D. A. ANGUS

ESG SOLUTIONS, CANADA, DOUG.ANGUS@ESGSOLUTIONS.COM

DOI Number: https://doi.org/10.66733/IEGL9743


Introduction

The energy transition will require more critical minerals, with many of these mineral deposits being located deep underground posing significant safety and technological challenges. Underground mining operations generally encounter adverse stress conditions, the resulting response can be seismic events occurring either very soon, or after sustained perturbation of the in-situ stress over several years. It has been recognized that closely monitoring and characterizing seismic response is non-negotiable for underground safety and therefore plays a key role in the operational success of a mine. With increasing depth, seismic hazard increases as larger magnitude events become more common, potentially leading to damaging rockbursts. As such, seismic monitoring systems have been installed at mines globally over the past several decades to help mine operators locate and understand rock failure induced by excavation.

Appropriately designed and optimized seismic arrays enable well-constrained event locations and hence provide greater confidence in calculated source parameters and greater accuracy in advanced seismic analysis methods such as tomographic imaging, source mechanism analysis, stress inversion and strain rate. The results of these analyses can then be evaluated in relation to known structures and excavations. The integration of Distributed Acoustic Sensing (DAS) fibre-optic monitoring with conventional seismic sensors has the potential to innovate the capture of both brittle anelastic and aseismic creep rock mass response. The integration and interpretation of different seismic parameters and methods are helping to improve short- and long-term seismic hazard assessment and to identify regions of high and low concern.

Mining context, seismic concepts and seismic systems

Mining context:

The underground mining environment is a dynamic system. As ore is extracted using either stoping or caving methods, adverse conditions can develop as stresses and strains within the rock mass respond to the creation of voids where ore is removed and to the development of mine infrastructure (e.g., drifts and shafts). As a general rule of thumb, in most parts of the world seismic hazard develops as mining progresses below a depth of 500 m, where the rock mass response has higher probability of being seismic rather than aseismic. This heuristic depends on the stress gradient in the crust and on the ratio of horizontal to vertical stress. Figure 1 shows the location of recorded seismicity and a geomechanical model of the stress evolution for Creighton Mine, Sudbury, Canada (adapted from Mercer & Bawden, 2005) as well as an example of support damage due to a rockburst at Nickel Rim Mine (adapted from Collins et al., 2025).

Figure 1: (a) Cross section view of seismic event locations over a three-month period (depths in feet), (b) contours of σ3 principal stress estimates and (c) example of support damage due to a rockburst at Nickel Rim mine.

As a result of stress perturbations, a variety of seismic event types can occur during mine operations. Seismicity in mines is predominantly earthquakes due to slip along pre-existing faults. These range over several orders of magnitude, from micro-earthquakes to larger-scale earthquakes that can be recorded by regional surface seismic networks. However, a variety of other seismic phenomena also pose risks, as summarized in Table 1.

Table 1: Expected seismic event types, mechanisms, and sizes.

Seismic concepts:

On a fundamental level, mine seismology is not complicated. The primary task is to identify and pick the arrival times of primary P- and S- body-wave phases recorded at seismic sensors and then, based on a calibrated velocity model and Green’s function (e.g., ray tracing or eikonal algorithm), inverting for the estimated source location and magnitude (Figure 2). The critical challenge is being able to automatically and robustly locate seismicity and estimate magnitudes in real-time, 24 hours a day, 7 days a week, 365 days a year, using seismic monitoring systems that operate in extremely harsh, noisy, and dynamic subsurface conditions.

The location of seismic events provides real-time information on how the rock mass is responding to development and extraction. The magnitude of a seismic event describes the strength of the inelastic failure at the source and is independent of the location of the observation. Magnitude scales are logarithmic and are extracted from the waveform amplitude of the recorded seismic signal. An increase of one unit of magnitude on a magnitude scale is equivalent to an increase of 10 times the amplitude recorded by a seismograph and approximately 30 times the energy.

Figure 2: (a) Three-component seismogram showing P- and S-wave phase arrival picks as well as T0 representing the estimated source initiation time, (b) seismic section for a mine seismic event with signals recorded on three- and one-component sensors highlighting increasing offset of P- and S-wave arrival times with distance from the source, and (c) example three-dimensional velocity model displaying ray paths between various sources and receivers.

