One of the ideals of mine seismologists is to observe and describe rock failure phenomena underground and to relate that to the seismic source mechanisms as determined through seismological analysis. In the case of some large fault slip events and Ortlepp shears, reasonable successes have been achieved (e.g. van Aswegen, 2013). There are, however some pitfalls and, in some cases, contradictory evidence from underground observations compared to the seismological information.
The importance of rock mechanics insight and underground observations is illustrated using two examples cases from platinum mines in the western Bushveld Complex In case 1 the moment tensor reflects the stope closure associated with partial pillar failure. The seismic data shows no evidence of the extremely violent pillar failure process observed underground. In the far field, the relatively slow stope closure process is the only process that yields detectable waves. The high-frequency waves associated with the extreme ejection velocities at the pillar face are (a) subdued, since most energy goes into the ejection process and (b) virtually completely attenuated by the rock mass between the sources and the sensors. A reasonable understanding of what happened is only achieved through a combination of underground observation and seismological analysis.
In case 2 the dynamic deformation process is completely different, namely the relatively slow closure of a tunnel associated with one or more seismic events. The deformation occurred in a manner exactly opposite to what could be expected in terms of the local stress orientation. A conceptual model is developed that involves a process of transpressional simple shear (consistent with the stress model), which causes closure of the tunnel in a direction nearly normal to the shear zone. The mechanism is consistent with the moment tensor in terms of the orientation of the one nodal plane and its large implosive component. The seismological data does not reveal the damage process, but is useful to constrain models to explain the damage.
One may conclude that, to gain understanding of seismic source and damage mechanisms, both underground studies and seismological analyses are required.