Abstract
Snow is a crucial factor impacting many areas, including freshwater resources, irrigation, floods, and avalanches. The study aims to compare the performances of unsupervised classification methods namely Normalized Difference Snow Index (NDSI), Red/Shortwave- infrared (R/SWIR), and Near-infrared/Shortwave-infrared (NIR/SWIR) band ratio methods in snow cover classification while determining the temporal change and trend in snow cover area during the past three decades on Erciyes Mountain. The study analyzed the snow cover on Erciyes Mountain from 1988 to 2020 using Landsat 5 TM and Landsat 8 OLI data with a spatial resolution of 30 meters. Three unsupervised classification methods were applied to atmospheric-corrected Landsat datasets to obtain snow cover maps. The results indicate a declining trend in snow cover on Erciyes Mountain due to global warming, with a significant reduction of about 87% over the past 32 years. Among the three classification methods, NDSI demonstrated the highest accuracy and Kappa coefficient, making it the most suitable approach for snow cover mapping in mountainous regions. The findings can supply valuable contributions to literature and assist in better use of accurate snow cover classification methods in high mountainous regions.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call