Tomographic Imaging Associated with a Mw 2.6 Fault-Slip Event in a Deep Nickel Mine

Abstract

One of the biggest challenges facing geoscientists is the ability to accurately predict failure within a rock mass. Conventionally, numerical modeling is performed to predict the response of the rock mass due to excavation. However, numerical modeling relies heavily on the estimated physical characteristics of the rock mass. Unless dense, costly sampling of the rock mass has been performed, the results of the modeling are not robust. Seismic tomography offers a unique advantage of monitoring the rock mass response over conventional numerical modeling because it is able to measure the true alteration in response to excavation (Westman, 2003). This paper utilizes a tomographic inversion scheme using the Fast Marching Method for raypath tracing and the Simultaneous Iterative Reconstruction Technique to solve the p-wave velocity model of an underground mine and surrounding rock mass. The inversion scheme presented is tested using a data set from Creighton Mine in Sudbury, Ontario, Canada and includes 9,270 distinct events over 62 days. A total of 53 geophones recorded 191,856 p-waves that are able to be used for inversion. Temporal monitoring of the seismic p-wave velocity in the vicinity of a known Mw 2.6 fault-slip event that occurred on March 14th is performed by creating tomograms of the axial plane at the depth of the event and of an oblique plane where a dense distribution of events occurred including the March 14th event. Tomograms are produced on a weekly basis leading up to the event and also on a daily basis three days before the event. The weekly tomograms reveal a decrease in p-wave velocity in the vicinity of the Mw 2.6 event as time approaches the event and then a significant increase 1,600 ft/sec larger than the background velocity the week of the event. The daily tomograms reveal a 1,200 ft/sec velocity increase in the same area from March 13th to March 14th, however, no trends in the daily or weekly tomograms prior to the date of the March 14th event suggest the known event is imminent.