Study for estimating spatial distributions of snow hydrological parameters usingmulti-temporal RADARSAT data

Abstract

The objective of this study is to develop an algorithm for quantitative estimation of snow parameter distributions using Synthetic Aperture Radar (SAR) data anda numerical scattering model that includes the Integral Equation Method (IEM) model and first-order radiative transfer solution. Simultaneous RADARSAT SAR observations were carried out during the winter seasons of 1996 and 1998. The study area is a part of the Hokuriku District of Japan that is well known for having ofwet snow with a complicated layer structure from the beginning of the winter season. We compared the results of simulated backscattering coefficients with the ground-truth data and observed backscattering coefficients. Since the agreement was good, we demonstrated the possibility of estimating roughness distributions of soil surface using images acquired just before the snowfall, assuming that thesoil conditions did not change during a winter season. Furthermore, we estimatedthe distributions of snow depth and roughness using images after a snowfall and estimated soil roughness distributions.