3-D Tomographic Amplitude Inversion For Compensating Transmission Losses in The Overburden
3-D Tomographic Amplitude Inversion For Compensating Transmission Losses in The Overburden
3-D Tomographic Amplitude Inversion For Compensating Transmission Losses in The Overburden
Barry Hung*
Kefeng Xin*
Sergey Birdus*
James Sun*
disadvantage because similar information is needed and reference amplitudes information from step (1),
by velocity updating anyway. In the following, we the tomographic inversion equation for velocity
describe the implementation of our method using update as described by Zhou et al (2003) can be
common image gathers (CIGs). We then apply this modified for tomographic amplitude correction with
method on synthetic and real data to demonstrate its this equation:
ability to compensate the amplitude loss caused by
transmission anomalies. Fm = a
(1)
offset range of 0 to 4.5km, obtained from the transmission anomalies were then utilized to
modeling using shots and receivers that have 25m mitigate the amplitude loss. Figure 10 depicts the
spacing and constant velocity of 2000 ms-1. It can compensation result where it can be seen that the
be observed that at locations where the raypaths do amplitudes of the reflectors below the first horizon
not pass through the transmission anomalies, such are nicely corrected. This is more evident when
as the location indicated by Figure 2(a), the viewing the depth slices from Figure 11 as the
amplitudes of the events do not vary with the offset. amplitudes along the horizon are more well-
On the contrary, at locations where the raypaths balanced after the correction.
pass through the transmission anomalies, the
amplitudes of the events below the absorption are
no longer constant at different offsets. This is CONCLUSIONS
illustrated by a selected location depicted in Figure
2(b). We have developed and demonstrated an inversion
method based on a post-migration common-image-
The data was then migrated with an amplitude gather tomography for correcting amplitude
preserving algorithm (Notfors, 2002) and the attenuation caused by transmission anomalies that
resulting CIGs of one subline are shown in Figure 3. can originate from anywhere within the overburden.
It can be seen from the horizons that amplitude Synthetic and real data results show that our method
anomalous streaks, similar to those observed by can be used to compensate for the amplitude loss
Harlan (1994), appear over offset and midpoint. associated with transmission anomalies.
Using the above-mentioned method to apply on the
CIGs to estimate the transmission anomalies, we
obtained a 3D volume of estimated transmission ACKNOWLEDGEMENTS
anomalies. Figure 4 displays one subline of the
transmission anomalies volume. It can be observed We thank OMV for providing us with the data and
that, apart from the boundary effects that are related CGGVeritas for the permission to publish this work.
to the size of the cells used in the calculation, the
transmission anomalies are well reconstructed. The
estimated anomalies can then be used to mitigate REFERENCES
the amplitude attenuation. Figure 5 shows the CIGs
after the compensation using our amplitude Brzostowski M. and McMechan G., 1992, 3-D
tomography method. The amplitude anomalous tomographic imaging of near-surface seismic
streaks that appeared in Figure 3 have clearly been velocity and attenuation: Geophysics, 57, 396-403.
removed. The effect of the compensation on AVO
is illustrated in Figure 6 where the amplitudes, Deal M., Matteucci G. and Kim Y., 2002, Turning
before and after our amplitude tomography, of the ray amplitude inversion: mitigating amplitude
second and fifth events are plotted against the attenuation due to shallow gas: 72nd Meeting, SEG,
offset. While the amplitudes of the events are Expanded Abstracts, 2078-2081.
almost restored to their original levels, indicated by
the solid lines, after the correction, the transmission González A. and Chambers R., 1994, Compensation
anomalies will have adverse effects on the AVO seismic amplitude for transmission losses in the
analysis if they are left uncorrected. overburden: 64th Meeting, SEG, Expanded
Abstracts, 1501-1504.
A real data set was then tested with our method.
Figure 7 displays the migrated stacked volume with Guo J., Zhou H., Young J. and Gray S., 2002,
four picked horizons. Amplitude dimming caused Toward accurate velocity. models by 3D
by gas in shallow region was observed in those tomographic velocity analysis: 64th Meeting,
areas circled by the dotted lines. The amplitude loss EAGE, B024.
makes it difficult for the reflectors to be identified
and interpreted for the AVO analysis. Using the Harlan W., 1994, Tomographic correction of
four picked events from the CIGs to perform the transmission distortions in reflected seismic
amplitude tomographic inversion, we obtained the amplitudes: 64th Meeting, SEG, Expanded
estimated transmission anomalies. Figure 8 shows Abstracts, 968-971.
the transmission anomalies extracted from one
subline. These correlate well with the velocity Notfors C., Zhang Y., Karazincir M., Sun J., Hung
pattern, as shown in Figure 9. The estimated B., 2002, Amplitude Preserving v(z) Prestack
Back to Menu
Kirchhoff Migration, Demigration and Modeling: Zhou H., Gray S., Young J., Pham D., and Zhang
64th Meeting, EAGE, B011. Y., 2003, Tomographic Residual Curvature
Analysis: The Process and its Components: 73rd
Taner M. and Koehler F., 1981, Surface consistent Meeting, SEG, Expanded Abstracts, 666-669.
corrections: Geophysics, 46, 17-22.
Back to Menu
Figure 1 - One subline of the 3D model used in the synthetic test. Five flat events are present in the model.
Transmission anomalies, which are constant along the inline direction, are represented by the
volumes colored in red.
Figure 2 - CMP gathers obtained from the modeling. The left gather, which is from crossline 4500, does
not have raypaths passing through the anomalies as shown in Fig 1.The right gather is from
crossline 5500 which have raypaths passing through the anomalies.
Back to Menu
Figure 3 - CIGs of one subline obtained after migration. Amplitude anomalous streaks are apparent.
Figure 5 - CIGs of the same subline shown in Fig 3 after amplitude compensation.
Back to Menu
Figure 6 - Plot of amplitude vs offset. The amplitudes of the second and fifth events are indicated by the
black and red curves respectively. The dotted lines and the circles indicate the amplitudes before
and after correction respectively. The solid lines are the original amplitudes.
Back to Menu
Figure 7 - Migrated stacked volume showing amplitude-dimming areas indicated by the dotted circles.
Four horizons, pointed out by the white arrows, are picked for inversion. Data courtesy of OMV
Figure 10 - Result after the application of our amplitude tomography method. The amplitudes are now
restored in the dimming areas.
Back to Menu
Figure 11 - Depth slices along the second horizon: (a) before the amplitude correction; (b) after the
amplitude correction.