You are on page 1of 6

The old and the new in seismic inversion

Brian Russell and Dan Hampson


Veritas Hampson-Russell

Introduction
Seismic inversion is a technique that has been in use by
geophysicists for over forty years. Early inversion techniques
transformed the seismic data into P-impedance (the product
of density and P-wave velocity), from which we were able to
make predictions about lithology and porosity. However,
these predictions were somewhat ambiguous since P-imped-
ance is sensitive to lithology, fluid and porosity effects, and it
is difficult to separate the influence of each effect. To perform
a less ambiguous interpretation of our inversion results,
we must perform full elastic inversion, in which we estimate
P-impedance, S -impedance (the product of density and
S-wave velocity) and density. The reason for this can be seen
in Figure 1, which plots the P and S-wave velocities as a
function of gas saturation. In this figure, it can be noted that
the P -wave velocity drops dramatically when gas is
introduced into the reservoir whereas the S -wave velocity
is largely unaffected by the introduction of the gas.
Figure 1. The effect of gas saturation on P and S-wave velocity.
This talk will present both a history of seismic inversion and
an overview of the inversion techniques themselves. Continued on Page 7

Brian Russell started his career as an exploration geophysicist with Chevron in 1976, and worked for
Chevron affiliates in both Calgary and Houston. He then worked for Teknica Resource Development Ltd.
and Veritas Seismic Ltd. in Calgary before co-founding Hampson-Russell Software Ltd. in 1987 with Dan
Hampson. Hampson-Russell develops and markets seismic inversion software which is used by oil and
gas companies throughout the world. Since 2002, Hampson-Russell has been a fully owned subsidiary of
VeritasDGC. Brian is currently Vice President of Veritas Hampson-Russell and is involved in both
geophysical research and training. He is also an Adjunct Professor in the Department of Geology and
Geophysics at the University of Calgary and is involved with CREWES (Consortium for Research in
Elastic Wave Exploration Seismology) and CHORUS (Consortium for Heavy Oil Research by University Scientists).
Brian was President of the CSEG in 1991, received the CSEG Meritorious Service Award in 1995, the CSEG medal in 1999,
and CSEG Honorary Membership in 2001. He served as chairman of The Leading Edge editorial board in 1995, technical
co-chairman of the 1996 SEG annual meeting in Denver, and as President of SEG in 1998. In 1996, Brian and Dan Hampson
were jointly awarded the SEG Enterprise Award, and in 2005 Brian received Life Membership from SEG.
Brian has presented numerous technical papers at geophysical conferences around the world, including the SEG, EAGE,
CSEG and ASEG conferences. His papers have been published in Geophysics, The Leading Edge, Exploration Geophysics
and The Journal of Petroleum Geology. His book Introduction to seismic inversion methods, based on course notes from
an SEG Continuing Education course, was published by the Society of Exploration Geophysicists in 1988.
Brian holds a B.Sc. in Geophysics from the University of Saskatchewan, a M.Sc. in Geophysics from the Durham University,
U.K., and a Ph.D. in Geophysics from the University of Calgary. He is registered as a Professional Geophysicist in Alberta.

DECEMBER LUNCHEON JANUARY LUNCHEON


DATE: December 18, 2006 January 22, 2007
TIME: 11:30 A.M. Lunch DHI / AVO best practices
LOCATION: Telus Convention Centre, Calgary methodology and applications
TICKETS: Contact CSEG office William A. Fahmy
TELEPHONE: 262-0015 or Fax: 262-7383 SEG / AAPG Fall 2006 Distinguished Lecturer

