File size: 3,253 Bytes
2ba7a65
7a9ca84
2ba7a65
 
 
e1e058d
 
 
2ba7a65
e1e058d
2ba7a65
e1e058d
2ba7a65
e1e058d
 
 
 
 
 
 
 
 
453417a
c24226d
15ad93c
e1e058d
2ba7a65
e1e058d
2ba7a65
e1e058d
453417a
c24226d
 
 
 
15ad93c
 
c24226d
 
 
15ad93c
c24226d
 
 
 
 
 
15ad93c
 
 
453417a
e1e058d
2ba7a65
e1e058d
 
 
 
 
 
 
 
 
 
 
 
15ad93c
 
2ba7a65
e1e058d
2ba7a65
e1e058d
 
 
19598e9
e1e058d
c24226d
2ba7a65
e1e058d
2ba7a65
e1e058d
 
 
2ba7a65
e1e058d
d24053f
 
 
2ba7a65
e1e058d
2ba7a65
e1e058d
 
d24053f
2ba7a65
e1e058d
7a9ca84
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
---
license: cc
---
# ACYD: Agricultural Crop Yield Dataset

## Overview
The Agricultural Crop Yield Dataset (ACYD) provides weekly, admin-2 level covariates for Argentina, Brazil, and the USA from 1979–2024.  
It is designed for **crop yield prediction and benchmarking**.

---

## Variables

### Weather Variables
- Temperature 2m max  
- Temperature 2m min  
- Wind speed 10m  
- Reference evapotranspiration (RefET)  
- Vapor pressure  
- Snow water equivalent  
- Solar radiation  
- Precipitation  

**Source:** 
- 1979-2024: AG-ERA5, collected daily and averaged weekly. 495 departments and no missing values.
**Processing:** Applied HYDE-3.5 cropland mask within admin boundaries for each year; values are weighted averages over cropland fraction.  

---

### Land Surface Variables

* Leaf Area Index (LAI) Low (crops, grass, shrubs, etc.)
* Leaf Area Index (LAI) High (forests, woody vegetation)
* Normalized Difference Vegetation Index (NDVI)

483 departments and there are some missing values (see below)

**Source:**

* LAI: ERA5 Landscape Reanalysis, collected weekly and averaged.
* NDVI: NOAA AVHRR Climate Data Record (CDR), collected weekly and averaged.

  * 1982–2013: AVHRR NDVI (Version 4, derived from AVHRR Surface Reflectance).
  * 2014–2024: NOAA Climate Data Record of AVHRR NDVI (updated product).

**Processing:**

Applied HYDE-3.5 cropland mask within admin boundaries for each year; values are weighted averages over cropland fraction. 

LAI does not have missing values; however, NDVI does have missing weeks, particularly during the years 1994 and earlier. Missing values are kept as NaN.

---

### Soil Variables
- Bulk density  
- Cation exchange capacity (CEC)  
- Clay content  
- Coarse fragments  
- Nitrogen content  
- Organic carbon density  
- Organic carbon content  
- Soil pH (H₂O)  
- Sand fraction  
- Silt fraction  

**Source:** SoilGrids / ISRIC (static soil properties) published in 2020. 500 departments and no missing value.
**Processing:** Aggregated at the admin-2 level **(no cropland mask applied)**.  

---

## Geography
- **Argentina**
  - weather: 495 departments,
  - Land Surface: 483 departments
  - soil: 500 departments
  - crop yields: vary 181-296 departments

---

## Crops Covered
- **Argentina:** Corn, Soybean, Sunflower, Wheat  
---

## Time Span
- weather, LAI: 1979–2024
- crop yields: 1970–2025 (depending on crop/country)
- soil: static (data was first published in 2020)

---

## Files
- **Raw files** — direct pulls from source datasets.  
- **Processed CSVs** — one file per variable, aggregated to admin-2 units, with weighted average on the boundaries.

---

## Licenses

This dataset (ACYD) integrates multiple open data sources.  
All preprocessing code and benchmark packaging is released under the **MIT License**.  
The underlying data retains its original licenses:

- ERA5 / AgERA5: Copernicus license (CC BY 4.0 equivalent)  
- NOAA AVHRR NDVI: Public Domain (U.S. Government work)  
- HYDE 3.5 Cropland: CC BY 4.0  
- USDA NASS (USA yields): Public Domain  
- IBGE (Brazil yields): CC BY 3.0  
- MAGyP (Argentina yields): CC BY 4.0  
- FAO (if included): CC BY-NC-SA 3.0  

Users must comply with these original licenses when reusing the dataset.