Table 2 provides a summary of typical mine seismicity. Note that the source radius estimates are based on the model of Brune (1970) and are generally valid when the seismic event is predominantly shear-slip failure. When seismic events occur close to excavations, tensile failure becomes significant, and so source sizes are better estimated using the approach of Cai et al. (1998).

Table 2: Approximate source characteristics assuming VP 5630 m/s and VS 3290 m/s and Brune model.

Seismic monitoring systems:

The seismic monitoring system is a hazard management tool with four key objectives: (a) early warning and real-time alerts, (b) provision of data to support re-entry and exclusion zones for people and equipment, (c) assistance in reducing mine operational downtime, and (d) rock mass characterization to aid ground support design and mine planning. Figure 3a shows a schematic of a typical seismic monitoring system consisting of borehole-deployed sensors powered by the mine electrical grid and connected to digitizers that acquire and digitize seismic signals. The digitizers and network switches are connected to the mine fibre communication network within protective junction boxes. The digitizers are synchronized to a GPS timing device at the surface using Precision Timing Protocol (PTP) to ensure accurate time-stamping of the seismic signals, which is necessary for accurate source locations. For a typical system, the flow of data starts with the sensor detecting an impulsive signal. The signal is acquired and digitized by a seismic recorder and the data are transferred along the communications network to an acquisition server for pre-processing. The data are then processed in real-time on a processing server to determine the event location and initial source parameters. The data can be interpreted on site and processed further with more advanced algorithms, with cloud-computing enabling the seismic service company to issue alerts, perform QC, and interpret results in real time.

Figure 3: (a) Schematic of mine seismic monitoring system for an underground mine, (b) three- and one-component seismic sensor (left and right respectively), and (c) example magnitude range for the various sensor types based on detectability and signal frequency content.

Mine seismic systems generally consist of a variety of seismic sensors, including uniaxial and triaxial component sensors, accelerometers, and geophones. Uniaxial sensors have single elements and smaller dimensions than triaxial sensors, enabling cost-efficient higher-density sensor placement in mines to achieve greater detectability. Uniaxial sensors are limited in the calculation of source parameters because they do not record the components of motion in three dimensions. Triaxial sensors have three orthogonal elements and are typically larger than uniaxial sensors. Due to cost considerations or channel restrictions, they have less dense placement in mines, reducing detectability but providing more complete information about the recorded vibration. Modern accelerometers consist of piezo-electric elements with internal amplifiers that measure voltage proportional to acceleration. They can be mounted in any orientation and have flat frequency responses, typically between 50 Hz and 5000 Hz. Due to the sensitivity of accelerometers, these sensors are generally used to record events of moment magnitude (MW) between -3.0 and -0.5 but can often be used for larger events for improved location accuracy using the P- and S-wave phase picks. For larger magnitude signals accelerometers may record saturated or clipped signals, limiting their utility beyond leveraging the P-wave phase arrivals. Geophones consist of magnets, coils, and masses and have no internal amplifier. They measure voltage proportional to velocity, with some sensors requiring specific orientations. The predominant geophone types in mines are 10 Hz and 15 Hz geophones that have flat frequency responses between 10 or 15 Hz up to 1500 Hz and are generally used to record MW between -0.5 and 1.5. Lower frequency geophones (e.g., 4.5 Hz or 2.0 Hz) are installed when larger magnitude events are expected (see Figure 3c for schematic range of sensor suitability) at surface and/or underground within the mine, installed within specially drilled vertical boreholes. Often, lower frequency sensors can be deployed on the surface to help mining operations assess mine-induced versus regional seismicity.

Seismic tools

There are a variety seismic tools to help mines manage hazard, and no one tool should be used alone. Each seismic parameter has associated uncertainty, and it is critical to understand the magnitude and cause of the uncertainty. Uncertainty is inherent in the data for a variety of factors: geological and geomechanical uncertainty that are used to build velocity models, limitations in the seismic array in terms of spatial coverage, geometry and sensor types used, dynamic mining environments that lead to signal-to-noise issues such as recording impulsive noise signals, as well as limiting factors related to technology, communication and contractual issues.

