Reliability analysis of an existing slope at a specific site considering rainfall triggering mechanism and its past performance records

Highlights

• A method is proposed to estimate real-time slope failure probability under a target rainfall.

• Bayesian updating of uncertainties is performed using slope performance records from past rainfall events.

• Uncertainties in soil and other parameters are significantly reduced by Bayesian updating.

• The challenge of overestimating slope failure probability is tackled.

Abstract

Accurate estimation of landslide probability or slope failure probability when there is a rainstorm event is crucial for assessment and mitigation of rainfall-induced landslide risk. Slope reliability methods provide a rigorous way of estimating landslide probability based on slope failure mechanism and probability theory. However, it is well recognized in literature that the landslide probability estimated from existing slope reliability analysis methods are often much larger than the observed landslide frequency. To improve accuracy of the slope failure probability, this study proposes a slope reliability analysis method for an existing slope at a specific site, which considers both rainfall triggering mechanism and the slope's performance records during previous rainfall events. It was found that the fact that the slope survived from previous rainfall events could be utilized to effectively reduce uncertainties in soil parameters and might reduce the estimated landslide probability by one to two orders of magnitudes. In addition, the proposed method provides a real-time slope failure probability for a given rainfall event, and the estimated slope failure probability varies as the considered rainfall event evolves with time.

Introduction

Rainfall-induced landslides are major geohazards in tropical and subtropical zones (e.g., Tang et al., 2019; Gong et al., 2021). With the rapid development of digital technologies, digital twin of slopes is emerging as a promising tool for life-cycle risk management of landslide hazards (e.g., Zhang et al., 2019; Cheverda et al., 2020). A digital twin of a slope is a virtual slope model that can be continuously improved by slope performance records and monitoring data obtained from its physical counterpart in reality (e.g., Grieves, 2015). It is valuable for evaluating actual slope performance and providing key information for decision-making and mitigation of landslide risk, such as landslide early warning (e.g., Baum and Godt, 2010; Chen and Zhang, 2014; Segoni et al., 2015; Chae et al., 2017; Kong et al., 2020).

An important task for a digital twin of slope is estimation of slope failure probability (P_f) when there is a rainstorm scenario. Slope reliability analysis methods provide a rigorous way of estimating P_f for existing slopes based on slope failure mechanism and probability theory (e.g., Li and Lumb, 1987; Baecher and Christian, 2003). Various factors and uncertainties can be properly incorporated in slope reliability analysis in a quantitative manner. However, rainfall triggering mechanism was often ignored in the previous reliability studies of slope stability. Therefore, the estimated P_f often does not reflect magnitudes of different rainfall events. Even when rainfall infiltration is considered, slope reliability analysis often produces an overestimated slope failure probability. Christian and Baecher (2011) summarized ten unresolved problems in geotechnical reliability analysis that hinders its wide applications. The top challenge in the list is “why are failures less frequent than our reliability studies predict?” The actual landslide probability is often relatively small. For example, the Geotechnical Engineering Office (GEO) in Hong Kong reported that the annual failure rate in 2016 is 0.07% and 0.527% for engineered and non-engineered slopes, respectively (Wai et al., 2018). It was found that the estimated P_f from previous reliability analysis are “at least an order of magnitude larger than the observed frequency”. One possible reason for such a difference might be attributed to huge uncertainties in soil parameters often observed from site investigation data, even when there are many measured data available (e.g., Tofani et al., 2017). To address this issue, existing information can be utilized to improve uncertainty quantification and prediction of P_f by slope reliability updating (e.g., Zhang et al., 2011; Papaioannou and Straub, 2012; Zhang et al., 2014; Schweckendiek et al., 2014; Li et al., 2015; Jiang et al., 2020). For example, Zhang et al. (2011) demonstrated that an existing slope in natural conditions can be deemed as a full-scale experiment under various triggering events, and its performance that it failed or survived from a past triggering event can be used as valuable information for updating P_f and improving its accuracy. Such performance records are also readily available in geotechnical practice. However, such performance records have rarely been utilized in estimating failure probabilities of rainfall-induced landslides, especially the survival records from multiple past rainfall events with different rainfall intensities, durations, and patterns. Furthermore, combinations of weak soil strength parameters may be generated when random variables or random fields are used to represent soil parameters in slope reliability analysis, and they result in slope failure before imposing any triggering factor, especially when the variability of soil strength parameters is large. This inevitably leads to overestimation of slope failure probability. The estimation of slope failure probability can be enhanced by utilizing the fact that an existing slope at a specific site is stable before imposing any triggering factor and after previous rainfall events.

This study aims to propose a slope reliability analysis method for estimating time-variant P_f of a slope existing at a specific site under a specific target rainfall scenario, such as a time-variant rainfall event. To improve estimation of P_f, the proposed method explicitly considers the fact that an existing slope at a specific site is stable before imposing any triggering factor, and it incorporates survival records from multiple past rainfall events for improving quantification of uncertainties in soils and prediction of P_f. The proposed method will be introduced in the next section, followed by its implementation procedure in Section 3. Then, it is illustrated using an infinite slope example under rainfall infiltration considering its past performance records in Section 4. Section 5 summarizes major conclusions drawn from this study.