Can the collapse of a fly ash heap develop into an air-fluidized flow? — Reanalysis of the Jupille accident (1961)

Abstract

A fly ash heap collapse occurred in Jupille (Liege, Belgium) in 1961. The subsequent flow of fly ash reached a surprisingly long runout and had catastrophic consequences. Its unprecedented degree of fluidization attracted scientific attention. As drillings and direct observations revealed no water-saturated zone at the base of the deposits, scientists assumed an air-fluidization mechanism, which appeared consistent with the properties of the material. In this paper, the air-fluidization assumption is tested based on two-dimensional numerical simulations. The numerical model has been developed so as to focus on the most prominent processes governing the flow, with parameters constrained by their physical interpretation. Results are compared to accurate field observations and are presented for different stages in the model enhancement, so as to provide a base for a discussion of the relative influence of pore pressure dissipation and pore pressure generation. These results show that the apparently high diffusion coefficient that characterizes the dissipation of air pore pressures is in fact sufficiently low for an important degree of fluidization to be maintained during a flow of hundreds of meters.

Effects of pore pressure in pyroclastic flows: Numerical simulation and experimental validation

“…Note that this is also close to the approach of Stilmant et al . [] for simulation of ash coke flows. However, equation is slightly different than their solution that uses a coefficient 2 D , instead of (see supporting information).…”

Mechanisms of Submarine Debris Flow Growth

“…However, when the flow is fully liquefied, grain frictional stresses are insignificant and the induced shear at the flow-bed interface may be assumed to be governed solely by the viscous stress of the fluidized mass (Campbell, 1978). As such, the degree of liquefaction of the laminar layer, 𝐴𝐴 𝐴𝐴f , may be used to quantify the relative contributions from grain frictional and fluid viscous shear stresses to the entrainment process (Iverson, 2012;Stilmant et al, 2015):…”

“…However, when the flow is fully liquefied, grain frictional stresses are insignificant and the induced shear at the flow‐bed interface may be assumed to be governed solely by the viscous stress of the fluidized mass (Campbell, 1978). As such, the degree of liquefaction of the laminar layer, $${\theta }_{\mathrm{f}}$$, may be used to quantify the relative contributions from grain frictional and fluid viscous shear stresses to the entrainment process (Iverson, 2012; Stilmant et al., 2015): $$${\theta }_{\mathrm{f}}={p}_{\mathrm{f}}/{\sigma }_{\mathrm{f}}$$$ where $${\sigma }_{\mathrm{f}}$$ is the gravitational stress at the base of the flow, which can be calculated by: $$${\sigma }_{\mathrm{f}}=\left({\rho }_{\mathrm{f}}-{\rho }_{\mathrm{w}}\right)\,g\,{h}_{\mathrm{f}}\,$$$ where $${h}_{\mathrm{f}}$$ is taken as the depth of the laminar layer. In our theoretical framework, we considered the role of the parameter of the liquefaction degree as follows: when $${\theta }_{\mathrm{f}}=0$$, the overriding flow can be assumed to behave as a frictional mass, whereby the flow‐bed interface is governed by the frictional stress.…”

Discretization of the divergence formulation of the bed slope term in the shallow-water equations and consequences in terms of energy balance

“…The shallow-water equations (SWE) constitute the state-of-the-art for large-scale modelling of a wide range of flow in hydraulic engineering and fluvial hydraulics [5,7,8,26,27,35], as well as for applications in other fields such as coastal engineering [30] or geophysical flow [29].…”