Quantifying the Performances of Simplified Physically Based Landslide Susceptibility Models: An Application Along the Salerno-Reggio Calabria Highway Open image in new window
Landslides are one of the most dangerous natural hazards in the world causing fatalities, destructive effects on properties, infrastructures, and environment. A correct evaluation of landslide risk is based on an accurate landslide susceptibility mapping that will affect urban planning, landuse planning, and infrastructure designs. Great effort has been devoted by the scientific community to develop landslide susceptibility models. Only few studies have been focused on defining accurate procedures for model selection, assessment, and inter-comparison. In this study we applied a methodology for objectively calibrate and compare different landslide susceptibility models in a framework based on three steps. The first step involves the automatic model parameter calibration based on different objective functions and the comparison of the models results in the ROC plane. The second step involves the intercomparison of a set of model performance indicators in order to exclude objective functions that provide the same information. Finally the third step involves a model parameter sensitivity analysis to understand how model parameter variations affect the model performances. In this study the three-step procedure was applied to compare two different simplified physically based landslide susceptibility models along the highway Salerno-Reggio Calabria in Italy. The model M2, able to consider the spatial variability of the soil depth respect to the model M1, coupled the distance to perfect classification index provided the most accurate result for the study area.
KeywordsLandslide susceptibility analysis Model performances evaluation ROC
This research was funded by the PON project no. 01_01503 “Integrated Systems for Hydrogeological Risk Monitoring, Early Warning and Mitigation Along the Main Lifelines”, CUP B31H11000370005, within the framework of the National Operational Program for “Research and Competitiveness” 2007–2013.
- Brabb EE (1984) Innovative approaches to landslide hazard and risk mapping. In: Proceedings of the 4th International Symposium on Landslides, vol 1. Canadian Geotechnical Society, Toronto, Ontario, Canada, 16–21 September, pp 307–324Google Scholar
- Cascini L, Bonnard C, Corominas J, Jibson R, Montero-Olarte J (2005) Landslide hazard and risk zoning for urban planning and development. In: Landslide risk management. Taylor and Francis, London, pp 199–235Google Scholar
- Cascini L, Ciurleo M, Di Nocera S (2016) Soil depth reconstruction for the assessment of the susceptibility to shallow landslides in fine-grained slopes. Landslides, 1–13Google Scholar
- Corominas J, Van Westen C, Frattini P, Cascini L, Malet JP, Fotopoulou S, Catani F, Van Den Eeckhaut M, Mavrouli O, Agliardi F, Pitilakis K (2014) Recommendations for the quantitative analysis of landslide risk. Bull Eng Geol Environ 73(2):209–263Google Scholar
- Dietrich WE, Bellugi D, Real De Asua R (2001) Validation of the shallow landslide model, SHALSTAB, for forest management. In: Wigmosta MS, Burges SJ (eds) Land use and watersheds: human influence on hydrology and geomorphology in urban and forest areas. American Geophysical Union, Washington, D.C. doi: 10.1029/WS002p0195 Google Scholar
- Formetta G, Capparelli G, Rigon R, Versace P (2014) Physically based landslide susceptibility models with different degree of complexity: calibration and verification. In: Ames DP, Quinn NWT, Rizzoli AE (eds) International Environmental Modelling and Software Society (iEMSs). 7th International Congress on Environmental Modelling and Software, San Diego, CA, USA, 15–19 June 2014. http://www.iemss.org/sites/iemss2014/papers/iemss2014_submission_157.pdf
- Jolliffe IT, Stephenson DB (eds) (2012) Forecast verification: a practitioner’s guide in atmospheric science. Wiley, University of Exeter, UKGoogle Scholar
- Kennedy J, Eberhart R(1995) Particle swarm optimization. In: Proceedings of the IEEE International Conference on Neural Networks, 1995, vol 4. IEEE, PerthGoogle Scholar
- Murdoch DJ, Chow ED (1996) A graphical display of large correlation matrices. Am Stat 50:178–180Google Scholar
- Naranjo JL, van Westen CJ, Soeters R (1994) Evaluating the use of training areas in bivariate statistical landslide hazard analysis: a case study in Colombia. ITC J 3:292–300Google Scholar
- Varnes DJ (1984) Landslide hazard zonation: a review of principles and practice. No. 3Google Scholar