little Wrist ASL B -10 10 10 10 10 10 0 ASL K -5 8 10 10 -10 -7 0 ASL M -10 0 -6 -6 -6 -10 0 ASL W -10 -10 10 10 10 -10 0 Stop Gesture -4 -7 10 10 10 10 0 Wave Right 0 0 10 10 10 10 10 Wave Left 0 0 10 10 10 10 10 Grab -10 -10 -10 -10 -10 -10 0 Knocking -10 -10 -10 -10 -10 -10 10 Arm side to side -10 -10 -10 -10 -10 -10 -10 Table 4.3: High Exertion Gestures Movement Range One observation made is that high exertion static gestures were considered to exertion the user more. As seen in the EMG graphs obtained during the experiment, the peak exertions for dynamic gestures occur in a smaller time frame and because of the variable nature of dynamic gestures there are areas where the gesture is actually in a rest posture. Another factor noticed in the experiment is the use of wrist movements may cause higher values as most muscles are used during the movement. However, these gestures are still shown to be uncomfortable as key features of these high exertion gestures are the stressful positions taken by the hand and sudden large movements in relation to the hand and wrist. 4.2.1 High Exertion 52 4.2.2 Low Exertion Table 4.4: Low Exertion Gestures (Activity Levels) Gesture Pod 1 Pod 2 Pod 3 Pod 4 Pod 5 Pod 6 Pod 7 Pod 8 Total ASL G 20 10 10 20 10 10 10 10 100 ASL H 10 20 40 20 10 10 40 10 160 ASL N 20 20 40 20 10 10 10 10 140 ASL O 10 20 40 30 10 10 10 10 140 ASL P 20 10 10 10 10 10 40 20 130 ASL S 50 20 50 20 10 10 10 10 180 ASL X 30 20 20 40 10 10 10 20 160 Fingers Closed 20 10 30 30 10 10 30 30 170 Thumbs up 20 20 30 30 20 10 30 30 190 Thumbs Down 20 10 30 30 10 10 20 20 150 Point 20 20 30 20 10 10 20 30 160 Hand point 20 20 40 30 10 10 20 20 170 Two-Finger Point 20 20 40 40 10 10 20 20 180 Lever Control 20 30 30 30 20 20 20 10 180 Wave Up 10 10 30 30 10 10 40 20 160 Lever Spin 20 40 40 30 10 10 10 10 170 Lever Pull 10 20 30 30 10 10 10 10 130 Switch 40 30 30 20 10 10 20 20 180 Pinch 20 30 30 40 20 10 10 20 180 Spread Pinch 20 20 20 20 10 10 10 20 130 Two-finger pinch 20 30 40 30 20 20 10 10 180 Index Finger Pull 20 10 10 20 20 10 30 10 130 ASL J 10 10 20 20 10 10 20 20 120 ASL Z 20 40 30 20 10 10 20 20 170 Moving Point 20 20 20 20 10 10 20 30 150 In the study the rest position demonstrated a value close to 0, and for the purposes of this study had been declared to be 0 for simplicity, however, in readings that demonstrated values of 10 to 20, the subject declared that the exertion and effort felt similar to the rest position, that is very relaxed and did not require exertion from the user. It is also noted that readings of 30 demonstrated both rest and exertion features, though it was not enough to be considered uncomfortable. 4.2.1 High Exertion Due to this, total EMG values below 200 were considered to be low exertion and 53 would be ideal for prolonged use in an interactive system. As previously noted, the relative positioning of the fingers can reflect exertion individually, hence why a value of 200 was selected. This accounts for gestures that relax certain aspects while exerting others. Several features noted in low exertion gestures show several key configurations, the first and most noticeable are gestures that closely resemble the rest position. In the case of the lever gestures, the user's hand is close to the rest gesture's shape, the fingers are in the middle movement range the thumb is neither stretched or flexed, the simulation of an object in the grasp of the user allows the hand to more closely resemble its rest position, allowing gestures with this feature to exert the user significantly less. The second feature is that the fingers should be close to each other in terms of flex; that the more fingers are in the same direction an increase in comfort can be expected. As such gestures that only require one finger to be extended are generally required less exertion from the user. The third feature is the use of the entire arm to make motions, for gestures that have stationary shapes but feature motion from the arm show little exertion on the user, however, an argument can be made that exertion will instead be placed on the shoulder, it is also a considerably larger muscle mass and is more resistant to exertion, though this is a possible future topic of study. Figure 4.4 also demonstrates that simulated object manipulations gestures, or object oriented gestures score consistently lower exertion values. It also does not follow the general trend of higher finger differences leading to higher exertion values. Further investigation and study into this observation can be considered for future work. 4.2.1 High Exertion 54 Table 4.5: Low Exertion Gesture Movement Range Gesture Thumb Axis Thumb Flex Index Middle Ring Little Wrist ASL G 0 0 0 -10 -10 -10 3 ASL H -10 10 10 10 -10 -10 4 ASL N -10 0 -6 -6 -10 -10 0 ASL O 0 0 -1 -1 -1 -1 0 ASL P 10 4 10 10 -10 -10 8 ASL S -10 -10 -10 -10 -10 -10 0 ASL X 0 -7 5 -10 -10 -10 0 Fingers Closed 0 -7 10 10 10 10 0 Thumbs up 10 10 -10 -10 -10 -10 0 Thumbs Down 10 10 -10 -10 -10 -10 0 Point -10 -10 10 -10 -10 -10 0 Hand point 10 10 10 10 10 10 0 Two-Finger Point -10 -10 10 10 -10 -10 0 Lever Control 1 1 1 1 1 1 0 Wave Up 0 0 10 10 10 10 10 Lever Spin 1 1 1 1 1 1 10 Lever Pull 1 1 1 1 1 1 0 Switch -10 -10 10 -10 -10 -10 0 Pinch 3 0 -3 -10 -10 -10 0 Spread Pinch 3 0 -3 -10 -10 -10 0 Two-finger pinch -5 -5 -5 -5 -10 -10 0 Index Finger Pull -10 -10 -10 -10 -10 -10 0 ASL J 10 10 -10 -10 -8 10 8 ASL Z -5 -10 7 -10 -10 -10 0 Moving Point -10 -10 10 -10 -10 -10 10 Table 4.5: Low Exertion Gesture Movement Range Table 4.5: Low Exertion Gesture Movement Range Some low exertion gestures share physical features with high exertion gestures, one such feature is the majority or all fingers in a fist shape, and the reason for this is the hand does not require additional movements that may cause the hand to exert more effort to maintain the shape and form. In dynamic gestures the action and movement of the hand is slower even when it reaches the range limit of their respective features. 55 Figure 4.4: Finger Difference vs Exertion (Dynamic) Figure 4.4: Finger Difference vs Exertion (Dynamic) 4.3 Summary In the study of current gesture database designs, several factors were considered, the gesture type, classified into static and dynamic gestures and then further categorized into iconic, metaphoric and deictic in order to identify and provide a frame work comparisons in later studies, Table 4.1. The exertion and movement range was then investigated through the use of a Myo armband sensory device to investigate the ergonomic design of the gestures. Through the use of these investigations, a guideline of how gestures should be made was formed. In the study of current gesture database designs, several factors were considered, the gesture type, classified into static and dynamic gestures and then further categorized into iconic, metaphoric and deictic in order to identify and provide a frame work comparisons in later studies, Table 4.1. The exertion and movement range was then investigated through the use of a Myo armband sensory device to investigate the ergonomic design of the gestures. Through the use of these investigations, a guideline of how gestures should be made was formed. The gestures were classed into iconic, metaphoric and deictic, and then further sub divided into static and dynamic. These classifications help understand the general meaning and specific application each gesture has. It was found that static gestures selected in these categories were either universal in nature or operate on conventional knowledge. Dynamic gestures relied heavily on context to convey information and due to their complex nature, 56 allowed for more abstract actions and shapes to be performed in relation to a command. Table 4.1 and Table 4.2 in Appendix 1 comprised of 66 gestures obtained from various databases, where Table 4.2 referred to the collection of ASL letters. From this categorization it can be noted that current development into gesture recognition is heading towards dynamic gestures that operate in a virtual 3 dimensional environment where the user of the system will interact with objects and tools to perform certain tasks or issue commands [56][58]. As seen in the study into exertion and finger positions, gestures that aim to interact with virtual or simple real world objects achieve lower exertion values than gestures that aim to replicate object shapes. Following the gesture classification, the exertion and movement range of each gesture was investigated through the use of an EMG sensory device, the Myo Armband. Discussions and Conclusions This thesis implemented a method of examining exertion with respect to gesture design with a goal of improving gesture interaction interfaces. The development of gesture classification aids in designing gesture on an application level while the exertion and gesture design found and identified flaws in current gesture design work. Through this study, improvements to current gesture designs have been created. 4.3 Summary The armband provided the EMG readings that correlated with the exertion the user experienced during the gesture making and maintaining process. Using this information, the relative exertion of the gesture could be investigated. In conjunction with the exertion investigation, the movement range of the gestures in reference to the fingers and wrist was also investigated. Using rest, extended and compressed postures as reference points, each gesture was examined in relation to the limits of the subject's physical characteristics. Using this information, high exertion gestures were identified and compared to their movement range analysis. It was discovered that high exertion gestures contained specific features. For static gestures, these were fingers being in vastly different orientations from each other, while in dynamic gestures these were gestures that heavily used the wrist in addition to reaching the upper physical limits of the hand. Following the investigation of high exertion gestures, investigation of low exertion gestures was carried out. For static gestures, the fingers tend to be in the same orientation as their adjacent fingers, as well as being close to the rest position in shape. For dynamic gestures, the inclusion of a simulated tool in the user hand helped to maintain the shape of the rest posture or a relaxed shape for the hand; this enabled these gestures to have rest periods in between active regions. Furthermore, simulated object manipulation gestures demonstrate much lower exertion values; this can be a consideration when designing gestures for computer interactions. 57 57 5.1 Design Guidelines From the study above observing high exertion and low exertion gestures, the following guidelines should be considered when designing gestures for recognition systems. 1. Intuitive: The gesture should be fairly universal in meaning if it does not invoke imagery, conventional gestures can be used, common gesture interactions with other technology can be considered if the use is widespread and considered easily recognizable. Such examples would be touch gestures used in touch-screen surfaces, as the technology is widespread and the cultural influence invoked by it has reached a larger audience. 2. Rest Posture: Dynamic gestures should consider movement ranges that fall in the rest area, this is the area of least exertion for the user and by placing most of the actions in 58 this region, and the exertion of the gesture on the user can be greatly reduced. Additionally, continuous gesture making should invoke rest posture as a means of starting and ending a gesture, this enables the system to better recognize the actions of the user and offers better control for the user to give meaningful commands to a recognition system. 3. Ease of Maintenance: For gestures that require a longer period of time to stay in one shape, the gesture should consider the following features: fingers should be mostly aligned in the same or similar orientation with their adjacent fingers, movements should only contain the necessary movement for a large enough physical change for the system to recognize, and only gestures that are close to rest posture should be held for an extended period of time. 3. Ease of Maintenance: For gestures that require a longer period of time to stay in one shape, the gesture should consider the following features: fingers should be mostly aligned in the same or similar orientation with their adjacent fingers, movements should only contain the necessary movement for a large enough physical change for the system to recognize, and only gestures that are close to rest posture should be held for an extended period of time. 4. Simulated object manipulation: When given context, gestures can provide more information than by verbal or other means; the use of virtual simulated object manipulation by gestures is a powerful tool that enables comfort, intuitiveness, and understanding. As demonstrated, gesture commands have a large development in gestures that simulate the operation of objects or tools in the virtual space. 5.1 Design Guidelines As observed in Figure 4.4 in Chapter 4, Simulated object manipulation Gestures show lower exertion values than most non-object oriented gestures. Due to simulated object manipulation gestures relating to concrete concepts, they are all classified as iconic gestures. One criticism for visual gesture recognition is the uncomfortable or unfamiliar employment of if in these systems; the attempt to replace more traditional input methods that are more efficient is a major factor in these criticisms. Designers and developers need to pay special attention to the designs and method of implementation, if the system serves no purpose beyond incorporating the technology into the product, then it will detract from the quality and experience. This study was made in the hope that these developers create systems that enhance the experience. As technology continues to develop, it is also important to create systems that are safe and comfortable for the users. The promotion of ergonomic gestures will help to improve upon the technology as well as create new conventions for later systems. As the technology sees more 59 widespread use, it is important to carefully design these gestures for these systems, both for the growth of the technology and future work that will be built upon it. widespread use, it is important to carefully design these gestures for these systems, both for the growth of the technology and future work that will be built upon it. 5.2 Possible Future Work This study investigates and presents gesture designs guidelines as a study of their context and effects on the user. Future work related to this study will see an expanded study into different or specific gesture types or databases. Furthermore, gesture recognition technology will further expand into other languages that the body can produce. Using this study as a basis, intelligent designs of gesture and body language for computer recognition can be carried out. Some of the following studies can be performed to further refine these guidelines. Investigation into specific target audiences, such as hearing impaired or mute persons will provide extensive materials into creating systems that can interpret language more effectively. As seen in this study, sign language gestures are not always comfortable for the user and several of these gestures are high exertion, thus extended use of these for any system will be uncomfortable and cause undo exertion on the user. Studies such as this can be applied to the gestures that will eventually be developed into systems for use for specific target audiences and users. Currently little study has been made into object oriented gestures, as observed in this thesis, simulated object manipulation demonstrated much lower exertion values than non-object oriented gestures. Study into this can yield valuable information into comfortable gesture designs for dynamic gestures. A study into databases with specific applications will further refine the guidelines for different applications. As discussed previously, context to gesture recognition is a valuable part of the information delivered by the system. Specific applications can include entertainment purposes, design software, navigation software, and control systems. At the basis of all of these applications, it can be inferred that the user may spend an extended period of time operating the system, hence the ergonomic design catered for these applications should be considered. By conducting studies such as these using subjects of expert in their fields, gesture recognition will see a much wider spread of use given that camera technology is also widespread and seeing an 60 increase in use. A later study can be conducted is which users can be given a list of commands and then asked to perform a gesture related to the command. This can add a layer of intuitiveness if a large sample of subjects were to be used in this study. 5.2 Possible Future Work By observing the user expectations, it is possible to compare and contrast with developer expectations to enhance the development and design of gestures in these systems. This is a possible extension of this thesis as this observes and analysis gestures in the developer domain. A study into user derived gestures can demonstrate more effective gestures. 5.3 Conclusion For the purposes of investigating gesture designs in computing systems for human-computer interactions, several current instructional gesture databases was obtained and investigated. With the goal of classifying and understanding gesture use in computer interactions, several aspects of gestures from these databases were investigated. This thesis then draws conclusions from the results obtained from this investigation. For this research, different aspects of gesture recognition were investigated. Current technology and trends were observed and the best approach to gesture recognition was considered based on this research. Notably, this included an examination of recognition techniques; required hardware and related body language interpreters such as gaze and touch gestures. With the conclusion that camera based systems will reach the largest number of audiences given its wide distribution, gesture types were investigated and examined. Through this examination, and understanding of the relation between computer interfaces and gestures can be observed. In order to effectively accomplish this, some research was made in how gestures and languages are related, terms, ideas and metrics from this section will be used to examine and observe gestures. Finally, the design of gestures was investigated, from here, different aspects of gestures were examined to help reconcile the terms required to understand how body language and gestures convey information to computing systems. In this study of gestures in current recognition systems, several public databases were 61 obtained and analyzed. In this analysis, the database gestures were extracted and categorized. Gestures that were similar in shape and function to ones already categorized were discarded. From this analysis, the gestures were separated into categorizations for further analysis. It was discovered that static gestures tend to use universal signs and shapes while dynamic gestures offer a greater range of information. With the gestures organized into categories, the exertion and movement range was analyzed using EMG sensor equipment and photo analysis. Through this analysis, exertion and finger locations were compared and contrasted. It was found that high exertion gestures involved sudden motions from one physical limit to another as well as opposite positioning to adjacent features. For low exertion gestures, the motions involved were more relaxed as well as finger positions in orientations close to the rest position. Furthermore, low exertion gestures have long periods of rest with only occasional exertion functions, however, it should also be noted that these exertion functions are still relatively low compared to readings seen in high exertion gestures. 5.3 Conclusion Findings and results of this study are similar to those found in [21] when comparing the analysis of the ASL gestures. With interest in gesture recognition increasing, development into recognition techniques is ever rising. All aspects of gesture recognition are being explored to determine its scope and potential, however much emphasis has been placed on the technical facets leaving the interaction aspect unexplored. However advance the technology it is still ultimately up to the end user of the system that determines much of that potential. Without considering the needs and comfort of the user, the technology’s potential will be held back. As with all technology, it is the users that control its future. 62 Compiled Gesture Database Gestures taken from multiple databases were compiled into this list, an image of the gesture and its description was provided to keep consistency during testing. Gesture Description Static: Iconic ASL American Sign Language, each symbol used in this set is a representation of a letter in the English alphabet. These are usually used in place of a keyboard to type or enter a character into an interactive system. Stop Gesture A forward facing palm that is normally associated with the command or request to stop. Description Description American Sign Language, each symbol used in this set is a representation of a letter in the English alphabet. These are usually used in place of a keyboard to type or enter a character into an interactive system. ASL ASL A forward facing palm that is normally associated with the command or request to stop. Stop Gesture Stop Gesture Stop Gesture 63 63 Okay Gesture Resembling the letters O and K, this is generally known to be related to a positive response to an inquiry. Fingers Closed The fingers are closed together simulating a flat surface or a floor. This gesture is sometimes accompanied by another gesture made by the other hand representing another object on this surface. Thumbs up The Fingers are balled up into a fist with the thumb facing up, this is associated with the concept of “Yes” in an inquiry Thumbs Down The Fingers are balled up into a fist with the thumb facing down, this is associated with the concept of “No” in an inquiry Deictic Point A single finger point into the direction of interest by the user, this is the most common deictic gesture and is used by the majority of databases that require spatial awareness as part of its interaction. Okay Gesture Okay Gesture Okay Gesture The fingers are closed together simulating a flat surface or a floor. This gesture is sometimes accompanied by another gesture made by the other hand representing another object on this surface. Fingers Closed Fingers Closed The Fingers are balled up into a fist with the thumb facing up, this is associated with the concept of “Yes” in an inquiry The Fingers are balled up into a fist with the thumb facing up, this is associated with the concept of “Yes” in an inquiry Thumbs up Thumbs up Thumbs Down Thumbs Down A single finger point into the direction of interest by the user, this is the most common deictic gesture and is used by the majority of databases that require spatial awareness as part of its interaction. Point Point 64 Hand point Using all fingers and thumb to point to specific direction, this is a highly uncommon pointing gestures but it is sometimes used when confirming four or more objects are related to the direction identified. Description Two-Finger Point Using two fingers to point in specific direction, this deictic gesture is normally interchangeable with the single point gesture; however, sometimes it is used to declare two objects in relation to the identified direction. Dynamic: Iconic Circular Right The hand is in a neutral position and makes a circular motion right at the wrist. This simulates an object rotating right. Circular Left The hand is in a neutral position and makes a circular motion left at the wrist. This simulates an object rotating left. Hand point Using all fingers and thumb to point to specific direction, this is a highly uncommon pointing gestures but it is sometimes used when confirming four or more objects are related to the direction identified. Hand point Using two fingers to point in specific direction, this deictic gesture is normally interchangeable with the single point gesture; however, sometimes it is used to declare two objects in relation to the identified direction. Two-Finger Point Two-Finger Point The hand is in a neutral position and makes a circular motion right at the wrist. This simulates an object rotating right. The hand is in a neutral position and makes a circular motion right at the wrist. This simulates an object rotating right. Circular Right Circular Right Circular Left Circular Left Circular Left 65 66 Dial Control The index and middle finger at pinched forward with a space in between representing an object between them; the hand is then moved in a twisting motion with the space in between the index and middle finger at the axis. Lever Control The fingers are curled up as if there was a bar between the palm and fingers. The entire arm is then moved up or down to simulate lever operation. Wave Right The hand is held with the fingers facing forward vertically and then the hand is moved to the right at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Left The hand is held with the fingers facing forward vertically and then the hand is moved to the left at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Up The hand is held with the fingers facing forward horizontally and then the hand is moved up at the wrist. Description A common use for this gesture is to signify the user brushing an object to the side. Dial Control The index and middle finger at pinched forward with a space in between representing an object between them; the hand is then moved in a twisting motion with the space in between the index and middle finger at the axis. Dial Control The fingers are curled up as if there was a bar between the palm and fingers. The entire arm is then moved up or down to simulate lever operation. Lever Control Lever Control The hand is held with the fingers facing forward vertically and then the hand is moved to the right at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Right Wave Right The hand is held with the fingers facing forward vertically and then the hand is moved to the left at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Left Wave Left The hand is held with the fingers facing forward horizontally and then the hand is moved up at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Up Wave Up 66 67 Wave Down The hand is held with the fingers facing forward horizontally and then the hand is moved down at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Lever Spin The fingers are curled up as if there was a bar between the palm and fingers. The hand is then rotated left or right, similar to dial control. Lever Pull The fingers are curled up as if there was a bar between the palm and fingers; the hand is then pulled towards the user in a puling motion. Switch The index finger is held forward and then either moved up to down to simulate the look of flipping a switch. Pinch The index finger and thumb move closer together until they touch. In touch interfaces these gestures is used to zoom out but in vision gesture interactions is used as an object select or grab function. Description Wave Down The hand is held with the fingers facing forward horizontally and then the hand is moved down at the wrist. A common use for this gesture is to signify the user brushing an object to the side. Wave Down Lever Spin Lever Spin Lever Pull Lever Pull Switch Switch The index finger and thumb move closer together until they touch. In touch interfaces these gestures is used to zoom out but in vision gesture interactions is used as an object select or grab function. Pinch Pinch Pinch 67 67 Spread Pinch The index finger and thumb move further apart from a touching position. In touch interfaces this gesture is often used with the zoom in feature, however, in vision gesture interaction systems; this is used to release an object under select. Grab The fingers are open and then are closed into a fist. This gesture is commonly use with the select command and used to select an object in virtual 3D space. Press down The palm is faced down and the entire hand is moved down. Used for pushing an object down or to signify the lowering of an object. Lift Up The palm is faced up and the entire hand is moved up. Used for raising an object up or to signify the raising of an object. Arm Rotate The entire arm is rotated back and forth. This gesture is commonly associated with object rotation. The index finger and thumb move further apart from a touching position. In touch interfaces this gesture is often used with the zoom in feature, however, in vision gesture interaction systems; this is used to release an object under select. Spread Pinch Spread Pinch Grab The fingers are open and then are closed into a fist. This gesture is commonly use with the select command and used to select an object in virtual 3D space. Grab Press down Press down Lift Up Lift Up Arm Rotate Arm Rotate Arm Rotate 68 Knocking A fist is made with the palm facing forward; the fist is then moved back and forth at the wrist. This gesture is used to simulate the knocking of a door. Two-finger pinch The index finger and middle finger are moved until they touch the thumb. Similar to the pinch gesture before, however this has some use as an alternate select. Description Index Finger Pull The palm is facing up with the index finger fully extended, it is then retracted. This gesture is usually associated with the “come here” command. Metaphoric Good-bye wave The palm is facing forward while the hand is moved side to side at the wrist. This can be associated with the concept of “good bye”, and ending sequence. A fist is made with the palm facing forward; the fist is then moved back and forth at the wrist. This gesture is used to simulate the knocking of a door. Knocking The index finger and middle finger are moved until they touch the thumb. Similar to the pinch gesture before, however this has some use as an alternate select. Two-finger pinch Two-finger pinch The palm is facing up with the index finger fully extended, it is then retracted. This gesture is usually associated with the “come here” command. Index Finger Pull Index Finger Pull Metaphoric Good-bye wave Good-bye wave Good-bye wave 69 69 Index-middle close The index and middle fingers are in a V-shape and then are closed together. This can be used to demonstrate two objects coming together. Right Grab The hand is tilted right and then moved into a fist centered in the middle. This was used to signify moving from the right to the center. Left Grab The hand is tilted left and then moved into a fist centered in the middle. This was used to signify moving from the left to the center. Open to close fingers The fingers are spread out then closed together, but still extended. This gesture was used to signify the ending or closing of an object. The index and middle fingers are in a V-shape and then are closed together. This can be used to demonstrate two objects coming together. Index-middle close Index-middle close The hand is tilted right and then moved into a fist centered in the middle. This was used to signify moving from the right to the center. Right Grab Right Grab Left Grab The hand is tilted left and then moved into a fist centered in the middle. This was used to signify moving from the left to the center. Left Grab Left Grab Open to close fingers Open to close fingers Open to close fingers 70 Spread Hand The fingers are closed together, extended and are then spread out. Description This gesture was used to signify the compression. Arm side to side The wrist and hand are in fixed position while the arm moves them side to side. This gesture used to signify the shaking or movement of an object. Wrist Tilt The hand is in neutral position and is tilted left and right at the wrist. This object was used to demonstrate the rotation or axial movement of an object. Deictic Tilt The hand tilted in the desired direction. This gesture is used as a command to signify when to move to the desired direction. The object or concept is instructed to stop, when the hand tilts to a direction, it is instructional for the object to move in that direction. The fingers are closed together, extended and are then spread out. This gesture was used to signify the compression. Spread Hand Spread Hand Arm side to side Arm side to side The hand is in neutral position and is tilted left and right at the wrist. This object was used to demonstrate the rotation or axial movement of an object. Wrist Tilt Wrist Tilt The hand tilted in the desired direction. This gesture is used as a command to signify when to move to the desired direction. The object or concept is instructed to stop, when the hand tilts to a direction, it is instructional for the object to move in that direction. Tilt Tilt 71 71 Fork Tilt The index and middle fingers are in a V-shape and then pointed at a desired direction. Palm With the fingers extended, the palm is pushed into the desired direction. Used as a command to push an object in the desired direction. Moving Point With the index finger extended, the entire hand is moved forward in the direction that the finger is pointing in. Table 4.1: Compiled Database Gestures Fork Tilt Fork Tilt Palm With the fingers extended, the palm is pushed into the desired direction. Used as a command to push an object in the desired direction. Palm With the index finger extended, the entire hand is moved forward in the direction that the finger is pointing in. Moving Point Table 4.1: Compiled Database Gestures Table 4.1: Compiled Database Gestures A table was made separately for the ASL lettering as they were shared similar descriptions. This was used as reference during the experiment. Description 72 ASL Letters Letter A B C D B C D 73 74 E F G H I F G H I 74 75 J K L M N K L M N 75 76 O P Q R S O P Q R S 76 T U V W X U V W X 77 Y Z Table 4.2: American Sign Language A-Z Table 4.2: American Sign Language A-Z 78 Myo Sensor Data In order to get the exertion the user experienced when making a specific gesture, two reference points were required, that is the relaxed state and the exertion state. These were used to compare how stressful the gestures were. Table 3.1: Rest and Exertion Points Reference Table As with the two reference states above, each gestures was this measured and recorded. Gesture Pod 1 Pod 2 Pod 3 Pod 4 Pod 5 Pod 6 Pod 7 Pod 8 Total Rest 0 0 0 0 0 0 0 0 0 Fist 110 70 40 80 40 20 40 60 460 Table 3.1: Rest and Exertion Points Reference Table Gesture Pod 1 Pod 2 Pod 3 Pod 4 Pod 5 Pod 6 Pod 7 Pod 8 Total Rest 0 0 0 0 0 0 0 0 0 Fist 110 70 40 80 40 20 40 60 460 Table 3.1: Rest and Exertion Points Reference Table As with the two reference states above, each gestures was this measured and record As with the two reference states above, each gestures was this measured and recorded. Table 4.3: Myo Armband sensor Readings Dynamic Table 4.3: Myo Armband sensor Readings Dynamic Myo Sensor Data 79 Table 4.1: Myo Armband sensor Readings Static (Activity Levels) Static: Pod 1 Pod 2 Pod 3 Pod 4 Pod 5 Pod 6 Pod 7 Pod 8 Total Iconic ASL A 80 40 50 40 0 5 0 10 225 ASL B 20 40 80 70 30 20 40 30 330 ASL C 80 50 50 40 10 10 10 10 260 ASL D 20 40 80 80 20 10 40 30 320 ASL E 40 40 90 70 20 20 30 20 330 ASL F 10 20 80 80 10 10 30 10 250 ASL G 20 10 10 20 10 10 10 10 100 ASL H 10 20 40 20 10 10 40 10 160 ASL I 10 50 40 30 20 10 20 20 200 ASL K 30 90 90 90 70 30 30 40 470 ASL L 20 40 40 30 10 20 20 20 200 ASL M 50 80 80 60 20 20 10 30 350 ASL N 20 20 40 20 10 10 10 10 140 ASL O 10 20 40 30 10 10 10 10 140 ASL P 20 10 10 10 10 10 40 20 130 ASL Q 40 20 20 10 10 10 110 60 280 ASL R 20 40 40 60 20 20 40 30 270 ASL S 50 20 50 20 10 10 10 10 180 ASL T 60 20 40 30 20 20 10 10 210 ASL U 30 40 50 60 30 20 30 20 280 ASL V 20 30 40 60 30 20 30 30 260 ASL W 50 50 70 80 40 30 30 40 390 ASL X 30 20 20 40 10 10 10 20 160 ASL Y 20 30 40 30 20 20 20 30 210 Stop Gesture 40 50 100 80 40 30 30 20 390 Okay Gesture 20 20 60 40 20 10 30 20 220 Fingers Closed 20 10 30 30 10 10 30 30 170 Thumbs up 20 20 30 30 20 10 30 30 190 Thumbs Down 20 10 30 30 10 10 20 20 150 Deictic Point 20 20 30 20 10 10 20 30 160 Hand point 20 20 40 30 10 10 20 20 170 Two-Finger Point 20 20 40 40 10 10 20 20 180 Table 4.1: Myo Armband sensor Readings Static (Activity Levels) 80 Gesture Pod 1 Pod 2 Pod 3 Pod 4 Pod 5 Pod 6 Pod 7 Pod 8 Rest 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Fist 1.00 0.64 0.36 0.73 0.36 0.18 0.36 0.55 Static: Iconic ASL A 0.73 0.36 0.45 0.36 0.00 0.05 0.00 0.09 ASL B 0.18 0.36 0.73 0.64 0.27 0.18 0.36 0.27 ASL C 0.73 0.45 0.45 0.36 0.09 0.09 0.09 0.09 ASL D 0.18 0.36 0.73 0.73 0.18 0.09 0.36 0.27 ASL E 0.36 0.36 0.82 0.64 0.18 0.18 0.27 0.18 ASL F 0.09 0.18 0.73 0.73 0.09 0.09 0.27 0.09 ASL G 0.18 0.09 0.09 0.18 0.09 0.09 0.09 0.09 ASL H 0.09 0.18 0.36 0.18 0.09 0.09 0.36 0.09 ASL I 0.09 0.45 0.36 0.27 0.18 0.09 0.18 0.18 ASL K 0.27 0.82 0.82 0.82 0.64 0.27 0.27 0.36 ASL L 0.18 0.36 0.36 0.27 0.09 0.18 0.18 0.18 ASL M 0.45 0.73 0.73 0.55 0.18 0.18 0.09 0.27 ASL N 0.18 0.18 0.36 0.18 0.09 0.09 0.09 0.09 ASL O 0.09 0.18 0.36 0.27 0.09 0.09 0.09 0.09 ASL P 0.18 0.09 0.09 0.09 0.09 0.09 0.36 0.18 ASL Q 0.36 0.18 0.18 0.09 0.09 0.09 1.00 0.55 ASL R 0.18 0.36 0.36 0.55 0.18 0.18 0.36 0.27 ASL S 0.45 0.18 0.45 0.18 0.09 0.09 0.09 0.09 ASL T 0.55 0.18 0.36 0.27 0.18 0.18 0.09 0.09 ASL U 0.27 0.36 0.45 0.55 0.27 0.18 0.27 0.18 ASL V 0.18 0.27 0.36 0.55 0.27 0.18 0.27 0.27 ASL W 0.45 0.45 0.64 0.73 0.36 0.27 0.27 0.36 ASL X 0.27 0.18 0.18 0.36 0.09 0.09 0.09 0.18 ASL Y 0.18 0.27 0.36 0.27 0.18 0.18 0.18 0.27 Stop Gesture 0.36 0.45 0.91 0.73 0.36 0.27 0.27 0.18 Okay Gesture 0.18 0.18 0.55 0.36 0.18 0.09 0.27 0.18 Fingers Closed 0.18 0.09 0.27 0.27 0.09 0.09 0.27 0.27 Thumbs up 0.18 0.18 0.27 0.27 0.18 0.09 0.27 0.27 Thumbs Down 0.18 0.09 0.27 0.27 0.09 0.09 0.18 0.18 Deictic Point 0.18 0.18 0.27 0.18 0.09 0.09 0.18 0.27 Hand point 0.18 0.18 0.36 0.27 0.09 0.09 0.18 0.18 Two-Finger Point 0.18 0.18 0.36 0.36 0.09 0.09 0.18 0.18 Table 4.2: Static Gesture Myo Armband Readings Normalized 81 Dynamic: Pod 1 Pod 2 Pod 3 Pod 4 Pod 5 Pod 6 Pod 7 Pod 8 Total Iconic Circular Right 40 60 50 40 20 20 30 30 290 Circular Left 30 50 50 30 20 10 30 30 250 Dial Control 20 40 40 30 20 20 20 20 210 Lever Control 20 30 30 30 20 20 20 10 180 Wave Right 40 70 100 100 50 40 30 30 460 Wave Left 60 20 40 20 20 20 80 40 300 Wave Up 10 10 30 30 10 10 40 20 160 Wave Down 50 30 30 20 10 10 50 40 240 Lever Spin 20 40 40 30 10 10 10 10 170 Lever Pull 10 20 30 30 10 10 10 10 130 Switch 40 30 30 20 10 10 20 20 180 Pinch 20 30 30 40 20 10 10 20 180 Spread Pinch 20 20 20 20 10 10 10 20 130 Grab 90 80 80 80 30 20 30 50 460 Press down 40 40 40 30 20 10 40 30 250 Lift Up 40 30 50 40 20 10 30 30 250 Arm Rotate 20 50 30 30 20 20 10 10 190 Knocking 60 60 50 50 20 20 40 40 340 Two-finger pinch 20 30 40 30 20 20 10 10 180 Index Finger Pull 20 10 10 20 20 10 30 10 130 ASL J 10 10 20 20 10 10 20 20 120 ASL Z 20 40 30 20 10 10 20 20 170 Metaphoric Good-bye wave 20 40 40 40 40 20 20 20 240 Index-middle close 20 20 40 40 20 20 30 30 220 Right Grab 40 40 50 40 20 20 20 30 260 Left Grab 30 40 40 40 20 10 40 30 250 Open to close fingers 40 30 70 50 20 20 30 30 290 Spread Hand 70 40 70 50 30 20 20 30 330 Arm side to side 60 60 70 40 20 20 10 20 300 Wrist Tilt 30 40 30 40 20 20 20 10 210 Deictic Tilt 30 20 20 30 20 20 20 40 200 Fork Tilt 30 20 40 40 30 20 30 30 240 Palm 20 30 40 40 30 20 20 10 210 Moving Point 20 20 20 20 10 10 20 30 150 25 22.5 30 32.5 22.5 17.5 22.5 27.5 200 Table 4 3: Myo Armband sensor Readings Dynamic 82 Dynamic: Iconic Circular Right 0.36 0.55 0.45 0.36 0.18 0.18 0.27 0.27 Circular Left 0.27 0.45 0.45 0.27 0.18 0.09 0.27 0.27 Dial Control 0.18 0.36 0.36 0.27 0.18 0.18 0.18 0.18 Lever Control 0.18 0.27 0.27 0.27 0.18 0.18 0.18 0.09 Wave Right 0.36 0.64 0.91 0.91 0.45 0.36 0.27 0.27 Wave Left 0.55 0.18 0.36 0.18 0.18 0.18 0.73 0.36 Wave Up 0.09 0.09 0.27 0.27 0.09 0.09 0.36 0.18 Wave Down 0.45 0.27 0.27 0.18 0.09 0.09 0.45 0.36 Lever Spin 0.18 0.36 0.36 0.27 0.09 0.09 0.09 0.09 Lever Pull 0.09 0.18 0.27 0.27 0.09 0.09 0.09 0.09 Switch 0.36 0.27 0.27 0.18 0.09 0.09 0.18 0.18 Pinch 0.18 0.27 0.27 0.36 0.18 0.09 0.09 0.18 Spread Pinch 0.18 0.18 0.18 0.18 0.09 0.09 0.09 0.18 Grab 0.82 0.73 0.73 0.73 0.27 0.18 0.27 0.45 Press down 0.36 0.36 0.36 0.27 0.18 0.09 0.36 0.27 Lift Up 0.36 0.27 0.45 0.36 0.18 0.09 0.27 0.27 Arm Rotate 0.18 0.45 0.27 0.27 0.18 0.18 0.09 0.09 Knocking 0.55 0.55 0.45 0.45 0.18 0.18 0.36 0.36 Two-finger pinch 0.18 0.27 0.36 0.27 0.18 0.18 0.09 0.09 Index Finger Pull 0.18 0.09 0.09 0.18 0.18 0.09 0.27 0.09 ASL J 0.09 0.09 0.18 0.18 0.09 0.09 0.18 0.18 ASL Z 0.18 0.36 0.27 0.18 0.09 0.09 0.18 0.18 Metaphoric Good-bye wave 0.18 0.36 0.36 0.36 0.36 0.18 0.18 0.18 Index-middle close 0.18 0.18 0.36 0.36 0.18 0.18 0.27 0.27 Right Grab 0.36 0.36 0.45 0.36 0.18 0.18 0.18 0.27 Left Grab 0.27 0.36 0.36 0.36 0.18 0.09 0.36 0.27 Open to close fingers 0.36 0.27 0.64 0.45 0.18 0.18 0.27 0.27 Spread Hand 0.64 0.36 0.64 0.45 0.27 0.18 0.18 0.27 Arm side to side 0.55 0.55 0.64 0.36 0.18 0.18 0.09 0.18 Wrist Tilt 0.27 0.36 0.27 0.36 0.18 0.18 0.18 0.09 Deictic Tilt 0.27 0.18 0.18 0.27 0.18 0.18 0.18 0.36 Fork Tilt 0.27 0.18 0.36 0.36 0.27 0.18 0.27 0.27 Palm 0.18 0.27 0.36 0.36 0.27 0.18 0.18 0.09 Moving Point 0.18 0.18 0.18 0.18 0.09 0.09 0.18 0.27 Table 4.4: Dynamic Gesture Myo Armband Readings Normalized 83 A program was used to graphically plot the data for later review. Myo Sensor Data A program was used to graphically plot the data for later review. A program was used to graphically plot the data for later review. Myo Sensor Readings Description Clenched Fist Moving to Rest ASL A ASL B Myo Sensor Readings Description Clenched Fist Moving to Rest ASL A ASL B Moving to Rest 84 ASL C ASL D ASL E ASL F ASL G ASL H ASL G ASL H 85 86 ASL I ASL K ASL L ASL M ASL N ASL O ASL L ASL M 86 87 ASL P ASL Q ASL R ASL S ASL T ASL U ASL T ASL U 87 87 ASL V ASL W ASL X ASL Y Stop (2 iterations) Stop (2 iterations) 88 Okay Fingers Closed Thumbs Up Thumbs Down Point Hand Point Thumbs Down Hand Point 89 Two-Finger Point Circular Left Circular Right Dial Control Lever Control Two-Finger Point Circular Left Circular Right Circular Right Lever Control Lever Control 90 Wave Left Wave Right Wave Up Wave Down Lever Spin Lever Pull Lever Pull 91 91 Switch Pinch Spread Pinch Grab Press Down Lift Up Arm Rotate Knocking Two-Finger Pinch Grab Press Down Lift Up Arm Rotate Knocking Two-Finger Pinch Arm Rotate Knocking Two-Finger Pinch 92 Index Finger Pull ASL J ASL Z Good-Bye Wave Index-Middle close Left Grab Good-Bye Wave Index-Middle close Left Grab 93 94 Right Grab Open to close finger Spread Hand Arm Side to Side Wrist Tilt Tilt Fork Tilt Palm Moving Point Table 4.5: Myo Armband EMG Plots Right Grab Open to close finger Spread Hand Arm Side to Side Wrist Tilt Tilt Fork Tilt Palm Moving Point Table 4 5: Myo Armband EMG Plots Tilt Tilt Fork Tilt Palm Moving Point Table 4.5: Myo Armband EMG Plots 94 Appendix 3 Gesture Movement Rating The hand was measured using the upper most limits and the lowest the fingers can compress as reference points. Once that was recorded, each finger was measured in reference to the limits. This was done because the limits and possible degree can vary from person to person, by using the subject's limits; it is possible to create comparisons with respect to exertion. Appendix 3 Gesture Movement Rating Gesture Thumb Axis Thumb Flex Index Middle Ring Little Wrist Rest 0 0 0 0 0 0 0 Fist -10 -10 -10 -10 -10 -10 0 Extended 10 10 10 10 10 10 0 Table 3.2: Rest, flex and clenched reference table Table 3.2: Rest, flex and clenched reference table Table 3.2: Rest, flex and clenched reference table 95 Table 4.6: Movement Range Table Static With the reference points established the process of measuring and recording the positioni Static: Thumb Axis Thumb Flex Index Middle Ring Little Wrist Iconic ASL A -8 0 -8 -8 -8 -8 0 ASL B -10 10 10 10 10 10 0 ASL C 0 0 1 2 1 1 0 ASL D 0 0 10 -3 -4 -5 0 ASL E 0 5 7 7 7 7 0 ASL F 0 4 -1 8 8 8 0 ASL G 0 0 0 -10 -10 -10 3 ASL H -10 10 10 10 -10 -10 4 ASL I -10 10 -10 -10 -8 10 0 ASL K -5 8 10 10 -10 -7 0 ASL L -10 10 10 -10 -10 -10 0 ASL M -10 0 -6 -6 -6 -10 0 ASL N -10 0 -6 -6 -10 -10 0 ASL O 0 0 -1 -1 -1 -1 0 ASL P 10 4 10 10 -10 -10 8 ASL Q 0 5 7 -10 -10 -10 10 ASL R -10 -10 10 10 -10 -10 0 ASL S -10 -10 -10 -10 -10 -10 0 ASL T -10 0 6 -10 -10 -10 0 ASL U -4 0 10 10 -10 -10 0 ASL V -4 0 10 10 -10 -10 0 ASL W -10 -10 10 10 10 -10 0 ASL X 0 -7 5 -10 -10 -10 0 ASL Y 10 10 -10 -10 -10 10 0 Stop Gesture -4 -7 10 10 10 10 0 Okay Gesture 0 0 -2 10 10 10 0 Fingers Closed 0 -7 10 10 10 10 0 Thumbs up 10 10 -10 -10 -10 -10 0 Thumbs Down 10 10 -10 -10 -10 -10 0 Deictic Point -10 -10 10 -10 -10 -10 0 Hand point 10 10 10 10 10 10 0 Two-Finger Point -10 -10 10 10 -10 -10 0 Table 4.6: Movement Range Table Static With the reference points established, the process of measuring and recording the positioning with respect to the reference points. Table 4.7: Movement Range Table Dynamic Table 4.7: Movement Range Table Dynamic Appendix 3 Gesture Movement Rating 96 Table 4.7: Movement Range Table Dynamic Dynamic: Thumb Axis Thumb Flex Index Middle Ring Little Wrist Iconic Circular Right 0 -7 10 10 10 10 10 Circular Left 0 -7 10 10 10 10 10 Dial Control 3 0 -3 -10 -10 -10 10 Lever Control 1 1 1 1 1 1 0 Wave Right 0 0 10 10 10 10 10 Wave Left 0 0 10 10 10 10 10 Wave Up 0 0 10 10 10 10 10 Wave Down 0 0 10 10 10 10 10 Lever Spin 1 1 1 1 1 1 10 Lever Pull 1 1 1 1 1 1 0 Switch -10 -10 10 -10 -10 -10 0 Pinch 3 0 -3 -10 -10 -10 0 Spread Pinch 3 0 -3 -10 -10 -10 0 Grab -10 -10 -10 -10 -10 -10 0 Press down 0 -7 10 10 10 10 0 Lift Up 0 -7 10 10 10 10 0 Arm Rotate -10 -10 -10 -10 -10 -10 10 Knocking -10 -10 -10 -10 -10 -10 10 Two-finger pinch -5 -5 -5 -5 -10 -10 0 Index Finger Pull -10 -10 -10 -10 -10 -10 0 ASL J 10 10 -10 -10 -8 10 8 ASL Z -5 -10 7 -10 -10 -10 0 Metaphoric Good-bye wave 10 10 10 10 10 10 10 Index-middle close -10 -10 10 10 -10 -10 0 Right Grab -10 -10 -10 -10 -10 -10 8 Left Grab -10 -10 -10 -10 -10 -10 8 Open to close fingers 10 10 10 10 10 10 0 Spread Hand 10 10 10 10 10 10 0 Arm side to side -10 -10 -10 -10 -10 -10 -10 Wrist Tilt 0 0 0 0 0 0 10 Deictic Tilt 0 0 5 5 5 5 10 Fork Tilt -10 -10 10 10 -10 -10 10 Palm 0 0 10 10 10 10 10 Moving Point -10 -10 10 -10 -10 -10 10 97 References [1]H. Duan, Q. Zhang and W. Ma, 'An approach to dynamic hand gesture modeling and real- time extraction', in 2011 IEEE 3rd International Conference on Communication Software and Networks (ICCSN), Xi'an, 2011, pp. 139-142. [2]J. Gallud, P. Villanueva, R. Tesoriero, G. Sabastian, S. Molina and A. Navarret, 'Gesture- Based Interaction: Concept Map and Application Scenarios', in 2010 Third International Conference on Advances in Human-Oriented and Personalized Mechanisms, Technologies and Services (CENTRIC), Nice, 2010, pp. 28-33. [3]L. Gupta and S. Ma, 'Gesture-based interaction and communication: automated classification of hand gesture contours', inSystems, Man, and Cybernetics, Part C: Applications and Reviews, IEEE Transactions on (Volume:31 , Issue: 1 ), IEEE, 2001, pp. 114-120. [4]M. Malgireddy, J. Corso, V. Govindaraju and D. Mandalapu, 'A Framework for Hand Gesture Recognition and Spotting Using Sub-gesture Modeling', in 2010 20th International Conference on Pattern Recognition (ICPR), Istanbul, 2010, pp. 3780-3783. [5]M. Panwar, 'Hand gesture recognition based on shape parameters', in 2012 International Conference on Computing, Communication and Applications (ICCCA), Dindigul, Tamilnadu, 2012, pp. 1-6. [6]H. Tauseef, M. Fahiem and S. Forhan, 'Browse Conference Publications > Visualisation, 2009. VIZ '09. ... Help Working with Abstracts Back to Results Recognition and Translation of Hand Gestures to Urdu Alphabets Using a Geometrical Classification', in Visualisation, 2009. VIZ '09. Second International Conference, Barcelona, 2009, pp. 213-217. [7]S. Jambhale and A. Khaparde, 'Gesture recognition using DTW & piecewise DTW', in Electronics and Communication Systems (ICECS), 2014 International Conference, Coimbatore, 2014, pp. 1-5. [8]M. Neggazi, M. Bengherabi, Z. Boulkenafet and A. Amira, 'An efficient FPGA implementation of Gaussian mixture models based classifier: Application to face recognition', in 2013 8th International Workshop on Systems, Signal Processing and their Applications (WoSSPA), Algiers, 2013, pp. 367 - 371. 98 [9]F. Jiang, Y. Gao, S. Wu, D. Zhao and C. Wang, 'Discriminating features learning in hand gesture classification', IET Computer Vision, vol. 9, no. 5, pp. 673-680, 2015. [10]F. Erden and A. Çetin, 'Hand gesture based remote control system using infrared sensors and a camera', IEEE Transactions on Consumer Electronics, vol. 60, no. 4, pp. 675-680, 2014. [11]M. Kaâniche and F. Brémond, 'Recognizing Gestures by Learning Local Motion Signatures of HOG Descriptors', IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 34, no. 11, pp. 2247-2258, 2012. [12] J. Gallud, P. Villanueva, R. Tesoriero, G. Sebastian, S. Molina and A. [21]D. Rempel, M. Camilleri and D. Lee, "The design of hand gestures for human–computer References Navarrete, "Gesture- Based Interaction: Concept Map and Application Scenarios", in Advances in Human-Oriented and Personalized Mechanisms, Technologies and Services (CENTRIC), 2010 Third International Conference, Nice, 2010, pp. 28-33. [13] J. Lothian Murray, R. Linden and D. Kendall, Sociology in our times. Toronto: Nelson Education, 2012. [14] J. Oates and A. Grayson, Cognitive and language development in children. Milton Keynes, U.K.: Open University, 2004. [15]M. Roetting and O. Zukunft, "Don't touch that tablet: An evaluation of gaze-based interfaces for tablet computers", in 2014 6th International Conference on Mobile Computing, Applications and Services (MobiCASE), TX, 2014, pp. 49 – 56. [16]V. Reddy, V. Raghuveer, J. Krishna and K. Chandralohit, "Finger gesture based tablet interface", in2012 IEEE International Conference on Computational Intelligence & Computing Research (ICCIC), Coimbatore, 2012, pp. 1 – 4. [17]N. Conci, P. Ceresato and F. De Natale, "Natural Human-Machine Interface using an Interactive Virtual Blackboard", in IEEE International Conference on Image Processing, 2007 ICIP 2007, San Antonio, TX, 2007, pp. V - 181 - V – 184. [18]K. Ouwehand, T. van Gog and F. Paas, "Effects of Gestures on Older Adults' Learning from Video-based Models", Applied Cognitive Psychology, vol. 29, no. 1, pp. 115-128, 2014. [19]K. Liebal, T. Behne, M. Carpenter and M. Tomasello, "Infants use shared experience to interpret pointing gestures", Developmental Science, vol. 12, no. 2, pp. 264-271, 2009. [20]Y. Shen, S. Ong and A. Nee, "Vision-Based Hand Interaction in Augmented Reality Environment",International Journal of Human-Computer Interaction, vol. 27, no. 6, pp. 523- 544, 2011. [21]D. Rempel, M. Camilleri and D. Lee, "The design of hand gestures for human–computer 99 interaction: Lessons from sign language interpreters", International Journal of Human- Computer Studies, vol. 72, no. 10-11, pp. 728-735, 2014. interaction: Lessons from sign language interpreters", International Journal of Human- Computer Studies, vol. 72, no. 10-11, pp. 728-735, 2014. [22]J. Lambrecht, H. Walzel and J. Kruger, "Robust finger gesture recognition on handheld devices for spatial programming of industrial robots", in RO-MAN, 2013 IEEE, Gyeongju, 2013, pp. 99 – 106. [23]X. Meng, C. Cheung, K. Ho, K. Lui, E. Lam and V. Tam, "Building smart cameras on mobile tablets for hand gesture recognition", in 2012 Sixth International Conference on Distributed Smart Cameras (ICDSC), Hong Kong, 2012, pp. 1 – 5. [24]W. Hurst, R. Poppe and J. References Angeren, "Drawing outside the lines: Tracking-based gesture interaction in mobile augmented entertainment", in 2015 7th International Conference on Intelligent Technologies for Interactive Entertainment (INTETAIN), Turin, 2015, pp. 79 – 87. [25]P. Treffner, M. Peter and M. Kleidon, "Gestures and Phases: The Dynamics of Speech-Hand Communication", Ecological Psychology, vol. 20, no. 1, pp. 32-64, 2008. [26]J. Vauclair and J. Imbault, "Relationship between manual preferences for object manipulation and pointing gestures in infants and toddlers", Developmental Science, vol. 12, no. 6, pp. 1060-1069, 2009. [27]J. Graham and S. Heywood, "The effects of elimination of hand gestures and of verbal codability on speech performance", European Journal of Social Psychology, vol. 5, no. 2, pp. 189-195, 1975. [28]Y. Lin, H. Chen, Y. Hsieh, M. Hu and W. Cheng, "Interactive human action search using body language", in Wireless and Optical Communications Conference (WOCC), 2012 21st Annual, Kaohsiung, 2012, pp. 30 - 31. [29]Z. Ren, J. Meng and J. Yuan, "Depth camera based hand gesture recognition and its applications in Human-Computer-Interaction", in 2011 8th International Conference on Information, Communications and Signal Processing (ICICS), Singapore, 2011, pp. 1 - 5. [30]V. Bhame, R. Sreemathy and H. Dhumal, "Vision based hand gesture recognition using eccentric approach for human computer interaction", in 2014 International Conference on Advances in Computing, Communications and Informatics (ICACCI), New Delhi, 2014, pp. 949 - 953. [31]J. Jong, J. Hong and C. Yen, "Persistence temperament associated with children playing math games between touch panel and embodied interaction", Journal of Computer Assisted Learning, vol. 29, no. 6, pp. 569-578, 2013. [32]A. Pereira, "Computer input devices: Design for well-being and productivity", Ph.D, 100 University of California, Berkeley, US, 2015. University of California, Berkeley, US, 2015. [33]S. Billinghurst and K. Vu, "Touch screen gestures for web browsing tasks", Computers in Human Behavior, vol. 53, pp. 71-81, 2015. [34]H. Ogata, Y. Sakamoto, T. Takei and A. Torige, "Hand Shape Discrimination in Touch Interface toward Personal Authentication on a Tablet Computer", in 2013 International Conference on Biometrics and Kansei Engineering, Tokyo, 2013, pp. 257 – 260. [35]H. Elluech, A. Wali and A. Alimi, "Smart Tablet Monitoring by a Real-Time Head Movement and Eye Gestures Recognition System", in 2014 International Conference on Future Internet of Things and Cloud, Barcelona, 2014, pp. 393 – 398. [36]K. Deshmane and S. Ghodake, "Android Software Based Multi-touch Gestures Recognition for Secure Biometric Modality", in 2015 International Conference on Information Technology (ICIT), Bhubaneswar, 2015, pp. References 107 - 112. [37]N. Aziz, "Children’s Interaction with Tablet Applications: Gestures and Interface Design", in International Journal of Computer and Information Technology, Loughborough, 2013, pp. 447-450. [38]J. Liang, Y. Han and Q. Hu, "Semi-supervised image clustering with multi-modal information", Multimedia Systems, vol. 22, no. 2, pp. 149-160, 2014. [39]D. Ghosh and S. Ari, "On an algorithm for Vision-based hand gesture recognition", Signal, Image and Video Processing, vol. 10, no. 4, pp. 655-662, 2015. [40]R. Ahmed, M. El Sayed, S. Gadsden, J. Tjong and S. Habibi, "Artificial neural network training utilizing the smooth variable structure filter estimation strategy", Neural Computer & Applications, vol. 27, no. 3, pp. 537-548, 2015.x [41]Y. Zhou, G. Jiang and Y. Lin, "A novel finger and hand pose estimation technique for real- time hand gesture recognition", Pattern Recognition, vol. 49, pp. 102-114, 2016. [42]Y. Fuh and H. Ho, "Highly flexible self-powered sensors based on printed circuit board technology for human motion detection and gesture recognition", Nanotechnology, vol. 27, no. 9, p. 095401, 2016. [43]F. Mod Ma'asum, S. Sulaiman and A. Saparon, "An Overview of Hand Gestures Recognition System Techniques", IOP Conf. Ser.: Mater. Sci. Eng., vol. 99, p. 012012, 2015. [44]E. Saykol, H. Türe, A. Sirvanci and M. Turan, "Posture labeling based gesture classification for Turkish sign language using depth values", Kybernetes, vol. 45, no. 4, pp. 604-621, 2016. 101 [45]S. Smithies, K. Novins and J. Arvo, "Equation entry and editing via handwriting and gesture recognition", Behaviour & Information Technology, vol. 20, no. 1, pp. 53-67, 2001. [46]M. Sun, X. Ren, H. Tu and F. Tian, "An Investigation Into the Relationship Between Texture and Human Performance in Steering and Gesture Input Tasks", International Journal of Human- Computer Interaction, vol. 30, no. 8, pp. 654-662, 2014. [47]D. Rozado, T. Moreno, J. San Agustin, F. Rodriguez and P. Varona, "Controlling a Smartphone Using Gaze Gestures as the Input Mechanismx", Human–Computer Interaction, vol. 30, no. 1, pp. 34-63, 2014. [48]H. Stern, J. Wachs and Y. Edan, "Optimal Hand Gesture Vocabulary Design Using Psycho- Physiological and Technical Factors", 7th International Conference on Automatic Face and Gesture Recognition (FGR06). [49]H. Stern, J. Wachs and Y. Edan, "Hand gesture vocabulary design: a multicriteria optimization",2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583). [50]P. Cash and A. Maier, "Prototyping with your hands: the many roles of gesture in the communication of design concepts", Journal of Engineering Design, vol. 27, no. [56]I. Hwang, H. Kim, J. Cha, C. Ahn, K. Kim and J. Park, "A gesture based TV control interface for visually impaired: Initial design and user study", 2015 21st Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV), 2015. References 1-3, pp. 118- 145, 2016. [51]L. Richland, "Linking Gestures: Cross-Cultural Variation During Instructional Analogies", Cognition and Instruction, vol. 33, no. 4, pp. 295-321, 2015. [52]L. Lanyon and M. Rose, "Do the hands have it? The facilitation effects of arm and hand gesture on word retrieval in aphasia", Aphasiology, vol. 23, no. 7-8, pp. 809-822, 2009. [53]K. Zhong, J. Kang, S. Qin and H. Wang, "Rapid 3D conceptual design based on hand gesture", 2011 3rd International Conference on Advanced Computer Control, 2011. [54]S. Bodiroža, H. Stern and Y. Edan, "Dynamic gesture vocabulary design for intuitive human- robot dialog", Proceedings of the seventh annual ACM/IEEE international conference on Human-Robot Interaction - HRI '12, 2012. [55]M. van Beurdena, W. IJsselsteijna and K. Hopfb, "User centered design of gesture-based interaction technology", 2011 3DTV Conference: The True Vision - Capture, Transmission and Display of 3D Video (3DTV-CON), 2011. [56]I. Hwang, H. Kim, J. Cha, C. Ahn, K. Kim and J. Park, "A gesture based TV control interface for visually impaired: Initial design and user study", 2015 21st Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV), 2015. 102 [57]O. Luzanin and M. Plancak, "Hand gesture recognition using low-budget data glove and cluster-trained probabilistic neural network", Assembly Automation, vol. 34, no. 1, pp. 94-105, 2014. [58]W. Zhu, L. Liu, S. Yin, S. Hu, E. Tang and S. Wei, "Motion-sensor fusion-based gesture recognition and its VLSI architecture design for mobile devices", International Journal of Electronics, vol. 101, no. 5, pp. 621-635, 2013. [59]Y. Zhiqi, "Gesture learning and recognition based on the Chebyshev polynomial neural network", 2016 IEEE Information Technology, Networking, Electronic and Automation Control Conference, 2016. [60]R. Agrawal and N. Gupta, "Real Time Hand Gesture Recognition for Human Computer Interaction", 2016 IEEE 6th International Conference on Advanced Computing (IACC), 2016. [61]"Imperial Computer Vision & Learning Lab (ICVL)", Iis.ee.ic.ac.uk, 2017. [Online]. Available: http://www.iis.ee.ic.ac.uk/ComputerVision/. [Accessed: 20 - Oct- 2016]. [62]Y. Wei, "Yichen Wei at Microsoft Research", Microsoft Research, 2017. [Online]. Available: http://research.microsoft.com/en-us/um/people/yichenw/handtracking/. [Accessed: 20 - Oct- 2016]. [63]L. Liu, "SKIG", Lshao.staff.shef.ac.uk, 2017. [Online]. Available: http://lshao.staff.shef.ac.uk/data/SheffieldKinectGesture.htm. [Accessed: 20 - Oct- 2016]. [64]I. Bullock, T. Feix and A. Dollar, "The Yale Human Grasping Data Set", Eng.yale.edu, 2017. [Online]. Available: http://www.eng.yale.edu/grablab/humangrasping/. [Accessed: 20 - Oct- 2016]. [65]P. Rogers, J. Rogers, and Y. Rogers. Interaction Design: Beyond Human-computer Interaction. New York, NY: J. Wiley & Sons, 2002. Print. [66]"Hand gesture database", Gti.ssr.upm.es, 2017. [Online]. Available: https://www.gti.ssr.upm.es/data/HandGesture_database.html. [Accessed: 07- Apr- 2017]. [67]M. References Kawulok, "Hand pose estimation", Database for hand gesture recognition, 2017. [Online]. Available: http://sun.aei.polsl.pl/~mkawulok/gestures/. [Accessed: 08- Apr- 2017]. [68"Welcome to Myo Support", Myo Support, 2017. [Online]. Available: https://support.getmyo.com/hc/en-us. [Accessed: 01- Mar- 2017]. [69]D. Cunningham and G. Romanes, Cunningham's textbook of anatomy, 1st ed. London: Oxford University Press, 1966. 103
https://openalex.org/W4393382805	https://www.researchsquare.com/article/rs-4147568/latest.pdf	English	null	Relationship between risk of locomotive syndrome and low back pain in Japanese postpartum women: a cross-sectional study	Research Square (Research Square)	2,024	cc-by	5,476	Relationship between risk of locomotive syndrome and low back pain in Japanese postpartum women: a cross-sectional study Yuu Kajiwara Kio University Daisuke Matsumoto Kio University Tomoe Inoue-Hirakawa Nagoya University Graduate School of Medicine Rika Kawabe Kyoto University Momoko Nagai-Tanima Kyoto University Tomoki Aoyama Kyoto University DOI: https://doi.org/10.21203/rs.3.rs-4147568/v1 DOI: https://doi.org/10.21203/rs.3.rs-4147568/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Additional Declarations: No competing interests reported. Additional Declarations: No competing interests reported. Page 1/15 Results Overall, 45 women (52.3%) had a high risk of locomotive syndrome. The high-risk locomotive syndrome group had significantly higher Oswestry Disability Index [10 (0–26)] than the non-locomotive syndrome group [4 (0–24)] (p < 0.001). However, no significant difference was observed between the two groups in terms of age, number of births, or proportion of women with low physical activity levels. Methods In this cross-sectional study, we included 86 women (30.0 ± 4.2 years) within 1 year postpartum. We assessed the locomotive syndrome risk using the stand-up test and 2-step test, physical activity using the International Physical Activity Questionnaire Short Form, and low back pain using the Oswestry Disability Index. The Oswestry Disability Index score and physical activity levels were compared between groups with and without the risk of locomotive syndrome. Conclusions The findings suggest that more than half of the women within 1 year after childbirth were at risk of locomotive syndrome, suggesting a relationship between locomotive syndrome and low back pain. Prevention of postnatal low back pain may necessitate addressing decline in prenatal motor function. The findings underscore the importance of early identification and intervention for locomotive syndrome risk in postpartum women to prevent future low back pain and improve mobility. Background Pregnancy and childbirth lead to various physiological and anatomical changes, often resulting in low back pain and decline in physical activity in postpartum women. Locomotive syndrome is reportedly associated with low back pain and physical activity levels. However, the prevalence of locomotive syndrome and related factors in postpartum women have not been thoroughly investigated. Therefore, this study aimed to investigate the relationship between locomotive syndrome risk and low back pain and physical activity in postpartum women. Background Pregnancy and childbirth can result in various physiological and anatomical changes, which may persist even after childbirth, thereby causing many physical symptoms such as low back pain (LBP) and muscle weakness [1–4]. LBP is a common symptom in prenatal and postpartum women, affecting approximately 50–80% of women during pregnancy [2–4]. While the symptoms may improve after Page 2/15 childbirth in many women with LBP, a notable proportion (23%) of women with LBP during pregnancy remain symptomatic after childbirth, and 10% can develop LBP a decade later [5, 6]. LBP persists after childbirth and impairs activities of daily living (ADL) and quality of life [7]. Therefore, postpartum women should understand and address the condition of locomotive organs. childbirth in many women with LBP, a notable proportion (23%) of women with LBP during pregnancy remain symptomatic after childbirth, and 10% can develop LBP a decade later [5, 6]. LBP persists after childbirth and impairs activities of daily living (ADL) and quality of life [7]. Therefore, postpartum women should understand and address the condition of locomotive organs. Exercise and physical activity (PA) are recommended for pregnant and postpartum women to reduce the risk of many pregnancy complications and promote overall health [8]. For example, the American College of Obstetrics and Gynecology (ACOG) recommends that women experiencing a healthy pregnancy should regularly engage in PA of moderate intensity for at least 20 to 30 min per day on most or all days of the week during pregnancy and the postpartum period [8]. Despite these recommendations, PA levels decrease in pregnant women [9, 10], with only 3% of women reaching the level of PA recommended by ACOG [11]. After childbirth, PA returns to the level undertaken in the second trimester; however, although light PA may increase, moderate- and high-intensity PA decrease. Postpartum women undertake less PA and are more sedentary compared to non-postpartum counterparts [1, 9, 10, 12]. Locomotive syndrome (LS) is a condition of reduced mobility due to impairment of locomotive organs, a concept proposed by the Japanese Orthopaedic Association (JOA) [13, 14]. LS is diagnosed using the following three modalities: the stand-up test, 2-step test, and 25-question geriatric locomotive function scale (GLFS-25). LS is reportedly associated with LBP and PA [15, 16]. A previous study reported that the prevalence of LS in young women was approximately 25% and that this prevalence increases concurrently with age [16]. Background In another study on nulliparous women in their 20s, only 43.6% of the participants met the reference values of the LS test score for Japanese women in their 20s [17, 18]. Therefore, LS poses a risk not only in older adults but also in young women. Postpartum women need to maintain adequate mobility for childcare and rehabilitation and for preventing future falls and fractures. However, despite the fact that LBP and decreased prenatal PA could increase the risk of LS, the prevalence of LS and related factors in postpartum women within 1 year after delivery have not been investigated. Therefore, this study aimed to investigate LS prevalence in postpartum women and clarify factors associated with LS risk including LBP. To our best knowledge, this is the first study to investigate LS in postpartum women. Participants In this cross-sectional study, the study participants were postpartum women who underwent childbirth within 1 year before enrollment. We recruited postpartum women at a local health event for pregnant and postpartum women in Aichi Prefecture, Japan, in February 2019. The inclusion criterion was an age > 20 years. We excluded pregnant women, women unable to understand the study explanation, and women with missing data. We explained our research aim and measurement items to the participants orally and in writing and obtained their written consent. Thereafter, the participants were asked to complete questionnaires and perform the LS risk tests (stand-up test and 2-step test), and their body composition Page 3/15 Page 3/15 was measured. Our questionnaire consisted of items on demographic data (age, height, weight, the period from the last birth, and number of births), the International Physical Activity Questionnaire Short Form (IPAQ-SF), and the Oswestry Disability Index (ODI) for assessing LBP assessment. All assessments were performed at the event venue where recruitment was conducted. LS risk test The main outcomes were the 2-step and stand-up test scores, which reflected the participants’ physical function. These tests were used to assess LS risk. The 2-step test involves measuring stride length in two steps. The participants were instructed to take two steps, each step as long as possible, and the measurement was recorded (Fig. 1). The 2-step test score was used to classify the LS risk level and was calculated using the following formula: length of the two steps (cm)/height (cm). In the stand-up test, the height of the lowest stool from which the participant could stand up from a sitting position using two legs or one leg was recorded. Stools of 40 cm and 20 cm heights were used. The participant was considered to have completed the trial if they succeeded in holding the final standing position for longer than 3 s without requiring any additional steps (Fig. 2). The criteria for LS Stage 2 were a 2-step test score < 1.1 or difficulty rising on both legs from a 20 cm-high stool in the stand-up test. The criteria for LS Stage 1 were a 2-step test score < 1.3 or difficulty rising on one leg from a 40 cm-high stool in the stand- up test (either leg) but the ability to rise on both legs from a 20 cm-high stool. Customarily, the LS test also includes the GLFS-25, a questionnaire focused on body pain and ADL in the most recent month. However, we did not administer this questionnaire in the present study, as it was originally developed for older people. A previous study revealed that the interquartile range of the GLFS- 25 score was lower in women in their 20s and 30s than the cut-off points of LS Stage 1 [18]. Body composition Appendicular skeletal muscle mass (ASM) was measured using bioelectrical impedance analysis with a body composition analyzer (MC-780A, TANITA, Tokyo, Japan), and the skeletal muscle mass index (SMI; ASM [kg]/height2 [m]) was calculated as an index of body composition. Body fat percentage and body mass index (BMI) were used as additional indices. PA The IPAQ-SF Japanese version is a valid and reliable assessment tool for PA [19]. The IPAQ-SF includes the frequency and duration (at least 10 min) of PA of high, moderate, and low intensity in a usual week and sedentary time during a typical day. PA level below 600 metabolic equivalent (MET) min/week was categorized as low PA, and that of more than 600 MET min/week was categorized as moderate and high PA. In addition, sedentary time in the IPAQ-SF was used as an index of inactivity. LBP Statistical analyses Participants who met the criteria for LS Stage 1 or 2 in either the 2-step or stand-up test were classified as the high-risk LS group, whereas those who did not were classified as the non-LS group. The Mann– Whitney U test was used to compare continuous data between the high-risk LS and non-LS groups. Categorical data are presented as number (percentage) and were assessed using the chi-square test. A P-value < .05 was considered significant. All analyses in this study were performed using IBM SPSS Statistics ver. 28 (IBM Corp, Armonk, NY, USA). LBP The ODI is a disease-specific self-report outcome tool used to measure functional disability related to LBP [20], and the reliability and validity of the Japanese version have been reported [21]. The ODI Page 4/15 Page 4/15 consists of the following 10 items: pain intensity, personal care, lifting, walking, sitting, standing, sleeping, sex life, social functioning, and traveling. For each item, the participant rates the level of disability on a scale from 0 to 5. The final score is calculated as follows: (total score) ×100 / (5 × number of items answered). Scores range from 0 (no disability) to 100 (most severe disability). The cut-off value for functional disability due to LBP was set to 12 points based on a previous study [22]. Results In this study, we initially recruited 102 postpartum women, of whom 16 were excluded because of missing data (unanswered questionnaire items). The remaining 86 women were divided into the high-risk LS and non-LS groups (Fig. 3. The mean age of all the participants was 30.0 ± 4.2 years, and 52.3% of the women had a high risk of LS. In the stand-up test, all 13 women who met the criteria for LS could stand up on both legs from a 20 cm-high stool and were classified as having LS Stage 1, and none of them were classified as having LS Stage 2. In the 2-step test, 30 women had a 2-step score < 1.3 (LS Stage 1), and three women had a 2-step score < 1.1 (LS Stage 2). Only one woman was at risk in both the stand-up and 2-step tests. The high-risk LS group had a significantly higher ODI than did the non-LS group. However, no significant difference was observed between the two groups in terms of age, BMI, the period from the last birth, number of births, low PA, sedentary time, SMI, and body fat percentage (Table 1). Comparing the subscale of the ODI between the two groups, the high-risk LS group had a significantly higher score for pain intensity and sitting (Table 2). Results Page 5/15 Table 1 Characteristics of postnatal women with and without a high risk of locomotive syndrome Overall (n = 86) High-risk LS group (n = 45) Non-LS group (n = 41) p-value Age (years) 30 (21–40) 31 (21–40) 28 (25–39) 0.11 Body mass index (kg/m²) 20.1 (15.8– 34.6) 20.3 (17.2– 34.6) 19.9 (15.8–25) 0.22 Period since the last birth (months) 7 (1–11) 7 (1–10) 7 (2–11) 0.22 Number of births (number) 0.66 1 58 (67.4) 32 (71.1) 26 (63.4) 2 23 (26.7) 10 (22.2) 13 (31.7) 3 5 (5.8) 3 (6.7) 2 (4.9) ODI score (%) 7.5 (0–26) 10 (0–26) 4 (0–24) < 0.001 Low physical activity (number) 53 (61.6) 26 (57.8) 27 (65.8) 0.44 Sedentary time (min/day) 240 (30–1140) 225 (30–1140) 240 (60–900) 0.71 Skeletal muscle mass index (kg/m²) 6.7 (5.6–8.7) 6.7 (5.9–8.7) 6.7 (5.6–7.5) 0.81 Body fat percentage (%) 25.9 (12.9– 48.6) 26.2 (17–48.6) 25.2 (12.9– 35.8) 0.27 Stand-up test (number) 14 (16.2) 14 (31.1) 0 (0.0) < 0.001 2-step test (number) 33 (38.4) 33 (73.3) 0 (0.0) < 0.001 2-step test score 1.3 (1.0–1.5) 1.2 (1.0–1.5) 1.4 (1.3–1.5) < 0.001 LS: Locomotive syndrome, ODI: Oswestry Disability Index; Mann–Whitney U test, median (range): age, BMI, period from the last birth, ODI score, sedentary time, skeletal muscle mass index, body fat percentage, 2-step test score; Chi-square test, n (%): the number of births, low physical activity, and stand-up test Table 1 Characteristics of postnatal women with and without a high risk of locom LS: Locomotive syndrome, ODI: Oswestry Disability Index; Mann–Whitney U test, median (range): age, BMI, period from the last birth, ODI score, sedentary time, skeletal muscle mass index, body fat percentage, 2-step test score; Chi-square test, n (%): the number of births, low physical activity, and stand-up test Page 6/15 Table 2 Subscale ODI score in postnatal women with and without a high risk of locomotive syndrome Overall (n = 86) High-risk LS group (n = 45) Non-LS group (n = 41) p-value Pain intensity 1 (0–4) 1 (0–4) 1 (0–3) 0.03 Personal care 0 (0–1) 0 (0–1) 0 (0–1) 0.74 Lifting 0 (0–2) 1 (0–2) 0 (0–2) 0.06 Walking 0 (0–1) 0 (0–1) 0 (0–1) 0.20 Sitting 1 (0–2) 1 (0–2) 0 (0–2) 0.02 Standing 1 (0–2) 1 (0–2) 0 (0–2) 0.09 Sleeping 0 (0–2) 0 (0–2) 0 (0–2) 0.33 Sex life 0 (0–5) (n = 71) 0 (0–5) (n = 33) 0 (0–1) (n = 38) 0.27 Social life 0 (0–2) (n = 82) 0 (0–2) (n = 40) 0 (0–1) (n = 42) 0.41 Traveling 0 (0–4) 0 (0–4) 0 (0–1) 0.09 ODI: Oswestry Disability Index, LS: Locomotive syndrome Mann–Whitney U test, median (range) Table 2 Table 2 Subscale ODI score in postnatal women with and without a high risk of locomotive syndrome Discussion A previous study has shown that the strength of the hip muscles, including the abductors and extensors, is decreased in individuals with non-specific chronic LBP [28]. These muscles are important for standing up or taking longer strides. In the present study, we did not examine the hip muscle strength; nonetheless, some participants with LBP might have had weakened hip muscles, which may have contributed to the LS risk. However, some participants with LS did not have LBP at the time of measurement. van Benten et al. reported that single-leg standing balance was reduced despite self-reported resolution of pregnancy- related pelvic girdle pain [29]. This suggests that even after having recovered from pelvic pain in the postpartum period, women may have difficulty in rising on one leg from a 40 cm-high stool or their maximum stride length may be reduced due to decline in their balance function, which may lead to LS. Although moderate-to-high PA levels were associated with a lower LS risk, we found no significant difference in PA levels between the high-risk LS and non-LS groups. Regarding PA, 61.6% of the women in this study had low PA levels, despite the ACOG recommendation for postpartum women [8]. In the non-LS group, 57% of the participants had a low PA level. Therefore, no significant difference was noted between the two groups in terms of PA levels. Most pregnant women decrease their PA levels throughout pregnancy [9, 10], and PA levels remain low after giving birth [1, 9, 10, 12]. Based on a previous study that identified determinants of changes in PA across the transition period to parenthood [30], possible barriers to postpartum PA include limited time owing to childcare or other tasks and physical difficulties such as pelvic floor disorder. Pelvic floor disorder is a common postpartum symptom, and many postpartum women experience urinary incontinence. Pelvic floor symptoms are considered a barrier to exercise participation [31]. However, provision of accurate information from a medical professional can facilitate PA during pregnancy and after childbirth. Therefore, medical professionals, including obstetricians, midwives, and physiotherapists, should develop approaches to increase PA of pregnant and postpartum women. In particular, it may be difficult for postpartum women to balance childcare responsibilities and find time for exercise. Discussion In this study, we investigated prevalence of LS and explored factors associated with LS risk in postpartum women. We found that 52.3% of postpartum women within 1 year after childbirth had a high risk of LS, which is approximately twice the proportion of non-pregnant women in their 20s (24.5%) and 30s (26.5%) and close to the proportion of those in their 60s (53.5%) [23]. This underscores the increased likelihood of impaired mobility in postpartum women compared with that in other counterpart women of the same age. Many intervention studies aimed at improving LS have been conducted. Notably, studies on electrical muscle stimulation of the quadriceps in older women [24] and hip flexor muscle strengthening [25] have shown improvements in the 2-step test scores. Similarly, a study on middle-aged individuals who performed squats and open-eyed single-legged standing revealed improvements in the stand-up test score [26]. Furthermore, a study conducted among older women who exercised their trunk muscles using training equipment showed improvements in both the 2-step and stand-up test scores [27]. In the Page 7/15 present study, 38.4% and 16.2% of the participants were classified as LS based on the 2-step and stand- up test, respectively, and only one participant met the criteria based on both tests. Although no significant difference was noted in the SMI, it may be possible to improve LS by changing the approach depending on which LS test’s criteria are met. present study, 38.4% and 16.2% of the participants were classified as LS based on the 2-step and stand- up test, respectively, and only one participant met the criteria based on both tests. Although no significant difference was noted in the SMI, it may be possible to improve LS by changing the approach depending on which LS test’s criteria are met. In the present study, only LBP was associated with LS risk. Previous studies targeting young and middle adulthood have revealed that LBP is related to LS [15, 16], suggesting LBP as a potential risk factor for LS, even in postpartum women. In terms of ADL, we suggested that the high-risk LS group was more significantly affected by pain than was the non-LS group, especially in the sitting position. In the stand-up test, starting from a sitting position might be influenced by LBP, potentially impacting the test results. Discussion In this study, no significant relationship was observed between LS risk and PA levels; however, the PA of mothers with children should be increased, as most women have low PA levels after childbirth. Page 8/15 Page 8/15 This study has several limitations. First, the study design was cross-sectional, and the causal relationship between LS risk and LBP is unknown. Second, we did not obtain GLFS-25 data; therefore, the results of only few previous studies can be compared to those of our study. Third, the participants were recruited at a local health event, potentially excluding women with severe LBP. Finally, the JOA revised the clinical decision limits and introduced a new LS stage, Stage 3, in 2020 [32]; there were only two stages when we conducted this survey in 2018. Hence, the highest stage in this report was Stage 2. However, the present study indicates that decreased locomotive function before or during pregnancy may be a contributing factor to LBP. Medical professionals, including obstetricians, midwives, and physiotherapists, should be vigilant in assessing LS risk among postpartum women and provide appropriate interventions to promote mobility and overall health. Further validation, including intervention studies to prevent the decline in locomotive function before pregnancy as this may prevent LBP after childbirth, is needed. Conclusion bb i i This study demonstrated that 52.3% of postpartum women within 1 year after childbirth were at risk of LS, suggesting a relationship between LS risk and LBP. The high prevalence of LS among postpartum women underscores the importance of early detection and intervention strategies to prevent or mitigate LS-related impairments in mobility. Future research should focus on longitudinal studies to establish causality between LBP and LS risk in postpartum women. Abbreviations Page 9/15 Abbreviations LBP low back pain ADL activities of daily living PA physical activity ACOG the American College of Obstetrics and Gynecology LS locomotive syndrome JOA the Japanese Orthopaedic Association GLFS-25 the 25-question geriatric locomotive function scale IPAQ-SF International Physical Activity Questionnaire Short Form ODI Oswestry Disability Index ASM Authors’ contributions YK, DM, and TI proposed the research ideas, and YK, DM, TI, MT, and TA designed the study. YK, DM, TI, and RK carried out the studies and drafted the manuscript. MT and TA revised the manuscript. All authors have read and approved the final manuscript. LBP Page 9/15 appendicular skeletal muscle mass SMI skeletal muscle mass index BMI body mass index IPAQ-SF, International Physical Activity Questionnaire Short Form MET, metabolic equivalent appendicular skeletal muscle mass SMI skeletal muscle mass index BMI body mass index IPAQ-SF, International Physical Activity Questionnaire Short Form MET, metabolic equivalent Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Ethics approval and consent to participate Written informed consent was obtained from each participant per the guidelines approved by the Research Ethics Committee and the Declaration of Human Rights (Helsinki, 1975). The study protocol was approved by the Research Ethics Committees of Kyoto University (approval number R1840) and Kio University (approval number R1-01). Funding Not applicable. Consent for publication Written informed consent for use was obtained from the model in the pictures. Competing interests The authors declare that they have no competing interests. Acknowledgements The authors are especially grateful to all the participants for their willingness to participate in the study. We thank the staff of “HAPPY MAMA FESTA” for assisting with this study. We are also grateful to the Page 10/15 members of Kio University, Kyoto University, and Nagoya University for their helpful suggestions and their help carrying out the measurements. members of Kio University, Kyoto University, and Nagoya University for their helpful suggestions and their help carrying out the measurements. References 1. Deering RE, Cruz M, Senefeld JW, Pashibin T, Eickmeyer S, Hunter SK. Impaired trunk flexor strength, fatigability, and steadiness in postpartum women. Med Sci Sports Exerc. 2018;50:1558–69. 2. Fast A, Shapiro D, Ducommun EJ, Friedmann LW, Bouklas T, Floman Y. Low-back pain in pregnancy. Spine (Phila Pa 1976). 1987;12:368–71. 3. Orvieto R, Achiron A, Ben-Rafael Z, Gelernter I, Achiron R. Low-back pain of pregnancy. Acta Obstet Gynecol Scand. 1994;73:209–14. 4. Morino S, Ishihara M, Umezaki F, Hatanaka H, Iijima H, Yamashita M, et al. Low back pain and causative movements in pregnancy: a prospective cohort study. BMC Musculoskelet Disord. 2017;18:416. 5. Norén L, Ostgaard S, Johansson G, Ostgaard HC. Lumbar back and posterior pelvic pain during pregnancy: a 3-year follow-up. Eur Spine J. 2002;11:267–71. 6. Elden H, Gutke A, Kjellby-Wendt G, Fagevik-Olsen M, Ostgaard HC. Predictors and consequences of long-term pregnancy-related pelvic girdle pain: a longitudinal follow-up study. BMC Musculoskelet Disord. 2016;17:276. 7. Gutke A, Lundberg M, Östgaard HC, Öberg B. Impact of postpartum lumbopelvic pain on disability, pain intensity, health-related quality of life, activity level, kinesiophobia, and depressive symptoms. Eur Spine J. 2011;20:440–8. 8. Physical activity and. exercise during pregnancy and the postpartum period: ACOG Committee Opinion. Obstet Gynecol. 2020;135:e178–88. 8. Physical activity and. exercise during pregnancy and the postpartum period: ACOG Committee Opinion. Obstet Gynecol. 2020;135:e178–88. 9. Hesketh KR, Evenson KR, Stroo M, Clancy SM, Østbye T, Benjamin-Neelon SE. Physical activity and sedentary behavior during pregnancy and postpartum, measured using hip and wrist-worn accelerometers. Prev Med Rep. 2018;10:337–45. 9. Hesketh KR, Evenson KR, Stroo M, Clancy SM, Østbye T, Benjamin-Neelon SE. Physical activity and sedentary behavior during pregnancy and postpartum, measured using hip and wrist-worn accelerometers. Prev Med Rep. 2018;10:337–45. 10. Borodulin K, Evenson KR, Herring AH. Physical activity patterns during pregnancy through postpartum. BMC Womens Health. 2009;9:32. 10. Borodulin K, Evenson KR, Herring AH. Physical activity patterns during pregnancy through postpartum. BMC Womens Health. 2009;9:32. 11. Borodulin KM, Evenson KR, Wen F, Herring AH, Benson AM. Physical activity patterns during pregnancy. Med Sci Sports Exerc. 2008;40:1901–8. 11. Borodulin KM, Evenson KR, Wen F, Herring AH, Benson AM. Physical activity patterns during pregnancy. Med Sci Sports Exerc. 2008;40:1901–8. 12. Evenson KR, Herring AH, Wen F. Self-Reported and objectively measured physical activity among a cohort of postpartum women: the PIN Postpartum Study. J Phys Act Health. 2012;9:5–20. 12. Evenson KR, Herring AH, Wen F. References Self-Reported and objectively measured physical activity among a cohort of postpartum women: the PIN Postpartum Study. J Phys Act Health. 2012;9:5–20. 13. Nakamura K. A super-aged society and the locomotive syndrome. J Orthop Sci. 2008;13:1–2. 13. Nakamura K. A super-aged society and the locomotive syndrome. J Orthop Sci. 2008;13:1–2. 14. Nakamura K, Ogata T. Locomotive syndrome: definition and management. Clin Rev Bone Min Metab. 2016;14:56–67. 14. Nakamura K, Ogata T. Locomotive syndrome: definition and management. Clin Rev Bone Min Metab 2016;14:56–67. 15. Hirano K, Imagama S, Hasegawa Y, Ito Z, Muramoto A, Ishiguro N. The influence of locomotive syndrome on health-related quality of life in a community-living population. Mod Rheumatol. 15. Hirano K, Imagama S, Hasegawa Y, Ito Z, Muramoto A, Ishiguro N. The influence of locomotive syndrome on health-related quality of life in a community-living population. Mod Rheumatol. Page 11/15 Page 11/15 2013;23:939–44. 16. Nishimura A, Ohtsuki M, Kato T, Nagao R, Ito N, Kato K, et al. Locomotive syndrome testing in young and middle adulthood. Mod Rheumatol. 2020;30:178–83. 17. Uesugi Y, Kanaya S, Nakanishi H, Naito Y. The relationship between locomotive syndrome risk, gait pattern, and standing posture in young Japanese women: a cross-sectional study. Healthc (Basel). 2020;8:565. 18. Yamada K, Ito YM, Akagi M, Chosa E, Fuji T, Hirano K, et al. Reference values for the locomotive syndrome risk test quantifying mobility of 8681 adults aged 20–89 years: a cross-sectional nationwide study in Japan. J Orthop Sci. 2020;25:1084–92. 19. Murase N, Katsumura T, Ueda C, Inoue S, Shimomitsu T. Validity and reliability of the Japanese version of the International Physical Activity Questionnaire. J Health Welf Stat. 2002;49:1–9. 19. Murase N, Katsumura T, Ueda C, Inoue S, Shimomitsu T. Validity and reliability of the Japanese version of the International Physical Activity Questionnaire. J Health Welf Stat. 2002;49:1–9. 20. Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine (Phila Pa 1976). 2000;25:2940-52. 20. Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine (Phila Pa 1976). 2000;25:2940-52. 21. Fujiwara A, Kobayashi N, Saiki K, Kitagawa T, Tamai K, Saotome K. Association of the Japanese Orthopaedic Association score with the Oswestry Disability Index, Roland-Morris Disability Questionnaire, and Short-Form 36. Spine (Phila Pa 1976). 2003;28:1601–7. 22. Tonosu J, Takeshita K, Hara N, Matsudaira K, Kato S, Masuda K, et al. The normative score and the cut-off value of the Oswestry Disability Index (ODI). Eur Spine J. 2012;21:1596–602. 23. Figures Figure 1 Figure 1 References Pelvic floor symptoms are an overlooked barrier to exercise participation: a cross-sectional online survey of 4556 women who are symptomatic. Phys Ther. 2022;102:pzab284. 32. Locomotive Challenge Council. Locomotive syndrome. In: Locomotive Challenge Council, Locomotive syndrome pamphlet 2020. Japanese Orthopaedic Association. 2020. https://locomo- joa.jp. Accessed 30 Apr 2023. 32. Locomotive Challenge Council. Locomotive syndrome. In: Locomotive Challenge Council, Locomotive syndrome pamphlet 2020. Japanese Orthopaedic Association. 2020. https://locomo- joa.jp. Accessed 30 Apr 2023. Figures References Yoshinaga S, Shiomitsu T, Kamohara M, Fujii Y, Chosa E, Tsuruta K. Lifestyle-related signs of locomotive syndrome in the general Japanese population: a cross-sectional study. J Orthop Sci. 2019;24:1105–9. 24. Nishikawa Y, Watanabe K, Kawade S, Takahashi T, Kimura H, Maruyama H, et al. The effect of a portable electrical muscle stimulation device at home on muscle strength and activation patterns in locomotive syndrome patients: a randomized control trial. J Electromyogr Kinesiol. 2019;45:46–52. 25. Sato H, Kondo S, Saito M, Saura R. Effects of strengthening the hip flexor muscles on walking ability and the locomotive syndrome rank test: an intervention study. J Orthop Sci. 2020;25:892–6. 26. Nishimura A, Ohtsuki M, Kato T, Nagao-Nishiwaki R, Senga Y, Kato K, et al. Is locomotion training effective for middle-aged workers? J Occup Health. 2021;63:e12303. 26. Nishimura A, Ohtsuki M, Kato T, Nagao-Nishiwaki R, Senga Y, Kato K, et al. Is locomotion training effective for middle-aged workers? J Occup Health. 2021;63:e12303. 27. Kato S, Demura S, Kurokawa Y, Takahashi N, Shinmura K, Yokogawa N, et al. Efficacy and safety of abdominal trunk muscle strengthening using an innovative device in elderly patients with chronic low back pain: a pilot study. Ann Rehabil Med. 2020;44:246–55. 28. Pizol GZ, Ferro Moura Franco K, Cristiane Miyamoto G, Nunes Cabral CM. Is there hip muscle weakness in adults with chronic non-specific low back pain? A cross-sectional study. BMC Musculoskelet Disord. 2023;24:798. 28. Pizol GZ, Ferro Moura Franco K, Cristiane Miyamoto G, Nunes Cabral CM. Is there hip muscle weakness in adults with chronic non-specific low back pain? A cross-sectional study. BMC Musculoskelet Disord. 2023;24:798. 29. van Benten E, Coppieters MW, Pool JJM, Pool-Goudzwaard AL. Differences in balance control despite self-reported resolution of pregnancy-related pelvic girdle pain. a cross-sectional study. Musculoskelet Sci Pract. 2022;62:102620. Page 12/15 30. Versele V, Stok FM, Dieberger A, Deliens T, Aerenhouts D, Deforche B, et al. Determinants of changes in women’s and men’s physical activity and sedentary behavior across the transition to parenthood: a focus group study. Int J Environ Res Public Health. 2022;19:2421. 31. Dakic JG, Cook J, Hay-Smith J, Lin KY, Ekegren C, Frawley HC. Pelvic floor symptoms are an overlooked barrier to exercise participation: a cross-sectional online survey of 4556 women who are symptomatic. Phys Ther. 2022;102:pzab284. 31. Dakic JG, Cook J, Hay-Smith J, Lin KY, Ekegren C, Frawley HC. 30. Versele V, Stok FM, Dieberger A, Deliens T, Aerenhouts D, Deforche B, et al. Determinants of changes in women’s and men’s physical activity and sedentary behavior across the transition to parenthood: a focus group study. Int J Environ Res Public Health. 2022;19:2421. Two-step test Participants are instructed to take two steps, each step as long as possible, and the length of two steps from toes to toes is recorded. The 2-step score is used to classify the LS risk level and is calculated using the following formula: length of the two steps (cm)/height (cm). A 2-step score of ≥1.3 indicates “no risk of LS”. LS Stage 1 is defined as 2-step score <1.3, while LS Stage 2 is defined as 2-step score <1.1. Page 13/15 Page 13/15 LS: locomotive syndrome Figure 2 Stand-up test (a) For test A, participants show no risk in completing the task. The criterion for LS Stage 1 is difficulty in completing the test A (either leg) but being able to complete (b) test B. The criterion for LS Stage 2 is difficulty in completing test B. LS: locomotive syndrome Figure 3 Flowchart of participant inclusion, exclusion, and grouping Stand-up test (a) For test A, participants show no risk in completing the task. The criterion for LS Stage 1 is difficulty in completing the test A (either leg) but being able to complete (b) test B. The criterion for LS Stage 2 is difficulty in completing test B. (a) For test A, participants show no risk in completing the task. The criterion for LS Stage 1 is difficulty in completing the test A (either leg) but being able to complete (b) test B. The criterion for LS Stage 2 is difficulty in completing test B. (a) For test A, participants show no risk in completing the task. The criterion for LS Stage 1 is difficulty in completing the test A (either leg) but being able to complete (b) test B. The criterion for LS Stage 2 is difficulty in completing test B. LS: locomotive syndrome Page 14/15 Figure 3 Flowchart of participant inclusion, exclusion, and grouping Stand-up test: The number of persons who cannot stand up on a single leg from a stool with a 40-cm height Two-step test: The number of persons with a 2-step test score < 1 LS: locomotive syndrome Figure 3 Flowchart of participant inclusion, exclusion, and grouping Stand-up test: The number of persons who cannot stand up on a single leg from a stool with a 40-cm height Stand-up test: The number of persons who cannot stand up on a single leg from a stool with a 40-cm height Two-step test: The number of persons with a 2-step test score < 1 LS: locomotive syndrome Page 15/15
https://openalex.org/W1902815548	https://jissn.biomedcentral.com/track/pdf/10.1186/1550-2783-12-S1-P28	English	null	Acute hemodynamic effects of a multi-ingredient performance supplement on brachial artery vasodilation and blood flow volume following elbow flexion exercise in healthy young men	Journal of the International Society of Sports Nutrition	2,015	cc-by	888	Vogel et al. Journal of the International Society of Sports Nutrition 2015, 12(Suppl 1):P28 http://www.jissn.com/content/12/S1/P28 Vogel et al. Journal of the International Society of Sports Nutrition 2015, 12(Suppl 1):P28 http://www.jissn.com/content/12/S1/P28 Vogel et al. Journal of the International Society of Sports Nutrition 2015, 12(Suppl 1):P28 http://www.jissn.com/content/12/S1/P28 Methods In a randomized double-blind, crossover, placebo- controlled design, 11 recreationally-active males (28.2 ± 5.0 y, 182.4 ± 5.7cm, 87.1 ± 10.3kg) ingested either 1 serving (14.5 g) of a MIPS (SUPP; Assault™, Muscle- pharm, Denver, CO) or a flavor-matched, visually identical placebo (PLA) and performed 3 sets of 15 arm curls at 30 minutes (30P) and 120 minutes (120P) post-supple- mentation. Brachial artery vessel diameter (VD) and blood flow volume (BFV) were measured via Doppler ultrasound at 0, 3, and 6 minutes post-exercise. Additionally, BP, HR, and BIA-determined extracellular water (ECW) and intra- cellular water (ICW) were assessed. Measurements taken following 30P and 120P were compared with both resting baseline (no treatment, no exercise) and active control (no treatment, exercise) values. Data were analyzed for all Acute hemodynamic effects of a multi-ingredient performance supplement on brachial artery vasodilation and blood flow volume following elbow flexion exercise in healthy young men group, time, and group × time interactions using 2-way repeated-measures ANOVA. Alpha was predetermined at p < 0.05. Results A significant (p < 0.05) group × time interaction was present for brachial artery VD, wherein SUPP increased to a greater extent than PLA at 0 minutes following 30P compared to both resting baseline (SUPP +0.09 ± 0.03cm; PLA +0.06 ± 0.03cm) and active control (SUPP +0.05 ± 0.04cm; PLA +0.02 ± 0.02cm) values. However, the increase in BFV at 0 minutes following 30P did not vary significantly between treatments from either resting baseline (p = 0.49) or active control (p = 0.27) values. No other variables had significant (p < 0.05) group × time interactions between any other time points. Background Nutritional supplements have received attention for increasing blood flow to skeletal muscle during exercise. L-arginine is often used for its vasodilatory effects, and supplementation with nitrates has recently become more popular for the same reason. The purpose of the present study was to determine the acute hemodynamic effects of a multi-ingredient performance supplement (MIPS) containing arginine and nitrates as compared to placebo following resistance exercise in healthy young men. Conclusion Acute supplementation with a multi-ingredient perfor- mance supplement containing arginine and nitrates may increase vasodilation synergistically with resistance exer- cise 30 minutes post-ingestion. However, it remains to be seen if increased vasodilation necessarily results in increased blood flow volume to working musculature. Acknowledgements This study was supported by MusclePharm, Corp. Acknowledgements Acknowledgements This study was supported by MusclePharm, Corp. * Correspondence: roxanne.vogel@musclepharm.com 1MusclePharm Sports Science Institute, Denver, CO, USA Full list of author information is available at the end of the article POSTER PRESENTATION Open Access Authors’ details 1 1MusclePharm Sports Science Institute, Denver, CO, USA. 2Department of Human Performance, Concordia University Chicago, River Forest, IL, USA. 3Department of Movement Science, Grand Valley State University, Allendale, MI, USA. 4Department of Biomedical Engineering, Widener University, * Correspondence: roxanne.vogel@musclepharm.com 1MusclePharm Sports Science Institute, Denver, CO, USA Full list of author information is available at the end of the article © 2015 Vogel et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http:// creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/ zero/1.0/) applies to the data made available in this article, unless otherwise stated. Page 2 of 2 Page 2 of 2 Vogel et al. Journal of the International Society of Sports Nutrition 2015, 12(Suppl 1):P28 http://www.jissn.com/content/12/S1/P28 Chester, PA, USA. 5The Hospitality College, Johnson and Wales University, Denver, CO, USA. 6Department of Human Performance and Sport, Metropolitan State University, Denver, CO, USA. 7Department of Sports Exercise Science, United States Sports Academy, Daphne, AL, USA. Published: 21 September 2015 doi:10.1186/1550-2783-12-S1-P28 Cite this article as: Vogel et al.: Acute hemodynamic effects of a multi- ingredient performance supplement on brachial artery vasodilation and blood flow volume following elbow flexion exercise in healthy young men. Journal of the International Society of Sports Nutrition 2015 12(Suppl 1):P28. Chester, PA, USA. 5The Hospitality College, Johnson and Wales University, Denver, CO, USA. 6Department of Human Performance and Sport, Metropolitan State University, Denver, CO, USA. 7Department of Sports Exercise Science, United States Sports Academy, Daphne, AL, USA. Published: 21 September 2015 doi:10.1186/1550-2783-12-S1-P28 Cite this article as: Vogel et al.: Acute hemodynamic effects of a multi- ingredient performance supplement on brachial artery vasodilation and blood flow volume following elbow flexion exercise in healthy young men. Journal of the International Society of Sports Nutrition 2015 12(Suppl 1):P28. doi:10.1186/1550-2783-12-S1-P28 doi:10.1186/1550-2783-12-S1-P28 Cite this article as: Vogel et al.: Acute hemodynamic effects of a multi- ingredient performance supplement on brachial artery vasodilation and blood flow volume following elbow flexion exercise in healthy young men. Journal of the International Society of Sports Nutrition 2015 12(Suppl 1):P28. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit
https://openalex.org/W2809303947	https://www.nature.com/articles/s41533-018-0091-9.pdf	English	null	“A palliative end-stage COPD patient does not exist”: a qualitative study of barriers to and facilitators for early integration of palliative home care for end-stage COPD	Npj primary care respiratory medicine	2,018	cc-by	12,859	BACKGROUND However, there is no unifying deﬁnition or common under- standing in literature of early integrated PC. This might be due to the polymorphous nature of integrated care itself.8 For this study, the meaning of early integrated PC can be derived from combining deﬁnitions of PC and Integrated Health Services given by the World Health Organization (WHO). Received: 19 December 2017 Revised: 28 May 2018 Accepted: 4 June 2018 1End-of-Life Care Research Group, Ghent University & Vrije Universiteit Brussel (VUB), Ghent, Belgium; 2Department of Internal Medicine, Ghent University, Ghent, Belgium; 3Department of Family Medicine and Primary Health Care, Ghent University, Ghent, Belgium; 4Department of Medical Oncology, Ghent University Hospital, Ghent, Belgium and 5Department of Respiratory Medicine, Ghent University Hospital, Ghent, Belgium Correspondence: Charlotte Scheerens (charlotte.scheerens@ugent.be) These authors contributed equally: Pype, Peter and Chambaere, Kenneth. www.nature.com/npjpcrm ARTICLE OPEN “A palliative end-stage COPD patient does not exist”: a qualitative study of barriers to and facilitators for early integration of palliative home care for end-stage COPD Charlotte Scheerens1,2, Luc Deliens1,2, Simon Van Belle1,4, Guy Joos2,5, Peter Pype1,3 and Kenneth Chambaere1,2 Early integration of palliative home care (PHC) might positively affect people with chronic obstructive pulmonary disease (COPD). However, PHC as a holistic approach is not well integrated in clinical practice at the end-stage COPD. General practitioners (GPs) and community nurses (CNs) are highly involved in primary and home care and could provide valuable perspectives about barriers to and facilitators for early integrated PHC in end-stage COPD. Three focus groups were organised with GPs (n = 28) and four with CNs (n = 28), transcribed verbatim and comparatively analysed. Barriers were related to the unpredictability of COPD, a lack of disease insight and resistance towards care of the patient, lack of cooperation and experience with PHC for professional caregivers, lack of education about early integrated PHC, insufﬁcient continuity of care from hospital to home, and lack of communication about PHC between professional caregivers and with end-stage COPD patients. Facilitators were the use of trigger moments for early integrating PHC, such as after a hospital admission or when an end-stage COPD patient becomes oxygen-dependent or housebound, positive attitudes towards PHC in informal caregivers, more focus on early integration of PHC in professional caregivers’ education, implementing advance care planning in healthcare and PHC systems, and enhancing communication about care and PHC. The results provide insights for clinical practice and the development of key components for successful practice in a phase 0–2 Early Integration of PHC for end-stage COPD (EPIC) trial, such as improving care integration, patients’ disease insight and training PHC nurses in care for end-stage COPD. npj Primary Care Respiratory Medicine (2018) 28:23 ; doi:10.1038/s41533-018-0091-9 Published in partnership with Primary Care Respiratory Society UK nd-of-Life Care Research Group, Ghent University & Vrije Universiteit Brussel (VUB), Ghent, Belgium; 2Department of Internal Medicine, Ghent University, Ghent, Belgium; epartment of Family Medicine and Primary Health Care, Ghent University, Ghent, Belgium; 4Department of Medical Oncology, Ghent University Hospital, Ghent, Belgium and epartment of Respiratory Medicine, Ghent University Hospital, Ghent, Belgium rrespondence: Charlotte Scheerens (charlotte.scheerens@ugent.be) ese authors contributed equally: Pype, Peter and Chambaere, Kenneth. BACKGROUND The WHO deﬁnition of PC incorporates: (1) encouraged collaboration between all professional caregivers (which we deﬁne as general practitioners, specialist physicians, nurses, physiotherapists, dentists, pharma- cists, midwives, and paramedics), in order to connect expertize; and (2) early assessment of PC, by integrating PC with disease based “curative” therapies.9 Furthermore, the WHO deﬁnition of Integrated Health Services emphasises on “the management and delivery of health services so that patients receive a continuum of preventive and curative services, according to their needs over time and across different levels of the health system.”10 Chronic obstructive pulmonary disease (COPD) is one of the leading causes of death,1 with an illness trajectory characterised by a progressive and inexorable decline interlaced with acute exacerbations.2 People with end-stage COPD (which we chose to deﬁne as ‘mostly GOLD stage III or IV3 and low to very low functioning’, although no clear deﬁnition of severe, very severe or end-stage COPD is available in literature), mainly suffer from symptoms such as dyspnea, pain, fatigue, anxiety and low mood, leading to a poor quality of life in the ﬁnal stages of the disease.4 Despite numerous therapies to treat symptoms, end-stage COPD impacts heavily on emotional and social functioning and daily activities.4 Their physical and psychosocial symptoms are poorly addressed.5 We know from former research that palliative care (PC), if integrated earlier than the ﬁnal weeks of life with standard care, can offer support for these symptoms as they in fact signal PC needs, and may have a positive impact on people with end-stage COPD.6,7 The Global initiative for chronic Obstructive Lung Disease (GOLD) also recommends early intergrated PC as a way to improve symptoms that reﬂect PC needs (such as dyspnea, anxiety, pain, and fatigue), which would potentially be better treated if PC was not only introduced in end-of-life situations.3 Moreover, patients with end-stage COPD often die in intensive care units in hospitals or nursing homes rather than at home,11 whereas end-stage COPD patients actually prefer home care,12 and patients with all kinds of diseases wish to die at home.13 If PC is provided at home by a PC nurse or PC professional, it can also improve quality of life and care,14,15 increase the chance of dying at home16 and reduce the burden of symptoms.17 Moreover, early Published in partnership with Primary Care Respiratory Society UK A palliative end-stage COPD patient does not exist C Scheerens et al. BACKGROUND 2 Barriers to early integration of PHC for end-stage COPD (Table 2) Disease trajectory of end-stage COPD. Because of the unpredict- able disease trajectory of end-stage COPD (1.a), people with end- stage COPD often experience unexpected exacerbations or other infections, and a sudden death. This made it difﬁcult to decide when or whether PHC is needed. According to participants in FG2gp and FG5cn, it was also unclear when to go from curative care to PHC as the deteriorating functioning of the patient is often invisible (1.b) to the professional caregiver as the disease evolves slowly. integrated palliative home care (PHC) can help to avoid hospital admissions and escalation of costs related to the ﬁnal months of life for people with end-stage diseases.18 Lastly, a qualitative study has shown that end-stage COPD patients have indicated needs for PHC and fully accept early integrated PHC.19 Early integrated PHC for end-stage COPD patients in practice, however, is not without its challenges as the unpredictable illness trajectory and chance of survival can interfere with its early integration.6 In Belgium, COPD accounted for 10.7% of all deaths, which made it the third most common cause of death in 2015.20 Furthermore, a study in Flanders revealed that of all deaths from end-stage COPD in 2013, only 37.3% were referred to PC, of which 7.2% to PHC. For half of the referred patients, time of onset of PC was only six days prior to death.21 Reasons for not referring end- stage COPD patients were according to the physicians due to a lack of time, because PC was not meaningful, or that PC needs were addressed in standard care.21 As research on implementing early integration of PHC for end-stage COPD is thin on the ground, with studies only exploring patients’ acceptance of integrated PHC,19 investigating PC and end-of-life discussions for COPD in general22 or not speciﬁcally focusing on end-stage COPD,16,23 a more detailed examination of challenges and possibilities for early integration of PHC for end-stage COPD is needed to gain insight into reasons why early integration of PHC for end-stage COPD is currently lacking and how to tackle this. I once saw a terminal COPD patient, with heavy exacerbations, as if he was almost gone, but he can now live further and wrestle through all of that again. BACKGROUND Blue lips, blue as… They rarely accept that when you tell them [that they are going to die] - No, no… I am not going to die. That is the denial, that is that denial (FG5, CN). You also have these [end-stage COPD] patients, we see that visually, whose health is declining. Blue lips, blue as… They rarely accept that when you tell them [that they are going to die] - No, no… I am not going to die. That is the denial, that is that denial (FG5, CN). Resistance towards care (2.b) was also mentioned, an attitude which depended on the patient’s personal context and personality. For example, some patients did not want further help from professional caregivers because they wanted to be left alone, while others refused it because of the wish to live life the way they wanted, thereby accepting the consequences. A participant explained that an end-stage COPD patient kept on smoking even when severely ill, stating it was too late for help anyhow. Other patients seemed to wait too long to contact a doctor, which made early integration of PHC impossible as they died before care could be given. npj Primary Care Respiratory Medicine (2018) 23 Participant characteristics (Table 1) But, information… there are many who do not want to hear it [information about further care possibilities such as palliative home care], they [the patient] tell us [professional caregivers] to leave them alone (FG7, CN). Three focus group interviews with GPs and four with CNs were held with a total of 28 GPs (n = 8, n = 10, n = 10) and 28 CNs (n = 4, n = 7, n = 5, n = 12) that attended one of seven focus groups. The majority of participants were between 40 and 60 years old, and 32 were male. Clinical working experience was variable, with the largest groups (each seventeen participants) working for zero to nine years and working for 20–29 years. 32 of 56 participants did not introduce PC to end-stage COPD patients in the past year. As we recruited GPs through local peer review groups and CNs through area-speciﬁc group meetings for six of the seven focus group conversations, not all participants of these focus groups met the predeﬁned inclusion criteria: 11 of 56 participants did not have ﬁve years or more clinical working experience, and nine participants did not care for at least three end-stage COPD patients in the last year. We also found out during the focus groups that ten participants were member of a PHC team as a PHC physician or PHC nurse, either currently or in the past, without our prior knowledge. Three focus group interviews with GPs and four with CNs were held with a total of 28 GPs (n = 8, n = 10, n = 10) and 28 CNs (n = 4, n = 7, n = 5, n = 12) that attended one of seven focus groups. The majority of participants were between 40 and 60 years old, and 32 were male. Clinical working experience was variable, with the largest groups (each seventeen participants) working for zero to nine years and working for 20–29 years. 32 of 56 participants did not introduce PC to end-stage COPD patients in the past year. BACKGROUND And I think that maybe that has something to do with it, that we [professional caregivers] don’t quite see it [deterioration] like that, right? (FG1, GP). Perceived patient attitudes. A lack of disease-insight (2.a) was mentioned, as some end-stage COPD patients did not seem to understand cognitively the severity of end-stage COPD and the possibility of death. This made it difﬁcult for professional caregivers to start talking about PHC because the patient did not grasp the need for it. Participants associated this attitude more with their end-stage COPD patients than patients with other diseases such as cancer. Denial of the severity of end-stage COPD even when aware of the possible negative consequences was another example of lacking disease insight: Perceived patient attitudes. A lack of disease-insight (2.a) was mentioned, as some end-stage COPD patients did not seem to understand cognitively the severity of end-stage COPD and the possibility of death. This made it difﬁcult for professional caregivers to start talking about PHC because the patient did not grasp the need for it. Participants associated this attitude more with their end-stage COPD patients than patients with other diseases such as cancer. Denial of the severity of end-stage COPD even when aware of the possible negative consequences was another example of lacking disease insight: y g Likewise, little is known about the opinions of involved parties in early integrating PHC for end-stage COPD, with the exception of patients' perspectives19 as well as those of pulmonologists.24 However, the views of general practitioners (GPs) and community nurses (CNs) on early integrated PHC for end-stage COPD are lacking. These perspectives are crucial in identifying speciﬁc reasons why implementing this type of care is difﬁcult in end- stage COPD25,26 as GPs and CNs in Belgium are active in primary and home care and well informed about PHC services. In order to gain useful data for clinical and policy-related solutions, this study aims to identify (1) barriers and (2) facilitators from the perspective of GPs and CNs for early integration of PHC in standard care for people with end-stage COPD. This qualitative study is performed as a part of a larger study to develop a complex phase 0–2 intervention trial on early integrated PHC for end-stage COPD (EPIC) in Flanders, Belgium. You also have these [end-stage COPD] patients, we see that visually, whose health is declining. Participant characteristics (Table 1) d h h l l d Finally, participants mentioned that the attitude towards PHC was one of resistance because of the perceived afﬁliation with death (2.c), as seen in this quotation: g p p y As we recruited GPs through local peer review groups and CNs through area-speciﬁc group meetings for six of the seven focus group conversations, not all participants of these focus groups met the predeﬁned inclusion criteria: 11 of 56 participants did not have ﬁve years or more clinical working experience, and nine participants did not care for at least three end-stage COPD patients in the last year. We also found out during the focus groups that ten participants were member of a PHC team as a PHC physician or PHC nurse, either currently or in the past, without our prior knowledge. We [professional caregivers] try to stimulate that [palliative home care] for our [end-stage COPD] patients, but it is really hard. Palliative care has a bad connotation, you know. When patients hear they are palliative, they believe they are going to die (FG5, CN). Professional caregiver practices. The lack of a coherent and proactive care plan (3.a) in professional caregiver practices formed npj Primary Care Respiratory Medicine (2018) 23 Published in partnership with Primary Care Respiratory Society UK A palliative end-stage COPD patient does not exist C Scheerens et al. 3 Table 1. Participant characteristics (Table 1) Characteristics of participating general practitioners and community nurses (n = 56) General practitioners Community nurses Total Characteristics FG1 (n = 8) FG2 (n = 10) FG3 (n = 10) FG4 (n = 4) FG5 (n = 7) FG6 (n = 5) FG7 (n = 12) 56 Sex Male 4 4 4 2 3 4 3 24 Female 4 6 6 2 4 1 9 32 Age ≤29 1 1 3 2 7 30–39 1 2 2 2 2 9 40–49 1 4 1 4 1 1 4 16 50–59 4 1 1 1 4 4 15 60–69 1 2 6 9 ≥70 Practice location Urban 8 10 4 2 24 Semi-urban or rural 10 7 3 12 32 Number of end-stage COPD patients cared for in the last year (1 non-response) (1 non-response) None 4 1 1 3 9 1–9 1 3 1 2 9 16 10–19 4 1 1 5 2 13 20–29 1 1 4 1 7 ≥30 2 7 9 Number of end-stage COPD patients introduced to palliative care in the last year None 5 2 7 1 4 2 11 32 1–3 2 8 3 3 2 18 4–6 1 3 1 5 7–9 ≥9 1 1 Active in a palliative home care team Yes 4a 3a 2b 1b 10 No 8 6 7 2 6 5 12 46 Clinical work experience (years) 0–4 1 2 1 4 3 11 5–9 1 1 0 1 3 6 10–19 1 2 2 2 1 3 11 20–29 3 3 2 2 2 2 3 17 ≥30 2 2 5 2 11 aGeneral practitioners were recruited as members of a LOK group. Without our prior knowledge, we found out they were part of a palliative home care team as palliative care physicians bCommunity nurses were recruited solely because of their experience as a community nurse. Without our prior knowledge, some of them have had experience as a palliative home care nurse in the past or as a second job One participant was recruited in FG4 because another participant canceled This participant was Table 1. Characteristics of participating general practitioners and community nurses (n = 56) palliative care physicians bCommunity nurses were recruited solely because of their experience as a community nurse. Without our prior knowledge, some of them have had experience as a palliative home care nurse in the past or as a second job. Participant characteristics (Table 1) Barriers according to general practitioners (GPs) (FG1gp, FG2gp, and FG3gp) and community nurses (CNs) (FG4cn, FG5cn, FG6cn, and FG7cn) for early integrating palliative home care in standard care for patients with end-stage COPD 5.a: Consultations: not enough time during consultations to start talking about palliative home care and further care (FG4cn) gp gp 2. Discharge from hospital to home situation without concrete guidelines (FG1gp, FG3gp) c: Reimbursement system for palliative home care service 1. Palliative status for palliative home care is based on predictability of death (FG2gp, FG5cn) 2. Palliative reimbursement of palliative home care is restricted to 3 months (FG2gp, FG3gp, and FG7cn) 1. Not knowing each other well enough for proper communication (FG2gp, FG3gp, FG5cn, and FG6cn) 2. Unclear who takes initiative to introduce palliative home care to end-stage COPD patients (FG3gp) 3. Not understanding each others' messages (FG2gp) g g g 6.b: Communication between caregiver and end-stage COPD patient 1. Not discussing palliative care (needs) in detail during consultations with end-stage COPD patients (FG2gp) 2. Difﬁculties for professional caregivers to talk about palliative care needs with their end-stage COPD patients (FG2gp, FG3gp, and FG4cn) 3. Patient–family relationship can prevent communication on palliative home care (FG1gp, FG5cn, and FG7cn) 4. Professional caregivers fear talking about palliative home care because of the patient’s reaction (FG5cn) The reason I would not immediately use PHC is that I need to know what they can offer in that context. So we want them to be able to offer comfort at a critical moment. But what can they do for someone who is suffocating? So then we need to hospitalize them after all (FG2, GP). agree. But, too often, you see general practitioners who think they have the answer, while they are totally wrong and that gives complications when it comes to patient compliance. Like when you [and end-stage COPD patient] show up with a specialist’s advice and your general practitioner says “hmm, you should not do that”. Come on, that cannot happen (FG5, CN). Related to this, professional caregivers did not clearly see the added value of early integrated PHC for end-stage COPD as PC is perceived to curtail all curative options for the patient. Stopping curative care and starting PHC was said to feel unnatural, especially for GPs, as they want to cure the patient. Participant characteristics (Table 1) One participant was recruited in FG4 because another participant canceled. This participant was a palliative home care nurse in the past, but is currently a full-time researcher on palliative care a barrier, ﬁrstly because professional caregivers experienced care coordination problems in the home situation of patients with end- stage COPD. For example: have to say “yeah, but that will not help, and that will not help either, and I sometimes ﬁnd it difﬁcult, that everyone [professional caregivers] has their opinion (FG1, GP). Secondly, conﬂicting therapies between professional caregivers were said to prevent the early integration of PHC as well: On improving care: you have the cleaning help, the family help, the nurses and so on, and they all have something to say about the [end-stage COPD] patient, like maybe you should try this or that sometime, maybe try that again, and then you, the general practitioner, arrives there, and there you are, with your scientiﬁc background and all the scientiﬁc evidence that you have learned, and all of those suggestions are ﬁred at and you To me, a good general practitioner is someone who does nothing. He only manages and says “I think you are suffering from that illness, you should go see that specialist physician.” I think that is great. Because they cannot know everything, I fully Published in partnership with Primary Care Respiratory Society UK Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2018) 23 A palliative end-stage COPD patient does not exist C Scheerens et al. 4 Table 2. Barriers according to general practitioners (GPs) (FG1gp, FG2gp, and FG3gp) and community nurses (CNs) (FG4cn, FG5cn, FG6cn, and FG7cn) for early integrating palliative home care in standard care for patients with end-stage COPD 1 Disease trajectory of COPD 1.a: Unpredictable exacerbations and death (FG1gp, FG2gp, FG3gp,FG5cn, FG7cn) 1.b: Invisible deterioration of functioning (FG2gp, FG5cn) 2 Perceived patient attitudes 2.a: Lack of disease insight: 1. Not understanding the severity of the disease or realizing the possibility of death (FG1gp, FG2gp, FG3gp, FG5cn, FG7cn) 2. Denial of the severity of the disease (FG2gp, FG5cn) 2.b: Resistance to care 1. The wish to be left on their own (FG2gp, FG5cn, FG6cn, FG7cn) 2. Published in partnership with Primary Care Respiratory Society UK Participant characteristics (Table 1) The wish to lead the life as they wished, accepting the consequences (FG1gp, FG3gp, FG5cn, FG7cn) 2.c: Resistance towards palliative (home) care because of the association with death (FG2gp, FG3gp, FG6cn, FG7cn) 3 Professional caregiver practices 3.a: Lack of a coherent and proactive care plan 1. No cooperation between professional caregivers involved in home care (FG1gp, FG4cn, and FG7cn) 2. Conﬂicting therapy and treatment between professional caregivers (FG3gp, FG5cn) 3.b: Insufﬁcient experience with and negative vision of palliative home care for end-stage COPD 1. No experience in clinical practice with palliative (home) care for end-stage COPD (FG1gp, FG2gp, FG3gp, FG5cn, and FG7cn) 2. Professional caregivers continue to give life-prolonging care as added value of palliative (home) care for people with end-stage COPD is not clear (FG1gp, FG5cn) 4 Education for professional caregivers Not enough focus on knowledge and advantages of palliative (home) care for end-stage COPD in professional caregivers’ basic and continuing education (FG2gp, FG3gp, FG5cn, and FG6cn) 5 Healthcare and palliative home care system characteristics 5.a: Consultations: not enough time during consultations to start talking about palliative home care and further care (FG4cn) 5.b: Coordination between hospital and home care 1. Lack of guidance on how to early integrate palliative home care to allow the patient to stay and die at home (FG1gp, FG2gp, and FG6cn) 2. Discharge from hospital to home situation without concrete guidelines (FG1gp, FG3gp) 5.c: Reimbursement system for palliative home care services 1. Palliative status for palliative home care is based on predictability of death (FG2gp, FG5cn) 2. Palliative reimbursement of palliative home care is restricted to 3 months (FG2gp, FG3gp, and FG7cn) 6 Communication 6.a: Inter-professional communication 1. Not knowing each other well enough for proper communication (FG2gp, FG3gp, FG5cn, and FG6cn) 2. Unclear who takes initiative to introduce palliative home care to end-stage COPD patients (FG3gp) 3. Not understanding each others' messages (FG2gp) 6.b: Communication between caregiver and end-stage COPD patient 1. Not discussing palliative care (needs) in detail during consultations with end-stage COPD patients (FG2gp) 2. Difﬁculties for professional caregivers to talk about palliative care needs with their end-stage COPD patients (FG2gp, FG3gp, and FG4cn) 3. Patient–family relationship can prevent communication on palliative home care (FG1gp, FG5cn, and FG7cn) 4. Professional caregivers fear talking about palliative home care because of the patient’s reaction (FG5cn) Table 2. Better explanation of the term early integrated palliative home care can help acceptance for end-stage COPD patients: talk about it as comfort care, psychosocial support (FG2gp) 5.b: Communication between professional caregivers: appoint a care coordinator who facilitates the care transition to early integrated palliative home care (FG3gp, FG5cn, FG6cn) p gp gp 2. Inform end-stage COPD patients clearly and ﬁrmly about their disease and future (FG4cn) 3. Better explanation of the term early integrated palliative home care can help acceptance for end-stage COPD patients: talk about it as comfort care, psychosocial support (FG2gp) 5.b: Communication between professional caregivers: appoint a care coordinator who facilitates the care transition to early integrated palliative home care (FG3gp, FG5cn, FG6cn) alive after three months, GPs and CNs saw this restriction of three months as a psychological obstacle to early integrating PHC: alive after three months, GPs and CNs saw this restriction of three months as a psychological obstacle to early integrating PHC: I wonder, if we, as general practitioners, would be better educated and could prescribe oxygen, how we could quickly move on to be giving oxygen. I think that would prevent a lot of hospitalizations (FG3, GP). Three months, right, if you want to request palliative care for three or six months, we do not know whether that will be the case [for a end-stage COPD patient], and that keeps you from proposing this [palliative home care] to the patient, because of that palliative status (FG3, GP). Healthcare and PHC system characteristics. Timeslots for profes- sional caregivers’ consultations which are too short (5.a), due to the fact that professional caregivers are paid per consultation, prevented discussions about early integrated P(H)C as this topic requires a lot of time to explain properly. Furthermore, coordina- tion between hospital and home care (5.b) was inefﬁcient, observed in a lack of guidance on how to early integrate PHC into the home situation of an end-stage COPD patient in order to keep the patient at home until death. Also, a lack of concrete guidance after discharge from hospital to home was mentioned, with end-stage COPD patients sometimes leaving the hospital without knowing what the next steps of care are: Communication. A lack of proper communication between the involved professional caregivers (6.a GP), pulmonologist, CNs and PHC nurses) was observed due to different roles and perspectives on care: Specialists also speak from an ivory tower. Published in partnership with Primary Care Respiratory Society UK I’m thinking of a woman [with end-stage COPD] now, who is terminal, and sure, she has a lot of pain and she uses tramadol [an opioid]. Step one in the treatment, according to them [pulmonologists], is medication because it suppresses the respiratory system. But, come on. That is easy to say behind your little desk, wearing your suit, is it not? (FG2, GP) I [general practitioner] never knew anything [of information given by someone] from the hospital for COPD (FG2, GP). Simply said: ‘go home and handle it [the situation where the end-stage COPD patient is in] yourself (FG2, GP). After hospital admission, a moment to reorganize care (FG2gp) 1.b After a couple of exacerbations (FG2gp) 1.c: When an end-stage COPD patient becomes oxygen-dependent (FG2gp, FG3gp) 1.d: When an end-stage COPD patient is confronted with loss of functioning and becomes housebound (FG1 gp, FG2gp, FG5cn) Increase knowledge about advantages of palliative home care for informal caregivers from patients with end-stage COPD (FG1gp, FG2gp, FG3gp, FG5cn) More focus on early integrated palliative home care for end-stage COPD and concrete implementation in clinical practice in education for professional caregivers (FG5cn) Start advance care planning as a standard procedure for end-stage COPD patients living at home (FG1gp, FG2gp, FG3gp, FG4cn, FG7cn) 5.a: Communication between professional caregivers and end-stage COPD patients 1. Talking about practical matters can help professional caregivers to start talking about palliative home care (FG2gp, FG3gp) 2. Inform end-stage COPD patients clearly and ﬁrmly about their disease and future (FG4cn) 3. Better explanation of the term early integrated palliative home care can help acceptance for end-stage COPD patients: talk about it as comfort care, psychosocial support (FG2gp) 5.b: Communication between professional caregivers: appoint a care coordinator who facilitates the care transition to early integrated palliative home care (FG3gp, FG5cn, FG6cn) ndard care for patients with end-stage COPD 1.a: Hospital admission 1. After hospital admission, a moment to start talking about the future (FG1gp, FG2gp, FG3gp, FG4cn, FG6cn) 2. After hospital admission, a moment to reorganize care (FG2gp) 1.b After a couple of exacerbations (FG2gp) 1.c: When an end-stage COPD patient becomes oxygen-dependent (FG2gp, FG3gp) 1.d: When an end-stage COPD patient is confronted with loss of functioning and becomes housebound (FG1 gp, FG2gp, FG5cn) Increase knowledge about advantages of palliative home care for informal caregivers from patients with end-stage COPD (FG1gp, FG2gp, FG3gp, FG5cn) More focus on early integrated palliative home care for end-stage COPD and concrete implementation in clinical practice in education for professional caregivers (FG5cn) Start advance care planning as a standard procedure for end-stage COPD patients living at home (FG1gp, FG2gp, FG3gp, FG4cn, FG7cn) 5.a: Communication between professional caregivers and end-stage COPD patients 1. Talking about practical matters can help professional caregivers to start talking about palliative home care (FG2gp, FG3gp) 2. Inform end-stage COPD patients clearly and ﬁrmly about their disease and future (FG4cn) 3. 5 Table 3. Facilitators according to general practitioners (GPs) (FG1gp, FG2gp, and FG3gp) and community nurses (CNs) (FG4cn, FG5cn, FG6cn, and FG7cn) for early integrating palliative home care in standard care for patients with end-stage COPD 1 Trigger moments 1.a: Hospital admission 1. After hospital admission, a moment to start talking about the future (FG1gp, FG2gp, FG3gp, FG4cn, FG6cn) 2. After hospital admission, a moment to reorganize care (FG2gp) 1.b After a couple of exacerbations (FG2gp) 1.c: When an end-stage COPD patient becomes oxygen-dependent (FG2gp, FG3gp) 1.d: When an end-stage COPD patient is confronted with loss of functioning and becomes housebound (FG1 gp, FG2gp, FG5cn) 2 Involvement of informal caregivers Increase knowledge about advantages of palliative home care for informal caregivers from patients with end-stage COPD (FG1gp, FG2gp, FG3gp, FG5cn) 3 Education for professional caregivers More focus on early integrated palliative home care for end-stage COPD and concrete implementation in clinical practice in education for professional caregivers (FG5cn) 4 Healthcare and palliative home care system characteristics Start advance care planning as a standard procedure for end-stage COPD patients living at home (FG1gp, FG2gp, FG3gp, FG4cn, FG7cn) 5 Communication 5.a: Communication between professional caregivers and end-stage COPD patients 1. Talking about practical matters can help professional caregivers to start talking about palliative home care (FG2gp, FG3gp) 2. Inform end-stage COPD patients clearly and ﬁrmly about their disease and future (FG4cn) 3. Better explanation of the term early integrated palliative home care can help acceptance for end-stage COPD patients: talk about it as comfort care, psychosocial support (FG2gp) 5.b: Communication between professional caregivers: appoint a care coordinator who facilitates the care transition to early integrated palliative home care (FG3gp, FG5cn, FG6cn) Table 3. Facilitators according to general practitioners (GPs) (FG1gp, FG2gp, and FG3gp) and community nurses (CNs) (FG4cn, FG5cn, FG6cn, and FG7cn) for early integrating palliative home care in standard care for patients with end-stage COPD tioners (GPs) (FG1gp, FG2gp, and FG3gp) and community nurses (CNs) (FG4cn, FG5cn, FG6cn, and FG7cn) ndard care for patients with end-stage COPD 1.a: Hospital admission 1. After hospital admission, a moment to start talking about the future (FG1gp, FG2gp, FG3gp, FG4cn, FG6cn) 2. Participant characteristics (Table 1) Next, insufﬁcient experience with and a negative vision of PHC for end-stage COPD (3.b) was noted during the focus groups as PHC in itself was either not well known or its usefulness for end-stage COPD was not clear to participants due to a lack of experience in PC for this particular patient group. During the focus groups, participants often claimed that ‘a palliative end-stage COPD patient does not exist’. Others asked the moderator to explain what PHC could do for people with end-stage COPD: Education for professional caregivers. Basic and continuing education about PHC and its advantages for end-stage COPD seemed to be lacking, which also inﬂuenced the barriers about professional caregiver practices: npj Primary Care Respiratory Medicine (2018) 23 Published in partnership with Primary Care Respiratory Society UK A palliative end-stage COPD patient does not exist C Scheerens et al. family members were involved. earlier. ACP is already a practice in nursing homes in Belgium: earlier. ACP is already a practice in nursing homes in Belgium: Because at that moment [going to a nursing home] there is a very important changeover in the life stage of a person. And because it is actually common to do advance care planning for someone who ends up in a nursing home. That is a procedure (FG2, GP). If his wife is not at home, then he [end-stage COPD patient] is incredibly chatty and he can pour out his heart: “and I do not want to live anymore and I want to die.” And when his wife gets back, the ﬁrst thing he says: “do not say anything, my wife is here”. But come on, we only get to talk to him for ﬁfteen minutes and the rest of the day he is with her. You realize he cannot discuss his illness with her, right? (FG5, CN). Reply: we should do this for all our chronic ill patients (FG2, GP). Lastly, all focus groups mentioned communication problems due to the terminology of PHC, stating that the term has negative connotations for end-stage COPD patients and professional caregivers, as it implies impending death. Communication. Enhancing communication from professional caregivers towards end-stage COPD patients (5.a) could be a facilitator, by using practical matters such as ‘where would you like to die’ as a way to start talking about early integrated PHC. Another possibility could be giving clear-cut information about end-stage COPD and future chances of survival in order to make the patient realise the severity of their disease: Facilitators for early integration of PHC for end-stage COPD (Table 3) Trigger moments. Participants expressed the need for trigger moments in the course of the disease trajectory of end-stage COPD as a way to facilitate early integration of PHC. Examples were after a hospital admission (1.a) as a moment to start talking about the future or reorganising care, a couple of exacerbations (1. b), when the patient becomes oxygen-dependent (1.c) or house- bound due to a loss of functioning (1.d). These moments were seen as turning points when the patient realises the severity of the disease more clearly: You do not need that [advance directive] if you are already at the point of dying. But for the things that might come. family members were involved. I think about an end-stage COPD patient who always says that “they [professional caregivers] will never put me on those machines [in the hospital] anyway, right?”. [General practitioner says:]”- Sure, but then we do really have to put that on paper, right?” And that is where you have a lead [to start talking about early integration of PHC]. Those practical questions are hints to talk about how far you want to go [in future care] (FG2, GP). Someone [end-stage COPD] who goes home after hospital and gets oxygen, that is an important thing to work on as a team [of professional caregivers]. And that [early integrated PHC] is something we [professional caregivers] could then discuss (FG2, GP). Patients should be correctly informed about further possibi- lities, about what medical care can still do for them. And then the conversation should mainly be about what the patient still wants and what he or she still expects and, and good agreements will have to be made about what will and what will no longer happen to the patient. And if hospitalization is out of the question, how are we [professional caregivers, patient and informal caregivers] going to organize the care package, and especially, with what objective? (FG4, GP). For example, I think that starting oxygen at home is quite the occasion [for early integrating PHC]. After all, it announces a huge phase (FG2, GP). Involvement of informal caregivers. Mainly GPs thought that providing more information about PHC and increasing positive attitudes towards it among informal caregivers (such as family members, volunteers of PHC teams) could encourage the latter to support the end-stage COPD patient with early integrated PHC: Furthermore, even if some respondents felt the need to change the term PHC to supportive home care, others stated that focusing on symptom management, comfort and psychosocial support in conversations with end-stage COPD patients could also help the latter to accept the content of P(H)C: We [general practitioners] are often asked by the family to come and talk without the patient being present. And then we discuss what will happen, what the palliative home care team could do, practical agreements (FG2, GP). Simply said: ‘go home and handle it [the situation where the end-stage COPD patient is in] yourself (FG2, GP). There was also confusion about who should take the initiative to early integrate PHC for the patient, along with miscommunica- tions in referral letters, cited by this quote: The reimbursement system for PHC (5.c) in Flanders is by law restricted to three months, with the possibility of making a second claim,27 which can be interpreted as an existing structural barrier for referring patients to PHC. To receive this reimbursement, a patient needs to have a legal palliative status, which depends on life expectancy, i.e., between three months and 24 h before death (this rule was still legal at the time of the focus groups). As the unpredictable disease trajectory of end-stage COPD makes it hard to predict whether a patient is in the ﬁnal three months of life, this was also for the participants seen as a barrier. Although early integrated PHC can be provided if patients do not have this status, and costs related to PHC are reimbursed even if patients are still Referral letters mentioning no possibilities for curative treat- ment for a cancer patient [from oncologists to general practitioners] often state: “referral to palliative team”. But with terminal COPD it [the referral letter from pulmonologists] just says “lung function borders livability”. (FG2, GP). Another barrier was communication between professional caregivers and patients with end-stage COPD (6.b), as talking about further care and PHC needs with the patient was mentioned to be difﬁcult during consultations, especially when patients’ npj Primary Care Respiratory Medicine (2018) 23 Published in partnership with Primary Care Respiratory Society UK A palliative end-stage COPD patient does not exist C Scheerens et al. A palliative end-stage COPD patient does not exist C Scheerens et al. 6 Published in partnership with Primary Care Respiratory Society UK Main ﬁndings The results of this study have revealed perceived barriers and facilitators from the perspective of general practitioners (GPs) and community nurses (CNs) to early integration of palliative home care (PHC) in standard care for people with end-stage COPD in Flanders, Belgium. The categories of barriers were (1) unpredict- able exacerbations and death in COPD and invisible deterioration of functioning; (2) perceived patient attitudes such as a lack of disease insight and resistance towards care; (3) professional caregiver practices with a lack of a coherent and proactive plan, insufﬁcient experience and a negative view of PHC for end-stage COPD; (4) not enough focus on knowledge and advantages of PHC and palliative care (PC) for end-stage COPD in professional caregivers’ basic and continuing education; (5) healthcare and PHC system characteristics: too short consultations, insufﬁcient coordination between hospital and home care, and a reimburse- ment system for PHC that is based on life expectancy; and (6) communication: a lack of and unclear communication about care possibilities for end-stage COPD patients between professional caregivers, and a lack of clear information about PHC between professional caregivers and their patients. Although it is known that PC and PHC increases quality of life for people with end-stage COPD when integrated early,7 the content of PHC needs adaptation if integrated before the terminal stage, depending on the disease population and the personal needs of the patient.37 Research has shown the need for management of troublesome symptoms and short-term PC if integrated early.6 Managing breathlessness or relieving psycho- social symptoms which are often seen in end-stage COPD despite receiving optimal medical care5 might require the involvement of other care professionals besides a PHC nurse, such as a physiotherapist, psychologist or social worker. Re-evaluating the content of PHC if given early and integrated for end-stage COPD is therefore necessary in order to be fully effective. family members were involved. You do not have to use the term palliative if you can say okay, from now on we will give you [end-stage COPD patient] maximum comfort and we will do everything to take care of you as good as possible without calling that directly palliative [care] (FG2, GP.) Education for professional caregivers. When talking about knowl- edge and care for end-stage COPD, there was an urgent need for more information about early integrated PHC for end-stage COPD and clinical implementation of it given in standard and further education of professional caregivers. This could better prepare professional caregivers in supporting end-stage COPD patients. Finally, improving communication between professional care- givers (5.b) by appointing a care coordinator to facilitate the information ﬂow between professional caregivers involved in hospital and home settings and integrating different professional caregivers’ perspectives could increase early integration of PHC. If specialist physician’s medical information would be combined with information from home and primary caregivers this could provide a better view of the patient’s personal, medical and social context. A care coordinator could also introduce the advantages of Healthcare and PHC system characteristics. Advance care planning (ACP) as a standard procedure in clinical practice for all end-stage COPD patients could facilitate conversations about the future, further wishes and needs. This could trigger professional caregivers and COPD after end-stage patients to think about integrating PHC Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2018) 23 A palliative end-stage COPD patient does not exist C Scheerens et al. 7 early integrated PHC to the patient and their informal caregivers, as this function could have more time for these conversations than other currently involved professional caregivers: early integrated PHC to the patient and their informal caregivers, as this function could have more time for these conversations than other currently involved professional caregivers: communication between professional caregivers and end-stage COPD patients by better informing the latter about PHC possibilities;24 5.b: improving communication between profes- sional caregivers by appointing a care coordinator.24 g y pp g Due to the speciﬁc focus on early integrated PHC for end-stage COPD, this study also identiﬁed new insights into barriers on conﬂicting therapies and insufﬁcient communication between professional caregivers and a lack of guidelines after hospital discharge. Main ﬁndings p g p The categories of facilitators were (1) trigger moments to start talking about early integration of PHC: such as after hospitalisa- tion, after a couple of exacerbations, when an end-stage COPD patient becomes oxygen-dependent or becomes housebound; (2) involvement of informal caregivers in early integrated PHC for COPD; (3) information about the advantages of early integrated PHC for end-stage COPD in professional caregivers’ education; (4) including advance care planning (ACP) as a part of healthcare and PHC systems and (5) enhancing communication between profes- sional caregivers by installing a care coordinator, and enhancing communication between professional caregivers and end-stage COPD patients by explaining better and giving practical examples of what early integrated PHC could mean for end-stage COPD. y y The large volume of results on professional caregiver-patient communication showed plenty of room for improvement. Participants claimed that some end-stage COPD patients did not fully understand the disease, sometimes refused care and often interpreted the term PC or PHC as a sign of impending death. However, previous research found that patients with end-stage COPD did express the desire to talk about end-of-life care30 and fully accepted PHC and early integrated PHC.19 This could thus mean that participants in our study either misinterpreted their end-stage COPD patients’ wishes and communication preferences about PHC, or that their end-stage COPD patients did not clearly share their care needs which would conﬁrm other research ﬁnding that patients often do not fully understand the severity of end- stage COPD,38 or did not know what future care they would prefer.22 In contrast, another study found that patients did convey the need for involvement and education about end-stage COPD and PC, which could improve PC communication.39 A previous trial tested patient feedback by giving self-reported patient questionnaires on end-of-life preferences for communication, therapy and experiences. These were then given to the involved professional caregivers which resulted in better patient- professional caregiver communication.40 As literature and the results of our study did not provide a clear answer to these communication issues between professional caregivers and patients, further testing of communication systems is suggested, while improving undergraduate and postgraduate education for professional caregivers on bad news delivery, ACP and shared decision making. family members were involved. A common denominator might be insufﬁcient or non- existent communication between hospital and home care settings.24 Professional caregivers active in hospital and home care might need to cooperate better and more often. By doing this, one could adjust care and therapies more adequately and better meet the patient’s wishes,34 while not forgetting to involve the patients and their informal caregivers in discussion about care. One option could be an electronic patient ﬁle accessible to the patients, their informal caregivers, and the professional caregivers in the hospital, the primary and the home care settings.35 This electronic patient ﬁle could contain a classiﬁcation system that emphasises patients’ (PHC) needs and functioning instead of the disease, such as the comprehensive ICF core set for COPD, developed by the World Health Organization.36 Another possibility could be organizing multidisciplinary consultations consistently, each time a serious deterioration of functioning occurs, similar to multidisciplinary oncology consultations in Belgium. More research is needed to explore whether these examples could work for early integrated PHC in end-stage COPD. Looking at each end-stage COPD patient to see which network can be provided and making connections with specialists' network. “Which nurse, which GP would you [the patient] like?” Then every end-stage COPD patient will have their own network up to informal care (FG5, CN). Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2018) 23 Main ﬁndings Interpretation of ﬁndings in relation to previous research The following barriers were in line with previous research on (communication about) PC in general, early integration of PHC or PC for end-stage COPD: 1.a: unpredictable exacerbations and death;6 3.b.2: continuation of life-prolonging care in end-stage COPD;28 5.a: lack of time during consultations to start talking about PHC and further care;28 5.b: no coordination between hospital and home care;24 6.b.1: not discussing PHC, PC and PC needs in detail during consultations;29 6.b.4: professional care- givers’ fear of talking about PHC because of the patient’s reaction.30 For facilitators we saw similarities with former studies on trigger moments 1.a: after hospital admission,31 and 1.c: when an end- stage COPD patient becomes oxygen-dependent;32 3) professional caregivers’ education, with the importance of providing more focus on (implementation of) early integrated PHC;24,33 4: health system and PHC system characteristics with reported advantages of ACP as a way to introduce PHC;24 5.a.2: enhancing npj Primary Care Respiratory Medicine (2018) 23 A palliative end-stage COPD patient does not exist C Scheerens et al. A palliative end-stage COPD patient does not exist C Scheerens et al. A palliative end-stage COPD patient does not exist C Scheerens et al. 8 The unpredictable disease trajectory of end-stage COPD was mentioned as a factor impeding (timely) referral to and conversations about early integrated PHC, somewhat conﬁrmed by research stating the need for clear identiﬁcation criteria for pulmonologists to introduce PC in a timely manner.24 The trigger points identiﬁed in the results of our study could respond to this need, as they signal an increase in PHC needs due to a decline in functioning of the end-stage COPD patient such as after a hospital admission, a couple of exacerbations, oxygen-dependency or becoming housebound. These trigger moments were not related to life expectancy as the latter was seen as an inappropriate basis for deciding whether early integrated PHC was needed, which is in line with a study proving that criteria to predict survival in end- stage COPD do not work.41 Moreover, at the time of the focus group conversations, eligibility for PHC in Belgium was dependent on a palliative status based on life expectancy (less than 3 months before death), which was seen as a psychological barrier to early integrating PHC as the unpredictability of COPD prevents professional caregivers from deciding whether an end-stage COPD-patient is likely to survive for three months. Strengths and limitations The research team involved in data analysis were people with different backgrounds including psychology, sociology, general practice, primary care, PHC, pulmonology, and oncology. This enhanced the interpretation of the data due to the multitude of perspectives. Furthermore, to the extent of our knowledge, this is the ﬁrst qualitative study reporting GPs’ and CNs’ insights into barriers to and facilitators for early integration of PHC for people with end-stage COPD. We obtained a varied sample of GPs and CNs with different backgrounds, care experience and perspectives on PHC. The high number (56) of participants in seven different focus groups also constituted a key strength of this study as it improved transferability of the ﬁndings beyond the context of the individual participants’ experiences. Finally, macro change by adapting the healthcare and PHC system would be needed, for example by disconnecting eligibility for palliative status and reimbursement of PHC-related costs from life expectancy and instead linking it to lower functioning, PC and PHC needs in end-stage COPD.27 Although the Flemish govern- ment has decided to change this system, the law has not yet been changed. Additionally, incorporating ACP as a standard procedure within early integrated PHC to facilitate patient-professional caregiver communication, and appointing care coordinators as an additional role in existing care could provide continuous support for end-stage COPD patients over different care settings.42 However this would require an economic costs and beneﬁts analysis. p p p However, it is worth noting that due to altering recruitment strategies at the start of the study not all participants reached the inclusion criteria we predeﬁned for the study. Eleven out of 56 did have less than ﬁve years working experience, and nine of 56 participants did not have COPD patients in their practice in the past year. This could have inﬂuenced the results as professional caregivers with less working experience or less experience with COPD patients might have faced difﬁculties in answering questions about early integration of PHC for end-stage COPD, as they could have lacked the clinical experience to relate their answers with. Nevertheless, we believe that due to the setting of a focus group, where groups were formed with other participants having many experience, this limitation did not compose any substantial problems to the quality of the conversations and to the results. Main ﬁndings Nevertheless, somewhat contrary to our results, a previous qualitative study with end-stage COPD patients found that admission for exacerbation was considered too chaotic and not an appropriate occasion to discuss PC, although it could be a milestone leading to PC discussions.31 Pulmonologists also stated that conversations about treatment preferences should be initiated when an end-stage COPD patient is stable.32 It is important to mention that the trigger moments in the results of our study could give rise to an opportunity for talking about early integrated PHC as these moments could help the end-stage COPD patient realise the severity of the disease, but initiating the conversations should take place when the end-stage COPD patient is back in a stable context, preferably at home, after the events had occurred. More research is needed to explore the feasibility of addressing PHC needs following the different triggers. strength of the answers from focus group conversations also relied on the vivid discussion between the participants who challenged each other in giving answers to questions. Therefore, we did not exclude the less experienced participants from the analysis as their participation helped in obtaining the results. y p p p g Furthermore, 32 of 56 participants had never introduced end- stage COPD patients to PC or PHC in the past year. This might have affected the results due to a lack of experience with PC or PHC for end-stage COPD between the participants. However, this does not mean that the participants did not know what PC or PHC can do for patients, as they had have experience with PC, but mainly for cancer patients. Therefore, we believe that these participants were able enough to form an opinion on why they never or hardly introduce PHC for end-stage COPD patients compared to cancer patients and what could be done to alter this. Another limitation of this study was the lack of insights from other professional caregivers involved in care for end-stage COPD patients, such as pulmonologists and physiotherapists. Neither did we consult patients with end-stage COPD or informal caregivers. However, gaining insight into early integrated PHC was the primary focus of the study and therefore we interviewed professional caregivers active in primary and home care. Notwithstanding these limitations, the results could provide valuable information on the development of feasible interven- tions, practical implementation and policy-related recommenda- tions on early integrated PHC for end-stage COPD. Main ﬁndings y g Furthermore, 32 of 56 participants had never introduced end- stage COPD patients to PC or PHC in the past year. This might have affected the results due to a lack of experience with PC or PHC for end-stage COPD between the participants. However, this does not mean that the participants did not know what PC or PHC can do for patients, as they had have experience with PC, but mainly for cancer patients. Therefore, we believe that these participants were able enough to form an opinion on why they never or hardly introduce PHC for end-stage COPD patients compared to cancer patients and what could be done to alter this. Another limitation of this study was the lack of insights from other professional caregivers involved in care for end-stage COPD patients, such as pulmonologists and physiotherapists. Neither did we consult patients with end-stage COPD or informal caregivers. However, gaining insight into early integrated PHC was the primary focus of the study and therefore we interviewed professional caregivers active in primary and home care. Notwithstanding these limitations, the results could provide valuable information on the development of feasible interven- tions, practical implementation and policy-related recommenda- tions on early integrated PHC for end-stage COPD. Implications for policy and practice, and future research Given the complexity of implementing early integrated PHC for end-stage COPD, we suggest a multilevel strategy approach in order to successfully change related policy and practice.42 The micro level could be adapted by increasing patients’ insight into (end-stage) COPD and early integrated PHC using government- funded campaigns about PC and PHC on national television which could raise awareness of PC and PHC among the general population. Meso-level changes could be on focusing professional caregivers’ basic and continuing education more on clinical PHC practice through obligatory internships in PC and PHC settings, enhancing knowledge about (end-stage) COPD, PC and PHC needs, advantages of early integrated PHC, and focusing on skills in communication and ACP. Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2018) 23 Strengths and limitations During the focus group conversations, the experienced professional caregivers inspired the less experienced participants to reﬂect critically on the questions asked by the moderator. The y As this study was performed to develop the phase 2 EPIC trial, the results suggested using a comprehensive PHC model in the intervention with inclusion criteria representing high PHC needs as a proxy to start early integrated PHC for end-stage COPD.43 Based on our results, these inclusion criteria representing high PHC needs could be GOLD III or IV combined with low functioning such as frequent hospitalisations for COPD, exacerbations due to COPD, becoming housebound or oxygen-dependent. Key compo- nents could cover several dimensions of appropriate PHC, from improving patient’s disease insight, to training the PHC team in knowledge and therapy for end-stage COPD, and integrating care by trying to improve cooperation and communication between involved professional caregivers. Previous interventions in early PC A palliative end-stage COPD patient does not exist C Scheerens et al. 9 and PHC for end-stage COPD mainly focused on caring for one symptom, for example managing breathlessness44 or provided training about one care aspect, such as nutrition.23 Instead, we suggest using several components to provide a holistic PHC approach, in order to fully tackle the lack of early integrated PHC for people with end-stage COPD. individual recruitment of GPs and CNs for focus group participation was difﬁcult and we had to change strategy. Only one focus group with CNs was composed by gathering independent CNs from the same region. This was done by contacting individual CNs from one urban area by phone through a contact lis fron the Flemish Professional Association for Independent Nurses which is available online: http://www. verplegingthuis.be/. individual recruitment of GPs and CNs for focus group participation was difﬁcult and we had to change strategy. Only one focus group with CNs was composed by gathering independent CNs from the same region. This was done by contacting individual CNs from one urban area by phone through a contact lis fron the Flemish Professional Association for Independent Nurses which is available online: http://www. verplegingthuis.be/. As a solution, we used beside the convenience sampling technique, a purposive sampling technique which allowed the research team to select participants based on the researchers’ judgment. We organized focus groups which consisted either of one regionally composed group of GPs or one regionally composed group of CNs. Recruitment strategy The participants were initially identiﬁed by a member of the research team and other key contacts who were either a GP, CN, or policy member of organisations involved in community care. Further recruitment for GPs was undertaken by contacting several people responsible for local peer review (LOK) GP groups in Flanders in person, by phone or e-mail. Further recruitment of CNs were recruited by contacting the “Wit-Gele Kruis”, a Flemish organisation for CNs. This organisation was asked to help with the recruitment by forwarding the call for participants to its members. With the help of this organisation, we contacted those responsible for regionally composed groups of CNs by phone to organise the focus group when that group had a meeting. Potential groups of GPs or CNs were invited to take part in a focus group with e-mails containing information about the study and participation. Suitable dates and venues were arranged with the people in charge of the local peer review GP groups or regionally composed CN group, if all group members agreed to participate in the study. A qualitative approach using focus groups was chosen for its group dynamic features that stimulate interaction between participants and allow the moderator to use more active interview techniques than with face-to- face interviews.45 This approach was supported by the methodological orientation of grounded theory,45 as we constructed new insights based on data obtained from the focus groups. The research protocol and topic guides were approved by the Ethics Committee of Ghent University Hospital (Reference: 2016/0171). Strengths and limitations This type of recruitment could offer us a sample of GPs and CNs representing a wide range of experience related to the topic (maximum variation sampling), even if not all participants would meet the inclusion criteria. For GPs these were local peer review GP groups (LOK): geographically determined groups of GPs from both individual and group practices. They meet four times a year to share and critically evaluate their medical practice (peer review) and to improve their quality of care. For CNs these were regionally composed groups from the National Association of Catholic Flemish Nurses and Midwifes (NVKVV) who meet monthly to discuss their practice and share work-related experiences.47 CONCLUSION Our study uncovered barriers in terms of the disease trajectory, patient attitudes, professional caregivers’ practices, the healthcare and PHC system and communication problems. Facilitators provided possibilities at many levels for a successful implementa- tion of early integrated PHC in practice or development of early integrated PHC interventions for end-stage COPD. This requires a multilevel approach with the involvement of professional caregivers active in hospital and home settings, while not forgetting to actively include end-stage COPD patients and informal caregivers in the process. Study setting The study was based in workplace settings in urban and semi-urban regions in Flanders, Belgium, in 2016, as it was part of the development of a phase 2 intervention on early integration of PHC for end-stage COPD in Flanders, Belgium. Study population and sampling Data collection The study population consisted of GPs and CNs involved in primary and home care settings. In selecting the participants, three criteria were stipulated as a guiding line for recruitment: (1) Dutch speaking; (2) at least ﬁve years experience as a GP or CN; (3) having cared for at least three end- stage COPD patients. We also took into account variation in semi-urban and urban areas. We then used ‘convenience sampling’. With this technique, the sample was composed of participants or groups who met the criteria and who were available or signed up ﬁrst.46 However, Recruitment and focus group conversations took place between March and September 2016. A semi-structured topic guide (pilot tested), consisting of four main questions and a set of prompts for each question, was developed and reviewed within a multidisciplinary research team of sociologists, a GP, a lung specialist and an oncologist (for content of the topic guide: Fig. 1). End-stage COPD was described during the focus group as mostly GOLD stage III or IV3 and low to very low functioning’. Introducon Introducon 1. What are the main problems for people with end-stage COPD, physically, psychologically and socially? - Which problems of people with end-stage COPD have the biggest impact on their well-being and health? 2. How is the current palliave home care for people with end-stage COPD organised? - Who is involved in the current care and what are their roles? - What is the quality level of communicaon and informaon towards people with end-stage COPD? 3. How could the current problems with PHC for people with end-stage COPD be beer addressed (through involvement of palliave home care teams)? - How could some soluons help improve quality of life for people with end-stage COPD? - Which professional caregiver could play a role in the improvement? 4. How could early integraon of palliave home care be implemented in standard care for people with end- stage COPD? Fig. 1 Topic list based on research questions Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2018) 23 1. What are the main problems for people with end-stage COPD, physically, psychologically and socially? - Which problems of people with end-stage COPD have the biggest impact on their well-being and health? - Who is involved in the current care and what are their roles? ACKNOWLEDGEMENTS 16. Enguidanos, S. M., Cherin, D. & Brumley, R. Home-based palliative care study. J. Soc. Work End. Life Palliat. Care 1, 37–56 (2005). The authors would like to thank all participating GPs and CNs for their efforts in providing data, Prof. Dr. Koen Pardon and Naomi Dhollander, MSc, for their efforts in organizing and executing the focus groups, as well as Helen White for her language editing. SBO/IWT (grant number 140009). 17. Gomes, B., Calanzani, N., Curiale, V., McCrone, P. & Higginson, I. J. Effectiveness and cost-effectiveness of home palliative care services for adults with advanced illness and their caregivers. Cochrane Database of Systematic Reviews (ed Gomes, B.) (Wiley, Chichester, 2013). CD007760. 18. Brian Cassel, J. et al. Effect of a home-based palliative care program on healthcare use and costs. J. Am. Geriatr. Soc. 64, 2288–2295 (2016). Data availability 14. Gomes, B., Calanzani, N. & Higginson, I. J. Beneﬁts and costs of home palliative care compared with usual care for patients with advanced illness and their family caregivers. JAMA 311, 1060 (2014). The data that support the ﬁndings of this study are available from the corresponding author (C.S) upon reasonable request. 15. Singer, A. E. et al. Populations and interventions for palliative and end-of-life care: a systematic review. J. Palliat. Med. 19, 995–1008 (2016). A palliative end-stage COPD patient does not exist C Scheerens et al. A palliative end-stage COPD patient does not exist C Scheerens et al. 10 Each focus group was moderated by one male senior researcher (sociologist or psychologist) and observed by one female junior researcher (sociologist or psychologist) who made ﬁeld notes, all experienced in conducting focus groups due to training in former education and conducting qualitative research in previous studies. The focus groups took place in a quiet room, were conducted in Flemish, lasted on average one and a half hours and were audio taped, for which all participants gave informed written consent. All participants ﬁlled in a short questionnaire regarding their own demographic characteristics, clinical experience and experience with care and PC or PHC for end-stage COPD. After conducting two focus groups with GPs, the research team slightly adapted the topic list by leaving out the ﬁrst question on perceived main problems for people with end-stage COPD, as this question did not lead to signiﬁcant information regarding early integrated PHC. We continued recruitment and sampling until data saturation was achieved. Saturation was deﬁned when no new themes on barriers and facilitators occurred during the focus group. COPD, and nocturnal hypoventilation—a consensus conference report. Chest. 116, 521–534 (1999). http://www.ncbi.nlm.nih.gov/pubmed/10453883. Accessed 29 Nov 2016. 3. Global Initiative for Chronic Obstructive Lung Disease (GOLD). From the Global Strategy for the Diagnosis, Management and Prevention of COPD. (2017). http:// goldcopd.org. 4. McSweeny, A. J., Grant, I., Heaton, R. K., Adams, K. M. & Timms, R. M. Life quality of patients with chronic obstructive pulmonary disease. Arch. Intern Med. 142, 473–478, http://www.ncbi.nlm.nih.gov/pubmed/7065785 (1982). Accessed 29 Nov 2016. 5. Gore, J. M., Brophy, C. J. & Greenstone, M. A. How well do we care for patients with end stage chronic obstructive pulmonary disease (COPD)? A comparison of palliative care and quality of life in COPD and lung cancer. Thorax 55, 1000–1006, http://www.ncbi.nlm.nih.gov/pubmed/11083884 (2000). Accessed 6 Feb 2017. 6. Maddocks, M., Lovell, N., Booth, S., Man, W. D.-C. & Higginson, I. J. Palliative care and management of troublesome symptoms for people with chronic obstructive pulmonary disease. Lancet 390, 988–1002 (2017). 7. Higginson, I. J. et al. An integrated palliative and respiratory care service for patients with advanced disease and refractory breathlessness: a randomised controlled trial. Lancet Respir. Med. 2, 979–987 (2014). AUTHOR CONTRIBUTIONS use and costs. J. Am. Geriatr. Soc. 64, 2288–2295 (2016). 19. Damps-Konstańska, I. et al. Acceptance of home support and integrated care among advanced COPD patients who live outside large medical centers. Appl. Nurs. Res. 31, 60–64 (2016). C.S., L.D., K.C., G.J., S.V.B., and P.P. were involved in the study conception and design and in obtaining ethical approvals. C.S. and K.C. were involved in the data collection. C.S., K.C., and P.P., experienced qualitative researchers with backgrounds in sociology (C.S. and K.C.) and medical, palliative home and primary care (P.P.) analysed the data. Analysis was discussed with all contributing authors. The ﬁrst and following drafts were written by C.S. All authors commented on the ﬁrst and following drafts, revised them critically, and agreed with the ﬁnal version. All authors are accountable for all aspects of the work. 20. Statbel. Causes of Death Belgium. 25 januari 2018. https://statbel.fgov.be/nl/ themas/bevolking/sterfte-en-levensverwachting/doodsoorzaken#news. Accessed April 11, 2018. 21. Scheerens, C. et al. Comparing the use and timing of palliative care services in COPD and lung cancer: a population-based survey. Eur. Respir. J. 51, 1702405 (2018). 22. Tavares, N., Jarrett, N., Hunt, K. & Wilkinson, T. Palliative and end-of-life care conversations in COPD: a systematic literature review. ERJ Open Res. 2017. https:// doi.org/10.1183/23120541.00068-2016. ADDITIONAL INFORMATION Competing interests: The authors declare no competing interests. 23. Berggren, E. et al. Early palliative home care: evaluation of an interprofessional educational intervention for district nurses and general practitioners about nutritional care. SAGE Open Med. 5, 205031211772646 (2017). Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. 24. Duenk, R. et al. The view of pulmonologists on palliative care for patients with COPD: a survey study. Int J. Chron. Obstruct Pulmon Dis. ume 12, 299–311 (2017). 25. van der Plas, A. G. et al. PaTz groups for primary palliative care: reinventing cooperation between general practitioners and district nurses in palliative care: an evaluation study combining data from focus groups and a questionnaire. BMC Fam. Pract. 15, 14 (2014). Data analysis The focus groups were completely transcribed verbatim. Then we used NVivo 9 software to code and analyse the data according to the research questions. Two researchers (CS and KC) ﬁrst read and coded the data in themes which were derived from the data from four full focus group transcripts and compared similarities and differences in their analyses until a primary coding framework was constructed. Then all seven focus group transcripts were independently read, compared with the primary coding framework by the two researchers and these results were discussed with all members of the research team. Codes were added, modiﬁed or merged when necessary. A third and fourth researcher (PP and LD) made ﬁnal changes to the codes, which were approved by the other two coding researchers. Once coding was ﬁnalized, all transcripts and the coding framework were revised and reﬁned, resulting in (sub)categories of barriers and facilitators. Quotations were selected and approved by the research team to illustrate the results. Transcripts were not sent back to participants for correction but respondent validation of the results was undertaken by sending the results of the study by e-mail to all participating GPs and CNs for consent. 8. Rand Europe. National Evaluation of the Department of Health’s Integrated Care Pilots. (RAND Europe, Cambridge, 2008). https://www.rand.org/content/dam/ rand/pubs/technical_reports/2012/RAND_TR1164.pdf. Accessed 17 April 2018. 9. Sepúlveda, C., Marlin, A., Yoshida, T. & Ullrich, A. Palliative Care: the World Health Organization’s global perspective. J. Pain Symptom Manage. 24, 91–96, http:// www.ncbi.nlm.nih.gov/pubmed/12231124 (2002). Accessed 11 April 2018. 10. INTEGRATED HEALTH SERVICES—WHAT AND WHY? Main Messages Making Health Systems Work. (2008). http://www.who.int/healthsystems/service_delivery_ techbrief1.pdf. Accessed 17 April 2018. 11. Cohen, J. et al. Differences in place of death between lung cancer and COPD patients: a 14-country study using death certiﬁcate data. npj Prim. Care Respir. Med. 27, 14 (2017). 12. Bereza, B. G., Troelsgaard Nielsen, A., Valgardsson, S., Hemels, M. E. & Einarson, T. R. Patient preferences in severe COPD and asthma: a comprehensive literature review. Int J. Chron. Obstruct Pulmon Dis. 10, 739–744 (2015). 13. Gomes, B., Calanzani, N., Gysels, M., Hall, S. & Higginson, I. J. Heterogeneity and changes in preferences for dying at home: a systematic review. BMC Palliat. Care 12, 7 (2013). Published in partnership with Primary Care Respiratory Society UK Data collection npj Primary Care Respiratory Medicine (2018) 23 Published in partnership with Primary Care Respiratory Society UK 11 27. Maetens, A. et al. Policy measures to support palliative care at home: a cross- country case comparison in three European Countries. J. Pain. Symptom Manage. 54, 523–529 (2017). 40. Au, D. H. et al. A randomized trial to improve communication about end-of-life care among patients with COPD. Chest 141, 726–735 (2012). 41. Marin, J. M. et al. Multicomponent indices to predict survival in COPD: the COCOMICS study. Eur. Respir. J. 42, 323–332 (2013). 28. Beernaert, K. et al. What Are Physicians’ Reasons for Not Referring People with Life- Limiting Illnesses to Specialist Palliative Care Services? A Nationwide Survey. PLoS One 10, e0137251 (2015). 42. Ogunbayo, O. J. et al. Understanding the factors affecting self-management of COPD from the perspectives of healthcare practitioners: a qualitative study. npj Prim. Care Respir. Med. 27, 54 (2017). 29. Curtis, J. R., Engelberg, R. A., Wenrich, M. D. & Au, D. H. Communication about palliative care for patients with chronic obstructive pulmonary disease. J. Palliat. Care 21, 157–164, http://www.ncbi.nlm.nih.gov/pubmed/16334970 (2005). Accessed 20 Nov 2017. 43. Duenk, R. G. et al. PROLONG: a cluster controlled trial to examine identiﬁcation of patients with COPD with poor prognosis and implementation of proactive pal- liative care. BMC Pulm. Med. 14, 54 (2014). 30. Curtis, J. R., Engelberg, R. A., Nielsen, E. L., Au, D. H. & Patrick, D. L. Patient- physician communication about end-of-life care for patients with severe COPD. Eur. Respir. J. 24, 200–205, http://www.ncbi.nlm.nih.gov/pubmed/15332385 (2004). Accessed 29 Nov 2016. 44. Farquhar, M. C. et al. The clinical and cost effectiveness of a Breathlessness Intervention Service for patients with advanced non-malignant disease and their informal carers: mixed ﬁndings of a mixed method randomised controlled trial. Trials 17, 185 (2016). 31. Seamark, D. et al. Is hospitalisation for COPD an opportunity for advance care planning? A qualitative study. Prim. Care Respir. J. 21, 261–266 (2012). 45. Mortelmans, D. Handboek Kwalitatieve Onderzoeksmethoden. Acco; 2013. 46. Palinkas, L. A. et al. Purposeful sampling for qualitative data collection and analysis in mixed method implementation research. Adm. Policy Ment. Health Ment. Health Serv. Res. 42, 533–544 (2015). 32. Sullivan, K. E., Hébert, P. C., Logan, J., O’Connor, A. M. & McNeely, P. D. What do physicians tell patients with end-stage COPD about intubation and mechanical ventilation? Chest 109, 258–264, http://www.ncbi.nlm.nih.gov/pubmed/8549194 (1996). Accessed 21 Nov 2017. 47. Nationaal Verbond der Katholieke Vlaamse Verpleegkundigen en Vroedvrouwen vzw (NVKVV). Year Report 2016. http://www.nvkvv.be/ﬁle?ﬂe=936027&ssn= d14695e827e4799e87c9125ad7c5e5c024077298. Published in partnership with Primary Care Respiratory Society UK © The Author(s) 2018 npj Primary Care Respiratory Medicine (2018) 23 REFERENCES 1. Mannino, D. M. COPD: epidemiology, prevalence, morbidity and mortality, and disease heterogeneity. Chest 121, 121S–126S, http://www.ncbi.nlm.nih.gov/ pubmed/12010839 (2002). Accessed 29 Nov 2016. 1. Mannino, D. M. COPD: epidemiology, prevalence, morbidity and mortality, and disease heterogeneity. Chest 121, 121S–126S, http://www.ncbi.nlm.nih.gov/ pubmed/12010839 (2002). Accessed 29 Nov 2016. 1. Mannino, D. M. COPD: epidemiology, prevalence, morbidity and mortality, and disease heterogeneity. Chest 121, 121S–126S, http://www.ncbi.nlm.nih.gov/ pubmed/12010839 (2002). Accessed 29 Nov 2016. 26. Yamagishi, A. et al. Providing palliative care for cancer patients: the views and exposure of community general practitioners and district nurses in Japan. J. Pain. Symptom Manage. 43, 59–67 (2012). 2. Porte, D., Leger, P., Hill, N. & Criner, G. Clinical indications for noninvasive positive pressure ventilation in chronic respiratory failure due to restrictive lung disease, 2. Porte, D., Leger, P., Hill, N. & Criner, G. Clinical indications for noninvasive positive pressure ventilation in chronic respiratory failure due to restrictive lung disease, Published in partnership with Primary Care Respiratory Society UK Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2018) 23 A palliative end-stage COPD patient does not exist C Scheerens et al. (2016). 33. Beernaert, K. et al. Family physicians’ role in palliative care throughout the care continuum: stakeholder perspectives. Fam. Pract. 32, cmv072 (2015). 34. Ahia, C. L. & Blais, C. M. Primary palliative care for the general internist: integrating goals of care discussions into the outpatient setting. Ochsner J. 14, 704–711, http://www.ncbi.nlm.nih.gov/pubmed/25598737 (2014). Accessed 22 Nov 2017. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/. 35. Shield, R. R. et al. Gradual electronic health record implementation: new insights on physician and patient adaptation. Ann. Fam. Med. 8, 316–326 (2010). 36. Jobst, A., Kirchberger, I., Cieza, A., Stucki, G. & Stucki, A. Content validity of the comprehensive ICF core set for chronic obstructive pulmonary diseases: an international delphi survey. Open Respir. Med J. 7, 33–45 (2013). 37. Temel, J. S. et al. Effects of early integrated palliative care in patients with lung and GI cancer: a randomized clinical trial. J. Clin. Oncol. 35, 834–841 (2017). 38. Schroedl, C. J. et al. A qualitative study of unmet healthcare needs in chronic obstructive pulmonary disease. A potential role for specialist palliative care? Ann. Am. Thorac. Soc. 11, 1433–1438 (2014). 39. Curtis, J. R. et al. Patients’ perspectives on physician skill in end-of-life care: differences between patients with COPD, cancer, and AIDS. Chest 122, 356–362, http://www.ncbi.nlm.nih.gov/pubmed/12114382 (2002). Accessed 21 Nov 2017. npj Primary Care Respiratory Medicine (2018) 23
https://openalex.org/W2775489671	https://jurnal.unimed.ac.id/2012/index.php/js/article/download/6310/7622	Indonesian	null	MENINGKATKAN HASIL BELAJAR BAHASA INDONESIA SISWA MELALUI PENERAPAN MODEL TGT DI KELAS IV-A SD NEGERI 058355 AMAN DAMAI	JS (Jurnal Sekolah)	2,017	cc-by-sa	3,740	MENINGKATKAN HASIL BELAJAR BAHASA INDONESIA SISWA MELALUI PENERAPAN MODEL TGT DI KELAS IV-A SD NEGERI 058355 AMAN DAMAI Esra Yunila Guru SD Negeri 058355 Aman Damai Surel : esra_yunila@gmail.com Abstract: Improving Learning Outcomes Indonesian Students Through Cooperative Learning Model Implementation Team Game Tournament In Class IV-A Elementary School No. 058355 Aman Damai. This study aims to determine whether learning outcomes can be improved by applying Cooperative Learning Model TipeTeam Game Tournamen (TGT) that can affect student learning outcomes Indonesian Language in Elementary School fourth grade A 058 355 Aman Damai. The subjects amounting to 22 students. From the observation by two people pengamatdengan apply Cooperative Learning Model Team Game Tournamen gained an average of student learning activities in the first cycle and the second cycle .. The results of data analysis showed the average value of the first cycle Formative I was 67.3 with individual mastery as 8 people while students in the class was not finished. Cycle II students' average score was 80.9 Formative II with 20 students completed the class individually and thoroughly. Keywords: Learning Model Team Game Tournament, Student Activities Abstrak: Meningkatkan Hasil Belajar Bahasa Indonesia Siswa Melalui Penerapan Model Pembelajaran KooperatifTipe Team Game Tournament Dalam Kelas IV-A SD Negeri No. 058355 Aman Damai.Penelitian ini bertujuan untuk mengetahui apakah hasil belajar dapat meningkat dengan menerapkan Model Pembelajaran Kooperatif TipeTeam Game Tournamen (TGT) mata pelajaran Bahasa Indonesia di kelas IV A SD Negeri 058355 Aman Damai. Subjek penelitian berjumlah 22 orang siswa.Dari hasil observasi diperoleh rata rata aktivitas belajar siswa pada siklus I dan siklus II.. Hasil analisis data menunjukan Siklus I rata rata nilai Formatif I adalah 67,3 dengan ketuntasan individu sebanyak 8 orang siswa sementara secara kelas belum tuntas. Siklus II rata rata nilai siswa Formatif II adalah 80,9 dengan 20 orang siswa tuntas secara individu dan secara kelas tuntas. Kata Kunci:Model Pembelajaran Team Game Tournament, Aktivitas Belajar Siswa bermartabat dalam rangka mencerdaskan kehidupan bangsa, bertujuan untuk berkembangnya potensi siswa agar menjadi Manusia yang beriman dan bertakwa kepada Tuhan Yang Maha Esa, berakhlak mulia, sehat, berilmu, cakap, kreatif, mandiri, dan menjadi warga negara PENDAHULUAN Undang-Undang Republik Indonesia No. 20 tahun 2003 tentang Sistem Pendidikan Nasional menyebutkan bahwa: Pendidikan nasional berfungsi mengembangkan kemampuan dan membentuk watak serta peradaban bangsa yang p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 93 Jurnal Sekolah (JS). Vol 1 (1) Desember 2016, hlm 93-102 yang demokratis serta bertanggung jawab. atau model pembelajaran kooperatif yang mudah diterapkan, melibatkan aktivitas seluruh siswa tanpa harus ada perbedaan status, melibatkan peran siswa sebagai tutor sebaya dan mengandung unsur permainan dan reinforcement. Pada model ini siswa memainkan permainan dengan anggota-anggota tim lain untuk memperoleh tambahan skor pada tim mereka. j Peneliti adalah guru kelas IV A SD Negeri No. 058355 Aman Damai, Kecamatan Batang Serangan. Berdasarkan pengamatan peneliti pada bidang studi Bahasa Indonesia di kelas penelitian, ternyata banyak siswa yang bermain-main saat pembelajaran dan kurang memperhatikan guru saat menjelaskan. Dari 22 siswa, hanya 5 orang yang memperhatikan serta terlibat dalam pembelajaran sehingga menyebabkan lebih dari 50% siswa tidak lulus kriteria ketuntasan minimal (KKM) yang ditetapkan sebesar 68.Peneliti sudah pernah menyampaikan masalah ini kepada orang tua siswa, namun sangat di sayangkan hal itu tidak banyak membantu. Orang tua siswa juga kurang mendukung dan merasa sulit menyuruh anak mereka belajar. Aktivitas belajar dengan permainan yang dirancang dalam pembelajaran kooperatif model TGT memungkinkan siswa dapat belajar lebih rileks disamping menumbuhkan tanggung jawab, kerjasama, persaingan sehat dan keterlibatan belajar. Rahmat menyatakan ada 5 komponen tahapan utama dalam pembelajaran kooperatif type TGT yaitu: 1. Penyajian kelas Pada awal pembelajaran guru menyampaikan materi dalam penyajian kelas, biasanya dilakukan dengan pengajaran langsung atau dengan ceramah, diskusi yang dipimpin guru. Pada saat penyajian kelas ini siswa harus benar-benar memperhatikan dan memahami materi yang disampaikan guru, karena akan membantu siswa bekerja lebih baik pada saat kerja kelompok dan pada saat game karena skor game akan menentukan skor kelompok. Masalah tersebut diatas, berdasarkan pengamatan peneliti malasah tersebut diakibatkan rendahnya aktivitas belajar siswa yang terlibat dalam proses pembelajaran sehinggasangat mempengaruhi hasil belajar siswaSD Negeri No. 058355 Aman Damai.modelpembelajaran Kooperatif Tipe Team Game Turnament peneliti pilih untuk mengundang ketertarikan dan keterlibatan siswa dalam proses belajar mengajar sehingga meningkatkan aktivitas belajar siswa untuk bidang studi Bahasa Indonesia. 2. PENDAHULUAN Kelompok (team) Kelompok biasanya terdiri dari 4 sampai 5 orang siswa yanganggotanya heterogen dilihat dari prestasi akademik, jenis Pembelajaran kooperatif model TGT adalah salah satu type Pembelajaran kooperatif model TGT adalah salah satu type p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 94 Esra Yunila, Meningkatkan Hasil Belajar ... kelamin dan ras atau etnik.Fungsi kelompok adalah untuk lebih mendalami materi bersama teman kelompoknya dan lebih khusus untuk mempersiapkan anggota kelompok agar bekerja dengan baik dan optimal pada saat game. rata skor memenuhi kriteria yang ditentukan. Team mendapat julukan “Super Team” jika rata- rata skor 45 atau lebih, “Great Team” apabila rata-rata mencapai 40-45 dan “Good Team” apabila rata-ratanya 30-40. Adapun rumusan masalah dari latar belakang di atas adalah sebagai berikut: (1) Apakah aktivitas belajar Bahasa Indonesia siswa meningkat dengan menerapkan model pembelajaran Kooperatif Tipe Team Game Turnamentdi kelas IV A SD Negeri No. 058355 Aman Damai Tahun Ajar 2015-2016? (2) Apakah hasil belajar Bahasa Indonesia siswa meningkat dengan menerapkan modelpembelajaran Kooperatif Tipe Team Game Turnament di kelas IV A SD Negeri No. 058355 Aman Damai Tahun Ajar 2015-2016? 3. Game Game terdiri dari pertanyaan- pertanyaan yang dirancang untuk menguji pengetahuan yang didapat siswa dari penyajian kelas dan belajar kelompok. Kebanyakan game terdiri dari pertanyaan-pertanyaan sederhana bernomor. Siswa memilih kartu bernomor dan mencoba menjawab pertanyaan yang sesuai dengan nomor itu. Siswa yang menjawab benar pertanyaan itu akan mendapat skor. Skor ini yang nantinya dikumpulkan siswa untuk turnamen mingguan. Adapun tujuan dari penelitian yang dapat ditarik dari rumusan masalah di atas adalah sebagai berikut: (1) Untuk mengetahui apakah aktivitas belajar Bahasa Indonesia siswa meningkat dengan menerapkan model pembelajaran Kooperatif Tipe Team Game Turnament di kelas IV A SD Negeri No. 058355 Aman Damai Tahun Ajar 2015-2016. (2) Untuk mengetahui apakah hasil belajar Bahasa Indonesia siswa meningkat dengan menerapkan model pembelajaran Kooperatif Tipe Team Game Turnament di kelas IV A SD Negeri No. 058355 Aman Damai Tahun Ajar 2015-2016. 4. Turnamen Biasanya turnamen dilakukan pada akhir minggu atau pada setiap unitsetelah guru melakukan presentasi kelas dan kelompok sudah mengejakan lembar kerja.Turnamen pertama guru membagi siswa ke dalam beberapa meja turnamen.Tiga siswa tertinggi prestasinya dikelompokkan pada meja I, tiga siswa selanjutnya pada meja II dan seterusnya. 5. Team Recognize (Penghargaan Kelompok) Guru kemudian mengumumkan kelompok yang menang, masing- masing team akan mendapat sertifikat atau hadiah apabila rata- 5. PENDAHULUAN Team Recognize (Penghargaan Kelompok) Guru kemudian mengumumkan kelompok yang menang, masing- masing team akan mendapat sertifikat atau hadiah apabila rata- 95 Jurnal Sekolah (JS). Vol 1 (1) Desember 2016, hlm 93-102 METODE tindakan pendahuluan yang berupa identifikasi permasalahan. Prosedur tersebut banyak diacu oleh guru dalam melaksanakan PTK dengan memuat bagan sebagai berikut: METODE Lokasi penelitian ini dilakukan di SD Negeri 058355 Aman Damaijalan Kwala Sawit, Kecamatan Batang Serangan. Pembelajaran yang diterapkan selama pengambilan data di kelas IV- A SD Negeri 058355 Aman Damai adalah Bahasa Indonesia. Waktu penelitian ini mulai Februari sampai dengan Mei Tahun 2016. Subjek dalam penelitian ini sebanyak I (satu) kelas yaitu kelas IV-A SD Negeri 058355 Aman Damai sebanyak 22 siswa.Alat pengumpul data dalam penelitian ini adalah instrument tes hasil belajar siswa dan instrumen aktivitas belajar siswa. GambarSpiral Tindakan Kelas Identifikasi Masalah Tindakan Revisi Perencana an Perencanaa n Refleksi Observasi Refleksi Observasi Tindakan Penelitian ini menggunakan Penelitian Tindakan Kelas (PTK). PTK adalah suatu bentuk kajian yang bersifat reflektif oleh pelaku tindakan yang dilakukan untuk meningkatkan kemantapan rasional dari tindakan mereka dalammelaksanakan tugas, memperdalam pemahaman terhadap tindakan-tindakan yang dilakukan itu, serta memperbaiki kondisi dimana praktek pembelajaran tersebut dilakukan. GambarSpiral Tindakan Kelas Analisis data yang digunakan pada penelitian ini digunakan analisis data deskriptif dengan membandingkan hasil belajar siswa sebelum tindakan dengan hasil belajar siswa setelah tindakan. Langkah-langkah pengolahan data sebagai berikut: Langkah-langkah pengolahan data sebagai berikut: Penelitian ini menggunakan model penelitian tindakan dari Kemmis dan Taggart yang berbentuk spiral dari siklus yang satu ke siklus yang berikutnya. Setiap siklus meliputi planning (rencana), action (tindakan), observation (pengamatan), dan reflection (refleksi).Langkah pada siklus berikutnya adalah perencanaan yang sudah direvisi, tindakan, pengamatan, dan refleksi.Sebelum masuk pada siklus 1 dilakukan 1. Merekapitulasi nilai pretes sebelum tindakan dan nilai tes akhir siklus I dan siklus II. 2. Menghitung nilai rata-rata atau persentase hasil belajar siswa sebelum dilakukan tindakan dengan hasil belajar setelah dilakukan tindakan pada siklus I dan siklus II untuk mengetahui adanya peningkatan hasil belajar. 3. 3. Penilaian p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 96 Esra Yunila, Meningkatkan Hasil Belajar ... a. Untuk penilaian aktivitas digunakan rumus sebagai berikut: Setelah data aktivitas siswa terkumpul sesuai dengan jumlah kegiatan belajar mengajar, maka data tersebut disusun kemudian data tersebut dirubah menjadi data prosentase. Untuk menganalisis data-data tersebut kemudian dianalisis dengan rumus sebagai berikut: % 100 ideal skor jumlah diperoleh skor jumlah %   Aktivitas PEMBAHASAN Penelitian ini dilaksanakan dalam dua siklus. Masing-masing siklus dilaksanakan dua kali pertemuan, pertemuan digunakan untuk pembahasan materi dengan alokasi waktu 2x40 menit, dan sebagian pertemuan akhir siklus digunakan untuk evaluasi dengan alokasi waktu 20 menit. Hal ini disesuaikan dengan jadwal pelajaran Bahasa Indonesia kelas IV. Sebelum dilaksanakan Siklus I dilakukan uji awal untuk menjajaki kemampuan awal siswa. Diperoleh hasil dengan rata-rata 50,5 dan nilai terendah 30 dan tertinggi 80, dengan KKM 75. b. Data nilai hasil belajar (kognitif) diperoleh dengan menggunakan rumus: b. Data nilai hasil belajar (kognitif) diperoleh dengan menggunakan rumus: Siklus pertama diawali dengan perencanaan penelitian yang meliputi pembuatan perangkat pembelajaran seperti : 100 soal seluruh Jumlah benar jawaban Jumlah Siswa Nilai   100 soal seluruh Jumlah benar jawaban Jumlah Siswa Nilai   c. Nilai rata-rata siswa dicari dengan rumus sebagai berikut: Keterangan : = Nilai rata-rata Σ= Jumlah nilai X N = Jumlah peserta tes N X X   X c. Nilai rata-rata siswa dicari dengan rumus sebagai berikut: N X X   1. Rencana Pelaksanaan Pembelajaran (RPP) 1 dan 2, Pembelajaran (RPP) 1 dan 2, 2. Lembar Kerja Siswa (LKS) 1 dan 2, Keterangan : 3. Alat bantu pembelajaran (gambar huruf), = Nilai rata-rata X 4. Lembar observasi aktivitas siswa, 5. Soal tes hasil belajar siswa. d. Ketentuan persentase ketuntasan belajar kelas % 100 K Sb kelas belajar Ketuntasan   d. Ketentuan persentase ketuntasan belajar kelas % 100 K Sb kelas belajar Ketuntasan   d. Ketentuan persentase ketuntasan belajar kelas % 100 K Sb kelas belajar Ketuntasan   Siklus I dilaksanakan selama 2 x pertemuan.Setiap pertemuan pembelajaran diterapkan model pembelajaran TGT. Selama menerapkan model pembelajaran TGTdilakukan pengamatan terhadap aktivitas siswa. ΣSb = Jumlah siswa yang mendapat nilai ≥ 75 ΣSb = Jumlah siswa yang mendapat nilai ≥ 75 ΣK = Jumlah siswa dalam sampel Data aktivitas belajar siswa siklus I ini menunjukkan bahwa aktivitas belajar membaca , mengerjakan LKS, bertanya pada p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 97 Jurnal Sekolah (JS). Vol 1 (1) Desember 2016, hlm 93-102 dan menganalisa kelemahan- kelemahan yang ada dalam pembelajaran menerapkan LKS melalui model pembelajaran TGT. Berdasarkan beberapa hal tersebut di atas maka dapat disimpulkan hasil dari refleksi Siklus I antara lain: guru dan teman masih kurang karena hanya mencapai 24% saja sementara aktivitas belajar yang tidak relevan dengan pembelajaran mencapai 17% yang menunjukkan suasana belajar siswa yang tidak kondusif. Dengan demikian diharapkan pada siklus II aktivitas belajar siswa lebih baik dari siklus I.  Temuan positif  Temuan positif a) Melalui penggunaan model pembelajaran TGT ini siswa terlihat lebih bergairah dalam belajar. Akhir Siklus I dilakukan tes hasil belajar atau disebut Formatif I, dengan data dapat dilihat Pada Tabel 1. Merujuk pada kesimpulan ini guru sebagai peneliti berusaha memperbaiki proses dan hasil belajar siswa melaluimodel pembelajaran TGT. b. Data nilai hasil belajar (kognitif) diperoleh dengan menggunakan rumus: Hasil belajar kognitif siswa yang diperoleh pada Siklus I selama dua pertemuan disajikan dalam tabel berikut: b) Dalam berdiskusi dan tanya jawab siswa terlihat mulai aktif, karena siswa diberi tanggung jawab untuk mengungkapkan pendapatnya.  Temuan negatif a) Sebagian siswa masih merasa malu-malu dalam mengungkapkan pendapatnya sehingga enggan untuk maju ke depan kelas mempresentasikan hasil kerjanya Tabel Distribusi Hasil Formatif I Nilai Frekuansi Persentase Rata-Rata 20 0 0% 40 3 14% 60 11 50% 80 5 23% 100 3 14% Jumlah 22 100% 67,3 b) Kualitas tanya jawab atau pendapat siswa belum maksimal, hal ini karena siswa-siswa tertentu yang selama ini pasif dalam pembelajaran agak kesulitan mengikuti alur pembelajaran dimana seperti tidak ada pendapat yang bisa disampaikan Merujuk pada Tabel tersebut, nilai terendah Formatif I adalah 20 dan tertinggi adalah 100. Merujuk pada KKM sebesar 75 maka hanya 8 dari 22 orang siswa mendapat nilai ketuntasan dan ketuntasan di dalam kelas belum tercapai. Sehingga pada siklus I ini penguasaan kompetensi siswa belum tercapai. c) Guru sendiri belum terbiasa dalam penggunaan model pembelajaran TGT sehingga pengambilan tindakan untuk mengatasi kesulitan siswa dalam pembelajaran tidak Data pada tabel dijadikan sebagai pemikiran bagi guru untuk mengevaluasi proses pembelajaran p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 98 Esra Yunila, Meningkatkan Hasil Belajar ... dapat langsung dilakukan oleh guru hingga menunggu refleksi yang dilakukan untuk siklus I. namun diberikan tindakan perbaikan dalam pembelajaran seperti yang telah disusun dalam perencanaan.Pengintegrasian tindakan perbaikan dilakukan pada kegiatan inti pembelajaran.Harapannya adalah aktivitas belajar siswa mengalami perbaikan dari siklus sebelumnya.Pengamatan terhadap aktivitas ini dilakukan oleh pengamat selama kerja kelompok kooperatif. Dari paparan deskripsi penelitian tindakan kelas siklus I, maka di dalam refleksi diupayakan perbaikan untuk meningkatkan proses pembelajaran dan aktivitas belajar siswa pada Siklus II sesuai dengan hasil refleksi. Data hasil observasi Siklus II disajikan dalam tabel.Merujuk pada tabel tersebut, terjadi perubahan aktivitas belajar siswa dibandingkan Siklus I karena perubahan yang terjadi cukup signifikan. Kegiatan menulis dan membaca dan mengerjakan LKS mendominasi dengan persentase 27% dan 32%, disusul kegiatan bertanya pada teman sebesar 19%, kemudian bertanya pada guru 17%. Hal ini menunjukkan bahwa kegiatan pembelajaran telah lebih banyak melibatkan siswa belajar yang artinya aktivitas belajar siswa meningkat terlihat dari menurunnya aktivitas bertanya pada guru dan naiknya aktivitas bertanya pada teman.Hasil ini juga didukung dengan menurunnya aktivitas siswa yang tidak relevan menjadi 5%. p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 b. Data nilai hasil belajar (kognitif) diperoleh dengan menggunakan rumus: Siklus kedua ini semua kegiatan tetap sama seperti pada Siklus I, hanya saja materi yang disampaikan berbeda dan dilakukan perbaikan kelemahan-kelemahan pada Siklus I. Tindakan perbaikan yang akan dilakukan pada Siklus II adalah sebagai berikut: a) Membantu siswa beradaptasi dengan alur pembelajaran, dimana setiap pendapat siswa dihargai dengan pujian ”bagus” atau meminta siswa lain bertepuk tangan. b) Untuk membantu siswa yang kesulitan merumuskan dan memfokuskan pembicaraannya maka di tampilkan gambar yang berhubungan dengan materi pembelajaran, sehingga sambil mengungkapkan pendapatnya siswa dapat melihat gambar yang dipasang guru. Akhir kegiatan belajar mengajar pada siklus II dilakukan tes hasil belajar atau disebut Formatif II.Datanya dapat dilihat pada Tabel berikut ini. c) Guru menganalisis kemungkinan- kemungkainan kesulitan siswa dalam Siklus II dan segera merencanakan tindakan yang dapat dilakukan langsung dalam pembelajaran. Pembelajaran dilaksanakan dengan langkah seperti Siklus I p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 99 Jurnal Sekolah (JS). Vol 1 (1) Desember 2016, hlm 93-102 Tabel Distribusi Hasil Formatif II Nilai Frekuansi Persentase Rata-Rata 20 0 0% 40 0 0% 60 2 9% 80 17 77% 100 3 14% Jumlah 22 100% 80,9 aktivitas siswa yang membaik serta hasil belajar siswa selama pelaksanaan proses belajar mengajar sudah berjalan dengan baik. Maka tidak diperlukan revisi terlalu banyak, tetapi yang perlu diperhatikan untuk tindakan selanjutnya adalah memaksimalkan dan mempertahankan apa yang telah ada dengan tujuan agar pada pelaksanaan proses belajar mengajar selanjutnya penerapan model pembelajaran TGT dapat meningkatkan proses belajar mengajar sehingga tujuan pembelajaran dapat tercapai dengan maksimal. aktivitas siswa yang membaik serta hasil belajar siswa selama pelaksanaan proses belajar mengajar sudah berjalan dengan baik. Maka tidak diperlukan revisi terlalu banyak, tetapi yang perlu diperhatikan untuk tindakan selanjutnya adalah memaksimalkan dan Merujuk pada Tabel tersebut, nilai terendah Formatif II adalah 60 dan tertinggi adalah 100. Merujuk pada KKM sebesar 75 maka 20 dari 22 orang siswa mendapat nilai ketuntasan atau ketuntasan klasikal tercapai sebesar 91%. Meningkatnya hasil belajar siswa ini sejalan dengan telah meningkatnya aktivitas belajar siswa. Merujuk dari penjabaran di atas peningkatan kualitas aktivitas belajar ditunjukkan dengan perubahan aktivitas Siklus I ke Siklus II.Dari tabel 1 dan table 2rata-rata aktivitas menulis dan membaca mengalami perubahan dari proporsi 38% menjadi 27%. Aktivitas mengerjakan dalam diskusi naik dari 24% menjadi 32%. b. Data nilai hasil belajar (kognitif) diperoleh dengan menggunakan rumus: Dari hasil observasi oleh dua orang pengamatdengan menerapkan modelpembelajaran Kooperatif Tipe Team Game Turnamentdiperoleh rata rata aktivitas belajar siswa pada siklus I dan siklus II menunjukan aktivitas mengerjakan LKS meningkat dari 24% menjadi 32%, bertanya kepada teman meningkat dari 10% menjadi 19% dan bertanya kepada guru meningkat dari 10% menjadi 17%. Sedangkan aktivitas yang tidak diharapkan seperti menulis dan membaca turun dari 38% menjadi 27% dan aktivitas yang tidak relevan dengan KBM turun dari 17% menjadi 5%. Peningkatan ini dikarenakan siswa sudah tidak cangung dan semakin tertarik dengan pembelajaran setelah model pembelajaran tersebut diterapkan 2. Dari hasil observasi oleh dua orang pengamatdengan menerapkan modelpembelajaran Kooperatif Tipe Team Game Turnamentdiperoleh rata rata aktivitas belajar siswa pada siklus I dan siklus II menunjukan aktivitas mengerjakan LKS meningkat dari 24% menjadi 32%, bertanya kepada teman meningkat dari 10% menjadi 19% dan bertanya kepada guru meningkat dari 10% menjadi 17%. Sedangkan aktivitas yang tidak diharapkan seperti menulis dan membaca turun dari 38% menjadi 27% dan aktivitas yang tidak relevan dengan KBM turun dari 17% menjadi 5%. Peningkatan ini dikarenakan siswa sudah tidak cangung dan semakin tertarik dengan pembelajaran setelah model pembelajaran tersebut diterapkan 2. b. Data nilai hasil belajar (kognitif) diperoleh dengan menggunakan rumus: Aktivitas bertanya pada teman naik dari 10% menjadi 19%.Dari ketiga aktivitas ini menunjukkan bahwa aktivitas siswa dalam belajar membaca mengalami peningkatan.Aktivitas bertanya kepada guru naik dari 10% menjadi 17% yang mengindikasikan siswa dapat konstruktif dalam berdiskusi dan saling membantu dalam belajar membaca. Aktivitas yang tidak relevan dengan kegiatan belajar mengajar turun dari 17% menjadi 5%.Dengan demikian aktivitas belajar siswa mengalami perbaikan dari siklus I ke siklus II. Merujuk dari penjabaran di atas peningkatan kualitas aktivitas belajar ditunjukkan dengan perubahan aktivitas Siklus I ke Siklus II.Dari tabel 1 dan table 2rata-rata aktivitas menulis dan membaca mengalami perubahan dari proporsi 38% menjadi 27%. Aktivitas mengerjakan dalam diskusi naik dari 24% menjadi 32%. Aktivitas bertanya pada teman naik dari 10% menjadi 19%.Dari ketiga aktivitas ini menunjukkan bahwa aktivitas siswa dalam belajar membaca mengalami Beberapa hal yang dapat dicatat dalam refleksi pembelajaran Siklus II adalah sebagai berikut: a) Siswa mulai aktif dalam diskusi dengan ditunjukkan oleh hasil observasi aktivitas belajarnya yang sedikit lebih baik dari pada Siklus I. b) Ketuntasan hasil belajar siswa meningkat dari 64% atau belum berhasil menjadi 91% atau dalam ketogori berhasil. c) Sikap konstruktif siswa menunjukkan respon yang tinggai pada penerapan model pembelajaran TGT. peningkatan.Aktivitas bertanya kepada guru naik dari 10% menjadi 17% yang mengindikasikan siswa dapat konstruktif dalam berdiskusi dan saling membantu dalam belajar membaca. Aktivitas yang tidak relevan dengan kegiatan belajar mengajar turun dari 17% menjadi 5%.Dengan demikian aktivitas belajar siswa mengalami perbaikan dari siklus I ke siklus II. d) Siswa mulai terbiasa mengungkapkan pendapatnya terlihat dari aktivitas belajar siswa dalam bertanya pada teman yang cukup dominan Siklus II guru telah menerapkan model pembelajaran TGTdengan baik dan dilihat dari nilai p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 100 Esra Yunila, Meningkatkan Hasil Belajar ... Penerapan model pembelajaran TGT dapat meningkatkan hasil belajar siswa, dilihat dari nilai rata-rata sebelum penerapan model pembelajaran TGT yaitu berupa nilai pretes adalah 50,5 dengan ketuntasan belajar yang dicapai 14%, setelah penerapan model pembelajaran TGT nilai siswa mengalami peningkatan. Berdasarkan hasil tes pada Siklus I, nilai rata-rata hasil belajar yang dicapai siswa adalah 67,3 dengan persentasi 64%. Setelah dilaksanakan Siklus II, maka hasil belajar siswa berdasarkan data Formatif II adalah rata-rata 80,9 dengan ketuntasan klasiklal mencapai 91%. ketuntasan hasil belajar siswa terjadi karena meningkatnya aktivitas belajar siswa serta siswa sudah tidak cangung dan semakin tertarik dengan pembelajaran setelah modelpembelajaran Kooperatif Tipe Team Game Turnamentditerapkan. p . KESIMPULAN Setelah data-data tes hasil belajar, dan aktivitas belajar siswa terkumpul kemudian dianalisis sehingga dapat disimpulkan antara lain: 1. Berdasarkan data penelitian setelah diterapkanmodelpembelajaran Kooperatif Tipe Team Game Turnament dan dianalisis sebanyak 12 orang siswa hasil belajarnya meningkat.Hasil analisis data menunjukan Siklus I rata rata nilai Formatif I adalah 67,3 dengan ketuntasan individu sebanyak 8 orang siswa sementara secara kelas belum tuntas. Siklus II rata rata nilai siswa Formatif II adalah 80,9 dengan 20 orang siswa tuntas secara individu dan secara kelas tuntas. Peningkatan dan 1. Berdasarkan data penelitian setelah diterapkanmodelpembelajaran Kooperatif Tipe Team Game Turnament dan dianalisis sebanyak 12 orang siswa hasil belajarnya meningkat.Hasil analisis data menunjukan Siklus I rata rata nilai Formatif I adalah 67,3 dengan ketuntasan individu sebanyak 8 orang siswa sementara secara kelas belum tuntas. Siklus II rata rata nilai siswa Formatif II adalah 80,9 dengan 20 orang siswa tuntas secara individu dan secara kelas tuntas. Peningkatan dan 1. Berdasarkan data penelitian setelah diterapkanmodelpembelajaran Kooperatif Tipe Team Game Turnament dan dianalisis sebanyak 12 orang siswa hasil belajarnya meningkat.Hasil Dari hasil yang diperoleh dalam penelitian ini, maka ada beberapa saran yang diajukan yaitu: 1. Bagi guru, tutor maupun peneliti berikutnya yang ingin menerapkan modelpembelajaran Kooperatif Tipe Team Game Turnamentdalam kegiatan belajar mengajar hendaknya melakukan pembagian kelompok dengan p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295 101 Jurnal Sekolah (JS). Vol 1 (1) Desember 2016, hlm 93-102 Slavin. E. Robert. (2005). Cooprative LearningTeori; Riset dan Praktik.Bandung: Nusa Media. (2005). 2. Bagi peneliti selanjutnya hendaknya lebih memahamimodelpembelajaran Kooperatif Tipe Team Game Turnamentsebagai salah satu upaya untuk mengaktifkan siswa belajar, meningkatkan aktivitas belajar siswa, serta meningkatkan hasil belajar siswa 2. Bagi peneliti selanjutnya hendaknya lebih memahamimodelpembelajaran Kooperatif Tipe Team Game Turnamentsebagai salah satu upaya untuk mengaktifkan siswa belajar, meningkatkan aktivitas belajar siswa, serta meningkatkan hasil belajar siswa DAFTAR RUJUKAN Arikunto, S., (2002), Prosedur Penelitian Suatu Pendekatan Praktek, Rineka Cipta, Jakarta. Aqib, Zainal. (2006), Penelitian Tindakan Kelas. Yrama Widya, Bandung. Istarani, (2011), Model Pembelajaran Inovatif (Referensi Guru Dalam Menentukan Model Pembelajaran). Media Persada, Medan. Joyce, B., Weil, M., dan Calhoun, (2009), Models of Teaching Edisi Kedelapan, Yogyakarta: Penerbit Pustaka Belajar. Joyce, B., Weil, M., dan Calhoun, (2009), Models of Teaching Edisi Kedelapan, Yogyakarta: Penerbit Pustaka Belajar. Sani, R.A., dan Sudiran, (2012), Meningkatkan Profesionalisme Guru Melalui Penelitian Tindakan Kelas, Citapustaka Media Perintis, Bandung 102 103 p-ISSN: 2355 - 1739 e-ISSN: 2407 - 6295
https://openalex.org/W3165133424	https://ieeexplore.ieee.org/ielx7/6287639/9312710/09439934.pdf	English	null	How Do Students Behave in a Gamified Course?—A Ten-Year Study	IEEE access	2,021	cc-by	15,666	Received April 13, 2021, accepted May 9, 2021, date of publication May 24, 2021, date of current version June 10, 2021. Digital Object Identifier 10.1109/ACCESS.2021.3083238 AMIR HOSSEIN NABIZADEH , JOAQUIM JORGE , (Senior Member, IEEE), SANDRA GAMA, AND DANIEL GONÇALVES INESC-ID, Instituto Superior Técnico, University of Lisbon, 1649-004 Lisbon, Portugal Corresponding author: Amir Hossein Nabizadeh (amir.nabizadeh@tecnico.ulisboa.pt) This work was supported in part by the National Funds through the Fundação para a Ciência e a Tecnologia (FCT) under Project UIDB/50021/2020, and in part by the Project GameCourse, Portugal, under Grant PTDC/CCI-CIF/30754/2017. This work was supported in part by the National Funds through the Fundação para a Ciência e a Tecnologia (FCT) under Proje UIDB/50021/2020, and in part by the Project GameCourse, Portugal, under Grant PTDC/CCI-CIF/30754/2017. ABSTRACT Gamiﬁed learning aims to motivate students using game elements. Although gamiﬁcation can enhance students’ enjoyment and engagement, it is unclear how different students behave in and interact with gamiﬁed contexts. To this end, we analyze how different students interact with a gamiﬁed course. We devised such an experimental course on Multimedia Content Production (MCP), and ran it for ten years. At each year, we modiﬁed it after students’ feedback from the previous year. We determined student groups applying clustering techniques to learner performance data, independently analyzed the resulting clusters in terms of behavior, engagement, performance, and also compared those pairwise. Our analysis identiﬁed four different student groups (proﬁles/clusters) according to their performance and interactions with the course across all years. We found out that the best performing students were those that had signiﬁcantly more interactions with course materials and consistently ranked highest. In addition, we found that performance indicators for students of all groups became stable within the ﬁrst month after course start, allowing ﬁnal grades to be predicted with high accuracy by then. Furthermore, all were deadline driven and became mainly active at the end of the semesters (indicating a lack of self-regulation skills). Moreover, we did not ﬁnd any speciﬁc relation between students’ groups and gaming proﬁles (Brainhex categories). Finally, we propose practical implications and guidelines for designing compelling gamiﬁed learning experiences. INDEX TERMS Brainhex categories, gamiﬁcation, gameful learning, student behavior, clustering. NDEX TERMS Brainhex categories, gamiﬁcation, gameful learning, student behavior, clustering. under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ VOLUME 9, 2021 is work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ II. RELATED WORK Different educational methods are introduced to support students and improve their learning activities toward their success. In that regard, several approaches aimed at enhanc- ing students’ performance and engagement with courses using the gameful learning. In [41], authors intended to boost the students’ motivation and engagement by gamifying a Moodle-based course about Uniﬁed Modeling Language (UML). This course was organized in ten levels considering its original curriculum. Levels were classiﬁed into syntax and semantic levels. Students were rewarded 100 and 200 points for completing tasks within syntax and semantics levels, respectively. The accumulated points of students were used to estimate their levels, and a level could be completed by obtaining the minimum required points for that level. When- ever a student could complete a level, he/she could get a new badge. The effectiveness of this course was evaluated in a single semester with 22 students in autumn 2017. The results showed that the gameful learning was successful and students were satisﬁed with the course and found it useful. RQ7 How does the ability to self-regulate vary between different student groups? Furthermore, while other researchers evaluate the effec- tiveness of their gamiﬁed courses in a single iteration, they fail to assess the consistency of obtaining similar results over the years. In this paper, we present a longitudinal study on a gamiﬁed course over a period of ten years: Multimedia Content Pro- duction (MCP). This is a MSc-level course, offered to stu- dents of the Computer Science Engineering MSc programme at Instituto Superior Técnico, the Engineering school of the University of Lisbon. It focuses on the digital representation and manipulation of different types of media (images, audio, video, etc.), but also on their creation and editing. It was gam- iﬁed for the ﬁrst time in the 2010-2011 school year, with the inclusion of game elements such as a leaderboard, experience points and levels, and achievements. The course was there- after revised annually, based on an analysis of the pedagogical outcomes from previous instances and student feedback. This led, over the years, to the inclusion and removal of other game elements (quests, a skill tree, etc.) and, more importantly, it gave us thorough and longitudinal insights on how different students react to a gamiﬁed learning experience. In another paper, authors studied the experience of undergraduate students’ with a gamiﬁed course about social networking technologies [23]. II. RELATED WORK This course was already devel- oped and in-used in a public university in the Northwest United States, and was designed for freshman students with no background on the subject. It included Experience Points (XP), a leaderboard, badges, and levels. In this course, all stu- dents’ activities were assigned a certain XP, and students were graded using their accumulated XP throughout the course. They leveled up by accomplishing a certain amount of XP and got minor rewards by collecting badges. The top 10 students (i.e. got the highest XP) were listed on the leaderboard. For the evaluation, a survey was carried out with 139 students. It was conducted at three times during the course (beginning- middle-end). The results conﬁrmed that the students had a positive inﬂuence about gameful learning on their achieve- ments, learning, and engagement with the course. In order to analyze how different students performed and interacted with the course, we applied cluster analysis at the end of each academic year. We also conducted a cluster-wise assessment to evaluate student behavior, engagement, and performance within each cluster independently. We later compared each cluster to other clusters of the same and other years. According to the students’ performance and behavior, we could mainly identify four student groups across all years. Our main contributions are as follows: Our main contributions are as follows: • Distinguishing and describing different students groups considering their performance and behavior. • Analyzing the achievements, performance, engagement, interactions and characteristics of each group and com- pare it to others. Aleksic-Maslac et al. used a tool called Kahoot [26] to create fun quizzes for an ICT course at the Zagreb School of Economics and Management (ZSEM) [2]. During a semester, six Kahoot quizzes were conducted, and each student got a total point by answering the questions of each quiz. At the end of the semester, both professors and students were asked to state their level of satisfaction. The results showed that using Kahoot was an efﬁcient method to motivate students for higher engagement. • Identifying how early we can predict each student’ group and performance within a reliable accuracy. • Analyzing whether student gaming proﬁles (Brainhex categories) have any speciﬁc relation with their groups. • Suggesting substantial instructions for designing a gam- iﬁed course. The remainder of this paper is structured as follows: Section 2 introduces related work. Section 3 explains the research methodology of this study. I. INTRODUCTION different mechanisms to inspire students by providing joyful and fun game-playing. Game-like mechanisms encourage students to engage more with courses and perform different learning activities to earn more rewards (e.g., points, lev- els, experience points). Doing different learning tasks boosts students’ learning and enhances their critical, collaboration, communication, and problem-solving skills [29]. Gamiﬁcation applies game design elements to non-game contexts [28] to increase users’ enjoyment, and subsequently boost their motivation and enhance their engagement within a gamiﬁed context [14], [44], [64]. Gamiﬁcation has been used in various contexts with different objectives, ranging from raising health awareness [17], teaching how to drive [33], to improving engagement with a course [10], [48], and enhancing enterprise risk management Education is one con- text that is widely explored by researchers. Although gamiﬁcation has shown promising results and studies demonstrate that it can enhance student performance, [32] and engagement [27], it is still not well explored how different students adjust to a gamiﬁed course and interact with it. Therefore, we intend to answer the following questions in our study: Despite innovative educational approaches, the traditional and current educational methods are often considered boring and ineffective by students [29]. Hence, the main challenges of these methods are about enhancing students’ motivation and engagement within a course [44]. Alternatively, gamiﬁed educational approaches attempt to tackle these challenges. These have a noticeable motivational power since they use RQ1 How many groups of students engage in a gamiﬁed course? RQ1 How many groups of students engage in a gamiﬁed course? RQ2 How do students of different groups collect experience points (XP), and badges? The associate editor coordinating the review of this manuscript and approving it for publication was Vlad Diaconita . RQ3 How do these groups differ, considering average ranks and ﬁnal grades? 81008 81008 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? RQ4 Are students deadline-driven regardless of group, or are they steadily engaged with the course throughout the semester? RQ4 Are students deadline-driven regardless of group, or are they steadily engaged with the course throughout the semester? RQ4 Are students deadline-driven regardless of group, or are they steadily engaged with the course throughout the semester? groups and the data used for that are detailed in Section 5. In Section 6, the groups’ behavior was analyzed and com- pared. I. INTRODUCTION Section 7 highlights some indications to design a more effective gamiﬁed course, while Section 8 presents our conclusion. RQ5 When does the performance of various groups become stable? RQ5 When does the performance of various groups become stable? RQ6 Is there a speciﬁc relation among different student groups and their gaming proﬁles (Brainhex cate- gories [55])? VOLUME 9, 2021 II. RELATED WORK Here, the gameful learning was used to sustain students’ motivation, and consequently, enhance their effectiveness in learning activities. During this course, stu- dents were engaged in a variety of learning activities and played with several SG. They received various topics with different level of difﬁculties in games, game-play, and home assignments. This course was carried out in Italy, Spain, and Netherlands, and it was concluded that a single game cannot be used to achieve all the gameful learning objectives (e.g. enhancing engagement, motivation, and performance). For that, more than one single game needs to be adopted in each course. g g g Pan et al. developed gamiﬁed forensic modules with intuitive designs and interactive dialogues [57]. They were designed for students with no related background to learn fundamental digital forensic content and explore the forensics procedures and technologies using interactive games. Here, the gameful learning was applied to keep students interested and engaged with the modules. Modules were implemented in a real computing environment with access to real forensics tools and evidence for enabling students to practice with forensics technologies. By the time of this paper, these mod- ules were still un-evaluated. The authors intended to measure the effectiveness of their modules in the summer and fall semesters of 2015 at their institution and two other partners’ colleges. The effect of gamiﬁcation on students’ engagement and performance was studied in [21]. This study included the results of participants in a one-term ICT course enrolled at a school of education. Participants were interviewed to examine the relationships among gamiﬁcation, engagement, and achievement. The results presented that gamiﬁed ele- ments had a positive motivational impact on engagement and indirectly affected the academic achievement. Another study examined participants’ willingness to join gamiﬁed activities where rewards were not directly tied to a course’s grades [4]. For that, over two semesters, an optional gaming activity was added in ﬁve sections of a course (experimental group), and four sections acted as a control group. The ﬁndings (collected by pre-, mid-, and post-surveys) presented an important difference between experimental and control groups regarding hours spent on gamiﬁed activities. In [22], authors analyzed the effect of gamiﬁcation on students of a management classroom. In this study, 44 elementary students attended a gamiﬁed management classroom (experimental group) and 42 attended a traditional classroom (control group). II. RELATED WORK Section 4 presents the elements of our gamiﬁed course as well as its evolution through the years. Our method to identify the students’ Another gamiﬁed course experiment was presented in [16]. This course was a short one (lasted one month) about promot- ing entrepreneurship among B.Sc., M.Sc., and Ph.D. students in the ﬁeld of Electronic Engineering at the University of 81009 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? Genoa. It was relied on using three Serious Games (SG) [1] for business, which were: Hos Shot Business (HSB), Enter- prise Game (EG), and SimVenture (SV). A range of topics was covered in this course having different level of difﬁcul- ties in games, game-play, and home assignments. Like other studies, the students’ ranks were determined considering their obtained cumulative scores. To evaluate this course, 34 stu- dents were divided into 16 teams. The results highlighted that their overall engagement with the course was high. Also, teachers had positive opinions about course acceptance and knowledge acquisition of the students. a survey with 18 participants was conducted for a quantitative evaluation, and a series of interviews with six participants was performed for a qualitative assessment. The results indicated that students were positive about badges, but had mixed feedback about the use of leaderboards. An immersive gamiﬁed course was explained in [40]. It was presented in the academic year 2013-14, and was designed for twelve sessions. In each session, students (space school trainees) should activate one of the spaceship systems. Depending on the complexity of a system, it could include different number of missions. More complex systems had more short missions to complete. Completing these missions resulted in getting badges, and sessions could be passed by concluding missions and getting a certain amount of points. These points could be also obtained through assignments. The assignments were optional and students could select them as they like (harder assignments had more points). In this study, all missions, assignments, and achievements were transferred to Youtopia platform, which was designed for motivating and engaging students [66]. The same authors conducted another gamiﬁed course using similar methodology two years later. This course was to stimulate entrepreneurial and innovative mindsets of B.Sc, M.Sc, and PH.D. students of non-business faculties, and to provide them with several signiﬁcant operational and the- oretical skills [3]. II. RELATED WORK Then, the differences between the experi- mental and control groups were examined regarding their divergent thinking and creative tendency. The results showed that the verbal divergent thinking and creativity of the exper- imental group were enhanced in comparison with the control group. In [25], authors assessed the students’ perception of the impact of badges and leaderboards in their motivation towards an introductory Software Engineering course. Hence, g Another gamiﬁed e-learning course was introduced in [35]. It was designed based on a structural gameful learning [42], [53] for the bachelor’s degree at the University of Plov- div. The structural gameful learning was obtained by using various game elements. This course was arranged for ten weeks, and each week was corresponding to a level. Levels included quests and assignments (seven individual and one group assignments). When students completed the assign- ments, they were awarded by points, and getting enough points would allow them to go to the next level. In the last level, students should answer to a Constructivist On-Line Learning Environment Survey (COLLES) [6]. To analyze the effectiveness of the gamiﬁed course, it was presented together with a standard e-learning course, and the students were allowed to select their preferred one. Out of 113 students, 41 of them selected the gamiﬁed course and only 27 could complete it. Results showed that the students, who used the gamiﬁed course, were less confused and had higher grades and engagement with the course than the ones using the standard e-learning course. In addition, the understandability of students that used the gamiﬁed course was more than their expectation. 81010 VOLUME 9, 2021 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? Dicheva et al. introduced a gamiﬁed course to enhance students’ engagement with the course and encouraging them to do self-study [31]. It was designed for a data structure subject, and was using a OneUp learning platform [30], [50], [60]. This platform supports the application of game design principles and elements for a course. In this gamiﬁed course, the students were awarded using badges and a vir- tual currency. Badges were used to reward students for their performance while the virtual currency was used to reward their engagement with the course (e.g. attending classes). A quasi-experiment [20] was conducted for the evaluation. II. RELATED WORK For that, 16 students (from fall 2017) formed the control group while 11 students (from spring 2018) shaped the exper- imental group. Both groups used the OneUp platform, but the gameful learning features were only activated for the experi- mental group. The log data of this platform together with the students’ ﬁnal grades were used to analyze their interactions and performance. In addition, a survey was conducted among the students in the experimental group to assess the usefulness of the gamiﬁed course. The results showed that the students found the gamiﬁed course useful while it increased their engagement and reduced their failing rates. how students (most motivated by badges) differ from oth- ers in terms of behavior and achievement goal orientation. The results indicated no important differences in that regard. In [24] and [32], authors presented mixed results. These studies used social networking and gameful learning in an undergraduate course, and analyzed their inﬂuence on the students’ behavior, engagement, and performance. The eval- uation results stated that the students using this undergraduate course performed better in completing their assignments than the ones using a traditional e-learning course, but the students using the traditional course could learn more. In addition, students’ engagement and grades remained low using the gamiﬁed course. Regardless of having negative or positive impact on the students’ motivation, engagement, and performance, only a few studies investigated how different students behaved in a gamiﬁed learning environment. For example, in [19] authors examined the inﬂuence of different personalities traits and learning styles on students’ engagement, perception, and performance on the course. Another instance is [46] where Lavoué et al. assessed whether the adaption of gaming fea- tures based on a player model enhance students’ participation and motivation. This experiment was conducted once using 266 participants. One year later, Mbabu studied how students having different personality and learning styles interacted with an adaptive gameful learning tool that was designed for an e-learning platform [47]. This study was also evaluated using 158 students within a few weeks. Barata et al. also proposed a gamiﬁed M.Sc. course to improve students’ engagement and motivation [7]. It included six main game elements, which were, experience points (XP), a leaderboard, levels, Badges, challenges, and a skill tree. The skill tree was a precedence tree, where each node was a learning task that resulted in XP upon completion. Initially, six nodes were unlocked. 1www.moodle.org II. RELATED WORK Posterior nodes could be accessed if the anterior ones were completed. Students’ behavior and interactions with the course was analyzed, and authors could distinguish four group of students. The evaluation results showed that the course enhanced students’ performance and participation. Besides the minority of studies that analyzed how students behaved in a gamiﬁed environment, they also often evalu- ated their courses in a single semester with a few students while obtaining consistent results over the years with a large number of students is ignored. III. RESEARCH METHODOLOGY TABLE 1. Game elements and changes over the years. FIGURE 1. Research methodology. course. Every year, the mentioned sets of data were collected from the enrolled students in the course (statistics are men- tioned in Table 1). These datasets were analyzed using the R programming language at the end of each semester. To analyze the collected data, we initially used an elbow technique together with a k-means clustering algorithm to identify the number of groups among students and distinguish their performance using their average accumulated XP over a semester. Then, we described the groups and used cluster analysis to compare them in terms of obtained badges, XP, ﬁnal grades, and ranks. Their engagement and interaction with the course were then analyzed using their XP and logs data. At ﬁrst, these sets of data were converted to binary sets (i.e. if a group member did an action or collected an XP in a day, it was considered as one, otherwise zero). These binary sets were then presented in form of density and scatter graphs to show groups’ engagement and interaction with the course. Next, the stability of groups’ performance was assessed using their ranks (calculated using average accumulated XP) over a semester. To evaluate whether groups had any relation with gaming proﬁles, we initially determined gaming proﬁles of groups’ members using brainhex categories (questionnaires) and then examined differences between each pair of groups using a statistical hypothesis test (p-values). Finally, the self- regulation skills of students were assessed using the sum of XP and badges that they got within a month of a semester and compare them with other months from the same semester. FIGURE 1. Research methodology. we modiﬁed the course and added/removed/edited the course elements for the next iteration. After ten iterations of the course, we reached a set of design implications that can be helpful in the setting up of new gamiﬁed courses. They are mentioned in Section VII. A general view of our research methodology is presented in Figure 1. III. RESEARCH METHODOLOGY III. RESEARCH METHODOLOGY Although many studies shown that the gameful learn- ing had positive impact on the students’ motivation and engagement with the course, a few studies presented that the gameful learning had no or even negative inﬂuence on the students. For instance, in [38], students were analyzed to assess how gameful learning inﬂuenced their course engage- ment and behavior. To this end, their performance, moti- vation, and satisfaction were measured four times during a 16-week semester. For the ﬁnal evaluation, the students across two courses were tested. One course was a gamiﬁed one using badges and a leaderboard while the other one was a non-gamiﬁed one (having similar curriculum). The results showed that the students using the gamiﬁed course were less satisﬁed, motivated and engaged with the course. In addition, these students had lower ﬁnal grades in compar- ison with the ones using the non-gamiﬁed course. In [37], gamiﬁcation was examined in relation to achievement goal orientation (students’ preferences to various goals, outcomes, and rewards). For this purpose, achievement badges were added to a Data Structures and Algorithms course (N = 278), and students’ feedback with various achievement goal orien- tation proﬁles was assessed. Moreover, the authors analyzed To accomplish the main goal of this study, which is under- standing how students of different groups behave in a gamiﬁed course, we have applied the Educational Design Research (EDR) approach [59] along with a formative study. EDR is an iterative development of solutions for practi- cal educational issues. These solutions can be educational products, policies, or processes. EDR not only tries to solve important issues facing educational practitioners, but it simul- taneously discovers new knowledge that can inform the work of others facing similar issues. Besides EDR, formative stud- ies [36] were applied to gather data that could be helpful to understand students’ behavior, diagnose problems, and improve the course design. Using the mentioned research approaches, we annually collected students’ interaction logs (e.g. what learning mate- rials were seen by a student, how many posts he/she made, etc.) with a Learning Management System (LMS) used in the course (Moodle1). We also gathered students’ perfor- mance data (e.g. obtained XP, badges, etc.) from our gamiﬁed 81011 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? TABLE 1. Game elements and changes over the years. TABLE 1. Game elements and changes over the years. IV. GAMIFIED COURSE In this study, we gamiﬁed a course named Multimedia Con- tent Production (MCP), which is designed for MSc students in the ﬁeld of Information Systems and Computer Engineering. The main goal by gamifying the MCP was to provide enjoy- ment for the students to keep them motivated and engaged with the course. In MCP, students attend both theoretical lec- tures and practical labs. In theoretical lectures, students learn about different formats (audio, video, image, etc.) from an engineering standpoint (compression, formats, etc.) but also with an emphasis on the creation of aesthetic, impactful and high-quality media (in the Lab classes) [10]. Besides lectures and labs, students also join the discussions and complete online assignments via Moodle. Besides the aforementioned data and analysis, we also collected students’ feedback at the end of each semester using questionnaires. This feedback could be divided in two parts. The ﬁrst part included students’ opinions about the course itself, such as if it was engaging, motivating, creative, etc. The second part referred to the quality of the course elements, such as if they were interesting, or contributed for a greater workload. After assessing how much we were successful in engag- ing, motivating, and providing enjoyment for students, 81012 81012 VOLUME 9, 2021 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 3. Dashboard. FIGURE 4. MCP leaderboard. FIGURE 2. Badges and their levels. FIGURE 3 Dashboard FIGURE 2. Badges and their levels. FIGURE 3. Dashboard. FIGURE 4. MCP leaderboard. FIGURE 4. MCP leaderboard. MCP is presented synchronized and identical across two university campuses. It is offered in English and is only available in the second semester of each academic year. This course has three sessions per week (each session lasts 1.5 hour), two theoretical lectures and one lab, and consists of various activities, such as quizzes, a multimedia presenta- tion, lab assignments, and several other activities. Instead of receiving traditional grades, students earn XP for completing different course activities, including the traditional evaluation elements existing in the pre-gamiﬁed course (a multimedia presentation, lab assignments, ﬁnal exam or quizzes) and the game elements. Performing speciﬁc course activities, such as attending lectures, ﬁnding bugs in class slides, or completing challenges, can result in obtaining badges. IV. GAMIFIED COURSE A badge can be achieved by completing an activity that might require a sin- gle iteration or up to three, with each iteration being worth a speciﬁc amount of XP and a badge (Badges are shown in Figure 2). Some of the badges award extra credit, as they reﬂect desirable behaviors but that cannot be mandatory due to the school’s bylaws, and some award 0 XP and are there for bragging rights only. Getting a certain amount of XP results in achieving a new experience level. MCP includes 20 levels while 10 is the minimum level to pass the course. At the end of each semester, levels are converted to a 20-point grading system, which is the norm in our university. As means to com- municate progress and provide feedback to students, a leader- board (Figure 4) and a dashboard were used (Figure 3), which include students’ XP, levels, and badges. For assessing stu- dents, MCP uses a quantum evaluation mechanism where students are awarded (by XP, badges, and levels) for perform- ing any kind of course’s activities (grade is granularized). Quantum grading allows students to transparently trace their progress whenever they participate in the course, down to the each XP. This is a marked departure from conventional so-called continuous evaluation schemes, which break exams and project assignments into smaller units in a rather discrete fashion. Thi h b diﬁd ll id i th FIGURE 4. MCP leaderboard. achieving the MCP goals (enhancing students’ enjoyment, motivation and engagement), the course was modiﬁed and its elements were added/removed/edited. Various game elements are used during the years. The four main game elements are experience points (XP), levels, achievements and a leaderboard. In the ﬁrst year, achieve- ments were the only game element to obtain XP. For that, students needed to perform a set of tasks that we intended to encourage them to do, such as completing challenges, ﬁnding bugs in course slides, and attending lecturers. Besides getting XP, completing these achievements could result in getting badges. Some achievements awarded extra XP, which allowed students to obtain high grades without performing all mandatory tasks. There were ≈3000 extra XP but stu- dents could only earn up to 1000. This extra XP enabled us to reward desirable behaviors of the students that were not mandatory by the university’s laws. IV. GAMIFIED COURSE Another main element of the MCP is a leaderboard (an online webpage) that allows students to monitor their progress and compare it with others (Figure 4). This leader- board also enables the students to clearly see what has been completed so far and what needs to be done. As presented in Table 1, it was used in all years without signiﬁcant changes. Skill Tree is a game element that was added to the course in the second year (2011-12). This element is designed to make the students more autonomous (select activities as they like). It is a precedence tree where each node refers to a learn- ing activity that results in XP upon completion (Figure 5). Another main element of the MCP is a leaderboard (an online webpage) that allows students to monitor their progress and compare it with others (Figure 4). This leader- board also enables the students to clearly see what has been completed so far and what needs to be done. As presented in Table 1, it was used in all years without signiﬁcant changes. This course has been modiﬁed annually considering the results that we got from analyzing the students’ performance and interactions with the course and their feedback (collected by questionnaires) about the quality of the course and its elements. Depending on how much we were successful in Skill Tree is a game element that was added to the course in the second year (2011-12). This element is designed to make the students more autonomous (select activities as they like). It is a precedence tree where each node refers to a learn- ing activity that results in XP upon completion (Figure 5). 81013 VOLUME 9, 2021 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 5. Skill tree. tools introduced in the classes. Up to three percent of total XP could be obtained from it. Anyhow, Avatar World was not a successful attempt since it could not be well-integrated with the course and students had little to do there. In addition, we found relatively low levels of students’ interest (X = 2.2, σ = 0.08) and engagement (X = 2.12, σ = 0.03) with it. To this end, it was removed from the course since 2015-16. Table1 shows all of the game elements as well as the changes that are made into the course over the years. IV. GAMIFIED COURSE In this table, we also present the number of students who enrolled in the course and number of students who dropped out (with- drew from the course). Here, we see that a ﬁnal exam is replaced with quizzes since 2013-14. This change was made since we noticed that several students had a low participation during a semester and became active only at the end of the semester, as the deadline looms. These quizzes allow students to assess themselves on a (almost) weekly basis and balance their participation through a semester. FIGURE 5. Skill tree. Initially, ﬁve nodes are unlocked and subsequent nodes could be accessed if the anterior ones are completed [11]. In 2012-13, since students stated (in the satisfaction question- naire) that the course needed more work than the other ones, we reduced 5% of total course XP from the ﬁnal multimedia presentation and added it to the Skill Tree. p p g Course modiﬁcation and elements can directly inﬂuence the students’ performance (e.g. XP and grades). In Figure 6, we show whether the students’ XP and grades were changed over the year. We can see some variations through the years, especially since 2015-16. Up to this year, MCP had a rigid major\minor structure. So, to take MCP students needed to enroll in the Multimedia minor or major, and consequently, would be enrolled in other courses in the area. It implied that the students taking MCP were highly interested in it. Since 2015-16 and by restructuring the MCP, students could enroll in any course they like, and their reasons for that might vary, from deep interest to a perception of easiness. Figure 6 presents that the median of XP and grades were slightly decreased after 2015-16. Furthermore, the 25th percentile and minimum value were reduced more drastically, which could be due to the course restructuring. In 2011-12, XP and grades are noticeably higher than the other years. It happened since the initial version of the Skill Tree was added to the course without having any limitation on its XP. So, if the students worked more, they could get more XP from the Tree (and subsequently could get higher grades). To avoid this situation, after 2011-12, the amount of XP that students could get from the Tree was limited to a speciﬁc amount. Skill Points were another game element that were only used in the second year. IV. GAMIFIED COURSE They were introduced to enhance the ﬂexibility of students for selecting their preferred paths through the Skill Tree. Each type of course media (i.e. image, text, audio) had a different Skill Point. Skills of the Tree were unblocked by obtaining a certain number of points on related media. Anyhow, we were not successful in introduc- ing these points due to lack of time management by students. As mentioned, these points were only used in 2011-12 and were dropped after that year. It needs to be noted that these points are different from XP (experience points), which can be obtained from various course activities. Also, these points are considered as XP in our analysis for 2011-12. Quest was used for ﬁve years (2012-13 to 2016-17) to promote the collaboration toward a common goal among students. It was an online riddle where students should start by manipulating a multimedia content to ﬁnd a URL for the next clue of the riddle. To enhance students’ participation, they should contribute at least once to obtain the XP, and their contributions were posted in the forums and graded by the professors. Here, the amount of obtained XP was proportional to the accomplished Quest level and number of active participants. The Quest was dropped after ﬁve years since the students’ feedback (ratings from 1 to 5) showed that, contrary to the ﬁrst years in which it was used, they did not ﬁnd it engaging (X = 3.07, σ = 0.47) neither interesting (X = 3.22, σ = 0.41) nor fun (X = 3.3, σ = 0.38). However, although the evolution of the course over the years caused meaningful performance changes among the students, we still do not know clearly how different students engaged with the course and interacted with it. In the next section, we explain a method that we used to identify var- ious groups of students considering their performance on the course. We also detail the data that we applied for this identiﬁcation. Finally, to boost the students’ creativity and autonomy, and also to customize their learning experience using what they learned in the MCP classes, an Avatar World was added to the course for two years (2013-15). It was a 3D virtual world that evolved and grew by obtaining XP (emerging new build- ings and characters). Students were represented by avatars, which could be customized or made by the techniques and B. CLUSTER PERFORMANCE In Section V-A, we could identify the optimal number of clusters for all years using the elbow technique and the accumulated XP of the students over time. In this section, we aim to present the average performance of each cluster using the same data. Here, we should state that each semester was different from another one in various terms, such as having different duration, learning activities, and evaluation elements. Therefore, the clusters from different years were generated using different collections of data. g g As presented in Figure 8, all students’ performance looked the same at the beginning of each semester since students were not fully enrolled in the MCP course or they still did not do any signiﬁcant learning activity. During a semester and by doing enough activities by the students, their groups became more distinct. The performance of all clusters sud- denly enhanced at the end of the semesters (i.e. end of May or beginning of June), which was due to obtaining a notice- able amount of XP from a multimedia presentation (it is a main course activity). After this presentation, since a little time (only two active weeks) and learning activities left, the clusters’ performance often did not change signiﬁcantly. In addition, the performance enhancement (at the end of the semesters) was larger in the ﬁrst three years because the ﬁnal exam was the main evaluation element that was located at the end of the course. After the ﬁrst three years and by having quizzes, the course workload became more balanced and that enhancement got smoother. In the next section, we describe each of these clusters and analyze them in detail. V. IDENTIFY STUDENTS’ GROUPS During the semester, we noticed that students were interacting with the course differently. Some were highly active during a semester and achieved higher grades, while some performed well at the beginning of the course but they lost their interest after a while and focused mainly on the major evaluation 81014 VOLUME 9, 2021 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 6. XP and Grade comparison over years. FIGURE 6. XP and Grade comparison over years. FIGURE 6. XP and Grade comparison over years. elements. There were also students that just performed enough to pass the course. Therefore, we found it of interest to identify different groups of students and assess how they interacted with the course. Considering groups rather than individuals also allows us for future extension of our course to a personalized one that can handle a large amount of data (a scalable course) for recommending gamiﬁed activities. In addition, it is of importance to ﬁnd whether groups’ per- formance is an adequate indicator to identify their groups. Hence, in this study, our main motivation is differentiat- ing the students’ groups considering their performance. For that, accumulated XP over a semester was estimated for each student, which allowed us to equally represent students with similar performance and clearly distinguish their ranks. In other words, if two students got similar amount of XP in the same day, with this data preparation method, we are able to distinguish their ranks since their previous XP was also taken into account. This approach was also used by Barata [11]. The amount of accumulated XP per day was used as attributes for the cluster analysis to group students by similarity of gaining XP. We used the K-means algorithm [39] for cluster analysis since it is easy to interpret and implement while has a linear complexity [51], [52], [54], [63]. that there were different groups of students interacting with the course. Interestingly, our results are compatible with the results that are presented in a PhD thesis [61]. In that thesis, author also identiﬁed four groups of students considering their performance. A. NUMBER OF CLUSTERS To identify the number of clusters (students’ groups) we used the elbow technique. It runs K-means algorithm K times on the data (K is the number of clusters), and for each value of K estimates the Sum of Squared Errors (SSE). The results will be presented in form of a line graph (like an arm), where the elbow on the arm indicates the optimal number of clusters for that data. As shown in Figure 7, four is one of the most promising number of clusters for all years except 2010-11. For that year, three is a good candidate for the number of clusters. It can be because of not having enough learning activities and evaluation elements in the ﬁrst year to clearly distinguish students’ groups from each others. Any- how, any of these numbers supports our initial assumption A. GROUP DESCRIPTION As presented above, we distinguish four main student groups according to performance (RQ1). The ﬁrst group is the Achievers, who concentrated on the achievements and obtained all the available XP. Due to the reason that they caught every opportunity to gain XP, they are called Achievers. These students had a higher XP accumulation curve (as shown in Figure 8), which presents that they were mainly ahead of other groups (positioned at the top of the Leaderboard). Regular students are the second group that their performance was good but lower than the Achievers, and balanced the achievements with the traditional assessment components. Their ﬁnal grades were also close to the Achiev- ers. Due to their behavior, they are named Regular students [8], [9]. In some studies, such as [10], they are called Late Awakeners since they seem to be like a fast rank loss in the beginning of the course followed by a progressive recovery (Figure 8-c, -f, -h). Furthermore, we compared the size of all groups. As pre- sented in Figure 9, Achievers and Underachievers (had the highest and lowest performance) were often in minority while the other two groups (had almost average performance) were the major ones. It matches most of the courses where only a few students out- or under-perform the rest of the students while the performance of the majority is around average. In terms of XP, the groups’ XP (Figure 9-a) was noticeably higher in 2011-12, which was due to not having any restric- tion on the XP gained from the Skill Tree. This problem was ﬁxed for the next years by restricting the achievable amount of XP from the Tree. Regarding the badges, the groups’ badges reduced since 2015-16, which could be due to course restructuring in that year. Before the course restructuring, the enrolled students often were highly interested in course while after the restructuring their reasons for course enrol- ment varied from high interest to perception of easiness (more detail in Section IV). So, they all were not that motivated to earn many badges. The performance of the third group (Disheartened group) was lower than the Regular group, and it looks that they ignored some of the course activities. They normally started the course at a pace similar to the Achievers, but soon they lost their interest and fell behind in terms of XP acquisition. B. GROUP ACHIEVEMENTS AND CHARACTERISTICS This section presents some of the groups’ statistics as well as showing their obtained XP, badges and ﬁnal grades. In Figure 9, we compared the identiﬁed groups in terms of obtaining XP and badges. In the same ﬁgure, we also compared their ﬁnal grades, ranks, and their size through the years. Here, we do not have the Underachievers group for the year 2010-11 since we only found three rigid groups among the students (see Figures 7 and 8). As shown in Figure 9 (sub-ﬁgures a to c), Achievers outperformed others in getting XP and badges, and had higher ﬁnal grades in all years. They also obtained better ranks constantly (sub-ﬁgure d) (RQ2 and RQ3). To compute the ranks, we used median on the accumulated XP of each group’s members. Median is opted since it is more robust than mean to outliers. In this computation, we ignored the students’ ranks during the ﬁrst four weeks of each semester (before March 15) due to having much variation (more details in Section VI-E). The groups’ ranks increased through years, which is due to having more students. VI. GROUP ANALYSIS VI. GROUP ANALYSIS In this section, we intend to describe each group, detail their characteristics and achievements, and analyze their 81015 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 7. Finding the optimal number of clusters via elbow technique and accumulated XP. 81016 FIGURE 7. Finding the optimal number of clusters via elbow technique and accumulated XP. 81016 81016 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 8. Average clusters performance using the accumulated XP. Here, black curves show the Achiever groups, green curves indicate the Regular, and blue curves present the Disheartened groups. Finally, red curves show the Underachievers. FIGURE 8. Average clusters performance using the accumulated XP. Here, black curves show the Achiever groups, green curves indicate the Regular, and blue curves present the Disheartened groups. Finally, red curves show the Underachievers. 81017 VOLUME 9, 2021 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? of the course activities. Finally, Underachievers had similar behavior to the Novice group and they did just enough to pass thee course. interaction, engagement, and self regulation skill. Besides these, we also evaluate how early we can predict the students’ groups and their performance with a high accuracy. In addi- tion, we assess whether the gaming proﬁle of the students had any relation with each student’s group. This analysis allows us to better understand what students of a given proﬁle like (Section VI-F). Finally, we study the students’ feedback to evaluate how much we were successful in achieving our goals. A. GROUP DESCRIPTION FIGURE 9. Comparing the groups XP, badges, rank, final grades, and size through the years. A. GROUP DESCRIPTION Their average Leaderboard positions, which was close to the Achievers at the beginning of a semester, dropped signiﬁ- cantly as the semester evolves. This is the reason that we called them Disheartened. Finally, the last group is named Underachievers since the students in this group were showing little interest and engagement with the course. They also had the lowest performance and just did enough activities to pass the course, and were mainly positioned at the bottom of the Leaderboard. Although the Achievers and Regular students had a high performance and engagement with the course, the other two groups were disengaged and their performance was relatively low [8]–[10]. We also analyzed the performance of groups in earning badges from different levels of the Skill Tree. In every year, for each level, we summed the number of obtained badges by each group and divided it by the size of that group. In Figure 10, each bar refers to a level of the Tree while every color indicates a semester (eight colors because the Tree is in-use since 2012-13). As presented in this ﬁg- ure, all groups had better performance in level one since the tasks were less complicated. Groups’ success got lower by raising the levels since the tasks became more chal- lenging. As expected, Achievers had the best performance in comparison with others in obtaining badges from all levels. As mentioned in Section V-A, authors of [61] also found four groups of students. It is of interest that our groups are compatible with the ones that are presented in [61]. The performance of Achievers matched the performance of the students in the Advanced group of that study. Both groups almost completed all the learning tasks and got most of the points. Regular group that had the second best performance can be compared with the Good Participants group. The ﬁnal grades of these groups were slightly lower than the Achievers and Advanced groups. The behavior of Disheartened was compatible with the Keen group. Both groups ignored some 81018 81018 VOLUME 9, 2021 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 9. Comparing the groups XP, badges, rank, final grades, and size through the years. FIGURE 9. Comparing the groups XP, badges, rank, final grades, and size through the years. C. COURSE ENGAGEMENT As shown in Figure 11, the densities of all groups in all years almost follow the same pattern. At the beginning of each semester, due to awarding an initial XP to all students as well as having simple tasks that students could get their XP, the density of all groups is one. Then, all densities dived drastically till end of April. After April, by getting close to the end of the semesters, students increased their activities and obtained more XP to get better grades or even pass the course. Within this time (after April) and by approaching the end the course, we can also observe some shifts in groups’ performance (change in average XP) that was never happened One of the current educational challenges is to keep students effectively engaged with a course [43]. Hence, in this section, we intend to show how much we were successful with that. For this purpose, we considered the density of activities and the collected XP by each group in every semester. For that, we initially built a binary matrix for every year in a way that whenever a group member did an action (e.g. made a post, got a badge, performed a task) or collected an XP in a day, we regarded it as one otherwise it considered as zero. Then, these matrices were used to generate the density graphs. 81019 81019 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 10. Earned badges from each level of the skill tree by each group. FIGURE 10. Earned badges from each level of the skill tree by each group. before April. It conﬁrms that the students are often deadline driven (RQ4). was quite sparse. It also presents that the Regular groups were interacting with the course more than the Disheartened ones but less than the Achievers. Therefore, we can conclude that there is a direct relation between the students’ interaction and their performance in the course. So, whenever a student is more active and interacts with the course frequently, he/she mainly gets a better grade than the ones with a lower level of interaction. This statement is also conﬁrmed by the results shown in Figure 14, where the students got higher ﬁnal grades whenever they interacted more with the course. D. STUDENT INTERACTION WITH THE COURSE Students’ interaction with the course is an informative type of data that we used to analyze whether there is a relation between the students’ interaction and their performance. For that, we employed the students’ log data the we collected from the Moodle. We then generated scatter plots using this data. In Figure 13, each dot indicates that a student interacted with the course in a day. As shown in this ﬁgure, Achievers interacted frequently while the Underachievers interaction C. COURSE ENGAGEMENT Similar pattern could be observed in Figure 12, where the density activity of all groups dropped until the end of April and raised after this month. In this graph, we ignored the initial XP that was awarded to all students. These XP were to motivate the students to continue with MCP and to solve the grade rounding issue at the end of the semester. Here, it is noticeable that the density of several groups became one (or close to one) at the end of the semesters, which shows that the groups were mainly active in that time, but sometimes without obtaining XP (compare the results in Figure 11 with 12). Like Figures 11 and 12, in Figure 13 we also observe that in April all groups interacted less with the course, which is due to the semester break. In addition, the groups’ interaction became sparse at the end of the semesters (≈after the ﬁrst week of June). It is because that around this time students delivered their multimedia presentation (one of the main course activities at the end of the semesters) and there was not much time (there are only two active weeks in June) and course activities left to complete. Hence, the groups’ interaction dropped signiﬁcantly. Moreover, in the ﬁrst ﬁve years, the initial interaction of the groups with the course was sparse while it became denser since 2015-16. It might be due to the course restructuring in this year and better distributing of the game elements throughout the course. Similarly to the results presented in Figure 9, in Figure 13 we also see that the Regular and Disheartened groups are often the larger ones while the other two are smaller. In both ﬁgures, we noticed that the students’ XP and activities were in the minimum in April. It is due to the reason that the semester break is in this month and stu- dents go for holidays for almost one week. Therefore, their activities, and subsequently their XP dropped signiﬁcantly in this month. Also, these graphs show that the students are mostly deadline-driven and become active at the end of the semesters. In order to have them well-engaged with a course, we need to deﬁne activities that are carefully distributed during a semester, which keeps the students interacting with the course continuously. E. PERFORMANCE STABILITY There are several models to determine the gaming proﬁles of the students, such as the one presented by Richard Bartle [12], [13] that divides students into four groups, or the one called Demographic Game Design 1 (DGD1) [15]. DGD1 was introduced by Chris Bateman, and is based on the Myers-Briggs personality model [49]. Among all the models, we selected the Brain- hex [55] since it is one of the most complete models that works based on the previous ones, such as DGD1. In addition, its questionnaire is available and online,2 which makes it easy to administrate and access. This model is based on neurobiological responses inherent to playing games [11]. It includes seven player archetypes, and classiﬁes players into primary (main) and secondary (sub) classes. The seven archetypes are: Achiever, Conqueror, Daredevil, Mastermind, Seeker, Socializer, and Survivor. FIGURE 14. Relation between students’ interaction and their final grades. Here, a year can be shown by three or four dots depending on the number of students’ groups for that year. Achievers +, Regular □, Disheartened ⃝, Underachievers △. made as soon as starting a course with a high accuracy since it gives more time to the professors to better guide different groups of students. Hence, in this section, we study how soon the students’ performance gets stable for having a more accurate identiﬁcation. For that, we initially estimated the accumulated XP of each student through a semester. This data enabled us to differentiate students that got similar XP in the same day. Then, the daily ranks of students were computed using the mentioned data. In Figure 15, we presented the students’ ranks through the semesters. Here, each group is highlighted in a different color, and the average ﬁnal grade of each group is mentioned on the right side of the graphs. In these graphs, students with better performance have lower ranks (located at the bottom of the graphs), and vice versa. As presented in Figure 15, the students’ ranks varied a lot in the ﬁrst four weeks of each semester. Then, they got stable till the end of the semesters (RQ5). It could be due to the reason that the students were new to the course and they still did not do much activities to gain XP (i.e. their ranks are similar). Hence, obtaining a small XP could affect their ranks signiﬁcantly. E. PERFORMANCE STABILITY One of the main challenges in educational environments is to identify the students’ groups (Proﬁles). It assists professors to provide more suitable learning materials for the students that match their competency levels. This identiﬁcation needs to be 81020 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 11. Density of obtained XP by groups. FIGURE 11. Density of obtained XP by groups. FIGURE 11. Density of obtained XP by groups. 81021 81021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? S G C FIGURE 12. Density of activities by groups. FIGUR D i f i i i b FIGURE 12. Density of activities by groups. FIGURE 12. Density of activities by groups. 81022 81022 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 13. Students Interaction with MCP. FIGURE 13. Students Interaction with MCP. 81023 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 14. Relation between students’ interaction and their final grades. Here, a year can be shown by three or four dots depending on the number of students’ groups for that year. Achievers +, Regular □, Disheartened ⃝, Underachievers △. TABLE 2. Student groups and the Brainhex classes. and their gamer proﬁles. The importance of this analysis is, the better we know what a proﬁle likes, the better we can provide gaming experiences for it. There are several models to determine the gaming proﬁles of the students, such as the one presented by Richard Bartle [12], [13] that divides students into four groups, or the one called Demographic Game Design 1 (DGD1) [15]. DGD1 was introduced by Ch i B d i b d h M B i li FIGURE 14. Relation between students’ interaction and their final grades. Here, a year can be shown by three or four dots depending on the number of students’ groups for that year. Achievers +, Regular □, Disheartened ⃝, Underachievers △. TABLE 2. Student groups and the Brainhex classes. TABLE 2. Student groups and the Brainhex classes. and their gamer proﬁles. The importance of this analysis is, the better we know what a proﬁle likes, the better we can provide gaming experiences for it. E. PERFORMANCE STABILITY After almost the ﬁrst month (≈March 15) and by earning a considerable amount of XP, the students’ ranks did not change noticeably anymore. This trend can be observed in all years and the course changes and various game elements did not inﬂuence this trend. It indicates that theoretically we should be able to identify the students’ groups and predict their performance with high accuracy around this time. According to our results for predicting the students’ groups and performance, which are not mentioned in this paper, we could achieve the accuracy of ≈85% (via Random Forest algorithm [56], [62]) around mid March. Therefore, soon after starting the course we are able to identify the students’ groups and performance accurately. To determine the Brainhex classes of the students (i.e. main-class and sub-class), we initially used the afore- mentioned questionnaire at the beginning of a semester. We then related the collected information to the students’ clusters [11]. The results for all years are shown in Figure 16. Although it was rough to determine the major classes in some years, Mastermind, Conqueror, and Achiever were the three major primary and secondary classes for all clusters (students’ groups). Mastermind refers to the students that enjoy solving puz- zles and strategic games while concentrating on the most efﬁ- cient decisions. Achiever addresses the goal-oriented students who like to collect special achievements and points. Finally, the Conqueror players like challenges, enjoy defeating tough rivals and beating others. We believe that the MCP was not attractive for this group since there was never an actual rival to defeat. Anyhow, it might come from the fact that the course Leaderboard encouraged the students’ competitiveness. As shown in Table 2, although the percentage of the three mentioned classes (main and sub) varied for each student group, in total around 60 to 70 percent of each group was formed by the mentioned classes and the remaining percent- age was composed of the other four classes. To show that there was not any signiﬁcant difference among the student groups considering the Brainhex classes, we estimated the p-values using all main and sub classes for each pair of student groups. As presented in Table 3, there was not 2http://www.survey.ihobo.com/BrainHex/ F. GAMING PROFILES To this end, we assessed the self-regulation skill of the students in all years (Figure 17). This assessment was based on the number of XP and badges earned by the students from beginning to the end of a semester. In Figure 17, we observe how students’ activities were biased towards the end of the semesters. Taking into account that there were only two active weeks in February and June FIGURE 18. Students’ score (1 to 5) to their effort distribution and regular study over a semester. and it became more ﬂexible. Nonetheless, this score was not signiﬁcantly high. H STUDENTS’ FEEDBACK FIGURE 17. Proportion of obtained XP and Badges per month. FIGURE 17. Proportion of obtained XP and Badges per month. FIGURE 17. Proportion of obtained XP and Badges per month. FIGURE 18. Students’ score (1 to 5) to their effort distribution and regular study over a semester. relation between students’ groups and their gaming proﬁles (RQ6). G. SELF-REGULATION In e-learning courses, whether gamiﬁed or not, one of the main issues is the lack of self-regulation skill of the students regardless of their groups (proﬁles). It results in missing a considerable amount of time during a semester while being pressured by the learning tasks and activities before the due dates. Thus, this skill can be considered as a sig- niﬁcant one for assisting students to better distribute their workload through a semester. To this end, we assessed the self-regulation skill of the students in all years (Figure 17). This assessment was based on the number of XP and badges earned by the students from beginning to the end of a semester. FIGURE 18. Students’ score (1 to 5) to their effort distribution and regular study over a semester. and it became more ﬂexible. Nonetheless, this score was not signiﬁcantly high. In Figure 17, we observe how students’ activities were biased towards the end of the semesters. Taking into account that there were only two active weeks in February and June, we immediately notice that in June students were more active than February, even if we deduct the effect of the ﬁnal exam in the ﬁrst two years. Also, the multimedia presentation that was in May/June might explain a bump in XP, but it could not justify an increase in badges since it only awarded XP. Fur- thermore, we can see that in May students were more active than April (the semester break is in April) and March. As it is shown, in all years, almost half of the rewards were obtained in May and June. Moreover, in 2011-12, the amount of XP obtained in June was signiﬁcantly higher than the rest of the years. It is because that the initial version of the Skill Tree was introduced and it was not well integrated with the course and its tasks were not well distributed over the semester. So it caused a bump in XP. F. GAMING PROFILES In this section, we intend to analyze whether there is a relation between the students’ groups (indicating their performance) VOLUME 9, 2021 81024 81024 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 15 Performance stability of students Black: Achievers Green: Regu egular, Blue: Disheartened, Red: Underachievers. Performance stability of students. Black: Achievers, Green: Regular, Blue: Disheartened, Red: Underachievers. 81025 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 16. Brainhex categories for each student’s cluster. TABLE 3. P-values across student groups. Significance = 0.05. an important difference among the student groups. We also ti t d th l ti id i TABLE 4. P-values across all years. Significance Level = 0.05. inﬂuenced the proportion of the classes through years. According to our results presented in Table 4, the proportion f th l did t h i iﬁ tl F ll th GURE 16. Brainhex categories for each student’s cluster. FIGURE 16. Brainhex categories for each student’s cluster. FIGURE 16. Brainhex categories for each student’s cluster. TABLE 4. P-values across all years. Significance Level = 0.05. TABLE 3. P-values across student groups. Significance = 0.05. TABLE 3. P-values across student groups. Significance = 0.05. TABLE 4. P-values across all years. Significance Level = 0.05. inﬂuenced the proportion of the classes through years. According to our results presented in Table 4, the proportion of the classes did not change signiﬁcantly. From all these results, we can conclude that there was not any speciﬁc an important difference among the student groups. We also estimated the p-values among consecutive years considering all Brainhex classes to assess whether the course changes VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 17. Proportion of obtained XP and Badges per month. relation between students’ groups and their gaming proﬁles (RQ6). G. SELF-REGULATION In e-learning courses, whether gamiﬁed or not, one of the main issues is the lack of self-regulation skill of the students regardless of their groups (proﬁles). It results in missing a considerable amount of time during a semester while being pressured by the learning tasks and activities before the due dates. Thus, this skill can be considered as a sig- niﬁcant one for assisting students to better distribute their workload through a semester. VII. DESIGN IMPLICATIONS After analyzing the students’ data for all years, we are in the position to suggest a set of design guidelines, which can be helpful to design a more effective gamiﬁed learning environment. In this section, we discuss them in detail. 1) Game elements made the course more interesting. 2) Game elements made the course more engaging. 3) MCP would be better off without game elements. 4) Game elements contribute for a greater workload when compared with traditional courses. H. STUDENTS’ FEEDBACK Moreover, in MCP, students needed to use various tools and software to complete assignments and tasks, which they found it useful for their future (Third state- ment). In this ﬁgure, we can see a notable drop in 2017-18. It could be due to not having the Quest element in the course anymore. Also, the peak in 2015-16 (for the Second and Third statements) is because of the course restructuring in that year, which caused that the students found the course more creative and valuable for the future. FIGURE 21. Students’ feedback on the course autonomy, creativity, and usefulness. FIGURE 19. Student feedback on the general aspects of MCP. FIGURE 20. Students’ feedback on the quality of game elements. In 2014-15 and 2015-16, we could not collect the students’ opinions about the first, second, and third statements. 3) The course taught me useful skills for my future. 3) The course taught me useful skills for my future. FIGURE 19. Student feedback on the general aspects of MCP. The results are summarized in Figure 21. As shown in this ﬁgure, students agreed that the course was autonomous and ﬂexible (First statement), and it highly triggered their creativ- ity (Second statement). Moreover, in MCP, students needed to use various tools and software to complete assignments and tasks, which they found it useful for their future (Third state- ment). In this ﬁgure, we can see a notable drop in 2017-18. It could be due to not having the Quest element in the course anymore. Also, the peak in 2015-16 (for the Second and Third statements) is because of the course restructuring in that year, which caused that the students found the course more creative and valuable for the future. FIGURE 19. Student feedback on the general aspects of MCP. FIGURE 21. Students’ feedback on the course autonomy, creativity, and usefulness. FIGURE 20. Students’ feedback on the quality of game elements. In 2014-15 and 2015-16, we could not collect the students’ opinions about the first, second, and third statements. FIGURE 21. Students’ feedback on the course autonomy, creativity, and usefulness. The students’ feedback on the quality of the game elements was collected using four statements. Like Figure 19, we asked the students to rate the statements using a ﬁve-point Likert scale. These statements were: A. MULTI ACTIVITIES Results are presented in the form of a line graph in Figure 20. A brief look at this ﬁgure shows that the students highly agreed that the course was more interesting with the elements and they made it more engaging. Furthermore, they strongly disagreed that the course could be more effective without the elements. Finally, they were not sure whether the elements made the course more demanding or not (Fourth statement). According to these results, we conclude that our game elements were well-designed and worked in a promis- ing way that the students did not feel exhausted using them. In MCP, we aimed at enhancing student autonomy, interac- tion, motivation, creativity, and engagement with the course. Through years, we have realized that these goals cannot be achieved using only a single learning activity and a combina- tion of activities is required. For example, the Leaderboard would motivate the students to be engaged more with the course for getting higher grades and ranks, and subsequently, it would inﬂuence their interaction, while the Skill Tree would enhance the students’ autonomy since it enables them to perform the preferred activities for obtaining rewards. Similar results were presented in [3], where the authors stated that a single game cannot be used to achieve all the gameful learning goals, such as enhancing engagement, motivation, and performance. In addition to the mentioned students’ opinions, we col- lected their feedback on how successful was the MCP in boosting the students’ autonomy, creativity, and whether they could learn some practical skills that could be useful for their future. For that, we asked the students to give us their feedback (rate from 1 to 5) on three statements that focused on the mentioned criteria. The three statements were: H. STUDENTS’ FEEDBACK The students’ feedback on the gamiﬁed MCP was collected since 2012-13. For this purpose, we provided a short ques- tionnaire to collect their opinions about various aspects of the course. Initially, we focused on their general feedback on the course, like if it was competitive, likable, creative, interesting, extendable to other subjects, or even if the students could learn more with it in comparison with the other courses. As presented in Figure 19, most of the students had a strong agreement on the creativity of the course and conﬁrmed that it was engaging, likable, extendable, interesting, motivating, and well-received. In spite of having their agreement on the mentioned aspects, they disagreed that they could learn more with the MCP than the other courses. It is worth mentioning that in 2019-20, where almost all universities and courses were negatively inﬂuenced by the Covid-19 pandemic, the students were positive about MCP and found it motivating and engaging (Figure 19). This, shows that MCP was robust to this issue, and the pandemic had almost no impact on its effectiveness. Also, the results presented in previous sections conﬁrmed that the students’ engagement and interaction did not change in 2019-20. We can conclude that the most of the students were lacking the self-regulation skill, and were mostly deadline driven (RQ7). It was also mentioned in their feedback (collected since 2012-13) that their effort was unbalanced and they did not study regularly through a semester (Figure 18). In 2015-16, the average score was slightly higher. It could be due to the reason that the course was restructured in that year, 81027 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? FIGURE 19. Student feedback on the general aspects of MCP. FIGURE 20. Students’ feedback on the quality of game elements. In 2014-15 and 2015-16, we could not collect the students’ opinions about the first, second, and third statements. Th t d t ’ f db k th lit f th l t 3) The course taught me useful skills for my future. The results are summarized in Figure 21. As shown in this ﬁgure, students agreed that the course was autonomous and ﬂexible (First statement), and it highly triggered their creativ- ity (Second statement). D. EARLY GROUP IDENTIFICATION We analyzed the performance (ranks) for each cluster across all years (Figure 15). We noticed that performance varied a lot at the start of each semester (≈ﬁrst four weeks). This could be due to lack of early engagement in collecting XP. Thus, even collecting a few XP would affect a student’s rank remarkably. After the middle of March (almost one month after semester start) and collecting signiﬁcant XP, student ranks became more stable. This implies that within a month after course start we can identify clusters and predict student performance with high accuracy. This enables us to tailor the gamiﬁed experience in the way that best suits different student groups early in the course. TABLE 5. Prediction accuracy of students’ performance with and without using genders data. B. QUANTUM ASSESSMENT MECHANISM 1) The course allowed me to get rewards for things I like to do. 1) The course allowed me to get rewards for things I like to do. As explained in Section VI-C, students are deadline-oriented and often become active at the end of the semester. This lag leads to idle time during the semester and a frenzied 2) The course allowed me to be creative. 81028 VOLUME 9, 2021 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? ﬂurry of activity just before deadlines. To avoid these issues and to better distribute students’ workloads over a semester, it is essential to have an assessment mechanism that awards students as they participate in a course. As mentioned in Section IV, MCP uses a quantum assessment mechanism, and students are granted (via XP, levels, and badges) when- ever they are performing course activities. This mechanism provides a transparent, traceable, and incremental continuous assessment of student performance down to the individual XP. Figure 8 shows that in the ﬁrst three years (when MCP grade was not well granularized) there was a bump in XP at the end of the semester because the main assessment took place at the end of the course. Since 2013-14 and by having quizzes instead of the ﬁnal exam, students became more active dur- ing the semester and the grade bump became smoother (i.e. performance curves became more linear). strategy to make the students interacting with the course right after starting it is of importance. One strategy could be placing simple learning tasks that have extra rewards (grades) at the beginning of the course to convince the students for interacting with it. F. BRAINHEX NOT SIGNIFICANT As presented in Section VI-F, Achiever, Conqueror, and Mas- termind were the three most frequent classes of Brainhex (main and sub) among all groups of the students. Therefore, we can conclude that there was not any speciﬁc relation between a single class and a students’ group. Our statement was also conﬁrmed in [9], where the authors predicted the students’ groups using different collection of data. They got the accuracy of 71.70% via the BayesNet algorithm [34] using the Brainhex classes, while after ignoring the BrainHex categories, they achieved the accuracy of 79.63%. They also used other algorithms, such as Logistic regression [45], for the prediction task. In some cases, the accuracy enhanced using Brainhex classes while in other cases the accuracy got worse. Hence, we can conclude that the Brainhex classes might not be an informative and signiﬁcant type of data, especially for the prediction task. G. GENDER NOT IMPORTANT During our study, we analyzed whether the students’ genders could inﬂuence the accuracy of predicting their performance, and consequently, their groups. For that, students’ perfor- mance was predicted using their logs data and their past grades (on the same course) with and without using gen- der information. Random forest [18], Naive Bayes [65], and K-Nearest Neighbor (KNN) [58] were used for the prediction task (Table 5). In both cases, the accuracy results were the same and did not change. In Table 5, we present, as a sample, the results for years 2010-11, 2014-15, and 2018-19. As can be seen, using or omitting gender information had no impact in the prediction’s accuracy. Values for the other years, while not show for brevity’s sake, follow the same pattern. There- fore, we can deduce that gender has no impact on predicting students’ performance and groups. C. FOUR STUDENT GROUPS The main goal of our study was to analyze how different student groups interacted with a gamiﬁed course. For that, we distinguished each grouping by applying clustering tech- niques to student performance data. As already presented in Figure 7, we identiﬁed four distinct clusters. Considering that the course was modiﬁed in each year, we conclude that the underlying four-cluster model is both stable and resilient to external changes. Interestingly, the same results were obtained by Nabizadeh [61]. He also conﬁrmed that four is the most promising cluster conﬁguration among all student populations considered. E. INITIAL ENGAGEMENT In our analysis, we noticed that the students were almost inactive in the ﬁrst two weeks of each semester, and then slowly started interacting with the course. This time can be considered as a golden time since if we get the students to work with the course as soon as starting a semester, we would be able to collect more interaction data. This data is essential to have an accurate prediction of the students’ groups and their performance. This accurate prediction in the early stages of the course gives us enough time to better guide the students by tailoring the gamiﬁed activities in the way the best match the students’ preferences and groups. Therefore, ﬁnding a VIII. CONCLUSION VIII. CONCLUSION In this paper, we studied how different groups of students behaved in a gamiﬁed environment and interacted with it. For that, we developed a gamiﬁed course called Multime- dia Content Production (MCP), and collected the students’ interaction data with it for ten years. Students’ groups were determined using clustering techniques, which were applied to the students’ performance data. In general, we often dis- tinguished four groups of students (clusters) in all years. 81029 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? By analyzing and comparing these groups, we noticed that all groups were deadline driven and became active as the deadline loomed. Also, the performance and ﬁnal grades of all groups were accurately predictable within the ﬁrst month after starting the course since within this time the students’ performance got stable and did not change noticeably any- more. Furthermore, we concluded that the students’ groups did not have any relations with their gaming proﬁles. Finally, several practical implications for designing a gamiﬁed course were suggested, such as having a continuous evaluation mechanism (i.e. awarding students as they participate in a course) throughout a semester that helps to keep students engaged with a course and distribute the workload. There are still a number of gaps that follow from our implications, such as how to engage students in early stages of a course, which would beneﬁt from further research. [13] R. A. Bartle, Designing Virtual Worlds. Indianapolis, IN, USA: New Riders, 2004. [14] S. Barzilai and I. Blau, ‘‘Scaffolding game-based learning: Impact on learn- ing achievements, perceived learning, and game experiences,’’ Comput. Educ., vol. 70, pp. 65–79, Jan. 2014. [15] C. Bateman and R. Boon, 21st Century Game Design (Game Development Series). Newton, MA, USA: Charles River Media, 2005. [16] F. Bellotti, R. Berta, A. De Gloria, E. Lavagnino, A. Antonaci, F. M. Dagnino, and M. Ott, ‘‘A gamiﬁed short course for promoting entrepreneurship among ICT engineering students,’’ in Proc. IEEE 13th Int. Conf. Adv. Learn. Technol., Jul. 2013, pp. 31–32. [17] P. Brauner, A. C. Valdez, U. Schroeder, and M. Zieﬂe, ‘‘Increase physical ﬁtness and create health awareness through exergames and gamiﬁcation,’’ in Proc. Int. Conf. Hum. Factors Comput. Informat. Berlin, Germany: Springer, 2013, pp. 349–362. [18] L. Breiman, ‘‘Random forests,’’ Mach. Learn., vol. 45, no. 1, pp. 5–32, 2001. [19] P. Buckley and E. REFERENCES [24] L. de-Marcos, A. Domínguez, J. Saenz-de-Navarrete, and C. Pagés, ‘‘An empirical study comparing gamiﬁcation and social networking on e-learning,’’ Comput. Educ., vol. 75, pp. 82–91, Jun. 2014. [1] A. Ahmed and M. J. D. Sutton, ‘‘Gamiﬁcation, serious games, simula- tions, and immersive learning environments in knowledge management initiatives,’’ World J. Sci., Technol. Sustain. Develop., vol. 14, nos. 2–3, pp. 78–83, Apr. 2017. [25] M. R. de Almeida Souza, K. F. Constantino, L. F. Veado, and E. M. L. Figueiredo, ‘‘Gamiﬁcation in software engineering education: An empirical study,’’ in Proc. IEEE 30th Conf. Softw. Eng. Educ. Training (CSEE&T), Nov. 2017, pp. 276–284. [2] K. Aleksic-Maslac, B. Sinkovic, and P. Vranesic, ‘‘Inﬂuence of gamiﬁca- tion on student engagement in education,’’ Int. J. Educ. Learn. Syst. vol. 2, pp. 76–82, May 2017. [26] R. Dellos, ‘‘Kahoot! A digital game resource for learning,’’ Int. J. Instruct. Technol. Distance Learn., vol. 12, no. 4, pp. 49–52, 2015. [3] A. Antonaci, F. M. Dagnino, M. Ott, F. Bellotti, R. Berta, A. De Gloria, E. Lavagnino, M. Romero, M. Usart, and I. Mayer, ‘‘A gamiﬁed collabo- rative course in entrepreneurship: Focus on objectives and tools,’’ Comput. Hum. Behav., vol. 51, pp. 1276–1283, Oct. 2015. [27] P. Denny, ‘‘The effect of virtual achievements on student engage- ment,’’ in Proc. SIGCHI Conf. Hum. Factors Comput. Syst., Apr. 2013, pp. 763–772. [4] D. D. Armier, Jr., C. E. Shepherd, and S. Skrabut, ‘‘Using game elements to increase student engagement in course assignments,’’ College Teach., vol. 64, no. 2, pp. 64–72, Apr. 2016. [28] S. Deterding, R. Khaled, L. E. Nacke, and D. Dixon, ‘‘Gamiﬁcation: Toward a deﬁnition,’’ in Proc. CHI Gamiﬁcation Workshop, Vancouver, BC, Canada, 2011, pp. 1–4. [5] P. Bajdor and L. Dragolea, ‘‘The gamiﬁcation as a tool to improve risk management in the enterprise,’’ Annales Univ. Apulensis, Ser. Oeconomica, vol. 13, no. 2, p. 574, 2011. [29] D. Dicheva, C. Dichev, G. Agre, and G. Angelova, ‘‘Gamiﬁcation in education: A systematic mapping study,’’ J. Educ. Technol. Soc., vol. 18, no. 3, pp. 75–88, 2015. [6] J. Baker, ‘‘Constructivist online learning environment survey,’’ in Hand- book of Research on Electronic Surveys and Measurements. Hershey, PA, USA: IGI Global, 2007, pp. 299–301. [30] D. Dicheva, K. Irwin, and C. Dichev, ‘‘OneUp: Supporting practical and experimental gamiﬁcation of learning,’’ Int. J. Serious Games, vol. 5, no. 3, pp. 5–21, Sep. 2018. [7] G. Barata, S. Gama, J. Jorge, and D. VIII. CONCLUSION Doyle, ‘‘Individualising gamiﬁcation: An investigation of the impact of learning styles and personality traits on the efﬁcacy of gamiﬁcation using a prediction market,’’ Comput. Educ., vol. 106, pp. 43–55, Mar. 2017. In the future, we want to analyze what students (especially the regular, disheartened, and underachiever groups) wanted to learn, what they learned, what their difﬁculties were, and how their performance can improve to reach the Achievers (the group with the best performance). To this end, we want to work on personalizing gamiﬁed activities for the regu- lar, disheartened, and underachiever groups to improve their learning performance. [20] D. T. Campbell and J. C. Stanley, Experimental and Quasi-Experimental Designs for Research. New York, NY, USA: Ravenio Books, 2015. [21] Ü. Çakıroğlu, B. Başıbüyük, M. Güler, M. Atabay, and B. Y. Memiş, ‘‘Gamifying an ICT course: Inﬂuences on engagement and academic performance,’’ Comput. Hum. Behav., vol. 69, pp. 98–107, Apr. 2017. [22] P.-Z. Chen, T.-C. Chang, and C.-L. Wu, ‘‘Effects of gamiﬁed classroom management on the divergent thinking and creative tendency of elementary students,’’ Thinking Skills Creativity, vol. 36, Jun. 2020, Art. no. 100664. [23] K. Davis, H. Sridharan, L. Koepke, S. Singh, and R. Boiko, ‘‘Learning and engagement in a gamiﬁed course: Investigating the effects of student characteristics,’’ J. Comput. Assist. Learn., vol. 34, no. 5, pp. 492–503, Oct. 2018. REFERENCES Nabizadeh, D. Gonçalves, S. Gama, J. Jorge, and H. N. Rafsanjani, ‘‘Adaptive learning path recommender approach using auxiliary learning objects,’’ Comput. Educ., vol. 147, Apr. 2020, Art. no. 103777. [53] A. H. Nabizadeh, J. P. Leal, H. N. Rafsanjani, and R. R. Shah, ‘‘Learning path personalization and recommendation methods: A survey of the state- of-the-art,’’ Expert Syst. Appl., vol. 159, Nov. 2020, Art. no. 113596. SANDRA GAMA is currently a Researcher with the Visualization and Multimodal Interfaces Group, INESC-ID. She is a Professor of computer science with the Instituto Superior Técnico, Uni- versity of Lisbon, Lisbon, Portugal. Her research interests include information visualization, mul- timodal user interfaces, and human–computer interaction. SANDRA GAMA is currently a Researcher with the Visualization and Multimodal Interfaces Group, INESC-ID. She is a Professor of computer science with the Instituto Superior Técnico, Uni- versity of Lisbon, Lisbon, Portugal. Her research interests include information visualization, mul- timodal user interfaces, and human–computer interaction. [54] A. H. Nabizadeh, A. M. Jorge, and J. P. Leal, ‘‘Estimating time and score uncertainty in generating successful learning paths under time constraints,’’ Expert Syst., vol. 36, no. 2, Apr. 2019, Art. no. e12351. [55] L. E. Nacke, C. Bateman, and R. L. Mandryk, ‘‘BrainHex: Preliminary results from a neurobiological gamer typology survey,’’ in Proc. Int. Conf. Entertainment Comput. Berlin, Germany: Springer, 2011, pp. 288–293. [56] T. M. Oshiro, P. S. Perez, and J. A. Baranauskas, ‘‘How many trees in a random forest?’’ in Proc. Int. Workshop Mach. Learn. Data Mining Pattern Recognit. Berlin, Germany: Springer, 2012, pp. 154–168. [57] Y. Pan, D. Schwartz, and S. Mishra, ‘‘Gamiﬁed digital forensics course modules for undergraduates,’’ in Proc. IEEE Integr. STEM Educ. Conf., Mar. 2015, pp. 100–105. [58] L. E. Peterson, ‘‘K-nearest neighbor,’’ Scholarpedia, vol. 4, no. 2, p. 1883, 2009. DANIEL GONÇALVES is currently a Researcher with the Visualization and Multimodal Interfaces Group, INESC-ID. He is a Professor of com- puter science with the Instituto Superior Técnico (IST/UL), Lisbon, Portugal. His research inter- ests include information visualization, personal information management, and human–computer interaction and accessibility. He is a member of the ACM and the Portuguese Computer Graphics Group (the National Eurographics Chapter). [59] T. Plomp, ‘‘Educational design research: An introduction,’’ in Educational Design Research. Enschede, The Netherlands: Netherlands Inst. for Cur- riculum Develop., 2013, pp. 11–50. [60] H. N. Rafsanjani, A. Ghahramani, and A. H. REFERENCES Fard, ‘‘Recom- mendation systems: A review,’’ Int. J. Comput. Eng. Res., vol. 3, no. 5, pp. 47–52, 2013. [63] L. Rokach and O. Maimon, ‘‘Clustering methods,’’ in Data Mining and Knowledge Discovery Handbook. Berlin, Germany: Springer, 2005, pp. 321–352. [38] M. D. Hanus and J. Fox, ‘‘Assessing the effects of gamiﬁcation in the classroom: A longitudinal study on intrinsic motivation, social comparison, satisfaction, effort, and academic performance,’’ Comput. Educ., vol. 80, pp. 152–161, Jan. 2015. [64] A. Todd, ‘‘Why gamiﬁcation is malarkey,’’ Morning Watch, Educ. Social Anal., vol. 44, nos. 1–2, pp. 1–7, Mar. 2017. [39] J. Hartigan and M. Wong, A K-Means Clustering Algorithm. Milton Keynes, U.K.: New Haven, 1979. [65] S. Xu, ‘‘Bayesian Naïve Bayes classiﬁers to text classiﬁcation,’’ J. Inf. Sci., vol. 44, no. 1, pp. 48–59, Feb. 2018. [40] M. Jakubowski, ‘‘Gamiﬁcation in business and education—Project of gamiﬁed course for university students,’’ in Proc. Annu. ABSEL Conf. Develop. Bus. Simulation Experiential Learn., 2014. [66] Youtopia. (2009). Youtopia Website. Accessed: Jul. 3, 2020. [Online]. Available: http://www.youtopia.com/info [41] M. Jurgelaitis, V. Drungilas, and L. Čeponien˙e, ‘‘Gamiﬁed moodle course for teaching UML,’’ Baltic J. Mod. Comput., vol. 6, no. 2, pp. 119–127, 2018. [42] K. M. Kapp, The Gamiﬁcation of Learning and Instruction: Game-Based Methods and Strategies for Training and Education. Hoboken, NJ, USA: Wiley, 2012. AMIR HOSSEIN NABIZADEH received the M.Sc. degree in information technology manage- ment from the University Technology Malaysia (UTM), in 2013, and the Ph.D. degree in computer science from the University of Porto, in 2018. He is currently a Postdoctoral Researcher with the INESC-ID, Instituto Superior Técnico, Uni- versity of Lisbon. His main research interests include machine learning, data mining, recom- mender systems, e-learning, and MOOC. AMIR HOSSEIN NABIZADEH received the M.Sc. degree in information technology manage- ment from the University Technology Malaysia (UTM), in 2013, and the Ph.D. degree in computer science from the University of Porto, in 2018. He is currently a Postdoctoral Researcher with the INESC-ID, Instituto Superior Técnico, Uni- versity of Lisbon. His main research interests include machine learning, data mining, recom- mender systems, e-learning, and MOOC. [43] A. Khan, O. Egbue, B. Palkie, and J. Madden, ‘‘Active learning: Engaging students to maximize learning in an online course,’’ Electron. J. E-Learn., vol. 15, no. 2, pp. 107–115, 2017. [44] S. Kiesler, R. E. Kraut, K. R. Koedinger, V. Aleven, and B. M. REFERENCES Mclaren, ‘‘Gamiﬁcation in education: What, how, why bother,’’ Acad. Exchange Quart., vol. 15, no. 2, pp. 1–5, 2011. [45] D. G. Kleinbaum, K. Dietz, M. Gail, M. Klein, and M. Klein, Logistic Regression. Berlin, Germany: Springer, 2002. [46] E. Lavoué, B. Monterrat, M. Desmarais, and S. George, ‘‘Adaptive gami- ﬁcation for learning environments,’’ IEEE Trans. Learn. Technol., vol. 12, no. 1, pp. 16–28, Jan./Mar. 2018. [47] K. M. Mbabu, R. O. Oboko, and S. Kamunya, ‘‘An adaptive gamiﬁcation tool for e-learning platform,’’ Open J. Inf. Technol., vol. 2, no. 2, pp. 41–52, Dec. 2019, doi: 10.32591/coas.ojit.0202.03041m. JOAQUIM JORGE (Senior Member, IEEE) received the Ph.D. degree from the Rensselaer Polytechnic Institute, in 1995. He coordinates the Interaction and Graphics Research Group, INESC-ID. He is currently a Computer Graph- ics and Multimedia Professor with the School of Engineering, Instituto Superior Técnico, Univer- sidade de Lisboa. His research interests include multimodal user interfaces, VR/AR/MR, 3D visu- alization, and advanced learning techniques. He is a Fellow of the Eurographics Association. He is a Distinguished Member and a Speaker of the ACM. He is the Editor-in-Chief of the Computers and Graphics and Graphics and Visual Computing journals. JOAQUIM JORGE (Senior Member, IEEE) received the Ph.D. degree from the Rensselaer Polytechnic Institute, in 1995. He coordinates the Interaction and Graphics Research Group, INESC-ID. He is currently a Computer Graph- ics and Multimedia Professor with the School of Engineering, Instituto Superior Técnico, Univer- sidade de Lisboa. His research interests include multimodal user interfaces, VR/AR/MR, 3D visu- alization, and advanced learning techniques. He is a Fellow of the Eurographics Association. He is a Distinguished Member and a Speaker of the ACM. He is the Editor-in-Chief of the Computers and Graphics and Graphics and Visual Computing journals. [48] C. I. Muntean, ‘‘Raising engagement in e-learning through gamiﬁcation,’’ in Proc. 6th Int. Conf. Virtual Learn. (ICVL), 2011, pp. 323–329. [49] I. B. Myers and M. H. McCaulley, Myers-Briggs Type Indicator: MBTI. Palo Alto, CA, USA: Consulting Psychologists Press, 1988. [50] A. H. Nabizadeh, A. M. Jorge, and J. P. Leal, ‘‘Long term goal oriented recommender systems,’’ in Proc. 11th Int. Conf. Web Inf. Syst. Technol. (WEBIST), Lisbon, Portugal, vol. 1, 2015, pp. 552–557. [51] A. H. Nabizadeh, A. Mário Jorge, and J. Paulo Leal, ‘‘RUTICO: Recommending successful learning paths under time constraints,’’ in Proc. Adjunct Publication 25th Conf. User Modeling, Adaptation Pers., Jul. 2017, pp. 153–158. [52] A. H. REFERENCES Gonçalves, ‘‘So fun it hurts— Gamifying an engineering course,’’ in Proc. Int. Conf. Augmented Cog- nition. Berlin, Germany: Springer, 2013, pp. 639–648. [31] D. Dicheva, K. Irwin, and C. Dichev, ‘‘OneUp: Engaging students in a gamiﬁed data structures course,’’ in Proc. 50th ACM Tech. Symp. Comput. Sci. Educ., Feb. 2019, pp. 386–392. [8] G. Barata, S. Gama, J. Jorge, and D. Gonçalves, ‘‘Gamiﬁcation for smarter learning: Tales from the trenches,’’ Smart Learn. Environ., vol. 2, no. 1, p. 10, Dec. 2015. [32] A. Domínguez, J. Saenz-de-Navarrete, L. de-Marcos, L. Fernández-Sanz, C. Pagés, and J.-J. Martínez-Herráiz, ‘‘Gamifying learning experi- ences: Practical implications and outcomes,’’ Comput. Educ., vol. 63, pp. 380–392, Apr. 2013. [9] G. Barata, S. Gama, J. Jorge, and D. Gonçalves, ‘‘Early prediction of student proﬁles based on performance and gaming preferences,’’ IEEE Trans. Learn. Technol., vol. 9, no. 3, pp. 272–284, Jul. 2016. [33] Z. Fitz-Walter, P. Wyeth, D. Tjondronegoro, and B. Scott-Parker, ‘‘Driven to drive: Designing gamiﬁcation for a learner logbook smartphone appli- cation,’’ in Proc. 1st Int. Conf. Gameful Design, Res., Appl., 2013, pp. 42–49. [10] G. Barata, S. Gama, J. Jorge, and D. Gonçalves, ‘‘Studying student dif- ferentiation in gamiﬁed education: A long-term study,’’ Comput. Hum. Behav., vol. 71, pp. 550–585, Jun. 2017. [34] N. Friedman, D. Geiger, and M. Goldszmidt, ‘‘Bayesian network classi- ﬁers,’’ Mach. Learn., vol. 29, no. 2, pp. 131–163, Nov. 1997. [11] G. Barata, S. Gama, J. A. P. Jorge, and D. J. V. Gonçalves, ‘‘Relating gaming habits with student performance in a gamiﬁed learning experi- ence,’’ in Proc. 1st ACM SIGCHI Annu. Symp. Comput.-Hum. Interact. Play, Oct. 2014, pp. 17–25. [35] M. Gachkova, M. Takev, and E. Somova, ‘‘Learning and assessment based on gamiﬁed e-course in moodle,’’ Math. Inform., vol. 61, no. 5, pp. 444–454, 2018. [36] C. V. Gipps, Beyond Testing: Towards a Theory of Educational Assessment. London, U.K.: Psychology Press, 1994, pp. 1–193. [12] R. Bartle, ‘‘Hearts, clubs, diamonds, spades: Players who suit MUDs,’’ J. MUD Res., vol. 1, no. 1, p. 19, 1996. 81030 81030 VOLUME 9, 2021 A. H. Nabizadeh et al.: How Do Students Behave in a Gamified Course? [37] L. Hakulinen and T. Auvinen, ‘‘The effect of gamiﬁcation on students with different achievement goal orientations,’’ in Proc. Int. Conf. Teach. Learn. Comput. Eng., Apr. 2014, pp. 9–16. [62] A. H. N. Rafsanjani, N. Salim, A. R. Aghdam, and K. B. REFERENCES Nabizadeh, ‘‘ISEA: IoT- based smartphone energy assistant for prompting energy-aware behav- iors in commercial buildings,’’ Appl. Energy, vol. 266, May 2020, Art. no. 114892. [61] A. H. N. Rafsanjani, ‘‘A long term goal recommender approach for learning environments,’’ Faculdade de Ciências da Universidade do Porto, Porto, Portugal, Tech. Rep. 101504217, 2018. 81031 VOLUME 9, 2021 VOLUME 9, 2021 VOLUME 9, 2021
https://openalex.org/W2135109214	https://europepmc.org/articles/pmc3707286?pdf=render	English	null	Prevalence and Correlates of Discomfort and Acceptability of Acupuncture among Outpatients in Chinese Acupuncture and Moxibustion Departments: A Cross-Sectional Study	Evidence-based complementary and alternative medicine	2,013	cc-by	5,472	Hindawi Publishing Corporation Evidence-Based Complementary and Alternative Medicine Volume 2013, Article ID 715480, 7 pages http://dx.doi.org/10.1155/2013/715480 Hindawi Publishing Corporation Evidence-Based Complementary and Alternative Medicine Volume 2013, Article ID 715480, 7 pages http://dx.doi.org/10.1155/2013/715480 Hindawi Publishing Corporation Evidence-Based Complementary and Alternative Medicine Volume 2013, Article ID 715480, 7 pages http://dx.doi.org/10.1155/2013/715480 Hindawi Publishing Corporation Evidence-Based Complementary and Alternative Medicine Volume 2013, Article ID 715480, 7 pages http://dx.doi.org/10.1155/2013/715480 Baoyan Liu,1 Huanfang Xu,1 Shengnan Guo,1,2 Jiani Wu,1 Jia Liu,1 Min Yee Lim,2,3 and Zhishun Liu1 1 Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing 100053, China 2 Beijing University of Chinese Medicine, Beijing 100029, China 3 China & Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798 Correspondence should be addressed to Zhishun Liu; liuzhishun@aliyun.com Received 4 February 2013; Revised 16 April 2013; Accepted 8 May 2013 Academic Editor: Jaung-Geng Lin Copyright © 2013 Baoyan Liu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Objective. This study aims to give a profile of discomfort and acceptability of acupuncture, including the prevalence and association with demographic and acupuncture-related factors. Methods. A cross-sectional study was conducted in Beijing, China. Outpatients of acupuncture and moxibustion departments were recruited using purposive sampling. 925 subjects were interviewed with an anonymous questionnaire. Multinomial and binary logistic regression were used to analyze factors affecting discomfort and acceptability of acupuncture. Results. The average VAS value of 925 subjects’ acupuncture discomfort was 2.66 ± 2.02, within the range of mild discomfort. Acupuncture was easily accepted by 81.1% of the subjects. Results of logistic regression were as follows: (1) subjects with a better knowledge of acupuncture, or a greater fear of pain or needles, experienced more “moderate to severe discomfort” and showed a decreased acupuncture acceptance (𝑃< 0.001 or 𝑃< 0.01); (2) Acupuncture with less discomfort or implemented by a more qualified doctor was easy to be accepted (𝑃< 0.001); (3) subjects aged 20–29 preferred to report “moderate to severe discomfort” while those aged 40–59 preferred to report “slight discomfort” (𝑃< 0.001). Conclusion. Acupuncture is an acceptable therapy with less discomfort, which can be greatly affected by fear of pain or needles, age, knowledge of acupuncture, and professional title of acupuncturist. Baoyan Liu,1 Huanfang Xu,1 Shengnan Guo,1,2 Jiani Wu,1 Jia Liu,1 Min Yee Lim,2,3 and Zhishun Liu1 1 Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing 100053, China 2 Beijing University of Chinese Medicine, Beijing 100029, China 3 China & Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798 Baoyan Liu,1 Huanfang Xu,1 Shengnan Guo,1,2 Jiani Wu,1 Jia Liu,1 Min Yee Lim,2,3 and Zhishun Liu1 1 Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing 100053, China 2 Beijing University of Chinese Medicine, Beijing 100029, China 3 China & Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798 Research Article Prevalence and Correlates of Discomfort and Acceptability of Acupuncture among Outpatients in Chinese Acupuncture and Moxibustion Departments: A Cross-Sectional Study Baoyan Liu,1 Huanfang Xu,1 Shengnan Guo,1,2 Jiani Wu,1 Jia Liu,1 Min Yee Lim,2,3 and Zhishun Liu1 1 Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing 100053, China 2 Beijing University of Chinese Medicine, Beijing 100029, China 3 China & Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798 1. Background Discomfort of acupuncture was measured by the question: “How is your feeling of discomfort caused by acupuncture?” Sub- jects were asked to answer the question within 10 minutes after acupuncture treatment. The severity of discomfort was marked using visual analogue scale (VAS), a numerical scale from 0 to 10 where 0 means no discomfort and 10 means the severest discomfort (see Figure 1). Severity of discomfort was evaluated by values of VAS [7]. A value of 0 on the VAS indicates no discomfort, 1 to 3 indicates mild discomfort, 4 to 6 moderate discomfort, and 7 to 10 indicates severe discomfort. In short, profiles of acupuncture discomfort and accept- ability are still uncertain. Moreover, personalized treatment is one of the acupuncture characteristics, and different con- ditions of patients directly influence the treatment protocol. Therefore, it is useful to understand factors associated with discomfort and acceptability of acupuncture. This pilot sur- vey focuses on the discomfort and acceptability of acupunc- ture, and the associated demographic and acupuncture- related factors, to provide useful suggestions on relieving discomfort of acupuncture and improving acceptance and personalized acupuncture treatment. 1. Background as sourness, numbness, distending, or pain [3]. According to TCM theory, sensations induced by acupuncture are closely related to deqi, a traditional acupuncture terminology which describes the connection between acupuncture needles and the energy pathways of the body and is essential for curative effect [4]. In our opinion, acupuncture sensations are in essence negative emotional experience for patient whether belonging to deqi or not. From this perspective, we named them discomfort of acupuncture. This study is designed to answer these two questions. To what degree do patients expe- rience discomfort of acupuncture? What are the influence factors? By finding answers for these two questions, acupunc- turists can change the acupuncture regimen accordingly, and thus improve patients’ experience of acupuncture. Acupuncture, as a main component of traditional Chinese medicine (TCM), has been adopted on diseases prevention and treatment for over 2,000 years in China. It exerts its effects through stimulating of acupoints with acupuncture needles, and thus triggers the body’s own ability to prevent diseases [1]. Due to its excellent efficacy, acupuncture has been increasingly accepted and used by practitioners and patients worldwidely. A fact is that acupuncture is a minimally invasive therapy, which may induce anxiety and fear to some patients [2]. When acupuncturists insert needles or perform acupuncture manipulation, patients may produce various sensations, such 2 2 Evidence-Based Complementary and Alternative Medicine 1 0 2 3 4 5 6 7 8 9 10 None Mild Moderate Severe Figure 1: Severity of discomfort of acupuncture by VAS. Though acupuncture is more and more used all over the world, there are few profiles on its discomfort, acceptability and influence factors. Discomfort of acupuncture strongly affects its acceptability. Our study showed that 44.2% (69/156) of the outpatients were reluctant to choose acupuncture because of fear of needles. In addition, acupuncture accept- ability could also be affected by patients’ characterizations, demographic data, and disease pattern: predominance, with physical symptoms such as diseases of the musculoskeletal system or injury, was the striking characteristic among acupuncture patients [5, 6]. Figure 1: Severity of discomfort of acupuncture by VAS. 2.2.2. Discomfort of Acupuncture. Discomfort of acupunc- ture refers to certain sensations felt by subjects during the treatment, such as aching, soreness, distension, heaviness, numbness, or dull pain that causes negative emotions, ner- vousness, fears, or hostility toward acupuncture. 2. Methods The data was collected by two postgraduate students who majored in acupuncture and received formal training on question- naire interviewing. To ensure research quality, a supervisor performed a spot check on completed questionnaires for completeness and consistency at the time of interview. 2.2.6. Data Collection. The questionnaire was first given to 5 outpatients to assure clarity of concepts for subjects. The data was collected by two postgraduate students who majored in acupuncture and received formal training on question- naire interviewing. To ensure research quality, a supervisor performed a spot check on completed questionnaires for completeness and consistency at the time of interview. 2. Methods 2.1. Study Design. A cross-sectional survey was conducted between May and August 2010 to assess the prevalence of dis- comfort and acceptability of acupuncture and associated fac- tors. The study was conducted with a purposive sampling in three hospitals in Beijing: Guang’anmen Hospital affiliated to China Academy of Chinese Medical Sciences, Dongzhimen Hospital affiliated to Beijing University of Chinese Medicine, and Dongfang Hospital affiliated to Beijing University of Chinese Medicine, which were selected from a total of six top traditional Chinese medicine hospitals firstly, and then every two acupuncture consulting rooms were randomly selected from each of the three hospitals. 2.2.3. Acceptability of Acupuncture. Acceptability of acupu- ncture was assessed by the question: “How is your accept- ability of acupuncture?” There were two options for selection, “Difficult to accept” or “Easy to accept.” 2.2.4. Demographic Profile. To understand the influence of demographic characteristics toward discomfort and accept- ability of acupuncture, the gender, age and education level were taken into consideration. Age was classified into seven segments: ≤19, 20–29, 30–39, 40–49, 50–59, 60–69, and ≥70. Educational level was determined based on the International Standard Classification of Education (ISCED) [8]. We classi- fied educational level into three categories: primary education and below (ISCED level: 0-1), secondary education (ISCED level: 2–4), and tertiary education (ISCED level: 5–8). Outpatients from the six consulting rooms at the three selected hospitals, who were willing to be interviewed and were able to complete the questionnaire, participated in the survey. The subjects were asked to do the survey within 10 minutes after completing acupuncture treatment. As we did not collect any personal identifiable information, voluntary provision of information was deemed to be consent. Subjects were invited to complete an anonymous questionnaire (for more details see Supplementary Material available online at http://dx.doi.org/10.1155/2013/715480) under the guidance of trained interviewers. A total of 928 subjects were recruited from May 4th to August 31st. Data from 925 subjects were included in the analysis. Three incomplete questionnaires were not included. 2.2.5. Acupuncture-Related Information. To explore acupu- ncture-related factors affecting discomfort and acceptability of acupuncture, five items were adopted in the questionnaire. These items included whether it was the first acupuncture experience, whether subjects were afraid of acupuncture needles, the extent of fear of pain, subjects’ knowledge of acupuncture, and professional title of acupuncturist. 2.2.6. Data Collection. The questionnaire was first given to 5 outpatients to assure clarity of concepts for subjects. 4. Results 4.1. Discomfort of Acupuncture. The average VAS value of 925 subjects’ acupuncture discomfort was 2.66 ± 2.02, within the range of mild discomfort. 146 subjects (15.8%) did not feel any discomfort; the majority of subjects (53.9%) felt slight discomfort; 188 subjects (20.3%) felt moderate discomfort; and 92 (9.9%) felt severe discomfort. “Moderate discomfort” and “severe discomfort” responses were analyzed together (Table 1). 4.6. Factors Affecting Acceptability of Acupuncture. The influ- ence of variables (i.e., discomfort of acupuncture, demo- graphic, and acupuncture-related factors) on acceptability of acupuncture was analyzed by binary logistic regression with backward stepwise. During the analysis procedure, variables of gender, educational level, and age were removed from the equation successively. The residual variables were signifi- cantly associated with acceptability of acupuncture, including discomfort of acupuncture, first acupuncture experience, fear of needles, fear of pain, knowledge of acupuncture, and professional title of acupuncturist (Table 3). 4.2. Acceptability of Acupuncture. Among 925 subjects, 750 subjects (81.1%) reported that acupuncture was easy to be accepted, and 175 (18.9%) reported that acupuncture was difficult to be accepted (Table 1). 4.2. Acceptability of Acupuncture. Among 925 subjects, 750 subjects (81.1%) reported that acupuncture was easy to be accepted, and 175 (18.9%) reported that acupuncture was difficult to be accepted (Table 1). 4.3. Demographic Profile. Out of 925 subjects (mean age ± standard deviation: 47.97 ± 2.00 years), 40.2% were male and 59.8% were female. 33.7% had completed tertiary education, 58.6% secondary, and 7.7% primary education (Table 1). 4.3. Demographic Profile. Out of 925 subjects (mean age ± standard deviation: 47.97 ± 2.00 years), 40.2% were male and 59.8% were female. 33.7% had completed tertiary education, 58.6% secondary, and 7.7% primary education (Table 1). Acupuncture discomfort and professional title of acupuncturist were positively associated with acupuncture acceptability. 3. Statistical Analysis Results showed that subjects reported more “moderate to severe discomfort” than “no discomfort” in several con- ditions: (1) if they had a fear of needles (OR = 0.26, 95% CI = 0.14–0.49); (2) if they were afraid of pain (a little versus not at all: OR = 0.22, 95% CI = 0.13–0.37; very much versus not at all: OR = 0.24, 95% CI = 0.14–0.41); (3) if they showed a better knowledge of acupuncture (a little versus not at all: OR = 0.43, 95% CI = 0.26–0.71, very well versus not at all: OR = 0.40, 95% CI = 0.21–0.77); and (4) if they were aged 20–29 (20–29 versus ≥70: OR = 0.35, 95% CI = 0.13–0.90). Statistical analysis was performed using SPSS statistics 18.0. Binary logistic regression with backward stepwise was used to assess the relationship between acceptability of acupuncture and demographic, discomfort, or acupuncture-related vari- ables. Chi-square test was used to test for association between acupuncture discomfort and demographic or acupuncture- related variables; variables that were shown to be significantly associated with discomfort of acupuncture were entered into multinomial logistic regression. 𝑃< 0.05 was considered statistically significant. In the comparisons between “slight discomfort” and “moderate to severe discomfort,” fear of needles, fear of pain and age rather than knowledge of acupuncture, showed significant differences. Subjects experienced more “moderate to severe discomfort” than “slight discomfort” if they were afraid of needles (OR = 0.50, 95% CI = 0.36–0.72) and pain (a little versus not at all: OR = 0.47, 95% CI = 0.32–0.69; very much versus not at all: OR = 0.50, 95% CI = 0.33–0.75), while subjects aged 40–49 reported more “slight discomfort” (40– 49 versus ≥70: OR = 1.85, 95% CI = 1.04–3.30). 4. Results Table 2 showed the multinomial logistic regression predicting the odds of reporting acupuncture discomfort as no discomfort or slight discomfort. 4.5. Factors Affecting Discomfort of Acupuncture. Results of chi-square tests showed that age (𝜒2 = 32.83, 𝑃= 0.001), fear of needles (𝜒2 = 34.15, 𝑃 < 0.01), knowledge of acupuncture (𝜒2 = 13.17, 𝑃 = 0.01) and fear of pain (𝜒2 = 53.75, 𝑃< 0.01), were significantly different among varying degrees of acupuncture discomfort. Table 2 showed the multinomial logistic regression predicting the odds of reporting acupuncture discomfort as no discomfort or slight discomfort. 2.2. Data Collection 2.2.1. Information Collected. We collected data on sub- jects’ discomfort of acupuncture (one item), acceptability of acupuncture (one item), demographic profile (three items), and acupuncture-related information (five items). 3 Evidence-Based Complementary and Alternative Medicine Evidence-Based Complementary and Alternative Medicine 4. Results A lower discomfort (no discomfort versus moderate to severe discomfort: OR = 3.11, 95% CI = 1.45– 6.72, 𝑃= 0.004; slight discomfort versus moderate to severe discomfort: OR = 1.90, 95% CI = 1.29–2.82, 𝑃= 0.001) and a higher professional title of acupuncturist (senior title versus primary title: OR = 3.22, 95% CI = 1.95–5.34, 𝑃< 0.001; middle title versus primary title: OR = 2.59, 95% CI = 1.58– 4.23, 𝑃< 0.001) were significantly associated with greater willingness to accept acupuncture, while other variables were significantly associated with a decreased willingness to accept acupuncture, referring primarily to first acupuncture experience (OR = 0.62, 95% CI = 0.39–0.99, 𝑃= 0.047), a greater fear of pain (very much versus not at all: OR = 0.18, 95% CI = 0.10–0.31, 𝑃< 0.001; a little versus not at all: OR = 0.42, 95% CI = 0.24–0.74, 𝑃< 0.001), a better knowledge of acupuncture (very well versus not at all: OR = 0.32, 95% CI = 0.16–0.61, 𝑃= 0.001; a little versus not at all: OR = 0.32, 95% CI = 0.18–0.56, 𝑃< 0.001), and fear of needles (OR = 0.34, 95% CI = 0.22–0.51, 𝑃< 0.001). 4.4. Acupuncture-Related Factors. There were 17.4% of sub- jects receiving acupuncture treatment for the first time, and 21.2% of subjects reported fear of needles. When asked how much they were afraid of pain, 32.9% of subjects chose “not at all”, 30.9% and 36.2% chose “very much” and “a little,” respectively. 4.4. Acupuncture-Related Factors. There were 17.4% of sub- jects receiving acupuncture treatment for the first time, and 21.2% of subjects reported fear of needles. When asked how much they were afraid of pain, 32.9% of subjects chose “not at all”, 30.9% and 36.2% chose “very much” and “a little,” respectively. Overall, most of the subjects were characterized by the following features: the previous acupuncture experience (82.6%), fear of acupuncture needles (78.8%), fear of pain (67.1%), little knowledge of acupuncture (80.8%), and seeking help from more qualified acupuncturists (82.9%). Details are shown in Table 1. 4.5. Factors Affecting Discomfort of Acupuncture. Results of chi-square tests showed that age (𝜒2 = 32.83, 𝑃= 0.001), fear of needles (𝜒2 = 34.15, 𝑃 < 0.01), knowledge of acupuncture (𝜒2 = 13.17, 𝑃 = 0.01) and fear of pain (𝜒2 = 53.75, 𝑃< 0.01), were significantly different among varying degrees of acupuncture discomfort. 5. Discussion In present study, a female predominance was observed (female: male = 1.48 : 1), and the age distribution displayed Evidence-Based Complementary and Alternative Medicine 4 Table 1: Characteristics of subjects. Variables Number % How’s your feeling of discomfort caused by acupuncture? No discomfort (VAS = 0) 146 15.8 Slight discomfort (1 ≤VAS ≤3) 499 53.9 Moderate to severe discomfort (4 ≤VAS ≤10) 280 30.3 How is your acceptability of acupuncture? Difficult to accept 175 18.9 Easy to accept 750 81.1 Gender Female 553 59.8 Male 372 40.2 Age ≤19 57 6.2 20–29 140 15.1 30–39 112 12.1 40–49 147 15.9 50–59 229 24.8 60–69 130 14.1 ≥70 110 11.9 Educational level Primary education and below (ISCED level: 0-1) 71 7.7 Secondary education (ISCED level: 2–4) 542 58.6 Tertiary education (ISCED level: 5–8) 312 33.7 First acupuncture experience: Is this your first acupuncture experience? No 764 82.6 Yes 161 17.4 Fear of needles: Are you afraid of acupuncture needles? No 729 78.8 Yes 196 21.2 Fear of pain: How much are you afraid of pain? Not at all 304 32.9 A little 335 36.2 Very much 286 30.9 Knowledge of acupuncture: How much do you know about acupuncture? Not at all 217 23.5 A little 530 57.3 Very well 178 19.2 Professional title of acupuncturist Resident physician 158 17.1 Attending physician 363 39.2 Chief physician 404 43.7 a peak at around fifties. Our findings were consistent with previous report [9] The educational levels of subjects in our easily, which was consistent with the finding that 81% of subjects considered acupuncture process to be comfortable Number % 146 15.8 499 53.9 280 30.3 175 18.9 750 81.1 553 59.8 372 40.2 57 6.2 140 15.1 112 12.1 147 15.9 229 24.8 130 14.1 110 11.9 71 7.7 542 58.6 312 33.7 764 82.6 161 17.4 729 78.8 196 21.2 304 32.9 335 36.2 286 30.9 217 23.5 530 57.3 178 19.2 158 17.1 363 39.2 404 43.7 Variables Variables Number How’s your feeling of discomfort caused by acupuncture? No discomfort (VAS = 0) 146 Slight discomfort (1 ≤VAS ≤3) 499 Moderate to severe discomfort (4 ≤VAS ≤10) 280 How is your acceptability of acupuncture? 5. Discussion Difficult to accept 175 Easy to accept 750 Gender Female 553 Male 372 Age ≤19 57 20–29 140 30–39 112 40–49 147 50–59 229 60–69 130 ≥70 110 Educational level Primary education and below (ISCED level: 0-1) 71 Secondary education (ISCED level: 2–4) 542 Tertiary education (ISCED level: 5–8) 312 First acupuncture experience: Is this your first acupuncture experience? No 764 Yes 161 Fear of needles: Are you afraid of acupuncture needles? No 729 Yes 196 Fear of pain: How much are you afraid of pain? Not at all 304 A little 335 Very much 286 Knowledge of acupuncture: How much do you know about acupuncture? Not at all 217 A little 530 Very well 178 Professional title of acupuncturist Resident physician 158 Attending physician 363 Chief physician 404 easily, which was consistent with the finding that 81% of subjects considered acupuncture process to be comfortable and relaxing [10]. a peak at around fifties. Our findings were consistent with previous report [9]. The educational levels of subjects in our study were mainly secondary education or higher. Majority of subjects believed that acupuncture induced slight or no discomfort (69.7%) and was easy to be accepted (81.1%). There were 11.4% of subjects considering that acupuncture was easy to be accepted, although they chose moderate or severe discomfort. This may be interpreted by subjects’ strong expectation for good effectiveness of acupuncture. Overall, acupuncture causes a little discomfort and it can be accepted Till now, there were few studies on acupuncture discom- fort and acceptability. Fear of pain or needles, as negative emotional experience, could cause greater discomfort. Multi- nomial regression analysis showed that subjects experienced more discomfort if he or she had a stronger fear toward pain or needles. Compared with subjects aged ≥70, subjects aged 20–29 preferred to report “moderate to severe discomfort” Evidence-Based Complementary and Alternative Medicine 5 Table 2: Multinomial logistic regression: Odds of subjects reporting less discomfort of acupuncture (𝑛= 925). Characteristic/subcategory Discomfort of acupuncture: No. 5. Discussion (%) No discomforta Slight discomforta No Slight Moderate to severe OR (95% CI) 𝑃value OR (95% CI) 𝑃value Age ≥70 20 (18.1) 51 (46.4) 39 (35.5) 1.00 1.00 ≤19 7 (12.3) 35 (61.4) 15 (26.3) 1.12 (0.37–3.41) 0.841 2.02 (0.94–4.32) 0.070 20–29 8 (5.7) 76 (54.3) 56 (40.0) 0.35 (0.13–0.90) 0.030 1.16 (0.66–2.03) 0.605 30–39 11 (9.8) 66 (58.9) 35 (31.3) 0.78 (0.32–1.94) 0.596 1.57 (0.86–2.85) 0.142 40–49 22 (15.0) 88 (59.9) 37 (25.1) 1.37 (0.62–3.02) 0.443 1.85 (1.04–3.30) 0.037 50–59 52 (22.7) 114 (49.8) 63 (27.5) 1.69 (0.84–3.39) 0.138 1.39 (0.82–2.37) 0.226 60–69 26 (20.0) 69 (53.1) 35 (26.9) 1.21 (0.55–2.65) 0.638 1.31 (0.71–2.39) 0.379 Fear of needles: Are you afraid of acupuncture needles? No 132 (18.1) 407 (55.8) 190 (26.1) 1.00 1.00 Yes 14 (7.2) 92 (46.9) 90 (45.9) 0.25 (0.14–0.48) <0.001 0.50 (0.36–0.72) <0.001 Fear of pain: How much are you afraid of pain? Not at all 79 (26.0) 171 (56.3) 54 (17.8) 1.00 1.00 A little 36 (10.7) 176 (52.5) 123 (36.7) 0.22 (0.13–0.37) <0.001 0.47 (0.32–0.69) <0.001 Very much 31 (10.8) 152 (53.1) 103 (36.0) 0.24 (0.14–0.41) <0.001 0.50 (0.33–0.75) 0.001 Knowledge of acupuncture: How much do you know about acupuncture? Not at all 51 (23.5) 110 (50.7) 56 (25.8) 1.00 1.00 A little 72 (13.6) 291 (54.9) 167 (31.5) 0.43 (0.26–0.71) 0.001 0.85 (0.58–1.24) 0.394 Very well 23 (12.9) 98 (55.1) 57 (32.0) 0.40 (0.21–0.77) 0.006 0.82 (0.51–1.32) 0.423 OR: odds ratio; CI: confidence interval; aThe reference category is “moderate to severe discomfort”. able 2: Multinomial logistic regression: Odds of subjects reporting less discomfort of acupuncture (𝑛= 925). while those aged 40–59 preferred to report “slight discom- fort.” It was unexpected that a better knowledge of acupunc- ture led to more discomfort. In clinical practice, some patients showed great willingness to acupuncture treatment because of good effectiveness; however, their nervousness did not decrease with treatment sessions. Considering this fact, the result of knowledge of acupuncture may be some- what understood. Nevertheless, further research is needed to understand the relationship between the knowledge and acupuncture discomfort. In addition, the previous acupunc- ture experience, educational level, and professional title of acupuncturist showed no significant influence on discomfort. so did the factor of gender. did not find a significant gender difference toward discomfort of acupuncture, which is similar to the result of systematic review. Evidence-Based Complementary and Alternative Medicine 6 Table 3: Binary logistic regression: Odds of subjects reporting better acceptance of acupuncture (𝑛= 925). Characteristic/Subcategory Acceptance of acupuncture Odds Ratios Difficult to accept Easy to accept OR (95% CI) 𝑃value First acupuncture experience: Is this your first acupuncture experience? No 135 (17.7) 629 (82.3) 1.00 Yes 40 (24.8) 121 (75.2) 0.62 (0.39–0.99) 0.047 Fear of needles: Are you afraid of acupuncture needles? No 95 (13.0) 634 (87.0) 1.00 Yes 80 (40.8) 116 (59.2) 0.34 (0.22–0.51) <0.001 Fear of pain: How much are you afraid of pain? Not at all 21 (12.0) 283 (93.1) 1.00 1.00 A little 63 (18.8) 272 (81.2) 0.42 (0.24–0.74) 0.002 Very much 91 (31.8) 195 (68.2) 0.18 (0.10–0.31) <0.001 Knowledge of acupuncture: How much do you know about acupuncture? Not at all 20 (9.2) 197 (90.8) 1.00 A little 119 (22.5) 411 (77.5) 0.32 (0.18–0.56) 0.001 Very well 36 (20.2) 142 (79.8) 0.32 (0.16–0.61) <0.001 Professional title of acupuncturist Resident physician 59 (37.3) 99 (62.7) 1.00 Attending physician 63 (17.4) 300 (82.6) 2.59 (1.58–4.23) <0.001 Chief physician 53 (13.1) 351 (86.9) 3.22 (1.95–5.34) <0.001 Discomfort of acupuncture: How is your discomfort feeling caused by acupuncture? Moderate to severe discomfort 87 (31.1) 193 (68.9) 1.00 Slight discomfort 79 (15.8) 420 (84.2) 1.90 (1.29–2.82) 0.001 No discomfort 9 (6.2) 137 (93.8) 3.11 (1.45–6.72) 0.004 OR: odds ratio; CI: confidence interval. Table 3: Binary logistic regression: Odds of subjects reporting better acceptance of acupuncture (𝑛= 925). our study sample may not be sufficiently representative. Sec- ondly, other factors that may affect acupuncture discomfort and acceptability were not included in this study, such as details of sociodemographic information (income, occupa- tion, marital status, etc.), diagnosis of diseases, needling location selecting, and the needling depth. In addition, only the severities of the sensations induced by acupuncture were recorded; the type of sensation was neglected. Acupuncture sensations were categorized into two types, namely, sensa- tions that cannot hurt as deqi, mainly including aching, soreness, and pressure, followed by tingling, numbness, dull pain, heaviness, warmth, fullness, and coolness, and sensations that can hurt as noxious stimulation, for example, sharp pain [17–19]. A certain type of sensation may contribute to both acupuncture discomfort and deqi. By neglecting the type of discomfort, we cannot differentiate useful discomfort constituting deqi from actual discomfort sensations. 5. Discussion Binary logistic regression analysis of acupuncture accept- ability showed that subjects could accept acupuncture more easily if they had less discomfort (no or slight discomfort) and were treated by acupuncturist with a higher professional title. However, they may have great difficulties in accepting acupuncture if they had a fear of pain or needles and a better knowledge of acupuncture. The more serious the negative experiences (e.g., discomfort, fear of pain and acupuncture) are, the more difficult the acupuncture is accepted by sub- jects. Patient without acupuncture experience showed more difficulty in accepting acupuncture treatment. However, the impact of the first acupuncture experience on acupuncture discomfort was little enough to be disregarded (𝑃= 0.047). Acupuncture discomfort, which is a negative emotion like pain, can be predicted to be influenced by gender. The relationship between gender differences and pain has been reported a lot, but the conclusions are different. Most reports showed that males exhibited greater pain tolerance than females [11, 12], but a systematic review failed to show a clear and consistent pattern of gender differences in pain sensitivity [13]. Paradigm to investigate the role of gender in pain perception was mainly based on laboratory-induced thermal, pressure, chemical, or visceral pain, but the application of the paradigms for clinic pain is questionable [13]. In our study, we Subjects could consider acupuncture easy to be accepted when they were treated by an acupuncturist with a higher professional title, consistent with the previous report [14]. Our results showed that better knowledge of acupuncture may cause more discomfort, and lead to a decreased willing- ness to acupuncture acceptance. A possible reason was that subjects gained acupuncture knowledge mainly from their own experience, in which negative experience tended to be 6 Evidence-Based Complementary and Alternative Medicine Evidence-Based Complementary and Alternative Medicine Finally, due to the cross sectional nature of this study, our findings should be interpreted as associations rather than implying causality. exaggerated. Further research is needed to understand the relationship of the knowledge, discomfort and acceptability of acupuncture. Overall, fear of needles and pain can cause more discom- fort, resulting in decreased acceptance. However, it is worth mentioning that pain endurance can be influenced by pain- related self-efficacy and positive self-instruction [11]. Positive outcome expectancy also indicates a marked improvement in patients’ self-reports of anxiety, pain and distress [15, 16]. Therefore, communication and encouragement before acupuncture can hopefully improve patients’ fear of pain and acupuncture, thus reducing discomfort and increasing acceptance. Our results showed that a senior professional title of acupuncturist had barely any impact on discomfort, but it can help to increase acupuncture acceptance.f Due to limited staff and resources, our study was com- pleted only in six acupuncture consulting rooms from three top TCM hospitals in Beijing; lower grade hospitals and communities were not included. Besides, purposive sampling instead of random sampling was used in the study. Therefore 7 Evidence-Based Complementary and Alternative Medicine 6. Conclusions population: the Taiwan experience,” Journal of Alternative and Complementary Medicine, vol. 12, no. 4, pp. 379–387, 2006. In conclusion, acupuncture is an acceptable therapy with less discomfort. Discomfort and acceptability of acupuncture can be affected by fear of pain, fear of needles, age, knowledge of acupuncture, and professional title of acupuncturist. Based on our results, methods to relieve discomfort and improve acceptance of acupuncture can be taken accordingly. [10] J. J. Mao, J. T. Farrar, K. Armstrong, A. Donahue, J. Ngo, and M. A. Bowman, “De qi: Chinese acupuncture patients’ experi- ences and beliefs regarding acupuncture needling sensation— an exploratory survey,” Acupuncture in Medicine, vol. 25, no. 4, pp. 158–165, 2007. [11] A. K. Schmitz, M. Vierhaus, and A. Lohaus, “Pain tolerance in children and adolescents: sex differences and psychosocial influences on pain threshold and endurance,” European Journal of Pain, vol. 17, no. 1, pp. 124–131, 2013. The authors declare no conflict of interests. The authors declare no conflict of interests. Authors’ Contribution [13] M. Racine, Y. Tousignant-Laflamme, L. A. Kloda, D. Dion, G. Dupuis, and M. Choini`ere, “A systematic literature review of 10 years of research on sex/gender and experimental pain perception—part 1: are there really differences between women and men?” Pain, vol. 153, no. 3, pp. 602–618, 2012. Z. Liu and B. Liu designed this study and revised the paper. HF. Xu drafted the paper and performed the statistical analysis. SN. Guo and JN. Wu participated in data acquisition. M. Y. Lim and J. Liu revised the paper. All the authors read and approved the final paper. [14] F. L. Bishop, Y. Massey, L. Yardley, and G. T. Lewith, “How patients choose acupuncturists: a mixed-methods project,” Journal of Alternative and Complementary Medicine, vol. 17, no. 1, pp. 19–25, 2011. Acknowledgment [15] R. Crow, H. Gage, S. Hampson, J. Hart, A. Kimber, and H. Thomas, “The role of expectancies in the placebo effect and their use in the delivery of health care: a systematic review,” Health Technology Assessment, vol. 3, no. 3, pp. 1–96, 1999. This study was supported by National Key Technology R&D Program during the Twelfth Five-year Plan Period (no. 2012BAI24B01). This study was supported by National Key Technology R&D Program during the Twelfth Five-year Plan Period (no. 2012BAI24B01). [16] J. J. Mao, K. Armstrong, J. T. Farrar, and M. A. Bowman, “Acupuncture expectancy scale: development and preliminary validation in China,” Explore, vol. 3, no. 4, pp. 372–377, 2007. Conflict of Interests [12] T. Thompson, E. Keogh, and C. C. French, “Sensory focusing versus distraction and pain: moderating effects of anxiety sensitivity in males and females,” Journal of Pain, vol. 12, no. 8, pp. 849–858, 2011. References [1] M. T. Cab´yoglu, N. Ergene, and U. Tan, “The mechanism of acupuncture and clinical applications,” International Journal of Neuroscience, vol. 116, no. 2, pp. 115–125, 2006. [17] K. K. K. S. Hui, T. N. Sporko, M. G. Vangel, M. Li, J. Fang, and L. Lao, “Perception of Deqi by Chinese and American acupuncturists: a pilot survey,” Chinese Medicine, vol. 6, article 2, 2011. [2] D. P. Lu and G. P. Lu, “Clinical management of needle- phobia patients requiring acupuncture therapy,” Acupuncture and Electro-Therapeutics Research, vol. 24, no. 3-4, pp. 189–201, 1999. [18] J. Park, H. Park, H. Lee, S. Lim, K. Ahn, and H. Lee, “Deqi sensation between the acupuncture-experienced and the Na¨ıve: a Korean study II,” American Journal of Chinese Medicine, vol. 33, no. 2, pp. 329–337, 2005. [3] H. MacPherson and A. Asghar, “Acupuncture needle sensations associated with De Qi: a classification based on experts’ ratings,” Journal of Alternative and Complementary Medicine, vol. 12, no. 7, pp. 633–637, 2006. [19] K. K. S. Hui, E. E. Nixon, M. G. Vangel et al., “Characterization of the “Deqi” response in acupuncture,” BMC Complementary and Alternative Medicine, vol. 7, article 33, 2007. [4] J. Kong, R. Gollub, T. Huang et al., “Acupuncture De Qi, from qualitative history to quantitative measurement,” Journal of Alternative and Complementary Medicine, vol. 13, no. 10, pp. 1059–1070, 2007. [5] V. Napadow and T. J. Kaptchuk, “Patient characteristics for outpatient acupuncture in Beijing, China,” Journal of Alternative and Complementary Medicine, vol. 10, no. 3, pp. 565–572, 2004. [6] A. Gould and H. MacPherson, “Patient perspectives on out- comes after treatment with acupuncture,” Journal of Alternative and Complementary Medicine, vol. 7, no. 3, pp. 261–268, 2001. [7] A. M. Kelly, “The minimum clinically significant difference in visual analogue scale pain score does not differ with severity of pain,” Emergency Medicine Journal, vol. 18, no. 3, pp. 205–207, 2001. [8] “International Standard Classification of Education (ISCED, 2011),” http://www.uis.unesco.org/Education/Documents /isced-2011-en.pdf. [9] F. P. Chen, Y. Y. Kung, T. J. Chen, and S. J. Hwang, “Demo- graphics and patterns of acupuncture use in the Chinese
