Probabilistic Analysis of Earthquake-induced Failure and Runout Behaviors of Rock Slopes with Discrete Fracture Network

Abstract

Earthquake-induced rock slope failure is one of the most destructive geohazards, posing serious threats to human lives and properties. The potential hazard of a rock slope, under the loading of earthquakes, can be assessed from its failure and runout characteristics, which are however affected by inherent fractures. To simulate the randomness of these inherent fractures, the discrete fracture network (DFN) modeling technique that is based on the Monte Carlo simulation (MCS) is often employed. It is worthwhile noting that different realizations of DFN can be sampled from different MCSs; however, only one realization of DFN is considered and studied in the conventional analysis. There exist differences between the arbitrarily selected DFN and true DFN; as such, the conventional deterministic analysis could cause uncertainty in DFN modeling and the potential hazard of a rock slope cannot be accurately evaluated. In such a circumstance, this paper proposes a probabilistic analysis of earthquake-induced failure and runout behaviors of fractured rock slopes, in which multiple realizations of DFN that are systematically generated with MCS are taken as inputs to the discrete element method (DEM) modeling of slope behaviors, thus, the uncertainty induced by DFN modeling can be explicitly considered and evaluated.