Geostatistical study of Italian submarine landslides

Submarine landslides are very large events occurring across both active and passive continental margin. They are sediment transport processes caused by submarine slope’s instability and the result of both internal structure changes and external dynamic conditions. The genesis and evolution of slope failure is controlled by different geological factors which may be considered as predisposing factors (e.g., seafloor morphology, lithology, type of sediment, presence of fluid or weak layers) and triggering factors (e.g., earthquakes). The Italian continental margins provide an excellent playfield to study submarine landslides because they have been identified and mapped in different morpho-tectonic and sedimentary contexts (e.g., accretionary prism, volcanic edifices, foredeep continental slope, upper slope in front of large deltas…). Between the 2007 and 2013, in the framework of MaGIC project funded by the Italian Civil Protection (DPC) a detailed mapping program of seafloor morphologies and features including landslides was carried out. The morphological features identified during the Magic Project will be used as a base to create a geodatabase of the Italian submarine landslides. This will include a careful reclassification of the different features, extraction of morphometric parameters and identification of triggering and preconditioning factors for the different case studies. This work has been funded by PNRR GEOSCIENCES IR project, and aim at use statistical methods to better classify the landslides that are relate than to different geological settings, and constrain the relation between different parameters and the triggers. When a relation will be established, we may build a step towards a better geohazard assessment, and may be define the likelihood of submarine landslide occurrences across the Italian Continental Margins, even in the areas when the landslides did not occurred yet. In fact, landslide hazard assessment requires the estimation of where, when (or how frequently) and how large a given landslide event may be. In marine realm this is very difficult because the cost of direct analysis is higher than in the subaerial counterpart. Therefore, a geostatistical approach on very large number of features could probably be the most realistic approach to the problem.