HWT-Hydro: Evaluation of Experimental Forecast and Nowcast Tools

During July 2014, the inaugural Hazardous Weather Testbed Hydrology Experiment was held at the National Weather Center in Norman, OK. Participating forecasters from the National Weather Service (17 in total) were provided access to a suite of flash flood and heavy rainfall prediction and monitoring tools from the Flooded Locations And Simulated Hydrographs (FLASH) project. These tools include Multi-Radar/Multi-Sensor quantitative precipitation estimates (QPE), short-range quantitative precipitation forecasts (QPF) from the High Resolution Rapid Refresh (HRRR) model, flash flood guidance, precipitation recurrence intervals, precipitable water, and outputs from distributed hydrologic models. The objectively analyzed skill of these tools for the duration of the experiment is presented. This skill is determined by comparing the tools to several sources of flooding observations, including NWS Storm Data, NWS Local Storm Reports, United States Geological Survey (USGS) stream gauges, Meteorological Phenomena Identification Near the Ground (mPING) reports, and surveys from the Severe Hazards Analysis and Verification Experiment (SHAVE). Objective results are compared to the subjective evaluations that were performed by the participants throughout the course of the experiment. Detailed analyses will include additional factors such as antecedent soil moisture conditions and geographic dependence of the tools' skill. These results will help guide future model developments in preparation for transitioning them to the NWS to improve and augment operational decision making with flash flood watches and warnings.