The maps were prepared using soil information from 23,273 soils samples, collected from 56 districts
covering seven provinces. These soil properties are combined with a stack of 168 remote
sensing-based soil covariates (SRTM DEM derivatives, climatic images, vegetation index etc.). Later
the spatial predictions were generated using a machine learning method and the random forest.
Please refer to the following journal article for the detailed procedure:
Hengl T, Mendes de Jesus J, Heuvelink GBM, Ruiperez Gonzalez M, Kilibarda M, Blagotić A, et al.
(2017) SoilGrids250m: Global gridded soil information based on machine learning. PLoS ONE 12(2):
e0169748. doi:10.1371/journal. pone.0169748