ReadingTimeMachine/rtm-sgt-ocr-v1 · Datasets at Hugging Face

Tain grateful to Cilles Chabricr for careful reading of the manuscript aud many useful sugecstion.

I am grateful to Gilles Chabrier for careful reading of the manuscript and many useful suggestion.

The financial support for this work is provided by the Canada Researcli Chairs program aud a NSERC Discovery eraut.

The financial support for this work is provided by the Canada Research Chairs program and a NSERC Discovery grant.

Performing au iutegral over o in eq. (5)) we find

Performing an integral over $\phi$ in eq. \ref{eq:irr_flux}) )

Ly rieht)). where D?=(47|1)?Le?siu?0 aud su,ναι,

we find ), where $D^2=(x^2+1)^2-4x^2\sin^2\theta$ and $x_{in}=1/\sin\theta$.

For ΕΠ) obeviug (6)) equation (A2)) can be rewrittenas eq. (7))

For $F_d(R)$ obeying \ref{eq:vis_dissip}) ) equation \ref{eq:1D}) ) can be rewrittenas eq. \ref{eq:irr_flux_mod}) )

with

lifetime (~20 Myr). which ts 1n turn short compared to the global gas consumption timescale of several Gyr.

lifetime $\sim20\,$ Myr), which is in turn short compared to the global gas consumption timescale of several Gyr.

To establish whether the replenishment achievable in supershells is sufficient to make this mechanism feasible. we first convert the replenishment rates given in refdeltarho— into a replenished mass per SN event.

To establish whether the replenishment achievable in supershells is sufficient to make this mechanism feasible, we first convert the replenishment rates given in \\ref{deltarho} into a replenished mass per SN event.

A replenishment rate proportional to the SFR density is also proportional to the rate of supernova type Il (SNID))

A replenishment rate proportional to the SFR density is also proportional to the rate of supernova type II ).

Converting a proportionality to SFR density into one depending on the SNII rate. Paxjj1. depends on the assumed IMF.

Converting a proportionality to SFR density into one depending on the SNII rate, $\dot{\rho}_{\rm SNII}$, depends on the assumed IMF.

From Hopkins&Beacom(2006) psxiy=(0.00915/42..) for the SalA IMF.

From \citet{HB:06} $\dot{\rho}_{\rm SNII}=(0.00915/M_{\odot})\,\dot{\rho}_*$ for the SalA IMF.

The replenishment rate A(*)=1.65. becomesp. Att)=L?LOPsxuM..

The replenishment rate $K(t)=1.6\dot{\rho}_*$ becomes $K(t)=174.9\dot{\rho}_{\rm SNII}\,M_{\odot}$.

The other extreme choice of IMF consistent with the normalization of the SFH (Hopkins&Beacom2006) is that of Baldry&Glazebrook(2003.hereaftertheBG IMF)..

The other extreme choice of IMF consistent with the normalization of the SFH \citep{HB:06} is that of \citet[hereafter the BG IMF]{Bal:03}. .

For the BG IMF PSU=(0.00532/A/.)p..

For the BG IMF $\dot{\rho}_{\rm SNII}=(0.0132/M_{\odot})\,\dot{\rho}_*$.

The recycled fraction is R=0.56 (Hopkins&Beacom2006).. changing the consumption term in Equation (1)) ο. Πρ.

The recycled fraction is $R=0.56$ \citep{HB:06}, changing the consumption term in Equation \ref{theeqn}) ) to $-1.44\dot{\rho}_*$.

The corresponding replenishment rate is A(f)109.1505x11AL...

The corresponding replenishment rate is $K(t)=109.1\dot{\rho}_{\rm SNII}\,M_{\odot}$.

These extremes imply that. depending on the IMF. sufficient gas replenishment to maintain a constant HI mass density with redshift would be achieved if each SN event caused the recombination and cooling of =10180 of gas.

These extremes imply that, depending on the IMF, sufficient gas replenishment to maintain a constant HI mass density with redshift would be achieved if each SN event caused the recombination and cooling of $\approx 110-180\,M_{\odot}$ of gas.

These IMFs are the extrema given the SFH normalization.. limits. and most reasonable IMFs should result in masses within this range.

