Multiplatform hydrometeorological analysis of a flash flood event

This chapter presents an integrated platform to simulate and predict flash floods. The platform consists of various state-of-the-art components, including remote sensing data assimilation and advanced atmospheric and hydrological models. The Local Analysis and Prediction System (LAPS) and the Chemical Hydrological Atmospheric Ocean wave System (CHAOS), including the Advanced Weather Research and Forecasting (WRF-ARW) model and the WRF-Hydro hydrological model, have been utilized. The efficiency of the platform was assessed in a flash flood event affected the suburban area of Mandra in western Attica, Greece. WRF-Hydro simulated the flood using forcing data, including four different precipitation estimations, obtained from CHAOS forecasts (CHAOS-hydro), the National Observatory of Athens X-band dual-polarization (XPOL) weather radar (XPOL-hydro), the Global Precipitation Measurement (GMP)/Integrated Multi-satellitE Retrievals GPM/IMERG-hydro, and LAPS nowcasts (LAPS-hydro). Comparisons with site-detailed postsurveys of flood extent revealed that XPOL-hydro well predicted flooded area with a 68% probability of detection (POD) while CHAOS-hydro and GPM/IMERG-hydro presented smaller POD, 43% and 30%, respectively. The nowcasting approach assessment showcased that 3 h before the maximum flood extent, LAPS-hydro yielded a 49% POD while it showed 61% and 67% POD 2 and 1 h before, respectively, comparable to POD resulted by XPOL-hydro.