Soil moisture estimation in Ferlo region (Senegal) using radar (ENVISAT/ASAR) and optical (SPOT/VEGETATION) data

Abstract

The sensitivity of the radar signal to the seasonal dynamics in the Sahel region is a considerable asset for monitoring surface parameters including soil moisture. Given the sensitivity of the radar signal to vegetation mass production, roughness and soil moisture, the main problem has been to estimate the contribution of these three parameters to the signal. This study aims to circumvent this problem by combining radar with optical data. The DMP (Dry Mater Product) extracted from SPOT data allowed to estimate vegetation mass production. Surface roughness was estimated from radar data during the dry season. Because during the dry season, radar signal is only conditioned by soil roughness in this region a Radiative Transfer Model (RTM) was used: it consists in a microwave scattering model of layered vegetation based on the first-order solution of the radiative transfer equation and it accounts for multiple scattering within the canopy, surface roughness of the soil, and the interaction between canopy surface and soil.This model was designed to account for the branch size distribution, leaf orientation distribution, and branch orientation distribution for each size. In this study, the RTM has been calibrated with ESCAT (European Radar Satellite Scatterometer) data, and has been used in order to estimate soil moisture.The results obtained have allowed to track the spatial and temporal dynamics of soil moisture on the one hand, and on the other hand the influence of geology and morphopedology on the spatial dynamics of the soil moisture variability. These results are promising despite the fact that the inversed RTM often faces difficulties to interpret the signal for saturated soils, giving an aberrant value of soil moisture more often than not. Keywords: Soil moisture, Radar remote sensing, ASAR, ERS, SPOT-VEGETATION, Ferlo