Deep South African Gold Mines Research Articles

Characteristics of Fault Rocks Within the Aftershock Cloud of the 2014 Orkney Earthquake (M5.5) Beneath the Moab Khotsong Gold Mine, South Africa

AbstractCores recovered during the International Continental Scientific Drilling Program project “Drilling into Seismogenic zones of M2.0 to M5.5 earthquakes in deep South African Gold Mines” include fault breccia from within the aftershock cloud of the 2014 Orkney earthquake (M5.5). The breccia and surrounding intrusive rocks, probably lamprophyres rich in talc, biotite, calcite, and amphibole, had high magnetic susceptibilities owing to the presence of magnetite. All of these characteristics can be attributed to fluid‐related alteration. Both the breccia and the lamprophyres had low friction coefficients and showed evidence of velocity strengthening, which is inconsistent with the occurrence of earthquakes. Variable amounts of talc, biotite, calcite, and amphibole within the lamprophyres might have produced complex frictional properties and spatial heterogeneity of fault stability. The altered lamprophyres may be the host rocks of the 2014 Orkney earthquake, but frictional complexity may have governed the magnitudes of the main‐ and aftershocks and their distributions.

Integration of underground mapping, petrology, and high-resolution microseismicity analysis to characterise weak geotechnical zones in deep South African gold mines

A highly-stressed shaft pillar is prone to large seismic events, falls of ground and rockbursts, which may cause injuries and loss of production, especially in weak geotechnical zones. It is thus important to identify weak geotechnical zones in order to mitigate risks. In this study, we present integrated studies (underground mapping, petrology, rock mechanics and high-resolution microseismicity analysis) to understand the different geotechnical zones in the shaft pillar of Cooke 4 mine in South Africa. The footwall of the remnant shaft pillar comprises the Upper Elsburg reef of the Mondeor Formation, while the Ventersdorp Contact Reef (VCR) of the Venterspost Formation and soft/weak lavas of the Westonaria Formation form the hangingwall. Results from underground mapping and microscopic analysis show that the shaft pillar is composed of quartzites, pebbly quartzites, argillaceous quartzites and conglomerates. Underground mapping further shows that the shaft pillar is characterized by several discontinuities, which vary from minor to macro scale fractures. Laboratory uniaxial compressive strength (UCS) tests indicate that quartzite has the strongest strength, followed by pebbly quartzite, argillaceous quartzite and lastly, conglomerate. Analysis of high-resolution acoustic emissions (AEs) clusters indicates that the majority of AEs are associated with the mining stope faces. The clusters show the formation of Ortlepp shears ahead of the stope, which is caused by the excavation-induced stress field. Microseismic data further reveal that the fracture turning-point occurs in the soft strata (weak hangingwall lavas). The integration of these datasets has allowed us to develop the fracture model for different geotechnical zones, which concurs with previous models developed for the similar underground environment (i.e., weak/soft lava hangingwall and quartzite/conglomerate footwall). This has major implications for future mining, support, production and safety.

Rockburst support in shallow-dipping tabular stopes at great depth

This paper investigates the unique problems associated with the design of rockburst support for shallow-dipping tabular excavations. These designs are particularly problematic when the stoping width is very small. In steep dipping orebodies, the layouts and mining methods can be selected to ensure that miners never enter the stopes. Only the access drives need to be protected by rockburst resistant support. In shallow-dipping orebodies, this problem is more difficult as miners enter the stopes and the entire hanging wall needs to be supported. A simple analytical model is used to investigate the implications for support design as a result of the convergence associated with the tabular geometry and the possibility of rocks being ejected during a rockburst. This illustrates that a support system is required that is initially stiff, but it should also be yieldable to survive the convergence in the back areas. By trial and error, the historic support solutions in the South African gold mines evolved into a system of timber packs and elongates to meet these requirements. When considering the three accepted key functions of modern rockburst support methodology namely, reinforce, retain and hold, the typical support design for these shallow-dipping orebodies does not meet all these requirements. The rock is highly fractured and fallouts occur between roofbolts during rockbursts as areal support is difficult to implement. Steel mesh is not used as it is often destroyed during cleaning operations. Solving this problem is of critical importance to ensure the viability of deep South African gold mines in the future.

