Using Multiple Precipitation Inputs for Flash-Flood Forecasting in Semiarid Environments

The WRF-hydro model was calibrated and evaluated for domains in Israel and in the United States to serve as the basis for an operational flood forecasting system. Previous studies assessed the advantages and limitations of one-way versus two-way coupled modeling systems for flood prediction; their results highlight the sensitivity of hydrological responses to different sources of precipitation data, and less so, to hydrologic model formulation. In most cases, when using observed precipitation data, the calibrated model closely simulated observed hydrographs. The two-way coupled WRF/WRF-Hydro modeling system improved both the precipitation and hydrological simulations as compared to the one-way WRF simulations. Findings from those studies suggested that the use of two-way atmospheric-hydrological coupling has the potential to improve the forecast of precipitation and, therefore, to further improve hydrological forecasts for early flood warning applications.

The objectives of this current study were to examine the advantages of the usage of various precipitation input for running a hydrological ensemble. We focused on two sources of precipitation forecast data: high-resolution coupled WRF forecasts, as well as offline quantitative precipitation nowcasts (QPN) produced by an operational, country-scale nowcasting system to serve as “synthetic radar-derived” precipitation inputs to WRF-Hydro. The nowcasting system leverages both traditional radar observations as well as proprietary data from unique data sources such as cellular microwave backhaul links.

Preliminary results demonstrate the advantage of using an ensemble of meteorological forcing from various sources for hydrological predictions. We will further demonstrate the efficacy of utilizing a coupled WRF-Hydro/nowcasting system to produce short-fuse forecasts with application to flood forecasting through selected cases studies of recent events in both Israel and in the United States.