Mathematical modeling and ultrasonic measurement of shale anisotropy and a comparison of upscaling methods from sonic to seismic.

摘要

The anatomy of shale is complicated because lithological heterogeneities are present at a very wide range of scales. Both the mineral and pore space contributions to the net seismic anisotropy of a shale are still subjects for research because of the extreme range of scales involved. There is no microscope of any form that allows all the details of shale to be studied first hand. This research takes what can be determined by microscopes and other available data, such as elastic properties, and develops a quantitative approach to understanding the seismic anisotropy of shale caused by the alignment of clay minerals and pores. In other words, when do the clay platelets dominate the anisotropy and when does the porosity play a role? This question is of interest for exploration purposes because the presence of cracks enhances the permeability. The answer depends upon the saturation of the pores. When the pores are water-filled, the mineral alignment dominates the anisotropy of the shale. When the pores are gas-filled, the pore alignment dominates the anisotropy produced by the mineral alignment to give a new signature to the shale anisotropy. This new signature includes a dramatic change in the S-wave anisotropy where a singularity point (a point where the two S-waves have the same velocity) is created giving a tell-tale signature of the gas.; The theoretical understanding of the effective media modeling is used to model Barnett Shale, which is one of the largest natural gas plays in the World. After the estimation of mineralogical assemblage using FTIR- and XRD techniques, forward modeling is used to calculate the elastic properties of the Barnett Shale facies. In order to extract the information about the microstructure of shale, the mineralogy-based elastic constants are matched against laboratory-measured elastic constants using inverse modeling by applying a minimization function.; Upscaling of heterogeneous elastic media requires accounting for elastic scattering and interaction among various elements of the heterogeneous media. Upscaling method based on pair correlation function approximation provides more accurate upscaling estimate of velocities at surface seismic exploration scales than Backus and simple averaging. The differences in the results are attributed to the energy loss due to elastic scattering.