The objective of this investigation was to develop procedures and analysis guidelines, based on appropriate laboratory experiments and numerical simulations, for the analysis of high frequency (HF) microseismic data collected at underground hardrock mines in the general frequency range of 40 to 400 kHz. These procedures and guidelines were then used in the analysis of HF microseismic data recorded at three underground mines in Ontario.

The results of laboratory experiments indicated that microseismicity associated with the microcracking of intact rock subjected to increasing deviatoric stresses was characterized by increasing source amplitude energies and dominant frequencies while slip along fractures was characterized by comparable source energies but lower source dominant frequencies. The differences in dominant frequencies observed during these tests was inferred to be a function of the dimensions of the source, i.e. the size of the induced microcracks or fracture displacements. Further laboratory experiments indicated that microseismic wave velocity increased with increased uniaxial compressive stress and that the attenuation of microseismic signals across fractures was increased by the presence of the fracture but the level of attenuation approached that of the intact rock as the fracture normal stresses and stiffnesses increased. The results of numerical simulations of event generation and propagation indicated that changes in source energy, source dominant frequency and the rockmass attenuation properties (described using the Q factor) all interacted to produce the characteristics of the waveforms recorded at the receiving transducer. Guidelines were developed, based on these results, to help identify the major changes in source parameters and rockmass Q factors based on the observed changes in received event parameters.

Analysis of field data collected at the Kidd Creek, Creighton and Copper Cliff North Mines indicated that increased event rate was the most obvious indication of transient stress changes in a local rockmass while analysis of the associated changes in received event parameters could help infer the predominant mode of deformation. The majority of received events were of two types:​ (1) brief immediate activity resulting from the passage of the direct seismic waves from a distant blast or seismic event and (2) longer sustained activity associated with the arrival of the transient deviatoric stress (changes resulting from the distant event) in the local area which may have occurred immediately or at some time after the distant event depending on the distance to the local area, the mining geometry and the intervening geology. The relative magnitude of the distant event and/or the level of stresses and incipient instability in the local area could be estimated by the duration and peak event rate of either type of induced HF microseismicity. Long term event parameter analysis could help identify stable stress changes, rockmass yielding and the movement of the source volume with respect to the receiving transducer.