Fluctuations of glacier extent in Lake Nam Co and Nyenchen Tanglha Mountains within a decade as detected by machine learning methods of image analysis for monitoring Lhasa terrane, Tibetan Plateau

Machine learning (ML) methods of satellite image analysis were applied in this study for geological-environmental analysis of glacier extent in Tibetan Plateau, China. The purpose of this work is to map the changes in glacier extent as a hydrological resource and its effects on land cover types using remote sensing data. A quantitative cartographic method of image analysis has been developed using ML algorithms and GRASS GIS scripts. Fluctuations of glacier extent are a key trigger for landscape dynamics in Tibetan Plateau. However, the links between spatio-temporal changes in snow and glacier, and associated land cover changes remain elusive. Six Landsat 8-9 multispectral satellite images covering Lhasa were evaluated. The images show fluctuation in glacier coverage from 2013 to 2023 with a 2-year gap between the observations, characterized by strong heterogeneities caused by climate changes. Glacier dynamics was evaluated for northern range of Nyenchen Tanglha Mountains and Lhasa Terrane, Tibetan Plateau, China. The results present an exploratory analysis of six images (on 2013, 2015, 2017, 2019, 2021 and 2023) for glaciological modelling using ML.