https://openalex.org/W4391818989	https://www.frontiersin.org/articles/10.3389/fbuil.2024.1363804/pdf?isPublishedV2=False	English	null	Experimental investigations on the compaction energy for a robotic rammed earth process	Frontiers in built environment	2,024	cc-by	12,102	Linus Paul Schmitz1†, Joschua Gosslar1†, Evelien Dorresteijn2, Dirk Lowke2,3 and Harald Kloft1 Linus Paul Schmitz1†, Joschua Gosslar1†, Evelien Dorresteijn2, Dirk Lowke2,3 and Harald Kloft1 1Institute for Structural Design, Technische Universität, Braunschweig, Germany, 2Institute for Building Materials, Concrete Construction and Fire Safety, Technische Universität, Braunschweig, Germany, 3Deaprtment of Materials Engineering, Technical University of Munich, Munich, Germany Rammed earth is a construction material with a long history of traditional manufacturing. Due to its low environmental impact, positive impact on indoor climate and completely recyclable nature, its demand is also increasing in modern construction industry. However, as a consequence of the predominantly manual manufacturing processes, the production of rammed earth components is both inefﬁcient and costly. Through the implementation of automated and robot-aided fabrication processes in the ﬁeld of rammed earth construction, the opportunity to advance the digitalization of the ﬁeld can raise to a new level. In this paper, general studies on the interrelation of process and material parameters and their inﬂuence on the compaction results were conducted as a basis for the development of a prototypic robotic manufacturing process. The results show that reducing the layer height can signiﬁcantly decrease the impact energy. Additionally, it was shown that there is a minimum number of strokes and a minimum ramming frequency required for sufﬁcient compaction. Furthermore, a possible workﬂow for a speciﬁc control of the required dry density through variation of the compaction energy with regard to the present moisture content was identiﬁed. Front. Built Environ. 10:1363804. doi: 10.3389/fbuil.2024.1363804 COPYRIGHT © 2024 Schmitz, Gosslar, Dorresteijn, Lowke and Kloft. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. KEYWORDS robotic rammed earth (RRE), earth construction, additive manufacturing (AM), digital fabrication, compaction energy, impact energy, layer height robotic rammed earth (RRE), earth construction, additive manufacturing (AM), digital fabrication, compaction energy, impact energy, layer height frontiersin.org OPEN ACCESS EDITED BY Pshtiwan Shakor, Sulaimani Polytechnic University, Iraq REVIEWED BY Anastasiia Puzatova, Immanuel Kant Baltic Federal University, Russia Sathvik Sharath Chandra, Dayananda Sagar College of Engineering, India CORRESPONDENCE Joschua Gosslar, j.gosslar@tu-braunschweig.de †These authors share ﬁrst authorship RECEIVED 31 December 2023 ACCEPTED 26 January 2024 PUBLISHED 14 February 2024 CITATION Schmitz LP, Gosslar J, Dorresteijn E, Lowke D and Kloft H (2024), Experimental investigations on the compaction energy for a robotic rammed earth process. Front. Built Environ. 10:1363804. doi: 10.3389/fbuil.2024.1363804 EDITED BY Pshtiwan Shakor, Sulaimani Polytechnic University, Iraq REVIEWED BY Anastasiia Puzatova, Immanuel Kant Baltic Federal University, Russia Sathvik Sharath Chandra, Dayananda Sagar College of Engineering, India CORRESPONDENCE Joschua Gosslar, j.gosslar@tu-braunschweig.de †These authors share ﬁrst authorship RECEIVED 31 December 2023 ACCEPTED 26 January 2024 PUBLISHED 14 February 2024 CITATION Schmitz LP, Gosslar J, Dorresteijn E, Lowke D and Kloft H (2024), Experimental investigations on the compaction energy for a robotic rammed earth process. Front. Built Environ. 10:1363804. doi: 10.3389/fbuil.2024.1363804 COPYRIGHT © 2024 Schmitz, Gosslar, Dorresteijn, Lowke and Kloft. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. EDITED BY Pshtiwan Shakor, Sulaimani Polytechnic University, Iraq TYPE Original Research PUBLISHED 14 February 2024 DOI 10.3389/fbuil.2024.1363804 TYPE Original Research PUBLISHED 14 February 2024 DOI 10.3389/fbuil.2024.1363804 1 Introduction The construction sector is responsible for about 40 percent of the global energy demand. This is due to the processing of materials and their transport to the construction site, as well as the construction, operation and demolition of the buildings themselves (United nations environment program, 2009). Because of the climate crisis and the limitation of natural resources, the construction industry is forced to ﬁnd alternative materials and construction processes (Khadka and Shakya, 2015). For this, rammed earth offers the potential of a sturdy and environmentally friendly option (Windstorm and Schmidt, 2013). In addition to the possible utilization of exclusively natural components, it is noteworthy that clay, which acts as the binder in earthen construction, is readily available in many parts of the world and can be effectively processed locally. This avoids costs as well as green-house gas emissions due to transport. Likewise, structures can be easily deconstructed and recycled at the end of their use-phase as long as no additives such as cement have been used (Khadka and Shakya, 2015; Giuffrida et al., 2019; Niroumand et al., 2021). In addition, earthen materials possess various advantages in terms of building physics: Massive earthen components provide excellent Frontiers in Built Environment 01 frontiersin.org 10.3389/fbuil.2024.1363804 10.3389/fbuil.2024.1363804 Schmitz et al. As a result of the radical change in the production process, current knowledge about the manufacturing of rammed earth has to be revised and complemented by customized experimental tests. Also, due to the digitalized process, basic input data is needed. This stands in contrast to current rammed earth practice, which is mostly based on the verbal transmission of knowledge with loose descriptions of geometric relations (Ávila et al., 2020) and characterized by a lack of standards (Morel et al., 2021; Kloft et al., 2023). To achieve an efﬁcient and optimized process, it is necessary to ﬁrst determine the required compaction energy to achieve the desired density in the material. Additionally, the minimum impact energy necessary for a sufﬁcient compaction must be identiﬁed. Following this logic, investigations into the compaction process will be undertaken within the scope of this research. The test results should give initial information about the compaction behavior in RRE and provide valuable data for further improvement of the general process. Likewise, relevant process parameters, their ranges and their inﬂuence on the compaction results are examined and deﬁned in more detail. 2 Principles of earth compaction The structural principle of the rammed earth technique is based on the concept of soil compaction. By applying compaction energy within the conﬁnes of a formwork, the loose soil is densiﬁed. Rammed earth material can be described as a three-phase system which consists of solid particles, water and air. The compaction process brings the particles closer together as the air is pressed out, resulting in a reduction of volume and, consequently, increased density (Reddy, 2022). Which is of importance as an increased density can lead to an increased compression strength. The compaction process is known from various applications of geotechnical engineering, such as embankments or pavement base courses (Volhard and Ulrich, 2002). Although the compaction tools, such as vibratory rollers differ (Reddy, 2022), the theoretical knowledge on the behavior of earthen materials in the context of construction can be transferred (Giuffrida et al., 2019; Niroumand et al., 2021). The amount of water bound in the material is of particular signiﬁcance since water absorbed by the clay particles affects the material’s plasticity, making compaction easier. Meaning, the water acts as a sort of lubricant (Attom, 1997; Reddy, 2022). Through subsequent drying, the material ﬁnally acquires its full strength (Jaquin et al., 2009). In our research, a robotic and fully automated process for building structural wall elements, called Robotic Rammed Earth (RRE), is being investigated. The main approach is to replace the time-consuming formwork related work with an active slip form combined with automated feeding and compaction processes (Kloft et al., 2023). In Figure 1 the setup is shown, consisting of a six-axis robot arm suspended from a three-axis gantry, guiding the slipform- and compaction unit (left) and a ﬁve-axis, CNC-controlled hopper for controlled feeding of loose material (right). Compared to traditional manufacturing, where the layer height is up to 15 cm (Hall and Djerbib, 2004b; Venkatarama Reddy and Prasanna Kumar, 2010), in the robotic rammed earth process this is reduced signiﬁcantly in order to reduce the required impact energy of the compaction process. This leads to a substantial reduction of the formwork loading (Schweiker et al., 2021; Kloft et al., 2023). The size of the slipform could be reduced to an absolute minimum that only the area of the active compaction is supported. Also, it is possible to achieve a high and constant compaction quality by an automated process (Kloft et al., 2023). 1 Introduction This will provide an initial foundation for further investigations, advancing the development of a robust manufacturing process with controllable and constant compaction quality. FIGURE 1 Robotic Rammed Earth process at the Institute of Structural Design of Technische Universität Braunschweig (Schweiker et al., 2021). FIGURE 1 Robotic Rammed Earth process at the Institute of Structural Design of Technische Universität Braunschweig (Schweiker et al., 2021). regulation of indoor climate due to a high thermal mass, as well as its open pore-structure (Giuffrida et al., 2019). Compared to concrete and masonry construction, rammed earth offers lower mechanical strength between 2–4 MPa. However, the lower strength in combination with very low embodied energy could outweigh materials with high strength and high embodied energy in the future. Possible applications include lower residential buildings with one to three stories (Hall and Djerbib, 2004a). Certainly, the manufacturing process remains predominantly artisanal, leading to construction costs that are disproportionately high for a material that is widely available and relatively inexpensive (Minke, 2021). In addition to the compaction work itself, the labor intensity for formwork related works is excessively high with current building techniques, even though the formwork is needed only in the moment of compaction (Keable, 2005). In order to counter these problems, initial approaches have already been developed for an automation of the production, process, for example, through the use of robots (Gomaa et al., 2023). 3.1.1 Impact energy Impact energy describes the kinetic energy applied to the material in one single stroke, and is of great importance in RRE. Due to the lateral pressure on the formwork during the compaction process, the formwork is loaded. The amount of pressure results from the increment of the individual compaction strokes (Schroeder, 2010; Kloft et al., 2023). Thus, the impact energy becomes an important factor when targeting to minimize the size and weight of the formwork as much as possible. Consequently, the impact energy must also be reduced in order to unload the formwork and raise efﬁciency. Nonetheless, the impact should not be excessively low; it must still possess enough strength to apply adequate compaction energy across the entire layer height. Due to many unknown parameters, it is difﬁcult to quantify a speciﬁc value for the impact energy applied by a hand guided pneumatic rammer. An approach for calculating the resulting impact force based on the piston area is provided by Burroughs (Burroughs, 2010). Making use of this procedure in combination with the stroke height, the impact energy of pneumatic rammers was estimated. The currently used rammer for the RRE process is much smaller than typically used rammers in construction. It should be noted that losses, e.g., due to friction, are not considered. It can be observed that for an increasing water content in the material, the achievable dry density ﬁrst increases to a certain peak and then decreases again. This tipping point is indicated as the Maximum Dry Density (MDD) for a speciﬁc compaction energy, which is reached at the so-called Optimum Moisture Content (OMC) (Reddy, 2022). In practical terms, when approaching the highest point of the compaction curve in relation to the dry density axis, the capillary strength of the clay impedes compaction. Meaning, the amount of compaction energy applied is not sufﬁcient to fully disrupt the structure of the clay (Schroeder, 2010). When further increasing the moisture content, compaction is limited by the presence of pore water or capillary tensions, so the compactor bounces back instead of achieving full compaction (Schroeder, 2010). It is also known that a higher compaction energy results in a higher MDD, which can only be achieved at a lower OMC (El Nabouch, 2017). 2 Principles of earth compaction In order to be able to achieve the highest possible density through the compaction process, it is very important to understand the interaction between various soil parameters and their reciprocal inﬂuence. The key parameters are moisture content, compaction energy and the individual material properties (El Nabouch, 2017; Ávila et al., 2020; Reddy, 2022). In Figure 2A the interrelation between material properties and process variables are illustrated schematically. The Proctor test is the most common method in the ﬁeld of soil mechanics to classify the compaction behavior of an earthen material. Originally, the test was developed to classify the quality Frontiers in Built Environment 02 frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 FIGURE 2 (A) Interconnection diagram (B) Schematic Proctor curve according to (Reddy, 2022). FIGURE 2 (A) Interconnection diagram (B) Schematic Proctor curve according to (Reddy, 2022). the level of compaction are the layer height, the impact energy and the travel speed. These three parameters combined result in the compaction energy, that is applied to the material. In the following paragraphs they will be explained individually. of compaction works in road construction. Since the properties of the treated material and the principle of ramming compaction are comparable, it is also suitable to be used in earthen construction (Schroeder, 2010). The test concept of the Proctor test is already used in the broad technical literature for rammed earth (Volhard and Ulrich, 2002; El Nabouch, 2017). It provides the optimum water content (OMC) for a soil to reach the maximum dry density (MDD). The tests makes use of iterative compaction of soil samples with increasing moisture content in a ﬁxed volume and with a clearly deﬁned number of strokes and layers (Abhilash et al., 2019). In Figure 2B, typical Proctor curves for two different compaction energy levels are shown schematically. 3 Materials and methods Thus, it can be assumed, that the impact energy delivered by the 4.5 kg hammer of the Proctor test is roughly equivalent to that of the pneumatic rammer used in the current compactor of the RRE process. This will be later considered when choosing the exact parameter levels. It is also clear that the impact energy is already signiﬁcantly lower than that generated by conventional compaction Frontiers in Built Environment 3.1.1 Impact energy For the standard Proctor test, the soil is compacted with a total compaction energy of 594 kJ/m³, while in a modiﬁed Proctor test the total compaction energy amounts to 2,653 kJ/m³ (Author Anonymous, 2012). The compaction with the Proctor-hammer, as well as hand guided rammers, can be calculated by the equation of a falling weight. However, it must be considered that depending on the individual strength of the worker the acceleration of the weight can be raised signiﬁcantly. Therefore, it is proposed to include gravity twice. In Table 1 the different impact energies are listed for comparison. frontiersin.org 3.1.2 Layer height The layer height describes the height of each compacted layer. All layers combined make up the printed rammed earth volume. Before each new layer can be compacted, the loose material is applied during the charging process. Here, the layer height can be customized by the amount of material poured out. In the context of the RRE process, the choice of the exact layer height can be inﬂuenced by several parameters. On one hand, a thinner layer reduces the necessary compaction energy per run, thereby minimizing the impact on formwork loading. This, in turn, allows for ramming with a lower impact or, alternatively, permits the application of more compaction energy per volume. On the other hand, thinner layers result in higher total layer numbers for a component. This increases the required production time, not least due to the higher number of feeding runs required. Likewise, the layer height cannot be reduced arbitrarily, since the maximum grain size becomes the limiting factor with coarser material compositions. The results of initial tests indicate that if the ratio of layer height to grain size is too low, the compression strength may be affected (Kloft et al., 2023). CEProctor EI · nL · nS V For the automated process with a pneumatic ramming device, which is guided by a robot along the manufactured building component, a slightly different equation is obtained. Considering the travel speed vT, the frequency f of the rammer, the width of the compacted area bA and the layer height hL the volume-related compaction energy is calculated as follows. CERobotic EI · f vT · bA · hL When it comes to speciﬁc literature, the comparison of the compaction energy on the construction site and in the laboratory is a discussed topic. The deviation is relevant because the quality and thus also the mechanical properties of the material are closely related to the manufacturing method or, respectively, the intensity of it. Also, the OMC for the optimal manufacturing results depends on the amount of compaction energy. Some estimate the compaction energy achieved on- site at around 1,500–2000 kJ/m³ (Abhilash et al., 2019). Elsewhere, the production methods are roughly compared with the proctor energy levels known from soil mechanics. 3.1.4 Compaction energy The value of the compaction energy is a main parameter for the production of rammed earth components and can be used to control the degree of compaction. This describes the amount of effort which is expended to compact the material in order to increase the density. The total applied compaction energy can be inﬂuenced by varying the layer height, impact energy and travel speed, or the number of strokes respectively. It is important to balance the parameters in a way that the compaction success is as high as possible. At the same time, however, the process should be kept as efﬁcient as possible. 3.1 Process parameters Previous to any practical works, robotic parameters were mapped, investigating their reciprocal inﬂuence regarding the compaction process. The input parameters directly inﬂuencing Frontiers in Built Environment 03 frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 1 Comparison of the impact energy of different rammer types. TABLE 1 Comparison of the impact energy of different rammer types. Rammer type Proctor 4.5 kg- hammer Proctor 15 kg- hammer Manual rammer Pneumatic rammer Pneumatic rammer (RRE) Calculation E = m x g x h E = m x g x h E = m x a x h E = p x A x h E = p x A x h Properties m = 4.5 kg h = 0.45 m m = 15 kg h = 0.60 m m = 15 kg h = 0.40 m a = 2 × 9.81 m2/s p = 0.625 bar A = 1,140 mm2 h = 0.15 m p = 0.630 bar A = 415 mm2 h = 0.07 m Impact energy 20 J 88 J 118 J 107 J 18 J equipment. The vibratory plate was left out of this consideration, as it is a different compaction method. equipment. The vibratory plate was left out of this consideration, as it is a different compaction method. The compaction energy CE is described as the total applied energy per volume V (Burroughs, 2010; Reddy, 2022). For the Proctor experiment, the following equation provides the compaction energy where EI is the impact energy, nL is the number of layers, nS is the number of strokes and V is the volume of the compacted sample. The compaction energy CE is described as the total applied energy per volume V (Burroughs, 2010; Reddy, 2022). For the Proctor experiment, the following equation provides the compaction energy where EI is the impact energy, nL is the number of layers, nS is the number of strokes and V is the volume of the compacted sample. 3.1.3 Travel speed The travel speed indicates how fast the robot guides the compaction tool along the component surface. Speed and compaction energy with a deﬁned ramming frequency exhibit an inverse relationship. Slower travel speed results in higher compaction energy introduced into the material per run, whereas faster speed leads to a reduction in introduced compaction energy. Since the experiments were performed in a manual setup, an alternative equivalent parameter had to be found to simulate the compaction by a moving rammer. Therefore, in the further course of the documentation, the reference used will be the number of strokes at one spot. When comparing these two values, a high number of strokes means a slow speed, while a small number of strokes means an accordingly faster speed. The travel speed indicates how fast the robot guides the compaction tool along the component surface. Speed and compaction energy with a deﬁned ramming frequency exhibit an inverse relationship. Slower travel speed results in higher compaction energy introduced into the material per run, whereas faster speed leads to a reduction in introduced compaction energy. Since the experiments were performed in a manual setup, an alternative equivalent parameter had to be found to simulate the compaction by a moving rammer. Therefore, in the further course of the documentation, the reference used will be the number of strokes at one spot. When comparing these two values, a high number of strokes means a slow speed, while a small number of strokes means an accordingly faster speed. 3.2 Material parameters Also essential for the manufacturing process is the composition of the material itself, with the main inﬂuencing factors being the grain size distribution as well as the clay and moisture content. The extent to which the material can be processed to a load-bearing component depends on these parameters. For a robust and adaptable manufacturing process, these factors should not serve as exclusion criteria; however, understanding the effects of individual parameters is crucial for making necessary process adjustments. 3.1.2 Layer height The standard Proctor level of 590 kJ/ m³ would be equivalent to manual processing and the modiﬁed test level of about 2,700 kJ/m³ would correspond to processing with pneumatic rammers (Burroughs, 2010; Ávila et al., 2020). Frontiers in Built Environment 3.2.1 Material composition Earthen materials consist of particles in different sizes which are aggregates (sand and gravel), silt and clay minerals, (Reddy, 2022; Frontiers in Built Environment 04 frontiersin.org 10.3389/fbuil.2024.1363804 Schmitz et al. Morel et al., 2021; Ávila et al., 2020; Schweiker et al., 2021). For the best compaction results, the material should have a well graded distribution, consisting of ﬁne and coarse portions (Houben and Guillaud, 1994; Hall and Djerbib, 2004a; Arrigoni et al., 2017). With a theoretically optimum distribution, the voids between larger particles are ﬁlled with particles of decreasing size. This leads to a maximum density which in turn results in increased strength (Matthew et al., 2012). Since the clay in the material acts as a natural binder, special attention is given to it in the general investigations. The clay particles have a very large speciﬁc surface area, which is why they are the main absorber of the contained water. Therefore, the clay content has a signiﬁcant inﬂuence on the OMC as well as the mechanical properties (Liu and Tong, 2017). the 15 kg-hammer is used. To avoid inﬂuences through the different sized heads of the hammers, an additional top plate is used for compacting the earth. By putting the plate in between during compaction, the impact energy is applied over the entire surface, regardless of which hammer is used. Figure 3B shows the test setup in the laboratory. the 15 kg-hammer is used. To avoid inﬂuences through the different sized heads of the hammers, an additional top plate is used for compacting the earth. By putting the plate in between during compaction, the impact energy is applied over the entire surface, regardless of which hammer is used. Figure 3B shows the test setup in the laboratory. To precisely determine the material moisture content, the material is dried with an oven at about 100°C, then the moisture content is measured and selectively remoistened to the desired level. Ready-mixed material is stored in sealable buckets and covered with foil to keep the moisture content as constant as possible. Immediately before the manufacturing of a sample, the material is poured out on a platform and again mixed thoroughly. The Material is then ﬁlled into the mold and the required number of strokes are applied consequently in a circular motion, using the proctor hammer. Following this procedure, the required number of layers are ﬁlled and compacted. 3.3.3 Dry density and compressive strength After each sample is compacted, the dry density has to be determined which is done by weighing the moist sample mm. The theoretical dry weight md of the sample is calculated by subtracting the water content w. The dry density ρd is then calculated in relation to the volume V (Author Anonymous, 2012). 3.2.1 Material composition After all layers have been compacted, the top piece is removed and the remaining protruding material is cut off with a metal ruler. In this procedure, achieving a ﬂat surface is crucial to ensure uniform volumes among the various samples. However, since only one material mixture is used in the investigations of this article, the clay content and the composition will not be assessed more deeply in the experiments. The material used for the experiments within this research is a common rammed earth mixture available in the industry. The maximum grain size is 22 mm and the raw material has an approximate particle density of 2,200 kg/m³. The clay content is estimated to be about 10%–12% (Okologisch Bauen. Natuürlich mit Lehm, 2024). 3.2.2 Moisture content Given the signiﬁcance of moisture content in inﬂuencing dry density results, special emphasis was placed on it during the experiments. According to the standard test procedure, the samples would be crushed after weighing in order to determine the moisture content directly (Author Anonymous, 2012). However, since the samples were stored for subsequent compressive strength testing, this step was neglected. Instead, the moisture content was monitored during the tests for each sample separately. The dry densities of the samples could be determined and classiﬁed accurately by drying material samples in an oven and determining the moisture content by relating the water content to the dry weight with the following equation. The moisture content is a very decisive factor for rammed earth construction. On the one hand, the compaction process is mainly determined by the moisture content, but on the other hand, also from a practical point of view, the moisture content has an inﬂuence on the workability of the material. The moisture content is deﬁned as the ratio of the water to the mass of the dry material (Schroeder, 2010). For a given compaction energy, the dry density will reach a maximum when the material is at the OMC. In practical cases, however, the perfect moisture content is at the discretion of the worker. For example, the German guidelines “Earth building rules” recommend that the material are processed in an earth-moist state (Volhard and Ulrich, 2002). Since the optimum moisture content is highly dependent on the processing method and the material composition used, it is difﬁcult to establish speciﬁc guideline values. However, even if the moisture content is at the optimum for a relatively low compaction energy, the material could practically still be too wet. w mw md Frontiers in Built Environment frontiersin.org 3.3.1 Setup and test procedure FIGURE 3 (A) Principle of experimental setup (B) Setup at the laboratory. FIGURE 4 Experimental program and parameter variation strategy. FIGURE 4 Experimental program and parameter variation strategy. FIGURE 4 Experimental program and parameter variation strategy. consistent across all phases, the objectives vary, and different parameters are altered. The range and the selection of the parameters used will be explained in the following sections. A graphical overview on the experimental program is provided in Figure 4. This illustrates the parameters considered and marks the extent to which they were varied or ﬁxed in the individual test phases. The parameter of strokes per layer holds particular signiﬁcance, it is varied in all three phases but depends on other variable parameters in order to regulate the overall compaction energy level. 2020), in some cases even up to 2,400 kg/m³ (Röhlen and Ziegert, 2020). Compressive strength values from 1.5 to 4.0 N/mm2 are speciﬁed for these cases (Röhlen and Ziegert, 2020). Others give signiﬁcantly lower values, between 1.5 and 2.5 N/mm2, with exceptions explicitly mentioned (Ávila et al., 2020). The large scatter of compressive strength results can be explained with a lack of standardization. There are many factors that affect the results, such as the size and shape of the samples, the compaction method, and the test procedure itself (Ávila et al., 2020). All parameter variations described in the following sections are listed in the Table A1. The samples are marked uniformly so that they can be clearly assigned later on. The labeling results from the respective phase and test series to which they belong. For example, the third sample of the second test series from Phase 2 would be labeled ‘P2.2.3’ accordingly. 3.3.1 Setup and test procedure ρd md V mm V · 1 + w ( ) The experimental concept of this research involves investigation on the inﬂuence of the abovementioned process relevant parameters on the compaction result. Therefore, the dry density was the chosen evaluation criterion in the experimental concept. For a ﬁnal evaluation, the main criterion of compressive strength would have to be tested. However, initially, the dry density will be used as criterion of the compaction quality. The results will be evaluated based on the compressive tests and ﬁndings from literature, applying the commonly approved indicator, the relationship between dry density and compressive strength (El Nabouch, 2017; Abhilash et al., 2019). It is assumed, that specimens with a higher density also have a higher compressive strength. Generally, the dry density ranges from about 1,700 up to 2,200 kg/m³ (Hall and Djerbib, 2004a; El Nabouch, 2017; Ávila et al., As a basis for the experimental setup, the proctor test setup is used, it consists of a 12.5 cm high, cylindrical steel mold with a diameter of 15 cm and a Proctor-hammer (Figure 3A). The hammers have a weight of 4.5 and 15 kg with a maximum drop height of 45 and 60 cm, respectively. By adding a clamp to the hammer, the drop height can be adjusted manually to achieve different impact energies. The 4.5 kg-hammer is used for most of the experiments. For specimens requiring a higher impact energy, Frontiers in Built Environment 05 frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 2020), in some cases even up to 2,400 kg/m³ (Röhlen and Ziegert, 2020). Compressive strength values from 1.5 to 4.0 N/mm2 are speciﬁed for these cases (Röhlen and Ziegert, 2020). Others give signiﬁcantly lower values, between 1.5 and 2.5 N/mm2, with exceptions explicitly mentioned (Ávila et al , 2020) The large consistent across all phases, the objectives vary, and different parameters are altered. The range and the selection of the parameters used will be explained in the following sections. A graphical overview on the experimental program is provided in Figure 4 This illustrates the parameters considered and marks the FIGURE 3 (A) Principle of experimental setup (B) Setup at the laboratory. FIGURE 4 Experimental program and parameter variation strategy. FIGURE 3 (A) Principle of experimental setup (B) Setup at the laboratory. FIGURE 3 (A) Principle of experimental setup (B) Setup at the laboratory. Frontiers in Built Environment 3.4.4 Phase 3–required compaction energy for a minimum dry density In the last experimental phase, it is planned to evaluate an operational area for the required compaction energy depending on the minimum required dry density and different moisture contents. The target is to provide a graphical solution process based on real test data for adjusting the robot parameters for an efﬁcient manufacturing process. Therefore, the focus is set on the successive increase of strokes per layer, respectively the compaction energy. Starting with only two strokes per layer were increased to a number of 200, which corresponds to a compaction energy of 2,716 kJ/m³ (for a number of three layers). Initially, the steps are positioned closer together to achieve a clearly discretized curve. Subsequently, the steps are enlarged to cover a broader range. In the last experimental phase, it is planned to evaluate an operational area for the required compaction energy depending on the minimum required dry density and different moisture contents. The target is to provide a graphical solution process based on real test data for adjusting the robot parameters for an efﬁcient manufacturing process. Therefore, the focus is set on the successive increase of strokes per layer, respectively the compaction energy. Starting with only two strokes per layer were increased to a number of 200, which corresponds to a compaction energy of 2,716 kJ/m³ (for a number of three layers). Initially, the steps are positioned closer together to achieve a clearly discretized curve. Subsequently, the steps are enlarged to cover a broader range. To obtain a suitable range for a robust manufacturing process, several moisture contents are also taken into account. Thus, three different moisture content levels have been chosen, which can be described as very dry, intermediate and wet. The exact values are set depending on the borders of workability of the material. To avoid an excessive increase of samples the parameters of impact energy and layer height are ﬁxed for the test series. The exact values chosen depend on the results of the second phase of the test series. Both the standard of 594 kJ/m³ and the modiﬁed compaction energy of 2,653 kJ/m³ level were considered. This decision was made to ensure that the trials are as close as possible to the proven standard procedure. Thus, the initial Proctor curves could be used as a reference to classify the results obtained in the following two test phases. 3.4.4 Phase 3–required compaction energy for a minimum dry density In addition, two further energy levels, each a reduced variant with a compaction energy of 297 and 1,619 kJ/m³, were investigated. These intermediate levels were added to have a broader and more diverse test area. The standard and modiﬁed Proctor levels are similar to the standard procedure. Accordingly, the reduced energy levels were realized by reducing the number of strokes per layer. To obtain a suitable range for a robust manufacturing process, several moisture contents are also taken into account. Thus, three different moisture content levels have been chosen, which can be described as very dry, intermediate and wet. The exact values are set depending on the borders of workability of the material. To avoid an excessive increase of samples the parameters of impact energy and layer height are ﬁxed for the test series. The exact values chosen depend on the results of the second phase of the test series. 3.4.2 Phase 1–effect of moisture content and compaction energy of 20 J, on the other hand, would more closely match the edge rammer currently used in the RRE process. of 20 J, on the other hand, would more closely match the edge rammer currently used in the RRE process. The tests will be performed only at one ﬁxed compaction energy level, which is chosen based on the results of the ﬁrst phase. According to that, the moisture content is set to the optimum of the selected compaction energy level. In the ﬁrst phase, the goal was to produce Proctor curves for the existing material according to the standardized procedure. The test series conducted here, serve to determine the OMC and the corresponding MDD for each compaction energy level. Additionally, the behavior of the MDD as compaction energy increases is examined. This will provide an initial insight into the minimum required and maximum suitable compaction energy. Furthermore, the compressive strength of the dried proctor samples was determined. 4.1 Phase 1–effect of moisture content and compaction energy The aim is to test whether the resulting dry density is inﬂuenced if a constant compaction energy is applied with a signiﬁcantly reduced impact energy. In case of a negative inﬂuence this would lead to the assumption, that the impact is too low to reach the bottom of the layer for a successful compaction. The same effect is tested by varying the layer height. In this case another possible effect could occur. With decreasing layer height for a ﬁxed volume height, the number of compaction passes increases. At this point the question arises if the underlying layers will be affected (positively) by compacting overlying layers. 3.4.1 Concept of the investigations The experimental concept comprises three phases. While the setup and procedure of the compaction experiments remain Frontiers in Built Environment frontiersin.org 06 Schmitz et al. 10.3389/fbuil.2024.1363804 Schmitz et al. 3.4.2 Phase 1–effect of moisture content and compaction energy 3.4.3 Phase 2–effect of process parameters at constant compaction energy In the second phase the inﬂuence of different impact energies and layer heights on the compaction results will be investigated. For an unaffected comparison, the test series is performed with a constant compaction energy and moisture content. The parameters of layer height and impact energy are varied in several levels and then combined in every possible constellation. To keep the total compaction energy the same for each sample the third parameter of strokes per layer is varied accordingly. frontiersin.org Frontiers in Built Environment 4.1.1 Dry density FIGURE 6 (A) Proctor curves for different compaction energy levels (B) Compaction energy and the corresponding maximum dry density. FIGURE 5 Effect of the (A) compaction energy and (B) the moisture content on the appearance of the proctor samples. FIGURE 5 Effect of the (A) compaction energy and (B) the moisture content on the appearance of the proctor samples. FIGURE 6 (A) Proctor curves for different compaction energy levels (B) Compaction energy and the corresponding maximum dry density. FIGURE 6 (A) Proctor curves for different compaction energy levels (B) Compaction energy and the corresponding maximum dry density. GURE 6 A) Proctor curves for different compaction energy levels (B) Compaction energy and the corresponding maximum dry density. OMC and MDD, the results for the standard level show a ﬂatter peak. densities are plotted depending on the speciﬁc moisture content of the sample. With an increase in moisture content, the dry density rises to a maximum before subsequently declining. Likewise, it can be recognized that with increasing compaction energy level the maximum of the dry density increases. At the same time, the OMC decreases. This causes the curves representing higher levels of compaction energy to move in positive y- and negative x-direction. The black line represents the zero-air void line (ZAV), which represents a saturation of 100%. The compaction curves will approach this line as the moisture content increases above the optimum (Schroeder, 2010), which can be conﬁrmed. It is also noticeable that the curves of the higher compaction energy have a more pronounced peak compared to the curves representing a lower compaction energy. In conventional manufacturing, this means that the operational area for production is much smaller with high compaction energy. From the maxima of the curves for the reduced standard (297 kJ/m³), standard (594 kJ/m³), reduced modiﬁed standard (1,619 kJ/m³) and modiﬁed (2,653 kJ/m³) compaction energy levels, it is possible to derive the maximum dry density of 2,009, 2,075, 2,155 and 2,187 kg/m³ and the corresponding OMC of 10.9, 10.3, 7.1 and 6.5 wt%, respectively. Where the values for the most test curves show a clear peak for the If the achieved maximum dry densities are plotted against the compaction energy, the result is a gradually ﬂattening ascending curve, see Figure 6B. With increasing compaction energy, the maximum achievable dry density also increases. 4.1.1 Dry density In the ﬁrst experimental phase, Proctor tests were undertaken for the four different compaction energy levels. For each level, ﬁve samples with increasing moisture content have been compacted by a ﬁxed procedure. Figure 5A shows the effect of the compaction energy on the on the appearance of the Proctor samples. The effect of higher compaction due to increased compaction energy is evident in the texture of the samples. The ones subjected to lower compaction energy exhibit more voids, ultimately resulting in lower density. The effect of the moisture content is shown in Figure 5B. Despite identical compaction energies applied to each sample in this series, the impact of increasing moisture content is evident. The surface becomes more plastic of the specimen with the highest moisture. Notably, the sample with a moisture content of 12.9 wt%, the moisture content in the material is so pronounced that the sample deforms under its own weight. This phenomenon is especially distinct in series with lower compaction energy. For the layer height the levels are chosen to be 2.5, 3.1, 4.2 and 6.3 cm. These values are obtained by dividing the total height of the test mold of 12.5 cm by the number of layers, which are 5, 4, 3 and 2 respectively. A layer height of under 2.5 cm is not suitable due to a maximum grain size of 22 mm. The range of layer heights is similar to previous investigations on RRE. The impact energies are varied at levels of 5, 10, 20, 40 and 80 J. The values were chosen this way to have a wide parameter range with a focus on lower values. For the impact energies from 5 to 20 J, the 4.5 kg-hammer is used with a corresponding drop height. Similarly, for the 40 and 80 J levels the 15 kg-hammer is used. In the context of traditional manufacturing, the maximum impact of 80 J would be in line with a hand-held pneumatic rammer. The value In Figure 6A the results of the four Proctor test series are shown. The measured values of the material moisture deviate to different extents from the intended values. This is due to the manual mixing and the inhomogeneous material consistency. The achieved dry Frontiers in Built Environment 07 frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 FIGURE 5 Effect of the (A) compaction energy and (B) the moisture content on the appearance of the proctor samples. 4.1.1 Dry density However, with increasing energy, the resulting gain in dry density decreases, and shows an asymptotic, which means that it approximates a value, which is the random close packing (RCP) of the material particles. Because of the described growing behavior, at a certain point it becomes ineffective to increase compaction energy further. For example, the maximum dry density between the reduced standard and the standard level increases by 3%, from 2,009 to 2,075 kg/m³, which is about 66 kg/m³ higher. This was realized by increasing the compaction energy by 200%, from 297 to 594 kJ/m³. Between the reduced modiﬁed and the modiﬁed level, the increase of dry density is only from 2,155 to 2,187 kg/m³. Though, the compaction energy here was raised by 1,034 kJ/m³. Thus, the additional input was more than three times higher while the achieved increase in dry density was only about 1.5%. In conclusion drawn from this curve, it can be stated that between the reduced modiﬁed and modiﬁed compaction energy levels, the Frontiers in Built Environment 08 frontiersin.org 10.3389/fbuil.2024.1363804 Schmitz et al. particles partially settle to the bottom, while the larger grains tend to stay on top. This could lead to problems in the process ﬂow before and during the feeding process. A moisture content of around 7 wt% is the OMC for the reduced modiﬁed compaction energy level, which is assumed to be an efﬁcient compaction energy. The workability can be classiﬁed as good with reduced tendency to segregation. It was observed that, even with compaction energies exceeding the OMC, i.e., for samples with a moisture content higher than the OMC, the material could still be compacted well. The limitation lies in the inability to achieve higher densities under such conditions. FIGURE 7 Effect of moisture content and compaction energy on compressive strength of rammed earth. A moisture content of 9 wt% is the OMC of the standard compaction level. At this moisture level, the material starts aggregating into smaller lumps, with the advantage that even the largest aggregates in the uncompacted material already have some clay particles adhering to them. As a result, there is virtually no segregation of the material, when mixed carefully. However, at this moisture content level, when higher compaction energies are applied, problems arise because the material is only plasticly deformed by the compactor after the maximum density is achieved. 4.1.2 Compressive strength Figure 7 displays the results of the compressive strength test on samples of the standard Proctor and the reduced modiﬁed Proctor test. These results should be seen as indicative as the test was performed once per sample. Nevertheless, the compressive strength results generally follow the Proctor curve, which proofs that a higher dry density leads to a higher compressive strength. This relationship between dry density and compressive strength is also found in the literature (El Nabouch, 2017; Abhilash et al., 2019). Furthermore, the individual compressive strength values are satisfactory. According to the German guidelines “Earth building rules”, a minimum of 2 N/mm2 is required for rammed earth structures, and all specimens meet this requirement. Considering the values from the literature and this compression test series, dry densities from around 2000 kg/m³ are considered good. 4.2 Phase 2–effect of process parameters at constant compaction energy The second phase of investigations, focusses on the variation of the parameters layer height, impact energy and the number of strokes. Based on the results of the ﬁrst phase, the following experiments were conducted at the reduced modiﬁed energy level of 1,619 kJ/m³ with a mean moisture content of 7.1 wt%. A rather high value was selected to simulate the experiments on a level with a higher compression strength than the absolute minimum. This is because the RRE process should be able to produce a compressive strength as high as possible. However, it was important to select a value which is still in a suitable range determined in the previous phase. 4.1.1 Dry density This implies that additional strokes cause the material to move or shift. As a result, when the moisture content further increases, the material becomes more difﬁcult to process. Above 11 wt%, the material appears almost wet and approaches the saturation limit, making it unsuitable for use. ratio of energy to yield starts to deteriorate signiﬁcantly. Nonetheless, it should be noted that with all compaction energy levels very appropriate densities have been achieved. 4.1.3 Workability Besides material compaction, consistency is crucial for efﬁcient manufacturing. In Figure 8 different degrees of material moisture are shown, ranging from dry to wet. As already found, for very dry mixtures, such as 3 and 5 wt% moisture, higher compaction energies are necessary. Even if processability during the process is good with exceptionally dry material conditions, it is observed that segregation occurs. The smaller In Figure 9, the results are plotted for the calculated dry density and the corresponding moisture content. For comparison, the Proctor curve of the compaction energy level used, is shown as well. It can be FIGURE 8 Different moisture contents of the material from dry to wet. FIGURE 8 Different moisture contents of the material from dry to wet. FIGURE 8 Different moisture contents of the material from dry to wet. Frontiers in Built Environment 09 frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 10.3389/fbuil.2024.1363804 Proctor curve in Figure 9, each series consists of points below and above the reference curve. This conﬁrms the assumption that the inﬂuence of the layer height in the compaction process is negligible. It is important to note, on one hand, that the investigated parameter range is very small. On the other hand, the layer heights in general are very low when compared to traditionally used layer heights. FIGURE 9 Classiﬁcation of the results from P2 to the corresponding Proctor curve. FIGURE 9 Classiﬁcation of the results from P2 to the corresponding Proctor curve. In Figure 10B, the test results are plotted as variation of the dry density at different impact energy levels. Here a signiﬁcant trend can be observed, with an increasing dry density at increasing impact energy up to 40 J. After that, the efﬁciency of compaction seems to drop. The highest dry densities were measured for the samples with an intermediate impact energy of 20 and 40 J. A further increase of the impact to 80 J results again lower dry densities. This can be explained by a very low stroke number as the sample with a layer height of 2.5 cm and an impact of 80 J per stroke has the lowest number of strokes per layer (nine). In conclusion, it seems, that an impact of 10–40 J will produce better results than a very high or low impact of 80 or 5 J, respectively. The highest mean values are observed for 20 and 40 J. 4.1.3 Workability However, it is important to acknowledge that the scatter for the 40 J series is considerably high in comparison to the others. This elevated scatter is primarily attributed to the outlier just described. Despite this, the overall variation of values in this conﬁguration is minimal. This suggests that the impact energy has a signiﬁcant effect. FIGURE 9 Classiﬁcation of the results from P2 to the corresponding Proctor curve. seen that the dry density scatters around the curve of the standard Proctor test. By averaging the values of the dry density and the moisture content, the center of the scatter plot aligns closely with the Proctor curve. At ﬁrst glance, this could lead to the assumption that there is no major inﬂuence of the investigated parameters. However, the scatter of values may still suggest some effect. In consideration of Figure 9, it is evident, that the achieved dry densities for the highest impact are all below the reference curve, conﬁrming the previous observations that the compaction with the highest hammer impact was the least efﬁcient. Especially the measurement points for the 5J-impact are particularly close to the Proctor curve. With one further exception, the remaining points are located above the reference curve. In conclusion, it seems that the variation of the dry density due to different impact energyies on a ﬁxed compaction energy level is not very high. Nevertheless, a slight trend can be determined. For a more detailed evaluation of the effect of these parameters, Figure 10A shows the effect of the layer height on the dry density individually. In order to identify the inﬂuence more clearly, the mean values were calculated. This shows that there is only very little effect. Comparing the results with regard to the standardized FIGURE 10 Distribution and mean values of dry density for (A) layer height and (B) impact energy levels. FIGURE 10 Distribution and mean values of dry density for (A) layer height and (B) impact energy levels. 10 Frontiers in Built Environment frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 10.3389/fbuil.2024.1363804 layer, which is caused by the variation of the layer height, the results are increasing steadily. This leads to the assumption that for a successful compaction process there is a minimum number of impacts required. In addition to that and in the context of the RRE process, the layer height could still be used to optimize the stroke rate in the compaction process. 4.3 P3–relation of compaction energy and moisture content The third experimental phase was undertaken in order to implement the parameters into the RRE approach. The layer height and the impact energy were ﬁxed. As the compaction is digitally controlled and monitored, it can be assumed, that the process runs with a constant impact energy per stroke. The best results in the previous phase 2 have been achieved by impacts of 20 and 40 J. For the RRE process however, the impact should be as low as possible due to the lateral loading of the formwork and the objective to reduce the end effectors weight. Therefore, reducing the impact energy to 10 J was considered a reasonable compromise. The layer height was set to an intermediate level of 4.2 cm. As described in the previous section, the impact energy slightly inﬂuences the achievable dry density. In contrast to that the layer height tends to have no effect. But after that, the question remains as to what is the reason for it. A plausible explanation for the lower results of the 5- and 80J-impacts could be attributed to the corresponding stroke numbers. Figure 11, illustrates the various impacts based on the applied strokes per layer. The best results were achieved with about 35–75 strokes per layer. The stroke numbers for the 80J-impact samples of course were very low due to the ﬁxed total compaction energy. Despite the fact that it is certainly possible to compact the entire layer height with high impact energy, the results are comparatively low. Also, the results for the 40J-impact are starting with lower dry densities. With the increase of strokes per 4.1.3 Workability FIGURE 11 Stroke numbers per layer and impact energy. 5 Discussion During the investigation, isolated effects could be detected. However, these are within a limited scope. With a larger parameter range of the layer height and the impact energy, the effects might have been more evident. Nevertheless, the range investigated in the present work was speciﬁcally chosen to ﬁt the context of the RRE process. This balance is crucial for optimizing parameters, encompassing both reduced impact for formwork unloading and increased speed to ensure a time-efﬁcient manufacturing process. For the application of the results, the transferability into practice as well as the scalability are important factors. The small-scale experiments were conducted under controlled conditions. This especially applies to the inﬂuence of the cylindrical formwork: under realistic conditions, a portion of the energy would be lost due to the deformation of the formwork, as well as the partial 4.3.1 Experimental study for increasing compaction energy Figure 12A, shows the effect of the compaction energy on the dry density for three moisture contents; (1) dry (3.7 wt%) (2) intermediate (7.6 wt%) and (3) wet (8.9 wt%). All curves show an asymptotic growth, which means that the gradient steadily increases and approaches a plateau. Compared to the Proctor curves from phase 1 the results are all within the expected range. It is noticeable, that the higher the material moisture content, the stronger the curves rise in the beginning. However, the growth also decreases FIGURE 12 (A) Compaction curves for three different moisture contents and selection of samples of (B) dry (MC 3.7%) (C) intermediate (MC 7.6%) and (D) wet (MC 8.9%) series. FIGURE 12 (A) Compaction curves for three different moisture contents and selection of samples of (B) dry (MC 3.7%) (C) intermediate (MC 7.6%) and (D) wet (MC 8.9%) series. FIGURE 12 (A) Compaction curves for three different moisture contents and selection of samples of (B) dry (MC 3.7%) (C) intermediate (MC 7.6%) and (D) wet (MC 8 9%) series Frontiers in Built Environment 11 frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 FIGURE 13 Graphical diagram for operational area for the RRE process. he RRE process FIGURE 13 Graphical diagram for operational area for the RRE process. compacted with a moisture content which is signiﬁcantly higher than the optimum (Burroughs, 2010). faster and the maximum value is reached earlier. Especially for the curve representing the wet material, the plateau is reached already at a compaction energy of about 600 kJ/m³, while the dryer mixes are still increasing at higher compaction energies. As illustrated on the left side of Figure 13, the compaction energy, depending on the moisture content and the required dry density, can be taken from the diagram. Starting on the horizontal level of the required dry density, the intersection with the actual moisture content can be determined. For example, the dry density derives from the structural requirements and the moisture content could be measured on site or derived from a live monitoring during the manufacturing process. Subsequently, the according compaction energy is marked by the vertical arrow. The layer height is selected based on the material properties and can generally be used to react on practical issues of the process. On the basis of these values, the travel speed can then be set. In Figures 12A–D selection of samples from each test series is shown. 4.3.1 Experimental study for increasing compaction energy In the two drier test series, the ﬁrst two samples (7.6 and 3.7 wt%) were not compacted enough to stay in shape. In the series with the wet material, on the other hand, two strokes per layer (6 in total) were already sufﬁcient for the cylinder to hold its shape. Nevertheless, the achieved density of 1,520 kg/m³ is still too low. total) were already sufﬁcient for the cylinder to hold its shape. Nevertheless, the achieved density of 1,520 kg/m³ is still too low. Regarding the third sample in each series, it can be observed that with intermediate and wet material, densities exceeding 2000 kg/m³ can be achieved with an energy input of 896 kJ/m³. However, with very dry material, this limit is not reached, even with almost twice the energy input of 1,630 kJ/m³. Here, again the strong correlation of the material moisture and the compaction process becomes evident. The density of the higher compacted samples of the wet series (purple line) are very high. It can also be observed that the density remains almost unchanged despite a doubling of the compaction energy energy. This illustrates well, how in this case the air was almost completely removed from the pores and further compaction is prevented by the water contained. Regarding the third sample in each series, it can be observed that with intermediate and wet material, densities exceeding 2000 kg/m³ can be achieved with an energy input of 896 kJ/m³. However, with very dry material, this limit is not reached, even with almost twice the energy input of 1,630 kJ/m³. Here, again the strong correlation of the material moisture and the compaction process becomes evident. If data of this kind can be generated quickly and easily for local materials, this could create data bases that can be used as input parameters for digital manufacturing processes, which could enable the enhancement of the process’s adaptability and uniformity. The density of the higher compacted samples of the wet series (purple line) are very high. It can also be observed that the density remains almost unchanged despite a doubling of the compaction energy energy. This illustrates well, how in this case the air was almost completely removed from the pores and further compaction is prevented by the water contained. Arrigoni, A., Pelosato, R., Dotelli, G., Beckett, C. T., and Ciancio, D. (2017). Weathering’s beneﬁcial effect on waste-stabilised rammed earth: a chemical and microstructural investigation. Constr. Build. Mater. 140, 157–166. doi:10.1016/j. conbuildmat.2017.02.009 Data availability statement yielding of the material. Similarly, the cylindrical shape is more convenient in contrast to the usual rectangular shape of wall elements. Irregularities due to compaction in the corners are thus avoided and not taken into account. Nevertheless, these restrictions are useful in the scope of the experimental tests in order to be able to exclude as many unknown factors as possible. However, further tests under real conditions must be carried out in the future to conﬁrm the results. Besides the mold also the manual Proctor hammers differ from the ramming devices on site. What should be mentioned ﬁrst is the relation between the surface of the compacted volume to the surface of the ramming device or hammer respectively. In most of the experiments a covering top plate was used, which means that the compaction is applied over the entire horizontal surface. Although, previous tests showed that this does not signiﬁcantly affect the result, the difference to realistic manufacturing conditions, where the rammer surface is much smaller than the surface of the compacted layer, which may allow the material to deﬂect. Pneumatic rammers have a much higher stroke frequency. Whether this inﬂuences the compaction process is not known and has been neglected here. The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author. Author contributions LS: Conceptualization, Investigation, Methodology, Visualization, Writing–original draft. JG: Conceptualization, Supervision, Writing–review and editing, Funding acquisition, Methodology, Project administration, Validation. ED: Supervision, Validation, Writing–review and editing. DL: Resources, Supervision, Writing–review and editing. HK: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing–review and editing. Acknowledgments The authors would like to thank the institute of geomechanics and geotechnics for the provision of test equipment, as well as the company Conluto for providing earth materials. 4.3.2 Deriving an operational range for the robotic rammed earth process In order to select the optimum compaction parameters for the robotic manufacturing process, the data of the previous section is analyzed. In Figure 13, the exemplary process is illustrated with the colored area indicating the operational range. It has to be mentioned that the interpolation of these curves was done graphically. For this purpose, the basic properties of the curves were considered, meaning the higher the moisture content, the ﬂatter is the gradient in the beginning and the later will the maximum dry density be reached. Furthermore, the maximum dry density increases with lower moisture content. However, it is important to keep in mind that the compressive strength might be affected if the material is Frontiers in Built Environment 12 frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 Publisher’s note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their afﬁliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. 6 Conclusion The author(s) declare ﬁnancial support was received for the research, authorship, and/or publication of this article. The present research activities are part of the project “GOLEHM2–Mobiler, robotischer Stampﬂehm”, by the initiative “GOLEHM”, funded by the WIR!-program for regional structural change, which is issued by the German Federal Ministry of Education and Research. In this paper, a comprehensive study was conducted to investigate the interrelation of process and material parameters and their impact on the compaction and strength of rammed earth with the aim of establishing an automated robotic rammed earth process. It was found, that with the same volume related compaction energy, the reduction of the impact energy (respectively increased stroke number) down to a value of 10 J is possible without major energy losses. Impact energies higher than 40 J result in slightly reduced densities due to the low number of compaction strokes. From this, it could be derived that a minimum number of strokes of 35–75 is required for sufﬁcient compaction. In contrast, the layer height has a negligible inﬂuence on the compaction process within the investigated parameter scope. This would allow to use this parameter for the process adjustment according to different material conditions. Furthermore, a range for the minimum required compaction energy, depending on the existing material moisture, could be identiﬁed. The studies showed that in a range of about 4–9 wt% material moisture, the applied compaction energy can be reasonably selected between approximately 150 to 2,000 kJ/m³, according to the required dry density. The ﬁndings of this work can be used as input data for the further development of the compaction unit of a Robotic Rammed Earth process. In future research on this topic, the ﬁndings on compaction behaviour could be implemented in the digitized process. By using live or rapid on- site measurements of moisture content and density, the compaction energy could be adjusted during production. This would improve the process robustness with regards to a uniform level of compaction regardless of local conditions and varying material properties. Conﬂict of interest The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Abhilash, H. N., and Morel, J.-C. (2019). “Stress–strain characteristics of unstabilised rammed earth,” in Earthen dwellings and structures. Editors B. Reddy, M. Mani, and P. Walker (Singapore: Springer Transactions in Civil and Environmental Engineering. Springer). doi:10.1007/978-981-13-5883-8_18 References Abhilash, H. N., and Morel, J.-C. (2019). “Stress–strain characteristics of unstabilised rammed earth,” in Earthen dwellings and structures. Editors B. Reddy, M. Mani, and P. Walker (Singapore: Springer Transactions in Civil and Environmental Engineering. Springer). doi:10.1007/978-981-13-5883-8_18 Frontiers in Built Environment 13 frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 Schmitz et al. Attom, M. F. (1997). The effect of compactive energy level on some soil properties. Appl. Clay Sci. 12, 61–72. doi:10.1016/S0169-1317(96)00037-3 Liu, Q., and Tong, L. (2017). Engineering properties of unstabilized rammed earth with different clay contents. J. Wuhan. Univ. Technol.-Mat Sci. Ed. 32, 914–920. doi:10. 1007/s11595-017-1690-y Author Anonymous (2012). Din 18127: baugrund, untersuchung von Bodenproben - proctorversuch - [2012-09-00]. German: DIN-Regelwerk Deutsche Normen. Matthew, R. H., Rick, L., and Meror, K. (2012). Modern earth buildings: materials, engineering, construction and applications, 33. Cambridge: Woodhead Publishing Series in Energy. Ávila, F., Puertas, E., and Gallego, R. (2020). Characterization of the mechanical and physical properties of unstabilized rammed earth: a review. Constr. Build. Mater. 270, 121435. doi:10.1016/j.conbuildmat.2020.121435 Minke, G. (2021). Building with earth: Design and technology of a sustainable architecture. Fourth and revised edition. Basel: Birkhäuser Verlag GmbH. Burroughs, S. (2010). Recommendations for the selection, stabilization, and compaction of soil for rammed earth wall construction. J. Green Build. 5, 101–114. doi:10.3992/jgb.5.1.101 Morel, J.-C., Charef, R., Hamard, E., Fabbri, A., Beckett, C., and Bui, Q. B. (2021). Earth as construction material in the circular economy context: practitioner perspectives on barriers to overcome. Philos. Trans. R. Soc. Lond B Biol. Sci. 376, 20200182. doi:10.1098/rstb.2020.0182 El Nabouch, R. (2017). Mechanical behavior of rammed earth walls under Pushover tests. HAL science, Lyon: Civil Engineering. Université Grenoble Alpes. Niroumand, H., Akbari, R., Khanlari, K., Gültekin, A. B., and Barcelo, J. A. (2021). A systematic literature review of rammed earth walls. Soil Mech. Found. Eng. 58, 295–301. doi:10.1007/s11204-021-09742-y Giuffrida, G., Caponetto, R., and Cuomo, M. (2019). An overview on contemporary rammed earth buildings: technological advances in production, construction and material characterization. IOP Conf. Ser. Earth Environ. Sci. 296, 012018. doi:10. 1088/1755-1315/296/1/012018 Okologisch Bauen. Naturlich mit Lehm (2024). Conluto - vielfalt aus Lehm. Available at: https://www.conluto.de/. Gomaa, M., Schade, S., Bao, D. W., and Xie, Y. M. (2023). Automation in rammed earth construction for industry 4.0: precedent work, current progress and future prospect. J. Clean. Prod. 398, 136569. doi:10.1016/j.jclepro.2023.136569 Reddy, B. V. V. (2022). Compressed earth block & rammed earth structures. Singapore: Springer Nature Singapore. Röhlen, U., and Ziegert, C. (2020). References Lehmbau-Praxis: planung und Ausführung, 3., aktualisierte und erweiterte Auﬂage. Beuth Praxis. Berlin, Wien, Zürich: Beuth Verlag GmbH. Hall, M., and Djerbib, Y. (2004a). Rammed earth sample production: context, recommendations and consistency. Constr. Build. Mater. 18, 281–286. doi:10.1016/j. conbuildmat.2003.11.001 Schroeder, H. (2010). Lehmbau: mit Lehm ökologisch planen und bauen; mit 55 Tabellen, 1. Wiesbaden: Auﬂ. Vieweg+Teubner. Hall, M., and Djerbib, Y. (2004b). Moisture ingress in rammed earth: Part 1—the effect of soil particle-size distribution on the rate of capillary suction. Constr. Build. Mater. 18, 269–280. doi:10.1016/j.conbuildmat.2003.11.002 Schweiker, M., Endres, E., Gosslar, J., Hack, N., Hildebrand, L., Creutz, M., et al. (2021). Ten questions concerning the potential of digital production and new technologies for contemporary earthen constructions. Build. Environ. 206, 108240. doi:10.1016/j.buildenv.2021.108240 Houben, H., and Guillaud, H. (1994). Earth construction: a comprehensive guide. London: CRAterre., & Intermediate Technology Publications. Jaquin, P. A., Augarde, C. E., Gallipoli, D., and Toll, D. G. (2009). The strength of unstabilised rammed earth materials. Géotechnique 59, 487–490. doi:10.1680/geot.2007. 00129 United nations environment program (2009) Building and climate change. Available at: https://www.uncclearn.org/wp-content/uploads/library/unep207.pdf. Accessed 20 June 2023 Venkatarama Reddy, B. V., and Prasanna Kumar, P. (2010). Embodied energy in cement stabilised rammed earth walls. Energy Build. 42, 380–385. doi:10.1016/j.enbuild. 2009.10.005 Keable, J. (2005). Rammed earth structures: a code of practice. London: Intermediate Technology. Volhard, F., and Ulrich, R. (2002). Lehmbau regeln: begriffe baustoffe bauteile, 2., korrigierte auﬂage. Wiesbaden: Vieweg+Teubner Verlag. Khadka, B., and Shakya, M. (2015). Comparative compressive strength of stabilized and un-stabilized rammed earth. Mater Struct. 49, 3945–3955. doi:10.1617/s11527-015- 0765-5 Windstorm, B., and Schmidt, A. (2013). A report of contemporary rammed earth construction and research in north America. Sustainability 5, 400–416. doi:10. 3390/su5020400 Kloft, H., Gosslar, J., and Fernández-Barba, D. (2023). Robotische Fabrikation von Bauteilen aus Stampﬂehm. Bonn. Frontiers in Built Environment 14 frontiersin.org 14 Schmitz et al. 10.3389/fbuil.2024.1363804 Appendix TABLE A1 Supplementary Material: Experimental data of Phases 1,2 and 3. Nr Compaction energy Layer height Impact force Strokes per layer Hammer Drop height Moisture content Dry unit weight Conpression strength - kJ/m³ cm J - kg cm % kg/m³ N/mm2 PHASE 1 P1.1.1 297 4.2 20 11 4.5 45 5.6 1769 P1.1.2 297 4.2 20 11 4.5 45 8.1 1924 P1.1.3 297 4.2 20 11 4.5 45 9.5 1999 P1.1.4 297 4.2 20 11 4.5 45 10.9 2009 P1.1.5 297 4.2 20 11 4.5 45 13.9 1956 P1.2.1 594 4.2 20 22 4.5 45 4.2 1951 P1.2.2 594 4.2 20 22 4.5 45 7.1 2017 2.37 P1.2.3 594 4.2 20 22 4.5 45 7.9 2064 3.78 P1.2.4 594 4.2 20 22 4.5 45 10.3 2075 2.77 P1.2.5 594 4.2 20 22 4.5 45 12.9 1993 P1.3.1 1,619 2.5 20 36 4.5 45 4.0 1996 P1.3.2 1,619 2.5 20 36 4.5 45 5.4 2038 2.22 P1.3.3 1,619 2.5 20 36 4.5 45 7.1 2,155 3.88 P1.3.4 1,619 2.5 20 36 4.5 45 9.7 2076 2.77 P1.3.5 1,619 2.5 20 36 4.5 45 12.2 1995 P1.4.1 2,653 2.5 20 59 4.5 45 2.9 2026 P1.4.2 2,653 2.5 20 59 4.5 45 5.1 2,123 P1.4.3 2,653 2.5 20 59 4.5 45 6.5 2,187 P1.4.4 2,653 2.5 20 59 4.5 45 8.4 2,128 P1.4.5 2,653 2.5 20 59 4.5 45 9.4 2070 PHASE 2 P2.1.1 1,619 6.3 5 358 4.5 11.3 6.2 2,110 P2.1.2 1,619 6.3 10 179 4.5 22.7 6.2 2,118 P2.1.3 1,619 6.3 20 89 4.5 45.3 7.4 2,133 P2.1.4 1,619 6.3 40 45 15 27.2 6.5 2,142 P2.1.5 1,619 6.3 80 22 15 54.4 7.4 2,117 P2.2.1 1,619 4.2 5 238 4.5 11.3 5.9 2083 P2.2.2 1,619 4.2 10 119 4.5 22.7 5.9 2,131 P2.2.3 1,619 4.2 20 60 4.5 45.3 6.7 2,158 P2.2.4 1,619 4.2 40 30 15 27.2 6.5 2,128 P2.2.5 1,619 4.2 80 15 15 54.4 7.4 2,129 P2.3.1 1,619 3.1 5 179 4.5 11.3 5.9 2,113 P2.3.2 1,619 3.1 10 89 4.5 22.7 6.7 2,125 P2.3.3 1,619 3.1 20 45 4.5 45.3 6.5 2,131 (Continued on following page) TABLE A1 Supplementary Material: Experimental data of Phases 1,2 and 3. 15 Frontiers in Built Environment 15 frontiersin.org Schmitz et al. 10.3389/fbuil.2024.1363804 TABLE A1 (Continued) Supplementary Material: Experimental data of Phases 1,2 and 3. Appendix Nr Compaction energy Layer height Impact force Strokes per layer Hammer Drop height Moisture content Dry unit weight Conpression strength - kJ/m³ cm J - kg cm % kg/m³ N/mm2 P2.3.4 1,619 3.1 40 22 15 27.2 6.2 2,176 P2.3.5 1,619 3.1 80 11 15 54.4 7.4 2,102 P2.4.1 1,619 2.5 5 143 4.5 11.3 6.5 2,131 P2.4.2 1,619 2.5 10 72 4.5 22.7 6.2 2,150 P2.4.3 1,619 2.5 20 36 4.5 45.3 6.2 2,158 P2.4.4 1,619 2.5 40 18 15 27.2 6.2 2,113 P2.4.5 1,619 2.5 80 9 15 54.4 6.7 2,101 PHASE 3 P3.1.0 27 4.2 10 2 4.5 22.7 7.6 1,491 P3.1.1 81 4.2 10 6 4.5 22.7 7.6 1,651 P3.1.2 149 4.2 10 11 4.5 22.7 7.6 1742 P3.1.3 299 4.2 10 22 4.5 22.7 7.6 1863 P3.1.4 598 4.2 10 44 4.5 22.7 7.6 1977 P3.1.5 896 4.2 10 66 4.5 22.7 7.6 2055 P3.1.6 1,222 4.2 10 90 4.5 22.7 7.6 2099 P3.1.7 1,630 4.2 10 120 4.5 22.7 7.6 2,125 P3.1.8 2,716 4.2 10 200 4.5 22.7 6.5 2,170 P3.3.1 27 4.2 10 2 4.5 22.7 8.9 1,520 P3.3.2 81 4.2 10 6 4.5 22.7 8.9 1726 P3.3.3 149 4.2 10 11 4.5 22.7 8.9 1861 P3.3.4 299 4.2 10 22 4.5 22.7 8.9 1997 P3.3.5 598 4.2 10 44 4.5 22.7 8.9 2084 P3.3.6 896 4.2 10 66 4.5 22.7 8.9 2098 P3.3.7 1,222 4.2 10 90 4.5 22.7 8.9 2096 P3.3.8 1,630 4.2 10 120 4.5 22.7 8.9 2,100 P3.2.1 27 4.2 10 2 4.5 22.7 3.7 1,497 P3.2.2 81 4.2 10 6 4.5 22.7 3.7 1,609 P3.2.3 149 4.2 10 11 4.5 22.7 3.7 1,685 P3.2.4 299 4.2 10 22 4.5 22.7 3.7 1748 P3.2.5 598 4.2 10 44 4.5 22.7 3.7 1837 P3.2.6 896 4.2 10 66 4.5 22.7 3.7 1895 P3.2.7 1,222 4.2 10 90 4.5 22.7 3.7 1926 P3.2.8 1,630 4.2 10 120 4.5 22.7 3.7 1976 P3.2.9 2,716 4.2 10 200 4.5 22.7 3.7 2019 P3.2.10 3,803 4.2 10 280 4.5 22.7 3.7 1994 TABLE A1 (Continued) Supplementary Material: Experimental data of Phases 1,2 and 3. Frontiers in Built Environment 16 frontiersin.org
https://openalex.org/W2773686719	http://www.scielo.br/pdf/ln/n102/1807-0175-ln-102-267.pdf	Portuguese	null	ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA	Lua nova	2,017	cc-by	11,647	ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA 267 Daniela Xavier Haj Mussi é pós-doutoranda em Ciência Política na Universidade de São Paulo, São Paulo, SP, Brasil. E-mail: <danixhm@gmail.com> Daniela Xavier Haj Mussi é pós-doutoranda em Ciência Política na Universidade de São Paulo, São Paulo, SP, Brasil. E-mail: <danixhm@gmail.com> Alvaro Bianchi é livre-docente do Departamento de Ciência Política da Unicamp, Campinas, SP, Brasil. E-mail: <albianchi@terra.com.br> http://dx.doi.org/10.1590/0102-267298/102 é livre-docente do Departamento de Ciência Política da Unicamp, Campinas, SP, Brasil. E-mail: <albianchi@terra.com.br> http://dx.doi.org/10.1590/0102-267298/102 é livre-docente do Departamento de Ciência Política da Unicamp, Campinas, SP, Brasil. E-mail: <albianchi@terra.com.br> é livre-docente do Departamento de Ciência Política da Unicamp, Campinas, SP, Brasil. E-mail: <albianchi@terra.com.br> http://dx.doi.org/10.1590/0102-267298/102 A Revolução Russa de 1917 é frequentemente lembrada como um evento político stricto sensu, como investida mili tar e instauração da ditadura do proletariado pelos bolche viques. Este artigo reconstrói uma trajetória interpretativa que se choca com esse senso comum, a do escritor italiano Antonio Gramsci no período imediatamente anterior e durante o ano revolucionário russo. Desde a chegada das primeiras notícias da revolução ocorrida na Rússia, é pos sível encontrar nos artigos de Gramsci a ideia da revolução como “ato de cultura”. O jovem jornalista de Turim buscava interpretar os acontecimentos russos sob o ponto de vista da capacidade de expansão de suas ideias-força, como um programa político que era, ao mesmo tempo, uma ruptura com os fundamentos culturais do mundo burguês. As principais balizas dessa interpretação estão postas pelos escritos reunidos no jornal La Città Futura, publi cado poucos dias antes de eclodir a Revolução de fevereiro de 1917 na Rússia, e pelo artigo “La Rivoluzione contro ‘Il Capitale’” na edição romana do Avanti!, em 24 de novem bro de 1917. Nesse intervalo percebe-se a permanência de Lua Nova, São Paulo, 102: 267-298, 2017 Lua Nova, São Paulo, 102: 267-298, 2017 1 O nome “Sotto la Mole” fazia referência ao imponente edifício projetado no século XIX por Alessandro Antonelli e construído entre 1863 e 1897 no centro de Turim. Gramsci também gostava de observar que a expressão sotto la mole (que além de “sob a Mole Antonelliana” poderia ser traduzida como “sob o fardo”) poderia ser entendida também em um sentido que prescindia do monumento, “como metáfora da condição dos trabalhadores, submetidos ao ônus da opressão capitalista” (Rapone, 2011, p. 62n). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA alguns temas, como a afirmação do caráter socialista e antija cobino da Revolução e de sua dimensão cultural, bem como uma consideração dos eventos na Rússia a partir de uma visão particularmente italiana, mais interessada em fazer deles um exemplo a ser seguido na península do que em interpretá-los. Nesse momento, o léxico gramsciano é nitidamente neoidealista. Logo após sua eclosão, a Revolução Russa é vista por ele como um “ato proletário”, um acontecimento que tem lugar “nas consciências de todos”, capaz de pro mover a “libertação dos espíritos” e a “instauração de uma nova consciência moral”. Também é percebida como um evento de alcance internacional, uma “luz que vem do oriente e irradia o velho mundo ocidental” (Gramsci, 1982, pp. 138-142). Um evento capaz de mudar a correlação de forças internacionais, e sua potência se “reflete sobre os outros proletariados, é a substância de uma vida nova, de uma nova autoridade” (Gramsci, 1982, p. 242). 268 Ao longo de 1917, embora o léxico permaneça substan cialmente o mesmo, uma reconstrução do ritmo do pensa mento gramsciano mostra como a influência neoidealista foi gradativamente absorvida em uma visão realista da polí tica influenciada, se não pelo pensamento, ao menos pela ação dos bolcheviques. O ponto de culminância é o artigo de novembro daquele ano, no qual Gramsci esforça-se para compreender o uso da força militar na Revolução como continuidade da ideia de ruptura cultural em sentido amplo que elaborara anteriormente. Ganhava destaque, assim, uma consistente crítica ao reformismo político da maioria da direção do Partito Socialista Italiano (PSI) e a convicção de que era chegada, também na Itália, a hora de agir. Em fins deste agitado período, neoidealismo e realismo aparecem unidos na análise da política por Gramsci. O resul tado é não apenas uma interpretação original da Revolução Russa, a qual guarda valor ainda hoje, mas a emergência de Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi uma nova concepção da política que iria resultar naquilo que Gramsci chamaria de “comunismo crítico” (Gramsci, 1984, p. 349). 2 O jornal socialista Avanti! era editado em Milão e continha uma página dedi cada a Turim, “Cronache Torinesi” [Crônicas de Turim], também conhecida por “edição do Piemonte”, cuja equipe de redação era dirigida desde 1902 por Ottavio Pastore (1888-1965). 3 Gramsci conhecia, desde 1914, o artigo “Les Idoles”, de Rolland, quando foi par cialmente traduzido e publicado no jornal Il Grido del Popolo. Este e outros artigos de Rolland seriam publicados alguns anos mais tarde na revista L’Ordine Nuovo. A guerra e os “ídolos da cultura” No final de 1915, Antonio Gramsci passou a atuar como jornalista na seção turinense do PSI, sua primeira expe riência orgânica como ativista político-partidário. Gramsci começou sua atividade jornalística como colaborador na coluna “Sotto la Mole”1 do jornal Avanti! 2, sendo esta sua primeira atividade sistemática no interior do movimento socialista. Suas crônicas, quase sempre anônimas ou assi nadas com pseudônimos, tratavam de diferentes aspectos da realidade de Turim, escritas como “crônicas da vida coti diana ou cultural da cidade, mais do que sobre os episódios da vida política e administrativa” (Rapone, 2011, p. 71). Até meados de 1917, porém, esta atividade se daria em “uma posição afastada na vida cotidiana do partido, tanto que nos relatos das assembleias das seções nunca está seu nome” (Rapone, 2011, p. 71). 269 Os artigos desta época evidenciam um Gramsci “his toriador do cotidiano”, consciente do contraste de seu texto com a “história dos séculos passados”, com a história “grande mestra da vida” (Gramsci, 1980, p. 45). Os limites da atividade de cronista se evidenciavam: esta era uma inter venção de tipo político-cultural imediato e não equivalente Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA à atividade disciplinada dos estudos e à pesquisa histórica de fôlego. Entre estes dois estilos, porém, Gramsci abraçou a carreira de jornalista como quem se libertava de uma prisão. Em carta escrita em 1916, para a irmã Grazietta, Gramsci falou com pesar do período anterior à atividade como cro nista, momento de isolamento provocado pelos problemas de saúde e financeiros e dedicado exclusivamente aos es tudos: “vivi por alguns anos fora do mundo, um pouco no sonho. Deixei que se perdessem, um por um, todos os laços que me uniam ao mundo e aos homens. Vivi totalmente para o cérebro e nada para o coração” (Gramsci, 2009, p. 166). As crônicas representavam, de certa forma, um processo de reconexão com o “mundo e os homens”, eram a atividade intelectual vivida fora do “sonho”, entrelaçada de maneira apaixonada à vida social (Gramsci, 2009, p. 167). p p Em sua atividade jornalística, Gramsci expressava sen sibilidade às transformações provocadas pela I Guerra Mundial na vida italiana e, particularmente, em Turim. Inspirava-se na reflexão do intelectual francês Romain Rolland sobre o papel político do “mundo cultural”, drasti camente transformado pelo conflito militar nesse período. Rolland criticava “les idoles de La Kultur”, intelectuais que “vivem no reino das ideias”, desconectados dos processos sociais e que, por este motivo, na guerra “não apenas foram os mais expostos ao contágio bélico, como contri buíram prodigiosamente para difundi-lo” (Rolland, 1953, p. 118; p. 125)3. Do isolamento, esses intelectuais passavam ao oportunismo dos que “aderem à concepção que serve melhor” e ao apoio – tímido ou mesmo aberto – ao con flito militar. Inspirado nessa crítica, Gramsci compôs muitas de suas crônicas nos anos 1915 e 1916 para revelar o opor tunismo dos representantes da vida intelectual de Turim, 270 Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi onde predominavam “os vulgares fraudadores da inteli gência” (Gramsci, 1980, p. 34). As crônicas eram, também, uma maneira de construir sua própria subjetividade polí tica. Gramsci retratava um universo cultural decadente em que figuras grotescas se enfileiravam: economistas, filósofos, literatos, clérigos, poetas. Seus artigos se dirigiam especial mente contra os representantes do positivismo na vida inte lectual do Piemonte e, como alternativa, projetava a neces sidade de fundir, em um único campo teórico, o socialismo e o pensamento neoidealista (Rapone, 2011). 4 Fundado em 1892, em Turim, o jornal semanário Il Grido del Popolo expressava as concepções do socialismo que se desenvolvia no Piemonte, de traços marcada mente intelectuais e racionalistas. Além de colaborador, Gramsci assumiu o posto de redator-chefe, entre agosto de 1917 e setembro de 1918, o último antes do jor nal ser fechado. Essa foi a primeira experiência do jovem socialista sardo à frente de um periódico (Paggi, 1970). 5 O PSI era um dos poucos partidos a sustentar a posição firme contra a guerra e o nacionalismo, posição que seria abalada em 1917, no contexto da derrota da Batalha de Caporetto. ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA As crônicas da coluna “Sotto la Mole” no jornal Avanti!, de Milão, tinham como objetivo passar em revista crítica uma vida cultural local. Nos artigos publicados no jornal turinense Il Grido del Popolo 4, entretanto, Gramsci buscou um ponto de vista mais amplo. No artigo “Socialismo e cultura”, de 29 de janeiro de 1916, assinado sob o pseudônimo “Alfa Gamma”, Gramsci tratou o tema da cultura em termos gerais e teóricos, como questão de conquista de autonomia indi vidual (“apoderar-se de si”) e de busca pela sincronia entre a existência natural e a existência consciente (Gramsci, 1980, p. 99). Apresentada de um ponto de vista histórico, a cultura era proposta como o desenvolvimento proble mático da “consciência da igualdade humana” entre “ple beus” e “nobres”. Nesse sentido, aparecia como atividade intimamente vinculada aos diferentes momentos em que se deu a elaboração de uma nova “base e razão histórica”, seja no surgimento da república democrática, na Antiguidade, seja no surgimento da república burguesa no contexto da Revolução Francesa (Gramsci, 1980). 271 Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA Nesse artigo do Il Grido del Popolo, Gramsci foi além da crítica dos intelectuais turinenses para explorar a questão de modo geral. O jornalista pretendia mostrar ao leitor que a cultura não é resultado de embates restritos aos intelec tuais, ou da “educação” oferecida por um grupo social a outro, mas surge e se desenvolve na história como insumo e produto dos próprios conflitos sociais. O “problema supremo da cultura”, afirmou, está justamente neste com por uma dimensão conflitiva onde existe, ao mesmo tempo, como “princípio e limite” dos momentos de luta. Dessa maneira, Gramsci procurava estabelecer o que entendia por “uma justa compreensão do conceito de cultura também em relação ao socialismo” (Gramsci, 1980, p. 102). p Esse empreendimento, ao mesmo tempo político e cultural, ganharia contornos mais precisos com a publi cação do jornal La Città Futura, no início de 1917. Já em meados de 1916, depois de quase dois anos da entrada da Itália na guerra, era generalizada a compreensão do “mons truoso preço, em vidas humanas e riquezas, que o país estava pagando” e a postura passiva diante dos acontecimentos parecia, agora, tão nociva quanto o apoio à participação no conflito (Arfè, 1965, p. 228). Assim como os demais partidos socialistas europeus, os socialistas italianos sentiam aumen tar as pressões nacionalistas e sofriam com a “flutuação de homens” no partido, dada a dificuldade de afirmar uma política capaz de orientar e engajar seus militantes, cada vez mais numerosos5. Em seus artigos, Gramsci caracterizou esse momento de impasse no interior do movimento socialista como uma “reviravolta”: “a ação política e administrativa, a eficácia energética [do partido] não está à altura da sua força efetiva” (Gramsci, 1980, p. 352). 272 Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi Daniela Xavier Haj Mussi, Alvaro Bianchi Daniela Xavier Haj Mussi, Alvaro Bianchi Gramsci criticava a política da direção do PSI, que con vertia a organização em um “fim em si mesmo”, voltada para uma ampliação numérica burocrática do partido, pouco preocupada com as “atitudes” e “sentimentos” de seus ati vistas, o que apenas a educação socialista poderia corrigir. Sendo um fim em si mesmo, o partido se revelava “um verda deiro obstáculo para o alcance do socialismo”, incapaz de oferecer respostas às necessidades de “unidade” e “coesão” no interior da massa social “amorfa, flutuante” (Gramsci, 1980, p. 352). 6 Giacinto Menotti Serrati (1874-1926) era um importante dirigente do PSI, por quem Gramsci nutria muito respeito, mas com quem polemizara muitas vezes. Ativo na Conferência de Zimmerwald, em 1917, Serrati seria responsável por apro ximar o partido italiano da Revolução Russa. 7 A responsabilidade do jornal foi, inicialmente, assumida pelo jovem ativista socia lista de Turim Andrea Viglongo, a quem Gramsci pediu para assumir a tarefa de conduzi-lo para dar-lhe um caráter mais homogêneo e afinado com uma função formativa (Gramsci, 1982). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA A ação direta poderia servir como ponto de “unidade” para as organizações socialistas por um tempo, mas não para sempre. A manutenção de suas fileiras deveria ser realizada por ideias gerais, horizontes distantes, capazes de impor uma disciplina constante e metódica (Gramsci, 1980). Ao longo de 1916, Gramsci acompanhara o debate aberto no interior da seção socialista de Turim a respeito da fundação de um novo jornal socialista na cidade, que de veria ser um “foco de entusiasmo” (Gramsci, 1980, p. 353), “a voz do partido que a cada dia alcançará espíritos novos, energias novas” (Rapone, 2011, p. 69). Dessa discussão parti ciparam muitos de seus companheiros mais próximos, como Angelo Tasca e Ottavio Pastore, em oposição às posições mais oficiais de dirigentes do PSI, como Giacinto Serrati6, que via nesta iniciativa o perigo da perda de controle do partido sobre sua imprensa (Gramsci, 1980). Para Gramsci, rejeitar a criação de um novo jornal era negar que o partido socialista estivesse “diante de uma reviravolta” na sociedade italiana, o que fazia com que o PSI ficasse aquém dos desa fios político-administrativos colocados pela expansão do par tido (Gramsci, 1980, p. 353). 273 Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA Nesse período, a atividade política de Gramsci era bas tante exterior às disputas internas do partido e mais vol tada para a vida cultural da seção socialista em Turim, onde ministrava conferências (Rapone, 2011, p. 71). Apesar disso, em sua atuação como jornalista, Gramsci demonstrava inte resse em participar dos debates partidários, especialmente para combater o economicismo e burocratismo que consu miam o PSI e colocavam seus dirigentes em “um labirinto sem saída” (Gramsci, 1980, p. 352). A criação de um novo periódico não tinha, portanto, fins puramente culturais ou jornalísticos: era uma forma de avançar na elaboração de uma “ideia geral”, de um “horizonte capaz de impor disci plina” aos ativistas socialistas (Gramsci, 1980, p. 353). Esse projeto daria seu primeiro passo nas páginas do opúsculo La Città Futura [A Cidade Futura] (Rapone, 2011, p. 68). Preparado desde fins de 1916 e publicado em fevereiro de 1917 a partir da orientação do comitê regio nal piemontês da Federazione Giovanile Socialista Italiana [Federação Jovem Socialista Italiana]7, o jornal de número único foi integralmente escrito por Gramsci. 274 A base de Gramsci para composição desse número único revelava seu vínculo com o ambiente operário socialista da capital do Piemonte – especialmente com os jovens. Além disso, foi o resultado da vivência prática na Universidade de Turim, de cultura fortemente positivista e de alguma proximidade com o “debate sobre a crise do mar xismo”, o qual conhecia ao menos desde 1914 em virtude do contato com o professor Annibale Pastore (Basile, 2014, p. 196; p. 202; D’Orsi; Chiarotto, 2012, pp. 43-44). Essa combinação é importante para compreender as nuances Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi que caracterizaram os argumentos expostos por Gramsci ao longo dos onze artigos que compuseram La Città Futura. Folhetim de educação e propaganda socialista dedicado aos jovens, La Città Futura se pretendia um “incentivo e um convite” (Gramsci, 1982, p. 3). Um incentivo a um ato de “independência e libertação” por meio do engajamento no movimento socialista, e um convite para pensar os limites nos quais o socialismo italiano esbarrava (Gramsci, 1982). “O futuro é dos jovens, a história é dos jovens” (Gramsci, 1982, p. 8 O filósofo neoidealista Benedetto Croce (1866-1952) foi editor da revista La Critica e o intelectual de maior destaque na Itália na primeira metade do século XX. Gramsci reconhece explicitamente a influência desse autor afirmando nos Quaderni del carcere que, na época em que redigiu La Città Futura, seu pensamento era “tendencialmente bastante croceano” (Gramsci, 1975, p. 1233). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA 3) – com essas palavras, Gramsci abria o jornal projetando um novo ambiente, no qual pudessem se com binar “energia” e “inteligência”, resultando na “mais per feita e frutífera afirmação” de um movimento socialista renovado (Gramsci, 1982, p. 16). A alternativa à “morte do socialismo” – expressão usada por Gramsci a partir de um artigo do filósofo Benedetto Croce (1993) publicado inicial mente em 1911 na revista La Voce e intitulado “La Morte del Socialismo” – era reencontrar a finalidade organização socialista, “engrossar sempre mais as fileiras e fechá-las” (Gramsci, 1982, p. 16)8. O objetivo era converter a situa ção defensiva na qual o socialismo estava diante de seus crí ticos em uma oportunidade para atrair e preparar jovens socialistas para um bom combate, capaz de superar o “senso comum”, o “terrível aplacador de espíritos” (Gramsci, 1982, pp. 6-7). 275 O texto de Gramsci nascia como proposição de um combate cultural e político de tipo popular, para “conceder a todos os cidadãos” a “atuação integral da própria perso nalidade humana” (Gramsci, 1982, p. 11). Com o artigo intitulado “Margini” [Margens], Gramsci apresentou no Lua Nova, São Paulo, 102: 267-298, 2017 Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA La Città Futura a ideia de que, entre as pessoas “comuns”, “o esforço feito para conquistar uma verdade” faria desta algo próprio, “mesmo se à nova enunciação não seja agregado nada fundamentalmente novo” (Gramsci, 1982, p. 23). Sua entonação colocava ênfase na vontade que se converte em sacrifício pela descoberta, na rejeição da acepção “determi nista da previsão” que reduzia o movimento socialista a uma atitude de passividade (Gramsci, 1982, p. 24). No processo da busca pela verdade, a existência de “modelos”, conti nuava, poderia ser muito útil desde que estes não fossem valorizados em termos absolutos (Gramsci, 1982). Assim como “a lei” e “o esquema”, o modelo permitiria a visão da totalidade (Gramsci, 1982). Este não poderia, entretanto, com sua lógica, substituir o movimento concreto do pensa mento que conduz à ação e vontade transformadoras. Sob esse olhar, a crise do socialismo, cuja manifestação se dava na deserção de muitos intelectuais, poderia ser expli cada de duas formas. Por um lado, como crise mais geral de “todos os ismos” (positivismo, futurismo, nacionalismo, neoidealismo etc.), concepções engajadas com as quais os intelectuais mantêm relação de exterioridade, “diletante na fé e diletante no saber” (Gramsci, 1982, p. 24). Por outro, como crise específica do socialismo como “visão livresca da vida”, na qual esta era “uma avalanche observada sempre à distância” (Gramsci, 1982, p. 25). 276 A crise do socialismo e do PSI era, para Gramsci, parte de uma crise geral e, ao mesmo tempo, um fenômeno espe cífico. Como parte da crise universal “dos ismos”, os socia listas sofriam com a debandada intelectual, já que se expan dia entre os intelectuais um ideal exterior de busca pela verdade, algo de “fora” da vida organizada, que os levava a trocar constantemente de partido e de posição política à medida que novos modismos eram lançados. Como crise singular, o socialismo italiano revelava sua incapacidade de entender esta “avalanche”, a qual assistia inerte, justificando Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi o imobilismo de maneira abstrata por meio da lógica de seu modelo determinista e afirmado como “científico” (Gramsci, 1982). Diante da crise dos intelectuais e da derrota do mito socialista – concluía – apenas a renovação interna do prole tariado poderia oferecer uma saída (Gramsci, 1982). 9 A menção ao projeto do pré-guerra não deixava de ser uma forma de lembrar os amigos Tasca, Togliatti e Terracini, que nesses dias serviam ao exército italiano no conflito. O projeto só seria concretamente retomado em fins no ano seguinte, com o término da guerra (D’Orsi; Chiarotto, 2012). 10 Claudio Treves (1869-1933) era advogado e jornalista em Turim, além de um importante dirigente socialista, figura central na revista Critica Sociale e líder da posição neutralista no interior do PSI. Em 1915, no calor das discussões sobre a participação da Itália na guerra, depois de uma troca de acusações e insultos, chegou a realizar um duelo com Benito Mussolini, no qual ambos se feriram, mas sobreviveram. ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA A expansão do proletariado e do movimento socialista em vários países durante a guerra era, para Gramsci, um claro sinal do potencial de renovação da vida popular e, tam bém, de intensificação de sua consciência – o que permitiria o surgimento de um novo mito, criado pela emergência do valor “número”, “massa”, um novo “mito de universalidade”. Essa novidade seria capaz de fazer com que os indivíduos se sentissem “partícipes de algo grandioso que está amadure cendo em cada nação, cada partido, cada seção, cada grupo” (Gramsci, 1982, pp. 26-27). Para ele, a ideia de igualdade – e não a de nacionalidade – permitiria essa renovação aos jovens, sendo Turim um laboratório do “surgimento de uma nova geração livre, sem preconceitos, que romperá a tra dição” (Gramsci, 1982, p. 32). 277 La Città Futura concluía retomando a ideia – surgida ainda antes da guerra nas conversas com outros jovens socia listas de Turim, tais como Angelo Tasca, Umberto Terracini e Palmiro Togliatti (este último menos engajado no movi mento socialista até então) – de “fundar uma revista de vida socialista para ser o abrigo das novas energias morais, do novo espírito […] idealista da nossa juventude” (Gramsci, 1982, p. 34)9. Esse jornal de número único, portanto, não era pensado como “mais um ensaio” cultural e filosófico, uma iniciativa puramente intelectual, mas funcionava como um chamado a “quem está convencido que o pensa mento e a cultura socialista têm muito que fazer ainda, e Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA que uma nova voz de jovens pode dizer muitas coisas ainda” (Gramsci, 1982, p. 34). Gramsci elaborava, aqui, o projeto de uma revista de pensamento e vida socialista cujo obje tivo definia: preparar uma nova geração socialista, capaz de superar o “relativismo” dos intelectuais – tanto os positi vistas, como os neoidealistas – e avançar na elaboração de um novo mito para preencher o vazio deixado pela falência do socialismo de bases deterministas e reformistas. Esse projeto entrava em rota de colisão com o socia lismo oficial do partido, cujo núcleo substancial se perso nificava na figura política de Claudio Treves (Gramsci, 1982; Rapone, 2011)10. Para Gramsci, o PSI havia assumido uma posição cômoda diante da vida política do país desde 1914, com a fórmula da neutralidade absoluta com relação à guerra, e com o predomínio no partido do absenteísmo político da posição reformista. No combate a essa posição, Gramsci lançou mão de referências renegadas pela “tradição teórica do socialismo italiano e internacional” (Rapone, 2011, p. 69), tal qual o pensamento neoidealista, que julgava central para compreender a cultura e os dilemas políticos do momento. O uso dessas referências recebeu críticas no interior do partido, mesmo entre os jovens, documentadas nos artigos do jornal da juventude do PSI, L’Avanguardia, no qual Gramsci foi acusado de “intelectualismo” e de ter ela borado “um jornal para iniciados”, “dificilmente compreen dido pelos leitores proletários” (Rapone, 2011, p. 69n). 278 No início de 1917, Gramsci considerava central dis cutir a “preparação cultural necessária para o desenvolvi mento da ação socialista” (Rapone, 2011, p. 70). A inicia tiva de La Città Futura em Turim era uma forma concreta, Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi prática, de atuar nesse sentido, buscando mostrar que “a simplicidade da linguagem” informativa não era o aspecto mais importante a ser pensado para favorecer a propaganda socialista, mas sim seu conteúdo formativo (Rapone, 2011, p. 105). Nesse jornal de número único, Gramsci propôs os termos desse conteúdo: pensar a relação entre disciplina e liberdade na organização socialista, redefinir “ordem” como um conceito-chave e agregador, incentivar os jovens a “sair da passividade e indiferença” e a assumir posição de prota gonistas da história (Rapone, 2011, p. 68). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA g p p Enfrentar a passividade e indiferença significava propor o problema da constituição de uma ordenação social nova, capaz de “substituir o dualismo pela unidade” e superar a falência das “construções sociais utópicas”, baseadas em um “futuro harmonicamente coordenado” (Basile, 2014, p. 205). As revoluções burguesas, nos séculos XVIII e XIX, haviam realizado este princípio utópico de maneira uni versal por meio do liberalismo. O aspecto paradoxal dessa realização evidenciava, entretanto, o caráter “nada abso luto ou rígido” da história, sendo que a realização plena do Estado suscitava uma nova antítese, o proletariado, e criava a necessidade de novas ideias-força (Basile, 2014, p. 205). Por esse motivo, Gramsci concluía: os socialistas não devem substituir uma ordem por outra. Devem, antes de mais nada, instaurar uma nova ordem em si mesmos (Gramsci, 1982, p. 11). 279 A Revolução na Rússia La Città Futura foi publicado em fevereiro de 1917, momento em que explodia uma Revolução de grandes proporções na Rússia. Apesar da apreensão, do sentimento de que existia a possibilidade de grandes explosões sociais resultantes dos impactos da guerra em toda a Europa, este foi um evento político imprevisto mesmo no ambiente socia lista. Não por acaso, a Rússia passara distante das atenções Lua Nova, São Paulo, 102: 267-298, 2017 11 O comentário de Gramsci nas páginas do Avanti! no dia 20 de abril sobre a repercussão da viagem de Morgari é seu primeiro artigo sobre a Revolução Russa “Morgari in Russia” [Morgari na Rússia] (Gramsci, 1982). 12 Trata-se de um artigo de abril de 1917, intitulado “Note sulla rivoluzione russa” [Notas sobre a Revolução Russa], publicado no jornal Il Grido del Popolo. 12 Trata-se de um artigo de abril de 1917, intitulado “Note sulla rivoluzione russa” [Notas sobre a Revolução Russa], publicado no jornal Il Grido del Popolo. 11 O comentário de Gramsci nas páginas do Avanti! no dia 20 de abril sobre a repercussão da viagem de Morgari é seu primeiro artigo sobre a Revolução Russa “Morgari in Russia” [Morgari na Rússia] (Gramsci, 1982). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA políticas de Antonio Gramsci até então. Com a revolução, seu programa para um socialismo criador “de uma ordem em si”, de uma consciência própria, se modificou no ritmo das notícias vinda do Leste. No início do ano, as informações sobre a revolução eram ainda escassas na Itália e reduziam-se em grande parte à reprodução de artigos publicados pelos jornais de Londres e Paris. Nas páginas do Avanti!, alguns artigos sobre os eventos em São Petersburgo eram assinados por Junior, pseudônimo de Vasilij Vasilevich Suchomlin, exilado russo partidário dos Socialistas Revolucionários em seu país natal. Para suprir os socialistas italianos de informações confiáveis, a direção do PSI encaminhou um telegrama ao deputado italiano Oddino Morgari, pedindo-lhe que fosse até Petrogrado e entrasse em contato com os revolucionários russos. Essa via gem fracassou e Morgari retornou à Itália em julho11. Apesar das dificuldades em interpretar os eventos que derrubaram o czarismo russo e instauraram um Governo Provisório, os primeiros argumentos de Gramsci sobre o assunto mostram seu esforço por aproximar a Revolução Russa da ideia de criação de uma “nova ordem”12. Pelos artigos escritos entre abril e maio de 1917, nota-se que ele tinha muito interesse em interpretar os acontecimentos do Leste para convertê-los em lições possíveis aos italianos. De acordo com Gramsci, sabia-se que a revolução havia sido “feita pelos proletários” e que existia “um comitê de dele gados operários que controla o trabalho dos entes admi nistrativos” (Gramsci, 1982, p. 138). Mas não era isso que dava à Revolução um caráter proletário. Para isso, era neces sário investigar os “fatores espirituais”, ou seja, verificar se 280 Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi a Revolução era também um “fenômeno de costumes”, um “fato moral” (Gramsci, 1982, p. 138). Esse programa de pesquisa – continuidade de suas ela borações anteriores – exigia que fosse investigada a capa cidade da revolução em promover uma nova cultura. A revo lução não deveria ser considerada apenas como um fato, algo mecânico, mas também como ato proletário em dire ção ao socialismo. Na diferença entre ato e fato é possível identificar uma chave de leitura neoidealista, inspirada em Giovanni Gentile (1937), que alguns anos antes havia desen volvido a noção de fato espiritual, objeto do conhecimento, só conhecido quando a objetividade se dissolve na ação real do sujeito do conhecimento. 13 “O ato, se não converter-se em um fato, se deve ser compreendido em sua natu reza atual, o ato puro, só pode ser pensamento. O fato é a negação do pensamento porque o próprio pensamento cria seu outro, a natureza. Apenas do ato se desce ao fato, se está fora do pensamento, no mundo da natureza. Não existem fatos espirituais, mas atos; na verdade, não existe se não o ato do espírito” (Gentile, 1937, p. 29). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA A subjetividade humana, aqui, determinava a natureza do fato13. O uso que Gramsci fez da noção de ato proletário servia, desse modo, para destacar a intervenção do proletariado na história como ação de um espírito superior a todos os interesses particulares. 281 Como fato moral e ato proletário, a revolução na Rússia era interpretada por Gramsci como de natureza essencial mente antijacobina. Ela não era a expressão de interesses particularistas de classe, nem um “fato violento”, como havia sido a revolução burguesa na França, muito menos se asse melhava a “um regime autoritário que substitui outro regime autoritário” (Gramsci, 1982, p. 138). Para o jovem jornalista, a nova revolução havia “destruído o autoritarismo e o subs tituído pelo sufrágio universal, estendendo-o até mesmo às mulheres” e, dessa maneira, instituído a “liberdade” e dado lugar para “a voz livre da consciência universal” (Gramsci, 1982, p. 139). A revolução era, aqui, um ato do espírito; Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ela está na consciência de todos e se transformará em decisão irrevogável apenas possa expressar-se em um ambiente de liberdade espiritual absoluta, sem o qual o sufrágio seria pervertido pela intervenção da polícia e pela ameaça da forca ou do exílio. O proletariado industrial já está preparado para essa passagem mesmo culturalmente; o proletariado agrícola, que conhece as formas tradicionais do comunismo comunal, está também preparado para a passagem a uma nova forma de sociedade (Gramsci, 1982, p. 139). O paralelo entre a Revolução Francesa de 1789 e a Revolução Russa que ocorria em 1917 era muito comum na imprensa da época e não se tratava de uma invenção de Gramsci. Na imprensa socialista, já no dia 19 de março, Il Grido del Popolo havia publicado um editorial, assinado por Nar, pseudônimo de Gaetano Polverelli, intitulado “L’89 di Russia” [O 1789 da Rússia]. A analogia com a França tam bém havia sido retomada logo em seguida, em 23 de março, pelo deputado e dirigente socialista Filippo Turati, em dis curso na Câmara de Deputados: 282 a Revolução Russa, cujo acontecimento parece recordar formidavelmente a revolução da França, de fato supera de uma só vez as fases de [17]89 e [17]93, abate todos os obstáculos e triunfa sem retorno e represálias possíveis (Turati, 1917, p. 13377). a Revolução Russa, cujo acontecimento parece recordar formidavelmente a revolução da França, de fato supera de uma só vez as fases de [17]89 e [17]93, abate todos os obstáculos e triunfa sem retorno e represálias possíveis (Turati, 1917, p. 13377). Antes mesmo de 1917, a analogia se tornara comum na Rússia e no movimento socialista internacional em virtude dos levantes de 1905. A novidade introduzida por Gramsci nesse paralelo histórico estava na recusa do jacobinismo. Além da interpretação antijacobina da revolução promovida Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi por Gaetano Salvemini (1905)14, Georges Sorel (1847-1922) e Benedetto Croce também eram fontes do antijacobinismo de Gramsci. A primeira delas fora a difusão, na Itália, da obra de Georges Sorel, Réflexions sur la violence [Reflexões sobre a violência] (2013), na qual o teórico do sindicalismo revolucionário recusava as “abominações” das revoluções burguesas e, em especial, os acontecimentos de 179315. Gramsci fez uma referência explícita a essa obra em um artigo de 30 de outubro de 1916 (Gramsci, 1980). 14 A interpretação de Gramsci da cultura no contexto da revolução na França possuía forte identidade com as pesquisas de Gaetano Salvemini (1873-1957), intelectual neoidealista, socialista, de origem meridional e radicado em Florença. Este concebia as grandes transformações políticas do final do século XVIII na França como resultado da interferência das grandes massas na política em um momento histórico, processo este que poderia ter resultados heróicos ou desas trosos, a depender da orientação seguida pela “multidão exasperada e frenética” (Salvemini, 1905, p. 138). Salvemini buscava se distanciar dos revolucionários franceses, girondinos e jacobinos, para mostrar que nenhum dos dois grupos fora responsável por sistematizar, isoladamente, o sucesso ou o fracasso das lutas polí ticas. A revolução, aqui, era pensada como obra intelectual em sentido amplo, realizada em grande medida pelos “filósofos pré-revolucionários”, responsáveis por refutar “as velhas regras no campo da vida prática” e abrir um espaço possível por onde as “massas incultas” poderiam caminhar em um contexto de crise das classes dominantes (Salvemini, 1905, p. 354). 14 A interpretação de Gramsci da cultura no contexto da revolução na França possuía forte identidade com as pesquisas de Gaetano Salvemini (1873-1957), intelectual neoidealista, socialista, de origem meridional e radicado em Florença. Este concebia as grandes transformações políticas do final do século XVIII na França como resultado da interferência das grandes massas na política em um momento histórico, processo este que poderia ter resultados heróicos ou desas trosos, a depender da orientação seguida pela “multidão exasperada e frenética” (Salvemini, 1905, p. 138). Salvemini buscava se distanciar dos revolucionários franceses, girondinos e jacobinos, para mostrar que nenhum dos dois grupos fora responsável por sistematizar, isoladamente, o sucesso ou o fracasso das lutas polí ticas. A revolução, aqui, era pensada como obra intelectual em sentido amplo, realizada em grande medida pelos “filósofos pré-revolucionários”, responsáveis por refutar “as velhas regras no campo da vida prática” e abrir um espaço possível por onde as “massas incultas” poderiam caminhar em um contexto de crise das classes dominantes (Salvemini, 1905, p. 354). 15 “Temos o direito de concluir, então, que não se pode confundir as violências sindicais exercidas ao longo das greves por proletários que desejam a derrubada do Estado com esses atos de selvageria que a superstição do Estado sugeriu aos revolucionários de 1793 quando tiveram o poder nas mãos e puderam oprimir os vencidos de acordo com os princípios que haviam recebido da Igreja e da realeza. Temos o direito de esperar que uma revolução social levada a cabo pelo sindicatos pelos puros sindicalistas não será manchada pelas abominações que marcaram as revoluções burguesas” (Sorel, 2013, p. 93). Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA As ideias de Sorel também se faziam sentir na obra de Benedetto Croce, o qual, em Cultura e vita morale [Cultura e vida moral], publicado em 1914, identificara o jacobinismo com a “mentalidade maçônica”, ou seja, com uma maneira de pen sar que simplifica tudo “em nome da razão, da liberdade, da humanidade, da fraternidade, da tolerância”, procedendo por meio de abstrações e “classificando fatos e homens por sinais exteriores e fórmulas” (Croce, 1993, p. 143). Gramsci conhecia muito bem esse livro do filósofo napolitano – a 283 – a 15 “Temos o direito de concluir, então, que não se pode confundir as violências sindicais exercidas ao longo das greves por proletários que desejam a derrubada do Estado com esses atos de selvageria que a superstição do Estado sugeriu aos revolucionários de 1793 quando tiveram o poder nas mãos e puderam oprimir os vencidos de acordo com os princípios que haviam recebido da Igreja e da realeza. Temos o direito de esperar que uma revolução social levada a cabo pelo sindicatos pelos puros sindicalistas não será manchada pelas abominações que marcaram as revoluções burguesas” (Sorel, 2013, p. 93). Lua Nova, São Paulo, 102: 267-298, 2017 Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ponto de extrair e incluir uma passagem dele em La Città Futura – e tomaria o título como inspiração, em 1918, para a proposta de criação de um Clube de Vida Moral em Turim, com vistas a “habituar os jovens que aderem ao movimento político e econômico socialista à discussão desinteressada dos problemas éticos e sociais” (Gramsci, 2009, p. 177). p ( p ) Essas eram as referências intelectuais que estruturavam, nos primeiros meses de 1917, a interpretação antijacobina de Gramsci a respeito da Revolução Russa. Retomando cer tas ideias que discutira nos meses anteriores em relação ao positivismo – e seguindo mais de perto o argumento de Sorel – o jornalista socialista afirmava que o jacobinismo era um fenômeno puramente burguês, pois estava desti nado a realizar os interesses particulares desta classe (Sorel, 2013). Como ato proletário, portanto, a revolução na Rússia se distanciava da experiência francesa. Mais do que inter pretar de maneira totalmente fiel os acontecimentos de Petrogrado, Gramsci desenvolvia o programa político que havia anunciado como problema nas reflexões sobre a cul tura, em 1916, e como necessidade para a reorganização da juventude socialista no início de 1917. Sua interpretação da revolução visava mostrar a impossibilidade de avanço de uma revolução operária sem a ruptura definitiva com o modelo político jacobino, ou seja, com o mundo político e cultural burguês. 284 O antijacobinismo como chave de leitura da Revolução Russa seria mais uma vez usado por Gramsci no artigo “Un po’ di Russia” [Um pouco da Rússia], publicado em 15 de maio de 1917, na contraposição da ação de uma “maio ria efetiva” aos “abusos de uma minoria facciosa e jaco bina” (Sorel, 2013, p. 280). Foi também nessa chave que o jornalista socialista interpretou as forças dirigentes da revolução no importante artigo publicado em 28 de julho de 1917, “I massimalisti russi” [Os maximalistas russos]. Expressando o apoio que os bolcheviques encontravam Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi entre os socialistas, Gramsci afirmou que os maximalistas eram “a continuidade da revolução, o ritmo da revolução, por isso, a própria revolução” (Gramsci, 1982, p. 265). Os bolcheviques expressavam o novo “estado de ânimo” sobre o qual Gramsci falara em 1916, eram a encarnação da “ideia -limite do socialismo” que não comportava um compromisso com o passado milenar (Gramsci, 1982). 16 Apenas em agosto, com a viagem à Itália de uma delegação representando os sovietes russos, da qual faziam parte Iosif Goldemberg e Aleksandr Smirnov, os socialistas italianos puderam entrar em contato direto com os revolucionários russos. No Avanti!, Gramsci resumiu os discursos de Goldenberg e Smirnov em uma manifestação que ocorreu em Turim no dia 13 de agosto, Il compito della rivo luzione russa [A tarefa da revolução russa](Cf. Gramsci, 1982). Sobre a calorosa recepção aos delegados dos sovietes em Turim, ver Spriano (1972). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA p Gramsci defendeu os maximalistas das campanhas de difamação e calúnia promovidas na imprensa comercial em todos os países ocidentais e afirmou seu papel político como imprescindível para que a revolução não se interrompesse no país (Gramsci, 1982). Preocupado em intervir para forta lecer uma posição socialista verdadeiramente autônoma, no início de agosto, Gramsci relatou de maneira entusiasmada em Il Grido del Popolo dois grandes comícios que reuniram dezenas de milhares de pessoas em Turim para receber uma delegação russa que percorria a Itália para falar sobre a “novidade” da revolução (Gramsci, 1982)16. Essas atividades promoviam, em sua opinião, um verdadeiro “espetáculo das forças proletárias e socialistas solidárias com a Rússia revo lucionária” (Gramsci, 1982, p. 274). A visita da delegação russa à Itália se dava em um momento de importante crise do Governo Provisório – incapaz de estabilizar-se, em um contexto de protestos de massa contra as novas ofensivas militares russas – e da escalada de repressão aos bolche viques que fez com que Lenin fugisse para a Finlândia para poder continuar atuando politicamente (Carr, 1964). O jor nal La Stampa, por sua vez, noticiava a fuga de Lenin (“que a esta altura está na Alemanha”) como uma confirmação da 285 Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA especulação sobre a infiltração alemã entre os extremistas da revolução, comemorando a “perda irreparável” de sua autoridade diante das massas russas (KERENSKI…, 1917). Na segunda metade de 1917, como redator e, logo em seguida, diretor do jornal Il Grido del Popolo, a intervenção de Gramsci se voltou contra a interrupção da revolução, con victo da necessidade de que esta avançasse e produzisse uma nova racionalidade e subjetividade históricas. A maneira que encontrou para fazer isso foi a defesa da Revolução Russa como um ato antijacobino, como processo no qual todas as ideias poderiam ter lugar e em que os mais diver sos grupos políticos poderiam ser criados em torno delas. Sua defesa da posição maximalista aparecia como defesa do “último elo lógico do devir revolucionário” (Gramsci, 1982, p. 265). A revolução era, portanto, um fenômeno eminen temente inclusivo, no qual a direção dos maximalistas sobre os demais grupos era resultado de sua maior capacidade de afirmação cultural. 286 Em agosto de 1917, Gramsci insistia na semelhança entre o que entendia por cultura e a revolução em curso. A revolução representava a “substituição de valores, de pessoas, de categorias, de classes” (Gramsci, 1982, p. 282). A essa altura, já havia concentrado seu interesse na elabo ração programático-partidária, mas o problema da cultura se mantinha como elo de ligação com sua origem como cro nista corrosivo da vida cotidiana. A crítica ao jacobinismo, entretanto, mantinha seu lugar. Para Gramsci, a ausência de jacobinismo teria impedido que os extremistas fossem “afo gados no sangue”, que estes não tivessem na Rússia o mesmo destino de Gracchus Babeuf na França. A revolução como ato proletário era o que permitia aos bolcheviques conver terem seu pensamento “em força operante na história”, a força vista aqui com resultado de um movimento amplo das consciências, o que seria impossível do ponto de vista burguês-jacobino. Apenas a revolução projetada como ato Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi proletário seria capaz de efetivar-se nas consciências, “enco rajá-las” e “conquistá-las”, tal como os partidários de Lenin faziam. Finalmente, a seu ver, surgira um novo “estado de ânimo” capaz de encarnar-se em uma “multidão de indiví duos” e as ideias haviam voltado a produzir “frutos no ter reno da ação” (Gramsci, 1982, pp. 266-267)17. 17 Gramsci retomaria, em setembro de 1917, a ideia de que Lenin e os bolchevi ques representavam o futuro da revolução: “Kerenski representa a fatalidade histó rica, enquanto Lenin representa o devir socialista e nós estamos com ele, com toda nossa alma” (Gramsci, 1982, p. 285). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA O antijacobinismo de Gramsci, portanto, guardava uma distância em relação aos intelectuais neoidealistas italianos e ao sindicalismo revolucionário. Diferente destes, Gramsci defendia a Rússia revolucionária como encarnação viva e efe tiva de um novo estado de ânimo histórico, como “uma con tínua mudança no bloco amorfo do povo”, em que “novas energias são suscitadas” e “novas ideias-força propagadas”. Diferente do modelo político burguês, os sujeitos dessa revolução eram indivíduos, ao mesmo tempo, autônomos e capazes de pensar e agir sem se comportar como “mino rias despóticas” (Gramsci, 1982, p. 267). Em 15 setembro de 1917, quando os bolcheviques enfrentaram de maneira vitoriosa as tropas lideradas pelo general Lavr Kornilov, Gramsci referiu-se, mais uma vez, a uma “revolução que ocorreu nas consciências” (Gramsci, 1982, p. 342). O uso da força para resistir à ofensiva militar contrarrevolucionária tinha natureza distinta do uso da força burguesa, nascia do movimento contínuo das consciências. Por isso, também, em 28 de setembro, Gramsci definiu Lenin como “o agi tador das consciências, o despertador das almas dormentes” (Gramsci, 1982, p. 359). 287 Semanas que parecem anos Em 7 de novembro, os bolcheviques assumiram o con trole dos pontos estratégicos de Petrogrado e prenderam Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA membros do Governo Provisório, enquanto outros, como Kerenski, fugiam (Carr, 1964). Na mesma noite, “o poder foi passado para as mãos do Soviet dos Deputados dos Operários, Soldados e Camponeses”. Nos dias seguintes, foram adotados decretos referentes à paz e distribuição de terras, e criou-se o Conselho de Comissários do Povo, o “primeiro governo dos operários e camponeses” (Carr, 1964, p. 100). As dificul dades e desafios políticos nesse novo contexto eram enor mes e o impacto da tomada do poder pelos bolcheviques na esfera internacional era gigantesco. Na Itália, esses eventos foram sentidos de maneira ainda mais intensa, em virtude da importante derrota militar para os alemães e o Império austro-húngaro, que ocorrera na semana anterior à revolução, na batalha de Caporetto (D’Orsi, 1985). As perdas humanas foram muitas, assim como o número de prisioneiros e feridos, mas a principal derrota desse confronto foi política, com uma grande reti rada do exército italiano da região do Rio Piave, o afasta mento do célebre general Luigi Cadorna de suas funções e a desmoralização do governo italiano e da campanha interventista. 288 Quando a notícia da tomada do poder pelos bolche viques chegou à Itália, portanto, esta encontrou um ambiente político devastado por uma derrota militar de grandes proporções e com sentimento de insatisfação muito difundido. A imprensa, para não falar da derrota, comemo rava a “retirada perfeita” das tropas italianas, sua recompo sição e o heroísmo dos soldados. O governo havia lançado mais um decreto de convocação às armas dos italianos nas cidos entre 1874 e 1899, mas a continuidade do conflito não conseguia se sustentar na opinião pública como antes. Poucos dias depois, em 9 de novembro, Cadorna foi substi tuído. As notícias do “grave conflito entre o Estado-Maior russo e os sovietes” inundavam a imprensa, estabelecendo Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi um incontrolável senso de comparação entre o que se pas sava nos dois países. Para Gramsci, essas foram semanas de grande mudança. Desde fins de outubro, o jovem jornalista sardo ocupava a posição de secretário da Comissão Executiva Provisória da seção socialista de Turim e de redator-chefe do Il Grido del Popolo, em virtude da prisão de Maria Giudice depois dos protestos massivos de agosto que tomaram a Itália, exi gindo pão e o fim da guerra (Gramsci, 2009). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA Em suas cartas deste período, Gramsci relatou o esforço por atuar como redator, para transformar o jornal em um veículo capaz de oferecer um “tratamento dos problemas concretos do pro grama” do partido e “fortalecer uma consciência unitária do proletariado italiano” (Gramsci, 2009, p. 170; p. 173). Desde o início de novembro, a censura sobre os jornais socialistas aumentara sensivelmente (Cortesi, 1969). Ao comentar a censura parcial ou total a uma série de artigos sobre a grave crise militar italiana, publicados na página de Turim do Avanti! e do Il Grido del Popolo, Gramsci ironizou o papel do censor em eliminar da opinião pública apenas a posição socialista e, em especial, aquela elaborada na capi tal do Piemonte (Gramsci, 1982). O jovem dirigente socia lista notava que a censura se intensificava justamente no momento em que as classes sociais e o governo reconheciam a necessidade de elaboração de “novas fórmulas” políticas capazes de substituir a decadência de ideias como “todos às armas” e “por um fronte único militar”, difundidas no início da guerra (Gramsci, 1982, p. 428). 289 Gramsci percebia aquelas como “semanas que parecem dois anos”, dada a intensidade com que os acontecimentos internacionais e nacionais impactavam a Itália, com a falên cia sucessiva das fórmulas políticas que as classes dominantes tentavam implementar em resposta à crise, tais como “pela resistência interna” e “pela concórdia nacional” (Gramsci, 1982, p. 429). Em 10 de novembro, a imprensa noticiou Lua Nova, São Paulo, 102: 267-298, 2017 Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA o “golpe de Estado na Rússia”, com a “sede do palácio de inverno ocupada pelos maximalistas”, o “apossamento do poder pelos sovietes”, a fuga de Kerenski e a prisão de mui tos ministros de seu governo (IL NUOVO…, 1917, p. 1). Em artigo publicado duas semanas depois, em 24 de no vembro, Gramsci relembrou o artigo do Il Grido del Popolo, de meses antes, no qual afirmara que a revolução não po deria “se interromper com a fase Kerenski” (Gramsci, 1982, p. 450). Em outro artigo da mesma edição, comentou o sentimento de inquietude e incerteza que predominava no movimento socialista italiano em relação à Revolução Russa, do “tumulto de intenções ainda incertas que se exprimem genericamente” do qual ele próprio tomava parte (Gramsci, 1982, p. 452). A revolução no Leste era vivida como uma espécie de abalo sísmico, um “tumulto” que não limitava à “mudança da fórmula política”, mas se realizava como processo de “interiorização, de intensificação da vida moral” (Gramsci, 1982, p. 453). Isso levava Gramsci, mais uma vez, à conclusão de que os socialistas italianos eram revolucionários na ação, mas reformistas no pensamento. A ação radical, porém, deveria encontrar o pensamento revolucionário, despren dido de qualquer reformismo (Gramsci, 1982). A revolução aguçava um sentimento de inquietude que nascera nos anos da guerra e se expressara na preocupação com a cultura, com a reforma do pensamento e dos métodos de ação socia listas (Gramsci, 1982). 290 Em 12 de novembro, a fração parlamentar do Partido Socialista, liderada por Filippo Turati e Claudio Treves, sob a pressão nacionalista depois de Caporetto, passou a assu mir uma atitude explicitamente nacionalista e passado a advogar a defesa da pátria, distanciando-se do neutralismo dos anos precedentes. Nas páginas de Critica Sociale, Turati e Treves publicaram o artigo, em novembro daquele ano, no qual afirmavam a necessidade de o proletariado defender a Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi pátria na hora do perigo (Treves; Turati, 1917; ver o comen tário de Spriano, 1972, p. 451). A fração intransigente-revo lucionária do partido socialista também se organizou para fazer frente à nova situação. Nos primeiros dias de novem bro, dirigentes da fração convocaram uma reunião secreta em Firenze, para discutir “a orientação futura de nosso par tido” (Spriano, 1972, p. 454). 18 Segundo Marx, “não se trata do grau maior ou menor de desenvolvimento dos antagonismos sociais decorrentes das leis naturais da produção capitalista. Trata-se dessas próprias leis, dessas tendências que atuam e se impõem com férrea neces sidade. O país industrialmente mais desenvolvido não faz mais do que mostrar ao menos desenvolvido a imagem de seu próprio futuro” (Marx, 2013, p. 78). E mais adiante afirmava: “Ainda que uma sociedade tenha descoberto a lei natural de seu desenvolvimento […], ela não pode saltar suas fases naturais de desenvolvimento, nem suprimi-las por decreto. Mas pode, sim, abreviar e mitigar as dores do parto” (Marx, 2013, p. 79). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA Gramsci, como visto, assumiu funções importantes na seção local do partido e participou do encontro representando Turim (Spriano, 1972). Na reu nião, o diretor de Il Grido del Popolo alinhou-se com aque les, como Amadeo Bordiga, que achavam que era neces sário agir, enquanto Serrati e outros se pronunciaram pela manutenção da antiga tática neutralista. A reunião terminou reafirmando os princípios do internacionalismo revolucio nário e a oposição à guerra, mas sem nenhuma orientação a respeito do que fazer (Spriano, 1972). Gramsci retornou a Turim convencido de que era o momento de agir e escreveu mais uma vez a respeito dos acontecimentos na Rússia no artigo intitulado “La rivolu zione contro ‘Il Capitale’” [A revolução contra “O Capital”]. “A revolução dos bolcheviques inseriu-se, definitivamente, na revolução geral do povo russo”, começou (Gramsci, 1982, p. 513). Depois de terem impedido que a revolução se estag nasse, os partidários de Lenin haviam chegado ao poder em condições de estabelecer “sua ditadura” e elaborar as “for mas socialistas às quais a revolução deverá finalmente ade quar-se, para continuar a desenvolver-se harmonicamente, sem grandes choques” (Gramsci, 1982, p. 513). 291 Até então, a palavra ditadura não fazia parte do léxico de seus artigos, mas mesmo aqui é possível ver seu esforço por interpretar o uso da força em um quadro mais geral, “cultural” em sentido amplo. Para Gramsci, a “revolução dos bolcheviques se baseava mais em ideologias do que em fatos” (Gramsci, 1982, p. 513). Justamente por isso, afir mou, não fora na letra de Marx que os maximalistas russos Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA teriam encontrado um guia para a ação. Na Rússia, escre via, O Capital era “o livro dos burgueses mais do que dos operários” (Gramsci, 1982, p. 513). Gramsci referia-se ao Prefácio de 1867, no qual Marx afirmara que as nações com maior desenvolvimento capitalista mostravam o caminho a ser percorrido pelas demais e não poderia saltar “fases natu rais”18. Tanto o determinismo reformista predominante no socialismo italiano como o marxismo legal russo tomavam como base essa elaboração e a usavam para defender a for mação de uma burguesia e de uma sociedade industrial plenamente desenvolvida antes de qualquer transformação socialista na Rússia. Os bolcheviques, segundo Gramsci, “não são marxistas” pois, embora não renegassem “o pen samento imanente, vivificador” de Marx, “renegam algumas afirmações do Capital”, ou seja, recusam “uma doutrinazinha exterior, de afirmações dogmáticas indiscutíveis” (Gramsci, 1982, pp. 513-514). O raciocínio de Gramsci sobre o reformismo se esten dia para pensar não apenas as organizações da classe traba lhadora. Gramsci via o reformismo como um momento do desenvolvimento da cultura e da política, vinculado ao avanço da conformação do capitalismo nos diversos países. No caso italiano, porém, constatava que a burguesia estava atrasada em relação ao proletariado. Durante décadas os trabalhadores italianos haviam desenvolvido uma concepção política própria, ainda que limitada, e que se materializava no reformismo. A burguesia local, por sua vez, apenas depois da entrada da Itália na guerra, em 1914, fora capaz 292 Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi de dar os primeiros passos no sentido de uma concepção econômica própria, da formação da “individualidade de classe”, que se materializava em uma forma de “reformismo burguês”, o nacionalismo econômico. Em um plano mais geral, para Gramsci, esses dois programas político-econô micos – reformismo e nacionalismo – eram equivalentes, dada sua “aparência revolucionária” e seu conteúdo conser vador (Gramsci, 1982, p. 454). Depois da Revolução Russa, entretanto, o proletariado italiano – assim como o russo – tinha a oportunidade de avançar e atingir a maturidade de pensamento no contato com o socialismo revolucionário (Gramsci, 1982). De um ponto de vista histórico, continuava, a este novo momento do pensamento das classes trabalhadoras, a burguesia deve ria opor a “doutrina liberal” desenvolvida e operante em países como Inglaterra e Estados Unidos (Gramsci, 1982). ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA Contudo, na Itália – como na Rússia – esta doutrina era simplesmente incapaz de se desenvolver. Por esse motivo, a principal lição da revolução na Rússia era a de que o prole tariado seria capaz de pular o momento revolucionário bur guês e mesmo realizar seus princípios, particularmente o da liberdade. 293 Em dezembro, portanto, Gramsci afirmou o caráter expansivo da revolução bolchevique em termos histó ricos e geográficos. Os “maximalistas” apropriaram-se do poder para evitar que a revolução estagnasse e agora se dedicavam à elaboração das formas socialistas pelas quais a revolução poderia se adaptar para continuar a se desen volver. A interpretação de Gramsci sobre Marx tinha inspi ração croceana, procurava aquilo que era morto e vivo em seu pensamento (Gramsci, 1982). Porém, diferente do que pensava o filósofo napolitano, para Gramsci, a superação do “cânone” marxista – ou de suas “incrustações positivistas e naturalistas” – não era uma atitude abstrata, puramente intelectual (Gramsci, 1982, p. 514). Lua Nova, São Paulo, 102: 267-298, 2017 Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA Gramsci considerava o marxismo um bom “cânone de crítica histórica” para períodos de “normalidade” do desen volvimento do capitalismo, nos quais se poderia observar e fazer previsões a respeito do avanço do “caos-povo para uma ordem de pensamento, sempre mais consciente da pró pria força” (Gramsci, 1982, pp. 513-514). A guerra, contudo, marcara um ponto de virada no “despertar das vontades”, em particular na Rússia, onde o “sofrimento acumulado” se convertera em um uníssono, mecanicamente e espiritual mente Gramsci, 1982, p. 415). Nesse novo contexto de lutas, a “pregação socialista” deveria adquirir um novo papel, pro jetando a experiência do povo russo internacionalmente, criando uma “vontade social” e consciência novas, dispostas a questionar a necessidade de “esperar” o desenvolvimento da burguesia. Mesmo sendo o pensamento de uma minoria, essa pregação poderia se converter no “meio” pelo qual o proletariado vivenciaria uma nova e inesperada experiência (Gramsci, 1982). 294 Nesse último artigo de 1917 sobre o tema, escrito com poucos meses de distância em relação às primeiras inter venções sobre a Revolução Russa, é notável o esforço de Gramsci por captar o significado mais geral dos aconteci mentos, especialmente no que se referia ao papel dos bol cheviques. As primeiras intervenções eram marcadas pela ideia de continuidade entre o Governo Provisório russo e os maximalistas, mas, ao final do ano, Gramsci falou de maneira incisiva sobre a necessidade de levar a cabo a crí tica do reformismo, inclusive por meio da força. Para essa mudança de abordagem, contribuía não apenas a con quista do poder pelos bolcheviques em outubro-novembro, mas também o balanço da atitude dos dirigentes socialis tas italianos diante dos grandes protestos contra a guerra de agosto, especialmente em Turim. No fim de 1917, por tanto, Gramsci sobrepunha de maneira sutil as duas reali dades nacionais, em uma comparação anunciada na ideia Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi de internacionalização da propaganda socialista e da cons ciência das massas proletárias nos diferentes países. Conclusões Fundado em 1892, em Turim, o jornal semanário Il Grido del Popolo expressava as concepções do socialismo que se desenvolvia no Piemonte, de traços marcadamente inte lectuais e racionalistas. Gramsci foi seu último redator-chefe, entre agosto de 1917 e setembro de 1918, e esta foi sua pri meira experiência à frente de um periódico (Paggi, 1970). Em outubro de 1918, um ano depois da tomada do poder pelos bolcheviques na Rússia, Il Grido del Popolo anunciaria o encerramento de suas atividades para dar lugar à publi cação de uma edição piemontesa do Avanti! – que se somava às edições romana e milanesa – na qual Gramsci passaria a atuar como jornalista. A ideia era manter, no novo jornal, o formato de “crônica” das questões envolvendo a cidade de Turim e toda a região do Piemonte, buscando uma amplia ção do enraizamento e organização local do PSI (Gramsci, 2009). Esse período coincidia com o fim da guerra e o começo do retorno dos soldados do front, dentre eles muitos socialistas (Fiori, 2003). 295 Com a recomposição da seção socialista de Turim, Gramsci assumiu então uma posição marginal em relação à direção do partido, dedicando-se integralmente a ativi dades de propaganda. Suas críticas às posições reformistas do PSI eram expressivas, mas não significavam ainda a dis posição para a organização ou participação ativa em alguma tendência ou fração interna (Terracini, 1976, p. 251). Sua atuação se voltou toda para a integração e desenvolvimento político na base da organização, especialmente aquela ope rária. O objetivo era educá-la de maneira paciente e dedi cada para que esta pudesse compreender e reagir diante dos equívocos da política reformista da organização. Foi nesse Lua Nova, São Paulo, 102: 267-298, 2017 Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA contexto que, tanto a reflexão sobre os intelectuais como aquela sobre os limites do Estado, ganhariam força. O ano 1917 esteve longe de representar, portanto, o ponto final da análise de Gramsci sobre a revolução na Rússia ou a crise do socialismo na Itália. Nos anos seguintes, muitas das interpretações formuladas no período aqui inves tigado seriam retomadas, reelaboradas ou mesmo descar tadas por Gramsci. A reconstrução das ideias do jovem jorna lista socialista no período 1916-1917, portanto, não alcança posições definitivas a respeito da sua interpretação sobre a Revolução Russa e as possibilidades revolucionárias na Itália. Os anos de 1916 e 1917 marcaram o início da atividade de Gramsci como jornalista da imprensa socialista italiana, período em que este buscou compatibilizar sua formação intelectual e erudição e a escrita de artigos voltados para um público ativista e partidário. Um período que coincidiu, além disso, com o aprofundamento da crise humanitária gerada pela I Guerra Mundial e com as grandes esperanças projetadas pela revolução em um país atrasado economica mente no interior da Europa. Até fins de 1916, a cultura e a crítica dos intelectuais deram o tom aos artigos de Gramsci. Tratava-se de reconstruir o “interior” do socialismo italiano, reencontrar seu “estado de ânimo”, iniciativa que se crista liza de maneira programática no opúsculo La Città Futura. A partir de março de 1917, porém, este programa é paula tinamente deslocado, passa a dar lugar a uma tentativa de interpretar a renovação “espiritual” socialista com base na experiência revolucionária em curso. As ferramentas intelectuais de que Gramsci dispunha até então para armar este projeto de reforma cultural em seu partido passam a ser usadas para construir uma nova intervenção analítica a res peito dos grandes acontecimentos. 296 Ao final desse período – especialmente depois da tomada do poder pelos bolcheviques na Rússia – essas “ferra mentas” passam por uma espécie de reforma antirreformista. Lua Nova, São Paulo, 102: 267-298, 2017 Daniela Xavier Haj Mussi, Alvaro Bianchi Daniela Xavier Haj Mussi, Alvaro Bianchi A cultura é aproximada da revolução não mais pela lógica da precedência temporal, mas como aspecto interno, fonte de legitimidade que se renova na experiência política das massas. De forma tímida, mas explícita, Gramsci introduz, em sua interpretação “culturalista” (ou “culturista”, termo usado na época), o elemento da força militar, até então considerado impróprio em um “ato de vontade” proletário. Daniela Xavier Haj Mussi é pós-doutoranda em Ciência Política pela USP. Alvaro Bianchi é livre-docente do Departamento de Ciência Política da Unicamp. Alvaro Bianchi é livre-docente do Departamento de Ciência Política da Unicamp. é livre-docente do Departamento de Ciência Política da Unicamp. 297 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA Cultura e revolução se encontram pela primeira vez em seus escritos e formam a base para o par conceitual “consenso ‑força” que o acompanharia até os anos 1930, em seus últi mos escritos como prisioneiro do fascismo. Daniela Xavier Haj Mussi é pós-doutoranda em Ciência Política pela USP. Bibliografia g ARFÈ, G. 1965. Storia del socialismo italiano: 1892-1926. Torino: Giulio Einaudi. BASILE, L. 2014. “Caro Maestro”, “Eccezionale studente”: sul rapporto di A. Gramsci con V. A. Pastore. Ipotesi e Riscontri. Giornale Critico della Filosofia Italiana, v. 10, n. 1, p.187-211. CARR, E. H. 1964. La rivoluzione bolscevica 1917-1923. Torino: Giulio Einaudi. CORTESI, L. 1969. Il socialismo italiano tra riforme e rivoluzione: dibatti congressuali del PSI 1892-1921. Bari: Laterza. CROCE, B. 1993 [1914]. Cultura e vita morale. Napoli: Bibliopolis. ______. 1915. Frammenti di etica. La Critica: Rivista di Letteratura, Storia e Filosofia, Napoli, v. 13, pp. 153-155. CUOCO, V. 1913 [1901]. Saggio storico sulla rivoluzione napoletana del 1899 seguito dal rapporto al Cittadino Carnot di Francesco Lomonoco: a cura di Fausto Nicolini. Bari: Laterza. (Coleção Scrittori d’Italia.) D’ORSI, A. 1985. La rivoluzione antibolscevica: fascismo, classi, ideologie (1917 1922) Mil F A li D’ORSI, A. 1985. La rivoluzione antibolscevica: fascismo, classi, ideologie (1917-1922). Milano: Franco Angeli. (1917-1922). Milano: Franco Angeli. Lua Nova, São Paulo, 102: 267-298, 2017 ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA D’ORSI, A.; CHIAROTTO, F. 2012. Introduzione. In: GRAMSCI, A. D’ORSI, A.; CHIAROTTO, F. 2012. Introduzione. In: GRAMSCI, A. Scritti della libertà: 1910-1926. Torino: Riuniti. Scritti della libertà: 1910-1926. Torino: Riuniti. FIORI, G. 2003. Vita di Antonio Gramsci. Nuoro: Ilisso. GENTILE, G. 1937 [1912]. L’atto del pensare como atto puro. Firenze: Sansoni. GRAMSCI, A. 1973. Scritti politici. Roma: Riuniti. ______. 1975. Quaderni del carcere. Torino: Giulio Einaudi. 4 v. ______. 1980. Cronache torinesi: 1913-1917. Torino: Giulio Einaudi. ______. 1982. La città futura: 1917-1918. Torino: Giulio Einaudi. ______. 1984. Il nostro Marx. Torino: Giulio Einaudi. ______. 2009. Epistolario I: gennaio 1906 – dicembre 1922. Roma: Istituto della Enciclopedia Italiana. ALVARO BIANCHI Resumo: Neste texto recuperam-se os artigos jornalísticos de Antonio Gramsci publicados entre 1915 e 1917, em que é abordado o tema da revolução. Para isso, reconstroem-se as análises gramscianas em perspectiva histórica, retomando a Revolução Francesa como paradigma e as mudanças de ênfase no contexto de outra revolução em curso, na Rússia. Evidencia-se o percurso analítico por meio do qual Gramsci modifica sua forma de pensar o problema cultural e político da revolução, em particular a reflexão sobre o uso da força militar com fins revolucionários. Conclui-se que Gramsci, neste período, inicia o desenvolvimento de um pensamento original sobre cultura e política, abrindo caminho para o que seria mais tarde sua elaboração sobre o conceito de hegemonia. Palavras-chave: Antonio Gramsci; Revolução Russa; Cultura; Política. ANTONIO GRAMSCI, A CULTURA SOCIALISTA E A REVOLUÇÃO RUSSA DANIELA XAVIER HAJ MUSSI della Enciclopedia Italiana. _. 2010. Cronache teatrali: 1915-1920. Torino: Nino Aragno. IL NUOVO colpo di Stato in Russia. 1917. La Stampa, Torino, p. 1, 10 nov. 1917. KERENSKI rassegna le dimissioni per l’impossibilità di ricostituire il Governo. 1917. La Stampa, Torino, p. 4, 5 ago. 1917. MARX, K. 2013. O capital: crítica da economia política. São Paulo: Boitempo. (Livro I: O processo de produção do capital). MISSIROLI, M. 1914. La monarchia socialista: estrema destra. Bari: Laterza. 298 PAGGI, L. 1970. Nella crisi del socialismo italiano. Roma: Riuniti. (Série Gramsci e il moderno principe, v. 1). RAPONE, L. 2011. Cinque anni che paiono secoli: Antonio Gramsci da socialismo al comunismo: 1914-1919. Roma: Carocci. ROLLAND, R. 1953. L’esprit libre. Paris: Albin Michel. SALVEMINI, G. 1905. La rivoluzione francese: 1788-1792. Milano: Signorelli & Palestrini. SOREL, G. 2013 [1908]. Réflexions sur la violence. Genève; Paris: Entremonde. SPRIANO, P. 1972. Storia di Torino operaia e socialista: da De Amicis a Gramsci. Torino: Giulio Einaudi. TERRACINI, U. 1976. Ricordi e riflessioni di un rivoluzionario professionale. Belfagor: Rassegna di varia umanità, v. 31, n. 3, pp. 249-266. TREVES, C.; TURATI, F. 1917. Proletariato e resistenza. Critica Sociale, v. 27, n. 21, pp. 266-267. TURATI, F. 1917. Per il nuovo regime in Russia. Atti parlamentari, Legislatura XXIV, 1ª sessione, Discussione della Camera dei Deputati, Tornata del 23 mar. 1917, pp. 13375-13377. Tornata del 23 mar. 1917, pp. 13375-13377. Lua Nova, São Paulo, 102: 267-298, 2017 Resumos \| Abstracts ANTONIO GRAMSCI, SOCIALIST CULTURE AND RUSSIAN REVOLUTION Abstract: This article retrieves Antonio Gramsci’s newspaper articles published between 1915 and 1917 where the theme of the revolution appears. To do so, it reconstructs the Gramscian analysis in historical perspective, retaking the French Revolution as a paradigm and the changes of emphasis in the context of a new revolution under way, now in Russia. It shows the analytical course through which Gramsci modifies his way of thinking the cultural and political problem of the revolution, in particular the reflection on the use of military force for revolutionary purposes. It concludes Resumos \| Abstracts that Gramsci, in this period, begins the development of an original thought about culture and politics, opening the way to what would later be his elaboration on the concept of hegemony. Keywords: Antonio Gramsci; Russian Revolution; Culture; Politics. Recebido: 01/05/2017 Aprovado: 04/09/2017 Resumos \| Abstracts Resumos \| Abstracts Resumos \| Abstracts that Gramsci, in this period, begins the development of an original thought about culture and politics, opening the way to what would later be his elaboration on the concept of hegemony. Keywords: Antonio Gramsci; Russian Revolution; Culture; Politics. Recebido: 01/05/2017 Aprovado: 04/09/2017
https://openalex.org/W4286233419	https://www.papersinphysics.org/papersinphysics/article/download/837/447	Lithuanian	null	When you can't count, sample! Computable entropies beyond equilibrium from basin volumes	arXiv (Cornell University)	2,022	cc-by	43,910	Papers in Physics, vol. 15, art. 150001 (2023) Papers in Physics, vol. 15, art. 150001 (2023) www.papersinphysics.org ISSN 1852-4249 Received: 15 July 2022, Accepted: 25 January 2023 Edited by: K. Daniels, L. A. Pugnaloni, J. Zhao Licence: Creative Commons Attribution 4.0 DOI: https://doi.org/10.4279/PIP.150001 3 Courant Institute of Mathematical Sciences, New York University, New York 10003, USA Mathias Casiulis1,2∗, Stefano Martiniani1–3† Mathias Casiulis1,2∗, Stefano Martiniani1–3† In statistical mechanics, measuring the number of available states and their probabilities, and thus the system’s entropy, enables the prediction of the macroscopic properties of a physical system at equilibrium. This predictive capacity hinges on the knowledge of the a priori probabilities of observing the states of the system, given by the Boltzmann distribution. Unfortunately, the successes of equilibrium statistical mechanics are hard to replicate out of equilibrium, where the a priori probabilities of observing states are, in general, not known, precluding the na¨ıve application of common tools. In the last decade, exciting developments have occurred that enable direct numerical estimation of the entropy and density of states of athermal and non-equilibrium systems, thanks to significant methodological advances in the computation of the volume of high-dimensional basins of attraction. Here, we provide a detailed account of these methods, underscoring the challenges present in such estimations, recent progress on the matter, and promising directions for future work. 2 Simons Center for Computational Physical Chemistry, Department of Chemistry, New York University, New York 10003, USA 1 Center for Soft Matter Research, Department of Physics, New York University, New York 10003, USA ∗mc9287@nyu.edu † sm7683@nyu.edu ∗mc9287@nyu.edu † sm7683@nyu.edu 1 Center for Soft Matter Research, Department of Physics, New York University, New York 10003, USA 2 Simons Center for Computational Physical Chemistry, Department of Chemistry, New York University, New York 10003, USA 3 Courant Institute of Mathematical Sciences, New York University, New York 10003, USA I Introduction tion for the microscopic states of the system. As- suming that the system being considered is in ther- mal equilibrium, the probability of observing any of its microscopic configurations is given by the Boltz- mann distribution, pC = Z−1 exp(−βEC), where Z = P C exp(−βEC) is the partition function, EC is the energy of the configuration and β is an inverse temperature imposed by a thermostat. Knowledge of the partition function affords us the ability to compute the free energy, F = −β−1 ln Z, which de- termines the thermodynamic stability of the state of the system, and allows us to derive all thermo- dynamic observables from its derivatives. From steam engines to LCD displays and polymer materials, the field of thermodynamics has played a pivotal role in the development of modern tech- nology. The key to this success has been the ef- ficacy with which equilibrium statistical mechan- ics predicts the macroscopic properties of physi- cal systems from knowledge of microscopic inter- actions alone. This predictive power hinges on the knowledge of the underlying probability distribu- An alternative, but completely equivalent [115], approach is to compute the entropy of the system, 1 Center for Soft Matter Research, Department of Physics, New York University, New York 10003, USA SG = − X C pC log pC, (1) (1) and deduce thermodynamics from its maximisa- tion. This definition of entropy was first introduced by Gibbs [116] and, when the states of the system 150001-1 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani occur with equal probability (i.e., pC = 1/Ωwith Ωthe volume of the region of accessible states), it reduces to the well-known Boltzmann entropy, SB = log Ω+ const. [117]. The entropy also allows us to predict the direction of spontaneous thermo- dynamic transformations, as the entropy of an iso- lated system can only increase with time: put sim- ply, heat flows from hot to cold [1,118]. generic dynamical system. To each of these struc- tures, we can associate a basin of attraction, i.e., the set of all initial conditions leading to the same structure via the dynamics. See Fig. 1 for an illus- tration. In this perspective we show how we can make use of these facts to arrive at a general protocol for es- timating the entropy of systems out of equilibrium, from granular to generic dynamical systems. I Introduction The key observation is that while Shannon’s entropy, Eq. 2, provides a universal definition of entropy, we do not know the a priori probabilities, pi, of observing a given state of the system, suggesting that its evaluation may not be possible. To the contrary, we show that the problem of evaluating these probabilities, pi, or alternatively the prob- lem of enumerating the number of possible states, Ω(e.g., if one wishes to compute a Boltzmann-like entropy, SB = log Ω+ const.), can be reduced to a tractable sampling problem. Indeed, as discussed at length in Sec. II, one can compute the a priori probability of observing a given stable structure by measuring the volume of its associated basin of at- traction [119] – and it is therefore possible to esti- mate the Shannon entropy for an arbitrary system. The state of affairs is not as simple far from equi- librium. Let us consider an arbitrary system evolv- ing according to some well-defined, but in general non-equilibrium, dynamics. In such a system, there is no guarantee that objects like temperature or energy can be defined, and there is generally no clear prescription for the probability distribution of microscopic configurations. Is it still possible to find analogies with equilibrium statistical mechan- ics? To address this question, we must introduce two theoretical notions. First, entropy can be universally defined as the average amount of surprise (or uncertainty) in- herent to a random variable (rare observations are more surprising, or informative, than common ones) [120]. Mathematically, Shannon [2] showed that this amounts exactly to a rewriting of the Gibbs entropy (1), typically referred to as the Shan- non entropy, In the following, we start by presenting the basin- sampling approach for entropy measurements in Sec. II. Then, in Sec. III, we highlight its recent successes in the context of granular packings. Fi- nally, in Sec. IV, we discuss exciting extensions of this idea that we believe shall shape the future of computational non-equilibrium physics. SS = − Ω X i=1 pi log pi, (2) (2) where pi is the probability of observing state i and the sum is over all accessible states, this time for an arbitrary statistical system with a generic (station- ary) probability distribution. This information- theoretic interpretation of entropy can be used to define, measure, and interpret entropies in any physical system, whatever its dynamics might be. I Introduction This is why there has been recent interest in direct measurements of this quantity using various algo- rithms, in no way restricted to the one we present here: for instance, the Shannon entropy of physi- cal systems can be estimated using ideas from data compression [3–7]. metic mean we have that metic mean we have that V = Ω X i=1 vi = Ω⟨v⟩, (3) (3) where vi is the volume of basin i in configuration space, and ⟨·⟩is the mean taken over all basins. As a result, one can write the number of basins as Ω= V/⟨v⟩. This seemingly innocent equation is in fact crucial, as it turns the intractable enu- meration problem of counting Ωinto a sampling problem, namely the computation of the average basin volume ⟨v⟩, which can be performed over a finite set of basins. Figure 1: Illustration of basins of attraction. Left: Ex- ample of a of 2d energy landscape. High-energy points are shown in blue, low-energy points in red. Right: In the same landscape, trajectories of steepest descent and ascent ˙x = ±∇U(x) are spawned from random points, and confined to an arbitrary region (blue circle). Tra- jectories converging to the same minimum belong to the same basin of attraction and are plotted in the same color. There are two main challenges in computing this quantity in the context of many-body sys- tems: on the one hand, high-dimensional geom- etry makes volume estimations difficult and, on the other hand, mean volume estimations by di- rect sampling are typically biased. We start by considering the challenges of estimating volumes in high dimensions, and how we can overcome them by means of suitably-modified free energy calcula- tions. inherent structures observed at finite temperature (viz., the configurations that the system relaxes to by steepest descent from finite temperature sam- ples), this quantity becomes the system’s configu- rational entropy (distinct from the vibrational con- tributions to the total entropy) [68]. By analogy, in any dynamical system, one can always define such an entropy, and compute it by counting basins of attraction. i Turning intractable counting into sam- pling While the probabilities of observing the states of a nonequilibrium system are in general not known a priori, it is a generic feature of dynamical sys- tems that for a given initial condition they will end up trapped in small regions of their state space, be they actual basins of a high-dimensional energy landscape, or more general dynamical at- tractors. Assuming that there are Ωsuch basins, one can readily compute a Boltzmann-like entropy, SB = log Ω+ const. In a thermodynamic system, it simply amounts to the entropy at zero temper- ature. If we restrict ourselves to a set of typical Second, the states of interest of a system can often be described by the stable structures of its dynamics, be they extrema of a high-dimensional function that can be reached by steepest descent (e.g., energy minima of a potential energy land- scape), or the fixed points and limit cycles of a 150001-2 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani Figure 1: Illustration of basins of attraction. Left: Ex- ample of a of 2d energy landscape. High-energy points are shown in blue, low-energy points in red. Right: In the same landscape, trajectories of steepest descent and ascent ˙x = ±∇U(x) are spawned from random points, and confined to an arbitrary region (blue circle). Tra- jectories converging to the same minimum belong to the same basin of attraction and are plotted in the same color. ii A starter in high-dimensional geometry: there is plenty of room in the corners The computation of high-dimensional basin vol- umes is a difficult task that cannot be accomplished by simple quadrature [8]. To illustrate this diffi- culty, we take the example of a simple shape with a volume that is known in any dimension: the hy- percube. In the top line of Fig. 2, we present 2d sketches of the aspect of hypercubes, r ∈[−a; a]d, in dimensions d = 2, 3, 4, and 5. In each case, following recommendations on drawings of high- dimensional convex volumes [127], we represent the d−dimensional hypercube with half-sidelength a = 1 by linking each of the 2d summits, which sit at a distance √ da from the center, to 2 of its neighbors using hyperbolas tangent to the largest hypersphere contained by the hypercube – namely the unit hypersphere. The reason for choosing hy- perbolas is that if one splits the cube into two ra- dial regions of equal volumes, Rin = r : r < R1/2 and Rout = r : r > R1/2 , where r is the distance from the center and R1/2 a threshold distance, the (d −1)-dimensional cross-section of each diagonal of the hypercube decays exponentially with r in Counting the number of basins of attractions is tantamount to enumerating all microstates in the microcanonical ensemble of statistical mechanics: while it is in principle a valid strategy, it is in practice numerically intractable. Until recently, this issue was believed to be impossible to over- come [8–10]. The problem has since been solved by the introduction of Monte Carlo methods capa- ble of integrating the volume of individual basins of attraction using ideas originally introduced for the calculation of the free energy of solids in sta- tistical mechanics [11]. Thanks to this new class of methods, pioneered by Frenkel, Xu [12], Asenjo [13], and Martiniani [14–17], alongside collabora- tors, it is now possible to approach these problems and measure quantities that could never have been computed with previous techniques. The gist of the method is as follows: since the accessible volume, V, of configuration space is tiled by Ωbasins of attraction, by definition of the arith- 150001-3 Papers in Physics, vol. 15, art. ii A starter in high-dimensional geometry: there is plenty of room in the corners The inside of each shape is colored with a density plot of the relative contribution w(r) of each radial shell inside the cube to the full volume, as obtained from Monte Carlo integration using 106 points uniformly drawn inside each cube. Bottom: Histograms of relative weights of radial shells w(r) for the hypercube (orange) and for the hyperball (purple). The lines highlight how large the intersection of the cube and a hyperspherical shell is at the distance that contains most of the cube’s volume. As reminded in the insets, this intersection here represents 1, 1, 0.78, and 0.62 times the surface of the full hyperspherical shell, respectively. Rout [127]. In other words, in large d, one corner of a hypercube has a vanishing volume over surface ratio, a property shared with hyperbolic objects, like Gabriel’s horn (also called Torricelli’s trumpet) in 3d [129]. To complete this picture, in the second row of Fig. 2, we plot in orange the distribution of mass of hypercubes with a = 1 along the radial direc- tion, w(r), and in purple the corresponding line for the hyperball, which is simply the surface area Sd of the (d −1)-sphere with radius √ da, normalised like w(r), i.e. by the volume of the cube, (2a)d. As the dimensionality of space increases, the volume of the hypercube concentrates deeper and deeper into disconnected tendril-like objects, at distances such that the intersection between the hypercube and the hypersphere becomes very small: in Fig. 2, as highlighted by lines and inset texts, the ratio between the value of the orange curve at its maxi- mum and the value of the purple curve at the same distance decreases with d. Furthermore, the con- tribution of the unit ball to the total volume of This is an illustration of the fact that, in high dimensions, even simple compact objects present tendril-, or tentacle-like regions [10, 15, 16, 18, 119] that extend very far but get very thin. Yet, some- what counter-intuitively, most of the volume of high-dimensional objects is contained in these ex- tended objects! This is shown in Fig. 2 by the den- sity plot inside each shape, that represents the ac- tual weight of each hyperspherical shell within each hypercube. ii A starter in high-dimensional geometry: there is plenty of room in the corners 150001 (2023) / M Casiulis & S Martiniani 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.5 1.0 1.5 2.0 2.5 r w(r) Sd(r)/2d d = 2 1 0.5 1.0 1.5 0.0 0.5 1.0 1.5 2.0 2.5 r w(r) Sd(r)/2d d = 3 1 0.5 1.0 1.5 0.0 0.5 1.0 1.5 2.0 2.5 r w(r) Sd(r)/2d d = 4 0.8 0.2 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 2.5 r w(r) Sd(r)/2d d = 5 0.6 0.4 Figure 2: High-dimensional volumes. Top: 2d sketches of hypercubes in d = 2, 3, 4, and 5. The outer black line links vertices located at a distance √ d from the center of the cube to midpoints located on the unit hypersphere with hyperbolic lines. The inside of each shape is colored with a density plot of the relative contribution w(r) of each radial shell inside the cube to the full volume, as obtained from Monte Carlo integration using 106 points uniformly drawn inside each cube. Bottom: Histograms of relative weights of radial shells w(r) for the hypercube (orange) and for the hyperball (purple). The lines highlight how large the intersection of the cube and a hyperspherical shell is at the distance that contains most of the cube’s volume. As reminded in the insets, this intersection here represents 1, 1, 0.78, and 0.62 times the surface of the full hyperspherical shell, respectively. 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.5 1.0 1.5 2.0 2.5 r w(r) Sd(r)/2d d = 2 1 0.5 1.0 1.5 0.0 0.5 1.0 1.5 2.0 2.5 r w(r) Sd(r)/2d d = 3 1 0.5 1.0 1.5 0.0 0.5 1.0 1.5 2.0 2.5 r w(r) Sd(r)/2d d = 4 0.8 0.2 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 2.5 r w(r) Sd(r)/2d d = 5 0.6 0.4 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.5 1.0 1.5 2.0 2.5 r w(r) Sd(r)/2d d = 2 1 Sd(r)/2d r r Figure 2: High-dimensional volumes. Top: 2d sketches of hypercubes in d = 2, 3, 4, and 5. The outer black line links vertices located at a distance √ d from the center of the cube to midpoints located on the unit hypersphere with hyperbolic lines. The squared distance of this point from the center, r2 = P x2 i is a sum of independent, identically drawn uniform random variables so that, per the central limit theorem, its distribution tends to a Gaussian with mean value µ = d⟨x2⟩= d/3, which asymptotically yields E[ √ r2] ∼ p d/3. 1This can be shown by a simple statistical argument: suppose that we uniformly draw a random vector within a d−dimensional hypercube, with coordinates (x1, . . . , xd). ii A starter in high-dimensional geometry: there is plenty of room in the corners While in d = 2, the shell that con- tributes the most to the volume is the unit circle, as the dimensionality increases, it shifts to larger and larger radial distances, at Rmax ∼ p d/3 in the limit of large d.1 150001-4 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 R/d 1/2 V(R)/V d 2 4 8 16 32 64 R = d / 3 Figure 3: Na¨ıve random sampling shortcomings. Direct Monte Carlo evaluation ˆV (R) of the volume of a hy- percube with actual volume V , using Ns = 105 points uniformly drawn from a hyperball with radius R, for di- mensions going from 2 to 64. Here the half-sidelength of the hypercube is a = 1. The dashed line indicates the asymptotic mode of the mass distribution of the cube, p d/3. 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 R/d 1/2 V(R)/V d 2 4 8 16 32 64 R = d / 3 the hypercube vanishes as the dimension increases. These last two properties make practical volume estimations extremely complicated: a na¨ıve Monte Carlo integration using uniformly sampled points in a hyperball is bound to fail since the vast majority of points will fall outside of the shape of interest, while an integration within a ball close to the center will yield only a tiny fraction of the overall volume, an example of the so-called curse of dimensional- ity [10,16,18]. y [ , , ] To more concretely illustrate how na¨ıve random sampling methods fail at estimating volumes in high dimensions, let us consider the volume estima- tion problem for a simple hypercube, using Monte Carlo integration. We sample Ns points uniformly drawn in a hyperball, B(R), with radius R cen- tered at the same point as the hypercube, measure the fraction fMC of points that fall within the hy- percube, and compute the corresponding volume, ˆV (R) = fMCVB(R). In general, one does not nec- essarily know the largest linear size of the object of interest, so R should, in principle, be varied. ii A starter in high-dimensional geometry: there is plenty of room in the corners We use this strategy to estimate the volume of hy- percubes with a = 1 by varying the ball radius, R, between 0 and the length of the longest diago- nal of the cube, √ d. In Fig. 3, we plot the mea- sured volume divided by the true volume of the hypercube, V = 2d, against R/ √ d, for d between 2 (mauve) and 64 (red), for Ns = 105. At low di- mensions of space, this method converges smoothly to VMC = V as R → √ d. As d increases, the vol- ume concentrates more and more around the mode of the distribution of mass of the cube, p d/3, and as a result the curves become step-like. However, as d increases, the measurement also becomes less reliable upon approaching R → √ d: it first dis- plays increasingly large fluctuations (d ≤16) then violently falls to 0 (d ≥32). The reason is that most points in the ball are actually sampled at radii such that the hypercube is already made of narrow spikes, as illustrated in Fig. 2. Indeed, the ratio of the volume of the hypercube to that of the hy- perball with radius √ d decays exponentially with dimension, V/VB( √ d) ∼ √ d exp(−d). Figure 3: Na¨ıve random sampling shortcomings. Direct Monte Carlo evaluation ˆV (R) of the volume of a hy- percube with actual volume V , using Ns = 105 points uniformly drawn from a hyperball with radius R, for di- mensions going from 2 to 64. Here the half-sidelength of the hypercube is a = 1. The dashed line indicates the asymptotic mode of the mass distribution of the cube, p d/3. simple example here, in d = 64, the best estimate with our choice of Ns would be roughly 50% off! This shows that simple Monte Carlo integration, while efficient in small dimensions, fails at captur- ing volumes in high dimensions. Note that the case illustrated here is actually fairly ideal: we know the location of the center of the hypercube, which makes the measurement easier, and the hypercube is a rather regular object. In real-world situations, a Monte Carlo integration of high-dimensional vol- umes is likely to be far worse. iii Free energy methods for volume compu- tations Written in this form, the volume can be interpreted as the parti- tion function of a free Brownian walker exploring the domain Γ, with a hard wall at the boundary, at an inverse temperature β. a control parameter that allows us to go continu- ously from an integral of unknown volume to one of known volume. For instance, one can tether ran- dom walkers to a reference point, r0, inside the basin of attraction (e.g., the minimum energy con- figuration) using harmonic springs of varying stiff- ness k, see Fig. 4. For a random walk constrained to remain within the domain of interest Γ, one can compute a volume vk weighted by the Boltzmann factor (viz., the corresponding partition function) as Z The most popular and earliest class of meth- ods for the computation of partition functions is based on thermodynamic integration (TI) [21, 22, 121], which consists of parameterizing the system’s Hamiltonian in such a way that we can “morph” an unknown partition function into one that we know how to compute analytically. In practice, this amounts to replacing a high-dimensional integral over phase space volume with a low-dimensional integral over one or more Hamiltonian parame- ters (each point in the integrand is obtained from an equilibrium simulation with a given choice of Hamiltonian parameters), or to estimating ratios of partition functions from equilibrium samples ob- tained from simulations with different Hamiltonian parameters [23,24], as we show presently. as vk = Z Γ dre−1 2 k\|r−r0\|2 (5) (5) and define the dimensionless basin free energy as the negative log-volume fk = −log vk. When k = 0, the walker is completely free to explore the basin volume, while for k →∞the walk is reduced to a small region surrounding r0 that fits entirely within the basin. Computing the basin volume amounts to measuring the dimensionless free energy difference between the walkers with k = 0 and k →∞so that fk=0 = fk→∞+ ( ˆfk=0 −ˆfk→∞) (6) (6) where we added the analytical reference free en- ergy fk→∞to the numerical estimate (denoted by a “hat”) of the free energy difference between k = 0 and k →∞. This is necessary because free ener- gies can only be computed numerically up to an additive constant equal for all k’s. iii Free energy methods for volume compu- tations It has been shown that volume estimations and enu- meration problems can be treated in a way that amounts to a free energy calculation in the spirit of the Frenkel-Ladd method [11]. The idea is that the volume v of a domain Γ can be rewritten as If we did not know that we were measuring a sim- ple cube, and ball-picked from spheres of increas- ing radius to produce the same kind of curve as in Fig. 3, in high dimension our estimate would be far off as the mode of the mass distribution of the cube moves further and further into the corners. In the v = Z Rd drOΓ(r) = Z Rd dre−βUΓ(r), (4) (4) 150001-5 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani r0 k1=0 r0 k2>0 r0 k3>k2 r0 reweighted Figure 4: Probing basin volumes with random walks. Different random walkers are linked by springs with different rigidities to the configuration at the minimum. The first three panels represent the probability density of the trajectories of such walkers, with rigidities growing from left to right. At very small (possibly negative) rigidities, the walker explores the outer rim of the basins, following tentacle-like structures, while at large rigidities the random walk typically explores a hypersphere around the minimum. Putting together and properly reweighing the information sampled by these walkers, one can faithfully recover the full volume as if it had been uniformly sampled (last panel). r0 k1=0 r0 k2>0 r0 k3>k2 r0 reweighted r0 k1=0 r0 k2>0 r0 k3>k2 r0 reweighted Figure 4: Probing basin volumes with random walks. Different random walkers are linked by springs with different rigidities to the configuration at the minimum. The first three panels represent the probability density of the trajectories of such walkers, with rigidities growing from left to right. At very small (possibly negative) rigidities, the walker explores the outer rim of the basins, following tentacle-like structures, while at large rigidities the random walk typically explores a hypersphere around the minimum. Putting together and properly reweighing the information sampled by these walkers, one can faithfully recover the full volume as if it had been uniformly sampled (last panel). where OΓ is the characteristic function of the do- main to integrate over, or “oracle”, which can be exponentiated to yield the potential UΓ which is 0 inside the domain and ∞outside it. iii Free energy methods for volume compu- tations The analytical In the context of volume estimations, the poten- tial UΓ that encodes the oracle is athermal, so that we can take β = 1 without loss of generality. Then, in the spirit of umbrella sampling [25,121], we can introduce simple biasing potentials that depend on 150001-6 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani values of k that are constrained to “live” closer to the hyperspherical core, as illustrated in Fig. 4. reference can be computed by a Gaussian integral at the largest stiffness value, kmax, provided that kmax is large enough that the corresponding ran- dom walk is essentially unaffected by the boundary of the domain. The dimensionless free energy difference in Eq. 6 can be computed by a simple thermodynamic in- tegration over k [121] or using more sophisticated estimators like the Multi-Bennet Acceptance Ra- tio Method (MBAR) [19], which should yield more statistically accurate results. The MBAR estima- tor hinges on the idea that one can always relate the free energy at one value of k with the free energy at every other value of k, via Intuitively, this importance sampling method should outperform brute-force Monte Carlo sam- pling because the steps of the walks are chosen to remain close to one another, so that even within the high-dimensional tentacles, the rejection rate of the steps will be much smaller than what we would get by ball-picking within a sphere that contains the shape of interest. In other words, this method samples points compactly within the volume being estimated, rather than throwing darts at random in a huge volume around it. ˆfk = −ln   K X m=1 Nm X a=1 exp[−βUk(ra)] P k′ Nk′ exp[ ˆfk′ −βUk′(ra)]  , (7) ˆfk = −ln   K X m=1 Nm X a=1 exp[−βUk(ra)] P k′ Nk′ exp[ ˆfk′ −βUk′(ra)]  , (7)  (7) Nevertheless, running independent biased ran- dom walks remains a poor strategy in high dimen- sions. Take for instance the example of the hyper- cube: Due to the rd−1 scaling for the surface of a hypersphere, most points of the walk will be con- centrated far away from the center of the cube, and instead will live close to the heaviest shell of the mass distribution at p d/3. iii Free energy methods for volume compu- tations In the context of basin volume calculations, we exchange coordinates between random walks with different stiffness, k, so that walkers at low stiffness can escape the tenta- cles by swapping coordinates with walkers at high ln h(r) = X k wk(r) h ln hk(r) + βUk(r) −( ˆfk −ˆf0) i , (8) (8) ( ) with wk(r) = hk(r)/ P k′ h′ k(r) being a set of nor- malised weights. with wk(r) = hk(r)/ P k′ h′ k(r) being a set of nor- malised weights. iii Free energy methods for volume compu- tations In high dimensions, this maximum lies far into the corners, so that a single random walk would typically spend very long times in one of the 2d corners. As a result, attempt- ing a direct estimation of v with independent free random walks would typically require either expo- nentially many (2d) realizations, or exponentially long equilibration times so that the walk can es- cape a given corner, reach the (typically tiny) con- vex core, and explore another corner, 2d times. where Uk(r) = −k\|r−r0\|2/2 is the biasing potential with spring stiffness k (but can in principle assume any shape), K is the total number of biased ran- dom walks, and Nm is the number of uncorrelated equilibrium samples obtained from the m-th ran- dom walk. This system of implicit equations can be solved numerically, typically using a self-consistent Newton-Raphson scheme [19]. Ideally, for the qual- ity of the MBAR solution to be as good as possible, one should measure a similar amount of equilibrium samples in every region of the volume being mea- sured. This problem boils down to the choice of Uk and we discuss it briefly in the next subsection. Finally, MBAR yields the optimal reweighing of the set of histograms hk(r), where r ≡\|r −r0\|, allowing for the reconstruction of the full density of states, which, in the context of volume mea- surements, amounts to the mass distribution, h(r), within the object of interest, that verifies To overcome this problem, one can take inspi- ration in the large body of work on free energy estimations at low temperature in rugged energy landscapes (see for instance [121, 122]). A typical strategy is to use parallel tempering [15, 23, 24], which amounts to running a collection of simula- tions with different control parameters (typically different temperatures), each called a “replica”, and allowing for configuration exchanges between high and low temperature replicas so that the low temperature replicas do not become trapped in a lo- cal region of the energy landscape, all while respect- ing the detailed balance condition. iv Measuring hypercubes To illustrate how effective this technique is com- pared to brute-force Monte Carlo, we now apply it to the measurement of high-dimensional hyper- cubes, r ∈[−a; a]d. For each dimensionality, d, we run a collection of Kd random walks – that we re- fer to as replicas – associated to spring constants 150001-7 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani (k1 = 0, k2, . . . , kKd), for Ns = 5 × 105 steps, at- tempting a parallel tempering coordinate exchange between replicas with neighboring values of k every 10 steps. The number of replicas at a given dimen- sionality, Kd, is set so that the histogram of sam- pled distances from the center, hki(r), overlaps sig- nificantly with the histogram of replicas with neigh- boring values of k, namely h(r)ki±1. Finally, to get absolute free energy values, we use the negative log-volume of the largest ball inscribed in the hypercube as the known reference in Eq. 6, instead of the free energy for a tight harmonic trap, so that the free energy becomes Finally, to get absolute free energy values, we use the negative log-volume of the largest ball inscribed in the hypercube as the known reference in Eq. 6, instead of the free energy for a tight harmonic trap, so that the free energy becomes fk=0 = fB(a) + ( ˆfk=0 −ˆfB(a)) (9) (9) where ˆfB(a) is the estimated free energy for a set of points sampled directly from the largest inscribed ball, B(a), whose volume is known analytically. The black line overlayed onto the histograms in Fig. 5 is the reconstructed mass distribution, or density of states, of the cube obtained using MBAR according to Eq. 8. Notice that, as expected, the region of large r leading to √ da = 5 contains very little volume. ± The scaling of the number of required replicas with dimensionality, d, can be estimated from the mean and variance of the radial distribution for a d- dimensional Gaussian, hk(r) ∼rd−1 exp (−kr2/2). We start by noting that for parallel tempering to be efficient, and for the MBAR estimation to be reliable, these distributions must display signifi- cant overlap. Choosing the replicas so that their modes are separated by the standard deviation of the narrowest walk, one can show that the required number of replicas, Kd, scales linearly with d (see App. iv Measuring hypercubes A). Recall, for context, that na¨ıve Monte Carlo sampling typically requires a number of sam- ples that grows exponentially with d. 1 2 3 4 5 2 4 6 8 10 r h(r) d=100 Figure 5: Biased walks in a hypercube. Each colored line is a histogram hk(r) of distances to the center for a random walk constrained to remain in a 100d hyper- cube with unit sidelength, a = 0.5, and subjected to a biasing potential Uk(r) = kr2/2. Color encodes the value of the rigidity k, from large (red) to small (pur- ple). Here, the K = 64 spring constants are chosen so that the modes of the distribution are equally spaced in r, from r = a = 0.5 (red) to r ≈3.1 > p d/3a (pur- ple). A 1d gaussian is sampled to obtain points near the center (gray line) to improve the MBAR solution. The reconstructed density of states of the cube is shown in black. Note that the 16 histograms to the right of the black line correspond to negative values of k. 1 2 3 4 5 2 4 6 8 10 r h(r) d=100 We show the output histograms, hk(r), for a practical implementation of these random walks in a d = 100-dimensional hypercube in Fig. 5. In this example, we used K = 64 replicas with k’s chosen so that the modes of the distributions are linearly spaced in r. Of these 64 k values, 48 were posi- tive, leading to k = 0, and 16 were negative (the most negative being k = −4.5), so that the bias- ing potential pushes these walkers further into the corners of the cube. As expected, high values of k (red histograms) yield narrower distributions closer to the origin, while lower ones (towards purple) are broader and distant from the origin. We chose kmax (the left-most curve in Fig. 5) so that its mode would coincide with that of the largest ball inscribed in the cube. Note how this leaves a significant gap in the histograms in the regions nearest to the origin due to the power- law rd−1 in hk (yet another manifestation of the curse of dimensionality). iv Measuring hypercubes In order to improve the MBAR free energy estimation, and the density of states reconstruction, we chose to sample indepen- dent points from hin(r) ∝r1−de−kinr2/2 , also con- strained to the inside of the cube, to accumulate samples near the origin (gray line in Fig. 5). The corresponding biasing potential is Uin = −(d − 1) ln r + kinr2/2, where we choose kin = 4 so that the distribution has width σ ∼1/√kin = 0.5. Figure 5: Biased walks in a hypercube. Each colored line is a histogram hk(r) of distances to the center for a random walk constrained to remain in a 100d hyper- cube with unit sidelength, a = 0.5, and subjected to a biasing potential Uk(r) = kr2/2. Color encodes the value of the rigidity k, from large (red) to small (pur- ple). Here, the K = 64 spring constants are chosen so that the modes of the distribution are equally spaced in r, from r = a = 0.5 (red) to r ≈3.1 > p d/3a (pur- ple). A 1d gaussian is sampled to obtain points near the center (gray line) to improve the MBAR solution. The reconstructed density of states of the cube is shown in black. Note that the 16 histograms to the right of the black line correspond to negative values of k. 150001-8 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani □ □□□□□□□□ □ □□□□□□□□ □ □□ □□ □ □□ □ 5 10 50 100 500 1000 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 d V/V MBAR MC Figure 6: Volume estimates. Ratio of measured volume ˆV to expected volume V of the unit cube against the dimensionality d, using brute-force Monte Carlo with Ns = 108 points in the smallest ball containing the cube (gray disks) and MBAR (red squares) with K = max(64, d/5) random walks linearly spaced in r, each with Ns = 5 × 105 proposed steps. Error bars are 95% confidence intervals for the mean, using statistics across 10 runs. □ □□□□□□□□ □ □□□□□□□□ □ □□ □□ □ □□ □ 5 10 50 100 500 1000 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 d V/V MBAR MC We use the same procedure when varying the di- mensionality of the cube. iv Measuring hypercubes The result for the ra- tio ˆV /V between the measured volume and the real volume of the cube, obtained using the fastM- BAR [20] implementations of MBAR, is shown in Fig. 6 as a function of the dimensionality (red squares). The results are compared with a brute- force Monte Carlo method using the same order of magnitude of total number of samples, Ns = 108 (gray disks). Notice that the x axis is in log scale. (g y ) g In spite of the finite number of samples, and of the rather crude choice of Kd, the error of the MBAR measurement grows roughly linearly with d and remains below 10% even as d reaches 500, while the Monte Carlo estimate essentially measures zero volume when dimensionality reaches d = 30, the point at which the ratio of the volume of the cube to that of the smallest circumscribed ball is roughly 1/Ns. Notice that in d = 500, that ratio is of the order of 10−156, meaning that na¨ıve Monte Carlo would be absolutely unfeasible. Here, using our ap- proach instead of na¨ıve MC sampling, with compa- rable numbers of samples, increases the maximum dimension for which a volume can be measured, albeit with some statistical error, from 30 to thou- sands of dimensions. Since the number of samples used per walk in this example is still rather modest (5 × 105), the statistical error can be reduced by making the random walks longer. d Figure 6: Volume estimates. Ratio of measured volume ˆV to expected volume V of the unit cube against the dimensionality d, using brute-force Monte Carlo with Ns = 108 points in the smallest ball containing the cube (gray disks) and MBAR (red squares) with K = max(64, d/5) random walks linearly spaced in r, each with Ns = 5 × 105 proposed steps. Error bars are 95% confidence intervals for the mean, using statistics across 10 runs. the shell of the cube that contributes most to its mass. In d = 1000, we find that this number is of the order of 1076. Even if one could somehow guess the radius of the shell that contributes most to a high-dimensional volume, direct Monte Carlo sampling of that shell alone is in general impossi- ble. iv Measuring hypercubes 6, in d = 2, 10, 50, 100, 200, 500, 1000 dimensions (b) Decimal log of the ratio between the mass distribu- tion and that of the smallest ball containing the whole cube for d = 1000. The dashed red lines indicates the location p d/3a of the maximum for d = 1000, and the value of the ratio it corresponds to in (b). We indicate the probability pMC of landing a single point inside the cube when drawing points uniformly from that specific hyperspherical shell 0 2 4 6 8 10 0 1 2 3 4 r h(r) (a) 2d 10d 50d 100d 200d 500d 1000d likely to be a worst-case scenario, as it assumes equally-spaced distributions in r that are all as narrow as the narrowest one, hkmax. One can in- stead use an iterative choice for the values of k’s such that the mode of hkn+1, r∗ n+1, lies at r∗ n + σn with σn the standard deviation of hkn. Assum- ing perfect Gaussian distributions for each random walk, and requiring that the values go up to the mode of the mass distribution of the cube minus the standard deviation, p d/3 −1/15 − p 4/(45d), some simple algebra leads to the asymptotic scaling Kd ∼d1/2. Since MBAR does not require the bias- ing potentials to be harmonic, it is likely that this scaling could be brought down even further with an altogether different choice of biasing potentials, e.g. by choosing a series of potentials of the form Ui(r) = r1−d exp[−k(r −ri)2], where r = \|r−r0\|, k is a fixed width, and ri is a tunable scalar distance from r0. 0 2 4 6 8 10 -100 -80 -60 -40 -20 0 r log10[h(r)/r d–1] d=1000 (b) pMC ≈10 -76 Second, one could accelerate the diffusion of the random walks by “cloud sampling”, which amounts to biasing the Monte Carlo sampling on the average weight of a larger number of trial points at each step [119]. Alternatively, one could introduce some deterministic drift in the random walks, so as to make them closer to recently-proposed piecewise- deterministic processes [26], or so-called Galilean Monte Carlo [27,28] methods, where particles travel following straight lines and bounce on walls instead of diffusing around. iv Measuring hypercubes In fact, we manage to reconstruct the density of states reliably up to points that would require 10120 Monte Carlo shots on a sphere, using only 108 points across all walks. Our approach also yields the mass distribution, as reconstructed by unbiasing the histograms of individual random walks, according to Eq. 8. In Fig. 7a we show the hypercube mass distribution, h(r), for a few dimensionalities up to d = 1000. As expected, as d increases, h(r) tends to a sharp peak centered at p d/3a, indicated by a red dashed line for d = 1000. Altogether, we have shown for a simple exam- ple that using importance sampling and a free en- ergy estimation method (here, MBAR) enables us to measure high-dimensional volumes much more efficiently than brute-force Monte Carlo by virtue of the correlations between successive positions of the random walkers, and by avoiding getting stuck in singular features of the domain thanks to parallel tempering. To illustrate the difficulty of measuring not just the volume, but also the mass distribution of the cube in such high dimensionality, in Fig. 7b we plot h(r)/rd−1 for d = 1000, which amounts to the prob- ability of landing inside the cube when sampling uniformly on the surface of the hypersphere with radius r. Because the intersection between the cube and a sphere becomes very small, we plot the base- 10 log of that ratio. A value of particular interest is that obtained for the mode of h(r), as it gives an estimate of how many trials would be needed to find a single point inside the cube if we were performing brute-force Monte Carlo, restricted to Going beyond the simple example given here, the contrast in efficiency can be made even starker by refining the precise design of the algorithm we pre- sented. First, one could choose a different set of 150001-9 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani 0 2 4 6 8 10 0 1 2 3 4 r h(r) (a) 2d 10d 50d 100d 200d 500d 1000d 0 2 4 6 8 10 -100 -80 -60 -40 -20 0 r log10[h(r)/r d–1] d=1000 (b) pMC ≈10 -76 Figure 7: Mass distribution estimate. (a) Mass dis- tribution of the unit-sidelength hypercube, a = 0.5 as estimated from MBAR, using the same runs as in Fig. v Unbiasing sampled volumes Recall that our original aim was to estimate the mean basin volume, ⟨v⟩, in Eq. 3. So, after having measured the volume of a large number of basins of attraction, we must deal with the issue that a na¨ıve average over the basin volumes of randomly sam- pled energy minima is, in general, biased. Indeed, if one samples configuration space uniformly and tags each point with its basin of attraction, each basin is sampled with probability pi = vi/V, that is to say, proportionally to its volume. Thus, to compute a Boltzmann entropy SB = log Ω+ const., we need to perform a fit of the observed basin volume dis- tribution using a putative functional form to undo the bias and obtain the true mean, ⟨v⟩. Concretely, if the measured (biased) distribution of volumes is called B(v), and the unbiased (true) distribution U(v), one can write Figure 8: Energy landscape of hard-WCA particles. Left: Snapshot of jammed packing of polydisperse disks with hard cores (dark shaded) plus soft repulsive coro- nas (light shaded). Right: Illustration of configura- tional space for jammed packings. The hatched re- gions are inaccessible due to hard-core overlaps. Single- colored regions with contour lines represent the basins of attraction of distinct minima. Blue region with solid dots indicates the coexisting unjammed fluid region (observed only for finite size systems) and hypotheti- cal marginally stable packings. Figures adapted from Ref. [14]. B(v) = NU(v)v, (10) (10) (10) where v is the bias proportional to volumes, and N is a normalisation factor. By integration, one finds that the unbiased mean is given by granular packings in two dimensions [12–15]. Due to their athermal nature, granular systems can- not be described by Gibbsian statistical mechan- ics [116]. Nevertheless, in the late 1980s, Edwards and Oakeshott [29] proposed that the collection of stable packings of a fixed number of particles in a fixed volume could play the role of an ensemble, and that one could arrive at a statistical mechanical formalism by making the assumption that all stable packings are equally probable once the system has settled in a jammed state. In other words, in the Edwards’ ensemble, jammed states occur with uni- form probability measure (also known as Edwards’ measure [30]). Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani v Unbiasing sampled volumes The existence of a Boltzmann-like distribution in volume and stress in granular me- dia has been supported by a number of experi- ments [31–38], lending credence to Edwards’ the- ory. However, theoretical checks of Edwards’ key hypothesis on the equiprobability of packings was largely considered as being not directly testable in simulations, as it seemingly required an explicit enumeration of all possible packings. N −1 = ⟨v⟩=   V Z 0 dv v B(v)   −1 , (11) (11) where one typically needs to fit B(v) to some func- tional form, for instance a parametric generalized Gaussian distribution, or a nonparametric kernel density estimate. Ω The Shannon entropy, SS = −PΩ i=1 pi log vi + const., is, instead, by definition the biased average of the negative log-volumes of basins, and it can be obtained directly with no assumptions. All in all, given a proper algorithmic basis, basin- volume measurements provide a generic way of computing the Gibbs-Shannon entropy of any dy- namical system whose properties are controlled by the ensemble of its steady-state structures. In the following, we shall first show how this idea was ap- plied to the special case of granular packings, and then how it could answer open questions in funda- mental physics. iv Measuring hypercubes The higher exploration efficiency afforded to us by this method is far from being free in general basin volume computations: in order for the ran- dom walk to remain within the basin, we need to query an “oracle” at every step to determine if we are inside or outside the shape of interest [17]. While this oracle is a simple geometric condition in the case of the hypercube, for basins of attraction in a many-body energy landscape we must solve for the path of steepest descent at every step of the ran- dom walk [123], meaning that a single Monte Carlo step typically requires hundreds of energy (func- tion) evaluations. In liquids, for instance, the cost of one such energy evaluation scales at best like the number of particles, N, which needs to be large for thermodynamic properties to be measured ac- curately. While costly, we will show in Sec. III that this cost is manageable in a practical example. Figure 7: Mass distribution estimate. (a) Mass dis- tribution of the unit-sidelength hypercube, a = 0.5, as estimated from MBAR, using the same runs as in Fig. 6, in d = 2, 10, 50, 100, 200, 500, 1000 dimensions. (b) Decimal log of the ratio between the mass distribu- tion and that of the smallest ball containing the whole cube for d = 1000. The dashed red lines indicates the location p d/3a of the maximum for d = 1000, and the value of the ratio it corresponds to in (b). We indicate the probability pMC of landing a single point inside the cube when drawing points uniformly from that specific hyperspherical shell. biasing potentials than the one used above, which reproduces the strategy of Ref. [14]. For instance, the scaling we used for the number of replicas is biasing potentials than the one used above, which reproduces the strategy of Ref. [14]. For instance, the scaling we used for the number of replicas is 150001-10 III The case of granular entropy This problem can be rephrased as a basin volume measurement: consider a system of polydisperse disks with a hard core and a soft corona, modelled as a shifted Weeks-Chandler-Anderson (WCA) [39] The first application of basin-volume calculations to measure entropy was performed in the context of 150001-11 Papers in Physics, vol. 15, art. III The case of granular entropy 150001 (2023) / M Casiulis & S Martiniani ■■ ■ ■ ■ ■ ■ ■ ■■ ■ ■ ■ ■ ■ ■ 0.82 0.83 0.84 0.85 0.86 10 15 20 25 30 ϕ S (a) SB SG ϕJ ■■ ■ ■ ■ ■ ■ ■ 0.82 0.83 0.84 0.85 0.86 0.00 0.02 0.04 0.06 0.08 0.10 ϕ λ (c) ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● -6 -4 -2 0 2 4 6 210 215 220 225 230 235 240 ln P –ln pi (b) ϕ 0.828 0.830 0.835 0.840 0.845 0.850 0.855 0.860 Figure 9: Checking Edwards’ hypothesis. III The case of granular entropy ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● -6 -4 -2 0 2 4 6 210 215 220 225 230 235 240 ln P –ln pi (b) ϕ 0.828 0.830 0.835 0.840 0.845 0.850 0.855 0.860 ■■ ■ ■ ■ ■ ■ ■ ■■ ■ ■ ■ ■ ■ ■ 0.82 0.83 0.84 0.85 0.86 10 15 20 25 30 ϕ S (a) SB SG ϕJ ■■ ■ ■ ■ ■ ■ ■ 0.82 0.83 0.84 0.85 0.86 0.00 0.02 0.04 0.06 0.08 0.10 ϕ λ (c) Figure 9: Checking Edwards’ hypothesis. III The case of granular entropy (a) Gibbs-Shannon entropy SG (blue) and Boltzmann entropy SB (orange) as a function of ϕ, as obtained from a parameter-free kernel-density estimate (KDE) fit of the distribution of basin volumes. The dashed curves are second-order polynomial fits. As ϕ approaches the unjamming density for N = 64 particles ϕJ = 0.82 (dashed gray line), SG tends to SB, implying equiprobability of all states precisely at unjamming. (b) Scatter plot of the log of the probability of landing in a given basin, pi (proportional to the volume vi) against the log of the pressure P measured in that basin. Power-law scaling relations (solid lines) are found for several densities indicated in the inset. (c) Exponents found for the power laws of (b), plotted against ϕ. The dashed black line is a linear fit, and the dashed gray line indicates ϕJ. The raw data from Ref. [14] was used here and re-analyzed independently. soft potential (see Fig. 8). Consider N such parti- cles, put them into a periodic box with sidelength L at a packing fraction ϕ. The system is endowed with a many-body energy landscape that, at high enough densities, contains many different basins. At the bottom of each basin lies a single local mini- mum, that can be interpreted as a zero-temperature configuration of the system. At the characteristic unjamming density ϕJ (the point at which the sys- tem goes from a liquid to a disordered solid), each minimum corresponds to a packing of non-frictional hard particles. Among these minima, there is a subset of mechanically stable configurations. De- pending on the system of interest, it is common to define mechanical stability through a non-zero bulk modulus, which is equivalent to saying that the number of contacts verifies Nc ≥d(Nnr −1), where Nnr is the number of particles that are not rattlers (mobile particles) [40], thus defining the en- semble of collectively jammed states [41]. A smaller subset of minima is that of states stable not only against compression, but also shear: the condi- tion of non-zero shear modulus this time imposes Nc ≥d(Nnr −1) + 1, which defines the ensemble of strictly jammed states [41]. In practice, we restrict ourselves to the latter case [14], which is a more broadly accepted definition of jamming. There- fore, by implementing the machinery of Sec. III The case of granular entropy (a) Gibbs-Shannon entropy SG (blue) and Boltzmann entropy SB (orange) as a function of ϕ, as obtained from a parameter-free kernel-density estimate (KDE) fit of the distribution of basin volumes. The dashed curves are second-order polynomial fits. As ϕ approaches the unjamming density for N = 64 particles ϕJ = 0.82 (dashed gray line), SG tends to SB, implying equiprobability of all states precisely at unjamming. (b) Scatter plot of the log of the probability of landing in a given basin, pi (proportional to the volume vi) against the log of the pressure P measured in that basin. Power-law scaling relations (solid lines) are found for several densities indicated in the inset. (c) Exponents found for the power laws of (b), plotted against ϕ. The dashed black line is a linear fit, and the dashed gray line indicates ϕJ. The raw data from Ref. [14] was used here and re-analyzed independently. III The case of granular entropy II to the energy landscape of hard-WCA particles, and restricting the set of relevant basins to mechani- cally stable configurations, one can estimate the Boltzmann and Gibbs-Shannon entropy of granu- lar packings. The main result, originally obtained in Ref. [14] for N = 64 hard-WCA disks, is reproduced in Fig. 9a. This plot shows that the Gibbs-Shannon entropy, SG = −PΩ i=1 pi log pi + const., and the Boltzmann entropy SB = log Ω+const., both eval- 150001-12 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani i Future work on granular packings As discussed in Sec. III, the basin-volume method led to a direct observation that Edwards’ hypothe- sis is only valid strictly at jamming in d = 2. There are however still several aspects of this problem that remain unchecked, and could be addressed by the very same method. Furthermore, as shown in Fig. 9b, the authors also found that above jamming, there exists a ro- bust power law relationship, pi ∼P −Nλ(ϕ), be- tween the probability of observing a packing and its pressure. This power law suggests a hierar- chical structure of the energy landscape of hard- WCA particles, where low-energy minima have large basin volumes and high-energy minima have small volumes, see Fig. 10 for an illustration of this property. First, these results were obtained within the iso- choric ensemble, approaching unjamming where the measured pressure vanishes, P →0+. It has been argued that the equiprobability of packings could break down when switching to the isobaric ensem- ble [42], viz., when allowing volume fraction fluctu- ations through particle inflation and deflation to maintain constant hydrostatic pressure, or more generally in the isostress ensemble that also con- strains the shear stresses. Finally, as shown in Fig. 9c, the exponent λ de- creases roughly linearly as the packing fraction ap- proaches jamming from above, and reaches 0 at ϕJ. In other words, at jamming, the volume of basins becomes a flat distribution that does not depend on the pressure in the system, adding to the evidence that all basins are equiprobable at ϕJ. As shown in App. B, this argument on λ can be made more formal, and the difference between the Boltzmann and Gibbs entropies can be written as The isobaric and isostress ensembles can be ex- plored within a basin volume framework. To do so, in the spirit of Parrinello-Rahman barostats [43– 45], the trick is to allow for deformations of the simulation box (both isotropic compressions and constant-volume deformations), and to replace the energy landscape, E, by an enthalpy-like landscape, H [40], where H = E −σ : εV0 (13) (13) lim N→∞ 1 N [SB(V ) −SG(V )] = O(λ2). (12) (12) explicitly contains the dependence on the full stress and strain tensors σ and ε through their Frobenius inner product A : B = P i,j AijBij, as well as a reference volume V0 of the simulation box. IV The road ahead: from packings to generic dynamical systems uated using basin volume measurements, converge precisely at ϕJ ≈0.82, implying that only at this density is Edwards’ hypothesis, pi = 1/Ω, verified. It can be checked through finite-size analysis [14] that this packing fraction is consistent with the lower range of values for which the system unjams at N = 64 following the same preparation protocol, so that the two entropies coincide only at unjam- ming. Note that, while the value of the unjam- ming density observed in compression experiments depends on the precise preparation protocol, what we report is a property of the energy landscape of jammed packings, which is a generic feature of the system independent of preparation recipes. In the rest of the paper, we propose perspectives for the basin-volume method, ranging from extensions of the problem of granular packings to completely unrelated problems. i Future work on granular packings Note that a subcase of this strategy is that in which only isotropic compression is allowed, ε = V/V0I, with I the identity tensor. In that case, only the trace Tr[σ] ≡−P participates in the box deformation term, leading to the usual definition of enthalpy with respect to the hydrostatic pressure P, namely H = E + PV . Basins are then defined through steepest descent paths that lead to the same en- thalpy minimum, in a configuration space compris- ing not only particle positions, but also d(d + 1)/2 In other words, in order for the two entropies to agree at a given density in the thermodynamic limit, the exponent λ must go to 0 as observed. All in all, a first application to granular packings has shown the potential of the method proposed in Sec. II. By replacing an intractable enumera- tion problem with a simpler sampling problem, the basin-volume method allowed us to test a hypoth- esis that had been left unchecked for 30 years. It also yielded insight into new physics at the jamming transition, by revealing a hierarchical structure of the basins of attraction above jamming. 150001-13 Papers in Physics, vol. 15, art. i Future work on granular packings 150001 (2023) / M Casiulis & S Martiniani ∝λ∝-γ () ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● - - - - - - - - - / ϕ () ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ●●●●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ●● ● ● ● ● ●● ●● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●●● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●●● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ●●● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ●● ●● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ●● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ●●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ●● ●●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ●● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●●● ● ●● ● ● ● ● ● ●● ●● ● ● ● ● ●●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ●●● ●● ● ● ● ●● ●●●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ●●● ●● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ●● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● - - - () ~ Figure 10: Hierarchical energy landscape of soft sphere packings. i Future work on granular packings (a) Sketch: the power law between basin vol- ume, vi, and pressure suggests that a minimum lying at Ei > 0 has a basin with volume vi ∝E−γ i with γ a positive exponent. We verify this scaling relationship in (b), using the same data as in Fig. 9. Panel (c) shows a smooth dependence of P as a function of E, with an asymptote P ∝ √ E at unjamming. Fig. 10(b), using the same data as in Fig. 9, and report a density-dependent power-law between the volume vi of a basin and the energy at the corre- sponding minimum. This power law results from P being a smooth function of the energy, as shown in Fig. 10(c), where we emphasize that unjamming is accompanied by an asymptote P ∝ √ E. This be- haviour is expected near unjamming, where over- laps are small so that the pair potential is well captured by a harmonic approximation, U(r) ∼ (r −r0)2 [119], and the pressure contribution from one interacting pair, as obtained from a virial ex- pression, reads p ∼r0(r−r0) ∼ √ U which, summed over, can be shown to yield P ∼ √ E as E →0. Fig. 10(b), using the same data as in Fig. 9, and report a density-dependent power-law between the volume vi of a basin and the energy at the corre- sponding minimum. This power law results from P being a smooth function of the energy, as shown in Fig. 10(c), where we emphasize that unjamming is accompanied by an asymptote P ∝ √ E. This be- haviour is expected near unjamming, where over- laps are small so that the pair potential is well captured by a harmonic approximation, U(r) ∼ (r −r0)2 [119], and the pressure contribution from one interacting pair, as obtained from a virial ex- pression, reads p ∼r0(r−r0) ∼ √ U which, summed over, can be shown to yield P ∼ √ E as E →0. Second, one may reasonably wonder whether the results are still observed in higher dimensions of space, notably in d = 3, where many open ques- tions remain to be answered to fully understand the ensemble of jammed states, thus constituting a topic of current research [50]. i Future work on granular packings ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● - - - - - - - - - / ϕ () ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ●●●●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ●● ● ● ● ● ●● ●● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●●● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●●● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ●●● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ●● ●● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ●● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ●●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● ● ● ●● ●●● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ●● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ●●● ● ●● ● ● ● ● ● ●● ●● ● ● ● ● ●●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ●●● ●● ● ● ● ●● ●●●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ●●● ●● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ●● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● - - - () ~ Figure 10: Hierarchical energy landscape of soft sphere packings. i Future work on granular packings (a) Sketch: the power law between basin vol- ume, vi, and pressure suggests that a minimum lying at Ei > 0 has a basin with volume vi ∝E−γ i with γ a positive exponent. We verify this scaling relationship in (b), using the same data as in Fig. 9. Panel (c) shows a smooth dependence of P as a function of E, with an asymptote P ∝ √ E at unjamming. ∝λ∝-γ () parameters describing simulation box deformations [40]. In practice, one would impose a finite but small pressure, and look at the asymptotic results as P →0+. Indeed, a technical difficulty in this approach is that, at exactly zero pressure, fluid so- lutions with arbitrarily large volumes trivially min- imize the enthalpy. One would then need to con- strain the maximal volume of the system to avoid converging exclusively to fluid states. l Finally, to perform basin volume calculations in the enthalpy landscape, a modification of the Monte Carlo algorithm used for the volume esti- mate is required. In the isochoric case presented above, the volumes of basins are computed by sam- pling configurations with constant N and V , using an oracle defined through a minimum of the energy, and umbrella sampling on the particles’ positional degrees of freedom. In the isobaric case, this time, one needs to sample configurations with constant N and P, with an oracle defined through a mini- mum of the enthalpy, and using umbrella sampling not only on the particles’ positions, but also on the degrees of freedom of the box shape. This requires the inclusion of independent shearing and stretch- ing (viz., box deformations) moves in our Monte Carlo sampling [46–49] as the building block of the random walks. Each of these moves is accepted or rejected according to a Monte Carlo criterion set by a combination of the oracle, which checks that the new point still falls to the same minimum of the enthalpy, Eq. 13, and the umbrella sampling biasing potentials that contain box deformations. Implementing this algorithm as part of a basin vol- ume calculation, while numerically costlier than the usual isochoric strategy, would constitute an impor- tant check on the validity of the Edwards hypoth- esis. i Future work on granular packings Finally, we observe that, in the isochoric en- semble and above jamming, the basin volumes are linked to the pressure of the packings via a power- law suggesting that configuration space is tiled by a hierarchy of basins (see Fig. 10(a)), which has neither been fully appreciated, nor explored. We check that such a hierarchical dependence exists in ii Beyond granular jams: glassy systems Based on our discussion thus far, a reader may think that basin volume calculations can only be used to measure granular entropies. In fact, these 150001-14 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani methods could be used to measure a number of in- dicators for the shapes of the basins, in addition to the radial mass distribution, as well as their neighborhoods. Such measurements could have deep fundamental implications, if one does not con- sider a granular (or zero-temperature) system, but a jammed liquid at a finite temperature. Indeed, in a finite-temperature glassy system, the shape of basins controls transition rates and relaxation times [51–54], which has prompted a large body of recent work on the enumeration and characteriza- tion of minima of rough landscapes [52,55–58]. tion (with or without anharmonic corrections), av- eraged over many IS [68]. Using this and an- other approach based on a Frenkel-Ladd-like com- putation, Berthier and coworkers recently produced compelling results suggesting that the change in Sconf for supercooled liquids in d = 3 is consis- tent with the existence of an ideal glass transition at TK > 0 [65,69–71]. Through basin volume calculations, it is possible to compute qi directly, and thereby obtain a general approach to estimating the configurational entropy of supercooled liquids, without resorting to any ap- proximations. In particular, this method does not need to assume large system sizes, and could al- low to precisely study finite-size effects in complex free energy landscapes, a problem that is nontrivial even in simple theories in 1d [66]. Achieving a clear picture of the basins of at- traction in the configuration space of a dense liq- uid in physical dimensions would therefore be very exciting, as it would open avenues for fundamen- tal checks regarding glassy dynamics. In an ex- act mean-field treatment of hard sphere glasses, a Kauzmann transition known as random first- order transition (RFOT) is attained [62–64]. In this framework the configurational entropy Sconf = log ΩF is defined in terms of the number of free en- ergy minima ΩF , so that the notion and number of “glassy states” is well defined and the ideal glass transition results from the population of glassy states becoming subextensive in system size. In finite dimensions, free energy minima are not in- finitely long-lived and the precise definition of a “glassy state” remains debated [65]. ii Beyond granular jams: glassy systems A popular take on the problem was proposed by Stillinger and We- ber [60,67,124], who suggested that the supercooled glassy states would reside close to the minima of the potential energy surface, known as inherent struc- tures (IS) of the liquid. For each IS there is an associated basin of attraction, that is, the set of all initial conditions leading to the i-th IS by steepest descent. The partition function can then be ex- pressed classically as a sum over the individual IS, Q = PΩ i=1 qi, where, for β = 1/kBT, Finally, in the context of glassy systems, free en- ergy methods could in principle be used to obtain not just the (Boltzmann weighted) volume of the basins of attraction of inherent structures, but also more complicated information about their geome- try, topology, and connectivity (e.g., what is the chromatic number for the tiling of basins in the en- ergy landscape? [130]). Understanding the proper- ties of the most probable paths between minima of the energy landscape is of particular interest to un- derstand the dynamics of relaxation in glasses, and could help bridge the gap between microscopic dy- namics and mesoscopic models that recently shed new light on the interpretation of the relaxation spectrum of glassy materials [72,73]. iii Generic dynamical systems: ecosys- tems, neural networks, and more. In the remainder of this paper, we discuss what we believe to be some of the most exciting opportu- nities for basin-volume methods, namely their ap- plications to generic dynamical systems. Indeed, going back to the method presented in Sec. II, we never explicitly relied on the existence of an un- derlying energy function. In principle, one can de- fine any dynamics of their choice, let them evolve, and classify initial conditions according to which steady-state or dynamical attractor they eventu- ally fall into, as sketched in Fig. 11. One can then always define the Shannon entropy associated with the volumes of these attractors, and use it as a generic descriptor of the system for a given set of qi = Z Γi e−βU(r)dr, (14) (14) and U(r) is the interaction potential. Until re- cently, qi could not be computed, and the ac- cepted operational definition for the configura- tional entropy was Sconf = Sliq −Sharm, where Sliq is the total entropy computed by thermody- namic integration, and Sharm is the vibrational entropy computed from the harmonic approxima- 150001-15 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani Figure 11: Generalised basins for dynamical systems. In a generic dynamical system, basin of attractions can be associated not only to fixed points, but also limit cycles, or any kind of stable attractor of the dynamics. Black arrows represent example trajectories, that fall into steady-state structures (white). These structures can be fixed points (yellow basin), simple limit cycles (green basins) or any complicated high-dimensional at- tractor (red basin). standing the structure of high-dimensional basins of attraction is also of great theoretical interest, as the problem of mapping high-dimensional basin volumes has been described as uncharted mathe- matical territory [10]. It is also a current challenge in cosmology, where characterising the number and volume of basins of attraction in axion landscapes constitutes a promising lead for elucidating the cos- mological constant problem [93]. A particular example of dynamical systems where basin volume methods could prove fruitful is the study of ecosystems, i.e. ensembles of popu- lations following Lotka-Volterra dynamics [94]. In the limit of symmetric antagonistic interactions, such systems have been shown to map onto glassy Hamiltonian systems, which leads to aging dynam- ics in a rugged landscape, the properties of which have been studied analytically at the mean-field scale [97–100]. iii Generic dynamical systems: ecosys- tems, neural networks, and more. Therefore, in this setting, one can apply the strategy of basin volume calculations to determine the existence and nature of phase tran- sitions in the limit of a finite number of species. Furthermore, since a Hamiltonian structure is not needed to define entropy via basin-volume calcu- lations, the same approach can also be used in the much more complicated case of non-reciprocal, not- all-antagonistic interactions. In this scenario, qual- itative results, both experimental and theoretical, have shown that stable structures tend to emerge in steady state [101–103]. Basin-volume approaches could provide insight into these systems, for in- stance by quantifying how many ecosystems are possible in steady-state for a given mix of inter- actions, or which ecosystem is the most typical. Figure 11: Generalised basins for dynamical systems. In a generic dynamical system, basin of attractions can be associated not only to fixed points, but also limit cycles, or any kind of stable attractor of the dynamics. Black arrows represent example trajectories, that fall into steady-state structures (white). These structures can be fixed points (yellow basin), simple limit cycles (green basins) or any complicated high-dimensional at- tractor (red basin). parameters. This approach is completely general and can be applied to any system with dynamics that are regular enough to admit attractors. We propose a few promising leads for such approaches. There are many examples of dynamical systems where such an approach would be relevant. For in- stance, mapping basins of attraction is an ubiqui- tous problem in control theory, where approximate methods for measuring basins have been proposed to determine the stability of the flight control-law of the F/A-18 Hornet aircraft with respect to large perturbations [74]. Enumerating the number of sta- tionary points, and their distribution, for certain classes of random functions, is a classical problem in mathematics and statistics [75–88,125]. In par- ticular, in combinatorial optimization, the size and connectedness of the space of solutions that satisfy a large number of constraints controls how easy it is to find a solution, which has implications in a variety of real-life problems ranging from com- puter science to optimal transport [89–92]. Under- parameters. This approach is completely general and can be applied to any system with dynamics that are regular enough to admit attractors. We propose a few promising leads for such approaches. There are many examples of dynamical systems where such an approach would be relevant. V Conclusions We derive a scaling for the required number of repli- cas, Kd, to correctly estimate the volume of a hy- percube in dimension d. Each of these replicas is attached to the center of the cube by a spring with rigidity k and samples positions according to a distribution that is well approximated by a d−dimensional Gaussian. The narrowest of these distributions is the one for kmax, hkmax(r), with variance In this perspective, we presented the principles, re- cent successes, and possible future applications of a powerful new approach to study disordered many- body systems through the estimation of the number and volume of basins of attractions. This approach enables the computation of the Shannon entropy, which appears as a natural extension of the ther- modynamic entropy of equilibrium systems, in a much broader class of systems. It has already been shown that basin-volume computations work in the specific case of granular packings, but there is an avenue ahead. Not only can this method lead to estimations of the features of complex energy land- scapes in thermal systems like supercooled liquids, molecular liquids or spin glasses, it also provides a systematic and quantitative way of characterising the state space of generic dynamical systems. At a time when numerical and experimental studies of a broad spectrum of non-equilibrium many-body systems are booming, in spite of the absence of a general theoretical framework to understand them, such a generic tool could prove to be invaluable. Our vision is that the theoretical ideas and pro- tocols presented herein will become new canonical forms of computation adopted in multiple areas of science and engineering. σ2 max = 1 kmax d −2Γ d+1 2 Γ d 2 ! . (15) (15) We can impose that the mode of hkmax coincide with the radius, a, of the largest hyperball inscribed in the hypercube, yielding the condition kmax = d −1 a2 . (16) (16) Finally, we know that the k = 0 replica will have most of its mass concentrated at a p d/3. We can therefore require that the number of replicas, Kd, is such that the interval [a; p d/3a] sums up to Kd × 2σmax. iii Generic dynamical systems: ecosys- tems, neural networks, and more. For in- stance, mapping basins of attraction is an ubiqui- tous problem in control theory, where approximate methods for measuring basins have been proposed to determine the stability of the flight control-law of the F/A-18 Hornet aircraft with respect to large perturbations [74]. Enumerating the number of sta- tionary points, and their distribution, for certain classes of random functions, is a classical problem in mathematics and statistics [75–88,125]. In par- ticular, in combinatorial optimization, the size and connectedness of the space of solutions that satisfy a large number of constraints controls how easy it is to find a solution, which has implications in a variety of real-life problems ranging from com- puter science to optimal transport [89–92]. Under- Finally, another exciting lead is the application of basin-volume methods to Deep Neural Networks (DNNs) [104]. It has been widely observed that the parameters learned by DNN during training corre- spond to “flat minima” of the loss-function land- scape [105–110]. Basin volume calculations may enable researchers to answer questions concerning the structure of the error landscapes of DNNs and to identify the relationship between the probability of finding a given solution, its flatness and its gen- eralization performance. Addressing these ques- tions would have a significant impact on our under- standing of generalization in deep learning systems with implications for high-stakes applications such as transportation, security and medicine. 150001-16 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani V Conclusions Putting all of this together, we arrive at the result: Kd = √ 3d −3 6 s d−2 Γ( d+1 2 ) Γ( d 2) d−1 , (17) (17) which for d →∞reduces to Kd ∼d/ √ 6. In other words, in the case of a hypercube, the number of replicas necessary to estimate the volume grows lin- early with d. Acknowledgements - We wish to thank Daniel Asenjo, Daan Frenkel, Johannes Gasteiger (n´e Klicpera), Fabien Paillusson, K. Julian Schrenk, Jacob Stevenson, and Praharsh Suryadevara for many invaluable contributions to the theoretical underpinnings and computer implementation of the methods described in this perspective. The au- thors acknowledge funding from the Simons Foun- dation through the Simons Center for Computa- tional Physical Chemistry, Department of Chem- istry, NYU. S. M. gracefully acknowledges support from the National Science Foundation grant IIS- 2226387. S. M. performed part of this work at the Aspen Center for Physics, which is supported by National Science Foundation grant PHY-1607611. B Implications of the power-law be- tween basin volumes and pres- sures We report how to deduce an expression for the difference between the Boltzmann and Gibbs en- tropies of granular packings from the power-law scaling between the volumes of basins and the pres- sures of the corresponding packings. Recall that the Gibbs entropy of a jammed configuration of N particles can be written as SG = −PΩ i pi ln pi − ln N! with a sum running over basins. Noticing that 150001-17 In order to simplify this expression, we use the re- sult of Ref. [14] that ln P is approximately normal- distributed, so that pi = vi/V, one gets SG = −⟨ln vi⟩B −ln N! + ln V, where the average is biased proportionally to the volume of the basins. The power-law relation shown in Fig. 9b entails a linear law between logs, −⟨ln vi⟩B = N(λ(ϕ)⟨ln P⟩B + c(ϕ)). Therefore, the Gibbs entropy reads B(P\|V )P Nλ ≈ 1 P √ 2πσ2 e−(ln P −µ)2 2σ2 +Nλ ln P . (24) (24) It is convenient to rewrite the integral in square brackets in Eq. 23 as It is convenient to rewrite the integral in square brackets in Eq. 23 as SG = Nλ⟨ln P⟩B + Nc −ln N! + ln V. (18) (18) Now, the Boltzmann entropy at a given value of pressure and volume can be computed as the log of the number of configurations Ω(P, V ), which can be written as Now, the Boltzmann entropy at a given value of pressure and volume can be computed as the log of the number of configurations Ω(P, V ), which can be written as Ξ = ∞ Z 0 dP ∞ Z 0 dΠB(Π\|V )xNλδ(Π −P). (25) (25) Ω(P, V ) = Ω(V ) P +δP Z P U(Π\|V )dΠ, (19) After some algebra, one finds After some algebra, one finds (19) Ξ = exp Nλµ + N 2λ2 σ2 2 . (26) (26) where U(P\|V ) is the unbiased distribution of pres- sures given a total volume V , and the integral over a small element imposes the value of P within the measure U. B Implications of the power-law be- tween basin volumes and pres- sures Since in our protocol the basins are sampled proportionally to their volume, B is linked to U via Using the empirical observation that σ2 = s/N and µ = µ∞+ m/N where s, µ∞, m are all O(1) [14], we find that lim N→∞ ln Ξ N = λµ∞+ λ2 s 2. (27) (27) B(P\|V ) = U(P\|V )v(P, ϕ, N) ⟨v⟩ , (20) Identifying µ∞with ⟨ln P⟩B, from Eqs. 18, 23 and 27 we finally get Identifying µ∞with ⟨ln P⟩B, from Eqs. 18, 23 and 27 we finally get (20) with v(P, ϕ, N) the volume of a basin at pres- sure P, packing fraction ϕ and N particles. Ac- cording to the results in Fig. 9b, these volumes verify v(P, ϕ, N) = P −Nλe−Nc. Recalling that Ω(V ) = V/⟨v⟩, the number of configurations with a fixed pressure reads with v(P, ϕ, N) the volume of a basin at pres- sure P, packing fraction ϕ and N particles. Ac- cording to the results in Fig. 9b, these volumes verify v(P, ϕ, N) = P −Nλe−Nc. Recalling that Ω(V ) = V/⟨v⟩, the number of configurations with a fixed pressure reads lim N→∞ 1 N [SB(V ) −SG(V )] = λ2 s 2. (28) (28) This expression shows that the two entropies only coincide when λ →0, but also that SB ≥SG, as observed in the measurements reported in Fig. 9 of the main text. Ω(P, V ) = VeNc P +δP Z P B(Π\|V )ΠNλdΠ. (21) (21) [1] R Clausius, ¨Uber eine ver¨anderte Form des zweiten Hauptsatzes der mechanischen W¨armetheorie, Annalen der Physik. 169, 481, (1854). [1] R Clausius, ¨Uber eine ver¨anderte Form des zweiten Hauptsatzes der mechanischen W¨armetheorie, Annalen der Physik. 169, 481, (1854). Furthermore, the number of configurations consid- ering all possible pressures at a given volume is sim- ply the sum Furthermore, the number of configurations consid- ering all possible pressures at a given volume is sim- ply the sum [2] C E Shannon, A mathematical theory of com- munication, The Bell system technical jour- nal 27, 379, (1948). Ω(V ) = ∞ Z 0 dPΩ(P, V ). (22) (22) [3] S Martiniani, P M Chaikin, D Levine, Quan- tifying hidden order out of equilibrium, Phys. Rev. X 9, 011031, (2019). B Implications of the power-law be- tween basin volumes and pres- sures The Boltzmann entropy at a given volume therefore reads SB(V ) = ln   ∞ Z 0 dP P +δP Z P B(Π\|V )ΠNλdΠ   [4] S Martiniani, Y Lemberg, P M Chaikin, et al., Correlation lengths in the language of computable information, Phys. Rev. Lett. 125, 170601, (2020). + Nc −ln N! + ln V. (23) (23) 150001-18 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani [5] R Avinery, M Kornreich, R Beck, Univer- sal and Accessible Entropy Estimation Using a Compression Algorithm, Phys. Rev. Lett. 123, 178102, (2019). [17] D Frenkel, K J Schrenk, S Martiniani, Monte Carlo sampling for stochastic weight func- tions, Proc. Nat. Ac. Sci. 114, 6924, (2017). [18] S S Ashwin, J Blawzdziewicz, C S O’Hern, et al., Calculations of the structure of basin vol- umes for mechanically stable packings, Phys. Rev. E 85, 061307, (2012). [6] A Cavagna, P M Chaikin, D Levine, et al., Vicsek model by time-interlaced compression: A dynamical computable information density, Phys. Rev. E 103, 062141, (2021). [19] M R Shirts, J D Chodera, Statistically opti- mal analysis of samples from multiple equi- librium states, J. Chem. Phys. 129, 124105, (2008). [7] S Ro, B Guo, A Shih, et al., Model-Free Mea- surement of Local Entropy Production and Extractable Work in Active Matter, Phys. Rev. Lett. 129, 220601, (2022). [20] X Ding, J Vilseck, C Brooks, Fast Solver for Large Scale Multistate Bennett Acceptance Ratio Equations, J. Chem. Theory Comput. 2019, 802, (2019). [8] D A Wiley, S H Strogatz, M Girvan, The size of a sync basin, Chaos 16, 015103, (2006). [9] S P Cornelius, W L Kath, A E Motter, Real- istic control of network dynamics, Nat. Com- mun. 4, 1, (2013). [21] J G Kirkwood, Statistical Mechanics of Fluid Mixtures, J. Chem. Phys. 3, 300, (1935). [10] Y Zhang, S H Strogatz, Basins with tentacles, Phys. Rev. Lett. 127, 194101, (2021). [22] A Gelman, X-L Meng, Simulating normaliz- ing constants: from importance sampling to bridge sampling to path sampling, Statist. Sci. 13, 185, (1998). [11] D Frenkel, A J Ladd, New Monte Carlo mehod to compute the free energy of arbitrary solids. Application to the fcc and hcp phases of hard spheres, J. Chem. Phys. 81, 3188, (1984). [23] A Bunker, B D¨unweg, Parallel excluded vol- ume tempering for polymer melts, Phys. Rev. E 63, 016701, (2000). B Implications of the power-law be- tween basin volumes and pres- sures [12] N Xu, D Frenkel, A J Liu, Direct determi- nation of the size of basins of attraction of jammed solids, Phys. Rev. Lett. 106, 245502, (2011). [24] H Fukunishi, O Watanabe, S Takada, On the hamiltonian replica exchange method for effi- cient sampling of biomolecular systems: Ap- plication to protein structure prediction, J. Chem. Phys. 116, 9058, (2002). [13] D Asenjo, F Paillusson, D Frenkel, Numerical calculation of granular entropy, Phys. Rev. Lett. 112, 098002, (20114). [25] J M Torrie, J P Valleau, Nonphysical sam- pling distributions in Monte Carlo free-energy estimation: Umbrella sampling, J. Comput. Phys. 23, 199, (1977). [14] S Martiniani, K J Schrenk, K Pamola, et al., Numerical test of the Edwards conjecture shows that all packings are equally probable at jamming, Nat. Phys. 13, 848, (2017). [26] A Chevallier, F Cazals, P Fearnhead, Effi- cient computation of the volume of a polytope in high dimensions using Piecewise Deter- ministic Markov Processes, ArXiv Preprint, 2202.09129, (2022). [15] S Martiniani, K J Schrenk, J D Stevenson, et al., Turning intractable counting into sam- pling: computing the configurational entropy of three-dimensional jammed packings, Phys. Rev. E 93, 012906, (2016). [27] J Skilling, Bayesian Computation in big spaces-nested sampling and Galilean Monte Carlo, AIP Conference Proceedings 1443, 145, (2012). [16] S Martiniani, K J Schrenk, J D Stevenson, et al., Structural analysis of high-dimensional basins of attraction, Phys. Rev. E 93, 031301, (2016). 150001-19 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani [28] M Griffiths, D J Wales, Nested Basin- Sampling, J. Chem. Theory Comput. 15, 6881, (2019). [40] C P Goodrich, S Dagois-Bohy, B P Tighe, et al., Jamming in finite systems: Stability, anisotropy, fluctuations, and scaling, Phys. Rev. Lett. 112, 145502, (2014). [29] S F Edwards, R B S Oakeshott, Theory of Powders, Physica A 157, 1080, (1989). [41] S Atkinson, F Stillinger, S Torquato, Detailed characterization of rattlers in exactly iso- static, strictly jammed sphere packings, Phys. Rev. E 88, 062208, (2013). [30] A Baule, F Morone, H J Herrmann, et al., Ed- wards statistical mechanics for jammed gran- ular matter, Rev. Mod. Phys. 90, 015006, (2018). [42] G-J Gao, J B lawzdziewicz, C S O’Hern, Frequency distribution of mechanically sta- ble disk packings, Phys. Rev. E 74, 061304, (2006). [31] T Aste, Volume fluctuations and geometri- cal constraints in granular packs, Phys. Rev. Lett. 96, 018002, (2006). B Implications of the power-law be- tween basin volumes and pres- sures [63] S Franz, G Parisi, Recipes for metastable states in spin glasses, Journal de Physique I 5, 1401, (1995). [52] V Ros, G Ben Arous, G Biroli, et al., Com- plex Energy Landscapes in Spiked-Tensor and Simple Glassy Models: Ruggedness, Arrange- ments of Local Minima, and Phase Transi- tions, Phys. Rev. X 9, 011003, (2019). [64] G Parisi, F Zamponi, Mean-field theory of hard sphere glasses and jamming, Rev. Mod. Phys. 82, 789, (2010). [65] L Berthier, M Ozawa, C Scalliet, Configu- rational entropy of glass-forming liquids, J. Chem. Phys. 150, 160902, (2019). [53] T Rizzo, Path integral approach unveils role of complex energy landscape for activated dy- namics of glassy systems, Phys. Rev. B 104, 094203, (2021). [66] C Rulquin, P Urbani, G Biroli, G Tarjus, M Tarzia, Nonperturbative fluctuations and metastability in a simple model: from observ- ables to microscopic theory and back, J. Stat. Mech. 2016, 023209, (2016). [54] G Folena, A Manacorda, F Zamponi, Intro- duction to the dynamics of disordered sys- tems: equilibrium and gradient descent, Lec- tures Notes for the Fundamental Problems in Statistical Physics XV Summer School (2021). [67] F H Stillinger, A topographic view of super- cooled liquids and glass formation, Science 267, 1935, (1995). [68] F Sciortino, Potential energy landscape de- scription of supercooled liquids and glasses, J. Stat. Mech. 2005, P05015, (2005). [55] V Ros, G Biroli, C Cammarota, Complexity of energy barriers in mean-field glassy sys- tems, EPL 126, 20003, (2019). [69] L Berthier, P Charbonneau, D Coslovich, A Ninarello, M Ozawa, S Yaida, Configura- tional entropy measurements in extremely su- percooled liquids that break the glass ceiling, Proc. Nat. Ac. Sci. 114, 11356, (2017). [56] V Ros, G Biroli, C Cammarota, Dynamical Instantons and Activated Processes in Mean- Field Glass Models, SciPost Phys. 10, 002, (2021). [57] B Lacroix-`a-chez-Toine, Y Fyodorov, Count- ing equilibria in a random non-gradient dy- namics with heterogeneous relaxation rates, J. Phys. A: Math. Theor. 55, 144001, (2022). [70] M Ozawa, G Parisi, L Berthier, Configura- tional entropy of polydisperse supercooled liq- uids, J. Chem. Phys. 149, 154501, (2018). [71] Y Nishikawa, M Ozawa, A Ikeda, P Chaud- huri, L Berthier, Relaxation Dynamics in the Energy Landscape of Glass-Forming Liquids, Phys. Rev. X 12, 021001, (2022). [58] J Kent-Dobias, J Kurchan, Complex complex landscapes, Phys. Rev. Research 3, 023064, (2021). B Implications of the power-law be- tween basin volumes and pres- sures [43] M Parrinello, A Rahman, Crystal Structure and Pair Potentials: A Molecular-Dynamics Study, Phys. Rev. Lett. 45, 413, (1980). [32] T Aste, T Di Matteo Emergence of Gamma distributions in granular materials and pack- ing models, Phys. Rev. E 77, 021309, (2008). [44] M Parrinello, A Rahman, Polymorphic tran- sitions in single crystals: A new molecular dynamics method, J. App. Phys. 52, 7182, (1981). [33] S McNamara, P Richard, S K De Richter, et al., Measurement of granular entropy, Phys. Rev. E 80, 031301, (2009). [34] J G Puckett, K E Daniels Equilibrating temperature-like variables in jammed granu- lar subsystems, Phys. Rev. Lett. 110, 058001, (2013). [45] M Parrinello, A Rahman, Strain fluctuations and elastic constants, J. Chem. Phys. 76, 2662, (1982). [46] R Najafabadi, S Yip, Observation of Finite- Temperature Bain Transformation (f.c.c. ↔ b.c.c.) in Monte Carlo Simulation of Iron, Scr. Metall. 17, 1199, (1983). [35] S-C Zhao, M Schr¨oter, Measuring the con- figurational entropy of a binary disc packing, Soft Matter 10, 4208, (2014). [36] E S Bililign, J E Kollmer, K E Daniels Protocol dependence and state variables in the force-moment ensemble, Phys. Rev. Lett. 122, 038001, (2019). [47] S Yashonath, C Rao, A monte carlo study of crystal structure transformations, Mol. Phys. 54, 245, (1985). [48] R J N Baldock, L B P´artay, A P Bart´ok, et al., Determining pressure-temperature phase diagrams of materials, Phys. Rev. B 93, 174108, (2016). [37] X Sun, W Kob, R Blumenfeld, et al., Friction-controlled entropy-stability competi- tion in granular systems, Phys. Rev. Lett. 125, 268005, (2020). [49] R J N Baldock, N Bernstein, K Michael Salerno, et al., Constant-pressure nested sam- pling with atomistic dynamics, Phys. Rev. E 96, 043311, (2017). [38] Y Yuan, Y Xing, J Zheng, et al., Experi- mental test of the Edwards volume ensemble for tapped granular packings, Phys. Rev. Lett. 127, 018002, (2021). [50] C Anzivino, M Casiulis, T Zhang, et al., Esti- mating random close packing in polydisperse and bidisperse hard spheres via an equilibrium model of crowding, J. Chem. Phys., In Press (2022). [39] J D Weeks, D Chandler, H C Andersen, Role of repulsive forces in determining the equi- librium structure of simple liquids, J. Chem. Phys. 54, 5237, (1971). 150001-20 150001-20 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani [51] T Castellani, A Cavagna, Spin-glass theory for pedestrians, J. Stat. Mech. 2005, P05012, (2005). B Implications of the power-law be- tween basin volumes and pres- sures [59] E Bitzek, P Koskinen, F G¨ahler, et al., Struc- tural relaxation made simple, Phys. Rev. Lett. 97, 170201, (2006). [72] C Scalliet, B Guiselin, L Berthier, Excess wings and asymmetric relaxation spectra in a facilitated trap model, J. Chem. Phys. 155, 064505, (2021). [60] F H Stillinger, T A Weber, Hidden structure in liquids, Phys. Rev. A 25, 978, (1982). [73] B Guiselin, C Scalliet, L Berthier, Micro- scopic origin of excess wings in relaxation spectra of supercooled liquids, Nat. Phys. 18, 468, (2022). [61] S D Cohen, A C Hindmarsh, P F Dubois, CVODE, a stiff/nonstiff ODE solver in C, Computers in Physics 10, 138, (1996). [62] P Charbonneau, J Kurchan, G Parisi, et al., Glass and jamming transitions: From ex- act results to finite-dimensional descriptions, Ann. Rev. Cond. Mat. Phys. 8, 265, (2017). [74] A Chakraborty, P Seiler, G J Balas, Sus- ceptibility of F/A-18 flight controllers to the fallingleaf mode: Nonlinear analysis, J. Guid. Control Dyn. 34, 73, (2011). 150001-21 150001-21 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani [75] M Kac, On the average number of real roots of a random algebraic equation, Bull. Am. Math. Soc. 49, 314, (1943). [86] Y V Fyodorov, High-dimensional random fields and random matrix theory, ArXiv Preprint, arXiv:1307.2379, (2013). [76] K Farahmand, On the average number of real roots of a random algebraic equation, Ann. Probab. 14, 702, (1986). [87] Y V Fyodorov, A Lerario, E Lundberg, On the number of connected components of ran- dom algebraic hypersurfaces, J. Geom. Phys. 95, 1, (2015). [77] E Bogomolny, O Bohigas, P Leboeuf, Distri- bution of roots of random polynomials, Phys. Rev. Lett. 68, 2726, (1992). [88] D Cheng, A Schwartzman, On the explicit height distribution and expected number of local maxima of isotropic Gaussian random fields, ArXiv Preprint, arXiv:1503.01328, (2015). [78] A Edelman, E Kostlan, How many zeros of a random polynomial are real?, Bull. Am. Math. Soc. 32, 1, (1995). [89] F Krzakala, J Kurchan, Landscape analysis of constraint satisfaction problems, Phys. Rev. E 76, 021122, (2007). [79] J M Rojas, On the average number of real roots of certain random sparse polynomial systems, Lectures in Applied Mathematics- American Mathematical Society 32, 689, (1996). [90] F Krzakala, L Zdeborov´a, Hiding Quiet So- lutions in Random Constraint Satisfaction Problems, Phys. Rev. Lett. 102, 238701, (2009). B Implications of the power-law be- tween basin volumes and pres- sures [80] Y V Fyodorov, Complexity of random en- ergy landscapes, glass transition, and abso- lute value of the spectral determinant of ran- dom matrices, Phys. Rev. Lett. 92, 240601, (2004). [91] L Zdeborov´a, M M´ezard, Constraint Satis- faction Problems with Isolated Solutions are Hard, J. Stat. Mech. 2008, P12004, (2008). [81] G Malajovich, J M Rojas, High probability analysis of the condition number of sparse polynomial systems, Theor. Comput. Sci. 315, 525, (2004). [92] M-H Tayarani-Narajan, A Pr¨ugel-Bennett, On the Landscape of Combinatorial Optimi- sation Problems, IEEE Transactions on Evo- lutionary Computation 18, 420, (2013). [82] J-M Aza¨ıs, M Wschebor, On the roots of a random system of equations. The theorem of Shub and Smale and some extensions, Found. Comput. Math. 5, 125, (2005). [93] T C Bachlechner, K Eckerle, O Janssen, et al., Axion landscape cosmology, Journal of Cosmology and Astroparticle Physics 2019, 062, (2019). [83] D Armentano, M Wschebor, Random systems of polynomial equations. The expected number of roots under smooth analysis, Bernoulli 15, 249, (2009). [94] A J Lotka, Contribution to the Theory of Periodic Reaction, J. Phys. Chem. 14, 271, (1910). [95] A J Lotka, Analytical Note on Certain Rhyth- mic Relations in Organic Systems, Proc. Natl. Acad. Sci. U.S.A. 6, 410, (1920). [84] Y V Fyodorov, G A Hiary, J Keating, Freezing transition, characteristic polynomi- als of random matrices, and the Riemann zeta function, Phys. Rev. Lett. 108, 170601, (2012). [96] V Volterra, Variazioni e fluttuazioni del nu- mero d’individui in specie animali conviventi, Mem. Acad. Lincei Roma 2, 31, (1926); Na- ture 118, 558 (1926). [85] Y V Fyodorov, C Nadal, Critical behavior of the number of minima of a random landscape at the glass transition point and the Tracy- Widom distribution, Phys. Rev. Lett. 109, 167203, (2012). [97] G Bunin, Ecological communities with Lotka- Volterra dynamics, Phys. Rev. E 95, 042414, (2017). 150001-22 150001-22 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani [98] G Biroli, G Bunin, C Cammarota, Marginally stable equilibria in critical ecosystems, New J. Phys. 20, 083051, (2018). [110] F Pittorino, C Lucibello, C Feinauer, et al., Entropic gradient descent algorithms and wide flat minima, J. Stat. Mech. 2021, 124015, (2021). [99] F Roy, G Biroli, G Bunin, et al., Numerical implementation of dynamical mean field the- ory for disordered systems: application to the Lotka–Volterra model of ecosystems, J. Phys. A: Math. Theor. B Implications of the power-law be- tween basin volumes and pres- sures 52, 484001, (2019). [111] S S Mannelli, G Biroli, C Cammarota, et al., Complex dynamics in simple neural net- works: Understanding gradient flow in phase retrieval, NeurIPS 33, 3265, (2020). [112] D M Ceperley, M Dewing, The penalty method for random walks with uncertain en- ergies, J. Chem. Phys. 110, 9812, (1999). [100] A Altieri, F Roy, C Cammarota, et al., Prop- erties of Equilibria and Glassy Phases of the Random Lotka-Volterra Model with Demo- graphic Noise, Phys. Rev. Lett. 126, 258301, (2021). [113] R L Stratonovich, On a Method of Calculat- ing Quantum Distribution Functions, Soviet Physics Doklady 2, 416, (1957). [101] M Barbier, J-F Arnoldi, G Bunin, et al., Generic assembly patterns in complex eco- logical communities, Proc. Natl. Acad. Sci. U.S.A. 115, 2156, (2018). [114] J Hubbard, Calculation of Partition Func- tions, Phys. Rev. Lett. 3, 77, (1959). [115] P Attard, Thermodynamics and statistical mechanics: equilibrium by entropy maximi- sation, Academic Press, London (2002). [102] G Bunin, Directionality and community-level selection, Oikos 130, 489, (2018). [103] M Barbier, C de Mazancourt, M Loreau, et al., Fingerprints of High-Dimensional Coex- istence in Complex Ecosystems, Phys. Rev. X 11, 011009, (2021). [116] J W Gibbs, Elementary principles in statisti- cal mechanics, Charles Scribner’s Sons, New York (1902). [117] M Kardar, Statistical Physics of Parti- cles, Cambridge University Press, Cambridge (2007). [104] A J Ballard, R Das, S Martiniani, et al., En- ergy landscapes for machine learning, Phys. Chem. Chem. Phys. 19, 12585-12603, (2017). [118] S Carnot, R´eflexions sur la puissance motrice du feu et sur les machines propres `a d´evelopper cette puissance, Annales Scien- tifiques de l’´E.N.S., Bachelier (1824). [105] S Hochreiter, J Schmidhuber, Simplify- ing neural nets by discovering flat minima, NeurIPS 7, (1994). [106] S Hochreiter, J Schmidhuber, Flat minima, Neur. Comput. 9, 1, (1997). [119] S Martiniani, On the complexity of energy landscapes: algorithms and a direct test of the Edwards conjecture, University of Cam- bridge, Cambridge (2017). [107] C Baldassi, F Pittorino, R Zecchina, Shap- ing the learning landscape in neural networks around wide flat minima, Proc. Natl. Acad. Sci. U.S.A. 117, 161, (2020). [120] T M Cover, Elements of information theory, John Wiley & Sons, New York (1999). [108] Y Feng, Y Tu, The inverse variance–flatness relation in stochastic gradient descent is criti- cal for finding flat minima, Proc. Natl. Acad. Sci. U.S.A. 118, e2015617118, (2021). [130] D Frenkel, Private communication. B Implications of the power-law be- tween basin volumes and pres- sures [121] D Frenkel, B Smit, Understanding molecular simulation: from algorithms to applications, Vol. I, Elsevier (2001). [109] S Zhang, I Reid, G P´erez, et al., Why Flatness Correlates With Generalization For Deep Neural Networks, Arxiv Preprint arXiv:2103.06219, (2021). [122] D Landau, K Binder, A Guide to Monte Carlo Simulations in Statistical Physics, Sec- ond Edition, Cambridge University Press, Cambridge (2005). 150001-23 150001-23 Papers in Physics, vol. 15, art. 150001 (2023) / M Casiulis & S Martiniani [123] D Wales, Energy landscapes: Applications to clusters, biomolecules and glasses, Cambridge University Press, Cambridge (2003). [127] S Artstein-Avidan, A Giannopoulos, V D Milman, Asymptotic Geometric Analysis, Part I, Mathematical Surveys and Mono- graphs, Volume 202, American Mathematical Society, Providence, Rhode Island (2015). [124] F H Stillinger, Energy landscapes, inherent structures, and condensed-matter phenom- ena, Princeton University Press, New Yersey (2015). [128] I Gel’Fand, G Shilov, Generalized functions, Vol. 1, Academic Press, New York (1968). [125] E Kostlan, On the expected number of real roots of a system of random polynomial equa- tions, Foundations of Computational Mathe- matics, World Scientific, Singapore (2002). [129] E W Weisstein, Gabriel’s Horn, In: MathWorld–A Wolfram Web Resource, mathworld.wolfram.com/GabrielsHorn.html [130] D Frenkel, Private communication. [126] A J Lotka, Elements of Physical Biology, Williams and Wilkins, Baltimore (1925). 150001-24 150001-24
https://openalex.org/W4389937529	https://www.mdpi.com/2073-4360/16/1/7/pdf?version=1702979582	English	null	Kinetics and Mechanism of Liquid-State Polymerization of 2,4-Hexadiyne-1,6-diyl bis-(p-toluenesulfonate) as Studied by Thermal Analysis	Polymers	2,023	cc-by	9,813	Citation: Galukhin, A.; Kachmarzhik, A.; Rodionov, A.; Mamin, G.; Gafurov, M.; Vyazovkin, S. Kinetics and Mechanism of Liquid-State Polymerization of 2,4-Hexadiyne- 1,6-diyl bis-(p-toluenesulfonate) as Studied by Thermal Analysis. Polymers 2024, 16, 7. https:// doi.org/10.3390/polym16010007 Academic Editors: Giulio Malucelli and Mariaenrica Frigione Received: 2 November 2023 Revised: 29 November 2023 Accepted: 18 December 2023 Published: 19 December 2023 Citation: Galukhin, A.; Kachmarzhik, A.; Rodionov, A.; Mamin, G.; Gafurov, M.; Vyazovkin, S. Kinetics and Mechanism of Liquid-State Polymerization of 2,4-Hexadiyne- 1,6-diyl bis-(p-toluenesulfonate) as Studied by Thermal Analysis. Polymers 2024, 16, 7. https:// doi.org/10.3390/polym16010007 Keywords: diacetylenes; polymerization; azide–alkyne cycloaddition; isoconversional analysis; differential scanning calorimetry; electron paramagnetic resonance Article Kinetics and Mechanism of Liquid-State Polymerization of 2,4-Hexadiyne-1,6-diyl bis-(p-toluenesulfonate) as Studied by Thermal Analysis Andrey Galukhin 1,* , Alexander Kachmarzhik 1, Alexander Rodionov 2, Georgy Mamin 2, Marat Gafurov 2 and Sergey Vyazovkin 3,* ,* , Alexander Kachmarzhik 1, Alexander Rodionov 2, Georgy Mamin 2, Marat Gafurov 2 kin 3,* 1 Alexander Butlerov Institute of Chemistry, Kazan Federal University, 18 Kremlevskaya Street, 420008 Kazan, Russia; sasha.kachma@mail.ru g * Correspondence: and_galuhin@mail.ru (A.G.); vyazovkin@uab.edu (S.V.) Abstract: A detailed investigation of the liquid-state polymerization of diacetylenes by calorimetric (DSC) and spectroscopic (in situ EPR) thermal analysis techniques is performed. Isoconversional kinetic analysis of the calorimetric data reveals that liquid-state polymerization is governed by a well-defined rate-limiting step as evidenced by a nearly constant isoconversional activation energy. By comparison, solid-state polymerization demonstrates isoconversional activation energy that varies widely, signifying multistep kinetics behavior. Unlike the solid-state reaction that demonstrates an autocatalytic behavior, liquid-state polymerization follows a rather unusual zero-order reaction model as established by both DSC and EPR data. Both techniques have also determined strikingly similar Arrhenius parameters for liquid-state polymerization. Relative to the solid-state process, liquid-state polymerization results in quantitative elimination of the p-toluenesulfonate group and the formation of p-toluenesulfonic acid and a polymeric product of markedly different chemical and phase composition. polymers polymers polymers 1. Introduction Polydiacetylenes are extensively studied because of their unique structural, spectral, and optical properties [1]. Their highly conjugated structure being exposed to chemi- cal [2–4], thermal [5–7], or mechanical [8–10] stimulus undergoes conformational changes, which alter chromatic properties of the material. Solid-state polymerization of diacetylenes is a topochemical reaction that requires specific orientation of the monomer molecules in the crystalline phase [11]. It proceeds as a 1,4-addition with intermediate formation of various radical species [12] to yield a rigid polymer backbone containing conjugated enyne fragments (Figure 1). Polymerization is usually initiated by raising temperature or supplying UV radiation. According to the literature data, solid-state thermally stimulated polymerization of diacetylenes demonstrates pronounced autocatalytic behavior [13–15]. When investigated in isothermal conditions, it goes through an induction period followed by a rapid reaction. In the solid-state, the reported ratio of the maximum reaction rate to the reaction rate during the induction period (rmax/ri), used as a characteristic of autocatalytic processes, reaches ~100 according to calorimetric [16] and spectroscopic [17] studies. A similar value of polymerization heat (~270 kJ mol−1) was reported for liquid-state reaction of some rigid diacetylenes [19]. It was also reported [19] that the induction period is absent in the case of the liquid-crystalline-state polymerization and that the properties of the polymers obtained differ from those for the materials obtained in the melt. It was tentatively suggested [19] that the structures of the polymers obtained in liquid-crystalline and isotropic melt polymerization differ as well. On the other hand, later studies of the molten state polymerization of unsymmetrically substituted diacetylenes by FTIR concluded that both reactions proceed in the same manner [20]. y j q y p y , heat of polymerization exceeds dramatically that of the reaction in the crystalline phase (~240 vs. ~140 kJ mol−1) [18]. A similar value of polymerization heat (~270 kJ mol−1) was reported for liquid-state reaction of some rigid diacetylenes [19]. It was also reported [19] that the induction period is absent in the case of the liquid-crystalline-state polymerization and that the properties of the polymers obtained diﬀer from those for the materials ob- tained in the melt. It was tentatively suggested [19] that the structures of the polymers obtained in liquid-crystalline and isotropic melt polymerization diﬀer as well. On the other hand, later studies of the molten state polymerization of unsymmetrically substi- tuted diacetylenes by FTIR concluded that both reactions proceed in the same manner [20]. Such intriguing diﬀerences between the polymerization processes in ordered and disordered phases and the absence of general insights into this issue have motivated us to y y p [20]. Such intriguing diﬀerences between the polymerization processes in ordered and disordered phases and the absence of general insights into this issue have motivated us to undertake a detailed study of the liquid-state polymerization of diacetylenes. As the sub- ject of our work, we have chosen 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) because, unlike other diacetylenes, its polymerization kinetics in the solid state are well studied and documented [16,17,21], which makes it a preferable reference system for our compar- ative liquid-state study. We have used diﬀerential scanning calorimetry (DSC) to study the polymerization process. The obtained calorimetric data were then subjected to com- prehensive isoconversional kinetic analysis [22] in order to determine the activation en- ergy, pre-exponential factor, activation entropy, and reaction model of the process. The calorimetric data were complemented by in situ EPR measurements. Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4 0/) https://www.mdpi.com/journal/polymers Polymers 2024, 16, 7. https://doi.org/10.3390/polym16010007 Polymers 2024, 16, 7 Polymers 2024, 16, x F 2 of 12 2 of 12 Figure 1. Scheme of the solid-state polymerization of diacetyle For thermally induced liquid-state polymerization Figure 1. Scheme of the solid-state polymerization of diacetylenes to polydiacetylene Figure 1. Scheme of the solid-state polymerization of diacetylenes to polydiacetylene sults seem to diﬀer markedly from those related to the solid-state reaction. that fo a et of dia etyle e ubje ted to the li uid y talli e tate oly Figure 1. Scheme of the solid-state polymerization of diacetylenes to polydiacetylenes. Figure 1. Scheme of the solid-state polymerization of diacetylenes to polydiacetylenes. sults seem to diﬀer markedly from those related to the solid-state reaction. It w that for a set of diacetylenes subjected to the liquid crystalline state polymeriz For thermally induced liquid-state polymerization of diacetylenes, the reported re- sults seem to diﬀer markedly from those related to the solid-state reaction. It was found that for a set of diacetylenes subjected to the liquid-crystalline-state polymerization, the heat of polymerization exceeds dramatically that of the reaction in the crystalline phase (~240 vs. ~140 kJ mol−1) [18]. A similar value of polymerization heat (~270 kJ mol−1) was reported for liquid-state reaction of some rigid diacetylenes [19]. It was also reported [19] that the induction period is absent in the case of the liquid-crystalline-state polymerization and that the properties of the polymers obtained diﬀer from those for the materials ob- tained in the melt. It was tentatively suggested [19] that the structures of the polymers obtained in liquid-crystalline and isotropic melt polymerization diﬀer as well. On the other hand, later studies of the molten state polymerization of unsymmetrically substi- tuted diacetylenes by FTIR concluded that both reactions proceed in the same manner For thermally induced liquid-state polymerization of diacetylenes, the reported results seem to differ markedly from those related to the solid-state reaction. It was found that for a set of diacetylenes subjected to the liquid-crystalline-state polymerization, the heat of polymerization exceeds dramatically that of the reaction in the crystalline phase (~240 vs. ~140 kJ mol−1) [18]. A combination of these two techniques allowed us to gain valuable insights into the mechanism of liquid- Such intriguing differences between the polymerization processes in ordered and disordered phases and the absence of general insights into this issue have motivated us to undertake a detailed study of the liquid-state polymerization of diacetylenes. As the subject of our work, we have chosen 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) because, unlike other diacetylenes, its polymerization kinetics in the solid state are well studied and documented [16,17,21], which makes it a preferable reference system for our comparative liquid-state study. We have used differential scanning calorimetry (DSC) to study the polymerization process. The obtained calorimetric data were then subjected to comprehensive isoconversional kinetic analysis [22] in order to determine the activation energy, pre-exponential factor, activation entropy, and reaction model of the process. The calorimetric data were complemented by in situ EPR measurements. A combination of these two techniques allowed us to gain valuable insights into the mechanism of liquid-state polymerization. p g g undertake a detailed study of the liquid-state polymerization of diacetylenes. As the sub- ject of our work, we have chosen 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) because, unlike other diacetylenes, its polymerization kinetics in the solid state are well studied and documented [16,17,21], which makes it a preferable reference system for our compar- ative liquid-state study. We have used diﬀerential scanning calorimetry (DSC) to study the polymerization process. The obtained calorimetric data were then subjected to com- prehensive isoconversional kinetic analysis [22] in order to determine the activation en- ergy, pre-exponential factor, activation entropy, and reaction model of the process. The calorimetric data were complemented by in situ EPR measurements. A combination of these two techniques allowed us to gain valuable insights into the mechanism of liquid- state polymerization. 2. Materials and Methods 2 Materials an 2.1. Materials Propargy 2. Materials and Methods 2.1. Materials Propargyl alcohol (99%, Sigma-Aldrich, Waltham, MA, USA), copper (I) chloride (99%, Sigma-Aldrich, Saint Louis, MI, USA), ammonium chloride (99.5%, Chimmed, Mos- cow, Russia), pyridine (99.5%, EKOS-1, Moscow, Russia), p-toluenesulfonyl chloride (TsCl, 99+%, Acros Organics, Geel, Belgium), potassium hydroxide (86%, Chimmed, Mos- cow, Russia), sodium sulfate (anhydrous, 99.5%, Chemprom-M, Yaroslavl, Russia), meth- anol (99.5%, Vekton, Saint Petersburg, Russia), dichloromethane (99.5%, Chemprom-M, Yaroslavl, Russia), benzene (99.8%, EKOS-1, Moscow, Russia), and silica gel (60 Å, Ma- chery-Nagel, Duren, Germany) were used as received. Acetone (99.5%, Chemprom-M, Ya- roslavl, Russia) was distilled over phosphorus pentoxide before use. Tetrahydrofuran (99.5%, EKOS-1, Moscow, Russia) and diethyl ether (99%, Kuzbassorgchem, Kemerovo, Russia) were distilled over potassium hydroxide before use. Deionized water (18.2 MΩ) Propargyl alcohol (99%, Sigma-Aldrich, Waltham, MA, USA), copper (I) chloride (99%, Sigma-Aldrich, Saint Louis, MI, USA), ammonium chloride (99.5%, Chimmed, Moscow, Russia), pyridine (99.5%, EKOS-1, Moscow, Russia), p-toluenesulfonyl chloride (TsCl, 99+%, Acros Organics, Geel, Belgium), potassium hydroxide (86%, Chimmed, Moscow, Russia), sodium sulfate (anhydrous, 99.5%, Chemprom-M, Yaroslavl, Russia), methanol (99.5%, Vek- ton, Saint Petersburg, Russia), dichloromethane (99.5%, Chemprom-M, Yaroslavl, Russia), benzene (99.8%, EKOS-1, Moscow, Russia), and silica gel (60 Å, Machery-Nagel, Duren, Germany) were used as received. Acetone (99.5%, Chemprom-M, Yaroslavl, Russia) was distilled over phosphorus pentoxide before use. Tetrahydrofuran (99.5%, EKOS-1, Moscow, Russia) and diethyl ether (99%, Kuzbassorgchem, Kemerovo, Russia) were distilled over potassium hydroxide before use. Deionized water (18.2 MΩ) was obtained using an Arium mini-instrument (Sartorius, Gottingen, Germany). Target monomer 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) was synthesized according to Figure 2. (99%, Sigma-Aldrich, Saint Louis, MI, USA), ammonium chloride (99.5%, Chimmed, Mos- cow, Russia), pyridine (99.5%, EKOS-1, Moscow, Russia), p-toluenesulfonyl chloride (TsCl, 99+%, Acros Organics, Geel, Belgium), potassium hydroxide (86%, Chimmed, Mos- cow, Russia), sodium sulfate (anhydrous, 99.5%, Chemprom-M, Yaroslavl, Russia), meth- anol (99.5%, Vekton, Saint Petersburg, Russia), dichloromethane (99.5%, Chemprom-M, Yaroslavl, Russia), benzene (99.8%, EKOS-1, Moscow, Russia), and silica gel (60 Å, Ma- chery-Nagel, Duren, Germany) were used as received. Acetone (99.5%, Chemprom-M, Ya- roslavl, Russia) was distilled over phosphorus pentoxide before use. Tetrahydrofuran (99.5%, EKOS-1, Moscow, Russia) and diethyl ether (99%, Kuzbassorgchem, Kemerovo, Russia) were distilled over potassium hydroxide before use. Deionized water (18.2 MΩ) was obtained using an Arium mini-instrument (Sartorius, GoĴingen, Germany). Target monomer 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) was synthesized according to Figure 2. was obtained using an Arium mini-instrument (Sartorius, GoĴingen, Germany). Target monomer 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) was synthesized according to Figure 2. Figure 2. 2 Materials an 2.1. Materials Propargy Synthesis of 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) from propargyl alcohol via hexa-2,4-diyne-1,6-diol. monomer Figure 2. Figure 2. Synthesis of 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) from propargyl alcohol via hexa-2,4-diyne-1,6-diol. Figure 2. Synthesis of 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) from propargyl alcohol via hexa-2,4-diyne-1,6-diol. Polymers 2024, 16, 7 3 of 12 Hexa-2,4-diyne-1,6-diol was synthesized from propargyl alcohol with Glaser cross- coupling [23]. Propargyl alcohol (20 g, 0.357 mol), pyridine (7.2 mL, 0.089 mol), copper (I) chloride (0.424 g, 4.28 mmol), and 200 mL of methanol were added to a round-bottom flask equipped with condenser and magnetic stirring bar. Obtained mixture was heated to 35 ◦C and then oxygen was bubbled through it for 4 h. After that, the reaction mixture was poured into a saturated aqueous ammonium chloride solution and extracted with several portions of diethyl ether. Combined diethyl ether extract was washed with 10 wt % sodium carbonate solution and dried over anhydrous sodium sulfate. After that, diethyl ether was evaporated, and the obtained crude product was recrystallized from benzene. The yield was 27%. The synthesized compound was additionally purified using silica gel column chromatography with diethyl ether as an eluent before the next synthetic step. 1H NMR (D2O): δ (ppm) 4.28 (4H, s, CH2). 13C NMR (D2O): δ (ppm) 49.83, 68.74, 77.33. ( 2 ) (pp ) ( , , 2) ( 2 ) (pp ) , , 2,4-Hexadiyne-1,6-diyl bis-(p-toluenesulfonate). The target monomer is sensitive to light; therefore, its synthesis, isolation, and purification procedures were carried out in the dark using a red lamp. Hexa-2,4-diyne-1,6-diol (2.00 g, 18.2 mmol) and p-toluenesulfonyl chloride (9.00 g, 47.2 mmol) were dissolved in 60 mL of tetrahydrofuran under continuous stirring, then the solution of potassium hydroxide (3.67 g, 65.3 mmol) in 30 mL of water was added dropwise while cooling the mixture on the water bath. Then, the reaction mixture was stirred for 5 h at room temperature. The reaction mixture was poured into 150 mL of ice water, producing red precipitate. The precipitate was filtered off and washed with ice water. The obtained product (the monomer) was purified using silica gel column chromatography with dichloromethane as an eluent. The yield was 61%. Tm (onset) was measured to be 94 ± 2 ◦C and the heat of melting was 36.3 ± 0.8 kJ mol−1. 1H NMR (CDCl3, δ, ppm): 2.45 (6H, s, CH3), 4.73 (4H, s, CH2), 7.35–7.80 (8H, m, Ar-H). 13C NMR (CDCl3, δ, ppm): 21.73, 57.46, 71.98, 72.18, 128.16, 130.01, 132.59, 145.60. IR (cm−1): 1361, 1368 (S=O), 1172 (C–O) (Figures S1–S3). 2.2. Methods A Bruker AVANCE III NMR spectrometer (Billerica, MA, USA) was run at 400 MHz to produce the 1H and 13C NMR spectra using CDCl3 as a solvent. The reported chem- ical shifts, delta (δ), are in units of parts per million (ppm). A Bruker Vertex 70 FTIR spectrometer was used to record IR spectra. p p A heat flux DSC 3+ (Mettler-Toledo, Greifensee, Switzerland) was employed for calori- metric measurements. Temperature, heat flow, and tau-lag calibrations were accomplished by utilizing indium and zinc standards. Nonisothermal experiments were performed under argon flow (80 mL min−1) in the temperature range 25–250 ◦C at the heating rates of 1.0, 2.0, 4.0, and 8.0 ◦C min−1 in 40 µL aluminum pans with pierced lids. Isothermal measurements were performed at 70, 75, 80, and 85 ◦C with the same DSC instrument in 40 µL aluminum pans with pierced lids. The mass of the sample in each run was 5.0 ± 0.1 mg. The mass loss of the sample during polymerization was evaluated by simultaneous TGA-DSC measure- ment with Netzsch STA 449 F1 Jupiter thermal analyzer (Selb, Germany). The measurement was conducted at 10.0 ◦C min−1 under argon flow of 75 mL min−1, as recommended by the manufacturer. The mass of the sample was ~5 mg. The mass loss of the sample in the temperature range of 50–200 ◦C did not exceed 2.5% (Figure S8). p g g A MiniFlex 600 diffractometer (Rigaku, Tokyo, Japan) with a D/teX Ultra detectorX-ray was used for powder diffraction (XRPD) measurements. The instrument radiation source was Cu Kα1 operated at 40 kV, 15 mA. XRPD data were collected at ambient temperature at the diffraction angle 2θ varying from 2◦to 100◦at 0.02◦steps and 0.24 s exposure time at each point without sample rotation. p p In situ EPR measurements were performed with microwave X-band (9.61 GHz) Bruker ESP–300 spectrometer equipped with a Bruker ER 4111 VT temperature attachment in a continuous wave mode. Monomer samples were placed in glass ampoules hermetically sealed in the atmosphere of argon. The concentration of paramagnetic centers in the Polymers 2024, 16, 7 4 of 12 final samples was estimated in a double resonator ER 4105DR (Bruker) using a standard 2,2-diphenyl-1-picrylhydrazyl calibration sample. final samples was estimated in a double resonator ER 4105DR (Bruker) using a standard 2,2-diphenyl-1-picrylhydrazyl calibration sample. where J[Eα, Ti(tα)] = Z tα tα−∆α exp −Eα RTi(t) dt (2) (2) and p is the number of the temperature programs, T(t). Statistical errors in Eα were evaluated as described elsewhere [25]. The obtained isoconversional values of Eα were then substituted into the equation of the compensation effect: ln Aα = a + bEα (3) (3) to estimate the values of the pre-exponential factor. The a and b parameters of the com- pensation effect were determined by fitting the lnAi and Ei pairs into Equation (3). The respective pairs were found by substituting the fi(α) reaction models into the linearized basic rate equation: ln dα dt −ln[ fi(α)] = ln Ai −Ei RT (4) (4) Each lnAi and Ei pair was found from the intercept and slope of Equation (4) while plugging different fi(α) models in the left-hand side and plotting it against the reciprocal temperature. As recommended [26], four reaction models (the power law (P2, P3, P4) and Avrami–Erofeev (A2) were utilized in the compensation effect calculations. p The numerical values of the integral reaction model were estimated by inserting Eα and Aα into Equation (5). (5) g(α) = ∑ α Aα J[Eα, Ti(tα)] (5) The isoconversional activation energy for the solid-state polymerization under isother- mal conditions was determined by the Friedman differential method [27] by plotting the left-hand side of Equation (6) against the reciprocal temperature. ln dα dt α,p = ln[Aα f (α)] − Eα RTα,p (6) (6) 3. Calculations Kinetic analysis was carried out according to the guidelines of the ICTAC Kinetic Committee [22]. The flexible integral isoconversional method of Vyazovkin [24] was employed to evaluate the activation energy Eα as a function of conversion (α) for the liquid-state polymerization studied in nonisothermal conditions. The use of this method eliminates the systematic error in Eα found for rigid integral methods when Eα reveals a significant dependence on α [24]. The advantage arises from the flexible integration conducted over small intervals of ∆α, within which the variability of Eα can be neglected. For the present calculations, ∆α was set as small as 0.01. Within each ∆α, Eα was determined as the value that corresponds to the minimum of the following function: Ψ(Eα) = p ∑ i=1 p ∑ j̸=i J[Eα, Ti(tα)] J Eα, Tj(tα) (1) (1) where 4. Results and Discussion DSC curves for isothermal solid-state polymerization (numbers denote temperature in ◦C) (A), and nonisothermal liquid-state polymerization of the studied monomer (numbers denote heating rates in ◦C min−1) (B). Figure 3A shows the DSC curves for solid-state polymerization measured in isother- mal conditions. The reaction generates 128 ± 4 kJ mol−1 of heat, which closely matches previously reported values for the same monomer [13,16,21]. It must be noted that the solid-state polymerization clearly demonstrates distinct autocatalytic behavior. It is man- ifested by a well-deﬁned induction period followed by acceleration. The duration of the induction period is seen to be up to about 6 h. Figure 3B presents temperature scans at diﬀerent heating rates of the liquid-state polymerization of the monomer under study. As one can see, the reaction starts above 120 °C, i.e., roughly 30 °C higher than the melting temperature of the monomer. Remarkably, the liquid-state reaction produces almost three times more heat than the reaction in the crystalline phase, namely, 360 ± 5 kJ mol−1. Con- sidering that the heat of melting of the monomer is 36.3 ± 0.8 kJ mol−1, such signiﬁcant diﬀerence between the reaction heat values cannot be explained simply by the diﬀerence in the phase states of the reacting monomer. Rather, it means that the chemistry of the reaction in these phases is completely diﬀerent. The net result of the solid-state reaction of diacetylenes proceeding with the formation of conjugated enyne structure (Figure 1) is the transformation of one π-bond into one σ-bond (per 1 monomer molecule). Apparently, the liquid-state reaction requires a larger number of breaking and forming bonds per mon- omer molecule. The liquid-state polymerization of the studied monomer can be proposed to occur as an intramolecular cyclization of diacetylenes, similar to the Hexadehydro Diels–Alder reaction [29]. In this case, two π-bonds are converted into two σ-bonds, which h ld d i ll l f h Figure 3A shows the DSC curves for solid-state polymerization measured in isothermal conditions. The reaction generates 128 ± 4 kJ mol−1 of heat, which closely matches previously reported values for the same monomer [13,16,21]. It must be noted that the solid- state polymerization clearly demonstrates distinct autocatalytic behavior. It is manifested by a well-defined induction period followed by acceleration. The duration of the induction period is seen to be up to about 6 h. 4. Results and Discussion Polymerization of diacetylenes is a highly exothermic reaction due to the formation of the conjugated polymer chains (Figure 1). Therefore, the reaction progress is conveniently followed by DSC. In the kinetics studies by DSC, nonisothermal conditions are generally preferred because they avoid some technical issues as well as the diffusion limitations commonly encountered in the later stages of the process [28]. Naturally, this approach has been taken to measure the kinetics of liquid-state polymerization. However, the kinetics Polymers 2024, 16, 7 5 of 12 ﬀusion ly this of the solid-state polymerization of diacetylenes has been routinely studied in isothermal conditions [13–15], including the present work. The reason is easy to understand by comparing the DSC data presented in Figure 3. The solid-state process has to be run below Tm = 94 ◦C. As seen from nonisothermal DSC data, lowering the heating rate nearly 10 times lowers the peak temperature by less than 50 ◦C. To bring the peak temperatures under 94 ◦C, the heating rates would have to be lowered roughly below 0.01 ◦C min−1, which lies below the range typically accessible by DSC instruments because of the signal to noise problem. pp q p y the kinetics of the solid-state polymerization of diacetylenes has been routinely studied in isothermal conditions [13–15], including the present work. The reason is easy to under- stand by comparing the DSC data presented in Figure 3. The solid-state process has to be run below Tm = 94 °C. As seen from nonisothermal DSC data, lowering the heating rate nearly 10 times lowers the peak temperature by less than 50 °C. To bring the peak temper- atures under 94 °C, the heating rates would have to be lowered roughly below 0.01 °C min−1, which lies below the range typically accessible by DSC instruments because of the signal to noise problem. Figure 3. DSC curves for isothermal solid-state polymerization (numbers denote temperature in °C) (A), and nonisothermal liquid-state polymerization of the studied monomer (numbers denote heating rates in °C min−1) (B). Figure 3. DSC curves for isothermal solid-state polymerization (numbers denote temperature in ◦C) (A), and nonisothermal liquid-state polymerization of the studied monomer (numbers denote heating rates in ◦C min−1) (B). Figure 3. DSC curves for isothermal solid-state polymerization (numbers denote temperature in °C) (A), and nonisothermal liquid-state polymerization of the studied monomer (numbers denote heating rates in °C min−1) (B). Figure 3. 4. Results and Discussion Figure 4. Isoconversional values of activation energy for solid- (squares) and liquid-state (circles) reactions Figure 4. Isoconversional values of activation energy for solid- (squares) and liquid-state (circles) reactions. On the contrary, the liquid-state polymerization demonstrates a nearly constant ef- fective activation energy in the range of α = 0.1–0.9, which signiﬁes the existence of a well- deﬁned rate-limiting step. The averaged value of the activation energy for this step is 106 ± 2 kJ mol−1. The natural logarithm of the pre-exponential factor Aα determined via com- pensation eﬀect (Equations (3) and (4)) is also nearly constant in the range of α = 0.1–0.9, On the contrary, the liquid-state polymerization demonstrates a nearly constant ef- fective activation energy in the range of α = 0.1–0.9, which signifies the existence of a well-defined rate-limiting step. The averaged value of the activation energy for this step is 106 ± 2 kJ mol−1. The natural logarithm of the pre-exponential factor Aα determined via compensation effect (Equations (3) and (4)) is also nearly constant in the range of α = 0.1–0.9, and its average value equals 22.3 ± 0.4 (Aα in s−1). and its average value equals 22.3 ± 0.4 (Aα in s−1). It is also instructive to compare the reaction models for solid- and liquid-state polymerizations. As already mentioned, solid-state polymerization follows the reaction model of an autocatalytic type (Figure 3A). It is well known that the integral reaction mod- els g(α) for the autocatalytic reactions (e.g., the Avrami–Erofeev models) present a sig- moid, i.e., distinctly nonlinear dependence on α [22]. Unfortunately, it is impossible to identify the exact form of g(α) for solid-state polymerization because it is not a single-step process. There is no such problem for the liquid-state reaction. For it, the experimental g(α) dependence is practically linear in a wide range of conversions (Figure 5). A linear g q It is also instructive to compare the reaction models for solid- and liquid-state polymer- izations. As already mentioned, solid-state polymerization follows the reaction model of an autocatalytic type (Figure 3A). It is well known that the integral reaction models g(α) for the autocatalytic reactions (e.g., the Avrami–Erofeev models) present a sigmoid, i.e., distinctly nonlinear dependence on α [22]. Unfortunately, it is impossible to identify the exact form of g(α) for solid-state polymerization because it is not a single-step process. There is no such problem for the liquid-state reaction. 4. Results and Discussion Figure 3B presents temperature scans at different heating rates of the liquid-state polymerization of the monomer under study. As one can see, the reaction starts above 120 ◦C, i.e., roughly 30 ◦C higher than the melting temperature of the monomer. Remarkably, the liquid-state reaction produces almost three times more heat than the reaction in the crystalline phase, namely, 360 ± 5 kJ mol−1. Considering that the heat of melting of the monomer is 36.3 ± 0.8 kJ mol−1, such significant difference between the reaction heat values cannot be explained simply by the difference in the phase states of the reacting monomer. Rather, it means that the chemistry of the reaction in these phases is completely different. The net result of the solid-state reaction of diacetylenes proceeding with the formation of conjugated enyne structure (Figure 1) is the transformation of one π-bond into one σ-bond (per 1 monomer molecule). Apparently, the liquid-state reaction requires a larger number of breaking and forming bonds per monomer molecule. The liquid- state polymerization of the studied monomer can be proposed to occur as an intramolecular cyclization of diacetylenes, similar to the Hexadehydro Diels–Alder reaction [29]. In this case, two π-bonds are converted into two σ-bonds, which should produce a proportionally large amount of heat. should produce a proportionally large amount of heat. The measured calorimetric data have been subjected to isoconversional kinetic anal- ysis to quantify the reactivity of the monomer in the solid- and liquid-state reactions and g The measured calorimetric data have been subjected to isoconversional kinetic analysis to quantify the reactivity of the monomer in the solid- and liquid-state reactions and to compare the kinetics for both processes. The calculated conversion dependencies of the activation energy Eα for both reactions are presented in Figure 4. It is seen that the reactions demonstrate very different trends in the effective activation energy change with conversion. The effective activation energy of solid-state polymerization increases from 68 to 95 kJ mol−1, highlighting its complex kinetics that definitely comprises more than Polymers 2024, 16, 7 6 of 12 h con- m 68 to one rate-limiting step. It is interesting to mention that previous studies of the solid- state polymerization of 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) by extraction [21], spectroscopic [17], and calorimetric techniques [16] report a single value of the activation energy equal to 94 ± 2 kJ mol−1. 4. Results and Discussion This means that the solid-state polymerization of the 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) was treated as a single-step process, which obviously is not the case as demonstrated by our analysis. It should be noted that in all previous studies the activation energy was calculated from the temperature dependence of the reaction time related to a specific conversion (e.g., half-conversion time t0.5) or the time related to the maximum polymerization rate (tmax, which is also roughly isoconversional). Such approaches produce inaccurate results because the respective equations are obtained by integration that assumes that the activation energy is independent of conversion and, thus, of time. This assumption holds true only for single-step processes. However, if the process rate is limited by more than one step, the resulting effective activation energy becomes conversion dependent [22,30], which invalidates the aforementioned assumption and introduces a systematic error in the resulting integrated equation for determining the activation energy. No such error occurs in the differential method used in the present work. limiting step. It is interesting to mention that previous studies of the solid-state polymer- ization of 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) by extraction [21], spectroscopic [17], and calorimetric techniques [16] report a single value of the activation energy equal to 94 ± 2 kJ mol−1. This means that the solid-state polymerization of the 2,4-hexadiyne-1,6- diyl bis-(p-toluenesulfonate) was treated as a single-step process, which obviously is not the case as demonstrated by our analysis. It should be noted that in all previous studies the activation energy was calculated from the temperature dependence of the reaction time related to a speciﬁc conversion (e.g., half-conversion time 𝑡଴.ହ) or the time related to the maximum polymerization rate (𝑡௠௔௫, which is also roughly isoconversional). Such ap- proaches produce inaccurate results because the respective equations are obtained by in- tegration that assumes that the activation energy is independent of conversion and, thus, of time. This assumption holds true only for single-step processes. However, if the process rate is limited by more than one step, the resulting eﬀective activation energy becomes conversion dependent [22,30], which invalidates the aforementioned assumption and in- troduces a systematic error in the resulting integrated equation for determining the acti- vation energy. No such error occurs in the diﬀerential method used in the present work. Figure 4. Isoconversional values of activation energy for solid- (squares) and liquid-state (circles) reactions Figure 4. Isoconversional values of activation energy for solid- (squares) and liquid-state (circles) reactions. 4. Results and Discussion For it, the experimental g(α) dependence is practically linear in a wide range of conversions (Figure 5). A linear dependence of g(α) on α means that the following holds true: ns that the following holds true: g(α) ≡ Z α 0 dα f (α) = Bα + C (7) (7) where B and C are constants. Differentiation of the expression (7) with respect to α leads to: f (α) = B−1 ≡B−1(1 −α)0 (8) (8) Polymers 2024, 16, 7 7 of 12 DSC B). Figure 5 Experimental g(α) data for the studied reaction and their linear approximation Figure 5. Experimental g(α) data for the studied reaction and their linear approximation. Figure 5 Experimental g(α) data for the studied reaction and their linear approximation Figure 5. Experimental g(α) data for the studied reaction and their linear approximation. Figure 5. Experimental g(α) data for the studied reaction and their linear approximation. A simple reaction scheme that can explain the zero-order kinetics is as follows. Let us consider a two-step process that involves the formation of intermediate I via a reaction of reactant A with catalyst B followed by conversion of I into product P (Figure 6). When the concentration of B is much lower than that of A, and k1 >> k2, then the product of A and B concentrations as well as the concentration of I becomes practically constant in a wide range of times. In this case, the rates of the corresponding reactions also remain nearly constant, giving rise to a linear dependence of the concentrations of A and P on time (Fig- Thus, a linear g(α) on α plot corresponds to a zero reaction-order dependence on (1 −α) [31]. The reaction rate profile of a zero-order reaction traces the temperature dependence of the reaction rate constant k(T), which means that under the constant heating rate conditions the rate rises exponentially with increasing temperature until the reaction completes. Since the DSC signal is directly proportional to the reaction rate, the respective reaction heat flow should be expected to rise exponentially and drop quickly once the reaction reaches completion. This behavior should result in generating asymmetric DSC peaks such as those measured experimentally for the reaction under study (Figure 3B). , g g p ( g ure 6), which constitutes the zero-order kinetics. 4. Results and Discussion The heat ﬂow generated by such a pro- cess of conversion A to P will also give rise to the zero-order reaction rate law. Although presently the mechanism of the liquid-state polymerization of diacetylenes is virtually unknown, it is not unreasonable to expect that this two-step reaction model is applicable f the polymerization is catalyzed by some impurities, e.g., trace amounts of copper-based catalyst used in diol synthesis or water. The presence of the laĴer one is detected by 1H NMR spectroscopy in the initial monomer sample (Figure S1). Such catalysis is not un- common, for example, in the cyanate esters polymerization that is known to be catalyzed by residual phenols or traces of water [32]. p p y y g A simple reaction scheme that can explain the zero-order kinetics is as follows. Let us consider a two-step process that involves the formation of intermediate I via a reaction of reactant A with catalyst B followed by conversion of I into product P (Figure 6). When the concentration of B is much lower than that of A, and k1 >> k2, then the product of A and B concentrations as well as the concentration of I becomes practically constant in a wide range of times. In this case, the rates of the corresponding reactions also remain nearly constant, giving rise to a linear dependence of the concentrations of A and P on time (Figure 6), which constitutes the zero-order kinetics. The heat flow generated by such a process of conversion A to P will also give rise to the zero-order reaction rate law. Although presently the mechanism of the liquid-state polymerization of diacetylenes is virtually unknown, it is not unreasonable to expect that this two-step reaction model is applicable if the polymerization is catalyzed by some impurities, e.g., trace amounts of copper-based catalyst used in diol synthesis or water. The presence of the latter one is detected by 1H NMR spectroscopy in the initial monomer sample (Figure S1). Such catalysis is not uncommon, for example, in the cyanate esters polymerization that is known to be catalyzed by residual phenols or traces of water [32]. EW 8 of 12 Figure 6. Simulation of concentration proﬁles for catalytic transformation of reactant A through two consecutive reactions with 𝑘ଵ𝑘ଶ= 100 ⁄ and 𝐶଴(𝐴)/𝐶଴(𝐵) = 100 . The results obtained with the online kinetics simulator [33]. 4. Results and Discussion Important information about the rate-limiting step of the studied reaction can also be Figure 6. Simulation of concentration profiles for catalytic transformation of reactant A through two consecutive reactions with k1/k2 = 100 and C0(A)/C0(B) = 100. The results obtained with the online kinetics simulator [33]. Figure 6. Simulation of concentration proﬁles for catalytic transformation of reactant A through two consecutive reactions with 𝑘ଵ𝑘ଶ= 100 ⁄ and 𝐶଴(𝐴)/𝐶଴(𝐵) = 100 . The results obtained with the online kinetics simulator [33]. Important information about the rate limiting step of the studied reaction can also be Figure 6. Simulation of concentration profiles for catalytic transformation of reactant A through two consecutive reactions with k1/k2 = 100 and C0(A)/C0(B) = 100. The results obtained with the online kinetics simulator [33]. Figure 6. Simulation of concentration proﬁles for catalytic transformation of reactant A through two consecutive reactions with 𝑘ଵ𝑘ଶ= 100 ⁄ and 𝐶଴(𝐴)/𝐶଴(𝐵) = 100 . The results obtained with the online kinetics simulator [33]. Figure 6. Simulation of concentration profiles for catalytic transformation of reactant A through two consecutive reactions with k1/k2 = 100 and C0(A)/C0(B) = 100. The results obtained with the online kinetics simulator [33]. Polymers 2024, 16, 7 8 of 12 Important information about the rate-limiting step of the studied reaction can also be obtained from the analysis of its activation entropy (∆S̸=) [34]. According to the activated complex theory [35] for a reaction, proceeding in a solution ∆S̸= can be calculated as ∆S̸= = R ln Aih kBT −1 (9) (9) where h and kB represent, respectively, the Planck and Boltzmann constants, and Ai is an intrinsic value of the pre-exponential factor. Considering the melt of the reacting monomer as a solution in itself, we can apply Equation (9) to the studied reaction. Since the intrinsic value of the pre-exponential factor is unknown, we have to use the effective one reported earlier. This yields the ∆S̸= value of −78 ± 3 J K−1 mol−1. One should be aware that the effective nature of the pre-exponential factor propagates into the calculated activation entropy as well [34]. Nevertheless, it is important to notice that the activation entropy has a sufficiently large negative value. It means that a transition state formed in the rate- limiting step is more ordered than the reactants, from which it is formed. This is typically found in bimolecular reactions [36]. 4. Results and Discussion It can also be expected in other reactions, whose rate- limiting step has an associative nature. It is worth noting that we have determined negative values of the activation entropy for the reactions of polymerization [37] and 1,3-dipolar cycloaddition [38]. The reaction product of liquid-state polymerization is a black brittle material insoluble in usual organic solvents. Its aqueous dispersion is strongly acidic, apparently, due to the presence of p-toluenesulfonic acid formed during the reaction. The presence of the p-toluenesulfonic acid among the products of the liquid-state reaction is also confirmed by FTIR (Figures S3–S5). The initial monomer features an absorption band at 1360 cm−1 that corresponds to the stretching vibration of the S=O bond in its sulfonate group. Upon polymerization, this band transforms into the 1175 cm−1 absorption band corresponding to the hydrated sulfonic acid group of p-toluenesulfonic acid [39]. Complete disappearance of the 1360 cm−1 absorption band in the liquid-state polymerization product signifies the quantitative nature of elimination of the p-toluenesulfonate group. Most likely, this liquid-state reaction is initiated by residual water detected in the initial monomer sample (Figure S1). It is remarkable that despite the fact that residual water is also present in solid- state polymerization, this process does not involve elimination of the p-toluenesulfonate group as evidenced by the presence of absorption at 1360 cm−1 and the absence of the 1175 cm−1 absorption band in the resulting polymerization product. This observation is consistent with the fact that in the solid state the reaction proceeds via 1,4-addition without involving the side-groups [16]. The 1175 cm−1 absorption band also does not appear after subjecting this product to the same nonisothermal heating schedule as the one used for liquid-state polymerization. Evidently, the exclusive formation of p-toluenesulfonic acid during liquid-state polymerization is caused by the mechanistic features of polymerization rather than thermal instability of the monomer. X-ray powder diffraction analysis reveals the semicrystalline nature of the liquid-state reaction product, whereas solid-state poly- merization results in a highly crystalline polymer (Figures S6 and S7). It is important to note that a part of the XRD patterns of the liquid-state reaction product coincides with those of crystalline p-toluenesulfonic acid hydrate (Figure S7). The acid is highly hygro- scopic, so it readily forms a hydrate upon exposure to the atmospheric humidity. 4. Results and Discussion Indeed, a single absorption line with a g-factor of 2.00 and a width of 5 Gauss appears in the EPR spectra (Figure 7) during isothermal measurements in the temperature range of 130–145 °C, which corresponds to the initial stages of the liquid- state polymerization (see Figure 3). The intensity of the detected EPR line increases with time. Also, the radical generation rate increases with increasing temperature. It should be noted that no signals were observed in the EPR spectra of the initial monomer. Upon cool- ing, the intensity of the absorption line of the reaction mixture remains nearly the same even after a few days of storage at room temperature. It means that the radicals formed are stable and the process of their formation is irreversible. Their concentration in the re- action mixture has been estimated to be around (6.67 ± 0.05) × 1018 spin g−1. This suggests that roughly one of two hundred monomeric units in the polymer bears a radical center. The radical centers, apparently, are embedded into the polymer matrix because their con- centration remains almost the same after extraction of low molecular product by tert-bu- tanol. Figure 7. Evolution of the EPR spectra during measurement at 130 °C. Measurement interval is ~15 min Figure 7. Evolution of the EPR spectra during measurement at 130 ◦C. Measurement interval is ~15 min. Figure 7. Evolution of the EPR spectra during measurement at 130 °C. Measurement interval is ~15 min Figure 7. Evolution of the EPR spectra during measurement at 130 ◦C. Measurement interval is ~15 min. Remarkably, the intensity of the EPR signal measured at different temperatures in- creases linearly with time up to 0.85 of relative intensity (Figure 8A). This means that just as the heat release rate does, the rate of generation of free radicals follows the zero-order kinetics (Equations (10) and (11)). dI dI dt = k(T) (10) I = k(T)t (11) (10) dt I = k(T)t (11) (11) The temperature dependence of the generation rate constant k(T) calculated from the linear part of the data presented in Figure 8A follows the Arrhenius equation (Figure 8B) and gives rise to the activation energy of 105 ± 6 kJ mol−1. The corresponding value of the natural logarithm of the pre-exponential factor A is 25.6 ± 1.9 (A in min−1) or 21.5 ± 1.9 (A in s−1). 4. Results and Discussion The presence of the acid in combination with the less than 2.5 % mass loss during polymeriza- tion (Figure S8) indicates that nearly all p-toluenesulfonic acid remains entrapped in the polymerization product. A proposed mechanism of the solid-state polymerization of diacetylenes includes the formation of certain reactive radical species [12,40]. Therefore, to obtain further insights into the liquid-state reaction mechanism, we have applied in situ EPR spectroscopy. Although, the lifetime of the respective radicals is too short to be detected even at ambient temperature. For example, the half-life of cumulene-type biradicals initially formed in the solid-state reaction does not exceed 5 h at −183 ◦C) [12]. Of course, it is unrealistic to detect such Polymers 2024, 16, 7 9 of 12 d in the detect radicals by in situ EPR in the liquid-state polymerization of diacetylenes that occurs on the scale of hours above 120 ◦C. However, we have assumed that such reactive species may participate in the radical-transfer processes with the formation of more stable secondary radical centers. Indeed, a single absorption line with a g-factor of 2.00 and a width of 5 Gauss appears in the EPR spectra (Figure 7) during isothermal measurements in the temperature range of 130–145 ◦C, which corresponds to the initial stages of the liquid-state polymerization (see Figure 3). The intensity of the detected EPR line increases with time. Also, the radical generation rate increases with increasing temperature. It should be noted that no signals were observed in the EPR spectra of the initial monomer. Upon cooling, the intensity of the absorption line of the reaction mixture remains nearly the same even after a few days of storage at room temperature. It means that the radicals formed are stable and the process of their formation is irreversible. Their concentration in the reaction mixture has been estimated to be around (6.67 ± 0.05) × 1018 spin g−1. This suggests that roughly one of two hundred monomeric units in the polymer bears a radical center. The radical centers, apparently, are embedded into the polymer matrix because their concentration remains almost the same after extraction of low molecular product by tert-butanol. on the scale of hours above 120 °C. However, we have assumed that such reactive species may participate in the radical-transfer processes with the formation of more stable sec- ondary radical centers. Clearly the 5. Conclusions Clearly, the polymerization kinetics determined from the heat release data as meas ured by DSC and from the free radical generation process as measured by EPR are strik- ingly similar. The activation energy is 106 ± 2 vs. 105 ± 6 kJ mol−1 and the natural logarith- mic value of the pre-exponential factor is 22.3 ± 0.4 vs. 21.5 ± 1.9 (Aα in s−1). In addition, both kinetics follow the zero-order reaction model. Apparently, either the radical genera- tion process limits the overall rate of polymerization and, thus, the heat release rate or both processes have the same rate-limiting step. 5. Conclusions For the ﬁrst time, a detailed investigation of the liquid-state polymerization of di- acetylenes has been performed. The thermally initiated liquid-state polymerization of 2,4- hexadiyne-1,6-diyl bis-(p-toluenesulfonate) demonstrates the thermodynamic and kinetic parameters that diﬀer signiﬁcantly from those for the solid-state reaction. In particular, the reaction heat values for liquid- and solid-state reactions diﬀer by 2.8 times (360 ± 5 vs. 128 ± 4 kJ mol−1). Isoconversional kinetic analysis applied to the calorimetric data reveals that liquid-state polymerization occurs via a well-deﬁned rate-limiting step characterized by the eﬀective activation energy of 106 ± 2 kJ mol−1, whereas the eﬀective activation en- ergy of the solid-state process varies in a wide range, suggesting that the corresponding rate is determined by more than one step. Also, the liquid-state polymerization follows a h l d d l h h l d d For the first time, a detailed investigation of the liquid-state polymerization of di- acetylenes has been performed. The thermally initiated liquid-state polymerization of 2,4-hexadiyne-1,6-diyl bis-(p-toluenesulfonate) demonstrates the thermodynamic and ki- netic parameters that differ significantly from those for the solid-state reaction. In particular, the reaction heat values for liquid- and solid-state reactions differ by 2.8 times (360 ± 5 vs. 128 ± 4 kJ mol−1). Isoconversional kinetic analysis applied to the calorimetric data reveals that liquid-state polymerization occurs via a well-defined rate-limiting step characterized by the effective activation energy of 106 ± 2 kJ mol−1, whereas the effective activation energy of the solid-state process varies in a wide range, suggesting that the corresponding rate is determined by more than one step. Also, the liquid-state polymerization follows a rather unusual zero-order reaction model, whereas the solid-state reaction demonstrates autocatalytic behavior. 4. Results and Discussion ( ) Clearly, the polymerization kinetics determined from the heat release data as measured by DSC and from the free radical generation process as measured by EPR are strikingly similar. The activation energy is 106 ± 2 vs. 105 ± 6 kJ mol−1 and the natural logarithmic value of the pre-exponential factor is 22.3 ± 0.4 vs. 21.5 ± 1.9 (Aα in s−1). In addition, both kinetics follow the zero-order reaction model. Apparently, either the radical generation process limits the overall rate of polymerization and, thus, the heat release rate or both processes have the same rate-limiting step. Polymers 2024, 16, 7 10 of 12 ± 1.9 (A Figure 8. Normalized intensity of EPR signal vs. time at diﬀerent temperatures (A) and temperature dependence of the generation rate constant (B). Figure 8. Normalized intensity of EPR signal vs. time at different temperatures (A) and temperature dependence of the generation rate constant (B). Figure 8. Normalized intensity of EPR signal vs. time at diﬀerent temperatures (A) and temperature dependence of the generation rate constant (B). Figure 8. Normalized intensity of EPR signal vs. time at different temperatures (A) and temperature dependence of the generation rate constant (B). Clearly the 5. Conclusions The liquid-state reaction results in the quantitative elimination of the p-toluenesulfonate group and the formation of p-toluenesulfonic acid and a polymeric product with stable free radical centers. The similarity of the kinetic parameters derived from the DSC and in situ EPR data, as well as the adherence of the respective kinetics to the same zero-order reaction model, apparently, indicates that the overall rate of polymeriza- tion is either limited by the radical generation or both processes have the same rate-limiting step. Further studies are needed to see whether the discovered striking differences between liquid- and solid-state polymerization are characteristic for other diacetylene monomers. autocatalytic behavior. The liquid-state reaction results in the quantitative elimination of the p-toluenesulfonate group and the formation of p-toluenesulfonic acid and a polymeric Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/polym16010007/s1. product with stable free radical centers. The similarity of the kinetic parameters derived Author Contributions: Conceptualization, A.G.; methodology, A.G., S.V. and M.G.; investigation, A.G., A.K., A.R., G.M. and M.G.; writing—original draft preparation, A.G. and A.R.; writing—review and editing, A.G., M.G. and S.V. All authors have read and agreed to the published version of the manuscript. Funding: This work was funded by the subsidy allocated to Kazan Federal University for the state assignment in the sphere of scientific activities, No. FZSM-2023-0020. The EPR measurements (G.M. and A.R.) were financially supported by the Ministry of Science and Higher Education of the Russian Federation, Agreement No. 075-10-2021-115 (15.SIN.21.0021). Institutional Review Board Statement: Not applicable. Institutional Review Board Statement: Not applicable. Data Availability Statement: The data presented in this study are available on request from the corresponding authors. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest. Polymers 2024, 16, 7 11 of 12 References 1. Qian, X.; Städler, B. Recent Developments in Polydiacetylene-Based Sensors. Chem. Mater. 2019, 31 1. Qian, X.; Städler, B. Recent Developments in Polydiacetylene-Based Sensors. Chem. Mater. 2019, 31, 1196–1222. [CrossRef] 2. Park, D.-H.; Heo, J.-M.; Jeong, W.; Yoo, Y.H.; Park, B.J.; Kim, J.-M. Smartphone-Based VOC Sensor Using Colorimetric Polydi- acetylenes. ACS Appl. Mater. Interfaces 2018, 10, 5014–5021. [CrossRef] [PubMed] Q p y y [ ] 2. Park, D.-H.; Heo, J.-M.; Jeong, W.; Yoo, Y.H.; Park, B.J.; Kim, J.-M. Smartphone-Based VOC Sensor Using Colorimetric Polydi- acetylenes. ACS Appl. Mater. Interfaces 2018, 10, 5014–5021. [CrossRef] [PubMed] 3. Park, D.-H.; Kim, B.; Kim, J.-M. A Tetrahydrofuran-selective Optical Solvent Sensor Based on Solvatochromic Polydiacetylene. Bull. Korean Chem. Soc. 2016, 37, 793–794. [CrossRef] 4. Wang, X.; Sun, X.; Hu, P.A.; Zhang, J.; Wang, L.; Feng, W.; Lei, S.; Yang, B.; Cao, W. Colorimetric Sensor Based on Self-Assembled Polydiacetylene/Graphene-Stacked Composite Film for Vapor-Phase Volatile Organic Compounds. Adv. Funct. Mater. 2013, 23, 6044–6050. [CrossRef] 5. Chance, R.R.; Baughman, R.H.; Müller, H.; Eckhardt, C.J. Thermochromism in a polydiacetylene crystal. J. Chem. Phys. 2008, 67, 3616–3618. [CrossRef] 6. Charoenthai, N.; Pattanatornchai, T.; Wacharasindhu, S.; Sukwattanasinitt, M.; Traiphol, R. Roles of head group architecture and side chain length on colorimetric response of polydiacetylene vesicles to temperature, ethanol and pH. J. Colloid Interface Sci. 2011, 360, 565–573. [CrossRef] [PubMed] 7. Lu, J.; Zhou, J.; Li, J. Tuned chromic process for polydiacetylenes vesicles: The influence of polymer matrices. Soft Matter 2011, 7, 6529–6531. [CrossRef] 8. Tomioka, Y.; Tanaka, N.; Imazeki, S. Surface-pressure-induced reversible color change of a polydiacetylene monolayer at a gas–water interface. J. Chem. Phys. 1989, 91, 5694–5700. [CrossRef] g 9. Carpick, R.W.; Sasaki, D.Y.; Burns, A.R. First Observation of Mechanochromism at the Nanometer S 1270–1278. [CrossRef] 10. Feng, H.; Lu, J.; Li, J.; Tsow, F.; Forzani, E.; Tao, N. Hybrid Mechanoresponsive Polymer Wires Under Force Activation. Adv. Mater. 2013, 25, 1729–1733. [CrossRef] 11. Enkelmann, V. Structural aspects of the topochemical polymerization of diacetylenes. Polydiacetylenes; Springer: Berlin/Heidelberg, Germany, 1984; pp. 91–136. g y pp 12. Kollmar, C.; Sixl, H. Reaction kinetics of the solid state polymerization in diacetylene crystals. J. Chem. Phys. 1987, 87, 5541–5553. [CrossRef] 13. McGhie, A.R.; Kalyanaraman, P.S.; Garito, A.F. Solid State Polymerization of a Diacetylene: 2,4-Hexadiyne-1,6-Diol bis (p- Toluenesulfonate) (PTS). Mol. Cryst. Liq. Cryst. 1979, 50, 287–299. [CrossRef] y q y 14. Njus, J.M.; Sandman, D.J. Solid-state NMR and the thermal polymerization of 2,4-hexadiyne-1,6-diol bis-(p-toluenesulfonate). References Solid State Nucl. Magn. Reson. 2006, 29, 251–257. [CrossRef] [PubMed] g Solid-state reaction kinetics in single-phase polymerizations. J. Chem. Phys. 1978, 68, 3110–3121. [CrossRef Baughman, R.H. Solid-state reaction kinetics in single-phase polymerizations. J. Chem. Phys. 1978, 68, 3110 16. Patel, G.N.; Chance, R.R.; Turi, E.A.; Khanna, Y.P. Energetics and mechanism of the solid-state polymerization of diacetylenes. J. Am. Chem. Soc. 1978, 100, 6644–6649. [CrossRef] 17. Chance, R.R.; Sowa, J.M. An examination of the thermal polymerization of a crystalline diacetylene using diffuse reflectance spectroscopy. J. Am. Chem. Soc. 1977, 99, 6703–6708. [CrossRef] 18. Tsibouklis, J.; Werninck, A.R.; Shand, A.J.; Milburn, G.H.W. The Synthesis, Polymerization and Second Order Non-Linear Optical Properties of Novel, Conjugated, Unsymmetrical Diacetylenes Showing Liquid Crystal Behaviour. Liq. Cryst. 1988, 3, 1393–1400. [CrossRef] 19. Schen, M.A.; Kotowski, K.; Cline, J. Rigid-rod derived amorphous polydiacetylenes. Polymer 1991, 32, 20. Fomina, L.; Allier, H.; Fomine, S.; Salcedo, R.; Ogawa, T. Synthesis and Molten-State Polymerization of Some Novel Conjugated Diacetylenes. Polym. J. 1995, 27, 591–600. [CrossRef] y y 21. Chance, R.R.; Patel, G.N. Solid-state polymerization of a diacetylene crystal: Thermal, ultraviolet, and γ-ray polymerization of 2,4-hexadiyne-1,6-diol bis-(p-toluene sulfonate). J. Polym. Sci. Polym. Phys. Ed. 1978, 16, 859–881. [CrossRef] 22. Vyazovkin, S.; Burnham, A.K.; Criado, J.M.; Perez-Maqueda, L.A.; Popescu, C.; Sbirrazzuoli, N. ICTAC Kinetics Committee recommendations for performing kinetic computations on thermal analysis data. Thermochim. Acta 2011, 520, 1–19. [CrossRef] p g p y 23. Li, J.J. Name Reactions: A Collection of Detailed Reaction Mechanisms; Springer: Berlin/Heidelberg, Germa f p g g y 24. Vyazovkin, S. Modification of the integral isoconversional method to account for variation in the activation energy. J. Comput. Chem. 2001, 22, 178–183. [CrossRef]i 25. Vyazovkin, S.; Wight, C.A. Estimating Realistic Confidence Intervals for the Activation Energy Determined from Thermoanalytical Measurements. Anal. Chem. 2000, 72, 3171–3175. [CrossRef] [PubMed] 26. Liavitskaya, T.; Birx, L.; Vyazovkin, S. Thermal Stability of Malonic Acid Dissolved in Poly(vinylpyrrolid 26. Liavitskaya, T.; Birx, L.; Vyazovkin, S. Thermal Stability of Malonic Acid Matrices. Ind. Eng. Chem. Res. 2018, 57, 5228–5233. [CrossRef] 26. Liavitskaya, T.; Birx, L.; Vyazovkin, S. Thermal Stability of Malonic Acid Dissolved in Poly(vinylpyrrolidone) and Other Polymeric Matrices. Ind. Eng. Chem. Res. 2018, 57, 5228–5233. [CrossRef] y , ; , ; y , y Matrices. Ind. Eng. Chem. Res. 2018, 57, 5228–5233. [CrossRef] 27. Friedman, H.L. Kinetics of thermal degradation of char-forming plastics from thermograv plastic. J. Polym. Sci. Part C Polym. Symp. References 1964, 6, 183–195. [CrossRef] p y y y p 28. Vyazovkin, S.; Achilias, D.; Fernandez-Francos, X.; Galukhin, A.; Sbirrazzuoli, N. ICTAC Kinetics Committee recommendations for analysis of thermal polymerization kinetics. Thermochim. Acta 2022, 714, 179243. [CrossRef] y p y , , [ ] 29. Hoye, T.R.; Baire, B.; Niu, D.; Willoughby, P.H.; Woods, B.P. The hexadehydro-Diels–Alder reaction. Nature 2012, 490, 208–212. [CrossRef] y p y 29. Hoye, T.R.; Baire, B.; Niu, D.; Willoughby, P.H.; Woods, B.P. The hexadehydro-Diels–Alder reaction. Nature 2012, 490, 208–212. [CrossRef] Polymers 2024, 16, 7 12 of 12 30. Vyazovkin, S. Isoconversional Kinetics of Thermally Stimulated Processes; Springer International Publishing: Berlin/Heidelberg, Germany, 2015; p. 236. 31. Galukhin, A.; Nikolaev, I.; Nosov, R.; Vyazovkin, S. Solid-state polymerization of a novel cyanate ester based on 4-tert- butylcalix[6]arene. Polym. Chem. 2020, 11, 4115–4123. [CrossRef] 32. Galukhin, A.; Vyazovkin, S. Cyanate Ester Polymerization: Overview of Mechanistic and Kinetic Aspects. Macromol. Chem. Phys. 2023, 224, 2200384. [CrossRef] Kinetics Simulator. Available online: https://www.stolaf.edu/depts/chemistry/courses/toolkits/126/js/ on 27 October 2023). 33. Mechanism-Based Kinetics Simulator. Available online: https://www.stolaf.edu/depts/chemistr kinetics/ (accessed on 27 October 2023). 34. Vyazovkin, S. Determining Preexponential Factor in Model-Free Kinetic Methods: How and Why? Molecules 2021, 26, 3077. [CrossRef] 35. Glasstone, S.; Laidler, K.; Eyring, H. The Theory of Rate Processes; Hammet, L., Ed.; McGRAW-Hizll Book Company, Inc.: New York, NY, USA; London, UK, 1941; p. 611. p 36. Benson, S. The Foundations of Chemical Kinetics, 1st ed.; McGraw Hill Book Company Inc.: New York, N p oundations of Chemical Kinetics, 1st ed.; McGraw Hill Book Company Inc.: New York, NY, USA, 1960. 36. Benson, S. The Foundations of Chemical Kinetics, 1st ed.; McGraw Hill Book Company Inc.: New York, NY, USA, 1960. 37. Galukhin, A.; Nikolaev, I.; Nosov, R.; Islamov, D.; Vyazovkin, S. Solvent-induced changes in the reactivity of tricyanate esters undergoing thermal polymerization. Polym. Chem. 2021, 12, 6179–6187. [CrossRef] 37. Galukhin, A.; Nikolaev, I.; Nosov, R.; Islamov, D.; Vyazovkin, S. Solvent-induced changes in the reac undergoing thermal polymerization. Polym. Chem. 2021, 12, 6179–6187. [CrossRef] g g p y y 38. Galukhin, A.; Aleshin, R.; Nosov, R.; Vyazovkin, S. Probing kinetic and mechanistic features of bulk azide–alkyne cycloaddition. Phys. Chem. Chem. Phys. 2023, 25, 10671–10677. [CrossRef] [PubMed] 38. Galukhin, A.; Aleshin, R.; Nosov, R.; Vyazovkin, S. Probing kinetic and m Phys. Chem. Chem. Phys. 2023, 25, 10671–10677. [CrossRef] [PubMed] y y , , [ ] [ ] 39. Salami, M.; Ezabadi, A. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. References A caffeine-based ionic liquid as a novel and eco-friendly catalyst for the synthesis of 1,8-dioxo- octahydroxanthenes under solvent-free conditions. Res. Chem. Intermed. 2019, 45, 3673–3686. [CrossRef]l y y 39. Salami, M.; Ezabadi, A. A caffeine-based ionic liquid as a novel and eco-friendly catalyst for octahydroxanthenes under solvent-free conditions. Res. Chem. Intermed. 2019, 45, 3673–3686. [Cros y 40. Warta, R.; Sixl, H. Optical absorption and fluorescence spectroscopy of stable diacetylene oligomer molecules. J. Chem. Phys. 1988, 88, 95–99. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://openalex.org/W4293447269	https://hal.science/hal-03766890/document	English	null	Rediscovery and redescription of Teneriffia quadripapillata Sig Thor (Acari: Trombidiformes: Teneriffiidae)	Acarologia	2,022	cc-by	5,040	To cite this version: Edward A Ueckermann, Juan Carlos de la Paz, David Hernández-Teixidor, Furkan Durucan. Redis- covery and redescription of Teneriﬀia quadripapillata Sig Thor (Acari: Trombidiformes: Teneriﬀiidae). Acarologia, 2022, 62 (3), pp.786-797. 10.24349/3jom-1y9d. hal-03766890 Distributed under a Creative Commons Attribution 4.0 International License HAL Id: hal-03766890 https://hal.science/hal-03766890v1 Submitted on 1 Sep 2022 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. Distributed under a Creative Commons Attribution 4.0 International License Acarologia is proudly non-profit, with no page charges and free open access Please help us maintain this system by encouraging your institutes to subscribe to the print version of the journal and by sending us your high quality research on the Acari. Subscriptions: Year 2022 (Volume 62): 450 € http://www1.montpellier.inra.fr/CBGP/acarologia/subscribe.php Previous volumes (2010-2020): 250 € / year (4 issues) Acarologia, CBGP, CS 30016, 34988 MONTFERRIER-sur-LEZ Cedex, France ISSN 0044-586X (print), ISSN 2107-7207 (electronic) Acarologia A quarterly journal of acarology, since 1959 Publishing on all aspects of the Acari All information: http://www1.montpellier.inra.fr/CBGP/acarologia/ acarologia-contact@supagro.fr Acarologia is proudly non-profit, with no page charges and free open access Please help us maintain this system by encouraging your institutes to subscribe to the print version of the journal and by sending us your high quality research on the Acari. Subscriptions: Year 2022 (Volume 62): 450 € http://www1.montpellier.inra.fr/CBGP/acarologia/subscribe.php Previous volumes (2010-2020): 250 € / year (4 issues) Acarologia, CBGP, CS 30016, 34988 MONTFERRIER-sur-LEZ Cedex, France ISSN 0044-586X (print), ISSN 2107-7207 (electronic) Acarologia A quarterly journal of acarology, since 1959 Publishing on all aspects of the Acari All information: http://www1.montpellier.inra.fr/CBGP/acarologia/ acarologia-contact@supagro.fr How to cite this article Ueckermann E. A. et al. (2022), Rediscovery and redescription of Teneriffia quadripapil- lata Sig Thor (Acari: Trombidiformes: Teneriffiidae). Acarologia 62(3): 786-797. https://doi.org/10.24349/ 3jom-1y9d Acarologia is proudly non-profit, with no page charges and free open access Please help us maintain this system by encouraging your institutes to subscribe to the print version of the journal and by sending us your high quality research on the Acari. Subscriptions: Year 2022 (Volume 62): 450 € http://www1.montpellier.inra.fr/CBGP/acarologia/subscribe.php Previous volumes (2010-2020): 250 € / year (4 issues) Acarologia, CBGP, CS 30016, 34988 MONTFERRIER-sur-LEZ Cedex, France ISSN 0044-586X (print), ISSN 2107-7207 (electronic) A quarterly journal of acarology, since 1959 Publishing on all aspects of the Acari All information: http://www1.montpellier.inra.fr/CBGP/acarologia/ acarologia-contact@supagro.fr Acarologia is under free license and distributed under the terms of the Creative Commons-BY The digitalization of Acarologia papers prior to 2000 was supported by Agropolis Fondation under the reference ID 1500-024 through the « Investissements d’avenir » programme (Labex Agro: ANR-10-LABX-0001-01) Please help us maintain this system by encouraging your institutes to subscribe to the print version of the journal and by sending us your high quality research on the Acari. Acarologia is under free license and distributed under the terms of the Creative Commons-BY ABSTRACT Teneriffia quadripapillata Sig Thor, the type species of Teneriffia Sig Thor (1911), is redescribed from new material collected on Tenerife, the island which is the type locality in the Canary Islands. Sig Thor’s original material was apparently destroyed. Keywords littoral mites; predatory mites; teneriffiids; redescription Edward A. Ueckermann a , Juan Carlos De La Paz b , David Hernández-Teixidor b,c , Furkan Durucan d,e a Unit for Environmental Sciences and Management, Potchefstroom Campus, North-West University, Private Bag X6001, Potchefstroom, 2520, South Africa. b Grupo de Investigaciones Entomológicas de Tenerife (GIET), La Laguna, Tenerife, Canary Islands, Spain. c Island Ecology and Evolution Research Group, Instituto de Productos Naturales y Agrobiología (IPNA-CSIC), Astrofísico Fco. Sánchez no. 3, 38206 La Laguna, Tenerife, Canary Islands, Spain. d German Centre for Marine Biodiversity Research (DZMB), Senckenberg am Meer, Wilhelmshaven, Germany. c Island Ecology and Evolution Research Group, Instituto de Productos Naturales y Agrobiología (IPNA-CSIC), Astrofísico Fco. Sánchez no. 3, 38206 La Laguna, Tenerife, Canary Islands, Spain. d German Centre for Marine Biodiversity Research (DZMB), Senckenberg am Meer, Wilhelmshaven, Germany. e Department of Aquaculture, Isparta University of Applied Sciences, 32260 Isparta, Turkey. e Department of Aquaculture, Isparta University of Applied Sciences, 32260 Isparta, Turkey. Introduction Sig Thor (1911) described Teneriffia quadripapillata as the type species of his new genus, Teneriffia, from Tenerife, Canary Islands, Spain. Sig Thor was a controversial, stubborn person and received both praise and criticism. The Swedish arachnologist, Dr. Lundblad suggested that microscope slides of his type material must be deposited in a public museum. Unfortunately, this wish did not come true as Sig Thor apparently suffered from depression and had determined in his will that all his preparations were to be destroyed after his death. However, his widow donated some of his material, preserved in alcohol, to the zoological museum where he was a curator (Natvig 1944). Since then, no mention was made if this alcohol material contains T. quadripapillata. According to article 75.2 of the International Commission on Zoological Nomenclature, designation of a neotype is invalid when there is no doubt about the species’ identity; in this case the “clasp” organ is a unique character of the species. Received 03 May 2022 Accepted 11 August 2022 Published 29 August 2022 Corresponding author Edward A. Ueckermann : edalbert@lantic.net Academic editor Auger, Philippe https://doi.org/ 10.24349/3jom-1y9d ISSN 0044-586X (print) ISSN 2107-7207 (electronic) Ueckermann E. A. et al. Licensed under Creative Commons CC-BY 4.0 Received 03 May 2022 Accepted 11 August 2022 Published 29 August 2022 Corresponding author Edward A. Ueckermann : edalbert@lantic.net Academic editor Auger, Philippe https://doi.org/ 10.24349/3jom-1y9d ISSN 0044-586X (print) ISSN 2107-7207 (electronic) Ueckermann E. A. et al. Licensed under Creative Commons CC-BY 4.0 y p g q p Strandtmann (1965) came across unpublished figures with more detail of T. quadripapillata drawn by Oudemans based on the type specimens and he published it. A redescription is necessary to add more detailed information and to confirm the presence of the “clasp” organ. The genus Teneriffia currently consists of five known species: T. quadripapillata, T. mexicana McDaniel, T. sebahatae Ueckermann & Durucan, T. aethiopica Zmudzinski et al. and T. hajiqanbari Paktinat-Saeij & Kazemi (Sig Thor 1911; McDaniel et al. 1976; Ueckermann and Durucan 2020; Zmudzinski et al. 2021 and Paktinat-Saeij and Kazemi 2022). Teneriffia quadripapillata was collected for the first time on the beach in La Orotava, Tenerife, in August 1909 (Sig Thor, 1911). For more than 100 years this species remained undetected. Recently, new samplings carried out on Tenerife have revealed the presence of this species and led to its redescription. Strandtmann (1965) came across unpublished figures with more detail of T. Figure 1 Photo of Teneriffia quadripapillata Sig Thor. Figure 1 Photo of Teneriffia quadripapillata Sig Thor. Figure 1 Photo of Teneriffia quadripapillata Sig Thor. Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d Introduction quadripapillata drawn by Oudemans based on the type specimens and he published it. A redescription is necessary to add more detailed information and to confirm the presence of the “clasp” organ. The genus Teneriffia currently consists of five known species: T. quadripapillata, T. mexicana McDaniel, T. sebahatae Ueckermann & Durucan, T. aethiopica Zmudzinski et al. and T. hajiqanbari Paktinat-Saeij & Kazemi (Sig Thor 1911; McDaniel et al. 1976; Ueckermann and Durucan 2020; Zmudzinski et al. 2021 and Paktinat-Saeij and Kazemi 2022). Teneriffia quadripapillata was collected for the first time on the beach in La Orotava, Tenerife, in August 1909 (Sig Thor, 1911). For more than 100 years this species remained undetected. Recently, new samplings carried out on Tenerife have revealed the presence of this species and led to its redescription. Localities • Locality A: La Barranquera, La Laguna, Tenerife. Beach with rocks and sand (Figure 2); coordinates 28.537908°N, 16.396246°W; sampling dates 21/07/20, 29/07/20, 01/08/20 and 10/08/20; collected by Juan Carlos de la Paz. • Locality B: Roque de las Bodegas, Taganana, Santa Cruz de Tenerife. Beach with large rocks, boulders, pebbles and sand (Figure 2); coordinates 28.571219°N, 16.203836°W; sampling dates 24/07/20 and 25/07/20; collected by Juan Carlos de la Paz. • Locality B: Roque de las Bodegas, Taganana, Santa Cruz de Tenerife. Beach with large rocks, boulders, pebbles and sand (Figure 2); coordinates 28.571219°N, 16.203836°W; sampling dates 24/07/20 and 25/07/20; collected by Juan Carlos de la Paz. • Locality not used in the redescription: Las Eras, Fasnia, Tenerife, Canary Islands, Spain. Beach with pebbles and sand; coordinates 28.19373443°N, 16.42278712°W; sampling dates 03/10/20; collected by Juan Carlos de la Paz. Figure 2 Map of localities. Figure 2 Map of localities. Figure 2 Map of localities. Localities Sampling and location This study was carried out on Tenerife, the largest of the Canary Islands, Spain. It is situated near the centre of this volcanic archipelago comprising seven major islands, off the south-west Atlantic coast of Morocco. Specifically, the sampling area is situated in the northeast of the island. The mites (Figure 1) were collected under rocks, in the supralittoral zone, about a meter above the limit of the last high tide. They were captured one by one with a fine brush moistened in sea water, then put in a small container with sea water. Under a stereomicroscope (Leica ES2), specimens were manually extracted and transferred to 100% ethanol for storage. 787 eckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d 62(3): 786-797. Description Female — (n=6). Idiosoma ovate, 648–782 long (from tip of naso to posterior margin of opisthosoma) and 412–476 wide at level of setae c (Figure 3A). Dorsal idiosoma – Peritremes situated anterolaterally on propodosoma consisting of 9 long and 7 shorter branches (Figure 3B). Naso small, situated between setae vi (Figure 3A), the latter arising from cuplike bothridia. Prodorsal shield 323–348 long and 171–190 wide, demarcated by fine transverse striae behind setae vi and between setae ve, longitudinal on rest of shield, bearing setae vi, ve, sci and sce, setae vi smooth, ve and sce serrate and sci coarsely serrate, the latter also inserted into bothridia surrounded by 10–13 vesicles forming a “rosette″. Flanking the prodorsum, are two pairs of equal eyes. Opisthosoma transversely striated medially and diagonal laterally, all setae serrate, each situated on small sclerite; setae c2 the longest setae on idiosoma. Lengths of setae: vi 79–97, ve 86–99, sci 93–110, sce 108–119, c1 59–67, c2 150–174, d 68–78, e 73–78, f 87–108, h1 92–108, h2 55–63. Distances between setae: vi–vi 33–42, ve–ve 147–176, sci –sci 134–158, sce–sce 127–149, c1–c1 128–158, c2–c2 254–307, d–d 160–174, e–e 112–141, f–f 81–105, h1–h1 44–49, h2–h2 130–169. Four pairs of ordinary circular cupules present with ia lateral to setae d, im lateral to setae e and ip lateral to setae f, and ih ventrally, lateral to setae ps4. Ventral idiosoma – Venter striate with coxal fields I-IV finely striated (Figure 4A). Anal opening situated caudally, each anal valve with four pairs of pseudanal setae: ps1 43–56, ps2 37, v2–h1 54, ps3 36–54 and ps4 37–43. Coxal formula: 6 to 7–6 to 7–6–5. All coxal setae slightly barbed; 1f, 2g, 3a, and 4f longest setae. Lengths of coxal setae: 1a 57–68, 1b 27–34, 1c 28–41, 1d 28–43, 1e 32–45, 1f 90–100, 1g 57–68; 2a 27–33, 2b 25–31, 2c 24–28, 2d 27–33, 2e 26–31, 2f 33–46, 2g 54–62; 3a 63–78, 3b 27–30, 3c 25–30, 3d 23–28, 3e 44–55, 3f 27–34; 4a 24–27, 4b 19–26, 4c 21–27, 4d 22–26, 4e 24–28, 4f 77–82, 4g 26–30. Genital valves with 6 pairs of genital setae (g1-6), varying between 13–22. Genital opening with 5 pairs of eugenital setae and 3 papillae (Figure 4B). Genital opening surrounded by 20–25 pairs of setae, aggenital setae included, aggenital setae (ag1-6) varying between 24—38, showing asymmetry with 6 setae on one side and 7 on the other in some specimens. Diagnosis Differential diagnosis — The character that differentiates this species from the rest of the genus is the presence of “clasps” on the venter of the gnathosoma (Figures 5D and 6). See discussion under Remarks. 34644-34659) and the Arachnology Collection of ARC-Plant Health and Protection (ARC- PPRI) (Accession numbers, Acy: 22/381 to Acy: 22/386). 34644-34659) and the Arachnology Collection of ARC-Plant Health and Protection (ARC- PPRI) (Accession numbers, Acy: 22/381 to Acy: 22/386). Preparation of specimens Mites were cleared in 90% lactic acid and mounted on microscope slides in PVA. They were then dried in an oven at 45-50 degrees Celsius for 24 hours. Line drawings were made from photographs of the specimens taken with a Zeiss Axioskop TM Research microscope equipped with a Zen Soft Imaging System with measuring tools and an Axiocam 208 color camera. The line drawings were prepared with a Wacom One 13” Pen display and edited using Adobe Illustrator CS5. Measurements are given in micrometres (μm). The material will be deposited in the Zoological Collection of the Department of Animal Biology, Edaphology and Geology at the University of La Laguna (DZUL) (collection numbers Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d 788 Taxonomy Family Teneriffiidae Sig Thor, 1911 Genus Teneriffia Sig Thor, 1911 Type species: Teneriffia quadripapillata, Sig Thor, 1911: 173. Type species: Teneriffia quadripapillata, Sig Thor, 1911: 173. Description All these setae slightly barbed. Gnathosoma – Palp 5-segmented; tarsus reduced, bearing 3 long serrate and 4 shorter smooth setae (1 stout), 1 minute solenidion and 1 microseta (Figure 5B). Tibia with 1 long, strong spur, o1 56–64, 2 subterminal spurs (o 2–3), 17–21 and 17–20 long and one serrate seta. Genu and femur each with one long, serrate seta, 80–93 and 105–119 long, respectively, oncophysis absent. Palp supracoxal, seta 7 long. Chelicerae 186–219 long with movable digit 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d 789 Figure 3 Teneriffia quadripapillata Sig Thor. A Dorsal view; B Peritreme (enlarged). Figure 3 Teneriffia quadripapillata Sig Thor. A Dorsal view; B Peritreme (enlarged). 26–33 long and setae cha and chb 39–48 and 44–50 long, respectively (Figure 5A). Gnathosoma ventrally with subcapitular setae m 59–70 long and 3 pairs of adoral setae, or1 18–24, or2 10–17 and or3 34–41. The unique “clasp” organ of this species on the venter of the gnathosoma is only outlined in Figure 5C but clearly visible in Figures 5D and 6. It is an internal organ that is clearly visible before the treatment with lactic acid (Also see video in supplementary material). Legs – Leg IV clearly longer than body. Each side of claws on tarsi I–II with 18 pectinations, empodium absent (Figure 7A), tarsal claws III–IV smooth with a tiny peg-like empodium between them (Figure 7B). Lengths of leg segments: Ta I 172–193, Ti I 139–160, Ge I 98–125, TF I 86–113, BF I, 89–103, Tr I 72–100, Cx I 174–220; Ta II 159–180, Ti II 118–131, Ge II 74–102, TF II 72–91, BF II 65–91, Tr II 75–88, Cx II 170–198; Ta III 208–223, Ti III 117–145, Ge III 96–107, TF III 86–98, BF III 71–88, Tr III 72–103, Cx III 155–189; Ta IV 250–302, constricted before trichobothrium, Ti IV 164–183, Ge IV 120–134, TF IV 111–125, BF IV 75–91, Tr IV 109–123, Cx IV 163–186. Trichobothrium III 102–118 and trichobothrium 121–130 long. Setal formulae of leg segments: Tarsi 27 (3 ω)-27 (3 ω)- 22+ 1 trich. –22 (1 ω) Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d a 62(3): 786-797. https://doi.org/10.24349/3jom-1 790 Figure 4 Teneriffia quadripapillata Sig Thor. A Ventral view; B Genitalia (enlarged). Figure 4 Teneriffia quadripapillata Sig Thor. A Ventral view; B Genitalia (enlarged). Description + 1 trich (Figure 7C); tibiae 14 (1ϕ) (1κ)-12(1ϕ)-12(1ϕ)-11(1ϕ); genua 10(1σ)-8(1σ)-7(1σ)-6; telofemora 5-5-4-4; basifemora 5-6-4-4; trochanters 1-2-2-2; coxae 6 or 7-8-6-6. + 1 trich (Figure 7C); tibiae 14 (1ϕ) (1κ)-12(1ϕ)-12(1ϕ)-11(1ϕ); genua 10(1σ)-8(1σ)-7(1σ)-6; telofemora 5-5-4-4; basifemora 5-6-4-4; trochanters 1-2-2-2; coxae 6 or 7-8-6-6. Male — (n=5). Idiosoma elongate-oval, body length 640–705, width at level of setae c2, 340–422. Dorsal idiosoma – Prodorsal shield oval, 308–316 long and 175–178 wide. Similar to that of female. Lengths of setae: vi 89–102, ve 81–105, sci 100–127, sce 104–113, c1 56–63, c2 154–164, d 56–77, e 67–78, f 87–103, h1 90–100 and h2 55–64. Distances between setae: vi–vi 30–38, ve–ve 147–166, sci-sci 136–149, sce–sce 128–141, c1–c1 140–145, c2-c2 246–269, d–d 153–170, e–e 118–130, f–f 79–93, h1–h1 37–42, h2–h2 133–145. Four pairs of ordinary circular cupules present, with ia lateral to setae d, im lateral to setae e and ip lateral to setae f, and ih ventrally, lateral to setae ps4. Ventral idiosoma – Striate. With 19–28 pairs of serrate setae, including aggenitals. Coxisternal shields I–II separated from III–IV by a striate band. Anogenital area with 6 pairs of aggenital setae but more setae closely associated with aggenital setae, 6 pairs of genital setae and internal male genitalia with 2 pairs of eugenital setae easily detected (Figure 8A). Genital setae smooth and aggenital setae serrate. Lengths of coxal setae: 1a 51–60, 1b 25–34, 1c 25–31, 1d 31– 46, 1e 36–47, 1f 88–104, 1g 59–70, 2a 25–31, 2b 23–27, 2c 25–31, 2d 25–35, 2e 22–31, 2f 55–66, 2g 33– 44, 3a 64–68, 3b 23–31, 3c 25–29, 3d 22–29, 3e 45–62, 3f Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d 791 Figure 5 Teneriffia quadripapillata Sig Thor. A Chelicera; B Palptarsus; C Venter of gnathosoma; D Clasps. Figure 5 Teneriffia quadripapillata Sig Thor. A Chelicera; B Palptarsus; C Venter of gnathosom D Clasps. Figure 5 Teneriffia quadripapillata Sig Thor. A Chelicera; B Palptarsus; C Venter of gnathosoma; D Clasps. 28–31, 4a22–28, 4b 19–24, 4c 21, 4d 21, 4e 22–25, 4f 71–87, 4g 22–30, g1 17–20, g2 16–21, g3 16–22, g4 17–22, g5 16–22, g6 17–20, ag1 26–29, ag2 24–28, ag3 24–29, ag4 28–30, ag5 30–34, ag6 26–35, ag7 26–35, ps1 34–40, ps2 39–45, ps3 37–48, ps4 35–56, supracoxal seta 5–8. Description 28–31, 4a22–28, 4b 19–24, 4c 21, 4d 21, 4e 22–25, 4f 71–87, 4g 22–30, g1 17–20, g2 16–21, g3 16–22, g4 17–22, g5 16–22, g6 17–20, ag1 26–29, ag2 24–28, ag3 24–29, ag4 28–30, ag5 30–34, ag6 26–35, ag7 26–35, ps1 34–40, ps2 39–45, ps3 37–48, ps4 35–56, supracoxal seta 5–8. Gnathosoma – Palp 5-segmented; tarsus reduced with 3 serrate and 4 smooth setae, 1 microseta and 1minute solenidion. Tibia with 1 long, strong spur, o1 60–65, 2 subterminal spurs (o2–3), 18–21 long and 16–19, respectively, and 1 serrate seta. Genu and femur each with 1 long serrate seta, 84–94 and 109–122 long, respectively. Palp supracoxal seta 5 long. Oncophysis absent. Chelicerae 188–200 long with movable digit 21–28 long and setae cha and chb 40–55 and 42–51 long, respectively. Ventrally, gnathosoma with subcapitular setae m 56–67 long and 3 pairs of adoral setae, or 1 18–23, or2 12–17 and or 3 35–44. Outlines of Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d 792 Figure 6 Teneriffia quadripapillata Sig Thor. Photo of ”clasps”. Figure 6 Teneriffia quadripapillata Sig Thor. Photo of ”clasps”. Figure 6 Teneriffia quadripapillata Sig Thor. Photo of ”clasps”. clasp organ vaguely visible in male. Palp coxa with one supracoxal seta, 4 long. clasp organ vaguely visible in male. Palp coxa with one supracoxal seta, 4 long. clasp organ vaguely visible in male. Palp coxa with one supracoxal seta, 4 long. Legs – Leg IV clearly longer than body. Each side of claws on tarsi I–II with 17–18 pectinations, empodium absent, tarsal claws III–IV smooth with a tiny peg-like empodium between them. Lengths of leg segments: Ta I 162–182, Ti I 142–149, Ge I 95–100, TF I 67–99, BF I 84–100, Tr I 60–77, Cx I 165–216; Leg I 775–923: Ta II 157–178, Ti II 117–126, Ge II 82–89, TF II 56–88, BF II 77–87, Tr II 66–79, Cx II 172–185; Leg II 727–832: Ta III 215–226, Ti III 135–150, Ge III 90–103, TF III 74–93, BF III 68–76, Tr III 76–102, Cx III 159–177; Leg III 817–927: Ta IV 256–266, Ti IV 160–179, Ge IV 109–134, TF IV 91–119, BF IV 73–95, Tr IV 91–122, Cx IV 156–167; Leg IV 936–1082, IP 3255– 3764. Description Setal formulae of leg segments: Tarsi 27 (3 ω)-27 (3 ω)-22+ 1 trich-22 (1 ω) + 1 trich; tibiae 14 (1ϕ) (1κ)-12(1ϕ)-12(1ϕ)-11(1ϕ), genua 10(1σ)-8(1σ)-7(1σ)-6; telofemora 5-5-4-4; basifemora 5- 6-4-4; trochanters 1-2-2-2; coxae 6 or 5-8-6-6 or 5. Trichobothrium III 107–125 and trichobothrium IV 126–136. Nymph — (n=3). Idiosoma elongate-oval, body length 535–601, width at level of setae c2, 293–425. Dorsal idiosoma – Prodorsal shield oval, 249–258 long and 148–154 wide. Similar to that of female. Lengths of setae: vi 73–78, ve 79–84, sci 84–109, sce 98–109, c1 48–50, c2 135–146, d 60–63, e 68–75, f 85–89, h1 88–91 and h2 51. Distances between setae: vi–vi 27–32, ve-ve 131–138, sci–sci 118–124, sce–sce 104–117, c1–c1 119–130, c2-c2 212–272, d–d 138–150, e–e 104–122, f–f 65–74, h1–h1 35–36, h2–h2 127–143. Four pairs of ordinary circular cupules present with ia lateral to setae d, im lateral to setae Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d a 62(3): 786-797. https://doi.org/10.24349/3jom-1 793 Figure 7 Teneriffia quadripapillata Sig Thor. A Tarsal claw leg I; B Tarsal claw legs III-IV; C Trichobothrium, legs III-IV. Figure 7 Teneriffia quadripapillata Sig Thor. A Tarsal claw leg I; B Tarsal claw legs III-IV; C Trichobothrium, legs III-IV. e and ip lateral to setae f, and ih ventrally, lateral to setae ps4. e and ip lateral to setae f, and ih ventrally, lateral to setae ps4. Ventral idiosoma – Striate. With 11–12 pairs of serrate setae, aggenitals included. Cox- isternal shields I–II separated from III–IV by a striate band. Anogenital area with 2 (or 2 on one side and 3 on the other side) pairs of aggenital setae, 2 pairs of genital setae (Figure 8B). Genital setae smooth and aggenital setae serrate. Lengths of setae: 1a 46–51, 1b 24, 1c 23, 1d 31– 33, 1e 28–31, 1f 67–70, 1g 47–55, 2a 22–25, 2b 20–21, 2c absent, 2d 22–26, 2e 21–25, 2f 42–46, 2g 27– 31, 3a 53–57, 3b 21–22, 3c 20–22, 3d 21, 3e 32–38, 3f 21–28, 4a 17–20, 4b 17, 4c 18–19, 4d absent, 4e absent, 4f 43-56, 4g 21–23, g1 16–17, g2 15–17, ag1 19–23, ag2 20–26, ag3 23–30, ag4 26–31, ps1 34–40, ps2 32–34, ps3 31, ps4 33–34, supracoxal seta 6–7. Gnathosoma – Palp 5-segmented; tarsus reduced with 3 serrate and 4 smooth setae, 1 minute solenidion, 1 microseta. Description Tibia with 1 long, strong spur, o1 50–54, 2 subterminal spurs (o2–3), 14–16 long and 15–16, respectively, and 1 serrate seta. Genu and femur each with 1 long serrate seta, 50–65 and 97–103 long, respectively. Palp supracoxal seta 5 long. Oncophysis absent. Chelicerae 148–156 long with movable digit 20–22 long and setae cha and chb 36–38 and 33–37 long, respectively. Ventrally, gnathosoma with subcapitular setae m 49–53 long and 3 pairs of adoral setae, or1 15–16, or2 9–12 and or3 27–35. Clasp organ outline on venter of gnathosoma. Palp coxa with 1 supracoxal seta, 4 long. Legs – Leg IV clearly longer than body. Each side of claws on tarsi I–II with 17 pectinations, Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d 794 Figure 8 Teneriffia quadripapillata Sig Thor. A Genitalia of male; B Genital opening of nymph (enlarged). Figure 8 Teneriffia quadripapillata Sig Thor. A Genitalia of male; B Genital opening of nymph (enlarged). empodium absent, tarsal claws III–IV smooth with a tiny peg-like empodium between them. Lengths of leg segments: Ta I 133–140, Ti I 113–121, Ge I 82–87, TF I 73–84, BF I 76–91, Tr I 51–69, Cx I 145–186; Leg I 673–778: Ta II 129–131, Ti II 90–100, Ge II 71–75, TF II 76–70, BF II 70–74, Tr II 58–71, Cx II 133–157; Leg II 627–678: Ta III 165–170, Ti III 107–110, Ge III 73–81, TF III 67–72, BF III 52–72, Tr III 63–74, Cx III 137–148; Leg III 664–727: Ta IV 184–204, Ti IV 119–131, Ge IV 86–107, TF IV 57–72, BF IV 59–69, Tr IV 72–102, Cx IV 132–152; Leg IV 709–837, IP 2 673–3 020. Setal formulae of leg segments: Tarsi 23 (2 ω)-23 (2 ω)-19+ 1 trich-18 (1 ω) + 1 trich; tibiae 12 (1ϕ) (1κ)-12(1ϕ)-12(1ϕ)-11(1ϕ), genua 10(1σ)-8(1σ)-7(1σ)-6; telofemora 5-5-4-4; basifemora 5- 6-4-3; trochanters 1-2-2-2; coxae 5-6-5-4. Trichobothrium III 96–106 and trichobothrium IV 120–124. Cruz de Tenerife, from large rocks, boulders, pebbles and sand, coordinates 28.571219°N, 16.203836°W, 25/07/20, coll. Juan Carlos de la Paz. Cruz de Tenerife, from large rocks, boulders, pebbles and sand, coordinates 28.571219°N, 16.203836°W, 25/07/20, coll. Juan Carlos de la Paz. Cruz de Tenerife, from large rocks, boulders, pebbles and sand, coordinates 28.571219°N, 16.203836°W, 25/07/20, coll. Juan Carlos de la Paz. Discussion The redescription of T. quadripapillata allowed us to confirm in principle the presence of the unique diagnostic character within the Teneriffia genus (“clasps” on the venter of the gnathosoma). The unpublished figures of T. quadripapillata by Oudemans (Strandmann, 1965) showed more characters than the original figures of Sig Thor and was apparently the first to point out the “clasps″. In their revision of the Teneriffiidae McDaniel et al. (1976) recognised only two monotypic genera, namely Teneriffia and Parateneriffia Sig Thor, 1911 but recognised 3 species in the former and 5 in the latter. These authors described the field with “clasps” as follows: “8 transverse chitinous clasps for muscles of the pharynx” and that it has not been mentioned in any other species up to 1976. Judson (1994) in his description of Neoteneriffiola coineaui described the pharynx as large and with irregular transverse ridges but the figure bears no resemblance to that of T. quadripapillata. An examination of the types of T. sebahatae clearly confirms the absence of “clasps” in this species. Zmudzinski et al. (2021) also confirms the absence of “clasps” in their new species T. aethiopica but suggested further studies about the presence or absence of “clasps” in the other representatives of the Teneriffiidae. According to Pakinat-Saeij & Kazemi (2022) “clasps” are also absent in their new species T. hajiqanbari. In a video clip produced (supplementary material) by the second author it is clear that the “clasps” are not sclerotised but show movement almost in a pump-action. Furthermore, this sampling on Tenerife detected this species again after more than 100 years, and also widen its known distribution on the island. Edward A. Ueckermann https://orcid.org/0000-0003-4213-4309 Juan Carlos De La Paz https://orcid.org/0000-0001-8524-6381 David Hernández-Teixidor https://orcid.org/0000-0003-3114-7892 Furkan Durucan https://orcid.org/0000-0002-6168-2135 Edward A. Ueckermann https://orcid.org/0000-0003-4213-4309 Juan Carlos De La Paz https://orcid.org/0000-0001-8524-6381 David Hernández-Teixidor https://orcid.org/0000-0003-3114-7892 Furkan Durucan https://orcid.org/0000-0002-6168-2135 Acknowledgements The Cabildo of Tenerife provided collecting permits. DHT is currently funded by the Cabildo de Tenerife, under the TFinnova Programme supported by MEDI and FDCAN funds. The manuscript was edited by Guido Jones, also funded by the Cabildo de Tenerife under the same programme. This work is based on the research supported in part by the National Research Foundation of South Africa (Grant Number 126938). Any opinion, findings and conclusions or recommendations expressed in the material are those of the authors and therefore the NRF does not accept any liability in regard thereto. Material examined Three females, two males and one nymph from rocks and sand at La Barranquera, La Laguna, Tenerife, coordinates 28.537908°N, 16.396246°W, 21/07/2020, coll. Juan Carlos de la Paz; two females, two males and one nymph from rocks and sand at La Barranquera, La Laguna, Tenerife, coordinates 28.537908°N, 16.396246°W, 29/07/2020, coll. Juan Carlos de la Paz; two females, two males from rocks and sand at La Barranquera, La Laguna, Tenerife, coordinates 28.537908°N, 16.396246°W, 10/08/2020, coll. Juan Carlos de la Paz; one female, two males and one nymph from rocks and sand at La Barranquera, La Laguna, Tenerife, coordinates 28.537908°N, 16.396246°W, 01/08/2020, coll. Juan Carlos de la Paz; four females and three males, Roque de las Bodegas, Taganana, Santa Cruz de Tenerife, from large rocks, boulders, pebbles and sand, coordinates 28.571219°N, 16.203836°W, 24/07/20, coll. Juan Carlos de la Paz; three females, one male and one nymph, Roque de las Bodegas, Taganana, Santa Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d 795 Judson M. 1994. Studies on the morphology and systematics of the Teneriffiidae (Acari, Prostigmata). 1. A new species of Neoteneriffiola from Namibia. Acarologia, 35(2): 115-134. McDaniel B., Morihara D., Lewis J.K. 1976. The family Teneriffiidae Thor, with a new species from Mexico. Ann. Entomol. Soc. Am., 69(3): 527-537. https://doi.org/10.1093/aesa/69.3.527 Natvig L.R. 1944. Entomology at the Royal Frederik’s University. A contribution to the history of Zlorian entomology in the period 1813-1907. Nor. J. Entomol., 7(1-2): 1-73. Paktinat-Saeij S., Kazemi S. 2022. Teneriffia hajiqanbari sp. nov. (Acari: Trombidiformes: Teneriffiidae), first record of the genus from Iran, with a key to world species of Teneriffia. Acarologia, 62(1): 262- 269. https://doi.org/10.24349/jkco-c4nu Strandtmann R.W. 1965. Additional notes on Teneriffiidae (Acarina: Prostigmata) with two previously unpublished plates by A.C. Oudemans. J. Kans. Entomol. Soc., 38 (3): 258-261. Thor S. 1911. A new acarine family (Teneriffiidae) and two new genera, one from Tenerife, the other from Paraguay. Zool. Anz., 38 (7-8): 171-179. g y ( ) Ueckermann E.A., Durucan F. 2020. Teneriffia sebahatae sp.nov. (Acari: Trombidiformes: Teneriffiidae), the first teneriffiid mite from Turkey. Syst. Appl. Acarol., 25(6): 1139-1146. https://doi.org/10.11158/ saa.25.6.15 Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d References Judson M. 1994. Studies on the morphology and systematics of the Teneriffiidae (Acari, Prostigmata). 1. A new species of Neoteneriffiola from Namibia. Acarologia, 35(2): 115-134. Natvig L.R. 1944. Entomology at the Royal Frederik’s University. A contribution to the history of Zlorian entomology in the period 1813-1907. Nor. J. Entomol., 7(1-2): 1-73. gy p ( ) Paktinat-Saeij S., Kazemi S. 2022. Teneriffia hajiqanbari sp. nov. (Acari: Trombidiformes: Teneriffiidae), first record of the genus from Iran, with a key to world species of Teneriffia. Acarologia, 62(1): 262- 269. https://doi.org/10.24349/jkco-c4nu 269. https://doi.org/10.24349/jkco-c4nu Strandtmann R.W. 1965. Additional notes on Teneriffiidae (Acarina: Prostigmata) with two previously unpublished plates by A.C. Oudemans. J. Kans. Entomol. Soc., 38 (3): 258-261. Strandtmann R.W. 1965. Additional notes on Teneriffiidae (Acarina: Prostigmata) with two previously unpublished plates by A.C. Oudemans. J. Kans. Entomol. Soc., 38 (3): 258-261. Ueckermann E. A. et al. (2022), Acarologia 62(3): 786-797. https://doi.org/10.24349/3jom-1y9d 796 Zmudzinski M., Skoracki M., Friedrich S. 2021. A new species of Teneriffiidae (Acariformes: Prostig- mata) from Ethiopia. Int. J. Acarol., 47(4): 317-326. https://doi.org/10.1080/01647954.2021.1908422 Zmudzinski M., Skoracki M., Friedrich S. 2021. A new species of Teneriffiidae (Acariformes: Prostig- mata) from Ethiopia. Int. J. Acarol., 47(4): 317-326. https://doi.org/10.1080/01647954.2021.1908422 797
