REMOTE SENSING OF SOIL PROPERTIES IN SEMI-ARID RANGELAND AREAS, SEVILLETA WLR, NEW MEXICO

Characterization of soil properties in rangeland areas has only been achieved in a general sense for large areas of the south western United States. Soil taxonomic units are very broadly defined and most soil mapping units are soil complexes, making it difficult to determine soil properties at a point of interest. Logistical constraints ensure that more detailed mapping with conventional techniques will be unlikely for most of this landscape. However, there is an increasing demand for more accurate soil information of such areas for monitoring the impacts of climate change, ecohydrological modeling, trafficability, land mine detection, dust sources, and waste disposal.

The technical limitations associated with producing larger scale soil maps means that remote sensing of soil properties is the only option for producing more detailed soil maps of rangeland areas. Remote sensing has been used to provide a quantitative measure of surface reflectance from different bands which is used to identify some soil properties such as particle size, iron oxide content and moisture. However, the satellite images are only showing the reflectance of the top 1-2 cm of a soil surface and the question remains as to what sub surface soil properties can be identified by this process.

We used the Surface Energy Balance Algorithms for Land (SEBAL) algorithm that solves the surface energy balance on an instantaneous time scale for every pixel of a satellite image to produce maps of root zone soil moisture. Landsat images from the growing season from several years were analyzed to identify recurring patterns in soil moisture and compared to existing soil and landform maps. Good correlation between soil map unit boundaries, landform boundaries and the patterns of soil moisture suggest that this is a useful tool for mapping semi arid rangeland soils.