Summary

A microseismic dataset was acquired in a mine during January 2011 by a network of 7 boreholes each having 4 three-components geophones. After initial preprocessing 488 events are identified with a minimum of 12 P-arrival time picks per event. These events are located with a grid search algorithm based on observed arrival times. Waveform crosscorrelation techniques are employed to identify events with similar waveforms and compute highly accurate differential arrival times. The latter are used for a combined double-difference based tomographic inversion to determine both the 3D velocity model and final event locations. The seismic ray distribution is used to select an optimal inversion grid so as to minimize the poorly sampled nodes but recover the expected velocity heterogeneity. The final 3D velocity model shows crosscuts between 0.45 km to 0.5 km as low velocity zones and more competent basement rock as high velocity zones. The inverted event locations are strongly clustered around the high velocity zone and are mostly absent near the low velocity zones between depths of 0.45 km and 0.5 km. This suggests that the high velocity zone represents the highly stressed more competent basement rock which can release seismic energy through brittle failure causing microseismic events while the low velocity zones represent the highly fractured and failed rocks due to the presence of the crosscuts and hence are mostly aseismic.

This content is only available via PDF.
You can access this article if you purchase or spend a download.