See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/249861027

QVOA analysis as an instrument for fracture

characterization

Article in SEG Technical Program Expanded Abstracts · January 2005

DOI: 10.1190/1.2142220

CITATION READS

1 8

1 author:

Tatiana Chichinina
Instituto Mexicano del Petroleo
46 PUBLICATIONS 155 CITATIONS

SEE PROFILE

All content following this page was uploaded by Tatiana Chichinina on 25 September 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.

QVOA analysis as an instrument for fracture characterization
Tatiana Chichinina*, Vladimir Sabinin and Gerardo Ronquillo-Jarillo, Instituto Mexicano del Petroleo
Summary
We present a method for estimation of preferred P-wave
attenuation direction in fractured media based on
azimuthally varying attenuation. The method can be
applied to Q-estimates extracted from multi-azimuthal 3D
seismic data involving surface reflection data and VSP. The
method proposed is called QVOA analysis (that means
variation of attenuation, or seismic quality factor Q,
versus offset and azimuth); it is based on an analytical
expression for P-wave attenuation as a function of wave-
propagation direction in HTI medium. The expression
appears to be similar to Rüger’s approximation for PP-
wave reflection coefficient as a function of incidence angle
and source-receiver-line azimuth. In this connection, the
proposed technique of QVOA analysis is similar to the
well-developed technique for azimuthal AVO analysis
(AVOA). An application of the method is illustrated on
synthetic data generated from Hudson’s effective-medium
model with a fluid-flow attenuation mechanism. Also we
predict relationships between attenuation anisotropy and
fractured medium properties.
Introduction
In seismic frequency range, anisotropy of attenuation was
observed in walk-around VSPs (Horne and MacBeth, 1997)
and from surface seismic reflection data (Lynn & Beckham,
1998; Clark et al., 2001). The theoretical studies of Hudson
et al. (1996), Thomsen (1995) and Chapman (2003)
provided effective-medium models, which explained
azimuthal attenuation variations in fractured media by
fluid-flow mechanisms. From all these theories it follows
that “the more attenuated azimuth” is perpendicular to the
aligned flow conduits.
We present the method based on azimuthally varying
attenuation, which we call QVOA by analogy with AVOA.
The method goal is to extract the azimuths of attenuation
maximum and minimum from multi-azimuth 3D P-wave
reflection data (or VSP). The minimum corresponds to the
fracture strike, which is assumed to be the direction of
maximum horizontal permeability (Lynn, 2004). To
illustrate and test the method, we use Hudson’s theory,
which incorporates the mechanism of attenuation due to
fluid flow between cracks and porous rock matrix (that is
the “equant-porosity” model). We consider that Hudson’s
attenuation mechanisms provide cos 2I -trend in the
azimuthally varying attenuation for seismic frequency
range, with the attenuation maximum in the crack-normal
direction, and consequently the minimum in the in-crack
direction.
SEG/Houston 2005 Annual Meeting
ANI 2.2
The method QVOA requires estimation of Q from seismic
data. For example, the well-known spectral ratio method
can be applied to azimuth-sectored CMP gathers in order to
estimate Q-values for each individual offset, as well as the
frequency shift method can be used (Quan & Harris, 1997).
An application of the spectral ratio method to a true-
relative-amplitude CMP gather (termed “QVO method”)
was developed by Dasgupta & Clark (1998), Clark et al.
(2001) and Hackert & Parra (2004).
Background
According to Chichinina et al. (2004), the dependence of P-
wave attenuation on wave-propagation direction can be
approximately expressed as
Q 1 (T , I ) | 'IN [1  2 g (1  cos 2 I sin 2 T )]2 , (1)
where T is the incidence angle, I is the source-receiver-
line azimuth with respect to symmetry axis; g (VS / VP ) 2 ,
where VS and VP are host rock’s S- and P-wave velocities;
'IN is the imaginary part of the complex normal weakness
~
' N ' N  i'IN (e.g., MacBeth, 1999). The normal weakness
' N as well as the tangential one, ' T , were introduced
originally for the non-attenuative fractured medium (TI) as
real quantities (Hsu and Schoenberg, 1993), see Appendix.
Taking square root from left-hand and right-hand sides
of equation (1), on can get linear dependence on sin 2 T ,
Q 1/ 2 (T ,I ) $ 0  %(I ) sin 2 T , (2)
where $ 0 is an intercept, $0 (' N ) (1  2 g ) , and %(I ) is
I 1/ 2
QVO gradient, %(I ) ('IN )1 / 2 2 g cos2 I , which represents cos 2I -
trend. We have introduced equation (2) for the attenuation
by analogy with Rüger’s (1997) weak-anisotropy
approximation for PP-wave reflection coefficient, which is
widely used in azimuthal AVO analysis. One can see that
the attenuation maximum is in the fracture normal direction
( I 0 , that is slow P-wave-velocity direction), and the
attenuation minimum is in the fracture strike direction
( I S / 2 , i.e., fast P-wave-velocity direction).
The QVO gradient (divided by $ 0 ) can be expressed as
B (I ) { %(I ) / $ 0 0.5 ( B A cos 2I  B A ) , (3)
A
where B is the gradient maximum (or symmetry-axis-
plane gradient divided by $ 0 ),
B A { %(0) / $ 0 2 g /(1  2 g ) . (4)
From here, one can see that the azimuth-variation
magnitude, B A , depends only on the host-rock parameter
VS / V P ( g (VS / VP ) 2 ).
We have derived equation (1) for Q 1 (T , I ) from the
equation for P-wave phase velocity (Schoenberg and
Douma, 1988, p.581)
127

