Method for Acoustic Anisotropy Interpretation in Shales When the Stoneley-Wave Velocity is Missing
- Ming Gu (Halliburton) | John Quirein (Halliburton) | Eric Murphy (Chesapeake Energy Corporation) | Saul Rivera Barraza (Chesapeake Energy Corporation) | Liwei Ou (Colorado School of Mines)
- Document ID
- Society of Petrophysicists and Well-Log Analysts
- Publication Date
- April 2016
- Document Type
- Journal Paper
- 140 - 155
- 2016. Society of Petrophysicists & Well Log Analysts
- 6 in the last 30 days
- 353 since 2007
- Show more detail
Characterizing shale mechanical properties requires five independent stiffness coefficients: C33, C44, C66, C11, and C13. In vertical wells, C33 and C44 are determined from the vertically propagating P- and S-waves. Because the commonly implemented models, such as the ANNIE model and the modified ANNIE (M-ANNIE 1) model use the Stoneley-wave velocity to interpret C66 and two other assumptions to predict C11 and C13, they are not applicable in cased-hole conditions when the Stoneley-wave cannot be measured.
In this paper, three new models without the Stoneley-wave as an input are introduced: the velocity regression model (V-reg), the further modified ANNIE (M-ANNIE 2), and the integration model combining both. The bases of the V-reg model are the observed near-linear relationships among measured 0°, 45°, and 90° P- and S-wave velocities. M-ANNIE 2, which is based on M-ANNIE 1, uses the linear relationship between the Thomsen compressional-and- shear-wave anisotropy parameters to replace the Stoneley-wave constraint to predict C66.
By applying the new models to the ultrasonic core data of multiple organic shales, their predictive power for the stiffness coefficients, elastic moduli, and closure stress are evaluated. Generally, the new models provide predictions that are as good as the M-ANNIE 1 and better predictions than the ANNIE and isotropic models. They provide good estimates for C66, with a small bias of approximately 1%. They also reduce the underestimation bias of the ANNIE and isotropic models. Finally, the log examples show that the new models yield predictions consistent with the Stoneley-wave models.
An accurate prediction of Young’s modulus and Poisson’s ratio is crucial for predicting fracture deformation, minimum horizontal stress (σhmin), and rock brittleness. Hence, it is important for selecting where to drill and perforate or designing a fracturing pumping strategy (Khan et al., 2012; Gokaraju et al., 2015). Elastic moduli computed from measured or predicted sonic-wave velocities are referred to as the small-strain elastic properties or “dynamic moduli,” in contrast to those measured in a rock mechanics laboratory with triaxial tests, which are referred to as the large-strain deformational properties or “static moduli.” Both the short-term brittle deformation during drilling/ fracturing and the long-term ductile deformation during production are related to the static moduli. So, all dynamic moduli must be transformed to static moduli based on the empirical relationships calibrated to laboratory data for their future use in drilling/completion designs.
|File Size||354 KB||Number of Pages||16|