The map is a hybrid of 30-second STATSGO over CONUS and 5-minute FAO elsewhere, which is available from the NCAR RAL website. The version of Noah model in GLDAS2 uses the FAO 16-category soil texture class. GLDAS uses the top layer soil parameter data for all layers, shown below with the images and links for the data files. The soil texture map was used if provided by the model developer, otherwise, it was computed based on the soil fractions.

In GLDAS2 products, our approach is to stay with model’s default parameter datasets as much as possible, as opposed to conforming models to use the GLDAS standard parameter datasets as done in GLDAS1. A look up table then provides the model specific soil parameters indexed based on a class. A texture class can be mapped from the fractions of sand, silt, and clay in a given grid cell for the model's soil texture classification scheme. Some LSMs use a soil texture classification scheme to specify soil parameters based on texture classes, while others derive soil hydrologic and thermal properties from sand, clay, and silt fractions.

Land Surface Models use a variety of methods to specify soil parameters.

Download sand and land