These IMFs are the extrema given the SFH normalization limits, and most reasonable IMFs should result in masses within this range.

Detailed measurements to confirm molecular gas formation within supershells are observationally challenging.

We use the limited data currently available to assess the replenishmentrates associated with supershells. and to establish whether at least one well-studied supershell achieves the required rate.

We use the limited data currently available to assess the replenishmentrates associated with supershells, and to establish whether at least one well-studied supershell achieves the required rate.

MeClure-Griffiths(2005) and Dawsonetal.(2008) have shown explicit cases of molecular clumps along the edges of supershells. suggestive of some degree of formation. with a significant amount of molecular material associated with the supershell walls.

\citet{McCG:05} and \citet{Daw:08} have shown explicit cases of molecular clumps along the edges of supershells, suggestive of some degree of formation, with a significant amount of molecular material associated with the supershell walls.

The supershell investigated by Dawsonetal.(2008) is associated with about 2«10°M.. of molecular gas. of which those authors estimate that 80% likely comes from a pre-existing giant molecular cloud.

The supershell investigated by \citet{Daw:08} is associated with about $2\times10^5\,M_{\odot}$ of molecular gas, of which those authors estimate that $80\%$ likely comes from a pre-existing giant molecular cloud.

Of the remaining xΕν10!M. of molecular gas it is difficult to determine how much is pre-existing and how much has been cooled and recombined by the expansion of the shell.

Of the remaining $\approx 4\times10^4\,M_{\odot}$ of molecular gas it is difficult to determine how much is pre-existing and how much has been cooled and recombined by the expansion of the shell.

We can use z|<101AF. as an upper limit to the replenishment rate.

We can use $\approx 4\times10^4\,M_{\odot}$ as an upper limit to the replenishment rate.

About 30 stars with stellar mass A.>7WM are required to form this supershell. including stars that may not.. yet have gone supernova.

About 30 stars with stellar mass $M_*> 7\,M_{\odot}$ are required to form this supershell, including stars that may not yet have gone supernova.

This gives =13002000ΑΓ, of molecular mass replenished per SN event. a limit comfortably encompassing the required rate.

This gives $\lesssim 1300 - 2000\,M_{\odot}$ of molecular mass replenished per SN event, a limit comfortably encompassing the required rate.

This upper limit could change significantly depending on the fraction of pre-existing molecular material and also on the fraction of stars that have not yet gone supernova.

Not all SNe lie within supershells. although Higdon&Lin-genfelter(2005) estimate that a minimum of65% of SNII should occur in superbubbles.increasing to =SO904 when the spatial and temporal correlations of stellar —clusters are considered.

Not all SNe lie within supershells, although \citet{HL:05} estimate that a minimum of $65\%$ of SNII should occur in superbubbles, increasing to $\approx 80-90\%$ when the spatial and temporal correlations of stellar clusters are considered.

If 80% of SNII are associated with supershells. for example. this would increase the required. replenishment rate per SN to z110230A...

If $80\%$ of SNII are associated with supershells, for example, this would increase the required replenishment rate per SN to $\approx 140-230\,M_{\odot}$.

But even if as few as of all SNII contribute in this way to the replenishment. the rate implied by the results of Dawsonetal.(2008) would still be sufficient.

But even if as few as of all SNII contribute in this way to the replenishment, the rate implied by the results of \citet{Daw:08} would still be sufficient.

This confirms that the necessaryreplenishment rates are likely to be achievable within supershells.

This confirms that the necessary replenishment rates are likely to be achievable within supershells.

The observed decline by a factor of two in the HI mass density may be a natural consequence of a replenishment rate about of that required to match consumption. as shown by the heavy solid line in Figure 2..

The observed decline by a factor of two in the HI mass density may be a natural consequence of a replenishment rate about of that required to match consumption, as shown by the heavy solid line in Figure \ref{fig:gasmodels}.

If the actual replenishment rate from supershells Hes somewhere between the required rate and our derived upper limit. though. there may in fact be too much newly replenished gas to allow any decline in the neutral gas mass density.

