On three-dimensional SPH modelling of large-scale landslides

Abstract

Lagrangian particle-based smoothed particle hydrodynamics (SPH) is increasingly widely used in landslide modelling. This paper investigates four important issues not addressed by previous studies on SPH modelling of large-scale landslides, i.e., convergence property, influence of constitutive parameters, scale effect and friction reduction, and influence of different treatments of the viscous effect. The GPU-acceleration technique is employed to achieve high-resolution three-dimensional (3D) modelling. The Baige landslide is investigated by comparing numerical results with field data, and detailed analyses on the four issues are provided. Suggestions on particle resolution, constitutive parameter, and formulations of viscous discretization are also presented for future SPH modelling of large-scale landslides.

Sensitivity and Calibration of Three‐Dimensional SPH Formulations in Large‐Scale Landslide Modeling

“…However, we are fully aware that our current work does not go so deep to demonstrate the benefits of employing 3D simulations comprehensively. In addition, the high‐efficient GPU acceleration is specifically beneficial in the varied and complex topography of real‐world terrain (Peng et al., 2022).…”

“…The aforementioned results are like the difference between small‐scale and large‐scale problems. In small‐scale problems, cohesion is very important and largely determines the final deposition (Bui et al., 2008; Peng et al., 2022). In contrast, in large‐scale problems, cohesion plays a lesser role.…”

“…This is because, in large‐scale landslides, the yield stress is mainly dependent on friction caused by the high normal stress due to the large thickness of sliding mass. Therefore, we can also infer that cohesion generally controls the final extension and cessation of the edge of deposits because of the small material height close to the margin where the friction influence is too small to eclipse the effect of cohesion (Peng et al., 2022).…”

“…A large regularization parameter leads to an extremely large apparent viscosity in the quasi-static flow regime that will severely reduce the numerical efficiency. Several previous test cases (S. Li et al, 2020;Peng et al, 2022) indicate that m = 100 s −1 can correctly simulate landslides. Therefore, in this study, we will set m = 100 s −1 with the RDP model for the Baige landslide.…”

Landslide mechanism and stability of an open-pit slope: The Manglai open-pit coal mine

“…Different types of landslide deformation failure modes have been studied via field investigations, numerical analyses, monitoring data, and model tests (Locat et al, 2017;Peng et al, 2021;Wu et al, 2021). According to the deformation mechanism and the different failure modes of landslides, landslides can be divided into three types according to the spatial development patterns of their sliding zones (Figure 2).…”