Three-dimensional numerical simulation of mud flow from a tailing dam failure across complex terrain

• Published in: Natural hazards and earth system sciences 2020-03, Vol.20 (3), p.727-741
• Main Authors: Yu, Dayu, Tang, Liyu, Chen, Chongcheng
• Format: Article
• Language: English
• Subjects: Accidents; Aerodynamics; Algorithms; Analysis; Channel flow; Computational fluid dynamics; Computer applications; Computer simulation; Constitutive models; Dam failure; Dams; Debris flow; Disasters; Ecological effects; Emergency preparedness; Emergency response; Emergency warning programs; Exact solutions; Failure; Finite volume method; Flow (Dynamics); Flow characteristics; Flow simulation; Fluid dynamics; Fluids; Hydrodynamics; Laboratories; Mathematical models; Mine tailings; Model accuracy; Mud; Mudflows; Navier-Stokes equations; Numerical analysis; Numerical simulations; Rheological properties; Rheology; Routing; Simulation; Terrain; Unmanned aerial vehicles; Viscosity
• ISSN: 1684-9981; 1561-8633; 1684-9981
• DOI: 10.5194/nhess-20-727-2020

A tailing dam accident can cause serious ecological disaster and property loss. Simulation of a tailing dam accident in advance is useful for understanding the tailing flow characteristics and assessing the possible extension of the impact area. In this paper, a three-dimensional (3-D) computational fluid dynamics (CFD) approach was proposed for reasonably and quickly predicting the flow routing and impact area of mud flow from a dam failure across 3-D terrain. The Navier–Stokes equations and the Bingham–Papanastasiou rheology model were employed as the governing equations and the constitutive model, respectively, and solved numerically in the finite volume method (FVM) scheme. The volume-of-fluid (VOF) method was used to track the interface between the tailings and air. The accuracy of the CFD model and the chosen numerical algorithm were validated using an analytical solution of the channel flow problem and a laboratory experiment on the dam-break problem reported in the literature. In each issue, the obtained results were very close to the analytical solutions or experimental values. The proposed approach was then applied to simulate two scenarios of tailing dam failures, one of which was the Feijão tailing dam that failed on 25 January 2019, and the simulated routing coincided well with the in situ investigation. Therefore, the proposed approach does well in simulating the flow phenomenon of tailings after a dam break, and the numerical results can be used for early warning of disasters and emergency response.