Non-Binary Snow Index for Multi-Component Surfaces

A Non-Binary Snow Index for Multi-Component Surfaces (NBSI-MS) is proposed to map snow/ice cover. The NBSI-MS based on the spectral characteristics of different Land Cover Types (LCTs), such as snow, water, vegetation, bare land, impervious, and shadow surfaces. This index can increase separability between values corresponding snow from other LCTs accurately delineate cover in non-binary maps. To test robustness NBSI-MS, regions Greenland France–Italy where interacts with highly diversified geographical ecosystems were examined. Data recorded by Landsat 5 TM, 8 OLI, Sentinel-2A MSI satellites used. performance was also compared against well-known Normalized Difference (NDSI), NDSII-1, S3, Water (SWI) methods evaluated Ground Reference Test Pixels (GRTPs) over non-binarized results. results show that achieved an overall accuracy (OA) ranging 0.99 1 kappa coefficient same range OA. precision assessment confirmed superiority method removing water surfaces relevant indices.