We apply a simulated annealing velocity modeling optimization to 22 km of multi-fold seismic reflection data from the Dixie Valley geothermal field. Our methods take into account strong lateral variations in velocity that are assumed not to exist i n conventional, post-stack migration processing methods. Simulated annealing is an iterative, computer intensive optimization method for two-dimensional velocity estimation which makes no assumptions on the direction of the subsurface velocity gradient. Employing this, we can develop velocity models from raw seismic data that accurately represent heterogeneous subsurface structure within thermal anomalies using no a priori data. We correlate velocity structure derived from the Dixie Valley seismic data with known geologic structure and seismic velocities derived from borehole data. Our results indicate six percent or less difference in velocity between models derived from simulated annealing and those derived from examination of well and surface .geological data. Our model also reflects lateral variations in velocity westward toward the surface trace of the Stillwater Fault. We present examples of applying simulated annealing to seismic reflection data from Dixie Valley and from the laterally complex offshore Santa Maria Basin. California.

Press the Back button in your browser, or search again.

Copyright 1997, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.