Development and application of a quantitative risk assessment to a very slow moving rock slope and potential sudden acceleration

The benefits of quantitative risk assessments for landslide management have been discussed and illustrated in several publications. However, there still are some challenges in its application for low-probability, high-magnitude events. These challenges are associated with the difficulties in populating our models for risk calculations, which largely require the input of expert opinion. This paper presents a quantitative risk assessment to a very slow moving rock slope within a dam reservoir in the Province of British Columbia, Canada. The assessment is focused on the risk to the population in the vicinity of the dam and the populated areas downstream. Expert opinions quantified the slope failure probabilities in the order of 10^?3 to 10^?1 per year for the smallest failure scenario considered and less than 10^?6 for a failure of the entire slope. However, these estimations are associated with high levels of uncertainty. Our approach starts with the calculation and assessment of the magnitude and probability of the potential slope failure consequences, minimizing the uncertainties associated with estimated slope failure probabilities. Then, these consequences and failure probabilities are combined to obtain a measure of risk. The uncertainty associated with the slope failure probabilities is managed by the estimation of plausible ranges for these. The calculated risk levels are then presented as ranges of values and assessed against adopted evaluation criteria. The consequence and risk assessment of the rock slope suggest that the risk to the population exposed in the vicinity of the dam and populated areas downstream is under adequate control. The probability of large consequence scenarios is extremely low, in the order of 10^?7 chance of an event causing more than 100 fatalities. We propose an observational technique to assess changes in risk levels and decide when to update the risk management approach or deploy emergency measures. The technique is focused on the detection of changes in the slope deformation patterns that would indicate an increase in the potential failure volumes or an imminent failure. It can be considered an extension to the current early warning system in place, easy to implement and enhanced with the strength of the comprehensive analysis required for a quantitative risk assessment.