1
Z99
Analysis of C-wave PSTM results from three 2D
4C lines acquired offshore West Africa
1
1
2
Fabio Mancini , Paul Williamson , Xiang-Yang Li and Hengchang Dai
1
2
2
Geoscience Research Centre, Total E&P UK, Crawpeel Road, Aberdeen AB12 3FG, UK
EAP, British Geological Survey, Murchison House, West Mains Road, Edinburgh EH9 3LA, UK
Abstract
We present an analysis of C-wave PSTM images obtained on three 2D
multicomponent (4C) lines acquired offshore West Africa. We applied an anisotropic PSTM
sequence developed by the Edinburgh Anisotropy Project (EAP), characterised by a complex
parameterisation and robust parameter estimation with a relatively fast workflow. We also
analyse the overall contribution of converted waves in this area. The geological setting is not
ideal for S-wave propagation due to the presence of low-velocity unconsolidated shales,
which behave like mud. Nonetheless C-waves give improved images in the deeper part of the
section and clear definition of the main faults, which leads us to believe that joint
interpretation of P- and C-wave sections can help improve interpreter confidence. These lines
were also processed (in parallel) by two contractors. There are large differences in the
imaging results, which highlights a diversity of approaches to C-wave processing and the high
sensitivity of C-waves to the parameters used.
Introduction
In this paper we investigate imaging with C-waves in a study area (offshore Africa)
where P-waves are distorted and attenuated by a combination of shallow gas pockets and subvertical faults reaching the surface, see map in Figure 1a. Since C-waves are known to give
better imaging results than P-waves in zones under gas clouds (e.g. Granli et al., 1999,
Mancini et al., 2004) it was decided to acquire three 2D multicomponent (4C) lines to assess
the imaging benefits of C-waves.
This test was also seen as a good opportunity to evaluate the approach and tools
developed by the EAP consortium. Their methodology is based on the assumption of a VTI
medium, from which they have formulated a Kirchhoff Pre-Stack Time Migration described
in terms of four parameters: vcmig, γ0, γeff, and χeff, respectively the C-wave migration velocity,
the vertical velocity ratio, the effective velocity ratio and the anisotropy parameter. All these
parameters, except γ0, are defined during migration velocity analysis using an interactive
velocity analysis tool. This way we follow a “lean” and fast processing flow optimising all
parameters directly in the PSTM domain. More on the theory of the anisotropic PSTM can be
found in Dai and Li (2003) and details on its application are available in Mancini et al. (2004).
Analysis of results
Geologically the area is composed of interbedded sequences of sand and clay. Due to
the high rate of subsidence shallow shales are very low-density and low-velocity and have
similar properties to mud. This zone near the water bottom has strong effects on the upcoming S-waves, causing receiver statics of about 200 ms (see the large sag between receivers
170 and 370 in the C-wave receiver stack shown in Figure 1c). In this area the near surface
vp/vs ratio was estimated to reach values of 20.
Correcting for these effects was one of the main problems in this project. We used an
approach based on shallow event correlation. We interpreted and correlated two events,
assumed to be the same, in the P and C-wave receiver stacks, shown in red on Figure 1b and
Figure 1c. Since the sag in the C-wave section is not structural we consider the correct
shallow structure to be the one shown by the P-wave data, so we need to time shift the CEAGE 67th Conference & Exhibition — Madrid, Spain, 13 - 16 June 2005
2
wave event to match the P-event transformed into C-wave arrival times. To do so consistently
from line to line we opted to use an average vertical vp/vs = 7, derived by event correlation
(vp/vs=ts0/tp0 with ts0=tc0-tp0; where tp0 and ts0 are the one way vertical travel times for P and Swaves and tc0 is the two way C-wave traveltime). The time-shifted P-event is shown in blue in
Figure 1d. We then calculated the differences in arrival times for each receiver between the
two events. After the correction we can verify that the static corrected C-wave event matches
the time shifted P-event, (blue line) Figure 1d. The red line here shows the original position of
the C-event, confirming that estimated average vp/vs is reasonable. This method proved to be
robust and consistent from line to line, giving excellent ties at the merging locations.
The remaining C-wave processing workflow followed in this study requires an initial
estimation of γ0 by event correlation in the P and C-wave stacked sections (pseudo zerooffset). After this step we move directly in the PSTM domain, in which we define the
remaining three parameters, vcmig, γeff, and χeff, via anisotropic migration velocity analysis on
selected Common Image Points (CIPs). For this task we make use of an interactive velocity
analysis tool, which allows the simultaneous picking of all the parameters. vcmig is responsible
for the flatness of the gather at near-middle offsets, χeff for the flatness of the gather at far
offsets, while γeff defines the symmetry of the events in the positive and negative offsets.
Although the full VTI parameterisation appears complex, with this workflow we are able to
define the model quite rapidly and efficiently.
b
a
c
d
Figure 1: (a) Map of the area. (b) P-wave receiver stack with shallow event interpretation, (c) Raw C-wave
receiver stack with interpretation and (d) Corrected C-wave receiver stack.
