Key points are not available for this paper at this time.
Abstract Rockslides and their cascading hazards, such as river blockage and outburst flooding, pose significant threats to infrastructure and human lives, particularly in mountainous and seismically active regions. To enhance hazard forecasting and risk mitigation, it is helpful to develop accurate numerical models for analysing the failure mechanisms of rock slopes under external loadings. Traditional two-dimensional numerical simulations have constraints in capturing the anisotropic stress distribution and complex evolution of failure mechanisms in three-dimensional space. Moreover, most three-dimensional numerical models typically treat rock slopes as homogeneous media, rarely accounting for the natural fractures within rock blocks. To address these limitations, this study introduces a novel three-dimensional material point method (MPM), combined with the discrete fracture network (DFN) to dynamically simulate the initiation and evolution of rockslides under seismic loading, along with the subsequent river blockage phenomenon. The DFN is explicitly incorporated into MPM using a projection algorithm by mapping fracture geometries onto the computational domain and then assigning the physical properties of fractures to specific material points, enabling a more realistic representation of fractured rock masses. The reliability of the proposed framework in addressing complex dynamic contact problems is verified through four numerical case studies. The motion of rockslides under seismic loading and the deposition of fragmented rock blocks within a river channel are then studied, revealing that fracture intensity and fracture orientation significantly influence the failure modes of rock slopes and the extent of river blockage. This research provides valuable insights into the complex failure processes of fractured rock slopes and interactions between rockslides and river systems, contributing to the advancement of hazard assessment and mitigation strategies.
Wu et al. (Sat,) studied this question.