%0 Dataset
%T A dataset of boundary and length of glaciers in the Three–River Headwaters region during 2000–2020
%J National Cryosphere Desert Data Center
%I National Cryosphere Desert Data Center(www.ncdc.ac.cn)
%U http://www.ncdc.ac.cn/portal/metadata/e30ec739-e8bb-4c38-8e77-0f7ea124526b
%W NCDC
%A Zhang Cong
%A Zhao Jiawei
%A Chai Chunmei
%A He Haotian
%A Sun Meiping
%A Yao Xiaojun
%K Glacier;glacier length;Landsat image;deep learning;Three–river headwaters region
%X The Three–River Headwaters region is located in the southern part of Qinghai Province, in the heart of the Tibetan Plateau. It is the source of the Yangtze River, the Yellow River and the Lancang River (known as the Mekong River outside China), and is known as the “Water Tower of China”. The glaciers developed in this area are important water sources for many rivers. This study first used Landsat TM/ETM+/OLI/OLI–2 remote sensing images and DEM data, combined with deep learning models, to achieve the preliminary automatic extraction of glacier boundaries in the Three–River Headwaters region at different periods (2000, 2005, 2010, 2015 and 2020). Then, manual visual interpretation was used to check and revise each glacier boundary, and the Randolph Glacier Inventory 7.0 glacier inventory data and the second glacier inventory data of China were referred to for the improvement of relevant attribute information, obtaining the multi–period glacier boundary dataset of this region. Finally, the glacier centerline method was used to extract the multi-period glacier length data of this region. The dataset contains two files: (1) the glacier inventory data in the Three–River Headwaters region; (2) the glacier length data in the Three–River Headwaters region. This dataset can be used as the basic data for studies on glacier change in the Three–River Headwaters region and the response of glaciers to climate change.