https://openalex.org/W1885450609	https://iris.unica.it/bitstream/11584/137089/1/A%20Novel_2015.pdf	English	null	A Novel High-Content Immunofluorescence Assay as a Tool to Identify at the Single Cell Level γ-Globin Inducing Compounds	PloS one	2,015	cc-by	8,184	RESEARCH ARTICLE OPEN ACCESS Citation: Durlak M, Fugazza C, Elangovan S, Marini MG, Marongiu MF, Moi P, et al. (2015) A Novel High- Content Immunofluorescence Assay as a Tool to Identify at the Single Cell Level γ-Globin Inducing Compounds. PLoS ONE 10(10): e0141083. doi:10.1371/journal.pone.0141083 Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This work was supported by the People Programme (Marie Curie Actions) of the European Union's Seventh Framework Programme FP7/2007- 2013/ under REA grant agreement n°289611 (HEM_ID Project) and by Fondazione Cariplo for grant n° 2012.0517 to AR. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. FG and IF are employees in Nerviano Medical Sciences S.r.l., which is a private sector partner in the HEM_ID A Novel High-Content Immunofluorescence Assay as a Tool to Identify at the Single Cell Level γ-Globin Inducing Compounds Marta Durlak1, Cristina Fugazza2, Sudharshan Elangovan2, Maria Giuseppina Marini3, Maria Franca Marongiu3, Paolo Moi4, Ivan Fraietta1, Paolo Cappella1¤, Gloria Barbarani2, Isaura Font-Monclus2, Mario Mauri5, Sergio Ottolenghi2, Fabio Gasparri1, Antonella Ronchi2* 1 Department of Biology, Nerviano Medical Sciences S.r.l., Nerviano, Milano, Italy, 2 Dipartimento di Biotecnologie e Bioscienze, Università degli studi di Milano-Bicocca, Milano, Italy, 3 Istituto di Ricerca Genetica e Biomedica del Consiglio Nazionale delle Ricerche, Cagliari, Italy, 4 Dipartimento di Sanità Pubblica, Medicina Clinica e Molecolare, Università degli Studi di Cagliari, Cagliari, Italy, 5 Dipartimento di Scienze della Salute, Università degli studi di Milano-Bicocca, Monza, Italy ¤ Current Address: FlowMetric Europe srl, Parco Tecnologico Padano, Lodi, Italy * antonella.ronchi@unimib.it ¤ Current Address: FlowMetric Europe srl, Parco Tecnologico Padano, Lodi, Italy * antonella.ronchi@unimib.it Abstract The identification of drugs capable of reactivating γ-globin to ameliorate β-thalassemia and Sickle Cell anemia is still a challenge, as available γ-globin inducers still have limited clinical indications. High-throughput screenings (HTS) aimed to identify new potentially therapeutic drugs require suitable first-step-screening methods combining the possibility to detect varia- tion in the γ/β globin ratio with the robustness of a cell line. We took advantage of a K562 cell line variant expressing β-globin (β-K562) to set up a new multiplexed high-content immunofluorescence assay for the quantification of γ- and β-globin content at single-cell level. The assay was validated by using the known globin inducers hemin, hydroxyurea and butyric acid and further tested in a pilot screening that confirmed HDACs as targets for γ- globin induction (as proved by siRNA-mediated HDAC3 knockdown and by treatment with HDACs inhibitors entinostat and dacinostat) and identified Heme-oxygenases as novel can- didate targets for γ-globin induction. Indeed, Heme-oxygenase2 siRNA knockdown as well as its inhibition by Tin protoporphyrin-IX (TinPPIX) greatly increased γ-globin expression. This result is particularly interesting as several metalloporphyrins have already been devel- oped for clinical uses and could be tested (alone or in combination with other drugs) to improve pharmacological γ-globin reactivation for the treatment of β-hemoglobinopathies. Editor: Andrew C. Wilber, Southern Illinois University School of Medicine, UNITED STATES Received: August 26, 2015 Accepted: October 5, 2015 Published: October 28, 2015 Copyright: © 2015 Durlak et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. A Novel Assay to Identify γ-Globin Inducing Compounds and infections in the developing countries [2–4]. SCA is caused by a missense mutation within the adult β-globin chain. Hemoglobin tetramers bearing this altered β chain (HbS) tend to polymerize within the Red Cell, under hypoxic conditions, conferring the typical sickle shape, leading to cell lysis, small vessel occlusion, pain crises and organ damage. In β-thalassemia, the reduced synthesis of β chains causes unbalanced accumulation of α-globin that precipitates, resulting in ineffective erythropoiesis and anemia [5]. Coinheritance of Hereditary Persistence of Fetal Hemoglobin (HPFH), a condition where the expression of the fetal HBG1/2 is main- tained postnatally, can ameliorate β-globinopathies, by reducing sickle hemoglobin polymers in SCA and the α/non-α chain imbalance in β-thalassemia[6]. This observation led to the intensive search for fetal hemoglobin (HbF) inducers that could mimic the beneficial effects observed in HPFH[7–9]. Genome-wide association studies identified three major gene loci (Xmn1-HBG2, HBS1L-MYB and BCL11A) accounting for the majority of inherited HbF vari- ance[10] but their exploitation as therapeutic targets is still distant. Another line of research focused on the development of drugs acting on γ-globin regulatory molecules: different classes of drugs (cytotoxic agents, HDAC inhibitors, DNA methyl transferase inhibitors) have been tested as HbF inducers but, despite the enormous effort in this direction and some encouraging results on some patients, no universal effective drugs have been found so far. Among them, hydroxyurea (HU) has been approved by the FDA for the treatment of SCA and has been recently considered for β-thalassemia, but its efficacy varies among patients. Indeed, about half of the patients do not reach therapeutic levels of HbF at HU doses of acceptable toxicity[11,12]. Other agents, such as short-chain fatty acids (Butyrate and its derivatives), 5-azacytabine, Deci- tabine and Tranylcypromine act on the epigenetic regulation of HbF, by inhibiting histones deacetylation or methylation of the HBG1/2, but their efficacy is still limited to a minority of patients[13–16]. Project (Marie Curie Actions of the European Union's Seventh Framework Programme FP7/2007-2013/ under REA grant agreement n°28961). The present research is the result of a collaborative project and MD is the Marie Curie fellow appointed to Nerviano Medical Sciences S.r.l. within this network. PC was a NMS employee at the time the research was performed (present address: FlowMetric Europe srl, Parco Tecnologico Padano, Via A. Einstein 26900, Lodi, Italy). Competing Interests: The authors have no competing interests related to this work. FG and IF are employees in Nerviano Medical Sciences S.r.l. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials. These observations point to the strong need to identify new compounds stimulating γ-glo- bin expression. With this goal in mind, we set up a high-content screening platform based on multiplexed imaging on a variant K562 cell line (β-K562) spontaneously expressing significant levels of β- globin. Simultaneous analysis of DNA content, adult hemoglobin HbA (α2β2) and fetal hemo- globin HbF (α2γ2) resulted in a robust and sensitive assay, capable of detecting changes at the single cell level in hemoglobinization and in γ/β ratio in response to drugs, as proved by the response of β-K562 to the known γ-globin inducers hemin, hydroxyurea and butyric acid and to two additional HDAC inhibitors: entinostat and dacinostat. The method was further validated by transfecting β-K562 with a panel of 70 siRNAs. Among them, we identified HMOX2, coding for Heme-oxygenase2 (HO-2), as a gene whose knockdown greatly increases γ globin levels, both in terms of percentage of expressing cells and of γ-globin accumulation per cell. Tin Protoporphyrin IX, a prototypical compound inhibiting HO-2, induced selective γ-globin accumulation in β-K562, suggesting that Heme-oxygenases could be a promising pharmacological target to ameliorate the α/β chains unbalance in β- hemoglobinopathies. Materials and Methods Cell lines and chemical treatments Introduction Sickle cell anemia (SCA) and β-thalassemia are among the commonest inherited diseases in humans, with more than 300,000 affected children born every year and with an estimated worldwide population of tens of millions patients suffering from these disorders [1]. The num- ber of these patients is increasing because of the decreased mortality from nutrition problems 1 / 14 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 NE \| DOI:10.1371/journal.pone.0141083 October 28, 20 Competing Interests: The authors have no competing interests related to this work. FG and IF are employees in Nerviano Medical Sciences S.r.l. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials. siRNA oligonucleotide transfections β-K562 cells were transfected with siRNA oligonucleotides (H-Silencer Select Druggable Genome siRNA Library V4, Ambion). A siRNA oligo targeting the proteasome subunit PSMC3 and a non-targeting oligo (siNTO) were used as positive and negative controls for transfection (siRNA sequences are listed in S2 Table). At least two siRNAs oligonucleotides per gene were transfected by using lipofectamine1 RNAiMAX (Invitrogen), as described in S1 File. Cell lines and chemical treatments ECACC-K562 (European Collection of Cell Cultures) and β-K562 (a kind gift of Prof. G. Fer- rari, HSR, Milano) were grown in standard conditions[17]. β-K562 were originally purchased from ATCC (CCL-243™). Doubling times were calculated on cells growing in exponential phase. β-K562 authentication was obtained by short tandem repeat fingerprinting (AmpFlSTR Identifiler Plus PCR Amplification kit -Applied Biosystems-), as described in [18]. For 2 / 14 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 A Novel Assay to Identify γ-Globin Inducing Compounds chemical treatments, 5 x 104 cells were exposed to increasing doses in 24-well plates. After four days, cells were analyzed by RTqPCR or high-content analysis. All experiments were per- formed in triplicate (at least two technical replicates per experiment). Chemicals and antibodies are listed in S1 Table. RNA Isolation and RT-PCR Total RNA from 106 cells was extracted with TRI Reagent (Applied Biosystems), treated with RQ1 DNase (Promega) for 30’ at 37°C and retrotranscribed (Applied Biosystems). Negative control reactions (RT-) gave no signal. Real time analysis was performed using ABI Prism 7500, (Applied Biosystems). Primers are listed in S3 Table. Immunofluorescence and high-content analysis Cells were collected and fixed in 3.7% paraformaldehyde for 20’ at RT, washed and permeabi- lized in staining buffer (PBS with 0.05% v/v Triton1 X-100 and 1% w/v powdered milk) for 30’. After washing in PBS, cells were incubated overnight at 4°C in staining buffer containing the appropriate antibodies and 1μg/ml Hoechst 33342. After washing, cells were resuspended in PBS and transferred to 96-well CELLSTAR1, Black/μClear1 plates (Greiner Bio-One). Plates were spun for 5’ at 2g to facilitate cell attachment, sealed and analyzed with the ArrayS- can VTI high-content screening reader (Thermo-Fisher Scientific). At least 600 cells were acquired in each well with a 20x magnification in three fluorescence channels (blue, green and red). The Molecular Translocation Bioapplication was used to determine the cell count per field, the nuclear area and intensity (based on the Hoechst staining in the blue channel) and the cytoplasmatic fluorescence intensity of β (green) and γ (red) globins. For simultaneous glo- bins/GlycophorinA staining, APC-anti-CD235 antibody was added for 2 hours to cells already stained for globins. Confocal microscopy K562 and β-K562 cells, stained as above, were transferred to a microscope glass slide and mounted with Mowiol (Sigma-Aldrich). Microphotographs were acquired with a confocal Zeiss microscope LSM710. Flow cytometry 106 cells were washed, fixed and permeabilized for 10’ on ice, then washed and incubated in PBS+1% milk for 20’. After washing, cells were stained overnight at 4°C in PBS+1% milk con- taining the appropriate antibodies. After washing, cells were analyzed with FACSCalibur (Bec- ton Dickinson). Identification and characterization of a K562 variant subclone expressing β-globin K562 are probably the most extensively used cellular model of a human “erythroid” cell. This line was established in 1975[19] from a patient with chronic myelogenous leukemia (CML) in blast crisis. K562 have been widely used to study the molecular regulation of embryonic and fetal globin genes and to assess the therapeutic potential of differentiation-inducing drugs[20]. However, the major limitation for their use in the study of the differential γ/β regulation is their “fetal-like” pattern of globin genes expression, since they exclusively express embryonic (HbGower-2, α2ε2) and fetal (HbF, α2γ2) hemoglobin[21,22]. While characterizing different K562 subclones, we came across a variant clone expressing the adult HBB, that we named β-K562. These cells are morphologically similar to ECACC (European Collection of Cell Culture) K562 cells, here considered as “prototypical” K562 cells (not shown). Moreover, K562 and β-K562 have a similar doubling time (S1A Fig) and are equally sensitive to drugs known to inhibit K562 proliferation: imatinib mesylate and dasatinib, two tyrosine kinase inhibitors targeting the bcr/abl fusion protein and doxorubicin, a DNA intercalating agent (S1B Fig). In addition, β-K562 fingerprinting characterization by using the STR AmpFlSTR Identifiler Plus, gave a profile substantially corresponding to K562 (Identity score[18] of 89.2%, not shown). Despite their bona fide “K562-like” profile, β-K562 do express the adult β-globin chain, as assessed by flow cytometry (FCM) analysis (S1C Fig). Based on this observation, we reasoned that the β-K562 subclone could be used to set up an immunofluorescence high-throughput, high-content screening platform to search for new genes/drugs modulating hemoglobinization and, in particular, the γ/β ratio. Statistical analysis Each experiment was statistically analyzed using a paired, two tailed Student-t-test. 3 / 14 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 A Novel Assay to Identify γ-Globin Inducing Compounds PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 Development of a multiplexed high-content assay for the quantification of γ- and β-globin content in β-K562 at the single-cell level The intensity value of signals is automatically assigned by the instrument and converted into a corresponding intensity of color. The relative scatter plots show the distribution of double γ-β- negative, single γ+β- positive, single γ-β+ positive and double γ+β+ positive cells (x axis: FITC-β- globin; y axis: PE-γ-globin). Numbers within plots refer to the averaged percentage of cells within each population from three independent experiments (n = 3). The relative st.errors are shown in panel C: p<0,05; * p<0,01; **p<0,001. B) Quantitative fluorescence imaging of single cells: cells numbered from 1 to 6 in panel A are taken as an example of γ-β- double negative (1 and 2), single γ+β- positive (5), single γ-β+ positive (4) and γ+β+ double positive (3 and 6). C) Statistical analysis (n = 3): γ-β- cells; red: γ+β- cells; yellow: γ+β+ cells; green: γ-β+ cells. D) RTqPCR on α, ε, γ- and β-globins. Histograms show the relative levels of expression normalized on glyceraldehyde-3-phosphate dehydrogenase (GAPDH). n3, statistical analysis: p<0,05; p<0,01; p<0,001. Fig 1. Analysis of γ/β globin levels by immunofluorescence and automated image capture. A) Image acquisition and analysis for β-K562 and K562. Merged signals of DNA (Hoechst-33342), β-globin and γ-globin are read in channel 1 (Ch1), channel 2 (Ch2) and channel 3 (Ch3), respectively (see also S1D Fig). Bar = 50μm. The intensity value of signals is automatically assigned by the instrument and converted into a corresponding intensity of color. The relative scatter plots show the distribution of double γ-β- negative, single γ+β- positive, single γ-β+ positive and double γ+β+ positive cells (x axis: FITC-β- globin; y axis: PE-γ-globin). Numbers within plots refer to the averaged percentage of cells within each population from three independent experiments (n = 3). The relative st.errors are shown in panel C: p<0,05; p<0,01; p<0,001. B) Quantitative fluorescence imaging of single cells: cells numbered from 1 to 6 in panel A are taken as an example of γ-β- double negative (1 and 2), single γ+β- positive (5), single γ-β+ positive (4) and γ+β+ double positive (3 and 6). C) Statistical analysis (n = 3): γ-β- cells; red: γ+β- cells; yellow: γ+β+ cells; green: γ-β+ cells. D) RTqPCR on α, ε, γ- and β-globins. Histograms show the relative levels of expression normalized on glyceraldehyde-3-phosphate dehydrogenase (GAPDH). n3, statistical analysis: p<0,05; p<0,01; *p<0,001. doi:10.1371/journal.pone.0141083.g001 expression of β-globin in β-K562 was further confirmed by RTqPCR (Fig 1D). Development of a multiplexed high-content assay for the quantification of γ- and β-globin content in β-K562 at the single-cell level 5x104 K562 or β-K562 were seeded in 24-well plates. Nuclei were stained with Hoechst-33342; γ- and β-globins were immunostained by using specific PE-anti γ and FITC-anti β-globin anti- bodies, respectively (S1D Fig). Cells were subsequently analyzed with an Array Scan VTI reader (Thermo-Fisher Scientific) and data were acquired and processed as shown in Fig 1A and 1B to obtain an automated and quantitative fluorescence imaging at a single cell level. The inten- sity of the staining is automatically converted in the corresponding intensity of colors: blue for Hoechst, green for β-globin and red for γ-globin. The detection threshold for the scoring of single γ+-, β+- and double β+γ+-cells was defined by using cells stained with the respective isotype controls (PE-IgG1 and FITC-IgG1, an example is shown in S1E Fig). When signals from the three single channels are merged (Fig 1A), the double expression of γ plus β results in an orange/yellow color of different intensity, depending on the amount of γ and β chains (see cells 3 and 6 in Fig 1B). This analysis allows measuring of both the percentage of single-positive (γ+ or β+) and of double positive (γ+β+) cells in each field. Moreover, the signal intensity per cell uncovers the intrinsic heterogeneity within the cell population. In a standard experiment, data are acquired from a minimum of 500 cells and plotted to give an immediate visual image of cell distribution with respect to globin accumulation per cell, as in Fig 1A, where a representative experiment is shown. The majority of K562 cells are γ+ (54.1% γ+β- + 3.1% γ+β+), the remaining being mostly γ-β-, with just a few marginally β+ cells (3.1% γ+β++0.7% γ-β+). In contrast, about 57% of β-K562 cells are γ+ (38.3% γ+β- + 18.7% γ+β+) and about 22% of cells are positive for β staining (18.7% γ+β+ + 3.6% γ-β+). Moreover, the Mean Fluo- rescence Intensity (MFI, Y axis) of γ+ cells is higher than in β-K562 (Fig 1A and S1C Fig). The PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 4 / 14 A Novel Assay to Identify γ-Globin Inducing Compounds Fig 1. Analysis of γ/β globin levels by immunofluorescence and automated image capture. A) Image acquisition and analysis for β-K562 and K562. Merged signals of DNA (Hoechst-33342), β-globin and γ-globin are read in channel 1 (Ch1), channel 2 (Ch2) and channel 3 (Ch3), respectively (see also S1D Fig). Bar = 50μm. Development of a multiplexed high-content assay for the quantification of γ- and β-globin content in β-K562 at the single-cell level Of interest, an intrinsic heterogeneity of the culture is present also for β signal: the large majority of β+ β -K562 cells also co-express γ-globin (18.7% of total cell population, corresponding to 85% of β+ cells, Fig 1C), whereas only few cells appear to be completely “switched” to β expression (3.6% of total cell population, corresponding to 15% of β+ cells, Fig 1C). Overall, these data confirm that β- K562 express β-globin and provide additional information on the population heterogeneity. PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 Validation of γ/β globin high-content assay by using the known hemoglobin inducers hydroxyurea and butyric acid To test the sensitivity of the method in detecting changes of γ- and β-globin levels, we treated β-K562 cells with the known Hemoglobin inducers, hydroxyurea (HU) and butyric acid (BA). We measured the response of β-K562 cells at different pharmacological concentrations of these drugs (S2 Fig) and we analyzed the same cell samples by both RTqPCR and immunofluores- cence. Fig 2 summarizes data relative to drug concentrations most commonly used in the litera- ture to induce hemoglobinization in K562. Both inducers, as expected, increase the hemoglobin content, reducing the number of dou- ble negative cells (Fig 2A and 2B). hydroxyurea induces both γ and β chains accumulation, as demonstrated by the increased percentage of γ+β+ double positive cells. Instead, butyric acid especially increases the percentage of γ+β- cells, suggesting a different mechanism of action for 5 / 14 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 A Novel Assay to Identify γ-Globin Inducing Compounds these compounds (Fig 2B). In addition to increasing the number of globins-positive cells, these drugs also strongly increase the proportion of highly fluorescent cells (S2B Fig) acting predom- inantly on γ-globin accumulation versus β-globin. At the mRNA level, both HU and BA stimulate γ expression by about 6–7 fold relative to untreated cells, whereas β-globin expression is essentially unchanged (Fig 2C). Finally, α-globin is moderately induced by both drugs, with BA eliciting the strongest increase (about four times). Overall, RTqPCR data confirm the effects observed at the protein levels, indicating the reliability of the assay. The visual microscopy analysis can provide further information on additional parameters at the single cell level, such as nuclear morphology (Fig 2D) and/or the expression of specific markers of interest. As an example, the quadruple staining of β-globin, γ-globin, nuclei and GlycophorinA shows increased levels of GlycophorinA in β-K562 exposed to HU (Fig 2D). Fig 2. High-content analysis of compound-induced changes in globins accumulation. β-K562 cells were treated with 800μM hydroxyurea and 900μ butyric acid (n = 3, a representative experiment is shown here) and the same cells were analyzed in parallel by immunofluorescence and by RTqPCR 4 d after the addition of the drugs. A) Immunofluorescence images (Bar = 50μm) and relative scatter plots. Data from three independent experiments are presented and statistically analyzed (B) as in Fig 1. C) RTqPCR on α-, γ- and β- globins. Histograms show the relative levels of expression relative to GAPDH. The transfection of a panel of siRNAs confirms the effectiveness of the immunomicroscopy platform to identify genes affecting hemoglobin synthesis and to test the efficacy of their modulators The transfection of a panel of siRNAs confirms the effectiveness of the immunomicroscopy platform to identify genes affecting hemoglobin synthesis and to test the efficacy of their modulators We then tested the efficacy of our method in identifying genes that could affect hemoglobiniza- tion and/or γ/β ratio by siRNA transfection. Firstly, we set up different types of controls: i) as a negative control we transfected a non-targeting oligo (siNTO); ii) as a positive transfection control we targeted the proteasome 26S subunit, ATPase3 (siPSMC3), the knockdown of which should severely affect cell growth; iii) we knocked-down γ-globin (siHBG1) and β-globin (siHBB) obtaining almost complete ablation of the respective signals (S3A Fig). To test the sensitivity in capturing changes in γ/β ratio, we knocked down HDAC3, repro- ducing the γ-globin promoter de-repression induced by treatment with butyric acid and its derivatives [23]. In HDAC3kd cells, we observed a marked increase in both γ- and β-globins, suggesting a broad mechanism of transcriptional de-repression (Fig 3A and 3B). In line with this, the cells treatment with HDAC inhibitors entinostat (MS-275) -an inhibitor of HDAC1 and HDAC3- and dacinostat (LAQ-824) induced γ-globin accumulation in a dose response manner (Fig 3C and 3D and S3B and S3C Fig). These results further confirm HDAC as targets for γ-globin reactivation and identify two additional inhibitors–in addition to BA (Fig 2)-, as potential therapeutic agents activating γ-globin. Fig 3. High-content γ/β globin analysis as readout of siRNA screening in β-K562 confirms HDAC as targets for γ-globin activation. A) Cells were transfected with a non-targeting oligo (siNTO) as negative control and with a siRNA directed to HDAC3. Two siRNAs were tested, with two technical replicates. C) β-K562 treated with two different HDAC inhibitors: entinostat and dacinostat (see also S3 Fig). A and C) Immunofluorescence images (Bar = 50μm) and relative scatter plots. Data from three independent experiments are presented and statistically analyzed (B and D) as in Fig 1. d i 10 1371/j l 0141083 003 Fig 3. High-content γ/β globin analysis as readout of siRNA screening in β-K562 confirms HDAC as targets for γ-globin activation. A) Cells were transfected with a non-targeting oligo (siNTO) as negative control and with a siRNA directed to HDAC3. Two siRNAs were tested, with two technical replicates. C) β-K562 treated with two different HDAC inhibitors: entinostat and dacinostat (see also S3 Fig). A and C) Immunofluorescence images (Bar = 50μm) and relative scatter plots. Validation of γ/β globin high-content assay by using the known hemoglobin inducers hydroxyurea and butyric acid D) Confocal analysis of β-K562 cells untreated or treated with HU as in panel A and subjected to a quadruple staining with Hoechst (blue), anti β (green), anti γ-globin (red) and anti-CD235a (white). Magnification: 20x. Right panel: 40x magnification of individual cells γ+CD235a+ or β+CD235a+ doub positive and γ+β+CD235a+ triple positive, respectively. d i 10 1371/j l 0141083 002 Fig 2. High-content analysis of compound-induced changes in globins accumulation. β-K562 cells were treated with 800μM hydroxyurea and 900μM butyric acid (n = 3, a representative experiment is shown here) and the same cells were analyzed in parallel by immunofluorescence and by RTqPCR 4 days after the addition of the drugs. A) Immunofluorescence images (Bar = 50μm) and relative scatter plots. Data from three independent experiments are presented and statistically analyzed (B) as in Fig 1. C) RTqPCR on α-, γ- and β- globins. Histograms show the relative levels of expression relative to GAPDH. D) Confocal analysis of β-K562 cells untreated or treated with HU as in panel A and subjected to a quadruple staining with Hoechst (blue), anti β- (green), anti γ-globin (red) and anti-CD235a (white). Magnification: 20x. Right panel: 40x magnification of individual cells γ+CD235a+ or β+CD235a+ double positive and γ+β+CD235a+ triple positive, respectively. doi:10.1371/journal.pone.0141083.g002 doi:10.1371/journal.pone.0141083.g002 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 6 / 14 A Novel Assay to Identify γ-Globin Inducing Compounds A Novel Assay to Identify γ-Globin Inducing Compounds Fig 4. HMOX2 siRNA-mediated knockdown and hemin or Tin-PPIX treatment have similar effects on β-K562 hemoglobinization levels. A) Cells were transfected with a non-targeting oligo (siNTO) as negative control and with a siRNA directed to HMOX2. Two siRNAs (see also S4A Fig) were tested, with two technical replicates. C) Cells were treated with 50μM of either hemin or Tin-PPIX. A, C) Immunofluorescence images (Bar = 50μm) and relative scatter plots. Data from three independent experiments are presented and statistically analyzed (B and D) as in Fig 1. E) RTqPCR on α-, γ- and β-globins from cells treated with hemin or Tin-PPIX. Histograms show levels of globins expression relative to GAPDH (n = 3). Fig 4. HMOX2 siRNA-mediated knockdown and hemin or Tin-PPIX treatment have similar effects on β-K562 hemoglobinization levels. A) Cells were transfected with a non-targeting oligo (siNTO) as negative control and with a siRNA directed to HMOX2. Two siRNAs (see also S4A Fig) were tested, with two technical replicates. C) Cells were treated with 50μM of either hemin or Tin-PPIX. A, C) Immunofluorescence images (Bar = 50μm) and relative scatter plots. Data from three independent experiments are presented and statistically analyzed (B and D) as in Fig 1. E) RTqPCR on α-, γ- and β-globins from cells treated with hemin or Tin-PPIX. Histograms show levels of globins expression relative to GAPDH (n = 3). doi:10.1371/journal.pone.0141083.g004 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 The transfection of a panel of siRNAs confirms the effectiveness of the immunomicroscopy platform to identify genes affecting hemoglobin synthesis and to test the efficacy of their modulators Data from three independent experiments are presented and statistically analyzed (B and D) as in Fig 1. doi:10.1371/journal.pone.0141083.g003 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 7 / 14 Discussion We present a reliable and sensitive assay based on the unique property of β-K562 that express both γ and β genes to perform a first step high-throughput screening (HTS) to identify genes/drugs influencing the γ/β globin ratio. This overcomes the major limitation of the available human ery- throid cell lines to study hemoglobin switching, i.e. their exclusive expression of embryonic/fetal genes. Indeed, HTS approaches published so far are almost exclusively based on cell lines trans- fected with a variety of reporters, under the control of γ and β globins promoters in the context of artificial genes/genomic arrangements[25],[26]. Recently established IPs-derived immortalized cell lines [27] represent a very promising tool for similar studies, but we feel that the easiness of growth and manipulation of β-K562 still represent a valuable advantage. Moreover, the growth factors independence of β-K562 prevents the possibility that globins expression could be influenced by subtle changes in growth conditions, and/or by other manipulations required to establish hES/hIPs. More physiological models for these studies are also available: mice carrying artificial chro- mosome constructs encompassing the entire human β-locus, some of them containing knock- in fluorescent reporters under the control of globins promoters[28–30]. The advantage of such models with respect to cell lines is counterbalanced by their reduced manageability making them unsuitable for first-step HTS. β-K562 cells are a valid tool for first-step screening because of their spontaneous expression of both γ- and β-globins from the intact β-locus (Fig 1). The presence of γ+β+ double positive cells and the plasticity in modulating γ and β expression suggest that in β-K562 the chromatin environment at the β locus is overall relatively accessible, making these cells particularly sensi- tive in detecting possible drugs/siRNA effects on γ activation. Regarding this issue, it is impor- tant to note that also in normal adult individuals there is a low proportion of HbF-positive cells–a few percent-, and immature cells in adults are known to transiently express γ-globin [31,32]. This suggests that a permissive environment for γ-globin expression may be present, although transiently, also in adult human cells, and could be modulated by drug treatment. Specific antibodies allow a reliable picture of the final readout of interest, i.e. the amount of β- and γ-globin protein upon different drugs treatments/genes manipulations at the single cell level, providing hints about the heterogeneity of the response. Is Heme-oxygenase a potential target for γ-globin induction? Given the above results, we undertook a pilot transfection screening on β-K562 with a panel of 70 siRNAs from the Ambion V4 library (S4 Table). Amongst them, siRNAs targeting the Heme-oxygenase (HO) coding gene HMOX2 (S4A Fig) gave a striking increase in globins with a prevalent accumulation of highly γ-globin expressing cells (Fig 4A and 4B). Heme-oxygenase catalyzes the conversion of heme to biliverdin (that, in turn, is immediately converted into bili- rubin), iron and carbon monoxide. Both hemin, and Heme-oxygenase work on heme pool homeostasis, albeit in an opposite way, the first by replenishing the heme pool and the latter by promoting its degradation[24]. We then reasoned that the effect of the pharmacological inhibi- tion of HO could result in globins stimulation similar to that known to be elicited by hemin (and possibly by the HMOX-2 knockdown shown in Fig 4). To test this hypothesis, we per- formed a dose/response treatment for hemin and Tin protoporphyrin IX (Tin-PPIX), here con- sidered as prototypical HO inhibitor (S4B Fig). Fig 4C and 4D shows a striking increase in the percentage of γ-globin expressing cells and in γ-globin accumulation obtained upon Tin-PPIX treatment. Interestingly, whereas hemin treatment significantly increases the percentage of PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 8 / 14 A Novel Assay to Identify γ-Globin Inducing Compounds double γ+β+ positive cells, Tin-PPIX seems to have a more selective activity on γ, as confirmed by the MFI values (S4C Fig). Instead, at the mRNA level, the effect of hemin and Tin-PPIX are very similar (at equal concentrations), suggesting that post-transcriptional effects may play additional roles in balancing the α/non-α ratio. PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 Discussion In parallel, RTqPCR provides information on the differential regulation (transcription/RNA processing and stability versus translation) of globin expression (as well as on any other gene of interest) elicited by drugs/treat- ments (Figs 2–4) and modulators (siRNA targeting of HDAC3 and HMOX2, Figs 3 and 4). The microscopy analysis can be further implemented to simultaneously analyze changes in multiple cellular parameters, such as morphology and/or expression of specific markers (Fig 3E), thus allowing the detection of possible effects of the tested drugs/treatments on different cellular processes. Importantly, the globin expression analysis is performed at the single cell level. This latter aspect is of particular relevance since the response to pharmacological agents that increase HbF is expected to involve either the accumulation of γ-globin in each single cell (due to transcriptional and/or translational effects) or the selection of subsets of erythroid dif- ferentiating “responder” cells, on the basis of a pre-existing heterogeneity. β-K562 cells, allowed an efficient automated transfection-based screening, that led to the confirmation of HDACs as “druggable” targets for γ reactivation[16] (Fig 3) and to the identifi- cation of Heme-oxygenases (Fig 4) as possible novel target. 9 / 14 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 A Novel Assay to Identify γ-Globin Inducing Compounds Regarding HDACs, we tested the effect on β-K562 of three different inhibitors: butyric acid (BA), a well known γ-inducing agent, entinostat (MS-275) and dacinostat (LAQ-824). Dacino- stat has been considered for myeloid leukemia treatment because it promotes apoptosis in CML and AML cells [33]. Entinostat, an inhibitor of HDAC1 and HDAC3, is an inducer of cell differentiation in AML cells, not associated with apoptosis induction[34–36]. Whereas BA appears to specifically lead to γ-globin expression, entinostat and dacinostat lead to an increase in both γ- and β-globins, suggesting a different selectivity profile of these three compounds. Concerning Heme-oxygenases, our study identified them as a novel class of potential drug- gable targets to reactivate γ-globin. Our results show that drug competitive inhibition of Heme- oxygenases by protoporphyrins (and thus presumably by other derivatives) is able to induce a substantial increase in γ-globin accumulation. In erythroid cells, the main function of heme is to serve as the oxygen-carrying moiety in hemoglobin (Hb), and heme biosynthesis is thus strictly coordinated with globin accumulation along with erythroid differentiation and maturation[37]. Discussion Heme, when present at high concentrations in the globin-unbound state (as in β-thalassemia), inactivates the heme-regulated eukaryotic initiation factor eIF2α-kinase (Heme Responsive Inhibitor, HRI), converting it into an inactive form. As active HRI inhibits the eIF2α translation initiation factor, excess heme increases globin synthesis, allowing a better balance between heme and globin chains[38]. Therefore, Heme-oxygenase inhibition, by increasing heme levels, is expected to favor globin synthesis. In addition, heme catabolism by oxidation is linked to genera- tion of biologically active molecules, such as iron, biliverdin, CO and NO[24,37]. The observation that both heme addition and inhibition of HO increase globin expression in β-K562 is consistent with the expected role of heme in relieving translational inhibition by HRI; however, the increase of globin mRNA levels, particularly of γ-globin (mRNA and protein) is not easily explained only by a simple effect on globin mRNA translation. It is possible that the translational effect of HRI also operates on other factors, for example transcription or chromatin factors regulating the HBG1/2. Previously, studies of HMOX1 deficiency demonstrated a clear effect on both stress and steady state erythropoiesis[39–42], but less is known about the possible specific role of the consti- tutive HMOX2 (the most abundant isoform expressed by K562[43]) in erythroid cells[37,44]. Of interest, a polymorphism in the HMOX1 gene was associated with high levels of fetal hemoglobin in Brazilian patients with sickle cell anemia[45]. Both HO-1 and HO-2 are sensitive (albeit to a different extent) to the competitive inhibition elicited by different Metalloporphyrins and the availability of different drugs with different selectivity make HOs an attractive target for pharmacological inhibition. This result is of par- ticular interest because different Metalloporphyrins have been developed and tested in clinics [46]. Our results suggest that their use, as single agent or in association with other known and FDA-approved HbF inducers, such as HU, should be explored as a promising tool to improve the α/non α globin chain imbalance in β-hemoglobinopathies. To address this question we are currently studying the effect of different Metalloporphyrins in mice carrying a complete human HBB locus transgene e and in ex-vivo cultures from thalassemic patients. S1 File. siRNAs oligonucleotide detailed transfection method. (PDF) S1 Table. Chemicals and antibodies. (PDF) S3 Table. List of primers used for RTqPCR. (PDF) S3 Table. List of primers used for RTqPCR. (PDF) S4 Table. List of genes tested by siRNA-mediated knockdown and selected from the Ambion-library. (PDF) Supporting Information A) HMOX2 knockdown: RTqPCR on cells transfected with a non targeting oligo (siNTO) and with two independent siRNA directed to HMOX2. B) Representative ArrayScan pictures of β- K562 cells treated with increasing doses of Hemin and or Tin-PPIX. Bar = 50μm. C) MFI plots as in S2 Fig. (TIF) S4 Fig. Hemin and Tin-PPIX have similar effects on β-K562 hemoglobinization levels. A) HMOX2 knockdown: RTqPCR on cells transfected with a non targeting oligo (siNTO) and with two independent siRNA directed to HMOX2. B) Representative ArrayScan pictures of β- K562 cells treated with increasing doses of Hemin and or Tin-PPIX. Bar = 50μm. C) MFI plots as in S2 Fig. (TIF) Supporting Information S1 Fig. Characterization of the β-K562 subclone by comparison with ECAAC-K562. A) Growth curves (n = 2). B) Response (IC50) to imatinib mesylate, dasatinib and doxorubicin (n3). C) FCM analysis: cells were stained with anti γ- and anti β-globin antibodies and with the corresponding isotype controls and read in FL-1 (FITC, green channel) or in FL-2 (PE, red channel). A representative experiment is shown. Immunofluorescence setup. D) In the immu- nofluorescence analysis, nuclei were stained with Hoechst-33342; HbF and HbA were immu- nostained by using specific anti γ- and anti β-globin antibodies and signals were acquired in 10 / 14 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 A Novel Assay to Identify γ-Globin Inducing Compounds three single channels: blue for Hoechst (Ch1), green for β-globin (Ch2) and red for γ-globin (Ch3), respectively, and then merged for analysis (Merge). E) Images acquired in the single channels for representative isotype controls and relative scatter plots. Bar = 50μm. (TIF) S2 Fig. Dose/response of β-K562 to Hydroxyurea and Butyric Acid. A) Representative ArrayScan pictures of β-K562 cells treated with increasing doses of HU and BA (n3). Bar = 50μm. B) Fluorescence intensity plots to better visualize the changes in mean fluores- cence intensity (MFI) of stained cells upon drugs treatment. Y axis: number of events (cells); X axis: fluorescence intensity for β-globin signal (upper panels) or γ-globin signal (lower panels), respectively. Green/Red curves: treated cells. Black curve: untreated cells. The vertical dotted line within each panel corresponds to the threshold set in Fig 2A. The MFI and the percentage of positive cells (%) are indicated within each panel. (TIF) S3 Fig. HDAC3 siRNA-mediated knockdown and HDAC inhibitors treatment in β-K562 confirm HDACs as targets for γ-globin activation. A) Cells were transfected with a non-tar- geting oligo (siNTO) as negative control and with a siRNA directed to PSMC3 as positive transfection control. As further control, siRNAs targeting γ- and β-globins greatly reduced the corresponding globins chains. For each gene, two siRNAs were tested, with two technical repli- cates (immunofluorescence images of representative experiments are shown). Bar = 50μm. Scatter plots are provided for each immunofluorescence image (n = 2). B) Representative ArrayScan pictures of β-K562 cells treated with increasing doses of entinostat and dacinostat. C) MFI plots as in S2 Fig. (TIF) S4 Fig. Hemin and Tin-PPIX have similar effects on β-K562 hemoglobinization levels. References 1. Weatherall D, Akinyanju O, Fucharoen S, Olivieri N, Musgrove P (2006) Inherited Disorders of Hemo- globin. In: Jamison DT, Breman JG, Measham AR, Alleyne G, Claeson M et al., editors. Disease Con- trol Priorities in Developing Countries. 2nd ed. Washington (DC). 1. Weatherall D, Akinyanju O, Fucharoen S, Olivieri N, Musgrove P (2006) Inherited Disorders of Hemo- globin. In: Jamison DT, Breman JG, Measham AR, Alleyne G, Claeson M et al., editors. Disease Con- trol Priorities in Developing Countries. 2nd ed. Washington (DC). 2. Weatherall DJ, Clegg JB (2001) Inherited haemoglobin disorders: an increasing global health problem. Bull World Health Organ 79: 704–712. PMID: 11545326 2. Weatherall DJ, Clegg JB (2001) Inherited haemoglobin disorders: an increasing global health problem. Bull World Health Organ 79: 704–712. PMID: 11545326 3. Sankaran VG, Nathan DG (2010) Thalassemia: an overview of 50 years of clinical research. Hematol Oncol Clin North Am 24: 1005–1020. doi: 10.1016/j.hoc.2010.08.009 PMID: 21075277 4. Flint J, Harding RM, Boyce AJ, Clegg JB (1998) The population genetics of the haemoglobinopathies. Baillieres Clin Haematol 11: 1–51. PMID: 10872472 5. Rivella S (2009) Ineffective erythropoiesis and thalassemias. Current Opinion in Hematology 16: 187– 194. doi: 10.1097/MOH.0b013e32832990a4 PMID: 19318943 6. Forget BG, Tuan D, Newman MV, Feingold EA, Collins F, Fukumaki Y, et al. (1983) Molecular studies of mutations that increase Hb F production in man. Prog Clin Biol Res 134: 65–76. PMID: 6198666 7. Suzuki M, Yamamoto M, Engel JD (2014) Fetal globin gene repressors as drug targets for molecular therapies to treat the beta-globinopathies. Mol Cell Biol 34: 3560–3569. doi: 10.1128/MCB.00714-14 PMID: 25022757 8. Perrine SP, Pace BS, Faller DV (2014) Targeted fetal hemoglobin induction for treatment of beta hemo- globinopathies. Hematol Oncol Clin North Am 28: 233–248. doi: 10.1016/j.hoc.2013.11.009 PMID: 24589264 9. Bauer DE, Kamran SC, Orkin SH (2012) Reawakening fetal hemoglobin: prospects for new therapies for the beta-globin disorders. Blood 120: 2945–2953. doi: 10.1182/blood-2012-06-292078 PMID: 22904296 10. Thein SL, Menzel S, Lathrop M, Garner C (2009) Control of fetal hemoglobin: new insights emerging from genomics and clinical implications. Hum Mol Genet 18: R216–223. doi: 10.1093/hmg/ddp401 PMID: 19808799 11. Pourfarzad F, von Lindern M, Azarkeivan A, Hou J, Kia SK, Esteghamat F, et al. (2013) Hydroxyurea responsiveness in beta-thalassemic patients is determined by the stress response adaptation of ery- throid progenitors and their differentiation propensity. Haematologica 98: 696–704. doi: 10.3324/ haematol.2012.074492 PMID: 23100274 12. Author Contributions Conceived and designed the experiments: FG AER. Performed the experiments: MD CF SE MGM MFM IF PC GB IFM MM. Analyzed the data: MD CF PM FG AER. Contributed reagents/materials/analysis tools: PC MM. Wrote the paper: PM SO FG AER. Conceived and designed the experiments: FG AER. Performed the experiments: MD CF SE MGM MFM IF PC GB IFM MM. Analyzed the data: MD CF PM FG AER. Contributed reagents/materials/analysis tools: PC MM. Wrote the paper: PM SO FG AER. Acknowledgments This work was supported by the People Programme (Marie Curie Actions) of the European Union's Seventh Framework Programme FP7/2007-2013/ under REA grant agreement n° PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 11 / 14 A Novel Assay to Identify γ-Globin Inducing Compounds 289611 (HEM_ID Project) and by Fondazione Cariplo for grant n° 2012–0517 to A.R. We thank Prof. Giuliana Ferrari and Dr. Angela Greco for scientific support. PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 References Maines MD (1997) The heme oxygenase system: a regulator of second messenger gases. Annu Rev Pharmacol Toxicol 37: 517–554. PMID: 9131263 25. Peterson KR, Costa FC, Fedosyuk H, Neades RY, Chazelle AM, Zelenchuk L, et al. (2014) A cell- based high-throughput screen for novel chemical inducers of fetal hemoglobin for treatment of hemo- globinopathies. PLoS One 9: e107006. doi: 10.1371/journal.pone.0107006 PMID: 25225870 26. Chan KS, Xu J, Wardan H, McColl B, Orkin S, Vadolas J (2012) Generation of a genomic reporter assay system for analysis of gamma- and beta-globin gene regulation. FASEB J 26: 1736–1744. doi: 10.1096/fj.11-199356 PMID: 22267339 27. Kurita R, Suda N, Sudo K, Miharada K, Hiroyama T, Miyoshi H, et al. (2013) Establishment of immortal- ized human erythroid progenitor cell lines able to produce enucleated red blood cells. PLoS One 8: e59890. doi: 10.1371/journal.pone.0059890 PMID: 23533656 28. McColl B, Kao BR, Lourthai P, Chan K, Wardan H, Roosjen M, et al. (2014) An in vivo model for analy- sis of developmental erythropoiesis and globin gene regulation. FASEB J 28: 2306–2317. doi: 10. 1096/fj.13-246637 PMID: 24443374 29. Papadopoulos P, Gutierrez L, van der Linden R, Kong ASJ, Maas A, Drabek D, et al. (2012) A dual reporter mouse model of the human beta-globin locus: applications and limitations. PLoS One 7: e51272. doi: 10.1371/journal.pone.0051272 PMID: 23272095 30. Patrinos GP, de Krom M, de Boer E, Langeveld A, Imam AM, Strouboulis J, et al. (2004) Multiple inter- actions between regulatory regions are required to stabilize an active chromatin hub. Genes Dev 18: 1495–1509. PMID: 15198986 31. Papayannopoulou T, Brice M, Stamatoyannopoulos G (1977) Hemoglobin F synthesis in vitro: evi- dence for control at the level of primitive erythroid stem cells. Proc Natl Acad Sci U S A 74: 2923–2927. PMID: 268643 32. Papayannopoulou T, Nakamoto B, Buckley J, Kurachi S, Nute PE, Stamatoyannopoulos G (1978) Ery- throid progenitors circulating in the blood of adult individuals produce fetal hemoglobin in culture. Sci- ence 199: 1349–1350. PMID: 628844 33. Weisberg E, Catley L, Kujawa J, Atadja P, Remiszewski S, Fuerst P, et al. (2004) Histone deacetylase inhibitor NVP-LAQ824 has significant activity against myeloid leukemia cells in vitro and in vivo. Leuke- mia 18: 1951–1963. PMID: 15496979 34. Blagitko-Dorfs N, Jiang Y, Duque-Afonso J, Hiller J, Yalcin A, Greve G, et al. (2013) Epigenetic priming of AML blasts for all-trans retinoic acid-induced differentiation by the HDAC class-I selective inhibitor entinostat. References Ronchi A, Ottolenghi S (2013) To respond or not to respond to hydroxyurea in thalassemia: a matter of stress adaptation? Haematologica 98: 657–659. doi: 10.3324/haematol.2013.084392 PMID: 23633538 13. Musallam KM, Taher AT, Cappellini MD, Sankaran VG (2013) Clinical experience with fetal hemoglobin induction therapy in patients with beta-thalassemia. Blood 121: 2199–2212; quiz 2372. doi: 10.1182/ blood-2012-10-408021 PMID: 23315167 14. Perrine SP, Castaneda SA, Chui DH, Faller DV, Berenson RJ, Siritanaratku N, et al. (2010) Fetal globin gene inducers: novel agents and new potential. Ann N Y Acad Sci 1202: 158–164. doi: 10.1111/j.1749- 6632.2010.05593.x PMID: 20712788 15. Banan M (2013) Hydroxyurea treatment in beta-thalassemia patients: to respond or not to respond? Ann Hematol 92: 289–299. doi: 10.1007/s00277-012-1671-3 PMID: 23318979 16. Bradner JE, Mak R, Tanguturi SK, Mazitschek R, Haggarty SJ, Ross K, et al. (2010) Chemical genetic strategy identifies histone deacetylase 1 (HDAC1) and HDAC2 as therapeutic targets in sickle cell dis- ease. Proc Natl Acad Sci U S A 107: 12617–12622. doi: 10.1073/pnas.1006774107 PMID: 20616024 17. Cantu C, Ierardi R, Alborelli I, Fugazza C, Cassinelli L, Piconese S, et al. (2011) Sox6 enhances ery- throid differentiation in human erythroid progenitors. Blood 117: 3669–3679. doi: 10.1182/blood-2010- 04-282350 PMID: 21263153 12 / 14 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 A Novel Assay to Identify γ-Globin Inducing Compounds 18. Somaschini A, Amboldi N, Nuzzo A, Scacheri E, Ukmar G, Ballinari D, et al. (2013) Cell line identity find- ing by fingerprinting, an optimized resource for short tandem repeat profile authentication. Genet Test Mol Biomarkers 17: 254–259. doi: 10.1089/gtmb.2012.0359 PMID: 23356232 19. Lozzio CB, Lozzio BB (1975) Human chronic myelogenous leukemia cell-line with positive Philadelphia chromosome. Blood 45: 321–334. PMID: 163658 20. Tsiftsoglou AS, Pappas IS, Vizirianakis IS (2003) Mechanisms involved in the induced differentiation of leukemia cells. Pharmacol Ther 100: 257–290. PMID: 14652113 21. Guerrasio A, Vainchenker W, Breton-Gorius J, Testa U, Rosa R, Thomopoulos P, et al. (1981) Embry- onic and fetal hemoglobin synthesis in K562 cell line. Blood Cells 7: 165–176. PMID: 6198015 22. Rutherford TR, Clegg JB, Weatherall DJ (1979) K562 human leukaemic cells synthesise embryonic haemoglobin in response to haemin. Nature 280: 164–165. PMID: 95354 23. Mankidy R, Faller DV, Mabaera R, Lowrey CH, Boosalis MS, White GL, et al. (2006) Short-chain fatty acids induce gamma-globin gene expression by displacement of a HDAC3-NCoR repressor complex. Blood 108: 3179–3186. PMID: 16849648 24. PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 References PLoS One 8: e75258. doi: 10.1371/journal.pone.0075258 PMID: 24116031 35. Maggio SC, Rosato RR, Kramer LB, Dai Y, Rahmani M, Paik DS, et al. (2004) The histone deacetylase inhibitor MS-275 interacts synergistically with fludarabine to induce apoptosis in human leukemia cells. Cancer Res 64: 2590–2600. PMID: 15059916 36. Stintzing S, Kemmerling R, Kiesslich T, Alinger B, Ocker M, Neureiter D (2011) Myelodysplastic syn- drome and histone deacetylase inhibitors: "to be or not to be acetylated"? J Biomed Biotechnol 2011: 214143. doi: 10.1155/2011/214143 PMID: 21629744 37. Abraham NG, Kappas A (2008) Pharmacological and clinical aspects of heme oxygenase. Pharmacol Rev 60: 79–127. doi: 10.1124/pr.107.07104 PMID: 18323402 38. Chen JJ (2007) Regulation of protein synthesis by the heme-regulated eIF2alpha kinase: relevance to anemias. Blood 109: 2693–2699. PMID: 17110456 39. Fraser ST, Midwinter RG, Berger BS, Stocker R (2011) Heme Oxygenase-1: A Critical Link between Iron Metabolism, Erythropoiesis, and Development. Adv Hematol 2011: 473709. doi: 10.1155/2011/ 473709 PMID: 22162689 13 / 14 PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 A Novel Assay to Identify γ-Globin Inducing Compounds 40. Fraser ST, Midwinter RG, Coupland LA, Kong S, Berger BS, Yeo JH, et al. (2015) Heme oxygenase-1 deficiency alters erythroblastic island formation, steady-state erythropoiesis and red blood cell lifespan in mice. Haematologica. 40. Fraser ST, Midwinter RG, Coupland LA, Kong S, Berger BS, Yeo JH, et al. (2015) Heme oxygenase-1 deficiency alters erythroblastic island formation, steady-state erythropoiesis and red blood cell lifespan in mice. Haematologica. 41. Koizumi S (2007) Human heme oxygenase-1 deficiency: a lesson on serendipity in the discovery of the novel disease. Pediatr Int 49: 125–132. PMID: 17445026 42. Kovtunovych G, Eckhaus MA, Ghosh MC, Ollivierre-Wilson H, Rouault TA (2010) Dysfunction of the heme recycling system in heme oxygenase 1-deficient mice: effects on macrophage viability and tissue iron distribution. Blood 116: 6054–6062. doi: 10.1182/blood-2010-03-272138 PMID: 20844238 43. Ding Y, Zhang YZ, Furuyama K, Ogawa K, Igarashi K, Shibahara S (2006) Down-regulation of heme oxygenase-2 is associated with the increased expression of heme oxygenase-1 in human cell lines. FEBS J 273: 5333–5346. PMID: 17064313 44. Seta F, Bellner L, Rezzani R, Regan RF, Dunn MW, Abraham NG, et al. (2006) Heme oxygenase-2 is a critical determinant for execution of an acute inflammatory and reparative response. Am J Pathol 169: 1612–1623. PMID: 17071585 45. Gil GP, Ananina G, Oliveira MB, Costa FF, Silva MJ, Santos MN, et al. PLOS ONE \| DOI:10.1371/journal.pone.0141083 October 28, 2015 References (2013) Polymorphism in the HMOX1 gene is associated with high levels of fetal hemoglobin in Brazilian patients with sickle cell ane- mia. Hemoglobin 37: 315–324. doi: 10.3109/03630269.2013.789438 PMID: 23725037 46. Dennery PA (2005) Metalloporphyrins for the treatment of neonatal jaundice. Curr Opin Pediatr 17: 167–169. PMID: 15800405 14 / 14
W4385995571.txt	https://zenodo.org/record/8233248/files/67-71.pdf	ta		சங்ககாலப் பெண்களின் நிலை	Zenodo (CERN European Organization for Nuclear Research)	2,023	cc-by	1,279	SHANLAX International Journal of Arts, Science and Humanities \[PPõ»¨ ö£sPÎß {ø» •øÚÁº Cµõ. öu#Á\|õuß EuÂ¨÷£µõ]›¯º, uªÌzxøÓ v¯õPµõ\ºPÀ¿›, ©xøµ •ßÝøµ \[P® øÁzxz uªÌ ÁÍºzu Põ»PmhzvÀ E¸ÁõÚ C»UQ¯[PÒ AUPõ» \‰P `Çø» A¨£i÷¯ ¤µv£¼US® Âu©õP Aø©QßÓÚ. AßøÓ¯ \‰P ÁõÌÂ¯À £s£õk ©»º: 11 P»õa\õµ® \|õPŸP® ©ßÚºPÎß Bm]•øÓ ¤Ó \|õmiÚ¸øh¯ ]Ó¨¤uÌ: 1 £øhö¯k¨¦z öuõÈÀ •øÓPÒ ÷ÁÍõsø©PÒ BQ¯ÁØøÓ AÔ¢x öPõÒÐ® ÁøP°À \[P C»UQ¯[PÒ Aø©QßÓÚ. ©õu®: BPìm CÆÁõÖ £À÷ÁÖ ÁøP¯õÚ Pmhø©¨¦PøÍ EÒÍhUQ¯ C»UQ¯[PÎÀ ö£sPÒ SÔzu £vÄPÒ £Ó¢x Â›¢x Á¸h®: 2023 Põn¨£kQßÓÚ. \[P Põ»® Gß£x SÔ¨£õP BnõvUP P-ISSN: 2321-788X \‰P©õS®. CUPõ»PmhzvÀ ö£sPÒ uÚUöPÚ J¸ uÛzu ÁõÌøÁ Pmhø©zx ÁõÌ¢x Á¢u @ÁøÍ°À £»¨ ö£sPÒ E-ISSN: 2582-0397 ¦»ÁºPÒ BPÄ® ]Ó¢u CÀ»zuµ]PÍõPÄ® ÂÍ[Q¯ {ø»°À ]» ö£sPÒ Ãµ ö\¯ÀPÎ¾® Dk£mkÒÍÚº Gß£øu²® DOI: AÔ¯ •iQÓx. AzuøP¯ ©UPÎß GzuøP¯ Ãµ¨£ß¦P÷Íõk https://doi.org/10.5281/ ÂÍ[QÚº Gß£uøÚ CUPmkøµ ÁÈAÔ¯»õ®. zenodo.8233248 öuõÀPõ¨¤¯® & ©PÎº uªÇPzvß •uÀ ¡»õPU P¸u¨£k® ¡À öuõÀPõ¨¤¯® BS®. C¢¡À C»UQ¯zvØS C»UPn® TÖÁx ©mk©À»õ©À ÁõÌUøPUS® C»UPn® TÖQÓx. AzuøP¯ C»UPn ¡¼À Aa\•® \|õÝ® ©hÝ® •ßxÖuÀ {a\•® ö£s£õºUS›¯ Gß£ (öuõÀ. ¡ 96) GßÖ ö£sPÐUS C»UPn® TÖQÓx. Aa\® \|õn® ©hß CøÁPÒ ©mk÷© ö£sPÐUS E›¯x GßÓõÀ Ãµ® ¯õ¸US›¯x GßÓ ÷PÒÂ \|®ªøh÷¯ C¸¢uõ¾® Âøh GßÚ÷Áõ BqUSz uõß GßÓ GÊu¨£hõu Âv¯õP Ax Aø©QÓx. öuõhUP Põ»Pmh[PÎÀ Bs ö£s GßÓ C¸£õÍ¸® ÷Ámøha \‰P©õP÷Á ÁõÌ¢uÚº Gß£÷u Áµ»õØÖ AÔbºPÎß P¸zx BS®. CzuøP¯ {ø»°À ÷Ámøha \‰P® •i¢x {»Äøha \‰P® öuõh[Q¯ ÷£õx ö£s Gß£ÁÒ J¸ SÔ¨¤mh Áøµ¯øµUSÒ ÁõÇ {º£¢vUP¨£mhõÀ. http://www.shanlaxjournals.com £UP® 67 ,yf;fpaq;fs; fhl;Lk; ngz; epiyg;ghL; CuØS £À÷ÁÖ Âu©õÚ Aµ]¯À \‰PÂ¯À Põµn[PÍõP Aø©QßÓÚ. CuøÚ÷¯ ö£s Gß£ÁÒ Cµshõ¢uµ Si©UPÍõP÷Á CßøÓ¯ \‰P® AÁºPøÍ¨ £õºUQÓx. öuõÀPõ¨¤¯® CzuøP¯ Põ»PmhzvÀ E¸ÁõÚ C»UPn® ¡À Gß£uõÀ PØ¦® Põ©•® \|õØ£õÀ JÊUP•® \|À¼vß ¦iu²® Â¸¢x ¦Ó¢u¸u¾® _ØÓ® K®£¾® ¤ÓÄ ©ßÚ QÇ@Áõß ©õs¦PÒ (öuõÀ. ¡ 150) PØ¦ JÊUP® \|Àö»sönÔ ¤Óøµ¨ ÷£o £õxPõzuÀ BQ¯¨ ö£sPÐUS ©mk÷© E›¯uõPz öuõÀPõ¨¤¯® SÔ¨¤kQÓx. Cx vmhªmka ö\#¯¨£mh \‰P JkUS •øÓPÒ BS®. {»® BqUS ©mk÷© GßÓ JØøÓ P¸zv¯À \‰PzvÀ £µÂ¯uõÀ ö£s ¦ÓUPoUP¨£mhõÀ Gß£øu AÔ¯ •iQÓx. ö£¸ø©²® EµÝ® BkE÷©Ú (öuõÀ.¡99) GßÓ öuõÀPõ¨¤¯ ¡Ø£õ ö£sPÐUS¨ ö£¸ø©²® EµÝ® (Eµ® EÒÍzvß Á¼ø© AÔÄ) C¸UP ÷Ásk® GßÖ TÖQÓx. CøÁ Hß ö£sPÐUS¨ ö£õ¸¢xÁuõP öuõÀPõ¨¤¯º TÓÂÀø». CuØS Ai¨£øh AßøÓ¯ \‰PzvÀ {»Â¯ BnõvUPa \‰P÷© Põµn©õP Aø©QßÓx. CuøÚU P¸zvÀ öPõs÷h AUPõ» \‰P ¯uõºzuzøu A¨£i÷¯ ¤µv£¼¨£õPz öuõÀPõ¨¤¯º £vÄ ö\#v¸UQÓõº GßÖ P¸u •iQÓx. £søhz uªÇPzvÀ \mh Ai¨£øh°À BhÁº ©PÎ›øh÷¯ \‰P \©zxÁU ÷Põm£õk C¸UPÂÀø»: AßÔ²® \•uõ¯zvÀ ©PÎ›øh¯ Phø©PÒ AøÚÁ¸øh¯ Phø©PÎßÖ® ÷ÁÓõÚøÁ. Bu»õÀ \‰PzvÀ BhÁºUS›¯ {ø»ø©ø¯ ©PÎº Aøh¯ ÂøÇuÀ Thõx Gß£øu öuÍÎuß Enµ¨£mhÚ. 68 £UP® (\[Põ» ÁõÌÂ¯À&£U375) GßÖ \|õß _¨¤µ©o¯ß P¸zx Gz;zzuUPuõP Aø©QÓx. BhÁ›ß E›ø©PøÍ G¢u ÁøP°¾® ö£sPÒ GkzxU öPõÒÍ÷Áõ uÚUPõÚ ÂvPøÍ «Ó÷Áõ E›ø© CÀ»õuÁÚõP¨ ö£s £õºUP¨£mhõÀ Gß£x \[PPõ» ÁõÌÂ¯À•øÓ¯õS®. C»UQ¯[PÎÀ ö£s AhUS•øÓ ö£õxÁõPa \[P C»UQ¯[PÎÀ ö£sPÒ SÔzu £vÄPÒ {øÓ¢x Põn QøhUQßÓÚ. AÁØÔÀ ö£s AhUS •øÓ²® ö£s£õØ ¦»ÁºPÎß ]Ó¨¦® ö£s Ãµ•® uÛzuÛ¯õP Aøh¯õÍ¨ £kzu¨£mk Á¸QßÓÚ CÁØÔÀ ö£s AhUS•øÓ SÔzxAÔ÷Áõ®. •¢}º ÁÇUP® ©PlE@Áõk CÀø» (öuõÀ.¡Ø£õ37) ªPz öuÎÁõPU PhÀ Ph¢x ö\ÀÁx ©PÎ¸US CÀø» GßÖ •u¼À öuõÀPõ¨¤¯º SÔ¨¤kQÓõº. FÈß E¸¨£ G¸UQ¯ ©PÎº ÁõøÇ¨ §Âß ÁøÍ•Ô ]uÓ (¦Ó®.227) GßÓ AiPÒ ntsp©õß GßÓ ©ßÚß CÓ¢x ÂmhuõÀ AÁÝøh¯ ©PÎº u[PÒ ©õº¤À AøÓ¢x öPõsk ÁõøÇ¨§ ©õv› u[PÒ ÁøÍ¯ÀPøÍ Eøhzx GÔ¢uõºPÒ GßÓ ÂÍUPzvøÚ u¸QßÓÚ. uÚUS E›¯ Bs CÓ¢x ÂmhõÀ Áis°À Ao¯UThõx GßÖ SÔ¨¤kÁøu \|õ® Põn •iQÓx. ÷©¾®, §ÁõÒ ÷PõÁ»º §Ähß EvµU öPõ#x Pmk AÈzu ÷Á[øP°ß ö©À¼¯À ©PÎ¸® CøÇPøÍ¢ uÚ÷µ (¦Ó®.224) GßÓ AiPÎÀ PõÁÀ AµnõP C¸¢u @ŒõÇß P›PõÀ ö£¸ÁÍzuõß GßÓ ©ßÚº CÓ¢uÄhß £]¯õÀ Áõk® u[PÒ PßÖPÐUS Cøh¯º S» ©UPÒ N`hypfpuh]; N`hk; rapd;]; fy;Y}hp> J}j;Jf;Fb SHANLAX International Journal of Arts, Science and Humanities ÷Á[øP ©µzvß uspºPøÍ §øÁ öÁmi EnÁõ# ÷£õmhÚº.A¨£i öÁmi¯¤ß ö©õmøh¯õ# ÷Á[øP ©µ® {ØS÷© Ax÷£õ» @ŒõÇÝøh¯ E›ø© ©PÎº u[PÒ AoP»ßPøÍ Pø»¢v¸¢uÚº GßÖ SÔ¨¤kQÓÚ. BhÁøµ CÇ¢u¨ ö£sPÒ CzuøP¯ AÁ» {ø»USz uÒÍ¨£mh `Çø» £õºUS®÷£õx ö£sPÐUPõÚ E›ø© •ØÔ¾® ©ÖUP¨£mhuõP Aø©¢v¸¨£uøÚ AÔ¯ •iQÓx. PnÁß CÓ¢u ¤ß \[P Põ» ö£sPÎß AÁ»® ö\õÀ¼ ©õshu {ø»¯õP EÒÍx. T¢uÀ öPõ#x SÖ¢öuõi }UQ AÀ¼ EnÂß ©øÚÂ÷¯õk CÛ÷¯ (¦Ó®.250) PnÁøÚ Co¢u ö£sPÎß uø»©°º }UP¨£kÁx® ¦À»›] EnÄ öPõsk ÁõÊÁx® AÁºPÎß {ø»¯õP C¸¨£vøÚ AÔ¯ •iQÓx. ©sqÖ ©Èz uø»z öus }º Áõµ öuõßÖ uõ® Ekzu A® £øPz öu›¯À ]Ö öÁÒ B®£À AÀ¼ Esq® PÈ P» ©PÎº ÷£õ», ÁÈ {øÚ¢v¸zuÀ, AuÛÝ® A›÷u! (¦Ó®.280) GßÖ AiPÎÀ uø»Áß CÓ¢x ÂmhõÀ AUPõ» ö£sPÒ ö©õmøh Aizx¨ ¦À»›]ø¯ Esk Ao©oPøÍ xÓ¢x øPik ÁõÌUøP ÷©ØöPõÒÍÁºPÒ GßÖ SÔ¨¤kQÓx. CÁØÖUöPÀ»õ® J¸ £i ÷©÷» ö\ßÖ \|õ[PÒ PnÁøÚ CÇ¢uõÀ _kPõmiÀ EshõUP¨£mh ÂÓS £kUøP G[PÐUS Cß£® ©¯US®. BÚõÀ E[PÐUS ÷Áskö©ßÓõÀ Ax xß£zøuz u¸®. G[PøÍ¨ ö£õÖzu Áøµ°À uõ©øµ §zu uhõP•® D•zv G›²® ÂÓS £kUøP²® JßÖuõß GßÖ ö£sP÷Í £õkÁuõP Aø©¢v¸¨£x Â¯¨¦US›¯uõP Aø©QÓx. CuøÚ, http://www.shanlaxjournals.com ö£¸[ Põmk¨ £so¯ P¸[ ÷Põmk D©® ~©US A›uõSP vÀ»; G©US G® ö£¸¢ ÷uõÒ PnÁß ©õ#¢öuÚ, A¸®¦ AÓ ÁÒ CuÌ AÂÌ¢u uõ©øµ \|Ò C¸® ö£õ#øP²® w²® KµØ÷Ó (¦Ó®.280) GßÓ £õhÀ AiPÒ EnºzxQßÓÚ. BnõvUP \‰PzvÀ ö£sPÎß ÷£õUSPÒ Aiø©zuÚ©õPÄ® AhUS •øÓPÍõPÄ® C¸¢xÒÍÚ Gß£øu AUPõ» ö£sP÷Í EÍÂ¯ÀŸv¯õP AÔ¯õ©À C¸¢xÒÍÚº Gß£øu AÔ¯ •iQÓx. öuõhUP Põ»PmhzvÀ uõ# ÁÈ \‰P©õP C¸¢x Á¢u ÷Áø»°À Bs ö£s C¸£õ»¸® \©©õP÷Á P¸u¨£mhÚ. \‰PzvÀ ö£sPøÍz öu#Á©õPÄ® £õºzuÚº. AøÚÁ¸® ÷Ámøh¯õi JßÓõP¨ £Qº¢xsk ©QÌ¢u Põ»zvÀ Bs ö£s \©zxÁ® {»Â¯x J¸ÁÝUS J¸zv J¸zvUS J¸Áß GßÓ EÓÄU Pmk¨£õkPÒ CÀ»õu _u¢vµ©õÚ EÓÄa `ÇÀ \‰PzvÀ HØÖU öPõÒÍ¨£mhx. GÚ÷Á SÇ¢øuPÒ uõ# EÓÂß ÁÈ°÷»÷¯ Aøh¯õÍ¨£kzu¨£mhÚ (\•uõ¯¨ £S¨£õ#Ä&£U42) CzuøP¯ {ø»°À C¸¢u ö£sPÒ Põ»\|PºÂÀ u¢øu öÁÎ \‰P©õP ©õØÓ® Aøh¢x P¸ÂPÒ Psk¤iUP¨£mh ¤ß ö£s RÌ{ø»¨£kzu¨£mhõÀ EØ£zvUPõÚ P¸ÂPÒ Psk¤iUP¨£mh ¤ß ö£sPøÍ Kµ® Pmi Âmk BsPÎß BvUP® EØ£zv°À PõÀ JßÔ¯x. ö£s Ãmk ÷Áø» ©mk÷© £õºzuõÀ ÷£õx® GßÓ {ø» E¸ÁõUP¨£mk Aiø©zuÚ® Á¾¨ö£ØÓx. Cx÷Á ö£sPÒ Bøna \õº¢x ÁõÊ® Aiø© {ø»ø¯ EÖv¨£kzv¯x (\•uõ¯¨ £S¨£õ#Ä& £U42) GßÖ ]Á©õø» µõ\õ SÔ¨¤kQÓõº. JßÓõP Cøn¢x EØ£zv°À Dk£mh Põ»•® ö£sPÎß SÇ¢øu ÷£ØÔß Põ»•® Cøn¢x ö£sPøÍ AhUS •øÓUSÒ uÒÎ¯x Gß£øu ³QUP •iQÓx. £UP® 69 ,yf;fpaq;fs; fhl;Lk; ngz; epiyg;ghL; Ãµ©PÎº \[P Põ» PmhzvÀ ö£sPÒ AhUS•øÓ ÁõÌøÁ ÁõÌ¢uõ¾® Ãµ® öŒÔ¢uÁºPÍõPÄ® ÁõÌ¢xÒÍ Gß£øu AÔ¯»õ®. CßÖ® ö\¸¨£øÓ ÷Pmk Â¸¨¦ØÖ ©¯[Q ÷ÁÀøPU öPõkzx öÁÎxÂ›zx EjC¨ Skª Gsön# }Â J¸©Pß AÀ»x CÀ÷»õÒ ö\¸•P ÷\|õUQa ö\ÀP GÚ Âk÷©! (¦Ó®: 279) GßÓ AiPÎÀ xUTº ©õ\õzv¯õº GßÓ ¦»Áº £õkQÓõº. •uÀ \|õÒ \|h¢u F›À u¢øu CÓ¢xÂmhõº. Auß ¤ß uÚx PnÁÝ® CÓ¢xÂmhõº. C¢{ø»°À \|õmøhU PõUP Gso¯ A¢u¨ ö£s ÷ÁÖ ÁÈ°ßÔ CßÖ® ÷£õº •µ_ J¼¨£øu ÷Pmk ©QÌa] öPõsk uß J÷µ ©PøÚ ]ßÚg]Ö £õ»PøÚ AøÇzx AÁÝUS öÁÒøÍ xo Ekzv AÁÚx uø»US Gsön# uhÂa^Â Skª•izx AÁøÚ ÷£õºUPÍ® ÷\|õUQ ÷£õ GßÖ AÝ¨¤ øÁzuõÒ GßÓ ö\#vø¯ @©ØSÔzu £õh»iPÎÀ CuøÚ AÔ¯ •iQßÓx. {»Áøh \‰P® {»Â¯ ÷ÁøÍ°À \|õk Gß£x uÚUPõÚ ö\õzuõP¨ £õºUP¨£kQÓx. AuÚõÀ GÁ¸øh¯ £øhö¯k¨¦US® J¸ \|õk Bs ö£s ÷£uªßÔ u¯õº {ø»°À C¸UP ÷Ásk® GßÓ P¸zuõUP® AøÚÁ›ß ö\|g]¾® BÇ©õP £v¢xÒÍuõÀ J¸ ö£s CzuøP¯ ö\¯À£õkPÎÀ Dk£kQÓõÒ Gß£øu AÔ¯ •iQÓx. PÎöÓÔ¢x £mhÚ öÚßÝ •ÁøP DßÓ bõßÔÝ® ö£›÷u Pspº ÷\|õßPøÇ ¯»®Á¸® öÁvµzx Áõßö£¯z y[Q¯ ]u›Ý® £»÷Á. (¦Ó®:277) 70 £UP® ]ØÔÀ \|Ø×s £ØÔ {ß©Pß ¯õskÍ÷Úõ GÚ ÂÚÄv Gß©Pß ¯õsk EÍÚõ°Ý® AÔ÷¯ß K¸® ¦¼ ÷\º¢x ÷£õQ¯ PÀ AøÍ ÷£õ» DßÓ Á°÷Óõ Cx÷Á ÷uõßÖÁß ©õ÷uõ ÷£õºUPÍzuõ÷Ú. (¦Ó®: 86) Áõk •ø» FÔa _µ¢uÚ Khõ¨ §møP Âhø» uõ#U÷P. (¦Ó® 295) £øh AÈ¢x, ©õÔÚß’ GßÖ £»º TÓ ‘©sk A©ºUS Eøh¢uÚß B°ß, Esh Gß •ø» AÖzvköÁß, ¯õß’ GÚa ]øÚ, ÁõöÍõk £k ¤n® ö£¯µõ ö\[PÍ® xÇÄ÷ÁõÒ, ]øu¢x ÷ÁÖ BQ¯ ©Pß QhUøP PõrE, DßÓ bõßÔÝ® ö£›x EÁ¢uÚ÷Í! (¦Ó®.278) BQ¯ £õhÀ AiPÒ AøÚzx÷© ö£sPÎß Ãµ ö\¯ÀPøÍ EnºzxQßÓÚ. uß©Pß CÓ¢u ö\#vø¯U ÷Pmhuõ# uß ©õºø£ AÖzx GÔ÷Áß GßÖ TÖÁx® ¦¼ C¸¢u SøPø¯¨ ÷£õ» AÁß Gß Á°ØÔÀ C¸¢uõß Gß£x® AÁºPÎß Ãµ©õP \‰P® £õºUQÓx. Esø©uõß ö£sPÒ uß Aiø© {ø»ø¯ Enµõ©À \|õk •UQ¯® Gß£øu÷¯ SÔU÷PõÍõPU öPõskÒÍÚº. AuÚõ÷»÷¯ AÁºPÒ Ãµzuõ¯õP A[RP›UP¨£kQÓõºPÒ. •iÄøµ \[P Põ»® Gß£x AUPõ» \‰P ÁõÌÄPøÍ ¤µv£¼US® Âu©õP Aø©QÓx. CvÀ ö£sPÎß ÁõÌÄ Gß£x uÛzxÁ©õÚ Aøh¯õÍ©õP ÂÍ[SQÓx. CÚUSÊ ÁõÌÄ Gß£x HÓUSøÓ¯ C¸£õ»¸US® \©©õP ÁõÌ¢x Á¢uÚº Gß£øu AÔ¯ •iQÓx. {»Äøhø©a \‰P©õP ©õÔ¯ ÷Áø»°À ö£sPÒ Cµshõ® uµ Si©UPÍõP÷Á BUP¨£mhÚº. N`hypfpuh]; N`hk; rapd;]; fy;Y}hp> J}j;Jf;Fb SHANLAX International Journal of Arts, Science and Humanities \[PPõ»¨ ö£sPÒ PnÁß CÓ¢u ¤ß¦ øPik ÁõÌøÁ ÷©ØöPõÒÁx® ©Ö©n® ö\#¯õ©À C¸¨£x® \‰P ÁÇUP©õP Aø©¢v¸¨£øu AÔ¯ •iQÓx. C»UPn ¡ÀPÐ® ö£sPÎß JkUS •øÓPÐUS HØÓõº ÷£õ» PØ¦ GßÓ ÷Põm£õmøh vo¨£uøÚ²® AÔ¯ •iQÓx. A÷u ÷ÁøÍ°À \‰PzvÀ ö£sPÒ {»® \|õk ÁõÌÄ GßÖ Á¸® ÷£õx AÁºPÒ Ãµ¨ ö£sPÍõP ©õÖÁuøÚ £»¨ £õhÀPÎÀ Põn•iQßÓÚ. C¸÷ÁÖ {ø»PÎÀ ö£sPÎß Aø©¨¦ C¸¨£uøÚ CUPmkøµ ÁÈ AÔ¯ •iQÓx. http://www.shanlaxjournals.com xøn{ßÓøÁ 1. \s•P®¤ÒøÍ. • (£.A) & öuõÀPõ¨¤¯®, •Àø» {ø»¯®, 9, £õµv\|Pº •uÀ öu¸, v.\|Pº, ö\ßøÚ. 2. _¨¤µ©o¯ß, \[PPõ» ÁõÌÂ¯À (\[Pz uªÇ›ß Aµ_•øÓ²® \‰P ÁõÌÄ®), {² ö\g_› ¦U PÄì (¤) ¼m, 41& ]m÷Põ Cshìi›¯À Gì÷hm, A®£zyº, ö\ßøÚ. 3. ö\£©õø» µõ\õ & \•uõ¯¨ £S¨£õ#Ä, øÁPøÓ £v¨£P®, vskUPÀ. 4. ©õoUPÁõ\Pß. bõ (E.B) & ¦Ó\|õÞÖ, E©õ£v¨£P®, ö\ßøÚ. £UP® 71
https://openalex.org/W4296955361	https://www.preprints.org/manuscript/202206.0400/v1/download	English	null	Securing Flows in the River Systems through Irrigation Water Use Efficiency—A Case Study from Karula River in the Ganga River System	Water	2,022	cc-by	15,630	Case Report Securing Flows in the River Systems through Irrigation Water Use Efficiency - A Case Study from Karula River in the Ganga River System Case Report Nitin Kaushal1, Suresh Babu1, Arjit Mishra1, Rajesh Bajpai1, Phanish Kumar Sinha2, Rama Kant Arya3, Dave Tick- ner4 and Conor Linstead4 Nitin Kaushal1, Suresh Babu1, Arjit Mishra1, Rajesh Bajpai1, Phanish Kumar Sinha2, Rama Kant Arya3, Dave Tick- ner4 and Conor Linstead4 1 World Wide Fund for Nature – India (WWF India) 2 Water Resources & Irrigation Expert and partner WWF India 3 Irrigation Expert and Partner WWF India 4 World Wide Fund for Nature UK *Corresponding author:nkaushal@wwfindia.net Abstract: The pressure on freshwater resources is leading to diminishing flows in some of the critical river systems across the globe and India is no exception and this is mainly because of water with- drawal for irrigation, which is often to the tune of 70% to 80% with some proportion for domestic and industrial use. While graduating from the concept of environmental flows and its assessment methodologies in India, the water-managers, the researchers and the conservationists are now mov- ing towards answering the next question if the rivers are to be revived, then where will the water come from, especially in the case of over-allocated rivers, including the river Ganga. While the log- ical way is to look at the biggest user of water, i.e. irrigation, it remains to be seen whether the irrigation water savings will actually lead to enhancing flows in a river, complementing the efforts towards maintaining e-flows in rivers, or whether it will lead to more area under agriculture, bring changes in cropping patterns towards more water-intensive crops or result in something else. This is a growing debate across the globe, where India is no exception, and there has been a wide range of opinions in this regard. This paper discusses the process, findings and lessons from a joint initia- tive involving farmers, the Uttar Pradesh state Irrigation and Water Resources Department, Bijnor District Administration and a conservation organisation to enhance flows in a rivulet, called karula River, which is part of the Ganga river system. Another objective of this paper is to look at the scalability and replicability of similar approaches in other irrigation command areas to benefit nearby river systems in general. Under this initiative, the team attempted to enhance flows in the river Karula by routing the saved water from irrigation supplies in a canal commanded area. Case Report Securing Flows in the River Systems through Irrigation Water Use Efficiency - A Case Study from Karula River in the Ganga River System This saving of water is being achieved due to ‘supply-side’ and ‘demand-side’ measures that are being adopted in the project area. With the objective of ensur- ing the sustainability of the initiative, efforts are made to form an institutional arrangement, through which this initiative can be sustained beyond the project support. Keywords: Ganga; environmental flows; river conservation; Ramganga; Karula; irrigation water use efficiency; Water Users Association; minor canal Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 b. reconciling multiple competing human demands for water, further com- pounded by changing lifestyles, market-driven processes and unplanned de- velopmental activities b. reconciling multiple competing human demands for water, further com- pounded by changing lifestyles, market-driven processes and unplanned de- velopmental activities c. ensuring sustainable water use, in line with Sustainable Development Goals 6 (SDGs) which calls for ‘ensuring availability and sustainable management of water and sanitation for all’ c. ensuring sustainable water use, in line with Sustainable Development Goals 6 (SDGs) which calls for ‘ensuring availability and sustainable management of water and sanitation for all’ Grill et al. (2015) concluded that, globally 48% of river volume is moderately to se- verely impacted by either flow regulation, fragmentation, or both. This situation calls for maintaining or restoring flow regimes, in the form of environmental flows, to ensure the maintenance of ecological integrity. The most referred definition of e-flows by Arthington et. al. (2018) is ‘the quantity, timing, and quality of freshwater flows and levels necessary to sustain aquatic ecosystems which, in turn, support human cultures, economies, sus- tainable livelihoods, and well-being’. g The inclusion of environmental flows in IWRM (Integrated Water Resources Man- agement) will result in increased effectiveness of environmental outcomes along with many benefits to social well-being and economic return (Hirji and Davis 2009). Environ- mental flows can form the basis for an integrated approach to water allocation and river operation. Identifying environmental flows is likely to provide a strong scientific and open process within river management and for water allocation decisions at a basin scale (Overton et. al. 2014). Currently 41% of global irrigation water use occurs at the expense of e-flows require- ments and India contributes to about 17.7% of global annual e-flows deficit (both in terms of the total annual deficit and the number of months with transgressions) (Jagermeyer et. al. 2017). The river basins in South Asia (including the Ganga basin), in the Mediterranean region, and the Sahel are most sensitive to irrigation improvements resulting from the combination of local crop types, climate and soil conditions and the current irrigation sys- tem. India has 18% of world population, having 4% of world’s freshwater, of which 80% is used in agriculture (Dhawan 2017). Excessive use of surface-water and ground-water for irrigation has led to a diminish- ing water-table and the transformation of perennial rivers into seasonal ones. 1. Introduction and context Rivers, wetlands and aquifers are a critical source of water for nature, biodiversity and human beings. In fact, these sources have their own inter-dependent ecosystems. All of these ecosystems face multiple challenges, in the wake of: y p g a. feeding a growing population in a changing climate, while also conserving and restoring nature Recent History of E-Flows Assessment and Implementation in India Recent History of E Flows Assessment and Implementation in India In the Indian context, a consortium of Indian Institutes of Technology, along with other partners developed Ganga River Basin Management Plan (GRBMP). This group defined E-Flows as, ‘a regime of flow in a river or stream that describes the temporal and spatial variation in quantity and quality of water required for freshwater as well as estuarine systems to perform their natural ecological functions (including sediment transport) and support the spiritual, cultural and livelihood activities that depend on these ecosystems’. WWF-India (World Wide Fund for Nature – India) has also been working towards E-Flows assessment and implementation, testing an assessment methodology with a multidisciplinary team of experts from other institutions and demonstrating field level interventions with local stakeholders. There are several initiatives from the government, civil society and academia who are working towards securing Environmental Flows in the river systems in India (updated from Gopal, 2013): o Minimum Flows – stipulations by Central Water Commission, Govt. of India 1992 o Deliberations and recommendations around E-Flows in Indian rivers by National Insti- tute of Technology and International Water Management institute in 2001 o E-Flows assessment (Water Quality Assessment Authority) – Govt. of India 2003-07 o Macro-level broad E-Flows assessment for Indian rivers by International Water Man- agement Institute 2007 o Upper Ganga E-Flows assessment by a multidisciplinary team & WWF-India 2008-10 o Upper Ganga E Flows assessment by a multidisciplinary team & WWF India 2008 10 o Aquatic species-centric E-Flows assessment for Upper Ganga by Wildlife Institute of India 2010-2011 o Hydrology-based E-Flows assessment for Upper Ganga by Alternate Hydro Energy Centre 2010-2011 o E-Flows assessment by consortia of IITs for Himalayan stretch of river Ganga 2011. The initiative was part of development of GRBMP o National Water Policy 2012, which called for maintaining E-Flows in river systems o E-Flows for river Ganga by a multidisciplinary team, led by WWF-India for Triveni Sangam, Prayagraj location, Kumbh 2013 (Tare Vinod et. al. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Stockle (2007) noted that withdrawing surface water implies changes to the natural hydrology of rivers and water streams, affecting the aquatic ecosystems associated with these water bodies. In the Indian context, the concept of e-flows is in the process of being mainstreamed in river basin management. However, barring a few exceptions, efforts have been largely centred around the Ganga river, within which the focus has been on developing the tech- nical foundations for e-flows assessment. Efforts are also being made to understand the tradeoffs, in cases where e-flows are to be maintained. However, the implementation of e-flows remains elusive and there is a particular need for practical case studies document- ing how irrigation management can aid maintaining e-flows. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Recent History of E-Flows Assessment and Implementation in India Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Tickner et. al. (2020) pointed out that case studies of environmental flows implemen- tation, successful or otherwise, provide valuable insights into barriers and enabling fac- tors, and illustrate the evolution and propagation of the practice of environmental flows globally. Kaushal et. al. (2019) documented approaches to understand and resolve poten- tial trade-offs between environmental flows objectives for the Ganga river in Uttar Pra- desh and agricultural water demand. They concluded that, contrary to common percep- tions, the increase in water needed to restore flows is likely to be small, compared to over- all water demand. Moreover, agricultural water use efficiency measures can ameliorate the potential adverse impact on farmers from changes in water allocation. The National Commission for Integrated Water Resources Development, Govern- ment of India had estimated total withdrawal/utilisation for 2010 for all types of uses as 710 BCM in a high projection scenario. Of this, irrigation accounted for nearly 78%, fol- lowed by domestic use of 6%, industries at 5%, power development at 3%, and other ac- tivities claimed about 8% including evaporation losses, and environment and navigational requirements (CWC 2020). With this background, it becomes imperative to engage with the irrigation and agriculture sector around water use efficiency, if freshwater resources (rivers, lakes and wetlands) are to be conserved. This paper reports on the process and lessons from an initiative to enhance flows in the Karula river, through the implementation of supply-side and demand-side measures in the Khanpur Minor canal command area in Bijnor district of Uttar Pradesh in India. The basic premise of this work is an ask – can we help secure e-flows in the river, through interventions in the irrigation sector, while maintaining sustainable and enhanced water and land productivity levels, with improved overall agricultural production? p y p g p Whilst the Karula river initiative is a case in point to discuss how flows can be en- hanced through promotion of Better Management Practices (BMPs) in agriculture and ir- rigation it is also an example of how flows in overallocated rivers can be secured by pilot- ing, upscaling and mainstreaming similar approaches in the command area of irrigation systems that offtake from rivers,diverting river waters for irrigation purposes. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 A carefully crafted approach encompassing demand-side management, while ensuring efficient irri- gation system and institutional support, can actually pave the way for managing trade- offs in a scenario where water re-allocation becomes inevitable in the wake of required e- flows releases from the dams and barrages. Recent History of E-Flows Assessment and Implementation in India 2013) o E-Flows initiative for Ramganga 2013 (Kaushal Nitin, Babu Suresh, Mishra Arjit, O’Keeffe Jay 2018) and continuing, led by WWF-India o E-Flows initiative for Ramganga 2013 (Kaushal Nitin, Babu Suresh, Mishra Arjit, O’Keeffe Jay 2018) and continuing, led by WWF-India o The Ministry of Environment, Forests & Climate Change (Government of India) in the standard Terms of Reference for conducting the Environmental Impact Assessment studies for any proposed River Valley and Hydro Project stipulated seasonal percent- age of E-Flows that are required to be maintained o Ganga Notification 2016 by Government of India to call for maintaining E-Flows in Ganga (National Mission Clean Ganga Gazette Notification, Government of India) o Ganga E-Flows Order 2018 and Amendment 2019 by Govt. of India, stipulating E- Flows values for Ganga river (E- Flows Gazette Order 2018 & Amendment 2019) o A joint initiative to assess E-Flows in all major rivers of Uttar Pradesh is underway (2019-22) by Uttar Pradesh Water Management & Regulatory Commission, Uttar Pra- desh State Water Resources Agency and World Wide Fund for Nature – India. Under this initiative, the E-Flows assessment is done for Sharda, Ghaghra (Saryu), Gomti, Rapti, Yamuna, Son, Gandak rivers and plus some additional sites on Ganga River (where E-Flows assessment was not done earlier). The purpose of this exercise is to in- form the exercise on River Basin Management Plans for these respective rivers. Project area Under the Karula river pilot project, the aspiration has been to enhance the diminish- ing flows in the Karula river, a tributary of the Ramganga River, from the saved water from the irrigated command area of a minor canal, called Khanpur Minor. The catchment area of Karula river is 957 sq. km., which is little over 4% of the catchment area of Ram- ganga basin (25,028 sq. km.). This canal system is operated and maintained by Uttar Pra- desh Irrigation & Water Resources Department (UPI&WRD) (Figure 1). Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Figure 1. line diagram to explain the Karula Pilot concept. Ramganga water resources are stored in a reservoir called Kalagarh dam, which is the second biggest dam-based reservoir (after Tehri dam) in the state of Uttarakhand, stor- ing over 2448 Million Cubic Metres of water (National Register of Large Dams 2019, Cen- tral Water Commission). The Kalagarh Multi-purpose Project was designed for irrigation, flood protection and production of electricity with an installed capacity of 198 MW (Utta- rakhand Irrigation Department). The major proportion of water in Kalagarh dam is allo- cated to augment the lower Ganga Canal System (85%) while the rest is allocated to small independent canal systems known as the Ramganga sub feeder (10%) and Pheeka canal system (5%) – as per the information from authorities. A l f h di i f R f i i i l h fl i h Not to scale map Figure 1. line diagram to explain the Karula Pilot concept. Not to scale map Figure 1. line diagram to explain the Karula Pilot concept. Ramganga water resources are stored in a reservoir called Kalagarh dam, which is the second biggest dam-based reservoir (after Tehri dam) in the state of Uttarakhand, stor- ing over 2448 Million Cubic Metres of water (National Register of Large Dams 2019, Cen- tral Water Commission). The Kalagarh Multi-purpose Project was designed for irrigation, flood protection and production of electricity with an installed capacity of 198 MW (Utta- rakhand Irrigation Department). The major proportion of water in Kalagarh dam is allo- cated to augment the lower Ganga Canal System (85%) while the rest is allocated to small independent canal systems known as the Ramganga sub feeder (10%) and Pheeka canal system (5%) – as per the information from authorities. Project area As a result of the diversion of Ramganga waters for irrigation canals, the flows in the Ramganga downstream barrage (Hareoli Barrage) are miniscule for the middle stretch of the Ramganga river. The lower stretch of the Ramganga, however, just before joining the Ganga, is relatively better due to contributions from the tributaries in the middle to lower stretches of the Ramganga River. The Ramganga sub feeder canal system takes off from Kho Barrage built on the Kho river in Bijnor district (see Figure-1). This Ramganga sub feeder main canal has a series of minor canal systems extending irrigation supplies to the farms in three districts of Uttar Pradesh; one such minor canal is called the Khanpur Minor Canal, having a designed dis- charge of over 3.5 cubic feet/second). The irrigation command area of this canal system largely falls in four villages (Khanpur, Meerapur, Rehtoli, Kolasagar) of Seohara Block in Bijnor district of Uttar Pradesh. Figure – 2 illustrates the location of the pilot area on the map of the country. 2. Approach and Methods The idea of the Karula pilot has been conceived keeping in view a stakeholder-centric participative approach, wherein the farmers, concerned state government institutions (Ut- tar Pradesh Irrigation and Water Resources Department) and district authorities (Bijnor district) were key stakeholders. Whilst the project team led and coordinated the entire task; the stakeholders, local knowledge and wisdom played a critical role, in terms of con- textual guidance, rapport building and farmer-level coordination. A three-pronged approach was adopted to implement the pilot activities, including: a. Demand-Side Management (promotion, demonstration and adoption of irri- gation water used in efficient ways and means, in terms of Better Manage- ment Practices, to save water) b. Supply-Side Management (rehabilitation of the entire canal system of Khan- pur Minor, including the construction of a passage from the tail-end of Minor to the riverbank of Karula) c. Institutional Strengthening (facilitation of the constitution of the Khanpur Minor Water Users Association and capacity building of command farmers to make them well-acquainted with various key provisions of the Uttar Pra- desh Participatory Irrigation Management Act, 2009 – under which the Wa- ter Users Associations are formed in the state) Key amongst the above three aspects of the approach has been the inclusion of socio- economic aspects, technical considerations, and stakeholder engagement. During imple- mentation of the three-pronged approach, these aspects were not only taken into account, but were of central focus. The Karula river initiative began with the assessment of baseline information per- taining to farmers, their landholdings, literacy rate, cropping cycle and cropping pattern, modes of irrigation, agricultural yield, input cost, profit margins, the status of canals and allied infrastructure, and more. With the increased understanding about the area, the work began, wherein the role of various stakeholders (including command farmers, Uttar Pradesh Irrigation and Water Resources Department (UPI & WRD), Bijnor District Ad- ministration and WWF India) was critical. Stakeholder engagement is seen as a means of contributing to improved water gov- ernance, where governance is defined as the policy and practices giving rise to particular forms of water management in different contexts (Wehn et. al. 2018). Various stakeholders under the Karula initiative had played an inclusive and iterative part in realising the larger objective. Although their responsibilities were distinct with overlapping roles, they did appreciate each other's contribution and collaborated to work for the larger water conser- vation goal. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 river. This passage is a mix of open earthen and lined channel, with some portion as un- derground-pipeline. river. This passage is a mix of open earthen and lined channel, with some portion as un- derground-pipeline. This paper documents the journey of the Karula river initiative from mid-2017 to mid-2021. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 doi:10.20944/preprints202206.0400.v1 Figure 2. Location map of the Karula pilot area. Figure 2. Location map of the Karula pilot area. The farmers in the catchment of the Karula river predominantly grow sugarcane, not only because of rich water resources and the presence of sugar-mills in the nearby areas (Seohara and Dhampur), but also due to the high economic value of sugarcane, as a cash crop, and the prevailing Minimum Support Price, which attracts farmers for assured in- comes. According to a broad estimate, about 67% of the Khanpur Minor command area, i.e., about 260 ha, grows sugarcane and on the rest of the command, the usual wheat- paddy is grown. (Landuse Map, WWF-India – available as Appendix 2) p y g ( p pp ) The key statistics of the Khanpur Minor Canal are tabulated in Table – 1: Table 1. Main Features of Khanpur Minor Canal System, including CCA (Culturable Command Area, the area which can be physically irrigated from a scheme and is fit for cultivation) and PPA (Proposed Protected Area, the area that is assured for irrigation by a scheme). S. No. Item 1 Length of Khanpur Minor Canal About 3 kilometres 2 CCA (Culturable Command Area) PPA (Proposed Protected Area) a. Rabi (Cropping season from July to October) b. Kharif (cropping season from November to March / April) 389 Hectare 148 ha 124 ha 3 Number of Farmers 311 4 Passage to connect tail-end of Canal with the nearest Karula river-bank (constructed as part of this initiative) Over 554 meters The tail end of the Khanpur Minor canal system is about 554 metres (acceptable route) from the left bank of the river Karula. Therefore, one of the tasks under this initiative was to construct a passage to connect the tail-end of the canal with the left bank of the Karula Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Knowledge Exchange programmes (Exposure Visits & on farm sessions) & efficient irrigation techniques implemented by farmers – since late 2017 SCOPING: Baseline generation, field surveys and deliberations with farmers in the general command area of Major Irrigation Project to identify a sub-system (Khanpur Minor) for intervention – early 2017 The upper and middle reach farmers, along with District Authorities helped in convincing tail-end farmers to have constructed the passage joining the Karula rivers – early 2018 Khanpur Minor WUA constituted – February 2021 The farmers upscaled Demand-side interventions in terms of more agriculture area under Better Management Practices – ongoing Farmers’ and Irrigation Department field functionaries’ capacity built on institutions to sustain the initiative – since early 2019 Preliminary deliberations (including obtaining permissions) with officials and field functionaries of Irrigation Department to select Khanpur Minor for intervention – mid 2017 Deliberations and formal meetings at Village level Institutions (to seek formal support for Karula initiative) in each of the villages falling within the command area of Khanpur Minor late 2017 Preliminary deliberations (including obtaining permissions) with officials and field functionaries of Irrigation Department to select Khanpur Minor for intervention – mid 2017 Deliberations and formal me level Institutions (to seek fo Karula initiative) in each of within the command area of late 2017 Deliberations and formal meetings at Village level Institutions (to seek formal support for Karula initiative) in each of the villages falling within the command area of Khanpur Minor late 2017 Preliminary deliberations (including obtaining permissions) with officials and field functionaries of Irrigation Department to select Khanpur Minor for intervention – mid 2017 Progressive farming techniques explored. Package of Practices developed. Knowledge Exchange programmes (Exposure Visits & on farm sessions) & efficient irrigation techniques implemented by farmers – since late 2017 SCOPING: Baseline generation, field surveys and deliberations with farmers in the general command area of Major Irrigation Project to identify a sub-system (Khanpur Minor) for intervention – early 2017 The upper and middle reach farmers, along with District Authorities helped in convincing tail-end farmers to have constructed the passage joining the Karula rivers – early 2018 Farmers’ and Irrigation Department field functionaries’ capacity built on institutions to sustain the initiative – since early 2019 Figure 3. Stakeholder’s engagement along with timelines and key steps and milestones. 2. Approach and Methods For instance, the supply side-interventions (rehabilitation and maintenance) on the Khanpur Minor canal is a Uttar Pradesh Irrigation and Water Resources Depart- ment task, but farmers and other stakeholders played a critical role in the overall super- vision and coordination. On the other hand, a passage was required to be constructed to connect the canal’s tail-end with the riverbank, which was purely a physical activity. Here, the technical guidance of the Uttar Pradesh Irrigation and Water Resources Department was obtained, yet the farmers played the key role, as they deliberated and finalised the alignment of the passage route. In this process, the involvement of district authorities was critical to pro- vide information about the rights (based on revenue records) on the land between the passage route. Only then could all stakeholders take the final call on the passage route and the work begin. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 ‘Social learning’ has been an added advantage of such stakeholder-centric ap- proaches. One of the most salient aspects of social learning is the collective – rather than individual – process of learning, knowledge co-creation and accumulation of wide expe- riences to generate a broader knowledge and evidence base, from which decisions can be taken (Wehn et. al. 2018). In terms of the Karula initiative, it has been a mutual learning for all stakeholders. For instance, whilst the team promoted trench-based sugarcane farm- ing in the Khanpur Minor canal command, farmers came up with the idea of multi-crop- ping by making use of moisture in the soil, and therefore growing other crops to maximise their economic gains. Some of the progressive farmers in the adjoining areas as well as the Department of Sugarcane, Government of Uttar Pradesh (Success Stories of Sugarcane de- velopment in District-Bijnor, Uttar Pradesh, 2018) were promoting these practices. As a result of knowledge exchange, exposure visits and personal initiatives, many farmers adopted this idea. This has also become a learning for their fellow farmers (even outside the com- mand area) The chronology of the stakeholders’ engagement process is illustrated in Figure 3. Figure 3. Stakeholder’s engagement along with timelines and key steps and milestones. Preliminary deliberations (including obtaining permissions) with officials and field functionaries of Irrigation Department to select Khanpur Minor for intervention – mid 2017 Deliberations and formal meetings at Village level Institutions (to seek formal support for Karula initiative) in each of the villages falling within the command area of Khanpur Minor late 2017 Progressive farming techniques explored. Package of Practices developed. a. Facilitate to identify passage route & its formation b. Convince command farmers to support this initiative ( Preferred route towards river Karula has a lot of encroachments by tail end farmers (mainly extension of farm boundaries). Therefore, sparing the space for passage route was one of the most challenging and complex tasks. p a. Agree on passage route . Convince fellow farmers for the initiative c. Support while formation of passage d. Maintain passage (as WUA function), beyond project duration Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Table 2. Summary of Roles, Challenges and Approach for this study. i. Supply Side Interventions Roles of various stakeholders Challenges Approach adopted to resolve WWF – India UPI &WRD Farmers District Authorities Canal works: a. Canal de- siltation b. Canal gate repair to ensure off- take of Designed Discharge throughout the canal c. Repair and Maintenance of Outlet heads d. Setting up hydrological monitoring system at Khanpur Minor and Karula River a. Permissions to carry out proposed work b. Technical guidance in carrying out canal works, i.e. repair & maintenance c. Technical supervision & monitoring of physical works d. Regular maintenance and repair post- a. Agree to become Ramganga Mitra b. Participate in field surveys on canal for identification of works c. Supervision of physical works on canals d. Report any issue to the authorities and team a. Support to carry out the work and provide contextual guidance, as required b. facilitate Institutional synergies, i.e. to facilitate support from other departments for the purpose of the work Canal system was in a dilapidated state, the passing of designed discharges from the head of the canal was not possible, plus several obstructions in the canal and therefore the tail- end area of canal generally remained un-fed Complete rehabilitation of canal system was done, including – head-works repair, canal desilting, fixing of outlet head- pipes, Gauges repair & establishing new Gauge, clearing of obstructions etc. Passage falls under tail-end village of command Series of Table 2. Summary of Roles, Challenges and Approach for this study. Canal works: a. Canal de- siltation b. Canal gate repair to ensure off- take of Designed Discharge throughout the canal c. Repair and Maintenance of Outlet heads d. Setting up hydrological monitoring system at Khanpur Minor and Karula River Canal-end to river-bank a. Permissions to carry out proposed work b. Technical guidance in carrying out canal works, i.e. repair & maintenance c. Technical supervision & monitoring of physical works d. Regular maintenance and repair post- intervention a. Agree to become Ramganga Mitra b. Participate in field surveys on canal for identification of works c. Supervision of physical works on canals d. Report any issue to the authorities and team c Canal works: a. Canal de- siltation b. Canal gate repair to ensure off- take of Designed Discharge throughout the canal c. Repair and Maintenance of Outlet heads d. Preferred route towards river Karula has a lot of encroachments by tail end farmers mainly extension of farm boundaries). Therefore, sparing the space for passage route was one of the most challenging and complex tasks. Passage falls under tail-end village of command. Series of deliberations held with farmers and they were exposed to (i) the benefits of adopting improved practices and (ii) how they can contribute to a healthy Karula. Farmers got convinced to provide passage, but requested that most of the passage route should be underground and part of it should be on the edges of the farms to avoid damage to crops Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 The engagement of various stakeholders, in a categorised (activity-based) manner is explained in the below table along with their specific roles, the challenges that were faced and how these challenges were overcome. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Setting up hydrological monitoring system at Khanpur Minor and Karula River Canal-end to river-bank passage work: a. Along with farmers and department, identify most preferred route from tail-end of canal to Karula riverbank b. Build consensus on the route and type of passage c. Construction of passage in accordance with consensus a c g o c p Canal works: a. Canal de- siltation b. Canal gate repair to ensure off- take of Designed Discharge throughout the canal c. Repair and Maintenance of Outlet heads d. Setting up hydrological monitoring system at Khanpur Minor and Karula River Canal-end to river-bank passage work: a. Along with farmers and department, identify most preferred route from tail-end of canal to Karula riverbank b. Build consensus on the route and type of passage c. Construction of passage in accordance with consensus a g p Canal works: a. Canal de- siltation b. Canal gate repair to ensure off- take of Designed Discharge throughout the canal c. Repair and Maintenance of Outlet heads d. Setting up hydrological monitoring system at Khanpur Minor and Karula River Canal-end to river-bank passage work: a. Along with farmers and department, identify most preferred route from tail-end of canal to Karula riverbank b. Build consensus on the route and type of passage c. Construction of passage in accordance with consensus a. Permissions t carry out proposed work b. Technical guidance in carryin out canal works, i.e repair & maintenance c. Technical supervision & monitoring of physical works d. Regular maintenance and repair post- intervention a. Convincing the farmers about passage formation, its route selection and support consensus building b. Technical supervision of passage constructio Canal works: a. Canal de- siltation b. Canal gate repair to ensure off- take of Designed Discharge throughout the canal c. Repair and Maintenance of Outlet heads d. Setting up hydrological monitoring system at Khanpur Minor and Karula River Canal-end to river-bank passage work: a. Along with farmers and department, identify most preferred route from tail-end of canal to Karula riverbank b. Build consensus on the route and type of passage c. Construction of passage in accordance with consensus a. Permiss carry out pro work b. Tech guidance in c out canal wor repair & maintenan c. Tech supervisio monitorin physical w d. Reg maintenanc repair po interventi a. Convi the farmers passage form its route sele and supp consensus bu b. Tech supervisio passage const c. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Repair and Maintenance of Outlet heads f Passage falls under tail-end village of command. Series of deliberations held with farmers and they were exposed to (i) the benefits of adopting improved practices and (ii) how they can contribute to a healthy Karula. Farmers got convinced to provide passage, but requested that most of the passage route should be underground and part of it should be on the edges of the farms to avoid damage to crops Preferred route towards river Karula has a lot of encroachments by tail end farmers (mainly extension of farm boundaries). Therefore, sparing the space for passage route was one of the most challenging and complex tasks. a. Along with a. Agree on passage route b. Convince fellow farmers for the initiative c. Support while formation of passage d. Maintain passage (as WUA function), beyond project duration a. Facilitate to identify passage route & its formation b. Convince command farmers to support this initiative towards rive Karula has a lo encroachments tail end farme (mainly extensi farm boundari Therefore, spar the space fo passage route one of the mo challenging a complex task a. Facilitate to identify passage route & its formation b. c. Construction of passage in accordance with consensus Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 ii. Demand Side Interventions Roles of various stakeholders Challenges Approach adopted to resolve WWF – India UPI &WRD Farmers Extension Agencies (Agriculture Science Center) a. Building the capacity of the farmers towards Better Management Practices (BMPs) in irrigation & agriculture for Sugarcane b. Demonstration of BMPs (Better Management Practices) & PoPs (Package of Practices) with farmers Cropping pattern vis-a-vis irrigation water delivery information, with respect to various reaches of the Khanpur Minor canal a. Agree to this initiative b. Participate in trainings and exposure c. Willingness to demonstrate BMPs & Package of Practices (PoPs) on their farms d. Implementation of BMPs & PoPs on their farms a. Progressive farming techniques b. Support in development of Package of Practices (PoPs) c. Knowledge Exchange, including exposure visits and on-farm sessions Sugarcane crop & flood-based irrigation is predominant in the region. Equitable distribution of water was a challenge. The situation aggravated by dilapidated state of canal & excess water being used by head-reach farmers leaving little for tail- enders. Surface water irrigation is 100% subsidized for farmer's welfare, so there was no economic incentive to use less water Being a cash crop, the recommendation for switching from sugarcane to another crop was deliberately not attempted. Therefore, the focus remained on improving the irrigation practices. The trench irrigation practice was introduced. Trench technique has not only resulted in reduction of canal water use but also reduced groundwater withdrawal, which certainly reduced input cost. In parallel, the farmers were sensitized for their role in reviving ii. Demand Side Interventions Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/pre doi:10.20944/preprints202206.0400.v1 recall’ method, this echo similar approach noted by Barton and Taron 2010, while con- ducting representative farm surveys in the irrigation command areas in Tungabhadra River Basin, India. There is a body of literature that talks about farmer-recall method as one of the means for conducting irrigation and agricultural surveys, especially in the absence of precise measuring and monitoring support. The analysis by Beegle et. al. 2011, as part of the work in three African countries, shows little evidence of recall bias impacting agriculture data quality at farm-level. They noted that, the results of their work allay some concerns about the quality of some types of agricultural data collected through recall over lengthy peri- ods. On the other hand, Wollburg et. al. 2020 find that, the recall length has a significant impact on reported outcomes in all areas of interest in agriculture surveys and analysis. They therefore suggested that, to reduce the risk of recall error and to improve the quality of key variables in agricultural surveys, shorter recall periods can be one of the solutions. The authors, therefore, collected the information from the farmers during different stages of sugarcane crop, i.e., during land preparation, sowing, input applications and harvesting. Whilst multiple visits and interactions could be resource and time intensive; but, since the Karula river initiative has been a 4-year one and the team happened to visit field numerous times, which made it possible for the team to visit the farms and have discussions with the farmers during different phases of the crop cycle. This aspect is in alignment with the suggestions made in previous studies and research (Wollburg et. al. 2021, Beegle et. al. 2011, Barton and Taron 2010). Besides the farmer surveys, for the pur- pose of validation of information related to water application at sample farms, the team also measured the discharge and water levels in the field channels and farms. g The farmer survey questionnaire was discussed with a small sample size, but with clear representation from all reaches of the canal system, i.e., two farmers each from the head, middle and tail end of the Khanpur Minor canal. The identification of head, middle and tail end of the canal is done by dividing the total length of the canal into three equal parts. ii. Demand Side Interventions Roles of various stakeholders Extension Agencies (Agriculture Science Center) a. Building the capacity of the farmers towards Better Management Practices (BMPs) in irrigation & agriculture for Sugarcane b. Demonstration of BMPs (Better Management Practices) & PoPs (Package of Practices) with farmers Cropping pattern vis-a-vis irrigation water delivery information, with respect to various reaches of the Khanpur Minor canal iii. Institutional strengthening (including constitution of Khanpur Minor Water Us- ers Association Roles of various stakeholders Challenges Approach adopted to resolve WWF – India UPI & WRD Farmers District Authorities a. Guide, support and facilitate the process for constitution of WUA, including –election process, voter list preparation & voter’s validation b. Trainings, Knowledge Exchange and Exposure Visits of command farmer’s to active WUAs in & outside the state a. Lead and coordinate the process for constitution of Khanpur WUA with ‘Government of Uttar Pradesh’ b. Conduction of elections e. Notify results & WUA constituted a. Khanpur Minor WUA constitution process b. Participate in the capacity building initiatives, including – training, exposure etc. Facilitate and support the WUA election process Although the State Government promulgated UP Participatory Irrigation Management Act’ 2009; but the process (farmer’s awareness, Voter-List preparation & its validation, election schedule etc.) for WUA formation was time-taking Series of awareness and training programmes were conducted. National & state-level exposure visits to successful WUAs were organised. The Voter List preparation and validation was facilitated. Khanpur Minor WUA is at place now. iii. Institutional strengthening (including constitution of Khanpur Minor Water Us- ers Association With an objective to assess the impact of Karula river initiative on the farmers with respect to (i) on-farm water management and water savings and (ii) agricultural produc- tivity and economic value of produce per unit of area, a detailed questionnaire (Appendix A) was developed. Based on this questionnaire, farmer surveys were conducted jointly by some of the authors between 2018-2019 (sugarcane cropping season). The farmer surveys were conducted through a combined approach, i.e., field-level measurements and ‘farmer Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 ideal (theoretical) crop water requirement is as per FAO (Food and Agriculture Organiza- tion) norms using CROPWAT – a tool for calculating crop water requirements. The mete- orological data of the nearest climatological station (Bareilly) was used. The rainfall data of the district Bijnor was taken and the value of crop coefficient “Kc” was taken from guidelines issued by CWC (Central Water Commission, Government of India) in 1984 (Technical Series 2: A Guide for Estimating Irrigation Water Requirement, Ministry of Ir- rigation, Water Management Division, New Delhi, May 1984). The theoretical irrigation water depth for sugarcane crop computed using FAO’s CROPWAT Program is calculated as 67.6 cm, including the 25% leaching requirement. Against this norm, the current irriga- tion water depth in control plots (without trench method) was calculated as 87.6 cm. The irrigation water depth in demo plots (with trench method) was calculated as 72.3 cm. The state of Uttar Pradesh promulgated the Uttar Pradesh Participatory Irrigation Management Act in the year 2009 and since then the constitution of Water Users Associa- tions (WUAs) at canal systems has been underway in a phased manner. So far, this work was done in project areas of the Uttar Pradesh Water Sector Restructuring Project (funded by the World Bank). Hence, WUA formation in this area (Khanpur Minor, around the Karula river) had not begun. Under the Karula initiative, WUA was considered as an ap- propriate participatory institutional mechanism to sustain and take forward this initiative. Work towards formation of WUA in the Khanpur Minor command area has been underway since 2018, with a series of awareness, sensitization and training programmes being conducted to build the capacity of farmers regarding WUA functioning, and its roles and responsibilities as per the Uttar Pradesh Participatory Irrigation Management Act, 2009 (UP PIM Act 2009). Exposure trips of farmers from Khanpur minor command area to successful WUAs in the state and outside the state have also been conducted. This way, a strong momentum was generated in favour of constituting the WUA and a critical mass of experts and vigilant farmers was readied to support the affairs of the WUA. Finally, in February 2021, the Khanpur Minor WUA was constituted with the unanimous election of its governing board members. The Khanpur Minor WUA was constituted following the provisions of the UP PIM Act 2009. 3. Results This section discusses the findings of a sample survey of farms at all reaches of the Khanpur Minor canal, i.e., head, middle and tail reaches of the canal. Farmers from both typologies of farms, i.e., where interventions are being carried out, and where agriculture is still being practised in a traditional manner, were interviewed. The data from these in- terviews were analysed and the results are presented in this section. This section, essentially narrates the following: This section, essentially narrates the following: 1. Water savings at farm level 2. Flows restored in the Karula river 3. Change in sugarcane productivity 4. Economic implications for the farmers and crop-water productivity Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 The unanimous election results indicated the overall positivity amongst the command farmers towards the initiative as well as the institutional setup. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/pre The farms where intervention (having BMPs) was made were noted as ‘Demonstra- tion-farms’ and the ones with usual agricultural practices (without BMPs) were named as ‘Control-farms’. It was noted while selecting the control farm, that both the control and demonstration farms belonged to similar specifications, except for sowing methods (with trencher and without trencher). Along with the field visit to all farms, detailed interactions based on the agreed questionnaire were conducted. Among all the command area farms, six sample demo plots (two each from head, middle and tail reaches of the canal) and correspondingly six control plots were selected to assess the impact and benefits of these interventions. Parallel to this, data on the running of the Khanpur Minor Canal and the water con- sumed in irrigation was collected on a fortnightly basis for the critical period. This analysis was carried out for kharif (July to October) and rabi (October to March) crop seasons. The hydrological observations at Khanpur Minor were carried out through monitor- ing of gauge levels, active channel width and velocity to calculate discharges, which were used for water accounting. The observed discharge data is not available for the Karula river, as there is no monitoring station on this small river. It therefore becomes imperative to establish baselines which could later be utilised for comparison with the volume of saved water from irrigation discharged in the Karula river to improve its health. The water used for sugarcane irrigation, both in demo and control fields, was com- pared with its ideal (theoretical) requirement. The actual discharge from tube wells with a 4-inch delivery pipe to the irrigation channel was measured at the site, using area veloc- ity method and volumetric measurement. On this basis, an average discharge of 0.5 cusecs (16 litres per second) was adopted. A primary survey was conducted to gather infor- mation regarding the actual running time of tube-wells in demo and control plots for each irrigation/season. The volume of water applied in a field was calculated by multiplying the discharge with water application time. The irrigation water depth applied to a plot was calculated by dividing the total volume of water applied by the area of the plot. The 3.2. Flows restored into the Karula river: Now, after the rehabilitation, the irrigation system is fully functional whenever the Khanpur Minor Canal gets water as per the roster1 issued by the UPI & WRD. The canal system is run as per the roster. The saved water from the canal is now released into the Karula river through the passage. The Khanpur Minor Canal generally runs for 6-8 months in a year (depending upon water availability in the reservoir and irrigation water demand by the command farmers). From May 2019 until June 2021, the Khanpur Minor canal, through the passage, discharged a total of 62.55 million litres of water saved from irrigation to the Karula river. This quantum of water flown into the river across 67 days from May 2019 to June 2021. The discharge from the tail end of the Khanpur canal into the Karula river within this period ranged from 0.12 -0.80 cusec, with an average flow rate of 0.42 cusec, which is 11% of the ‘designed discharge’ of Khanpur Minor canal. Graph-2 (A, B and C) shows the temporal variation in saved water discharged into the Karula river since May 2019. Graph 2 A&B: Flows Dependability Curve of Observed Discharges in Karula River at critical point & Hydrological Variations in Karula river over 2017 to 2021. by-turn fashion, within the irrigation system. 1 Roster – the mechanism of irrigation scheduling which defines the date and time of water distribution for various canals, in a turn- by-turn fashion, within the irrigation system. 3.1. Water savings at farm level: The sugarcane crop raised using traditional practices (primarily, flood irrigation) consumed more water, whereas the crop raised using Better Management Practices (BMPs), including trench-based technique, consumed less water. The analysis of data shows average water savings to the tune of 17.4% using the trench method of sowing, (with the range between 40% and 10%) as shown in graph – 1. The saving of water can be attributed to the larger spacing among cane rows in the trench method. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Graph 1: Comparison of irrigation water depths applied in control and demo plots. 0 20 40 60 80 100 1 2 3 4 5 6 Average 84 84 90 90 84 94 88 60 60 81 72 77 85 72 Water depth in cm Control plot vs Demo Plot CHART SHOWING COMPARISON OF WATER DEPTHS Control Plots Demo Plots Graph 1: Comparison of irrigation water depths applied in control and demo plots. 1 Roster – the mechanism of irrigation scheduling which defines the date and time of water distribution 3.2. Flows restored into the Karula river: 4.0 4.7 4.7 2.8 6.75.5 26.5 4.4 1.71.5 0.0 5.0 10.0 15.0 20.0 25.0 30.0 5月-19 7月-19 9月-19 11月-19 1月-20 3月-20 5月-20 7月-20 9月-20 11月-20 1月-21 3月-21 5月-21 Time series volume of saved water Discharge in Karula, Million litres 3 23 41 4 19 40 0 5 10 15 20 25 30 35 40 45 50 2019 2020 2021 A n n u a l V o l u m e t r e n d d i s c h a r g e d i n t o K a r u l a r i v e r Total running days Total Volume discharged, Million litres 3 23 41 4 19 40 0 5 0 5 0 5 0 5 0 5 0 2019 2020 2021 A n n u a l V o l u m e t r e n d d i s c h a r g e d i n t o K a r u l a r i v e r Total running days Total Volume discharged, Million litres Graph 2 A&B: Flows Dependability Curve of Observed Discharges in Karula River at critical point & Hydrological Variations in Karula river over 2017 to 2021. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Graph 2C: A snapshot of minimum flows in Karula, Khanpur Minor discharge and number of days when saved water was released into Karula river. 0 10 20 30 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec No. of days water discharged from canal to Karula Flow, cusec Flow in Karula at Karula - Khanpur canal tail confluence Minimum flows in Karula (2017-2021), cusec Average flow, Cusec Khanpur minor canal design discharge, cusec No. of days water discharged into karula from canal Graph 2C: A snapshot of minimum flows in Karula, Khanpur Minor discharge and number of days when saved water was released into Karula river. 0 10 20 30 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec No. of days water discharged from canal to Karula Flow, cusec Flow in Karula at Karula - Khanpur canal tail confluence Minimum flows in Karula (2017-2021), cusec Average flow, Cusec Khanpur minor canal design discharge, cusec No. 3.2. Flows restored into the Karula river: of days water discharged into karula from canal Graph 2C: A snapshot of minimum flows in Karula, Khanpur Minor discharge and number of days when saved water was released into Karula river. Graph 2D: Flows Dependability Curve of Observed Discharges in Karula River at Rahtauli village point & hydrological variations in Karula river from 2017 to 2021. 05 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Dsicharge, cusec Dependability, % Karula Dicharge at Rahtauli, Cusec Karula Dicharge at Rahtauli, Cusec 05 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Dsicharge, cusec Dependability, % Karula Dicharge at Rahtauli, Cusec Karula Dicharge at Rahtauli, Cusec Graph 2D: Flows Dependability Curve of Observed Discharges in Karula River at Rahtauli village point & hydrological variations in Karula river from 2017 to 2021. From Graph 2 D, which shows the flow duration curve, it can be inferred that, at 90% dependability (leanest flows), about 3 cusec water is available, whereas minimum average flows of 3.6 cusec are observed in the month of June in the Karula, near the tail end of the Khanpur canal. It is also evident here that the saved water from irrigation discharged into the river Karula accounts for 7% of minimum lean season flows. It can be seen that except during the monsoon months (June to October) saved water from irrigation is discharged into river Karula during all the lean season months. With further adoption of Better Man- agement Practices in the remaining sugarcane area in command and the scaling up of trench-based interventions, it is expected that more water will be contributed by the Khan- pur command to the Karula river. 3.3. Changes in sugarcane productivity: The data around sugarcane yield per unit area was discussed with the farmers. The figures around changes in yield (reported by the farmers) vary, depending upon the level/degree of adoption/adherence to Better Management Practices suggested, in addi- tion to the adoption of the trench-based practice by individual farmers. Therefore, there may be some variations in the outcome or productivity levels. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Graph 3: Comparison of productivity of sugarcane (Better Management Practices including trench vs traditional methods). 962.5 1061.3 957.5 1009.0 1012.5 1062.5 1010.9 750 900 800 750 850 850 816.7 0.0 200.0 400.0 600.0 800.0 1000.0 1200.0 Set 1 Set 2 Set 3 Set 4 Set 5 Set 6 Average Productivity, Q/ha Demo versus Control Plot Sets Productivity (Qtl/ha) of sugarcane (trench vs traditional) Productivity (qtl/ ha) in demo plot including secondary crop Productivity in control Plot Productivity (qtl/ ha) in demo plot including secondary crop Productivity in control Plot Graph 3: Comparison of productivity of sugarcane (Better Management Practices including trench vs traditional methods). Graph 3: Comparison of productivity of sugarcane (Better Management Practices including trench vs traditional methods). In this case, of the six farms sampled, the general average trend of agricultural productivity enhancement is about 23.8%, with the range between 34% and 19%; Graph 3 exhibits the degree of change in sugarcane productivity. 3.4. Economic implications for farmers and crop-water productivity: 3.4. Economic implications for farmers and crop-water productivity: Farmers have benefited in terms of earnings as well. The average income, in terms of unit area, is to the tune of Rs. 117,000/ha, whereas the range is Rs. 162,039/ha to Rs. 91,884/ha (Graph 4). Graph 4: Comparison of income per hectare (Better Management Practices including trench vs tra- ditional method). 269400 284598 256454 249628 290125 276115 271053 107361 192714 146520 130425 174100 173200 154053 0 50000 100000 150000 200000 250000 300000 350000 Set 1 Set 2 Set 3 Set 4 Set 5 Set 6 Average Income in Rs / ha Comparison of income per Ha from Trench and Traditionally Sown Sugarcane Demo Plot Control Plot Graph 4: Comparison of income per hectare (Better Management Practices including trench vs tra- ditional method). Income per unit of water consumed (irrigation applied) was enhanced by 117 % (on average) in farms using BMPs than a traditionally sown farm (Graph 5). A farmer, on average, gets additional income of Rs. 20.60 on every cubic metre of irrigation water used in the trench method. This is mainly due to a reduction in input costs (less fertiliser/pesti- cide, fuel etc.) and increase in yield and higher returns. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Graph 5: Comparison of income per cubic metre of water use (Better Management Practices includ- ing trench vs traditional methods). 45.21 47.43 31.73 34.73 34.13 36.06 38.22 12.79 22.96 16.28 14.56 20.69 18.47 17.62 0.00 10.00 20.00 30.00 40.00 50.00 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Average Income INR/cubic meter Income (in Rs) on per cubic metre of water use Income per cubic meter of water in Demo Plot Income per cubic meter of water in Control Plot Graph 5: Comparison of income per cubic metre of water use (Better Management Practices includ- ing trench vs traditional methods). The productivity per unit area may be attributed to the spacing between rows, which allows better aeration and provides space to grow freely, which results in cane plants of larger circumference and height, and weighing more, with greater sugar content. The per unit less water consumption may also be attributed to the heavier cane, providing greater yield of more value – with less water used. 3.4. Economic implications for farmers and crop-water productivity: Besides changes in sugarcane productivity and saving in irrigation water, the trench method offers opportunities to the farmers to grow a second crop in the sugarcane fields, simultaneously, between the ridges. Most of the farmers grow mustard or black-gram (urad) as an additional crop. These crops are not provided additional irrigation as their less water requirement is easily met with the soil moisture regime of the sugarcane crop. Farmers can use the additional crop for their consumption as well as to gain extra income from it. It has been calculated from demo farm data that the average income of multi- cropped sugarcane fields is around 20% higher (with the range between 15% and about 26%) than the single sugarcane crop sown with Better Management Practices, including the trench method, as shown in Graph 6. Graph 6: Percentage increase in income due to secondary crop with sugarcane. 24.7 15.9 20.7 15.3 25.9 19.9 20.4 0.0 5.0 10.0 15.0 20.0 25.0 30.0 1 2 3 4 5 6 Average Increase in income (%) Demo Plot Set Number % increase in income due to secondary crop with sugarcane Graph 6: Percentage increase in income due to secondary crop with sugarcane. The secondary crop, on an average, contributes to around 17% (with the range be- tween 13% and about 20%) of the total income of trench method sugarcane cultivation with multi-cropping. (Graph 7). Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Graph 7: Percentage Contribution of sugarcane and secondary crops in total income. 80.2 86.3 82.9 86.7 79.5 83.4 83.2 19.8 13.7 17.1 13.3 20.5 16.6 16.8 0.0 20.0 40.0 60.0 80.0 100.0 1 2 3 4 5 6 Average % income % Contribution of sugarcane and other crop in total income % contribution from sugarcane % contribution from other crop Graph 7: Percentage Contribution of sugarcane and secondary crops in total income. If the secondary crops had been sown alone, it would have consumed 15 cm irriga- tion water depth per hectare (assuming 50% area covered in sugarcane field is by the sec- ondary crops, which consumes 30 cm water for maturity). This is totally saved by the irrigation water provided to the sugarcane. The total water requirement of both crops, if each crop is sown alone, comes to 87 cm (72cm+15cm). Thus, the saving of 15 cm water of 87 cm is 17.2%. 3.4. Economic implications for farmers and crop-water productivity: The sugarcane crop raised using trencher tool already has a saving of 17.4% over traditional sowing, hence the multi-cropping scenario offers total water saving of 34.6% over the traditional raising of sugarcane crop. From the river conservation and water management perspective, the major outcome and impact of this initiative is the water savings from irrigation and release of that water into the river Karula through the passage. There are two set of calculations – total water savings at farm level in view of using Better Management Practices (BMPs), including trench-based sugarcane farming, and actual water discharge data (from the gauge near the riverbank on the passage). These calculations are shown in the following table (Table No. 3): Table 3. Current Gains due to Pilot Project Interventions. Water savings from farm – Unit Area (in cubic meters / hectare) Potential water saving if trench- based sugarcane adopted in all farms in Khanpur command (cu m) Water released into Karula river from passage (in cu m) [Observed Data] 1,570 246,490 (from about 157 ha) 62,550 Table 3. Current Gains due to Pilot Project Interventions. Water savings from farm – Unit Area (in cubic meters / hectare) Potential water saving if trench- based sugarcane adopted in all farms in Khanpur command (cu m) Water released into Karula river from passage (in cu m) [Observed Data] 1,570 246,490 (from about 157 ha) 62,550 Table 3. Current Gains due to Pilot Project Interventions. Water released into Karula river from passage (in cu m) [Observed Data] Water saved to the tune of 62,550 cubic metres (25% of potential water savings) has found its way into the Karula river, thereby enhancing its flows. There are substantial conveyance (seepage) losses and unaccounted withdrawals, which has significantly re- duced the overall volume of actual water released into Karula river. However, this means that there is an opportunity to bridge this inefficiency gap, so that the net gains can be enhanced. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Herein, there have been favourable changes in terms of water use requirement from the demand-side and an efficient irrigation canal system, which ensured reliable water supply to the farmers. This has led to the achievement of saving water meant for irrigation and its release into the Karula river, besides benefitting the farmers economically. The specific values of water released into the river Karula would remain a dynamic figure, as there are several associated and external factors that would influence this. Some of these key factors could be: a. The quantum of water flows in the Khanpur Minor canal, which may vary depending upon ✔ Availability of water in the main/parent canal ✔ Irrigation demand by farmers within the Khanpur command area ✔ Unauthorised withdrawals from the Khanpur Minor canal b. State of maintenance of Khanpur Minor c. Rainfall in the local catchment d. Maintenance of passage structure To sustain such an effort beyond the project duration is indeed a challenging ask, as there would be an apprehension that the situation would be back to ‘business-as-usual’ once the external support is withdrawn. To overcome this challenge, the formation of the Water Users Association (WUA) as per the provision of the Uttar Pradesh Participatory Irrigation Management Act 2009 was facilitated. In February 2021, the Khanpur Minor WUA was constituted and the elections for Executive/Governing Body (comprising of President, Secretary, Treasurer and other office bearers) of the WUA were unanimous. This is indicative of positivity amongst command farmers about the institutional support for this initiative, besides bringing them permanent solutions to the operation and mainte- nance of the Khanpur Minor canal system. Parallel to the efforts to form the WUA at the Khanpur Minor level, the capacity building of the farmers about roles, responsibilities and functions of WUA was done through training programmes, exposure visits to successful WUAs in the state and at the national level. This has helped in mobilising a ‘critical-mass’, who is now ready to take up the affairs of the WUA. However, the WUA is only recently established and further sup- port will be needed for it to become fully sustainable in financial and institutional terms. The Karula initiative was planned in such a way that the process for enhancement of flows in the Karula river fits within the current mechanism of irrigation scheduling and allocations and does not overwhelmingly change existing farm practices. 4. Discussion The Karula pilot was envisaged as a unique initiative, but under the backdrop of a well-debated idea – whether efficient irrigation water use can actually aid flows enhance- ment into the rivers and ultimately support the maintenance of e-flows in the rivers. On the other hand, there were externalities, which had the potential to disrupt the aspired outcomes of this initiative. However, a carefully developed stakeholder-led initiative has begun to deliver on the stated objectives, i.e., enhancing the flows in the river Karula. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 b. There has been another concern that farmers may tend to increase area un- der agriculture using water saved from the application of Better Manage- ment Practices (including trench use) in sugarcane farming. Nevertheless, the team still faced a situation where, since the Khanpur Minor canal did not feed all the farms in the middle to tail-end, saturation of the command area was bound to happen –once the demand-side and supply-side inter- ventions were applied in the command area, the saved water in the head to middle reaches of the canal would be used by the tail-enders. As this was well-understood since inception and there was no hurried and strict re- sponse from the team to ensure that the saved water fed immediately into the river, the team worked with the tail-end farmers and assured them that they could use the water from the canal as well as from the passage for irri- gation (by adopting trench-based technique), while letting the remaining water discharge into the river. The tail-end farmers agreed and this strategy worked well. c. The other consideration in the Karula pilot is the promotion of local and scalable ideas to manage the demand-side aspect and not really call for hi- tech, expensive means of pressure irrigation (drip and sprinkler), at least in the early phases of the project. The idea was not to introduce something totally new to the area, but to bring some of the improvisations that are rare but known amongst the progressive farmers in and around that district. However, at a later stage, a few farmers proposed the idea of demonstrat- ing pressure irrigation techniques and the team agreed to facilitate these. Various scientific studies have suggested that water from seepage through unlined canals recharges groundwater (Mirudhula K. 2014) and helps build shallow aquifers that are generally used as a source for irrigation. Infiltration from the canals recharges the aq- uifer directly and partially compensates for water uptake from plants and evaporation (Arumi J.L. et.al. 2009). The idea behind this project was to support conjunctive use and reduce overall water withdrawal (canal and groundwater for irrigation), combined with improved practices in irrigation and agriculture, which is likely to reduce the losses from evapotranspiration, a matter of further investigation. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 This would mean that the envisaged objective is likely to achieve partial success in terms of actually maintaining the e-flows for a river. Therefore, the initiative may not achieve the full suite of e-flows requirements (locations, timing and quantity of flows) for the Karula river, but it certainly aids to enhance the flows in the river in times of need, like the lean season of November to June. Some local factors that worked in favour of the Karula pilot were: a. Farmers in this area largely grow sugarcane (a water intensive crop) and the produce is insured by the Central and State government through Fair and Remunerative Price (FRP) and State Advised Price (SAP). Additionally, sugar mills that buy sugarcane are mandated to purchase crops from farmers within a specified radius known as the Cane Reservation Area at the FRP, which serves as defined market linkage for this cash crop (Niti Aayog). The team was fully aware of this fact – due to the availability of water and as- sured purchase of produce by the government through sugar mills, farmers would not switch to another water-intensive crop, which is a general appre- hension otherwise. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 rivers and wetlands. The saved water from irrigation, if conveyed to these freshwater re- sources, is likely to aid improvement of flows in the rivers. Arriving at such a stage is a critical milestone for maintaining e-flows in a river, because the most important question for e-flows maintenance is where the water for e-flows will come from, especially in over- allocated river basins. The irrigation water use efficiency initiative, as that of Khanpur Minor, could theoretically be upscaled at the extent of the Karula basin –about 65% (625 sq. km.) of catchment area of the Karula river grows sugarcane (as depicted in Appendix- 2). The extrapolations show that there is a potential of saving about 68 million cubic metres of water from about 70% of sugarcane farms within the Karula catchment. Whilst all the sugarcane farms in the Karula catchment may not be supported by surface-irrigation fa- cilities (that could have otherwise directly demonstrated enhancing flows in Karula); however, potentially lesser groundwater pumping in view of application of Better Man- agement Practices would certainly benefit the aquifer and river from these savings. This is likely to contribute to river discharges through enhanced base-flows. Moreover, there are about 30 minor irrigation canals in the adjoining areas of Khanpur Minor and these are all fed by the Ramganga Canal. If this initiative could be up-scaled in these irrigation sub-systems, then more water could be augmented into the Karula river. rivers and wetlands. The saved water from irrigation, if conveyed to these freshwater re- sources, is likely to aid improvement of flows in the rivers. Arriving at such a stage is a critical milestone for maintaining e-flows in a river, because the most important question for e-flows maintenance is where the water for e-flows will come from, especially in over- allocated river basins. The irrigation water use efficiency initiative, as that of Khanpur Minor, could theoretically be upscaled at the extent of the Karula basin –about 65% (625 sq. km.) of catchment area of the Karula river grows sugarcane (as depicted in Appendix- 2). The extrapolations show that there is a potential of saving about 68 million cubic metres of water from about 70% of sugarcane farms within the Karula catchment. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Whilst all the sugarcane farms in the Karula catchment may not be supported by surface-irrigation fa- cilities (that could have otherwise directly demonstrated enhancing flows in Karula); however, potentially lesser groundwater pumping in view of application of Better Man- agement Practices would certainly benefit the aquifer and river from these savings. This is likely to contribute to river discharges through enhanced base-flows. Moreover, there are about 30 minor irrigation canals in the adjoining areas of Khanpur Minor and these are all fed by the Ramganga Canal. If this initiative could be up-scaled in these irrigation sub-systems, then more water could be augmented into the Karula river. Whilst the apprehension may be valid that even if the water from irrigation is saved, it may lead to ‘enhancing-area-under-irrigation’ and/or push for ‘adoption-of-more-wa- ter-intensive-crops’, in certain circumstances, the Karula initiative has proved that a care- fully designed participative programme can actually bear desired results in terms of en- hanced flows. The Karula initiative demonstrates an alternative to promoting radical changes (suggesting newer cropping patterns or promoting pressure-irrigation in the early stages) in a short time span, without much rapport building with the stakeholders. It would be lot more prudent to look for local solutions (trench-based sugarcane farming, other package of practices including application of bio-pesticides and bio-fertilizers) and promote them in the project area. Once the benefits for the farmers are proven, they would come forward to support other forthcoming propositions as well. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 The groundwater serves the function of discharging base-flows into the river, espe- cially during lean season. It was observed that excess infiltration from the flood irrigation technique (earlier prevalent in the command area), though, may be recharging shallow aquifers to some extent, but would also be increasing the overall evapotranspiration (ET) losses. Post field interventions, the volume of canal water applied has reduced, which may affect infiltration, but will also reduce the overall groundwater abstractions, subsequently helping in stabilising groundwater levels in the long run and will continue to feed the river through base-flows. Following the interventions in the Khanpur Minor command area to reduce abstractions, increase efficiency, and connect the canal tail to the river, the water has a more direct route to the river which augments riverine flows in its leanest flows periods. However, there are larger river-groundwater interactions in play too, which impact the riverine baseflows. Precise and conclusive information in regard to the exact benefits to the river and to the catchment can be inferred through long-term hydro- logical and hydro-geological monitoring. Initiatives like the Karula river pilot can influence larger irrigation systems, as in a general scenario, the tail-ends of irrigation canals (in gravity-based systems) are close to Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Going forward, the team is now aspiring to upscale this initiative to about 16,000 hectare of Culturable Command Area (CCA) in the state of Uttar Pradesh, where the Ganga water resources feed the irrigation canals. This three-year programme will explore new leads, ideas, challenges, and opportunities, which would be worth narrating to the wider audience for their information, understanding and uptake. It is fully recognized that the rejuvenation of some of the world’s most populated and contested river systems continues to remain a challenging task, if the tributaries, rivulets, and wetlands in such river basins are not considered. It is in this context that the Karula pilot initiative is a pointer for policymakers and water-managers for the future. It is hoped that initiatives of this sort will help in curbing water-scarcity and will ensure wiser use of this precious resource. Moreover, the overall local ecology is set to benefit in this process as well! Author Contributions: NK conceived the idea of the paper; he also structured, drafted and finalized the paper. SB reviewed the paper and provided valuable inputs to the paper and its structure. AM and RB reviewed the paper and provided crucial information and data for the paper. PKS and RKA reviewed and provided detailed inputs about results for farm level information and water saving data analysis. DT and CL reviewed the paper and provided critical feedback and inputs to the paper which has improved the quality of paper. NK, AM, RB, SB, PKS, RKA were part of Karula Initiative implementation team on ground. Funding: This initiative was supported by HSBC Water Programme (2017-2022). Acknowledgments: The authors are grateful to Mr. Ravi Singh, Secretary General & Chief Executive Officer of WWF India for his constant motivation and support. The authors are also thankful to Dr. Sejal Worah, Programme Director of WWF India for her continuous encouragement and for being the source of inspiration to take up unique and new initiatives, like the one for the Karula river. WWF India’s work on e-flows has been possible because of active support and valuable contribu- tions from various partners, who have been part of this journey. The partners include several tech- nical, scientific and academic institutions, civil society organisations and individual experts from the country. These entities include the Indian Institute of Technology, Kanpur, Integrated Natural Resources Management Consultants, Indian Institute of Technology, Banaras Hindu University, Varanasi, H.N.B. 5. Conclusions As lessons learnt from the Karula initiative, the following takeaway points are made, which may not be conclusive for further replication of similar ideas, but are certainly key pointers for future considerations: a. Integrated approach: rather than merely looking at a single aspect, a holistic and comprehensive view works better. For instance, instead of simply work- ing on demand-side aspects, both supply-side aspects and institutional strengthening were also taken-up and this helped to achieve the objective. In addition, engagement with all key stakeholders, including the irrigation de- partment, district authorities, local agriculture science centres and farmers, was critical for a transformational change a. Equity and Ownership: a saturation of canal commanded area, in terms of access to irrigation water across the various ends of the canal (head-middle- tail) is a necessary and critical step in such exercises and therefore this should be acknowledged to get wholehearted support from the farmers across all reaches within the canal system. Such considerations also allow better buy- in and sense of ownership amongst the farmers in the entire canal command area b. Monitoring: the monitoring of the transformation is a critical aspect and if this is done in a joint fashion, it adds value not only for the initiative, but also better informs the stakeholders about the change that is in the offing c. Scalability: considering a unit for proof-of-concept that is scalable, is criti- cal, as the demonstration at an optimum unit has far better potential of up- scaling, and therefore mainstreaming Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Garhwal University,Srinagar, Central Inland Fisheries Research Institute,Allaha- bad and People’s Science Institute, Dehradun. Besides these, key government institutions like the National Mission for Clean Ganga and Central Water Commission,Government of India, the Irriga- tion and Water Resources Department of state of Uttar Pradesh, and concerned district administra- tions (Bijnor and Moradabad) also contributed to the work. The authors and the WWF India team are indebted to senior officials of the Uttar Pradesh Irrigation & Water Resources Department, in- cluding Mr. VK Rathi, Mr. AK Singh and Mr. RP Singh for being supportive of the idea of the Karula initiative and also for providing their valuable guidance and contextual knowledge to carry out the work. The authors also thank Mr. DP Singh and Mr. Naresh Kumar, the senior field officials of the department for extending all field-based support to the initiative. The local field team of WWF India, including Mr. Anar Singh Yadav and Mr. Deepak Kumar contributed to the Karula initiative in a big way and the authors thank them for their contributions. The authors also thank WWF India partners, Mr. Ravindra Kumar and Mr. DK Dudeja for their continuous guidance and support in carrying out the e-flows work. The authors thank scores of farmers from the command areas of Khanpur Minor and adjoining canals for supporting, adopting and demonstrating the irrigation water management approaches proposed by the team. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manu- script; or in the decision to publish the results. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Appendix – 1 Landuse Map of Khanpur Command Area Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Appendix – 2 Land-use & Land-cover Map of Karula River Basin Appendix – 2 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Appendix – 3 Illustration of combination of Supply-side Intervention and Demand-side Intervention leading to enhanc Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 Appendix – 4 Khanpur Minor command area map with location of control and demo farms Khanpur Minor command area map with location of control and demo farms Appendix A Karula River Pilot Appendix A Karula River Pilot Joint Farmer Surveys Objective a. to understand the agriculture and irrigation practices in both demonstr b. to ascertain the water-use at both categories of farms during watering a tum of water that is used c. to understand the agricultural productivity and its economic value, wh so that net economic gains can be assessed 1. Basic details 1.1 Date: 1.2 Name of Farmer: 1.3 Crop type: 1.4 Farm size: 1.5 Location on canal (H/M/T): 1.6 Outlet Head Number 2. Irrigation water application 2.1 Name of crop: 2.2 Method of Irrigation (flooding, basin, furrow etc.): 2.3 Source of Irrigation (canal, tube well, well etc.): 2.4 Total time of irrigation (calculated from irrigation time per water 2.5 Total water depth applied: 3. Input details and costing 3.1 Expense on seeds: 3.2 Expense on labour (harrowing, ploughing, harvesting): 3.3 Expense on compost: 3.4 Expense on Fertilizers: 3.5 Expense on Weedicides/pesticides: 4. Productivity and economic value 4.1 Sugarcane productivity per unit area: 4.2 Other crop productivity per unit area: 4.3 Market rate per quintal of sugarcane: 4.4 Market rate per quintal of other crop: Objective a. to understand the agriculture and irrigation practices in both demonstration farms and control farms b. to ascertain the water-use at both categories of farms during watering and understand the variation in quan- tum of water that is used c. to understand the agricultural productivity and its economic value, while calculating the entire input costing; so that net economic gains can be assessed c. to understand the agricultural productivity and its economic value, while calculating the entire input costing; so that net economic gains can be assessed 1. Basic details 1.1 Date: 1.2 Name of Farmer: 1.3 Crop type: 1.4 Farm size: 1.5 Location on canal (H/M/T): 1.6 Outlet Head Number 2. Irrigation water application 2. Irrigation water application 2.1 Name of crop: 2.2 Method of Irrigation (flooding, basin, furrow etc.): 2.3 Source of Irrigation (canal, tube well, well etc.): 2.4 Total time of irrigation (calculated from irrigation time per watering and number of waterings per crop): 2.5 Total water depth applied: 3. Input details and costing 3.1 Expense on seeds: 3.2 Expense on labour (harrowing, ploughing, harvesting): 3.3 Expense on compost: 3.4 Expense on Fertilizers: 3.5 Expense on Weedicides/pesticides: 4. Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 References Water and Agriculture in India: Background paper for the South Asia expert panel during the Global Forum for Food and Agriculture 2017. OAV – German Asia-Pacific Business Association and The Energy and Resources Institute. The paper can be accessed at 7. Dhawan Vibha (2017). Water and Agriculture in India: Background paper for the South Asia expert panel during the Global Forum for Food and Agriculture 2017. OAV – German Asia-Pacific Business Association and The Energy and Resources Institute. The paper can be accessed at p p https://www.oav.de/fileadmin/user_upload/5_Publikationen/5_Studien/170118_Study_Water_Agriculture_ 8. Ganga River Basin Management Plan – Main Document (2015). Consortia of IITs and partners. The document can be accessed at: g g ( ) p at: https://nmcg.nic.in/writereaddata/fileupload/13_GRBMP%20-%20MPD.pdf at: https://nmcg.nic.in/writereaddata/fileupload/13_GRBMP%20-%20MPD.pdf 9. Giordano, M.; Turral, H.; Scheierling, S. M.; Tréguer, D. O.; McCornick, P. G. (2017). Beyond “More Crop per Drop”: evolving thinking on agricultural water productivity. Colombo, Sri Lanka: International Water Management Institute (IWMI); Washington, DC, USA: The World Bank. 53p. (IWMI Research Report 169). doi: 10.5337/2017.202. License: Creative Commons Attribution CC BY 3.0 IGO. The paper can be accessed at: http://documents1.worldbank.org/curated/ru/716711491456318468/pdf/114060-WP-P128451-PUBLIC-IWMIResearchReport.pdf 9. Giordano, M.; Turral, H.; Scheierling, S. M.; Tréguer, D. O.; McCornick, P. G. (2017). Beyond “More Crop per Drop”: evolving thinking on agricultural water productivity. Colombo, Sri Lanka: International Water Management Institute (IWMI); Washington, DC, USA: The World Bank. 53p. (IWMI Research Report 169). doi: 10.5337/2017.202. License: Creative Commons Attribution CC BY 3.0 IGO. The paper can be accessed at: http://documents1.worldbank.org/curated/ru/716711491456318468/pdf/114060-WP-P128451-PUBLIC-IWMIResearchReport.pdf 10. Gopal Brij (2013) Environmental Flows – an introduction for Water Resources Managers. National Institute of Ecology, New Delhi India and CapNet, Pretoria South Africa. The publication can be accessed at: 10. Gopal Brij (2013) Environmental Flows – an introduction for Water Resources Managers. National Institute of Ecology, New Delhi India and CapNet, Pretoria South Africa. The publication can be accessed at: http://aquaticecosystems.org/library/environmental-flows/ http://aquaticecosystems.org/library/environmental-flows/ 11. Grill, G., Lehner, B., Lumsdon, A. E., MacDonald, G. K., Zarfl, C., and Reidy Liermann, C. (2015). An index-based framework for assessing patterns and trends in river fragmentation and flow regulation by global dams at multiple scales. Environ. Res. Lett. 10:015001. doi: 10.1088/1748-9326/10/1/015001. The paper can be accessed at: 11. Grill, G., Lehner, B., Lumsdon, A. E., MacDonald, G. K., Zarfl, C., and Reidy Liermann, C. (2015). An index-based framework for assessing patterns and trends in river fragmentation and flow regulation by global dams at multiple scales. Environ. Res. Lett. 10:015001. doi: 10.1088/1748-9326/10/1/015001. Appendix A Karula River Pilot Productivity and economic value 4.1 Sugarcane productivity per unit area: 4.2 Other crop productivity per unit area: 4.3 Market rate per quintal of sugarcane: 4.4 Market rate per quintal of other crop: References 1. Arthington AH, Bhaduri A, Bunn SE, Jackson SE, Tharme RE, Tickner D, Young B, Acreman M, Baker N, Capon S, Horne AC, Kendy E, McClain ME, Poff NL, Richter BD and Ward S (2018) The Brisbane Declaration and Global Action Agenda on Environmental Flows (2018). The paper can be accessed at: ( ) p p https://www.frontiersin.org/articles/10.3389/fenvs.2018.00045/full 2. Arumi Luis Jose, Rivera D., Holzapfel E., Boochs P., Billib M., Fernald A. (2009). Effect of the irrigation canal network on surface and groundwater interactions in the lower valley of the Cachapoal River, Chile. Chilean Journal of Agricultural Research 69 (1): 12-20 (January-March 2009). The paper can be accessed at: http://www.bioline.org.br/pdf?cj09002 2. Arumi Luis Jose, Rivera D., Holzapfel E., Boochs P., Billib M., Fernald A. (2009). Effect of the irrigation canal network on surface and groundwater interactions in the lower valley of the Cachapoal River, Chile. Chilean Journal of Agricultural Research 69 (1): 12-20 (January-March 2009). The paper can be accessed at: http://www.bioline.org.br/pdf?cj09002 3. Barton David N. and Teron Avinandan (2010). Valuing irrigation water using survey-based methods in the Tungabhadra River basin, India. NINR & IWMI. December 2010 Irrigation and Drainage Systems 24(3):265-277. The paper can be accessed at https://www.researchgate.net/publication/225355850_Valuing_irrigation_water_using_survey- 3. Barton David N. and Teron Avinandan (2010). Valuing irrigation water using survey-based methods in the Tungabhadra River basin, India. NINR & IWMI. December 2010 Irrigation and Drainage Systems 24(3):265-277. The paper can be accessed at 4. Beegle Kathleen, Carletto Calogero, Himelein Kristen (2011). Reliability of Recall in Agricultural Data. Policy Research Working Paper. Poverty and Inequality Team, Development Research Group, The World Bank. The document can be accessed at https://elibrary.worldbank.org/doi/pdf/10.1596/1813-9450-5671 5. Central Water Commission (2020). Abstract on Water Sector 2020. Water Related Statistics Directorate, Information System Organisation, Water Planning and Projects Wing, Central Water Commission, Ministry of Jal Shakti, Government of India. The document can be accessed at http://cwc.gov.in/sites/default/files/abstract-water-sector-2020.pdf 5. Central Water Commission (2020). Abstract on Water Sector 2020. Water Related Statistics Directorate, Information System Organisation, Water Planning and Projects Wing, Central Water Commission, Ministry of Jal Shakti, Government of India. The document can be accessed at http://cwc gov in/sites/default/files/abstract-water-sector-2020 pdf 6. Department of Sugarcane, Government of Uttar Pradesh (2018). Success stories of Sugarcane development in District- Bijnor, Uttar Pradesh. The document can be accessed at 2018035156Sugarcane department success story.pdf (rkvy.nic.in) g p y p ( y ) 7. Dhawan Vibha (2017). IJERTV3IS091115.pdf p 22. National Mission Clean Ganga, Government of India – Gazette Order (2018). Environmental Flows Order. The document can be accessed at: https://nmcg nic in/writereaddata/fileupload/28 190717 pdf https://nmcg.nic.in/writereaddata/fileupload/28_190717.pdf https://nmcg.nic.in/writereaddata/fileupload/28_190717.pdf 23. National Mission Clean Ganga Authority Notification, Government of India – Gazette Notification (2016). The document can be accessed at: https://nmcg.nic.in/writereaddata/fileupload/47_AuthorityNotification.pdf 24. Central Dam Safety Organisation, Central Water Commission (2019). National Register of Large Dams. Dam Safety Monitoring Directorate, CWC, Ministry of Jal Shakti, Government of India. The document can be accessed at ttp://cwc.gov.in/sites/default/files/nrld06042019.pdf 25. Smith Martin – Land & Water Management Division of the Food and Agriculture Organization (1996). CROPWAT – A Computer Programme for Irrigation Planning & Management. Irrigation & Drainage Paper 46. The document can be accessed at: https://books.google.it/books?id=p9tB2ht47NAC&pg=PP1&source=kp_read_button&redir_esc=y#v=onepage&q&f=false https://books.google.it/books?id=p9tB2ht47NAC&pg=PP1&source=kp_read_button&redir_esc=y#v=one 26. Stockle O Claudio (2002). Environmental Impact of irrigation: A Review. State of Washington Water Research Center – Washington State University PO Box 643002 Pullman Washington. The paper can be accessed at g y g p p http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.488.4861&rep=rep1&type=pdf 27. Tickner, D., Kaushal, N., Speed, R. A., Tharme, R. E., eds. (2020). Implementing Environmental Flows: Lessons for Policy and Practice. Lausanne: Frontiers Media SA. doi: 10.3389/978-2-88966-039-1. The document can be accessed at https://www.frontiersin.org/articles/10.3389/fenvs.2020.00106/full 28. OECD Water Governance Initiative (2014). Scoping Note: Stakeholders Engagement for effective Water Governance. The document can be accessed at: 28. OECD Water Governance Initiative (2014). Scoping Note: Stakeholders Engagement for effective Water Governance. The document can be accessed at: https://www.riob.org/sites/default/files/IMG/pdf/OECD_WGI_WG1_Stakeholder_Engagement___scoping_note_FINAL.pdf https://www.riob.org/sites/default/files/IMG/pdf/OECD_WGI_WG1_Stakeholder_Engagement___scoping_note_FINAL.pdf 29. O’Keeffe J, Kaushal N, Bharati L, Smakhtin V (2012). Assessment of the Environmental Flows for the Upper Ganga Basin. World https://www.riob.org/sites/default/files/IMG/pdf/OECD_WGI_WG1_Stakeholder_Engagement___scoping_note_FINAL.pdf 29. O’Keeffe J, Kaushal N, Bharati L, Smakhtin V (2012). Assessment of the Environmental Flows for the Upper Ganga Basin. World Wide Fund for Nature – India. The report can be accessed at: h //d2391 l 4h h l df /d l d / f fl df 29. O’Keeffe J, Kaushal N, Bharati L, Smakhtin V (2012). Assessment of the Environmental Flows for the Upper Ganga Basin. World Wide Fund for Nature – India. The report can be accessed at: https://d2391rlyg4hwoh.cloudfront.net/downloads/wwf_e_flows_report.pdf 30. Qi J, ST Brantley, SW Golladay (2020). Simulated irrigation reduction improves low flow in streams – A case study in the Lower Flint River Basin. Journal of Hydrology: Regional Studies (Vol. 28). The paper can be accessed at: https://www sciencedirect com/science/article/pii/S2214581819301156 30. Qi J, ST Brantley, SW Golladay (2020). Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 16. Ja¨germeyr, J., Pastor A., Biemans H., Gerten D. (2017) Reconciling irrigated food production with environmental flows for Sustainable Development Goals implementation. Nat. Commun. 8, 15900 doi: 10.1038/ncomms15900 (2017). The paper can be accessed at: 16. Ja¨germeyr, J., Pastor A., Biemans H., Gerten D. (2017) Reconciling irrigated food production with environmental flows for Sustainable Development Goals implementation. Nat. Commun. 8, 15900 doi: 10.1038/ncomms15900 (2017). The paper can be accessed at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524928/pdf/ncomms15900.pdf https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524928/pdf/ncomms15900.pdf https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5524928/pdf/ncomms15900.pdf 17. Kaushal N, Mishra A, Babu S, O’Keeffe J (2018). Environmental Flows for a healthy Ramganga. Wo India. The document can be accessed at: https://www.wwfindia.org/news_facts/wwf_publications/?19842/Environmental-Flows-for-a-Healthy-Ramganga https://www.wwfindia.org/news_facts/wwf_publications/?19842/Environmental-Flows-for-a-Healthy-Ramganga 18. Kaushal N, Babu S, Mishra A, Ghosh N, Tare V, Kumar R, Sinha PK and Verma RU (2019) Towards a Healthy Ganga— Improving River Flows Through Understanding Trade Offs. Frontiers Environ. Sci. 7:83. The paper can be accessed at https://www.frontiersin.org/articles/10.3389/fenvs.2019.00083/full p g 19. Megdal BS, Eden S, Shamir E (2017). Water Governance, Stakeholder Engagement, and Sustainable Water Resources Management. MDPI Water 2017, 9, 190; doi:10.3390/w9030190. 20. Ministry of Jal Shakti, Government of India – Press Information (2019). Inadequacy of Ecological Flows Notification. The Press Note can be accessed at: h // b / l I 20. Ministry of Jal Shakti, Government of India – Press Information (2019). Inadequacy of Ecological Flows Notification. The Press Note can be accessed at: https://pib.gov.in/PressReleasePage.aspx?PRID=1595554 21. Mirudhula K. (2014). Impact of Lined/Unlined Canal on Groundwater Recharge in the Lower Bhavani Basin. International Journal of Engineering Research & Technology (IJERT) ISSN: 2278-0181. Vol. 3 Issue 9, September- 2014. The paper can be accessed at: https://www.ijert.org/research/impact-of-linedunlined-canal-on-groundwater-recharge-in-the-lower-bhavani-basin- IJERTV3IS091115.pdf https://www.ijert.org/research/impact-of-linedunlined-canal-on-groundwater-recharge-in-the-lower-bh IJERTV3IS091115.pdf References The paper can be accessed at: p p https://iopscience.iop.org/article/10.1088/1748-9326/10/1/015001 p p https://iopscience.iop.org/article/10.1088/1748-9326/10/1/015001 p p https://iopscience.iop.org/article/10.1088/1748-9326/10/1/015001 p p https://iopscience.iop.org/article/10.1088/1748-9326/10/1/015001 12. Harvey, N. (2014). The efficiency dilemma. Denver, Colorado, USA: Colorado Foundation for Water Education (CFWE). The article can be accessed at: 12. Harvey, N. (2014). The efficiency dilemma. Denver, Colorado, USA: Colorado Foundation for Water Education (CFWE). The article can be accessed at: https://www.watereducationcolorado.org/publications-and-radio/headwaters-magazine/fall-2014-eastern-p dilemma/ 13. Hijri R and Davis R (2009) Environmental Flows in Water Resources Policies, Plans, and Projects Case Studies. Natural Resources Management Series, The World Bank Environment Department. The paper can be accessed at: g p p p https://openknowledge.worldbank.org/bitstream/handle/10986/18381/488710NWP0Box31C10EDP01170Env0flows.pdf?sequence=1 &isAllowed=y 14. I.C. Overton, D.M. Smith, J. Dalton, S. Barchiesi, M.C. Acreman, J.C. Stromberg & J.M. Kirby (2014) Implementing environmental flows in integrated water resources management and the ecosystem approach, Hydrological Sciences Journal, 59:3-4, 860-877, DOI: 10.1080/02626667.2014.897408 https://www.tandfonline.com/doi/full/10.1080/02626667.2014.897408 tandfonline.com/doi/full/10.1080/02626667.2014.89740 https://www.tandfonline.com/doi/full/10.1080/02626667.2014.897408 15. Ja¨germeyr, Gerten D., Heinke J., Schaphoff S., Kummu M., Lucht W. (2015). Water savings potentials of irrigation systems: global simulation of processes and linkages. Hydrology and Earth System Sciences, 19, 3073–3091, 2015. This paper can be accessed at: https://www.hydrol-earth-syst-sci.net/19/3073/2015/hess-19-3073-2015.pdf https://www.hydrol-earth-syst-sci.net/19/3073/2015/hess-19-3073-2015.pdf https://www.hydrol-earth-syst-sci.net/19/3073/2015/hess-19-3073-2015.pdf https://www.hydrol-earth-syst-sci.net/19/3073/2015/hess-19-3073-2015.pdf https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2584147/ p g p 35. Wollburg Philip, Tiberti Marco, Zezza Alberto (2020). Recall length and measurement error in agricultural surveys. Develop- ment Data Group, The World Bank. Food Policy. Published by Elsevier Ltd. The paper can be accessed at p g p 35. Wollburg Philip, Tiberti Marco, Zezza Alberto (2020). Recall length and measurement error in agricultural surveys. Develop- ment Data Group, The World Bank. Food Policy. Published by Elsevier Ltd. The paper can be accessed at https://reader.elsevier.com/reader/sd/pii/S0306919220302098?to- ken=CFFCE80FF12E44EC3240539B55C94128D81FA2CD2F5BEE70DE4E20929B71B50AA2E4751DE58A1DAC204FA758203584 F0&originRegion=eu-west-1&originCreation=20220421063541 35. Wollburg Philip, Tiberti Marco, Zezza Alberto (2020). Recall length and measurement erro ment Data Group, The World Bank. Food Policy. Published by Elsevier Ltd. The paper can ment Data Group, The World Bank. Food Policy. Published by Elsevier Ltd. The paper can be accessed at https://reader.elsevier.com/reader/sd/pii/S0306919220302098?to- ken=CFFCE80FF12E44EC3240539B55C94128D81FA2CD2F5BEE70DE4E20929B71B50AA2E4751DE58A1DAC204FA758203584 F0&originRegion=eu-west-1&originCreation=20220421063541 IJERTV3IS091115.pdf Simulated irrigation reduction improves low flow in streams – A case study in the Lower Flint River Basin. Journal of Hydrology: Regional Studies (Vol. 28). The paper can be accessed at: 30. Qi J, ST Brantley, SW Golladay (2020). Simulated irrigation reduction improves low flow in streams – A case study in the Lower Flint River Basin. Journal of Hydrology: Regional Studies (Vol. 28). The paper can be accessed at: 31. Stockle Claudio O. (2007) Environmental Impacts of irrigation – an overview. State of Washington Water Resources Centre, Washington State University. The paper can be accessed at: http //citeseerx ist psu edu/ iewdoc/download?doi 10 1 1 488 4861&rep rep1&type pdf g y p p http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.488.4861&rep=rep1&type=pdf 32. Tare V, Sanand SG, Agarwal IC, Chopra R, Gautam A, Kaushal N, Babu S, Behera S, Joshi KD, Tyagi P, Kumar R, Mathur RP, Sinha R, Tripathi S, Jain A, Srivastava R, Ohri A (2013). World Wide Fund for Nature – India. The report can be accessed at: https://d2391rlyg4hwoh.cloudfront.net/downloads/environmental_flows_for_kumbh_2013_at_triveni_sangam__allahabad.pdf p yg g p 33. Uta Wehn, Kevin Collins, Kim Anema, Laura Basco-Carrera & Alix Lerebours (2018) Stakeholder engagement in water governance as social learning: lessons from practice, Water International, 43:1, 34-59, DOI: 10.1080/02508060.2018.1403083 https://doi.org/10.1080/02508060.2018.1403083 p g 34. Ward A. Frank and Velazquez MP (2008). Water conservation in irrigation can increase water use. Proceedings of National Academy of Science USA. 2008 Nov 25; 105(47): 18215–18220. This paper can be accessed at: p g 34. Ward A. Frank and Velazquez MP (2008). Water conservation in irrigation can increase water use. Proceedings of National Academy of Science USA. 2008 Nov 25; 105(47): 18215–18220. This paper can be accessed at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2584147/ 35. Wollburg Philip, Tiberti Marco, Zezza Alberto (2020). Recall length and measurement error in agricultural surveys. Develop- ment Data Group, The World Bank. Food Policy. Published by Elsevier Ltd. The paper can be accessed at https://reader.elsevier.com/reader/sd/pii/S0306919220302098?to- ken=CFFCE80FF12E44EC3240539B55C94128D81FA2CD2F5BEE70DE4E20929B71B50AA2E4751DE58A1DAC204FA758203584 F0&originRegion=eu-west-1&originCreation=20220421063541 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 https://reader.elsevier.com/reader/sd/pii/S0306919220302098?to- ken=CFFCE80FF12E44EC3240539B55C94128D81FA2CD2F5BEE70DE4E20929B71B50AA2E4751DE F0&originRegion=eu-west-1&originCreation=20220421063541 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2584147/ 35. Wollburg Philip, Tiberti Marco, Zezza Alberto (2020). Recall length and measurement error in agricultural surveys. Develop- ment Data Group, The World Bank. Food Policy. Published by Elsevier Ltd. The paper can be accessed at https://reader.elsevier.com/reader/sd/pii/S0306919220302098?to- ken=CFFCE80FF12E44EC3240539B55C94128D81FA2CD2F5BEE70DE4E20929B71B50AA2E4751DE58A1DAC204FA758203584 F0&originRegion=eu-west-1&originCreation=20220421063541 Preprints (www.preprints.org) \| NOT PEER-REVIEWED \| Posted: 29 June 2022 doi:10.20944/preprints202206.0400.v1 doi:10.20944/preprints202206.0400.v1
https://openalex.org/W2098122966	https://biblio.ugent.be/publication/5863968/file/5863990.pdf	English	null	Arguments for and against self and non-self root recognition in plants	Frontiers in plant science	2,014	cc-by	6,816	INTRODUCTION the inaccessibility of the belowground root system, recent efforts addressed this still open question. Nevertheless, the exploration of molecular mechanisms of root identity recognition is lim- ited. With next generation sequencing methods becoming more available in research practices, it seems only timely to address this question by using such state of the art techniques, for which proteomics and metabolomics approaches could also prove useful. Competition among coexisting plants—most restrictively deﬁned as a negative interaction among individuals with reduced growth, survival, or fecundity of neighbors as a consequence (Casper and Jackson, 1997)—is all about the availability of space, nutri- ents, water, and light. This contest is thought to be, at least in part, responsible for the plant diversity in different ecosystems (Goldberg and Barton, 1992; Wilson and Tilman, 1993). More- over, it is, if anything, a showcase for the remarkable adaptive plasticity of plants, i.e., their ability to alter their morphology and physiology in response to environmental stimuli (Bradshaw, 1965; reviewed in Hodge, 2009; Ford, 2014). Here, knowledge on root–root dynamics between interacting plants will be summarized and new advances will be discussed that cannot only enhance the understanding of plant evolu- tion and biology, but can also have an impact on ecology and agriculture. g Roots are pivotal for plant survival because they ensure the uptake of nutrients and water and they secure ﬁxation in the soil; hence, the growing interest in the study of belowground plant competition. Plants that grow together in one soil volume depend on the same resources and rearrange their root systems to gain access to these limited supplies (Robinson, 1994). Indeed, root systems develop differently when neighboring roots are present and their growth responses vary. These responses are determined by species, relatedness, even genotype, and by self or non-self identity of the competing roots (reviewed in Schenk et al., 1999; Chen et al., 2012). The latter indicate interactions among roots of the same individual plant (“self”) or of different plants (“non- self”), whereby plants also seem to be able to recognize kin (Dudley and File, 2007). Thus, roots possess a so-called “iden- tity recognition.” However, how do roots recognize other roots? Although root–root interaction studies are extremely complex due to the many factors that inﬂuence root competition and MINI REVIEW ARTICLE MINI REVIEW ARTICLE Correspondence: Keywords: root–root interaction, root competition, identity recognition, self/non-self recognition, root growth stephen.depuydt@ghent.ac.kr Stephen Depuydt 1,2,3 1 Ghent University Global Campus, Incheon, South Korea 2 Department of Plant Biotechnology and Bioinformatics, Ghent University, Ghent, Belgium 3 Department of Plant Systems Biology, Flanders Institute for Biotechnology, Ghent, Belgium Root–root interaction research gained more and more attention over the past few years. Roots are pivotal for plant survival because they ensure uptake of water and nutrients. Therefore, detection of adjacent roots might lead to competitive advantages. Several lines of experimental evidence suggest that roots have ways to discriminate non-related roots, kin, and—importantly—that they can sense self/non-self roots to avoid intra-plant competition. In this mini-review, the existence of self/non-self recognition in plant roots will be discussed and the current knowledge on the mechanisms that could be involved will be summarized. Although the process of identity recognition is still not completely understood, interesting data are available and emerging new technologies will certainly aid to better understand this research ﬁeld that can have an important biological, ecological, and agricultural impact. Edited by: Edited by: Boris Rewald, University of Natural Resources and Life Sciences, Vienna, Austria Boris Rewald, University of Natural Resources and Life Sciences, Vienna, Austria Reviewed by: Catharina Meinen, Georg-August-Universität Göttingen, Germany Harsh Bais, University of Delaware, USA Correspondence: Stephen Depuydt, Ghent University Global Campus, 119 Songdomunhwa-Ro, Yeonsu-Gu, Incheon 406840, South Korea e-mail: h d d @ h k Correspondence: Stephen Depuydt, Ghent University Global Campus, 119 Songdomunhwa-Ro, Yeonsu-Gu, Incheon 406840, South Korea e-mail: ROOT–ROOT INTERACTIONS: HOW THE ROOT SYSTEM RESPONDS TO NEIGHBORING ROOTS A lot of experimental evidence suggests that plants alter their root growth in the presence of other plants (for a review, see Schenk et al., 1999). Pioneering work on root interaction focused mainly on spatial segregation, such as intraspeciﬁcally in Parthenium argentatum (guayule; Muller, 1946) or Prunus persica (peach) trees (Bini and Chisci, 1961), or interspeciﬁcally, such as Juglans nigra (black walnut) roots that exclude Solanum lycopersicum (tomato) roots (Massey, 1925). Roots can also be attracted to other roots; for instance, Fragaria vesca (wild strawberry) roots are drawn to Glechoma hederacea (ground ivy) roots, whereas the ivy roots avoid the strawberry roots (de Kroon, 2007). In addition, root elongation responses also occur: for instance, elongation of Fragaria chiloensis (beach strawberry) roots is stimulated upon November 2014 \| Volume 5 \| Article 614 \| 1 www.frontiersin.org www.frontiersin.org Identity recognition in plants Depuydt contact with ground ivy (Semchenko et al., 2007b). Analysis of the overall root biomass of natural grassland systems revealed overyielding, no effect, or even underyielding when mixtures are compared with monocultures (Faget et al., 2013). In crops as well, effects on root growth by neighboring roots are clear, not only intraspeciﬁcally, such as for Glycine max (soybean) and Allium cepa (onion; Raper and Barber, 1970; Baldwin and Tinker, 1972), but also when intercropped. For instance, when certain Zea mays (maize) and soybean species are grown together, the roots of each plant tend to keep away from each other and become shallower than those in systems intercropped with their kin. Remarkably, not every maize variety responds in the same manner to the presence of the same soybean species (Fang et al., 2011). Similarly, roots of a Beta vulgaris (beet) variety grow faster and deeper than legume roots grown in the same soil, providing a competitive advantage (Tosti and Thorup-Kristensen, 2010). Furthermore, roots can accumulate in the top soil, such as in mixed grassland species (Mamolos et al., 1995). The root density in the top soil of Acacia saligna (orange wattle) trees intercropped with Sorghum bicolor (sorghum) is also higher than that of monocultures (Lehmann et al., 1998). Intriguingly, root allocations might be inﬂuenced by kin recognition, i.e., the ability to discriminate siblings from strangers. In Cakile edentula (sea rocket) and Impatiens pallida (pale touch-me-not), root alloca- tion is larger and smaller in groups of strangers than of siblings, respectively (Dudley and File, 2007; Murphy and Dudley, 2009). ROOT–ROOT INTERACTIONS: HOW THE ROOT SYSTEM RESPONDS TO NEIGHBORING ROOTS In conclusion, responses (Figure 1A) vary in several experi- ments, indicating that roots sense the presence of other roots and that identity recognition might be important in altering root growth. contact with ground ivy (Semchenko et al., 2007b). Analysis of the overall root biomass of natural grassland systems revealed overyielding, no effect, or even underyielding when mixtures are compared with monocultures (Faget et al., 2013). In crops as well, effects on root growth by neighboring roots are clear, not only intraspeciﬁcally, such as for Glycine max (soybean) and Allium cepa (onion; Raper and Barber, 1970; Baldwin and Tinker, 1972), but also when intercropped. For instance, when certain Zea mays (maize) and soybean species are grown together, the roots of each plant tend to keep away from each other and become shallower than those in systems intercropped with their kin. Remarkably, not every maize variety responds in the same manner to the presence of the same soybean species (Fang et al., 2011). Similarly, roots of a Beta vulgaris (beet) variety grow faster and deeper than legume roots grown in the same soil, providing a competitive advantage (Tosti and Thorup-Kristensen, 2010). Furthermore, roots can accumulate in the top soil, such as in mixed grassland species (Mamolos et al., 1995). The root density in the top soil of Acacia saligna (orange wattle) trees intercropped with Sorghum bicolor (sorghum) is also higher than that of monocultures (Lehmann et al., 1998). Intriguingly, root allocations might be inﬂuenced by kin recognition, i.e., the ability to discriminate siblings from strangers. In Cakile edentula (sea rocket) and Impatiens pallida (pale touch-me-not), root alloca- tion is larger and smaller in groups of strangers than of siblings, IDENTITY RECOGNITION AND ITS IMPORTANCE In plants, identity recognition has been unequivocally demon- strated for several biological processes that affect plant ﬁtness, reproduction, and/or survival. For instance approximately 60% of the angiosperms show self-incompatibility which ultimately serves to prevent self-fertilization (for reviews, see Kitashiba and Nasrallah, 2014; Sawada et al., 2014). Moreover, host recognition systems of parasitic plants (Cardoso et al., 2011) and recognition of potential pathogens basically relies on the ability to discrimi- nate “self” and “non-self” (Sanabria et al., 2008). As shown above, identity recognition is also of great impor- tance for the outcome of belowground interactions. Self/non- self recognition had ﬁrst been reported for the desert shrub Ambrosia dumosa (burro-weed). Roots of Ambrosia stop growing when root systems from other Ambrosia plants (i.e., the same plant population) are encountered, seemingly as an avoidance response, but not when roots from the same physiological indi- vidual (i.e., self roots) are sensed (Mahall and Callaway, 1991, 1992). In contrast, roots of Larrea tridentata (creosote bush), also FIGURE 1 \| Common root responses to neighboring plants and possible mechanisms of root identity recognition. (A) Range of root responses toward neighboring roots. Two plants are depicted (plant 1 and 2) of which plant 2 shows a differential root response because it is neighbored with plant 1. Plants 1 and 2 can be either of the same species or of different species. Examples (both intra- and interspeciﬁcally) of interacting species, are provided for each response: overproliferation of the root system, i.e., root biomass changes which may include main root length increases, more adventitious rooting, more and higher order lateral roots, etc.; in accumulation in the top soil; spatial segregation; ro proposed mechanisms in self/non-self root recogni of electrical or hormonal signals (Schenk et al., 199 associated microorganisms and/or their secreted s and Vanderleyden, 2000); and soluble metabolites i et al., 2010; Caffaro et al., 2011; Fang et al., 2013; S non-self root–root interaction is shown. rooting, more and higher order lateral roots, etc.; increased lateral rooting; accumulation in the top soil; spatial segregation; root attraction. (B) Four proposed mechanisms in self/non-self root recognition: resonant ampliﬁcation of electrical or hormonal signals (Schenk et al., 1999; Falik et al., 2003); root associated microorganisms and/or their secreted substances (Steenhoudt and Vanderleyden, 2000); and soluble metabolites in root exudates (Biedrzycki et al., 2010; Caffaro et al., 2011; Fang et al., 2013; Semchenko et al., 2014). A non-self root–root interaction is shown. IDENTITY RECOGNITION AND ITS IMPORTANCE Nonetheless, a clear conclusion could be drawn, namely that to be recognized as self roots, they must be physiologically attached. Detached roots, even when they originate from the same and, thus, genetically identical individual, are recognized as non-self (Mahall and Callaway, 1991; Gruntman and Novoplansky, 2004; Falik et al., 2006; Nord et al., 2011). In contrast, kin recog- nition or recognition of the same species/population, but not of the same individual would occur via different mechanisms. In Arabidopsis, photosensory receptors distinguish between light signals from kin and other neighbors and allow leaf reposi- tioning to decrease light competition (Crepy and Casal, 2014). The distinction between strangers and siblings could also be based on genetic similarity, although both in Pisum sativum (pea) and Buchloe dactyloides (buffalograss; Falik et al., 2003; Gruntman and Novoplansky, 2004) intermediate responses and phenotypes during self/non-self root experiments hint at certain overlaps. Novoplansky, 2004) do not indicate shoot or reproductive mass changes, so root growth inhibition is not always paralleled by obvious aboveground modiﬁcations and could be species speciﬁc. Besides biomass alterations, quick physiological responses can be mediated by root identity recognition. In pea, root competition does not affect photosynthesis, although leaf dark respiration is halved, whereas root respiration increases in the vicinity of non- self roots (Meier et al., 2013). Root overproduction, at the expense of reproductive or shoot biomass, suggests that regulation of the identity recognition can be an important means to increase crop yields. In the cases in which the tragedy of the commons had been observed, isolation of plants from each other could enhance yield (e.g., biomass, seeds, fruits, and ﬂowers) with the same input of water and nutrients (Maina et al., 2002). Moreover, data derived from transcriptomics techniques, only recently applied in the ﬁeld of root identity recognition, can prove useful. Thus far, dif- ferentially expressed gene sets have been reported for intraspe- ciﬁc and interspeciﬁc competition of Arabidopsis thaliana (thale cress) plants and for Centaurea maculosa (spotted knapweed; Broz et al., 2008; Biedrzycki et al., 2011; Masclaux et al., 2012; Schmid et al., 2013). The existence of a core gene set involved in identity recognition, as suggested by Schmid et al. (2013), merits further research. Moreover, identity recognition seems to be evolutionarily conserved because it has been reported already in spermatophytes (Gorelick and Marler, 2014). IDENTITY RECOGNITION AND ITS IMPORTANCE The molecular biology behind root identity recognition should be tested exhaus- tively, for example, by employing deep sequencing methods. Biedrzycki et al. (2011) and Schmid et al. (2013) demonstrated that the molecular responses of root and pathogen recognition overlap. Comparison of their datasets with datasets of plant growth promoting rhizobacteria that can accelerate the growth and vegetative phase of plants (Poupin et al., 2013) can be most relevant for crops. The results might have great applications in agricultural practices, in addition to the discovery of the The impact of root growth inhibition by other plants, whether they are self, strangers, or kin, can be intuitively explained in terms of “space defense” and resource availability. Indeed, inhi- bition is less demanding than direct competition for the same nutrients in the shared space (for a review, see Schenk et al., 1999). In contrast, root growth overproliferation might maximize the nutrient uptake, but could also affect propagation. In this so- called “tragedy of the commons” that is demonstrated in soy- bean (Gersani et al., 2001) and Phaseolus varigaris (Kenya beans; Maina et al., 2002), the root overproliferation response reduces the reproductive biomass without competitive advantages. How- ever, other studies (Holzapfel and Alpert, 2003; Gruntman and Table 1 \| Overview of self/non-self root recognition studies: parameters analyzed, outcome, and used species. Parameter Effect Species Reference Root elongation rate Decline for non-self roots, no effect for self roots Ambrosia dumosa Mahall and Callaway (1991) Decline for both self and non-self roots Larrea tridentata Mahall and Callaway (1991) Root growth (length and/or number) No effect Andropogon gerardii Markham and Halwas (2011) Reduced for non-self, no effect for self Arabidopsis thaliana Biedrzycki et al. (2010) Fewer and shorter roots toward self Buchloe dactyloides Gruntman and Novoplansky (2004) Lateral roots More and longer lateral roots toward non-self Pisum sativum Falik et al. (2003) Root segregation Roots avoid non-self roots, no effect for self roots Arabidopsis thaliana Caffaro et al. (2011) Spatial segregation for self roots Fragaria chiloensis Holzapfel and Alpert (2003) Attraction for same genotype, avoidance for different genotypes Oryza sativa Fang et al. (2013) No effect Fragaria vesca Semchenko et al. (2007b) Avoidance for self and non-self Glechoma hederacea Semchenko et al. (2007b) Root biomass No effect of neighboring plants Avena sativa Semchenko et al. (2007a) Self-inhibition Glycine max Gersani et al. (2001) Less biomass in presence of self roots Trifolium repens Falik et al. IDENTITY RECOGNITION AND ITS IMPORTANCE FIGURE 1 \| Common root responses to neighboring plants and possible mechanisms of root identity recognition. (A) Range of root responses toward neighboring roots. Two plants are depicted (plant 1 and 2) of which plant 2 shows a differential root response because it is neighbored with plant 1. Plants 1 and 2 can be either of the same species or of different species. Examples (both intra- and interspeciﬁcally) of interacting species, are provided for each response: overproliferation of the root system, i.e., root biomass changes which may include main root length increases, more adventitious FIGURE 1 \| Common root responses to neighboring plants and possible mechanisms of root identity recognition. (A) Range of root responses FIGURE 1 \| Common root responses to neighboring plants and possible mechanisms of root identity recognition. (A) Range of root responses toward neighboring roots. Two plants are depicted (plant 1 and 2) of which plant 2 shows a differential root response because it is neighbored with plant 1. Plants 1 and 2 can be either of the same species or of different species. Examples (both intra- and interspeciﬁcally) of interacting species, are provided for each response: overproliferation of the root system, i.e., root biomass changes which may include main root length increases, more adventitious FIGURE 1 \| Common root responses to neighboring plants and possible mechanisms of root identity recognition. (A) Range of root responses toward neighboring roots. Two plants are depicted (plant 1 and 2) of which plant 2 shows a differential root response because it is neighbored with plant 1. Plants 1 and 2 can be either of the same species or of different species. Examples (both intra- and interspeciﬁcally) of interacting species, are provided for each response: overproliferation of the root system, i.e., root biomass changes which may include main root length increases, more adventitious November 2014 \| Volume 5 \| Article 614 \| 2 Frontiers in Plant Science \| Functional Plant Ecology Identity recognition in plants Depuydt a desert shrub, tolerate neither other Larrea nor Ambrosia roots in their proximity (Mahall and Callaway, 1991, 1992). Since these ﬁrst ﬁndings, self/non-self identity recognition has been studied in various species, but no uniform responses are observed. In some species, non-self roots seem to promote root growth traits, whereas root growth is not enhanced by self roots (Table 1). SELF/NON-SELF RECOGNITION IN ROOTS: PITFALLS The interpretation of some of the experiments concerning self/non-self recognition remains somewhat controversial (for a review, see Chen et al., 2012). As already mentioned above (see also Table 1), different species are used to study self/non- self identity recognition in plants, making it difﬁcult to draw clear conclusions due to likely species-speciﬁc and genotype- speciﬁc effects (Fang et al., 2011). Moreover, the root growth strategies of the species under study might inﬂuence the outcome of the experiments. For instance, in the strawberry/ivy experi- ments, strawberries grow clonally and always spread widely within plant communities, which may well affect whether a neighboring root will be attracted or avoided (de Kroon, 2007; Faget et al., 2013). Moreover, several parameters have been analyzed during root recognition research, such as root biomass, adaptation of root architecture/morphology, and root length (Table 1). Most studies focus on root biomass, but root architecture may well be the primary and quickest response that does not necessarily impose an altered photosynthate allocation when compared to mere root growth, as demonstrated in Phaseolus vulgaris (common bean; Nord et al., 2011). Split-root experimental systems have been used to study self/non-self recognition in root (Gersani et al., 2001; Maina et al., 2002; O’Brien et al., 2005), but the effects of pot volume and nutrient levels, which are important factors determining root growth, are difﬁcult to correct and will inﬂuence responses to strangers and/or identity recognition. Indeed, several results can be rationalized as responses to soil volume (Schenk, 2006; Hess and de Kroon, 2007; Markham and Halwas, 2011). Nonetheless, other experiments have unequivocally demonstrated root mass changes that depend solely on the identity of the interacting root, as, for instance, in buffalograss (Gruntman and Novoplansky, 2004). In addition, plants would react to available resources rather than to the presence of a neighbor that will, while growing, deplete the same soil zone from soil nutrients (Semchenko et al., 2007a). Indeed, roots grow preferentially where supplies are most accessible (Gersani et al., 1998; Hodge, 2009), the probable reason for avoidance of other root systems. Therefore, nutrient levels and detection of the presence of other root systems are often con- founded as well (O’Brien et al., 2005; Klemens, 2008; Fang et al., 2011). MECHANISMS OF IDENTITY RECOGNITION IN PLANTS For self-incompatibility, speciﬁc ligands are involved (Sawada et al., 2014). Volatile cues from self cuttings of Artemisia triden- tata (sagebrush) increase herbivore resistance when compared to volatiles from non-self cuttings (Karban and Shiojiri, 2009). In addition, light signals mediate discrimination between kin and neighbors, leading to leaf repositioning which requires auxin biosynthesis (Crepy and Casal, 2014). Regarding root commu- nication (Figure 1B), mediation through electrical signals has been proposed (Schenk et al., 1999). Furthermore, experimen- tal data in pea demonstrate that hormonal rhythms might be implicated (Falik et al., 2003), as corroborated by Gruntman and Novoplansky (2004) who concluded that an unknown physiolog- ical mechanism (i.e., electrical or hormonal rhythm) might be responsible for root discrimination in buffalograss. Differential internal oscillatory signals and their resonant ampliﬁcation would lead to the recognition of a non-self root. Alternatively, perception of neighboring roots has been proposed to be attributed to asso- ciated microorganisms and their secreted substances (Steenhoudt and Vanderleyden, 2000). Transcriptomics data have conﬁrmed this hypothesis by the striking overlap of genes associated with plant reactions to neighbors and with responses to pathogens (Biedrzycki et al., 2011; Schmid et al., 2013). Surprisingly, how- ever, in axenic cultures of Arabidopsis plants exposed only to root exudates—i.e., the mixture of compounds that are actively secreted or passively released by roots (Bais et al., 2006)—of strangers, the induction of lateral root formation is higher than that after exposure to sibling exudates. This observation sug- gests that a soluble chemical, originating from root exudation, might be responsible for identity recognition (Biedrzycki and Bais, 2010a,b; Biedrzycki et al., 2010). By means of the root secretion inhibitor sodium orthovanadate, which blocks active root secretion of several phenolic compounds in Arabidopsis, seedlings no longer recognize strangers, implying that active secretion by roots is required for kin recognition (Biedrzycki et al., 2010). However, self/non-self recognition is not inﬂu- enced by the secretion inhibitor, conﬁrming that two separate identity recognition mechanisms exist. Caffaro et al. (2011) demonstrated that self and non-self exudates similarly reduce root growth, but that addition of activated charcoal, speciﬁ- cally reverses the effect of self roots on root growth, indicating that reduced amounts of secondary metabolites in the medium affected self/non-self recognition. Consistently, root proximity is important for interaction responses in Oryza sativa (rice). MECHANISMS OF IDENTITY RECOGNITION IN PLANTS Exclusion of aerial interactions by shoot separation experiments hinted at the induction of interactions by root exudates that diffuse into the medium rather than by physical contact (Fang et al., 2013). In Deschampsia cespitosa (tufted-hair grass), root exudates have also been demonstrated as cues of neighbor identity that control root mass and morphology (Semchenko et al., 2014). Noteworthy, root-object recognition might occur via allelopathic root exudates, as shown in pea (Falik et al., 2005), but, according to recent evidence in rice, could also be mediated via different processes that require physical contact of IDENTITY RECOGNITION AND ITS IMPORTANCE (2006) Overproliferation toward non-self Phaseolus varigaris Maina et al. (2002) www.frontiersin.org November 2014 \| Volume 5 \| Article 614 \| 3 November 2014 \| Volume 5 \| Article 614 \| 3 www.frontiersin.org www.frontiersin.org Identity recognition in plants Depuydt mechanisms responsible for identity recognition that have long been elusive. the root tip with the obstacle (Fang et al., 2013). Unraveling the exact nature of the signals that trigger identity recogni- tion would be a gigantic leap forward in root–root interaction studies. Frontiers in Plant Science \| Functional Plant Ecology REFERENCES Bais, H. P., Weir, T. L., Perry, L. G., Gilroy, S., and Vivanco, J. M. (2006). The role of root exudates in rhizosphere interactions with plants and other organisms. Annu. Rev. Plant Biol. 57, 233–266. doi: 10.1146/annurev.arplant.57.032905. 105159 Baldwin, J. P., and Tinker, P. B. (1972). A method for estimating the lengths and spatial patterns of two interpenetrating root systems. Plant Soil 37, 209–213. doi: 10.1007/BF01578497 Biedrzycki, M. L., and Bais, H. P. (2010a). Kin recognition in plants: a mysterious behaviour unsolved. J. Exp. Bot. 61, 4123–4128. doi: 10.1093/jxb/erq250 Biedrzycki, M. L., and Bais, H. P. (2010b). Kin recognition: another bio- logical function for root secretions. Plant Signal. Behav. 5, 401–402. doi: 10.4161/psb.5.4.10795 Biedrzycki, M. L., Jilany, T. A., Dudley, S. A., and Bais, H. P. (2010). Root exudates mediate kin recognition in plants. Commun. Integr. Biol. 3, 28–35. doi: 10.4161/cib.3.1.10118 Biedrzycki, M. L., Lakshmannan, V., and Bais, H. P. (2011). Transcriptome analysis of Arabidopsis thaliana plants in response to kin and stranger recognition. Plant Signal. Behav. 10, 1515–1524. doi: 10.4161/psb.6.10.16525 Bini, G., and Chisci, P. (1961). Some observations on the effects of roots of peach and pear. Riv. Ortoﬂorofrutt. Ital. 45, 345–352 [in Italian with English summary]. Moreover, quantiﬁcation of belowground interactions is difﬁ- cult, certainly at the level of the individual root. Although beyond the scope of this minireview, recent advances in imaging technol- ogy might be helpful. A transparent gel system is now developed that allows imaging and three-dimensional reconstruction to quantitatively assess root growth parameters during interaction studies (Fang et al., 2013). Likewise, ﬂuorescent markers and horizontal minirhizotrons imaging systems (Faget et al., 2009, 2012) have proven successful to study maize, Lolium multiﬂo- rum (Italian ryegrass), and soybean interactions. As a drawback, genetically modiﬁed plants are required that, hence, hamper ecological applications. These non-destructive technologies out- compete the mere analysis of root biomass and are promising alternatives for root–root interaction and root identity recogni- tion studies. Borch, K., Bouma, T. J., Lynch, J. P., and Brown, K. M. (1999). Ethylene: a regulator of root architectural responses to soil phosphorus availability. Plant Cell Environ. 22, 425–431. doi: 10.1046/j.1365-3040.1999.00405.x Bradshaw, A. D. (1965). Evolutionary signiﬁcance of phenotypic plasticity in plants. Adv. Genet. 13, 115–155. doi: 10.1016/S0065-2660(08)60048-6 Broz, A. K., Manter, D. K., Callaway, R. M., Paschke, M. W., and Vivanco, J. M. (2008). A molecular approach to understanding plant–plant interactions in the context of invasion biology. Funct. REFERENCES Plant Biol. 35, 1123–1134. doi: 10.1071/FP08155 Caffaro, M. M., Vivanco, J. M., Gutierrez Boem, F. H., and Rubio, G. (2011). The effect of root exudates on root architecture in Arabidopsis thaliana. Plant Growth Regul. 64, 241–249. doi: 10.1007/s10725-011-9564-3 Cahill, J. F. Jr., McNickle, G. G., Haag, J. J., Lamb, E. G., Nyanumba, S. M., and St. Clair, C. C. (2010). Plants integrate information about nutrients and neighbors. Science 328, 1657. doi: 10.1126/science.1189736 Cardoso, C., Ruyter-Spira, C., and Bouwmeester, H. J. (2011). Strigolactones and root infestation by plant-parasitic Striga, Orobanche and Phelipanche spp. Plant Sci. 180, 414–420. doi: 10.1016/j.plantsci.2010.11.007 In conclusion, exciting advances in the ﬁeld of self/non- self recognition of roots have been made over the recent years. New imaging technologies will not only aid to analyze the root response in a non-destructive way, but will also allow kinetics studies that will help to understand the mechanisms of root identity recognition and to avoid the confusion of the effects of root interactions with those of nutrients and root volume. The identiﬁcation of a core set of genes involved in neighbor detection merits further research and functional analyses. New and high-resolution chemical analysis techniques, besides state- of-the-art techniques used to measure electrical signals in planta, as well as molecular biological approaches should be utilized to clarify root–root identity recognition. The obtained results can have an enormous impact on the research in plant biology and development as well as on the agricultural and ecological research ﬁelds and practice. Casper, B. B., and Jackson, R. B. (1997). Plant competition underground. Annu. Rev. Ecol. Syst. 28, 545–570. doi: 10.1146/annurev.ecolsys.28.1.545 Chen, B. J. W., During, H. J., and Anten, N. P. R. (2012). Detect thy neighbor: identity recognition at the root level in plants. Plant Sci. 195, 157–167. doi: 10.1016/j.plantsci.2012.07.006 Crepy, M. A., and Casal, J. J. (2014). Photoreceptor-mediated kin recognition in plants. New Phytol. doi: 10.1111/nph.13040 [Epub ahead of print]. de Kroon, H. (2007). How do roots interact? Science 318, 1562–1563. doi: 10.1126/science.1150726 Dudley, S. A., and File, A. L. (2007). Kin recognition in an annual plant. Biol. Lett. 3, 435–438. doi: 10.1098/rsbl.2007.0232 Faget, M., Herrera, J. M., Stamp, P., Aulinger-Leipner, I., Frossard, E., and Liedgens, M. (2009). The use of green ﬂuorescent protein as a tool to identify roots in mixed plant stands. Funct. Plant Biol. 36, 930–937. doi: 10.1071/ FP09125 Faget, M., Liedgens, M., Feil, B., Stamp, P., and Herrera, J. M. SELF/NON-SELF RECOGNITION IN ROOTS: PITFALLS For example, common bean plants will change their root system architecture and produce fewer roots in soil patches that are occupied by neighboring roots (Nord et al., 2011), possibly in relation to the phosphorus concentration in the soil, which is nearly immobile and inﬂuences the developmental plasticity of roots (e.g., Borch et al., 1999; reviewed in Ticconi and Abel, 2004). Nonetheless, transcriptomics analysis of Arabidopsis in the presence or absence of competing Hieracium pilosella (mouse- ear hawkweed) clearly indicate that sensing neighboring roots occurs before resource depletion is discovered (Schmid et al., 2013). November 2014 \| Volume 5 \| Article 614 \| 4 Frontiers in Plant Science \| Functional Plant Ecology Depuydt Depuydt Identity recognition in plants Costas Stathopoulos, and Martine De Cock for help in preparing it. Stephen Depuydt was a postdoctoral fellow of the Research Foundation-Flanders. Costas Stathopoulos, and Martine De Cock for help in preparing it. Stephen Depuydt was a postdoctoral fellow of the Research Foundation-Flanders. Costas Stathopoulos, and Martine De Cock for help in preparing it. Stephen Depuydt was a postdoctoral fellow of the Research Foundation-Flanders. In addition to nutrients and soil volume, water availability must be considered as well. In Ambrosia dumosa, intraspeciﬁc water competition is thought to be the reason for growth reduc- tion when self roots are recognized (Mahall and Callaway, 1992). Hence, local changes in the microclimate should also be taken into account; for instance, root temperature gradients affect root productivity and lead to top soil accumulation (Füllner et al., 2012). Indeed, these problems have been recognized and highly controlled experiments have been set up in which plants are grown in preconditioned liquid media (Biedrzycki et al., 2010; Caffaro et al., 2011). Nevertheless, because these systems remain artiﬁcial, their relevance in natural soil systems can be ques- tioned. Alternatively, the use of clonal ramet pairs (Holzapfel and Alpert, 2003; Gruntman and Novoplansky, 2004; Semchenko et al., 2007b) was suggested to circumvent the above mentioned problems. However, although pot volume and nutrient levels could indeed be kept constant, disconnected ramets would still be considered to access only half the amount of nutrients (i.e., two one-root system plants) as compared to connected plants (a single two-root system plant) that have access to the full amount of nutrients. These approaches have thus been criticized as well (Hess and de Kroon, 2007). REFERENCES (2012). Root growth of maize in an Italian ryegrass living mulch studied with a non-destructive method. Eur. J. Agron. 36, 1–8. doi: 10.1016/j.eja.2011.08.002 method. Eur. J. Agron. 36, 1–8. doi: 10.1016/j.eja.2011.08.002 Faget, M., Nagel, K. A., Walter, A., Herrera, J. M., Jahnke, S., Schurr, U., et al. (2013). Root–root interactions: extending our perspective to be more inclusive of the range of theories in ecology and agriculture using in-vivo analyses. Ann. Bot. 112, 253–266. doi: 10.1093/aob/mcs296 ACKNOWLEDGMENTS Vertical gradient in soil temperature stimulates development and increases biomass accumulation in barley. Plant Cell Environ. 35, 884–892. doi: 10.1111/j.1365-3040.2011.02460.x O’Brien, E. E., Gersani, M., and Brown, J. S. (2005). Root proliferation and seed yield in response to spatial heterogeneity of below-ground competition. New Phytol. 168, 401–412. doi: 10.1111/j.1469-8137.2005. 01520.x Gersani, M., Abramsky, Z., and Falik, O. (1998). Density-dependent habitat selec- tion in plants. Evol. Ecol. 12, 223–234. doi: 10.1023/A:1006587813950 Gersani, M., Brown, J. S., O’Brien, E. E., Maina, G. M., and Abramsky, Z. (2001). Gersani, M., Brown, J. S., O’Brien, E. E., Maina, G. M., and Abramsky, Z. (2001). Tragedy of the commons as a result of root competition. J. Ecol. 89, 660–669. doi: 10.1046/j.0022-0477.2001.00609.x Tragedy of the commons as a result of root competition. J. Ecol. 89, 660–669. doi: 10.1046/j.0022-0477.2001.00609.x Poupin, M. J., Timmermann, T., Vega, A., Zuñiga, A., and González, B. (2013). Effects of the plant growth-promoting bacterium Burkholderia phytoﬁrmans PsJN throughout the life cycle of Arabidopsis thaliana. PLoS ONE 8:e69435. doi: 10.1371/journal.pone.0069435 Goldberg, D. E., and Barton, A. M. (1992). Patterns and consequences of interspe- ciﬁc competition in natural communities: a review of ﬁeld experiments with plants. Am. Nat. 139, 771–801. doi: 10.1086/285357 p Gorelick, R., and Marler, T. E. (2014). Kin recognition by roots occurs in cycads and probably in conifers. Commun. Integr. Biol. 7:e28009. doi: 10.4161/cib. 28009 Raper, C. D., and Barber, S. A. (1970). Rooting systems of soybeans. I. Dif- ferences in root morphology among varieties. Agron. J. 62, 581–584. doi: 10.2134/agronj1970.00021962006200050009x Gruntman, M., and Novoplansky, A. (2004). Physiologically mediated self/non- self discrimination in roots. Proc. Natl. Acad. Sci. U.S.A. 101, 3863–3867. doi: 10.1073/pnas.0306604101 g j Robinson, D. (1994). The responses of plants to non-uniform supplies of nutrients. New Phytol. 127, 635–674. doi: 10.1111/j.1469-8137.1994. tb02969.x Hess, L., and de Kroon, H. (2007). Effects of rooting volume and nutrient availabil- ity as an alternative explanation for root self/non-self discrimination. J. Ecol. 95, 241–251. doi: 10.1111/j.1365-2745.2006.01204.x Sanabria, N., Goring, D., Nürnberger, T., and Dubery, I. (2008). Self/nonself perception and recognition mechanisms in plants: a comparison of self- incompatibility and innate immunity. New Phytol. 178, 503–514. doi: 10.1111/j. 1469-8137.2008.02403.x Hodge, A. (2009). Root decisions. Plant Cell Environ. 32, 628–640. doi: 10.1111/j.1365-3040.2008.01891.x Sawada, H., Morita, M., and Iwano, M. (2014). Self/non-self recogni- tion mechanisms in sexual reproduction: new insight into the self- incompatibility system shared by ﬂowering plants and hermaphroditic animals. Biochem. Biophys. Res. Commun. 450, 1142–1148. ACKNOWLEDGMENTS doi: 10.1016/j.bbrc.2014. 05.099 Holzapfel, C., and Alpert, P. (2003). Root cooperation in a clonal plant: connected strawberries segregate roots. Oecologia 134, 72–77. doi: 10.1007/s00442-002- 1062-x Karban, R., and Shiojiri, K. (2009). Self-recognition affects plant communica- tion and defense. Ecol. Lett. 12, 502–506 [Erratum Ecol. Lett. 12, 999]. doi: 10.1111/j.1461-0248.2009.01313.x Schenk, H. J. (2006). Root competition: beyond resource depletion. J. Ecol. 94, 725– 739. doi: 10.1111/j.1365-2745.2006.01124.x Kitashiba, H., and Nasrallah, J. B. (2014). Self-incompatibility in Brassicaceae crops: lessons for interspeciﬁc incompatibility. Breed. Sci. 64, 23–37. doi: 10.1270/jsbbs.64.23 Schenk, H. J., Callaway, R. M., and Mahall, B. E. (1999). Spatial root segrega- tion: are plants territorial? Adv. Ecol. Res. 28, 145–180. doi: 10.1016/S0065- 2504(08)60032-X Klemens, J. A. (2008). Kin recognition in plants? Biol. Lett. 4, 67–68. doi: 10.1098/rsbl.2007.0518 Schmid, C., Bauer, S., Müller, B., and Bartelheimer, M. (2013). Belowground neighbor perception in Arabidopsis thaliana studied by transcriptome analysis: roots of Hieracium pilosella cause biotic stress. Front. Plant Sci. 4:296. doi: 10.3389/fpls.2013.00296 Lehmann, J., Peter, I., Steglich, C., Gebauer, G., Huwe, B., and Zech, W. (1998). Below-ground interactions in dryland agroforestry. For. Ecol. Manage. 111, 157– 169. doi: 10.1016/S0378-1127(98)00322-3 Semchenko, M., Hutchings, M. J., and John, E. A. (2007a). Challenging the tragedy of the commons in root competition: confounding effects of neigh- bour presence and substrate volume. J. Ecol. 95, 252–260. doi: 10.1111/j.1365- 2745.2007.01210.x Mahall, B. E., and Callaway, R. M. (1991). Root communication among desert shrubs. Proc. Natl. Acad. Sci. U.S.A. 88, 874–876. doi: 10.1073/pnas.88.3.874 Mahall, B. E., and Callaway, R. M. (1992). Root communication mechanisms and intracommunity distributions of two Mojave desert shrubs. Ecology 73, 2145– 2151. doi: 10.2307/1941462 Mahall, B. E., and Callaway, R. M. (1992). Root communication mechanisms and , , y, ( ) intracommunity distributions of two Mojave desert shrubs. Ecology 73, 2145– 2151. doi: 10.2307/1941462 Semchenko, M., John, E. A., and Hutchings, M. J. (2007b). Effects of physical connection and genetic identity of neighbouring ramets on root-placement patterns in two clonal species. New Phytol. 176, 644–654. doi: 10.1111/j.1469- 8137.2007.02211.x Maina, G. G., Brown, J. S., and Gersani, M. (2002). Intra-plant versus inter-plant root competition in beans: avoidance, resource matching or tragedy of the commons. Plant Ecol. 160, 235–247. doi: 10.1023/A:1015822003011 Semchenko, M., Saar, S., and Lepik, A. (2014). Plant root exudates mediate neighbour recognition and trigger complex behavioural changes. New Phytol. 204, 631–637. doi: 10.1111/nph.12930 Mamolos, A. P., Elisseou, G. K., and Veresoglou, D. S. (1995). ACKNOWLEDGMENTS The author wishes to thank Soﬁe Goormachtig for helpful com- ments on the manuscript, and Kirsten De Wilde, Hilde Nelissen, Falik, O., de Kroon, H., and Novoplansky, A. (2006). Physiologically-mediated self/nonself root discrimination in Trifolium repens has mixed effects on November 2014 \| Volume 5 \| Article 614 \| 5 www.frontiersin.org Depuydt Identity recognition in plants Masclaux, F. G., Bruessow, F., Schweizer, F., Gouhier-Darimont, C., Keller, L., and Reymond, P. (2012). Transcriptome analysis of intraspeciﬁc competition in Arabidopsis thaliana reveals organ- speciﬁc signatures related to nutrient acquisition and general stress response pathways. BMC Plant Biol. 12:227. doi: 10.1186/1471-2229- 12-227 plant performance. Plant Signal. Behav. 1, 116–121. doi: 10.4161/psb.1.3. 2639 Falik, O., Reides, P., Gersani, M., and Novoplansky, A. (2003). Self/non-self discrimination in roots. J. Ecol. 91, 525–531. doi: 10.1046/j.1365-2745.2003. 00795.x Falik, O., Reides, P., Gersani, M., and Novoplansky, A. (2005). Root naviga- tion by self inhibition. Plant Cell Environ. 28, 562–569. doi: 10.1111/j.1365- 3040.2005.01304.x Massey, A. B. (1925). Antagonism of the walnuts (Juglans nigra and J. cinerea L.) in certain plant association. Phytopathology 15, 773–784. Fang, S., Clark, R. T., Zheng, Y., Iyer-Pascuzzi, A. S., Weitz, J. S., Kochian, L. V., et al. (2013). Genotypic recognition and spatial responses by rice roots. Proc. Natl. Acad. Sci. U.S.A. 110, 2670–2675. doi: 10.1073/pnas.1222821110 Meier, I. C., Angert, A., Falik, O., Shelef, O., and Rachmilevitch, S. (2013). Increased root oxygen uptake in pea plants responding to non-self neighbors. Planta 238, 577–586. doi: 10.1007/s00425-013-1910-4 Muller, C. H. (1946). Root Development and Ecological Relations of Guayule. USDA Technical Bulletin 923. Washington, DC: United States Department of Agriculture, 1–114. Fang, S., Gao, X., Deng, Y., Chen, X., and Liao, H. (2011). Crop root behavior coordinates phosphorus status and neighbors: from ﬁeld studies to three- dimensional in situ reconstruction of root system architecture. Plant Physiol. 155, 1277–1285. doi: 10.1104/pp.110.167304 g Murphy, G. P., and Dudley, S. A. (2009). Kin recognition: competition and cooperation in Impatiens (Balsaminaceae). Am. J. Bot. 96, 1990–1996. doi: 10.3732/ajb.0900006 Ford, E. D. (2014). The dynamic relationship between plant architecture and competition. Front. Plant Sci. 5:275. doi: 10.3389/fpls.2014.00275 Nord, E. A., Zhang, C., and Lynch, J. P. (2011). Root responses to neighbour- ing plants in common bean are mediated by nutrient concentration rather than self/non-self recognition. Funct. Plant Biol. 38, 941–952. doi: 10.1071/ FP11130 Füllner, K., Temperton, V. M., Rascher, U., Jahnke, S., Rist, R., Schurr, U., et al. (2012). November 2014 \| Volume 5 \| Article 614 \| 7 Received: 05 September 2014; accepted: 20 October 2014; published online: 06 November 2014. Citation: Depuydt S (2014) Arguments for and against self and non-self root recogni- tion in plants. Front. Plant Sci. 5:614. doi: 10.3389/fpls.2014.00614 This article was submitted to Functional Plant Ecology, a section of the journal Frontiers in Plant Science. Copyright © 2014 Depuydt. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the orig- inal author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. ACKNOWLEDGMENTS Depth of root activity of coexisting grassland species in relation to N and P additions, measured using nonradioactive tracers. J. Ecol. 83, 643–652. doi: 10.2307/2261632 Steenhoudt, O., and Vanderleyden, J. (2000). Azospirillum, a free-living nitrogen- ﬁxing bacterium closely associated with grasses: genetic, biochemical and ecological aspects. FEMS Microbiol. Rev. 24, 487–506. doi: 10.1111/j.1574- 6976.2000.tb00552.x Markham, J., and Halwas, S. (2011). Effect of neighbour presence and soil volume on the growth of Andropogon gerardii Vitman. Plant Ecol. Divers. 4, 265–268. doi: 10.1080/17550874.2011.618515 November 2014 \| Volume 5 \| Article 614 \| 6 Frontiers in Plant Science \| Functional Plant Ecology Depuydt Identity recognition in plants Ticconi, C. A., and Abel, S. (2004). Short on phosphate: plant surveillance and countermeasures. Trends Plant Sci. 9, 548–555. doi: 10.1016/j.tplants.2004. 09.003 Received: 05 September 2014; accepted: 20 October 2014; published online: 06 November 2014. Citation: Depuydt S (2014) Arguments for and against self and non-self root recogni- tion in plants. Front. Plant Sci. 5:614. doi: 10.3389/fpls.2014.00614 Tosti, G., and Thorup-Kristensen, K. (2010). Using coloured roots to study root interaction and competition in intercropped legumes and non-legumes. J. Plant Ecol. 3, 191–199. doi: 10.1093/jpe/rtq014 This article was submitted to Functional Plant Ecology, a section of the journal Frontiers in Plant Science. This article was submitted to Functional Plant Ecology, a section of the journal Frontiers in Plant Science. Wilson, S. D., and Tilman, D. (1993). Plant competition and resource avail- ability in response to disturbance and fertilization. Ecology 74, 599–611. doi: 10.2307/1939319 Copyright © 2014 Depuydt. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the orig- inal author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Conﬂict of Interest Statement: The author declares that the research was con- ducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. November 2014 \| Volume 5 \| Article 614 \| 7 www.frontiersin.org www.frontiersin.org
https://openalex.org/W4385494032	https://pressto.amu.edu.pl/index.php/eip/article/download/39020/33197	English	null	Women’s Reproductive Health Rights in Poland. Between a Druggists’ Conscience Clause and Their Legal Duty to Provide Contraceptives	Ethics in Progress	2,023	cc-by-sa	6,869	Ewa Nowak (Adam Mickiewicz University, Poznań; ewanowak@amu.edu.pl) ORCID: 0000-0002-5722-7711 Women’s Reproductive Health Rights in Poland. Between a Druggists’ Conscience Clause and Their Legal Duty to Provide Contraceptives Women’s Reproductive Health Rights in Poland. Between a Druggists’ Conscience Clause and Their Legal Duty to Provide Contraceptives (Adam Mickiewicz University, Poznań; kinga.ciereszko@amu.edu.pl) Kinga Ciereszko ORCID: 0000-0002-8016-1140 (Adam Mickiewicz University, Poznań; kinga.ciereszko@amu.edu.pl) Kinga Ciereszko ORCID: 0000-0002-8016-1140 Ethics in Progress (ISSN 2084-9257). Vol. 14 (2023). No. 1, Art. #6, pp. 94-109. DOI:10.14746/eip.2023.1.6 Creative Commons BY-SA 4.0 Ethics in Progress (ISSN 2084-9257). Vol. 14 (2023). No. 1, Art. #6, pp. 94-109. DOI:10.14746/eip.2023.1.6 Creative Commons BY-SA 4.0 Karolina Napiwodzka (Adam Mickiewicz University, Poznań; karolina.napiwodzka@amu.edu.pl) ORCID: 0000-0003-1845-7992 1 This article is an extended and revised version of „Health vs. Conscience Clause: Examining Access to Contraceptives and the Observance of Women’s Reproductive Rights in Poland,” originally published in the Journal of Public Value 1(2021):153–164. In addition, the research findings were reported at three international conferences on the basis of peer-reviewed submissions: “Fair Medicine and Artificial Intelligence: Chances, Challenges, Consequences”, Tübingen, 3-5.03.2021; “The Korea Association for Public Value (KAPV) International Conference on Public Values”, Seoul, 25.03.2022; American Educational Research Association (AERA) International Annual Meeting: “Accepting Educational Responsibility”, Washington D.C., 8-12.04.2021. Kay Hemmerling (Institute for Moral-Democratic Competence, Mühlbeck/Leipzig; kayhemmerling@aol.com) Abstract: This article recommends the promotion of moral competence in the health and pharmacy professions to enable them to respect human and patient health rights with a focus on the provision of reproductive and sexual health care services. In certain cultures, health care and drug providers follow their conscientious objection (conscience clause) and decline to perform specific health services, including the provision of legal contraceptives in cases protected by legal and human rights. Such malpractices may violate patients’ and purchasers’ legitime rights. The article also presents findings obtained in Poland with N=121 women experimentally interviewed to examine their experiences as contraception purchasers, to assess their preference concerning facing human vs. robotic pharmacists, to manage the risk of refusal argued by the conscientious objection, and to score their moral competence with one of the dilemmas included in the MCT by G. Lind. This study demonstrated that purchasers with higher C-score (C for moral competence) would not just prefer a robotic pharmacist without a ‘conscience’ but, rather, a competent sales staff able to instruct the patient and advice her on any related queries. It further results that participants with higher moral competence are thus less likely to trust the medical expertise of artificial intelligence. We conclude that public institutions in pluralistic societies must manage normative reproductive health contexts more inclusively, and the election, education, and practice of health professionals in the public health care sector require the development of a normative mindset toward respecting the rights of all patients instead of respecting them selectively at the diktat of particularistic conscience. Keywords: Conscience clause; access to legal contraception; reproductive and sexual human rights; women’s rights; Moral Competence Test; thought experiment. 94 Kinga Ciereszko, Karolina Napiwodzka, Ewa Nowak, Kay Hemmerling 1. Reproductive and Sexual Health in Terms of Human Rights Reproductive and sexual health are aspects of human health, as one of the core public and global values1. These values, as well as the corresponding, reproductive healthcare and equal access to its services, were for the first time articulated and proclaimed in the Cairo Declaration on Population and Development (1994). The latter urges governments “to help support the provision of reproductive health and family planning services as widely as possible. We further urge Governments to ensure that all population and development policies and programmes in our countries safeguard internationally recognized human rights” (Cairo Declaration, cf., § 4). It is further declared that “the empowerment of women and the improvement of their political social, economic and health status are highly important ends in themselves. Reproductive and sexual health is related to multiple human rights, including the right to life, the right to be free from torture, the right to health, the right to privacy, the right to education, and the prohibition of discrimination” (United Nations Human Rights). Not only equal rights, but also reduction of maternal mortality rates and decreasing the risks related to abortion (“a major public health concern for women all over the world,” according to § 6), sexual violence, unplanned pregnancies, and untreatable diseases, all belong to the scope of ‘the new generation’ human rights with their focus on contraception accessibility (Rudolf 2016), provision and distribution supported by social consensus, legal rights and governmental policies. Further, individual reproductive and sexual health represent public health and wellbeing conceptualized by public health ethics (according to The Oxford Handbook of Public Health Ethics) as noninstrumentalizable. Human rights belong to general moral principles so that to be properly ranked, recognized, and respected, a high-type moral orientation (Lind 2019) and the ability to make a “principled” moral judgment must be developed in people. Occasionally named ‘postconventional’ in Kohlberg’s approach, general principles can be universalized consensually by public democratic discourse as a modern “faculty of principles,” as Habermas puts it. Principles remain “accumulated in postconventional discourses of justification” (Habermas 2003, 275). Accordingly, an advanced stage of conscience development would be “principled conscience” (Stage 6 in terms of Kohlberg), able to manage conflicts between human rights (fundamental rights, respectively) and particular contents, specifically these of particular conscience (e.g., Drozd 2013; Merks et al. 2015; Erstad 2019; Flynn 2008; Wicclair 2006; Nelson 2005; Yoder 2007; Curlin 2004). 1. Reproductive and Sexual Health in Terms of Human Rights “Violations of women’s sexual and reproductive health rights are often deeply engrained 95 Women’s Reproductive Health Rights in Poland in societal values pertaining to women’s sexuality. Patriarchal concepts of women’s roles within the family mean that women are often valued based on their ability to reproduce” (CEDAW § 16). CEDAW “guarantees women equal rights in deciding freely and responsibly on the number and spacing of their children and to have access to the information, education and means to enable them to exercise these rights,” including access to contraception. “The human rights of women include their right to have control over and decide freely and responsibly on matters related to their sexuality, including sexual and reproductive health, free of coercion, discrimination and violence” (cf.) which might not be obvious for subjects with insufficient moral competence, rather than being due to their moral or religious affiliations. It was one of the characteristics of modernity to take health out of the confines of religion and charity and make it a key element of action of the state and the rights of citizenship. The process, initially within the context of the constitution of the nation state, today needs to go global as a key dimension of global justice. Global health needs to move out of the charity mode of bilateral aid and philanthropy into the realm of rights, citizenship, and a global contract (Kickbush 2004, 631). We then suggest that both the violation of fellow citizens’ human rights and core values, as well as meeting an equilibrium or a reasonable disagreement (Wilkinson et al. 2016) between the subject’s own vs. fellow subjects’ principles, may depend on the subjects’ moral competence scores, but remains unrelated to the subjects’ moral or religious opinions (Mishtal 2009). This may apply to health and pharmacy professionals. 2. The Janus Face of Conscience In the history of philosophy, theology, and psychology, the concept of conscience has had opposed connotations, ranging from subjective, irrational certainty as opposed to reasoning in terms of objective and universal norms (as in the case of Sophocles’ Antigone) to the inner instance of independent judgment (“tribunal of conscience” and “the inherent judge of oneself” in Kant), the tension between “id” and “superego” according to Freud, and, finally, a faculty for judgment that can be socialized, that is, evolve from the particular certainty to more universal justifications of its judgments. According to Hegel, Conscience expresses the absolute right of subjective self-consciousness, namely, to know (…) what is right and duty and to acknowledge nothing but what it thus knows to be good while asserting that what it thus knows and wills is in truth right and duty. As this unity of subjective knowledge and that which is in and for itself, conscience is a sanctuary, and that would be sacrilege to touch it. However, if the conscience of a particular individual is in accordance with this idea of conscience, if what it holds to be good (...) is really good, this can be seen only from the content of this desire for good. In turn, what is right and duty, as in and of itself reasonable (...), is essentially not the individual’s proper interest (...) For this reason, the state cannot recognize conscience in its original form (Hegel 2009, § 137). 96 96 Kinga Ciereszko, Karolina Napiwodzka, Ewa Nowak, Kay Hemmerling Conscience is not always satisfied with verdicts in the field of a subject’s private conduct outside the public-institutional sphere. It may pretend to the overall validity of its verdicts and hence show dictatorial inclinations. This, in turn, would undermine the foundations of a modern, pluralistic society and its public institutions, as well as position such claims for validity at the antipodes of the “principled conscience” advocated by Kohlberg (and implicitly by Habermas). This subjectivity, as the abstracted self-determination and pure certainty only of oneself, volatilizes all determinacy of right [and] duty (...) manifesting here just as a judgmental power to determine (...) from itself only what is good and what is not good (…) [This type of] Conscience (...) is par excellence that of being on the verge of turning to evil. (Hegel 2009, § 138 – § 139; square brackets added). 2. The Janus Face of Conscience However, this does not imply that modern individuals are deprived of the right to their conscience. Although the right to conscience freedom is not, in Luhmann’s words, a ‘supra-positive right from the otherworld’, it has been included in the human and fundamental rights list. “Everyone has the right to their conscience” (Luhmann 1965, 261), which finds justification in Hegel’s “philosophical jurisprudence”: Morality, ethical life, and state interest [including the Rechtsstaat – K.C. et al.] (...) are each distinctive right since each form is a determination and reality of freedom. They can only collide insofar as they stand on the same line of being rights; if the moral standpoint of the spirit were not also a right and freedom in one of its forms, it could not come into collision with the right of personality or another (Hegel 2009, § 30). In this my deepest and most personal certainty, where my belief has its origin and locus, I am free for myself against others [bin ich frei für mich gegen andere], and the sort of belief or other grounds, be they emotional or reflective, is irrelevant here [sind hier gleichgültig] (...) After all, there is an essential distinction between this very inward location of conscience, in which I remain with and for myself [as a moral subject – K.C. et al.], and its content (Hegel 1993, 337). This content can be of various normative qualities – not necessarily equal to the principles most universally accepted, but instead rooted in the beliefs limited to a specific confession, tradition, culture, etc. Indeed, if employees of public institutions such as public health care would like to be governed by the verdict of conscience when their duty from the legal and professional- ethical perspective is to honor the rights of the patient, then the right of morality and the statutory law of right would ‘stand on the same line’ and collide one against another (see also McGraw 2010). There are several solutions in this respect, ranging from the fact that prospective medical professionals will not choose a branch in which they could face conflicts of conscience up to the invalidation of the conscience clause by associations of medical professionals as contrary to the professional medical ethos (e.g., Stahl & Emanuel 2017) which would accelerate the amendment on the law. 2. The Janus Face of Conscience In certain countries, 97 Women’s Reproductive Health Rights in Poland it is regulated in the following way: health providers can exercise their right to conscience clause under the condition that they or their medical facility’s management direct the patient to alternative providers of the relevant health services (e.g., Kane Tiernan 2021). Frequently, this involves urgent, even life-saving health care services. Such a solution results from the physician’s and medical facility’s information obligation toward the patient, who has the right to be informed. Further, there would be also an option to perform as a “double self-identity,” (Orr et al. 2021) which is an illusory solution. 3. Conscience Clause for Pharmacists as One of the Human and Fundamental Rights Two types of conscience clause (conscientious objection, conscientious objector, opt-out choice) were distinguished: 1) Negative: if a health care provider acts illegally (against legal regulations, e.g., against a patient’s consent or rights) (this is a criminal act under article 192 of Criminal Law in Poland); and 2) Positive: if a health care provider refrains from action and follows their conscience (Kubicki 2008). Conscience clauses for medical professionals (including pharmacists) have a strong international normative background. The American Convention on Human Rights (Article 12, “Freedom of onscience and religion”) claims that “Everyone has the right to freedom of conscience and of religion” (compare Puppinck 2017; Mishtal 2009; Major 1992). The European Charter of Fundamental Rights (2000, Article 10, “Freedom of thought, conscience and religion”) claims that “1. Everyone has the right to freedom of thought, conscience and religion. This right includes freedom to change religion or belief and freedom, either alone or in community with others and in public or in private, to manifest religion or belief, in worship, teaching, practice and observance. 2. The right to conscientious objection is recognised, in accordance with the national laws governing the exercise of this right.” This right corresponds to the right in Article 9 of ECHR in accordance with Article 52(3). Certain scholars unambiguously advocate for “protecting right of conscience for pharmacists” which means “protecting this right for everyone” in pluralist and democratic contexts; “if this right can be taken away from pharmacists, it can be taken away from anyone,” Rudd stresses (Rudd 2007, 1904; see also Herbe 2002). The conscience clause for pharmacists is practiced in Canada, USA (e.g., Arizona, Illinois and Wisconsin (Achey & Robertson 2021; Erstad 2019; Bradley 2009; Yoder 2007; Wicclair 2006; Nelson 2005). United Kingdom and Poland belong to the group (Radlińska & Kolwitz 2015; Merks et al. 2015). Thus, the domestic right of medical professionals (including pharmacists) to a conscience clause has a solid foundation in human and fundamental rights. It deserves to be emphasized that not every refusal to sell medicinal products is conditioned on a pharmacist’s conscience clause. 3. Conscience Clause for Pharmacists as One of the Human and Fundamental Rights According to the Polish Pharmaceutical Bill of 2001, § 95, pharmacists may refuse the dispensation of medicinal products for 98 Kinga Ciereszko, Karolina Napiwodzka, Ewa Nowak, Kay Hemmerling various reasons, for example, when the patient’s life or health can be put at risk, when medicinal products can be used for non-medical purposes, when the prescription might be manipulated (not authentic), when the customer is a child under 13 years, etc. Furthermore, a pharmacist is free to follow their personal conscience, therefore, to refuse the provision of contraceptives according to extra-legal normative criteria. Pharmacists’ Code of Ethics of 2012, § 3 is one of them. To date, pharmacists’ conscientious objection has yet to be given a statutory basis. Nevertheless, in 2017 the Ombudsman of the Republic of Poland received a response from the Minister of Health that “the lack of relevant legal provisions regulating the practice of the pharmacy profession is not sufficient premise to conclude that pharmacists cannot claim the ‘conscience clause’” (source: https://bip. brpo.gov.pl/pl/content/minister-zdrowia-farmaceuci-mog%C4%85-stosowa%C4%87- klauzul%C4%99-sumienia). Further, in 2021, BAS (the Polish Research Office) has positively evaluated a draft law that would introduce a conscience clause for pharmacists and owners of drug stores. According to the law elaborated and delivered by the Federation of Polish Catholic Pharmacists, they will have the rights 1) to refuse to sell drugs should they be incompatible with their conscience, and 2) not to order such drugs. In practice, only procreation-related medicines are in contradiction to the conscience clause. Therefore, the law is supposed to restrict access to contraception, thus breaching women’s reproductive rights. (…) The draft law provides one exception – medicaments must be given out when customer’s life or health is threatened. Independent experts, however, argue that the draft law is a legal nonsense (source: https://astra.org.pl/1574- revision-v1/). Thus, the following situation actually comes into play in Poland: there is no explicit legislation on the conscience clause for pharmacists, yet it is not at all uncommon for pharmacists to use the conscience clause. 8% (…) admitted they had refused to fill a prescription due to their beliefs, while in the opinion of 15%, they would exercise the right to conscientious objection if it were legally sanctioned. This difference may imply that about 7% of the participants sometimes dispense medications against their consciences. A relatively high rate 3. Conscience Clause for Pharmacists as One of the Human and Fundamental Rights This finds three extra-legal bases: (1) Pharmacists’ Code of Ethics of 2012, Article 3; (2) the 2017 opinion of the Minister of Health encouraging pharmacists to use the conscience clause in spite of the lack of relevant legislation; (3) Developed in 2014, the Declaration of Faith of Catholic Physicians and Medical Students on the Subject of Human Sexuality and Fertility (to date, several thousand medical professionals and students have voluntarily signed it). This situation raises practical consequences that are troublesome for persons entitled to purchase contraceptives and challenges their legal and human rights with regard to health (more precisely, reproductive and sexual health). Likewise, as in health care (especially in gynecology), directing a patient with a prescription to another pharmacy to dispense prescribed drugs is viewed as complicity in an act reprehensible from the perspective of pharmaceutical conscience. Some pharmacists regard contraception as one of the methods of pregnancy termination, instead of prevention. Piecuch et al. (2014) conducted a research study with N = 126 Polish pharmacists. Of them, 8% (…) admitted they had refused to fill a prescription due to their beliefs, while in the opinion of 15%, they would exercise the right to conscientious objection if it were legally sanctioned. This difference may imply that about 7% of the participants sometimes dispense medications against their consciences. A relatively high rate 99 Women’s Reproductive Health Rights in Poland (17%) of participants who could not clearly state whether they would apply the conscience clause in their job if it were legally sanctioned may be explained in terms of situation-specific reaction. Those participants may have never experienced a direct conflict of conscience but they do not exclude it in the future (Piecuch et al. 2014, 314). (17%) of participants who could not clearly state whether they would apply the conscience clause in their job if it were legally sanctioned may be explained in terms of situation-specific reaction. Those participants may have never experienced a direct conflict of conscience but they do not exclude it in the future (Piecuch et al. 2014, 314). Not only hospitals and clinics, but also pharmacies may rise serious “concerns about the nature and future of Polish democratization” process (Zielińska 2000). Facing this, we decided to conduct a pilot study to examine how it looks from the perspective of female customers and their experiences with the pharmacists’ conscientious objection. 5. Findings A total of 80.17% of the female population in Poland (with higher education diplomas) are purchasers of contraceptives; 19.83% are not. Contraceptives’ purchasers among the most religious participants: 39.67%. 34.0% of the most religious participants are regular purchasers, 23.4% are irregular purchasers, 42.5% are not purchasers or purchase for curative purposes only. 70.25% of all participants purchase contraceptives at local (‘next door’) pharmacies. Most probably, they are familiar with the pharmacists, which may reduce the risk of refusal on the basis of conscientious objection. 8.96% of participants (who have ever tried to purchase contraceptives) faced a refusal based on a pharmacist’s conscience clause. On the question “Should pharmacists be entitled to refuse to sell contraceptives to legal purchasers?” 88.43% answered “No,” while 9.09% answered “Difficult to say.” Only 2.48% marked the answer “Yes.” The participants demonstrated self-awareness concerning their right to reproductive and sexual health as at least potentially colliding with pharmacists’ conscientious objection. As a part of the survey, the participants were presented the alternative of a ‘human’ vs. a ‘robotic’ pharmacist providing contraception. They were invited to a thought experiment participation. 67.77% of the participants welcomed artificial pharmacists, since they cannot raise a personal moral conscientious objection. However, personal preferences regarding contraception purchase and on being clients at robotic pharmacies were lower. Only 39.67% of participants showed their willingness to buy contraceptives for their personal purpose from AI. On the question, “Should an automatic pharmacist also give medical advice and recommendations, for which it is hard to ask a human pharmacist or even a doctor? (e.g., as a printout together with a receipt)?,” the following answers were collected: N=121 % Yes (= unlimited trust) N=23 19.01% Yes, but it is better to additionally ask a pharmacist/doctor for medical recommendations (= limited trust) N=62 51.24% No (= zero trust) N=23 19.01% Difficult to say / N/A N=13 10.74% Figure 1: Participants’ trust concerning a ‘robotic’ pharmacist. y Figure 1: Participants’ trust concerning a ‘robotic’ pharmacist. Furthermore, one-dilemma (1/2 MCT)-based scoring of participants moral competence was experimentally conducted. The more demanding dilemma of the “Doctor” was not used to avoid associations with medical life-and-death contexts, as between October 2020 and present time, abortion, contraception, euthanasia, etc., are constant issues in street protests, ideological battles, and human and women’s rights campaigns in Poland, so a more neutral dilemma was considered to be exceptionally suited for use in our pilot study. 4. Research Design, Method and Sample Description The questionnaire “Pharmacy or a Drug Dispensing Machine?” with 37 items was developed for and addressed only to females, while excluding males and persons who do not identify as female; the sex rubric included options for ‘female’ or ‘‘person who identifies as female’ as eligible for female pharmacological contraception use (Cartwright & Nancarrow 2022). Its subject was the experience of purchasing pharmacological contraceptives intended for females and available only by prescription (excluding any nonpharmacological means, e.g., sterilization, condoms, etc.) by female purchasers. Specifically, the research was designed to examine how many female subjects entitled to purchase female contraceptives confronted a pharmacist’s conscience-related obstacles when trying to purchase contraceptives. This characterization of the study group gives the study the qualification of women studies (not gender studies). Further, ½ of the Moral Competence Test developed by Georg Lind (the Workers’ dilemma only; Polish MCT version was validated by Nowak et al. and certified by G. Lind in 2009) with 6 proarguments and 6 counterarguments to rank (on the Likert scale -4 to 4) was experimentally included. The questionnaire was uploaded to the e-platform Survio.pl. A pilot study with adult female participants was conducted on February 2021. The entire data set was additionally peer- reviewed and processed according to Lind’s algorithm. The study carried out was pilot and exploratory in character. A final total N=121 women randomly completed the survey. Only female participants were addressed, as usually it is females that are engaged in contraception purchase and are facing restrictions of their rights; 98.35% were females and 1.65% identified themselves as females; aged from 21 to 62. Educational background: 86.78% of participants with a university diploma; 9.09% of university students; 4.13% of participants with a high school leaving certificate. Demographic description: only Polish nationality, 96.6% of them being residents of Poland, 3.4% of Poland and third countries. Residential locality size: 82.3% living in cities > 100,000 inhabitants; 17.7% in smaller localities. 100 Kinga Ciereszko, Karolina Napiwodzka, Ewa Nowak, Kay Hemmerling 5. Findings Its validity according to moral orientations’ criterion in Kohlberg’s and 101 Women’s Reproductive Health Rights in Poland Lind’s sense was confirmed. Though we should note that the moral competence score “C” must be a mean value for both dilemmas, one of them more, and one less challenging test participants’ moral cognitive processes and their emotions. Dealing with moral dilemmas results in emotional and cognitive dissonance, and dealing with the latter might be demanding for subjects. Lind’s sense was confirmed. Though we should note that the moral competence score “C” must be a mean value for both dilemmas, one of them more, and one less challenging test participants’ moral cognitive processes and their emotions. Dealing with moral dilemmas results in emotional and cognitive dissonance, and dealing with the latter might be demanding for subjects. Figure 2: Moral orientations according to Kohlberg. Orientation 6 corresponds to ‘principled conscience’ and is represented in MCT by four arguments. Participants tend to rate their normative quality superior. Figure 2: Moral orientations according to Kohlberg. Orientation 6 corresponds to ‘principled conscience’ and is represented in MCT by four arguments. Participants tend to rate their normative quality superior. Moral competence is defined as a personal ability to make judgments, decisions, or arrive at solutions to a problem (e.g., conflict, dilemma) based on self-chosen and internally prioritized moral principles, and to act accordingly. The C-score may range between 0-100 points (Lind 2019). Below we present the correlations between participants’ C-scores and their behaviors as 1) regular, 2) irregular, and 3) therapeutical purchasers and users of contraception. Interestingly, regular purchasers’ C-scores are lower than the C-scores of irregular and therapeutical purchasers. The differences are partially significant (not lower than 8 C-points). However, these are approximate and incomplete values, as the full “C” (for two dilemmas) would hypothetically be lower than the “C” measured for the “Workers’ dilemma” in the present pilot study. 102 Kinga Ciereszko, Karolina Napiwodzka, Ewa Nowak, Kay Hemmerling Figure 3: Buying and using contraceptives in correlation with participants’ C-score. Figure 3: Buying and using contraceptives in correlation with participants’ C-score. Further, participants with higher ‘half’ C-scores are more likely to mistrust (or are undecided), while those with lower C-scores are more likely to trust in artificial pharmacists as competent advisers in reproductive and sexual health related issues. The difference between “Fully trusting” and “Undecided” participants in terms of their “C” (for moral competence) was significant. 5. Findings It suggests that participants with higher moral competence are less likely to risk their health due to robotic devices replacing medical and pharmacy professionals. Figure 4: Participants’ trust in robotic pharmacy in correlation with C-competence. Figure 4: Participants’ trust in robotic pharmacy in correlation with C-competence. Additionally, we present C-scores measured in 2020 in Polish healthcare students (Nowak et al. 2021) for both of dilemmas, with the same instrument (Moral Competence Test). The following table shows that healthcare students achieve sufficient “C” (higher than the minimum 20 C-points recommended by G. Lind) to manage typical sociomoral conflicts and normative controversies. However, conflicts between high-type orientations such, as principles, might be still challenging when subjects have to deal with conscientious objection. 103 Women’s Reproductive Health Rights in Poland N Main C-score Minimum C-score Maximum C-score Standard deviation Workers‘ dilemma C-score 115 44,61516 4,5455 91,95876 23,62329 Doctor’s dilemma C-score 115 40,93825 19,3182 92,08791 26,76602 Figure 5. C-scores of Polish healthcare students (Nowak et al. 2021). Conclusions and Discussion For instance, the immediate postsocialist period between 1989 and 1993, strongly influenced by Catholic nationalism, brought critical transformations in policies regulating reproductive and sexual health and rights (…) Of particular importance in regulating reproductive and sexual conduct was the implementation of the Conscience Clause law. The clause, written into the postsocialist Medical Code of Ethics in 1991, provides that ‘a physician can withhold health services which are not in agreement with his conscience’, but must make a referral elsewhere where there are ‘realistic possibilities of obtaining such a healthcare’ (Mishtal 2009, 163). the immediate postsocialist period between 1989 and 1993, strongly influenced by Catholic nationalism, brought critical transformations in policies regulating reproductive and sexual health and rights (…) Of particular importance in regulating reproductive and sexual conduct was the implementation of the Conscience Clause law. The clause, written into the postsocialist Medical Code of Ethics in 1991, provides that ‘a physician can withhold health services which are not in agreement with his conscience’, but must make a referral elsewhere where there are ‘realistic possibilities of obtaining such a healthcare’ (Mishtal 2009, 163). Thirty years later, women have difficulty not only in exercising their right to abortions in cases permitted by law (Mishtal 2009, 171), but also in exercising their right to obtain legal contraceptives, both due to the health providers’ conscience clause and because there is a lack of respect for their conscience and their rights. Twenty years after Mishtal’s research, pharmacists and gynecologists continue to declare, “here in our clinic we absolutely do not condone contraception, because those are the rules of how we provide care,” and so they follow their particular “contraceptive conscience,” as one of Mishtal’s interviewees admitted (Mishtal 2009, 173). Democracy, as a set of public institutions and procedures, such as deliberative and discursive, is a permanent project everywhere. In Eastern Europe, the grassroots, axiological, and normative self-constitutionalization of societies after 1989 occurs not only in the public spheres but also much more profoundly. It might be an antagonistic process in all political cultures, including mature democracies, when individuals, professionals, or citizens contest values they do not prioritize for themselves (Bozeman 2018; Ogorevc et al. 2019). However, axiological particularities and polarities should not discredit public institutions and human and fundamental rights, including those concerning health and health care. Conclusions and Discussion The presented findings demonstrate that contraception purchasers and users are usually aware of the normative conflicts between such high-type, general, human rights- labelled principles as health and pharmacy professionals’ right to the conscientious objection on the one hand and, on the other hand, a series of reproductive and sexual health-related values and human rights. The findings for the Polish sample (N=121) examined in this paper reflect the complex normative situation of a society seeking a balance and consensus between traditional, e.g., predominantly Christian values with their claims to universal validity and the plurality of values and principles. Persons and professionals unable to recognize and accept this plurality (which might be demanding due to the moral competence level, regardless of particular moral affiliations, opinions, and attitudes) are prone to ‘unresolvable’ conflicts and may potentially violate fellow citizens’ human and legal rights – as contraception in Poland belongs to the legal medical services. Our study demonstrated that a quiet high moral competence (though experimentally scored for one dilemma only) may correlate with various and even dichotomous preferences on contraception use among its female purchasers in Poland. In contrast, low moral competence would produce a normative trap for all citizens, including health and pharmacy professionals, patients, their relatives, and society as a whole. If pharmacists (and, generally, all health professions) are eligible to prioritize their moral conscience clause over their fellow citizens’ human and legal rights, and yet were unable to find a balance, and to respect fellow citizens’ rights and values (without resigning from conscience protection), this could permanently stir up conflicts and painful dilemmas between conscience clause, obligations to patients, and patient rights (Drozd 2013; Flynn 2008). This also would increase discomfort in society, due to violation of the legal, human, and fundamental rights of purchasers who need contraceptives for health-related purposes; would sharpen social inequalities, violate democratic difference and principle pluralism, and postpone public discourse or deliberation, which are needed to elaborate reasonable consensus or reasonable dissensus on choice-friendly laws. Enabling and empowering future health and pharmacy professionals to provide patients with medicinal products in just and fair way (Yoder 2007) beyond particularisms and biases would be one of the priorities of medical education. Public axiologies are dynamic spheres, especially in societies in transition from 104 Kinga Ciereszko, Karolina Napiwodzka, Ewa Nowak, Kay Hemmerling nondemocratic to democratic constitutions. Polish society has been in democratic transition since 1989, with various effects. Conclusions and Discussion The latter show increasing complexity and specificity in line with advances in health sciences, including pharmacology, and, finally, due to the progressive recognition of autonomous self-determination and rights of groups previously controlled by other groups and power centers (Gyrd-Hansen 2004). The European Parliament resolution on the 25th anniversary of the International Conference on Population and Development (ICPD25) (Nairobi Summit) was voted on June 25, 2021. Among others, the resolution shall urge the EU member states to improve women’s “control over their bodies, their health, and their fertility,” their ability „to define their role in society if sustainable growth and development are to follow such steep population growth,” and their access to “comprehensive reproductive healthcare,” “contraception and emerging contraceptive methods” (European Parliament 2021). Arguably, the progress in medicine seems to be ahead of the normative development of societies, especially those familiar with more traditional values: to this day, one can still hear public complaints voiced by religious authorities about the availability of oral contraception, in vitro fertilization technology, prenatal and preimplantation diagnostics discovered decades ago. 105 Women’s Reproductive Health Rights in Poland In the face of the differentiation and antagonization of priorities and values, the “legitimation crisis” (Habermas 1988) is also intensifying, which means that it is not the values but the principles that allow the balance of different, sometimes opposite values in parallel. Recognizing such a balance would be the first deal essential for democratization. However, ensuring a multitude of values through general principles is not enough. Improving institutional (instead of particular) “public value governance” (Huijbregts et al. 2022) and fostering professionals’ ability to manage this plurality in everyday practice of common and public institutions such as public health care would be the second deal that is essential for democratization. This ability corresponds to a well-developed moral competence (e.g., Lind 2012) and “principled judgment” as its core performance. References Achey T. S. & Robertson A. T. 2021. “Conscientious Objection: A Review of State Pharmacy Laws and Regulations,” Hospital Pharmacy 57(2):268–272. https://doi. org/10.1177/00185787211024217 American Medical Association 2006. „Broader Conscience Clauses Would Expand Physicians’ Right of Refusal.” URL: www.amaassn.org/amednews/2006/03/06/ prl20306.htm (retrieved 1 March 2021). Bozeman B. 2018. “Public Values: Citizens’ Perspective,” Public Management Review 21(6):817–838. https://doi.org/10.1080/14719037.2018.1529878 Bradley C. T. 2009. “Emergency Contraception and Physicians’ Rights of Conscience: A Review of Current Legal Standards in Wisconsin,” WMJ 108(3):156–160. PMID: 19552354. Carr B., Gates M. F., Mitchell A., & Shah R. 2012. “Giving Women the Power to Plan Their Families,” The Lancet 380:80–82. https://doi.org/10.1016/S0140- 6736(12)60905-2 Cartwright T. & Nancarrow C. 2022. „A Question of Gender: Gender Classification in International Research,” International Journal of Market Research 64(5):575–593. https://doi.org/10.1177/14707853221108663 Curlin F. A. et al. 2004. ”Religion, Conscience, and Controversial Clinical Practices,” The New England Journal of Medicine 356:593–600. Drozd M. 2013. “Right of the Pharmacist to Conscientious Objection in Light of the Applicable of Legal Regulations,” Law, Constitution, Jurisprudence 16:267–280. Erstad B. L. 2019. “The Conscience of a Pharmacist,” American Journal of Pharmaceutical Education 83(2):7301. https://doi.org/10.5688/ajpe7301 Flynn D. 2008. “Pharmacist Conscience Clauses and Access to Oral Contraceptives,” Journal of Medical Ethics 34(7):517–520. 106 Kinga Ciereszko, Karolina Napiwodzka, Ewa Nowak, Kay Hemmerling Field H. M. 2000. “Increasing Access to Emergency Contraceptive Pills Through State 27 Law Enabled Dependent Pharmacist Prescribers,” UCLA Women’s Law Journal 11(1):141–253. Gyrd-Hansen D. 2004. “Investigating the Social Value of Health Changes,” Journal of Health Economics 23(6):1101–1116. https://doi.org/10.1016/j. jhealeco.2004.02.002 Habermas J. 2003. Truth and Justification, trans. B. Fultner. Cambridge, Mass.: The MIT Press. Habermas J. 1988. Legitimation Crisis, trans. T. McCarthy. Cambridge, UK: Polity Press. ̈ Hegel G. W. F. 2009. Grundlinien der Philosophie des Rechts, hg. von K. Grotsch & E. Weisser-Lohmann, GW, Bd. 14,1. Copy Rights by Nordrhein-Westfälische Akademie der Wissenschaften und der Künste, Düsseldorf. Hamburg: Felix Meiner. Hegel G. W. F. 1993. Der Begriff der Religion, in idem, Vorlesungen über die Philosophie der Religion. Einleitung. Der Begriff der Religion, hg. von W. Jaeschke, Hamburg: Felix Meiner. Herbe D. W. 2002. ”The Right to Refuse: A Call for Adequate Protection of a Pharmacist’s Right to Refuse Facilitation of Abortion and Emergency Contraception,” Journal of Law and Health 17(1):76–102. Huijbregts R., George B., & Bekkers V. 2022. “Public Values Assessment as a Practice: Integration of Evidence and Research Agenda,” Public Management Review 24(6):840–859. https://doi.org/10.1080/14719037.2020.1867227 Kickbusch I. 2004. References ”From Charity to Rights: Proposal for Five Action Areas of Global Health,” Journal of Epidemiol Community Health 58:630–631. http://dx.doi. org/10.1136/jech.2004.021121 Kubicki L. 2008. „Sumienie lekarza jako kategoria prawna,” Prawo i Medycyna 4(1):5– Lind G. 2000. “Moral Regression in Medical University Students and Their Learning Environment,” Revista Brasileira de Educação Medica 24(3):24–33. Lind G. 2005. “The Cross-Cultural Validity of the Moral Judgment Test: Findings from 28 Cross-cultural Studies,” paper presented at the APA Congress, August 18-21, 2005. Washington, D.C. Lind G. et al. 2010. “Testing for Moral Judgment Competence,” in G. Lind et al. (Eds.), Moral Judgment and Social Education (pp. 79–105). Rutgers, NJ: Transaction. Lind G. 2012. “Moral Competence and Democratic Ways of Life,” in W. G. Weber et al. (Eds.), Democratic Competences and Social Practices (pp. 62–85). Springer VS. (Eds.), Democratic Competences and Social Practices (pp. 62–85). Springer VS. Lind G. 2019. How to Teach Moral Competence. New: Discussion Theater. Berlin: Logos. Lind G. 2019. How to Teach Moral Competence. New: Discussion Theater. Berlin: Logos. Lind G. 2019. How to Teach Moral Competence. New: Discussion Theater. Berlin: Logos. Luhmann N. 1965. „Die Gewissensfreiheit und das Gewissen,“ Archiv des Öffentlichen Rechts 90(3):257–286. Major M. F. 1992. “Conscientious Objection and International Law: A Human Right?,” Case Western Reserve Journal of International Law 24(2):349–378. 107 Women’s Reproductive Health Rights in Poland McGraw B. 2010. “Public Reason and Religious Conflict,” in Faith in Politics: Religion and Liberal Democracy (pp. 126–175). Cambridge University Press. McGraw B. 2010. “Public Reason and Religious Conflict,” in Faith in Politics: Religion and Liberal Democracy (pp. 126–175). Cambridge University Press. McGraw B. 2010. “Public Reason and Religious Conflict,” in Faith in Politics: Religion and Liberal Democracy (pp. 126–175). Cambridge University Press. McGraw B. 2010. “Public Reason and Religious Conflict,” in Faith in Politics: Religion and Liberal Democracy (pp. 126–175). Cambridge University Press. Merks P. et al. 2015. „Conscience Clause for Pharmacists from the Perspective of Pharmacists Working in Poland and in the UK,” Farmacja Polska 71(8):2–9. Mishtal J. 2009. “Matters of ‘Conscience’,” Medical Anthropology Quarterly 23(2):161– 183. https://doi.org/10.1111/j.1548-1387.2009.01053.x Nelson J. 2005. „Freedom of Choice for Everyone: The Need for Conscience Clause Legislation for Pharmacists,” University of St Thomas Law Journal 3(1), #8. Nowak E., Barciszewska A.-M., Lind G., Hemmerling K., & Kukolja-Taradi S. 2020. „What Would Hippocrates Say? References Improving Moral Competence in Prospective Health Care Providers 1985-2020,” paper presented at the Annual Meeting “The Power and Possibilities for the Public Good When Researchers and Organizational Stakeholders Collaborate,” American Educational Research Association (AERA), San Francisco, April 8-12, 2021. Ogorevc M., Murovec N. et al. 2019. “Questioning the Differences between General Public vs. Patient Based Preferences Towards EQ-5D-5L defined Hypothetical Health States,” Health Policy 123(2):166–172. https://doi.org/10.1016/j. healthpol.2017.03.011 Orr Z. et al. 2021. „The Challenges and Dilemmas of Local Translators of Human Rights: The Case of Disability Rights among Jewish Ultra-Orthodox Communities,“ Journal of Human Rights 20(3):339–355. https://doi.org/10.1080/14754835.2020.18682 94 Pellegrino E. D. 2002. “A Physician’s Conscience, Conscience Clauses, and Religious Belief: A Catholic Perspective,” Fordham Urban Law Journal 30(1):221–244. URL: https://ir.lawnet.fordham.edu/ulj/vol30/iss1/13 Piecuch A., Gryka M., Kozłowska-Wojciechowska M. 2014. “Attitudes towards Conscientious Objection among Community Pharmacists in Poland,” International Journal of Clinical Pharmacy 36(2):310–315. https://doi.org/10.1007/s11096- 013-9878-5 Polis C. B., Bradley S. E. K., Bankolea A. et al. 2016. “Typical-Use Contraceptive Failure Rates in 43 Countries with Demographic and Health Survey Data: Summary of a Detailed Report,” Contraception 94(1):11–17. https://doi.org/10.1016/j. contraception.2016.03.011 Puppinck G. 2017. “Conscientious Objection and Human Rights. A Systematic Analysis,” Law and Religion 1(1):1–75. https://doi.org/10.1163/24682993-12340001 Rudolf E. I. 2016. “Zdrowie reprodukcyjne kobiet – rozważania terminologiczne,” in A. Szlagowska (Ed.), Problemy zdrowia reprodukcyjnego kobiet. Wstęp do badań (pp. 9–26). Quaestio/Medical University of Wrocław. 108 Kinga Ciereszko, Karolina Napiwodzka, Ewa Nowak, Kay Hemmerling Soo Hyun Lim, Hae In Jang, Dong-Yun Lee et al. 2016. “Recent Trends in Contraceptive Use Among Korean Adolescents: Results from a Nationwide Survey from Year 2013 to 2015,” Obstetrics and Gynecology Science 59(6):519–524. https://doi. org/10.5468/ogs.2016.59.6.519 Stahl R. Y. & Emanuel E. J. 2017. “Physicians, Not Conscripts – Conscientious Objection in Health Care,” The New England Journal of Medicine 376:1380–1385. https://doi. org/10.1056/NEJMsb1612472 Tiernan B. K. 2021. “Reevaluating Conscience Clauses,” The Journal of Medicine and Philosophy: A Forum for Bioethics and Philosophy of Medicine 46(3):297–312. https://doi.org/10.1093/jmp/jhab005 Wicclair M. R. 2006. „Pharmacies, Pharmacist and Conscientious Objection,” Kennedy Institute Ethics Journal 16(3):225–250. Wilkinson D., Truog R. & Savulescu J. 2016. “In Favour of Medical Dissensus: Why We Should Agree to Disagree About End-of-Life Decisions,” Bioethics 2:109–118. PMCID: PMC4864446. https://doi.org/10.1111/bioe.12162 Yoder J. D. 2007. „Pharmacists’ Right of Conscience: Strategies for Showing Respect for Pharmacists’ Beliefs While Maintaining Adequate Care for Patients,” Valparaiso University Law Review 41(2):975–1025. Zielińska E. 2000. ”Between Ideology, Politics, and Common Sense: The Discourse of Reproductive Rights in Poland,” in S. References Gal & G. Klingman (Eds.), Reproducing Gender: Politics, Publics, and Everyday Life after Socialism (pp. 23–57). Princeton: Princeton University Press. Żuradzki T. 2016. ”Uzasadnienie sprzeciwu sumienia: lekarze, poborowi, żołnierze,” Diametros 47:98–128. https://doi.org/10.13153/diam.47.2016.871 109
