Abstract: The Qiangtang endorheic basin, the largest endorheic basin in the Tibetan Plateau, is experiencing a rapid evolution of its lake system structure and water quantity owing to climate change. Considering the impacts of climate change on the boundaries of endorheic units, we propose a novel delineation algorithm introducing the Joint Elevation-Area Threshold (JEAT) to address delineation issues caused by climate change. Firstly, by observing the reorganizations and fluctuations in lake levels in the Qiangtang endorheic zone, we determined that the JEAT was 10 m and 50 km², respectively. Next, a total of 163 endorheic units with reasonable contribution area and permanent boundaries were identified. The results demonstrated that the algorithm is suitable for extracting endorheic units on a large scale, and the extraction accuracy is superior to that of other existing algorithms and datasets. Subsequently, 163 endorheic units were categorized into five main types based on weak watershed type, with the upstream-downstream connected type (Type Ⅱ) and alpine type (Type Ⅳ) being the main ones. Pit type (Type Ⅴ) watersheds tended to mostly coexist with other types, exhibiting a more intense response to the connectivity characteristics of large depressions at different periods.