Abstract: There are numerous glaciers and glacial lakes in the Koshi River basin. The coverage rate of glaciers reaches 9.78% and the number of glacial lakes is 221. The distribution and change of glacial lakes in this region are the most important factors for predicting the disaster of glacial lake outburst floods (GLOFs), and can also reflect the changes of climate and glacial water resources in alpine frigid areas to a certain extent. The dataset includes the vectorized outlines of glacial lakes with an area of no less than 0.05 km2 during six periods in the Koshi River basin, that is, 1960s, 1970s, 1980s, 1990s, 2000s and 2010s, which were acquired by using multi-source remote sensing image (Corona, Landsat MSS/TM/ETM+/OLI) and topographic maps to conduct a visual interpretation. We controlled the accuracy of interpretation within one pixel. Lying between 26.86°–29.14°N and 85.39°–88.95°E, the study area is composed of several sub-basins including Indrawati, Sun Koshi, Tama Koshi, Likhu, Dudh Koshi, Arun and Tamor. The dataset has a long time series, type and geometric properties of glacial lakes, which can be used as the basic data for assessing the temporal-spatial changes of glacial lakes and the changes in water volume of glacial lakes in the Koshi River basin. In addition, it can also be used as an important basis for the risk assessment of GLOFs, the changes of water resources, climate change and the relationship between glacial lakes and glaciers in this region.
Keywords: glacial lake; Koshi River; visual interpretation; Tibetan Plateau