Master of Science
Crosshole seismic tomography has been used by Vale to provide geophysical images of mineralized massive sulphides in the Eastern Deeps deposit at Voisey's Bay. High resolution seismic images are presented by applying acoustic waveform tomography to these data. In waveform tomography, an initial model is required which predicts the first arrival traveltimes to within a half-cycle of the lowest frequency in the data. Because seismic velocity anisotropy can be significant, the initial model must quantify the anisotropy, as well as the velocity, in order to meet the half-cycle criterion.
In our case study, difficulties were encountered in generating an accurate anisotropy model through traveltime tomography, and the starting model for waveform tomography failed the half-cycle criterion. We formulate a new, semi-global approach for finding the best-fit 1-D elliptical anisotropy model using simulated annealing, and successfully apply this technique to the Vosiey's Bay dataset, as well as synthetically generated datasets.
Afanasiev, Michael Victor, "Estimating Subsurface Anisotropy: Combining Waveform Tomography and Simulated Annealing" (2012). Electronic Thesis and Dissertation Repository. Paper 1000.