Quantitative risk assessment of slow-moving landslides from the viewpoint of decision-making: A case study of the Three Gorges Reservoir in China

Abstract

A large number of landslides in the Three Gorges Reservoir area in China have been reactivated by reservoir impoundment since 2003. Many of them were activated as slow-moving landslides, which caused severe damages to buildings and also pose a significant risk for residents. In this study, a quantitative procedure has been proposed to analyze and assess the landslide risks to buildings and lives by considering the Outang landslide as an example. First, the landslide segmentation was done by considering the sliding history, macroscopic deformation, and monitoring data of displacement. The SLOPE/W and SEEP/W modules of the Geostudio software were used to establish the geotechnical model and analyze the seepage, respectively. Under different scenarios of annual reservoir regulation and return period of rainfall events as determined by Pearson type III (P3) distribution, failure probabilities of every segmented part were obtained with Monte Carlo simulation. Second, the vulnerability of the elements at risk was quantitatively estimated using a model that incorporated the landslide intensity and resistance of the exposed elements. Then, through the analysis and comparison of scenario-based landslide risks, the evolution of landslide risk influenced by different combinations of triggering factors (i.e., reservoir water level and rainfall) was examined. Finally, landslide risks before and after the implementation of several risk reduction alternatives were compared to determine the benefits of alternatives. The risk maps displayed that the risk was the greatest for the scenario of the rapid decline of the reservoir water level in combination with heavy rainfall; the corresponding total annual economic and population risks of the landslide were estimated in 96.88 million ¥ and 216 lives, respectively. In this scenario, the economic risk level of the west deformation zone was the highest, and the total population risk was the highest in the lower unit of the landslide. Results of benefit-cost analysis suggested that real-time monitoring and resident relocation seem to be the most effective alternatives for risk reduction concerning the Outang landslide.