POST-STACK STEEP DIP ENHANCEMENT AIDS SEARCH AROUND SALT STRUCTURES

Davis W. Ratcliff
Amoco Production Co.
Houston

Naide Pan
PGS Tensor Geophysical Inc.
Houston

Recent interest in drilling targets near and beneath subsurface salt bodies has challenged geophysicists to improve imaging of steeply dipping reflectors of seismic energy.

An important new imaging technology is post-stack steep dip enhancement (SDE).

It is an interpretive tool that has obvious application along the flanks of salt. And it applies to other structures with steeply dipping horizons, such as faults.

In effect, SDE compensates for a bias in unmigrated seismic data toward reflections from flat or gently dipping features.

Reflections from steeply dipping horizons are commonly weakened by a number of phenomena during acquisition and processing.

These include common depth point (CDP) smear, velocity dependence on dip, receiver arrays, stacking, and filtering.

DMO EFFECTS

One technique geophysicists use to preserve steep-dip reflection strength is dip moveout (DMO), which compensates for at least some of the harmful effects of CDP smear and velocity dependence on dip. DMO preserves reflection strength of steeply dipping events during stack.

Even after DMO, however, steeply dipping events in seismic data sets often remain weaker than energy from flat or more gently dipping reflectors.1 This can be due to premature truncation of the DMO operator, particularly in data processed with DMO algorithms used during 1980-90.

Post-stack SDE compensates for energy loss of steeply dipping events that have been attenuated by faulty DMO algorithms, receiver arrays, and improper stacking.

The technique is based on the recognition that steeply dipping salt-face reflections are weak in 2D and 3D data processed during 1980-90. It isolates the weakened energy, boosts it up, adds it back to the unmigrated stack, and migrates the enhanced seismic data.

The result is a much-improved image of steeply dipping horizons, such as the flanks of salt bodies. A sharp image of the salt-sediment interface can be crucial in targeting well bores.

An alternative to post-stack SDE is to return to field reels and reprocess the unstacked data with modem DMO algorithms. Because it involves processing many times more traces, however, doing so is much more costly than remigrating stacked data with SDE.

For SDE to work, some amount of steep-dip energy must be preserved in the 2D or 3D stack. A crucial assumption is that the DMO algorithm did not completely filter out the steeply dipping events.

The geophysicist must be able to distinguish weak salt signatures and fault planes in the data set from the stronger energy of more gently dipping events. He or she therefore must study much seismic data and conduct modeling to see how steep-dip energy should look. The energy can be digitized and migrated in a modeling routine, then compared with events in the seismic data.

SDE is a structural imaging tool, not a relative amplitude preserving technique. The geophysicist must not boost up energy of events not evident in the data, which is why recognition of salt and fault plane signatures is so important.

The figures show how post-stack SDE has improved images of salt domes in the Gulf of Mexico.

PROCESSING SEQUENCE

Fig. 1 shows the processing sequence for a 2D seismic line. Section 1 is the finished product: a salt dome imaged via steep-dip migration with post-stack SDE.

Section 2 is the starting point: an unmigrated stack with 1987 DMO processing. Gently dipping events are strong on both sides of the dome, and there are hints of the unmigrated position of the salt face reflection on both sides.

Section 3 is the step taken to determine whether steep-dip energy exists in the seismic data. The DMO stack is run through a dip filtering procedure to remove gentle dips. The procedure makes the steep-dip energy stand out, as the section shows.

All the linear energy that stands out in Section 3 is contained in Section 2, but it is very weak. What Section 3 shows is the salt face signature, the unmigrated response of a salt face reflection. The unmigrated reflection from a fault plane, or fault plane signature, appears as well. The section also shows a sediment signature, which in this case happens to be the unmigrated position of hydrocarbons.

The steep-dip signature plot is a diagnostic step taken to determine whether steep-dip energy is preserved in the stack. If the energy is preserved in the stack, it can be boosted up by increasing the amplitude strength of the steep-dip energy.

In basic post-stack SDE, the boosted-up steep-dip energy is added back to the unmigrated DMO stack in Section 2. Then the whole section is migrated to produce the image of Section 1.

FURTHER STEPS

Fig. 2 shows further steps that can be taken in conjunction with post-stack SDE.

In Section 4 of the figure, the steep-dip signature plot is migrated with the steep-dip migration algorithm. The result is a quick look the salt shape.

Here, the fault plane an sediment signatures migrate to their correct positions relative to the salt. This step sometimes produces more noise in the data than is apparent here. Although the salt, sediment, and fault plane signatures appear in their proper positions, the salt-sediment interface remains ambiguous because of steeply dipping multiples associated with the salt face reflection.

In Section 5, dip-dependent deconvolution has suppressed the multiples in Section 4. The process partitions the stacked data into different dip bands, deconvolves them, and puts them back together. In essence, it reduces the reverberations on the salt face apparent in Section 4.

Fig. 3 provides another example of how post-stack SDE enhances 2D seismic imaging of a salt body. Section 6 is an unmigrated DMO stack processed in 1988. Section 8 shows the result of migration of the DMO stack without SDE.

In Section 7, SDE has been applied to the unmigrated DMO stack. When it is migrated, the salt body image--shown in Section 9--is much clearer than the image in Section 8 (without SDE).

3D APPLICATIONS

Applying post-stack SDE to 3D seismic data requires an algorithm that works in three dimensions.

The algorithm in effect performs a two-pass [see equation] transform, which is a procedure that transforms the time domain 3D data set into a dip-angle domain 3D data set. In this domain, the steeply dipping salt face reflections are easier to identify in three dimensions.

Once this is accomplished, the energy can be boosted up and run through dip-dependent deconvolution in three dimensions.

Fig. 4 shows unmigrated sections from three 3D seismic lines over the same salt dome, with and without SDE. The salt face appears as deep as 7.5 sec in the examples after SDE. This deeper energy is associated with turning wave energy, reflections from dips greater than 90. After SDE is applied, the reflectors with dips greater than 90 are easier to identify.

Fig. 5 compares 3D migration with and without SDE, again over a common salt body. On every line, the salt-sediment interface is stronger and more continuous after SDE.

And Fig. 6 shows an example of a very dear salt-body image produced with post-stack SDE applied to 3D data.

CONCLUSIONS

Post-stack SDE is not a solution to every imaging problem that occurs with salt or other structures presenting the migration challenges of steep dips.

It is one technique in a rapidly growing body of technology that is enabling explorationists to identify prospects around salt and drill accurately to them as never before.

The key to post-stack SDE is recognizing salt signatures, fault plane signatures, and gently dipping sediments in data sets processed in the 1980s. Once recognized, the energy from steep-dip reflections can be boosted up and added back to the stack for migration.

It's an interpretive process that requires area experience and judgment. Handled properly, it can help produce salt body images of unprecedented sharpness and accuracy.

ACKNOWLEDGMENTS

The authors would like to thank Todd Jones and John Etgen of Amoco Production for implementing the 3D steep-dip migration algorithm on the seismic data. They also thank George McMillian and Richard Crider of Amoco Production and Tim Hagen of PGS Tensor Geophysical for their assistance in the implementation of the 2D and 3D SDE procedures.

REFERENCE

Ratcliff, D.W., Gray, S-H., Whitmore, N.D., "Seismic imaging of salt structures in the Gulf of Mexico," The Leading Edge, Vol. 11, No. 4, April 1992, pp. 15-31.