Full-wave Seismic Analysis: Source Comparisons, Land Streamer Tests, and Converted-wave Processing

Gabriela M. Suarez Barbero

Two full-wave multicomponent seismic analyses are undertaken in this thesis. The first case assesses two land streamer surveys conducted by the CREWES project in the foothills of the Canadian Rocky Mountains. The first land streamer survey initially generated promising data with prominent reflections at around 50-60 ms and 300 ms. A second survey consisted of a comparison between land streamer and planted geophone data. Comparison of raw shots, amplitude spectra and stacking sections showed that the vertical channel data quality was similar for both datasets, while the planted geophone provided the highest quality radial channel data. In a 3D seismic survey acquired in this same area, seismic processing techniques for acquisition footprint removal constituted a way of improving the seismic images quality that are necessary for imaging near-surface structures and stratigraphy. The 2D and 3D seismic data correlated quite well and provided very promising pictures of the near surface.

The second analysis considered three 3-C 2D seismic lines acquired with different seismic sources (dynamite, heavy vibrator, and light vibrator) in the Spring Coulee area in southern Alberta. The objective was to compare sources for full-wave imaging. We found that the explosive source provided deeper penetration, higher frequency content and better resolution, but was highly variable from source to source location. The heavy vibroseis data appears to be more consistent, with similar characteristics to the dynamite especially in the shallow parts of the sections. The lighter vibrator data proved to be an efficient source for acquiring good quality P-wave data to 1000 ms, but not for converted wave data.