Observation of numerous aftershocks of an M w 1.9 earthquake with an AE network installed in a deep gold mine in South Africa (original) (raw)
Related papers
Proceedings of the Ninth International Conference on Deep and High Stress Mining, 2019
In 2014, a M5.5 earthquake ruptured the range of depths between 3.5 km and 7 km near Orkney, South Africa. The main and aftershocks were very well monitored in the nearfield by dense, surface, strong motion meters and a dense underground seismic network in the deep gold mines. The mechanism of this M5.5 earthquake was left-lateral strike-slip faulting, differing from typical mining-induced earthquakes with normalfaulting mechanisms on the mining horizons shallower than 3.5 km depth. To understand why such an unusual event took place, the aftershock zone was probed by full-core NQ drilling during 2017-2018, with a total length of about 1.6 km, followed by in-hole geophysical logging, core logging, core testing, and monitoring in the drilled holes. These holes also presented a rare opportunity to investigate deep life. In addition, seismogenic zones of M2-M3 earthquakes were probed on mine horizons that were also very well monitored by acoustic emission networks. This paper reviews the early results of the project.
Journal of Geophysical Research: Solid Earth, 2015
We observed very small repeating earthquakes with À5.1 ≤ M w ≤ À3.6 on a geological fault at 1 km depth in a gold mine in South Africa. Of the 851 acoustic emissions that occurred on the fault during the 2 month analysis period, 45% were identified as repeaters on the basis of waveform similarity and relative locations. They occurred steadily at the same location with similar magnitudes, analogous to repeaters at plate boundaries, suggesting that they are repeat ruptures of the same asperity loaded by the surrounding aseismic slip (background creep). Application of the Nadeau and Johnson (1998) empirical formula (NJ formula), which relates the amount of background creep and repeater activity and is well established for plate boundary faults, to the present case yielded an impossibly large estimate of the background creep. This means that the presently studied repeaters were produced more efficiently, for a given amount of background creep, than expected from the NJ formula. When combined with an independently estimated average stress drop of 16 MPa, which is not particularly high, it suggests that the small asperities of the presently studied repeaters had a high seismic coupling (almost unity), in contrast to one physical interpretation of the plate boundary repeaters. The productivity of such repeaters, per unit background creep, is expected to increase strongly as smaller repeaters are considered (∝ M o À1/3 as opposed to M o À1/6 of the NJ formula), which may be usable to estimate very slow creep that may occur on intraplate faults.
Tectonophysics, 2018
We investigated the activities of very small (−5.10 ≤ M W ≤ −2.35) repeating earthquakes that occurred during 14 months on a geological fault at a 1 km depth in the Cooke 4 gold mine in South Africa. Out of 4214 acoustic emissions occurring on the fault, 1785 (42.36% of the total) were identified as repeaters on the basis of the waveform similarity and the proximity of hypocenters. Quite a few large groups, having up to 57 members, were recognized. Activities of some groups certainly continued throughout the 14 months, but we also found some repeater groups that newly emerged or disappeared during this period. Both types often coexisted within a distance as small as~0.5 m, within which background fault creep should be coherent. Some disappearing groups showed a decrease in the event magnitude with time. These temporal changes of repeater activities may imply the formation and dissipation of topographical asperity contacts on the fault. Our results suggest that ultrasensitive observation of acoustic emissions can reveal the physical evolution of asperities.
Stress drops and radiated seismic energies of microearthquakes in a South African gold mine
Journal of Geophysical Research, 2007
1] We estimate stress drops and radiated seismic energies of 20 microearthquakes (0.0 < M W < 1.3) in a South African gold mine to investigate their rupture characteristics and scaling relationships to large earthquakes. We analyze seismograms of borehole accelerometers recorded with high sampling rate (15 kHz) within 200 m of the hypocenters at the depth of 2650 m. The waveform data have very high signal-to-noise ratio and no significant later phases are observed at all stations. Corner frequencies and quality factors of the anelastic attenuation Q are estimated from spectra of velocity seismograms by assuming the omega squared model of . We also investigate moment tensors for double couple solutions and volumetric components from the waveform inversion. Static stress drops of the 20 earthquakes calculated from the model of Madariaga (1976) are from 3.2 to 88 MPa and scaled energies (= E R /M o ; the ratio of the radiated energy E R to the seismic moment M o ) are from 4.2 Â 10 À6 to 1.1 Â 10 À4 . We find that both the static stress drops and the scaled energies of the analyzed earthquakes are comparable to those values of larger earthquakes. Our results indicate that the dynamic rupture processes of these microearthquakes are similar to those of larger earthquakes.