If the actual replenishment rate from supershells lies somewhere between the required rate and our derived upper limit, though, there may in fact be too much newly replenished gas to allow any decline in the neutral gas mass density.

A possible resolution in this scenario would be increasing the proportionality between the gas outflow rates and the SER as redshift decreases.

A possible resolution in this scenario would be increasing the proportionality between the gas outflow rates and the SFR as redshift decreases.

This is not unreasonable. as the SFH is becoming progressively more dominated by lower-mass galaxies with decreasing redshift (Juneauetal.2005;Panteral.Mobasheret2008).

This is not unreasonable, as the SFH is becoming progressively more dominated by lower-mass galaxies with decreasing redshift \citep{Jun:05,Pan:07,Mob:08}.

. Galaxies with stellar masses AL,3007:X107"A£.. dominate the SFH at :<1 (Mobasheretal.2008).

Galaxies with stellar masses $M_* \lesssim 10^{10}\,M_{\odot}$ dominate the SFH at $z\lesssim 1$ \citep{Mob:08}.

.. Such low-mass galaxies lose more mass in gas outflows in proportion to their SFR than high-mass galaxies. simply due to the former's shallower potential wells (e.g..Dekel&Silk1986:ΜαςLow&Ferrara1999:Tolstoy 2000).

Such low-mass galaxies lose more mass in gas outflows in proportion to their SFR than high-mass galaxies, simply due to the former's shallower potential wells \citep[e.g.,][]{DS:86,MF:99,FT:00}.

. This effect may contribute to the slow decline in the HI mass density.

This effect may contribute to the slow decline in the HI mass density.

We have treated a number of complex physical processes in very general terms.

While being cautious of oversimplification. we have attempted to capture the essential interactions between star formation. recycling from stellar evolutionary. processes. ISM processes of heating and ionization. recombination. cooling and molecule formation. together with infall from the IGM. and outflow of ISM material.

While being cautious of oversimplification, we have attempted to capture the essential interactions between star formation, recycling from stellar evolutionary processes, ISM processes of heating and ionization, recombination, cooling and molecule formation, together with infall from the IGM, and outflow of ISM material.

Most of this complexity 1s concealed within the replenishment factor A(f).

Most of this complexity is concealed within the replenishment factor $K(t)$.

One issue is that stellar winds and SNe contribute to all components of the ISM rather than solely to page.

One issue is that stellar winds and SNe contribute to all components of the ISM rather than solely to $\rho_{\rm SFG}$.

In a “galactic fountain" (Shapiro&Field1976;HouckBreg- 1990). infalling gas will contribute to. and outflowing gas will strip from. all components.

In a “galactic fountain" \citep{SF:76,HB:90}, infalling gas will contribute to, and outflowing gas will strip from, all components.

If recycled gas includes a component that never subsequently forms stars (such as some recycled gas in the ionized phase being ejected from the galaxy before contributing to star formation). the factor |Rp.(f) in Equation (1)) will be reduced. and (f) will need to be increased to compensate.

If recycled gas includes a component that never subsequently forms stars (such as some recycled gas in the ionized phase being ejected from the galaxy before contributing to star formation), the factor $+R\dot{\rho}_*(t)$ in Equation \ref{theeqn}) ) will be reduced, and $K(t)$ will need to be increased to compensate.

Our quantitative results strongly depend on the assumed gas outflow rate.

Variations by a factor of two or so in. either direction will still result in a constant or slowly varying HI mass density. as long as a proportionality with the SFR of the host galaxies remains (assuggestedbyVeilleuxetal.2005).

Variations by a factor of two or so in either direction will still result in a constant or slowly varying HI mass density, as long as a proportionality with the SFR of the host galaxies remains \citep[as suggested by][]{Vei:05}.

. The chosen outflow rate is an effective average over all star forming galaxies. and is consistent with observed trends (e.g..Martin1999;Pettinietal.2000:Veilleux 2005).

The chosen outflow rate is an effective average over all star forming galaxies, and is consistent with observed trends \citep[e.g.,][]{Mar:99,Pet:00,Vei:05}.

. While individual galaxies show a large observed scatter between outflow rates and SERs. for the ensemble properties of the total population this assumption should be robust.

