SoilGrids — global gridded soil information

Share on:

A system for automated soil mapping based on global compilation of soil profile data and publicly available remote sensing data.

SoilGrids is a system for automated soil mapping based on state-of-the-art spatial predictions methods. SoilGrids predictions are based on globally fitted models using soil profile and environmental covariate data. Currently, serves a collection of updatable soil property and class maps of the world at 1 km / 250 m spatial resolutions produced using automated soil mapping based on machine learning algorithms. aims at becoming OpenStreetMap and/or OpenWeatherMap for soil data. SoilGrids data is available publicly under the Open DataBase License.

For the most up to date version of SoilGrids refer to:


SoilGrids1km and SoilGrids250m are outputs of a system for automated global soil mapping developed within ISRIC's workstream on 'Soil Information Brokering'. This workstream is intended to facilitate global soil data initiatives and to serve as a bridge between global and local soil mapping.

SoilGrids have been documented in detail in:

Data download:

SoilGrids data (GeoTiffs) can be obtained either via the web-mapping interface at, via FTP, via the REST service and/or via a smartphone App. To download the complete global maps please use the FTP service:

Further reading:

WorldGrids — Global Environmental Layers

SoilGrids, and similar digital soil mapping projects, are based on using remote sensing data products. In the period 2012–2016 the main focus of global soil mapping were MODIS land products, SRTM Digital Elevation Model and various climatic data products. Beyond 2017 the focus of global soil mapping is slowly shifting toward finer and finer resolutions (Landsat, Aster, ALOS, Sentinel satellites). From the times when soil mapping was purely based on geomorphological interpretation of landscape, we have come to an era when soil mapping is primarily based on remote sensing data sources. The data portal aims at facilitating using remote sensing and bottom-up produced covariates of especial interest to soil mapping. Global data sets are available at various resolutions from 100 m to 10 km. Diversity of DEM derivatives, climatic images, MODIS derivatives, lithological and parent material maps will be made available via ISRIC's Web Mapping/Coverage Service.

The majority of covariate layers produced and/or pre-processed by ISRIC are original contributions and can not be found on other data repositories. To contribute with a new layer/predictor to this repository please contact Luís de Sousa.

Further reading: