Compensation of Terrain Azimuthal Slope Effects in Geophysical Parameter Studies Using Polarimetric SAR Data

The existence of significant terrain relief reduces the accuracy of algorithms intended to extract geophysical parameters from polarimetric synthetic aperture radar (SAR) images. For this reason most inversion studies attempting to estimate biomass, crop type, snow-cover, surface roughness, or soil moisture, from SAR data are best done in areas of flat earth. In areas which require compensation for topography, limited corrections have been made in the past for range slopes using a coregistered digital elevation model (DEM). No corrections have been made, however, for terrain slope-induced polarimetric effects in the azimuthal (or along-track) direction. Recent methods, using polarimetric SAR data, of measuring terrain slopes in the azimuthal direction have been developed. This article presents an algorithm to use these slope estimates to compensate for the effects of azimuthal terrain slopes. The algorithm involves rotation of the polarimetric covariance matrix to achieve maximum azimuthal symmetry. Evidence is given to substantiate the degree to which azimuthal slope-effects have been removed from the data. NASA/AIRSAR P-, L-, and C-band data, obtained for open terrain in Death Valley National Park, California (1994), are used to evaluate the compensation algorithm.