SMAP Radiometer Soil Moisture Downscaling in CONUS

SMAP (Soil Moisture Active/Passive) and SMOS (Soil Moisture Ocean Salinity) provide soil moisture observations that can be used for studying the global hydrological cycle, agriculture, ecology, and land atmosphere interactions. SMAP provides soil moisture at two grid scales; 36 km (which is close to its native radiometer spatial resolution) and an enhanced grid resolution of 9 km. However, these scales are not compatible with some agricultural and watershed applications that require a higher spatial resolution. This study applied a downscaling algorithm to the SMAP Level-2 radiometer 36 km product and improved the grid resolution to 1 km over the CONUS (Contiguous United States). The downscaling algorithm is built on the thermal inertial relationship between daily temperature change and averaged soil moisture modulated by Normalized Difference Vegetation Index (NDVI). The average soil moisture and thermal inertia model functions were developed by using data from NLDAS (North America Land Data Assimilation System) and LTDR (Land Long Term Data Record) for 1981 – 2016. The algorithm is applied with the 1 km MODIS Aqua LST product and the downscaled SMAP 1 km soil moisture was validated by in situ soil moisture measurements from the ISMN (International Soil Moisture Network). The validation variables show improved accuracy of the downscaled soil moisture.