While individual galaxies show a large observed scatter between outflow rates and SFRs, for the ensemble properties of the total population this assumption should be robust.

The proposed replenishment through the supershell mechanism is not inconsistent with some simultaneous— replenishment through infall.

The proposed replenishment through the supershell mechanism is not inconsistent with some simultaneous replenishment through infall.

Metallicity considerations. which we do not address here. do require infall of some low metallicity gas (Erb2008). and gas infall in local galaxies is well established (e.g..Bland-Hawthornetal.2007:Sancisi 2008).. although the observed infall rate is insufficient to match consumption.

Metallicity considerations, which we do not address here, do require infall of some low metallicity gas \citep{Erb:08},, and gas infall in local galaxies is well established \citep[e.g.,][]{Bla:07,San:08}, , although the observed infall rate is insufficient to match consumption.

therefore. a PAGD sar (Ilrivnak 1995. νοκ 1993).

therefore, a PAGB star (Hrivnak 1995, Kwok 1993).

Even on a low resolution sceterum. Lrivnak (1995) could see the enhancement. of lines of s-process elements.

Even on a low resolution spectrum, Hrivnak (1995) could see the enhancement of lines of s-process elements.

From the UDV photometry. Arkhipova ct al. (

From the UBV photometry, Arkhipova et al. (

2003) found. this star to be pulsating variable with a period of 90 davs.

2003) found this star to be pulsating variable with a period of $\sim$ 90 days.

Long term monitoring of this star using high resolution spectra has been carried out by. IxIochkova. Panchuck ‘Tavolzhanskava (2010) with following interesting findings.

Long term monitoring of this star using high resolution spectra has been carried out by Klochkova, Panchuck Tavolzhanskaya (2010) with following interesting findings.

The strong absorption lines such as low excitation line of Ballat 6141 not only show asvounetries in the profile with short wavelength: side of the profile showing extended wing than the red wine: these strong lines also show large amplitude profile variations (with time) caused by variations in blue wing while red wing remained unchanged.

The strong absorption lines such as low excitation line of at 6141 not only show asymmetries in the profile with short wavelength side of the profile showing extended wing than the red wing; these strong lines also show large amplitude profile variations (with time) caused by variations in blue wing while red wing remained unchanged.

The spectrum contains C» lines most likely formed. in. the circumstellar shell.

The spectrum contains $_{2}$ lines most likely formed in the circumstellar shell.

At the epoch of largest asvmmetry in strong. line the €» (0:1). band head at A 5635 is seen in emission.

At the epoch of largest asymmetry in strong line the $_{2}$ (0;1) band head at $\lambda$ 5635 is seen in emission.

The cores of hwdrogen lines show larger variations in radial velocities ( 8 kms 1) while weak metallic lines show smaller amplitude variations in radial velocities (~ 1 i).

The cores of hydrogen lines show larger variations in radial velocities $\sim$ 8 $^{-1}$ ) while weak metallic lines show smaller amplitude variations in radial velocities $\sim$ 1 $^{-1}$ ).

Molecular Cs lines remain stationary with time: the shift in. circumstellar features relative to systemic velocity. gives an expansion velocity Voss Of 15.0 |.

Molecular $_{2}$ lines remain stationary with time; the shift in circumstellar features relative to systemic velocity gives an expansion velocity $_{exp}$ of 15.0 $^{-1}$.

Our spectrum taken on Dec 27. 2009 also exhibits the features mentioned in Wlochkova. Panchuck Tavolzhanskava (2010).

Our spectrum taken on Dec 27, 2009 also exhibits the features mentioned in Klochkova, Panchuck Tavolzhanskaya (2010).

This star was analysed by Van Winckel and. Revniers (2000). (hereinafter: WR2000) who found it moderately metal-poor Fe/H] of 0.82 dex and showing enhancement of s-process elements.

This star was analysed by Van Winckel and Reyniers (2000) (hereinafter WR2000) who found it moderately metal-poor [Fe/H] of $-$ 0.3 dex and showing enhancement of s-process elements.

The present analysis covers additional elements Na. Mg and Zn and uses larger number of lines for many species (see Table S).

