A simple and effective method for correcting soil moisture and precipitation estimates using AMSR-E measurements

Recent studies have demonstrated the possibility to use surface soil moisture satellite products in order to correct precipitation estimates. The overall approaches are generally based on adjusting the precipitation rate into a water balance model in order to match observed and simulated soil moisture. In the present study, a recently proposed approach was improved and assessed over a 4-year period in West Africa. Three different satellite precipitation products (CMORPH, TRMM-3B42 and PERSIANN) were used in this study. The new algorithm was evaluated over three 0.25 × 0.25° areas in Niger, Mali and Benin and precipitation estimates (before and after correction) were compared to observed ground-based precipitation measurements. An assessment was also conducted at the regional scale using 48 independent rainfall stations located in Togo, Niger and Burkina Faso. The proposed methodology showed that, after correction, the three satellite precipitation products converge to the ground-based precipitation measurements, demonstrating the robustness of the methodology. In addition, soil moisture estimates derived from the methodology were found to be in good agreement with ground-based soil moisture measurements (R2 > 0.78 in Niger, R2 > 0.54 in Mali and R2 > 0.64 in Benin).