In this study, I use the cross-hole method to examine the relationship between fractures in limestone reservoirs, where the presence of fractures has been determined by <I>a priori </I>information, and the shear-wave anisotropy of the rockmass. I process and interpret multicomponent seismic data from two producing areas: the East Fitts field Oklahoma, where a multi-azimuthal cross-hole survey is used to image the Hunton and Viola reservoir formations at depths of 3000-4000 feet; and the Iatan East Howards field, Texas, where two cross-hole azimuths are used to image the Clearfork reservoir formation at a similar depth. I apply a numerical method to measure the shear-wave splitting parameters, <I>qS1</I> polarization and time delay between <I>qS1</I> and <I>qS2</I> arrivals, and best-fit parameters for reservoir crack strike and crack density are determined by forward modelling. The observed <I>qS1</I> polarizations at three azimuths from the East Fitts site can be related to propagation through two thin layers, representing the Hunton and Viola reservoir zones, each containing distributions of micro-cracks with a best-fit strike of N35°E. The observed time delays between <I>qS1</I> and <I>qS2</I> arrivals can also be modelled by lower crack densities distributed throughout the layered model. Polarization measurements at one cross-hole azimuth from the Iatan site agree with <I>a priori </I>reservoir fracture direction of N60°E-N85°E. Results from the second azimuth fall within the expected zone of behaviour for cracks striking ±12.5° of N170°E. A measured <I>qS1</I> polarization direction of N170°E from near-offset VSP data at the Iatan site also disagrees with the known reservoir fracture direction. The Iatan measurements are incompatible with hexagonal anisotropic symmetry with a horizontal axis of symmetry. No improvements in the resolution of the anisotropic parameters of fractured rocks and reservoirs was achieved by using the cross-hole method although propagation through the near-surface is avoided. Acceptable model solutions have been found for the observations, however, the problems of non-uniqueness are inherent in the forward modelling approach.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:641286 |
| Date | January 1996 |
| Creators | Baptie, Brian J. |
| Publisher | University of Edinburgh |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/1842/10725 |
Page generated in 0.0129 seconds