The present analysis covers additional elements Na, Mg and Zn and uses larger number of lines for many species (see Table 8).

Since the solar abundances usec in WR2000 are dillerent from our work. we have transformed. these abundances to solar abundances of Asplund et al. (

Since the solar abundances used in WR2000 are different from our work, we have transformed these abundances to solar abundances of Asplund et al. (

2005) to facilitate comparison.

All elements agree within £0.15 dex.

All elements agree within $\pm$ 0.15 dex.

IRAS 22223]4827's progenitor was most probably a thermally pulsing ACB star.

IRAS 22223+4327's progenitor was most probably a thermally pulsing AGB star.

This is. indicated. by the C/O ratio of unity and the about one dex enrichment. of the s-process elements.

This is indicated by the C/O ratio of unity and the about one dex enrichment of the $s$ -process elements.

Two of our program stars LRAS 17279-1119 ancl LIGAS 22223|4327 show significant s-process enhancement.

Two of our program stars IRAS 17279-1119 and IRAS 22223+4327 show significant s-process enhancement.

We have compared the spectra of these two objects with HAS 07140-2321 with similar temperature but without s-process enhancement in Although this object has been mentioned in several papers on PAGB stars. a contemporary abundance analysis using high-quality digital spectra. modern model atmospheres aud refined atomic data has not been undertaken.

We have compared the spectra of these two objects with IRAS 07140-2321 with similar temperature but without s-process enhancement in Although this object has been mentioned in several papers on PAGB stars, a contemporary abundance analysis using high-quality digital spectra, modern model atmospheres and refined atomic data has not been undertaken.

Abundance data from Ixodaira et al. (

Abundance data from Kodaira et al. (

1970) show strong ellects of dust-eas winnowing.

1970) show strong effects of dust-gas winnowing.

Yet. this star unlike other stars exhibiting severe dust-gas winnowing does not have an infrared excess.

Yet, this star unlike other stars exhibiting severe dust-gas winnowing does not have an infrared excess.

Llowever. the star is a spectroscopic binary. as are many or even all other stars exhibiting severe dust-gas winnowing.

However, the star is a spectroscopic binary, as are many or even all other stars exhibiting severe dust-gas winnowing.

gas Is dissipated from the disk.

gas is dissipated from the disk.

On the contrary. for Vega-like objects no giant planet needs to be formed and/or migrate inward.

On the contrary, for Vega-like objects no giant planet needs to be formed and/or migrate inward.

The gas may dissipate and still the planetesimal in the external part of the disk may produce dust by collisions.

We tentatively analyzed two small sub-sets of Vega-like objects: the Vega-like stars with planets and the Vega-like group with no Doppler detected planets.

The first group is composed of 7 stars: 6 with 70 jm excess detected by Spitzer (HD33636.HD50554.52265.82943.HD128311and117176:Beichmanetal.2006) and € Eri with infrared and submillimieter excesses (Greavesetal.1998:Zuckerman2001).

The first group is composed of 7 stars: 6 with 70 $\mu$ m excess detected by Spitzer \citep[HD 33636, HD 50554, HD 52265, HD 82943, HD 128311 and HD 117176; ][]{beichman06} and $\epsilon$ Eri with infrared and submillimieter excesses \citep{greaves98,zuckerman01}.

. In the second group we include 5 stars without Exoplanets detected by the Doppler technique (Santosetal.2004:Gilli2006) and showing infrared excess in 24 or 70 jim (HD7570.HD38858.Beichmanetal.2006:Bryden 2006).

In the second group we include 5 stars without Exoplanets detected by the Doppler technique \citep{santos04,gilli06} and showing infrared excess in 24 or 70 $\mu$ m \citep[HD 7570, HD 38858, HD 69830, HD 76151 and HD 115617; ][]{beichman06,bryden06}.

. The median metallicity of Vega-like stars with planets is +0.07 dex and the dispersion is 0.16 dex.

The median metallicity of Vega-like stars with planets is $+$ 0.07 dex and the dispersion is 0.16 dex.

For the Vega-like objects without planets these values are: 20.08 and 0.18 dex. respectively.

For the Vega-like objects without planets these values are: $-$ 0.08 and 0.18 dex, respectively.

