Using numerical weather prediction and high-performance hydrologic-hydraulic models for flood forecasting

Flood forecasting plays a critical role in inundation warning and emergency response; however, issuing such forecasting in detail (e.g., pixel-wise inundation information) usually requires massive computational resources. Here we present an attempt that couples the Weather Research and Forecasting (WRF) model, WRF-Hydrological modeling system (WRF-Hydro), with the Two-dimensional Runoff Inundation Toolkit for Operational Needs (TRITON). By introducing the high-performance modeling framework, we wish to preserve and utilize the physics in hydrologic-hydraulic modeling as much as possible to provide flood forecasts. Our experiment in the southwestern region of Taiwan shows some promising results that correctly identify locations and depths of historical flood events at lead times of 6 to 12 hours. We expect that the coupled modeling framework, in addition to a better understanding of the complex mechanisms and interactions between hydrometeorology and flood disasters, can put a more sophisticated, real-time flood forecast in practice.