December 2006 CSEG RECORDER 5


Luncheon Contd
The old and the new in seismic inversion
Continued from Page 5

Seismic interpretation Post-stack seismic inversion


The seismic reflection method was developed in the first quarter The seismic traces in the stacked seismic section shown in Figure
of the twentieth century and was used initially as a tool for iden- 2 can be modelled as the convolution of the earths reflectivity
tifying structures, such as anticlines, which could act as trapping and a bandlimited seismic wavelet, which can be written
mechanisms for hydrocarbon reservoirs. Such a structure can be
seen in Figure 2, which shows a seismic line recorded over a gas- (1)
charged sand. The high on the picked structure corresponds to where st is the seismic trace, wt is the seismic wavelet and rt is the
the gas sand. The inserted curve is the P -wave sonic log. But this reflectivity. The reflectivity, in turn, is related to the acoustic
structural interpretation is ambiguous when it comes to identi- impedance of the earth by
fying gas sands. By the 1970s, geophysicists had started to realize
that information was contained in the amplitudes of the seismic (2)
reflections themselves. This information could be correlated with
porosity changes, lithology changes, or even fluid changes where rPi is the zero-offset P -wave reflection coefficient at the ith
within the subsurface of the earth. interface of a stack of N layers and ZPi=iVPi is the ith -impedance
of the ith layer, where is density, VP is P-wave velocity and *
The amplitude anomalies on a seismic section became known as
denotes convolution. Lindseth (1979) showed that if we assume
bright-spots and were considered to correlate well with gas
that the recorded seismic signal is as given in equation (2), we
sands. A typical bright-spot is highlighted by the rectangle in
can invert this equation to recover the P-impedance using the
Figure 1, and is associated with the gas sand. Unfortunately,
recursive equation given by
bright-spots were also ambiguous with respect to identifying
fluid anomalies, which lead to the development of the AVO
(amplitude variations with offset) technique, to be discussed (3)
later. However, let us first look at how we can invert the
section shown in Figure 2. By applying equation (3) to a seismic trace we can effectively
transform, or invert, the seismic reflection data to P-impedance.
However, as also recognized by Lindseth, there are a number of
problems with this procedure. The most severe problem is that
the recorded seismic trace is not the reflectivity given in equation
(2) but rather the convolutional model given in equation (1).
The effect of the bandlimited wavelet is to remove the low
frequency component of the reflectivity, meaning that it cannot be
recovered by the recursive inversion procedure of equation (3).

After proper processing and scaling of the seismic data, an intuitive


approach to recovering the low frequency component is to simply
extract this component from well log data and add it back to the
seismic. However, this is a fairly ad-hoc procedure, and a more
recent approach to inversion is called model-based inversion
Figure 2. A seismic section from Alberta, in which the picked even between a time (Russell and Hampson, 1991). In model-based inversion we start
of 600 and 650 ms represents a seismic structure and the rectangle highlights an
amplitude anomaly, or bright spot, associated with the gas sand.
with a low frequency model of the P-impedance and then perturb
this model until we obtain a good fit between the seismic data and

Figure 3. The extracted wavelet from the seismic data of Figure 3, where (a) is the time response and (b) is the frequency response.

Continued on Page 8

December 2006 CSEG RECORDER 7


Luncheon Contd
The old and the new in seismic inversion
Continued from Page 7

a synthetic trace computed by applying equations (1) and (2). Both In Figure 4, notice that although the model-based results looks a
recursive and model-based inversion use the assumption that we little more geologically reasonable (it has a blockier, less
have extracted a good estimate of the seismic wavelet. The smoothed appearance, and less dramatic swings), both inver-
extracted wavelet from the stacked section of Figure 2 is shown in sions show a low-impedance zone at the gas sand zone, which is
Figure 3, where the time domain response of the wavelet is shown to be expected. However, there are also low impedance zones
on the left, and its frequency domain response on the right. elsewhere on both inversions, probably due to shales. Thus, low
impedance associated with bright amplitudes is not an unam-
Figure 4 shows the inverted sections for the seismic line in Figure biguous indicator of a gas sand.
2, where the top inversion was done using the recursive technique
and the bottom inversion was done using model-based inversion.
Pre-stack simultaneous inversion
The standard seismic data processing flow involves trans-
forming a set of CMP gathers into a stacked section. For example,
a few of the processed gathers that were used to create the
section shown in Figure 2 are shown in Figure 5, along with a
cut-out of the stacked section within the zone corresponding to
these gathers.