It seems that when a Vega-like star has a planet the metallicity increases slightly.

However the small number of objects available as well as the dispersions prevent us from giving any statistical significance to this mitial trend.

However the small number of objects available as well as the dispersions prevent us from giving any statistical significance to this initial trend.

Greavesetal.(2007) proposed that the solid-mass (1.e.. metals) content in primordial disks. called Ms. is the fundamental parameter that regulates the planet/disk formation.

\citet{greaves07} proposed that the solid-mass (i.e., metals) content in primordial disks, called $_{\rm S}$, is the fundamental parameter that regulates the planet/disk formation.

If Ma is small. the star will form a Vega-like disk. while if Ms is larger. a giant planet may be formed.

If $_{\rm S}$ is small, the star will form a Vega-like disk, while if $_{\rm S}$ is larger, a giant planet may be formed.

Table | of Greavesetal.(2007) shows the range of metallicity and the final configurations (planet+debris.

Table 1 of \citet{greaves07} shows the range of metallicity and the final configurations $+$ debris,

ISM.

According to the models by ? a lower rratio would indicate larger densities.

According to the models by \citet{snijders07} a lower ratio would indicate larger densities.

In Fig.

13 the 12.81 line ratios are also lower in the center than around it.

\ref{fig:line-ratios} the 12.81 line ratios are also lower in the center than around it.

This ratio ranges between 0.002 and 0.013 in the center, which indicate densities larger than 106cm? in a >5 Myr old starburst system with solar metallicity and relatively high (q=8x10%) ionization parameter (?,theirFig.5)..

This ratio ranges between 0.002 and 0.013 in the center, which indicate densities larger than $10^6~\3cm$ in a $>$ 5 Myr old starburst system with solar metallicity and relatively high $\rm q=8\times10^8$ ) ionization parameter \citep[][their Fig.5]{snijders07}.

Based on our observed rratios, the rratios predicted with the model by ?,theirequation2 are about ten times larger than the observed ratios.

Based on our observed ratios, the ratios predicted with the model by \citet[][their equation 2]{pereira10} are about ten times larger than the observed ratios.

These can be a consequence of the about 10 times higher extinction found in NGC 4945 than in the sample of galaxies used by ?..

These can be a consequence of the about 10 times higher extinction found in NGC 4945 than in the sample of galaxies used by \citet{pereira10}.

The lline ratios obtained with the extinction correction in Fig. 13))

The line ratios obtained with the extinction correction in Fig. \ref{fig:line-ratios}) )

are just ~9% larger than without correction.

are just $\sim$ larger than without correction.

This relatively small change after the extinction correction is because even in a high-extinction situation the differential extinction between aand lis small, given that both lines are closely spaced in wavelength and not in one of the silicate absorption features.

This relatively small change after the extinction correction is because even in a high-extinction situation the differential extinction between and is small, given that both lines are closely spaced in wavelength and not in one of the silicate absorption features.

On the other hand, because of their larger differential (wavelength) extinction, the rratios corrected for extinction do change significantly from a factor —50 in the center (where the extinction is larger) to a factor 3 away from the center (where the extinction is lower).

On the other hand, because of their larger differential (wavelength) extinction, the ratios corrected for extinction do change significantly from a factor $\sim$ 50 in the center (where the extinction is larger) to a factor 3 away from the center (where the extinction is lower).

Note that only the ratios from our co-added fluxes (Table 1)) can be compared to other galactic nuclei, as the 10x10 aperture is comparable to the size probed in any of the more distant galaxy nuclei.

Note that only the ratios from our co-added fluxes (Table \ref{tab-c4:fluxes-fov}) ) can be compared to other galactic nuclei, as the $\times$ 10 aperture is comparable to the size probed in any of the more distant galaxy nuclei.

The rratio at the position of the H?O mega maser is about lower than the ratio obtained from the fluxes of the 10x10 co-added spectrum (Table 1)).

The ratio at the position of the $_2$ O mega maser is about lower than the ratio obtained from the fluxes of the $\times$ 10 co-added spectrum (Table \ref{tab-c4:fluxes-fov}) ).