-
f Beyond Kirchhoff: New Technologies for Improved Seismic Imaging
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, PGCE 2006, Nov 2006, cp-256-00038
Abstract
The Kirchhoff algorithm is a standard tool for imaging subsurface structures. These methods are efficient, versatile and can produce accurate images in moderately complex geology. As the drive to image hydrocarbon reservoirs moves to increasingly complex areas, the limitations of Kirchhoff algorithms become evident. Typically, these limitations include: use of single-arrival travel times, ray-paths computed with a high-frequency approximation, anti-alias filters that can only be correct for a single dip angle. These<br>limitations further complicate the demanding problem of preserving the amplitude response. The Kirchhoff algorithm can be enhanced; for example using multiple-arrival travel times or using band-limited travel times. However, such improvements can only incrementally improve the image quality and at great cost to algorithmic efficiency. Recent years have seen exciting developments in both imaging algorithms and computer hardware. These have lead to the implementation of alternative methods that can potentially overcome the limitations of Kirchhoff particularly those where multi-pathing is an issue. New imaging technologies based on solutions to the wave equation have provided significant improvements in areas with complex geology. For example, Wave Equation Migration (WEM) has become routine in areas such as sub-salt exploration in the Gulf of Mexico and is now also providing benefits in other exploration areas. Another technique, Beam Migration is based on wavefield continuation. The method retains strengths of Kirchhoff migration such as steep dip and anisotropy handling, but can also image multiple arrivals. Finally, a solution based on the full 2-way wave equation has recently been implemented. Although at an early stage of development, this Reverse Time Migration promises highest fidelity images in the most demanding situations. In this paper, we describe each of these migration methods and show examples of their use on synthetic and field data.