https://openalex.org/W2108569176	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0103268&type=printable	English	null	Forest Structure in Low-Diversity Tropical Forests: A Study of Hawaiian Wet and Dry Forests	PloS one	2,014	cc-by	12,907	Rebecca Ostertag1, Faith Inman-Narahari2, Susan Cordell3, Christian P. Giardina 1 Department of Biology, University of Hawai‘i at Hilo, Hilo, Hawai‘i, United States of America, 2 Department of Natural Resources and Environmental Management University of Hawai‘i at Ma¯noa, Honolulu, Hawai‘i, United States of America, 3 Institute of Pacific Islands Forestry, Pacific Southwest Research Station, USDA Forest Service Hilo, Hawai‘i, United States of America, 4 Department of Ecology and Evolutionary Biology, University of California Los Angeles, Los Angeles, California, United States o America Abstract This is an open-access article distributed under the terms of the Creative Comm unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: The major funding for this research came from National Science Foundation’s EPSCoR Grants No. 0554657 and No. 0903833 to the University of Hawai ‘i. Major in-kind support was provided by the Pacific Southwest Research Station of the United States Forest Service. Logistical or financial support that was supplemental was provided for by the Smithsonian Tropical Research Institute Center for Tropical Forest Science, and the University of California, Los Angeles. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Email: ostertag@hawaii.edu Abstract The potential influence of diversity on ecosystem structure and function remains a topic of significant debate, especially for tropical forests where diversity can range widely. We used Center for Tropical Forest Science (CTFS) methodology to establish forest dynamics plots in montane wet forest and lowland dry forest on Hawai‘i Island. We compared the species diversity, tree density, basal area, biomass, and size class distributions between the two forest types. We then examined these variables across tropical forests within the CTFS network. Consistent with other island forests, the Hawai‘i forests were characterized by low species richness and very high relative dominance. The two Hawai‘i forests were floristically distinct, yet similar in species richness (15 vs. 21 species) and stem density (3078 vs. 3486/ha). While these forests were selected for their low invasive species cover relative to surrounding forests, both forests averaged 5–.50% invasive species cover; ongoing removal will be necessary to reduce or prevent competitive impacts, especially from woody species. The montane wet forest had much larger trees, resulting in eightfold higher basal area and above-ground biomass. Across the CTFS network, the Hawaiian montane wet forest was similar to other tropical forests with respect to diameter distributions, density, and aboveground biomass, while the Hawai‘i lowland dry forest was similar in density to tropical forests with much higher diversity. These findings suggest that forest structural variables can be similar across tropical forests independently of species richness. The inclusion of low-diversity Pacific Island forests in the CTFS network provides an ,80-fold range in species richness (15–1182 species), six-fold variation in mean annual rainfall (835–5272 mm yr21) and 1.8-fold variation in mean annual temperature (16.0–28.4uC). Thus, the Hawaiian forest plots expand the global forest plot network to enable testing of ecological theory for links among species diversity, environmental variation and ecosystem function. Citation: Ostertag R, Inman-Narahari F, Cordell S, Giardina CP, Sack L (2014) Forest Structure in Low-Diversity Tropical Forests: A Study of Hawaiian Wet and Dry Forests. PLoS ONE 9(8): e103268. doi:10.1371/journal.pone.0103268 Editor: Bruno He´rault, Cirad, France Received January 18, 2014; Accepted June 30, 2014; Published August 27, 2014 Copyright: 2014 Ostertag et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright: 2014 Ostertag et al. Introduction evenness—varies with species diversity is itself understudied, likely an effect of the paucity of studies of the structure of low-diversity tropical forests. Some have hypothesized that forest structure and species-richness might be related, if structure acts as a habitat scaffold or template that precedes and enables species assembly and diversity by providing an increased variety of habitat niches (e.g., nurse logs for seedlings, perches for birds that disperse seeds, climbing structures for vines [17,18]. Alternatively, higher diversity may enhance forest structure, if more species correspond to a wider variety of size classes, strata, and crown architectures [17]. Both processes are not mutually exclusive and may operate simultaneously, creating a positive feedback cycle that would enhance diversity and influence various forest structural attributes. Recent efforts have examined some structural variables, such as latitudinal trends in height across forests e.g., [19,20] and the effects of diversity and spatial scale on standing forest biomass [12], but very low-diversity tropical forests were not considered in these analyses. The tropical forests in the Hawaiian Islands represent a low-diversity extreme, as a result of its young High species richness is a hallmark of many tropical forests [1,2]. Indeed, the latitudinal gradient and equatorial peak in plant diversity has attracted attention for centuries e.g., [1,3,4,5]. Numerous studies have focused on the causes of high diversity in tropical forests [1,6,7,8,9,10], and theories have been formulated to explain how species or functional diversity in turn affects ecosystem function [11,12]. However, these linkages have rarely been tested, and not all tropical forests are diverse. For example, legume-dominated swamp forests, peat forests, pine savannas, and oceanic islands that are geographically isolated can have low to very low diversity [13,14,15,16]. Such low-diversity forests are understudied, and there is no clear answer to the simple question of whether the structure of a low-diversity tropical forest would be expected to be similar to or different from that of a high-diversity tropical forest with comparable climate. Indeed, the question of how forest structure—i.e., physiognomy, basal area, density, diameter size class distributions, biomass, and 1 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e103268 August 2014 \| Volume 9 \| Issue 8 \| e103268 Low-Diversity Hawaiian Forests geological origins [21] and extreme isolation from continental land masses: at approximately 4000 km from the nearest continent, Hawai‘i is the world’s most isolated archipelago. Introduction The resulting native flora in Hawai‘i is disharmonic (i.e., missing many functional groups) and is about 90% endemic [22]. While long- term plot-based ecological measurements across the tropics have focused on high-diversity forests, there have been surprisingly few data from low-diversity tropical forests [23,24,25,26]. Such low- diversity forests present many interesting contrasts to other tropical forests, and within Hawai‘i they also fall across striking environ- mental gradients (Table 1). but we used this study design, to test a prediction based on the previous literature that Hawaiian dry forest would have greater stem density, lower diversity, and smaller diameter trees than wet forest [32]. To place our findings in a broader context, we also asked: 2) Can the extremely low forests of Hawai‘i have similar structural attributes to more diverse tropical forests? To examine this question, we compared Hawaiian forests with others in the CTFS network enabling the comparison of forest structural variables across a range of environments and diversity levels [1,28,31,33,34,35]. If Hawaiian forests converge with other tropical forests, the importance of climate in determining forest structure is highlighted. The aim of this study was to: 1) characterize and compare two extremely low-diversity Hawaiian forests, montane wet and lowland dry forest, and 2) compare the structural attributes of these two forests to more diverse tropical forests within the Center for Tropical Forest Science (CTFS) permanent plot network. Including the Hawaiian plots as part of a cross-plot analysis allows, for the first time, examination of forest structure along a diversity gradient that varies almost 80-fold across large-scale plots with consistent measurement protocols. Study Sites Environmental Conditions Large variation in elevation, rainfall, temperature and soils among forests that are geographically close 1–2 High light levels in intact wet, mesic, and dry forest (1.9–40% diffuse light transmission) 3–9 Species Composition and Diversity Patterns A global biodiversity hotspot due to high endemism and number of endangered species 10–11 Same species distributed in many habitats differing in environmental conditions, demonstrating exceptional phenotypic plasticity 10–14 Tree ferns common and often the understory dominant in wet forests at all elevations, whereas outside of Hawai‘i they tend to be more restricted 10 Monodominance by a few canopy species 15 Autecology of Plant Species Metrosideros polymorpha dominant in wet forests throughout succession (as pioneer and late successional species) 15–17 Extremely slow growth of primary pioneer species, M. polymorpha (1–2 mm/year diameter) 18–20 Nurse logs serve as a substrate for seedling regeneration 21 Dieback and regeneration of canopy dominant M. polymorpha in cohorts contribute strongly to gap dynamics 16, 22 Trophic Interactions Evolution without land mammals 23, 24 Documented extinctions of plants, pollinators and dispersers may influence present day evenness and rarity measures 23 Animal dispersal of seeds conducted entirely by birds before human contact 24, 25 Apparently low rates of insect herbivory 26 and seed predation 27 Presence of invasive weeds, ungulates, and birds may alter present-day plant-animal interactions 25–26, 28 Superscripts refer to references listed in Table S5 in File S2. doi:10.1371/journal.pone.0103268.t001 Table 1. Distinctive structural and demographic features of Hawaiian forests. Environmental Conditions Large variation in elevation, rainfall, temperature and soils among forests that are geographically close 1–2 High light levels in intact wet, mesic, and dry forest (1.9–40% diffuse light transmission) 3–9 Species Composition and Diversity Patterns A global biodiversity hotspot due to high endemism and number of endangered species 10–11 Same species distributed in many habitats differing in environmental conditions, demonstrating exceptional phenotypic plasticity 10–14 Tree ferns common and often the understory dominant in wet forests at all elevations, whereas outside of Hawai‘i they tend to be more restricted 10 Monodominance by a few canopy species 15 Autecology of Plant Species Metrosideros polymorpha dominant in wet forests throughout succession (as pioneer and late successional species) 15–17 Extremely slow growth of primary pioneer species, M. polymorpha (1–2 mm/year diameter) 18–20 Nurse logs serve as a substrate for seedling regeneration 21 Dieback and regeneration of canopy dominant M. Study Sites polymorpha in cohorts contribute strongly to gap dynamics 16, 22 Trophic Interactions Evolution without land mammals 23, 24 Documented extinctions of plants, pollinators and dispersers may influence present day evenness and rarity measures 23 Animal dispersal of seeds conducted entirely by birds before human contact 24, 25 Apparently low rates of insect herbivory 26 and seed predation 27 Presence of invasive weeds, ungulates, and birds may alter present-day plant-animal interactions 25–26, 28 Superscripts refer to references listed in Table S5 in File S2. doi:10.1371/journal.pone.0103268.t001 PLOS ONE \| www.plosone.org 2 August 2014 \| Volume 9 \| Issue 8 \| e103268 Table 1. Distinctive structural and demographic features of Hawaiian forests. p p doi:10.1371/journal.pone.0103268.t001 Superscripts refer to references listed in Table S5 in File S2. Study Sites In 2008 and 2009, we established two forest dynamics plots (FDPs) on Hawai‘i Island – one within montane wet forest (MWF) and one within lowland dry forest (LDF), to initiate the Hawai‘i Permanent Plot Network (HIPPNET; Fig. 1). We focused our study on Hawai‘i Island, because it has the greatest area of intact forests, a complete map of lava flow ages, and excellent infrastructure for ecological studies. As the youngest island in the archipelago (,700,000 years), it has had the least time for plant colonization and subsequent speciation, and thus has lower species richness relative to its size than the older islands [36]. We selected areas in excellent ecological condition that are representative of a given forest type, with high native species cover, and a commitment by ownership to long-term conservation objectives. Notably, all forests in Hawai‘i are affected to some degree by altered trophic interactions due to invasion of non-native species or extinction of the native species [37], but this is not unique to Hawai‘i [38]. Non-native stems that were encountered were measured for percent cover, and then controlled mechanically or We used the initial census of large-scale permanent plots in Hawai‘i to examine structural and floristic characteristics of two forests that are geographically close but located in widely contrasting environments. The two Hawaiian forest types exam- ined in this first census were montane wet forest (MWF) and lowland dry forest (LDF). Many studies have shown that forests established in areas with higher rainfall or temperature have higher diversity [1,27,28], and also greater basal area, tree height, and above-ground biomass [28,29,30,31]. Further, forests in higher rainfall areas tend to have a greater representation of larger trees, but lower tree densities [32]. We therefore ask: 1) How do the two Hawaiian forests compare in terms floristic and life form composition, stand structure, species diversity, and non- native species cover? Our study was not designed to specifically examine the effects of climate on forest structure and composition, Table 1. Distinctive structural and demographic features of Hawaiian forests. p p doi:10.1371/journal.pone.0103268.t001 Autecology of Plant Species Presence of invasive weeds, ungulates, and birds may alter present-day plant-animal interactions 25–26, 28 Presence of invasive weeds, ungulates, and birds may alter present-day plant-animal interactions 25–26, 28 August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org 2 Contour map of the two 4-ha forest plots on Hawai‘i Island. Pa¯lamanui site in west Hawai‘i is lowland dry forest (LDF; left panel he dominant canopy tree Diospyros sandwicensis and the open canopy and understory structure of small trees and shrubs); Laupa¯hoehoe t Hawai‘i is montane wet forest (MWF; right panel showing Metrosideros polymorpha tree and Cibotium spp. tree fern understory). 1/journal.pone.0103268.g001 Low-Diversity Hawaiian Forests Low-Diversity Hawaiian Forests Figure 1. Contour map of the two 4-ha forest plots on Hawai‘i Island. Pa¯lamanui site in west Hawai‘i is lowland dry forest (LDF; left panel showing the dominant canopy tree Diospyros sandwicensis and the open canopy and understory structure of small trees and shrubs); Laupa¯hoehoe plot in east Hawai‘i is montane wet forest (MWF; right panel showing Metrosideros polymorpha tree and Cibotium spp. tree fern understory). doi:10.1371/journal.pone.0103268.g001 chemically (see ‘‘Plot Establishment and Vegetation Measure- ments’’ below) and were not considered in the census of stems. classified as hydrous, ferrihydritic, isothermic Acrudoxic Hydru- dands (websoilsurvey.nrcs.usda.gov). Rainfall at the MWF is dominated by tradewind-driven precipitation [40]. Interpolated mean annual precipitation, based on analysis of climate station data over 30 years, is 3440 mm with no distinct dry season [41] and mean annual air temperature is 16uC [42]. The forest consists of evergreen broad-leaved trees, and the ,25–28 m canopy is dominated by Metrosideros polymorpha (Myrtaceae; Fig. 1) and to a lesser extent, Acacia koa (Fabaceae). Vegetation at the MWF is highly representative of this forest type in Hawai‘i [43] (see references in Table 1). Montane wet forest (MWF). The 4-ha Laupa¯hoehoe FDP (19u55’ N, 155u17’ W) is located within the state-owned Laupa¯hoehoe Natural Area Reserve section of the Hawai‘i Experimental Tropical Forest (HETF) on the northeast slope of Mauna Kea volcano. Permits were obtained for work in the HETF through the Institute of Pacific Islands Forestry and the Hawai‘i Division of Forestry and Wildlife/Department of Land and Natural Resources. The mean elevation of the plot is 1120 m.a.s.l. with slopes of 0–20%, and the overall direction of downslope is northwards towards the Pacific Ocean. The substrate within the plot is 4000-14,000 years old [39]. Autecology of Plant Species Soils were formed from weathered volcanic material, and are deep, rocky, and moderately well-drained silty clay loam in the Akaka series, and The dominant pre-human contact disturbance regime in this forest type was single-to multiple-tree falls, with the maximum gap size averaging 21.5 m2 [44]. Larger openings coincide with dieback due to cohort senescence of older M. polymorpha stands August 2014 \| Volume 9 \| Issue 8 \| e103268 August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org 3 Low-Diversity Hawaiian Forests 25%, 3: 25–50%, 4: 50–100%. Non-native trees with stems $ 1 cm at 130 cm were individually mapped. The DBH of the largest stem of non-native trees ,5 cm was estimated to the nearest centimeter and measured to the nearest centimeter if . 5 cm. For trees with multiple stems, we counted the total number of stems $1 cm at 130 cm. After the non-native trees were mapped, they were girdled and sprayed with herbicide. We did not spray herbicide on the grasses in the LDF, nor the vine Passiflora tarminiana in the MWF. [45]. Following contact, large A. koa trees were occasionally harvested for traditional canoe building. In modern times, limited A. koa logging occurred in the HETF but was restricted to , 100 m of an unimproved road that traverses areas. There is no evidence of logging within the MWF [46], which .500 m from the road. Non-native wild pigs disturb soils while rooting, as well as tree ferns [47], with damage over a large area. [45]. Following contact, large A. koa trees were occasionally harvested for traditional canoe building. In modern times, limited A. koa logging occurred in the HETF but was restricted to , 100 m of an unimproved road that traverses areas. There is no evidence of logging within the MWF [46], which .500 m from the road. Non-native wild pigs disturb soils while rooting, as well as tree ferns [47], with damage over a large area. Lowland dry forest (LDF). The 4-ha Pa¯lamanui FDP is an example of one of the world’s most endangered forest types, and is located on a privately-owned tract of dry forest on the northwest slope of Huala¯lai Volcano in the district of North Kona (240 m elevation, 19u44’ N, 155u59’ W). SI~ number of species shared in both sites (no: of species in MWFzno: of species in LDF) Autecology of Plant Species A memorandum of understand- ing was established with the land owners and managers, the Pala¯mani Group, for permission to conduct research in the lowland dry forest site. The mean elevation of the plot is 240 m.a.s.l. Geological substrate in the Pa¯lamanui area consists of ‘a‘a¯ lava with scattered pa¯hoehoe flows dating to 1,500–3,000 years old [48]. Soils developing at this site are shallow, rocky, highly organic, and classified as euic, isothermic, shallow Lithic Ustifolist (websoilsurvey.nrcs.usda.gov). Interpolated mean annual precipitation at the LDF site is 835 mm [41,49], with large within- and between-year variability [50]. For the LDF, major rainfall events typically occur in the winter as low pressure storms (‘‘Kona lows’’) while summers tend to be dry and characterized by small convective storms. Mean daily air temperature is approximately 20uC (wrcc.dri.edu). Native vegetation consists of evergreen broad- leaved trees and shrubs that form an open-canopy forest that reaches heights of ,7–8 m dominated by Diospyros sandwicensis (Ebenaceae) and Psydrax odorata (Rubiaceae; Fig. 1). One species (Erythrina sandwicensis) is drought deciduous and is only represented by a few individuals. Data Analyses To estimate above-ground biomass (AGB) for the two plots, we used site-specific and species-specific information whenever possible for wood specific gravity, tree height, and DBH (equations derived from 52,54,55,56; see Table S1 in File S2). When these were not available, we compiled data from global databases, utilizing equations based on other sites, and in some cases for other species from within the same genera [56,57]. Previous studies have reported that genus means are reasonable proxies for species values for specific gravity (r2.0.70; [58,59]). Pre-contact disturbance regimes likely included tree falls. Following contact, selective harvesting of valuable woods (e.g., sandalwood) occurred throughout the area but we do not know of any logging that occurred within the plot. In the last 200 years, much of the lowland dry forest in Hawai‘i has been subjected to grazing and browsing by exotic ungulates, with remnants impacted by wildfire carried by non-native grasses [51]. These factors have reduced the native forest to a fraction of its original extent [52]. While the area containing the FDP has not been burned or significantly browsed by ungulates, the surrounding area is a matrix of degraded LDF and open grassland, and in 2009, a fence and firebreak were installed around the area to protect it from ungulates and fire. Species richness and diversity. Species area curves were generated by plotting cumulative number of species against area for the 20 m620 m quadrats. Rarefaction analyses were based on 999 permutations (PRIMER-E v. 6, PRIMER-E Ltd, Plymouth, UK), which randomized the sampling order and resulted in a robust average curve. We present several indices: Sobs (the observed number of species), Chao 1 based on rare species (non- parametric), and Michaelis-Menten (parametric), given uncertain- ty in the ideal estimator [60,61,62]. We used the program EstimateS 9.1.0 to calculate species diversity indices and an estimate of error. We report Fisher’s alpha, Shannon diversity index, and Simpson’s index (inverse form) following standard formulas [63]. Overlap in species composition between the two sites was determined using the Sørenson similarity index (SI): Data Analyses Stand structure. We determined stand structural character- istics based on DBH measurements. We considered multiple- stemmed plants as single individuals for the calculation of stem density, and summed the basal area of all stems for the calculation of basal area (m2/ha). For each species, we calculated relative abundance (RA, %) as the number of individuals of that species/ total number of individuals, relative dominance (RD, %) as the basal area of that species/total basal area, and relative frequency (RF, %) as the number of quadrats with that species/total number of quadrats. Above-ground biomass. To estimate above-ground biomass (AGB) for the two plots, we used site-specific and species-specific information whenever possible for wood specific gravity, tree height, and DBH (equations derived from 52,54,55,56; see Table S1 in File S2). When these were not available, we compiled data from global databases, utilizing equations based on other sites, and in some cases for other species from within the same genera [56,57]. Previous studies have reported that genus means are reasonable proxies for species values for specific gravity (r2.0.70; [58,59]). Above-ground biomass. To estimate above-ground biomass (AGB) for the two plots, we used site-specific and species-specific information whenever possible for wood specific gravity, tree height, and DBH (equations derived from 52,54,55,56; see Table S1 in File S2). When these were not available, we compiled data from global databases, utilizing equations based on other sites, and in some cases for other species from within the same genera [56,57]. Previous studies have reported that genus means are reasonable proxies for species values for specific gravity (r2.0.70; [58,59]). To determine tree height, we applied species-specific equations of [54] giving the relationship of tree height vs. DBH, to each individual tree for 12 of the MWF species and 4 of the LDF species (Table S1 in File S2). For the other species, we used the general wet and dry forest equations [55] to determine tree height. We used these tree height estimates to calculate AGB for each tree using published equations that also included DBH and wood specific gravity. Hawai‘i-specific equations for AGB were available for 5 MWF species and 4 LDF species. For another two species of the LDF, D. sandwicensis and P. odorata, equations were available that were developed specifically from our study site [52] (Table S1 in File S2). Above-ground biomass. Plot Establishment and Vegetation Measurements We applied field methodology developed by the Center for Tropical Forest Science global FDP network [53]. Both of our 4- ha FDPs (2006200 m) were oriented north-south and located at the center of a 16 ha buffer area, with all edges at least 100 m from any road or major trail where possible. From 2008 to 2009, we tagged all live, native woody plants $1 cm diameter at breast height (DBH, at 130 cm), and mapped tagged plants relative to 5 m65 m grids installed throughout the plots. Each tagged plant was identified to species and measured for DBH. More detailed methods are in Methods S1 in File S1. Finally, we estimated and mapped cover of abundant non- native herbaceous, shrub and tree species, which will be important for understanding long-term vegetation change. At each site, we chose six abundant focal species or life forms that were considered ‘‘invasive pests’’ according to their Hawai‘i Weed Risk Assessment scores (Daehler 2004; www.botany.hawaii.edu/faculty/daehler/ wra/full_table.asp). Percent cover within each 565 m subquadrat was estimated in the following categories: 0: absent, 1: ,5%, 2: 5– SI~ number of species shared in both sites (no: of species in MWFzno: of species in LDF) August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org 4 Low-Diversity Hawaiian Forests Table 2. Diversity and forest structure characteristics of plots in the Center of Tropical Forest Science global plot network, including the Hawaiian plots, arranged in order of descending species richness. CTFS plot location Plot code Latitude Mean annual rain (mm) Dry season months Mean elevation (m) Mean annual temp (6C) Land type Plot size (ha) No. of species No. of families Mean species/ family Fisher’s a/ha H’/ha Trees/ha Basal area (m2/ha) Dom. of most common family Above- ground biomass (Mg/ha)7 Lambir, Malaysia LAM 4.19 2664 0 170 26.6 I 52 1182 83 14.2 165 2.40 6915 43.5 41.0 497.2 Yasunı´, Ecuador YAS 20.69 3081 0 230 28.4 M 50 1114 81 13.8 187 2.44 3026 33.0 14.9 282.4 Pasoh, Malaysia PAS 2.98 1788 1 80 28.0 M 50 814 82 9.93 124 2.31 6708 31.0 28.2 339.8 Khao Chong, Thailand KHA 7.54 19851 32 140 27.3 M 24 593 na na na Na 5063 na na na Korup, Cameroon KOR 5.07 5272 3 200 26.7 M 50 494 62 7.97 48.0 1.75 6580 32.0 16.1 na Ituri, Dem. Rep. Plot Establishment and Vegetation Measurements of Congo3 ITU 1.44 1700 3–4 780 23.1 M 404 445 na na na Na 7200 na na na Palanan, Philippines PAL 17.04 3379 4–5 110 23.5 I 16 335 60 5.58 43.4 1.92 4125 39.8 52.8 290.1 Bukit Timan, Singapore BUK 1.25 2473 0 150 26.9 I 2 329 62 5.31 60.0 1.90 5950 34.5 38.4 na BCI, Panama BCI 9.15 2551 3 140 26.9 M 50 299 58 5.16 34.6 1.62 4168 32.1 11.4 306.5 Mo Singto, Thailand MOS 14.43 2200 5 770 23.0 M 30.5 262 na na na Na Na na na na Huai Kha Khaeng, Thailand HKK 15.63 1474 6 590 24.1 M 50 251 58 4.33 23.3 1.50 1450 31.2 21.2 211.2 La Planada, Colombia LPL 1.16 4084 0 1840 19.0 M 25 240 54 4.44 30.6 1.72 4216 29.8 14.9 177.6 Dinghushan, China DIN 23.16 1985 0 350 20.9 M 20 210 na na na Na 3581 na na na Sinharaja, Sri Lanka SIN 6.4 5016 0 500 22.6 I 25 204 46 4.43 24.4 1.53 7736 45.6 26.7 357.9 Doi Inthanon, Thailand DOI 18.52 1908 6 1700 21.4 M 15 192 na na na Na 4913 na na na Luquillo, Puerto Rico LUQ 18.33 3548 0 380 22.8 I 16 138 47 2.94 13.5 1.45 4194 38.3 17.3 276.1 Nanjenshan, Taiwan NAN 22.06 3582 0 320 23.5 I 3 125 41 3.05 15.6 1.64 12133 36.3 32.3 na Ilha do Cardoso, Brazil5 ILH 225.1 2100 0 5 22.4 M 10 106 na na na Na Na na na na Mudumalai, India MUD 11.6 1250 6 1050 22.8 M 50 72 29 2.48 6.20 0.944 510 25.5 28.2 174.2 Laupa¯hoehoe, USA LAU 19.93 3440 1 1150 16.0 I 4 21 15 1.40 2.58 1.82 3078 67.36 37.4 247.9 Pa¯lamanui, USA PLN 19.74 835 12 240 20.0 I 4 15 15 1.00 1.18 1.00 3487 8.6 74.2 29.4 Southern hemisphere latitudes are negative; land types are island (I) and mainland (M); dry season months are as those with ,100 mm precipitation (Richards 1996). Data are from [104] and ctfs.si.edu unless indicated by footnotes. 1Mean annual rainfall data for the nearby city of Songkhla, Thailand (www.world-climates.com) 2Kira T (1998) NPP Tropical Forest: Khao Chong, Thailand, 1962–1965. Data set. Available on-line [http://www.daac.ornl.gov] from Oak Ridge National Laboratory Distributed Active Archive Center, Oak Ridge, Tennessee, U.S.A. August 2014 \| Volume 9 \| Issue 8 \| e103268 Plot Establishment and Vegetation Measurements 3Average of 4 plots (2 monodominant forest, 2 mixed forest) 4Divided into four 10-ha plots 5Data from Ferreira de Lima RA, Oliveira AAD, Martini AMZ, Sampaio D, Souza VC, Rodrigues RR (2011) Structure, diversity, and spatial patterns in a permanent plot of a high restinga forest in Southeastern Brazil. Acta Botanica Brasilica 25: 633–645. 6Basal area including tree ferns; 36.1m2/ha without tree ferns 7Chave J and 37 others (2008) Assessing evidence for a pervasive alteration in tropical tree communities. PLoS Biology 6: 455–462. doi:10.1371/journal.pone.0103268.t002 August 2014 \| Volume 9 \| Issue 8 \| e103268 August 2014 \| Volume 9 \| Issue 8 \| e103268 5 Low-Diversity Hawaiian Forests Figure 2. Life form distribution of stems and biomass by diameter size intervals. In (A) Hawaiian montane wet forest (MWF) and (B) lowland dry forest (LDF), stems represent the number of main stems (i.e., one per individual, not including other multiple stems). In (C) MWF and (D) (LDF), biomass calculations were made for all stems (including multiple stemmed individuals). Diameter classes are 1–4.99 cm, 5 - ,9.99 cm, 10– 29.99 cm, 30–59.99 cm, 60–99.99 cm, and $100 cm. doi:10.1371/journal.pone.0103268.g002 Figure 2. Life form distribution of stems and biomass by diameter size intervals. In (A) Hawaiian montane wet forest (MWF) and (B) lowland dry forest (LDF), stems represent the number of main stems (i.e., one per individual, not including other multiple stems). In (C) MWF and (D) (LDF), biomass calculations were made for all stems (including multiple stemmed individuals). Diameter classes are 1–4.99 cm, 5 - ,9.99 cm, 10– 29.99 cm, 30–59.99 cm, 60–99.99 cm, and $100 cm. doi:10.1371/journal.pone.0103268.g002 M. polymorpha occurred in both forests; it was the second most common species in MWF but was represented by only 5 individuals in the 4-ha LDF plot. Species richness was 21 in the MWF and 15 in the LDF. Fifteen families were represented at each site, and the canopy trees at the two sites were from different families, though four families were represented in the understory or the midstory at both sites (Euphorbiaceae, Fabaceae, Myrta- ceae, and Rubiaceae). Forest type comparisons. We compiled data for 19 additional mainland and island CTFS tropical plots for which climate and structure data were available (Table 2). Differences between the Hawai‘i plots and other CTFS plots were assessed using one sample t-tests. Differences in the characteristics of island and mainland plots were assessed using Wilcoxon signed-rank tests [64]. Results The plots differed in their distribution of plant life forms. In the MWF, 68% of stems were trees, 4.5% were shrubs, and 28% were tree ferns, accounting for 45%, 8.3% and 46% of the basal area respectively. In the LDF, 82% of stems were trees and 18% were shrubs, accounting for 95% and 5% of basal area respectively (Fig. 2). In the MWF, a large proportion of stems (31%) were growing on non-soil substrates, primarily tree ferns, logs or rocks, Plot Establishment and Vegetation Measurements These statistics were analyzed with JMP v. 6 [29,65]. August 2014 \| Volume 9 \| Issue 8 \| e103268 Comparison of Floristics and Life Forms in Hawaiian Forests The two Hawai‘i forests were distinct in floristic composition (Table S2 in File S2). The plots had a very low Sørenson similarity index of 0.06 (a value of 1 would indicate complete overlap). Only August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e103268 6 Low-Diversity Hawaiian Forests Table 3. Statistics on abundance, basal area, and frequency of the species in the Laupa¯hoehoe (montane wet forest) plot, with data displayed on an absolute and a relative b Laupa¯hoehoe montane wet forest Species No. individuals Basal area (m2/ha) Presence (no. of quadrats) Relative abundance (%) Relative dominance (%) Relative frequency (%) IV (%) METPOL 2631 25.2 100 21.4 37.5 10.2 69.1 CIBGLA 2274 17.7 100 18.5 26.4 10.2 55.1 CHETRI 3320 4.17 100 27.0 6.20 10.2 43.4 CIBMEN 1076 13.2 100 8.74 19.6 10.2 38.6 COPRHY 972 0.585 99 7.90 0.870 10.1 18.9 ILEANO 965 0.466 99 7.84 0.692 10.1 18.7 ACAKOA 141 5.49 57 1.15 8.16 5.84 15.1 BROARG 271 0.0454 74 2.20 0.067 7.58 9.85 MYRLES 237 0.0571 70 1.93 0.085 7.17 9.18 VACCAL 255 0.0328 51 2.07 0.0488 5.23 7.35 HEDHIL 43 0.0200 29 0.349 0.0297 2.97 3.35 PERSAN 35 0.0084 28 0.284 0.0125 2.87 3.17 CIBCHA 34 0.232 17 0.276 0.345 1.74 2.36 CLEPAR 19 0.00230 17 0.154 0.00342 1.74 1.90 MELCLU 13 0.00235 11 0.106 0.00349 1.13 1.24 PSYHAW 10 0.00219 9 0.0812 0.00326 0.922 1.01 MYRSAN 6 0.00166 6 0.0487 0.00246 0.615 0.666 PIPALB 4 0.00226 4 0.0325 0.00335 0.410 0.446 TREGRA 2 0.000107 2 0.0162 0.000159 0.205 0.221 LEPTAM 2 0.0000682 2 0.0162 0.000101 0.205 0.221 ANTPLA 1 0.000154 1 0.0081 0.000229 0.102 0.111 Total 12311 67.3 August 2014 \| Volume 9 \| Issue 8 \| e103268 7 Low-Diversity Hawaiian Forests Total 13946 8.64 Presence based on 100 20620 m quadrats per plot; species sorted by importance value (IV), which is the sum of the three relative measures (max 300%); species abbreviations as in Table S2 in File S2. doi:10.1371/journal.pone.0103268.t004 Table 4. Statistics on abundance, basal area, and frequency of the species in the Pa¯lamanui (lowland dry forest) plot, with data displayed on an absolute and a relative Pa¯lamanui lowland dry forest Species No. individuals Basal area (m2/ha) Presence (no. August 2014 \| Volume 9 \| Issue 8 \| e103268 Comparison of Floristics and Life Forms in Hawaiian Forests of quadrats) Relative abundance (%) Relative dominance (%) Relative frequency (%) IV (%) DIOSAN 2208 6.41 99 15.8 74.2 18.3 108.3 PSYODO 8640 1.27 100 62.0 14.7 18.5 95.2 DODVIS 2301 0.359 94 16.5 4.15 17.4 38.1 SOPCHR 5 0.21 4 0.0359 2.38 0.741 3.20 SANPAN 275 0.156 32 1.97 1.81 5.93 9.70 OSTANT 147 0.0900 40 1.05 1.04 7.41 9.50 WIKSAN 88 0.0890 44 0.631 1.03 8.15 9.81 EUPMUL 134 0.023800 54 0.961 0.275 10.0 11.2 SENGAU 70 0.013300 27 0.502 0.154 5.00 5.66 METPOL 12 0.013 9 0.0860 0.152 1.67 1.90 MYOSAN 54 0.007480 26 0.387 0.087 4.81 5.29 SIDFAL 1 0.0008 1 0.0072 0.00947 0.185 0.202 PLEHAW 1 0.0007 1 0.0072 0.00843 0.185 0.201 ERYSAN 2 0.000513 2 0.0143 0.00594 0.370 0.391 PITTER 8 0.00048 7 0.0574 0.00551 1.30 1.36 Total 13946 8.64 Presence based on 100 20620 m quadrats per plot; species sorted by importance value (IV), which is the sum of the three relative measures (max 300%); species abbreviations as in Table S2 in File S2. doi:10.1371/journal.pone.0103268.t004 8 Low-Diversity Hawaiian Forests Table 5. Aboveground biomass listed by species for the two Hawai‘i forest plots; species abbrev Table 5. Aboveground biomass listed by species for the two Hawai‘i forest plots; species abbreviations as in Table S2 in File S2. Laupa¯hoehoe montane wet forest Pa¯lamanui lowland dry forest Species Biomass (Mg/ha) Relative biomass (%) Species Biomass (Mg/ha) Relative biomass (%) METPOL 186 74.9 PSYODO 15.3 51.9 ACAKOA 31.1 12.5 DIOSAN 10.5 35.8 CHETRI 12.4 4.99 METPOL 1.40 4.78 CIBMEN 10.9 4.39 DODVIS 0.921 3.14 CIBGLA 4.55 1.83 OSTANT 0.525 1.79 COPRHY 1.59 0.64 SANPAN 0.359 1.22 ILEANO 1.27 0.51 WIKSAN 0.181 0.615 CIBCHA 0.184 0.07 MYOSAN 0.109 0.372 MYRLES 0.109 0.04 SOPCHR 0.0446 0.152 VACCAL 0.0947 0.04 SENGAU 0.0398 0.135 HEDHIL 0.0456 0.02 EUPMUL 0.0199 0.068 BROARG 0.0394 0.02 PITTER 0.00435 0.0148 PERSAN 0.0124 0.00499 PLEHAW 0.00217 0.00740 PSYHAW 0.00384 0.00155 SIDFAL 0.00187 0.00637 MELCLU 0.00343 0.00138 ERYSAN 0.000871 0.00297 CLEPAR 0.00327 0.00132 MYRSAN 0.00319 0.00128 PIPALB 0.00274 0.00110 Total 29.4 ANTPLA 0.000287 0.000116 LEPTAM 0.00017 0.0000685 TREGRA 0.0000703 0.0000283 Total 247.9 doi:10.1371/journal.pone.0103268.t005 doi:10.1371/journal.pone.0103268.t005 with 17% of all individuals growing on dead tree ferns (Table S3 in File S2). In contrast, in the LDF, all trees were growing on soil or broken lava, and tree ferns were absent. Comparison of Floristics and Life Forms in Hawaiian Forests Comparison of Community Structure in Hawaiian Forests For both the Hawaiian MWF and LDF, rarefaction curves indicated that a 1 ha sample was sufficient to capture 90% of the species present in the larger 4-ha area (Fig. 3). In the MWF, diversity values for the plot were 2.46 (Fisher’s alpha), 1.98 (Shannon), and 5.74 (Simpson); in the LDF values were 1.66 (Fisher’s alpha), 1.15 (Shannon), and 2.29 (Simpson). When viewed graphically, there was no overlap in any index value between the two forests (Fig. 4): the Hawaiian MWF was more diverse than the LDF. In the MWF, species evenness was higher than in the LDF, primarily because P. odorata in the LDF had a relative abundance of over 60%. In contrast the forests were similar in the abundance of uncommon species, and ,20% of species were rare, i.e., having #1 stem/ha (Tables 3–4). Comparison of Stand Structure in Hawaiian Forests The MWF had larger trees and lower stem density than the LDF (307861.21 and 348761.40 stems/ha respectively; Table 3). The tree size class distributions differed between the two forests as expected based on their contrasting climates: the LDF had mainly small stems and the MWF had a much more even spread of size classes (Fig. 2). Because the stems in the LDF were small, total basal area and biomass values were low. Thus, the MWF had a nearly eight-fold higher basal area than the LDF (67.3 vs. 8.6 m2/ ha respectively; Tables 3–4), and tree ferns accounted for 31.2 m2/ha basal area. Above-ground biomass in the MWF was also more than eight times higher than the LDF (248 Mg/ha vs. 29.4 Mg/ha respectively; with 15.6 Mg/ha in the MWF account- ed for by tree ferns; Table 5). The above-ground biomass value for the MWF was consistent with that previously estimated for surrounding forest in the same reserve [66]. In both forest types, the two most common canopy species represented 87–88% of biomass (Table 5). In the MWF, the very large trees ($60 cm) made up the greatest proportion of the biomass, but in the LDF the majority of the biomass was in the 1–5 cm size class (Fig. 2). More multi-stemmed individuals make up the LDF, a mean of 3.2 stems/individual, compared to 1.4 stems/individual in the MWF plot) (Table S4 in File S2). Comparing Hawaiian Wet and Dry Forests Comparing Hawaiian Wet and Dry Forests Examination of how the two Hawaiian forests compared in terms of composition and structure matched and extended the paradigm for differences between mature wet and dry forests described on other tropical islands [32]. As predicted, our results for Hawai‘i are in agreement with comparisons of mature wet and dry forests in Puerto Rico [32]: wet forest had larger diameter trees, greater basal area, and higher biomass than dry forest, and differences between wet and dry forest in tree density, dominance, and species richness were minor (Tables 3–4). Indeed, across many forests, biomass and basal area are typically correlated with climatic variables such as MAT, MAP, and water deficit within and across sites [31,70,71,72,73,74,75]. However, other variables such as substrate age and type [76] are likely to have contributed to structural and floristic differences between LDF and MWF and we cannot ascribe results solely to climate. While both forests occur on young lava, the much higher rainfall of the MWF and variation in substrate type and texture contributed to greater soil development. Lava flow age and substrate type are important determinants of successional stage in Hawai‘i [39]. In addition, differences in disturbance regimes between the two sites may have influenced their forest structure, and their invasive species cover. In the MWF, almost a third of stems were found growing on a substrate other than soil, such as nurse logs and living tree ferns, which likely reflects preferential survivorship on those substrates. Further, canopy dieback of Metrosideros polymorpha [45], wind storms [44], and invasive animals [77] may be important factors influencing forest structure. Canopy gaps are larger in the MWF due to the much larger and taller trees that make up the canopy. ( ) In contrast to biodiversity, the structural comparisons across the CTFS network revealed complicated patterns. The MWF was similar to other CTFS sites with respect to tree size class distribution (Fig. 9), and was not significantly different from other CTFS plots with respect to standing above-ground biomass/ha. However, the MWF had 35% lower stem density than the all- forest mean of 47336722 SE (t = 2.29, P = 0.039, n = 14). Further, the MWF had a 92% higher basal area than the mean of other tropical FDPs due to its high tree fern abundance (t = 221.6, P, 0.0001). Non-native Species in the Plots Invasive species made up a larger presence in LDF than MWF (Figs.5–6). The grass Pennisetum setaceum was most widespread in the LDF and the herbaceous weed Persicaria punctata was most common in the MWF, where it tended to dominate low-lying boggy areas. In the LDF there was a greater overall weed cover, particularly of woody weeds (Fig. 7; see also Methods S1 in File S1). In the MWF there were only a few stems that qualified for DBH measurements (.5 cm). F. uhdei averaged 32.3 cm (n = 1) and P. cattleianum averaged 6.2 cm (n = 1). In the LDF, average DBH for Grevillia robusta was 20.068.2 cm SE (n = 27), for Leucaena leucocephalum it was 5.960.8 cm SE (n = 7), and for Schinus terebinthius it was 11.061.5 cm SE (n = 20). PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e103268 9 Low-Diversity Hawaiian Forests Figure 3. Species accumulation curves. Species number is shown cumulatively, as additional 20 m620 m quadrats are sampled, until the entire 4-ha plot is represented (100 quadrats), for Hawaiian montane wet forest (MWF) and lowland dry forest (LDF). Three rarefaction techniques are used: Sobs (observed species number), Chao 1, and MM (Michaelis-Menten). doi:10.1371/journal.pone.0103268.g003 Figure 3. Species accumulation curves. Species number is shown cumulatively, as additional 20 m620 m quadrats are sampled, until the entire 4-ha plot is represented (100 quadrats), for Hawaiian montane wet forest (MWF) and lowland dry forest (LDF). Three rarefaction techniques are used: Sobs (observed species number), Chao 1, and MM (Michaelis-Menten). doi:10.1371/journal.pone.0103268.g003 Discussion Diversity of both Hawai‘i forests was very low relative to other CTFS forests, including those on islands and with dry climates (Fig. 8). Across the CTFS network, the mean Fisher’s alpha per ha 6 SE was 59.7616.6 (n = 13), and the two Hawaiian forests were statistical outliers, with diversity values more than 2 SD lower (t = 3.45 and 3.53, P,0.005). The MWF had approximately 15% as many species as the most comparable island site with tropical wet forest (Luquillo, Puerto Rico). Compared with the next two driest CTFS sites, the Hawai‘i LDF had 21% of the number of species found in the Mudumalai, India plot and just 6% of the number of species found at the Huai Kha Khaeng, Thailand plot (Table 2). Comparing Hawaiian Wet and Dry Forests When tree ferns were excluded, the basal area of the Hawaiian MWF was within the range of that for other FDPs. For the LDF, stem density was not significantly different than the all- forest mean (t = 1.79, P = 0.097, n = 16, Table 2), but the LDF was an outlier with its very low basal area (t = 10.40, P,0.001, n = 19) and above-ground biomass (t = 7.12, P,0.001, n = 12). The LDF was especially distinctive in having virtually all small stems (Fig. 9). August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e103268 10 Low-Diversity Hawaiian Forests re 4. Species diversity indices. Fisher’s alpha, Shannon index, and Simpson index for the Hawaiian montane wet forest (MWF) orest (LDF). Each 20620 m subplot is shown, with the values being cumulative and number above each line representing the entire Values are the diversity index and standard deviation, as estimated by the program EstimateS. 0.1371/journal.pone.0103268.g004 Low Diversity Haw Figure 4. Species diversity indices. Fisher’s alpha, Shannon index, and Simpson index for the Hawaiian montane wet forest (MWF) and lowland dry forest (LDF). Each 20620 m subplot is shown, with the values being cumulative and number above each line representing the entire plot area (4- ha). Values are the diversity index and standard deviation, as estimated by the program EstimateS. doi:10.1371/journal.pone.0103268.g004 Figure 4. Species diversity indices. Fisher’s alpha, Shannon index, and Simpson index for the Hawaiian montane wet forest (MWF) and lowland dry forest (LDF). Each 20620 m subplot is shown, with the values being cumulative and number above each line representing the entire plot area (4- ha). Values are the diversity index and standard deviation, as estimated by the program EstimateS. doi:10.1371/journal.pone.0103268.g004 Figure 4. Species diversity indices. Fisher’s alpha, Shannon index, and Simpson index for the Hawaiian montane wet forest (MWF) and lowland dry forest (LDF). Each 20620 m subplot is shown, with the values being cumulative and number above each line representing the entire plot area (4- ha). Values are the diversity index and standard deviation, as estimated by the program EstimateS. doi:10.1371/journal.pone.0103268.g004 August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org 11 Low-Diversity Hawaiian Forests Figure 5. Invasive species cover distribution. Map showing percent cover and locations of invasive species in the MWF. Comparing Hawaiian Wet and Dry Forests Each grid square represents one 565-m subquadrat white: absent, light grey: present to ,5%, medium grey: 5–25%, dark grey: 25–50%, black: .50% cover). doi:10.1371/journal.pone.0103268.g005 Figure 5. Invasive species cover distribution. Map showing percent cover and locations of invasive species in the MWF. Each grid square represents one 565-m subquadrat white: absent, light grey: present to ,5%, medium grey: 5–25%, dark grey: 25–50%, black: .50% cover). doi:10.1371/journal.pone.0103268.g005 There is also evidence of pig rooting that may affect seedling regeneration preferences [47,78], explaining why many stems grow on substrates other than soil. The LDF site is currently fenced from ungulates but the large proportion of multi-stemmed trees and the higher prevalence of regeneration by sprouting suggest adaptation to disturbance [79,80,81]. can significantly alter forest functioning [50,51,52,92,94]. A debate in invasion biology is whether invaders owe their success to their introducing a new function to the community (e.g., N- fixing species) or are simply better competitors [92] and we argue that it is the latter case at our sites. At our sites, invasive grasses were widespread, but, woody invaders are a greater competitive threat (Figs. 7). In the LDF, Pennisetum setaceum is widespread, but the vegetation is still dominated by woody species with moderate canopy closure, and reduction of grass cover and fire prevention will reduce its competitive effect in the future. While non-native grasses and herbs are more common than non-native trees across the MWF, their abundance is strongly related to boggy areas, canopy openings, and pig disturbance, and these patches are not likely to expand, but rather to be shaded out in the long term. In MWF, woody invaders such as Psidium cattleianum represent much greater threats based on their extreme abundance elsewhere, and traits such as shade tolerance, vegetative repro- duction, and animal-dispersed fruits [93,94]. At present, the MWF has limited cover of woody invaders (Fig. 7), and in that respect is in better condition that the LDF. The MWF site distinguishes itself in its abundance and dominance of tree ferns, which form a distinct mid-canopy layer approximately 5 m above the ground. Notably, tree ferns also make up a large proportion of stand basal area or stem density in some temperate rain forests [82,83,84] and tropical cloud forests [85,86,87,88] but not in other CTFS sites. Comparing Hawaiian Wet and Dry Forests In Hawai‘i, tree ferns are common in wet forests at all elevations, and are particularly abundant in areas with more well-developed organic soils as opposed to young lava flows. While the dynamics of tree ferns have not been well studied in tropical environments [89], in Hawaiian forests tree ferns undoubtedly influence forest function, due to their long lifespans, high frond area, slow growth [90], and slow decomposition rates [91]. They also play a critical role as a substrate for tree seedlings [47,78]. Clearly, site-specific properties influence the structure and species composition between the two sites, but our study also highlights that at the island scale (1 million ha), climate likely exerts a strong influence, both directly and indirectly [67]. These differences matched patterns found in continental forests, where diversity measures as well as structural measures correlate negatively with the length or severity of the dry season [68,69]. Unlike other CTFS plots that are not heavily impacted by non- native plant species, all forests in Hawai‘i have been invaded to some degree. We purposefully chose sites with low non-native species abundance, but cataloged cover before removal for future long-term studies. Because we are removing the invasive species after data collection, we are not examining the consequences of invasion, but previous work in Hawai‘i has shown that invaders August 2014 \| Volume 9 \| Issue 8 \| e103268 August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org 12 Low-Diversity Hawaiian Forests Figure 6. Map showing percent cover and locations of invasive species in the LDF. Each grid square represents one 565-m subquadrat (white: absent, light grey: present to ,5%, medium grey: 5–25%, dark grey: 25–50%, black: .50% cover). doi:10.1371/journal.pone.0103268.g006 Figure 6. Map showing percent cover and locations of invasive species in the LDF. Each grid square represents one 565-m subquadrat (white: absent, light grey: present to ,5%, medium grey: 5–25%, dark grey: 25–50%, black: .50% cover). doi:10 1371/journal pone 0103268 g006 Figure 6. Map showing percent cover and locations of invasive species in the LDF. Each grid square represents one 565-m subquadrat (white: absent, light grey: present to ,5%, medium grey: 5–25%, dark grey: 25–50%, black: .50% cover). doi:10.1371/journal.pone.0103268.g006 Figure 7. Combined invasive species cover. In each 565 m subquadrat a cover score from 0–4 was given based on cover classes (see Methods). Comparing Hawaiian Wet and Dry Forests The y axis represents the average cover class across the 400 subquadrats, separated by life form: grasses, herbaceous, or woody (shrubs and trees). The combined cover represents the species shown in Figure 5. doi:10.1371/journal.pone.0103268.g007 Figure 7. Combined invasive species cover. In each 565 m subquadrat a cover score from 0–4 was given based on cover classes (see Methods). The y axis represents the average cover class across the 400 subquadrats, separated by life form: grasses, herbaceous, or woody (shrubs and trees). The combined cover represents the species shown in Figure 5. doi:10.1371/journal.pone.0103268.g007 August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org 13 ture and Diversity across Tropical Forests Globally e Hawaiian forest data allowed for the examination of the was previously available. One of the most striking conclu our study is that, despite the extremely low species richne e 8. Comparisons of species richness and stem density across a series of CTFS plots. Black bars represent continents and o ent islands. Abbreviations as in Table 2. Data from Losos and Leigh, Jr. (2004) and www.ctfs.si.edu. .1371/journal.pone.0103268.g008 Low-Diversity Hawaiia Low-Diversity Hawaiian Forests Figure 8. Comparisons of species richness and stem density across a series of CTFS plots. Black bars represent continents and open bars represent islands. Abbreviations as in Table 2. Data from Losos and Leigh, Jr. (2004) and www.ctfs.si.edu. doi:10.1371/journal.pone.0103268.g008 Figure 8. Comparisons of species richness and stem density across a series of CTFS plots. Black bars represent continents and open bars represent islands. Abbreviations as in Table 2. Data from Losos and Leigh, Jr. (2004) and www.ctfs.si.edu. doi:10.1371/journal.pone.0103268.g008 Structure and Diversity across Tropical Forests Globally The Hawaiian forest data allowed for the examination of the question of how forest structure varies across species diversity gradients across a much wider range of tree species diversity than was previously available. Acknowledgments The Hawai‘i Permanent Plot Network thanks the USFS Institute of Pacific Islands Forestry (IPIF) and the Hawai‘i Division of Forestry and Wildlife/ Department of Land and Natural Resources for permission to conduct research within the Hawai‘i Experimental Tropical Forest; the Pala¯manui Group, especially Roger Harris, for access to the lowland dry forest site. We thank N. DiManno, L. Ellsworth, B. Hwang, R. Moseley, M. Murphy, K. Nelson-Kaula, M. Nullet, C. Perry, J. Schulten, M. Snyder, and J. VanDeMark for logistical assistance, among the many others who served as project interns and volunteers (see hippnet.hawaii.edu); G. Asner, S. Davies, T. Giambelluca, J. Mascaro, D. Metcalfe, J. Michaud, and J. Thompson for technical advice and/or comments on the manuscript In conclusion, Hawaiian forests have among the lowest species richness and highest endemism rates globally, but in a number of key structural variables both of these forests were similar to even the highest diversity tropical forests in the CTFS network. Future work could examine the evolutionary consequences of such a limited species pool. Biodiversity theory developed in high- diversity tropical forests emphasizes that competitive interactions among species are unlikely on evolutionary time scales because any given two species are rarely consistent neighbors [9]. However, in low-diversity forest any two given species have far greater potential for competitive interactions than in high-diversity tropical forests [102,103]. The addition of Hawai‘i to the global plot network enables investigations of the consequences of such 12. Chisholm RA, Muller-Landau HC, Abdul Rahman K, Bebber DP, Bin Y, et al. (2013) Scale-dependent relationships between tree species richness and ecosystem function in forests. Journal of Ecology 101: 1214–1224. Comparing Hawaiian Wet and Dry Forests One of the most striking conclusions of our study is that, despite the extremely low species richness of the Hawai‘i FDPs, some structural variables, particularly those for the wet forest, were well within the range of values for the world’s most The Hawaiian forest data allowed for the examination of the question of how forest structure varies across species diversity gradients across a much wider range of tree species diversity than August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org 14 Low-Diversity Hawaiian Forests ONE \| www.plosone.org 15 August 2014 \| Volume 9 \| Issue 8 \| August 2014 \| Volume 9 \| Issue 8 \| e103268 August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org 15 Low-Diversity Hawaiian Forests Figure 9. A reverse-cumulative distribution of basal area by size class. Size classes are: $1 cm, $10 cm, $30 cm, and $60 cm. Data shown for the Hawaiian montane wet forest (LAU) and lowland dry forest (PLN) (top row) and for selected other CTFS plots. Tree ferns (found only at LAU) are symbolized by the gray bars. Island sites are open bars and continental sites are filled bars. Abbreviations as Table 2. Data from Losos and Leigh, Jr. (2004) and www.ctfs.si.edu. doi:10.1371/journal.pone.0103268.g009 differences across a very wide range in species diversity and environmental gradients. diverse tropical forests (Table 2). For example, stem densities for Hawaiian MWF and LDF were similar to those of the hyper- species-rich Yasun??? FDP (Fig. 8), while biomass and basal area of the Hawai‘i MWF (excluding tree ferns) were similar to those of the higher diversity forests in the CTFS network (Fig. 9). The inclusion of tree ferns increased basal area values by 52%, but only increased biomass by 11% (Tables 3 and 4). Notably, the LDF had among the lowest basal area and biomass in the CTFS network, consistent with this site having the lowest precipitation of all FDPs (Table 2). It should also be noted that the LDF is dry year-round, while other dry sites in the CTFS network are seasonally dry (Table 2). Supporting Information File S1 Methods S1. Detailed methods and description of situations where field site conditions dictated a different or entirely new methodology by adopted than standardized CTFS protocol in [1]. (DOCX) File S2 Supporting tables. Table S1. Values and equations used for estimating aboveground biomass (AGB) in the montane wet forest (MWF) and lowland dry forest sites (LDF). Table S2. Species $1 cm diameter at breast height recorded in Laupa¯hoe- hoe (montane wet forest) plot with canopy dominants in bold. Table S3. Percentage of individuals in the Laupa¯hoehoe (montane wet forest) plot growing on each substrate type. Table S4. Size and multiple stem characteristics of the species species in Laupa¯hoehoe (montane wet forest) and Pa¯lamanui (lowland dry forest) plots; species abbreviations as in Table S4. Table S5. References from Table 1. File S2 Supporting tables. Table S1. Values and equations used for estimating aboveground biomass (AGB) in the montane wet forest (MWF) and lowland dry forest sites (LDF). Table S2. Species $1 cm diameter at breast height recorded in Laupa¯hoe- hoe (montane wet forest) plot with canopy dominants in bold. Table S3. Percentage of individuals in the Laupa¯hoehoe (montane wet forest) plot growing on each substrate type. Table S4. Size and multiple stem characteristics of the species species in Laupa¯hoehoe (montane wet forest) and Pa¯lamanui (lowland dry forest) plots; species abbreviations as in Table S4. Table S5. References from Table 1. The low floristic richness and population structure of the Hawaiian forest plots represented strong convergence with other island forests. Hawaiian forests had fewer species per family and greater average population densities for each species, as seen in other very isolated sites [95]. High relative dominance values were consistent with island forests having greater dominance by the most common family than mainland tropical forests (Table 2). In the MWF and LDF, 37% and 74% of basal area respectively were accounted for by a single canopy dominant species. On average 20% of species were rare in Hawaiian forests (defined as #1 tree/ ha), by contrast with 42% on average across other high-diversity forests [96]. It is likely that the patterns of high basal area dominance in Hawaiian forests arose due to the biogeographic consequences of isolation, but we cannot rule out species loss due to human disturbance and invasive species of multiple trophic levels [97,98,99,100,101]. (DOCX) Author Contributions Conceived and designed the experiments: RO FIN SC CPG LS. Performed the experiments: RO FIN SC CPG LS. Analyzed the data: RO FIN LS. Contributed reagents/materials/analysis tools: RO FIN SC CPG LS. Wrote the paper: RO FIN SC CPG LS. y J gy 13. Torti SD, Coley PD, Kursar TA (2001) Causes and consequences of monodominance in tropical lowland forests. American Naturalist 157: 141– 153. 1. Gentry AH (1988) Changes in plant community diversity and floristic composition on environmental and geographical gradients. Annals of the Missouri Botanical Garden 75: 1–34. 11. Hooper DU, Chapin FS, Ewel JJ, Hector A, Inchausti P, et al. (2005) Effects of biodiversity on ecosystem functioning: A consensus of current knowledge. Ecological Monographs 75: 3–35. 2. Givnish TJ (1999) On the causes of gradients in tropical tree diversity. Journal of Ecology 87: 193–210. g g p y gy 5. Hawkins BA, Rodrı´guez MA´ , Weller SG (2011) Global angiosperm family richness revisited: linking ecology and evolution to climate. Journal of Biogeography 38: 1253–1266. 4. Mittelbach GG, Schemske DW, Cornell HV, Allen AP, Brown JM, et al. (2007) Evolution and the latitudinal diversity gradient: speciation, extinction and biogeography. Ecology Letters 10: 315–331. 6. Condit R (1998) Tropical forest densus plots: Methods and results from Barro Colorado Island, Panama and a comparison with other plots. Berlin, Germany, and Georgetown, Texas, USA: Springer-Verlag and R. G. Landes Company. 211 p. 8. Givnish TJ (1999) On the causes of gradients in tropical tree diversity. Journal of Ecology 87: 193–210. 1. Gentry AH (1988) Changes in plant community diversity and floristic composition on environmental and geographical gradients. Annals of the Missouri Botanical Garden 75: 1–34. 2. Givnish TJ (1999) On the causes of gradients in tropical tree diversity. Journal of Ecology 87: 193–210. 3. Phillips O, Gentry AH, Hall P, Sawyer S, Vasquez R (1994) Dynamics and species richness of tropical rain forests. Proceedings of the National Academy of Sciences 91: 2805–2809. 4. Mittelbach GG, Schemske DW, Cornell HV, Allen AP, Brown JM, et al. (2007) Evolution and the latitudinal diversity gradient: speciation, extinction and biogeography. Ecology Letters 10: 315–331. 5. Hawkins BA, Rodrı´guez MA´ , Weller SG (2011) Global angiosperm family richness revisited: linking ecology and evolution to climate. Journal of Biogeography 38: 1253–1266. 6. Condit R (1998) Tropical forest densus plots: Methods and results from Barro Colorado Island, Panama and a comparison with other plots. Berlin, Germany, and Georgetown, Texas, USA: Springer-Verlag and R. G. Landes Company. 211 p. 7. Rosenzweig ML (1995) Species diversity in space and time. Cambridge, UK: Cambridge University Press. Low-Diversity Hawaiian Forests Price JP, Wagner WL (2004) Speciation in Hawaiian angiosperm lineages: cause, consequence, and mode. Evolution 58: 2185–2200. 68. Ashton PS (2004) Floristics and vegetation of the forest dynamics plots. In: Losos E, Leigh EG Jr, editors. Tropical forest diversity and dynamism: Findings from a large-scale plot network. Chicago, Illinois, USA: University of Chicago Press. 37. Ziegler AC (2002) Hawaiian natural history, ecology, and evolution. Honolulu, HI: University of Hawaii Press. 38. Harrison RD (2011) Emptying the forest: Hunting and the extirpation of wildlife from tropical nature reserves. BioScience 61: 919–924. 69. Davidar P, Puyravaud JP, Leigh EG Jr (2005) Changes in rain forest tree diversity, dominance and rarity across a seasonality gradient in the Western Ghats, India. Journal of Biogeography 32: 493–501. 39. Vitousek PM, Farrington H (1997) Nutrient limitation and soil development: experimental test of a biogeochemical theory. Biogeochemistry 37: 63–75. p g y g y 40. Juvik SP, Juvik JO, Paradise TR (1998) Atlas of Hawaii, 3rd edition. Honolulu, HI: University of Hawaii Press. 70. Clark JS, McLachlan JS (2003) Stability of forest diversity. Nature 423: 635– 638. 41. Giambelluca TW, Chen Q, Frazier AG, Price JP, Chen Y-L, et al. (2013) Online Rainfall Atlas of Hawai‘i. Bulletin of the American Meteorological Society 94: 313–316. 71. Clark DA, Piper SC, Keeling CD, Clark DB (2003) Tropical rain forest tree growth and atmospheric carbon dynamics linked to interannual temperature variation during 1984–2000. Proceedings of the National Academy of Sciences 100: 5852–5857. y 42. Crews TE, Kitayama K, Fownes JH, Riley RH, Herbert DA, et al. (1995) Changes in soil phosphorus fractions and ecosystem dynamics across a long chronosequence in Hawaii. Ecology 76: 1407–1424. 72. Feeley KJ, Wright SJ, Supardi MNN, Kassim AR, Davies SJ (2007) Decelerating growth in tropical forest trees. Ecology Letters 10: 1–9. 43. Tosi JAJ, Watson V, Bolan˜os R (2002) Life zone maps Hawaii, Guam, American Samoa, Northern Mariana Islands, Palau, and the Federated States of Micronesia. San Jose, Costa Rica and Hilo, Hawaii, USA: Tropical Science Center and the Institute of Pacific Islands Forestry, USDA Forest Service. 73. Clark DB, Clark DA, Oberbauer SF (2010) Annual wood production in a tropical rain forest in NE Costa Rica linked to climatic variation but not to increasing CO2. Global Change Biology 16: 747–759. g g gy 74. Stegen JC, Swenson NG, Enquist BJ, White EP, Phillips OL, et al. Low-Diversity Hawaiian Forests 14. Ghazoul J, Sheil D (2010) Tropical rain forest ecology, diversity, and conservation. Oxford: Oxford University Press. 46. Friday JB, Scowcroft PG, Ares A (2008) Responses of native and invasive plant species to selective logging in an Acacia koa-Metrosideros polymorpha forest in Hawai’i. Applied Vegetation Science 11: 471–482. 15. Whittaker RJ, Triantis KA, Ladle RJ (2008) A general dynamic theory of oceanic island biogeography. Journal of Biogeography 35: 977–994. 47. Murphy M, Inman-Narahari F, Ostertag R, Litton CM (2014) Invasive feral pigs impact native tree ferns and woody seedlings in Hawaiian forest. Biological Invasions 16: 63–71. g g p y g g p y 16. Kier G, Kreft H, Lee TM, Jetz W, Ibisch PL, et al. (2009) A global assessment of endemism and species richness across island and mainland regions. Proceedings of the National Academy of Sciences 106: 9322–9327. 48. Moore RB, Claque DA, Rubin M, Bohrson WA (1987) Hualalai volcano: A preliminary summary of geologic, petrologic, and geophysical data. In: Decker RW, Wright TL, Stauffer PH, editors. Volcanism in Hawaii. Washington, DC: U.S. Geological Service Professional Paper 1350, U’. Government Printing Office. pp. 571–585. g y 17. Terborgh J (1985) The vertical component of plant species diversity in temperate and tropical forests. The American Naturalist 126: 760–776. p p 18. Spies TA (1998) Forest structure: A key to the ecosystem. Northwest Science 72: 34–39. 49. Giambelluca T, Chen Q, Frazier A, Price J, Chen Y-L, et al. (2011) The Rainfall Atlas of Hawai‘i. Available: http://rainfall.geography.hawaii.edu. Accessed 2014 Aug 10. 19. Whitmore TC (1998) An introduction to tropical rain forests. Oxford, UK: Oxford Press. 20. Moles AT, Bonser SP, Poore AGB, Wallis IR, Foley WJ (2011) Assessing the evidence for latitudinal gradients in plant defence and herbivory. Functional Ecology 25: 380–388. g 50. Thaxton JM, Cole TC, Cordell S, Cabin RJ, Sandquist DR, et al. (2010) Native species regeneration following ungulate exclusion and nonnative grass removal in a remnant Hawaiian dry forest. Pacific Science 64: 533–544. 21. Price JP, Clague DA (2002) How old is the Hawaiian biota? Geology and phylogeny suggest recent divergence. Proceedings of the Royal Society of London, Series B Biological Sciences 269: 2429–2435. 51. Freifelder RR, Vitousek PM, D’Antonio CM (1998) Microclimate change and effect on fire following forest-grass conversion in seasonally dry tropical woodland. Biotropica 30: 286–297. 22. Wagner WL, Herbst DR, Sohmer S (1999) Manual of the flowering plants of Hawaii. 2nd ed. Low-Diversity Hawaiian Forests Journal of Tropical Ecology 22: 481– 59. Chave J, Alonso D, Etienne RS (2006) Theoretical biology: Comparing models of species abundance. Nature 441: E1. gy 30. Slik JWF, Aiba S-I, Brearley FQ, Cannon CH, Forshed O, et al. (2010) Environmental correlates of tree biomass, basal area, wood specific gravity and stem density gradients in Borneo’s tropical forests. Global Ecology and Biogeography 19: 50–60. 60. Gotelli NJ, Colwell RK (2001) Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters 4: 379–391. 61. Chiarucci A, Enright NJ, Perry GLW, Miller BP, Lamont BB (2003) Performance of nonparametric species richness estimators in a high diversity plant community. Diversity and Distributions 9: 283–295. 31. Toledo M, Poorter L, Pen˜a-Claros M, Alarco´n A, Balca´zar J, et al. (2011) Climate and soil drive forest structure in Bolivian lowland forests. Journal of Tropical Ecology 27: 333–345. p y y 62. Colwell RK, Mao CX, Chang J (2004) Interpolating, extrapolating, and comparing incidence-based species accumulation curves Ecology 85: 2717– 2727. 32. Lugo AE, Scatena FN, Silver WL, Molina Colo´n S, Murphy PG (2002) Resilience of tropical wet and dry forests in Puerto Rico. In: Gunderson LH, Pritchard L Jr, editors. Resilience and the behavior of large-scale systems. Washington, DC: Island Press. pp. 195–225. 63. Magurran A (2004) Measuring biological diversity. Oxford, UK: Blackwell Publishing. 33. Bellingham PJ, Stewart GH, Allen RB (1999) Tree species richness and turnover throughout New Zealand forests. Journal of Vegetation Science 10: 825–832. 64. Sokal RR, Rohlf FJ (1994) Biometry: The principles and practice of statistics in biological research. 3rd edition. New York, New York: W.H. Freeman. 65. SAS Institute (2005) JMP introductory guide, release 6. Cary, NC: SAS Institute. 34. Sagar R, Singh JS (2006) Tree density, basal area and species diversity in a disturbed dry tropical forest of northern India: implications for conservation. Environmental Conservation 33: 256–262. 66. Asner GP, Hughes RF, Varga TA, Knapp DE, Kennedy-Bowdoin T (2009) Environmental and biotic controls over aboveground biomass throughout a tropical rain forest. Ecosystems 12: 261–278. 35. Stegen JC, Swenson NG, Enquist BJ, White EP, Phillips OL, et al. (2011) Variation in above-ground forest biomass across broad climatic gradients. Global Ecology and Biogeography: 744–754. 67. Brenes-Arguedas T, Roddy A, Coley P, Kursar T (2010) Do differences in understory light contribute to species distributions along a tropical rainfall gradient? Oecologia: 1–14. 36. References 7. Rosenzweig ML (1995) Species diversity in space and time. Cambridge, UK: Cambridge University Press. 8. Givnish TJ (1999) On the causes of gradients in tropical tree diversity. Journal of Ecology 87: 193–210. 2. Givnish TJ (1999) On the causes of gradients in tropical tree diversity. Journal of Ecology 87: 193–210. gy 9. Hubbell SP (2001) The unified neutral theory of biodiversity Princeton, New Jersey, USA: Princeton University Press. 9. Hubbell SP (2001) The unified neutral theory of biodiversity Princeton, New Jersey, USA: Princeton University Press. 3. Phillips O, Gentry AH, Hall P, Sawyer S, Vasquez R (1994) Dynamics and species richness of tropical rain forests. Proceedings of the National Academy of Sciences 91: 2805–2809. 10. Zimmerman JK, Thompson J, Brokaw N (2008) Large tropical forest dynamics plots: testing explanations for the maintenance of species diversity. Chichester, UK: Wiley-Blackwell. 4. Mittelbach GG, Schemske DW, Cornell HV, Allen AP, Brown JM, et al. (2007) Evolution and the latitudinal diversity gradient: speciation, extinction and biogeography. Ecology Letters 10: 315–331. 11. Hooper DU, Chapin FS, Ewel JJ, Hector A, Inchausti P, et al. (2005) Effects of biodiversity on ecosystem functioning: A consensus of current knowledge. Ecological Monographs 75: 3–35. 5. Hawkins BA, Rodrı´guez MA´ , Weller SG (2011) Global angiosperm family richness revisited: linking ecology and evolution to climate. Journal of Biogeography 38: 1253–1266. 6. Condit R (1998) Tropical forest densus plots: Methods and results from Barro Colorado Island, Panama and a comparison with other plots. Berlin, Germany, and Georgetown, Texas, USA: Springer-Verlag and R. G. Landes Company. 211 p. August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org 16 Low-Diversity Hawaiian Forests Low-Diversity Hawaiian Forests Honolulu, HI: Bishop Museum. p 52. Litton CM, Sandquist DR, Cordell S (2006) Effects of non-native grass invasion on aboveground carbon pools and tree population structure in a tropical dry forest of Hawaii. Forest Ecology and Management 231: 105–113. 23. Condit R (1995) Research in large, long-term tropical forest plots. Trends in Ecology and Evolution 10: 18–21. gy g 53. Condit R (1998) Tropical forest census plots. Berlin: Springer-Verlag, Berlin, Germany. gy 24. Connell JH, Green PT (2000) Seedling dynamics over thirty-two years in a tropical rain forest. Ecology 81: 568–584. p gy 25. Hobbie JE, Carpenter SR, Grimm NB, Gosz JR, Seastedt TR 54. Asner GP, Hughes RF, Mascaro J, Uowolo AL, Knapp DE, et al. (2011) High- resolution carbon mapping on the million-hectare island of Hawai‘i. Frontiers in Ecology and the Environment 9: 434–439. 25. Hobbie JE, Carpenter SR, Grimm NB, Gosz JR, Seastedt TR (2003) Th long term ecological research program. BioScience 53: 21–32. long term ecological research program. BioScience 53: 21–32. 26. Hubbell SP (2004) Two decades of research on the BCI Forest Dynamics Plot. In: Losos EC, Leigh EG Jr, editors. Tropical forest diversity and dynamism: Findings from a large-scale plot network. Chicago: University of Chicago Press. pp. 8–30. 55. Chave J, Andalo C, Brown S, Cairns MA, Chambers JQ, et al. (2005) Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia 145: 87–99. 56. Zanne AE, Lopez-Gonzalez G, Coomes DA, Ilic J, Jansen S, et al. (2009) Towards a worldwide wood economics spectrum. Dryad Digital Repository doi: 10.5061/dryad.5234. pp 27. Clinebell RR, Phillips OL, Gentry AH, Stark N, Zuuring H (1995) Prediction of neotropical tree and liana species richness from soil and climatic data. Biodiversity and Conservation 4: 56–90. 57. Chave J, Coomes D, Jansen S, Lewis SL, Swenson NG, et al. (2009) Towards a worldwide wood economics spectrum. Ecology Letters 12: 351–366. 28. Malhi Y, Wood D, Baker TR, Wright J, Phillips OL, et al. (2006) The regional variation of aboveground live biomass in old-growth Amazonian forests. Global Change Biology 12: 1107–1138. 58. Slik JWF (2006) Estimating species-specific wood density from the genus average in Indonesian trees. Journal of Tropical Ecology 22: 481–482. g gy 29. Saatchi SS, Houghton RA, Dos Santos Alvala´ RC, Soares JV, Yu Y (2007) Distribution of aboveground live biomass in the Amazon basin. Global Change Biology 13: 816–837. average in Indonesian trees. Low-Diversity Hawaiian Forests Low-Diversity Hawaiian Forests Dry Tropical Forests. Cambridge, UK: Cambridge University Press. pp. 326– 345. 90. Durand LZ, Goldstein G (2001) Growth, leaf characteristics, and spore production in native and invasive tree ferns in Hawaii. American Fern Journal 91: 25–35. 76. Vitousek P, Asner GP, Chadwick OA, Hotchkiss S (2009) Landscape-level variation in forest structure and biogeochemistry across a substrate age gradient in Hawaii. Ecology 90: 3074–3086. 91. Amatangelo KL, Vitousek PM (2009) Contrasting predictors of fern versus angiosperm decomposition in a common garden. Biotropica 41: 154–161. gy 77. Cole RJ, Litton CM, Koontz MJ, Loh RK (2012) Vegetation recovery 16 years after feral pig removal from a wet Hawaiian forest. Biotropica 44: 463–471. 92. Mack MC, D’Antonio CM, Ley RE (2001) Alteration of ecosystem nitrogen dynamics by exotic plants: A case study of C4 grasses in Hawaii. Ecological Applications 11: 1323–1335. 78. Inman-Narahari F, Ostertag R, Cordell S, Giardina CP, Nelson-Kaula K, et al. (2013) Seedling recruitment factors in low-diversity Hawaiian wet forest: towards global comparisons among tropical forests. Ecosphere 4: 24. http://dx. doi.org/10.1890/ES1812-00164.00161. 93. Huenneke LF, Vitousek PM (1989) Seedling and clonal recruitment of the invasive tree, Psidium cattleianum: implications for management of native Hawaiian forests. Biological Conservation 53: 199–211. 79. Busby PE, Vitousek PM, Dirzo R (2010) Prevalence of tree regeneration by sprouting and seeding along a rainfall gradient in Hawai’i. Biotropica 42: 80– 86. g 94. Hughes RF, Denslow JS (2005) Invasion by an N2-fixing tree, Falcataria moluccana, alters function, composition, and structure of wet lowland forests of Hawai‘i. Ecological Applications 15: 1615–1628. 80. Kammesheidt L (1999) Forest recovery by root suckers and above-ground sprouts after slash-and-burn agriculture, fire and logging in Paraguay and Venezuela. Journal of Tropical Ecology 15: 143–157. 95. Gravel D, Canham CD, Beaudet M, Messier C (2006) Reconciling niche and neutrality: The continuum hypothesis. Ecology Letters 9: 399–409. neutrality: The continuum hypothesis. Ecology Letters 9: 399–409 96. Hubbell SP, Foster RB (1986) Commonness and rarity in a neotropical forest: Implications for tropical tree conservation. In: Soule ME, editor. Conservation biology: The science of scarcity and diversity: Sunderland, Mass, USA Illus Paper Maps. Sunderland, Massachussetts: Sinauer Associates, Inc. pp. 205– 231. 81. Bellingham PJ, Sparrow AD (2009) Multi-stemmed trees in montane rain forests: their frequency and demography in relation to elevation, soil nutrients, and disturbance. Journal of Ecology 97: 472–483. J gy 82. Low-Diversity Hawaiian Forests Lehmann A, Leathwick JR, Overton JM (2002) Assessing New Zealand fern diversity from spatial predictions of species assemblages. Biodiversity and Conservation 11: 2217–2238. 97. Lach L (2003) Invasive ants: Unwanted partners in ant-plant interactions? Annals of the Missouri Botanical Garden 90: 91–108. 83. Coomes DA, Allen RB, Bentley WA, Burrows LE, Canham CD, et al. (2005) The hare, the tortoise and the crocodile: the ecology of angiosperm dominance, conifer persistence and fern filtering. Journal of Ecology 93: 918–935. 98. Nogueira SSdC, Nogueira-Filho SLG, Bassford M, Silvius K, Fragoso JMV (2007) Feral pigs in Hawai’i: Using behavior and ecology to refine control techniques. Appl Anim Behav Sci 108: 1–11. conifer persistence and fern filtering. Journal of Ecology 93: 918–9 84. Bellingham PJ, Richardson SJ (2006) Tree seedling growth and survival over 6 years across different microsites in a temperate rain forest. Canadian Journal of Forest Research 36: 910–918. q pp 99. Nogueira-Filho S, Nogueira S, Fragoso J (2009) Ecological impacts of feral pigs in the Hawaiian Islands. Biodiversity and Conservation 18: 3677–3683. y 100. Shiels A, Drake D (2010) Are introduced rats (Rattus rattus) both seed predators and dispersers in Hawaii? Biological Invasions: 1–12. 85. Tanner EVJ (1983) Leaf demography and growth of the tree-fern Cyathea pubescens Mett ex Kuhn in Jamaica Bot J Linn Soc 87: 213–227. 101. Shiels AB (2011) Frugivory by introduced black rats (Rattus rattus) promotes dispersal of invasive plant seeds. Biological Invasions 13: 781–792. 86. Bernabe N, Williams-Linera G, Palacios-Rios M (1999) Tree ferns in the interior and at the edge of a Mexican cloud forest remnant: Spore germination and sporophyte survival and establishment. Biotropica 31: 83–88. dispersal of invasive plant seeds. Biological Invasions 13: 781–7 102. Gilbert B, Lechowicz MJ (2004) Neutrality, niches, and dispersal in a temperate forest understory. Proceedings of the National Academy of Sciences of the United States of America 101: 7651–7656. 87. Arens NC (2001) Variation in performance of the tree fern Cyathea caracasana (Cyatheaceae) across a successional mosaic in an Andean cloud forest. American Journal of Botany 88: 545–551. 103. Hubbell SP (2005) Neutral theory in community ecology and the hypothesis of functional equivalence. Functional Ecology 19: 166–172. y 88. Williams-Linera G, Palacios-Rı´os M, Herna´ndez-Go´mez R (2005) Fern richness, tree species surrogacy and fragments complementarity in a Mexican tropical montane cloud forest. Biodiversity and Conservation 14: 119–133. 104. Low-Diversity Hawaiian Forests (2011) Variation in above-ground forest biomass across broad climatic gradients. Global Ecology and Biogeography 20: 744–754. J p Center and the Institute of Pacific Islands Forestry, USDA Forest Serv 44. Kellner JR, Asner GP (2009) Convergent structural responses of tropical forests to diverse disturbance regimes. Ecology Letters 12: 1–11. to diverse disturbance regimes. Ecology Letters 12: 1–11. gy g g p y 75. Martinez-Yrizar A (1995) Biomass distribution and primary productivity of tropical dry forests. In: Bullock SH, Mooney HA, Medina E, editors. Seasonally 45. Mueller-Dombois D (2000) Rain forest establishment and succession in the Hawaiian Islands. Landsc Urban Plann 51: 147–157. PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e103268 17 August 2014 \| Volume 9 \| Issue 8 \| e103268 Low-Diversity Hawaiian Forests Losos E, Leigh EG Jr, editors (2004) Tropical forest diversity and dynamism: Findings from a large-scale plot network. Chicago, Illinois: University of Chicago Press. p y 89. Jones MM, Olivas Rojas P, Tuomisto H, Clark DB (2007) Environmental and neighbourhood effects on tree fern distributions in a neotropical lowland rain forest. Journal of Vegetation Science 18: 13–24. PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e103268 PLOS ONE \| www.plosone.org 18
https://openalex.org/W4247911670	https://figshare.com/articles/thesis/Semi-automated_Extraction_of_Urban_Road_Networks_by_Geometric_Analysis_of_IKONOS_Imagery/14647335/1/files/28127508.pdf	English	null	Semi-automated Extraction of Urban Road Networks by Geometric Analysis of IKONOS Imagery	null	2,021	cc-by	26,641	Semi-automated Extraction of Urban Road Networks by Geometric Analysis of IKONOS Imagery A Thesis submitted to the School of Graduate Studies of Ryerson Universit in partial fulfillment of the requirements for the degree of Master of Applied Science in the Program of Civil Engineering Toronto, Ontario, Canada © 2003 Haibin Dong Toronto, Ontario, Canada © 2003 Haibin Dong RyERSONUKiVu-wHi-L.^ UMI Number: EC53444 UMI UMI Microform EC53444 Copyright2009by ProQuest LLC All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 UMI INFORMATION TO USERS The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. UMI ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 Haibin Dong I further authorize Ryerson University to reproduce this thesis by photocopying or by other means, in total or in part, at the request of other institutions or individuals for the purpose of scholarly research. Declaration I hereby declare that I am the sole author of this thesis. I authorize Ryerson University to lend this thesis to other institutions or individuals for the purpose of scholarly research. i Haibin Dong Department of Civil Engineering Ryerson University Haibin Dong Haibin Dong Department of Civil Engineering Ryerson University Department of Civil Engineering 11 Borrower’s page Ryerson University requires the signatures of all persons using or photocopying this thesis. Please sign helow, and give address and date. Name of Borrowers Date Address Signature Ill Abstract This thesis addresses the topic of semi-automated extraction of urban road networks from high-resolution satellite imagery. Research on this topic is mainly motivated by the use of geographic information systems in transportation (GIS-T), and the need for reliable data acquisition methods and to update GIS-T databases. To this end, 1-m spatial resolution IKONOS imagery provides a new data source to collect the spatial models of citywide road networks. In this thesis, a novel methodology of a semi-automated road extraction using high- resolution satellite imagery over urban areas is developed. The main objective of this research is to extract urban road networks from a single IKONOS image. To detect the road features from a highly complex scene, a multiscale analysis of the optimal image was performed. To extract roads and their networks, the knowledge of road geometry is exploited in an interactive environment. The key advantage of the developed method is the full employment of a human and a computer’s abilities for fast and precise road extraction from high-resolution satellite imagery. The results show that the presented method enables reliable road extraction over urban areas. The potential applications exemplified in case studies indicate that the high-resolution satellite imagery offers an efficient and precise source for geographic and transportation databases. Based on this research, the limitations and future work for the prototype system are discussed. IV Acknowledgements First of all, I would like to thank my supervisor Dr. Jonathan Li for his kind supervision and many good ideas. Special thanks for all fiuitful discussions, invaluable suggestions and critical remarks. I would also like to thank him for bringing me to the active research field of remote sensing for transportation. Many thanks go to Dr. Said M. Easa, my co supervisor, for his critical guidance and patience. His serious and significant research had a great influence on my work. I appreciate that he gave me initial research topic, ideas regarding extraction of various types of horizontal curves, transportation knowledge, and times of discussion. Their efforts to ensure my financial support during my study are also appreciated. I also attribute my accomplishments to Dr. Michael Chapman, Dr. Ahmed El-Rabbany, Dr. Songnian Li, and other faculty and staff members in the Department of Civil Engineering, for their encouragements, in particular, Desmond Rogan for kindly helping me many times to solve computer and network-related problems. Many thanks are also due to the department secretaries, especially Kim Kritzer and Dianne Mendonca, for their kindness and assistance. I would like to acknowledge all my fellow graduate student at Ryerson University, who have helped me in countless ways and made the period of study an enjoyable time, in particular, Artur Fidera, Hongmei Zhao, Jun Nie, Mohamed Abdalla, Mohammed El- Diasty, Wenglong He, Xiangqian Gao, Yinfeng Li, and Yu Li. The financial support partially provided by the School of Graduate Studies is gratefully acknowledged. Thanks also go to Anthony Sani of Spatial Geo-Link, Inc. in Toronto for his support with the latest version of ERDAS Imagine software and the Greater Toronto Airports Authority (GTAA) for providing the IKONOS satellite imagery used in my research. Last, but not least, I wish to express my unfathomable gratitude to my parents, for their understanding and their enduring patience during my studies. I could have never completed this thesis without their greatest support. VI List of Figures Figure 2.1 Overlooks and profiles of (a) step edge, (b) line edge, and (c) comer edge 8 Figure 3.1 Scheme for the proposed approach of extracting road networks...................... 24 Figure 3.2 Histograms of stretch methods............................................................................32 Figure 3.3 IKONOS Images of Kampsax, Denmark resampled at scale (a) 1:25,000, (b) 1:10,000, and (c) 1:2,500...............................................................................................37 Figure 3.4 Modulus maxima of continuous WT of ID signal,........................................... 39 Figure 3.5 Four test directions for the Canny algorithm......................................................41 Figure 3.7 Classical image Circle for the wavelet edge detection...................................... 47 Figure 3.9 Geometric properties of a circle.................................................................. 52 Figure 3.10 Schematic representation for establishing a simple horizontal curve.............53 Figure 3.11 Schematic representation for establishing a reverse horizontal curve............56 Figure 3.12 Schematic representation for establishing a compound horizontal curve 56 Figure 3.13 Flow diagram for establishing reverse and compound horizontal curves 59 Figure 3.16 Detection of double roadsides........................................................................... 63 Figure 4.1 SDS(a = 2) enhanced image of study area 1...................................................... 68 Figure 4.2 SDS(a = 2) enhanced image of study area 2 ......................................................69 Figure 4.3 Original image of study area 1 without any enhancement.................................71 Figure 4.4 Histogram of original image................................................................................72 Figure 4.5 Histogram of the SDS-enhanced image..............................................................72 Figure 4.6 SDS-DS-enhanced image of study area 1 depicted in Figure 4.3.....................74 Figure 4.7 Histogram after DS-enhanced.............................................................................74 Figure 4.8 SDS-PCA enhanced image of study area 1 depicted in Figure 4.3...................76 Figure 4.9 DS-PCA-enhanced image depicted in Figure 4.3...............................................77 Figure 4.10 The difference image of the PCA image and DS-PCA image...................... 78 Figure 4.11 Edge image of the image of study area 1 (Figure 4.9) at wavelet scale 1.......79 Figure 4.12 Edge image of the image of study area 1 (Figure 4.9) at wavelet scale 2.......80 Figure 4.13 Edge image of the image of study area 1 (Figure 4.9) at wavelet scale 3.......81 Figure 4.14 Results of establishing horizontal curves at complex freeway interchange.. 82 Figure 4.15 Results of establishing reverse horizontal curves in urban residential area.. 83 Figure 4.16 Results of establishing compound horizontal curves in urban highway area.84 Figure 4.17 Results of establishing spiral horizontal curves in urban highway area 85 Figure 4.18 Segmented freeway interchange in the image of study area 1........................ 87 Figure 4.19 Segmented road network in the image of study area 2.................................... 88 Figure 4.20 Vectorized freeway interchange in the image of study area 1.........................89 Figure 4.21 Vectorized road network in the image of study area 2.................................... Table of Contents Declaration.............................................................................................................................. ü Borrower’s page.....................................................................................................................in Abstract...................................................................................................................................iv Acknowledgements.................................................................................................................v Table of Contents..................................................................................................................vii List of Figures......................................................................................................................... x List of Tables......................................................................................................................... xi List of Abbreviations............................................................................................................xii 1 INTRODUCTION.........................................................................................................1 1.1 Motivation.....................................................................................................................1 1.2 Problem Statement.......................................................................................................3 1.3 Research Objectives..................... 4 1.4 Thesis Organization.....................................................................................................5 2 LITERATURE REVIEW............................................................................................6 2.1 Multiscale Edge Detection.......................................................................................... 7 2.2 Parametric Feature Detection.....................................................................................10 2.3 Automatic Methods.................................................................................................... 13 2.3.1 Use of Single Image...........................................................................................13 2.3.2 Use of Multi source Data....................................................................................14 2.3.3 Use of Multiscale Images...................................................................................15 2.3.4 Use of Combination of Multisource Datasets and Multiscale Images 16 2.4 Semi-automatic Methods......................................................................................... 16 2.4.1 Active Tracing....................................................................................................17 2.4.2 Snakes................................................................................................................17 2.4.3 Template Matching............................................................................................ 18 2.4.4 Classification.......................................................................................................18 2.5 Summary of Existing Methods................................................................................19 vn 3 METHODOLOGY...................................................................................................... 21 3.1 Input Data Description............................................................................................... 21 3.2 General Strategy for Semi-automated Road Extraction........................................... 23 3.3 Image Preprocessing...................................................................................................25 3.3.1 Interpretation of Image Data............................................................................. 25 3.3.2 Standard Deviation Stretch................................................................................30 3.3.3 Decorrelation Stretch.........................................................................................34 3.4 Wavelet Edge Detection............................................................................................ 35 3.4.1 Frequency Analysis............................................................................................35 3.4.2 Time-Scale Analysis..........................................................................................35 3.4.3 Dyadic Wavelet Transform................................................................................38 3.4.4 Regularity Analysis............................................................................................38 3.4.5 Multiscale Edge Detection.................................................................................38 3.5 Geometric Feature Detection..................................................................................... 47 3.5.1 Extraction of Simple Horizontal Circular Curves............................................49 3.5.2 Extraction of Reverse and Compound Curves................................................. 55 3.5.3 Extraction of Spiral Curves................................................................................58 3.5.4 Extracting Second Roadside..............................................................................62 3.6 Connection to Desktop GIS Environment..................................................................64 4 CASE STUDIES............................................................................................................67 4.1 Study Areas................................................................................................................. 67 4.2 Image Preprocessing...................................................................................................70 4.2.1 Standard Deviation Stretch................................................................................70 4.2.2 Decorrelation Stretch.........................................................................................73 4.2.3 Principal Components Analysis........................................................................ 73 4.2.4 Results and Discussion......................................................................................75 4.3 Wavelet Edge Detection............................................................................................ 75 4.4 Extraction of Road Curves......................................................................................... 80 4.4.1 Extraction of Simple CircularCurves................................................................81 Vlll 4.4.3 Extraction of Spiral Curves..............................................................................84 4.4.4 Results and Discussion....................................................................................85 4.5 Extraction of Road Networks....................................................................... 86 4.6 Performance Evaluation............................................................................................ 88 5 CONCLUSIONS AND RECOMMENDATIONS..................................................97 5.1 Summary.................................................................................................................... 97 5.2 Limitations.................................................................................................................99 5.3 Conclusions.............................................................................................................. 100 5.4 Recommendations for Further Research............................................................... 102 BIBLIOGRAPHY................................................................................................................103 IX List of Tables Table 3.1 Specifications of IKONOS imagery....................................................................22 Table 3.2 Weights of three conversion methods from RGB to gray-scale.........................27 Table 4.1 Sample image information...................................................................................70 List of Tables Table 3.1 Specifications of IKONOS imagery....................................................................22 Table 3.2 Weights of three conversion methods from RGB to gray-scale.........................27 Table 4.1 Sample image information...................................................................................70 List of Tables List of Figures 90 Figure 4.22 Extracted road centerlines using the method based on FMM (Li et al., 2003). ......................................................................................................................................... 92 Figure 4.23 Overlay of the extracted highways on the input image of study area..1........ 95 Fi 4 24 O l f th t t d d t k th i t i f t d 2 96 List of Tables List of Tables Table 3.1 Specifications of IKONOS imagery....................................................................22 Table 3.2 Weights of three conversion methods from RGB to gray-scale.........................27 Table 4.1 Sample image information...................................................................................70 Table 3.1 Specifications of IKONOS imagery....................................................................22 Table 3.2 Weights of three conversion methods from RGB to gray-scale.........................27 Table 4.1 Sample image information...................................................................................70 XI List of Abbreviations ID..........................................................One-Dimensional 2D..........................................................T wo-Dimensional 3D..........................................................Three-Dimensional 4D..........................................................Four-Dimensional ANN......................................................Artificial Neural Network DEM..................................................... Digital Elevation Model DN........................................................ Digital Number DS.........................................................Decorrelation Stretch DSM..................................................... Digital Surface Model DWT......................................................Dyadic Wavelet Transform FMM..................................................... Fuzzy Mathematical Morphology GIS........................................................ Geographic Information Systems GIS-T.................................................... Geographic Information Systems in Transportation HT......................................................... Hough Transform INSAR.................................................. Interferometric SAR PCA......................................................... Principal Component Analysis SAR.......................................................Synthetic Aperture Radar SDS.......................................................Standard Deviation Stretch WT........................................................ Wavelet Transform 1.1 Motivation From a practical point of view, research on automatic road extraction in urban areas is mainly motivated by the use of geographic information systems in transportation (GIS-T) and the need for data acquisition and update for GIS-T databases. Applications of road data of urban areas include analyses and simulations of traffic flow, estimation of air and noise pollution, street maintenance, and so on. From a scientific perspective, the extraction of roads in complex urban environments is one of the challenging issues in remote sensing and photogrammetric computer vision, since many tasks related to automatic scene interpretation are involved. Factors greatly influencing the scene complexity are, for instance, the number of different objects, the amount of their interrelations, and the variability of both. Moreover, each factor - and thus the scene complexity - is related to a particular scale. Although traditional aerial photographs can provide maximum detail and precision, satellite imagery has already greatly benefited areas, such as functional planning of highway and railway networks. Since 2000, high-resolution satellite imagery has been used in various mapping and engineering applications. Moreover, satellite imagery over one area can be recaptured continuously over days or weeks. Satellite remote sensing is a fast and cost-effective way to collect digital images of road networks over a considerably large urban area. The new generation of high-resolution imaging satellites (e.g., IKONOS, QuickBird) have presented more details and precision on urban man-made features (e.g., roads, buildings). For example, roads appear to be elongated regions in Im resolution IKONOS imagery rather than linear features in 30m resolution Landsat TM imagery. Nevertheless, extract useful spatial object information (e.g., buildings, roads, and vegetation) is a difficult task. Although human beings are good at abstracting useful information they need, the accuracy is lost during the vectorization due to mistakes in human interaction (e.g., misshaping and misplacement). In general, man-made objects have regularities in size, shape, and surface. These regularities facilitate the extraction of man-made objects from remote sensing imagery. A good survey of existing approaches can be found in Zhang (2003). The geometry of a road is a good example of this regularity. Thus far, many approaches have been developed to extract roads from aerial and satellite imagery. These approaches are based on either spatial, spectral, or both characteristics of the road. In low-resolution satellite imagery (e.g., Landsat TM, SPOT HRV), roads appear as linear or curvilinear features, to which some contour finding methods can be applied. In medium-resolution satellite imagery (e.g., 1RS-1 C/ID and MOMS-2), these roads appear as nearly homogenous elongated regions. In high-resolution satellite imagery (e.g., IKONOS and QuickBird), the roads appear as elongated regions with more details (e.g., traffic markings, manholes, curbs, cars, and shadows). The selection of extraction approaches should correspond with the level of image resolution. So far, many existing approaches, working on high- resolution satellite imagery, are actually based on the algorithms developed for lower resolution satellite imagery following a downgrading preprocessing to reduce the image spatial resolution. This compromises the accuracy of the results and the full exploitation of high-resolution data. Extracting roads and their networks from high-resolution satellite imagery and taking advantage of most data spectra should be a new research topic in remote sensing. 1.2 Problem Statement In most cases, users prefer to work with a single image to extract all desired information. The new generation of high-resolution satellites can provide stable and economical high- resolution image source. However, in high-resolution satellite imagery, detailed urban features (e.g., traffic markings, manholes, curbs, cars, and shadows) make urban road extraction more difficult. Fully automated methods have proven to be more sensitive to the details than semi-automated methods considering the reliability and accuracy of the results. As such, semi-automated methods may be more operational. Most urban highways are designed and constructed using relevant policies of highway geometric design. With high-resolution satellite imagery, reestablishing geometric design parameters and reconstructing road networks from imagery are possible. One of the major responsibilities of highway survey for existing highways is to reestablish the centerlines of highways, of which curve segments are much more complex than line segments. The curved highway segments can be either horizontal or vertical. However, most remote sensing images are overhead, or nearly overhead, snapshots of views parallel to the plane of local Earth’s surface. The vertical alignment of the road is unclear without auxiliary information, such as topographie data. However, compared to horizontal alignment, the scale of the vertical alignment eould be ignored especially in most urban areas. To maximize the retrieval of useful information, and suppress the unwanted or redundant data, three major image-processing techniques will be studied and applied. They are radiometric, spectral, and spatial enhancement. information, such as topographie data. However, compared to horizontal alignment, the scale of the vertical alignment eould be ignored especially in most urban areas. To maximize the retrieval of useful information, and suppress the unwanted or redundant data, three major image-processing techniques will be studied and applied. They are radiometric, spectral, and spatial enhancement. 1.3 Research Objectives In this study, the road and its network extraction from high-resolution satellite imagery is the focus. The applications of remotely sensed data and analysis in geomatics and transportation areas are also illustrated. By fully exploiting high-resolution satellite image data, the ability of abstraction of human, and the high-accuracy and high-performance of digital processing technology, the objective of this study is to develop a semi-automatic road extraction method that can generate high accuracy data source for GIS-T databases. Specifically, the following tasks will be accomplished in this research: (1) To create a single-band image from a single true-colour IKONOS image by image preprocesing. (2) To generate an edge image appropriate for desired feature extraction from the single-band image using multiscale image analysis. (3) To extract the road networks by geometric analysis of linear and curve road edges presented in the edge image. (4) To approximate the road networks for the applications in such areas as remote sensing and GIS using the extracted road geometric parameters. 1.4 Thesis Organization This thesis consists of five chapters. After this introduction, Chapter 2 introduces the basic concepts and describes previous related work, including semi-automatic and automatic road extraction methods, and the key techniques of image processing and analysis in remote sensing. Chapter 3 presents a detailed description of the methodologies developed and used in this thesis. In general, these methodologies are grouped into data collection, image pre-processing, wavelet multiscale analysis, feature extraction, and data integration, respectively. Chapter 4 includes case studies of the methodologies on IKONOS satellite imagery. The experimental results and system performance evaluation are presented. In Chapter 5, the summary and conclusions are presented based on the results of this research work followed by some recommendations for further research. 2 LITERATURE REVIEW The extraction of urban road networks from remotely sensed images has drawn considerable attention lately. The existing approaches use a wide variety of strategies using different resolution aerial or satellite images. Semi-automated schemes require human interaction to provide interactively some information to control the extraction. Roads are then extracted by automatic algorithms. Automatic methods usually extract reliable hypotheses for road segments through edge and line detection and then establish connections between road segments to form road networks. This chapter tries to review several existing techniques. As a complete road extraction methodology, low-level to high-level procedures are to be followed. These procedures are to find road pixels, road segments (road pixel group), road (road segment group), and road networks (road group), consecutively. Section 2.1 reviews the multi scale edge detection techniques that are useful for finding road pixels in imagery. Section 2.2 reviews the parametric feature detection techniques that are useful for finding of road segments. The automated and semi-automated methods dedicated to perform partial or complete road extraction procedures are reviewed and summarized in Sections 2.3,2.4 and 2.5, respectively. 2.1 Multiscale Edge Detection An edge could be defined as points where the image intensity has sharp transitions (Mallat, 1998). Edges are the most important elements for identifying the feature by the human vision system. This can be explained by that when drawing a picture, the first thing to draw is the profile of targeted object, then, on which, colours and/or textures are added to further describe more properties. In most cases, the edges already tell what an object is or what an object looks like. For a computer vision system, many analytical methods are performed on the edge image generated rather than directly on the raw image. The edge image can depict the structural and morphological properties of important features abstracted from raw imagery. The edge detection operation is important because it is used to locate where the roads might be. A survey on edge detection techniques can be found in Ziou and Tabbone (1997). According to them, the most common types of image intensity variations are steps, lines and comers (see Figure 2.1); and the majority of existing edge detection algorithms are applied to step edges which are the most common edges in images. In low-resolution satellite images, a road detection algorithm is used to find the line edge, while in high-resolution satellite images where a road segment is depicted as elongated homogeneous area, it is used to find the two parallel step edges. (a) / / I .%<>’ # : sr.i % # (b) (c) Figure 2.1 Overlooks and profiles of (a) step edge, (b) line edge, and (c) comer edge. / / I .%<>’ # : sr.i % # (b) (c) (a) (a) (b) (c) Figure 2.1 Overlooks and profiles of (a) step edge, (b) line edge, and (c) comer edge. It is a fact that one edge detector with one scale cannot identify all edges in an image (Ziou and Koukam, 1995). The multi-scale edge detection offers a solution for using only one detector with multiple scales (Ziou and Tabbone, 1997). A step towards multiscale image analysis techniques was motivated by the need for a preprocessing operation to regularize the differentiation operators (Torre and Poggio, 1980). Canny (1986) proposed a method that determines an “optimal smoothing filter” for multiscale detection of step edges. He pointed out that the smoothing filter could be approximated by the first-order derivative of a Gaussian kernel. He also introduced the concept of non-maxima suppression and hysteresis threshold. Much of the research for the development of wavelets was carried out in the thirties of 20*’ century, it was only in recent years that wavelet applications have been widely found. The two-dimensional (2D) multiscale discrete wavelet transform (WT) analysis and associated fast filter bank algorithm were first founded in Mallat (1989). Mallat (1992a) argued that the fundamental weakness of the classical edge detectors is that they create binary edge images and lose most image information and create instabilities in further processing. Although the wavelet approach is equivalent to the Canny edge detector that is based on the gradient analysis on imagery, it manipulates the edges at multiple scales, and integrates the information at these scales to locate edges precisely, and to reduce noise simultaneously. Most edge detectors perform poor at comers, where two edges intersect. Chen et al. (1995) proposed a WT-based comer indicator that locates the comer points based on the orientation variance. Multiscale image analysis is based on either intensity or edge images. According to the summary in Hay et al. (2003), multiscale image analysis is composed of two steps: (1) the multiscale representations of imageiy and (2) information extraction. Mayer (1996a) created multiscale images with different smoothness kemel fi"om the source image, such that abstraction of desired features is to be found on appropriately scaled image. Mayer and Steger (1996b) introduced a concept of multi-abstraction that is another representation space. An abstraction is defined as the increase of the level of simplification and emphasis. Substmcture (minor stmctures) at each level should be neglected. Mayer and Steger (1998c) used their method to detect curvilinear stmctures in image. The idea of 2D spatial pattem is considered that the signal in an image is the composite of pattems at different scale superimposed on each other. The large-scale pattems, such as land and sea, can be regard as background, on which small-scale pattems, such as buildings and roads, are added. Hoffinann (2001) used multi-scale IKONOS images to establish a hierarchical network of urban features based on the classification. pattems, such as land and sea, can be regard as background, on which small-scale pattems, such as buildings and roads, are added. Hoffinann (2001) used multi-scale IKONOS images to establish a hierarchical network of urban features based on the classification. Although multiscale edge detection is an ideal way to achieve desired edges, the selection of the appropriate scale is still dependent on human interpretation, which is one of the main manual operations. However, several efforts have been done to do it automatically. Lindeberg (1998) introduced the concept of seal e-space edge, in which an automatic mechanism for selection of scale levels is applied in detecting one-dimensional (ID) image features, such as edges and ridges. The strength of edge response was the measure derived from third-order derivative in the gradient direction. Zhang and Bao (2002) reported a better resulting edge map when using a multiplication of adjacent scale edge images, such that the noise is further suppressed and the dislocation is improved. Jung and Scharcanski (2003) separated noise and edges with adaptive criteria in scale-space using the WT. Their results show that WT is capable of suppressing noise without the priori knowledge of noise level while still keeping sharp edges. 2.2 Parametric Feature Detection The edges detected by edge detectors are simply composed of the pixels considered as belonging to an edge based upon same definition. There is no intemal spatial relationship 10 among these edge pixels in an edge image. Finding and describing the relationship of these pixels with parameters are the tasks of parametric feature detection. among these edge pixels in an edge image. Finding and describing the relationship of these pixels with parameters are the tasks of parametric feature detection. The pixels representing for edges of features can be linked using a simple contour-tracing algorithm (Gonzalez and Woods, 1993). They are encoded with the Freeman chain model (Freeman^ 1975) and saved in a Region Search Map with which structural and morphological information about the regions could be retrieved for further applications. In this study, roads are of particular features of interest to be detected. In low-resolution satellite imagery, roads are at most only a few pixels wide and appear as light or dark lines. In high-resolution satellite imagery, roads are modeled as bright homogeneous regions bordered by parallel edges (Baumgartner et al., 1996), in which roads can be approximately treated as polygons. Steger (1996) extracted the curved lines in aerial images by computing the Taylor polynomial approximation, and then determined the width of them. His solution still cannot give the parameters of the curves although the line can be given to subpixel accuracy. Dong et al. (1996) did some similar work in edge detection using WT, and edge orientation recognition using a statistical method, but on Landsat TM and SPOT HRV images. Ji (1996) applied dyadic WT to delineate the agricultural field boundaries from Landsat TM imagery. As the author inferred, his method was area-dependent and did not work well over urban areas. 11 Rosin and West (1989) presented a method to segment edges into lines and arcs. Their method requires two initial points at the ends of a chain of points to detect the arcs on this chain. Keller et al. (1995) introdueed a multiscale method for the detection of curvilinear structures in 2D image data, which is more sensitive to linear edges, while suppressed in curve seen from their sample results. Dori and Liu (1998) developed a vector-based arc segmentation method that computes the center of a potential arc. They categorized the arc detection methods into two main families. 2.3 Automatic Methods Automated methods are not autonomous methods. There can be human interactions before and after computer-based phases. Many of them have been developed. All of them try to utilize as much information as possible from available data to accomplish the goal automatically. Thus, the methodologies are categorized by the strategy of how they utilize the dataset accordingly. They work with either (1) single image, or (2) multisource data, or (3) multiscale images, or (4) a combination of (2) and (3). 2.2 Parametric Feature Detection The methods in the first family are based on the Hough transform (HT) and work directly on the original pixels of the images. This family of methods proves to be quite robust in the presence of noise, but requires quite intensive computation. The second family works on the chains of points or the polygonal approximation of such chains. The curvature of these chains can be estimated. Dosch et al. (2000) surveyed the two above-mentioned methods for arc and circle detection, and presented an improved method inspired by surveyed methods. However, these three methods were designed to work well under ideal condition, such as in a graphics document. They are insufficient to deal with complex context of features and noise in remotely sensed imagery. Mayer (1998b) described a way of detecting curvilinear structures in image which convolves image with second partial derivatives and finds the maximum direction as the normal directions of a curve. Baker (1998) explored the approach to detecting the features through multi-cues, including steps, lines and comers. Carreira et al. (2002) used WT edge detector and four 12 dimensional (4D) HT to detect and classify perceptual primitives on close-range photographs. dimensional (4D) HT to detect and classify perceptual primitives on close-range photographs. 2.3.1 Use of Single Image To reduce system errors and cost, many existing methods tried to extract the road from a single image. A road extraction system RoadF, developed and introduced by Zlotnick and Gamine (1993), was based on geometric constraints (parallel edge pairs of road) for automatic road finding, tracking, and linking from aerial images. Trinder and Wang (1998) presented a method, based on Marr’s theory of vision (Marr 1982) and human knowledge, of automatic road extraction from aerial images. Three processing levels were involved, including parallel edge pair detection, linking and formation of road stmcture with knowledge of geometric and radiometric properties express as mles in Prolog, and recognition of roads. Xiong (2001) introduced an optimization-based method for extracting road networks. It is a method involved dynamic programming and 13 supervised classification. Doucette et al. (2001) used a neural network clustering method to extract the centerlines and to construct self-organized road networks on high-resolution classified aerial image. Li et al. (2003) presented an automated approach to urban road network extraction based on fuzzy mathematical morphology (FMM), the performance of which was easily affected by the complexity of context. The dislocations of road centerlines were obvious near the shadows or road intersections. 2.3.2 Use of Multisource Data If a single data source is insufficient to retrieve the desired information or to achieve required accuracy, use of multisource data is one solution. For the purpose of road extraction, most methods use more than a single image, such as multiple images or another form of ancillary data. Barzohar and Cooper (1996) presented an approach built on geometric- probabilistic models to find main roads in aerial images. A map was needed for their road estimation method. Steger et al. (1997) presented an approach of completion of road networks from aerial images, using a graph representation of the road network and Digital Surface Model (DSM) data. Ruskoné and Airault (1997) designed an approach, based on its automatic seeding, of detection of road networks from aerial images. It can, reportedly, extract 40% of the road networks. Klang (1998) used a knowledge source from road databases to discriminate roads from rivers, railways or other lines in the image, and the roads in the database were employed to initialize snake seeds. Fiset et al. (1998) developed a map-image matching method that uses artificial neural network (ANN) on 14 SPOT imagery and rasterised old maps to find the updated segments of the road. Zhang et al. (2001) presented a knowledge-based system for automatic extraction of 3D roads from stereo aerial images. In contrast to other approaches, the developed system integrates processing of colour image data and information from spatial databases, extracts and fuses multi-cues, takes into account context information, employs existing knowledge, rules, and models, and treats each road accordingly. 2.3.4 Use of Combination of Multisource Datasets and Multiscale Images Some newly developed methods exploit the advantages of use of both multisource datasets and multiscale images. Hinz et al. (1999) used a multiscale strategy to find initial hypotheses in rural area, and digital elevation model (DEM) data and road markings were used to help identify urban roads. Hellwich and Wiedemann (2000) used fusion technology on multisource data to extract the road network in rural areas. These data include the Im-resolution panchromatic aerial photographs, hyperspectral imagery, interferometric synthetic aperture radar (InSAR) data, and DEM. Continuing the work in (Hinz et al., 1999), Hinz and Baumgartner (2003) integrated high-resolution aerial imagery, DSM and context model to extract road networks over urban areas. 2.3.3 Use of Multiscale Images Generally, a method used to extract roads is only appropriate for an image with a certain scale level. Different scale levels can be modeled and interpreted differently and accordingly. Multiscale images derived from a single image provide an ancillary means to analyze this image. Heipke et al. (1995) used multiresolution analysis applied on aerial imagery to construct a hierarchical approach of detect roads in rural areas. Baumgartner et al. (1997) presented a multiresolution approach for automatic extraction of roads from aerial imagery. This approach is based on the extraction of edges in an aerial image and the extraction of lines in a resolution-reduced image. Baumgartner et al. (1999a) reported a three-module scheme for automatic road extraction in rural areas. The three modules are (a) multiscale detection of most part of a road network, (b) global grouping by the network characteristics of roads, and (c) completion of the road network by analysis of path length. Baumgartner et al. (1999b) made use of versions of an aerial image with different resolutions to model and detect a network of intersections and links of the rural roads. 15 Amini et al. (2002) used a morphological algorithm to segment and convert an image into binary one, whose resolution was reduced by WT to extract the skeleton of roads. The final roadsides are found by the searching roadside algorithm using the original image. 2.4 Semi-automatic Methods Semi-automatic methods integrate manual operations with automatic modules in the workflow. These methods are categorized into four kinds: (1) active tracing, (2) active contour or snakes, (3) template matching, and (4) classification. 16 2.4.1 Active Tracing This category of methods tries to trace the road extension by starting from a user defined point and/or road direction. McKeown and Denlinger (1988) developed a method that requires a start point and direction to trace a linked road in aerial imagery. Vosselman and Knecht (1995) also used tracing and a Kalman filter by the user-specified start point and direction. Given a starting point and direction, Geman and Jedynak (1996) traced the road network using “active testing” technique on a medium-resolution satellite imagery where road can be searched as a ID structure. Shukla et al. (2002) used a Canny edge detector to generate thin feature edges, with which the path-following technique is performed at the direction determined by a cost minimization technique. 2.4.3 Template Matching Some methods regard a road as the concatenation of road segments with certain geometric properties. According to the geometric properties of a road segment, a template is designed and used to find the matched road segments in the image. Park and Kim (2001) presented a template-matching algorithm for road extraction fi’om 1-m IKONOS imagery. Given a road seed on the centerline, the algorithm used an adaptive least- squares method to compare the template and window on image. The shortcoming of the algorithm is that it is sensitive to the position of the seed on the centerline, the presence of shadows, and the curvature of road segments. Dal Poz and Silva (2002) presented a semi-automated method for extracting road segments and centrelines from medium- and high-resolution images based on both active template testing and edge analysis. 2.4.2 Snakes This category requires the seeds, set by a user, near the interested road to outline the road contour. Gruen and Li (1995a) applied WT in a SPOT image to sharpen the edges, where seed points were given by the user, then road tracking and linking are performed automatically and optimization is done by dynamic programming. Gruen and Li (1995b) designed a scheme based on either dynamic programming or least-squares B-spline snakes. In this semi-automatic technique, roads on aerial images, with many details and other objects distributed, cannot be processed appropriately. Gruen and Li (1997) used 3D least-squares B-spline snakes to detect boundaries of the roads in rural areas using aerial images. An approach based on multiscale detection, and geometry-constrained edge extraction using a snake algorithm was proposed and advanced by Mayer et al. 17 (1998) and Laptev et al. (2000), in which down-sampling to a coarse-scale image was performed in order to find edge points which are then used as snake seeds. Auclair et al. (2001) worked with active contour, or “snakes” approach, but automates the seeding procedure by deriving existing topographic database. (1998) and Laptev et al. (2000), in which down-sampling to a coarse-scale image was performed in order to find edge points which are then used as snake seeds. Auclair et al. (2001) worked with active contour, or “snakes” approach, but automates the seeding procedure by deriving existing topographic database. 2.4.4 Classification Classification methods utilize the spectral characteristics of roads in the image. The results are produced by a certain classification method, which divides the image into road areas and non-road areas. Chiesa (2001) developed a software package, called Veridian System that works with classification and extraction of transportation features in Landsat 18 TM imagery. However, this system does not offer additional benefit over manual extraction and vectorization as admitted by the author. TM imagery. However, this system does not offer additional benefit over manual extraction and vectorization as admitted by the author. 2.5 Summary of Existing Methods Road data are important for many applications. However, manual extraction of road networks is time-consuming and sometimes inaccurate. That is why many automated and semi-automated extraction approaches have been developed. The road segment in high- resolution satellite imagery of urban areas is susceptible to traffic markings, vehicles, sidewalks, curbs, shadows, etc. Methods based on tracing consider the parallel edges of road, and/or the profile of the road cross-section. Methods, like dynamic programming or snakes-based, were only designed for low-resolution imagery where roads appear to be linear features, and results are error-prone because the algorithms merely count on the local intensity of image if disturbances are present. Multi-images or auxiliary data can assist in the extraction on the road features or even 3D information of the roads from 2D data. However, in flat urban areas, it is not preferable considering the availability, cost, and accuracy of the extra data, and aspects of efficiency. Among the existing road extraction approaches, most of them work well on roads with relative simplicity in rural areas. Most of them use aerial imagery. Most of them assume that roads are linear or rectangle features on imagery, and regard implicitly the curve segment as the links of piecewise straight segments. Those studies on curve features 19 detection only work with ideal simple image (e.g., like graphic documents) or small-scale curve (e.g., the circle/arc whose diameter radius is less than the dimension of the image). Many algorithms merely generate results as the group of pixels belonging to the road or its centerline, not the higher-level description for each road. Most of those algorithms, aiming at detection of high-level semantic objects, simply model road segments as lines or rectangular shapes, even along the curved segments. Those methods, considering a curve as higher order curvilinear feature rather than linked linear pieces, have to rely on the either active tracking or active contour (snakes) to approximate the curves. They are inaccurate and greatly sensitive to the disturbing objects and noise. Many methods are designed to extract the roads with the characteristics of elongated linear or rectangle features. Nevertheless, some natural features (e.g., rivers) and man- made features (e.g., open ditches, running tracks) also satisfy this model. For practical applications, there should be involvement with human knowledge and interaction, more or less, integrated with automated steps. 2.5 Summary of Existing Methods From the summary of the literature review, this study focuses on the use of a single high- resolution satellite imagery, employs a multiscale analysis technique, and is based on the geometric curvilinear feature detection for semi-automated road extraction. Chapter 3 and 4 will give the details on the new methodology and its implementation with two case studies. 20 3 METHODOLOGY This chapter describes a semi-automated knowledge-based approach of road network extraction. Section 3.1 introduces some specifieations of high-resolution IKONOS imagery used in this study. Section 3.2 gives a brief deseription of the concept and procedures of the road extraction method. To assist further edge detection, some image preprocessing is used and discussed in Section 3.3. The multiscale edge detection based on wavelet transform is described in Section 3.4. The parametric feature detection methods for extracting the road network are presented in Section 3.5. Finally, Section 3.6 describes some methods that implement some useful applications for the proposed approach. 3.1 Input Data Description The advent, over the last few years, of the third generation of high spatial resolution imaging satellites (e.g., 1-m IKONOS, 60-cm QuickBird) could stimulate the development of remote sensing of urban areas further. The data produced by these sensors facilitate superior discrimination of the urban features (Donnay et al., 2001). Many areas appear to be spectrally heterogeneous in images from the new, finer spatial resolution sensors. This implicates the more inherent spatial complexity of urban scenes, and the potential to extract urban features. 21 Space Imaging’s IKONOS satellite is the world's first commercial satellite to collect panchromatic images with 1 -m resolution and multispectral images with 4-m resolution. It orbits the Earth every 98 minutes at an altitude of approximate 680 km. IKONOS was launched into a sun-synchronous orbit, passing a given longitude at about the same local time (10:30 A.M.) daily. IKONOS can produce 1-m imagery of the same location on the Earth every 3 days. Table 3.1 lists the main spectral and spatial characteristics of the IKONOS sensor. Using data fusion technology, these bands, with 1-m and 4-m spatial resolution listed in Table 3.1, can be transformed into 1-m resolution multispectral imagery, such as the true colour IKONOS image used in this study. Fundamentally, the proposed approach works with a single-band image, to which the IKONOS multispectral image should be converted before the application. Road network extraction from IKONOS imagery has significant applicability in transportation. It is a means of creating and updating GIS-T databases, upon which abundant information could be utilized or analyzed for transportation purposes, such as traffic management, infrastructure pi arming, road safety analysis, and route guidance. The research work required the orthophotos not only for the integration with GIS databases, but also for precise geometric primitives to achieve good results. The linear and arc features are the most important parametric road segments in the image, in which each location of pixel should be linked to the real object coordinates. 22 Table 3.3 Specifications of IKONOS imagery. Band Spectral Range (Mm) Spatial Resolution (m) Radiometric Resolution (bit) Temporal Resolution (day) Blue 0.45-0.52 1 Green 0.51-0.60 Red 0.63-0.70 Near-Infrared 0.76-0.85 Panchromatic 0.45-0.90 3.2 General Strategy for Semi-automated Road Extraction Using multispectral IKONOS imagery, the urban environment is one of the major problems preventing conventional spectral analysis. The high feature density of the urban scene causes problems, like shadows and mixed-pixel effects. Although conventional multispectral classification methods can be applied successfully to rural areas, these methods are error-prone for applications in urban areas. In this section, a novel strategy for semi-automated road extraction is proposed. The strategy is illustrated in Figure 3.1. The main objective is to derive robust and efficient methods that process raw image data to extract road edge segments, find the edge correspondences across images, and transfer them to 2D object space that can be used in a desktop GIS environment. The scheme is designed as a bottom-up process, in which the available information and organizational process are introduced at several layers of image processing. 23 High-Resolution Satellite Image Image Preprocessing (SDS, PCA and DS) Grey Image Edge Detection (Wavelet Transform Based) Binary Edge Image Detecting Features (Line Segments by HT, Curve Segments by Developed Curve Extraction Methods) Human operator Intervention Raster Image V ector Image F eature Parameters Figure 3.1 Scheme for the proposed approach of extracting road networks. Image Preprocessing (SDS, PCA and DS) Edge Detection (Wavelet Transform Based) Detecting Features (Line Segments by HT, Curve Segments by Developed Curve Extraction Methods) Human operator Intervention V ector Image Raster Image Figure 3.1 Scheme for the proposed approach of extracting road networks. The core of the system prototype consists of three main components: (1) edge extraction, (2) parametric road segment detection, and (3) road network reconstruction. There are two separate processes in the parametric segment detection component: detection of straight road segments and curve road segments. Each segment is important and possesses particular features. On the other hand, these two processes are related to each 24 other in terms of the computation of their attributes. Straight road segment detection is the premise of the curved road segment detection, whose results will be fed back to rectify straight road segments. This study focuses not only on object detection, but also on object recognition and identification. At the detection level, the objective is to separate objects discretely. At the recognition level, the objective is to determine what the objects are. At the identification level, the objective is to identify the objects. 3.3 Image Preprocessing The image preprocessing techniques enhance the edge features in images, which facilitate the edge extraction process. Image enhancement aims to visually amplify these slight differences to make them readily observable (Lillesand and Kiefer, 2000). 3.2 General Strategy for Semi-automated Road Extraction In this study, the three levels refer, more specifically, to edge detection, road segment recognition, and road parameter identification. 3.3.1 Interpretation of Image Data The images used in the case studies are subsets of an IKONOS image, which is original image without any enhancement. The subsets were cut with the original pixel value intact. To fully exploit the visibility for human vision and computer vision to perform image analysis, image enhancement is necessary. The visual spectrum for human eyes is the electromagnetic energy whose wavelength is approximately between 0.4-0.7 pirn . In 25 terms of colour perception of the human vision, the visual spectrum is from violet to red. Within the range of visual spectrum, human eyes are not uniformly sensitive to the intensity of the light. The peak sensitivity is located at about 0.55 //m , which is in green colour (Mather 1999). Most remote sensing systems can acquire data concurrently from several channels (multi spectral) to hundreds of channels (hyperspectral) (Sonka et al., 1999). Most optical sensors and cameras are designed to separate the visual energy into three primary colours: red, green, and blue. The primary colours are also called additive colours (Coren and Ward, 1989), which means a colour can result from the weighted addition of three primary components. For example, the colours displayed on television or computer screen is the mixture of red, green, and blue. Image displayed with these three bands are called natural-colour images (ERDAS, 1999), or normal-colour images (Lillesand and Kiefer, 2000), which approximate the colour observation that appears to humans. Gray, or colourless, scene can be simply produced by the combination of three primary colours with the same weight. Most airborne and spacebome sensors have the capability of detecting multiband data (e.g., the true-colour IKONOS images used in this study). Since WT or Canny edge detectors only inherently process single-band image, integrating as useful much information as possible from multiple bands into a single band (gray-level) image is the primary purpose of the image preprocessing stage in this study. There are many approaches to convert multiband imagery to single-band imagery. The most common approach is based on the theory and experiments on human vision. For human vision, the 26 pixel value in gray-level image represents the “brightness” of the energy projected on this position. Although it could be argued that the brightness is mostly related to the intensity, other dimensions, like wavelength and duration, affect the brightness of human perception as well (Coren and Ward, 1989). 3.3.1 Interpretation of Image Data Three common methods can be used to convert the true-colour images to gray-level images. All of them use linear combination of the three primary colours with different weights (Table 3.2). “Luma” is the classic method that complies with the industry video standards (Jack, 1996). “Luminance” is a new measure that mixes the RGB with new weights according to the new technology in the manufacturing of monitors (ITU, 1990). The “Average” method simply uses the arithmetical average of three primary colours, which makes sense when human perception will not be taken into account, and all the data from all bands have the same importance. Digital images are typically stored in the format of array of primary colours, which provides the possibility to restore and display the original colour scene. Many theories use colour space to represent the colours. RGB cube is one of 3D representation of colour space. The size of the cube is the maximum value for red, green, or blue. These values in this space, called digital numbers (DN), correspond to the average radiance measured at each pixel. DNs are simply the positive integers that result from quantitizing the original electrical signal from sensor into positive integer values (Lillesand and Kiefer, 2000). For example, for data of 8-bit-per-pixel in radiometric resolution, the range of DN is from 0 to 255 (which equals to 2^-1). The number of colours that can be discreetly (i.e., digitally) 27 Table 3,2 Weights of three conversion methods from RGB to gray-scale. Gray-Scale Method Red Green Blue Luma 0.299 0.587 0.114 Luminance 0.212671 0.715160 0.072169 Average 1/3 1/3 1/3 Table 3,2 Weights of three conversion methods from RGB to gray-scale. described is then equal to 2^'=16,777,216. Gray “colour” is located on the diagonal line from the origin to the opposite comer of this RGB cube. This means that identical values of red, green, and blue are assigned to a gray-level pixel. There is no limitation for computers to recognize the binary “data spectrum” in full range mode. The range from energy-free to maximum radiometric value is entirely visible to the computer. However, the results are highly sensitive to the data distribution in the image band(s). For example, the above-mentioned colour-to-gray methods, generally speaking, use weighted averages, which lead to a counteractive effect when there is little or negative correlation between two bands. 3.3.1 Interpretation of Image Data Thus, conversion from multi-band to single band inevitably loses information, which lies in different bands of the original image. Fischer (1969) pointed out that colour photography offers a great potential for use in many fields in Earth sciences. It is more satisfactory and promising for recognizing significant features than the panchromatic photography. The preprocessing stage focuses on finding as much information as possible from multi-band images; eliminating as much redundancy and false signal as possible; and integrating the information into a single band image for further edge and feature analysis and extraction. 28 28 It is possible to reduce the original data dimensions into fewer dimensions without losing too much information, or retaining as much information as possible. For example, in extracting features from the natural-colour image, redundant work will be done if all RGB bands are processed individually, because high correlation in visual bands exists in the normal image. The resulting features will overlap, which introduces difficulty in unifying or separating them. The difficulty is avoidable if a subset of the data could be used in place of the full dataset. The colour-to-gray methods mentioned above are not ideal, at least in the remote sensing area. It is possible to reduce the original data dimensions into fewer dimensions without losing too much information, or retaining as much information as possible. For example, in extracting features from the natural-colour image, redundant work will be done if all RGB bands are processed individually, because high correlation in visual bands exists in the normal image. The resulting features will overlap, which introduces difficulty in unifying or separating them. The difficulty is avoidable if a subset of the data could be used in place of the full dataset. The colour-to-gray methods mentioned above are not ideal, at least in the remote sensing area. Principal components analysis (PCA) has been used to enhance the visibility by the transformation from the original axis of observed variables to the space of principal axis (Mather, 1976). All of conversion methods have weakness. “Luma” and “Luminance” methods (shown in Table 3.2) could only be applied on RGB bands, while many data source are available with up to more than three visible bands, or even hyperbands. They only focus on human perception, which may not be optimal for digital image processing. 3.3.1 Interpretation of Image Data Although the “Average” method has no limit on the number of bands, it works well only on highly correlated bands. If little or negative correlation is shown in multiband, its side effect will show that the useful information or contrast is eliminated in the consequential gray-level image. 29 To transform an «-band image to primary component space, the first principal component (PCI) is the axe on which observed variables project with largest variance. The second principal component (PC2) is uncorrelated (orthogonal) to the first one and has the second largest amount of variance. The rest components are thus selected iteratively with descending order in terms of variance, until « mutually uncorrelated primary components have been chosen. To transform an «-band image to primary component space, the first principal component (PCI) is the axe on which observed variables project with largest variance. The second principal component (PC2) is uncorrelated (orthogonal) to the first one and has the second largest amount of variance. The rest components are thus selected iteratively with descending order in terms of variance, until « mutually uncorrelated primary components have been chosen. Data volume compression is one of common applications of PCA. As a compression method, PCA is a “lossy” method due to those information transformed in high-order primary components are to be discarded with respect to the low percentages in the total information. When one principal componet out of three bands is used, two thirds of original data space has been eliminated. 3.3.2 Standard Deviation Stretch The digital number (DN) is the value associated with a pixel in a digital image, corresponding to the value of some physical quantity such as the radiance in a particular band, measured at the detector (Rees, 1999). The range of possible DN values is usually from zero to 2" -1 , where « is the number of binary bits available, in terms of radiometric resolution (Richards and Jia, 1999). The Standard Deviation Stretch (SDS) is a radiometric processing method, which is widely used to enhance the radiometric characteristics of the original image. Its main 30 effect on the image is the increased intensity contrast compared to the original image that does not use the available full DN range. There are several reasons that the SDS is preferred for the enhancement of image intensity. As previously mentioned, for the 8-bit radiometric resolution imagery, the possible DN values range from 0 to 255. In this research work, all of the sample and experimental data are in 8-bit radiometric resolution, which means that the possible DNs are always between the ranges. The data in each band are not distributed in the full range. In other words, the minimum value of data is larger than zero, and/or the maximum value of data is less than the possible maximum value. The Min/Max Stretch simply expands the data to the full range of possible values by the linear transformation: DNoriginai = thc DN of thc original data, DNmox = the maximum DN of the original data, and DNmw = the minimum DN of the original data The normal distribution is the most encountered data probability model in many natural phenomena. The mean value of normally distributed data is at the peak of the frequency curve, as shown in Figure 3.2. The amounts of data below and above the mean value are balanced. The distribution curve could be represented in the function below (ERDAS, 1999): 31 >20 moan +2o Stored data file values 256 Original Histogram most of the data 2o moan D Stretched data file values 255 standard Deviadon Stretch values stretched over 255 are not displayed I ^ most of the data ^ /n -2o m##n 2o 2S6 stretched data me values Mln/Max Stretch Figure 3.2 Histograms of stretch methods (ERDAS, 1999). Figure 3.2 Histograms of stretch methods (ERDAS, 1999). 3.3.2 Standard Deviation Stretch TTC (3-2) Where, x = the quantity’s distribution that is being approximated, n and e = mathematical constants, and H and a = parameters controlling the location and shape of the probability. The SDS uses the DNs at which the number of standard deviations stands above and below the mean DN value of original data, rather than the maximum and minimum DN values. Equation 3-2 is modified as: DNStretched 2 S S (D N q ,,^-D N .„ ) D N ,., -D N .„ (3-3) (3-3) 32 32 Where: cr = standard deviation, n = number of standard deviation, ^^+ncT ~ ^he DN value na above the mean DN value, and = the DN value no" below the mean DN value. Where: ^^+ncT ~ ^he DN value na above the mean DN value, and = the DN value no" below the mean DN value. For example, if the data are normally distributed, the data within two standard deviations (±2cr) around the mean includes about 95% of all data (ERDAS, 1999). From the histogram of the normal distribution (see Figure 3.2), the percentage of these extreme data is small, less than 5%. After using Min/Max stretch, where the data range expand from 0 to 255, these 5% data cover the radiometric space of more than 1/3 of the range. The majority of data are squeezed into the radiometric space of less than 2/3 of 255. Obviously, most of the data are not fully stretched to exploit the ability of human/computer vision. The standard deviation (s^ of ^ sample values (Q) is calculated by (ERDAS, 1999): Sq - ^(Qi -M qŸ '•=' . (3-4) Sq - ^(Qi -M qŸ '•=' . (3-4) (3-4) (3-4) in which, jUq is the mean of the k sample values, which is given by: 33 3.4.1 Frequency Analysis Fourier transform is the most commonly used frequency analysis of a signal. Its transformation operation on a signal/in L^(R) is: = (3-6) = (3-6) where, co = the frequency, and ere, co = the frequency, and The signal/can be represented by inverse Fourier transforms: The signal/can be represented by inverse Fourier transforms: The signal/can be represented by inverse Fourier transforms: f{t) = ^ r F { o ) ) e '‘^d(o (3-7) ZTt (3-7) However, the Fourier transform cannot analyze the local frequency of a signal, which makes it impossible to measure the local variation. 3.3.3 Decorrelation Stretch Decorrelation stretch (DS), also called saturation stretch, is a spectral image enhancement technique. However, it works on the principal component space, not on the original spectral space. It exaggerates the least correlated portion of spectral data, improves the intensity and saturation without changing the distribution of hues (Gillespie et al., 1987). Three steps are involved in DS: (1) Principal components analysis (PCA). (2) Gaussian stretch is applied as contrast equalization on the principal components, whose histograms are, then, Gaussian distributed. (3) Inverse PCA transforms the Gaussian stretched PCs into the original spectral spaces. The DS is the most commonly used image enhancement technique that makes visual interpretation easier. Of course, this method will benefit the machine vision as well. The proposed multilayer-to-gray conversion involves two general steps: The DS is the most commonly used image enhancement technique that makes visual interpretation easier. Of course, this method will benefit the machine vision as well. The proposed multilayer-to-gray conversion involves two general steps: (1)The source multilayer image is taken as the input of DS, which will generate a multilayer image. (2)The multilayer is taken as the input of PCA, which will generate the final output single-layer image. 34 3.4.2 Time-Scale Analysis An important property of the wavelet transform is that the window of analysis is adapted locally to the phenomena under investigation, such that it is able to provide information on local signals (Ranchin et al., 2001). In this context, the WT methods provide a 35 promising avenue of research and applications. They not only allow efficient data compression while preserving the original spectral values, but also can be used to fuse images at different resolutions (Donnay et ah, 2001). The WT replaces Fourier transforms sinusoidal infinite waves by wavelets, a family of waves generated by translations and dilations. Two arguments (time and scale) are considered; while Fourier transform only takes one argument (frequency) Wavelet analysis is the time-scale view of a signal. In general, the scale parameter for time- ffequency/time-scale analysis has the relationship with the frequency parameter in that: low scale means high frequency, and high scale means low frequency. Time-scale is a more natural way for people to observe and analyze the signal obtained. The scale used in maps is one example. In mapping industry, small-scale represents a global view, and large-scale represents a detailed view. The scale domain also applies to remote sensing and image analysis. In large-scale (high-resolution) satellite imagery, the cars on highway are discernable, while in small-scale (low-resolution) satellite imagery, these ears may be regarded as noise or too small to detect. Hence, the scale magnitude affects the performance of feature recognition. On the other hand, when elassifying a satellite imagery, low resolution often simplifies the elassification since the area, with low-resolution both in spatial and spectral, could appear homogenous and is within the same class. Even in a single image, one kind of features can be separated from others by its distinct scale size. Such features, in remotely sensed imagery, may represent highways, buildings, cars, vegetation textures, and even noise, which are exemplified in descending order of their scales. Figure 3.3, showing the images with scales of 1:25,000,1:1,000, 36 Figure 3.3 IKONOS Images of Kampsax, Denmark resampled at scale (a) 1:25,000, (b) 1:10,000, and (c) 1:2,500 Figure 3.3 IKONOS Images of Kampsax, Denmark resampled at scale (a) 1:25,000, (b) 1:10,000, and (c) 1:2,500 1:2,500, respectively, illustrates how different scales can make different understanding of the imagery. 1:2,500, respectively, illustrates how different scales can make different understanding of the imagery. The wavelet transform is defined by: 3.4.4 Regularity Analysis Unlike Fourier transform analysis, the WT can analyze the local regularity of a signal. WT modulus maxima are related to the singularities of the signal. An ID wavelet analysis illustrates the ideas behind the singularity detection in Figure 3.4. 3.4.3 Dyadic Wavelet Transform Dyadic wavelets transform (DWT) involves sampling the scale s to be the dyadic number y J e Z . Time is not sampled. The DWT of signal f i s defined by substituting s by 2^, y G Z in Equation 3-9: wf(u,v) = ( 3-10) V2-' 2-' wf(u,v) = ( V2-' 2-' 3-10) with The wavelet transform is defined by: Wf{u,s) >= M (3-8) Wf{u,s) >= M (3-8) (3-8) (3-8) where xj/ is the zero average function, u and s are the time and scale arguments respectively. The wavelet is derived by the translations and dilations by: (3-9) (3-9) To accelerate the computations, the dyadic WT is often used. To accelerate the computations, the dyadic WT is often used. 37 3.4.5 Multiscale Edge Detection The magnitude of first differencing or gradient can be computed by finding the partial derivatives and and then determining the composite gradient (Mather, 1999) by 38 2 1 0 60 100 160 200 250 Figure 3.4 Modulus maxima of continuous WT of ID signal, Modulus maxima are in light gray in grayscale printing. (http://cas.ensmp.ff/~chaplais/Wavetour_presentation/Regularite/Detection_of_singulariti es.html) 2 1 0 60 100 160 200 250 Figure 3.4 Modulus maxima of continuous WT of ID signal, Modulus maxima are in light gray in grayscale printing. Modulus maxima are in light gray in grayscale printing. (http://cas.ensmp.ff/~chaplais/Wavetour_presentation/Regularite/Detection_of es.html) Modulus maxima are in light gray in grayscale printing. (http://cas.ensmp.ff/~chaplais/Wavetour_presentation/Regularite/Detection_of_singulariti es.html) Modulus maxima are in light gray in grayscale printing. (http://cas.ensmp.ff/~chaplais/Wavetour_presentation/Regularite/Detection_of_singulariti es.html) Modulus maxima are in light gray in grayscale printing. (http://cas.ensmp.ff/~chaplais/Wavetour_presentation/Regularite/Detection_of_singulariti es.html) (3-12) (3-12) The direction of the composite gradient is represented by the angle Û, which is given by The direction of the composite gradient is represented by the angle Û, which is given by 0 = tan-1 Ap , J 0 = tan-1 Ap , (3-13) J 0 = tan-1 Ap , (3-13) J (3-13) In terms of multiscale edge detection, it is useful to introduce the Canny algorithm first, as a classical but dominant method. As Mallat (1998) mentioned, his wavelet version of In terms of multiscale edge detection, it is useful to introduce the Canny algorithm first, as a classical but dominant method. As Mallat (1998) mentioned, his wavelet version of 39 edge detection is equivalent to the Canny method. Besides, the preliminary work for this thesis was also based on the Canny edge detector. edge detection is equivalent to the Canny method. Besides, the preliminary work for this thesis was also based on the Canny edge detector. The input format required by the Canny edge detector is a gray-scale image, and the binary image is the output format, in which the significantly detected edges are supposed to be in white, while non-edge areas are supposed to be in black. The Canny algorithm is a fast and convenient edge detector for multiscale edge detection. The Canny edge detector uses two tunable thresholds to divide the edge and non-edge pixels. The pixels whose DN lie above high threshold are set to edge pixels (strong edge), while the pixels whose DN lies below a low threshold are set to non-edge pixels. 3.4.5 Multiscale Edge Detection The pixels between these two thresholds are weak edge pixels, which could be output as final edge only if they show strong connection with strong edge pixels. Any noises with little connection to the true edge are eliminated. By tuning the criterion of connectivity, the output edge image is adaptive in accordance with the preferred scale of edge. The gradient vector is composed of partial derivatives ofp(x, y)\ The gradient vector is composed of partial derivatives ofp(x, y)\ / a _ \ V p = dx (3-14) / a _ \ V p = dx (3-14) (3-14) In the Canny algorithm, a point (xo, yo) in image p(x, y) is defined as an edge point when the modulus of gradient vector is locally maximum, which means along the direction of In the Canny algorithm, a point (xo, yo) in image p(x, y) is defined as an edge point when the modulus of gradient vector is locally maximum, which means along the direction of Vj9. The modulus of gradient vector at (xo, yo) is larger than the modulus of its ID 40 neighbors. This step of non-maximum suppression attempts to thin the qualified edge to a single pixel width. neighbors. This step of non-maximum suppression attempts to thin the qualified edge to a single pixel width. Here, the procedures applied for the Canny edge detection are summarized as follows: Step 1 : Smoothing. The image is pre-processed by a Gaussian filter to eliminate the noise and smooth the coarse appearence. Step 1 : Smoothing. The image is pre-processed by a Gaussian filter to eliminate the noise and smooth the coarse appearence. Step 2: Differentiation. A first derivative operator is applied in the x and y directions, respectively. Thus, the edges that mark areas of different brightness are strengthened. Non-edge pixels are dimmed because the gradient magnitudes with respect to their neighboring pixels are low. This step starts off by reducing the angle of gradient to one of the four sectors shown in Figure 3.5. The algorithm uses its eight neighboring pixels. At each pixel, the center element of the neighborhood is compared with its two neighbors along line of the gradient given by the sector value. If the pixel value is non-maximum, that is, not greater than the neighbors, it is suppressed. 112.5 67.5' 157.5; :2.5‘ 247.5' 292.5' Figure 3.5 Four test directions for the Canny algorithm. Figure 3.5 Four test directions for the Canny algorithm. 41 Step 3: Thresholding. Canny edge detector uses double thresholds to generate a binary image. That is, unlike the single threshold method, if the pixel value lies above the upper limit, it is accepted as an edge pixel. If the pixel value lies below the lower limit, it is rejected. A pixel with values between the upper and lower limits is considered an edge if it exhibits strong connection with other edge pixels. The WT-based edge detector employs the criteria of the Canny algorithm in the steps including smoothing and connecting, but the wavelet version has its own characteristics in the calculations and its advantages and disadvantages. The main advantage is the multiscale nature of the wavelet analysis. The wavelet smoothing kernel 6 is dilated at different scales of the wavelet. Two wavelets are to be constructed because of two directions in image processing: (3-15) O X (3-16) dy (3-15) (3-16) The scale used in dyadic wavelet transform is simplified to{2-' }^.^2 5 as with Fast Fourier The scale used in dyadic wavelet transform is simplified to{2-' }^.^2 5 as with Fast Fourier The scale used in dyadic wavelet transform is simplified to{2-' }^.^2 5 as with Fast Fourier Transforms does, to limit the levels of scale. neighbors. This step of non-maximum suppression attempts to thin the qualified edge to a single pixel width. The DWT on image p(x, y) is denoted by W'^piu, v,2^ ) =< p(x, y), {x-u ,y-v)> = pif/\, (u, v) (3-19) W'^piu, v,2^ ) =< p(x, y), {x-u ,y-v)> = pif/\, (u, v) (3-19) (3-19) The scaled convolution kernels could be written as The scaled convolution kernels could be written as The scaled convolution kernels could be written as ^2J ^2/ (3-20) ^2J ^2/ (3-20) (3-20) The two wavelets in Equations 3-16 and 3-17 are transformed into a scaled format: he two wavelets in Equations 3-16 and 3-17 are transformed into a scaled format: The two wavelets in Equations 3-16 and 3-17 are transformed into a scaled format: (3-21) dx -2 _ ,• d02‘_ (3.22) ¥ 2> — 2"^ dy (3-21) (3.22) Equation 3-20 can be transformed into: Equation 3-20 can be transformed into: Equation 3-20 can be transformed into: 43 W ^ p { u ,v ,V ) = 2^ — {p 02‘ )(M,v) O U d — (j7^2/)(w,v) ,0 V = 2^'V(pÛ20(u,v) (3-23) W ^ p { u ,v ,V ) = 2^ — {p 02‘ )(M,v) O U d — (j7^2/)(w,v) ,0 V = 2^'V(pÛ20(u,v) (3-23) (3-23) Then, the modulus of the wavelet components is proportional to the gradient vector of p smoothed by 62^ : Then, the modulus of the wavelet components is proportional to the gradient vector of p smoothed by 62^ : Mp{u,v,V) = ^ \w ^ p (u ,v ,r f +\W ^p{u,v,V)\ (3-24) Mp{u,v,V) = ^ \w ^ p (u ,v ,r f +\W ^p{u,v,V)\ (3-24) Mp{u,v,V) = ^ \w ^ p (u ,v ,r f +\W ^p{u,v,V)\ (3-24) The angle, a , of the wavelet vector is calculated by: a = tan W^p{u,v,T) W'p(u,v,2^)^ (3-25) (3-25) The range of the angle falls into 0 < Ap{u,v,2^) < 2n-, where: The range of the angle falls into 0 < Ap{u,v,2^) < 2n-, where: (3-26) (3-26) (3-26) 44 Now that the wavelet equivalents of modulus Mp(u,v,2^) and angle Àp(u,v,2^) are obtained, the same procedure is followed as with a Canny algorithm to find the local modulus maxima of wavelet transform. That is, it finds the maximum modulus of the local ID neighborhood at (uo, vg) along the direction of Ap(u,v,2-'). The local maximum pixel is the edge pixel at the scale of 2^. neighbors. This step of non-maximum suppression attempts to thin the qualified edge to a single pixel width. = = (3-17) (3-17) 42 42 — - - ,k ¥i> (^, y) = ¥ 2j i- x - y ) —* - - ,k ¥i> (^, y) = ¥ 2j i- x - y ) (3-18) where k = \ ,2, indicating the x and y directions, respectively. where k = \ ,2, indicating the x and y directions, respectively. where k = \ ,2, indicating the x and y directions, respectively. neighbors. This step of non-maximum suppression attempts to thin the qualified edge to a single pixel width. Now that the wavelet equivalents of modulus Mp(u,v,2^) and angle Àp(u,v,2^) are obtained, the same procedure is followed as with a Canny algorithm to find the local modulus maxima of wavelet transform. That is, it finds the maximum modulus of the local ID neighborhood at (uo, vg) along the direction of Ap(u,v,2-'). The local maximum pixel is the edge pixel at the scale of 2^. A similar chaining method as the Canny algorithm is applied to the local modulus maxima of WT. The two neighbouring maxima are linked by a vector, which should be perpendicular to the angle direction Ap(u,v,2-' ) at each point. Due to the uncertainty on the scale at which the qualified edges could be identified, the criteria should be selected by human interpreter according to the presence of edges of desired features, such as roadsides. Generally, for a small scale, the result is quite noise- sensitive, resulting in many fine structures, and short twisted lines. For large scales, coarse edges are obtained, but some of them have large delocalization errors (Ziou and Tabbone, 1997). The Vision Laboratory at the University of Algarve provides an interesting synthetic image (see Figure 3.6a) for testing edge detectors. This test image contains a vertical squarewave grating, a ring, and two thin diagonal lines. Figures 3.6b and 3.6c are the resultant edge maps from Figure 3.6a produced by the Canny edge detector and WT edge detectors used in this research, respectively. 45 I I (a) (b) (c) Figure 3.6 Edge detection results (a) test image, (b) Canny detected edges, and (c) WT detected edges. I I (a) (b) (c) Figure 3.6 Edge detection results (a) test image, (b) Canny detected edges, and (c) WT d d d (c) (b) (a) Figure 3.6 Edge detection results (a) test image, (b) Canny detected edges, and (c) WT detected edges. By comparing Figures 3.6b and 3.6c, the pros and cons of each method can be stated as follows: (1) The WT-based method is able to correctly detect the edges at the intersection of the two edges, where both edges are continuous and cross each other, while one of the two edges is broken by the Canny method. (2) The WT-based method is able to detect the line edge (see the diagonal lines in Figure 3.6c), while in Figure 3.6b, produced by the Canny method, a single line edge becomes double lines. neighbors. This step of non-maximum suppression attempts to thin the qualified edge to a single pixel width. (2) The WT-based method is able to detect the line edge (see the diagonal lines in Figure 3.6c), while in Figure 3.6b, produced by the Canny method, a single line edge becomes double lines. (2) The WT-based method is able to detect the line edge (see the diagonal lines in Figure 3.6c), while in Figure 3.6b, produced by the Canny method, a single line edge becomes double lines. Figure 3.7 demonstrates of Mallet’s wavelet edge detector, which is applied to an image with a single circle. Figure 3.7 demonstrates of Mallet’s wavelet edge detector, which is applied to an image with a single circle. 46 The original image is on top. WT angle for a non WT Horizontal Vertical WT zero modulus WT WT modulus modulus maxima m Figure 3.7 Classical image Circle for the wavelet edge detection. (http://cas.ensmp.fr/~chaplaisAVavetour_presentation/ondelettes%20dyadiques/Circle.ht ml) The original image is on top. WT angle for a non WT Horizontal Vertical WT zero modulus WT WT modulus modulus maxima m The original image is on top. Figure 3.7 Classical image Circle for the wavelet edge detection. (http://cas.ensmp.fr/~chaplaisAVavetour_presentation/ondelettes%20dyadiques/Circle.ht ml) 3.5 Geometric Feature Detection The proposed algorithm for road extraction is based on the extraction of road’s geometric parameters (Easa et al., 2003). In most cases, the intersection of two roads, or turn point on one road, should be an arc of a circle. The reason is that the turning the speed of 47 vehicle is to be kept the same and smoothly once on a curve track. In terms of dynamics, once the object is moving, and rotating around one center, the linear velocity and rotational velocity keep constant on rotation. An arc is an ideal curve for this purpose. It starts from (changing direction) an original road, ends in another road (other direction). Obviously, both the start and end point are not only part of the start and end road segments, but are also on the arc. In other words, the curve is a connector of two straight lines; the slope of the start point on arc must be same as the start line (road), likewise with end point to end line (road). In terms of calculus, continuous first order derivative leads a smooth curve. From road safety perspective, a horizontal curve must have a minimum radius that ensures vehicle stability. The minimum radius depends on the road design speed, the superelevation and maximum side friction factor. Based on the law of dynamics, the radius of a circular curve is related to other variables (Easa, 2002), where, R = minimum radius (m), where, R = minimum radius (m), V= vehicle speed (km/h), e = curve superelevation (m/m), and / = side friction factor. 48 To determine the minimum radius of a horizontal curve, the variables used in Equation 3- 28 would be the design speed, maximum superelevation, and maximum side friction factor for V, e, and f, respectively. In IKONOS imagery, two types of road arcs could be found. They are called comer arc and turning arc. The comer arc is at a street comer (e.g., at a signalized intersection), which is only visible on high-resolution satellite images (e.g., IKONOS or QuickBird). In low-to-medium resolution satellite images (e.g., Landsat TM and SPOT), this type of arc may not be discernable since the comer is almost similar to a right angle. That is, the radius is too small, in pixel length, to be measured. The tuming arc, such as a highway horizontal curve, which would appear in images of all resolutions since road alignments is generally discemable. This study focuses on the second type of road curves. In this study, horizontal tuming arcs are further categorized into four subtypes. Each subtype has a different strategy developed for its extraction. They are: (1) simple circular arcs, (2) reverse arcs, (3) compound arcs, (4) spiral arcs. 3.5.1 Extraction of Simple Horizontal Circular Curves For simple horizontal circular curves, the proposed algorithm simply detects the lines that are tangent to the horizontal circular curve at any points selected, and then finds the arc 49 49 that connects these lines. The Hough transform, a well-adapted algorithm in image feature detection, was used. For man-made objects (e.g., roads) in IKONOS imagery, their shapes appear to be simple, regular, and primitive, which can be accurately described mathematically. The standard Hough transform (Trucco and Verri, 1998) was applied to detect straight lines, which can be represented by (Figure 3.8) that connects these lines. The Hough transform, a well-adapted algorithm in image feature detection, was used. For man-made objects (e.g., roads) in IKONOS imagery, their shapes appear to be simple, regular, and primitive, which can be accurately described mathematically. The standard Hough transform (Trucco and Verri, 1998) was applied to detect straight lines, which can be represented by (Figure 3.8) p = x cos 6 +y sin 9 (3-28) where pF distance from the origin to the line to be detected, angel between the x -axis and a line passing the origin and perpendicular to the line to be detected. angel between the x -axis and a line passing the origin and perpendicular to the line to be detected. In a binary image, every pixel, whose location is (x, y) and value is non-zero, may belong to many lines (with parameters p and 9) passing through this pixel. A counter for every line passing through this pixel is incremented by one. Thus, after scanning all image pixels, the accumulator contains the number of pixels that every line has. Any line having pixels greater than a given threshold is a candidate line in this image. Obviously, although a line in an image might not be a perfect straight line (somewhere broken, noisy, or distorted), the statistical result can filter out these imperfections. A line recognizable to human eyes should have a large quantity of associated pixel members. The accumulator for recording these counters is a 2D array A(p, 9) that is initialized to zeros. In a binary image, every pixel, whose location is (x, y) and value is non-zero, may belong to many circles (with parameters Xo,yo, and i?) whose perimeters pass through it (Figure 50 Figure 3.8 Geometric properties of a straight line. Figure 3.8 Geometric properties of a straight line. 3.9). 3.5.1 Extraction of Simple Horizontal Circular Curves The equation of a circle is given by 3.9). The equation of a circle is given by 3.9). The equation of a circle is given by (x - XoŸ+ (y - yoŸ = (3-29) (x - XoŸ+ (y - yoŸ = (3-29) where Xg, yo = coordinates of the center of the circle, where Xg, yo = coordinates of the center of the circle, where Xg, yo = coordinates of the center of the circle, where Xg, yo = coordinates of the center of the circle, where Xg, yo = coordinates of the center of the circle, R = radius of the circle. R = radius of the circle. R = radius of the circle. Similar to line detection, the proposed method first constructs an accumulator initialized with zeros, and then scans all pixels to find the candidate circles. This time, there are three unknown parameters (xo, yo, R), and, therefore, a 3D array accumulator is established to record the counting. The arc is only part of a circle. In IKONOS imagery, no perfect circle exists and many separate arcs may belong to one circle. Therefore, arc detection and circle detection are similar. 51 Figure 3.9 Geometric properties of a circle. Figure 3.9 Geometric properties of a circle. A simple curve is a circular curve connecting two tangents that intersect at the point of intersection (PI) (Figure 3.10). The beginning of the curve is the point of curvature (PC) and the end of the curve is the point of tangency (PT). In practice, finding specific tangents (lines) and the curve (arc) connecting them is more localized. That is, it does not need to find all lines and arcs in an image. There are many of them, and the difficulty is how to determine the threshold criteria for an image. Even if a road is short and its length is below the threshold, it still could be a candidate for investigation. An arc, whose center is located out of the image, cannot be detected. Therefore, automatic line and arc detection in the whole image is not feasible for the puiposes of this study. Instead, the searching area for lines is limited. The user is allowed to select the sides of a road, so that only two mouse clicks are required for detecting an arc (Figure 3.10). 3.9). The equation of a circle is given by As noted before, the accumulator array is 3D, which means that arc 52 PC Line A PT Road Segment 1 max ■^V-Road Segment 2 Line B' Figure 3.10 Schematic representation for establishing a simple horizontal curve, Line A Road Segment 1 Line B' Figure 3.10 Schematic representation for establishing a simple horizontal curve, detection requires greater memory and time than line detection. detection requires greater memory and time than line detection. detection requires greater memory and time than line detection. To make the search results more accurate and to satisfy computation and computer storage limitations, a self-adaptive threshold is introduced to find a candidate line. Two filters are applied to the search: (1) the line must pass through the selected point and (2) the direction of the line must be the same as (or nearest to) the direction of the line in the small point window. Thus, a line is found finally on which the small point window (a or b in Figure 3.10) resides. Once two intersecting roads have been found, the arc connecting them can be easily found. Its centre must reside on the bisector line (Line B in Figure 3.10) of their 53 intersection angle /, and its radius R must be the distance from the curve centre to either of the two lines. Now only the centre point of this arc is an independent unknown parameter, and C lies on the Line B. The problem is reduced to a ID analysis. The amount of computation and memory also could be greatly reduced. The centre of the searched circle must lie on Line B. It is assumed that the user would click the initial points on the straight part of the road, not on the arc. Then, the user picks the initial point (point a in Figure 3.10) which is less distant to the intersection point 7, and the algorithm will automatically makes an assistant line (Line A) perpendicular to the road line on which this initial point lies. The new intersection point (Cmax) generated by the assistant line and Line B is the farthest possible centre for the arc being searched. To find the center of the searched arc, a ID accumulator is allocated to record the rank of each possible arc. Each possible arc is the one whose centre is between PI and Cmax, and whose radius is the distance between its centre and either of the two tangent lines. Only one independent unknown is involved because the centre lies on Line A. A statistical method, like Hough transform line detection, can tolerate imperfection of the image and determine the most likely 'actual' arc. After scanning all pixels which each arc passes, if not zero, the counter for this arc is incremented by integer one. detection requires greater memory and time than line detection. The larger the radius, the more pixels on the chord of the arc would be scanned. To make each arc have an equal chance, after scanning, the counter number is divided by the arc length of each arc. Thus, the arc that has the maximum relative hit count is selected. This results in the only one arc which is the best arc connecting the tangent lines. Once the best arc is determined, all corresponding parameters are saved, including its radius, beginning point, and end point. 54 54 3.5.2 Extraction of Reverse and Compound Curves A reverse horizontal curve consists of two consecutive circular arcs in opposite directions. The radii of the two arcs can be same or different. A compound horizontal curve consists of two circular arcs in the same direction. The radii of the two arcs cannot be the same, otherwise they are portions of a simple horizontal circular curve. The sketches of reverse curve and compound curve are drawn in Figures 3.11 and 3.12, respectively. These subtypes of horizontal curves cannot be established using the algorithm developed for simple curves because this algorithm requires two straight lines to extract the simple curve. The two arcs of the reverse or compound curves cannot be separately identified as two simple circular curves because their common part is a tangent point, not a straight line. If the reverse or compound curves are separated by a short tangent, the algorithm for simple curves can be applied to establish the two arcs separately. However, the common part of the reverse or compound curve is normally a common tangent point (point f in Figures 3.11 and 3.12). To establish these types of curves, other algorithms were developed to recognize these two subtypes. The algorithm for reverse curves is described below, but the same principles can be applied to compound curves. The procedure for establishing reverse horizontal curves is illustrated in Figure 3.11. If the road is wide enough, two reverse horizontal curves can be shown in the image, each one represents each side of the road. This is the case in high-resolution satellite imagery where a road is shown as a band rather than a line. The reverse curve has two arcs (with same radii or not) with one common tangent point. The objective here is to find the 55 Line B Road Segment 1 -Line C C2 R2. Line D /R2 Line A '? Road Segment 2 Figure 3.11 Schematic representation for establishing a reverse horizontal curve. Line B Road Segment 1 -Line C Line D Line A ' Road Segment 2 Figure 3.11 Schematic representation for establishing a reverse horizontal curve. Line B Line A R1 Lirie C Road Segment 2 " Road Segment 1 Figure 3.12 Schematic representation for establishing a compound horizontal curve. Line B Line A Figure 3.12 Schematic representation for establishing a compound horizontal curve. Figure 3.12 Schematic representation for establishing a compound horizontal curve. 3.5.2 Extraction of Reverse and Compound Curves The center of the second arc must be on Line C, and on the line (Line D) perpendicular to the common tangent Line B. These two lines (Line C and Line D) generate an intersection point, which is the centre of the second arc. (f) Calculate the percentage of matching pixels of the arc under consideration in the image. (f) Calculate the percentage of matching pixels of the arc under consideration in the image. There are many possibilities of such combinations for a reverse or compound arc. By trying every candidate, and comparing their percentage of match in the image, the combination with the maximum possibility can be selected as our final reverse or compound curve. The logical flow of the above procedure, which involves a three dimensional loop (distance over tangent, radius of first arc, and common tangent angle), is shown in Figure 3.13, coupled with Figures 3.11 and 3.12 to illustrate the procedures. 3.5.2 Extraction of Reverse and Compound Curves gure 3.12 Schematic representation for establishing a compound horizontal curv 56 starting tangent point e, ending tangent point d, and common tangent point ^ along one reverse curve of one road side as well as the radii and centers of the two arcs. The algorithm involves the following steps to automatically establish the reverse curve illustrated in Figure 3.11. These steps are also applicable for compound curve illustrated in Figure 3.12: (a) Find two straight lines representing the start and end tangents of the reverse curve (b) Select the first initial point a. To reduce the volume of computations, it is recommended that the initial point be selected on the side with the arc that has smaller radius (first arc). Then, select the second point b. As shown in Figures 3.11 and 3.12, two initial points a and b are marked on the two tangents of the reverse curve. The following steps are completed by the computer automatically. (c) Select every point on the first straight line (from a toward the first arc) to be the candidate of the starting tangent point. Make a line perpendicular to the starting line and through the candidate tangent point, such as Line A, on which the center of the candidate first arc must be located. (d) Try all possible radii and all possible common tangent points on the circumference of the candidate first arc. Using every possible common tangent point (such as/), make a common tangent line shared by both arcs, such as Line B, which will intersect with the end tangent at c. The distance cf must equal the distance cd. Thus, the position of the candidate end tangent point d can be located on the end tangent. (d) Try all possible radii and all possible common tangent points on the circumference of the candidate first arc. Using every possible common tangent point (such as/), make a common tangent line shared by both arcs, such as Line B, which will intersect with the end tangent at c. The distance cf must equal the distance cd. Thus, the position of the candidate end tangent point d can be located on the end tangent. 57 (e) Draw a line perpendicular to the end tangent line at d, such as Line C. 3.5.3 Extraction of Spiral Curves A spiral is a curve with a uniformly changing radius. It is used in highway to overcome the abrupt change in direction that occurs where the alignment changes from a tangent to a circular curve. The spiral curve either connects one circular curve and one tangent, or connects two circular curves of different radii. A sketch of a spiral curve, which connects a tangent and a circular curve, is drawn in Figure 3.14. The cubic spiral forms the first order approximation of the spiral whose curvature (reciprocal of the radius) is linearly related to its arc length. The circular curve, spiral 58 (cubic) curve, and straight line are represented in (%, space by Equations 3-31, 3-32, and 3-33, respectively. Their relationships are illustrated in Figure 3.15. (cubic) curve, and straight line are represented in (%, space by Equations 3-31, 3-32, and 3-33, respectively. Their relationships are illustrated in Figure 3.15. ( x - x o Y + ( y - y o y — (3-30) ( x - x o Y + ( y - y o y — (3-30) No Final angle? ,Yes No Maximum radiu.«s? Yes No Last point on first tangent? Yes Select initial angle Calculate the percentage of matching Pij^ Select a point (centre of circle) on Line A with radius Select a point on the first tangent with distance 2}-i=0 If P i j i > P m a x , then P m a x = P ijk . Store P „ ,a x and corresponding parameters, Rj, Rj^Cj and Cj. The curve corresponding to the final P^ax is selected along with its parameters. Flow diagram for establishing reverse and compound horizontal curves. Select a point on the first tangent with distance 2}-i=0 Select a point (centre of circle) on Line with radius Final angle? Maximum radiu.«s? Yes Last point on first tangent? The curve corresponding to the final P^ax is selected along with its parameters. The curve corresponding to the final P^ax is selected along with its parameters. Figure 3.13 Flow diagram for establishing reverse and compound horizontal curves. igure 3.13 Flow diagram for establishing reverse and compound horizontal curve 59 Tangent to Spiral Spiral Curve Circular Curve Figure 3.14 Spiral curve. 3.5.3 Extraction of Spiral Curves Tangent to Spiral Spiral Curve y = ax^ y = 0 (3-31) (3-32) y = ax^ y = 0 (3-31) (3-32) (3-31) (3-32) where Ji = unknown radius of the circular curve, where Ji = unknown radius of the circular curve, yo = unknown center of the circular curve, yo = unknown center of the circular curve, a = unknown constant of the cubic spiral. a = unknown constant of the cubic spiral. The spiral curve is approximated by a polynomial curve as indicated by Equation 3-31. It is possible to detect the spiral curve in an image and its parameter by modifying a Hough transform. The basic idea is also based on the finding the parameter of the spiral statistically in an image, like 2D ( p , Û) line searching or 3D (xo,yo,.R) circle searching using a Hough transform. Thus, a Hough spiral curve detector can be designed to work in 4-dimensional ( x q , yo, R, Ô) searching space, which is almost impossible with the computation power of the hardware (recall the discussion for HT-based circle detection). By extending the algorithm of detecting simple arc, it is possible to reduce the searching space to 3D searching as follows; 60 Road Segment 2 Road Segment 1 J l Line A (x-xoŸ+{y-yo) y=ox^ X Figure 3.15 Symmetrical spiral curves approximated by cubic spiral. Road Segment 2 Figure 3.15 Symmetrical spiral curves approximated by cubic spiral. (1) Let x-axis be on the one of the straight road segment (tangent to spiral); (2) Let the tangent to spiral be the origin of coordinate system; (3) The coordinates of the intersecting point A of circular curve and spiral (cubic) curve are given by (%, ax^). At Point A, the first order derivatives of both curves must be the same for smooth transition. (4) The first order derivatives of the cubic curve (y=ca^) and road segment 1 at origin O must equal to 0. (5) Center C must locate on the bisector line (Line A) of the angle intersected by road segment 1 and 2. (5) Center C must locate on the bisector line (Line A) of the angle intersected by road segment 1 and 2. Thus, the constant a in Equation 3-31 of spiral (cubic) curve, center C of the circular curve, and coordinates of A and B can be obtained by satisfying the above conditions. 3.5.3 Extraction of Spiral Curves Further details on the extraction of spiraled horizontal curves are presented in Dong et al. (2003). 61 3.5.4 Extracting Second Roadside Once the edge of one side of a road arc has been found, in terms of the parameters of radius and centre position, the next step is to find the other side of that road arc. The same way used for the first side can be applied again to extract the second arc on the same road, and, subsequently, the centerline can be determined. In this way, each arc on the same road is treated as an individual one, not related. Hence, the error which occurred when extracting the first arc will not affect the precision of the second arc. Of course, there is no guarantee that second arc is error-free without previous error for first arc. To extract both roadsides separately, four points should be clicked on the image. However, this process is inefficient and fault-prone because much human interference is involved. From the road design perspective, the width of a road is normally constant, especially over the circular arc. The sight distance is much shorter on the curve than it is on the straight part of a road, because the view angle does not allow the driver to perceive the change of road width, not to mention that obstacles, such as buildings or trees, located at the comer may block the view sight. In Figure 3.16, the inner arc is detected with radius R and centre C. The road width is set to W (distance between the two roadsides) is constant along the road arc. It is obvious that the outer curve is an arc with the centre C and radius R+W. If the case that the first arc detected is outer one, the inner curve is to be an arc with the centre C and radius R-W. 62 In both cases, the positions of both sides of arc road are located at C, which has been known from detecting the first arc. What is unknown is only the road width W. Once one side of a road is determined, the computer can find the other side of the same road, either inward or outward. To reduce the search space and the chance of errant detection, a user-defined parameter is needed: which roadside is selected, inner or outer. If it is the inner side, the direction of search is outward along the radial direction, and vice versa. 3.5.4 Extracting Second Roadside The search space is also limited because of the actual road width in the real world. It is assumed that the minimum width of the road is one lane (approximately 4 m wide), and maximum width is 8 lanes (approximately 32-m wide). On IKONOS imagery, the search space is converted to pixel units. The conversion is straightforward: In both cases, the positions of both sides of arc road are located at C, which has been known from detecting the first arc. What is unknown is only the road width W. Once one side of a road is determined, the computer can find the other side of the same road, either inward or outward. To reduce the search space and the chance of errant detection, a user-defined parameter is needed: which roadside is selected, inner or outer. If it is the inner side, the direction of search is outward along the radial direction, and vice versa. The search space is also limited because of the actual road width in the real world. It is assumed that the minimum width of the road is one lane (approximately 4 m wide), and maximum width is 8 lanes (approximately 32-m wide). On IKONOS imagery, the search space is converted to pixel units. The conversion is straightforward: = Road _ Width{meter) Re solution _ o f _\m age{meter i pixel) (3-33) (3-33) Pixel _Width(pixel) Figure 3.16 Detection of double roadsides. Figure 3.16 Detection of double roadsides. 63 For example, the corresponding road width in IKONOS imagery varies from 4 to 32 pixels. 3.6 Connection to Desktop GIS Environment Spatial data acquisition and interpretation are bottlenecks to GIS-T databases. Remote sensing has been one of the most effective means of spatial data collection. According to Wilkinson (1996), the interface between GIS and remote sensing can be envisaged in one of three different ways: (1) Remote sensing can be used as a tool to gather data sets for the use in GIS; (2) GIS data sets can be used as ancillary information with which to improve the products derived from remote sensing, and (1) Remote sensing can be used as a tool to gather data sets for the use in GIS; (3) Remote sensing data and GIS data can be used together for modeling and analysis. Using vector data format, the feature boundaries are converted to straight-sided polylines that approximate the original curves. These polylines are encoded by determining the coordinates of their vertices, which can be connected later to form the curves (Sonka et al., 1999). The automated creation and update of urban road networks, with semi-automated methods, can be faster, less error-prone and less tedious than using traditional manual vectorization from imagery. Here, the road network extracted in this research is in 2D. In 64 most urban areas, the overlook view approximates the urban scene closely, because the roads are flat and continuous with respect to their change of height. most urban areas, the overlook view approximates the urban scene closely, because the roads are flat and continuous with respect to their change of height. Since the parameters of each detected road segments are precisely known using the proposed approach, these parameters can be used to generate the data compatible with the existing GIS-T databases. For example, a shape file is a widely used vector data format. It can store the vectors with the types of points, polylines, polygons in 2D or 3D domain. In the case studies, the parameters of both line and curve road segments are used to generate the shape files which can be read by most popular GIS software. Figure 3.17 shows how the vector model constructs an arc connecting two straight lines, that simulates the turning part of a road. The left side is the extracted edge or centerline of a road segment. This is saved in raster format, pixel by pixel. The right side is the illustrated vector format of the segment. 3.6 Connection to Desktop GIS Environment Only vertices (black dots) are saved (encoded) in persistent media, the other part (dashed line) can be inteipolated linearly when reconstructing the road (decoding). If the vertices are sampled within appropriate distances, the reconstructed shape could represent the original curve without noticeable coarse steps. The sampling procedure can be completed by the computer. The distance of the neighboring vertices can be determined by the curvature of the segment between the two vertices. Generally, the larger the curvature, the smaller the distance to be adopted. For example, in Figure 3.17, the straight segment (with curvature 0) only needs two vertices at the start and end points to be represented, while the simple circular curve (with the finite radius) needs 8 vertices for its representation. 65 Figure 3.17 Vectorized polyline approximation of linear and curve segments. Figure 3.17 Vectorized polyline approximation of linear and curve segments. 66 4 CASE STUDIES 4.1 Study Areas To test the proposed algorithms, an experiment was designed and conducted to extract two types of horizontal curves from IKONOS imagery. One standard frame IKONOS imagery, acquired in August 2001, was provided by the Greater Toronto Airports Authority (GTAA). The IKONOS imagery was geometrically corrected in two dimensions (X and Y) in the Universal Transverse Mercator (UTM) projection coordinate system, zone 17° in the Geodetic Reference System of 1980 (GRS80) and North American Horizontal Datum of 1983 (NAD83). The imagery used in this study is a pan sharpened true-colour IKONOS image, which has three colours and 1-m spatial resolution. Two areas were selected in this study. Both images are the subsets of the IKONOS imagery covering west Toronto, Canada. Figures 4.1 and 4.2 show these two images, enhanced by standard deviation stretch. Figure 4.1 covers a typical highway network at the intersection area of Highway 401 and Highway 427. The major disturbances are the vehicles, shadows of highways, and marking lines. Figure 4.2 covers a typical residential area. To comply with the requirement of wavelet dyadic transform, each image size is cut into a window consisting of 1024x1024 (2'°x2'°) pixels. Although processing arbitrary size is preferred and can be implemented by overlaying multiple intermediate square (2"- 67 g Figure 4.1 SDS(a = 2) enhanced image of study area 1. Figure 4.1 SDS(a = 2) enhanced image of study area 1. sized) images, the procedure is not discussed in the research session. Table 4.1 lists the brief information of each image. sized) images, the procedure is not discussed in the research session. Table 4.1 lists the brief information of each image. The approach previously described in Section 3.5 was implemented in a PC environment on a Visual C++ platform and OpenCV code framework. The user interface was designed 68 to facilitate the initialization of seed points and selection of curve type for the user. The complete flow of processes on the image data has been illustrated briefly in Figure 3.1. Prior to the road extraction, image preprocessing (Section 4.2), and wavelet edge detection (Section 4.3), are performed. Since the two steps are not image-specific, the intermediate results produced are shown only on the sample highway image. %» m Figure 4.2 SDS(a = 2) enhanced image of study area 2 %» m Figure 4.2 SDS(a = 2) enhanced image of study area 2 69 Table 4.1 Sample image information. 4 CASE STUDIES Table 4.1 Sample image information. Highway Area (Figure 4.1) Residential Area (Figure 4.2) Image Information Layers 3(R, G,B) 3(R, G,B) Image Size (pixel) 1024x1024 512x512 Data Type Unsigned 8-bit Unsigned 8-bit Map Information Upper Left Coordinates (X,Y) Pixel Size (m) (614293.013, 4836858.060) (611074.01,4842065.06) 1.0 1.0 Projection Information Datum NAD27 (Canada) NAD27 (Canada) NAD27 (Canada) 4.2.1 Standard Deviation Stretch The original image (Figure 4.3) of the study area 1 is the raw image that does not give appropriate brightness and contrast for both human operator and computer to see the features. The histogram (Figure 4.4) of the raw image shows that band data (green band) are mostly distributed at the lower DN range (the mean is 71.0), although the pixel values range from 0 to 255 (from the smallest to largest possible DN). Roughly speaking, seen from the shape of histogram its data is normally distributed. The standard deviation (a ) of the normal distribution is 31.4, which means most of data are within the range approximately from 10 (mean- 2cr ) to 130 (mean+2cr). Let the mean value be the approximate measure of brightness, and the standard deviation be the approximate 70 Figure 4.3 Original image of study area 1 without any enhancement. Figure 4.3 Original image of study area 1 without any enhancement. measure of contrast, neither of the brightness and contrast of the raw image are not suitable for human or computer vision analysis. Applying SDS on the raw normally distributed data with 2, the SD-enhanced image is displayed in Figure 4.1, which gives a much clearer scene with highways, buildings, vegetation, vehicles, and other features, than the original image (Figure 4.3). The 71 71 improvement of visibility can be inteipreted by histogram of the SDS-enhanced image as shown in Figure 4.5. Using green band as an example, the min/max values are still the same as the original image, but the mean value has moved from 71 to 127.2, which is around the middle value of the range. The new standard deviation cr of the SDS enhanced image is 58.3, which means the most data are distributed roughly from 10 to 245. These statistical values also proved that SDS can significantly enhance the brightness and contrast of the original image. '■■■ 0 47.5004 62025 "1 2 5 5 ''\':T 0 " :-^ 66.8443 255 255 ^ 0 7019964 15415 21271 7789 (a) Red (b) Green (c) Blue Figure 4.4 Histogram of original image (Min = all zero; max = all 255; mean = 48, 71, 67; a = 30,31,33). 127217 12B.4B9 127.171 (a) Red (b) Green (c) Blue Figure 4.5 Histogram of the SDS-enhanced image. 4.2.1 Standard Deviation Stretch (median = 125, 132, 136; a = 59, 58, 61) '■■■ 0 47.5004 62025 "1 2 5 5 ''\':T 0 " :-^ 66.8443 255 255 ^ 0 7019964 15415 (a) Red (b) Green (c) Blue Figure 4.4 Histogram of original image (Min = all zero; max = all 255; mean = 48, 71, 67; a = 30,31,33). '■■■ 0 47.5004 62025 "1 2 5 5 ''\':T 0 " :-^ 66.8443 255 255 ^ 0 7019964 15415 (a) Red (b) Green (c) Blue 4.2.2 Decorrelation Stretch It is noticed in Figure 4.5 that the mean value of each RGB band has been moved to middle of the radiometric range (around 127-128) after the SDS. This means that all RGB bands are SDS-enhanced individually and equally, and only the contrast of intensity of the RGB image has been enhanced by the SDS. Applying decorrelation stretch (DS) on SDS enhanced image (Figure 4.1) is the next step to further increase the contrast between colours. Figure 4.6 shows the image processed by the DS. Its contrast of colours has been visually enhanced. By comparison of the histograms of the DS-enhanced image to the histograms of the SDS-enhanced image, it can be found that the median values of red and blue bands have moved significantly towards lower DN and higher DN, while the mean values of red and blue bands do not change much. This phenomenon shows the visual contrast of different colours can be increased without changing the hue by the DS. Figure 4.4 Histogram of original image Figure 4.4 Histogram of original image (Min = all zero; max = all 255; mean = 48, 71, 67; a = 30,31,33). 21271 7789 127217 12B.4B9 127.171 (a) Red (b) Green (c) Blue Figure 4.5 Histogram of the SDS-enhanced image. (median = 125, 132, 136; a = 59, 58, 61) 21271 7789 127217 12B.4B9 127.171 (a) Red (b) Green (c) Blue Figure 4.5 Histogram of the SDS-enhanced image. (b) Green (b) Green (c) Blue (c) Blue Figure 4.5 Histogram of the SDS-enhanced image. (median = 125, 132, 136; a = 59, 58, 61) 72 4.2.3 Principal Components Analysis The final image needed for edge detection should be a gray-scale image. Both the SDS enhanced image (see Figure 4.1) and the SDS-DS-enhanced image (see Figure 4.6) include three bands. PCA can be applied to either of them at this stage to generate a single-band (gray-scale) image. In this study, both of the single-band images were generated and compared in order to find which one is optimal with respect to contrast. It could be found, according to the results of ERDAS Imagine, that 93.6% (SDS-enhanced) or 87.8% (SDS-DS-enhanced) of the information has been transformed into the first principal component for image of study area 1, and likewise for image of study area 2. 73 I Figure 4.6 SDS-DS-enhanced image of study area 1 depicted in Figure 4.3. Figure 4.6 SDS-DS-enhanced image of study area 1 depicted in Figure 4.3. 32518»! 72B2 127339 255 0 30282^ 128.403 9907 255 ;0 41759»" 120.385 255 (a) Red (b) Green (c) Blue Figure 4.7 Histogram after DS-enhanced (median = 119, 134,140; o= 62, 59, 62) (c) Blue Figure 4.7 Histogram after DS-enhanced (median = 119, 134,140; o= 62, 59, 62) 74 4.2.4 Results and Discussion The two gray-scale images generated by PCA and DS-PCA are shown in Figures 4.8 and 4.9, respectively. By employing change detection technique provided by ERDAS Imagine 8.5, the difference between the PCA image and DS-PCA image can be visually examined. Figure 4.10 shows such a difference image displaying the change of the two images. The green (light gray in gray image) areas are where the DN (brightness) in DS- PCA image is larger (brighter) than the DN (brightness) in PCA image; the red (dark gray in gray image) areas are where the DN (brightness) in DS-PCA image is smaller (darker) than the DN (brightness) in PCA image; black areas are of no change. It can be seen that, in DS-PCA image, most of the highway areas are highlighted, and its background areas are dimmed, in comparison with the PCA image. When viewing the PCA image (Figure 4.8), the highways are generally brighter than their background areas. By DS-PCA processing, bright becomes brighter, dark becomes darker, and the contrast of the whole gray-scale image is enhanced by DS-PCA processing compared to PCA processing. Thus, the optimal gray-scale image for edge detection is to be the DS-PCA image (Figure 4.9). 4.3 Wavelet Edge Detection The wavelet edge detection used in this study was implemented by LastWave, a signal processing (wavelet oriented) software developed by Emmanuel Bacry, CNRS (http://www.cmap.polvtechnique.ff/~bacrv/LastWave/). 75 I Figure 4.8 SDS-PCA enhanced image of study area 1 depicted in Figure 4.3. Figure 4.8 SDS-PCA enhanced image of study area 1 depicted in Figure 4.3. The gray-scale image (Figure 4.9) is used as the input image of wavelet edge detector. The wavelet edge detector performs dyadic wavelet decomposition up to 3 levels (/ = 1, 2,3) or more. At each wavelet transform scale, extrema (local modulus maxima along the gradient direction) are computed. A threshold value is given to eliminate those extrema whose modulus is smaller than the threshold. Figures 4.11, 4.12 and 4.13 are the binary edge images at three scales (/ = 1,2,3), respectively. 76 76 compared to the edge image at scale 2 (Figure 4.12). Further checking the overlay of the gray-scale image (Figure 4.9) and the edge image at scale 3 (Figure 4.13) tells that there are dislocations of some edges to the true feature boundaries. The dislocations are due to the smoothing effect used at each wavelet decomposition scale. The edge image at scale 2 (Figure 4.12) keeps the most edges of desired features (roads). The dislocations of edge compared to the edge image at scale 2 (Figure 4.12). Further checking the overlay of the gray-scale image (Figure 4.9) and the edge image at scale 3 (Figure 4.13) tells that there are dislocations of some edges to the true feature boundaries. The dislocations are due to the smoothing effect used at each wavelet decomposition scale. The edge image at scale 2 (Figure 4.12) keeps the most edges of desired features (roads). The dislocations of edge Figure 4.10 The difference image of the PCA image and DS-PCA image (Green areas: DS-PCA>PCA; Red areas: DS-PCA<PCA; Black areas: DS-PCA=PCA). Figure 4.10 The difference image of the PCA image and DS-PCA image (Green areas: DS-PCA>PCA; Red areas: DS-PCA<PCA; Black areas: DS-PCA=PCA). Figure 4.10 The difference image of the PCA image and DS-PCA image (Green areas: DS-PCA>PCA; Red areas: DS-PCA<PCA; Black areas: DS-PCA=PCA). 78 78 % M Figure 4.9 DS-PCA-enhanced image depicted in Figure 4.3. Figure 4.9 DS-PCA-enhanced image depicted in Figure 4.3. The edge image at scale 1 presents almost all intensity variations (whose moduli are above the specified threshold) of Figure 4.9. 4.3 Wavelet Edge Detection The edges of almost all features make it difficult for both human and computer to discriminate the desired features. The edge image at scale 3 looks like the most elegant one of the three edge images. However, some important boundaries of desired features (e.g., roads) disappear or become broken 77 Figure 4.11 Edge image of the image of study area 1 (Figure 4.9) at wavelet scale 1. Figure 4.11 Edge image of the image of study area 1 (Figure 4.9) at wavelet scale 1. are unnoticeable by checking the overlay image. By observing and analyzing the edge images at all scales, the decision can be made that the edge image at scale 2 will be the best representation of the feature boundaries. 79 Figure 4.12 Edge image of the image of study area 1 (Figure 4.9) at wavelet scale 2. Figure 4.12 Edge image of the image of study area 1 (Figure 4.9) at wavelet scale 2. 4.4 Extraction of Road Curves Although the background computation works on the edge image, the user input and results are displayed on the SD-enhanced colour image (e.g., Figure 4.1), which gives the most natural and comfortable view to facilitate the human operation. 80 Figure 4.13 Edge image of the image of study area 1 (Figure 4.9) at wavelet scale 3. gure 4.13 Edge image of the image of study area 1 (Figure 4.9) at wavelet scale 3 4.4.1 Extraction of Simple Circular Curves The image for the simple horizontal curve is a freeway interchange that has many simple curves (Figure 4.14). As noted, all interchange curves were accurately identified using the proposed methodology. The identification was done sequentially (one curve at a time). The parameters identified for each curve included the centre, radius, start angel, and end 81 angel. Extracting simple circular curves is fast. It spent about 3 seconds to get the largest- scale curve (in radius) in Figure 4.14. Thus, most of the time consumption for the road network extraction in the overall region is in the session of user interaction, such as clicking the seed points. I m m i ^0 Figure 4.14 Results of establishing horizontal curves at complex freeway interchange. Figure 4.14 Results of establishing horizontal curves at complex freeway interchange. 82 4.4.2 Extraction of Reverse and Compound Curves The image for the reverse curve is a subset of the IKONOS imagery. As noted, the algorithm correctly identified the two arcs of the reverse curve (arc A and B in Figure 4.15), and two arcs of compound curves (arc A and B in Figure 4.16). Similar to simple curves, the algorithm can extract the parameters of the curve from the image, including the coordinates of the centers, radii, start point, end point, and common point. Computation on the reverse or compound curve is time-consuming. For example, it requires about 3 minutes to achieve the reverse curve in Figure 4.15. : m . Figure 4 15 Results of establishing reverse horizontal curves in urban residential area : m . Figure 4.15 Results of establishing reverse horizontal curves in urban residential area. Figure 4.15 Results of establishing reverse horizontal curves in urban residential area. 83 msm itM tSS. 'MmU fMi Figure 4.16 Results of establishing compound horizontal curves in urban highway area. 4.16 Results of establishing compound horizontal curves in urban highway area. Figure 4.16 Results of establishing compound horizontal curves in urban highway area. 4.4 J Extraction of Spiral Curves The image for the spiral curve is a subset of the IKONOS imagery. As noted, the algorithm correctly identified the simple circular curve (arc A) and its two wing spiral curves (curve B and Q (see Figure 4.17). Here, the spiral curve is approximated by a cubic polynomial. 84 84 Figure 4.17 Results of establishing spiral horizontal curves in urban highway area. Figure 4.17 Results of establishing spiral horizontal curves in urban highway are Figure 4.17 Results of establishing spiral horizontal curves in urban highway area. 4.4.4 Results and Discussion As shown in Figure 4.14, almost all of the arcs can be extracted. The developed method can accurately establish simple, reverse, compound, and siralled curves, even for a complex freeway interchange. Some curves could not be extracted using the proposed methodology. An example is the transition curves (seen in the middle of Figure 4.1), which might be combination of spiral curve and circular curve(s). 85 The time consumed on computation of simple and complex arcs varies dramatically. The extraction of a simple circular curve is almost instant, while the extraction of a reverse or compound curve is much slower. It is mainly due to the difference between using ID and 3D searching spaces. However, the time spent on reverse and compound arc is still tolerable, and the number of reverse and compound arcs is much less than the number of simple circular curves in a road network. 4.5 Extraction of Road Networks The individual arc, simple or complex, with its centerline and/or two sides can be extracted from the image one by one. A road, visible in an image, is composed of straight segments and curved segments, or only a straight segment transiting the image. In the former case, the human operator instructs to close a road after extracting the last curve on it, and then starts with another road until all the roads of interest have been extracted. All the roads extracted together represents the road network. Note that the road-like feature in the upper-right part of study area 1 (Figure 4.1) is actually an open ditch, which should not be extracted as part of the road network although its alignments (arc F and G in Figure 4.14) can also be measured by the proposed method. Human knowledge is always needed to exclude those non-road but road-like features, which is difficult for automated methods. It is also noteworthy that the storage of all geometric parameters for each road network is less than 10KB. Figures 4.18 and 4.19 are examples of all extracted freeway interchange and road network in the image, respectively. The uncompressed sizes of the files for the two images are 3MB and 768KB. 86 Figure 4.18 Segmented freeway interchange in the image of study area 1. Figure 4.18 Segmented freeway interchange in the image of study area 1. The retrieved geometric parameters of all line and curve segments were used to create the vector data files. Each road is represented as three polylines, which are two roadsides and one centerline. Figure 4.20 and 4.21 show the vector data displayed by ERDAS Imagine. The sizes of the files for the two vector files are 431KB and 54KB. The sizes of the files for the two vector files are 431KB and 54KB. 87 Figure 4.19 Segmented road network in the image of study area 2. Figure 4.19 Segmented road network in the image of study area 2. 4.6 Performance Evaluation This section discusses the performance evaluation of the developed method of road network extraction . Internal self-diagnosis and external evaluation of the obtained results are of major importance for the relevance of automatic methods for practical applications. However, only relatively little work has been carried out in this area. The measures for self-diagnosis should take into account the reliability of the generated road primitive as 88 well as the consistency between them. External evaluation is conducted by comparing the extracted roads with manually measured reference data. well as the consistency between them. External evaluation is conducted by comparing the extracted roads with manually measured reference data. The method of road extraction from high-resolution satellite imagery could be used in various applications. The extracted highway strictly complies with the highway design, which gives the chance that the orientation of the road at any point is geometrically Figure 4.20 Vectorized freeway interchange in the image of study area 1. Figure 4.20 Vectorized freeway interchange in the image of study area 1. 89 Figure 4.21 Vectorized road network in the image of study area 2. Figure 4.21 Vectorized road network in the image of study area 2. known. The orientation is strictly parallel to the centerline, the width of the road at any point is also known. If the centerline is represented as I (x, y), the road width at any point on the centerline is W (I). A unique 2D road model could be reconstructed. It is assumed that the width of the arc part of a road is constant. Two linked neighboring arcs could have different widths. The width along the straight segment, connecting these two arcs, changes linearly in order to smoothly change the width from one arc to another. The two ends of a road are exceptions, because the arcs connecting them are outside of the image 90 and, therefore, not visible. In such cases, the width is assumed to be the same as the width of the adjacent arc. In transportation applications, roads are the area of interest. Only the objects on the road are to be analyzed. Unfortunately, road segments projected on the satellite image are, most likely, irregularly shaped, and not in the shape of simple rectangles. The contour of the road of interest is the most desirable in such cases. 4.6 Performance Evaluation The method presented here is a fast and precise way to trace the outline of a road in a satellite image. Some highways in Figure 4.18 are samples of interest areas. Other highways of non-interest and non highway areas are blacked out. Now, in such a selective image, it is much easier for automatic algorithms to perform some tasks, such as detection of vehicles, marking lines or manholes, analysis of traffic flow, and etc. The results could be more convincing because the focus is simply on the roads of interest, not on the entire image. Another advantage is the abundant auxiliary information, such as geometric parameters, central line and width of the road, are provided natively when the results are extracted. The information can be used accordingly as the criterion in various analyses. For example, to extract the vehicles on the highway, even if the area of the interested highway of interest is the only object displayed in the image (see Figure 4.18), it is still difficult to achieve the goal. The raster data in digital format are insufficient to provide accuracy with such a small scale, with only several pixels per vehicle. The detection algorithm would be sensitive to the extraneous object seen at this scale, such as the noise or non vehicle objects. The knowledge of the direction of the vehicle is the most important cue for ease of detection. The major axis of the rectangle of the moving vehicle body is 91 parallel to the local direction of the road. Since the direction of any position on a given road is known already. The searching will become a 2D (x, y) loop reduced from a 3D (x, y,0) loop. This will greatly increase both the speed and accuracy of the detection of algorithm. Given the same IKONOS images over the same residential area (see Figure 4.2), the resultant images (see Figures 4.21 and 4.22) are generated by the semi-automated method developed in this study and an automated method, based on Fuzzy Mathematical Morphology (FMM). This permits the analysis of the geometric and radiometric properties of roads and other non-road features in the imagery (Li et al, 2003). The Figure 4.22 Extracted road centerlines using the method based on FMM (Li et al., 2003). Figure 4.22 Extracted road centerlines using the method based on FMM (Li et al., 2003). 4.6 Performance Evaluation Figure 4.22 Extracted road centerlines using the method based on FMM (Li et al., 2003). 22 Extracted road centerlines using the method based on FMM (Li et al., 2003). Figure 4.22 Extracted road centerlines using the method based on FMM (Li et al., 2003). Figure 4.22 Extracted road centerlines using the method based on FMM (Li et al., 2003). 92 morphological structure element, an elongated rectangle window, was used to detect the straight road segments automatically. morphological structure element, an elongated rectangle window, was used to detect the straight road segments automatically. From the result image (Figure 4.22) of this automated method, some disadvantages of this kind of method are listed below: From the result image (Figure 4.22) of this automated method, some disadvantages of this kind of method are listed below: (1) The rectangle structure element is too simple to be used to detect all geometries. Some of the curve comers in Figure 4.2 will nearly become right angles in the detected centerline image (Figure 4.22) because of lack of curve structure elements, which cannot be designed as easily as the rectangle element with respect to the variation in curvature. (1) The rectangle structure element is too simple to be used to detect all geometries. Some of the curve comers in Figure 4.2 will nearly become right angles in the detected centerline image (Figure 4.22) because of lack of curve structure elements, which cannot be designed as easily as the rectangle element with respect to the variation in curvature. (2) The centerline and roadside extracted are not as smooth as the real world. It is mainly due to the imperfections (e.g., trees, cars and shadows) on the road. (3) The resultant image (Figure 4.22) is actually a raster edge image, in which individual road has not been identified. (3) The resultant image (Figure 4.22) is actually a raster edge image, in which individual road has not been identified. (4) The time spent on FMM computation is increased by the order O (n^), where n is the image size in one dimension. For example, with double the dimension of an image, the time would be 8 times longer. This is also a general time scale for such methods, which automatically scan all pixels and local directions in the entire image. 4.6 Performance Evaluation (4) The time spent on FMM computation is increased by the order O (n^), where n is the image size in one dimension. For example, with double the dimension of an image, the time would be 8 times longer. This is also a general time scale for such methods, which automatically scan all pixels and local directions in the entire image. The method developed in this study has none of the above-mentioned problems. Its advantages over the automated method are: (1) Both straight and curved elements can be identified accurately. 93 (2) The method is not easily affected by the disturbances on the road. (3) The results can be the individual roads, and their geometric parameters. (4) The time increase in terms of size depends on how many new curve elements can be seen on the enlarged area. It is in the order of 0 ( n \ where n is the image size in one dimension. (2) The method is not easily affected by the disturbances on the road. (2) The method is not easily affected by the disturbances on the road. (3) The results can be the individual roads, and their geometric parameters. (4) The time increase in terms of size depends on how many new curve elements can be seen on the enlarged area. It is in the order of 0 ( n \ where n is the image size in one dimension. (4) The time increase in terms of size depends on how many new curve elements can be seen on the enlarged area. It is in the order of 0 ( n \ where n is the image size in one dimension. If one is familiar with the manual vectorizing on an image, the steps in this method are similar, but much easier. First, when vectorizing the straight road segments, the human operator does not need to pinpoint the correct pixel on the side of road, since the system will identify the appropriate point to click around the edge of road. Second, there is no need to vectorize the curved road segments, which is the most difficult in manual vectorizing, since it is done automatically by the method presented in this study, based on the type of curve. Third, the human operator only needs to vectorize one side of a road, because the other side can be extracted automatically by this system. 5 CONCLUSIONS AND RECOMMENDATIONS In this thesis, a semi-automated method of urban road network extraction based on geometric analysis of IKONOS imagery has been presented. In this chapter, the major achievements of the developed approach are summarized in Section 5.1. Some limitations are discussed in Section 5.2. Based on the previous discussions, conclusions are drawn in Section 5.3. Finally, recommendations for future research are given in Section 5.4. 4.6 Performance Evaluation An external evaluation is essential for the results from automatic road network extraction. The manually plotted or extracted data can be the reference data for evaluating the automated method (Heipke et al., 1997; Wiedemann and Hinz, 1999). Since the method introduced in this study can achieve a high accuracy for urban road maps, its results are a candidate for reference data for the evaluation of other automated extraction, or even manually plotted or extracted methods. It is appropriate to check the validity and accuracy of the results of the proposed road extraction method by an internal evaluation. That is, the resultant vector data is 94 overlapped on the original image (enhanced if necessary) in order to check the performance visually by the user. Figures 4.23 and 4.24 are such overlay images Figure 4.23 Overlay of the extracted highways on the input image of study area 1, Figure 4.23 Overlay of the extracted highways on the input image of study area 1, 95 ü s m Figure 4.24 Overlay of the extracted road networks on the input image of study area 2 ü s m Figure 4.24 Overlay of the extracted road networks on the input image of study area 2 Figure 4.24 Overlay of the extracted road networks on the input image of study area 2. 96 5.1 Summary In this research work, a semi-automated extraction method of urban road networks from high-resolution satellite imageiy has been developed and implemented. This novel approach fully exploits the high-accuracy image information, the knowledge of road geometry, and the abstraction ability of humans to achieve more reliable and accurate results. The developed methodology differentiates itself from the traditional semi- or fully-automated approaches in two aspects: (1) the feature detection is based on the geometric characteristics of roads, in particular, the curve road segments; and (2) the feature parameters of lines and various types of horizontal road alignment are extracted precisely, and the 2D vector or symbolized model of road network can be reestablished. With this generic methodology, both urban and rural highways, even road networks in urban residential areas can be extracted from a single IKONOS image. 97 Most of the algorithms for detecting the linear and curve features are implemented to test and demonstrate the practicality of the new proposed approach. This potential technique can be used in mapping highways using available satellite images, extracting and exporting features into GIS databases. This can be done with either vectorized data in widely recognizable format, or the geometric parameters in order to save the storage space. For the road network located in the urban and suburban areas (in this case, the Greater Toronto Area), the primitive structure of highway geometry is relatively simple because of the limited space. Normally, simple horizontal curves are dominant in number in such areas due to the simplicity of design and construction, and limitation of the available space for the designed highway. Some spiral curves can be found around the suburban areas which are far away from downtown and the mobility is the first concern in design. In residential areas, reverse curves are often seen, because accessibility is important factor to be concerned in urban areas. The extracted arcs are regarded as the joints of the straight road segments. Once all the parameters of the joint arcs have been achieved, a complete 2D horizontal road can be reconstructed by the parameters of its joints. Extracting a road without curvature segments in the image view field is achieved simply by detecting one linear segment. Extracting residential roads are more difficult than extracting highways. 5.1 Summary There are three main factors influencing the results: (1) the straight segments of a road are not strictly connected by ideal curve(s) somewhere; (2) the detection of straight road segments is 98 disturbed by the trees along streets or their shadow; and (3) many residential roads show a white border with one or two pixels width on one or both sides, which makes the detection of road width difficult. A visible street edge may appear to be irregularly dashed line. However, in most cases of the case studies, they can be detected Hough transform and the statistical method for curve detection. This is also one of the advantages of semi-automated methods over fully- automated ones, which are highly sensitive to these urban disturbances on high-resolution imagery. disturbed by the trees along streets or their shadow; and (3) many residential roads show a white border with one or two pixels width on one or both sides, which makes the detection of road width difficult. A visible street edge may appear to be irregularly dashed line. However, in most cases of the case studies, they can be detected Hough transform and the statistical method for curve detection. This is also one of the advantages of semi-automated methods over fully- automated ones, which are highly sensitive to these urban disturbances on high-resolution imagery. 5.2 Limitations Although the objective of this research was to develop a generic methodology of road network extraction, only a prototype was studied or implemented for the time being. There are some limitations listed as follows: Although the objective of this research was to develop a generic methodology of road network extraction, only a prototype was studied or implemented for the time being. There are some limitations listed as follows: (1) Although the spiral curve is considered as a basic type, its detection is not implemented for all types of spirals because it involves various combinations with circular curves. (2) Only those basic types of road curves were considered. Other complex curves are most likely the combination of the basic curves (circular, reverse, compound, and spiral curves). The proposed methodology is based on the premise of successful detection of tangent lines, which does not exist within the combination of curves. (3) The connection of curves (grouping of road segments) is only implemented with simple circular curves. Other curve types are only tested individually. (3) The connection of curves (grouping of road segments) is only implemented with simple circular curves. Other curve types are only tested individually. 99 (4) If one arc passes the boundary of an image, it is impossible to be detected using the proposed methodology, although some strategies can achieve acceptable results. (4) If one arc passes the boundary of an image, it is impossible to be detected using the proposed methodology, although some strategies can achieve acceptable results. (4) If one arc passes the boundary of an image, it is impossible to be detected using the proposed methodology, although some strategies can achieve acceptable results. (5) It is assumed that the intersection angle of the two tangents are subtended to the curve. Actually, two tangents can be parallel, and reverse subtended. The methodology should be modified to allow such special cases, 5.3 Conclusions This thesis has presented a new semi-automated method for the extraction of urban road networks from a single IKONOS image. The method is based on multiscale analysis and knowledge of road geometry. Its applications to real case studies show the benefits provided by the method. Several findings and conclusions can be made as follows: (1) The image pre-processing is a general and necessary step to employ the power of both human and computer vision. Here, the enhancement of the contrast is the first priority in image preprocessing, since enhancement will assist the next step of feature edge detection. The standard deviation stretch is used to linearly enhance the intensity contrast. The DS-PCA is used to enhance the spectral contrast if necessary. The DS-PCA method is not always better than the PGA method to obtain a single-band image. The contrast enhancement is expected to increase the global contrast. (2) As a method of spatial enhancement, edge detection is used to extract the primitive features in this research. Only sharp variations that outline important (2) As a method of spatial enhancement, edge detection is used to extract the primitive features in this research. Only sharp variations that outline important 100 features in the image remained. Since the features of interest (roads) in the study are large-scale, the small-scale features should be eliminated. The most common multiscale edge detectors, Canny and WT-based methods, are considered and compared. Canny method is impressive in its fast speed, and the fact it is not limited in image size. However, wavelet transform generates more satisfactory edges than the Canny method. (3) Hough transform is the basis of the proposed feature detection methodology. HT automatic line detection is used directly in this work to find the straight segments of roads. The HT-based automatic circle detection is not suitable for the large- scale arc detection in this study. Several approaches were developed to extract the basic types of highway curves according to the highway geometry. (4) The basic idea of extracting road network is to find each road’s curves, then extract those which are connected by straight lines. All the roads in the region of the image can be extracted likewise. The proposed method works best with high- resolution imagery. 5.3 Conclusions (5) This proposed methodology was designed to work with high-resolution or median-resolution image, on which the road width is at least three pixels wide for IKONOS Im-resolution imagery, every single lane road is about 3~4m wide; for medium-resolution image, roads with width of 5m or above can be detected. Otherwise, the road can still be extracted if it is straight or with visible curved segments. From a practical point of view, the developed approach can be used as an operational and reliable road extraction method, whose output provides useful highway data. 1 0 1 5.4 Recommendations for Further Research As discussed above, some limitations exist with the developed method. Future work may include: As discussed above, some limitations exist with the developed method. Future work may include: • To establish a whole road network within a certain region, a complete extraction system should consider the road’s 2D spatial relationship. • Consider the use of auxiliary information (e.g., shadow of the road, or DSM data) to retrieve the relative or absolute elevation of each road, thus a 3D road model can be established. • Once the 2D road has been extracted and its heights along the road are known, it is possible to detect the parameters of vertical alignments of the road using the same principle as with the detection of horizontal alignments. Of course, algorithms for extracting various vertical curves need to be developed. 102 BIBLIOGRAPHY Amini, J., M.R. Saradjian, J.A.R. Blais, C. Lucas and A. Azizi, 2002. Automatic road side extraction from large scale imagemaps, International Journal o f Applied Earth Observation and Geoinformation, 4(2):95-107. Aronoff, S., 1989. Geographic Information Systems: A Management Perspective. WDL Publications, Ottawa, Canada. Auclair, F., D. Ziou, C. Armenakis and S. Wang, 2001. Automated correction and updating of road databases from high-resolution imagery. Canadian Journal o f Remote Sensing, 27(l):76-89. Bâhr, H.B., 2001. Image segmentation for change detection in urban environments. In: Donnay, J.P., M.J. Barnsley, and P.A. Longley (eds.): Remote Sensing and Urban Analysis, Taylor & Francis, London. Baker, S., 1998. Design and Evaluation o f Feature Detectors, PhD Dissertation, Graduate School of Arts and Sciences, Columbia University, New York. Barnsley, M.J., L. M(j)ller-Jensen and S.L. Barr, 2001. Inferring urban land use by spatial and structural pattern recognition. In: Donnay, J.P., M.J. Barnsley, and P.A. Longley, (eds.) Remote Sensing and Urban Analysis, Taylor & Francis, London. Barzohar, M. and D.B. Cooper, 1996. Automatic finding of main roads in aerial images by using geometric-stochastic models and estimation, IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(7):707-721. Baumgartner, A., C. Steger, C. Wiedemann, H. Mayer, W. Eckstein and H. Ebner, 1996. Update of roads in GIS from aerial imagery: verification and multi-resolution QXivdiCiion, International Archives o f Photogrammetry and Remote Sensing, 31 (B3):53-58. Baumgartner, A., C. Steger, H. Mayer and W. Eckstein, 1997. Multi-resolution, semantic objects, and context for road extraction. Semantic Modeling for the Acquisition o f Topographic Information from Images and Maps, Birkhauser Verlag, Basel, Switzerland, pp. 140-156. Baumgartner, A., W. Eckstein, C. Heipke, S. Hinz, H. Mayer, B. Radig, C. Steger and C. Wiedemann, 1999a. T-REX: TUM Research on Road Extraction, In: Festschrift fur Prof. Dr.-Ing. Heinrich Ebner zum 60. Geburtstag, Hsg.: C. Heipke und H. Mayer, Technische Universitat München, Lehrstuhl fur Photogrammetrie und Fernerkundung, pp. 43-64. 103 Baumgartner, A., C. Steger, H. Mayer, W. Eckstein and H. Ebner, 1999b. Automatic road extraction in rural areas. In: International Archives o f Photogrammetry and Remote Sensing, 32 (3-2W5): 107-112 Canny, J., 1986. A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 8(6):679-698. Carreira, M.J., M. Mirmehdi, B.T. Thomas and Marta Penas, 2002. Perceptual primitives from an extended 4D Hough transform. Image and Vision Computing, 20(13-14):969- 980. Chen, C.H., J.S, Lee and Y.N. Sun, 1995. Wavelet transformation for gray-level comer detection. Pattern Recognition, 28(6);853-861. BIBLIOGRAPHY Chiesa, C., 2001. Multispectral Feature-space approaches to change detection and road extraction. In: Proceedings in Road Centerline Extraction & Maintenance Specialist Meeting, CLEM2001, August 6-7, Santa Barbara, CA. (http://www.ncgia.ucsb.edu/ncrst/meetings/clem2001/proceedings.html) Chui, C.K., 1997. Wavelets: A Mathematical Tool for Signal Analysis, SIAM Monographs on Mathematical Modeling and Computation, Society for Industrial & Applied Mathematics, Philadelphia, PA. Colwell, R.N., 1961. Some practical applications of multiband spectral reconnaissance, American Scientist, 49:9-36. Coren, S. and L.M. Ward, 1989. Sensation & Perception, 3"^^ Edition, Harcourt Brace Jovanovich. Couloigner, 1. and T. Ranchin, 1998. Extraction of urban network from high-spatial- resolution imagery using multiresolution analysis and wavelet transform. Wavelet Applications in Signal and Imaging Processing VI, SPIE Proceedings, 3458:103-112. Dal Poz A. P. and M. A. O. Silva, 2002. Active testing and edge analysis for road centreline extraction. International Archives o f Photogrammetiy and Remote Sensing, 34 (3B): B-44 ff. Dekker, A.G., V.E. Brando and J.M. Anstee, 2001. Imaging spectrometry of water. In: Van der Meer, F.D., and S.M. de Jong (Eds.), Imaging Spectrometty: Basic Principles and Prospective Applications. Bookseries Remote Sensing and Digital Image Processing. 4:307 - 359. Dong, H., S.M. Easa, and J. Li. 2003. Approximate extraction of spiraled horizontal cur\'cs from satellite imagery, ASCE., Journal o f Surveying Engineer, (under review). Dong, Y., B. Forster and C. Ticehurst, 1996, Street orientation detection and recognition in Landsat TM and SPOT HRY imagery. Pattern Recognition Letters, 18(8):759-769. 104 Donnay, J.P., M.J. Barnsley and P.A. Longley, (eds.) 2001. Remote Sensing and Urban Analysis, Taylor & Francis, London. Dori, D. and W. Liu, 1998. Stepwise recovery of arc segmentation in complex line environments. International Journal on Document Analysis and Recognition, 1(1):62-71. Dosch, Ph., G. Masini and K. Tombre, 2000. Improving arc detection in graphics recognition. In: Proceedings o f 15th International Conference on Pattern Recognition (ICPR'2000), Barcelona, Spain, pp. 243-246 Doucette, P., P. Agouris, A. Stefanidis, and M. Musavi, 2001. Self-organised clustering for road extraction in classified imagery, ISPRS Journal of Photogrammetiy and Remote Sensing, 55(5-6):347-358. Easa, S., 2002. Geometric Design. Civil Engineering Handbook, W.F. Chen and J.Y. Liew eds., CRC Press, Boca Raton, FL, Chapter 63. Easa, S., H. Dong and J. Li, 2003. Use of satellite imagery to establish simple and reverse highway horizontal curves, ASCE Journal o f Surveying Engineering (accepted). ERDAS, 1999. ERDAS Field Guide. ERDAS, Inc., Atlanta, USA. Fischer, W.A., 1969. Examples of remote sensing applications to engineering. BIBLIOGRAPHY In: Special Report 102, Remote Sensing and Its Application to Highway Engineering, Highway Research Board. Fiset, R., F. Cavayas, M.C. Mouchot, B. Solaiman, and R. Desjardins, 1998. Map-image matching using a multi-layer perception: the case of the road network, ISPRS Journal o f Photogrammetry and Remote Sensing, 53(2):76-84. Freeman, J., 1975. The modeling of spatial relations, Computer Graphics and Image Processing, 4:156-171. Geman, D. and B. Jedynak, 1996. An active testing model for tracking roads in satellite images. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18( 1 ): 1 -14. Gillespie, A.R., A.B. Kahle and R.E. Walker, 1987. Color enhancement of highly correlated images. 1. Decorrelation and HSI Contrast Stretches. Remote Sensing of Environment, 20:209-235. Gonzales, R.C. and R.E. Woods, 1992. Digital Image Processing, Addison-Wesley, Reading, Massachusetts. Gonzalez, R.C. and R.E. Woods, 1993. Digital Image Processing, Addison-Wesley, New York. 105 Gruen, A. and H. Li, 1995a. Road extraction from aerial and satellite images by dynamic programming, JSPRS Journal o f Photogrammetry and Remote Sensing, 50(4): 11 -20. Gruen, A. and H. Li, 1995b. Automation of road extraction from space and aerial images, In Proceeding o f the 16'^ Asian Conference on Remote Sensing, November 20-24, Thailand, pp. E-3-1 to E-3-7. Gruen, A. and H. Li, 1997. Linear feature extraction with 3-D LSB-Snakes, In Automatic Extraction o f Man-Made Objects from Aerial and Space Images (II), Basel, Switzerland, Birkhauser Verlag, pp 287-298. Hay, G.J., T. Blaschke, D.J. Marceau and A. Bouchard, 2003, A comparison of three image-object methods for the multiscale analysis of landscape structure, ISPRS Journal o f Photogrammetry and Remote Sensing, 57:327-345. Heipke, C., C., Steger and R. Multhammer, 1996. A hierarchical approach to automatic road extraction from aerial imagery. In McKeown, D. M. Jr., I. J. Dowman (Eds). Integrating Photogrammetrie Techniques with Scene Analysis and Machine Vision II, Proceedings of SPIE 2486:222-231. Heipke, C., H. Mayer, C. Wiedemann and O. Jamet, 1997. Evaluation of automatie road extraction. In: International Archives o f Photogrammetry and Remote Sensing, 32( 3- 2W3):47-56. Hellwich, O., and C. Wiedemann, 2000. Object extraction from high-resolution multisensor image data. In Proceedings o f the 5'^^ International Conference Fusion oj Earth Data, Sophia Antipolis, January 26-28, pp. 105-115. Hinz, S., A. Baumgartner, C. Steger, H. Mayer, W. Eekstein, H. Ebner and B. Radig, 1999. Road extraction in rural and urban areas. In: Forstner, W., C.E. Liedtke, and J. Büekner (eds.). BIBLIOGRAPHY Semantic Modeling for the Acquisition o f Topographic Information from Images and Maps, SMATI '99, München, pp. 133-153. Hinz, S. and A. Baumgartner, 2003. Automatic extraction of urban road networks from multi-view aerial imagery, ISPRS Journal o f Photogrammetry and Remote Sensing, 58(l-2):83-98. Hofmann, P., 2001. Detecting urban features from IKONOS data using an object-oriented approach. In Proceedings o f P‘ Annual Conference o f the Remote Sensing & Photogrammetry Society, 12-14 September, pp. 28-33. Hyvarinen, A., J. Karhunen and E. Oja, 2001, Independent Component Analysis, John Wiley & Sons, Toronto. ITU-R Recommendation BT.709, 1990. Basic Parameter Values for the HDTV Standard for the Studio and for International Programme Exchange, ITU, 1211 Geneva 20, Switzerland. 106 Jack, K., 1996. Video Demystified. Hightext Publications, Solano Beach, CA 92075. Ji, C.Y., 1996. Delineating agricultural field boundaries from TM imagery using dyadic wavelet transforms, ISPRS Journal o f Photogrammetiy and Remote Sensing, 51(6):268-283. Jung, C.R. and J. Scharcanski, 2003. Adaptive image denoising and edge enhancement in scale-space using the wavelet transform. Pattern Recognition Letters, 24(7):965-971. Kauth, R.J. and G. Thomas, 1976. The Tasselled Cap - a graphic description of the spectral-temporal development of agricultural crops as seen by Landsat In Proceedings o f the Symposium on Machine Processing o f Remotely-Sensed Data 1976, Purdue University, West Lafayette, 46:41-51. Klang, D., 1998. Automatic detection of changes in road databases using satellite imagery. International Archives of Photogrammetry & Remote Sensing, 32:293-298. Roller, T., G.Grieg, G. Szekely and D. Dettwiler, 1995. Multiscale detection of curvilinear structures in 2D and 3D image data. In Proceedings o f the 5' International Conference on Computer Vision, pp 864—869. Rurhar, L., R. Schmidt, S. Dury and A. Skidmore, 2001. Imaging spectrometry and vegetation science. In: Van der Meer F.D., and S.M. de Jong (Eds.), Imaging Spectrometry: Basic Principles and Prospective Applications, 4:111-155. Laptev, I., H. Mayer, T. Lindeberg, W. Eckstein, C. Steger and A. Baumgartner, 2000. Automatic extraction of roads from aerial images based on scale space and snakes. Machine Vision and Applications, 12(1):23-31. Li, J., Y. Li., H. Dong and S. Easa, 2003. Automated extraction of urban road centerlines from IRONOS imagery based on fuzzy mathematical moiphology, Photogrammetrie Engineering and Remote Sensing (under review). Lillesand, T. and R. Riefer, 2000. Remote Sensing and Image Interpretation, 4'’ Edition, John Wiley & Sons, New York. Lindeberg, T., 1998. Edge detection and ridge detection with automatic scale selection, IJCV 30:117-154. Liu, J.G. Mallat, S. and W.L. Hwang, 1992a. Singularity detection and processing with wavelets, IEEE Transaction Information Theory, 38(2):617-643. BIBLIOGRAPHY and J.M. McMoore, 1996, Direct decorrelation stretch technique for RGB colour composition. International Journal o f Remote Sensing, 17:1005-1018. Mallat, S, 1989. A theoiy for multiresolution signal decomposition: The wavelet representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, ll(7):674-693. Mallat, S. and W.L. Hwang, 1992a. Singularity detection and processing with wavelets, IEEE Transaction Information Theory, 38(2):617-643. 107 Mallat, S., 1992b. Characterization of signals from multiscale edges, IEEE Transactions on Pattern Analysis and Machine Intelligence, 14(7):710-732. Mallat, S., 1996. Wavelets for a vision, In Proceedings o f the IEEE, 84(4):604-614. Mallat, S., 1998, A Wavelet Tour of Signal Processing, Academic Press, San Diego Toronto. Marr, D., 1982. Vision, Freeman, San Francisco, CA. Mather, P.M., 1976. Computational Methods o f Multivariate Analysis in Physical Geography, Wiley, New York. Mather, P.M., 1999. Computer Processing o f Remotely-Sensed Images: An Introduction, John Wiley & Sons, New York. Mayer, H., 1996a, Abstraction and scale-space events in image understanding, International Archives o f Photogrammetry and Remote Sensing, 31 (B3/III):523-528. Mayer, H. and Steger, C., 1996b. A New Approach for Line Extraction and its Integration in a Multi-Scale, MultiAbstraction -Level Road Extraction System. In: lAPR TC-7 Workshop: Mapping Buildings, Roads and other Man-Made Structures from Images, Oldenbourg Verlag, Vienna, Austria, pp. 331—348. Mayer, H., A. Baumgartner and C. Steger, 1998a. Tutorial on road extraction from aerial imagery. In: CVonline: The Evolving, Distributed, Non-Proprietary, On-Line Compendium o f Computer Vision. Mayer, H., I. Laptev and A. Baumgartner, 1998b. Multiscale and snakes for automatic road extraction. In: Proceedings o f the 5’’' European Conference on Computer Vision, pp. 720-733. Mayer, H., C. Steger, 1998c. Scale-Space Events and Their Link to Abstraction for Road Extraction, ISPRS Journal o f Photogrammetry and Remote Sensing, 53(2):62-75 Mayer, H., 1999. Automatic object extraction from aerial imagery - a survey focusing on buildings. Computer Vision and Image Understanding, 74(2):138-149. McKeown, D.M., and J.L. Denlinger, 1988. Cooperative methods for road tracking in aerial imagery. Computer Vision and Pattern Recognition, pp. 662-672. Park, S.R. and T. Kim, 2001. Semi-automatic road extraction algorithm from IKONOS images using template matching. In Proceedings o f the 22"^ Asian Conference on Remote Sensing, 5-9 November, Singapore. Phasomkusolsil, S., N. Hinsamooth, F. Cheevasuvit, K. Dejhan, S. Mittatha, S. Chitwong and A. Somboonkaew, 1998, Principal component analysis for multi-resolution 108 images, In Proceedings o f Asian Conference on Remote Sensing. (http://www.gisdevelopment.com/aars/acrs/1998/ps3/ps3018pf.htm). images, In Proceedings o f Asian Conference on Remote Sensing. BIBLIOGRAPHY (http://www.gisdevelopment.com/aars/acrs/1998/ps3/ps3018pf.htm). Price, K., 1999. Road grid extraction and verification. International Archives o f Photogrammetry and Remote Sensing, 32(3-2W5):I01-106. Quegan, S. and 1. Rhodes, 1994. Relating plarimetric SAR data to surface properties — the MAC-Europe experiment. In: Mather, P.M. (ed.), 159-174. Ranchi, T., L. Wald and M, Mangolini, 2001. Improving the spatial resolution of remotely-sensed images by means of sensor fusion: a general solution using the ARSIS method. In: Donnay, J.P., M.J. Barnsley, and P.A. Longley, 2001. Remote Sensing and Urban Analysis, Taylor & Francis, London. Rees, W.G., 1999. The Remote Sensing Data Book, Cambridge University Press, Cambridge, England. Richards, J.A. and X. Jia, 1999. Remote Sensing Digital Image Analysis, 3” edition. Springer-Verlag, Berlin. Roger, R.E., 1996. Principal components transform with simple, automatic noise adjustment, InternationalJournal o f Remote Sensing, 17:2719-2727. Rosin, P. L. and G.A.West, 1989. Segmentation of edges into lines and arcs. Image and Vision Computing, 7(2): 109-114. Ruskoné, R. and S. Airault, 1997. Toward an automatic extraction of the road network by local interpretation of the scene. Photogrammetrie Week, Wichmann Verlag, Heidelberg, Germany, pp. 147-157. Sânchez-Âvila, C., 2003, Wavelet domain signal deconvolution with singularity- preserving regularization. Mathematics and Computers in Simulation, 61(3-6): 165- 176. Shukla, V., R. Chandrakanth and R. Ramachandran, 2002. Semi-automatic road extraction algorithm for high resolution images using path following approach. In Proceedings o f the 3'''^ Indian Conference on Computer Vision, Graphics and Image Processing, Ahmedabad, India. Sonka, M., V. Hlavac and R. Boyle, 1999. Image Processing, Analysis, and Machine Vision, 2"** edition, Brooks-Cole, Pacific Grove, CA. Steger, C., 1996. Extraction of curved lines from images. In Proceedings o f the 13"’ International Conference on Pattern Recognition, pp. 251-255. Steger, C., H. Mayer and B. Radig, 1997. The role of grouping for road extraction. In: A. Gmen, E. Baltsavias and O. Henricsson (eds). Automatic Extraction o f Man-Made 109 Objects from Aerial and Space Images (II), Birkhauser Verlag, Basel, Switzerland, pp. 245-256. Objects from Aerial and Space Images (II), Birkhauser Verlag, Basel, Switzerland, pp. 245-256. Torre, V. and T.A. Poggio, 1980, On edge detection, IEEE Transaction on Pattern Analysis and Machine Intelligence, 8(2):147-163. Trinder, J.C. and Y. Wang, 1998. Automatic road extraction from aerial images. Digital Signal Processing, 8(4):215-224. Vosselman, G. and J. de Knecht, 1995. Road tracing by profile matching and Kalman filtering. In Proceedings Workshop on Automatic Extraction o f Man-Made Objects from Aerial and Space Images, Birkhauser Verlag, Basel, Switzerland, pp. Zlotnick, A. and P.D. Gamine, 1993. Finding road seeds in aerial images. Computer Vision and Image Understanding, 57(2):243-260. BIBLIOGRAPHY 265-274. Wang, Y. and Q. Zheng, 1998, Recognition of roads and bridges in SAR images. Pattern Recognition, 31(7):953-962. Wiedemann, C. and H. Mayer, Automatic verification of roads in digital images using profiles, Mustererkennung 1996, Springer-Verlag, Berlin, Germany, pp. 609-618. Wiedemann, C., C. Heipke, H. Mayer and O. Jamet, 1998. Empirical evaluation of automatically extracted road axes. In: K. J. Bowyer and P. J. Phillips (eds). Empirical Evaluation Methods in Computer Vision, IEEE Computer Society Press, Los Alamitos, California, pp. 172-187. Wiedemann, C. and S. Hinz, 1999. Automatic extraction and evaluation of road networks from satellite imagery. International Archives of Photogrammetry and Remote Sensing, 32(3-2W5):95-100. Wiedemann, C. and H. Ebner, 2000, Automatic completion and evaluation of road networks. International Archives o f Photogrammetry and Remote Sensing, 33(Part B):979-986. Wiemker, R., A. Speck, D. Kulbach, H. Spitzer and J. Beinlein, 1997. Unsupervised robust change detection on multispectral imagery using spectral and spatial features. In Proceedings o f the 3'^ International Airborne Remote Sensing Conference and Exhibition, Copenhagen, Denmark, 1:640-647. Wilkinson, G.G., 1996. A review of current issues in the integration of GIS and remote sensing data. International Journal o f Geographical Information Systems, 10(I):85- 101. Xiong, D., 2001. Optimization-based method for road network extraction, In Proceedings o f Road Centerline Extraction & Maintenance Specialist Meeting {CLEM200I), August 6-7, Santa Barbara, CA. (http://www.ncgia.ucsb.edu/ncrst/meetings/clem2001/proceedings.html) Zhang, C:, E. Baltsavias and A. Gruen, 2001. Knowledge-based image analysis for 3D road reconstruction, Asian Journal o f Geoinfomatics, 1(4):3-14. 1 1 0 Zhang, L. and P. Bao, 2002. Edge detection by scale multiplication in wavelet domain, Pattern Recognition Letters, 23(14): 1771-1784. Zhang, L. and P. Bao, 2002. Edge detection by scale multiplication in wavelet domain, Pattern Recognition Letters, 23(14): 1771-1784. Zhang, C., 2003. Updating of cartographic road databases by images analysis, PhD Thesis, November, ETH Zurich. Ziou, D. and A. Koukam, 1995. The selection of edge detectors using local image structure. In Proceedings o f the 7''' IEEE International Conference on Tools with Artificial Intelligence, Virginia, pp. 366-370. Ziou, D. and S. Tabbone, 1997. Edge Detection Techniques - An Overview, Technical Report, No. 195, Department of Mathematics & Informatics, University of Sherbrooke, Quebec. Zlotnick, A. and P.D. Gamine, 1993. Finding road seeds in aerial images. Computer Vision and Image Understanding, 57(2):243-260. I ll
https://openalex.org/W4238450744	https://www.solid-earth.net/6/1207/2015/se-6-1207-2015.pdf	English	null	Poroelastic responses of confined aquifers to subsurface strain changes and their use for volcano monitoring	null	2,015	cc-by	20,479	Poroelastic responses of conﬁned aquifers to subsurface strain and their use for volcano monitoring K. Strehlow, J. H. Gottsmann, and A. C. Rust School of Earth Sciences, University of Bristol, Wills Memorial Building, Bristol BS8 1RJ, UK K. Strehlow, J. H. Gottsmann, and A. C. Rust School of Earth Sciences, University of Bristol, Wills Memorial Building, Bristol BS8 1RJ, UK Received: 11 May 2015 – Published in Solid Earth Discuss.: 9 June 2015 Revised: 18 September 2015 – Accepted: 21 October 2015 – Published: 10 November 2015 Abstract. Well water level changes associated with mag- matic unrest can be interpreted as a result of pore pressure changes in the aquifer due to crustal deformation, and so could provide constraints on the subsurface processes caus- ing this strain. We use ﬁnite element analysis to demonstrate the response of aquifers to volumetric strain induced by pres- surized magma reservoirs. Two different aquifers are invoked – an unconsolidated pyroclastic deposit and a vesicular lava ﬂow – and embedded in an impermeable crust, overlying a magma chamber. The time-dependent, fully coupled models simulate crustal deformation accompanying chamber pres- surization and the resulting hydraulic head changes as well as ﬂow through the porous aquifer, i.e. porous ﬂow. The simulated strain leads to centimetres (pyroclastic aquifer) to metres (lava ﬂow aquifer) of hydraulic head changes; both strain and hydraulic head change with time due to substan- tial porous ﬂow in the hydrological system. aquifer and are commonly neglected in analytical models. These ﬁndings highlight the need for numerical models for the interpretation of observed well level signals. However, simulated water table changes do indeed mirror volumetric strain, and wells are therefore a valuable addition to monitor- ing systems that could provide important insights into pre- eruptive dynamics. 1 Introduction Pre-, syn- and post-eruptive changes in water levels have been reported for several volcanoes (Newhall et al., 2001). Many processes can lead to water level changes, including meteorological inﬂuences (e.g. rainfall or barometric pres- sure), the injection of magmatic ﬂuids (e.g. suggested for Campi Flegrei by Chiodini et al., 2012), the opening and closing of fractures (e.g. proposed for Kilauea by Hurwitz and Johnston, 2003) or heating of the aquifer (suggested for Campi Flegrei by Gaeta et al., 1998). Well level changes are particularly sensitive to cham- ber volume, shape and pressurization strength, followed by aquifer permeability and the phase of the pore ﬂuid. The depths of chamber and aquifer, as well as the aquifer’s Young’s modulus also have signiﬁcant inﬂuence on the hy- draulic head signal. While source characteristics, the distance between chamber and aquifer and the elastic stratigraphy de- termine the strain ﬁeld and its partitioning, ﬂow and coupling parameters deﬁne how the aquifer responds to this strain and how signals change with time. p g y Here, we focus on the very commonly suggested mech- anism of strain-induced water level changes. Examples in- clude well level changes of more than 9 m preceding the 2000 eruption of Usu volcano, Japan (Matsumoto et al., 2002) and the water level rise of more than 85 m in a geothermal well at Kraﬂa volcano, Iceland, associated with a dyke intrusion in 1977 (Stefansson, 1981). The observations can be explained by poroelasticity (Wang, 2000). Compression or dilatation of an elastic porous medium leads to a decrease or an increase in pore space, respectively, which in turn inﬂuences the pore pressure and thereby the water level. Hence, measured well level changes can be interpreted as strain-induced changes in pore pressure in the aquifer due to crustal deformation. This We ﬁnd that generic analytical models can fail to capture the complex pre-eruptive subsurface mechanics leading to strain-induced well level changes, due to aquifer pressure changes being sensitive to chamber shape and lithological heterogeneities. In addition, the presence of a pore ﬂuid and its ﬂow have a signiﬁcant inﬂuence on the strain signal in the Poroelastic responses of conﬁned aquifers to subsurface stra and their use for volcano monitoring K. Strehlow, J. H. Gottsmann, and A. C. Rust School of Earth Sciences, University of Bristol, Wills Memorial Building, Bristol BS8 1RJ, UK Correspondence to: K. Strehlow (karen.strehlow@bristol.ac.uk) Received: 11 May 2015 – Published in Solid Earth Discuss.: 9 June 2015 Revised: 18 September 2015 – Accepted: 21 October 2015 – Published: 10 November 2015 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1208 is in line with observations of water level changes accom- panying seismic events (e.g. Roeloffs, 1996; Jonsson et al., 2003; Shibata et al., 2010) or crustal spreading, as observed at the Juan de Fuca Ridge (Davis et al., 2001). due to the injection of hot magmatic ﬂuids, using one-way coupling of solid deformation and porous ﬂow (e.g. Tode- sco et al., 2004; Hurwitz et al., 2007; Hutnak et al., 2009; Rinaldi et al., 2010; Fournier and Chardot, 2012). Rutqvist et al. (2002) have developed a two-way coupled code and applied it to problems related to carbon dioxide injection in aquifers and the disposal of nuclear waste in porous media. Whether and how the pure deformational strain caused by magma bodies would induce water level changes has not yet been explored numerically. The full, two-way coupling of ﬂuid and solid mechanics required has so far been avoided in volcanological applications, and so the effect of solid defor- mation on pore pressure and porous ﬂow has been neglected. In volcanic environments, many processes can lead to sub- stantial strain, including pressure changes in magma reser- voirs and intruding dykes. Information about the local strain ﬁeld is therefore highly valuable for volcano monitoring and eruption forecasting, as it could allow derivation of these sub- surface magmatic processes (e.g. Voight et al., 2006; Linde et al., 2010; Bonaccorso et al., 2012). However, strain data are difﬁcult to interpret and strainmeters are complex and expensive installations. The described poroelastic relations raise the question whether wells in aquifers can provide ad- ditional information on the subsurface strain ﬁeld and if we could even use them as cheaper and somewhat simpler strain- meters. We investigate the phenomenon of poroelastic responses to magmatic strain to better understand the hydrological sig- nals one might observe in wells on a volcano before and dur- ing eruptions. We assess to what extent conﬁned aquifers can serve as indicators of stress/strain partitioning in the shallow crust due to reservoir pressure changes and therefore if they could provide a tool to scrutinise pre-eruption processes. Previous studies have indeed utilized the poroelastic be- haviour of aquifers to infer magmatic processes from ob- served water level changes at volcanoes (e.g. Shibata and Akita, 2001; Takahashi et al., 2012). 2.1 Theory We present a set of generic models using ﬁnite element anal- ysis to perform parametric studies on several volcanic set- tings with an inﬂating magma chamber affecting overlying rock layers and hydrology. The models solve a series of con- stitutive equations that result from the full coupling of contin- uum mechanics equations for stress–strain relations of a lin- ear elastic material with Darcy’s law and mass conservation within the porous ﬂow theory (for details see Biot, 1962; Wang, 2000; COMSOL, 2013). The calculations are based on the Navier equation for a solid: However, oversimpliﬁcation of the coupling between solid and ﬂuid mechanics may make these models inadequate. An example is the 2000 Usu eruption (Matsumoto et al., 2002), where pre-eruptive water table changes of several me- tres were observed in two wells at different locations simul- taneously with a radial ground deformation of about 2 cm recorded about 8 km from the summit over the course of 2 weeks prior to the eruption. The water level changes were interpreted as a result of this crustal deformation. However, the two different wells apparently give inconsistent informa- tion about the source of strain: only one of the two well level changes agrees with the model proposed by Matsumoto et al. (2002). In order to make reasonable monitoring interpreta- tions based on well level data, we therefore need to improve our understanding of how these hydrological signals are gen- erated and identify the relative importance of the parameters that affect them. Changes in the hydrological conditions in volcanic areas are usually interpreted as a result of changes in the magmatic system, but the effects of non-magmatic pa- rameters on the pressure-response in the aquifer should also be considered. −∇· σ = F V , (1) (1) −∇· σ = F V , with σ being the stress tensor and F V a body force. Iner- tia terms in the Navier equation are neglected as the solid deformation is treated as quasi-static. The solid mechanics equations assume linear elasticity and do not allow for ma- terial failure, hence only work for sufﬁciently small strains. The stress tensor σ is related to the strain tensor ϵ and the pore pressure pf by a generalized Hooke’s law: σ −σ0 = C : (ϵ −ϵ0) −αpfI. (2) (2) Here, C is the drained elasticity tensor and α is the Biot– Willis coefﬁcient. Table 1 gives a list of all symbols used in this study. Table 1 gives a list of all symbols used in this study. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain A method of assess- ing the strain sensitivity of an aquifer is to track water level changes as a result of predictable excitations such as Earth tides or measured barometric variations. The known strain sensitivity is then used to derive volumetric strain from ob- served water level changes during unrest, and combining this with analytical deformation models such as the Mogi model (Mogi, 1958), inferences can be made on magmatic drivers behind the level changes. Published by Copernicus Publications on behalf of the European Geosciences Union. Published by Copernicus Publications on behalf of the European Geosciences Union. Fluid ﬂow is described by mass conservation In both Eqs. (2) and (4), the terms including the Biot– Willis coefﬁcient describe the coupling between solid de- formation and ﬂuid ﬂow, which manifests in stress absorp- tion by the ﬂuid and pore pressure changes due to the increase/decrease of pore space resulting from volumetric changes of the porous medium. The coupling parameter α is a measure of the strength of this coupling (with values between the porosity of the medium and 1), and is deﬁned by the volume of ﬂuid expelled from/sucked into a porous medium when subject to volumetric change. Fluid mechani- cal properties remain unchanged, including permeability and porosity. The coupling is achieved solely by the pore pres- sure and stress effects, as well as the expression for the spe- ciﬁc storage, which includes elastic properties of both pore ρfS ∂pf ∂t + ∇· (ρfv) = Q −ρfα ∂ϵvol ∂t (4) and Darcy’s law: v = −κ µ(∇pf + ρfg∇D). (5) ρfS ∂pf ∂t + ∇· (ρfv) = Q −ρfα ∂ϵvol ∂t (4) and Darcy’s law: v = −κ µ(∇pf + ρfg∇D). (5) ρfS ∂pf ∂t + ∇· (ρfv) = Q −ρfα ∂ϵvol ∂t (4) and Darcy’s law: v = −κ µ(∇pf + ρfg∇D). (5) (5) Here, ρf is the ﬂuid density, S is the speciﬁc storage, v is ﬂuid ﬂow velocity, Q is a source/sink term, ϵvol is the volumetric strain, κ is aquifer permeability, µ is water viscosity, ρf is water density, g is acceleration of gravity, and D is eleva- tion. The equations for ﬂuid ﬂow only consider single-phase, single-component ﬂow. The aquifer is considered to be fully saturated and perfectly conﬁned at all times. Here, ρf is the ﬂuid density, S is the speciﬁc storage, v is ﬂuid ﬂow velocity, Q is a source/sink term, ϵvol is the volumetric strain, κ is aquifer permeability, µ is water viscosity, ρf is water density, g is acceleration of gravity, and D is eleva- tion. The equations for ﬂuid ﬂow only consider single-phase, single-component ﬂow. The aquifer is considered to be fully saturated and perfectly conﬁned at all times. Solid Earth, 6, 1207–1229, 2015 2.1 Theory Strain is given through the displacement vector (u): Analytical solutions exist for only a few comparatively simple poroelastic problems (e.g. Rice and Cleary, 1976). Numerical modelling of pressure changes in hydrological systems has focused on pressure and temperature transients in hydrothermal systems and resulting ground deformation ϵ = 1 2[(∇u)T + ∇u + (∇u)T∇u]. (3) (3) Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain Table 1. Symbols. b Vertical semi-axis of ellipsoidal chamber (m) ϵvol Volumetric strain (1) C Drained elasticity tensor (Pa) κ Permeability of the aquifer (m2) daq Aquifer thickness (m) µ Viscosity of pore ﬂuid (Pas) dc Cap rock thickness (m) νaq Drained Poisson’s ratio of the aquifer (1) dist Distance aquifer – magma chamber (m) νc Poisson’s ratio of the cap rock (1) distcﬂip dist-value that changes sign of strain in the aquifer (m) νh Poisson’s ratio of the host rock (1) D Elevation (m) ρaq Drained density of the aquifer (kgm−3) Eaq Drained Young’s modulus of the aquifer (Pa) ρc Density of the cap rock (kgm−3) Ec Young’s modulus of the cap rock (Pa) ρf Density of pore ﬂuid (kgm−3) Eh Young’s modulus of the host rock (Pa) ρh Density of the host rock (kgm−3) ERc Ratio of cap rock to aquifer stiffness (1) σ Stress tensor (Pa) ERcﬂip ERc-value that changes sign of strain in the aquifer (1) 8 Porosity of the aquifer (1) ERh Ratio of host rock to aquifer stiffness (1) χf Compressibility of pore ﬂuid (Pa−1) F V Body force (N) g Acceleration of gravity (ms−2) h Hydraulic head (m) 1h Hydraulic head change (m) 1href Hydraulic head change in reference simulation (m) I Unity matrix (1) K Drained bulk modulus of the aquifer (Pa) L Radial distance domain centre – aquifer (m) 1P Magma chamber pressurization (Pa) pf Fluid pore pressure (Pa) Q Source/sink (kgm−3 s−1) r Radius of the spherical magma chamber (m) S Speciﬁc storage (Pa−1) Tf Temperature of pore ﬂuid (◦C) t Time (s) u Displacement (m) V Magma chamber volume (m3) v Fluid ﬂow velocity (ms−1) z z-coordinate (m) zaq Depth of aquifer top (m) zCH Depth of magma chamber top (m) α Biot–Willis coefﬁcient (1) β Magma compressibility (Pa−1) ϵ Strain (1) K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1209 Fluid ﬂow is described by mass conservation K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain ﬂuid and solid matrix: and assuming a magma compressibility of β = 10−11 Pa−1, a pressurization of 10 MPa could correspond to a volume change of 1V = 100 000m3. 1V is not simulated in the presented models, as the magma chamber is represented by a pressurized cavity. The volume change calculated with Eq. (7) serves as a ﬁrst order estimation and a guide to corresponding magmatic processes. Note however, that this equation does not account for several additional processes and does not deliver a perfectly accurate volume change of the source. The resulting deformation of the surround- ing material is calculated by discretizing the model domain to solve the constitutive equations for continuum mechanics for stress–strain relations of a linear elastic material. Bound- ary conditions are also taken from Hickey and Gottsmann (2014): the Earth’s surface is treated as a free surface, the bottom boundary is ﬁxed and the lateral boundary has a roller condition (free lateral, but no vertical displacement). We then adapt this model set-up for our purposes by adding a shallow, rectangular, poroelastic aquifer, which is saturated with wa- ter. The internal boundary conditions bordering the aquifer domain are (a) no ﬂow and (b) continuous stress and dis- placement. Note that changing the lateral aquifer boundary condition to a ﬁxed pressure instead of ”no ﬂow”, hence al- lowing water to leave or enter the domain, does not change the results of this study. The initial pore pressure is set as hy- drostatic: pf = ρfgz, with z being the depth coordinate. The duration of the time-dependent simulation is 1000 days. We solve the full set of coupled equations, giving solid displace- ment u and ﬂuid pore pressure pf. To demonstrate its mean- ing for water table changes that could be observed during volcanic unrest, we present all results as hydraulic head h, which is proportional to pore pressure: S = φχf + (α −φ)(1 −α) K , (6) (6) with φ being the porosity of the medium, χf the ﬂuid com- pressibility, and K the drained bulk modulus of the solid ma- trix. This set of equations is solved for solid deformation (u) and ﬂuid pressure (pf) using the structural mechanics and Darcy’s law modules of the ﬁnite element analysis soft- ware package COMSOL Multiphysics, version 5.0. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain These modules provide the equations for solid deformation and ﬂuid ﬂow, respectively, and have been coupled by manu- ally including the coupling terms in Eqs. (2) and (4) as de- scribed above and appropriately deﬁning the speciﬁc storage. It should be noted that the readily provided poroelasticity module (which applies the same equations) has been avoided, as this causes problems when gravity is not neglected. In the provided coupling, COMSOL treats solid and ﬂuid as one unit, meaning they compute the average density of the porous domain and gravity then acts on this. This causes downward displacement of the solid matrix and the gravity part of Darcy’s law causes downward porous ﬂow. While it is possible to equilibrate this ﬂow by initializing hydrostatic pressure (using ﬂuid density), this does not prevent the solid displacement, as this needs to be balanced using initial litho- static pressure (using average density). As one needs to use different densities it is not possible to equilibrate ﬂow and displacement at the same time in this module when gravity is turned on. In our solution, we switch on gravity only in the Darcy’s law module, hence the solid matrix is not affected by the gravitational stresses, and the ﬂuid pressure is initialized as hydrostatic. This way we can take ﬂow following topo- graphic gradients into account. h = pf ρf ∗g −z. (8) h = pf ρf ∗g −z. (8) www.solid-earth.net/6/1207/2015/ 1210 www.solid-earth.net/6/1207/2015/ 2.2 Model set-up It represents the maximum water level change in a small diameter well (ideally a piezometer) in a conﬁned aquifer; its initial value is 0 m. As a well moves with the ground, we subtract the vertical ground displacement from this hy- draulic head change prior to using the data in order to obtain the relative water level change that would be measured in a well. Note here that, when “initial” responses are shown, the time referred to is 10−6 s. The ﬁnal model set-up is shown in Fig. 1; reference values of geometric parameters can be found in Table 2. As a starting point to investigate hydrological responses to magma chamber inﬂation, we build a 2-D-axisymmetric model geometry in COMSOL Multiphysics following Hickey and Gottsmann (2014), who provide guidelines for volcano deformation modelling using ﬁnite element analy- sis. The initial model consists of a linear elastic solid block with an embedded spherical cavity, representing a magma chamber at depth. This cavity is pressurized by applying a boundary load, which is stepped up over 10−8 s. We as- sume almost instantaneous pressurization for simplicity and to more easily recognise the different inﬂuences of parame- ters. Magma chamber pressurization can be generated by the injection of fresh magma, vesiculation, thermal expansion of the magma, melting of country rocks or volume changes dur- ing crystallization (Fagents et al., 2013). Using the relation for temperature-independent volume changes The linear elastic material surrounding the magma cham- ber, from here on called “host rock”, has elastic properties of a general granitic crust. Depth- and temperature-dependent changes of Young’s modulus of the crust are ignored for sim- plicity. We chose a Young’s modulus of 30 GPa, representing an average value for crustal rocks in arcs for depths up to 8 km as derived from seismic velocity data (Gottsmann and Odbert, 2014). We test for two typical aquifer types found in volcanic regions: unconsolidated pyroclastic deposits, com- monly composed of coarse ash to ﬁne lapilli sized clasts, and 1P = 1 β 1V V (7) Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain Table 2. Input parameters: reference values and ranges for parametric studies (where performed). 2.2 Model set-up Note that elastic prop- erties of poroelastic layers are always required to be the drained parameters (i.e. measured under constant pore pres- sure). However, very few data exist on poroelastic parameters so we used the dry Young’s moduli and Poisson’s ratios in- stead and increased respective ranges in parametric sweeps to account for this unknown error. Some experimental data on the elastic properties of porous volcanic rocks are provided by Heap et al. (2014) for permeable tuff in the Neapolitan area; they fall within the here explored range of parameters. Within the different layers, material properties are consid- ered isotropic and homogeneous. Standard water parameters are also given in Table 2. 2.2 Model set-up Parameter Reference value Range Aquifer depth zaq 200 m 100–2200 m Aquifer thickness daq 200 m 50–450 m Chamber top depth zCH 3 km 2–5 km Chamber radius (spherical) r 1 km 0.5–1.5 km Distance chamber – aquifer dist 2.6 km 1.6–4.6 km Vertical semi-axis b 1 km 0.25–2 km Aquifer lateral onset L 0 km 0–8 km Cap rock Young’s modulus Ec 70 MPa 0.01–10 GPa Host rock Young’s modulus Eh 30 GPa 0.1–100 GPa Aquifer Young’s modulus – pyroclastic Eaq 10 MPa 0.5–100 MPa Aquifer Young’s modulus – lava ﬂow Eaq 50 GPa 0.5–100 GPa Cap rock Poisson’s ratio νc 0.45 Host rock Poisson’s ratio νh 0.25 Aquifer Poisson’s ratio – pyroclastic νaq 0.275 0.15–0.4 Aquifer Poisson’s ratio – lava ﬂow νaq 0.225 0.1–0.35 Cap rock densityρc 1800 kgm−3 Host rock densityρh 2600 kgm−3 Aquifer density – pyroclastic ρaq 2000 kgm−3 Aquifer density – lava ﬂow ρaq 2800 kgm−3 Aquifer permeability – pyroclastic κ 5 × 10−11 m2 10−14 – 10−7 m2 Aquifer permeability – lava ﬂow κ 5 × 10−12 m2 10−14 – 10−9 m2 Aquifer porosity – pyroclastic φ 0.35 Aquifer porosity – lava ﬂow φ 0.1 Biot–Willis coefﬁcient – pyroclastic α 0.7 0.45–1 Biot–Willis coefﬁcient – lava ﬂow α 0.2 0.1–1 Water density ρf 1000 kgm−3 changed acc. to temperature changes, see Table 3 Water viscosity µ 10−3 Pas changed acc. to temperature changes, see Table 3 Water compressibility χf 4 × 10−10 Pa−1 changed acc. to temperature changes, see Table 3 Pressurization value 1P 10 MPa 1–100 MPa 1211 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain vesicular basaltic lava ﬂows. Vesicular here means a sufﬁ- ciently connected porosity of the lava to serve as an aquifer; note that while the petrological porosity might be higher, only the connected pores matter for the ﬂuid ﬂow. These two types differ substantially in their elastic and ﬂuid ﬂow prop- erties, which have signiﬁcant inﬂuence on the observed sig- nals. The layer above the aquifer, from here on called “cap rock”, has elastic properties of a soft, impermeable clay. In- put material properties for the reference simulation are given in Table 2; we used medians of parameter ranges found in the literature (Freeze and Cherry, 1979; Fetter, 1994; Wang, 2000; Gercek, 2007; Gudmundsson, 2011; Adam and Oth- eim, 2013; Geotechdata.info, 2013). 2.3 Parametric studies and sensitivity analysis In the parametric studies we investigated the effects of mag- matic source properties as well as poroelastic and geomet- ric properties of the aquifer (Table 2). Ranges for material properties of the aquifer were taken from literature (Freeze and Cherry, 1979; Fetter, 1994; Wang, 2000; Gercek, 2007; Gudmundsson, 2011; Adam and Otheim, 2013; Geotech- data.info, 2013). Ranges for geometric parameters can of course never cover the whole natural variation, we attempted to cover a reasonable range to be able to make general statements about the inﬂuence of certain parameters. A K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1212 Table 3. Temperature dependent water properties for a pressure of 4.5 MPa (calculated using Verma, 2003). able 3. Temperature dependent water properties for a pressure of 4.5 MPa (calculated using Verma, 2003). Table 3. Temperature dependent water properties for a pressure of 4.5 MPa (calculated using Verma, 2003). Table 3. Temperature dependent water properties for a pressure of 4.5 MPa (calculated using Verma, 2003). Temperature Density ρf Viscosity µ Compressibility χf (◦C) (kgm−3) (Pas) (Pa−1) 10 1001.80 1.30 × 10−3 4.73 × 10−10 40 994.14 6.53 × 10−4 4.37 × 10−10 70 979.70 4.05 × 10−4 4.46 × 10−10 100 960.40 2.83 × 10−4 4.83 × 10−10 200 866.89 1.35 × 10−4 8.64 × 10−10 300 19.46 1.98 × 10−5 2.60 × 10−7 400 15.44 2.44 × 10−5 2.38 × 10−7 500 13.07 2.87 × 10−5 2.31 × 10−7 20km r zCH 20km daq zaq Host rock: Eh,νh,ρh Aquifer: Eaq,νaq,ρaq,κ,φ,α,μf,χf,ρf,Tf Cap rock: Ec,νc,ρc ΔP Roller free surface fixed boundary Axis of Symmetry L b Figure 1. 2-D axisymmetric model set-up: a boundary load 1P is applied on a cavity at depth, with the radius r for the spherical case or vertical semi-axis b for the ellipsoidal case, respectively. This strains the surrounding linear elastic host rock (granitic crust), the poroelastic aquifer and the overlying linear elastic cap rock (clay). The water-saturated aquifer is modelled as either a vesicular lava ﬂow or unconsolidated pyroclasts. An aquifer not covering the chamber but starting at some lateral distance L is realized by setting the darker grey region impermeable. The bottom boundary is ﬁxed, the upper boundary is treated as a free surface, the lateral bound- aries have a roller condition. There is no ﬂow outside the aquifer; stress and displacement at the internal boundaries are continuous. An extract of the ﬁnite element mesh is shown only for illustration. The mesh density is ﬁner around the cavity, at aquifer boundaries and the free surface. 20km r zCH 20km daq zaq Host rock: Eh,νh,ρh Aquifer: Eaq,νaq,ρaq,κ,φ,α,μf,χf,ρf,Tf Cap rock: Ec,νc,ρc ΔP Roller free surface fixed boundary Axis of Symmetry L b 20km r zCH 20km daq zaq Host rock: Eh,νh,ρh Aquifer: Eaq,νaq,ρaq,κ,φ,α,μf,χf,ρf,Tf Cap rock: Ec,νc,ρc ΔP Roller free surface fixed boundary Axis of Symmetry L b to recognise general patterns. www.solid-earth.net/6/1207/2015/ www.solid-earth.net/6/1207/2015/ Solid Earth, 6, 1207–1229, 2015 1212 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain Table 3. Temperature dependent water properties for a pressure of 4.5 MPa (calculated using Verma, 2003). Temperature Density ρf Viscosity µ Compressibility χf (◦C) (kgm−3) (Pas) (Pa−1) 10 1001.80 1.30 × 10−3 4.73 × 10−10 40 994.14 6.53 × 10−4 4.37 × 10−10 70 979.70 4.05 × 10−4 4.46 × 10−10 100 960.40 2.83 × 10−4 4.83 × 10−10 200 866.89 1.35 × 10−4 8.64 × 10−10 300 19.46 1.98 × 10−5 2.60 × 10−7 400 15.44 2.44 × 10−5 2.38 × 10−7 500 13.07 2.87 × 10−5 2.31 × 10−7 20km r zCH daq zaq Host rock: Eh,νh,ρh Aquifer: Eaq,νaq,ρaq,κ,φ,α,μf,χf,ρf,Tf Cap rock: Ec,νc,ρc ΔP Roller free surface Axis of Symmetry L b to recognise general patterns. As mentioned above, instanta- neous pressurization is assumed in the simulations for sim- plicity, however chambers in reality will more likely pres- surise over longer time periods. We therefore compare hy- draulic head changes produced by the reference simulation with the poroelastic response to a chamber that inﬂates over 100 days. When sweeping over one parameter, all others are kept constant. This entails that in all geometric sweeps, the dis- tance between magma chamber top and aquifer was ﬁxed, except for the sweep over magma chamber depth because this distance is such an important parameter it would have other- wise overwhelmed the pure effects of, for example, aquifer thickness. When investigating the effects of magma chamber shape, we changed the vertical semi-axis b of an ellipsoidal chamber, which then deﬁnes the horizontal semi-axis via the K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain As mentioned above, instanta- neous pressurization is assumed in the simulations for sim- plicity, however chambers in reality will more likely pres- surise over longer time periods. We therefore compare hy- draulic head changes produced by the reference simulation with the poroelastic response to a chamber that inﬂates over 100 days. y When sweeping over one parameter, all others are kept constant. This entails that in all geometric sweeps, the dis- tance between magma chamber top and aquifer was ﬁxed, except for the sweep over magma chamber depth because this distance is such an important parameter it would have other- wise overwhelmed the pure effects of, for example, aquifer thickness. When investigating the effects of magma chamber shape, we changed the vertical semi-axis b of an ellipsoidal chamber, which then deﬁnes the horizontal semi-axis via the constant chamber volume. Pore ﬂuid (H2O) temperature was effectively changed by varying its density, viscosity and com- pressibility. We used the program provided by Verma (2003) to calculate these parameters for varying temperatures and a pressure of 4.5 MPa, which represents average lithostatic pressure in the aquifer that was kept at 200 m depth (Table 3). In a subset of simulations, the central portion of the aquifer is replaced with an area of zero permeability out to a radial distance L but with the same poroelastic properties as the aquifer, to avoid numerical errors at the inner boundary of the aquifer. Aquifer density and porosity have a negligible inﬂuence and have not been included in the parametric study results. Figure 1. 2-D axisymmetric model set-up: a boundary load 1P is applied on a cavity at depth, with the radius r for the spherical case or vertical semi-axis b for the ellipsoidal case, respectively. This strains the surrounding linear elastic host rock (granitic crust), the poroelastic aquifer and the overlying linear elastic cap rock (clay). The water-saturated aquifer is modelled as either a vesicular lava ﬂow or unconsolidated pyroclasts. An aquifer not covering the chamber but starting at some lateral distance L is realized by setting the darker grey region impermeable. The bottom boundary is ﬁxed, the upper boundary is treated as a free surface, the lateral bound- aries have a roller condition. There is no ﬂow outside the aquifer; stress and displacement at the internal boundaries are continuous. An extract of the ﬁnite element mesh is shown only for illustration. www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain The mesh density is ﬁner around the cavity, at aquifer boundaries and the free surface. Figure 1. 2-D axisymmetric model set-up: a boundary load 1P is applied on a cavity at depth, with the radius r for the spherical case or vertical semi-axis b for the ellipsoidal case, respectively. This strains the surrounding linear elastic host rock (granitic crust), the poroelastic aquifer and the overlying linear elastic cap rock (clay). The water-saturated aquifer is modelled as either a vesicular lava ﬂow or unconsolidated pyroclasts. An aquifer not covering the chamber but starting at some lateral distance L is realized by setting the darker grey region impermeable. The bottom boundary is ﬁxed, the upper boundary is treated as a free surface, the lateral bound- aries have a roller condition. There is no ﬂow outside the aquifer; stress and displacement at the internal boundaries are continuous. An extract of the ﬁnite element mesh is shown only for illustration. The mesh density is ﬁner around the cavity, at aquifer boundaries and the free surface. Figure 1. 2-D axisymmetric model set-up: a boundary load 1P is applied on a cavity at depth, with the radius r for the spherical case or vertical semi-axis b for the ellipsoidal case, respectively. To investigate the importance of parameters on hydraulic head change, we performed a sensitivity analysis. The in- ﬂuence of lateral distance L between magma chamber and aquifer onset has not been included in this analysis due to the lack of comparable signals (i.e. comparing the central hy- draulic head change is not possible as the aquifer only starts at some radial distance) – it will be discussed in detail later. To assess the sensitivity of head changes to changes in in- dividual parameter values, we compare the range of relative hydraulic head change ( 1h 1href ) produced by varying the re- spective parameter. To account for the change in time and non-geometric source property explored is its pressuriza- tion strength, for which we investigated a range of 3 or- ders of magnitude in an attempt to cover a sufﬁcient range non-geometric source property explored is its pressuriza- tion strength, for which we investigated a range of 3 or- ders of magnitude in an attempt to cover a sufﬁcient range Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ K. 3.1 Reference simulation The described model was run for each aquifer type, using ref- erence values of parameters given in Table 2, with a magma chamber pressurization of 10 MPa. In both aquifer types, the pressurization of the magma chamber induces a fall in hy- draulic head, which is strongest directly above the magma chamber and decreases with radial distance from the chamber (Fig. 2). At distances larger than 5 km from the axis of sym- metry – hereafter termed “centre” – the initial head change is of opposite sign compared to the central areas, but the am- plitude of the head rise here is small in comparison with the central signal. There is also a comparatively small vertical gradient in the hydraulic head values. Whilst the pattern of the head change is the same in both aquifers, the absolute value of the signal differs substantially. In the pyroclastic aquifer, the maximum head fall is about 1.4 cm, while the hydraulic head in the lava ﬂow aquifer falls by a maximum of 6 m. The initial hydraulic head change proﬁle perfectly mirrors the strain curves (Fig. 2), illustrating that strain is the driver for the head changes. The aquifer is subject to dila- tion (positive strain), with a maximum value centrally above the chamber, which changes to compression (negative strain) with radial distance. Like for the hydraulic head, the two aquifers show similar patterns in strain, but different absolute values. Maximum volumetric strain in the pyroclastic aquifer is about 3 microstrain, while it is 13 microstrain in the lava ﬂow aquifer. tre. Despite its lower permeability, ﬂow speeds are higher by more than a factor of two in the lava ﬂow aquifer - as a result of the larger pressure gradient. Fluid ﬂow is impor- tant because it equilibrates pressure in the aquifer and is re- sponsible for the changes of strain and hydraulic head sig- nals with time. Figure 3 shows the change with time of hy- draulic head and volumetric strain, respectively, in a point in the aquifer centrally above the chamber. As water ﬂows away from the centre, hydraulic head continues to fall in the pyro- clastic aquifer until it reaches an equilibrium value of about −4 cm. Volumetric strain decreases and changes sign to com- pression after about 10 days; this compression increases and reaches an equilibrium value of about 10 microstrain. 3.1 Reference simulation In the lava ﬂow aquifer, hydraulic head increases, also tending to- wards an equilibrium value of about −4 cm with time; vol- umetric strain increases and also evolves to an equilibrium value. Whilst time-dependent changes take place almost until the end of simulation duration (1000 days) in the pyroclastic aquifer, the values in the lava ﬂow aquifer reach equilibrium after less than 10 days. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1213 0 5 10 15 20 −15 −10 −5 0 5 0 5 10 15 20−1 0 1 2 3 Radial distance (km) Hydraulic head change (10-3m) Volumetric strain (microstrain) (a) 0 5 10 15 20 −7 −6 −5 −4 −3 −2 −1 0 1 0 5 10 15 20−2 0 2 4 6 8 10 12 14 Volumetric strain (microstrain) Hydraulic head change (m) (b) Radial distance (km) Figure 2. Results of the reference simulation, shown as the initial (i.e. 10−6 s) hydraulic head change (blue, solid line) and volumetric strain (red, dashed line) along proﬁles through the two aquifer types due to a magma chamber pressurization of 10 MPa. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. Both aquifers show a fall in hydraulic head mirroring the dilatational strain curves. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain low et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1213 s to subsurface strain 0 5 10 15 20 −7 −6 −5 −4 −3 −2 −1 0 1 0 5 10 15 20−2 0 2 4 6 8 10 12 14 Volumetric strain (microstrain) Hydraulic head change (m) (b) Radial distance (km) 0 5 10 15 20 −15 −10 −5 0 5 0 5 10 15 20−1 0 1 2 3 Radial distance (km) Hydraulic head change (10-3m) Volumetric strain (microstrain) (a) Figure 2. Results of the reference simulation, shown as the initial (i.e. 10−6 s) hydraulic head change (blue, solid line) and volumetric strain (red, dashed line) along proﬁles through the two aquifer types due to a magma chamber pressurization of 10 MPa. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. Both aquifers show a fall in hydraulic head mirroring the dilatational strain curves. space, this has been done for different locations in the model domain and at different times during the simulation. Table 4. Parameter groups deﬁnition for the ranking resulting from sensitivity analysis. Parameter h∗= 1h 1href Group A h∗≥2.5 or h∗≤−0.5 B 1.9 < h∗< 2.5 or −0.5 < h∗< 0.1 C 0.1 ≤h∗≤1.9 Table 4. Parameter groups deﬁnition for the ranking resulting from sensitivity analysis. Table 4. Parameter groups deﬁnition for the ranking resulting from sensitivity analysis. Parameter h∗= 1h 1href Group A h∗≥2.5 or h∗≤−0.5 B 1.9 < h∗< 2.5 or −0.5 < h∗< 0.1 C 0.1 ≤h∗≤1.9 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1214 0 200 400 600 800 1000 −0.06 −0.04 −0.02 0 0 200 400 600 800 1000 −20 −10 0 10 Time (days) Hydraulic head change (m) Volumetric strain (microstrain) 1 2 3 4 5 6 7 8 0 Radial distance (km) Darcy’s velocity flow field Velocity magnitude (10-9 m/s) (a) 1 2 3 4 0 5 0 2 4 6 8 10 −7 −6 −5 −4 −3 −2 −1 0 0 2 4 6 8 10 13 13.1 13.2 13.3 Time (days) Hydraulic head change (m) Volumetric strain (microstrain) 1 2 3 4 5 6 7 8 0 Radial distance (km) Darcy’s velocity flow field Velocity magnitude (10-9 m/s) (b) 2 4 6 8 0 10 Figure 3. Upper graphs: porous ﬂow pattern shown for the reference simulation at t = 0.1 days. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. Arrows indicate ﬂow direction at the point where the arrow is attached, their length is proportional to ﬂow velocity (note: different scales for a and b), colours show velocity magnitude. Lower graphs show hydraulic head and strain development with time in the centre of the aquifers. Note the different time scale – ﬂow processes are faster in the lava ﬂow aquifer. 0 2 4 6 8 10 −7 −6 −5 −4 −3 −2 −1 0 0 2 4 6 8 10 13 13.1 13.2 13.3 Time (days) Hydraulic head change (m) Volumetric strain (microstrain) 1 2 3 4 5 6 7 8 0 Radial distance (km) Darcy’s velocity flow field Velocity magnitude (10-9 m/s) (b) 2 4 6 8 0 10 0 200 400 600 800 1000 −0.06 −0.04 −0.02 0 0 200 400 600 800 1000 −20 −10 0 10 Time (days) Hydraulic head change (m) Volumetric strain (microstrain) 1 2 3 4 5 6 7 8 0 Radial distance (km) Darcy’s velocity flow field Velocity magnitude (10-9 m/s) (a) 1 2 3 4 0 5 Figure 3. Upper graphs: porous ﬂow pattern shown for the reference simulation at t = 0.1 days. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. Arrows indicate ﬂow direction at the point where the arrow is attached, their length is proportional to ﬂow velocity (note: different scales for a and b), colours show velocity magnitude. Lower graphs show hydraulic head and strain development with time in the centre of the aquifers. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain Note the different time scale – ﬂow processes are faster in the lava ﬂow aquifer. Group A Group B Δh / Δhreference −10 0 10 20 30 κ: 10-14-10-7 m2 Eaq: 0.5-100 MPa α: 0.45-1 Tf: 10-500°C νaq: 0.15-0.4 ΔP:1-100 MPa bch: 0.25-2 km Vch: 0.125-3.375 km3 zch: 2-5 km daq:50-450 m zaq:100-2200 m initial 1 day 10 days Group C (a) κ: 10-14-10-9 m2 Eaq: 0.5-100 GPa α: 0.1-1 Tf: 10-500°C νaq: 0.1-0.35 ΔP:1-100 MPa bch: 0.25-2 km Vch: 0.125-3.375 km3 zch: 2-5 km daq:50-450 m zaq:100-2200 m (b) Figure 4. Exemplary plots used for the sensitivity analysis, show- ing the inﬂuence of changing a parameter (whilst keeping all others constant) on the central, initial hydraulic head change. (a) Pyroclas- tic aquifer, (b) lava ﬂow aquifer. Dashed lines/grey areas indicate the priority bounds and groups that were used to rank parameters according to their importance (Table 4). For speciﬁcation of sym- bols please refer to Table 1. Group A Group B Δh / Δhreference −10 0 10 20 30 κ: 10-14-10-7 m2 Eaq: 0.5-100 MPa α: 0.45-1 Tf: 10-500°C νaq: 0.15-0.4 ΔP:1-100 MPa bch: 0.25-2 km Vch: 0.125-3.375 km3 zch: 2-5 km daq:50-450 m zaq:100-2200 m Group C (a) different locations in the aquifer, respectively. As the inﬂu- ence of many parameters varies in time and space, the rank- ing of parameters according to their signiﬁcance is a two-step procedure. First, three parameter groups A, B, and C are de- ﬁned based on the inﬂuence of a parameter on hydraulic head change in one location at one of the three tested simulation times (Table 4 and Fig. 4). Group A has the strongest inﬂu- ence on hydraulic head change, C the least. 3.2 Sensitivity analysis Figure 3 illustrates the ﬂuid ﬂow pattern in the simula- tions, which are very different for the two different aquifer types. In the pyroclastic aquifer, ﬂuid ﬂow is away from the centre, while ﬂow in the lava ﬂow aquifer is towards the cen- Figure 4 shows the range of relative hydraulic head change ( 1h 1href ) produced by varying individual parameters, demon- strating the signiﬁcant differences in the sensitivity of hy- www.solid-earth.net/6/1207/2015/ Solid Earth, 6, 1207–1229, 2015 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain Parameters are then ranked into four priority groups based on the number of tested occasions (locations and simulation times) in which they belong to a certain parameter group: initial 1 day 10 days κ: 10-14-10-9 m2 Eaq: 0.5-100 GPa α: 0.1-1 Tf: 10-500°C νaq: 0.1-0.35 ΔP:1-100 MPa bch: 0.25-2 km Vch: 0.125-3.375 km3 zch: 2-5 km daq:50-450 m zaq:100-2200 m (b) – priority 1 – parameters that belong to group A in ≥85% of tests; – priority 2 – parameters that belong to group A in > 50% of tests; – priority 3 – parameters that belong to groups B or C in ≥50% of tests, but belong to group C in < 85% of tests; Figure 4. Exemplary plots used for the sensitivity analysis, show- – priority 4 – parameters that belong to group C in ≥85% of tests. Figure 4. Exemplary plots used for the sensitivity analysis, show- ing the inﬂuence of changing a parameter (whilst keeping all others constant) on the central, initial hydraulic head change. (a) Pyroclas- tic aquifer, (b) lava ﬂow aquifer. Dashed lines/grey areas indicate the priority bounds and groups that were used to rank parameters according to their importance (Table 4). For speciﬁcation of sym- bols please refer to Table 1. Following the deﬁnitions of the priority groups we can rank the investigated parameters as follows: Following the deﬁnitions of the priority groups we can rank the investigated parameters as follows: – priority 1 – pressurization value, volume and aspect ra- tio of the chamber; draulic head change to the different parameters. For both aquifers, plots as shown in Fig. 4 were produced for four – priority 2 – temperature of the pore ﬂuid, permeability and Biot–Willis coefﬁcient of the aquifer; draulic head change to the different parameters. For both aquifers, plots as shown in Fig. 4 were produced for four www.solid-earth.net/6/1207/2015/ www.solid-earth.net/6/1207/2015/ Solid Earth, 6, 1207–1229, 2015 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1215 −3 −2 −1 0 1 2 3 −0.3 −0.2 −0.1 0 0.1 Hydraulic head change (m) (a) κ=10-14 m2 κ=10-12 m2 κ=10-10 m2 Eaq=1MPa Eaq=10MPa Eaq=100MPa Time (log(d)) −6 −4 −2 0 2 −8 −6 −4 −2 0 Hydraulic head change (m) (b) Eaq=1GPa Eaq=10GPa Eaq=100GPa Time (log(d)) Figure 5. Inﬂuence of Young’s modulus and permeability of the aquifer on the central hydraulic head change and its evolution with time. (a) Pyroclastic aquifer, (b) lava ﬂow aquifer. −6 −4 −2 0 2 −8 −6 −4 −2 0 Hydraulic head change (m) (b) Eaq=1GPa Eaq=10GPa Eaq=100GPa Time (log(d)) −3 −2 −1 0 1 2 3 −0.3 −0.2 −0.1 0 0.1 Hydraulic head change (m) (a) κ=10-14 m2 κ=10-12 m2 κ=10-10 m2 Eaq=1MPa Eaq=10MPa Eaq=100MPa Time (log(d)) Figure 5. Inﬂuence of Young’s modulus and permeability of the aquifer on the central hydraulic head change and its evolution with time. (a) Pyroclastic aquifer, (b) lava ﬂow aquifer. – priority 3 – chamber depth, aquifer depth and Young’s modulus of the aquifer; – priority 3 – chamber depth, aquifer depth and Young’s modulus of the aquifer; response decreases again when increasing Eaq from 10 to 100 GPa (Fig. 5b). Note that, in the pyroclastic aquifer, a change in ﬂow behaviour can be seen when increasing the stiffness: as long as Eaq is smaller or equal to 10 MPa, hy- draulic head falls with time as it does in the reference sim- ulation. When the stiffness is increased to 100 MPa, the hy- draulic head behaviour with time is comparable to that in the lava ﬂow aquifer, where ﬂow is towards the centre of the domain; hence hydraulic head increases with time (Fig. 5a). The permeability determines porous ﬂow velocity and hence how fast the hydraulic head signal changes with time. A larger κ leads to a quicker change in head, visible for exam- ple in Fig. 5a for Eaq = 100MPa: while the hydraulic head signal after 1 day is still at the initial value for a permeabil- ity of 10−14 m2, it is decreased by half for a permeability of 10−10 m2. – priority 4 – Poisson’s ratio and thickness of the aquifer. www.solid-earth.net/6/1207/2015/ This ranking is, however, only a relative one – even those parameters of the last priority group have a non-negligible in- ﬂuence on the resulting hydraulic head change. Furthermore, the ranking of a parameter depends partly on the range of values tested for that parameter. This is particularly impor- tant in interpreting the sensitivity to the Biot–Willis coefﬁ- cient (α). Due to the scarcity of experimental data for this parameter, the sweeps in both aquifer cases were performed almost over the whole mathematical range of α between the porosity and 1. However, the true value of α for natural soft rocks should be close to 1, while it is close to the porosity for hard rocks. Therefore, although ranked here as priority 2, in reality the Biot–Willis coefﬁcient might belong in a lower priority group. More information on the individual inﬂuence of α can be found in Appendix A. The non-monotonous inﬂuence of the Young’s modulus stems from the fact that not only its absolute, but also its value relative to the surrounding lithology is important. We therefore also performed parametric sweeps over the Young’s Moduli of the host and cap rock, Eh and Ec, respectively, while Eaq is kept constant. The ratios of elastic properties of the three rock layers are deﬁned as 3.3 Results of parametric studies The parametric sweeps provided a number of interesting in- sights; we are focusing here on describing the most important ones. ERh = Eh Eaq (9) and ERc = Ec Eaq . (10) (9) 3.3.1 Inﬂuence of material properties and and Of the aquifer’s elastic properties, namely the Poisson’s ratio ν and the Young’s modulus Eaq, only the latter is signiﬁ- cant for the poroelastic response to applied strain. The most important hydraulic property is the permeability κ. Figure 5 shows the inﬂuence of these two important material proper- ties on the hydraulic head response and its change with time. The initial hydraulic head change in an aquifer is identical for different permeabilities, but can be changed by orders of magnitude by changing the aquifer stiffness. For values between 1 and 10 000 MPa, a higher Young’s modulus, i.e. a stiffer aquifer, leads to a larger hydraulic head response. However, this relationship is not monotonous as the head Of the aquifer’s elastic properties, namely the Poisson’s ratio ν and the Young’s modulus Eaq, only the latter is signiﬁ- cant for the poroelastic response to applied strain. The most important hydraulic property is the permeability κ. Figure 5 shows the inﬂuence of these two important material proper- ties on the hydraulic head response and its change with time. ERc = Ec Eaq . (10) (10) Figure 6a and b show the hydraulic head change depending on ERc for different ERh values. A larger ERh for a ﬁxed Eaq indicates a stiffer host rock and results in a smaller strain in the aquifer and hence smaller hydraulic head change. The rel- ative cap rock stiffness ERc has negligible inﬂuence when it is small (generally: ERc < 0.1; for small ERh: ERc < 0.01). However, it becomes increasingly important when the cap rock stiffness is close to or larger than that of the aquifer (ERc > 1): a stiff cap rock can decrease the hydraulic head The initial hydraulic head change in an aquifer is identical for different permeabilities, but can be changed by orders of magnitude by changing the aquifer stiffness. For values between 1 and 10 000 MPa, a higher Young’s modulus, i.e. a stiffer aquifer, leads to a larger hydraulic head response. However, this relationship is not monotonous as the head www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain Figure 7a shows the strain proﬁle for a simulation with a sufﬁciently large ERc to ﬂip the sign of the signal. This change in sign is due to the strain jump at the host rock–aquifer boundary, where the dilatational strain in the host rock is turned into compression in the aquifer in the sign-ﬂipped case. change and even change its sign to a head rise. This “sign- ﬂipped” signal increases with larger ERc, as can be seen for the pyroclastic aquifer for ERc values larger than 100 (Fig. 6a). For small ERh values, this effect of ERc already takes place at ERc values larger than 0.1, visible for the lava ﬂow aquifer (Fig. 6b). When sweeping over the Young’s modulus of the aquifer, these elastic ratios are changed as well and contribute to the resulting head change which leads to the non-monotonous inﬂuence of Eaq. q By sweeping ERc of the pyroclastic aquifer together with sweeping other parameters, we found that the ERc value at which the strain sign is ﬂipped (“ERc ﬂip”) is determined by the geometry of the system, in particular by the distance be- tween aquifer and magma chamber and the thickness of the cap rock (Fig. 6c and d). The shorter the chamber-aquifer distance and the thicker the cap rock layer, the smaller is ERc ﬂip. Figure 7b shows the hydraulic head change along a hori- zontal proﬁle in an aquifer with a sign-ﬂipped signal. In con- trast to the reference case, the central head change here is positive and changes sign twice: at about 3 km radial distance to a fall, and again at about 6 km to a head rise – mirroring sign-ﬂipped volumetric strain. It is common that aquifers are heated in volcanic settings. Figure 8 shows the substantial inﬂuence of changing the pore ﬂuid temperature on the initial hydraulic head change and its evolution with time, especially when temperatures are above the pressure-dependent boiling point and the aquifer pores are no longer ﬁlled with liquid water, but steam. With in- creasing temperature of liquid water, the initial hydraulic head change is reduced in the lava ﬂow aquifer, whereas it is very slightly increased in the pyroclastic aquifer (Fig. 8). K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1216 (a) (b) stronger cap rock ERh=0.002 ERh=0.01 ERh=0.02 ERh=0.1 ERh=0.2 ERh=1 ERh=2 ERh=10 ERh=50 ERh=100 ERh=500 ERh=1000 ERh=5000 ERh=10000 0 1 2 3 −10 −5 0 5 Δh (m) 1 2 3 −0.1 0 0.1 stronger caprock 0 1 2 3 0 0.2 0.4 0.6 0.8 1 dist=1km dist=1.6km dist=2km dist=2.6km (c) Δh (m) log(ERc) 0 1 2 3 0 0.2 0.4 0.6 0.8 dc=0.2km dc=1.2km dc=2.2km dc=3.2km (d) Δh (m) log(ERc) Δh (m) −4 −3 −2 −1 0 −25 −20 −15 −10 −5 0 5 log(ERc) log(ERc) stronger cap rock Figure 6. Inﬂuence of the elastic stratigraphy on the central, initial hydraulic head change, shown using the ratios of Young’s Moduli ERc = Ec Eaq , and ERh = Eh Eaq . (a) Pyroclastic aquifer, (b) lava ﬂow aquifer. Also shown is the behaviour of central initial head change in the pyroclastic aquifer with ERc for different distances between chamber and aquifer (c) and the behaviour of central initial head change in the pyroclastic aquifer with ERc for different cap rock thicknesses (d). (b) ERh=0.002 ERh=0.01 ERh=0.02 ERh=0.1 ERh=0.2 ERh=1 ERh=2 stronger caprock Δh (m) −4 −3 −2 −1 0 −25 −20 −15 −10 −5 0 5 log(ERc) stronger cap rock (b) ERh=0.002 ERh=0.01 ERh=0.02 ERh=0.1 ERh=0.2 ERh=1 ERh=2 0 stronger caprock 0 1 2 3 0 0.2 0.4 0.6 0.8 dc=0.2km dc=1.2km dc=2.2km dc=3.2km (d) Δh (m) log(ERc) Δh (m) −4 −3 −2 −1 0 −25 −20 −15 −10 −5 0 5 log(ERc) stronger cap rock Δh (m) log(ERc) Figure 6. Inﬂuence of the elastic stratigraphy on the central, initial hydraulic head change, shown using the ratios of Young’s Moduli ERc = Ec Eaq , and ERh = Eh Eaq . (a) Pyroclastic aquifer, (b) lava ﬂow aquifer. Also shown is the behaviour of central initial head change in the pyroclastic aquifer with ERc for different distances between chamber and aquifer (c) and the behaviour of central initial head change in the pyroclastic aquifer with ERc for different cap rock thicknesses (d). settings, ERh is larger for the pyroclastic aquifer, leading to a smaller strain compared to the lava ﬂow aquifer. The rela- tive cap rock stiffness ERc then determines the strain change at the aquifer–cap rock boundary, but it also inﬂuences the strain change at the aquifer–host rock boundary. www.solid-earth.net/6/1207/2015/ Solid Earth, 6, 1207–1229, 2015 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain (a) shows vertical strain proﬁles through the centre of the domain from 2 km depth to the surface, for different elastic stratigra- phies: the two reference simulations and a setting in which the sign of strain (and consequently hydraulic head) is ﬂipped from dilation to compression due to a sufﬁciently stiff cap rock (ERc = 1000). (b) shows hydraulic head change and volumetric strain along a horizontal proﬁle through the aquifer in the reference pyroclastic aquifer in comparison to the aquifer, where sign of strain and head is ﬂipped. −4 −3 −2 −1 0 1 2 3 −0.4 −0.3 −0.2 −0.1 0 T=10°C T=100°C T=200°C T=300°C T=400°C T=500°C Time (log(d)) Hydraulic head change (m) steam water T=300°C T=400°C T=500°C (a) −4 −3 −2 −1 0 1 2 3 −6 −5 −4 −3 −2 −1 0 T=10°C T=100°C T=200°C T=300°C T=400°C T=500°C Time (log(d)) Hydraulic head change (m) steam water T=300°C T=400°C T=500°C (b) Figure 8. Central hydraulic head change and its evolution with time for different pore ﬂuid temperatures. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. −4 −3 −2 −1 0 1 2 3 −0.4 −0.3 −0.2 −0.1 0 T=10°C T=100°C T=200°C T=300°C T=400°C T=500°C Time (log(d)) Hydraulic head change (m) steam water T=300°C T=400°C T=500°C (a) −4 −3 −2 −1 0 1 2 3 −6 −5 −4 −3 −2 −1 0 T=10°C T=100°C T=200°C T=300°C T=400°C T=500°C Time (log(d)) Hydraulic head change (m) steam water T=300°C T=400°C T=500°C (b) Figure 8. Central hydraulic head change and its evolution with time for different pore ﬂuid temperatures. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. aquifer the opposite relation is true. Interestingly, the order of magnitude of hydraulic head change is the same in the two different aquifer types when steam saturated, while there is a 2 order of magnitude difference in the signals for the water saturated aquifers. In both aquifers, hydraulic head change increases with increasing temperature of the steam. Flow ve- locities are up to 1 order of magnitude faster in steam satu- rated pyroclastic aquifers when compared to water aquifers – lava ﬂow aquifers show 1 order of magnitude higher ﬂow velocities when saturated with water. Additionally, a change in ﬂow behaviour is visible when conditions in the pyroclas- tic aquifer are changed: in a steam aquifer, hydraulic head increases with time, while it falls in the water-aquifer. the aquifers. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain For steam-ﬁlled pores (above 300 ◦C), the initial central hy- draulic head change in the pyroclastic aquifer is 1 order of magnitude larger than for liquid water – in the lava ﬂow The elastic stratigraphy determines the strain distribution in the domain, visible in Fig. 7a that shows vertical strain proﬁles from 2 km depth to the surface for different settings. Strain in the host rock is larger in the pyroclastic reference case – despite the fact that the same host rock is used in the reference simulations, the Young’s modulus of the aquifer hence also inﬂuences the strain in the underlying rock. The graph also illustrates how strain changes at the boundary be- tween different elastic mediums: strain increases when hit- ting a stiffer medium and vice versa. ERh determines this change at the host rock–aquifer boundary. In the reference Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1217 0 5 10 15 20 −0.01 0 0.01 0.02 0 5 10 15 20−5 −4 −3 −2 −1 0 1 2 3 Reference Sign-flipped Radial distance (km) Hydraulic head change (m) Volumetric strain (microstrain) −2 −1.5 −1 −0.5 0 −10 −5 0 5 10 15 20 Reference − Pyroclastic Reference − Lava flow Sign-flipped strain Volumetric strain (microstrain) aquifer (b) Depth (km) (a) Figure 7. (a) shows vertical strain proﬁles through the centre of the domain from 2 km depth to the surface, for different elastic stratigra- phies: the two reference simulations and a setting in which the sign of strain (and consequently hydraulic head) is ﬂipped from dilation to compression due to a sufﬁciently stiff cap rock (ERc = 1000). (b) shows hydraulic head change and volumetric strain along a horizontal proﬁle through the aquifer in the reference pyroclastic aquifer in comparison to the aquifer, where sign of strain and head is ﬂipped. 0 5 10 15 20 −0.01 0 0.01 0.02 0 5 10 15 20−5 −4 −3 −2 −1 0 1 2 3 Reference Sign-flipped Radial distance (km) Hydraulic head change (m) Volumetric strain (microstrain) (b) −2 −1.5 −1 −0.5 0 −10 −5 0 5 10 15 20 Reference − Pyroclastic Reference − Lava flow Sign-flipped strain Volumetric strain (microstrain) aquifer Depth (km) (a) Figure 7. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain For the lava ﬂow aquifer, the absolute magma chamber depth has no inﬂuence on hydraulic head change as long as the distance between aquifer and magma chamber is constant. The smaller this distance, the larger is the corre- sponding hydraulic head fall. aquifer the opposite relation is true. Interestingly, the order of magnitude of hydraulic head change is the same in the two different aquifer types when steam saturated, while there is a 2 order of magnitude difference in the signals for the water saturated aquifers. In both aquifers, hydraulic head change increases with increasing temperature of the steam. Flow ve- locities are up to 1 order of magnitude faster in steam satu- rated pyroclastic aquifers when compared to water aquifers This relation is somewhat more complicated for the py- roclastic aquifers, where hydraulic head depends on the dis- tance between chamber and aquifer but also on the chamber depth. The central hydraulic head in the pyroclastic aquifers (Fig. 9a) is positive (hence sign-ﬂipped) for a sufﬁciently small distance between aquifer and chamber, then switches sign to a head fall at a value distﬂip, which increases with fur- ther increasing distance. The value distﬂip is larger, the deeper the chamber. For a constant distance, hydraulic head change is larger for deeper chambers, if hydraulic head change is positive. Extrapolating the curves shows that if the hydraulic head change is negative, it is larger for shallower chambers. – lava ﬂow aquifers show 1 order of magnitude higher ﬂow velocities when saturated with water. Additionally, a change in ﬂow behaviour is visible when conditions in the pyroclas- tic aquifer are changed: in a steam aquifer, hydraulic head increases with time, while it falls in the water-aquifer. 3.3.2 Inﬂuence of the geometry 0 5 10 15 20 −80 −60 −40 −20 0 20 b=250m b=500m b=750m b=1000m b=1250m b=1500 b=2000 Radial distance (km) Hydraulic head change (m) (b) 0 5 10 15 20 −0.15 −0.1 −0.05 0 0.05 b=250m b=500m b=750m b=1000m b=1250m b=1500 b=2000 Radial distance (km) Hydraulic head change (m) (a) Figure 10. Inﬂuence of changing the aspect ratio of a spheroidal chamber (with constant zCH and V ) on the initial hydraulic head change proﬁle through the aquifer. Oblate shapes have b < 1000 m, prolate chambers correspond to b > 1000 m. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. 1km and zaq = 200m, see Appendix B), the maximum head change is no longer central but laterally offset by up to 1 km. the maximum hydraulic head fall after this time (at the re- spective locations) in aquifers starting at 2 km radial distance is about 50 % larger in the pyroclastic aquifer and about 50 % smaller in the lava ﬂow aquifer. This difference is strongest close to the lateral aquifer boundary facing the domain cen- tre, where head falls are largest – with radial distance, the head change proﬁle of the shorter aquifer approximates the reference proﬁle. In the pyroclastic aquifer the difference be- tween reference and shorter aquifer is negligible after the ﬁrst kilometre, while in the lava ﬂow aquifers head values differ considerably from each other over longer distances. We also evaluated the inﬂuence of the shape of the magma chamber by incorporating tests for a prolate and oblate spheroid. Although chamber volumes are constant, the shape can change the hydraulic head signal by 1 order of magni- tude. Figure 10 shows that the amplitude is highest for oblate chambers, intermediate for a sphere and smallest for prolate chambers. Instead of having an “inﬁnite” aquifer covering the whole volcano, we also varied the lateral distance L between the centre of the model and the onset of the aquifer, realized by a zero permeability zone in the centre of the domain (compare Fig. 1). The initial hydraulic head in these shorter aquifers equals the respective value at the same location in the reference aquifer. After some time, however, the head signal in the shorter aquifer differs from the reference case. 3.3.2 Inﬂuence of the geometry To demonstrate combined geometric effects we plot the cen- tral initial hydraulic head change vs. the distance between magma chamber and aquifer for different chamber radii and absolute chamber depths in Fig. 9. The larger the chamber radius the larger is the resulting hydraulic head change in Normally, the maximum hydraulic head fall is directly above the chamber. However, when considering a hydraulic head proﬁle through the pyroclastic aquifer for a shallow magma chamber without a sign-ﬂipped strain (e.g. zCH = www.solid-earth.net/6/1207/2015/ Solid Earth, 6, 1207–1229, 2015 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strai 1218 2 2.5 3 3.5 4 4.5 −0.2 0 0.2 0.4 0.6 Radius=0.5km Radius=1km Radius=1.5km dist (km) Hydraulic head change (m) zCH=3km (a) zCH=4km zCH=5km 2 2.5 3 3.5 4 4.5 −40 −30 −20 −10 0 Radius=0.5km Radius=1km Radius=1.5km dist (km) Hydraulic head change (m) (b) all zCH Figure 9. Inﬂuence of the geometry, i.e. chamber radius, chamber depth zCH, and distance (dist) between aquifer and chamber, on the initial central hydraulic head change. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. 2 2.5 3 3.5 4 4.5 −0.2 0 0.2 0.4 0.6 Radius=0.5km Radius=1km Radius=1.5km dist (km) Hydraulic head change (m) zCH=3km (a) zCH=4km zCH=5km 2 2.5 3 3.5 4 4.5 −40 −30 −20 −10 0 Radius=0.5km Radius=1km Radius=1.5km dist (km) Hydraulic head change (m) (b) all zCH Figure 9. Inﬂuence of the geometry, i.e. chamber radius, chamber depth zCH, and distance (dist) between aquifer and chamber, on the initial central hydraulic head change. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. 0 5 10 15 20 −0.15 −0.1 −0.05 0 0.05 b=250m b=500m b=750m b=1000m b=1250m b=1500 b=2000 Radial distance (km) Hydraulic head change (m) (a) 0 5 10 15 20 −80 −60 −40 −20 0 20 b=250m b=500m b=750m b=1000m b=1250m b=1500 b=2000 Radial distance (km) Hydraulic head change (m) (b) Figure 10. Inﬂuence of changing the aspect ratio of a spheroidal chamber (with constant zCH and V ) on the initial hydraulic head change proﬁle through the aquifer. Oblate shapes have b < 1000 m, prolate chambers correspond to b > 1000 m. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. 3.3.2 Inﬂuence of the geometry Figure 11 shows the head changes after 10 days of simula- tion in the pyroclastic aquifer and after 1 day in the lava ﬂow aquifer, respectively (the different timescales were used to account for the faster processes in the latter case). Compared to the proﬁle of hydraulic head in the reference simulation, For larger values of L, the pyroclastic aquifers also differ in a comparable manner from the reference case, but become indistinguishable on the centimetre scale at distances larger than 6 km. Lava ﬂow aquifers starting at 4 km or further ra- dial distance are all signiﬁcantly different from the reference case as they show positive head changes (up to 10 cm), while the reference aquifer at this time shows negative values ev- erywhere less than about 10 km radial distance from the cen- tre. Figure 12 shows the different ﬂow patterns in the ﬁrst 8 km of three different aquifers after 0.1 days of simulation. Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1219 0 5 10 15 20 −0.025 −0.02 −0.015 −0.01 −0.005 0 0.005 Reference L=2km L=4km L=6km L=8km Radial distance (km) Hydraulic head change (m) t=10d (a) 0 5 10 15 20 −1.5 −1 −0.5 0 0.5 Reference L=2km L=4km L=6km L=8km Radial distance (km) Hydraulic head change (m) t=1d (b) Figure 11. Inﬂuence of lateral distance L between chamber and aquifer on the hydraulic head change proﬁle through the aquifer after t = 10d (pyroclastic aquifer) and t = 1d (lava ﬂow aquifer), respectively. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain : Poroelastic responses of conﬁned aquifers to subsurface strain 1219 0 5 10 15 20 −1.5 −1 −0.5 0 0.5 Reference L=2km L=4km L=6km L=8km Radial distance (km) Hydraulic head change (m) t=1d (b) 0 5 10 15 20 −0.025 −0.02 −0.015 −0.01 −0.005 0 0.005 Reference L=2km L=4km L=6km L=8km Radial distance (km) Hydraulic head change (m) t=10d (a) Figure 11. Inﬂuence of lateral distance L between chamber and aquifer on the hydraulic head change proﬁle through the aquifer after t = 10d (pyroclastic aquifer) and t = 1d (lava ﬂow aquifer), respectively. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. 3.3.3 Inﬂuence of source pressurization For L = 2km, the aquifer shows a ﬂow pattern similar to the reference simulation, although ﬂow in the pyroclastic aquifer shows an upward component that is absent in the ref- erence simulation. This component vanishes at later simula- tion times: after 1 day, ﬂow is horizontal. For L = 6km and L = 8km, ﬂow in the pyroclastic aquifer has a downward component, which is slightly stronger for the larger L and also vanishes at later simulation times. Two ﬂow directions can be observed in the lava ﬂow aquifer for L = 6km – one towards and one away from the volcano. At later simulation times, the ﬂow towards the volcano diminishes and then all ﬂow is away from the centre of the domain. For L = 8km, ﬂow in the lava ﬂow is completely reversed compared to the reference case: instead of ﬂowing towards the centre, water ﬂows away from it. Flow velocities are generally slower in the shorter aquifers (i.e. for larger values of L). For L = 2km, the aquifer shows a ﬂow pattern similar to the reference simulation, although ﬂow in the pyroclastic aquifer shows an upward component that is absent in the ref- erence simulation. This component vanishes at later simula- tion times: after 1 day, ﬂow is horizontal. For L = 6km and L = 8km, ﬂow in the pyroclastic aquifer has a downward component, which is slightly stronger for the larger L and also vanishes at later simulation times. Two ﬂow directions can be observed in the lava ﬂow aquifer for L = 6km – one towards and one away from the volcano. At later simulation times, the ﬂow towards the volcano diminishes and then all ﬂow is away from the centre of the domain. For L = 8km, ﬂow in the lava ﬂow is completely reversed compared to the reference case: instead of ﬂowing towards the centre, water ﬂows away from it. Flow velocities are generally slower in the shorter aquifers (i.e. for larger values of L). The initial hydraulic head response linearly depends on pres- surization strength of the source. We assumed instantaneous pressurization for simplicity; however, real magma chambers will more likely inﬂate over longer time periods. In Fig. 3.3.2 Inﬂuence of the geometry Darcy’s velocity flow field Velocity magnitude (10-9 m/s) Radial distance (km) 6 8 10 12 14 Radial distance (km) 8 10 12 14 16 Radial distance (km) 2 4 6 8 10 Darcy’s velocity flow field Velocity magnitude (10-9 m/s) Radial distance (km) 6 8 10 12 14 Radial distance (km) 8 10 12 14 16 Radial distance (km) 2 4 6 8 10 (a) (b) L=2km L=6km L=8km 0.1 0.2 0.3 0.4 0 0.5 0.2 0.4 0.6 0.8 0 1 1 2 3 4 0 5 0.1 0.2 0.3 0.4 0 0.5 0.1 0.2 0.3 0.4 0 0.5 10 20 30 40 0 50 Figure 12. Inﬂuence of lateral distance L between chamber and aquifer on the ﬂow pattern in the aquifers, here shown for t = 0.1 d for the ﬁrst 8 km of each aquifer. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. Arrows indicate ﬂow direction at the point where the arrow is attached, their length is proportional to ﬂow velocity (different scales for a and b), colours show velocity magnitudes. Note that arrows are upscaled for better visibility; therefore, they cross the upper aquifer boundary. However, this is only a plotting effect, there is no ﬂow leaving the aquifer. Darcy’s velocity flow field Veloc Radial distance (km) 6 8 10 12 14 Radial distance (km) 8 10 12 14 16 Radial distance (km) 2 4 6 8 10 (a) (a) (b) Figure 12. Inﬂuence of lateral distance L between chamber and aquifer on the ﬂow pattern in the aquifers, here shown for t = 0.1 d for the ﬁrst 8 km of each aquifer. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. Arrows indicate ﬂow direction at the point where the arrow is attached, their length is proportional to ﬂow velocity (different scales for a and b), colours show velocity magnitudes. Note that arrows are upscaled for better visibility; therefore, they cross the upper aquifer boundary. However, this is only a plotting effect, there is no ﬂow leaving the aquifer. www.solid-earth.net/6/1207/2015/ 3.3.3 Inﬂuence of source pressurization 13 we compare central hydraulic head evolution due to instan- taneous pressurization with hydraulic head evolution due to a pressurization that is stepped up over 100 days (reaching the same maximum pressurization value). In this simulation, hydraulic head in the pyroclastic aquifer decreases more or less in parallel to the increase of chamber pressurization, un- til it reaches its maximum fall after 100 days. The maximum fall and its further development with time is not distinguish- able from the hydraulic head evolution in the reference sim- ulation at the respective time. Hydraulic head change in the lava ﬂow aquifer at ﬁrst also falls more or less in parallel www.solid-earth.net/6/1207/2015/ Solid Earth, 6, 1207–1229, 2015 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1220 0 50 100 150 −5 0 0 50 100 150−0.1 0 Time (days) (b) Δh (m), inst. pressurisation Δh (m), pressurisation over 100d 0 50 100 150 −5 0 0 50 100 150−0.1 0 Time (days) (b) Δh (m), inst. pressurisation Δh (m), pressurisation over 100d 0 200 400 600 800 1000 −0.05 −0.04 −0.03 −0.02 −0.01 0 0 200 400 600 800 1000 −0.05 −0.04 −0.03 −0.02 −0.01 0 Time (days) Δh (m), inst. pressurisation (a) Δh (m), pressurisation over 100d Time (days) 20 60 100 0 5 10 ΔP (MPa) Figure 13. Comparison of the development of central hydraulic change with time due to an instantaneous chamber pressurization and a pressurization over 100 d (with the same maximum pressurization value). (a) Pyroclastic aquifer, (b) lava ﬂow aquifer. Inlet in (a) shows how the pressurization is stepped up over time. 0 200 400 600 800 1000 −0.05 −0.04 −0.03 −0.02 −0.01 0 0 200 400 600 800 1000 −0.05 −0.04 −0.03 −0.02 −0.01 0 Time (days) Δh (m), inst. pressurisation (a) Δh (m), pressurisation over 100d Time (days) 20 60 100 0 5 10 ΔP (MPa) Figure 13. Comparison of the development of central hydraulic change with time due to an instantaneous chamber pressurization and a pressurization over 100 d (with the same maximum pressurization value). (a) Pyroclastic aquifer, (b) lava ﬂow aquifer. Inlet in (a) shows how the pressurization is stepped up over time. high nonlinearity of the problem. 3.3.3 Inﬂuence of source pressurization The model also ignores any hydrological sources and sinks, such as meteoric recharge, which can signiﬁcantly inﬂuence well level observations in reality. with the increase of chamber pressurization, but reaches its maximum earlier at about 50 days, when pressurization slows down (due to the deﬁnition of the step function; compare in- let in Fig. 13a). This maximum fall of about 8 cm is signiﬁ- cantly smaller than the maximum (initial) fall in the reference simulation of about 6 m. Note though that the hydraulic head in the reference simulation is already equilibrated to −4 cm at 50 days. After reaching its maximum, the hydraulic head fall decreases and reaches approximately the same equilib- rium value as the reference simulation when stepping up of pressurization is complete. The discussed models are most applicable to conﬁned aquifers that do not undergo extensive heating during the ob- servation period (e.g. aquifers at some distance of the vol- canic centre). They present a good opportunity to better un- derstand poroelastic aquifer responses that have been used for monitoring. Their advantage over previous models is the full two-way coupling of ﬂow and linear elastic behaviour and that we are able to simulate various geometries. The comparatively short computation time (on the order of 10 to 15 min per simulation depending on geometric complex- ity and number of necessary time steps) allows the study of a large number of parameters and their inﬂuence on hydraulic head changes and ﬂow pattern. 4.1 Model limitations In order to investigate poroelastic aquifer responses to crustal deformation, we made some simplifying assump- tions. For one, the presented models only consider single- phase, single-component ﬂow under constant temperature conditions. However, our parametric studies have shown that the pore ﬂuid properties signiﬁcantly inﬂuence the resulting head changes. Hydrothermal systems can contain steam, wa- ter and a number of solutes, and temperatures can change substantially. This can also affect the solid matrix, as its mechanical behaviour may deviate from elastic when it is sufﬁciently heated. Additionally, the injection of hydrother- mal ﬂuids into the aquifer can lead to a pore pressure in- crease, heating and further deformation (see e.g. Fournier and Chardot (2012) for a one-way-coupled model). We fo- cused on the pure poroelastic response, but the poroelastic, heating and phase change processes superimpose. www.solid-earth.net/6/1207/2015/ 4.2 General aspects Our simulations show that neither injections of ﬂuids nor ﬂow within a fracture – hereafter termed “fracture ﬂow” – is needed to induce hydraulic head changes of several me- tres in an aquifer. Volumetric strain induced by a quasi- instantaneous magma chamber pressurization causes imme- diate hydraulic head changes in local aquifers. Dilation above the chamber, due to ground uplift, leads to a fall in pore pressure, while the accompanying compression at more than 5 km distance from the centre of the uplift induces a head rise. Poroelastic processes are therefore a reasonable ap- proach to interpret rapid and large water level changes ob- served at volcanoes and they should not be ignored when studying hydrological systems in volcanic areas. For the same source and model geometry we could observe large differences between the two typical aquifer types. These dif- ferences are mainly due to the different elastic properties of the aquifers: the pyroclastic aquifer is much softer than the lava ﬂow aquifer and therefore strain attenuation is stronger, hence the resulting hydraulic head change is smaller. Secondly, the aquifer was fully saturated and conﬁned. To keep this study feasible, we did not investigate uncon- ﬁned aquifers as this would imply a non-saturated perme- able zone, and the coupling of linear elastic behaviour with non-saturated porous ﬂow is associated with a high computa- tional effort and often the solvers fail to converge due to the Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain Strain partitioning in the crust is reg- ulated by the elastic properties of the different layers; both the absolute and relative elastic properties of the aquifer and its surrounding lithology have a complex inﬂuence on the strain and head signals. A special case occurs when the cap rock is sufﬁciently stronger than the aquifer. A stiff cap rock prevents the dilation of the aquifer and turns the strain into compression, hence causing sign-ﬂipped signals. In the ref- erence set-up, the cap rock needs to be 2 orders of magnitude stiffer than the aquifer; this could be fulﬁlled if an unconsol- idated, permeable pyroclastic layer is overlain by a lava ﬂow. But other, perhaps more common, geological settings exist in which a sign-ﬂipped response can be expected, as the ge- ometry plays an important role as well: the thicker the cap rock, the smaller is the necessary ratio of cap rock to aquifer stiffness to change the sign of strain. For example, a cap rock that is only 3 times stiffer than the aquifer can already lead to a sign-ﬂip if the aquifer is about 1 km deep. For our reference pyroclastic aquifer, this means a cap layer with a stiffness of more than 30 MPa, such as another pyroclastic layer that is slightly stiffer (e.g. due to a different lithology, rate of con- solidation or grain size distribution) – a situation particularly feasible at stratovolcanoes. Fluid ﬂow leads to the changes of strain and hydraulic head signals with time. Hydraulic head continues to fall in the pyroclastic aquifer as water ﬂows away from the cen- tre, while the opposite ﬂow direction in the lava ﬂow aquifer leads to a decrease of the initial hydraulic head fall with time (i.e. head increases). In both aquifers the equilibrium hy- draulic head is approximately balancing the change in eleva- tion (about 4 cm in the centre), as this is the equilibrated wa- ter level signal that would be recorded in the well that moved upwards with the ground. When the Young’s modulus of the pyroclastic aquifer is sufﬁciently high (Fig. 5), the poroe- lastic response of this aquifer also becomes large enough to overcome the topographic gradient; hence the ﬂow pat- tern changes to ﬂow towards the volcano, causing hydraulic head in the centre to increase with time. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain strain relations. Indeed, the initial strain results of our poroe- lastic simulations agree with solutions calculated with just the structural mechanics module applying undrained param- eters, which can be derived from the drained parameters with the Biot–Willis coefﬁcient. However, the equilibrium strain signal in the aquifers does not correspond to the elastic so- lution calculated with drained parameters, except when one ignores gravitational ﬂow, as ﬂow down the slopes of the up- lifted volcano adds strain changes that are not taken into ac- count by purely elastic formulations. Both the strain-induced pressure gradient in the aquifers and the topographic gradient due to the ground uplift in- duce porous ﬂow; groundwater ﬂows from larger to smaller pressure/hydraulic head and from higher to lower elevation (Eq. 5). The chamber pressurization in our reference simula- tion leads to a central ground uplift of about 4 cm, leading to a topographic gradient that is opposing the pressure gradient induced by the strain: the topographic gradient points away from the centre, while the pressure gradient points towards the centre. As the hydraulic head change in the pyroclastic aquifer is smaller than this uplift, gravitational ﬂow domi- nates over strain-induced ﬂow, and hence ﬂuid ﬂow is away from the centre. In the lava ﬂow aquifer, strain-induced hy- draulic head changes are two orders of magnitude larger than the topographic change; therefore ﬂow in this aquifer is to- wards the centre, following the dominant pressure gradient. Despite its lower permeability, the lava ﬂow aquifer shows higher ﬂow velocities, as the strain-induced pressure gradient in this stiffer aquifer is large enough to overcome not only the topographic change but also the difference in permeability. The above ﬁndings highlight the necessity of a full cou- pling of ﬂuid and solid mechanics. Both the effect of ground deformation on the pore ﬂuid, as well as the inﬂuence of a pore ﬂuid on strain in the solid matrix need to be considered to fully understand well level and/or strain signals. y g Parametric studies have shown that poroelastic aquifer re- sponses are complex processes that are strongly inﬂuenced by source, geometrical and aquifer parameters as well as the elastic stratigraphy. Chamber radius and pressurization de- termine the strength of the deformation source and the sub- surface strain it causes. www.solid-earth.net/6/1207/2015/ www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1221 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1222 signal, i.e. if the aquifer is rather soft, a sufﬁciently close aquifer-source distance can lead to a sign-ﬂipped strain (be- cause ERc ﬂip is changed) and hence sign-ﬂipped hydraulic head change in the aquifer. that starts at 2 km radial distance from the centre, hydraulic head is up to 50 % smaller than the reference value (where L = 0km). This effect is due to locally faster ﬂow in the shorter aquifer. The positive hydraulic head changes in the lava ﬂow aquifers starting at distances larger than 4 km are due to very different ﬂow processes caused by changed ini- tial pressure gradients. Generally, the initial hydraulic head is negative in the ﬁrst 5 km of lateral distance and is positive at locations further from the centre; this pressure proﬁle causes ﬂow towards the centre in the lava ﬂow aquifer. However, if the aquifer onset falls in the “positive head area”, the driving pressure gradient and hence the ﬂow direction are reversed. For lava ﬂow aquifers that onset near the transition zone, two ﬂow directions can be observed – one towards and one away from the centre of the volcano. These two directions are due to the maximum initial head change being not directly at, but lateral offset from the lateral aquifer boundary. This can be seen in the proﬁle of the initial hydraulic head change in the reference simulations: beyond 5 km from the centre, hy- draulic head change ﬁrst increases with distance before de- creasing again. While this comparatively small gradient is negligible in the reference simulation, as ﬂow is dominated by the much stronger gradient towards the centre, it matters for ﬂow in aquifers that start close to the transition zone from positive to negative strain – at least early in the simulation. The elastic properties of the solid matrix as well as the pore ﬂuid together with the Biot–Willis coefﬁcient of the aquifer determine the initial pressure response of the aquifer to the strain. Permeability then determines the velocity of pressure equilibration and gravitational ﬂow and therefore the development of head and strain signals with time. Of particular interest is the inﬂuence of pore ﬂuid tempera- ture. It can change the hydraulic head response by 1 or- der of magnitude as well as inﬂuence the ﬂow behaviour in the aquifer. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain This is especially important in volcanic envi- ronments, where heat ﬂow is high and therefore temperature changes are likely. Changing the temperature means chang- ing compressibility, density and viscosity of the water. We at- tempted to distinguish their individual inﬂuence with simula- tions in which only one of the three parameters was changed to a value corresponding to steam, while the others were kept at values corresponding to water. The viscosity does not in- ﬂuence the initial head fall, but affects the speed of equili- bration, which is slower for a liquid water viscosity than for a lower steam viscosity (compare Eq. 5). Changing the phase of the pore ﬂuid does not have this straightforward effect however, as ﬂow velocities are also determined by the ini- tial pressure gradients. These are inﬂuenced by ﬂuid density and compressibility: decreasing the density of the pore ﬂuid increases the initial hydraulic head change, while increasing the compressibility decreases it. At higher temperatures high enough for a phase change from liquid to steam, ﬂuid density is reduced, while its compressibility rises. We therefore see a complex combination of these two effects. In the pyroclastic aquifer, the density effect dominates (hence hydraulic head change is larger), while in the lava ﬂow aquifer the com- pressibility effect is more important (hence hydraulic head change is smaller). The stronger initial hydraulic head fall in the steam saturated pyroclastic aquifer is then large enough to overcome the topographic gradient, such that ﬂow is towards the volcano. Therefore, as opposed to the reference simula- tion, the initial fall in hydraulic head diminishes with time in the soft aquifer just as it does in the stiffer aquifers. Hy- draulic head falls in the lava ﬂow and pyroclastic aquifers are of the same order of magnitude when saturated with steam, suggesting that the elastic properties of the solid matrix are less important and the processes are now governed by the ﬂuid properties. While we investigated the temperature ef- fect, other processes could also change pore ﬂuid properties, such as dissolved minerals, and thereby play a role in deter- mining the hydraulic head change. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain These considerations show how the entire process is mostly governed by the elastic properties of the aquifer, not its hydraulic properties. Strain changes simultaneously with hydraulic head due to the poroelastic nature of the aquifers. As water ﬂows away, the pyroclastic aquifer responds to the removal of pore ﬂuid with compaction – explaining the change of strain from di- lation to compaction. The volumetric strain increase in the lava ﬂow aquifer stems from the initial stress absorption by the pore ﬂuid (ﬁnal term in Eq. 2), which manifests as the pore pressure change. With equilibration of the pressure in the aquifer this stress absorption effect vanishes and strain approximates an equilibrium value that represents the strain value in an elastically equivalent, but dry material. Here, stress absorption of the ﬂuid leads to an initial strain reduc- tion by about 15 %; this value can be increased by increasing the Biot–Willis coefﬁcient. The subsurface stress and strain ﬁelds are also substan- tially dependent on the shape of the chamber. For oblate chambers, the aquifer area that is exposed to vertical stress is larger than for prolate chambers and it is therefore sub- ject to stronger strain. Additionally, the centre of the oblate chambers is shallower than the centre of the prolate chambers (as the depth of the chamber top is ﬁxed in the simulations). The distance between aquifer and magma chamber is another factor contributing to the strength of the strain ﬁeld affecting the aquifer. Generally, the closer the aquifer to the source, the stronger the strain and hence its pressure response. However, if elastic properties are close to values causing a sign-ﬂipped As pointed out by, for example, Rice and Cleary (1976), the initial elastic response (i.e. instantaneous deformation) of a porous medium can be calculated with the pure elastic solution by using undrained elastic parameters in the stress– www.solid-earth.net/6/1207/2015/ Solid Earth, 6, 1207–1229, 2015 Solid Earth, 6, 1207–1229, 2015 4.3.1 General considerations We have shown that wells can reﬂect the deformation at volcanoes, suggesting that their implementation in volcano monitoring systems could provide insights into subsurface processes causing the strain. However, prior to the inter- pretation of well signals, one needs to carefully consider that water levels can also be changed by several other pro- cesses, e.g. meteorological inﬂuences (rainfall), hydrother- mal ﬂuid injection, heat transferred conductively through the crust or changes in ﬂow conditions due to the opening or closure of fractures. These processes can also act simulta- neously and overcome hydraulic head changes caused by a poroelastic response. Under certain circumstances the differ- ent processes can be distinguished. First, the general hydro- logical behaviour – i.e. the meteorological responses – and up-to-date meteorological information should be tracked and therefore be reasonably well known if wells are to be in- cluded in a monitoring system. Then, a water level response to strain will be a transient signal on top of the background behaviour. Well level monitoring can form an important com- ponent for volcano monitoring in conjunction with geophys- ical or geochemical observations to track magma reservoir evolution. For example, ground deformation data will be use- ful for identifying hydrothermal injections. When hydrother- mal ﬂuids from a magma reservoir are injected into sur- rounding rocks, hydraulic head in the hydrothermal systems will rise and the ground will be uplifted (e.g. Todesco et al., 2004; Hurwitz et al., 2007; Rinaldi et al., 2010; Fournier and Chardot, 2012), while strain due to chamber inﬂation gen- erally leads to a water level fall together with ground up- lift. Temperature sensors in monitoring wells provide addi- tional valuable information, as hydrothermal ﬂuid injections should lead to relatively fast temperature effects compared to heat transferred from an inﬂating chamber or a dyke intrud- ing through the crust, which would reach the aquifer much later – in fact, after the poroelastic response is already equili- brated. More complicated is the recognition of fracture ﬂow. A hydraulic head change caused by rapid volumetric strain- ing may be distinguished from ﬂow-induced changes on the basis of timescales of the changes: the pore pressure response to strain is instantaneous, ﬂow processes are slower. How- ever, a gradual evolution of volumetric strain (see Fig. 13) Yet another limitation lies in the fact that chamber inﬂation generally is not instantaneous. 4.3.1 General considerations As the focus of this study was to identify the different inﬂuences of model parameters on hydraulic head signals, we assumed instantaneous pressur- ization for simplicity. However, it is clear that the analysis of signals during long-term inﬂation is even more complicated as one needs to decipher ﬂow and inﬂation effects – again emphasizing that well data should ideally only be used in conjunction with surface deformation data. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain will result in a slow hydraulic head response. Again, the combination with other monitoring data will help constrain the strain rate to distinguish between reservoir induced level changes from those by ﬂuid ﬂow. in soft aquifers, where ﬂow mostly follows the topographic changes, are comparable to instantaneous pressurization, the hydraulic head signals in stiffer, strain-dominated aquifers are reduced as the ﬂow quickly equilibrates strain-induced pressure changes. Flow works against the increased pressur- ization and the rate of change of inﬂation determines which effect dominates, i.e. whether hydraulic head continues to fall (pressurization dominates) or starts to reach its equilibrium value (ﬂow dominates). Our parametric studies show how poroelastic aquifer re- sponses are inﬂuenced by a variety of source, geometrical and aquifer parameters, which each have the potential to sig- niﬁcantly alter the signal amplitude and development with time and space making the poroelastic processes highly com- plex. Consequently, a change in any of these parameters could lead to a change in an observed hydraulic head. In ad- dition, the porous ﬂow alters the initial hydraulic head signal with time. Therefore, not all observed aquifer pressure tran- sients are necessarily related to a change in the magmatic system, which needs to be carefully considered when inter- preting observed water level changes. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain Porous ﬂow in the lava ﬂow aquifer and therefore evo- lution of signals with time is also signiﬁcantly inﬂuenced Horizontal ﬂow directions in the softer pyroclastic aquifers are not affected by the changed initial strain and head gradi- ent for different L, as ﬂow is dominated by the topographic gradient, which still points away from the volcano. How- ever, the vertical component of ﬂow changes. In the reference simulation, the vertical gradient in hydraulic head change is overwhelmed by the horizontal gradient; hence lateral ﬂow is dominant. In the shorter aquifers (i.e. with larger L val- ues), the vertical gradients become more important and lead to up- or downward components of ﬂow, respectively. That hydraulic head changes are up to 50 % larger close to the lateral boundary of the shorter pyroclastic aquifers than in the reference aquifer is due to the difference in mechani- cal properties of the central, impermeable portion and the outer permeable portion of the aquifer. While initially elas- tically equivalent, the change of head and strain with time due to porous ﬂow in the outer aquifer leads to a mechan- ical boundary at the lateral aquifer onset. Especially in the pyroclastic aquifer, where strain undergoes signiﬁcant ﬂow- induced changes, this discontinuity in strain, head and ﬂow is hard to simulate (compare Fig. C1a) and hence numeri- cal signals close to this boundary should be interpreted with caution. The strain discontinuity is negligible in the lava ﬂow aquifer, where strain does not change much with ﬂow (com- pare Fig. C1b). We only brieﬂy studied the effect of long-term inﬂa- tion, but results show that the time scale of pressurization is non-negligible as ﬂow processes act simultaneously with the response to increased pressurization and can signiﬁ- cantly change the signals. While hydraulic head responses Porous ﬂow in the lava ﬂow aquifer and therefore evo- lution of signals with time is also signiﬁcantly inﬂuenced by the lateral distance between the magma chamber and the aquifer, even though initial hydraulic head values at respec- tive locations are the same. For L = 2km, i.e. in an aquifer Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ 1223 4.3.2 Strain sensitivity and ﬂuid ﬂow If the level changes are thought to be caused by strain, our models suggest that volumetric strain in the aquifer can be directly inferred from measured water level changes, as the simulated initial hydraulic head change perfectly mirrors the strain. This requires a known strain sensitivity, the change of hydraulic head in the aquifer in metres per unit applied strain, which can be assessed by tracking water level changes as a result of predictable excitations. Figure 14 shows the theoret- ical strain sensitivity of the two aquifers used in the reference simulation, determined by dividing the simulated hydraulic head change by the volumetric strain. This has been done along proﬁles through the aquifers. The very small strain and head values close to the transition zone from dilatational to compressional strain lead to numerical errors in the deter- mined strain sensitivities in these locations (which can be reduced by increasing the mesh density), but in general we calculate a consistent value. Strain sensitivity of the pyro- clastic aquifer is about −5×103 m; the lava ﬂow aquifer has a strain sensitivity about 2 orders of magnitude larger. However, the inﬂuence of ﬂow on strain sensitivity is prob- lematic; Fig. 15 shows how the theoretically calculated strain sensitivity changes with time: in the pyroclastic aquifer it quickly increases to about −250 × 103 m before it changes sign to a value of about 100 × 103 m, followed by a de- crease – due to the change of strain from dilation to com- pression associated with the removal of pore ﬂuid. In the lava ﬂow aquifer, strain sensitivity shows a decrease (approaching www.solid-earth.net/6/1207/2015/ www.solid-earth.net/6/1207/2015/ Solid Earth, 6, 1207–1229, 2015 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1224 1224 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 0 5 10 15 20 −5.5 −5 −4.5 Radial distance (km) Strain sensitivity (103m) (a) 0 5 10 15 20 −550 −500 −450 Radial distance (km) Strain sensitivity (103m) (b) Figure 14. Strain sensitivity in the aquifers, determined by dividing simulated hydraulic head change by the volumetric strain, along a pro- ﬁle through the aquifers. (a) Pyroclastic aquifer, (b) lava ﬂow aquifer. Very small strains close to the transition zone from dilatational to compressional strain lead to numerical errors (can be reduced with increasing mesh density). 0 5 10 15 20 −5.5 −5 −4.5 Radial distance (km) Strain sensitivity (103m) (a) Figure 14. Strain sensitivity in the aquifers, determined by dividing simulated hydraulic head change by the volumetric strain, along a pro- ﬁle through the aquifers. (a) Pyroclastic aquifer, (b) lava ﬂow aquifer. Very small strains close to the transition zone from dilatational to compressional strain lead to numerical errors (can be reduced with increasing mesh density). 0 20 40 60 80 100 −300 −200 −100 0 100 200 Time (days) (a) Strain sensitivity (103m) 0 2 4 6 8 10 −500 −400 −300 −200 −100 0 Time (days) (b) Strain sensitivity (103m) Figure 15. Strain sensitivity in the aquifers, determined at a point centrally above the chamber for different simulation times. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. The value signiﬁcantly changes with time, depending on ﬂow processes in the aquifers. 0 2 4 6 8 10 −500 −400 −300 −200 −100 0 Time (days) (b) Strain sensitivity (103m) Figure 15. Strain sensitivity in the aquifers, determined at a point centrally above the chamber for different simulation times. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. The value signiﬁcantly changes with time, depending on ﬂow processes in the aquifers. 0 50 100 150 200 250 300 −5 0 5 0 50 100 150 200 250 3003.5 4 4.5 Reference Caprock Youngs modulus Caprock = 10GPa Time (d) Hydraulic head change (m) Surface Displacement (m) Figure 16. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain volcanoes (e.g. Folch and Gottsmann, 2006; Manconi et al., 2007; Geyer and Gottsmann, 2010). Our investigation of the elastic stratigraphy has shown that hydraulic head change and consequently derived strain can also signiﬁcantly deviate from signals in a homogeneous crust. Especially in settings with a sign-ﬂipped signal, i.e. where the dilatational strain in the aquifer is turned into compression by a stiff cap rock, this inﬂuence becomes crucial. The hydraulic head rises and hence interpretation of the hydraulic head data alone would suggest a deﬂating chamber, while it is really inﬂating. We simulated ground deformation signals with the aim of inves- tigating whether they – if available – could aid with this prob- lem. Figure 16 shows the central hydraulic head and surface displacement signals with time for the pyroclastic aquifer with two different cap rocks: a soft one (Ec = 70 MPa) and a stiff one (Ec = 10 GPa). The latter leads to a change of sign of the volumetric strain. The surface deformation however does not change sign and shows inﬂation of the ground in both cases and can hence be used to indicate that the strain in the aquifer is sign-ﬂipped. help decipher ﬂow processes. While these tests usually pro- vide only a local value, high-resolution time series of head data can in fact provide information on the effective perme- ability of the aquifers. If a poroelastic head response has been identiﬁed, one can observe the equilibration of the pressure change with time – giving information on groundwater ﬂow velocities. In any case, observing the ﬂow behaviour in local aquifers, by installing several observation wells, is a valuable addition to existing monitoring efforts as they can reveal ﬂow patterns caused by head changes, be they strain-induced or caused by other (volcanic) processes. Finally, it is important to know aquifer geometry as the models show that the ﬂow pattern can strongly depend on the lateral distance from the aquifer to the source. 4.3.3 Application of analytical volcano deformation models Even if strain sensitivity has been accurately used to infer volumetric strain, we still face the problem of interpretation of this signal. To invert for the source of volumetric strain, analytical volcano deformation models can be applied. How- ever, these models commonly assume a source in a homoge- neous half space and some only consider spherical or point- like chambers (e.g. Dzurisin, 2007, and references therein). Additionally, all previous approaches to model the deforma- tion due to reservoir inﬂation treated their data as a result of dry deformation. Our results underline that any model using these simpliﬁcations will likely be inadequate when used for interpretation of poroelastic processes. Firstly, the assump- tion of a spherical chamber is a likely source for substantial mistakes and several chamber shapes have to be tested, as chamber shape is one of the two most important parameters inﬂuencing the signal. But even if chamber shape is taken into account, the assumptions of a homogeneous half space and dry deformation can lead to further misinterpretation. The above considerations hint that the apparent inconsis- tency of observed well data and model predictions in the 2000 Usu case (Matsumoto et al., 2002) probably stems from the simpliﬁcations used in the applied Mogi model, which assumes a spherical pressure source in a homogeneous, dry half space. In summary, while water level data can be a valuable addi- tion to monitoring systems and give indications on subsur- face strain, one needs to be careful when interpreting the head as well as strain data. We need to take into account that many parameters inﬂuence water level changes and that most of the commonly used analytical dry deformation mod- els might fail to explain them. www.solid-earth.net/6/1207/2015/ Central vertical surface deformation and hydraulic head change with time for the pyroclastic aquifer overlain by different cap rocks, showing the effect of a sign-ﬂipped strain in comparison to the reference case. 0 50 100 150 200 250 300 −5 0 5 0 50 100 150 200 250 3003.5 4 4.5 Reference Caprock Youngs modulus Caprock = 10GPa Time (d) Hydraulic head change (m) Surface Displacement (m) zero), which is comparable to the decrease of the hydraulic head change. Hence, the strain sensitivity value determined from aquifer responses to known strains only provides accu- rate strains when applied to the initial hydraulic head change, as it does not take ﬂow and resulting poroelastic processes into account, and dense time series of well data (and ideally simultaneously recorded ground deformation data) are nec- essary to catch this response. y p The better the local hydrology is known, the more value lies in well monitoring. Our simulations show how different soft and stiff aquifers behave in a strain ﬁeld. Hence, knowl- edge of the lithology and/or determination of the strain sen- sitivity is important to discriminate between aquifer types. A low sensitivity value would indicate an aquifer similar to the presented pyroclastic example, which entails crucial infor- mation: the topographic gradient due to ground deformation can easily dominate over a strain-induced pressure gradient, and if this is the case, dilatational strain will quickly change to compression. Additionally, softer aquifers are more prone to the sign-ﬂip effect. Figure 16. Central vertical surface deformation and hydraulic head change with time for the pyroclastic aquifer overlain by different cap rocks, showing the effect of a sign-ﬂipped strain in comparison to the reference case. Information about ﬂow in the aquifer is important and the acquisition of permeability data, e.g. via pumping tests, should be part of hydrological monitoring efforts as it can www.solid-earth.net/6/1207/2015/ should be part of hydrological monitoring efforts as it can Solid Earth, 6, 1207–1229, 2015 1225 5 Conclusions The stress absorption of a pore ﬂuid leads to a reduction of initial strain in the aquifer when compared to an elasti- cally equivalent dry layer. If the initial strain is used to infer the magmatic source based on a model for dry deformation, its strength can therefore be underestimated. “Dry” strain is reached in the lava ﬂow aquifer after porous ﬂow has equili- brated the strain-induced pressure gradient. So, this problem could be solved when sufﬁciently dense time series of hy- draulic head data are available: strain sensitivity can be com- bined with the evolution of signals with time to infer initial as well as equilibrium “dry” strain. In aquifers that are dom- inated by the topographic gradient, the presence of the pore ﬂuid even leads to a reversal of dilatational to compressional strain and the application of dry deformation models is not possible. In this study we presented fully coupled numerical models to investigate the interaction between solid mechanics and ﬂuid ﬂow in porous media. We have shown that strain due to the inﬂation of a magma chamber leads to signiﬁcant hydraulic head changes and porous ﬂow in the local hydrology. The ﬂexibility of the ﬁnite element analysis method allowed us to perform extensive parametric studies providing detailed in- sights in these poroelastic processes. Parameters controlling aquifer behaviour are in order of importance (i) the shape, volume and pressurization strength of the magma chamber (ii) the phase of the pore ﬂuid and the permeability of the aquifer (iii) chamber and aquifer depths and the aquifer’s Young’s modulus. Magmatic source properties and the dis- tance between chamber and aquifer determine the strain ﬁeld; strain partitioning is deﬁned by the elastic stratigraphy of the crust. Elastic and ﬂow parameters of the aquifer deﬁne its re- sponse to this strain and how head and strain signals change with time due to porous ﬂow. The third assumption of a homogeneous half space is pre- carious as volcanoes are strongly heterogeneous – several previous studies have already shown that mechanical hetero- geneities in the subsurface affect the ground deformation at www.solid-earth.net/6/1207/2015/ Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ Appendix A: Biot–Willis coefﬁcient (a) pyroclastic aquifer (b) lava ﬂow aquifer. The different dependence of 1h on α for the two aquifers can be mathematically explained by considering an order of magnitude analysis of the deﬁnition of the speciﬁc storage (Eq. 6): Radial distance (km) Radial distance (km) Radial distance (km) Figure C1. Volumetric strain after 1000 days along proﬁles through the aquifers for the reference case in comparison to an aquifer with a central impermeable portion out to a radial distance of 2 km. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. S = φχf + (α −φ)(1 −α) K = φχf + (α −φ)(1 −α)3(1 −2ν) E ≈10−1 × 10−10 + 10−1 × (1 −α) × 100 × 10−1 107 (for the pyroclastic aquifer) = 10−11 + 10−9 × (1 −α). aquifer is no longer central, but laterally offset by up to 1 km (shown for zCH = 1km and zaq = 200m in Fig. B1). E ≈10−1 × 10−10 + 10−1 × (1 −α) × 100 × 10−1 107 (for the pyroclastic aquifer) (for the pyroclastic aquifer) Appendix A: Biot–Willis coefﬁcient 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 4 −0.07 −0.06 −0.05 −0.04 −0.03 −0.02 −0.01 0 0.01 Distance from Centre (km) Hydraulic head change (m) The inﬂuence of the Biot–Willis coefﬁcient is quite com- plex, as it deﬁnes the coupling terms in the constitutive equa- tions and is involved in the deﬁnition of speciﬁc storage of the aquifer as well. Figure A1 shows the effect of varying the coupling parameter on the initial central hydraulic head change for the two different aquifer types. In the pyroclastic aquifer, the head fall ﬁrst strongly decreases with increasing α, then reaches a plateau at α = 0.8 before decreasing mini- mally when approaching α = 1. In the lava ﬂow aquifer, the hydraulic head change is larger for larger α. 0.4 0.5 0.6 0.7 0.8 0.9 1 −0.022 −0.02 −0.018 −0.016 −0.014 −0.012 Hydraulic head change (m) Biot-Willis coefficient (a) 0 0.2 0.4 0.6 0.8 1 −25 −20 −15 −10 −5 0 Biot-Willis coefficient Hydraulic head change (m) (b) Figure A1. Dependence of central, initial hydraulic head change on the Biot–Willis coefﬁcient. (a) pyroclastic aquifer (b) lava ﬂow aquifer. 0.4 0.5 0.6 0.7 0.8 0.9 1 −0.022 −0.02 −0.018 −0.016 −0.014 −0.012 Hydraulic head change (m) Biot-Willis coefficient (a) 0 0.2 0.4 0.6 0.8 1 −25 −20 −15 −10 −5 0 Biot-Willis coefficient Hydraulic head change (m) (b) Figure B1. Hydraulic head change proﬁle in the pyroclastic aquifer for zCH = 1km. Radial distance (km) 0 5 10 15 20 −15 −10 −5 0 5 Reference L=2km (a) Volumetric strain (microstrain) 0 5 10 15 20 −5 0 5 10 15 Reference L=2km Radial distance (km) Volumetric strain (microstrain) (b) 2 2.4 −5 0 5 1.95 2 2.05 6.2 6.6 7 Figure C1. Volumetric strain after 1000 days along proﬁles through the aquifers for the reference case in comparison to an aquifer with a central impermeable portion out to a radial distance of 2 km. (a) pyroclastic aquifer, (b) lava ﬂow aquifer. 0 5 10 15 20 −5 0 5 10 15 Reference L=2km Radial distance (km) Volumetric strain (microstrain) (b) 1.95 2 2.05 6.2 6.6 7 Radial distance (km) 0 5 10 15 20 −15 −10 −5 0 5 Reference L=2km (a) Volumetric strain (microstrain) 2 2.4 −5 0 5 Biot-Willis coefficient Figure A1. Dependence of central, initial hydraulic head change on the Biot–Willis coefﬁcient. Appendix C: Strain discontinuity at the lateral aquifer boundary = 10−11 + 10−9 × (1 −α). So, in the pyroclastic aquifer for α ≤0.9, φχf is 1 order of magnitude smaller than the right summand, which therefore dominates the deﬁnition of S. For α approaching 1, both terms become important. For the lava ﬂow aquifer, E ≈1010 and therefore the right summand has the order of magnitude 10−11 × (1 −α) and φχf is the dominating term in the def- inition of S for all α. Therefore, in the pyroclastic aquifer, changing α changes the coupling terms and the speciﬁc stor- age, while in the lava ﬂow aquifer changing α has almost no effect on S. When the central portion of the aquifer is replaced with an area of zero permeability, the change of head and strain with time due to porous ﬂow in the outer aquifer leads to a me- chanical boundary at the lateral aquifer onset. Especially in the pyroclastic aquifer, where strain undergoes signiﬁcant ﬂow-induced changes, this leads to a discontinuity in strain (Fig. C1). Appendix B: Extra information on the inﬂuence of chamber depth For a shallow magma chamber, in a situation with no sign- ﬂipped strain, the maximum head change in the pyroclastic Solid Earth, 6, 1207–1229, 2015 K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1226 However, using common analytical deformation models for the interpretation of this strain information is problem- atic, as several assumptions of these models can lead to substantial misinterpretation. They are only applicable for a comparatively homogeneous crust (i.e. Eaq ≈Ec ≈Eh), when one either accounts for ﬂuid-induced strain reduction or considers an aquifer with very little strain reduction. The shape of the chamber needs to be taken into account as well. One aim of this study was to investigate the accuracy of the method to combine strain sensitivities with deformation models to interpret observed hydraulic head changes. Our models show that volumetric strain in the aquifer can indeed be inferred from hydraulic head changes using strain sen- sitivities, under certain conditions. Firstly, other causes for hydraulic head change have to be excluded, ideally by con- sulting other monitoring systems. Dense time series of well level data need to be acquired in order to account for ﬂow processes and to measure the initial hydraulic head change. Additionally, we need to ensure that strain sensitivities have been accurately determined and have not changed with time due to changes in the hydrology. The hydraulic head signal is very sensitive to source vol- ume, shape and pressurization value. This suggests that if we have a detailed knowledge on the hydrology, some infor- mation about the source can be gained from hydraulic head changes – although solutions will always be non-unique. Our analysis has shown the necessity of numerical models to account for the large number of parameters that signiﬁ- cantly inﬂuence the results. Nevertheless, well water levels and groundwater ﬂow reﬂect subsurface strain and therefore are a valuable complement to other monitoring systems. Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1227 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 4 −0.07 −0.06 −0.05 −0.04 −0.03 −0.02 −0.01 0 0.01 Distance from Centre (km) Hydraulic head change (m) Figure B1. Hydraulic head change proﬁle in the pyroclastic aquifer for zCH = 1km. References Hickey, J. and Gottsmann, J.: Benchmarking and developing numer- ical ﬁnite element models of volcanic deformation, J. Volcanol. Geoth. Res., 280, 126–130, 2014. Adam, L. and Otheim, T.: Elastic laboratory measurements and modeling of saturated basalts, J. Geophys. Res.-Sol. Ea., 118, 840–851, 2013. Hurwitz, S. and Johnston, M. J. S.: Groundwater level changes in a deep well in response to a magma intrusion event on Kilauea Volcano, Hawai’i, Geophys. Res. Lett., 30, 2173, doi:10.1029/2003GL018676, 2003. Bonaccorso, A., Calvari, S., Linde, A., Sacks, S., and Boschi, E.: Dynamics of the shallow plumbing system investigated from borehole strainmeters and cameras during the 15 March, 2007 Vulcanian paroxysm at Stromboli volcano, Earth Planet. Sc. Lett., 357–358, 249–256, 2012. Hurwitz, S., Christiansen, L., and Hsieh, P. A.: Hydrother- mal ﬂuid ﬂow and deformation in large calderas: Inferences from numerical simulations, J. Geophys. Res., 112, B02206, doi:10.1029/2006JB004689, 2007. Biot, M. A.: Mechanics of Deformation and Acoustic Propagation in Porous Media, J. Appl. Phys., 33, 1482–1498, 1962. Hutnak, M., Hurwitz, S., Ingebritsen, S. E., and Hsieh, P. A.: Nu- merical models of caldera deformation: Effects of multiphase and multicomponent hydrothermal ﬂuid ﬂow, J. Geophys. Res. -Sol.Ea., 114, B04411, doi:10.1029/2008JB006151, 2009. Chiodini, G., Caliro, S., De Martino, P., Avino, R., and Gherardi, F.: Early signals of new volcanic unrest at Campi Flegrei caldera? Insights from geochemical data and physical simulations, Geol- ogy, 40, 943–946, 2012. Jonsson, S., Segall, P., Pedersen, R., and Björnsson, G.: Post- earthquake ground movements correlated to pore-pressure tran- sients, Nature, 424, 179–183, 2003. COMSOL: COMSOL Subsurface Flow Module User’s Guide, Ver- sion 4.4, COMSOL, 2013. Davis, E., Wang, K., Thomson, R., Becker, K., and Cassidy, J.: An episode of seaﬂoor spreading and associated plate deformation inferred from crustal ﬂuid pressure transients, J. Geophys. Res., 106, 21953–21963, 2001. Linde, A. T., Sacks, S., Hidayat, D., Voight, B., Clarke, A., Elsworth, D., Mattioli, G., Malin, P., Shalev, E., Sparks, S., and Widiwijayanti, C.: Vulcanian explosion at Soufrière Hills Vol- cano, Montserrat on March 2004 as revealed by strain data, Geophys. Res. Lett., 37, L00E07, doi:10.1029/2009GL041988, 2010. Dzurisin, D.: Volcano Deformation, Springer, 2007. Fagents, S., Greggs, T., and Lopes, R. (Eds.): Modeling Volcanic Processes, The Physics and Mathematics of Volcanism, Cam- bridge University Press, 2013. Manconi, A., Walter, T. R., and Amelung, F.: Effects of mechanical layering on volcano deformation, Geophys. J. Int., 170, 952–958, 2007. Fetter, C.: Applied Hydrogeology, Macmillan Company, New York, 1994. K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1228 Acknowledgements. The research leading to these results has received funding from the People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme (FP7/2007–2013) under the project NEMOH, REA agreement no. 289976. Additional funding was provided by the MED-SUV project, under grant agreement no. 308665, and the VUELCO project under grant agreement no. 282759, both part of the European Union’s Seventh Framework Programme. We thank Micol Todesco, Maurizio Battaglia and Maurizio Bonafede for their constructive reviews that helped to signiﬁcantly improve the manuscript. Gercek, H.: Poisson’s ratio values for rocks, Int. J. Rock. Mech. Min., 44, 1–13, 2007. Geyer, A. and Gottsmann, J.: The inﬂuence of mechanical stiff- ness on caldera deformation and implications for the 1971–1984 Rabaul uplift (Papua New Guinea), Tectonophysics, 483, 399– 412, 2010. Gottsmann, J. and Odbert, H.: The effects of thermomechanical heterogeneities in island arc crust on time-dependent preerup- tive stresses and the failure of an andesitic reservoir, J. Geophys. Res.-Sol. Ea., 119, 6, 4626–4639, 2014. Gudmundsson, A.: Rock fractures in geological processes, Cam- bridge University Press, 2011. Edited by: M. Heap Heap, M. J., Baud, P., Meredith, P. G., Vinciguerra, S., and Reuschlé, T.: The permeability and elastic moduli of tuff from Campi Flegrei, Italy: implications for ground deformation mod- elling, Solid Earth, 5, 25–44, doi:10.5194/se-5-25-2014, 2014. References Matsumoto, N., Sato, T., Matsushima, N., Akita, F., Shibata, T., and Suzuki, A.: Hydrological anomalies associated with crustal de- formation before the 2000 eruption of Usu volcano, Japan, Geo- phys. Res. Lett., 29, 1057, doi:10.1029/2001GL013968, 2002. Folch, A. and Gottsmann, J.: Faults and ground uplift at active calderas, Geol. Soc. Lond. Spec. Publ., 269, 109–120, 2006. Fournier, N. and Chardot, L.: Understanding volcano hydrother- mal unrest from geodetic observations: insights from nu- merical modeling and application to White Island Volcano, New Zealand, J. Geophys. Res.-Sol. Ea., 117, B11208, doi:10.1029/2012JB009469, 2012. Mogi, K.: Relations between the eruptions of various volcanoes and the deformations of the ground surfaces around them, B. Earthq. Res. I. Tokyo, 36, 99–134, 1958. Newhall, C., Albano, S., Matsumoto, N., and Sandoval, T.: Roles of groundwater in volcanic unrest, J. Geol. Soc. Phil, 56, 69–84, 2001. Freeze, R. and Cherry, J.: Groundwater, Prentice-Hall, Englewood Cliffs, N. J., 1979. Gaeta, F. S., De Natale, G., and Rossano, S.: Genesis and evolu- tion of unrest episodes at Campi Flegrei caldera: The role of the thermal ﬂuid-dynamical processes in the geothermal system, J. Geophys. Res., 103, 20921–20933, 1998. Rice, J. R. and Cleary, M. P.: Some Basic Stress Diffusion Solutions for Fluid-Saturated Elastic Porous Media With Compressible Constituents, Rev. Geophys. Space Phys., 14, 227–241, 1976. Rinaldi, A., Todesco, M., and Bonafede, M.: Hydrothermal instabil- ity and ground displacement at the Campi Flegrei caldera, Phys. Earth Planet. In., 178, 155–161, 2010. Geotechdata.info: Soil Young’s Modulus, available at: http: //geotechdata.info/parameter/soil-elastic-young-modulus.html, 2013. Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ www.solid-earth.net/6/1207/2015/ K. Strehlow et al.: Poroelastic responses of conﬁned aquifers to subsurface strain 1229 Takahashi, H. T. S., Yamaguchi, T. R. I., Okazaki, N., and Akita, F.: Volcanic strain change prior to an earthquake swarm observed by groundwater level sensors in Meakan-dake, Hokkaido, Japan, J. Volcanol. Geoth. Res., 215, 1–7, 2012. Roeloffs, E.: Poroelastic techniques in the study of earthquake- related hydrologic phenomena, Adv. Geophys., 37, 135–195, 1996. Rutqvist, J., Wu, Y.-S., Tsang, C.-F., Bodvarsson, G.: A modelling approach for analysis of coupled multiphase ﬂuid ﬂow, heat transfer, and deformation in fractured porous rock, Int. J. Rock. Mech. Min., 39, 429–442, 2002. Todesco, M., Rutqvist, J., Chiodini, G., Pruess, K., and Olden- burg, C. M.: Modeling of recent volcanic episodes at Phlegrean Fields (Italy): Geochemical variations and ground deformation, Geothermics, 33, 531–547, 2004. Shibata, T. and Akita, F.: Precursory changes in well water level prior to the March, 2000 eruption of Usu volcano, Japan, Geo- phys. Res. Lett., 28, 1799–1802, 2001. Verma, M.: Steam tables for pure water as an ActiveX component in Visual Basic 6.0, Comput. Geosci., 29, 1155–1163, 2003. Shibata, T., Matsumoto, N., Akita, F., Okazaki, N., Takahashi, H., and Ikeda, R.: Linear poroelasticity of groundwater levels from observational records at wells in Hokkaido, Japan, Tectono- physics, 483, 305–309, 2010. Voight, B., Linde, A. T., Sacks, I. S., Mattioli, G. S., Sparks, R. S. J., Elsworth, D., Hidayat, D., Malin, P. E., Shalev, E., Widiwi- jayanti, C., Young, S., Bass, V., Clarke, A., Dunkley, P., John- ston, W., McWorther, N., Neuberg, J., and Williams, P.: Un- precedented pressure increase in deep magma reservoir trig- gered by lava-dome collapse, Geophys. Res. Lett., 33, L03312, doi:10.1029/2005GL024870, 2006. Stefansson, V.: The Kraﬂa geothermal ﬁeld, in: Geothermal Sys- tems: Principles and Case Histories, Wiley, New York, 273–294, 1981. Wang, H. F.: Theory of Poroelasticity with Applications to Geome- chanics and Hydrogeology, Princeton University Press, 2000. Solid Earth, 6, 1207–1229, 2015 www.solid-earth.net/6/1207/2015/ www.solid-earth.net/6/1207/2015/