Triggers:

The trigger represents the most certain attribute recorded on a seismic system as there are no assumptions (a trigger is a trace gather which contains a possible seismic or blast signal). A seismic monitoring system is configured to listen to the rock mass and record impulsive signals using an acquisition setup consisting of one or more digitizers. These systems can run in continuous or triggered mode. In continuous mode, the acquisition unit continuously digitizes the analogue signal from the sensors and passes the digital signal to an acquisition server that saves the digital data to disk. An algorithm, such as a short-term/long-term average (STA/LTA) calculation (e.g., Withers et al., 1998), scans the data for impulsive signals above a preset amplitude threshold (e.g., Figure 4a). If a preset number of signals exceed the threshold the server assembles the signals into a trigger. In trigger mode, the acquisition unit digitizes and temporarily triggers any impulsive signals with amplitude above a preset threshold and sends the temporary trigger to the acquisition server. The acquisition server then compares all the triggers from the various acquisition units and, if the trigger times fall within a preset association interval, assembles the temporary triggers into one trigger.

Monitoring the located trigger rates is an effective tool to monitor seismicity levels and aftershock decay due to planned blasting as well as significant large seismic events (see Figure 4c). Like a Geiger counter, trigger activity tracker tools can be configured to monitor aftershock activity within the whole mine, or specified zones within the mine, to evaluate potential hazard and whether seismic activity has returned to background levels.

Figure 4: (a) Schematic of STA/LTA function for a hypothetical waveform, (b) snapshot of digitizers and (c) example display of trigger activity over a 24-hour period, highlighting aftershocks due to a production blast and 2.5 hours later, aftershocks due to a large seismic event.

Locations:

For all triggers, automatic routines pick the primary P- and S-wave phase arrivals on the recorded signals, which are subsequently located using a calibrated velocity model and a ray-based forward modelling algorithm. A modified Simplex-Geiger (e.g., Lomax et al., 2009) and/or a coherency-based migration algorithm (e.g., Gharti et al., 2010) are typically used given the required speed and robustness in finding a solution in real-time. When triaxial sensors are available, application of a polarization filter (e.g., Montalbetti & Kanasewich, 1970) to rotate the three-component signal into local ray coordinates is recommended for improved location accuracy and to reduce any potential array geometry artifacts. Figure 5a shows an example of automatically located seismicity for a period of three months at a nickel mine in northern Ontario. Location accuracy and precision is achieved through iterative calibration of velocity models using known production blast data.

Figure 5: (a) Automatic locations of seismic events for a three-month period at a nickel mine, (b) iterative calibration of velocity models using blast data enables location algorithms to achieve necessary accuracy and precision for mine operations, and (c) classification of located triggered signals for seismic hazard management catalogue.

Due to the dynamic nature of mining, signals recorded within the mine consist of seismic events and production/development blasts as well as noise such as drilling noise, ore pass noise, ventilation noise, etc. Seismic catalogues are filtered in real-time to distinguish seismic events and blasts from sources of noise to facilitate seismic hazard management as well as blast production monitoring.

Uncertainty in location stems from a variety of factors, such as error in phase picks, uncertainty in velocity model, and limitations of forward modelling algorithm (e.g., Usher et al., 2013; Angus et al., 2014) and array geometry relative to an event. Despite these uncertainties, locations provide mine operations valuable information on the rock mass response (see Figure 6). The locations of significant events, as well as spatial and temporal clustering of events, can provide details on seismic hazard and map de-stressed zones (e.g., Simser et al., 2015), map potential damage thresholds onto mine workings using peak particle velocity PPV (e.g., Collins et al., 2026) and cumulative absolute displacement (e.g., Mendecki, 2019), and seismic tomography to map spatial and temporal changes in seismic velocities to image changes in the stress field (e.g., Bailey et al., 2026, Angus et al., 2026).

