SoilGrids250m

Global spatial predictions of soil properties and classes at 250 m resolution.

This repository hosts the source code used to create the SoilGrids version published in 2017 (Hengl et al. 2017). The maps generated in 2017 can be downloaded from here. For information about the most recent version of SoilGrids, please visit the project page. The new maps are displayed in the new portal soilgrids.org.

Citation:

• 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

What can you find on this github repository:

• R scripts documenting processing steps,
• Sample code explaining the modelling framework,
• Functions for Cross-validation of ensemble models with examples,
• Examples of predictions, outputs and visualizations,
• Technical specifications of the system,