SoilGrids is a system for automated soil mapping based on global soil profile and environmental covariate data. SoilGrids represents a collection of updatable soil property and class maps of the world at 1 km and 250 m spatial resolution produced using automated soil mapping based on machine learning algorithms. It aims at becoming OpenStreetMap and/or OpenWeatherMap for soil data. SoilGrids predictions are updated on a regular basis (at least every few months).
SoilGrids is a global soil data product generated at ISRIC — World Soil Information as a result of international collaboration (see list of contributing agencies) and as a proposed contribution to the Global Soil Partnership initiative. SoilGrids1km and SoilGrids250m are outputs of a system for automated global soil mapping developed within the Global Soil Information Facilities framework. This system is intended to facilitate global soil data initiatives and to serve as a bridge between global and local soil mapping.
Important notice: the SoilGrids layers are of limited thematic (relatively wide confidence limits) and spatial accuracy and still contain artifacts and missing pixels. Help us improve these maps by contributing georeferenced soil profiles (points) and covariate data (rasters with global coverage at resolutions of 1 km, 250 m, 100 m or better). As soon as we receive new ground observations and/or new covariates, we can re-run the predictions to produce an updated version of SoilGrids. Read more...
Last update: June 2016.
Under preparation: update SoilGrids250m (end of 2016).
Hengl, T., Mendes de Jesus, J., Heuvelink, G. B.M., Ruiperez Gonzalez, M., Kilibarda, M. et al. (2016?) SoilGrids250m: global gridded soil information based on Machine Learning. Earth System Science Data (ESSD) in review.