Failure analysis of an infinite unsaturated soil slope

Increases in rainfall and groundwater level rising can cause more frequent failures of unsaturated soil slopes, resulting in large-scale landslides. The aim of this paper is to propose a methodology for prediction of the failure of an infinite soil slope subject to steady unsaturated flow conditions. Probabilistic models for soil porosity, friction angle, matric suction and saturation are developed. The probability of slope failure for different groundwater levels and infiltration (or evaporation) intensities is investigated. Numerical results show that, for a slope with deep groundwater, the slope failure is mainly controlled by the rainfall infiltration, whereas for shallow groundwater, the slope failure is governed by the location of the groundwater level. A merit of the proposed methodology is that the fluctuation of the soil porosity along the depth of a slope is modelled as a Gaussian random field. The paper concludes that the proposed methodology can be used for prediction of slope failures under steady unsaturated flow conditions. Accurate prediction of slope failures can prevent catastrophic consequences of landslides.