Project Name

Observational Studies in South African Mines to Mitigate Seismic Risks


Republic of South Africa

Date of R/D Signed

February 26, 2010

Project Site

Deep gold mines in Far West Rand area (about 60 km west of Johannesburg) and the Klerksdorp area

Term of Cooperation

August 6, 2010 to August 5, 2015 (5 years)

Implementing Organization

Council of Science and Industrial Research, Council for Geoscience, Wits University

Implementing organization (Japanese side)

Ritsumeikan University, Tokyo University, Tohoku University, Kagoshima University, Kyoto University, Advanced Industrial Science and Technology, Tono Research Institute of Earthquake Science

Project Background

The Republic of South Africa (hereinafter RSA) boasts an excellent resource base for Gold, Platinum and Coal, and its mining industry employs over 500,000 people. Around 60% of these employees work in the mines, and are exposed to many risks and dangers. Mine accidents have long been one of RSA’s most serious work place issues, but with government and corporate commitments to improve the safety and health conditions the mine workers, the indicators have improved. In the 1980s the number of fatalities ranged from 677 to 855 per annum (1.00 to 1.20 deaths per 1000 workers per annum). By the early 2000s, the number of fatalities had declined to 246 to 290 per year, partly owing to a decrease in the work force, although there was also a significant decline in the risk faced by workers (0.56 to 0.75 deaths per 1000 workers per annum). Despite the progress, the rate of decline has stopped short of satisfactory and further improvements need to be made to the safety and health conditions of mine workers.

About 40% of mine fatalities are caused by rock falls and rock bursts. The latter are associated with seismic events triggered by the mining activities. These seismic events not only affect the underground workings, but also cause earthquake shaking on the surface that is sometimes strong enough to cause buildings and other structures to collapse.

The occurrence of mining-related seismic events generally increases as the depth of mining increases. RSA has many goldmines that are over 3000m deep, which makes RSA prone to mining-related seismic activity. The excavation of tunnels and ore induces very high stresses in the surrounding rock mass. The stresses may exceed the strength of the rock, causing it to fracture. The fractured rock mass is supported and reinforced, but from time to time instabilities occur that result in rock falls and rock bursts. In order to mitigate the damage caused by seismic events, it is necessary to understand the rock fracture and failure mechanisms. With better understanding of mining-related seismic phenomena, mine planning and production activities can be improved leading to safer work places.

Japanese scientists have long been active within some of the deepest mines of RSA, carefully studying and monitoring seismic activities underground. It has been shown that seismic events can be recorded at a close range within the mines, and that under certain conditions, the location and size of seismic events can be foreseen. The deep gold mines of RSA are uniquely suitable for such research, and with the scientific understandings obtained, scientists will be able to better understand seismic mechanisms and advance their efforts to forecast seismic event.

From this background, the government of RSA requested the launch of a project aimed at advancing the forecast of seismic events and improving technologies to mitigate the damage. In response to the request, a Japanese research group led by Ritsmeikan University submitted the project proposal to the Japan Science and Technology Agency (JST) as a “Science and Technology Research Partnership for Sustainable Development; SATREPS”, which was later adopted as an official project of JST.

Parallel with the endorsement of the proposal, the request from the government of RSA was also accepted by the Japanese government, to which JICA responded by conducting a pre-project survey in October 2009. The two governments agreed upon the basic project outline later proposed by JICA, and in February 2010 a Memorandum of Understanding (MOU) was signed by JICA South Africa Office and the government of RSA, marking the official launch of the proposed project.


  1. Rock properties at seismic sources are elucidated.
  2. Understanding of the preparation and forerunners of earthquakes in gold mines is enhanced.
  3. Reliability of seismic hazard assessment in gold mines is improved.
  4. Reliability of strong ground motion predictions in gold mines is improved.
  5. Estimation of the locations of seismic events, and damage assessment of seismic disasters become more accurate.

Project Activities

1-1 Observe and collect rock samples at seismic sources and its surroundings
1-2 Investigate rock properties in laboratory
2-1 Install monitoring systems for micro-fracturing and rock deformation, and sensitive long period seismometers in the vicinity of impending earthquake sources
2-2 Monitor the activities of micro fracturing
2-3 Monitor the accumulation and release of stress in and around the impending earthquake source
2-4 Analyse the monitored data to clarify the forerunning phenomena and its characteristics
3-1 Evaluate spatiotemporal changes in stress and rock mass stability based on the data produced by mine’s microseismic network
3-2 Upgrade the scheme of seismic hazard assessment by calibrating existing schemes with the data obtained through the activities 2-2, 2-3 and 3-1.
4-1 Install accelerometers and high capacity strainmeter in the vicinity of impending earthquake sources to monitor the dynamic slip and stress change
4-2 Analyse the data obtained through the activity 4-1 to clarify the process of rock mass failure and strong motion generation
4-3 Install strong motion meters and geophones in underground working places to monitor strong motion at disaster sites
4-4 Compare strong ground motions at the source with those at damage sites to clarify the characteristics of site amplification of strong motion
4-5 Compare the monitored dynamic stress change and fault slip with existing lab-experimental results to clarify the scaling relationship in dynamic rupture process
5-1 Install seismic stations in the Far West Rand mining district (Carletonville area) on the surface to monitor surface ground motion caused by mine tremors
5-2 Upgrade the Data Centre in the Silverton offices of the Council for Geoscience
5-3 Develop and validate a parametric model that will be capable of predicting strong ground motion