Exploring the effect of absence selection on landslide susceptibility models: A case study in Sicily, Italy

Abstract

A statistical approach was employed to model the spatial distribution of rainfall-triggered landslides in two areas in Sicily (Italy) that occurred during the winter of 2004-2005. The investigated areas are located within the Belice River basin and extend for 38.5 and 10.3 km2, respectively. A landslide inventory was established for both areas using two Google Earth images taken on October 25th 2004 and on March 18th 2005, to map slope failures activated or reactivated during this interval. Geographic Information Systems (GIS) were used to prepare 5 m grids of the dependent variables (absence/presence of landslide) and independent variables (lithology and 13 DEM-derivatives). Multivariate Adaptive Regression Splines (MARS) were applied to model landslide susceptibility whereas receiver operating characteristic (ROC) curves and the area under the ROC curve (AUC) were used to evaluate model performance. To evaluate the robustness of the whole procedure, we prepared 10 different samples of positive (landslide presence) and negative (landslide absence) cases for each area. Absences were selected through two different methods: (i) extraction from randomly distributed circles with a diameter corresponding to the mean width of the landslide source areas; and (ii) selection as randomly distributed individual grid cells. A comparison was also made between the predictive performances of models including and not including the lithology parameter.The models trained and tested on the same area demonstrated excellent to outstanding fit (AUC > 0.8). On the other hand, predictive skill decreases when measured outside the calibration area, although most of the landslides occur where susceptibility is high and the overall model performance is acceptable (AUC > 0.7). The results also showed that the accuracy of the landslide susceptibility models is higher when lithology is included in the statistical analysis. Models whose absences were selected using random circles showed a significantly better performance when learning and validation samples were extracted from the same area; whereas, conversely, no significant difference was observed when testing the models outside the training area

Exploiting Maximum Entropy method and ASTER data for assessing debris flow and debris slide susceptibility for the Giampilieri catchment (north‐eastern Sicily, Italy)

“…As regards the adopted statistical technique, MaxEnt performed in susceptibility modelling with the same skill and accuracy of frequently adopted presence/absence methods, without requiring complex and time consuming negative multi-extraction routines before the actual analyses (e.g., Costanzo et al 2014;Lombardo et al, 2014;Conoscenti et al 2016). …”

“…According to some published papers (Aronica et al, 2012;Peres and Cancelliere, 2014;Penna et al, 2014;Schilirò et al, 2015) the regolithic layer is characterised on average by a Unit Weight of 20-25 kNm -3 , a porosity of near 35%, a permeability of 2 X 10 -5 m s -1 , a cohesion ranging from 0 to 6 kPa and an internal friction angle of 35 degrees. The 2009 event has been the study case of a number of papers dealing with either hydrologic, physical or stochastical modelling for shallow landslide susceptibility assessment (e.g., Aronica et al, 2012;Lombardo et al, 2014;Peres and Cancelliere, 2014;Cama et al, 2015;Lombardo et al, 2015;Schilirò et al, 2015, Cama et al, 2016, as well as of specific studies focused on mapping techniques (Ardizzone et al, 2012;Mondini et al, 2011;Ciampalini et al, 2015), rainfall thresholds (Gariano et al, 2015), land-use change effect or tectonic control (Goswami et al, 2011;De Guidi and Scudero, 2013).…”

“…Besides, it is worth to mention that the conditions for landslide activation are to be searched also into cells which are actually located outside the failure zone (typically uphill from crowns; Rotigliano et al, 2011). The use of LIPs has been verified as an effective tool in assessing landslide susceptibility in previous applications to the Messina 2009event (Lombardo et al, 2014Cama et al, 2015;Lombardo et al, 2015, Cama et al 2016. The final inventory consisted of 824 and 297 flow and slide LIPs, respectively.…”

The propagation of inventory-based positional errors into statistical landslide susceptibility models

“…Thus, statistically discriminating landslide locations from non-landslide locations becomes increasingly difficult since non-landslide observations are regularly represented by random spatial observations (e.g. Conoscenti et al, 2016;Goetz et al, 2015a). The observed decreasing predictive performances provided quantitative evidence for this assumption (CV and SCV in Fig.…”

The influence of systematically incomplete shallow landslide inventories on statistical susceptibility models and suggestions for improvements

“…samples not used to train the model), they are frequently considered to summarize the capability of a predictive model to identify landslide-prone areas (Chung and Fabbri 2003;Remondo et al 2003;Beguería 2006;Guzzetti et al 2006b). Such inventory-based metrics are also taken into account to evaluate different classification algorithms (Goetz et al 2015;Steger et al 2016a) or the utility of specific predictor combinations (Iovine et al 2014;Conoscenti et al 2016), the spatial transferability of modelling results (Petschko et al 2014b;Lombardo et al 2014), the influence of sample sizes (Petschko et al 2014b;Heckmann et al 2014;Hussin et al 2016), the effect of sampling strategies (Regmi et al 2014;Conoscenti et al 2016;Hussin et al 2016) or the impact of data set qualities (Galli et al 2008;Fressard et al 2014).…”