Spatial Variability of Input Motion in Stochastic Slope Stability

Abstract

The slope stability has been widely analysed in the past, which demonstrates the significance of the subject in human life, especially in terms of life safety and economy. This is even more important in earthquake-prone areas where statically stable slopes may be triggered by the dynamic loads. Pseudo-static (PS) approaches are most used in the first stages of the regular assessment of seismic slope stability analysis. However, the spatial variability of the PS coefficient has not been considered so far in the realm of PS analyses, meaning that the value of the seismic coefficient is assumed to be constant at every location within a field while this is not true in the real case. The consideration of the variation of the PS coefficient is especially relevant to landslides in wide areas. This research aims at addressing this issue considering the stochastic nature of soils in seismic slope stability analysis within the framework of the limit equilibrium method (LEM) of slices, Monte Carlo (MC) simulation and random fields, termed 2D-RLEM. Results of parametric studies are presented, through which the sensitivity of stochastic slope stability problem to various factors, including different levels of spatially variable PS loading, the scale of fluctuation (SOF) of the PS coefficient random field, etc, are explored. It was concluded that the effect of different values assigned to the coefficient of variation and the SOF of the PS coefficient on the resulting slope probabilities of failure was more tangible for larger earthquakes.