A brief overview of the seismicity in a deep level goldmine is given. The direct influence of mining on the state of seismicity is demonstrated with some case studies. To gain a clearer picture of the mechanisms involved, it was necessary to categorize certain event-types. It will be shown that only a limited number of them (normally geology-related events) are a real threat to the mine-workers safety. The mechanisms behind the different event-types indicate which counter-measures arc appropriate. Backfill in low stoping width environment, preconditioning blasts and adjusted mining geometry are measures to minimize the rate of exposure of the mine-worker to such violent rock-failures.
The goldmine Western Deep Levels, Ltd. is situated in the Western Transvaal, about 70 km west of Johannesburg. The mine forms part of the West Wits Line, which consists of two major conglomerate formations, the Ventersdorp Contact reef ("VCR") and the Carbon Leader reef ("CLR"). The VCR is mined between 1600 and 2400 m below surface and the CLR between 2300 and 3500 m below surface. Both reefs dip approximately 21 degrees towards the South and are nearly parallel to each other and separated by 800 to 1000 m of quartzitic rock. They are intersected by numerous dykes and some faults of minor throw (< 50 m). The company has adapted longwall-mining as a general concept which allows a 10 - 12 m face-advance per month. The mine experiences an increase of seismicity with depth which is reflected in the support requirements for safe mining operations. These circumstances led to the introduction of stabilizing pillars in 1980 and backfill in 1987. The seismic network of the mine records more than 560 events (M >0) per month at the present. More than two-thirds of which locate on the WDL-mine.
Seismicity in deep level mining is determined by four factors namely depth, production, mining geometry and the presence of geological discontinuities such as dykes and faults. One factor may dominate the whole picture, however. The link between production blasting and mining induced seismicity is obvious in Figure 1. This concentration of seismic events during blasting might give the impression that only minor events are triggered by blasting and that large rockbursts occur on shift. This is, however, not true. Firstly, at least 40 % of all seismic events occur outside blasting time, although they are spread over a period of 20 hours. Secondly, their frequency-magnitude distribution is very similar (Fig. 2). Their b-values (slope of the curve logN= a - b * M, where N represents the number of events larger than a magnitude M) differ slightly- 0,80 during blasting and 0,74 outside blasting time. Only in the magnitude range M > 3 do we find a little dominance outside blasting time, indicating a longer lasting process of stress-redistribution which can result in a larger seismic event. Large seismic events (M >3) have been observed only on
geological features,
abutments,
and planned pillars and isolated pillars.