Figure 6: Application of source locations for risk management: (a) event mapping for spatial and temporal clustering, (b) seismic hazard mapping of peak particle velocity (PPV) from seismic catalogue (red representing relatively higher hazard and blue representing lower hazard), and (c) seismic tomography to image changes in P- and S-wave velocities using variations in source-sensor travel-times as a proxy for changes in stress.

Source parameters:

Source parameters, such as magnitude, energy, source dimension, and stress drops (e.g., Gibowicz and Kijko, 1994), provide physical details about the inelastic failure process. Much of the uncertainty in source parameters relates both to the cumulative propagation of errors due to location uncertainty and to signal-to-noise issues, as well as to imposing kinematic or quasi-dynamic seismic source models to approximate the far-field displacement time-frequency signal characteristics (see Figure 7a and b). The seismic moment (M0) and moment magnitude (MW) are derived from the source spectra using the low frequency plateau (Ω0) and are generally robust calculations yielding greater accuracy when estimated from a minimum of four well-distributed sensors. Accuracy improves when more sensors are available enabling a greater sampling of the seismic radiation (e.g., Glazer, 2016).

Many mines utilize a local magnitude, such as Richter or Nuttli magnitudes, or some empirically derived local magnitude that uses a combination of magnitude and energy values extracted from the seismic waveforms. These local magnitudes require calibration and may not be easily comparable between mine sites. Magnitude and energy provide information on the damage potential of a seismic event. Source size estimates are useful in terms of assessing regions that may have been damaged as well as improving ground support design. Stress parameters, such as apparent stress, estimate the stress at the source based on the energy release and moment and can be considered a proxy for stress changes in the rock mass (see Figure 7c). The ES/EP ratio (i.e. ratio of the shear energy to the compressional energy) represents a somewhat controversial source parameter which has been used to estimate the degree of tensile (or non-shear) failure (e.g., Morkel et al., 2019). Although subjective, the ES/EP ratio can provide a quick estimate of the failure mechanism, whether failure is predominantly shear or composed of volumetric/tensile components (e.g., Simser & Butler, 2022).

Figure 7: Evaluation of source parameters: (a) three-component waveforms windowed around the P- (red) and S-wave (blue) phases as well as sampling signal noise (green), (b) example displacement spectra showing data and model fits, and (c) example data consisting of 725 events over a two week period with the seismic events sized according to magnitude and contoured according to apparent stress (red being high and blue being low apparent stress).

Mechanism geometry:

To evaluate the geometric aspect of failure, the seismic moment tensor (e.g., Gilberd 1970; Backus & Mulcahy 1976a,b) has been used effectively in mine seismology to understand source kinematics. Studies of large mine tremors have confirmed that seismic events generally occur because of slip along pre-existing failure planes (e.g., McGarr, 1971; Spottiswoode, 1984). However, due to the proximity of excavation surfaces to rock mass instabilities, inelastic failure can generate the potential for non-shear sources (e.g., Wong & McGarr, 1990; Feignier & Young, 1992; McGarr, 1992; Baker & Young, 1997). These non-double-couple sources are expected to have significantly large volumetric components of failure (i.e. net change in rock volume at the earthquake source), typically 30% or more.

From an operational perspective, the ability to provide fast (e.g., Trifu et al., 1998) as well as automated (e.g., Smith-Boughner et al., 2020) seismic moment tensor inversion allows for assessing the orientation of failure as well as distinguishing between shear-type fault damage and volumetric/tensile type damage that occur near excavation surfaces. This is important because it allows mine sites to assess regions within the mine that may have experienced damaging rockbursts or strainbursts as well as hazards related to uncontrolled slip along faults within the mine. Figure 8a and b show examples of moment tensor solutions (represented as beachballs) as well as orientations of incremental compressional strain directions.

Although moment tensor inversion provides information about source mechanism, there is ambiguity in terms of fracture plane orientation and slip direction. Utilizing clusters of moment tensors, the stress state can be estimated to yield the direction of slip relative to the shear stress on the fracture plane most likely to fail (e.g., Gephart and Forsyth,1984; Vavrycuk, 2014) as shown in Figure 8c. Furthermore, stress inversion allows estimating the local evolution of the principal stress orientations and relative stress magnitudes that can be compared with geomechanical models or other non-seismic measurements. Ongoing development of moment tensor theory is focused on enhancements to the moment tensor approach where the simplifying assumption of the point-source approximation breaks down. This typically occurs where seismic events occur near excavation surfaces where boundary effects are non-negligible and there are more degrees of freedom available for rock deformation (e.g., Linzer et al., 2020, Malovichko, 2022).

