Probabilistic analysis is gaining more attention in the field oflandslide hazard assessment, due to the possibility of taking into account estimation uncertainties and spatial variability ofgeological, geotechnical, geomorphological and seismological parameters. In this paper, an implementation ofa simple approach to derive probabilistic earthquake triggered landslide hazard maps is described. The method is based on the simplified Newmark slope stability model, applied on a pixel-by-pixel basis, which fully integrates into current GIS computational environments. Uncertainties and fluctuations in input parameters are considered by treating these quantities as statistical distributions. Various probability density functions can be simulated through Monte Carlo techniques on a pixel-by-pixel basis, and the simulated samples are retained through all the computing steps. This allows the resulting quantities to be cast into probabilistic hazard maps, without restrictions about the symmetry or the mathematical complexity ofthe underlying distributions. First results on a test landslide site in Southern Italy show good performances for realistic landslide hazard zonation. The simplicity ofthe adopted framework allows the current approach to be easily expanded and improved the current approach. r 2002 Elsevier Science Ltd. All rights reserved.
Probabilistic modeling of uncertainties in earthquake-induced landslide hazard assessment
Refice A;
2002
Abstract
Probabilistic analysis is gaining more attention in the field oflandslide hazard assessment, due to the possibility of taking into account estimation uncertainties and spatial variability ofgeological, geotechnical, geomorphological and seismological parameters. In this paper, an implementation ofa simple approach to derive probabilistic earthquake triggered landslide hazard maps is described. The method is based on the simplified Newmark slope stability model, applied on a pixel-by-pixel basis, which fully integrates into current GIS computational environments. Uncertainties and fluctuations in input parameters are considered by treating these quantities as statistical distributions. Various probability density functions can be simulated through Monte Carlo techniques on a pixel-by-pixel basis, and the simulated samples are retained through all the computing steps. This allows the resulting quantities to be cast into probabilistic hazard maps, without restrictions about the symmetry or the mathematical complexity ofthe underlying distributions. First results on a test landslide site in Southern Italy show good performances for realistic landslide hazard zonation. The simplicity ofthe adopted framework allows the current approach to be easily expanded and improved the current approach. r 2002 Elsevier Science Ltd. All rights reserved.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.