Towards the prediction of seismic anisotropy in sedimentary rocks.
Siska Valcke
School of Earth Sciences, University of Leeds, Leeds, LS2 9JT, UK
siska@earth.leeds.ac.uk
Modern seismic processing of hydrocarbon reservoirs requires detailed information on seismic anisotropy in sedimentary rocks. Seismic anisotropy is the change in seismic wave velocity and polarisation depending on the propagation direction through anisotropic media. Successful interpretation of seismic anisotropy depends on the knowledge of macro- and microstructural features that control the elasticity and therefore the seismic wave propagation through rocks. This contribution concentrates on the microstructures that may cause anisotropy in sedimentary rocks. It is generally accepted what these include:
Unfortunately, the relative contribution of each of these factors is less well established and most previous studies have tended to ignore sedimentary rocks. The present understanding of how microstructural variables influence seismic anisotropy in sedimentary rocks is mainly based on ultrasonic velocity measurements and on theoretical models in which microstructural variables are incorporated. Both approaches have restrictions, such as experimental conditions in the former and determining boundary conditions in the latter. An alternative approach, presented here, is to predict seismic properties by averaging single crystal elastic constants of minerals according to their LPO and modal volume fraction in the rock aggregate.
Both SEM Electron Backscattered Diffraction (EBSD) and X-ray Texture Goniometry (XTG) are tested as quantitative LPO measuring techniques for sedimentary rocks. Although EBSD is a promising technique for future LPO measurements in polymineralic sedimentary rocks, there are currently problems in measuring low symmetry phases such as feldspars and micas. However, the LPO of micas can successfully be identified with XTG. The measured LPO’s are near-random for quartz to non-random for dolomite, micas and clays. The seismic properties calculated from these data show maximum anisotropy values for P-waves, ranging from 1.5 % in sandstones to 6 % in shales. These results can then be used to model the effect of thin multilayers (e.g. in siltstones) on seismic wave propagation. Such models are based on simplified lower and upper boundary calculations for long and short wavelength propagation through a stack of anisotropic layers. Finally, ultrasonic velocity measurements, which are sensitive to the total seismic response of all contributing microstructural features, are being performed on the same rocks from which the predictions have been made. Together with the predictions, these measurements allow an estimate of the relative contribution of LPO and layering to the seismic anisotropy.
The results of this study will be a helpful tool in further developing the predictions of seismic anisotropy in sedimentary rocks, leading to a better understanding and processing of seismic reservoir data.