SPAM ADM1 extraction by raster files with GAUL
Ulrike mentioned that the Adm1 GGS map is not really up to date in terms of shapes for ll adm1 units. Now I checked and realized that indeed a lot of these shapes were missing. So all the zonal statistics would not work and would lead to an underestimation when extracting values from raster files (crop disaggregation).
Justification of GAUL: Higher accuracy of boundaries than GADM and it is more relevant for linking with agricultural FAO data. https://www.geos.ed.ac.uk/~gisteac/proceedingsonline/GISRUK2012/Papers/presentation-35.pdf
So I will have to perform a zonal statistic for all the SPAM GeoTIFF 2010 using GAUL adm1 as the shapefile so that I have production, harvested, and physical area for all the different adm1 units worldwide.
Then I have to redo all the crop disaggregation.
And then subset it to EU and update it in the Validation dataset