Application and validation of bivariate GIS-based landslide susceptibility assessment for the Vitravo river catchment (Calabria, south Italy)

Abstract

The Calabria (Southern Italy) region is characterized by many geological hazards among which landslides, due to the geological, geomorphological, and climatic characteristics, constitute one of the major cause of significant and widespread damage. The present work aims to exploit a bivariate statistics-based approach for drafting a landslide susceptibility map in a specific scenario of the region (the Vitravo River catchment) to provide a useful and easy tool for future land planning. Landslides have been detected through air-photo interpretation and field surveys, by identifying both the landslide detachment zones (LDZ) and landslide bodies; a geospatial database of predisposing factors has been constructed using the ESRI ArcView 3.2 GIS. The landslide susceptibility has been assessed by computing the weighting values (Wi) for each class of the predisposing factors (lithology, proximity to fault and drainage line, land use, slope angle, aspect, plan curvature), thus evaluating the distribution of the landslide detachment zones within each class. The extracted predisposing factors maps have then been re-classified on the basis of the calculated weighting values (Wi) and by means of overlay processes. Finally, the landslide susceptibility map has been considered by five classes. It has been determined that a high percentage (61%) of the study area is characterized by a high to very high degree of susceptibility; clay and marly lithologies, and slope exceeding 20°in inclination would be much prone to landsliding. Furthermore, in order to ascertain the proposed landslide susceptibility estimate, a validation procedure has been carried out, by splitting the landslide detachment zones into two groups: a training and a validation set. By means of the training set, the susceptibility map has first been produced; then, it has been compared with the validation set. As a result, a great majority of LDZ-validation set (85%) would be located in highly and very highly susceptible areas. The predictive power of the model is considered reliable, since more than 50% of the LDZ fall into 20% of the most susceptible areas. The reliability of the susceptibility map is also suggested by computing the SCAI index, true positive and false positive rates; nevertheless, the most susceptible areas are overestimated. As a whole, the results indicate that landslide susceptibility assessment based on a bivariate statistics-based method in a GIS environment may be useful for land planning policy, especially when considering its cost/benefit ratio and the need of using an easy tool.

Mass movements inventory map of the Rubbio stream catchment (Basilicata – South Italy)

“…In areas where Flysch lithotypes crop out, the landslide type is closely linked to the prevailing lithology and the degree of fracturing. Moreover, a chaotic and disorganized tectonic assemblage (joints, faults, foliation and bedding), further promotes mass movements (Conforti et al, 2012a). Structural discontinuities provide high permeability and deep weathering of rocks, which have a strong influence on slope instability.…”

“…Catenacci, 1992;Catani, Casagli, Ermini, Righini, & Menduni, 2005;Conforti, Robustelli, Muto, & Critelli, 2012a;Conforti, Pascale, Robustelli, & Sdao, 2014a;Conforti, Muto, Rago, & Critelli, 2014b;Guzzetti, 2000;Luca, Robustelli, Conforti, & Fabbricatore, 2011). In the Basilicata region (southern Italy), mass movements cause considerable damage to human activity and property every year (Bentivenga, Palladino, & Caputi, 2012;Caniani, Pascale, Sdao, & Sole, 2008;Conforti, Pascale, Pastore, Pepe, Sdao & Sole, 2012b;Conforti, Pascale, Pepe, Sdao, & Sole, 2013;Dal Sasso et al, 2014;Pascale, Sdao, & Sole, 2010;Pascale et al, 2013;Sdao & Simeone, 2007).…”

Forward logistic regression for earth-flow landslide susceptibility assessment in the Platani river basin (southern Sicily, Italy)

“…) Binary logistic regression (BLR) Atkinson and Massari (1998), Ayalew and Yamagishi (2005), Bai et al (2010), Can et al (2005), Carrara et al (2008), Chauan et al (2010), Conforti et al (2012), Dai and Lee (2002), Davis and Ohlmacher (2002), Erener and Düzgün (2010), Mathew et al (2009), Nandi and Shakoor (2009), Nefeslioglu et al (2008, Ohlmacher and Davis (2003), Van den Eckhaut et al (2006 Classification and regression trees (CART) Felicísimo et al (2012), Vorpahl et al (2012) Artificial neuronal networks (ANN) Aleotti and Chowdhury (1999), Ermini et al (2005), Lee et al (2004), Pradhan and Lee (2010) Original Paper exploited to compare the fitting of the model having only the constant term (all the β p are set to 0) with the fitting of the model that includes all the considered predictors with their estimated non-null coefficients so as to verify if the increase in likelihood is significant; in this case, at least one of the p coefficients is to be expected as different from zero (Hosmer and Lemeshow 2000). By exponentiating the β's, odds ratios (OR) for the independent variables are derived: these are measures of association between the independent variables and the outcome of the dependent, and directly express how much more likely (or unlikely) it is for the outcome to be positive (unstable cell) for unit changing of the considered independent variable.…”

Geomorphological map of the Crotone Province (Calabria, South Italy)

“…Landforms created by mass movement are widespread in the study area, and locally play an important role in the present-day landscape evolution, leading to an increase in hazards (Iovine et al, 2010;Conforti et al, 2011) (Figure 5a). A total of 379 landslides have been recognized and they occupy a surface of 121 km 2 , corresponding to more than 7% of the whole study area.…”

“…A total of 379 landslides have been recognized and they occupy a surface of 121 km 2 , corresponding to more than 7% of the whole study area. Some active landslides represent a reactivation of pre-existing dormant landslides (Conforti et al, 2011). Landslide distribution is closely related to the geological and geomorphological framework, including structural factors (alternating weak and hard rocks), tectonic setting, local relief and fluvial downcutting.…”