ABSTRACT
Rainfall-induced shallow landslides pose significant risks in unsaturated, highly weathered residual soils, where spatial variability in permeability critically affects slope stability. This study examines the influence of permeability heterogeneity on infiltration and failure behavior using a coupled hydromechanical Random Finite Element Method (RFEM) with Monte Carlo simulations. Permeability is modeled as a log-normal random field with varying coefficients of variation (60%, 75%, 90%) and vertical correlation lengths (0.5–15 m). Simulations on a 30° infinite slope with 3 m depth under sustained rainfall evaluate the temporal evolution of factor of safety ( https://www.w3.org/1998/Math/MathML" display="inline"> F o S ) and probability of failure. Results show that higher CoV and shorter correlation lengths lead to earlier, more uncertain failures and broader distributions of failure time. The framework effectively captures the temporal progression of instability, with the most probable time of failure decreasing as spatial variability intensifies, underscoring the need to account for heterogeneity in slope stability predictions.
