A statistical approach for the integrated analysis of mine induced seismicity and numerical stress estimates, a case study-Part II: evaluation of the relations

The results of this component of the study indicated that all but three of the developed relations met the statistical requirements for relations of this type and all were shown to be quite stable. In terms of prediction, it was demonstrated that this could best be achieved through the application of Discriminant Function Analysis. This technique was shown to be capable of suggesting which events were generated under relatively high-stress conditions, though the technique was less successful at correctly identifying events generated under lower stress conditions. Interpretation focussed on the relation between the stress conditions at the source and two particular variables from the seismic database and concluded that the observed variations in source strength were primarily controlled by spatial variations in confinement. The importance of a rigorous methodology was also discussed, as was potential ways in which the relations of the types formed in this study could impact mine design decisions.