A probabilistic hydrometeorological forecasting chain for operational warning procedures in Marche Region

An operational hydrometeorological forecasting chain has been developed and implemented to support the civil protection activities of the Multi-risk Functional Centre of Marche Region. The chain includes a distribute hydrological model (Continuum) feed by observed meteorological variables from different sources: ground stations, weather radars and satellites (snow cover from Sentinel 2, MODIS and HSAF, and soil moisture from ASCAT). The precipitation field is obtained using a merging algorithm that fuse rain gauge data and weather radars (Modified Conditional Merging). In the forecasting configuration the chain ingests weather forecast (QPF and other meteorological variables) from different sources producing an ensemble of streamflow forecast (COSMO-LAMI 5 km, WRF 1.5 km, HRES 9 km). An interesting feature of the hydrometeorological forecasting systems on small and medium catchments is the possibility to feed the model with quantitative prediction issued by expert forecasters. They consider the meteorological uncertainty by using the output of various meteorological models combined with their knowledge of the territory, of its climatic peculiarities and on the meteorological situation to give their best quantitative estimate of expected precipitation amount and maxima. Part of the forecasting chain is an interactive tool that allows to create different scenarios to mitigate floods by acting in advance on some of the dams present in the area and used for hydropower production and water supply. Modelling upgrade, an activity of the STREAM project, was financed by the European Regional Development Fund  inside the  Interreg IT-HR programme.  In this work the performances of the forecasting chain will be presented on a set of several past events.