The standard processing of seismic-monitoring data includes location of seismic sources and the evaluation of basic source parameters (i.e. seismic potency and radiated energy). In some cases, the seismic data is of high quality, making assessment of advanced characteristics of sources possible. The routine evaluation of source mechanisms improves the interpretation of recorded data in terms of geological structures and stress field.

Regular assessment of seismic and ground motion hazard is important at rockburst-prone mines. This requires a consistent seismic database. The modelling of seismicity for past and future mining steps helps to interpret the recorded seismic data and allows for the future extension of the forecast of seismic hazard compared to the forecast based on observed seismicity only.

IMS takes responsibility for maintaining the seismic database according to state-of-the-art standards. The location and source parameters of historical events are updated as the processing algorithms are improved. This eliminates spurious sudden changes in the source parameters marked by arrows in the following time history. Velocities and quality factors of seismic waves are regularly revisited and updated as necessary. Cloud storage of the seismic database is established.

The spatial distribution of probabilities of exceedance of potentially damaging ground motion is calculated based on relevant recorded seismic data. The calculations utilise ground motion prediction equations (GMPE) for peak ground velocity (PGV) or cumulative absolute displacement (CAD), which needs to be calibrated for the mine.

This method is described in the following work:

Mendecki, A.J., 2018. Mapping seismic ground motion hazard in mines, Keynote lecture. Proc. of the 9th Rockburst and Seismicity in Mines Symposium, Santiago, Chile.

Routine Evaluation of Source Mechanisms

This includes automatic evaluation of the seismic source mechanisms for all appropriate events, and estimation of location uncertainty of all events. An experienced seismologist manually evaluates the source mechanisms of large events.

Routine Modelling and Interpretation of Seismicity (monthly)

Modelling of stresses induced by mining and associated expected seismicity is performed monthly. The recorded seismicity is compared to the modelled one, which facilitates the interpretation.

This method is described here:

Malovichko, D. and Basson, G., 2014. Simulation of mining induced seismicity using Salamon-Linkov method, Proc. of the 7th International Conference on Deep and High Stress Mining, Sudbury, Canada (eds. M. Hudyma and Y. Potvin), ACG, pp. 667-680.

Assessment of Seismic and Ground Motion Hazard for Planned Mining (quarterly)

The spatial distribution of probabilities of occurrence of potentially damaging seismic events or associated ground motion is evaluated for the planned mining sequence. Usually this is done every quarter for four–eight future quarterly mining steps. The forecast of seismic hazard issued in the previous quarter is tested against the observed seismic data.

This method is described here:

Malovichko, D., 2017. Assessment and testing of seismic hazard for planned mining sequences, Proc. of the 8th International Conference on Deep and High Stress Mining, Perth, Australia (ed. J. Wesseloo), ACG, pp. 61-77.

SCAN is revolutionary new technique that continuously monitors the rock mass response to mining by using the seismic noise generated by mining activities. The method uses the seismic noise generated by mining activity to measure miniscule (0.01%) changes in seismic velocity. The seismic velocities are calibrated in terms of known stress, such as atmospheric air pressure change. This method effectively turns each sensor into a volumetric stress-meter. As this method requires no additional hardware, and only uses continuous seismic data, SCAN is immediately available to existing IMS customers. SCAN does not require any microseismic events, and continuously provides spatial stress change information. This improves seismic our understanding of how the rock mass is responding to mining.

References
Olivier, G., Brenguier, F., Campillo, M., Roux, P., Shapiro, N.M. and Lynch, R., 2015. Investigation of coseismic and postseismic processes using in situ measurements of seismic velocity variations in an underground mine, Geophysical Research Letters, 42.

Olivier, G., Brenguier, F., Campillo, M., Lynch, R. and Roux, P., 2015. Body-wave reconstruction from ambient seismic noise correlations in an underground mine, Geophysics, 80 (3).

Noise from mechanical sources and rock falls from ore passes are typically recorded at mines. These types of ground vibrations do not add value for seismic analysis and interpretation but add to the amount of data that needs to be manually processed. Ideally these would not be recorded in the first place but as that is not always possible. The objective is to automatically remove these events from the database.

The available information for discrimination before processing includes:

• number of triggers
• trigger times
• sensor positions
• trigger duration
• maximum amplitude
• dominant frequency

Combining some of these parameters may give patterns that match mine operations. For example, trigger time and sites triggered may match the time and location of working ore passes. Rock falls could also be identified by looking at the combination of dominant period and maximum amplitude, as these are typically low frequency and do not generate high ground motions.

These parameters can be calibrated per mine and used to filter events as they are brought into the database.