https://openalex.org/W3122962155	https://www.scielo.br/j/bjmbr/a/F9TJM5QQMSrysSwGKTqtfSz/?format=pdf&lang=en	Portuguese	null	Is soft drink consumption associated with gestational hypertension? Results from the BRISA cohort	Brazilian Journal of Medical and Biological Research	2,021	cc-by	6,671	J.M.A. Barbosa1 00 , A.A.M. da Silva1 00 00, G. Kac2 00 , V.M.F. Simões1 00 , H. Bettiol3 00 , R.C. Cavalli4 00 , M.A. Barbieri3 00 , and C.C.C. Ribeiro1,5 00 1Departamento de Saúde Pública, Universidade Federal do Maranhão, São Luís, MA, Brasil 2Departamento de Nutric¸ão Social Aplicada, Instituto de Nutric¸ão Josué de Castro, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brasil 3Departamento de Puericultura e Pediatria Faculdade de Medicina de Ribeirão Universidade de São Paulo 1Departamento de Saúde Pública, Universidade Federal do Maranhão, São Luís, MA, Brasil 2Departamento de Nutric¸ão Social Aplicada, Instituto de Nutric¸ão Josué de Castro, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brasil 3Departamento de Puericultura e Pediatria, Faculdade de Medicina de Ribeirão, Universidade de São Paulo, Ribeirão Preto, SP, Brasil 4Departamento de Ginecologia e Obstetrícia Faculdade de Medicina de Ribeirão Preto Universidade de São Paulo Departamento de Nutric¸ão Social Aplicada, Instituto de Nutric¸ão Josué de Castro, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brasil 3Departamento de Puericultura e Pediatria, Faculdade de Medicina de Ribeirão, Universidade de São Paulo, Ribeirão Preto, SP, Brasil 4Departamento de Ginecologia e Obstetrícia, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto SP Brasil 5Departamento de Odontologia II, Universidade Federal do Maranhão, São Luís, MA, Brasil Abstract It is still unknown whether excessive consumption of sugar-sweetened beverages may be linked to gestational hypertensive disorders, other than preeclampsia. This study investigated the association between soft drink consumption and hypertension during pregnancy, analyzing the relationship from the perspective of counterfactual causal theory. Data from pregnant women of the BRISA cohort were analyzed (1,380 in São Luis and 1,370 in Ribeirão Preto, Brazil). The explanatory variable was the frequency of soft drink consumption during pregnancy obtained in a prenatal interview. The outcome was gestational hypertension based on medical diagnosis, at the time of delivery. A theoretical model of the association between soft drink consumption and gestational hypertension was constructed using a directed acyclic graph. Marginal structural models (MSM) weighted by the inverse of the probability of soft drink consumption were also employed. Using Poisson regression analysis, high soft drink consumption (X7 times/week) was associated with gestational hypertension in São Luís (RR=1.48; 95%CI: 1.03–2.10), in Ribeirão Preto (RR=1.51; 95%CI: 1.13–2.01), and in the two cohorts combined (RR=1.45; 95%CI: 1.16–1.82) compared to lower exposure (o7 times/week). In the MSM, the association between high soft drink consumption and gestational hypertension was observed in Ribeirão Preto (RR=1.63; 95%CI: 1.21–2.19) and in the two cohorts combined (RR=1.51; 95%CI: 1.15–1.97), but not in São Luís (RR=1.26; 95%CI: 0.79–2.00). High soft drink consumption seems to be a risk factor for gestational hypertension, suggesting that it should be discouraged during pregnancy. Key words: Gestational hypertensive disorders; Soft drink consumption; Cohort study; Marginal structural model; Pregnancy Brazilian Journal of Medical and Biological Research (2021) 54(1): e10162, http://dx.doi.org/10.1590/1414-431X202010162 ISSN 1414-431X Research Article Brazilian Journal of Medical and Biological Research (2021) 54(1): e10162, http://dx.doi.org/10.1590/1414-431X202010162 ISSN 1414-431X Research Article 1/9 Is soft drink consumption associated with gestational hypertension? Results from the BRISA cohort J.M.A. Barbosa1 00 , A.A.M. da Silva1 00 00, G. Kac2 00 , V.M.F. Simões1 00 , H. Bettiol3 00 , R.C. Cavalli4 00 , M.A. Barbieri3 00 , and C.C.C. Ribeiro1,5 00 Data collection Gestational hypertension has a faster evolution and distinct pathophysiological mechanisms compared to hypertension outside pregnancy. Metabolic adaptations are present in pregnancy and inﬂammatory responses are exacerbated in gestational hypertension (10). Therefore, it is biologically plausible that the risk of high soft drink consumption associated with gestational hypertension is somewhat different from the risk of high soft drink con- sumption associated with hypertension outside pregnancy. To the best of our knowledge, only two studies con- ducted on pregnant women have found that high daily sugar intake in pregnancy was associated with pre- eclampsia (5,11). Furthermore, when considering the asso- ciation between added sugars or SSBs and preeclampsia separately, the association was strongest for SSBs (5). Another study pointed out that intake of added sugars is a risk factor for preeclampsia, but this association lost statistical signiﬁcance in the multiple logistic regression model, and only the association between SSBs and preeclampsia remained signiﬁcant (OR=1.33; 95%CI: 1.10– 1.61) (11). Gestational hypertension has a faster evolution and distinct pathophysiological mechanisms compared to hypertension outside pregnancy. Metabolic adaptations are present in pregnancy and inﬂammatory responses are exacerbated in gestational hypertension (10). Therefore, it is biologically plausible that the risk of high soft drink consumption associated with gestational hypertension is somewhat different from the risk of high soft drink con- sumption associated with hypertension outside pregnancy. Data were collected during an interview when a structured questionnaire was applied and clinical exam- inations in the prenatal period (baseline) and at delivery (from January 2010 to June 2011) were performed. During the prenatal period, 1,447 women were interviewed in the São Luís cohort and 1,400 in the Ribeirão Preto cohort. A total of 1,381 women in São Luís and 1,370 in Ribeirão Preto were interviewed again during the ﬁrst 24 h after delivery. One pregnant woman in São Luís was excluded from the study because she did not provide information about hypertension during pregnancy. To the best of our knowledge, only two studies con- ducted on pregnant women have found that high daily sugar intake in pregnancy was associated with pre- eclampsia (5,11). Furthermore, when considering the asso- ciation between added sugars or SSBs and preeclampsia separately, the association was strongest for SSBs (5). Introduction Dietary factors seem to play an important role in the etiology of HDP (3,4). High scores of the Healthy Eating Index adapted for use in pregnant women have been shown to protect against preeclampsia (OR=0.87; 95%CI: 0.76–1.00) (3). A meta-analysis concluded that women with high energy intake were at higher risk of developing preeclampsia compared to their counterparts (4). The main source of energy intake during pregnancy is exces- sive sugar consumption, especially soft drinks (5). High soft drink consumption has been associated with over- weight/obesity, metabolic syndrome, and type II diabetes (6). All these metabolic disorders are associated to HDP (7). Systematic reviews (8,9) have shown that high Hypertensive disorders of pregnancy (HDP) include: 1) chronic hypertension (of any cause that precedes preg- nancy); 2) gestational hypertension (high blood pressure (BP) after 20 weeks of gestation in the absence of preeclampsia); 3) preeclampsia-eclampsia (high BP after 20 weeks of gestation with proteinuria or any of the severe features of preeclampsia); and 4) chronic hypertension with superimposed preeclampsia (chronic hypertension in association with preeclampsia) (1). HDP are a major cause of maternal mortality and perinatal outcomes such as prematurity, perinatal asphyxia, and low birth weight (1), and also increase the risk of future diseases in the mother and child (2). Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Soft drinks and gestational hypertension 2/9 in order to determine gestational age (GA) (13) and sin- gleton pregnancy. consumption of soft drinks and other sugar-sweetened beverages (SSBs) are associated with increased BP and hypertension in non-pregnant women. Exposure variables, confounding factors, mediators, and outcome High intake of soft drinks during pregnancy, analyzed by conventional regression models, seems to be asso- ciated with more severe hypertension (preeclampsia) (5,11). The present study was conducted to investigate the association between soft drink consumption during pregnancy and gestational hypertension in a prenatal cohort, analyzing these relationships from the perspective of counterfactual causal theory. The exposure was soft drink consumption during pregnancy up to 22/25 weeks of GA. This information was obtained from answers to the following two questions: 1) On how many days of the week do you consume soft drinks? 2) How many times a day do you take soft drinks? The frequencies of weekly intake (zero to seven days a week) and of daily intake (once to six times a day) of soft drinks reported by the women were multiplied and categorized as a dichotomous variable corresponding to no or low consumption (o7 times per week) or high consumption (X7 times per week) (14). This cutoff point was assumed since the majority of previous studies has shown that daily SSBs intake is associated to worse health outcomes, such as obesity and hypertension (8,9,15). Tertiles of soft drink consumption (measured in times per week) were also calculated to test a dose- response gradient in the association between soft drink consumption and gestational hypertension, as follows: for São Luís: 1st tertile (no consumption), 2nd tertile (once a week), and 3rd tertile (two or more times per week); and for Ribeirão Preto: 1st tertile (up to once a week), 2nd tertile (two to six times per week), and 3rd tertile (seven or more times per week). Data collection Another study pointed out that intake of added sugars is a risk factor for preeclampsia, but this association lost statistical signiﬁcance in the multiple logistic regression model, and only the association between SSBs and preeclampsia remained signiﬁcant (OR=1.33; 95%CI: 1.10– 1.61) (11). The following information was collected in the prenatal period: age, family income, smoking habit during preg- nancy, daily and weekly soft drink consumption, reported weight before pregnancy, and height measured with an Altura Exatas (TBW, Brazil) stadiometer. Additional infor- mation was obtained after delivery: schooling, skin color, parity, and hypertension based on medical diagnosis. Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Material and Methods Family income was divided into tertiles as follows: for São Luís: 1st tertile (US$ 446.80), 2nd tertile (US$ 755.80), and 3rd tertile (US$ 1,162.70); and for Ribeirão Preto: 1st tertile (US$ 552.30), 2nd tertile (US$ 872.10), and 3rd tertile (US$ 1,279.10), considering a US dollar exchange value of 1.7 for the Brazilian real (July 2010). Parity was categorized as 1, 2–3, and X4 deliveries. Smoking habit in the current pregnancy was categorized as yes or no. Pre-gestational body mass index (BMI) was calculated according to the WHO (World Health Organization) criterion (17) and assigned to three categories: p24.9, 25–29.9, or X30 kg/m2. To estimate the total effect (non-mediated effect) of the association between soft drink consumption and hyper- tension during pregnancy, the following variables were suggested to be included in the minimally sufﬁcient adjustment set for confounding, by the application of the back-door criterion: maternal schooling, family income, and age of the pregnant women. According to the back- door criterion, to estimate this effect, skin color, parity, BMI, junk food consumption, and smoking should not enter the model. Junk food intake was considered a potential confound- ing factor for the association between soft drink consump- tion and gestational hypertension since these foods are usually consumed together with soft drinks and are asso- ciated with higher BMI (18). Junk food intake was based on consumption of the following items: hamburger, cheese- burger, or cheese and meat sandwich, sausage, hot dog and salami, ham, baloney, potato chips, packaged salty snacks, and popcorn. The frequency of consumption of each junk food item was ﬁrst categorized as follows: up to three times a month, once to twice a week, and X3 times a week. The sum of each of these categories of consumption was calculated for all food items and was then divided into tertiles as follows: for both São Luís and Ribeirão Preto, 1st tertile (up to three times a month), 2nd tertile (once to twice a week), and 3rd tertile (X3 times a week). The chi-squared test was performed to compare the characteristics between pregnant women in the São Luís and Ribeirão Preto cohorts, considering a P-value o0.05 as statistically signiﬁcant. Poisson regression analysis with robust adjustment of variance was used to test the association between soft drink consumption and hyper- tension during the current pregnancy. The level of sig- niﬁcance was set at 0.05. Ethical aspects p The study was approved by the Research Ethics Committee of the University Hospital of the Federal University of Maranhão (#4771/2008-30) and by the Research Ethics Committee of the University Hospital, Faculty of Medicine of Ribeirão Preto, University of São Paulo (#4116/2008). Material and Methods Risk ratios and their respective conﬁdence intervals (95%CI) were estimated. For the MSM analysis, we ﬁrst estimated a logistic regression model to predict soft drink consumption including as predictors variables signiﬁcantly associated with this exposure (age, schooling, junk food consump- tion, and smoking habit during pregnancy) (Table 1). Based on this model, we calculated the probability of soft drink consumption for each individual. A marginal struc- tural model (Poisson regression with robust standard errors) based on the counterfactual logic including soft drink consumption during pregnancy as the exposure and gestational hypertension as the outcome was then esti- mated, weighted by the inverse of the probability of soft drink consumption. All analyses were carried out using STATA software version 12.0 (StataCorp, USA). Material and Methods This study is part of a large research project that aimed to evaluate risk factors for preterm birth in a cohort of pregnant women selected during the ﬁrst weeks of pregnancy named BRISA (12) (Brazilian Ribeirão Preto and São Luís Birth Cohort Studies, Portuguese acronym). This cohort was carried out in two cities: São Luís, in the Brazilian Northeast and Ribeirão Preto, in the Brazilian Southeast. A convenience sample was carried out due to the lack of registries of pregnant women or women receiving pre- natal care that were necessary to obtain a random sample representing the pregnant women population in São Luís and in Ribeirão Preto; thus, the reference population consisted of pregnant women users of prenatal care of all major public and private hospitals in both cities. At baseline, the pregnant women were invited for an interview that was held from the 22nd to the 25th week of gestation. At delivery (follow-up), the women were reinterviewed after the birth of their babies. The outcome was hypertension during the current pregnancy based on medical diagnosis (yes or no). This information was obtained from the response to the following question: ‘‘Did you have hypertension (high blood pres- sure) in the current pregnancy, diagnosed by a doctor or a nurse?’’ Sociodemographic and economic variables were con- sidered to be confounding factors since these variables Inclusion criteria were having been submitted to obstetric ultrasound before the 20th week of pregnancy Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Soft drinks and gestational hypertension 3/9 are ancestral (common factors) regarding both exposure (16) and outcome (3). Maternal age was categorized as 14–19, 20–24, 25–29, 30–34, and X35 years. Skin color was self-reported and categorized as white, black, brown (which represents miscegenation of black and white Brazilians), and yellow (Asian peoples). In addition to the relationships between the observed variables described above, two non-observed variables, salt consumption and total energy intake, were included in the theoretical model of association between soft drink consumption and gestational hypertension, since they might be important confounders (1,4). Since soft drinks are important sources of sodium and of discretionary calories (6), these two variables were considered as mediators of the effect of soft drinks on gestational hypertension (Figure 1). Maternal schooling was categorized as p4, 5–8, 9–11, and X12 years. Theoretical model and data analysis Two analytical strategies were employed to answer the question of the present study. Causal diagrams were ﬁrst elaborated to guide the analysis of the effect of soft drink consumption on hypertension during pregnancy, with control for confounding factors in the most appropriate manner and at the same time avoiding unnecessary adjustments (19). Second, marginal structural models (MSM) estimated with weighting by the inverse of the probability of soft drink consumption were used to estimate the association between soft drink consumption and hypertension during pregnancy. MSM are used for the causal analysis of observational data as long as the assumptions of exchangeability, positivity, well-deﬁned exposures, and a correct model speciﬁcation are satisﬁed (20). A theoretical model was constructed to analyze the association between soft drink consumption and gesta- tional hypertension using directed acyclic graphs (DAG) elaborated using Dagitty 2.2s (owww.dagitty.net4) software. Results In São Luís, 16.9% of the 1,380 pregnant women reported gestational hypertension and 9.1% consumed Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 4/9 Soft drinks and gestational hypertension Figure 1. Theoretical model of the association between soft drink consumption and hypertension during the current pregnancy based on a directed acyclic graph (DAG). Figure 1. Theoretical model of the association between soft drink consumption and hypertension during the current pregnancy based on a directed acyclic graph (DAG). soft drinks X7 times per week. In the Ribeirão Preto cohort, 14.0% of the 1,370 pregnant women were hyper- tensive and 30.4% consumed soft drinks X7 times per week (Table 1). The median value of consumption in Ribeirão Preto was 6 times per week, while in São Luís the median value of consumption was 2 times per week (data not shown in tables). 95%CI: 1.01–1.57; P=0.038). A dose-response relationship was also demonstrated between soft drink consumption and gestational hypertension in São Luís (RR=1.16; 95% CI: 1.01–1.33; P=0.036), in Ribeirão Preto (RR=1.20; 95% CI: 1.00–1.44; P=0.039), and in the two cohorts combined (RR=1.12; 95%CI: 1.00–1.25; P=0.033) (Table 2). Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Discussion Age, schooling, junk food consumption, and smoking habit were predictors of soft drink consumption by the pregnant women (Supplementary Table S1). The results of the present study suggested that high soft drink consumption (X7 times per week) inﬂuenced gestational hypertension, even after adjustment for con- founding variables chosen based on the back-door crite- rion in a DAG. High soft drink consumption was associated with gestational hypertension in each city of the BRISA cohort and also in the combined analysis of the two cities. A dose response gradient between soft drink consumption and gestational hypertension was also observed. In the Poisson regression analysis, soft drink consump- tion X7 times per week was associated with gestational hypertension in São Luís (RR=1.48; 95%CI: 1.03–2.10; P=0.030), in Ribeirão Preto (RR=1.51; 95%CI: 1.13–2.01; P=0.005), and in the two cohorts combined (RR=1.45; 95% CI: 1.16–1.82; P=0.001) (Supplementary Table S2). In the MSM, soft drink consumption X7 times per week was signiﬁcantly associated with hypertension in the Ribeirão Preto cohort (RR=1.63; 95%CI: 1.21–2.19; P= 0.001) and in the two cohorts combined (RR=1.51; 95% CI: 1.15–1.97; P=0.002), but not in the São Luís cohort alone (RR=1.26; 95%CI: 0.79–2.00; P=0.328) (Table 2). An association between high soft drink consumption and gestational hypertension for the Ribeirão Preto cohort and in the joint analysis of the two cities was also present in the MSM, suggesting a causal effect. However, for the São Luís cohort no association between high soft drink consumption and gestational hypertension was detected in the MSM. It is important to note that the group exposed to high soft drink consumption in São Luís (9.1%) was smaller compared to Ribeirão Preto (30.4%) and this may help explain this difference between the two cities. In addition, the median value of the consumption in Ribeirão Preto was 3 times higher than the median value in Soft drink consumption divided into tertiles was also tested to verify the consistency of our ﬁndings. The lowest tertile of exposure was compared with the two other cate- gories (medium and highest tertiles). The highest tertile of soft drink consumption was associated with gestational hypertension in São Luís (RR=1.35; 95%CI: 1.02–1.78; P=0.032), in Ribeirão Preto (RR=1.45; 95%CI: 1.04–2.01; P=0.026), and in the two cohorts combined (RR=1.26; Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 5/9 Soft drinks and gestational hypertension Table 1. Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Discussion Descriptive characteristics of the pregnant women, comparing the São Luís and Ribeirão Preto cohorts, 2010-2011. Gestational hypertension São Luís (n, %) Ribeirão Preto (n, %) Chi-squared P-value Age (years) o0.001 14–19 169 (12.3) 215 (15.7) 20–24 449 (32.5) 387 (28.3) 25–29 424 (30.7) 417 (30.4) 30–34 243 (17.6) 221 (16.1) X35 95 (6.9) 130 (9.5) Schooling (years) 0.070 p4 24 (1.7) 57 (4.2) 5–8 146 (10.6) 337 (24.6) 9–11 1056 (76.5) 862 (62.9) X12 154 (11.2) 114 (8.3) Family income (tertiles) 0.326 1st 450 (33.6) 413 (33.4) 2nd 445 (33.2) 413 (33.4) 3rd 445 (33.2) 412 (33.2) Skin color 0.077 White 223 (16.2) 700 (51.5) Black 208 (15.1) 152 (11.2) Brown 925 (68.0) 500 (36.8) Yellow 23 (1.67) 6 (0.5) Parity (number of deliveries) 0.133 1 711 (51.5) 676 (49.3) 2–3 592 (42.9) 593 (43.3) X4 77 (5.6) 101 (7.4) Smoking habit 0.100 No 1321 (95.7) 1190 (87.1) Yes 59 (4.3) 176 (12.9) Pre-gestational BMIa o0.001 Normal weight 941 (74.5) 809 (61.2) Overweight 248 (19.6) 329 (24.9) Obesity 75 (5.9) 185 (13.9) Soft drink consumptionb 0.039 o7 times per week 1243 (90.9) 943 (69.6) X7 times per week 124 (9.1) 411 (30.4) Junk food consumption (tertiles)c 0.222 1st 654 (47.5) 519 (37.9) 2nd 419 (30.4) 463 (33.9) 3rd 306 (22.1) 385 (28.2) Gestational hypertension o0.001 No 1147 (83.1) 1178 (86.0 Yes 233 (16.9) 192 (14.0) Numbers may not add up to total because of missing values. aBody mass index: normal weight: p24.9 kg/m2, overweight 25–29.9 kg/m2, and obesity X30 kg/m2. bMultiplying times per day vs days per week and reported in times per week. cTertile of junk food consumption classiﬁcation for both São Luís and Ribeirão Preto: 1st tertile (up to three times a month), 2nd tertile (once to twice a week), and 3rd tertile (X3 times a week). Table 1. Descriptive characteristics of the pregnant women, comparing the São Luís and Ribeirão Preto cohorts, 2010-2011. Numbers may not add up to total because of missing values. aBody mass index: normal weight: p24.9 kg/m2, overweight 25–29.9 kg/m2, and obesity X30 kg/m2. bMultiplying times per day vs days per week and reported in times per week. cTertile of junk food consumption classiﬁcation for both São Luís and Ribeirão Preto: 1st tertile (up to three times a month), 2nd tertile (once to twice a week), and 3rd tertile (X3 times a week). has been shown in systematic review of studies including non-pregnant women (8,9). Hormonal and metabolic abnormalities occur naturally during gestation resulting in increased low-density lipo- protein (LDL)-cholesterol and triglyceride levels, and also in decreased high-density lipoprotein (HDL)-cholesterol (22). Discussion has been shown in systematic review of studies including non-pregnant women (8,9). Hormonal and metabolic abnormalities occur naturally during gestation resulting in increased low-density lipo- protein (LDL)-cholesterol and triglyceride levels, and also in decreased high-density lipoprotein (HDL)-cholesterol (22). São Luís, a fact that may also help explain the present ﬁndings. has been shown in systematic review of studies including non-pregnant women (8,9). g Association of high soft drink consumption and hypertension has been reported in non-pregnant women (21). Our results suggested that soft drink consumption also seemed to be a risk factor for HDP, conﬁrming what Hormonal and metabolic abnormalities occur naturally during gestation resulting in increased low-density lipo- protein (LDL)-cholesterol and triglyceride levels, and also in decreased high-density lipoprotein (HDL)-cholesterol (22). Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Soft drinks and gestational hypertension 6/9 Table 2. Marginal structural models to estimate the association of soft drink consumption with hypertension during pregnancy in the São Luís and Ribeirão Preto cohorts, 2010–2011. Table 2. Marginal structural models to estimate the association of soft drink consumption with hypertension during pregnancy in the São Luís and Ribeirão Preto cohorts, 2010–2011. g p yp g p g y São Luís and Ribeirão Preto cohorts, 2010–2011. Gestational São Luís Ribeirão Preto São Luís and Ribeirão Preto hypertension RR 95%CI P-value RR 95%CI P-value RR 95%CI P-value X7 times/week 1.26 0.79–2.00 0.328 1.63 1.21–2.19 0.001 1.51 1.15–1.97 0.039 2nd tertile 1.08 0.79–1.48 0.589 0.91 0.64–1.30 0.622 1.19 0.92–1.54 0.182 3rd tertile 1.35 1.02–1.78 0.032 1.45 1.04–2.01 0.026 1.26 1.01–1.57 0.038 Dose response 1.16 1.01–1.33 0.036 1.20 1.00–1.44 0.039 1.12 1.00–1.25 0.033 RR: risk ratio; 95%CI: conﬁdence interval. Only variables with a P-value o0.05 were considered to be statistically signiﬁcant. can increase oxidative stress during pregnancy (24) as well as gestational hypertension (34). These physiological adaptations increase oxidative stress during pregnancy that may result in a higher risk of hypertension (23). Oxidative stress in gestation may be unbalanced by excessive pregestational weight and by an unhealthy diet (high sugar/high fatty content) resulting in oxidative damage (24). Furthermore, SSBs have been consistently implicated in increased triglycerides and HDL- cholesterol levels in randomized clinical trials (25). Further- more, those beverages have been clearly associated to obesity risk (15). In the physiopathology of HDP there is an angiogenic imbalance, with involvement of oxidative stress and endothelial dysfunction (32). Discussion The mechanism of the association between high soft drink consumption and gestational hypertension shown here might be mediated by vascular changes. Supporting this hypothesis, con- sumption of added fructose has been shown to be asso- ciated with endothelial dysfunction, reduction of protective vascular factors, and increased oxidative stress, meta- bolic disorders, and high BP in an animal model (35). A possible effect of high soft drink consumption on gestational hypertension was suggested in the MSM. Previous studies have already suggested that soft drink consumption may be associated with more severe signs and symptoms of HDP. However, these results were obtained by studying associations between total energy consumption (1) or sugar consumption (10) and pre- eclampsia using conventional regression models. Our data did not distinguish whether soft drinks were sugar- or artiﬁcially-sweetened. However, it is much more probable that consumption of the caloric version predom- inated, since in Brazil only 11% of the population consume exclusively the artiﬁcially sweetened version (36). Artiﬁ- cially sweetened drinks have also been implicated in systemic metabolic changes, such as excessive weight gain, metabolic syndrome (26), hypertension (37), and preterm birth (26) through still unknown mechanisms. Thus, we cannot totally discard that other mechanisms not triggered by added sugars could be involved in the association between soft drinks consumption and hyper- tension shown here. The high sodium content of soft drinks (6) may also be a mechanism capable of explaining our results, considering that sodium retention is also characteristic of HDP (38). However, the effect of restrict- ed salt consumption on HDP has not been supported by evidence (1). Interestingly, cohort studies have also associated high SSB consumption during pregnancy with preterm birth (26). Since maternal hypertension is a recognized risk factor for preterm birth (27), these data may suggest that soft drink consumption is a common risk factor for both HDP and preterm birth. A possible mechanism to explain the association shown here between high soft drink consumption and gestational hypertension is the high sucrose content of soft drinks, which may contain added fructose from invert sugars (28). Greater consumption of added fructose has been associated with increased inﬂammatory response, dyslipidemia, insulin resistance, and increased systolic BP (29). Randomized clinical trials have shown that SSB consumption for three weeks results in increased C-reactive protein and is associated with metabolic disorders (dyslipidemia and insulin resistance) (30). Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Discussion Furthermore, intake of sugary drinks is associated with increased mean BP (31). These inﬂammatory and metabolic disorders that are linked to obesity, dyslipidemia, and insulin resistance are also present in HDP (32). Studies have shown little difference in the rate of increased weight gain between consumption of high fructose corn syrup and sucrose deriv- atives from sugarcane (33); furthermore being overweight It is possible that the association detected here between high soft drink consumption and gestational hypertension may be spurious, with soft drink consump- tion only representing an inappropriate dietary pattern. However, adjustment for junk food consumption in the MSM minimized the possibility that this association between greater soft drink consumption and gestational hypertension would only reﬂect other inappropriate eating habits. A limitation of the current study is that no adjustment was made for dietary data that may also be involved in the association between soft drink consumption and Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Soft drinks and gestational hypertension 7/9 the two cities were similar to those obtained in public maternity hospitals in Brazil (39) and in one cohort study conducted in the United Kingdom (40). gestational hypertension, especially sodium consumption and total calorie intake, information that was not available in the studies. However, these non-observed variables were included in the theoretical model and, according to the back-door criterion, adjustment for these variables to estimate the total effect of soft drink consumption on gestational hypertension would not be necessary. Another limitation is the impossibility of distinguishing between different HDP types. Thus, it was not possible to exclude the possibility that preeclampsia was the only disorder associated with high soft drink consumption. Furthermore, aggregation of data for some categories of maternal school- ing might have decreased our ability to ﬁnd differences between categories of maternal schooling according to soft drink consumption and might thus have decreased our ability to adjust for confounding. An important strength of the present study was its prospective design, in which soft drink consumption was measured before the outcome. Another positive aspect was the construction of a theoretical causal model based on the back-door criterion in a DAG, which allowed us to select an appropriate minimum set of confounders. Click to view [pdf]. Click to view [pdf]. 4. Schoenaker DAJM, Soedamah-Muthu SS, Mishra GD. The association between dietary factors and gestational hyper- tension and pre-eclampsia: a systematic review and meta- analysis of observational studies. BMC Med 2014; 12: 157, doi: 10.1186/s12916-014-0157-7. 3. Rifas-Shiman SL, Rich-Edwards JW, Kleinman KP, Oken E, Gillman MW. Dietary quality during pregnancy varies by mater- nal characteristics in Project Viva: a US cohort. J Am Diet Assoc 2009; 109: 1004–1011, doi: 10.1016/j.jada.2009.03.001. Acknowledgments The authors thank FAPEMA (Maranhão State Research and Scientiﬁc and Technological Development Foundation), FAPESP (São Paulo State Research Foundation; #2008/ 53593-0), CNPq (National Council for Scientiﬁc and Tech- nological Development), CAPES (Coordination for the Improvement of Higher Education Personnel; Finance Code 001), and FAEPA (Foundation for the Support for Education, Research and Assistance of Hospital das Clínicas of the Ribeirão Preto Medical School of the University of São Paulo). Reporting of hypertension based on medical diagnosis may also be pointed out as a potential source of infor- mation bias. However, this information was considered to be reliable since all pregnant women of the two cohorts attended more than six prenatal visits. In addition, prevalence rates of gestational hypertension reported in Discussion The use of the MSM increased our ability to make causal infer- ence regarding the association detected here between soft drink consumption and gestational hypertension, as long as the assumptions for causal interpretation had been fulﬁlled. Greater soft drink consumption seemed to be a risk factor for gestational hypertension regardless of adjust- ment for socioeconomic factors, smoking, and junk food consumption, suggesting that high consumption of these beverages should be discouraged during pregnancy. Obstetric advice should include guidance on healthy eating habits for pregnant women emphasizing the need to restrict consumption of these beverages rich in added sugars in order to avoid metabolic disorders and the development of gestational hypertension. Total intake of soft drinks during pregnancy would also be relevant but it has not been assessed. However, daily frequency of SSBs consumption has been the exposure most frequently studied in the meta-analyses that showed association between SSB and hypertension in non-preg- nant populations (8,9). As a limitation, soft drink consump- tion was assessed only once at 22/25 weeks of gestation; thus, we cannot be sure that the same frequency of this unhealthy behavior had been maintained throughout the remaining gestational period. Exposure to high soft drink intake in the beginning of gestation was consistently associated with gestational hypertension in the two birth cohorts, showing a dose-response gradient. Finally, the non-probabilistic convenience sample of our study was prone to selection bias, limiting the external generalizability of the ﬁndings; however, our results showed consistency in the association between soft drink consumption during gestation and gestational hypertension in two different Brazilian cities. 5. Clausen T, Slott M, Solvoll K, Drevon CA, Vollset SE, Henriksen T. High intake of energy, sucrose, and 3. Rifas-Shiman SL, Rich-Edwards JW, Kleinman KP, Oken E, Gillman MW. Dietary quality during pregnancy varies by mater- nal characteristics in Project Viva: a US cohort. J Am Diet Assoc 2009; 109: 1004–1011, doi: 10.1016/j.jada.2009.03.001. 4. Schoenaker DAJM, Soedamah-Muthu SS, Mishra GD. The association between dietary factors and gestational hyper- tension and pre-eclampsia: a systematic review and meta- analysis of observational studies. BMC Med 2014; 12: 157, doi: 10.1186/s12916-014-0157-7. 5. Clausen T, Slott M, Solvoll K, Drevon CA, Vollset SE, Henriksen T. High intake of energy, sucrose, and References Am J Clin Nutr 2014; 100: 65– 79, doi: 10.3945/ajcn.113.081521. 10. Roberts JM, Gammill HS. Pre-eclampsia: recent insights. Hypertension 2005; 46: 1243–1249, doi: 10.1161/01.HYP. 0000188408.49896.c5. 26. Englund-Ögge L, Brantsaeter AL, Haugen M, Sengpiel V, Khatibi A, Myhre R, et al. Association between intake of artiﬁcially sweetened and sugar-sweetened beverages and preterm delivery: a large prospective cohort study. Am J Clin Nutr 2012; 96: 552–559, doi: 10.3945/ajcn.111.031567. 11. Borgen I, Aamodt G, Harsem N, Haugen M, Meltzer HM, Brantsæter AL. Maternal sugar consumption and risk of preeclampsia in nulliparous Norwegian women. Eur J Clin Nutr 2012; 66: 920–925, doi: 10.1038/ejcn.2012.61. 27. Seely EW, Ecker J. Chronic hypertension in pregnancy. Circulation 2014; 129: 1254–1261, doi: 10.1161/CIRCU LATIONAHA.113.003904. 12. da Silva AA, Simões VM, Barbieri MA, Cardoso VC, Alves CMC, Thomaz EBAF, et al. A protocol to identify non- classical risk factors for preterm births: the Brazilian Ribeirão Preto and São Luís prenatal cohort (BRISA). Reprod Health 2014; 11: 79, doi: 10.1186/1742-4755-11-79. 28. Johnson RK, Appel LJ, Brands M, Howard BV, Lefevre M, Lustig RH, et al. Dietary sugars intake and cardiovascular health: a scientiﬁc statement from the American Heart Association. Circulation 2009; 120: 1011–1020, doi: 10.11 61/CIRCULATIONAHA.109.192627. 13. Verburg BO, Steegers EAP, De Ridder M, Snijders RJM, Smith E, Hofman A, et al. New charts for ultrasound dating of pregnancy and assessment of fetal growth: longitudinal data from a population-based cohort study. Ultrasound Obstet Gynecol 2008; 31: 388–396, doi: 10.1002/uog.5225. 29. Pollock NK, Bundy V, Kanto W, Davis CL, Bernard PJ, Zhu H, et al. Greater fructose consumption is associated with cardiometabolic risk markers and visceral adiposity in adolescents. J Nutr 2012; 142: 251–257, doi: 10.3945/ jn.111.150219. 14. Menezes CC, Ribeiro CCC, Alves CMC, Tomaz EBAF, Franco MM, Batista RFL, et al. Soft drink consumption and periodontal status in pregnant women. J Periodontol 2019; 90: 159–166, doi: 10.1002/JPER.16-0388. 30. Aeberli I, Gerber PA, Hochuli M, Kohler S, Haile SR, Gouni- Berthold I, et al. Low to moderate sugar-sweetened beverage consumption impairs glucose and lipid metabolism and promotes inﬂammation in healthy young men: a randomized controlled trial. Am J Clin Nutr 2011; 94: 479– 485, doi: 10.3945/ajcn.111.013540. 15. Malik VS, Popkin BM, Bray GA, Despres JP, Willett WC, Hu FB. Sugar-sweetened beverages and risk of metabolic syndrome and type 2 diabetes. Diabetes Care 2010; 33: 2477–2483, doi: 10.2337/dc10-1079. 16. Vereecken CA, Inchley J, Subramanian SV, Snijders RJ, Smith E, Hofman A, et al. References 1. American College of Obstetretrcians and Gynecologists, Task Forec on Hypertension in Pregnancy. Hypertension in pregnancy. Report of the American College of Obstetricians and Gynecologist’ task force on hypertension in pregnancy. Obstet Gynecol 2013; 122: 1122–1131, doi: 10.1097/01. AOG.0000437382.03963.88. 2. Ramakrishnan A, Lee LJ, Mitchell LE, Agopian AJ. Maternal hypertension during pregnancy and the risk of congenital heart defects in offspring: a systematic review and meta- analysis. Pediatr Cardiol 2015; 36: 1442–1451, doi: 10.10 07/s00246-015-1182-9. 2. Ramakrishnan A, Lee LJ, Mitchell LE, Agopian AJ. Maternal hypertension during pregnancy and the risk of congenital heart defects in offspring: a systematic review and meta- analysis. Pediatr Cardiol 2015; 36: 1442–1451, doi: 10.10 07/s00246-015-1182-9. Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Soft drinks and gestational hypertension 8/9 polyunsaturated fatty acids is associated with increased risk of preeclampsia. Am J Obstet Gynecol 2001; 185: 451–458, doi: 10.1067/mob.2001.116687. 21. Sayon-Orea C, Martinez-Gonzalez MA, Gea A, Alonso A, Pimenta AM, Bes-Rastrollo MB. Baseline consumption and changes in sugar-sweetened beverage consumption and the incidence of hypertension: The SUN Project. Clin Nutr 2015; 34: 1133–1140, doi: 10.1016/j.clnu.2014.11.010. 6. Malik VS, Popkin BM, Bray GA, Després JP, Hu FB. Sugar- sweetened beverages, obesity, type 2 diabetes mellitus, and cardiovascular disease risk. Circulationi 2010; 121: 1356– 1364, doi: 10.1161/CIRCULATIONAHA.109.876185. 22. Mankuta D, Elami-Suzin M, Elhayani A, Vinker S. Lipid proﬁle in consecutive pregnancies. Lipids Health Dis 2010; 9: 58, doi: 10.1186/1476-511X-9-58. 7. Carpenter MW. Gestational diabetes, pregnancy hyperten- sion, and late vascular disease. Diabetes Care 2007; 30: S246–S250, doi: 10.2337/dc07-s224. 23. Lima ACS, Afonso JC. A Química do Refrigerante [in Portuguese]. Quı´mica Nova Escola 2009; 31: 1–6. 8. Malik AH, Akram Y, Shetty S, Malik SS, Njike VY. Impact of sugar-sweetened beverages on blood pressure. Am J Cardiol 2014; 113: 1574–1580, doi: 10.1016/j.amjcard.2014.01.437. 24. Figueroa A, Agil A. Changes in plasma’s oxidative stress and antioxidant activity, measured with melatonin levels, and its relationship to newborns from obese and diabetic pregnancies. J Diabetes Metab 2012; S4. 9. Xi B, Huang Y, Reilly KH, Li S, Zheng R, Barrio-Lopez MT, et al. Sugar-sweetened beverages and risk of hypertension and CVD: a dose-response meta-analysis. Br J Nutr 2015; 113: 709–717, doi: 10.1017/S0007114514004383. 25. Te Morenga LA, Howatson AJ, Jones RM, Mann J. Dietary sugars and cardiometabolic risk: systematic review and meta-analyses of randomized controlled trials of the effects on blood pressure and lipids. References The relative inﬂuence of individual and contextual socio-economic status on consumption of fruit and soft drinks among adolescents in Europe. Eur J Pub Health 2005; 15: 224–232, doi: 10.1093/eurpub/cki005. 31. Grasser EK, Dulloo A, Montani JP. Cardiovascular respon- ses to the ingestion of sugary drinks using a randomised cross-over study design: Does glucose attenuate the blood pressure-elevating effect of fructose? Br J Nutr 2014; 112: 183–192, doi: 10.1017/S0007114514000622. 17. WHO (World Health Organization). Physical status: the use and interpretation of anthropometry. Geneva: WHO (Who Thechnical Report Series, No. 854. 1997. 32. Mori T, Shinohara K, Wakatsuki A, Watanabe K, Fujimaki A. Adipocytokines and endothelial function in preeclamptic women. Hypertens Res 2010; 33: 250–254, doi: 10.1038/ hr.2009.222. 18. Rosenheck R. Fast food consumption and increased caloric intake: a systematic review of a trajectory towards weight gain and obesity risk. Obes Rev 2008; 9: 535–547, doi: 10.1111/j.1467-789X.2008.00477.x. 33. Venturine Filho WG. Bebidas não alcoólicas: ciência e tecnologia. Blucher. Vol 2, 2010. 19. Schisterman EF, Cole SR, Platt RW. Overadjustment bias and unnecessary adjustment in epidemiologic studies. Epidemiology 2009; 20: 488–495, doi: 10.1097/EDE.0b013 e3181a819a1. 34. Alba JJF, Páez CM, Sánchez ÁV, Pazos ES, Macías MCG, Negro ES, et al. Sobrepeso y obesidad como factores de riesgo de los estados hipertensivos del embarazo: estudio de cohortes retrospectivo [in Spanish]. Nutr Hosp 2018; 35: 874–880, doi: 10.20960/nh.1702. ̌ 20. Cole SR, Hernán MA. Constructing inverse probability weights for MSMs. Am J Epidemiol 2008; 168: 656–664, doi: 10.1093/aje/kwn164. 35. Akar F, Uludağ O, Ayď in A, Aytekin YA, Elbeg S, Tuzcu M, et al. High-fructose corn syrup causes vascular dysfunction Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 Soft drinks and gestational hypertension 9/9 associated with metabolic disturbance in rats: protective effect of resveratrol. Food Chem Toxicol 2012; 50: 2135– 2141, doi: 10.1016/j.fct.2012.03.061. 38. Roberts JM, Balk JL, Bodnar LM, Belizán JM, Bergel E, Martinez A. Nutrient involvement in preeclampsia. J Nutr 2003; 133: 1684S-1692S, doi: 10.1093/jn/133.5.1684S. 36. IBOPE (Brazilian acronym for Brazilian Institute of Public Opinion and Statistics). Web site. http://www.ibope.com.br/ pt-br/noticias/Paginas/aspx. Published 2008. Date accessed: 23 nov. 2015. 39. Oliveira ACM, Graciliano NG. Síndrome hipertensiva da gravidez e diabetes mellitus gestacional em uma materni- dade pública de uma capital do Nordeste brasileiro, 2013: prevalência e fatores associados. Epidemiol Serv Saúde 2015; 24: 441–451. 37. Cheungpasitporn W, Thongprayoon C, Edmonds PJ, Srivali N, Ungprasert P, Kittanamongkolchai W, et al. Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162 associated with metabolic disturbance in rats: protective effect of resveratrol. Food Chem Toxicol 2012; 50: 2135– 2141, doi: 10.1016/j.fct.2012.03.061. References Sugar and artiﬁcially sweetened soda consumption linked to hyperten- sion: a systematic review and meta-analysis. Clin Exp Hypertens 2015; 37: 587–593, doi: 10.3109/10641963. 2015.1026044. 40. Staley JR, Bradley J, Silverwood RJ, Howe LD, Tilling K4, Lawlor DA, et al. Associations of blood pressure in pregnancy with offspring blood pressure trajectories during childhood and adolescence: ﬁndings from a prospective study. J Am Heart Assoc 2015; 4: 1–13, doi: 10.1161/JAHA.114.001422. Braz J Med Biol Res \| doi: 10.1590/1414-431X202010162
https://openalex.org/W2796756739	https://dspace.mit.edu/bitstream/1721.1/142544.2/3/10240_2022_Article_131.pdf	English	null	Hecke correspondences for smooth moduli spaces of sheaves	Publications mathématiques de l'IHES/Publications mathématiques	2,022	cc-by	38,081	ABSTRACT We deﬁne functors on the derived category of the moduli space M of stable sheaves on a smooth projective surface (under Assumptions A and S below), and prove that these functors satisfy certain commutation relations. These relations allow us to prove that the given functors induce an action of the elliptic Hall algebra on the K-theory of the moduli space M, thus generalizing the action studied by Nakajima, Grojnowski and Baranovsky in cohomology. Lunei, pentru cinci luni minunate Lunei, pentru cinci luni minunate HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES by ANDREI NEGUT, © IHES and Springer-Verlag GmbH Germany, part of Springer Nature 2022 https://doi.org/10.1007/s10240-022-00131-1 1. Introduction 1.1. Let S be a smooth projective surface over an algebraically closed ﬁeld of characteristic 0 (henceforth denoted by C). An important family of moduli spaces asso- ciated to S are the Hilbert schemes of points on S, which we will denote by Hilbn(S): these are smooth 2n-dimensional algebraic varieties, which parameterize colength n ideal sheaves I ⊂OS. The Betti numbers of Hilbert schemes were computed in [10, 16], and this lead to a very interesting observation: the generating series of the Betti numbers of Hilbn(S), as n goes from 0 to ∞, matches the Poincaré polynomial of (an appropriately graded version of) the Fock space of the inﬁnite-dimensional Heisenberg algebra. Cou- pled with expectations from mathematical physics, the way to understand this interesting connection is to consider all Hilbert schemes together: Hilb(S) = ∞ n=0 Hilbn(S) and construct an action and construct an action (1.1) Heisenberg algebra ↷H∗(Hilb(S)) = ∞ n=0 H∗(Hilbn(S)) (1.1) This was achieved by Grojnowski in [17] and Nakajima in [23]. We will mostly use the formulation of loc. cit., in which the generators {a±n}n∈N of the Heisenberg algebra act on H∗(Hilb(S)) by the so-called Hecke correspondences This was achieved by Grojnowski in [17] and Nakajima in [23]. We will mostly use the formulation of loc. cit., in which the generators {a±n}n∈N of the Heisenberg algebra act on H∗(Hilb(S)) by the so-called Hecke correspondences (1.2) C• n = (I′ ⊂nx I) (1.2) © IHES and Springer-Verlag GmbH Germany, part of Springer Nature 2022 https://doi.org/10.1007/s10240-022-00131-1 © IHES and Springer-Verlag GmbH Germany, part of Springer Nature 2022 https://doi.org/10.1007/s10240-022-00131-1 A. NEGUT, where I′ ⊂nx I means that the ideal sheaves I,I′ ⊂OS are contained inside each other, and their quotient is a length n sheaf supported at a single (but arbitrary) closed point x ∈S. Nakajima considered the projection maps where I′ ⊂nx I means that the ideal sheaves I,I′ ⊂OS are contained inside each other, and their quotient is a length n sheaf supported at a single (but arbitrary) closed point x ∈S. 1. Introduction Nakajima considered the projection maps C• n p+ pS p− Hilb(S) S Hilb(S) I′ ⊂nx I I′ x I eﬁned I′ ⊂nx I I′ x I C• n p+ pS p− Hilb(S) S Hilb(S) I′ ⊂nx I I′ x I C• n p+ pS p− Hilb(S) S Hilb(S) I′ ⊂nx I I′ x I and deﬁned (1.3) a±n = (p∓× pS)∗◦p∗ ± : H∗(Hilb(S)) →H∗(Hilb(S)) ⊗H∗(S) I and deﬁned and deﬁned a±n = (p∓× pS)∗◦p∗ ± : H∗(Hilb(S)) →H∗(Hilb(S)) ⊗H∗(S) a±n = (p∓× pS)∗◦p∗ ± : H∗(Hilb(S)) →H∗(Hilb(S)) ⊗H∗(S) a±n = (p∓× pS)∗◦p∗ ± : H∗(Hilb(S)) →H∗(Hilb(S)) ⊗H∗(S) (1.3) The main result of [23] is that the operators above satisfy the deﬁning relations of the Heisenberg algebra, with H∗(S) as a “parameter space”. Under this action, H∗(Hilb(S)) is isomorphic to the Fock space of the Heisenberg algebra, which explains in a represen- tation theoretic way the classical formulas for the Betti numbers. The main result of [23] is that the operators above satisfy the deﬁning relations of the Heisenberg algebra, with H∗(S) as a “parameter space”. Under this action, H∗(Hilb(S)) is isomorphic to the Fock space of the Heisenberg algebra, which explains in a represen- tation theoretic way the classical formulas for the Betti numbers. 1.2. After constructing the action (1.1), one might wonder how to generalize it. One direction would be to replace singular cohomology by other homology theories. While the construction extends almost verbatim to Chow groups, in algebraic K-theory things are not so simple. The reason for this is that the correspondences C• n are very badly behaved for n > 1 (one does not even know if they are equidimensional in general) so the initial question is to ﬁnd suitable replacements. Problem: deﬁne analogues of the operators (1.3) in K-theory, which satisfy the deﬁning relations of the deformed Heisenberg satisfy the deﬁning relations of the deformed Heisenberg algebra One approach toward this problem was undertaken in [11, 31] for S = A2, where the authors used equivariant localization computations to construct an action (1.4) A ↷KHilb(S) = ∞ n=0 KHilbn(S) (1.4) (see Section 3.3 for our conventions on K-theory). In the elliptic Hall algebra formula- tion of [31], the algebra A is generated by elements {e±n,k}(n,k)∈N×Z, with the subalgebra generated by the e±n,0’s isomorphic to the deformed Heisenberg algebra, thus explaining our interest in constructing an action (1.4). In [25], we realized the operators e±n,k using certain K-theory classes on the correspondence (see Section 3.3 for our conventions on K-theory). In the elliptic Hall algebra formula- tion of [31], the algebra A is generated by elements {e±n,k}(n,k)∈N×Z, with the subalgebra generated by the e±n,0’s isomorphic to the deformed Heisenberg algebra, thus explaining our interest in constructing an action (1.4). (1.3) In [25], we realized the operators e±n,k using certain K-theory classes on the correspondence (1.5) Z• n = (I0 ⊂x I1 ⊂x ··· ⊂x In) (1.5) HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES which can thus be thought of as a resolution of C• n of (1.2). While loc. cit. only proved results for S = A2, the correspondence (1.5) is well-deﬁned for any smooth surface S, and we are thus poised to solve the Problem stated above. Even more so, the operators e±n,k naturally lift to the derived category, and are thus related to various categoriﬁcations of Hecke algebras (see [15] for an overview). 1.3. Another direction for generalizing (1.1) is to replace Hilbert schemes by other moduli spaces of coherent sheaves on S. An important example is given by moduli spaces of stable sheaves (see Section 2.8 for more details), in which the analogue of the action (1.1) was worked out in [2]. The main purpose of the present paper is to construct an action (1.4) in the aforementioned setup of stable sheaves. Thus, let S be a smooth pro- jective surface with an ample divisor H, and also ﬁx (r,c1) ∈N × H2(S,Z). Consider the moduli space M of H-stable sheaves on the surface S with the numerical invariants r,c1 and any c2 ∈Z. We make the following two assumptions throughout the present paper: Assumption A: gcd(r,c1 · H) = 1 (1.6) Assumption S: either ωS ∼= OS, or c1(ωS) · H < 0 (1.7) Assumption A: gcd(r,c1 · H) = 1 Assumption S: either ωS ∼= OS, or c1(ωS) · H < 0 (1.6) Assumption A implies that M is representable, i.e. there exists a universal sheaf U on M × S. Assumption S implies that M is smooth, which allows us to deﬁne DM = Db(Coh(M)) = Perf(M) DM = Db(Coh(M)) = Perf(M) In the following formulas, we will denote by F the coherent sheaves on S that are pa- rameterized by the moduli space M. With this in mind, let us recall the following moduli spaces of ﬂags of sheaves that we studied in [27] and [28], respectively: Z1 = (F0 ⊂x F1) Z• 2 = (F0 ⊂x F1 ⊂x F2) where F ′ ⊂x F means that F ′ ⊂F and F/F ′ is the length 1 skyscraper sheaf at the closed point x ∈S. We recall the scheme structures of Z1 and Z• 2 in Section 2, and in particular, the fact that they are both smooth. This implies that the maps Z1 p+ pS p− M S M F0 ⊂x F1 F0 x F1 Z1 p+ pS p− M S M F0 ⊂x F1 F0 x F1 Z1 p+ pS p− M S M F1 A. NE Z• 2 π+ π− Z1 Z1 A. NEGUT, Z• 2 π+ π− Z1 Z1 F0 ⊂x F1 ⊂x F2 F0 ⊂x F1 F F0 ⊂x F1 F1 ⊂x F2 Z1 Z1 induce direct and inverse image functors between the derived categories DM, DZ1, DZ• 2. We may combine these spaces into more complicated diagrams of the form induce direct and inverse image functors between the derived categories DM, DZ1, DZ• 2. We may combine these spaces into more complicated diagrams of the form Z• 2 π+ π− ... π+ π− Z• 2 π+ π− Z1 p+×pS Z1 Z1 Z1 p− M × S M Z• 2 π+ π− ... π+ π− Z• 2 π+ π− Z1 p+×pS Z1 Z1 Z1 p− M × S M M where the number of copies of Z1 contained in the middle row is denoted by n. Then for an arbitrary sequence d1,...,dn ∈Z, we consider the functor (1.8) e(d1,...,dn) : DM →DM×S e(d1,...,dn) : DM →DM×S (1.8) given by the composition given by the composition given by the composition given by the composition DZ1 (p+×pS)∗ DZ1 ⊗Ld1 DZ1 π+∗π∗ − DZ1 ... 1 More precisely, consider the universal sheaves U0 ⊂U1 on Z1 × S; then L is deﬁned as the push-forward of the quotient U1/U0 to Z1 via the ﬁrst projection. (1.6) DZ1 ⊗Ld2 DZ1 π+∗π∗ − DZ1 ⊗Ldn DM×S DM p∗ − where L is the line bundle on Z1 with ﬁbers (S,F1/F0).1 Note thated :=e(d) are the derived category versions of the operators e1,d of Section 1.2, but it is crucial to us that e(d1,...,dn) is different from the compositioned1 ◦···◦edn (see Propositions 2.27 and 3.4). With this in mind, our main result is the following. where L is the line bundle on Z1 with ﬁbers (S,F1/F0).1 Note thated :=e(d) are the derived category versions of the operators e1,d of Section 1.2, but it is crucial to us that e(d1,...,dn) is different from the compositioned1 ◦···◦edn (see Propositions 2.27 and 3.4). With this in mind, our main result is the following. Theorem 1.4. — For any n ∈N and any d1,...,dn,k ∈Z, consider the functors Theorem 1.4. — For any n ∈N and any d1,...,dn,k ∈Z, consider the functors (1.9) e(d1,...,dn) ◦ek and ek ◦e(d1,...,dn) : DM →DM×S×S of (3.9), (3.10). There are explicit functors g0,...,gn : DM →DM×S×S such that (1.9) of (3.9), (3.10). There are explicit functors g0,...,gn : DM →DM×S×S such that 1 More precisely, consider the universal sheaves U0 ⊂U1 on Z1 × S; then L is deﬁned as the push-forward of the quotient U1/U0 to Z1 via the ﬁrst projection. 1 More precisely, consider the universal sheaves U0 ⊂U1 on Z1 × S; then L is deﬁned as the push-forward of the quotient U1/U0 to Z1 via the ﬁrst projection. (1.11) Items (1)–(3) in Theorem 1.4 imply that these maps satisfy the following relations: Items (1)–(3) in Theorem 1.4 imply that these maps satisfy the following relations: 12) [e(d1,...,dn),ek] = ∗◦ n i=1 ⎧ ⎨ ⎩ − k≤a<di e(d1,...,di−1,a,di+k−a,di+1,...,dn) if di > k di≤a<k e(d1,...,di−1,a,di+k−a,di+1,...,dn) if di < k HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES (1) g0 =e(d1,...,dn) ◦ek and gn =ek ◦e(d1,...,dn); (2) for all i ∈{1,...,n}, there exist explicit natural transformations (1) g0 =e(d1,...,dn) ◦ek and gn =ek ◦e(d1,...,dn); ( ) g ( 1, , n) g ( 1, , n) (2) for all i ∈{1,...,n}, there exist explicit natural transformations (1.10) ⎧ ⎪⎨ ⎪⎩ gi−1 →gi if di > k gi−1 ←gi if di < k gi−1 ∼= gi if di = k (1.10) ⎧ ⎪⎨ ⎪⎩ gi−1 →gi if di > k gi−1 ←gi if di < k gi−1 ∼= gi if di = k (1.10) ⎧ ⎪⎨ ⎪⎩ gi−1 →gi if di > k gi−1 ←gi if di < k gi−1 ∼= gi if di = k (1.10) (3) for all i ∈{1,...,n}, the cone of the natural transformation in the previous item has a ﬁltration with the associated graded object given by the functor (3) for all i ∈{1,...,n}, the cone of the natural transformation in the previous item has a ﬁltration with the associated graded object given by the functor ⎧ ⎪⎪⎪⎪⎨ ⎪⎪⎪⎪⎩ di−1 a=k ∗◦e(d1,...,di−1,a,di+k−a,di+1,...,dn) if di > k k−1 a=di ∗◦e(d1,...,di−1,a,di+k−a,di+1,...,dn) if di < k Above, we abuse notation by writing : M × S →M × S × S for the identity on M times the diagonal embedding S →S × S. Above, we abuse notation by writing : M × S →M × S × S for the identity on M times the diagonal embedding S →S × S. 1.5. To gain more insight on the meaning of the categorical relations in (1)–(3) above, let us consider the maps induced by (1.8) at the level of K-theory groups: (1.11) e(d1,...,dn) : KM →KM×S Conjecture 1.8. — For any n,m ∈N, we have Conjecture 1.8. — For any n,m ∈N, we have e(0,...,0) n zeroes ◦e(0,...,0) m zeroes ∼=e(0,...,0) m zeroes ◦e(0,...,0) n zeroes e(0,...,0) ◦e(0,...,0) ∼=e(0,...,0) ◦e(0,...,0) as functors DM →DM×S×S. (1.12) — For any n,m ∈N, we have e(0,...,0) n zeroes ◦e(0,...,0) m zeroes ∼=e(0,...,0) m zeroes ◦e(0,...,0) n zeroes as functors DM →DM×S×S. Conjecture 1.8. — For any n,m ∈N, we have e(0,...,0) n zeroes ◦e(0,...,0) m zeroes ∼=e(0,...,0) m zeroes ◦e(0,...,0) n zeroes s functors DM →DM×S×S. (1.12) We show in Section 4 that these relations are sufﬁcient to conclude that the maps (1.11) induce an action of the elliptic Hall algebra A (see [4, 30], and Section 4 for a review) on the K-theory groups of the moduli space M. We thus conclude: We show in Section 4 that these relations are sufﬁcient to conclude that the maps (1.11) induce an action of the elliptic Hall algebra A (see [4, 30], and Section 4 for a review) on the K-theory groups of the moduli space M. We thus conclude: Corollary 1.6 (See Theorem 4.15 for details). — There exists an action A ↷KM, in the sense of Deﬁnition 4.14. This notion of action is deﬁned so that composing maps KM →KM×S treats the second factor of S as a “parameter space”. We prove the Corollary only upon tensoring KM with Q, but we expect it remains true over Z. The parameters q1,q2 of the elliptic Hall algebra A are such that q1 + q2 = [1 S] q = q1q2 = [ωS] q1 + q2 = [1 S] q = q1q2 = [ωS] as elements of KS. Therefore, KS plays the role of ground ring of the algebra A. elements of KS. Therefore, KS plays the role of ground ring of the algebra A. A. NEGUT, 1.7. As a consequence of Corollary 1.6 and the deﬁning relations in A, the maps 1.7. As a consequence of Corollary 1.6 and the deﬁning relations in A, the maps e(0,...,0) n zeroes : KM →KM×S e(0,...,0) : KM →KM×S n zeroes satisfy the relations of the deformed Heisenberg algebra. This solves the Problem in Sec- tion 1.2 in the context of moduli spaces of stable sheaves (strictly speaking, the operators above only give half of the Heisenberg algebra, with the other half provided by the trans- posed correspondences). It is natural to propose the following. satisfy the relations of the deformed Heisenberg algebra. This solves the Problem in Sec- tion 1.2 in the context of moduli spaces of stable sheaves (strictly speaking, the operators above only give half of the Heisenberg algebra, with the other half provided by the trans- posed correspondences). It is natural to propose the following. Conjecture 1.8. as functors DM →DM×S×S. Theorem 1.4 proves the m = 1 case of Conjecture 1.8, for any n. It would be inter- esting to compare Conjecture 1.8 with other categoriﬁcations of Heisenberg algebras in the literature. For example, such a categoriﬁcation was proposed in [6] when the surface is of ADE type (generalized to all surfaces in [19]), but using the language of derived categories of symmetric powers of the surface S. It is thus not easy, but would be very interesting, to compare our results with those of loc. cit. One can generalize Conjecture 1.8 by recalling that the generators e−n,k ∈A satisf (1.13) e−n,k = qgcd(n,k)−1e(d1,...,dn), where di = ki n − k(i −1) n + δn i −δ1 i (1.13) 6 implies the following formula, for all (n,k),(n′,k′) ∈N × Z: Corollary 1.6 implies the following formula, for all (n,k),(n′,k′) ∈N × Z: [e−n,k,e−n′,k′] = ∗◦(a certain linear combination of products of e−n′′,k′′’s) where the coefﬁcients of the linear combination match the analogous structure constants in the algebra A (see (4.6) and (4.8)). We expect such formulas to categorify. Conjecture 1.9. — Suppose (n,k),(n′,k′) ∈N × Z are such that kn′ −k′n = 1. Then there exists a natural transformation between the functors DM e−n′,k′◦e−n,k −−−−−→DM×S×S and DM e−n,k◦e−n′,k′ −−−−−→DM×S×S e(d1,...,dn) with the di’s as in (1.13)) whose cone is ∗◦e−n−n′,k+k′. (wheree−n,k =e(d1,...,dn) with the di’s as in (1.13)) whose cone is ∗◦e−n−n′,k+k′. (wheree−n,k =e(d1,...,dn) with the di’s as in (1.13)) whose cone is ∗◦e−n−n′,k+k′. 1.10. Working withe(d1,...,dn) : DM →DM×S instead of e(d1,...,dn) : KM →KM×S reveals new features, such as the presence of categoriﬁed knot invariants (namely Kho- vanov homology). Indeed, following the principles laid out in [15], the natural transfor- mations that appear in Theorem 1.4 can be interpreted as yielding skein exact triangles HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES between certain objects in the category C of afﬁne Soergel bimodules (in [14], we make this precise by realizing Proposition 2.39 in C ). The construction of the functors (1.8) was given in terms of derived schemes in [28], and Corollary 1.6 was conjectured therein. In the present paper, we show that Assumption S implies that many of the spaces featured in loc. cit. are local complete in- tersections. Therefore, one can work in classical algebraic geometry rather than derived algebraic geometry. (1.14) is injective, where A∗ taut denotes the subring of the Chow ring of the Hilbert scheme of arbitrarily many points on S that is generated by the tautological classes. This is a partial version of the Beauville-Voisin conjecture [34] for hyperkähler manifolds. is injective, where A∗ taut denotes the subring of the Chow ring of the Hilbert scheme of arbitrarily many points on S that is generated by the tautological classes. This is a partial version of the Beauville-Voisin conjecture [34] for hyperkähler manifolds. 1.11. The structure of the present paper is the following. In Section 2 we recall the moduli space of stable sheaves M, introduce the moduli spaces of ﬂags of sheaves Zλ and several versions of the moduli space of quadruples Y, and state a number of geometric properties of Zλ and Y (such as dimension, irreducibility, smoothness, l.c.i.- ness, Cohen-Macaulay-ness and normality). In Section 3, we use these properties to set up and prove Theorem 1.4. In Section 4, we go from derived categories to K-theory and prove Corollary 1.6. Finally, in Sections 5 and 6 we prove all the aforementioned geometric properties: dimension estimates and irreducibility in Section 5 and all other properties in Section 6. as functors DM →DM×S×S. By combining our Corollary 1.6 with the results of [29], we obtain the identiﬁcation of the Carlsson-Okounkov operator KM →KM with a W-algebra in- tertwiner (see loc. cit. for deﬁnitions). When the surface is S = A2 (and K-theory is under- stood equivariantly), this is enough to prove a mathematical incarnation of the deformed AGT correspondence for U(r) gauge theory with matter, see [26] (inspired by [21, 32] and the physics literature). We observe that Theorem 1.4 also holds when the moduli space of stable sheaves is replaced by the Hilbert scheme of points on an arbitrary smooth quasiprojective surface (we do not need the surface to be projective, as all the push-forward maps in Section 1.3 are proper). The modiﬁcations required to make the argument work are minimal, and we leave the details to the interested reader. When the surface S is K3, we revisit some of the methods in the present paper in [20] in order to prove that the cycle class map (1.14) A∗ taut →H∗(Hilb(K3)) A∗ taut →H∗(Hilb(K3)) A∗ taut →H∗(Hilb(K3)) A∗ taut →H∗(Hilb(K3)) 2. The moduli space of sheaves 2.1. All our schemes will be projective over an algebraically closed ﬁeld of char- acteristic 0, henceforth denoted by C. A scheme will be called smooth if all of its local rings are regular. A closed embedding Z →X is called regular if the ideal of Z in X is A. NEGUT, locally generated by a regular sequence. In the present paper, we will often encounter closed embeddings which arise as the zero loci of sections locally generated by a regular sequence. In the present paper, we will often encounter closed embeddings which arise as the zero loci of sections σ : OX −→W where W is a locally free sheaf on the scheme X. We will often abuse terminology and refer to either σ or σ ∨: W∨→OX as “the section”, with the latter having the advantage that the zero locus is deﬁned as the scheme (2.1) Z(σ) = SpecX OX Im σ ∨ (2.1) If X is a Cohen-Macaulay scheme, we have the well-known inequality If X is a Cohen-Macaulay scheme, we have the well-known inequality (2.2) dimZ(σ) ≥dimX −rank W (2.2) Since all schemes in our paper will be Cohen-Macaulay, we give the followin e all schemes in our paper will be Cohen-Macaulay, we give the following. Deﬁnition 2.2. — The section σ is called regular if equality holds in (2.2). Deﬁnition 2.5. — With the notation as above, assume that there exists a subsheaf L ⊂Ker η∗(σ ∨) such that both L and η∗(W∨) L are locally free. In this case, if the induced section such that both L and η∗(W∨) L are locally free. In this case, if the induced section (2.6) η∗(W∨) L σ ′∨ −→OX′ (2.6) is regular, then we call (2.3) a derived ﬁber square with excess. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Deﬁnition 2.4. — If the sections σ and η∗(σ) are both regular, then we call (2.3) a derived ﬁber square. We will also use this terminology when the horizontal arrows in (2.3) are embeddings of zero loci followed by smooth maps. Deﬁnition 2.4. — If the sections σ and η∗(σ) are both regular, then we call (2.3) a derived ﬁber square. We will also use this terminology when the horizontal arrows in (2.3) are embeddings of zero loci followed by smooth maps. If σ is a regular section and X, X′ are Cohen-Macaulay schemes, then (2.2) implies that η∗(σ) is regular if and only if (2.5) dimZ(η∗(σ)) −dimX′ = dimZ(σ) −dimX Deﬁnition 2.2. — The section σ is called regular if equality holds in (2.2). Indeed, the usual deﬁnition of regularity (that the coordinates of σ in any local trivialization of W form a regular sequence) is equivalent to that of Deﬁnition 2.2 over Cohen-Macaulay schemes. In this case, we have a quasi-isomorphism OZ(σ) = OX Im σ ∨ q.i.s. ∼= ··· −→∧2W∨−→W∨σ ∨ −→OX (the maps in the complex above are given by contraction with σ ∨) and we call Z(σ) →X a complete intersection. 2.3. Given a map of schemes η : X′ →X, we may form the ﬁber square 2.3. Given a map of schemes η : X′ →X, we may form the ﬁber square 2.3. Given a map of schemes η : X′ →X, we may form the ﬁber square (2.3) Z(η∗(σ)) X′ η Z(σ) X (2.3) (2.3) where where where (2.4) η∗(W∨) η∗(σ ∨) −−−→OX′ (2.4) η∗(W∨) η∗(σ ∨) −−−→OX′ (2.4) η∗(W∨) η∗(σ ∨) −−−→OX′ (2.4) is the pull-back of σ ∨. However, we note the very important fact that is the pull-back of σ ∨. However, we note the very important fact that σ regular does not imply η∗(σ) regular is the pull-back of σ ∨. However, we note the very important fact that σ regular does not imply η∗(σ) regular σ regular does not imply η∗(σ) regular in general, because regular sequences are not preserved under pull-back. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES is regular, then we call (2.3) a derived ﬁber square with excess. In the setting of Deﬁnition 2.5, the locally free sheaf L is called the excess bun- dle, and its rank is precisely the difference between the two sides of (2.5). 2.6. Given a locally free sheaf V on a scheme X, consider the projective bundle PX(V) = Proj symmetric algebra of V PX(V) = Proj symmetric algebra of V Projective bundles are among the easiest examples of moduli spaces in algebraic geom- etry, in the sense that they represent the functor of line bundle quotients of the vector bundle V. More speciﬁcally, this means that there is a natural identiﬁcation Maps(T,PX(V)) 1-to-1 ←−→ (2.7) Maps(T,PX(V)) 1-to-1 ←−→ (2.7) Maps(T,PX(V)) 1-to-1 ←−→ (2.7) T φ−→X, line bundle L on T, surjection φ∗(V) ↠L (2.7) T φ−→X, line bundle L on T, surjection φ∗(V) ↠L Assume that we have a map W →V of locally free sheaves on X. In the present paper, we will encounter local complete intersection morphisms of the form (2.8) Z(σ) ι→PX(V) ρ↠X (2.8) where the map ι is cut out by the section (2.9) σ : ρ∗(W) −→ρ∗(V) taut −→O(1) (2.9) and the map denoted by taut is the tautological morphism on PX(V). and the map denoted by taut is the tautological morphism on PX(V). A. NEGUT, Remark 2.7. — We note a convenient abuse of terminology in calling (2.9) a “sec- tion”, as it will appear again throughout the present paper. Given a line bundle L and a vector bundle E on a scheme X, whenever we refer to a map L σ−→E as a section, we are implicitly referring to the induced section O →E ⊗L−1. The same terminology will apply to the dual map E∨→L−1 in relation to E∨⊗L →O. as a section, we are implicitly referring to the induced section O →E ⊗L−1. The same terminology will apply to the dual map E∨→L−1 in relation to E∨⊗L →O. Since the map ρ in (2.8) is smooth, the considerations of the preceding subsection apply to the composed map Z(σ) →X. More speciﬁcally, given any map η : X′ →X, the ﬁber square analogous to (2.3) is derived iff σ and η∗(σ) are regular. 2.8. Consider a smooth projective surface S and an ample divisor H ⊂S. is regular, then we call (2.3) a derived ﬁber square with excess. The Hilbert polynomial of a coherent sheaf F on S is given by PF(n) := χ(S,F ⊗O(nH)) = an2 + bn + c where a,b,c are rational numbers that one can compute from the Hirzebruch-Riemann- Roch theorem. One can ﬁnd formulas for these numbers in the Appendix to [27], but the only thing we will need in the present paper is that they can be expressed in terms of S,H and the rank r and Chern classes c1,c2 of F. If F is torsion free, then a > 0, and one deﬁnes the reduced Hilbert polynomial as where a,b,c are rational numbers that one can compute from the Hirzebruch-Riemann- Roch theorem. One can ﬁnd formulas for these numbers in the Appendix to [27], but the only thing we will need in the present paper is that they can be expressed in terms of S,H and the rank r and Chern classes c1,c2 of F. If F is torsion free, then a > 0, and one deﬁnes the reduced Hilbert polynomial as pF(n) = PF(n) a A torsion-free coherent sheaf F on S is called (Gieseker H-) stable if for all proper subsheaves G ⊂F, we have the inequality A torsion-free coherent sheaf F on S is called (Gieseker H-) stable if for all proper subsheaves G ⊂F, we have the inequality pG(n) < pF(n) whenever n ≫0. Since the reduced Hilbert polynomials are monic and quadratic, stabil- ity is determined by checking certain inequalities for the linear term and constant term coefﬁcients. Note that stability depends on the ample divisor H, but we will ﬁx a choice throughout this paper. Deﬁnition 2.9 (See [18]). — Let M(r,c1,c2) denote the quasiprojective variety which corepresents the moduli functor of stable sheaves on S with the invariants r,c1,c2. It is straightforward to compute the tangent spaces to the variety M(r,c1,c2), and the interested reader can refer to Section 6.10 in the Appendix. A simple consequence of this computation is the following well-known fact. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Proposition 2.10. — Under Assumption S, M(r,c1,c2) is smooth of dimension (2.10) const + 2rc2 where const only depends on S,H,r,c1 (the speciﬁc formula can be found in [27]). Proposition 2.10. — Under Assumption S, M(r,c1,c2) is smooth of dimension (2.10) const + 2rc2 where const only depends on S,H,r,c1 (the speciﬁc formula can be found in [27]). 2.11. Recall Assumption A of (1.6), which states that gcd(r,c1 · H) = 1. An im- portant consequence of this assumption is the existence of a universal sheaf U(r,c1,c2) M(r,c1,c2) × S (2.11) U(r,c1,c2) M(r,c1,c2) × S (2.11) U(r,c1,c2) M(r,c1,c2) × S (2.14) with W and V locally free sheaves on M × S. In fact, we can take (2.15) V = π ∗ 1 π1∗ U ⊗π ∗ 2 (O(nH)) ⊗π ∗ 2 (O(−nH)) (2.15) for n ≫0, where π1 : M × S →M and π2 : M × S →S are the standard projections. 2.15. A set partition is an equivalence relation on a ﬁnite ordered set. We will represent set partitions symbolically, for example (x,y,z) will refer to the partition of a 3-element set into distinct 1-element subsets, while (x,y,x) (respectively (x,x,x)) refers to the equivalence relation which sets the ﬁrst and the last element (respectively all elements) equivalent to each other. The size of a set partition λ, which is denoted by \|λ\|, is the number of elements of the underlying set. Proposition 2.14 ([27]). — There exists a short exact sequence (2.14) 0 →W →V →U →0 A. NEGUT, We will write U for the universal sheaf on M × S obtained as the disjoint union of the universal sheaves (2.11) over all c2; we assume that the latter are compatible with each other as c2 varies, as explained in [27, Section 5.9]. This implies that the moduli spaces of ﬂags of sheaves that we will introduce in the next subsection also have universal sheaves which are contained inside each other in the obvious way. Because the universal sheaf U is ﬂat over M, it inherits certain properties from the stable sheaves it parameterizes, such as having homological dimension 1. Indeed, any stable sheaf of rank r > 0 is torsion free, and any torsion free sheaf on a smooth projective surface has homological dimension 1, see [18, Example 1.1.16]. The fact that this also holds in families is made explicit by the following result. Proposition 2.14 ([27]). — There exists a short exact sequence (2.11) U(r,c1,c2) which is ﬂat over M(r,c1,c2), and has the universal property (2.13). In what follows, given a coherent sheaf F on a scheme T × S, we will denote by Ft its ﬁber over {t} × S, for any closed point t ∈T. Then we have the following fact (see [18]). which is ﬂat over M(r,c1,c2), and has the universal property (2.13). In what follows, given a coherent sheaf F on a scheme T × S, we will denote by Ft its ﬁber over {t} × S, for any closed point t ∈T. Then we have the following fact (see [18]). Proposition 2.12. — Under Assumption A, M(r,c1,c2) is a projective variety, which represents the moduli functor of stable sheaves on S with the invariants r,c1,c2. The functorial description of M(r,c1,c2) entails the existence of natural bijections Maps(T,M(r,c1,c2)) 1-to-1 ←−→ coherent sheaves F on T × S, ﬂat over T, (2.12) such that Ft is stable with invariants (r,c1,c2),∀t ∈T ∼ such that Ft is stable with invariants (r,c1,c2),∀t ∈T ∼ for all schemes T. Above, we write F ∼F ′ if F ′ = F ⊗π ∗ 1 (L) for some line bundle L on T, and π1 : T × S →T denotes the standard projection. The word “represents” in Proposition 2.12 means that the 1-to-1 correspondence (2.12) is given by (2.13) T φ−→M(r,c1,c2) ⇝ F = (φ × IdS)∗(U(r,c1,c2)) (2.13) Because of the equivalence relation ∼in (2.12), the universal sheaf U(r,c1,c2) is not unique, but may be tensored with any line bundle pulled back from M(r,c1,c2). Because of the equivalence relation ∼in (2.12), the universal sheaf U(r,c1,c2) is not unique, but may be tensored with any line bundle pulled back from M(r,c1,c2). 2.13. From now on, we will ﬁx (r,c1) ∈N × H2(S,Z) and set 2.13. From now on, we will ﬁx (r,c1) ∈N × H2(S,Z) and set M = ∞ c2=−∞ M(r,c1,c2) M = ∞ c2=−∞ M(r,c1,c2) Note that the Bogomolov inequality implies that c2 is bounded below by r−1 2r c2 1. Note that the Bogomolov inequality implies that c2 is bounded below by r−1 2r c2 1. A. NEGUT, are stable with invariants (r,c1),∀t ∈T, are stable with invariants (r,c1),∀t ∈T, together with the data in (a) and (b) below ∼ where ∼is the equivalence relation induced by tensoring F0,...,Fn by one and the same line bundle pulled back from T. In the formula above, (a) and (b) refers to where ∼is the equivalence relation induced by tensoring F0,...,Fn by one and the same line bundle pulled back from T. In the formula above, (a) and (b) refers to (a) maps x1,...,xn : T →S such that xi = xj if i ∼j in λ, (b) line bundles L1,...,Ln on T such that Fi/Fi−1 ∼= i ∗(Li), where i : T → T × S denotes the graph of the map xi : T →S from (a) (a) maps x1,...,xn : T →S such that xi = xj if i ∼j in λ, (b) line bundles L1,...,Ln on T such that Fi/Fi−1 ∼= i ∗(Li), where i : T → T × S denotes the graph of the map xi : T →S from (a) Proposition 2.19 will establish (by induction on n) that the functor (2.17) is indeed representable, which implies the existence of the scheme stipulated in Deﬁnition 2.16. However, to keep our notation simple, we will denote points of this scheme as (2.16) instead of the more complicated notation (2.17). In other words, all the constructions from now on will be done at the level of closed points, and the family versions are left as exercises to the interested reader. 2.17. If λ is a partition on an n element ordered set, then we will use the notation \|λ and λ\| \|λ and λ\| \|λ and λ\| for the partition on the n −1 element ordered set obtained by dropping the ﬁrst (respec- tively last) element of λ. Then we may consider the maps for the partition on the n −1 element ordered set obtained by dropping the ﬁrst (respec- tively last) element of λ. Then we may consider the maps (2.18) Zλ π− Z\|λ × S# and Zλ π+ Zλ\| × S# (2.18) Zλ π− Z\|λ × S# and Zλ π+ Zλ\| × S# (2.18) given by forgetting the ﬁrst (respectively last) sheaf in the ﬂag (2.16). The number # is 0 or 1 depending on whether the ﬁrst (respectively last) element of λ is or is not equivalent to some other element of λ. Therefore, the map π−(resp. Deﬁnition 2.16. — For a set partition λ of size n, we will consider the scheme Deﬁnition 2.16. — For a set partition λ of size n, we will consider the scheme Zλ = (F0 ⊂x1 F1 ⊂x2 ··· ⊂xn Fn) stable coherent sheaves (2.16) (2.16) for some x1,...,xn ∈S such that xi = xj if i ∼j in λ where F ′ ⊂x F means that F ′ ⊂F and F/F ′ ∼= Cx. If n = 0, we write Z∅= M where F ′ ⊂x F means that F ′ ⊂F and F/F ′ ∼= Cx. If n = 0, we write Z∅= M. where F ′ ⊂x F means that F ′ ⊂F and F/F ′ ∼= Cx. If n = 0, we write Z∅= M. We will often write Zn = Z(x1,...,xn) and Z• n = Z(x,...,x) Deﬁnition 2.16 should be read as “the scheme Zλ represents the functor” Maps(T,Zλ) 1-to-1 ←−→ ﬂags of coherent sheaves (F0 ⊂F1 ⊂··· ⊂ (2.17) on T × S, ﬂat over T, such that F0,t,...,Fn,t HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES are stable with invariants (r,c1),∀t ∈T, are stable with invariants (r,c1),∀t ∈T, are stable with invariants (r,c1),∀t ∈T, π+) remembers the point x1 ∈S (resp. xn ∈S) if and only if # = 1. For example: given by forgetting the ﬁrst (respectively last) sheaf in the ﬂag (2.16). The number # is 0 or 1 depending on whether the ﬁrst (respectively last) element of λ is or is not equivalent to some other element of λ. Therefore, the map π−(resp. π+) remembers the point x1 ∈S (resp. xn ∈S) if and only if # = 1. For example: Z(x,y,x,z) π− Z(y,x,z) but Z(y,x,x,z) π− Z(x,x,z) × S Z(x,y,x,z) π− Z(y,x,z) but Z(y,x,x,z) π− Z(x,x,z) × S Z(x,x,z) × S As is clear from the example above, in the case of the arrow on the left, there is no reason to remember the point x ∈S “forgotten” from the set partition λ = (x,y,x,z), since it As is clear from the example above, in the case of the arrow on the left, there is no reason to remember the point x ∈S “forgotten” from the set partition λ = (x,y,x,z), since it A. NEGUT, can be recovered from \|λ = (y,x,z). In other words, our convention when # = 0 is that π−(resp. π+) composed with the graph of the map that remembers x1 (resp. xn) yields the analogous map Zλ →Z\|λ × S (resp. Zλ →Zλ\| × S) to the one we would have deﬁned in the # = 1 case. 2.18. There exists a natural map pi : Zλ →M which only remembers the sheaf Fi in the ﬂag (2.16). We will write Ui = (pi × IdS)∗(U), and note that Proposition 2.14 implies that there exists an exact sequence 2.18. There exists a natural map pi : Zλ →M which only remembers the sheaf Fi in the ﬂag (2.16). We will write Ui = (pi × IdS)∗(U), and note that Proposition 2.14 implies that there exists an exact sequence 0 →Wi →Vi →Ui →0 (2.19) on Zλ × S, where Wi and Vi are locally free. While the map pi is not ﬂat, (2.19) is short exact because all three of the coherent sheaves in (2.14) are ﬂat over M. In the following Proposition, we consider any set partition λ of size n, and let Zλ denote the moduli space (2.16), whose points will be denoted by F0 ⊂··· ⊂Fn. Proposition 2.19. — The map π−of (2.18) can be realized as the diagonal arrow in Proposition 2.19. — The map π−of (2.18) can be realized as the diagonal arrow in Zλ π− PZ\|λ×S(V1) ρ Z\|λ × S if # = 1, and (2.20) (2.20) if # = 1, and Zλ π− PZ\|λ(x∗(V1)) ρ Z\|λ if # = 0 (2.21) (2.21) (in the latter case, x : Z\|λ →Z\|λ × S is the graph of the function Z\|λ →S that remembers the support point we forget when going from λ to \|λ) where →is the closed embedding cut out by the following composition of maps of locally free sheaves on the projectivization: ρ∗(W1) →ρ∗(V1) taut ↠O(1) if # = 1, (2.22) ρ∗(x∗(W1)) →ρ∗(x∗(V1)) taut ↠O(1) if # = 0 (2.23) (2.23) The line bundle L1 on Zλ is the restriction of O(1) from the projectivization. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Similarly, the map π+ of (2.18) can be realized as the diagonal arrow in Zλ π+ PZλ\|×S(W∨ n ⊗ωS) ρ Zλ\| × S if # = 1, and (2.24) if # = 1, and (2.24) Zλ π+ PZλ\|(x∗(W∨ n ⊗ωS)) ρ Zλ\| if # = 0 (2.25) (we write ωS for the canonical line bundle on S, and its various pullbacks) where →is the closed embedding cut out by the following composition of maps of locally free sheaves on the projectivization: ρ∗(V∨ n ⊗ωS) →ρ∗(W∨ n ⊗ωS) taut ↠O(1) if # = 1, and (2.26) ρ∗(x∗(V∨ n ⊗ωS)) →ρ∗(x∗(W∨ n ⊗ωS)) taut ↠O(1) if # = 0 (2.27) (2.27) The line bundle Ln on Zλ is the restriction of O(−1) from the projectivization. The line bundle Ln on Zλ is the restriction of O(−1) from the projectivization. Proof. — When λ has size 1, the Proposition was proved in [27, Deﬁnition 2.5, Proposition 2.8, Proposition 2.10], and the general case will follow the same logic. (2.19) Then a map T →PZ\|λ×S(V1) T →PZ\|λ×S(V1) consists of the data of ¯φ and x1 as above, together with a line bundle L1 on T and a surjective homomorphism consists of the data of ¯φ and x1 as above, together with a line bundle L1 on T and a surjective homomorphism τ : φ∗(V1) ↠L1 τ : φ∗(V1) ↠L1 Finally, a map from T to the subscheme Zλ of (2.20) consists of ¯φ, x1, L1 and τ as above, such that the following composition vanishes: Finally, a map from T to the subscheme Zλ of (2.20) consists of ¯φ, x1, L1 and τ as above, such that the following composition vanishes: φ∗(W1) →φ∗(V1) τ↠L1 φ∗(W1) →φ∗(V1) τ↠L1 Since pull-back is right-exact, this is equivalent to a surjective map φ∗(U1) ↠L1 ⇔ 1∗◦( ¯φ × IdS)∗(U1) ↠L1 ⇔ 1∗(F1) ↠L1 By adjunction, this datum is equivalent to a surjection F1 ↠1 ∗(L1) on T×S. Letting F0 be the kernel of this surjection, this precisely completes the ﬂag (2.28) to the ﬂag (2.17). By [27, Proposition 5.5], F0 is stable if and only if F1 is stable. □ 2.20. Consider an arbitrary set partition λ of size n and the scheme Zλ of Deﬁ- nition 2.16. Let π1 : Zλ × S →Zλ and π2 : Zλ × S →S denote the standard projections. For each i ∈{1,...,n}, we have the following short exact sequence on Zλ × S: 2.20. Consider an arbitrary set partition λ of size n and the scheme Zλ of Deﬁ- nition 2.16. Let π1 : Zλ × S →Zλ and π2 : Zλ × S →S denote the standard projections. For each i ∈{1,...,n}, we have the following short exact sequence on Zλ × S: 2.20. Consider an arbitrary set partition λ of size n and the scheme Zλ of Deﬁ- nition 2.16. Let π1 : Zλ × S →Zλ and π2 : Zλ × S →S denote the standard projections. For each i ∈{1,...,n}, we have the following short exact sequence on Zλ × S: (2.29) 0 →Ui−1 →Ui →Ai ⊗Oi →0 (2.29) where Ai = π ∗ 1 (Li), and i ⊂Zλ × S denotes the graph of the map where Ai = π ∗ 1 (Li), and i ⊂Zλ × S denotes the graph of the map pi S : Zλ →S that remembers the i-th support point xi. (2.19) We will prove the statements pertaining to π−, and leave those pertaining to π+ as exercises to the interested reader. Moreover, we will only prove the case # = 1, as # = 0 is analogous. We interpret the Proposition as follows: having constructed Z\|λ which represents the functor (2.17) for the set partition \|λ, we must show that the scheme Zλ deﬁned as the closed embedding (2.20) represents the functor (2.17) for the set partition λ. Recall that a map φ : T →Z\|λ × S consists of a ﬂag (2.28) F1 ⊂F2 ⊂··· ⊂Fn (2.28) of coherent sheaves on T × S which are ﬂat over T, together with maps x1,x2,...,xn : T →S and line bundles L2,...,Ln on T such that Fi/Fi−1 ∼= i ∗(Li) for all i ∈{2,...,n}, where i : T →T × S denotes the graph of xi. We write x1 for the map T →S arising from the and line bundles L2,...,Ln on T such that Fi/Fi−1 ∼= i ∗(Li) for all i ∈{2,...,n}, where i : T →T × S denotes the graph of xi. We write x1 for the map T →S arising from the A. NEGUT, second factor of Z\|λ × S and 1 for the graph of x1. Therefore, we may write ¯ 1 second factor of Z\|λ × S and 1 for the graph of x1. Therefore, we may write second factor of Z\|λ × S and 1 for the graph of x1. Therefore, we may write second factor of Z\|λ × S and 1 for the graph of x1. Therefore, we may write φ = ( ¯φ × IdS) ◦1 φ = ( ¯φ × IdS) ◦1 where ¯φ : T →Z\|λ represents the ﬂag (2.28). Then a map where ¯φ : T →Z\|λ represents the ﬂag (2.28). (2.19) If i and j are equivalent elements of the set partition λ, then i = j. We may upgrade (2.29) to the following commutative diagram that remembers the i-th support point xi. If i and j are equivalent elements of the set partition λ, then i = j. We may upgrade (2.29) to the following commutative diagram HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES of sheaves on Zλ × S with all rows and columns exact: of sheaves on Zλ × S with all rows and columns exact: 2.30) 0 0 0 0 Ui−1 Ui Bi ⊗Oi 0 0 Vi−1 Vi Bi 0 0 Wi−1 Wi Bi ⊗Ii 0 0 0 0 (2.30) 0 0 0 0 Ui−1 Ui Bi ⊗Oi 0 0 Vi−1 Vi Bi 0 0 Wi−1 Wi Bi ⊗Ii 0 0 0 0 (2.30) 0 where Ii denotes the ideal of the graph i inside Zλ × S, and the line bundle (2.31) Bi = π ∗ 1 Li ⊗pi∗ S (O(nH)) ⊗π ∗ 2 (O(−nH)) (2.31) is isomorphic to the quotient Vi/Vi−1 due to (2.15). Note that Bi\|i ∼= Ai\|i ∼= Li. In the Proposition below, we will restrict the diagram (2.30) to 1 and n. is isomorphic to the quotient Vi/Vi−1 due to (2.15). Note that Bi\|i ∼= Ai\|i ∼= Li. In the Proposition below, we will restrict the diagram (2.30) to 1 and n. Proposition 2.21. — Let λ be a set partition of size n. Proposition 2.21. — Let λ be a set partition of size n. (−) If p1 S is ﬂat, then there exists an injective map of sheaves L1 ⊗p1∗ S (ωS) →W0\|1 with locally free quotient, such that the following composition vanishes: (2.32) L1 ⊗p1∗ S (ωS) →W0\|1 →V0\|1 L1 ⊗p1∗ S (ωS) →W0\|1 →V0\|1 (2.32) (+) If pn S is ﬂat, then there exists an injective map of sheaves L−1 n →(Vn\|n)∨with locally free quotient, such that the following composition vanishes: (+) If pn S is ﬂat, then there exists an injective map of sheaves L−1 n →(Vn\|n)∨with locally free quotient, such that the following composition vanishes: (2.33) L−1 n →(Vn\|n)∨→(Wn\|n)∨ L−1 n →(Vn\|n)∨→(Wn\|n)∨ (2.33) (2.33) In both formulas above, the second arrows are induced by (2.19). In both formulas above, the second arrows are induced by (2.19). Remark 2.22. — The ﬂatness of p1 S,...,pn S is proved in Proposition 6.2 for any λ of size n ≤3, but we expect it to hold for any set partition λ. A. NEGUT, Proof. — Let us ﬁrst prove (−). Upon restricting (2.30) (for i = 1) to 1 : Zλ → Zλ × S, we obtain the following diagram of Tor sheaves with exact rows and columns: (2.34) U0\|1 U1\|1 B1\|1 V0\|1 V1\|1 B1\|1 B1 ⊗T or1(I1,O1) W0\|1 W1\|1 B1 ⊗I1\|1 0 B1 ⊗T or2(O1,O1) T or1(U0,O1) T or1(U1,O1) B1 ⊗T or1(O1,O1) (2.34) The injectivity of the maps denoted by →in the diagram above is due to the fact that W1,V0,V1,B1 are locally free, except for the injectivity of the horizontal map in the bottom left corner, which follows from the fact that U1 has homological dimension 1 (see (2.19)). We have the following ﬁber square: Zλ 1 p1 S Zλ × S p1 S×IdS S S × S The assumption that p1 S is a ﬂat morphism implies that The assumption that p1 S is a ﬂat morphism implies that T or2(O1,O1) = p1∗ S (T or2(O,O)) = p1∗ S (ωS) on Zλ. The latter equality in the equation above is a standard and straightforward exer- cise, which follows from the fact that S →S×S is a smooth embedding of codimension 2. If we recall the fact that B1\|1 ∼= L1, then we conclude that B1 ⊗T or2(O1,O1) ∼= L1 ⊗p1∗ S (ωS) (2.35) (2.35) Since B1 ⊗T or2(O1,O1) injects into the kernel of W0\|1 →V0\|1 (as can be seen from diagram (2.34)), this establishes the vanishing of the composition (2.32). The fact that the sheaf Since B1 ⊗T or2(O1,O1) injects into the kernel of W0\|1 →V0\|1 (as can be seen from diagram (2.34)), this establishes the vanishing of the composition (2.32). The fact that the sheaf W0\|1 L1 ⊗p1∗ S (ωS) ∼= Ker W1\|1 ↠B1 ⊗I1\|1 is locally free follows from the facts that W1 is locally free, and I1\|1 has homological dimension 1 (since it is the pull-back of the deﬁning ideal of the codimension 2 regular embedding : S →S × S under the ﬂat morphism p1∗ S ). is locally free follows from the facts that W1 is locally free, and I1\|1 has homological dimension 1 (since it is the pull-back of the deﬁning ideal of the codimension 2 regular embedding : S →S × S under the ﬂat morphism p1∗ S ). HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES To deal with the (+) case, let us restrict (2.30) (for i = n) to n and then dualize: (2.36) (Vn−1\|n)∨ (Vn\|n)∨ (Bn\|n)∨ (Wn−1\|n)∨ (Wn\|n)∨ (Bn ⊗In\|n)∨ (2.36) The top-most row is exact, due to the fact that restriction and dualization preserve short exact sequences of locally free sheaves. Because pn S : Zλ →S is ﬂat, we have In\|n ∼= pn∗ S (1 S) ⇒ (In\|n)∨∼= pn∗ S (T 1 S ) and the right-most vertical map in (2.36) is zero. Therefore, the rank 1 sub-bundle (Bn\|n)∨∼= L−1 n embeds in the kernel of (Vn\|n)∨→(Wn\|n)∨, thus implying the van- ishing of the composition (2.33). Moreover, the quotient (Vn\|n)∨ L−1 n ∼= (Vn−1\|n)∨ (Vn\|n)∨ L−1 n ∼= (Vn−1\|n)∨ is locally free by construction. □ □ □ is locally free by construction. is locally free by construction. Remark 2.23. — A straightforward analogue of the proof given above shows that statement (−) (respectively (+)) of Proposition 2.21 would still hold with the numbers 0,1 (respectively n −1,n) replaced by i −1,i, for any i ∈{1,...,n}. 2.24. We will now state certain geometric properties of several schemes Zλ with n = \|λ\| ≤4, to be proved in Section 6. Let k be the number of equivalence classes of λ. Deﬁnition 2.25. — The expected dimension of (a connected component of) Zλ is Proposition 2.26. — Z(x) is smooth and irreducible of expected dimension. Deﬁnition 2.25. — The expected dimension of (a connected component of) Zλ is const + c2(F0) + c2(Fn) + k for any (F0 ⊂x1 ··· ⊂xn Fn) ∈Zλ. for any (F0 ⊂x1 ··· ⊂xn Fn) ∈Zλ. Moreover, consider the map π : Zλ →M × Sn which takes (F0 ⊂x1 ··· ⊂xn Fn) to (Fn,x1,...,xn). In what follows, whenever we say that Zλ is irreducible, we actually mean that π −1(C) is irreducible for every connected component C of M × Sn. Proposition 2.26. — Z(x) is smooth and irreducible of expected dimension. A. NEGUT, Proposition 2.27. — Z(x,x) is smooth and irreducible of expected dimension. Moreover, the fol- lowing ﬁber square is derived with excess: (2.37) Z(x,x) π− π+ Z(x) Z(x) M × S F0 ⊂x F1 ⊂x F2 F1 ⊂x F2 F0 ⊂x F1 (F1,x) F0 ⊂x F1 ⊂x F2 F1 ⊂x F2 F0 ⊂x F1 (F1,x) (2.37) The excess bundle is L2 ⊗L−1 1 ⊗p∗ S(ωS), where Z(x,x) pS−→S is the map that records x. Proposition 2.28. — Z(x,y) is l.c.i. and irreducible of expected dimension. Moreover, the following ﬁber square is derived: (2.38) Z(x,y) Z(y) × S Z(x) × S M × S × S F0 ⊂x F1 ⊂y F2 (F1 ⊂y F2,x) (F0 ⊂x F1,y) (F1,x,y) F0 ⊂x F1 ⊂y F2 (F1 ⊂y F2,x) (F0 ⊂x F1,y) (F1,x,y) (2.38) Proposition 2.29. — Z(x,x,x) is l.c.i. and irreducible of expected dimension. Moreover, the fol- lowing ﬁber square is derived: Z(x,x,x) Z(x,x) Z(x,x) Z(x) F0 ⊂x F1 ⊂x F2 ⊂x F3 F1 ⊂x F2 ⊂x F3 F0 ⊂x F1 ⊂x F2 F1 ⊂x F2 Z(x,x,x) Z(x,x) Z(x,x) Z(x) F0 ⊂x F1 ⊂x F2 ⊂x F3 F1 ⊂x F2 ⊂x F3 F0 ⊂x F1 ⊂x F2 F1 ⊂x F2 Proposition 2.30. — Z(x,x,y) and Z(y,x,x) are l.c.i. and irreducible of expected dimension. More- over, the following ﬁber squares are derived: Z(x,x,y) Z(y) × S Z(x,x) × S M × S × S F0 ⊂x F1 ⊂x F2 ⊂y F3 (F2 ⊂y F3,x) (F0 ⊂x F1 ⊂x F2,y) (F2,x,y) Z(y,x,x) Z(x,x) × S Z(y) × S M × S × S F0 ⊂y F1 ⊂x F2 ⊂x F3 (F1 ⊂x F2 ⊂x F3,y) (F0 ⊂y F1,x) (F1,x,y) Proposition 2.31. — Z(x,y,x) is Cohen-Macaulay and irreducible of expected dimension. Proposition 2.31. — Z(x,y,x) is Cohen-Macaulay and irreducible of expected dimension. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Proposition 2.32. — Z(x,x,x,x) is l.c.i. and has two irreducible components of expected dimension. Deﬁnition 2.25. — The expected dimension of (a connected component of) Zλ is Moreover, the following ﬁber squares are derived: Z(x,x,x,x) Z(x,x,x) Z(x,x) Z(x) F0 ⊂x F1 ⊂x F2 ⊂x F3 ⊂x F4 F1 ⊂x F2 ⊂x F3 ⊂x F4 F0 ⊂x F1 ⊂x F2 F1 ⊂x F2 Z(x,x,x,x) Z(x,x) Z(x,x,x) Z(x) F0 ⊂x F1 ⊂x F2 ⊂x F3 ⊂x F4 F2 ⊂x F3 ⊂x F4 F0 ⊂x F1 ⊂x F2 ⊂x F3 F2 ⊂x F3 Proposition 2.33. — Z(x,x,y,x) and Z(x,y,x,x) are Cohen-Macaulay and irreducible of expected dimension. Moreover, the following ﬁber squares are derived: Z(x,x,y,x) Z(x,y,x) Z(x,x) Z(x) F0 ⊂x F1 ⊂x F2 ⊂y F3 ⊂x F4 F1 ⊂x F2 ⊂y F3 ⊂x F4 F0 ⊂x F1 ⊂x F2 F1 ⊂x F2 Z(x,y,x,x) Z(x,x) Z(x,y,x) Z(x) F0 ⊂x F1 ⊂y F2 ⊂x F3 ⊂x F4 F2 ⊂x F3 ⊂x F4 F0 ⊂x F1 ⊂y F2 ⊂x F3 F2 ⊂x F3 Finally, we will need the normality of some of the varieties Zλ. Z(x,x,y,x) Z(x,y,x) Z(x,x) Z(x) F0 ⊂x F1 ⊂x F2 ⊂y F3 ⊂x F4 F1 ⊂x F2 ⊂y F3 ⊂x F4 F0 ⊂x F1 ⊂x F2 F1 ⊂x F2 Z(x,y,x,x) Z(x,x) Z(x,y,x) Z(x) F0 ⊂x F1 ⊂y F2 ⊂x F3 ⊂x F4 F2 ⊂x F3 ⊂x F4 F0 ⊂x F1 ⊂y F2 ⊂x F3 F2 ⊂x F3 F0 ⊂x F1 ⊂x F2 ⊂y F3 ⊂x F4 F1 ⊂x F2 ⊂y F3 ⊂x F4 F0 ⊂x F1 ⊂x F2 F1 ⊂x F2 F0 ⊂x F1 ⊂y F2 ⊂x F3 ⊂x F4 F2 ⊂x F3 ⊂x F4 F0 ⊂x F1 ⊂y F2 ⊂x F3 F2 ⊂x F3 Finally, we will need the normality of some of the varieties Zλ. Finally, we will need the normality of some of the varieties Zλ. Proposition 2.34. — The scheme Zλ is normal for any Proposition 2.34. — The scheme Zλ is normal for any 9) λ ∈ (x,y),(x,x,y),(x,y,x),(y,x,x),(x,x,y,x),(x,y,x,x) (2.39) λ ∈ (x,y),(x,x,y),(x,y,x),(y,x,x),(x,x,y,x),(x,y,x,x) (2.39) (2.39) 2.35. Let us consider the spaces Y, Y−, Y+, Y−+ which parameterize diagrams 2.35. Let us consider the spaces Y, Y−, Y+, Y−+ which parameterize diagram F1 y F0 x y F2 F ′ 1 x A. NEGUT, (2.41) F1 y F0 x y F2 x F3 F ′ 1 x (2.42) F2 y F0 x F1 x y F3 F ′ 2 x (2.43) F2 y F0 x F1 x y F3 x F4 F ′ 2 x A. Deﬁnition 2.25. — The expected dimension of (a connected component of) Zλ is NEGUT, (2.41) F1 y F0 x y F2 x F3 F ′ 1 x (2.41) F1 F0 F ′ 1 F2 y F0 x F1 x y F3 F ′ 2 x (2.42) F3 (2.43) F ′ 2 respectively, of stable coherent sheaves where each successive inclusion is colength 1 and supported at the point indicated on the diagram. Strictly speaking, Y, Y−, Y+, Y−+ are functors which associate to a scheme T diagrams of ﬂat families of stable coherent sheaves (2.40), (2.41), (2.42), (2.43) on T × S, satisfying all the standard properties. On these schemes, we have the line bundles with ﬁbers respectively, of stable coherent sheaves where each successive inclusion is colength 1 and supported at the point indicated on the diagram. Strictly speaking, Y, Y−, Y+, Y−+ are functors which associate to a scheme T diagrams of ﬂat families of stable coherent sheaves (2.40), (2.41), (2.42), (2.43) on T × S, satisfying all the standard properties. On these schemes, we have the line bundles with ﬁbers Li = (S,Fi/Fi−1) and where the notation is applicable, line bundles and where the notation is applicable, line bundles L′ i = (S,F ′ i /Fi−1) or L′ i = (S,Fi/F ′ i−1) Note that L1L2 = L′ 1L′ 2 ∈Pic(Y),Pic(Y−) (2.44) L2L3 = L′ 2L′ 3 ∈Pic(Y+),Pic(Y−+) (2.45) HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES 2.36. Consider the maps 2.36. Consider the maps (2.46) Y π↑ Z(x,y) Y− π↑ Z(x,y,x) Y+ π↑ Z(x,x,y) Y−+ π↑ Z(x,x,y,x) (2.46) obtained by remembering only the middle and top part of (2.40)–(2.43), and (2.47) Y π↓ Z(y,x) Y− π↓ Z(y,x,x) Y+ π↓ Z(x,y,x) Y−+ π↓ Z(x,y,x,x) (2.47) obtained by remembering only the middle and bottom part of (2.40)–(2.43). In Proposi- tion 2.37 we will show that the maps (2.46) and (2.47) are representable, which together with Proposition 2.19 shows that the functors Y, Y−,Y+, Y−+ are themselves repre- sentable. Let us ﬁrst introduce some notation pertaining to Y. The short exact sequence 0 −→U1/U0 = x ∗(L1) −→U2/U0 −→U2/U1 = y ∗(L2) −→0 consists of coherent sheaves on Z2 × S which are ﬂat over Z2 = Z(x,y). Here and in what follows, x and y denote the graphs of the maps consists of coherent sheaves on Z2 × S which are ﬂat over Z2 = Z(x,y). Here and in what follows, x and y denote the graphs of the maps px S,p y S : Z2 →S px S,p y S : Z2 →S which record the points x and y, respectively. Let us write Z2 × S pr−→Z2 Z2 × S pr−→Z2 for the standard projection. Then the short exact sequence for the standard projection. Then the short exact sequence (2.48) 0 −→L1 −→E := pr∗(U2/U0) −→L2 −→0 (2.48) consists of locally free sheaves of ranks 1,2,1, respectively, on Z2. The composition consists of locally free sheaves of ranks 1,2,1, respectively, on Z2. The composition pr∗(E) ⊗Iy →pr∗(E) −→U2/U0 −→y ∗(L2) vanishes (on account of Iy being the ideal sheaf of functions which vanish on any coher- ent sheaf of the form y ∗(...)), and therefore induces a map vanishes (on account of Iy being the ideal sheaf of functions which vanish on any coher- ent sheaf of the form y ∗(...)), and therefore induces a map pr∗(E) ⊗Iy −→U1/U0 = x ∗(L1) A. NEGUT, By adjunction, this gives rise to a map By adjunction, this gives rise to a map x∗(pr∗(E)) ⊗x∗(Iy) −→L1 which can be rewritten as (2.49) E ⊗(px S × p y S)∗(I) −→L1 E ⊗(px S × p y S)∗(I) −→L1 (2.49) where I is the ideal of the diagonal : S →S×S. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Indeed, the identiﬁcation x∗(Iy) ∼= (px S × p y S)∗(I) stems from the ﬁber square where I is the ideal of the diagonal : S →S×S. Indeed, the identiﬁcation x∗(Iy) ∼= (px S × p y S)∗(I) stems from the ﬁber square (2.50) Z2 y py S Z2 × S py S×IdS S S × S (2.50) Proposition 6.2 asserts that the vertical maps are ﬂat, hence Iy = (p y S × Id)∗(I). The preceding discussion applies equally well to the spaces Y−, Y+, Y−+ instead of Y (the sheaves U0,U1,U2,L1,L2 must be replaced by U1,U2,U3,L2,L3 in the case of Y+ and Y−+, see the notation in (2.40), (2.41), (2.42), (2.43)). Proposition 2.37. — With the notation above, we have Proposition 2.37. — With the notation above, we have Proposition 2.37. — With the notation above, we have (2.51) Y π↑or π↓ ι↑or ι↓ PZ2(E) ρ Z2 (2.51) Y π↑or π↓ ι↑or ι↓ PZ2(E) ρ Z2 (2.51) Y π↑or π↓ ι↑or ι↓ PZ2(E) ρ Z2 (2.51) where ι↑and ι↓are cut out by the following section: where ι↑and ι↓are cut out by the following section: where ι↑and ι↓are cut out by the following section: HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES (1) a map ρ : T →Z2, i.e. a ﬂat family of stable coherent sheaves (F0 ⊂F1 ⊂F2) on T × S, together with line bundles L1, L2 on T and points x,y : T →S such that F1/F0 ∼= x ∗(L1) and F2/F1 ∼= y ∗(L2). ∗ (2) a line bundle O(1) on T and a surjective homomorphism ∗ (2) a line bundle O(1) on T and a surjective homomorphism E ↠O(1) (2.52) (2 2) ( ) where E = pr∗(F2/F0) with pr : T × S →T the standard projection. If we let H denote the kernel of (2.52), then one requires that the composition where E = pr∗(F2/F0) with pr : T × S →T the standard projection. If we let H denote the kernel of (2.52), then one requires that the composition (2.53) H ⊗(x × y)∗(I) →L1 be 0, where x × y : T →S × S is the product of the maps in item (1). We must show that any datum as in items (1)–(2) above gives rise to a square (2.40). Item (1) yields the commutative diagram with exact rows and column 0 pr∗(L1) pr∗(E) pr∗(L2) 0 0 x ∗(L1) ι F2/F0 π y ∗(L2) 0 on T × S. The kernel of (2.52) gives rise to a sub line bundle H ⊂E, and the question is when y ∗(H) is a subsheaf of F2/F0 (this subsheaf would be F′ 1/F0, and this would complete the square (2.40) by constructing the bottom sheaf). It is easy to see that this happens precisely when the following composition vanishes: ν : pr∗(H) ⊗Iy →pr∗(E) ↠F2/F0 Since π ◦ν = 0, the map ν takes values in x ∗(L1), so the question is when the map Since π ◦ν = 0, the map ν takes values in x ∗(L1), so the question is when the map pr∗(H) ⊗Iy →x ∗(L1) vanishes. By adjunction, this happens when vanishes. By adjunction, this happens when (2.54) H ⊗x∗(Iy) →L1 H ⊗x∗(Iy) →L1 where ι↑and ι↓are cut out by the following section: σ : H ⊗ρ∗ (px S × p y S)∗(I) −→ρ∗E ⊗(px S × p y S)∗(I) (2.49) −−→ρ∗(L1) = Ker ρ∗(E) ↠O(1). The same formulas hold for the spaces Y−, Y+, Y−+ instead of Y, cing Z2 by the corresponding spaces in (2.46) and (2.47). σ : H ⊗ρ∗ (px S × p y S)∗(I) −→ρ∗E ⊗(px S × p y S)∗(I) (2.49) −−→ρ∗(L1) σ : H ⊗ρ∗ (px S × p y S)∗(I) −→ρ∗E ⊗(px S × p y S)∗(I) (2.49) −−→ρ∗(L1) r ρ∗(E) ↠O(1) The same formulas hold for the spaces Y Y Y instead of Y ⊗ρ (pS pS) ( ) ρ ⊗(pS pS) ( ) ρ ( 1) with H = Ker ρ∗(E) ↠O(1). The same formulas hold for the spaces Y−, Y+, Y−+ instead of Y, but replacing Z2 by the corresponding spaces in (2.46) and (2.47). with H = Ker ρ∗(E) ↠O(1). The same formulas hold for the spaces Y−, Y+, Y−+ instead of Y, but replacing Z2 by the corresponding spaces in (2.46) and (2.47). Proof. — We will only prove the required statement for Y, as the cases of Y−, Y+, Y−+ are analogous (equivalently, they follow from base change in the derived ﬁber squares (6.40)–(6.41)). Also, we will only prove the case of the map π ↑, as the situation is symmetric by replacing (↑,x,y,L1,L2) with (↓,y,x,L′ 1,L′ 2). If we deﬁne the scheme Y by (2.51), then maps T →Y are in one-to-one correspondence with Proposition 2.41. — Y is smooth and irreducible of expected dimension. (2.54) vanishes. Since (x × y)∗(I) = x∗(Iy) (as a consequence of the sentence preceding diagram (2.50)), the vanishing of (2.54) is equivalent to the vanishing of (2.53). □ vanishes. Since (x × y)∗(I) = x∗(Iy) (as a consequence of the sentence preceding diagram (2.50)), the vanishing of (2.54) is equivalent to the vanishing of (2.53). □ 2.38. Now that we have deﬁned the schemes Y, Y−, Y+, Y−+, let us consider their basic properties. An important fact is the following geometric observation. A. NEGUT, Proposition 2.39. — The composition F1/F0 →F2/F0 ↠F2/F ′ 1 (see the notation in (2.40)) induces the following map of line bundles on Y: Proposition 2.39. — The composition F1/F0 →F2/F0 ↠F2/F ′ 1 (see the notation in (2.40)) induces the following map of line bundles on Y: (2.55) L1 →L′ 2 (2.55) If we interpret this map as a section of L′ 2 ⊗L−1 1 , then its zero locus consists of If we interpret this map as a section of L′ 2 ⊗L−1 1 , then its zero locus consists of (2.56) (F1,x) = (F ′ 1,y) →Y, (2.56) and is isomorphic to Z(x,x). The analogous result holds for Y−,Y+,Y−+. and is isomorphic to Z(x,x). The analogous result holds for Y−,Y+,Y−+. Proof. — By the functor-of-points description, a map from a scheme T into the zero locus of (2.55) consists of a square (2.40) (of coherent sheaves on T × S, ﬂat over T) such that the corresponding induced map (2.57) pr∗(F1/F0) →pr∗(F2/F ′ 1) (2.57) vanishes, where pr : T × S →T denotes the projection (here we have used the facts that F1/F0 ∼= x ∗(L1), F2/F ′ 1 ∼= x ∗(L′ 2) and pr ◦x = Id). Because the sheaves F1/F0, and F2/F ′ 1 are ﬂat of length 1 over T, the map pr∗is an isomorphism on local sections, hence the induced map F1/F0 →F2/F ′ 1 vanishes. This implies that F1 ⊂F ′ 1 as subsheaves of F2, hence we obtain short exact sequences vanishes, where pr : T × S →T denotes the projection (here we have used the facts that F1/F0 ∼= x ∗(L1), F2/F ′ 1 ∼= x ∗(L′ 2) and pr ◦x = Id). Because the sheaves F1/F0, and F2/F ′ 1 are ﬂat of length 1 over T, the map pr∗is an isomorphism on local sections, hence the induced map F1/F0 →F2/F ′ 1 vanishes. This implies that F1 ⊂F ′ 1 as subsheaves of F2, hence we obtain short exact sequences 0 →F1/F0 f−→F ′ 1/F0 g−→F ′ 1/F1 →0 0 →F ′ 1/F1 f ′ −→F2/F1 g′ −→F2/F ′ 1 →0 The homomorphism f is an injection x ∗(L1) → y ∗(L′ 1), which implies that x = y as maps T →S (the statement is local on both S and T). Since x is a closed embedding, x ∗is an exact functor and so the homomorphisms f and g′ induce line bundle homomorphisms L1 →L′ 1 and L2 ↠L′ 2 on T. Since L1L2 ∼= L′ 1L′ 2 ∼= detE with E = pr∗(F2/F0), this is only possible if the aforementioned line bundle homomorphisms are isomorphisms (in other words, if ab is a unit in a certain ring, then both a and b are units). Therefore, we must have L1 = L′ 1 as subsheaves of E, hence F1 = F ′ 1 as subsheaves of F2, which precisely yields a point of Z(x,x). The analogous results for Y−,Y+,Y−+ instead of Y are proved by the same argument. □ 2.40. and is isomorphic to Z(x,x). The analogous result holds for Y−,Y+,Y−+. The following results will be proved in Section 6. The expected dimension of the varieties Y, Y−, Y+, Y−+ is deﬁned to be the dimension of the respective spaces on the bottom of (2.46) or (2.47) (which is equal to the actual dimension of these spaces by Propositions 2.28, 2.30, 2.31, 2.33). 2.40. The following results will be proved in Section 6. The expected dimension of the varieties Y, Y−, Y+, Y−+ is deﬁned to be the dimension of the respective spaces on the bottom of (2.46) or (2.47) (which is equal to the actual dimension of these spaces by Propositions 2.28, 2.30, 2.31, 2.33). HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Proposition 2.42. — Y−and Y+ are l.c.i. and irreducible of expected dimension. Proposition 2.42. — Y−and Y+ are l.c.i. and irreducible of expected dimension. Proposition 2.42. — Y−and Y+ are l.c.i. and irreducible of expected dimension. Proposition 2.43. — Y−+ is l.c.i. and has two irreducible components of expected dimension. Using the results above, we will prove the following fact in Section 6. Proposition 2.44. — The schemes Y, Y−, Y+, Y−+ are reduced. Proposition 2.42. — Y−and Y+ are l.c.i. and irreducible of expected dimension. Proposition 2.43. — Y−+ is l.c.i. and has two irreducible components of expected dimension. U i h l b ill h f ll i f i S i 6 position 2.43. — Y−+ is l.c.i. and has two irreducible components of expected dimension. Proposition 2.43. — Y−+ is l.c.i. and has two irreducible components of expected dimension. Using the results above, we will prove the following fact in Section 6. Proposition 2.44. — The schemes Y, Y−, Y+, Y−+ are reduced. Our main interest in the above geometric properties is the following result. Proposition 2.45. — The map π ↑: Y →Z2 = Z(x,y) has the property that Proposition 2.45. — The map π ↑: Y →Z2 = Z(x,y) has the property that Riπ ↑ ∗(OY) = OZ2 if i = 0 0 if i > 0 (2.58) of a single po that f∗(OX) = To ob reduced by P (which is pro the ﬁber of π is isomorphi Remark placed by th sheaves F by nested Hilbe blow-up of it 3. Derived 3.1. We recall tha free sheaves is equivalent inclusion fun (3.1) is fully faithf A. NEGUT, of a single point, and the normality of Y implies that h is an isomorphism. We conclude that f∗(OX) = h∗◦g∗(OX) = h∗(OY′) = OY. □ of a single point, and the normality of Y implies that h is an isomorphism. We conclude that f∗(OX) = h∗◦g∗(OX) = h∗(OY′) = OY. □ To obtain formula (2.58) for i = 0, we apply Claim 2.46 for X = Y (which is reduced by Proposition 2.44), Y = Z2 (which is normal by Proposition 2.34) and f = π ↑ (which is proper by Proposition 2.37). Moreover, the proof of Proposition 2.37 shows that the ﬁber of π ↑above a closed point (F0 ⊂x F1 ⊂y F2) (F0 ⊂x F1 ⊂y F2) □ □ Remark 2.47. — All the results in Section 2 hold with the moduli space M re- placed by the Hilbert schemes of points Hilb(S): all one needs to do is to replace stable sheaves F by ideal sheaves I everywhere. In particular, Proposition 2.45 shows that the nested Hilbert scheme Z2 = {(I0 ⊂I1 ⊂I2)} has rational singularities, and Y is the blow-up of its singular locus (see the proof of Proposition 6.9). (2.58) The analogous properties hold for π ↓. Moreover, the analogous properties hold with the scheme Y replaced by the schemes Y−, Y+, Y−+ of (2.46) and (2.47). The analogous properties hold for π ↓. Moreover, the analogous properties hold with the scheme Y replaced by the schemes Y−, Y+, Y−+ of (2.46) and (2.47). Proof. — Recall from Proposition 2.37 that π ↑= ρ ◦ι↑, where ι↑is a closed em- bedding and ρ : PZ2(E) ↠Z2 is a P1-bundle. Therefore, we have a short exact sequence (2.59) 0 →Kernel →OPZ2(E) →ι↑ ∗(OY) →0 (2.59) of coherent sheaves on PZ2(E). Because PZ2(E) is a P1-bundle over Z2, we have of coherent sheaves on PZ2(E). Because PZ2(E) is a P1-bundle over Z2, we h Riρ∗(OPZ2(E)) = 0 ∀i ≥1 Riρ∗(Kernel) = 0 ∀i ≥2 (the latter equality would actually hold for any coherent sheaf on PZ2(E)). Therefore, in the long exact sequence in cohomology associated to (2.59) (the latter equality would actually hold for any coherent sheaf on PZ2(E)). Therefore, in the long exact sequence in cohomology associated to (2.59) ··· →Riρ∗(OPZ2(E)) →Riρ∗◦ι↑ ∗(OY) →Ri+1ρ∗(Kernel) →··· the spaces on the left and on the right are 0 for any i ≥1. This implies (2.58) for i ≥1. As for the case i = 0, let us recall the following well-known fact. the spaces on the left and on the right are 0 for any i ≥1. This implies (2.58) for i ≥1. As for the case i = 0, let us recall the following well-known fact. Claim 2.46. — Assume X is reduced, Y is normal, and f : X →Y is a proper morphism with all ﬁbers connected and non-empty. Then f∗(OX) = OY. Proof. — Stein factorization implies that we can express f as Proof. — Stein factorization implies that we can express f as X g −→Y′ h −→Y with g∗(OX) = OY′ and h a ﬁnite morphism. However, because the ﬁbers of f = h ◦g are connected, so are the ﬁbers of h. Since h is ﬁnite, this implies that all the ﬁbers of h consist with g∗(OX) = OY′ and h a ﬁnite morphism. However, because the ﬁbers of f = h ◦g are connected, so are the ﬁbers of h. Since h is ﬁnite, this implies that all the ﬁbers of h consist 3. Derived categories 3.1. Recall that all our schemes are projective over C. Consider the following: Db(Coh(X)) = the bounded derived category of coherent sheaves on X P f(X) th d i d t f f t l X 3.1. Recall that all our schemes are projective over C. Consider the following Db(Coh(X)) = the bounded derived category of coherent sheaves on X Perf(X) = the derived category of perfect complexes on X We recall that a complex is perfect if it is quasi-isomorphic to a ﬁnite complex of locally free sheaves of ﬁnite rank on X (this notion is usually called “strictly perfect”, although it is equivalent to the more general notion of “perfect” on projective varieties). The natural inclusion functor (3.1) Perf(X) →Db(Coh(X)) (3.1) is fully faithful, and is an equivalence if the scheme X is smooth. We will write is fully faithful, and is an equivalence if the scheme X is smooth. We will write DX = Perf(X) = Db(Coh(X)) if X is smooth The functors between derived categories associated to a morphism f : X →Y are The functors between derived categories associated to a morphism f : X →Y are (1) Db(Coh(X)) f∗−→Db(Coh(Y)) if f is proper f (1) Db(Coh(X)) f∗−→Db(Coh(Y)) if f is proper f (2) Perf(X) f∗−→Perf(Y) if f is proper and l.c.i. f ∗ (2) Perf(X) f∗−→Perf(Y) if f is proper and l.c.i. (3) Db(Coh(Y)) f ∗ −→Db(Coh(X)) if f has ﬁnite Tor dimension f ∗ (4) Perf(Y) f ∗ −→Perf(X) if f is arbitrary (4) Perf(Y) f ∗ −→Perf(X) if f is arbitrary HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES The existence of the push-forward on the second line is non-trivial, and we refer the reader to [33] for an overview. As for the pull-back on the third line, we note that we always have a left derived functor D−(Coh(Y)) f ∗ −→D−(Coh(X)) on the derived categories of bounded above complexes of coherent sheaves over projec- tive schemes. However, in order to ensure that the functor f ∗takes bounded complexes to bounded complexes, one needs a strong assumption, such as f having ﬁnite Tor di- mension, i.e. T orOY−mod i (OX,−) = 0 for all i large enough. on the derived categories of bounded above complexes of coherent sheaves over projec- tive schemes. However, in order to ensure that the functor f ∗takes bounded complexes to bounded complexes, one needs a strong assumption, such as f having ﬁnite Tor di- mension, i.e. T orOY−mod i (OX,−) = 0 for all i large enough. Proposition 3.2. — Consider a derived ﬁber square as in Deﬁnition 2.4: Proposition 3.2. — Consider a derived ﬁber square as in Deﬁnition 2.4: A. NEGUT, Having established (3.3) and (3.4) at the level of unbounded complexes, the desired conclusion follows from the fact that Perf and Db(Coh) are full subcategories of the de- rived category of unbounded complexes, together with the fact that the pull-back and push-forward functors associated to ι,ι′,η,η′ are well-deﬁned on these subcategories due to items (1)–(4) that precede the statement of Proposition 3.2. □ 3.3. Associated to a scheme X, we have the K-theory groups 3.3. Associated to a scheme X, we have the K-theory groups K0(X) = Grothendieck group of Db(Coh(X)) K0(X) = Grothendieck group of Perf(X) If X is smooth, these groups are isomorphic and we will write If X is smooth, these groups are isomorphic and we will write KX = K0(X) = K0(X) if X is smooth Associated to a morphism f : X →Y, we have group homomorphisms f∗, f ∗between the various K-theory groups, as in the items (1)–(4) of Section 3.1. Moreover, (3.3) and (3.4) yield equalities of maps between K-theory groups, under the assumptions in Propo- sition 3.2. However, the following only holds in K-theory. Proposition 3.4 ([13, Proposition 2.2]). — Consider a derived ﬁber square with excess, as in Deﬁnition 2.5. Then (3.6) ι∗η∗= η′ ∗ ∧•(L) · ι′∗ : K0(X′) →K0(Z) (3.6) where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. Proposition 3.2. — Consider a derived ﬁber square as in Deﬁnition 2.4: Proposition 3.2. — Consider a derived ﬁber square as in Deﬁnition 2.4: (3.2) Z′ ι′ η′ X′ η Z ι X (3.2) Z′ ι′ η′ X′ η Z ι X (3.2) where η is proper. Suppose that Z →X and Z′ →X′ are closed embeddings cut out by regular sections σ and η∗(σ) of locally free sheaves W and η∗(W), respectively. Then we have equivalences where η is proper. Suppose that Z →X and Z′ →X′ are closed embeddings cut out by regular sections σ and η∗(σ) of locally free sheaves W and η∗(W), respectively. Then we have equivalences (3.3) η∗ι∗∼= ι′ ∗η′∗: Perf(Z) →Perf(X′) (3.4) ι∗η∗∼= η′ ∗ι′∗: Db(Coh(X′)) →Db(Coh(Z)) (3.3) The same equivalences hold if the regular embeddings ι and ι′ are replaced by smooth morphisms such as the projectivization of a locally free sheaf coming from X. The same equivalences hold if the regular embeddings ι and ι′ are replaced by smooth morphisms such as the projectivization of a locally free sheaf coming from X. Proof. — The equivalences (3.3) and (3.4) are well-known to hold in the derived category of unbounded complexes, as long as the maps η and ι are Tor independent: (3.5) TorOX,x−mod i (OZ,z,OX′,x′) = 0 ∀i ≥1 (3.5) for all points z ∈Z and x′ ∈X′ which map to the same point x ∈X. Since ι is a regular embedding, we may replace OZ,z by the Koszul complex of the section σ ∨: W∨ x →OX,x, and so (3.5) is equivalent to the complex for all points z ∈Z and x′ ∈X′ which map to the same point x ∈X. Since ι is a regular embedding, we may replace OZ,z by the Koszul complex of the section σ ∨: W∨ x →OX,x, and so (3.5) is equivalent to the complex ... σ ∨ −→∧2(W∨ x ) ⊗OX,x OX′,x′ σ ∨ −→W∨ x ⊗OX,x OX′,x′ σ ∨ −→OX′,x′ being exact everywhere except at the right-most place. However, the above is none other than the Koszul complex of the section η∗(σ) of the locally free sheaf η∗(W), and its exactness follows from our assumption that the section η∗(σ) is regular. A. NEGUT, where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. Equality (3.6) is called the “excess intersection formula”, and it does not generally lift to the derived category. One reason for this is that formula (2.6) does not naturally give rise to a map s : L →O whose Koszul complex categoriﬁes ∧•(L). The case when (3.6) lifts to the derived category, with the section being s = 0, is treated in [1] and linked with the relation between Z′ and the derived ﬁber product Z ×L X X′. 3.5. Let us recall the schemes Z1 and Z• 2 of Section 2.15. It was shown in [27, 28] that these schemes are smooth (see also Propositions 2.26 and 2.27). Because of Proposi- tion 2.19, the maps p± and π± of Section 1.3 are all proper and l.c.i., which allow us to deﬁne functors (3.7) e(d1,...,dn) : DM →DM×S (3.7) for all d1,...,dn ∈Z, according to the composition immediately after (1.8). for all d1,...,dn ∈Z, according to the composition immediately after (1.8). HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Remark 3.6. — Analogously, one can deﬁne the transposed functor Remark 3.6. — Analogously, one can deﬁne the transposed functor Remark 3.6. — Analogously, one can deﬁne the transposed functor (3.8) f(d1,...,dn) : DM →DM×S (3.8) f(d1,...,dn) : DM →DM×S (3.8) (3.8) f(d1,...,dn) : DM →DM×S by the composition going the other way (left-to-right): by the composition going the other way (left-to-right): by the composition going the other way (left-to-right): DZ1 ⊗Ld1−r DZ1 π−∗π∗ + DZ1 ⊗Ld2−r DZ1 ... DZ1 π−∗π∗ + DZ1 ⊗Ldn−r DZ1 (p−×pS)∗ DM p∗ + DM×S followed by tensoring with ω⊗(1−r) S detU ⊗n and shifting complexes by rn. followed by tensoring with ω⊗(1−r) S detU ⊗n and shifting complexes by rn. followed by tensoring with ω⊗(1−r) S detU ⊗n and shifting complexes by rn. In Section 4, we will show that the functors e(d1,...,dn) categorify the elliptic Hall algebra introduced in [4]. Our main tool is the commutation relation between the func- tors (3.7) that we stipulated in Theorem 1.4. To set up the relation, let us consider the compositions (we will often writeek =e(k)) DM ek−→DM×S2 e(d1,...,dn)×IdS2 −−−−−−−→DM×S1×S2 (3.9) DM e(d1,...,dn) −−−−→DM×S1 ek×IdS1 −−−→DM×S1×S2 (3.10) (3.9) (3.10) and denote them bye(d1,...,dn) ◦ek andek ◦e(d1,...,dn), respectively. where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. In formulas (3.9) and (3.10), we set S1 = S2 = S, but we use different notations for the two copies of S to emphasize the fact that the functore(d1,...,dn) takes values in the ﬁrst factor of S × S, while the functor ek takes values in the second factor. and denote them bye(d1,...,dn) ◦ek andek ◦e(d and denote them bye(d1,...,dn) ◦ek andek ◦e(d1,...,dn), respectively. In formulas (3.9) and (3.10), we set S1 = S2 = S, but we use different notations for the two copies of S to emphasize the fact that the functore(d1,...,dn) takes values in the ﬁrst factor of S × S, while the functor ek takes values in the second factor. Proof of Theorem 1.4. — We will begin with the case n = 1, then n = 2 and ﬁnally general n. The fact that the ﬁber square (2.38) is derived, combined with Proposition 3.2, implies that the compositionsed ◦ek andek ◦ed are given by the following correspondences, respectively: Ld 1Lk 2 Z(x,y) p− p+×px S×py S M × S × S M Lk 1Ld 2 Z(y,x) p− p+×px S×py S M × S × S M M M M A. NEGUT, Above, we write px S,p y S for the maps Z2 →S that remember the points x,y, respectively. The way to interpret the diagrams above as functors is the following: start from DM, pull-back to Z2, tensor by the line bundle on top, and then push-forward to DM×S×S. For example, in the case of the diagram on the left,ed ◦ek equals DM×S×S (p+×px S×py S)∗ ←−−−−−−Db(Coh(Z(x,y))) ⊗Ld 1Lk 2 ←−−−Perf(Z(x,y)) p∗ − ←−DM Note that the middle arrow consists of tensoring with a line bundle, followed by the fully faithful functor (3.1). This aspect is a necessary technicality, because the scheme Z2 is not smooth (in the case at hand, we could have gotten away with only using the category Db(Coh) since the map p−is an l.c.i. morphism, but we will apply the notation above in situations where the morphism which plays the role of p−will not be l.c.i.). Consider the scheme Y, together with the maps (2.46) and (2.47): Y π↑ π↓ Z(x,y) Z(y,x) Y π↑ π↓ Z(x,y) Z(y,x) Y π↑ π↓ Z(x,y) Z(y,x) Z(y,x) Because of Proposition 2.45, we have R•π ↑ ∗(OY) = OZ(x,y) and R•π ↓ ∗(OY) = OZ(y,x) in the derived category. where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. Therefore, the compositionsed ◦ek andek ◦ed can be alternatively given by the following correspondences: Ld 1Lk 2 Y M × S × S M L′ 1 kL′ 2 d Y M × S × S M M M M M × S × S (the map that points left, respectively right, remembers the sheaf F0, respectively F2 in the notation of (2.40)). Recall from (2.44) the identity L1L2 = L′ 1L′ 2 on Y. The map of line bundles L1 →L′ 2 of (2.55) gives rise to maps (the map that points left, respectively right, remembers the sheaf F0, respectively F2 in the notation of (2.40)). Recall from (2.44) the identity L1L2 = L′ 1L′ 2 on Y. The map of line bundles L1 →L′ 2 of (2.55) gives rise to maps (3.11) ⎧ ⎪⎨ ⎪⎩ Ld 1Lk 2 →L′ 1 kL′ 2 d if d > k Ld 1Lk 2 ←L′ 1 kL′ 2 d if d < k Ld 1Lk 2 ∼= L′ 1 kL′ 2 d if d = k (3.11) which induces a natural transformation between the functorsed ◦ek andek ◦ed, as required in item (2) of Theorem 1.4. In general, the cone of a composition of maps in any trian- gulated category has a ﬁltration with associated graded object given by the direct sum of which induces a natural transformation between the functorsed ◦ek andek ◦ed, as required in item (2) of Theorem 1.4. 2 The analogous statement in an abelian category would be that, given a ﬂag of objects V0 ⊂··· ⊂Vn, the quotient Vn/V0 has a ﬁltration with associated graded object given by the direct sum of the individual quotients Vi/Vi−1. where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. In general, the cone of a composition of maps in any trian- gulated category has a ﬁltration with associated graded object given by the direct sum of HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES the cones of the individual maps.2 Thus, the cone of the maps denoted by →and ←in (3.11) has a ﬁltration with associated graded object given by the sum of the cones the cones of the individual maps.2 Thus, the cone of the maps denoted by →and ←in (3.11) has a ﬁltration with associated graded object given by the sum of the cones ⎧ ⎪⎪⎪⎪⎪⎨ ⎪⎪⎪⎪⎪⎩ La+1−k 1 L′ 2 d−a−1 →La−k 1 L′ 2 d−a ⊗(L1L2)k with a ∈{k,...,d −1} if d > k L′ 1 a−dLk−a 2 ←L′ 1 a−d+1Lk−a−1 2 ⊗(L1L2)d with a ∈{d,...,k −1} if d < k By Proposition 2.39, the cone of the maps denoted by →and ←above is precisely the line bundle La 1Ld+k−a 2 on the scheme Z(x,x), where a goes over the indexing sets featured in either of the two situations above. The correspondence La 1Ld+k−a 2 Z(x,x) p− p+×px S M × S × S M × S M M × S × S M × S M is precisely ∗◦e(a,d+k−a), and this implies item (3) of Theorem 1.4. The proof of Theo- rem 1.4 in the case n = 1 is now complete. is precisely ∗◦e(a,d+k−a), and this implies item (3) of Theorem 1.4. The proof of Theo- rem 1.4 in the case n = 1 is now complete. Let us now deal with the case n = 2. The fact that the ﬁber squares in Propo- sition 2.30 are derived, combined with Proposition 3.2, implies that the compositions g0 =e(d1,d2) ◦ek and g2 =ek ◦e(d1,d2) are given by the following correspondences, respec- tively: Ld1 1 Ld2 2 Lk 3 Z(x,x,y) p− p+×px S×py S M × S × S M Lk 1Ld1 2 Ld2 3 Z(y,x,x) p− p+×px S×py S M × S × S M The maps denoted by p+ and p−only remember the sheaves F0 and F3, respectively, from points (F0 ⊂··· ⊂F3) of either Z(x,x,y) or Z(y,x,x). We will also consider the functor M M M M × S × S M The maps denoted by p+ and p−only remember the sheaves F0 and F3, respectively, from points (F0 ⊂··· ⊂F3) of either Z(x,x,y) or Z(y,x,x). where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. We will also consider the functor A. NEGUT, g1 : DM →DM×S×S given by the following correspondence g1 : DM →DM×S×S given by the following correspondence Ld1 1 Lk 2Ld2 3 Z(x,y,x) p− p+×px S×py S M × S × S M Ld1 1 Lk 2Ld2 3 Z(x,y,x) p− p+×px S×py S M × S × S M M M M × S × S Because of Proposition 2.45, the functors g0 and g1 can be given by the following corre- spondences, respectively: Ld1 1 Ld2 2 Lk 3 Y+ M × S × S M and Ld1 1 L′ 2 kL′ 3 d2 Y+ M × S × S M Ld1 1 L′ 2 kL′ 3 d2 Y+ M × S × S M while the functors g1 and g2 can be given by the following correspondences, respectively: L′ 1 kL′ 2 d1Ld2 3 Y− M × S × S M Ld1 1 Lk 2Ld2 3 Y− M × S × S M and L′ 1 kL′ 2 d1Ld2 3 Y− M × S × S M L′ 1 kL′ 2 d1Ld2 3 Y− M × S × S M M Depending on whether d2 > k or d2 < k or d2 = k, we obtain natural transformations g0 →g1 or g0 ←g1 or g0 ∼= g1 by using the map (2.55) on the space Y+ (cf. (3.11)). Similarly, depending on whether d1 > k or d1 < k or d1 = k, we obtain natural transformations g1 →g2 or g1 ←g2 or g1 ∼= g2 by using the map (2.55) on the space Y−. This establishes item (2) of Theorem 1.4. To prove item (3), one needs to proceed as in the n = 1 case, and we will present the detailed argument for the natural transformation g0 →g1 in the case d2 > k. where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. The cone of this natural transformation arises from the cone of the map of HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES line bundles line bundles line bundles line bundles Ld1 1 ⊗ Ld2 2 Lk 3 →L′ 2 kL′ 3 d2 on Y+, which in turn has a ﬁltration whose associated graded object is the direct sum of the cones on Y+, which in turn has a ﬁltration whose associated graded object is the direct sum of the cones Ld1 1 La 2Lk 3L′ 3 d2−a ⊗ L2 L′ 3 →O as a ∈{k,...,d2 −1} (3.12) The analogue of Proposition 2.39 for Y+ instead of Y implies that the cone of the map (3.12) is the structure sheaf of the subscheme Z(x,x,x) →Y+ given by the condition x = y and F2 = F ′ 2. Therefore, the cone in (3.12) is the correspondence Ld1 1 La 2Ld2+k−a 3 on Z(x,x,x). Since the ﬁber square in Proposition 2.29 is derived, the functor DM →DM×S×S induced by this correspondence is ∗◦e(d1,a,d2+k−a), precisely as stipulated in item (3) of Theorem 1.4. Now that we have established Theorem 1.4 for n = 1 and n = 2, let us tackle the case of general n. Consider the functors g0 =e(d1,...,dn) ◦ek and gn =ek ◦e(d1,...,dn). For any i ∈{1,...,n −1} let gi denote the following composition: DM p∗ − DZ1 ρ∗ − DZ1 ... DZ1 ⊗Ldn−i+1 DZ1 π+∗π∗ − DZ1 ⊗Ldn Perf(Z(x,y,x)) ⊗Lk mid Db(Coh(Z(x,y,x))) ρ+∗ DZ1 (p+×pS)∗ DZ1 ⊗Ld1 DZ1 ... DZ1 π+∗π∗ − DZ1 ⊗Ldn−i DM×S Perf(Z(x,y,x)) Db(Coh(Z(x,y,x))) A. NEGUT, where we consider the maps ρ± and the line bundle Lmid as in the following diagram: Lmid Z(x,y,x) ρ+ ρ− Z1 Z1 (S,F2/F1) F0 ⊂x F1 ⊂y F2 ⊂x F3 F0 ⊂x F1 F2 ⊂x F3 Lmid Z(x,y,x) ρ+ ρ− Z1 Z1 (S,F2/F1) F0 ⊂x F1 ⊂y F2 ⊂x F3 F0 ⊂x F1 F2 ⊂x F3 Z1 Z1 The required statements about the natural transformations g0 ↔g1 and gn−1 ↔gn are proved just like in the case n = 2 in the previous paragraph, so we will not review them. Similarly, the required statements about the natural transformations gi−1 ↔gi are all proved just like in the case n = 3 and i = 2, which we will now explain. where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. According to the diagram above, g1 is given by the correspondence The required statements about the natural transformations g0 ↔g1 and gn−1 ↔gn are proved just like in the case n = 2 in the previous paragraph, so we will not review them. Similarly, the required statements about the natural transformations gi−1 ↔gi are all proved just like in the case n = 3 and i = 2, which we will now explain. According to the diagram above, g1 is given by the correspondence Lk mid Z(x,x) π− Ld2 Z(x,y,x) ρ+ ρ− Ld3 Z1 Z1 p− M Lk mid Ld1 Z(x,x) π+ π− Ld2 Z(x,y,x) ρ+ ρ− Ld3 Z1 p+×px S×py S Z1 Z1 p− M × S × S M Z1 M × S × S One should read the diagram above as inducing a functor DM →DM×S×S by composing the pull-back and push-forward functors corresponding to the solid arrows, read right- to-left, all the while at every step tensoring by the line bundle which is displayed with a dotted line above every space. Similarly, g2 is given by the correspondence One should read the diagram above as inducing a functor DM →DM×S×S by composing the pull-back and push-forward functors corresponding to the solid arrows, read right- to-left, all the while at every step tensoring by the line bundle which is displayed with a dotted line above every space. Similarly, g2 is given by the correspondence Lk mid Ld1 Z(x,y,x) ρ+ ρ− Ld2 Z(x,x) π+ π− Ld3 Z1 p+×px S×py S Z1 Z1 p− M × S × S M M × S × S HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Because the ﬁber squares in Proposition 2.33 are derived, we may invoke Proposition 3.2 in order to conclude that g1 and g2 are given by the following correspondences: Ld1 1 Lk 2Ld2 3 Ld3 4 Z(x,y,x,x) M × S × S M Ld1 1 Ld2 2 Lk 3Ld3 4 Z(x,x,y,x) M × S × S M and Ld1 1 Lk 2Ld2 3 Ld3 4 Z(x,y,x,x) M × S × S M Z(x,x,y,x) M M M respectively. However, Proposition 2.45 applied to Y−+ implies that we may replace the line bundles above on Z(x,x,y,x) and Z(x,y,x,x) by their pullbacks to Y−+ without changing the correspondences. Therefore, the required natural transformation between g1 and g2 is induced by a map of line bundles respectively. where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. However, Proposition 2.45 applied to Y−+ implies that we may replace the line bundles above on Z(x,x,y,x) and Z(x,y,x,x) by their pullbacks to Y−+ without changing the correspondences. Therefore, the required natural transformation between g1 and g2 is induced by a map of line bundles Ld1 1 Ld2 2 Lk 3Ld3 4 and Ld1 1 L′ 2 kL′ 3 d2Ld3 4 on Y−+. The construction of this map, as well as the proof of the fact that its cone has the properties stipulated in item (3) of Theorem 1.4, are achieved by using the map (2.55) in the same way as we have already seen in the cases n = 1 and n = 2. □ on Y−+. The construction of this map, as well as the proof of the fact that its cone has the properties stipulated in item (3) of Theorem 1.4, are achieved by using the map (2.55) in the same way as we have already seen in the cases n = 1 and n = 2. □ 3.7. At the level of K-theory, the functors (3.7) and (3.8) give rise to maps (3.13) e(d1,...,dn) : KM →KM×S (3.14) f(d1,...,dn) : KM →KM×S for all d1,...,dn ∈Z. We also consider the power series of maps for all d1,...,dn ∈Z. We also consider the power series of maps (3.15) h±(z) = ∞ k=0 h± k z±k : KM →KM×S[[z∓1]] (3.15) deﬁned as the composition deﬁned as the composition h±(z) : KM pull-back −−−−→KM×S multiplication by ∧•U(q−1−1)/z −−−−−−−−−−−−−−−−→KM×S h±(z) : KM pull-back −−−−→KM×S multiplication by ∧•U(q−1−1)/z −−−−−−−−−−−−−−−−→KM×S Above, recall that U is the universal sheaf on M × S, and q = [ωS] ∈KS (we abuse notation and use the symbol q also for the pull-back of the canonical line bundle to KM×S). The meaning of the wedge power is the following. Proposition 2.14 allows one to write [U] = [V] −[W] ∈KM×S, where V and W are locally free sheaves. Since the total A. where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. By a simple diagram chase, we have eded′ !!! = (p+ × pS)∗ Ld · (p−× pS)∗◦(p+ × pS)∗(Ld′ · p∗ −) By Propositions 2.27 and 3.4 (the latter Proposition applies because all the maps in (3.20) are closed embeddings followed by smooth maps), we have (p−× pS)∗◦(p+ × pS)∗= π+∗ " 1 −L2q L1 · π ∗ − # Therefore, we have Therefore, we have eded′ !!! = (p+ × pS)∗ $ Ld · π+∗ " 1 −L2q L1 · π ∗ − Ld′ · p∗ − #% □ and it is easy to see that the right-hand side is e(d,d′) −q · e(d−1,d′+1). and it is easy to see that the right-hand side is e(d,d′) −q · e(d−1,d′+1). □ As an immediate consequence of Theorem 1.4, we conclude that the operators (1.11) satisfy the identities (1.12) of operators KM →KM×S×S, as well as the opposite Lie bracket relations for the commutator of f(d1,...,dn) with fk. where ∧•(L) = rank L i=0 (−1)i[∧i(L)] for the excess bundle L. NEGUT, exterior power is supposed to be multiplicative, we deﬁne exterior power is supposed to be multiplicative, we deﬁne ∧• U(q−1 −1) z = ∧• V zq ∧• W z ∧• V z ∧• W zq ∈KM×S(z), ∧• U(q−1 −1) z = ∧• V zq ∧• W z ∧• V z ∧• W zq ∈KM×S(z), where ∧• V z = rank V i=0 [∧iV] (−z)i We showed in [27] that the operators (3.13), (3.14), (3.15) satisfy the relations We showed in [27] that the operators (3.13), (3.14), (3.15) satisfy the relations (3.16) ∧• −w z · O ∞ k=−∞ ek zk h±(w) = ∧• −z w · O h±(w) ∞ k=−∞ ek zk (3.17) ∧• −z w · O ∞ k=−∞ fk zk h±(w) = ∧• −w z · O h±(w) ∞ k=−∞ fk zk (3.18) ∞ k=−∞ ek zk , ∞ l=−∞ fl wl = ∞ k=−∞ zk wk ∗◦ h+(z) −h−(w) 1 −q−1 (3.18) where O is the K-theory class of the diagonal inside S × S. We will now work out some relations among the operators (3.13) and (3.14). Proposition 3.8. — We have the following identities of operators KM →KM×S: Proposition 3.8. — We have the following identities of operators KM →KM×S: (3.19) e(d1,...,dn)e(d′ 1,...,d′ n′) !!! = e(d1,...,dn,d′ 1,...,d′ n′) −q · e(d1,...,dn−1,d′ 1+1,...,d′ n′) (3.19) for all d1,...,dn,d′ 1,...,d′ n′ ∈Z. The analogous formula to (3.19) holds with the e’s replaced by f ’s, once one replaces the product in the LHS by the opposite product. for all d1,...,dn,d′ 1,...,d′ n′ ∈Z. The analogous formula to (3.19) holds with the e’s replaced by f ’s, once one replaces the product in the LHS by the opposite product. Proof. — We will prove (3.19) for n = n′ = 1, and then the general case will be clear, in light of the fact that the operators (3.13) arise as successive push-forwards and pull-backs between Z1 and Z• 2 (cf. the diagram after (1.8)). Consider the diagram (3.20) Z• 2 π− π+ Z1 p+×pS p− M Z1 p−×pS p+×pS M × S M × S (3.20) HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES ation as in Section 1.3. By a simple diagram chase, we have with the notation as in Section 1.3. 4. The A action 4.1. We will now recall the algebraic structure that governs the composition of the operators (3.13), (3.14), (3.15). Our presentation will be an adaptation of [4] and [30]. Consider two formal parameters q1 and q2, let q = q1q2 and deﬁne [n] = 1 + q−1 + ··· + q−n+1 Throughout this section, we will often encounter the ring (4.1) K = Z[q±1 1 ,q±1 2 ] Sym ([1],[2],[3],...) (4.1) (4.1) where Sym refers to Laurent polynomials which are symmetric in q1 and q2. where Sym refers to Laurent polynomials which are symmetric in q1 and q2. Deﬁnition 4 2 ([30]) Consider the K algebra A with generators Deﬁnition 4.2 ([30]). — Consider the K-algebra A with generators Deﬁnition 4.2 ([30]). — Consider the K-algebra A with generators {Ek,Fk,H± l }k∈Z,l∈N modulo the fact that the H± l ’s all commute, and the following relations (4.2) (z −wq1)(z −wq2) z −w q E(z)E(w) = = z −w q1 z −w q2 z −wq E(w)E(z) (4.2) A. NEGUT, (4.3) (z −wq1)(z −wq2) z −w q E(z)H±(w) = = z −w q1 z −w q2 z −wq H±(w)E(z) (4.4) [[Ek+1,Ek−1],Ek] = 0 ∀k ∈Z A. NEGUT, , (4.3) (z −wq1)(z −wq2) z −w q E(z)H±(w) = = z −w q1 z −w q2 z −wq H±(w)E(z) (4.3) (z −wq1)(z −wq2) z −w q E(z)H±(w) = (4.3) = z −w q1 z −w q2 z −wq H±(w)E(z) (4.4) [[Ek+1,Ek−1],Ek] = 0 ∀k ∈Z (4.4) together with the opposite relations for F(z) instead of E(z), as well as (4.5) [E(z),F(w)] = δ z w (1 −q1)(1 −q2) H+(z) −H−(w) 1 −q−1 (4.5) where where E(z) = k∈Z Ek zk , F(z) = k∈Z Fk zk , H±(z) = ∞ l=0 H± l z±l H+ 0 = 1 and H− 0 = c−1, and note that c is a central element of A. We will set H+ 0 = 1 and H− 0 = c−1, and note that c is a central element of A. We will set H+ 0 = 1 and H− 0 = c−1, and note that c is a central element of A. 4.3. The algebra A is known by many names, including the Ding-Iohara-Miki al- gebra ([8, 22]), the double shufﬂe algebra ([12]), the stable limit of trigonometric DAHA ([7]), and many other incarnations in representation theory and mathematical physics. However, the main description we will be concerned with is that of the elliptic Hall alge- bra of [4]. Speciﬁcally, the following is the main theorem of [30]. Theorem 4.4. (4.8) Formula (4.8) shows how, after inverting (1 −q1)(1 −q2), any En,k for gcd(n,k) = 1 can be obtained from successive commutators of E±1,k (there is a similar formula in [4] which deals with the non coprime case, see [28] for our conventions). Formula (4.8) shows how, after inverting (1 −q1)(1 −q2), any En,k for gcd(n,k) = 1 can be obtained from successive commutators of E±1,k (there is a similar formula in [4] which deals with the non coprime case, see [28] for our conventions). Remark 4.5. — Another important case of (4.6) is when (n,k) and (n′,k′) are pro- portional. In order to state the relation, let us set for any coprime integers n,k exp − ∞ s=1 Pns,ks sxs = 1 + ∞ s=1 Ens,ks (−x)s Then the elements Pn,k ∈A deﬁned by the formula above are multiples of the generators considered in [4] (see [26, Section 2.6] for a formula of the precise multiple). In terms of the Pn,k’s, formula (4.6) when (n,k) and (n′,k′) are proportional reads (4.9) [Pn,k,Pn′,k′] = δ0 n+n′(1 −qs 1)(1 −qs 2) · s(−sign n)(1 −c−\|n\|) 1 −q−s (4.9) where s = gcd(n,k). In all representations of A considered in the present paper, we will set c = qr for a natural number r (called the “level” of the representation) and therefore the fraction in (4.9) is an element of K. As a consequence of this fact, we conclude that the algebra A contains a deformed Heisenberg algebra for each slope k n. 4.6. In [24, Proposition 6.2], we deﬁned elements (albeit denoted by Xm1,...,m 4.6. In [24, Proposition 6.2], we deﬁned elements (albeit denoted by Xm1,...,mk) (4.10) E(d1,...,dn) d1,...,dn∈Z ∈A 4.10) E(d1,...,dn) d1,...,dn∈Z ∈A (4.10) for all d1,...,dn ∈Z. (4.1) — Deﬁne the elliptic Hall algebra ([4], see [28, Theorem 3.5] for our conven- tions) to be generated by elements {En,k}(n,k)∈Z2\(0,0), modulo the relations (4.6) [En,k,En′,k′] = (1 −q1)(1 −q2) ⎛ ⎝ t≥1, t i=1 ni=n+n′, t i=1 ki=k+k′ (n′,k′)↷(n1,k1)↷...↷(nt,kt)↷(n,k) En1,k1 ...Ent,kt · coeff ⎞ ⎠ (4.6) for any (n′,k′)↷(n,k) (the notation ↷means that (n,k) can be reached clockwise from (n′,k′) without crossing the negative y axis) and certain scalars coeff ∈K. Then upon localization with respect to the element (1−q1)(1−q2) ∈K, the algebra A becomes isomorphic to the elliptic Hall algebra via the assignment for any (n′,k′)↷(n,k) (the notation ↷means that (n,k) can be reached clockwise from (n′,k′) without crossing the negative y axis) and certain scalars coeff ∈K. for any (n′,k′)↷(n,k) (the notation ↷means that (n,k) can be reached clockwise from (n′,k′) without crossing the negative y axis) and certain scalars coeff ∈K. Then upon localization with respect to the element (1−q1)(1−q2) ∈K, the algebra A becomes isomorphic to the elliptic Hall algebra via the assignment crossing the negative y axis) and certain scalars coeff ∈K. Then upon localization with respect to the element (1−q1)(1−q2) ∈K, the algebra A becomes isomorphic to the elliptic Hall algebra via the assignment (4.7) E−1,k = Ek, E1,k = Fk, 1 + ∞ l=1 E0,±l (−zq)±l 1 + ∞ l=1 E0,±l (−z)±l = H±(z) (4.7) HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES We will henceforth abuse notation and refer to En,k as elements of A. If the triangle with vertices (0,0), (n,k), (n + n′,k + k′) has no lattice points inside and on any of the edges, then (4.6) takes the following simple form: (4.8) [En,k,En′,k′] = −(1 −q1)(1 −q2)En+n′,k+k′ (4.8) It is an easy exercise to show that we have the following property for all d1,...,dn,d′ 1,...,d′ n′ ∈Z (see [24, Proposition 6.5]): (4.11) E(d1,...,dn)E(d′ 1,...,d′ n′) = E(d1,...,dn,d′ 1,...,d′ n′) −q · E(d1,...,dn−1,d′ 1+1,...,d′ n′) Meanwhile, we have for all (n,k) ∈N × Z (see [28, Proposition 2.3]): (4.11) E(d1,...,dn)E(d′ 1,...,d′ n′) = E(d1,...,dn,d′ 1,...,d′ n′) −q · E(d1,...,dn−1,d′ 1+1,...,d′ n′) (4.11) n n Meanwhile, we have for all (n,k) ∈N × Z (see [28, Proposition 2.3]): (4.12) E−n,k = qgcd(n,k)−1E(d(n,k) 1 ,...,d(n,k) n ) where (4.12) E−n,k = qgcd(n,k)−1E(d(n,k) 1 ,...,d(n,k) n ) (4.13) d(n,k) i = ki n − k(i −1) n + δn i −δ1 i A. NEGUT, for all i. As shown in the proof of Proposition 2.3 of loc. cit., any E(d1,...,dn) can be written as a linear combination of products of E−n,k’s using relation (4.11). for all i. As shown in the proof of Proposition 2.3 of loc. cit., any E(d1,...,dn) can be written as a linear combination of products of E−n,k’s using relation (4.11). Remark 4.7. — Analogously, one can deﬁne elements F(d1,...,dn) ∈A instead of E(d1,...,dn) by replacing E−n,k with En,k in (4.12). Remark 4.7. — Analogously, one can deﬁne elements F(d1,...,dn) ∈A instead of E(d1,...,dn) by replacing E−n,k with En,k in (4.12). The elements (4.10) were constructed using the isomorphism between A and a double shufﬂe algebra (see [24]), and this allows us to prove the following. Proposition 4.8. — For any d1,...,dn,k ∈Z, we have (note that E(k) = Ek) Proposition 4.8. — For any d1,...,dn,k ∈Z, we have (note that E(k) = Ek) (4.14) [E(d1,...,dn),Ek] = (1 −q1)(1 −q2) n i=1 ⎧ ⎪⎨ ⎪⎩ − k≤a<di E(d1,...,di−1,a,di+k−a,di+1,...,dn) if di > k di≤a<k E(d1,...,di−1,a,di+k−a,di+1,...,dn) if di < k (4.14) There is no summand corresponding to any i for which di = k. Proposition 4.8 may be proved by embedding the negative half of A (i.e. the sub- algebra generated by {Ek}k∈Z) in the algebra Aq,t of [5] via the last formula of loc. cit. Alternatively, we recall that the elliptic Hall algebra A acts on the K-theory group of the moduli space of rank r framed sheaves on P2 ([11, 31], see [25] for our conventions). This statement may be interpreted as an equivariant version of the results in the present paper, and our proof carries through in the equivariant case. Therefore, (1.12) implies that relation (4.14) holds in the level r representation of A. Since any non-zero element of A acts non-trivially in some level r representation for r large enough, this implies that relations (4.14) hold in A. Proposition 4.9. — Relations (4.2) and (4.4) both follow from (4.14). Proposition 4.9. — Relations (4.2) and (4.4) both follow from (4.14). Proof. — Taking the coefﬁcient of z−mw−n in relation (4.2) shows it is equivalent to Proof. — Taking the coefﬁcient of z−mw−n in relation (4.2) shows it is equivalent to Em+3En− q1 + q2 + 1 q Em+2En+1+ 1 q1 + 1 q2 + q Em+1En+2−EmEn+3 = = EnEm+3− 1 q1 + 1 q2 + q En+1Em+2+ q1 + q2 + 1 q En+2Em+1−En+3Em = EnEm+3− 1 q1 + 1 q2 + q En+1Em+2+ q1 + q2 + 1 q En+2Em+1−En+3Em HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES for all m,n ∈Z. We may rewrite the relation above as for all m,n ∈Z. There is no summand corresponding to any i for which di = k. We may rewrite the relation above as " Em+3En −Em+2En+1 1 q + 1 + q + + Em+1En+2 1 q + 1 + q −EmEn+3 # − − " EnEm+3 −En+1Em+2 1 q + 1 + q + + En+2Em+1 1 q + 1 + q −En+3Em # = = (1 −q1)(1 −q2) En+1Em+2 q −En+2Em+1 −Em+2En+1 + Em+1En+2 q In the right-hand side, we may convert every EmEn into E(m,n) −qE(m−1,n+1) using relation (4.11), and so the formula above becomes equivalent to [Em+3,En] − 1 q + 1 + q [Em+2,En+1]+ + 1 q + 1 + q [Em+1,En+2] −[Em,En+3] = (1 −q1)(1 −q2)· " (E(n+1,m+2) + E(m+1,n+2)) 1 q + q −E(n,m+3)− + 1 q + 1 + q [Em+1,En+2] −[Em,En+3] = (1 −q1)(1 −q2)· " (E(n+1,m+2) + E(m+1,n+2)) 1 q + q −E(n,m+3)− −E(n+2,m+1) −E(m+2,n+1) −E(m,n+3) # It is easy to check that the formula above follows by applying (4.14) to the four Lie brack- ets in the left-hand side. As for (4.4), it follows from the fact that It is easy to check that the formula above follows by applying (4.14) to the four Lie brack- ets in the left-hand side. As for (4.4), it follows from the fact that [Ek−1,Ek+1] = E(k−1,k+1) + E(k,k) together with [E(k−1,k+1) + E(k,k),Ek] = [E(k−1,k+1),Ek] + [E(k,k),Ek] = E(k−1,k,k+1) −E(k−1,k,k+1) = 0 h are special cases of (4.14). □ □ both of which are special cases of (4.14). both of which are special cases of (4.14). □ both of which are special cases of (4.14). 4.10. We would like to compare the abstract algebra elements E(d1,...,dn) ∈A of the previous Subsections with the explicit maps e(d1,...,dn) : KM →KM×S from Section 3.7. The goal would be to say that we have an action of the algebra A on KM, but this A. NEGUT, statement requires care: taken literally, having an action means that any element of A gives rise to a homomorphism KM →KM. In formula (3.13), we see that this is not quite the case, so we must adapt the notion of “action”. Deﬁnition 4.11. — For any group homomorphisms x,y : KM →KM×S, deﬁne xy\| = KM y−→KM×S x×IdS −−→KM×S×S ∗ −→KM×S Deﬁnition 4.11. — For any group homomorphisms x,y : KM →KM×S, deﬁne xy\| = KM y−→KM×S x×IdS −−→KM×S×S ∗ −→KM×S where : M × S →M × S × S is IdM times the diagonal embedding. Also deﬁne [x,y] = $ KM y−→KM×S2 x×IdS2 −−−→KM×S1×S2 % − − $ KM x−→KM×S1 y×IdS1 −−−→KM×S1×S2 % where S1 = S2 = S, but we use different labels in order to emphasize the fact that x takes values in the ﬁrst factor of S and y takes values in the second factor. where S1 = S2 = S, but we use different labels in order to emphasize the fact that x takes values in the ﬁrst factor of S and y takes values in the second factor. The notions in Deﬁnition 4.11 satisfy associativity, in the sense that (4.15) (xy\|)z\| = x(yz\|)\| (4.15) as homomorphisms KM →KM×S. Therefore, we will unambiguously use the notation xyz\|. We also have the following version of the Jacobi identity as homomorphisms KM →KM×S. Therefore, we will unambiguously use the notation xyz\|. We also have the following version of the Jacobi identity (4.16) [[x,y],z] + [[y,z],x] + [[z,x],y] = 0 (4.16) as homomorphisms KM →KM×S×S×S. as homomorphisms KM →KM×S×S×S. Deﬁnition 4.12. — If two homomorphisms x,y : KM →KM×S satisfy (4.17) [x,y] = ∗◦z (4.17) for some homomorphism z : KM →KM×S, we will write for some homomorphism z : KM →KM×S, we will write for some homomorphism z : KM →KM×S, we will write (4.18) [x,y]red = z [x,y]red = z (4.18) Note that formula (4.18) is unambiguous, since ∗: KS →KS×S is injective (as it has a left inverse given by projection to the ﬁrst factor), and so z in (4.17) is unique. Note that formula (4.18) is unambiguous, since ∗: KS →KS×S is injective (as it has a left inverse given by projection to the ﬁrst factor), and so z in (4.17) is unique. Note that formula (4.18) is unambiguous, since ∗: KS →KS×S is injective (as it has a left inverse given by projection to the ﬁrst factor), and so z in (4.17) is unique. We have the following version of the Leibniz rule (4.19) [x,yz\|]red = [x,y]redz\| + y[x,z]red\| (4.19) (4.19) HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES which is an equality of homomorphisms KM →KM×S. Formula (4.19) follows from base change in the derived ﬁber square which is an equality of homomorphisms KM →KM×S. Deﬁnition 4.11. — For any group homomorphisms x,y : KM →KM×S, deﬁne Formula (4.19) follows from base change in the derived ﬁber square S S × S ×IdS S × S IdS× S × S × S (IdS × )∗◦( × IdS)∗= ∗◦∗ and we leave the details to the interested reader. Similarly, (4.16) implies (4.20) [[x,y]red,z]red + [[y,z]red,x]red + [[z,x]red,y]red = 0 as homomorphisms KM →KM×S. (4.20) 4.13. For the remainder of the paper, the parameters q1,q2 over which the alge- bra A is deﬁned will be reinterpreted such that q1 + q2 = [1 S], q = q1q2 = [ωS] While q1,q2 are not themselves well-deﬁned in KS, any symmetric Laurent polynomial in q1,q2 is indeed well-deﬁned. In other words, we obtain a ring homomorphism While q1,q2 are not themselves well-deﬁned in KS, any symmetric Laurent polynomial in q1,q2 is indeed well-deﬁned. In other words, we obtain a ring homomorphism K →KS K →KS where K is the ring (4.1). In order for the ring homomorphism above to be well-deﬁned, we need the quantity 1 + q−1 + ··· + q−n+1 to be invertible in KS, and the solution to achieving this is to work with K-theory with Q coefﬁcients (indeed, q is unipotent in KS, so 1 + q−1 + ··· + q−n+1 is equal to n times a unit). Deﬁne [X,Y]red = [X,Y] (1 −q1)(1 −q2) (4.21) [X,Y]red = [X,Y] (1 −q1)(1 −q2) (4.21) (4.21) for all X,Y ∈A, and note that the right-hand side is well-deﬁned due to the fact that all commutators in the algebra A are multiples of (1 −q1)(1 −q2), see (4.6). for all X,Y ∈A, and note that the right-hand side is well-deﬁned due to the fact that all commutators in the algebra A are multiples of (1 −q1)(1 −q2), see (4.6). Deﬁnition 4.14. — An action A ↷KM is an abelian group homomorphism Deﬁnition 4.14. — An action A ↷KM is an abelian group homomorphism (4.22) A −→Hom(KM,KM×S) (4.22) such that for all X,Y ∈A, we have such that for all X,Y ∈A, we have (4.23) (XY) = (X)(Y)\| (4.23) (XY) = (X)(Y)\| (4.24) [(X),(Y)]red = ([X,Y]red) (XY) = (X)(Y)\| A. NEGUT, The compatibility between relations (4.23) and (4.24) is visible when restricting the latter formula to the diagonal : M × S →M × S × S, as follows: (X)(Y)\| −(Y)(X)\| = ∗∗ " XY −YX (1 −q1)(1 −q2) # = (4.25) = XY −YX (1 −q1)(1 −q2) · ∧•(N ∨ :S→S×S) = (XY −YX) (4.25) Because the normal bundle N:S→S×S is isomorphic to the tangent bundle to S, hence the exterior algebra of its dual equals (1 −q1)(1 −q2) = [∧•(1 S)]. Because the normal bundle N:S→S×S is isomorphic to the tangent bundle to S, hence the exterior algebra of its dual equals (1 −q1)(1 −q2) = [∧•(1 S)]. Theorem 4.15. — There exists an action A ↷KM as in Deﬁnition 4.14, where Theorem 4.15. — There exists an action A ↷KM as in Deﬁnition 4.14, where (E(d1,...,dn)) = e(d1,...,dn) and (F(d1,...,dn)) = f(d1,...,dn) Moreover, the series H±(z) ∈A of (4.7) act by (3.15). (E(d1,...,dn)) = e(d1,...,dn) and (F(d1,...,dn)) = f(d1,...,dn) Moreover, the series H±(z) ∈A of (4.7) act by (3.15). Moreover, the series H±(z) ∈A of (4.7) act by (3.15). Proof. — Recall from [4] (see [28] for our notations) that a K-basis of A is given by Proof. — Recall from [4] (see [28] for our notations) that a K-basis of A is given by (4.26) Ev = En1,k1 ...Ent,kt Ev = En1,k1 ...Ent,kt (4.26) Ev = En1,k1 ...Ent,kt as v = {(n1,k1),...,(nt,kt)} goes over all convex paths of lattice points, i.e. Claim 4.16. — For any lattice points (n′,k′)↷(n,k), we have (4.29) [(En,k),(En′,k′)]red = = t≥1, t i=1 ni=n+n′, t i=1 ki=k+k′ (n′,k′)↷(n1,k1)↷...↷(nt,kt)↷(n,k) (En1,k1)...(Ent,kt) !!! · coeff (4.29) where the coefﬁcients in the right-hand side match the analogous ones in the right-hand side of (4.6), but with q1 and q2 interpreted as the Chern roots of 1 S. To conclude the proof of Theorem 4.15, we must show that relations (4.29) imply that satisﬁes properties (4.23) and (4.24); we will only take care of the former of these, as the latter is analogous and so left to the interested reader. We will use the “straightening” argument of [4]: one may deﬁne the elements (4.26) of A for an arbitrary path v of lattice points. The convexiﬁcation of v is the path consisting of the same collection of lattice points, but ordered in clockwise order of slope, starting from the negative y axis. The area of a path, denoted by a(v), is deﬁned as the area between the path v and its convexiﬁcation. Therefore, we will show that (4.30) (Ev) = (En1,k1)...(Ent,kt) !!! (4.30) holds for an arbitrary path v, by induction on a(v) (this claim will establish (4.23)). The base case is when a(v) = 0, i.e. v is already convex, in which case (4.30) is simply the deﬁnition of (Ev). A non-convex path v may be written as v = ...,(ni,ki),(ni+1,ki+1),... v = ...,(ni,ki),(ni+1,ki+1),... for some i such that ki/ni < ki+1/ni+1. Let v′ denote the path obtained from v by switching the lattice points (ni,ki) and (ni+1,ki+1). It is elementary to observe that a(v′) < a(v), and moreover the straightening argument of [4] shows that for some i such that ki/ni < ki+1/ni+1. Let v′ denote the path obtained from v by switching the lattice points (ni,ki) and (ni+1,ki+1). It is elementary to observe that a(v′) < a(v), and moreover the straightening argument of [4] shows that (4.31) Ev = Ev′ + (1 −q1)(1 −q2) v′′ Ev′′ · coeff (4.31) where the sum goes over paths with a(v′′) < a(v), and the coefﬁcients are induced from (4.6). Because of (4.29) and the induction hypothesis, we have where the sum goes over paths with a(v′′) < a(v), and the coefﬁcients are induced from (4.6). Because of (4.29) and the induction hypothesis, we have (4.32) ...(Eni,ki)(Eni+1,ki+1)... !!! (4.21) such that the slopes ki/ni are ordered clockwise starting from the negative y axis (formula (4.9) implies that En,k and En′,k′ commute if (n,k) ∈R>0(n′,k′), so we may order them arbitrarily). Thus, we deﬁne the action : A →Hom(KM,KM×S) by (E−n,k) = qgcd(n,k)−1e(d(n,k) 1 ,...,d(n,k) n ) of (3.13) (4.27) (En,k) = qgcd(n,k)−1f(d(n,k) 1 ,...,d(n,k) n ) of (3.14) (4.28) (4.28) for any n > 0, k ∈Z, where d(n,k) i are deﬁned in (4.13). Moreover, we have for any n > 0, k ∈Z, where d(n,k) i are deﬁned in (4.13). Moreover, we have (E0,k) : KM pull-back −−−−→KM×S multiplication by ∧kU −−−−−−−−−−→KM×S Then requirement (4.23) forces us to set Then requirement (4.23) forces us to set Then requirement (4.23) forces us to set (Ev) = (En1,k1)...(Ent,kt) !!! Together with K-linearity, this completely determines . Together with K-linearity, this completely determines . Together with K-linearity, this completely determines . HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Claim 4.16. — For any lattice points (n′,k′)↷(n,k), we have A. NEGUT, Proof of Claim 4.16. — Let us ﬁrst give a slightly imprecise argument. By Theo- rem 4.4, the commutation relations (4.6) can be deduced from relations (4.2)–(4.5). As the latter hold with Ek,Fk,H± k replaced by ek = (Ek), fk = (Fk), h± k = (H± k ) of Sec- tion 3.7 (this fact was proved in [27]), one might conclude the claim. The imprecision in the previous paragraph is that, strictly speaking, Theorem 4.4 requires one to invert the scalar (1 −q1)(1 −q2). When q1 and q2 are set equal to the Chern roots of 1 S, this scalar is manifestly a zero divisor, so inverting it is off the table. However, the ace up our sleeve is the operation (x,y) →[x,y]red for homomorphisms x,y : KM →KM×S, and its counterpart (4.21) for elements of A. We thus revisit the argument of [30], which consisted of the following main steps: (1) Deﬁne En,k ∈A as iterated reduced commutators (4.21) of E±1,l ∈A via (4.8) (and its generalization to the case when gcd(n,k) > 1, see [28]). In the geo- metric case, we emulate this procedure and deﬁne (En,k) : KM →KM×S as iterated reduced commutators (4.18) of (E±1,l) : KM →KM×S. (2) Show that relations (4.6) among the elements En,k ∈A thus constructed can be obtained from relations (4.2)–(4.5). While combinatorially quite involved, this part of the argument of [30] only uses the Lebniz rule and Jacobi identities. By (4.19) and (4.20), the same argument therefore applies equally well to show that relations (4.29) among the operators (En,k) : KM →KM×S can be ob- tained from the geometric counterparts of relations (4.2)–(4.5), which we know to hold. The only thing that remains to be proved is that the operators (En,k) constructed as in item (1) above match the operators in the right-hand sides of (4.27) and (4.28). This happens because relations (4.11) and (4.14) in A are matched by relations (3.19) and (1.12) between operators KM →KM×S. □ Claim 4.16. — For any lattice points (n′,k′)↷(n,k), we have = (Ev′) + ∧•(N ∨ →S×S) v′′ (Ev′′) · coeff (4.32) with the coefﬁcients in the RHS being the same ones as in (4.31) (indeed, the two com- putations (4.31) and (4.32) take as inputs (4.6) and (4.29), respectively, so it should be no surprise that they produce the same coefﬁcients). Formula (4.32) yields the induction step of (4.30), which as we have seen establishes (4.23). A. NEGUT, 5. Dimension estimates 5.1. The main purpose of the present Section is to obtain certain estimates on the dimensions of various strata of the schemes Zλ of Deﬁnition 2.16, when n = \|λ\| ≤4. In what follows, the dimension of a scheme will refer to the maximum of the dimensions of its irreducible components. Consider the map Zλ π−→M × Sn which takes a ﬂag (2.16) to (Fn,x1,...,xn). The target of this map is smooth but dis- connected, with the dimension of its connected components depending linearly on the second Chern class of Fn. Thus, the phrase “dimension (resp. number of irreducible com- ponents) of Zλ” will refer to the dimension (resp. number of irreducible components) of π −1(C) for any ﬁxed connected component C of M × Sn. The analogous terminology will apply to locally closed subsets of Zλ. which takes a ﬂag (2.16) to (Fn,x1,...,xn). The target of this map is smooth but dis- connected, with the dimension of its connected components depending linearly on the second Chern class of Fn. Thus, the phrase “dimension (resp. number of irreducible com- ponents) of Zλ” will refer to the dimension (resp. number of irreducible components) of π −1(C) for any ﬁxed connected component C of M × Sn. The analogous terminology will apply to locally closed subsets of Zλ. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES 5.2. Any stable sheaf F of rank r > 0 on a smooth surface is torsion-free, hence it injects into its double dual F →F ∨∨. The double dual is a reﬂexive sheaf, hence locally free (over a smooth surface). Therefore, we have an injection (5.1) F →V F →V (5.1) with V locally free. Assumption A implies that F is stable iff V is stable (see [27, Propo- sition 5.5]), and since stable sheaves over an algebraically closed ﬁeld are simple, this implies that the inclusion (5.1) is unique up to constant multiple. The quotient V/F is supported at ﬁnitely many distinct closed points x1,...,xk ∈S, hence (5.2) V/F ∼= Q1 ⊕··· ⊕Qk (5.2) where Qi is a ﬁnite length sheaf supported at the closed point xi ∈S. The length di of the sheaf Qi will be called the defect of F at the point xi, since it measures how far F is from being locally free in the vicinity of xi. where Qi is a ﬁnite length sheaf supported at the closed point xi ∈S. The length di of the sheaf Qi will be called the defect of F at the point xi, since it measures how far F is from being locally free in the vicinity of xi. Deﬁnition 5.3. — Let Quotd denote the moduli space of quotients of a rank r locally free sheaf V, which have length d and are supported at a given closed point x ∈S. As the notation suggests, the variety Quotd does not depend (up to isomorphism) on the choices of V,x,S, as long as the latter is a smooth surface. Theorem 5.4 ([2, 9]). — If d > 0, Quotd is irreducible of dimension rd −1. Theorem 5.4 ([2, 9]). — If d > 0, Quotd is irreducible of dimension rd −1. In order to produce a stable sheaf F with prescribed defects at distinct points x1,...,xk ∈S, one must start from a stable locally free sheaf V with invariants (r,c1), and modify it according to a point in the irreducible algebraic variety Quotd1 × ··· × Quotdk. This means that the locally closed subscheme which parameterizes sheaves with defects d1,...,dk at given distinct points x1,...,xk (and no defect anywhere else) has codimension r(d1 + ··· + dk) + k in M × Sk. 5.5. Recall the schemes introduced in Deﬁnition 2.16, namely 5.5. Recall the schemes introduced in Deﬁnition 2.16, namely Zλ = (F0 ⊂p1 F1 ⊂p2 ··· ⊂pn Fn) such that pi = pj if i ∼j in λ (5.3) where λ is a set partition of size n. Assume that the set partition λ consists of k distinct parts with multiplicities n1,...,nk, and we will use the notation x1,...,xk for the points of S corresponding to the distinct parts of λ. All the sheaves in a ﬂag (5.3) have isomorphic double duals, hence they are all contained in one and the same locally free sheaf V. Therefore, we let (5.4) Zdef d1,...,dk λ ⊂Zλ (5.4) A. NEGUT, be the locally closed subscheme where the support points x1,...,xk ∈S of the ﬂag (5.3) are all distinct, and the sheaf Fn has defect di at the point xi for all i (note that Fn may have arbitrary defect at points different from x1,...,xk). Deﬁnition 5.6. — Let Quotd,...,d+n denote the moduli space of quotients V ↠Q ↠ ↠Q Deﬁnition 5.6. — Let Quotd,...,d+n denote the moduli space of quotients V ↠Qd+n ↠··· ↠Qd Deﬁnition 5.6. — Let Quotd,...,d+n denote the moduli space of quotients V ↠Qd+n ↠··· ↠Qd where V is locally free, and the sheaves Qd+n,...,Qd have lengths d + n,...,d, respectively, and are supported at one and the same given closed point x ∈S. where V is locally free, and the sheaves Qd+n,...,Qd have lengths d + n,...,d, respectively, and are supported at one and the same given closed point x ∈S. Theorem 5.4 ([2, 9]). — If d > 0, Quotd is irreducible of dimension rd −1. We will give an explicit construction of the schemes Quotd,...,d+n in the next sub- section, as well as some dimension estimates when n ≤4 (as opposed from the scheme in Deﬁnition 5.3, their dimensions grow wildly for large n, and they are far from being irreducible). The construction will not depend on V,x,S, and it will be clear that the same principle as in the previous subsection applies. Namely, in order to construct a ﬂag (5.3) which lies in the subscheme (5.4), one (1) chooses distinct points x1,...,xk ∈S (2) chooses a stable sheaf V which is locally free at x1,...,xk (3) chooses a point in the variety (3) chooses a point in the variety (3) chooses a point in the variety Quotd1,...,d1+n1 × ··· × Quotdk,...,dk+nk (ni denotes the multiplicity of xi in λ) which determines the sheaves F0,...,Fn as modiﬁcations of V in the vicinity of the chosen points x1,...,xk. More rigorously, the discussion above implies the following. More rigorously, the discussion above implies the following. Lemma 5.7. — The scheme Zdef d1,...,dk λ is locally isomorphic to Lemma 5.7. — The scheme Zdef d1,...,dk λ is locally isomorphic to M × Sk × Quotd1,...,d1+n1 × ··· × Quotdk,...,dk+nk As a consequence of Lemma 5.7, we obtain the following result (recall the conven- tions of Section 5.1 when discussing dimension and the number of irreducible compo- nents of locally closed subsets of Zλ, such as Zdef d1,...,dk λ ). Corollary 5.8. — For any sequence of defects d1,...,dk ≥0 and any set partition λ whose parts have multiplicities n1,...,nk ∈N, we have dimZdef d1,...,dk λ = const + r(cﬁrst 2 + clast 2 ) + k+ (5.5) dimZdef d1,...,dk λ = const + r(cﬁrst 2 + clast 2 ) + k+ (5.5) + k i=1 dimQuotdi,...,di+ni + 1 −2rdi −rni HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES where cﬁrst 2 , clast 2 are the second Chern classes of the sheaves labeled F0 and Fn in (5.3), and const is the same constant that appears in (2.10). Proof. Theorem 5.4 ([2, 9]). — If d > 0, Quotd is irreducible of dimension rd −1. — Because the second Chern class of the stable sheaf V (deﬁned in item (2) before the statement of Lemma 5.7) is c2 = clast 2 − k i=1 di = cﬁrst 2 − k i=1 (di + ni) the dimension of the moduli space parameterizing V is the dimension of the moduli space parameterizing V is const + r(cﬁrst 2 + clast 2 ) −r k i=1 (2di + ni) The choice of the support points x1,...,xk adds 2k dimensions, and the choice of ﬂag F0 ⊂··· ⊂Fn ⊂V with the given defects adds k i=1 dimQuotdi,...,di+ni dimensions to our estimate (according to item (3) above Lemma 5.7). □ 5.9. Let us now describe the schemes Quotd and Quotd,...,d+n explicitly. Since these do not depend on the choice of the locally free sheaf V, or of a closed point x on a smooth surface S, we may as well consider V = O⊕r, S = A2 and x = (0,0). Therefore, we have Quotd = O⊕r A2 ↠Q,length Q=d,supp Q=(0,0) Quotd,...,d+n= O⊕r A2 ↠Qd+n ↠··· ↠Qd,length Qi=i,supp Qi=(0,0) Quotd,...,d+n= O⊕r A2 ↠Qd+n ↠··· ↠Qd,length Qi=i,supp Qi=(0,0) As vector spaces, we may consider isomorphisms Qi ∼= Ci, and the OA2 = C[x,y] module structure on Qi can be packaged by providing two nilpotent commuting endomorphisms X,Y ∈End(Ci). A surjective homomorphism O⊕r A2 ↠Qi is the same datum as r vectors v1,...,vr ∈Ci which are cyclic. Recall that cyclicity means that the vector space Ci is generated by polynomials in X and Y acting on linear combinations of v1,...,vr. In other words, let us construct the i × r matrix v = (v1,...,vr) ∈Hom(Cr,Ci) and let End0(Ci) denote the variety of nilpotent endomorphisms of Ci. Then and let End0(Ci) denote the variety of nilpotent endomorphisms of Ci. Then Quotd = (X,Y,v) ∈End0(Cd) × End0(Cd)× × Hom(Cr,Cd) s.t. [X,Y] = 0, v cyclic /G × Hom(Cr,Cd) s.t. [X,Y] = 0, v cyclic /G × Hom(Cr,Cd) s.t. [X,Y] = 0, v cyclic /G A. NEGUT, where we take the quotient by the G = GLd action in order to factor out the ambiguity of the chosen isomorphism Q ∼= Cd. Explicitly, g ∈G acts by where we take the quotient by the G = GLd action in order to factor out the ambiguity of the chosen isomorphism Q ∼= Cd. Theorem 5.4 ([2, 9]). — If d > 0, Quotd is irreducible of dimension rd −1. Explicitly, g ∈G acts by (X,Y,v) ⇝(gXg−1,gYg−1,gv) and the cyclicity of v implies that the G action is free. The quotient by G is geomet- ric, because the fact that we restrict to the open subscheme of cyclic triples (X,Y,v) is precisely equivalent to restricting to the locus of stable points (as in geometric invariant theory) with respect to the inverse of the determinant character. Similarly, and the cyclicity of v implies that the G action is free. The quotient by G is geomet- ric, because the fact that we restrict to the open subscheme of cyclic triples (X,Y,v) is precisely equivalent to restricting to the locus of stable points (as in geometric invariant theory) with respect to the inverse of the determinant character. Similarly, Quotd,...,d+n = (X,Y,v) ∈End0(Cd+n ↠··· ↠Cd)× (5.6) × End0(Cd+n ↠··· ↠Cd)× × Hom(Cr,Cd+n) s.t. [X,Y] = 0, v cyclic /P (5.6) where End0(Cd+n ↠··· ↠Cd) denotes the variety of nilpotent endomorphisms of Cd+n which preserve a ﬁxed ﬂag of quotients of the speciﬁed dimensions (and P ⊂GLd+n de- notes the subgroup of automorphisms with the same property). If one is apprehensive about taking the quotient of a quasiprojective algebraic variety by the non-reductive group P, then one must employ the following trick. Instead of ﬁxing a ﬂag of quotients, let it vary and replace the datum (X,Y,v) by (X,Y,v,Cd+n ↠Vd+n−1 ↠··· ↠Vd) where dimVi = i and the maps X,Y ∈End0(Cd+n) are required to preserve the ﬂag {Vi}d≤i<d+n. Then the P quotient should be replaced by a GLd+n quotient. We will refrain from making this modiﬁcation, so as to keep the presentation clear. 5.10. We ﬁx a basis of Ci so that the endomorphism X takes the form 5.10. We ﬁx a basis of Ci so that the endomorphism X takes the form 5.10. We ﬁx a basis of Ci so that the endomorphism X takes the form 5.10. We ﬁx a basis of Ci so that the endomorphism X takes the form (5.7) ⎛ ⎜⎜⎜⎜⎝ ∗ ∗ 0 0 0 ∗ ∗ 0 0 0 ∗ ∗ 0 0 0 ∗ ∗ ∗ 0 0 ∗ ∗ ∗ ∗ 0 ⎞ ⎟⎟⎟⎟⎠ = X0 0 x X′ (5.7) where the block in the top left is a d × d matrix, and the lower triangular block in the bottom right is n × n. We will use the notations X0,x,X′ for the three blocks, as pictured above, and employ the analogous notations for Y. By analogy with the constructions in Section 5.9, we deﬁne the afﬁne algebraic variety where the block in the top left is a d × d matrix, and the lower triangular block in the bottom right is n × n. We will use the notations X0,x,X′ for the three blocks, as pictured above, and employ the analogous notations for Y. By analogy with the constructions in Section 5.9, we deﬁne the afﬁne algebraic variety Commn = (X,Y) ∈End0(Cn ↠··· ↠C1)× × End0(Cn ↠··· ↠C1) s.t. [X,Y] = 0 Commn = (X,Y) ∈End0(Cn ↠··· ↠C1)× × End0(Cn ↠··· ↠C1) s.t. [X,Y] = 0 for any d ≥0 and n ∈{1,2,3,4}. Before we prove Proposition 5.13, let us present its main consequence. Theo- rem 5.4 and Proposition 5.11 tell us that Quotd and Stackn are equidimensional of di- mensions rd −1 + δ0 d and −1, respectively, so Proposition 5.13 yields dimQuotd,...,d+n ≤2rd + rn −2 + δ0 d for all n ≤4. This leads to the following improvement of Corollary 5.8. Corollary 5.14. — (a) For any sequence of defects d1,...,dk ≥0 and any set partition λ whose parts all have multiplicity ≤4, we have (5.10) dimZdef d1,...,dk λ ≤const + r(cﬁrst 2 + clast 2 ) + k HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES and the stack and the stack Stackn = Commn/B Stackn = Commn/B where B ⊂GLn denotes the Borel subgroup of lower triangular matrices. The variety Commn has been studied in the literature, and we refer the reader to [3] for a survey, where the author proves that the dimension and number of irreducible components of Commn grow wildly as n increases. However, we are interested only in the particular cases n ∈{1,2,3,4}, when this variety is behaved rather mildly. Proposition 5.11. — When n ≤3, the variety Commn is irreducible of dimension n(n + 1) Proposition 5.11. — When n ≤3, the variety Commn is irreducible of dimension n(n + 1) 2 −1 = dimB −1 When n = 4, it has the dimension above, but consists of two irreducible components. For example, we have Comm3 = Spec C[x21,x31,x32,y21,y31,y32]/(x32y21 −y32x21) the intuition being that {xij,yij}3≥i>j≥1 are the entries of the matrices X and Y, and the expression x32y21 −y32x21 is the only non-zero entry of the commutator [X,Y]. Similarly, Comm4 is a subvariety of afﬁne space with 12 coordinates {xij,yij}4≥i>j≥1, cut out by 3 equations. Its irreducible components are given by (1) the subvariety Z1 ⊂Comm4 cut out by x32 = y32 = 0. Note that two of the three sub-diagonal coefﬁcients of [X,Y] vanish identically on Z1, thus dimZ1 = 10 −1 = 9 (2) the subvariety Z2 ⊂Comm4 deﬁned as the closure of the locus (x32,y32) ̸= (0,0), for which it is not hard to observe that (2) the subvariety Z2 ⊂Comm4 deﬁned as the closure of the locus (x32,y32) ̸= (0,0), for which it is not hard to observe that dimZ2 = 12 −3 = 9 dimZ2 = 12 −3 = 9 It is straightforward to prove that Comm4 is generically reduced. Its singular locus is precisely the intersection Z1 ∩Z2, which has codimension 1 in Comm4. 5.12. Consider the natural map (notation as in (5.7)) (5.8) Quotd,...,d+n ζ Quotd × Stackn (X,Y,v) →(X0,Y0,v0) × (X′,Y′) (5.8) A. NEGUT, where v0 ∈Hom(Cr,Cd) is the projection of the cyclic vector v ∈Hom(Cr,Cd+n) onto the quotient Cd+n ↠Cd (it is easy to see that projection preserves cyclicity). (5.9) (5.10) and equality holds if and only if d1 = ··· = dk = 0. and equality holds if and only if d1 = ··· = dk = 0. and equality holds if and only if d1 = ··· = dk = 0. (b) With notation as in (a), the number of irreducible components of Zdef 0,. λ (b) With notation as in (a), the number of irreducible c (b) With notation as in (a), the number of irreducible components of Z , , λ of top dimension is 2#, where # is the number of parts of λ of multiplicity 4. is 2#, where # is the number of parts of λ of multiplicity 4. Proof. — Statement (a) is an immediate consequence of (5.5) and (5.9). For state- ment (b), we invoke Lemma 5.7 to reduce the problem to the fact that Quot0,...,n has 1 irreducible component of top dimension if n ∈{1,2,3}, and 2 such components if n = 4. To see this, recall that Quot0,...,n is an open subset of a rank rn afﬁne bundle over Stackn. As shown in Proposition 5.11, the latter stack is irreducible if n ≤3, while if n = 4 its two irreducible components were described in items (1) and (2) of the previous subsec- tion. The open condition that determines Quot0,...,n in the said afﬁne bundle is given by requiring that v of (5.6) be cyclic, and it is easy to see that this open condition does not “miss” any of the two components. □ Proof of Proposition 5.13. — Consider the locally closed subset 1) Sn,d,r,i = dim ζ −1 (X0,Y0,v0) × (X′,Y′) = i ⊂Quotd × Stackn (5.11) To prove the proposition, it sufﬁces to show that (5.12) codim Sn,d,r,i ≥i −r(d + n) (5.12) codim Sn,d,r,i ≥i −r(d + n) (5.15) Therefore, condition (5.12) is trivial unless Therefore, condition (5.12) is trivial unless (5.16) n > r 2) rn extra coordinates that upgrade v0 ∈Hom(Cr,Cd) to v ∈Hom(Cr,Cd+n) The data in (1)–(2) above provide 2dn + rn afﬁne coordinates, and the quotient by the unipotent part of the group P (i.e. the block strictly lower triangular matrices, in the notation of (5.7)) subtracts dn dimensions, on account of the P-action being free. After accounting for all the generators and relations above, we see that (5.15) dim ζ −1(point (5.13)) = n(r + d) −# of independent equations in (5.14) (5.15) dim ζ −1(point (5.13)) = n(r + d) −# of independent equations in (5.14) (5.16) (5.16) n > r Given that we ﬁxed X0,Y0,X′,Y′, one may think of (5.14) as a linear map Given that we ﬁxed X0,Y0,X′,Y′, one may think of (5.14) as a linear map Given that we ﬁxed X0,Y0,X′,Y′, one may think of (5.14) as a linear map (5.17) Hom(Cd,Cn) × Hom(Cd,Cn) μ−→Hom(Cd,Cn) μ(x,y) = (x · Y0 −Y′ · x) −(y · X0 −X′ · y) (5.17) HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES for all n,d,r,i. The ﬁber of ζ over a point for all n,d,r,i. The ﬁber of ζ over a point (5.13) (X0,Y0,v0) × (X′,Y′) ∈Quotd × Stackn (5.13) (X0,Y0,v0) × (X′,Y′) ∈Quotd × Stackn consists of a choice of (5.13) consists of a choice of consists of a choice of (1) n × d blocks x,y as in the bottom left corner of (5.7), which satisfy (1) n × d blocks x,y as in the bottom left corner of (5.7), which satisfy (5.14) x · Y0 + X′ · y = y · X0 + Y′ · x (5.14) x · Y0 + X′ · y = y · X0 + Y′ · x and (5.17) The number of independent equations in (5.14) is equal to dim Im μ. Claim 5.15. — With the notation as above, we have Claim 5.15. — With the notation as above, we have (5.18) dimIm μ ≥n · dim(Im X0 + Im Y0) dimIm μ ≥n · dim(Im X0 + Im Y0) (5.18) (5.18) Proof. — Let us consider the ﬂag of quotients Cn πn↠Cn−1 πn−1 ↠··· π1↠C1 which is preserved by the nilpotent maps X′ and Y′, and consider the commutative diagram Hom(Cd,Ci) × Hom(Cd,Ci) μi Hom(Cd,Ci) Hom(Cd,Ci−1) × Hom(Cd,Ci−1) μi−1 Hom(Cd,Ci−1) Hom(Cd,Ci) × Hom(Cd,Ci) μi Hom(Cd,Ci) A. NEGUT, Above, the horizontal arrows are deﬁned by the same formula as (5.17), and the vertical arrows are the operators of composition with πi. Let K = Ker πi ∼= C. Since all the maps in the commutative diagram are linear, it sufﬁces to show that Above, the horizontal arrows are deﬁned by the same formula as (5.17), and the vertical arrows are the operators of composition with πi. Let K = Ker πi ∼= C. Since all the maps in the commutative diagram are linear, it sufﬁces to show that (5.19) dim Im μi ∩Hom(Cd,K) ≥dim(Im X0 + Im Y0) (5.19) Because X′\|Ci and Y′\|Ci annihilate K, then for any Because X′\|Ci and Y′\|Ci annihilate K, then for any we have μi(x,y) = xY0 −yX0. Then we ﬁx a basis of Im X0 +Im Y0 consisting of linearly independent vectors in Im X0 ∪Im Y0. We may deﬁne x,y ∈Hom(Cd,K) ∼= (Cd)∨to take any values on these basis vectors, which precisely implies (5.19). □ we have μi(x,y) = xY0 −yX0. Then we ﬁx a basis of Im X0 +Im Y0 consisting of linearly independent vectors in Im X0 ∪Im Y0. We may deﬁne x,y ∈Hom(Cd,K) ∼= (Cd)∨to take any values on these basis vectors, which precisely implies (5.19). □ The essential feature which allowed the proof of Claim 5.15 to work was that X′ and Y′ are strictly lower triangular. Since the commuting nilpotent matrices X0 and Y0 also preserve a full ﬂag of subspaces of Cd, a completely analogous argument yields The essential feature which allowed the proof of Claim 5.15 to work was that X′ and Y′ are strictly lower triangular. Since the commuting nilpotent matrices X0 and Y0 also preserve a full ﬂag of subspaces of Cd, a completely analogous argument yields (5.20) dimIm μ ≥d · dim(Im X′ + Im Y′) (5.20) so we obtain the estimate so we obtain the estimate # of independent equations in (5.14) ≥ ≥max(n · dim(Im X0 + Im Y0),d · dim(Im X′ + Im Y′)) Claim 5.16. (5.18) — For any 4 ≥n > r, the locus of points (5.13) such that ≥max(n · dim(Im X0 + Im Y0),d · dim(Im X′ + Im Y′)) ≥max(n · dim(Im X0 + Im Y0),d · dim(Im X + Im Y )) Claim 5.16. — For any 4 ≥n > r, the locus of points (5.13) such that Claim 5.16. — For any 4 ≥n > r, the locus of points (5.13) such that max(n · dim(Im X0 + Im Y0),d · dim(Im X′ + Im Y′)) = j ∈N max(n · dim(Im X0 + Im Y0),d · dim(Im X′ + Im Y′)) = j ∈N max(n · dim(Im X0 + Im Y0),d · dim(Im X′ + Im Y′)) = j ∈N has codimension ≥d(n −r) −j in Quotd × Stackn, with the following exception. There is an irre- ducible component G of the locus above for n = 4, d = 2, r = 1, j = 4 which has codimension = 2(4−1)−5 instead of ≥2(4−1)−4. However, outside of codimension 1, points of G have the property that dimIm μ ≥5 instead of ≥4. which has codimension = 2(4−1)−5 instead of ≥2(4−1)−4. However, outside of codimension 1, points of G have the property that dimIm μ ≥5 instead of ≥4. It is straightforward to see that formula (5.15) and Claim 5.16 imply the required inequality (5.12) (even the presence of G does not violate this fact, because even though G has dimension one larger than expected, outside of codimension 1 points of G have the property that dimIm μ is one larger than expected). HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Proof of Claim 5.16. — We may assume d > 0, since the d = 0 case is trivial. Be- cause the matrices X0 and Y0 have a cyclic vector v0 ∈Hom(Cr,Cd), then Proof of Claim 5.16. — We may assume d > 0, since the d = 0 case is trivial. Be- cause the matrices X0 and Y0 have a cyclic vector v0 ∈Hom(Cr,Cd), then dim(Im X0 + Im Y0) ≥d −r Indeed, the existence of a proper subspace of Cd that contains Im X0, Im Y0 and the r columns of v0 contradicts the cyclicity of v0. The fact that n(d −r) ≥d(n −r) establishes Claim 5.16 when n ≤d. (5.18) Therefore, we are left to deal with the case Indeed, the existence of a proper subspace of Cd that contains Im X0, Im Y0 and the r columns of v0 contradicts the cyclicity of v0. The fact that n(d −r) ≥d(n −r) establishes Claim 5.16 when n ≤d. Therefore, we are left to deal with the case (5.21) n > d n > d n > d (5.21) Taken together with the inequality (5.16), as well as the fact that we only consider n ∈ {1,2,3,4}, this means that we are left with ﬁnitely many cases. Claim 5.17. — The locally closed substack Ln,λ ⊂Stackn consisting of (X′,Y′) with the property that dim(Im X′ + Im Y′) = λ has codimension Claim 5.17. — The locally closed substack Ln,λ ⊂Stackn consisting of (X′,Y′) with the property that dim(Im X′ + Im Y′) = λ has codimension codim L2,λ = 2 if λ = 0 0 if λ = 1, codim L3,λ = ⎧ ⎪⎨ ⎪⎩ 5 if λ = 0 1 if λ = 1 0 if λ = 2, codim L4,λ = ⎧ ⎪⎪⎪⎨ ⎪⎪⎪⎩ 9 if λ = 0 3 if λ = 1 1 if λ = 2 0 if λ = 3 Claim 5.18. — The locally closed subscheme Md,μ ⊂Quotd consisting of (X0,Y0,v0) such that dim(Im X0 + Im Y0) = μ is empty unless μ ∈{d −r,...,d −1}. Moreover, codim M1,μ ≥ 0 if μ = 0, codim M2,μ ≥ 3 if μ = 0 0 if μ = 1, codim M3,μ ≥ ⎧ ⎪⎨ ⎪⎩ 8 if μ = 0 2 if μ = 1 0 if μ = 2 The proof of Claims 5.17 and 5.18 are straightforward exercises, which we leave to the interested reader. For example, the μ = 0 case of Claim 5.18 holds because the subscheme {X0 = Y0 = 0} ⊂Quotd has codimension d2 −1. Indeed, this subscheme is nothing but the Grassmannian of d dimensional quotients of r dimensional space, so it has dimension d(r −d), whereas Quotd has dimension rd −1. Note that Claim 5.16 reduces to the inequality codim Ln,λ + codim Md,μ ≥d(n −r) −max(dλ,nμ) (5.22) A. NEGUT, for all 0 ≤λ < n, max(0,d −r) ≤μ < d and 1 ≤d,r < n ≤4. We leave it to the inter- ested reader to show that inequality (5.22) follows from Claims 5.17 and 5.18, with one exception: the inequality fails for n = 4, d = 2, r = 1, λ = 2, μ = 1, and in fact, in this case the difference between the LHS and RHS of (5.22) is equal to −1. (5.21) The failure of the inequality is due to the locally closed subset G ⊂Quot2 × Stack4 consisting of points (X0,Y0,v0) × (X′,Y′) such that X0,Y0 are generic, but Im X′ + Im Y′ has dimension 2. After a change of basis, points of G take the form X = ⎛ ⎜⎜⎜⎜⎜⎜⎝ 0 0 s 0 0 0 0 0 x1 0 0 0 x2 0 0 0 x3 x4 x5 0 ⎞ ⎟⎟⎟⎟⎟⎟⎠ Y = ⎛ ⎜⎜⎜⎜⎜⎜⎝ 0 0 t 0 0 0 0 0 y1 0 0 0 y2 0 0 0 y3 y4 y5 0 ⎞ ⎟⎟⎟⎟⎟⎟⎠ X = ⎛ ⎜⎜⎜⎜⎜⎜⎝ 0 0 s 0 0 0 0 0 x1 0 0 0 x2 0 0 0 x3 x4 x5 0 ⎞ ⎟⎟⎟⎟⎟⎟⎠ Y = ⎛ ⎜⎜⎜⎜⎜⎜⎝ 0 0 t 0 0 0 0 0 y1 0 0 0 y2 0 0 0 y3 y4 y5 0 ⎞ ⎟⎟⎟⎟⎟⎟⎠ with (s,t) ̸= (0,0) and y1x2 = x1y2, y1x4 + y2x5 = x1y4 + x2y5. It is easy to show that G is irreducible (compare it with Z1 of Section 5.10). Moreover, G contains the point x1 = ··· = x5 = y1 = ··· = y5 = 1, for which it is elementary to show that dimIm μ ≥5. The fact that the latter inequality holds on G outside of codimension 1 is a consequence of lower semicontinuity of rank. This establishes the ﬁnal sentence of Claim 5.16. □ 5.19. Since the locally closed subsets Zdef d1,...,dk λ stratify the scheme Zλ, Corol- lary 5.14 implies that 5.19. Since the locally closed subsets Zdef d1,...,dk λ stratify the scheme Zλ, Corol- lary 5.14 implies that (5.23) dimZλ = const + r(cﬁrst 2 + clast 2 ) + k HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Proposition 5.21. — Z split (x,x) has dimension Proposition 5.21. — Z split (x,x) has dimension Proposition 5.21. — Z split (x,x) has dimension const + r(cﬁrst 2 + clast 2 ) −1 and a single irreducible component of top dimension. Proof. — Consider the stratiﬁcation into locally closed subsets Proof. — Consider the stratiﬁcation into locally closed subsets Z split (x,x) = ∞ d=0 Z split,def d (x,x) Z split (x,x) = ∞ d=0 Z split,def d (x,x) where the d-th stratum parameterizes those ﬂags as in (5.24) where F2 has defect d at x. It sufﬁces to show that where the d-th stratum parameterizes those ﬂags as in (5.24) where F2 has defect d at x. It sufﬁces to show that (5.25) dimZ split,def d (x,x) ≤const + r(cﬁrst 2 + clast 2 ) −1 (5.25) with equality if and only if d = 0. The splitting condition in (5.24) may be realized as follows. Consider the codimension 2 substack with equality if and only if d = 0. The splitting condition in (5.24) may be realized as follows. Consider the codimension 2 substack (5.26) Stack′ 2 = X′ = Y′ = 0 ⊂Stack2 (5.26) and the following ﬁber square where the map on the right is (5.8): Quot′ d,d+1,d+2 ζ ′ Quotd,d+1,d+2 ζ Quotd × Stack′ 2 Quotd × Stack2 It is easy to see that It is easy to see that Z split,def d (x,x) is to Quot′ d,d+1,d+2 as Zdef d (x,x) is to Quotd,d+1,d+2 (as per Lemma 5.7). With this in mind, (5.25) is a consequence of the following analogue of Proposition 5.13: (as per Lemma 5.7). With this in mind, (5.25) is a consequence of the following analogue of Proposition 5.13: dimQuot′ d,d+1,d+2 ≤dimQuotd + dimStack′ 2 + r(d + 2) = dimQuotd + dimStack2 −2 + r(d + 2) To prove the inequality above, one follows the proof of Proposition 5.13 closely, but with X′ = Y′ = 0 throughout. In particular, (5.16) and (5.21) reduce the problem to the case r = d = 1, in which case Claim 5.16 is trivial. As for the statement of a single irreducible component of top dimension, this is because one only has equality in (5.25) for d = 0, and one argues as in the proof of Corollary 5.14. (5.23) whenever \|λ\| ≤4. Keeping in mind Deﬁnition 2.25, we conclude the following. whenever \|λ\| ≤4. Keeping in mind Deﬁnition 2.25, we conclude the following. Corollary 5.20. — If \|λ\| ≤4, then Zλ has expected dimension. The number of its irreducible components of top dimension is 1 if λ ̸= (x,x,x,x) 2 if λ = (x,x,x,x) Let us now consider Z(x,y), which parameterizes ﬂags (F0 ⊂x F1 ⊂y F2). It was shown in [28, Claim 4.22] that the singular locus of Z(x,y) is given by (5.24) Z split (x,x) = (F0 ⊂x F1 ⊂x F2) such that F2/F0 is split (5.24) where split means that the length 2 sheaf F2/F0 supported at x is a direct sum of length 1 skyscraper sheaves Cx. We will now estimate the dimension of (5.24). where split means that the length 2 sheaf F2/F0 supported at x is a direct sum of length 1 skyscraper sheaves Cx. We will now estimate the dimension of (5.24). HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES □ To prove the inequality above, one follows the proof of Proposition 5.13 closely, but with X′ = Y′ = 0 throughout. In particular, (5.16) and (5.21) reduce the problem to the case r = d = 1, in which case Claim 5.16 is trivial. As for the statement of a single irreducible component of top dimension, this is because one only has equality in (5.25) for d = 0, and one argues as in the proof of Corollary 5.14. □ □ A. NEGUT, 6. Geometric properties 6.1. We will now use the dimension estimates from the previous Section to obtain various geometric properties of the varieties Zλ, which we have invoked in Section 2. Proposition 6.2. — If Zλ is Cohen-Macaulay of expected dimension (see Deﬁnition 2.25), then the map pi S : Zλ →S which remembers the i-th support point is ﬂat ∀i. Proposition 6.2. — If Zλ is Cohen-Macaulay of expected dimension (see Deﬁnition 2.25), then the map pi S : Zλ →S which remembers the i-th support point is ﬂat ∀i. Proof. — Since S is smooth and Zλ is Cohen-Macaulay, the Miracle Flatness The- orem asserts that all we need to show is that the ﬁbers of the morphism pi S : Zλ →S all have dimension equal to dimZλ −2. By upper semicontinuity, it sufﬁces to show that all the ﬁbers have dimension ≤dimZλ −2. This is proved in the same way as Corollary 5.8, since ﬁxing one of the support points of the ﬂag merely replaces the number k in the right-hand side of (5.5) by k −2. □ Proof of Proposition 2.26. — See [27, Proposition 2.10] for smoothness, and Corol- lary 5.20 for the dimension and irreducibility statements. □ Proof of Proposition 2.27. — See [28, Proposition 4.21] for smoothness, and Corol- lary 5.20 for the dimension and irreducibility statements. Let us now show that the ﬁber square (2.37) is derived with excess. By Proposition 2.19, we may subdivide this ﬁber square into 6.1) Z(x,x) π+ ι′ PZ(x)(∗(V1)) ρ′ Z(x) Z(x) ι PM×S(V1) ρ M × S (6.1) where : Z(x) →Z(x) × S is the graph of the map pS : Z(x) →S. The rightmost ﬁber square in (6.1) is already derived, because the maps ρ and ρ′ are smooth. However, the leftmost ﬁber square in (6.1) is not derived, because where : Z(x) →Z(x) × S is the graph of the map pS : Z(x) →S. The rightmost ﬁber square in (6.1) is already derived, because the maps ρ and ρ′ are smooth. However, the leftmost ﬁber square in (6.1) is not derived, because (6.2) codim ι −codim ι′ = dimZ(x,x) −dimZ(x) −dimZ(x) + dim(M × S) = 1 (6.2) (this follows from the dimension statements of Propositions 2.26 and 2.27, which have already been proved). 6. Geometric properties Recall from Proposition 2.19 that ι is cut out by the regular section ∗(W ) ∗(V ) O(1) (this follows from the dimension statements of Propositions 2.26 and 2.27, which have already been proved). Recall from Proposition 2.19 that ι is cut out by the regular section ρ∗(W1) →ρ∗(V1) ↠O(1) hence is a complete intersection of codimension equal to the rank of W1. On the other hand, ι′ is cut out by the section hence is a complete intersection of codimension equal to the rank of W1. On the other hand, ι′ is cut out by the section W1\| →V1\| ↠O(1) = L1 HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES In Proposition 2.21, we showed that the section above factors through In Proposition 2.21, we showed that the section above factors through (6.3) W1\| L2 ⊗p∗ S(ωS) −→O(1) = L1 (6.3) where the left-hand side is a locally free sheaf of rank 1 less than the rank of W1. By (6.2), the section (6.3) is therefore regular, and this establishes the claim that the square (2.37) is derived with excess bundle L2 ⊗L−1 1 ⊗p∗ S(ωS). □ where the left-hand side is a locally free sheaf of rank 1 less than the rank of W1. By (6.2), the section (6.3) is therefore regular, and this establishes the claim that the square (2.37) is derived with excess bundle L2 ⊗L−1 1 ⊗p∗ S(ωS). □ Proof of Propositions 2.28, 2.29, 2.30. — See Corollary 5.20 for the dimension and irreducibility statements. The fact that the schemes in question are l.c.i. is proved by showing that the ﬁber squares in the statements of the Propositions are derived. The lat- ter statement is an immediate consequence of Proposition 2.19 together with the fact that the spaces in the four corners of each square satisfy the analogue of (2.5) (itself an imme- diate consequence of the already proved fact that the northeast, southeast and southwest corners of these squares are l.c.i. of expected dimension, while the northwest corner has expected dimension, see Corollary 5.20). □ 6.3. We will now prove that the variety Z(x,y,x) is Cohen-Macaulay. 6.3. We will now prove that the variety Z(x,y,x) is Cohen-Macaulay. The dimension of Z(x) × S is equal to const + r(c2(F0) + c2(F1)) + 3 and therefore the dimension of the projective bundle (6.4) is equal to const + r(c2(F0) + c2(F1)) + 2 + rank W1 To obtain the closed embedding ι in (6.4), we impose as many equations as the number of coordinates of the section (6.5), so the expected dimension of Z(x,y) is const + r(c2(F0) + c2(F1)) + 2 + rank W1 −rank V1 = = const + r(c2(F0) + c2(F2)) + 2 Since this is equal to the actual dimension of Z(x,y) by (5.23), we conclude that the coor- dinates of the section σ form a regular sequence Since this is equal to the actual dimension of Z(x,y) by (5.23), we conclude that the coor- dinates of the section σ form a regular sequence = regular sequence formed by the coordinates of σ (6.6) In order to study in detail, recall that the vector bundle V1 is given by (2.15): from now on we will set n = 0 in (2.15), in order to keep our formulas legible (otherwise, one would have to often tensor our formulas by line bundles coming from S, but this has no substantial effect on our argument). We have a map of locally free sheaves φ : V1 →L1 on Z(x) that is a particular case of the middle row of diagram (2.30) when i = 1. Putting all of these constructions together, we may consider the map of line bundles In order to study in detail, recall that the vector bundle V1 is given by (2.15): from now on we will set n = 0 in (2.15), in order to keep our formulas legible (otherwise, one would have to often tensor our formulas by line bundles coming from S, but this has no substantial effect on our argument). We have a map of locally free sheaves φ : V1 →L1 on Z(x) that is a particular case of the middle row of diagram (2.30) when i = 1. Putting all of these constructions together, we may consider the map of line bundles (6.7) φ ◦σ : O(−1) taut −→W1 ⊗ω−1 S →V1 ⊗ω−1 S φ⊗Id −−→L1 ⊗ω−1 S (6.7) on P. 6.3. We will now prove that the variety Z(x,y,x) is Cohen-Macaulay. Proof of Proposition 2.31. — See Corollary 5.20 for the dimension and irreducibility statements. As a warm-up to proving the fact that Z(x,y,x) is Cohen-Macaulay, let us recall the explicit realization of Z(x,y) as a local complete intersection. We do so not only because the computation will serve as valuable illustration, but we will need a speciﬁc realization of Z(x,y) as the zero locus of a regular section. Therefore, consider the smooth scheme Z(x) × S and take the map Z(x,y) π+ Z(x) × S (F0 ⊂x F1 ⊂y F2) (F0 ⊂x F1,y) (F0 ⊂x F1 ⊂y F2) A particular case of Proposition 2.19 says that this map can be realized as A particular case of Proposition 2.19 says that this map can be realized as (6.4) Z(x,y) π+ ι P := PZ(x)×S(W∨ 1 ⊗ωS) Z(x) × S (6.4) Z(x,y) π+ ι P := PZ(x)×S(W∨ 1 ⊗ωS) Z(x) × S (6.4) where the closed embedding ι is cut out by the section where the closed embedding ι is cut out by the section (6.5) σ : O(−1) →W1 ⊗ω−1 S →V1 ⊗ω−1 S where the closed embedding ι is cut out by the section (6.5) σ : O(−1) →W1 ⊗ω−1 S →V1 ⊗ω−1 S (6.5) σ : O(−1) →W1 ⊗ω−1 S →V1 ⊗ω−1 S (6.5) A. NEGUT, and O(1) denotes the tautological line bundle on the projective bundle in (6.4). Above and hereafter, we abuse notation by writing W1 for the vector bundle on Z(x) × S, as well as for its pull-back to P. The dimension of Z(x) × S is equal to and O(1) denotes the tautological line bundle on the projective bundle in (6.4). Above and hereafter, we abuse notation by writing W1 for the vector bundle on Z(x) × S, as well as for its pull-back to P. 6.3. We will now prove that the variety Z(x,y,x) is Cohen-Macaulay. Since Ker φ = V0 is locally free, the map of line bundles φ ◦σ may be thought of as one of the coordinates of the section σ in any local trivialization. Therefore, it is one of the entries of the regular sequence , so let us compute it explicitly on a closed point of the projective bundle (6.4). We may work locally, so we assume that x and y are points of A2 = Spec C[s1,s2] (alternatively, we need to take generators of the maximal ideals at the closed points x and y, but the explanation is analogous) given by coordinates (x1,x2) and (y1,y2), respectively. As we have seen in the proof of Proposition 2.21, the map taut in (6.7) is identiﬁed with (S,F2/F1) ⊗T or2(Oy,Oy) →T or0(W1,Oy) HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES where y : P →P × S is the graph of the map that remembers the point y. The Tor groups above may be computed using the following length 2 resolution of Oy: O s2−y2,y1−s1 −−−−−→O ⊕O s1−y1,s2−y2 −−−−−→O q.i.s. ∼= Oy Therefore, the map denoted taut in (6.7) takes a generator v of the one-dimensional vector space (S,F2/F1) to the section (s1 −y1)τ1 + (s2 −y2)τ2, where τ1,τ2 ∈(S,F1) have the property that τ1 = (s2 −y2)f and τ2 = (y1 −s1)f for some preimage f ∈(S,F2) of v ∈(S,F2/F1) (global sections exist due to our assumption that n = 0 in (2.15), otherwise we would need to tensor F1 and F2 by a very ample line bundle). Therefore, the entire composition (6.7) takes the generator v to the image of (s1 −y1)τ1 + (s2 −y2)τ2 in (S,F1/F0) In coordinates, if we identify (S,F2/F0) with C2, then the OS = C[s1,s2]-module struc- ture on this 2-dimensional vector space is given by matrices In coordinates, if we identify (S,F2/F0) with C2, then the OS = C[s1,s2]-module struc- ture on this 2-dimensional vector space is given by matrices (6.8) s1 = y1 0 a1 x1 s2 = y2 0 a2 x2 (6.8) and so the assignment (6.7) takes 1 0 ⇝ 0 (x1 −y1)a2 −(x2 −y2)a1 Therefore, we conclude that (6.9) (x1 −y1)a2 −(x2 −y2)a1 (6.9) is one of the elements of the regular sequence (6.6). 6.3. We will now prove that the variety Z(x,y,x) is Cohen-Macaulay. Since over local Noetherian rings, one may permute the order of elements of a regular sequence, we will assume (6.9) to be the last element of the regular sequence. is one of the elements of the regular sequence (6.6). Since over local Noetherian rings, one may permute the order of elements of a regular sequence, we will assume (6.9) to be the last element of the regular sequence. Claim 6.4. — Before imposing Equation (6.9), the other elements of the regular sequence cut out a regular local ring. In more mathematical terms, the local rings of Z(x,y) are quotients of regular local rings by the single equation (6.9). Indeed, the claim follows from the fact (proved in [28]) that the tangent spaces to Z(x,y) have expected dimension, except at a closed point such that x = y and F2/F0 is split, where the dimension of the tangent space jumps by 1. In the local coordinates (6.9), this corresponds to a1 = a2 = x1 −y1 = x2 −y2 = 0: therefore, Equation (6.9) fails to cut down the dimension of the tangent spaces on the split locus. Since Z(x) is smooth, this means that all other elements in the regular sequence do cut out regular subschemes, and regularity is broken precisely by (6.9). A. NEGUT, Armed with the discussion above, we are ready to analyze the scheme Z(x,y,x): Armed with the discussion above, we are ready to analyze the scheme Z(x,y,x): Z(x,y,x) π+ Z(x,y) (F0 ⊂x F1 ⊂y F2 ⊂x F3) (F0 ⊂x F1 ⊂y F2) (F0 ⊂x F1 ⊂y F2 ⊂x F3) A particular case of Proposition 2.19 says that this map can be realized as (6.10) Z(x,y,x) π+ ι P′ := PZ(x,y)(x∗(W∨ 2 ⊗ωS)) Z(x,y) (6.10) where x : Z(x,y) →Z(x,y) × S is the graph of the map that remembers the point x, and the closed embedding ι is cut out by the section (6.11) σ : O(−1) →x∗(W2 ⊗ω−1 S ) →x∗(V2 ⊗ω−1 S ) (6.11) (we abuse notation by identifying the vector bundles on Z(x,y) above with their pull-backs to P′) where O(1) denotes the tautological line bundle on P′. By (5.23), the dimension of Z(x,y) is equal to (we abuse notation by identifying the vector bundles on Z(x,y) above with their pull-backs to P′) where O(1) denotes the tautological line bundle on P′. (6.12) such that N has rank 2 and the closed subscheme such that N has rank 2 and the closed subscheme (6.13) φ ◦σ = 0 →P′ (6.13) is Cohen-Macaulay of codimension 1. Once we do so, the fact that Z(x,y,x) is Cohen- Macaulay of the expected dimension follows from Claim 6.5 with i = 1. is Cohen-Macaulay of codimension 1. Once we do so, the fact that Z(x,y,x) is Cohen- Macaulay of the expected dimension follows from Claim 6.5 with i = 1. To construct the map (6.12), we go back to diagram (2.30) for i = 2. Its middle row consists of locally free sheaves, so we may restrict it to x: 0 →x∗(V1) →x∗(V2) →L2 →0 (the right-most term is L2 because we assumed n = 0 in (2.15), otherwise we could have had to twist by a line bundle). The push-out of the short exact sequence above with respect to the tautological map x∗(V1) ↠L1 that we have on Z(x,y) yields (the right-most term is L2 because we assumed n = 0 in (2.15), otherwise we could have had to twist by a line bundle). The push-out of the short exact sequence above with respect to the tautological map x∗(V1) ↠L1 that we have on Z(x,y) yields 0 x∗(V1) x∗(V2) φ L2 0 0 L1 N ⊗x∗(ωS) L2 0 where N is deﬁned by the short exact sequence on the bottom row. This deﬁnes the map (6.12). We will now write out the entries of the map φ explicitly in coordinates, just as we did in the discussion immediately preceding Claim 6.4. We still work locally, so assume that x and y are points of A2 = Spec C[s1,s2] given by coordinates (x1,x2) and (y1,y2), respectively. Then we identify (S,F3/F0) with C3, and then the OS = C[s1,s2]-module structure on this 3-dimensional vector space is given by matrices (6.14) s1 = ⎛ ⎝ x1 0 0 b1 y1 0 c1 a1 x1 ⎞ ⎠ s2 = ⎛ ⎝ x2 0 0 b2 y2 0 c2 a2 x2 ⎞ ⎠ (6.14) The matrices (6.8) are precisely the bottom right 2 × 2 blocks of (6.14). Therefore, the composition φ ◦σ : O(−1) →N is given explicitly in coordinates by The matrices (6.8) are precisely the bottom right 2 × 2 blocks of (6.14). 6.3. We will now prove that the variety Z(x,y,x) is Cohen-Macaulay. By (5.23), the dimension of Z(x,y) is equal to ) Z(x,y) is equal to const + r(c2(F0) + c2(F2)) + 2 and therefore the dimension of the projective bundle (6.10) is equal to and therefore the dimension of the projective bundle (6.10) is equal to and therefore the dimension of the projective bundle (6.10) is equal to const + r(c2(F0) + c2(F2)) + 1 + rank W2 const + r(c2(F0) + c2(F2)) + 1 + rank W2 To obtain the closed embedding ι in (6.10), we impose as many equations as the number of coordinates of (6.11), so the expected dimension of Z(x,y,x) is const + r(c2(F0) + c2(F2)) + 1 + rank W2 −rank V2 = = const + r(c2(F0) + c2(F3)) + 1 const + r(c2(F0) + c2(F2)) + 1 + rank W2 −rank V2 = However, according to (5.23), the actual dimension of Z(x,y,x) is 1 bigger than the above expected dimension. Therefore, Z(x,y,x) is an almost complete intersection, i.e. cut out by one more equation than its dimension. We have the following criterion for when such schemes are Cohen-Macaulay. Claim 6.5. — Consider a Cohen-Macaulay local ring R and a collection of elements f0,...,fn ∈R such that the quotient ring R/(f0,...,fn) has codimension n in R. If R/(f0,...,fi) is Cohen-Macaulay of codimension i in R for some i ≥0, then R/(f0,...,fn) is Cohen-Macaulay. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES The claim is proved by induction and the well-known fact that if an element f in a Cohen-Macaulay local ring R has the property that dimR/(f ) = dimR −1, then f is a non-zero divisor and R/(f ) is Cohen-Macaulay. We will apply Claim 6.5 to our situation by constructing a map of locally free sheaves on P′: (6.12) φ : x∗(V2 ⊗ω−1 S ) ↠N (6.12) (6.12) Therefore, the composition φ ◦σ : O(−1) →N is given explicitly in coordinates by ⎛ ⎝ 1 0 0 ⎞ ⎠⇝ ⎛ ⎝ 0 b1(x2 −y2) −b2(x1 −y1) b1a2 −a1b2 ⎞ ⎠ A. NEGUT, Therefore, we conclude that the subscheme (6.13) is obtained by imposing the equations b1(x2 −y2) = b2(x1 −y1) and b1a2 = a1b2 on the local rings of a projective bundle over the scheme Z(x,y). As we have seen in Claim 6.4, the local rings of the scheme Z(x,y) were obtained by imposing the equation a1(x2 −y2) = a2(x1 −y1) of (6.9) in a regular local ring. Therefore, we are in a particular case of the following general situation. Claim 6.6. — Suppose R is a regular ring with given elements a1,b1,a2,b2,d1,d2, and let I = (a1d2 −a2d1,b1d2 −b2d1,a1b2 −b1a2). Then R/I is Cohen-Macaulay of codimension 2 if (1) a1b2 −b1a2 is not a zero-divisor in R/(d1,d2) (2) R/(a1d2 −a2d1,b1d2 −b2d1) is codimension 2 in R Using Claim 6.4, let us show that the hypotheses of Claim 6.6 hold with a1,b1,a2,b2 as in (6.14) and d1 = x1 −y1, d2 = x2 −y2. For item (1), we observe that imposing d1 = d2 = 0 has the effect of setting the support points x,y equal to each other, which cuts out the smooth subscheme Z(x,x) →Z(x,y). Since P′\|Z(x,x) is smooth, it does not have any non-trivial zero-divisors, so must show that (6.15) a1b2 −b1a2 ̸= 0 (6.15) a1b2 −b1a2 ̸= 0 (6.15) in the local rings of P′\|Z(x,x). If (6.15) failed to hold, then in the local rings of P′\|Z(x,x). If (6.15) failed to hold, then in the local rings of P′\|Z(x,x). If (6.15) failed to hold, then # equations cutting out ι\|Z(x,x) ≤# equations cutting out ι −1 and so and so and so dimZ(x,x,x) = dimP′\|Z(x,x) −# equations cutting out ι\|Z(x,x) ≥ ≥dimP′ −# equations cutting out ι = dimZ(x,y,x) which would contradict (5.23). As for item (2), we must show that b1d2 −b2d1 is not a zero-divisor in the local rings of P′. This is the case because b1,b2 are linear coordinates in a projective bundle over Z(x,y), so the only way b1d2 −b2d1 could be a zero divisor is if d1 and d2 were zero-divisors. If this were the case, then Z(x,x) would have the same dimension as Z(x,y), which would contradict (5.23). (6.12) Proof of Claim 6.6. — By the Auslander-Buchsbaum formula, it is enough to show that the ring R/I has projective dimension 2 as an R-module. In fact, we claim that a HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES projective resolution is given by projective resolution is given by 0 −→R2 ⎛ ⎜⎜⎜⎝ b1 b2 −a1 −a2 d1 d2 ⎞ ⎟⎟⎟⎠ −−−−−−−−−→R3 d1a2 −a1d2 d1b2 −b1d2 a1b2 −a2b1 −−−−−−−−−−−−−−−−−−−−−−−−−−−−→ R −→R/I −→0 R −→R/I −→0 Exactness at R and R/I is obvious. As for exactness at R2, if the ﬁrst map failed to be injective then a1b2−a2b1, d1b2−b1d2 and d1a2−a1d2 would all be zero-divisors (hence zero in the regular ring R) which is not allowed by property (1). It remains to prove exactness at R3. Assume that we have (x,y,z) ∈R3 such that (d1a2 −a1d2)x + (d1b2 −b1d2)y + (a1b2 −a2b1)z = 0 Clearly, (a1b2 −a2b1)z ∈(d1,d2), so property (1) implies z = d1m+d2n. Therefore, we may rewrite the relation above as Clearly, (a1b2 −a2b1)z ∈(d1,d2), so property (1) implies z = d1m+d2n. Therefore, we may rewrite the relation above as (d1a2 −a1d2)(b1m + b2n + x′) + (d1b2 −b1d2)(−a1m −a2n + y′)+ + (a1b2 −a2b1)(d1m + d2n) = 0 (d1a2 −a1d2)(b1m + b2n + x′) + (d1b2 −b1d2)(−a1m −a2n + y′)+ + (a1b2 −a2b1)(d1m + d2n) = 0 where x′ = x −b1m −b2n and y′ = y + a1m + a2n. The relation above reduces to (d1a2 −a1d2)x′ + (d1b2 −b1d2)y′ = 0 + (a1b2 −a2b1)(d1m + d2n) = 0 where x′ = x −b1m −b2n and y′ = y + a1m + a2n. The relation above reduces to (d1a2 −a1d2)x′ + (d1b2 −b1d2)y′ = 0 = x −b1m −b2n and y′ = y + a1m + a2n. The relation above reduces to (d1a2 −a1d2)x′ + (d1b2 −b1d2)y′ = 0 (d1a2 −a1d2)x′ + (d1b2 −b1d2)y′ = 0 and then item (2) implies that we have x′ = (d1b2 −b1d2)u and y′ = −(d1a2 −a1d2)u for some u. Therefore, we have and then item (2) implies that we have x′ = (d1b2 −b1d2)u and y′ = −(d1a2 −a1d2)u for some u. (6.16) which form a stratiﬁcation of the variety Z(x,x,y,x). To show normality of this variety, one may ignore all strata of codimension 2 and higher. which form a stratiﬁcation of the variety Z(x,x,y,x). To show normality of this variety, one may ignore all strata of codimension 2 and higher. (1) when x = y and F4 has non-zero defect > 0 at x, Corollary 5.14 shows that the corresponding locally closed subset of (6.16) has codimension ≥2, hence can be ignored; (1) when x = y and F4 has non-zero defect > 0 at x, Corollary 5.14 shows that the corresponding locally closed subset of (6.16) has codimension ≥2, hence can be ignored; g ; (2) when x = y and F4 is locally free near x, the scheme Z(x,x,y,x) is locally isomor- phic to M × , where (2) when x = y and F4 is locally free near x, the scheme Z(x,x,y,x) is locally isomor- phic to M × , where (2) when x = y and F4 is locally free near x, the scheme Z(x,x,y,x) is locally isomor- phic to M × , where = (O⊕r ↠Q4 ↠Q3 ↠Q2 ↠Q1), supp Q4 = {x,x,y,x}, supp Q3 = {x,y,x},supp Q2 = {y,x},supp Q1 = {x} Compare with the scheme Quot0,1,2,3,4 of Deﬁnition 5.6. We must prove that is normal. Since we may work locally, we assume that the base surface is S = A2, and we will normalize x = (0,0) and y = (a,b). By analogy with Section 5.9, the scheme parameterizes triples (X,Y,v) where X = ⎛ ⎜⎜⎝ 0 0 0 0 x21 a 0 0 x31 x32 0 0 x41 x42 x43 0 ⎞ ⎟⎟⎠, Y = ⎛ ⎜⎜⎝ 0 0 0 0 y21 b 0 0 y31 y32 0 0 y41 y42 y43 0 ⎞ ⎟⎟⎠ X = ⎛ ⎜⎜⎝ 0 0 0 0 x21 a 0 0 x31 x32 0 0 x41 x42 x43 0 ⎞ ⎟⎟⎠, Y = ⎛ ⎜⎜⎝ 0 0 0 0 y21 b 0 0 y31 y32 0 0 y41 y42 y43 0 ⎞ ⎟⎟⎠ such that [X,Y] = 0, and v ∈Hom(Cr,C4) is cyclic for X,Y. Therefore, is an open subset of an afﬁne bundle over the afﬁne variety such that [X,Y] = 0, and v ∈Hom(Cr,C4) is cyclic for X,Y. (6.12) Therefore, we have ⎛ ⎝ x y z ⎞ ⎠= ⎛ ⎝ x′ + b1m + b2n y′ −a1m −a2n d1m + d2n ⎞ ⎠= ⎛ ⎝ b1(m −d2u) + b2(n + d1u) −a1(m −d2u) −a2(n + d1u) d1(m −d2u) + d2(n + d1u) ⎞ ⎠ □ which shows that the triple (x,y,z) came from the image of the 3 × 2 matrix Proof of Proposition 2.32 and 2.33. — See the Proof of Propositions 2.28, 2.29, 2.30. □ Proof of Proposition 2.32 and 2.33. — See the Proof of Propositions 2.28, 2.29, 2.30. □ Proof of Proposition 2.32 and 2.33. — See the Proof of Propositions 2.28, 2.29, 2.30. □ 6.7. We will now prove the normality of some of the schemes Zλ for \|λ\| ≤4. 6.7. We will now prove the normality of some of the schemes Zλ for \|λ\| ≤4. 6.7. We will now prove the normality of some of the schemes Zλ for \|λ\| ≤4. Proof of Proposition 2.34. — Since all the schemes in question are Cohen-Macaulay, it sufﬁces to show that they are singular in codimension ≥2. Let us note that the required statement for λ = (x,y) follows from Proposition 5.21, which shows that the singular locus Z split (x,x) ⊂Z(x,y) A. NEGUT, has codimension 3. As for the other λ’s in (2.39), we will only prove the case λ = (x,x,y,x), as the analysis in the other cases is analogous and no more difﬁcult. Below, we will list certain locally closed subsets of (6.16) Z(x,x,y,x) = (F0 ⊂x F1 ⊂x F2 ⊂y F3 ⊂x F4) (6.16) (6.16) Therefore, is an open subset of an afﬁne bundle over the afﬁne variety = (X,Y) as above, such that [X,Y] = 0 = (X,Y) as above, such that [X,Y] = 0 It sufﬁces to show that is normal. As an afﬁne variety, it is cut out by ay21 −bx21 = 0 ay32 −bx32 = 0 x32y21 −y32x21 = 0 x43y32 −y43x32 + bx42 −ay42 = 0 x42y21 + x43y31 −y42x21 −y43x31 = 0 ay21 −bx21 = 0 ay32 −bx32 = 0 x32y21 −y32x21 = 0 x43y32 −y43x32 + bx42 −ay42 = 0 x42y21 + x43y31 −y42x21 −y43x31 = 0 ay21 −bx21 = 0 ay32 −bx32 = 0 x32y21 −y32x21 = 0 HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES in the 14-dimensional space of entries a,b,xij,yij. Using Macaulay2, one can check that is a Cohen-Macaulay irreducible afﬁne variety of dimension 10. The tangent space to at a given closed point (X,Y) is the kernel of the map μ : (∂X,∂Y) →[X,∂Y] + [∂X,Y] μ : (∂X,∂Y) →[X,∂Y] + [∂X,Y] where (∂X,∂Y) runs over the 14-dimensional afﬁne space of pairs of matrices with the same pattern of zeroes as (X,Y). Thus, is smooth at a point (X,Y) iff μ has 10-dimensional kernel, so let us see when this happens. where (∂X,∂Y) runs over the 14-dimensional afﬁne space of pairs of matrices with the same pattern of zeroes as (X,Y). Thus, is smooth at a point (X,Y) iff μ has 10-dimensional kernel, so let us see when this happens. Assume ﬁrst that (a,b) ̸= (0,0), and by taking appropriate linear com- binations of X,Y we may assume (a,b) = (1,0). Then one can successively solve the equation Ker μ = 0 for the entries of the matrices (∂X,∂Y) one by one, concluding that the kernel of μ fails to be 10-dimensional when x43 = y21 = y31 = y32 = y42 = y42 = x31 −x32x21 = 0. The dimension of this locus is 5, and after we add 2 dimensions to reverse the choice (a,b) = (1,0), we conclude that is smooth in codimension 3 on the locus (a,b) ̸= (0,0). ̸ Now assume (a,b) = (0,0). The corresponding subvariety is precisely Comm4 ⊂, and we have seen at the end of Section 5.10 that it is 9- dimensional and has two irreducible components Z1 and Z2. (6.16) Thus, it sufﬁces to take a generic point (X,Y) in each of these components, and show that the map μ has 10-dimensional kernel at the chosen point. It is straightforward to show that the following choices will do: X = ⎛ ⎜⎜⎝ 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 ⎞ ⎟⎟⎠, Y = ⎛ ⎜⎜⎝ 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 ⎞ ⎟⎟⎠ and X = ⎛ ⎜⎜⎝ 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 ⎞ ⎟⎟⎠, Y = ⎛ ⎜⎜⎝ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ⎞ ⎟⎟⎠ and (3) when x ̸= y and F4 has non-zero defect at both x and y, Corollary 5.14 shows that the corresponding locally closed subset has codimension ≥2, hence can be ignored; (3) when x ̸= y and F4 has non-zero defect at both x and y, Corollary 5.14 shows that the corresponding locally closed subset has codimension ≥2, hence can be ignored; (4) when x ̸= y and F4 is locally free near x, the scheme Zλ is locally isomorphic to Z1 × Quot0,1,2,3, and thus normal (indeed, Quot0,1,2,3 is open in an afﬁne bundle over Stack3, and the latter is easily seen to be normal); (5) when x ̸= y and F4 has non-zero defect at x but is locally free near y, the scheme Zλ is locally isomorphic to Z(x,x,x) × Quot1, so it sufﬁces to show that Z(x,x,x) is (5) when x ̸= y and F4 has non-zero defect at x but is locally free near y, the scheme Zλ is locally isomorphic to Z(x,x,x) × Quot1, so it sufﬁces to show that Z(x,x,x) is A. NEGUT, normal near any point with defect ≥1. To this end, consider the stratiﬁcation ∞ normal near any point with defect ≥1. To this end, consider the stratiﬁcation ∞ Z(x,x,x) = ∞ d=0 Zdef d (x,x,x) in terms of the defect of F4 at x. As shown in Corollary 5.14, the open sub- set Zdef 0 (x,x,x) is the only top-dimensional stratum; as it is locally isomorphic to M × Quot0,1,2,3, it is normal. (6.16) Tracing through the proof of Proposition 5.13 shows that other strata can have codimension 1 only if r = 1 and d = 1, in which case the moduli space of stable sheaves may be replaced with the Hilbert scheme of points on S. Therefore, it sufﬁces to show that the scheme ′ pa- rameterizing ﬂags of ideals (I0 ⊂x I1 ⊂x I2 ⊂x I3) is normal near any ideal I3 of defect precisely 1 at x. Since the problem is local, we may assume S = A2 and x = (0,0), in which case ′ may be described by analogy with Section 5.9 as the space of triples (X,Y,v) such that X = ⎛ ⎜⎜⎝ a 0 0 0 x21 0 0 0 x31 x32 0 0 x41 x42 x43 0 ⎞ ⎟⎟⎠, Y = ⎛ ⎜⎜⎝ b 0 0 0 y21 0 0 0 y31 y32 0 0 y41 y42 y43 0 ⎞ ⎟⎟⎠ such that [X,Y] = 0, and v ∈C4 is cyclic for X,Y. As before, one may ex- plicitly write down the quadratic equations among the {xij,yij,a,b}4≥i>j≥1 and conclude that ′ is normal. The method of proof is analogous to that in item (2) above, so we leave it as an analogous exercise to the interested reader. □ such that [X,Y] = 0, and v ∈C4 is cyclic for X,Y. As before, one may ex- plicitly write down the quadratic equations among the {xij,yij,a,b}4≥i>j≥1 and conclude that ′ is normal. The method of proof is analogous to that in item (2) above, so we leave it as an analogous exercise to the interested reader. □ 6.8. We will now study the schemes Y, Y−, Y+, Y−+ of Section 2.35. 6.8. We will now study the schemes Y, Y−, Y+, Y−+ of Section 2.35. Proposition 6.9. — (a) The schemes Y, Y−, Y+, Y−+ have expected dimension, i.e. the dimension of the respective spaces on the bottom of (2.46) or (2.47). (b) The schemes Y, Y−, Y+, Y−+ have 1, 1, 1, 2 irreducible components of expected dimen- sion, respectively. Proposition 6.9. — (a) The schemes Y, Y−, Y+, Y−+ have expected dimension, i.e. the dimension of the respective spaces on the bottom of (2.46) or (2.47). f p p f ( ) ( ) (b) The schemes Y, Y−, Y+, Y−+ have 1, 1, 1, 2 irreducible components of expected dimen- sion, respectively. (6.16) However, over any point of V1 we have F3/F1 split, and so there exists a whole P1 in Y−+ above points of V1. As dimV1 = dimZ(x,x,x,x) = dimY−+ −1, this contributes an irreducible component of top dimension to Y−+. □ where V1 and V2 lie above the irreducible components Z1 and Z2 of Stack4 (see items (1)–(2) in Section 5.10). The map π ↑has inverse image a single point over the generic point of V2. However, over any point of V1 we have F3/F1 split, and so there exists a whole P1 in Y−+ above points of V1. As dimV1 = dimZ(x,x,x,x) = dimY−+ −1, this contributes an irreducible component of top dimension to Y−+. □ where V1 and V2 lie above the irreducible components Z1 and Z2 of Stack4 (see items (1)–(2) in Section 5.10). The map π ↑has inverse image a single point over the generic point of V2. However, over any point of V1 we have F3/F1 split, and so there exists a whole P1 in Y−+ above points of V1. As dimV1 = dimZ(x,x,x,x) = dimY−+ −1, this contributes an irreducible component of top dimension to Y−+. □ 6.10. In the next subsection, we will prove that the scheme Y is smooth. To do so, we will explicitly describe the tangent space to a closed point (2.40) of Y and compute its dimension. Let us recall that the tangent space to the moduli space M at a point F ∈Coh(S) is given by 6.10. In the next subsection, we will prove that the scheme Y is smooth. To do so, we will explicitly describe the tangent space to a closed point (2.40) of Y and compute its dimension. Let us recall that the tangent space to the moduli space M at a point F ∈Coh(S) is given by (6.17) TanFM = Ext1(F,F) TanFM = Ext1(F,F) (6.16) p p ( ) ( ) (b) The schemes Y, Y−, Y+, Y−+ have 1, 1, 1, 2 irreducible components of expected dimen- sion, respectively. Recall that when we say that Y (or any of the other 3 schemes) has # irreducible components of expected dimension, what we actually mean is that it has # such irre- ducible components over each connected component of the moduli space M. Proof. — The map Y π↑ −→Z(x,y) is surjective. Over a closed point (F0 ⊂x F1 ⊂y F2) ∈Z(x,y), the ﬁber of this map is either a single point, or a copy of P1. The latter happens if and only if x = y and F2/F0 is split, so we conclude that the only points where the ﬁbers jump are those of Z split (x,x). Since the locus Z split (x,x) has codimension 3 in Z(x,y) (see Proposition 5.21), and the dimensions of the ﬁbers above such points are all 1, this implies HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES that Y will have the same dimension as Z2 (in Section 6.11, we will show that Y is actually smooth, so the map π ↑can be thought of as the blow-up of the singular locus of Z(x,y)). The fact that Y has a single irreducible component of top dimension is clear, since Z(x,y) is irreducible. The proof for the schemes Y−, Y+, Y−+ is analogous. For example, the ﬁbers of π ↑: Y−+ −→Z(x,x,y,x) consist of a single point or a copy of P1, with the latter situation only happening over closed points consist of a single point or a copy of P1, with the latter situation only happening over closed points (F0 ⊂x F1 ⊂x F2 ⊂y F3 ⊂x F4) ∈Z(x,x,y,x) s.t. x = y and F3/F1 is split Since the locus of such points is contained in Z(x,x,x,x), which has dimension 1 less than Z(x,x,y,x) by Propositions 2.32 and 2.33, the dimension of Y−+ is the same as that of Z(x,x,y,x). As for the irreducible components of top dimension in Y−+, we note that one of them is the closure of the locus x ̸= y. But recall that Z(x,x,x,x) = V1 ∪V2 where V1 and V2 lie above the irreducible components Z1 and Z2 of Stack4 (see items (1)–(2) in Section 5.10). The map π ↑has inverse image a single point over the generic point of V2. (6.19) (6.19) Since stable sheaves are simple (see [18]), we have Since stable sheaves are simple (see [18]), we have Since stable sheaves are simple (see [18]), we have dimHom(F,F) = 1 because Hom(F,F) ∼= C (6.20) dimExt2(F,F) = ε because Ext2(F,F) ∼= Hom(F,F ⊗ωS)∨ (6.21) dimHom(F,F) = 1 because Hom(F,F) ∼= C (6.20) dimExt2(F,F) = ε because Ext2(F,F) ∼= Hom(F,F ⊗ωS)∨ (6.21) where the latter isomorphism is Serre duality, and the number ε is 1 or 0 depending on which situation of Assumption S we are in (ε = 1 for ωS ∼= OS and ε = 0 for c1(ωS) · H < 0). Therefore, we conclude that (6.22) dimExt1(F,F) = 1 + ε + γ + 2rc2 (6.17) Indeed, the functor-of-points description (2.12) implies that a tangent vector at F ∈M is a coherent sheaf on S × Spec C[ν]/(ν2) which is ﬂat over the second factor and restricts to F when one sets ν = 0. In other words, a tangent vector is a coherent sheaf G on S with an morphism ν : G →G that squares to 0, such that Indeed, the functor-of-points description (2.12) implies that a tangent vector at F ∈M is a coherent sheaf on S × Spec C[ν]/(ν2) which is ﬂat over the second factor and restricts to F when one sets ν = 0. In other words, a tangent vector is a coherent sheaf G on S with an morphism ν : G →G that squares to 0, such that G/Im ν ∼= F The ﬂatness condition on G implies that TorC[ν]/(ν2) 1 (C[ν]/(ν),G) = 0, and so 0 −→G/Im ν ·ν−→G −→G/Im ν −→0 (6.18) A. NEGUT, is a short exact sequence, which precisely gives rise to an element of Ext1(F,F). The moduli space M is smooth precisely when the dimensions of the tangent spaces (6.17) are locally constant in F (for a more rigorous presentation of the smoothness of the moduli space via obstruction theory, we refer the reader to [18]). We have is a short exact sequence, which precisely gives rise to an element of Ext1(F,F). The moduli space M is smooth precisely when the dimensions of the tangent spaces (6.17) are locally constant in F (for a more rigorous presentation of the smoothness of the moduli space via obstruction theory, we refer the reader to [18]). We have (6.19) dimHom(F,F) −dimExt1(F,F) + dimExt2(F,F) = χ(F,F) (6.19) dimHom(F,F) −dimExt1(F,F) + dimExt2(F,F) = χ(F, (6.22) (6.22) where χ(F,F) = −γ −2rc2 can be computed using the Hirzebruch-Riemann-Roch theorem, and the constant γ only depends on S,H,r,c1. 6.11. Following [27, relation (2.23)], the tangent space to Z1 at a closed point (F0 ⊂x F1) is the vector space of pairs of the form: (w0,w1) ∈Ker Ext1(F0,F0) ⊕Ext1(F1,F1) ψ−→Ext1(F0,F1) (6.23) where the arrow is the difference of the two natural maps induced by the inclusion F0 ⊂F1. These maps ﬁt into the diagram below with exact rows and columns: where the arrow is the difference of the two natural maps induced by the inclusion F0 ⊂F1. These maps ﬁt into the diagram below with exact rows and columns: (6.24) Ext1(F1,F0) Ext1(F0,F0) Ext2(Cx,F0) Ext1(F1,F1) Ext1(F0,F1) Ext2(Cx,F1) Ext1(F1,Cx) Ext1(F0,Cx) Ext2(Cx,Cx) (6.24) where we write F1/F0 = Cx for the skyscraper sheaf at the closed point x ∈S. The di- mensions of the Ext spaces in the diagram above may be computed as in the previous subsection: where we write F1/F0 = Cx for the skyscraper sheaf at the closed point x ∈S. The di- mensions of the Ext spaces in the diagram above may be computed as in the previous subsection: dimExt1(F0,F0) = 1 + ε + γ + 2rcﬁrst 2 dimExt1(F1,F1) = 1 + ε + γ + 2rclast 2 HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Since the dimension of Ext1(F1,F0) is 2 less than that of the tangent space to Z1, this implies that the map dpS is surjective. is a surjective map. Since the dimension of Ext1(F1,F0) is 2 less than that of the tangent space to Z1, this implies that the map dpS is surjective. Proof of Proposition 2.41. — Since we already proved the dimension and irreducibil- ity statements in Proposition 6.9, it remains to prove that Y is smooth. By analogy with the discussion above, we showed in [28, relation (4.37)] that the tangent space to Z2 at a closed point (F0 ⊂x F1 ⊂y F2) consists of triples of the form (w0,w1,w2) ∈Ker Ext1(F0,F0) ⊕Ext1(F1,F1) ⊕Ext1(F2,F2) −→Ext1(F0,F1) ⊕Ext1(F1,F2) (6.28) where the arrow is the alternating sum of the four natural maps induced by the inclusions F0 ⊂F1 ⊂F2. In loc. cit., we also showed that dim space of triples (6.28) = 1 + ε + γ + r(cﬁrst 2 + clast 2 ) + 2 + δ split F2/F0 where cﬁrst 2 and clast 2 are the second Chern classes of the sheaves denoted by F0 and F2, respectively (meanwhile, the Kronecker δ symbol is 1 if x = y and F2/F0 is split, i.e. ∼= C⊕2 x , and 0 otherwise). We conclude that the dimensions of the tangent spaces to Z2 jump by 1 precisely on the split locus. The differential of the map (6.29) Z2 p1 S×p2 S −−→S × S, (F0 ⊂x F1 ⊂y F2) →(x,y) (6.29) admits a presentation analogous to (6.26). It was shown in [28] that the differential dp1 S × dp2 S is surjective if and only if either x ̸= y or x = y and F2/F0 is split. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES dimExt1(F0,F1) = 1 + γ + r(cﬁrst 2 + clast 2 ) dimExt1(F1,F0) = ε + γ + r(cﬁrst 2 + clast 2 ) where cﬁrst 2 and clast 2 are the second Chern classes of the sheaves denoted by F0 and F1, respectively. The fact that the kernel of ψ in (6.23) has the expected dimension 1 + ε + γ + r(cﬁrst 2 + clast 2 ) + 1 then follows from the elementary facts below: where cﬁrst 2 and clast 2 are the second Chern classes of the sheaves denoted by F0 and F1, respectively. The fact that the kernel of ψ in (6.23) has the expected dimension 1 + ε + γ + r(cﬁrst 2 + clast 2 ) + 1 then follows from the elementary facts below: (1) the image of ψ coincides with the kernel of the dotted arrow 2 (2) the target Ext2(Cx,Cx) of the dotted arrow is 1 dimensional (3) the dotted arrow is non-zero if and only if ε = 0 These facts were proved in [27]. As shown in loc. cit., a pair as in (6.23) contains precisely the same information as a commutative diagram with exact rows: (6.25) 0 F1 G1 F1 0 0 F0 G0 F0 0 (6.25) In this language, the differential of the map pS : Z1 →S is given by In this language, the differential of the map pS : Z1 →S is given by (6.26) diagram (6.25) dpS −→ 0 →F1/F0 →G1/G0 →F1/F0 →0 ∈TanxS diagram (6.25) dpS −→ 0 →F1/F0 →G1/G0 →F1/F0 →0 ∈Tanx (6.26) where we use the fact that F1/F0 ∼= Cx and the fact that there exists a canonical isomor- phism TanxS = Ext1(Cx,Cx). A diagram in the kernel of dpS is one in which the extension G1/G0 splits, which precisely means that the diagram (6.25) allows one to insert an extra row, as follows: 0 F1 G1 F1 0 0 F0 H F1 0 0 F0 G0 F0 0 The ability to insert the middle row into the diagram above is equivalent to saying that the pair (6.23) comes from one and the same element in the vector space Ext1(F1,F0) situated in the top left corner of diagram (6.24). We conclude that (6.27) Ext1(F1,F0) ↠Ker dpS (6.27) A. NEGUT, is a surjective map. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES By combining the discussion above with the moduli functor presentation of the scheme Y, we see that Tan(F0⊂F1,F′ 1⊂F2)Y is the space of quadruples (w0,w1,w′ 1,w2) ∈Ext1(F0,F0) ⊕Ext1(F1,F1) ⊕Ext1(F ′ 1,F ′ 1)⊕ ⊕Ext1(F2,F2) which satisfy the four properties below: which satisfy the four properties below: (1) w0 and w1 (or w′ 1) map to the same element of Ext1(F0,F1) (or Ext1(F0,F ′ 1)) (2) w1 (or w′ 1) and w2 map to the same element of Ext1(F1,F2) (or Ext1(F ′ 1,F2)) (3) dpS(w0,w1) = dpS(w′ 1,w2) ∈Ext1(Cx,Cx) (4) dpS(w0,w′ 1) = dpS(w1,w2) ∈Ext1(Cy,Cy) 1 (3) dpS(w0,w1) = dpS(w′ 1,w2) ∈Ext1(Cx,Cx) (4) dpS(w0,w′ 1) = dpS(w1,w2) ∈Ext1(Cy,Cy) By analogy with (6.23) and (6.24), consider the vector space By analogy with (6.23) and (6.24), consider the vector space (6.30) A = Ker " Ext1(F0,F0) ⊕Ext1(F2,F2) ψ′ −→Ext1(F0,F2) # (6.30) HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES where ψ′ is the difference of the two natural maps in the diagram below with exact rows and columns: (6.31) Ext1(F2,F0) Ext1(F0,F0) Ext2(Q,F0) Ext1(F2,F2) Ext1(F0,F2) Ext2(Q,F2) Ext1(F2,Q) Ext1(F0,Q) Ext2(Q,Q) (6.31) (here F2/F0 = Q is a length 2 sheaf which is ﬁltered by Cx and Cy). By analogy with our analysis of (6.24), it is easy to show that the image of the map ψ′ in (6.30) coincides with the kernel of the dotted arrow, and so (6.32) dimA = 1 + ε + γ + r(cﬁrst 2 + clast 2 ) + dimExt2(Q,Q) (6.32) The dimension of Ext2(Q,Q) is 4 if x = y and Q is split, and 2 otherwise. Items (1)–(4) above imply that we have a Cartesian diagram of vector spaces: (6.33) Tan(F0⊂F1,F′ 1⊂F2)Y a′ b′ Tan(F0⊂yF′ 1⊂xF2)Z2 (b,dp2 S,dp1 S) Tan(F0⊂xF1⊂yF2)Z2 (a,dp1 S,dp2 S) A ⊕TanxS ⊕TanyS (6.33) where the maps a,a′ forget w1 and the maps b,b′ forget w′ 1. where the maps a,a′ forget w1 and the maps b,b′ forget w′ 1. Claim 6.12. — The map a is injective, unless x = y in which case Ker a is one-dimensional and spanned by (0,w1,0), where w1 represents the following extension: 6.34) 0 −→F1 (inclusion,0) −−−−−→F2 ⊕Cx F1 (0,projection) −−−−−→F1 −→0 (6.34) (the middle space requires ﬁxing isomorphisms F2/F1 ∼= F1/F0 ∼= Cx). The image of the extension (6.34) under dp1 S × dp2 S is equal to (v,v) ∈TanxS ⊕TanxS, where v ∈Ext1(Cx,Cx) is the class of the extension 0 →F1/F0 →F2/F0 →F2/F1 →0. We will ﬁrst show how Claim 6.12 allows us to prove that all tangent spaces to Y have dimension ≤than (6.35) dimY Prop. 6.9 = dimZ2 Prop. 2.28 = 1 + ε + γ + r(cﬁrst 2 + clast 2 ) + 2 (which would conclude the proof of Proposition 2.41) and then prove the claim (6.35) dimY Prop. 6.9 = dimZ2 Prop. 2.28 = 1 + ε + γ + r(cﬁrst 2 + clast 2 ) + 2 (6.35) (which would conclude the proof of Proposition 2.41) and then prove the claim (which would conclude the proof of Proposition 2.41) and then prove the cla A. NEGUT, Case 1: when x ̸= y or x = y and F2/F0 is not split (i.e. we are at a smooth point of Z2), Claim 6.12 implies that (a,dp1 S,dp2 S) is injective. Then the fact that diagram (6.33) is Cartesian implies that the map a′ is also injective, which implies dimTan(F0⊂F1,F′ 1⊂F2)Y ≤dimTan(F0⊂yF′ 1⊂xF2)Z2 dimTan(F0⊂F1,F′ 1⊂F2)Y ≤dimTan(F0⊂yF′ 1⊂xF2)Z2 Because of Proposition 6.9 and the fact that we are on the smooth locus of Z2, the dimen- sions of the two tangent spaces must be equal. Case 2: when x = y and F2/F0 is split, the dimensions of the vector spaces in (6.33) are (6.33) are ? a′ b′ Cd+1 (b,dp2 S,dp1 S) Cd+1 (a,dp1 S,dp2 S) Cd+2 ⊕C2 ⊕C2 where d is the number in the right-hand side of (6.35). The goal is to show that the Carte- sian product of the diagram, namely the vector space ?, has dimension d. If F1 = F ′ 1, then the northeast and southwest corners are naturally identiﬁed, as are the maps a and b. By Claim 6.12, we can decompose Cd+1 = C ⊕Cd, where a(C) = 0 and a\|Cd is injective. Moreover, dp1 S(C) = dp2 S(C) = 0, which implies that dp1 S × dp2 S\|Cd is surjective (this follows from the surjectivity of dp1 S ×dp2 S, proved in [28, relation (4.47)]). We conclude that a point of ? is of the form where d is the number in the right-hand side of (6.35). (6.34) The goal is to show that the Carte- sian product of the diagram, namely the vector space ?, has dimension d. If F1 = F ′ 1, then the northeast and southwest corners are naturally identiﬁed, as are the maps a and b. By Claim 6.12, we can decompose Cd+1 = C ⊕Cd, where a(C) = 0 and a\|Cd is injective. Moreover, dp1 S(C) = dp2 S(C) = 0, which implies that dp1 S × dp2 S\|Cd is surjective (this follows from the surjectivity of dp1 S ×dp2 S, proved in [28, relation (4.47)]). We conclude that a point of ? is of the form (l,l′,v) ∈C ⊕C ⊕Cd such that dp1 S(v) = dp2 S(v). The latter equality imposes two non-trivial linear conditions on v, so we conclude that the dimension of ? is 1 + 1 + d −2 = d. such that dp1 S(v) = dp2 S(v). The latter equality imposes two non-trivial linear conditions on v, so we conclude that the dimension of ? is 1 + 1 + d −2 = d. If F1 ̸= F ′ 1, then it sufﬁces to prove that Im a and Im b are transversal d- dimensional subspaces of A. Consider a point of A given by a pair of extensions (6.36) 0 F2 G2 F2 0 0 F0 G0 F0 0 (6.36) 0 F2 G2 F2 0 0 F0 G0 F0 0 (6.36) The diagram above induces an extension at the level of quotients The diagram above induces an extension at the level of quotients (6.37) 0 →Q →H →Q →0 where Q = F2/F0 and H = G2/G0. The p (6.37) 0 →Q →H →Q →0 where Q = F2/F0 and H = G2/G0. The pair of extensions (6.36) lies in Im a iff (6.37) (6.38) H has a length 2 subscheme compatible with F1/F0 ⊂Q (6.38) H has a length 2 subscheme compatible with F1/F0 ⊂Q HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES (and similarly for Im b, if we replace F1 by F ′ 1). Fix a vector space isomorphism (and similarly for Im b, if we replace F1 by F ′ 1). Fix a vector space isomorphism F2/F0 ∼= C2 with respect to which F1/F0 is the ﬁrst standard coordinate line and F ′ 1/F0 is the second coordinate line. The rank 4 coherent sheaf H is determined on the local neighborhood of x ∈S by two commuting 4 × 4 matrices X and Y, whose only non-zero entries are allowed to be in the bottom left 2 × 2 block, as below: X = ⎛ ⎜⎜⎝ 0 0 0 0 0 0 0 0 x1 x2 0 0 x3 x4 0 0 ⎞ ⎟⎟⎠ and Y = ⎛ ⎜⎜⎝ 0 0 0 0 0 0 0 0 y1 y2 0 0 y3 y4 0 0 ⎞ ⎟⎟⎠ Condition (6.38) is equivalent to x3 = y3 = 0, while the analogous condition with F1 replaced by F ′ 1 is equivalent to x2 = y2 = 0. (l,l′,v) ∈C ⊕C ⊕Cd Taken together, this would prove the desired fact that the images of a and b are transverse codimension two subspaces of A, as soon as we prove that the map A α−→Ext1(Q,Q) (6.36) →(6.37) A α−→Ext1(Q,Q) (6.36) →(6.37) is surjective. To this end, note that the dimension of A is given by (6.32), dimExt2(Q,Q) = 4, dimExt1(Q,Q) = 8. Therefore, it sufﬁces to show that is surjective. To this end, note that the dimension of A is given by (6.32), dimExt2(Q,Q) = 4, dimExt1(Q,Q) = 8. Therefore, it sufﬁces to show that dimKer α ≤1 + ε + γ + r(cﬁrst 2 + clast 2 ) −4 By analogy with (6.27), we have a surjective map Ext1(F2,F0) β↠Ker α A simple application of Hirzebruch-Riemann-Roch shows that dimExt1(F2,F0) = ε + γ + r(cﬁrst 2 + clast 2 ). Therefore, it sufﬁces to show that the kernel of the map β has dimension ≥3. To this end, consider the following piece of the Ext long exact sequence corresponding to the short exact sequence 0 →F0 →F2 →Q →0: Hom(F2,F2) Hom(F2,Q) ρ Ext1(F2,F0) Ext1(F2,F2) Hom(F0,Q) Ext1(F0,F0) Hom(F0,Q) Ext1(F0,F0) Consider the 4-dimensional subspace V ⊂Hom(F2,Q) which consists of a ﬁxed homo- morphism with kernel F0, composed with an arbitrary endomorphism of Q ∼= C2 x. Any element of V maps to 0 in both Ext1(F0,F0) and Ext1(F2,F2), so ρ(V) lies inside the Consider the 4-dimensional subspace V ⊂Hom(F2,Q) which consists of a ﬁxed homo- morphism with kernel F0, composed with an arbitrary endomorphism of Q ∼= C2 x. Any element of V maps to 0 in both Ext1(F0,F0) and Ext1(F2,F2), so ρ(V) lies inside the A. NEGUT, kernel of β. However, the map ρ has a 1-dimensional kernel since Hom(F2,F2) = C, so we conclude that Ker β has dimension at least 3. Proof of Claim 6.12. — It is enough to show that any triple (0,w1,0) satisfying Proof of Claim 6.12. — It is enough to show that any triple (0,w1,0) satisfying 0 and w1 map to the same element in Ext1(F0,F1) 0 and w1 map to the same element in Ext1(F0,F1) w1 and 0 map to the same element in Ext1(F1,F2) must have w1 equal to a multiple of the extension (6.34). (l,l′,v) ∈C ⊕C ⊕Cd Indeed, the natural long exact sequences imply that it sufﬁces to show that any w1 ∈Ext1(F1,F1) which lies in the intersection of the images of Ext1(Cx,F1) and Hom(F1,Cy) is a multiple of (6.34), where Cx = F1/F0 and Cy = F2/F1. In other words, if we have a diagram (6.39) 0 F1 F2 taut Cy 0 0 F1 G F1 α taut 0 0 F1 H β Cx 0 (6.39) (the maps denoted “taut” are the projection maps F1 ↠Cx and F2 ↠Cy that give rise to the ﬂag F0 ⊂F1 ⊂F2) where the middle short exact sequence is the pull-back of both the top and the bottom short exact sequences, we must show that the middle short exact sequence is a multiple of (6.34). As the sheaves F0,F1,F2,H all have the same reﬂexive hull, and since the reﬂexive hull is stable, we may regard F0,F1,F2,H as subsheaves of the same stable vector bundle V. (the maps denoted “taut” are the projection maps F1 ↠Cx and F2 ↠Cy that give rise to the ﬂag F0 ⊂F1 ⊂F2) where the middle short exact sequence is the pull-back of both the top and the bottom short exact sequences, we must show that the middle short exact sequence is a multiple of (6.34). As the sheaves F0,F1,F2,H all have the same reﬂexive hull, and since the reﬂexive hull is stable, we may regard F0,F1,F2,H as subsheaves of the same stable vector bundle V. (1) If H ̸= F2 (as subsheaves of V), then the two sides of the inclusion F1 ⊆H ∩F2 have the same colength as subsheaves of V, so the inclusion above is an equality. Similarly, the two sides of the inclusion have the same colength as subsheaves of V, so the inclusion above is an equality. Similarly, the two sides of the inclusion G ⊆(H ∩F2) ⊕F1 = F1 ⊕F1 have the same colength as subsheaves of V ⊕V, so the inclusion above is an equality. But then the short exact sequence 0 →F1 →G →F1 →0 is split, so w1 = 0. have the same colength as subsheaves of V ⊕V, so the inclusion above is an equality. But then the short exact sequence 0 →F1 →G →F1 →0 is split, so w1 = 0. (2) if H = F2 (as subsheaves of V), then x = y. (l,l′,v) ∈C ⊕C ⊕Cd Since Hom(F1,F2) is one dimen- sional, any two injections F1 →F2 are scalar multiples of each other, which HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES implies that β = λ · taut for some λ ∈C× (notations as in (6.39)). Then the extension G is equal to λ times the extension (6.34), as we needed to show. □ 6.13. Let us now consider the schemes Y−, Y+, Y−+ in relation to Y. Proof of Propositions 2.42 and 2.43. — Since we already proved the dimension and irreducibility statements in Proposition 6.9, it remains to prove that Y−, Y+ and Y−+ are l.c.i. This is a consequence of the claim that the following squares are all derived, which we will now prove: (6.40) Y+ Z(x,x) Y Z(x) (6.41) Y−+ Z(x,x) Y− Z(x) (6.40) Y+ Z(x,x) Y Z(x) Y− Z(x,x) Y Z(x) (6.41) Y−+ Z(x,x) Y− Z(x) Y−+ Z(x,x) Y+ Z(x) In all cases above, the arrow on the left is the only map one can write which forgets a single sheaf (in the notation of (2.40)–(2.43)), while the arrow on the right is the unique map which forgets the same sheaf as the arrow on the left. Y− Z(x,x) Y Z(x) Y−+ Z(x,x) Y+ Z(x) (6.40) (6.41) In all cases above, the arrow on the left is the only map one can write which forgets a single sheaf (in the notation of (2.40)–(2.43)), while the arrow on the right is the unique map which forgets the same sheaf as the arrow on the left. In all cases above, the arrow on the left is the only map one can write which forgets a single sheaf (in the notation of (2.40)–(2.43)), while the arrow on the right is the unique map which forgets the same sheaf as the arrow on the left. We will only prove the fact that the ﬁrst square in (6.40) is derived, since all other cases are analogous. Consider the map on the left of the square Y+ −→Y (6.42) With the notation in (2.40) and (2.42), we note that the ﬁbers of this map consist of all ways to append a sheaf F−1 ⊂x F0 to diagram (2.40). Just like in Proposition 2.19, one sees that the map (6.42) factors as With the notation in (2.40) and (2.42), we note that the ﬁbers of this map consist of all ways to append a sheaf F−1 ⊂x F0 to diagram (2.40). Just like in Proposition 2.19, one sees that the map (6.42) factors as Y+ ι PY(x∗(V0)) Y Y+ ι PY(x∗(V0)) Y where x : Y →Y × S is the graph of the map px S that records the support point x ∈S. The closed embedding ι is cut out by the composition where x : Y →Y × S is the graph of the map px S that records the support point x ∈S. The closed embedding ι is cut out by the composition σ : x∗(W0) −→x∗(V0) −→O(1) A. NEGUT, and just like in Proposition 2.21, one may show that the section σ factors through a locally free sheaf of rank 1 less, as follows: and just like in Proposition 2.21, one may show that the section σ factors through a locally free sheaf of rank 1 less, as follows: σ : x∗(W0) ↠ x∗(W0) L1 ⊗px∗ S (ωS) σ ′ −→O(1) (the argument requires the fact that px S is ﬂat, which is proved by estimating the dimensions of its ﬁbers, akin to the proof of Proposition 6.2). Because of (2.2), we obtain: (the argument requires the fact that px S is ﬂat, which is proved by estimating the dimensions of its ﬁbers, akin to the proof of Proposition 6.2). Because of (2.2), we obtain: dimY+ −dimY ≥r However, Proposition 2.41 implies that Y is smooth, while Proposition 6.9 implies that dimY+ = dimY + r. Therefore, we actually have equality in the inequality above. This is a particular case of Deﬁnition 2.2, hence the section σ ′ is regular. However, this is precisely the same section that describes the map Z(x,x) →Z(x). By Deﬁnition 2.4, this precisely says that the ﬁrst ﬁber square in (6.40) is derived. □ Proof of Proposition 2.44. — The scheme Y is reduced because it is smooth. As for the other schemes, they are local complete intersections, so it sufﬁces to prove that their generic points are reduced. (6.42) In the case of Y−and Y+, they are irreducible, and the generic point corresponds to a diagram (2.41)–(2.42) with x ̸= y. Near such a point, Y− and Y+ are isomorphic to Z(y,x,x) and Z(x,x,y), respectively. Since the latter schemes are normal (due to Proposition 2.34), reducedness follows. The same argument applies to the irreducible component of Y−+ which is the closure of the locus of diagrams (2.43) with x ̸= y. As for the other irreducible component, we recall that it corresponds to diagrams (2.43) with x = y and F3/F1 a split length 2 sheaf. Therefore, the second component is locally isomorphic to V1 × P1 where V1 ⊂Z(x,x,x,x) is the irreducible component consisting of (F0 ⊂x F1 ⊂x F2 ⊂x F3 ⊂x F4) such that F3/F1 is a split length 2 sheaf. It sufﬁces to show that V1 is generically reduced. As a consequence of Corollary 5.14, the generic point of V1 corresponds to F4 locally free. Lemma 5.7 implies that near such a point, V1 is locally isomorphic to the smooth moduli space M4 = {F4} times the component Z1/B ⊂Comm4/B that we studied in Section 5.10. As we noted therein, Z1 is generically reduced, so we are done. □ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. HECKE CORRESPONDENCES FOR SMOOTH MODULI SPACES OF SHEAVES Alexander Minets, Georg Oberdieck, Alexei Oblomkov, Andrei Okounkov, Hiraku Nakajima, Francesco Sala, Olivier Schiffmann, Richard Thomas, Alexander Tsymbal- iuk and the anonymous referee for many interesting discussions on the subject. I would like to thank MSRI, Berkeley, for their hospitality while this paper was being written in the Spring semester of 2018. I gratefully acknowledge the support of NSF grants DMS- 1600375 and DMS-1440140. Acknowledgements I would like to thank Mina Aganagic, Roman Bezrukavnikov, Tom Bridgeland, Eu- gene Gorsky, Sergei Gukov, Tamas Hausel, Ivan Loseu, Alina Marian, Davesh Maulik, REFERENCES 1. D. ARINKIN, A. C ˘ALD˘ARARU and M. HABLICSEK, Formality of derived intersections and the orbifold HKR isomorphism, J. Algebra, 540 (2019), 100–120. . V. BARANOVSKY, Moduli of sheaves on surfaces and action of the oscillator algebra, J. Differ. Geom., 55 (2000), 3. R. BASILI, On commuting varieties of upper triangular matrices, Commun. Algebra, 45 (2017), 1533–1541. BURBAN and O. SCHIFFMANN, On the Hall algebra of an elliptic curve I, Duke Math. J., 161 (2012), 1171–1231 5. E. CARLSSON, E. GORSKY and A. MELLIT, The Aq,t algebra and parabolic ﬂag Hilbert schemes, Math. Ann., 376 (2020), 1303–1336. . CAUTIS and A. LICATA, Heisenberg categoriﬁcation and Hilbert schemes, Duke Math. J., 161 (2012), 2469–254 7. I. CHEREDNIK, Double Afﬁne Hecke Algebras, Cambridge University Press, Cambridge, 2005. xii+434 pp. ISBN 0-521- 60918-6. 8. J. DING and K. IOHARA, Generalization of Drinfeld quantum afﬁne algebras, Lett. Math. Phys., 41 (1997), 181–193. . G. ELLINGSRUD and M. LEHN, Irreducibility of the punctual quotient scheme of a surface, Ark. Mat., 3 245–254. LLINGSRUD and A. STRØMME, On the homology of the Hilbert schemes on points in the plane, Invent. Math., 8 7), 343–352. 11. B. FEIGIN and A. TSYMBALIUK, Heisenberg action in the equivariant K-theory of Hilbert schemes via Shufﬂe Algebra, Kyoto J. Math., 51 (2011), 831–854. A. NEGUT, 12. B. FEIGIN, K. HASHIZUME, A. HOSHINO, J. SHIRAISHI and S. YANAGIDA, A commutative algebra on degenerate CP1 and MacDonald polynomials, J. Math. Phys., 50, 095215 (2009). 13. L. FU and M.-T. NGUYEN, Orbifold products for higher K-theory and motivic cohomology, Doc. Math., 24 (2019), 1769–1810. 14. E. GORSKY and A. NEGUT, , The trace of the afﬁne Hecke category, 2201.07144. 15. E. GORSKY, A. NEGUT, and J. RASMUSSEN, Flag Hilbert schemes, colored projectors and Khovanov-Rozansky homol- ogy, Adv. Math., 378, 107542 (2021). 115 pp. 16. L. GÖTTSCHE, The Betti numbers of the Hilbert scheme of points on a smooth projective surface, Math. Ann., 286 (1990), 193–207. 17. I. GROJNOWSKI, Instantons and afﬁne algebras I. The Hilbert scheme and vertex operators, Math. Res. Lett., 3 (1996), 275–291. 18. D. HUYBRECHTS and M. LEHN, The Geometry of Moduli Spaces of Sheaves, 2nd ed., Cambridge University Press, 2010. ISBN 978-0-521-13420-0. 19. A. KRUG, Symmetric quotient stacks and Heisenberg actions, Math. Z., 288 (2018), 11–22. 20. D. MAULIK and A. NEGUT, , Lehn’s formula in Chow and conjectures of Beauville and Voisin, J. Inst. Math. Jussieu, 21 (2022), 933–971. REFERENCES https://doi.org/10.1017/S1474748020000377. 21. D. MAULIK and A. OKOUNKOV, Quantum Groups and Quantum Cohomology, Astérisque, vol. 408, 2019. ix+209 pp. ISBN 978-2-85629-900-5. 22. K. MIKI, A (q,γ ) analog of the W1+∞algebra, J. Math. Phys., 48, 123520 (2007). 23. H. NAKAJIMA, Heisenberg algebra and Hilbert schemes of points on projective surfaces, Ann. Math., 145 (1997), 379–388. 24. A. NEGUT, , The shufﬂe algebra revisited, Int. Math. Res. Not., 22 (2014), 6242–6275. 25. A. NEGUT, , Moduli of ﬂags of sheaves and their K-theory, Algebr. Geom., 2 (2015), 19–43. 27. A. NEGUT, , Shufﬂe algebras associated to surfaces, Sel. Math. (N.S.), 25, 36 (2019). 57 pp. EGUT, , W-algebras associated to surfaces, Proc. Lond. Math. Soc. (2022). https://doi.org/10.1112/plms.12435. 28. A. NEGUT, , W-algebras associated to surfaces, Proc. Lond. Math. Soc. (2022). https://doi.org/ 29. A. NEGUT, , AGT relations for sheaves on surfaces, 1711.00390. 30. O. SCHIFFMANN, Drinfeld realization of the elliptic Hall algebra, J. Algebraic Comb., 35 (2012), 237–262 31. O. SCHIFFMANN and E. VASSEROT, The elliptic Hall algebra and the equivariant K-theory of the Hilbert scheme of A2, Duke Math. J., 162 (2013), 279–366. 32. O. SCHIFFMANN and E. VASSEROT, Cherednik algebras, W-algebras and the equivariant cohomology of the moduli space of instantons on A2, Publ. Math. Inst. Hautes Études Sci., 118 (2013), 213–342. 33. B. TOËN, Proper local complete intersection morphisms preserve perfect complexes, 1210.2827. 34. C. VOISIN, On the Chow ring of certain algebraic hyper-Kahler manifolds, Pure Appl. Math. Q., 4 (2008), 613–649. A. N. Department of Mathematics MIT Cambridge, MA USA and Simion Stoilow Institute of Mathematics Bucharest Romania andrei.negut@gmail.com
https://openalex.org/W4293254786	https://www.researchsquare.com/article/rs-1109324/latest.pdf	English	null	Nomogram Models Based on the Gene Expression in Prediction of Breast Cancer Bone Metastasis	Journal of healthcare engineering	2,022	cc-by	5,681	Nomogram Models Based on Gene Expression in Prediction of Breast Cancer Bone Metastasis Teng-di Fan Ningbo Medical Center Lihuili Hospital Di-kai Bei Ningbo Medical Center Lihuili Hospital Song-wei Li (  lisongweitg@163.com ) Ningbo Medical Center Lihuili Hospital Research Article Keywords: Breast cancer, bone metastasis, gene signature, prognosis Posted Date: November 30th, 2021 DOI: https://doi.org/10.21203/rs.3.rs-1109324/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/19 Page 1/19 Abstract Objective: To design a weighted co-expression network and build gene expression signature-based nomogram (GESBN) models for predicting the likelihood of bone metastasis in breast cancer (BC) patients. Methods: Dataset GSE124647 was used as a training set, and GSE14020 was taken as a validation set. In the training cohort, limma package in R was adopted to obtain differentially expressed genes (DEGs) between BC non-bone metastasis and bone metastasis patients, which were used for functional enrichment analysis. After weighted co-expression network analysis (WGCNA), univariate Cox regression and Kaplan-Meier plotter analyses were performed to screen potential prognosis-related genes. Then, GESBN models were constructed and evaluated. Further, the expression levels of genes in the models were explored in the training set, which was validated in GSE14020. Finally, the prognostic value of hub genes in BC was explored. Results: A total of 1858 DEGs were obtained. WGCNA result showed that the blue module was most significantly related to bone metastasis and prognosis. After survival analyses, GAJ1, SLC24A3, ITGBL1, and SLC44A1 were subjected to construct a GESBN model for overall survival. While GJA1, IGFBP6, MDFI, ITGFBI, ANXA2, and SLC24A3 were subjected to build a GESBN model for progression-free survival. Kaplan-Meier plotter and receiver operating characteristic analyses presented the reliable prediction ability of the models. Besides, GJA1, IGFBP6, ITGBL1, SLC44A1, and TGFBI expressions were significantly different between the two groups in GSE124647 and GSE14020. The hub genes had a significant impact on patient prognosis. Conclusion: Both the four-gene signature and six-gene signature could accurately predict patient prognosis, which may provide novel treatment insights for BC bone metastasis. Introduction Breast cancer (BC) is one of the most prevalent malignancies and the major cause of cancer-associated deaths of women worldwide [1]. BC is considered to have the highest diagnostic rate in cancer, with more than 1.6 million new cases detected a year, accounting for approximately one-quarter of all cancers in women [2]. Despite the substantial improvements in prognosis have been achieved due to better therapeutic approaches over the past 20 years, BC is still one of the most common causes of cancer deaths among females. According to statistics, a total of 522,000 women died from BC in 2012, while 627,000 individuals died from this disease in 2018 [3, 4]. Nearly 70% of BC patients developed bone metastasis, leading to osteolytic and osteoblastic cancers [5]. Tumor cells secreted factors including parathyroid hormone in the bone to create an environment conducive to osteolysis instead of the direct destruction of bone [6]. In addition, bone metastasis often contributes to adverse events such as hypercalcemia, spinal cord compression, fractures, and pain, which severely affect the quality of life in BC patients [7]. Magnetic resonance imaging, computed tomography, Page 2/19 Page 2/19 Page 2/19 X-ray are conventional imaging methods to detect bone metastasis but fail to sense tiny tumor masses and negligible tumor-induced osteolysis [8]. Besides, various metastatic bone lesions which resulted from BC are hard to eradicate by adjuvant localized radiotherapy or surgical intervention [9]. Therefore, clinicians are faced with a big challenge in the diagnosis and treatment of bone metastatic BC. Currently, pathologic diagnosis has been considered as the gold standard for diagnosis and pathology- based classification, which is critical in guiding the treatment of BC patients [10]. Nevertheless, gene expression also represents an essential role in the prognosis of patients, thus providing clinically relevant information and targeted therapies [11, 12]. Hence, it’s imperative to find novel biomarkers to improve the early diagnosis and prognosis of patients with bone metastatic BC by using public databases of gene expression data. In this study, 1858 differentially expressed genes (DEGs) between BC non-bone metastasis and BC bone metastasis were obtained, of which 76 genes were probably related to bone metastasis and prognosis based on weighted co-expression network analysis (WGCNA). With univariate Cox regression and Kaplan- Meier plotter analyses, a four-gene expression signature-based nomogram model for overall survival (OS) and a six-gene expression signature-based nomogram model for progression-free survival (PFS) were constructed, respectively. Introduction Kaplan-Meier plotter and receiver operating characteristic (ROC) analyses validated the potential value of the models for predicting patient survival. Finally, the expression of the genes in nomogram models and the prognostic value of hub genes were initially explored and validated. In this study, 1858 differentially expressed genes (DEGs) between BC non-bone metastasis and BC bone metastasis were obtained, of which 76 genes were probably related to bone metastasis and prognosis based on weighted co-expression network analysis (WGCNA). With univariate Cox regression and Kaplan- Meier plotter analyses, a four-gene expression signature-based nomogram model for overall survival (OS) and a six-gene expression signature-based nomogram model for progression-free survival (PFS) were constructed, respectively. Kaplan-Meier plotter and receiver operating characteristic (ROC) analyses validated the potential value of the models for predicting patient survival. Finally, the expression of the genes in nomogram models and the prognostic value of hub genes were initially explored and validated. Data mining from gene expression omnibus (GEO) database GEO database (https://www.ncbi.nlm.nih.gov/geo/) was used to obtain the BC microarray dataset by setting the following filter: (1) more than 50 samples with bone metastasis information; (2) with survival data; (3) with expression profiling data. Finally, GSE124647 dataset was chosen as a training set to identify the DEGs between non-bone metastasis and bone metastasis samples. The platform was Affymetrix Human Genome U133A Array (GPL96). In total, there were 140 samples containing clinical and RNA-seq expression data in the GSE124647. Besides, 65 samples in GSE14020 dataset were used as validation cohorts to verify the expression levels of key genes. Normalized gene expression was measured as log2-based transformation. WGCNA WGCNA is a systemic method that uses gene expression data to build a scale-free network [13]. A weighted co-expression network with the expression profile data of the DEGs was built using the WGCNA package of R. Following this, we screened the key module related to BC bone metastasis and prognosis, and then extracted the genes for further analysis. WGCNA is a systemic method that uses gene expression data to build a scale-free network [13]. A weighted co-expression network with the expression profile data of the DEGs was built using the WGCNA package of R. Following this, we screened the key module related to BC bone metastasis and prognosis, and then extracted the genes for further analysis. The expression levels of prognostic genes in nomogram models The expression levels of key genes between BC non-bone metastasis and BC bone metastasis groups in GSE124647 were firstly explored using t-test. Then, GSE14020 as a validation dataset was used to assess the differential expression of the key genes in two groups. Nomogram model construction and model effectiveness evaluation Using the “survival” package in R, univariate Cox regression analysis was performed to obtain the potential prognostic genes based on OS and PFS. Only genes that had a significant impact on OS or PFS were considered to pass univariate Cox regression analysis screening. In addition, the prognostic value of the significant genes obtained in the univariate Cox regression analysis was evaluated by Kaplan-Meier plotter analysis. Only genes with statistical significance in OS or PFS analyses were considered to pass the screening. The intersected genes generated in univariate Cox regression and Kaplan-Meier plotter analyses were then entered into the construction of nomogram models in terms of OS and PFS using the “rms” package in R. The calibration curves were drawn to measure the performance of the two models. The genes which had the greatest contribution were selected as hub genes. After that, the patients were divided into high-risk or low-risk groups using the optimal cut-off value of risk score, which was calculated by the “maxstat” package in R. And Kaplan-Meier plotter analyses were adopted to assess the survival difference between the two groups using “survfit” function of “survival” package in R. Moreover, the time-independent ROC analyses were conducted to further evaluate the prognostic value of the nomogram models by using the “pROC” package in R. P <0.05 was considered as significantly different. The area under curve (AUC) was used as an indicator of prognostic accuracy. Validation of the prognostic value of the hub genes Kaplan-Meier plotter (http://kmplot.com/analysis/index.php?p=background) is capable to assess the effect of 54,000 genes on survival in 21 cancer types. We used this database to verify the prognostic significance of the hub genes in BC. Survival curves were generated by the Kaplan-Meier method using the log-rank test. Log-rank P value less than 0.05 was statistically significant. Identification and functional enrichment analysis of DEGs The R package limma was used to screen the DEGs between BC non-bone metastasis and BC bone metastasis groups. \|log2 FC\|>1 and P-value <0.05 were set as the filtering parameters. Then, gene ontology (GO) including biological process (BP), cellular component (CC), and molecular function (MF), and KEGG were carried out to determine the major biological functions of these DEGs in the database for annotation, visualization, and integrated discovery (DAVID) (https://david.ncifcrf.gov/summary.jsp). P <0.05 was considered statistically significant. Page 3/19 Page 3/19 WGCNA To further analyze the module, we calculated the eigengenes of each module and merged the modules by setting a height of 0.25. Finally, a total of 4 modules were acquired (Figure 3C-3D). The genes in the grey module could not be incorporated into any other module. Next, Pearson’s correlation coefficients of the module eigengene of each module and the sample characteristics were calculated. The blue module with 76 genes was closely related to bone metastasis and survival status (Figure 3E). Thus, the genes in the blue module were chosen for further analysis. Identification and functional enrichment of DEGs Identification and functional enrichment of DEGs Page 4/19 Taking BC non-bone metastasis samples as a control group, 1858 DEGs in the training set including 992 up-regulated and 866 down-regulated genes were generated according to the selection criteria. The volcano plot and heat map of the DEGs were presented in Figure 1A, and Figure 1B, respectively. Taking BC non-bone metastasis samples as a control group, 1858 DEGs in the training set including 992 up-regulated and 866 down-regulated genes were generated according to the selection criteria. The volcano plot and heat map of the DEGs were presented in Figure 1A, and Figure 1B, respectively. To have a biological understanding of these DEGs, they were subjected to the DAVID database for GO annotation and KEGG pathway enrichment analysis. The top enriched GO terms in BPs were signal transduction, positive regulation of transcription from RNA polymerase II promoter, and immune response, and those in CCs were cytoplasm, cytosol, and extracellular exosome (Figure 2A-2B). The major MFs were protein binding, Poly(A) RNA binding, and identical protein binding (Figure 2C). In the KEGG pathway enrichment analysis, these genes were mainly involved in the MAPK signaling pathway, proteoglycans in cancer, and focal adhesion (Figure 2D). WGCNA We incorporated the expression profile of integrated DEGs with clinical traits of the BC samples to construct a gene co-expression network. Clinical characteristics including sample group, PFS time, OS time, OS status, and PFS status were clustered with expression matrix (Figure 3A). Then, we chose the optimal β=6 to ensure that network was scale-free (β was a soft-thresholding parameter that could emphasize strong correlations between genes and penalize weak correlations). After choosing the power of 2, the adjacency was transformed into a topological overlap matrix (TOM), which could measure the network connectivity of a gene defined as the sum of its adjacency with all other genes for the network gene ration, and the corresponding dissimilarity (1-TOM) was calculated (Figure 3B). Based on TOM, the average linkage hierarchical clustering was conducted to cluster genes by setting the minimum number of genes for each gene network module to 30. To further analyze the module, we calculated the eigengenes of each module and merged the modules by setting a height of 0.25. Finally, a total of 4 modules were acquired (Figure 3C-3D). The genes in the grey module could not be incorporated into any other module. Next, Pearson’s correlation coefficients of the module eigengene of each module and the sample characteristics were calculated. The blue module with 76 genes was closely related to bone metastasis and survival status (Figure 3E). Thus, the genes in the blue module were chosen for further analysis. We incorporated the expression profile of integrated DEGs with clinical traits of the BC samples to construct a gene co-expression network. Clinical characteristics including sample group, PFS time, OS time, OS status, and PFS status were clustered with expression matrix (Figure 3A). Then, we chose the optimal β=6 to ensure that network was scale-free (β was a soft-thresholding parameter that could emphasize strong correlations between genes and penalize weak correlations). After choosing the power of 2, the adjacency was transformed into a topological overlap matrix (TOM), which could measure the network connectivity of a gene defined as the sum of its adjacency with all other genes for the network gene ration, and the corresponding dissimilarity (1-TOM) was calculated (Figure 3B). Based on TOM, the average linkage hierarchical clustering was conducted to cluster genes by setting the minimum number of genes for each gene network module to 30. Evaluation of the GESBN models The GESBN score was calculated for each patient in the training set. Patients were ranked based on their risk scores and assigned into two groups as high-risk and low-risk of bone metastases. Using the survival package in R for survival analysis, the results showed that the OS rate of patients in the high-risk group was low, and the difference between the two groups was statistically significant (Figure 8A) (P <0.001). Similarly, an unfavorable PFS was observed in the high-risk group patients (Figure 8B) (P <0.001), suggesting that two nomogram models could predict survival well. Further, the time-independent ROC curves were drawn using the pROC package in R. In terms of OS, the AUCs of the 1-, 3-, and 5-year survival rates were 0.68, 0.62, and 0.72, respectively (Figure 8C). For PFS, the AUCs of the 1-, 3-, and 5-year survival rates were 0.73, 0.88, 0.94, respectively (Figure 8D), indicating that nomograms had a good predictive ability. The expression levels of genes in nomogram models Due to the predictive ability of nomogram models for both OS and PFS, we explored the expression levels of these key genes. In the training dataset of GSE124647, the expression levels of all the prognosis- related genes were significantly different between control and bone-metastasis groups (Figure 9A) (all P <0.05). In the validation dataset of GSE14020, GJA1, IGFBP6, ITGBL1, SLC44A1, and TGFBI expressions in the bone-metastasis group were different from those in the control group. The differences were statistically significant (Figure 9B) (P<0.05). Construction of the gene expression signature-based nomogram (GESBN) model Univariate Cox regression and Kaplan-Meier plotter analyses were carried out on 140 patients in the GSE124647 to evaluate the association of 76 gene expression profiles in the blue module with patient OS and PFS. In univariate Cox regression analysis, significant genes related to OS were SLC44A1, SLC24A3, PDGFC, ITGBL1, and GJA1 (Figure 4A) (all P <0.05). Eleven genes including MDFI, IGFBP6, GJA1, ANXA2, SLC24A3, TGFBI, CELA2A, CELA2B, CLEC11A, PPEF2, and SLC44A1 were notably linked to PFS (Figure Page 5/19 Page 5/19 4B) (all P <0.05). However, only four genes related to OS, and six genes related to PFS with statistical differences were extracted in Kaplan-Meier plotter analysis (Figure 5 and Figure 6) (all P <0.05). Taken together, GAJ1, SLC24A3, ITGBL1, and SLC44A1 were defined as potential prognostic genes for OS. GJA1, IGFBP6, MDFI, ITGFBI, ANXA2, and SLC24A3 were potential genes correlated with PFS. These prognostic genes were then subjected to the construction of nomogram models based on OS and PFS (Figure 7A- 7B). For OS, SLC44A1 had the greatest contribution, which could reach 100 points, while MDFI contributed most to PFS. Therefore, SLC44A1 and MDFI were considered as hub genes. To ensure the accuracy of the GESBN models, the calibration curves were drawn. The calibration curves showed good agreement between prediction and observation in the probability of 1-, 3- and 5- year survival (Figure 7C- 7D), indicating that the accuracy of the nomogram model was reliable. 4B) (all P <0.05). However, only four genes related to OS, and six genes related to PFS with statistical differences were extracted in Kaplan-Meier plotter analysis (Figure 5 and Figure 6) (all P <0.05). Taken together, GAJ1, SLC24A3, ITGBL1, and SLC44A1 were defined as potential prognostic genes for OS. GJA1, IGFBP6, MDFI, ITGFBI, ANXA2, and SLC24A3 were potential genes correlated with PFS. These prognostic genes were then subjected to the construction of nomogram models based on OS and PFS (Figure 7A- 7B). For OS, SLC44A1 had the greatest contribution, which could reach 100 points, while MDFI contributed most to PFS. Therefore, SLC44A1 and MDFI were considered as hub genes. To ensure the accuracy of the GESBN models, the calibration curves were drawn. The calibration curves showed good agreement between prediction and observation in the probability of 1-, 3- and 5- year survival (Figure 7C- 7D), indicating that the accuracy of the nomogram model was reliable. Discussion BC is a heterogenous tumor driven by various molecular progression pathways [14]. Analyses of BC progression showed that bone is the first metastatic site of this disease possibly due to the favorable chemokine milieu or microenvironment in the bone, as well as the intrinsic molecular characteristics of cancer cells [15, 16]. Although these hypotheses are promising, biological information and anatomical characteristics are still the basis for clinicians to determine prognosis; however, the predictors of bone metastasis remain uncertain clinically. [17, 18]. Some gene signature-based prognostic prediction models for BC patients have been reported via repurposing and analysis of microarray data [19, 20]. These models were built for predicting OS for BC patients but lack the prediction of bone metastasis. By using GEO accession number GSE124647, we obtained 1858 DEGs between BC non-bone metastasis and bone metastasis groups. After screening the prognosis-related genes, we constructed a four-gene expression signature-based nomogram model and a six-gene expression signature-based nomogram model. We firstly conducted a differential analysis of the GSE124647 dataset in relation to BC bone metastasis and employed functional enrichment analysis to these DEGs, which were found to be mainly related to signal transduction, and positive regulation of transcription in terms of BP. CCs were mainly enriched in cytoplasm, and cytosol. MFs were mainly protein binding, and Poly(A) RNA binding. The potential pathways that they were involved in were MAPK signaling pathway and proteoglycans in cancer. Based on WGCNA, 76 genes in the blue module were initially selected for the following prognostic analysis. After univariate Cox regression and Kaplan-Meier plotter analyses, OS nomogram including GJA1, SLC24A3, ITGBL1, and SLC44A1, and PFS nomogram including GJA1, IGFBP6, MDFI, TGFBI, ANXA2, and SLC24A3 were constructed. Then, its reliable prognostic ability for the OS and PFS was confirmed by Kaplan-Meier plotter and ROC analyses. AUC can be used to assess the accuracy and predictive capacity of biomarkers in diagnostic tests [21]. After confirming the predictive value of nomograms, we validated the prognostic significance of hub genes in the models. We found that high SLC44A1 expression was significantly related to favorable OS, PFS, and DMFS. However, patients in the high MDFI group predicted worse PFS. The solute carrier (SLC) superfamily contains various membrane-bound transporters which are required to transport a wide variety of substrates over biological membranes, and the dysregulated expression of these transporters may be related to cancer metastasis [22]. Validation of the prognostic value of hub genes Page 6/19 Based on the nomogram result, SLC44A1 and MDFI were the hub genes. Kaplan-Meier plotter was performed to verify the effect of SLC44A1 and MDFI on OS, PFS, and DMFS in BC. Patients in the high SLC44A1 expression group tended to have favorable OS, PFS, and DMFS (Figure 10A-10C) (P <0.05). Although MDFI expression was not significantly linked to OS and DMFS of the BC patients (Figure 10D, 10F) (P >0.05), its high expression predicted worse PFS (Figure 10E) (P <0.01). These results indicated that SLC44A1 and MDFI might be potential biomarkers for BC. 10F) (P >0.05), its high expression predicted worse PFS (Figure 10E) (P <0.01). These results indicated that SLC44A1 and MDFI might be potential biomarkers for BC. Conclusion Based on the construction of a weighted co-expression network for DEGs between BC non-bone metastasis and bone metastasis, we screened the key module and related genes to investigate a prognostic nomogram model for bone metastatic BC. The study provided some potent biomarkers of BC bone metastasis and enable the prediction of patient survival. We also found that SLC44A1 and MDFI were the hub genes in BC bone metastasis, which might be the therapeutic targets for this disease. Due to the limited clinical characteristics of patients included in the GSE124647 Due to the limited clinical characteristics of patients included in the GSE124647 GSE14020, we failed to verify the independent prognostic value of nomograms. Besides, experiments were not conducted due to various objective reasons, which would be implemented when the conditions are available in the future. Despite of these limitations, our study employed the bioinformatic approaches to assess the potential genes related to BC bone metastasis and predict the prognosis of metastatic BC patients. Funding: No funding was received for conducting this study. Author contributions: TD is mainly responsible for conception and design, DK is mainly responsible for data collection and assembly, and SW is mainly responsible for data analysis and interpretation. All authors wrote manuscript together. All authors read and approved the final manuscript. Acknowledgement: Not applicable. Ethics approval and consent to participate: Not applicable. Ethics approval and consent to participate: Not applicable. Consent for publication: Not applicable. Availability of data and materials: The dataset used and/or analyzed during the current study is available from the corresponding author on reasonable request. Competing interests: The authors have no conflicts of interest to declare. Funding: No funding was received for conducting this study. Discussion SLCO1B1 was found to be highly expressed in colon cancer, and its expression level was significantly associated with the degree of differentiation in this type of cancer [23]. SLCO1B3 overexpression may be linked to hormone-dependent growth mechanisms, and the expression of this transporter could serve as a valid prognostic factor for BC [24]. As a member of SLC superfamily, SLC44A1 is a mitochondrial protein mediating choline transport, and is preferentially expressed in neurons and oligodendrocytes [25]. Besides, high activity of the SLC44A1 promoter has been proved to participate in the occurrence of papillary glioneuronal tumors [26]. Our study revealed an Page 7/19 Page 7/19 important finding that high SLCO4A1 expression contributed to the favorable clinical outcome for BC metastasis patients. MDFI is a transcription factor that negatively regulates myogenic family proteins [27]. Previous study has demonstrated that the loss of MDFI was related to human BC and myeloid neoplasm via negative regulation of Wnt pathway [28]. In this study, high MDFI expression led to poor PFS for patients with metastatic BC. References Page 8/19 Page 8/19 1. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, et al. 2015Global cancer statistics, 2012.CA Cancer J Clin, 65:87-108. doi:10.3322/caac.21262. 2. Jemal A, Bray F, Center MM, Ferlay J, Ward E, et al. 2011Global cancer statistics.CA Cancer J Clin, 61:69-90. doi:10.3322/caac.20107. 3. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, et al. 2015Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012.Int J Cancer, 136:E359-386. doi:10.1002/ijc.29210. 4. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, et al. 2018Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.CA Cancer J Clin, 68:394-424. doi:10.3322/caac.21492. GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.CA Cancer J Clin, 68:394-424. doi:10.3322/caac.21492. 5. Coleman RE. 2006Clinical features of metastatic bone disease and risk of skeletal morbidity.Clin Cancer Res, 12:6243s-6249s. doi:10.1158/1078-0432.CCR-06-0931. 6. Chirgwin JM, Guise TA. 2000Molecular mechanisms of tumor-bone interactions in osteolytic metastases.Crit Rev Eukaryot Gene Expr, 10:159-178. 6. Chirgwin JM, Guise TA. 2000Molecular mechanisms of tumor-bone interactions in osteolytic metastases.Crit Rev Eukaryot Gene Expr, 10:159-178. 7. Gralow J. 2005Evolving role of bisphosphonates in women undergoing treatment for localized and advanced breast cancer.Clin Breast Cancer, 5 Suppl:S54-62. doi:10.3816/cbc.2005.s.005. 7. Gralow J. 2005Evolving role of bisphosphonates in women undergoing treatment for localized and advanced breast cancer.Clin Breast Cancer, 5 Suppl:S54-62. doi:10.3816/cbc.2005.s.005. 8. Xing H, Bu W, Zhang S, Zheng X, Li M, et al. 2012Multifunctional nanoprobes for upconversion fluorescence, MR and CT trimodal imaging.Biomaterials, 33:1079-1089. doi:10.1016/j.biomaterials.2011.10.039. 8. Xing H, Bu W, Zhang S, Zheng X, Li M, et al. 2012Multifunctional nanoprobes for upconversion fluorescence, MR and CT trimodal imaging.Biomaterials, 33:1079-1089. doi:10.1016/j.biomaterials.2011.10.039. 9. Albain KS, Swann RS, Rusch VW, Turrisi AT, 3rd, Shepherd FA, et al. 2009Radiotherapy plus chemotherapy with or without surgical resection for stage III non-small-cell lung cancer: a phase III randomised controlled trial.Lancet, 374:379-386. doi:10.1016/S0140-6736(09)60737-6. 9. Albain KS, Swann RS, Rusch VW, Turrisi AT, 3rd, Shepherd FA, et al. 2009Radiotherapy plus chemotherapy with or without surgical resection for stage III non-small-cell lung cancer: a phase III randomised controlled trial.Lancet, 374:379-386. doi:10.1016/S0140-6736(09)60737-6. 10. Prat A, Pineda E, Adamo B, Galvan P, Fernandez A, et al. 2015Clinical implications of the intrinsic molecular subtypes of breast cancer.Breast, 24 Suppl 2:S26-35. doi:10.1016/j.breast.2015.07.008. 10. Prat A, Pineda E, Adamo B, Galvan P, Fernandez A, et al. References 2015Clinical implications of the intrinsic molecular subtypes of breast cancer.Breast, 24 Suppl 2:S26-35. doi:10.1016/j.breast.2015.07.008. 11. Li B, Cui Y, Diehn M, Li R. 2017Development and Validation of an Individualized Immune Prognostic Signature in Early-Stage Nonsquamous Non-Small Cell Lung Cancer.JAMA Oncol, 3:1529-1537. doi:10.1001/jamaoncol.2017.1609. 11. Li B, Cui Y, Diehn M, Li R. 2017Development and Validation of an Individualized Immune Prognostic Signature in Early-Stage Nonsquamous Non-Small Cell Lung Cancer.JAMA Oncol, 3:1529-1537. doi:10.1001/jamaoncol.2017.1609. 12. Bing Z, Tian J, Zhang J, Li X, Wang X, et al. 2016An Integrative Model of miRNA and mRNA Expression Signature for Patients of Breast Invasive Carcinoma with Radiotherapy Prognosis.Cancer Biother Radiopharm, 31:253-260. doi:10.1089/cbr.2016.2059. 12. Bing Z, Tian J, Zhang J, Li X, Wang X, et al. 2016An Integrative Model of miRNA and mRNA Expression Signature for Patients of Breast Invasive Carcinoma with Radiotherapy Prognosis.Cancer Biother Radiopharm, 31:253-260. doi:10.1089/cbr.2016.2059. 13. Liu S, Song A, Wu Y, Yao S, Wang M, et al. 2021Analysis of genomics and immune infiltration patterns of epithelial-mesenchymal transition related to metastatic breast cancer to bone.Transl Oncol, 14:100993. doi:10.1016/j.tranon.2020.100993. 13. Liu S, Song A, Wu Y, Yao S, Wang M, et al. 2021Analysis of genomics and immune infiltration patterns of epithelial-mesenchymal transition related to metastatic breast cancer to bone.Transl Oncol, 14:100993. doi:10.1016/j.tranon.2020.100993. 14. Esteva FJ, Sahin AA, Cristofanilli M, Arun B, Hortobagyi GN. 2002Molecular prognostic factors for breast cancer metastasis and survival.Semin Radiat Oncol, 12:319-328. doi:10.1053/srao.2002.35251. Page 9/19 15. Kang Y, Siegel PM, Shu W, Drobnjak M, Kakonen SM, et al. 2003A multigenic program mediating breast cancer metastasis to bone.Cancer Cell, 3:537-549. doi:10.1016/s1535-6108(03)00132-6. Page 9/19 16. Jones DH, Nakashima T, Sanchez OH, Kozieradzki I, Komarova SV, et al. 2006Regulation of cancer cell migration and bone metastasis by RANKL.Nature, 440:692-696. doi:10.1038/nature04524. 17. Hess KR, Pusztai L, Buzdar AU, Hortobagyi GN. 2003Estrogen receptors and distinct patterns of breast cancer relapse.Breast Cancer Res Treat, 78:105-118. doi:10.1023/a:1022166517963. 18. Millar EK, Graham PH, O'Toole SA, McNeil CM, Browne L, et al. 2009Prediction of local recurrence, distant metastases, and death after breast-conserving therapy in early-stage invasive breast cancer using a five-biomarker panel.J Clin Oncol, 27:4701-4708. doi:10.1200/JCO.2008.21.7075. 19. Li J, Wang W, Xia P, Wan L, Zhang L, et al. 2018Identification of a five-lncRNA signature for predicting the risk of tumor recurrence in patients with breast cancer.Int J Cancer, 143:2150-2160. doi:10.1002/ijc.31573. 20. Iuliano A, Occhipinti A, Angelini C, De Feis I, Lio P. References 2018Combining Pathway Identification and Breast Cancer Survival Prediction via Screening-Network Methods.Front Genet, 9:206. doi:10.3389/fgene.2018.00206. 21. Hanley JA, McNeil BJ. 1982The meaning and use of the area under a receiver operating characteristic (ROC) curve.Radiology, 143:29-36. doi:10.1148/radiology.143.1.7063747. 22. Sutherland R, Meeson A, Lowes S. 2020Solute transporters and malignancy: establishing the role of uptake transporters in breast cancer and breast cancer metastasis.Cancer Metastasis Rev, 39:919- 932. doi:10.1007/s10555-020-09879-6. 23. Pressler H, Sissung TM, Venzon D, Price DK, Figg WD. 2011Expression of OATP family members in hormone-related cancers: potential markers of progression.PLoS One, 6:e20372. doi:10.1371/journal.pone.0020372. 24. Muto M, Onogawa T, Suzuki T, Ishida T, Rikiyama T, et al. 2007Human liver-specific organic anion transporter-2 is a potent prognostic factor for human breast carcinoma.Cancer Sci, 98:1570-1576. doi:10.1111/j.1349-7006.2007.00570.x. 25. Michel V, Bakovic M. 2012The ubiquitous choline transporter SLC44A1.Cent Nerv Syst Agents Med Chem, 12:70-81. doi:10.2174/187152412800792733. 26. Bridge JA, Liu XQ, Sumegi J, Nelson M, Reyes C, et al. 2013Identification of a novel, recurrent SLC44A1-PRKCA fusion in papillary glioneuronal tumor.Brain Pathol, 23:121-128. doi:10.1111/j.1750-3639.2012.00612.x. 27. Elabd C, Ichim TE, Miller K, Anneling A, Grinstein V, et al. 2018Comparing atmospheric and hypoxic cultured mesenchymal stem cell transcriptome: implication for stem cell therapies targeting intervertebral discs.J Transl Med, 16:222. doi:10.1186/s12967-018-1601-9. 28. Cigognini D, Corneo G, Fermo E, Zanella A, Tripputi P. 2007HIC gene, a candidate suppressor gene within a minimal region of loss at 7q31.1 in myeloid neoplasms.Leuk Res, 31:477-482. doi:10.1016/j.leukres.2006.09.007. Figures Figures Page 10/19 Figure 1 Differentially expressed genes between breast cancer non-bone metastasis and bone metastasis patients from the GSE124647 dataset. (A). The volcano plot of the 992 up-regulated (red triangle) genes and 866 down-regulated (green tringle) genes. (B) The heat map of the top 50 significant DEGs. Figure 1 Differentially expressed genes between breast cancer non-bone metastasis and bone metastasis patients from the GSE124647 dataset. (A). The volcano plot of the 992 up-regulated (red triangle) genes and 866 down-regulated (green tringle) genes. (B) The heat map of the top 50 significant DEGs. Page 11/19 igure 2 Figure 2 Page 11/19 Functional enrichment analysis of differentially expressed genes. (A) Biological process. (B) Cellular component. (C) Molecular function. (D) KEGG pathway. component. (C) Molecular function. (D) KEGG pathway. Figure 3 Weighted co-expression network analysis. (A) Dendrogram of sample clustering and heatmap of clinica traits of all breast cancer samples in a dataset of GSE124647. (B) The optimal β value result graph. (C) Module eigengene dendrogram. The horizontal axis represents a color block, and each of the different color blocks represents a different module, and the vertical axis represents the height of the dendrogram based on the expression value. (D) Clustering of module eigengenes. (E) The correlation between gene Figure 3 Figure 3 Weighted co-expression network analysis. (A) Dendrogram of sample clustering and heatmap of clinical traits of all breast cancer samples in a dataset of GSE124647. (B) The optimal β value result graph. (C) Module eigengene dendrogram. The horizontal axis represents a color block, and each of the different color blocks represents a different module, and the vertical axis represents the height of the dendrogram based on the expression value. (D) Clustering of module eigengenes. (E) The correlation between gene modules and sample characteristics. Page 12/19 igure 4 orest plot of univariate Cox regression analyses of prognosis. (A). Significant genes related to overall urvival. (B). Significant genes related to progression-free survival. Figure 4 Forest plot of univariate Cox regression analyses of prognosis. (A). Significant genes related to overall survival. (B). Significant genes related to progression-free survival. Page 13/19 Figure 5 Prognostic genes related to overall survival by Kaplan-Meier plotter. Abbreviations: L, Low; H, High; HR, Hazard ratio. Figure 5 Prognostic genes related to overall survival by Kaplan-Meier plotter. Abbreviations: L, Low; H, High; HR, Hazard ratio. Prognostic genes related to overall survival by Kaplan-Meier plotter. Abbreviations: L, Low; H, High; HR, Page 14/19 Page 14/19 Figure 6 Prognostic genes related to progression-free survival by Kaplan-Meier plotter. Abbreviations: L, Low High; HR, Hazard ratio. Figure 6 Prognostic genes related to progression-free survival by Kaplan-Meier plotter. Abbreviations: L, Low; H, High; HR, Hazard ratio. Page 15/19 Figure 7 Construction and calibration of gene expression signature-based model. (A) The four-gene-based nomogram model on overall survival. (B) The six-gene-based nomogram model on progression-free survival. (C). Calibration curve for predicting patient overall survival at 1-, 3- and 5-year. (D). Calibration curve for predicting patient progression-free survival at 1-year. Figure 7 Construction and calibration of gene expression signature-based model. (A) The four-gene-based nomogram model on overall survival. (B) The six-gene-based nomogram model on progression-free survival. (C). Calibration curve for predicting patient overall survival at 1-, 3- and 5-year. (D). Calibration curve for predicting patient progression-free survival at 1-year. Page 16/19 gure 8 ognostic evaluation of nomogram models in GSE124647. Kaplan-Meier plotter (A). overall survival rve and (B). progression-free survival curve between high-risk and low-risk bone metastatic patients. . Receiver operating characteristic (ROC) analysis of the four-gene signature. (D). ROC analysis of the -gene signature. Abbreviations: L, Low; H, High; HR, Hazard ratio. Figure 8 Figure 8 Prognostic evaluation of nomogram models in GSE124647. Kaplan-Meier plotter (A). overall survival curve and (B). progression-free survival curve between high-risk and low-risk bone metastatic patients. (C). Receiver operating characteristic (ROC) analysis of the four-gene signature. (D). ROC analysis of the six-gene signature. Abbreviations: L, Low; H, High; HR, Hazard ratio. Page 17/19 Figure 9 Expression levels of genes in nomograms. (A). GSE 124647. (B). GSE14020. Figure 9 Expression levels of genes in nomograms. (A). GSE 124647. (B). GSE14020. Expression levels of genes in nomograms. (A). GSE 124647. (B). GSE14020. Page 18/19 Figure 10 Prognostic value of the hub genes in breast cancer. The effect of SLC44A1 on (A). overall survival, (B). progression-free survival, and (C). distant metastasis-free survival in breast cancer patients. The effect of MDFI on (D). overall survival, (E). progression-free survival, and (F) distant metastasis-free survival. Abbreviations: HR, Hazard ratio. Figure 10 Prognostic value of the hub genes in breast cancer. The effect of SLC44A1 on (A). overall survival, (B). progression-free survival, and (C). distant metastasis-free survival in breast cancer patients. The effect of MDFI on (D). overall survival, (E). progression-free survival, and (F) distant metastasis-free survival. Abbreviations: HR, Hazard ratio. Page 19/19 Page 19/19 Page 19/19
https://openalex.org/W4286377743	https://zenodo.org/records/6876362/files/CURE_D4.2.pdf	English	null	CURE System Design (CURE Deliverable D4.2)	Zenodo (CERN European Organization for Nuclear Research)	2,021	cc-by	3,958	Ref. Ares(2022)172392 - 11/01/2022 Ref. Ares(2022)172392 - 11/01/2022 CURE System Design Deliverable D4.2 DATE 31 December 2021 ISSUE 2.0 GRANT AGREEMENT no 870337 DISSEMINATION LEVEL PU PROJECT WEB-SITE http://cure-copernicus.eu/ AUTHORS Mario Dohr (GeoVille) Samuel Carraro (GeoVille) Johannes Schmid (GeoVille) Michal Opletal (GISAT) AUTHORS Mario Dohr (GeoVille) Samuel Carraro (GeoVille) Johannes Schmid (GeoVille) Michal Opletal (GISAT) Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 1 of 17 LIST OF FIGURES Figure 1: CURE SYSTEM architecture ............................................................................................. 4 Figure 2: Activity diagram for the creation of a customer ............................................................ 5 Figure 3: Activity diagram for the authentication process and a service call .............................. 6 Figure 4: Component based workflow diagram for a service call ................................................ 7 Figure 5: Graph of an example DAG ............................................................................................. 10 Figure 6: WEkEO Dataset .............................................................................................................. 14 LIST OF TABLES Table 1: CURE SYSTEM components .............................................................................................. 3 LIST OF ACRONYMS App Application API Application Programming Interface CPU Central Processing Unit CCSI Copernicus Core Service Interface DB Database DIAS Data and Information Access Services DAG Directed Acyclic Graphs RAM Random-Access memory Regex Regular expression REST Representational State Transfer WP Work Package Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 1 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 1 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 1 of 17 LIST OF FIGURES Figure 1: CURE SYSTEM architecture ............................................................................................. 4 Figure 2: Activity diagram for the creation of a customer ............................................................ 5 Figure 3: Activity diagram for the authentication process and a service call .............................. 6 Figure 4: Component based workflow diagram for a service call ................................................ 7 Figure 5: Graph of an example DAG ............................................................................................. 10 Figure 6: WEkEO Dataset .............................................................................................................. 14 LIST OF TABLES LIST OF ACRONYMS App Application API Application Programming Interface CPU Central Processing Unit CCSI Copernicus Core Service Interface DB Database DIAS Data and Information Access Services DAG Directed Acyclic Graphs RAM Random-Access memory Regex Regular expression REST Representational State Transfer WP Work Package LIST OF ACRONYMS Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 2 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 2 of 17 2 SYSTEM DESIGN INFRASTRUCTURE 2 SYSTEM DESIGN INFRASTRUCTURE 1.1 Purpose of the document Deliverable 4.2 is the final delivery of Task 4.2 Preparation of CURE System Design in (WP) 4 Cure System Development. CURE Prototype System Design will describe the architectural design of the CURE Prototype in detail based on the usage of DIAS as a Service. That comprises the set-up of hardware components like processing units and storage as well as the software design including the interaction between those components. The System Design also includes the evaluation of the most suiting DIAS based on the given requirements from D4.1. Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 3 of 17 2.1 CURE SYSTEM Architecture The CURE SYSTEM is based on a loose microservice architecture. Table 1 lists the main system components (microservices) of the CURE SYSTEM infrastructure. Table 1: CURE SYSTEM components Component Technology Authentication & Authorization OAuth2 Server API Gateway Python FLASK, Gunicorn and nginx Interface description language Swagger Databases PostgreSQL Message queue RabbitMQ Job scheduling system Apache Airflow / Celery (Flower) Status manager Python module Container virtualization Docker Logging Python module Storage WEkEO S3-Bucket Monitoring & Tests Grafana, API Integration tests Figure 1 shows the CURE SYSTEM architecture and depicts how the different components are related to each other. Figure 1 shows the CURE SYSTEM architecture and depicts how the different components are related to each other. Figure 1 shows the CURE SYSTEM architecture and depicts how the different components are related to each other. Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 4 of 17 Figure 1: CURE SYSTEM architecture Figure 1: CURE SYSTEM architecture 2.2 CURE SYSTEM Service Submission Workflow • Endpoint for frontend logins: https://services.geoville.com/cure/v1/auth/login • Endpoint for registration: https://services.geoville.com/cure/v1/auth/get_bearer_token • Endpoint for access token: https://services.geoville.com/cure/v1/crm/customers/create • Required information: User information (name etc.), email and password • Response: Access token, client-id and client-secret 2.2.2 Service Submission After getting access to the infrastructure, a user can submit a new service order (e.g. app 1). • Endpoint: https://services.geoville.com/cure/v1/app1 • Required information: o Access token if the API is directly used (using the LOG-IN Endpoint in an UI will obtain the Access Token automatically) o service specific parameters (region of interest, date, etc.) • Response: Order-id Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 7 of 17 Figure 4: Component based workflow diagram for a service call Figure 4: Component based workflow diagram for a service call 2.2.1 Authentication As a first step, a user must get access to the CURE SYSTEM infrastructure. Therefore, a user can apply the authentication API endpoint. As a first step, a user must get access to the CURE SYSTEM infrastructure. Therefore, a user can apply the authentication API endpoint. • Endpoint for frontend logins: https://services.geoville.com/cure/v1/auth/login • Endpoint for registration: https://services.geoville.com/cure/v1/auth/get_bearer_token • Endpoint for access token: https://services.geoville.com/cure/v1/crm/customers/create • Endpoint for access token: https://services.geoville.com/cure/v1/crm/customers/create • Required information: User information (name etc.), email and password • Response: Access token, client-id and client-secret 2.2.2 Service Submission . Se ce Sub ss o fter getting access to the infrastructure, a user can submit a new service order (e.g. app After getting access to the infrastructure, a user can submit a new service order (e.g. app 1). 2.2 CURE SYSTEM Service Submission Workflow The aim of this section is to get a first rough overview of the CURE SYSTEM infrastructure and the workflow of a service order. In order to use a CURE SYSTEM service a user must perform the following steps: Registration: Create a user (see Figure 2) Authentication: Get access to the CURE SYSTEM infrastructure (see Figure 3) Service Submission: Submit a new service order (see Figure 3) Service Monitoring: Get the order status of the submitted service Retrieve Results: Download the result of the service order Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 5 of 17 A more detailed workflow of a typical service submission is shown in Figure 3 and Figure 4. Figure 2 shows the activity diagram for the authentication process and the submission of a Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 5 of 17 A more detailed workflow of a typical service submission is shown in Figure 3 and Figure 4. Figure 2 shows the activity diagram for the authentication process and the submission of a service. Figure 3 highlights the service call workflow from a component-related viewpoint. A more detailed workflow of a typical service submission is shown in Figure 3 and Figure 4. Figure 2 shows the activity diagram for the authentication process and the submission of a service. Figure 3 highlights the service call workflow from a component-related viewpoint. Figure 2: Activity diagram for the creation of a customer Figure 2: Activity diagram for the creation of a customer Figure 2: Activity diagram for the creation of a customer Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 6 of 17 Figure 3: Activity diagram for the authentication process and a service call Figure 3: Activity diagram for the authentication process and a service call Figure 3: Activity diagram for the authentication process and a service call Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 7 of 17 Figure 4: Component based workflow diagram for a service call 2.2.1 Authentication As a first step, a user must get access to the CURE SYSTEM infrastructure. Therefore, a user can apply the authentication API endpoint. • Endpoint: https://services.geoville.com/cure/v1/services/order_status • Required information: • Required information: • Required information: o Access token if the API is directly used (using the LOG-IN Endpoint in an UI will obtain the Access Token automatically) o Order id of the individual service order o Access token if the API is directly used (using the LOG-IN Endpoint in an UI will obtain the Access Token automatically) o Order-id of the individual service order • Response: Order status and a link to the final result file if the process is successful The endpoint supports multiple order states which are listed below: The endpoint supports multiple order states which are listed below: • FAILED: An unexpected error occurred during the execution of the service • SUCCESS: Service calculation was successful • QUEUED: Submitted request is in a waiting position (waiting list) • RECEIVED: API received the service request and created an order-id • RUNNING – Service is currently running • INVALID – Indicates missing satellite data for the requested date or tile 2.2.3 Service Monitoring After successfully submitting a new order, a user can monitor the status of the service. Therefore, the CURE SYSTEM API provides the oder_status endpoint. • Endpoint: https://services.geoville.com/cure/v1/services/order_status • Endpoint: https://services.geoville.com/cure/v1/app1 • Endpoint: https://services.geoville.com/cure/v1/app1 • Required information: • Required information: o Access token if the API is directly used (using the LOG-IN Endpoint in an UI will obtain the Access Token automatically) o Access token if the API is directly used (using the LOG-IN Endpoint in an UI wi obtain the Access Token automatically) o service specific parameters (region of interest, date, etc.) • Response: Order-id • Response: Order-id Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 8 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 8 of 17 2.2.4 Retrieve Results After receiving a successful state from the oder_status endpoint, a user can access the service result by using the download link provided by the endpoint - typically a HTTP download link. The link to the result or an error notification will be also sent to the user via email. The link to the result or an error notification will be also sent to the user via email. Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 9 of 17 3.1 Authentication & Authorization The CURE SYSTEM user authentication and user management is based on the OAuth2 standard. Therefore, the CURE SYSTEM provides an OAuth server and a database for managing users, clients, access rights and access tokens. The CURE SYSTEM API provides a set of endpoints which are required to perform common authentication and authorization operations. Among others, this includes for example: • Creating OAuth clients • Client login • Access token generation • Token validation • Resetting passwords • Setting scopes • Creating OAuth clients • Client login • Access token generation • Token validation • Resetting passwords • Setting scopes • Creating OAuth clients • Client login • Access token generation • Token validation • Resetting passwords • Setting scopes • Creating OAuth clients • Access token generation • Token validation • Resetting passwords • Setting scopes To ensure appropriate authorization standard, the CURE SYSTEM supports user scopes. These scopes are used to grant access to individual CURE SYSTEM services. 3.2 API Gateway The API Gateway is the entry point to the CURE SYSTEM infrastructure. The RESTful API provides all endpoints which are required to interact with the system. The endpoints are divided into scopes (e.g.: geo-services, custom-relation-management services, etc.) 3 CURE SYSTEM COMPONENTS This section describes the individual system components in detail and provides useful information to use and extend the CURE SYSTEM - typically a HTTP download link. 3.3 Interface Description Language The CURE SYSTEM API is documented with the Swagger software. Moreover, the Swagger UI supports a straightforward workflow to run, debug and test all CURE SYSTEM API endpoints. Please visit the Swagger UI for more details (https://services.geoville.com/cure/v1/). 3.5 Message Queue As already mentioned, the CURE SYSTEM infrastructure is based on a loose microservice architecture. The communication between these microservices is implemented using a message broker software. In particular, the CURE SYSTEM architecture uses the open-source message brokers RabbitMQ. To exchange messages between the microservices, CURE SYSTEM provides a Python module which supports the two main applications of messaging – publishing messages and receiving messages. • Publisher: Sends messages to a RabbitMQ queue • Receiver: Reads messages from a RabbitMQ queue 3.4 Databases The CURE SYSTEM infrastructure uses object-relational PostgreSQL databases. Moreover, in order to support geo-related issues, the PostGIS extension is used. PostGIS is a spatial database extension for the PostgreSQL DBMS. The extension adds support for spatial operations and geometries such as points, lines, polygons, multi-polygons and geometry collections. Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 10 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 10 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 10 of 17 Page 10 of 17 3.6.1 DAG Apache Airflow supports the creation of workflows. A workflow is formulated as Directed Acyclic Graphs (DAG). Usually, such a DAG is a collection of tasks whereas each DAG is represented as a Python script. In Figure 5 an example DAG is visually presented. Each rectangle represents a task which can be either a Python function, a Bash command or a Docker container. Therefore, before the main Docker container of a specific app runs, previous tasks can take care of obtaining the paths to the input data using GISAT’s Open Search API (CCSI), preprocessing, data formatting, etc. In the end, the result can be uploaded to a WEkEO S3-Bucket and potential interim data can be deleted to save space. Figure 5: Graph of an example DAG 3.6.2 Operator A DAG consists of several tasks which are also called operators. Airflow supports various types of operators. Common operators which are used in the current CURE SYSTEM installation are listed below: Figure 5: Graph of an example DAG Figure 5: Graph of an example DAG Figure 5: Graph of an example DAG 3.6 Job Scheduler The scheduling system is one of the most important components of the CURE SYSTEM infrastructure. It allows the execution of multiple parallel workflows and tasks. The CURE SYSTEM scheduler is based on the Apache Airflow workflow management platform. The following sub-sections address some of the most important Apace Airflow components. 3.7 Status Manager The status manager is a Python based module which listens to events that affect the status of a service order. It updates the order status in a database accordingly. For specific status updates, additional actions are triggered (e.g.: e-mail notifications). 3.6.5 Monitoring Airflow provides a powerful monitoring tool, which helps to monitor, start, delete and debug workflows and DAGs. 3.6.3 Worker One advantage of Apache Airflow is the support of distributed system architectures. Therefore, so called workers are running on worker nodes. Jobs can be transformed from the main Airflow application to worker instances by using message protocols. To put it simply, each CURE app can run on a separate virtual machine and gets managed by the airflow scheduler machine. 3.6.4 Parallelism Apache Airflow provides powerful tools and configuration options to run workflows and even DAGs in parallel. This helps to manage huge workloads. 3.6.2 Operator p A DAG consists of several tasks which are also called operators. Airflow supports various types of operators. Common operators which are used in the current CURE SYSTEM installation are listed below: p A DAG consists of several tasks which are also called operators. Airflow supports various types of operators. Common operators which are used in the current CURE SYSTEM installation are listed below: Python Operator: Executes Python callable and commands. Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 11 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 11 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 11 of 17 Bash operator: Executes commands in a Bash shell. Bash operator: Executes commands in a Bash shell. Docker operator: Executes a command inside a docker container. 3.8 Container virtualization For OS-level virtualization the CURE system uses Docker. As already mentioned, Apache Airflow provides an operator for efficiently running Docker containers. Docker images can be stored on a Docker image hub (registry.cure.geoville.com). Therefore, it is important to use correct tags. Please consider the following recommendations: • Small images are desirable • In general, each piece of code should also be checked regarding the recommendations listed below: o Use the provided logging module (Python) or the logging API o Make use of only one programming language o Follow Coding standards (e.g. PEP8 for Python) o Include sufficient error handling o Documentation with comments o Avoid the creation of temporary data if possible and do not forget the final deletion of it o Create tests cases Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 12 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 12 of 17 Copernicus for Urban Resilience in Europe o All parameters which should be able to get set by the user need to have a respective command-line argument o All parameters which should be able to get set by the user need to have a respective command-line argument 3.9 Logging 3.9 ogg g The CURE system supports two different techniques for logging: The CURE system supports two different techniques for logging: • Python based module which provides a command for logging. • API endpoint: A REST endpoint which allows logging by applying HTTP requests. Generally, both logging mechanisms send messages to a queue. A standalone program (GeoVille_MS_Logging_Saver) reads the log data from this queue and stores the messages in a database. In order to support traceability and debugging, the logging module provides different log- levels: In order to support traceability and debugging, the logging module provides different log- levels: • INFO: Confirmation that things are working as expected. • WARNING: Indicates that something unexpected happened. The software is still working as expected • ERROR: Indicates a serious problem. The software has not been able to perform some function. 3.11 Monitoring & Tests For monitoring and obtaining OS level statistics, the CURE system uses numerous Grafana dashboards. These are helpful to detect defects at an early stage. Moreover, integration tests for various stages are scheduled to continuously check the system health. 3.10 Storage The CURE system must be able to handle large amounts of geo-data. This data should be accessible (read, write) by using standard technologies (e.g.: boto3). Therefore, the CURE system uses an object storage which is easy to use and location independent (WEkEO S3 Buckets). 3.12 Copernicus core service interface Copernicus Core Service Interface is a python-based service module responsible for searching input data paths across the DIAS. CCSI provides a unified OpenSearch interface across multiple collections and provides harmonized output. CCSI is intended to provide access to the products generated by the CURE system and stored on WEkEO. Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 13 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 13 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 P 13 f 17 Page 13 of 17 3.13 Cure Portal 3.13 Cure Portal Cure portal is a web-based application that allows the user to interact with CURE products. The application will be based on the concept of storyline. Each storyline will represent a cure application or their combination to underline added value of the CURE products. The set of interactive maps, graphs, charts, tables, and other elements will be used to demonstrate the product’s values and their benefits for the user. Storylines will be developed around the demonstration products generated for frontrunners cities. Part of the Portal will be user registrations and limited access for registered users to generate their own products by using certain CURE applications. Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 14 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 14 of 17 4 DIAS SELECTION 4 DIAS SELECTION Smart decisions in a complex and strongly linked world need to be based on high value information extracted from multiple datasets by well suited data preparation steps. To derive all the insights two technical components are essential: • scalable and reliable IT infrastructure and • fast accessible, heterogeneous big datasets For this purpose, in the context of Copernicus the EU has initiated data platforms called DIAS (Data and Information Access Services). These platforms are meant to offer All-in-One Access to satellite imagery and high values Services called Copernicus Services. • fast accessible, heterogeneous big datasets For this purpose, in the context of Copernicus the EU has initiated data platforms called DIAS (Data and Information Access Services). These platforms are meant to offer All-in-One Access to satellite imagery and high values Services called Copernicus Services. After a careful platform selection process, the consortium decided to implement its work horse for processing chains on WEkEO. WEkEO is the EU's Copernicus DIAS reference service for environmental data, virtual environments for data processing and skilled user support. As highlighted in Figure 6, WEkEO offers a large range of pre-processed Sentinel satellite fleet data, data from Copernicus Services, and additional in-situ data. In total 235 open datasets are available ranging over many thematic and geographic areas. Figure 6: WEkEO Dataset Figure 6: WEkEO Dataset Besides the data availability, WEkEO offers additional computation infrastructure in terms of typical cloud virtualized engines. They are arranged along various virtual processing environments, suitable to serve the distributed processing system of CURE. CURE’s backend system is based on fully virtualized containers offering: • API based system interaction for job execution as well as data access Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 15 of 17 • central storage access components • central job execution components • central monitoring and logging • user authentication • high degree of scalability through central container orchestration Additionally, in many projects undertaken by CURE’s consortium members WEkEO has proven service maturity, excellent stability and fast support. Therefore, WEkEO is an ideal partner for a project such as CURE. Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 15 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 15 of 17 • central storage access components • central job execution components • central monitoring and logging • user authentication • high degree of scalability through central container orchestration Additionally, in many projects undertaken by CURE’s consortium members WEkEO has proven service maturity, excellent stability and fast support. Therefore, WEkEO is an ideal partner for a project such as CURE. • high degree of scalability through central container orchestration Additionally, in many projects undertaken by CURE’s consortium members WEkEO has proven service maturity, excellent stability and fast support. Therefore, WEkEO is an ideal partner for a project such as CURE. Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 16 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 16 of 17 Copernicus for Urban Resilience in Europe CURE System Design Deliverable D4.2 Page 16 of 17 5 CONCLUSION This document presents the current version of the CURE System Design and its components and functionalities. This first version of the System will be further developed, improved, and adjusted in accordance with initial specification and planned project schedule and according to specific needs of CURE applications. As a next step the development of the CURE prototype, based on User-Requirements and System design, will be done, following an agile software development approach by iteratively build a system with step-by-step adding of functionalities (components and services). Necessary changes will be will in close iteration with the work conducted in the frame of WP3 and WP2.
