2026-10-12T06:18:29Z https://www.scidb.cn/oai

10.57760/sciencedb.j00001.01043 2024-09-27T15:37:37Z

2024-09-27 A dataset of glacier inventory in four Central Asian Countries during 2022–2023 doi:10.57760/sciencedb.j00001.01043 en The four Central Asian countries including Kazakhstan, Kyrgyzstan, Tajikistan and Uzbekistan located in inland arid regions are characterized by an average annual precipitation of approximately 100–400 mm and a severe water shortage. Glaciers developed in high mountainous areas are important water resources in Central Asia, and glacier meltwater contributes up to 40–60% of river recharge. Affected by global warming, glaciers in Central Asia are experiencing rapid retreat. The latest Randolph Glacier Inventory (RGI) 7.0 dataset, although providing detailed information on glaciers in Central Asia, primarily relies on remote sensing images from 2002, thereby failing to capture the current state of glaciers. In this study, a new glacier inventory for four Central Asia countries is compiled based on Landsat OLI images acquired between 2022 and 2023. Specifically, a combination of a convolutional neural network and a Transformer model is employed for automated glacier extraction. Subsequently, the manual visual revision method is used to meticulously verify the glacier boundaries individually. Finally, the relevant attribute information of glaciers is supplemented with reference to the RGI specifications. The results show that there are 22,662 glaciers in four Central Asian countries, with a total area of 12,873.33 km2. Tajikistan has the largest number and area of glaciers (11528 and 7624.48 km2), followed by Kyrgyzstan with 8916 glaciers and 4340.26 km2, Kazakhstan (1697 glaciers and 822.94 km2), and finally Uzbekistan (521 glaciers and 85.65 km2). This dataset provides valuable data support for researches on glacier change and scientific utilization of glacier water resources in Central Asia. remote sensing; deep learning; glacier inventory; Central Asia guo zhen yang Yao Xiaojun Cheng Yishuai Liu Yidan PUBLIC https://creativecommons.org/licenses/by/4.0/ dataset http://www.doi.org/10.11922/11-6035.csd.2024.0151.zh Science Data Bank