Morphometric and magnitude-frequency analysis of a regional inventory of submarine landslides in the tectonically-controlled Calabrian margin (Italy)

Abstract

Morphometric analysis of submarine landslide may be used for marine geohazard assessment. The collection of high-resolution multibeam bathymetry on the Calabrian Continental margin has allowed the identification and characterization of some 420 landslides over an area of 87 km2. The large number of landslides is due to the fact that this margin is tectonically and seismically extremely active, thus providing a favorable setting for predisposing and/or triggering processes and events such as high sedimentation rate, steep seafloor slopes, frequent ground shaking, etc. However it is important the statistical analysis is realized on very high-resolution data and using standardized interpretative and morphometric measuring procedures, in order to avoid bias due to spatial and vertical resolution or due to focusing on specific features or lack of data in very shallow or very deep water. In this work, we show the results of a detailed morphometric and magnitude-frequency analysis of a large database, extending from the very shallow water to the base of the continental slope. The aim of the study is to infer which parameters or relationship among parameters such as size, failure area/volume, depth and slope gradient, etc. might control submarine instability processes. Moreover, the study of the distribution of landslides at regional level (volcanic/non volcanic setting, within canyon or in open slope) can provide valuable information in order to infer the role of not-measurable parameters such as sedimentation rate, seismicity, dominating sedimentary processes, etc. Finally the zoning of distribution and characters of past landslides may represent a first-order assessment of areas where sedimentary instabilities, and their direct or indirect effects, are more likely to occur. This approach is essential for submarine landslides, where detailed in situ measurements are often not feasible due to high-cost, technical difficulties and the very large number of features to be investigated