QVOA analysis as an instrument for fracture characterization
V (i) 2 VP (1  ' N [1  2 g sin 2 i]2  ' T g sin 2 2i ) ,
2 (5)
(where sin 2 i 1  cos2 I sin 2 T ), in which we have substituted
~
instead of ' N the complex normal weakness ' N . We
assume that the tangential weakness ' T is real, because the
imaginary part of the complex weakness '~ T , which is
responsible for the intra-crack viscous dissipation, turns to
zero (MacBeth, 1999; Pointer et al., 2000). The phase
velocity becomes complex, and the attenuation 1/Q can be
expressed from it as: 1 / Q Im(V 2 ) / Re(V 2 ) , where Re and
Im denote real and imaginary parts, respectively (e.g.,
Carcione et al., 1998). Then the exact version of equation
(2) is
Q 1 / 2 [V P / V (T , I )] ( $ 0  %(I ) sin 2 T ) . (6)
The approximation (2) can be used instead of the exact
equation (6), assuming
( $ 0  %(I ) sin 2 T ) !! [VP / V (T , I )] | 1 ,
that is valid for weak anisotropy media.
The equations (2)-(4) are the QVOA-method background.
An illustration and validity of the method will be shown on
synthetic data.
Application of the method to synthetic data
For estimation of P-wave Q-anisotropy from wide-azimuth
3D reflection data, one should sector by azimuth the data,
as it is commonly used for azimuthal AVO analysis. Figure
1 shows a scheme for 6 source-receiver lines selected for a
certain CMP.
Fig. 1: CMP scheme with 6 source-receiver lines, each with
a different azimuth angle I k . Symmetry-axis azimuth is I 0 .
The model-input azimuth of the symmetry axis was set by
I 0 75 $ ; the source–receiver azimuths are I1 0 $ ,
I2 36 $ , …, I6 180$ ( 'I 36 $ ).
For the modeling, we have generated synthetic Q-data
following the “exact” equation (6), in which the value of
'IN was calculated for the “equant-porosity” model
(Hudson et al., 1996); the details and model-input
parameters are given in Appendix. For each individual
SEG/Houston 2005 Annual Meeting
ANI 2.2
source-receiver line, I Ik , an individual value of QVO
gradient, % k , should be determined.
This is the first step of the method shown in Figure 2. The
Q-data was recalculated to y Q 1 / 2 and plotted versus
x sin 2 T , squared sine of incidence angle T . According to
equation (2), the Q-data was fitted by a linear dependence
y $ 0  % x (using the least squares method). And so, the
line intercept and slope gave the value of $ 0 and the QVO-
gradient value % .
(7)
Fig.2: The scheme of QVO-gradient extraction. The line slope
gives the gradient value % ; the line intercept is $ 0 . Solid line
corresponds to the Q-data, and dashed – to the linear fit. The
plot is given for the source-receiver azimuth I3 72 $ for gas-
filled cracks with model-input parameters from Appendix.
This procedure was applied to each of six I k -azimuth Q-
data, and that yielded six pairs of QVO-gradient value %
and intercept value $ 0 . Then each % was divided by its
$ 0 : % k / $ 0 k , k 1, 2, ..., 6 ; these six values were plotted
versus azimuth, see Figure 3.
Figure 3 shows the next step of the method, which consists
in determination of attenuation minimum and maximum
directions from the QVO-gradient values after fitting
cos 2I -type function, as it follows from equation (3). For
example, in the case of brine-filled cracks, the fit is
y 0.548 cos 2(I  75$ )  0.548 , in which the azimuth
I0 75$ is the QVO-gradient maximum and points the
attenuation maximum. Then, the orthogonal direction,
I 0  90 $ 165 $ , indicates the attenuation-minimum
direction, or fracture-strike orientation, which is considered
to be the direction of maximum horizontal permeability (or
preferred fluid-flow direction).
In Figure 3, the QVO-gradient-maximum value is greater
for gas-filled cracks than for liquid-filled cracks (i.e., with
brine or oil). That is due to the term [VP / V (T , 0)] in
equation (6), which always gives larger B A -values for gas-
filled cracks than for liquid-filled cracks (because, for gas-
128

