A DOWNSCALING SCHEME FOR DERIVING SPATIALLY CONTINUOUS FINE-RESOLUTION SOIL MOISTURE DATA BASED ON GAP-FREE LAND SURFACE TEMPERATURE

Soil moisture (SM) downscaling has become more and more crucial for assisting the application of the coarse-resolution SM product, such as climate change, sustainable development of agroforestry, efficient management of water resources, and monitoring of natural hazards. The main idea of downscaling methods lies on the help of fine-resolution auxiliary data, such as the widely used, land surface temperature (LST) and normalized difference vegetation index (NDVI). However, in the downscaling process, the ancillary data, especially for the daily LST, is strongly affected by cloud cover, resulting high frequency of blank areas in the final downscaled SM products. By contrast, the impact is usually omitted or paid less attention in current downscaling studies. To obtain the spatially continuous fine-resolution SM product, this study firstly introduced an annual temperature cycle (ATC) model to fill the gaps in daily Moderate Resolution Imaging Spectroradiometer (MODIS) LST product induced by cloud cover. Then the 36-km SM product from Soil Moisture Active Passive (SMAP) satellite mission was downscaled from 36-km to 1-km spatial resolution with the synergistic use of the filled LST and MODIS NDVI to estimate spatially continuous fine-resolution SM product.