Soil Moisture Estimation at 500m using Sentinel-1: application to African sites

Abstract

This paper proposes a change detection approach for the estimation of soil water content, at a spatial resolution of 0.5 km and a temporal resolution of 6 days. The algorithm proposes a soil moisture index between 0 and 1. 0 corresponds to the driest context, 1 corresponds to the wettest context. The approach is being tested on different study sites with sentinel-1 radar data. Unlike the classic change detection approach, the algorithm takes into account the effects of land use, vegetation development and the seasonal context. Results show a good correlation between satellite estimations and true measurements for the African studied regions.