A limit equilibrium fracture zone model to investigate seismicity in coal mines

This paper explores possible synergies between techniques used to minimise seismicity in deep South African gold mines and their applicability to control coal bumps. The paper gives a summary of the techniques used in the deep gold mines and a critical appraisal if these are useful in coal mines. The techniques typically include control of mining rate, preconditioning, optimisation of extraction sequences and centralised blasting. Of particular interest to the coal bump problem is an experimental limit equilibrium fracture zone model implemented in a displacement discontinuity code. This was recently developed for the gold mines to enable the interactive analysis of complex tabular mine layout extraction sequences. The model specifically accommodates energy dissipation computations in the developing fracture zone near the edges of these excavations. This allows the released energy to be used as a surrogate measure of ongoing seismic activity and addresses a number of the weaknesses in the traditional usage of this quantity as a criterion for the design of seismically active layouts. This paper investigates the application of the model to a hypothetical coal longwall layout and the specific problem of coal bumps.

Detection of potential methane gas pathways in deep South African gold mines

At a number of gold mines in South Africa, methane gas has been encountered when drilling into faults and/or dykes extending to depths beyond 7000 m. Methane gas has been reported to migrate up through structures from within the basin to the mine working environments (∼3000 m depth) and caused explosions. The Booysens Shale is one of the possible source rocks for hydrocarbons and it forms the footwall to the gold-bearing Ventersdorp Contact Reef (VCR, ∼1.5 m thick). The Booysens Shale lies at depths between 3500 and 4500 m below the surface and can be best described as a base of the divergent clastic wedge, hosting the quartzite and conglomerate units that sub-crop against the VCR towards the west of the gold mining areas. Geometric attributes (dip and dip azimuth) and instantaneous attributes (phase, frequency and envelope) computed for the Booysens Shale and VCR horizons provide insight into structures that extend from the Booysens Shale into the overlying mining level (i.e. VCR). These attributes successfully mapped the structures (faults and dykes) that have displaced both the VCR and Booysens Shale horizons. The edge-detection attributes were particularly useful in delineating faults with throws below the traditional seismic resolution criteria (e.g. a quarter of the dominant wavelength) that were not visible in the conventional seismic interpretation. The structural analysis using seismic attributes gives a new visual representation of geological structures that may be probable conduits for hydrocarbons, particularly methane, migrating from the Booysens Shale to the VCR mining levels.

The management of heat flow in deep mines

AbstractWith the ever‐increasing depth of mines, the management of heat has become a key issue for their design and operation. There are two main sources of heat: heat transfer from the rock mass into the mine workings and heat associated with mining operations. The principles of heat transfer from the rock mass are discussed and basic relationships presented. Sources of heat linked to mining operations are discussed. It is shown that in deep‐level mines, heat transfer from the rock mass accounts more than 75 % of total mine heat load. In highly mechanized coal mines, heat from the use of mining machinery is also significant. Some models of heat flow prediction for deep gold mines are presented. It is shown that in the case of deep mines, control of heat flow is more important than increasing refrigeration capacity. Examples of heat flow management methods are given.Furthermore, the paper deals with the effects of heat on the human body and mine cooling strategies for deep mines. In detail the effects of heat on a worker are examined, heat stress and heat tolerance discussed and methods of assessing heat stress in different mining situations presented. Experiences from deep South African gold mines highlight the adverse effects of heat stress environment on safety and labour productivity. The principal methods of cooling of deep and ultra deep mines are discussed. It is shown that auto‐compression of ventilation air is a deciding factor governing the choice of surface or underground cooling of ventilation air. In the case of deep and ultra deep mines, the use of chilled service water and ice slurry has shown to be the most cost effective means of mine cooling. In the case of ice slurry as cooling medium advantage is taken of the latent heat of ice which significantly reduces the amount of water required for mine cooling and hence the cost of pumping the water to surface. Cooling strategies for moderately deep, deep and ultra deep mines are discussed. Examples of cooling of deep long mine tunnels are given.