https://openalex.org/W3161373196	https://www.frontiersin.org/articles/10.3389/fphys.2021.651105/pdf	English	null	Thioredoxin-1 Is a Target to Attenuate Alzheimer-Like Pathology in Diabetic Encephalopathy by Alleviating Endoplasmic Reticulum Stress and Oxidative Stress	Frontiers in physiology	2,021	cc-by	11,083	Thioredoxin-1 Is a Target to Attenuate Alzheimer-Like Pathology in Diabetic Encephalopathy by Alleviating Endoplasmic Reticulum Stress and Oxidative Stress Yu Guo†, Chenghong Zhang†, Chunyang Wang, Yufei Huang, Jingyun Liu, Haiying Chu, Xiang Ren, Li Kong* and Haiying Ma* Citation: Guo Y, Zhang C, Wang C, Huang Y, Liu J, Chu H, Ren X, Kong L and Ma H (2021) Thioredoxin-1 Is a Target to Attenuate Alzheimer-Like Pathology in Diabetic Encephalopathy by Alleviating Endoplasmic Reticulum Stress and Oxidative Stress. Front. Physiol. 12:651105. doi: 10.3389/fphys.2021.651105 Keywords: diabetic encephalopathy, Alzheimer’s disease, thioredoxin-1, endoplasmic reticulum stress, oxidative stress Edited by: Mohsin Khan, Temple University, United States Edited by: Mohsin Khan, Temple University, United States Reviewed by: Dhanendra Tomar, Temple University, United States Sadia Mohsin, Lewis Katz School of Medicine at Temple University, United States Correspondence: Haiying Ma mahaiying@dmu.edu.cn Li Kong kongli@dmu.edu.cn †These authors have contributed equally to this work Mohsin Khan, Temple University, United States Reviewed by: Dhanendra Tomar, Temple University, United States Sadia Mohsin, Lewis Katz School of Medicine at Temple University, United States Correspondence: Haiying Ma mahaiying@dmu.edu.cn Li Kong kongli@dmu.edu.cn †These authors have contributed equally to this work Reviewed by: Dhanendra Tomar, Temple University, United States Sadia Mohsin, Lewis Katz School of Medicine at Temple University, United States Correspondence: Haiying Ma mahaiying@dmu.edu.cn Li Kong kongli@dmu.edu.cn †These authors have contributed equally to this work Correspondence: Haiying Ma mahaiying@dmu.edu.cn Li Kong kongli@dmu.edu.cn Correspondence: Haiying Ma mahaiying@dmu.edu.cn Li Kong kongli@dmu.edu.cn †These authors have contributed equally to this work Specialty section: This article was submitted to Redox Physiology, a section of the journal Frontiers in Physiology Received: 08 January 2021 Accepted: 20 April 2021 Published: 17 May 2021 Edited by: Mohsin Khan, Temple University, United States Edited by: Mohsin Khan, Temple University, United States Varying degrees of central nervous system neuropathy induced by diabetes mellitus (DM) contribute to a cognitive disorder known as diabetic encephalopathy (DE), which is also one of the independent risk factors for Alzheimer’s disease (AD). Endoplasmic reticulum stress (ERS) plays a critical role in the occurrence and development of DE and AD. However, its molecular mechanism remains largely unknown. This study aims to investigate whether thioredoxin-1 (Trx-1) could alleviate DE and AD through ERS, oxidative stress (OS) and apoptosis signaling pathways. Mice were randomly divided into a wild-type group (WT-NC), a streptozotocin (STZ)-treated DM group (WT-DM), a Trx-1-TG group (TG-NC) and a Trx-1-TG DM group (TG-DM). Diabetic animals showed an increase in the time spent in the target quadrant and the number of platform crossings as well as AD-like behavior in the water maze experiment. The immunocontent of the AD-related protein Tau and the levels of cell apoptosis, β-amyloid (Aβ) plaque formation and neuronal degeneration in the hippocampus of the diabetic group were increased. Some key factors associated with ERS, such as protein disulﬁde isomerase (PDI), glucose-regulated protein 78 (GRP78), inositol-requiring enzyme 1α (IRE1α), tumor necrosis factor receptor-associated factor 2 (TRAF2), apoptosis signal-regulating kinase-1 (ASK1), c-Jun N-terminal kinase (JNK), protein kinase RNA (PKR)-like ER kinase (PERK), and C/EBP homologous protein (CHOP), were upregulated, and other factors related to anti-oxidant stress, such as nuclear factor erythroid 2-related factor (Nrf2), were downregulated in the DM group. Moreover, DM caused an increase in the immunocontents of caspase-3 and caspase-12. However, these changes were reversed in the Trx-1-tg DM group. Therefore, we conclude that Trx-1 might be a key factor in alleviating DE and AD by regulating ERS and oxidative stress response, thus preventing apoptosis. Department of Histology and Embryology, College of Basic Medical Sciences, Dalian Medical University, Dalian, China ORIGINAL RESEARCH published: 17 May 2021 doi: 10.3389/fphys.2021.651105 ORIGINAL RESEARCH published: 17 May 2021 doi: 10.3389/fphys.2021.651105 INTRODUCTION most of the UPR signaling pathways are activated, including the following: (1) after inositol-requiring enzyme 1α (IRE1α) is activated, tumor necrosis factor receptor-associated factor 2 (TRAF2) forms a complex with apoptosis signal-regulating kinase-1 (ASK1), which activates c-Jun N-terminal kinase (JNK) and subsequently caspase-12 on the ER membrane to induce apoptosis; (2) activating transcription factor 6 (ATF6) is released from the ER membrane, which leads to its nuclear translocation and upregulation of ER stress response genes; (3) glucose- regulated protein 78 (GRP78) activation induces the protein kinase RNA (PKR)-like ER kinase (PERK) eukaryotic initiation factor 2α (Eif2α) pathway to activate the expression of C/EBP homologous protein (CHOP) to initiate apoptosis (Zahraa et al., 2015; Almanza et al., 2018); additionally, the activation of PERK signaling can induce a conformational change in nuclear factor erythroid 2-related factor (Nrf2) that triggers the dissociation of the Kelch-like protein 1 (keap1)-Nrf2 complex and regulates cell oxidation (Zhu et al., 2015). In addition, it has been reported that Nrf2 is involved in increasing endogenous antioxidant levels and inhibiting apoptosis (Lim et al., 2014; Glory and Averill-Bates, 2016). It can be seen that there is a clear interaction between OS and ERS. OS exacerbates ERS, while ERS can also cause and exacerbate OS (Thummayot et al., 2016; Yang et al., 2016). Many studies have shown that both type 1 and type 2 diabetes can cause varying degrees of central nervous system dysfunction and cognitive decline, which is called diabetic encephalopathy (DE) (Stranahan, 2015). The molecular mechanism of the cognitive dysfunction in DE has been widely studied, but it is still not clear. A large number of studies have shown that endoplasmic reticulum stress (ERS) and oxidative stress (OS) are involved in the pathogenesis of DE (Biessels et al., 2002; Díaz- Gerevini et al., 2014; Wang et al., 2016). Lupachyk et al. (2013) found CHOP deﬁcient mice displayed improved sciatic nerve oxidative-nitrative stress and attenuated peripheral neuropathy as compared with their wild-type (WT) littermates in the setting of diabetes, suggesting dysregulated UPR induced by prolonged ER stress is implicated in the development of diabetic peripheral neuropathy. While another research shown that the up-regulation of CHOP in hippocampal neurons of diabetic mice may promote neuronal apoptosis and account for the damaged learning and memory ability of diabetic mice (Zhao et al., 2015). Dabidi et al. INTRODUCTION found that OS aﬀects brain tissue with the progress of the disease course, the damage of brain tissue gradually accumulation, degeneration and even necrosis of nerve cells, and then central nervous system DE occurs as a result of chronic, irreversible damage to the system (Roshan et al., 2013). y g Thioredoxin-1 (Trx-1) is a small 12 kDa multifunctional protein having a redox-active disulﬁde/dithiol within its active site sequence, -Cys-Gly-Pro-Cys-, and operates together with NADPH and thioredoxin reductase (Holmgren, 1985). In the regulatory region of the Trx-1 gene, there are several promoter- speciﬁc transcription factor 1 binding motifs, an antioxidant- responsive element, and a putative cAMP-responsive element (CRE) (Bai et al., 2003). Trx-1 as a disulﬁde-reducing system low molecular weight protein with redox properties, and it plays an important role in regulating redox reactions in the human body (Matés and Sánchez-Jiménez, 2000; Bai et al., 2003). Trx- 1, a major isoform of Trx that is ubiquitously expressed in many cell types and is found in various subcellular localizations (Nagarajan et al., 2016), has been reported to be a regulator of ERS (Bai et al., 2007). Some studies have also shown that Trx- 1 can protect nerves by inhibiting the stress response of the ER (Luo et al., 2012; Zeng et al., 2014). During the pathogenic process of DM, long-term high glucose stimulation promotes the occurrence and development of ERS and OS, inhibits the activity of Trx-1, causes the dissociation of the Trx-1-ASK1 complex, activates the ASK1-mediated apoptotic pathway, and causes cell apoptosis (Zhu et al., 2018). As shown in several studies that: a decrease in neuronal Trx-1 in AD brains, Aβ neurotoxicity might be mediated by oxidation of Trx-1 and subsequent activation of the ASK1 cascade (Akterin et al., 2006). Deregulation of Trx-1 antioxidant systems could be important events in AD pathogenesis. Moreover, Trx-1 is also considered to be one of the antioxidant enzymes regulated by Nrf2 (Nakamura, 2004). Therefore, we reasoned that Trx-1/Nrf2 could play a key protective role in DE. g y A large number of epidemiological studies have shown that diabetes signiﬁcantly increases the risk of acquiring Alzheimer’s disease (AD); pathological changes similar to AD, such as Aβ deposits, neuroﬁbrillary tangles, and damaged or lost neuronal synapses, are also found in patients with DE (Selkoe, 2001; Zargar et al., 2014). Citation: Citation: Guo Y, Zhang C, Wang C, Huang Y, Liu J, Chu H, Ren X, Kong L and Ma H (2021) Thioredoxin-1 Is a Target to Attenuate Alzheimer-Like Pathology in Diabetic Encephalopathy by Alleviating Endoplasmic Reticulum Stress and Oxidative Stress. Front. Physiol. 12:651105. doi: 10.3389/fphys.2021.651105 May 2021 \| Volume 12 \| Article 651105 1 Frontiers in Physiology \| www.frontiersin.org Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. Frontiers in Physiology \| www.frontiersin.org INTRODUCTION Other studies have shown that diabetes can promote the abnormal modiﬁcation of Tau protein and that ERS and OS are also involved in the pathogenesis of AD: an upregulation of PDI during ER stress associated with multiple neurodegenerative diseases including AD represents an adaptive response to protect neuronal cells, their results here indicate that human PDI, a key component of the protein quality control system in the endoplasmic reticulum (ER) (Han et al., 2009; Xu et al., 2013), can directly interact with human Tau on the ER, and could counteract the toxicity caused by abnormal Tau aggregation in the ER by binding to Tau protein and inhibiting Tau ﬁbrillization (Clodfelder-Miller et al., 2006; Asano et al., 2007; Aubert et al., 2015). In addition, most studies indicate that Tau is involved in the neurodegeneration associated with oxidative stress in AD: in a Drosophila model of human tauopathy (Tau R406W), a reduction in the gene dosage of thioredoxin reductase or mitochondrial superoxide dismutase 2 (SOD2) promotes Tau-induced neurodegenerative histological abnormalities and neuronal apoptosis. On the contrary, overexpression of these antioxidant enzymes or treatment with vitamin E decreases the Tau-induced neuronal death (Stamer et al., 2002). These ﬁndings suggest that DE and AD may have the same mechanisms. High glucose levels in the body disrupt the internal balance of the ER, resulting in the accumulation of unfolded or misfolded proteins, which leads to ERS (Jin et al., 2016; Wu et al., 2016), and the main response to ERS is the activation of the unfolded protein response (UPR) (Dan et al., 2017). At this point, In this study, we veriﬁed the AD-like pathological changes and cognitive dysfunction of DE, including AD-related protein expression, neuronal apoptosis and neuronal degeneration, in the mouse hippocampus. More importantly, the eﬀects of Trx-1 May 2021 \| Volume 12 \| Article 651105 2 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. upregulation on the abovementioned factors and ERS, OS were studied. Our results suggest that Trx-1 can alleviate DE and AD by regulating the ERS and OS induced by DM. 10 s. The probe test was performed 24 h after the last acquisition trial. In the probe test, the platform was removed from the tank, and the mice were allowed to swim freely for 60 s. Then, the visual platform experiment was immediately performed to test the swimming speed of the mice. Animal Experiments Male C57/B6 wild-type (WT) and C57/B6 human-Trx-1 transgenic (TG) mice approximately 6∼8 weeks of age were used in the experiments. Mice were obtained from the SPF Animal Center of Dalian Medical University, housed in plastic cages and maintained on a 12 h light/dark cycle with free access to food and water. Mice were randomly selected from the wild-type and transgenic groups for the diabetic model groups. The mice were randomly divided into a wild-type group (WT-NC), a streptozotocin (STZ)-induced DM group (WT-DM), a Trx-1-tg group (TG-NC) and a Trx-1-tg DM group (TG-DM) (n = 6). The diabetic group was fasted for 12 h and then intraperitoneally injected with streptozotocin (STZ) solution (50 mg/kg) for ﬁve consecutive days. Then, they were fed a normal diet; after 72 h, their blood glucose was measured on three consecutive days, and fasting was carried out 8 h before each measurement. When the blood glucose was greater than 11.1 mmol/L, the diabetic model was considered to be successfully established. The normal control group was injected intraperitoneally with the same amount of citric acid-sodium citrate buﬀer, and all experimental materials were obtained at 12 weeks after the completion of the model. All procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (NIH), and all protocols were approved by the Institutional Animal Care and Use Committee of Dalian Medical University. g y The TdT-Mediated dUTP Nick-End Labeling (TUNEL) assay was performed on mice hippocampal tissues following the manufacturer’s instructions (TransGen Biotech, China). First, deparaﬃnized tissue sections were washed with phosphate- buﬀered saline (PBS) for 5 min. Then, 100 µl of cell permeation solution (0.1% Triton X-100) was added dropwise to the sample area to be tested, the tissue was incubated at room temperature for 30 min, and the excess liquid was removed. The tissues were incubated with 50 µl of a well-mixed labeling solution and 2 µl of terminal deoxynucleotidyl transferase (TDT) at 37◦C for 60 min in the dark to allow the tailing reaction to occur and then washed with PBS three times for 5 min each. Cell permeation buﬀer was added to the sample area to be tested three times for 5 min each. Finally, the TUNEL-stained slides were observed immediately upon completion of the assay. Fluorescent cells were quantiﬁed by ImageJ software. Immunohistochemistry The sections were ﬁrst dewaxed and hydrated, and then the sections were subjected to citrate antigen repair. An appropriate amount of blocking endogenous peroxidase was added dropwise to each section, and the sections were incubated for 15 min at room temperature and then washed with PBS 3 times for 3 min each; the sections were blocked with goat serum solution for 15 min at room temperature and subsequently incubated overnight at 4◦C with a rabbit Aβ (Aβ1−40, Aβ1−42, 1:200, novusbio) antibody. The cells were washed with PBS three times for 3 min each and then treated with an appropriate amount of biotin-labeled sheep anti-rabbit/mouse IgG, incubated for 15 min at room temperature, and washed with PBS 3 times for 3 min each. An appropriate amount of streptozotocin-peroxidase was added, incubated for 15 min at room temperature, and washed with PBS three times for 3 min each. Diaminobenzidine (DAB) solution was applied to the sections for 10 s–5 min, and the sections were washed three times with PBS. Hematoxylin was used to stain the cell nuclei. Five random slides were selected from each group, and ﬁve randomly selected visual ﬁelds in the hippocampal region from each slide were observed. The mean optical density was quantiﬁed by ImageJ software. Frontiers in Physiology \| www.frontiersin.org Morris Water Maze The Morris water maze (MWM) test was used to evaluate spatial learning and memory. The MWM consists of a circular pool with a diameter of 120 cm and a depth of 45 cm. The tank was divided into four equal quadrants with the platform (10 cm × 10 cm) located in the center of the northeast quadrant. The platform was hidden 1 cm below the water surface, and a non-toxic white dye was used to make the water opaque. The testing room was illuminated with a constant intensity light source and kept quiet during the experiments, and diﬀerent ﬁgures and objects were hung on the walls (⃝, □, △and ×). All swimming trials were recorded with a video camera suspended from the ceiling and analyzed using EthoVision XT 9.0 software (Noldus, Netherlands). Real-Time Quantitative PCR Total RNA was extracted using TRIzol reagent (Takara, China) and transcribed using a Reverse Transcription Kit (Applied Biosystems, United States) according to the manufacturer’s instructions. mRNA samples were mixed with primers and 2× TransStart Top Green qPCR SuperMix (Transgene) in a total volume of 20 µl. The thermal cycling conditions used in the protocol were 30 s at 94◦C followed by 45 cycles at 94◦C for 5 s, 60◦C for 30 s, and the dissociation stage. Gene expression levels were analyzed relative to the level of the GAPDH gene transcript. The primers used were as follow (see Table 1). TABLE 2 \| Blood glucose test results for each group of mice (mmol/L). TABLE 2 \| Blood glucose test results for each group of mice (mmol/L). WT-NC WT-DM TG-NC TG-DM 1 day 6.83 ± 1.28 6.77 ± 0.87 6.23 ± 1.76 6.80 ± 0.32 4 days 6.50 ± 0.56 21.56 ± 2.35* 6.38 ± 1.43 15.93 ± 1.12# 4 weeks 7.38 ± 0.38 26.7 ± 2.31 7.85 ± 0.61 21.44 ± 0.12## 8 weeks 7.62 ± 1.19 25.68 ± 0.65 7.95 ± 0.56 17.17 ± 0.68## 12 weeks 7.75 ± 1.56 25.89 ± 3.19 7.62 ± 2.36 20.14 ± 1.22## p < 0.01, comparison between the WT-NC group and the WT-DM group; #p < 0.05, ##p < 0.01, comparison between the TG-NC group and the TG- DM group. Western Blot Acquisition tests were performed with the mice for ﬁve consecutive days. Each mouse was placed into the water in all four quadrants on each test day. During the experiment, the time for the mice to ﬁnd the platform in the water was recorded. This time is the escape latency period. The longest swimming time for the mice that could not ﬁnd the platform was 60 s. If the platform was not found within 60 s, the experimenter guided the mouse to the platform and allowed it to stay on the platform for The samples were thawed, washed in ice-cold PBS and sonicated in KeyGen lysis assay buﬀer (KeyGen Biotech, China). The samples were then sonicated, incubated on ice for 30 min and centrifuged at 10,000 × g for 20 min at 4◦C. The protein concentration in the supernatant was determined by a Pierce BCA Protein Assay Kit (Life Technologies, United States). Equal quantities of protein were separated May 2021 \| Volume 12 \| Article 651105 3 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. by 10% [for the ASK-1 (1:1,000, Cell Signaling Technology, United States), p-ASK-1, IRE1α (Wanleibio, China), p-IRE1α, TRAF2 (1:1,000, Wanleibio), JNK (1:1,000, Wanleibio), p-JNK (1:1,000, Wanleibio), GRP78 (1:1,000, Wanleibio), CHOP (1:1,000, Wanleibio), Nrf2 (1:1,000, Wanleibio), p-PERK, PERK (1:1,000, Wanleibio), Tau (1:1,000, Cell Signaling Technology), and p-Tau (1:1,000, Cell Signaling Technology) antibodies] or 12% SDS–PAGE (for the cleaved caspase-12 (1:1,000, Wanleibio), cleaved caspase-3 (1:1,000, Wanleibio) antibodies) and transferred to a polyvinylidene diﬂuoride membrane (Millipore Corp., United States). The membrane was soaked in 5% skim milk (in PBS, pH 7.2) or 5% bovine serum albumin (BSA) (in PBS, pH 7.2) overnight at 4◦C and then incubated with primary antibodies (1:1000) followed by peroxidase-conjugated anti-mouse or anti-rabbit IgG antibody (1:10,000, KPL, Gaithersburg, United States). The epitope was visualized by an ECL Western blot detection kit (Millipore, United States) and imaged with a ChemiDocTM XRS and Image LabTM Software (Bio-Rad Laboratories, Inc. Hercules, United States). K2HPO4 + KH2PO4 + 0.1 mM ethylenediaminetetraacetic acid (EDTA) (pH, 7.0) + 0.1% BSA and then centrifuged. The obtained brain supernatants were used for further investigations. The absorbance of each well was determined by a microplate reader, and the activity was calculated according to a formula. The SOD, MDA, GSH-PX and CAT activities were measured. The activities of the studied antioxidant enzymes in tissue extracts were calculated in U/mg of protein. Statistical Analysis All values are expressed as the mean ± standard deviation (SD). The statistical analyses were completed with one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test by GraphPad Prism 5.0 software (GraphPad Software, Inc., La Jolla, CA, United States). Diﬀerences were considered signiﬁcant at p < 0.05. Blood Glucose Changes in Each Group of Mice Slides containing sections of brain tissue were dewaxed in xylene two times, immersed in 100% ethanol two times for 5 min each, immersed in 70% alcohol for 2 min, and then rinsed with fresh ddH2O two times for 1 min each. Nine parts ddH2O was mixed with 1 part Solution B (potassium permanganate); the slides were added to this mixture and incubated for 10 min. Then, eight parts ddH2O was mixed with one part Solution C (Fluoro- Jade) (Merck Millipore, United States) and 1 part Solution D (DAPI) and placed in a Coplin jar in the dark or low light; the slides were incubated in this solution for 10 min. The slides were then rinsed in ddH2O three times for 1 min each and dried on a slide warmer at 50–60◦C for at least 5 min. The dry slides were then cleared by brief immersion in xylene. Finally, the FJC slides were observed immediately upon completion of the assay, and the ﬂuorescence intensity was quantiﬁed by ImageJ software. There were no signiﬁcant diﬀerences in mobility, body weight, or blood glucose levels between the mice in each group prior to STZ modeling. Compared with the WT-NC group, blood glucose in the WT-DM group was signiﬁcantly higher (p < 0.01), and compared with the TG-NC group, blood glucose in the TG- DM group was signiﬁcantly higher (p < 0.05) after 3 days of STZ injection, and the blood glucose was persistently high (see Table 2). TABLE 1 \| PCR primer sequences. Gene Primer sequences Nrf2 F:5′-GCCTTACTCTCCCAGTGAATAC-3′ R:5′-CTCCCAAATGGTGCCTAAGA-3′ Trx-1 F: 5′-GGAATGGTGAAGCAGATCGAG-3′ R:5′-ACGCTTAGACTAATTCATTAAT-3′ GAPDH F: 5′-GAGCCCTTCCACAATGCCAAAGTT-3′ R:5′-TGTGATGGGTGTGAACCACGAGAA-3′ F, forward primer; R, reverse primer. High Expression of Trx-1 Gene in Transgenic Mice Improves Spatial Learning and Memory p < 0.01, and p < 0.001, respectively). In addition, the TG- NC and TG-DM groups also showed diﬀerences on the next 3 days (p < 0.01, p < 0.01 and p < 0.05, respectively) (see Figure 1B). In the ascending platform test, which was carried out 24 h later, we found that the platform crossings and the time spent in the target quadrant in the WT-DM group were signiﬁcantly reduced compared with those of the WT-NC group (p < 0.001, p < 0.0001, respectively) and the TG-DM group (p < 0.05, p < 0.01, respectively) (see Figures 1C,D). Finally, in the swimming speed and visual platform tests, there were no signiﬁcant diﬀerences among the groups (see Figures 1E,F). In the present experiment, we aimed to enhance the anti- ERS and antioxidative stress eﬀects of Trx-1 by transgenically increasing the expression of the Trx-1 gene in mice. Real-time PCR was used to detect Trx-1. The results showed that the Trx-1 gene was upregulated in both the TG-NC group and TG- DM group compared with the WT-NC and WT-DM groups (p < 0.0001 and p < 0.001, respectively), and upregulated in TG-NC group compared with the TG-DM group (p < 0.05) (see Figure 1A). Biochemical Analysis Determination of hippocampal tissue catalase (CAT), superoxide dismutase (SOD), malondiadehyde (MDA) and glutathione peroxidase (GSH-PX) activities. According to the manufacturer’s instructions, the collected organs were homogenized in 50 mM phosphate buﬀer cooled to 4◦C containing May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 4 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. Trx-1 Alleviates the Hippocampal Neurodegeneration and Hippocampal Apoptosis Induced by DM To evaluate whether the DM mice exhibited cognitive impairment and whether Trx-1 overexpression positively aﬀected this impairment, we used the Morris water maze test, a well- established test for spatial learning and memory, to assess spatial learning and memory in each group of mice. We found that there was a downward trend in escape latency on the second day in all groups, and there were no signiﬁcant diﬀerences among the groups. From the third to the ﬁfth day, except for the WT-DM group, the escape latency of the other three groups still showed a downward trend; however, the WT-DM group had a smaller trend than the other two groups. There were signiﬁcant diﬀerences in the escape latency between the WT- DM group and the WT-NC group (p < 0.01, p < 0.01, and p < 0.01, respectively), as well as the TG-DM group (p < 0.01, We evaluated the apoptosis of neurons by the TUNEL assay to verify the damage to neurons in the hippocampus induced by DM. We found that the number of TUNEL-positive cells with red ﬂuorescence signals was much higher in the STZ- treated transgenic and non-transgenic DM mice than in the control DM mice without STZ treatment (see Figure 2A). Quantitative analysis showed that the number of TUNEL- positive cells was signiﬁcantly increased in the WT-DM group compared with the WT-NC group (p < 0.001), but in the TG-DM group, the number of positive cells was signiﬁcantly decreased compared with the WT-DM group (p < 0.01) (see Figure 2B). FIGURE 1 \| Trx-1 mRNA levels in the hippocampal tissues of the mice and neurobehavioral analysis of Trx-1 transgenic mice. (A) Real-time quantitative PCR of Trx-1 mRNA levels in mouse hippocampal tissue. (B) Latency to platform, ##p < 0.01: WT-NC vs WT-DM; &&p < 0.01: WT-DM vs TG-DM; p < 0.05, p < 0.01: TG-NC vs TG-DM. (C) Platform crossing (probe trial), (D) Time spent in the target quadrant (probe trial). (E) Latency to platform (visual trial). (F) Swimming speed. Tukey’s post hoc test for one-way ANOVA was used to determine Trx-1 mRNA levels. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, p < 0.001, and **p < 0.0001. FIGURE 1 \| Trx-1 mRNA levels in the hippocampal tissues of the mice and neurobehavioral analysis of Trx-1 transgenic mice. Trx-1 Alleviates the Hippocampal Neurodegeneration and Hippocampal Apoptosis Induced by DM (A) Real-time quantitative PCR of Trx-1 mRNA levels in mouse hippocampal tissue. (B) Latency to platform, ##p < 0.01: WT-NC vs WT-DM; &&p < 0.01: WT-DM vs TG-DM; p < 0.05, *p < 0.01: TG-NC vs TG-DM. (C) Platform crossing (probe trial), (D) Time spent in the target quadrant (probe trial). (E) Latency to platform (visual trial). (F) Swimming speed. Tukey’s post hoc test for one-way ANOVA was used to determine Trx-1 mRNA levels. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, p < 0.001, and **p < 0.0001. May 2021 \| Volume 12 \| Article 651105 5 Frontiers in Physiology \| www.frontiersin.org Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. Guo et al. FIGURE 2 \| Trx-1 signiﬁcantly inhibits DM-induced apoptosis and neural degeneration in the hippocampal region. (A) Positive cells with a red ﬂuorescence signal (white arrows) were present in the hippocampal as determined by the TUNEL assay. (B) Quantitative statistical analysis of TUNEL-positive cells. (C) Representative images of FJC-positive staining (white arrows) in the hippocampal regions. (D) Quantitative statistical analysis of FJC-positive cells. Tukey’s post hoc test for one-way ANOVA used for TUNEL test. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, p < 0.001, **p < 0.0001. FIGURE 2 \| Trx-1 signiﬁcantly inhibits DM-induced apoptosis and neural degeneration in the hippocampal region. (A) Positive cells with a red ﬂuorescence signal (white arrows) were present in the hippocampal as determined by the TUNEL assay. (B) Quantitative statistical analysis of TUNEL-positive cells. (C) Representative images of FJC-positive staining (white arrows) in the hippocampal regions. (D) Quantitative statistical analysis of FJC-positive cells. Tukey’s post hoc test for one-way ANOVA used for TUNEL test. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, p < 0.001, *p < 0.0001. not in the other two groups (see Figure 3A), and the Aβ plaque density in the TG-DM group was signiﬁcantly lower than that in the WT-DM group (p < 0.001) (see Figure 3B). Western blot analysis was performed to detect the immunocontent of Tau and p-Tau. The results showed that the p-Tau immunocontent levels were signiﬁcantly increased in the WT-DM group compared with the WT-NC group (p < 0.001). Trx-1 Alleviates the Hippocampal Neurodegeneration and Hippocampal Apoptosis Induced by DM However, the p-Tau immunocontent was decreased in the TG-DM group compared with the WT-DM group (p < 0.01) (see Figure 3C). Fluoro-JadeC (FJC) staining was used to label damaged cells to further determine the severity of neuronal degeneration. The number of FJC-positive cells in the hippocampal regions of the WT-DM group was signiﬁcantly increased compared with that of the WT-NC group (p < 0.0001), and the number of FJC-positive cells in the hippocampal CA1 and CA3 regions of the TG-DM group was signiﬁcantly decreased compared with that of the WT- DM group (p < 0.01) (see Figures 2C,D). Trx-1 Affects DM-Induced ERS Marker Proteins and Inhibits DM-Induced Activation of PERK Pathway-Associated Proteins To examine AD-like pathological changes in DE and the eﬀect of Trx-1 on the pathological changes, the formation of Aβ plaques in the brain and the load of amyloid plaques in mice were quantiﬁed by immunohistochemistry (IHC) with antibodies against Aβ. Through observation and calculation of the density of Aβ-positive plaques in the hippocampus, we found that Aβ plaques were present in the WT-DM and TG-DM groups but To investigate the mechanism by which the ERS PERK receptor- related pathway is involved in the eﬀects of Trx-1, Western blot analysis was performed to detect the immunocontent of PERK, p-PERK, CHOP and caspase-3. The results showed that the p-PERK, CHOP and caspase-3 immunocontent levels were May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 6 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. Guo et al. FIGURE 3 \| Trx-1 reduced Aβ burden and Tau phosphorylation in DM mice. (A) Representative images of Aβ deposits in the hippocampus. (B) Quantitative statistical analysis of Aβ density value in the hippocampus. (C) Representative Western blot images and analysis of Tau and p-Tau. Tukey’s post hoc test for one-way ANOVA used for Aβ density value and Western blot immunocontent. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.01, *p < 0.001. FIGURE 3 \| Trx-1 reduced Aβ burden and Tau phosphorylation in DM mice. (A) Representative images of Aβ deposits in the hippocampus. (B) Quantitative statistical analysis of Aβ density value in the hippocampus. (C) Representative Western blot images and analysis of Tau and p-Tau. Tukey’s post hoc test for one-way ANOVA used for Aβ density value and Western blot immunocontent. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.01, **p < 0 001 signiﬁcantly increased in the WT-DM group compared with the WT-NC group (p < 0.001, p < 0.05, and p < 0.05, respectively) (see Figures 4A–D). However, the p-PERK, CHOP and caspase- 3 immunocontent levels were signiﬁcantly decreased in the TG-DM group compared with the WT-DM group. (p < 0.01, p < 0.01, p < 0.01, respectively) (see Figures 4A–D). ER homeostasis (Jui-Ching et al., 2008; Li-Rong et al., 2013). We further examined the eﬀect of Trx-1 on DM-induced ERS. The results showed that Trx-1 inhibited the DM-induced decrease in GRP78 levels (see Figures 4E,F). Trx-1 Affects DM-Induced ERS Marker Proteins and Inhibits DM-Induced Activation of PERK Pathway-Associated Proteins Moreover, the DM- induced activation of PDI expression was inhibited by Trx-1 (see Figures 4E,G). These data suggest that Trx-1 attenuated DM-induced ERS by enhancing the expression of GRP78 and inhibiting the expression of PDI. p p p y g GRP78 and PDI, both ER-resident chaperones, are markers of ERS that play crucial roles in the dynamic regulation of May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 7 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. FIGURE 4 \| Effects of DM and Trx-1 on the protein immunocontent of GRP78 and PDI and changes in the activation of PERK, CHOP and caspase-3 in the PERK receptor-related signaling pathway in DM mice. (A) Representative protein expression bands of PERK, p-PERK, CHOP and caspase-3. (B) Representative Western blot analysis of p-PERK/PERK. (C) Representative Western blot analysis of caspase-3. (D) Representative Western blot analysis of CHOP. (E) Representative protein expression bands of GRP78 and PDI. (F) Representative Western blot analysis of GRP78. (G) Representative Western blot analysis of PDI. Tukey’s post hoc test for one-way ANOVA was used to analyze the Western blot immunocontent. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, and p < 0.001. FIGURE 4 \| Effects of DM and Trx-1 on the protein immunocontent of GRP78 and PDI and changes in the activation of PERK, CHOP and caspase-3 in the PERK receptor-related signaling pathway in DM mice. (A) Representative protein expression bands of PERK, p-PERK, CHOP and caspase-3. (B) Representative Western blot analysis of p-PERK/PERK. (C) Representative Western blot analysis of caspase-3. (D) Representative Western blot analysis of CHOP. (E) Representative protein expression bands of GRP78 and PDI. (F) Representative Western blot analysis of GRP78. (G) Representative Western blot analysis of PDI. Tukey’s post hoc test for one-way ANOVA was used to analyze the Western blot immunocontent. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, and p < 0.001. Trx-1 Inhibits DM-Induced Activation of IRE1α Pathway-Associated Proteins Trx-1 Inhibits DM-Induced Activation of IRE1α Pathway-Associated Proteins was used to detect Nrf2. The results showed that both the Nrf2 gene and immunocontent were downregulated in the WT-DM group compared with the WT-NC group (p < 0.001 and p < 0.05, respectively) and upregulated in the TG-DM group compared with the WT-DM group (p < 0.01 and p < 0.001, respectively) (see Figures 6A–C). y We next investigated the mechanism by which the ERS IRE1α receptor-related pathway is involved in the eﬀects of Trx-1. Western blot analysis was performed to detect the immunocontent of IRE1α, p-IRE1α, TRAF2, ASK1, p-ASK1, JNK, p-JNK, and caspase-12. The results showed that the p-IRE1α, TRAF2, p-ASK1, p-JNK, and caspase-12 immunocontent levels were signiﬁcantly increased in the WT-DM group compared with the WT-NC group. (p < 0.05, p < 0.0001, p < 0.001, p < 0.001, and p < 0.001, respectively) (see Figures 5A–F). However, the abovementioned protein immunocontent levels were signiﬁcantly decreased in the TG-DM group compared with the WT-DM group (p < 0.01, p < 0.0001, p < 0.001, p < 0.001, and p < 0.01, respectively) (see Figures 5A–F). ( g ) Statistically signiﬁcant decreases in the activities of SOD, CAT and GSH-PX were observed in the WT-DM group compared with the WT-NC group (p < 0.001, p < 0.05 and p < 0.0001, respectively), but the MDA activities in the WT-DM group was signiﬁcantly increased compared with it in the WT-NC group (p < 0.001). Furthermore, the SOD, CAT, and GSH-PX activities in the TG-DM group were signiﬁcantly increased compared with those in the WT-DM group (p < 0.01, p < 0.05, and p < 0.05, respectively), but statistically signiﬁcant decreases in the activities of MDA was observed in the TG-DM group compared with the WT-DM group (p < 0.05), and CAT, GSH-PX were also signiﬁcantly increased in the TG-NC group compared with the WT-NC group (p < 0.01 and p < 0.01, respectively) (see Figures 6D–G). Trx-1 Upregulates the Gene and Protein Levels of Nrf2, Antioxidant Defense Parameters and Decreases Phosphorylated NFκB Levels in the Hippocampus of DM Mice DISCUSSION results in neuroﬁbrillary tangle production and neuronal cell dysfunction and death (Ittner et al., 2010). As a molecular chaperone induced by ERS, PDI has been found to coexist with neuroﬁbrillary tangles in the brains of patients with AD and is thought to prevent neurotoxicity associated with ERS and protein misfolding (Li-Rong et al., 2013). It has been reported that PDI and Tau can form a 1:1 complex to prevent the abnormal aggregation of Tau under physiological conditions (Li-Rong et al., 2013). In our study, we found that the immunocontent of p-Tau and PDI in the hippocampus of diabetic mice was signiﬁcantly increased, indicating that diabetes can promote the activation of Tau, and the increase in PDI also indicates that diabetes can exacerbate ERS (see Figures 3, 4). DM is one of the most common diseases aﬀecting humans. The pathological changes of DE are similar to those of AD (Sima, 2010). Moreover, epidemiological studies have shown that diabetic patients have a higher risk of developing AD than healthy people, and there are some common mechanisms and interactions between AD and DE (MacKnight et al., 2002; Fukazawa et al., 2013). These mechanisms have not been fully elucidated. In this study, we explored whether DE caused by DM has AD-like pathological changes. We found that Trx-1 overexpression in transgenic mice alleviated the STZ- induced dysfunction in learning and memory and reduced cell degeneration and apoptosis. This process was achieved by the upregulated Trx-1 exerting resistance to ERS and OS. DM is one of the most common diseases aﬀecting humans. The pathological changes of DE are similar to those of AD (Sima, 2010). Moreover, epidemiological studies have shown that diabetic patients have a higher risk of developing AD than healthy people, and there are some common mechanisms and interactions between AD and DE (MacKnight et al., 2002; Fukazawa et al., 2013). These mechanisms have not been fully elucidated. In this study, we explored whether DE caused by DM has AD-like pathological changes. We found that Trx-1 overexpression in transgenic mice alleviated the STZ- induced dysfunction in learning and memory and reduced cell degeneration and apoptosis. This process was achieved by the upregulated Trx-1 exerting resistance to ERS and OS. A large number of studies have shown that ERS can contribute to neuronal death in AD (Gerakis and Hetz, 2018). Trx-1 Upregulates the Gene and Protein Levels of Nrf2, Antioxidant Defense Parameters and Decreases Phosphorylated NFκB Levels in the Hippocampus of DM Mice Finally, we tested the activation of NFκB, and the results showed that the immunocontent of phosphorylated NFκB was signiﬁcantly increased in the WT-DM group compared with the WT-NC group (p < 0.001) and signiﬁcantly decreased in the TG- DM group compared with the WT-DM group (p < 0.001) (see Figures 6H,I). Changes in the immunocontent of Nrf2 partly reﬂected the degree and development of antioxidant stress. Real-time PCR May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 8 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. Guo et al. FIGURE 5 \| Changes in the activation of IRE1α, TRAF2, ASK1, JNK, and caspase-12 in the IRE1α receptor-related pathway signaling pathway in DM mice. (A) Representative protein expression bands of IRE1α, TRAF2, ASK1, JNK, and caspase-12. (B) Representative Western blot analysis of p-IRE1α/IRE1α. (C) Representative Western blot analysis of caspase-12. (D) Representative Western blot analysis of TRAF2. (E) Representative Western blot analysis of p-ASK1/ASK1. (F) Representative Western blot analysis of p-JNK/JNK. Tukey’s post hoc test for one-way ANOVA was used to analyze the Western blot immunocontent. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, p < 0.001, and **p < 0.0001. FIGURE 5 \| Changes in the activation of IRE1α, TRAF2, ASK1, JNK, and caspase-12 in the IRE1α receptor-related pathway signaling pathway in DM mice. (A) Representative protein expression bands of IRE1α, TRAF2, ASK1, JNK, and caspase-12. (B) Representative Western blot analysis of p-IRE1α/IRE1α. (C) Representative Western blot analysis of caspase-12. (D) Representative Western blot analysis of TRAF2. (E) Representative Western blot analysis of p-ASK1/ASK1. (F) Representative Western blot analysis of p-JNK/JNK. Tukey’s post hoc test for one-way ANOVA was used to analyze the Western blot immunocontent. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, p < 0.001, and **p < 0.0001. Frontiers in Physiology \| www.frontiersin.org DISCUSSION Tukey’s post hoc test for one-way ANOVA was used to analyze the mRNA levels, Western blot immunocontent, and antioxidant defense parameters. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, p < 0.001, and **p < 0.0001. also disturbs the Trx-1/ASK-1 axis, since the enhanced inhibitory binding of Txnip to Trx-1 releases ASK-1, which then induces endothelial cells (EC) apoptosis. Therefore, the regulation of Trx- 1 in the pathogenesis of diabetes may also be accomplished by modulating ERS (Zschauer et al., 2013; see Figure 8). It has also been reported that the expression of Trx-1 is reduced in AD, a fact that suggests that the reduced level of Trx-1 may be related to the neurodegenerative mechanisms of AD: its reduction may contribute to the neurodegenerative mechanisms of AD, which may be related to the interaction of Trx-1 with ASK1 (Akterin et al., 2006). And our results also found that the expression level of Trx-1 was reduced in the hippocampus of mice in the TG-DM group compared with the TG-NC group (see Figure 1) and which is consistent with the results reported above, and the extent of ASK1 phosphorylation was also attenuated in TG-DM compared with WT-DM, Trx-1 may play a key role in AD through the regulation of ASK1 (see Figure 5). NIT-1 cells treated with STZ at diﬀerent stages, in which the expression of GRP78 protein increased at ﬁrst and then decreased (Wang et al., 2007; Cao et al., 2014; Zhao et al., 2015). The change in GRP78 expression partly reﬂects the development process of ERS. In this study, the expression of GRP78 in the WT-DM group decreased signiﬁcantly (see Figure 4), indicating that the development of ERS had reached a later stage, which may also be key to AD-like changes. Trx-1, an important antioxidant protein, not only protects cell activity by exerting its oxidoreductase activity through conserved Cys32 and Cys35, reducing oxidized proteins through thiol disulﬁde exchange reactions (Nagarajan et al., 2016), but also plays an important role in cell signal transduction and gene transcription (Bai et al., 2003). Other studies have shown that Trx-1 overexpression in mice can signiﬁcantly prolong survival during septicemia by inhibiting ERS (Bai et al., 2007; Chen et al., 2016). DISCUSSION GRP78 is a marker of ERS; under non-stress conditions, it binds to three ERS receptors, IRE1α, PERK, and ATF6, and inhibits their activation (Yi et al., 2017). In response to ERS, GRP78 dissociates from these transmembrane proteins, causing them to activate and trigger the UPR (see Figure 7). Moreover, some studies have shown that the upregulation of GRP78 may decrease with time, such as in the hippocampus of STZ-treated diabetic mice and in islets and We also observed Aβ deposition only in the hippocampus of diabetic and transgenic mice, which supported the conclusion that diabetes is a risk factor for AD. The microtubule- associated protein Tau is a cytoskeletal protein mainly expressed by neurons and is primarily located in the axonal chamber (MacKnight et al., 2002). One of the characteristics of AD is the accumulation of misfolded Tau protein in neurons, which May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 9 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. FIGURE 6 \| Effects of Trx-1 on the gene and protein levels of Nrf2, the activation of NFκB and antioxidant defense parameters. (A) Representative protein expression bands of Nrf2. (B) Real-time quantitative PCR of Nrf2 mRNA levels in mice hippocampal tissue. (C) Representative Western blot analysis of Nrf2. (D) Antioxidant defense parameter: SOD activity analysis. (E) Antioxidant defense parameter: CAT activity analysis. (F) Antioxidant defense parameter: GSH-PX activity analysis. (G) Antioxidant defense parameter: MDA activity analysis. (H) Representative bands of the expression of NFκB. (I) Representative Western blot analysis of p-NFκB/NFκB. Tukey’s post hoc test for one-way ANOVA was used to analyze the mRNA levels, Western blot immunocontent, and antioxidant defense parameters. For each group, n = 6. The vertical lines represent the standard error of the mean. p < 0.05, p < 0.01, p < 0.001, and ***p < 0.0001. FIGURE 6 \| Effects of Trx-1 on the gene and protein levels of Nrf2, the activation of NFκB and antioxidant defense parameters. (A) Representative protein expression bands of Nrf2. (B) Real-time quantitative PCR of Nrf2 mRNA levels in mice hippocampal tissue. (C) Representative Western blot analysis of Nrf2. (D) Antioxidant defense parameter: SOD activity analysis. (E) Antioxidant defense parameter: CAT activity analysis. (F) Antioxidant defense parameter: GSH-PX activity analysis. (G) Antioxidant defense parameter: MDA activity analysis. (H) Representative bands of the expression of NFκB. (I) Representative Western blot analysis of p-NFκB/NFκB. Frontiers in Physiology \| www.frontiersin.org DISCUSSION Our previous studies have shown that sulforaphane (SF) can protect diabetic hippocampal neurons by upregulating Trx-1 expression (Tang et al., 2020). Indeed, in animal model of STZ-induced diabetes in mice, ROS were increased and Trx-1 activity was signiﬁcantly decreased without a change in expression or protein levels in the diabetic animals in comparison to untreated littermates (Schulze et al., 2004). One may speculate that this glucose-dependent upregulation of Txnip In ERS signaling pathways, p-IRE1α is thought to initiate apoptosis and is signiﬁcantly related to cell death (Kim et al., 2017). Activated IRE1α has been shown to recruit the adaptive molecules TRAF2 and ASK1 to form a complex, activate caspase- 12 on the ER and activate JNK to initiate apoptosis (see Figure 7; Szegezdi et al., 2006). Trx-1 has been reported to act as an endogenous inhibitor of ASK1 (Saitoh et al., 1998); when Trx-1 May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 10 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. FIGURE 7 \| Schematic illustration of the DM-induced ERS and OS. DM induced apoptosis through activation of the signaling pathway IRE1α, TRAF2, ASK1, JNK, caspase-12, and the pathway PERK, CHOP, caspase-3. Meanwhile, the reduction of Nrf2 induced by activation of PERK and the increase of NFκB induced by activation of TRAF2 contribute to enhance oxidative stress, further lead to a increase of PDI, p-TAU and accumulation of misfolded proteins which in turn increased the activation of the receptor IRE1α and PERK. FIGURE 7 \| Schematic illustration of the DM-induced ERS and OS. DM induced apoptosis through activation of the signaling pathway IRE1α, TRAF2, ASK1, JNK, caspase-12, and the pathway PERK, CHOP, caspase-3. Meanwhile, the reduction of Nrf2 induced by activation of PERK and the increase of NFκB induced by activation of TRAF2 contribute to enhance oxidative stress, further lead to a increase of PDI, p-TAU and accumulation of misfolded proteins which in turn increased the activation of the receptor IRE1α and PERK. FIGURE 8 \| Schematic illustration of the role of Trx-1 in regulating DM-induced ERS and OS. On the one hand, the upregulation of Trx-1 increased Nrf2 and inhibited NFκB, leading to the increase of anti-oxidant enzymes and the decrease of oxidases to exert antioxidant stress and anti-endoplasmic reticulum stress, which reduced misfold protein production. DISCUSSION These could be potential mechanisms by which the decline in learning and memory induced by diabetes was inhibited in transgenic diabetic mice. In summary, DE causes AD-like pathological changes, and the Trx-1 transgene alleviated these pathological changes and improved learning and memory function in DE. The upregulation of Trx-1 increased Nrf2 and inhibited NFκB and ASK1 phosphorylation, which exerted anti-endoplasmic reticulum stress, antioxidant stress, and antiapoptotic eﬀects (see Figure 8). Therefore, Trx-1 may be a potential target for regulating ERS and OS in DE and AD. Furthermore, Nrf2 is considered to be a substrate of PERK. The phosphorylation of PERK can trigger conformational changes in the Nrf2 protein and trigger the dissociation of the Keap1-Nrf2 complex and the translocation of Nrf2 into the nucleus, thus upregulating the expression of antioxidant genes (Zhu et al., 2015), and this activates the transcription of more than 200 stress/antioxidant enzymatic genes such as glutathione, glutathione reductase, thioredoxin, thioredoxin-S- transferase, and catalase (Thimmulappa et al., 2002; Aldana et al., 2003; Datta et al., 2017; Periyasamy and Shinohara, 2017). Therefore, there is a clear correlation between OS and ERS through the PERK/Nrf2 pathway (see Figure 7; Thummayot et al., 2016). Thus, interfering with the Nrf2 response leads to the accumulation of damaged proteins in the ER, which leads to PERK-dependent apoptosis, suggesting that OS can induce ERS, and excessive ERS can also cause or exacerbate OS through PERK/Nrf2 crosstalk (Hayashi et al., 2005). In this study, we also conﬁrmed that at both the protein and gene levels, the expression of Nrf2 in the WT-DM group was lower than that in the WT- NC group, and other oxidative stress-related indicators, such as SOD, CAT and GSH-PX, showed the same trend as Nrf2, but the MDA showed the reverse trend as Nrf2 (see Figure 6). SOD is an important antioxidant enzyme, which can eﬀectively scavenge free radicals and thus resist the damage of oxygen free radicals to cells (Wang et al., 2018). Therefore, SOD plays a very important role in the balance of oxidation and antioxidation, and its activity indirectly reﬂects the ability of the body to scavenge oxygen free radicals. DISCUSSION On the other hand, the upregulation of Trx-1 inhibited ASK1 phosphorylation, decreased the expression of JNK and caspase-12, further exerted antiapoptotic effects. FIGURE 8 \| Schematic illustration of the role of Trx-1 in regulating DM-induced ERS and OS. On the one hand, the upregulation of Trx-1 increased Nrf2 and inhibited NFκB, leading to the increase of anti-oxidant enzymes and the decrease of oxidases to exert antioxidant stress and anti-endoplasmic reticulum stress, which reduced misfold protein production. On the other hand, the upregulation of Trx-1 inhibited ASK1 phosphorylation, decreased the expression of JNK and caspase-12, further exerted antiapoptotic effects. May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 11 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. is reduced it binds ASK1. Thus, overexpression of Trx-1 may repress JNK activation by suppressing ASK1 (see Figure 8). An increase in the ERS has been shown to result in the activation of NFκB. The IRE1α-TRAF2 complex recruits I kappa B (IκB) kinase, which phosphorylates IκB and leads to NFκB degradation, leading to the exacerbation of inﬂammatory reactions and OS (Zha et al., 2015). Recent study demonstrated that CHOP could also modulate NFκB activation by decreasing IκB degradation and p65 translocation (see Figure 7). We further examined the expressions of NFκB in the present study. We demonstrated that Trx-1 overexpression inhibited the NFκB activation in DE and AD induced DM (Figure 6). This data suggested that Trx-1 overexpression was able to inhibit the NFkB signaling pathway (see Figure 8). On the other hand, GRP78 induces the PERK pathway and activates the downstream expression of CHOP and caspase-3 to initiate apoptosis (see Figure 7; Sankrityayan et al., 2019). All of these factors may eventually lead to neuronal apoptosis, and IRE1α, TRAF2, ASK1, JNK, caspase-12, PERK, CHOP, and caspase-3 were activated in this study, indicating that ERS was induced by diabetes (see Figure 4). through the upregulation of Trx-1. In turn, the elevated Nrf2 aﬀected SOD, CAT, MDA and GSH-PX, reversing the decrease in SOD, CAT and GSH-PX activity and increase in MDA, thereby counteracting DM-induced ERS and OS and reducing PDI as well as misfolded protein production, the degree of neuronal apoptosis and degeneration and the production of Aβ plaques, indicating that Trx-1 has antidiabetic eﬀects and opposes AD-like changes (see Figures 4, 6). DISCUSSION MDA is the product of lipid peroxidation in brain tissue, especially during brain injury, a large amount of oxygen free radicals can make lipid peroxidation, so the increase of MDA content further indicates that oxidative stress is active in diabetes, thus proving the importance of oxidative stress in the occurrence and development of DE (Wang et al., 2018). These results showed that in the later stages of ERS, ERS and OS induce and promote each other. Therefore, in this study, in Trx-1 transgenic mice (both transgenic control and transgenic diabetic mice), the elevated Nrf2 levels were achieved ETHICS STATEMENT The animal study was reviewed and approved by All procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (NIH), and all protocols were approved by the Institutional Animal Care and Use Committee of Dalian Medical University. FUNDING This work was supported by the National Natural Science Foundation of China (grant number 31300812), the Natural Science Foundation of Liaoning Province (grant number 20180550468), and the Liaoning Provincial Program for Top Discipline of Basic Medical Sciences. DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s. AUTHOR CONTRIBUTIONS All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication. Frontiers in Physiology \| www.frontiersin.org REFERENCES Holmgren, A. (1985). Thioredoxin. Annu. Rev. Biochem. 54, 237–271. Ittner, L. M., Ke, Y. D., Delerue, F., Bi, M., Gladbach, A., van Eersel, J., et al. (2010). Dendritic function of tau mediates amyloid-beta toxicity in Alzheimer’s disease mouse models. Cell 142, 387–397. doi: 10.1016/j.cell.2010. 06.036 Akterin, S., Cowburn, R. F., Miranda-Vizuete, A., Jiménez, A., Bogdanovic, N., Winblad, B., et al. (2006). Involvement of glutaredoxin-1 and thioredoxin-1 in beta-amyloid toxicity and Alzheimer’s disease. Cell Death Diﬀer. 13, 1454–1465. doi: 10.1038/sj.cdd.4401818 Jin, J. K., Blackwood, E. A., Azizi, K., Thuerauf, D. J., Fahem, A. G., Hofmann, C., et al. (2016). ATF6 decreases myocardial ischemia/reperfusion damage and links er stress and oxidative stress signaling pathways in the heart. Circ. Res. 120:862. doi: 10.1161/circresaha.116.310266 Aldana, J. P. A., Marcovich, R., Singhal, P., Reddy, K., Morgenstern, N., El-Hakim, A., et al. (2003). Immune response to laparoscopic bowel injury. J. Endourol. 17, 317–322. doi: 10.1089/089277903322145503 120:862. doi: 10.1161/circresaha.116.310266 Almanza, A., Carlesso, A., Chintha, C., Creedican, S., Doultsinos, D., Leuzzi, B., et al. (2018). Endoplasmic reticulum stress signalling - from basic mechanisms to clinical applications. FEBS J. 286, 241–278. Matés, JM., and Sánchez-Jiménez, FM. (2000). Role of reactive oxygen species in apoptosis: implications for cancer therapy. Int. J. Biochem. Cell Biol. 32, 157–170. doi: 10.1016/s1357-2725(99)00088-6 Asano, T., Fujishiro, M., Fujishiro, M., Nakatsu, Y., Yoneda, M., Kamata, H., et al. (2007). Role of phosphatidylinositol 3-kinase activation on insulin action and its alteration in diabetic conditions. Biol. Pharmaceutic. Bull. 30:1610. doi: 10.1248/bpb.30.1610 Jui-Ching, C., Wu, M. L., Huang, K. C., and Lin, W. W. (2008). HMG- CoA reductase inhibitors activate the unfolded protein response and induce cytoprotective GRP78 expression. Cardiovasc. Res. 80, 138–150. doi: 10.1093/ cvr/cvn160 Kim, J., Song, H., Heo, H. R., Kim, J. W., Kim, H. R., Hong, Y., et al. (2017). Cadmium-induced ER stress and inﬂammation are mediated through C/EBP– DDIT3 signaling in human bronchial epithelial cells. Exp. Mol. Med. 49:e372. doi: 10.1038/emm.2017.125 Aubert, L., Pichierri, S., Hommet, C., Camus, V., Berrut, G., de Decker, L., et al. (2015). Association between comorbidity burden and rapid cognitive decline in individuals with mild to moderate Alzheimer’s disease. J. Am. Geriatr. Soc. 63, 543–547. doi: 10.1111/jgs.13314 Lim, J. L., Wilhelmus, M. M., de Vries, H. E., Drukarch, B., Hoozemans, J. J., and van Horssen, J. (2014). Antioxidative defense mechanisms controlled by Nrf2: state-of-the-art and clinical perspectives in neurodegenerative diseases. Arch. Toxicol. 88, 1773–1786. REFERENCES Recent insights into the role of unfolded protein response in er stress in health and disease. Front. Cell Dev. Biol. 5:48. doi: 10.3389/fcell.2017.00048 Periyasamy, P., and Shinohara, T. (2017). Age-related cataracts: role of unfolded protein response, ca2+ mobilization, epigenetic DNA modiﬁcations, and loss of Nrf2/Keap1 dependent cytoprotection. Prog. Retin. Eye Res. 60, 1–19. doi: 10.1016/j.preteyeres.2017.08.003 Datta, S., Cano, M., Ebrahimi, K., Wang, L., and Handa, J. T. (2017). The impact of oxidative stress and inﬂammation on RPE degeneration in non-neovascular AMD. Prog. Retinal Eye Res. 60, 201–218. doi: 10.1016/j.preteyeres.2017.03.002 Roshan, V. D., Hosseinzadeh, S., Mahjoub, S., Hosseinzadeh, M., and Myers, J. (2013). Endurance exercise training and diferuloyl methane supplement: changes in neurotrophic factor and oxidative stress induced by lead in rat brain. Biol. Sport 30, 41–46. doi: 10.5604/20831862.1029820 Díaz-Gerevini, G. T., Repossi, G., Dain, A., Tarres, M. C., Das, U. N., and Eynard, A. R. (2014). Cognitive and motor perturbations in elderly with longstanding diabetes mellitus. Nutrition 30, 628–635. doi: 10.1016/j.nut.2013.11.007 Saitoh, M., Nishitoh, H., Fujii, M., Takeda, K., Tobiume, K., Sawada, Y., et al. (1998). Mammalian thioredoxin is a direct inhibitor of apoptosis signal- regulating kinase (ASK) 1. EMBO J. 17, 2596–2606. doi: 10.1093/emboj/17.9. 2596 Fukazawa, R., Hanyu, H., Sato, T., Shimizu, S., Koyama, S., Kanetaka, H., et al. (2013). Subgroups of Alzheimer’s disease associated with diabetes mellitus based on brain imaging. Demen. Geriatr. Cogn. Disord. 35, 280–290. doi: 10.1159/ 000348407 Gerakis, Y., and Hetz, C. (2018). Emerging roles of ER stress in the etiology and pathogenesis of Alzheimer’s disease. FEBS J. 285, ,995–1011. Sankrityayan, H., Oza, M. J., Kulkarni, Y. A., Mulay, S. R., and Gaikwad, A. B. (2019). ER stress response mediates diabetic microvascular complications. Drug Discovery Today 24, 2247–2257. doi: 10.1016/j.drudis.2019.08.003 Glory, A., and Averill-Bates, D. A. (2016). The antioxidant transcription factor Nrf2 contributes to the protective eﬀect of mild thermotolerance (40◦C) against heat shock-induced apoptosis. Free Radic. Biol. Med. 99, 485–497. doi: 10.1016/j. freeradbiomed.2016.08.032 Schulze, P. C., Yoshioka, J., Takahashi, T., He, Z., King, G. L., and Lee, R. T. (2004). Hyperglycemia promotes oxidative stress through inhibition of thioredoxin function by thioredoxin-interacting protein. J. Biol. Chem. 279, 30369–30374. doi: 10.1074/jbc.m400549200 Han, C., Wang, L., and Wang, C. (2009). Domain a’ of protein disulﬁde isomerase plays key role in inhibiting α-synuclein ﬁbril formation. Cell Stress Chaperones 15, 415–421. doi: 10.1007/s12192-009-0157-2 Selkoe, D. J. (2001). Alzheimer’s disease: genes, proteins, and therapy. Physiol. Rev. 81, 741–766. REFERENCES doi: 10.1007/s00204-014- 1338-z Bai, J., Nakamura, H., Kwon, Y. W., Hattori, I., Yamaguchi, Y., Kim, Y. C., et al. (2003). Critical roles of thioredoxin in nerve growth factor-mediated signal transduction and neurite outgrowth in PC12 cells. J. Neurosci. 23, 503–509. doi: 10.1523/jneurosci.23-02-00503.2003 Bai, J., Nakamura, H., Kwon, Y. W., Tanito, M., Ueda, S., Tanaka, T., et al. (2007). Does thioredoxin-1 prevent mitochondria- and endoplasmic reticulum- mediated neurotoxicity of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine? Antioxid Redox Signal. 9, 603–608. doi: 10.1089/ars.2006.1513 Li-Rong, X., Liu, X. L., Chen, J., and Liang, Y. (2013). Protein disulﬁde isomerase interacts with tau protein and inhibits its ﬁbrillization. PLoS One 8:e76657. Luo, F. C., Zhou, J., Qi, L., Wang, S. D., Nakamura, H., Yodoi, J., et al. (2012). Induction of endoplasmic reticulum stress and the modulation of thioredoxin- 1 in formaldehyde-induced neurotoxicity. NeuroToxicology 33, 290–298. doi: 10.1016/j.neuro.2012.02.004 Biessels, G. J., ter Laak, M. P., Hamers, F. P., and Gispen, W. H. (2002). Neuronal Ca2+ disregulation in diabetes mellitus. Eur. J. Pharmacol. 447, 201–209. doi: 10.1016/s0014-2999(02)01844-7 Lupachyk, S., Watcho, P., Stavniichuk, R., Shevalye, H., and Obrosova, I. G. (2013). Endoplasmic reticulum stress plays a key role in the pathogenesis of diabetic peripheral neuropathy. Diabetes 62, 944–952. doi: 10.2337/db12-0716 Cao, Y., Hao, Y., Li, H., Liu, Q., Gao, F., Liu, W., et al. (2014). Role of endoplasmic reticulum stress in apoptosis of diﬀerentiated mouse podocytes induced by high glucose. Int. J. Mol. Med. 33, 809–816. doi: 10.3892/ijmm.2014.1642 Chen, G., Chen, G., Li, X., Huang, M., Li, M., Zhou, X., et al. (2016). Thioredoxin-1 increases survival in sepsis by inﬂammatory response through suppressing endoplasmic reticulum stress. Shock 46, 67–74. doi: 10.1097/shk. 0000000000000570 MacKnight, C., Rockwood, K., Awalt, E., and McDowell, I. (2002). Diabetes mellitus and the risk of dementia, Alzheimer’s disease and vascular cognitive impairment in the Canadian Study of Health and Aging. Demen. Geriatr. Cogn. Disord. 14, 77–83. doi: 10.1159/000064928 Clodfelder-Miller, B., Zmijewska, A. A., Johnson, G. V., and Jope, R. S. (2006). Tau is hyperphosphorylated at multiple sites in mouse brain in vivo after streptozotocin-induced insulin deﬁciency. Diabetes 55, 3320–3325. doi: 10. 2337/db06-0485 Nagarajan, N., Oka, S., and Sadoshima, J. (2016). Modulation of signaling mechanisms in the heart by thioredoxin 1. Free Radic. Biol. Med. 109, 125–131. doi: 10.1016/j.freeradbiomed.2016.12.020 Nakamura, H. (2004). Thioredoxin as a key molecule in redox signaling. Antioxid Redox Signal. 6, 15–17. doi: 10.1089/152308604771978309 Dan, L., Korhonen, L., Eriksson, O., and Kõks, S. (2017). ACKNOWLEDGMENTS YG wants to thank, in particular, the patience, care, understanding and support from Zhanwei Sun all the time. May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 12 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. REFERENCES Drug Targets 13, 290–298. Zeng, X. S., Jia, J. J., Kwon, Y., Wang, S. D., and Bai, J. (2014). The role of thioredoxin-1 in suppression of endoplasmic reticulum stress in Parkinson disease. Free Radic. Biol. Med. 67, 10–18. Thimmulappa, R. K., Mai, K. H., Srisuma, S., Kensler, T. W., Yamamoto, M., and Biswal, S. (2002). Identiﬁcation of Nrf2-regulated genes induced by the chemopreventive agent sulforaphane by oligonucleotide microarray. Cancer Res. 62, 5196–5203. Zha, X., Yue, Y., Dong, N., and Xiong, S. (2015). Endoplasmic reticulum stress aggravates viral myocarditis by raising inﬂammation through the IRE1- associated NF-κB pathway. Can. J. Cardiol. 31, 1032–1040. Thummayot, S., Tocharus, C., Suksamrarn, A., and Tocharus, J. (2016). Neuroprotective eﬀects of cyanidin against Aβ-induced oxidative and ER stress in SK-N-SH cells. Neurochem. Int. 101, 15–21. doi: 10.1016/j.neuint.2016. 09.016 Zhao, Y., Yan, Y., Zhao, Z., Li, S., and Yin, J. (2015). The dynamic changes of endoplasmic reticulum stress pathway markers GRP78 and CHOP in the hippocampus of diabetic mice. Brain Res. Bull. 111, 27–35. Zhu, W., Wang, X. R., Du, S. Q., Yan, C. Q., Yang, N. N., Lin, L. L., et al. (2018). Anti-oxidative and anti-apoptotic eﬀects of acupuncture: role of thioredoxin-1 in the hippocampus of vascular dementia rats. Neuroence 379:S0306452218302203. Wang, C., Lü, G., Li, Y., Zhao, S., and Huang, L. (2018). Impairment of spatial learning and memory and changes of oxidative stress in hippocampus from Type 1 diabetic mice. Zhong Nan Da Xue Xue Bao Yi Xue Ban 43, 469–474. Zhu, Y. F., Li, X. H., Yuan, Z. P., Li, C. Y., Tian, R. B., Jia, W., et al. (2015). Allicin improves endoplasmic reticulum stress-related cognitive deﬁcits via PERK/Nrf2 antioxidative signaling pathway. Eur. J. Pharmacol. 762, 239–246. Wang, M., Zhao, X. R., Wang, P., Dai, Y., Huang, H., Zhu, H. F., et al. (2007). Glucose regulated proteins 78 protects insulinoma cells (NIT-1) from death induced by streptozotocin, cytokines or cytotoxic T lymphocytes. Int. J. Biochem. Cell Biol. 39, 2076–2082. doi: 10.1016/j.biocel.2007.05.022 Zschauer, T. C., Matsushima, S., Altschmied, J., Shao, D., Sadoshima, J., and Haendeler, J. (2013). Interacting with thioredoxin-1–disease or no disease? Antioxidants Redox Signal. 18, 1053–1062. Wang, Z., Huang, Y., Cheng, Y., Tan, Y., Wu, F., Wu, J., et al. (2016). Endoplasmic reticulum stress-induced neuronal inﬂammatory response and apoptosis likely plays a key role in the development of diabetic encephalopathy. Oncotarget 7, 78455–78472. REFERENCES Hayashi, T., Saito, A., Okuno, S., Ferrand-Drake, M., Dodd, R. L., and Chan, P. H. (2005). Damage to the endoplasmic reticulum and activation of apoptotic machinery by oxidative stress in ischemic neurons. J. Cereb. Blood Flow Metab. 25, 41–53. doi: 10.1038/sj.jcbfm.9600005 Sima, A. A. F. (2010). Encephalopathies: the emerging diabetic complications. Acta Diabetol. 47, 279–293. doi: 10.1007/s00592-010-0218-0 Stamer, K., Vogel, R., Thies, E., Mandelkow, E., and Mandelkow, E. M. (2002). Tau blocks traﬃc of organelles, neuroﬁlaments, and APP vesicles in neurons Stamer, K., Vogel, R., Thies, E., Mandelkow, E., and Mandelkow, E. M. (2002). Sta e , ., Voge , R., es, ., a de kow, ., a d a de kow, . . ( 00 ). Tau blocks traﬃc of organelles, neuroﬁlaments, and APP vesicles in neurons May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 13 Thioredoxin-1 Attenuates Alzheimer-Like Pathology in DE Guo et al. and enhances oxidative stress. J. Cell Biol. 156, 1051–1063. doi: 10.1083/jcb. 200108057 Yi, S., Shi, W., Wang, H., Ma, C., Zhang, X., Wang, S., et al. (2017). Endoplasmic reticulum stress PERK-ATF4-CHOP pathway is associated with hypothalamic neuronal injury in diﬀerent durations of stress in rats. Front. Neurosci. 11:152. doi: 10.3389/fnins.2017.00152 Stranahan, M. A. (2015). Models and mechanisms for hippocampal dysfunction in obesity and diabetes. Neuroence 309, 125–139. doi: 10.1016/j.neuroscience. 2015.04.045 Zahraa, M. A., Cruz, G. L., and Dickhout, J. G. (2015). Crosstalk between the unfolded protein response and NF- κ B-mediated inﬂammation in the progression of chronic kidney disease. J. Immunol. Res. 2015, 1–10. Szegezdi, E., Logue, S. E., Gorman, A. M., and Samali, A. (2006). Mediators of endoplasmic reticulum stress-induced apoptosis. EMBO Rep. 7, 880–885. doi: 10.1038/sj.embor.7400779 g y Zargar, M. A., Sheikh, I. A., Ganie, S. A., Ali, R., Singh, L. R., Gan, S. H., et al. (2014). molecular linkages between diabetes and Alzheimer’s disease: current scenario Zargar, M. A., Sheikh, I. A., Ganie, S. A., Ali, R., Singh, L. R., Gan, S. H., et al. (2014). molecular linkages between diabetes and Alzheimer’s disease: current scenario and future prospects. CNS Neurol. Disord. Drug Targets 13, 290–298. Tang, L., Ren, X., Han, Y., Chen, L., Meng, X., Zhang, C., et al. (2020). Sulforaphane attenuates Apoptosis of hippocampal neurons induced by high glucose via regulating endoplasmic reticulum. Neurochem. Int. 136:104728. doi: 10.1016/j. neuint.2020.104728 molecular linkages between diabetes and Alzheimer’s disease: current scenario and future prospects. CNS Neurol. Disord. Frontiers in Physiology \| www.frontiersin.org REFERENCES Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Wu, H., Ye, M., Yang, J., and Ding, J. (2016). Modulating endoplasmic reticulum stress to alleviate myocardial ischemia and reperfusion injury from basic research to clinical practice: a long way to go. Int. J. Cardiol. 223, 630–631. Copyright © 2021 Guo, Zhang, Wang, Huang, Liu, Chu, Ren, Kong and Ma. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Copyright © 2021 Guo, Zhang, Wang, Huang, Liu, Chu, Ren, Kong and Ma. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Xu, L. R., Liu, X. L., Chen, J., and Liang, Y. (2013). Protein disulﬁde isomerase interacts with tau protein and inhibits its ﬁbrillization. PLoS One 8:e76657. doi: 10.1371/journal.pone.0076657 Yang, B., Xu, Y., Hu, Y., Luo, Y., Lu, X., Tsui, C. K., et al. (2016). Madecassic acid protects against hypoxia-induced oxidative stress in retinal microvascular endothelial cells via ros-mediated endoplasmic reticulum stress. Biomed. Pharmacother. 84, 845–852. May 2021 \| Volume 12 \| Article 651105 Frontiers in Physiology \| www.frontiersin.org 14
https://openalex.org/W1592357273	https://www.redalyc.org/pdf/742/74232097011.pdf	Portuguese	null	Religião, educação e economia em Max Weber	Civitas - Revista de Ciências Sociais	2,014	cc-by	7,002	Civitas - Revista de Ciências Sociais ISSN: 1519-6089 civitas@pucrs.br Pontifícia Universidade Católica do Rio Grande do Sul Brasil de Carvalho Filho, Juarez Lopes Religião, educação e economia em Max Weber Civitas - Revista de Ciências Sociais, vol. 14, núm. 3, septiembre-diciembre, 2014, pp. 540-555 Pontifícia Universidade Católica do Rio Grande do Sul Porto Alegre, Brasil Disponível em: http://www.redalyc.org/articulo.oa?id=74232097011 Disponível em: http://www.redalyc.org/articulo.oa?id=74232097011 Como citar este artigo Número completo Mais artigos Home da revista no Redalyc Sistema de Informação Científica Rede de Revistas Científicas da América Latina, Caribe , Espanha e Portugal Projeto acadêmico sem fins lucrativos desenvolvido no âmbito da iniciativa Acesso Aberto Sistema de Informação Científica Rede de Revistas Científicas da América Latina, Caribe , Espanha e Portugal Projeto acadêmico sem fins lucrativos desenvolvido no âmbito da iniciativa Acesso Aberto Artigos Artigos Religion, education and the economy in Max Weber Carvalho Filho – Religião, educação e economia em Max Weber Introdução Religion, education and the economy in Max Weber Juarez Lopes de Carvalho Filho* Juarez Lopes de Carvalho Filho* Resumo: Este artigo propõe relacionar a religião, a educação e a economia em Max Weber. Apesar de não ter escrito uma obra específica sobre a educação, é possível verificar em escritos como Confucionismo e taoísmo um esboço de tipos ideais da educação encontrados em qualquer sistema de formação, que Weber articulava aos tipos de dominação legitima (racional-legal, tradicional, carismática), seja política seja religiosa. Cada tipo de educação desenvolve nos alunos uma cultura, uma conduta de vida, ou as qualidades que convêm a um tipo de dominação político-religiosa. Esses elementos dos tipos ideais de educação articulados a uma cultura religiosa exercem um papel importante na formação de sistemas econômicos, notadamente de uma cultura capitalista. Como Weber demonstrou na Ética protestante e o “espírito” do capitalismo, existe uma real adaptação, de uma educação religiosa (ascética) ao capitalismo e à sua cultura. Palavras-chave: Educação confucianista. Educação ascética. Cultura capitalista. Conduta de vida Abstract: This paper proposes to relate religion, education and the economy in Max Weber. Despite not having a specific written work on education, it is possible to read texts like Confucianism and Taoism as an outline of ideal types of education, present in any training system. Weber has articulated these types of education with the types of (rational-legal, traditional and charismatic) legitimate political as well as religious domination. Each type of education develops in students a culture, a conduct of life, or the qualities befitting a type of political and religious domination. These elements of the ideal types of education articulated a religious culture play an important role in the formation of economic systems, especially in a capitalist culture. As demonstrated in Weber’s Protestant ethic and the “spirit” of Capitalism, there is a real adaptation of a religious (ascetic) education to capitalism and its culture. Keywords: Confucian education. Ascetic education. Capitalist culture. Conduct of life. Civitas Porto Alegre v. 14 n. 3 p. 540-555 set.-dez. 2014 A matéria publicada neste periódico é licenciada sob forma de uma Licença Creative Commons - Atribuição 4.0 Internacional. http://creativecommons.org/licenses/by/4.0/ * Doutor em Sciences Sociales et Économiques no Institut Catholique de Paris (França), é professor na Universidade Federal do Maranhão em São Luis, MA, Brasil <juarez.lopes@ gmail.com>. * Doutor em Sciences Sociales et Économiques no Institut Catholique de Paris (França), professor na Universidade Federal do Maranhão em São Luis, MA, Brasil <juarez.lopes@ gmail.com>. 541 J. L. Sociologia da religião: uma sociologia da dominação A partir das relações entre o protestantismo e o capitalismo, Weber amplia o problema de uma sociologia comparada das religiões, em uma história comparada do capitalismo. Deste último, encontramos elementos essenciais nos estudos de sociologia das religiões. Esses estudos são a continuidade direta do estudo consagrado ao protestantismo. Ou seja, a Ética econômica das religiões mundiais e a Ética protestante e o espírito do capitalismo formam um conjunto coerente que, atualizando os “processos de racionalização religiosa em longo prazo”, fornece a chave do desenvolvimento específico da cultura ocidental: o protestantismo constituiria o desfecho do processo de “desencantamento do mundo” inaugurado pelo judaísmo (Grossein, 2000a, p. 78). A curiosidade histórica de Weber se estendeu a todas as épocas. Ele analisou tanto a ação dos fatores políticos e sociais como dos fatores religiosos, tendo estes um lugar privilegiado nos dispositivos teóricos de seu projeto de uma sociologia histórica e comparativa. De 1911 a 1914, ele se dedica aos esboços de uma sociologia da religião consagrados sobre a ética econômica das religiões universais. A partir de 1916, ele publica seus estudos nos Archiv für Sozialwissenschaft und Sozialpolitik, temendo que suas reflexões não fossem retomadas após a Primeira Guerra. Os artigos deveriam aparecer simultaneamente com Wirtschaft und Gesellschaft contido no Grundriss der Sozialökonomik. Eles eram destinados a explicar e completar a parte consagrada à sociologia da religião. Logo da publicação dessa versão, ainda imperfeita, Weber acrescenta uma Introdução que ele redige em 1915 e na qual ele empreende uma espécie de primeiro resumo provisório dos principais resultados; no mesmo ano ele redige as Considerações intermediárias. Em 1920, antes de morrer, Weber reúne os estudos de sociologia da religião em três volumes intitulados Coletânea de estudos de sociologia das religiões. Para a reedição, Weber rever a Ética protestante e Confucionismo e taoísmo e insere um “Prefácio”. Ele fala em termos de uma mise en œuvre do método aplicado em Ética protestante e na Coletânea de estudos de sociologia das religiões; a análise insiste sobre os efeitos que a religião enquanto sistema de regulamentação da vida exerce sobre a conduta de vida cotidiana das massas. No início de seus estudos sobre a sociologia da religião, Weber insiste sobre a originalidade dos produtos culturais da idade capitalista; existe nele a ideia de que esses produtos culturais sejam ligados a uma época e a uma estrutura de certo tipo de sociedade. Introdução Civitas, Porto Alegre, v. 14, n. 3, p. 540-555, set.-dez. 2014 542 Introdução Este artigo propõe relacionar a religião, a educação e a economia em Max Weber. Nesse sentido, a reflexão se situa ao mesmo tempo no campo da sociologia das religiões, da sociologia da educação e da sociologia econômica. É muito comum falar de uma sociologia das religiões e de uma sociologia econômica desenvolvidas por esse autor. O mesmo não se pode dizer de uma sociologia da educação, talvez por dois motivos: em primeiro, Weber não escreveu nenhuma obra específica sobre a educação, como foi o caso de Émile Durkheim; em segundo, pelo fato de o texto Confucionismo e taoísmo, no qual Weber esboça uma tipologia da educação, ter ficado muito tempo desvinculado de seu projeto intelectual e científico. Neste texto, Weber relacionou os tipos de educação com os tipos de dominação legitima (racional-legal, tradicional, carismática), seja política seja religiosa, sem deduzir os primeiros dos segundos. O propósito do texto é demonstrar, na teoria de Weber, como uma religião adquire uma posição de orientação de conduta de vida das ca- madas socialmente dominantes antes de se tornar um fator de afeiçoamento de uma sociedade. E como numa cultura tradicional, como a chinesa, os determinantes de sua conduta de vida, seu estilo de vida, se impõem para além de seu próprio círculo, até impregnar a sociedade inteira (Grossein, 2000b, p. XVIII). Apresenta-se inicialmente elementos constitutivos do projeto de uma sociologia das religiões em Weber, reconhecidamente uma das maiores preocupações do sociólogo alemão. Em seguida, expõe-se sua compreensão da ética econômica das religiões mundiais. A relação entre ética econômica das religiões mundiais e a gênese de uma cultura capitalista amplia os horizontes das argumentações de Weber, numa série de comparações a fim de definir a singularidade do capitalismo moderno e do Ocidente. Enquanto a Ética protestante busca as origens do “espírito” do capitalismo no protestantismo ascético (que figura apenas como um elemento causal), a explicação weberiana a partir da ética das religiões se funda em elementos multicausais. Por último, descreve-se a tipologia sociológica da educação esboçada em Confucionismo e taoísmo, colocada em correspondência com os modos de dominação: cada tipo de educação desenvolve nos alunos a cultura, a conduta de vida, as qualidades que convêm a cada tipo de dominação político-religiosa. O projeto de Weber parece ser o mesmo: buscar, pelo viés de uma sociologia histórica e comparada das religiões mundiais, a singularidade da civilização ocidental dominada pelo processo de racionalização crescente. Sociologia da religião: uma sociologia da dominação Ele engloba numa visão de conjunto as relações das principais religiões com a economia e a estratificação social de seu universo social. 543 J. L. Carvalho Filho – Religião, educação e economia em Max Weber No que concerne a sua observação da religião, Weber não se preocupa, como Durkheim, em As formas elementares da vida religiosa, com a busca de uma definição prévia. Em Economia e sociedade, enfatizando a impossibi- lidade de definir a priori a religião, ele concebe-a como “uma maneira particular do modo de agir em comunidade” (Weber, 1995, p. 145). Weber entende ser necessário analisar as condições e os efeitos de tal compor- tamento. Nesse sentido, tanto Weber quanto Durkheim rompem com a pretensão fenomenológica de captar uma essência da religião, uma vez que o sociólogo francês define a religião como um dos mais primitivos fenômenos sociais. Weber, nas suas análises sobre a religião, expõe duas observações que implicam em grandes consequências. A primeira é a seguinte: As formas mais elementares do comportamento motivado por fatores religiosos ou mágicos são orientadas para o mundo terrestre. Os atos ditados pela religião ou pela magia devem ser realizados ‘a fim de se conquistar [...] a felicidade e uma longa vida na terra’ (Weber, 1995, p. 145-146). Para Willaime (2012, p. 50) “essa afirmação representa uma reversão considerável com relação a todas as perspectivas teóricas que identificam interesses religiosos e interesses pelo além a uma falta de distância evidente com o próprio discurso religioso”. Para Weber “mesmo através de referências a uma ou outra manifestação do além, a religião diz respeito à vida terrestre” (Weber, 1995, p. 145-146). A segunda observação é uma ruptura: é que Weber se recusa a associar o religioso ao irracional: “Os atos motivados pela religião ou pela magia são atos, ao menos relativamente, racionais”. Segundo Willaime (2012, p. 50), uma das principais contribuições de Weber consiste em mostrar que existem diferentes tipos de racionalidade e que a racionalização da própria religião exerceu um papel fundamental no surgimento da modernidade. Dois conceitos são importantes para entender a abordagem weberiana da religião: “agrupamento hierocrático” e “bens de redenção”. Por agrupamento hierocrático ele entende um grupo no interior do qual se exerce um modo particular de dominação sobre os homens. Em efeito, Weber inscreve sua sociologia das religiões no âmbito de uma sociologia da dominação. Ele observa de modo particular, os modos do exercício do poder religioso. Sociologia da religião: uma sociologia da dominação Assim, “um modo de agir em comunidade”, é uma forma de dominação sobre os homens. Weber coloca duas principais características da religião Civitas, Porto Alegre, v. 14, n. 3, p. 540-555, set.-dez. 2014 544 apreendida como fenômeno social: o vínculo social e o tipo de poder que ela gera. Desse fato, a sociologia da religião weberiana busca definir os tipos de “comunalização religiosa”, assim como os tipos de dominação religiosa. Os tipos de comunalização religiosa se apresentam na famosa distinção entre Igreja e “seita”, apreendidas como dois modos de existência social da religião. A igreja seria uma instituição burocrática de salvação aberta a todos, na qual é exercida a autoridade de função do padre, que coabita em perfeita simbiose com a sociedade global. A seita seria uma associação voluntária de crentes em ruptura mais ou menos marcada pelo ambiente social. Nesta última forma de associação prevalece uma autoridade religiosa do tipo carismático. Quanto aos tipos de autoridade religiosa, Weber os elaborou a partir da observação de diferentes formas de legitimação do poder na vida social. A dominação é acompanhada de uma forma legítima, cuja função é normalizar o que está estabelecido. Essa legitimidade é uma crença social que valida o poder exercido pelo dominante. Como sabemos, Weber distingue três tipos de dominação legítima: 1) a dominação tradicional, funda sua legitimidade, sua crença, no caráter sagrado da tradição. Por exemplo: o poder patriarcal no seio dos grupos domésticos, o poder dos senhores na sociedade feudal. 2) a dominação carismática procede de uma personalidade dotada de uma aura excepcional. O chefe carismático funda a crença no seu poder no caráter sagrado, na virtude heroica ou no valor exemplar de sua pessoa. 3) a dominação racional-legal, repousa sua crença na legalidade do direito abstrato e impessoal. O tipo mais exemplar desse tipo de dominação é a “dominação pela direção administrativo-burocrática”, cuja submissão é expressa num código (código civil), numa regra universal e funcional. No campo religioso esses modos de dominação legítima definiriam os tipos ideais do padre, do feiticeiro e do profeta. O padre representaria a autoridade religiosa de função exercida dentro de uma estrutura burocrática de salvação. O feiticeiro a autoridade religiosa que exerce suas competências como autêntico portador de uma tradição junto a uma clientela que o reconhece como tal. E o profeta, a autoridade religiosa pessoal daquele que impõe sua legitimidade por meio de uma revelação que o mesmo divulga. Sociologia da religião: uma sociologia da dominação Willaime (2012, p. 55) ressalva que essa tipologia das formas de autoridade religiosa exige uma utilização cuidadosa de seus termos; contudo, seu poder heurístico é grande e muitos sociólogos das religiões referem-se a ela. Ela serve para perceber que as análises de Weber ultrapassam as relações de um ethos do protestantismo puritano e o desenvolvimento de certa racionalidade econômica. J. L. Carvalho Filho – Religião, educação e economia em Max Weber 545 Religião e ética econômica No texto Introdução, referido anteriormente, Weber define o projeto de pesquisa e os conceitos que lhe servirão a elaborar seu estudo sobre a “ética econômica das religiões universais”. Ele formula, ainda, alguns de seus resultados provisórios. Por “religiões universais”, ele entende as cinco religiões e sistemas de regras de vida condicionadas pela religião que conseguiram reagrupar em torno de uma ideia religiosa ou de uma moral religiosa uma massa de fiéis: o confucionismo, o hinduísmo, o budismo, o cristianismo e o islamismo. Weber acrescenta o judaísmo, por conter “as condições históricas preliminares decisivas para o entendimento do cristianismo e do islamismo e pela sua significação histórica e autônoma para a evolução da moderna ética econômica do Ocidente” (Weber, 1996, p. 331), ainda que o povo judeu tenha sempre constituído uma minoridade, geralmente um povo de párias. Para o cristianismo ele remete aos seus trabalhos sobre a ética protestante. Convém sublinhar que Weber entende por “ética econômica” não a “teoria moral dos tratados das teologias, que pode servir somente de instrumento de conhecimento, mas as incitações práticas à ação enraizadas nas articulações psicológicas e pragmáticas das religiões” (Weber, 1996, p. 332). Dito de outra maneira, não se trata de expor a teologia moral dessas religiões, mas de compreender os encadeamentos psicológicos e pragmáticos que exerceram um papel importante nas motivações praticas das atividades em geral e na economia, em particular. Para compreender a ética econômica assim entendida, Weber examina a determinação religiosa do modo de vida das camadas sociais, cujos estilos de vida foram pelo menos predominantemente decisivos para certas religiões (Weber, 1996, p. 333). No entanto, ele afirma que “a ética econômica” não é uma simples “função” de uma forma de organização econômica; e a recíproca também não é verdadeira, ou seja, a ética econômica não marca sem ambiguidade a forma de organização econômica. Assim, o que Weber deixa entender é que os fatores que determinam a conduta de vida é um elemento entre tantos da ética econômica. Em contrapartida ele afirma que o modo de vida determinado religiosamente é, em si, profundamente influenciado pelos fatores econômicos e políticos que operam dentro de determinados limites geográficos, políticos, sociais e nacionais. Iríamos perder-nos nessas discussões se tentássemos demonstrar essas dependências em toda a singularidade. Só podemos no caso, tentar retirar os elementos diretivos na conduta de vida das camadas sociais que influenciaram mais fortemente a ética prática de suas respectivas religiões. o modo de vida determinado religiosamente é, em si, profundamente influenciado pelos fatores econômicos e políticos que operam dentro de determinados limites geográficos, políticos, sociais e nacionais. Iríamos perder-nos nessas discussões se tentássemos demonstrar essas dependências em toda a singularidade. Só podemos no caso, tentar retirar os elementos diretivos na conduta de vida das camadas sociais que influenciaram mais fortemente a ética prática de suas respectivas religiões. Esses elementos marcaram os aspectos mais Religião e ética econômica Esses elementos marcaram os aspectos mais 546 Civitas, Porto Alegre, v. 14, n. 3, p. 540-555, set.-dez. 2014 característicos da ética prática, as características que distinguem uma ética das outras; e, ao mesmo tempo, foram importantes para a respectiva ética econômica (Weber, 1996, p. 333). Assim, na Introdução, Weber declina as condutas de vida das grandes religiões do mundo: no confucionismo a ética de uma camada de prebendários letrados seculares e mandarins; aquele que não pertence a essa camada culta não conta. A ética do corpo dessa camada determinou a conduta de vida chinesa; o hinduísmo era esposado por uma casta hereditária de letrados cultos que, além de qualquer cargo oficial, exerceram uma espécie de culto ritualista das almas para indivíduos e comunidade. Essa ética religiosa desta camada, segundo ele, determinou o modo de vida chinês, muito além da própria camada. O budismo se propagou através dos monges mendicantes e itinerantes, estritamente contemplativos e recusando o mundo. Os brâmanes, educados no Veda, formavam, como portadores da tradição, o estamento religioso plenamente aceito. O islamismo foi, desde suas origens, a religião de guerreiros que queriam conquistar o mundo, uma ordem cavalheiresca de cruzados disciplinados. O judaísmo, após o Exílio, foi a religião de um “povo pária” cívico, e na Idade Média, a religião de intelectuais treinados na literatura e ritual, característica particular do judaísmo. Essa camada representava uma intelligentsia de “pequenos burgueses” racionalistas. O cristianismo, finalmente, começou sua carreira como doutrina de pobres artesãos jornaleiros itinerantes, mas em todos os períodos de sua formação foi sempre uma religião particularmente urbana e essencialmente burguesa. Após essa descrição, que não chega a ser uma “tipologia” sistemática da religião, Weber se preocupa em levar em conta as lógicas intrínsecas das diversas esferas de atividades: econômicas, políticas, religiosa, estética, e de se respeitar sua coerência interna de um ponto de vista ideal-típico. É nesse sentido que devemos ler suas considerações sobre as relações da religião com as esferas econômica, artística, erótica e científica. Assim, como observa Giddens (2011, p. 233): “a relação entre o conteúdo das crenças religiosas e as formas das atividades econômicas características de uma dada ordem social são muitas vezes indiretas, sofrendo a influência de outras instituições existentes no seio dessa mesma ordem”. Religião e ética econômica De todo modo, é justo afirmar que o ponto de partida dos estudos weberianos de sociologia das religiões é fundado na seguinte questão: em que medida as concepções religiosas influenciaram o comportamento econômico das diversas sociedades? Weber quis demonstrar que as condutas dos homens das diversas sociedades não são inteligíveis senão no quadro da concepção 547 J. L. Carvalho Filho – Religião, educação e economia em Max Weber geral que esses homens fizeram de suas existências; os dogmas religiosos e sua interpretação são partes constitutivas dessa visão de mundo. É preciso compreendê-los para compreender o comportamento dos indivíduos, dos seus grupos sociais e, notadamente, suas atividades econômicas. Por conseguinte, Weber quis provar que as concepções religiosas revelam, efetivamente, determinadas condutas econômicas, uma das causas de transformações econômicas das sociedades (Aron, 1988, p. 530). No Prefácio que Weber colocou na abertura da Coletânea de estudos de sociologia das religiões (1920-1921)1 ele precisa, desde o início, uma das questões fundamentais de sua obra: [...] qual encadeamento de circunstâncias levou a que o Ocidente precisamente, e apenas ele, tenha visto aparecer sobre seu solo fenô- menos culturais que se inscreveram em uma direção de desen- volvimento, que revestiu –pelo menos gostamos de assim pensar– uma significação e uma validade universais (Weber, 1996, p. 489). É no processo sistemático de racionalização da conduta que Weber identifica a especificidade ocidental, e isso em diferentes domínios: a economia, as ciências, direito, a arte, a arquitetura, o estado. Essa noção de racionalidade, que recobre ao mesmo tempo a progressão de uma eficácia (utilizar sistematicamente meios adaptados para obter um fim), uma progressão da coerência lógica dos discursos (racionalidade dos meios intelectuais ou da ciência), uma progressão da racionalidade ética (redação dos mandamentos divinos do judaísmo antigo) que acompanha ao processo de desmagificação do mundo, Weber aplicou, em primeiro lugar à economia. Assim, ele postulava que se existem diferentes formas de capitalismo, é apenas no Ocidente que ele deu o surgimento a um tipo de homem especializado, tendo uma formação específica: empreendedor burguês, definido por certa relação com o trabalho e pela racionalidade de sua atividade econômica (Weber, 1996, p. 492). 1 Trata-se de estudos sobre a ética protestante (Ética protestante e o espírito do capitalismo; e As seitas protestantes e o espírito do capitalismo). Religião e ética econômica Isso posto, Weber reconhece que o ponto mais difícil a apreender é: como certos conteúdos de crenças religiosas condicionaram a gênese de uma “mentalidade econômica”, ou seja, o ethos de uma forma de economia, e isso tomando como exemplo relações do ethos econômico moderno com a ética racional do protestantismo ascético” (Weber, 1996, p. 504). Como ele mesmo afirma, ele se ocupa apenas de uma variante da relação causal. Os estudos sobre a Ética econômica das religiões mundiais, percorrendo as relações das Civitas, Porto Alegre, v. 14, n. 3, p. 540-555, set.-dez. 2014 548 religiões de civilização mais importantes com a economia e a estrutura social que lhes cercam, tentam seguir as duas causalidades o mais longe possível, numa perspectiva histórica, a fim de encontrar os pontos de comparação com o desenvolvimento ocidental (Weber, 1996, p. 504). Nas Considerações intermediárias que abre a coletânea de estudos de sociologia das religiões, Weber desenvolve uma “teoria dos graus e orientações da rejeição do mundo”. Utilizando o procedimento ideal-típico ele se esforça para esclarecer a interrogação “sobre aquilo que motivou o nascimento e o desenvolvimento das éticas religiosas da rejeição do mundo e sobre as direções que elas tomaram; e, por conseguinte, sobre o ‘sentido’ possível dessa negação” (Weber, 1996, p. 411). Construindo adequadamente os tipos racionais, ele esperava dar uma “contribuição à tipologia e à sociologia do racionalismo”. Como ele afirma: O racional, entendido no sentido de uma coerência lógica ou teleológica de uma tomada de posição teórico-intelectual ou ético- prática, exerce realmente um poder sobre os homens por mais limitado e instável que esse poder seja e tenha sido frente a outras forças da vida histórica (Weber, 1996, p. 412). Para Weber as interpretações religiosas que foram criadas pelos intelectuais tendo como alvo a racionalidade, foram fortemente submetidas ao imperativo da coerência (Weber, 1996, p. 412). Em Considerações Weber se dedica, também, a esclarecer a noção de “ascese” e de “mística” que lhe permitem desenvolver uma tipologia da “rejeição ao mundo”. Ele distingue, assim, uma ascese aplicada (ascese intramundana) ao mundo e uma ascese fugitiva ao mundo. Ou seja, uma mística orientada para o mundo e uma mística fugaz ao mundo (Weber, 1996, p. 414). Após esta descrição, Weber expõe o problema da tensão entre a religião e as outras atividades, que o ocupou mais do que qualquer outro problema. Religião e ética econômica As relações de tensão opõem: a religião e o mundo; a religião e a esfera econômica, a religião e o sistema político; a religião e a sexualidade; e a religião e o império do conhecimento intelectual. Para o propósito deste artigo, nos referimos aqui especificamente à tensão com a economia e o domínio do conhecimento. As tensões com a economia, a segunda tensão apresentada no texto, tomaram diversas formas: oposição aos juros e à usura, defesa da esmola e da vida reduzida às necessidades estritas, hostilidade do comércio que não poderia agradar a Deus (Weber, 1996, p. 422-423). Segundo Weber, 549 J. L. Carvalho Filho – Religião, educação e economia em Max Weber Toda operação primitiva de influência sobre os espíritos e sobre as divindades, que elas fossem mágicas ou mistagógicas, tiveram interesses especiais. Lutaram pela riqueza, bem como pela vida, saúde, honra, descendência e, possivelmente, melhorias do destino no outro mundo (Weber, 1996, p. 420). Toda operação primitiva de influência sobre os espíritos e sobre as divindades, que elas fossem mágicas ou mistagógicas, tiveram interesses especiais. Lutaram pela riqueza, bem como pela vida, saúde, honra, descendência e, possivelmente, melhorias do destino no outro mundo (Weber, 1996, p. 420). Weber declina alguns exemplos: foi o caso das religiões fenícias e védicas, da religião popular chinesa, do judaísmo e do islamismo antigos. Tal promessa foi também ofertada aos leigos hindu e budista. Em contrapartida, as religiões sublimadas da salvação tiveram relações cada vez mais tensas com a economia racionalizada (Weber, 1996, p. 421). Como se vê, a rejeição da atividade econômica é marcada de uma parte pela religião institucional, pelo fato de que toda organização institucional necessita de meios materiais e econômicos, o que pode influenciar por sua vez a religiosidade. Dito de outra forma, os conflitos da ética religiosa da fraternidade com a economia impessoal não impede, como Weber afirmava, [...] o paradoxo de toda ascese racional, ou seja, que ela própria cria a riqueza que ela rejeitava, estendeu a mesma armadilha para os monges de todos os tempos. Em todo lugar, templos e claustros se tornaram eles próprios, por sua vez, lugares de economia racional por excelência (Weber, 1996, p. 422). [...] o paradoxo de toda ascese racional, ou seja, que ela própria cria a riqueza que ela rejeitava, estendeu a mesma armadilha para os monges de todos os tempos. Religião e ética econômica Em todo lugar, templos e claustros se tornaram eles próprios, por sua vez, lugares de economia racional por excelência (Weber, 1996, p. 422). E a economia monacal, atesta Weber, é uma economia racional por excelência. Como o ascetismo recai sempre na contradição que faz com que seu caráter racional o leve à acumulação das riquezas, ele explica que a piedade calvinista puritana valorizou religiosamente a atividade econômica e o sucesso material, contribuindo desse modo, defendendo-se a si mesma, para o desenvolvimento de uma lógica econômica que iria dissolver as motivações religiosas que haviam contribuído para fazê-lo surgir. A outra orientação de rejeição do mundo nas diversas esferas que interessa aqui para o nosso propósito (a última da análise weberiana) é o domínio do conhecimento reflexivo. É com o intelecto que a religião entra em maior tensão. Aí também se pode constatar que em certos períodos da Historia os padres foram os agentes e mesmo os criadores da cultura, quer por constituírem o elemento capaz de ler e escrever, de que precisavam os chefes políticos, quer por deterem uma espécie de monopólio pedagógico. Observa Weber, que quanto mais uma religião abandona seus aspectos mágicos e místicos para se tornar uma “doutrina”, mais se desenvolve sob a forma de um conhecimento teológico ou apologético. Entretanto, por sua vez, uma religião acomodada a Civitas, Porto Alegre, v. 14, n. 3, p. 540-555, set.-dez. 2014 550 um tipo de conhecimento especulativo puramente metafísico, não é raro que ela considere a pesquisa empírica, notadamente com base nas ciências da natureza, como sendo mais compatível com seus interesses que a filosofia. Ou seja, a religião podia conciliar-se com a metafísica; mas dificilmente pode encontrar um ponto de entendimento com as disciplinas que desencantam o mundo, não somente porque estas se desinteressam em geral do problema da significação, mas porque elas dão origem a uma técnica puramente mecânica e a uma consciência racional dos problemas, de sorte que a religião se vê cada vez mais relegada entre as forças irracionais ou antirracionais, que exigem o “sacrifício do intelecto”. As religiões tentaram encontrar o ponto de encontro, afirmando que o conhecimento que lhe é próprio se situa numa outra esfera que o conhecimento puramente científico e que se baseiam muito mais na intuição ou na iluminação carismática do que no raciocínio. Religião e ética econômica Isso não impede que o problema da teodiceia (que muito preocupou Weber e do qual ele trata no final do texto) se apresente sob nova roupagem com o desenvolvimento da cultura no sentido de um progresso e de um aperfeiçoamento da humanidade (Weber, 1996, p. 448). Pode se citar o trabalho de Guy Vincent (2009) e de Nildo Viana (2004). 3 Exemplo do texto Os letrados chineses, que figura no capítulo 16 da coletânea de texto de Max Weber, publicado no Brasil como Ensaios de Sociologia, originalmente publicada em inglês e com organização e introdução de H. H. Gerth e C. Wright Mills. 2 Pode-se citar o trabalho de Guy Vincent (2009) e de Nildo Viana (2004).i 4 Em português esse capítulo é traduzido por Letrados chineses. Conserva-se aqui a expressão corps de lettrés, da tradução científica francesa de Confucionismo e taoísmo de Catherine Colliot-Thélène em colaboração com Jean-Pierre Grossein, por indicar a constituição de um grupo social, marcado pelo “espírito” do corpo, uma disposição, uma conduta de vida, um habitus, um ethos. Palavras evocadas que perpassam a obra de Weber, no que concerne à educação religiosa e à cultura capitalista. A religião e tipos sociológicos da educação Os trabalhos no campo da sociologia da educação explorando as análises de Weber são ainda muito restritos.2 Os que aparecem nesse sentido são de cunho exegético que tentam garimpar nas obras do autor as passagens e a articulação com os grandes temas presentes na sua obra. É importante lembrar que na Reproduction: éléments pour une théorie du système d’enseignement Pierre Bourdieu e Jean-Claude Passeron, em 1970, já se referem às análises weberianas da cultura chinesa, em particular à cultura de letrados, especificamente ao estudo sobre o Confucionismo e o taoísmo, texto no qual Weber expõe os tipos de educação. O interesse de Bourdieu e Passeron é o de construir uma teoria geral dos campos como “estrutura da relação objetiva”, estabelecendo uma homologia “estruturais e funcionais”. Essa escassez provavelmente se deve primeiro ao fato do autor não ter escrito um texto específico sobre a educação. Segundo, porque, um dos textos em que ele define de maneira breve, mas pertinente, os grandes tipos de educação, Confucionismo e taoísmo, além de mal conhecido, é geralmente desvinculado do seu projeto intelectual e com capítulos separados.3 551 J. L. Carvalho Filho – Religião, educação e economia em Max Weber Para compreender as referências de Weber sobre a educação, é importante situar esse objeto no conjunto do seu projeto de uma sociologia histórica e comparada das religiões do mundo e as éticas protestantes e o espírito de capitalismo. Embora não se possa deduzir dos tipos de dominação legítima os tipos de educação, não foi isso que fez Weber; nem pretender que seu propósito fosse estabelecer uma “tipologia sociológica dos objetivos e dos meios pedagógicos”, coisa que ele mesmo se preocupou em ressalvar (Weber, 2000, p. 177); sua definição da educação e seus tipos nos ajudam a compreender a teoria da educação esboçada por ele. O estudo de Weber sobre o Confucionismo e taoísmo faz parte dos três grandes estudos comparativos de sociologia das religiões que ele publicou a partir de 1915 sob o título geral de Ética econômica das religiões mundiais. Os quatro primeiros capítulos de Confucionismo e taoísmo são dedicados às bases sociológicas da China: a cidade, o príncipe e deus; Estado feudal e estado prebendário; Administração e regime agrário; Auto-administração, direito e ausência de relações capitalista. A religião e tipos sociológicos da educação O quinto capítulo Corps de lettrés,4 apresenta a longa formação de um sistema de exames, a caráter literário, que dão unicamente acesso ao corpo dos mandarins, ou seja, à classe dos funcionários civis. Como Weber destaca: Na China, depois de mais de vinte séculos, a posição social foi determinada pela qualificação às funções oficiais mais do que pela posse; uma qualificação que foi estabelecida através de uma formação (Bildung) e, em particular, sancionada pelos exames. A China foi o país que fez da formação literária o critério de estima social, de maneira mais exclusiva do que a Europa na época humanista ou do que, em último lugar, a Alemanha (Weber, 2000, p. 161). Assim, na China, a classe de letrados sempre foi “a classe dominante”. O imperador se dirige a eles como “Senhores”. Sua formação é laica, dife- rentemente do que se passou na Índia (Ibid., p. 162). Como relembra o sociólogo francês Guy Vincent (2009), foi para explicar o que é esse sistema de formação levando em conta seu método de cons- 552 Civitas, Porto Alegre, v. 14, n. 3, p. 540-555, set.-dez. 2014 trução de tipos ideais, que Weber julga indispensável uma tipologia da educação. Ele os caracteriza da seguinte forma: trução de tipos ideais, que Weber julga indispensável uma tipologia da educação. Ele os caracteriza da seguinte forma: No domínio dos objetivos da educação os dois polos extremos historicamente foram os seguintes: de um lado, o despertar de um carisma (quer se trate de qualidades heroicas ou de dons mágicos); de outro, a transmissão de uma instrução especializada. O primeiro corresponde à estrutura carismática, o segundo à estrutura burocrático-racional (moderna) de dominação. Os dois não estão sem conexão ou transição entre si. O herói guerreiro ou o mágico havia igualmente necessidade de uma instrução especializada. [...] Entre esses contrários radicais, encontramos os tipos de educação que querem cultivar nos alunos um tipo determinado de conduta de vida, que ela seja laica ou clerical. Em todos os casos, uma conduta de vida corresponde a um estatuto (ständisch) (Weber, 2000, p. 177). Apesar de sermos tentados a aplicar essa tipologia aos tipos “puros” de dominação legítima, não foi o que fez Weber. A religião e tipos sociológicos da educação Guy Vincent argumenta que, na Introdução quando Weber (1996) retoma o que havia esboçado em Economia e sociedade e define a dominação (autoridade) carismática, a dominação (autoridade) tradicional, e enfim, a dominação racional-legal, ele indica, a cada vez, exemplo de relação de dominação legítima: o profeta, o chefe de guerra para o primeiro; o pai de família, o marido, o príncipe e seus ministros para o segundo; o agente de um agrupamento burocrático e o cidadão para o terceiro. No entanto, não encontramos o mestre (professor) e o aluno. Nesse sentido, quando Weber se refere à educação, esse problema está sempre ligado às suas análises sobre os agrupamentos de poder político ou hierocrático ou de um sistema econômico (Vincent, 2009, p. 77). Essa consideração serve para evitarmos o atalho apressado da relação entre escola e democracia e a abstração da educação, da autoridade e do poder pedagógico. Vincent alerta para o fato de que a estrutura da tipologia de Weber elaborada de três termos (dois opostos e um intermediário) é histórica, apesar de ser geral. Ou seja, ela engloba todos os sistemas de educação que se pode observar na história, o que é confirmada nos inúmeros exemplos declinados por Weber em Confucionismo e taoísmo. E mais, a tipologia da educação é dita sociológica, na medida em que recobrindo os diferentes aspectos da atividade social (politica, econômica etc.) ela é colocada em relação com os modos de dominação (Vincent, 2009, p. 77-78). Para Weber, os objetivos de cada tipo de educação devem corresponder a partir dos elementos essenciais desses tipos de dominação. Mas, os tipos de educação são construídos de objetivos e de meios. O exame dos meios mise 553 J. L. Carvalho Filho – Religião, educação e economia em Max Weber en œuvre em cada tipo de educação permite a Weber melhor caracterizar e detalhar a análise das escolas na China confucianista ou da escola alemã (Vincent, 2009, p. 78). Em Confucionismo e taoísmo (p. 177-178), encontramos um termo chave, ou um conceito, para caracterizar o objetivo de cada tipo de educação: “despertar”, “educar”, “dar uma educação, uma cultura”. O exercício carismático do ascetismo mágico antigo e os julgamentos dos heróis que os feiticeiros e heróis guerreiros aplicavam aos rapazes tentava ajudar o noviço a adquirir uma “nova alma”, no sentido animista, e, portanto, a renascer. A religião e tipos sociológicos da educação As tentativas de educação especializada visam a treinar os alunos para que eles adquiram uma aptidão prática às atividades administrativas, quer ela seja no exercício da magistratura, da condução de um escritório, de um atelier, de um laboratório científico ou industrial ou, ainda, no exército disciplinado. Quanto à pedagogia do cultivo (Kultivationspäedagogik), busca educar um tipo de homem culto (Kulturmensch), diferente segundo o ideal da cultura da camada dominante. Isso significa dizer: um homem que tenha uma conduta de vida interior e exterior determinada. A China confucianista abandona a educação carismática em proveito da aquisição de uma qualificação. Esta consiste na posse de uma cultura literária aprofundada e do modo de pensamento que convém ao homem de qualidade; essa qualidade é mais específica que a conferida aos alunos, no Ocidente, na formação humanista. Da mesma forma que no Ocidente, ela condiciona o acesso às funções dotadas de um poder de comando na administração civil e militar. Ela é controlada e monopolizada pelo poder político, mas ela confere ao mesmo tempo aos alunos a marca de seu pertencimento social ao corpo de “homens cultivados”. Isso dito, não havia na China o treinamento especializado e racional que veio a ser completado e, em parte, substituir essa qualificação cultural no Ocidente (Weber, 2000, p. 178-179). Essa espécie de qualificação é adquirida na escola. O sistema de formação comporta três níveis, que por analogia podem ser chamados de primário, secundário e nível superior. A todos os níveis correspondem exames de redação, estilo e domínios das obras dos autores clássicos. Desde o nível médio são exigidas dissertações que demandam capacidade de análises filológica e erudita de um texto. No nível superior, a maior parte dos sujeitos se vincula às questões do imperador que tratam da administração, mas não no mesmo sentido estrito da técnica como a nossa. Os textos objetos dos exames são textos clássicos. É bom lembrar que Confucio é um letrado. A palavra chinesa que traduzimos por confucionismo significa “doutrina dos letrados”. Weber precisa que não existe, na China, uma “religião” autônoma, de racionalização de uma crença popular suscetível de constituir um poder autônomo. Não existe, por consequência, uma Civitas, Porto Alegre, v. 14, n. 3, p. 540-555, set.-dez. 2014 554 hierocracia suscetível de concorrer à burocracia. Os textos que estudam aqueles chamados de alunos não são textos sagrados. A religião e tipos sociológicos da educação A filosofia de Confucio é uma sabedoria cuja preocupação principal foi fazer reinar a ordem no estado, formando homens virtuosos. No capítulo que segue (cap. 6), sobre a Orientação confucianista da vida, Weber acentua o caráter intramundano do confucionismo e a ausência de metafísica nessa doutrina. Essa doutrina intramundana se opõe às religiões da salvação para o “au-delà du monde”, mesmo quando se trata do puritanismo. Weber sublinha, ainda, as diferenças da compreensão da educação na Grécia antiga e no puritanismo. Com isso ele pretende demonstrar que o estudo dos textos antigos canonizados, a relação com os textos escritos como foi praticado na educação confucianista impede de pensar por si mesmo. A razão própria do confucionismo diz respeito a um racionalismo da ordem. ARON, Raymond. Les étapes de la pensée sociologique. Paris: Gallimard, 1998. BOURDIEU, Pierre; PASSERON, Jean-Claude. La reproduction: éléments pour une théorie du système d’enseignement. Paris: Minuit, 1999. Considerações finais Os temas religião, educação (expressa nas fórmulas habitus, disposição, conduta de vida) e economia se encontram com muita frequência no conjunto da obra de Weber, no projeto de elaboração de uma sociologia comparada das religiões presente nos estudos históricos da ética econômica das religiões mundiais, que se apresenta como continuação do estudo consagrado à ética protestante. Como observa Colliot-Thélène (2006, p. 85-86) habitus, “espírito”, disposição, são empregados de modo idêntico; designam geralmente o substrato mental das condutas sociais, pelo qual é preciso entender a disposição a se comportar de maneira determinada. Weber explica que o termo “espírito” no “espírito do capitalismo” tem a mesma significação.i Seu objetivo era justificar um dos aspectos da relação causal, a saber, as maneiras pelas quais certas crenças religiosas determinadas condicionam o ethos de uma forma econômica. Em relação a uma maior proporção de protestantes detentores de capitais, ele reconhece explicitamente que é por razões históricas junto às quais o pertencimento confessional aparece como causa primeira dos fenômenos econômicos, mas, até certo ponto, como sua consequência. O propósito de Weber não era mostrar que o protestantismo engendrou o capitalismo moderno, mas demonstrar as afinidades existentes entre um certo tipo de protestantismo, notadamente o calvinismo puritano e o espírito empresarial. Referências ARON, Raymond. Les étapes de la pensée sociologique. Paris: Gallimard, 1998. BOURDIEU, Pierre; PASSERON, Jean-Claude. La reproduction: éléments pour une théorie du système d’enseignement. Paris: Minuit, 1999. 555 J. L. Carvalho Filho – Religião, educação e economia em Max Weber CALLIOT-THÉLÈNE, Catherine. La sociologie de Max Weber. Paris: La Découverte, 2006. GIDDENS, Anthony. Capitalismo e moderna teoria social. Lisboa: Editorial Presença, 2011. GROSSEIN, Jean-Pierre. Présentation. In: Max Weber. Sociologie des religion. Gallimard: Paris, 2000a. p. 51-129. GROSSEIN, Jean-Pierre. Présentation. In: Max Weber. Confucianisme et taoïsme. Gallimard: Paris, 2000b. p. I-XXV. PASSERON, Jean Claude. Introduction à Max Weber. In: Max Weber. Sociologie des religions. Paris: Gallimard, 1996. p. 1-49. VIANA, Nildo. Weber: tipos de educação e educação burocrática. Guanicus, v. 1, p. 117-132, 2004. VINCENT, Guy. Les types sociologiques d’éducation selon Max Weber. Revue Française de Pédagogie, v. 168, p. 75-82, 2009 <http://rfp.revues.org/1755#text> (5 jan. 2014). WEBER, Max. Confucianisme et taoïsme. Paris: Gallimard, 2000. WEBER, Max. Sociologie des religions. Paris: Gallimard, 1996. WEBER, Max. L’Étique protestante et l’“esprit” du capitalisme: suivi d’autres essais. Paris: Gallimard, 2003. WEBER, Max. Ensaios de sociologia. Rio de Janeiro: Guanabara, 1982. WEBER, Max. Économie e société. v. 1. Paris: Plon, 1995. WILLAIME, Jean-Paul. Sociologia das religiões. São Paulo: Unesp, 2012. Autor correspondente: Juarez Lopes de Carvalho Filho Av. dos Portugueses s/nº, bloco 6, sala 3 Campus do Bacanga 65085-580 São Luis, MA, Brasil Recebido em: 24 abr. 2014 Aprovado em: 26 jun. 2014
https://openalex.org/W4385556750	https://www.mdpi.com/2073-4352/13/8/1208/pdf?version=1691047464	English	null	Effects of Tetrafluorocyclohexa-1,3-Diene Ring Position on Photoluminescence and Liquid-Crystalline Properties of Tricyclic π-Conjugated Molecules	Crystals	2,023	cc-by	13,054	crystals crystals Citation: Ohsato, H.; Yamada, S.; Yasui, M.; Konno, T. Effects of Tetraﬂuorocyclohexa-1,3-Diene Ring Position on Photoluminescence and Liquid-Crystalline Properties of Tricyclic π-Conjugated Molecules. Crystals 2023, 13, 1208. https:// doi.org/10.3390/cryst13081208 Academic Editor: Ingo Dierking Received: 10 July 2023 Revised: 28 July 2023 Accepted: 29 July 2023 Published: 3 August 2023 Citation: Ohsato, H.; Yamada, S.; Yasui, M.; Konno, T. Effects of Tetraﬂuorocyclohexa-1,3-Diene Ring Position on Photoluminescence and Liquid-Crystalline Properties of Tricyclic π-Conjugated Molecules. Crystals 2023, 13, 1208. https:// doi.org/10.3390/cryst13081208 Keywords: ﬂuorine; tetraﬂuorocyclohexa-1,3-diene; photoluminescence; liquid crystal; tricyclic molecule; aggregation Haruka Ohsato, Shigeyuki Yamada * , Motohiro Yasui and Tsutomu Konno * Faculty of Molecular Chemistry and Engineering, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan; m1673006@edu.kit.ac.jp (H.O.); myasui@kit.ac.jp (M.Y.) * Correspondence: syamada@kit.ac.jp (S.Y.); konno@kit.ac.jp (T.K.) Abstract: Tetraﬂuorocyclohexa-1,3-diene ring-containing tricyclic π-conjugated molecules are promis- ing negative-dielectric-anisotropy guest species for vertical-alignment-type liquid-crystalline (LC) displays. Building on our previous work reporting the excellent photoluminescence (PL) properties of tricyclic π-conjugated molecules with central tetraﬂuorocyclohexa-1,3-diene rings, we herein synthe- sized four analogous molecules with terminal tetraﬂuorocyclohexa-1,3-diene rings from commercially available precursors and investigated the effects of substituent type and diene ring position on PL and LC properties using microscopic and spectroscopic methods. One of the prepared molecules exhibited a relatively planar molecular structure and formed herringbone-type aggregates via π/F and CH/π interactions instead of forming stacked aggregates via π/π stacking interactions, thus exhibiting relatively strong PL in solution and crystalline states. Moreover, the PL color of this compound depended on the electronic character of its terminal substituents along the long molecular axis. Of the four prepared species, two featured terminal ethyl groups and formed one or more LC phases. The PL properties of these phases indicated that the related phase transition induced changes in the aggregate structure, PL wavelength, and PL color. Our results expand the applicability of CF2CF2 moiety-containing tricyclic compounds as functional molecules for the fabrication of next-generation PL, LC, and PL-LC materials. g y 2. Materials and Methods 2 1 General Characterization 2.1. General Characterization 2. Materials and Methods 2.1. General Characterization Melting points (Tm) were measured on a Shimadzu DSC-60 Plus instrument using least three heating/cooling cycles at a scan rate of 5.0 °C·min–1. 1H and 13C nuclear magne resonance (NMR) spectra were recorded on a Bruker AVANCE III 400 spectrometer ( 400.13 MHz, 13C: 100.61 MHz) in chloroform-d (CDCl3). Chemical shifts were reported the basis of the residual proton or carbon signal of CHCl3 (δH = 7.26 ppm, δC = 77 ppm parts per million (ppm). 19F-NMR (376.46 MHz) spectra were recorded on the Bru AVANCE III 400 spectrometer in CDCl3 using trichlorofluoromethane (CFCl3, δF = 0 ppm) as an internal standard. Infrared (IR) spectra were acquired using the KBr meth on a JASCO FT/IR-4100 type A spectrometer. High-resolution mass spectra (HRMS) w recorded on a JEOL JMS-700MS spectrometer using fast atom bombardment (FAB+) me 2.1. General Characterization Melting points (Tm) were measured on a Shimadzu DSC-60 Plus instrument using least three heating/cooling cycles at a scan rate of 5.0 °C·min–1. 1H and 13C nuclear magnet resonance (NMR) spectra were recorded on a Bruker AVANCE III 400 spectrometer (1H 400.13 MHz, 13C: 100.61 MHz) in chloroform-d (CDCl3). Chemical shifts were reported o the basis of the residual proton or carbon signal of CHCl3 (δH = 7.26 ppm, δC = 77 ppm) parts per million (ppm). 19F-NMR (376.46 MHz) spectra were recorded on the Bruk AVANCE III 400 spectrometer in CDCl3 using trichlorofluoromethane (CFCl3, δF = 0.0 ppm) as an internal standard. Infrared (IR) spectra were acquired using the KBr metho on a JASCO FT/IR-4100 type A spectrometer. High-resolution mass spectra (HRMS) we recorded on a JEOL JMS-700MS spectrometer using fast atom bombardment (FAB+) meth ods. Column chromatography was performed using Wakogel® 60N (38–100 μm), and thin layer chromatography was performed using the corresponding silica gel plates (silica g Melting points (Tm) were measured on a Shimadzu DSC-60 Plus instrument using at least three heating/cooling cycles at a scan rate of 5.0 ◦C·min−1. 1H and 13C nuclear magnetic resonance (NMR) spectra were recorded on a Bruker AVANCE III 400 spectrom- eter (1H: 400.13 MHz, 13C: 100.61 MHz) in chloroform-d (CDCl3). Chemical shifts were reported on the basis of the residual proton or carbon signal of CHCl3 (δH = 7.26 ppm, δC = 77 ppm) in parts per million (ppm). 1. Introduction Fluorinated organic molecules have drawn much attention as the structural compo- nents of pharmaceuticals [1,2] and agrichemicals [3,4], as well as major constituents of liquid crystals [5–7] and optoelectronic materials [8,9]. This popularity is due to the unique properties of ﬂuorine [10], namely, its highest electronegativity among all elements (4.0 on the Pauling scale), second smallest atomic radius (147 pm according to Bondi [11]), and the high dissociation energy of C–F bonds (105.4 kcal·mol−1). In view of these properties, the introduction of ﬂuorine into molecular structures enhances latent functions or promotes the emergence of new ones and is, therefore, a powerful approach for the development of novel organic functional materials. Academic Editor: Ingo Dierking Received: 10 July 2023 Revised: 28 July 2023 Accepted: 29 July 2023 Published: 3 August 2023 g Our group has developed efﬁcient and selective synthetic routes to various ﬂuorinated organic molecules [12,13] including those exhibiting photoluminescence (PL) and liquid- crystalline (LC) properties [14]. The results obtained so far indicate that the introduction of ﬂuorine atoms substantially increases PL intensity in the solid state and induces the emergence of mesophases between crystalline (Cry) and isotropic (Iso) phases. Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Previously, we prepared a tricyclic molecule with a central tetraﬂuorocyclohexa-1,3- diene ring (1a) as a guest molecule with negative dielectric anisotropy to develop vertical alignment-type LC materials [15–18] and showed that 1a exhibits blue PL in the crystalline https://www.mdpi.com/journal/crystals Crystals 2023, 13, 1208. https://doi.org/10.3390/cryst13081208 Crystals 2023, 13, 1208 2 of 16 t lli and solution states. On this basis, we synthesized analogous tricyclic molecules with controlled electron density along the long molecular axis (1b and 1c) and revealed that their PL behavior is greatly affected by the electron density distribution, which, in turn, is inﬂuenced by the electronic properties of terminal substituents (Figure 1) [19]. and solution states. On this basis, we synthesized analogous tricyclic molecules with c trolled electron density along the long molecular axis (1b and 1c) and revealed that th PL behavior is greatly affected by the electron density distribution, which, in turn, is fluenced by the electronic properties of terminal substituents (Figure 1) [19]. 1. Introduction trolled electron density along the long molecular axis (1b and 1c) and revealed that the PL behavior is greatly affected by the electron density distribution, which, in turn, is i fluenced by the electronic properties of terminal substituents (Figure 1) [19]. ic molecules 1a–c with central tetrafluorocyclohexa Figure 1. Previously synthesized tricyclic molecules 1a–c with central tetrafluorocyclohexa di i Figure 1. Previously synthesized tricyclic molecules 1a–c with central tetrafluorocyclohexa-1,3-diene rings. Figure 1. Previously synthesized tricyclic molecules 1a–c with central tetrafluorocyclohexa- diene rings. Figure 1. Previously synthesized tricyclic molecules 1a–c with central tetrafluorocyclohexa di i Figure 1. Previously synthesized tricyclic molecules 1a–c with central tetrafluorocyclohexa-1,3-diene rings. diene rings. Figure 1. Previously synthesized tricyclic molecules 1a–c with central tetrafluorocyclohexa- di i Figure 1. Previously synthesized tricyclic molecules 1a–c with central tetrafluorocyclohexa-1,3-diene rings. diene rings. diene rings. Building on the abovementioned results, we herein synthesized and characteri molecules 2a–d to examine how PL and LC properties are affected by the position of tetrafluorocyclohexa-1,3-diene ring in the tricyclic structure and the electronic proper Building on the abovementioned results, we herein synthesized and characterized molecules 2a–d to examine how PL and LC properties are affected by the position of the tetraﬂuorocyclohexa-1,3-diene ring in the tricyclic structure and the electronic properties of terminal substituents (Figure 2) [15,18]. Building on the abovementioned results, we herein synthesized and characterize molecules 2a–d to examine how PL and LC properties are affected by the position of th tetrafluorocyclohexa-1,3-diene ring in the tricyclic structure and the electronic properti of terminal substituents (Figure 2) [15,18]. l y g y p p of terminal substituents (Figure 2) [15,18]. Figure 2. Structures of tricyclic molecules 2a–d with terminal tetrafluorocyclohexa-1,3-diene rin Figure 2. Structures of tricyclic molecules 2a–d with terminal tetraﬂuorocyclohexa-1,3-diene rings. Figure 2. Structures of tricyclic molecules 2a–d with terminal tetrafluorocyclohexa-1,3-diene ring Figure 2. Structures of tricyclic molecules 2a–d with terminal tetraﬂuorocyclohexa-1,3-diene rings. g y 2. Materials and Methods 2 1 General Characterization 2.1. General Characterization 19F-NMR (376.46 MHz) spectra were recorded on the Bruker AVANCE III 400 spectrometer in CDCl3 using trichloroﬂuoromethane (CFCl3, δF = 0.00 ppm) as an internal standard. Infrared (IR) spectra were acquired using the KBr method on a JASCO FT/IR-4100 type A spectrometer. High-resolution mass spectra (HRMS) were recorded on a JEOL JMS-700MS spectrometer using fast atom bombard- ment (FAB+) methods. Column chromatography was performed using Wakogel® 60N (38–100 µm), and thin-layer chromatography was performed using the corresponding silica gel plates (silica gel 60F254, Merck, Darmstadt, Germany). Crystals 2023, 13, 1208 3 of 16 2.2. Materials molecules wer The target molecules were synthesized according to a previously reported method from readily available precursors, namely, dimethyl 2,2,3,3-tetraﬂuorosuccinate (2a and 2d; Scheme 1a) [18] and 4-bromo-3,3,4,4-tetraﬂuorobut-1-ene (2b and 2c, Scheme 1b) [15]. molecules were synthesized according to a previously reported method ailable precursors, namely, dimethyl 2,2,3,3-tetrafluorosuccinate (2a and 18] and 4-bromo-3,3,4,4-tetrafluorobut-1-ene (2b and 2c, Scheme 1b) [15]. y p , y, y , , , ( Scheme 1a) [18] and 4-bromo-3,3,4,4-tetraﬂuorobut-1-ene (2b and 2c, Scheme 1b) [15]. 2d; Scheme 1a) [18] and 4-bromo-3,3,4,4-tetrafluorobut-1-ene (2b and 2c, Scheme 1b) [15]. Scheme 1. Syntheses of (a) 2a and 2d, and (b) 2b and 2c. Detailed synthetic procedures are provided in Schemes S1 and S2, and Figures S1– S42 shown in the Supplementary Materials. Characterization data are presented below (for 2a–d) and in the Supplementary Materials (for other molecules). Scheme 1. Syntheses of (a) 2a and 2d, and (b) 2b and 2c. Detailed synthetic procedures are provided in Schemes S1 and S2, and Figures S1– shown in the Supplementary Materials. Characterization data are presented below 2a–d) and in the Supplementary Materials (for other molecules). Scheme 1. Syntheses of (a) 2a and 2d, and (b) 2b and 2c. Scheme 1. Syntheses of (a) 2a and 2d, and (b) 2b and 2c. es of (a) 2a and 2d, and (b) 2b and 2c. Scheme 1. Syntheses of (a) 2a and 2d, and (b) 2b and 2c. es of (a) 2a and 2d, and (b) 2b and 2c. Scheme 1. Syntheses of (a) 2a and 2d, and (b) 2b and 2c. nthetic procedures are provided in Schemes S1 and S2, and Figures S1– e Supplementary Materials. Characterization data are presented below the Supplementary Materials (for other molecules) Detailed synthetic procedures are provided in Schemes S1 and S2, and Figures S1–S42 shown in the Supplementary Materials. g y 2. Materials and Methods 2 1 General Characterization 2.1. General Characterization Characterization data are presented below (for 2a–d) and in the Supplementary Materials (for other molecules). Crystals 2023, 13, 1208 4 of 16 2.2.1. 4-Ethyl-5,5,6,6-tetraﬂuoro-1-[4-(4-n-propylphenyl)phenyl]cyclohexa-1,3-diene (2a) 2.2.1. 4-Ethyl-5,5,6,6-tetraﬂuoro-1-[4-(4-n-propylphenyl)phenyl]cyclohexa-1,3-diene (2a) Yield: 90% (0.25 g, 0.67 mmol); yellow solid; Tm: 92 ◦C; 1H-NMR (CDCl3): δ 0.98 (t, J = 7.2 Hz, 3H), 1.20 (t, J = 7.6 Hz, 3H), 1.69 (sext, J = 7.6 Hz, 2H), 2.40 (q, J = 7.2 Hz, 2H), 2.64 (t, J = 7.2 Hz, 2H), 6.09 (d, J = 6.0 Hz, 1H), 6.39 (d, J = 6.0 Hz, 1H), 7.27 (d, J = 7.8 Hz, 2H), 7.53 (d, J = 8.4 Hz, 4H), 7.62 (d, J = 7.6 Hz, 2H); 13C-NMR (CDCl3): δ 11.4, 13.8, 21.6, 24.5, 37.7, 114.0 (tt, J = 251.87, 26.82 Hz), 114.1 (tt, J = 251.9, 26.8 Hz), 123.1 (t, J = 9.2 Hz), 125.8 (t, J = 8.79 Hz), 126.8, 126.9, 127.4, 129.0, 131.7, 133.6 (t, J = 22.0 Hz), 137.6, 137.8 (t, J = 21.9 Hz), 141.5, 142.3; 19F-NMR (CDCl3): δ −126.57 (d, J = 4.76 Hz, 2F), −122.23 (d, J = 4.86 Hz, 2F). The above characterization data were consistent with those reported previously [15,18]. 2.2.2. 5,5,6,6-Tetraﬂuoro-1-(4-methoxyphenyl)phenylcyclohexa-1,3-diene (2b) , , , ( yp y )p y y , ( ) Yield: 80% (0.53 g, 1.6 mmol); white solid; Tm: 130 ◦C; 1H-NMR (CDCl3): δ 3.86 (s, 3H), 6.03–6.12 (m, 1H), 6.37–6.46 (m, 2H), 7.00 (d, J = 8.8 Hz, 2H), 7.51–7.63 (m, 6H); 13C-NMR (CDCl3): δ 55.3, 110.0–116.0 (m, 2C of CF2CF2), 114.3, 122.9 (t, J = 25.8 Hz), 124.7 (t, J = 8.0 Hz), 126.7, 127.7, 128.1, 129.9 (t, J = 11.8 Hz), 130.9, 132.6, 136.4 (t, J = 22.0 Hz), 141.6, 159.5; 19F-NMR (CDCl3): δ −121.29 (s, 2F), −121.67 (s, 2F); IR (KBr): ν 3026, 2968, 2844, 1649, 1604, 1576, 1530, 1445, 1399, 1312, 1289, 1202, 1183, 1021, 1011, 879, 787 cm−1; HRMS (FAB) calculated for C19H14F4O [M]+: 334.0980, found: 334.0980. Crystal data for C19H14F4O (M = 334.30 g/mol): orthorhombic, space group P 21 21 21, a = 5.5586(7) Å, b = 9.2038(15) Å, c = 29.282(4) Å, α = 90◦, β = 90◦, γ = 90◦, V = 1498.1(4) Å3, Z = 4, T = 173 K, µ(MoKα) = 0.710 mm−1, Dcalc = 1.482 g/cm3, 98,894 reﬂections measured (3.042◦≤2θ ≤27.480◦), 7212 unique (Rint = 0.0476, Rsigma = 0.0950), which were used in all calculations. g y 2. Materials and Methods 2 1 General Characterization 2.1. General Characterization The ﬁnal R1 was 0.0631 (I > 2σ(I)) and wR2 was 0.1269 (all data). 2.2.3. 5,5,6,6-Tetraﬂuoro-1-{4-(triﬂuoromethyl)phenyl}phenylcyclohexa-1,3-diene (2c) Yield: 70% (0.15 g, 1.4 mmol); white solid; Tm: 138 ◦C; 1H-NMR (CDCl3): δ 6.06–6.18 (m, 1H), 6.40–6.54 (m, 2H), 7.60 (d, J = 8.8 Hz, 2H), 7.64 (d, J = 8.8 Hz, 2H), 7.72 (s, 4H); 13C-NMR (CDCl3): δ 112.7 (tt, J = 249.4, 27.2 Hz), 113.4 (tt, J = 253.0, 26.5 Hz), 124.2 (q, J = 272.1 Hz), 123.4 (t, J = 25.6 Hz), 125.5 (t, J = 8.0 Hz), 125.8 (q, J = 3.6 Hz), 127.3, 127.4, 127.9, 129.7 (q, J = 32.3 Hz), 129.8 (t, J = 11.8 Hz), 132.6, 136.1 (t, J = 22.7 Hz), 140.4, 143.6; 19F-NMR (CDCl3): δ −62.43 (s, 3F), −121.31 (s, 2F), −121.72 (s, 2F); IR (KBr): ν 3088, 2362, 1919, 1690, 1616, 1502, 1425, 1274, 1210, 968, 875, 794, 739, 729 cm−1; HRMS (FAB) calculated for C19H11F7 [M]+: 372.0749, found: 372.0759. 2.2.4. 4-Ethyl-5,5,6,6-tetrafluoro-1-[4-{4-(n-octyloxy)phenyl}phenyl]cyclohexa-1,3-diene(2d) Yield: 83% (0.32 g, 0.69 mmol); pale-yellow solid; Tm: 71 ◦C; 1H-NMR (CDCl3): δ 0.90 (t, J = 6.8 Hz, 3H), 1.19 (t, J = 7.2 Hz, 3H), 1.26–1.42 (m, 8H), 1.48 (quin, J = 8.0 Hz, 2H), 1.81 (quin, J = 7.2 Hz, 2H), 2.40 (q, J = 7.2 Hz, 2H), 4.00 (t, J = 6.8 Hz, 2H), 6.09 (d, J = 6.0 Hz, 1H), 6.38 (d, J = 6.0 Hz, 1H), 6.98 (d, J = 8.8 Hz, 2H), 7.48–7.62 (m, 6H); 13C-NMR (CDCl3):δ 11.5, 14.1, 21.7, 22.7, 26.1, 29.26, 29.31, 29.4, 31.8, 68.1, 110–125 (m, 2C of CF2CF2), 114.9, 123.1 (t, J = 8.8 Hz), 125.6 (t, J = 8.8 Hz), 126.7, 127.5, 128.0, 131.3, 132.6, 133.7 (t, J = 23.1 Hz), 137.8 (t, J = 21.2 Hz), 141.3, 159.1; 19F-NMR (CDCl3): δ –123.55 (s, 2F), –127.90 (s, 2F); IR (KBr): ν 3038, 2926, 2852, 1885, 1654, 1606, 1579, 1529, 1500, 1253, 1132, 907, 864 cm−1; HRMS (FAB) calculated for C28H32F4O [M]+: 460.2389, found: 460.2382. 2.6. Theoretical Calculations All computations were performed using the Gaussian 16 program set [23] with density functional theory (DFT) at the level of the M06-2X hybrid functional [24] and the 6-31+G(d) (for all atoms) basis set with a conductor-like polarizable continuum model (CPCM) [25] for CHCl3. Theoretical vertical transitions were calculated using the time-dependent DFT (TD-DFT) method at the same theoretical level using the same solvation model. 2.3. Single-Crystal X-ray Diffraction (XRD) Single-crystal XRD patterns were recorded on an XtaLAB AFC11 diffractometer (Rigaku, Tokyo, Japan). The reﬂection data were integrated, scaled, and averaged using CrysAlisPro software (v. 1.171.39.43a; Rigaku Corporation, Akishima, Japan), and empiri- cal absorption corrections were applied using the SCALE 3 ABSPACK scaling algorithm (CrysAlisPro). Structures were identiﬁed using a direct method (SHELXT-2018/2 [20]), re- ﬁned using a full-matrix least-squares method (SHELXL-2018/3 [21]), and visualized using OLEX2 [22]. The crystallographic data were deposited in the Cambridge Crystallographic Data Center (CCDC) database (CCDC 2269760 for 2b) and can be obtained free of charge Crystals 2023, 13, 1208 5 of 16 5 of 16 from the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: +44-1223-336033; e-mail: deposit@ccdc.cam.ac.uk. from the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: +44-1223-336033; e-mail: deposit@ccdc.cam.ac.uk. 2.4. Photophysical Properties JASCO V-750 absorption (JASCO, Tokyo, Japan) and FP-6600 fluorescence (JASCO, Tokyo, Japan) spectrometers were used to acquire solution-phase ultraviolet/visible (UV/vis) absorption and PL spectra. A Quantaurus-QY C11347-01 instrument (Hamamatsu Photon- ics, Hamamatsu, Japan) was used for PL quantum yield measurements, and a Quantaurus- Tau ﬂuorescence lifetime spectrometer (C11367-34; Hamamatsu Photonics, Japan) was employed for PL lifetime determination. 2.5. LC Properties Polarizing optical microscopy (POM) measurements were carried out using an Olym- pus BX53 microscope (Tokyo, Japan) equipped with cooling and heating stages (10,002 L, Linkam Scientiﬁc Instruments, Surrey, UK) to assess LC properties. Thermodynamic prop- erties were assessed using differential scanning calorimetry (DSC; DSC-60 Plus, Shimadzu, Kyoto, Japan) at heating and cooling rates of 5.0 ◦C·min−1 under N2. Variable-temperature powder X-ray diffraction (VT-PXRD) analyses were carried out using an X-ray diffractome- ter (Rigaku, MiniFlex600, Tokyo, Japan) equipped with an X-ray tube (Cu Kα, λ = 1.54 Å) and semiconductor detector (D/teX Ultra2). The sample powder was mounted on a non- reﬂecting silicon plate set on a benchtop stage (Anton Paar, BTS-500). The temperature, heating/cooling rate, and X-ray exposure time were controlled. 3. Results and Discussion (a) 4° 19° (b) (c C(sp3)···F: 302.3 pm CH···p: 284.5 pm C(sp2)···F: 304.4 pm o a b c (b) CH···p: 286.7 pm O···H: 270.5 pm o a b c (c) CH···p: 286.7 pm O···H: 270.5 pm C(sp3)···F: 302.3 pm CH···p: 284.5 pm C(sp2)···F: 304.4 pm Figure 3. (a) Molecular structure and (b,c) packing of 2b in the crystalline lattice. Display notation: space-filling model for rearmost molecules, ball-and-stick model for middle molecules, and wire- frame model for frontmost molecules. Figure 3. (a) Molecular structure and (b,c) packing of 2b in the crystalline lattice. Display notation: space-ﬁlling model for rearmost molecules, ball-and-stick model for middle molecules, and wire- frame model for frontmost molecules. Compound 2b crystallized in an orthorhombic system (P 21 21 21 space group) and featured a unit cell with four molecules. The dihedral angle between the two aromatic rings of the biphenyl moiety was approximately 4°, and that between the tetrafluorocy- clohexa-1,3-diene ring and the biphenyl moiety was approximately 19° (Figure 3a). In 1b, which has a central tetrafluorocyclohexa-1,3-diene ring, the dihedral angle between the cyclohexa-1,3-diene ring and the adjacent aromatic ring was at least 31° [19]. On the basis of the molecular structures of 1b and 2b, we concluded that the change in the position of the tetrafluorocyclohexa-1,3-diene ring from central to terminal favored a more planar structure. The space-filling model representation in Figure 3b suggests that the π/F inter- actions [26,27] between the π-electrons and F atoms of tetrafluorocyclohexa-1,3-diene re- sulted in the formation of a stacked structure along the a-axis. The C(sp2)···F interatomic distance corresponding to the π/F interaction (304.4 pm) was shorter than the sum of van der Waals radii (317 pm) of carbon (170 pm) and fluorine (147 pm) atoms [10] The mole- Compound 2b crystallized in an orthorhombic system (P 21 21 21 space group) and featured a unit cell with four molecules. The dihedral angle between the two aromatic rings of the biphenyl moiety was approximately 4◦, and that between the tetraﬂuorocyclohexa- 1,3-diene ring and the biphenyl moiety was approximately 19◦(Figure 3a). In 1b, which has a central tetraﬂuorocyclohexa-1,3-diene ring, the dihedral angle between the cyclohexa- 1,3-diene ring and the adjacent aromatic ring was at least 31◦[19]. On the basis of the molecular structures of 1b and 2b, we concluded that the change in the position of the tetraﬂuorocyclohexa-1,3-diene ring from central to terminal favored a more planar structure. 3. Results and Discussion Compounds 2a and 2d, featuring an ethyl group attached to the longitudinal molecular terminal, were synthesized from the readily available dimethyl 2,2,3,3-tetraﬂuorosuccinate according to a reported procedure (Scheme 1a) [18]. The reaction of dimethyl tetraﬂu- orosuccinate with 4-(4-n-propylphenyl)phenylmagnesium bromide in THF at −78 ◦C overnight followed by hydrolysis under acidic conditions afforded ketoester 3a in 64% yield. Compound 3a was treated with 3.6 equivalents of vinylmagnesium chloride in Et2O, and the reaction mixture was stirred overnight at reﬂux to afford 4,4,5,5-tetraﬂuoroocta-1,7- diene (4a) in 37% yield. In the presence of a second-generation Grubbs catalyst, the ring-closing metathesis of 4a in CH2Cl2 (40 ◦C, 24 h) furnished 1-aryl-4-ethyl-5,5,6,6- tetraﬂuorocyclohex-2-ene-1,4-diol (5a) in 49% yield. The 24 h exposure of 5a in methanol to H2 at room temperature resulted in catalytic hydrogenation and furnished 1-aryl-4-ethyl 2,2,3,3-tetraﬂuorocyclohexan-1,4-diol (6a) in 70% yield. Subsequent dehydration with phosphoryl chloride in pyridine at 90 ◦C for 24 h produced 2a in 90% yield. The octyloxy chain-bearing structural analog 2d was prepared by a similar procedure starting with the addition of 4-(4-octyloxyphenyl)phenylmagnesium bromide. Compound 2b, featuring an electron-donating methoxy group, and 2c, featuring an electron-withdrawing triﬂuoromethyl (CF3) group at the longitudinal molecular end, were synthesized according to a previously reported procedure (Scheme 1b) [15]. The Barbier-type nucleophilic addition of 1,1,2,2-tetraﬂuorobut-3-enyllithium (prepared in situ from 4-bromo-3,3,4-4-tetraﬂuorobut-1-ene and LiBr-free MeLi) to p-anisaldehyde in tetrahydrofuran (THF) at −78 ◦C for 2 h gave tetraﬂuorohomoallyl alcohol 7b in 70% yield. The oxidation of 7b with Oxone® in the presence of sodium 2-iodobenzenesulfonate Crystals 2023, 13, 1208 Crystals 2023 13 x FOR 6 of 16 6 of 16 (pre-IBS; 5 mol%) in acetonitrile at 90 ◦C for 16 h afforded 1-aryl-2,2,3,3-tetraﬂuoropent-4- en-1-one (8b) in 77% yield. Compound 8b was treated with allylmagnesium bromide in THF at −78 ◦C for 2 h to produce 4-aryl-5,5,6,6-tetraﬂuoroocta-1,7-diene-4-ol (9b) in 58% yield. Compound 9b underwent ring-closing metathesis upon treatment with a second- generation Grubbs catalyst (3 mol.%) to furnish 4-aryl-5,5,6,6-cyclohex-1-en-4-ol (10b) in 73% yield. The dehydration of 10b with phosphoryl chloride in pyridine at 90 ◦C for 24 h produced the target methoxy-substituted species (2b) in 80% yield. The CF3-substituted 2c was synthesized using the same reaction sequence. 1-one (8b) in 77% yield. Compound 8b was treated with allylmagnesium bromide in THF at −78 °C for 2 h to produce 4-aryl-5,5,6,6-tetrafluoroocta-1,7-diene-4-ol (9b) in 58% yield. 3. Results and Discussion Compound 9b underwent ring-closing metathesis upon treatment with a second-genera- tion Grubbs catalyst (3 mol.%) to furnish 4-aryl-5,5,6,6-cyclohex-1-en-4-ol (10b) in 73% yield. The dehydration of 10b with phosphoryl chloride in pyridine at 90 °C for 24 h pro- duced the target methoxy-substituted species (2b) in 80% yield. The CF3-substituted 2c was synthesized using the same reaction sequence. i y g q Compounds 2a–d were purified by column chromatography (eluent: hexane/EtOAc = 3/1 for 2a or 10/1 for 2b–d) and recrystallization from a 1:1 (v/v) mixture of CH2Cl2 and hexane. The molecular structures of the target molecules were conﬁrmed by NMR spectroscopy, IR spectroscopy, and HRMS, and the related purities were sufﬁcient for photophysical and LC property analyses. Compounds 2a–d were purified by column chromatography (eluent: hexane/EtOAc = 3/1 for 2a or 10/1 for 2b–d) and recrystallization from a 1:1 (v/v) mixture of CH2Cl2 and hexane. The molecular structures of the target molecules were confirmed by NMR spec- troscopy, IR spectroscopy, and HRMS, and the related purities were sufficient for photo- physical and LC property analyses. Among 2a–d, only the methoxy-substituted 2b furnished single crystals appropriate for X-ray crystallographic analysis upon recrystallization, whereas 2a, 2c, and 2d did not furnish single crystals even after multiple recrystallizations. Figure 3 shows the crystal structure of 2b obtained by X-ray structure analysis. Among 2a–d, only the methoxy-substituted 2b furnished single crystals appropriate for X-ray crystallographic analysis upon recrystallization, whereas 2a, 2c, and 2d did not furnish single crystals even after multiple recrystallizations. Figure 3 shows the crystal structure of 2b obtained by X-ray structure analysis. Figure 3. (a) Molecular structure and (b,c) packing of 2b in the crystalline lattice. Display notation: space-filling model for rearmost molecules, ball-and-stick model for middle molecules, and wire- frame model for frontmost molecules. (b) (c) (a) C(sp3)···F: 302.3 pm CH···p: 284.5 pm C(sp2)···F: 304.4 pm o a b c CH···p: 286.7 pm O···H: 270.5 pm o a b c 4° 19° Figure 3. (a) Molecular structure and (b,c) packing of 2b in the crystalline lattice. Display notation: space-ﬁlling model for rearmost molecules, ball-and-stick model for middle molecules, and wire- frame model for frontmost molecules. 3.2. Photophysical Properties 3.2. Photophysical Properties Figure 4 shows the UV/vis absorption spectra, PL spectra, and PL color chromaticity diagrams (as deﬁned by the Commission Internationale de l’Eclailage (CIE)) of 2a–d, and Table 1 lists the related photophysical data. Figure 4 shows the UV/vis absorption spectra, PL spectra, and PL color chromatic diagrams (as defined by the Commission Internationale de l’Eclailage (CIE)) of 2a–d, a Table 1 lists the related photophysical data. Figure 4. (a) Ultraviolet/visible absorption spectra (concentration: 1.0 × 10–5 mol·L−1) and (b) pho luminescence (PL) spectra (concentration: 1.0 × 10–6 mol·L−1) of 2a–d measured in chlorofor (CHCl3). Inset: photographs of PL in CHCl3 solution under UV irradiation (λex = 365 nm). (c) Co mission Internationale de l’Eclailage (CIE) chromaticity diagram for PL colors of 2a–d. 2a 2b 2d 2c (a) (b) (c) 2a 2b 2c 2d CHCl3 solution (1.0×10–5 mol L–1) CHCl3 solution (1.0×10–6 mol L–1) 2a 2b 2c 2d 2a 2b 2c 2d Figure 4. (a) Ultraviolet/visible absorption spectra (concentration: 1.0 × 10–5 mol·L−1) and (b) pho- toluminescence (PL) spectra (concentration: 1.0 × 10–6 mol·L−1) of 2a–d measured in chloroform (CHCl3). Inset: photographs of PL in CHCl3 solution under UV irradiation (λex = 365 nm). (c) Com- mission Internationale de l’Eclailage (CIE) chromaticity diagram for PL colors of 2a–d. (a) (b) 2a 2b 2c 2d CHCl3 solution (1.0×10–5 mol L–1) CHCl3 solution (1.0×10–6 mol L–1) 2a 2b 2c 2d 2a 2b 2c 2d (a) 2a 2b 2c 2d CHCl3 solution (1.0×10–5 mol L–1) (b) CHCl3 solution (1.0×10–6 mol L–1) 2a 2b 2c 2d 2a 2b 2c 2d (b) (a) Figure 4. (a) Ultraviolet/visible absorption spectra (concentration: 1.0 × 10–5 mol·L−1) and (b) phot luminescence (PL) spectra (concentration: 1.0 × 10–6 mol·L−1) of 2a–d measured in chlorofor (CHCl3). Inset: photographs of PL in CHCl3 solution under UV irradiation (λex = 365 nm). (c) Com mission Internationale de l’Eclailage (CIE) chromaticity diagram for PL colors of 2a–d. 2a 2b 2d 2c (c) Figure 4. (a) Ultraviolet/visible absorption spectra (concentration: 1.0 × 10–5 mol·L−1) and (b) pho- toluminescence (PL) spectra (concentration: 1.0 × 10–6 mol·L−1) of 2a–d measured in chloroform (CHCl3). Inset: photographs of PL in CHCl3 solution under UV irradiation (λex = 365 nm). (c) Com- mission Internationale de l’Eclailage (CIE) chromaticity diagram for PL colors of 2a–d. 2a 2b 2d 2c (c) Figure 4. 3. Results and Discussion The space-ﬁlling model representation in Figure 3b suggests that the π/F interactions [26,27] between the π-electrons and F atoms of tetraﬂuorocyclohexa-1,3-diene resulted in the formation of a stacked structure along the a-axis. The C(sp2)···F interatomic distance corresponding to the π/F interaction (304.4 pm) was shorter than the sum of van der Waals radii (317 pm) of carbon (170 pm) and ﬂuorine (147 pm) atoms [10]. The molecule Crystals 2023, 13, 1208 Crystals 2023, 13, x FOR 7 of 16 7 o 7 of 16 7 of represented by the space-ﬁlling model formed molecular packings featuring two pairs of CH/π interactions [28] with the molecule represented by the ball-and-stick model along the b-axis direction. The C(sp2)···H interatomic distance corresponding to the CH/π interaction worked (284.5 pm) was also shorter than the sum of the van der Waals radii (290 pm) of carbon (170 pm) and hydrogen (120 pm). In addition to the short distance between the C(sp2) and H atoms, the carbon atom of the methoxy group was in close contact with the ﬂuorine atom at a distance (302.3 pm) shorter than the sum of carbon (170 pm) and ﬂuorine (147 pm) van der Waals radii. The molecule represented by the ball-and-stick model also formed a stacked structure with the molecule represented by the wire-frame model along the a-axis via CH/π interactions (short contact: 286.7 pm) and O/H hydrogen bonds (short contact: 270.5 pm) (Figure 3c). Accordingly, herringbone-type packing structures were formed through multiple intermolecular interactions. However, unlike the packing structure of 1b [19], which features a central cyclohexa-1,3-diene ring, the packing structure of 2b did not feature intermolecular π/π stacking. pairs of CH/π interactions [28] with the molecule represented by the ball-and-stick mo along the b-axis direction. The C(sp2)···H interatomic distance corresponding to the CH interaction worked (284.5 pm) was also shorter than the sum of the van der Waals ra (290 pm) of carbon (170 pm) and hydrogen (120 pm). In addition to the short distan between the C(sp2) and H atoms, the carbon atom of the methoxy group was in close co tact with the fluorine atom at a distance (302.3 pm) shorter than the sum of carbon (1 pm) and fluorine (147 pm) van der Waals radii. 3. Results and Discussion The molecule represented by the ball-an stick model also formed a stacked structure with the molecule represented by the wi frame model along the a-axis via CH/π interactions (short contact: 286.7 pm) and O hydrogen bonds (short contact: 270.5 pm) (Figure 3c). Accordingly, herringbone-ty packing structures were formed through multiple intermolecular interactions. Howev unlike the packing structure of 1b [19], which features a central cyclohexa-1,3-diene ri the packing structure of 2b did not feature intermolecular π/π stacking. 3.2. Photophysical Properties 3.2. Photophysical Properties 3.2. Photophysical Properties 3.2. Photophysical Properties (a) Ultraviolet/visible absorption spectra (concentration: 1.0 × 10–5 mol·L−1) and (b) phot luminescence (PL) spectra (concentration: 1.0 × 10–6 mol·L−1) of 2a–d measured in chlorofor (CHCl3). Inset: photographs of PL in CHCl3 solution under UV irradiation (λex = 365 nm). (c) Com mission Internationale de l’Eclailage (CIE) chromaticity diagram for PL colors of 2a–d. Figure 4. (a) Ultraviolet/visible absorption spectra (concentration: 1.0 × 10–5 mol·L−1) and (b) pho- toluminescence (PL) spectra (concentration: 1.0 × 10–6 mol·L−1) of 2a–d measured in chloroform (CHCl3). Inset: photographs of PL in CHCl3 solution under UV irradiation (λex = 365 nm). (c) Com- mission Internationale de l’Eclailage (CIE) chromaticity diagram for PL colors of 2a–d. Crystals 2023, 13, 1208 Table 1. P 8 of 16 Table 1. Photophysical data of 2a–d in CHCl3 solution. Molecule λabs [nm] 1 (ε [103, L·mol−1·cm−1]) λPL [nm] 2 (ΦPL) 3 τ [ns] kr [108, s−1] 4 knr [108, s−1] 5 CIE (x, y) 2a 330 (18.5) 437 (0.94) 2.08 4.52 0.28 (0.155, 0.102) 2b 337 (27.6) 463 (0.60) 1.84 3.29 2.15 (0.153, 0.193) 2c 320 (37.4) 416 (0.24) 0.73 3.30 10.44 (0.157, 0.036) 2d 337 (28.9) 463 (0.89) 2.11 4.21 0.53 (0.150, 0.145) 1 Concentration: 1.0 × 10−5 mol·L−1. 2 Concentration: 1.0 × 10−6 mol·L−1. 3 Measured using an integrat- ing sphere. 4 Radiative deactivation rate constant (kr) = ΦPL/τ. 5 Nonradiative deactivation rate constant (knr) = (1 – ΦPL)/τ. (ε [103, L·mol 1·cm 1]) (ΦPL) 3 [108, s 1] 4 [108, s 1] 5 (x, y) 2a 330 (18.5) 437 (0.94) 2.08 4.52 0.28 (0.155, 0.102) 2b 337 (27.6) 463 (0.60) 1.84 3.29 2.15 (0.153, 0.193) 2c 320 (37.4) 416 (0.24) 0.73 3.30 10.44 (0.157, 0.036) 2d 337 (28.9) 463 (0.89) 2.11 4.21 0.53 (0.150, 0.145) 1 Concentration: 1.0 × 10−5 mol·L–1. 2 Concentration: 1.0 × 10−6 mol·L−1. 3 Measured using an integrating sphere. 4 Radiative deactivation rate constant (kr) = ΦPL/τ. 5 Nonradiative deactivation rate constant (knr) = (1 – ΦPL)/τ. Compound 2a, possessing ethyl and n-propyl substituents at longitudinal molecular Table 1. Photophysical data of 2a–d in CHCl3 solution. , L mol cm ]) (ΦPL) [10 , s 330 (18.5) 437 (0.94) 2.08 4.5 Compound 2a, possessing ethyl and n-propyl substituents at longitudinal molecular terminals, exhibited a single absorption band with a maximum absorption wavelength (λabs) of ~330 nm in CHCl3. 3.2. Photophysical Properties 3.2. Photophysical Properties The high-energy PL of 2c corresponded to dark-blue color represented by CIE chro- maticity coordinates of (x, y) = (0.157, 0.036) (Figure 4c). In contrast, the low-energy PL of 2b and 2d emitted from ICT states corresponded to light-blue color with CIE coordinates of (x, y) = (0.153, 0.189). The quantum yields (ΦPL) and PL lifetimes (τ) of 2a–d were determined as 0.24–0.94 and ~2.11 ns, respectively. This value of τ indicates that the light emitted by 2a–d was ﬂuorescent. Among the four compounds, 2c exhibited the lowest ΦPL (0.24) and a very short τ (<1.0 ns). The radiative (kr) and nonradiative (knr) deactivation rate constants of 2c were calculated from ΦPL and τ as 3.30 × 108 s−1 and 10.44 × 108 s−1, respectively. Notably, kr was not signiﬁcantly different between 2a and d, whereas the knr of 2c was 5–37 times higher than those of other derivatives. These results suggested the occurrence of ﬂuorescence reabsorption (self-absorption) in 2c, which resulted in decreased Φ d i d k Table 2. Theoretical data of 2a–d obtained using Gaussian software with time-dependent density functional theory 1. Molecule HOMO Energy [eV] LUMO Energy [eV] Theoretical λcalcd [nm] Oscillator Strength (f) Theoretical Transition (Probability) 2a −7.33 −1.50 331 1.00 HOMO→LUMO (85%) HOMO–1→LUMO (12%) 2b −7.22 −1.66 338 0.90 HOMO→LUMO (77%) HOMO–1→LUMO (20%) 2c −7.75 −1.76 325 0.87 HOMO→LUMO (91%) HOMO–2→LUMO (6%) 2d −7.14 −1.48 335 1.04 HOMO→LUMO (78%) HOMO–1→LUMO (19%) 1 Calculated at the M06-2X/6-31+G(d) level of theory using a conductor-like polarizable continuum model for CHCl3. Table 2. Theoretical data of 2a–d obtained using Gaussian software with time-dependent density functional theory 1. According to Figure 5, the HOMOs of 2a–d were spread throughout the π-conjugated structure, whereas the LUMOs were localized on the tetraﬂuorocyclohexa-1,3-diene ring of the tricyclic π-conjugated framework. Substituents at longitudinal molecular ends affected the HOMO and LUMO energies, e.g., electron-donating substituents such as alkoxy groups increased the HOMO energy, whereas electron-withdrawing substituents had the opposite effect. The alkoxy group at the opposite end did not affect the energy of the LUMO, as this orbital was localized on the tetraﬂuorocyclohexa-1,3-diene ring, whereas the electron-donating ethyl group introduced into the tetraﬂuorocyclohexa-1,3-diene skeleton increased the LUMO energies of 2a and 2d. 3.2. Photophysical Properties 3.2. Photophysical Properties The absorption wavelengths (λcalcd) of 2a–d determined by TD-DFT calculations (331 nm for 2a, 338 nm for 2b, 325 nm for 2c, and 335 nm for 2d) were close to the measured λabs values listed in Table 1. The transitions from the ground to the ﬁrst excited states were calculated to be of π–π* HOMO→LUMO and HOMO–1/HOMO–2→LUMO types. yp When a solution of 2a in CHCl3 was excited by irradiation with UV light at λabs (330 nm), a single PL band with a maximum PL wavelength (λPL) of approximately 437 nm was observed (Figure 4b). Compared to that of 2a, the PL band of 2b with an electron- donating methoxy group (λPL = 463 nm in CHCl3) was red-shifted by 26 nm, whereas the PL band of 2c with an electron-withdrawing CF3 group (λPL = 416 nm) was substantially blue-shifted. Similar to the methoxy-substituted 2b, the n-octyloxy-substituted 2d exhibited PL (λPL = 463 nm). In the case of 2c with a large HOMO–LUMO overlap, radiative deactivation probably occurred from the locally excited state, whereas 2b or 2d with a locally existing LUMO luminesced through the radiative deactivation of the intramolecular charge transfer (ICT) excited state, which can be reasonably explained by the Lippert– Mataga plot [29,30] shown in Figure S49. The high-energy PL of 2c corresponded to dark-blue color represented by CIE chro- maticity coordinates of (x, y) = (0.157, 0.036) (Figure 4c). In contrast, the low-energy PL of 2b and 2d emitted from ICT states corresponded to light-blue color with CIE coordinates of (x, y) = (0.153, 0.189). The quantum yields (ΦPL) and PL lifetimes (τ) of 2a–d were determined as 0.24–0.94 and ~2.11 ns, respectively. This value of τ indicates that the light emitted by 2a–d was ﬂuorescent. Among the four compounds, 2c exhibited the lowest ΦPL (0.24) and a very short τ (<1.0 ns). The radiative (kr) and nonradiative (knr) deactivation rate constants of 2c were calculated from ΦPL and τ as 3.30 × 108 s−1 and 10.44 × 108 s−1, respectively. Notably, kr was not signiﬁcantly different between 2a and d, whereas the knr of 2c was 5–37 times higher than those of other derivatives. These results suggested the occurrence of ﬂuorescence reabsorption (self-absorption) in 2c, which resulted in decreased ΦPL and increased knr. 3.2. Photophysical Properties 3.2. Photophysical Properties Compound 2b, featuring a strongly electron-donating methoxy group, exhibited a red-shifted λabs of 337 nm, whereas 2c, featuring a strongly electron- withdrawing CF3 group, exhibited a blue-shifted λabs of 320 nm. In CHCl3, the λabs of 2d with ethyl and n-octyloxy groups as longitudinal terminal substituents was 337 nm, i.e., equal to that of 2b. ited a single absorption band with a maximum absorption wavelength in CHCl3. Compound 2b, featuring a strongly electron-donating methoxy a red-shifted λabs of 337 nm, whereas 2c, featuring a strongly electron- 3 group, exhibited a blue-shifted λabs of 320 nm. In CHCl3, the λabs of 2d -octyloxy groups as longitudinal terminal substituents was 337 nm, i.e., b. ical vertical transition was modeled using Gaussian software [23] with q The theoretical vertical transition was modeled using Gaussian software [23] with time-dependent density functional theory (TD-DFT). Figure 5 shows the distributions of the highest occupied molecular orbitals (HOMOs) and lowest unoccupied molecular orbitals (LUMOs) for 2a–d, with the related theoretical data summarized in Table 2. The detailed orbital distributions are shown in Figures S44–47 in Supplementary Materials. ical vertical transition was modeled using Gaussian software [23] with density functional theory (TD-DFT). Figure 5 shows the distributions of pied molecular orbitals (HOMOs) and lowest unoccupied molecular or- for 2a–d, with the related theoretical data summarized in Table 2. The distributions are shown in Figures S44–47 in Supplementary Materials. Figure 5. Highest occupied molecular orbital (HOMO, left) and lowest unoccupied molecular orbital (LUMO, right) distributions of (a) 2a, (b) 2b, (c) 2c, and (d) 2d. HOMO (–7.33 eV) LUMO (–1.50 eV) HOMO (–7.22 eV) LUMO (–1.66 eV) HOMO (–7.75 eV) LUMO (–1.76 eV) HOMO (–7.14 eV) LUMO (–1.48 eV) (a) 2a (b) 2b (c) 2c (d) 2d Figure 5. Highest occupied molecular orbital (HOMO, left) and lowest unoccupied molecular orbital (LUMO, right) distributions of (a) 2a, (b) 2b, (c) 2c, and (d) 2d. HOMO (–7.33 eV) (a) 2a (a) 2a LUMO (–1.50 eV) LUMO (–1.50 eV) HOMO (–7.22 eV) (b) 2b (b) 2b LUMO (–1.66 eV) LUMO (–1.76 eV) LUMO (–1.48 eV) ccupied molecular orbital (HOMO, left) and lowest unoccupied molecular orbital ributions of (a) 2a, (b) 2b, (c) 2c, and (d) 2d. Figure 5. Highest occupied molecular orbital (HOMO, left) and lowest unoccupied molecular orbital (LUMO, right) distributions of (a) 2a, (b) 2b, (c) 2c, and (d) 2d. Crystals 2023, 13, 1208 9 of 16 Table 2. 3.2. Photophysical Properties 3.2. Photophysical Properties Theoretical data of 2a–d obtained using Gaussian software with time-dependent density functional theory 1. Molecule HOMO Energy [eV] LUMO Energy [eV] Theoretical λcalcd [nm] Oscillator Strength (f) Theoretical Transition (Probability) 2a −7.33 −1.50 331 1.00 HOMO→LUMO (85%) HOMO–1→LUMO (12%) 2b −7.22 −1.66 338 0.90 HOMO→LUMO (77%) HOMO–1→LUMO (20%) 2c −7.75 −1.76 325 0.87 HOMO→LUMO (91%) HOMO–2→LUMO (6%) 2d −7.14 −1.48 335 1.04 HOMO→LUMO (78%) HOMO–1→LUMO (19%) 1 Calculated at the M06-2X/6-31+G(d) level of theory using a conductor-like polarizable continuum model for CHCl3. According to Figure 5, the HOMOs of 2a–d were spread throughout the π-conjugated structure, whereas the LUMOs were localized on the tetraﬂuorocyclohexa-1,3-diene ring of the tricyclic π-conjugated framework. Substituents at longitudinal molecular ends affected the HOMO and LUMO energies, e.g., electron-donating substituents such as alkoxy groups increased the HOMO energy, whereas electron-withdrawing substituents had the opposite effect. The alkoxy group at the opposite end did not affect the energy of the LUMO, as this orbital was localized on the tetraﬂuorocyclohexa-1,3-diene ring, whereas the electron-donating ethyl group introduced into the tetraﬂuorocyclohexa-1,3-diene skeleton increased the LUMO energies of 2a and 2d. The absorption wavelengths (λcalcd) of 2a–d determined by TD-DFT calculations (331 nm for 2a, 338 nm for 2b, 325 nm for 2c, and 335 nm for 2d) were close to the measured λabs values listed in Table 1. The transitions from the ground to the ﬁrst excited states were calculated to be of π–π* HOMO→LUMO and HOMO–1/HOMO–2→LUMO types. When a solution of 2a in CHCl3 was excited by irradiation with UV light at λabs (330 nm), a single PL band with a maximum PL wavelength (λPL) of approximately 437 nm was observed (Figure 4b). Compared to that of 2a, the PL band of 2b with an electron- donating methoxy group (λPL = 463 nm in CHCl3) was red-shifted by 26 nm, whereas the PL band of 2c with an electron-withdrawing CF3 group (λPL = 416 nm) was substantially blue-shifted. Similar to the methoxy-substituted 2b, the n-octyloxy-substituted 2d exhibited PL (λPL = 463 nm). In the case of 2c with a large HOMO–LUMO overlap, radiative deactivation probably occurred from the locally excited state, whereas 2b or 2d with a locally existing LUMO luminesced through the radiative deactivation of the intramolecular charge transfer (ICT) excited state, which can be reasonably explained by the Lippert– Mataga plot [29,30] shown in Figure S49. 3.2. Photophysical Properties 3.2. Photophysical Properties Molecule λPL [nm] (ΦPL) 1 τave [ns] τ1 [ns] τ2 [ns] kr [108, s−1] 2 knr [108, s−1] 3 CIE (x, y) 2a 509 (0.99) 1.84 – – 5.38 0.054 (0.265, 0.570) 2b 463 (0.63) 2.43 2.06 4.47 2.56 1.55 (0.162, 0.200) 2c 413 (0.31) 3.04 – – 1.03 2.26 (0.155, 0.175) 2d 458 (0.93) 3.19 – – 2.92 0.21 (0.182, 0.423) 1 Measured using an integrating sphere. 2 Radiative deactivation rate constant (kr) = ΦPL/τ. 3 Nonradiative deactivation rate constant (knr) = (1 −ΦPL)/τ. 1 Measured using an integrating sphere. 2 Radiative deactivation rate constant (kr) = ΦPL/τ. 3 Nonra- diative deactivation rate constant (knr) = (1 − ΦPL)/τ. 1 Measured using an integrating sphere. 2 Radiative deactivation rate constant (kr) = ΦPL/τ. 3 Nonradiative deactivation rate constant (knr) = (1 −ΦPL)/τ. Crystalline 2a with two alkyl groups at longitudinal molecular ends exhibited green PL with a single PL band at λPL around 509 nm, which was red-shifted relative to the value in CHCl3 solution by 72 nm. However, the PL behavior of 2b–d did not substantially change upon the transition from the CHCl3 solution to the crystalline state. In the crystal- line state, 2b crystallized mainly via CH/π, π/F, and hydrogen bonds; π/π stacking be- tween the intermolecular aromatic rings was not observed. The similarity between the λPL and ΦPL values observed in the crystalline state and CHCl3 solution was ascribed to the absence of π/π stacking interactions in crystalline 2b, which suppressed the nonradiative deactivation induced by the formation of molecular aggregates. Given that crystalline 2c and 2d also exhibited PL behavior similar to that in the CHCl3 solution state, we concluded that their conjugated structures were also not involved in intermolecular interactions, alt- hough their crystal structures have not yet been elucidated. On the other hand, we inferred that crystalline 2a, which featured a PL wavelength and PL color different from those ob- served in the solution state, interacted with the π-conjugated site through the formation of molecular aggregates, unlike in the dilute solution, although the crystal structure of 2a also remained veiled. PL lifetime measurements showed that the τ of crystalline 2a–d was 1.84–3.19 ns and, therefore, also indicative of fluorescence. The PL decays of 2a, 2c, and 2d were well modeled by a mono-exponential function, and the related PL originated from a single excited state. 3.2. Photophysical Properties 3.2. Photophysical Properties Most molecules exhibiting luminescence in solution generally experience lumines- cence quenching through intermolecular energy transfer at high concentrations or in the solid state. However, 2a–d exhibited strong luminescence even in the crystalline state. Crystals 2023, 13, 1208 10 of 16 Figure 6 shows the PL spectra of crystalline 2a–d, the related CIE chromaticity diagram, and photographs of crystals under 365 nm UV light. The corresponding photophysical data are summarized in Table 3. ure 6 shows the PL spectra of crystalline 2a–d, the related CIE chromaticity diagram, and photographs of crystals under 365 nm UV light. The corresponding photophysical data are summarized in Table 3. Figure 6 shows the PL spectra of crystalline 2a–d, the related CIE chromaticity diagram, and photographs of crystals under 365 nm UV light. The corresponding photophysical data are summarized in Table 3. ure 6 shows the PL spectra of crystalline 2a–d, the related CIE chromaticity diagram, and photographs of crystals under 365 nm UV light. The corresponding photophysical data are summarized in Table 3. Figure 6. (a) PL spectra of crystalline 2a–d. (b) CIE chromaticity diagram and photographs of 2a–d crystals under 365 nm ultraviolet light. 2d 2a 2c 2b (a) (b) Crystal 2a 2b 2c 2d Figure 6. (a) PL spectra of crystalline 2a–d. (b) CIE chromaticity diagram and photographs of 2a–d crystals under 365 nm ultraviolet light. 2d 2a 2c 2b (b) (a) Crystal 2a 2b 2c 2d Figure 6. (a) PL spectra of crystalline 2a–d. (b) CIE chromaticity diagram and photographs of 2a–d crystals under 365 nm ultraviolet light. Figure 6. (a) PL spectra of crystalline 2a–d. (b) CIE chromaticity diagram and photographs of 2a–d crystals under 365 nm ultraviolet light. Table 3. Photophysical data of crystalline 2a–d. Table 3. Photophysical data of crystalline 2a–d. Table 3. Photophysical data of crystalline 2a–d. Molecule λPL [nm] (ΦPL) 1 τave [ns] τ1 [ns] τ2 [ns] kr [108, s–1] 2 knr [108, s–1] 3 CIE (x, y) 2a 509 (0.99) 1.84 – – 5.38 0.054 (0.265, 0.570) 2b 463 (0.63) 2.43 2.06 4.47 2.56 1.55 (0.162, 0.200) 2c 413 (0.31) 3.04 – – 1.03 2.26 (0.155, 0.175) 2d 458 (0.93) 3.19 – – 2.92 0.21 (0.182, 0.423) 1 Measured using an integrating sphere. 2 Radiative deactivation rate constant (kr) = ΦPL/τ. 3 Nonra- diative deactivation rate constant (knr) = (1 − ΦPL)/τ. Table 3. Photophysical data of crystalline 2a–d. 3.2. Photophysical Properties 3.2. Photophysical Properties In contrast, the PL decay of 2b was fitted by a biexponential function assuming a radiative deactivation pathway from any two excited states, although the re- lated excited-state details remain unknown Crystalline 2a with two alkyl groups at longitudinal molecular ends exhibited green PL with a single PL band at λPL around 509 nm, which was red-shifted relative to the value in CHCl3 solution by 72 nm. However, the PL behavior of 2b–d did not substantially change upon the transition from the CHCl3 solution to the crystalline state. In the crystalline state, 2b crystallized mainly via CH/π, π/F, and hydrogen bonds; π/π stacking between the intermolecular aromatic rings was not observed. The similarity between the λPL and ΦPL values observed in the crystalline state and CHCl3 solution was ascribed to the absence of π/π stacking interactions in crystalline 2b, which suppressed the nonradiative deactivation induced by the formation of molecular aggregates. Given that crystalline 2c and 2d also exhibited PL behavior similar to that in the CHCl3 solution state, we concluded that their conjugated structures were also not involved in intermolecular interactions, although their crystal structures have not yet been elucidated. On the other hand, we inferred that crystalline 2a, which featured a PL wavelength and PL color different from those observed in the solution state, interacted with the π-conjugated site through the formation of molecular aggregates, unlike in the dilute solution, although the crystal structure of 2a also remained veiled. PL lifetime measurements showed that the τ of crystalline 2a–d was 1.84–3.19 ns and, therefore, also indicative of ﬂuorescence. The PL decays of 2a, 2c, and 2d were well modeled by a mono-exponential function, and the related PL originated from a single excited state. In contrast, the PL decay of 2b was ﬁtted by a biexponential function assuming a radiative deactivation pathway from any two excited states, although the related excited-state details remain unknown. lated excited-state details remain unknown. Compared with the previously reported 1b and 1c with a central tetraﬂuorocyclohexa- 1,3-diene ring [19], 2b and 2c featured a shorter (by 20–25 nm) λabs in CHCl3, which was Crystals 2023, 13, 1208 11 of 16 ocyclo- 11 of 16 ocyclo- ascribed to the signiﬁcantly increased LUMO level of the latter molecules. However, λPL was found to be almost the same, except for 2c, which had a CF3 group at the molecular ter- minal. 3.3. LC Properties 3.3. LC Properties The previously reported 1a–c exhibited transitions only between their Cry and Iso phases upon heating and cooling, i.e., no LC phases were observed [15]. To understand how the position of the tetraﬂuorocyclohexa-1,3-diene ring in tricyclic molecules affects their LC properties, we used POM and DSC to examine the LC behavior of 2a–d, which showed PL in both dilute solution and crystalline states (Figures S53–S56 in Supplementary Materials). Compounds 2b and 2c exhibited only a Cry→Iso phase transition but did not form any mesophase upon heating and cooling. In contrast, for 2a and 2d, a ﬂuid bright-ﬁeld POM image was observed between the Cry and Iso phases, indicating the formation of an LC phase upon cooling (2a) or heating/cooling (2d). Figure 7 shows the DSC curves of 2a and 2d and the POM images of the corresponding mesophases. Table 4 lists the phase transition behaviors of 2a–d, namely, their phase sequences, as well as phase transition temperatures and enthalpies in the second heating and cooling processes. The previously reported 1a–c exhibited transitions only between their Cry and Iso phases upon heating and cooling, i.e., no LC phases were observed [15]. To understand how the position of the tetrafluorocyclohexa-1,3-diene ring in tricyclic molecules affects their LC properties, we used POM and DSC to examine the LC behavior of 2a–d, which showed PL in both dilute solution and crystalline states (Figures S53–S56 in Supplemen- tary Materials). Compounds 2b and 2c exhibited only a Cry→Iso phase transition but did not form any mesophase upon heating and cooling. In contrast, for 2a and 2d, a fluid bright-field POM image was observed between the Cry and Iso phases, indicating the for- mation of an LC phase upon cooling (2a) or heating/cooling (2d). Figure 7 shows the DSC curves of 2a and 2d and the POM images of the corresponding mesophases. Table 4 lists the phase transition behaviors of 2a–d, namely, their phase sequences, as well as phase transition temperatures and enthalpies in the second heating and cooling processes. Exo. Endo. Scan rate: 5 ℃min–1 Heating Cooling 2a –1 –2 Exo. Endo. Heating Cooling Scan rate: 5 ℃min–1 2d –1 –2 –3 (a) (b) Exo. Endo. Scan rate: 5 ℃min–1 Heating Cooling 2a –1 –2 Exo. Endo. 3.2. Photophysical Properties 3.2. Photophysical Properties In the crystalline state, the λPL of 2c was blue-shifted relative to that of 1c, although almost identical λPL values were observed for 2b and 1b. The CHCl3 solution-phase ΦPL values of 2b and 2c exceeded those of 1b and 1c. In contrast, the opposite trend was ob- served in the crystalline state, i.e., the ΦPL values of 2b and 2c were lower than those of 1b and 1c. In 2b and 2c, which greatly differ from 1b and 1c [19], the biphenyl moiety was pla- nar and formed a herringbone structure because of CH/π interactions. We concluded that weak intermolecular interactions did not lead to molecular motion suppression, resulting in decreased ΦPL. Accordingly, the positional change of the tetraﬂuorocyclohexa-1,3-diene ring in the tricyclic scaffold had a relatively large effect on the crystalline-state behavior, and the position of this ring affected intermolecular interactions and, hence, the extent of molecular motion inhibition and ΦPL. was ascribed to the significantly increased LUMO level of the latter molecules. However, λPL was found to be almost the same, except for 2c, which had a CF3 group at the molecular terminal. In the crystalline state, the λPL of 2c was blue-shifted relative to that of 1c, alt- hough almost identical λPL values were observed for 2b and 1b. The CHCl3 solution-phase ΦPL values of 2b and 2c exceeded those of 1b and 1c. In contrast, the opposite trend was observed in the crystalline state, i.e., the ΦPL values of 2b and 2c were lower than those of 1b and 1c. In 2b and 2c, which greatly differ from 1b and 1c [19], the biphenyl moiety was planar and formed a herringbone structure because of CH/π interactions. We concluded that weak intermolecular interactions did not lead to molecular motion suppression, re- sulting in decreased ΦPL. Accordingly, the positional change of the tetrafluorocyclohexa- 1,3-diene ring in the tricyclic scaffold had a relatively large effect on the crystalline-state behavior, and the position of this ring affected intermolecular interactions and, hence, the extent of molecular motion inhibition and ΦPL. 3.3. LC Properties 3.3. LC Properties Heating Cooling Scan rate: 5 ℃min–1 2d –1 –2 –3 (a) (b) (c) (d) 2a (N, 93 ℃) 2d (SmC, 55 ℃) 2d (SmA, 100 ℃) 2d (SmA, 120 ℃) 100mm 100mm 100mm 100mm P A P A P A P A Figure 7. Differential scanning calorimetry (DSC) curves of (a) 2a and (b) 2d recorded during the second heating and cooling processes at a scan rate of 5 ◦C·min−1 under N2. Polarizing optical microscopy textures in the mesophases of (c) 2a and (d) 2d. (a) (b) Exo. Endo. Scan rate: 5 ℃min–1 Heating Cooling 2a –1 –2 (a) Exo. Endo. Heating Cooling Scan rate: 5 ℃min–1 2d –1 –2 –3 (b) 2a (c) 2a (N, 93 ℃) 100mm P A (d) 2d (SmC, 55 ℃) 2d (SmA, 100 ℃) 2d (SmA, 120 ℃) 100mm 100mm 100mm P A P A P A (c) (d) 2d (SmA, 120 ℃) 2a (N, 93 ℃) 2d (SmC, 55 ℃) 2d (SmA, 100 ℃) Figure 7. Differential scanning calorimetry (DSC) curves of (a) 2a and (b) 2d recorded during the second heating and cooling processes at a scan rate of 5 ◦C·min−1 under N2. Polarizing optical microscopy textures in the mesophases of (c) 2a and (d) 2d. Crystals 2023, 13, 1208 12 of 16 12 of 16 Table 4. Phase transition data of 2a–d during the second heating and cooling processes. Table 4. Phase transition data of 2a–d during the second heating and cooling processes. Molecule Process Phase Transition Temperatures [◦C] and Enthalpies [kJ·mol−1] 1 2a Heating Cry 92 (14.0) Iso Cooling Cry 81 (−4.9) N 90 (−6.2) Iso 2b Heating Cry 130 (16.4) Iso Cooling Cry 93 (−12.1) Iso 2c Heating Cry 138 (11.8) Iso Cooling Cry 98 (−8.2) Iso 2d Heating Cry 71 (12.6) SmA 110 (0.88) N 133 (0.72) Iso Cooling Cry 40 (−8.1) SmC 68 (−1.1) SmA 114 (−1.0) N 136 (−0.85) Iso 1 Determined by DSC (scan rate: 5 ◦C·min−1, atmosphere: N2). Abbreviations: Cry, crystal; Iso, isotropic; N, nematic; SmA, smectic A; SmC, smectic C phase. Molecule Process Phase Transition Temperatures [◦C] and Enthalpies [kJ·mol−1] 1 1 Determined by DSC (scan rate: 5 ◦C·min−1, atmosphere: N2). Abbreviations: Cry, crystal; Iso, isotropic; N nematic; SmA, smectic A; SmC, smectic C phase. In the case of the 2a mesophase, POM revealed that a ﬂuid four-brush Schlieren texture formed at 90 ◦C after the slow cooling from the dark-ﬁeld-image Iso phase. 3.3. LC Properties 3.3. LC Properties Given that POM indicated the formation of a nematic (N) phase with only orientational order, the mesophase appearing during the cooling of 2a was classiﬁed as the N phase. Further cooling from the N-phase state of 2a resulted in ﬂuidity loss at 81 ◦C and a phase transition to the hard Cry phase. In the case of the 2d mesophase, the nonﬂuidic bright-ﬁeld POM image corresponding to the Cry phase changed to a ﬂuidic fan-shaped POM image at 71 ◦C upon heating. Further heating induced an optical texture change to a Schlieren-patterned N phase at 110 ◦C followed by a phase transition to the Iso phase in the dark-ﬁeld POM image at 133 ◦C. Upon cooling, the N-phase Schlieren texture appeared at 136 ◦C, and a transition to a phase with a fan-shaped texture occurred at 114 ◦C. Upon further cooling, a broken fan-shaped texture was observed at 68 ◦C, followed by a phase transition to the nonﬂuidic Cry phase at 40 ◦C. The fan-shaped optical texture observed in the mesophase of 2d is characteristic of the smectic (Sm) phase, which has an orientational and positional order. Notably, in the case of 2d, the Sm phase appeared at a lower temperature than the N phase. p Further insights into the LC phases exhibited by 2a and 2d were provided by VT- PXRD measurements. The pattern of 2a recorded after cooling from the Iso phase and holding at 70 ◦C featured no Cry phase peaks but contained a halo peak centered around 2θ = 18◦(Figure S57). This result strongly suggests that the mesophase appearing in 2a is the N phase without positional order. PXRD measurements were also performed for 2d at 124, 89, and 49 ◦C after cooling from the Iso phase. A halo peak centered around 18◦ was also observed in the pattern recorded at 124 ◦C, and the mesophase appearing at this temperature was determined to be the N phase (Figure S57). The PXRD pattern recorded at 89 ◦C featured a sharp peak at 3.75◦and a weak peak at 7.45◦(Figure 8a). W 13 o Figure 8. Powder X-ray diffraction patterns of 2d recorded at (a) 89 and (b) 49 °C. 3.75°(001) d= 2.35 nm 7.45° (002) SmA (89 ℃) 2.35 nm ≡ SmC (49 ℃) 3.95°(001) d= 2.23 nm 7.85° (002) 2.23 nm (a) (b) Figure 8. 3.4. PL Properties of 2d in Various Molecular Aggregation States ope ie of i a iou o e u a Agg e Compound 2d, which forms various mesop Compound 2d, which forms various mesophases, was selected to investigate PL behav- ior changes associated with the phase transition-induced alterations in molecular aggregate structure. PL behavior was examined using a ﬂuorescence spectrometer equipped with a self-made temperature control unit. The samples were cooled from the Iso phase and held for 5 min at each temperature during cooling. Figure 9 shows the thus obtained PL spectra and CIE chromaticity diagrams, and Table 5 summarizes the related photophysical data. p p g havior changes associated with the phase transition-induced alterations in mol gregate structure. PL behavior was examined using a fluorescence spectrometer with a self-made temperature control unit. The samples were cooled from the and held for 5 min at each temperature during cooling. Figure 9 shows the thu PL spectra and CIE chromaticity diagrams, and Table 5 summarizes the relat physical data. Figure 9. (a) PL spectra of 2d recorded at different temperatures upon cooling. (b) CIE c diagram for PL color of 2d at different temperatures. (a) (b) 2d (Cooling process) Cry 40 SmC 68 SmA 114 N 136 Iso Cry (25 ℃) SmC (50 ℃) SmA (100 ℃) N (130 ℃) Cry Sm A Sm C N Figure 9. (a) PL spectra of 2d recorded at different temperatures upon cooling. (b) CIE chromaticity diagram for PL color of 2d at different temperatures. (b) Cry Sm A Sm C N (a) (b) (a) ( 2d (Cooling process) Cry 40 SmC 68 SmA 114 N 136 Iso Cry (25 ℃) SmC (50 ℃) SmA (100 ℃) N (130 ℃) Figure 9. (a) PL spectra of 2d recorded at different temperatures upon cooling. (b) CIE c diagram for PL color of 2d at different temperatures. Figure 9. (a) PL spectra of 2d recorded at different temperatures upon cooling. (b) CIE chromaticity diagram for PL color of 2d at different temperatures. Table 5. Photophysical data of 2d in various phases. Temp. [◦C]/Phase λPL [nm] I/IN 1 CIE (x, y) 130/N 466 7.8 (0.156, 0.171) 100/SmA 470 1.6 (0.152, 0.217) 50/SmC 468 2.0 (0.158, 0.199) 25/Cry 454 1.0 (0.164, 0.121) 1 PL intensity I of each phase with respect to the PL intensity (IN) of the N phase. Table 5. Photophysical data of 2d in various phases. 3.3. LC Properties 3.3. LC Properties Powder X-ray diffraction patterns of 2d recorded at (a) 89 and (b) 49 ◦C. (a) (b) 3.75°(001) d= 2.35 nm 7.45° (002) SmA (89 ℃) 2.35 nm ≡ (a) SmC (49 ℃) 3.95°(001) d= 2.23 nm 7.85° (002) 2.23 nm (b) igure 8. Powder X-ray diffraction patterns of 2d recorded at (a) 89 and (b) 49 ° Figure 8. Powder X-ray diffraction patterns of 2d recorded at (a) 89 and (b) 49 ◦C. Crystals 2023, 13, 1208 13 of 16 °C 13 of 16 °C These diffraction peaks corresponded to the plane indices of (hkl) = (001) and (002). The peak at 3.75◦in the low-angle region corresponded to a d-spacing of 2.35 nm, according to Bragg’s equation, which was consistent with the longitudinal molecular length of 2d (Figure 8a) This consistency of the interlayer distance with the molecular length agreed with the formation of a smectic A (SmA) phase with a layered periodic structure wherein the long molecular axis was oriented in the direction of the layer normal. In the pattern recorded at 49 ◦C, the peak of the (001) plane appeared at 3.95◦and corresponded to a d-spacing of 2.23 nm, which was shorter than the molecular length along the long molecular axis (2.35 nm) (Figure 8b). This result indicated the presence of a smectic C (SmC) phase featuring a tilt angle with respect to the layer normal (Figure 8b). These diffraction peaks corresponded to the plane indices of (hkl) = (001) The peak at 3.75° in the low-angle region corresponded to a d-spacing of 2.35 n ing to Bragg’s equation, which was consistent with the longitudinal molecular 2d (Figure 8a) This consistency of the interlayer distance with the molecular leng with the formation of a smectic A (SmA) phase with a layered periodic structur the long molecular axis was oriented in the direction of the layer normal. In t recorded at 49 °C, the peak of the (001) plane appeared at 3.95° and correspond spacing of 2.23 nm, which was shorter than the molecular length along the long axis (2.35 nm) (Figure 8b). This result indicated the presence of a smectic C (Sm featuring a tilt angle with respect to the layer normal (Figure 8b). 3 4 PL P ti f 2d i V i M l l A ti St t 4. Conclusions Tricyclic π-conjugated molecules with terminal tetraﬂuorocyclohexa-1,3-diene rings and different substituents introduced at the longitudinal molecular ends (2a–d) were synthesized in ﬁve steps from dimethyl 2,2,3,3-tetraﬂuorosuccinate or 4-bromo-3,3,4,4- tetraﬂuorobut-1-ene and evaluated in terms of their photophysical and LC behaviors. All four molecules exhibited PL in both dilute solutions and crystalline states. In dilute solutions, the PL wavelength varied in the range of 416–463 nm, which reﬂected the effect of substituent electron-donating/withdrawing nature on molecular orbital energy. ΦPL was maximal (0.94) for 2a and minimal (0.24) for 2c, which had the shortest λPL. The low ΦPL observed in the latter case was ascribed to self-absorption caused by the overlap of absorption and PL spectra. In the crystalline state, the PL behaviors of 2b–d were similar to those in dilute solution, whereas 2a, which had two alkyl groups at both ends, exhibited green PL with substantially red-shifted λPL. Regarding phase transition behavior, a mesophase was observed for 2a and 2d with an ethyl group at one molecular end. Only the N phase with an orientational order appeared in the case of 2a, whereas the Sm phase with both orientational and positional orders, as well as the N phase, appeared in the case of 2d. The N phase observed for 2d exhibited weak blue PL during cooling. The PL intensity increased upon the N→SmA phase transition during cooling, did not substantially change upon the SmA→SmC phase transition, and strongly increased upon the SmC→Cry phase transition on further cooling. Concomitantly, the PL color changed from dark blue to light blue, i.e., temperature-responsive PL behavior was observed. The results described herein expand the applicability of CF2CF2-containing tricyclic molecules as next-generation PL, LC, and PL-LC materials. Supplementary Materials: The following supporting information can be downloaded from https://www.mdpi.com/article/10.3390/cryst13081208/s1: Scheme S1. Synthetic procedure of 2a and 2d starting from commercially available dimethyl 2,2,3,3-tetraﬂuorosuccinate. Scheme S2. Synthetic procedure of 2b and 2c starting from commercially available 4-bromo-3,3,4,4-tetraﬂuorobut- 1-ene. Figures S1–S42. 1H-, 13C-, and 19F-NMR spectra; Figure S43. ORTEP-type crystal structure of 2b; Figures S44–S47. HOMO-1/HOMO-2, HOMO, and LUMO distributions and differential density between HOMO and LUMO; Figure S48. UV/vis absorption and PL spectra of 2a–d in CHCl3 solution; Figure S49. PL spectra of 2a–c in different solvents and related Lippert–Mataga plots; Figure S50. PL decay proﬁles of 2a–d in CHCl3 solution; Figure S51. Excitation and PL spectra of 2a–d in crystalline states; Figure S52. 3.4. PL Properties of 2d in Various Molecular Aggregation States ope ie of i a iou o e u a Agg e Compound 2d, which forms various mesop In the case of 2d, a PL band with λPL ≈466 nm appeared in the N phase; however, the related PL intensity (IN) decreased because of the accelerated nonradiative deactivation by micro-Brownian motion upon heating. The N→SmA phase transition observed upon cooling induced a 2.0-fold PL intensity increase (I/IN = 2.0) along with a slight red shift in λPL. No signiﬁcant change was observed in λPL or PL intensity upon the transition to the SmC phase, whereas the transition to the Cry phase induced a blue shift of λPL by 12 nm and a 7.8-fold increase in PL intensity relative to the N phase (I/IN = 7.8). The CIE chromaticity diagram shown in Figure 9b demonstrates that the PL color of 2d changed from dark blue to light blue owing to the phase transition-induced alteration of the molecular aggregate structure. Crystals 2023, 13, 1208 14 of 16 4. Conclusions PL decay proﬁles of 2a–d in crystalline states; Figures S53–S56. DSC thermograms and POM images for 2a–d; Figure S57. VT-PXRD patterns of 2a and 2d recorded at different temperatures; Table S1. Crystallographic data for 2b; Tables S2–S5. Cartesian coordinates for 2a–d; Tables S6–S9. Phase transition behaviors of 2a–d observed by DSC. Author Contributions: Conceptualization, H.O., S.Y., and T.K.; methodology, H.O., S.Y., and T.K.; validation, H.O., S.Y., and T.K.; formal analysis, H.O., S.Y., and T.K.; investigation, H.O., S.Y., and T.K.; resources, S.Y. and T.K.; data curation, H.O., S.Y., and T.K.; writing—original draft preparation, H.O., S.Y., and T.K.; writing—review and editing, H.O., S.Y., M.Y., and T.K.; Visualization, H.O., S.Y., and T.K.; supervision, T.K.; project administration, T.K.; funding acquisition, S.Y. and T.K. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Funding: This research received no external funding. Data Availability Statement: Data supporting the presented ﬁndings are contained within the article and Supplementary Materials. Acknowledgments: The authors express their sincere gratitude to Tosoh Finechem Corporation for providing 4-bromo-3,3,4,4-tetraﬂuorobut-1-ene and to Profs. Sakurai and Shimizu (Kyoto Institute of Technology) for help with VT-PXRD measurements. The authors acknowledge the use of equipment shared in the MEXT project to promote public utilization of advanced research infrastructure (program for supporting the introduction of the new sharing system), grant number JPMXS0421800222. Conﬂicts of Interest: The authors declare no conﬂict of interest. 15 of 16 15 of 16 Crystals 2023, 13, 1208 References 2017, 15, 1495–1509. [CrossRef] [PubMed] bearing a CF2CF2 fragment with negative dielectric anisotropy. Org. Biomol. Chem. 2017, 15, 1495 1509. [CrossRef] [PubMed] 16. Yamada, S.; Hashishita, S.; Konishi, H.; Nishi, Y.; Kubota, T.; Asai, T.; Ishihara, T.; Konno, T. New entry for ﬂuorinated carbocycles: Unprecedented 3,6-disubstituted 1,1,2,2-tetraﬂuorocyclohexane derivatives. J. Fluorine Chem. 2017, 200, 47–58. [CrossRef] 17. Yamada, S.; Tamamoto, K.; Kida, T.; Asai, T.; Ishihara, T.; Konno, T. Rational design and synthesis of a novel laterally- tetraﬂuorinated tricyclic mesogen with large negative dielectric anisotropy. Org. Biomol. Chem. 2017, 15, 9442–9454. [CrossRef] [PubMed] g g g py g 16. Yamada, S.; Hashishita, S.; Konishi, H.; Nishi, Y.; Kubota, T.; Asai, T.; Ishihara, T.; Konno, T. New entry fo Unprecedented 3,6-disubstituted 1,1,2,2-tetraﬂuorocyclohexane derivatives. J. Fluorine Chem. 2017, 200 p , , , , y J , , [ ] 17. Yamada, S.; Tamamoto, K.; Kida, T.; Asai, T.; Ishihara, T.; Konno, T. Rational design and synthesis of a novel laterally- tetraﬂuorinated tricyclic mesogen with large negative dielectric anisotropy. Org. Biomol. Chem. 2017, 15, 9442–9454. [CrossRef] [PubMed] 18. Kumon, T.; Hashishita, S.; Kida, T.; Yamada, S.; Ishihara, T.; Konno, T. Gram-scale preparation of negative-type liquid crystals with a CF2CF2-carbocycle unit via an improved short-step synthetic protocol. Beilstein J. Org. Chem. 2018, 14, 148–154. [CrossRef] 19. Ohsato, H.; Morita, M.; Yamada, S.; Agou, T.; Fukumoto, H.; Konno, T. Aggregation-induced enhanced ﬂuorescence by hydrogen bonding in π-conjugated tricarbocycles with a CF2CF2-containing cyclohexa-1,3-diene skeleton. Mol. Syst. Des. Eng. 2022, 7, 1129–1137 [CrossRef] 18. Kumon, T.; Hashishita, S.; Kida, T.; Yamada, S.; Ishihara, T.; Konno, T. Gram-scale preparation of negative-type liquid crystals with a CF2CF2-carbocycle unit via an improved short-step synthetic protocol. Beilstein J. Org. Chem. 2018, 14, 148–154. [CrossRef] 19. Ohsato, H.; Morita, M.; Yamada, S.; Agou, T.; Fukumoto, H.; Konno, T. Aggregation-induced enhanced ﬂuorescence by hydrogen bonding in π-conjugated tricarbocycles with a CF2CF2-containing cyclohexa-1,3-diene skeleton. Mol. Syst. Des. Eng. 2022, 7, 1129–1137. [CrossRef] 20. Sheldrick, G.M. SHELXT-integrated space-group and crystal-structure determination. Acta Crystallogr. Sect. A Found. Adv. 2015, 71, 3–8. [CrossRef] 21. Sheldrick, G.M. Crystal structure reﬁnement with SHELXL. Acta Crystallogr. Sect. C Struct. Chem. [PubMed] 22. Dolomanov, O.V.; Bourhis, L.J.; Gildea, R.J.; Howard, J.A.K.; Puschmann, H. OLEX2: A complete structure solution, reﬁnement, and analysis program. J. Appl. Crystallogr. 2009, 42, 339–341. [CrossRef] y p g pp y g 23. References 1. Inoue, M.; Sumii, Y.; Shibata, N. Contribution of organoﬂuorine compounds to pharmaceuticals. ACS Omega 2020, 5, 10633–10640. [CrossRef] [PubMed] ] [ ] ; Moore, P.R.; Swallow, S.; Gouverneur, V. Fluorine in medicinal chemistry. Chem. Soc. Rev. 2008, 37, 320–330. ] 2. Purser, S.; Moore, P.R.; Swallow, S.; Gouverneur, V. Fluorine in medicinal chemistry. Chem. Soc. Rev. 2008, 37, 320–330. [CrossRef] [PubMed] 3. Ogawa, Y.; Tokunaga, E.; Kobayashi, O.; Hirai, K.; Shibata, N. Current contributions of organoﬂu agrochemical industry. iScience 2020, 23, 101467. [CrossRef] 4. Jeschke, P. The unique role of ﬂuorine in the design of active ingredients for modern crop protection. ChemBioChem 2004, 5, 570–589. [CrossRef] [ ] 5. Zhou, X.; Kang, S.W.; Kumar, S.; Li, Q. Self-assembly of discotic liquid crystal porphyrin into more controllable ordered nanostructure mediated by ﬂuorophobic effect. Liq. Cryst. 2009, 36, 269–274. [CrossRef] 6. Wang, H.; Bisoyi, H.K.; Urbas, A.M.; Bunning, T.J.; Li, Q. Reversible circularly polarized reﬂection in a self-organized helical superstructure enabled by a visible-light-driven axially chiral molecular switch. J. Am. Chem. Soc. 2019, 141, 8078–8082. [CrossRef] 6. Wang, H.; Bisoyi, H.K.; Urbas, A.M.; Bunning, T.J.; Li, Q. Reversible circularly polarized reﬂection in a self-organized helical superstructure enabled by a visible-light-driven axially chiral molecular switch. J. Am. Chem. Soc. 2019, 141, 8078–8082. [CrossRef] 7. Wang, H.; Bisoyi, H.K.; Li, B.X.; McConney, M.E.; Bunning, T.J.; Li, Q. Visible-light-driven halogen bond donor based molecular switches: From reversible unwinding to handedness inversion in self-organized soft helical superstructures. Angew. Chem. Int. Ed. 2020, 59, 2684–2687. [CrossRef] p y g y J , , [ ] 7. Wang, H.; Bisoyi, H.K.; Li, B.X.; McConney, M.E.; Bunning, T.J.; Li, Q. Visible-light-driven halogen bond donor based molecular switches: From reversible unwinding to handedness inversion in self-organized soft helical superstructures. Angew. Chem. Int. Ed. 2020, 59, 2684–2687. [CrossRef] 8. Kirsch, P. Modern Fluoroorganic Chemistry: Synthesis, Reactivity, Applications, 2nd ed.; Kirsh, P., Ed.; WILEY-VCH: Weinheim, Germany, 2013; pp. 247–298. y pp 9. Hird, M. Fluorinated liquid crystals–properties and applications. Chem. Soc. Rev. 2007, 36, 2070–20 q y p p pp 10. O’Hagan, D. Understanding organoﬂuorine chemistry. An introduction to the C–F bond. Chem. Soc. Rev. 2008, 37, 308–319. [CrossRef] [PubMed] [ ] [ ] 11. Bondi, A. van der Waals volumes and radii. J. Phys. Chem. 1964, 68, 441–451. [CrossRef] [ ] [ ] 11. Bondi, A. van der Waals volumes and radii. J. Phys. Chem. 1964, 68, 441–451. References [CrossRef] y 12. Konno, T. Triﬂuoromethylated internal alkynes: Versatile building blocks for the preparation of var molecules. Synlett 2014, 25, 1350–1370, and references cited therein. [CrossRef] 13. Ohsato, H.; Kawauchi, K.; Yamada, S.; Konno, T. Diverse synthetic transformations using 4-bromo-3,3,4,4-tetraﬂuorobut-1-ene and its applications in the preparation of CF2CF2-containing sugars, liquid crystals, and light-emitting materials. Chem. Rec. 2023, 23, e202300080, and references cited therein. [CrossRef] 14. Yamada, S.; Konno, T. Development of donor-π-acceptor-type ﬂuorinated tolanes as compact condensed phase luminophores and applications in photoluminescent liquid-crystalline molecules. Chem. Rec. 2023, 23, e202300094, and references cited therein. [CrossRef] [PubMed] pp p [CrossRef] [PubMed] 15. Yamada, S.; Hashishita, S.; Asai, T.; Ishihara, T.; Konno, T. Design, synthesis and evaluation of new ﬂuorinated liquid crystals bearing a CF2CF2 fragment with negative dielectric anisotropy. Org. Biomol. Chem. 2017, 15, 1495–1509. [CrossRef] [PubMed] 15. Yamada, S.; Hashishita, S.; Asai, T.; Ishihara, T.; Konno, T. Design, synthesis and evaluation of new ﬂuorinated liquid crystals bearing a CF2CF2 fragment with negative dielectric anisotropy. Org. Biomol. Chem. 2017, 15, 1495–1509. [CrossRef] [PubMed] 16. Yamada, S.; Hashishita, S.; Konishi, H.; Nishi, Y.; Kubota, T.; Asai, T.; Ishihara, T.; Konno, T. New entry for ﬂuorinated carbocycles: Unprecedented 3,6-disubstituted 1,1,2,2-tetraﬂuorocyclohexane derivatives. J. Fluorine Chem. 2017, 200, 47–58. [CrossRef] 17. Yamada, S.; Tamamoto, K.; Kida, T.; Asai, T.; Ishihara, T.; Konno, T. Rational design and synthesis of a novel laterally- tetraﬂuorinated tricyclic mesogen with large negative dielectric anisotropy Org Biomol Chem 2017 15 9442 9454 [CrossRef] 15. Yamada, S.; Hashishita, S.; Asai, T.; Ishihara, T.; Konno, T. Design, synthesis and evaluation of new ﬂuorinated liquid crystals bearing a CF2CF2 fragment with negative dielectric anisotropy. Org. Biomol. Chem. 2017, 15, 1495–1509. [CrossRef] [PubMed] 16. Yamada, S.; Hashishita, S.; Konishi, H.; Nishi, Y.; Kubota, T.; Asai, T.; Ishihara, T.; Konno, T. New entry for ﬂuorinated carbocycles: Unprecedented 3,6-disubstituted 1,1,2,2-tetraﬂuorocyclohexane derivatives. J. Fluorine Chem. 2017, 200, 47–58. [CrossRef] 15. Yamada, S.; Hashishita, S.; Asai, T.; Ishihara, T.; Konno, T. Design, synthesis and evaluation of new ﬂuorinated liquid crystals bearing a CF2CF2 fragment with negative dielectric anisotropy. Org. Biomol. Chem. 2017, 15, 1495–1509. [CrossRef] [PubMed] 16. Yamada, S.; Hashishita, S.; Konishi, H.; Nishi, Y.; Kubota, T.; Asai, T.; Ishihara, T.; Konno, T. New entry for ﬂuorinated carbocycles: 15. Yamada, S.; Hashishita, S.; Asai, T.; Ishihara, T.; Konno, T. Design, synthesis and evaluation of new ﬂuorinated liquid crystals bearing a CF2CF2 fragment with negative dielectric anisotropy. Org. Biomol. Chem. 28. Tsuzuki, S.; Fujii, A. Nature and physical origin of CH/π interaction: Signiﬁcant difference from conventional hydrogen bonds. Phys. Chem. Chem. Phys. 2008, 10, 2584–2594. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. y y , , [ ] [ ] 29. Mataga, N.; Kaifu, Y.; Koizumi, M. The solvent effect on the ﬂuorescence spectrum changes of solute-solvent interactions during the lifetime of the excited solute molecule. Bull. Chem. Soc. Jpn. 1955, 28, 690–691. [CrossRef] 30. Mataga, N.; Kaifu, Y.; Koizumi, M. Solvent effects on ﬂuorescence spectra and dipole moments of excited molecules. Bull. Chem. Soc. Jpn. 1956, 29, 465–470. [CrossRef] y y , , [ ] [ ] 29. Mataga, N.; Kaifu, Y.; Koizumi, M. The solvent effect on the ﬂuorescence spectrum changes of solute-solvent interactions during the lifetime of the excited solute molecule. Bull. Chem. Soc. Jpn. 1955, 28, 690–691. [CrossRef] 30 M t N K if Y K i i M S l t ff t ﬂ t d di l t f it d l l B ll Ch 30. Mataga, N.; Kaifu, Y.; Koizumi, M. Solvent effects on ﬂuorescence spectra and dipole moments of excit Soc. Jpn. 1956, 29, 465–470. [CrossRef] y y , , [ ] [ ] 29. Mataga, N.; Kaifu, Y.; Koizumi, M. The solvent effect on the ﬂuorescence spectrum changes of solute-so the lifetime of the excited solute molecule. Bull. Chem. Soc. Jpn. 1955, 28, 690–691. [CrossRef] the lifetime of the excited solute molecule. Bull. Chem. Soc. Jpn. 1955, 28, 690 691. [CrossRef] 30. Mataga, N.; Kaifu, Y.; Koizumi, M. Solvent effects on ﬂuorescence spectra and dipole moments of excited molecules. Bull. Chem. Soc. Jpn. 1956, 29, 465–470. [CrossRef] References Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.; Petersson, G.A.; Nakatsuji, H.; et al. Gaussian 16, Revision B.01; Gaussian, Inc.: Wallingford, CT, USA, 2016. 24. Hohenstein, E.G.; Chill, S.T.; Sherrill, C.D. Assessment of the performance of the M05-2X and M06-2X exchange-correlation functionals for noncovalent interactions in biomolecules. J. Chem. Theory Comput. 2008, 4, 1996–2000. [CrossRef] [PubMed] 25. Li, H.; Jensen, J.H. Improving the efﬁciency and convergence of geometry optimization with the polarizable continuum model: 24. Hohenstein, E.G.; Chill, S.T.; Sherrill, C.D. Assessment of the performance of the M05-2X and M06-2X exchange-correlation functionals for noncovalent interactions in biomolecules. J. Chem. Theory Comput. 2008, 4, 1996–2000. [CrossRef] [PubMed] 25. Li, H.; Jensen, J.H. Improving the efﬁciency and convergence of geometry optimization with the polarizable continuum model: New energy gradients and molecular surface tesselation. J. Comput. Chem. 2004, 25, 1449–1462. [CrossRef] [PubMed] 26. Rybalova, T.V.; Bagryanskaya, I.Y. C–F···π, F···H, and F···F intermolecular interactions and F-aggregations: Role in crystal , ; J , J p g y g g y p p New energy gradients and molecular surface tesselation. J. Comput. Chem. 2004, 25, 1449–1462. [CrossRef] [PubMed] 26. Rybalova, T.V.; Bagryanskaya, I.Y. C–F···π, F···H, and F···F intermolecular interactions and F-aggregations: Role in crystal engineering of ﬂuoroorganic compounds. J. Struct. Chem. 2009, 50, 741–753. [CrossRef] g g g p 27. Kawahara, S.; Tsuzuki, S.; Uchimaru, T. Theoretical study of the C–F/π interaction: Attractive interac alkane and an electron-deﬁcient π-system. J. Phys. Chem. A 2004, 108, 6744–6749. [CrossRef] Crystals 2023, 13, 1208 16 of 16 Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://openalex.org/W4251963450	https://peerj.com/articles/4709v0.5/submission	English	null	Peer Review #3 of "Validity and reliability of the WIMU inertial device for the assessment of the vertical jump (v0.3)"	null	2,018	cc-by	6,405	Manuscript to be reviewed Validity and reliability of the WIMU inertial device for the assessment of the vertical jump José Pino 1 Javier Garcia Corresp., 2, 3 Sergio J Ibañez 3 José Pino 1 , Javier Garcia Corresp., 2, 3 , Sergio J Ibañez 3 1 Faculty of Sports Sciences, Universidad de Murcia, Spain 2 Facultad de Educación, Universidad Autonoma de Chile, Chile 3 Faculty of Sports Sciences, University of Extremadura, Spain 1 Faculty of Sports Sciences, Universidad de Murcia, Spain 2 Facultad de Educación, Universidad Autonoma de Chile, Chile 3 Faculty of Sports Sciences, University of Extremadura, Spain Corresponding Author: Javier Garcia Email address: jagaru@unex.es The aim of this study was to test the validity and reliability of the inertial device WIMU (Realtrack Systems SL, SPA) for the assessment of the vertical jump, counter movement jump and squat jump. Fifteen soccer players were evaluated in two identical sessions separated by one week. In each session, participants performed three jumps of each type. The flight time was quantified by the inertial device WIMU and by a force platform (Twin plates, Globus) at the same time. For the analysis of reliability of the flight time of the CMJ and the SJ, the intraclass correlation coefficient (ICC) was used. The calculation of the concurrent validity was performed by using the Pearson correlation coefficient (r). This analysis was complemented with the realization of the Bland-Altman plots. For the analysis of reliability, the coefficient of variation and the standard error of the means were calculated. The analysis presented a high validity and reliability of the device. The results show the inertial device WIMU (Realtrack Systems SL, SPA) as a useful tool for measuring the jump capacity of the athletes, presenting immediate results in real time, on any type of surface and in a simple way since it does not need cables. PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed 1 ABSTRACT 2 3 BACKGROUNDː The aim of this study was to test the validity and reliability of the inertial device 4 WIMUTM (Realtrack Systems SL, SPA) for the assessment of different vertical jump modalities. 5 METHODSː Fifteen soccer players were evaluated into two identical sessions separated by one 6 week. In each session, participants performed three jumps of each type. The flight time was 7 quantified by the inertial device WIMUTM and by a force plate (Twin plates, Globus) at the same 8 time. The intraclass correlation coefficient (ICC) was used for the analysis of reliability of the 9 flight time of the Counter Movement Jump and the Squat Jump. For the analysis of reliability, the 10 coefficient of variation and the standard error of the means were calculated. The calculation of the 11 concurrent validity was performed by using the Pearson correlation coefficient (r). This analysis 12 was complemented with the realization of Bland-Altman plots. 13 RESULTSː The analysis presented a high validity (CMJ: ICC= 0.97, r= 0.95; SJ: ICC= 0.96, r: 14 0 93) and reliability (CMJ: CV 3 1; SJ: r 2 5) of the device ABSTRACT 3 BACKGROUNDː The aim of this study was to test the validity and reliability of the inertial device 4 WIMUTM (Realtrack Systems SL, SPA) for the assessment of different vertical jump modalities. 3 BACKGROUNDː The aim of this study was to test the validity and reliability of the inertial device 4 WIMUTM (Realtrack Systems SL, SPA) for the assessment of different vertical jump modalities. 5 METHODSː Fifteen soccer players were evaluated into two identical sessions separated by one 6 week. In each session, participants performed three jumps of each type. The flight time was 7 quantified by the inertial device WIMUTM and by a force plate (Twin plates, Globus) at the same 8 time. The intraclass correlation coefficient (ICC) was used for the analysis of reliability of the 9 flight time of the Counter Movement Jump and the Squat Jump. For the analysis of reliability, the 10 coefficient of variation and the standard error of the means were calculated. The calculation of the 11 concurrent validity was performed by using the Pearson correlation coefficient (r). This analysis 12 was complemented with the realization of Bland-Altman plots. 13 RESULTSː The analysis presented a high validity (CMJ: ICC= 0.97, r= 0.95; SJ: ICC= 0.96, r: 14 0.93) and reliability (CMJ: CV= 3.1; SJ: r= 2.5) of the device. 13 RESULTSː The analysis presented a high validity (CMJ: ICC= 0.97, r= 0.95; SJ: ICC= 0.96, r: 14 0.93) and reliability (CMJ: CV= 3.1; SJ: r= 2.5) of the device. 15 CONCLUSIONSː The results showed an almost perfect relation between the inertial device 16 WIMUTM and the contact platform. WIMUTM is an useful tool for measuring the jump capacity of 17 the athletes, presenting results in real time in a simple way since it does not need cables. 15 CONCLUSIONSː The results showed an almost perfect relation between the inertial device 16 WIMUTM and the contact platform. WIMUTM is an useful tool for measuring the jump capacity of 17 the athletes, presenting results in real time in a simple way since it does not need cables. 18 18 19 INTRODUCTION 20 One of the most commonly used indicators to assess physical fitness in different in both the general 21 population and high performance sports, and for identifying young talents (1) is the vertical jump. 22 Similarly, it has been used to predict the risk of injury and is associated with muscle power or 23 neuromuscular fatigue (2, 3). In fact, jump performance plays a crucial role in both the 24 predominantly anaerobic and aerobic disciplines, as well as in the mixed disciplines (4, 5). This 25 can be measured by different tools or systems (force plates, video analysis, photoelectric cells, 26 contact mat, video cameras, etc.) (6-8) and more recently with phone apps as MyJump (1). 27 Although they are highly accurate measurement systems, they involve some drawbacks. For 28 example, force plates are difficult to use in field test while the video analysis does not generate 29 immediate results (1). Similarly, video analysis has a major drawback, especially if only one 30 camera is used. If the plane that records the camera moves, the accuracy will be affected (9). There 31 are several means for evaluating vertical jump, therefore it is important to choose the one that 32 bestfits to complete the task, regarding precision, cost, reliability or duration (6). Accelerometers, 33 however, are a highly portable, lightweight, easy to use and accessible tool for almost any coach 34 (7, 10). 35 The vertical jump height is measured as the difference between the position of the center of mass 36 of the individual in the starting position (a standing position) and its position at the maximum 37 height (11). This height can be measured according to the flight time (12). Flight time is the raw 38 data that both devices recorded and is used in several studies (13-15) and reported as the more 39 accurate method to calculate jump height (16). The estimation of jumping performance is done 40 after the event. Also, the jump height is an indirect technique to measure jump height or muscle 41 power and as (8) suggest, the different methods used in the calculation could be led to a systematic 35 The vertical jump height is measured as the difference between the position of the center of mass 36 of the individual in the starting position (a standing position) and its position at the maximum 37 height (11). PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed 42 error. Not all the measurement systems are equally accurate, in fact it has been reported that, due 43 to the different angles of the knees and ankles at the time of the jump starts, there may be an error 44 of 2.2 cm (15). WIMUTM is an inertial device designed for monitoring the physical activity for 45 athletes of different disciplines. 46 The control of the validity and reliability of the WIMUTM device is necessary for its acceptance as 47 an accurate measurement system for assessing the vertical jump. The validity and reliability refer 48 to the device’s ability to measure what it is designed to measure (validity), and to always measure 49 the same event inthe same way (reliability) (17). Therefore, the objective of this study was to 50 evaluate the validity and reliability of the WIMUTM device for assessing SJ and CMJ. It has been 51 hypothesized that the WIMUTM device will display a high validity and reliability for measuring 52 the different vertical jumps used in the study. 19 INTRODUCTION This height can be measured according to the flight time (12). Flight time is the raw 38 data that both devices recorded and is used in several studies (13-15) and reported as the more 39 accurate method to calculate jump height (16). The estimation of jumping performance is done 40 after the event. Also, the jump height is an indirect technique to measure jump height or muscle 41 power and as (8) suggest, the different methods used in the calculation could be led to a systematic PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) 54 Experimental approach to the problem 55 The aim of this study was to test the validity and reliability of the device WIMU (Realtrack 56 Systems SL, SPA) to measure the performance in the vertical jump. In order to do that, the tests 57 for measuring the flight time during a slow cycle of stretching and shortening the muscle (CMJ) 58 and the explosive concentric muscle actions (SJ) were conducted by 15 subjects into two identical 59 sessions. Data collections were separated by one week. Jumps were measured with the WIMUTM 60 device and the force plate at the same time (Twin Plates, Globus Sport and Health Technologies 61 LLC, ITA), considered as the gold standard. The election of flight time as variable of comparison 62 is because is the direct data that provide both systems. Jump height is an indirect measurement of PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed 63 a formula based on flight time. WIMUTM was attached to a belt and fixed on the lower back 64 (Choukuo et al., 2014). 63 a formula based on flight time. WIMUTM was attached to a belt and fixed on the lower back 64 (Choukuo et al., 2014). 66 The participants in the study were 15 soccer players at early stages (N=15, age = 14.74 ± 0.23 67 years old, height = 164.34 ± 4.04 cm, weight = 65.7 ± 3.35 kg). All the participants belonged to 68 the same soccer team participating at a regional category in Spain. The team trains three sessions 69 of two hours per week, with a total volume of six hours of training. The jumps were performed 70 during the competitive season. All subjects and their parents were informed about the study 71 procedures and their possible risks, giving their consent to participate before testing. The ethics 72 committee of the University approved the study (nº 67/2017). PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed 85 position. After that, the examiner made them wait for three seconds and then cheered them up 86 verbally to jump as high as possible, without making any type of counter movement. In all jumps 87 at takeoff, participants were asked to leave the floor at the same time and always with their knees 88 extended, and land in similarly extended position (7, 18). A total of six jumps were evaluated for 89 each participant, measured with the inertial device WIMU and the contact plate (figure 1). 85 position. After that, the examiner made them wait for three seconds and then cheered them up 86 verbally to jump as high as possible, without making any type of counter movement. In all jumps 87 at takeoff, participants were asked to leave the floor at the same time and always with their knees 88 extended, and land in similarly extended position (7, 18). A total of six jumps were evaluated for 89 each participant, measured with the inertial device WIMU and the contact plate (figure 1). 90 WIMUTM (Figure 2). For this study, trials have been recorded by the accelerometer at a sampling 91 frequency of 1000 Hz. Accelerometers and gyroscopes integrated in the device were used to 92 correct and calculate the vertical acceleration recordings. All the information is represented in its 93 specific software QÜIKOTM, which allows automatic analysis. A mathematic algorithm was 94 developed in order to calculate flight time using total acceleration signal (vector sum of three axis 95 of the accelerometer and information recorded by gyroscopes). Jump height can be calculated with 96 Bosco, Luhtanen (12) equation. 90 WIMUTM (Figure 2). For this study, trials have been recorded by the accelerometer at a sampling 91 frequency of 1000 Hz. Accelerometers and gyroscopes integrated in the device were used to 92 correct and calculate the vertical acceleration recordings. All the information is represented in its 93 specific software QÜIKOTM, which allows automatic analysis. A mathematic algorithm was 94 developed in order to calculate flight time using total acceleration signal (vector sum of three axis 95 of the accelerometer and information recorded by gyroscopes). Jump height can be calculated with 96 Bosco, Luhtanen (12) equation. 97 Force Plate. 73 Procedure. 74 In both sessions of data acquisition, conditions were the same. During the week of training, the 75 jumps were performed three days after the last game. Before jumping, warming-up was equal in 76 both sessions. A standard warm-up of 15 minutes, consisting of five minutes of continuous 77 running, specific mobility of the lower body for another 5 minutes, active stretching of the lower 78 extremities, and vertical jumps for the last 5 minutes (1, 7). Previously, the subjects had been 79 instructed in performing the different jumps by the same examiner. 80 Testing procedures: The subjects performed each CMJ starting from a static standing position, 81 with their arms on their hips and with their knees extended during the flight. Once in position, the 82 subjects were instructed (approximately a 90° angle) as quickly as possible and then jump as high 83 as possible. In SJ, subjects began in the same standing position as in the CMJ, also with their hands 84 on their hips. In this position, they were asked to flex their hips (aprox. 90 °) and maintain this PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed The force plate “Twin Plates” (Globus Sport and Health Technologies LLC, ITA) 98 (240 x 400 mm) records the data at a frequency of 1000 Hz with an error of less than 1%. It was 99 used at the same time as the inertial device WIMUTM. The plate was connected to a laptop 100 computer for the real-time feedback via the software “Ergo system” (Globus Sport and Health 101 Technologies LLC, ITA). Flight time is automatic calculated by the force platform software. 103 A first descriptive analysis with averages and standard deviation was performed to characterize 104 the sample. The intraclass correlation coefficient (ICC) (2.1) was used for the analysis of reliability 105 of the flight time of the CMJ and the SJ. The Coefficient of Variation (CV) was used to analyze 106 the reliability of the instrument. The absolute reliability was determined by calculating the indexes PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed 128 The descriptive analysis showed that the flight time registered by the inertial device WIMUTMwas 129 almost identical than the one registered by the plate (Table 1), but there were no significant 130 differences between both devices. The average flight time in the CMJ was: force plate = 437.62 131 ms; WIMUTM device = 437.31 ms. For the SJ, flight time was: force plate = 416.72 ms; WIMUTM 132 device = 416.11 ms. The results showed an almost perfect relation between the inertial device 133 WIMUTM and the contact platform, both in the CMJ (ICC (2.1) = 0.97, 95% CI: 0.96–0.98, P < 134 0.001) as in the SJ (ICC (2.1) = 0.96, 95% CI: 0.94–0.97, P < 0.001). The reliability analysis 135 showed a SEM% of 2.2% for the CMJ and of 1.4% for the SJ. The SRD% is 6.2% for the CMJ 136 and 3.9% for the SJ. The CVs were very low for both periods of flight time of jump (CMJ: 3.1%; 137 SJ: 2.5%) (Table 1). In addition, the Pearson correlation coefficient is almost perfect in both cases 138 (r >.9) (Table 1 and Figure 3 and 4). 139 Bland-Altman plots show an average systematic trend of 1.31 (0.29%) milliseconds between the 140 force plate and the inertial device in the CMJ; and an average systematic trend of 0.61 (0.13%) in 141 the SJ. The trend is the average difference between the two measures. As these values are positive, 142 the force plate gets higher values than the inertial device (Figures 5 and 6). Manuscript to be reviewed 107 Standard Error of Measurement (SEM) [SEM= SD√(1-ICC) where SD is the standard deviation 108 of day one and day two] and the Minimum Real Change (SRD) [SRD= 1.96 x √2xSEM] (17). Both 109 SEM and the SRD were calculated in absolute terms and in percentage for easier interpretation 110 and comparison with other measuring devices. Similarly, both the SEM and the SRD became 111 percentages to facilitate comparison with other studies. The percentages were calculated according 112 to the following equation: SEM% = (SEM/mean flight time of the two repetitions)·100; and 113 SRD%=(SRD/mean flight time of the two repetitions)·100. All the reliability tests were performed 114 for the force plate and the inertial device. Similarly, the SEM and the SRD are indicators that 115 express the absolute reliability of the device, shown at the same measurement unit of the 116 instrument. Besides, the SEM results are highly independent from the population under study, not 117 as the ICC (17). To extrapolate the results of the study, it has been chosen to express these indexes 118 in absolute terms, thus being able to compare the results with those of other instruments more 119 easily (19). 120 The calculation of the concurrent validity was performed by using the Pearson correlation 121 coefficient (r). This analysis was complemented with the realization of the Bland-Altman plots. 122 This representation indicates the degree of agreement between both instruments, not only the 123 degree of relation (20). 124 The level of significance was established at P < 0.05. All of the analyses were performed by using 125 the statistical package SPSS 21.0 for Windows (IBM Co., USA), except for the Bland-Altman 126 plots, which were performed by using the software Graphpad Prism (Graphpad, Inc., USA). 124 The level of significance was established at P < 0.05. All of the analyses were performed by using 125 the statistical package SPSS 21.0 for Windows (IBM Co., USA), except for the Bland-Altman 126 plots, which were performed by using the software Graphpad Prism (Graphpad, Inc., USA). PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) 143 DISCUSSION 144 The objective of the present study was to evaluate the validity and reliability of the WIMUTM 145 device for the assessment of the flight time of squat jump and countermovement jump. The results 146 show the high concurrent validity and reliability that the WIMUTM device presents compared with 147 a force plate. The WIMUTM device has proven to be a useful instrument. The results from this 148 study conclude that WIMUTM provides similar flight times in squat jump and countermovement 149 jump to the criterion method. PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed 150 In previous studies, correlation coefficient has been the main method to assess validity and 151 reliability. It has been suggested that other methods, as Bland-Altman plots, provide more relevant 152 information about agreement between two measures (20). For example, it has been probed that 153 Myotest and Optojump have a high correlation according to ICCs (0.98), but with a systematic 154 bias of around 7 cm in vertical jump (7). In fact, a constant error will not be detected on the 155 correlation analysis and, therefore, it cannot be concluded that devices are accurate (9). The data 156 presented in the Bland-Altman plots show that most of the jumps are close to the media of the 157 differences between instruments in both jump modalities, showing a high level of agreement (20) 158 and a high correlation between them. The ICC presents a great precision of measurements (>0.93). 159 Similarly, by analyzing the reliability of the devices, excellent CVs (< 10%) can be appreciated 160 (11, 19). The SEM% shows very low percentages of absolute error (21). 161 Several studies have examined the validity of different methods of analysis of the vertical jump in 162 comparison to force plates. Previous studies have misreported the vertical jump performance, 163 found discrepancies between devices compared with force plates of around 10 cm measuring flight 164 time (16, 22) or measuring jump height, from recorded flight time, in contact mats (16, 23). It has 165 also been reported that mastery of the jumping technique may affect the jump performance (23). 166 On the other hand, contact mats measure flight time according to the moment when the subject 167 leaves the ground, and miss some data of the initial rise of the centre of mass before the take-off 168 (23). Also, the differences could be due to the different ascending and descending phases and 169 landing (16). In fact, some studies report as limitation the lack of gyroscope that can detect the 170 inclination of the body in take-off and landing (7, 25). 171 Studies have shown lower values for the validity (ICCs scores), of their devices to those found 172 with the WIMUTM. Choukou, Laffaye (11) studies the validity of the Myotest in comparison to a 171 Studies have shown lower values for the validity (ICCs scores), of their devices to those found 172 with the WIMUTM. Manuscript to be reviewed Force plates are 178 considered the most accurate tools for measuring the vertical jump, since they allow identifying 179 the moment of take-off with great precision (25, 27, 28). This device, force platform, is enable to 180 measure, in eccentric and concentric phase of the movement, the force and power production (23), 181 allowing more detailed analysis of subjects´ training. WIMUTM allows providing a similar analysis 182 of force plates (as results concluded) and, contrary to force plates, WIMUTM still records data of 183 more variables during flight time (such as G force in take-off and landing or inclination, among 184 others), which makes the analysis more detailed. These results demonstrate the ability of the 185 WIMUTM system to play the measurements that a force plate makes, having the advantage of being 186 lightweight and portable. These measurements are valid for both the CMJ and the SJ. Due to its 187 properties of size and weight, this device can be easily placed in any segment of the body to 188 measure the vertical jump (center of mass, hips, back, lower body, etc.). The data collection is 189 done with a single computer wirelessly connected to the device, so a great time preparing for the 190 test is not needed. In addition, subjects do not have to be connected by any cable or have to take 191 off and land in a delimited area. Moreover, thanks to the specific software, the device provides 192 immediate feedback. 193 CONCLUSIONS 193 CONCLUSIONS 194 Because of the ability of this device to collect data on different types of jump and the relevant 195 information from these, WIMUTM is a valuable tool for controlling the training and competition. 194 Because of the ability of this device to collect data on different types of jump and the relevant 195 information from these, WIMUTM is a valuable tool for controlling the training and competition. Manuscript to be reviewed Choukou, Laffaye (11) studies the validity of the Myotest in comparison to a PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed 173 force plate, finding ICC values between 0.86-0.96. Casartelli, Müller (7), by studying the same 174 device, finds similar results. The analysis of reliability shows very low CVs, better than those 175 found when the reliability of mobile applications was developed (26), photoelectric cells (27), or 176 other inertial devices such as accelerometers (7, 11). When the results are compared to the ones 177 obtained with high-speed cameras, the results are practically the same (25). Force plates are 178 considered the most accurate tools for measuring the vertical jump, since they allow identifying 179 the moment of take-off with great precision (25, 27, 28). This device, force platform, is enable to 180 measure, in eccentric and concentric phase of the movement, the force and power production (23), 181 allowing more detailed analysis of subjects´ training. WIMUTM allows providing a similar analysis 182 of force plates (as results concluded) and, contrary to force plates, WIMUTM still records data of 183 more variables during flight time (such as G force in take-off and landing or inclination, among 184 others), which makes the analysis more detailed. These results demonstrate the ability of the 185 WIMUTM system to play the measurements that a force plate makes, having the advantage of being 186 lightweight and portable. These measurements are valid for both the CMJ and the SJ. Due to its 187 properties of size and weight, this device can be easily placed in any segment of the body to 188 measure the vertical jump (center of mass, hips, back, lower body, etc.). The data collection is 189 done with a single computer wirelessly connected to the device, so a great time preparing for the 190 test is not needed. In addition, subjects do not have to be connected by any cable or have to take 191 off and land in a delimited area. Moreover, thanks to the specific software, the device provides 192 immediate feedback. 193 CONCLUSIONS 173 force plate, finding ICC values between 0.86-0.96. Casartelli, Müller (7), by studying the same 174 device, finds similar results. The analysis of reliability shows very low CVs, better than those 175 found when the reliability of mobile applications was developed (26), photoelectric cells (27), or 176 other inertial devices such as accelerometers (7, 11). When the results are compared to the ones 177 obtained with high-speed cameras, the results are practically the same (25). Manuscript to be reviewed 196 Throughout control of the jumps, changes produced by a particular training program can be 197 detected, being able to redirect or modify it according to the results (7). It can also be used to 198 determine the fatigue accumulated during the same work, adapting the breaks between sets for a 199 full recovery (10, 29). The WIMUTM device does not need cables, so it greatly facilitates its 200 placement in the subject's body as well as the freedom of movement. 200 placement in the subject's body as well as the freedom of movement. 201 REFERENCES. 202 1. Balsalobre-Fernández C, Glaister M, Lockey RA. The validity and reliability of an iPhone app for 203 measuring vertical jump performance. Journal of sports sciences. 2015;33(15):1574-1579. 202 1. Balsalobre-Fernández C, Glaister M, Lockey RA. The validity and reliability of an iPhone app for 203 measuring vertical jump performance Journal of sports sciences 2015 33(15) 1574 1579 204 2. Jimenez-Reyes P, Pareja-Blanco F, Rodriguez-Rosell D, Marques MC, Gonzalez-Badillo JJ. 205 Maximal Velocity as a Discriminating Factor in the Performance of Loaded Squat Jumps. International 206 journal of sports physiology and performance. 2016;11(2):227-234. 207 3. Stevenson JH, Beattie CS, Schwartz JB, Busconi BD. Assessing the Effectiveness of Neuromuscula 208 Training Programs in Reducing the Incidence of Anterior Cruciate Ligament Injuries in Female Athletes A 209 Systematic Review. The American journal of sports medicine. 2015;43(2):482-490. 207 3. Stevenson JH, Beattie CS, Schwartz JB, Busconi BD. Assessing the Effectiveness of Neuromuscular 208 Training Programs in Reducing the Incidence of Anterior Cruciate Ligament Injuries in Female Athletes A 209 Systematic Review. The American journal of sports medicine. 2015;43(2):482-490. 210 4. Brumitt J, Heiderscheit BC, Manske RC, Niemuth PE, Rauh MJ. Off-season training habits and 211 preseason functional test measures of division iii collegiate athletes: a descriptive report. International 212 journal of sports physical therapy. 2014;9(4):447-455. 213 5. Hartman MJ, Clark B, Bemben DA, Kilgore JL, Bemben MG. Comparisons between twice-daily 214 and once-daily training sessions in male weight lifters. International journal of sports physiology and 215 performance. 2007;2(2):159-169. 216 6. Bui HT, Farinas MI, Fortin AM, Comtois AS, Leone M. Comparison and analysis of three different 217 methods to evaluate vertical jump height. Clinical physiology and functional imaging. 2015;35(3):203- 218 209. 219 7. Casartelli N, Müller R, Maffiuletti NA. Validity and reliability of the myotest accelerometric 220 system for the assessment of vertical jump height. The Journal of Strength & Conditioning Research. 221 2010;24(11):3186-3193. 222 8. Dias JA, Dal Pupo J, Reis DC, Borges L, Santos SG, Moro AR,Borges Jr NG. Validity of two methods 223 for estimation of vertical jump height. The Journal of Strength & Conditioning Research. 224 2011;25(7):2034-2039. 225 9. Magnúsdóttir Á, Karlsson B. Comparing three devices for jump height measurement in a 226 heterogeneous group of subjects. The Journal of Strength & Conditioning Research. 2014;28(10):2837- 227 2844. 228 10. Sato K, Smith SL, Sands WA. Validation of an accelerometer for measurin 228 10. Sato K, Smith SL, Sands WA. 201 REFERENCES. Validation of an accelerometer for measuring sport performance. 229 The Journal of Strength & Conditioning Research. 2009;23(1):341-347. The Journal of Strength & Conditioning Research. 2009;23(1):341-347. 229 The Journal of Strength & Conditioning Research. 2009;23(1):341-347. 230 11. Choukou M-A, Laffaye G, Taiar R. Reliability and validity of an accelerome 230 11. Choukou M-A, Laffaye G, Taiar R. Reliability and validity of an accelerometric system for 231 assessing vertical jumping performance. Biology of sport. 2014;31(1):55-62. 230 11. Choukou M A, Laffaye G, Taiar R. Reliability and validity of an accelerometric system for 231 assessing vertical jumping performance. Biology of sport. 2014;31(1):55-62. 232 12. Bosco C, Luhtanen P, Komi PV. A simple method for measurement of mechanical power in 233 jumping. European journal of applied physiology and occupational physiology. 1983;50(2):273-282. 234 13. Balsalobre-Fernández C, Tejero-González CM, del Campo-Vecino J, Bavaresco N. The concurrent 235 validity and reliability of a low-cost, high-speed camera-based method for measuring the flight time of 236 vertical jumps. The Journal of Strength & Conditioning Research. 2014;28(2):528-533. 234 13. Balsalobre-Fernández C, Tejero-González CM, del Campo-Vecino J, Bavaresco N. The concurrent 235 validity and reliability of a low-cost, high-speed camera-based method for measuring the flight time of 236 vertical jumps. The Journal of Strength & Conditioning Research. 2014;28(2):528 PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed 237 14. García-López J, Morante JC, Ogueta-Alday A, Rodríguez-Marroyo JA. The type of mat (Contact vs. 238 Photocell) affects vertical jump height estimated from flight time. The Journal of Strength & 239 Conditioning Research. 2013;27(4):1162-1167. 240 15. Garcia-Lopez J, Peleteiro J, Rodgriguez-Marroyo J, Morante J, Herrero J, Villa J. The validation of 241 a new method that measures contact and flight times during vertical jump. International journal of 242 sports medicine. 2005;26(4):294-302. 240 15. Garcia-Lopez J, Peleteiro J, Rodgriguez-Marroyo J, Morante J, Herrero J, Villa J. The validation of 241 a new method that measures contact and flight times during vertical jump. International journal of 242 sports medicine. 2005;26(4):294-302. 243 16. Aragón LF. Evaluation of four vertical jump tests: Methodology, reliability, validity, and accuracy. 244 Measurement in physical education and exercise science. 2000;4(4):215-228. 243 16. Aragón LF. Evaluation of four vertical jump tests: Methodology, reliability, validity, and accuracy. 244 Measurement in physical education and exercise science. 2000;4(4):215-228. 245 17. Weir JP. Quantifying test-retest reliability using the intraclass correlation coefficient and the 246 SEM. Journal of strength and conditioning research. 2005;19(1):231-240. 245 17. Weir JP. Quantifying test-retest reliability using the intraclass correlation coefficient and the 246 SEM. Journal of strength and conditioning research. 2005;19(1):231-240. 247 18. Häkkinen K, Komi PV. Effect of explosive type strength training on electromyographic and force 248 production characteristics of leg extensors muscles during concentric and various stretch-shortening 249 cycle exercises. Scand J Sports Sci. 1985;7(2):65-76. 247 18. Häkkinen K, Komi PV. Effect of explosive type strength training on electromyographic and force 248 production characteristics of leg extensors muscles during concentric and various stretch-shortening 249 cycle exercises. Scand J Sports Sci. 1985;7(2):65-76. 250 19. Atkinson G, Nevill AM. Statistical methods for assessing measurement error (reliability) in 251 variables relevant to sports medicine. Sports medicine. 1998;26(4):217-238. 250 19. Atkinson G, Nevill AM. Statistical methods for assessing measurement error (reliability) in 251 variables relevant to sports medicine. Sports medicine. 1998;26(4):217-238. 252 20. Bland JM, Altman D. Statistical methods for assessing agreement between two methods of 253 clinical measurement. The lancet. 1986;327(8476):307-310. 252 20. Bland JM, Altman D. Statistical methods for assessing agreement between two methods of 253 clinical measurement. The lancet. 1986;327(8476):307-310. 254 21. Collado-Mateo D, Adsuar JC, Olivares PR, Cano-Plasencia R, Gusi N. Using a dry electrode EEG 255 device during balance tasks in healthy young-adult males: Test-retest reliability analysis. Somatosensory 256 & motor research. 2015;32(4):219-226. 257 22. Manuscript to be reviewed Moir GL. Three different methods of calculating vertical jump height from force platform data in 258 men and women. Measurement in Physical Education and Exercise Science. 2008;12(4):207-218. 259 23. Buckthorpe M, Morris J, Folland JP. Validity of vertical jump measurement devices. Journal of 257 22. Moir GL. Three different methods of calculating vertical jump height from force platform data in 258 men and women. Measurement in Physical Education and Exercise Science. 2008;12(4):207-218. 259 23. Buckthorpe M, Morris J, Folland JP. Validity of vertical jump measurement devices. Journal of 260 sports sciences. 2012;30(1):63-69. p ; ( ) 261 24. Leard JS, Cirillo MA, Katsnelson E, Kimiatek DA, Miller TW, Trebincevic K, Garbalosa J. Validity of 262 two alternative systems for measuring vertical jump height. The Journal of Strength & Conditioning 263 Research. 2007;21(4):1296-1299. PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) 285 Figure 6. Bland-Altman plots for force plate and WIMU in CMJ. The central line represents the absolute average difference 286 between instruments. Short-dashed lines represent the upper and lower 95% limits of agreement. 283 Figure 5. Bland-Altman plots for force plate and WIMU in SJ. The central line represents the absolute average difference between 284 instruments. Short-dashed lines represent the upper and lower 95% limits of agreement. Manuscript to be reviewed Manuscript to be reviewed Manuscript to be reviewed Figure 1 A graphic example of flight time data of WIMU and force plate. Squat jump recorded by WIMU and Force platform. Squat jump recorded by WIMU and Force platform. Squat jump recorded by WIMU and Force platform. Manuscript to be reviewed 282 Figure 4. Concurrent validity between force plate and WIMU: SJ. 283 Figure 5. Bland-Altman plots for force plate and WIMU in SJ. The central line represents the absolute average difference between 284 instruments. Short-dashed lines represent the upper and lower 95% limits of agreement. 285 Figure 6. Bland-Altman plots for force plate and WIMU in CMJ. The central line represents the absolute average difference 286 between instruments. Short-dashed lines represent the upper and lower 95% limits of agreement. 285 Figure 6. Bland-Altman plots for force plate and WIMU in CMJ. The central line represents the absolute average difference 286 between instruments. Short-dashed lines represent the upper and lower 95% limits of agreement. PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed Manuscript to be reviewed Figure 2 WIMU inertial device WIMU inertial device Photo credit: Sergio Ibáñez and Javier García. PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed Manuscript to be reviewed Concurrent validity between force plate and WIMU: CMJ. Concurrent validity between force plate and WIMU: CMJ. PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Figure 3 Concurrent validity between force plate and WIMU: CMJ. Correlation of CMJ (ms). Manuscript to be reviewed Figure 4 Concurrent validity between force plate and WIMU: SJ. Concurrent validity between force plate and WIMU: SJ. Correlation of SJ (ms). PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed Manuscript to be reviewed Figure 5 Bland-Altman plots for force plate and WIMU in SJ. PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed Figure 6 Bland-Altman plots for force plate and WIMU in CMJ. Bland-Altman plots for force plate and WIMU in SJ. The central line represents the absolute average difference between instruments. Short- dashed lines represent the upper and lower 95% limits of agreement. Counter Movement Jump (Flight time) (ms). PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Bland-Altman plots for force plate and WIMU in CMJ. The central line represents the absolute average difference between instruments. Short- dashed lines represent the upper and lower 95% limits of agreement. Squat Jump (Flight time) (ms). PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed Table 1(on next page) Table 1(on next page) Table 1(on next page) Descriptive Statistics (Flight time in ms and jumping performance in cm), intraclass correlation coefficients, reliability of the inertial device and flight time correlation of the CMJ and the SJ. PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018) Manuscript to be reviewed Manuscript to be reviewed Table 1. Descriptive Statistics(Flight time in ms and jumping performance in cms), intraclass correlation coefficients, reliability of the inertial device and flight time correlation of the CMJ and the SJ WIMU GLOBUS ICC IC 95% r SEM SEM% SRD SRD% CV% Time 436.31±13.70 437.62±14.91 CMJ Performance 23.34±0.02 23.48±0.02 0.97 0.96–0.98 0.95 9.69 2.2 26.85 6.2 3.1 Time 416.11±10.70 416.72±12.50 SJ Performance 21.23±0.01 21.29±0.02 0.96 0.94–0.97 0.93 5.60 1.4 16.35 3.9 2.5 1 PeerJ reviewing PDF \| (2017:06:18499:4:0:NEW 9 Apr 2018)
https://openalex.org/W2744892005	https://www.scielo.br/j/eagri/a/dPR6q5yK8q4n8ygY8MZYLGr/?lang=en&format=pdf	English	null	TEMPORAL VARIABILITY IN ACTIVE REFLECTANCE SENSOR-MEASURED NDVI IN SOYBEAN AND WHEAT CROPS	Engenharia agrícola	2,017	cc-by	5,762	2Corresponding author. Western Paraná State University (UNIOESTE), Cascavel - PR, Brazil. E-mail: eduardo souza@unioeste br 2Corresponding author. Western Paraná State University (UNIOESTE), Cascavel - PR, Brazil. E-mail: eduardo souza@unioeste br 2Corresponding author. Western Paraná State University (UNIOESTE), Cascavel - PR, Brazil. E-mail: eduardo.souza@unioeste.br ABSTRACT: Optimization of N management is one of the great challenges to be overcome in grain production, as it is directly related to productivity and can also cause environmental damage. Precision agriculture aims to solve this problem by applying nitrogen fertilizer at varying rates. Reflectance sensors are instruments capable of estimating N needs in various crops, including grain crops. However, it is not clear how these sensors perform under varying solar radiation and cloud cover, due to a lack of research on their temporal variability. Thus, this study examined the temporal variability of the NDVI (normalized difference vegetation index), as measured by an active reflectance sensor, in both soybean and wheat crops. The NDVI data were collected using a GreenSeeker sensor every 15 minutes over 12 or 14 consecutive hours. Incident solar radiation was recorded using an Instrutherm MES-100 pyranometer. In all experiments in soybean and wheat, NDVI was negatively influenced by irradiation, showing higher values at the beginning and end of the day. Changes in cloud cover also affected NDVI values during the experiments. KEYWORDS: precision agriculture, remote sensing, vegetation index, NDVI _________________________ 1 Paraná Federal Institute (IFPR), Foz do Iguaçu-PR, Brazil. 3 Federal Technological University of Paraná (UTFPR), Medianeira-PR, Brazil. 4 Federal Technological University of Paraná (UTFPR), Santa Helena-PR, Brazil. Received in: 5-16-2016 Accepted in: 3-8-2017 Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 _________________________ 1 Paraná Federal Institute (IFPR), Foz do Iguaçu-PR, Brazil. Doi:http://dx.doi.org/10.1590/1809-4430-Eng.Agric.v37n4p771-781/2017 HUMBERTO M. BENEDUZZI1, EDUARDO G. SOUZA2, CLAUDIO L. BAZZI3, Doi:http://dx.doi.org/10.1590/1809-4430-Eng.Agric.v37n4p771-781/2017 Journal of the Brazilian Association of Agricultural Engineering ISSN: 1809-4430 (on-line) Journal of the Brazilian Association of Agricultural Engineering ISSN: 1809-4430 (on-line) Journal of the Brazilian Association of Agricultural Engineering ISSN: 1809-4430 (on-line) INTRODUCTION Optimizing N management remains one of the greatest challenges in grain production. Because productivity is directly related to N availability (BARRACLOUGH et al., 2010), N is often applied in excess (DELLINGER et al., 2008, SCHMIDT et al., 2011), causing financial losses and environmental damage due to leaching and groundwater contamination (BURKART & JAMES, 1999; HONG et al., 2007). Among the technologies used in precision agriculture is the application of fertilizer at varying rates, in which the appropriate amount of fertilizer is identified and applied at each point of the cultivated area (THRIKAWALA et al., 1999; MULLA, 2013; SAPKOTA et al., 2014). This approach can not only increase productivity but also reduce its environmental impact (SCHARF et al., 2011; LI et al., 2014). Reflectance sensors are a type of instrument that can estimate the N needs of various types of crops, including grain crops (SCHMIDT et al., 2009; WINTERHALTER et al., 2013). These types of instruments have logistical and economic advantages since they can be used to rapidly cover large areas, quickly generating results at a lower cost compared to other techniques (SCHARF et al., 2002, SCHARF et al., 2011). Several studies have already shown that passive sensors are influenced by the time of day, since their measurements depend on the level of solar radiation (SOUZA et al., 2004; SOUZA et al., 2010; ERDLE et al., 2011). On the other hand, active sensors use their own light source and so have the ability to work at any time of day without interference from solar radiation or atmospheric variations (TRIMBLE, 2013). However, there are studies indicating that even active sensors can be influenced by the time of day and the weather conditions, creating distortions in the collected data (SCHARF et al., 2010; KIPP et al., 2014; OLIVEIRA & SCHARF, 2014; ELSAYED et al., 2015). Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 772 Humberto M. Beneduzzi, Eduardo G. Souza, Claudio L. Bazzi, et al. To confront this challenge, this work investigated the changes in the measurements made by an active reflectance sensor in soybean and wheat crops during different times of day, with consequent variation in solar radiation and cloud cover. MATERIAL AND METHODS This research was conducted in soybean and wheat fields at different stages of development in an agricultural area located outside of the city of Céu Azul - Paraná state - Brazil, whose approximate central geographic location is 25°06'32” S, 53°49'55” W. The average altitude of this area is 752 m, and the area comprises approximately 15.5 ha on a soil classified as a typic Dystroferric Red Latosol (EMBRAPA, 2013). The area has been cultivated using no-till practices for over 10 years in a rotation of soybean, wheat, corn and oat crops grown for commercial purposes. Data relating to soybean (variety Dom Mario 5958 RR2) were collected in the R1 and R2 stages, while the data for wheat (variety Catuara) were obtained in stages V3, V8 and V10.5. In the soybean, the R1 stage indicates the beginning of the reproductive phase, and this is marked by the beginning of bloom, when plants have at least one open flower at any node. At the R2 stage full bloom occurs, with an open flower in one of the last two stem nodes with fully developed leaf (EMBRAPA, 2007). The wheat in V3 stage presents tillers formed and spiral wound sheets. At the V8 stage the flag leaf is visible, but still curled or rolled up. The V10.5 indicates the end of the heading stage, when the plants presents all the ears outside the sheaths (EMBRAPA, 2014). To collect normalized difference vegetation index (NDVI) data, a GreenSeeker 505 active optical sensor was used. This sensor detects electromagnetic radiation in the (visible) red wavelength of 656 nm and in the near-infrared at 774 nm (NTECH, 2008). The NDVI is then calculated using the following equation: (1) where, where, where, (1) NIR is the radiation reflectance measured by the sensor in the near-infrared wavelength, and RED is the radiation reflectance measured by the sensor in the red wavelength. NIR is the radiation reflectance measured by the sensor in the near-infrared wavelength, and RED is the radiation reflectance measured by the sensor in the red wavelength Incident solar radiation (or irradiation) measurements were performed with an Instrutherm MES-100 pyranometer. The incident radiation parameter was used to estimate the influence of clouds in each of the data collection periods. We used the ITI (instantaneous transparency index), proposed by SOUZA et. al. (2006), which aims to indicate the percentage of incident radiation of expected irradiation for a given time. MATERIAL AND METHODS The ITI was calculated using the methodology developed by SOUZA et al. (2006), according to [eq. (2)]: The ITI was calculated using the methodology developed by SOUZA et al. (2006), according t [ (2)] (2) where, (2) where, I is the expected theoretical radiation, and I0 is the real radiation measured by the pyranometer at the same time. To estimate the theoretical radiation for the date of data collection within each plant growth stage, a polynomial regression function was developed for each day. This function was based on actual radiation values measured during periods without clouds. The data were normalized to allow the inclusion of the data of different magnitudes on the same graph. To this end, the average method proposed by SWINDELL (1997) was used: Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 Temporal variability in active reflectance sensor-measured NDVI in soy and wheat crops Temporal variability in active reflectance sensor-measured NDVI in soy and wheat crops 773 (3) (3) where, Vn is the normalized value for a particular sample, found by dividing the sample value by the average sample value; V is the value of the variable in case I, and V is the value of the variable in case I, and V is the value of the variable in case I, and j represents the number of cases. j represents the number of cases. Data were collected using the same methods for both wheat and soybean. Data were collected with a stationary GreenSeeker sensor positioned parallel to the ground, with the same position maintained throughout the day (Figure 1). Preliminary tests were conducted to ensure that the support used did not influence sensor readings. The sensor was maintained at a distance of 80 cm from the crop canopy, which was within the range recommended by the instrument manufacturer (NTECH, 2008). FIGURE 1. GreenSeeker sensor mounted on soybean in R1 stage The times used for the beginning and end of data collection took into account the time of year in order to capture the entire insolation period. In soybean, this period was from 7 am to 9 pm, and in wheat, it was 7 am to 7 pm. During each stage researched, measurements were performed every 15 minutes. Thus, the data were always collected at minutes 0, 15, 30 and 45 of each hour. The FIGURE 1. GreenSeeker sensor mounted on soybean in R1 stage FIGURE 1. MATERIAL AND METHODS NDVI = a + bstage + cirr + d(irrITI) where, NDVI = a + bstage + cirr + d(irrITI) where, NDVI = a + bstage + cirr + d(irrITI) (4) where, NDVI = a + bstage + cirr + d(irrITI) (4) where, (4) a, b, c, d are parameters generated by the multiple regression model, a, b, c, d are parameters generated by the multiple regression model, stage represents the developmental stage of the crop, stage represents the developmental stage of the crop, irr is the irradiance (W m-2), and ITI refers to the instantaneous transparency index. The model defined by exploratory data analysis for wheat (Equation 5) indicates that NDVI as a dependent variable is influenced by the stage of development of the crop, solar irradiation, ITI and ITIITI. NDVI = a + bstage + cirr + dITI + eITI2 where, NDVI = a + bstage + cirr + dITI + eITI2 where, (5) a, b, c, d and e are the parameters generated by the multiple regression model, stage represents the developmental stage of the crop, a, b, c, d and e are the parameters generated by the multiple regression model, t t th d l t l t f th a, b, c, d and e are the parameters generated by the multiple regression model, a, b, c, d and e are the parameters generated by the mu stage represents the developmental stage of the crop, stage represents the developmental stage of the crop, irr is irradiance (W m-2), and ITI is the instantaneous transparency index. MATERIAL AND METHODS GreenSeeker sensor mounted on soybean in R1 stage The times used for the beginning and end of data collection took into account the time of year in order to capture the entire insolation period. In soybean, this period was from 7 am to 9 pm, and in wheat, it was 7 am to 7 pm. During each stage researched, measurements were performed every 15 minutes. Thus, the data were always collected at minutes 0, 15, 30 and 45 of each hour. The NDVI values obtained for each sample were composed of the average of 100 readings taken by the GreenSeeker sensor continuously over 10 seconds. Analyzing the influence of crop row orientation on vegetation indices, SOUZA et al. (2004) concluded that NDVI variability increased when crops were sown in north-south rows. Thus, in this experiment, data collection was carried out on crops growing in north-south rows. The GreenSeeker sensor was positioned over the row, perpendicular to sunrise and sunset, with the front facing north. Soon after data collection with the GreenSeeker, a pyranometer positioned above the sensor was used to measure incident radiation. Data collected in the field were statistically analyzed through exploratory analysis. Coefficients of variation (CV) were ranked according to PIMENTEL & GARCIA (2002). CVs ≤ 10% were deemed low (homoscedastic), while 10% < CV ≤ 20% was average, 20% < CV ≤ 30% was high, and CV > 30% was very high (heteroscedastic). Regression analysis was performed using Statistica 12.0 software (STATSOFT, 2014), with NDVI modeled as a function of the stage of development and irradiation. We used the regression best subsets method, in which the adjusted Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 Humberto M. Beneduzzi, Eduardo G. Souza, Claudio L. Bazzi, et al. 774 rto M. Beneduzzi, Eduardo G. Souza, Claudio L. Bazzi, et coefficient of determination (adjusted R2) was used as a selection criterion of the best model among all possible subsets (MONTGOMERY et al., 2012). coefficient of determination (adjusted R2) was used as a selection criterion of the best model among all possible subsets (MONTGOMERY et al., 2012). The following model (Equation 4) was defined by exploratory analysis conducted using data from soybean, and it assumes that the dependent variable, NDVI, is influenced by the developmental stage of the crop, solar irradiation and irradiation ITI. Soybean The NDVI coefficients of variation determined in soybean stages R1 and R2 were 2.8% and 1.8%, respectively, which can be considered low (Table 1). The variability of NDVI was significantly smaller than the variability of its constituent bands, what is to be expected (TUCKER, 1979; PINTER et al., 1985). The red band (656 nm) showed a much higher coefficient of variation than that for the near-infrared band (774 nm). TABLE 1. Descriptive statistics for the data grouped by soybean stage. TABLE 1. Descriptive statistics for the data grouped by soybean stage. In both growth stages, the soybean NDVI curve showed similar behavior throughout the day, with higher values at the beginning and end of the day and smaller values close to noon, when solar irradiation was greater (Figure 2a). This observation is in agreement with the results obtained by OLIVEIRA & SCHARF (2014). It was also determined that there was a greater dispersion of NDVI Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 ability in active reflectance sensor-measured NDVI in soy and wheat crops 7 Temporal variability in active reflectance sensor-measured NDVI in soy and wheat crops 775 values in the R1 stage than in the R2 stage, which indicates that as crop development continues, NDVI variability decreases (Table 1, Figures 2a and 2b). a) NDVI variation b) Boxplot graph FIGURE 2. a) Variation in NDVI throughout the day for each soybean stage; b) Boxplot graphics for NDVI grouped by stage. b) Boxplot graph FIGURE 2. a) Variation in NDVI throughout the day for each soybean stage; b) Boxplot graphics for NDVI grouped by stage. The R2 values obtained for the theoretical irradiation fitting curve for the R1 and R2 soybean stages were greater than 99%. The ITI was calculated as a function of the time of day (Figures 3a and 3b). Irradiation was further reduced by cloud cover during the experiment in R2 stage compared with R1. a) R1 Stage b) R2 Stage FIGURE 3. Measured irradiation, theoretical irradiation and instantaneous transparency index (ITI) for R1 (a) and R2 (b) soybean development stages. a) R1 Stage b) R2 Stage FIGURE 3. Measured irradiation, theoretical irradiation and instantaneous transparency index (ITI) for R1 (a) and R2 (b) soybean development stages. Soybean Comparing Figures 2a and 4a, it can be seen that during the experiment on R1 stage, at approximately 15:00, a sharp drop in irradiation occurred, with a simultaneous and marked increase in NDVI, indicating a clear association between these two variables. This can also be seen in the experiment on R2 stage, by comparing the graphs in Figures 2a and 4b, particularly at 15:30, when again, a sharp drop in irradiation occurs together with a peak in NDVI. The variation of the solar radiation throughout the day affected the red band more than the near-infrared band (Figures 4a and 4b). In addition, sudden changes in cloud cover captured by ITI also held more influence over the red band. This response agrees with what SOUZA et. al. (2006) have shown. The NDVI was inversely correlated with ITI (Figure 4c) and irradiation (Figure 4d), in both stages. From the model that was produced from regression analysis (Equation 6), the R2 value was 0.842, i.e., 84.2% of the variability in NDVI can be explained by the stage of development, irradiation and ITI. The fact that the variability in NDVI cannot be fully explained using these three Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 Humberto M. Beneduzzi, Eduardo G. Souza, Claudio L. Bazzi, et al. 776 variables suggests that additional research may need to consider other variables, such as temperature and humidity. NDVI= 0.9153 - 0.004280stage - 0.0001123irr + 0.00006211irrITI (6) 0 0,5 1 1,5 2 07:00 09:00 11:00 13:00 15:00 17:00 19:00 21:00 Normalized values Time of day NDVI Red NIR Irradiation ITI a) R1 Stage 0 0,5 1 1,5 2 2,5 07:00 09:00 11:00 13:00 15:00 17:00 19:00 21:00 Normalized values Time of day NDVI Red NIR Irradiation ITI b) R2 Stage R1 stage ITI:NDVI: r2 = 0.1695; y = 0.9088 - 0.0297x R2 stage ITI:NDVI: r2 = 0.0362; y = 0.8878 - 0.0104x ITI NDVI 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 0,83 0,84 0,85 0,86 0,87 0,88 0,89 0,90 0,91 0,92 0,93 R1 stage R2 stage c) NDVI by ITI R1 stage Rad:NDVI: y = 0,9118 - 5,6018E-5x; r2 = 0,8743 R2 stage Rad:NDVI: y = 0,8973 - 3,8769E-5x; r2 = 0,6786 Irradiation (W m-2) NDVI -200 0 200 400 600 800 1000 1200 0,83 0,84 0,85 0,86 0,87 0,88 0,89 0,90 0,91 0,92 0,93 R1 stage R2 stage d) NDVI by solar irradiation FIGURE 4. Soybean NDVI, red, near-infrared (NIR), irradiation and instantaneous transparency index (ITI) as a function of time of day during R1 (a) and R2 (b) stages of soybean development (normalized data); NDVI as a function of ITI (c); Correlation between NDVI and solar irradiation (d). 0 0,5 1 1,5 2 2,5 07:00 09:00 11:00 13:00 15:00 17:00 19:00 21:00 Normalized values Time of day NDVI Red NIR Irradiation ITI b) R2 Stage 0 0,5 1 1,5 2 07:00 09:00 11:00 13:00 15:00 17:00 19:00 21:00 Normalized values Time of day NDVI Red NIR Irradiation ITI a) R1 Stage R2 Stage a) R1 Stage R1 stage ITI:NDVI: r2 = 0.1695; y = 0.9088 - 0.0297x R2 stage ITI:NDVI: r2 = 0.0362; y = 0.8878 - 0.0104x ITI NDVI 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 0,83 0,84 0,85 0,86 0,87 0,88 0,89 0,90 0,91 0,92 0,93 R1 stage R2 stage c) NDVI by ITI R1 stage Rad:NDVI: y = 0,9118 - 5,6018E-5x; r2 = 0,8743 R2 stage Rad:NDVI: y = 0,8973 - 3,8769E-5x; r2 = 0,6786 Irradiation (W m-2) NDVI -200 0 200 400 600 800 1000 1200 0,83 0,84 0,85 0,86 0,87 0,88 0,89 0,90 0,91 0,92 0,93 R1 stage R2 stage c) NDVI by ITI d) NDVI by solar irradiation FIGURE 4. NDVI, red, near-infrared (NIR), irradiation and instantaneous transparency index (ITI) as a function of time of day during R1 (a) and R2 (b) stages of soybean development (normalized data); NDVI as a function of ITI (c); Correlation between NDVI and solar irradiation (d). Wheat In wheat, the NDVI coefficients of variation recorded in stages V3, V8 and V10.5 were 2.8%, 3.3% and 6.8%, respectively. The rise in variability as the season progressed contrasts with the trend observed in soybean. Furthermore, the average NDVI values were lower in later stages (Table 2, Figure 5b). The decline in NDVI in stage V10.5 can be partially explained by the attenuation of the green tint inherent to a maturing wheat crop, which in turn causes increased reflectance in the red band. Nonetheless, the reason why the NDVI in V8 stage was less than that for V3 was not clear. Similar to the results obtained in soybean, the variability recorded in the red band (656 nm) was much higher than that in the near-infrared band (774 nm). The variability of NDVI, in turn, was less than that of the individual bands at stages V3 and V8 but greater than the variability in red at stage V10.5. Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 777 TABLE 2. Descriptive statistics for the data grouped by wheat stage. TABLE 2. Descriptive statistics for the data grouped by wheat stage. In all three wheat experiments, the NDVI values were higher at the beginning and end of the day, similar to what was observed in soybean. In stage V10.5, NDVI was well below that recorded in the other two stages; it also fell more quickly with each passing hour, followed by a sharp rise at the end of the day (Figure 5a). In all three wheat experiments, the NDVI values were higher at the beginning and end of the day, similar to what was observed in soybean. In stage V10.5, NDVI was well below that recorded in the other two stages; it also fell more quickly with each passing hour, followed by a sharp rise at the end of the day (Figure 5a). FIGURE 5. a) Variation in NDVI throughout the day for each stage; b) Boxplots for NDVI grouped by wheat stage. FIGURE 5. a) Variation in NDVI throughout the day for each stage; b) Boxplots for NDVI grouped by wheat stage. The R2 values obtained for the adjustment curve of the theoretical irradiation (Figures 6a, 6b and 6c) in stages V3, V8 and V10.5 were all above 99.9%. The instantaneous transparency index (ITI) was then calculated as a function of time of day. Wheat Cloud cover reduced irradiation the most in stage V3, followed by stages V10.5 and V8. The R2 values obtained for the adjustment curve of the theoretical irradiation (Figures 6a, 6b and 6c) in stages V3, V8 and V10.5 were all above 99.9%. The instantaneous transparency index (ITI) was then calculated as a function of time of day. Cloud cover reduced irradiation the most in stage V3, followed by stages V10.5 and V8. Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 Humberto M. Beneduzzi, Eduardo G. Souza, Claudio L. Bazzi, et al. 778 0 0,2 0,4 0,6 0,8 1 1,2 1,4 0 200 400 600 800 1000 07:00 09:00 11:00 13:00 15:00 17:00 Instantaneous Transparency Index Irradiation (W m-2) Measured Irradiation Theoretical Irradiation ITI a) V3 Stage 0,0 0,2 0,4 0,6 0,8 1,0 1,2 0 200 400 600 800 1000 07:00 09:00 11:00 13:00 15:00 17:00 Instantaneous Transparency Index Irradiation (W m-2) Measured Irradiation Theoretical Irradiation ITI b) V8 Stage 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 0 200 400 600 800 1000 07:00 09:00 11:00 13:00 15:00 17:00 Instantaneous Transparency Index Irradiation (W m-2) Measured Irradiation Theoretical Irradiation ITI c) V10.5 Stage FIGURE 6. Measured irradiation, theoretical irradiation and instantaneous transparency index (ITI) for V3 (a), V8 (b) and V10.5 (c) wheat development stages. 0,0 0,2 0,4 0,6 0,8 1,0 1,2 0 200 400 600 800 1000 07:00 09:00 11:00 13:00 15:00 17:00 Instantaneous Transparency Index Irradiation (W m-2) Measured Irradiation Theoretical Irradiation ITI 0 0,2 0,4 0,6 0,8 1 1,2 1,4 0 200 400 600 800 1000 07:00 09:00 11:00 13:00 15:00 17:00 Instantaneous Transparency Index Irradiation (W m-2) Measured Irradiation Theoretical Irradiation ITI a) V3 Stage b) V8 Stage 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 0 200 400 600 800 1000 07:00 09:00 11:00 13:00 15:00 17:00 Instantaneous Transparency Index Irradiation (W m-2) Measured Irradiation Theoretical Irradiation ITI c) V10.5 Stage FIGURE 6. Measured irradiation, theoretical irradiation and instantaneous transparency index (ITI) f V3 ( ) V8 (b) d V10 5 ( ) h t d l t t c) V10.5 Stage FIGURE 6. Measured irradiation, theoretical irradiation and instantaneous transparency index (ITI) for V3 (a), V8 (b) and V10.5 (c) wheat development stages. Wheat In wheat, the variation in the solar radiation throughout the day also affected the red band more than near-infrared band (Figures 7a, 7b and 7c), similar to what was observed in soybean. Abrupt changes in cloud cover, indicated by the ITI readings, also held more influence over the red band. The NDVI was inversely correlated with irradiation (Figure 8b) in all three stages, and it showed a direct correlation with ITI only in stage V10.5 (Figure 8a). The regression model (Equation 7) had an R2 value of 0.803; i.e., 80.3% of the variability in NDVI could be explained by the developmental stage of the crop, irradiation and ITI. Again, the fact that the variability of NDVI cannot be explained by these variables indicates the need for studies that also consider other variables, such as temperature and humidity. NDVI = 0.871578 - 0.076001stage - 0.000100irr + 0.231785ITI - 0.114141ITI2 (7) NDVI = 0.871578 - 0.076001stage - 0.000100irr + 0.231785ITI - 0.114141ITI2 (7) Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 NDVI = 0.871578 - 0.076001stage - 0.000100irr + 0.231785ITI - 0.114141ITI2 (7) 0 0,5 1 1,5 2 2,5 3 07:00 09:00 11:00 13:00 15:00 17:00 19:00 Normalized values Time of day NDVI Red NIR Irradiation ITI a) V3 Stage 0 0,5 1 1,5 2 2,5 07:00 09:00 11:00 13:00 15:00 17:00 19:00 Normalized values Time of day NDVI Red NIR Irradiation ITI b) V8 Stage 0 0,5 1 1,5 2 2,5 3 07:00 09:00 11:00 13:00 15:00 17:00 19:00 Normalized values Time of day NDVI Red NIR Irradiation ITI a) V3 Stage 0 0,5 1 1,5 2 2,5 07:00 09:00 11:00 13:00 15:00 17:00 19:00 Normalized values Time of day NDVI Red NIR Irradiation ITI b) V8 Stage Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 201 Temporal variability in active reflectance sensor-measured NDVI in soy and wheat crops 779 0 0,5 1 1,5 2 2,5 3 07:00 09:00 11:00 13:00 15:00 17:00 19:00 Normalized values Time of day NDVI Red NIR Irradiation ITI c) V10.5 Stage FIGURE 7. Normalized data collected in wheat V3 (a), V8 (b) and V10.5 (c) stages: NDVI, red, near-infrared, irradiation and ITI. FIGURE 7. Normalized data collected in wheat V3 (a), V8 (b) and V10.5 (c) stages: NDVI, red, near-infrared, irradiation and ITI. CONCLUSIONS NDVI values measured by the GreenSeeker sensor on soybean and wheat crops throughout the day were negatively influenced by irradiation and were higher in the early morning and at the end of the day. The NDVI was also affected by changes in cloud cover during the experiments, which had a positive influence on the values for wheat but a negative influence on the values for soybean. Among the variables studied, solar irradiation best explained the variation in NDVI for both crops. Wheat V3 stage ICI:NDVI: y = 0,8366 + 0,012x; r2 = 0,0261 V8 stage ICI:NDVI: y = 0,8116 + 0,0154x; r2 = 0,0089 V10.5 stage ICI:NDVI: y = 0,6243 + 0,0942x; r2 = 0,2431 ITI NDVI 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 0,60 0,62 0,64 0,66 0,68 0,70 0,72 0,74 0,76 0,78 0,80 0,82 0,84 0,86 0,88 0,90 V3 stage V8 stage V10.5 stage a) NDVI by ITI V3 stage Rad:NDVI: y = 0,8641 - 8,5307E-5x; r2 = 0,5979 V8 stage Rad:NDVI: y = 0,8612 - 0,0001x; r2 = 0,9498 V10.5 stage Rad:NDVI: y = 0,7376 - 0,0001x; r2 = 0,2944 Irradiation (W m-2) NDVI -100 0 100 200 300 400 500 600 700 800 900 1000 0,60 0,62 0,64 0,66 0,68 0,70 0,72 0,74 0,76 0,78 0,80 0,82 0,84 0,86 0,88 0,90 V3 stage V8stage V10.5 stage b) NDVI by solar irradiation FIGURE 8. NDVI as a function of ITI (a); Correlation between NDVI and solar irradiation (b) V3 stage ICI:NDVI: y = 0,8366 + 0,012x; r2 = 0,0261 V8 stage ICI:NDVI: y = 0,8116 + 0,0154x; r2 = 0,0089 V10.5 stage ICI:NDVI: y = 0,6243 + 0,0942x; r2 = 0,2431 ITI NDVI 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 0,60 0,62 0,64 0,66 0,68 0,70 0,72 0,74 0,76 0,78 0,80 0,82 0,84 0,86 0,88 0,90 V3 stage V8 stage V10.5 stage V3 stage Rad:NDVI: y = 0,8641 - 8,5307E-5x; r2 = 0,5979 V8 stage Rad:NDVI: y = 0,8612 - 0,0001x; r2 = 0,9498 V10.5 stage Rad:NDVI: y = 0,7376 - 0,0001x; r2 = 0,2944 Irradiation (W m-2) NDVI -100 0 100 200 300 400 500 600 700 800 900 1000 0,60 0,62 0,64 0,66 0,68 0,70 0,72 0,74 0,76 0,78 0,80 0,82 0,84 0,86 0,88 0,90 V3 stage V8stage V10.5 stage b) NDVI by solar irradiation b) NDVI by solar irradiation FIGURE 8. NDVI as a function of ITI (a); Correlation between NDVI and solar irradiation (b) REFERENCES Barraclough PB, Howarth JR, Jones J, Lopez-Bellido R, Parmar S, Shepherd CE, Hawkesford MJ (2010) Nitrogen efficiency of wheat: genotypic and environmental variation and prospects for improvement. European Journal of Agronomy 33(1):1-11. Barraclough PB, Howarth JR, Jones J, Lopez-Bellido R, Parmar S, Shepherd CE, Hawkesford MJ (2010) Nitrogen efficiency of wheat: genotypic and environmental variation and prospects for improvement. European Journal of Agronomy 33(1):1-11. Burkart MR, James DE (1999) Agricultural nitrogen contribution to hypoxia in the Gulf of Mexico. Journal of Environmental Quality 28(3):850-859. Burkart MR, James DE (1999) Agricultural nitrogen contribution to hypoxia in the Gulf of Mexico. Journal of Environmental Quality 28(3):850-859. Dellinger AE, Schmidt JP, Beegle DB (2008) Developing Nitrogen Fertilizer Recommendations for Corn Using an Active Sensor. Agronomy Journal 100(6):1546-1552. Dellinger AE, Schmidt JP, Beegle DB (2008) Developing Nitrogen Fertilizer Recommendations for Corn Using an Active Sensor. Agronomy Journal 100(6):1546-1552. Elsayed S, Rischbeck P, Schmidhalter U (2015) Comparing the performance of active and passive reflectance sensors to assess the normalized relative canopy temperature and grain yield of drought- stressed barley cultivars. Field Crops Research 177:148-160. Elsayed S, Rischbeck P, Schmidhalter U (2015) Comparing the performance of active and passive reflectance sensors to assess the normalized relative canopy temperature and grain yield of drought- stressed barley cultivars. Field Crops Research 177:148-160. Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 Humberto M. Beneduzzi, Eduardo G. Souza, Claudio L. Bazzi, et al. 780 EMBRAPA - Empresa Brasileira de Pesquisa Agropecuária (2007) Ecofisiologia da Soja. Londrina, Embrapa. (Circular Técnica 48) EMBRAPA - Empresa Brasileira de Pesquisa Agropecuária (2013) Sistema brasileiro de classificação de solos. Brasília, Embrapa. 3 ed. EMBRAPA - Empresa Brasileira de Pesquisa Agropecuária (2014) Informações Técnicas para Trigo e Triticale - Safra 2015. Brasília, Embrapa. Erdle K, Mistele B, Schmidhalter U (2011) Comparison of active and passive spectral sensors in discriminating biomass parameters and nitrogen status in wheat cultivars. Field Crops Research 124(1):74-84. Hong N, Scharf PC, Davis JG, Kitchen NR, Sudduth KA (2007) Economically optimal nitrogen rate reduces soil residual nitrate. Journal of Environmental Quality 36(2):354-362. Kipp S, Mistele B, Schmidhalter U (2014) The performance of active spectral reflectance sensors as influenced by measuring distance, device temperature and light intensity. Computers and Electronics in Agriculture 100:24-33. Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 REFERENCES Li F, Miao Y, Feng G, Yuan F, Yue S, Gao X, Liu Y, Liu B, Ustin SL, Chen X (2014) Improving estimation of summer maize nitrogen status with red edge-based spectral vegetation indices. Field Crops Research 157:111-123. Montgomery DC, Peck EA, Vining GG (2012) Introduction to linear regression analysis. New York, John Wiley & Sons, 5 ed. 672p. Mulla DJ (2013) Twenty five years of remote sensing in precision agriculture: Key advances and remaining knowledge gaps. Biosystems Engineering 114(4):358-371. NTech Industries (2008) GreenSeeker model 500 series datasheet. Available: http://www.ssmgps.com/pdf/DataSheet_Model_500_GS_Sensor.pdf. Accessed: Nov 21, 2014. Oliveira LF, Scharf PC (2014) Diurnal variability in reflectance measurements from cotton. Crop Science 54(4):1769-1781. Pimentel FG, Garcia GH (2002) Estatística aplicada a experimentos agronômicos e florestais. Piracicaba, Biblioteca de Ciências Agrárias Luiz de Queiroz, 307p. Pinter PJ, Jackson RD, Ezra CE (1985) Sun-angle and canopy-architecture effects on the spectral reflectance of six wheat cultivars. International Journal of Remote Sensing 6(12):1813-1825. Sapkota TB, Majumdar K, Jat ML, Kumar A, Bishnoi DK, Mcdonald AJ, Pampolino M (2014) Precision nutrient management in conservation agriculture based wheat production of Northwest India: Profitability, nutrient use efficiency and environmental footprint. Field Crops Research 155:233-244. Scharf PC, Schmidt JP, Kitchen NR, Sudduth KA, Hong SY, Lory JA, Davis JG (2002) Remote sensing for nitrogen management. Journal of Soil and Water Conservation 57(6):518-524. Scharf PC, Kitchen NR, Sudduth KA, Lory JA Stevens WG Oliveira LF, Shannon DK Palm H, Davis JG, Vories ED, Dunn DJ, Jones AJ (2010) Precision nitrogen fertilizer management of maize and cotton using crop sensors. In: World Congress of Soil Science, Soil Solutions for a Changing World, Brisbane. Proceedings… Scharf, P. C.; Shannon, D. K.; Palm, H. L.; Sudduth, K. A.; Drummond, S. T.; Kitchen, N. R.; Mueller, L. J.; Hubbard, V. C.; Oliveira, L. F. (2011) Sensor-based nitrogen applications out- performed producer-chosen rates for corn in on-farm demonstrations. Agronomy Journal 103(6):1683-1691. Schmidt JP, Dellinger AE, Beegle DB (2009) Nitrogen recommendations for corn: an on-the-go sensor compared with current recommendation methods. Agronomy Journal 101(4):916-924. Eng. Agríc., Jaboticabal, v.37, n.4, p.771-781, jul./ago. 2017 781 Temporal variability in active reflectance sensor-measured NDVI in soy and wheat crops Schmidt J, Beegle D, Zhu Q, Sripada R (2011) Improving in-season nitrogen recommendations for maize using an active sensor. Field Crops Research 120(1):94-101. Schmidt J, Beegle D, Zhu Q, Sripada R (2011) Improving in-season nitrogen recommendations for maize using an active sensor. REFERENCES Field Crops Research 120(1):94-101. Souza EG, Orlando GA, Uribe-Opazo MA, Silva ED, Scharf P (2004) Influência da direção de semeadura do milho nas variáveis reflectância e índice de vegetação verde normalizado. Revista Brasileira de Engenharia Agrícola e Ambiental 8(1):79-84. Souza EG, Scharf PC, Sudduth KA, Hipple JD (2006) Using a field radiometer do estimate instantaneous sky clearness. Revista Brasileira de Engenharia Agrícola e Ambiental 10(2):369-373. Souza EG, Scharf PC, Sudduth KA (2010) Sun Position and Cloud Effects on Reflectance and Vegetation Indices of Corn. Agronomy Journal 102(2):734-744. STATSOFT. Statistica for windows. V. 12.0. Tulsa, StatSoft. Available: http://www.statsoft.com. Accessed: Sep 10, 2014. Swindell J (1997) Mapping the spatial variability in the yield potential of arable land through GIS analysis of sequential yield maps. In: European Conference on Precision Agriculture, Warwick. Proceedings… Oxford, Bios Scientific, p827-834. Thrikawala S, Weersink A, Gary K, Fox G (1999) Economic feasibility of variable-rate technology for nitrogen on corn. American Journal of Agricultural Economics 81(4):914-927. Trimble (2013) GreenSeeker system datasheet. Available: http://trl.trimble.com/docushare/dsweb/Get/Document-475150/022503- 11231_GreenSeeker_DS_MarketSmart_USL_0114_LR.pdf. Accessed: Apr 5, 2014. p 11231_GreenSeeker_DS_MarketSmart_USL_0114_LR.pdf. Accessed: Apr 5, 2014. Tucker CJ (1979) Red and Photographic Infrared Linear Combinations for Monitoring Vegetation. Remote Sensing of Environment 8(2):127-150. Tucker CJ (1979) Red and Photographic Infrared Linear Combinations for Monitoring Vegetation. Remote Sensing of Environment 8(2):127-150. Winterhalter L, Mistele B, Schmidhalter U (2013) Evaluation of active and passive sensor systems in the field to phenotype maize hybrids with high-throughput. Field Crops Research 154:236-245. Winterhalter L, Mistele B, Schmidhalter U (2013) Evaluation of active and passive sensor systems in the field to phenotype maize hybrids with high-throughput. Field Crops Research 154:236-245.
https://openalex.org/W4220969544	https://ejournal.pnc.ac.id/index.php/infotekmesin/article/download/677/pdf	Indonesian	null	Efek Penambahan Butanol Terhadap Emisi dan Temperatur Gas Buang Mesin Bensin EFI Menggunakan EGR	Infotekmesin	2,022	cc-by	2,757	Jurnal Infotekmesin Vol.13, No.01, Januari 2022 p-ISSN: 2087-1627, e-ISSN: 2685-9858 DOI: 10.35970/infotekmesin.v13i1.677, pp.8-12 Jurnal Infotekmesin Vol.13, No.01, Januari 2022 p-ISSN: 2087-1627, e-ISSN: 2685-9858 DOI: 10.35970/infotekmesin.v13i1.677, pp.8-12 Jurnal Infotekmesin Efek Penambahan Butanol Terhadap Emisi dan Temperatur Gas Buang Mesin Bensin EFI Menggunakan EGR Firman Lukman Sanjaya1, Syarifudin2, Faqih Fatkhurrozak3 1, 2, 3 Program Studi Teknik Mesin, Politeknik Harapan Bersama Tegal 1,2, 3 Jln.Mataram No. 9, Pesurungan Lor, Kota Tegal, 52147, Indonesia E-mail: sanjaya.ilzamy14@gmail.com1, syarifudin@poltektegal.ac.id2, faqih.fatkhurrozak@poltektegal.ac.id3 Abstrak Info Naskah: Naskah masuk: 18 Mei 2021 Direvisi: 17 Agustus 2021 Diterima: 12 Januari 2022 Tingginya angka oktan dan kandungan oksigen pada butanol dapat memperbaiki pembakaran lebih sempurna dan gas sisa hasil pembakaran mesin lebih ramah lingkungan. Penelitian ini menguji efek penambahan butanol 5% sampai dengan 15% dengan interval 5% pada mesin bensin dengan sistem EGR berbahan bakar premium terhadap emisi dan temperatur gas buang. Alat ukur analisis gas buang digunakan untuk mengukur gas hasil pembakaran. Temperatur gas buang diukur menggunakan termokopel yang terpasang pada exhaust manifold. Campuran P85B15 menurunkan kadar CO sebesar 68,11% dan HC sebesar 37,50% dibanding P100. Namun, emisi CO2 dan temperatur gas buang meningkat. Penggunaan hot dan cold EGR secara umum meningkatkan emisi gas buang mesin. Namun, temperatur gas buang menurun sebesar 4,72% saat mesin menggunakan cold EGR dan 1,33 % saat mesin meggunakan hot EGR. Penulis korespondensi: Firman Lukman Sanjaya E-mail: firmanlukman@poltektegal.ac.id Abstract Keywords: butanol; gasoline engine; emissions and exhaust gas; temperature. The high octane number and oxygen content of butanol can improve combustion more completely and the residual gas resulting from engine combustion is more environmentally friendly. This study examines the effect of adding 5% to 15% butanol at 5% intervals on a gasoline engine with an EGR system with premium fuel on emissions and exhaust gas temperatures. The exhaust gas analyzer was used to measure the combustion gases. Exhaust gas temperature was measured using a thermocouple mounted on the exhaust manifold. The mixture of P85B15 reduced CO levels by 68.11% and HC by 37.50% compared to P100. However, CO2 emissions and exhaust gas temperatures increased. The use of hot and cold EGR generally increased engine exhaust emissions. However, the exhaust gas temperature decreased by 4.72% when the engine used cold EGR and 1.33% when the engine used hot EGR. 8 p-ISSN: 2087-1627, e-ISSN: 2685-9858 Gambar 1. Diagam Alir Penelitian 1. Pendahuluan Kendaraan bermotor semakin meningkat seiring dengan tingginya kebutuhan dan populasi manusia. Hal ini mengakibatkan tingginya penggunaan bahan bakar fosil dan terjadinya krisis energi dan polusi udara yang memberi dampak negatif pada kesehatan manusia [1],[2],[3]. Solusi masalah ini beralih pada bahan bakar alternatif yang ramah lingkungan dan dapat diperbaharui seperti butanol [4],[5]. Butanol terbuat dari bahan-bahan nabati sehingga dapat diperbaharui [6], [7]. Butanol menghasilkan pembakaran sempurna sehingga gas sisa hasil pembakaran lebih ramah lingkungan. Hal ini karena butanol teroksigenisasi sehingga rambat nyala api lebih cepat [8],[9]. Namun, penambahan butanol meningkatkan emisi NOx (Nitrogen Oxide) karena tingginya temperatur pada ruang bakar mengenai campuran nitrogen dan oksigen pada saat proses pembakaran [10]. Solusi pengurangan emisi NOx adalah penggunaan sistem EGR (Exhaust Gas Resirculation) dengan mensirkulasikan kembali sebagian sisa hasil gas pembakaran ke silinder [11]. Penggunaan sistem EGR menurunkan temperatur gas buang mesin sehingga terbentuknya emisi NOx menurun [12]. Gambar 1. Diagam Alir Penelitian Beberapa peneliti memaparkan bahwa bahan bakar bensin yang dicampur dengan butanol dapat memberikan keuntungan. Penambahan butanol dapat memperbaiki pembakaran di ruang bakar [13], hal ini karena butanol memiliki kandungan oksigen yang melimpah sehingga kecepatan rampat nyala api meningkat. Butanol mampu menstabilkan proses pembakaran sehingga emisi gas buang dapat perbaiki [14], tetapi emisi NOx meningkat. Emisi NOx dapat dikurangi dengan menyalurkan kembali sebagian sisa gas hasil pembakaran ke silinder dengan sistem EGR [15],[16]. Hal ini disebabkan karena kandungan oksigen yang masuk ke ruang bakar lebih sedikit sehingga panas spesifik meningkat dan temperatur pembakaran menurun [17] [2]. Tabel 1. Spesifikasi mesin Tabel 1. Spesifikasi mesin Tabel 2. Prosentase campuran bahan bakar Tabel 2. Prosentase campuran bahan bakar Tujuan penelitian yang diusulkan adalah mengurangi ketergantungan bahan bakar bensin dengan mengganti bahan bakar alternatif butanol. Selain itu, penelitian ini menggunakan mesin bensin yang ditambahan sistem EGR. Hal ini untuk mengurangi efek negatif dari penggunaan butanol sebagai bahan bakar berupa emisi NOx. Oleh karena itu, penelitian ini menguji efek penambahan butanol pada bahan bakar premium terhadap emisi dan temperatur gas buang pada mesin bensin dengan menggunakan sistem EGR. Tabel 3. Karakteristik Bahan Bakar Tabel 3. Karakteristik Bahan Bakar 2. Metode P li i Penelitian ini dilakukan melalui beberapa langkah yaitu studi pustaka, persiapan alat dan bahan, setting mesin dan alat ukur, running test mesin, pengujian mesin dan pengambilan data menggunakan berbagai variasi campuran bahan bakar, analisis data dan pembahasan, serta kesimpulan. Adapun diagram alir penelitian ditunjukan pada Gambar1. Mesin bensin EFI digunakan dalam penelitian ini. Mesin bensin EFI digunakan dalam penelitian ini. Spesifikasi mesin ditunjukan pada Tabel 1. Prosentase campuran bahan bakar ditunjukan pada Tabel 2 dan karakteristik bahan bakar ditampilkan pada Tabel 3. Penelitian ini menguji penggunaan butanol pada emisi dan temperatur sisa gas pembakaran pada mesin bensin 9 p-ISSN: 2087-1627, e-ISSN: 2685-9858 Penggunaan hot dan cold EGR meningkatkan emisi lebih tinggi daripada tanpa EGR. Namun, cold EGR meningkatkan emisi CO tertinggi sebesar 135,59% dibanding hot ataupun tanpa EGR. Hasil pengujian diaplikasikan dalam bentuk grafik yang ditunjukan pada Gambar 3. menggunakan sistem EGR. Kecepatan mesin yang digunakan konstan yaitu 3000 rpm. Gambar 2. menunjukan susunan dan alur pengujian mesin bensin. Gambar 2. Experimental Set-Up Gambar 2. Experimental Set-Up P100 P95B5 P90B10 P85B15 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 2,2 2,4 CO (%) Bahan Bakar (%) Tanpa EGR Hot EGR Cold EGR Gambar 3. Hasil uji emisi CO P100 P95B5 P90B10 P85B15 100 110 120 130 140 150 160 170 180 190 200 210 220 HC (ppm) Bahan Bakar (%) Tanpa EGR Hot EGR Cold EGR Gambar 4. Hasil uji emisi HC P100 P95B5 P90B10 P85B15 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 2,0 2,2 2,4 CO (%) Bahan Bakar (%) Tanpa EGR Hot EGR Cold EGR Gambar 3. Hasil uji emisi CO Gambar 2. Experimental Set-Up Gambar 3. Hasil uji emisi CO P100 P95B5 P90B10 P85B15 100 110 120 130 140 150 160 170 180 190 200 210 220 HC (ppm) Bahan Bakar (%) Tanpa EGR Hot EGR Cold EGR Gambar 4. Hasil uji emisi HC Mixer digunakan untuk mencampurkan senyawa kimia bahan bakar premium dan butanol agar tercampur dengan baik. Campuran bahan bakar didistribusikan oleh pompa bahan bakar menuju injektor dan dikabutkan sebelum masuk ke ruang bakar. Pada exhaust manifold terdapat pipa yang terhubung pada intake manifold yang digunakan untuk mensirkulasikan kembali beberapa sisa gas buang ke ruang bakar. Proses tersebut merupakan sistem EGR. Sisa hasil pembakaran keluar melaui exhaust manifold dan kadar emisi CO, HC, CO2 diukur menggunakan alat ukur analisa gas buang Stargass 88. 3.1 Emisi Karbon Monoksida (CO) 3.1 Emisi Karbon Monoksida (CO) Penggunaan butanol pada bahan bakar premium menyempurnakan pembakaran diruang bakar sehingga emisi CO dapat dikurangi. Hal ini karena butanol meningkatkan kandungan oksigen pada ruang bakar sehingga nyala api merampat lebih cepat [9]. Penurunan emisi CO tertinggi sebesar 68,11% saat menggunakan bahan bakar P85B15 dibanding P100. Penggunaan EGR meningkatkan proses heterogen udara dan bahan bakar yang masuk ke selinder sehingga sedikit merusak pembakaran dan terbentuknya emisi CO lebih tinggi [8]. 2. Metode P li i Termokople dipasang pada exhaust manifold untuk mengukur temperatur gas buang melewatinya dan ditampilkan pada display. 3.2 Emisi Hidro Karbon (HC) Emisi HC mengalami penurunan akibat butanol yang ditambahkan pada bahan bakar. Butanol memiliki oksigen yang melimpah dibanding pemium membuat pembakaran dalam silinder lebih baik. Kandungan oksigen yang tinggi pada bahan bakar meningkatkan perambatan nyala api pada silinder sehingga pembakaran lebih cepat dan merata [19] [20]. Penurunan emisi HC tertinggi sebesar 37,50% saat menggunakan campuran bahan bakar P85B15 dibanding P100. Penggunaan EGR memperburuk oksidasi diruang bakar sehingga pelepasan panas saat proses pembakaran berkurang. Hal ini meningkatkan produksi emisi HC. Penggunaan cold EGR meningkatkan emisi HC lebih tinggi dibanding hot EGR [21] [22]. Sistem cold EGR meningkatkan emisi HC tertinggi hingga 44,55% dibanding hot maupun tanpa EGR. Hasil pengujian dipaparkan pada Gambar 4. dalam bentuk grafik. 3.3 Emisi Karbondioksida (CO2) Uji Emisi CO2 ini dilakukan pada mesin bensin dengan atau tanpa hot/cold EGR pada campuran premium dan butanol. Gambar 5 merupakan grafik hasil uji emisi CO2 dan secara umum penambahan butanol meningkatkan 10 p-ISSN: 2087-1627, e-ISSN: 2685-9858 emisi CO2. Campuran P95B5 menyebabkan peningkatan CO2 tertinggi sebesar 48,84% dibanding dengan P100. Tingginya prosentase oksigen pada butanol bereaksi pada atom karbon yang tidak terbakar selama proses pembakaran sehingga pembentukan emisi CO2 lebih tinggi. Emisi CO2 yang tinggi mengindikasi bahwa proses pembakaran lebih baik [9] [19]. Penggunaan EGR dapat meningkatkan emisi CO2 karena EGR mengirim kembali gas sisa hasil pembakaran berupa CO2 dan H2O. Hal ini dapat meningkatkan emisi CO2 yang keluar dari ruang bakar [2]. Peningkatan tertinggi terjadi pada bahan bakar P85B5 saat menggunakan cold EGR sebesar 32,21% dibanding hot maupun tanpa EGR. temperatur gas buang lebih rendah. Penggunaan cold EGR mampu menurunkan EGR lebih efektif daripada hot EGR [21] [24]. Penggunaan Cold EGR mengalmai penurunan temperatur gas buang tertinggi sebesar 4,72% dibanding hot maupun tanpa EGR. 4. Kesimpulan Hasil pengujian menunjukan penggunaan bahan bakar premium dan butanol menurunkan emisi CO dan HC tertinggi sebesar 68,11% dan 37,50% dibanding premium murni. Namun, emisi CO2 dan temperatur gas buang mesin bensin meningkat masing-masing 48,84% dan 4,92% dibanding premium murni. Penggunaan hot dan cold EGR secara umum meningkatkan emisi gas buang mesin. Namun, temperatur gas buang menurun sebesar 4,72% saat mesin menggunakan cold EGR dan 1,33 % saat mesin meggunakan hot EGR. P100 P95B5 P90B10 P85B15 7,5 8,0 8,5 9,0 9,5 10,0 10,5 11,0 11,5 12,0 12,5 CO2 (%) Bahan Bakar (%) Tanpa EGR Hot EGR Cold EGR Gambar 5. Hasil uji emisi CO2 P100 P95B5 P90B10 P85B15 640 645 650 655 660 665 670 675 680 685 Temperatur Gas Buang ( oC) Bahan Bakar (%) Tanpa EGR Hot EGR Cold EGR Gambar 6. Hasil uji temperatur gas buang P100 P95B5 P90B10 P85B15 7,5 8,0 8,5 9,0 9,5 10,0 10,5 11,0 11,5 12,0 12,5 CO2 (%) Bahan Bakar (%) Tanpa EGR Hot EGR Cold EGR Gambar 5. Hasil uji emisi CO2 Daftar Pustaka [1] Q. Tang, P. Jiang, C. Peng, H. Chang, and Z. Zhao, “Comparison and analysis of the effects of spark timing and lambda on a high-speed spark ignition engine fuelled with n- butanol/gasoline blends,” Fuel, vol. 287, no. August, p. 119505, 2021, doi: 10.1016/j.fuel.2020.119505. j [2] F. Fatkhurrozak and Syaiful, “Effect of Diethyl Ether (DEE) on Performances and Smoke Emission of Direct Injection Diesel Engine Fueled by Diesel and Jatropha Oil Blends with Cold EGR System,” IOP Conf. Ser. Mater. Sci. Eng., vol. 494, no. 1, 2019, doi: 10.1088/1757- 899X/494/1/012005. Gambar 5. Hasil uji emisi CO2 [3] F. Fatkhurrozak, F. L. Sanjaya, "Pengaruh Dietyl Ether pada Mesin Diesel Berbahan Bakar Diesel dan Jatropa Terhadap Smoke Opacity", vol. 9, no. 1, pp. 5–7, 2020. P100 P95B5 P90B10 P85B15 640 645 650 655 660 665 670 675 680 685 Temperatur Gas Buang ( oC) Bahan Bakar (%) Tanpa EGR Hot EGR Cold EGR Gambar 6. Hasil uji temperatur gas buang [4] F. L. Sanjaya, “Pengaruh Penambahan Butanol sebagai Campuran Bahan Bakar Premium terhadap Torsi dan Daya Mesin Bensin dengan Sistem EGR,” vol. 1, no. 1, pp. 7–10, 2020, doi: 10.35970/accurate.v1i1.175. [5] Z. Tian, X. Zhen, Y. Wang, D. Liu, and X. Li, “Combustion and emission characteristics of n-butanol-gasoline blends in SI direct injection gasoline engine,” Renew. Energy, vol. 146, pp. 267–279, 2020, doi: 10.1016/j.renene.2019.06.041. [6] S. Syarifudin and S. Syaiful, “Pengaruh Penggunaan Energi Terbarukan Butanol Terhadap Penurunan Emisi Jelaga Mesin Diesel Injeksi Langsung Berbahan Bakar Biodiesel Campuran Solar Dan Jatropa,” Infotekmesin, vol. 10, no. 1, pp. 18–22, 2019, doi: 10.35970/infotekmesin.v10i1.20. [7] K. Liu, B. Deng, Q. Shen, J. Yang, and Y. Li, “Optimization based on genetic algorithms on energy conservation potential of a high speed SI engine fueled with butanol– gasoline blends,” Energy Reports, vol. 8, pp. 69–80, 2022, doi: 10.1016/j.egyr.2021.11.289. Gambar 6. Hasil uji temperatur gas buang 3.4 Temperatur Gas Buang Gambar 6 menunjukan hasil uji Temperatur gas buang mesin bensin meningkat saat bahan bakar ditambahkan butanol. Angkat oktan yang tinggi pada butanol mengakibatkan peningkatan temperatur langkah kompresi sehingga temperatur meningkat di ruang bakar [8]. Selain itu, butanol merupakan bahan bakar teroksigenisasi sehingga kecepatan rambat nyala api dan temperatur dalam ruang bakar meningkat [19] [23]. Temperatur gas buang tertinggi sebesar 4,92% pada penggunaan campuran bahan bakar P85B15 dibanding P100. Namun, temperatur gas buang dapat perbaiki dengan sistem EGR. Gas sisa hasil pembakaran yang disalurkan kembali ke silinder oleh EGR menurunkan temperatur puncak silinder sehingga [8] F. Sanjaya, S. Syaiful, and N. SINAGA, “Effect of butanol on performances and exhaust gas emissions of gasoline engine with egr system,” Int. J. Innov. Eng. Technol., vol. 13, no. 4, pp. 117–125, 2019. [9] M. N. A. M. Yusoff et al., “Performance and emission characteristics of a spark ignition engine fuelled with butanol isomer-gasoline blends,” Transp. Res. Part D Transp. Environ., vol. 57, no. September, pp. 23–38, 2017, doi: 10.1016/j.trd.2017.09.004. [10] J. Cha, J. Kwon, Y. Cho, and S. Park, “The effect of Exhaust Gas Recirculation (EGR) on combustion stability, engine performance and exhaust emissions in a gasoline engine,” KSME Int. J., 2001, doi: 10.1007/BF03185686. [11] Syarifudin, Syaiful, and E. Yohana, “Effect of butanol on 11 p-ISSN: 2087-1627, e-ISSN: 2685-9858 fuel consumption and smoke emission of direct injection diesel engine fueled by jatropha oil and diesel fuel blends with cold EGR system,” SHS Web Conf., vol. 49, p. 02010, 2018, doi: 10.1051/shsconf/20184902010. Premium-Butanol Blends on Fuel Consumption and Emissions on Gasoline Engine with Cold EGR System,” J. Phys. Conf. Ser., vol. 1373, no. 1, pp. 11–17, 2019, doi: 10.1088/1742-6596/1373/1/012019. Premium-Butanol Blends on Fuel Consumption and Emissions on Gasoline Engine with Cold EGR System,” J. Phys. Conf. Ser., vol. 1373, no. 1, pp. 11–17, 2019, doi: 10.1088/1742-6596/1373/1/012019. [19] M. S. M. Zaharin, N. R. Abdullah, H. H. Masjuki, O. M. Ali, G. Najafi, and T. Yusaf, “Evaluation on physicochemical properties of iso-butanol additives in ethanol-gasoline blend on performance and emission characteristics of a spark-ignition engine,” Applied Thermal Engineering, vol. 144. pp. 960–971, 2018, doi: 10.1016/j.applthermaleng.2018.08.057. [12] L. Chen, T. Li, T. Yin, and B. Zheng, “A predictive model for knock onset in spark-ignition engines with cooled EGR,” Energy Convers. Manag., vol. 87, pp. 946–955, 2014, doi: 10.1016/j.enconman.2014.08.002. [13] Y. Li, Z. Ning, C. fon F. Lee, J. Yan, and T. H. 3.4 Temperatur Gas Buang Lee, “Effect of acetone-butanol-ethanol (ABE)–gasoline blends on regulated and unregulated emissions in spark-ignition engine,” Energy, vol. 168, pp. 1157–1167, 2019, doi: 10.1016/j.energy.2018.12.022. [20] H. Sharudin, N. R. Abdullah, G. Najafi, R. Mamat, and H. H. Masjuki, “Investigation of the effects of iso-butanol additives on spark ignition engine fuelled with methanol- gasoline blends,” Appl. Therm. Eng., 2017, doi: 10.1016/j.applthermaleng.2016.12.017. [14] Z. Chen, Y. Zhang, X. Wei, Q. Zhang, Z. Wu, and J. Liu, “Thermodynamic process and performance of high n- butanol/gasoline blends fired in a GDI production engine running wide-open throttle (WOT),” Energy Conversion and Management, vol. 152. pp. 57–64, 2017, doi: 10.1016/j.enconman.2017.09.037. [21] A. S. Ayodhya, V. T. Lamani, P. Bedar, and G. N. Kumar, “Effect of exhaust gas recirculation on a CRDI engine fueled with waste plastic oil blend,” Fuel, vol. 227, no. X, pp. 394–400, 2018, doi: 10.1016/j.fuel.2018.04.128. [15] H. Wei, T. Zhu, G. Shu, L. Tan, and Y. Wang, “Gasoline engine exhaust gas recirculation - A review,” Appl. Energy, vol. 99, no. 2015, pp. 534–544, 2012, doi: 10.1016/j.apenergy.2012.05.011. [22] C. Hergueta, M. Bogarra, A. Tsolakis, K. Essa, and J. M. Herreros, “Butanol-gasoline blend and exhaust gas recirculation, impact on GDI engine emissions,” Fuel, vol. 208, pp. 662–672, 2017, doi: 10.1016/j.fuel.2017.07.022. [16] T. Li, T. Yin, and B. Wang, “Anatomy of the cooled EGR effects on soot emission reduction in boosted spark-ignited direct-injection engines,” Appl. Energy, vol. 190, pp. 43–56, 2017, doi: 10.1016/j.apenergy.2016.12.105. [23] G. Dhamodaran, G. S. Esakkimuthu, Y. K. Pochareddy, and H. Sivasubramanian, “Investigation of n-butanol as fuel in a four-cylinder MPFI SI engine,” Energy, vol. 125, pp. 726– 735, 2017, doi: 10.1016/j.energy.2017.02.134. [17] D. Agarwal, S. K. Singh, and A. K. Agarwal, “Effect of Exhaust Gas Recirculation (EGR) on performance, emissions, deposits and durability of a constant speed compression ignition engine,” Appl. Energy, vol. 88, no. 8, pp. 2900–2907, 2011, doi: 10.1016/j.apenergy.2011.01.066. [24] S. Verma, L. M. Das, S. C. Kaushik, and S. S. Bhatti, “The effects of compression ratio and EGR on the performance and emission characteristics of diesel-biogas dual fuel engine,” Appl. Therm. Eng., vol. 150, no. January, pp. 1090– 1103, 2019, doi: 10.1016/j.applthermaleng.2019.01.080. pp j p gy [18] F. L. Sanjaya, Syaiful, and D. N. Sinaga, “Effect of 12
https://openalex.org/W2951267508	https://link.springer.com/content/pdf/10.1140/epjc/s10052-015-3844-4.pdf	English	null	The top quark right coupling in the tbW-vertex	European physical journal. C, Particles and fields	2,015	cc-by	10,384	Gabriel A. González-Sprinberg1,a, Jordi Vidal2,b Gabriel A. González-Sprinberg1,a, Jordi Vidal2,b 1 Facultad de Ciencias, Instituto de Física, Universidad de la República, Iguá 4225, Montevideo 11600, Uruguay 2 Departament de Física Teòrica, Universitat de València, and Instituto de Física Corpuscular (IFIC), Centro Mixto Universitat de València-CSIC, Burjassot, 46100 València, Spain Gabriel A. González Sprinberg , Jordi Vidal 1 Facultad de Ciencias, Instituto de Física, Universidad de la República, Iguá 4225, Montevideo 11600, Uruguay 2 Departament de Física Teòrica, Universitat de València, and Instituto de Física Corpuscular (IFIC), Centro Mixto Universitat de València-CSIC, Burjassot, 46100 València, Spain Received: 14 October 2015 / Accepted: 11 December 2015 / Published online: 26 December 2015 © The Author(s) 2015. This article is published with open access at Springerlink.com t →bW +, is known. For instance, many extensions of the SM predict new decay modes that may be accessible at LHC. The top quark was detected for the ﬁrst time at TEVATRON [4,5], where many of its physical properties where measured and some bound on the anomalous Wtb couplings were set [6–8]. Nowadays, top physics is intensively investigated in theoretical research [3] and at the LHC [1,2,9], as the reader can veriﬁed in the ATLAS and CMS web pages. In particu- lar, the measurements of the different helicity components of the W in the top decay allows one to study the tbW Lorentz vertex structure [10]. These studies were extended in recent years [11–15]. In these references the longitudinal and trans- verse helicities of the W coming from the top decay were investigated and they show that a precise determination of the Lorentz form factors of the vertex can be done with a suitable choice of observables. The most general parametrization of the on-shell vertex needs four couplings, but in the SM three of them are zero at tree level while only the usual left cou- pling VL is not zero and has a value close to one [16]. This is not the case in extended models where, in addition, some of these couplings can also be sensitive to new CP-violation mechanisms. Abstract The most general parametrization of the tbW vertex includes a right coupling VR that is zero at tree level in the standard model. This quantity may be measured at the Large Hadron Collider where the physics of the top decay is currently investigated. a e-mail: gabrielg@ﬁsica.edu.uy b e-mail: vidal@uv.es Gabriel A. González-Sprinberg1,a, Jordi Vidal2,b This coupling is present in new physics models at tree level and/or through radiative correc- tions, so its measurement can be sensitive to non-standard physics. In this paper we compute the leading electroweak and QCD contributions to the top VR coupling in the stan- dard model. This value is the starting point in order to sep- arate the standard model effects and, then, search for new physics. We also propose observables that can be addressed at the LHC in order to measure this coupling. These observ- ables are deﬁned in such a way that they do not receive tree level contributions from the standard model and are directly proportional to the right coupling. Bounds on new physics models can be obtained through the measurements of these observables. Eur. Phys. J. C (2015) 75:615 DOI 10.1140/epjc/s10052-015-3844-4 Regular Article - Theoretical Physics Regular Article - Theoretical Physics Regular Article - Theoretical Physics The top quark right coupling in the tbW-vertex Gabriel A. González-Sprinberg1,a, Jordi Vidal2,b 2 Right top tbW coupling in the SM from diagrams tw0w, t Hw, bww0, bwH, w0tb and Htb are ultraviolet(UV)divergent.However,whensummingthemup by pairs (i.e. tw0w+t Hw, bww0 +bwH, and w0tb+ Htb), the result is ﬁnite. The electroweak contributions of all the diagrams are given in Appendix A in terms of parametric integrals. For the UV divergent diagrams we present the sum of the two diagrams that cancels the divergence, as can be seen in Eqs. (46), (52), and (57). There, the ﬁrst (second) term, in each of these expressions, corresponds to the UV safe sector of the ﬁrst (second) diagram, while the third one corresponds to the sum of the UV divergent part of the two diagrams, which results in a ﬁnite contribution, as it should. Considering the most general Lorentz structure for on-shell particles, the MtbW amplitude for the t(p) →b(p′)W +(q) decay can be written in the following way: MtbW = − e sin θw √ 2 ϵμ∗ub(p′) γμ(VL PL + VR PR) +iσμνqν MW (gL PL + gR PR) ut(p), (1) (1) where the outgoing W + momentum, mass and polarization vector are q = p −p′, MW, and ϵμ, respectively. The form factors VL and VR are the left and right couplings, respec- tively, while gL and gR are the so called tensorial or anoma- lous couplings. All the contributions can be written as All the contributions can be written as V ABC R = α Vtb rb I ABC, (2) (2) The expression (1) is the most general model indepen- dent parametrization for the tbW + vertex. Within the SM, the W couples only to left particles then, at tree level, all the couplings are zero except for VL, which is given by the Kobayashi–Maskawa matix element VL = Vtb ≃1 [16]. At one loop all of these couplings receive contributions from the SM. For example, the real and imaginary parts of the gL and gR couplings were calculated in the SM [18] and in a general alligned-2HDM [19]. where rb = mb/mt and I ABC is an integral shown in Appendix A for all the diagrams that contribute to VR. As expected, all the contributions are proportional to the bottom mass through rb. Whenoneoftheparticlescirculatingintheloopisaphoton the integral can be performed analytically. 1 Introduction Top quark physics is now a high statistics physics, mainly due to the huge amounts of data coming from the large hadron collider (LHC) run I and, now, run II [1,2]. It is strongly believed that, due to its very high mass, the top quark will be a window to new physics [3]. This can easily be understood in the effective lagrangian approach, where the new physics contributions can be parametrized in a series expansion in terms of the parameter mt/, where mt is the top quarks mass and is the new physics scale. Besides, most of the top quark properties and couplings are known with a precision far lower than the other quarks and than the other standard model (SM) particles. It is the only quark that weakly decays before hadronization but, up to now, only one decay mode, The right top coupling is largely unknown and deserves a careful study. The other two couplings, usually called ten- sorial couplings, were investigated at the LHC [17] and will not be considered here. The predictions for the SM, for the two-Higgs-doublet model (2HDM) and other extended mod- els where recently considered in Refs. [18–20]. In this paper we ﬁrst compute the right coupling VR in the SM at leading order, and deﬁne appropriate observables in order to have direct access to it. The former one-loop calculation is needed in order to disentangle SM and new physics effects in the observables. Next, we will introduce a set of observables that allows one to perform a precise search of the VR coupling. We obtain a combination of observables directly proportional to it in such a way that they are not dominated by the leading tree level SM contribution VL. These observables can be an 123 12 3 3 Eur. Phys. J. C (2015) 75 :615 615 Page 2 of 11 t b A B C W + µ Fig. 1 One-loop contributions to the VR coupling in the t →bW + decay important tool in order to measure new physics contributions to VR. This paper is organized as follows. In the next section we introduce a precise deﬁnition of the right coupling and com- putetheﬁrst order QCDandelectroweak(EW) contributions. In Sect. 3 we present and discuss a set of observables that can be measured with LHC data, both in the polarization matrix and in the spin correlations. 1 Introduction Finally, we discuss the results and present our conclusions in Sect. 4. Fig. 1 One-loop contributions to the VR coupling in the t →bW + decay 2 Right top tbW coupling in the SM Then the contri- bution can be written as V ABC R = α 8π Vtb Q A I ABC 0 , (3) The one-loop electroweak and QCD contributions to the VR coupling can be calculated just by considering the vertex corrections shown in Fig. 1 and extracting from them the Lorentz structure corresponding to the VR coupling. Note that this Lorentz structure is not present in the SM lagrangian so, at one loop, there is no need for any counter-term for this contribution. For this reason, the SM one-loop contribution to the VR coupling is ﬁnite. This is similar to what happens for the top gL,R couplings [18,19]. (3) with Q A being the charge of the A-quark circulating in the loop in units of \|e\| (for the γ tb diagram, Q A = Qt · Qb). The I ABC 0 analytical expressions, as well as their limits for rb →0, are shown in Appendix B. All these expressions were used as a check of our computations. The leading QCD contribution can easily be obtained from Eq. (56) just by substituting the couplings of the photon by the gluon ones, so we get We will denote each diagram by the label ABC according with the particles running in the loop. In addition to the usual notation for the SM particles we use the symbols w0 and w for the neutral and charge would-be Goldstone bosons, respectively, and H for the SM Higgs. There are 18 diagrams that have to be considered and the one-loop VR value one gets from them is ﬁnite, without the need of renormalization as already stated. In particular, the one-loop contributions V gtb R = −αs 8π CF VtbI γ tb 0 , (4) (4) with I γ tb given in Eqs. (62) and (67), and CF = 4/3 is the color factor. 123 123 Page 3 of 11 615 Eur. Phys. J. 3.1 Observables in the W rest frame Top properties were studied in previous works by means of the observables that we just mentioned. One of the ﬁrst possi- bilities are the angular asymmetries for the t →W + b decay, with the W + decaying leptonically. The normalized charged lepton angular distribution in the W rest frame can be written as as 1 d d cos θl = 3 8 1 + cos θl 2 F+ + 3 8 1 −cos θl 2 F− +3 4 sin2 θl F0, (5) (5) where F0, F± are the normalized partial widths of the top decay into the W helicity states, and θl is the angle between the charged lepton momentum in the W rest frame and the W momentum in the t rest frame. Then the asymmetries are deﬁned, in terms of a new parameter z, as follows [10,21]: With the set of values of Ref. [16], the numerical values for the contribution to VR of each diagram and the complete one-loop SM value are given in Table 1, for Vtb = 1. In this table the contributions from the UV divergent dia- grams are summed up to get a ﬁnite result; the ﬁrst (second) quantity between brackets corresponds to the ﬁnite contribu- tion of the ﬁrst (second) diagram, and the third one (which is logarithmic) is the ﬁnite sum of the two UV divergent parts. Az = N(cos θl > z) −N(cos θl < z) N(cos θl > z) + N(cos θl < z). (6) (6) As can be seen in Table 1, even though the contribution of most of the diagrams is of the order of 10−5 for the real part of the VR coupling, the total EW contribution is, at the end, two orders of magnitude smaller due to the accidental cancellations among the diagrams. The QC D contribution is real and four orders of magnitude bigger than the real EW one so that the real part of the VR coupling is dominated by the former. The imaginary part, instead, remains of order 10−5, and it is purely EW. The z parameter allows one to separate the helicity frac- tions F0, F±. The value z = 0 gives the usual forward– backward asymmetry while z = ±(22/3 −1) deﬁnes the A∓ asymmetries, respectively. The ﬁrst one, A0 depends only on the F± helicity fractions while A± depends on both F0 and F±. 3.1 Observables in the W rest frame The helicity fractions, F0 and F±, can be computed in terms of the tbW couplings given in Eq. (1), and they can be found, for example, in Ref. [22]. Performing the θl integra- tion of Eq. (6) for a given value of z, the numerator of the asymmetry Az in terms of VL and VR is 2 Right top tbW coupling in the SM C (2015) 75 :615 615 Table 1 Contribution to VR from the different diagrams Diagram Contribution to VR t ZW 2.01 × 10−5 tγ W −1.10 × 10−5 t HW 0 tw0w (−3.05 × 10−5) + +(0.64 × 10−5) = −1.55 × 10−5 t Hw +(0.86 × 10−5) t Zw 0.10 × 10−5 tγ w 0.69 × 10−5 bW Z (1.12 + 8.24 i) × 10−5 bWγ (8.34 −4.25 i) × 10−5 bW H 0 bww0 (−0.72 + 3.73 i × 10−5) + +(−0.10 −2.99 i × 10−5) = (1.01 −0.35 i) × 10−5 bwH +(1.83 −1.09 i × 10−5) bwZ (0.00 + 0.31 i) × 10−5 bwγ (−4.47 + 2.29 i) × 10−5 Ztb −2.30 × 10−5 γ tb −2.78 × 10−5 w0tb (−0.24 × 10−5) + +(0.20 × 10−5) = −1.03 × 10−5 Htb +(−0.99 × 10−5) (EW) (0.06 + 6.23 i) × 10−5 gtb(QC D) 2.68 × 10−3 (QC D + EW) (2.68 + 0.06 i) × 10−3 the fact that this coupling comes from a lagrangian term that has the same parity and chirality properties than the leading coupling VL so that the observables receive contributions from both terms. These observables are the angular asym- metries in the W rest frame [10,11,21,22], angular asymme- tries in the top rest frame [11,22–24] and spin correlations [11,22,25,26]. Our strategy will be to deﬁne observables directly proportional to VR considering the dependencies on the coupling terms. Similar ideas were widely applied when investigating tau physics dipole moments [27,28]. For top decays, one way to suppress the VL contribution is to deﬁne observables where only right polarized quarks contribute, but this polarization is not accessible to the present facilities and experiments. Given the results shown in the previous section, from now on we always assume that the imaginary part of the VR coupling is negligible. 3.2 Observables in the top rest frame Fig. 2 Plot of the coefﬁcients of V 2 L and V 2 R from Eq. (7) in terms of z The angular distribution of the decay products for the weak process t →W +b carries information as regards the spin of the decaying top. Then angular asymmetries can be built to test the Lorentz structure of the vertex. We will follow the same procedures as described previously, in order to optimize the sensitivity of the observables to VR but, in this case, for the asymmetries in the top rest frame. 0.10 0.05 0.00 0.05 0.10 0.000 0.001 0.002 0.003 0.004 0.005 VR Az A A AR Fig. 3 Dependence on the VR coupling for AR (blue-solid), Eq. (8), and for the usual A+ (purple-dashed) and A−(black-dot-dashed) asym- metries, Eq. (6) y p For the top decay t →W + b →l+νb, q ¯q′b, the angular distribution of the product X = l+, ν, q, ¯q′, W +, b, in the top rest frame, is given by 1 d d cos θX = 1 2 1 + αX cos θX , (9) (9) Fig. 3 Dependence on the VR coupling for AR (blue-solid), Eq. (8), and for the usual A+ (purple-dashed) and A−(black-dot-dashed) asym- metries, Eq. (6) where θX is the angle between the momentum of X and the top spin direction, and αX are the spin-analyzer powers, given in Refs. [11,22–24,29] in terms of the couplings shown in Eq. (1). Then an asymmetry can be deﬁned by One can now choose the value of the z parameter, within the range (−1, 1), to make zero the leading VL contribution. In this way we get the maximum sensitivity to VR. The coef- ﬁcients of V 2 L and V 2 R from Eq. (7) are plotted in Fig. 2. Az X ≡N(cos θX > z) −N(cos θX < z) N(cos θX > z) + N(cos θX < z) = 1 2 αX(1 −z2) −2z . (10) (10) L R There, it can be seen that for z = zR = −0.17 the coef- ﬁcient of V 2 L cancels, leaving the VR term, in Eq. (7), as the leading one. 3 Observables V 2 L 2.48z3 + 2.00z2 −11.45z −2.00 + 0.56zVLVR +V 2 R 2.48z3 −2.00z2 −11.45z + 2.00 . (7) In general, the LHC observables considered in the literature are not very sensitive to the right coupling VR. This is due to (7) 123 3 3 Eur. Phys. J. C (2015) 75 :615 615 Page 4 of 11 1.0 0.5 0.0 0.5 1.0 5 0 5 z Coefficient of VL 2 Coefficient of VR 2 Fig. 2 Plot of the coefﬁcients of V 2 L and V 2 R from Eq. (7) in terms of z 0.10 0.05 0.00 0.05 0.10 0.000 0.001 0.002 0.003 0.004 0.005 VR Az A A AR Fig. 3 Dependence on the VR coupling for AR (blue-solid), Eq. (8), and for the usual A+ (purple-dashed) and A−(black-dot-dashed) asym- metries, Eq. (6) 1.0 0.5 0.0 0.5 1.0 5 0 5 z Coefficient of VL 2 Coefficient of VR 2 Fig. 2 Plot of the coefﬁcients of V 2 L and V 2 R from Eq. (7) in terms of z 1.0 0.5 0.0 0.5 1.0 5 0 5 z Coefficient of VL 2 Coefficient of VR 2 Fig 2 Plot of the coefﬁcients of V 2 and V 2 from Eq (7) in terms of z we show the VR dependence of AR and also, for compari- son, the dependence of A±, with the V 2 L leading contribution subtracted. As can be seen there AR may allow more precise bounds on VR. Besides, it is directly proportional to VR and eliminates other uncertainties that the VL dependence in A± may introduce. For polarized top it is possible to deﬁne asymmetries with respect to the normal and transverse spin directions. These were studied in Ref. [11], but they are not sensitive to the VR coupling so we are not going to consider them in our analysis. 3.2 Observables in the top rest frame For this zR the new asymmetry AR ≡AZ(z = zR) is proportional to VR and has the form For z = 0, one gets the usual forward–backward asym- metry: AR(VL, VR) = −0.01VLVR + 0.48V 2 R 1.12 V 2 L + V 2 R −0.07VLVR ≃−0.01VR, (8) AX ≡N(cos θX > 0) −N(cos θX < 0) N(cos θX > 0) + N(cos θX < 0) = αX 2 . (11) (11) (8) where the last expression is the value of the asymmetry taking VL = 1 and assuming \|VR\| << 1. As can be seen this asymmetry is directly proportional to VR so that a non-zero measurement of it is a direct test of VR ̸= 0. Sensitivities to VR and VL for this asymmetry have already been given for X = l+, b, ν in the t-channel single top pro- duction in Refs. [11,30]. In order to deﬁne a new asymmetry directly proportional to the VR coupling one can again extract the SM leading contribution, given by the V 2 L term in the Az X asymmetry, and make it zero. For the X = l, ν, b cases, Eq. (10) is Note that the leading contribution to the usual A± asym- metries comes from V 2 L so that in order to be sensitive to VR one needs to subtract this SM central value. In Fig. 3 Eur. Phys. J. C (2015) 75 :615 Page 5 of 11 615 Page 5 of 11 615 615 0.10 0.05 0.00 0.05 0.10 0.0025 0.0020 0.0015 0.0010 0.0005 0.0000 VR Al zl Fig. 5 Dependence on the VR coupling for the Azl l asymmetry, Eq. (16) 0.10 0.05 0.00 0.05 0.10 0.0025 0.0020 0.0015 0.0010 0.0005 0.0000 VR Al zl Fig. 5 Dependence on the VR coupling for the Azl l asymmetry, Eq. (16) 1.0 0.5 0.0 0.5 1.0 1.0 0.5 0.0 0.5 1.0 z Coefficient of VL 2 from Al z Coefficient of VL 2 from Ab z Coefficient of VL 2 from Az Fig. 4 Plot of the coefﬁcients of V 2 L , in terms of z, from Eqs. 3.2 Observables in the top rest frame (12), (13), and (14) 0.10 0.05 0.00 0.05 0.10 0.0025 0.0020 0.0015 0.0010 0.0005 0.0000 VR Al zl 1.0 0.5 0.0 0.5 1.0 1.0 0.5 0.0 0.5 1.0 z Coefficient of VL 2 from Al z Coefficient of VL 2 from Ab z Coefficient of VL 2 from Az Fig. 5 Dependence on the VR coupling for the Azl l asymmetry, Eq. (16) 0.10 0.05 0.00 0.05 0.10 0.000 0.001 0.002 0.003 0.004 0.005 VR Ab zb 0.10 0.05 0.00 0.05 0.10 0.000 0.001 0.002 0.003 0.004 0.005 VR Ab zb Fig. 6 Dependence on the VR coupling for the Azb b asymmetry, Eq. (17) Fig. 4 Plot of the coefﬁcients of V 2 L , in terms of z, from Eqs. (12), (13), and (14) Az l = −1 2 V 2 L + V 2 R −0.06VLVR V 2 L −1 2z2 −z + 1 2 + V 2 R −0.16z2 −z + 0.16 + 0.06VLVR 1 2z2 + z −1 2 , (12) (12) Fig. 6 Dependence on the VR coupling for the Azb b asymmetry, Eq. (17) Az b = 0.20 V 2 L + V 2 R −0.06VLVR × V 2 L z2 −4.93z −1 + V 2 R −z2 −4.93z + 1 + 0.31zVLVR , (13) Az b = 0.20 V 2 L + V 2 R −0.06VLVR × V 2 L z2 −4.93z −1 + V 2 R −z2 −4.93z + 1 + 0.31zVLVR , (13) Az ν = 0.16 V 2 L + V 2 R −0.06VLVR × V 2 L z2 −6.30z −1 + V 2 R 3.15z2 −6.30z −3.15 + VLVR −0.19z2 + 0.39z + 0.19 . (14) 0.10 0.05 0.00 0.05 0.10 0.15 0.005 0.004 0.003 0.002 0.001 0.000 VR Az Fig. 7 Dependence on the VR coupling for the Azν ν asymmetry, Eq. (18) 0.10 0.05 0.00 0.05 0.10 0.15 0.005 0.004 0.003 0.002 0.001 0.000 VR Az (13) Az ν = 0.16 V 2 L + V 2 R −0.06VLVR × V 2 L z2 −6.30z −1 + V 2 R 3.15z2 −6.30z −3.15 + VLVR −0.19z2 + 0.39z + 0.19 . 3.3 Spin correlations in t ¯t production (22) Az1z2 X ¯X′ ≡1 σ 1 z1 d(cos θX) 1 z2 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ + z1 −1 d(cos θX) z2 −1 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ − z1 −1 d(cos θX) 1 z2 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ − 1 z1 d(cos θX) z2 −1 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ . (20) 2 1.0 1.0 Fig. 9 Values of z1 Az1z2 ll′ asymmetry, Eq coefﬁcient C z± 2 = −2z1 ± C C Fig. 9 Values of z1 and z2 that make zero the VL leading terms of the Az1z2 ll′ asymmetry, Eq. (22), for different values of the spin correlation coefﬁcient C z± 2 = −2z1 ± C2(1 −z2 1)2 + 4z2 1 C(1 −z2 1) . (22) (22) The different regions of integration for this asymmetry are plotted in Fig. 8. There, the number of events on each of the regions is collected and the sign of their contribution to the asymmetry is also indicated. The particular case z1 = z2 = 0 corresponds to the usual spin correlation asymmetry: Similarly to what happened for Az l , in Eq. (12), for values of z1 and z2 satisfying the previous equation, the coefﬁcient of V 3 L VR in the asymmetry also accidentally cancels. Then there remains the V 2 R coupling as the leading contribution. Solutions to Eq. (22) are plotted in Fig. 9, where it can be seen that the values of z1 and z2 are restricted to the ﬁrst quadrant (for the z+ 2 solution) or to the third one (for the z− 1 solution). In addition to the trivial solutions of Eq. (22), i.e. z1 = ±1, z2 = 0, which reproduces the usual All′ asymme- try, one can improve the computation by ﬁnding the values of z1 and z2 that, satisfying Eq. (22), also maximize the coef- ﬁcient of the leading V 2 R term in Eq. (20), namely: AX ¯X′= N cos θX cos θ ¯X′ > 0 −N cos θX cos θ ¯X′ < 0 N cos θX cos θ ¯X′ > 0 + N cos θX cos θ ¯X′ < 0 . 3.2 Observables in the top rest frame (14) (14) Figure 4 shows the behavior of the V 2 L coefﬁcient from the Al, Ab, and Aν asymmetries. We can choose z in order to make the leading coefﬁcients of Eqs. (12), (13), and (14) zero. These z values are Fig. 7 Dependence on the VR coupling for the Azν ν asymmetry, Eq. (18) zl = √ 2 −1, zb = −0.20, zν = −0.16, (15) (15) one can see that, unfortunately, for the Al asymmetry the same value of z as the one that cancels the coefﬁcient of V 2 L also cancels the VLVR term, in such a way that a poorer sen- sitivity to the coupling will be expected from this observable because the surviving term is V 2 R instead of VLVR. and then the new asymmetries are and then the new asymmetries are Azl l = −0.28V 2 R V 2 L + V 2 R −0.06VLVR ≃−0.28V 2 R, (16) Azb b = 0.39V 2 R −0.01VLVR V 2 L + V 2 R −0.06VLVR ≃−0.01VR, (17) Azν ν = 0.02VLVR −0.33V 2 R V 2 L + V 2 R −0.06VLVR ≃0.02VR, (18) (16) In Figs. 5, 6 and 7 we show the dependence on VR for the new Azl l , Azb b , and Azν ν observables. We have explicitly checked that the usual forward–backward asymmetries for the same decay product have a very similar behavior as the ones shown in the ﬁgures, once the leading V 2 L contribution is removed. However, the new observables again have the advantage that are proportional to VR so that their measure- ment may provide a direct bound or measurement of the VR (17) (18) where in the last expression we show the value of the asym- metries for VL = 1 and assuming \|VR\| << 1. From Eq. (12) 12 3 Eur. Phys. J. C (2015) 75 :615 615 Page 6 of 11 z1,z2 1.0 0.5 0.0 0.5 1.0 1.0 0.5 0.0 0.5 1.0 Cos X θ Cos X θ Fig. 8 Regions of integration and sign of the contribution to the spin correlation asymmetry Az1 z2 X ¯X′ deﬁned in Eq. (20), depending on the point (z1, z2) z1,z2 1.0 0.5 0.0 0.5 1.0 1.0 0.5 0.0 0.5 1.0 Cos X θ Cos X θ coupling with no assumption on the value of the V 2 L leading term. 3.3 Spin correlations in t ¯t production The top–antitop spin correlations depend on the Lorentz structure of the tbW vertex. This structure can be studied through the measurement of the angular distributions of the decay products for the t →W +b and ¯t →W −¯b processes, which carry information on the top and antitop spin correla- tion terms. In particular, using the notation of previous sec- tion, the double angular distribution—of the decay products X, from top, and ¯X′, from antitop—can be written as [25,26]: Fig. 8 Regions of integration and sign of the contribution to the spin correlation asymmetry Az1 z2 X ¯X′ deﬁned in Eq. (20), depending on the point (z1, z2) 1 σ dσ d cos θX d cos θ ¯X′ = 1 4 1 + C αX α ¯X′ cos θX cos θ ¯X′ , (19) z2 C 1 C 0.5 C 0.2 z2 C 1 C 0.5 C 0.2 1.0 0.5 0.0 0.5 1.0 1.0 0.5 0.0 0.5 1.0 z1 z2 z2 C 1 C 0.5 C 0.2 z2 C 1 C 0.5 C 0.2 1.0 0.5 0.0 0.5 1.0 1.0 0.5 0.0 0.5 1.0 z1 z2 Fig. 9 Values of z1 and z2 that make zero the VL leading terms of the Az1z2 ll′ asymmetry, Eq. (22), for different values of the spin correlation coefﬁcient C (19) where θX (θ ¯X′) is the angle between the momentum of the decay product X ( ¯X′) and the momentum of the top (antitop) quark in the t ¯t center-of-mass frame, αX and α ¯X′ are the spin-analyzer powers of particles X and ¯X′, respectively. C is a coefﬁcient that weights the spin correlation between the quark and the antiquark. Then one can deﬁne the asymmetry Az1z2 X ¯X′ ≡1 σ 1 z1 d(cos θX) 1 z2 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ + z1 −1 d(cos θX) z2 −1 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ − z1 −1 d(cos θX) 1 z2 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ − 1 z1 d(cos θX) z2 −1 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ . (20) 2 C 1 1.0 0.5 0.0 0.5 1.0 1.0 z1 Fig. 3.3 Spin correlations in t ¯t production 9 Values of z1 and z2 that make zero the VL leading terms of the Az1z2 ll′ asymmetry, Eq. (22), for different values of the spin correlation coefﬁcient C z± 2 = −2z1 ± C2(1 −z2 1)2 + 4z2 1 C(1 −z2 1) . (22) Az1z2 X ¯X′ ≡1 σ 1 z1 d(cos θX) 1 z2 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ + z1 −1 d(cos θX) z2 −1 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ − z1 −1 d(cos θX) 1 z2 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ − 1 z1 d(cos θX) z2 −1 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ . (20) 2 C 1 1.0 0.5 0.0 0.5 1.0 1.0 z1 Fig. 9 Values of z1 and z2 that make zero the VL leading terms of the Az1z2 ll′ asymmetry, Eq. (22), for different values of the spin correlation coefﬁcient C z± 2 = −2z1 ± C2(1 −z2 1)2 + 4z2 1 C(1 −z2 1) . (22) Az1z2 X ¯X′ ≡1 σ 1 z1 d(cos θX) 1 z2 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ + z1 −1 d(cos θX) z2 −1 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ − z1 −1 d(cos θX) 1 z2 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ − 1 z1 d(cos θX) z2 −1 d(cos θ ¯X′) dσ d cos θX d cos θ ¯X′ . (20) 2 C 1 1.0 0.5 0.0 0.5 1.0 1.0 z1 Fig. 9 Values of z1 and z2 that make zero the VL leading terms of the Az1z2 ll′ asymmetry, Eq. (22), for different values of the spin correlation coefﬁcient C z± 2 = −2z1 ± C2(1 −z2 1)2 + 4z2 1 C(1 −z2 1) . 3.3 Spin correlations in t ¯t production (21) For the Az1z2 X ¯X′ asymmetry, Eq. (20), there is a range of values of z1 and z2 that cancel the leading V 4 L contribution, in such a way that the observable becomes proportional to the VR coupling. For the X = l, ¯X′ = l′ case, these values satisfy the relation 2.00 V 2 R z1z2 + 0.08 C z2 1 1 −z2 2 + z2 2 −1 . (23) (23) 123 Eur. Phys. J. C (2015) 75 :615 Page 7 of 11 615 Page 7 of 11 615 0.15 0.10 0.05 0.00 0.05 0.10 0.15 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 VR All zll Fig. 10 Dependence on VR coupling for Az1z2 ll′ in Eq. (20), for C = 0.4 and z1 = z2 = zll′ = 0.29 0.2 0.1 0.0 0.1 0.2 0.3 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.0030 VR Fig. 11 Dependence on the VR coupling for Az1z2 νl′ , Eq. (20), for C = 0.4 and z1 = −z2 = zνl′ = 0.17 0.15 0.10 0.05 0.00 0.05 0.10 0.15 0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 VR All zll Fig. 10 Dependence on VR coupling for Az1z2 ll′ in Eq. (20), for C = 0.4 and z1 = z2 = zll′ = 0.29 Fig. 11 Dependence on the VR coupling for Az1z2 νl′ , Eq. (20), for C = 0.4 and z1 = −z2 = zνl′ = 0.17 In Fig. 11 we show the dependence of the spin correlation asymmetry Azνl′ zνl′ νl′ on VR, for z1, z2 given by Eq. (28) with C = 0.4. Note that the sensitivity of both asymmetries shown in Figs. 10 and 11 is rather similar. This can be trivially done and the result is z1 = z2 = zll′ ≡± 1 + 2 C −2 C √ 1 + C, (24) (24) Another spin correlation considered in the literature is the angular distribution of the top (antitop) decay products deﬁned as [31]: which corresponds to the intersection of the curves of Fig. 9 with the diagonal line of the ﬁrst and third quadrants. 1 σ dσ d cos ϕX ¯X′ = 1 4 1 + D αX α ¯X′ cos ϕX ¯X′ , (29) (29) These values of z1, z2 give a maximum sensitivity of the Az1z2 ll′ asymmetry to the VR coupling. ll In Fig. 3.3 Spin correlations in t ¯t production 10 we show the dependence on the VR coupling for the Az1z2 ll′ asymmetry in Eq. (20), for C = 0.4 [11], and z1 = z2 = zll′ = 0.29, given by Eq. (24). where ϕX ¯X′ is the angle between the momentum of the X particle in the t rest frame, and that of the X′ one, in the ¯t rest frame. D is the spin correlation coefﬁcient. Then one can construct the following asymmetry: The same procedure followed here can be used for other decay products of the W ±. For t ¯t →l ν b ¯ν l′ ¯b ﬁnal state, the values of z1 and z2 that cancel the coefﬁcient of the leading V 4 L term in the Az1z2 νl′ asymmetry (X = ν, ¯X′ = l′ in Eq. (20)) satisfy a quadratic equation, ˜Az X ¯X′ ≡N(cos ϕX ¯X′ > z) −N(cos ϕX ¯X′ < z) N(cos ϕX ¯X′ > z) + N(cos ϕX ¯X′ < z) = 1 2 αX α ¯X′(1 −z2) −2z . (30) (30) z1z2 = −β(1 −z2 1)(1 −z2 2), β = 0.08C, (25) (25) For z = 0 one gets the usual forward–backward spin cor- relation asymmetry (25) For z = 0 one gets the usual forward–backward spin cor- relation asymmetry and the solution is and the solution is ˜AX ¯X′ = 1 2 D αX α ¯X′. (31) (31) z± 2 = 1 2β(1 −z2 1) z1 ± z1 + 4β2(1 −z1)2 . (26) (26) For both W ± leptonic decays, X = l and ¯X′ = l′, and following similar procedures to the previous sections, we can ﬁnd the value of z that cancels the V 4 L leading term in Eq. (30): Here, −1 ≤z1 ≤0 for the z+ 2 solution and 0 ≤z1 ≤1 for the z− 2 solution of Eq. (26), in order to satisfy \|z± 2 \| < 1. In that case, these z1,2 values do not cancel the VR term of the asymmetry, which remains as the leading one: z = ˜zll′ ≡1 D 1 − 1 + D2 . (32) (32) −0.12 z1z2 −C 4 (1 −z2 1)(1 −z2 2) VR. (27) (27) This value also cancels the VR term, so that the V 2 R contri- bution is the dominant one. 4 Conclusions We computed the SM one-loop QCD and electroweak con- tribution to VR. Due to accidental cancellations between the diagrams, the leading contribution is mainly coming from QCD, it is real and of the order of 10−3. Any measurement of observables that may lead to VR higher than 10−3 should be interpreted as new physics effects. Fig. 12 Dependence on the VR coupling for the ˜Az ll′ asymmetry, Eq. (33), for D = −0.29 and z = ˜zll′ given by Eq. (32) We also have proposed new observables that may provide a direct measurement of the right coupling. We found that for several angular asymmetries considered in the literature it is possible to deﬁne new observables, with an optimal choice of parameters, in such a way that they become a direct probe of VR. These observables include angular asymmetries in the W rest frame, angular asymmetries in the top rest frame and also spin correlations. All the new observables deﬁned in this paper are proportional to VR and thus suitable for a direct determination of this coupling. In some cases the behavior shown by the expressions for our observables, presents a better sensitivity to VR (like it is shown in Fig. 3). While the asymmetries usually considered in the literature have leading contributions from VL, the new asymmetries that we have studied here have as a leading term the VR right coupling we are interested in. These asymmetries can be measured with LHC data, where a huge number of top events are being collected, in order to obtain a direct measurement on the standard model contribution to the right top quark. 0.2 0.1 0.0 0.1 0.2 0.3 0.005 0.004 0.003 0.002 0.001 0.000 VR Fig. 13 Dependence on the VR coupling for the ˜Az νl′ asymmetry, Eq. (35), for D = −0.29 and z = z = ˜zνl′ given by Eq. (34) 0.2 0.1 0.0 0.1 0.2 0.3 0.005 0.004 0.003 0.002 0.001 0.000 VR Fig. 13 Dependence on the VR coupling for the ˜Az νl′ asymmetry, Eq. (35), for D = −0.29 and z = z = ˜zνl′ given by Eq. (34) where again, in the last term, we show the asymmetry for VL = 1 and \|VR\| << 1. This asymmetry can be seen in Fig. 12. 3.3 Spin correlations in t ¯t production Then, for D = −0.29 [11], the ˜Az ll′ asymmetry is This value also cancels the VR term, so that the V 2 R contri- bution is the dominant one. Then, for D = −0.29 [11], the ˜Az ll′ asymmetry is One can easily ﬁnd the values that maximize this coefﬁ- cient and simultaneously verify Eq. (26): ˜A˜zll′ ll′ = −0.19V 2 L V 2 R + 0.01VLV 3 R −0.13V 4 R V 2 L −0.06VLVR + V 2 R 2 ≃−0.19V 2 R z1 = −z2 = zνl′ ≡± 1 + 1 2β −1 2β 1 + 2β (28) (33) 3 3 615 Page 8 of 11 615 Page 8 of 11 Eur. Phys. J. C (2015) 75 :615 0.3 0.2 0.1 0.0 0.1 0.2 0.3 0.020 0.015 0.010 0.005 0.000 VR All’ z Fig. 12 Dependence on the VR coupling for the ˜Az ll′ asymmetry, Eq. (33), for D = −0.29 and z = ˜zll′ given by Eq. (32) 0.2 0.1 0.0 0.1 0.2 0.3 0.005 0.004 0.003 0.002 0.001 0.000 VR Fig. 13 Dependence on the VR coupling for the ˜Az νl′ asymmetry, Eq. (35), for D = −0.29 and z = z = ˜zνl′ given by Eq. (34) 0.3 0.2 0.1 0.0 0.1 0.2 0.3 0.020 0.015 0.010 0.005 0.000 VR All’ z Fig. 12 Dependence on the VR coupling for the ˜Az ll′ asymmetry, Eq (33), for D = −0.29 and z = ˜zll′ given by Eq. (32) 0.3 0.2 0.1 0.0 0.1 0.2 0.3 0.020 0.015 0.010 0.005 0.000 VR All’ z have the advantage of being directly proportional to the VR term that we want to test. 4 Conclusions Analogously, for W leptonic decays with a neutrino– lepton ﬁnal state, X = ν, ¯X′ = l′, the z value that cancels the V 4 L term in Eq. (30) is Acknowledgments This work has been supported, in part, by the Ministerio de Ciencia e Innovación, Spain, under Grants FPA2011- 23897 and FPA2011-23596; by Ministerio de Economía y Competitivi- dad, Spain, under Grants FPA2014-54459-P and SEV-2014-0398; by Generalitat Valenciana, Spain, under Grant PROMETEOII2014-087; and by CSIC and Pedeciba, Uruguay. z = ˜zνl′ = −3.15 D ⎛ ⎝1 − 1 + D 3.15 2 ⎞ ⎠, (34) (34) and then the ˜Az νl′ asymmetry, for D = −0.29, is ˜A˜zνl′ νl′ = 0.006V 3 L VR −0.068V 2 L V 2 R + 0.006VLV 3 R (V 2 L −0.062VLVR + V 2 R)2 ≃0.006VR. and then the ˜Az νl′ asymmetry, for D = −0.29, is ˜A˜zνl′ νl′ = 0.006V 3 L VR −0.068V 2 L V 2 R + 0.006VLV 3 R (V 2 L −0.062VLVR + V 2 R)2 ≃0.006VR. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecomm ons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Funded by SCOAP3. (35) The last term in this equation is given for VL = 1 and \|VR\| << 1. This asymmetry is depicted in Fig. 13. Note that in this case the sensitivity to VR is lower than the one obtained in some of the previous observables due to the small coefﬁcients in the numerator of Eq. (35). The last term in this equation is given for VL = 1 and \|VR\| << 1. This asymmetry is depicted in Fig. 13. Note that in this case the sensitivity to VR is lower than the one obtained in some of the previous observables due to the small coefﬁcients in the numerator of Eq. (35). 123 Appendix A Diagram contributions (40) with I bW H = 0, (51) I bww0 + I bwH = 1 16πs2w 1 r2w 1 0 dx 1 0 dy × x2y 1 −x −r2 b (1 −x(1 −2y)) BZ −(1 −r2 b)x2y(1 −x) BH +x log BH BZ , (52) BZ = x (x(y −1) + 1)r2 b + x −1 y −r2 z (y −1) −r2 w(x −1) [x(y −1) + 1] , (37) CZ = (x −1)(xy −1)r2 b −r2 w(x −1)x(y −1) +r2 z xy + x(y −1)(xy −1), (38) Aγ , Bγ , Cγ = {AZ, BZ, CZ} (rz →0), (39) {AH, BH, CH} = {AZ, BZ, CZ} (rz →rh). Appendix A Diagram contributions (40) with rx ≡mx mt (41) and taking the usual deﬁnition for the SM couplings at = −ab = 1, vb = −1+4s2 w 3 and vt = 1−8s2 w 3 , (42) the expression for the contribution of each diagram is given by I t ZW = − 1 32πs2w × 1 0 dx × 1 0 dy 2x2y [vt(1 + 2xy) −at(5 −2xy)] AZ , (43) I tγ W = −1 8π Qt × 1 0 dx 1 0 dy 2x2y(1 + 2xy) Aγ , (44) I t HW = 0, (45) I tw0w + I t Hw = 1 16 2 1 2 1 dx 1 dy I bW H = 0, (51) I bww0 + I bwH = 1 16πs2w 1 r2w 1 0 dx 1 0 dy × x2y 1 −x −r2 b (1 −x(1 −2y)) BZ −(1 −r2 b)x2y(1 −x) BH +x log BH BZ , (52) I bwZ = 1 32πc2w × 1 0 dx 1 0 dy 2x2y(vb −ab) BZ , (53) I bwγ = −1 8π Qb × 1 0 dx 1 0 dy 2x2y Bγ , (54) I Ztb = 1 32πc2ws2w 1 0 dx 1 0 dy × x [ab(1 + xy) −vb(1 −xy)] [at(1 + xy) −vt(1 −xy)] CZ , (55) I γ tb = 1 2π Qb Qt × 1 0 dx 1 0 dy x(1 −xy)2 Cγ , (56) I w0tb + I Htb = 1 16πs2w 1 0 dx 1 0 dy x2(1 −x)(1 −y) × −1 CZ + 1 CH + x r2w log CZ CH . (57) BZ = x (x(y −1) + 1)r2 b + x −1 y −r2 z (y −1) −r2 w(x −1) [x(y −1) + 1] , (37) CZ = (x −1)(xy −1)r2 b −r2 w(x −1)x(y −1) +r2 z xy + x(y −1)(xy −1), (38) Aγ , Bγ , Cγ = {AZ, BZ, CZ} (rz →0), (39) {AH, BH, CH} = {AZ, BZ, CZ} (rz →rh). Appendix A Diagram contributions (40) with I bW H = 0, (51) I bww0 + I bwH = 1 16πs2w 1 r2w 1 0 dx 1 0 dy × x2y 1 −x −r2 b (1 −x(1 −2y)) BZ −(1 −r2 b)x2y(1 −x) BH +x log BH BZ , (52) (51) rx ≡mx mt (41) I bwZ = rx ≡mx mt (41) H Z I bwZ = 1 32 2 × 1 dx 1 dy 2x2y(vb −ab) B , (53) (41) (53) and taking the usual deﬁnition for the SM couplings 32πc2w 0 0 y BZ ( ) 1 1 2 and taking the usual deﬁnition for the SM couplings at = −ab = 1, vb = −1+4s2 w 3 and vt = 1−8s2 w 3 , (42) the expression for the contribution of each diagram is given by the expression for the contribution of each diagram is given by I Ztb = 1 32πc2ws2w 0 dx 0 dy × x [ab(1 + xy) −vb(1 −xy)] [at(1 + xy) −vt(1 −xy)] I t ZW = − 1 32πs2w × 1 0 dx × 1 0 dy 2x2y [vt(1 + 2xy) −at(5 −2xy)] AZ , (43) I tγ W = −1 8π Qt × 1 0 dx 1 0 dy 2x2y(1 + 2xy) Aγ , (44) I t HW = 0, (45) H 1 1 1 1 × [ b( y) b( y)] [ t( y) t( y)] CZ , (55) I γ tb = 1 2π Qb Qt × 1 0 dx 1 0 dy x(1 −xy)2 Cγ , (56) I w0tb + I Htb = 1 16πs2w 1 0 dx 1 0 dy x2(1 −x)(1 −y) × −1 CZ + 1 CH + x r2 log CZ CH . Appendix A Diagram contributions Wecheckedthattheexpressionsoftheobservablesdeﬁned for the VR coupling in this section show a behavior compa- rable to the one obtained from the expressions of the observ- ables deﬁned in the literature, once the VL leading contri- bution is removed. Moreover, the observables deﬁned here Using the following deﬁnitions for the denominators: AZ = x2 (y −1)r2 b + 1 y −r2 w(y −1) −r2 z (x −1), (36) 123 Eur. Phys. J. C (2015) 75 :615 Page 9 of 11 615 BZ = x (x(y −1) + 1)r2 b + x −1 y −r2 z (y −1) −r2 w(x −1) [x(y −1) + 1] , (37) CZ = (x −1)(xy −1)r2 b −r2 w(x −1)x(y −1) +r2 z xy + x(y −1)(xy −1), (38) Aγ , Bγ , Cγ = {AZ, BZ, CZ} (rz →0), (39) {AH, BH, CH} = {AZ, BZ, CZ} (rz →rh). (40) with rx ≡mx mt (41) and taking the usual deﬁnition for the SM couplings at = −ab = 1, vb = −1+4s2 w 3 and vt = 1−8s2 w 3 , (42) the expression for the contribution of each diagram is given by t ZW 1 1 I bW H = 0, (51) I bww0 + I bwH = 1 16πs2w 1 r2w 1 0 dx 1 0 dy × x2y 1 −x −r2 b (1 −x(1 −2y)) BZ −(1 −r2 b)x2y(1 −x) BH +x log BH BZ , (52) I bwZ = 1 32πc2w × 1 0 dx 1 0 dy 2x2y(vb −ab) BZ , (53) I bwγ = −1 8π Qb × 1 0 dx 1 0 dy 2x2y Bγ , (54) I Ztb = 1 32πc2ws2w 1 0 dx 1 0 dy × x [ab(1 + xy) −vb(1 −xy)] [at(1 + xy) −vt(1 −xy)] CZ , BZ = x (x(y −1) + 1)r2 b + x −1 y −r2 z (y −1) −r2 w(x −1) [x(y −1) + 1] , (37) CZ = (x −1)(xy −1)r2 b −r2 w(x −1)x(y −1) +r2 z xy + x(y −1)(xy −1), (38) Aγ , Bγ , Cγ = {AZ, BZ, CZ} (rz →0), (39) {AH, BH, CH} = {AZ, BZ, CZ} (rz →rh). Appendix A Diagram contributions (57) (55) (56) (57) × −x3y 1 + y −r2 b(1 −y) AZ + (1 −rb2) x3y(1 −y) AH + x log AH AZ , (46) I t Zw = 1 32πc2w × 1 0 dx 1 0 dy 2x2y(at −vt) AZ , (47) I tγ w = −1 8π Qt × 1 0 dx 1 0 dy 2x2y Aγ , (48) I bW Z = 1 32πs2w × 1 0 dx 1 0 dy 2x2y [v (1 + 2xy) a (5 2xy)] Appendix B Contribution of diagrams with a photon I tγ W 0 = −2 rb 1 + (1 −r2 w −r2 b −)(1 −r2 w −3r2 b −) 4r2 b log r2 w 1 −r2w + →− , (58) I tγ w 0 = −1 rb (1 −r2 b −r2 w −) log 1 + r2 w −r2 b + 2rw + →− , (59) bW 2 −2 1 −r2 b −r2 + 2 2 Appendix B Contribution of diagrams with a photon × −x3y 1 + y −r2 b(1 −y) AZ Appendix B Contribution of diagrams with a photon , (46) (47) (48) I tγ W 0 = −2 rb 1 + (1 −r2 w −r2 b −)(1 −r2 w −3r2 b −) 4r2 b log r2 w 1 −r2w + →− , (58) I tγ w 0 = −1 rb (1 −r2 b −r2 w −) log 1 + r2 w −r2 b + 2rw + →− , (59) I t Zw = 1 32πc2w × 1 0 dx 1 0 dy 2x2y(at −vt) AZ , (47) I tγ w = −1 8π Qt × 1 0 dx 1 0 dy 2x2y Aγ , (48) I bW Z = 1 32πs2w × 1 0 dx 1 0 dy ×2x2y [vb(1 + 2xy) −ab(5 −2xy)] BZ , (49) I bWγ = 1 8π Qb × 1 0 dx 1 0 dy 2x2y(1 + 2xy) Bγ , (50) + →− , (58) I tγ w 0 = −1 rb (1 −r2 b −r2 w −) log 1 + r2 w −r2 b + 2rw + →− , (59) I bWγ 0 = 2 rb + −2 1 −r2 b −r2 w + rb(1 + r2 b −r2w + )2 (1 −r2 w)2 −r2 b(1 + r2 w) + (1 −r2 w) −2r2 b(1 −r2 w + ) log 2r2 b 1 2 2 + I t Zw = 1 32πc2w × 1 0 dx 1 0 dy 2x2y(at −vt) AZ , (47) I tγ w = −1 8π Qt × 1 0 dx 1 0 dy 2x2y Aγ , (48) I bW Z = 1 32πs2w × 1 0 dx 1 0 dy 2 + →− , (58) I tγ w 0 = −1 rb (1 −r2 b −r2 w −) log 1 + r2 w −r2 b + 2rw + →− , (59) 2 2 1 2 2 + (59) 32πsw 0 0 ×2x2y [vb(1 + 2xy) −ab(5 −2xy)] BZ , (49) I bWγ = 1 8π Qb × 1 0 dx 1 0 dy 2x2y(1 + 2xy) Bγ , (50) I bWγ 0 = 2 rb + −2 1 −r2 b −r2 w + rb(1 + r2 b −r2w + )2 (1 −r2 w)2 −r2 b(1 + r2 w) + (1 −r2 w) −2r2 b(1 −r2 w + ) log 2r2 b 1 −r2w −r2 b + I bWγ = 1 8π Qb × 1 0 dx 1 0 dy 2x2y(1 + 2xy) Bγ , (50) −r2 b(1 + r2 w) + (1 −r2 w) −2r2 b(1 −r2 w + ) log 2r2 b 1 −r2w −r2 b + 123 3 Eur. Appendix A Diagram contributions I γ tb 0 = 2 rb log 1 −r2 w + r2 b + 1 −r2w + r2 b − , (62) (62) 10. F. del Aguila, J. Aguilar-Saavedra, Precise determination of the Wtb couplings at CERN LHC. Phys. Rev. D 67, 014009 (2003). arXiv:hep-ph/0208171 with = (1 −r2w)2 + r4 b −2r2 b(1 −r2w). 11. J. Aguilar-Saavedra, J. Bernabéu, W polarisation beyond helicity fractions in top quark decays. Nucl. Phys. B 840, 349–378 (2010). arXiv:1005.5382 In the limit rb →0, the formulas get the simplest expres- sion: 12. J. Drobnak, S. Fajfer, J.F. Kamenik, New physics in t − decay at next-to-leading order in QCD. Phys. Rev. D 82, 114008 (2010). arXiv:1010.2402 I tγ W 0 rb→0 −−−−−→− rb (1 −r2w)3 × (3 −8r2 w + 5r4 w) + 4r2 w(1 −2r2 w) log(rw) , (63) I tγ w 0 rb→0 −−−−−→ 2 rb (1 −r2w)2 1 −r2 w + r2 w log(r2 w) , (64) I tγ W 0 rb→0 −−−−−→− rb (1 −r2w)3 × (3 −8r2 w + 5r4 w) + 4r2 w(1 −2r2 w) log(rw) , (63) I tγ w 0 rb→0 −−−−−→ 2 rb (1 −r2w)2 1 −r2 w + r2 w log(r2 w) , (64) I bWγ 0 rb→0 −−−−−→2 rb πi 2 + r2 w + r4 w 1 −r2w +2 r2 w 1 −2r2 w + 2r6 w −r8 w (1 −r2w)4 log(rw) + 1 −2 −r2 w(5 −4r2 w + 2r2 w −2rw + r8 w) (1 −r2w)4 log(1 −r2 w) , (65) 13. S.D. Rindani, P. Sharma, Probing anomalous tbW couplings in single-top production using top polarization at the large hadron collider. JHEP 1111, 082 (2011). arXiv:1107.2597 (63) 14. Q.-H. Cao, B. Yan, J.-H. Yu, C. Zhang, A general analysis of Wtb anomalous couplings. arXiv:1504.0378 15. A.V. Prasath, R.M. Godbole, S.D. Rindani, longitudinal top polar- isation measurement and anomalous Wtb coupling. Eur. Phys. J. C 75(9), 402 (2015). arXiv:1405.1264 16. Particle Data Group Collaboration, K. Olive et al., Review of par- ticle physics. Chin. Phys. C38, 090001 (2014) 17. M. Moreno Llácer, Search for CP violation in single top quark events with the ATLAS detector at LHC. Appendix A Diagram contributions Phys. J. C (2015) 75 :615 615 Page 10 of 11 References 1. F.-P. Schilling, Top quark physics at the LHC: a review of the ﬁrst two years. Int. J. Mod. Phys. A 27, 1230016 (2012). arXiv:1206.4484 2. R. Hawkings, Top quark physics at the LHC. Comptes Rendus Physique 16, 424–434 (2015) (60) 3. W. Bernreuther, Top quark physics at the LHC. J. Phys. G 35, 083001 (2008). arXiv:0805.1333 I bwγ 0 = −2 rb π i 1 −r2 b −r2 w + (1 + r2 b −r2w + ) − →− + 1 −r2 b −r2 w + (1 + r2 b −r2w + ) log 2r2 b 1 −r2 b −r2w + + →− , (61) I γ tb 0 = 2 rb log 1 −r2 w + r2 b + 1 −r2w + r2 b − , (62) with = (1 −r2w)2 + r4 b −2r2 b(1 −r2w). 4. CDF Collaboration, F. Abe et al., Observation of top quark pro- duction in ¯pp collisions. Phys. Rev. Lett. 74, 2626–2631 (1995). arXiv:hep-ex/9503002 I bwγ 0 = −2 rb π i 1 −r2 b −r2 w + (1 + r2 b −r2w + ) − →− + 1 −r2 b −r2 w + (1 + r2 b −r2w + ) log 2r2 b 1 −r2 b −r2w + + →− , (61) 5. D0 Collaboration, S. Abachi et al., Observation of the top quark. Phys. Rev. Lett. 74, 2632–2637 (1995). arXiv:hep-ex/9503003 6. C. Deterre, W vertex at the Tevatron. Nuovo Cim. C035N3, 125– 129 (2012). arXiv:1203.6802 7. D0 Collaboration, V. M. Abazov et al., Search for anomalous Wtb = 1.96 TeV. Phys. Lett. B 708, 21–26 (2012). arXiv:1110.4592 (61) 8. D0 Collaboration, V. M. Abazov et al., Search for anomalous Wtb couplings in single top quark production. Phys. Rev. Lett. 101, 221801 (2008). arXiv:0807.1692 9. C. Bernardo, N.F. Castro, M.C.N. Fiolhais, H. Gonçalves, A.G.C. Guerra, M. Oliveira, A. Onofre, Studying the Wtb vertex structure using recent LHC results. Phys. Rev. D 90(11), 113007 (2014). arXiv:1408.7063 I γ tb 0 = 2 rb log 1 −r2 w + r2 b + 1 −r2w + r2 b − , (62) with = (1 −r2w)2 + r4 b −2r2 b(1 −r2w). Appendix A Diagram contributions PhD thesis, Valencia U., IFIC, 2014 I bWγ 0 rb→0 −−−−−→2 rb πi 2 + r2 w + r4 w 1 −r2w +2 r2 w 1 −2r2 w + 2r6 w −r8 w (1 −r2w)4 log(rw) + 1 −2 −r2 w(5 −4r2 w + 2r2 w −2rw + r8 w) (1 −r2w)4 log(1 −r2 w) , (65) 18. G.A. González-Sprinberg, R. Martinez, J. Vidal, Top quark ten- sor couplings. JHEP 07, 094 (2011). arXiv:1105.5601. [Erratum: JHEP05,117(2013)] 19. L. Duarte, G.A. González-Sprinberg, J. Vidal, Top quark anoma- lous tensor couplings in the two-Higgs-doublet models. JHEP 1311, 114 (2013). arXiv:1308.3652 (65) 20. W. Bernreuther, P. Gonzalez, M. Wiebusch, The top quark decay vertex in standard model extensions. Eur. Phys. J. C 60, 197–211 (2009). arXiv:0812.1643 I bwγ 0 rb→0 −−−−−→ −rb 1 −r2w 2π i (1 + r2 w) + log r2 b 1 −r2w +r2 w log r2 w 1 −r2w , (66 21. B. Lampe, Forward-backward asymmetry in top quark semilep- tonic decay. Nucl. Phys. B 454, 506–526 (1995) y y 22. J. Aguilar-Saavedra, J. Carvalho, N.F. Castro, F. Veloso, A. Onofre, Probing anomalous Wtb couplings in top pair decays. Eur. Phys. J. C 50, 519–533 (2007). arXiv:hep-ph/0605190 (66) p p 23. B. Grzadkowski, Z. Hioki, New hints for testing anomalous top quark interactions at future linear colliders. Phys. Lett. B 476, 87– 94 (2000). arXiv:hep-ph/9911505 I γ tb 0 rb→0 −−−−−→−4 rb 1 −r2w log rb 1 −r2w , (67) (67) Eur. Phys. J. C (2015) 75 :615 Page 11 of 11 615 615 24. R.M. Godbole, S.D. Rindani, R.K. Singh, Lepton distribution as a probe of new physics in production and decay of the t quark and its polarization. JHEP 0612, 021 (2006). arXiv:hep-ph/0605100 28. J. Bernabeu, G. González-Sprinberg, M. Tung, J. Vidal, The Tau weakmagneticdipolemoment.Nucl.Phys.B436,474–486(1995). arXiv:hep-ph/9411289 25. T. Stelzer, S. Willenbrock, Spin correlation in top quark pro- duction at hadron colliders. Phys. Lett. B 374, 169–172 (1996). arXiv:hep-ph/9512292 29. B. Grzadkowski, Z. Hioki, Decoupling of anomalous top decay vertices in angular distribution of secondary particles. Phys. Lett. B 557, 55–59 (2003). arXiv:hep-ph/0208079 26. G. Mahlon, S.J. Parke, Angular correlations in top quark pair pro- duction and decay at hadron colliders. Phys. Rev. D 53, 4886–4896 (1996). arXiv:hep-ph/9512264 30. J. Aguilar-Saavedra, J. Carvalho, N. Castro, M. Fiolhais, A. Onofre et al., Study of ATLAS sensitivity to asymmetries in single top events. Appendix A Diagram contributions Nuovo Cim. B 123, 1323–1324 (2008) 27. J. Bernabéu, G. González-Sprinberg, J. Vidal, Normal and trans- verse single tau polarization at the Z peak. Phys. Lett. B 326, 168– 174 (1994) 31. W. Bernreuther, A. Brandenburg, Z. Si, P. Uwer, Top quark pair production and decay at hadron colliders. Nucl. Phys. B 690, 81– 137 (2004). arXiv:hep-ph/0403035 123 123 123
https://openalex.org/W3032423260	https://nottingham-repository.worktribe.com/preview/4031300/Papadopoulos-Ontology%20of%20Resistance-To%20Production-Final.pdf	English	null	The ontology of resistance: Power, tactics and making do in the Vila Rubim market	Urban studies	2,020	cc-by	10,015	Abstract The paper re-examines the relation between power and resistance by investigating the reconstruction of the Vila Rubim market, one of the established markets in the city of Vitória in Brazil. After a fire that destroyed large parts of the market—probably the most significant event in its history—the market had to be fully rebuilt and the broader local area had to be redeveloped. Empirical materials were collected through ethnographic fieldwork, interviews and visual and archival research. The destruction and reconstruction of the Vila Rubim market unleashed a fierce struggle between the city council and the market's traders. We argue that the traders' resistance to urban management held the primacy in shaping the outcome of this conflict by initiating a multiplicity of space making practices. We reframe resistance as ontological, that is as the practice of creating a material position, of making a world that allows an alternative form of life to emerge beyond given power relations. Rather than in acts of protest, the stallholders of the Vila Rubim market engaged in mundane tactics which created alternative ontologies of existence in urban space. he Ontology of Resistance: Power, Tactics and Making Do in the Vila Rubim Mark A.d.P. Carrieri, D. Papadopoulos, E.A. Quaresma Jr, & A.R.L.d. Silva Urban Studies Keywords Ontological Organising, Primacy of Resistance, Tactics, Urban Space, Make Do, Management Ontological Organising, Primacy of Resistance, Tactics, Urban Space, Make Do, Management The Mercado da Vila Rubim is one of the few established markets in the city of Vitória in Brazil. It was destroyed by a fire in 1994. The market had to be fully rebuilt and the broader local area redesigned. Its reconstruction lasted 13 years. During this period and similar to its beginnings, stallholders and shopkeepers were selling their goods in the adjacent streets rather in the market itself. This period is the focus of this paper. The fire was a turning point in the life of the market—a myriad of new urban management practices and new forms of resistance emerged after the destruction. The oppositional relation between the city council and the stallholders is articulated through urban space and its material, ontological remaking. It is a continuous struggle, in which ontological resistance, a key term that we will develop throughout this paper, occurs almost daily in and around the market. We use the term ontological resistance to describe everyday political practices that rather than focussing solely on traditional representational politics, engage directly in the remaking of the material fabric of urban environments. Ontology here denotes that such everyday politics operate within the constraints of given material spaces and that its primary starting point is the transformation of the immediate material conditions of everyday existence. Alternative food growing practices, citizen science, urban gardens, squatting and repurposing, the caretaking of local ecological systems, community technoscience, alternative and traditional knowledge systems, the creation of DIY technologies, making, mending, repairing and hacking are examples of such ontological practices that are often connected to politics of resistance. The paper explores such politics in relation to urban space and argues that direct interventions in the material order of the urban environment can be a powerful tool in the articulation of contemporary politics of resistance that often drives how urban conflicts unfold. We engage with concepts developed in the works of Foucault, de Certeau and Deleuze to create a theoretical vocabulary that captures how alternative material practices, ontological organising as we will call it, give birth to everyday worlds of existence that come to challenge established power relations. We start with Foucault’s thesis on the primacy of resistance and his elusive claim that resistance comes first (section 1). Keywords We discuss these ideas with a specific focus on how resistance unfolds and erupts within the realm of mundane urban life. de Certeau’s concept of tactics provides a useful framework for translating the primacy of resistance to an everyday practice (section 2). In a third theoretical step we deploy the work of Deleuze to reframe tactics as a practice that reconfigures the ontological fabric of mundane urban life, something that we develop conceptually as ontological organizing (section 3). It is important to note here that our main focus is not on these three theories as such but on the attempt to assemble a conceptual framework for understanding ontological resistance in the case of the Vila Rubim market. In the sections that follow we present and analyse the empirical materials that we have collected through our fieldwork on the history of the Vila Rubim market: we start with a discussion of how the market has evolved since its beginning (section 4), the events of the its destruction in 1994 (section 5) and the subsequent conflict between the stallholders and city management over its redevelopment (section 6). We then come to discuss how ontological organizing unfolded through different cycles of conflict that erupted during this process (sections 7-9). In the final concluding sections of the paper we discuss the world making capacities of ontological organising and reframe resistance as a form of politics that does not primarily make claims directly to power but changes the underlying material conditions by which power operates (section 10). Ontological resistance is about creating many minor alternative worlds within urban space and it is here that we see the implications of this form of organising for rethinking and practising resistance in a way that could potentially enhance its efficacy in today’s tumultuous urban life (section 11). 1 The primacy of resistance in urban space Although there is significant scholarship on the emergence of non-oppositional ontologies, affective geographies and their complex material configurations (Braun, 2008; Clark, 2011; Tolia-Kelly, 2006; de Vries and Rosenow, 2015), there is less exploration of what this really means for an everyday politics of resistance within urban space. This paper approaches this question through an empirical case study by simultaneously advancing a theoretical discussion of resistance driven by the work of de Certeau and Deleuze. Here we conceptualise resistance as the creation of everyday material alternatives, an approach that stands against ‘standard definitions of resistance within geography, which identify it as the resisting of domination or oppression, building on well-established political movements’ (Legg, 2018: 28). Such a conceptualisation of resistance starts with problematizing dominant interpretations of Foucault's analysis (1978) of power and resistance which assume the existence of a merely coercive power and treats resistance as a mere reaction to it. Such an approach to power and resistance has shaped our understanding of conflict in urban space: see, for example, Ju and Tang (2010) on grassroots environmental groups against the South Korea government, Lauermann and Vogelpohl (2019) on protest campaigns against the organisation of mega-events in Boston and Hamburg, Davies and Blanco (2017) on contentious anti- austerity politics, Pearsall (2013) on anti-gentrification struggles in New York or work describing different modes of resistance to surveillance (Swanlund and Schuurman, 2019; Grommé, 2016). Although this work is important and, indeed, necessary for understanding and visualising oppositional politics, it has been also problematized because it 'draws a strict contrast between the diabolic world of power and the liberating world of resistance' (Fleming and Spicer, 2008: 304). The second line of interpretation of Foucault’s work states that there is no real division of power and resistance and implies that ‘[t]he operations of power, domination and resistance are seen as integrally rolled up in articulations of society and space, resulting in the entanglement of resisting and dominating practices’ (Routledge, 2009: 647). Power and resistance are seen as indivisible, following Foucault's (1978: 95) mantra that wherever there is power there is also resistance. Resistance is always an inflection of power. Spinney (2010), for example, analyses urban cycling and argues that practices which have been widely seen as a performance of resistance are a performance of a power that is always retransforming and reshaping local actions. 1 The primacy of resistance in urban space It is from this second line of interpretation that our discussion of the dynamics of power and resistance in urban space starts. de Certeau (1984) provides us with valuable help for understanding relations of power in this context: strategies seek to prevent the tactics that attempt to appropriate the city and its meanings by those who move beyond acceptable limits. Strategies act upon the actions of such individuals or groups attempting to transform their everyday modes of existence. And beyond that, Carrieri et al. (2009) have shown in their discussion of the introduction of citizens' disciplining programmes in Belo Horizonte through the city's 'Code of Postures' and the 'Centro Vivo' that strategic actions not only attempt to direct the everyday actions of people but also to shape the coordinates in which potential resistance can take place (see also Carrieri and Murta, 2011). de Certeau (1984: xix) defines strategy as ‘the calculus of force-relationships which becomes possible when a subject of will and power (a proprietor, an enterprise, a city, a scientific institution) can be isolated from an "environment." A strategy assumes a place that can be circumscribed as proper (propre) and thus serve as the basis for generating relations with an exterior distinct from it.’ This interplay between power/strategy and resistance/tactics seems to be able to reveal the mechanics of city governance and how urban space is made (Foucault, 1980: 149). From this point of view, urban space is organized through strategic practices that respond and subsequently provoke the tactics of its inhabitants. And this is the reason why the second line of interpretation mentioned earlier appears to be plausible and is so widespread: power and resistance are conceived as inseparable, each one responds to the other in an eternal linear succession of strategies and tactics. But to what extent is this the case? What if we consider resistance as the outside of an existing order of power relations (Deleuze, 2006) that drives transformation and changes its very configuration? What if resistance has the primacy and tactics shape strategy? The starting point of this approach can be still found in Foucault’s work. He miraculously says that '[i]f if there was no resistance, there would be no power relations. Because it would simply be a matter of obedience. You have to use power relations to refer to the situation where you're not doing what you want. 1 The primacy of resistance in urban space So resistance comes first, and resistance remains superior to the forces of the process; power relations are obliged to change with the resistance. So I think that resistance is the main word, the key word, in this dynamic' (Foucault, 1997: 167). In this last turning point, Foucault assigns to power the ability to define the structure of social action and potentially shape how future actions will be articulated. But paradoxically this capacity of power relations to structure social action is not driven by power itself but by resistance (Checchi, 2014). This paper advances this argument, namely that resistance does not reshape power when it directly opposes it but when it creates alternatives that escapes it. How can we evade power relations in order to materialise transformative resistance? How can this be done within urban space? 2 Evading city management: Tactics of resistance Urban space is permeated by divergent discourses and practices capable of managing transactions that daily construct it. One of the most prominent of these discourses is urban management and planning (Limena, 2001). Urban management is generative in its nature: it gives birth to a multiplicity of other practices that vary enormously, from discourses about where certain businesses can and cannot be located (Parker, 2011) or debates on how to accommodate citizen disagreement (Özdemir and Tasan-Kok, 2017) to the management of movements of people (King, 2010), to tightly linking areas of cities to specific activities (Carrieri et al., 2009; Carrieri and Murta, 2011), and so on. Urban planning and its actors can be associated with what de Certeau (1984) considers as a practice of a 'subject with will and power' who incites individuals or social groups to specific forms of behaviour and ways of acting. Nonetheless, an actor with power does not mean that it owns the power it possess—see an example in Wideman and Masuda (2018) on strategic government interventions of the city council of Vancouver (see also Williams et al., 2012). Such social actors, we could say with Deleuze (2006: 75), 'are not sources or essences, and have neither essence nor interiority. They are practices or operating mechanisms which do not explain power, since they presuppose its relations and are content to "fix" them, as part of a function that is not productive but reproductive.' When seen only from the perspective of strategy, urban space appears to be managed by the 'powerful' and marked by injustice (Williams, 2016). From this point of view the relationship of power and resistance remains trapped, as discussed earlier, in an eternal dialectic and complicity: power tries to fix resistance 'in the diagram,' as Deleuze would have said, and resistance tries to confront power. When we talk about resistance as confrontation to power then we propose to talk about protest, opposition, or revolt. In this configuration, power and 'resistance' (as protest, opposition, or revolt) are in tight and infrangible connection, one opposing the other. In fact, one could say that 'resistance' as opposition becomes part of fixed relations and enters the nexus of existing order of forces—Checchi (2015) provides an insightful analysis of these different readings and meanings of resistance in his work on Foucault. 2 Evading city management: Tactics of resistance However, in this paper we try to conceptualise resistance in a different way: as 'local, unstable and diffuse', as something that does not emerge in direct relation to power and does not 'emanate from a central point or unique locus of sovereignty' (Deleuze, 2006: 73). Resistance here is not what opposes power but what exits power by creating a locus that exists outside of existing power relations. If Deleuze (2006: 75) argues that ‘there is no state, only state control, and the same holds for all other cases’ then in a similar manner we can say that there is no city but urban management and the same applies to all other organizations within. When institutions are seen as open forms of practices, power and resistance are no longer just in a tight inseparable relation because none of them simply ‘owns’ the power over the other. As we will show later in our case study there are no stallholders against the city council and when the former act in ways that cannot be predicted, expected or calculated by the city management, their actions emanate from outside of the existing order of power relations within the city. With de Certeau we can call these actions ‘tactics’. What characterises tactics is not that it opposes strategy's plans but that it subtracts itself from the locus of strategy. 'I call a "tactic" […] a calculus which cannot count on a "proper" (a spatial or institutional localization), nor thus on a border-line distinguishing the other as a visible totality. The place of a tactic belongs to the other. […] The "proper" is a victory of space over time. On the contrary, because it does not have a place, a tactic depends on time—it is always on the watch for opportunities that must be seized "on the wing"' (de Certeau, 1984: xix). Tactics are tricks for evading the fixed order of power relations. Tactics survive not because they confront power but because they use cunning to escape from it (Smith et al., 2015; Papadopoulos et al., 2008). Tactics open continuously new ways of existence, create new worlds. ‘Power’ has to follow and recapture these spaces. 2 Evading city management: Tactics of resistance It is in this sense that tactics exemplify the primacy of resistance that we mentioned earlier: Tactics, as resistance, can be conceived as an creative force which cannot be explained by the given order of power fixed by strategy but introduces new unexpected ways of acting. Tactics interrupt the generativity of urban management and create new ‘organizational ontologies.’ 3 Making do and the organizational ontology of tactics If strategy lies within the existing order of forces, tactics are 'multiform, resistance, tricky and stubborn procedures that elude discipline' (de Certeau, 1984: 96). de Certeau offers a glimpse on these everyday actions that are characterised by the art of 'making do': by showing how people find in their daily activities ways to overcome prescriptions using ingredients that can be found within fixed power relations; by carrying out ‘coups’ that are disguised in particular forms of reading, making, walking, relating, speaking, cooking, selling or mundane organizing. Studies such as that of Pinder (2011) on walking, of Junquilho et al. (2012) on public schools, of Carrieri et al. (2019) on circuses, of Kokkinidis (2014) on workers’ self-management or of Spinney (2010) on urban cycling show how the (re-)appropriation of power is founded on these micro-liberties that emerge as people engage in their everyday life 'since they lack their own space, have to get along in a network of already established forces and representations (for a discussion see Machado et al., 2017). People have to make do with what they have' (de Certeau, 1984: 18). de Certeau here helps clarifying Foucault's cryptic assertion on the primacy of resistance mentioned earlier. Resistance and power operate with the same stuff of life but they are not in direct conflict. With the work of Deleuze (2006), Checchi (2015) and Papadopoulos et al. (2008) we propose to read resistance (that is resistance as resistance, not as opposition) as a movement of exit and escape of the prevalent power relations in a certain field. Resistance is here not about direct confrontation but about evading the conditions in which power relations operate by arranging the mundane conditions of existence in ways that allow for such an exit to take place. In order to facilitate such an escape a community needs to organize its material independence, food, shelter, care, access to resources, alternative forms of cooperation, suitable modes of exchange. The making do with what is available to the community in each certain moment in order to be able to resist is what we call organizational ontology, a term which we borrow here (Papadopoulos and Tsianos, 2013; Papadopoulos, 2018) to describe the practice of resisting power by organizing alternative forms of life: alterative ontologies. In our use of the concept of ontology and materiality we rely on existing debates in anthropology (e.g. Holbraad et al., 2014), geography (e.g. 3 Making do and the organizational ontology of tactics Braun, 2008), urban studies (e.g. McFarlane, 2011; Williams, 2016), organization studies (e.g. Carlile et al., 2013; Silva and Silva, 2019), science and technology studies (e.g. Woolgar and Lezaun, 2013) and political theory (e.g. Braun and Whatmore, 2010). Although it is beyond the scope of the paper to engage with these very diverse debates directly, they inform our discussions of the term and the way we deploy it here. In particular, when we refer to ontology we mean the more-than-human capacity of certain actors (such as a group of humans or members of an animal species or certain objects etc) to change the material configuration of their concrete space of existence (Papadopoulos, 2018). Within this framework, one needs to deconstruct the idea of a deliberate intention to remain outside an existing order of power because it reifies the existence of a centrally organized subject (vis-à-vis power). The inside and the outside loose gradually meaning as communities create alternative ontologies of existence. The subject of tactics is not collectively organized as such, that is as a proper subject. Rather it often acts without even articulating its own refusal of power. In this sense resistance is not just dissent or withdrawal from power as a symbolic act. Resistance is a practice of creating—literally, that is ontologically—a position, of making a world that allows an alternative form of existence beyond given power to emerge (Whitson, 2007). When resistance is resistance (and not opposition, protest or revolt) it is with de Certeau the tactics of everyday life that destabilise power and change its fixed order. Insert Figure 1 In the decades after the establishment of the market, traders occupied the space around its original 1928 building selling many different products at wooden stands often described by our interviewees as precarious and temporary. In 1969 the vendors who were trading with fresh fruit and vegetables were transferred to three new sheds (named on Figure 1 as 'Galpão 1' and 'Galpões 2 and 3') very close to the market's original building. In 1970 an area around these new sheds was given also to traders who worked in the original building selling other types of products. This made the demolition of the original building possible and allowed the construction of an open public space, which was named as the ‘Manoel Rosindo square’. The city council prohibited its use for selling products. But this would not last long: 'We organized a meeting. There, I said this: "Let's break into that square there. The beggars are there, we kick them out and occupy the place." Everybody was there. Each one took a stall [...] it happened through three invasions. ' (Interview E14, 55). This episode led the Manoel Rosindo square to be commonly called 'Praça da Feirinha'/'Market Square'. The installation of makeshift booths and the somehow disorderly use of the square as a market continued until the 1990s as we can see in Figure 2. There are people waiting for the bus very close to the fresh food market stalls. The transient stalls of the vendors changed the material formation and use of the square permanently. This happened without any formal negotiation. Instead of opposing and challenging formally the city council's decision to keep the square free of trading, stallholders gradually occupied the space. Instead of opposition we see the silent introduction of the ontological and a-subjective tactics of everyday life. 4 Vila Rubim market, before the fire The Mercado de Vila Rubim market in Vitória, Brazil was inaugurated in the homonymous neighbourhood in 1928. It is located in the centre of Vitória, the capital of the south-eastern state of Espírito Santo with a population of 327,000 (IBGE, 2010). The economy of the region is driven by mining, the steel industry, cellulose production and petrochemicals, all tightly linked to the ports of Vitória and the adjacent Tubarão. The economic wealth of the region is in stark contrast to the decline of Vitoria’s inner city. A gentrification strategy was designed to address this issue by promoting historical heritage and transforming this widely ‘demonized’ area--including Vila Rubim and the market--to the guarantor of the cultural identity of Vitória (Botelho, 2005). We investigated the history and development of the market and the surroundings through a multiplicity of qualitative methods since 2006. In the first phase of the ethnographic study, fieldwork in the market was carried out for three months, followed by a much longer engagement with the traders of fruit and vegetables produce. The materials collected from this research fed into an extensive field diary. Alongside fieldwork, written and photographic documents were collected from the Vila Rubim Traders Association (ACVR), from the Association of Stallholders of the Vila Rubim market, from the traders themselves and from the Municipal Public Archives and the State Public Archives. In the second phase of the study we continued fieldwork and conducted interviews with customers, suppliers and traders: 24 stallholders (five fresh market dealers, five former traders and 14 belonging to other segments); five suppliers (three of fresh goods); nine customers; two representatives of associations; eleven employees; three police officers. In the current paper we will primarily focus on some of the visual materials underpinned by fieldnotes and interviews. Insert Figure 1 Insert Figure 2 'The stallholders' as an actor was not formed as a socio-political subject before the invasion of the square. In fact, this subject doesn’t exist as such. In tactics, subjects and their organizational ontologies do not exist before their transformative practices of urban space. In other words, subjects exist to the extent that they materially change space. There was not a collective subject of 'The Traders of Manoel Rosindo Square' before the square was occupied. It formed after the event: 'There was a moment when the mayor gave us 24 hours to get out. Then everybody met: "What are we going to do? We received a notice to leave." [...] We sought a local politician of that time. We explained the situation to him, and he said this: "Oh! Form an Association."[...] Then we went to court. It is there for 12 years now' (Interview E14, 51). And even after the event the presence of such a subject was rather transient. It existed in the form of an association, but the association itself was not the locus of transformative action. It was occasionally mobilised in negotiations with urban management but action was gravitating primarily around the mundane practices of the a-subjective community of the stallholders trying to make do with the existing conditions in Vila Rubim. It is important to emphasize here that when we refer to the ‘the stallholders’ throughout this paper we do not mean a formally and coherently organized social group or a unitary and distinctive collective subject. We rather refer to a multiplicity of actors that forms a community of action as long as it engages in transformative ontological organizing. Such actor appears at certain moments as a powerful and seemingly solid and unified subject but it is in fact a very diverse, occasionally ambiguous and internally differentiated community that coheres around its participation in the making of a specific form of life. Thus, there was neither a ‘subject’ that took over the square at once by decision nor did it take over the square entirely and permanently. Rather, the becoming a ‘subjectivity’ of the stallholders unfolds as they remake the square through a series of mundane material transformations and form gradually a fluid community of common experience and action (Stephenson and Papadopoulos, 2006). Sidewalks change their function as stallholders set up boxes of products and umbrellas to increase shadows for their goods and clients. Insert Figure 2 Some of these makeshift stalls become semi-permanent changing the conditions of visibility and movement of the square. Bricks, stones, wooden planks, corrugated metal, fabric, pieces of plastic. Traffic around the square changes, more people passing and stopping, cars parking, noise, occasional shouting, honking, other sounds and smells. The use of the inner part of the square behind the stalls changes too, to accommodate people who use it to sustain a life on it rather than just visiting it recreationally. As the vendors make do with the materiality of the space there is available in order to extend the occupation of the square, they ‘constantly manipulate events in order to turn them into “opportunities”’ as de Certeau says (Silva and Silva, 2019). Day by day, year by year, square meter by square meter. In this process their subjectivity as “the stallholders of Vila Rubim” adapts and transforms to their practices and to the creation of their alternative material conditions of existence. Rather than through the formation of a clear-cut social subject, it is through time and materiality that tactics transforms space. This is the cunning of tactics. 5 The explosion In the meantime, the precarious status quo of those who were selling their products in the three main sheds built in 1969 changed unexpectedly in a tragic way. In a moment in which the Vila Rubim market seemed to have found a relatively stable arrangement of its permanently transient and shifting existence a violent event changed its very material configuration. The tactics of the traders have taken advantage of several inconsistencies in the regulations of urban management and its incapacity to enforce these regulations: they perverted the strategies that defined what could or could not be sold. The initial planning was that only fresh produce should be sold in the sheds. Gradually though, several other goods were traded in the market, including fireworks. This led to a situation which became a neuralgic point in the relationship between the traders' tactics and the strategy of urban management: in July 1994 a fire destroyed much of the market. This event is often called 'the explosion' by many of our interviewees. 6 The primacy of ontological resistance: Occupy and transform After the explosion urban management designated a place where the traders could resume selling their products. The shops that housed the old market vendors were almost completely destroyed, in addition to sheds 2 and 3 (Galpões 2 and 3 of Figure 1), which were mainly used for the trade of fresh produce. The stallholders were allocated in booths built on Nair Silva Azevedo Avenue (see Figure 1), a nearby market area 200 meters away from the original sheds. However, according to our interviews, after a few months there was a further subversion of urban management by the stallholders: because the new designated space was isolated and sales decreased they returned to the destroyed market and they built new stalls to work offhand (as it can be seen for one small part of the market in Figure 4). 'When the explosion happened, I worked [...] there, near the petrol station. The government arranged that for us. We worked, more or less, for about 60 days. We saw that this wasn't the way to go. Then we went back to the street. [...] My shop was right here, but it was all blown up [...] everything was burned, but we made a little shed inside where we were before. Then, they removed us from there and built for us something provisional, until we return. This was the only way. The city council made a promise, we leave from here, to work there. When they did [the city council constructed the provisional sheds], we went back to the same location' (Interview E16, 48). After the explosion urban management designated a place where the traders could resume selling their products. The shops that housed the old market vendors were almost completely destroyed, in addition to sheds 2 and 3 (Galpões 2 and 3 of Figure 1), which were mainly used for the trade of fresh produce. The stallholders were allocated in booths built on Nair Silva Azevedo Avenue (see Figure 1), a nearby market area 200 meters away from the original sheds. Instead of entering into a power struggle with the city council about the relocation of the market to Nair Silva Azevedo Avenue the traders simply returned to the original site. Instead of seeking an institutional conflict they organized themselves by reclaiming and reshaping the old market. Insert Figure 3 Insert Figure 3 In Figure 3, we can see what happened after the explosion of the fireworks stored on the site. Half of the long shed and shops disappeared along with much of two parts of the main building. Our conversations with stallholders reveal that it is important to realize that the issue here is whether there should have been fireworks on the site or not. The urban management's effort to 'fix' the everyday life of the market in a stable order and to impose clear-cut rules (for example by controlling which commodities should be traded and which not) is irrelevant from the perspective of people who are making do with what is available (Correia et al., 2018). The organizational ontologies and the tactics of the stallholders eluded strategy as they tried to support themselves and their families and to sustain their existence day by day. Fireworks—as well as the existence of many other commodities that we do not know of—was a minor part of these tactics, one though which changed the life of the market entirely. And although the explosion destroyed the market, one could say that the fireworks where immanent to it. An ethical judgement is inappropriate here. The stallholders respected the necessities of their own everyday conditions of existence. They had to make do in order to be able to continue to make a life. 6 The primacy of ontological resistance: Occupy and transform They, in fact, subtracted themselves from a direct power confrontation and they gradually started materialising their position through the making of new stalls around the destroyed market. Insert Figure 4 In the centre of Figure 4 is the cleared area that was previously occupied by sheds 2 and 3 and around them we can see some of the traders' improvised booths. On the top right of the lot it is possible to see a rooftop of shed 3 that housed 'Supermarket A' the only store that survived the explosion. Apart from 'Supermarket A' almost all of the rest of trading in the Vila Rubim market was done through the improvised ontological tactics of the stallholders as they went back to the area around the destroyed building. Nonetheless, strategy rises again as the city council approves a new management agreement and commits to rebuild the sheds. Meanwhile, traders were supposed to move from where they were to work in purpose build wooden huts erected by the city council just in front of the market. The persistence of the vendors to evacuate the space of the Nair Silva Azevedo Avenue and to relocate their activities around the original site sets up the material constraints against which the city council made its decision. Resistance to the original decision of the city council drives change and becomes a force for co-constructing urban space. Here we can see the meaning of the primacy of ontological resistance discussed earlier: firstly, resistance became the primary force of change because it did not confine itself to the coordinates that where defined by power (that is urban management’s decision to relocate the traders to Nair Silva Azevedo Avenue). Secondly, resistance and the evasion of power wasn’t just protest and opposition to the plans of city council; rather it become an ontological force, in the sense that it materialised through reclaiming the space around the market. The refusal of the traders took the form of an organisational ontology as a 'politics of matter' (Papadopoulos, 2014): by reshaping the space around the market, installing semi-permanent makeshift booths, transforming materially the surrounding streets and sidewalks, changing the flows of people and traffic, modulating the affective economy around the market, and, finally, by modifying the conditions of visibility. The tactics of the traders appear as an anti-discipline: a form of micro-liberty that holds the primacy over the strategy of the council. The primacy of ontological resistance in our case study means that it created a new practical and material space of existence that urban management could not neglect or bypass. 7 From ontological organizing to representational politics and back The traders remained in their provisional stalls, eventually co-constructed by the city council, for several years. During that period, stallholders included a new political repertoire to their practices. In addition to the ontological tactics discussed so far they engaged also in formal representational politics. The first type of this politics focussed on activities that attempted to change the image of the stallholders and to connect to local instituted politics. Many of these political actions were performed by the Vila Rubim Traders Association but used by the traders as a whole, such as organising public meetings to express their political ideas, handing out promotional materials, distributing leaflets and hanging posters. The second set of political activities involved an intense engagement with the media which occurred in association with the commemoration of the anniversary of the explosion. The press was invited, the city council was also present, there were speeches and a cake was made specifically for the occasion. The media coverage was mobilised to expose to the public the failures of urban management in relation to the delay in the re-construction of the destroyed sheds. The politics of representation inside existing power relations is very different to the politics of resistance as ontological tactics that we described in the previous sections. Our case study shows that they are not exclusive and in fact they often are part of the mobilisations of a certain community as the one described here. However, as much as representational politics was important in the actions of the community, we argue that such politics would not have much traction without the long ontological organizing that allowed the traders to materialise— literally—alternatives to the one proposed by the city council. It is this materialisation that effectively facilitated the emergence of representational politics that in turn strengthened the collective actions of the stallholders and allowed them to create alliances within the neighbourhood and to expose and oppose the strategy of urban management on a formal political level. We are used to recognise representational oppositional politics as the quintessential form of politics while here we argue that this politics is often impossible without the ontological tactics of resistance. Of course, ontological resistance might be not always the most appropriate tool for alternative political action. 7 From ontological organizing to representational politics and back In certain cases representational politics or, indeed, other forms of politics and social movement action that aim to include alternative positions in mainstream instituted power might be necessary. Rather than elevating one type of politics to the centre of the practice of resistance we seek to oppose the implicit universalism of representational politics as the main pathway to politics and to open our sensibility to other forms of political practice that might be becoming increasingly pertinent for an effective performance of resistance in today’s social and political conditions—for a further discussion see Tsianos et al. (2012); Papadopoulos (2018). 8 The reconstruction of the market and the provisional dominance of urban management It is in this context and with much delay that the definite plans for the reconstruction of the destroyed sheds emerged. As the city council included the demolition of 'Supermarket A' (the only trading store that was not destroyed by the fire and was still operating in the market) a new struggle began when the owners opposed the council's plans. This conflict resulted in a court case that lasted until 2002, eight years after the fire. The impasse was resolved when the owner of 'Supermarket A' acquired a space just opposite of the destroyed sheds and built a new store there (see also Figure 1), thus withdrawing from the sheds and leaving them empty. Afterwards and when all legal issues were settled the city council carried out the reconstruction of the sheds according to a business plan which included the reinvention of the market's profile and activities and a reorientation towards tourism (Botelho, 2005). Since then, handicrafts began to be the predominant merchandise traded in the sheds. The new facilities include 38 stores and a mezzanine where public events can be performed. These plans initiated a new cycle of strategic and tactical actions. This new cycle of conflict revolves around attempts to gentrify the broader neighbourhood (Butler, 2007). As in many other similar cases, one of the main drivers behind gentrification is the removal of portions of the population to less central or disputed areas of the city (Lees, 2012) and the maintenance of a specific social order (Uitermark et al., 2007). Economic interests dictated this decision and resulted in segregating those who were not classified as being able to sustain the new business development plan that was promoted by urban management. The plan for the Vila Rubim market was based on an urban planning blueprint that tried to introduce local craftsmen into a larger commercial area that should provide a new touristic hub gravitating around cultural and historic heritage. In discussing the work of Foucault, Díaz (2012) states that prevalent rationalities of urban power are built to exclude the supposedly 'non-rational', that which is incomprehensible, everything that diverges. 8 The reconstruction of the market and the provisional dominance of urban management City management, as the bearer of legitimate institutional reason of urban planning, invented administrative practices that portrayed and classified the traders, the street peddlers, the working families of the local community, the invisible citizens of the neighbourhood as subjects that cannot support the new business plan and do not fit within the limits of what is reasonable in this new situation for development. As in many other urban redevelopments (see for example a similar case in Belo Horizonte, Carrieri et al., 2009) also in the Vila Rubim market the attempt was to restrict, control, delimit the field of action of those who cannot be inserted into the new configuration of urban space. The new sheds resemble a small suburban shopping centre, with more ambient spaces and fewer stores (the more than 200 stalls that previously occupied the two sheds were reduced to only 38). This is similar to what de Certeau (1984) calls as a proper: a space victory over time. Only few of the traders who sold their goods in the booths in front of the sheds built by the city council during the previous cycle of resistance after the fire were selected and transferred to shops inside the sheds while the remaining booths were relocated by the city council to the Manoel Rosindo square, mentioned earlier. The stallholders accepted the transfer on the promise that the square would be also reconstructed and new permanent stalls made of brick would be built. It seems that the phase of reconstruction in the history of Vila Rubim was now dominated by urban management's strategies at the expense of the ontological tactics of the traders. now dominated by urban management's strategies at the expense of the ontological tactics of the traders. 9 The long duration of resistance With de Certeau (de Certeau, 1984: xix) we have argued earlier that tactics does not have a “proper place,” that is a fixed and stable position from which it acts within a certain situation. This is reserved for strategy and power. The tactics of resistance rely on looking for the right moment for action in order to have a chance to be successful. Finding a window of opportunity extends the timeframe of resistance and extends its duration to include periods in which resistance might seem to be dormant or even inexistent. Ontological organizing depends on time in order to be able to transform existing ontologies outside the order of power. In the contrary, opposition, protest and revolt are space bound and they almost react instantly to power. Resistance has a long duration and can only unfold when the right moment is there. The apparent dominance of strategy that we described in the precious section was only ephemeral. The stallholders waited, seemingly accepting the decision to expel them from Vila Rubim, and then at a certain moment they returned back. Silently but decisively. Figure 5 shows the return of fruit and vegetables stalls in front of the new reconstructed market. Insert Figure 5 Insert Figure 5 Once again the traders reclaimed the space of the market by—literally—remaking the spatial order of movement and visibility. Commonly used as an extension of the counter, the produce crates occupy a part of the sidewalk as before, allegorically and functionally, a practice that is, as ever before, prohibited by the city council. '[...] some time ago, the city council officer used to come to grab the tray, the stallholders used to go against him, to beat him, you understand [...]. Today this does not exist [the fights]. [...] The officer is passing by all the time, with the truck. [He says:] "Hey! If I arrive here tomorrow and those goods are here [pointing to the side walk], I do not want to know who is the owner, I will take it." [...] but here is like this: it is very difficult to sell goods that are inside [not presented on the sidewalk] [...], so we stop for two or three days, without putting anything outside. Then, you look around and there they are again: "Gee, one of the other guys put [the goods on the sidewalk]." [...]. "I'll put mine too." In a little while, about 15 days pass and the officer returns [...] and everyone gathers all [the goods] without any protest' (Interview E04, 52). Ontological resistance does not only change the materiality of public space and of people’s movements in it, the shape of the streets around the market and how one walks but also the field of vision (Smith et al., 2015). People in cars or buses passing in front of the market are unable to see the small shopping centre that was the outcome of the restoration of the destroyed market sheds. Many of our interviewees said that after years of reconstruction the Vila Rubim market returned to the shape it had before the fire. This return reveals another implicit dimension of the temporality of ontological resistance. The future is not simply what lies ahead, because such future is often decided and executed by power and strategy. Rather, from the perspective of tactics the future means often to return back in order to reclaim the past in a different way within the present (Papadopoulos, 2018). Ontological resistance operates within the present but its duration is long and its movements often hybridise the linear temporal trajectories which power relies on. 10 Between ontological tactics and symbolic power: Urban interspace Although the market survived many attacks on its existence over the course of several decades it finds itself today again in a very precarious situation. 'Here [the market] was like a bus terminal [...] So here was where one could change to other buses, the movement here was intense. Trade was very good, nowadays it does no longer exist. There are direct terminals there. [...] And everyone was shocked with this' (Interview E10, 46). There is a progressive abandonment of the central areas of the city towards its suburbs and the Vila Rubim market could become a victim of this urban transformation. Throughout this paper we tried to describe how the traders of the market and their creative evasion of urban management became a force that literally contributed to the making of urban space. The ontological fabric of urban space is co-constructed; it is an interstitial field shaped by ontological resistance and urban management in a continuous process of transformation (Swyngedouw and Kaika, 2003). Urban space exists as interspace. Vila Rubim’s whole history, especially since the 1960s, was made in these interstices between resistance and power: in the initial planning of the three sheds, each of the dozens of available stores should be occupied by one dealer of fruit and vegetables. However, many traders occupied more than a single space and began selling various products such as clothes, handicrafts, religious artefacts etc. By inserting new material configurations that silently defied council regulations the tactics of the traders created a precedence that has since then accompanied the market in all its developments, including of course the 1994 fire. In Figure 1 one can see how these ontological tactics established a different meaning of the Vila Rubim market: the reconstructed sheds 1, 2 and 3 between Rua Jair Andrade and Rua Orlando Rocha are formally considered as the Vila Rubim market according to the addresses and documents available at the city council. However, for people who work and come to shop or visit the market, Vila Rubim includes not only the sheds, but also all other stalls of Jair Andrade Street, the Manoel Rosindo Square, the adjacent supermarkets, groceries and shops and all other transient vendors of the area. 10 Between ontological tactics and symbolic power: Urban interspace This resilience of the ontological materialization of the meaning of Vila Rubim prevailed over attempts to delimit the symbolic and physical space of the market and to confine it to the actual space of the shopping mall. On the map in Figure 1 the Vila Rubim market corresponds to the light grey shaded area with a circumference of approximately 200 meters, a space which exceeds by far the actual space of the Vila Rubim market itself. As King (2010) shows, urban management has power over the symbolic and representational processes that occur within city space. And although this power over the symbolic and the process of naming and classifying has the capacity to create spaces of exclusion and inclusion and to impose new forms of social order, ontological tactics have throughout Vila Rubim’s history undermined the instalment of any fixed and final meaning of how specific parts of the area are called or what they are used for. Against power’s rein over the symbolic the ontological tactics of the traders had constantly shifted the dominant meanings and made them permeable to other interpretations, other practices and other forms of existence (Silva and Silva, 2019). Meaning in urban interspace cannot be owned, at least not indefinitely by any actor. Meaning is constructed in the indeterminate and creative interspace of symbolic power and ontological resistance. 11 Conclusion: Urban space making as world making Resistance happens when social practice materializes—literally, ontologically—alternatives which do not exist as such within the range of possibilities of strategy and power. This might sound contradictory but one has to consider, as Checchi (2014: 209) says, that ultimately ‘resistance desires not to be recognized as such.' And exactly this is what happened several times in the history of the Vila Rubim market. Representation is never permanently fixed and meaning is never ultimately given by power. After all these years of reconstruction, Vila Rubim seemed to have returned to its past, to the times of makeshift wooden stalls, in contrast to the 'modern' commercial tourist hub that was supposed to become. Ultimately, the traders did not manage to reverse the plan for reconstructing the market as an open shopping area and as a tourist attraction. But they managed to reinsert themselves into these new conditions and to re-establish their livelihoods. Resistance in this case is not primarily about dialogue and even less about dialectics. Resistance (almost) never delivers a final and clear-cut resolution to a conflict. Rather, resistance in our case is about the subtraction from a conflict in order to create vitally divergent conditions of existence that cannot be avoided by power. It resembles the creation of ‘forms of life’ (Wittgenstein, 1958: 226) which cannot be bypassed or neglected. This is the reason that we have called the practices that establishes alternative forms of life ontological. Resistance creates ontological constraints to power: it changes power not by addressing it directly but by changing the material conditions by which power operates. The target of ontological resistance is not power but world making. With Ranciere (1998: 42) we can say that '[p]olitics is not made up of power relationships; it is made up of relationships between worlds.' But here we need to take the meaning of 'world' literally and not metaphorically: as relationships between material worlds. When resistance is at stake, we make different worlds: alterontologies (Papadopoulos, 2018; Papadopoulos, 2011). The people who work and visit the Vila Rubim market know that the market encompasses—physically, affectively, materially, symbolically—an assemblage much more complex, large and open- ended than what the city council has planned (Farias, 2011). In our analysis of the history of the Vila Rubim market we tried to uncover another type of resistance—ontological resistance—which is often less visible than common forms of protest, opposition and revolt that are addressed towards power. 11 Conclusion: Urban space making as world making And although we think that universalising ontological resistance as a form of political practice is neither possible nor desirable, we would claim that perhaps it is the changing and troubling nature of today’s social conflicts that allows us—and indeed obliges us—to invent alternative political practices such as ontological resistance. We therefore tried to generalise our case study to theory and to capture this practice within the conceptual framework of ontological organising, hoping that this might be a useful tool for approaching and practising such forms of resistance in diverse locations and geographies. Within this framework we tried to show that urban space is not the outcome of power antagonisms, clashing interests and struggles between less powerful and more powerful actors; urban space is the outcome of the practices of those who try to create their own minor worlds inside it. References Botelho TR (2005) Revitalização de centros urbanos no Brasil: uma análise comparativa das experiências de Vitória, Fortaleza e São Luís. EURE (Santiago) 31(93): 53-71. n B (2008) Environmental issues: inventive life. Progress in Human Geography 32(5): 667-679 Braun B and Whatmore S (2010) Political matter: technoscience, democracy, and public life. Minneapolis, MN: University of Minnesota Press. p , y Butler T (2007) For gentrification? Environment and Planning A 39: 162-181. ( ) g g Carlile PR, Nicolini D, Langley A, et al. (2013) How matter matters: objects, artifacts, and materiality in organization studies. Oxford: Oxford University Press. Carrieri AdP, Maranhão CMSdA and Murta IBD (2009) Crítica ao manejo humano em Belo Horizonte. Revista de Administração Pública 43: 1315-1342. Carrieri AdP, Michelatto Natt ED and Aguiar ARC (2019) Brazilian canvas circuses: A field in constant motion and symbolic transformation. Revista de Administração da UFSM 12(2): 233-254. Carrieri AdP and Murta IBD (2011) Cleaning up the City: A Study on the Removal of Street Vendors from Downtown Belo Horizonte, Brazil. Canadian Journal of Administrative Sciences / Revue Canadienne des Sciences de l'Administration 28(2): 217-225. ( ) Checchi M (2014) Spotting the Primacy of Resistance in the Virtual Encounter of Foucault and Deleuze. Foucault Studies 18: 197‐212. Checchi M (2015) The Primacy of Resistance: A Historical and Conceptual Exploration. University of Leicester: PhD Dissertation. Clark N (2011) Inhuman nature: sociable life on a dynamic planet. London: SAGE. Correia GFA, Pereira HG and Carrieri AdP (2018) “Ser um ambulante é necessidade que nós temos de trabalhar”: cotidiano e identificação de trabalhadores pipoqueiros de Belo Horizonte. Cadernos De Psicologia Social Do Trabalho 21(2): 165-180. Davies JS and Blanco I (2017) Austerity urbanism: Patterns of neo-liberalisation and resistance in six cities of Spain and the UK. Environment and Planning A 49(7): 1517-1536. p g ( ) e Certeau M (1984) The Practice of Everyday Life. Berkeley, CA: University of California Press de Vries LA and Rosenow D (2015) Opposing the opposition? Binarity and complexity in political resistance. Environment and Planning D: Society and Space 33(6): 1118-1134. g y p ( ) Deleuze G (2006) Foucault. Minneapolis: University of Minnesota Press. g y p ( ) Deleuze G (2006) Foucault. Minneapolis: University of Minnesota Press. Díaz E (2012) A Filosofia de Michel Foucault. São Paulo: Uniesp. Farias I (2011) The politics of urban assemblages. City 15: 365–374. References Papadopoulos D, Stephenson N and Tsianos V (2008) Escape routes: Control and subversion in the 21st century. London: Pluto Press. Papadopoulos D and Tsianos V (2013) After citizenship: Autonomy of migration, organisational ontology and mobile commons. Citizenship Studies 17(2): 178-196. Parker M (2011) Organizing the Circus: The Engineering of Miracles. Organization Studies 32(4): 555- 569. Pearsall H (2013) Superfund Me: A Study of Resistance to Gentrification in New York City. Urban Studies 50(11): 2293-2310. D (2011) Errant paths: the poetics and politics of walking. Environment and Planning D: Society and Space 29: 672-692. Rancière J (1998) Disagreement. Politics and philosophy. Minneapolis: University of Minnesota Press. R tl d P (2009) R i t I G D J h t R P tt G t l ( d ) Th Di ti f H Rancière J (1998) Disagreement. Politics and philosophy. Minneapolis: University of Minnesota Press. Routledge P (2009) Resistance. In: Gregory D, Johnston R, Pratt G, et al. (eds) The Dictionary of Human Geography. Hoboken: John Wiley & Sons, pp.646-647. Silva CLRd and Silva ARLd (2019) Organization/Sociomaterialidade, Poder e Conexoes em Redes de Acao no Organizar do Artesanato. RAC - Revista de Administracao Contemporanea 23: 454- 475. Smith W, Higgins M, Kokkinidis G, et al. (2015) Becoming invisible: The ethics and politics of imperceptibility. Culture and Organization. DOI: 10.1080/14759551.2015.1110584. 1-20. Spinney J (2010) Performing Resistance? Re-Reading Practices of Urban Cycling on London's South Bank. Environment and Planning A: Economy and Space 42(12): 2914-2937. Stephenson N and Papadopoulos D (2006) Analysing everyday experience. Social research and political change. London: Palgrave Macmillan. Swanlund D and Schuurman N (2019) Resisting geosurveillance: A survey of tactics and strategies for spatial privacy. Progress in Human Geography 43(4): 596-610. p p y g g p y ( ) Swyngedouw E and Kaika M (2003) The making of ‘glocal’ urban modernities. City 7: 5–21. Swyngedouw E and Kaika M (2003) The making of ‘glocal’ urban modernities. City 7: 5–21. Tolia-Kelly DP (2006) Affect - an ethnocentric encounter? Exploring the 'universalist' imperative of Emotional/affectual geographies. Area 38(2): 213-217. Tolia-Kelly DP (2006) Affect - an ethnocentric encounter? Exploring the 'universalist' imperative of Emotional/affectual geographies. Area 38(2): 213-217. g g p ( ) Tsianos V, Papadopoulos D and Stephenson N (2012) This is class war from above and they are winning it: what is to de done? Rethinking Marxism 24(3): 448-457. References Fleming P and Spicer A (2008) Beyond Power and Resistance: New Approaches to Organi Politics. Management Communication Quarterly 21(3): 301-309. Foucault M (1978) The history of sexuality. Vol.1: An introduction. London: Penguin. Foucault M (1978) The history of sexuality. Vol.1: An introduction. London: Penguin. Foucault M (1980) Power/knowledge. Selected interviews and other writings, 1972-1977. New York: Pantheon. Foucault M (1980) Power/knowledge. Selected interviews and other writings, 1972-1977. New York: Pantheon. Foucault M (1997) The essential works of Michel Foucault, 1954-1984, vol. 1. Ethics, Subjectivity and Truth. New York: The New Press. Grommé F (2016) Provocation: Technology, resistance and surveillance in public space. Environment and Planning D: Society and Space 34(6): 1007-1024. Holbraad M, Pedersen MA and Viveiros de Castro E (2014) "The politics of ontology: anthropological positions" Theorizing the Contemporary. Cultural Anthropology Online. nd Tang S-Y (2010) External Legitimacy, Goal Congruence and Collective Resistance. Urban Studies 48(4): 811-825. Junquilho GS, Almeida RAd and Silva ARLd (2012) As "artes do fazer" gestão na escola pública: uma proposta de estudo. Cadernos EBAPE.BR 10: 329-356. p p King AD (2010) O que está acontecendo com a pesquisa sobre cidades mundiais? Estudos Históricos (Rio de Janeiro) 23: 247-260. Kokkinidis G (2014) Spaces of possibilities: workers' self-management in Greece. Organization 22(6): 847-871. Lauermann J and Vogelpohl A (2019) Fast Activism: Resisting Mobile Policies. Antipode. DOI: 10.1111/anti.12538. Lees L (2012) The geography of gentrification: Thinking through comparative urbanism. Progress in Human Geography 36(2): 155–171. g p y ( ) Legg S (2018) Subjects of truth: Resisting governmentality in Foucault’s 1980s. Environm Planning D: Society and Space 37(1): 27-45. Limena MMC (2001) Cidades Complexas no Século XXI: ciência, técnica e arte. São Paulo em Perspectiva 15: 37-44. Machado FCL, Fernandes TA and Silva ARLd (2017) Michel de Certeau e estudios organizacionais: uma leitura do cenário brasileiro. Caderno de Administração 25(2): 24-43. McFarlane C (2011) Assemblage and critical urban theory. City 15: 204–224. Özdemir E and Tasan-Kok T (2017) Planners’ role in accommodating citizen disagreement: The case of Dutch urban planning. Urban Studies. DOI: 10.1177/0042098017726738. p g Papadopoulos D (2011) Alter-ontologies: Towards a constituent politics in technoscience. Social Studies of Science 41(2): 177-201. Papadopoulos D (2014) Politics of Matter: Justice and Organisation in Technoscience. Social Epistemology: A Journal of Knowledge, Culture and Policy 28(1): 70-85. Papadopoulos D (2018) Experimental Practice. Technoscience, Alterontologies, and More-Than-Social Movements. Durham, NC: Duke University Press. Woolgar S and Lezaun J (2013) A turn to ontology in science and technology studies? [Special issue]. Social Studies of Science 43(3): 321–462. References Uitermark J, Duyvendak JW and Kleinhans R (2007) Gentrification as a governmental strategy: social control and social cohesion in Hoogvliet, Rotterdam. Environment and Planning A 39: 125- 141. Whitson R (2007) Hidden struggles: spaces of power and resistance in informal work in Argentina. Environment and Planning A 39: 2916-2934. g Wideman TJ and Masuda JR (2018) Toponymic assemblages, resistance, and the politics of planning in Vancouver, Canada. Environment and Planning C: Politics and Space 36(3): 383-402. Williams A, Cloke P and Thomas S (2012) Co-Constituting Neoliberalism: Faith-Based Organisations, Co-Option, and Resistance in the UK. Environment and Planning A: Economy and Space 44(6): 1479-1501. Williams M (2016) Searching for actually existing justice in the city. Urban Studies 54(10): 2217-2231. Wittgenstein L (1958) Philosophical investigations. New York: Prentice Hall.
