Depth Imaging of Foothills Seismic Data

Our CSEG luncheon talk (October 11, 2000) will give highlights of the CSEG publication entitled “Depth Imaging of Foothills Seismic Data”. The book and its accompanying short course (given at five Canadian universities) were both products of a CSEG Superfund Project and a CSEG Fiftieth Anniversary Project. The talk will review the complex structural geology of the Canadian Foothills, the concept of prestack depth migration, and the necessity of iterative interpretation in the development of a seismic velocity model for migration. Detailed explanations, data examples, and computer codes on Foothills imaging can be found in our CSEG book.

The structural geology of the Western Canadian Foothills is dominated by a series of thrust faults, complex folds, and steeply dipping formations. In order to better understand the Foothills and the possible structural petroleum traps resulting from faulting and folding, one can examine the beautiful outcrops in the Front Ranges of the Rocky Mountains. Having gained an appreciation for the complexity of these structures and structural types, we can derive geological models through seismic imaging. In estimating our models, we should be careful to check for agreement with the seismic data, outcrops, and well information, and for geological correctness, which should include attention toward structural and thrust fault evolution of structures over geological time.

Depth imaging is an essential tool in the petroleum exploration of such complicated geology. Due to the structural complexity and steep dips of formations, migration is a process that is necessary for the correct positioning of seismic reflection events. Migration is defined as a processing step for placing seismic reflectors in their true subsurface location and has been successfully used in Foothills exploration for decades.

In our discussions we compare and contrast four of our favorite imaging techniques including phase-shift, f-x, Kirchhoff, and reverse-time depth migration methods. All of these have relative merits and shortcomings, many of which have been discussed by Sam Gray and colleagues. In several cases, poststack migrations can provide useful first approximations to the subsurface geology. However, we almost always want to end up performing prestack depth migration to obtain the most accurate images. This approach of initially using poststack depth migration followed by iterative applications of prestack depth migration was used successfully by Ratcliff et al. (The Leading Edge, 1994, p. 163-172) for the imaging of Gulf of Mexico salt intrusions. We have found that this approach is also valid for the imaging of Foothills structures. Prestack depth migration is the method that makes the fewest incorrect assumptions in the depth imaging of complex structures. More specifically, it is the preferred method for the following reasons:

1. Prestack migration does not use common midpoint stacking, which will smear reflections from steeply dipping layers.
2. Unlike time migration, depth migration can accommodate the lateral seismic velocity variations found in foothills geology.
3. Migration from topography can be readily incorporated into prestack depth migration, thereby obviating many of the problems with conventional stacking analysis.
4. Prestack depth migration can itself be used as a velocity analysis tool – either through focusing analysis or smiles/frowns curvature analysis.
5. Depth images are amenable to structural balancing or “palinspastic restoration” which is a recommended step in the development of geological models.
6. Anisotropy can be readily incorporated into prestack depth migration techniques.

We evaluate these aspects of prestack depth migration through the use of synthetic and real seismograms. In examining the advantages of prestack migration from topography we look at models and real data from the Shaw Basing area of the Alberta Foothills. In searching out problems with images, we examine the adequacy of algorithms, the preprocessing of data, and the issues of migration from topography and near-surface velocity variation. The Shaw Basing case history basically summarizes research presented by Yan and Lines at the SEG 2000 annual meeting.

Finally, we examine the effects of anisotropy on seismic imaging, while revisiting much of the research done by Don Lawton and co-investigators in the Fold-Fault Research Project (FRP) at the University of Calgary. FRP researchers have shown with physical models and real data that anisotropy generally causes seismic images to be mispositioned when isotropic techniques are used. The solution to the mispositioning problem is to use anisotropic migration algorithms with the correct anisotropic rock property parameters. Both the development of anisotropic algorithms and the estimation of physical parameters are important problems in the correct imaging of Foothills seismic data – especially when shale layers are present. The advantages of anisotropic depth migration are shown on both model and real data.

In our presentation, we attempt to present a complete picture of depth imaging for Foothills data. Nevertheless, many problems still remain. Many of these problems involve the successful integration of processing and interpretation, the effective handling of anisotropy, and the collection and imaging of 3-D Foothills seismic data. The future will undoubtedly involve many interesting challenges for those involved in depth imaging of Foothills seismic data.