The assumption behind stacking is that the amplitudes on the


gather do not show much variation, so that stacking can be
considered as simply a noise cancellation technique. However, if
we look carefully at the gathers in Figure 5 around the zone of
interest (630 ms) it is clear that the amplitudes show a lot of vari-
ation from the near offset on the left to the far offset on the right.
In this case, we see an increase in amplitude, often associated
with an anomaly in which the impedance of the gas layer is less
than that of the surrounding shales.

The reason behind this can be seen in Figure 6, which shows that
an incident P-wave at an angle results in reflected and trans-
mitted P and S-waves. This is called mode conversion, and the
amplitudes of the reflected and transmitted waves can be
computed using the Zoeppritz equations (Zoeppritz, 1919).

Figure 4. Inversion of the seismic data shown in Figure 2, where (a) shows re c u r-
sive inversion and (b) shows model-based inversion, and the colour bar on the right.

Figure 5. Several of the CDP gathers used to create the stacked section of Figure 2 Figure 6. Mode conversion of an incident P-wave on an elastic boundary.
are shown at the top of this diagram, with their location on the final stack shown at
the bottom.

Continued on Page 9

8 CSEG RECORDER December 2006


Luncheon Contd
The old and the new in seismic inversion
Continued from Page 8

Figure 7. Crossplots of (a)


ln(ZD) vs ln(ZP) and (b)
ln(ZS) versus ln(ZP)
where, in both cases, a best
straight line fit has been
added. The deviations
away from this straight
line, LD and LS, are the
desired fluid anomalies.

(a) (b)

Figure 8. Simultaneous
P-impedance inversion of
the pre-stack data shown
in Figure 5, where the
ellipse highlights the gas
sand zone.

Figure 9. The VP/VS ratio


found by dividing
the inverted P and S-
impedance sections from
simultaneous inversion,
where the ellipse
highlights the gas
sand zone.

Continued on Page 10

December 2006 CSEG RECORDER 9


Luncheon Contd
The old and the new in seismic inversion
Continued from Page 9

The Zoeppritz equations are a set of four equations in four least-squares procedure can be implemented to extract the three
unknowns that are difficult to intuitively interpret. However, as reflectivity terms from the pre-stack seismic data.
shown by Aki and Richards (2002) a linearized version of these
equations can be written for the reflected P-wave in which we After we have extracted the three reflectivities in equation (5), they
divide the response into three terms. In the original Aki-Richards can be inverted using the post-stack inversion method described in
equation, the three terms were weighted values of changes in P- the last section. This is referred to as independent inversion.
wave velocity, S -wave velocity and density. However, their equa- However, Hampson et al. (2005) developed a new approach that
tion was reformulated by Wiggins et al. (1983) as uses a modification of equation (5) and allows us to invert directly
for P-impedance, S -impedance, and density. This method is
(4) referred to as simultaneous inversion. It was also the goal of this
work to extend the model-based post-stack impedance inversion
w h e re is a linearized approximation method described in the last section to perform pre-stack inversion.
to the zero - o ffset P-wave reflection coefficient, Although the mathematics of this approach will not be described
here (the interested reader is encouraged to read the expanded
and . abstract by Hampson et al., 2005), one of the key assumptions in
simultaneous inversion is that we can build linear relationships
Equation (4) forms the basis for the AVO (amplitude variations between the logarithms of P-impedance and S-impedance, LP and
with offset) method, in which the terms A and B, called intercept LS, and between LP and the logarithm of the density reflectivity, LD.
and gradient, are extracted from the seismic data using a That is, we are looking for deviations away from this linear fit
weighted stack method and are cros plotted and analyzed for given by LS and LD, as illustrated in Figure 7.
fluid anomalies. We will not discuss AVO in this article, but rather The inverted P -impedance from the simultaneous inversion of
jump directly to pre-stack inv sion, which can be considered as a the pre-stack data in Figure 5 is shown in Figure 8, and displays
quantitative extension of AVO. less of an impedance drop than the model-based post-stack
The Aki-Richards equation was re-formulated by Fatti et al. (1994) inversion shown in Figure 4. This is to be expected, since the
as a function of zero - o ffset P-wave reflectivity RP0, zero-offset S- amplitudes used in the post-stack inversion were increased due
wave reflectivity RS0 and density reflectivity RD in the form to the AVO effect. As in post-stack inversion, the P-impedance on
its own is not a gas sand indicator.
(5)
However, when we combine the plot shown in Figure 8 with the
w h e re ratio of the inverted P and S -impedances (which gives the VP/VS
ratio since the density terms cancel) shown in Figure 9, the
interpretation becomes more clear. Associated with a drop in
and
P-impedance is a drop in the VP/VS ratio, which is generally an
RP0 is equivalent to the A term in equation (4), and the other two indicator of a gas sand.
reflectivity terms are given by

