An integrated analysis framework for surface runoff, infiltration, slope stability and slope real-time monitoring

Abstract

Buildup of pore-water pressure due to water infiltration during an extreme rainfall event is a major cause of many landslides worldwide. A competent simulation model can greatly contribute to the reliable landslide prediction and prevention. Rainfall-induced slope stability is dependent on many factors, including slope topography, rainfall hyetograph, surface runoff characteristics, soil properties, and sub-surface conditions. This thesis aims at developing an integrated analysis framework for the rainfall-induced slope stability problems.

A fully-coupled groundwater-surface water interaction numerical model, HydroGeoSphere, is used to compute the transient surface runoff and sub-surface pore-water pressure responses due to rainfall simultaneously without the need to make assumptions on the infiltration rate. Therefore, rainfall hyetograph can be used directly as an input parameter in the numerical model. The computed pore-water pressure as a function of time is used as input to the computer program OpenSees. OpenSees is used to calculate slope displacements using finite element methods. Parameter estimation methods are applied to calibrate the model parameters on the basis of field data. Assessment of slope stability can be made according to the model results. It is proven that the integrated model provides a more comprehensive and reliable approach for slope safety analysis, real-time monitoring, and hazard assessment.

A case study on a full-scale instrumented slope in Hong Kong is also presented to evaluate the integrated framework. The results obtained from the numerical simulation are compared with the field observation data and slope stability of the site is assessed. The integrated model is capable to provide reliable results on many aspects.

Statistical analyses using Kalman Filtering are also carried out to conduct real-time slope monitoring and prediction for comparison. There are both advantages and disadvantages for the integrated model and Kalman Filtering in real-time slope monitoring and prediction.