3D edge detection seismic attributes used to map potential conduits for water and methane in deep gold mines in the Witwatersrand basin, South Africa

Inrushes of ground water and the ignition of flammable gases pose risks to workers in deep South African gold mines. Large volumes of water may be stored in solution cavities in dolomitic rocks that overlie the Black Reef (BLR) Formation, while there are several possible sources for methane, namely, coal seams, kerogen found in some gold ore bodies, or methane introduced by igneous intrusions. Potential conduits that may transport water and methane to underground workings were mapped using 3D reflection seismic data. Edge detection attributes successfully identified many faults, some with displacements as small as 10 m. Faults that displace the Ventersdorp Contact Reef (VCR) and the BLR horizons were of special interest, as known occurrences of fissure water and methane in underground workings show a good correlation with faults that were imaged on the VCR and BLR horizons. Because there are uncertainties in determining the linkage of faults with aquifers and methane sources, it is considered prudent to assume that all structures that displace the VCR and BLR horizons are potential conduits.

The management of heat flow in deep mines (part 2)

AbstractIn Part 1 of “The Management of Heat Flow in Deep Mines” the sources of heat, mechanism of heat transfer and strategies of controlling heat transfer have been discussed. Part 2 deals with the effects of heat on the human body and mine cooling strategies for deep mines. In detail the effects of heat on a worker are examined, heat stress and heat tolerance discussed and methods of assessing heat stress in different mining situations presented. Experiences from deep South African gold mines highlight the adverse effects of heat stress environment on safety and labour productivity. The principal methods of cooling of deep and ultra deep mines are discussed. It is shown that auto‐compression of ventilation air is a deciding factor governing the choice of surface or underground cooling of ventilation air. In the case of deep and ultra deep mines, the use of chilled service water and ice slurry has shown to be the most cost effective means of mine cooling. In the case of ice slurry as cooling medium advantage is taken of the latent heat of ice which significantly reduces the amount of water required for mine cooling and hence the cost of pumping the water to surface. Cooling strategies for moderately deep, deep and ultra deep mines are discussed. Examples of cooling of deep long mine tunnels are given.

Constraining the far-field in situ stress state near a deep South African gold mine

We present and test a new technique for determining the far-field virgin state of stress near the TauTona gold mine. The technique we used to constrain the far-field stress state follows an iterative forward modelling approach that combines observations of drilling-induced borehole failures in borehole images, boundary element modelling of the mining-induced stress perturbations, and forward modelling of borehole failures based on the results of the boundary element modelling. Using this approach, we constrained a range of principal stress orientations and magnitudes that are consistent with all the observed failures and other stress indicators. We found that the state of stress is a normal faulting regime ( S v * ⩾ S Hmax * ⩾ S hmin * ) with principal stress orientations that are slightly deviated from vertical and horizontal and, therefore, denoted with a (*). The maximum principal stress, S v * , is deviated ∼10° from vertical plunging towards the NNW with a magnitude gradient of ∼27 MPa/km. The intermediate principal stress, S Hmax * , is inclined ∼10° from horizontal plunging towards an azimuth of ∼156° and has a magnitude gradient of ∼24 MPa/km. The least principal stress, S hmin * , is inclined ∼5° from horizontal plunging towards an azimuth of 247° and has a magnitude gradient of ∼14 MPa/km. This stress state indicates that the crust is in a state of frictional faulting equilibrium, such that normal faulting is likely to occur on cohesionless pre-existing fault planes that are optimally oriented to the stress field. Modelling of breakout rotations and gaps in breakout occurrence associated with recent fault slip on critically stressed faults located >100 m from the mine further confirmed this stress state.

