Local indexes, based on a nationwide threshold, for rainfall-induced landslides

The Italian territory is affected by rainfall-induced landslides, which can result in loss of lives and widespread damage. Empirical rainfall thresholds are widely used for predicting the occurrence of failures triggered by intense or prolonged rainfall at national and regional scale, even if they have empirical limitations for small geographical areas. This limitations depend on the size of the landslide catalogue available for the definition of statistically robust rainfall threshold over small areas (Peruccacci et al. 2012). Martinotti et al. 2017 theorized, designed and tested the Ensemble-Non-Exceedance Probability (E-NEP) algorithm, which exploits standard rainfall records obtained from rain gauges and a given rainfall threshold to quantitatively assess the landslide occurrence probability over time. Using the rainfall threshold defined by Peruccacci et al. 2017 for the entire Italian territory, they applied the E-NEP algorithm to analyse the landslides triggered during a time period of torrential rain between 1 and 6 September 2014 in the Gargano Promontory (Puglia) and found that the E-NEP metrics provided better diagnostics than the single metrics often used for landslide forecasting.