Land use-driven parameter regionalization for distributed hydrological modeling in data-scarce basins: A case study of Mekong River Basin

Abstract

[Objective] To address the long-standing challenge of parameter identification in hydrological modeling for data-scarce regions within transnational water conservancy projects, a parameter transfer method for a raster-based distributed hydrological model based on land use types is proposed. [Methods] Using land use types as basic units, a “land use-parameter” mapping relationship was constructed to form a structured parameter lookup table. This approach overcame the dependence of traditional parameter regionalization on gauging station density, enabling robust cross-basin parameter transfer under similar underlying surface conditions. The Cambodian section of the Mekong River Basin was selected as a case study. A land use type-parameter lookup table was established through zonal calibration in a reference river basin, which was then transferred to the target river basin for validation. [Results] In the reference river basin(Kompong Cham station), the Nash-Sutcliffe efficiency(NSE) values during the calibration and validation periods were 0. 87 and 0. 86, respectively, with relative errors(RE) in total runoff volume of 5. 71% and 8. 61%. After the parameters were transferred to the target river basin, the NSE reached 0. 87 at the Thnous Loung station on the Das River with an RE of 8. 37%, and the NSE was 0. 89 at the Kralanh station on the Sreng River with an RE of 7. 01%. The accuracy was significantly higher than that of the instantaneous unit hydrograph method(NSE of 0. 76 and 0. 68) and the antecedent precipitation index method(NSE of 0. 32 and 0. 74), demonstrating superior capability in simulating flood peak processes and maintaining water balance. [Conclusion] The proposed method realizes physically based cross-basin parameter transfer using land use as a carrier, fills the gap in hydrological model parameters in data-scarce regions, and provides an effective reference for process simulation in distributed hydrological modeling.

Keywords

parameter regionalization / ungauged basins / land use homogeneity / distributed hydrological model / data-scarce regions / hydrological non-stationarity / Mekong River Basin / climate change