Scatterometer and multispectral-based image pansharpening for improved spatial resolution snow cover maps

Abstract

Snow cover monitoring in the Western Himalayan region is essential for understanding regional climate variability, hydrological processes, and water resource management. This study investigates changes in snow cover using the integration of optical and microwave satellite datasets. Data obtained from MODIS (Moderate Resolution Imaging Spectroradiometer) and SCATSAT-1 (Scatterometer Satellite-1) were combined to improve the reliability of snow mapping in complex mountainous terrain. To enhance spatial detail and reduce the limitations of individual datasets, five pan-sharpening techniques—Brovey Transform (BT), Gram–Schmidt (GS), Intensity Hue Saturation (IHS), Nearest Neighbor Diffusion (NND), and Principal Component Analysis (PCA)—were applied before classification. Snow and non-snow areas were identified using the Support Vector Machine. The integration of microwave backscatter with optical reflectance improved the discrimination between snow and non-snow surfaces and enhanced the spatial clarity of the imagery. The results demonstrate that pan-sharpened images provide better boundary delineation and more reliable classification compared to optical data alone. Change detection analysis highlights variations in snow distribution across the Himalayan region. The findings confirm that the fusion of microwave and optical satellite data significantly improves snow cover mapping, particularly under cloud cover, and provides a reliable approach for monitoring snow dynamics in mountainous regions.