Constraining the Far-Field Stress State near a Deep South African Gold Mine

The deep gold mines in the Witwatersrand Basin of South Africa are some of the deepest underground mines in the world. For over a century, numerous, often deadly, mining induced earthquakes have been observed. In this work, we develop and test a new technique for determining the virgin state of stress near the TauTona gold mine. This work was completed as part of the Natural Earthquake Laboratory in South African Mines (NELSAM) project. The technique we use to constrain the far-field stress state follows an iterative forward modelling approach that combines observations of drilling induced borehole failures in borehole images, boundary element modelling of the mining-induced stress perturbations, and forward modelling of borehole failures based on the results of the boundary element modelling. The final result is a well constrained stress state consistent with all the observed stress indicators. We find that the stress state is a normal faulting regime with principal stress orientations that are slightly deviated from vertical and horizontal (denoted with a *). The maximum principal stress, Sv*, is deviated 0-20o from vertical towards the NNW and has a magnitude gradient of 27 ± 0.3 MPa/km. The intermediate principal stress, SHmax*, is inclined 0-20o from horizontal with an azimuth of 145o to 168o and has a magnitude gradient of 21.5 to 26 MPa/km. The least principal stress, Shmin*, is inclined 0-10o from horizontal with an azimuth of 235o to 258o and has a magnitude gradient of 13 to 15.5 MPa/km. This stress state indicates that the crust is in a state of frictional faulting equilibrium, such that normal faulting is likely to occur on pre-existing fault planes that are optimally oriented to the stress field. The stress concentrations caused by the mining activities can dramatically alter the range of fault orientations upon which fault slip could occur.

Stress change prior to the major events in the 1989 earthquake swarm off the eastern Izu Peninsula, Japan

AbstractWe investigate a temporal change in a stress parameter of earthquakes that occurred in the 1989 swarms off the eastern Izu Peninsula in Japan. We use the energy index (EI), which have been monitored in deep South African gold mines for predicting major events, as an estimate of apparent stress that is proportional to the ratio of seismic energy (E) to seismic moment (Mo).EImeasures an excess or shortage inEwith respect to the empirical relationship betweenEandMo. We check thatEIis almost proportional to the apparent stress for the ranges ofMoand frequency for our analysis, althoughEis underestimated due to the artifact of limited frequency band of monitoring. The largest events (M= 5.2 and 5.5) took place off the tip of a vertical crack, which opened associated with magma intrusion. While the opening continued to load the source areas of theM5.2 and 5.5, we find significant decrease inEIprior to the events. Based on the experimental result that the stress decreases when yielding takes place prior to final failure, we interpret this observation as yielding in the vicinity of the large earthquake hypocenters, following a rapid increase in stress caused by magma intrusion.

Characterization of iron- and sulphide mineral-oxidizing moderately thermophilic acidophilic bacteria from an Indonesian auto-heating copper mine waste heap and a deep South African gold mine.

Iron- and chalcopyrite-oxidizing enrichment cultures were obtained at 50 degrees C from acidic, high-temperature, copper/gold mine environments in Indonesia and South Africa. Over 90% copper yield was obtained from chalcopyrite concentrate with the Indonesian enrichment in 3 months with 2% solids concentration, when pH was maintained at around 2. Neither addition of silver cations nor an enhanced nutrient concentration influenced chalcopyrite leaching. Excision and sequencing of bands from denaturing gradient gel electrophoresis of the amplified partial 16S rRNA gene showed that the enrichment cultures from different environments in South Africa and Indonesia were very simple, and similar. Chalcopyrite concentrate supported a simpler and different community than Fe2+. The members of the enrichment cultures were closely related to Sulfobacillus yellowstonensis and Sulfobacillus acidophilus.

Identification of iron-reducing Thermus strains as Thermus scotoductus.