Figure 8: Application of moment tensor and stress inversion: (a) beach ball representation of inverted moment tensors from a mine seismic dataset assisting in the interpretation of fault slip, (b) distribution of P-axes from a cluster of moment tensor inversions as a proxy for incremental strain, and (c) results from stress inversion using the cluster of moment tensor solutions to estimate clockwise from top left) the compressional (P) and tensile strain (T) axes, failure planes and failure mechanisms using the Hudson plot representation.

Aseismic tools

Distributed Acoustic Sensing (DAS) technology has been gaining interest in mine geotechnical engineering. DAS offers unique capabilities in that both high frequency data (up to 10 kHz) to detect seismicity and low frequency data (down to 0.001 Hz) to detect slow-strain (i.e., relative strain) can be achieved with a single interrogator unit (e.g., for underground mining applications see Dande et al., 2024; Dande et al., 2025). DAS fibre can be deployed within mines to compliment discrete sensor (e.g., geophone and accelerometer) based systems, where sensor deployment is too challenging or costly (such as in the hanging wall of steeply dipping orebodies where most infrastructure and access for sensor installation is in the footwall). Furthermore, relative-strain measurements allow for capturing aseismic deformation which can be used to calibrate geomechanical models as well as monitor the rock mass behaviour during mine development. Figure 9 shows the results of a pilot DAS project at Onaping Depth mine (adapted from Dande et al., 2024). Figure 9a is an example of a blast signal recorded along a fibre as well as existing seismic system. Figure 9b compares event and blast locations using only the seismic system (red) and using both the seismic and fibre systems (blue). Figure 9c displays a heatmap of the DAS energy computed from 1-10 Hz slow-strain along the uphole and downhole fibre lines. Although deployed fibre may break as microcracking initiates (i.e., when strains reach mm/m) along the fibre, slow strain provides the ability to map rock mass heterogeneity through observations of strain behaviour.

Figure 9: DAS pilot project at Onaping Depth mine (adapted from Dande et al., 2024): (a) Blast signal recorded by both fibre and seismic system, traces are sorted by blast distance to sensors, (b) comparison of event locations using seismic system only (red) and using both seismic and fibre data (blue), and (c) heatmap of DAS energy data computed using 1-10 Hz slow-strain data.

Conclusions

Mine seismology is primarily a safety tool, focused on alerting and early warning to support safe mining operations, and re-entry after large events or blasting while at the same time helping to minimize operational downtime. With high-quality seismic catalogues, characterization of the rock mass through seismicity rates and locations, evaluation of failure mechanisms, and mapping of the seismic velocity and signal attenuation fields can assist in validating designs and adjustments to mine sequences and rates. DAS fibre-optic seismic and slow-strain measurements show potential as cost-effective monitoring enhancements. Integration of seismicity with geomechanics allows validation and calibration of numerical models and improved understanding of the rock mass response, as well as testing of hypotheses.

Acknowledgements

The author would like to thank colleagues at ESG Solutions for their invaluable contribution to supporting the material presented in this paper as well as the CSEG Distinguished Lecture Tour Seismic and Aseismic tools for mine safety, 2025-2026: Willem de Beer (Australia), Septian Prahastudhi (Indonesia), Dave Collins (Canada), as well as Suresh Dande and Denise Furtado (USA). The author also thanks Brad Simser (Glencore) and Alex Hall (Agnico-Eagle), and David Beck (Beck Engineering) for their feedback on DAS technology from an operator and consultancy perspective.

References

Angus, D.A., A. Aljaafari, P. Usher and J.P. Verdon (2014) Seismic waveforms and velocity model heterogeneity: towards full-waveform microseismic location algorithm, Journal of Applied Geophysics, 111, 228-233.