QVOA analysis as an instrument for fracture characterization
filled cracks, ' N -value is always greater, as it is shown in
Figure 4).
Fig. 3: Six QVO-gradient values for gas-filled cracks (squares)
and for oil- or brine-filled cracks (circles); their cos 2I -type fit
(solid line). Dashed line shows the QVO gradient B (I ) calculated
from the approximations (3)-(4); B A is its maximum value.
As it is shown in Figure 4 and follows from the expression
for ' N given in Appendix, the change in crack parameters
(fluid type, crack-aspect-ratio and crack density) results in
the change of ' N -value. The normal weakness ' N affects
P-wave velocity anisotropy much, but, however, its impact
on P-wave attenuation anisotropy is found to be
insignificant.
Fig. 4 : Calculated values of ' N for the cases of oil, brine (dashed
line) and gas (solid line) infill. Curve “dry” is for empty cracks, for
which fluid bulk modulus N f 0 . ' N is divided by crack density
e. The expression for ' N and crack parameters used are given in
Appendix.
Q-anisotropy and fractured medium properties
From eq. (1)-(2), one can infer that $ 0 | 1 / Q|| and the
QVO-gradient maximum, B A { %(0) / $ 0 , is
SEG/Houston 2005 Annual Meeting
ANI 2.2
B A | ( 1 / Q A  1 / Q || )
, 1 / Q ||
(8)
where 1 / Q A is the symmetry-plane attenuation, and 1 / Q | | –
the isotropy-plane one. Therefore, the magnitude of
attenuation-anisotropy may be expressed by B A .
The value of B A depends on the host-rock parameter
VS / V P , as it follows from equation (4). Figure 5 shows the
great growth of B A -value with the change of VS / VP -value
from 0.4 to 0.6. There are calculations for gas-filled cracks
and for brine- and/or oil- filled cracks (the latter two
coincide).
Fig. 5: The same QVO gradient as in Figure 3, calculated for
VS / V P = 0.4, 0.5 and 0. 6.
The QVO-gradient maximum B A is more sensitive to the
change of VS / VP -value (from 0.4 to 0.6), than to the change
of crack-infill type (from brine/ oil to gas). This is the main
feature of Q-anisotropy (on the contrary to velocity
anisotropy). The attenuation-anisotropy magnitude strongly
depends on the host-rock VS / VP -parameter, while the
dependence on fracture parameters is weak. Such
dependence on VS / VP makes Q-anisotropy much stronger
than crack-related velocity anisotropy:
1
Q || / Q A (V || / V A ) 2 !! V || / V A .
[1  2(VS / VP ) 2 ]2
This ensures the validity of the approximation for
attenuation, equation (6), by the fulfillment the requirement
(7). The approximation is ideal for liquid-filled cracks due
to its small ' N -value, and, therefore, the right part of the
condition (7), [VP / V (T ,I )] o 1 , is perfectly fulfilled.
Conclusions
We have developed the analytical expressions for
estimating anisotropic attenuation in fractured HTI media
and defined relations between attenuation anisotropy and
fracture properties. Attenuation is represented by an
equation analogous to Ruger’s approximation for the P-
wave reflection coefficient for TI media. We propose using
this attenuation approximation to estimate properties that
129

