A physically-based inversion algorithm for retrieving soil moisture in passive microwave remote sensing

Surface soil moisture plays a key role in water and energy exchanges at the land–atmosphere interface. Previous research on the microwave remote sensing of soil moisture focused mostly on the forward modeling problem. This paper presents a unique inversion algorithm (Hong’s algorithm) for soil moisture retrieval with reduced degrees of freedom of the Fresnel equation; the algorithm is based on the characteristics of the polarization ratio for rough surfaces around the Brewster angle and Hong’s approximation, which is an approximate relationship between vertically and horizontally polarized reflectivities. Using the daily observations of the Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E) taken at 0.25° × 0.25° pixel, the soil moisture is estimated globally over the land surface using a relation between the complex dielectric constant and the soil moisture after retrieving the surface roughness and complex dielectric constant. The surface soil moisture products obtained from Hong’s algorithm are validated using in situ observations carried out in Oklahoma and Georgia in 2003. The official AMSR-E soil moisture products are compared to assess Hong’s algorithm using identical brightness temperature data sets. The biases and the root mean squared error calculated in comparison with the ground observations are in the range of −0.059 m3/m3 to 0.01 m3/m3 and 0.023 m3/m3 to 0.065 m3/m3, respectively. Hong’s algorithm estimates the soil moisture within the estimated accuracy of AMSR-E surface soil moisture, 0.06 m3/m3, without requiring a priori information about the roughness and the dielectric constants of the surface. With a minimal use of forward models, this research could also be applied to global soil moisture retrieval in the Soil Moisture and Ocean Salinity satellite mission and the Soil Moisture Active/Passive mission.