Flash Flood Warning Technology and Metrics

The Flash Flood Monitoring and Prediction (FFMP) system is providing forecasters with rainfall and flash flood guidance information on the scale of small basins. As FFMP gets accepted into the National Weather Service forecast environment, a number of issues are being considered with renewed interest such as the accuracy of radar-derived precipitation and the ability of flash flood guidance to truly represent the processes in small, fast-response basins. Recent flash flood cases involving a number of different hydrologic situations have tested the capabilities of FFMP and pointed to strengths and weaknesses of the current program. Training and the evolution of FFMP functionality will address weaknesses, particularly those associated with remotely-sensed precipitation and flash flood guidance. FFMP shows tremendous promise for rapidly-evolving situations, such as those when concurrent severe weather may distract forecasters from the developing flood threat, particularly in fast-response basins such as those in urban areas and fire-burn regions.

Flash flood warning lead time metrics are receiving increased attention within the NWS. As routine use FFMP becomes commonplace, warnings will potentially show greater specificity in time and space. At the same time, the more vague warnings covering multiple counties for many hours should decrease. Thus, a situation can arise where more specific and accurate warnings do not necessarily result in longer lead times. FFMP will force the interpretation of metrics for the flash flood phenomenon which, unlike severe weather, has been characterized by somewhat subjective definition and verification criteria.

The Cooperative Program for Operational Meteorology, Education and Training (COMET®) at the University Corporation for Atmospheric Research (UCAR) has been involved with cutting-edge, graduate-level training in hydrometeorology for more than a decade. The newly developed flash flood workshop will assist FFMP users with understanding its input, its products, and how to optimize its functionality with some of the user-interactive functions.