Multisensor monitoring and early warning of precipitation in mountain catchments prone to debris flow events

Five major debris flow events occurred in the Central Italian Alps in 2012 (Val Rabbia), 2018 (Rio Rotiano), 2020 (Torrente Vallaro), 2021 (Torrente Blé) and 2022 (Torrenti Re di Niardo e Cobello) were monitored with a multi-sensor and multi-system approach to assess their probability of occurrence and the potential of early warning systems. The five events caused one victim and severe damages to a camping site, buildings, road and energy infrastructures, structural flood control systems  and the environment and the measured point rainfall intensity had a frequency between 1 over 10 to 200 years, with the 2022 event being an exceptional outlier. Monitoring systems included two C-band radars, raingauges, IR and MW satellite sensors, water level sensors, video cameras with geophysical sensors (geophones and infrasound). Operational results of MOLOCH, a non-hydrostatic high-resolution  0.0113 degrees (1.25 km) meteorological model were analysed to assess the predictability of the events. The conducted analyses indicate the reliability of radar reflectivity, processed by considering also the delay in the atmosphere to ground rainfall induced by the falling velocity of raindrops, in capturing the timing and the spatial pattern of rainfall, although the Z(R) transformation still needs event-based or event-type calibration. Satellite images processed through the MASHA algorithm were effective in the synoptic-scale event of 2018 but still not always for some convective events.  The same happens for the MOLOCH meteorological models. The results, although promising, indicate that  the predictability of such debris flow events in mountain areas, on average, is still problematic and merging the different sources of information is needed for an effective early warning.