Analysis of the backscattering coefficient of salt-affected soils using modeling and RADARSAT-1 SAR data

Salt significantly changes the backscattering coefficient of wet soil. This is easily observed on RADARSAT-1 synthetic aperture radar (SAR) images acquired over a salty depression located in the Egyptian desert. The aim of this paper is to use backscattering models to understand the behavior of the backscattering coefficient over salt-affected soils and to evaluate the possibility of monitoring the salt content. Simulations conducted over salt-affected soils show a lower sensitivity of these models to soil moisture compared to nonaffected soils. Besides, there is no model suitable to represent the variation of the backscattering coefficient due to changes in soil salinity. These results are discussed with regard to the magnitude of the two components of the dielectric constant (epsilon' and epsilon). Based on a series of relationships, we propose a parametric formulation that allows us to determine the salinity acting on RADARSAT-1 SAR images without any use of backscattering models.