Angus, D.A., S. Prahastudhi, E. Bailey, R. William, W. de Beer, D. Collings and A. Senso (2026) Application of time-lapse passive seismic tomography in cave mining, SANIRE Symposium, Kempton Park, Gauteng, South Africa, 7-8 July.

Bailey, E., Y. Yanagimura, E. Martyniuk, D. Collins and D. Angus (2026) Time-lapse passive seismic tomography at New Afton C-Zone Mine, Caving 2026, Ulaanbaatar, Mongolia, 11-13 August.

Backus, G. and M. Mulcahy (1976a), Moment tensors and other phenomenological descriptions of seismic sources – I. Continuous displacements, Geophys. J. R. astr. Soc., 46, 341–362.

Backus, G. and M. Mulcahy (1976b) Moment tensors and other phenomenological descriptions of seismic sources – II. Discontinuous displacements, Geophys. J. R. astr. Soc., 47,301–330.

Baker, C. and R.P. Young (1997) Evidence for extensile crack initiation in point source time-dependent moment tensor solutions, Bull. Seism. Soc. Am., 87, 1442–1453.

Brune, J.N. (1970) Tectonic stress and the spectra of seismic shear waves from earthquakes, Journal of Geophysical Research, 75, 4997-5009.

Cai, M.C., P.C. Kaiser and D. Martin (1998) A tensile model for the interpretation of microseismic events near underground openings, Pure and Applied Geophysics, 153, 67-92.

Collins, D.S., R. Noorani, K. Hutton and D. Angus (2026) Examples of hazard identification using induced seismicity monitoring and analysis in underground mining, 60th US Rock Mechanics/Geomechanics Symposium, Tucson, USA, 21-24 June, 368.

Collins, D.S., B. Simser, P. Yadav, T. Butler and D. Angus (2025) Determination of maximum slip velocity as a possible rock burst hazard parameter, Proc. 11th International Symposium on Rockbursts and Seismicity in Mines, Lule, Sweden, 24-29 August, 651-665.

Dande, S., E. Bailey, E. Forges, T. Butler, A. Hall and B. Simser (2025) Integrating fiber-optic DAS and conventional seismic network for SMTI: a mining case study, Proc. 11th International Symposium on Rockbursts and Seismicity in Mines, Lule, Sweden, 24-29 August, 434-449.

Dande, S., E. Forbes, T. Butler, K. Graham, C. Hawryluck, A. Hall, B. Simser, R. Chauvet and A. Cherubini (2024) Utilising distributed acoustic sensing for monitoring rock mass stress conditions in underground mining: a case study, Deep. Mining, Montreal, Quebec, Canada, 375-390.

Feignier, B. and R.P. Young (1992) Moment tensor inversion of induced microseismic events: evidence of non–shear failures in the -4 ≤ M ≤ -2 moment magnitude range, Geophys. Res. Lett., 19, 1503–1506.

Gephart, J.W. and D.W. Forsyth (1984) An improved method for deter-mining the regional stress tensor using earthquake focal mechanism data: application to the San Fernando earthquake sequence, Journal of Geophysical Research, 89, 9305–9320.

Gibowicz, S.J. and A. Kijko (1994) An introduction to mining seismology, Academic Press.

Gharti, H.N., V. Oye, M. Roth and D. Kuhn (2010) Automated microearthquake location using envelope stacking and robust global optimization, Geophysics, 75(4), MA27-MA46.

Gilbert, F. (1970) Excitation of the normal modes of the earth by earthquake sources, Geophys. J. R. astr. Soc., 22, 223–226.

Glazer, S.N. (2016) Mine seismology: data analysis and interpretation. Palabora mine caving process as revealed by induced seismicity, Springer.

Linzer, L.M., M.W. Hildyard, S.M. Spottiswoode and J. Wesseloo (2020) Do stopes contribute to the seismic source? Journal of the Southern African Institute of Mining and Metallurgy, 120(1), 81-90.

Lomax, A., A. Michelini and A. Curtis (2011) Earthquake location, direct, global-search methods, Encyclopedia of Complexity and System Science, Part 5, 2449-2473.

Malovichko, D. (2022) Description of seismic sources in underground mines: theory, Bull. Seism. Soc. Am., 110(5), 2124-2137.

