Mauro Rossi, Michele Santangelo, Massimiliano Alvioli, Ivan Marchesini, Mauro Cardinali, Francesco Bucci, Federica Fiorucci, 2019, Composite landslide susceptibility maps, SGI 2019, Parma, Italia, 16/09/2019,
URL: http://www.cnr.it/prodotto/i/429795

Landslides are heterogeneous natural processes. Given the significant variability of the different landslide types and in particular of their dimensional characteristics, style of movement and triggering mechanisms, multiple models and approaches have been developed in the literature. At present, no single model or modelling approach is applicable to all the different landslide types. This also limits the prediction of the spatial occurrence of landslides achieved through susceptibility models which aims at estimating the probability of the spatial landslide occurrence. Such models divide and classify a territory on the basis of its propensity to specific landslide types, using relationships linking the occurrence of landslides with their conditioning factors. In this work, we identify and apply a methodology for the combined modelling of susceptibility posed by different landslide types. The methodology initially envisages the modelling of the susceptibility for the individual landslide types using differentiated approaches, calibration, evaluation and application phases and subsequently their combination. In the modelling, the pixel is used as the base mapping unit, while the slope unit is used as a mapping unit for a subsequent spatial susceptibility aggregation. The result maps, here defined as "composite susceptibility maps", subdivide a territory according to its propensity to failure of one or more landslide types. The proposed model was applied as part of the Project "Paesaggi Sicuri", funded by the Italian Ministry of Cultural Heritages and Activities (MIBACT), in a portion of the UNESCO site "Porto Venere, Cinque Terre e Isole" characterized by the occurrence of different landslides types, namely slides, falls and flows. The derived composite susceptibility map allowed a more realistic analysis of the geo-hydrological vulnerability being able to consider jointly the different landslides types occurring in the study area.

Data from https://intranet.cnr.it/people/