Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Crack density tensor inversion for analysis of changes in rock frame architecture

Hall, Stephen LU ; Kendall, J. Michael ; Maddock, Jamie and Fisher, Quentin (2008) In Geophysical Journal International 173(2). p.577-592
Abstract
This paper presents a development of the use of multi-axial ultrasonic data for the quantification of extrinsic, crack or grain-contact induced elastic anisotropy in core samples with application to a UKCS oil field. An approach for inversion of multi-axial velocity measurements is presented, which extends the previous work by Sayers (2002) for the determination of second- and fourth-order crack density tensors from inversion of multi-axial ultrasonic velocity data. The extensions to the inversion approach provide improved consideration of data uncertainties, by using all available P- and S-wave data and also permit the inclusion of an orthorhombic background anisotropy in the inversion [e.g. due to intrinsic lattice preferred... (More)
This paper presents a development of the use of multi-axial ultrasonic data for the quantification of extrinsic, crack or grain-contact induced elastic anisotropy in core samples with application to a UKCS oil field. An approach for inversion of multi-axial velocity measurements is presented, which extends the previous work by Sayers (2002) for the determination of second- and fourth-order crack density tensors from inversion of multi-axial ultrasonic velocity data. The extensions to the inversion approach provide improved consideration of data uncertainties, by using all available P- and S-wave data and also permit the inclusion of an orthorhombic background anisotropy in the inversion [e.g. due to intrinsic lattice preferred orientation(LPO) effects]. The latter aspect leads to estimates of the extrinsic anisotropy, that is, the quantified crack density tensors, that are 'unpolluted' by the effects of the intrinsic anisotropy, thus permitting extrinsic and intrinsic anisotropies to be distinguished. For the samples considered, the extrinsic crack-induced anisotropy is strong relative to that of the intrinsic LPO effects, and the pre-dominant crack-set is commonly aligned parallel to the depositional fabric (which is generally horizontal). However, the LPO and extrinsic anisotropies are in general aligned, which indicates a linked origin. Furthermore, a strong correlation is observed between the degree of VTI anisotropy and the modal content of micas in the samples, which cannot be explained solely by the LPO effect. In fact, it is shown that increased horizontal (grain parallel) crack densities are associated with higher mica content. These horizontal cracks in the mica-rich samples often show moderate-to-strong variations in crack density with loading that might be detected in situ with non-hyperbolic moveout. Additionally, three samples show significant pressure sensitivity of the vertical crack sets indicating that loading-induced azimuthal anisotropy might also exist in some layers in situ and be detectable using azimuthal AVO type analyses. Analysis of the fourth-order crack density tensor allows insight into the relative sizes of the normal and tangential crack compliances, B-N and B-T. For one case it is found that B-N >= B-T (i. e. the crack-sets are more compliant in compression than in shear). For another sample B-N < B-T; this sample had been cleaned prior to the analysis, which appears to facilitate shear in the cracks. This conclusion may have significant implications for the use of analogue samples acquired at the surface where organic products, which lead to the reduced shear compliance, are not present. (Less)
Please use this url to cite or link to this publication:
author
; ; and
publishing date
type
Contribution to journal
publication status
published
subject
in
Geophysical Journal International
volume
173
issue
2
pages
577 - 592
publisher
Oxford University Press
external identifiers
  • scopus:42349114347
ISSN
0956-540X
DOI
10.1111/j.1365-246X.2008.03748.x
language
English
LU publication?
no
id
892de3f6-8e80-4b3c-845e-2ed4435b8173 (old id 2441133)
date added to LUP
2016-04-04 14:23:49
date last changed
2022-01-30 01:55:41
@article{892de3f6-8e80-4b3c-845e-2ed4435b8173,
  abstract     = {{This paper presents a development of the use of multi-axial ultrasonic data for the quantification of extrinsic, crack or grain-contact induced elastic anisotropy in core samples with application to a UKCS oil field. An approach for inversion of multi-axial velocity measurements is presented, which extends the previous work by Sayers (2002) for the determination of second- and fourth-order crack density tensors from inversion of multi-axial ultrasonic velocity data. The extensions to the inversion approach provide improved consideration of data uncertainties, by using all available P- and S-wave data and also permit the inclusion of an orthorhombic background anisotropy in the inversion [e.g. due to intrinsic lattice preferred orientation(LPO) effects]. The latter aspect leads to estimates of the extrinsic anisotropy, that is, the quantified crack density tensors, that are 'unpolluted' by the effects of the intrinsic anisotropy, thus permitting extrinsic and intrinsic anisotropies to be distinguished. For the samples considered, the extrinsic crack-induced anisotropy is strong relative to that of the intrinsic LPO effects, and the pre-dominant crack-set is commonly aligned parallel to the depositional fabric (which is generally horizontal). However, the LPO and extrinsic anisotropies are in general aligned, which indicates a linked origin. Furthermore, a strong correlation is observed between the degree of VTI anisotropy and the modal content of micas in the samples, which cannot be explained solely by the LPO effect. In fact, it is shown that increased horizontal (grain parallel) crack densities are associated with higher mica content. These horizontal cracks in the mica-rich samples often show moderate-to-strong variations in crack density with loading that might be detected in situ with non-hyperbolic moveout. Additionally, three samples show significant pressure sensitivity of the vertical crack sets indicating that loading-induced azimuthal anisotropy might also exist in some layers in situ and be detectable using azimuthal AVO type analyses. Analysis of the fourth-order crack density tensor allows insight into the relative sizes of the normal and tangential crack compliances, B-N and B-T. For one case it is found that B-N &gt;= B-T (i. e. the crack-sets are more compliant in compression than in shear). For another sample B-N &lt; B-T; this sample had been cleaned prior to the analysis, which appears to facilitate shear in the cracks. This conclusion may have significant implications for the use of analogue samples acquired at the surface where organic products, which lead to the reduced shear compliance, are not present.}},
  author       = {{Hall, Stephen and Kendall, J. Michael and Maddock, Jamie and Fisher, Quentin}},
  issn         = {{0956-540X}},
  language     = {{eng}},
  number       = {{2}},
  pages        = {{577--592}},
  publisher    = {{Oxford University Press}},
  series       = {{Geophysical Journal International}},
  title        = {{Crack density tensor inversion for analysis of changes in rock frame architecture}},
  url          = {{http://dx.doi.org/10.1111/j.1365-246X.2008.03748.x}},
  doi          = {{10.1111/j.1365-246X.2008.03748.x}},
  volume       = {{173}},
  year         = {{2008}},
}