Using Polarimetric and Dual-Pol RADARSAT-2 data for Soil Moisture Estimation
Francois Charbonneau(1) and Melanie Trudel(2)
(1) Canada Centre for Remote Sensing, 588 Booth St, K1A 0Y7 Ottawa, Canada
(2) École de Technologie Supérieure, 1100 rue Notre-Dame Ouest, H3C 1K3 Montréal, QC, Canada
Knowing the water balance properties of an aquifer is important in various domains like security and health, weather forecast and climate change. Spatio-temporal surface soil moisture analysis from SAR data can be used as input for hydro-geological models characterizing a basin. In this study, we are using RADARSAT-2 fully polarimetric data for generating a coarse soil roughness classification and for reducing the vegetation effect on the backscattering intensity. Integrate Equation Model (IEM) combined with the use of HH and VV magnitudes are used to generate a look-up table (LUT). Surface soil moisture estimations are generated by two different LUT inversion methods, and then compared to in situ measurements. The coarse soil roughness classification helps to constrain the search range in the LUT, which limits the number of local minimums. The soil moisture error can be handling by the hydro-geological data assimilation model.
This research is part of an Earth observation applications development, of Natural Resources Canada - Groundwater Mapping Program, focusing on modelling the polarimetric SAR signal to geophysical parameters to improve and support the study of groundwater, specifically for aquifer recharge modeling.