GIS-Based Probabilistic Slope Stability Assessment Using Shear Band Propagation

Abstract

Abstract This paper describes a GIS-based application of shear band propagation to model slope instability that combines rapid analysis over large areas using pixel-based analysis with an approach capable of modelling the observed slope failures. Slope instability modelling can be performed over large study areas through application within a GIS. This approach typically involves calculating the limit equilibrium factor of safety using an infinite slope geomechanical model for a dense grid of pixels covering the study area. The result is a quantitative and spatially resolute estimate of susceptibility to slope failure representative of the seabed gradient, sediment strength and density, and other conditioning or triggering factors. For input to a spatial risk assessment, factor of safety can be converted into annual probability of failure using a Monte Carlo assessment to capture the uncertainties in the inputs and the modelling approach. Shear band propagation offers an alternative to limit equilibrium and provides a simple approach to modelling to the long and relatively thin slope failures typically observed in the offshore environment. This method is applied probabilistically, resulting in a numerical map of annual probability of slope failure through shear band propagation. The complexities of applying a two-dimensional model within a three-dimensional spatial framework are explored in this paper. This GIS-based probabilistic slope stability assessment using shear band propagation has been successfully employed as part of a landslide risk assessment of the Azeri-Chirag-Gunashli (ACG) development, Caspian Sea, where pixel-based annual probability of failure estimates contribute to a landslide risk score for each geohazard province within the area.