QVOA analysis as an instrument for fracture characterization
indicate fracture direction. The algorithm is intended to
estimate attenuation attributes from measurements of
attenuation in seismic reflection data and VSP. The method
was illustrated on synthetic Q-data, and the accuracy of the
attenuation-anisotropy approximation was approved.
However, attenuation is difficult to estimate from seismic
data, and without an application to real data, the reliability
of the method cannot be evaluated. The next step of the
method development should demonstrate the QVOA
method on seismic data and analyze the errors in the
estimated parameters. The method QVOA has a great
potential because it uses relative characteristics of
attenuation and not its absolute values which are known to
be deficient in accuracy.
References
Carcione, J.M., Helle H.B., and Zhao T., 1998, Effects of
attenuation and anisotropy on reflection amplitude versus
offset: Geophysics, 63, 1652-1658.
Chapman, M., 2003, Frequency – dependent anisotropy due to
meso-scale fractures in the presence of equant porosity:
Geophysical prospecting, 51, 369-379.
Chichinina, T., Sabinin, V., and Ronquillo-Jarillo, G., 2004, P-
wave attenuation anisotropy in fracture characterization:
numerical modeling for reflection data: 74th Ann. Internat. Mtg:
Soc. of Expl. Geophys., 143–146.
Clark, R. A., Carter, A. J., Nevill, P. C. and Benson, P. M., 2001,
Attenuation measurements from surface seismic data:
azimuthal variation and time–lapse case studies: 63rd Meeting,
European Association Exploration Geophysicists, Extended
Abstracts, paper L28.
Dasgupta, R., Clark, R. A., 1998, Estimation of Q from surface
seismic reflection data: Geophysics, 63, 2120–2128.
Hackert, C.L. and Parra, J.O., 2004, Improving Q estimates from
seismic reflection data using well-log-based localized spectral
correction: Geophysics, 69, 1521-1529.
Horne, S. and MacBeth, C., 1997, AVA observations in
walkaround VSPs, 67th Ann. Internat. Mtg: Soc. of Expl.
Geophys., 290-293.
Hsu, C.-J. and Schoenberg, M., 1993, Elastic waves through a
simulated fractured medium: Geophysics, 58, 964-977.
Hudson, J. A., Liu, E., Crampin, S., 1996, The mechanical
properties of materials with interconnected cracks and pores:
Geophysical Journal International, 124, 105-112.
ANI 2.2
Pointer, T., Liu, E., and Hudson, J.A., 2000,. Seismic wave
propagation in cracked porous media: Geophysical Journal
International, 142, 199-231.
Quan Y., Harris J.M., 1997, Seismic attenuation tomography using
the frequency shift method: Geophysics, 62, 895-905.
Rüger, A., 1997, P -wave reflection coefficients for transversely
isotropic models with vertical and horizontal axis of symmetry:
Geophysics, 62, 713-722.
Schoenberg M., Douma J., 1988, Elastic wave propagation in
media with parallel fractures and aligned cracks: Geophys.
Prosp., 36, 571–590.
Thomsen, L., 1995, Elastic anisotropy due to aligned cracks in
porous rock: Geophysical Prospecting, 43, 805–829.
Appendix: Model input parameters
In equation (1), the attenuation 1/Q is expressed through
the imaginary part of the complex weakness '~ N , which
was calculated for the equant-porosity model of Hudson et
al. (1996), at seismic frequency f 30 Hz. The small crack
aspect ratio, D 0.001 , is chosen, because it gives
relatively large attenuation values (e.g., Maultzsch et al.,
2003). In Figure 6, vertical lines 1-3 signify the values of
the attenuation, 1/Q, and corresponding values of
(permeability K r * porosity I p ), chosen for the cases of
gas, brine and oil crack fill. Other parameter values are
given in Table 1, for example, fluid viscosity K f , fluid P-
wave velocity V f and fluid density U f .
Fig. 6: The attenuation 1/Q vs
K r * I p , in the isotropy plane.
The input crack parameters
D c / a 0.001 , c 10 4 m and
e 0.095 . Other parameters
are given in Table 1.
Normal and tangential weaknesses ' N and ' T are
Lynn, H., 2004, The winds of change. Anisotropic rocks – their calculated as: ' N 4e /[3 g (1  g )] and ' T 16e >33  2 g @ ,
1 K
preferred direction of fluid flow and their associated seismic
signatures – Part 1/ The Leading Edge, p.1156-1162. where e is crack density, K N f /(SPD (1  g )) , N f – fluid
Lynn, H. and Beckham W., 1998, P- wave Azimuthal variations in bulk modulus (see Table 1), P – host-rock shear modulus,
attenuation, amplitude and velocity in 3D field data:
Implications for mapping horizontal permeability anisotropy: P =1.47.1010 Pa; VP = 4000 m/s, VS = 2000 m/s, U = 2550 kg/m3.
68th Annual International Meeting, SEG, Expanded Abstracts,
193 – 196. Table 1. Crack-fill and other model parameters
MacBeth, ɋ., 1999, Azimuthal variation in P -wave signatures Crack Vf Uf Nf K Pa·s K r
f Ip
due to fluid flow: Geophysics, 64, 1181-1192. Model
Maultzsch S., Horne S., Archer S., Burkhardt H. 2003, Effects of m/s kg/m3 Pa mD %
an anisotropic overburden on azimuthal amplitude analysis in 1) Gas 620 65 2,50·107 0,00002 1 1
horizontal transverse isotropic media // Geophysical 2 ) Brine 1710 1100 3,22·109 0,001 10 10
Prospecting, 51, 61-74. 3 ) Oil 1250 800 1,25·109 0,02 100 10

