Downscaling of SMOS data using NDVI, Elevation, and Sand Fraction

Surface soil moisture information at high spatial resolution is necessary for better forecasting and understanding of various hydrological, meteorological and ecological models. Microwave remote sensing systems show great potential in retrieving soil moisture information on daily basis. However, major limitations using passive microwave systems are due to lower spatial resolution. Accurate fine-scale soil moisture observations are needed at a consistent basis to be used for local and regional scale models. In the absence of consistent high resolution soil moisture datasets, downscaling procedures enable to convert coarse resolution surface soil moisture estimates to high and liable resolution soil moisture estimates. Surface soil moisture distributions and dynamics depend greatly on vegetation (NDVI), topographic (EL), and sand (SF) features. The downscaling algorithm is based on the understanding of each of these physical parameter (NDVI, EL, and SF) and coarse remote sensing data and how they impact soil moisture retrievals. Results suggest that not all physical parameter (NDVI, El, and SF) affect surface soil moisture equally, since every region has its own soil composition. Unhealthy vegetation can be due to high sand fraction or seasonal change, or vice versa.