Earthquake-induced landslides susceptibility assessment of the Abruzzo Region (Central Italy): a geomorphological approach.

Earthquakes and consequent phenomena represent a major issue for human activities planning and fulfilment; within this framework, Earthquake-Induced Landslides (EILs) constitute major hazards, being often responsible for the greatest damages up to overshadowing those caused by the solely ground motion. Due to its geomorphological dynamics and strong seismicity, the Abruzzo Region (Central Italy) is severely affected by EILs, whose distribution results from the interaction between the seismic shaking and the local physiographic and geological-structural setting of the area. In this context, the present work focuses on the realisation of an EILs susceptibility map, following a heuristic approach combined with a statistical analysis, integrated using GIS technology. This approach leads to the identification of nine instability factors, including morphometric, lithological, geomorphological, and tectonic elements. These factors are analysed and assigned proper expert-based weights after the critical evaluation of literature data and available landslide inventories. Subsequently, they are combined into a preliminary susceptibility map wherein high/low numerical values correspond to a high/low propensity of the slope to fail. A statistical analysis is then executed on these values to derive the optimal number of classes by performing an unsupervised classification of preliminary susceptibility values. The total Within Clusters Sum of Squares (WCSS) and the Between Clusters Sum of Squares (BCSS) are hence computed, and the optimal number is derived by applying the elbow method. An absolute susceptibility scale is then introduced, with values ranging from the minimum to the maximum potential values of the area. Subsequently, statistics of clusters are analysed through violin plot diagrams and compared to classes of the absolute scale; in this phase, an algorithm is also applied to achieve the best differentiation among classes. The final map is created by grouping preliminary values in seven susceptibility classes ranging from very low to very high. The stepwise approach here presented has been applied to the whole Abruzzo Region, with ongoing specific site investigations in the areas severely affected by the 2009 L’Aquila earthquake and the 2016-2017 Central Italy seismic sequence. The applied methodology could constitute a scientific tool to better define situations that potentially lead to hazards following an earthquake and, consequently, to develop sustainable territorial management, loss-reduction measures, and post-earthquake reconstruction plans.