Numerical simulation of wave propagation and snow failure from explosive loading

TL;DR

This study uses numerical simulations of acoustic wave propagation in snow to understand how explosions induce snow failure, aiding avalanche control efforts.

Contribution

It introduces a coupled acoustic-poroelastic model to simulate snow response to explosions, linking stress fields to snow failure mechanisms.

Findings

Simulated accelerations correlate with field measurements.

Acoustic waves induce stress exceeding snow strength, causing failure.

The model helps predict snow failure locations during explosions.

Abstract

Avalanche control by explosion is a widely applied method to minimize the avalanche risk to infrastructure in snow-covered mountain areas. However, the mechanisms involved leading from an explosion to the release of an avalanche are not well understood. Here we test the hypothesis that weak layers fail due to the stress caused by propagating acoustic waves. The underlying mechanism is that the stress induced by the acoustic waves exceeds the strength of the snow layers. We compare field measurements to a numerical simulation of acoustic wave propagation in a porous material. The simulation consists of an acoustic domain for the air above the snowpack and a poroelastic domain for the dry snowpack. The two domains are connected by a wave field decomposition and open pore boundary conditions. Empirical relations are used to derive a porous model of the snowpack from density profiles of the field experiment. Biot's equations are solved in the poroelastic domain to obtain simulated accelerations in the snowpack and a time dependent stress field. Locations of snow failure were identified by comparing the principal normal and shear stress fields to snow strength which is assumed to be a function of snow porosity. One air pressure measurement above the snowpack was used to calibrate the pressure amplitude of the source in the simulation. Additional field measurements of air pressure and acceleration measurements inside the snowpack were compared to individual field variables of the simulation. The acceleration of the air flowing inside the pore space of the snowpack was identified to have the highest correlation to the acceleration measurements in the snowpack.