Regionalization of catchment hydrological model parameters for global water resources simulations

Abstract Parameter regionalization of hydrological models is one the most commonly used methods for prediction over ungauged catchments. Although there were many regional studies, no clear conclusion on best-performing method global modelling. The objective this study to determine an appropriate global-scale scheme (GSRS) To end, performance five with two different average options, weighting approaches, and seven efficiency thresholds (i.e. Kling-Gupta (KGE) values measure model performances) was compared thousands catchments based four conceptual models. Results nine from Global Earth Observation Integrated Water Resource Assessment (EartH2Observe) project selected validate accuracy GSRS in estimating runoff. results show that: (1) Spatial proximity Inverse Distance Weighting output option offers best result when using KGE ? 0.5 as threshold all models, (2) regionalization-based simulation schemes (RGHSs), i.e. proposed combining consistently performs better than EartH2Observe estimation runoff catchments, varying degrees improvement median, upper lower quartiles, whiskers each metric, (3) long-term annual water resources estimated by RGHSs range between 42,592 46,810 km3/yr.