"There are many factors that influence where and how deep a mining company will dig for gold, but in general “as you go deeper it gets more expensive and time-consuming”, says Steve McKinnon, a mining engineer at Queen's University in Kingston, Ontario, Canada, who specializes in designing deep mines. It also gets more dangerous.The rule of thumb is that the sun influences the first 500 feet or so of temperature and after that, the temperature increases 1 oF for every 100 ft in depth.
Mining at depths such as those of TauTona (2.4 miles/3.9 km) presents many unique challenges in protecting the miners, says McKinnon. First of all, it's hot. The temperature at TauTona's deepest levels is a stifling 58 °C. Air conditioning brings the temperature down to a toasty but more tolerable 28 °C.
Then there is the risk that digging can fracture the rock around the pit, triggering a seismic event. “Sometimes that fracture process can be very violent, because the rock behaves in a brittle manner,” says McKinnon. “There have been events larger than magnitude 5” — equivalent to a moderate earthquake.In other words, the ground support is far more sophisticated than Steven Speilberg would have you believe - wood beams. Granted, wood beams used to be used, but there are much better materials nowadays.
To minimize the risk, mines use 'yielding supports' that are able to deform as the tunnel walls move yet still retain their ability to support the structure. There are also networks of seismic sensors that constantly monitor the rock and develop a seismic 'fingerprint' for the region — any departure from an established baseline requires that workers be pulled out until seismic readings return to normal.
Engineers specializing in rock mechanics, such as McKinnon, also try to design mines in such a way as to minimize and control the forces exerted on the rock. Taking into account data on local stresses on the rock, and other factors such as the presence of nearby faults, they calculate the most favourable layout and method of ore extraction to minimize fracturing. This information helps engineers determine which bits of rock can be safely removed and which need to be left behind to support the rest — in much the same way as an architect decides where to put the pillars in a cathedral to hold up the enormous roof. This design modelling is “really at the limit of our technical ability now”, says McKinnon."In other words, there is a large amount of engineering involved. Without it, you are relying on luck. You can find luck in a casino, but there is precious little of it thousands of feet underground.