Downscaling SMOS soil moisture data using geospatial technology

Abstract

Downscaling of SMOS data using Geospatial Technology was carried out to develop a system to enhance the spatial resolution of soil moisture which otherwise is available at~ 40 Km, by incorporating various ancillary information, at a much higher spatial scale. The current spatial resolution of SMOS data has limited uses due to the coarse resolution. Two study sites one each in Udham Singh Nagar district (Rudrapur area) and Nainital District (Haldwani area) were selected for developing downscaling approach. Point wise soil moisture was interpolated in Quantum GIS software by employing Inverse Distance Weightage algorithm. LANDSAT 8 images were used to generate land use and land cover map of the study region. Soil information was retrieved from the earlier studies carried out in the department. Soil layer and land use map were merged in order to analyze the moisture regime of the area. Soil moisture at each polygon was extracted from interpolated soil moisture layer. Multivariate model was used to downscale the soil moisture values of SMOS at 1km spatial resolution by using the layers of LST and EVI as input. Results indicate that downscaled moisture exhibited quite good agreement with measured soil moisture (R2, 0.924 RMSE 0.0487).