https://openalex.org/W2758685180	https://www.matec-conferences.org/10.1051/matecconf/201712403001/pdf	English	null	Design of a Multi-layer Lane-Level Map for Vehicle Route Planning	MATEC web of conferences	2,017	cc-by	4,447	1 Introduction lanes). Thus, the map model is supposed to be an abstract representation of the map database [4]. In terms of map model, there are a number of map standards containing specific explanations on the map model which we can refer to. In traditional road level map, GDF, KiWi, Navteq are mainstream digital map standards which have been employed widespread for decades [5], however, these road-level model are not directly applicable to lane level. RNDF is the road network definition file designed for DARPA urban challenge, in which the basic structure segment-lane-waypoint are included to provide basic information to driverless vehicles [6]. Nowadays, OpenDRIVE and NDS are well-known map standard providers, but their complicated map model and inaccessibility to public (NDS) cause difficulties for practical uses. Besides the standards above, Qing Zhu et al. [7] establish a 3-D road network which is composed of three parts: roadway centerline, carriageway, and lane. Tao Zhang et al. [4] design a lane-level map model including details on intersections and lanes. Jiang Liu et al. [8] generate enhanced intersection model and use circular to illustrate the virtual lane inside the intersection. Thus, few studies are able to provide an applicable map model with completeness and simplicity for route planning. Digital maps play an important role in current vehicle applications, e.g. vehicle localization and route planning. With more attention on intelligent transportation system, advanced driving assistance system and even driverless vehicle, the improvements in accuracy and richness are required in digital map technology [1]. When it is available to determine the vehicles’ position precisely in lane level, there are a lot of probable benefits we may obtain, i.e. the transportation officials and researchers may determine distinctions in traffic conditions for different lanes on a freeway via probe vehicles [2]. y p Most of existing digital maps are based on road level data, which may illustrate basic information and provide useful applications for users but ignore some precise details that are essential for vehicles’ high precision localization and path guidance. The improvements of accurate localization sensors such as RTK GPS and other on-board sensors such as IMU and LiDAR made the enhanced maps become possible. On the purpose of increasing utility of digital maps in advanced vehicle applications, creating the lane-level map is becoming a prevailing interest among researchers worldwide [3]. Design of a Multi-layer Lane-Level Map for Vehicle Route Planning Abstract. With the development of intelligent transportation system, there occurs further demand for high precision localization and route planning, and simultaneously the traditional road-level map fails to meet with this requirement, by which this paper is motivated. In this paper, the three-layer lane-level map architecture for vehicle path guidance is established, and the mathematical models of road-level layer, intermediate layer and lane-level layer are designed considering efficiency and precision. The geometric model of the lane-level layer of the map is characterized by Cubic Hermite Spline for continuity. A method of generating the lane geometry with fixed and variable control points is proposed, which can effectively ensure the accuracy with limited number of control points. In experimental part, a multi-layer map of an intersection is built to validate the map model, and an example of a local map was generated with the lane-level geometry. © The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0 (http://creativecommons.org/licenses/by/4.0/). MATEC Web of Conferences 124 , 03001 ( 2017 ) MATEC Web of Conferences 124 , 03001 ( 2017 ) MATEC Web of Conferences 124 , 03001 ( 2017 ) ICTTE 2017 DOI: 10.1051/matecconf/201712403001 1 Introduction Compared with the traditional road level map, the lane level map is enhanced with massive data, which are capable of improving the accuracy from about 10 meter to decimetre level or even centimetre level and illustrating more precise geometry in line with the real situations. Another significant issue about lane-level maps is the geometrical representation of the lanes. Expanding from traditional maps, some lane illustrations in enhanced maps continue using the polyline [9]. Du Jie et al. [2] employ the piecewise polyline to approximate the centreline in a lane. Betaille et al. [3] in their lane-level enhanced map apply the clothoids for the description of the road lanes in a digital map. Circular arc spline is also applied in the map geometrical model [10]. Compared to The map model is supposed to be created for routing planning before geometrical representation because vehicle navigation is established on a network consisting of vertices (e.g., intersections) and edges (e.g., roads, MATEC Web of Conferences 124 , 03001 ( 2017 ) MATEC Web of Conferences 124 , 03001 ( 2017 ) ICTTE 2017 DOI: 10.1051/matecconf/201712403001 where, a W is the whole road-level network, a C is the set of road-level intersections, and a R is the set of roads. there methods mentioned, Cubic Hermite Spline ˄CHS˅ may possess some advantages, because it can generate various lines, such as, straight lines, arcs, and even clothoids [11], therefore the whole map would share one type of line to accurately generate the geometrical shapes of different lanes. And there have been several studies employing the CHS to describe the shape of lanes, however, the method mentioned in [4] applies CHS to the whole lanes and even virtual lanes in the intersections without distinguishing the control points with different functions. The road-level intersection is defined as, , , , a c c a c P E T (2) (2) where, cP is the set of road-level nodes entering this intersection, and c E is the set of road-level nodes leaving this intersection. a T , road-level traffic matrix, indicating if there is a topological connection between two nodes, defined as, where, cP is the set of road-level nodes entering this intersection, and c E is the set of road-level nodes leaving this intersection. 1 Introduction a T , road-level traffic matrix, indicating if there is a topological connection between two nodes, defined as, ,1 , ,1 ,1,1 , ,1 , ,1, , . c c i c a a i a c j a j a i j P P E E ª º « » « » « » ¬ ¼ t t T t t (3) This paper is organized as below. Firstly, in section 2, the multi-layer lane-level map model is proposed, which includes three layer and other significant details. Then a method of geometrical representation is established by applying the CHS. Finally, the experimental validations are performed in section 4. (3) where, the element , . a i j t represents whether a vehicle could drive from the entering node ,c i P to leaving node ,c j E , and owns mathematical expression , . , a i j a a f m t , as shown in Figure 2, in which af denotes whether (=1) or not (=0) a vehicle may drive from ,c i P to ,c j E , and a m is the way of passing, e.g. turning left, turning right, going straight, U-turn. 2.2 Intermediate layer Figure 1. Multi-layer map model Figure 1. Multi-layer map model In this layer, a logic connection between upper and lower layer is designed. On the purpose of providing higher precision data and conserving the advantages of traditional maps, the intermediate layer serves as a library of corresponding relationship from road level to lane level. 2 Multi-layer lane-level map model In terms of convenience of geographical illustration and efficiency of further applications like route planning, we propose a multi-layer lane-level map model, which contains three layers: road level layer, intermediate layer, and lane level layer. The road-level road is unidirectional, which means in this model, there generally are two roads between two adjacent intersections. A road is expressed as, Road-level layer in this map model reserves most of traditional mathematical expressions for current existing map model, containing roads and intersections, which is intended for making use of ripe routing algorithms based on road level maps. The intermediate layer acts as a bridge between the upper layer and the lower one, in which relationship between some sets is stored for routing planning and other applications. And the third layer is designed to express the lane-level details, not only the lane sets on a road and more detailed intersections, but also geometrical elements such as high- precision points on the centre line of a lane, lane lines and so on. An example is Figure 1 p , , a r r a r P E Q (4) (4) where, rP is the set of road-level nodes entering this road, and r E is the set of road-level nodes leaving this road. a Q includes the road class rk and road length rl . Figure 2. Traffic matrix and intermediate layer Figure 1. Multi-layer map model Figure 2. Traffic matrix and intermediate layer 2.3 Micro layer in lane level Lane-level layer should not only possess more data with higher accuracy which could supplement the deficiency of road-level layer, but also provide information of complete road network for navigation. Besides nodes, control points also play an important role in layered map, especially lane-level layer. Control points in intersections and lanes both contain precise positions, tangent vectors, indices, and other attributes, and are defined as, To achieve this goal, the lane-level layer is defined as, , m m m W C R (6) To achieve this goal, the lane-level layer is defined as, , m m m W C R (6) , m m m W C R where, m C is the set of lane-level intersections, and m R is the set of roads. where, m C is the set of lane-level intersections, and m R is the set of roads. , ^ ` , , , s s s s n s u v Q (12) (12) where, sn is the serial number, and s u is the position which is two or three-dimension vector, and sv is the tangent vector on the node, and s Q involves attributes, such as lateral sequence number of the lane, line types of left lane line and right lane line for control points on a lane. The road-level intersection is defined as, , , m m m m c P E T , , m m m m c P E T (7) (7) where, m P is the set of lane-level nodes entering this intersection, and m E is the set of lane-level nodes leaving this intersection. m T , lane-level traffic matrix, indicating whether there is a topological connection between two lane-level nodes, which is defined as, where, m P is the set of lane-level nodes entering this intersection, and m E is the set of lane-level nodes leaving this intersection. m T , lane-level traffic matrix, indicating whether there is a topological connection between two lane-level nodes, which is defined as, Particularly, in a common scenario, the nodes at the start/end of a lane share the same positions and tangent direction with the first and the last control points. ,1 , ,1,1 , ,1 ,1 , ,1, , . 2.1 Road-level layer According to intuition that a road network is supposed to include roads and intersections, thus, a road-level layer is expressed as follow, In an intersection c , cP is the set of its road-level incoming nodes. And without losing generality, let us assume this intersection has four road-level incoming , a a a W C R (1) , a a a W C R (1) 2 2 MATEC Web of Conferences 124 , 03001 ( 2017 ) ICTTE 2017 MATEC Web of Conferences 124 , 03001 ( 2017 ) DOI: 10.1051/matecconf/201712403001 2.4 Node and control point nodes from different incoming directions, which could expressed as ,1 ,4 , , c c c P P P . There are nodes in road-level and lane-level (Notice that nodes in intermediate layer come from the other two layers.), which only exist at the connection knots between intersections and roads(lanes) and are defined as, And generally there are at least one lanes for each road at an intersection. As Figure 2 shows, the road-level node ,1 cP represents two lane-level nodes ,1,1 ,1,2 , m m P P . ( ) ^ ` , N N N u v (11) Thus, intermediate layer is expressed as follow, (11) , , ,1 , , , , i c i m i m i k P P P (5) (5) where, N represents the entering or leaving nodes in this map model, including cP , c E in the road-level intersections, rP , r E in the road-level roads, and m P , m E in the lane-level intersections. N u is the location of the node, and N v is the traffic direction of the node. 2.3 Micro layer in lane level m m m m m m m m i m m i m m m j m j m i j P P E E ª º « » « » « » ¬ ¼ t t T t t (8) (8) 3 Geometrical representation The detailed geometry in road level is not necessary because the detailed information and precise geometry is presented in lane level and it is easy to obtain the lane- level geometrical information via intermediate layer from road level and on the other hand, the road level nodes are easy to locate, simple lines such as straight line could illustrate the connection and topology in road level completely. Thus, according to our multi-layer model, the lane-level geometry is discussed as below. where, the element , . m m m i j t implies whether and how a vehicle could drive from the entering node , m m i P to leaving node , m m j E ,and its mathematical expression is , . , , m m m i j m m c f m t S , in which mf denotes whether (=1) or not (=0) a vehicle may drive from , m m i P to , m m j E , and m m is the way of passing which shares the same definition with that in road-level. The c S represents the set of control points on the centre line of the virtual lane between two lane-level nodes in an intersection, as Figure 2. 3.2.1 Fixed Control Points The fixed control points f S are determined by the lane attributes. Between two adjacent fixed control points the lane attributes, except the shape of the lane, keep the same, and the lane attributes include speed limit, lateral sequence number of the lane, line types of lane lines, and etc. Some details are displayed as below. Figure 3. Control Points on lanes (Red dashed line indicates the position where the type of lane lines changes.) Figure 3. Control Points on lanes (Red dashed line indicates the position where the type of lane lines changes.) Thus, we get the complete series of control points, including fixed control points and other variable shape control points, which are combined to illustrate the lane- level geometry. 3.2 Control Points Choosing The control points serve as the shape points in the lane level geometry, and simultaneously some of them are supposed to be feature points, of which on two sides there are some different attributes, at particular positions, e.g., the control points at the start/end of a lane. Though this kind of control points own the same form with the common control points, because these do not change with other variable control points, we call them the fixed control points. Figure 4. Shape control points choosing Figure 4. Shape control points choosing The stepwise approach is started at ,1 ,1 = t d s s , then from each chosen shape control point, successively choose the next shape control point, the details are listed in Table 1. Table 1 Pseudocode: stepwise method for shape control points 1 set ,1 ,1 , , = , , , , , t d t t t t j t N S s S s s s ; 2 while 1 3 if t S return d S ; 4 set , = d c d end S S , 2 k ; 5 while 1 6 . = d n t k S S ; 7 , , , , , , d c d n d c d n f x o s s s s ; 8 > @ , , , 0,1 1, , 2 max min , d c d n t t t x j k e j f x j j s s u u S ; 9 if h e T t ,break; 10 else = 1 k k ; 11 if = t k S ,break; 12 end if; 13 end if; 14 end while; 15 , = 1 d n t k s S ; 16 , = d d d n S S s , , , = , , t t d c d n S S s s ; 17 end while; 3.2.2 Variable Control Points The determination of shape control points occurs in two adjacent fixed control points in a lane. Assume that a series of sampled points representing the centreline of a lane are known, expressed as ,1 , , = , , , , t t t t j t N S s s s , and the first and last points are fixed points, i.e. ,1 , , t t t N f s s S , a stepwise approach on determining the series of shape control points ,1 , , = , , , , d d d d i d N S s s s is proposed. MATEC Web of Conferences 124 , 03001 ( 2017 ) ICTTE 2017 MATEC Web of Conferences 124 , 03001 ( 2017 ) ICTTE 2017 DOI: 10.1051/matecconf/201712403001 the position extracted from ,t j s which are located among two adjacent shape control points, as shown in Figure 4. the position extracted from ,t j s which are located among two adjacent shape control points, as shown in Figure 4. It is easy to get the i i Q n u , 1 1 i i Q n u , i i Q n c v , 1 1 i i Q n c v , thus at each control point the first order continuity is realized, which may offer convenience for vehicle path planning due to global tangential continuity. Figure 4. Shape control points choosing 3.1 Lane-level geometry In order to achieve the global conformity and continuity, as we mentioned before, CHS curve is employed in this part. CHS is a spline in which each segments between two control points is a three order polyline represented by Hermite form. The lane-level road m r contains all lanes on the road, thus we define, , , , m in out r c c r L Q (9) (9) where, L is the set of lanes, and in c and out c are the indices of entrance and exit intersections according to traffic directions on the road respectively, and r Q is the attributes of the road. Assuming that two given control points are is , +1 is , where , , i i i i n s u v , and definitions of symbols are the same with Equation (12), then the CHS generated by these two points is presented as, Each lane in the road, i.e. element in the set of lanes, is defined as, 3 2 3 2 3 2 1 3 2 1 2 3 1 2 2 3 T i i i i x x x x x Q n f x x x x x ª º ª º « » « » « » « » « » « » « » « » « » ¬ ¼ ¬ ¼ u v u v (13) 3 2 3 2 3 2 1 3 2 1 2 3 1 2 2 3 T i i i i x x x x x Q n f x x x x x ª º ª º « » « » « » « » « » « » « » « » « » ¬ ¼ ¬ ¼ u v u v (13) where, 1 i i i x n n n n . , l l l S Q (10) (10) (13) where, lS is the set of control points on the centre line of the lane, and l Q is the set of attributes, including lane width wl , lane length el and speed limit V . where, 1 i i i x n n n n . 3 MATEC Web of Conferences 124 , 03001 ( 2017 ) viewed in MapInfo and employed easily by other applications. viewed in MapInfo and employed easily by other applications. Figure 5. An example of map based on multi-layer map model Acknowledgement Figure 5. An example of map based on multi-layer map model This work is supported by International Science & Technology Cooperation Program of China under contract No.2016YFE0102200. gure 5. An example of map based on multi-layer map model In the other experimental test, our test vehicle is equipped with RTK GPS and IMU system. The RTK GPS is BD982 from Trimble, and IMU is produced in Oxts, the model is RT2502 which can measure the position and orientation accurately. In order to measure the centreline of lanes on a road, the vehicle in our test is required to drive along the centreline as close as possible. Then the vehicle state space is established for applying Kalman Filter to obtain the precise results of locations regarded as the true value of points on centrelines. The nodes and fixed control points are determined by static vehicle with equipment running. When all the fixed lane- level control points are measured, the stepwise method of variable shape control points is executed to pick up the other control points for lane-level geometrical representation. References 1. K. Wevers, S. Dreher. “Digital Maps for Lane Level Positioning”. 15th World Congress on Intelligent Transport Systems and ITS America's 2008 Annual Meeting,(2008). 2. J. Du, M.J. Barth. “Next-generation automated vehicle location systems: positioning at the lane level”. IEEE Trans. Intell. Transp. Syst. 9(1): 48–57, (2008). ( ) 3. D. Betaille, R. Toledo-Moreo. “Creating enhanced maps for lane-level vehicle navigation”. IEEE Trans. Intell. Transp. Syst. 11(4):786-798, (2011). 4. T. Zhang, S. Arrigoni, M. Garozzo, et al. “A lane- level road network model with global continuity”. Transp. Res. Part C, 71:32-50, (2016). Figure 6 A lane level local map with geometry 5. T. Zhang, D. Yang, T. Li, et al. “An improved virtual intersection model for vehicle navigation at intersections,” Transp. Res. Part C, 19(3):413–423, (2011). 6. J. Bohren et al., “Little Ben : The Ben Franklin Racing Team’s Entry in the 2007 DARPA Urban Challenge,” J. F. Robot., 25(9):598–614, (2008). 7. Q. L, Y. Zhu, “Hierarchical laneϋ oriented 3D roadϋ network model,” Int. J. Geogr. Inf. Sci., 22(5):479–505, (2008). Figure 6 A lane level local map with geometry 8. J. Liu, B. Cai, Y. Wang, et al. “Generating Enhanced Intersection Maps for Lane Level Vehicle Positioning based Applications,” 13th COTA Int. Conf. Transp. Prof., 96(Cictp):2395–2403, (2013). As Figure 6 shows, the geometry representation can illustrate the shape of lanes and even intersections, increasing utility compared to the traditional geometry. 9. EDMap Consortium, “Enhanced Digital Mapping Project Final Report,” (2004). 4 Experimental Validation Two experiments are conducted to validate our theoretical model and geometrical representation respectively. In order to demonstrate the validity of multi-layer map model, we combine the aerial photography and ground calibration to establish a local map including a complete intersection which is displayed as Figure 5. The left graph is the road-level layer with the background of an aerial photo, in which the points symbolized by star are road-level nodes. The right one is the lane-level map, and the details on lanes corresponding to the road they belonging to are shown. Besides, the traffic matrix and other information are stored in TAB format which can be In this stepwise method, the distances between two adjacent shape control points are variable, which is determined by, > @ , , 1 , 0,1 1, , 2 max min d i d i h t x j N T e j f x t s s u (13) (13) Where, hT is the threshold designed, which can specify the accuracy of the geometrical representation, and e is the measurement of the gap between CHS curve generated and corresponding sampled point, and t j u is 4 4 DOI: 10.1051/matecconf/201712403001 MATEC Web of Conferences 124 , 03001 ( 2017 ) MATEC Web of Conferences 124 , 03001 ( 2017 ) ICTTE 2017 planning algorithm based on this layered map model [12], and due to lane-level layer in this model contains all detailed information for routing, a route planning method directly employing lane-level map is also designed [13]. Then, there is a large-scale map we are building in a South-eastern province, China, which is completely based on this our layered model. 4 Conclusions and future researches 10. A. Schindler, G. Maier, F. Janda. “Generation of high precision digital maps using circular arc splines,” IEEE Intell. Veh. Symp, 246–251. (2012). This paper proposes a multi-layer map model for route planning, in which we integrate the traditional road-level map with enhanced lane-level map, we also design an intermediate layer to connect the other two layers. According to advantages of CHS curve in presenting lane-level geometry, a stepwise approach to generating lanes is discussed and a geometrical system including fixed control points and shape control points is performed. A test to validate the feasibility of the map model and geometrical representation is conducted. 11. A. Chen, A. Ramanandan, J.A. Farrell. “High- precision lane-level road map building for vehicle navigation.” IEEE Position Location and Navigation Symposium.1035-1042, (2010). 12. C. Liu, K, Jiang, Z. Xiao, Z. Cao, D. Yang, “Lane- Level Route Planning Based on a Multi-Layer Map Model. ” to be published. 13. C. Liu, T. Zhang, S. Arrigoni, K. Jiang, F. Chelia, D. Yang, “Addressing the route planning problem on the lane-level road network,” to be published. There are some ongoing researches based on this multi-layer map model. We have finished the route 5
https://openalex.org/W4243835907	https://www.researchsquare.com/article/rs-4634/latest.pdf	English	null	Effective methods for reactivating inactive blood donors: a stratified randomized controlled study	Research Square (Research Square)	2,019	cc-by	8,189	Effective methods for reactivating inactive blood donors: a stratified randomized controlled study Jian OU-YANG PLA North Military Command Region General Hosp Chun-Hua BEI Guangzhou Blood Center Hua-Qin LIANG Guangzhou Blood Center Bo HE Guangzhou Blood Center Jin-Yan CHEN Guangzhou Blood Center Yong-Shui FU (  fuyongshui@sina.com ) Guangzhou Blood Center Research article License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Version of Record: A version of this preprint was published at BMC Public Health on April 10th, 2020. See the published version at https://doi.org/10.1186/s12889-020-08594-9. Page 1/21 Abstract Background: Recruiting of sufficient numbers of donors of blood products is vital worldwide. In this study we assessed the efficacy and cost-effectiveness of telephone calls and SMS reminders for re- recruitment of inactive blood donors. Methods: This single-centre, non-blinded, parallel randomized controlled trial in Guangzhou, China included 11,880 inactive blood donors whose last donation was between January 1 and June 30, 2014. The donors were randomly assigned to one of two intervention groups (telephone call or short message service [SMS] communications) or to a control group without intervention. SMS messages with altruistic appeal were adopted in the SMS group; in addition to altruistic appeal, reasons for deferral of blood donation were also asked in the telephone group. All participants were followed up for 1 year. The primary outcome was re-donation rate, and rates in different groups were compared by intention-to-treat (ITT) analysis and estimation of the average treatment effect on the treated (ATT). Secondary outcomes were the self-reported deterrents. Other outcomes included the re- donation interval, and the incremental cost-effectiveness ratio (ICER) of telephone calls and SMS reminders on re-recruitment. Results: ITT analysis revealed no significant differences in the re-donation rate among the three groups. ATT estimations indicated that among compliers, telephone calls significantly increased re-donation compared to both SMS reminders and no intervention. Donor return behaviour was positively associated with receiving reminders successfully, being male, older age, and previous donation history. The SMS reminder prompted donors to return sooner than no reminder within 6 months, and according to ICER calculations, SMS reminders were more cost-effective than telephone calls. Donors reported time constraints as the most main causes of self-deferral in the telephone group, and altruistic appeal had a positive effect on these donors. Conclusions: Interventions to reactivate inactive blood donors can be effective, with telephone calls prompting more donors to return but at a greater cost than SMS messages. SMS reminder with altruistic appeal can urge donors to re-donate sooner within 6 months than no reminder. Background Blood products play a vital role in saving lives in a wide variety of medical conditions. Along with the rapid development of the economy and improvement of modern medicine in China, the demand for blood products has continued to grow. Many cities in China have faced a “blood shortage” dilemma, in which the blood supply cannot meet the clinical demand [1]. Therefore, effective strategies for recruiting sufficient numbers of blood donors are critically needed. It has been widely established that repeat donors have a lower transfusion-transmissible infection risk [2], and this reduced risk is maintained in donors who have not donated blood for 5 years. [3]. In addition, individuals with a previous donation experience are more likely to restart donations in the future than are first-time donors [2]. The approach of reactivating inactive donors is nonetheless challenging. The percentage of donations from repeat donors in China was reported to be 34-40% [4-6], which is lower than that in the United States (68%) [7] and that in England (55%)[8]. It is essential to determine effective methods for reactivating lapsed donors (defined as those who have made at least one donation within the last 24 months, but not within the previous 12 Page 2/21 Page 2/21 months) and inactive donors (those who have made at least one donation but have not donated within the previous 24 months) [9] in order to maintain an adequate blood supply. The most common and accessible reminders for promoting the return of blood donors include telephone calls, cell phone short message service (SMS) messages, mailings, and e-mails, which may help to support the intrinsic motivations of donors, thereby increasing their commitment to donation [10]. Aside from its wide availability, low cost, and convenience, SMS messaging has been proven to be an effective intervention for a variety of health behaviours [11, 12], including blood donation [13, 14]. Upon receiving SMS message, donors may recall positive feelings from previous blood donations, thus increasing their desire to repeat the experience [11]. However, instead of sending only generic information via SMS messages, telephone calls have the advantage of personalizing communication with donors. Godin et al. found that a first phone call reminder could encourage first-time donors to return [15]. Sinclair et al. reported that the use of an adapted motivational interview via telephone calling could increase the chance of future donation [10]. Background Donors also reviewed their donation experience in consideration of their wider motivations for giving and extended this line of thinking to problem-solving solutions to perceived barriers [10]. Eliminating the deterrents for inactive donors is a critical retention strategy. Research has shown that participants are more likely to donate again after they have been invited to report their reasons for not donating [16]. Reports from different countries have indicated that medical reasons, time constraints, fear (of needles/bleeding), and negative physical reactions are the most frequently self-reported deterrents among lapsed and inactive donors. [17-21]. Blood donors report multiple motivations for blood donation [22-26], and campaigns to encourage them to donate should focus on multiple perspectives for different groups. However, in the era of information overload, SMS messaging has the disadvantage of being easily ignored by recipients; hence, the recruitment message needs to be short and simple to understand. Altruistic appeal is a common and acceptable way of recruiting blood donors, which has been widely adopted in various campaigns. Therefore, in the present study, SMS messages with an altruistic appeal that emphasized “saving a life” were sent in an attempt to re-recruit inactive blood donors. Meanwhile, telephone calls were made to re- recruit inactive donors by calling them with not only an altruistic appeal but also questions regarding the reasons why they stopped and providing corresponding solutions. The objective of this stratified, randomized controlled trials was to assess the efficacy and cost- effectiveness of telephone calls and SMS messages for blood donor re-recruitment. A secondary objective was to explore the self-reported reasons for deferral among donors who received the telephone call. Other aims included evaluating donor return according to demographic characteristics, the time to return among the different interventions, and the cost of telephone calls versus SMS messages to former blood donors. Study design, setting and participants The Guangzhou Blood Center is one of the largest blood centres in China along with the Beijing and Shanghai Centers. A total of 263,681 donors donated blood during 2014, of which 179,964 (68.3%) then became inactive and 83,717 (31.7%) donated again before 2016 [27]. This single-centre, non-blinded parallel randomized controlled trial involved two intervention groups (telephone or SMS reminders) and a no-intervention control group. Figure 1 shows a flow chart of the study design. During the experiment period, donors who had donated blood after July 1, 2014 were continuing to receive phone call and/or SMS reminders occasionally from Guangzhou Blood Center. Therefore, in order to avoid contaminations, both whole blood and apheresis platelet donors whose last donations were between January 1 and June 30, 2014 were eligible for the screening. The age range for blood donors in mainland China is 18-55 years. Those aged above 50 years old were excluded from the screening based on the previous finding that most older individuals are unlikely to donate again due to physical reasons [28]. All data were provided by the Guangzhou Blood Center through the Blood Donation and Supply System [27]. The calculated sample size needed for each group was 2,252 [29], and in order to better detect significant differences and balance the sample size in each tier, the actual sample size was 3,960 in each group (11,880 total). All participants were stratified into 18 tiers by age (20-30, 31-40, and 41-50 years), gender (male and female), and frequency of prior donation, which refers to number of times a donor donated before becoming inactive (one time, two or three times, and four times or more). Based on a computer- generated list of random numbers, the first 220 eligible participants in each tier were assigned to the telephone group, the 221st to 440th were assigned to the SMS group, and the 441st to 660th were assigned to the control group. Methods Page 3/21 Page 3/21 Interventions and endpoint The experimental period lasted from October 20 to November 10, 2016. The details of the recruitment method were described in the pilot study [29]. The start time was set at the day that an intervention was made. Thank you again for your support!” The message was sent via the SMS platform of the Guangzhou Blood Center. Message receipts, which stated if a message was received successfully or not were retrieved from the SMS platform within 48 hours. All participants, whether they received the message or not, remained on the list for further follow- up as described below. Follow-up and outcome measures The donation activity of each participant was followed for 365 days from the recruitment day. All participants could be followed via the Blood Donation and Supply System in which their blood donation records in Guangzhou could be checked. The primary outcome was to identify the occurrence of the first next blood donation attempt among all participants within the 1-year follow-up and evaluate the return rates according to donor characteristics (gender, age and past donation frequency). A participant was classified as a re-activated donor if he/she made at least one subsequent donation by the end of the 1- year follow-up period; otherwise, the donor was classified as “no return”. The secondary outcomes were the main self-reported reasons for deferral given by donors during the telephone calls. Other outcomes included the re-donation interval for each group after recruitment, the efficacy of each intervention, and the cost-effectiveness of telephone calls versus SMS reminders on re-recruitment. “Dear donors, Thank you for your donation through which your love brought hope to those helpless patients and your donated blood reignited the fire in their lives. If you can, please consider donating blood again to save a life. Thank you again for your support!” Telephone call reminder Telephone interviews were conducted over a 1-month period by two interviewers (O-Y and BEI, staff members at the Guangzhou Blood Center with the responsibility of blood donor recruitment). The interviews were conducted using pre-designed questionnaires and lasted 2.4-21.3 min (mean±SD, 4.8±1.2 min). Prior to this study, the interviewers summarized the barriers to donation frequently mentioned by participants in the pilot study [29], discussed the challenges that arose during the interviews, reviewed the optimal response techniques, and practiced via role playing to ensure adherence to the script. Page 4/21 Donors who could not be reached by telephone because the phone number was wrong were marked as non-responders. Donors with a disconnected phone line or who did not answer were called two more times on subsequent days before being classified as “no answer”. Donors who answered the phone but refused the interview request were marked as “refusal”. All participants, including those marked as non- responders, no answer, and refusal, were further followed up as described below. After contact was successfully made, a brief and scripted interview was delivered with the donors’ permission (Additional file 1). SMS reminder In the SMS group, participants received a text message making an altruistic appeal. As follow: Statistical analysis The database and foundation for analyses were established by recoding data in an Excel software (2013, Microsoft Corporation, Redmond, WA, USA) file and importing into the Statistical Package for Social Sciences software (SPSS Statistics version 23 for Windows, SPSS Inc., Armonk, NY, USA) and The R Page 5/21 Page 5/21 Page 5/21 Project for Statistical Computing (R version 3.6.1). For intention-to-treat (ITT) analysis, the re-donation rate was calculated by dividing the number of participants who donated again during the follow-up period by the corresponding number of initially randomized donors. Because of the large disparity in the intervention received rates between the two interventional groups, the ITT result might have masked a true effect on the re-donation rate among those who received reminders as intended. Therefore, Project for Statistical Computing (R version 3.6.1). For intention-to-treat (ITT) analysis, the re-donation rate was calculated by dividing the number of participants who donated again during the follow-up period by the corresponding number of initially randomized donors. Because of the large disparity in the intervention received rates between the two interventional groups, the ITT result might have masked a true effect on the re-donation rate among those who received reminders as intended. Therefore, estimation of the effects of the interventions on inactive blood donors while accounting for compliance with assigned intervention was also conducted. Previous studies defined four compliance types on the basis of individuals’ treatment assignment status and potential treatment receipt status [30-32]. In this study, strict adherence to the intervention assignment meant that those in the control group did not receive any telephone call or SMS message; meanwhile, participants in the telephone group did not receive an SMS message and vice versa. Thus, in this case, there were no directly observed always-takers (defined as those who will always implement the treatment, regardless of the group to which they are assigned), nor defiers (defined as those who will not implement if assigned to the treatment group but will implement if assigned to the control group). The participants did include compliers (defined as those who will implement the treatment when assigned to the treatment group but will not implement if assigned to the control group) and those could still be never-takers (defined as those who will never implement, regardless of the treatment assignment). Statistical analysis In other words, this was a one-sided non-compliance situation, with only compliers (who received the telephone or SMS reminders successfully in the intervention groups, and who were in the control group) and never-takers (who failed to receive the telephone or SMS reminders in the intervention groups) [33]. Therefore, the average treatment effect on the treated (ATT) were also estimated [33]. Chi-square test was used to identify statistical differences of the re-donation rate among groups and conduct paired comparisons between contact methods, and Bonferroni correction was applied. R Package “ATE” was used to estimate the ATT the ATT among compliers under the intervention and control conditions (random assignment was used as an instrumental variable that telephone or SMS group coded as 1, control group coded as 0; complier in the telephone or SMS group was coded as 1, never-taker and those in the control group were coded as 0). The Kruskal-Wallis test was used to determine whether the re-donation intervals within 30, 90, 180, 270 and 365 days were affected by different reminders, and Mann-Whitney U test was applied for comparisons of two groups. Binary logistic regression analyses were adopted to identify associations of donor characteristics with donor return behaviour to determine the best predictors of future donation; odds ratio (ORs) and 95% confidence intervals (CIs) were calculated. The incremental cost-effectiveness ratio (ICER) was applied to compare the cost-effectiveness of telephone calls and SMS reminders. All hypothesized differences were considered statistically significant if the P-values from two-tailed tests were < 0.05. Study Participants Study Participants Twenty-nine donors in the telephone group, 37 in the SMS group, and 26 in the control group found to have donated blood before the day of recruitment. They were all excluded and replaced by an equal number of matched participants in the same tier. In the telephone group, 40% of the participants were successfully interviewed, 20.8% could not be reached due to an incorrect number, 36.1% did not answer the phone and 3.1% refused to participate. Since the rate at which lapsed blood donors answered telephone interviews from Guangzhou Blood Center was 35-45%, the response rate in the present study was close to the real percentage [27]. In the SMS group, 79.3% participants received the message successfully, and 20.7% did not (successful intervention rates in telephone and SMS groups: 40.0% vs. 79.3%, P<0.001). Ethics considerations Page 6/21 All procedures were reviewed and approved by Institutional Review Board of the Guangzhou Blood Center. The registration ID for this study on ClinicalTrial.gov is: NCT03366441. This study is reported according to the CONSORT guidelines (Additional file 2). Effects of interventions on donor return For ITT analysis, the re-donation rates were 8.1% (n = 322) in the telephone group, 8.5% (n = 337) in the SMS group, and 7.4% (n = 291) in the control group. Chi-square test showed no significant difference in the re-donation rates among the three groups (P = 0.154). The re-donation rates within compliers in the telephone and SMS groups were 11.7% (185/1583) and 8.6% (270/3142), respectively. Table 1 shows the ATT estimation results that among those who received the intervention successfully, the telephone call was estimated to significantly increase re-donation by 2.3 percentage points compared to SMS reminder, and by 6.0 compared to no intervention. Cost-effectiveness of telephone and SMS reminders The ICERs for the telephone and SMS reminders were evaluated from the bottom-up approach. In the telephone group, all interviews were completed in 7,694 minutes, while recruiters waited on hold three times for a total of 7,131 minutes. The hourly pay for one recruiter was RMB¥60 in the Guangzhou Blood Center, and therefore, the cost was RMB¥1 for a recruiter to work for 1 minute. The telephone merchant charged RMB¥0.22 for the first 3 minutes and RMB¥0.11 for every 1 minute thereafter for one call, whereas answering a call was free of charge. Thus, it cost RMB¥678.6 totally for 1,583 calls. The average cost per participant in this group was: (See Equation 1 in the Supplementary Files) In the SMS group, it took 5 minutes for one recruiter to send all the messages. The system maintenance cost for the automatic message sending system is RMB¥2523 per year, and the SMS operator charges RMB¥0.05 for each message successfully sent. Thus, the total cost was RMB¥157.1 for sending 3,142 messages successfully. The average cost per participant in this group was: (see Equation 2 in the Supplementary Files) Table 6 presents the ICER estimation results, which indicated that the SMS reminder was more cost- effective than the telephone call. Interaction between covariates and receipt of reminders on donor return Interaction between covariates and receipt of reminders on donor return Table 2 summarizes the basic information of the reactivated donors. Logistic regression analysis showed that donors who were older, those with a larger donation frequency before recruitment, or those who accepted the interventions successfully were more likely to re-donate (Table 3). Table 4 indicates the associations of re-donation and donors who were successfully interviewed, received the message and were in the control group. Donor return behaviour was positively associated with receiving reminders, being male, being of older age, and having a previous donation history. Among those participants who were successfully contacted, older donors (1.02, CI: 1.00–1.04, P=0.019) and those with a greater past donation frequency (1.08, CI: 1.03–1.13, P=0.001) in the telephone group; as well as male donors (1.33, CI: 1.03–1.71, P=0.028) and those with a greater past donation frequency (1.11, CI: 1.07–1.16, P<0.001) in the SMS group were more likely to return. Page 7/21 Page 7/21 Impact of reminders on time to return Table 5 shows the 1-year re-donation intervals for the three groups. The Kruskal-Wallis test revealed significant differences in the time to re-donation only within a 180-day interval among the three groups (P=0.023), but not within the other intervals (data not shown). The Mann-Whitney U test indicated that participants returned to donate sooner in the SMS group (76.7±50.9) than those in the control group (90.9±51.2) within the 180-day interval (Z=2.730, P=0.006). Discussion Blood donor retention and re-enrolment are critical for the collection of a sufficient blood supply but are challenges in clinical practice. The use of a stratified, randomized trial design in the present study allowed us to test the efficacy and cost-effectiveness of telephone call and SMS message reminders for prompting inactive blood donors to donate blood again. Altruism, as a genuine part of human nature, has been found to be the most frequent motivator driving people to donate blood [23-25]. In the SMS group, inactive donors received a short text message on their cell phones that contained an altruistic appeal that emphasized how they could “save a life”, and in the telephone group, in addition to the altruistic appeal, inactive donors were also questioned about the reasons why they stopped donating and provided corresponding solutions according to their answers. ITT analysis showed that the differences in the re- donation rates among all participants in the three groups were not statistically significant. ATT estimations revealed that among those who received the interventions successfully, the telephone call was more effective than the SMS reminder or no intervention. Our results also showed that donor reactivation was positively associated with receiving reminders. Moreover, participants in the SMS group returned to donate sooner than control participants (P=0.006) within 6 months and based on the calculated ICERs, SMS reminders were more cost-effective than telephone calls. In summary, interventions to promote inactive donors’ return to give blood are appropriate. As mentioned above, an altruistic (‘save a life’) message was used in the SMS group in this study, because that help-seeking message can evoke empathy, create positive emotional feelings in the reader and connect them to the recipient of their help, which eventually increases helpful behaviour [34]. Nonetheless, it has been proven that blood donation is driven by multiple motives [22, 23, 35], and the results of the present study suggest that a message with an altruistic appeal might not be strong enough to prompt action among donors. Gemelli et al. found that sending a personalized post-donation message was effective for retaining donors [14]. Notably, the SMS messages sent via the Guangzhou Blood Center automatic message sending system are all personalized. After donation each donor receives a message including his/her name, blood type and blood test results confirming their eligibility to donate (donors who are ineligible are informed by phone call). Self-reported reasons for blood donation deferral The distribution of self-reported reasons for deferral is shown in Figure 2. Those who reported time constraints were more likely to return after a phone call reminder than those who claimed other deterrents (14.2% vs. 10.6%, P = 0.037). Medical reasons included multiple different causes (Table 7). Donors who believed they had an inadequate health status were not able to reveal more specific details. Group- sponsored donation is a special form of donation in China, which is defined as blood donation organized by universities, companies, governmental agencies and any other groups. Seventy-seven (43.0%) former donors who had donated during a group-sponsored event had not donated again because they had missed the blood donation activity organized by their affiliation, 66 (36.9%) did not donate again because their affiliations stopped organizing the blood donation activity, and 36 (20.1%) reported that they did not re-donate because there is no “quota” for them to donate. Donors who simply did not want to donate again did not provide more information even when they were further asked about the reasons. Among the “other reasons”, “adverse reaction” was reported by the highest percentage of donors (23/60, 38.3%). Page 8/21 Page 8/21 Table 8 compares the willingness of former donors to re-donate and the actual re-donation rates according to the different self-reported deterrents to re-donation. Table 8 compares the willingness of former donors to re-donate and the actual re-donation rates according to the different self-reported deterrents to re-donation. Discussion In this study, we not only used an altruistic appeal but also communicated with the donors to better understand their self-reported reasons for blood donation deferral. The altruistic appeal via the telephone call had a significantly greater positive effect on those who reported time constraints than on those who claimed other deterrents kept them from donating again, and in multiple studies, time constraints were the most frequently stated factor preventing donors from continuing to donate blood [21, 38, 39]. One study found that people believed that spending time on blood donation had no more or less value than any other moment in their day [17]. They might have a positive attitude towards a request for blood donation, but did not take corresponding action due to a lack of urgency or motivation [40]. When we mentioned the idea of “saving a life” to emphasize the urgency of the need as well as revisit their original motivation for donating, they returned to donate. Therefore, blood donation agencies should make efforts to minimize the time required for donation, to implement more extensive and flexible opening hours, and also to convince donors of the importance of donation. An altruistic appeal could not effectively reactivate those who reported medical reasons for their donation deferral. Self-perceived inadequate health status, adverse reaction to blood donation, and becoming unhealthy after blood donation represented particular barriers to blood donation and seem to have similarities in China and other countries. In the Chinese traditional culture, people believe that blood is vital to human life (the Mother of Qi) and loss of blood equates to ruining one’s constitution [41]. Once donors experience adverse events or even they simply feel tired after donation, they likely deem that they experienced substantial detrimental effects or long-term consequences from blood donation. In addition, some donors may have mentioned medical barriers as a "false" reason that is more socially acceptable than stating that they do not have time [38]. To develop a strategy to recruit these donors, more specific psychological research should be carried out. Blood donors in China typically fit into one of two types: those who spontaneously donate at blood collection sites, and those who donate through a group donation. Discussion Therefore, the message sent for the purpose of recruiting inactive blood donors becomes personalized if sent via the automatic message sending system, which might increase the re-donation rate. Moreover, male donors or those with a greater past donation frequency were more likely to return after they received a message with an altruistic appeal, and thus, SMS reminders can be used to target these donors. The SMS reminder is overall an effective and convenient strategy for reactivating inactive donors, with the additional advantage of being cost- effective. After the study period, SMS messages were also sent to those who could not be reached in the telephone group and those in the control group. Page 9/21 Although the successful contact rate for the telephone group was much lower than that for the SMS group, the donors in the telephone group were more likely to return once they received the call successfully. Moreover, the effect was greater on those with a higher past donation frequency and older donors. Previous studies proved that the number of previous returns of a donor is positively associated with future return [6]. As more donations are made, the perception of oneself as a donor becomes internalized and serves as a motivating force for repeat donation [36]. Blood donors who gave more than 4 donations a year considered blood donation as an act of altruism and promised to continue donating blood in the absence of benefits and rewards [37]. Therefore, altruistic appeal, by either telephone or SMS reminder, was effective for these donors. Although the successful contact rate for the telephone group was much lower than that for the SMS group, the donors in the telephone group were more likely to return once they received the call successfully. Moreover, the effect was greater on those with a higher past donation frequency and older donors. Previous studies proved that the number of previous returns of a donor is positively associated with future return [6]. As more donations are made, the perception of oneself as a donor becomes internalized and serves as a motivating force for repeat donation [36]. Blood donors who gave more than 4 donations a year considered blood donation as an act of altruism and promised to continue donating blood in the absence of benefits and rewards [37]. Therefore, altruistic appeal, by either telephone or SMS reminder, was effective for these donors. Conclusions In conclusion, inactive blood donors may be encouraged to re-donate after receiving telephone calls or SMS reminders. Telephone calls may be more effective than SMS messages for reactivating inactive donors, but the effectiveness of each method must be weighed with the corresponding costs. More detailed studies are needed to evaluate the effects of the contents of the SMS message for reactivating inactive donors. Moreover, future studies should also focus on strategies for re-recruiting those who stopped donating because of a self-perception of inadequate health status. List Of Abbreviations ITT: intention-to-treat; ATT: average treatment effect on the treated; ICER: incremental cost-effectiveness ratio Discussion State-owned and state-run enterprises and governmental agencies will compensate workers with either subsidies (small amounts of money for nutritional supplementation) or a few days off (with or without paid vacation), and hence, some groups will limit the number of donors. In light of these findings, the objectives of group donors for blood donation might include a combination of motives, such as modestly self-serving incentives, instead of pure altruism. In addition, with the convenience of the donation process occurring at ones’ place of work, Page 10/21 Page 10/21 donors do not perceive blood donation as time-consuming or something that must be done too far away. According to these factors, individuals rarely donate again at blood collection sites, if their group stops organizing a blood donation activity or has no quota. In the present study, most group donors refused to donate at blood collection sites because they thought it too inconvenient, but they would donate again at their group. donors do not perceive blood donation as time-consuming or something that must be done too far away. According to these factors, individuals rarely donate again at blood collection sites, if their group stops organizing a blood donation activity or has no quota. In the present study, most group donors refused to donate at blood collection sites because they thought it too inconvenient, but they would donate again at their group. Our study has some limitations. First, we only used an altruistic appeal in the SMS group, and altruism might not be the main driver for blood donation. Further research using different messages to re-recruit donors is needed to determine which type of message is the optimal intervention. Secondly, telephone and SMS reminders are different interventions, and thus, we cannot determine precisely which factor prompted the donors to return. Future research that explores how inactive donors interact with the interventions is needed to determine which factor is more effective. Moreover, even with delivery confirmation receipts, as available in newer smartphones, one cannot know with certainty that messages were read and understood. In addition, just as we mentioned above, donation records from outside of Guangzhou could not be confirmed. Therefore, the donation frequency history for inactive donors might not be correct, and the re-donation rate, especially for those who cited moving out of Guangzhou as a deterrent, might be underestimated. Consent for publication Not applicable. Acknowledgements The author thanks Mr. HE Jun who supported this project. The author would also like to thank all blood donors for giving their blood. Availability of data and material The datasets used and analysed during the current study are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. Funding This work was supported by the Natural Science Foundation of Guangdong Province 2019 (General Program: 2019A1515011505), the Western Medicine Guide Project of Health and Family Planning Commission of Guangzhou Municipality (20161A010073), the Major Project of Health Bureau of Guangzhou (20141A031001), the Key Laboratory of Guangzhou Science Technology and Innovation Commission (201509010009) and The Key Medical Disciplines and Specialties Program of Guangzhou. The funding agencies had no role in study design, data collection or analysis, decision to publish or preparation of the manuscript. Authors’ contributions OYJ and BCH did the recruitment and led the data analysis. OYJ wrote the first draft of the manuscript. FYS led the coding of the key messages. HB and CJY aided in data analysis. RX and LHQ were involved in critically revising the manuscript. All authors have read and approved the manuscript. Ethic approval and consent to participate All procedures were reviewed and approved by Institutional Review Board of the Guangzhou Blood Center (Approval Number: GZBC20161001). Oral informed consents were only retrieved from donors in telephone group but not in other two groups, since it was a low-risk study comparing operational recruitment techniques that would have been applied to the donors in any case, except for randomization; besides all data were processed anonymously. The ethics committee (The Institutional Review Board of Page 11/21 Page 11/21 the Guangzhou Blood Center) approved of the verbal consent procedure for participants in telephone group, and waiving the need for consent for the other two groups. All donors were not informed of the study goals or provided explanations about the nature of the interventions, otherwise they might have changed their return behaviour and thus compromised the results of the study. The Institutional Review Board of the Guangzhou Blood Center also approved these. The registration ID for this study on ClinicalTrial.gov is: NCT03366441. References Page 12/21 Page 12/21 1. Liang XH, Zhou SH, Fan YX, Meng QL, Zhang ZY, Gao Y, Li YJ, Liu Z: A survey of the blood supply in China during 2012-2014. Transfusion medicine 2019, 29(1):28-32. 2. Guo N, Wang J, Ness P, Yao F, Dong X, Bi X, Mei H, Li J, He W, Lu Y et al: Analysis of Chinese donors' return behavior. Transfusion 2011, 51(3):523-530. 3. Schreiber GB, Glynn SA, Damesyn MA, Wright DJ, Tu Y, Dodd RY, Murphy EL: Lapsed donors: an untapped resource. Transfusion 2003, 43(1):17-24. 4. Shi L, Wang J, Liu Z, Stevens L, Sadler A, Ness P, Shan H: Blood donor management in china. Transfusion Medicine and Hemotherapy 2014, 41(4):273-282. 5. O'Brien SF, Shao ZJ, Osmond L, Yi QL, Li CY, An QX: Donor motivation in Xi'an, China: comparison with Canadian donors. Vox sanguinis 2013, 104(3):200-206. 6. Guo N, Wang J, Yu Q, Yang T, Dong X, Wen G, Tiemuer MH, Li J, He W, Lv Y et al: Long-term return behavior of Chinese whole blood donors. Transfusion 2013, 53(9):1985-1991. 7. Whitaker B, Rajbhandary S, Harris A: The 2013 AABB blood collection, utilization, and patient blood management survey report. Bethesda (MD): AABB; 2013. 8. Lattimore S, Wickenden C, Brailsford SR: Blood donors in England and North Wales: demography and patterns of donation. Transfusion 2015, 55(1):91-99. 9. Veldhuizen I, Follea G, de Kort W: Donor cycle and donor segmentation: new tools for improving blood donor management. Vox sanguinis 2013, 105(1):28-37. 10. Sinclair KS, Campbell TS, Carey PM, Langevin E, Bowser B, France CR: An adapted postdonation motivational interview enhances blood donor retention. Transfusion 2010, 50(8):1778-1786. 11. Hall AK, Cole-Lewis H, Bernhardt JM: Mobile text messaging for health: a systematic review of reviews. Annual review of public health 2015, 36:393-415. 12. Porto-Ferreira FA, de Almeida-Neto C, Murphy EL, Montebello SC, Nogueira FA, Koga da Silva EM, MacFarland W, Custer B: A randomized trial to evaluate the use of text messaging, letter, and telephone call reminders to improve return of blood donors with reactive serologic tests. Transfusion 2017, 57(1):102-107. 13. Saleem S, Wasim A, Sabih S, Khan AF, Rizvi MH, Jillani UA, Syed MJ, Mumtaz M, Mumtaz Y, Shehzad AM et al: Assessing Acceptability of Short Message Service Based Interventions towards Becoming Future Voluntary Blood Donors. Journal of blood transfusion 2014, 2014:567697. 14. References Gemelli CN, Carver A, Garn A, Wright ST, Davison TE: Evaluation of the impact of a personalized postdonation short messaging service on the retention of whole blood donors. Transfusion 2018, 58(3):701-709. 15. Godin G, Amireault S, Vezina-Im LA, Germain M, Delage G: The effects of a phone call prompt on subsequent blood donation among first-time donors. Transfusion 2011, 51(12):2720-2726. 15. Godin G, Amireault S, Vezina-Im LA, Germain M, Delage G: The effects of a phone call prompt on subsequent blood donation among first-time donors. Transfusion 2011, 51(12):2720-2726. 16. Ferguson E, France CR, Abraham C, Ditto B, Sheeran P: Improving blood donor recruitment and retention: integrating theoretical advances from social and behavioral science research agendas. Transfusion 2007, 47(11):1999-2010. Page 13/21 17. Duboz P, Cuneo B: How barriers to blood donation differ between lapsed donors and non-donors in France. Transfusion medicine 2010, 20(4):227-236. 18. Klinkenberg EF, Romeijn B, de Kort WL, Merz EM: Reasons to end the donor career: a quantitative study among stopped blood donors in the Netherlands. Transfusion medicine 2018, 28(3):200-207. 19. Weidmann C, Müller-Steinhardt M, Schneider S, Weck E, Klüter H: Characteristics of Lapsed German Whole Blood Donors and Barriers to Return Four Years after the Initial Donation. Transfusion Medicine and Hemotherapy 2012, 39(1):9-15. 20. Piersma TW, Bekkers R, Klinkenberg EF, De Kort W, Merz EM: Individual, contextual and network characteristics of blood donors and non-donors: a systematic review of recent literature. Blood transfusion 2017, 15(5):382-397. 21. Charbonneau J, Cloutier MS, Carrier E: Why Do Blood Donors Lapse or Reduce Their Donation's Frequency? Transfusion medicine reviews 2016, 30(1):1-5. 22. Ferguson E, Atsma F, de Kort W, Veldhuizen I: Exploring the pattern of blood donor beliefs in first-time, novice, and experienced donors: differentiating reluctant altruism, pure altruism, impure altruism, and warm glow. Transfusion 2012, 52(2):343-355. 23. Bagot KL, Murray AL, Masser BM: How Can We Improve Retention of the First-Time Donor? A Systematic Review of the Current Evidence. Transfusion medicine reviews 2016, 30(2):81-91. 24. Charbonneau J, Cloutier MS, Carrier E: Whole blood and apheresis donors in Quebec, Canada: Demographic differences and motivations to donate. Transfusion and apheresis science 2015, 53(3):320-328. 25. Shi L, Wang JX, Stevens L, Ness P, Shan H: Blood safety and availability: continuing challenges in China's blood banking system. Transfusion 2014, 54(2):471-482. 26. Czeizler A, Garbarino E: Give blood today or save lives tomorrow: Matching decision and message construal level to maximize blood donation intentions. References Health marketing quarterly 2017, 34(3):175- 186. 27. Guangzhou Blood Center: PAss Aladdin System. In.: Passsoft CO. LTD.; 2017. 27. Guangzhou Blood Center: PAss Aladdin System. In.: Passsoft CO. LTD.; 2 28. LIU Z: Analysis of populational distribution status in Haixi City from 2006 to 2010 (in Chinese). Qinghai Medical Journal 2012, 42(5):65-66. 29. Ou-Yang J, He B, Rong X, Bei CH: Can inactive blood donors be re-recruited? A stratified randomised pilot study. Transfusion medicine 2017, 27(6):421-427. 30. Angrist JD, Imbens GW, Rubin DB: Identification of Causal Effects Using Instrumental Variables. Journal of the American Statistical Association 1996, 91(434):444-455. 31. Holland PW: Statistics and Causal Inference. Journal of the American Statistical Association 2012, 81(396):945-970. 32. Frangakis CE, Rubin DB: Principal stratification in causal inference. Biometrics 2015, 58(1):21-29. 32. Frangakis CE, Rubin DB: Principal stratification in causal inference. Biom 33. Frölich M, Melly B: Identification of Treatment Effects on the Treated with One-Sided Non- Compliance. Econometric Reviews 2013, 32(3):384-414. Page 14/21 Page 14/21 34. Chou EY, Murnighan JK: Life or death decisions: framing the call for help. PloS one 2013, 8(3):e57351. 35. Bednall TC, Bove LL: Donating blood: a meta-analytic review of self-reported motivators and deterrents. Transfusion medicine reviews 2011, 25(4):317-334. 35. Bednall TC, Bove LL: Donating blood: a meta-analytic review of self-reported motivators and deterrents. Transfusion medicine reviews 2011, 25(4):317-334. 36. Campbell AV, Tan C, Boujaoude FE: The ethics of blood donation: does altruism suffice? Biologicals : journal of the International Association of Biological Standardization 2012, 40(3):170-172. 37. Charng H, Pilivin J, Callero P: Roles identity and reasoned action in the prediction of repeated behaviour. Social Psychology Quarterly 1998, 51(4):303-317. 37. Charng H, Pilivin J, Callero P: Roles identity and reasoned action in the prediction of repeated behaviour. Social Psychology Quarterly 1998, 51(4):303-317. 38. Wevers A, Wigboldus DH, de Kort WL, van Baaren R, Veldhuizen IJ: Characteristics of donors who do or do not return to give blood and barriers to their return. Blood transfusion 2014, 12 Suppl 1:s37-43. 39. Piliavin JA: Why do they give the gift of life? A review of research on blood donors since 1977. Transfusion 1990, 30(5):444-459. 39. Piliavin JA: Why do they give the gift of life? A review of research on blood donors since 1977. Transfusion 1990, 30(5):444-459. 40. References Wevers A, Wigboldus DH, van den Hurk K, van Baaren R, Veldhuizen IJ: Increasing first-time blood donation of newly registered donors using implementation intentions and explicit commitment techniques. Vox sanguinis 2015, 108(1):18-26. 40. Wevers A, Wigboldus DH, van den Hurk K, van Baaren R, Veldhuizen IJ: Increasing first-time blood donation of newly registered donors using implementation intentions and explicit commitment techniques. Vox sanguinis 2015, 108(1):18-26. 41. Tison GH, Liu C, Ren F, Nelson K, Shan H: Influences of general and traditional Chinese beliefs on the decision to donate blood among employer-organized and volunteer donors in Beijing, China. Transfusion 2007, 47(10):1871-1879. Tables Table 1 Average treatment effect on the treated estimations among three groups Point Estimate Standard Error 95% CI Z P Telephone vs. SMS 0.023 0.008 0.007, 0.038 2.885 0.004 Telephone vs. Control 0.060 0.009 0.041, 0.078 6.363 <0.001 SMS vs. Control 0.004 0.011 -0.017, 0.025 0.372 0.710 Table 2 Summary of reactivated donors among all participants in the three groups Table 1 Average treatment effect on the treated estimations among three groups Point Estimate Standard Error 95% CI Z P Telephone vs. SMS 0.023 0.008 0.007, 0.038 2.885 0.004 Telephone vs. Control 0.060 0.009 0.041, 0.078 6.363 <0.001 SMS vs. Control 0.004 0.011 -0.017, 0.025 0.372 0.710 erage treatment effect on the treated estimations among three gro Table 1 Average treatment effect on the treated estimations amon Table 2 Summary of reactivated donors among all participants in the three groups Table 2 Summary of reactivated donors among all participants in the three groups Page 15/21 Telephone group -1 SMS group -2 Control group -3 P 1 vs. 2 1 vs. 3 2 vs. 3 Gender Male 164 (51.1) 186 (55.2) 158 (54.3) 0.292 0.428 0.822 Female 157 (48.9) 151 (44.8) 133 (45.7) Age, years* 36.8±8.0 37.0±8.0 37.3±7.5 0.873 0.203 0.150 Previous donations, n* 4.8±6.5 4.5±3.5 4.4±3.1 0.250 0.167 0.597 Type of re-donation Whole blood 306 (95.3) 328 (97.3) 282 (96.9) 0.003 0.363 0.388 Apheresis platelet 15 (4.7) 9 (2.7) 9 (3.1) Additional donations, n 1 291 (90.7) 305 (90.5) 269 (92.4) 0.213 0.318 0.752 ≥2 30 (9.3) 32 (9.5) 22 (7.6) Data presented as no. (%), unless otherwise stated. Data presented as mean ± standard deviation. Data presented as no. (%), unless otherwise stated. Data presented as mean ± standard deviation. Table 3 Logistic regression analysis of associations of groups, donor characteristics, and intervention status with re-donation among all participants OR (95% CI) 1 vs. 2 vs. 3 OR (95% CI) 1 vs. 2 OR (95% CI) 1 vs 3 OR (95% CI) 2 vs. Tables 3 Group Telephone call-1 0.92 (0.76, 1.11) 1.10 (0.93, 1.31) 0.80 (0.55, 1.00) - SMS-2 0.83 (0.66, 1.04) reference - 1.17 (0.89, 1.55) Control-3 reference - reference reference Gender Male 1.14 (0.99, 1.31) 1.13 (0.96, 1.33) 1.10 (0.93, 1.30) 1.20 (1.02, 1.41)* Female reference reference reference reference Age 1.03 (1.02, 1.04) 1.03 (1.02, 1.04) 1.03 (1.02, 1.04) 1.03 (1.02, 1.04) Donation history 1.08 (1.06, 1.10) 1.11 (1.08, 1.13) 1.07 (1.05, 1.09) 1.06 (1.04, 1.08) Status of intervention# Successful 1.56 (1.30, 1.88) 1.56 (1.29, 1.88) 2.01 (1.60, 2.55)** 1.02 (0.77, 1.35) Failed reference reference reference reference Coding: Telephone call = 1, SMS = 2, Control = 3; Male = 1, Female = 2; Status of intervention (Successful) =1, (Failed/Control) = 2 Table 3 Logistic regression analysis of associations of groups, donor characteristics, and intervention status with re-donation among all participants Coding: Telephone call = 1, SMS = 2, Control = 3; Male = 1, Female = 2; Status of intervention (Successful) =1, (Failed/Control) = 2 #: Referred to those who accepted the interventions successfully; : P<0.05; : P<0.001 Page 16/21 Page 16/21 Table 4 Logistic regression analysis of associations of groups and donor characteristics with donor reactivation among those who were successfully contacted OR (95% CI) 1 vs. 2 vs. 3 OR (95% CI) 1 vs. 2 OR (95% CI) 1 vs 3 OR (95% CI) 2 vs. Coding: Telephone call = 1, SMS = 2, Control = 3; Male = 1, Female = 2 Tables 3 Group Telephone call-1 1.63 (1.34, 1.98)* 1.33 (1.09, 1.63) 1.62 (1.33, 1.97) - SMS-2 1.20 (1.00, 1.43)* reference - 1.20 (1.01, 1.43)* Control-3 reference - reference reference Gender Male 1.25 (1.07, 1.45)* 1.28 (1.05, 1.56)* 1.19 (0.99, 1.44) 1.24 (1.04, 1.48)* Female reference Reference reference Reference Age 1.02 (1.01, 1.03)** 1.01 (1.00, 1.03)* 1.03 (1.02, 1.04) 1.02 (1.10, 1.03) Donation history 1.06 (1.04, 1.08) 1.10 (1.06, 1.13) 1.05 (1.03, 1.07) 1.06 (1.04, 1.08) Coding: Telephone call = 1, SMS = 2, Control = 3; Male = 1, Female = 2 Table 4 Logistic regression analysis of associations of groups and donor characteristics with donor reactivation among those who were successfully contacted Coding: Telephone call = 1, SMS = 2, Control = 3; Male = 1, Female = 2 : P<0.05; : P<0.001 Table 5 Donors’ 1-year re-donation intervals (in days) among the three groups (mean±SD) Intervals for reactivated donors who were enrolled Intervals for reactivated donors who were successfully contacted Telephone group 1 to 365 (157.4±102.8) 1 to 365 (163.3±95.8) SMS group 1 to 365 (141.5±102.7) 1 to 365 (155.5±101.4) Control group 1 to 365 (151.7±99.2) Not applicable Table 5 Donors’ 1-year re-donation intervals (in days) among the three groups nors’ 1-year re-donation intervals (in days) among the three group Table 5 Donors’ 1-year re-donation intervals (in days) among the t Table 6 Incremental cost-effectiveness ratio estimations of telephone and SMS groups Table 6 Incremental cost-effectiveness ratio estimations of telephone and SMS groups Page 17/21 Page 17/21 Page 17/21 Cost per participant (C) (RMB¥) Effectiveness (E) (%) DC DE Ratio (DC/DE) Control group - 7.4 - - - Telephone group 3.9 8.1 3.9 0.007 557.1 SMS group 0.7 8.4 0.7 0.010 70 Table 7 Medical reasons for the lack of re-donation reported by donors who were successfully contacted by telephone (n, %) Table 7 Medical reasons for the lack of re-donation reported by donors who were successfully contacted by telephone (n, %) Self-perception of inadequate health status 197 (40.8) Pregnancy/lactation 124 (25.7) Confirmed diagnosis of severe disease 43 (8.8) Self-perception as being too old to donate 42 (8.7) Becoming unhealthy after blood donation 33 (6.8) Confirmed diagnosis of anaemia 23 (4.8) Other temporary reasons for deferral 21 (4.3) *Other temporary reasons for deferral included reasons such as ineligible weight, menstrual disorder etc. Tables Table 8 Summary of willingness to re-donate and re-donation status of donors contacted by telephone according to the different reported reasons for deferral (n, %) Table 8 Summary of willingness to re-donate and re-donation status of donors contacted by telephone according to the different reported reasons for deferral (n, %) Page 18/21 Total Willing to re- donate Not willing to re- donate Uncertain Actually re- donated Time constraints 485 444 (91.5) 5 (1.0) 36 (7.4) 69 (14.2) Medical reasons 483 266 (55.1) 150 (61.0) 67 (13.9) 37 (7.7) Self-perception of inadequate health status (unrelated to donation) 197 125 (63.5) 38 (19.3) 31(15.7) 20 (8.7) Pregnancy/lactation 124 101(81.4) 7 (5.6) 16 (12.9) 5 (4.0) Self-perception as being too old to donate 42 14 (33.3) 14 (33.3) 14 (33.3) 3 (7.1) Becoming unhealthy after blood donation (related to donation) 33 7 (21.2) 22 (66.7) 4 (12.1) 3 (9.1) Moving away from Guangzhou 195 55 (28.2) 136 (69.7) 4 (2.1) 18 (9.3) Group-sponsored donation 179 160 (89.4) 7 (3.9) 12 (6.7) 26 (14.5) Not wanting to donate again 65 25 (38.5) 21 (32.3) 19 (29.2) 6 (7.7) Being far away from blood collection locations 56 53 (94.6) 1 (1.8) 2 (3.6) 12 (21.4) Forgetting to donate 31 30 (96.8) 1 (3.2) 0 (0.0) 6 (19.4) Inability to be prioritized to receive blood 27 8 (29.6) 15 (55.6) 4 (14.8) 4 (14.8) Adverse reaction 23 3 (12.5) 2 (8.7) 18 (75.0) 1 (4.2) Additional Files - Additional file 1 - Interview script and examples of responses - Contents are the recruitment message scripts of the telephone groups. - Contents are the recruitment message scripts of the telephone groups. - CONSORT 2010 checklist of information to include when reporting a randomised trial - Contents are the CONSORT 2010 checklists of this study. - CONSORT 2010 checklist of information to include when reporting a randomised trial - Contents are the CONSORT 2010 checklists of this study. Figures Page 19/21 gure 1 Figure 2 Distribution of self-reported reasons for deferral. Figure 1 Flow chart of the study design. Flow chart of the study design. Flow chart of the study design. Page 20/21 Figure 2 Distribution of self-reported reasons for deferral. Supplementary Files Figure 2 Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. Equation1.jpg BMCAdditionalfile2.doc Equation2.jpg BMCAdditionalfile1.docx Page 21/21
https://openalex.org/W1873164654	https://publications.goettingen-research-online.de/bitstream/2/38349/2/c5cp03013g.pdf	English	null	The optical properties and quantum chemical calculations of thienyl and furyl derivatives of pyrene	Physical chemistry chemical physics/PCCP. Physical chemistry chemical physics	2,015	cc-by	7,733	a Geoscience Centre of the University of Go¨ttingen, Dept. Applied Geology, Goldschmidtstr. 3, 37077 Go¨ttingen, Germany. E-mail: krzysztof.idzik@pwr.wroc.pl b Functional Materials and Devices, Fraunhofer Institute for Applied Polymer Research, Geiselbergstr. 69, 14476 Potsdam-Golm, Germany c Czestochowa University of Technology, Faculty of Electrical Engineering, Armii Krajowej 17, 42-201 Czestochowa, Poland d VENITUR Sp. z o.o., Wawozowa 34 B, 31-752 Krakow, Poland e Polish Academy of Sciences, Institute of Physical Chemistry, Kasprzaka 44/52, 01-224 Warsaw, Poland Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Cite this: Phys.Chem.Chem.Phys., 2015, 17, 22758 Krzysztof R. Idzik,a Piotr J. Cywin´ski,b Wojciech Kuznik,c Jaroslaw Frydel,d Tobias Lichaa and Tomasz Ratajczyke A detailed electrochemical, photophysical and theoretical study is presented for various new thienyl and furyl derivatives of pyrene. Their optical properties are described based on UV-VIS absorption and both steady-state and time-resolved fluorescence spectroscopy. DFT and TDDFT calculations are also presented to support experimental data. The calculations results show that HOMO–LUMO orbitals are delocalized uniformly between aromatic core and aryl substituents. Good electrochemical stability of thienyl and furyl hybrids of pyrene confirm their potential application for light emitting electrochemical cells or spintronics mainly due to their beneficial optical and charge transport properties in electrochromic devices. In order to demonstrate this potential, an OLED device is presented. Synthesized compounds included in this OLED device both facilitate electron transport and act as a light emitting layer. Received 26th May 2015, Accepted 21st July 2015 liquid crystals,37,38 and photoactive polypeptides.39 Pyrene is a flat aromatic molecule and exhibits excellent fluorescence properties. Its pure blue emission already permits its exploita- tion for its use in OLEDs.40 The pyrene core is particularly interesting in this context, since it offers numerous possibilities for peripheral group modifications with profound consequences on the condensed-matter structure. For example, unsubstituted pyrene forms monoclinic crystals, while tetraethynylpyrene deriva- tives of phenylethynyl form liquid crystalline columnar phases,41 and pyrenes similarly substituted with diethyleneglycolether deri- vatives display a liquid phase at room temperature.42 Pyrene is an efficient luminescent material, which has been investigated for several decades. However, self-quenching can be observed at high concentrations or pure crystals due to a molecular aggregation via p-stacking.43–45 However, the luminescence properties of the pyrene core might be modulated using the proper molecular substitution, e.g., with heterocyclic units. PCCP View Article Online View Journal \| View Issue This journal is ©the Owner Societies 2015 PCCP PCCP 1. Introduction Organic materials possessing an extended p-conjugation have received immense attention in recent years. This is due to their unique photophysical and charge transport properties, which make them potential materials for applications in electronic devices such as organic light emitting diodes (OLEDs),1–8 organic photovoltaics (OPV),9–15 or organic thin film transistors (OTFT).16–20 Further, they are attractive due to their promising two photon absorption21–24 and nonlinear optical21–24 characteristics. Fluorescence probes based on pyrene moieties have attracted considerable attention over the past decades.25–32 The pyrene ring is of special interest because its hydrogen atoms can easily be substituted using a wide range of chemical groups, particularly aryl groups such as thiophene or furan. The pyrene structure is well studied and has been intensively examined in the fields of biology, chemistry and physics. Moreover, due to its electro- conductive properties, pyrene and its derivatives have been successfully applied as micro-environmental sensors,33–36 Organic p-conjugated structures containing thiophene units play an important role in the search for new materials and their novel applications46 such as organic solar cells47,48 or organic field eﬀect transistors.49 Thiophenes with well-defined chemical structures have recently attracted great attention not only as an example of model compounds for conducting materials, but also as a new class of functional p-electron systems.50–56 Furan-based polymers are among the most widely used p-conjugated systems in organic electronic devices. For example, polymers containing furyl groups show an excellent performance when acting as building materials for a highly eﬃcient OLEDs.57 Another reason for using furan is its lower resonance stabilization 22758 \| Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 This journal is ©the Owner Societies 2015 enzene. Thus, chromophores with furan th the details of the organic materials and their yer. Fig. 2 Electronic absorption spectra for all pyrene thiophene derivatives in chloroform (in ex. coeﬃcient units). PCCP View Article Online View Article Online Fig. 1 OLED structure with the details of the organic materials and their functions used in each layer. Paper Paper Fig. 1 OLED structure with the details of the organic materials and their functions used in each layer. Paper Fig. 2 Electronic absorption spectra for all pyrene thiophene derivatives in chloroform (in ex. coeﬃcient units). Fig. 1 OLED structure with the details of the organic materials and their functions used in each layer. Fig. 2 Electronic absorption spectra for all pyrene thiophene derivatives in chloroform (in ex. coeﬃcient units). energy compared to benzene. 3.1. Synthesis The chemical structure and synthetic route to the pyrene deriva- tives are illustrated in Scheme 1. Bromination of pyrene (1) with one to four equivalents of bromine gave the mono-, di-, tri-, and tetrabromopyrenes, respectively. The bromination resulted in 1-bromopyrene (2), 1,6-dibromopyrene (3) or 1,4-dibromopyrene (4), 1,3,6-tribromopyrene (5), and 1,3,6,8-tetrabromopyrene (6),21 respectively. Mono- and dibromopyrene were prepared in DCM, while in the case of tri- and tetrabromopyrene we used nitro- toluene as a solvent. The cross-coupling reaction of these bromo- pyrenes with 2-(tributylstannyl)furan under the conditions of the Stille reaction gave mono-, bis-, tris-, and tetrakis(furyl) pyrenes 7–11, respectively. For the preparation of di(furyl)pyrenes we used 2.2. Optical measurements Electronic absorption spectra were collected on a UV-VIS absorp- tion spectrometer Lambda 35 (Perkin Elmer, Rodgau, Germany). The spectra were corrected with solvent absorption spectra to obtain final absorption spectra for the studied compounds. Steady-state fluorescence emission spectra were collected on an FLS920-stm spectrometer (Edinburgh Instruments, Livingstone, United Kingdom). The spectra were detector response corrected. Fluorescence decays were also acquired on the FLS920-stm spectrometer using the Time Correlated Single Photon Counting (TCSPC) technique with a sub-nanosecond pulsed LED (EPLED 320) as an excitation source. Fluorescence decay times were determined from the decays using the least squares fit method. The fitting was assumed to be correct when the goodness-of-fit value w2 was lower than 1.2. Fluorescence quantum yields were measured on a C9920-02G absolute quantum yield (F) measurement system from Hamamatsu (Hamamatsu Photonics Deutschland GmbH, Herrsching am Ammersee, Germany). All measurements were performed using a 3 ml quartz cuvette (Hellma GmbH, Germany) with 1 cm light path. All measurements were obtained for samples with optical densities below 0.15. poly(styrenesulfonate) (Sigma-Aldrich). Synthesized low molecular materials were used for the preparation of the light-emitting layer (EML). All measurements were performed using laboratory power supply EA-PS 2032-025 (Fig. 1). 2.1. Synthesis All chemicals, reagents, and solvents were used as received from commercial sources without further purification. Melting points of the synthesized dyes were determined using a Stuart smp 20 melting point apparatus. Fig. 3 Normalized fluorescence emission spectra for all dyes in chloroform. 2.3. Quantum-chemical calculations Fig. 4 Electronic absorption spectra for all dyes in chloroform (in ex. coeﬃcient units). All quantum mechanical calculations were made at the DFT level using the GAUSSIAN09 package80 and GABEDIT 2.4.6 Graphical User Interface.81 The 6-31G basis set and the B3LYP functional were used. The polarizable continuum model (PCM) was used in all calculations to account for the solvent influence (chloroform).82 The absorption and emission spectra and excited state geometries were obtained from time dependent DFT (TDDFT) singlet calcula- tions. The reported lowest unoccupied molecular orbitals (LUMO) energy values are read as DFT highest occupied molecular orbital (HOMO) level reduced by the TDDFT energy gap. 1. Introduction Thus, chromophores with furan moieties display a molecular hyperpolarizability, which is higher than those of their their benzene analogues.58 Oligofurans with well-defined structures have also received great attention, not only exemplarily as model compounds for conducting polyfurans, but also as a new class of functional p-electron systems. Over the last decade, highly eﬃcient blue OLEDs have attracted considerable attention due to their potential applica- tions in full color ultra-thin flat panel displays.69–75 Organic light sources are currently made from either low molecular weight organic materials or polymers. For the former, the layer structure of the OLED is usually deposited on a rigid base made from glass or metal.76–79 Considering the current literature, one can conclude that, when compared to other heteroaromatics such as thiophene or pyrrole, furan has great potential for further studies.59–66 This may partially be attributed to the chemical lability of furan and its derivatives. For example, furan serves as a diene in the [4+2] Diels–Alder reactions under milder conditions than those for thiophene and pyrrole.67 On the other hand, the relatively low aromaticity of furan compared to those of thiophene and pyrrole may give rise to different electronic properties in the resulting furan-based materials.68 Suzuki, Stille, Kumada, and Negishi coupling reactions are well-described synthetic methods that are often used to prepare electroactive compounds. In this paper, we also present a series of symmetric molecules in which a pyrene core is connected to thiophene and furan units. All compounds were obtained using the Stille cross-coupling reaction. Furthermore, we describe DFT cheme 1 Synthesis of thienyl and furyl derivatives of pyrene Scheme 1 Synthesis of thienyl and furyl derivatives of pyrene. Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 \| 22759 Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 \| 22759 This journal is ©the Owner Societies 2015 View Article Online PCCP Paper Fig. 3 Normalized fluorescence emission spectra for all dyes in chloroform. calculations and TDDFT simulations of molecular materials based on a pyrene core which is connected to thiophene and furan units. We also report the results of UV-VIS and fluores- cence spectroscopy measurements for the obtained compounds. Additionally, we present a fully functioning OLED device using a synthesized low molecular weight material to act as an electron transporting and light emitting layer (Fig. 1). This journal is ©the Owner Societies 2015 2.4. OLED preparation Indium tin oxide (ITO)-coated glass with a sheet resistance of about 8–12 O sq1 (Sigma-Aldrich) was used for preparing the OLED, while the cathode was made from Al. The hole-conductive layer (HTL) was PEDOT: PSS poly(3,4-ethylenedioxythiophene): Fig. 4 Electronic absorption spectra for all dyes in chloroform (in ex. coeﬃcient units). 22760 \| Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 This journal is ©the Owner Societies 2015 a mi t re of 1 6 dibromop rene (3) and 1 4 dibromop rene (4) conditions ith moderate to good ields All compo nds st died i Table 1 Absorption and emission properties of the studied thienyl derivatives of pyrene Name/struktur labs (nm) e (M1 cm1) lem (nm) F (%) Dn (cm1) t (ns) kf (108 s1) knr (108 s 292 2 4000 435 19.7 3893 0.40 4.9 20.1 372 34 000 461 268 28 000 470 22.0 3354 0.46 4.8 17.0 315 31 000 406 36 000 295 27 000 435 22.2 3893 0.61 3.6 12.8 372 34 000 461 280 25 000 405 15.1 4044 0.95 1.6 8.9 348 26 000 305 27 000 446 18.1 3220 0.43 4.2 19.0 390 40 000 464 Paper PCC Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. View Article Onlin Table 1 Absorption and emission properties of the studied thienyl derivatives of pyrene 292 372 Thi a m We pyre stan The Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. a mixture We succes pyrene (9). stannyl)thi The react Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. uly 2015. Downloaded on 08/12/2015 13:31:58. Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. ed on 28 July 2015. Downloaded on 08/12/2015 13:31:58. 268 315 406 Creative Commons Attribution 3.0 Un le. Published on 28 July 2015. Downloaded on 08/12/2015 1 Access Article. Published on 28 July 2015. Downloaded on 0 Open Access Article. Published on 28 July 2015. Dow 295 372 ss Article. Published on 28 July This article is licensed under a Open Access Article. 2.4. OLED preparation This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 5. Downloaded on 08/12/2015 13:31:58. ed on 28 July 2015. Downloaded on 08/12/2015 13:31:58. 330 32 000 479 65.8 2379 1.8 3.7 1.9 430 32 000 330 430 Creative Commons Attribution 3.0 Unp le. Published on 28 July 2015. Downloaded on 08/12/2015 1 Access Article. Published on 28 July 2015. Downloaded on 0 Open Access Article. Published on 28 July 2015. Dow 299 388 t c e. ub s ed o 8 Ju y 0 his article is licensed under a Cr Open Access Article. Published on 28 Jul 289 362 315 410 2.4. OLED preparation Published on 28 Jul 280 348 305 390 305 390 a mixture of 1,6-dibromopyrene (3) and 1,4-dibromopyrene (4). We successfully isolated 1,6-di(furyl)pyrene (8) from 1,4-di(furyl)- pyrene (9). Instead of 2-(tributylstannyl)furan we used 2-(tributyl- stannyl)thiophene to prepare the thienyl derivative of pyrenes 12–15. The reactions were conducted under easy-to-perform, mild a mixture of 1,6-dibromopyrene (3) and 1,4-dibromopyrene (4). We successfully isolated 1,6-di(furyl)pyrene (8) from 1,4-di(furyl)- pyrene (9). Instead of 2-(tributylstannyl)furan we used 2-(tributyl- stannyl)thiophene to prepare the thienyl derivative of pyrenes 12–15. The reactions were conducted under easy-to-perform, mild conditions with moderate to good yields. All compounds studied in this work were obtained by the Stille cross-coupling methodology according to the procedure reported in our previous work (Scheme 1).83–86 Table 7 presents the melting points of the synthesized dyes. Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 \| 22761 This journal is ©the Owner Societies 2015 Table 2 Absorption and emission properties of the studied furyl derivatives of pyrene Name/structure labs (nm) e (M1 cm1) lem (nm) F (%) Dn (cm1) t (ns) kf (108 s1) knr (108 s1) 297 31 000 433 67.1 2546 1.6 4.2 2.1 390 42 000 463 PCCP Paper View Article Online Name/structure labs (nm) e (M1 cm1) lem (nm) F (%) Dn (cm1) t (ns) kf (108 s1) knr (108 s 297 31 000 433 67.1 2546 1.6 4.2 2.1 390 42 000 463 330 32 000 479 65.8 2379 1.8 3.7 1.9 430 32 000 299 30 000 433 68.7 2679 1.9 3.6 1.6 388 33 000 463 289 29 000 403 38.2 2810 3.3 1.2 1.9 362 29 000 423 315 35 000 460 63.2 2651 1.7 3.7 2.3 410 39 000 480 Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 297 390 uly 2015. Downloaded on 08/12/2015 13:31:58. 330 430 299 388 289 362 315 410 Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 3.2. Opti Electronic derivatives for pyrene absorption Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 22762 3.2. Electr deriv for p absor deriv Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. 3.2. Optical measurements The derivative containing four thiophenes exhibits similar absorption features to pyrene; however, the spectrum displays a significant red shift. This feature can be associated with pyrene or furan electron transitions mixed with the transitions charac- teristic for thiophene. Depending on the derivative, the higher energy transitions (presumably p - p) can be observed in the Electronic absorption spectra collected for thienyl and furyl derivatives are shown in Fig. 2 and in Fig. 4, respectively. Unlike for pyrene itself, for which three bands are observed in its absorption spectrum, most of the studied thienyl and furyl derivatives have only two absorption bands. 22762 \| Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 This journal is ©the Owner Societies 2015 Fig. 7 Experimental (top panel) and TDDFT fluorescence spectra of the pyrene derivatives K1–K5. PCCP View Article Online PCCP View Article Online Fig. 5 Normalized fluorescence emission spectra for all dyes in chloroform. Fig. 7 Experimental (top panel) and TDDFT fluorescence spectra of the pyrene derivatives K1–K5. Paper PCCP Paper Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Fig. 5 Normalized fluorescence emission spectra for all dyes in chloroform. Fig. 7 Experimental (top panel) and TDDFT fluorescence spectra of the pyrene derivatives K1–K5. range of 280–315 nm, while the lower energy transitions (presumably n - p) can be found in the spectral range between 350 nm and 405 nm. The n - p* transitions can be associated with the presence of heteroatoms (O or S) within the dye structures. A bathochromic shift can also be observed with an increasing number of electron-donating thiophene or furan moieties within the chemical dye structure. However, for the thienyl derivatives the observed shift is lower than the one for furyl derivatives. For the higher energy band the shift is approximately 35 nm for thienyl and 50 nm for furyl derivatives, while for the lower energy band the shift is respectively larger with 55 nm and 75 nm, respectively. Both, hyperchromic effect and hypochromic effect are also observed for molecules with different numbers of thiophene moieties. Both effects can be associated with the structural symmetry. In non-symmetrical systems, the absorption increases with the number of thio- phene moieties, while for symmetrical molecules it decreases. This journal is ©the Owner Societies 2015 3.2. Optical measurements Depending on the derivative and the investigated absorption band, molar extinction coefficients (e) determined on the basis of the Beer–Lambert law were found in the range of 24 000– 40 000 M1 cm1 for thienyl derivatives and in the range of 29 000–42 000 M1 cm1 for furyl derivatives. Noticeably, the e values are basically in the same order of magnitude as the corresponding values for pyrene itself. The summary of para- meters determined for thienyl and furyl derivatives based on absorption measurements are presented in Table 1 and Table 2, respectively. p y The steady-state fluorescence emission spectra collected for thienyl and furyl derivatives are shown in Fig. 3 and in Fig. 5, respectively. For both classes of compounds, a bathochromic Fig. 6 Experimental (top panel) and TDDFT electronic absorption spectra of the pyrene derivatives K1–K5. l (t l) d TDDFT l t i b ti t f th d i ti K1 K5 Fig. 6 Experimental (top panel) and TDDFT electronic absorption spectra of the pyrene derivatives K1–K5. Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 \| 22763 This journal is ©the Owner Societies 2015 View Article Online PCCP Paper Importantly, pyrene emission is still dominating in the emis- sion observed for derivatives with only one side group, but for the derivative with four substituents the spectrum is almost completely merged into one bold spectrum, even though the separate peaks can still be recognized. The spectra collected for derivatives with 2 and 3 substitutes are more structured. The merging eﬀect can be associated with decreasing energies due to radiationless deactivation. In all cases, the third band on the shift is observed with an increasing number of electron-donating (thiophene or furan) groups and the relative ratios between corresponding peaks are diﬀerent for diﬀerent derivatives. Noticeably, when two substituents are present within the chemical structure either in cis or trans form, the fluorescence emission spectra are identical, which shows that the position of substituent has no eﬀect on the emission frequencies. This eﬀect has been observed both for thienyl and furyl derivatives. eﬀect has been observed both for thienyl and furyl derivatives. to radiationless deactivation. 3.2. Optical measurements In all cases, the third band on the Table 3 Molecular parameters derived from (TD)DFT calculations HOMO [eV] LUMO (=Eg HOMO) [eV] Eg (absorption) [eV nm1] Eg (emission) [eV nm1] Average dihedral angle between the central moiety and the substituents in ground/excited state [deg] K1 5.23 2.25 2.97/417 2.39/518 133/153 K2 5.15 2.47 2.68/463 2.16/573 130/144 K3 5.23 2.25 2.98/416 2.39/518 131/152 K4 5.30 2.08 3.22/384 2.65/467 134/159 K5 5.18 2.37 2.81/440 2.28/544 131/148 Table 4 Graphical representation of HOMO and LUMO orbitals in the studied molecules in the ground and excited states Compound HOMO LUMO HOMO – excited state LUMO – excited state K1 K2 K3 K4 K5 Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 22764 \| Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 This journal is ©the Owner Societies 2015 Table 3 Molecular parameters derived from (TD)DFT calculations HOMO [eV] LUMO (=Eg HOMO) [eV] Eg (absorption) [eV nm1] Eg (emission) [eV nm1] Average dihedral angle between the central moiety and the substituents in ground/excited state [deg] K1 5.23 2.25 2.97/417 2.39/518 133/153 K2 5.15 2.47 2.68/463 2.16/573 130/144 K3 5.23 2.25 2.98/416 2.39/518 131/152 K4 5.30 2.08 3.22/384 2.65/467 134/159 K5 5.18 2.37 2.81/440 2.28/544 131/148 Table 4 Graphical representation of HOMO and LUMO orbitals in the studied molecules in the ground and excited states Compound HOMO LUMO HOMO – excited state LUMO – excited state K1 K2 K3 K4 K5 Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licenc Table 3 Molecular parameters derived from (TD)DFT calculations Table 3 Molecular parameters derived from (TD)DFT calculations 22764 \| Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 This journal is ©the Owner Societies 2015 Compound HOMO LUMO HOMO – excited state LUMO – excited state K1 K2 K3 K4 K5 Ope ccess t c e. ub s ed o 8 Ju y 0 5. This article is licensed under a Creativ 22764 \| Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 This journal is ©the Owner Societies 2015 View Article Online View Article Online Paper Fig. 8 TDDFT (bottom panel) and experimental (top) absorption spectra of the compounds K6–K10. Paper PCCP ve Commons Attribution 3.0 Unported Licence. tom panel) and experimental (top) absorption spectra of the compounds K6–K10. PCCP Fig. 8 TDDFT (bottom panel) and experimental (top) absorption spectra of the compounds K6–K10. spectra stays virtually unaﬀected and only a red shift is observed for it. Independent from the spectral features similar for both classes of compounds, significant diﬀerences in photophysical properties can be noticed when quantum yields (F) and decay times (t) are considered. The thienyl derivatives exhibit significantly lower F (70% vs. 20%) and much shorter fluorescence lifetimes (from 0.4 to 0.95 ns compared to B2 ns) when compared to their furan-substituted counterparts. Notice- ably, the fluorescence quantum yield is virtually independent of the number of substituents and their orientation (cis or trans). Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. For the thienyl derivatives the non-radiative processes rate constant knr is on average 5 times larger than the radiative process rate constant kf, while for the furyl derivatives kf was on average two times greater than knr. This result is a clear evidence for the domination of the non-radiation processes in the deactivation path of thienyl derivatives. In comparison to furyl derivatives, the thienyl derivatives have larger Stokes shifts on average of about 1000 cm1. This is another evidence for the domination of non-radiative processes in the thienyl based systems. All the photophysical properties determined for thienyl and furyl pyrene derivatives based on fluorescence measure- ments are also summarized in Table 1 and Table 2, respectively. Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 \| 22765 3.3. Quantum-chemical methods 3.3.1. Thienyl derivatives of pyrene. The results of the DFT calculations are presented in Fig. 6 and 7 as well as in Tables 3 and 4. All of the studied molecules are optically active, they absorb light in the UV-violet spectral range and emit in the violet-blue part of the spectrum. Noticeably, the theoretical TDDFT predictions are in good agreement with experimental absorption and fluorescence spectra apart from the under- estimation of the energy gap by 0.35 eV (equivalent to approxi- mately 70 nm). However, this deviation is often encountered in DFT calculations. Nonetheless, the trend along the series K1–K5 is well reproduced in the theoretically derived spectra, including the very close overlapping of K1 and K3 in both absorption and emission spectra. The order of the rising energy gap is K2 o K5 o K1 = K3 o K4. This is clearly connected to the number of thiophene substituents adding their electrons to the molecular conjugate system. Interestingly, the overlap of K1 and K3 spectra indicate that in the case of an isomer mixture the relative position of the substituents does not influence the energy levels of HOMO and LUMO orbitals as the substituents contribute to the molecular orbitals in a similar manner (see also Tables 3 and 4). Fig. 9 TDDFT (bottom panel) and experimental (top) fluorescence spectra of the compounds K6–K10. K6 and K8 in bottom panel closely overlap. HOMO and LUMO orbitals are distributed throughout the whole molecule for all studied compounds (Table 4), which is due to the partially planar structures of both ground and Fig. 9 TDDFT (bottom panel) and experimental (top) fluorescence spectra of the compounds K6–K10. K6 and K8 in bottom panel closely overlap. Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 \| 22765 This journal is ©the Owner Societies 2015 View Article Online PCCP Paper 3.3.2. Furyl derivatives of pyrene. The optical properties of the studied compounds K6–K10 obtained from TDDFT calcula- tions confirm the influence of each additional furan substituent excited states (Table 3). The large overlap between HOMO and LUMO orbitals results in quite eﬃcient absorptions and emissions. 3.4. OLED preparation Fig. 1 shows the layer structure of the OLED device with details on the used organic compounds and the function of each layer. Indium tin oxide (ITO)-coated glass (size 5 5 cm) with a sheet resistance of about 8–12 O sq1 was used as the backing material. ITO deposited on the glass was cleaned by sonication in a detergent solution and then rinsed using de-ionized water. An aqueous solution of PEDOT/PSS (poly(3,4-ethylenedioxy- thiophene)/poly(styrenesulfonate)) (HTL) forming a hole con- ductive layer, was deposited onto ITO. Then water was removed under low pressure in a nitrogen atmosphere. The light-emitting layer (EML) (thienyl and furyl derivatives of pyrene) was deposited Fig. 1 shows the layer structure of the OLED device with details on the used organic compounds and the function of each layer. Indium tin oxide (ITO)-coated glass (size 5 5 cm) with a sheet resistance of about 8–12 O sq1 was used as the backing material. ITO deposited on the glass was cleaned by sonication in a detergent solution and then rinsed using de-ionized water. Fig. 10 Eﬀect of the organic light-emitting diodes made from 1,3,6,8- tetra(2-furyl)-pyrene. 4. Conclusions In this work we have presented photophysical properties and theoretical calculations for two new classes of low molecular weight materials based on a pyrene core substituted either with thiophene or furan units. All studied molecules can absorb light in the UV or violet visible range and emit in the blue-violet spectral region. We observed that the replacement of the thiophene by furan increases the fluorescence quantum yield, most likely due to the more planar structure of the former allowing for a eﬃcient contribution of the p electrons. Furthermore, we described DFT calculations and TDDFT simulations for the presented materials. These quantum-chemical calculations fully support obtained experimental data. HOMO and LUMO orbitals are delocalized uniformly on the pyrene core and its aryl substituents. Theoretical calculations provided values for the HOMO–LUMO gaps for the neutral states of the compounds. Compounds described in this paper fulfill the technological requirements for materials to be successfully applied in the con- struction of organic–electronic devices as demonstrated by the OLED device presented in this paper. In order to verify the practical usefulness of the synthetized materials we built a prototype diode (OLED) using thienyl and furyl derivatives of pyrene and also bithienyl derivatives of triazine. Obtained OLEDs emitted blue light at a voltage of 12 V and a yellow light at a voltage of 24 V. Following the physico-chemical measurements and a successful practical application in OLEDs, the synthesized heterocyclic units can be used as innovative, efficient, cost-effective and environmentally friendly (mercury free) materials for light sources. These units can be flexibly designed and emit a broad spectrum of visible light, when appropriately tuned. The manufacturing of this type of light sources is one of the global research priorities, because it is based on a combination of new organic materials and economic produc- tion methods, thus enabling a variety of innovative products to be designed. Considerable progress in nanotechnology that has been made in recent years has resulted in the launch of the first electronic products made of organic materials. 3.3. Quantum-chemical methods Table 5 Molecular parameters derived from (TD)DFT calculations HOMO [eV] LUMO (=Eg HOMO) [eV] Eg (absorption) [eV nm1] Eg (emission) [eV nm1] Average dihedral angle between the central moiety and the substituents in ground/excited state [deg] K6 5.10 2.32 2.78/446 2.33/531 150/164 K7 4.95 2.53 2.42/512 2.07/599 154/162 K8 5.09 2.29 2.80/443 2.33/531 149/168 K9 5.23 2.14 3.09/402 2.63/471 150/168 K10 5.02 2.43 2.59/478 2.18/568 151/162 Table 6 Graphic representation of HOMO and LUMO orbitals in the studied molecules in the ground and excited states Compound HOMO LUMO HOMO – excited state LUMO – excited state K6 K7 K8 K9 K10 Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Table 5 Molecular parameters derived from (TD)DFT calculations Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 22766 \| Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 This journal is ©the Owner Societies 2015 / / / K8 5.09 2.29 2.80/443 2.33/531 149/168 K9 5.23 2.14 3.09/402 2.63/471 150/168 K10 5.02 2.43 2.59/478 2.18/568 151/162 Table 6 Graphic representation of HOMO and LUMO orbitals in the studied molecules in the ground and excited states Compound HOMO LUMO HOMO – excited state LUMO – excited state K6 K7 K8 K9 K10 Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:5 This article is licensed under a Creative Commons Attribution 3.0 Unport 22766 \| Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 This journal is ©the Owner Societies 2015 Paper on the cathode, which was an aluminum plate of size 0.5 0.5 cm. Plates were superimposed onto each other. Then, the device was connected to a DC power source and a 24 V voltage was passed through it to give a current equal to 0.2 A. A bright light with a color characteristic for the corresponding pyrene derivatives was observed under such conditions. Fig. 10 displays the working diode prepared using 1,3,6,8-tetra(2-furyl)-pyrene as an active emitting layer. on the absorption and emission properties (Fig. 8 and 9, Table 5). The TDDFT-predicted UV-VIS absorption spectra presented in Fig. 8 are in very good agreement with experimental data apart from the underestimation of the energy gap by approximately 0.4 eV commonly encountered in such calculations. 3.3. Quantum-chemical methods Nevertheless, the relative spectral positions of the lowest energy absorption peak, attributed by the TDDFT calculations to the HOMO–LUMO transition, are reproduced well in the calculation. The order of the rise in the energy gap is K7 o K10 o K6 = K8 o K9 and this is also found in the emission spectra, both theoretically derived and experimentally observed (Fig. 9). It is quite clear that the HOMO– LUMO energy gap and the related optical properties depend on the number of furene substituents attached to the pyrene molecule. Interestingly, the spectra measured for the K6 and K8 compounds with two furyl units in trans- and cis- positions respectively, are nearly overlapping. Apparently, the contribution of the furyl substituents to the crucial HOMO and LUMO orbitals is independent of their position in the compounds, as can be seen in Table 6. on the cathode, which was an aluminum plate of size 0.5 0.5 cm. Plates were superimposed onto each other. Then, the device was connected to a DC power source and a 24 V voltage was passed through it to give a current equal to 0.2 A. A bright light with a color characteristic for the corresponding pyrene derivatives was observed under such conditions. Fig. 10 displays the working diode prepared using 1,3,6,8-tetra(2-furyl)-pyrene as an active emitting layer. Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 28 C. J. Tonzola, A. P. Kulkarni, A. P. Giﬀord, W. Kaminsky and S. A. Jenekhe, Adv. Funct. Mater., 2007, 17, 863–874. 2 Y. Sun, N. C. Giebink, H. Kanno, B. Ma, M. E. Thompson and S. R. Forrest, Nature, 2006, 440, 908–912. 29 L. Duan, L. Hou, T. W. Lee, J. Qiao, D. Zhang, G. Dong, L. Wang and Y. J. Qiu, J. Mater. Chem., 2010, 20, 6392–6407. 3 J. M. Hancock, A. P. Giﬀord, Y. Zhu, Y. Lou and S. A. Jenekhe, Chem. Mater., 2006, 18, 4924–4932. 30 C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett., 1987, 51, 913–915. 4 P. L. Burn, S. C. Lo and D. W. Samuel, Adv. Mater., 2007, 19, 1675–1688. 31 P. L. Burn, S. C. Lo and D. W. Samuel, Adv. Mater., 2007, 19, 1675–1688. 5 Y. Y. Lyu, O. Kwak, J. Kwon, S. H. Lee, D. Kim, C. Lee and K. Char, Adv. Mater., 2008, 20, 2720–2729. 32 S. Tao, Y. Zhou, C. S. Lee, X. Zhang and S. T. Lee, Chem. Mater., 2010, 22, 2138–2141. 6 C. J. Tonzola, A. P. Kulkarni, A. P. Giﬀord, W. Kaminsky and S. A. Jenekhe, Adv. Funct. Mater., 2007, 17, 863–874. 33 L. Pu, Chem. Rev., 2004, 104, 1687. 34 R. Martinez-Manez and F. Sancenon, Chem. Rev., 2003, 103, 4419. 7 S. Tao, Y. Zhou, C. S. Lee, X. Zhang and S. T. Lee, Chem. Mater., 2010, 22, 2138–2141. 35 H. Yuasa, N. Miyagawa, T. Izumi, M. Nakatani, M. Izumi and H. Hashimoto, Org. Lett., 2004, 6, 1489. 8 L. Duan, L. Hou, T. W. Lee, J. Qiao, D. Zhang, G. Dong, L. Wang and Y. J. Qiu, J. Mater. Chem., 2010, 20, 6392–6407. 36 H. Abe, Y. Mawatari, H. Teraoka, K. Fujimoto and M. Inouye, J. Org. Chem., 2004, 69, 495. 9 X. Zhan, Z. A. Tan, B. Domercq, Z. An, X. Zhang, S. Barlow, Y. Li, D. Zhu, B. Kippelen and S. R. Marder, J. Am. Chem. Soc., 2007, 129, 7246–7247. 37 Y. Hassheider, S. A. Benning, H. S. Kitzerow, M. F. Achard and H. Bock, Angew. Chem., Int. Ed., 2001, 40, 2060. 10 A. B. F. Martinson, T. W. Hamann, M. J. Pellin and J. T. Hupp, Chem. – Eur. J., 2008, 14, 4458–4467. 38 V. De Halleux, J. P. Calbert, P. Brocorens, J. Cornil, J. P. Declercq, J. L. Bre´das and Y. Geerts, Adv. Funct. Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Mater., 2004, 14, 649. 11 M. Velusamy, J.-H. Huang, Y.-C. Hsu, H.-H. Chou, K.-C. Ho, P.-L. Wu, W.-H. Chang, J. T. Lin and C.-W. Chu, Org. Lett., 2009, 11, 4898–4901. 39 G. Jones and V. I. Vullev, Org. Lett., 2002, 4, 4001. 12 Z. Ning and H. Tian, Chem. Commun., 2009, 5483–5495. 40 J. Daub, R. Engl, J. Kurzawa, S. E. Miller, S. Schneider, A. Stockmann and M. R. Wasielewski, J. Phys. Chem. A, 2001, 105, 5655. 13 A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo and H. Pettersson, Chem. Rev., 2010, 110, 6595–6663. 41 M. C. Jobanputra, M. F. Durstock and S. F. Clarson, J. Appl. Polym. Sci., 2003, 87, 523. 14 J. N. Cliﬀord, E. Martı´nez-Ferrero, A. Viterisi and E. Palomares, Chem. Soc. Rev., 2011, 40, 1635–1646. 15 X. W. Zhan, A. Facchetti, S. Barlow, T. J. Marks, M. A. Ratner, M. R. Wasielewski and S. R. Marder, Adv. Mater., 2011, 23, 268–284. 42 L. T. Cheng, W. Tam, S. H. Stevenson, G. R. Meredith and G. Rikken, J. Phys. Chem., 1991, 95, 10643. 43 A. K.-Y. Jen, Y. Cai, P. V. Bedworth and S. R. Marder, Adv. Mater., 1997, 9, 132. 16 V. C. Sundar, J. Zaumseil, V. Podzorov, E. Menard, R. L. Willett, T. Someya, M. E. Gershenson and J. A. Rogers, Science, 2004, 303, 1644–1647. 44 B. R. Cho, K. N. Son, S. J. Lee and T. J. Kang, Tetrahedron Lett., 1998, 39, 3167. 45 S. S. P. Chom and C. H. Shen, Tetrahedron Lett., 1997, 38, 6407. 17 Y. M. Sun, Y. Q. Liu and D. B. Zhu, J. Mater. Chem., 2005, 15, 53–65. 46 H. Sirringhaus, Adv. Mater., 2005, 17, 2411. 18 M. Mas-Torrent and C. Rovira, J. Mater. Chem., 2006, 16, 433–436. 47 A. Mishra, C. Q. Ma and P. Ba¨uerle, Chem. Rev., 2009, 109, 1141. 19 B. J. Jung, N. J. Tremblay, M.-L. Yeh and H. E. Katz, Chem. Mater., 2011, 23, 568–582. 48 W. Ma, C. Yang, X. Gong and K. Lee, Adv. Funct. Mater., 2005, 15, 1617. 20 K. Niimi, S. Shinamura, I. Osaka, E. Miyazaki and K. Takimiya, J. Am. Chem. Soc., 2011, 133, 8732–8739. 49 F. Padinger, R. S. Rittberger and N. S. Sariciftci, Adv. Funct. Mater., 2003, 13, 85. 21 F. Terenziani, A. Painelli, C. Katan, M. Charlot and M. Blanchard-Desce, J. Am. Chem. Soc., 2006, 128, 15742–15755. 50 Y. Shirota, J. Mater. Acknowledgements This work was supported within the research grant no 2011/01/N/ ST4/02251 from National Science Centre, Poland. Calculations have Fig. 10 Eﬀect of the organic light-emitting diodes made from 1,3,6,8- tetra(2-furyl)-pyrene. Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 \| 22767 This journal is ©the Owner Societies 2015 View Article Online PCCP Paper 24 M. Nakano, R. Kishi, K. Yoneda, Y. Inoue, T. Inui, Y. Shigeta, T. Kubo and B. Champagne, J. Phys. Chem. A, 2011, 115, 8767–8777. been carried out in Wroclaw Centre for Networking and Super- computing (http://www.wcss.wroc.pl, grant No.135). The presented study was partially funded by the German Federal Ministry of Environment (promotional reference No. 0325417, Reaktherm). 25 Y. Sun, N. C. Giebink, H. Kanno, B. Ma, M. E. Thompson and S R. Forrest, Nature, 2006, 440, 908–912. 26 J. M. Hancock, A. P. Giﬀord, Y. Zhu, Y. Lou and S. A. Jenekhe, Chem. Mater., 2006, 18, 4924–4932. Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 27 Y. Y. Lyu, J. Kwak, O. Kwon, S. H. Lee, D. Kim, C. Lee and K. Char, Adv. Mater., 2008, 20, 2720–2729. 1 C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett., 1987, 51, 913–915. This journal is ©the Owner Societies 2015 Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence 79 M. S. AlSalhi, J. Alam, L. A. Dass and M. Raja, Int. J. Mol. Sci., 2011, 12, 2036. 58 L. T. Cheng, W. Tam, S. H. Stevenson, G. R. Meredith and G. Rikken, J. Phys. Chem., 1991, 95, 10643. 80 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision B.01, Gaussian, Inc., Wallingford CT, 2010. 59 H. Tsuji, C. Mitsui, L. Ilies, Y. Sato and E. Nakamura, J. Am. Chem. Soc., 2007, 129, 11902. 60 Y. Miyata, M. Terayama, T. Minari, T. Nishinaga, T. Nemoto, S. Isoda and K. Komatsu, Chem. – Asian J., 2007, 2, 1492. J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, 61 J. C. Bijleveld, B. P. Karsten, S. G. J. Mathijssen, M. M. Wienk, D. M. de Leeuw and R. A. J. Janssen, J. Mater. Chem., 2011, 21, 1600. g , p , J J y , , Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, 62 Y. Li, P. Sonar, S. P. Singh, W. Zeng and M. S. Soh, J. Mater. Chem., 2011, 21, 10829. K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, 63 O. Gidron, Y. Diskin-Posner and M. Bendikov, J. Am. Chem. Soc., 2010, 132, 2148. J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Chem., 2000, 10, 1. 22 F. Terenziani, G. D’Avino and A. Painelli, ChemPhysChem, 2007, 8, 2433–2444. 51 P. Liu, Y. Shirota and Y. Osada, Polym. Adv. Technol., 2000, 11, 512. 23 F. Rizzo, M. Cavazzini, S. Righetto, F. De Angelis, S. Fantacci and S. Quici, Eur. J. Org. Chem., 2010, 4004–4016. 52 P. Liu, H. Nakano and Y. Shirota, Liq. Cryst., 2001, 28, 581. 22768 \| Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 This journal is ©the Owner Societies 2015 View Article Online Paper PCCP 53 P. Liu, X. B. Wang, Y. M. Zhang, X. P. Zhou and W. J. Deng, Synth. Met., 2005, 155, 565. 75 Y. Z. Wu, X. Y. Zheng, W. Q. Zhu, R. G. Sun, X. Y. Jiang, Z. L. Zhang and S. H. Xu, Appl. Phys. Lett., 2003, 83, 5077. 54 P. Liu, M. S. Huang, W. Z. Pan, Y. M. Zhang and W. J. Deng, J. Lumin., 2006, 121, 109. 76 L. Duan, L. Hou, T. Lee, J. Qiao, D. Zhang, G. Dong and L. Wang L, Mater. Chem., 2010, 20, 6392. 55 P. Liu, M. M. Liu and W. J. Deng, Synth. Commun., 2006, 36, 685. 77 G.-F. Wang, X.-M. Tao, J. H. Xin and B. Fei, Nanoscale Res. Lett., 2009, 4, 613. 56 P. Liu, Y. G. Wang, W. Z. Pan, M. S. Huang and W. J. Deng, Chin. J. Chem., 2006, 24, 1305. 78 C. h. Zhong, C. h. Duan, F. Huang, H. Wu and Y. Cao, Chem. Mater., 2011, 23, 326. 57 M. C. Jobanputra, M. F. Durstock and S. F. Clarson, J. Appl. Polym. Sci., 2003, 87, 523. Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 \| 22769 Open Access Article. Published on 28 July 2015. Downloaded on 08/12/2015 13:31:58. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, 64 O. Gidron, A. Dadvand, Y. Sheynin, M. Bendikov and D. F. Perepichka, Chem. Commun., 2011, 47, 1976. R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, 65 L. Huo, Y. Huang, B. Fan, X. Guo, Y. Jing, M. Zhang, Y. Li and J. Hou, Chem. Commun., 2012, 48, 3318. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, 66 C. Mitsui, J. Soeda, K. Miwa, H. Tsuji, J. Takeya and E. Nakamura, J. Am. Chem. Soc., 2012, 134, 5448. sion B.01, Gaussian, Inc., Wallingford CT, 2010. 67 M. Jones Jr, Organic Chemistry, 3rd edn, 2005. 81 A.R. Allouche, Gabedit is a free graphical user interface for computational chemistry packages. Available at: http://gabe dit.sourceforge.net/. 68 A. Juric, A. Sabljic and N. Trinajstic, J. Heterocycl. Chem., 1984, 21, 273. 69 Z. Q. Gao, C. S. Lee, I. Bello, S. T. Lee, R. M. Chen, T. Y. Luh, J. Shi and C. W. Tang, Appl. Phys. Lett., 1999, 74, 865. 82 S. Miertus, E. Scrocco and J. Tomasi, J. Chem. Phys., 1981, 55, 117. 83 P. Machata, P. Rapta, V. Lukes, K. R. Idzik, T. Licha, R. Beckert and L. Dunsch, Electrochim. Acta, 2014, 122, 57–65. 70 A. P. Kulkarni, A. P. Giﬀord, C. J. Tonzola and S. A. Jenekhea, Appl. Phys. Lett., 2005, 86, 061106. 71 J. M. Shi and C. W. Tang, Appl. Phys. Lett., 2002, 80, 3201. 84 K. R. Idzik, P. Ledwon, T. Licha, W. Kuznik, M. Lapkowski and J. Frydel, Dyes Pigm., 2014, 103, 55–61. 72 D. Gebeyehu, K. Walzer, G. He, M. Pfeiﬀer, K. Leo, J. Brandt, A. Gerhard, P. Sto¨ssel and H. Vestweber, Synth. Met., 2005, 148, 205. 85 K. R. Idzik, T. Licha, V. Lukes, P. Rapta, J. Frydel, M. Schaﬀer, E. Taeuscher, R. Beckert and L. Dunsch, J. Fluoresc., 2014, 24(1), 153–160. 73 C. Hosokawa, H. Higashi, H. Nakamura and T. Kusumoto, Appl. Phys. Lett., 1995, 67, 3853. 86 P. Ledwon, M. Lapkowski, T. Licha, J. Frydel and K. R. Idzik, Synth. Met., 2014, 191, 74–82. 74 C. H. Liao, M. T. Lee, C. H. Tsai and C. H. Chen, Appl. Phys. Lett., 2005, 86, 203507. Phys. Chem. Chem. Phys., 2015, 17, 22758--22769 \| 22769 This journal is ©the Owner Societies 2015
https://openalex.org/W2749896097	http://www.scielo.br/pdf/aob/v25n4/1413-7852-aob-25-04-00137.pdf	English	null	STRUTHERS’ LIGAMENT AND SUPRACONDYLAR HUMERAL PROCESS: AN ANATOMICAL STUDY AND CLINICAL IMPLICATIONS	Acta Ortopédica Brasileira	2,017	cc-by	5,759	ABSTRACT Objetivo: Determinar a frequência e as características anatômicas do ligamento de Struthers e do processo supracondilar do úmero e avaliar sua implicação clínica na neuropatia compressiva do nervo mediano. Método: Foram dissecados 60 membros superiores de 30 cadáveres de adultos, 26 do sexo masculino e quatro do sexo feminino, 15 previamente preservados em formol e glicerina e 15 dissecados a fresco no Laboratório de Anatomia. A relação do ligamento de Struthers com o nervo mediano e a artéria e veias braquiais, foi documentada com desenhos e fotografias. Resultados: O processo supracondilar do úmero não foi encontrado em nenhum dos 60 braços dissecados. O ligamento de Struthers foi identificado em seis membros (dois bilaterais); em todos havia inserção alta do músculo pronador redondo. Conclusão: O ligamento de Struthers é uma estrutura aponeurótica que pode estar ou não associada ao processo supracondilar do úmero e representa local de possível compressão do nervo mediano no terço inferior do braço. Nível de Evidência IV, Série de Casos. Objective: The objective of this study was to determine the frequency and anatomical characteristics of Struthers’ ligament and the su pracondylar humeral process and evaluate the clinical implications in compressive neuropathy of the median nerve. Method: We dissected 60 arms from 30 cadavers (26 males and 4 females): 15 were previously preserved in formalin and glycerin and 15 were dissected fresh in the Anatomy Laboratory for this paper. The relationships between Struthers’ ligament and the median nerve and brachial artery and veins were documented with drawings and photos. Results: The supracondylar humeral process was not found in any of the 60 dissected arms. Struthers’ ligament was identified in six arms (two bilateral); in all cases high insertion of the pronator teres muscle was observed. Conclusion: Struthers’ ligament is an aponeurotic structure that may or may not be associated with the supracondylar humeral process, and is an important potential site of median nerve compression in the lower third of the arm. Level of Evidence IV, Case Series. Descritores: Ligamentos. Síndromes de compressão nervosa. Nervo mediano. Úmero. Keywords: Ligaments. Nerve compression syndromes. Median nerve. Humerus. STRUTHERS’ LIGAMENT AND SUPRACONDYLAR HUMERAL PROCESS: AN ANATOMICAL STUDY AND CLINICAL IMPLICATIONS Edie Benedito Caetano1, João José Sabongi Neto2, Luiz Angelo Vieira1, Maurício Ferreira Caetano2, José Eduardo de Bona1, Thais Mayor Simonatto1 Edie Benedito Caetano1, João José Sabongi Neto2, Luiz Angelo Vieira1, Maurício Ferreira Caetano2, J Thais Mayor Simonatto1 1. Pontifícia Universidade Católica de São Paulo, Faculdade de Ciências Médicas e da Saúde, Sorocaba, SP, Brazil. 2. Conjunto Hospitalar de Sorocaba, Department of Hand Surgery, Sorocaba, SP, Brazil. Citation: Caetano EB, Sabongi Neto JJ, Vieira LA, Caetano MF, Bona JE, Simonatto TM. Struthers' ligament and supracondylar humeral process: an anatomical study and clinical implications. Acta Ortop Bras. [online]. 2017;25(4):137-42. Available from URL: http://www.scielo.br/aob. Work conducted at the Pontifícia Universidade Católica de São Paulo, Faculdade de Ciências Médicas e da Saúde, Sorocaba, SP, Brazil. Correspondence: Pontifícia Universidade Católica de São Paulo, Faculdade de Ciências Médicas e da Saúde. PUC-SP/FCMS. Rua Joubert Wey, 290. Sorocaba, SP, Brazil. 18030-070. ediecaetano@uol.com.br Original Article DOI: http://dx.doi.org/10.1590/1413-785220172504168330 Original Article DOI: http://dx.doi.org/10.1590/1413-785220172504168330 idade Católica de São Paulo, Faculdade de Ciências Médicas e da Saúde, Sorocaba, SP, Brazil. Católica de São Paulo, Faculdade de Ciências Médicas e da Saúde. PUC-SP/FCMS. Rua Joubert Wey, 290. Sorocaba, SP, Brazil. 18030-070. INTRODUCTION phylogenetically considered to be a remnant of the supracondylar foramen found in reptiles, marsupials, and some mammals.2,3 Kessel and Rang4 consider that from the embryonic point of view, Struthers’ ligament is a remaining vestige of the tendon of the latissimus-condyloid muscle tendon, which is found in some climbing animals and serves as an anchor for the pronator teres muscle. In the lower mammals, the tunnel of osteo-fibrous tissue formed by the humerus, the supracondylar process and Struthers’ ligament protects the nerves and blood vessels that extend to the forearm.4 Its occurrence in humans is very rare, in only 0.7–2.5% of the population.4-6 It is more frequent in women and Europeans and Struthers’ ligament was described by the anatomist John Struthers1 in 1848; this fibrous band extends from a bone spike located on the anteromedial face of the lower third of the humerus known as the supracondylar process and is part of the medial epicondyle of the humerus. Struthers’ ligament passes over the median nerve and the brachial artery, and can cause compression of these structures. This ligament may be present even when the supracondylar process is absent, and even when it is present may not cause the compression of these structures. The supracondylar process of the humerus has been described by anatomists and anthropologists and is All authors declare no potential conflict of interest related to this article. Article received in 08/24/2016, approved in 01/24/2017. 137 Acta Ortop Bras. 2017;25(4):137-42 is extremely rare in Black individuals.7 Some authors have reported familial occurrence of this bone spike.5,6 is extremely rare in Black individuals.7 Some authors have reported familial occurrence of this bone spike.5,6 ranged from 28 to 77 years, 17 were white and 13 were non-white. Forearms deformed by trauma, congenital malformations, and scars were excluded. The dissection was performed through a midline incision in the arm and forearm, and two flaps including the skin and subcutaneous tissue were folded back to the radial and ulnar sides, respectively. This same process was repeated for the fascia of the arm and forearm, exposing all the musculature. The median nerve was identified in the proximal third of the arm in the medial margin of the brachial biceps muscle and dissected distally, analyzing the presence of any fibrous bands, Struthers’ ligament, and the supracondylar process of the humerus that could narrow its passage. RESULTS In all 60 dissected arms we recorded that in the middle third of the arm, the median nerve crossed in front of the brachial artery in a lateral-to-medial direction and proceeded toward the cubital fossa, where it was positioned medially to the brachial artery and the tendon of the biceps brachii muscle. The supracondylar process of the humerus was not found in any of the 60 dissected arms. Struthers’ ligament was identified in 6 arms (two bilateral) and in all cases the pronator teres muscle had a high insertion. In the right arm of one cadaver we identified high insertion of the pronator teres muscle where it came from a cord-shaped ligament positioned on the median nerve and brachial artery, inserting into the diaphysis of the humerus. (Figure 1A) On the left side the ligament originated from the same place and was inserted proximally, but had no relation with the median nerve and brachial artery. (Figure 1B) Unlike Struthers’ ligament, the arcade of Struthers is an aponeurotic or musculoaponeurotic structure which extends from the medial intermuscular septum to the medial head of the triceps brachii muscle. (Figure 2A) A case similar to the situation in Figure 1B was recorded in the left arm of another cadaver. (Figure 2B) In two arms from one cadaver, we identified an anatomical variation consisting of high insertion of the humeral head of the pronator teres muscle which was inserted through a short ligament in the diaphysis of the humerus causing pressure on the median nerve and brachial artery, but there was no bone spike. (Figures 3A and 3B) We identified two similar cases on the right arm of a recently-deceased cadaver and in the left arm of another cadaver preserved in formaldehyde and glycerin; Struthers’ ligament was composed of a fibrous lamina originating in the medial epicondyle and adjacent brachial fascia, moving upwards and passing over the median nerve and brachial artery to insert itself into the fascia of the brachial muscle and the humeral shaft. (Figures 4A and 4B) INTRODUCTION The dissection continued distally on the forearm where we analyzed the presence of nerve compression from the bicipital aponeurosis, between the humeral and ulnar heads of the pronator teres muscle and through the archway formed between the radial and ulnar humerus inserts of the su perficial flexor muscle, and also identified the Gantzer muscle, the Martin-Gruber anastomosis, and possible anatomical variations. These are part of studies that have already been published or are forthcoming. The anatomical variations were identified, recorded, and photographed. We used a Keeler 2.5x magnifying glass. The study was approved by the institutional review board under process number 1,611,295. Struthers’ ligament and the arcade of Struthers are two different anatomical structures which are often confused. The arcade of Struthers was first described in 1973 by Kane et al.8 and has some times been defined as a thickening of the brachial fascia, and at other times as an aponeurotic or musculoaponeurotic structure which extends from the medial intermuscular septum to the medial head of the triceps brachii muscle at a variable distance above the medial epicondyle of the humerus. The arcade of Struthers can cause compression of the ulnar nerve. Compression of the median nerve in the elbow is usually caused by the presence of fibrous bands, which may be seen in four different anatomic locations in the following order of frequency:9 between the superficial and deep heads of the pronator teres muscle, in the arcade formed by the proximal insertions of the superficial flexor muscle, in the bicipital aponeurosis (lacertus fibrosus), and in Struthers’ ligament which may or may not be associated with the supracondylar process of the humerus. Clinically, it is not easy to identify the exact location of compression. The most common cause of median nerve compression syndrome in this region occurs between the humeral and ulnar heads of the pronator teres muscle. However, the decreased muscle strength of the pronator teres suggests compression above the elbow. Some provocative tests can be used to differentiate the location of the nerve compression:9,10 1. Pronation of the forearm against resistance with the elbow flexed and then gradually extended indicates compression between the two heads of the pronator teres muscle. INTRODUCTION To test the pronator teres muscle alone, the elbow must be extended and supported on a flat surface, with the patient prone and the forearm against resistance with the arm in neutral rotation and the wrist flexed (to relax the superficial muscular flexor of the fingers). Pain and paresthesia show the involvement of the pronator teres muscle in compression of the median nerve. 2. Independent flexion of the middle finger against resistance reproduces the paresthetic symptoms in the area where the median nerve innervates indicates compression in the arch formed between the proximal insertions of the superficial flexor muscle. 3. Elbow flexion and supination of the forearm against resistance reproduces the symptoms and indicates compression by the bicipital aponeurosis (lacertus fibrosus). 4. Compression by Struthers’ ligament is usually associated with pain in the forearm that is accentuated during extension of the wrist. The elbow is flexed against resistance and simultaneously palpating the area 5 to 10 cm above the medial epicondyle in an attempt to palpate the supracondylar process of the humerus. Radiological examination rules out the presence of the supracondylar process. Tinel’s sign can be useful to find the location of the compression. The results of electrophysiological examinations are consistent with a nerve compression at the elbow, suggesting but not confirming the exact location of the compression.9 Only surgical exploration of the nerve can identify the structure responsible for nerve compression.9,10 The objective of this study was to analyze 60 limbs from 30 cadavers in order to identify the presence of Struthers’ ligament and the supracondylar process of the humerus, along with the possibility that these anatomical variations may be responsible for compression of the median nerve. Acta Ortop Bras. 2017;25(4):137-42 DISCUSSION A) In the right arm of one corpse, we identified high insertion of the pronator teres muscle (a) from which a cord-shaped ligament originated (b) and was positioned on the median nerve (c) and the brachial artery (d) inserting into the diaphysis of the humerus. B) On the left side, the ligament originated in the same place and was inserted proximally but had no relation to the median nerve or brachial artery. The pronator teres muscle (a); Struthers’ ligament (b); median nerve (c); brachial artery (d). of the lower third of the left and right arms; it was more extensive on the right side with positive Tinel’s sign at the level of the tumor. On this side there was a limitation of 35 degrees extension and 30 degrees supination. Pain was exacerbated with passive supination and active pronation with simultaneous extension of the elbow. There were no signs of vascular compression. The left arm had a limitation of 15 degrees of extension and 20 of supination, and negative Tinel’s sign with no symptoms of neurovascular compression. Radiological examination showed the presence of a 3-cm bone spike in the right arm and a 1.5-cm bone spike in the left arm. Surgery showed that the pronator teres muscle was inserted abnormally via a short ligament in the supracondylar process of the humerus. (Figure 5A) Disinsertion of the muscle and resection of the bone spike resulted in decompression of the median nerve. (Figure 5B) Clinical improvement was already evident in the second month after surgery. After 18 months, the patient was completely asymptomatic. The pronator teres muscle was reinserted in the medial epicondyle of the humerus. In our dissections we identified in both arms from a single cadaver an anatomical variation identical to that registered in the clinical case described above. The pronator teres muscle was inserted the same way via a short ligament in the diaphysis of the humerus, but there was no bone spike. (Figures 4A and B) We do not know the medical history of this individual, but the narrowing of the space where the nerve passed was evident. Aydinlioglu et al.16 described the rare case of a 21-year-old woman who complained of pain, sensory disturbances, and loss of motor function in the area where the median nerve innervated in both arms. Clinical examination identified the presence of a painful bone spur in the distal third of the humerus. DISCUSSION We dissected 60 forearms of 30 adult cadavers belonging to the Anatomy Department Laboratory to conduct this study; 26 corpses were male and four were female, 15 had previously been preserved in formaldehyde and glycerin and 15 were fresh cadavers. Ages Nerve compression in the elbow region is generally called pronator teres syndrome because compression most frequently occurs between the two heads of this muscle.9 Tubbs et al.11 considered this nomenclature to be incorrect when compression occurs 138 Acta Ortop Bras. 2017;25(4):137-42 Figure 2. A) The arcade of Struthers (a) is a musculoaponeurotic structure which extends from the medial intermuscular septum to the medial head of the triceps brachii muscle (b). It is positioned anterior to the ulnar nerve (c) and may compress this structure. B) A case similar to the situation in Figure 1B was recorded in the left arm of another cadaver. Pronator teres muscle (a). Struthers’ ligament (b). Median nerve (c). Brachial artery (d). Figure 1. A) In the right arm of one corpse, we identified high insertion of the pronator teres muscle (a) from which a cord-shaped ligament originated Figure 2. A) The arcade of Struthers (a) is a musculoaponeurotic structure which extends from the medial intermuscular septum to the medial head of the triceps brachii muscle (b). It is positioned anterior to the ulnar nerve (c) and may compress this structure. B) A case similar to the situation in Figure Figure 2. A) The arcade of Struthers (a) is a musculoaponeurotic structure which extends from the medial intermuscular septum to the medial head of the triceps brachii muscle (b). It is positioned anterior to the ulnar nerve (c) and may compress this structure. B) A case similar to the situation in Figure 1B was recorded in the left arm of another cadaver. Pronator teres muscle (a). Struthers’ ligament (b). Median nerve (c). Brachial artery (d). Figure 2. A) The arcade of Struthers (a) is a musculoaponeurotic structure which extends from the medial intermuscular septum to the medial head of the triceps brachii muscle (b). It is positioned anterior to the ulnar nerve (c) and may compress this structure. B) A case similar to the situation in Figure 1B was recorded in the left arm of another cadaver. Pronator teres muscle (a). Struthers’ ligament (b). Median nerve (c). Brachial artery (d). Figure 1. Acta Ortop Bras. 2017;25(4):137-42 DISCUSSION Radiological and electrophysiological through Struthers’ ligament, bicipital aponeurosis, or through the archway of the superficial flexor, and suggested that the correct name would be compressive neuropathies proximal to the median nerve and not pronator teres syndrome. Struthers’ ligament and supracondylar process of the humerus can be asymptomatic and differ from the osteocartilagenous exostosis because they do not have a cartilaginous cap and histologically are normal bone in continuity with the humeral cortex.6 The cartilaginous cap can be visualized using magnetic resonance imaging.6,12 Clinical abnormal ities caused by the supracondylar process of the humerus were first described by Soliere,13 who described sensory and motor changes in the median nerve which was compressed by the supracondylar process of the humerus in a 19-year-old man. Suranyi14 reported the case of a 60-year-old patient exhibiting progressive weakness, pain, and numbness in the left forearm and hand. Clinical examination showed that these changes occurred in the area of distribution of the median nerve. Electrophysiological examination showed impairment of the nerve in the segment immediately proximal to the elbow. Surgery revealed that Struthers’ ligament was the cause of the compression of the median nerve. The bone spike was not found via palpation or x-ray examination. Sectioning of the Struthers’ ligament relieved the symptoms. Caetano and Brandi15 reported the case of a 26-year-old patient with pain in the shoulder, forearm, and hand as well as numbness of the hand for six months; he was unable to fully extend his left or right elbows. During palpation the presence of a hard and painful tumor was noted on the antero-medial surface through Struthers’ ligament, bicipital aponeurosis, or through the archway of the superficial flexor, and suggested that the correct name would be compressive neuropathies proximal to the median nerve and not pronator teres syndrome. Struthers’ ligament and supracondylar process of the humerus can be asymptomatic and differ from the osteocartilagenous exostosis because they do not have a cartilaginous cap and histologically are normal bone in continuity with the humeral cortex.6 The cartilaginous cap can be visualized using magnetic resonance imaging.6,12 Clinical abnormal ities caused by the supracondylar process of the humerus were first described by Soliere,13 who described sensory and motor changes in the median nerve which was compressed by the supracondylar process of the humerus in a 19-year-old man. DISCUSSION Suranyi14 reported the case of a 60-year-old patient exhibiting progressive weakness, pain, and numbness in the left forearm and hand. Clinical examination showed that these changes occurred in the area of distribution of the median nerve. Electrophysiological examination showed impairment of the nerve in the segment immediately proximal to the elbow. Surgery revealed that Struthers’ ligament was the cause of the compression of the median nerve. The bone spike was not found via palpation or x-ray examination. Sectioning of the Struthers’ ligament relieved the symptoms. Caetano and Brandi15 reported the case of a 26-year-old patient with pain in the shoulder, forearm, and hand as well as numbness of the hand for six months; he was unable to fully extend his left or right elbows. During palpation the presence of a hard and painful tumor was noted on the antero-medial surface 139 Acta Ortop Bras. 2017;25(4):137-42 Figure 3. A-B) In one cadaver, we identified a bilateral anatomical variation consisting of high insertion of the humeral head of the pronator teres muscle (a) which was inserted through a short ligament (b) in the diaphysis of the humerus causing pressure on the median nerve (c) and brachial artery (d), but no bone spike was seen. Figure 3. A-B) In one cadaver, we identified a bilateral anatomical variation consisting of high insertion of the humeral head of the pronator teres muscle (a) which was inserted through a short ligament (b) in the diaphysis of the humerus causing pressure on the median nerve (c) and brachial artery (d), but no bone spike was seen. Figure 3. A-B) In one cadaver, we identified a bilateral anatomical variation consisting of high insertion of the humeral head of the pronator teres muscle (a) which was inserted through a short ligament (b) in the diaphysis of the humerus causing pressure on the median nerve (c) and brachial artery (d), but no bone spike was seen. Figure 4. We identified two similar cases (A) in the right arm of a recent ly-deceased cadaver and (B) in the left arm of another cadaver preserved in formaldehyde and glycerin; the Struthers’ ligament was composed of a Figure 3. DISCUSSION A-B) In one cadaver, we identified a bilateral anatomical variation consisting of high insertion of the humeral head of the pronator teres muscle (a) which was inserted through a short ligament (b) in the diaphysis of the humerus causing pressure on the median nerve (c) and brachial artery (d), but no bone spike was seen. Figure 4. We identified two similar cases (A) in the right arm of a recent ly-deceased cadaver and (B) in the left arm of another cadaver preserved in formaldehyde and glycerin; the Struthers’ ligament was composed of a fibrous lamina (b) originating in the medial epicondyle and adjacent brachial fascia, moving upwards and passing over the median nerve (c) and brachial artery (d) to insert itself into the fascia of the brachial muscle and the humeral shaft. Pronator teres (a). Figure 4. We identified two similar cases (A) in the right arm of a recent ly-deceased cadaver and (B) in the left arm of another cadaver preserved in formaldehyde and glycerin; the Struthers’ ligament was composed of a fibrous lamina (b) originating in the medial epicondyle and adjacent brachial fascia, moving upwards and passing over the median nerve (c) and brachial artery (d) to insert itself into the fascia of the brachial muscle and the humeral shaft. Pronator teres (a). examination confirmed the diagnosis of bilateral compression of the median nerve caused by Struthers’ ligament. These authors stated that this was the first reported case of bilateral compression of the median nerve caused by this ligament. The patient underwent surgical decompression of the nerve on both sides and symptoms were relieved after two weeks. These authors described the im portance of removing the adjacent periosteum to avoid regrowth in the supracondylar process. Lordan et al.17 reported the case of a 13-year-old boy with a history of 4 weeks of vague pain in the left forearm after hitting his elbow during sports activity. Palpation identified a hard mass along the distal humerus. X-ray revealed a bone spike 5 cm proximal to the elbow. The clinical history showed the boy had paresthesia in the ipsilateral thumb, index finger, middle and radial half of the ring finger, as well as pain in the forearm during pronossupination. Physical examination revealed mild weakness in grip strength, but no atrophy of the thenar muscles or sensory deficits of the hand. Because the neurological symptoms were persistent, the patient underwent surgery. Acta Ortop Bras. 2017;25(4):137-42 AUTHORS’ CONTRIBUTIONS: Each author made significant individual contributions to this manuscript. EBC (0000-0003-4572-3854), LAV (0000-0003- 4406-2492), and MBFC (0000-0003-0994-2128)* were the main contributors in drafting the manuscript. JEB (0000-0002-2416-7281)* and TMS (0000-0002- 0654-5373)* conducted the bibliographical research. All authors participated in dissecting and photographing the cadavers. EBC and JJSN (0000-0002- 0554-1426)* reviewed the manuscript, contributed to the intellectual concept, coordinated the study, and were responsible for directing the other participants. *ORCID (Open Researcher and Contributor ID). DISCUSSION In one cadaver, we identified a bilateral anatomical variation consisting of high insertion of the humeral head of the pronator teres muscle (a) which inserted via a short ligament (b) in the supracondylar process of the humerus, compressing the median nerve (c). B) Surgery showing the pronator teres muscle disinserted (a) from the supracondylar process of the humerus (b) and compressing the median nerve (c). Detail: supracondylar process (a) and median nerve (b). Figure 5. A) Clinical case. In one cadaver, we identified a bilateral anatomical variation consisting of high insertion of the humeral head of the pronator teres muscle (a) which inserted via a short ligament (b) in the supracondylar process of the humerus, compressing the median nerve (c). B) Surgery showing the pronator teres muscle disinserted (a) from the supracondylar process of the humerus (b) and compressing the median nerve (c). Detail: supracondylar process (a) and median nerve (b). Figure 6. A) High insertion of the pronator teres muscle (a) associated with high origin of the anterior interosseous nerve (above the elbow joint) (b). Median nerve (c). B) The Struthers’ ligament (a) originated in the medial epicondyle (b) and moved proximally, fanning out and inserting into the brachial fascia and the internal brachial ligament and passing over the ulnar nerve (c) in precisely the place where the nerve passed from the anterior compartment to the posterior of the arm. Figure 6. A) High insertion of the pronator teres muscle (a) associated with high origin of the anterior interosseous nerve (above the elbow joint) (b). Median nerve (c). B) The Struthers’ ligament (a) originated in the medial epicondyle (b) and moved proximally, fanning out and inserting into the brachial fascia and the internal brachial ligament and passing over the ulnar nerve (c) in precisely the place where the nerve passed from the anterior compartment to the posterior of the arm. or high division of the brachial artery in this cadaver. The proximity of the ligament and the supracondylar process to the brachial artery and veins can cause symptoms resulting from the compression of these structures, with ischemic episodes of pain and changes in the arterial pulses which have been previously described.4,6 Some authors4,23 have indicated that Struthers’ ligament can compress the ulnar nevus, but consider this phenomenon to be very rare. DISCUSSION We did identify a case not belonging to this series (but rather in a demonstration of access routes to the elbow) in which Struthers’ ligament originated in the medial epicondyle and moved proximally, fanning out and inserting into the brachial fascia and the internal brachial ligament and passing over the ulnar nerve in precisely the place where the nerve passed from the anterior compartment to the posterior of the arm. (Figure 6B) Gessini et al.24 considered compression of the median nerve by Struthers’ ligament to be very rare, and reported that in a series of 228 patients with compressive syndromes of the median nerve, only three cases occurred above the elbow: one case involving Struthers’ ligament and two involving the bicipital aponeurosis, 201 cases of carpal tunnel syndrome, 21 cases involving the pronator teres muscle, and three involving the anterior interosseous nerve. DISCUSSION Struthers’ ligament and supracondylar process were identified as responsible for the symptoms. The ligament was sectioned and the supracondylar process removed, and the symptoms consequently disappeared. These authors also stated that compression of the median nerve by Struthers’ ligament and the supracondylar process should be considered in cases where symptoms persist after decompression of the median nerve in the carpal tunnel. Petret et al.18 presented the case of a professional tennis player with a stress fracture in the supracondylar process of the humerus who underwent surgery to avoid possible displacements and neurovascular complications. These authors state that this was the first report of a stress fracture in the supracondylar process, and believed that the excessive traction of the pronator teres caused the fracture, which was seen in both X-ray and MR imaging. Jelev et al.19 reported noticing an unusually high insertion for the pronator teres muscle during routine anatomic dissection of the right arm of a 53-year-old female cadaver; this insertion had two bone origins, one in the medial epicondyle and a smaller insertion in the supracondylar process of the humerus, with tendinous arch (Struthers’ ligament) extending between them with the median nerve and brachial vessels passing through this arch. They no ticed that the musculocutaneous nerve was absent and that the coracobrachialis, brachialis, and biceps brachialis muscles received innervation from the median nerve. The anatomical relationships between the supracondylar spur and Struthers’ ligament and the neighboring neurovascular structures have been clearly demonstrated in MR imaging by some authors.20,21 Other associated anatomical variations may occur: high insertion of the pronator teres, high division of the brachial artery, low insertion of the coracobrachialis muscle, or high origin of the anterior interos seous nerve.22 In the left arm of a cadaver with a high origin for the pronator teres muscle, we noted low insertion of the coracobrachialis muscle and high origin of the anterior interosseous nerve, (Figure 6A) but did not identify Struthers’ ligament, the supracondylar process, 140 Acta Ortop Bras. 2017;25(4):137-42 or high division of the brachial artery in this cadaver. DISCUSSION The proximity of the ligament and the supracondylar process to the brachial artery and veins can cause symptoms resulting from the compression of these structures, with ischemic episodes of pain and changes in the arterial pulses which have been previously described.4,6 Some authors4,23 have indicated that Struthers’ ligament can compress the ulnar nevus, but consider this phenomenon to be very rare. We did identify a case not belonging to this series (but rather in a demonstration of access routes to the elbow) in which Struthers’ ligament originated in the medial epicondyle and moved proximally, fanning out and inserting into the brachial fascia and the internal brachial ligament and passing over the ulnar nerve in precisely the place where the nerve passed from the anterior compartment to the posterior of the arm (Figure 6B) Gessini et al 24 considered the elbow: one case involving Struthers’ ligament and two involving the bicipital aponeurosis, 201 cases of carpal tunnel syndrome, 21 cases involving the pronator teres muscle, and three involving the anterior interosseous nerve. CONCLUSION Struthers’ ligament is a rare aponeurotic structure that may or may not be associated with the supracondylar process of the humerus and may compress the median nerve against the deeper structures, changing the normal course of the nerve; it is consequently one Figure 5. A) Clinical case. In one cadaver, we identified a bilateral anatomical variation consisting of high insertion of the humeral head of the pronator teres muscle (a) which inserted via a short ligament (b) in the supracondylar process of the humerus, compressing the median nerve (c). B) Surgery showing the pronator teres muscle disinserted (a) from the supracondylar process of the humerus (b) and compressing the median nerve (c). Detail: supracondylar process (a) and median nerve (b). Figure 6. A) High insertion of the pronator teres muscle (a) associated with high origin of the anterior interosseous nerve (above the elbow joint) (b). Median nerve (c). B) The Struthers’ ligament (a) originated in the medial epicondyle (b) and moved proximally, fanning out and inserting into the brachial fascia and the internal brachial ligament and passing over the ulnar nerve (c) in precisely the place where the nerve passed from the anterior compartment to the posterior of the arm. Figure 5. A) Clinical case. Acta Ortop Bras. 2017;25(4):137-42 REFERENCES 14. Suranyi L. Median nerve compression by Struthers ligament. J Neurol Neurosurg Psychiatry. 1983;46(11):1047-9. 1. Struthers J. On a peculiarity of the humerus and humeral artery. Mon J Med Sci. 1848;28:264-7. 15. Caetano EB, Brandi S. Compressão do nervo mediano por processo supra condilar do úmero. Rev Bras Ortop. 1989;24(9):323-6. 2. Dwight T. A bony supracondyloid foramen in man. Am J Anat 3. Kolb LW, Moore RD. Fracture of the supracondyloid process of the humerus. J Bone Joint Surg Am. 1967;49:532-8. 16. Aydinlioglu A, Cirak B, Akpinar F, Tosun N, Dogan A. Bilateral median nerve compression at the level of Struthers' ligament. Case report. J Neurosurg. 2000;92(4):693-6. 4. Kessel L, Rang M. Supracondilar spur of the humerus. J Bone Joint Surg Br. 1966;48(4):765-8. 17. Lordan J, Rauh P, Spinner RJ. The clinical anatomy of the supracondylar spur and the ligament of Struthers. Clin Anat. 2005;18(7):548-51. 5. Barnard LB, McCoy SM. The supra condyloid process of the humerus. J Bone Joint Surg Am. 1946;28(4):845-50. 18. Pedret C, Balius R, Alomar X, Vilaró J, Ruiz-Cotorro A, Minoves M. Stress fracture of the supracondylar process of the humerus in a professional tennisplayer. Clin J Sport Med. 2015;25(1):e20-2. 6. Engber WD, McBeath AA, Cowle AE. The supracondylar process. Clin Orthop Relat Res. 1974;(104):228-31. 7. Terry RJ. On the racial distribution of the supracondyloid variation. Am J Phys Anthropol. 1930;14:459-62. 19. Jelev L, Georgiev GP. Unusual high-origin of the pronator teres muscle from a Struthers' ligament coexisting with a variation of the musculocutaneous nerve. Rom J Morphol Embryol. 2009;50(3):497-9. 8. Kane E, Kaplan EB, Spinner M. Observations of the course of the ulnar nerve in the arm. Ann Chir. 1973;27(5):487-96. 20. Ay S, Bektas U, Yilmaz C, Diren B. An unusual supracondylar process syndrome. J Hand Surg Am. 2002;27(5):913-5. 9. Spinner M. Injuries to the major branches of the peripheral nerves of the Philadelphia: WB Saunders; 1978. p. 160-227. Philadelphia: WB Saunders; 1978. p. 160-227. 21. Newman A. The supracondylar process and its fracture. Am J Roentgenol Radium Ther Nucl Med. 1969;105(4):844-9. 10. Rengachary SS. Entrapment neuropathies. In: Wilkins RH, Rengachary SS, editors. Neurosurgery. New York: McGraw- Hill; 1985. p. 1771-95. editors. Neurosurgery. New York: McGraw- Hill; 1985. p. 1771-9 22. Gunther SF, DiPasquale D, Martin R. Struthers' ligament and associa ted median nerve variations in a cadaveric specimen. Yale J Biol Med. 1993;66(3):203-8. 11. Tubbs RS, Marshall T, Loukas M, Shoja MM, Cohen-Gadol AA. CONCLUSION Struthers’ ligament is a rare aponeurotic structure that may or may not be associated with the supracondylar process of the humerus and may compress the median nerve against the deeper structures, changing the normal course of the nerve; it is consequently one of the rare potential sites for nerve compression which narrow the space where nerves pass and consequently can cause motor and sensory symptoms. 141 Acta Ortop Bras. 2017;25(4):137-42 Acta Ortop Bras. 2017;25(4):137-42 REFERENCES The sublime bridge: anatomy and implications in median nerve entrapment. J Neurosurg. 2010;113(1):1102. bridge: anatomy and implications in median nerve entrapment. J Neurosurg. 2010;113(1):1102. ( 23. Casadei R, Ferraro A, Ferruzzi A, Innao V, Mercuri M. Supracondylar process of the humerus: four cases. Chir Organi Mov. 1990;75(3):265-77. 12. Meschan I. Roentgen signs in diagnostic imaging. 2nd ed. Philadelphia: WB Saunders; 1985. 24. Gessini L, Jandolo B, Pietrangeli A. Entrapment neuropathies of the median nerve at and above the elbow. Surg Neurol. 1983;19(2):112-6. 13. Soliere SB. Nevralgia del nervo mediano da processo supra epitrocleare. Chir Organi Mov. 1929;14:171-80. 142 Acta Ortop Bras. 2017;25(4):137-42
https://openalex.org/W2619064611	https://www.frontiersin.org/articles/10.3389/fimmu.2017.00628/pdf	English	null	IL-10 Dysregulation in Acute Mountain Sickness Revealed by Transcriptome Analysis	Frontiers in immunology	2,017	cc-by	6,443	Bao Liu1,2,3, Jian Chen1,2,3, Long Zhang4, Yixing Gao1,2,3, Jianhua Cui5, Erlong Zhang1,2,3, Gang Xu1,2,3, Yan Liang4, Yu Liang4, Jian Wang4 and Yuqi Gao1,2,3* 1 Institute of Medicine and Hygienic Equipment for High Altitude Region, College of High Altitude Military Medicine, Third Military Medical University, Chongqing, China, 2 Key Laboratory of High Altitude Medicine, PLA, Chongqing, China, 3 Key Laboratory of High Altitude Environmental Medicine, Third Military Medical University, Ministry of Education, Chongqing, China, 4 BGI-Shenzhen, Shenzhen, China, 5 Research Center of PLA for Prevention and Treatment of High Mountain Sickness, The 18th Hospital of PLA, Xinjiang, China Acute mountain sickness (AMS), which may progress to life-threatening high-altitude cerebral edema, is a major threat to millions of people who live in or travel to high alti- tude. Although studies have revealed the risk factors and pathophysiology theories of AMS, the molecular mechanisms of it do not comprehensively illustrate. Here, we used a system-level methodology, RNA sequencing, to explore the molecular mechanisms of AMS at genome-wide level in 10 individuals. After exposure to high altitude, a total of 1,164 and 1,322 differentially expressed transcripts were identified in AMS and non-AMS groups, respectively. Among them, only 328 common transcripts presented between the two groups. Immune and inflammatory responses were overrepresented in participants with AMS, but not in non-AMS individuals. Anti-inflammatory cytokine IL10 and inflammation cytokines IF17F and CCL8 exhibited significantly different genetic con- nectivity in AMS compared to that of non-AMS individuals based on network analysis. IL10 was downregulated and both IF17F and CCL8 were upregulated in AMS individu- als. Moreover, the serum concentration of IL10 significantly decreased in AMS patients after exposure to high altitude (p = 0.001) in another population (n = 22). There was a large negative correlation between the changes in IL10 concentration, r(22) = −0.52, p = 0.013, and Lake Louise Score. Taken together, our analysis provides unprecedented characterization of AMS transcriptome and identifies that genes involved in immune and inflammatory responses were disturbed in AMS individuals by high-altitude exposure. The reduction of IL10 after exposure to high altitude was associated with AMS. Keywords: interleukin-10, acute mountain sickness, RNA-seq, inflammatory response, immune response Edited by: Kai Fang, University of California, Los Angeles, United States Edited by: Kai Fang, University of California, Los Angeles, United States Reviewed by: Alain Simard, Université de Moncton, Canada Richa Hanamsagar, Massachusetts General Hospital, United States Correspondence: Yu Liang liangy3456@sina.com; Jian Wang wangjian@genomics.org.cn; Yuqi Gao gaoy66@yahoo.com Reviewed by: Alain Simard, Université de Moncton, Canada Richa Hanamsagar, Massachusetts General Hospital, United States Reviewed by: Alain Simard, Université de Moncton, Canada Richa Hanamsagar, Massachusetts General Hospital, United States Reviewed by: Alain Simard, Université de Moncton, Canada Richa Hanamsagar, Massachusetts General Hospital, United States Correspondence: Yu Liang liangy3456@sina.com; Jian Wang wangjian@genomics.org.cn; Yuqi Gao gaoy66@yahoo.com Correspondence: Yu Liang liangy3456@sina.com; Jian Wang wangjian@genomics.org.cn; Yuqi Gao gaoy66@yahoo.com Specialty section: This article was submitted to Inflammation, a section of the journal Frontiers in Immunology Received: 11 October 2016 Accepted: 11 May 2017 Published: 30 May 2017 IL Bao Liu1,2,3, Jian Chen1,2,3, Long Zhang4, Yixing Gao1,2,3, Jianhua Cui5, Erlong Zhang1,2,3, Gang Xu1,2,3, Yan Liang4, Yu Liang4, Jian Wang4* and Yuqi Gao1,2,3* Original Research published: 30 May 2017 doi: 10.3389/fimmu.2017.00628 INTRODUCTION Liu B, Chen J, Zhang L, Gao Y, Cui J, Zhang E, Xu G, Liang Y, Liang Y, Wang J and Gao Y (2017) IL-10 Dysregulation in Acute Mountain Sickness Revealed by Transcriptome Analysis. Front. Immunol. 8:628. doi: 10.3389/fimmu.2017.00628 Acute mountain sickness (AMS) is the most common disease caused by the lower pressure and reduced oxygen amounts at high altitudes (above 2,500 m) and can progress to high-altitude cerebral edema in severe cases, which has a high mortality (1, 2). A significant increase in the number of sojourns and mountaineers has been observed in high altitudes (3, 4). Although the awareness of altitude-related health hazards increased, the prevalence of AMS was ~16–100% in different study May 2017 \| Volume 8 \| Article 628 Frontiers in Immunology \| www.frontiersin.org 1 Liu et al. Interleukin-10 in AMS designs (3, 5, 6); the median of AMS incidences without prophy- laxis was 60% in randomized trails (7). designs (3, 5, 6); the median of AMS incidences without prophy- laxis was 60% in randomized trails (7). to a high altitude (5,300 m) by bus over a 72-h period, including a 1-day stopover at 3,000 m. Individuals for protein validation set traveled by train from 300 to 3,658 m in ~48 h. Vital signs, including heart rate, blood pressure, and oxygen saturation, were measured at 7:00 p.m. prior to ascent. Upon arrival at high alti- tudes, the same measurable parameters were performed at 7:00 p.m. daily for 5 days, and mean values were calculated. During the exposure to high altitudes, all volunteers did the regimented daily life and avoided any exercises or physical labor. Hypoxia is a principal etiological factor for AMS and initiates a pathophysiology process and causes its symptoms. Insufficient cerebrospinal compliance (8), alterations in fluid balance (9), activation of nociceptors induced by free radicals (10), and vaso- genic edema caused by increased capillary permeability (11, 12) have all been associated with AMS development, but the biologi- cal pathways and exact molecular mechanisms underlying AMS remain unknown (13).f Blood was drawn from a peripheral vein in the morning soon after waking before ascending to high altitudes and after 3 days high-altitude exposures. All blood samples of RNA sequencing set were stored at −80°C with RNA protecting tube for subse- quent analysis. Serum samples of validation set were separated at 3,000 rpm and stored at −80°C for subsequent use. INTRODUCTION Comparison of transcriptome differences in samples from individuals with and without AMS can reveal distinct differences in gene regulation changes related to AMS at genome-wide level. Transcriptome of whole blood is supposed to constitute an acces- sible window to the multiorgan transcriptome by several tran- scriptome profiling studies of different conditions and diseases (14, 15). In this study, we used RNA sequencing to create whole blood transcriptomes to assess the gene expression landscape in response to acute altitude hypoxia and used this to define the molecular changes involved in AMS. Self-assessment questionnaires of the Lake Louise Scoring System (16) were performed at 8:00 a.m. daily for 5 days when volunteers had arrived at high altitudes. AMS was diagnosed through the use of Lake Louise Scoring System, which comprises a questionnaire and a scorecard that determine severity. Individuals with a Lake Louise Score (LLS) ≥ 3 (including a headache score of ≥1) were designated as having AMS, while individuals without a headache or with LLS < 3 were considered to be non-AMS. More clear study procedure is depicted in Figure 1A. RNA Sequencing and Analysis q g y Total RNAs were extracted from 20 whole blood samples from individuals before and after exposure to high altitude. mRNA-seq libraries were constructed according to the TruSeq RNA Sample Prep Kit v2 (Illumina) and sequenced on an Illumina HiSeq 2000 sequencer, following the manufacturer’s instructions. We used the following criteria to filter the raw reads sequenced from the Illumina HiSeq 2000 sequencer: (1) remove reads with adaptors; (2) remove reads in which unknown bases are more than 10%; and (3) remove low-quality reads (the percentage of low-quality bases is over 50% in a read, we define the low-quality base to be the base whose sequencing quality is not more than 5). Then, clean reads were aligned to human genome build hg19 using TopHat (v1.4.1) Study Oversighth g The study was reviewed and approved by the Third Military Medical University Ethics Committee, China. Excluding peo- ple with lung, heart, and blood diseases, 10 individuals were recruited to provide samples for RNA sequencing, another population encompassed 22 individuals were constituted for protein validation set. All of them were healthy young men who had never previously been at high altitude. The study was thor- oughly explained to all individuals who agreed to participate, and all participants signed informed consent forms before their examinations. Enzyme-Linked Immune Sorbent Assay (ELISA) ( ) We measure the concentration of IL10, IL17F, and CCL8 in plasma collected from individuals in transcriptomic analyses set using ELISA with commercially available kits (product number: IL10, ELH-IL10-1, IL17F, ELH-IL17F-1, CCL8, ELH- MCP2-1, RayBiotech, GA, USA) according to the manufacturer’s instructions. Study Procedure Volunteers for transcriptomic analyses set were assembled at a low-altitude starting point (1,300 m), then made a rapid ascent Figure 1 \| The distribution of Lake Louise Scores and symptoms during the investigation period. (A) Schematic of study procedure. (B) Scatter plot for the LLS of all individuals. The numbers correspond to the days when acute mountain sickness (AMS) was assessed. The red dashed line denotes the threshold of diagnosis for AMS. Figure 1 \| The distribution of Lake Louise Scores and symptoms during the investigation period. (A) Schematic of study procedure. (B) Scatter plot for the LLS of all individuals. The numbers correspond to the days when acute mountain sickness (AMS) was assessed. The red dashed line denotes the threshold of diagnosis for AMS. May 2017 \| Volume 8 \| Article 628 2 Frontiers in Immunology \| www.frontiersin.org Liu et al. Interleukin-10 in AMS (17), which built on the ultrafast short read mapping program Bowtie (v1.1.2) (18). The basic information of RNA-seq data set is listed in Table S1 in Supplementary Material. was analyzed before and after arrival at high altitude with chemi- luminescent immunometric assay using an IMMULITE® 1000 autoanalyzer (Siemens, Germany) with commercially available kits (catalog number: LKXPZ, Siemens Healthcare Diagnostics Products Ltd.), which is a method of diagnostic use in the study of inflammatory disease. We used reads per kilobase million as a normalization method in following analysis (19). The expression value of each transcript was calculated; differentially expressed transcripts were selected by comparison of postexposure and preexposure expression val- ues in each group using a density-based pruning algorithm whose key idea is based on the observation that differentially expressed genes tend to have average expression values across conditions. And it confirmed that differentially expressed genes between two conditions are usually located in the boundary region in the 2D feature space of average gene expression versus average difference of gene expression (20). Statistical Analysis y Statistical analyses were performed with the use of SPSS soft- ware, version 19·0. Normality was assessed for all data sets by the Shapiro–Wilk’s test. To examine the difference between the physiological indices of AMS and non-AMS groups, a one-way ANOVA with repeated measures was employed. Age and AMS severity were calculated with an independent t-test. Paired t-test or Wilcoxon signed-rank test analysis was used to compare IL10, IL17F, and CCL8 protein expression of AMS and non-AMS indi- viduals before and after exposure to high altitude, respectively. A Pearson’s product–moment correlation was run to assess the relationship between the changes in IL10 concentration and LLS score in individuals of protein validation set. We used the gene ontology (GO) tool DAVID (version 6.8 Beta) (21) to determine enriched GO terms in the differentially expressed gene set. Then, the Gene Functional Classification tool from the DAVID suite was used to cluster enriched GO terms (biological processes) with the classification stringency set to medium and all other options set to default. According to the instructions of DAVID, we selected the groups with a higher score (scores ≥ 1.3) for subsequent analysis, which indicates that the gene members in the group are involved in more important (enriched) terms in a given study. Identification of Distinct Genetic Expression Patterns in the AMS Group We carried out further analyses on those transcripts that showed expression pattern differences in the AMS and non-AMS tran- scriptome landscapes to uncover the pathophysiological process of AMS. The results of a density-based pruning algorithm analysis of the transcriptome data for each group pre- and postexposure to high altitude showed that 1,164 and 1,322 transcripts differen- tially expressed in the non-AMS and AMS groups, respectively (Tables S3 and S4 in Supplementary Material). Among them, only 328 common transcripts presented between non-AMS and AMS groups (Figure 2A), indicating a substantially different signature of gene expression regulation between the two groups. RESULTS Clinical Characteristics of the Individuals In transcriptomic analysis set, five individuals were diagnosed as AMS and five as non-AMS, while 12 volunteers were desig- nated as AMS and 10 as non-AMS in validation set (Table S2 in Supplementary Material). The onset of symptoms commenced 3–24 h after arrival at the highest altitude, peaked in severity between 48 and 72 h (Figure 1B). During the 5 days of the study, headaches were the most common symptom. To further investigate the potential functional interactivity between gene products and compare the different regulatory modes between AMS and non-AMS groups, we used the biologi- cal modules condensed from enriched GO terms to construct gene networks for each group within differentially expressed genes of participants with AMS.i We investigated five biological modules including immune response, inflammatory response, leukocyte activation, sprouting angiogenesis, and response to oxidative stress, which associated with the inflammatory and immune responses under hypoxia (22–24). For both AMS and non-AMS groups, expression data from high altitude of genes in these modules were used to calcu- late the topological overlap matrix from given expression data to get the topological overlap between xi and xj to measure cluster- ing or shared neighbors (25). The connections among genes were depicted using VisANT (26). The connectivity, which is defined as the sum of a gene’s connection strengths with all other genes in the network, was compared between AMS and non-AMS groups of all genes within their respective network. We found that rapid ascent to a high-altitude environment initiated a cascade of physiological responses: oxygen saturation decreased substantially once individuals were at high altitude; heart rate and blood pressure increased; the mean hemoglobin concentration was higher in individuals postexposure to high altitude in contrast to preexposure to high altitude. But, changes in these clinical features have no significant difference between AMS and non-AMS groups except AMS severity (Table 1). Measurement of Serum IL10 with Clinical Diagnosis Assayhi We used functional annotation clustering analysis to deter- mine the biological module for the differentially expressed transcripts and found that the genes whose transcripts were differentially expressed in the AMS group pre- and postexposure Diagnosis Assayh The protein expression of IL10, which exhibited significantly dif- ferent genetic connectivity between AMS and non-AMS groups, May 2017 \| Volume 8 \| Article 628 Frontiers in Immunology \| www.frontiersin.org 3 Interleukin-10 in AMS Liu et al. Table 1 \| Demographic data and physiological indices of study subjects and comparison in AMS and non-AMS groups. Transcriptomic analyses set Protein validation set Variable Non-AMS group (N = 5) AMS group (N = 5) P-value Non-AMS group (N = 10) AMS group (N = 12) P-value Age, years 0.35 0.31 Median 22 21 24 23.5 Range 20–23 20–23 22–32 21–26 SpO2 (%) 0.12 0.63 Plain 97 (1.0) 96 (2.2) 98 (0.8) 98 (0.5) Plateau 81 (1.2) 77 (7.4) 89 (1.5) 89 (1.8) HR (beats/min) 0.06 0.40 Plain 65 (10.1) 76 (13.8) 69 (5.6) 64 (10.8) Plateau 89 (7.3) 101 (14.8) 87 (9.0) 86 (9.7) SBP (mmHg) 0.96 0.74 Plain 116 (15.0) 116 (14.4) 113 (8.3) 111 (8.9) Plateau 124 (12.2) 125 (13.4) 127 (5.9) 128 (8.7) DBP (mmHg) 0.60 0.91 Plain 58 (5.7) 60 (5.5) 66 (5.6) 69 (6.9) Plateau 74 (7.6) 75 (7.4) 77 (5.4) 76 (6.6) Hemoglobin (g/L) 0.92 0.14 Plain 141 (14.7) 140 (5.1) 153 (8.4) 149 (6.9) Plateau 158 (11.5) 160 (6.2) 168 (7.7) 163 (10.8) AMS severity LLS 2.0 (1.2) 9.4 (2.3) <0.001 1.9 (0.6) 5.8 (1.1) <0.001 Values are mean (SD). A P-value of less than 0.05 was considered to indicate statistical significance. P-Values were calculated with the use of one-way ANOVA with repeated measures, except for age and AMS severity, which were calculated with the unpaired t-test. SpO2, blood oxygen saturation; AMS, acute mountain sickness; HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure; LLS, Lake Louise Score. Table 1 \| Demographic data and physiological indices of study subjects and comparison in AMS and non-AMS groups. ographic data and physiological indices of study subjects and comparison in AMS and non-AMS groups. Protein validation set Values are mean (SD). A P-value of less than 0.05 was considered to indicate statistical significance. P-Values were calculated with the use of one-way ANOVA with repeated measures, except for age and AMS severity, which were calculated with the unpaired t-test. SpO2, blood oxygen saturation; AMS, acute mountain sickness; HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure; LLS, Lake Louise Score. Diagnosis Assayh Figure 2 \| Differentially expressed transcripts and overrepresented biological processes in non-acute mountain sickness (AMS) and AMS groups. (A) Venn diagram showing the number of transcripts equally and differentially expressed in whole blood cells derived from non-AMS and AMS individuals. (B) The top five representative biological modules of differentially expressed transcripts for non-AMS and AMS groups. Figure 2 \| Differentially expressed transcripts and overrepresented biological processes in non-acute mountain sickness (AMS) and AMS groups. (A) Venn diagram showing the number of transcripts equally and differentially expressed in whole blood cells derived from non-AMS and AMS individuals. (B) The top five representative biological modules of differentially expressed transcripts for non-AMS and AMS groups. postexposure to high altitude, a change in the interleukin genes occurred in individuals with AMS, with downregulation of IL2, IL4, IL6ST, IL7, IL7R, IL10, IL17B, IL32, and IL23R and upregula- tion of IL13 and IL17F. The results indicated that inflammatory and immune responses are important pathophysiological pro- cesses in AMS.i to high altitude were more enriched in the biological processes relating to sprouting angiogenesis, immune responses, inflamma- tory responses, apoptosis, and erythropoiesis. In contrast, genes whose transcripts were differentially expressed in the non-AMS group were associated with antigen processing and presentation, cell migration, sprouting angiogenesis, erythropoiesis, and iron homeostasis (Figure 2B; Table S5 in Supplementary Material). Intriguingly, there were no genes enriched in inflammatory and immune responses in the non-AMS group. Moreover, pre- and To further explore the AMS-specific transcriptional regulatory patterns for inflammatory and immune responses, we constructed networks in AMS and non-AMS groups. We further calculated May 2017 \| Volume 8 \| Article 628 Frontiers in Immunology \| www.frontiersin.org 4 Liu et al. Interleukin-10 in AMS the connectivity value for each gene in the networks on the basis of the AMS and non-AMS expression data from high altitude. In the AMS group, the strongest connections were present among five func- tional annotation clusters in the network (Figure 3A). To determine whether these connections were also present in the non-AMS group network, we looked at the network conservation. The connections that showed gene coexpression relationships in the AMS group were essentially absent in the non-AMS group (Figure 3B).fl To further assess our findings with cytokine expression, we measured serum IL10 by chemiluminescent immunometric assay in another population encompassed 22 individuals. In participants with AMS, IL10 protein expression was significantly downregulated after exposure to high altitude. Diagnosis Assayh By contrast, there were no significant changes in non-AMS individuals (Figure 5A). There was a large negative correlation between the changes in IL10 concentration (IL10 concentration postexposure to high altitude − IL10 concentration preexposure to high altitude), r(22) = −0.52, p = 0.013, and Lake Louise Score (Figure 5B). y g Differential connectivity reflects a distinct pattern of genetic expression. We examined the connectivity values of the genes in networks of the AMS and non-AMS groups and made a comparison. We identified 43 genes with significant differences between the two groups (Table S6 in Supplementary Material). Specifically, the cytokines IL10, CCL8, CCR7, and IL17F possessed substantial differ- ential connectivity between AMS and non-AMS groups (Figure 3). In the AMS group, CCL8 and IL17F present an upregulation and down- regulation of IL10 and CCR7 (Figure 4; Table S4 in Supplementary Material). And we found that dual-specificity protein phosphatase (DUSP1), involved in the production of IL10 26, has an upregulation in the AMS group, but its homolog DUSP19 is downregulated in the non-AMS group. Taken together, our results show that individuals with AMS have a distinct genetic profile in which anti-inflammatory cytokine production was reduced. DISCUSSION In the present study, we performed RNA sequencing in whole blood cells of individuals who were rapidly exposed to high altitudes and analyzed the sequencing data of individuals before and after high- altitude exposure in AMS and non-AMS groups. Inflammatory response and immune response were specific transcriptional altera- tion that occurred in participants with AMS, as revealed by functional annotation analyses of the differentially expressed transcripts.l It has been demonstrated in previous studies that inflammatory response is associated with AMS (30, 31). In this study, we have illustrated the questions using a system-level methodology and demonstrated that genes associated with immune and inflam- matory response, including chemokine and their receptors, CD and HLA molecules, and inflammatory cytokines were disturbed in AMS individuals by high-altitude exposure acutely. Among them, the anti-inflammatory cytokine IL10 presents significantly different connectivity between AMS and non-AMS individuals and the changes in IL10 after exposure to high alti- tude present strong correlation with AMS. In our recent study, Serum IL10 in AMS and Non-AMS Groups A meta-analysis in our laboratory has demonstrated that dexa- methasone, an immunosuppressive drug, has been shown to be quite effective in preventing and treating AMS (27). Moreover, cell assays demonstrated that dexamethasone can promote the production of IL10 (28, 29). Combined with our transcriptome analyses results, we speculated that IL10 production may be associated with AMS. Figure 3 \| Cluster visualization identifies acute mountain sickness (AMS)-specific transcriptional regulatory patterns. (A) Connections of differentially expressed genes enriched in five functional clusters that were overrepresented in AMS are depicted. (B) Connections from A that are present in the AMS group but absent in the non-AMS group. Node color represents the enriched biological process. Node size reflects the number of direct connections a gene has within the network. Genes enriched in different clusters are connected by gray lines, and genes enriched in each cluster are connected by colored lines. The cytokines with significant differential connectivity are indicated by red arrows. Figure 3 \| Cluster visualization identifies acute mountain sickness (AMS)-specific transcriptional regulatory patterns. (A) Connections of differentially expressed genes enriched in five functional clusters that were overrepresented in AMS are depicted. (B) Connections from A that are present in the AMS group but absent in the non-AMS group. Node color represents the enriched biological process. Node size reflects the number of direct connections a gene has within the network. Genes enriched in different clusters are connected by gray lines, and genes enriched in each cluster are connected by colored lines. The cytokines with significant differential connectivity are indicated by red arrows. May 2017 \| Volume 8 \| Article 628 5 Frontiers in Immunology \| www.frontiersin.org Liu et al. Interleukin-10 in AMS re 4 \| Quantification of IL10, IL17F, and CCL8 protein using enzyme-linked immunosorbent assay acute mountain sickness (AMS) (N = 5) and AMS (N = 5) individuals in transcriptome assay set before and after exposure to high altitude, respectively. Serum IL10 in AMS and Non-AMS Groups Figure 4 \| Quantification of IL10, IL17F, and CCL8 protein using enzyme-linked immunosorbent assay acute mountain sickness (AMS) (N = 5) and non AMS (N 5) indi id als in transcriptome assa set before and after e pos re to high altit de respecti el Figure 4 \| Quantification of IL10, IL17F, and CCL8 protein using enzyme-linked immunosorbent assay acute mountain sickness (AMS) (N = 5) and non-AMS (N = 5) individuals in transcriptome assay set before and after exposure to high altitude, respectively. May 2017 \| Volume 8 \| Article 628 Frontiers in Immunology \| www.frontiersin.org 6 Liu et al. Liu et al. Interleukin-10 in AMS Figure 5 \| Comparison of IL10 protein expression in independent non-acute mountain sickness (AMS) and AMS individuals by paired t-test. (A) Illustration of IL10 expression in individuals with AMS (N = 12) and non-AMS (N = 10), respectively, before and after exposure to elevated altitudes. (B) Pearson correlation for the changes in IL10 concentration (IL10 concentration postexposure to high altitude − IL10 concentration preexposure to high altitude) and Lake Louise Score. Figure 5 \| Comparison of IL10 protein expression in independent non-acute mountain sickness (AMS) and AMS individuals by paired t-test. (A) Illustration of IL10 expression in individuals with AMS (N = 12) and non-AMS (N = 10), respectively, before and after exposure to elevated altitudes. (B) Pearson correlation for the changes in IL10 concentration (IL10 concentration postexposure to high altitude − IL10 concentration preexposure to high altitude) and Lake Louise Score. we found that the concentration of inflammatory cytokines presents positive correlation with AMS, such as IL6, TNF-α, and IL-1β (32). Maybe the increase in these inflammatory cytokines was a result of the decrease in production of anti-inflammatory cytokine IL10.h countervailing regulation, the inflammatory signal induced by hypobaric hypoxia is converted and readily amplified in an organism, and leukocytes undergo continuous cellular rolling and adherence to endothelium promoted by chemokine, such as CCL8 (43) upregulated in AMS. This results in increased vascular permeability and ultimately leads to clinical manifestation conse- quent to vasogenic edema (44, 45). The molecular signals for IL10 production were observed in the transcriptional landscape. DUSP1 is enriched in response to oxidative stress and limits IL10 production by negatively regulating p38 phosphorylation (33). It has been shown to be induced by hypoxia and has antioxidant properties (34). Serum IL10 in AMS and Non-AMS Groups We note here that DUSP1 is upregulated in our AMS group, but DUSP19, a homolog of DUSP1, was downregulated in the non-AMS group. Another IL10 regulatory mechanism involves the inhibition of T-cell differentiation, which normally secretes IL10. One protein that is involved in the maintenance of Treg cell function is CCR7 (35), and this was downregulated in the AMS group. In contrast, its ligand CCL19 was upregulated in the non-AMS group. These findings suggested that the IL10 production could be diminished in participants with AMS both through the signaling molecule DUSP1 and by genes that control T cell differentiation. Dexamethasone, an immunosuppressive drug, has been shown to be quite effective in preventing and treating AMS (27), although the mechanism by which this is achieved is unclear. Recent studies were reported that dexamethasone can promote the production of IL10 (28, 29). Results from our study demonstrated that IL10 may be a key point for AMS, and thus may be relevant to the effectiveness of dexamethasone. However, dexamethasone is a rather non-specific, generalized suppressor of inflammation. Thus targeted therapy with IL-10 might be more beneficial in AMS prevention.l i We demonstrated that inflammatory response and immune response were specific transcriptional alterations that occurred in participants with AMS and anti-inflammatory cytokine IL10 reduction presents a large positive correlation with AMS; the more the reduction of IL10, the more the severity of AMS. However, only the young health men were included in this study because they are the main part of population who travel to high altitudes for recreation, work, and pilgrimage. Therefore, further investigations in more individuals to confirm these results to exclude potential bias should include females, high altitude residents and old age people. And the more detailed mechanism of IL10 reduction in AMS after exposure to high altitude need to be studied in future. f Immune system is a highly regulated system that is sensitive to several extrinsic factors including environmental stress (36). Consistently, previous data from human studies and animal models indicate that T-cell function is dampened following expo- sure to hypoxia (37, 38). The concept that hypoxia can induce inflammation has also gained credence in recent studies (39). Especially, hypoxia coupled with reduced intrinsic IL10 could activate angiogenesis pathway excessively and result in disease (40, 41). Frontiers in Immunology \| www.frontiersin.org Serum IL10 in AMS and Non-AMS Groups The present study observed that hypoxia could be as a driver for immune and inflammatory responses in AMS patients and revealed that the reduction of anti-inflammatory cytokine IL10, after high altitude exposure, was associated with AMS. REFERENCES 14. Liew CC, Ma J, Tang HC, Zheng R, Dempsey AA. The peripheral blood transcriptome dynamically reflects system wide biology: a potential diagnostic tool. J Lab Clin Med (2006) 147:126–32. doi:10.1016/j.lab.2005. 10.005 1. Hackett PH, Roach RC. High-altitude illness. N Engl J Med (2001) 345:107–14. doi:10.1056/NEJM200107123450206 1. Hackett PH, Roach RC. High-altitude illness. N Engl J Med (2001) 345:107–14. doi:10.1056/NEJM200107123450206 2. Bartsch P, Swenson ER. Clinical practice: acute high-altitude illnesses. N Engl J Med (2013) 368:2294–302. doi:10.1056/NEJMcp1214870 2. Bartsch P, Swenson ER. Clinical practice: acute high-altitude illnesses. N Engl J Med (2013) 368:2294–302. doi:10.1056/NEJMcp1214870 15. Kohane IS, Valtchinov VI. Quantifying the white blood cell transcriptome as an accessible window to the multiorgan transcriptome. Bioinformatics (2012) 28:538–45. doi:10.1093/bioinformatics/btr713 3. MacInnis MJ, Carter EA, Freeman MG, Pandit BP, Siwakoti A, Subedi A, et al. A prospective epidemiological study of acute mountain sickness in Nepalese pilgrims ascending to high altitude (4380 m). PLoS One (2013) 8:e75644. doi:10.1371/journal.pone.0075644 16. Roach RC, Bärtsch P, Oelz O, Hackett PH. Lake Louise AMS Scoring Consensus Committee. The Lake Louise acute mountain sickness scoring system. In: Sutton JR, Houston CS, Coates G, editors. Hypoxia and Molecular Medicine. Burlington, VT: Charles S. Houston (1993). p. 272–4. 4. Salazar H, Swanson J, Mozo K, White AC Jr, Cabada MM. Acute mountain sickness impact among travelers to Cusco, Peru. J Travel Med (2012) 19:220–5. doi:10.1111/j.1708-8305.2012.00606.x 17. Trapnell C, Pachter L, Salzberg SL. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics (2009) 25:1105–11. doi:10.1093/bioinformatics/ btp120 fi 5. McDevitt M, McIntosh SE, Rodway G, Peelay J, Adams DL, Kayser B. Risk determinants of acute mountain sickness in trekkers in the Nepali Himalaya: a 24-year follow-up. Wilderness Environ Med (2014) 25:152–9. doi:10.1016/j. wem.2013.12.027 18. Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol (2009) 10:R25. doi:10.1186/gb-2009-10-3-r25 f 6. Gaillard S, Dellasanta P, Loutan L, Kayser B. Awareness, prevalence, medi- cation use, and risk factors of acute mountain sickness in tourists trekking around the Annapurnas in Nepal: a 12-year follow-up. High Alt Med Biol (2004) 5:410–9. doi:10.1089/ham.2004.5.410 19. Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat Methods (2008) 5:621–8. doi:10.1038/nmeth.1226 20. Hu J, Xu J. Density based pruning for identification of differentially expressed genes from microarray data. BMC Genomics (2010) 11(Suppl 2):S3. doi:10.1186/1471-2164-11-S2-S3 7. AUTHOR CONTRIBUTIONS The Supplementary Material for this article can be found online at http://journal.frontiersin.org/article/10.3389/fimmu. 2017.00628/full#supplementary-material. YG, YL, and JW conceived and designed the study. YG and YL oversaw laboratory analyses and JC provided the overall YG, YL, and JW conceived and designed the study. YG and YL oversaw laboratory analyses and JC provided the overall CONCLUSION In summary, we have explored the gene expression pattern of AMS pattern at currently the most detailed level of resolution. Our data suggest that inflammatory response and immune response were specific transcriptional alterations that occurred in participants with AMS. Cytokine regulatory changes are associated with AMS, and anti-inflammatory cytokine IL10 reduction presents a large positive correlation with AMS; the IL10 is a central cytokine during the resolution phase of inflammation and a general suppressor of cytokines that inhibit pro-inflammatory responses from the innate and adap- tive immune pathways and prevents tissue lesions caused by exacerbated adaptive immune responses (42). Suppression of the anti-inflammatory response in the AMS group was accom- panied by a decrease in production of IL10. In the absence of May 2017 \| Volume 8 \| Article 628 7 Liu et al. Interleukin-10 in AMS more the reduction of IL10, the more the severity of AMS. Thus, targeted therapy with IL-10 might be more beneficial in AMS prevention. supervision of the study. LZ, BL, YG, YL, and EZ did the labo- ratory experiments or contributed to the statistical analysis, or both GX and JC contributed to sample and physical data collections. JC, BL, and LZ drafted the report. All the authors contributed to the interpretation of results, critical revision of the manuscript, and approved the final manuscript. YG is the guarantor. ETHICS STATEMENT This work was supported by the National Key Basic Research Program of China (973 Program, grant No. 2012CB518201), by the Key Projects in the Military Science and Technology Pillar Program during the 12 Five-year Plan Period (AWS14C007), and by the Ministry of Health of PR of China (grant No. 201002012). The study was reviewed and approved by the Third Military Medical University Ethics Committee, China. Excluding people with lung, heart, and blood diseases, we recruited, enrolled, and obtained informed consent from 32 healthy young male volun- teers, who had never previously been at high altitude. DATA ACCESSION The raw data have been deposited to Gene Expression Omnibus (GEO) under accession GSE75665. REFERENCES Waeber B, Kayser B, Dumont L, Lysakowski C, Tramer MR, Elia N. Impact of study design on reported incidences of acute mountain sickness: a systematic review. High Alt Med Biol (2015) 16:204–15. doi:10.1089/ham.2015.0022 8. Wilson MH, Davagnanam I, Holland G, Dattani RS, Tamm A, Hirani SP, et al. Cerebral venous system and anatomical predisposition to high-altitude headache. Ann Neurol (2013) 73:381–9. doi:10.1002/ana.23796 21. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc (2009) 4:44–57. doi:10.1038/nprot.2008.211 9. Loeppky JA, Icenogle MV, Maes D, Riboni K, Hinghofer-Szalkay H, Roach RC. Early fluid retention and severe acute mountain sickness. J Appl Physiol (2005) 98:591–7. doi:10.1152/japplphysiol.00527.2004 22. Wood JG, Johnson JS, Mattioli LF, Gonzalez NC. Systemic hypoxia promotes leukocyte-endothelial adherence via reactive oxidant generation. J Appl Physiol (1999) 87:1734–40. 10. Lopez JI, Holdridge A, Mendizabal JE. Altitude headache. Curr Pain Headache Rep (2013) 17:383. doi:10.1007/s11916-013-0383-2 23. Krock BL, Skuli N, Simon MC. Hypoxia-induced angiogenesis: good and evil. Genes Cancer (2011) 2:1117–33. doi:10.1177/1947601911423654 11. DiPasquale DM, Muza SR, Gunn AM, Li Z, Zhang Q, Harris NS, et al. Evidence for cerebral edema, cerebral perfusion, and intracranial pressure elevations in acute mountain sickness. Brain Behav (2016) 6:e00437. doi:10.1002/brb3.437 f 24. Nathan C, Cunningham-Bussel A. Beyond oxidative stress: an immunolo- gist’s guide to reactive oxygen species. Nat Rev Immunol (2013) 13:349–61. doi:10.1038/nri3423 12. Van Osta A, Moraine JJ, Melot C, Mairbaurl H, Maggiorini M, Naeije R. Effects of high altitude exposure on cerebral hemodynamics in normal subjects. Stroke (2005) 36:557–60. doi:10.1161/01.STR.0000155735.85888.13 25. Ravasz E, Somera AL, Mongru DA, Oltvai ZN, Barabasi AL. Hierarchical organization of modularity in metabolic networks. Science (2002) 297:1551–5. doi:10.1126/science.1073374 13. Lu H, Wang R, Xiong J, Xie H, Kayser B, Jia ZP. In search for better pharma- cological prophylaxis for acute mountain sickness: looking in other directions. Acta Physiol (2015) 214:51–62. doi:10.1111/apha.12490 26. Hu Z, Mellor J, Wu J, Yamada T, Holloway D, Delisi C. VisANT: data-integrat- ing visual framework for biological networks and modules. Nucleic Acids Res (2005) 33:W352–7. doi:10.1093/nar/gki431 May 2017 \| Volume 8 \| Article 628 Frontiers in Immunology \| www.frontiersin.org 8 Liu et al. Interleukin-10 in AMS 38. Mishra KP, Ganju L. Influence of high altitude exposure on the immune system: a review. Immunol Invest (2010) 39:219–34. doi:10.3109/08820131003681144 27. Tang E, Chen Y, Luo Y. Dexamethasone for the prevention of acute mountain sickness: systematic review and meta-analysis. REFERENCES Int J Cardiol (2014) 173:133–8. doi:10.1016/j.ijcard.2014.03.019 39. Eltzschig HK, Carmeliet P. Hypoxia and inflammation. N Engl J Med (2011) 364:656–65. doi:10.1056/NEJMra0910283 28. Agarwal SK, Marshall GD Jr. Dexamethasone promotes type 2 cytokine pro- duction primarily through inhibition of type 1 cytokines. J Interferon Cytokine Res (2001) 21:147–55. doi:10.1089/107999001750133159 40. Dace DS, Khan AA, Kelly J, Apte RS. Interleukin-10 promotes pathological angiogenesis by regulating macrophage response to hypoxia during develop- ment. PLoS One (2008) 3:e3381. doi:10.1371/journal.pone.0003381 29. Barrat FJ, Cua DJ, Boonstra A, Richards DF, Crain C, Savelkoul HF, et al. In vitro generation of interleukin 10-producing regulatory CD4(+) T cells is induced by immunosuppressive drugs and inhibited by T helper type 1 (Th1)- and Th2-inducing cytokines. J Exp Med (2002) 195:603–16. doi:10.1084/ jem.20011629 41. Lai Z, Kalkunte S, Sharma S. A critical role of interleukin-10 in modulating hypoxia-induced preeclampsia-like disease in mice. Hypertension (2011) 57:505–14. doi:10.1161/HYPERTENSIONAHA.110.163329 42. Ouyang W, Rutz S, Crellin NK, Valdez PA, Hymowitz SG. Regulation and functions of the IL-10 family of cytokines in inflammation and disease. Annu Rev Immunol (2011) 29:71–109. doi:10.1146/ annurev-immunol-031210-101312 30. Boos CJ, Woods DR, Varias A, Biscocho S, Heseltine P, Mellor AJ. High alti- tude and acute mountain sickness and changes in circulating endothelin-1, interleukin-6, and interleukin-17a. High Alt Med Biol (2016) 17:25–31. doi:10.1089/ham.2015.0098 31. Julian CG, Subudhi AW, Wilson MJ, Dimmen AC, Pecha T, Roach RC. Acute mountain sickness, inflammation, and permeability: new insights from a blood biomarker study. J Appl Physiol (2011) 111:392–9. doi:10.1152/ japplphysiol.00391.2011 43. O’Boyle G, Brain JG, Kirby JA, Ali S. Chemokine-mediated inflammation: identification of a possible regulatory role for CCR2. Mol Immunol (2007) 44:1944–53. doi:10.1016/j.molimm.2006.09.033 44. Roach RC, Hackett PH. Frontiers of hypoxia research: acute mountain sick- ness. J Exp Biol (2001) 204:3161–70. 32. Song TT, Bi YH, Gao YQ, Huang R, Hao K, Xu G, et al. Systemic pro-in- flammatory response facilitates the development of cerebral edema during short hypoxia. J Neuroinflammation (2016) 13:63. doi:10.1186/ s12974-016-0528-4 45. Schoonman GG, Sandor PS, Nirkko AC, Lange T, Jaermann T, Dydak U, et al. Hypoxia-induced acute mountain sickness is associated with intracellular cerebral edema: a 3 T magnetic resonance imaging study. J Cereb Blood Flow Metab (2008) 28:198–206. doi:10.1038/sj.jcbfm.9600513 33. Saraiva M, O’Garra A. The regulation of IL-10 production by immune cells. Nat Rev Immunol (2010) 10:170–81. Frontiers in Immunology \| www.frontiersin.org REFERENCES doi:10.1038/nri2711 f Conflict of Interest Statement: The authors declare that the research was con- ducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. 34. Hoffmann MS, Singh P, Wolk R, Narkiewicz K, Somers VK. Obstructive sleep apnea and intermittent hypoxia increase expression of dual spec- ificity phosphatase 1. Atherosclerosis (2013) 231:378–83. doi:10.1016/j. atherosclerosis.2013.09.033 35. Forster R, Davalos-Misslitz AC, Rot A. CCR7 and its ligands: balancing immu- nity and tolerance. Nat Rev Immunol (2008) 8:362–71. doi:10.1038/nri2297 Copyright © 2017 Liu, Chen, Zhang, Gao, Cui, Zhang, Xu, Liang, Liang, Wang and Gao. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. 36. Muhie S, Hammamieh R, Cummings C, Yang D, Jett M. Transcriptome char- acterization of immune suppression from battlefield-like stress. Genes Immun (2013) 14:19–34. doi:10.1038/gene.2012.49 37. Meehan RT. Immune suppression at high altitude. Ann Emerg Med (1987) 16:974–9. doi:10.1016/S0196-0644(87)80743-6 May 2017 \| Volume 8 \| Article 628 Frontiers in Immunology \| www.frontiersin.org 9
https://openalex.org/W2637874198	https://pureadmin.qub.ac.uk/ws/files/187286546/353_2593_1_PB.pdf	English	null	Rewriting King Lear in a Diasporic Context, Rewriting the Nation: Second Generation and Life Goes On	Odisea	2,017	cc-by	10,242	Queen's University Belfast - Research Portal: Queen s University Belfast - Research Portal: Link to publication record in Queen's University Belfast Research Portal y Link to publication record in Queen's University Publisher rights g Copyright 2016 the authors. py g This is an open access article published under a Creative Commons Attribution License (https://creativecommons.org/ which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited General rights General rights Copyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights. Take down policy Th R h P t Take down policy The Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made to ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the Research Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk. Open Access This research has been made openly available by Queen's academics and its Open Research team. We would love to hear how access to this research benefits you. – Share your feedback with us: http://go.qub.ac.uk/oa-feedback Published in: Odisea Document Version: Publisher's PDF, also known as Version of record REWRITING KING LEAR IN A DIASPORIC CONTEXT, REWRITING THE NATION: SECOND GENERATION AND LIFE GOES ON42 Rosa María García Periago, Universidad de Murcia Email: rosagperiago@um.es Abstract: This article explores two film adaptations of King Lear located in London in a diasporic community: Second Generation (Jon Sen 2003) and Life Goes On (Sangeeta Datta 2009). This paper examines the ways in which King Lear has to be modified to suit Non–Resident Indians. Following a diasporic framework, this article sheds light on the striking parallelisms and connections between both movies via the presence of a mother figure, two nostalgic Lears, the appearance of Muslim characters and the transformation of an extremely tragic dénouement by a happy ending. The role of the mother is especially significant since it hints at the Indian nation and the long–held association between the mother figure and India in mainstream Hindi cinema. The main hypothesis of this paper is that Second Generation and Life Goes On use Shakespeare’s King Lear, which deals with the division of the kingdom, as a prism through which to approach partition. Both films relocate the action to the UK, more specifically London, since it has one of the largest Indian diasporic communities. Alluding to the colonial legacy of partition and Shakespeare and being made by diasporic filmmakers, they become postcolonial – or rather transnational works. Curiously enough, not only is King Lear rewritten and rein- vented, but also Shakespeare, partition, and, ultimately, the nation, although the films offer different – and contradictory – perspectives and alternatives. Keywords: Shakespeare, King Lear, adaptation, diaspora, partition. Abstract: This article explores two film adaptations of King Lear located in London in a diasporic community: Second Generation (Jon Sen 2003) and Life Goes On (Sangeeta Datta 2009). This paper examines the ways in which King Lear has to be modified to suit Non–Resident Indians. Following a diasporic framework, this article sheds light on the striking parallelisms and connections between both movies via the presence of a mother figure, two nostalgic Lears, the appearance of Muslim characters and the transformation of an extremely tragic dénouement by a happy ending. The role of the mother is especially significant since it hints at the Indian nation and the long–held association between the mother figure and India in mainstream Hindi cinema. The main hypothesis of this paper is that Second Generation and Life Goes On use Shakespeare’s King Lear, which deals with the division of the kingdom, as a prism through which to approach partition. 42 The research of this article was done under the auspices of the research project FFI2015–68871–P “Shake- speare and the 20th century: War, Cultural Memory and New Media”. Open Access Thi h Open Access This research has been made openly available by Queen's academics and its Open Research team. We would love to hear how access to this research benefits you. – Share your feedback with us: http://go.qub.ac.uk/oa-feedback Open Access This research has been made openly available by Queen's academics and its Open Research team. We would love t this research benefits you. – Share your feedback with us: http://go.qub.ac.uk/oa-feedback Download date:24. Oct. 2024 Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 REWRITING KING LEAR IN A DIASPORIC CONTEXT, REWRITING THE NATION: SECOND GENERATION AND LIFE GOES ON42 Both films relocate the action to the UK, more specifically London, since it has one of the largest Indian diasporic communities. Alluding to the colonial legacy of partition and Shakespeare and being made by diasporic filmmakers, they become postcolonial – or rather transnational works. Curiously enough, not only is King Lear rewritten and rein- vented, but also Shakespeare, partition, and, ultimately, the nation, although the films offer different – and contradictory – perspectives and alternatives. Keywords: Shakespeare, King Lear, adaptation, diaspora, partition. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 52 Resumen: Este artículo analiza dos adaptaciones cinematográficas del Rey Lear que tienen lugar en Londres en una comunidad diaspórica: Second Generation (Jon Sen 2003) y Life Goes On (Sangeeta Datta 2009). El artículo examina las diferentes formas en las que El Rey Lear se tiene que modificar para adaptarse a la diáspora india. Gracias a un marco teórico de la diáspora, el artículo se centra en los paralelismos entre las dos películas a través de la presencia de la madre, dos “reyes” caracterizados por la nostalgia, la aparición de musulmanes y la transformación de un final trágico en uno feliz. El papel de la madre es especialmente relevante, ya que hace alusión a la India y a la asociación entre la figura materna y la India en el cine de Bollywood. La hipótesis principal es que Second Generation y Life Goes On utilizan El Rey Lear, que trata sobre la división del reino, como un prisma a través del cual aproximarse a la partición en la India. Ambas localizan la acción en Reino Unido, más concretamente en Londres, ya que tiene una de las comunidades diaspóricas indias más extensas. Las películas, al aludir al legado colonial de la partición y de Shakespeare y al ser realizadas por directores diaspóricos, se convierten en trabajos postcoloniales e incluso transnacionales. Lo curioso es que no sólo se reescribe y reinventa Shakespeare a través del Rey Lear, sino también la par- tición en la India y, en última instancia, la nación, a pesar de que las películas ofrecen perspectivas y alternativas diferentes – e incluso contradictorias. Palabras clave: Shakespeare El Rey Lear adaptación diáspora partición p p y Palabras clave: Shakespeare, El Rey Lear, adaptación, diáspora, partición. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 REWRITING KING LEAR IN A DIASPORIC CONTEXT, REWRITING THE NATION: SECOND GENERATION AND LIFE GOES ON42 This paper takes as a premise the necessity to explore King Lear in a diasporic context to delve into a different understanding of Shakespeare and to reflect on Indian and diasporic social and political conditions. As necessary first steps in examining these diasporic Lears, it is worth analysing the significant role King Lear has always played on the Indian stage. Poonam Trivedi has argued that “Lear–like themes of banishment, suffering and exile” (2010: 1) have been dominant in India. In fact, Trivedi alludes to an Indian folk tale of an elderly maharajah who gets depressed when he tests his daughters’ love before dividing his kingdom. When hearing that his youngest daughter claimed that she loved him like salt – a necessity – he burst in anger and disinherited her in favour of the other two. Besides, the topics of banishment and exile are also part and parcel of the great Indian epics, Ramayana and Mahabharata. Apart from these allusions, Alessandra Marino (2013: 72) even claims that the partition in King Lear resembles the division of India and Pakistan in 1947. Consequently, Lear’s story is not unknown to Indian audiences and plays a crucial role in the country. On the Indian stage, the adaptations of King Lear range from English performances – such as some made in 1832 for the Chowringhee Theatre – to Parsi theatre (localized) productions with happy endings, namely Atipidacharita in 1880. Parsi Theatre productions evoked a clear case of cultural resistance to the oppressive, imperial system governing in India at the time. As Sisir Kumar Das notices (2005: 47), the transformations of Shake- speare’s oeuvre were required to address local audiences. Songs, dances, the change of language and, above all, changes in the plot were necessary to interest locals. But King Lear was not only used in the colonial period, since there were also post–colonial versions of the Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 53 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 53 play. For instance, St. Stephen’s College Shakespeare Society performed Lear in English in 1962, with a young Roshan Seth as Lear. Ebrahim Alkazi, produced a Raja Lear, in Urdu translation, for the National School of Drama in 1964, “a production that has become a benchmark of the universalized Shakespeare” (Trivedi 2010: 1). The play was translated and produced in several languages, including Bengali, Hindi, Malayalam or Tamil. 43 Read for instance Ania Loomba, ‘Local–manufacture made–in–India Othello fellows’: Issues of race, hybridity and location in post–colonial Shakespeares’. In Ania Loomba and Martin Orkin (eds.) Post–colonial Shakespeares. London and New York: Routledge, 143–163. 44 See Rosa M. García–Periago, “English Shakespeares in India: 36 Chowringhee Lane and The Last Lear”, Borrowers and Lenders: The Journal of Shakespeare and Appropriation 10.1 (2015) to read about the similari- ties between the films. REWRITING KING LEAR IN A DIASPORIC CONTEXT, REWRITING THE NATION: SECOND GENERATION AND LIFE GOES ON42 The well–known Kathakali King Lear has been the object of in–depth research.43 Curiously enough, Shakespeare’s play has not enjoyed the same trajectory on the Indian screen. There is a mismatch between the numerous stage productions and the scarce film adaptations or offshoots. Probably the serious nature of Shakespeare’s play has contributed to the lack of adaptations of King Lear in Hindi mainstream cinema. If the troubles between the family in King Lear are reminiscent of the political scene in India, as Trivedi claims (2010: 1), the scarcity of visual materials in what is often called an “escapist” cinema is understandable. Only two movies – 36 Chowringhee Lane (Aparna Sen 1981) and The Last Lear (Rituparno Ghosh 2007) – belonging to Indian alternative or parallel cinema contain Lear figures and are freely based on the play.44 Given the lack of resources, it is worth mentioning the existence of two film adaptations of Shakespeare’s King Lear in a diasporic context: Second Generation (2003), directed by Jon Sen and script by Neil Biswas and Life Goes On (2009), directed by Sangeeta Datta. Both are set in London and revolve around first and second generation diasporic beings. This essay attempts to show the versatility of Shakespeare and how the rewriting of the play deepens not only our understanding of Shakespeare, but also of India and Indian–di- asporic culture. The exploration of the role of women, first generation diasporic Lears, the role of Muslims in the films and, finally, the transformation of the endings gear toward not only a reinterpretation of the text, but also of the nation. In fact, in Second Generation and Life Goes On, King Lear becomes a site for the construction of a national (or transnational) identity in a diasporic context. DIASPORIC FILMMAKING Second Generation and Life Goes On should be labelled instances of accented cinema (Naficy 2001: 2) since they follow the modes of production, direction and distribution of other diasporic movies. Jon Sen is a British filmmaker and Neil Biswas is a British–Asian scriptwriter who benefitted from the Indian hype that invaded Britain in 2002 and 2003 with productions such as Tim Supple’s Twelfth Night (2002). As a BBC producer claimed, “for some reason, the white middle classes [were] eating up Asian culture like fish and chips” (in Greenhalgh and Shaunnessy 2007: 99). Second Generation was a two–part TV version of King Lear commissioned by Channel Four. Neil Biswas was himself a diasporic individual, so he was familiar with two different cultures, the one of the homeland as well Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 54 as the one of the host country. Even the music – composed by the Asian underground composer Nitin Sawhney – was also part of the diasporic project. Life Goes On (2009) was directed by the Indian born but London based Sangeetta Datta. Her aim was to make a freewheeling adaptation of King Lear in modern times and in a British Asian context. According to Hamid Faficy, one of the main features of accented cinema is low budget production. These projects are often “non–commercial and usually artisanal and collective in their mode of production” (2001: 45). It is worth emphasizing that all these productions move away from artistic takes and tend to use conventional cinematic shots; they do not aim for aesthetic originality, but for content. While Channel Four was the main source of financing of Second Generation, Storm Glass Productions – a small company – produced Life Goes On. The latter was mainly screened in film festivals – it was in fact released in Mumbai film festival – and in cinemas mostly targeted at diasporic clienteles. As Keith Corson (2015: 135) claims, the main feature of the film circuit is the avoidance of popular products in favour of art films that would not find an audience otherwise. Thus, it is clear that Life Goes On is an art film, taking part in the film circuit. The cast in Second Generation and Life Goes On needs to be analysed because it appears as an interesting example of the ghosting of actors. DIASPORIC FILMMAKING The most familiar example of “ghosting” of roles alludes to the process by which the audience expects to see the same actor/actress in the same role again and “producers expect to market the show more easily as a result” (Hatchuel 2011: 25). Extremely interesting is the case of Parminder Nagra who, before being the female protagonist in Second Generation, played the role of Jesminder (Jess) in Bend It Like Beckham (Gurinder Chadha 2002) and Viola in Twelfth Night (Tim Supple 2003), and thus revolutionized diasporic audiences. Consequently, she was the perfect ac- tress for Second Generation, since audiences were already familiar with her in this kind of role. Another instance of ghosting is Om Puri, who, curiously enough, plays crucial roles in Second Generation and Life Goes On; he is King Lear in Second Generation and the fool in Life Goes On. Having acted in similar roles before, such as the successful diasporic movie East is East (Damien O’Donnell 1999), Om Puri becomes the suitable actor for such films. Regular cinemagoers are inevitably “haunted” by these ghosts of previous performances. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 NOSTALGIC LEARS In fact, his nostalgia for a mythic origin and for a sense of rootedness is developed even more after Mr Kahn’s death – the Gloucester–like figure. His imaginary national identity makes him idealise and venerate the homeland. For that reason, only when he returns to mother India are his problems solved; the passage to India becomes compulsory for the Lear figure. Life Goes On equally locates King Lear in London. Although according to Suzuki Tadashi, “Lear’s tragedy of solitude and madness must be brought forth not as specific to his kinship in distant time and space but as relevant to any old man living in any age in any country” (quoted in Carruthers 2009: 99), Life Goes On concentrates again on a first generation diasporic Lear – called Sanjay Banerjee – who works as a doctor. Like Second Generation, the film opens with the mother figure, paring down the original text significantly. Instead of focusing on Lear’s story, the film rather revolves around the conse- quences caused by the mother’s death and how the main characters were affected by such a loss – Dr. Sanjay Banerjee, the three daughters Lolita, Tuli and Dia (Gonerill, Reagan and Cordelia respectively) and their long–time friend Alok, who stands for the fool in King Lear. Dr. Sanjay Banerjee’s nostalgia is evident from the outset of the movie when he is following a cricket match between India and Pakistan on the radio, while his wife and his friend Alok are watching it live. Throughout the film, this idea of a nostalgic Lear is only reinforced; the audience learns during Manju’s funeral that he holds a powerful position in the Hindu community, like a king. He is a strict person who clings to Hindu values and traditions and plans to pass them on his three daughters with more or less luck. But the past constantly invades Banerjee; history and nostalgia mingle in this diasporic being through nightmares. With a sense of foreignness within the newly inhabited country, he cannot but remember his homeland. Nevertheless, he cannot feel at peace since when he is sleeping, he is constantly disturbed by awful memories associated with the partition of India. NOSTALGIC LEARS Second Generation and Life Goes On are characterised by the presence of Lear figures who are Non–Resident Indians. Set in London in the 1990s, Second Generation deals with a first generation migrant called Mr Sharma (Lear), who abandoned Calcutta in the 1960s to settle in south–east London. The death of Mr Sharma’s wife inevitably triggered his de- cision to abandon his homeland with his three daughters Pria, Rina and Heere – Gonerill, Reagan and Cordelia respectively. Starring Om Puri in the leading role, Sharma–ji appears as a typical diasporic individual, full of nostalgia for the mother country. According to Ato Quayson and Girish Daswani, “nostalgia is now commonly associated with rupture from, and the desire to one day return to, a place called home. It is also associated with the mourning of return, at least to a home as one remembers it” (2013: 17). Despite living in London, he feels alienated from his host country UK and constantly longs to return to his homeland India. Sharma–ji’s alienation can be noticed in his endless attempts to maintain Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 55 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 55 Indian customs, traditions and even food. Not surprisingly, the Lear figure is the owner of one of the most important curry businesses in London. Perhaps one of the distinguishing features of this diasporic rewriting of Shakespeare’s King Lear is Lear’s illness from the outset of the film. As Kenneth Muir observes (1966: 32), in the original source text, King Lear’s madness starts once he feels rejected by his daughters and feels alone in the world and without possessions. However, Jon Sen’s movie highlights the fact that the Lear figure is ill from the beginning. In fact, the opening scene depicts an ailing father suffering from a coma, already showing his malaise. The film anticipates his identity crisis, extremely connected to his traumatic experience. According to Paromita Chakravarti, “the Lear plot becomes the template of a narrative of death and lost legacies framed by nostalgia for a disappeared world” (2014: 139), and it is precisely this nostalgia what haunts Mr. Sharma. There is no difference between the Lear depicted at the beginning and the one halfway through the movie; if anything, his nostalgia increases and he feels completely disoriented. NOSTALGIC LEARS Curiously enough, the most important scene in Shakespeare’s King Lear – the division of the kingdom – is totally or partially eliminated in Second Generation and Life Goes On. In Second Generation, for instance, this scene is not included. Sharma–ji never gives up his power and authority; instead his daughters Pria and Rina take advantage of the situation, and simply take over when he is in hospital in the Intensive Care Unit with a coma. As Marino claims, “they secretly try to carry out their plan to sell the factory” (2013: 172). Obviously, the omission of this scene entails the victimisation of this King Lear. While Shakespeare’s Lear is responsible for his fate and destiny, Mr Sharma is a mere victim of his daughters, and is betrayed by them. Life Goes On presents a parallel development of the story. The division of the kingdom does not take place at the beginning, but halfway through the movie. Interestingly, there is no division of the kingdom as such. Mr Banerjee simply informs his daughters of his estates, bank shares and divides his wife’s jewellery among the three of them. This scene makes perfect sense in the movie after Manju’s death, as he may feel his death is not long to come. It is no surprise that the scene does not take place entirely, but is interrupted by Dia’s unwillingness to listen to all the properties. If the division of the kingdom is so central in King Lear, its total or partial omission in Second Generation and Life Goes On is suspicious, and connected with nostalgia. The total or partial absence of the partition of the sovereignty provides a channel through which to explore the partition of India, since it is an allegory. Such was the trauma caused by Partition and the ensuing bloodshed in India that it can be too harsh and difficult to deal with it in movies, even via the allegory of the division of the kingdom in King Lear. Second Generation for instance shows a grim vision of London, but an idealized one of India in which there is no division of the kingdom and, for that reason, there is no partition either. The country is the home of Hindus and Muslims alike, even deleting all the allusions to Islam at the end of the film, as if all the inhabitants of India had to be Hindus. NOSTALGIC LEARS In the words of Földváry, “broken, fragmented images of a happy childhood are continued by equally fragmented images of violence and terror, with no continuity between them and the main narrative of the film to offer any explanation or consolation” (2013: 306). Thus, although the knowledge of his youngest daughter’s relationship with a Muslim man (Imtiaz) entails a kind of madness in this Lear counterpart, his on–going nostalgic feelings are over present. Dr Sanjay Banerjee recalls the trauma originated by partition, talks on TV about it and needs to find a way to release his pain. What this diasporic Lear emphasizes is the Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 56 similarity between the way he feels in a host country such as England and the homeland India, since partition made citizens lose everything they had, and were like refugees in their own land. Consequently, Banerjee’s nostalgia cannot be solved by returning to the current homeland, but should be settled in the diasporic country. Unlike Sharma–ji, the similarities between Lear and Dr Sanjay Banerjee go beyond the main themes and part of the plot. Ellen Dengel–Janic and Johanna Roering (2008: 212) have realized that Second Generation explores the dynamics of King Lear by retaining the three sisters and the confrontation between the father and the youngest daughter, but Life Goes On seems to go further. The film includes a play–within–the–film in which Dia plays the role of Cordelia. Interestingly, Dia is an onstage Cordelia and a real life one. The audience and Dr Banerjee learn through Manju that Lolita and Tuli referred to their father as Lear since his long–time favourite was clearly Dia. In fact, after spending the night alone, lost in the middle of nowhere trying to make sense of his life again, he sees Dia and quotes from King Lear: “Pray do not mock me, I am a very foolish old man” (4.7: 69–70). This quotation inevitably reminds us of another diasporic Lear, Miss Violet Stoneham, the protagonist of 36 Chowringhee Lane. Like Dr Banerjee, she equally quotes from King Lear to shed light on her solitude and nostalgia, but with a very different outcome. The presence of Shakespeare’s King Lear in Life Goes On then plays a paramount importance, and is linked to the bond with the land. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 MOTHER FIGURES AND NATION Much of the discussion around King Lear has centred around the absence of a mother figure – or even mother figures if the subplot with Edmund and Edgar is also taken into account. Coppélia Kahn, for instance, in her infamous essay entitled “The Absent Mother in King Lear” (1985) tried to uncover the hidden mother in Lear’s inner self. She realised that unlike one of the main sources of King Lear, Shakespeare’s play conspicuously omits the mother figure so that “the play articulates a patriarchal conception of the family in which children owe their existence to their fathers alone; the mother’s role in procreation is eclipsed by the father’s” (1985: 4). In spite of the fact that there is no literal mother in King Lear, Kahn pointed out the imprint of mothering on the main character through hysteria. Janet Adelman in one chapter within her book entitled Suffocating Mothers: Fantasies of Maternal Origin in Shakespeare’s Plays, ‘Hamlet’ to ‘The Tempest’ (1992) cast light on Lear’s tragedy as being caused by maternal deprivation. Her analysis equally emphasized the role of Cordelia as a nurturing mother.i Interestingly, the two films under analysis are characterized by the presence of a mother, who needs to be analyzed in depth to understand the implications. Instead of starting with the division of the kingdom, Second Generation’s opening scene focuses on the Lear–figure called Sharma–ji being at hospital with a coma. From the outset, the movie centers around Lear’s mind and hallucinations that revolve around his late wife Sonali, who committed suicide to escape the strict patriarchal codes of her society and community. Sonali’s way of dying pursues Sharma–ji, who is constantly haunted by these hard and terrible memories highlighting a guilt complex. Thus, he cannot but feel guilty for his wife’s death. Second Generation frequently incorporates shots of the deceased Sonali, always wearing the traditional Indian outfit saree. Really striking are the resemblances between the mother and the youngest daughter Heere. For Alessandra Marino, “mother and daughter become confused in the eye of patriarchy” (2013: 176). Apart from the obvious physical resemblances – emphasized through Parminder Nagra performing both roles – their personalities are very much alike since they are rebellious. NOSTALGIC LEARS The partial inclusion of the division of 57 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 57 the kingdom in Life Goes On equally bears a resemblance with the partition of British India. During the reading of his wife’s will, Mr Sanjay Banerjee gets too irritated when his youngest daughter Dia claims she is not in the mood of doing it. This scene is constructed just after Dia’s confession of her relationship with a Muslim boy – Imtiaz – to her father. Father and daughter overreact at this moment and demonstrate the ways in which the past is recovered in the present. Mr Sanjay Banerjee – via Dia’s ‘betrayal’ with a Muslim – is reminded of the pain and trauma he felt during Partition when his best friend Imtiaz also betrayed him. The current situation in Life Goes On brings to life the suffering and pain of the historical past as well as the unsolved or even insolvable problems. Yet, this scene somewhat evokes a gloss of nostalgia over these events since they are connected with an ideal homeland. Both interestingly present complex, diasporic Lear figures yearning for an idealised mother country and looking for their identity once lost during Partition. MOTHER FIGURES AND NATION While Sonali did not satisfy the conventional image of female roles in a patriarchal society by committing suicide, Heere also presents herself as an unruly woman due to her relationship with a white man called Jack. In Life Goes On, the mother figure is “an image presented with Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 58 unquestioning nostalgia” (Földváry 2013: 309). To begin with, this nostalgia is boosted from the beginning for it only takes five minutes for Manju to die in Life Goes On. Like Sonali in Second Generation, all the images and shots showing Manju are presented as flashbacks. Paring down Shakespeare’s play considerably, the movie revolves around Manju’s death, funeral and the consequences it has rather than around the division of the kingdom, which does not appear as such. Hardly any negative aspect is shown for most of the film; Manju is depicted as a paragon of virtue. She was a domestic woman who regarded home as an inclusive territory, was in charge of passing on Bengali culture and values – as can be seen in her wardrobe full of sarees, when she supports the Indian National cricket team or when she sings Indian songs. She also united the family around her and, above all, stood for mother nature. Whenever the mother appears, she is always taking care of plants and flowers, highlighting her nurturing role. Like in Second Generation, it is the youngest daughter Dia who resembles her mother the most and maintains her spiritual essence, and also becomes a primordial nurturing force via the flowers and plants of the house. The mother’s presence carries a considerable weight in Second Generation and Life Goes On.i The mother in these two movies clearly invokes the symbolic figure of Mother India or the nation; it could refer “to the country as a geographical and political unit” (Földváry 2013: 309). The mother – or rather, the suffering mother – already became a metonym for the na- tion in the 19th century when India was struggling for political autonomy and independence. This idea was part of the anticolonial movement principle. In mainstream Hindi cinema, the mother figure stood for the nation especially after the independence of India and, above all, after the worst holocaust in India – partition. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 MOTHER FIGURES AND NATION One typical portrayal of the female figure was as a “powerless” woman, “an object to be acted upon” (Virdi 2003: 67). The Hindu–Muslim strife triggered nostalgia. There was an idealized mythification of the mother as an attempt to reinvent an ideal Indian past. With the intention to resolve contradictions between different communities, the mother was idealized to provide a sense of unity to the population. As Virdi notes, the female figure “works as a symbol of unity, a Pan–Indian consciousness that Hindi cinema strives to project at a moment when political fragmentation is imminent” (2003: 72). But the mother figure equally entailed sacrifice, forbearance and resilience. The association between mother and nation in Hindi mainstream cinema has long existed. It probably started with Mehboob Khan’s Mother India (1957), which revolved around a mother who sacrificed her own son in favour of the nation. The main character endured all the problems stoically, both nurturing, but at the same time punishing her children until in the last act of selflessness, shot her youngest son to death for turning bandit and doing mischievous acts. Starring Nargis in the lead role, she became the mother par excellance in Bollywood cinema, and immortalized the mother figure. What is commemorated in Mother India is precisely a past where women have timelessly and constantly served the nation. As Sinha claims, “Mehboob’s Mother India went on to become a poster child for the patriotism inspired by the new Indian nation–state” (2006: 248). It managed to represent the major themes of the Nehruvian era but, above all, excelled in connecting woman, land and family well–being. In the subsequent decades, the mother figure continued being idealized. From the suffering mother in Amitabh Bachchan’s movies usually portrayed by Nirupa Roy to the mother performed by Jaya Bachchan in the more recent Bollywood films – Kabhi Kushi Kabhie Gham (Karan Johar 2001) or Kal Ho Naa Ho (Nikhil Alvani 2003), the association between mother and nation is still pertinent. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 59 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 59 Manju in Life Goes On follows in the footsteps of these idealized mothers in Hindi mainstream cinema and seems to be an extension of the resilient mother figure, but the character gives fresher insight to the gender issue. MOTHER FIGURES AND NATION Manju is the “figurehead of the familial and communal” (Bhattacharya 2011: 136) and not only stands for Mother India, but also for Mother Earth. On the one hand, she holds the family together. Her daughters Lolita, Tuli and Dia always tell their mother their dose of problems. Lolita is in an unhappy mar- riage, Tuli is in a lesbian relationship and Dia has a Muslim partner. On the other hand, Manju is a metonym for Mother Earth because when she appears, land, flowers and plants are involved. Curiously enough, the spectators discover throughout the movie that Manju had been unfaithful to the Lear figure with the long time friend Alok. Interestingly, there is no condemnation of the mother, but complete understanding on the daughters’ behalf due to her constant loneliness. Never is Manju criticized in Life Goes On; in fact, the one who is blamed by the Lear figure is Alok for betraying their friendship. These nuances in the mother figure show an evolution in the treatment of gender in this hybrid cinema since it has tended to be very conservative and patriarchal. The mother’s death involves a re–organization and reconstitution of the family. Life Goes On does not show the division of the kingdom as such, because Manju’s death triggers a partition within the family; the family members are lost for a while, wandering around until Manju’s spirit reunites them all again. Partition is over–present in Dr. Banerjee’s mind. He frequently has nightmares which constantly bring to light the lost past with a Muslim friend and how partition put closure to that. There is a recreation of the trauma of Partition and its massacres via the mother figure. If Manju is a metonym of the nation that indirectly provokes a partition with her death – an allegory of King Lear’s partition and India’s partition into three countries (India, Pakistan and Bangladesh) – then, the solution provided in the movie is to come to terms with the mother’s death – and, therefore, the nation’s partition – in an example of Bhabha’s infamous “third space,” London. Second Generation’s jingoistic appropriation of the national space through the woman’s body is clear from the outset. The film’s opening scene depicts an ailing Lear figure unable to forget his past. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 MOTHER FIGURES AND NATION Interestingly, the audience discovers that the main cause of his distress is his wife’s traumatic passing – since she had committed suicide not to abide by patriarchal rules – and his unwillingness to comply with the long rooted Hindu tradition of throwing the ashes of the dead to the Ganges river to absorb the impurities. Immediately after the death, he and his three daughters leave the country and move to the diasporic location par excellence, London. Consequently, Mr. Sharma not only abandons his wife’s body, but also ‘Mother’ India. The film forcefully reminds us of what is at stake here – the association between the mother and nation. During the twenty years Mr. Sharma has lived in London, he has not visited India once, which could be attributed to his inability to forgive his wife and, by extension, the nation. In spite of the fact he seeks to continue with his traditions in the West, he has not resolved his inner conflicts with his wife and nation, and his memory constantly reminds him of them; he is the victim of hallucinations and fits of anger. In the words of Marino, “he experiences a controversial relationship with the country that is at one and the same time the agent responsible for splitting his motherland and the ‘stepmoth- er’ to which he turns, and faces the difficulties involved in raising his children in such a culturally different environment” (2013: 172). Mr. Sharma’s problems are aggravated due Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 60 to the similarities between his wife and youngest daughter. If Sonali had suffered from depression and had killed herself showing a rebellious personality, the diasporic version of Cordelia – Heere – equally appears as an independent spirit turning away from the imposed Hindu traditions, having a relationship with an Englishman called Jack and opting for exile from her family. “Sharma’s troubled relation with Heere then emerges as a transposition of the intimate and conflicting dialogue the man has with Sonali during his illness” (Marino 2013: 276). Not until Mr. Sharma resolves the conflicts with his daughter Heere will he manage to come to terms with his wife – and the repressed past that keeps haunting the present – and ‘Mother India’. Second Generation’s climax with Mr. MOTHER FIGURES AND NATION Sharma, Heere and Sam in the homeland reveals both the reconciliation with his deceased wife Sonali as well as with the Indian nation. The ‘stepmother’ England paves the way for the mother country and the mother figure where Mr. Sharma is at ease again without anxiety, madness and/ or depression. Curiously enough, what Second Generation seems to do is to reconcile Mr. Sharma with the mother/nation and, by extension, King Lear with the absent mother. Second Generation toys with the idea of the nation, even through the actor who per- forms the role of King Lear’s Fool, Roshan Seth. Apart from establishing the connection between the mother and the nation, the film points out one of the active agents of building the Indian nation – Nehru – via the actor. The choice of actor is not random, but based on a long list of performances in which Seth has played Nehru’s role. In the words of Gau- tam Basu Thakur, Roshan Seth is “without a doubt the quintessential ‘face’ of Nehru on Indian celluloid” (2011: 86). Roshan Seth has starred as Prime Minister Nehru on endless occasions on the Indian screen: Gandhi (Richard Attenborough 1982), Food For Ravens (Griffiths 1997) and The Last Days of the Raj (Hindmarch 2007) are only some instances. Gender and nationalist discourses are then entangled in these two movies. According to Sangeeta Datta, “nationalist discourse constitutes the female body as a privileged signifier and various struggles are waged over the meaning and ownership of that body” (2000: 73). The link between mother and nation acquires a new dimension in a diasporic context where the homeland tends to be venerated – at least by the first generation migrants – in what Tölöyan names “exilic nationalism” (2010: 34). Diasporic citizens consider not living in the mother country as a loss or deficiency. If the homeland entails the cultural identity of a country, the mother figure is in charge of transmitting these values in a diasporic context. Yet, as can be imagined, this bond is not exempt from problems since there is a clear patriarchal attitude linked to it. The interesting idea is that Life Goes On seems to challenge this view. MUSLIM CHARACTERS One of the remarkable similarities between Second Generation and Life Goes On has to do with the presence of Muslim characters. While Jon Sen’s film includes King Lear’s subplot with Gloucester – Mr. Khan – and his two sons Edmund (Firoz) and Edgar (Sam) in a Muslim family, Datta’s movie basically turns the Prince of Burgundy into a Muslim character, but is not interested in reproducing the subplot. Curiously enough, Second Gener- ation’s Edgar (Sam) inevitably reminds the audience of Nahum Tate’s version of King Lear due to his relationship with Cordelia’s counterpart – Heere – and their happy resolution. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 61 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 61 In Second Generation, Mr. Hansaab Khan amalgamates two characters in one – Kent and Gloucester – and is given a terrible ending. If Kent is banished from the kingdom, Mr. Khan is fired from the curry company owned by Mr. Sharma. Furthermore, like Kent, he always supports Sharma–ji. But Mr. Khan is also indebted to Gloucester since he has two sons like him – Firoz and Sam – who are the doubles of Edmund and Edgar respectively. A twist in the plot occurs when Mr. Khan dies. Unlike Gloucester who fantasizes with death at Dover, Mr. Khan commits suicide since he felt professionally redundant and emotionally lonely. Gloucester’s mere fantasy is transformed into a painful and real act that achieves intensity to sustain audience contemplation. The moment before committing suicide shows an emotional character after watching videos of the longed for past. Endless close–ups of the character crying after finding his son Firoz is having an affair with Reena or after learning about the relationship between Heere and Sam and, especially after feeling isolated advance and prepare the audience for the tragic ending. A moved and touched Khan phones his sons but he is so unlucky that nobody answers it. The audience cannot avoid thinking that no one can be so wretched in life since apart from losing his wife in childbirth twenty years ago, he is abandoned by his long–life friend and even by his sons. In fact, when Sam notices the phone call, it is just too late and his father has already committed suicide. MUSLIM CHARACTERS Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 62 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago The Hindu–Muslim strife is equally present in the Sam–Heere relationship à la Nahum Tate’s Edgar–Cordelia style. Grippingly, the film opens with scenes of Heere having a relationship with a white man called Jack, but the audience soon learns that she had pre- viously had a love affair with Sam Khan. As Geraldine Harris notes, “these two had been lovers but in the past were pushed apart by parental disapproval arising from religious difference” (2006: 97). In spite of the long–lasting friendship between the two families, it had not been enough to hold them together, and religious difference got in the way. Sam and Heere emerge in Second Generation as hybrid beings trying to reconcile Indian culture with British culture, which is certainly a site of conflict. It is precisely this hybridity that characterizes them what allows them to be together again. Hybridity here should not be taken as the celebratory, triumphalist term Gayatri Chakravorty Spivak warns us against, but as a site of transformation and change “that entertains difference without an assumed or imposed hierarchy” (Bhabha 1993: 4). Although hybridity has become a buzzword in recent criticism, it generally entails conflict between two different cultures. Sam and Heere realize their similarities weigh more than their differences and their desire to be together pushes them to move forward. Religious conflict arises again when Mr. Khan learns about their union. Given that Heere is a Hindu whereas his son is a Muslim, he cannot but oppose to the relationship. That said, this impossibility of being together makes Sam and Heere look like Romeo and Juliet more than Edgar and Cordelia. The death of the Muslim patriarch gave Sam and Heere free rein to their love and union, Heere made amends to her father, decided to go back to India and, interestingly, all the references to Muslims disappear. Besides, even Sam Khan seems to have forgotten his religious background at the film’s ending. Together with Heere and his father–in–law, Sam Khan has a happy–ever–after finale if it was not for the complete erosion of his customs, values and religion that the movie seems to promote. MUSLIM CHARACTERS The viewers discover the horror of the scene via a close–up of the white legs hanging on a chair and through the medium shot of Sam, feeling guilty about the dreadful death. By exploring his sudden and unexpected passing, we discover curious and interesting findings. It is – to say the least appalling – that the only moment in which the rewriting is actually more depressing or bleaker than the original source is when the death of a Muslim character is involved. Gloucester was comforted by his son Edgar at the end of the play whereas Mr. Khan dies completely alone, abandoned by his sons and even with unresolved conflicts with them. Taking into account the fact that Second Generation ends as a ‘feel–good’ movie in a light–hearted tone, Mr. Khan’s suicide needs to be explored; this death has not taken place at random, but for ulterior – and not naïve – motives. Second Generation obviously spotlights the religious dimension and religious difference at this point. It seems that the Muslim patriarch is marginalized and stigmatized in the movie to the extent he becomes a victim of this diasporic society, which is mainly Hindu, and kills himself. There are two people who commit suicide in Jon Sen’s film, Mr. Khan – a Muslim – and Sonali – Shar- ma–ji’s wife. In a way, the movie seems to establish a parallelism between Muslims and women, since they both suffer the ‘stigma’ of being outside the male Hindu norm. y g g Second Generation appears as a contradictory site of “Muslim diasporic invisibility and hypervisibility” (Alsultany and Shohat 2013: 9). Although it depicts the alienation and isolation of Muslim characters through Mr. Khan’s suffering and subsequent dreadful death, the movie at the same time contributes to the increasing Islamophobia that started after 9/11 terrorist attacks and the unsavory polemics that ensued. Mr. Khan ends up being punished and, with his funeral, Muslims immediately disappear and die out. Even Sam Khan seems to forget his religion at the end of the film, where everybody seems to be a Hindu citizen in the homeland. Besides alluding to the growing wave of Islamophobia, the drama of partition is also inevitably evoked with the over–present religious conflict. Partition is not such a distant memory, but remains unsolved; it is still a tragedy in the backdrop. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 ENDING, NATION AND PARTITION The most significant difference between Second Generation and Life Goes On and King Lear has to do with the ending. The tragic dénouement is transformed into a happy ending in which all the problems are solved. Jon Sen’s film finishes with an understanding of most of the characters. The Goneril figure Pria Sharma reveals that she was the first one to see her mother’s corpse, which has influenced her existence. Rina gets divorced since she cannot continue in an unhappy and meaningless marriage. But the film focuses most on Mr. Sharma, Heere and Sam. They abandon London and decide to go back to the home- land, which contributes to their acquisition of a deeper sense of belonging to a culture. Mr. Sharma’s imaginary national identity finally finds its meaning in the motherland, erasing all his nostalgia and hallucinations. Throwing his wife’s ashes to the Ganges River and going back to India help him to solve his drama. But the serial equally promotes a homogeneous, conservative – and even difficult – representation of migrant identities with the ending provided for Heere and Sam; their cultural hybridity intermingling Western and Eastern traits is controversially reduced at the end. Second Generation problematically favours the idea that hybridity is impossible and the only alternative Heere and Sam have is to embrace Indianness. All their identity conflicts will die out once they return to India and escape the “corruptions” of British life and diaspora. “What is striking here is a significant change in Heere’s character: we are left with Heere bargaining in the local market, followed by a domestic scene in which she is serving food to her father and husband–to–be (Dengel–Janic and Eckstein 2008: 56). The twofold nature of British–Asian cultural identity no longer becomes manifest in Second Generation’s ending. The feminist stance from the beginning paves the way for the typical insertion of the woman into a patriarchal culture and society. Gender and nation are linked once again but, unfortunately, there is no challenge of patriar- chal values, but even reinforcement of them, causing more blatant stereotypes and clichés. Life Goes On presents a parallel development of the ending, since the tragicomedy smooths the way for a happy resolution. 45 See Richard Burt, “All that Remains of Shakespeare in Indian Film” in Dennis Kennedy and Yong Li Lan’s Shakespeare in Asia: Contemporary Performance. Cambridge: Cambridge University Press to read about a considerable number of adaptations that include stagings of Shakespeare within the films. MUSLIM CHARACTERS So tough was partition for Indian citizens that the solution Second Generation offers is to go back to the homeland, leave in peace there but erasing all the Muslim traces on the way.i Datta’s film Life Goes On includes one main Muslim character, who also happens to be the boyfriend of Cordelia’s counterpart Dia. If the cliché of forbidden unions was certainly explored in Second Generation, it was given a greater insight in Life Goes On. In fact, there is an emphasis on the strong Hindu feelings the patriarch Dr Sanjay Banerjee has. He belongs to the Hindu community in the diaspora and seems to have a good reputation there for he is highly respected. Dr. Sanjay Banerjee still recalls partition, which he regards as an upsetting and painful experience in his life and frequently has nightmares associated with it. Flashbacks in the narrative depict the horror and violence experienced during partition. Dr. Banerjee’s attitude towards Muslims is therefore related to the trauma of partition. Aware of this, Dia/Cordelia tries to conceal from her father the fact that her boyfriend is a Muslim doctor. In a flashback, the audience discovers that, as usual, it was Manju the first to learn about Dia’s union to Imtiaz, but she died too soon to know about Dia’s pregnancy. It is precisely the discovery of the relationship between Dia and Imtiaz what causes his temporary madness; he is just unable to forgive the agony of partition and sees Muslims as active participants of his current distress. Yet, as will be seen, Shakespeare’s King Lear allows Dr. Sanjay Banerjee to overcome his fears, accept his daughter’s Muslim boyfriend and, overall, change radically his feelings towards Muslims and rediscover his identity in the host country. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 63 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago ENDING, NATION AND PARTITION Perhaps one of the main differences between the two rewritings of King Lear has to do with the inclusion of a play–within–the–film, which follows in the footsteps of other post–colonial revisions of Shakespeare’s plays, namely Shakespeare Wallah (James Ivory 1965) or In Othello (Roysten Abel 2001).45 Dia actually played the role of Cordelia in a play within the film. Trained at the Royal Shakespeare Company (RSC), Dia and the rest of the actors of the production render Shakespeare with a perfect British accent. Thus, the production seems to propose an identification of Shakespeare with high–brow culture, since he is still presented as a British icon. Only two moments are shown of the production: the well–known beginning in which Cordelia utters the word “nothing” and immediately after, the performance moves towards Cordelia’s forgiveness of her father and subsequent union and reconciliation. The production’s accent on parent–child relationship is crucial for the development of the story, and advances the happy ending. As Paromita Chakravarti claims, “the reconciliation scene between Lear and Cordelia helps to resolve bitter conflicts between Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 Rewriting King Lear in a Diasporic Context, Rewriting... 64 Rosa María García Periago Sanjay and Diya, opening up a space for intergenerational dialogue and sharing” (2014: 140). It is precisely the emphasis on the reconciliation in the production what paves the way for an exploration of the conflict between father and daughter and the successful resolution. pl g Life Goes On’s happy ending has a myriad of nuances and interesting interpretations when looked at it through the lens of South Asian popular culture. Manju’s funeral entails not only the harmony between father and daughter but also within the whole family. The forgiveness of Alok’s affair with Manju and the acceptance of Dia’s Muslim boyfriend gear towards a blissful finale. But it is necessary to deepen into the complexities of this conclusion. Dr. Sanjay Banerjee, by forgiving his wife and his friend in the present and by consenting to his daughter’s Muslim partner, is also reconciling himself with his traumatic past associated with partition. It is worth pointing out how the Lear figure even connected his daughter’s boyfriend Imtiaz with his childhood friend also called Imtiaz, whom he had seen killing a neighbor during partition. Therefore, admitting the present implicitly means acknowledging the past; past and present are thus reconciled in Lear’s world. ENDING, NATION AND PARTITION The problems, conflicts and tensions faced by the Lear figure in his mother country during partition come to an end during Manju’s funeral. If Manju stands for the homeland, her death can be inter- preted as the “death” of conflicts within the diasporic individuals and even the “re–birth” of the nation, as death in Hinduism is associated with re–birth. Sen and Datta’s choice of King Lear as a vehicle through which to portray the tensions faced by diasporic beings shows that their rewritings of the tale are going to be full of nuances and implications. Rewriting Lear is tantamount to rewriting the nation. Yet, the conclusions provided by the movies differ considerably and are even contradictory. While Second Gen- eration’s ending seems to condemn hybridity, Life Goes On favours it and even promotes it. Second Generation focuses on the return to the homeland of Sharma–ji, Heere and Sam erasing their cultural differences and even encouraging cultural purity – as if that were possible. In contrast, Life Goes On ends with all the main characters together in London, the third space, which allows for reconciliation with the host country, the homeland, but, above all the self. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 CONCLUSION Traces of the underlying Indian history remained beneath the surface of Life Goes On and Second Generation. Both are based on King Lear, which becomes the perfect allegory for partition since it deals with the division of the kingdom. The hidden presence of parti- tion in these two movies and the reasons for its non–explicitness, have to tell us as much about diasporic Indians and their traumas as about Shakespeare. If Shakespeare has been frequently associated with colonial India, partition is a product of colonial legacy. Thus, rewriting Shakespeare and making him postcolonial entails rewriting the past. The film adaptations echo Indian historical and cultural past and diasporic fears to finally rewrite and reinterpret Shakespeare, King Lear and, above all, their own nation. In Life Goes On and Second Generation, Shakespeare’s King Lear is transformed into a story with a happy ending, and becomes a tool to create unions above all divisions and heals wounds created by a collective amnesia about partition. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 65 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago REFERENCES 36 CHOWRINGHEE LANE. Dir. Aparna Sen. India. 1981. ADELMAN, J. 1992. Suffocating. Mothers: Fantasies of Maternal Origin in Shake- speare’s Plays, Hamlet to The Tempest. New York: Routledge. ALSULTANY, E. and E. SHOHAT. 2013. Between the Middle East and the Americas. The Cultural Politics of Diaspora. Michigan: University of Michigan. BEND IT LIKE BECKHAM. Dir. Gurinder Chadha. UK. 2003. BEND IT LIKE BECKHAM. Dir. Gurinder Chadha. UK. 2003. BHABHA, H. 1993. The Location of Culture. New York: Routledge. BHATTACHARYA, N. 2011. “Imagined Subjects: Law, Gender and Citizenship in Indian Cinema.” Bollywood and Globalization: Indian Popular Cinema, Nation, and Diaspora. Ed. Rajeshwari V. Pandharipande. London, New York and Delhi: Anthem Press. 129–144. BURT, R. 2010. “All that Remains of Shakespeare in Indian Film.” Shakespeare in Asia: Contemporary Performance. Ed. Dennis Kennedy and Yong Li Lan. Cambridge: Cambridge University Press. 73–108. CARRUTHERS, I. 2009. “Fooling with Lear”. Replaying Shakespeare in Asia. Eds. Poonam Trivedi and Minami Ryuta. New York and London: Routledge. 97–118. CHAKRAVARTI, P. 2014. “Interrogating ‘Bollywood Shakespeare’: Reading Rituparno Ghosh’s The Last Lear”. Bollywood Shakespeares. Eds. Craig Dionne and Parmita Kapadia. London and New York: Palgrave Macmillan. 127–145. CORSON, K. 2015. “Little Indias: Diasporic Communities in the US and the Con- sumption of Bollywood.” Shaping Indian Diaspora: Literary Representations and Bollywood Consumption. Eds. Cristina M. Gámez–Fernández and Deena Dwivedi. London: Lexington Books. 133–146. DAS, S. K. 2005. “Shakespeare in Indian Languages.” India’s Shakespeare: Translation, Interpretation and Performance. Eds. Poonam Trivedi and Dennis Bartholomeusz. Newark: University of Delaware Press. 47–73. DATTA, S. 2000. “Globalisation and Representations of Women in Indian Cinema”, Social Scientist 28.3/4: 71–82. DENGEL–JANIC, E. and L. ECKSTEIN. 2008. “Bridehood Revisited: Disarming Concepts of Gender and Culture in Recent Asian British Film”. Multi–ethnic Britain 2000+ New Perspectives in Literature, Film and the Arts. Eds. Lars Eckstein, Barbara Korte, Eva Ulrike Pirker and Christoph Reinfandt. Amsterdam: Rodopi. 45–64. DENGEL–JANIC, E. and J. ROERING. 2008. “Re–Imaging Shakespeare in Second Generation – A British– Asian Perspective on Shakespeare’s King Lear.” Drama and Cultural Change: Turning Around Shakespeare. Eds. Matthias Bauer and Angelika Zirker. Trier: WVT. 211–219. EAST IS EAST. Dir. Damien O’Donnell. UK. 1999. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 66 FOOD FOR RAVENS. Dir. Trevor Griffiths. UK. 1997. FOOD FOR RAVENS. Dir. Trevor Griffiths. UK. 1997. FÖLDVÁRY, K. 2013. “Postcolonial Hybridity: The Making of a Bollywood Lear in London,” Shakespeare 9.3: 304–312. GANDHI. Dir. Richard Attenborough. UK. 1982. GARCÍA–PERIAGO, R.M. 2015. “English Shakespeares in India: 36 Chowringhee Lane and The Last Lear,” Borrowers and Lenders: The Journal of Shakespeare and Appropriation 9. GREENHALGH, S. and R. SHAUNNESSY. 2006. “Our Shakespeares: British Televi- sion and the Strains of Multiculturalism.” Screening Shakespeare in the Twenty–First Century. Eds. Mark Thornton Burnett and Ramona Wray. Edinburgh: Edinburgh University Press. 90–112. HARRIS, G. 2006. Beyond Representation: Television Drama and the Politics and Aesthetics of Identity. Manchester: Manchester University Press. HATCHUEL, S. 2011. Shakespeare and the Cleopatra/Caesar Intertext: Sequel, Con- flation, Remake. Lanham: Fairleigh Dickinson University Press. IN OTHELLO. Dir. Roysten Abel. India. 2001. KABHIE KUSHI KABHI GHAM. Dir. Karan Johar. India. 2001. KAHN, COPPÉLIA. 1985. “The Absent Mother in King Lear.” Rewriting the Renais- sance: The Discourses of Sexual Difference. Eds. Margaret Ferguson, Maureen Quilligan, and Nancy Vickers. Chicago: University of Chicago Press. 33–49. KAL HO NAA HO. Dir. Nikhil Alvani. India. 2003. LIFE GOES ON. Dir. Sangeeta Datta. UK. 2009. LOOMBA, A. 1998. “Local–manufacture made–in–India Othello fellows’: Issues of race, hybridity and location in post–colonial Shakespeares’”. Post–colonial Shake- speares. Eds. Ania Loomba and Martin Orkin. London and New York: Routledge. 143–163. MARINO, A. 2013. “Cut’n’mix King Lear: Second Generation and British–Asian Identities.” Shakespeare and Conflict: A European Perspective. Eds. Carla Dente and Sara Soncini. London and New York: Palgrave Macmillan. 170–183. MOTHER INDIA. Dir. Mehboob Khan. India. 1957. MUIR, K. 1966. “Madness in King Lear”, Shakespeare Survey 13: 30–40. NAFICY, H. 2001. An Accented Cinema: Exilic and Diasporic Filmmaking. Princeton: Princeton University Press. QUAYSON, A. and G. DASWANI. 2013. A Companion to Diaspora and Transnation- alism. Wiley Blackwell. SECOND GENERATION. Dir. Jon Sen. UK. 2003. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 67 Rewriting King Lear in a Diasporic Context, Rewriting... Rosa María García Periago 67 SHAKESPEARE, W. 1986. Complete Works. Edited by Stanley Wells et al. Oxford: Clarendon Press. SHAKESPEARE, W. 1986. Complete Works. Edited by Stanley Wells et al. Oxford Clarendon Press. SHAKESPEARE WALLAH. Dir. James Ivory. UK. 1965. SINHA, M. 2006. Specters of Mother India. The Global Restructuring of an Empire. Duke: Duke University Press. SPIVAK, G. C. 2003. A Critique of Postcolonial Reason. Cambridge: Harvard Uni- versity Press. THAKUR, G. 2011. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67 FOOD FOR RAVENS. Dir. Trevor Griffiths. UK. 1997. “Globalization and the Cultural Imaginary: Constructions of Subjectivity, Freedom and Enjoyment in Popular Indian Cinema.” Bollywood and Globalization: Indian Popular Cinema, Nation and Diaspora. Eds. Rini Bhattacha- rya Mehta and Rajeshwari Pandharipande. London, New York and Delhi: Anthem Press. 75–92. THE LAST DAYS OF THE RAJ. Dir. Carl Hindmarch. UK. 2007. THE LAST LEAR. Dir. Rituparno Ghosh. India. 2007. TÖLÖYAN, K. 2010. “Beyond the Homeland: From Exilic Nationalism to Diasporic Transnationalism”. The Call of the Homeland: Diaspora Nationalisms, Past and Present. Eds. Allon Gal, Athena S. Leoussi and Anthony D. Smith. Leiden and Boston: Brill. 27–46. TRIVEDI, P. 2010. “Shakespeare in India: History of King Lear in India”. MIT Global Shakespeares. Video and Performance Archives. http://globalshakespeares.mit.edu/ blog/2010/06/22/history–of–king–lear–in–india/ Last accessed 07/08/2015. TWELFTH NIGHT. Dir. Tim Supple. UK. 2002. VIRDI, J. 2003. The Cinematic ImagiNation: Indian Popular Films as Social History. New Brunswick, New Jersey and London: Rutgers University Press. Odisea, nº 17, ISSN 1578–3820, 2016, 51–67
https://openalex.org/W4390077808	https://ijape.iaescore.com/index.php/IJAPE/article/download/20827/13116	English	null	Techno-economic assessment and wind energy potential of Nagad in Djibouti	International Journal of Applied Power Engineering	2,024	cc-by-sa	8,983	International Journal of Applied Power Engineering (IJAPE) Vol. 13, No. 1, March 2024, pp. 91~101 ISSN: 2252-8792, DOI: 10.11591/ijape.v13.i1.pp91-101 International Journal of Applied Power Engineering (IJAPE) Vol. 13, No. 1, March 2024, pp. 91~101 ISSN: 2252-8792, DOI: 10.11591/ijape.v13.i1.pp91-101 International Journal of Applied Power Engineering (IJAPE) Vol. 13, No. 1, March 2024, pp. 91~101 ISSN: 2252-8792, DOI: 10.11591/ijape.v13.i1.pp91-101 International Journal of Applied Power Engineering (IJAPE) Vol. 13, No. 1, March 2024, pp. 91~101 ISSN: 2252-8792, DOI: 10.11591/ijape.v13.i1.pp91-101 91  91 Techno-economic assessment and wind energy potential of Nagad in Djibouti Abdoulkader Ibrahim Idriss1, Ramadan Ali Ahmed2, Hamda Abdi Atteyeh1, Abdou Idris Omar1, Tahir Cetin Akinci3 Abdoulkader Ibrahim Idriss1, Ramadan Ali Ahmed2, Hamda Abdi Atteyeh1, Abdou Idris Omar1, Tahir Cetin Akinci3 1Department of Electrical and Energy Engineering, Faculty of Engineering, Université de Djibouti, Djibouti, Djibouti 2Department of Electrical and Energy Engineering, Institute of Industrial Technology, Université de Djibouti, Djibouti, Djibout Abdou Idris Omar1, Tahir Cetin Akinci3 1Department of Electrical and Energy Engineering, Faculty of Engineering, Université de Djibouti, Djibouti, Djibouti 2Department of Electrical and Energy Engineering, Institute of Industrial Technology, Université de Djibouti, Djibouti, Djibouti 3Department of Electrical Engineering, Istanbul Technical University, Istanbul, Turkey Article Info Article history: Received Jun 12, 2023 Revised Jul 27, 2023 Accepted Aug 2, 2023 Keywords: Economic analysis Nagad Wind power Wind speed Wind turbine Keywords: Economic analysis Nagad Wind power Wind speed Wind turbine This is an open access article under the CC BY-SA license. This is an open access article under the CC BY-SA license. Corresponding Author: Abdoulkader Ibrahim Idriss Department of Electrical and Energy Engineering, Faculty of Engineering, Université de Djibouti Street Djanaleh, BP 1904, Djibouti, Djibouti Email: abdoulkader_ibrahim_idriss@univ.edu.dj Corresponding Author: Abdoulkader Ibrahim Idriss Department of Electrical and Energy Engineering, Faculty of Engineering, Université de Djibouti Street Djanaleh, BP 1904, Djibouti, Djibouti Email: abdoulkader_ibrahim_idriss@univ.edu.dj ABSTRACT The use of small scaled horizontal and vertical axis wind turbines in urban installation is increasing over the world. However, in Djibouti, the latter is still in the development phase. The paper presents a techno-economical analysis and wind energy potential for the period of five years (2015-2019) in Nagad based on actual measured wind speed data collected every 10 min at 10 m height. The energy pattern factor method has been used to estimate the Weibull parameters. With this method, the mathematical complexity is reduced with a minimization of the error at any heights and locations when calculating the wind power density. At 50 m height, the shape parameter showed a small variation for different periods. The scale parameter values of 7.78 m/s and 4.8 m/s were obtained in the hot and cold seasons, respectively. The results showed that the Nagad site is suitable for wind power development. According to the economic viability, RX30, Vestas V20, Enercon, Nordex N27, and Vestas V44 wind turbines are recommended for the Nagad site due to their low energy price ranging from 0.05$/kWh to 0.31$/kWh. This is 2-6 times cheaper than the average local tariff of electricity in Djibouti. Corresponding Author: Abdoulkader Ibrahim Idriss Department of Electrical and Energy Engineering, Faculty of Engineering, Université de Djibouti Street Djanaleh, BP 1904, Djibouti, Djibouti Email: abdoulkader_ibrahim_idriss@univ.edu.dj Journal homepage: http://ijape.iaescore.com 1. INTRODUCTION They showed that the studied sites receive the greatest amount of solar radiation compared to other places in the world with the global radiation value of 2,898 kWh/ (m². year). In addition to that, the sites have an encouraging potential to develop a hybrid power system for any application. Queen et al. [15] have investigated the advantages of integrating natural source of energy from the renewable energies to the prevailing electric power systems. For the two studied standard IEEE system (IEEE 14 bus and IEEE 30 bus), the price of electricity acquired from the grid is lowered by 30% with incorporated renewable energy systems. In addition, Sakhrieh et al. [16] provided a techno-economical study of the optimized hybrid system includes photovoltaics, a biogas generator, batteries, and a diesel generator in the rural sites. The levelized cost of energy of 0.06$/kWh and a net present cost of 2,100,000$ have been estimated for the optimized hybrid system. In Egypt, Abdelrahman et al. [17] have examined the wind energy potential and the economic feasibility to develop the first wind farm at Elkharga Oasis. With 50 MW wind farm at Elkharga, they concluded that the cost of energy is much cheaper by > 50% than the current tariff in Egypt. Daoudi et al. [18] have demonstrated the economic viability of the two onshore wind farms located in the province of Tantan. The results have shown that the two wind farms have a good potential to develop the wind farm with the cost of the production values of 3.45$/kWh and 3.87$/kWh in Tantan-1 and Tantan-2, respectively. The total potential of wind energy is estimated at 1,300 GW in the Sub-Saharan African regions. The use of micro and small wind turbines in urban and rural installations is increasing over the world, and it is still in the developing phase in East Africa. Djibouti imports all its energy needs while the country has a high potential in renewable energies, untapped until now. Recent surveys conducted as part of the strategy to combat poverty reveals that 49.7% of sedentary households (99.5% of which are in urban area) use electricity to power lighting systems with an average consumption estimated at 228 kWh/year per capita. However, this prominence of energy for Djiboutian households’ masks huge disparities in access linked to the availability and high production costs of around 52 DJF/kWh (0.32$/kWh set by Electricité de Djibouti). 1. INTRODUCTION The future of the East African population is uncertain due to two major challenges which are the lack of access to modern energy services and the vulnerability associated with climate change. These challenges are at the origin of the reflections engaged to reinvent the energy future by carrying out the necessary transitions towards energy systems that allow responsible growth and reconcile economic development, environmental protection, and the reduction of inequalities. Access to electricity is a key indicator of a country's level of development. Generation of electricity from renewable energy such as wind, sun (solar thermal and photovoltaic), hydro, and geothermal can play a major role in electricity production in Eastern African countries. Several programs are defined by the Sustainable Energy for All [1], [2] as well as the renewable capacity statistics from the International Renewable Energy Agency [3] to offer a new form of planning centered on needs and to redefine the energy model of the region and the associated policies. Wind energy is the most non-polluting, sustainable and can potentially make a significant contribution to developing countries with poor infrastructure for power generation. The cost, performance, and reliability of renewable energy technologies are significantly Journal homepage: http://ijape.iaescore.com  92 ISSN: 2252-8792 improved to the point that they can now compete with conventional energy sources in several applications [4], [5]. Numerous studies have been done to assess the wind speed characteristic and wind power potential in the world [6], [7] and especially in Africa [8]–[10]. According to [11], the levelized cost of electricity and net present cost in Yanbu region of Saudi Arabia are estimated as (0.0885$/kWh and 23.8$) for Enercon E-126 EP4 wind turbine that leads their corresponding values of (0.142$/kWh and38.3$) for WES 30 turbine. Abd in [12] has provided a strategy based on a weather change to find the optimal designing and modelling for four types of wind energy conversion system models using HOMER software. The study has focused on the technical, economic, and profitability calculation for any renewable energy system. In Saswat et al. [13], India have demonstrated the effectiveness of the hybrid PV/solar/wind power system, which is given the best and most efficient alternative to conventional energy sources. Idriss et al. [14] conducted the potential of wind and solar energy in two rural sites in Djibouti which are Herkalou and Lake Assal. Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101 1. INTRODUCTION The country survived until May 2011 on electricity produced from imported petroleum products. Consequently, high production costs are a barrier to access to energy sources for the poorest of the population. In the Republic of Djibouti, a few studies have been conducted to analyze wind power generation [19], more attention should concentrate on the resource of wind speed potential in the urban, peri-urban, and rural areas of the country. During the implementation of the wind project, it is recommended to evaluate the character of the wind speed data, the feasibility depending on the location, the characteristics of wind turbines (horizontal or vertical axis models, cut-in and cut-out velocity, rated velocity, power energy output, and capacity factor), the cost analysis (initial cost, maintenance costs during the lifetime of the turbine), and the energy potential before any wind energy system. Researchers have studied the integration of micro, small, and mid-sized wind systems using several statistical and probability distribution analyses of wind speed data [20]–[22]. According to the International Electrotechnical Commission (IEC:614-00-12), the Weibull 2-parameters are becoming a standard indicator of probability distribution function (PDF) to describe the wind characteristics [23]. Others have investigated by using several numerical methods for fitting the wind speed data as well as the graphical (GM), the moment (MM), and the energy pattern factor (EPFM) methods [24]–[26]. In this work, the wind potential assessment, and the economic feasibility of using commercially available wind turbines are evaluated for the Nagad peri-urban site, located in the southern part of Djibouti- city. Potential Djibouti sites for wind energy generation have not been thoroughly explored as the development of wind project continues to be hampered by the lack of reliable and accurate wind datasets in many parts of Djibouti, as well as the lack of both qualified human resources and accessibility of mountainous and hostile areas of the northern and southern regions for scientists and researchers. Therefore, the objective of this paper is to evaluate the cost of energy production from micro, small and mid-sized wind turbines which will serve as a benchmark in the national energy plan. The statistics and costs provided by this study have enabled government officials and potential investors to propose a strategy to reduce the weight of energy consumption bills in the household budget and to make energy more accessible for all. 2.1. Study area: Nagad site description and wind data 2.1. Study area: Nagad site description and wind data Nagad is a coastal site (at 11.3124° N and latitude 43.0739° E, altitude 8 m) located near to the International Airport of Djibouti. The site is situated in the south of Djibouti city and is identified as a peri- urban area. Due it’s to rather close proximity to the equator, Djibouti is classified as a hot and humid country. This type of climate receives the highest amount of solar radiation compared to other regions. The high level of solar radiation also causes high air temperatures. The meteorological actual wind data were collected and analyzed hourly and every 10 min taken by a mast at 10 m height for a period of five years from 2015 to 2019 using Vantage Pro2 equipment. It includes a mast, an anemometer, a wind vane, a thermometer, and a barometer. The equipment was installed in July 2014 on the roof of the Department of Electrical and Energy in the Faculty of Engineering. Daily average wind speed, wind directions, and temperatures were measured in 10 min time intervals, at 10 m height above the ground level. Figure 1(a) displays a histogram that illustrates the monthly average temperature (including minimum and maximum). This graph highlights the cold and hot seasons. The period extends from October to April where the climate is rather pleasant, with average temperatures between 18 °C and 34 °C during the day and 17 °C and 32 °C at night. The second season extends from May to September, with very high temperatures. Ranging from 24 °C to 46 °C during the day, and at night it ranges from 20 °C to 42 °C. The difference between the minimum and maximum temperatures in the hot summer months is slightly higher than in the cooler months. Figure 1(b) shows a monthly average contour map of diurnal mean wind speeds over the studied period. The average wind speeds are higher during the hot period (5-7.5 m/s from 12 p.m. to 8 p.m.) than during the coldest one (4-6 m/s from 12 p.m. to 8 p.m.). The wind characteristics information for a specific site can be combined visually in a wind rose. The latter allows the three essential pieces of information to be grouped in a single graph: wind speed, direction, and frequency by sector. Int J Appl Power Eng - Estimating the wind energy, for the first time, in Nagad owing to its windy and less strict topography locations - Estimating the wind energy, for the first time, in Nagad owing to its windy and less strict topography locations Illustrating the performance comparisons of nine wind turbines with various technologies (horizontal an vertical) and characteristics - Evaluating the economic viability by analysing the cost of energy production - Evaluating the economic viability by analysing the cost of energy production The rest of this paper is subdivided into four sections. Section 2 presents the methodology including the site description and the wind data analysis are presented. In section 3, the results are discussed. Finally, the section 4 presents the conclusion and the recommendation of this study. 1. INTRODUCTION The fundamental contributions and research originality of this paper can be summarized as: Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101 ISSN: 2252-8792 93  Int J Appl Power Eng 2.1. Study area: Nagad site description and wind data Figure 1(c) reveals an omnidirectional wind with a dominant wind from the west (5%, 9 m/s) generated by the Ethiopian highlands in summer. This wind which is called "Khamsin" is dry and hot. During the cold season, the city is swept by easterly winds (11.5%, 11 m/s) generated by the trade winds, humid winds from Arabia, and the Gulf of Aden. The annual mean wind speed is 4.28 m/s during the period considered, here 5 years. These representations are the first approach, allowing to have a quick overview of the wind profile of a site for a wind energy application. Further analyses of wind data were performed in Figure 1(d). The EPFM method is adopted in this study to describe the power density function (PDF) and the cumulative density function (CDF). The EPFM method requires less computation, and easier implementation to calculate the wind power density with less error at any height and location. Based on the latter method, the scale and shape parameters are estimated, which will be discussed in section 3. 2.2.1. Weibull parameters and energy pattern factor method (EPFM) 𝑘= 1 + 3.69 (𝐸𝑝𝑓)2 𝑐= 𝑣 𝛤 (1+1 𝑘) 𝑘= 1 + 3.69 (𝐸𝑝𝑓)2 (4) 𝑐= 𝑣 𝛤 (1+1 𝑘) (5) 𝑘= 1 + 3.69 (𝐸𝑝𝑓)2 (4) 𝑐= 𝑣 𝛤 (1+1 𝑘) (5) (4) (5) With Γ(.) is the gamma function. With Γ(.) is the gamma function. With Γ(.) is the gamma function. With Γ(.) is the gamma function. (a) (b) (c) (d) Figure 1. Nagad site wind data covering 2015-2019 (a) monthly mean temperature of Nagad’s wind speed data, (b) monthly (and daily diurnal) contour map of the hourly mean wind speed data, (c) wind rose diagram, and (d) PDF and CDF curves compared to the observed wind data (b) (a) (b) (a) (c) (d) (c) (d) Figure 1. Nagad site wind data covering 2015-2019 (a) monthly mean temperature of Nagad’s wind speed data, (b) monthly (and daily diurnal) contour map of the hourly mean wind speed data, (c) wind rose diagram, and (d) PDF and CDF curves compared to the observed wind data 2.2.1. Weibull parameters and energy pattern factor method (EPFM) Several mathematical models have been used to assess the wind speed. In this investigation, the 2-parameters Weibull function is chosen, because it gives a good fit and better measurement of probability distribution function than other statistical methods [27]. The PDF and the CDF of the 2-parameters Weibull can be expressed as: 𝑓(𝑣) = ( 𝑘 𝑐) ( 𝑣 𝑐) 𝑘−1 𝑒𝑥𝑝[−( 𝑣 𝑐) 𝑘 ] (1) 𝑓(𝑣) = ( 𝑘 𝑐) ( 𝑣 𝑐) 𝑘−1 𝑒𝑥𝑝[−( 𝑣 𝑐) 𝑘 ] (2) 𝑓(𝑣) = ( 𝑘 𝑐) ( 𝑣 𝑐) 𝑘−1 𝑒𝑥𝑝[−( 𝑣 𝑐) 𝑘 ] 𝑓(𝑣) = ( 𝑘 𝑐) ( 𝑣 𝑐) 𝑘−1 𝑒𝑥𝑝[−( 𝑣 𝑐) 𝑘 ] (1) (2) (2) Techno economic assessment and wind energy potential of Nagad (Abdoulkader Ibrahim Idriss) where v is the wind speed, c (scale) and k (shape) are the Weibull parameters, respectively. Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss) where v is the wind speed, c (scale) and k (shape) are the Weibull parameters, respectively. Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss) where v is the wind speed, c (scale) and k (shape) are the Weibull parameters, respectively. where v is the wind speed, c (scale) and k (shape) are the Weibull parameters, respectively where v is the wind speed, c (scale) and k (shape) are the Weibull parameters, respectively.  94 ISSN: 2252-8792 To calculate k and c values, several methods are proposed in the literature [28], [29]. In this work, the EPFM is selected because it is a valuable and direct method that does not demand repetitions, simple and easy to tool and formulate. 𝐸𝑝𝑓 is defined the as ratio between the mean of cubic wind speed (𝑣3 ̅̅̅) to the cube of mean wind speed (𝑣3). The 𝐸𝑝𝑓 can be calculated as (3) [21]. 𝐸𝑝𝑓= 𝑣3 ̅̅̅̅ 𝑣3 = 1 𝑛∑ 𝑣𝑖3 𝑛 𝑖=1 (1 𝑛∑ 𝑣𝑖 𝑛 𝑖=1 ) 3 (3) 𝐸𝑝𝑓= 𝑣3 ̅̅̅̅ 𝑣3 = 1 𝑛∑ 𝑣𝑖3 𝑛 𝑖=1 (1 𝑛∑ 𝑣𝑖 𝑛 𝑖=1 ) 3 (3) Then, the values of k and c are determined by (4) and (5). Then, the values of k and c are determined by (4) and (5). 2.2.2. Wind speed extrapolation with height 𝑊𝑃𝐷= 𝑃 (𝑣) 𝐴 = 1 2 𝜌 𝑐3𝛤 (1 + 3 𝑘) (𝑊/𝑚²) (11) (11) While wind energy density (WED) is defined as the power density over a period, energy density can be evaluated using (12). Where T is a period of time (hour). e wind energy density (WED) is defined as the power density over a period, energy density can be ated using (12). Where T is a period of time (hour). 𝑊𝐸𝐷= 𝑊𝑃𝐷× 𝑇 (𝑊ℎ/𝑚²) (12) 𝑊𝐸𝐷= 𝑊𝑃𝐷× 𝑇 (𝑊ℎ/𝑚²) 𝑊𝐸𝐷= 𝑊𝑃𝐷× 𝑇 (𝑊ℎ/𝑚²) (12) To evaluate the efficient and best-suited wind turbine for the Nagad site, nine commercial turbines of the site were selected. Average power output (𝑃𝑒.𝑎𝑣𝑒), capacity factor (CF), and accumulated annual energy (AEP) are the important performance parameters to study wind speed installed in a given site. The 𝑃𝑒.𝑎𝑣𝑒 and CF of a wind turbine are calculated by (13) and (14). 𝑃𝑒.𝑎𝑣𝑒= 𝑃𝑒𝑅( 𝑒−(𝑣𝑐 𝑐) 𝑘 −𝑒−(𝑣𝑟 𝑐) 𝑘 (𝑣𝑟 𝑐) 𝑘 −(𝑣𝑐 𝑐) 𝑘 −𝑒 −( 𝑣𝑓 𝑐) 𝑘 ) (13) (13) ( 𝑐) ( 𝑐) 𝐶𝐹= 𝑃𝑒.𝑎𝑣𝑒/𝑃𝑒𝑅 (14) 𝐶𝐹= 𝑃𝑒.𝑎𝑣𝑒/𝑃𝑒𝑅 (14) 𝐶𝐹= 𝑃𝑒.𝑎𝑣𝑒/𝑃𝑒𝑅 𝐶𝐹= 𝑃𝑒.𝑎𝑣𝑒/𝑃𝑒𝑅 (14) Where 𝑣𝑐, 𝑣𝑟 and 𝑣𝑓 are the cut-in, rated, and cut-off wind speeds respectively. PeR is the rated electrical power of the wind turbine. The AEP is calculated over a period by using (15) [32]. Where t is the time, example for one a year period it has 8760 in hours. 𝐴𝐸𝑃= 𝐶𝐹× 𝑃𝑒𝑅 × 𝑡 (𝑘𝑊ℎ) (15) 𝐴𝐸𝑃= 𝐶𝐹× 𝑃𝑒𝑅 × 𝑡 (𝑘𝑊ℎ) (15) 2.2.2. Wind speed extrapolation with height The wind speed increases with heights, the power law is used for its simplicity in this study to extrapolate wind speed at different altitudes [30]. The wind speed variation with height can be mathematically expressed as (6). 𝑣(ℎ) = 𝑣0 ( ℎ ℎ0) 𝛼 (6) 𝑣(ℎ) = 𝑣0 ( ℎ ℎ0) 𝛼 (6) Where 𝑣0 is the wind speed at the initial height ℎ0 and α is the power-law commonly admitted to be 1/7. As the wind speed varies with height, similarly, the Weibull parameters c and k are also functioning of hub height. Using (7) and (8), the parameters can be calculated. Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101 ISSN: 2252-8792  95 ISSN: 2252-8792 95  Int J Appl Power Eng 𝑐(ℎ) = 𝑐0 × ( ℎ ℎ0) 𝑛 (7) 𝑘(ℎ) = 𝑘0 × (1−0.088𝑙𝑛( ℎ ℎ0)) (1−0.088𝑙𝑛( ℎ 10)) (8) 𝑐(ℎ) = 𝑐0 × ( ℎ ℎ0) 𝑛 𝑘(ℎ) = 𝑘0 × (1−0.088𝑙𝑛( ℎ ℎ0)) (1−0.088𝑙𝑛( ℎ 10)) (7) ℎ0 𝑘(ℎ) = 𝑘0 × (1−0.088𝑙𝑛( ℎ ℎ0)) (1−0.088𝑙𝑛( ℎ 10)) 𝑘(ℎ) = 𝑘0 × (1−0.088𝑙𝑛( ℎ ℎ0)) (1−0.088𝑙𝑛( ℎ 10)) (8) are Weibull parameters at ℎ0. The exponent 𝑛 is given by (9). Where 𝑐0 and 𝑘0 are Weibull parameters at ℎ0. The exponent 𝑛 is given by (9). 𝑛= (0.37−0.088𝑙𝑛(𝑐0)) (1−0.088 𝑙𝑛( ℎ 10)) (9) 𝑛= (0.37−0.088𝑙𝑛(𝑐0)) (1−0.088 𝑙𝑛( ℎ 10)) (9) 𝑛= (0.37−0.088𝑙𝑛(𝑐0)) (1−0.088 𝑙𝑛( ℎ 10)) (9) (9) 2.2.3. Evaluation of wind power density, energy density, capacity factor, power, and accumulated annual energy of the wind turbines The wind power is mathematically expressed as (10) [31]. 2.2.3. Evaluation of wind power density, energy density, capacity factor, power, and accumulated annual energy of the wind turbines The wind power is mathematically expressed as (10) [31]. 2.2.3. Evaluation of wind power density, energy density, capacity factor, power, and accumulated annual energy of the wind turbines The wind power is mathematically expressed as (10) [31]. 𝑃(𝑣) = 1 2 𝜌𝐴 𝑣3(W) (10) 𝑃(𝑣) = 1 2 𝜌𝐴 𝑣3(W) (10) The air density ρ is assumed to be 1.225 kg/m3 and the rotor area of the turbine is A (m²). The wind power density (WPD) in a selected site at a period with Weibull parameters can be expressed as (11). 3. RESULTS AND DISCUSSION To evaluate the availability of wind power at a site, the plot of the probability density function (PDF) curve is essential. The annual average k and c Weibull parameters are calculated using the probability distribution for the considerate wind speeds, which are then plotted with the cumulative density function curve see Figure 1(d). The most probable wind speed occurs at a speed of 2.5 m/s with a probability of 16.2%, while wind speed greater than 10 m/s shows a very low probability. Table 1 presents the monthly, seasonal, and yearly average of wind speed, the Weibull parameters (c and k), and the power and energy densities (WPD and WED) for the studied site at 10 m height. The results show that the whole year is divided into two seasons: cold (from October to April) and hot (from May to September) seasons. The highest mean wind speeds were observed in July and August with values of 4.99 m/s and 5.30 m/s, respectively. The lowest was obtained in April with a value of 3.38 m/s. The EPFM value of parameter k ranges from 1.71 in April to 1.99 in August while parameter c varies between 3.38 m/s in April to 5.30 m/s in August. The higher values of c are observed during the hot season (5.08 m/s) and lower during the cold season (4.59 m/s). The mean seasonal values of k parameter are observed to be 1.72 and 1.65 corresponding to the cold and hot seasons, respectively. In addition, the lower values of wind power densities and energies are obtained as 68.32 W/m² and 49.19 kWh/m² respectively in April while the higher values of 222.56 W/m² and 164.84 kWh/m² are computed in July, respectively. Also, the yearly mean power and energy densities are 111.49 W/m² and 976.65 kWh/m²/year, respectively. Table 1. 2.2.4. Wind energy cost analysis gy y In a given site, to assess the feasibility of the wind farm, the cost of energy is the most important parameter to evaluate the economic viability. For evaluating wind energy cost, several methods have been discussed in [33]. However, the present value of costs (PVC) method is commonly used, and it is adopted in this study. The PVC is given as (16). 𝑃𝑉𝐶= 𝐼+ 𝐶𝑜𝑚𝑟[ 1+𝑖 𝑟−𝑖] × [1 −( 1+𝑖 1+𝑟) 𝑙𝑡 ] −𝑆( 1+𝑖 1+𝑟) 𝑙𝑡 ($) (16) (16) The cost of produced energy (in kWh) by turbines at the respective location was estimated as 20% for the investment cost (I), the inflation (i) and interest (r) rates are 2% and 11.2% [34], with the lifetime (lt) of wind turbines is 20 years. The operation, maintenance, and repair cost (Comr) are 15% (minimum cost) and 25 % (maximum cost). The scrap value (S) is 10 %. The cost per kWh of electricity generated (UCE) can be determined by (17) [33]. Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss) ISSN: 2252-8792  96 𝑈𝐶𝐸= 𝑃𝑉𝐶 𝐴𝐸𝑃 ( $ 𝑘𝑊ℎ) (17) 3. RESULTS AND DISCUSSION Mean wind speed, Weibull parameters, WPD and WED over the considered period Period 𝒗 (m/s) k (-) c (m/s) WPD (W/m²) WED (kWh/m²) Jan 4.7 1.91 5.6 125.52 93.39 Feb 3.99 1.83 4.82 93.87 63.08 Mar 3.89 1.82 4.58 92.73 68.99 Apr 3.38 1.71 4.33 68.32 49.19 May 3.39 1.72 4.13 68.53 50.99 Jun 3.89 1.83 4.59 94.05 67.71 Jul 4.9 1.98 5.42 222.56 164.84 Aug 5.30 1.99 6.11 222.06 165.21 Sep 3.39 1.72 4.33 68.53 49.34 Oct 3.99 1.83 4.82 94.09 70.00 Nov 3.89 1.83 4.59 93.78 67.52 Dec 3.99 1.83 4.82 93.94 69.89 Cold season 4.10 1.72 4.59 94.88 482.78 Hot season 4.54 1.65 5.08 136.10 499.78 Annual 4.05 1.83 4.84 111.49 976.65 Table 1. Mean wind speed, Weibull parameters, WPD and WED over the considered period Period 𝒗 (m/s) k (-) c (m/s) WPD (W/m²) WED (kWh/m²) nthly mean wind speed and WPD at 10 m and 50 m 3.2. Monthly mean wind speed and WPD at 10 m and 50 m 3.2. Monthly mean wind speed and WPD at 10 m and 50 m The characteristics of the wind turbines are given in Table 3. Here, HAWT and VAWT mean the type of Horizontal and Vertical Axis Wind Turbines, respectively. Seven HAWTs namely Aeolos, Antaris, RX30, Vestas V20, Enercon, Nordex, and Vestas V44, and two VAWTs namely Turby and Sun surf were used to estimate the energy production for all the considered years and were analyzed. Table 3. Characteristics of selected commercial wind turbines Type Turbine model Vc (m/s) Vr (m/s) Vf (m/s) Rotor diameter (m) Hub height (m) PeR (kW) HAWT Antaris 2.8 13 25 4 12 3.50 Aeolos 3.5 12 25 8 18 10 RX30 3 10 25 16 18 30 Vestas V20 5 17.50 25 20 24 100 Enercon 3 12 25 16.20 30 55 Nordex N27 3 13 25 27 30 150 Vestas V44 5 17 20 44 50 600 VAWT Turby 3.5 12 14 2 15 2.5 Sun surf 1.8 8 25 9 18 10 The annual and seasonal mean variation of 𝑣̅, k, and c are shown at different hub heights corresponding to the wind turbines heights 12 m, 15 m, 18 m, 24 m, 30 m, and 50 m. At 50 m, the mean wind speed value is 5.37 m/s, k and c values are 1.96 and 7.41 m/s, respectively. The shape parameter shows a small variation for different heights and periods, whereas the scale parameter values vary between 5.30 m/s (at 12 m height) to 7.78 m/s at 50 m in the hot season. For the cold season, the c parameter variation ranges from 4.8 m/s (at 12 m height) to 7.15 m/s at 50 m, which is a sign that the site is suitable for wind applications. For all heights, it is noted that the hot season presents high wind characteristics than the cold season. Thus, Pe,ave, CF, and AEP given by these wind turbines are calculated using (13), (14), and (15). Additionally, Figure 3 shows the monthly variation of estimated CF and AEP of wind turbines. For the monthly analysis, the outcomes can be listed as: - The mean capacity factors and annual energy output values were recorded in August, as 28.97% and 215 kWh for Aeolos small-scale wind turbine, whereas the minimum values were 22.80% and 59.3 kWh for Antaris turbine, given in Figure 3(a). 3.2. Monthly mean wind speed and WPD at 10 m and 50 m The WPD classification is commonly used [35] and is established for classifying the wind potential. The wind power resource can be divided into 7 categories from poor (class=1) to excellent (class=5, 6, and 7) [7], [36]. According to Table 2 and as shown in Figure 2(a), for the July and August, Nagad site shows good wind energy with WPD values higher than 600 W/m² (𝑣 =7.2 m/s) at 50 m height. The mean WPD is higher than 350 W/m² in January while for the other months it ranges from 200 to 300 W/m² (with 𝑣 varies from 4.5 to 5.3 m/s). ( ) In Figure 2(b) the annual wind energy density was recorded at 339 W/m² (moderate, class 3 with 𝑣 =5.4 m/s) while it was measured at 297 W/m² (marginal, class 2 with 𝑣 =5.2 m/s) in the cold season and 400 W/m² (moderate, class 3 with 𝑣 =5.7 m/s) in the hot season. From the above classification, the Nagad site has reasonable wind energy resources and is suitable for harnessing wind turbines applications. Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101 Table 2. Classification of WPD [35], [36] Wind class At 10 m height At 50 m height Wind class At 10 m height At 50 m height 𝑣 (m/s) WPD (W/m²) 𝑣 (m/s) WPD (W/m²) 𝑣 (m/s) WPD (W/m²) 𝑣 (m/s) WPD (W/m²) 1 <4.4 <100 <5.6 <200 5 6.0-6.4 250-300 7.5-8.0 500-600 2 4.4-5.1 100-150 5.6-6.4 200-300 6 6.4-7.0 300-400 8.0-8.8 600-800 3 5.1-5.6 150-200 6.4-7.0 300-400 7 >7.0 >400 >8.8 >800 4 5.6-6.0 200-250 7.0-7.5 400-500 Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101 Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101 ISSN: 2252-8792 97  Int J Appl Power Eng (a) (b) Figure 2. Extrapolation for (a) monthly and (b) annual and seasonal results of mean wind speed and power density at selected heights (a) (b) (b) Figure 2. Extrapolation for (a) monthly and (b) annual and seasonal results of mean wind speed and power density at selected heights 3.3. Performance of nine small wind turbines: monthly, yearly and seasonal analysis In this part of the study, the performances of the nine different wind turbines were calculated. These turbines are selected based on their availability in the Republic of Djibouti and for their tower heights. 3.2. Monthly mean wind speed and WPD at 10 m and 50 m In August, the highest capacity factor was 32.33% for the Nordex turbine followed by 21.22% for Vestas V44 and 13.20% for Vestas V20. Also, for the same month, the minimum AEP values of about 3.60 MWh, 9.82 MWh, and 9.47 MWh are observed for the Nordex, Vestas V20, and Vestas V44 respectively. - The mean CF and AEP generated by different mid-sized wind turbines namely Nordex, Vestas V20, and Vestas V44 at three hub heights are estimated and shown in Figure 3(d). In August, the highest capacity factor was 32.33% for the Nordex turbine followed by 21.22% for Vestas V44 and 13.20% for Vestas V20. Also, for the same month, the minimum AEP values of about 3.60 MWh, 9.82 MWh, and 9.47 MWh are observed for the Nordex, Vestas V20, and Vestas V44 respectively. (a) (b) (c) (d) Figure 3. Monthly variation of CF and AEP for (a) Aeolos and Antaris, (b) Sun surf and Turby, (c) Enercon and RX30, and (d) Nordex, Vestas V44, and Vestas V20 (b) (a) (b) (a) ( ) (c) (d) (d) Figure 3. Monthly variation of CF and AEP for (a) Aeolos and Antaris, (b) Sun surf and Turby, (c) Enercon and RX30, and (d) Nordex, Vestas V44, and Vestas V20 The local average electricity cost value is 0.32$/kWh in Djibouti. Comparing the estimated tariffs obtained by the RX30, Vestas V20, Enercon, Nordex N27, and Vestas V44 wind turbines with the local cost 3.2. Monthly mean wind speed and WPD at 10 m and 50 m In this case, depending on the hub height of the wind turbine and PeR of the turbine, the Aeolos turbine produces maximum energy compared to Antaris. - Figure 3(b) depicts the mean CF and AEP for the VAWT turbines. The maximum mean CF value is obtained as 50.86% in August and a minimum value of 27.29% in April for the Sun surf turbine. The maximum mean AEP value is obtained as 378.4 kWh in August while the minimum value is computed as 196.6 kWh in April. The result shows that the Sun surf turbine generates more energy than the Turby wind turbine. - For the Enercon turbine, the mean AEP was maximum for August with 1515 kWh and least for April with 713 kWh as described in the Figure 3(c). The highest mean CF occurs in August with a value of 37.03% and the lowest mean CF occurs in April with a value of 17.99%. On observing the RX30 turbine, it can be - For the Enercon turbine, the mean AEP was maximum for August with 1515 kWh and least for April with 713 kWh as described in the Figure 3(c). The highest mean CF occurs in August with a value of 37.03% and the lowest mean CF occurs in April with a value of 17.99%. On observing the RX30 turbine, it can be Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss)  98 ISSN: 2252-8792 observed that the maximum AEP available in August was 899.2 kWh (with CF=40.28%) and the minimum mean AEP value of 402.8 kWh (with CF=18.64%) was present in April. - The mean CF and AEP generated by different mid-sized wind turbines namely Nordex, Vestas V20, and Vestas V44 at three hub heights are estimated and shown in Figure 3(d). In August, the highest capacity factor was 32.33% for the Nordex turbine followed by 21.22% for Vestas V44 and 13.20% for Vestas V20. Also, for the same month, the minimum AEP values of about 3.60 MWh, 9.82 MWh, and 9.47 MWh are observed for the Nordex, Vestas V20, and Vestas V44 respectively. - The mean CF and AEP generated by different mid-sized wind turbines namely Nordex, Vestas V20, and Vestas V44 at three hub heights are estimated and shown in Figure 3(d). 4. CONCLUSION AND RECOMMENDATIONS The performance comparisons and techno-economic analysis of HAWT and VAWT wind turbines in a hot climate have been studied at Nagad peri-urban site in Djibouti city. The wind data covering 2015-2019, recorded every 10 min, were analyzed for the wind speed characteristics and economic feasibility of wind systems. The results show that the average wind speeds are higher during the hot period (5-7.5 m/s from 12 p.m. to 8 p.m.) than the coldest one (4-6 m/s from 12 p.m. to 8 p.m.) at 10 m hub height. The annual mean wind speed over the considered period was 4.05 m/s. The performance of the selected empirical EPFM method was observed and was close to the measured wind speed data of Nagad k values ranges from 1.65 to 1.99 while c values vary between 4.13 m/s to 6.11 m/s at 10 m. At 50 m, the annual mean wind speed was 5.37 m/s, k was 1.96 and c was 7.41 m/s. For the cold season, c was 7.15 m/s, which is a sign that the Nagad site is suitable for wind power technologies development. At 50 m hub height, the annual WED was recorded at 339 W/m² while it was measured at 297 W/m² in the cold season and 400 W/m² in the hot season. It is important to note that Nagad site was classified as class 3 and was considered suitable to harness wind power. In August, the highest CF was 32.33% for the Nordex turbine followed by 21.22% for Vestas V44 and 13.20 % for Vestas V20. The minimum AEP values of about 3.60 MWh, 9.82 MWh, and 9.47 MWh are observed for the Nordex, Vestas V20, and Vestas V44, respectively. p y The economic analysis showed that RX30, Vestas V20, Enercon, Nordex N27, Vestas V44 wind turbines are recommended for Nagad due to a low price of energy between 0.05$/kWh and 0.31$/kWh. For Vestas V44, Nordex N27, and Enercon, the price of energy is 2-6 times cheaper than the average local tariff of electricity. The vertical turbines are not recommended for the Nagad site. REFERENCES [1] International Bank for Reconstruction and Development/The World Bank, “Tracking SDG7: The Energy Progress Report,” 2018. [Online]. Available: https://openknowledge.worldbank.org/handle/10986/29812. [1] International Bank for Reconstruction and Development/The World Bank, “Tracking SDG7: The Energy Progress Report,” 2018. [Online]. Available: https://openknowledge.worldbank.org/handle/10986/29812. [1] International Bank for Reconstruction and Development/The World Bank, “Tracking SDG7 [Online]. Available: https://openknowledge.worldbank.org/handle/10986/29812. [ ] p p g g [2] D. L. McCollum et al., “Connecting the sustainable development goals by their energy inter-linkages,” Environmental Researc Letters, vol. 13, no. 3, p. 033006, Mar. 2018, doi: 10.1088/1748-9326/aaafe3. [2] D. L. McCollum et al., “Connecting the sustainable development goals by the Letters, vol. 13, no. 3, p. 033006, Mar. 2018, doi: 10.1088/1748-9326/aaafe3. [2] D. L. McCollum et al., “Connecting the sustainable development goals by their energy inter-linkages,” Environmental Research Letters, vol. 13, no. 3, p. 033006, Mar. 2018, doi: 10.1088/1748-9326/aaafe3. p N. S. Ouedraogo, “Opportunities, Barriers and Issues with Renewable Energy Development in Africa: a Comprehensib Review,” Current Sustainable/Renewable Energy Reports, vol. 6, pp. 52–60, Jun. 2019, doi: 10.1007/s40518-019-00130-7. gy p pp [4] T. R. Ayodele, A. S. O. Ogunjuyigbe, and T. O. Amusan, “Techno-economic analysis of utilizing wind energy for water pumping in some selected communities of Oyo State, Nigeria,” Renewable and Sustainable Energy Reviews, vol. 91, pp. 335–343, Aug. 2018, doi: 10.1016/j.rser.2018.03.026. [4] T. R. Ayodele, A. S. O. Ogunjuyigbe, and T. O. Amusan, “Techno-economic analysis of utilizing wind energy for water pumping in some selected communities of Oyo State, Nigeria,” Renewable and Sustainable Energy Reviews, vol. 91, pp. 335–343, Aug. 2018, doi: 10.1016/j.rser.2018.03.026. , j [5] Z. H. Hulio, W. Jiang, and S. Rehman, “Techno - Economic assessment of wind power potential of Hawke’s Bay using Weibull parameter: A review,” Energy Strategy Reviews, vol. 26, p. 100375, Nov. 2019, doi: 10.1016/j.esr.2019.100375. j [5] Z. H. Hulio, W. Jiang, and S. Rehman, “Techno - Economic assessment of wind power potential of Hawke’s Bay using Weibull parameter: A review,” Energy Strategy Reviews, vol. 26, p. 100375, Nov. 2019, doi: 10.1016/j.esr.2019.100375. [6] Y. Z. Alharthi, M. K. Siddiki, and G. M. Chaudhry, “Resource Assessment and Techno-Economic Analysis of a Grid-Connected Solar PV-Wind Hybrid System for Different Locations in Saudi Arabia,” Sustainability, vol. 10, no. 10, p. 3690, Oct. 2018, doi: 10.3390/su10103690. [7] M. Gul, N. Tai, W. Huang, M. H. Nadeem, and M. 3.4. Cost of energy production: comparison and analysis The latter remark is highlighted in bold in Table 4. Their cost reveals that the horizontal wind turbines are viable renewable energy sources to serve local communities and to intensify the production of electricity in the Nagad site. price of electricity (0.32$/kWh), indicate that the estimated tariffs are lower than the local cost price of electricity. The latter remark is highlighted in bold in Table 4. Their cost reveals that the horizontal wind turbines are viable renewable energy sources to serve local communities and to intensify the production of electricity in the Nagad site. ACKNOWLEDGEMENT The authors of this article wish to acknowledge the financial support from the University of Djibouti and the contributions of Eng. Omar Abdoulkader Mohamed (INSTAD). The authors of this article wish to acknowledge the financial support from the University of Djibouti and the contributions of Eng. Omar Abdoulkader Mohamed (INSTAD). Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idris 3.4. Cost of energy production: comparison and analysis Table 4 shows the annual and seasonal computed values of PVC ($) and the UCE ($/kWh) while the corresponding cost of Comr are 15% and 25% of the total investment. The results show that the maximum PVC ranged from 18136.79$ to 47953.69$ for RX30 and Vestas V44 wind turbines. The Antaris turbine produces the highest annual cost of 2.31$/kWh and the Vestas V44 generates the lowest annual cost of 0.034$/kWh, followed by Nordex N27 with a cost of 0.060$/kWh. As indicated in Table 4, the highest cost of electricity in the cold season was 6.28$/kWh for the Antaris wind turbine, and the lowest was 0.09$/kWh for the Vestas V44. Hence, with regards to the hot season, the highest value of UCE was 3.42$/kWh for Antaris, and the lowest value of UCE was 0.05$/kWh for Vestas V44. Table 4. Results of PVC (in $) and UCE (in $/kWh) for each turbine (min. and max. represent 15% and 25% of the total investment) Type Turbine model PVC($) Annual Cold season Hot season UCE ($/kWh) UCE ($/kWh) UCE ($/kWh) Min. Max. Min. Max. Min. Max.. Min. Max. HAWT Antaris 19691.89 20447.24 2.22 2.31 6.05 6.28 3.30 3.42 Aeolos 20849.23 21648.97 0.67 0.69 1.81 1.88 0.99 1.03 RX30 17466.80 18136.79 0.12 0.13 0.33 0.35 0.19 0.20 Vestas V20 28653.04 29752.12 0.21 0.22 0.61 0.63 0.30 0.31 Enercon 19481.20 20228.55 0.081 0.085 0.21 0.22 0.12 0.13 Nordex N27 32780.19 34037.58 0.058 0.060 0.15 0.16 0.08 0.09 Vestas V44 46182.22 47953.69 0.033 0.034 0.08 0.09 0.04 0.05 VAWT Turby 16302.32 16927.65 2.35 2.44 6.34 6.58 3.54 3.68 Sun surf 37179.34 38605.47 0.59 0.62 1.53 1.58 0.93 0.97 Table 4. Results of PVC (in $) and UCE (in $/kWh) for each turbine (min. and max. represent 15% and 25% of the total investment) The local average electricity cost value is 0.32$/kWh in Djibouti. Comparing the estimated tariffs obtained by the RX30, Vestas V20, Enercon, Nordex N27, and Vestas V44 wind turbines with the local cost The local average electricity cost value is 0.32$/kWh in Djibouti. Comparing the estimated tariffs obtained by the RX30, Vestas V20, Enercon, Nordex N27, and Vestas V44 wind turbines with the local cost Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101 99  ISSN: 2252-8792 Int J Appl Power Eng price of electricity (0.32$/kWh), indicate that the estimated tariffs are lower than the local cost price of electricity. REFERENCES Yu, “Assessment of Wind Power Potential and Economic Analysis at Hyderabad in Pakistan: Powering to Local Communities Using Wind Power,” Sustainability, vol. 11, no. 5, p. 1391, Mar. 2019, doi: 10.3390/su11051391. W. Sawadogo et al., “Current and future potential of solar and wind energy over Africa using the RegCM4 CORDEX-COR ensemble,” Climate Dynamics, vol. 57, pp. 1647–1672, Sep. 2021, doi: 10.1007/s00382-020-05377-1. [8] W. Sawadogo et al., “Current and future potential of solar and wind energy over Africa using the RegCM4 CORDEX-CORE ensemble,” Climate Dynamics, vol. 57, pp. 1647–1672, Sep. 2021, doi: 10.1007/s00382-020-05377-1. [9] A. I. Idriss, R. A. Ahmed, A. I. Omar, R. K. Said, and T. C. Akinci, “Wind energy potential and micro-turbine performance analysis in Djibouti-city, Djibouti,” Engineering Science and Technology, an International Journal, vol. 23, no. 1, pp. 65–70, A. I. Idriss, R. A. Ahmed, A. I. Omar, R. K. Said, and T. C. Akinci, “Wind energy potential and micro-turbine performan analysis in Djibouti-city, Djibouti,” Engineering Science and Technology, an International Journal, vol. 23, no. 1, pp. 65–7 Feb. 2020, doi: 10.1016/j.jestch.2019.06.004. j j [10] N. M.-A. Mutombo and B. P. Numbi, “Assessment of renewable energy potential in Kwazulu-Natal province, South Africa,” Energy Reports, vol. 5, pp. 874–881, Nov. 2019, doi: 10.1016/j.egyr.2019.07.003. gy p pp j gy [11] A. Sundaram, A. A. Mas’ud, H. Z. Al Garni, and S. Adewusi, “Assessment of off-shore wind turbines for application in Saudi Arabia,” International Journal of Electrical and Computer Engineering (IJECE), vol. 10, no. 5, pp. 4507–4513, Oct. 2020, doi: 10.11591/ijece.v10i5.pp4507-4513. j pp 2] M. K. Abd, “Economic viability and profitability assessments of WECS,” International Journal of Electrical and Compu Engineering (IJECE), vol. 10, no. 2, pp. 1220–1228, Apr. 2020, doi: 10.11591/ijece.v10i2.pp1220-1228. [13] S. S. Saswat, S. Patra, D. P. Mishra, S. R. Salkuti, and R. N. Senapati, “Harnessing wind and solar PV system to build hybrid power system,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 12, no. 4, pp. 2160–2168, Dec. 2021, doi: 10.11591/ijpeds.v12.i4.pp2160-2168. 4] A. I. Idriss, R. A. Ahmed, A. I. Omar, H. A. Atteyeh, M. H. Ibrahim, and T. C. Akinci, “Wind and solar energy potential Herkalou and Lake Assal locations, Djibouti,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol.  100 ISSN: 2252-8792 no. 1, pp. 461–470, Mar. 2023, doi: 10.11591/ijpeds.v14.i1.pp461 pp jp pp [15] H. J. Queen, J. Jayakumar, and T. J. REFERENCES 2697, Oct. 2018, doi: 10.3390/en11102697. case study of Adama Zuria woreda, Ethiopia, Scientific African, vol. 14, p. e01045, Nov. 2021, doi: 10.1016/j.sciaf.20 [32] M. M. Alayat, Y. Kassem, and H. Çamur, “Assessment of Wind Energy Potential as a Power Generation Source: A Ca Eight Selected Locations in Northern Cyprus,” Energies, vol. 11, no. 10, p. 2697, Oct. 2018, doi: 10.3390/en11102697 g yp g p [33] J. F. N. Touafio, O. Sanda, S. Malenguinza, J. M’Boliguipa, and R. M. Mouangue, “Analysis of a wind turbine project in the city of Bouar (Central African Republic),” Scientific African, vol. 8, p. e00354, Jul. 2020, doi: 10.1016/j.sciaf.2020.e00354. [34] Bangue Centrale De Djibouti, “Bangue Centrale De Djibouti Rapport Annuel 2018,” 2018. [Online]. Available: https://banque- centrale.dj/wp-content/uploads/2019/10/BCD-Rapport-Annuel-2018_Proof-3.pdf. 5] M. Irwanto, N. Gomesh, M. R. Mamat, and Y. M. Yusoff, “Assessment of wind power generation potential in Perlis, Malaysi Renewable and Sustainable Energy Reviews, vol. 38, pp. 296–308, Oct. 2014, doi: 10.1016/j.rser.2014.05.075. [36] A. N. Celik, “A Techno-Economic Analysis of Wind Energy in Southern Turkey,” International Journal of Green Energy, vol. 4, no. 3, pp. 233–247, May 2007, doi: 10.1080/15435070701338358. REFERENCES Deepika, “Comparative techno-economic analysis of power system with and without renewable energy sources,” Indonesian Journal of Electrical Engineering and Computer Science, vol. 24, no. 3, pp. 1260–1268, Dec. 2021, doi: 10.11591/ijeecs.v24.i3.pp1260-1268. [16] A. Sakhrieh, J. Al Asfar, and N. A. Shuaib, “An optimized off-grid hybrid system for power generation in rural areas,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 13, no. 2, pp. 865–872, Jun. 2022, doi: 10.11591/ijpeds.v13.i2.pp865-872. jp pp [17] M. A. Abdelrahman, R. H. Abdel-Hamid, M. A. Abo Adma, and M. Daowd, “Techno-economic analysis to develop the first wind farm in the Egyptian western desert at Elkharga Oasis,” Clean Energy, vol. 6, no. 1, pp. 211–225, Feb. 2022, doi: 10.1093/ce/zkac006. [18] M. Daoudi et al., “Techno-economic feasibility of two onshore wind farms located in the desert area: a case study,” Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, vol. 45, no. 3, pp. 7544–7559, Aug. 2023, doi: 10.1080/15567036.2023.2222679. [19] O. A. Dabar, M. O. Awaleh, D. Kirk-Davidoff, J. Olauson, L. Söder, and S. I. Awaleh, “Wind resource assessment and economic analysis for electricity generation in three locations of the Republic of Djibouti,” Energy, vol. 185, pp. 884–894, Oct. 2019, doi: 10.1016/j.energy.2019.07.107. j gy 0] R. Rathi, C. Prakash, S. Singh, G. Krolczyk, and C. I. Pruncu, “Measurement and analysis of wind energy potential using fuz based hybrid MADM approach,” Energy Reports, vol. 6, pp. 228–237, Nov. 2020, doi: 10.1016/j.egyr.2019.12.026. thi, C. Prakash, S. Singh, G. Krolczyk, and C. I. Pruncu, “Measu y pp , gy p pp j gy [21] M. A. Saeed, Z. Ahmed, J. Yang, and W. Zhang, “An optimal approach of wind power assessment using Chebyshev metric for determining the Weibull distribution parameters,” Sustainable Energy Technologies and Assessments, vol. 37, p. 100612, Feb. 2020, doi: 10.1016/j.seta.2019.100612. j [22] R. Blažević, H. Hadžiahmetović, and I. Ahmović, “Statistical Analysis and Assessment of Wind Energy Potential in Sarajevo, Bosnia and Herzegovina,” Tehnicki vjesnik - Technical Gazette, vol. 28, no. 5, pp. 1511–1518, Oct. 2021, doi: 10.17559/TV- 20200720163605. [23] S. A. Akdağ and Ö. Güler, “A novel energy pattern factor method for wind speed distribution parameter estimation,” Energy Conversion and Management, vol. 106, pp. 1124–1133, Dec. 2015, doi: 10.1016/j.enconman.2015.10.042. [24] A. H. Shaban, A. K. Resen, and N. Bassil, “Weibull parameters evaluation by different methods for windmills farms,” Energy Reports, vol. 6, pp. 188–199, Feb. 2020, doi: 10.1016/j.egyr.2019.10.037. p pp j gy 5] O. M. REFERENCES Salim, H. T. Dorrah, and M. A. Hassan, “Wind speed estimation based on a novel multivariate Weibull distribution,” IE Renewable Power Generation, vol. 13, no. 15, pp. 2762–2773, Nov. 2019, doi: 10.1049/iet-rpg.2019.0313. [26] B. A. Çakmakçı and E. Hüner, “Evaluation of wind energy potential: a case study,” Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, vol. 44, no. 1, pp. 834–852, Mar. 2022, doi: 10.1080/15567036.2020.1811810. [27] Q. Deltenre, T. De Troyer, and M. C. Runacres, “Techno‐economic comparison of rooftop‐mounted PVs and small wind turbines: a case study for Brussels,” IET Renewable Power Generation, vol. 13, no. 16, pp. 3142–3150, Dec. 2019, doi: 10.1049/iet- rpg.2019.0032. pg [28] A. Gungor, M. Gokcek, H. Uçar, E. Arabacı, and A. Akyüz, “Analysis of wind energy potential and Weibull parameter estimation methods: a case study from Turkey,” International Journal of Environmental Science and Technology, vol. 17, pp. 1011–1020, Feb. 2020, doi: 10.1007/s13762-019-02566-2. [29] X.-Y. An, Z. Yan, and J.-M. Jia, “A new distribution for modeling wind speed characteristics and evaluating wind power potential in Xinjiang, China,” Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, May 2020, doi: 10.1080/15567036.2020.1758250. [30] İ. Mert, F. Üneş, C. Karakuş, and D. Joksimovic, “Estimation of wind energy power using different artificial intelligence techniques and empirical equations,” Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, vol. 43, no. 7, pp. 815–828, Apr. 2021, doi: 10.1080/15567036.2019.1632981. , p , 1] C. S. Ambaye, K. T. Beketie, D. Y. Ayal, and Z. T. Dame, “Multiple criteria application in determining wind power potential: p [31] C. S. Ambaye, K. T. Beketie, D. Y. Ayal, and Z. T. Dame, “Multiple criteria application in determining wind power potential: A case study of Adama Zuria woreda, Ethiopia,” Scientific African, vol. 14, p. e01045, Nov. 2021, doi: 10.1016/j.sciaf.2021.e01045. [ ] y , , y , , p pp g p p case study of Adama Zuria woreda, Ethiopia,” Scientific African, vol. 14, p. e01045, Nov. 2021, doi: 10.1016/j.sciaf.2021.e01045. [32] M M Alayat Y Kassem and H Çamur “Assessment of Wind Energy Potential as a Power Generation Source: A Case Study of case study of Adama Zuria woreda, Ethiopia,” Scientific African, vol. 14, p. e01045, Nov. 2021, doi: 10.1016/j.sciaf.2021.e01045. [32] M. M. Alayat, Y. Kassem, and H. Çamur, “Assessment of Wind Energy Potential as a Power Generation Source: A Case Study of Eight Selected Locations in Northern Cyprus ” Energies, vol. 11, no. 10, p. BIOGRAPHIES OF AUTHORS Abdoulkader Ibrahim Idriss is an assistant professor at Faculty of Electrical and Energy Engineering, Université de Djibouti, Djibouti. He holds a Ph.D. degree in Photonic Engineering in France (Université de Franche-Comté, Besançon) with a specialization in optical nano-antennas for the inspection of photonic structures. His research areas are materials, photonics, and nanomaterials for renewable energy. He is Dean of the Faculty of Engineering. He was also the Director of the Logistic and Transport Centre (Centre of Excellence), financed by the World Bank, from 2019 to December 2021. He is currently a professor in the department of electrical engineering. He is also Guest Editor of the Special Issue Big Data in Renewable Energy for Renewable and Sustainable Energy Reviews published by Elsevier’s Ltd. He can be contacted at email: abdoulkader_ibrahim_idriss@univ.edu.dj. Int J Appl Power Eng, Vol. 13, No. 1, March 2024: 91-101 101  ISSN: 2252-8792 Int J Appl Power Eng Ramadan Ali Ahmed received a Ph.D. in nanoscience’s. He completed his Ph.D. in Laboratory of Molecular Physics at the Faculty of Science and Technic at the Université de Franche-Comté (France) in 2010. He is assistant-professor at the Faculty of Science since 2002 and also gives courses at the Faculty of Engineering in electricity and magnetism. Dean of the faculty of science since 2012 at the Université de Djibouti, he is currently Dean of the Institute of Industrial Technology. His current research is focused on the fields of renewable energies, photonics, and mechanical engineering. He can be contacted at email: ramadan_ali@univ.edu.dj. Hamda Abdi Atteyeh obtained her Bachelor of Engineering (Electrical) and M.Sc. in Electrical and Energy from the Faculty of Engineering at the Université de Djibouti. Her research interests include electrical power, electrical distribution systems, system engineering, energy efficiency, and system reliability. She can be contacted at email: hamdaabdi1998@gmail.com. Abdou Idris Omar is a lecturer in the Electrical Engineering Department, Université de Djibouti, Djibouti, since 2015, and he has been a senior lecturer since 2015 as a Fluid Mechanics and Energetics Engineer. He obtained his engineering’s degree from the Department of Mechanics-Energetics of the ENSIAME, one of the European "Grandes écoles" of engineering located in the city of Valenciennes in the north of France. He is currently Ph.D. in fluid mechanics and hydraulics in the Department of Energy System of the Faculty of Engineers at the Université de Djibouti, Djibouti. Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss) BIOGRAPHIES OF AUTHORS His main areas of research interest are energy efficiency in buildings, energy design, thermal comfort, bioclimatic design, applications of computational fluid dynamics, and renewable energy. He can be contacted at email: abdou_idriss_omar@univ.edu.dj. Tahir Cetin Akinci received B.S. degrees in Electrical Engineering, M.S. degrees, and Ph.D. degrees from Marmara University, respectively in 2000, 2003, and 2009. His research interests include electrical power systems, power electronics, computer programming, artificial neural network, machine learning, and signal processing technique. It is also actively researching in the Istanbul Technical University (ITU) cognitive system lab. He has been a visiting professor at the University of California Riverside (UCR), WCGEC since 2021. He has been working as a professor in the Electrical Engineering Department of Istanbul Technical University, Turkey. He can be contacted at email: akincitc@itu.edu.tr. Techno-economic assessment and wind energy potential of Nagad … (Abdoulkader Ibrahim Idriss)
https://openalex.org/W2919727945	https://europepmc.org/articles/pmc6690421?pdf=render	English	null	Control of stress and damage in structures by piezoelectric actuation: 1D theory and monofrequent experimental validation	Structural control & health monitoring/Structural control and health monitoring	2,019	cc-by	9,713	Control of stress and damage in structures by piezoelectric actuation: 1D theory and monofrequent experimental validation Juergen Schoeftner \| Andreas Brandl \| Hans Irschik Institute of Technical Mechanics, Johannes Kepler University Linz, Linz, Austria Summary Summary This contribution presents novel results on feed‐forward control of stress in piezoelectric structures by means of piezoelectric actuation. For that sake, we focus on a one‐dimensional benchmark problem, a piezoelectric transducer that is excited by a piezoelectric stack actuator. We investigate the following problem: Is it possible to actuate the piezoelectric transducer in such a manner that the dominant axial stress component is nullified. In order to find a theo- retical solution for this question, we discretize our system as a two‐degree‐of‐ freedom (2DOF) model. The equations of motion are transformed into the dif- ferential equations for the inner forces by taking advantage of the constitutive relations, which relate displacement, stress, and electric field. Finally, we find a mathematical relation for the piezoelectric transducer excitation in order to annihilate the transducer force. A static and a frequency‐dependent approxi- mate solution for the transducer actuation signal are derived. The latter solu- tion reduces the inner force drastically in a certain frequency range. After numerical results for the force‐control algorithm are presented, we finally experimentally verify our theory: First, the force‐controlled configuration is exposed to a monofrequent harmonic excitation test run for 30 min, showing no sign of fatigue or material failure, because the transducer force is below the ultimate tensile strength. Then, the system is excited by the same harmonic excitation again, but the control signal for the piezoelectric transducer is turned off. The result is a visible damage of the piezoelectric transducer, leading to a significant change of the first eigenfrequency. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2019 The Authors. Structural Control and Health Monitoring Published by John Wiley & Sons Ltd. Revised: 7 November 2018 Accepted: 11 January 2019 Revised: 7 November 2018 Accepted: 11 January 2019 Received: 6 April 2018 DOI: 10.1002/stc.2338 Struct Control Health Monit. 2019;26:e2338. https://doi.org/10.1002/stc.2338 Correspondence Correspondence Juergen Schoeftner, Johannes Kepler University Linz, Institute of Technical Mechanics, Altenbergerstrasse 69, Linz 4040, Austria. Funding information Austrian Science Fund FWF, Grant/ Award Number: P26762‐N30 1 \| INTRODUCTION 2 of 14 SCHOEFTNER ET AL. Smart or intelligent structures are equipped with multifunctional materials. These structures often take advantage of the piezoelectric effect in order to track the displacement of a flexible system. For fundamentals of structural piezoelectric- ity, see, for example, Yang, Preumont, Safari and Akdogan, and Moheimani and Fleming,1-4 particularly, displacement tracking, that is, the control of structural displacements, has gained an increasing interest in the last years in literature and engineering practice. Shape control denotes a particular case of displacement tracking, namely, that the desired dis- placement of the flexible system or at least the deflection at several locations shall be nullified. For literature overviews on shape control and displacement tracking, the reader is referred to Irschik, Irschik et al., Irschik and Krommer,5-7 and also Irschik et al.8 for recent results on displacement tracking of predeformed structures. Shape control and displace- ment tracking are feed‐forward control methods, where one asks for the control actuation in order that a certain desired displacement field is obtained. In case of piezoelectric control elements, one asks for the actuating electric field and its spatial distribution. It is assumed that the imposed load (given boundary excitations, imposed forces, applied tempera- ture distributions, or imposed electric fields) as well as the parameters of the structure under consideration is completely known. For the piezoelectric control actuation, the converse piezoelectric effect, or an analogous physical actuation effect, is utilized. In theoretical considerations, the influence of the much smaller direct piezoelectric effect, which is frequently exploited in structural health monitoring for the sake of measuring structural displacements, is usually neglected, or it is taken into account in an approximate manner in the actuation context. The direct piezoelectric effect is needed for automatic control algorithms, which however are not in the focus of the present contribution but which should be superimposed when the above assumptions for successful feed‐forward control are not satisfied in a sufficient manner. Driven by the recent encouraging technology development in the field of piezoelectric transducers and as a problem that is complementary to the above problem of controlling displacements, our group in the last years has studied the question, how to design piezoelectric control devices, so that a certain stress distribution can be achieved. 1 \| INTRODUCTION 2 of 14 Surprisingly, to our best knowledge, other groups hardly discussed this stress control problem in the open literature, although stress is known to be the main factor for failure and breakdown of structures, see Weber and Basu et al.9,10 Exceeding a certain critical stress level will cause irreparable damage and structural failure already under static conditions. Under a periodic dynamic excitation, damage under a relatively small number of stress cycles will appear after this critical stress level has been approached, which is called low‐cycle fatigue. But, depending on the type of material under consideration, this critical stress level is known to decrease with the number of stress cycles. The corresponding structural failure under a high number of stress cycles at a lower stress level is known as high‐cycle fatigue. For fundamentals on fatigue, the reader is referred, for example, to Suresh, Schijve,11,12 and Weisshaar.13 A first theoretical three‐dimensional (3D) framework for stress control has been formulated in Irschik,14 where stress control and displacement tracking strategies have been discussed at a continuum mechanics level, see also, for example, Irschik et al.15 for a short account on 3D dynamic stress compensation. More recently, Schoeftner and Irschik16 showed how to achieve a desired stress distribution for a base‐excited one‐dimensional bar with continuously distributed mass and stiffness by a proper temporal and spatial distribution of the piezoelectric actuation. Control of stress resultants, such as of bending moments, in piezoelectrically actuated continuous beam‐type structures has been treated in Schoeftner.17 In Irschik, Irschik et al., Schoeftner and Irschik, and Schoeftner,14-17 first 3D and 1D theoretical and numerical results for the feed‐forward stress control by piezoelectric actuation have been presented, where it has been assumed that the necessary spatial and temporal distribution of the piezoelectric actuation can be realized every- where in the structure under consideration. No experimental results have been presented so far to validate these theo- retical findings in reality. The latter findings, which were obtained at a continuum mechanics level, however have given us confidence for designing laboratory experiments, which can be modeled at the level of structural mechanics and which can give clear evidence for the appropriateness of the concept of controlling stresses in structures by piezoelectric actuation. The first outcomes of these laboratory experiments are presented subsequently, where the corresponding structural system is described in detail below. KEYWORDS control of fatigue and damage, dynamics of structures, piezoelectric actuation, stress control, structural control - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2019 The Authors. Structural Control and Health Monitoring Published by John Wiley & Sons Ltd. Struct Control Health Monit. 2019;26:e2338. https://doi.org/10.1002/stc.2338 wileyonlinelibrary.com/journal/stc 1 of 14 Struct Control Health Monit. 2019;26:e2338. https://doi.org/10.1002/stc.2338 proportional to the internal force of each transducer. One end of the piezoelectric transducer is connected to a single point mass, whereas the other end is connected to a piezoelectric stack actuator, which serves as a robust excitation source of the transducer, see Figure 1 below. proportional to the internal force of each transducer. One end of the piezoelectric transducer is connected to a single point mass, whereas the other end is connected to a piezoelectric stack actuator, which serves as a robust excitation source of the transducer, see Figure 1 below. Our paper is organized as follows. First, the system is approximated as a two‐degree‐of‐freedom (2DOF) system, where the two masses of the model replace the masses of the stack actuator, of the piezoelectric transducer and the end mass, and where the dynamic interaction between stack actuator and piezoelectric transducer is taken into account. Applying Newton's fundamental law of dynamics to the free‐body diagrams of the two equivalent masses, we set up their equations of motion not only in terms of the accelerations of the masses but also in terms of the internal forces in the stack actuator and in the piezoelectric transducer. The two equations of motion are accompanied by two linear constitutive equations, which relate the internal forces with the displacements and the velocities of the single masses and with the electric fields in the two elements. In the present structural mechanics framework, where we neglect the direct piezoelectric effect and we assume the axial component of the electric field to be predominant in the two pie- zoelectric devices, the electric fields are replaced by corresponding imposed electric voltages. Assuming that the motion of the system starts from rest and transforming the four relations into the Laplace domain, we eventually find a theo- retical condition, how to electrically actuate the piezoelectric transducer in a feed‐forward framework, such that the internal force in the transducer, that is, the axial stress, is completely nullified. We first demonstrate in a numerical study that our theory yields the expected results, that is, that the force of the piezoelectric transducer vanishes in the numerical experiments when the theoretically derived piezoelectric actuation is applied to the transducer. As an impor- tant by‐product, our theoretical results demonstrate that it would not even be necessary to identify the numerical values of the parameters of the 2DOF model from the corresponding laboratory experiment, substituting the identified values afterwards into the theoretical solution, but that it is sufficient to perform preliminary transfer function measurements. This is because we succeed in identifying the theoretical stress control solution as the ratio of specific transfer functions of the system, which can be experimentally determined in the frequency domain easily. Then, we realize the laboratory setup, for which we identify the parameters of the corresponding 2DOF model on the one hand side and we determine the required transfer functions at a low excitation level on the other hand side. The experimental setup eventually verifies our proposed stress control method: We show that the piezoelectric trans- ducer, as long as the theoretically derived piezoelectric actuation signal for stress control is applied (control on), remains undamaged during a 30‐min test run. In contrast, in the uncontrolled configuration (control off), the internal force in the piezoelectric transducer exceeds a certain critical limit after a comparatively low number of cycles and thus is irrep- arably damaged. This failure results in a significant lowering of the frequency response peak of the system, and a crack in the piezoelectric transducer becomes clearly visible. The results of the tests are described in some detail. The paper ends with concluding remarks and a short outlook. 1 \| INTRODUCTION 2 of 14 Because the latter has section‐wise constant properties and only longitudinal vibra- tions are considered, the axial stress can be assumed to be uniformly distributed over the cross section and to be 3 of 14 SCHOEFTNER ET AL. 1 \| INTRODUCTION 2 of 14 In contrast to our previous theoretical considerations in Irschik, Irschik et al., Schoeftner and Irschik, and Schoeftner,14-17 only a part of the considered system is subjected to the piezoelectric control actuation, and only the stress in that part is controlled. Hence, our subsequent formulations are novel also from a theoretical point of view. A first theoretical three‐dimensional (3D) framework for stress control has been formulated in Irschik,14 where stress control and displacement tracking strategies have been discussed at a continuum mechanics level, see also, for example, Irschik et al.15 for a short account on 3D dynamic stress compensation. More recently, Schoeftner and Irschik16 showed how to achieve a desired stress distribution for a base‐excited one‐dimensional bar with continuously distributed mass and stiffness by a proper temporal and spatial distribution of the piezoelectric actuation. Control of stress resultants, such as of bending moments, in piezoelectrically actuated continuous beam‐type structures has been treated in Schoeftner.17 In Irschik, Irschik et al., Schoeftner and Irschik, and Schoeftner,14-17 first 3D and 1D theoretical and numerical results for the feed‐forward stress control by piezoelectric actuation have been presented, where it has been assumed that the necessary spatial and temporal distribution of the piezoelectric actuation can be realized every- where in the structure under consideration. No experimental results have been presented so far to validate these theo- retical findings in reality. The latter findings, which were obtained at a continuum mechanics level, however have given us confidence for designing laboratory experiments, which can be modeled at the level of structural mechanics and which can give clear evidence for the appropriateness of the concept of controlling stresses in structures by piezoelectric actuation. The first outcomes of these laboratory experiments are presented subsequently, where the corresponding structural system is described in detail below. In contrast to our previous theoretical considerations in Irschik, Irschik et al., Schoeftner and Irschik, and Schoeftner,14-17 only a part of the considered system is subjected to the piezoelectric control actuation, and only the stress in that part is controlled. Hence, our subsequent formulations are novel also from a theoretical point of view. In the present contribution, we intend to annihilate the longitudinal stress of a piezoelectric control device, which we call the piezoelectric transducer. \| MODELING AND THEORETICAL FEED‐FORWARD STRE In this section, we derive theoretical conditions for a successful feed‐forward (open loop) control of stress in a structural element. As already mentioned, we intend to annihilate the longitudinal stress in a comparatively light piezoelectric transducer (subscript T), which is actuated by a more massive piezoelectric stack actuator (subscript A). The stack is connected to the transducer, where at the free end of the piezoelectric transducer, there is a single attached mass, see Figure 1a. The piezoelectric elements in Figure 1a are characterized by continuously distributed mass, stiffness, and damping. We now discretize the system in Figure 1a as a 2DOF system, see Figure 1b, where we follow the usual procedures and assumptions of structural mechanics. The two single masses of this model, which replace the stack actuator, the FIGURE 1 (a) Sketch of a piezoelectric transducer with attached mass, actuated by a piezoelectric stack actuator; (b) two‐ degree‐of‐freedom model FIGURE 1 (a) Sketch of a piezoelectric transducer with attached mass, actuated by a piezoelectric stack actuator; (b) two‐ degree‐of‐freedom model SCHOEFTNER ET AL. piezoelectric transducer, and the single attached mass, are denoted by mA and mT, respectively. Note that the dynamic interaction between stack actuator and piezoelectric transducer is taken into account in the 2DOF model. Moreover, because the two piezoelectric elements have section‐wise constant properties each and because only lon- gitudinal vibrations are considered, the axial stresses are assumed to be uniformly distributed over the respective cross sections. Thus, the notion of stresses can be directly replaced by the internal forces in the transducer, F T, and in the stack actuator, FA, respectively. Hence, our goal is to nullify the internal force FT by a control electric field that is applied to the piezoelectric transducer, in the presence of a given electric field in the stack actuator. In order to find a theoretical solution, Newton's law of dynamics is applied to the two effective masses first. In the Laplace domain, these two equations of motion read as follows: mA s2 xA ¼ −FA þ FT; (1) mT s2 xT ¼ −FT: (2) mA s2 xA ¼ −FA þ FT; (1) mT s2 xT ¼ −FT: (2) (1) mA s2 xA ¼ −FA þ FT; (2) mT s2 xT ¼ −FT: (2) mT s2 xT ¼ −FT: The coordinate of the Laplace transform is denoted by s, and the Laplace‐transformed absolute displacements of the two effective masses are xT and xA, respectively. For the free‐body diagrams leading to Equations (1) and (2), the reader is referred to Figure 2. The two equations of motion (1) and (2) are accompanied by two linear constitutive equations, which, at the structural mechanics level, relate the internal forces with the displacements and the dominant electric field compo- nents. We consider linear viscoelastic behavior of the piezoelectric elements, in order to include light damping. In the present structural mechanics framework, neglecting the direct piezoelectric effect, and because the axial component of the electric field is predominant in the two piezoelectric devices, the electric fields are replaced by corresponding imposed electric voltages, see the sketch of the 2DOF system in Figure 1b. The constitutive behavior thus can be for- mulated as FT ¼ kT þ sdT ð Þ xT −xA ð Þ þ cTV T; (3) FA ¼ kA þ sdA ð ÞxA þ cAVA: (4) FT ¼ kT þ sdT ð Þ xT −xA ð Þ þ cTV T; (3) (3) FT ¼ kT þ sdT ð Þ xT −xA ð Þ þ cTV T; FA ¼ kA þ sdA ð ÞxA þ cAVA: (4) (4) FA ¼ kA þ sdA ð ÞxA þ cAVA: The elastic stiffness of an element is denoted by k, and d is a viscous material parameter. The Laplace‐transformed electric voltages applied to the two piezoelectric elements are denoted as V, and the effective piezoelectric constants at the structural mechanics level are denoted by c. The four relations stated in Equations (1)–(4) can be used in various directions. A direct usage is to assume that the electric voltages in both piezoelectric elements be given and to compute the corresponding two displacements and two internal forces. Special direct solutions that are of interest for our later derivations follow by assuming that one of the two electric voltages in the two piezoelectric elements does vanish and to compute corresponding transfer func- tions. The notion of transfer function refers to the ratio between the Laplace transform of a certain output entity (super- scripts x, v, and F for displacement, velocity, and force, respectively) and the Laplace‐transformed corresponding excitation. Particularly, let Gx TA and Gv TA be the transfer functions for the displacement and for the velocity of the end mass due to VA, when there is VT = 0, and let Gx TT and Gv TT be the transfer function for the displacement due to VT, when VA = 0, Gx TA ¼ xT VA ; Gx TT ¼ xT VT ; Gv TA ¼ sxT VA ; Gv TT ¼ sxT V T : (5) (5) FIGURE 2 Free‐body diagram of the two‐degree‐of‐freedom model FIGURE 2 Free‐body diagram of the two‐degree‐of‐freedom model FIGURE 2 Free‐body diagram of the two‐degree‐of‐freedom model SCHOEFTNER ET AL. Anyway, setting mT = 0 in Equation (2), from which follows FT = 0, and inserting Equation (4) into Equation (1), the transfer function for the displacement of the mass mA due to VA becomes Gx AA mT¼0 ¼ − cA mAs2 þ dAs þ kA : (8 Gx AA mT¼0 ¼ − cA mAs2 þ dAs þ kA : (8) (8) The transfer function for the internal force in the stack actuator reads, see also Equations (1) and (8): The transfer function for the internal force in the stack actuator reads, see also Equations (1) and (8): GF AA mT¼0 ¼ mAs2 cA mAs2 þ dAs þ kA : (9) (9) Solutions of so‐called inverse problems related to Equations (1)–(4) also can be easily found, too. Shape control and displacement tracking are treated by requiring that the displacements should become certain desired functions of s, for example, that they should vanish; then, the electrical voltages VA, VT and the internal forces FA, FT are taken as the four unknowns. For our present stress control problem, we assume that the actuation voltage VA in the stack actuator is given, and we treat the voltage VT in the piezoelectric transducer as an unknown quantity, which must assure that our present control goal is satisfied, namely, that the internal force in the piezoelectric transducer should vanish, FT ¼ 0: (10) (10) FT ¼ 0: Hence, xT ¼ 0; (11) xT ¼ 0; (11) xT ¼ 0; (11) follows directly from Equation (2), and follows directly from Equation (2), and follows directly from Equation (2), and follows directly from Equation (2), and xA ¼ − cA mAs2 þ dAs þ kA V A ¼ Gx AA mT¼0 VA (12) xA ¼ − cA mAs2 þ dAs þ kA V A ¼ Gx AA mT¼0 VA (12) (12) follows from Equations (1), (4), and (10). 5 of 14 Then, it turns out that the ratio χT of the latter transfer functions becomes Then, it turns out that the ratio χT of the latter transfer functions becomes χT ¼ Gx TA Gx TT ¼ Gv TA Gv TT ¼ cA kT þ dTs ð Þ cT mAs2 þ dAs þ kA ð Þ: (6) (6) In a similar manner, we define the transfer function of the transducer force as the quotient of the Laplace‐ transformed internal force FT to the input voltages VA (when VT = 0) and VT (when VA = 0) as In a similar manner, we define the transfer function of the transducer force as the quotient of the Laplace‐ transformed internal force FT to the input voltages VA (when VT = 0) and VT (when VA = 0) as GF TA ¼ FT VA ; GF TT ¼ FT VT : (7) (7) In Equation (6), the ratio is equal to the ratio of the corresponding transfer functions for the velocities of the mass and of the corresponding transfer functions for the internal forces in the piezoelectric transducer. The former fact holds, because, in the Laplace domain, velocity is obtained as sxT, and the latter follows from Equation (2). Note that the ratio χT can be obtained from laboratory experiments in the frequency domain under low‐level excitations in a straightfor- ward manner. Another interesting case in the present context is the behavior of the system, when we set mT = 0 in Equations (1)–(4). In reality, this corresponds to a setup, in which the piezoelectric transducer and the attached end mass are absent, which will hardly change the value of the effective mass of the model mA, because the stack actuator is taken as comparatively heavy. SCHOEFTNER ET AL. We note that an alternative formulation and a solution for the transducer voltage in the time domain, which also takes into account the initial conditions, are presented in the Appendix. Here, we are only interested in the steady‐state solution, so initial disturbances will be attenuated after a while. It is interesting to note that the attached block mT does not affect the outcome, see Equation (14) or Equation (34). Furthermore, we have to remark here that the annihilation of stress does not necessarily mean also the annihilation of the displacement in general, as Equations (10) and (11) suggest. Here, we are interested in preventing the end mass to move; according to Newton's law, no force may act on the mass. Because we do not allow any external forces, a vanishing transducer force F T = 0 means the end mass mT to be at rest. The relation in Equation (14) represents the feed‐forward solution of our stress control problem, that is, the Laplace transform of the necessary electric control voltage in order that the internal force in the piezoelectric transducer vanishes. An interesting approximate form of the control law in Equation (14), which is valid in the lower frequency domain, is obtained by neglecting the frequency dependent components: VT;sta ¼ −χT 0 ð ÞVA with χT 0 ð Þ ¼ kTcA kAcT : (15) (15) In the following, this is shortly denoted as static solution and used for comparison sake. Inserting Equation (12) into Equation (2) yields the stack actuation force FA ¼ mAs2cA mAs2 þ dAs þ kA V A ¼ GF AA mT¼0 VA; (13) (13) (10)–(12) into Equation (3) yields tution of Equations (10)–(12) into Equation (3) yields and substitution of Equations (10)–(12) into Equation (3) yields and substitution of Equations (10)–(12) into Equation (3) yields and substitution of Equations (10)–(12) into Equation (3) yields VT ¼ −χT VA with χT ¼ cA kT þ s dT ð Þ cT mAs2 þ kA þ s dA ð Þ: (14) (14) 3.2 \| Three methods for stress control In this section, we present three feed‐forward control algorithms. These are denoted as dynamic control, static control, and monofrequent control with tuning and explained in the following: • Dynamic control: The transfer function Equation (14) is applied for the piezo control voltage VT,dyn(s) = −χT(s) VA(s), which should yield (theoretically) a perfect annihilation of the mass displacement and of the transducer force, see Equations (10) and (11). This should hold for any desired s = jω. • Dynamic control: The transfer function Equation (14) is applied for the piezo control voltage VT,dyn(s) = −χT(s) VA(s), which should yield (theoretically) a perfect annihilation of the mass displacement and of the transducer force, see Equations (10) and (11). This should hold for any desired s = jω. • Static control: The Laplace variable s is set to 0, so the transfer function in Equation (14) becomes a constant, see Equation (15), VT,sta(s) = −χT(0)VA(s). It turns out that this control voltage yields a very low displacement of the attached mass and of the transducer force in the low‐frequency domain, see Figure 4. • Static control: The Laplace variable s is set to 0, so the transfer function in Equation (14) becomes a constant, see Equation (15), VT,sta(s) = −χT(0)VA(s). It turns out that this control voltage yields a very low displacement of the attached mass and of the transducer force in the low‐frequency domain, see Figure 4. • Monofrequent control with tuning: Here, we are mainly interested in annihilating the transducer force only at f = 4300 Hz, from which follows: • Monofrequent control with tuning: Here, we are mainly interested in annihilating the transducer force only at f = 4300 Hz, from which follows: f = 4300 Hz, from which follows: VT;mono sð Þ ¼ −χT jωM ð ÞVA sð Þ: (16) (16) We split up the complex‐valued function χT(jωM) into χT(jωM) = χT(0) F(jωM), which clarifies that the static portion χT(0) is real valued and the function F (jωM) is complex valued in general. Inserting Equations (6) and (15) into F (jωM) = χT(jωM)/χT(0) and setting s = jωM, one finds for F (jωM), We split up the complex‐valued function χT(jωM) into χT(jωM) = χT(0) F(jωM), which clarifies that the static portion χT(0) is real valued and the function F (jωM) is complex valued in general. mode is hardly visible in the diagram and the dynamics beyond the second mode are out of our interest. The validated and modified parameters for the setup are listed in Table 1. mode is hardly visible in the diagram and the dynamics beyond the second mode are out of our interest. The validated and modified parameters for the setup are listed in Table 1. 3.1 \| Validation of the parameters In this section, we validate and parameterize our mathematical model. This can be done either by recalculation (e.g., the stiffness is computed by means of the geometry and the Young's modulus) or by relying on the manufacturers datasheet (for the piezoelectric transducer18 and for the stack actuator19). In our study, we weighed the masses of the stack, the transducer, and the attached mass and used the stiffness values and the piezoelectric coefficients from the manufac- turer's datasheet in a first step. Then, in a later step, we compared the frequency response functions (FRFs) Hv TA ω ð Þ ¼ Gv TA s¼ j ω and Hv TT ω ð Þ ¼ Gv TT s¼ j ω of the simulation results to the experimental ones. If the results do not match properly, we modified the stiffness, mass, and damping values until we obtain a good agreement to the experi- mental FRFs. Hence, also for our low‐dimensional model, the limit of acceptability is considered to be 8000 Hz, which is between the first and second eigenfrequency. It is noted that the Fourier transform H(ω) can be considered as a spe- cial form of the Laplace transform G(s) taking the values on the positive imaginary axis only, see Shin and Hammond.20 We further modified the system parameters for the 2DOF model in order to find a better match between simulation and experimental results, see Figure 3. Two longitudinal vibration modes are visible: The fundamental eigenfrequency is at f 1 = 4990 Hz and the second one at f 2 ≈11500 Hz. For the first mode, the transducer and actuator mass vibrate in phase (when the motion/velocity of the actuator mass is much smaller, see the blue peak in Figure 3), in contrast to the antiphase oscillations for the second mode. The higher eigenmodes are irrelevant for our study because the third FIGURE 3 Transfer functions (black: stack actuation, blue: piezoelectric transducer actuation) of (a) the experiment and (b) the simulation model with the identified parameters listed in Table 1 FIGURE 3 Transfer functions (black: stack actuation, blue: piezoelectric transducer actuation) of (a) the experiment and (b) the simulation model with the identified parameters listed in Table 1 SCHOEFTNER ET AL. 3.2 \| Three methods for stress control Inserting Equations (6) and (15) into F (jωM) = χT(jωM)/χT(0) and setting s = jωM, one finds for F (jωM), TABLE 1 Parameters used in the numerical study Variable (unit) Value mA (kg) 0.156 mT (kg) 0.109 kA (N/m) 495 ⋅106 kT (N/m) 156 ⋅106 dA (Ns/m) 259.84 dT (Ns/m) 224.00 cA (N/V) 8.12 cT (N/V) 9 TABLE 1 Parameters used in the numerical study TABLE 1 Parameters used in the numerical study Variable (unit) Value mA (kg) 0.156 mT (kg) 0.109 kA (N/m) 495 ⋅106 kT (N/m) 156 ⋅106 dA (Ns/m) 259.84 dT (Ns/m) 224.00 cA (N/V) 8.12 cT (N/V) 9 FIGURE 4 (a,b) Frequency response F T due to the actuation amplitude VA = 1 V, (c) necessary transducer amplitude VT (red: without control VT = 0 V, dark blue dashdot: dynamic control, see Equation (14), light blue dashed: static control, see Equation (15), blue: monofrequent control with tuning, see Equation (16)) FIGURE 4 (a,b) Frequency response F T due to the actuation amplitude VA = 1 V, (c) necessary transducer amplitude VT (red: without control VT = 0 V, dark blue dashdot: dynamic control, see Equation (14), light blue dashed: static control, see Equation (15), blue: monofrequent control with tuning, see Equation (16)) SCHOEFTNER ET AL. F jωM ð Þ ¼ 1 þ dT=kT ð ÞjωM −mA=kA ð ÞmAω2 M þ dA=kA ð ÞjωM þ 1 ½ ; with ωM ¼ 2πf M; f M ¼ 4300 Hz: (17) with ωM ¼ 2πf M; f M ¼ 4300 Hz: (17) (17) The alternative representation of the function F jωM ð Þ ¼ Aejφ demonstrates that F (jωM) can be split up into the amplitude A* that represents the matching gain, whereas the complex phase angle φ* represents a phase shift in the frequency domain; that is, in the time domain, this means a time delay of the system. From Equations (15) and (17), it follows for χT(jωM) in Equation (16), χT jωM ð Þ ¼ kTcA kAcT 1 þ dT=kT ð ÞjωM −mA=kA ð ÞmAω2 M þ dA=kA ð ÞjωM þ 1 ½ : (18) (18) One observes from Equation (18) that static control (see Equation (15)) is a special case of monofrequent control with ωM = 0, that is, F (0) = 1. 3.3 \| Harmonic analysis for stress control As a result of numerical computations, Figure 4 shows the amplitude of the transducer force F T if the piezo transducers voltage is not present (i.e., VT = 0V [red], control off) and for the three above control laws (control dynamic: dark blue dashdot, static control: light blue dashed, monofrequent control with tuning: blue). The actuation amplitude is VA = 1 V. The corresponding transducer voltage signals VT for dynamic, static, and monofrequent control are obtained by Equations (14)–(16), see Figure 4c. However, the dynamic control method is the only frequency‐dependent voltage and represents a transfer function with a peak at f ¼ ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ kA=mA p = 2π ð Þ ¼ 8965 Hz, see Equation (14). One recognizes the reso- nance frequencies at f 1 = 4990 Hz and f 2 ≈11000 Hz. The dynamic control method, Equation (14), yields a perfect annihilation of the transducer force (blue): The response is 0 for the whole frequency range. If the static control formula is applied for stress control, Equation (15), the force can be significantly reduced for frequencies f < 7500 Hz in comparison with the uncontrolled configuration. Especially, the force at the first eigenfrequency f 1 is lowered by 69% (from 66.3 to 20.6 N/V). In the higher frequency domain, however, this control method fails: The response at the second eigenfrequency is even amplified. Monofrequent control with tuning, see Equation (16), shows that the transducer force vanishes at the matching frequency f M = 4300 Hz. As long as the excitation frequency is close to the matching frequency, one observes a substantial force reduction: The force at the eigenfrequency is only 6.9 N/V. Assuming a discrepancy of the matching and the excitation frequency still leads to a sufficient reduction of the transducer force: Changing the excitation frequency to 4600 Hz (i.e., 7% higher than the matching frequency) yields FT = 0.4 N, which is still much lower than the amplitude without control or if the static control method is applied. Nevertheless, for the low‐frequency regime, the method static control works well. An improvement can be achieved by means of monofrequent control with tuning, where the control tracking variable FT can be annihilated at the specific frequency ωM. 3.2 \| Three methods for stress control One observes from Equation (18) that static control (see Equation (15)) is a special case of monofrequent control with ωM = 0, that is, F (0) = 1. 4.2 \| Description of the test procedure In this section, we experimentally verify the force control theory developed in Section 2. We show that a force‐controlled transducer does not suffer any damage, whereas an uncontrolled setup suffers an irreparable mechanical breakdown. As an objective criterion, it is evident that a significant mechanical damage is related to a notable loss of stiffness; therefore, a significant change of the eigenfrequency is an indicator for the damage. The test procedure, which is divided into five steps, is depicted in Figure 6 and explained in the following. FIGURE 5 (a) Schematic presentation; (b) photograph and electronic devices for the experimental setup FIGURE 5 (a) Schematic presentation; (b) photograph and electronic devices for the experimental setup Figure 5 shows the vertical arrangement of our setup. The consideration of gravity effects in Equations (1) and (2) does not play a significant role in our study, because the influence of gravity is much smaller than the deflection due to stack actu- ation, in particular close to the first resonance. In order to verify our stress control theory experimentally, the velocity of the attached mass vT is measured (device Polytec PDV‐100). We recalculate the axial transducer force, where the attached mass is glued onto the transducer cross section, as F T = mTs2xT, see Equation (2). The excitation and control signals in time‐ domain VA, VT are generated by utilizing dSpace ACE 1103 (hardware) and dSpace Control Desk (software). The corre- sponding real‐time C‐code (i.e., Equations (14)–(16) with the corresponding analog–digital output blocks for the communi- cation is generated from a MATLAB/Simulink model, in which the feed‐forward control model is implemented. The sampling time of the dSpace hardware is Ts = 1/20,000 s; this means that the sampling frequency is f s = 20,000Hz. PULSE Front‐end 3560D from Brüel&Kjær is used as a measurement device to read in the actuation and the sensor signals. The measurement software is PULSE LabShop (version 18). The sampling time of the measurement device is Ts = 1/64,000s. FIGURE 6 The five steps of the test procedure 4.1 \| Description of the experimental setup A photograph and a scheme of the experimental setup are shown in Figure 5. Figure 5b shows the attached mass (1), which is glued onto the piezoelectric transducer (2). The transducers are an SA Series 150‐V Piezo Stack Actuator SA070718 from the company piezo drive.19 The piezoelectric stack actuator (3) (Piezocomposite—stack‐type actuator series PStVS with preload, PSt 1000/25/40 VS 35, piezosystem jena18) serves as the excitation source and is rigidly con- nected to the lower surface of the transducer. The copper wires transmit the excitation VA and the control signal VT to the stack actuator and to the transducer, respectively. As glue material, an electrically isolating two‐component adhesive (Loctite Hysol 9466 A&B) is used. The electrodes of the transducer and of the stack actuator are linked to the linear power amplifiers (AE Techron 7224 power amplifier), which amplify the input signal from the control unit by a factor 20. Lossless copper wires transmit the necessary voltage signals VA and VT to the stack actuator and to the transducer electrodes of the actuation. 9 of 14 9 of 14 SCHOEFTNER ET AL. 4.2.4 \| Step 4‐Destruction phase Step 4 is the destruction phase, when the same excitation voltage for the stack as in Phase 3 is applied. This time, the electrodes of the piezoelectric transducer are short‐circuited VT = 0 V (control off). According to the manufacturer of the transducers,18 damage should occur if the transducer force exceeds 10–20% of the blocking force F block ≈1,800 N. 4.2.3 \| Step 3‐FRF measurements after 30‐min test run but before damage After the test run, we have to prove that the transducer is not damaged. In addition to obvious cracks of the surface, we need to define an objective criterion. Hence, we measure the FRFs again. 4.2.1 \| Step 1‐FRF measurements before 30‐min test run In a first step (Step 1), we measure the FRFs Hv TA ω ð Þ and Hv TT ω ð Þ of the undamaged 2DOF system (i.e., Hv TA ω ð Þ is the amplitude of velocity of the attached mass, if the stack actuation is excited by VA (with a short‐circuited piezo transducer VT = 0), and Hv TT ω ð Þ is the attached mass velocity if the piezoelectric transducer is harmonically excited by VT (with a short‐circuited stack actuator VA = 0)). 10 of 14 SCHOEFTNER ET AL. SCHOEFTNER ET AL. transducer VT = 0), and Hv TT ω ð Þ is the attached mass velocity if the piezoelectric transducer is harmonically excited by VT (with a short‐circuited stack actuator VA = 0)). transducer VT = 0), and Hv TT ω ð Þ is the attached mass velocity if the piezoelectric transducer is harmonically excited by VT (with a short‐circuited stack actuator VA = 0)). 4.2.5 \| Step 5‐FRF after destruction In order to detect damage and failure, we perform an FRF analysis, observing changes of the system properties. Signif- icant changes of the velocity response, in particular a reduction of the first resonance frequency and peak modifications, are clear indicators for the transducer damage. 4.2.2 \| Step 2‐30‐min force‐controlled test run Then, we perform a 30‐min test run by controlling the force of the piezoelectric transducer using monofrequent control with tuning, where we used the measured transfer functions Hv TA ω ð Þ and Hv TT ω ð Þ from Section 4.2.1 for calculating the desired transfer ratio χT at ω = ωM, compare Equations (6) and (14). Among the three control methods that we intro- duced in Section 3.2, we prefer monofrequent control because it represents an acceptable compromise between satisfy- ing stress control results on the one hand and it is relatively insensitive to parameter uncertainties on the other hand. The excitation (stack actuation) is VA = 40 sin(ωMt), with an excitation frequency f M = 4300 Hz that is close to the first resonance frequency. The control signal of the piezoelectric transducer theoretically is realized by transforming Equation (16) back into time domain. During the half‐hour test run (Step 2), the experimental setup is subjected to 7.74 ⋅106 stress cycles (i.e., 30 min at 4300 Hz) when the fatigue stress limit (=endurance limit) is achieved for ceramic materials. 4.3 \| Experimental results Figure 7 shows the experimental results of Steps 2 (30‐min test run in the controlled case VT ≠0) and 4 (destruction phase for VT = 0). The envelopes of the applied sinusoidal stack and piezo‐transducers voltages VA,enve and VT,enve are shown in Figure 7a (gray and red signals for the uncontrolled case, gray and blue signals for the case monofrequent control with tuning). Figure 7b–e shows the envelopes of the velocity and of the transducer force of the uncontrolled (Figure 7c and 7e) and the controlled (Figure 7b and 7d) configurations. One observes that the maximum velocity ampli- tude of the force‐controlled configuration is only vmax = 7.6 mm/s (Figure 7b, blue), which means a transducer force of F T max = mT(2π4300)⋅0.0076 ≈22.3 N that is much lower than the necessary damage force F damage ≈0.1 F block = 180 N. It is noted that according to the manufacturer's datasheet, the limit tensile force is about 10% to 20% of the blocking force.18 If the control method is deactivated from the beginning, one observes a very large velocity amplitude peak v = 105.6 mm/s, which means a transducers force of F T ≈310.4 N (17% of the blocking force, see Figure 7c, red). Indeed, at t = 36.1 s, the transducer suffers damage, see Figure 7. Finally, the FRFs for Steps 1, 3, and 5 are shown in Figure 8a. For Steps 1 and 3, the change of the first eigenfrequency before and immediately after the 30‐min test run is negligible; that is, there is no proof for damage. In Step 5, however, one recognizes a remarkable shift of the first eigenfrequency from 4990 Hz (undamaged configura- tion) to 3100 Hz (damaged configuration, dashed lines), which clearly indicates damage of the transducer. Further 11 of 14 11 of 14 SCHOEFTNER ET AL. FIGURE 7 Envelope curves for (a) the voltage actuation and the piezoelectric control signal, (b) velocity of the attached mass, and (c) zoom (without control red, with control blue) FIGURE 7 Envelope curves for (a) the voltage actuation and the piezoelectric control signal, (b) velocity of the attached mass, and (c) zoom (without control red, with control blue) FIGURE 8 (a) Frequency response functions before (Step 1: blue/black), after the 30‐min test run (Step 3: light blue/gray), and after the damage (Step 5: dashed blue/black). 4.3 \| Experimental results (b) The test specimen shows the horizontal crack of the piezoelectric transducer (compare the damaged transducer (right) to the undamaged transducer (left)) FIGURE 8 (a) Frequency response functions before (Step 1: blue/black), after the 30‐min test run (Step 3: light blue/gray), and after the damage (Step 5: dashed blue/black). (b) The test specimen shows the horizontal crack of the piezoelectric transducer (compare the damaged transducer (right) to the undamaged transducer (left)) damage indicators are shown in Figure 8b (left: undamaged; right: damaged), where one can observe a horizontal crack line on the surface of the transducer. that the transducer force is nullified or at least considerably reduced despite a piezoelectric actuation in a stack actuator is present, see Figures 1 and 5. In a first step, we discretize this system as a 2DOF model. One degree of freedom stands for the stack actuator; the other degree of freedom models the attached mass and the piezoelectric transducer. Using this simple structural model, we first present a theoretical relation for the piezoelectric voltage actuation that is needed, such that the force in the transducer can be eliminated, see Equation (14). Interestingly enough, it turns out that the attached mass remains at rest in the controlled configuration, see Equation (11), which should give raise to various practically appealing applications. We then prove our theory by numerical computations. In these computations, it is demonstrated that it should be indeed possible in practice to nullify the transducer force for a given stack actuation, see Figure 4. Finally, an experimental setup proves our developed stress control method, see Figure 5. Our test specimen is harmon- ically excited close to the first resonance frequency. It is shown that the endurance stress limit (fatigue limit) is not exceeded and the device does not experience any damage, if the developed stress control method is applied. Contrary, the uncontrolled configuration suffers an irreparable mechanical damage, if the specimen exceeds a certain stress limit, see Figures 7 and 8. To the best knowledge of the authors, our study is the first experimental verification of feed‐forward stress control by piezoelectric actuation in the open literature. We understand it as a further step toward the develop- ment of so‐called ageless structures. ORCID Juergen Schoeftner https://orcid.org/0000-0003-2990-4102 Andreas Brandl https://orcid.org/0000-0003-1990-2584 Hans Irschik https://orcid.org/0000-0001-6593-243X Juergen Schoeftner https://orcid.org/0000-0003-2990-4102 Andreas Brandl https://orcid.org/0000-0003-1990-2584 Hans Irschik https://orcid.org/0000-0001-6593-243X Hans Irschik https://orcid.org/0000-0001-6593-243X 5 \| CONCLUSION A main goal of the present contribution is the experimental verification of feed‐forward stress control. We investigate, how a force‐excited piezoelectric transducer with an attached mass can be piezoelectrically actuated in such a manner, 12 of 14 SCHOEFTNER ET AL. REFERENCES 1. Yang J. Analysis of Piezoelectric Devices. Hackensack, NJ: World Scientific; 2006. 2. Preumont A. Mechatronics: Dynamics of Electromechanical and Piezoelectric Systems. Springer; 2006. 3. Safari A, Akdogan EK (Eds). Piezoelectric and Acoustic Materials for Transducer Applications. Springer; 2008. https://www.springer.com/ de/book/9780387765389 3. Safari A, Akdogan EK (Eds). Piezoelectric and Acoustic Materials for Transducer Applications. Springer; 2 de/book/9780387765389 4. Moheimani SOR, Fleming AJ. Piezoelectric Transducers for Vibration Control and Damping. London: Springer; 2006. review on static and dynamic shape control of structures by piezoelectric actuation. Eng Struct. 2002;24(1):5‐ 5. Irschik H. A review on static and dynamic shape control of structures by piezoelectric actuation. Eng St 6. Irschik H, Krommer M, Nader M, Schöftner J, Zehetner C. Active and passive shape control of structures. In: Del Grosso E, Basso P, eds. Proceedings 5th European Conference on Structural Control (EACS 2012). Genua, Italy: Erredi Grafiche Editoriali; 2012 Paper No. 146, 8 6. Irschik H, Krommer M, Nader M, Schöftner J, Zehetner C. Active and passive shape control of structures. In: Del Grosso E, Basso P, eds. Proceedings 5th European Conference on Structural Control (EACS 2012). Genua, Italy: Erredi Grafiche Editoriali; 2012 Paper No. 146, 8 pages, 2012. https://www.amazon.com.au/structures-Proceedings-european-Conference-structural/dp/8895023137 Proceedings 5th European Conference on Structural Control (EACS 2012). Genua, Italy: Erredi Grafiche Editoriali; 2012 Paper No. 146, pages, 2012. https://www.amazon.com.au/structures-Proceedings-european-Conference-structural/dp/8895023137 pages, 2012. https://www.amazon.com.au/structures-Proceedings-european-Conference-structural/dp/8895023137 7. Irschik H, Krommer M. A review on static and dynamic shape control of structures: the period 2002‐2012. In: Adam C, Heuer R, Lenhardt W, Schranz C, eds. Proceedings Vienna Congress on Recent Advances in Earthquake Engineering and Structural Dynamics 2013 (VEESD 2013). Vienna, Austria; 2013 Paper No. 581, 2013. https://fodok.jku.at/fodok/publikation.xsql?PUB_ID=45429 8. Irschik H, Krommer M, Zehetner C. Displacement tracking of pre‐deformed smart structures. Smart Struct Syst. 2016;18(1):139‐154. ce tracking control scheme for MR dampers. Struct Control Health Monit. 2015;22(12):1373‐1375. 9. Weber F. Robust force tracking control scheme for MR dampers. Struct Control Health Monit. 2015;22(1 10. Basu B, Bursi OS, Casciati F, et al. A European Association for the Control of Structures joint perspective. Recent studies in civil structural control across Europe. Struct Control Health Monit. 2014;21(12):1414‐1436. 11. Suresh S. Fatigue of Materials, 2nd Edition. Cambridge University Press; 1998. 11. Suresh S. Fatigue of Materials, 2nd Edition. Cambridge University Press; 1998. 12. Schijve J. Fatigue of Structures and Materials, 2nd Edition. Springer; 2009. 13. Weisshaar T., Aerospace structures—an introduction to fundamental problems. ACKNOWLEGDEMENT J. Schoeftner acknowledges support from the Austrian Science Fund FWF via the Project P26762‐N30. J. Schoeftner acknowledges support from the Austrian Science Fund FWF via the Project P 17. Schoeftner J. Bending moment tracking and the reduction of the axial stress in vibrating beams by piezoelectric actuation. Acta Mechanica. 2017;228(11):3827‐3838. 17. Schoeftner J. Bending moment tracking and the reduction of the axial stress in vibrating beams by piezoelectric actuation. Acta Mechanica. 2017;228(11):3827‐3838. 18. Website of the manufacturer of the piezoelectric transducer: https://www.piezodrive.com/actuators/150v‐piezo‐stack‐actuators/ (February 26th, 2018). 18. Website of the manufacturer of the piezoelectric transducer: https://www.piezodrive.com/actuators/150v‐piezo‐stack‐actuators/ (February 26th, 2018). 19. Website of the manufacturer of the piezoelectric stack: https://www.piezosystem.com/fileadmin/Piezocomposite/Datenblaetter/Aktoren/ en/Stapel/VS/PSt_1000_VS_35_ds_Rev00_2017_01_09.pdf (February 26th, 2018) 19. Website of the manufacturer of the piezoelectric stack: https://www.piezosystem.com/fileadmin/Piezocom en/Stapel/VS/PSt_1000_VS_35_ds_Rev00_2017_01_09.pdf (February 26th, 2018) 20. Shin K, Hammond JK. Fundamentals of Signal Processing for Sound and Vibration Engineers. West Suss How to cite this article: Schoeftner J, Brandl A, Irschik H. Control of stress and damage in structures by piezoelectric actuation: 1D theory and monofrequent experimental validation. Struct Control Health Monit. 2019;26:e2338. https://doi.org/10.1002/stc.2338 REFERENCES West Lafayette, in: Purdue University; 2011 (https://engi- neering.purdue.edu/AAECourses/aae352/2013/AAE%20352%20Course%20Text%20Weisshaar%202011.pdf). 14. Irschik H. Generation of Transient Desired Displacement or Stress Fields in Force Loaded Solids and Structures by Smart Actuation, in: Proc. XXXV Summer School Advanced Problems in Mechanics (APM2007). Saint‐Petersburg, Russia; 2007. https://fodok.jku.at/fodok/ publikation.xsql?PUB_ID=24741 15. Irschik H, Gusenbauer M, Pichler U. Dynamic stress compensation by smart actuation. In: Smith RC, ed. Proceedings of SPIE on Smart Structures and Materials 2004: Modeling, Signal Processing, and Control. San Diego, CA; 2004 SPIE vol. 5383, Paper No. 386, 2004. 16. Schoeftner J, Irschik H. Stress tracking of piezoelectric bars by eigenstrain actuation. J Sound Vib. 2016 SCHOEFTNER ET AL. APPENDIX Finally we present an alternative derivation for our force control theory. Contrary to section 1 when all signals are considered as Laplace‐transformed signals, here the displacements xA, xT, the voltages VA, VT and the forces F A, F T (see Equations (19)–(34)) are time‐domain signals. We present a general solution for the necessary voltage actuation of the transducer, if the system does not start from rest and if the desired transducer force is non‐zero, see Equation (28). Applying Newton's law for the two masses (see Figure 1) and solving for the acceleration one finds Applying Newton's law for the two masses (see Figure 1) and solving for the acceleration, (19) xA;tt ¼ −FA þ FT ð Þ=mA (19) xA;tt ¼ −FA þ FT ð Þ=mA xT;tt ¼ −FT=mT: (20) xT;tt ¼ −FT=mT: (20) The spring‐law (without damping) reads, see Figure 2 FT ¼ kT xT −xA ð Þ þ cTV T (21) FA ¼ kA xA þ cAV A (22) ng cAVA yields FT ¼ kT xT −xA ð Þ þ cTV T (21) FT ¼ kT xT −xA ð Þ þ cTV T (21) FA ¼ kA xA þ cAV A (22) FA ¼ kA xA þ cAV A (22) FA ¼ kA xA þ cAV A (22) Multiplying Equation (19) by kA and adding cAVA,tt yields Multiplying Equation (19) by kA and adding cAVA,tt yields kAxA;tt þ cAVA;tt ¼ −FA þ FT ð Þ kA mA þ cAV A;tt; (23) (23) nd multiplying Equation (20) by kT and adding (cTVT,tt −kTxA,tt) yields and multiplying Equation (20) by kT and adding (cTVT,tt −kTxA,tt) yields and multiplying Equation (20) by kT and adding (cTVT,tt −kTxA,tt) yields and multiplying Equation (20) by kT and adding (cTVT,tt −kTxA,tt) yields and multiplying Equation (20) by kT and adding (cTVT,tt −kTxA,tt) yields kT xT;tt −xA;tt þ cTVT;tt ¼ −FT kT mT −kTxA;tt þ cTVT;tt: (24) (24) Note that the left hand sides of Equations (23) and (24) are FA,tt and FT,tt, respectively. Substituting Equation (19) into (24) and replacing the left‐hand sides of Equations (23) and (24), one finds the following two coupled second order differential equations for the transducer force F T and for the stack actuator force F A Note that the left hand sides of Equations (23) and (24) are FA,tt and FT,tt, respectively. 17. Schoeftner J. Bending moment tracking and the reduction of the axial stress in vibrating beams by piezoelectric actuation. Acta Mechanica. 2017;228(11):3827‐3838. APPENDIX Substituting Equation (19) into (24) and replacing the left‐hand sides of Equations (23) and (24), one finds the following two coupled second order differential equations for the transducer force F T and for the stack actuator force F A FA;tt þ kA mA FA ¼ FT kA mA þ cAVA;tt (25) (25) FT;tt þ kT m−1 T þ m−1 A FT ¼ kT mA FA þ cTV T;tt: (26) (26) Transforming Equations (25) and (26) into a fourth order differential equation for the piezoelectric transducer force yields F T(t), one finds SCHOEFTNER ET AL. cT mA kT VT;tttt þ cT kA kT VT;tt ¼ −cAVA;tt þ mA kT FT;tttt þ kA kT þ mA mT þ 1 FT;tt þ kA mT FT:1 (27) (27) Equation (27) states how to choose the actuation voltage, if the transducer force F T(t) follows a certain path (in general F T(t) ≠0) and if the voltage actuation VA (that is considered as disturbance source here) is known a‐priori. Performing integration with respect to time twice, we find the second order ordinary differential equation for the piezoelectric transducer voltage VT(t) cT mA kT VT;tt þ cT kA kT VT ¼ −cAVA þ mA kT FT;tt þ kA kT þ mA mT þ 1 FT þ kA mT ∬FT dt dt þ C1t þ C0; (28) (28) which requires 2 initial conditions VT(0), VT,t(0) and the specification of the integration constants C0 and C1. Taking into account the spring law at t = 0 FT 0 ð Þ ¼ kTðxT 0 ð Þ −xA 0 ð ÞÞ þ cTV T 0 ð Þ →VT 0 ð Þ ¼ 1 cT½FT 0 ð Þ −kT xT 0 ð Þ −xA 0 ð Þ ð Þ (29) (29) FT;t 0 ð Þ ¼ kT xT;t 0 ð Þ −xA;t 0 ð Þ þ cTVT;t 0 ð Þ →VT;t 0 ð Þ ¼ 1 cT FT;t 0 ð Þ −kT xT;t 0 ð Þ −xA;t 0 ð Þ (30) (30) one finds the initial conditions for VT(0), VT,t(0). APPENDIX The integration constants C0 and C1 can be calculated by further time‐ derivatives of Equations (29) and (30), when VT,tt(0) and VT,ttt(0) from Equations (31) and (32) FT;tt 0 ð Þ ¼ kT xT;tt 0 ð Þ −xA;tt 0 ð Þ þ cTVT;tt 0 ð Þ →VT;tt 0 ð Þ ¼ 1 cT FT;tt 0 ð Þ þ kT mT FT 0 ð Þ þ kT mAðkAxA 0 ð Þ þ cAVA 0 ð ÞÞ (31) (31) and FT;ttt 0 ð Þ ¼ kT xT;ttt 0 ð Þ −xA;ttt 0 ð Þ þ cTVT;ttt 0 ð Þ →V T;ttt 0 ð Þ ¼ 1 cT FT;ttt 0 ð Þ þ kT mT FT;t 0 ð Þ þ kT mA kAxA;t 0 ð Þ þ cAV A;t 0 ð Þ (32) (32) are inserted into Equation (28) and its time‐derivation. A special case solution is found if all initial conditions are zero and one demands F T(t) to vanish (i.e. force annihi- lation). Then one finds from Equation (28) A special case solution is found if all initial conditions are zero and one demands F T(t) to vanish (i.e. force annihi- lation). Then one finds from Equation (28) cT mA kT VT;tt þ cT kA kT VT ¼ −cAVA with VT 0 ð Þ ¼ VT;t 0 ð Þ ¼ 0 (33) (33) Transformation into the Laplace domain yields Transformation into the Laplace domain yields bVT ¼ − cAkT cT mAs2 þ kA ð Þ bVA (34) (34) where the tilde‐symbol denotes the corresponding signals in the Laplace domain. One observes that Equation (34) is equal to Equation (14) if damping is neglected. Furthermore we see that the attached transducer mass mT does not influence the solution. If the configuration does not start from rest, e.g. xT(0) ≠0, it can be shown by steps (28) to (34) that the steady‐state solution is not affected as long as a little amount of damping is considered which is always the case in reality (as in our experimental setup). where the tilde‐symbol denotes the corresponding signals in the Laplace domain. One observes that Equation (34) is equal to Equation (14) if damping is neglected. Furthermore we see that the attached transducer mass mT does not influence the solution. If the configuration does not start from rest, e.g. APPENDIX xT(0) ≠0, it can be shown by steps (28) to (34) that the steady‐state solution is not affected as long as a little amount of damping is considered which is always the case in reality (as in our experimental setup). rrection added on 18 March 2019, after first online publication: equations (27) and (28) have been corrected.