SEG/Houston 2005 Annual Meeting 130

EDITED REFERENCES

Note: This reference list is a copy-edited version of the reference list submitted by the
author. Reference lists for the 2005 SEG Technical Program Expanded Abstracts have
been copy edited so that references provided with the online metadata for each paper will
achieve a high degree of linking to cited sources that appear on the Web.

QVOA analysis as an instrument for fracture characterization

REFERENCES
Carcione, J. M., H. B. Helle, and T. Zhao, 1998, Effects of attenuation and anisotropy on
reflection amplitude versus offset: Geophysics, 63, 1652-1658.
Chapman, M., 2003, Frequency-dependent anisotropy due to meso-scale fractures in the
presence of equant porosity: Geophysical Prospecting, 51, 369-379.
Chichinina, T., V. Sabinin, and G. Ronquillo-Jarillo, 2004, P-wave attenuation anisotropy
in fracture characterization: numerical modeling for reflection data: 74th Annual
International Meeting, SEG, Expanded Abstracts, 143–146.
Clark, R. A., A. J. Carter, P. C. Nevill, and P. M. Benson, 2001, Attenuation
measurements from surface seismic data: azimuthal variation and time–lapse case
studies: 63rd Annual Conference, EAGE, Extended Abstracts, L28.
Dasgupta, R., R. A. Clark, 1998, Estimation of Q from surface seismic reflection data:
Geophysics, 63, 2120–2128.
Hackert, C. L. and J. O. Parra, 2004, Improving Q estimates from seismic reflection data
using well-log-based localized spectral correction: Geophysics, 69, 1521-1529.
Horne, S. and C. MacBeth, 1997, AVA observations in walkaround VSPs: 67th Annual
International Meeting, SEG, Expanded Abstracts, 290-293.
Hsu, C.-J. and M. Schoenberg, 1993, Elastic waves through a simulated fractured
medium: Geophysics, 58, 964-977.
Hudson, J. A., E. Liu, and S. Crampin, 1996, The mechanical properties of materials with
interconnected cracks and pores: Geophysical Journal International, 124, 105-112.
Lynn, H., 2004, The winds of change. Anisotropic rocks – their preferred direction of
fluid flow and their associated seismic signatures – Part 1: The Leading Edge, 23,
1156-1162.
Lynn, H., and W. Beckham, 1998, P-wave azimuthal variations in attenuation, amplitude
and velocity in 3D field data: Implications for mapping horizontal permeability
anisotropy: 68th Annual International Meeting, SEG, Expanded Abstracts, 193 –
196.
MacBeth, C., 1999, Azimuthal variation in P -wave signatures due to fluid flow:
Geophysics, 64, 1181-1192.
Maultzsch S., S. Horne, S. Archer, and H. Burkhardt, 2003, Effects of an anisotropic
overburden on azimuthal amplitude analysis in horizontal transverse isotropic
media: Geophysical Prospecting, 51, 61-74.
Pointer, T., E. Liu, and J. A. Hudson, 2000, Seismic wave propagation in cracked porous
media: Geophysical Journal International, 142, 199-231.
Quan, Y., and J. M. Harris,1997, Seismic attenuation tomography using the frequency
 shift method: Geophysics, 62, 895-905.
Rüger, A., 1997, P-wave reflection coefficients for transversely isotropic models with
vertical and horizontal axis of symmetry: Geophysics, 62, 713-722.
Schoenberg, M., and J. Douma, 1988, Elastic wave propagation in media with parallel
fractures and aligned cracks: Geophysical Prospecting, 36, 571–590.
Thomsen, L., 1995, Elastic anisotropy due to aligned cracks in porous rock: Geophysical
Prospecting, 43, 805–829.

View publication stats