Detecting on-set of Snow Melt and Elevation Based Distribution of Wet Snow Using EOS-04 SAR Data in Bhaga Sub-Basin

In the present study, snow and wet snow cover is derived using Landsat- 8 and EOS-04 at 30 m and 18 m resolution respectively. The study region used is Bhaga sub-basin in the Indus basin of western Himalayan region. HH polarization of Medium Resolution ScanSAR (MRS) from Aug-2022 to May-2023 was used using threshold-based technique assuming -3 dB as threshold from previous studies to arrive at temporal maps of wet snow cover product. For time-series plots of each individual scenes of MRS data, wet snow pixels were calculated and the melt onset was detected near first week of April in 2023 in Bhaga sub-basin using the present study. For each available scenes, mostly separated by 17 days, for each elevation band of 500 m interval, from 2000 m to 6500 m, were plotted to see elevation wise distribution as well as temporal distribution of wet snow pixels. It is found that as time approaches towards May (summer season), the number of wet snow pixels are drastically increased in consent with the summer conditions in Himalayas. To compare wet snow product derived from EOS-04 with the snow cover derived from Landsat-8, Normalized Difference Snow Index (NDSI) was computed using Landsat-8 data. The comparison was made at common coarser resolution and it was found that $4.99 \%$ of the total wet snow pixels were outside the snow cover region indicating commission error of the same magnitude with respect to NDSI-based snow cover. Overall, the method underestimates the wet snow cover, for which further study using parameters like slope of the terrain, detailed land use-land cover information and backscatter modelling can improve the calculation. Further, using dual polarization or multi polarization data from missions like NASA-ISRO dual-frequency SAR (NISAR), can enhance the diffentiability of wet snow from other surfaces.