Automating nested watershed delineation over the world considering endorheic basins and islands

Global-scale nested watershed datasets are foundational for hydrological studies. However, their construction is not yet fully automated, making the process labor-intensive and time-consuming, primarily due to the lack of algorithms capable of encoding endorheic watersheds and islands. This study developed an innovative approach to automating the global delineation of nested watersheds accounting for endorheic basins and islands based on the widely used Pfafstetter coding system. Key advancements include combining adjacent endorheic watersheds into collections for systematic coding and developing automatic hierarchical delineation procedures for islands. This approach makes the delineation of endorheic basins and islands easy, reproducible, and more reasonable. Utilizing this approach, we developed a global nested watershed dataset, named NWEI (Nested Watershed dataset considering Endorheic basins and Islands), based on the Multi-Error-Removed Improved-Terrain (MERIT) Hydro datasets. Compared with the watershed boundaries reported from the Global Runoff Data Center (GRDC), 93% of the 4,301 watersheds globally exhibited relative area errors below 0.05. In comparison to the state-of-the-art HydroBASINS dataset, NWEI demonstrated improved detail and accuracy in delineated watershed boundaries. The automated nature of the proposed approach facilitated the rapid updating of NWEI using newer Digital Elevation Model (DEM) datasets.