A suction stress-based probabilistic model for wetting induced shallow landslides

Tasnia Shehrin

doi:10.7273/000004247

Precipitation of rainwater is one of the most influential elements that can trigger shallow landslides in the mountainous region. The prevailing idea about landslide occurrence under rainfall was landslides occurs only due to increase in positive pore water pressure results from the precipitation of water during intense wetting event and thereby, a landslide can only happen below the groundwater table. However, the recent advantage of technical instrumentation in real environmental setting provides evidence that landslides can happen under partially saturated conditions, above the water table without increasing positive pore water pressure. Instead, slope stability can be affected by the reduction of negative pore water pressure and the increase of saturation. Besides, soil properties can vary spatially and with depth. There might be uncertainties in soil properties due to field and laboratory measurement errors and deviation of results from the prediction model to reality. Thus a slope stability model is essential that can consider soil variability and estimate slope failure due to change in saturation under partially saturated conditions above the water table. This study developed a suction stress-based probabilistic model that can predict the effect of different steady-state flux rate on the likelihood of landslides associated with rainfall. A factor of safety equation modified with suction stress was used in this study to capture the effect of negative pore-water pressures on shear strength. To overcome the uncertainties and variabilities associated with soil properties, a probabilistic framework was developed by applying a probability distribution function and range for each soil parameter. The properties (e.g. mean, standard deviation, correlation between soil parameters) that need to describe the distribution function were determined based on literature review. The suction stress-based probabilistic slope stability analysis was compared with traditional slope stability analysis for a hypothetical cross-section to understand the performance of this model. Moreover, the effect of depth of the water table on shallow landslides was evaluated. A sensitivity analysis was performed to determine the effect of the input parameters on slope stability. Lastly, two case studies were presented in this study to examine and validate the model performance on precipitation induced shallow landslides.

A suction stress-based probabilistic model for wetting induced shallow landslides

Files and links (1)

Abstract

Metrics

Details