Implementation of a vector-based river network routing scheme in the community WRF-Hydro modeling framework for flood discharge simulation

Continental-scale flood discharge modeling requires a high level of efficiency and flexibility. To this end, this study documents the implementation and application of a vector-based river routing model in the community WRF-Hydro modeling framework. Using Hurricane Ike as a case study, the hybrid vector–grid modeling framework's sensitivity to the land grid resolution and the coupling interface is assessed. Results show the model is more sensitive to the coupling interface than the grid resolution, and a 1-km land grid with an area-weighted coupling interface exhibits the optimal simulation results. A geographic information system (GIS) based approach is adopted to improve the regional representativeness of the flow travel time estimation. The model's computational efficiency and complexity are compared to a grid-based routing scheme, demonstrating its advantages for large-scale “offline” hydrological applications with GIS-supported features. Trade-offs between the modeling efficiency and complexity are then discussed to inform future large-scale flood prediction applications. •A vector-based river routing model (RAPID) is implemented as a new routing option in WRF-Hydro.•A real-world application is presented to simulate inland flood discharges for Hurricane Ike.•The hybrid model shows higher sensitivity to grid-to-vector coupling than to land grid resolution.•A GIS-based approach to estimating flow travel time is adopted for the regional simulation.•The framework is computationally efficient but now only supports “offline” hydrological applications.