Seismic monitoring data acquired at mines can be used to infer the characteristics of coseismic strain, including amount, orientation of principal directions and type (deviatoric, volumetric explosive or implosive). This strain corresponds to plastic strain increment if seismic events represent episodes of sudden deformation in confined environment. If sources of seismic events involve the convergence of excavations, then coseismic strain depends on the amount of the convergence. These peculiarities of coseismic strain are taken into account in the proposed method for its direct incorporation into a numerical stress model. The method is verified using analytical cases of confined inelastic deformation as well as brittle failure around a tunnel. The method can be used to assess the redistribution of stresses and change in the loading of tunnels in seismically active mines. A synthetic mining example is presented to illustrate the impact on tunnel stress.
Seismic systems installed at mines record seismic waves radiated by episodes of sudden (i.e., durations on the order of fractions of a second) deformation within the rockmass. The analysis of these recorded waves make it possible to assess the time, location, amount and intensity of deformation, as well as its geometrical characteristics (e.g., direction of principal strains, proportion of volumetric and deviatoric shape change). This information is highly valuable for understanding the overall rockmass deformation induced by mining and the associated redistribution of stresses. It is quite logical that the characteristics of rockmass deformation inferred from seismic data are often utilized when performing numerical stress modelling. There are fundamentally two ways to do this.
Firstly, seismic data can help to improve the input parameters of the stress models. This is often done through qualitative or quantitative comparing of various aspects of seismic data (e.g., location of sources) with the corresponding modelling parameters (e.g., areas of high deviatoric stress) – see, for example, the works of Lachenicht (2001);
Beck and Brady (2002); Beck et al. (2006); Spottiswoode et al. (2008); O'Connor et al. (2010); Arndt et al. (2013); Kalenchuk (2022); Malovichko and Rigby (2024). The improved (calibrated) stress model is expected to deliver more realistic distribution of stresses for the current mining step and should provide better prediction of the stress redistribution for the future planned mining steps.
