Passive microwave soil moisture downscaling using NLDAS and MODIS data

The soil moisture retrieved from Advanced Microwave Scanning Radiometer (AMSR-E) is an important variable that has been applied in agriculture, hydrology, climate and weather. However, low spatial resolution (1/4 degree) of AMSR-E derived soil moisture cannot fulfill the requirements of high spatial resolution. In this paper, we studied the relationship between three factors: daily surface temperature range, Normalized Difference Vegetation Index (NDVI) and daily averaged soil moisture, using the 30-year period North America Land Data Assimilation System (NLDAS) Land Surface Temperature (LST) and soil moisture products (1/8 degree); MODIS (Moderate-Resolution Imaging Spectroradiometer) and AVHRR (Advanced Very High Resolution Radiometer) NDVI products (1/20 degree) in Oklahoma. We derived relationships between temperature difference and soil moisture under different vegetation conditions then derived the downscaled soil moisture by comparing the difference between AMSR-E and MODIS retrieved soil moisture. This soil moisture downscaling method not only provides disaggregated soil moisture data for passive microwave sensors, but also for active microwave sensors, such as Soil Moisture Active Passive Mission (SMAP).