and . Based on equation (5), a

(a) (b)

Figure 10. The results of cross-plotting the VP/VS ratio seen in Figure 6 against the P-impedance seen in Figure 5, where (a) shows the cross-plot, with a red zone high-
lighted, and (b) shows the resulting highlighted zone on the seismic section.
Continued on Page 11

10 CSEG RECORDER December 2006


Luncheon Contd
The old and the new in seismic inversion
Continued from Page 10

To show this more conclusively, Figure 10 show the results of a Hampson, D., Russell, B., and Bankhead, B., 2005, Simultaneous inversion of pre-stack
seismic data: Ann. Mtg. Abstracts, Society of Exploration Geophysicists.
crossplot of a small portion of the two sections shown in Figures
Lindseth, R. O., 1979, Synthetic sonic logs - A process for stratigraphic interpretation:
8 and 9. On the crossplot itself, shown on the left of Figure 10, Geophysics, 44 , no.1, 3-26.
we have highlighted a zone in which both P -impedance and
Russell, B. and Hampson, D., 1991, A comparison of post-stack seismic inversion
VP/VS ratio are low. This zone is displayed on the seismic methods: Ann. Mtg. Abstracts, Society of Exploration Geophysicists, 876-878.
section on the right-hand image in Figure 10. Notice the excel- Wiggins, R., Kenny, G.S., and McClure, C.D., 1983, A method for determining and
lent definition of the gas sand zone. displaying the shear-velocity reflectivities of a geologic formation: European patent
Application 0113944.
The example we have been considering is a 2D example and the Zoeppritz, K.,1919, Erdbebenwellen VIIIB, On the reflection and propagation of
method discussed in this talk can be applied to 3D datasets to seismic waves: Gottinger Nachrichten, I, 66-84.
give a spatial image of the reservoir. Such an example, taken
from the Gulf of Mexico, will be
shown in the expanded version of
this talk given during the CSEG
luncheon.

Conclusions
In this talk, we have discussed the
history of seismic amplitude inver-
sion, from its origin as a post-stack
process to the most recent develop-
ments which involves the simulta-
neous inversion of pre-stack seismic
data. Although post-stack inversion
is a powerful and robust method, it
suffers from the fact that its final
p roduct, P-impedance, does not
allow us to discriminate between
lithology, porosity and fluid effects.
This limitation was removed with the
development of both the AVO tech-
nique and simultaneous inversion of
p re-stack data. The simultaneous
inversion method that we discussed
is based on the assumptions that
reflectivity as a function of angle can
be given by the Aki-Richards equa-
tion, and that there is a linear rela-
tionship between the logarithm of
P-impedance and both S-impedance
and density. We illustrated our inver-
sion methods using a gas sand
example from Alberta. R

References
Aki, K., and Richards, P.G., 2002, Quantitative
S e i s m o l o g y, 2nd Edition: W.H. Freeman and
Company.
Fatti, J., Smith, G., Vail, P., Strauss, P., and Levitt,
P., 1994, Detection of gas in sandstone reservoirs
using AVO analysis: a 3D Seismic Case History
Using the Geostack Technique: Geophysics, 59,
1362-1376.

December 2006 CSEG RECORDER 11

You might also like