2012
Mining-induced earthquakes pose a risk to workers in deep mines, while natural earthquakes pose a risk to everywhere, but especially near plate boundaries. A five year Japanese-South African collaborative project entitled 'Observational studies to mitigate seismic risks in mines' commenced in August 2010. Here we report on the achievements of the first 18 months of the project. Faults at Ezulwini, Moab-Khotsong and Driefontein gold mines considered likely to become seismically active during mining activity were modelled using pre-existing geological information supplemented by cores and camera images from new boreholes. As of 30 January 2012, about 90% of about 70 planned boreholes totalling more than 2 km in length had been drilled at project sites to locate fault zones accurately and to deploy sensors. Acoustic emission sensors, geophones, accelerometers, strain-and tilt meters, and controlled seismic sources were installed to monitor the deformation of the rock mass, the accumulation of damage during the earthquake preparation phase, and changes in stress produced by the propagation of the rupture front. The suite of sensors has greater sensitivity and dynamic range than those typically used in civil or mining engineering applications, making it possible to record very small changes in stress and strain as well as violent rock mass deformation associated with large seismic events. These data sets will be integrated with measurements of stope closure, strong ground motion in stopes, and seismic data recorded by the mine-wide network.
Radiation efficiency and apparent stress of small earthquakes in a South African gold mine
Journal of Geophysical Research, 2005
1] Nine triaxial borehole accelerometers were installed within 200 m along a 2650-m-deep haulage tunnel in the Mponeng gold mine in South Africa. We analyzed the high sample rate recordings (15 kHz) to determine source parameters of small earthquakes in the mine. To study the source processes, we carried out multiple time window waveform inversions for five larger events (0.8 < M < 1.4) that occurred within 150 m of the stations. From the inversion results we could determine the fault planes for all five events and estimate the range of rupture speed. We can conclude that rupture speeds were faster than 2.5 km/s (65% of the shear wave velocity). The radiation efficiency is written as a function of the rupture speed and becomes greater with increasing rupture speed. This study indicates that radiation efficiencies of small earthquakes in the South African gold mine are almost equal to those of larger natural earthquakes. We also calculated radiated seismic energies and static stress drops of the five events to investigate their apparent stresses. Apparent stresses of the five events were from 1 to 10 MPa and static stress drops were 1 to 20 MPa. We found that the source parameters (rupture speed, apparent stress, radiation efficiency, and static stress drop) did not largely differ from values for larger natural earthquakes. This suggests that the dynamic rupture processes of these small events were similar to those of the larger natural earthquakes.
Multidisciplinary Monitoring of the Entire Life Span of an Earthquake in South African Gold Mines
2005
deep gold mines. In the second field experiment, the authors successfully monitored the entire strain history within a hundred metres from the hypocentres, associated with a few seismic events with M>2. However, there were no close strong-motion meters available to locate asperities; only a single strainmeter was available, so the authors were not able to locate the strain-change source; no in situ stress measurements were carried out at the site, and no information was available to constrain strength. In order to address these deficiencies, from 2003 to 2004, the authors deployed new experimental instrument arrays at fault bracket/stabilising pillars. Multiple strainmeters, arrays of strong ground-motion meters, sensitive thermometers to monitor seismic heat generation, and fault displacement meters were installed. Successful monitoring began, but the authors learnt that they had to develop instruments for much quicker drilling and installation, especially at highly stressed pil...
