The accurate and rapid prediction of breach outflow rates of landslide dams is crucial for effective risk assessment and mitigation strategies. However, challenges arise due to the complexities associated with breaching mechanisms. In this study, 12 flume tests were conducted on landslide dams composed of various material types, including fine-grained, coarse-grained, well-graded, and gap-graded materials, each with two dry densities. This study comprehensively investigated erosion, deposition, and breach evolution, providing the foundation for the development of breach evolution models. A comprehensive logic diagram, focusing on shear stress and grain size distribution, was proposed to forecast the failure modes of landslide dams. The study also suggested a method to assess the longitudinal breach evolution based on the prediction of the Erosion Point (EP) and Deposition Point (DP), and to assess the lateral breach evolution by considering both steep and gentle breach side slopes. Our findings indicate that fine-grained materials undergo layer erosion, coarse-grained materials remain stable, well-graded materials develop erosion pits, and gap-graded materials are prone to piping. The existence of an interface can result in piping channels, even when the overall hydraulic gradient is insufficient to initiate piping. Incorporating the longitudinal slope angle and its influence on both driving and resisting forces into the traditional erosion rate equation can significantly enhance its accuracy, especially when coarse particles are present. Coarse particles play a pivotal role in erosion and deposition behavior and reshaping the breach evolution as the initiation of coarse particles is dominated by rolling instead of slipping in fine particles. Deposition on the downstream slope of a dam can significantly influence the longitudinal breach process, leading to changes in the downstream topography and grain size distribution. The proposed method for assessing longitudinal breach evolution unifies the previous methodologies by predicting the positions of EP and DP based on erosion and deposition.
Read full abstract