Thermus strain SA-01, previously isolated from a deep (3.2 km) South African gold mine, is closely related to Thermus strains NMX2 A.1 and VI-7 (previously isolated from thermal springs in New Mexico, USA, and Portugal, respectively). Thermus strains SA-01 and NMX2 A.1 have also been shown previously to grow using nitrate, Fe(III), Mn(IV) or S(O) as terminal electron acceptors and to be capable of reducing Cr(VI), U(VI), Co(III), and the quinone-containing compound anthraquinone-2,6-disulfonate. The objectives of this study were to determine the phylogenetic positions of the three known metal-reducing Thermus strains and to determine the phylogenetic significance of metal reduction within the genus Thermus. Phylogenetic analyses of 16S rDNA sequences, BOX PCR genomic fingerprinting, and DNA-DNA reassociation analyses indicated that these strains belong to the previously described genospecies T. scotoductus. The morphologies and lipid fatty acid profiles of these metal-reducing strains are consistent with their identification as T. scotoductus; however, the T. scotoductus strains tested in this study evinced a wide intraspecies variability in some other phenotypic traits, e.g., carbon substrate utilization and pigmentation. Iron reduction occurred in all strains of T. scotoductus tested except the mixotrophic, sulfur-oxidizing strain IT-7254. Thermus strains belonging to other species did not reduce Fe(III) to Fe(II) or reduced it only poorly.

Jacques Pierre Friederich Sellschop

Jacques Pierre Friederich Sellschop

Manuel Rocha Medal Recipient Simulating the Time-dependent Behaviour of Excavations in Hard Rock

Although hard rock is not usually associated with large creep deformation, data collected from the tunnels and stopes of the deep South African gold mines illustrates significant time-dependent behaviour. Apart from application in mining, a better understanding of the time-dependent behaviour of crystalline rock is required to analyse the long term stability of nuclear waste repositories and to design better support for deep civil engineering tunnels in these rock types. To illustrate the subtle problems associated with using viscoelastic theory to simulate the time-dependent behaviour of hard rock, a viscoelastic convergence solution for the incremental enlargement of a tabular excavation is discussed. Data on the time-dependent deformation of a tunnel developed in hard rock further illustrates the limitations of the theory, as it is unable to simulate the fracture zone around these excavations. To simulate the rheology of the fracture zone, a continuum viscoplastic approach was developed and implemented in a finite difference code. This proved more successful in modelling the time-dependent closure of stopes and squeezing conditions in hard rock tunnels. A continuum approach, however, has limitations in areas where the squeezing behaviour is dominated by the time-dependent behaviour of prominent discontinuities such as bedding planes. To overcome this problem, a viscoplastic displacement discontinuity technique was developed. This, combined with a tessellation approach, leads to more realistic modelling of the time-dependent behaviour of the fracture zone around excavations.

Archaeal diversity in waters from deep South African gold mines.

A culture-independent molecular analysis of archaeal communities in waters collected from deep South African gold mines was performed by performing a PCR-mediated terminal restriction fragment length polymorphism (T-RFLP) analysis of rRNA genes (rDNA) in conjunction with a sequencing analysis of archaeal rDNA clone libraries. The water samples used represented various environments, including deep fissure water, mine service water, and water from an overlying dolomite aquifer. T-RFLP analysis revealed that the ribotype distribution of archaea varied with the source of water. The archaeal communities in the deep gold mine environments exhibited great phylogenetic diversity; the majority of the members were most closely related to uncultivated species. Some archaeal rDNA clones obtained from mine service water and dolomite aquifer water samples were most closely related to environmental rDNA clones from surface soil (soil clones) and marine environments (marine group I [MGI]). Other clones exhibited intermediate phylogenetic affiliation between soil clones and MGI in the Crenarchaeota. Fissure water samples, derived from active or dormant geothermal environments, yielded archaeal sequences that exhibited novel phylogeny, including a novel lineage of Euryarchaeota. These results suggest that deep South African gold mines harbor novel archaeal communities distinct from those observed in other environments. Based on the phylogenetic analysis of archaeal strains and rDNA clones, including the newly discovered archaeal rDNA clones, the evolutionary relationship and the phylogenetic organization of the domain Archaea are reevaluated.

