This repository contains the code used for data input preparation and model output analysis for the paper: Grogan D.S., Frolking S., Wisser D., Prusevich A., Glidden S. (2021, accepted) Global gridded crop harvested area, production, yield, and monthly physical area data circa 2015. Scientific Data (doi XXXX).
Data products analyzed in the paper and by this repository's code are available for download here:
Frolking, S., Wisser, D., Grogan, D., Proussevitch, A. & Glidden, S. GAEZ+_2015 Crop Harvest Area. Harvard Dataverse https://doi.org/10.7910/DVN/KAGRFI (2020).
Frolking, S., Wisser, D., Grogan, D., Proussevitch, A. & Glidden, S. GAEZ+_2015 Crop Yield. Harvard Dataverse https://doi.org/10.7910/DVN/XGGJAV (2020).
Frolking, S., Wisser, D., Grogan, D., Proussevitch, A. & Glidden, S. GAEZ+_2015 Crop Production. Harvard Dataverse https://doi.org/10.7910/DVN/KJFUO1 (2020).
Grogan, D., Prusevich, A., Frolking, S., Wisser, D. & Glidden, S. GAEZ+_2015 Monthly Cropland Data: Global gridded monthly crop physical area for 26 irrigated and rainfed crops. MyGeoHub https://doi.org/10.13019/J2BH-VB41 (2021).
This perl script executes the algorithm comprising "Step 6" in the paper, which applies the MIRCA2000 monthly crop calendars to the GAEZ+ 2015 annual cropland harvest area data
This perl script aggregates the Hydrosheds 5-minute river network into smaller basins of the size 20-200 km2. These basins are used for validation by cropland_validation_HYDE_PUB.R, with results appearing in Table 6 and Figures 7 and 8.
This R code compares the GAEZ+ 2015 harvest area to FAOSTAT 2015 harvest area statistics by country. This code produces Tables 3 and 4.
This R code compares GAEZ+ 2015 crop yield for maize, rice, soybean, and wheat to the GDHY data product. This code produces Figures 3 and 4.
This R code compares GAEZ+ 2015 physical cropland area to the HYDE v2.3 data product's cropland area. This code produces the results shown in Table 6 and Figures 7 and 8.
This R code compares GAEZ 2010 cropland extent to the MapSPAM data product, and produces Figure 10.
This R script compares year 2010 crop harvest area as generated by GAEZ_2010 to the FAOSTAT-reported crop harvest area. Results produce Table 3 and Table 4 in Grogan et al 2021 publication
cropland_spacialAgg_validation_PUB.R This R script uses linear regression to validate the spatial distribution of GAEZ+2015 crops by comparing spatial aggregates of cropland area to spatial aggregates of cropland area published by the HYDE data product. HYDE data can be downloaded from: https://dataportaal.pbl.nl/downloads/HYDE/