Evaluation of surface soil moisture distribution by means of SAR remote sensing techniques and conceptual hydrological modelling

Abstract

This paper discusses the attempts to derive surface soil moisture information for two catchments under different climates in Europe based on active microwave remote sensing techniques and conceptual hydrological modelling. The two catchments were selected as test sites during the MAC-EUROPE'91 campaign. The first catchment, Slapton Wood, is situated in south-west England. The second catchment, Virginiolo, is situated in Tuscany, Italy. For both sites, multitemporal full polarimetric AIRSAR (C-, L- and P-band) data are available for the summer of 1991. Based on ground-truth data collected during the flights, simple SAR-soil moisture relationships are established. This allows the estimation of the spatial variation of surface soil moisture at the field scale. From hydrological and meteorological data acquired during the experiment, a conceptual water and energy balance model was calibrated. Simulation results for soil moisture are then compared with SAR-based soil moisture estimates. The comparison between sampled, modelled and SAR-derived soil moisture at the transect scale is acceptable, but the spatial patterns of soil moisture predicted by the conceptual model and the SAR images are different owing to the influence of vegetation. The results are also useful in pointing out some of the major problems encountered in the application of SAR data at the catchment scale.