From SMOS/SMAP Surface Soil Moisture to Precipitation Rate

Surface soil moisture intrinsically contains information about previous rain event. Different research teams (W. Crow (USDA, US), L. Brocca (IRPI, Italy), T. Pellarin (IGE, France), N. Wanders (Utrecht, NL)) investigated the way to use satellite soil moisture measurements to derive a precipitation product. At IGE (Grenoble Alpes University), a methodology based on the assimilation of remotely sensed soil moisture measurements into a simple soil moisture - precipitation model was developed: the PrISM methodology (Precipitation Inferred from Soil Moisture). The concept of the PrISM methodology consists on generating a first-guess soil moisture time-series based on a given satellite precipitation product and to exploit the difference between satellite and simulated soil moisture by increasing or decreasing the amount of water of the original satellite precipitation product.

Results were assessed over various regions (US, Europe, India, Australia and Africa) and using different soil moisture measurements (SMOS-L3, SMAP, ASCAT, SMOS-IC) as well as a merged product SMOS&SMAP. Globally, the PrISM precipitation product improves initial precipitation product and is consequently a powerful tool to provide near real time precipitation product. The following illustration gives an overview of the performance of three real-time satellite precipitation products (CMORPH, PERSIANN and TRMM-RT) versus SMOS-PrISM products assessed over 20 raingauge stations in Burkina Faso (2012) in term of correlation, RMSE and bias (red boxes). The blue boxes (PrIMS products) indicate the improvement of the performances of the three real-time satellite precipitation products due to SMOS soil moisture information. PrISM precipitation products (0.25°, 3 hours) are available on CATDS from 2010 to 2017, for Africa and Australia.