The first line we worked on was Line 2, although it was not positioned in an area of
high P-wave disturbance. Figure 2a shows the P-wave results (after hydrophone and vertical
geophone (PZ) summation and PSTM), in which we can notice the numerous faults reaching
the surface and some shadowing effects (for example at CDP 2550). The vertical faults are
well imaged but the dome structure in the central part is less well defined. Figure 2b shows
the final PSTM C-wave image obtained at the GRC. This image is quite noisy, generally
lower frequency (even in the shallow part) but has a better definition of the top of the dome.
Faults are very well defined, probably better than in the PZ section, due to the difference in Cwave raypaths.
Figure 3a shows the P-wave results for Line 1. In this area the fault-shadowing effects
are quite strong on the left part of the section, due to possible gas-leakage through the faults.
Looking at the GRC C-wave section, Figure 3b, we see good continuity of the events in the
shadow-zone and a much better definition of the shale-dome.
3
a
b
Figure 2: (a) PZ PSTM stack, (b) GRC C-wave PSTM image
a
b
Figure 3: (a) PZ PSTM stack, (b) internal C-wave PSTM image
Figure 4a and Figure 4b show the C-wave results for Line 1 from contractor 1 and
contractor 2. The processing flow followed by the two contractors also included C-wave
PSTM but parameterisation and parameter estimation was different in each case. The three Cwave images look comparable in the shallow zone, apart from variations due to the static
corrections, but they differ significantly in the definition of the shale-dome structure (about
2.5 ms C-time). Similar differences were observed for Line 2, while results for Line 3 (not
shown here) were more comparable.
What we think is an important result of this comparison is to notice that there is a
diversity of approaches to C-wave processing. Furthermore, C-wave imaging requires a
higher number of parameters than P-wave imaging and, as a consequence of this, the results
are more sensitive to the parameterisation used. This leads to large differences in the imaging
results depending on the approach followed. In this test the results obtained with the EAP
approach were generally preferable.
a
b
Figure 4: (a) contractor 1 PSTM image; (b) contractor 2 PSTM image
Joint P and C-wave data analysis
In Figure 5a we show a structural interpretation on the PZ image for Line 2. The same
faults interpreted on the P-wave section can be mapped more easily on our C-wave image,
EAGE 67th Conference & Exhibition — Madrid, Spain, 13 - 16 June 2005
4
Figure 5b. As mentioned before, C-wave raypaths allow better imaging of the shallow faults.
On this section we marked two red circles showing additional discontinuities. While the one
on the left is probably due to loss of resolution in the PSTM image due to the effects of the
superficial reef, the one in the middle seems related to faulting. So while the C-wave section
on its own does not appear to have greatly improved the imaging, differences can be seen
compared to the PZ section. These differences between P- and C-images can be a source of
valuable information. Joint interpretation of the two sections could improve interpreter
confidence in fault placement.
a
b
Figure 5: Structural interpretation on the PZ image (a) and on the GRC C-wave image (b). Yellow circles
indicate the position of the reefs, the red circles indicate discontinuities possibly related to faults,
the black circle indicates an out-of-plane event.
Discussions and conclusions
The area studied presents several challenges to P-wave imaging: fault shadowing
effects, high absorption due to gas-pockets and low-density, low-velocity shales. C-waves did
not solve all the imaging problems; we did not see the dramatic improvement reported in
more "classic" studies of gas-affected areas. These low-density shales, comparable to mud in
many respects, caused problems for shear-wave propagation, resulting in loss of resolution.
Nonetheless we saw improvements in the images of the deeper regions and in the definition of
the sub-vertical faults. There are differences in the images obtained with the two wave-types,
which require careful interpretation but could be a source of valuable information.
We notice large differences between the three C-wave images produced for each line
by the GRC and by two contractors. These differences highlighted the diversity of processing
approaches used in this case, corresponding to the increased number of parameters in
comparison to P-wave imaging. Parameterisation and parameter estimation differ from
contractor to contractor. Due to this complexity C-wave results are more sensitive to the
procedure used compared to P-waves. In this test the EAP approach provided the most
satisfactory results. This may be due to the parameterisation and robust parameter estimation,
but, due to the nature of the dataset, we cannot attribute the improvements with certainty.
Acknowledgement
The authors would like to acknowledge Total E&P UK and the Total subsidiary in
West Africa and partners for permission to publish this paper. We also thank Jean-Luc Boelle,
Alain Lortscher and Christian Chappey for their help with this work.
References
Dai H., and Li, X.-Y., 2003, Migration velocity analysis of C-wave using INMO-CIP gathers
of PKTM: a case study from the Gulf of Mexico, 65th Mtg EAGE Conference, Exp. Abstr.
Granli, J. R., Arntsen, B., Sollid, A., and Hilde, E., 1999, Imaging through gas-filled
sediments using marine shear wave data, Geophysics, 64, 668-677.
Mancini, F., Li., X.-Y., Dai H., Ziolkowski, A., and Pointer, T., 2004, C-wave anisotropic
imaging using PSTM: a case example from the North Sea, 66th Mtg EAGE Conference, Exp. Abst.