Landsat 8 snow coverage data for the central and eastern sections of the Himalayas from 2013 to 2020

The the Himalayas, located at the southern edge of the Qinghai Tibet Plateau, is the highest mountain in the world. Its snow cover and its changes affect the regional climate, water resources and ecological environment. In order to better understand the snow cover changes in the region and overcome the problems of insufficient detail description in snow monitoring in complex mountainous areas using low - and medium resolution remote sensing, Landsat 8 clear sky condition data with a spatial resolution of 30m was used for snow monitoring. Due to the complex terrain in mountainous areas and the common overestimation problem of normalized snow cover index method in the Himalayas, support vector machine (SVM) classification method was applied to select snow cover feature training samples with different terrain, shadow and other conditions for snow cover classification. Combined with auxiliary data such as glacial lakes, surface water bodies and spatial neighborhood analysis for data post-processing, a 30 meter resolution snow cover dataset was constructed for the middle and eastern sections of the Himalayas from 2013 to 2020. By comparing Sentinel-2 high-resolution snow classification, the correlation coefficient of snow coverage in a 900 * 900 m grid is above 0.95, with a root mean square error of about 0.1%. The snow identification results constructed by the two have good consistency. The snow coverage dataset includes 607 snow data sets covering the central and eastern sections of the Himalayas over the past 8 years, mainly distributed during the winter snow season from October of the current year to April of the following year. This data set can provide verification and optimization basis for spatio-temporal and feature analysis of snow cover, and low and medium resolution snow cover data, and support research on climate change, water resources management and ecological benefits in the the Himalayas and downstream areas.