CAPABILITY OF MULTITEMPORAL ERS-1 SAR DATA FOR WET-SNOW MAPPING

The backscattering behavior of the snow cover was investigated on the basics of nine European Remote Sensing-1 Synthetic Aperture Radar (ERS -1 SAR) images of a southeastern Quebec study area in Canada (center l attitude 47 degrees 30' N center longitude 69 degrees 45'W), which wer e acquired in the winters of 1992 and 1993. Comparative field measurem ents on physical properties of the snow cover were carried out in 25 t est sites. Dry snow shores backscattering coefficients similar to thos e of snow-free surfaces, whereas backscattering is clearly reduce for wet snow. An algorithm was developed for oct-snow mapping by means of C-band SAR, based on multitemporal ERS-1 data. The SAR-derived snow ma ps could be successfully applied for runoff modeling during a full per iod of snow melt. The accuracy of the method for wet-snow zu mapping d emonstrates the high operational potential of SAR. C-band ERS-1 SAR ha s been proved to be a suitable device for snow monitoring. It is of gr eat interest for studies in hydrology. For our experimental conditions , the algorithm operates with a relative false detection ratio of 1.33 % for the study area. (C) Elsevier Science Inc., 1997.