https://openalex.org/W3118999030	https://zenodo.org/records/1526274/files/article.pdf	English	null	Lord Kelvin.—II	Scientific American	1,910	public-domain	5,310	USES OF BALATA. USES OF BALATA. Balata is a substance belonging to the rubber-like products, and which is very similar to gutta-percha. It is obtained from the milky juice of the "Bully tree" (Sapota muelleri Belk.), found chiefly in the Guianas and Venezuela. This substance is obtained by felling the trees; although nowadays the bark is also cut open and the juice collected in wooden vessels. While a medium-sized tree, according to the old method (by felling) produced a quantity of from 6 to 12 pounds of balata in one process, only 0.6 to 1 pound of bal sam is obtained by the method of incision; but in the latter case the bark may be cut every year. According to a Russian patent, an insulating sub stance for the covering of cables, etc., which may be used as a surrogate for gutta-percha, is obtained by boiling 45 parts of asphalt, adding thereto, and stir ring the mass uninterruptedly, 40 parts of colophony, until a thick mass is formed; thereupon 10 parts of turpentine and 5 parts of linseed oil are poured into mass, which must be left boiling until the required lent our has been reached. Jelotong is considerably inferior to gutta-percha, and is chiefly used as a filling material for low classes of rubber and gutta-percha goods. The article is principally brought upon the European markets by the Borneo-Sumatra Maatschappij Company, which con trols the jelotong product almost entirely, via Singa pore. London is the principal jelotong market, only small quantities being handled in Hamburg. The local importers of jelotong do not seek a market for the product outside of Germany; the supplies reach ing Hamburg, mostly via London, are almost exclu sively sold to inland manufacturers. Another German patent prescribes that caoutchouc be dissolved in turpentine oil; then shellac or asphalt to be added, the mass to be heated until becoming uni form. After removing the mass from the fire, rice flour, watery agar-agar solUtion, and umbra is added, while it must be stirred until it congeals. Thereupon the mass is manipulated between rolls. In this man ner a surrogate for gutta-percha is obtained, which may be vulcanized with sulphur, but preserves the property of being plastic, and which can be kneaded when heated. In drying the juice in the air it is transformed into a whitish-pink or reddish substance. B Y PRO F. J. A. E WIN G. J Concluded from Supplement No. 1799, page 407 Concluded from Supplement No. 1799, page 407 that is, turned so that she heads successively on all points. The permanent magnetism will cause an er ror of the compass which will be of the same . nature as you woul̡ find if you placed a compass needle on a fixed pivot and disturbed it by turning a bar mag- net slowly round a vertical axis. This error will reach a maximum twice in the revolution, once to one side and once to the other side-in other words, once in each semicircle. Hence it is called the semicircu lar error. The permanent magnetism of the ship has a vertical component, and this causes not only semi circular error, but also a heeling error, namely, a deflection of the compass when the ship 'inclines to either side. made the card as light as he could get it-a mere aluminium rim tied by silk threads to a small cen tral boss, just as the rim of a bicycle wheel is tied to the nave by wire spokes, and from the silk-thread spokes he hung short pieces of magnetized knitting needle to serv'O) as the magnets. The result was that not only was the total weight very small, but it was nearly all in the rim, where it is most useful for giving moment of inertia and consequent slowness of period. Magnets and all, the card only weighs 180 grains for a 10-inch size, and yet its period of oscillation is much longer than that of the old stand ard compass, while its friction error is less. IT is time now to turn to Lord Kelvin's work in navigation. Taking the two oldest aids to naviga tion, the compass and the sounding-line, he revolu tionized them both. Where most men would have thought there was nothing left for invention to do, he found much. He has earned profound gratitude for appliances which add immeasurably to the secur ity of all who go to sea. B Y PRO F. J. A. E WIN G. He has been called the best friend the sailor e':er had; and it is said that a bluejacket was once overheard to remark, "I don't know who this Thomson may be, but every sailor ought to pray for him every night." It was about 1873 that he began to study the com pass seriously, partly because he had undertaken to write an article on it for Good Words, and partly because he had occasion to prepare, for the Royal Society, a biographical sketch of his friend Archi bald Smith, containing an account of Smith's work on the theory of the perturbation of the compass caused by the magnetism of iron ships. Kelvin's first patent for an improved compass was taken out in 1876. Another admirable feature of Kelvin's invention was his method of keeping the compass always level and free from pendulum-like oscillation. He hung the bowl, as usual, from gimballs, but with knife edges instead of the usual round spindles at the trun nions, and under the card he provided a chamber at the bottom of the bowl partly filled with castor oil. You see this in the glass-bowl now on the table. There is a glass partition to separate the place where the compass card stands from .the lower part of the bowl, and in the lower part is the castor-oil. Its function is to damp out any oscillation of the bowl that may tend to be set up by the rolling or pitching of the ship, and it does so by dissipating the energy of such swings. At the same time the knife-edge gimballs leave the compass perfectly free to take up true level. By a combination of three sets of correcting mag nets, two horizontal and one vertical, Kelvin ob tained complete neutralization of the disturbing effect of the ship's permanent magnetism, both as respects semicircular error .in change of the ship's course and heeling error as she heels or rolls. From time to time, if the condition of ]lerfect compensation is to be maintained, the ]losition of these various cor rectors has to be altered, because of changes which take ]llace in the so-called permanent magnetism of the shiP. The navigator has always to be on the look-out for the gradual development of errors from this cause, however perfectly the first adjustment has been carried out. He found the compass full of serious defects. B Y PRO F. J. A. E WIN G. For one thing it was very unsteady-that is to say, it was liable to be set swinging through a large angle when the ship rolled. Sometimes an attempt was made to reduce this unsteadiness by intro.ducing fric tion at the pivo.t, which, in a way, made matters worse by causing the compass to stick, pointing in a wrong direction. Under a mistaken idea of what would lead to steadiness, the card was made heavy and the needles long, and the long needles made it impossible to correct the compass properly for the magnetism of the ship. This was the most serious de f ect of all. In iron ships, and especially in ironclads, the compass is at the mercy of disturbing influences, which do much to mask the true directive force of the earth's magnetic field. To neutralize these is in dispensable; the way to do it, as a matter of theory, haq been pointed out, but it was only through the radical change in .construction which we owe to Kel vin that it became possible to carry the process into effect. We have next to consider the effeds of induced magnetism. The most important of these arise from the fact that the ship is a long body of magnetizable material turning in a horizontal plane, and there fore subject to the inductive influence of the horizon tal component of the earth's magnetic field. Think of what would happen if we were to take a pivoted compass needle and place it above or below a bar of soft iron, and slowly turn the bar round in a hori zontal plane. We are to think of the bar as having no appreciable magnetic hysteresis, so that in every position it is the induced effect only with which we have to do. What will be the nature of the devi ation? When the bar points nOI:.th, and again when it points south, there is no deflection of ·the needle, for though the magnetism of the bar is then at its strongest, the field due to it is in the line with the undisturbed earth field; also when the bar points east or west there is no deflection, for the bar then takes up no magnetism; but between these points, namely, when the bar is point N.E., S.E., S.W., or N.W., the deflection is at its maximum. SCIENTIFIC AMERICAN SUPPLEMENT No. 1800. JULY 2. 1910. in consequence of its high price, among which balata plays an important r61e. in consequence of its high price, among which balata plays an important r61e. creased. The materials so inspissated-are thcn, either alone or mixed, stirred with caoutchouc and oils, while the temperature is gradually raised. Finally, the mass is well kneaded in heated kneading machines. USES OF BALATA. The latter is viscid, extraordinarily flexible, more elastic than gutta-percha, can be cut as easily as the latter, and is vulcanizable with sulphur. Friction causes electri fication. At 49 deg. C. (118 deg. F.) balata can be kneaded, and it melts at 149 deg. C. (300 deg. F.). When heated, its odor is like that of gutta-percha. The substance is easily soluble in pure CS" and can therefore be purified without difficulty by evaporation of the filtered solution. According to Sparlich's ex periments, it contains 88.5 per cent carbon and 11.3 per cent hydrogen. Balata has been known in Europe since 1859 only. It was first used in industries in Great Britain, but is now considerably consumed in 'Germany, particularly for making electric insulators, beltings, for soles and heels of shoes, and in dental practice. For soles and heels of shoes it is boiled with gutta-percha. An Amenican syndicate is now reported as extract ing, by a certain chemical process, the rubber con tained in the jelotong; the yield is said to amount to 10 per cent. Of late the German Telegraph Administration has, by way of experiment, laid a cable in which the in sulating gutta-percha has been substituted by "Gutta Gentzsch," an artificial gutta-percha, composed of pure rubber and a certain kind of palm wax. "Gutta Gentzsch" is said to be a favorable surrogate for the real article and not to be inferior as regards insulat ing properties. Besides, the price of the cable was about 35 per cent lower than one insulated with nat ural gutta-percha would have cost. SlJPPLY OF JELOTONG. The average price of jelotong, such as Bandjermas sin, Pontianak, and Sarawak, during the last five years has been 45 to 5 0 pfennigs per kilo (10.7 to 11.9 cents per 2.2 pounds). The price is now 58 to 60 pfennigs (1'l.8 to 14.3 cents) per kilo. Pontianak is the lowest grade, and is generally quoted half a cent below other kinds. At times the price has been as low as 9.1h. cents per kilo. 10 10 * From Ill .. ,<,comi Kelvin lecture, delivercdl\t the Institution of Elec trical Engineers. T\,X'HNIC' AL SUBSTITUTES. Several methods have been applied to produce arti .ticial gutta-percha surrogates. According to two Ger man patents, water, either with or without an addi tion of salt, is dripped on suitable substances-wax, rellin, asphalt, tar, or pitch-heated to a temperature above 100 deg. C. (212 deg. F.), whereby the melting point and lent our of the latter is considerably in- © 1910 SCIENTIFIC AMERICAN, INC. B Y PRO F. J. A. E WIN G. Jackson's lectures for graduate students on the organization and administration of public service com panies have this year dealt more particularly with questions of value of plant, the theory of so-called in tangible values, the relation of revenues to value of the plant, depredation, and the like; and next year the lectures will be directed more to the theory under lying methods or ('harging for servil'p by publil' servi('p (·ompanies. wIth parti('ulal' referen('p to ('harges for electric light and power but with collateral considera- tion of railroad and tramway charges and charges for gas and the service of otller public utilities. Prof. Wickenden will originate a course of lectures on il lumination, photometry, and illuminating engineering which will become a part of the optional curriculum for undergraduates and graduate students. advanced lectures of this year on the high-voltage alternating transmission and utilization of power. The general treatment of the transmission circuit contained in his lectures of this year will be r.epeated and ex tended, and more attention will be given to the con ditions arising from the utilization of the power. Prof. Jackson's lectures for graduate students on the organization and administration of public service com panies have this year dealt more particularly with questions of value of plant, the theory of so-called in tangible values, the relation of revenues to value of the plant, depredation, and the like; and next year the lectures will be directed more to the theory under lying methods or ('harging for servil'p by publil' servi('p (·ompanies. wIth parti('ulal' referen('p to ('harges for electric light and power but with collateral considera- Some interesting f,;,cts are being brought out by investigations of the effect of high voltages on in sulating material by Mr. H. S. Osborne, who is carry ing out work for the degree of Doctor of Engineering at the Massachusetts Institute of Technology. At a recent meeting of the Boston section of the American Institute of Electrical Engineers, which was held at the electrical engineering lab()ratories of the Institute of Technology, Mr. Osborne lectured on the results of his experimental research. He briefly presented the re sults in an interesting and effective manner, and set forth theories of the effect of high voltage on inSUlat ing materials which are in closer harmony with ex nerimental facts than the theorieR usually stated . The lednres of Prof. B Y PRO F. J. A. E WIN G. as the period of oscillation of the card, the disturbance would become so great as to make steering by compass im possible. It was to secure steadiness in this sense that Kelvin strove to give his compass card a long period of oscillation, recognizing that the right way to obtain steadiness was to make the lleriod much longer than the period of the slowest rolling motion liable to occur in a ship, at the same time keeping the friction as small as possible. The problem of secur ing a steady, frictionless compass was a problem where, as in tke invention of the mirror galvano meter, his genius for practical dynamics guided him In the navigational sounding machine we have an other invention of first-rate importance, second only to the compass in practical value to sailors, and re markable for its extreme simplicity. It was his cable laying experience that first led Kelvin to take an in terest in d̢ep-sea sounding. The process, as then car ried out, was a laborious one. The line was a rope an inch and a half in circumference, and though it car ried a very heavy sinker, the resistance to its motion through the water was so great that it took a long time to reach the bottom. For the same reason the ship had to be stopped while the line ran out, and, except in shallow water, while it was being heaved in. many hands were needed, and much time was spent in making a cast. Hence it came about that the operation of sounding, beyond the use of the hand lead in quite shallọ water, was but little resorted to as an aid tv navigation, notwithstanding the impor tance of the indications it could give in such cases as when a ship was approaching land in a fog or in circumstances which made the exact position uncer: tain, when the depth might be anything up to, say, one or two hundred fathoms. In attempting this account of the work 'of Kelvin in telegraphy and navigation, I am embarrassed by its volume and its range. The time has proved far too short for a fitting notice of discoveries and inven tions so various, so fundamental, so far-reaching in their practical effects. B Y PRO F. J. A. E WIN G. The evolution of the Kelvin compass, in its main features, took about five years; but a longer task lay before the inventor in overcoming the professional conservatism of sailors, the objections of the so-called practical man, active hostility in, some quarters, and the passive resistance of official inertia. Gradually the compass came to be used in merchant vessels of the best appointed class. Enlightened navigators such as Capt. Lecky, the author of the well-known "Wrin kles," became its enthusiastic advocates. Foreign ad miralties took it up, and in our own service individ ual officers were quick to see its merits. Capt. Fisher, now Admiral of the Fleet Lord Fisher, was warm in its praise after observing its behavior in ships under his command, first in the "Northampton" in rough weather and afterward in the "Inflexible" during the firing of heavy guns in the bombardment of Alexan dria. That was in 1882; but it was no.t until No vember, 18S!:!, that the superintendent of the Compass Department of the Admiralty was in a position to inform Lord Kelvin that his 10-inch compass was to be adopted as the standard compass for the navy. This was twelve years after the date of his patent, and more than eleven years after he had laid the invention formally before the First Lord. The way of the inventor, like that of the transgressor, may still be hard, but I trust it is not so hard now as it was then. One does not care to dwell on the specta cle of a Kelvin spending his strength in dishearten ing effort as the sea beats against a cliff. It is pain ful to read the correspondence and discussion of these weary years. One does it with increased ad miration of the infinite patience which at last secured to us the benefits of his practical genius. This machine has become a standard navigational appliance., The length of wire in common use is 300 fathoms. A strand of seven tine steel wires, which gives greater flexibility, is now substituted for the single wire. It runs out under a regulated tension, supplied by a rope brake, which retards the rotation of the drum on which the wire is wound. B Y PRO F. J. A. E WIN G. so that the magnetism induced on it will act as a counter-balance. This is the Flinders bar, so called because its use was pointed out by Capt. j<'linders as early as 1801. It has generally to be fixed in front of the binnacle, and in Kelvin's com pass it is made in several separate lengths of soft iron, which can be put together to make up a bar giving' any necessary amount of correcting effect. The use of the Kelvin compass may now be said to be universal, except that in the navy a modified form, due to Capt. Chetwynd, with a card immersed in liquid, is taking the place of the Kelvin dry card in the newer ships as being steadier still under gun-fire. The system of correction remains substantially un changed, and the compass continues to embody the same mechanical features as formed the basis of Kel vin's invention. The main function of the Flinders bar is to correct the semicircular error due to induced vertical mag netism. So far as the heeling error is concerned it also helps, but in practice it is found convenient to correct a part of the heeling error due to induced magnetism' by means of the same kind of permanent magnet correctors as I have already described in speaking of the heeling error due to permanent mag netism, namely, vertical magnet bars placed in a can in the binnacle directly under the center of the compa!ls card. The number and height of these bars has therefore to be altered from time to time, as the ship moves to regions where the vertical force is different. When the heeling error is fully corrected we escape one cause of the unsteadiness which il com pass shows when a ship rolls, for we escape the mag netic cause of ol5cillation, namely, the alternate mag netic pull to port and starboard; but a purely dy namical cause of unsteadiness necessarily remains, arising from the fact that the point of suspension of a compass card must be placed some way from the center of gravity to hold the car level against the dipping action of the earth's magnetic field. Conse quently, every roll to either side applies a mechanical couple tending to set up oscillation, and if the period of the roll were the same, or nearly the same. B Y PRO F. J. A. E WIN G. So in a Rhip'l'l cOm]laRR thiR error. due to the purely transient magnetism induced by the horizontal component of Another feature is that the bowl and gimballs as a whole is hung from springs to withstand vibration caused by the action of the screw, or in warships by gun fire. Now as to the correction for the magnetism of the ship. Let me indicate very briefly the nature of that problem, and how it is solved. An iron ship is a great magnet, or rather a great aggregate of many magnets. Her magnetism at any instant springs from two causes. First, there is the more or less permanent part, which she takes up first wheu she is built; it depends to a great extent oil how her head lay while she was on the stocks. Then there is the induced part, which changes with every change of course-a transient effect due to the induction of the earth's magnetic field. Strictly speaking, the induced magnetism is not entirely transient, nor is the other by any means entirely permanent; but the ideal divij;)ion into transient and permanent is a highly llseful one ]lrovided we under stand the limitation within which it is to be acce]lted. Now think of what happens when the ship is "swung," He recognized that for this purpose the needles must be short. Further, that for steadiness what was wanted was a long period of horizontal oscilla tion-in other words, small magnetic moment rela tively to the moment of inertia of the card; but, to keep the frictional error down, the weight of the card, including the needles, should be small. So he © 1910 SCIENTIFIC AMERICAN, INC. SCIENTIFIC AMERICAN SUPPLEMENT No. 1800. 11 JUI.y 2, ] 910. the earth's field, has its maximum on these four courses, one in each quadrant, and for that reason it is called the quadrantal error. later a compact form of navigational sounding ma chine by which flying soundings are taken without stopping the ship. to the right solution. In the case of the compass it was rendered difficult by the fact that other condi tions, apparently antagonistic, had at the same time to be satisfied, in order that the correction of mag netic errors might be completely carried out. B Y PRO F. J. A. E WIN G. When the sinker touches bottom the tension is at once seen to 'slacken, or rather felt to slacken by a sailor who keeps a little rod of wood lightly pressed against the wire while it runs out; the drum is stopped, and the wire is slowly wound in again by hand, or in the latest naval type by electric motor. Lord Kelvin's final improvements in the machine were made only a year or so before his death; they were, in fact, his last ser.ious inventive work. They include a large hori zontal ' dial for reading the number of fathoms of wire out, and with this it is often practicable to tell tl:J.e depth very closely without resorting to a depth £age at all; for in the modern machine the action is so uniform that, at any given speed of ship, a definite relation holds between the depth and the length of wire out, and by finding this relation once for all a table can be prepared by which the speed is Imown, and so when the length of wire out is ob served the depth may be at once inferred. This sys tem is now in regular use in the navy. A pair of the Kelvin machines stands on the bridge; the wire runs out along a boom at either side and over an ingeni ously designed pulley or fair-lead; whenever sound ings are wanted they can be taken systeJ)1atically and in quick succession while the ship proceeds at undim inished speed, and the depth is called out for the information of the navigating officer almost as soon as the wire had stopped running out. Alike in the navy and the merchant service, there -is no difficulty in making it a matter of routine to keep the sounding machines going incessantly when near shore or within, say, a hundred fathoms in thick weather. One more of these disturbing causes remains to be mentioned. The vertical component of the earth's fluid induces magnetism as well as the horizontal component, and gives rise to an additional error of two kinds, namely, a further semicircular error and a further heeling error. These are distinct from the semicircular error and heeling error due to perma nimt magnetism, and the right way to correct them is to fix a bar of soft iron in a vertical positiont near the binnacle. B Y PRO F. J. A. E WIN G. It is due, as we have seen, to the ship's being a long body, extending fore and aft, and it is corrected by balancing this excess of· fore and aft iron by other iron, placed quite near the compass and on either side of it. The two balls which you see on the side of the Kelvin binnacle are the correctors for quad rantal error. They are adjusted, in the first place, by selecting a suitable size of ball, and then placing them nearer to or farther from the compass until, on swing ing the ship, the quadrantal error disappears. The possibility of correcting the quadrantal error in this way had been pointed out by Airy as early as 184 0; but with the old form of compass card and needles it could not be done, because of the excessive length and large magnetic moment of the needles. To apply the method to a compass of the old pattern would have needed globes of impracticable size, not a few inches in diameter as these are, but weighing tons. Kelvin, with his short needles on a light card, made it possible to carry out the process, and so gave the world, for the first time, a compass that would point truly to the magnetic north, notwithstanding all the ,;Jerturbations due to permanent and induced mag netism in the iron of the ship. in a flying sounding the wire streams out behind, taking an oblique course to the bottom, and the length of wire that runs out is greatly in excess of the depth. To read the depth directly, Thomson invented sev eral forms of depth gage, the simplest of which is a long narrow glass tube, closed at the top, and coated inside with chromate of silver or some other chemi cal which is discolored by the action of sea water. This tube is put in a protecting case, which is at tached near the sinker, and as it descends the in creased pressure forces the sea water up into it, com pressing the air, and indicating the depth by the height to which the chemical lining is discolored. Accordingly, the depth is read off by laying the tube against a scale, when the line is drawn on board. 1 That is Lo say, vertical when the ship is on even keelr or perpendicular to the deck. B Y PRO F. J. A. E WIN G. Yet we have dealt only with a very small part of the whole achievement of a man not less remarkable for sustained industry than for outstanding originality-a man incessant in action and in thought-of whom it may be truly said that there is no department of physics on which he has not left an abiding impress. I have said nothing of the lofty flights of scien tific imagination, which are, perhaps, his highest title to fame; but I have said enough to show that Kelvin was no mere philosopher with head in the clouds. He was quick to recognize a real need, quic!r also to see how the need should be met. He found material for invention in the most commonplace appliances, be cause his mental haliit was in everything to seek for the how and the why and to ask himself in what way the thing might be done better. He had an infinite faculty of taking pains, of adhering to a purpose until he secured its full accomplishment, of going on from improvement to improvern,ent in pursuit of the more perfect result, and with all this a courage and hope fulness that never accepted defeat. I 'have spoken already of Thomson's study of the forces acting on a cable during its submersion. Ap plying these principles to the sounding-line, he recog nized that to make the line slip down quickly it should have the smallest possible and the smoothest possible surface, and this led him to use a single wire of steel-the steel of high tensile strength used in pianofortes. In 1872 he demonstrated th'e practica bility of using wire by taking a sounding and finding bottom at 2,700 fathoms in the Bay of Biscay with a ';JO-pound sinker and a single wire of No. 22 gage. He soon devised a suitable drum and winding-in wheel for deep-sea use, and from this was developed 1 That is Lo say, vertical when the ship is on even keelr or perpendicular to the deck. advanced lectures of this year on the high-voltage alternating transmission and utilization of power. The general treatment of the transmission circuit contained in his lectures of this year will be r.epeated and ex tended, and more attention will be given to the con ditions arising from the utilization of the power. Prof. B Y PRO F. J. A. E WIN G. Harold Pender for graduate students will next year extend the discussion contained in his A brief paper was presented by Dr. Paul Heroult in which the hypothesis was brought out with great force that gases in steel are harmless, and that they do not produce blowholes. Blowholes are the result of disen gagement of carbon monoxide, due to the interaction of ('arbon and iron peroxide at the timi' of solidi fica- tion. they mer, Hydrogen, )litrogen, ptc., gases arp harmless; are contained to abont the same extent in Besse baSiC, acid open-hearth, crucible or electric steel. © 1910 SCIENTIFIC AMERICAN, INC.
https://openalex.org/W2120939731	https://europepmc.org/articles/pmc2918619?pdf=render	English	null	Tian Xian Liquid (TXL) induces apoptosis in HT-29 colon cancer cell in vitro and inhibits tumor growth in vivo	Chinese medicine	2,010	cc-by	4,903	© 2010 Liu et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Tian Xian Liquid (TXL) induces apoptosis in HT-29 colon cancer cell in vitro and inhibits tumor growth in vivo Qing Liu1, Yao Tong1, Stephen Cho Wing Sze1, Wing Keung Liu2, Lam Lam1, Ellie Shihng Meir Chu1, Christine Miu Ngan Yow3 Abstract Background: Tian Xian Liquid (TXL) is a Chinese medicine decoction and has been used as an anticancer dietary supplement. The present study aims to investigate the effects of TXL on the apoptosis of HT-29 cells and tumor growth in vivo. Method: HT-29 colon cancer cells were treated with gradient dilution of TXL. The mitochondrial membrane potential was measured by JC-1 assay. The release of cytochrome c from mitochondrial and apoptosis-related proteins Bax, Bcl-2, cleaved caspase-3, 9 were examined by Western blot analysis. HT-29 cells were implanted in nude mice to examine the effects of TXL on tumor growth. Result: TXL inhibited HT-29 xenografted model and showed a strong and dose-dependent inhibitory effect on the proliferation of HT-29 cells. Mitochondrial membrane potential was reduced by TXL at the concentration of 0.5% above. For Western blot analysis, an increase in Bax expression and a decrease in Bcl-2 expression were observed in TXL-treated cells. TXL treatment increased the protein level of cleaved casepase-3 and caspase-9, and the release of cytochrome c in cytoplasm was up-regulated as well. Conclusion: TXL significantly inhibits cell proliferation in the HT-29 cells and HT-29 xenografted model via the mitochondrial cell death pathway. Liu et al. Chinese Medicine 2010, 5:25 http://www.cmjournal.org/content/5/1/25 Open Access Correspondence: tongyao@hkucc.hku.hk 1School of Chinese Medicine, Li Ka Shing Faculty of Medicine, University of Hong Kong, Pokfulam, Hong Kong SAR, China © 2010 Liu et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background procaspase-9, thereby triggering caspase-9 activation which subsequently cleaves the effector caspase-3[5,6]. Colorectal carcinoma increased up to four folds in the past decade and the mortality is rising [1]. Much pro- gress has been achieved in alternative medicine [2] such as Chinese medicine. Tian Xian Liquid (TXL), an aqueous extraction of Chi- nese medicinal herbs including Radix Ginseng, Cordyceps, Radix Astragali, Radix Glycyrrhizae, Rhizoma Dioscorea, Margarita, Fructus Lycii, Ganoderma, Fructus Ligustri Lucidi, Herba Scutellariae Barbatae, has been used as an anticancer dietary supplement for more than a decade [7]. Previous experiments reported that TXL had inhibitory effects on human cervical carcinoma C-33A cells and human lung carcinoma H1299 cells[7]. The present study aims to investigate the effects of TXL on the apoptosis of HT-29 cells and tumor growth in vivo. Most methods of chemotherapy for cancer induce cancer cell apoptosis. Excessive apoptosis causes hypo- trophy such as ischemic damage whereas insufficient apoptosis leads to uncontrolled cell proliferation such as cancer [3]. Chemotherapeutic agents may cause mito- chondrial dysfunction leading to depolarization of the inner mitochondrial membrane potential (Δψm) [4], triggering the caspases cascade by releasing several cas- pase activators. Among them, cytochrome c activates caspases by forming a complex with Apaf-1 and DAPI staining DAPI (Sigma, USA) (4’ 6-diamidino 2-phenylindole)- stained nuclei were observed with fluorescence micro- scopy. HT-29 cells (70-80% confluent) in 24-well uncoated plates were exposed to 0.5% and 1% TXL for 24 hours respectively. Cells were fixed with 4% parafor- maldehyde for 30 minutes and incubated with 1 μg/mL DAPI solution for 30 minutes in the dark. Stained cells were imaged under a fluorescence microscope (Carl Zeiss, Germany). Preparation of TXL Tian Xian Liquid (TXL) (Batch number: L2-171040) was provided by China-Japan Feida Union Company Ltd. and stored away from light at 4°C. TXL was diluted and incorporated into the cell culture medium RPMI 1640. Residues were removed by filtration. Western blot The HT29 cells were incubated with increasing concen- trations of TXL (0, 0.5%, 0.75%, 1%) for 48 hours. For the time-course experiment, HT29 were treated with 1% TXL for 12, 24, or 48 hours. Cellular levels of cleaved caspase-3, 9 (Cell Signaling Technology, USA) Bax/Bcl-2 cytochrome C and glyceraldehyde 3-phosphate dehydro- genase (GAPDH) (Santa Cruz Biotechnology, USA) were determined by Western blot. Lysates were prepared from 1 × 107 cells by dissolving cell pellets in 100 μl of lysis buffer. Lysates were centrifuged (Eppendorf, Ger- many) at 18000× g for 15 minutes and the supernatant was collected. The protein concentration was estimated with the Bio-Rad protein assay kit (Bio-Rad, USA) using bovine serum albumin as a standard. Sample proteins were resolved by 10% sodium dodecylsulfate polyacryla- mide gel (Bio-Rad, USA) electrophoresis and then elec- trophoretically transferred to PVDF membrane (Millipore, USA) and blocked with 5% BSA (Sigma, USA). Subsequently the primary antibodies caspase-9, cleaved caspase3, Bax, Bcl-2, cytochrome C and GAPDH were added. After overnight incubation at 4°C the blots were washed, exposed to HRP-conjugated cor- responding secondary antibodies for one hour and finally were visualized by ECL Advanced Solution (GE Healthcare Life Sciences, USA). Digital images were cap- tured by Gel Doc™gel documentation system (Bio-Rad, USA) and intensity was quantified using Quantity-One software version 4.62(Bio-Rad, USA). Cell proliferation Cell proliferation was assessed in vitro with 3-(4,5- Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide (MTT) according to the manufacturer’s protocol (Roche, USA). HT-29 cells (10000 per well) were incubated in triplicates in a 96-well plate. TXL was serially diluted with RPMI1640 and the final concentrations were 0.25, 0.5, 1, 2 and 5%. The plates were incubated with or without TXL for 24 and 48 hours. At the end of the incubation, cells were exposed to MTT (10 μL, 5 mg/ mL in phosphate-buffered saline) in culture medium for four hours at 37°C. The supernatant was removed and 150 μL DMSO (Sigma, USA) was added to dissolve the formazan crystals. The absorbance was measured at 595 nm with an ELISA plate reader (Bio-Rad, USA). Assessment of apoptosis by determination of mitochondrial membrane potential Mitochondrial membrane potential was assessed by 5, 5’, 6, 6’-tetrachloro-1, 1’, 3, 3’tetraethylbenzimidazolylcarbo- cyanine iodide (JC-1) according to the manufacturer’s protocol (Biotium, USA). After trypsinization and centri- fugation (500× g)(Eppendorf, Germany) for ten minutes at room temperature, the pellets of cell culture with or without TXL were re-suspended in RPMI 1640 medium (1 ml), stained with 5 mg/ml JC-1 for 30 minutes at 37°C in the dark, washed twice in phosphate buffered saline (PBS) and re-suspended in 0.5 ml PBS. Δψm depletion was observed under a fluorescence Cell culture Human colon cancer cell HT-29 (ATCC® Number:HTB- 38™) was obtained from the American Type Culture Page 2 of 7 Liu et al. Chinese Medicine 2010, 5:25 http://www.cmjournal.org/content/5/1/25 Collection (ATCC, USA) and cultured in RPMI 1640 (Hyclone, USA) supplemented with fetal bovine serum (10%), penicillin (100 units/ml) and streptomycin (100 mg/ml) (Hyclone, USA) in a humidified incubator (37°C) containing 95% air and 5% CO2. Trypsin (Hyclone, USA) was used for trypsination. microscope. A green filter was used for green-fluores- cent monomer at depolarized membrane potentials and a red filter for orange-fluorescent J-aggregate at hyper- polarized membrane potentials. Collection (ATCC, USA) and cultured in RPMI 1640 (Hyclone, USA) supplemented with fetal bovine serum (10%), penicillin (100 units/ml) and streptomycin (100 mg/ml) (Hyclone, USA) in a humidified incubator (37°C) containing 95% air and 5% CO2. Trypsin (Hyclone, USA) was used for trypsination. To measure the quantitative change of mitochondrial potential, we applied JC-1 with fluorescence plate reader. Briefly, cells (1 × 105) in 100 μl culture medium/ well were seeded in black 96-well plate (Nunc, Den- mark) and treated with TXL (0.15, 0.3, 0.6, 1.25 and 2.5%). After 24 and 48 hours incubation, JC-1 (5 μg/ml) was added for the last 30 minutes of treatment. Cells were washed twice with PBS to remove unbound dye. The concentration of retained JC-1 dye was measured (490 nm excitation/600 nm emission) with a lumines- cence spectrometer (PerkinElmer, USA). Anti-proliferative and apoptotic effects of TXL on HT-29 cells To investigate the anti-proliferative effects of TXL on HT-29 cells, we treated the HT-29 cells with TXL in a gradient of doses (0.25-5%) and cell proliferation after two days was assessed with the MTT assay in triplicates. The results were consistent. TXL inhibited HT-29 cell proliferation in a dose-dependent manner (Figure 1). Treatment of TXL (1%) for 48 hour significantly inhib- ited (38.47%; P < 0.05, P = 0.011) cell proliferation. Effects of TXL on cell nuclear morphology Nuclear staining with DAPI was used to determine apoptosis-inducing activity of TXL in HT-29 cells. After TXL (1%) treatment, HT-29 cells underwent typical morphologic changes of apoptosis including nuclear condensation and formation of apoptosis bodies (Figure 1). In vivo tumor-growth inhibition studies The experiment was repeated three times with similar results. CTL: control. Figure 1 TXL’s inhibitory effects on cellular growth and Figure 1 TXL’s inhibitory effects on cellular growth and apoptosis in HT-29 cells. (A) Apoptosis in HT-29 after TXL Statistical analysis Data were presented as mean and standard deviation (SD). When one-way ANOVA showed significant differ- ences among groups, Tukey’s post hoc test was used to determine the specific pairs of groups that were statisti- cally different. A level of P < 0.05 was considered statis- tically significant. Analysis was performed with the software SPSS version 16.0 (SPSS Inc, USA). J-aggregation fluorescence whereas TXL-treated cells had green fluorescence (Figure 2A). JC-1 staining was used to determine mitochondrial integrity. To quantify the change of mitochondrial potential, we applied JC-1 with fluorescence plate reader. The green to red fluores- cence ratio significantly decreased at 48 hours in a dose- dependent manner (Figure 2B). TXL triggers interaction between Bcl-2 and Bax and releases cytochrome c Low Δψm is regulated by Bcl-2 family proteins [11]. We studied the effects of TXL on the expression of Bax and Bcl-2 which are important for mitochondrial membrane permeablization. In this study, the HT-29 cells were incubated with increasing concentrations of TXL (0, 0.5%, 0.75%, 1%) for 48 hours. For the time-course experiment, HT29 were treated with 1% TXL for 12, 24, or 48 h. Cell lysates were prepared for western blot ana- lysis. After 48 hours, HT-29 cells treated with 1% TXL showed significant up-regulation (P = 0.003) in Bax expression (Figure 3) while significant down-regulation (P = 0.013) in Bcl-2 expression (Figure 3). TXL (0.75%) and TXL (1%) increased the Bax/Bcl-2 ratio by 1.4 and 2.8 folds respectively in HT-29 cells. Stability of mito- chondrial membrane is influenced by the interactions among Bcl-2 family proteins, thereby affecting the release of cytochrome c from mitochondria to and In vivo tumor-growth inhibition studies The experiment was approved by the Department of Health, Hong Kong SAR, China and the Committee on the Use of Live Animals in Teaching and Research (CULATR) of Li Ka Shing Faculty of Medicine, Univer- sity of Hong Kong. Six-week-old female nude mice were purchased from the Laboratory Animal Unit, University of Hong Kong and kept under sterile conditions in Liu et al. Chinese Medicine 2010, 5:25 http://www.cmjournal.org/content/5/1/25 Page 3 of 7 accordance with the institutional guidelines of animal care. The HT-29 carcinoma was established in nude mice by injecting the suspensions of HT-29 (1 × 106 cells per animal) [8] cells subcutaneously into the right flank of each animal. When the tumors became palpable (size: 18 mm3) after xenografting, mice were divided into three groups (n = 8) by a random numbered table: (1) Control group orally administered with 200 μl PBS); (2) 5-fluorouracil (5-FU) (Choongwae, Korea) group (injected intraperitoneally with 5-FU, 30 mg per kg of body weight) three times a week [9,10]; (3) TXL group (orally administered with 200 μl TXL daily for 14 days. To evaluate the antitumor activity of TXL, we measured the tumor volume with a digital caliper six times every week (from day1 to day 6 and from day 8 to day 14) and calculated using the formula: (longest diameter) × (shortest diameter)2 × 0.5. The body weights of all ani- mals were recorded throughout the experiment to assess drug toxicity. Figure 1 TXL’s inhibitory effects on cellular growth and apoptosis in HT-29 cells. (A) Apoptosis in HT-29 after TXL treatment for 48 hours was determined by staining the cell with DAPI. Apoptotic cell exhibiting characteristic chromatin condensation were observed by fluorescence microscopy. (B) Cell proliferation was assessed after 24 and 48 hours with the MTT assay as described in Methods. Results (optical densities) are expressed as mean and standard deviation (n = 3). The experiment was repeated three times with similar results. CTL: control. Figure 1 TXL’s inhibitory effects on cellular growth and apoptosis in HT-29 cells. (A) Apoptosis in HT-29 after TXL treatment for 48 hours was determined by staining the cell with DAPI. Apoptotic cell exhibiting characteristic chromatin condensation were observed by fluorescence microscopy. (B) Cell proliferation was assessed after 24 and 48 hours with the MTT assay as described in Methods. Results (optical densities) are expressed as mean and standard deviation (n = 3). Treatment of TXL reduces the mitochondrial membrane potential JC-1, a cationic dye, produces red fluorescent J-aggre- gates in mitochondria with high Δψm and green fluores- cence with low Δψm. Most control cells had red Page 4 of 7 Liu et al. Chinese Medicine 2010, 5:25 http://www.cmjournal.org/content/5/1/25 0.03) decreased from day 13 to day 15. The difference in tumor size in the TXL group (P = 0.933) was not signif- icant compared with the 5-FU group (P = 0.99). Figure 2 TXL’s effect on the depolarization of HT-29 mitochondria. (A) JC-1 staining observed by fluorescence microscopy. TXL treated cells showed a majority of cells stained green dye due to low mitochondrial membrane potential. (B) Effect of TXL on the depolarization of HT-29 mitochondria was also measured by fluorescence plate reader using JC-1. The cells were exposed to increasing concentrations of TXL. Data represent mean and standard deviation of three individuals with asterisks denoting significant differences between controls and TXL-exposed cells (P < 0.05, P < 0.01). CTL: control. Discussion Data represent mean and standard deviation of three individuals with asterisks denoting significant differences between controls and TXL-exposed cells (P < 0.05, *P < 0.01). CTL: control. subsequently accumulation in the cytosol [12]. The cyto- sol levels of cytochrome c in HT-29 cells were examined with Western blot. In HT-29 cells treated with TXL, cytochrome c significantly increased in a dose-depen- dent (P = 0.0096) and time-dependent manner (P = 0.001). A previous study showed that mitochondrial mem- brane disruption and the release of cytochrome c was controlled by Bcl-2 family protein [11]. Bcl-2 and other pro-apoptotic factors prevent mitochondrial membrane disruption while Bax promotes these events. To clarify whether Bcl-2 family was changed in TXL treated HT-29 cell to activate the release of cytochrome c, we examined the expression level of Bcl-2 and Bax with or without the TXL treatment. An increasing Bax and a decreasing Bcl-2 were observed in a time-dependent manner after exposed to 1% TXL. Our results showed that TXL induced apoptosis by increasing the Bax/Bcl-2 ratios. These observations confirmed that TXL induced apoptosis in colon cancer via the mitochondrial path- way. The above concomitant molecular events in TXL- treated HT-29 cells result in remarkable apoptosis pro- cess. Further in vivo and in vitro studies are needed to clarify the protein interactions, thereby delineating the upstream regulatory events, such as the Wnt signaling pathway which is important factor in the development of the majority of colorectal cancers[14]. Discussion Most chemotherapeutic drugs induce cancer cell apop- tosis whereby a cell activates its own destruction by initiating a series of cascading events including the loss of the mitochondrial transmembrane potential [6]. A rapid collapse of mitochondrial transmembrane electri- cal potential Δψm is always found in chemotherapeutic agents-induced apoptosis in cancer cells [13]. The pre- sent study demonstrated that TXL-induced apoptosis was related to the collapse of the mitochondrial mem- brane potential Δψm. Our study showed the depletion of Δψm (Figure 2) and the activation of caspase-3 of HT-29 treated with TXL. Mitochondria participate in apoptosis induction by releasing several caspase activators. Among them, cyto- chrome c activates caspases by forming a complex with Apaf-1 and procaspase-9, thereby triggering caspase-9 activation which subsequently cleaves the effector cas- pase-3 [6]. The present study found that 1% of TXL induced the cleavage of caspase-3 to its active form, namely p17 (Figure 4). The fragment, p17 (17 kDa), was accumulated after 24 hours of TXL treatment. In this study, we also observed that TXL remarkably increased the release of cytochrome c from the mitochondria to the cytosol in HT-29 cells. Levels of cytochrome c in the cytosolic fraction increased dramatically when the dosage of TXL was 0.5% or above. These results suggest a direct link between the mitochondria and the TXL- induced apoptosis. Figure 2 TXL’s effect on the depolarization of HT-29 mitochondria. (A) JC-1 staining observed by fluorescence microscopy. TXL treated cells showed a majority of cells stained green dye due to low mitochondrial membrane potential. (B) Effect of TXL on the depolarization of HT-29 mitochondria was also measured by fluorescence plate reader using JC-1. The cells were exposed to increasing concentrations of TXL. Data represent mean and standard deviation of three individuals with asterisks denoting significant differences between controls and TXL-exposed cells (P < 0.05, *P < 0.01). CTL: control. Figure 2 TXL’s effect on the depolarization of HT-29 mitochondria (A) JC-1 staining observed by fluorescence Figure 2 TXL’s effect on the depolarization of HT-29 mitochondria. (A) JC-1 staining observed by fluorescence microscopy. TXL treated cells showed a majority of cells stained green dye due to low mitochondrial membrane potential. (B) Effect of TXL on the depolarization of HT-29 mitochondria was also measured by fluorescence plate reader using JC-1. The cells were exposed to increasing concentrations of TXL. TXL induces caspase-3 and caspase-9 cleavage To confirm the induction of the mitochondrial-mediated apoptosis, we examined the activation of the intrinsic initiator caspase-9 and casapse-3 using western blot. TXL (0.75% and 1%) induced the cleavage of caspase-3 to its active form, i.e. p17 (17 kDa) which was found after 24 hours of TXL treatment (Figure 4). As shown in Figure 4, caspases-9 in HT29 cells treated with TXL was activated, as judged by the decrease of the procas- pases-9 and the increase of their cleavage products. In vivo effects of TXL on HT-29 tumor growth To determine the antitumor efficacy of TXL as a single agent therapy, we examined the growth of HT-29 cells in immunocompromised mice. Compared with mice orally administrated with 200 μl PBS as control group, treatment with the TXL and 5-FU significantly inhibited tumor growth (Figure 5A). After treatment of nude mice with TXL, the tumor size was significantly (P = Liu et al. Chinese Medicine 2010, 5:25 http://www.cmjournal.org/content/5/1/25 Page 5 of 7 Figure 3 Effect of TXL on cytochrome c. Total protein from HT-29 treated with 1%TXL for 12, 24 and 48 hours (A) or indicated concentration of TXL for 48 hours (B) were analyzed by Western blot with specific antibodies against Bax and Bcl-2. Protein from cytosolic fraction of HT-29 which has been treated with TXL (1%) for 12, 24 and 48 hours (C) or indicated concentration of TXL for 48 hours (D) were analyzed by Western blot with specific antibodies against cytochrome c. GAPDH antibody was used as control for equal loading. The relative expressions of proteins were quantified using Bio-Rad Quatity-One software. Results are expressed as mean and standard deviation (n = 3),P < 0.05 *P < 0.01compared with control group. Figure 3 Effect of TXL on cytochrome c. Total protein from HT-29 treated with 1%TXL for 12, 24 and 48 hours (A) or indicated concentration of TXL for 48 hours (B) were analyzed by Western blot with specific antibodies against Bax and Bcl-2. Protein from cytosolic fraction of HT-29 which has been treated with TXL (1%) for 12, 24 and 48 hours (C) or indicated concentration of TXL for 48 hours (D) were analyzed by Western blot with specific antibodies against cytochrome c. GAPDH antibody was used as control for equal loading. The relative expressions of proteins were quantified using Bio-Rad Quatity-One software. Results are expressed as mean and standard deviation (n = 3),P < 0.05 *P < 0.01compared with control group. Figure 4 TXL Induces caspase-3 and caspase-9 cleavage. Protein from HT-29 which has been treated with 1%TXL for 12, 24 and 48 hours (A) and protein from HT-29 which has been treated with indicated concentration of TXL for 48 hours (B) were analyzed by Western blot with specific antibodies against caspase-9 and cleaved caspase-3. GAPDH antibody was used as control for equal loading. The relative expressions of proteins were quantified using Bio-Rad Quatity-One software. Authors’ contributions Because 5-fluorouracil (5-FU) as common choice for single-agent chemotherapy of advanced colon cancer is also recognized for its toxicity including fatigue, diarrhea and sometimes myelosuppression, much attention has been focused on exploring complementary and alterna- tive medicine. This study may provide a platform for evaluation the function of Chinese medicine decoctions on treatment of cancer, which has no significant side effect. To further evaluate the potential of TXL as an adjuvant agent in colon cancer chemotherapy, we are studying TXL’s effect on attenuation of 5-FU-induced side effect and the synergistic anti-tumor effect of the 5-FU/TXL. QL performed the experiments, analyzed data and drafted the manuscript. YT and SCWS designed the study and revised the manuscript. ESMC and LL conducted the in vivo experiments. WKL designed the in vivo experiment and prepared the human colon cancer cells. CMNY designed the in vitro experiment. All authors read and approved the final manuscript. Author details 1School of Chinese Medicine, Li Ka Shing Faculty of Medicine, University of Hong Kong, Pokfulam, Hong Kong SAR, China. 2School of Biomedical Sciences, Faculty of Medicine, Chinese University of Hong Kong, Shatin, NT, Hong Kong SAR, China. 3Medical Laboratory Science Section, Department of Health Technology and Informatics, Hong Kong Polytechnic University, Hung Hom, Hong Kong SAR, China. Abbreviations TXL: Tian Xian Liquid; 5-FU: 5-fluorouracil; Δψm: mitochondrial membrane potential; MTT: 3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide; DAPI: 4’ 6-diamidino 2-phenylindole; JC-1: 5, 5’, 6, 6’-tetrachloro-1, 1’, 3, 3’tetraethylbenzimidazolylcarbocyanine iodide; PBS: phosphate buffered saline; SD: standard deviation; GAPDH: glyceraldehyde 3-phosphate dehydrogenase. Acknowledgements This research was supported by a grant from Seed Funding Programme for Applied Research (200807160015), Small Project Funding (200807176239), the University of Hong Kong and the contract research funding from China- Japan Feida Union Company Ltd. Competing interests This research has received a grant from China-Japan Feida Union Company Ltd. This research has received a grant from China-Japan Feida Union Company Ltd. This research has received a grant from China-Japan Feida Union Company Ltd. References 1. Sung JJ, Lau JY, Goh KL, Leung WK: Increasing incidence of colorectal cancer in Asia: implications for screening. Lancet Oncol 2005, 6:871-876. 1. Sung JJ, Lau JY, Goh KL, Leung WK: Increasing incidence of colorectal cancer in Asia: implications for screening. Lancet Oncol 2005, 6:871-876. 2. Nelson PS, Montgomery B: Unconventional therapy for prostate cancer: good, bad or questionable? Nature Rev 2003, 3:845-858. 3. Thompson CB: Apoptosis in the pathogenesis and treatment of disease. Science 1995, 267:1456-1462. 4. Zamzami N, Marchetti P, Castedo M, Decaudin D, Macho A, Hirsch T, In vivo effects of TXL on HT-29 tumor growth Results are expressed as means and standard deviation (n = 3),P < 0.05 *P < 0.01 compared with control group. Figure 4 TXL Induces caspase-3 and caspase-9 cleavage. Protein from HT-29 which has been treated with 1%TXL for 12, 24 and 48 hours (A) and protein from HT-29 which has been treated with indicated concentration of TXL for 48 hours (B) were analyzed by Western blot with specific antibodies against caspase-9 and cleaved caspase-3. GAPDH antibody was used as control for equal loading. The relative expressions of proteins were quantified using Bio-Rad Quatity-One software. Results are expressed as means and standard deviation (n = 3),P < 0.05 *P < 0.01 compared with control group. Liu et al. Chinese Medicine 2010, 5:25 http://www.cmjournal.org/content/5/1/25 Page 6 of 7 Figure 5 Effect of TXL on tumor volume (A) and body weight (B) in HT-29-bearing mouse model. HT-29 cells were injected in nude mice as described in Methods. A total of 24 mice were divided into three groups. Data are expressed as mean and standard deviation (n = 8). Significantly difference compared with control group:+ P < 0.05, *++ P < 0.01. Figure 5 Effect of TXL on tumor volume (A) and body weight (B) in HT-29-bearing mouse model. HT-29 cells were injected in nude mice as described in Methods. A total of 24 mice were divided into three groups. Data are expressed as mean and standard deviation (n = 8). Significantly difference compared with control group:+ P < 0.05, **++ P < 0.01. We studied TXL’s effects on the growth of HT-29 cell lines grown in vitro and compared those results with its effect on tumor growth in vivo. We found that TXL attenuated the growth of xenografted HT-29 tumors in vivo. The injection of 5-fluorouracil (5-FU), a common choice for single-agent chemotherapy of advanced colon cancer [15,16] on nude mice significantly attenuated tumor growth (Figure 5A). However, during the 5-FU treatment, the body weight of nude mice was signifi- cantly decreased on day 10 (P < 0.05, P = 0.038) (Figure 5B). The components of TXL such as Radix Ginseng, Cordyceps, Radix Astragali, Fructus Lycii, Ganoderma are commonly used in China as immune-stimulating agents. Whether TXL’s immunomodulation effect and protection effect on stomach and digestive system redu- cing toxic side effects of 5-FU will be investigated in the future. Liu et al. Chinese Medicine 2010, 5:25 http://www.cmjournal.org/content/5/1/25 Liu et al. Chinese Medicine 2010, 5:25 http://www.cmjournal.org/content/5/1/25 6. Elmore S: Apoptosis: a review of programmed cell death. Toxicol Pathol 2007, 35:495-516. 7. Sun A, Chia JS, Chiang CP, Hsuen SP, Du JL, Wu CW, Wang WB: The chinese herbal medicine Tien-Hsien liquid inhibits cell growth and induces apoptosis in a wide variety of human cancer cells. J Altern Complement Med 2005, 11:245-256. 8. Ninomiya I, Terada I, Yoshizumi T, Takino T, Nagai N, Morita A, Fushida S, Nishimura G, Fujimura T, Ohta T, Miwa K: Anti-metastatic effect of capecitabine on human colon cancer xenografts in nude mouse rectum. Int J Cancer 2004, 112:135-142. 9. Wada H, Nagano H, Yamamoto H, Arai I, Ota H, Nakamura M, Damdinsuren B, Noda T, Marubashi S, Miyamoto A, et al: Combination therapy of interferon-alpha and 5-fluorouracil inhibits tumor angiogenesis in human hepatocellular carcinoma cells by regulating vascular endothelial growth factor and angiopoietins. Oncol Rep 2007, 18:801-809. 10. Tai IT, Dai M, Owen DA, Chen LB: Genome-wide expression analysis of therapy-resistant tumors reveals SPARC as a novel target for cancer therapy. J Clin Invest 2005, 115:1492-1502. y 11. Cory S, Adams JM: The Bcl2 family: regulators of the cellular life-or-death switch. Nature Rev 2002, 2:647-656. 11. Cory S, Adams JM: The Bcl2 family: regulators of the cellular life-or-death switch. Nature Rev 2002, 2:647-656. 12. Gross A, McDonnell JM, Korsmeyer SJ: BCL-2 family members and the mitochondria in apoptosis. Genes Dev 1999, 13:1899-1911. 13. Sen S, D’Incalci M: Apoptosis. Biochemical events and relevance to cancer chemotherapy. FEBS Lett 1992, 307:122-127. 14. Morin P, Sparks A, Korinek V, Barker N, Clevers H, Vogelstein B, Kinzler K: Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC. Science 1997, 275:1787-1790. 15. Wilke HJ, Van Cutsem E: Current treatments and future perspectives in colorectal and gastric cancer. Ann Oncol 2003, 14(Suppl 2):ii49-55. 16. Louvet C, Andre T, Tigaud JM, Gamelin E, Douillard JY, Brunet R, Francois E, Jacob JH, Levoir D, Taamma A, et al: Phase II study of oxaliplatin, fluorouracil, and folinic acid in locally advanced or metastatic gastric cancer patients. J Clin Oncol 2002, 20:4543-4548. doi:10.1186/1749-8546-5-25 Cite this article as: Liu et al.: Tian Xian Liquid (TXL) induces apoptosis in HT-29 colon cancer cell in vitro and inhibits tumor growth in vivo. Chinese Medicine 2010 5:25. doi:10.1186/1749-8546-5-25 Cite this article as: Liu et al.: Tian Xian Liquid (TXL) induces apoptosis in HT-29 colon cancer cell in vitro and inhibits tumor growth in vivo. Chinese Medicine 2010 5:25. Conclusion 3. Thompson CB: Apoptosis in the pathogenesis and treatment of disease. Science 1995, 267:1456-1462. TXL significantly inhibits cell proliferation in the HT-29 cells and HT-29-bearing mouse model. TXL-induced apoptosis is likely achieved through the mitochondrial cell death pathway as indicated by a reduction in mito- chondrial membrane potential, and the decrease of Bcl- 2/Bax ratio and the release of cytochrome c followed by the activation of caspase-3 and caspase-9. 4. Zamzami N, Marchetti P, Castedo M, Decaudin D, Macho A, Hirsch T, Susin SA, Petit PX, Mignotte B, Kroemer G: Sequential reduction of mitochondrial transmembrane potential and generation of reactive oxygen species in early programmed cell death. J Exp Med 1995, 182:367-377. 5. 5. Gamen S, Anel A, Pineiro A, Naval J: Caspases are the main executioners of Fas-mediated apoptosis, irrespective of the ceramide signalling pathway. Cell Death Differ 1998, 5:241-249. Page 7 of 7 Page 7 of 7 Liu et al. Chinese Medicine 2010, 5:25 http://www.cmjournal.org/content/5/1/25 Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit
https://openalex.org/W2616845280	https://europepmc.org/articles/pmc5597566?pdf=render	English	null	Compositional design and Taguchi optimization of hardness properties in silicone-based ocular lenses	Progress in biomaterials	2,017	cc-by	4,334	Prog Biomater (2017) 6:67–74 DOI 10.1007/s40204-017-0065-y Prog Biomater (2017) 6:67–74 DOI 10.1007/s40204-017-0065-y ORIGINAL RESEARCH Mohammad Hanifeh1 • Mojgan Zandi1 • Parvin Shokrollahi1 • Mohammad Atai2 • Ebrahim Ghafarzadeh3 • Fahimeh Askari1 Mohammad Hanifeh1 • Mojgan Zandi1 • Parvin Shokrollahi1 • Mohammad Atai2 • Ebrahim Ghafarzadeh3 • Fahimeh Askari1 Received: 4 July 2016 / Accepted: 14 March 2017 / Published online: 15 May 2017 The Author(s) 2017. This article is an open access publication Abstract A multi-component acrylate-based copolymer system especially designed for application as ocular lenses is developed through free-radical, bulk polymerization of a system containing hydroxyethyl methacrylate, methyl methacrylate, triethylene glycol dimethacrylate, dimethyl itaconate, 3-(trimethoxysilyl) propylmethacrylate, Polyhe- draloligomeric silsesquioxane-acrylate (POSS-acrylate) and AIBN as an initiator. The progress of the reaction was monitored by Fourier transform infrared spectroscopy (FTIR). The effect of increasing concentration of the components on the hardness of the synthesized lenses was measured by Shore Durometer before and after immersion in PBS solutions. Extraction test method was performed to analyze the biocompatibility of the fabricated lenses. In this research the Taguchi method was employed to achieve the optimal hardness property which plays a critical role in ﬁnal application of the lens materials. The Taguchi trial for ocular lens hardness was conﬁgured in an L16 orthogonal array, by ﬁve control factors, each with four level settings. The results showed that 3-(trimethoxysilyl) propyl methacrylate decreases and 2-hydroxyethylmethacrylate increases, polyhedraloligomeric silsesquioxane with a cage-like structure, methyl methacrylate and dimethyl itaconate increase the hardness. Proliferation and growth of the cells showed that there is no toxic substance extracted from the lenses which can interfere with the cell growth. Keywords Silicone acrylate Polyhedraloligomeric silsesquioxane-acrylate Contact lens Hardness Taguchi method Keywords Silicone acrylate Polyhedraloligomeric silsesquioxane-acrylate Contact lens Hardness Taguchi method 3 Lassonde School of Engineering, York University, 4700 Keele Street, 11 Arboretum Ln, Toronto, ON M3J 1P3, Canada & Mojgan Zandi m.zandi@ippi.ac.ir 2 Science Department, Iran Polymer and Petrochemical Institute, Pazhoohesh Blvd., Tehran-Karaj Hwy, Tehran 1497713115, Iran 1 Biomaterials Department, Iran Polymer and Petrochemical Institute, Pazhoohesh Blvd., Tehran-Karaj Hwy, Tehran 1497713115, Iran Introduction In the past few years, a majority of 60% of worldwide population reported to use visual aids such as glasses or contact lenses which would rise steadily in the future (Morgan et al. 2009; Jan-Willem Bruggink et al. 2013). Therefore, a market over tens of billions of dollars per year would be predicted for this global market in ophthalmic industry (Nichols 2009). As a brief history, ocular lenses have developed by the earliest ideas from Leonardo da Vinci to the present day. Firstly, they were made of glasses and lens materials gradually became friendlier by invention of hydrogels which were more comfortable and adjustable. High water content lenses should be thicker for more durability but increasing the thickness impairs the oxygen permeability. In the twenty-ﬁrst century, silicone hydrogels allowed more than 90% of available O2 to reach the eyes and allowed the lenses to be thinner (Gasson and Moriss 2003). These days, the ocular lenses and color lenses are worn by many people for different purposes. 1 Biomaterials Department, Iran Polymer and Petrochemical Institute, Pazhoohesh Blvd., Tehran-Karaj Hwy, Tehran 1497713115, Iran There are three types of lenses; rigid, soft and hybrid lenses due to selected materials and also ﬁnal applications (Mannis et al. 2003). One of the most interesting tech- niques for lens preparation is in situ polymerization of multi-functional acrylic monomers, because a variety of 12 3 Prog Biomater (2017) 6:67–74 68 polymerization methods and systems are possible. Free- radical bulk polymerization of vinyl monomers, charac- terized by auto-acceleration, has been widely investigated. hydrophilicity and water uptake of synthesized lens mate- rials. The Taguchi trial for ocular lens hardness was con- ﬁgured in L16 orthogonal array, by ﬁve control factors, each with four level settings. Therefore, the molar con- centration of initiator (AIBN) and the amount of triethylene glycol dimethacrylate (TEGDMA) as cross-linker has been considered as ﬁxed values of 5.0 mol% and 2.8 mol%, respectively. Table 1 demonstrates the selected factors and levels and Table 2 shows 16 L used in the experimental design of Taguchi method. Understanding the chemical composition and effect of each component on the mechanical properties of the lenses is crucial to design the ocular lenses with the best possible performances. Lenses based on silicone acrylate monomers include the wide range of lenses; soft hydrogel lenses and rigid gas permeable hard ocular lenses (RGPL) (Mannis et al. 2003). Sample preparation and polymerization The amounts of components (based on the data in Table 1) were poured into the Petri dish at 23 C for 30 min being homogeneously stirred using a magnetic stirrer. The weight of each batch was about 5 g. The resulting homogeneous mixture was transferred into the Teﬂon coated moulds and placed in the vacuum oven. Polymerization was followed in two steps; ﬁrstly they were heated up to 65 C for 24 h and secondly, they were post-cured at 45 C for 48 h to complete the reaction. A two-step polymerization protocol was performed where in the ﬁrst step the system was below gelation (60-70 C), and after that it was under diffusion control (40-50 C). Methods FTIR spectroscopy (Equinox 55, Brucker, Germany), hardness Durometer (Shore D) (Santam; Iran), vacuum oven were used to investigate the effect of composition on hardness of the synthesized lens materials. The degree of conversion was calculated from Eq. (1) Materials and methods FTIR analysis was used to identify functional groups well as evaluation of degree of conversion of involved monomers (Sangermano et al. 2012). For this technique, the specimens were pulverized into a ﬁne powder and 5 mg of the ground powder were thoroughly mixed with KBr powder and then pressed using pelletingpresswith a load of 10 tons over 1 min to obtain a pellet. The pellets were placed into a sample holder and into the FTIR spectrophotometer. Introduction The RGPL are made in rod-shape blank with a max- imum initial diameter of 5 cm and then cut to form discs with desired thickness. This rod-shape blanks are purchased by vision centers. Ocular lenses are produced by cutting this rod-shape blanks on a lathe, polishing and ﬁnishing due to required curvatures. In other word, RGD lenses due to the presence of the blade on the surface are opaque and they need extra ﬁnal ﬁnishing process. Therefore, suitable hardness of materials guarantees the quality of the cutting, ﬁnishing and polishing steps of the ocular lenses. Taguchi method is one of the best option to determinethe optimal number of tests when the number of variables or factors is high (Roy 1968). Taguchi with combined orthogonal arrays was used to study the effect of ﬁve fundamental constituents of lenses, including hydroxyethyl methacrylate, methyl methacrylate, triethyleneglycol dimethacrylate, dimethyl itaconate, 3-(trimethoxysilyl) propyl methacrylate, on hardness. Materials In this research hydroxyethyl methacrylate (HEMA), 3-(trimethoxysilyl) propyl methacrylate (TMSPMA), methyl methacrylate (MMA) and tetraethylene glycol dimethacrylate (TGDMA) purchased from Merck, dime- thyl itaconate (DMI) and AIBN obtained from Sigma- Aldrich Chemicals, Germany, POSS-acrylate from Hybrid Plastics in USA were used. 3-[4,5-Dimethylthiazol-2-yl]- 2,5-diphenyltetrazolium bromide (MTT) and DMSO were obtained from Sigma-Aldrich Chemicals, Germany. The measurements were carried out under the following conditions: 32 scans, 4 cm-1 resolution, in range of 300–4000 cm-1 wavelength. The percentage of unreacted carbon–carbon double bonds (C=C) was determined from the ratio of the absorbance intensity of aliphatic C=C (peak at 1637 cm-1) against the absorbance intensity of the carbonyl group C=O (1720 cm-1) as a reference before and after polymerization. This experiment was carried out in triplicate. The degree of conversion was determined by subtracting the percentage of C=C from 100%, according to the formula: DC% ¼ ð1 ð1636=1720cm1Þ peak area after curing ð1636=1720cm1Þ peak area before curingÞ 100% Experimental design Taguchi method-based design experiments has been used to study the component concentration effects on hardness, ð1Þ Prog Biomater (2017) 6:67–74 69 Level Factor POSS(g) DMI(g) TMSPMA(g) MMA(g) HEMA(g) 1 0 0.15 0.2 0 24 0 2 0.05 0.20 0.3 0 28 0.05 3 0.10 0.25 0.4 0 32 0.10 4 0.15 0.30 0.5 0 36 0.15 medium is applied into the test system. In this research the extract test method was performed to evaluate the biocom- patibility of the fabricated lenses. The biocompatibility tests were conducted on the extracts obtained from the samples 1, 7 and 16 as shown in Table 2. The selected samples were immersed in PBS solution at 37 C for 28 days. The soup applied into the system test containing L929 ﬁbroblast cells and cell behavior was studied after 24 h in accordance with ISO 10993. MTT assay was used to indicate the level of viability. The survival of ﬁbroblast cells into the extracted was followed by 3-[4,5-dimethylthiazol-2-yl]-2,5- diphenyltetrazolium bromide. A 100 lL MTT solution with 900 lL medium was incubated with the cells in polystyrene culture plate wells (as control), and the wells with synthe- sized lenses at 37 C for 3 h. Next, dimethyl sulfoxide solution was added into the dissolved formazan crystals. The absorbance values of formazan solutions were measured using an ELIZA reader at 570 nm (Bio-Tek ELx800). . Table 2 L16 orthogonal array, by ﬁve control factors, with four level settings Factor level HEMA POSS DMI TMSPMA MMA 1 1 1 1 1 1 2 2 2 2 2 1 3 3 3 3 3 1 4 4 4 4 4 1 5 4 3 2 1 2 6 3 4 1 2 2 7 2 1 4 3 2 8 1 2 3 4 2 9 2 4 3 1 3 10 1 3 4 2 3 11 4 2 1 3 3 12 3 1 2 4 3 13 3 2 4 1 4 14 4 1 3 2 4 15 1 4 2 3 4 16 2 3 1 4 4 Table 2 L16 orthogonal array, by ﬁve control factors, with four level settings Hardness measurements Free radical polymerization initiated by heat was achieved by heating the monomers at 65 C for 24 h. As it is shown in Fig. 1 the radical attacks the double bonds of monomer, Hardness is the ability of materials to resist scratching and indentation which is very important in ocular lens materi- als. To study the surface properties, especially hardness, Shore Durometer hardness was used (ISO 868 2003). In this technique, three specimens of each group were made as discs with a diameter of 13 mm and a height of 4 mm and the hardness was calculated on four different points on the top and the bottom surfaces of each sample for 15 s. To consider the effect of environment, hardness as a model before and after exposure in PBS solution was measured. Fig. 1 Schematic presentation of copolymerization of multi-compo- nent (meth)acrylate monomers Biocompatibility test Therefore, multi-functional POSS-acrylate can be polymerized in combination with HEMA, DMI, MMA, TMSPMA and TGDMA to improve the mechanical prop- erties as well as oxygen transmission without sacriﬁcing the other essential lens properties. and the electron is transferred to another part. A newly formed radical attacks another monomer again and the process will be progressed and polymerization occurs. Since the materials containing silicone and oxygen have been of interest to increase the oxygen permeability (Tighe 2013, Lee et al. 2015), we introduced polyhedraloligomeric silsesquioxane (POSS)-acrylate into the polymeric back- bone. Therefore, multi-functional POSS-acrylate can be polymerized in combination with HEMA, DMI, MMA, TMSPMA and TGDMA to improve the mechanical prop- erties as well as oxygen transmission without sacriﬁcing the other essential lens properties. As time progresses, the degree of conversion is increased and after 48 h it reaches 92% for 4 samples and 4 different points on each sample. Six- teen samples are available in L16 orthogonal array (Table 2), all have been tested. The average hardness was measured before and after immersing in phosphate-buf- fered saline (PBS). The samples were immersed in PBS for 28 days and the hardness was measured every 7 days. The average of these values is shown in Fig. 3. To the best of our knowledge, no studies have been published on multi-component acrylate tethered polyhe- draloligomeric silsesquioxane used in ocular lenses. Effect of monomer content on hardness was studied. The analysis of variance for the values of hardness test is pre- sented in Table 4. Formulation 16 of Table 2 was used to determine the suitable reaction time. The samples prepared at different temperatures were evaluated by FTIR spectroscopy to consider the reaction progress. Infrared absorption spectra obtained at different reaction times are shown in Fig. 2. Biocompatibility test The biocompatibility of medical devices is evaluated by either direct contact test or extract test methods. In direct contact test, the samples are applied directly into the test systems such as cells or skin of animals. In extract test, the samples are immersed into the some kind of extraction media and allowed media to extract any residual chemicals which might be toxic. Then, the extraction Fig. 1 Schematic presentation of copolymerization of multi-compo- nent (meth)acrylate monomers 12 Prog Biomater (2017) 6:67–74 70 and the electron is transferred to another part. A newly formed radical attacks another monomer again and the process will be progressed and polymerization occurs. Since the materials containing silicone and oxygen have been of interest to increase the oxygen permeability (Tighe 2013, Lee et al. 2015), we introduced polyhedraloligomeric silsesquioxane (POSS)-acrylate into the polymeric back- bone. Therefore, multi-functional POSS-acrylate can be polymerized in combination with HEMA, DMI, MMA, TMSPMA and TGDMA to improve the mechanical prop- erties as well as oxygen transmission without sacriﬁcing the other essential lens properties. for 4 samples and 4 different points on each sample. Six- teen samples are available in L16 orthogonal array (Table 2), all have been tested. The average hardness was measured before and after immersing in phosphate-buf- fered saline (PBS). The samples were immersed in PBS for 28 d d th h d d 7 d Th Fig. 2 Infrared absorption spectra of samples at different reaction times; (1) 12 h (2) 24 h (3) 36 h (4) 48 h (5) 60 h and (6) 72 h Table 3 Degree of conversion in different times Time (h) 72 60 48 36 24 12 Dc (%) 100 97 92 84 71 32 As time progresses, the degree of conversion is increased and after 48 h it reaches 92% g Fig. 2 Infrared absorption spectra of samples at different reaction times; (1) 12 h (2) 24 h (3) 36 h (4) 48 h (5) 60 h and (6) 72 h and the electron is transferred to another part. A newly formed radical attacks another monomer again and the process will be progressed and polymerization occurs. Since the materials containing silicone and oxygen have been of interest to increase the oxygen permeability (Tighe 2013, Lee et al. 2015), we introduced polyhedraloligomeric silsesquioxane (POSS)-acrylate into the polymeric back- bone. Effect of trimethoxysilyl propyl methacrylate on hardness The calculated values for degree of conversion at dif- ferent times are shown in Table 3. We expected that by entering the Si–O group into the polymer backbone, hardness would be reduced. In this study, the amount of silicone-based monomer (TMSPMA) varies from 0.2 to 0.5 g. Figure 3 shows that by increasing the amount of TMSPMA from 0.2 g in level 1 to 0.5 g in level 2, Shore D hardness has decreased by 12 units. In Fig. 4, the hardness values obtained for different amount of TMSPMA are provided after and before water absorption. 123 Hardness study Hardness is the ability of materials to resist scratching and indentation which is very important in ocular lens materi- als. In this study, Shore Durometer hardness was measured 123 71 Prog Biomater (2017) 6:67–74 Fig. 3 The average hardness is measured before and after immersing in PBS Fig. 3 The average hardness is measured before and after immersing in PBS Fig. 3 The average hardness is measured before and after immersing in PBS Table 4 Analysis of variance for the values of hardness test Factor (%) Participation Pure sum of squares Variance ratio Variance Sum of squares Degree of freedom (P) (S’) (F) (V) (S) (F) MMA 27 58 2176/53 223 728 2186 3 TMSPMA 42 84 3379 346 1129 3389 3 DMI 9 69 746 79 258 774 3 POSS 13 29 1049 108 352 1058 3 Error 6 57 – – 3/29 749 147 Table 4 Analysis of variance for the values of hardness test The Si–O bond has 1.63 A˚ length and 130 degrees angle bond while C–C bond possesses 1.54 A˚ length and of 112 degrees bond angle (Kratky and Laggner 2002). Longer covalent bond with greater angle reduces the energy required to change the conﬁguration of the molecules and Si–O molecules are more readily moved compared to C–C molecules. In addition to the easier movement of Si–O groups and because the lengths of branches attached to the alpha carbon and the molecular free volumes increase, there would develop easy and ready chain mobility (Mirau et al. 2001). Low energy conﬁguration in macro-scale is comparable with reduction in hardness (Fig. 4). around the C–C bond. In turn, this increases the required energy for movement and makes the polymer hard and brittle (Mirau et al. 2001). Figure 5 shows that the hard- ness increases from 73 in Shore D scale to 80; a 7 unite in Shore D. Effect of dimethyl itaconate on hardness In the sample formulation, 0.15–0.3 g DMI per specimen was used. Increasing DMI increased the hardness of lens materials from 74 Shore D to 80.5 Shore D, as it is pre- sented in Fig. 6. There are two large groups on the alpha carbon of DMI which increases the stiffness of the ocular lens materials (Saiz et al. 1988). Large groups because of steric hindrance with adjacent groups require more energy for torsion bond angle. So these groups restricted the conformational changes and also molecular mobility. Effect of methymethacrylate on hardness Methyl groups attached to the alpha carbon of acrylic monomers leads to steric hindrance and prevents rotation 12 3 3 72 Prog Biomater (2017) 6:67–74 Effect of polyhedraloligo silsesquioxane on hardness Introducing POSS-acrylate increased Shore D hardness value from 73.4 to 79.6 Shore D. Existence of ten acrylate arms on th b id f POSS th l l t t Biocompatibility study As it is evident in Fig. 8, the cells grow and proliferate in the extract and in comparison with the control, pro- liferation and growth of the cells show that there is no toxic substance in the extract media to interfere with the Fig. 4 The effect of 3-methoxypropyl methacrylate alkoxysilyl on hardness hardness before and after water absorption Fig. 5 The effect of methymetacrylate on hardness before and after water absorption Fig. 6 The effect of dimethyl itaconate hardness before and after water absorption Fig. 7 The effect of polyhedraloligo silsesquioxane on hardness before and after water absorption 72 Prog Biomater (2017) 6:67 74 Fig. 4 The effect of 3-methoxypropyl methacrylate alkoxysilyl on hardness hardness before and after water absorption Fig. 6 The effect of dimethyl itaconate hardness before and after water absorption Fig. 6 The effect of dimethyl itaconate hardness before and after water absorption Fig. 6 The effect of dimethyl itaconate hardness before and after water absorption Fig. 4 The effect of 3-methoxypropyl methacrylate alkoxysilyl on hardness hardness before and after water absorption Fig. 7 The effect of polyhedraloligo silsesquioxane on hardness before and after water absorption Fig. 5 The effect of methymetacrylate on hardness before and after water absorption Fig. 7 The effect of polyhedraloligo silsesquioxane on hardness before and after water absorption Fig. 5 The effect of methymetacrylate on hardness before and after water absorption Effect of polyhedraloligo silsesquioxane on hardness As it is evident in Fig. 8, the cells grow and proliferate in the extract and in comparison with the control, pro- liferation and growth of the cells show that there is no toxic substance in the extract media to interfere with the cell growth. The L929 cells into the extracts exhibited their typical spindle morphology and ﬁlopodia around the cells indicating the beginning of cellular expansion (Oian et al.). Introducing POSS-acrylate increased Shore D hardness value from 73.4 to 79.6 Shore D. Existence of ten acrylate arms on the cage bridge of POSS causes the molecule to act as a crosslinking agent in the polymerization reaction. By increasing the amount of POSS, crosslinking density would increases which leads to lower segment mobility (Sperling 1992; Kopesky et al. 2004). As it is shown in Fig. 7, the reduced mobility segment has increased Shore D hardness value by 6 units. Evaluated MTT results demonstrated that the level of viability conﬁrms the results of microscopy. 123 Conclusion In this research, the experimental design Taguchi method was performed to evaluate the effect of acrylate-based components on hard ocular lenses and a multi-component silicone acrylate nanocomposite containing POSS. L16 orthogonal array was used in experimental design. The effect of increasing concentrations of the components on the scale Shore D hardness of samples before and after exposing in PBS was measured. which is the most pronounced effect among all monomers followed by MMA. POSS, methyl methacrylate and di- methyl itaconate increases the hardness 13, 27 and 9%, respectively. According to the results from Taguchi method, tri- methoxy silyl propyl methacrylate percentage and acrylate have great inﬂuence on controlling the hardness, while 2 hydroxymethyl methacrylate has the less impact about 3%. About 6% of total hardness observed in this study was due to errors in the test. PBS had no signiﬁcant change on Fig. 8 Light microscope images of L929 ﬁbroblast cell culture in extracts from ocularlenses. (a) Control, (b) an extract prepared from a sample number of 1 in Table 2 (c) an extract prepared from a sample number of 7 in Table 2 and (d) an extract prepared from a sample number of 16 in Table 2 Prog Biomater (2017) 6:67–74 73 73 Prog Biomater (2017) 6:67–74 Fig. 8 Light microscope images of L929 ﬁbroblast cell culture in extracts from ocularlenses. Effect of polyhedraloligo silsesquioxane on hardness (a) Control, (b) an extract prepared from a sample number of 1 in Table 2 (c) an extract prepared from a sample number of 7 in Table 2 and (d) an extract prepared from a sample number of 16 in Table 2 Conclusion which is the most pronounced effect among all monomers followed by MMA. POSS, methyl methacrylate and di- methyl itaconate increases the hardness 13, 27 and 9%, respectively. which is the most pronounced effect among all monomers followed by MMA. POSS, methyl methacrylate and di- methyl itaconate increases the hardness 13, 27 and 9%, respectively. In this research, the experimental design Taguchi method was performed to evaluate the effect of acrylate-based components on hard ocular lenses and a multi-component silicone acrylate nanocomposite containing POSS. L16 orthogonal array was used in experimental design. The effect of increasing concentrations of the components on the scale Shore D hardness of samples before and after exposing in PBS was measured. According to the results from Taguchi method, tri- methoxy silyl propyl methacrylate percentage and acrylate have great inﬂuence on controlling the hardness, while 2 hydroxymethyl methacrylate has the less impact about 3%. About 6% of total hardness observed in this study was due to errors in the test. PBS had no signiﬁcant change on hardness and the difference values are negligible. Cyto- toxicity test showed that the growth and proliferation of The tri-methoxy silyl propyl methacrylate contributes the hardness by 42% (having an impact on hardness), 12 3 74 Prog Biomater (2017) 6:67–74 Lee SE, Kim SR, Park M, Tighe BJ (2015) Oxygen permeability of soft contact lenses in different pH, osmolality and buffering solution. Int J Ophthalmol 8:1037–1042 cells into the extracted media are comparable with the cells behavior in culture medium and the cells did not sense any toxic substances. Mannis MJ, Zadnik K, Coral-Ghanem C, Kara-Jose N (2003) Contact lenses in ophthalmic practice. Springer, New York Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://crea tivecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Mirau PA, Jelinski LW, Bovey FA (2001) Macromolecules, structure. In: Meyers R (ed) Encyclopedia of physical science and technology, 3rd edn. Academic Press, London, pp 857-901 Morgan PB, Woods CA, Tranoudis IG, Efron N, Knajian R, Grupcheva CN, Jones D, Tan KO, Pesinova A, Ravn O (2009) International contact lens prescribing in 2008. Contact Lens Spectr 24:28–32 Nichols J (2009) Contact lenses. Conclusion Contact Lens Spectr 24:24–32 Roy RK (1968) A primer on the taguchi method. Society of Manufacturing Engineers, Dearborn, Michigan References Saiz E, Horta A, Gargallo L, Hernandez-Fuentes I, Radic D (1988) Dipole moment and conformational analysis of itaconate poly- mers. Macromolecules 21:1736–1740 Bruggink JW (2013) CBS Web magazine, 20 September Bruggink JW (2013) CBS Web magazine, 20 September Gasson A Moriss J (2003) The Contact Lens Manual; A Practical Bruggink JW (2013) CBS Web magazine, 20 September Gasson A, Moriss J (2003) The Contact Lens Manual; A Practical Guide to Fitting. Butterworth-Heinemann, London Gasson A, Moriss J (2003) The Contact Lens Manual; A Practical Guide to Fitting. Butterworth-Heinemann, London Sangermano M, Meier P, Tzavalas S (2012) Infrared spectroscopy as a tool to monitor radiation curing, infrared spectroscopy. Mater Sci Eng Technol 325–336 ISO 868 (2003) Plastics and ebonite-determination of indentation hardness by means of a durometer, Shore hardness Sperling LH (1992) Introduction to physical polymer science. Wiley, New York Kopesky ET, Haddad TS, Cohen RE, McKinley GH (2004) Thermomechanical properties of poly (methyl methacrylate) containing tethered and untethered polyhydraloligomeric- silsesquioxanes. Macromolecules 37:8992–9004 Tighe BJ (2013) A decade of silicone hydrogel development: surface properties, mechanical properties, and ocular compatibility. Eye Contact Lens 39:4–12 Kratky O, Laggner P (2002) Encyclopedia of physical science and technology. Academic, New York 123
https://openalex.org/W2031547642	https://journal.unila.ac.id/index.php/tropicalsoil/article/download/108/pdf	English	null	Soil Chemical and Biological Characteristics for Diagnostic the Potency of Acid Dry Land for Soybean Extensification	Jurnal Tanah Tropika/Jurnal tanah Tropika dan Kualitas Lingkungan	2,013	cc-by	7,069	Jauhari Syamsiyah, Sumarno, Suryono, Nur Echsan Muhamat Rajaband Ida Aryaningrum rtmentof Soil Science, Faculty of Agriculture, Universitas Sebelas Maret, Jl. Ir. Sutami No.36A, Jebres, Kota Surakarta, Jawa Tengah 57126, Indonesia; e-mail: ninukts@staff. uns.ac.id 1Departmentof Soil Science, Faculty of Agriculture, Universitas Sebelas Maret, Jl. Ir. Sutami No.36A, Jebres, Kota Surakarta, Jawa Tengah 57126, Indonesia; e-mail: ninukts@staff. uns.ac.id Received 21 Juny 2017/ accepted 01 September 2017 ABSTRACT Soil fertility is a crucial factor determining the growth and yield of plants. The increase of nutrient content and availability in soil can be achieved by fertilization. A field experiment was conducted using a Randomized Com- pletely Block Design (RCBD) with two factors and three replications in order to study the effects of Mixed Source of Fertilizer (MSF) application on the nutrient contents in Vertisol and its relationship to the growth and yield of mustard. The first factor was the three MSF formulas (F1, F2, F3) and second factor was the doses of MSF (0; 2.5; 5.0; 7.5; 10 Mg ha-1) applied to the soil. At the end of the experiment, the soil pH, CEC, organic-C, total-N, available- P and exchangeable-K contents were measured. The results show that there are no significant differences on the soil chemical characteristics, such as pH, organic-C content, available-P, exchangeable-K, -Ca and -Mg measured after application of different MSF formulas to the soil. Meanwhile, the increase of MSF doses applied to the soil significantly increases organic-C content, total-N, available-P and exchangeable-K in the soil. The significant increase of available-P (by 29.13%) and total-N (by 24.1%) occured after application of MSF at 5.0 Mg ha-1 and the increase of exchangeable-K (by 50%) is achieved after application of 7.5 Mg ha-1, in comparison to that without MSF application. The height and fresh weight of mustard increase in accordance with the increase of MSF doses applied. The application of 10.0 Mg ha-1 MSF results in the highest height and fresh weight of the mustard up to 63.9% and 620%, respectively. The height and fresh weight of mustard are positively correlated to the total-N, available-P and exchangeable-K in the soil. The MSF is an alternative fertilizer that can be used to improve Vertisol fertility and plant growth. Keywords: Coconut husk ash, feldspar, mixed source fertilizer, phosphate rock,Vertisol J Trop Available online at: http://journal.unila.ac.id/index.php/tropicalsoil DOI: 10.5400/jts.2017.v22i3.139-148 J Trop Available online at: http://journal.unila.ac.id/index.php/tropicalsoil DOI: 10.5400/jts.2017.v22i3.139-148 Kata kunci: Abu sabut kelapa, feldspar, campuran berbagai sumber pupuk, batuan fosfat, Vertisol Keywords: Coconut husk ash, feldspar, mixed source fertilizer, phosphate rock,Vertisol The Effects of Mixed Source Fertilizer Application on Vertisol Fertility and Growth of Mustard auhari Syamsiyah, Sumarno, Suryono, Nur Echsan Muhamat Rajaband Ida Aryaningrum INTRODUCTION Central Java (Subagyo et al. 2007). Vertisol is potential for agricultural land use because it has relatively good fertility status with high Cation Exchange Capacity (CEC), base saturation, water holding capacity, and pH (i.e. pH 6-8.5) (Deckers et al. 2001). Meanwhile, the availability of phosphorus in dry Vertisol is generally low because the phosporus is bound by Ca, and the amount of available-K is low as well (Munir 1996). A previous study also showed that Vertisol that is dominated by clay fraction has neutral pH, low organic-C and total- N contents, medium amount of P extracted using Olsen and HCl 25%, low amount of K extracted using HCl 25%, high CEC and high amount of exchangeable base cations (Harsanti et.al.2003). The availability of nutrients in soil is a factor that can affect the growth and yield of plants. The appropriate amount and the use efficiency of fertilizers applied will have a pronounced effect on crop production since fertilization will affect the nutrient content in soil (Elamin and Elagib 2001; Kuntyastuty et al. 2011). The use of organic and inorganic fertilizers has positive and negative effects on soil and plant growth. The disadvantages of using organic fertilizers among others are slow release fertilizer (release the nutrients gradually) and low nutrient content (Sentana 2010). On the other hand, the use of organic fertilizers also has several advantages namely providing balanced nutrients to plants, including macro and micro nutrients (Tisdale et al. 2010), preventing harmful effects caused by nutritional excess of certain plants, increasing the availability of nutrients in soil that is triggered by the increase of microbial activity, improving the nature of soil structure and root development as well as increasing the ability of soil to hold water (Nieder and Benbi 2005; Brady and Weil 2008). Some previous studies showed various effects of using organic fertilizers on nutrient content or availability in soil. The study of Kuntyastuti (2011) reported that organic fertilizer application showed no significant effect on organic-C content and macronutrient availability in soil after the first harvesting, and only increased the amounts of available-P and exchangeable-K after the second growing season. Besides, the study of Han et al. (2006) showed that the use of manure increased the concentration of nitrogen, available-P, exchangeable-K, -Ca and -Mg in soil during the plant nursery period. INTRODUCTION In order to increase the fertility status of Vertisol and to overcome the lack of organic and inorganic fertilizer supplies, use of natural resources that are potential for fertilizers can be an option. The potential natural resources include Azolla and quail manure as sources of N, phosphate rocks as a source of P, coconut, coconut husk and feldspar rocks as sources of K, dolomite as a source of Ca and Mg, and sulfuric element as a source of S. By mixing these natural resources, it is expected to result in a fertilizer formula that has positive effects in increasing soil productivity and fertility. This study aimed to evaluate the effects of various Mixed Source Fertilizer (MSF) formulas and their doses on soil fertility status of Vertisol and its relationship to the growth of mustard. Study Site A field experiment on Vertisol was conducted in Karanganyar, Central Java, Indonesia in April until December 2016. On the other hand, inorganic fertilizers hold an important role in food production. In Indonesia inorganic fertilizers has been used intensively since the development of intensive farming system in which farmers depend a lot on inorganic fertilizer for their farms. Inorganic fertilizers contain high concentrations of nutrients, which can be absorbed immediately by plants, as well as they are practical and easy to apply (Tisdale et al. 2010). However, inappropriate and continuous use of inorganic fertilizers will cause various problems for soil and environment, such as nutrient loss, soil acidification, reduction of beneficial microbial population (Chen 2006), land productivity degradation (Khairatun et al. 2013) and environmental pollution (Suharsi et al. 2002; Kassir et al. 2012). ABSTRAK Kesuburan tanah merupakan faktor yang sangat menentukan pertumbuhan dan hasil tanaman. Peningkatan hara dalam tanah dapat dicapai dengan pemupukan. Penelitian lapangan dilakukan dengan menggunakan Rancangan Kesuburan tanah merupakan faktor yang sangat menentukan pertumbuhan dan hasil tanaman. Peningkatan hara dalam tanah dapat dicapai dengan pemupukan. Penelitian lapangan dilakukan dengan menggunakan Rancangan Kelompok Lengkap (RKL) dengan dua faktor perlakuan dan tiga ulangan, untuk mengkaji penggunaan Mixed Source Fertilizer (MSF) pada kandungan hara dalam tanah Vertisol dan hubungannya dengan pertumbuhan dan hasil sawi. Perlakuan yang dicobakan adalah formula MSF (F1, F2 dan F3) dan lima dosis MSF (0,25; 5,0; 7,5 dan 10 Mg ha-1). pH tanah, C-organik, CEC, N-total, P-tersedia, Kation tertukar diukur pada akhir penelitian. Tidak ada perbedaan yang nyata yang diperoleh dari berbagai formula MSF pada pH tanah, C-organik, P-tersedia, K-, Ca- and Mg-dapat ditukar. Peningkatan dosis MSF nyata meningkatkan C-organik tanah, N-total, P-tersedia dan K-dapat ditukar. Pemberian MSF 5,0 Mg ha-1dalam waktu 40 hari nyata meningkatkan P-tersedia (29,13%) dibandingkan kontrol. Sedangkan peningkatan N-total (24,1%), dan K-dapat ditukar (50%) dari kontrol tercapai dengan pemberian MSF 7,5 Mg ha-1. Tinggi tanaman dan berat segar sawi meningkat sejalan dengan meningkatnya dosis MSF yang diberikan. Pemberian MSF 10,0 Mg ha-1 memberikan tinggi tanaman dan berat segar sawi tertinggi, berturut turut mencapai 63,9% and 620% . Tinggi tanaman dan berat segar sawi berkorelasi posistif dengan N-total, P-tersedia dan K-dapat ditukar di dalam tanah. Campuran berbagai sumber pupuk (Mixed Source Fertilizer) adalah pupuk alternatif yang dapat digunakan untuk memperbaiki status kesuburan tanah dan pertumbuhan tanaman. J Trop Soils, Vol. 22, No. 3, 2017: 139-148 ISSN 0852-257X 140 Experimental Design The study was arranged in a Completely Randomized Block Design (RCBD) with two factors and three replications. The first factor was 3 MSF formulas, namely F1 (50% quail manure, 20% phosphate rock, 18% coconut husk ash, 6% dolomite, 1% sulfuric element, 5% feldspar), F2 (60% quail manure, 20% phosphate rock, 14% coconut husk ash, 6% dolomite, 1% sulfuric element) and F3 (20% quail manure, 30% Azolla, 16% phosphate rock, 23% coconut husk ash, 10% dolomite, 1% sulfuric element). Each MSF formula was made by mixing all the ingredients including composted Azolla, matured quail manure, phosphate rock, sulfur and feldspar (from Banjarnegara) and filtered coconut Vertisol covers approximately 2.1 million hectares of land in Indonesia (Subagyo et al. 2004). It spreads in many areas in Indonesia including 141 J Trop Soils, Vol. 22, No. 3, 2017: 139-148 before and after the experiment. The chemical properties of the soil samples including total-N (Kjedhal), available-P (Bray-1), exchangeable base cations (NH4OAc 1 N pH 7), organic-C (Walkley and Black) and CEC (NH4OAc 1 N pH 7) were analyzed at the Laboratory of Soil Chemistry and Soil Fertility, Agriculture Faculty, Sebelas Maret University. husk. All the ingredients were mixed homogeneously in accordance with their doses to obtain MSF formulas that have high nutrient content. The second factor was 5 doses of MSF applied, namely D0 (no fertilizer), D1 (2.5 Mg ha-1), D2 (5 Mg ha-1), D3 (7.5 Mg ha-1), and D4 (10 Mg ha-1). Data Analysis The MSF was applied on each plot according to the treatment three days before planting. The size of each plot was 1 m × 1.5 m. The mustard seeds of Choy Green variety (in accordance with the variety of mustard planted by farmers around the study site) were planted, i.e. 2 seeds for each planting hole, with the planting distance of 20 cm × 20 cm. The nurturing of plants was carried out in accordance with the local farmer practices. The data were analyzed using F-variance test at 5% and 1% significance levels, continued with Duncan’s Multiple Range Test (DMRT) test, and correlation test (Gomez and Gomez 2007). Characteristics of Vertisol and Mixed Source Fertilizers (MSFs) Dose (Mg ha-1) MSF Formula F1 F2 F3 pH CEC cmol(+) kg-1 pH CEC cmol(+) kg-1 pH CEC cmol(+)kg-1 0 6.9±0.1 50.68±11.24 7.1±0.1 42.27±7.38 7.1±0.2 51.84±1.02 2.5 7.0±0.2 49.89±15.01 6.9±0.2 45.17±4.45 7.0±0.1 48.73±3.16 5.0 6.9±0.1 48.25±4.32 7.0±0.2 44.72±7.01 6.9±0.2 44.38±5.30 7.5 7.0±0.2 52.67±7.92 7.1±0.1 44.23±6.01 7.0±0.0 50.22±6.85 10.0 6.9±0.1 43.58±9.68 6.9±0.1 48.73±2.55 7.2±0.1 44.30±5.01 F1 F2 F3 J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility Table 3. pH and CEC of the soil applied with Mixed Source Fertilizers. Table 3. pH and CEC of the soil applied with Mixed Source Fertilizers. Table 3. pH and CEC of the soil applied with Mixed Source Fertilizers. that clay particles have a negative charge, therefore, the soil with high clay content will have high CEC (Hanafiah 2005; Nursyamsi and Suprihati 2005). The Vertisol used in the current study contains low organic matter, i.e. 1.19%, which is in line with the study of Munir (1996) that suggested that Vertisols have low organic-C content. This condition causes the availability of nutrients in the Vertisols is low. According to Prasetyo and Suriadikarta (2006), soil organic matter is an important component of soil fertility and nutrient source for plants (Tisdale et al. 2010). The low content of organic matter in the Vertisol used in the current study results in a moderate total-N content and low amount of available-P. The amount of exchangeable-K is also low because most of K are fixed by 2:1 clay minerals that are commonly found in Vertisols (Borchardt 1989). although the K content in the MSF F1 was higher than that in other MSF formulas (Table 2). It is supposed that by adding 5% feldspar in the MSF F1 will result in better advantage than other formulations especially on K content. Feldspar is the source of K in soil (Tisdale et al. 2010). The high total-N content in F1 formula is correlated with the high organic C content in this formula and low C/N ratio. The low C/N ratio indicates that the decomposition process of the MSF F1 has been going well (Brady and Weil 2008). Furthermore, organic materials will be decomposed into amino acids and then heterotrophic microorganisms decompose the amino acids into ammonium as the main form of inorganic- N in soil (Suntoro 2003; Tisdale et al. 2008). Characteristics of Vertisol and Mixed Source Fertilizers (MSFs) Five plant samples were selected for each plot and the height of plants was measured started from soil surface till the end of the longest leaf at the maximum vegetative growth. In addition, the fresh weight of plants was measured after harvesting. Soil samples were taken from the experimental plots Vertisol used in the experiment has a neutral pH, clay texture, and high CEC (Table 1). The high CEC of the soil is due to the high clay content in the soil. The study of Brady and Weil (2008) indicated Table 1. Characteristics of Vertisol used in the current study. Note: ) According to the criteria proposed by the Soil Research Institute (2005) Soil Properties Value Range of Criteria Criteria pH H2O 6.7 6.6-7.5 Neutral Organic-C (%) 0.67 <1 Very low* CEC (cmol(+) kg-1) 43.45 >40 Very high* Total-N (%) 0.37 0.21-0.5 Moderate* Available-P (ppm) 3.1 <5 Very low* Exchangeable-K (cmol(+) kg-1) 0.14 0.1-0.3 Low* Texture Sand (%) 8.76 Clay Silt (%) 24.77 Clay (%) 66.47 Table 1. Characteristics of Vertisol used in the current study. Table 2. Characteristics of Mixed Source Fertilizer (MSF) Formulas. Characteristics MSF Formula Standard values regulated by the Ministry of Agriculture Number 140/10/2011 F1 F2 F3 pH 6.41 ± 0.17 6.29 ± 0.26 6.74 ± 0.20 4 – 8 Organic-C (%) 21.0 ± 0.18 19.6 ± 0.60 19.8 ± 0.39 > 12 N (%) 1.72 ± 0.17 1.50 ± 0.08 1.52 ± 0.07 < 6 P2O5 (%) 1.21 ± 0.09 1.17 ± 0.06 1.10 ± 0.08 < 6 K2O (%) 1.85 ± 0.07 1.42 ± 0.16 1.39 ± 0.10 < 6 C/N Ratio 12.2 ± 0.56 13.1 ± 0.23 13.0 ± 0.09 15 – 25 Table 2. Characteristics of Mixed Source Fertilizer (MSF) Formulas. 142 J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility able 3. pH and CEC of the soil applied with Mixed Source Fertilizers. 42 J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility Table 3. pH and CEC of the soil applied with Mixed Source Fertilizers. Soil pH Application of MSF with various formulas and doses did not significantly affect the pH of Vertisol (p > 0.05). The range of soil pH after application of MSF was 6.9 up to 7.2, which is considered as neutral. There was no significant difference on soil pH after application of the three MSF formulas although the soil pH applied with MSF F3 was slightly The organic-C contents in the three MSF formulas were quite high and above the minimum organic-C content in organic fertilizers according to the regulation proposed by the Indonesian Ministry of Agriculture, i.e. 12%. The N, P, and K contents of the three MSF formulas were almost the same, Figure 1. Organic-C content in the soil applied with various doses and formulas of MSF. D: dose, F: Formula. The same letters above the bart chart indicate no significant difference at 95%. 1.20 1.00 0.80 0.60 0.40 0.20 0.00 Organic-C (%) DOF 1 DOF 2 DOF 3 D1F 1 D1F 2 D1F 3 D2F 1 D2F 2 D2F 3 D3F 1 D3F 2 D3F 3 D4F 1 D4F 2 D4F 3 Treatment a a ab abc abc abc cde bcde bcde e bcd bcde abcd 1.01 de Figure 1. Organic-C content in the soil applied with various doses and formulas of MSF. D: dose, F: Formula. The same letters above the bart chart indicate no significant difference at 95%. 143 J Trop Soils, Vol. 22, No. 3, 2017: 139-148 indicated that for 40 days of incubation, the mineralization of C from sewage sludge is only <30%. Consequently, there has not been much functional groups of organic acids such as COOH and OH released into the soil. According to Brady and Weil (2008), the dissociation of the open hydroxyl ions on the functional groups of organic matter will produce a very negative charge that further determines the soil CEC. This is evident that there is a correlation between soil CEC and soil organic matter content. higher than that applied with MSF F1 and F2. This phenomenon is due to the three MSF formulas have almost similar properties (Table 2). However, the soil pHs after application of MSFs that were measured at the end of the experiment in general were higher than the soil pHs before application of MSFs. This was probably due to the effects of dolomite and coconut husk applied in the three MSF formulas. Soil pH Dolomite contains Ca and Mg that can release OH ions that further could increase soil pH (Nurhayati 2013). In addition, application of coconut husk ash can increase soil pH (Risnah et al. 2013). Total-N increases were 62.9% in the soil applied with MSF F1, 57.4% in the soil applied with MSF F2 and 37.0% in the soil applied with MSF F3, although there was no significant difference among the three organic- C contents. In the soil applied with MSF F1, the highest organic-C content was obtained at 7.5 Mg ha-1, whereas in the soils applied with MSF F2 and F3 were occurred at 10.0 Mg ha-1. The MSF F1 contained higher organic C than MSF F2 and F3, so that the application of 7.5 Mg ha-1 has increased soil organic C, which is equal to the application of 10.0 Mg ha-1 MSF F2 or F3. The quail manure and Azolla are the sources of organic matter so that their applications to the soil can increase the soil organic C. According to Nugroho and Firmansyah (2016), quail manure has high C content (16.53%). Likewise, the application of Azolla will increase soil fertility by increasing the soil organic matter content (Mandal 1999). The results of analysis of variance showed that the MSF formula significantly affected the total-N content in the soil. Figure 2 shows that application of MSF F3 results in the highest total-N content in the soil with the increase of 24.2% in comparison to that in the soil applied with MSF F2 and 51.9% in comparison to that in the soil applied with MSF F1. This phenomenon is due to MSF F3 contains Azolla that is the source of N in soil. The result of this study is in line with the study of Mandal (1999) that indicated that Azolla application increases the availability of N in soil. The increase of total-N content in the soil is in line with the increase of MSF doses applied. Figure 3 shows that the application of MSF at 2.5 and 5.0 Mg ha-1 resulted in an insignificant increase on the total-N in the soil. The application of MSF at 7.5 Figure 4. The amount of available-P in the soil applied with various doses of MSF. The letters above the bar chart indicate no significant different at 5% significance level. 7 6 5 4 3 2 1 0 Available-P (ppm) 0 2.5 5 7.5 10 MSF (Mg ha-1) a ab bc cc cc Figure 4. The amount of available-P in the soil applied with various doses of MSF. Cation Exchange Capacity (CEC) The soil organic-C content significantly increased after application of various formulas and doses of MSF (p<0.001). Application of 2.5 Mg ha-1 MSF increased soil organic C content by 27.8% in the soil applied with MSF F1, 29.8% in the soil applied with MSF F2 and 15.7% in the soil applied with MSF F3 in comparison to that in the soil without MSF application, although the three organic-C contents were not statistically different (Figure 1). The significant increase of soil organic-C content occurred after application of 5.0 Mg ha-1 MSF. The The results of analysis of variance showed that the doses, MSF formulas and their interaction did not significantly affect the soil CEC (p > 0.05). This is because the three MSF formulas have almost similar properties (Table 2) and contain the same organic matterials, such as quail manure, coconut husk ash and Azolla. Allegedly, these organic materials have not all been decomposed within 40 days of application of MSF to the soil. This result is in line with the study of Giacomini et al. (2015) that Figure 2. Total-N content in the soil applied with various formulas of MSF. F: Formula. The same letters above the bar chart indicate no significant difference at 5% significance level. 0.5 0.4 0.3 0.2 0.1 0.0 Total-N (%) F1 F2 F3 MSF Formula a b b 0.5 0.4 0.3 0.2 0.1 0.0 Total-N (%) F1 F2 F3 MSF Formula a b b F3 Figure 2. Total-N content in the soil applied with various formulas of MSF. F: Formula. The same letters above the bar chart indicate no significant difference at 5% significance level. 0.5 0.4 0.3 0.2 0.1 0.0 Total-N (%) 0 2.5 5 7.5 10 MSF (Mg ha-1) a a a b c Figure 3. Total-N content in the soil applied with various doses of MSF. The same letters above the bar chart indicate no significant difference at 5% significance level. 0.5 0.4 0.3 0.2 0.1 0.0 Total-N (%) 0 2.5 5 7.5 10 MSF (Mg ha-1) a a a b c Figure 3. Total-N content in the soil applied with various doses of MSF. The same letters above the bar chart indicate no significant difference at 5% significance level. 144 J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility Total-N The letters above the bar chart indicate no significant different at 5% significance level. Figure 5. The amount of exchangeable-K in the soil applied with various doses of MSF. The letters above the bar chart indicate no significant different at 5% significance level. 0.25 0.20 0.15 0.10 0.05 0.00 Exchangeable-K (cmol(+) kg-1) 0 2.5 5 7.5 10 a a ab b b MSF (Mg ha-1) Figure 5. The amount of exchangeable-K in the soil applied with various doses of MSF. The letters above the bar chart indicate no significant different at 5% significance level. 145 J Trop Soils, Vol. 22, No. 3, 2017: 139-148 Mg ha-1 was able to increase the total-N by 23.33% in comparison to that in the soil without MSF application, although the highest total N content in the soil was obtained after application of 10.0 Mg ha-1 MSF with the increase of 30% in comparison to that in the soil without MSF application. This result is due to the contribution of nitrogen derived from organic compounds in the MSF. The result showed that there was a significantly positive correlation between total-N content in the soil and soil organic matter content (p <0.005; r = 0.445), indicating that the total-N content in the soil increases with the increase of soil organic matter content. According to Wahyudi (2009), decomposed organic materials will produce a number of proteins and amino acids that will be further mineralized into ammonium (NH4 +) or nitrate (NO3 -), and organic materials are the largest contributor of N to the soil. suggested that soil organic matter is the source of P in soil for plant uptake. Exchangeable-K The amount of exchangeable-K in the soil after application of various MSF formulas was not different (p>0.05), however, it increased in accordance with the MSF doses applied (p<0.01) (Figure 5). The application of 5.0 Mg ha-1 MSF increased the amount of exchangeable-K in the soil, which was similar to that in the soil applied with 2.5 Mg ha-1 MSF and without MSF application. This phenomenon was due to the addition of coconut husk ash as the source of K to the MSF was not enough to provide an additional amount of exchangeable-K to the soil. The significant increase of exchangeable- K was achieved after application of 7.5 Mg ha-1 MSF, in which the amount of exchangeable-K was not different from that in the soil applied with 10.0 Mg ha-1MSF. This result is in line with the study of Risnah (2013), which shows that the application of 150% coconut ash effectively increases the amount of exchangeable-K in soil although it is not different from that in the soil applied with 200% coconut ash within 4 months. The increase of MSF dose that is not followed by the increase of exchangeable-K in the soil observed in this study indicates that there are other K sources that contribute to the increase of the exchangeable-K in the soil, such as quail manure and feldspar. Available-P The result of analysis of variance showed that the MSF doses were significantly (p <0.01) affected the amount of available-P in the soil. Meanwhile, the MSF formulas and the interaction between doses and formula of MSF showed no significant effect (p > 0.05) on the amount of available-P in the soil. The amount of available-P increased by 69% after application of MSF at 5 Mg ha-1 in comparison to that in the soil without MSF application. The amount of available-P in the soil applied with 10.0 Mg ha-1 MSF is not significantly different from that in the soil applied with 5.0 Mg ha-1 MSF. The results indicated that organic fertilizers combined with phosphate rock can improve the amount of available- P in the soil. Exchangeable-Ca and -Mg Plant Height (cm) Fresh Weight of Plant (g) Formula p> 0.05 p> 0.05 F1 20.29 ± 3.28 a 72.31 ± 40.70 c F2 20.27 ± 3.92 a 72.41 ± 42.70 c F3 20.20 ± 3.66 a 58.75 ± 33.44 d Dose MSF (Mg ha-1) p< 0.001 p< 0.001 0 14.68 ± 0.60 p 16.31 ± 1.89 p 2.5 19.16 ± 0.28 q 47.18 ± 5.69 q 5.0 20.86 ± 0.29 r 66.76 ± 8.81 r 7.5 22.52 ± 0.12 s 92.91 ± 11.93 s 10 24.06 ± 0.41 t 115.96± 18.20 t Note: The numbers followed by the same letters indicate no significant difference at 5% significance level. Note: The numbers followed by the same letters indicate no significant difference at 5% significance level. Note: The numbers followed by the same letters indicate no significant difference at 5% significance level. Note: The numbers followed by the same letters indicate no significant difference at 5% significance level. Note: The numbers followed by the same letters indicate no significant difference at 5% significance level. range of 13.06 up to 19.21 cmol (+) kg-1, while the amount of exchangeble-Mg is in the range of 2.1 up to 5.7 cmol (+) kg-1. the plant height (p>0.05). This phenomenon is due to the three MSF formulas have almost similar characteristics (Table 2). The fresh weight of mustard was significantly influenced by the doses (p<0.05) and formulas (p<0.05) of MSF applied to the soil. The increased doses of MSF caused an increase on the fresh weight of mustard. The applications of 2.5 Mg ha-1 and 10.0 Mg ha-1 MSF to the soil increased the fresh weight of mustard by 189% and 610.1%, respectively, in comparison to that without MSF application. This phenomenon is due to the increase on the amount of available-N, - P and -K in the soil that can be taken up by the plants. The results showed that there are positive correlations between fresh weight of mustard and total soil N (p<0.001; r = 0.714), available-P (p <0.001; r = 0.564) and exchangeable-K (p <0.001; r = 0.422). In addition, the increase of fresh weight of mustard is due to the increase of plant height (p <0.001; r = 0.926). The Growth of Mustard The result of analysis of variance showed that the increase of MSF doses applied to the soil significantly increased the height of mustard (p <0.01). The application of 2.5 Mg ha-1 MSF increased the plant height by 30%, while the highest plant height was obtained after application of 10.0 Mg ha-1 MSF, which was 63.9% higher than that without MSF application (Table 5). This phenomenon is due to the higher dose of fertilizer applied to the soil, the more nutrient will be absorbed by plants, consequently the plant growth will also increase (Rizqiani et al. 2007). The plant height is positively correlated to total soil N (p<0.001, r = 0.637), available-P (p<0.001, r = 0.547*), and exchangeable-K (p<0.05; r = 0.369 ). Nitrogen, phosphor, and potassium are essential macronutrients for plant growth and development (Tisdale et al. 2010). The more nitrogen absorbed by the plant, the growth of the plant will increase as well as the plant height (Erawan et al.2013). Phosphorus plays a role in the transfer of energy used for the whole plant metabolic activities (Marschner 2006). In addition, phosophorus can stimulates root growth and a good root system. Potassium (K) plays a role in enzyme activity, absorption and transport of water and nutrients from soil to plants (Marschner 2006; Tisdale et al.2010). Exchangeable-Ca and -Mg The application of various MSF formulas and doses did not affect the amount of exchangeable- Ca (p>0.05) and -Mg (p>0.05) in the soil within 40 days of application to the soil (Table 4). The result indicated that the addition of 5.5 %, 10% and 16% dolomite or equivalent to 550, 1000 and 1600 kg ha-1 dolomite on 10 Mg ha-1MSF dose did not provide significant addition to the amount of exchangeable-Ca and -Mg in the soil in comparison to other MSF doses applied. Similarly, the addition of other Ca and Mg sources such as 2 to 6 Mg ha-1 quail manure on 10 Mg ha-1MSF dose also did not provide significant increases on the amounts of exchangeable-Ca and -Mg at 40 days of application of MSF to the soil, in comparison to other MSF doses applied. The study of Tabitha et al. (2017) showed that a significant increase on the amount of exchangeable-Ca and -Mg in soil is occurred after application of quail manure at 8.45 Mg ha-1. The amount of exchangeable-Ca is in the At a dose of 5.0 Mg ha-1 MSF, the phosphate rock was completely dissolved within 40 days of application of MSF to the soil. However, at the doses of 7.5 and 10.0 Mg ha-1 MSF, some of the phosphate rock was not completely solubilized, so that it was unavailable for plants. According to Sulaeman et al. (2002), natural phosphate rocks are considered as the long-term fertilizer that releases relatively slow the available-P to the soil. Therefore, to increase the solubility of phosphate rocks, it is advisable to add biological fertilizers containing phosphate solubilizing microbes (Stamford et al 2006). The increase on the amount of available-P in the soil after MSF application is also caused by the increase of organic matter content in the soil since the soil organic matter content is positively correlated to the amount of available-P in the soil (p <0.001, r = 0.538**). The study of Brady and Weil (2008) 146 146 J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility Table 4. The amount of exchangeable-Ca and -Mg in the soil applied with various MSF for- mulas and doses. Table 5. The height and fresh weight of mustard applied with various formulas and doses of MSF. Exchangeable-Ca and -Mg Plant Height (cm) Fresh Weight of Plant (g) Formula p> 0.05 p> 0.05 F1 20.29 ± 3.28 a 72.31 ± 40.70 c F2 20.27 ± 3.92 a 72.41 ± 42.70 c F3 20.20 ± 3.66 a 58.75 ± 33.44 d Dose MSF (Mg ha-1) p< 0.001 p< 0.001 0 14.68 ± 0.60 p 16.31 ± 1.89 p 2.5 19.16 ± 0.28 q 47.18 ± 5.69 q 5.0 20.86 ± 0.29 r 66.76 ± 8.81 r 7.5 22.52 ± 0.12 s 92.91 ± 11.93 s 10 24.06 ± 0.41 t 115.96± 18.20 t Note: The numbers followed by the same letters indicate no significant difference at 5% significance level. Note: The numbers followed by the same letters indicate no significant difference at 5% significance level. Exchangeable-Ca (cmol(+) kg -1) Exchangeable-Mg (cmol(+) kg -1) Formula p> 0.05 F1 16.92 ± 0.85 a 3.77 ± 1.29 a F2 16.67 ± 1.21a 4.02 ± 1.18 a F3 16.93 ± 1.05 a 3.73 ± 0.90 a Dose MSF (Mg ha -1) p> 0.05 p> 0.05 0 16.02 ± 2.64 b 5.99 ± 0.71 b 2.5 16.71 ± 2.26 b 6.02 ± 3.23 b 5.0 16.98 ± 2.77 b 6.09 ± 1.89 b 7.5 17.14 ± 5.04 b 6.14 ± 2.20 b 10 17.63 ± 0.08 b 6.47 ± 1.54 b J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility 146 J Syamsiah et al.: Effects of Mixed Source Table 4. The amount of exchangeable-Ca and -Mg in the soil applied with various MSF for- mulas and doses. Note: The numbers followed by the same letters indicate no significant difference at 5% significance level. Exchangeable-Ca (cmol(+) kg -1) Exchangeable-Mg (cmol(+) kg -1) Formula p> 0.05 F1 16.92 ± 0.85 a 3.77 ± 1.29 a F2 16.67 ± 1.21a 4.02 ± 1.18 a F3 16.93 ± 1.05 a 3.73 ± 0.90 a Dose MSF (Mg ha -1) p> 0.05 p> 0.05 0 16.02 ± 2.64 b 5.99 ± 0.71 b 2.5 16.71 ± 2.26 b 6.02 ± 3.23 b 5.0 16.98 ± 2.77 b 6.09 ± 1.89 b 7.5 17.14 ± 5.04 b 6.14 ± 2.20 b 10 17.63 ± 0.08 b 6.47 ± 1.54 b Table 5. The height and fresh weight of mustard applied with various formulas and doses of MSF. CONCLUSIONS The application of MSF with various doses significantly increases total-N, available-P, and exchangeable-K in the soil but it does not increase the pH and CEC of the soil. There is a difference on the amount of total-N in the soil and the fresh weight of mustard after application of the three MSF formulas. The application of MSF with various formulas and doses significantly increases organic- C content in the soil. The increase of total-N and available-P is positively correlated to soil organic-C content. The application of 5.0 Mg ha-1 MSF The differences in the MSF formulas applied to the soil did not show significant differences on 147 J Trop Soils, Vol. 22, No. 3, 2017: 139-148 Entry JA, BH Wood, JH Edwards and CW Wood. 1997. Influence of organic byproducts and nitrogen source on chemical and microbiologial status of an agricultural soil. Biol Fertil Soils 24: 196-204. significantly increases the amount of available-P (by 29.13%) and total-N content (by 24.1%) in the soil in comparison to that in the soil without MSF application. Meanwhile, the significant increase of exchangeable-K (by 50%) is achieved after application of 7.5 Mg ha-1 MSF. The height and fresh weight of mustard increase in accordance with the increase of MSF doses applied. The application of 10.0 Mg ha-1 MSF results in the highest height and fresh weight of the mustard up to 63.9% and 620%, respectively. The height and fresh weight of mustard are positively correlated to the total-N, available-P and exchangeable-K in the soil. The MSF is an alternative fertilizer that can be used to improve Vertisol fertility and plant growth. Erawan D, WO Yani and A Bahrun. 2013. Pertumbuhan dan hasil tanaman sawi (Brassica juncea L.) pada berbagai dosis pupuk urea. Agroteknos 3: 19-25. GiacominiSJ, VLG Simon, Celso Aita, LD Bastos, DA Weiler and M Redin. 2015. Carbon and nitrogen mineralization in soil combining sewage sludge and straw. R Bras Ci Solo 39: 1428-1435. Gomez KA and AA Gomez. 2007. Prosedur Statistik untuk Penelitian Pertanian Edisi Kedua. Universitas Indonesia, UI Press (in Indonesian). Han SH, JY An, J Hwang, SB Kim and BB Park. 2012. The effects of organic manure and chemical fertilizer on the growth and nutrient concentrations of yellow poplar (Liriodendron tulipifera Lin.) in a nursery system. Forest Sci Technol 12: 137-143. ACKNOWLEDGEMENT Hanafiah KA. 2005. Dasar-Dasar Ilmu Tanah. PT Raja Grafindo Persada. Jakarta (in Indonesian). I would like to thank Ministry of Research, Technology, and Higher Education of Republic of Indonesia for the research funding, and The Dean of Faculty of Agriculture and The Rector of Sebelas Maret University for supporting me in finishing the project. Jacobsen ST. 1995. Aerobic decomposition of organic waste 2. Value of compost as fertilizer. Resour Conserv Recy 13: 15-31. Kassir LN, Lartiges B and N Ouaini. 2012. Effects of fertilizer industry emissions on local soil contamination: a case study of a phosphate plant on the east Mediterranean coast. Environ Technol 33 : 873-85. J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility Sentana S. 2010. Pupuk organik, peluang dan kendalanya. Seminar Nasional Teknik Kimia Kejuangan. Pengembangan Teknologi Kimia untuk Pengolahan Sumber Daya Alam Indonesia Yogyakarta (in Indonesian). Nieder R and DK Benbi. 2008. Carbon and Nitrogen in the Terrestrial Environment. USA: Springer Science and Bussines Media B.V. Nugroho WA and MA Firmansyah. 2016. Pengaruh jenis dan dosis pupuk kandang terhadap pertumbuhan dan produksi bawang merah di lahan kering dataran rendah. Seminar Nasional Inovasi Teknologi Pertanian Banjarbaru, pp. 906-913 (in Indonesian). Stamford NP, PR Santos, CES Santos, ADS Freitas, SHL Dias and MA Lira Jr. 2006. Agronomic effectiveness of biofertilizers with phosphate rock, sulphur and Acidithiobacillus for yam bean grown on a Brazilian tablel and acidic soil. J Bioresour Technol 98: 1311- 1318. Nurhayati. 2013. Pengaruh jenis amelioran terhadap efektivitas dan infektivitas mikrob pada tanah gambut dengan kedelai sebagai tanaman indikator. J Floratek 40: 124-139 (in Indonesian). Subagjo H. 1983. Pedogenesis dua pedon Grumosol (Vertisols) dari bahan volkanik gunung Lawu dekat Ngawi dan Karanganyar. Pemberitaan Penelitian Tanah dan Pupuk 2: 8-18 (in Indonesian). Nursyamsi D and Suprihati. 2005. Soil chemical and mineralogical characteristics and its relationship with the fertilizers requirement for rice (Oriza sativa), maize (Zea mays) and soybean (Glycine max). Bul Agron 33: 40-47. Subagyo H, N Suharta and AB Siswanto. 2004. Tanah- tanah pertanian di Indonesia. In: A Adimihardja, LI Amien, F Agus and D Djaenudin (eds). Sumberdaya Lahan Indonesia dan Pengelolaannya. Pusat Penelitian dan Pengembangan Tanah dan Agroklimat, Bogor, pp. 21-66 (in Indonesian) Prasetyo BH and DA Suriadikarta. 2006. Karakteristik, potensi, dan teknologi pengelolaan tanah ultisol untuk pengembangan pertanian lahan kering di Indonesia. Litbang Pertanian 2: 39 (in Indonesian). Suharsih P, Setyanto P and AK Makarim. 2002. Emisi gas metan pada lahan sawah irigasi inceptisol akibat pemupukan nitrogen pada tanaman padi. Penelitian Pertanian Tanaman Pangan 22: 67-75 (in Indonesian) Prasetyo BH. 2007. Perbedaan sifat-sifat tanah vertisol dari berbagai bahan induk. J Ilmu-Ilmu Pertanian Indonesia 9: 20-31 (in Indonesian). Sulaeman, Suparto and Eviati. 2002. Enhancing the reactivity of phosphate rocks by acidulation. J Tanah dan Iklim 20: 69-79. RisnahS, P Yudono and A Syukur. 2013. Pengaruh abu sabut kelapa terhadap ketersediaan k di tanah dan serapan k pada pertumbuhan bibit kakao. J Ilmu Pertanian 16: 79-91. Suntoro WA. 2003. Peranan bahan organik terhadap kesuburan tanah dan upaya pengelolaannya. Pidato Pengukuhan Guru Besar. REFERENCES Khairatun N and RD Ningsih. 2013. Penggunaan pupuk organik untuk mengurangi pupuk anorganik dan peningkatan poduktivitas padi di lahan pasang surut. Seminar Nasional Inovasi Teknologi Pertanian, pp. 297-304 (in Indonesian). Abdurachman A, A Dariah and A Mulyani. 2008. Strategi dan teknologi pengelolaan lahan kering mendukung pengadaan pangan nasional. J Litbang Pertanian 27: 43-49 (in Indonesian). Afandi FN, Bambang S and Y Nuraini. 2015. Pengaruh pemberian berbagai jenis bahan organik terhadap sifat kimia tanah pada pertumbuhan dan produksi tanaman ubi jalar di Entisol Ngrangkah Pawon, Kediri. J Tanah dan Sumberdaya Lahan 2: 237-244 (in Indonesian). Khasawneh FE and EC Doll. 1978. The use of phosphate rock for direct application to soils. Adv Agron 30: 159-206. Kuntyastuti H, AWijanarko, RD Purwaningrahayu and A Taufiq. 2011. Pengaruh residu pupuk organik dan NPK terhadap perubahan dan kondisi tanah Vertisol Ngawi pada tanaman kedelai. Seminar Hasil Penelitian Tanaman Aneka Kacang dan Umbi, pp. 177-188 (in Indonesian). Borchardt G. 1989. Smectites. In: Mineral in Soil Environments. Second Edition. Madison: Soil Science Society of America, pp. 675-721. Brady NC and RR Weil. 2008. The Nature and Properties of Soil. 14 th Edition. USA: Pearson. Kusuma ME. 2012. Pengaruh takaran pupuk kandang kotoran burung puyuh terhadap pertumbuhan dan hasil tanaman sawi putih (Brassica juncea L.). J Ilmu Hewani Tropika 1: 7-11 (in Indonesian). Chand S, Anwar M and DD Patra. 2006. Influence of long-term application of organic and inorganic fertilizer to build up soil fertility and nutrient uptake in mint mustard cropping sequence. Commun Soil Sci Plant Anal 37: 63-76. Mandal B, PLG Vlek and LN Mandal. 1999. Beneficial effect of blue green algae and azolla excluding supplying nitrogen on wetland rice field. Biol Fertil Soils 28: 329-342. Deckers J, O Spaargaren and F Nachtergaele. 2001. Vertisols: Genesis, properties and soilscape management for sustainable development. In: Syers JKF, WT Penning De Vries and P Nyamudeza (eds). The Sustainable Management of Vertisols, IBSRAM Proceeding No. 20, pp. 3-20. Marschner H. 2012. Mineral Nutrition of Higher Plant. USA: Academic press. Mielean N and A Riosenberg. 2006. The effect of emissions of fertilizer production on the environment contamination by cadmium and arsenic in Southern Brazil. Environ Pollut 143: 335- 340. Elamin AE and MA Elagib. 2001. Comparative study of organic and inorganic fertilizers on forage corn (Zea mays L.) grown on two soil types. Qatar Univ Sci J 21: 47-54. Munir. 1996. Tanah-Tanah Utama di Indonesia. Pustaka Jaya. Jakarta (in Indonesian). REFERENCES 148 J Syamsiah et al.: Effects of Mix J Syamsiah et al.: Effects of Mixed Source Fertilizer Application on Vertisol Fertility Fakultas Pertanian, Universitas Sebelas Maret, Surakarta (in Indonesian). RisnahS, Y Prapto and S Abdul. 2013. Pengaruh abu sabut kelapa terhadap ketersediaan k di tanah dan serapan k pada pertumbuhan bibit kakao. J Ilmu Pertanian 16: 79-91 (in Indonesian). Tisdale SL, WL Nelson, JD Beatonand JL Havlin. 2010. Soil fertility and fertilizerFifth Edition.New York:Macmillan Publishing Company. Rizqiani NF, E Ambarwati and NW Yuwono. 2007. Pengaruh dosis dan frekuensi pemberian pupuk organik cair terhadap pertumbuhan dan hasil buncis (Phaseolus vulgaris L.) dataran rendah. J Ilmu Tanah dan Lingkungan 7: 43-53 (in Indonesian). Wahyudi I. 2009. Manfaat bahan organik terhadap peningkatan ketersediaan fosfor dan penurunan toksisitas aluminium di ultisol. [Disertasi]. Universitas Brawijaya, Malang (in Indonesian).
https://openalex.org/W4237618576	https://www.frontiersin.org/articles/10.3389/fpsyg.2019.02428/pdf	English	null	For whom does determinism undermine moral responsibility? Surveying the conditions for free will across cultures	null	2,019	cc-by	11,279	Edited by: Andrea Bender, University of Bergen, Norway Reviewed by: Joshua Knobe, Yale University, United States Annelie Rothe-Wulf, University of Freiburg, Germany Reviewed by: Joshua Knobe, Yale University, United States Annelie Rothe-Wulf, University of Freiburg, Germany Correspondence: Ivar R. Hannikainen ivar.hannikainen@gmail.com Correspondence: Ivar R. ORIGINAL RESEARCH published: 05 November 2019 doi: 10.3389/fpsyg.2019.02428 For Whom Does Determinism Undermine Moral Responsibility? Surveying the Conditions for Free Will Across Cultures Ivar R. Hannikainen1*, Edouard Machery2, David Rose3, Stephen Stich4, Christopher Y. Olivola5, Paulo Sousa6, Florian Cova7, Emma E. Buchtel8, Mario Alai9, Adriano Angelucci9, Renatas Berniûnas10, Amita Chatterjee11, Hyundeuk Cheon12, In-Rae Cho12, Daniel Cohnitz13, Vilius Dranseika14, Ángeles Eraña Lagos15, Laleh Ghadakpour16, Maurice Grinberg17, Takaaki Hashimoto18, Amir Horowitz19, Evgeniya Hristova17, Yasmina Jraissati20, Veselina Kadreva17, Kaori Karasawa18, Hackjin Kim21, Yeonjeong Kim22, Minwoo Lee21, Carlos Mauro23, Masaharu Mizumoto24, Sebastiano Moruzzi25, Jorge Ornelas26, Barbara Osimani27, Carlos Romero15, Alejandro Rosas López28, Massimo Sangoi9, Andrea Sereni29, Sarah Songhorian30, Noel Struchiner1, Vera Tripodi31, Naoki Usui32, Alejandro Vázquez del Mercado15, Hrag A. Vosgerichian19, Xueyi Zhang33 and Jing Zhu34 Hannikainen ivar.hannikainen@gmail.com 1 Department of Law, Pontiﬁcal Catholic University of Rio de Janeiro, Rio de Janeiro, Brazil, 2 Department of History and Philosophy of Science, University of Pittsburgh, Pittsburgh, PA, United States, 3 Department of Philosophy, Florida State University, Tallahassee, FL, United States, 4 Department of Philosophy, Rutgers University, New Brunswick, NJ, United States, 5 Tepper School of Business, Carnegie Mellon University, Pittsburgh, PA, United States, 6 Institute of Cognition and Culture, Queen’s University, Belfast, United Kingdom, 7 Department of Philosophy, University of Geneva, Geneva, Switzerland, 8 Department of Psychology, The Education University of Hong Kong, Tai Po, Hong Kong, 9 Department of Pure and Applied Sciences, University of Urbino Carlo Bo, Urbino, Italy, 10 Institute of Psychology, Vilnius University, Vilnius, Lithuania, 11 School of Cognitive Science, Jadavpur University, Kolkata, India, 12 Department of Philosophy, Seoul National University, Seoul, South Korea, 13 Department of Philosophy and Religious Studies, Utrecht University, Utrecht, Netherlands, 14 Institute of Philosophy, Vilnius University, Vilnius, Lithuania, 15 Instituto de Investigaciones Filosóﬁcas-UNAM, Mexico City, Mexico, 16 Independent Researcher, Tehran, Iran, 17 Department of Cognitive Science and Psychology, New Bulgarian University, Soﬁa, Bulgaria, 18 Department of Social Psychology, University of Tokyo, Tokyo, Japan, 19 Department of History, Philosophy and Judaic Studies, Open University of Israel, Ra’anana, Israel, 20 Department of Philosophy, American University of Beirut, Beirut, Lebanon, 21 Department of Psychology, Korea University, Seoul, South Korea, 22 Sloan School of Management, Massachusetts Institute of Technology, Cambridge, MA, United States, 23 CLOO Behavioral Insights Unit, Porto, Portugal, 24 School of Knowledge Science, Japan Advanced Institute of Science and Technology, Ishikawa, Japan, 25 Department of Philosophy and Communication Studies, University of Bologna, Bologna, Italy, 26 Faculty of Social Sciences and Humanities, Universidad Autónoma de San Luis Potosí, San Luis Potosí, Mexico, 27 Munich Center for Mathematical Philosophy, Ludwig Maximilians Universität, Munich, Germany, 28 Department of Philosophy, National University of Colombia, Bogotá, Colombia, 29 Faculty of Philosophy, Scuola Universitaria Superiore IUSS, Pavia, Italy, 30 Faculty of Philosophy, Vita-Salute San Raffaele University, Milan, Italy, 31 Department of Philosophy and Educational Sciences, University of Turin, Turin, Italy, 32 Department of Humanities, Mie University, Tsu, Japan, 33 School of Humanities, Southeast University, Nanjing, China, 34 School of Information Management, Sun Yat-sen University, Guangzhou, China Specialty section: This article was submitted to Cognitive Science, a section of the journal Frontiers in Psychology Received: 30 May 2019 Accepted: 14 October 2019 Published: 05 November 2019 Citation: Hannikainen IR, Machery E, Rose D, Stich S, Olivola CY, Sousa P, Cova F, Buchtel EE, Alai M, Angelucci A, Berniûnas R, Chatterjee A, Cheon H, Cho I-R, Cohnitz D, Dranseika V, Eraña Lagos Á, Ghadakpour L, Grinberg M, Hashimoto T, Horowitz A, Hristova E, Jraissati Y, Kadreva V, Karasawa K, Kim H, Kim Y, Lee M, Mauro C, Mizumoto M, Moruzzi S, Ornelas J, Osimani B, Romero C, Rosas López A, Sangoi M, Sereni A, Songhorian S, Struchiner N, Tripodi V, Usui N, Vázquez del Mercado A, Vosgerichian HA, Zhang X and Zhu J (2019) For Whom Does Determinism Undermine Moral Responsibility? Surveying the Conditions for Free Will Across Cultures. Front. Psychol. 10:2428. doi: 10.3389/fpsyg.2019.02428 Philosophers have long debated whether, if determinism is true, we should hold people morally responsible for their actions since in a deterministic universe, people are arguably not the ultimate source of their actions nor could they have done otherwise if initial conditions and the laws of nature are held ﬁxed. To reveal how non-philosophers ordinarily reason about the conditions for free will, we conducted a cross-cultural and cross-linguistic survey (N = 5,268) spanning twenty countries and sixteen languages. Overall, participants tended to ascribe moral responsibility whether the perpetrator Philosophers have long debated whether, if determinism is true, we should hold people morally responsible for their actions since in a deterministic universe, people are arguably not the ultimate source of their actions nor could they have done otherwise if initial conditions and the laws of nature are held ﬁxed. To reveal how non-philosophers ordinarily reason about the conditions for free will, we conducted a cross-cultural and cross-linguistic survey (N = 5,268) spanning twenty countries and sixteen languages. Overall, participants tended to ascribe moral responsibility whether the perpetrator November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 1 Conditions for Free Will Across Cultures Hannikainen et al. lacked sourcehood or alternate possibilities. However, for American, European, and Middle Eastern participants, being the ultimate source of one’s actions promoted perceptions of free will and control as well as ascriptions of blame and punishment. By contrast, being the source of one’s actions was not particularly salient to Asian participants. Finally, across cultures, participants exhibiting greater cognitive reﬂection were more likely to view free will as incompatible with causal determinism. INTRODUCTION universe at time t1 entail the state of the universe at time t2. Thus, agents never initiate the causal sequence resulting in their behavior; rather, people’s intentions, desires, and reasons are themselves the result of prior events. Hence, determinism also precludes what is known as ultimate sourcehood of one’s own acts (Pereboom, 2001): For example, if in a deterministic universe I make a salad instead of a burger for lunch, the causal chain of events that can be traced back from my action does not stop at my intentions, desires, and beliefs, but regresses to events that predate my own existence. The question of whether free will is compatible with determinism has fueled an intricate debate among philosophers (Strawson, 1963; Frankfurt, 1969; Fischer et al., 2009), which recent scientiﬁc work on human volition has enlivened (Dennett, 2004; Mele, 2008; Caruso, 2012). The crux of the free will problem can be outlined succinctly: In a deterministic universe, all events are the consequence of past events and the laws of nature. Then, if our universe is deterministic and every human action is the consequence of past events and the laws of nature, do people exercise control over their behavior in the way required for moral responsibility? Philosophical debate on this question has clustered around two proposed conditions for free will: alternate possibilities and ultimate sourcehood. The same is true of people’s achievements and crimes: In a deterministic universe, agents may proximately cause them, but their ultimate source traces back to the beginning of time. Some incompatibilist philosophers view this lack of sourcehood as undermining free will and moral responsibility (van Inwagen, 1983; Pereboom, 2001), while some compatibilist philosophers do not. In turn, non-philosophers have been shown to endorse incompatibilism when given an abstract description of a deterministic universe (Nichols and Knobe, 2007), and to endorse compatibilism when given a concrete description of a deterministic universe (Nahmias et al., 2005). According to the principle of alternate possibilities, free will and moral responsibility depend upon the ability to do otherwise. Since determinism implies that agents could not have done otherwise once initial conditions and the laws of nature are held ﬁxed, it follows that free will and moral responsibility are incompatible with determinism. INTRODUCTION This conclusion was challenged when philosopher Harry Frankfurt (1969) – in an ingenious appeal to counterfactual intervention – provided an inﬂuential argument against the principle of alternate possibilities, illustrated in the following thought experiment: Suppose that I want to stay home all day on Sunday to rest for the busy week ahead. Come Sunday, I cancel my plans to go hiking with friends. Instead, I spend the day watching a movie, cooking a meal, and taking a long nap on the couch. Unbeknownst to me, the door to my apartment was jammed and I would not have been able to go hiking, or leave at all, had I tried. In this circumstance I could not have done otherwise; but did I freely stay home anyway? According to the principle of alternate possibilities, I did not; but Frankfurt had the inﬂuential insight that we should think of my behavior as being freely willed despite my lack of alternate possibilities. Do people ordinarily conceive of free will as Frankfurt does? Some evidence has shown that North Americans typically agree with Frankfurt’s assessment that alternate possibilities are unnecessary for free will or moral responsibility (Miller and Feltz, 2011). Thus, existing empirical evidence reveals that laypeople hold agents morally responsible, even while acknowledging that the agent could not have done otherwise (Miller and Feltz, 2011), and did not originate the causal sequence resulting in action (Nichols and Knobe, 2007). However, past studies share an important limitation: With one notable exception (Sarkissian et al., 2010), they have been administered only to relatively small and homogeneous samples of respondents in the United States. Since cultures diﬀer considerably in their values (Hofstede et al., 2010) and philosophical judgments (Machery et al., 2004; Machery, 2017), whether past ﬁndings generalize across cultures remains very much an open question. Furthermore, given growing concerns about the replicability of key results in experimental psychology and philosophy (Cova et al., 2018), conducting a highly powered, cross-cultural generalization of core ﬁndings is a timely and important enterprise. Thus, our primary goal in this paper is to draw on a large and diverse sample of international respondents in order to examine intuitions regarding the conditions for free will and moral responsibility across cultures. In addition, the present study oﬀers two novel contributions. Determinism doesn’t merely entail the absence of alternate possibilities. Citation: We discuss these ﬁndings in light of documented cultural differences in the tendency toward dispositional versus situational attributions. Keywords: free will, compatibilism, cognitive style, situationism, dispositionism, sourcehood, alternate possibilities Frontiers in Psychology \| www.frontiersin.org INTRODUCTION In a deterministic universe, the laws of nature together with a complete description of the state of the November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 2 Conditions for Free Will Across Cultures Hannikainen et al. First, motivated by evidence of cultural variation in explanatory style, we examine whether Asians and Westerners diﬀer in their reasoning about free will and moral responsibility. Numerous experiments on social attribution have shown that people typically allude to an agent’s internal characteristics, such as her skills, desires, or personality, when attempting to explain why and how she behaves (Heider, 1958; Weiner, 1985) – a tendency known as dispositionism. For instance, when asked to describe a medicine mix-up at a children’s hospital, many participants say the accident happened because a particular pharmacy worker was “careless” or “irresponsible” (Chiu et al., 2000). Yet, a handful of comparative studies have provided compelling evidence that dispositionism may not be universally shared (Miller, 1984; Morris and Peng, 1994; Choi et al., 1999): In particular, East Asians are more likely to highlight features of the situation. For instance, when asked to describe the same hospital incident, Chinese participants tended to attribute this occurrence to extrinsic causes: e.g., the clinic’s failure to “assure the quality of the medicine” or “monitor the performance of the workers.” These diﬀerences in attributional style have been observed in naturalistic contexts too (Morris and Peng, 1994): When explaining a gruesome murder, an American newspaper (The New York Times) focused more on intrinsic features, while a Chinese newspaper (World Journal) focused more on situational factors. reward-sensitive (Smillie, 2013), as evidenced by neuroimaging studies of gambling behavior (Cohen et al., 2005). These contrasting styles of information and reward processing point to a potential explanation for the link between personality and compatibilist judgments: While introverts may tend to reﬂect more on the implications of determinism for free will, extraverts may be more immediately motivated to punish the perpetrator – consistent, also, with evidence that exacting third- party punishment is rewarding (de Quervain et al., 2004). To investigate this hypothesis, participants were asked to complete the Cognitive Reﬂection Test – a performance-based measure of individuals’ tendency to reconsider their initial intuitions (Frederick, 2005). In the sections that follow, we report the methods and results of our large-scale, cross-cultural study. INTRODUCTION We end with a general discussion that connects our ﬁndings to our two novel contributions, and which emphasizes the importance of our ﬁndings, and cross-cultural work more generally, for philosophical methodology. 1Due to a technical problem, the AS scenario was not administered in Colombia. Participants We invited 5,268 participants (46% women; age: Median = 26, Min = 18, Max = 88) from 21 diﬀerent locations around the world to complete a survey as part of an international collaboration. Supplementary Table S1 provides details regarding the demographic proﬁle, language, and medium of administration for each location. Developmental evidence on the emergence of free will beliefs has revealed corresponding cultural diﬀerences: Between ages four and eleven, children in the United States appear to strengthen their belief that individuals can choose to violate norms (e.g., to make a classmate cry), whereas Nepali children show no such change (Chernyak et al., 2013). In comparison to the characteristically Western sense that agents choose whether to act on desires and social norms, Asian cultures are more likely to treat such situational factors as direct constraints on human behavior. Taken together, these ﬁndings suggest that intuitions regarding the conditions for free will may be diﬀerent in Asian cultures. Procedure As part of a battery of questions concerning philosophical thought experiments, participants were randomly assigned to read one of two hypothetical murder scenarios narrated in the primary local language. For each site, we sought a bilingual translator to prepare all stimuli in the local language, except where previously validated and published translations were already available (as noted in the section “Measures”). Our second novel contribution is to characterize the motivational and cognitive underpinnings of compatibilist versus incompatibilist intuitions. Previous empirical research has shown that, at least among Americans, extraverts are more likely to view free will as compatible with determinism, while introverts are more likely to see them as incompatible (Feltz and Cokely, 2009) – a pattern that holds even among philosophers with expert knowledge about this very issue (Schulz et al., 2011). Half of our participants (n = 2,381) read the “Actual Sequence” (AS) scenario, previously employed in inﬂuential studies of moral psychology (Nichols and Knobe, 2007; Feltz et al., 2009; Cova et al., 2012; Rose and Nichols, 2013; Feltz and Cova, 2014; Murray and Nahmias, 2014).1 The AS scenario opens by situating the reader in a deterministic universe: Thus far, no research has sought to clarify why extraverts and introverts might diﬀer by probing into their divergent reasoning styles – and our present study helps to ﬁll this gap: Leading theories of extraversion point not only to greater interpersonal engagement, but also to diﬀerences in decision- making style (Depue and Collins, 1999). Speciﬁcally, introverts tend to adopt a rational thinking style (e.g., “I enjoy intellectual challenges”), while extraversion is more associated with an experiential thinking style (e.g., “I like to rely on my intuitive impressions”) and a dependence on habitual over goal-directed learning mechanisms (Pacini and Epstein, 1999; Skatova et al., 2013). A related literature suggests that extraverts are more Imagine a universe in which everything that happens is completely brought about by whatever happened before it. This is true from the very beginning of the universe, so what happened in the beginning of the universe brings about what happened next, and so on right up until the present. For example, 1 day John decided to have vegetable soup at lunch. Like everything else, this decision was completely brought about by what happened before it. Frontiers in Psychology \| www.frontiersin.org (5) CRT score: the sum of correct answers (ranging from 0: least reﬂective to 3: most reﬂective), The TIPI is a short-form of the ﬁve-factor model of personality that has been validated in several languages (Muck et al., 2007; Oshio et al., 2012; Renau et al., 2013; Chiorri et al., 2014), yielding a further individual diﬀerence measure: The year is 3072. A group of mad scientists has invented a sophisticated device that can monitor what is going on in a person’s mind. The device works at a distance by sending and receiving signals from a special chip that can be easily implanted into a person’s brain. With the device, the scientists can change a person’s decisions to engage in speciﬁc actions by simply sending signals to the special chip implanted in the person’s head and thereby manipulating the activation of the person’s neurons. (6) Extraversion: the two-item average on the Ten Item Personality Inventory (1: most introverted, 7: most extraverted). (6) Extraversion: the two-item average on the Ten Item Personality Inventory (1: most introverted, 7: most extraverted). Finally, we asked participants to provide the following sociodemographic information: age, gender, native language, nationality, country of residence, educational attainment, religiosity, and political orientation. One day, the scientists had a person inﬁltrate a clinic to ﬁnd people so that the chip could be secretly implanted in them. Martin is one of the subjects who receive the implant. Below we report how sample size was determined, all data exclusions, all manipulations, and all measures relevant to the present research question. Participants completed four additional tasks that were unrelated to (and therefore not analyzed for) the current study. This study was reviewed and approved by the Institutional Review Board at the University of Pittsburgh. The next day, while monitoring Martin’s thoughts, the scientists see that Martin is deliberating on a matter of great concern: whether to kill his friend Adam, who is having an aﬀair with Martin’s wife. The scientists agree that they will let Martin make his own decision, but that, if he decides not to kill Adam, they will make him change his mind by sending signals that reinforce his desire and reasons to kill Adam. In other words, regardless of Martin’s own ﬁnal decision, Martin will kill Adam, because the scientists are set on interfering if necessary. Martin decides to kill Adam and ends up killing Adam. The scientists didn’t have to interfere. Exclusion Criteria To identify participants who failed to understand the scenarios, we asked a comprehension question after each scenario. Following the CI scenario, participants were asked whether the scientists in that scenario had to interfere for the murder to take place, and they were excluded if they (incorrectly) responded that the scientists had to interfere. Following the AS scenario, participants were asked whether the murder “was brought about by whatever happened before it” and whether “it had to happen” in a single question, and they were excluded if they (incorrectly) responded that the murder was not brought about by whatever happened before it and that it did not have to happen. Procedure So, if everything in this universe was exactly the same up until John made his 1Due to a technical problem, the AS scenario was not administered in Colombia. Frontiers in Psychology \| www.frontiersin.org November 2019 \| Volume 10 \| Article 2428 3 Conditions for Free Will Across Cultures Hannikainen et al. decision, then it had to happen that John would decide to have vegetable soup at lunch. (4) Punishment: i.e., how much punishment he deserved (1: no punishment – 7: severe punishment). (4) Punishment: i.e., how much punishment he deserved (1: no punishment – 7: severe punishment). decision, then it had to happen that John would decide to have vegetable soup at lunch. (4) Punishment: i.e., how much punishment he deserved (1: no punishment – 7: severe punishment). Thus, the ﬁrst paragraph describes a deterministic universe in which agents (a) could not have done otherwise, and (b) are not the source of their actions. Then, the second paragraph describes a hypothetical agent carrying out a murder in those precise conditions: At the end of the survey, participants completed two further measures: the original (three-item) Cognitive Reﬂection Test (CRT; Frederick, 2005) and the Ten Item Personality Inventory (TIPI; Gosling et al., 2003). The CRT is a short assessment of participants’ cognitive style through three open-ended questions. For instance, one question asks: “If it takes 5 machines 5 min to make 5 widgets, how long would it take 100 machines to make 100 widgets?” Many participants provide the incorrect, intuitive answer that springs to mind, i.e., “100 min,” while others provide the correct answer, “5 min.” As an individual diﬀerence measure of cognitive style, we calculated: In this universe, a man named Bill has become attracted to his secretary, and he decides that the only way to be with her is to kill his wife and three children. Before he leaves on a business trip, he sets up a bomb that destroys his house and kills his family while he is away. A separate group of participants (n = 2,887) read a variant of the “Counterfactual Intervener” (CI) case (adapted from Frankfurt, 1969; Miller and Feltz, 2011) – in which the perpetrator is the source of the murder, but happens to lack alternate possibilities: (5) CRT score: the sum of correct answers (ranging from 0: least reﬂective to 3: most reﬂective), The Actual Sequence (AS) Scenario In the AS scenario, participants were more divided about whether the perpetrator acted freely (47%, 95% CI [44%, 49%]) and they tended to deny control (M = 3.31, SD = 2.38) – with both attributions signiﬁcantly below chance/midpoint, ps < 0.005. Still, attributions of blame (M = 4.67, SD = 2.41) and punishment (M = 5.04, SD = 2.32) remained signiﬁcantly above the scale midpoint, ps < 0.001. Therefore, in line with previous studies on the problem of free will (Nichols and Knobe, 2007; Miller and Feltz, 2011), people revealed compatibilist responses to both concrete scenarios – whether the agent was characterized as lacking alternate possibilities or sourcehood. Analytic Approach Statistical analyses were performed in R 3.5.1. In section “Separate Analyses,” we report the results for the AS and CI scenarios separately. Then, in section “Comparative Analyses,” we report comparative analyses of AS and CI data. In every regression and ANCOVA model, we enter age and gender as covariates to mitigate the impact of diﬀerences in sample composition across countries. Unless otherwise noted, regressions and ANCOVAs are random coeﬃcients and slopes models, estimated using the lme4 package (Bates et al., 2015), with participants nested within countries. P-values for the ﬁxed eﬀects were calculated using the Satterthwaite approximation in lmerTest (Kuznetsova et al., 2017). National and regional comparisons were based on estimated marginal means using the emmeans package (Lenth et al., 2019), with signiﬁcance levels adjusted using the Tukey method for familywise error correction. Mediation and moderated mediation analyses were conducted using the mediation package (Tingley et al., 2014) and conﬁdence intervals were derived through 5000 quasi-Bayesian Monte Carlo simulations (see Imai et al., 2010). Once again, one-way ANCOVAs entering site as a factor revealed diﬀerences in perceived control, F(19, 1577) = 6.13, η2 p = 0.07, blame, F(19, 1576) = 6.65, η2 p = 0.08, and punishment, F(19, 1576) = 5.03, η2 p = 0.06, while a corresponding logistic regression revealed diﬀerences in ascriptions of freedom, χ2(df = 19) = 86.9, all ps < 0.001. The ANCOVAs across world regions suggested that Asians ascribed signiﬁcantly more blame and punishment than either Middle Easterners or Europeans, ps < 0.001, but no more freedom or control, ps > 0.20 (see Table 2). There were few reliable diﬀerences between Europeans, Middle Easterners and Americans (only 2 of the 12 pairwise comparisons revealed signiﬁcant or suggestive diﬀerences). Raw data and analysis scripts are available at the following link on the Open Science Framework: https://osf.io/t3gzv/. Personality and Cognitive Style Turning to individual diﬀerences, we partially replicated the previously reported link between extraversion and free will judgments in the AS case (Feltz and Cokely, 2009, 2019), as shown in Table 1: Extraverts were more likely to ascribe freedom, r = 0.09, 95% CI [0.04, 0.13], and they attributed greater control, r = 0.11, 95% CI The Counterfactual Intervener (CI) Scenario This diﬃculty in conceiving causal determinism was expected in light of recent work on the ‘intrusion of intuitive metaphysics’ (Rose et al., 2015): namely, people’s strong conviction that the actual world is indeterministic hinders the suspension of disbelief required to engage in the AS scenario’s thought experiment. ( ) Despite lacking alternate possibilities, the perpetrator was seen as acting freely by a signiﬁcant majority of participants (82%, 95% CI [81%, 84%], binomial test: p < 0.001). Moreover, attributions of control (M = 5.00, SD = 2.28), blame (M = 5.90, SD = 1.64), and punishment (M = 5.99, SD = 1.44) were all signiﬁcantly above the scale midpoint (all one-sample t-test ps < 0.001). Thus, extending prior research (Miller and Feltz, 2011), participants tended to share Frankfurt’s (1969) intuition that alternate possibilities are not necessary for free will and moral responsibility. On one hand, since these participants are construing the universe as indeterministic, their responses cannot shed light on their intuitions regarding the conditions for free will and responsibility. Therefore, throughout the section “Results,” we only report analyses of the subsamples who demonstrated comprehension: 2,525 and 1,645 participants for the CI and AS cases, respectively. At the same time, the stark diﬀerence in comprehension rates across these two scenarios introduces the risk of an assignment bias, which could jeopardize the comparative analyses reported in section “Comparative Analyses.” To address this concern, in Supplementary Analysis 1 we provide statistical analyses for the full samples, and ﬁnd that many of our results (including our key ﬁnding) are unchanged when we do so. Separate ANCOVAs with site as factor revealed substantial diﬀerences in ascriptions of control, F(20, 2375) = 12.5, η2 p = 0.11, blame, F(20, 2370) = 18.9, η2 p = 0.16, and punishment across sites, F(20, 2363) = 9.47, η2 p = 0.08, all ps < 0.001, while a corresponding logistic regression revealed diﬀerences in ascriptions of freedom, χ2(df = 20) = 217.4, p < 0.001. To evaluate whether intuitions varied across world regions, we repeated the above analyses with respondents grouped by world region (rather than by site): Asian respondents ascribed less freedom, control, blame, and punishment than European, Middle Eastern, or American participants, all ps < 0.005 (see Table 2). In addition, European participants ascribed less control than either Middle Eastern or American participants, ps < 0.001. Power Analysis We heed Benjamin et al.’s (2017) recommendation of treating results as statistically signiﬁcant if their associated p-values are lower than 0.005 and as suggestive if they lie between 0.005 and 0.05. A power analysis for small eﬀects (r = 0.10) with 80% power revealed a target sample size of 782 participants to uncover suggestive evidence, and 1,325 participants to uncover a signiﬁcant eﬀect. Measures This study made use of four dependent measures, presented consecutively. Two measures assessed non-moral aspects of free will, namely: (1) Freedom: i.e., whether or not the agent acted freely when he killed his victim (1: yes, 0: no), and Nearly one in three participants (31%; n = 736) failed to comprehend the AS scenario – in contrast to one in eight who misunderstood the CI scenario (n = 362). That is, numerous participants ignored or disregarded the instructions stipulating “a universe in which everything that happens is completely brought about by whatever happened before it,” and (incorrectly) reported that the murderer’s decision “was not brought about by whatever happened before it” (emphasis added). (2) Control: i.e., how much control he had over killing his victim (1: no control – 7: complete control). In two further questions, we examined whether participants judged that the agent should be held morally responsible: (3) Blame: i.e., to what extent he was blameworthy (1: not at all – 7: extremely), and November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 4 Conditions for Free Will Across Cultures Hannikainen et al. RESULTS Separate Analyses Summary statistics are displayed in Table 1 and Supplementary Table 2. 1 November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 5 Conditions for Free Will Across Cultures Hannikainen et al. TABLE 1 \| Means, standard deviations, and correlations with conﬁdence intervals. CI AS 1 2 3 4 5 6 M SD M SD 1. Freedom 0.83 0.38 0.47 0.50 – 0.51∗∗ [0.47, 0.55] 0.46∗∗ [0.42, 0.50] 0.44∗∗ [0.40, 0.48] −0.16∗∗ [−0.21, −0.11] 0.09∗∗ [0.04, 0.13] 2. Control 5.00 2.28 3.31 2.38 0.31∗∗ [0.27, 0.34] – 0.50∗∗ [0.47,0.54] 0.45∗∗ [0.41,0.49] −0.17∗∗ [−0.22, −0.12] 0.11∗∗ [0.06, 0.16] 3. Blame 5.90 1.64 4.67 2.41 0.39∗∗ [0.36, 0.42] 0.28∗∗ [0.24, 0.31] – 0.80∗∗ [0.78, 0.81] −0.12∗∗ [−0.17, −0.07] 0.04 [−0.01, 0.09] 4. Punishment 5.98 1.44 5.04 2.32 0.44∗∗ [0.40, 0.47] 0.27∗∗ [0.24, 0.31] 0.64∗∗ [0.62, 0.67] – −0.10∗∗ [−0.15, −0.06] 0.03 [−0.02, 0.07] 5. CRT score 1.43 1.19 1.56 1.21 0.04∗ [0.00, 0.08] 0.00 [−0.04, 0.04] 0.08∗∗ [0.04, 0.12] 0.09∗∗ [0.05, 0.13] – −0.13∗∗ [−0.18, −0.08] 6. Extraversion 4.05 1.46 3.99 1.49 −0.01 [−0.05, 0.02] 0.02 [−0.02, 0.06] 0.02 [−0.02, 0.06] −0.01 [−0.05, 0.03] −0.06∗∗ [−0.10, −0.02] – M and SD represent mean and standard deviation, respectively. Values in square brackets indicate the 95% conﬁdence interval for each correlation separately by condition: CI (below diagonal; gray background); AS (above diagonal). ∗p < 0.05; ∗∗p < 0.005. TABLE 1 \| Means, standard deviations, and correlations with conﬁdence intervals. M and SD represent mean and standard deviation, respectively. Values in square brackets indicate the 95% conﬁdence interval for each correlation separately by condition: CI (below diagonal; gray background); AS (above diagonal). ∗p < 0.05; ∗∗p < 0.005. TABLE 2 \| Regression-based pairwise comparisons between world regions on each dependent measure. TABLE 2 \| Regression-based pairwise comparisons between world regions on each dependent measure. CI AS B t p B t p Freedom∗ −1.35 −9.88 < 0.001 0.04 0.39 0.98 Asia Europe Control −0.39 −3.38 0.004 0.26 1.92 0.22 Blame −0.31 −3.73 0.001 0.81 5.97 < 0.001 Punishment −0.46 −6.35 < 0.001 0.64 4.92 < 0.001 Freedom∗ −1.17 −4.59 < 0.001 0.20 0.95 0.78 Asia Middle east Control −1.31 −6.47 < 0.001 0.11 0.45 0.97 Blame −0.72 −5.00 < 0.001 1.00 4.05 < 0.001 Punishment −0.66 −5.18 < 0.001 1.08 4.57 < 0.001 Freedom∗ −2.06 −9.94 < 0.001 −0.10 −0.67 0.91 Asia N. and S. 2When simultaneously entering both CRT scores and extraversion ratings as predictors, we ﬁnd that CRT scores predict judgments of freedom (β = −0.33, 95% CI [−0.48, −0.18], z = −4.44, p < 0.001), control (β = −0.43, 95% CI [−0.57, −0.30], t = −6.43), blame (β = −0.27, 95% CI [−0.43, −0.11], t = −3.48, p = 0.004), and punishment (β = −0.24, 95% CI [−0.40, −0.08], t = 3.04, p = 0.007). Extraversion independently predicted judgments of freedom (β = 0.12, 95% CI [0.00, 0.24], z = −1.96, p = 0.042) and control (β = 0.19, 95% CI [0.06, 0.32], t = 2.86, p = 0.007), but not blame (β = 0.05, 95% CI [−0.08, 0.18], t = 0.79, p = 0.43) or punishment (β = 0.03, 95% CI [−0.12, 0.17], t = 0.40, p = 0.69). RESULTS America Control −0.95 −7.17 < 0.001 −0.02 −0.10 1 Blame −0.90 −9.61 < 0.001 0.34 1.89 0.23 Punishment −0.72 −8.69 < 0.001 0.28 1.61 0.37 Freedom∗ −0.18 0.69 0.90 0.15 0.74 0.88 Europe Middle east Control −0.92 −4.69 < 0.001 −0.15 −0.61 0.93 Blame −0.41 −2.97 −0.016 0.19 0.78 0.86 Punishment −0.19 −1.58 0.39 0.44 1.88 0.24 Freedom∗ −0.71 −3.25 0.006 −0.15 −0.97 0.77 Europe N. and S. America Control −0.55 −4.35 < 0.001 −0.28 −1.55 0.41 Blame −0.59 −6.54 < 0.001 −0.46 −2.59 0.047 Punishment −0.26 −3.20 0.008 −0.36 −2.08 0.16 Freedom∗ −0.89 −2.89 0.020 −0.3 −1.29 0.57 Middle East N. and S. America Control 0.36 1.72 0.31 −0.13 0.47 0.97 Blame −0.18 −1.23 0.61 −0.66 −2.40 0.077 Punishment −0.06 −0.47 0.97 −0.80 −3.05 0.013 ∗coefﬁcients for the Freedom measure are on the logit scale. In line with prior research (Pacini and Epstein, 1999), personality was linked to cognitive style—with extraverts scoring lower on the CRT (α = 0.73) than introverts, r(4006) = −0.09 [−0.12, −0.06], p < 0.001. We therefore sought to understand [0.06, 0.16], both ps < 0.001. However, they did not attribute greater blame, r = 0.04, 95% CI [−0.01, 0.09], or punishment, r = 0.02, 95% CI [−0.02, 0.07], than introverts, both ps > 0.05. November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 6 Conditions for Free Will Across Cultures Hannikainen et al. whether cognitive style better predicted compatibilist judgments than did extraversion. greater control, and also deserved greater blame and harsher punishment than the perpetrator in the AS case (ps < 0.005 with ﬁve exceptions).3 Indeed, CRT scores systematically correlated with all four attributions: More reﬂective participants were less likely to ascribe freedom, r = −0.16, 95% CI [−0.21, −0.11], they attributed less control, r = −0.17, 95% CI [−0.22, −0.13], and also viewed the perpetrator as less worthy of blame, r = −0.12, 95% CI [−0.17, −0.07], or punishment, r = −0.10, 95% CI [−0.15, −0.06], all ps < 0.001 – even after controlling for diﬀerences in extraversion.2 In sum, we successfully replicated some eﬀects of extraversion – but, overall, cognitive style revealed a more robust association with judgments of free will and moral responsibility than did personality (see also the Supplementary Analysis 1). 3Only ﬁve out of 56 tests revealed p-values > 0.005, a result that might be expected by chance alone: Germany-blame p = 0.057, Germany-punishment p = 0.064. Italy- blame p = 0.026. Switzerland-control p = 0.061. Portugal-control p = 0.020. As seen in Figure 1, every eﬀect trended in the predicted direction, ps < 0.10. The Role of Control Across Cultures The Role of Control Across Cultures Attribution theory (Heider, 1958; Weiner, 1985) predicts that the greater tendency to hold agents with sourcehood morally responsible (e.g., in the CI condition) arises from the perception that the agent exercised greater control over their behavior (see also Monroe et al., 2014; Monroe and Malle, 2019). Speciﬁcally, the CI condition, in which the counterfactual intervener did not have to interfere, may invite an intrinsic attribution – i.e., for participants to see the murderer’s actions as a reﬂection of his evil character. Meanwhile, the AS case – in which behavior is ultimately explained by antecedent causes and the laws of nature – invites an extrinsic attribution, and the corresponding assessment that the agent should not be held morally responsible. RESULTS In contrast, among most Asian countries, we found no corresponding eﬀect of scenario: Rather, Asian participants tended to assign comparable blame and punishment to both perpetrators, i.e., in mainland China, Hong Kong, India, and Indonesia, ps > 0.05. Moreover, we failed to ﬁnd any diﬀerence in judgments about whether the agent was in control (for mainland China, Hong Kong, India, and South Korea, ps > 0.250) or acted freely (for mainland China, Hong Kong, and Indonesia, ps > 0.250). A multivariate meta-analysis with site and dependent measure as cross-classiﬁed random eﬀects conﬁrmed this pattern of cultural moderation: There was substantial heterogeneity across world regions, QM(df = 3) = 12.5, p = 006. Speciﬁcally, though no diﬀerences emerged between Europe, North and South America, and the Middle East, zs < 1, ps > 0.50, Asia revealed a reduced eﬀect across dependent measures and sites, r = −0.23, 95% CI [−0.40, −0.06] z = −2.66, p = 0.008 (see Figure 1). Comparative Analyses The AS and CI scenarios diﬀer in a few respects; yet, the most salient philosophical diﬀerence concerns the question of ultimate sourcehood (Pereboom, 2001). Neither perpetrator could have done otherwise, but they diﬀer with regards to sourcehood: The perpetrator in the AS case lacks sourcehood, while the perpetrator in the CI case is the ultimate source of his actions. Does the presence of sourcehood promote the perception that an agent acted freely and should be held morally responsible? To shed light on this question, we compare responses to both thought experiments below. Cultural Differences To begin with, we conducted a logistic regression (for the dichotomous freedom measure) and two-way ANCOVAs (for the control, blame, and punishment measures), entering scenario, site, and the scenario × site interaction. These analyses revealed main eﬀects of scenario, Fs > 271, and site, Fs > 6, on all dependent measures, all ps < 0.001. The main eﬀect of scenario indicated that participants were more likely to ascribe freedom OR = 6.99, 95% CI [5.75, 8.55], and they ascribed greater control B = 1.56, 95% CI [1.41, 1.73], blame B = 1.18, 95% CI [1.05, 1.32], and punishment B = 0.93, 95% CI [0.80, 1.06], to the perpetrator in the CI condition than in the AS condition, all ps < 0.001. To evaluate this prediction, we averaged blame and punishment ratings to form an index of moral responsibility (Cronbach’s α = 0.86), and tested whether attributions of control statistically mediate the diﬀerence in moral responsibility ascription across scenarios. As with all of our prior models, both our mediator and outcome models incorporated age and gender as covariates: We also observed a scenario × site interaction in every model: freedom, χ2(19, N = 3987) = 174.8; control, F(19, 4005) = 8.11, ηp2 = 0.04; blame, F(19, 3999) = 8.94, ηp2 = 0.04; and punishment, F(19, 3992) = 7.59, ηp2 = 0.04, all ps < 0.001 – pointing toward cultural diﬀerences in the import of sourcehood for free will. control = a × scenario + x1 × age + x2 × gender + i Tests of the simple eﬀect of scenario in each site revealed systematic regional variation, dovetailing with previously reported East-West diﬀerences (Miller, 1984; Morris and Peng, 1994; Choi et al., 1999): Throughout American, European, and Middle Eastern countries, participants overwhelmingly judged that the perpetrator in the CI case acted more freely, exercised (mediator model) y1 × age + y2 × gender + i (outcome model) We calculated the proportion of the total eﬀect that is mediated (i.e., prop. mediated). The total eﬀect equals the sum of the direct eﬀect (expressed by coeﬃcient c in the outcome model), and the indirect eﬀect via the mediator (which amounts to a × b, the November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 7 Conditions for Free Will Across Cultures Hannikainen et al. FIGURE 1 \| Mean ascriptions of freedom, control, blame and punishment by scenario. 4Note that the proportion of the eﬀect that is mediated can be negative when the direct and indirect eﬀects have opposite signs. prop. mediated = ab/(ab + c) Averaging the indirect eﬀect estimate, ab, across 5000 Monte Carlo simulations yields the average causally mediated eﬀect, ACME. As expected, control attributions mediated the eﬀect of scenario on moral responsibility judgments among American (n = 726; ACME = 0.72, 95% CI [0.47, 0.98]; prop. mediated = 0.52, 95% CI [0.31, 0.98]), European (n = 1711; ACME = 0.57, 95% CI [0.44, 0.72]; prop. mediated = 0.41, 95% CI [0.29, 0.59]), and Middle Eastern (n = 255; ACME = 1.34, 95% CI [0.56, 2.17], prop. mediated = 0.66, 95% CI [0.43, 1.15]) participants; ps < 0.005. In each of these world regions, people viewed the perpetrator in the CI scenario as having exercised greater control than the perpetrator in the AS scenario. This diﬀerence in perceived control accounted for part of the eﬀect of scenario on moral responsibility evaluations – as predicted by theories of attributional reasoning. Cultural Differences Observed means for each location (/site) and their 95% conﬁdence intervals are plotted on the x-axis. A dotted vertical line represents the scale midpoint, and world region means are displayed using solid and dashed vertical lines. FIGURE 1 \| Mean ascriptions of freedom, control, blame and punishment by scenario. Observed means for each location (/site) and their 95% conﬁdence intervals are plotted on the x-axis. A dotted vertical line represents the scale midpoint, and world region means are displayed using solid and dashed vertical lines. product of the eﬀect of scenario in the mediator model and the eﬀect of judged control in the outcome model); therefore: (Miller, 1984; Morris and Peng, 1994; Choi et al., 1999; Chiu et al., 2000; Feinberg et al., 2019). A wide literature reveals that Asian individuals gravitate toward situational (and not dispositional) attributions – which could explain why they evaluate these two crimes in a very similar way – i.e., by viewing both perpetrators as subject to extrinsic pressures that drove them to act. prop. mediated = ab/(ab + c) Moderated Mediation Model As demonstrated in section “Personality and Cognitive Style,” reﬂective participants ascribed less moral responsibility to the perpetrator in the AS case. No corresponding association was shown for the CI case – if anything, reﬂective participants ascribed slightly more moral responsibility to the perpetrator, when merely lacking alternate possibilities (see Table 1 and Figure 2). Furthermore, the mitigating eﬀect of cognitive reﬂection emerged only among participants who grasped determinism, whereas the associations with personality generalized to those who misunderstood the AS scenario and perceived the perpetrator’s behavior as causally indetermined (see Supplementary Analysis 1). In contrast, the mediation model failed to predict the data obtained throughout Asian countries (n = 1302; ACME = −0.08, 95% CI [−0.01, 0.17], p = 0.092; prop. mediated = −0.11, 95% CI [−5.57, 5.06], p = 0.86).4 This result dovetails with abundant evidence that Asian cultures diﬀer in their explanatory style Taken together, these results may indicate that cognitive reﬂection supports the conclusion that ultimate sourcehood is a condition for free will and moral responsibility (see, e.g., Pereboom, 2001). Perhaps, reﬂective individuals are more likely to engage in dispositionist reasoning, and ascribe control when the agent in question is seen as the ultimate source of her behavior. November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 8 Conditions for Free Will Across Cultures Hannikainen et al. FIGURE 2 \| Mean ascriptions by scenario and CRT score. We plot observed means for each dependent measure, grouped by location (/site) and score on the CRT. Point size is proportional to the number of observations. A dotted horizontal line represents the scale midpoint, and linear trends by world region are displayed using solid and dashed lines. FIGURE 2 \| Mean ascriptions by scenario and CRT score. We plot observed means for each dependent measure, grouped by location (/site) and score on the CRT. Point size is proportional to the number of observations. A dotted horizontal line represents the scale midpoint, and linear trends by world region are displayed using solid and dashed lines. Moderated Mediation Model Indeed, reﬂective participants perceived a greater diﬀerence in freedom, OR = 2.08, 95% CI [1.62, 2.67], z = 5.76, and control, B = 0.47, 95% CI [0.30, 0.64], t = 5.42, across scenarios; ps < 0.001.5 Meanwhile, in the same models, the interaction between extraversion and scenario was far from signiﬁcant (freedom: OR = 0.92, 95% CI [0.77, 1.11], z = −0.88, p = 0.38; control B = −0.14, 95% CI [−0.31, 0.04], t = −1.53, p = 0.13). selectively ascribed less moral responsibility to the agent in the AS scenario than did intuitive individuals, and this eﬀect was mediated by diﬀerences in the degree to which they viewed the agent as exercising control. In several Asian countries, participants did not perceive a diﬀerence in control across scenarios – which we interpreted in light of their documented emphasis on situationist explanations. Might reﬂection therefore play a qualitatively distinct role in attitudes toward the free will problem in these cultural contexts? To the contrary, even throughout Asian countries, cognitive reﬂection supported incompatibilist responses to the AS case6 (freedom OR = 1.33, 95% CI [1.06, 1.67], z = 2.43, p = 0.015; control B = 0.42, 95% CI [0.18, 0.66], t = 3.38, p < 0.001) and moderated the indirect (ACME = 0.09, 95% CI [0.04, 0.14], p < 0.001) eﬀect of scenario on moral responsibility. Does the greater diﬀerence in control ascriptions among reﬂective participants explain why they tended to selectively exculpate the agent in the AS scenario on measures of moral responsibility? To assess this hypothesis, we entered CRT score and the CRT×scenario interaction into the mediator and outcome models deﬁned in section “The Role of Control Across Cultures.” We can now ask whether the magnitudes of c (the direct eﬀect) and ab (the indirect eﬀect) reliably depend on participants’ CRT score. If cognitive reﬂection underlies dispositionist reasoning, we should observe that the indirect eﬀect via control is larger among reﬂective (than unreﬂective) participants. In sum, reﬂective participants ascribed less freedom and control – and derivatively, assigned reduced moral responsibility – to the perpetrator in the AS condition (but not in the CI condition). This eﬀect of cognitive reﬂection accounted for previously reported associations with extraversion, and emerged even throughout Asian cultures that tended not to sharply distinguish the AS and CI scenarios. 5Corresponding eﬀects emerged for blame and punishment: blame B = 0.36, 95% CI [0.20, 0.53], t = 4.36; punishment B = 0.32, 95% CI [0.18, 0.46], t = 4.42; both ps < 0.001. 6Looking exclusively at Asian countries, CRT scores predicted freedom r = −0.12, p = 0.014; control r = −0.14, p = 0.006; blame r = −0.13, p = 0.009; and punishment r = −0.15, p = 0.004, attributions in the AS condition. Moderated Mediation Model Importantly, failure Indeed, as depicted in Figure 3, CRT score moderated the indirect eﬀect of scenario on moral responsibility, B = 0.13, 95% CI [0.09, 0.17], p < 0.001. In other words, reﬂective individuals November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 9 Hannikainen et al. Conditions for Free Will Across Cultures FIGURE 3 \| Moderated mediation diagram. Cognitive style moderates the indirect and direct effects of scenario on attributions of moral responsibility via perceived control (both ps < 0.005). on diagram. Cognitive style moderates the indirect and direct effects of scenario on attributions of moral responsibility via perceived FIGURE 3 \| Moderated mediation diagram. Cognitive style moderates the indirect and direct effects of scenario on attributions of moral responsibility via perceived control (both ps < 0.005). result from a causal chain originating at the beginning of the universe, explanations of this sort – implying sourcehood – seem particularly unsatisfactory and incomplete. In contrast, from a situationist perspective, whether the agent could be seen as the source of her action may be largely irrelevant: Instead, a situationist may think of others’ behavior as the product of extrinsic pressures – from momentary upheaval, to the way they were raised, social norms or fate – and thus perceive both agents, in the CI and AS cases, as similar in matters of free will and moral responsibility. to comprehend the scenarios is unlikely to explain these results, since we excluded participants who failed our comprehension checks in each condition. Though not predicted by our theory, cognitive reﬂection also moderated the direct eﬀect of scenario, B = 0.28, 95% CI [0.18, 0.37], p < 0.001, perhaps because incompatibilist reasoning is not based exclusively on assessments of agent control, but also on constructs we did not observe in this study, such as beliefs about the agent’s genuine or second-order desires (Frankfurt, 1969; Cullen, 2018). Throughout American, European, and Middle Eastern cultures, dispositionist explanations seemed to prevail: Participants tended to hold the perpetrator with sourcehood morally responsible, but ascribed less moral responsibility to the perpetrator without sourcehood – and this eﬀect was even larger among participants who exhibited greater cognitive reﬂection. Moderated Mediation Model Meanwhile, consistent with past evidence (Miller, 1984; Morris and Peng, 1994; Choi et al., 1999; Chiu et al., 2000), Asian cultures showed signs of both dispositionist and situationist reasoning: Overall, they tended not to distinguish between agents with and without sourcehood, suggesting a default preference for situationist explanations. Still, as with other world regions, reﬂective participants throughout Asia were more likely to selectively exculpate the agent lacking sourcehood. It could be that Asians tended intuitively toward situationist explanations, but were more likely to conjure alternative dispositionist explanations upon further reﬂection. Frontiers in Psychology \| www.frontiersin.org November 2019 \| Volume 10 \| Article 2428 Limitations First, throughout our comparative analyses we emphasized sourcehood as the primary factor driving the diﬀerence between scenarios. Yet, the AS and CI scenarios diﬀered in several other ways (e.g., the length of the vignette, the perpetrator’s motives), which could have contributed to the greater free will and moral responsibility ascriptions in the CI case. Second, participants faced acute comprehension diﬃculties in the AS scenario – introducing potential assignment bias into the comparative analyses. Even though the stimuli dictated that “everything that happens is completely brought about by whatever happened before it,” nearly a third of participants denied that the murderer’s decision had been brought about by prior events. This high rate of comprehension failure has been documented in previous studies, and attributed to overpowering indeterministic assumptions (Rose et al., 2015) as well as to a morally motivated denial of determinism (Clark et al., 2019). It may also have been aggravated by the complexity with which determinism was introduced in our vignette. In closing, our ﬁndings have certain implications for philosophical methodology: Debates about the problem of free will have generally focused on establishing the predominant intuition as an element in empirically informed argumentation. Given the present evidence of substantial individual and cultural variability, this approach seems misguided (Machery, 2017). Clearer insights into the free will problem may instead be gleaned by understanding the psychological and cultural bases of disagreement concerning the tension between determinism and free will. The results of our Supplementary Analysis 1 help to alleviate both concerns. (1) The eﬀect of cognitive reﬂection fully reversed among participants who misunderstood the AS scenario: Reﬂective participants who incorrectly construed the scenario as indeterministic ascribed more free will and moral responsibility than did intuitive participants – in line with our favored interpretation that (perceived) sourcehood promotes attributions of free will and moral responsibility under cognitive reﬂection. (2) When re-running the comparative analyses based on scenario assignment, regardless of comprehension, the scenario diﬀerence in free will and moral responsibility ascriptions remained signiﬁcant – suggesting that assignment bias alone did not produce the diﬀerence between scenarios. Still, in future research, we aim to employ maximally matched stimuli, and pre-test instructions that facilitate comprehension and balance exclusion rates across conditions. DATA AVAILABILITY STATEMENT The datasets generated for this study can be found in the Open Science Framework: https://osf.io/t3gzv/. ETHICS STATEMENT The studies involving human participants were reviewed and approved by the University of Pittsburgh. The patients/participants provided their written informed consent to participate in this study. Third, because our study did not speciﬁcally measure dispositional and situational explanatory styles, we submit that the observed diﬀerence between Asia and other world regions could instead be driven by other dimensions of cultural psychology, such as individualism versus collectivism (Triandis, 1995), the prevailing self-concept (Markus and Kitayama, 1991; Heine and Hamamura, 2007), or analytic versus holistic thinking (Nisbett et al., 2001; Feinberg et al., 2019). DISCUSSION At the aggregate level, we found that participants blamed and punished agents whether they only lacked alternate possibilities (Miller and Feltz, 2011) or whether they also lacked sourcehood (Nahmias et al., 2005; Nichols and Knobe, 2007). Thus, echoing early ﬁndings, laypeople did not take alternate possibilities or sourcehood as necessary conditions for free will and moral responsibility. Yet, our study also revealed a dramatic cultural diﬀerence: Throughout the Americas, Europe, and the Middle East, participants viewed the perpetrator with sourcehood (in the CI scenario) as freer and more morally responsible than the perpetrator without sourcehood (in the AS scenario). Meanwhile, South and East Asian participants evaluated both perpetrators in a strikingly similar way. We interpreted these results in light of cultural variation in dispositional versus situational attributions (Miller, 1984; Morris and Peng, 1994; Choi et al., 1999; Chiu et al., 2000). From a dispositionist perspective, participants may be especially attuned to the absence of sourcehood: When an agent is the source of their action, people may naturally conjure dispositionist explanations that refer to her goals, desires (e.g., because “she wanted a new life”) or character (e.g., because “she is ruthless”). In contrast, when actions Our study also shed new light on the understanding of individual diﬀerences in compatibilist beliefs. Several studies previously reported that extraverts and introverts diﬀer in their assessments of whether free will and determinism are compatible (Feltz and Cokely, 2009; Schulz et al., 2011), but yielded limited insight into why. Building on evidence that extraverts and introverts diﬀer in their cognitive style, we found that a tendency toward cognitive reﬂection largely November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 10 Conditions for Free Will Across Cultures Hannikainen et al. CONCLUSION subsumed the previously reported eﬀect of personality. This ﬁnding helps us interpret extraverts’ and introverts’ divergent attitudes toward the free will problem as the result of associated diﬀerences in trait reﬂectivity: In the wake of a heinous crime, we are immediately motivated to condemn the perpetrator, prior to reasoning about whether she exercised free will or control (Nichols and Knobe, 2007). While reﬂective individuals may evaluate the implications of determinism for free will, concluding that sourcehood (but not alternate possibilities) is a condition for free will, less reﬂective participants may readily attribute moral responsibility, and even ascribe free will, motivated by their initial punitive drive (Clark et al., 2014, 2017). After more than a decade of research on the cognitive science of free will, we have learned that laypeople’s beliefs vary systematically, depending on how (Nichols and Knobe, 2007) and who (Feltz and Cokely, 2009; Schulz et al., 2011) you ask. Yet, prior research on judgments of free will suﬀered from an important limitation: Most past studies relied on small and homogeneous North American samples (but see Sarkissian et al., 2010). The present work addressed this limitation by surveying thousands of participants in twenty countries and sixteen languages. In so doing, our study documented individual and cultural variations in views about the problem of free will. First, in most world regions, people ascribe greater free will to an agent who merely lacks alternate possibilities (in a Frankfurt case) than to one who also lacks sourcehood (in a deterministic universe). By contrast, being the source of one’s actions is not particularly salient throughout Asian cultures – perhaps due to the greater tendency toward situationist over dispositionist explanations of behavior. Second, reﬂective participants were more likely to treat ultimate sourcehood as a condition for free will, even in predominantly situationist cultures. Importantly, the cultural diﬀerence – i.e., between situationist and dispositionist cultures – was not the result of variations in cognitive style. Rather, culture and reﬂectivity independently contributed to lay judgments about the conditions for free will and moral responsibility. Frontiers in Psychology \| www.frontiersin.org REFERENCES doi: 10.1037//0033- 2909.125.1.47 Frankfurt, H. G. (1969). Alternate possibilities and moral responsibility. J. Philos. 66, 829–839. Frederick, S. (2005). Cognitive reﬂection and decision making. J. Econ. Perspect. 19, 25–42. doi: 10.1257/089533005775196732 Clark, C. J., Baumeister, R. F., and Ditto, P. H. (2017). Making punishment palatable: belief in free will alleviates punitive distress. Conscious. Cogn. 51, 193–211. doi: 10.1016/j.concog.2017.03.010 Gosling, S. D., Rentfrow, P. J., and Swann, W. B. Jr. (2003). A very brief measure of the big-ﬁve personality domains. J. Res. Personal. 37, 504–528. doi: 10.1016/ s0092-6566(03)00046-1 Clark, C. J., Luguri, J. B., Ditto, P. H., Knobe, J., Shariﬀ, A. F., and Baumeister, R. F. (2014). Free to punish: a motivated account of free will belief. J. Personal. Soc. Psychol. 106:501. doi: 10.1037/a0035880 Heider, F. (1958). The Psychology of Interpersonal Relations. New York, NY: Wiley. Heine, S. J., and Hamamura, T. (2007). In search of east asian self-enhancement. Heider, F. (1958). The Psychology of Interpersonal Relations. New York, NY: Wiley. Heine, S. J., and Hamamura, T. (2007). In search of east asian self-enhancement. Personal Soc Psychol Rev 11 4–27 doi: 10 1177/1088868306294587 Heine, S. J., and Hamamura, T. (2007). In search of east asian self-enhanc Personal. Soc. Psychol. Rev. 11, 4–27. doi: 10.1177/1088868306294587 Personal. Soc. Psychol. Rev. 11, 4–27. doi: 10.1177/1088868306294587 Clark, C. J., Winegard, B. M., and Baumeister, R. F. (2019). Forget the folk: moral responsibility preservation motives and other conditions for compatibilism. Front. Psychol. 10:215. doi: 10.3389/fpsyg.2019.00215 Hofstede, G., Hofstede, G. J., and Minkov, M. (2010). Cultures and Organizations: Software of the Mind Revised and Expanded New York NY: McGraw Hill Hofstede, G., Hofstede, G. J., and Minkov, M. (2010). Cultures and Organizations: Software of the Mind. Revised and Expanded. New York, NY: McGraw-Hill. Hofstede, G., Hofstede, G. J., and Minkov, M. (2010). Cultures and Organizations: Software of the Mind. Revised and Expanded. New York, NY: McGraw-Hill. Software of the Mind. Revised and Expanded. New York, NY: McG Cohen, M. X., Young, J., Baek, J. M., Kessler, C., and Ranganath, C. (2005). Individual diﬀerences in extraversion and dopamine genetics predict neural reward responses. Cogn. Brain Res. 25, 851–861. doi: 10.1016/j.cogbrainres. 2005.09.018 Imai, K., Keele, L., and Tingley, D. (2010). A general approach to causal mediation analysis. Psychol. Methods 15, 309–334. doi: 10.1037/a0020761 Kuznetsova, A., Brockhoﬀ, P. B., and Christensen, R. H. B. (2017). lmerTest package: tests in linear mixed eﬀects models. J. Stat. Softw. 82, 1–26. SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpsyg. 2019.02428/full#supplementary-material REFERENCES Depue, R. A., and Collins, P. F. (1999). Neurobiology of the structure of personality: dopamine, facilitation of incentive motivation, and extraversion. Behav. Brain Sci. 22, 491–517. doi: 10.1017/s0140525x99002046 Bates, D., Maechler, M., Bolker, B., and Walker, S. (2015). Fitting linear mixed- eﬀects models using lme4. J. Stat. Softw. 67, 1–48. Feinberg, M., Fang, R., Liu, S., and Peng, K. (2019). A world of blame to go around: cross-cultural determinants of responsibility and punishment judgments. Personal. Soc. Psychol. Bull. 45, 634–651. doi: 10.1177/0146167218794631 Benjamin, D. J., Berger, J. O., Johannesson, M., Nosek, B. A., Wagenmakers, E. J., Berk, R., et al. (2017). Redeﬁne statistical signiﬁcance. Nat. Hum. Behav. 2, 6–10. Personal. Soc. Psychol. Bull. 45, 634–651. doi: 10.1177/01461672 Feltz, A., and Cokely, E. (2019). Extraversion and compatibilist intuitions: a ten- year retrospective and meta-analyses. Philos. Psychol. 32, 388–403. doi: 10.1080/ 09515089.2019.1572692 y p year retrospective and meta-analyses. Philos. Psychol. 32, 388–403. doi: 10.1080/ 09515089.2019.1572692 Caruso, G. D. (2012). Free Will and Consciousness: A Determinist Account of the Illusion of Free Will. Lanham, MA: Lexington Books. Chernyak, N., Kushnir, T., Sullivan, K. M., and Wang, Q. (2013). A comparison of american and nepalese children’s concepts of freedom of choice and social constraint. Cogn. Sci. 37, 1343–1355. doi: 10.1111/cogs. 12046 Feltz, A., and Cokely, E. T. (2009). Do judgments about freedom and responsibility depend on who you are? Personality diﬀerences in intuitions about compatibilism and incompatibilism. Conscious. Cogn. 18, 342–350. doi: 10.1016/j.concog.2008.08.001 Chiorri, C., Bracco, F., Piccinno, T., Modaﬀeri, C., and Battini, V. (2014). Psychometric properties of a revised version of the ten item personality inventory. Eur. J. Psychol. Assess. 31, 109–119. doi: 10.1027/1015-5759/a000215 Feltz, A., Cokely, E. T., and Nadelhoﬀer, T. (2009). Natural compatibilism versus natural incompatibilism: back to the drawing board. Mind Lang. 24, 1–23. doi: 10.1111/j.1468-0017.2008.01351.x Feltz, A., and Cova, F. (2014). Moral responsibility and free will: a meta-analysis. Conscious. Cogn. 30, 234–246. doi: 10.1016/j.concog.2014.08.012 Chiu, C. Y., Morris, M. W., Hong, Y. Y., and Menon, T. (2000). Motivated cultural cognition: the impact of implicit cultural theories on dispositional attribution varies as a function of need for closure. J. Personal. Soc. Psychol. 78:247. doi: 10.1037//0022-3514.78.2.247 Fischer, J. M., Kane, R., Pereboom, D., and Vargas, M. (2009). Four Views on Free Will. Hoboken, NJ: John Wiley & Sons. Choi, I., Nisbett, R. E., and Norenzayan, A. (1999). Causal attribution across cultures: variation and universality. Psychol. Bull. 125:47. ACKNOWLEDGMENTS the manuscript. All authors were involved in data collection, and approved the ﬁnal manuscript for publication. We thank James Andow, Hugo Viciana and the members of the Pittsburgh Empirical Philosophy Lab, particularly Wesley Buckwalter, for feedback on previous versions of this manuscript. AUTHOR CONTRIBUTIONS EM, SS, and DR designed the study. IH analyzed and interpreted the data under the supervision of EM and CO. IH drafted the manuscript. EM, SS, DR, CO, EB, FC, and PS critically revised November 2019 \| Volume 10 \| Article 2428 11 Conditions for Free Will Across Cultures Hannikainen et al. FUNDING This research was made possible through the support of a grant (IH-112) from the Fuller Theological Seminary/Thrive Center, in concert with the John Templeton Foundation, awarded to EM and SS. The opinions expressed in this publication are those of the authors and do not necessarily reﬂect the views of the Fuller Thrive Center or the John Templeton Foundation. REFERENCES Cova, F., Bertoux, M., Bourgeois-Gironde, S., and Dubois, B. (2012). Judgments about moral responsibility and determinism in patients with behavioural variant of frontotemporal dementia: still compatibilists. Conscious. Cogn. 21, 851–864. doi: 10.1016/j.concog.2012.02.004 Lenth, R., Singmann, H., Love, J., Buerkner, P., and Herve, M. (2019). Package “emmeans”: Estimated Marginal Means, aka Least-Squares Means. Machery, E. (2017). Philosophy Within its Proper Bounds. Oxford: Oxford University Press. Cova, F., Strickland, B., Abatista, A., Allard, A., Andow, J., Attie, M., et al. (2018). Estimating the reproducibility of experimental philosophy. Rev. Philos. Psychol. 1–36. Machery, E., Mallon, R., Nichols, S., and Stich, S. P. (2004). Semantics, cross- cultural style. Cognition 92, B1–B12. Markus, H. R., and Kitayama, S. (1991). Culture and the self: implications for cognition, emotion, and motivation. Psychol. Rev. 98:224. doi: 10.1037//0033- 295x.98.2.224 Cullen, S. (2018). When do circumstances excuse? moral prejudices and beliefs about the true self drive preferences for agency-minimizing explanations. Cognition 180, 165–181. doi: 10.1016/j.cognition.2018. 06.021 Mele, A. R. (2008). Free Will and Luck. Oxford: Oxford University Press. Miller, J. G. (1984). Culture and the development of everyday social explanation. y Miller, J. G. (1984). Culture and the development of everyday social explanation. r, J. G. (1984). Culture and the development of everyday social e de Quervain, D. J., Fischbacher, U., Treyer, V., and Schellhammer, M. (2004). The neural basis of altruistic punishment. Science 305:1254. doi: 10.1126/science. 1100735 J. Personal. Soc. Psychol. 46:961. doi: 10.1037/0022-3514.46.5.961 Miller, J. S., and Feltz, A. (2011). Frankfurt and the folk: an experimental investigation of frankfurt-style cases. Conscious. Cogn. 20, 401–414. doi: 10. 1016/j.concog.2010.10.015 Dennett, D. C. (2004). Freedom Evolves. Penguin Books: Westminster. November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 12 Conditions for Free Will Across Cultures Hannikainen et al. Monroe, A. E., Dillon, K. D., and Malle, B. F. (2014). Bringing free will down to earth: people’s psychological concept of free will and its role in moral judgment. Conscious. Cogn. 27, 100–108. doi: 10.1016/j.concog.2014.04.011 Rose, D., and Nichols, S. (2013). The lesson of bypassing. Rev. Philos. Psychol. 4, 599–619. doi: 10.1007/s13164-013-0154-3 Sarkissian, H., Chatterjee, A., De Brigard, F., Knobe, J., Nichols, S., and Sirker, S. (2010). Is belief in free will a cultural universal? Mind Lang. 25, 346–358. doi: 10.1111/j.1468-0017.2010.01393.x g j g Monroe, A. E., and Malle, B. F. (2019). People systematically update moral judgments of blame. J. Personal. Soc. Psychol. 116, 215–236. REFERENCES doi: 10.1037/ pspa0000137 Schulz, E., Cokely, E. T., and Feltz, A. (2011). Persistent bias in expert judgments about free will and moral responsibility: a test of the expertise defense. Conscious. Cogn. 20, 1722–1731. doi: 10.1016/j.concog.2011.04.007 Morris, M. W., and Peng, K. (1994). Culture and cause: american and chinese attributions for social and physical events. J. Personal. Soc. Psychol. 67:949. doi: 10.1037/0022-3514.67.6.949 Skatova, A., Chan, P. A., and Daw, N. D. (2013). Extraversion diﬀerentiates between model-based and model-free strategies in a reinforcement learning task. Front. Hum. Neurosci. 7:525. doi: 10.3389/fnhum.2013.00525 ( ) between model-based and model-free strategies in a reinforcement learning task. Front. Hum. Neurosci. 7:525. doi: 10.3389/fnhum.2013.00525 Smillie, L. D. (2013). Extraversion and reward processing. Curr. Direct. Psychol. Sci. 22, 167–172. doi: 10.1177/0963721412470133 Muck, P. M., Hell, B., and Gosling, S. D. (2007). Construct validation of a short ﬁve-factor model instrument: a self-peer study on the german adaptation of the ten-item personality inventory (TIPI-G). Eur. J. Personal. Assess. 23, 166–175. Smillie, L. D. (2013). Extraversion and reward processing. Curr. Direct. Psychol. Sci. 22, 167–172. doi: 10.1177/0963721412470133 Murray, D., and Nahmias, E. (2014). Explaining away incompatibilist intuitions. Philos. Phenomenol. Res. 88, 434–467. doi: 10.1111/j.1933-1592.2012.00609.x Strawson, P. F. (1963). Freedom and resentment. Proc. Br. Acad. 48, 1–25 Nahmias, E., Morris, S., Nadelhoﬀer, T., and Turner, J. (2005). Surveying freedom: folk intuitions about free will and moral responsibility. Philos. Psychol. 18, 561–584. doi: 10.1080/09515080500264180 Tingley, D., Yamamoto, T., Hirose, K., Keele, L., and Imai, K. (2014). mediation: r package for causal mediation analysis. J. Stat. Softw. 59, 1–38. Triandis, H. C. (1995). New Directions in social psychology. Individualism & Collectivism. Boulder, CO: Westview Press. Triandis, H. C. (1995). New Directions in social psychology. Individualism & Collectivism. Boulder, CO: Westview Press. Nichols, S., and Knobe, J. (2007). Moral responsibility and determinism: the cognitive science of folk intuitions. Nous 41, 663–685. doi: 10.1111/j.1468-0068. 2007.00666.x van Inwagen, P. (1983). An Essay on Free Will. Oxford: Clarendon Press. van Inwagen, P. (1983). An Essay on Free Will. Oxford: Clarendon Press. Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychol. Rev. 92:548. doi: 10.1037//0033-295x.92.4.548 Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychol. Rev. 92:548. doi: 10.1037//0033-295x.92.4.548 Nisbett, R. E., Peng, K., Choi, I., and Norenzayan, A. (2001). Culture and systems of thought: holistic versus analytic cognition. Psychol. Rev. 108:291. Frontiers in Psychology \| www.frontiersin.org November 2019 \| Volume 10 \| Article 2428 REFERENCES doi: 10.1037/ /0033-295x.108.2.291 Conﬂict of Interest: The authors declare that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Oshio, A., Abe, S., and Cutrone, P. (2012). Development, reliability, and validity of the japanese version of the ten item personality inventory (TIPI-J). Japanese J. Personal. 21, 40–52. doi: 10.2132/personality.21.40 Pacini, R., and Epstein, S. (1999). The relation of rational and experiential information processing styles to personality, basic beliefs, and the ratio-bias phenomenon. J. Personal. Soc. Psychol. 76:972. doi: 10.1037/0022-3514.76.6.972 Copyright © 2019 Hannikainen, Machery, Rose, Stich, Olivola, Sousa, Cova, Buchtel, Alai, Angelucci, Berniûnas, Chatterjee, Cheon, Cho, Cohnitz, Dranseika, Eraña Lagos, Ghadakpour, Grinberg, Hashimoto, Horowitz, Hristova, Jraissati, Kadreva, Karasawa, Kim, Kim, Lee, Mauro, Mizumoto, Moruzzi, Ornelas, Osimani, Romero, Rosas López, Sangoi, Sereni, Songhorian, Struchiner, Tripodi, Usui, Vázquez del Mercado, Vosgerichian, Zhang and Zhu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Pereboom, D. (2001). Living Without Free Will. Cambridge: Cambridge University Press. Renau, V., Oberst, U., Gosling, S. D., Rusiñol, J., and Chamarro, A. (2013). Translation and validation of the ten-item-personality inventory into spanish and catalan. Aloma.Revista Psicol. Ciències Educació Esport 31, 85–97. Rose, D., Buckwalter, W., and Nichols, S. (2015). Neuroscientiﬁc prediction and the intrusion of intuitive metaphysics. Cogn. Sci. 41, 482–502. doi: 10.1111/cogs. 12310 November 2019 \| Volume 10 \| Article 2428 Frontiers in Psychology \| www.frontiersin.org 13
https://openalex.org/W2076407210	https://bmcinfectdis.biomedcentral.com/counter/pdf/10.1186/s12879-014-0610-4	English	null	Spatiotemporal clustering, climate periodicity, and social-ecological risk factors for dengue during an outbreak in Machala, Ecuador, in 2010	BMC infectious diseases	2,014	cc-by	12,297	RESEARCH ARTICLE Open Access * Correspondence: stewarta@upstate.edu 1Department of Microbiology and Immunology, Center for Global Health and Translational Science, State University of New York Upstate Medical University, 750 East Adams St, Syracuse, NY 13210, USA Full list of author information is available at the end of the article Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Spatiotemporal clustering, climate periodicity, and social-ecological risk factors for dengue during an outbreak in Machala, Ecuador, in 2010 Anna M Stewart-Ibarra1*, Ángel G Muñoz2,3, Sadie J Ryan1,4,5, Efraín Beltrán Ayala6,7, Mercy J Borbor-Cordova8, Julia L Finkelstein9,10, Raúl Mejía11, Tania Ordoñez6, G Cristina Recalde-Coronel8,11 and Keytia Rivero11 © 2014 Stewart-Ibarra et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Abstract Background: Dengue fever, a mosquito-borne viral disease, is a rapidly emerging public health problem in Ecuador and throughout the tropics. However, we have a limited understanding of the disease transmission dynamics in these regions. Previous studies in southern coastal Ecuador have demonstrated the potential to develop a dengue early warning system (EWS) that incorporates climate and non-climate information. The objective of this study was to characterize the spatiotemporal dynamics and climatic and social-ecological risk factors associated with the largest dengue epidemic to date in Machala, Ecuador, to inform the development of a dengue EWS. Methods: The following data from Machala were included in analyses: neighborhood-level georeferenced dengue cases, national census data, and entomological surveillance data from 2010; and time series of weekly dengue cases (aggregated to the city-level) and meteorological data from 2003 to 2012. We applied LISA and Moran’s I to analyze the spatial distribution of the 2010 dengue cases, and developed multivariate logistic regression models through a multi-model selection process to identify census variables and entomological covariates associated with the presence of dengue at the neighborhood level. Using data aggregated at the city-level, we conducted a time-series (wavelet) analysis of weekly climate and dengue incidence (2003-2012) to identify significant time periods (e.g., annual, biannual) when climate co-varied with dengue, and to describe the climate conditions associated with the 2010 outbreak. Results: We found significant hotspots of dengue transmission near the center of Machala. The best-fit model to predict the presence of dengue included older age and female gender of the head of the household, greater access to piped water in the home, poor housing condition, and less distance to the central hospital. Wavelet analyses revealed that dengue transmission co-varied with rainfall and minimum temperature at annual and biannual cycles, and we found that anomalously high rainfall and temperatures were associated with the 2010 outbreak. Conclusions: Our findings highlight the importance of geospatial information in dengue surveillance and the potential to develop a climate-driven spatiotemporal prediction model to inform disease prevention and control interventions. This study provides an operational methodological framework that can be applied to understand the drivers of local dengue risk. Keywords: Dengue fever, Aedes aegypti, GIS, Social-ecological, Climate, Spatial, Temporal, Wavelet analysis, Ecuador, Early warning system © 2014 Stewart-Ibarra et al.; licensee BioMed Central Ltd. Background Importantly, these analyses also need to consider the spatial and temporal scales of ongoing data collection and surveillance activities to ensure that the model out- puts can support an operational EWS. Dengue fever is the most significant mosquito-borne viral disease globally, and has rapidly increased in incidence, geographic distribution, and severity in recent decades [1-3]. The disease is caused by four distinct dengue virus serotypes (DENV 1-4) that are transmitted primarily by the female Aedes aegypti mosquito, with Aedes albopictus as a secondary vector. Common disease manifestations range from asymptomatic to moderate febrile illness, with a smaller proportion of patients who progress to severe illness characterized by hemorrhage, shock and death [4]. Integrated vector control and surveillance remain the principle strategies for disease prevention and control in endemic regions, as no vaccine or specific medical treat- ment are yet available. Macro social and environmental drivers have facilitated the global spread and persistence of dengue, including growing vulnerable urban popula- tions, global trade and travel, climate variability, and inad- equate vector control [5-8]. However, we have a limited understanding of the relative effects of these drivers at the local level, restricting our ability to predict and respond to site-specific dengue outbreaks. The National Institute of Meteorology and Hydrology (INAMHI) of Ecuador is coordinating efforts with the Ministry of Health (Ministerio de Salud Pública – MSP) to develop an operational dengue EWS for coastal regions of Ecuador, where the disease is hyper-endemic [25]. Our previous studies in southern coastal Ecuador demon- strated the potential to develop a dengue EWS that incorporates climate and non-climate information. We found that the magnitude and timing of dengue out- breaks were associated with anomalies in local climate, the El Niño Southern Oscillation (ENSO), the virus serotypes in circulation, and vector abundance [26]. Local field studies showed that dengue risk also depended on household risk factors (e.g., access to piped water infra- structure, demographics, water storage behaviors, housing conditions) [22]. Our recent advances in seasonal climate forecasts indicate that the forecasts in this region have considerable skill (i.e., predictive ability) [27,28]. Building on these previous studies, this study was conducted to characterize the spatiotemporal dynamics, climatic and social-ecological risk factors associated with the largest dengue epidemic (2010) on record in the coastal city of Machala, Ecuador, an important site for dengue surveil- lance in the region. Background Early warning systems (EWS) for dengue and other climate-sensitive diseases are decision-support tools that are being developed to improve the ability of the public health sector to predict, prevent, and respond to local disease outbreaks [9,10]. An EWS incorporates environ- mental data (e.g., climate, altitude, sea surface temperature), epidemiological surveillance data, and other social- ecological data in a spatiotemporal prediction model that generates operational disease risk forecasts, such as seasonal risk maps. Previous studies have demonstrated the utility of this approach for vector-borne diseases, including for dengue [11-13], malaria [14-16] and rift valley fever [17]. Maps and other model outputs are linked to an epidemic alert and response systems, trig- gering a chain of preventive interventions when an alert threshold is reached. Abstract This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 2 of 16 Page 2 of 16 Methods Study Area Machala, El Oro Province, is a mid-sized coastal port city (pop. 241,606) [29] located in southern coastal Ecuador, 70 kilometers north of the Peruvian border and 186 kilometers south of the city of Guayaquil (the epi- center of historical dengue outbreaks in the region). Dengue is an emerging disease in this region, with the first cases of dengue hemorrhagic fever (DHF) reported in 2005. The disease is now hyper-endemic, with year round transmission and co-circulation of all four sero- types. Recent multi-country studies showed that Machala had the highest Ae. aegypti larval indices of ten sites in other countries in Latin America and Asia [30,31] Given the high burden of disease, the high volume of people and goods moving across the Ecuador-Peru border, and prox- imity to Guayaquil, Machala is a strategic location to monitor and understand dengue transmission dynamics. One of the first steps in developing an EWS is to characterize the spatiotemporal dynamics and the covar- iates associated with historical disease transmission. This is often done by developing GIS base maps of epidemio- logical, environmental, and social data to identify risk factors; and through time series analyses of epidemio- logical and climate data. These analyses require cross- institutional integration of expertise and data, including epidemiological and entomological data from ministries of health, climate information from national institutes of meteorology, and social-ecological spatial data from national census bureaus. Previous studies indicate that associations among climate, socioeconomic indicators and dengue risk vary by location and time, indicating the need for analyses of dengue risk that consider the local context to explain transmission mechanisms [18-24]. In 2010, DENV-1 caused the largest dengue epidemic to date, with over 4,000 cases reported in El Oro prov- ince [32] (Figure 1A). In Machala, there were 2,019 cases of dengue fever (and 77 DHF) or an incidence of 84 dengue cases (and 3 DHF cases) per 10,000 population per year, compared to 25 dengue cases per 10,000 popu- lation per year from 2003 to 2009. The greatest burden Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 3 of 16 Page 3 of 16 Figure 1 Time series of dengue and local climate conditions in 2010 and historically in Machala, Ecuador. Data sources The following data from Machala were included in ana- lyses: neighborhood-level georeferenced dengue cases, national census data, and entomological surveillance data from 2010; and time series of weekly dengue cases (aggregated to the city-level) and meteorological data from 2003 to 2012. These data were examined to identify potential social-ecological and climate variables associated with the presence of dengue fever during the 2010 out- break in Machala, Ecuador. Epidemiological data were provided by INAMHI through a collaborative project with Epidemiological data INAMHI provided a map of georeferenced dengue cases from Machala in 2010, de-identified and aggregated to neighborhood-level polygons (n = 253) to protect the identity of individuals [34]. This map was generated from individual records of clinically suspected cases of dengue fever and DHF (aggregated as total dengue fever) reported to a mandatory MSP disease surveillance system, and the map included 83% of all dengue cases (n = 1,674) reported in 2010. Reported dengue cases were defined based on a clinical diagnosis. INAMHI also provided data for weekly dengue cases from Machala from 2003 to 2012 for the wavelet analysis described below. Methods Study Area (A) Weekly reported cases of dengue in 2010 and weekly average cases from 2003 to 2012; (B) weekly averages of rainfall and minimum air temperature (Tmin) in 2010 compared to the climatology (1986 to 2013 average conditions). Figure 1 Time series of dengue and local climate conditions in 2010 and historically in Machala, Ecuador. (A) Weekly reported cases of dengue in 2010 and weekly average cases from 2003 to 2012; (B) weekly averages of rainfall and minimum air temperature (Tmin) in 2010 compared to the climatology (1986 to 2013 average conditions). of disease (58%) during the epidemic was among individ- uals under 20 years of age (Figure 2). The number of cases in older adults (i.e., 11% of cases reported by people over 50) indicated a strong force of infection, and that dengue was a relatively new disease in the population. The epidemic occurred during a wetter than average year, with elevated vector indices. Cumulative rainfall from January to April 2010 was 56% above the 1986 to 2009 average. The percent of households with Ae. aegypti juveniles (House Index) was 21.7 ± 4.11 (mean ± 95% CI) in 2010 compared to 14.3 ± 4.70 from 2003 to 2009 [33]. the MSP that was sponsored by the Ecuadorian govern- ment. Accordingly, no formal ethical review was required, as the data used in this analysis were de-identified and aggregated to the neighborhood- and city-level, as described below. Entomological data ll Vector surveillance data for Ae. aegypti from 2010 was obtained from the National Service for the Control of Vector-Borne Diseases of the MSP, and included quar- terly House Indices (percent of households with Ae. aegypti juveniles) and Breteau Indices (number of con- tainers with Ae. aegypti juveniles per 100 households). The average Breteau Index during the first two quarters of 2010 (January to June), was the vector index that was most strongly associated with dengue presence (1) or absence (0) (Pearson correlation, r = 0.2, p = 0.001) and this period corresponded with the peak of the epidemic; accordingly, we selected this variable to test in the multi- variate model (Table 1). HCI ¼ CR þ CW þ CF ð Þ – min ½ = max – min ð Þ ð1Þ Using individual and household census records, we recoded selected census variables and calculated parame- ters as the percent of households or percent of the popula- tion per census sector (n = 558 census sectors). The data element dictionary of recoded variables in Spanish is pre- sented in Additional file 1: Table S1. To scale the sector- level polygon data to neighborhood-level polygons, we used the ‘isectpolypoly’ tool in Geospatial Modeling Envir- onment [35,36]. We estimated neighborhood population Social-ecological risk factors We extracted individual and household-level data from the 2010 Ecuadorian National Census [29] to test the Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 4 of 16 Page 4 of 16 Figure 2 Dengue incidence (per 10,000 population per year) by age and gender for (A) dengue fever and (B) dengue hemorrhagic fever in Machala in 2010. Figure 2 Dengue incidence (per 10,000 population per year) by age and gender for (A) dengue fever and (B) dengue hemorrhagic fever in Machala in 2010. hypothesis that social-ecological variables were associated with the presence of dengue (Table 1). We calculated a composite normalized housing condition index (HCI) for each household by combining variables for the condition of the roof (CR), condition of the walls (CW), and condi- tion of the floors (CF) (Equation 1). Each of the three vari- ables ranged from 1 to 3, where 3 indicated poor condition. When summed, the values of the composite index ranged from 3 (min) to 9 (max), and we normalized the index from 0 to 1, where 1 indicated the worst housing condition. by calculating the area-weighted sum, and estimated all other parameters by calculating area-weighted means. The neighborhood population estimates were also used to calculate neighborhood dengue prevalence and population density parameters. Other variables Average distance to the central hospital (km) Average Breteau Index during the first two quarters of 2010 the Granja Santa Ines weather station located in Machala (3°17'16” S, 79°54'5” W, 5 meters above sea level) and operated by INAMHI. The weekly climatology (1985- 2013) and weather during the study period are shown in Figure 1B. Weekly average rainfall and minimum temperature from 2003 to 2012 were included in the wavelet analysis, since it has been shown that these two climate variables explain an important part of the total variance of dengue cases in coastal Ecuador [26]. the Granja Santa Ines weather station located in Machala (3°17'16” S, 79°54'5” W, 5 meters above sea level) and operated by INAMHI. The weekly climatology (1985- 2013) and weather during the study period are shown in Figure 1B. Weekly average rainfall and minimum temperature from 2003 to 2012 were included in the wavelet analysis, since it has been shown that these two climate variables explain an important part of the total variance of dengue cases in coastal Ecuador [26]. autocorrelation [37] that identifies significant clusters (hot or cold spots) and outliers (e.g., nonrandom groups of neighborhoods with above or below the expected dengue prevalence). Previous studies have used Moran’s I and LISA to test the spatial distribution of dengue transmission [38], including in Ecuador [39], allowing for comparison between studies. Social-ecological risk factors Census data aggregated to the neighborhood-level were examined to identify potential social-ecological variables associated with the presence of dengue fever, including population density, human demographic characteristics, and housing condition (Table 1). We hypothesized that the presence or absence of dengue was associated with one or more of these factors; each factor was presented as a suite of census variables, representing testable vari- able ensemble hypotheses in a model selection frame- work - a modeling strategy that has been previously described [22]. Variables for the average distance to the public hospital (Teofilo Davila Hospital, the provincial hospital located in the city center) and the Breteau Climate data l Daily meteorological data (rainfall and minimum air temperature) during the study period were provided by Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 5 of 16 Page 5 of 16 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Table 1 Social-ecological parameters (mean and standard deviation - SD) tested in logistic regression models to predict dengue presence (1) and absence (0) at the neighborhood level in Machala in 2010 Parameter Mean SD Population density More than four people per bedroom (% households) 14.6% 6.4% Population density (people per square kilometer) 10,864 5,302 More than one other household sharing the home (% households) 2.2% 1.6% People per household 3.88 0.52 Demographics Receive remittances (% households) 10.8% 3.2% People emigrate for work (% households) 2.2% 1.2% Mean age of the head of the household (years) 45.2 3.0 Head of the household has primary education or less (% households) 35.9% 12.9% Afro-Ecuadorian (% population) 9.6% 6.9% Head of the household is unemployed (% households) 23.0% 5.3% Head of household is a woman (% households) 30.3% 4.5% Housing conditions Housing condition index (HCI), 0 to 1, where 1 is poor condition 0.29 0.10 No access to municipal garbage collection (% households) 8.0% 12.3% No piped water inside the home (% households) 34.4% 18.7% No access to sewerage (% households) 22.4% 27.6% No access to paved roads (% households) 26.7% 22.3% People drink tap water (% households) 32.8% 11.7% Rental homes (% households) 24.6% 10.6% Other variables Average distance to the central hospital (km) 2.36 1.27 Average Breteau Index during the first two quarters of 2010 28.6 2.15 Wavelet analysis We centered all variables and selected the best-fit models (GLM, family = binomial, link = logit) using glmulti, an R package for multimodel selection [40]. All possible unique models were tested and ranked based on Akaike’s Informa- tion Criterion (AIC) modified for small sample sizes (AICc) (Equation 2). We compared the top ranked model to the global model, which included all proposed variables as model parameters. To understand the time-frequency variability of dengue and climate during the 2010 epidemic, we conducted a wavelet analysis of a 10-year time series of weekly inci- dent dengue cases (2003-2012), rainfall and minimum temperature (Figure 4A). Wavelet analyses are ideal for noisy, non-stationary data, such as dengue cases data, which demonstrate strong seasonality and interannual variability (yearly changes) [41,42]. These analyses iden- tify significant temporal scales (i.e., defined here as pe- riods whose associated wavelet power is statistically significant for at least two continuous years; Figure 4) over time for a given variable, such as 2-year cycles or annual seasonal cycles of dengue transmission. Cross wavelet and wavelet coherency allowed us to compare two time series, such as climate and dengue, and to identify synchronous periods or signals. AIC ¼ 2k−ln L ð Þ AICc ¼ AIC þ 2k k þ 1 ð Þ n−k−1 ð2Þ ð2Þ Where k is the number of parameters in the model, n is the sample size, and L is the maximized likelihood function for the model. Where k is the number of parameters in the model, n is the sample size, and L is the maximized likelihood function for the model. Parameter estimates and 95% confidence intervals (CI) were calculated for variables in the top ranked model (Table 2 Model A). Variance inflation factors (VIF) were calculated to assess multi-colinearity and model disper- sion. We found that inclusion of the parameters for housing condition and piped water together led to over- dispersion and highly inflated variance in the best-fit model (Table 2 Model B). Based on the strong linear correlation between housing condition and access to piped water (Figure 3), we replaced these variables with the residuals of housing condition regressed on access to piped water, and re-ran glmulti to identify the best- fit model. The inclusion of this variable enabled testing The pre-processing of the time series data for Machala followed a two-step methodology described elsewhere in detail [27,28,43]. Wavelet analysis Table 2 The parameters included in the best-fit logistic regression models to predict the presence (1) or absence (0) of dengue in neighborhoods in Machala in 2010 Parameter Estimate 95% CI SE VIF P al rs included in the best-fit logistic regression models to predict the presence (1) or absence (0) rhoods in Machala in 2010 the best-fit logistic regression models to predict the presence (1) or absence (0) h l i 2010 y Exploratory spatial analysis We applied Moran’s I with inverse distance weighting (ArcMap 10.1) to epidemiological dengue data from 2010 to test the hypothesis that dengue cases were ran- domly distributed in space. Moran’s I is a global measure of spatial autocorrelation, that provides an index of dispersion from -1 to +1, where -1 is dispersed, 0 is random, and +1 is clustered. We identified the locations of significant dengue hot and cold spots using Anselin Local Moran’s I (LISA) with inverse distance weighting (ArcMap 10.1). The LISA is a local measure of spatial Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 6 of 16 for the effect of housing condition beyond that which was explained only by access to piped water. Index were also tested in the model to assess geographic differences, potential underreporting, and other factors (e.g., microclimate, vector control) not captured by the census variables. Wavelet analysis First, we quality-controlled the time series using a standard R package [44] to identify outliers and inconsistent values (e.g., minimum temperatures > maximum temperatures, negative precipitation values or negative frequency of dengue cases). Outliers were de- fined as data points at least three standard deviations above or below the mean. To account for real outliers (e.g., not artifacts produced by human, instruments, or transmission errors), we compared suspicious values with data from nearby climate stations. Entries that we Table 2 The parameters included in the best-fit logistic regression models to predict the presence (1) or absence (0) of dengue in neighborhoods in Machala in 2010 Parameter Estimate 95% CI SE VIF P value Model A. Intercept 0.75 0.46 – 1.05 0.15 < 0.001 Head of household is a woman 7.77 0.73 – 15.17 3.67 1.14 0.034 Age of head of household 0.10 0.00 – 0.21 0.05 1.06 0.051 Residual of HCI regressed on households with no access to piped water inside the home 9.04 3.98 – 14.37 2.64 1.05 < 0.001 Distance to central hospital −0.0005 −0.0007 – 0.0 0.0001 1.20 < 0.001 Model B. Intercept −7.59 −14.24 – −1.32 3.28 0.021 Head of household is a woman 7.30 0.11 – 14.83 3.74 1.18 0.051 Age of head of household 0.14 0.002 – 0.26 0.07 1.60 0.052 No piped water inside the home −3.18 −6.08 – −0.4 1.44 3.64 0.027 HCI 9.16 4.06 – 14.56 2.66 3.11 0.001 Distance to central hospital −0.001 −0.001 – 0.0 0.0 1.31 <0.001 (Model A.) The best fit model, which included the residual of the HCI regressed on the variable for no access to piped water inside the home. (Model B.) The same model is shown with separate parameters for the HCI and no access to piped water inside the home to indicate the direction of the effects of the parameters in the model. VIF values indicate a high degree of multicollinearity in model B compared to model A. High values of HCI indicate poor housing condition. Distance to central hospital (Model A.) The best fit model, which included the residual of the HCI regressed on the variable for no access to piped water inside the home. (Model B.) The same model is shown with separate parameters for the HCI and no access to piped water inside the home to indicate the direction of the effects of the parameters in the model. VIF values indicate a high degree of multicollinearity in model B compared to model A. High values of HCI indicate poor housing condition. (Model A.) The best fit model, which included the residual of the HCI regressed on the variable for no access to piped water inside the home. (Model B.) The same model is shown with separate parameters for the HCI and no access to piped water inside the home to indicate the direction of the effects of the parameters in the model. VIF values indicate a high degree of multicollinearity in model B compared to model A. High values of HCI indicate poor housing condition. Page 7 of 16 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 7 of 16 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Figure 3 Scatter plot and linear fit for the composite normalized housing condition index (HCI) versus the percentage of households with no piped water inside the home, for neighborhoods with cases of dengue (red circle) and without dengue (black triangle) in Machala in 2010. incidence using a linear interpolation of local population data from the 2001 and 2010 national censuses. The final step in data pre-processing involved the normalization of the three variables to constrain variability. Dengue incidence and rainfall time series had non-normal probability density functions, thus they were percentile-transformed [34]. The wavelet analysis of dengue and climate data enabled us to identify common periodicity patterns (e.g., annual or biannual signals) and anomalous climate conditions dur- ing the 2010 outbreak. We used Morlet wavelets in Matlab [49] to compute the time series’ wavelet power spectrum and to identify significant periods for each variable, cross- wavelet power to identify periods where dengue-rainfall and dengue-temperature had high common power, and the coherence spectra to identify local co-variability of dengue-rainfall and dengue-temperature [50]. Significance testing (p ≤0.05) was conducted using an AR1 back- ground noise for the first two spectra, and a Monte Carlo approach to compute the significance levels in the coher- ence spectrum. Distance to central hospital Statistically significant regions are dis- played enclosed by a solid black line in the wavelet plots; and cones of influence (COI), where edge effects increase the uncertainty of the analysis, are shown as a lighter shaded region (Figures 4, 5 and 6). The arrows represent the relative phase, which is indicative of the lags between the two time series, as determined by frequency and time [51].The direction of the arrows can be used to quantify the phase relationship: arrows pointing horizontally to the Figure 3 Scatter plot and linear fit for the composite normalized housing condition index (HCI) versus the percentage of households with no piped water inside the home, for neighborhoods with cases of dengue (red circle) and without dengue (black triangle) in Machala in 2010. deemed to be uncorrectable were flagged as missing values. Then we used the R package ‘RHTestsV4’ [45-48] to detect and correct temporal inhomogeneities in these variables. The climate time series did not need substantial correc- tions. Weekly dengue case data were transformed to weekly Figure 4 Dengue and climate wavelets. (A) Normalized time series of weekly dengue incidence, rainfall, and minimum temperature from 2003- 2012, and the wavelet power spectrum for (B) dengue incidence, (C) rainfall, and (D) minimum temperature. The black box indicates weeks 1-15 in 2010, when 75% of the cases from the epidemic were reported. Statistically significant regions are displayed enclosed by a solid black line in the wavelet plots; and cones of influence (COI), where edge effects increase the uncertainty of the analysis, are shown as a lighter shaded region. Figure 4 Dengue and climate wavelets. (A) Normalized time series of weekly dengue incidence, rainfall, and minimum temperature from 2003- 2012, and the wavelet power spectrum for (B) dengue incidence, (C) rainfall, and (D) minimum temperature. The black box indicates weeks 1-15 in 2010, when 75% of the cases from the epidemic were reported. Statistically significant regions are displayed enclosed by a solid black line in the wavelet plots; and cones of influence (COI), where edge effects increase the uncertainty of the analysis, are shown as a lighter shaded region. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 8 of 16 Figure 5 As in Figure 4, but for the cross wavelet power spectrum for (A) dengue-rainfall and (B) dengue-minimum temperature. Distance to central hospital Arrows pointing horizontally to the right (left) indicate that the two variables are in (anti-) phase. Figure 5 As in Figure 4, but for the cross wavelet power spectrum for (A) dengue-rainfall and (B) dengue-minimum temperature. Arrows pointing horizontally to the right (left) indicate that the two variables are in (anti-) phase. right (left) indicate that the two variables are in (anti-) phase. When the signals of two time series are in phase, their maximum amplitudes occur simultaneously. skewed, with 35% of neighborhoods (n = 89) reporting zero cases (Additional file 2: Figure S1). Dengue cases during the epidemic were significantly clustered (Moran’s I = 0.03, p <0.001). Findings from the LISA analysis indi- cated that there were significant dengue hotspots (n = 15 high-high neighborhoods) in west-central Machala, and a smaller number of significant outliers (n = 2 high-low neighborhoods, n = 5 low-high neighborhoods) (p <0.05, Figure 7b). Results Spatial analyses and social-ecological risk factors Spatial analyses and social-ecological risk factors Average neighborhood dengue incidence in 2010 was 76.7 ± 14.3 (95% CI) per 10,000 population (range: 0 to 775.8) (Figure 7A). The distribution was heavily left Figure 6 As in Figure 5 but for wavelet coherence spectrum for (A) dengue-rainfall and (B) dengue-minimum temperature. Figure 6 As in Figure 5 but for wavelet coherence spectrum for (A) dengue-rainfall and (B) dengue-minimum temperature. Figure 6 As in Figure 5 but for wavelet coherence spectrum for (A) dengue-rainfall and (B) dengue-minimum temperature. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 9 of 16 Page 9 of 16 Figure 7 The spatial distribution of dengue transmission in Machala, Ecuador in 2010: (A) Dengue incidence (cases per 10,000 population in 2010) per neighborhood, and (B) significant hot spots (high-high) and outliers (high-low and low-high) identified through LISA analysis (p ≤0.05). Figure 7 The spatial distribution of dengue transmission in Machala, Ecuador in 2010: (A) Dengue incidence (cases per 10,000 population in 2010) per neighborhood, and (B) significant hot spots (high-high) and outliers (high-low and low-high) identified through LISA analysis (p ≤0.05). The top ranked model to predict the presence of dengue, identified through the multimodel selection process, was a better fit than the global model that included all of the proposed social-ecological variables (global model AICc = 311, top ranked model AICc = 291.3, ΔAICc = 19.7). The top ranked model included the residual of the HCI regressed on no access to piped water, distance from the central hospital, and demographics of the heads of households (i.e., older age, female gender) (Table 2 Model A, Figure 8, Additional file 3: Figure S2). We also presented the model with the HCI and access to piped water as separate variables, to indicate the direction of the effect of each variable (Table 2 Model B). Neighborhoods were more likely to report dengue if they had poor housing condition and had greater access to piped water inside the home. When we compared neighbor- hoods with similar housing conditions, neighborhoods were more likely to report dengue cases if they had greater access to piped water inside the home (Figure 3). Inclusion of the residual variable in the model (Table 2 Model A) reduced multicollinearity, as indicated by the low VIFs. Stewart-Ibarra et al. Results BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 10 of 16 Page 10 of 16 Figure 8 Social-ecological parameters from the 2010 census that were included in the top logistic regression model to predict the presence of dengue, clockwise from top left: the average age of the head of the household, the housing condition index, distance to the central hospital, the percent of households with a female head of household, the percent of households with no piped water inside the home. Figure 8 Social-ecological parameters from the 2010 census that were included in the top logistic regression model to predict the presence of dengue, clockwise from top left: the average age of the head of the household, the housing condition index, distance to the central hospital, the percent of households with a female head of household, the percent of households with no piped water inside the home. Figure 8 Social-ecological parameters from the 2010 census that were included in the top logistic regression model to predict the presence of dengue, clockwise from top left: the average age of the head of the household, the housing condition index, distance to the central hospital, the percent of households with a female head of household, the percent of households with no piped water inside the home. Figure 8 Social-ecological parameters from the 2010 census that were included in the top logistic regression model to predict the presence of dengue, clockwise from top left: the average age of the head of the household, the housing condition index, distance to the central hospital, the percent of households with a female head of household, the percent of households with no piped water inside the home. Figure 8 Social-ecological parameters from the 2010 census that were included in the top logistic regression model to predict the presence of dengue, clockwise from top left: the average age of the head of the household, the housing condition index, distance to the central hospital, the percent of households with a female head of household, the percent of households with no piped water inside the home. Page 11 of 16 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 11 of 16 Multiple best-fit models were within the predetermined threshold criteria of ΔAICc ≤2 of the top model and weights greater than 1.5% (Additional file 4: Table S2). Discussion D i Dengue is the most important mosquito-borne viral disease globally, and has increased in incidence and distri- bution despite ongoing vector control interventions during the last three decades [1-3]. To date, we have a limited understanding of the spatiotemporal dynamics of dengue transmission, particularly at the local scale, due to the complex, non-stationary relationships among dengue infection, climate, vector, and virus strain dynamics [41,52-54]; and the geographic and temporal variation in the social-ecological conditions that influence risk [18-21]. More robust analysis tools, such as wavelet analyses and multimodel inference, and the increasing availability of geospatial epidemiological, climate, and social-ecological data have increased our ability to explore these dynamics. Studies such as this provide critical information to im- prove disease surveillance and to develop an EWS and other evidence-based interventions. The rainfall and minimum temperature spectra in Figure 4C,D demonstrated a strong annual signal (1-year periodic band), in agreement with the annual dengue cycle in the region. There was no evidence of relevant changes in variability for minimum temperature; the correspond- ing signal in the power spectrum is continuous around the annual band. In contrast, there were fluctuations in the ~1-year band for precipitation, likely associated with periods of low precipitation. Beginning around 2006, both climate variables demonstrated significant power around the 2-year band, a feature that is most notice- able in the rainfall data, particularly in recent years. The cross-wavelet power spectra for dengue and rainfall (Figure 5A), and dengue and minimum temperature (Figure 5B) showed regions in the time-frequency space with high common power in the 1-year and 2-year bands, suggesting a relationship between climate and dengue incidence at both time scales. The corresponding wavelet coherence spectra, however, indicated that the dengue and rainfall co-vary mostly in the 2-year band (Figure 6A), while dengue and minimum temperature co-vary mostly in the 1-year band (Figure 6B). This suggests that temperature and rainfall have well-differentiated roles in dengue trans- mission. The directions of the arrows in the plots indicate a slow change of phase in the co-variability of dengue and rainfall in the 2-year band, approaching in-phase behavior In this study, we found that neighborhoods with certain social-ecological conditions were more likely to have cases of dengue during the largest outbreak to date in El Oro Province. Dengue cases were clustered in neighborhood- level transmission hotspots near the city center during the epidemic. Results In addition to the parameters included in the top model, the following variables were included in competing best-fit models: Breteau Index, population density, households with people who emigrate for work, and households without access to paved roads. in late 2009, and we observed synchronized co-variability for the 1-year band in 2009 and at the start of 2010. These results highlight the distinct roles of these climate variables in dengue transmission at different temporal scales, and the importance of the phase and timing of climate variables with respect to dengue transmission. We found that the 2010 epidemic episode could be characterized by a combination of annual and bi-annual signals in dengue transmission and climate variables. The outbreak was characterized by a combination of in-phase variability of above normal minimum temperatures, and quasi-in-phase above normal rainfall episodes associated with the late 2009 to early 2010 moderate El Niño event (see arrows pointing right along the 1-year band at the bottom of Figures 5 and 6). The times series (1995-2010) of monthly anomalies in dengue cases from El Oro province, ENSO, temperature and rainfall have been previously described (See Figure 2 in Stewart Ibarra & Lowe 2013) [26]. This analysis demonstrated that the observed effect (quasi-simultaneity in the variability of dengue, temperature and rainfall) was present in early 2010, but not in any other year of the period under study (Figures 5 and 6). Temporal climate analyses We found that multiple temporal scales were involved in local dengue transmission dynamics, as shown in the wavelet power spectrum for dengue incidence (Figure 4B). In wavelet analyses, strong significant signals at a certain frequency are associated with persistent (quasi) periodic cycles in the time series (e.g., a 1-year band indicates pres- ence of annual cycles). There was a strong and significant signal for the ~2-year periodic band for dengue incidence. There was also a significant signal for the ~1-year periodic band, although it was less frequent (e.g., 2003, 2006, and 2011). Signals around and above the 4-year periodic band were not considered, as they fell inside of the COI (Figure 4B). These results suggest that dengue period- icity in this locality is not only annual (~1 year), but that there is also an important biannual cycle (~2 year), that may reflect typical time scales of extrinsic (e.g., climate) and intrinsic (e.g., immunologic) processes involved in the occurrence of dengue for this region. Discussion D i Previous studies that used spatial clustering statis- tics also found evidence of significant clustering of den- gue transmission across the urban landscape [18,56-58]. A previous study in Guayaquil, Ecuador, identified neighborhood-level dengue hot and coldspots, and found that the location of hotspots shifted over the 5-year period, highlighting the spatially dynamic nature of den- gue risk and the importance of multiyear studies [39]. Longitudinal field studies in Thailand found evidence of fine-scale spatial and temporal clustering of dengue virus serotypes and transmission at the school and household levels [59,60]. Focal transmission patterns are likely associated with the limited flight range of the Ae. aegypti mosquito. Recent studies in Peru demonstrated the im- portance of human movement patterns in determining spatial dengue transmission dynamics within an urban area [61,62]. At a regional scale, dengue outbreaks are likely influenced by human movement north and south along the Ecuador-Peru border. Future studies should continue to investigate the regional effects of cross-border movement of people and goods, and the local effects of intra-urban movement between work, school, and home to better understand the spatial dynamics of dengue transmission. gy g Neighborhoods were also at greater risk of dengue if they were closer to the central hospital, reflecting either spatially biased reporting and/or a true increase in transmission near the city center. This variable was also significant in all of the top models (Additional file 4: Table S2). Given the small size of the city of Machala (~5 km across) and easy access to low-cost public transportation, travel time to the hospital was not likely to be a limiting factor. However, people from lower income communities may be less likely to seek medical care due to the cost of medicine and the high cost of missing work, leading to underreporting from the urban periphery. It is also possible that people residing near the city center in Machala were at greater risk because they may have been less willing to cooperate with vector control technicians (E. Beltran, pers. comm.), due in part to the misconception that dengue is a problem of poor communities at the urban periphery [55]. Households in these areas may also be at greater risk because they store water as a secondary water source, as described above. These findings highlight the complexity of the cultural and behavioral factors influencing dengue risk and the importance of local-level studies that consider the social context. Discussion D i Risk factors included poor housing condition, greater access to piped water inside the home, less distance to the central hospital, and demographics of the heads of households (i.e., older age, female gender). In analyses of 10 years of weekly epidemiological and climate data, we found that dengue, rainfall and minimum temperature co-varied and had common power at 1-year Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 12 of 16 Page 12 of 16 access to piped water inside the home (likely associated with older, established communities with access to urban infrastructure). This apparent paradoxical relationship suggests that household water storage behaviors played an important role in the 2010 dengue outbreak. In our experience low-income households in Machala with access to piped water tend to store water in containers in the patio as a secondary water source, since water supply interruptions are common. These secondary water con- tainers are often uncovered, and the containers become ideal Ae. aegypti larval habitat during the rainy season. In contrast, low-income households without access to piped water are likely to store water in containers as their primary water source (e.g., 55 gallon drums), frequently filling and emptying the containers and thus preventing Ae. aegypti from developing into adult mosquitoes. and 2-year cycles, with quasi-synchronized higher than average rainfall and minimum temperatures likely contrib- uting to the 2010 dengue outbreak. This study contributes to ongoing efforts by INAMHI and the MSP of Ecuador to develop a dengue prediction model and early warning system. Findings from this study will inform the development of dengue vulnerability maps and climate- driven dengue seasonal forecasts that provide the MSP with information to target high-risk regions and seasons, allowing for more efficient use of scarce resources [9]. Spatial dynamics and social-ecological risk factors In Machala, a relatively small and heterogeneous city, there was evidence of unequal exposure or unequal reporting of dengue. During the epidemic, dengue trans- mission was focused in hotspots in the west-central urban sector, a middle- to low-income residential area with mod- erate access to urban infrastructure. Although people had access to basic services, our previous studies suggest that dengue control in these communities may be limited by the cost of household vector control, lack of social cohesion, and limited engagement with local institutions [55]. Discussion D i Our findings are consistent with a previous longitudinal field study of household risk factors for Ae. aegypti in Machala, where it was found that poor housing condition and access to piped water inside the home were positively associated with the presence of Ae. aegypti pupae [22]. This prior study found that Ae. aegypti were more abundant in the central urban area that had better access to infrastruc- ture than in the urban periphery [22]. Interestingly, the same risk factors emerged in the study presented here and the prior field study despite differences in rainfall (i.e., the field study was conducted one year after the epidemic, during a drier than average year) and differences in spatial scale (i.e., household- versus neighborhood level). These findings indicate that high-risk households could be identi- fied and targeted using a combination of census data and We found that the combination of HCI and access to piped water was the most important risk factor for dengue transmission, as indicated by the magnitude of the best-fit model parameter estimate (Table 2 Model A); this param- eter was also a significant variable in all other top models (Additional file 4: Table S2). Neighborhoods were more likely to report dengue if they had poor housing condi- tions (likely associated with lower income) and greater Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 13 of 16 Page 13 of 16 in this analysis suggests the potential to monitor the climate in this region to identify future time periods with synchronous climate conditions similar to 2010, that may increase the risk of a dengue outbreak. a locally adapted rapid survey of housing conditions, similar to the Premise Condition Index, an aggregate index measuring house condition, patio condition, and patio shade, which has been validated in other countries [63,64]. The HCI and the combined HCI-water access variables developed in this study should be explored and validated as dengue predictors in future studies in this region. Our results indicate that the 2-year band in precipitation is an important component in the co-variability of dengue incidence for the period under study, although its role in the 2010 dengue epidemic requires further investigation. This periodic band is not unique to Machala. Discussion D i Two to three-year cycles of dengue transmission have been reported in other parts of the world [41,53], particularly in years associated with El Niño events. The statistically significant 2-year band is present in the dengue power spectrum for the entire time series. This is not true for rainfall or minimum temperature, whose variability in the region is strongly associated with El Niño-Southern Oscillation (ENSO). This suggests that although ENSO has a strong influence in the occurrence of dengue epi- demics in coastal Ecuador, other variables (e.g., immunity) are also involved in the process and/or that there is a persistent mechanism for the climate’s biannual contribu- tion in the dengue spectrum. It is interesting to note that similar 2-year cycles have been reported for dengue and malaria in mountain locations in Peru [72], but not along the Peruvian Coast or Amazon [73]. We hypothesize that the biannual signal found in Peru and Machala is related to an additional climate mode present over the Andes in this region [28] in addition to ENSO. Machala may be uniquely situated to capture climate signals from ENSO and the so-called [28] Andean mode, given its proximity to the Andean foothills and the strong coupled climate- ocean system (i.e., teleconnections) present in the region. Ae. aegpyti juvenile indices were included in two of the top seven best-fit models to predict the presence of dengue in neighborhoods (Additional file 4: Table S2). A previous study in El Oro Province found that Ae. aegypti indices (House Index) were positively associated with dengue outbreaks at the province level [26]. Although pupal or adult indices are considered better predictors of dengue risk than larval indices [65], our findings suggest that larval indices may have some predictive power in this region. In Ecuador, entomological surveillance is limited to larval indices, and neighborhoods are rarely sampled in consecutive periods in a given year due to limited re- sources. These findings highlight the need for additional studies of the vector-dengue dynamics in this region and local evaluations of the robustness of vector abundance measures in order to strengthen cost-effective entomo- logical surveillance systems. Climate and dengue periodicity The wavelet analysis in this study provided a nuanced understanding of the relationships among local dengue transmission and climate variables at multiple temporal scales. The analysis of 10 years of weekly epidemiological and climate data from Machala provided evidence of significant 1-year and 2-year cycles in dengue, rainfall and minimum temperature. The 1-year cycles of minimum temperature and rainfall likely contributed to the annual dengue cycles observed in the power spectrum. This find- ing was expected, as previous studies have documented significant annual dengue cycles in this region [26]. Inter- estingly, we also found evidence of 2-year cycles in the rainfall wavelet power spectrum that were likely associated with biannual cycles of dengue transmission, a pattern that was previously undocumented in Ecuador. References 1. Dick OB, Martín JLS, Montoya RH, del Diego J, Zambrano B, Dayan GH: The history of dengue outbreaks in the Americas. Am J Trop Med Hyg 2012, 87:584–593. 2. San Martín JL, Brathwaite O, Zambrano B, Solórzano JO, Bouckenooghe A, Dayan GH, Guzmán MG: The epidemiology of dengue in the Americas over the last three decades: a worrisome reality. Am J Trop Med Hyg 2010, 82:128–135. Conclusions The results of this study highlight the importance of incorp- orating climate and social-ecological information with georeferenced and clinically validated epidemiological data in a dengue surveillance system. Investigators in Ecuador are exploring the development of web-based GIS for na- tional dengue surveillance using open-access software. GIS is an effective tool to integrate diverse data streams, such as dynamic, real-time epidemiological and climate data with static vulnerability maps generated from census data. Open access tools are especially important in resource-limited set- tings, and analysis packages targeted to dengue are becom- ing available [76]. Web-based GIS tools have been developed for global dengue surveillance, such as the CDC’s DengueMap, and for local dengue surveillance research projects [77,78]. National-level dengue GIS initiatives have been developed in countries such as Mexico [79], where Ministry of Health practitioners and software developers jointly designed the software platform. This collaborative ap- proach to integrate diverse data streams will ideally provide public health decision-makers with information to assess intervention programs, allocate resources more efficiently, and provide the foundation for an operational dengue EWS. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 14 of 16 the immune, nutritional, or health status of the population. We are currently collaborating with the MSP to improve dengue diagnostic infrastructure in the region and to reduce the time lag between epidemiological reporting and vector control interventions. Importantly, the MSP is undergoing a reorganization and decentralization process to merge the health and vector control divisions at the local level, with the goal of improving information flows and linking responses to evidence-based interventions. the immune, nutritional, or health status of the population. We are currently collaborating with the MSP to improve dengue diagnostic infrastructure in the region and to reduce the time lag between epidemiological reporting and vector control interventions. Importantly, the MSP is undergoing a reorganization and decentralization process to merge the health and vector control divisions at the local level, with the goal of improving information flows and linking responses to evidence-based interventions. Competing interests The authors declare that they have no competing interests Acknowledgements Many thanks to colleagues at the MSP and INAMHI for supporting ongoing climate–health initiatives in Ecuador. This work was funded by the National Secretary of Higher Education, Science, Technology and Innovation of Ecuador (SENESCYT), grant to INAMHI for the project “Surveillance and climate modeling to predict dengue in urban centers (Guayaquil, Huaquillas, Portovelo, Machala),” and the Global Emerging Infections Surveillance and Response System (GEIS), grant #P0001_14_UN. AGM used computational resources from the Latin American Observatory of Extreme Events (www. ole2.org) and Centro de Modelado Científico (CMC), Universidad del Zulia. The following institutes that participated in this study also form part of the Latin American Observatory partnership (http://ole2.org): International Research Institute for Climate and Society (IRI), Earth Institute, Columbia University, New York, NY, USA; and Centro de Modelado Científico (CMC), Universidad del Zulia, Maracaibo, Venezuela; Escuela Superior Politécnica del Litoral, Guayaquil, Ecuador; National Institute of Meteorology and Hydrology, Guayaquil, Ecuador. Author details 1 f 1Department of Microbiology and Immunology, Center for Global Health and Translational Science, State University of New York Upstate Medical University, 750 East Adams St, Syracuse, NY 13210, USA. 2International Research Institute for Climate and Society (IRI), Earth Institute, Columbia University, New York, NY, USA. 3Centro de Modelado Científico (CMC), Universidad del Zulia, Maracaibo, Venezuela. 4Department of Geography, Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA. 5School of Life Sciences, College of Agriculture, Engineering, and Science, University of KwaZulu-Natal, Durban, South Africa. 6The National Service for the Control of Vector-Borne Diseases, Ministry of Health, Machala, El Oro Province, Ecuador. 7Facultad de Medicina, Universidad Técnica de Machala, Machala, El Oro Province, Ecuador. 8Escuela Superior Politécnica del Litoral, Guayaquil, Ecuador. 9Division of Nutritional Sciences, Cornell University, Ithaca, NY, USA. 10Center for Geographic Analysis, Harvard University, Cambridge, MA, USA. 11National Institute of Meteorology and Hydrology, Guayaquil, Ecuador. Received: 12 June 2014 Accepted: 4 November 2014 Additional file 4: Table S2. 6. Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL, Drake JM, Brownstein JS, Hoen AG, Sankoh O, Myers MF, George DB, Jaenisch T, Wint GRW, Simmons CP, Scott TW, Farrar JJ, Hay SI: The global distribution and burden of dengue. Nature 2013, 496:504–507. 6. Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL, Drake JM, Brownstein JS, Hoen AG, Sankoh O, Myers MF, George DB, Jaenisch T, Wint GRW, Simmons CP, Scott TW, Farrar JJ, Hay SI: The global distribution and burden of dengue. Nature 2013, 496:504–507. 7. Messina JP, Brady OJ, Scott TW, Zou C, Pigott DM, Duda KA, Bhatt S, Katzelnick L, Howes RE, Battle KE, Simmons CP, Hay SI: Global spread of dengue virus types: mapping the 70 year history. Trends Microbiol 2014, 22:138–146. Limitations Although this study revealed patterns of climate and social- ecological conditions as important drivers of dengue transmission, this study has some limitations. It should be noted that non-climate factors that were undocumented in this study (e.g., population immunity, vector control interventions) are also key drivers of interannual variability in dengue [26,74,75] and most likely influenced the 2010 outbreak. The 1-year of spatially explicit epidemiological data constrained our ability to assess whether the social- ecological factors associated with the spatial distribution of dengue transmission were consistent in time. The 10-year time series of weekly dengue data was not available at the appropriate spatial scale for this analysis. With multiple years of data, we could evaluate whether dengue transmis- sion at the beginning of the dengue season or at the beginning of an epidemic is more likely to begin in neigh- borhoods with similar characteristics, to assess whether there are persistent high-risk, hotspot neighborhoods that trigger outbreaks. The analyses were also limited by a lack of laboratory confirmation for cases or information about Indeed, our analyses suggest that the 2010 dengue epidemic could be related to a timely coincidence of above normal minimum temperatures and above normal rainfall episodes during the moderate 2009 to 2010 El Niño event. Previous studies in this region have shown that Ae. aegypti abundance is associated with rainfall and minimum temperature [22]. In 2010, rainfall from February and March, the peak of dengue season, was almost double the long-term average (89% and 81% above average, respect- ively), likely increasing the availability of mosquito larval habitat. Temperature and temperature fluctuations influ- ence rates of mosquito development and virus replication [66-71]. The slow rate of climate phase change observed 9. Kuhn K, Campbell-Lendrum D, Haines A, Cox J: Using climate to predict infectious disease epidemics. Geneva: World Health Organization; 2005. Additional files Additional file 1: Table S1. Spanish dictionary of census variables evaluated in the multivariate model to predict the presence of dengue. Additional file 2: Figure S1. Histogram showing the density distribution of neighborhood dengue incidence in Machala, 2010 (n = 253). Additional file 3: Figure S2. Scatter matrix of parameters included in the top logistic regression model to predict the presence of dengue in neighborhoods in Machala in 2010. Additional file 4: Table S2. Top competing logistic regression models (ΔAICc < 2 or Weight > 1.5%) from multi-model selection to predict the presence (1) and absence (0) of dengue at the neighborhood level in Machala in 2010. Additional file 1: Table S1. Spanish dictionary of census variables evaluated in the multivariate model to predict the presence of dengue. Additional file 1: Table S1. Spanish dictionary of census variables evaluated in the multivariate model to predict the presence of dengue. 3. World Health Organization: Global Strategy for Dengue Prevention and Control, 2012–2020. Geneva: WHO; 2012 [http://www.who.int/ denguecontrol/9789241504034/en/] 3. World Health Organization: Global Strategy for Dengue Prevention and Control, 2012–2020. Geneva: WHO; 2012 [http://www.who.int/ denguecontrol/9789241504034/en/] Additional file 2: Figure S1. Histogram showing the density distribution of neighborhood dengue incidence in Machala, 2010 (n = 253). Additional file 2: Figure S1. Histogram showing the density distribution of neighborhood dengue incidence in Machala, 2010 (n = 253). 4. World Health Organization: Dengue: Guidelines for Diagnosis, Treatment, Prevention and Control. Geneva: WHO; 2009 [http://www.who.int/tdr/ publications/documents/dengue-diagnosis.pdf] Additional file 3: Figure S2. Scatter matrix of parameters included in the top logistic regression model to predict the presence of dengue in neighborhoods in Machala in 2010. Additional file 3: Figure S2. Scatter matrix of parameters included in the top logistic regression model to predict the presence of dengue in neighborhoods in Machala in 2010. 5. Gubler DJ: Epidemic dengue/dengue hemorrhagic fever as a public health, social and economic problem in the 21st century. Trends Microbiol 2002, 10:100–103. Additional file 4: Table S2. Top competing logistic regression models (ΔAICc < 2 or Weight > 1.5%) from multi-model selection to predict the presence (1) and absence (0) of dengue at the neighborhood level in Machala in 2010. 10. Thomson MC, Garcia-Herrera R, Beniston M (Eds.): Seasonal Forecasts, Climatic Change and Human Health: Health and Climate. Springer; 2008. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Anyamba A, Chretien J-P, Small J, Tucker CJ, Formenty PB, Richardson JH, Britch SC, Schnabel DC, Erickson RL, Linthicum KJ: Prediction of a Rift Valley fever outbreak. Proc Natl Acad Sci 2009, 106:955–959. 39. Castillo KC, Körbl B, Stewart A, Gonzalez JF, Ponce F: Application of spatial analysis to the examination of dengue fever in Guayaquil, Ecuador. Procedia Environ Sci 2011, 7:188–193. 18. Almeida AS, Medronho RA, Valencia LIO: Spatial analysis of dengue and the socioeconomic context of the city of Rio de Janeiro (southeastern Brazil). Rev Saúde Publica 2009, 43:666–673. 40. Calcagno V, de Mazancourt C: Glmulti: an R Package for easy automated model selection with (generalized) linear models. J Stat Softw 2010, 34:1–29. 41. Cazelles B, Chavez M, McMichael AJ, Hales S: Nonstationary influence of El Niño on the synchronous dengue epidemics in Thailand. PLoS Med 2005, 2:e106. 19. De Mattos Almeida MC, Caiaffa WT, Assunçao RM, Proietti FA: Spatial vulnerability to dengue in a Brazilian urban area during a 7-year surveillance. J Urban Health 2007, 84:334–345. 42. Cazelles B, Chavez M, de Magny GC, Guégan J-F, Hales S: Time-dependent spectral analysis of epidemiological time-series with wavelets. J R Soc Interface 2007, 4:625–636. 20. Mondini A, de Moraes Bronzoni RV, Nunes SHP, Neto FC, Massad E, Alonso WJ, Lázzaro ES, Ferraz AA, de Andrade Zanotto PM, Nogueira ML: Spatio-temporal tracking and phylodynamics of an urban dengue 3 outbreak in Sao Paulo, Brazil. PLoS Negl Trop Dis 2009, 3:e448. 43. Muñoz ÁG, Ruiz D, Ramírez P, León G, Quintana J, Bonilla A, Torres W, Pastén M, Sánchez O: Risk management at the Latin American Observatory. In Risk Manag - Curr Issues Chall. Edited by Banaitiene N. InTech; 2012:532–556. 21. da Teixeira GM, da Costa MCN, Ferreira LDA, Vasconcelos PFC, Cairncross S: Dynamics of dengue virus circulation: a silent epidemic in a complex urban area. Trop Med Int Health 2002, 7:757–762. 44. Muñoz ÁG, López P, Velásquez R, Monterrey L, León G, Ruiz F, Recalde C, Cadena J, Mejía R, Paredes M: An environmental watch system for the Andean Countries: El Observatorio Andino. Bull Am Meteorol Soc 2010, 91:1645–1652. 22. Stewart Ibarra AM, Ryan SJ, Beltrán E, Mejía R, Silva M, Muñoz Á: Dengue vector dynamics (Aedes aegypti) influenced by climate and social factors in Ecuador: implications for targeted control. PloS One 2013, 8:e78263. 45. Aguilar E, Sigró J, Brunet M: RClimdex con funcionalidades extras de control de calidad. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 32. Real J, Mosquera C: Detección del virus dengue en el ecuador. una visión epidemiológica. Periodo, 1988–2012. Guayaquil: Instituto Nacional de Higiene y Medicina Tropical del Ministerio de Salud Publica; 2012:2012. 11. Lowe R, Bailey TC, Stephenson DB, Jupp TE, Graham RJ, Barcellos C, Carvalho MS: The development of an early warning system for climate-sensitive disease risk with a focus on dengue epidemics in southeast Brazil. Stat Med 2013, 32:864–883. southeast Brazil. Stat Med 2013, 32:864–883. 33. Servicio Nacional para el Control de Enfermedades Transmitidas por Artropodos (SNEM), Ministerio de Salud Publica (MSP): Informe general por provincias, cantones/localidades. programa de control y vigilancia del Aedes aegypti. Indices Breteau y Casa por ciclos. Guayaquil. 12. Racloz V, Ramsey R, Tong S, Hu W: Surveillance of dengue fever virus: a review of epidemiological models and early warning systems. PLoS Negl Trop Dis 2012, 6:e1648. 34. Hartter J, Ryan SJ, MacKenzie CA, Parker JN, Strasser CA: Spatially explicit data: stewardship and ethical challenges in science. PLoS Biol 2013, 11:e1001634. 13. Yu H-L, Yang S-J, Yen H-J, Christakos G: A spatio-temporal climate-based model of early dengue fever warning in southern Taiwan. Stoch Environ Res Risk Assess 2011, 25:485–494. 35. Beyer HL: Geospatial Modelling Environment v0. 7.0. 2012. 14. Grover-Kopec E, Kawano M, Klaver RW, Blumenthal B, Ceccato P: An online operational rainfall-monitoring resource for epidemic malaria early warning systems in Africa. Malar J 2013, 4:5. 36. Beyer HL, Jenness J, Cushman SA: Components of spatial information management in wildlife ecology: software for statistical and modeling analysis. In Spatial Complexity, Informatics, and Wildlife Conservation. Japan: Springer; 2010:245–253. 15. Ruiz D, Connor SJ, Thomson MC: A multimodel framework in support of malaria surveillance and control. In Seasonal Forecasts, Climatic Change and Human Health: Health and Climate. Springer; 2008:101–125. 37. Anselin L: Local indicators of spatial association—LISA. Geogr Anal 1995, 27:93–115. p g 16. Thomson MC, Mason SJ, Phindela T, Connor SJ: Use of rainfall and sea surface temperature monitoring for malaria early warning in Botswana. Am J Trop Med Hyg 2005, 73:214–221. 38. de Oliveira MA, Ribeiro H, Castillo-Salgado C, de Oliveira MA, Ribeiro H, Castillo-Salgado C: Geospatial analysis applied to epidemiological studies of dengue: a systematic review. Rev Bras Epidemiol 2013, 16:907–917. 17. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Manual de Uso, version 1.0 2009:12. [http://cmc.org.ve/ descargas/Cursos/CRRH/Manual_rclimdex_extraQC.r.pdf] 23. Moore CG, Cline BL, Ruiz-Tiben E, Lee D, Romney-Joseph H, Rivera-Correa E: Aedes aegypti in Puerto Rico: environmental determinants of larval abundance and relation to dengue virus transmission. Am J Trop Med Hyg 1978, 27:1225–1231. 46. RHTestsV4 software [http://etccdi.pacificclimate.org/RHtest/RHtestsV4_ UserManual_20July2013.pdf] y 24. Barrera R, Amador M, MacKay AJ: Population dynamics of Aedes aegypti and dengue as influenced by weather and human behavior in San Juan, Puerto Rico. PLoS Negl Trop Dis 2011, 5:e1378. 47. Wang XL, Chen H, Wu Y, Feng Y, Pu Q: New techniques for detection and adjustment of shifts in daily precipitation data series. J Appl Meteorol Climatol 2010, 49:2416–2436. 25. Ministerio de Salud Publica de Ecuador Dirección Nacional de Vigilancia Epidemiológica: Anuario Epidemiológico, 1994-2014. Quito; 2014. [http:// www.salud.gob.ec/direccion-nacional-de-vigilancia-epidemiologica/] 48. Wang XL: Accounting for autocorrelation in detecting mean shifts in climate data series using the penalized maximal t or f test. J Appl Meteorol Climatol 2008, 47:2423–2444. 49. Wang XL: Penalized maximal f test for detecting undocumented mean shift without trend change. J Atmospheric Ocean Technol 2008, 25:368–384. 26. Stewart Ibarra AM, Lowe R: Climate and non-climate drivers of dengue epidemics in southern coastal ecuador. Am J Trop Med Hyg 2013, 88:971–981. 50. Torrence C, Compo GP: A practical guide to wavelet analysis. Bull Am Meteorol Soc 1998, 79:61–78. 27. Muñoz ÁG, Stewart-Ibarra AM, Ruiz Carrascal D: Desarrollo de modelos de pronóstico experimental: análisis socio ecológico de riesgo a dengue y análisis estadístico de patrones climáticos, entomológicos y epidemiológicos en modelos de dengue. In Technical Report for the CLIDEN Project. Quito: INAMHI-SENESCYT; 2013. 51. Grinsted A, Moore JC, Jevrejeva S: Application of the cross wavelet transform and wavelet coherence to geophysical time series. Nonlinear Process Geophys 2004, 11:561–566. 52. Lourenço J, Recker M: Natural, persistent oscillations in a spatial multi-strain disease system with application to dengue. PLoS Comput Biol 2013, 9:e1003308. 28. Recalde-Coronel GC, Barnston AG, Muñoz ÁG: Predictability of December-April rainfall in coastal and Andean Ecuador. J Appl Meteorol Climatol 2014, 53(6):1471–1493. 53. Reiner RC, Stoddard ST, Forshey BM, King AA, Ellis AM, Lloyd AL, Long KC, Rocha C, Vilcarromero S, Astete H, Bazan I, Lenhart A, Vazquez-Prokopec GM, Paz-Soldan VA, McCall PJ, Kitron U, Elder JP, Halsey ES, Morrison AC, Kochel TJ, Scott TW: Time-varying, serotype-specific force of infection of dengue virus. Proc Natl Acad Sci 2014, 111:E2694–E2702. 29. Authors’ contributions 8. Kyle JL, Harris E: Global spread and persistence of dengue. Annu Rev Microbiol 2008, 62:71–92. AMSI, AGM, MJBC, and RM conceived of the investigation. EBA, TO, GCRC and KR compiled the data used in analyses. AMSI, AGM, and SJR conducted analyses and drafted the manuscript. All co-authors, AMSI, AGM, MJBC, JLF, RM, TO, GCRC, KR, assisted with interpretation of the data, provided feedback for this manuscript, and read and approved the final manuscript. 9. Kuhn K, Campbell-Lendrum D, Haines A, Cox J: Using climate to predict infectious disease epidemics. Geneva: World Health Organization; 2005. 10. Thomson MC, Garcia-Herrera R, Beniston M (Eds.): Seasonal Forecasts, Climatic Change and Human Health: Health and Climate. Springer; 2008. Page 15 of 16 Page 15 of 16 Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Page 16 of 16 Page 16 of 16 78. Moreno-Sanchez R, Hayden M, Janes C, Anderson G: A web-based multimedia spatial information system to document Aedes aegypti breeding sites and dengue fever risk along the US–Mexico border. Health Place 2006, 12:715–727. 79. Hernández-Ávila JE, Rodríguez M-H, Santos-Luna R, Sánchez-Castañeda V, Román-Pérez S, Ríos-Salgado VH, Salas-Sarmiento JA: Nation-wide, web-based, geographic information system for the integrated surveillance and control of dengue fever in Mexico. PLoS ONE 2013, 8:e70231. doi:10.1186/s12879-014-0610-4 Cite this article as: Stewart-Ibarra et al.: Spatiotemporal clustering, climate periodicity, and social-ecological risk factors for dengue during an outbreak in Machala, Ecuador, in 2010. BMC Infectious Diseases 2014 14:610. 78. Moreno-Sanchez R, Hayden M, Janes C, Anderson G: A web-based multimedia spatial information system to document Aedes aegypti breeding sites and dengue fever risk along the US–Mexico border. Health Place 2006, 12:715–727. 56. Galli B, Chiaravalloti Neto F: Temporal-spatial risk model to identify areas at high-risk for occurrence of dengue fever. Rev Saúde Pública 2008, 42:656–663. 57. Vazquez-Prokopec GM, Kitron U, Montgomery B, Horne P, Ritchie SA: Quantifying the spatial dimension of dengue virus epidemic spread within a tropical urban environment. PLoS Negl Trop Dis 2010, 4:e920. 79. Hernández-Ávila JE, Rodríguez M-H, Santos-Luna R, Sánchez-Castañeda V, Román-Pérez S, Ríos-Salgado VH, Salas-Sarmiento JA: Nation-wide, web-based, geographic information system for the integrated surveillance and control of dengue fever in Mexico. PLoS ONE 2013, 8:e70231. 79. Hernández-Ávila JE, Rodríguez M-H, Santos-Luna R, Sánchez-Castañeda V, Román-Pérez S, Ríos-Salgado VH, Salas-Sarmiento JA: Nation-wide, web-based, geographic information system for the integrated surveillance and control of dengue fever in Mexico. PLoS ONE 2013, 8:e70231. 58. Hu W, Clements A, Williams G, Tong S: Spatial analysis of notified dengue fever infections. Epidemiol Infect 2011, 139:391–399. 59. Endy TP, Nisalak A, Chunsuttiwat S, Libraty DH, Green S, Rothman AL, Vaughn DW, Ennis FA: Spatial and temporal circulation of dengue virus serotypes: a prospective study of primary school children in Kamphaeng Phet, Thailand. Am J Epidemiol 2002, 156:52–59. doi:10.1186/s12879-014-0610-4 Cite this article as: Stewart-Ibarra et al.: Spatiotemporal clustering, climate periodicity, and social-ecological risk factors for dengue during an outbreak in Machala, Ecuador, in 2010. BMC Infectious Diseases 2014 14:610. 60. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Yoon I-K, Getis A, Aldstadt J, Rothman AL, Tannitisupawong D, Koenraadt CJM, Fansiri T, Jones JW, Morrison AC, Jarman RG, Nisalak A, Mammen MP Jr, Thammapalo S, Srikiatkhachorn A, Green S, Libraty DH, Gibbons RV, Endy T, Pimgate C, Scott TW: Fine scale spatiotemporal clustering of dengue virus transmission in children and Aedes aegypti in rural Thai villages. PLoS Negl Trop Dis 2012, 6:e1730. 61. Stoddard ST, Morrison AC, Vazquez-Prokopec GM, Soldan VP, Kochel TJ, Kitron U, Elder JP, Scott TW: The role of human movement in the transmission of vector-borne pathogens. PLoS Negl Trop Dis 2009, 3:e481. 62. Vazquez-Prokopec GM, Stoddard ST, Paz-Soldan V, Morrison AC, Elder JP, Kochel TJ, Scott TW, Kitron U: Usefulness of commercially available GPS data-loggers for tracking human movement and exposure to dengue virus. Int J Health Geogr 2009, 8:68. 63. Andrighetti MTM, Galvani KC, da Gracca Macoris ML: Evaluation of Premise Condition Index in the context of Aedes aegypti control in Marília, São Paulo. Brazil. Dengue Bull 2009, 33:167. 64. Tun-Lin W, Kay BH, Barnes A: The Premise Condition Index: a tool for streamlining surveys of Aedes aegypti. Am J Trop Med Hyg 1995, 53:591–594. 65. Focks DA: A review of entomological sampling methods and indicators for dengue vectors. Geneva: World Health Organization; 2003. 66. Pant CP, Yasuno M: Field studies on the gonotrophic cycle of Aedes aegypti in Bangkok, Thailand. J Med Entomol 1973, 10:219–223. 67. Watts DM, Burke DS, Harrison BA, Whitmire RE, Nisalak A: Effect of temperature on the vector efficiency of Aedes aegypti for dengue 2 virus. Am J Trop Med Hyg 1986, 36:143–152. 68. Rueda LM, Patel KJ, Axtell RC, Stinner RE: Temperature-dependent development and survival rates of Culex quinquefasciatus and Aedes aegypti (Diptera: Culicidae). J Med Entomol 1990, 27:892–898. 69. Bar-Zeev M: The effect of temperature on the growth rate and survival of the immature stages of Aedes aegypti (L.). Bull Entomol Res 1958, 49:157–163. 70. Thu HM, Aye KM, Thein S: The effect of temperature and humidity on dengue virus propagation in Aedes aegypti mosquitos. Southeast Asian J Trop Med Public Health 1998, 29:280–284. 71. Lambrechts L, Paaijmans KP, Fansiri T, Carrington LB, Kramer LD, Thomas MB, Scott TW: Impact of daily temperature fluctuations on dengue virus transmission by Aedes aegypti. Proc Natl Acad Sci 2011, 108:7460–7465. 72. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Instituto Nacional de Estadística y Censos (INEC): Censo de Poblacion Y Vivienda. Quito: 2010 [http://www.ecuadorencifras.gob.ec/] 30. Arunachalam N, Tana S, Espino F, Kittayapong P, Abeyewickrem W, Wai KT, Tyagi BK, Kroeger A, Sommerfeld J, Petzold M: Eco-bio-social determinants of dengue vector breeding: a multicountry study in urban and periurban Asia. Bull World Health Organ 2010, 88:173–184. 54. Cummings DAT, Irizarry RA, Huang NE, Endy TP, Nisalak A, Ungchusak K, Burke DS: Travelling waves in the occurrence of dengue haemorrhagic fever in Thailand. Nature 2004, 427:344–347. 31. Quintero J, Brochero H, Manrique-Saide P, Barrera-Pérez M, Basso C, Romero S, Caprara A, Cunha JCDL, Ayala EB, Mitchell-Foster K, Kroeger A, Sommerfeld J, Petzold M: Ecological, biological and social dimensions of dengue vector breeding in five urban settings of Latin America: a multi-country study. BMC Infect Dis 2014, 14:38. 55. Stewart Ibarra AM, Luzadis VA, Borbor Cordova MJ, Silva M, Ordoñez T, Beltrán Ayala E, Ryan SJ: A social-ecological analysis of community perceptions of dengue fever and Aedes aegypti in Machala, Ecuador. BMC Public Health 2014, 14:1135. Stewart-Ibarra et al. BMC Infectious Diseases 2014, 14:610 http://www.biomedcentral.com/1471-2334/14/610 Chowell G, Cazelles B, Broutin H, Munayco CV: The influence of geographic and climate factors on the timing of dengue epidemics in Perú, 1994-2008. BMC Infect Dis 2011, 11:164. 73. Chowell G, Munayco CV, Escalante AA, McKenzie FE: The spatial and temporal patterns of falciparum and vivax malaria in Perú: 1994-2006. Malar J 2009, 8:142. 73. Chowell G, Munayco CV, Escalante AA, McKenzie FE: The spatial and temporal patterns of falciparum and vivax malaria in Perú: 1994-2006. Malar J 2009, 8:142. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color ﬁgure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and take full advantage of: 74. Hay SI, Myers MF, Burke DS, Vaughn DW, Endy T, Ananda N, Shanks GD, Snow RW, Rogers DJ: Etiology of interepidemic periods of mosquito- borne disease. Proc Natl Acad Sci 2000, 97:9335–9339. • Convenient online submission 75. Wearing HJ, Rohani P: Ecological and immunological determinants of dengue epidemics. Proc Natl Acad Sci 2006, 103:11802–11807. 76. Delmelle EM, Zhu H, Tang W, Casas I: A web-based geospatial toolkit for the monitoring of dengue fever. Appl Geogr 2014, 52:144–152. 77. Chang AY, Parrales ME, Jimenez J, Sobieszczyk ME, Hammer SM, Copenhaver DJ, Kulkarni RP: Combining Google Earth and GIS mapping technologies in a dengue surveillance system for developing countries. Int J Health Geogr 2009, 8:49.
https://openalex.org/W2893702958	https://hal.science/hal-02356302/document	English	null	New Adaptive Prognostics Approach Based on Hybrid Feature Selection with Application to Point Machine Monitoring	Proceedings of the Annual Conference of the Prognostics and Health Management Society	2,018	cc-by	6,011	A new adaptive prognostics approach based on hybrid feature selection with application to point machine monitoring Vepa Atamuradov, Kamal Medjaher, Pierre Dersin, Noureddine Zerhouni, Fatih Camci Fatih Camci To cite this version: Vepa Atamuradov, Kamal Medjaher, Pierre Dersin, Noureddine Zerhouni, Fatih Camci. A new adap- tive prognostics approach based on hybrid feature selection with application to point machine mon- itoring. Annual Conference of the Prognostics and Health Management Society 2018, Sep 2018, Philadelphia, United States. pp.1-9. hal-02356302 HAL Id: hal-02356302 https://hal.science/hal-02356302v1 Submitted on 8 Nov 2019 L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. 4Advanced Micro Devices (AMD), USA fatih.camci@amd.com (Ashasi-Sorkhabi et al. 2017), breaking systems (Lee 2017), overhead contact lines (Brahimi et al. 2017), tilting actuators (Martin et al. 2017) and point machines (Böhm 2017). an author's http://oatao.univ-toulouse.fr/22003 an author's https://doi.org/10.1234/phmconf.2018.v10i1.500 Atamuradov, Vepa and Medjaher, Kamal and Dersin, Pierre and Zerhouni, Noureddine and Camci, Fatih A new adaptive prognostics approach based on hybrid feature selection with application to point machine monitoring. (2018) In: Annual Conference of the Prognostics and Health Management Society 2018, September 2018 (Philadelphia, United States). ABSTRACT This paper proposes a new adaptive prognostics approach consisting of hybrid feature selection and remaining-useful- life (RUL) estimation steps for railway point machines. In step-1, different time-domain based features are extracted and the best ones are selected by the hybrid feature selection method. Then, a degradation model is fitted to each of the selected features and the parameters are estimated. In step-2, the RUL of the component is predicted by using the proposed adaptive prognostics approach. The adaptive prognostics is based on the weighted likelihood combination of the estimated model parameters. The model parameters each of which estimated by curve fitting are used in the calculation of the likelihood probability weights. Then, an adaptive degradation model is built by using the weighted combination of the model parameter estimates and the component RUL is estimated. The proposed approach is validated on in-field point machine sliding-chair degradation and the results are discussed. Railway turnout system, which consists of sliding-chair plates, point machine, stock rails and locking systems are used to control the train turnouts at a distance (Eker et al. 2011). In literature, the point machine failure diagnostics (Atamuradov, Medjaher, Lamoureux, et al. 2017; García Márquez, Roberts, and Tobias 2010) and prognostics (Letot et al. 2015) have been studied extensively. However, there still remain many problems that need to be studied to increase the accuracy while minimizing the uncertainty in RUL prediction. One of the key steps in the development of robust and accurate fault prognostics is the selection of good prognostics features. In literature, feature evaluation and selection techniques are classified as a) inherent: which uses ranking metrics to filter out least interesting feature (e.g. trendability, monotonicity (J. B. Coble 2010) and seperability (Camci et al. 2013), etc.), b) consistent: which filters out the least correlated feature from the given feature population, and c) hybrid: which is the combination of inherent and/or consistent techniques (Lei et al. 2018). The authors in (Javed et al. 2015) proposed an inherent feature selection technique to increase the prognostics accuracy. In (Liao 2014) an inherent feature evaluation metric was integrated with genetic algorithm (GA) to discover good prognostics features for RUL prediction. The authors in (J. Coble and Hines 2009) developed a hybrid feature selection technique for prognostics based on the linearly weighted combination of the inherent and consistent techniques. In this proposed technique, the weights were Vepa Atamuradov et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 3FEMTO-ST Institute, UMR CNRS 6174 – UFC/ENSMM 15B av. des Montboucons, 25000 Besançon, France zerhouni@ens2m.fr 3FEMTO-ST Institute, UMR CNRS 6174 – UFC/ENSMM 15B av. des Montboucons, 25000 Besançon, France zerhouni@ens2m.fr 4Advanced Micro Devices (AMD), USA fatih.camci@amd.com 1Production Engineering Laboratory (LGP), INP-ENIT 47 Av. d’Azereix, 65000 Tarbes, France vepa.atamuradov@enit.fr kamal.medjaher@enit.fr 2ALSTOM Transport, 93400 Saint-Ouen, France pierre.dersin@transport.alstom.com benjamin.lamoureux@transport.alstom.com A New Adaptive Prognostics Approach based on Hybrid Feature Selection with Application to Point Machine Monitoring Vepa Atamuradov1, Kamal Medjaher1, Pierre Dersin2, Noureddine Zerhouni3 and Fatih Camci4 1Production Engineering Laboratory (LGP), INP-ENIT 47 Av. d’Azereix, 65000 Tarbes, France vepa.atamuradov@enit.fr kamal.medjaher@enit.fr 1Production Engineering Laboratory (LGP), INP-ENIT 47 Av. d’Azereix, 65000 Tarbes, France vepa.atamuradov@enit.fr kamal.medjaher@enit.fr 1Production Engineering Laboratory (LGP), INP-ENIT 47 Av. d’Azereix, 65000 Tarbes, France vepa.atamuradov@enit.fr kamal.medjaher@enit.fr 2. PROPOSED APPROACH The failure prognostics can be defined as the process of predicting the remaining time (RUL) at which a component will no longer perform a particular function. The authors in (Omer F Eker and Camci 2013; Omer Faruk Eker et al. 2011), developed a state duration based prognostics approach for point machine monitoring. The developed approach gave better RUL prediction results when compared with different prognostics tools. A data-driven failure prognostics model was proposed by (Letot et al. 2015) for point machine monitoring based on the power signals to predict the RUL. A similar data-driven prognostics approach based on a Bayesian parameter update was also proposed in (Ashasi- Sorkhabi et al. 2017), for train gearbox monitoring using vibration signals. A failure of train braking system was studied in (Lee 2017). The authors developed an air leakage detection and prediction approach based on a density-based clustering and logistic function. Since prognostics approaches deal with the prediction of the future component health states, the uncertainties in system parameters, nominal system model, degradation model, RUL prediction, and failure threshold should be well quantified in component health assessment (Atamuradov, Medjaher, Dersin, et al. 2017; Sankararaman and Goebel 2015). In this section, the hybrid feature selection and the adaptive prognostics steps will be explained in detail. The overall scheme of the proposed approach is depicted in Figure 1. Physical system dry-sliding chair failure mode modelling Figure 1. Step1: a) Raw measurements (!" is the #$% sample), b) extracted feature population (&',") and c) selected features (&),", where + < -), Step2: Adaptive prognostics. &.,. &.,"/. &.," &),. &),"/. &)," &.,. &.,"/. &.," &',. &',"/. &'," !. !"/. !" a) Raw measurements c) Selected features b) Extracted features Physical system dry-sliding chair failure mode modelling Step-1: hybrid feature selection le), Step-2: Prognostics Model definition RUL prediction dry-sliding chair failure mode modelling dry-sliding chair failure mode modelling Step-1: hybrid feature selection To fill the aforementioned gaps in the literature, this paper proposes a new adaptive prognostics approach based on hybrid feature selection for railway point machine sliding- chair degradation. The proposed approach is composed of two steps. In step-1, a hybrid feature selection method is developed. It is based on the affinity matrix and inherent feature evaluation. The affinity matrix is built to calculate the features’ relative importance weights (RIWs). The inherent feature evaluation deals with the calculation of monotonicity, correlation and robustness metrics of each feature. 1. INTRODUCTION An improvement of reliability, availability and passenger safety has been one of the main concerns for many years in the railway industry. Thus, it is important to develop predictive maintenance strategies to monitor railway infrastructures such as bogies (Ashasi-Sorkhabi et al. 2017; Atamuradov, Medjaher, Dersin, et al. 2017), gearbox Vepa Atamuradov et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. degradation model parameters are estimated and updated at each prediction time, iteratively, to estimate the RUL. optimized by utilizing the GA. However, despite the good optimization performance, feature selection techniques based on heuristic algorithms, e.g. GA, might be computationally expensive, particularly if there is a big amount of feature samples. Hence, the development of computationally efficient feature selection techniques is necessary to improve the failure prognostics accuracy. Then, the selected prognostics features can be used to train the prognostics tools for RUL prediction. The paper contains four sections. After the introduction, Section 2, describes the main steps of the proposed prognostics approach. Section 3 presents the experimental rig setup, data collection and the results of the proposed approach. Section 4 concludes the paper. 2.1. Hybrid feature selection where # is the length of the given features R and S. 2. PROPOSED APPROACH Then, a hybrid fitness function is constructed by combining the weighted (with RIWs) inherent feature metrics and the features are ranked accordingly. The features with the highest hybrid ranking value are selected and used in prognostics. In step-2, a degradation model is defined to each of the selected features and the model parameters are estimated. Then, a likelihood probability of each parameter is calculated by using the estimated model parameters of each feature. Afterward, an adaptive degradation model is constructed by using the weighted combination of the estimated model parameters with the likelihood probabilities. The adaptive Figure 1. Step1: a) Raw measurements (!" is the #$% sample), b) extracted feature population (&',") and c) selected features (&),", where + < -), Step2: Adaptive prognostics. le), 2.2. Adaptive prognostics approach T&&CUCDV'×' = W X C&R = S GCHD(&J, &K7 C&R Y S! ! (4)! (4)! (4)! In this study, a polynomial function with a degree of 3 is used to model the sliding chair degradation due to its good degradation representability. This model is given in equation (11). The steps of the adaptive prognostics approach are illustrated in Figure 2. where GCHD(&J, &K7 is the Euclidean distance between the features &J and &K from the feature population with a size of - . The relative importance weight :1 of the C$%(ZC = E … -7 feature is then derived by using the exponential membership function (5). (11) &(D7‰Š = ‹ × DŒ • Ž × DP • • × D • • (11) where &(D7‰Š is the model output at time D and ‹, Ž, •, • are the model parameters to be estimated. The model parameters of each of the selected features are estimated by using a curve fitting toolbox of MATLAB. Then, the estimated parameters are used to build an adaptive degradation model for RUL prediction. :1 = ?[R \OE × ] GCHD(&1,.7 ' 1Q. - ^ (5) (5) In step-2, inherent metrics such as monotonicity (-_U1 ), correlation (`_@@1) and robustness (a_b1) are calculated by using equations (7), (8) and (9). The features monotonicity metric is used to extract increasing or decreasing trend information. A similar work based on the Dempster-Shafer evidence theory to build a prior model for battery degradation was proposed in (He et al. 2011). The belief measure was assigned to each of the estimated parameters of the corresponding feature, by comparing the parameter confidence intervals. The basic idea was to assign a more belief weight to the parameter interval that includes the other parameter intervals to be used in parameter combination. However, the disadvantage of this evidence theory-based approach is that if there are no any such interval subsets, then it combines the parameters with the equal weights resulting in a simple weighted arithmetic mean combination. -_U1(&17 = cd e f fg3 hi "/. O e f fg3 ji "/. dk (6) (6) where -_U1 is the monotonicity value for the C$% feature (&1) with length of # . The absolute value of the difference between number of positive (e G G&C l X7 and negative (e G G&C < X7 derivatives gives the monotonicity value. 2.1. Hybrid feature selection where u_v is the covariance of C$% feature (&1) with the time vector F, and w is the standard deviation. The robustness metric stands for the features’ resistance to the measurement noise and it is calculated by decomposing the feature into trend (Hx__D>?Gy&1) and residual (@?H&1) components by using equations (8) and (9). First, time-domain based features such as skewness, root mean square (rms), kurtosis, mean, standard deviation (stdev), variance (var), crest factor (crfactor) and peak-to- peak (p2p) are extracted from the raw measurements. The features in different scales are normalized before selection by using equations (1) and (2). First, time-domain based features such as skewness, root mean square (rms), kurtosis, mean, standard deviation (stdev), variance (var), crest factor (crfactor) and peak-to- peak (p2p) are extracted from the raw measurements. The features in different scales are normalized before selection by using equations (1) and (2). @?H&1 = &1 O Hx__D>?Gy&1! (8) a_b1(&17 =c ] z9{(/\| }~•g3 g3 \|7 € • " k (9) @?H&1 = &1 O Hx__D>?Gy&1! (8) (8) 01 = (23/456(2377 (489(237/456(2377 ; :>?@?!1,$ = A3,B 489(A37* (1) 01 = (23/456(2377 (489(237/456(2377 ; :>?@?!1,$ = 456(A37 A3,B (2) (1) a_b1(&17 =c ] z9{(/\| }~•g3 g3 \|7 € • " k (9) (9) (2) where Hx__D>?Gy&1 is the smoothed feature, # is the length of C$% feature (&1). Then, the hybrid ranking function is built by using equation (10), which is the combination of the inherent metrics weighted by the corresponding relative importance weights. &1,$ is the C th feature data point at time index D(D = E, … , F7,F is the feature length and 01 is the Cth normalized feature. &1,$ is the C th feature data point at time index D(D = E, … , F7,F is the feature length and 01 is the Cth normalized feature. The hybrid feature selection is carried out in two-steps. In step-1, the affinity matrix (4) is built using the Euclidean distance (3). >a@‚UƒCU„1 = ] †:1 × -_U1, :1 × `_@@1, :1 × a‡b1ˆ ' 1Q. (10) (10) Finally, the >a@‚UƒCU„ vector is sorted in descending order starting from the highest relevant feature to the lowest relevant feature. Once the feature ranking step is completed, the top best (&)7 features are selected and used in prognostics. GCHDI&J, &KL = MN(&J,1 O &K,17P " 1Q. (3) (3) where # is the length of the given features R and S. 2.2. Adaptive prognostics approach The features correlation measures the linearity statistics between the failure propagation and time. where -_U1 is the monotonicity value for the C$% feature (&1) with length of # . The absolute value of the difference between number of positive (e G G&C l X7 and negative (e G G&C < X7 derivatives gives the monotonicity value. The features correlation measures the linearity statistics between the failure propagation and time. The difference between our approach and the work in (He et al. 2011) is that the calculation of parameter likelihood weights are not limited to the confidence interval length. `_@@1(&1, F17 = m nop(A3,q37 rg3rs3 t (7) (7) Figure 2. The adaptive prognostics approach steps. Figure 2. The adaptive prognostics approach steps. Instead, in our work, the estimated parameters get varying likelihood weights, as follows: Instead, in our work, the estimated parameters get varying likelihood weights, as follows: The adaptive degradation parameters are updated at each 5- time stamp, then the adaptive RUL is estimated. The RUL prediction accuracy (‘Tuu) is calculated by using equation (14) (Tobon-Mejia, Medjaher, and Zerhouni 2012). · If there is no parameter confidence interval that includes the other parameter intervals, then each parameter gets a varying likelihood weights proportional to their values. (14) ‘Tuu = . š ] O›œ•žaŸ ¡(b7 O aŸ ¢(b7ž aŸ ¡(b7 £ š ¤Q. (14) · If one of the parameter intervals includes the other parameter interval(s) or has wider interval length, then a more likelihood weight is assigned to this parameter(s). where ¥ is the number of data points used in RUL prediction. For the best prediction performance, the ‘Tuu produces 1, and 0 for the worst. Note that, our approach does not compare the parameter intervals, but only the likelihood of the estimated parameters. Note that, our approach does not compare the parameter intervals, but only the likelihood of the estimated parameters. If one of the estimated parameters is bigger than the others, then, theoretically, it should have the wider length of the confidence interval (i.e. the estimated parameter is the mean of the confidence interval estimates). 3. APPLICATION AND RESULTS If one of the estimated parameters is bigger than the others, then, theoretically, it should have the wider length of the confidence interval (i.e. the estimated parameter is the mean of the confidence interval estimates). This section explains the experimental rig setup and data collection procedures for point machine and presents the proposed approach results. Let’s assume that there are &) selected features and the ?‘),’ = “‹),’, Ž),’, •),’, •),’, ” = E, • • ,–— are the estimated initial parameters from each features’ degradation model. Then, the likelihood probability weight for the ?‘),’Q. (i.e. the 1st estimated parameter of model + ) is calculated by using equation (12). 3.1. System description and data collection In this study, we investigated the dry sliding-chair failure mode of the point machine, which is generated by an accelerated aging procedure (i.e. a manual contamination process such as soiling or scratching out the grease) of the sliding-chair plates. Figure 3 a) shows the in-field experimental test-rig setup, Figure 3 b) the turnout system and Figure 3 c) installed sensors for data acquisition. ˜),’Q. = ?‘),. ] ?‘),. ) 1Q. = ‹),. ] ‹),. ) 1Q. ;N ˜+,E = E ) 1Q. (12) (12) Sliding-chair plates are the metal assets of the turnout system that assist the point machine drive rods in moving the rail blades easily. The dry sliding-chair degradation data were generated on the real turnout system with 12 sliding-chair plates, in total. At first, all 12 plates were individually lubricated and the point machine was run 10 times in each movement to get the first healthy (fault-free) measurements. The same equation (12) is used to build the likelihood probability weights for the other Ž),., •),., •),. parameters. After the calculation of ˜),’ values, the adaptive degradation model parameters R., RP, RŒ, R™ can be estimated by the weighted arithmetic mean function, which is given in (13). R’ = N ˜),’ × ‹),’; *” = E, … ,– ) 1Q. (13) (13) Afterward, the accelerated aging procedure took place by contaminating the three farthest (10th, 11th, and 12th) plates from the point machine to get an initial faulty state. Figure 3. a) Experimental setup, b) railway turnout system and c) installed sensors. a) in-field Experimental setup b) Railway turnout system c) Installed sensors Force sensor Current sensor Voltage sensor Proximity sensor Sliding-chair plates Point machine Proximity sensor Figure 3. a) Experimental setup, b) railway turnout system and c) installed sensors. Voltage sensor Proximity sensor Figure 3. a) Experimental setup, b) railway turnout system and c) installed sensors. Proximity sensor The second faulty state was generated by contaminating the 9th plate after the first process. After each step of the contamination process, the point machine was run 10 times from normal-to-reverse (forth) and reverse-to-normal (back) positions to collect the measurements. The contamination on sliding-chair plates results in variation of performance measurement signals (e.g. force, current, voltage, etc.) due to the increasing friction force against the turnout driving rod force applied to move the blades. 3.1. System description and data collection The accelerated aging procedure was repeated until a final and complete sliding- chair failure state was reached. Note that no trains went through the turnout system during the data acquisition operation. It was temporarily reserved for experimentation purposes only. The force and current sensor measurements are the most commonly used data in the literature for point machine diagnostics and prognostics (García Márquez and Schmid 2007). In this study, the force measurements are used to validate the proposed approach. Table 3 shows the results of the model goodness-of-fit statistics (R2) and the estimated parameters for each of the features by using the curve fitting toolbox of MATLAB. The R-statistics indicates that the polynomial model is suitable to represent the degradation of the sliding-chair plate. Table 1. Affinity matrix and calculated relative importance weights (step-1). Table 1. Affinity matrix and calculated relative importance weights (step-1). F1 F2 F3 F4 F5 F6 F7 F8 F1 0 3.10 2.44 2.52 2.59 2.01 4.63 2.66 F2 3.10 0 1.11 0.97 1.02 1.81 2.79 0.97 F3 2.44 1.11 0 0.38 0.26 0.71 3.16 0.60 F4 2.52 0.97 0.38 0 0.37 0.96 2.83 0.40 F5 2.59 1.02 0.26 0.37 0 0.88 3.05 0.41 F6 2.01 1.81 0.71 0.96 0.88 0 3.56 1.13 F7 4.63 2.79 3.16 2.83 3.05 3.56 0 2.75 F8 2.66 0.97 0.60 0.40 0.41 1.13 2.75 0 w 0.08 0.22 0.33 0.34 0.34 0.24 0.05 0.32 Table 2. Hybrid feature ranking results (step-2). Monotonicity Correlation Robustness hRanking F5 0.48 0.85 0.75 0.71 F3 0.50 0.76 0.71 0.67 F8 0.46 0.82 0.71 0.65 F4 0.34 0.64 0.78 0.61 F6 0.48 0.85 0.69 0.50 F2 0.22 0.60 0.74 0.35 F1 0.02 0.45 0.66 0.09 F7 0.26 0.34 0.76 0.07 Table 1. Affinity matrix and calculated relative importance weights (step-1). F1 F2 F3 F4 F5 F6 F7 F8 F1 0 3.10 2.44 2.52 2.59 2.01 4.63 2.66 F2 3.10 0 1.11 0.97 1.02 1.81 2.79 0.97 F3 2.44 1.11 0 0.38 0.26 0.71 3.16 0.60 F4 2.52 0.97 0.38 0 0.37 0.96 2.83 0.40 F5 2.59 1.02 0.26 0.37 0 0.88 3.05 0.41 F6 2.01 1.81 0.71 0.96 0.88 0 3.56 1.13 F7 4.63 2.79 3.16 2.83 3.05 3.56 0 2.75 F8 2.66 0.97 0.60 0.40 0.41 1.13 2.75 0 w 0.08 0.22 0.33 0.34 0.34 0.24 0.05 0.32 Table 2. Hybrid feature ranking results (step-2). Table 2. Hybrid feature ranking results (step-2). 3.2. Results and Discussions Estimated parameters including the 95% confidence interval bounds with R-statistics. a) Figure 4. a) Raw measurements, b) extracted features and c normalized features. skewness rms kurtosis mean crfactor p2p Cycle Cycle Cycle Cycle Cycle Cycle Cycle Cycle stdev var b) a) c) Degradation Cycle p g confidence interval bounds with R-statistics. R2 Parameter Low bound Mean Upper bound F5 0.98 ‹. 9.882e-07 1.541e-06 2.093e-06 Ž. -0.000135 -5.012e-05 3.475e-05 •. -0.00466 -0.00096 0.00273 •. -0.0157 0.0276 0.071 F3 0.97 ‹P 1.119e-06 1.734e-06 2.348e-06 ŽP -0.000168 -7.363e-05 2.077e-05 •P -0.00494 -0.000833 0.00328 •P -0.0136 0.0347 0.0828 F8 0.97 ‹Œ 4.814e-07 9.391e-07 1.397e-06 ŽŒ -4.991e-05 2.039e-05 9.07e-05 •Œ -0.0059 -0.00286 0.00019 •Œ 0.0370 0.07295 0.1089 Before triggering the prognostics tool, a faulty state from the degradation data should be detected first. In this paper, the faulty state was obtained by projecting the F5-F3-F8 feature combination in the representation space (Soualhi, Medjaher, and Zerhouni 2015) as depicted in Figure 5. rms Cycle Before triggering the prognostics tool, a faulty state from the degradation data should be detected first. In this paper, the faulty state was obtained by projecting the F5-F3-F8 feature combination in the representation space (Soualhi, Medjaher, and Zerhouni 2015) as depicted in Figure 5. Before triggering the prognostics tool, a faulty state from the degradation data should be detected first. In this paper, the faulty state was obtained by projecting the F5-F3-F8 feature combination in the representation space (Soualhi, Medjaher, and Zerhouni 2015) as depicted in Figure 5. mean Cycle kurtosis mean State-3: Sever fault State-1: No fault State-2: Incipient fault Figure 5. State detection by representation space projection. State-3: Sever fault State-1: No fault State-2: Incipient fault Cycle Cycle c) Degradation Cycle Figure 5. State detection by representation space projection. The representation space of the feature combination allows identifying the health state transitions of the sliding-chair degradation. Since the training features F5, F3 and F8 have correlated degradation pattern, it was assumed that they have the same cycle number where an incipient fault occurs. After the detection of the incipient fault, which is at cycle 69, the feature degradation models (x., xP, xŒ) are trained and the initial parameters are estimated as shown in Table 4. Table 4. Initial parameter estimates after fault detection. Table 4. Initial parameter estimates after fault detection. 3.2. Results and Discussions y g ( p ) Monotonicity Correlation Robustness hRanking F5 0.48 0.85 0.75 0.71 F3 0.50 0.76 0.71 0.67 F8 0.46 0.82 0.71 0.65 F4 0.34 0.64 0.78 0.61 F6 0.48 0.85 0.69 0.50 F2 0.22 0.60 0.74 0.35 F1 0.02 0.45 0.66 0.09 F7 0.26 0.34 0.76 0.07 Figure 4 shows the extracted features and normalized features from the raw measurements (equations (1) and (2)). The hybrid feature selection step-1 results are given in Table 1 and step-2 results are given in Table 2. Table 2 presents the ranked features (F1{skw}, F2{krt}, F3{rmse}, F4{avg}, F5{stdev}, F6{Var}, F7{crst} and F8{p2p}) in descending order. Then, the first three features F5, F3 and F8 were selected as the best prognostics features and used in model training for the RUL prediction. Figure 4. a) Raw measurements, b) extracted features and c) normalized features. Table 3. Estimated parameters including the 95% confidence interval bounds with R-statistics. R2 Parameter Low bound Mean Upper bound F5 0.98 ‹. 9.882e-07 1.541e-06 2.093e-06 Ž. -0.000135 -5.012e-05 3.475e-05 •. -0.00466 -0.00096 0.00273 •. -0.0157 0.0276 0.071 F3 0.97 ‹P 1.119e-06 1.734e-06 2.348e-06 ŽP -0.000168 -7.363e-05 2.077e-05 •P -0.00494 -0.000833 0.00328 •P -0.0136 0.0347 0.0828 F8 0.97 ‹Œ 4.814e-07 9.391e-07 1.397e-06 ŽŒ -4.991e-05 2.039e-05 9.07e-05 •Œ -0.0059 -0.00286 0.00019 •Œ 0.0370 0.07295 0.1089 Before triggering the prognostics tool, a faulty state from the degradation data should be detected first. In this paper, the faulty state was obtained by projecting the F5-F3-F8 feature combination in the representation space (Soualhi, Medjaher, and Zerhouni 2015) as depicted in Figure 5. Figure 5. State detection by representation space projection. The representation space of the feature combination allows identifying the health state transitions of the sliding-chair degradation. Since the training features F5, F3 and F8 have correlated degradation pattern, it was assumed that they have the same cycle number where an incipient fault occurs. After the detection of the incipient fault, which is at cycle 69, the feature degradation models (x., xP, xŒ) are trained and the initial parameters are estimated as shown in Table 4. Table 4. Initial parameter estimates after fault detection. ‹ Ž • • 0¦(x.7 -4.258e-07 8.055e-05 -0.0024 0.0243 0§(xP7 2.688e-07 -6.901e-06 -0.00024 0.0174 0¨(xŒ7 -1.38e-06 0.0001895 -0.0055 0.0763 skewness rms kurtosis mean crfactor p2p Cycle Cycle Cycle Cycle Cycle Cycle Cycle Cycle stdev var b) a) c) Degradation Cycle State-3: Sever fault State-1: No fault State-2: Incipient fault Table 3. 3.2. Results and Discussions ‹ Ž • • 0¦(x.7 -4.258e-07 8.055e-05 -0.0024 0.0243 0§(xP7 2.688e-07 -6.901e-06 -0.00024 0.0174 0¨(xŒ7 -1.38e-06 0.0001895 -0.0055 0.0763 Figure 4. a) Raw measurements, b) extracted features and c) normalized features. By using the equation (12) the likelihood probability weights of the training model parameters were calculated and the results are given in Table 5. From the initial parameter estimates (Table 5), the combined parameters can be estimated by using equation (13). The combined parameters R., RP, RŒ, R™ for the adaptive degradation model (‚-) are given in Table 6. Figure 6 shows that the combined parameters and their confidence intervals (C.I.) are adapted to the change of the initial model parameter estimates and their C.Is. The parameter updating is iteratively repeated as new data points are available until the end-of-life (EoF) threshold value (see Figure 2). Table 5. Calculated likelihood probability weights. x. ˜.,’Q. xP ˜P,’Q. xŒ ˜Œ,’Q. ‹. 0.1355 ‹P 0.2177 ‹Œ 0.6468 Ž. 0.5102 ŽP 0.1890 ŽŒ 0.3008 •. 0.1316 •P 0.5170 •Œ 0.3515 •. 0.3400 •P 0.3281 •Œ 0.3319 Table 6. Combined adaptive degradation model parameters. R. RP RŒ R™ ‚- -1.4038e-06 1.6133e-04 -0.0044 0.0569 Figure 7 shows the RUL prediction results for the models x.(Ra‡©E7, xP(Ra‡©ª7, xŒ(Ra‡©¦7 and ‚-(‚a‡©7 , whereas Table 7 presents the RUL prediction accuracies (‘TuuxE, ‘Tuu«P, ‘Tuu«Œ, ‘Tuu¬'7. As can be seen from the given Table 7 , the proposed adaptive prognostics approach improved the RUL prediction accuracy, which proves the applicability in railway point machine monitoring. Figure 6. Estimated parameter confidence intervals (C.I.). Table 7. RUL prediction accuracy. ‘Tuu«. ‘Tuu«P ‘Tuu«Œ ‘Tuu¬' ‘Tuu 0.84 0.83 0.91 0.92 Figure 7. RUL prediction results for a) m1, b) m2, c) m3 and d) adaptive degradation model. the given Table 7 , the proposed adaptive prognostics approach improved the RUL prediction accuracy, which proves the applicability in railway point machine monitoring. By using the equation (12) the likelihood probability weights of the training model parameters were calculated and the results are given in Table 5. By using the equation (12) the likelihood probability weights of the training model parameters were calculated and the results are given in Table 5. From the initial parameter estimates (Table 5), the combined parameters can be estimated by using equation (13). The combined parameters R., RP, RŒ, R™ for the adaptive degradation model (‚-) are given in Table 6. 3.2. Results and Discussions Figure 6 shows that the combined parameters and their confidence intervals (C.I.) are adapted to the change of the initial model parameter estimates and their C.Is. The parameter updating is iteratively repeated as new data points are available until the end-of-life (EoF) threshold value (see Figure 2). Figure 6. Estimated parameter confidence intervals (C.I.). Table 5. Calculated likelihood probability weights. x. ˜.,’Q. xP ˜P,’Q. xŒ ˜Œ,’Q. ‹. 0.1355 ‹P 0.2177 ‹Œ 0.6468 Ž. 0.5102 ŽP 0.1890 ŽŒ 0.3008 •. 0.1316 •P 0.5170 •Œ 0.3515 •. 0.3400 •P 0.3281 •Œ 0.3319 Table 6. Combined adaptive degradation model parameters. R. RP RŒ R™ ‚- -1.4038e-06 1.6133e-04 -0.0044 0.0569 Figure 7 shows the RUL prediction results for the models x.(Ra‡©E7, xP(Ra‡©ª7, xŒ(Ra‡©¦7 and ‚-(‚a‡©7 , whereas Table 7 presents the RUL prediction accuracies (‘TuuxE, ‘Tuu«P, ‘Tuu«Œ, ‘Tuu¬'7. As can be seen from Table 5. Calculated likelihood probability weights. Figure 6. Estimated parameter confidence intervals (C.I.). Table 7. RUL prediction accuracy. ‘Tuu«. ‘Tuu«P ‘Tuu«Œ ‘Tuu¬' ‘Tuu 0.84 0.83 0.91 0.92 Table 7. RUL prediction accuracy. Table 7. RUL prediction accuracy. hows the RUL prediction results for the models , xP(Ra‡©ª7, xŒ(Ra‡©¦7 and ‚-(‚a‡©7 , able 7 presents the RUL prediction accuracies Tuu«P, ‘Tuu«Œ, ‘Tuu¬'7. As can be seen from Table 7. RUL prediction accuracy. ‘Tuu«. ‘Tuu«P ‘Tuu«Œ ‘Tuu¬' ‘Tuu 0.84 0.83 0.91 0.92 Figure 7. RUL prediction results for a) m1, b) m2, c) m3 and d) adaptive degradation model. Figure 7 shows the RUL prediction results for the models x.(Ra‡©E7, xP(Ra‡©ª7, xŒ(Ra‡©¦7 and ‚-(‚a‡©7 , whereas Table 7 presents the RUL prediction accuracies (‘TuuxE, ‘Tuu«P, ‘Tuu«Œ, ‘Tuu¬'7. As can be seen from Figure 7 shows the RUL prediction results for the models x.(Ra‡©E7, xP(Ra‡©ª7, xŒ(Ra‡©¦7 and ‚-(‚a‡©7 , whereas Table 7 presents the RUL prediction accuracies (‘TuuxE, ‘Tuu«P, ‘Tuu«Œ, ‘Tuu¬'7. As can be seen from Figure 7. RUL prediction results for a) m1, b) m2, c) m3 and d) adaptive degradation model. Brahimi, Mehdi, Kamal Medjaher, Mohammed Leouatni, and Noureddine Zerhouni. 2017. “Prognostics and Health Management for an Overhead Contact Line System - A Review.” International Journal of Prognostics and Health Management: 1–16. ACKNOWLEDGEMENT This research was supported by a grant from ENIT, Production Engineering Laboratory (LGP), funded by ALSTOM. Point machine and Li-ion degradation datasets were taken from the projects (# 108M275 and # 113M093), which were supported by The Scientific and Technological Research Council of Turkey (TUBITAK). Eker, Omer F, and Fatih Camci. 2013. “State Based Prognostics with State Duration Information.” Quality and Reliability Engineering International. Eker, Omer Faruk et al. 2011. “A Simple State-Based Prognostic Model for Railway Turnout Systems.” IEEE Transactions on Industrial Electronics 58(5): 1718–26. 4. CONCLUSION In this paper, a new adaptive prognostics approach based on a hybrid feature selection method was proposed for point machine sliding chair monitoring. A polynomial model was defined for each of the selected features and the model parameters were estimated. Then, the adaptive degradation model was built based on the likelihood probability weights calculated by using the initial model parameter estimates of the selected prognostics features. The model parameters were updated, iteratively, and the RULs were estimated. The results showed that the proposed prognostics approach improved the RUL prediction accuracy for the sliding-chair degradation. Camci, F., K. Medjaher, N. Zerhouni, and P. Nectoux. 2013. “Feature Evaluation for Effective Bearing Prognostics.” Quality and Reliability Engineering International 29(4): 477–86. http://doi.wiley.com/10.1002/qre.1396. Coble, Jamie Baalis. 2010. “Merging Data Sources to Predict Remaining Useful Life--an Automated Method to Identify Prognostic Parameters.” http://trace.tennessee.edu/utk_graddiss/683. As a future work, we plan to extend the proposed approach and to develop an adaptive system-level prognostics approach based on the extracted features from different components for condition monitoring and predictive maintenance. Coble, Jamie, and J. Wesley Hines. 2009. “Identifying Optimal Prognostic Parameters from Data: A Genetic Algorithms Approach.” Proceedings of the Annual Conference of the Prognostics and Health Management Society: 1–11. https://www.phmsociety.org/sites/phmsociety.org/file s/phm_submission/2009/phmc_09_69.pdf. REFERENCES García Márquez, Fausto Pedro, C Roberts, and Andrew Mark Tobias. 2010. “Railway Point Mechanisms: Condition Monitoring and Fault Detection.” Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit 224(1): 35–44. http://pif.sagepub.com/lookup/doi/10.1243/09544097J RRT289. Ashasi-Sorkhabi, Ali, Stanley Fong, Guru Prakash, and Sriram Narasimhan. 2017. “A Condition Based Maintenance Implementation for an Automated People Mover Gearbox.” International Journal of Prognostics and Health Management: 1–14. Atamuradov, Vepa, Kamal Medjaher, Benjamin Lamoureux, et al. 2017. “Fault Detection By Segment Evaluation Based On Inferential Statistics For Asset Monitoring.” In Annual Conference of the Prognostics and Health Management Society 2017, , 1–10. García Márquez, Fausto Pedro, and Felix Schmid. 2007. “A Digital Filter-Based Approach to the Remote Condition Monitoring of Railway Turnouts.” Reliability Engineering and System Safety 92(6): 830–40. Atamuradov, Vepa, Kamal Medjaher, Pierre Dersin, et al. 2017. “Prognostics and Health Management for Maintenance Practitioners - Review, Implementation and Tools Evaluation.” International Journal of Prognostics and Health Management 8(3): 1–31. He, Wei, Nicholas Williard, Michael Osterman, and Michael Pecht. 2011. “Prognostics of Lithium-Ion Batteries Based on Dempster-Shafer Theory and the Bayesian Monte Carlo Method.” Journal of Power Sources 196(23): 10314–21. http://dx.doi.org/10.1016/j.jpowsour.2011.08.040. Böhm, T. 2017. “Remaining Useful Life Prediction for Railway Switch Engines Using Classification Techniques.” International Journal of Prognostics and Health Management 8(Special Issue 7). https://www.scopus.com/inward/record.uri?eid=2- s2.0- 85037702039&partnerID=40&md5=45468c92ee66f1 8acca25e8d99c67764. Javed, Kamran, Rafael Gouriveau, Noureddine Zerhouni, and Patrick Nectoux. 2015. “Enabling Health Monitoring Approach Based on Vibration Data for Accurate Prognostics.” IEEE Transactions on Industrial Electronics 62(c): 647–56. http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm? arnumber=6824783. arnumber=6824783. Lee, Wan-jui. 2017. “Anomaly Detection and Severity Prediction of Air Leakage in Train Braking Pipes.” International Journal of Prognostics and Health Management. Lei, Yaguo et al. 2018. “Machinery Health Prognostics: A Systematic Review from Data Acquisition to RUL Prediction.” Mechanical Systems and Signal Processing 104(May): 799–834. https://doi.org/10.1016/j.ymssp.2017.11.016. Letot, C. et al. 2015. “A Data Driven Degradation-Based Model for the Maintenance of Turnouts: A Case Study.” IFAC-PapersOnLine 28(21): 958–63. http://dx.doi.org/10.1016/j.ifacol.2015.09.650. Liao, Linxia. 2014. “Discovering Prognostic Features Using Genetic Programming in Remaining Useful Life Prediction.” IEEE Transactions on Industrial Electronics 61(5): 2464–72. Martin, Andrea De, Andrea Dellacasa, Giovanni Jacazio, and Massimo Sorli. 2017. “Integrated Health Monitoring for the Actuation System of High-Speed Tilting Trains.” International Journal of Prognostics and Health Management: 1–12. Sankararaman, Shankar, and Kai Goebel. 2015. “Uncertainty in Prognostics and Health Management: An Overview.” International Journal of Prognostics and Health Management: 1–11. Soualhi, Abdenour, Kamal Medjaher, and Noureddine Zerhouni. 2015. “Bearing Health Monitoring Based on Hilbert – Huang Transform , Support Vector Machine , and Regression.” IEEE Transactions on Instrumentation and Measurement 64(1): 52–62. Tobon-Mejia, D. A., K. Medjaher, and N. Zerhouni. 2012. “CNC Machine Tools Wear Diagnostic and Prognostic by Using Dynamic Bayesian Networks.” Mechanical Systems and Signal Processing 28: 167–82. http://dx.doi.org/10.1016/j.ymssp.2011.10.018.

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.