Earthquake-induced landslide scenarios for seismic microzonation: application to the Accumoli area (Rieti, Italy)

Abstract

Scenarios of earthquake-induced landslides are necessary for seismic microzonation (SM) studies since they must be integrated with the mapping of instability areas. The PARSIFAL (Probabilistic Approach to pRovide Scenarios of earthquake‐Induced slope FAiLures) approach provides extensive analyses, over tens to thousands of square kilometers, and is designed as a fully comprehensive methodology to output expected scenarios which depend on seismic input and saturation conditions. This allows to attribute a rating, in terms of severity level, to the landslide-prone slope areas in view of future engineering studies and designs. PARSIFAL takes into account first-time rock- and earth-slides as well as re-activations of existing landslides performing slope stability analyses of different failure mechanisms. The results consist of mapping earthquake-induced landslide scenarios in terms of exceedance probability of critical threshold values of co-seismic displacements (P[D ≥ Dc|a(t),ay]). PARSIFAL was applied in the framework of level 3 SM studies over the municipality area of Accumoli (Rieti, Italy), strongly struck by the 2016 seismic sequence of Central Apennines. The use of the PARSIFAL was tested for the first time to screen the Susceptibility Zones (ZSFR) from the Attention Zones (ZAFR) in the category of the unstable areas, according to the guidelines by Italian Civil Protection. The results obtained were in a GIS-based mapping representing the possibility for a landslide to be induced by an earthquake (with a return period of 475 years) in three different saturation scenarios (i.e. dry, average, full). Only 41% of the landslide-prone areas in the Municipality of Accumoli are existing events, while the remaining 59% is characterized by first-time earth- or rock-slides. In dry conditions, unstable conditions or P[D ≥ Dc|a(t),ay] > 0 were for 54% of existing landslides, 17% of first-time rock-slides and 1% of first-time earth-slides. In full saturation conditions, the findings are much more severe since unstable conditions or P[D ≥ Dc|a(t),ay] > 0 were found for 58% of the existing landslides and for more than 80% of first-time rock- and earth-slides. Moreover, comparison of the total area of the ZAFR versus ZSFR, resulted in PARSIFAL screening reducing of 22% of the mapped ZAFR.