Probabilistic slope stability analysis considering the non-stationary and spatially variable permeability under rainfall infiltration-redistribution

The non-stationary and spatially variable permeability k_s can have a significant influence on the process of rainwater infiltration and redistribution in a soil slope and further affect the slope stability and the slope failure time after a rain event. However, there has been limited research on the influence of non-stationary and spatially variable k_s on the stability and the failure time of soil slopes. This paper aims to investigate how the non-stationary and spatially variable k_s affects the soil slope stability and the most probable failure time (MPFT) of the slope considering the rainwater infiltration and redistribution. To achieve this purpose, an integrated probabilistic analysis framework is proposed by coupling the non-stationary k_s random field into a slope model for the probabilistic slope seepage analysis based on Monte Carlo simulations. An unsaturated soil slope subjected to rainfall is taken as an illustrative example to demonstrate the influence of non-stationary and spatially variable k_s on the slope stability. The results show that the location of the slip surface with minimum factor of safety continuously changes during the process of the rainwater infiltration-redistribution. Ignoring the effect of decreasing trend of k_s will result in an underestimation of the failure probability of slope even if the spatial variability of k_s is considered. The MPFT of slope will be earlier if the non-stationary characteristics of k_s is considered.