Novel WRF-Hydro runoff simulation method considering optimal river network and underlying surface data

Abstract

Accurate river network and underlying surface data are crucial for runoff simulation, and the generation effect of river networks is directly affected by the resolution of digital elevation model (DEM). However, the distortion of elevation resampling with different resolutions changes the river network structure. In addition, the spatial resolution and timeliness of the default underlying surface data in the weather research and forecasting model (WRF) are poor, and thus cannot meet the application needs of accurate hydrological forecasting. To overcome this issue, this study proposes a runoff simulation method by combining optimal river network and underlying surface data. The method introduces multiple metrics to evaluate the simulation effect of the river network based on multiresolution topographic and geomorphic data, and the WRF-Hydro is used to simulate the runoff process under different topographic and geomorphic scenarios. The Yuehe River Basin is selected to perform the experiments, and results show that the river network discrepancy can well reflect the simulation effect of the river network in WRF-Hydro GIS. The river network discrepancy obtained by replacing the elevation data SRTM1 DEM is 1.24%, which demonstrates that the simulation effect of the river network is the best. In addition, the mean values of the determination coefficient (R²) and Nash–Sutcliffe efficiency coefficient (NSE) of the proposed method are increased by 5.1% and 16.58%, respectively, when compared with the existing methods. Such results demonstrate the good prospect of the runoff simulation method proposed in this study.