Flash flood detection through a multi-stage probabilistic warning system for heavy precipitation events

摘要

The deadly combination of short to no warning lead times and thevulnerability of urbanized areas makes flash flood events extremelydangerous for the modern society. This paper contributes to flash floodearly warning by proposing a multi-stage warning system for heavyprecipitation events based on threshold exceedances within a probabilisticframework. It makes use of meteorological products at different resolutions,namely, numerical weather predictions (NWP), radar-NWP blending, and radarnowcasting. The system is composed by two main modules. First, a EuropeanPrecipitation Index based on a simulated Climatology (EPIC) andprobabilistic weather forecasts is calculated to pinpoint catchments at riskof upcoming heavy precipitation. Then, a Probabilistic Flash Flood GuidanceSystem (PFFGS) is activated at the regional scale and uses more accurateinput data to reduce the estimation uncertainty.The system is tested for a high flow event occurred in Catalonia (Spain) inNovember 2008 and results from the different meteorological input data arecompared and discussed. The strength of coupling the two systems is shown inits ability to detect areas potentially at risk of severe meteorologicalconditions and then monitoring the evolution by providing more accurateinformation with higher spatial-temporal resolution as the event approaches.