Comparison of ground motion from tremors and explosions in deep gold mines
Journal of Geophysical Research, 1990
Seismic body waves, from tamped chemical explosions, two with yields of 50 and one of 150 kg, were compared with corresponding data from three mining-induced tremors with a view to testing methods of discriminating between the two types of events. Detonated at depths of about 2 km, all three explosions generated P waves for which the low-frequency spectral asymptotes agree well with corresponding results scaled down from nuclear shots at the Nevada Test Site. For the two smaller explosions, recorded underground at hypocentral distances ranging from 234 to 871 m, the P wave corner frequencies, defined by the intersections of the low-and high-frequency asymptotes, are also in fine agreement with scaled-down results from the Nevada Test Site; for the 150-kg explosion, which was recorded only at the surface, it appears that 2.39 km of upward propagation caused the corner frequency to be at least a factor of 4 lower than anticipated due to attenuation. All three explosions generated S waves that appear to be a consequence of deviatoric stress release in the immediate environs of the explosions. The three tremors analyzed here generated P and S waves whose spectra and source parameters agree well with standard earthquake source models. With regard to discrimination based on P wave spectra, we conclude that for events of fixed low-frequency spectral asymptotes, the explosions typically have higher corner frequencies than tremors or earthquakes, although counterexamples certainly exist. Interestingly, the 150-kg explosion was identified as such on the basis of P and S wave polarities that are incompatible with the normally expected double-couple source model; instead, these initial motions are consistent with an explosion in conjunction with normal faulting. The body wave spectra of this explosion and those of a nearby tremor, however, were indistinguishable.
Source parameters of acoustic emission events and scaling with mining-induced seismicity
Journal of Geophysical Research, 2003
1] A series of uniaxial compression tests were performed on 96-mm-diameter quartzite samples, 242 mm in length, to understand the scaling of rock fracture processes. Nine acoustic emission (AE) sensors glued to each sample monitored the AEs resulting from microcracking within the samples. In contrast to previous AE studies, the sensors were calibrated as velocity transducers so that the output could be compared to mining-induced seismicity and natural earthquakes. A new hybrid, relative moment tensor method was applied to obtain source mechanism solutions for eight clusters of events. Once the AE rate accelerated prior to failure, the event positions are associated with the observed failure planes. The moment tensors were found to have double-couple components, indicating that shearing was occurring. The stress drop appears to be constant over the range of moments suggesting the self-similar scaling of the fracture response from the laboratory sample to mine seismicity and natural earthquakes over a wide range of length scales. Similar conclusions can be drawn by considering the apparent stress and source radius. The scaling of the peak velocity and peak acceleration parameters is apparently consistent with mining-induced seismicity but is considerably affected by high-frequency attenuation and the limited bandwidth. This is confirmed by the frequency-magnitude plot, which has a slope of unity. The fracture processes in the laboratory are similar to those occurring underground near stope faces and pillars in deep-level South African gold mines, where there is a high vertical compression and low confinement. Citation: Sellers, E. J., M. O. Kataka, and L. M. Linzer, Source parameters of acoustic emission events and scaling with mininginduced seismicity,
Journal of Geophysical Research: Solid Earth, 2015
Using a network of sensitive high-frequency acoustic emission sensors, we observed foreshock activity of an Mw 2.2 earthquake (main shock) in a deep gold mine in South Africa. Foreshock activity, which selectively occurred on a part of the rupture plane of the forthcoming main shock, lasted for at least 6 months until the main shock. Rock samples recovered from the main shock source region showed evidence of ancient hydrothermal alteration on the main shock rupture plane, suggesting that the foreshock activity occurred on a preexisting weakness. The foreshocks during 3 months before the main shock were concentrated in three clusters (F1-F3), which we interpret as representing localized preslip at multiple sites. While the location of mining area, the source of stress perturbations, changed with time, the locations of foreshock clusters did not change, suggesting that the preslip patches were controlled by strength heterogeneity rather than stress distribution. Activity over the entire foreshock area was generally constant, but the largest cluster (F2) showed accelerated activity starting at least 7 days before the main shock, while mining stress did not increase in this period. The main shock initiated at a point close to F1, away from F2. All the six foreshocks during the final 41 h occurred in F1 and F2 and in-between. These suggest that in the last stage of the preparation process of the main shock, preslip patches interacted with each other through the stress concentration ahead of the expanding preslip patch (F2), which should be the only driving force of the preparation process under the constant external load.