McGarr, A. (1971) Violent deformation of rock near deep–level tabular excavations–seismic events, Bull. Seism. Soc. Am., 61, 1453–1466.

McGarr, A. (1992) An implosive component in the seismic moment tensor of a mining–induced tremor, Geophys. Res. Lett., 19, 1579–1582.

Mendecki, A.J. (2019) Simple GMPE for underground mines, Acta Geophysica, 67, 837-847.

Mercer, R.A. and W.F. Bawden (2005) A statistical approach for the integrated analysis of mine-induced seismicity and numerical stress estimates, a case study – Part I: Developing the relations, International Journal of Rock Mechanics and Mining Science, 42, 47-72.

Montalbetti, J.M. and E.R. Kanasewich (1970) Enhancement of teleseismic body phases with a polarization filter, Geophys. J. R. astr. Soc., 21, 119-129.

Morkel, I.G., J. Wesseloo and Y. Potvin (2019) The validity of Es/Ep as a source parameter in mining seismology, Proceedings of the 9th International Conference on Deep and High Stress Mining, Johannesburg, South Africa, 385-398.

Simser, B., R. Deredin, A. Jalbout and T. Butler (2015) Use of microseismic monitoring data as an aid to rock mechanics decision making and mine design verification, 49th US Rock Mechanics/Geomechanics Symposium, San Francisco, California, 28 June – 1 July, 374.

Simser, B., and T. Butler (2022) The value of recording small mining induced microseismic events with examples from Glencore’s Nickel Rim South Mine, 10th Symposium on Rockbursts and Seismicity in Mines, Tucson, Arizona, 24–29 April.

Smith-Boughner, L., I. Nizkous, I. Leslie, S. Braganza, I. Pinnock, J. Matassa, V. Shumila, G. Viegas and D. Angus (2020) A fast, automated seismic moment tensor inversion approach for mine-induced microseismic data, First Break, 38, 75-82.

Spottiswoode, S.M. (1984) Source mechanisms of mine tremors at Blyvooruitzicht gold mine, Rockbursts and seismicity in mines, eds. N.C. Gay and E.H. Wainwright, Symp. Ser.No. 6, S. Afr. Inst. Min. Metal., Johannesburg, 29–37.

Trifu, C-I., D. Angus and V. Shumila (1998) A fast evaluation of the seismic moment tensor for induced seismicity, Bull. Seis. Soc. Am., 90(6), 1521-1527.

Usher, P.J., D.A. Angus and J.P. Verdon (2013) Influence of velocity model and source frequency on microseismic waveforms: some implications for microseismic locations, Geophysical Prospecting, 61, 334-345.

Vavrycuk, V. (2014) Iterative joint inversion for stress and fault orientation from focal mechanisms, Geophysical Journal International, 199,69-77.

Withers, M., R. Aster, C. Young, J. Beiriger, M. Harris, S. Moore and J. Trujillo (1998) A comparison of selected trigger algorithms for automated global seismic phase and event detection, B.S.S.A., 88(1), 95-106.

Wong, I.G. and A. McGarr (1990) Implosional failure in mining-induced seismicity: a critical review, Rockbursts and seismicity in mines, ed. C. Fairhurst, Balkema, Rotterdam, 45–51.

About the Author

Doug Angus is VP of Mining at ESG Solutions, a company that provides world-class acquisition and data services for real-time rock-mass characterization to improve mine safety and operational efficiency. He received his BSc(Eng), MSc(Eng) and PhD from Queen’s University, gaining his MSc in applied seismology focused on moment tensor inversion of mine seismic data and gaining his PhD in theoretical seismology focused on wave propagation in anisotropic heterogenous media under the supervision of Colin Thomson and Gerhard Pratt. He spent a year at New Mexico State University, Las Cruces, working on lithosphere/asthenosphere imaging. From 2006-2016, he worked as an academic focused on integrating seismology with geomechanics, rock physics and multi-phase fluid flow; first as a post-doctoral research associate at Bristol University (2006-2008) and subsequently as Associate Professor of Seismic Geomechanics at the University of Leeds (2008-2016).