Classification of Dry and Wet Snow of Glaciers in Western Himalayas Using SAR Polarimetric Decomposition—An Application of Satellite Image Processing

Monitoring snow cover is an essential element in regulating the hydrological cycle, regional energy balances, and water use. Environmental modeling includes the monitoring of snow surface dynamics. Optical remote sensing data are useful for mapping snow cover, but their use is restricted by solar illumination conditions. On the other hand, Synthetic Aperture Radar (SAR) can record snow moisture changes in any weather. Even though there are techniques for identifying dry or wet snow in the literature, these methods do not take into account the case of defrosting in the snow cover. The present study involves the mapping of dry and wet snow in the region of Himachal Pradesh which includes the Samudra Tapu glacier, Bara Shigri, Chhota Shigri, and Gepang Gath glaciers located in the Chandra basin. By using dual polarimetric synthetic aperture radar (SAR) data, it is possible to distinguish between dry and wet snow. Sentinel-1 Single Look Complex (SLC) data is used for the polarimetric analysis in the present study. H-α decomposition for dual pole is done which gives accurate results for the classification of wet and dry snow. From the obtained results, wet snow and dry snow were classified using unsupervised k-means cluster classification and it is validated using optical data obtained from Landsat—8. From results obtained after a comparison between SAR and optical data, it can be seen that SAR data gives better and more accurate results.