Application of Sentinel-1 VH and VV and Sentinel-2 for soil moisture studies

The results of application of microwave and optical satellite data for soil moisture (SM) assessment are presented. The research has been carried out from 2015 to 2016 at Biebrza Wetlands test site located in North-East Poland, designated by Ramsar Convention as Wetlands of International Importance. A regression models based on Sentinel-1 backscattering coefficients (sigma degrees) have been developed to generate the soil moisture (SM) maps over Biebrza Wetlands. The optical data from Sentinel-2 have been used for the classification of wetlands vegetation habitats to improve SM predictions. The wetland vegetation differed, there were reeds, sedge-moss, sedges, grass-herbs, and grass. The majority of the changes occurred in moist habitats, while anthropogenic appeared more stable during study period. The observed changes were referred to moving/grazing changes and weather effects causing droughts/floods. SM differed from 30% during the drought season in 2015 to 95% in the wet season in 2016. It has been examined the impact of biomass and SM on microwave signal under changing soil moisture and vegetation growth conditions. Vegetation biomass has been characterized by measured in-situ LAI and by vegetation indices calculated from Sentinel-2, Terra MODIS data. The impact of SM and LAI on sigma degrees calculated from Sentinel-1 data showed that LAI dominates the influence on sigma degrees when SM is low. The analysis have been done to estimate the threshold of the SM values which dominate the backscatter. This study demonstrates the capability of Sentinel-1/2 data to estimate SM, offering an important advantage for wetlands monitoring.