Essential Elements for an Early Warning System to Detect Flowslides in Pyroclastic Deposits

Air-fall pyroclastic deposits on steep slopes in Campania (Southern Italy) are periodically sub-jected to rainfall-induced landslides that may evolve into catastrophic flowslides. To protect built-up areas, Early Warning Systems (EWSs) were implemented. Existing EWSs are essentially based on pluviometric thresholds or models which are unable to accurately monitor the physical phenomena which are responsible for flow-slide generation in pyroclastic deposits. Over the last 20 years, landslides with no evolution in flow-slide occurred in this area and the alarms generated by existing EWSs in the cases of rainfall were false and very expensive, thus, lowering population trust in EWSs. To improve the existing EWSs, two complex mod-els for pyroclastic soils from Cervinara and Sarno slopes are proposed in the paper, capable of simulating physical phenomena (such as, the saturation increase due to rainwater infiltration, mechanical degradation and undrained instability), control instability phenomena (landslide) and evaluate the post-failure evolution