Performance of soil moisture retrieval algorithms using multiangular L band brightness temperatures

Abstract

The Soil Moisture and Ocean Salinity (SMOS) mission of the European Space Agency was successfully launched in November 2009 to provide global surface soil moisture and sea surface salinity maps. The SMOS single payload is the Microwave Imaging Radiometer by Aperture Synthesis (MIRAS), an L band two‐dimensional aperture synthesis interferometric radiometer with multiangular and polarimetric imaging capabilities. SMOS‐derived soil moisture products are expected to have an accuracy of 0.04 m3/m3 over 50 × 50 km2 and a revisit time of 3 days. Previous studies have remarked the necessity of combining SMOS brightness temperatures with auxiliary data to achieve the required accuracy. However, the required auxiliary data and optimal soil moisture retrieval setup need yet to be optimized. Also, the satellite operation mode (dual polarization or full polarimetric) is an open issue to be addressed during the commissioning phase activities. In this paper, an in‐depth study of the different retrieval configurations and ancillary data needed for the retrieval of soil moisture from future SMOS observations is presented. A dedicated L2 Processor Simulator software has been developed to obtain soil moisture estimates from SMOS‐like brightness temperatures generated using the SMOS End‐to‐End Performance Simulator (SEPS). Full‐polarimetric brightness temperatures are generated in SEPS, and soil moisture retrievals are performed using vertical (Tvv) and horizontal (Thh) brightness temperatures and using the first Stokes parameter (TI). Results show the accuracy obtained with the different retrieval setups for four main surface conditions combining wet and dry soils with bare and vegetation‐covered surfaces. Soil moisture retrievals using TI exhibit a significantly better performance than using Thh and Tvv in all scenarios, which indicates that the dual‐polarization mode should not be disregarded. The uncertainty of the ancillary data used in the minimization process and its effect on the retrievals is thoroughly evaluated, and an optimum soil moisture retrieval configuration is devised.