Improving hydrological model performance and parameter transferability for data-scarce Himalayan basins to inform water resource decisions

Sparse and unrepresentative rain gauge networks in Himalayan regions limit accurate spatial precipitation estimation, highlighting the need for satellite-based precipitation (SPP) for reliable hydrological simulations. The SWAT model was employed to calibrate hydrological process parameters in the Koshi Basin, considering varying drainage areas (DA) and the inclusion or exclusion of elevation bands (EB). These parameters were then transferred to the physically similar but distant Karnali Basin to simulate river discharge, and vice versa. This study underscores that (1) integrating gauge and SPP markedly improves hydrological model performance, particularly when EB is considered; (2) performance of spatial transferability depends on DA thresholds and the configurations of donor and receiver basins, emphasizing the need for basin-specific considerations; and (3) averaging the annual water balance results from models with varying DA thresholds reduces the uncertainty associated with spatial transferability, underscoring its potential as a valuable tool for hydrological estimation in data-scarce basins.