Observation of mining-induced faults in an intact rock mass at depth

Stoping of a highly-stressed remnant in a deep South African gold mine was accompanied by considerable seismic activity and some significant rockbursts. The larger seismic events were registered some 60 km away at a World-wide Standard Seismograph Network (WSSN) station. Several shear ruptures were encountered subsequently in a tunnel driven to follow the mining faces. A careful study based on detailed exploration of two of these ruptures proved them to be the source of two of the larger rockbursts. The ruptures were essentially planar features of considerable areal extent with shear displacements of up to 100 mm of a normal dip–slip sense. Certain striking features were revealed by a scanning electron microscopic study of some of the fresh cataclastic rock-flour which formed part of the comminuted filling of these ruptures. These provided strong evidence for extremely violent fracturing phenomena in the faulting process. A proper understanding of these phenomena could provide useful insight into the source mechanism of rockbursts and the control of the accompanying damage. Because of the many similarities in morphology of the faults and in the nature of the seismic signals it could be argued that the rockburst rupture is a microcosm of the pristine earthquake. Understanding of the mechanics of the rockburst shear rupture might therefore have some relevance in the study of the earthquake source.

Strike stabilizing pillars as a regional support strategy for ultra-deep gold mines

Strike stabilizing pillars are included in the mine layout of a number of deep South African gold mines as a means of providing regional support with the principal aim of controlling rockbursts. Large seismic events associated with stabilizing pillars can cause extensive damage to working areas. Mining-induced seismicity recorded at Western Deep Levels Limited has been analysed in an attempt to improve the design of stabilizing pillars, and thereby reduce their associated seismic hazard. This work revealed that the vast majority of stabilizing pillars, regardless of their dimensions and those of their adjacent stopes, will, at some time, give rise to seismic events of magnitude, M≥2. Contrary to expectations, this work strongly indicates that the rock mass in the near vicinity of the mined out areas does not behave in an elastic manner. Consequently, the currently employed design methodologies, based on elastic principles, should not provide the only criteria when designing strike stabilizing pillars.

Violent failure of a remnant in a deep South African gold mine

The violent failure of a peninsular remnant at a depth of 2300 m below surface occurred in a mine in the Carletonville Goldfield of South Africa, severely damaging a stope mining the Ventersdorp Contact Reef (VCR). At the rockburst site the VCR is 1–2 m thick with a lava hangingwall and quartzite/conglomerate footwall. The remnant had been formed as the result of a fault and `roll' encountered during mining. Observations at the rockburst site led us to conclude that the seismic event, with local magnitude of 2.1, resulted from failure of the remnant with attendant movement into the workings. The event could not be explained by a single shear slip. Two different damage mechanisms were identified. Firstly, the face and footwall on the east side of the remnant were violently ejected into the void between the original face and first line of timber packs following failure and dilation of the remnant and its foundation. Secondly, the hangingwall on the south side of the remnant fragmented and collapsed when subjected to intense seismic shaking. This response was due to the presence of a bedding-parallel fault and calcite-coated joints in the vicinity of a `roll'. The stope support system failed to contain the seismically fragmented rock.

Seismic imaging of rock mass responses to excavation

Three case studies are presented which illustrate rock mass characterization using a combination of tomographic images of spatial variation in seismic p-wave velocity with induced seismicity. In the controlled case of a tunnel excavation, known stress changes and fracturing were successfully mapped using velocity images and the location of induced seismicity. Within a stabilizing pillar in a deep South African gold mine, an active-source velocity image resolved the location of stress concentrations ahead of the mining face, and fracturing around the stope. Velocity images of pre-existing, seismically active faults were also computed by simultaneously inverting the induced seismicity data for velocity structure and hypocentral location. The resulting velocity image was interpreted in terms of the expected deformation and seismic potential along the faults. The potential role of velocity imaging and induced seismicity in validating geomechanical models is discussed, along with a possible strategy for incorporating the technique into an integrated geomechanical investigation of induced seismicity.