Norman Daniel Whitmore, Jr.
Norman Daniel (Dan) Whitmore received a Ph.D. in geophysics in 1995 at the University of Tulsa for his research into the methods for plane wave imaging. He has 40 years of experience in geophysical research and applications in the areas of seismic modeling, processing, velocity estimation, multicomponent imaging and 3D imaging. He served as an individual contributor, mentor, instructor, consultant, and leader in seismic imaging technologies at Amoco, ConocoPhillips, and currently for Petroleum Geo-Services.
He joined Amoco Research in 1974 and began working in the pioneering group that developed finite difference wavefield modeling. He developed one of the earliest implementations of reverse time migration (RTM) in 1979 and demonstrated the ability of RTM to image arbitrary subsurface complexity and image data beyond 90 dip. In 2011 he received the Society of Exploration Geophysicists Reginald Fesseden Award for the development of RTM. Upon joining ConocoPhillips in 2001, Dan achieved the rank of geoscience fellow. He was an advisor in 3D imaging, 3D plane wave migration, and 3D tomographic velocity estimation. He served as the team leader of the Elastic Wave Technology project with specific emphasis on joint PP-PS imaging of North Sea reservoirs and providing technical input to the planning for a Life of Fields permanent monitoring system at Ekofisk Field. He joined Petroleum Geo-Services in 2009 and is presently a geophysical advisor in the depth imaging department, providing technical guidance and depth-imaging support in the application of 3D RTM to wide- and full-azimuth data.
Dan pioneered the use of reverse-time migration (RTM). In the migration workshop at the 1982 SEG Annual Meeting, Dan stunned the audience with the first publicly shown RTM image, of an overhanging Gulf Coast salt dome. Th is application and the subsequent developments of RTM are documented by Levin in “Principle of reversetime migration” (GEOPHYSICS, 1984). Shortly after Whitmore’s initial presentation, three nearly identical RTM algorithms were presented by McMechan (1983), Baysal et al. (1983), and Whitmore (1983). Dan’s use of prestack RTM on the VSP depth migration of a salt dome received Honorable Mention for Best Paper in GEOPHYSICS in 1986. While RTM is today recognized as the most general depth migration method, Dan’s pioneering research and application to production processing took place three decades earlier. Whitmore has dozens of other depth-migration papers in SEG publications, a widely cited PhD thesis on plane wave depth-migration, and has recently shown how RTM can utilize multiple refl ections for imaging.
2015 SEG Honorary Lecturer, North America
Concepts and applications of imaging with multiples and primaries
The goal of seismic acquisition and processing is to create a well-sampled image of the subsurface. Broadband acquisition has expanded the useable low and high frequencies of the seismic signal. Multistreamer acquisition provides a much greater receiver sampling. Wide- and full-azimuth acquisition and longer offsets have improved the azimuthal and angular illumination. The increased receiver coverage allows for a more complete sampling of the total wavefield. Multicomponent sensor technology both provides broadband acquisition and allows for separation of the downgoing and upgoing wavefields at the receivers, which subsequently can be used in imaging. Alternatively, there has been an increase in the use of ocean-bottom seismic, where the receiver sampling is relatively sparse, but the source sampling is typically dense. This presentation discusses imaging methods that take advantage of these advances in acquisition.
There is an array of imaging methods to choose from, including (1) fast beam methods used in model building, (2) Kirchhoff methods to address high-resolution imaging, (3) wavefield depth extrapolation and migration (WEM) used in more complex media, and (4) reverse time migration (RTM) for complex media and arbitrary dip.
This talk gives an overview of the principles and applications of RTM and WEM, which employ wavefield propagation of incident and reflected wavefields as part of their imaging process. It focuses on the combined use of single and higher order scattered wavefields, where the downgoing and upgoing wavefields can be used to construct subsurface images using both multiples and primaries.
The presentation covers the following topics:
- Prerequisites of processing before imaging, including designature, deghosting, wavefield separation, and possibly surface-related multiple estimation
- WEM and RTM imaging methods, including discussions on wavefield extrapolation, imaging conditions, and subsurface angle decompositions
- Imaging of primary and secondary (multiple) reflections using upgoing and downgoing separated wavefields. Also included is the estimation of surface-related multiple noise and subtraction in the image space.
- Beyond first-order imaging, including a discussion of trends in using inversion to achieve full- wavefield imaging
A number of examples will be shown, including:
- Imaging of primaries and multiples in deepwater and wide azimuth scenarios
- High-resolution imaging of primaries and multiples in shallow-water scenarios
- Imaging of ocean-bottom seismic, comparing the imaged upgoing primaries, mirror images, and full wavefield images using multiples
Biography Citation for the Reginald Fessenden Award
The early 1970s saw many talented scientists leave behind their classical training and take up the challenges of turning reflection seismology, particularly seismic data processing, into a modern, computationally intensive, scientific discipline. Among these scientists was a young math and physics graduate of Wartburg College in Iowa. While finishing his master’s degree in mathematics at Iowa State, hoping to get a PhD in Operations Research, he happened to be visited by Ken Kelly, another Iowa State alum, who was already paving new ground in seismic modeling at Amoco Production Research Company in Tulsa, Oklahoma with the likes of Rusty Alford, and Sven Treitel. Ken must have seen something in Dan, and the interview trip must have gone smoothly, as soon Dan found himself in Tulsa; the year was 1974.
Dan’s first assignment was to help with Amoco’s pioneering work in finite-difference wave propagation. Something about watching those early simulations of wave propagation created an inspiration for Dan. We don’t know for sure, but maybe one of those days while looking at how a modeled wavefield propagated and reflected from the many layers and boundaries in the model, he must have said to himself: “I can image the subsurface by running those waves backwards in TIME, and that will be the best seismic imaging method ever.” Other groups were already moving waves backward in depth, but as we all know, that requires considerable re-engineering of the wave equation into something it doesn’t want to be. By 1978, Dan had a working 2D poststack reverse time migration code. Of course it was revolutionary compared to the other methods of the day; so Dan was charged with deploying that new capability to Amoco’s business groups. His groundbreaking work was not reported in the geophysical literature until 1983, when the much cited “Iterative depth migration by backward time propagation” was published in SEG’s Annual Meeting Expanded Abstracts.
While the name changed a little bit, from “backward time” to “reverse time”, RTM has been with us ever since. Dan never lost his love of working on real data and solving real business/seismic imaging challenges. During his career at Amoco, Dan worked on just about every complex imaging problem encountered by the business: from the salt domes of the Gulf coast to the thrust belt of Wyoming to the North Sea to the Canadian foot hills and many more. From the Tulsa Research Center, Dan was a relentless force behind moving migration algorithms into the prestack domain. Dan’s work on time-delay encoding schemes for imaging, described in his 1995 PhD thesis, again illustrated his ability to produce practical solutions to complex problems at hand. Dan combined RTM as well as many other popular imaging algorithms with line and plane-wave sources creating computationally efficient approaches to prestack depth migration before computers enabled prestack migration with individual shot records. After leaving Amoco in 1999, Dan joined Phillips, later ConocoPhillips, where he continued his work on depth migration and velocity analysis, serving as Geoscience Fellow until 2009. He then joined PGS as a principal research geophysicist.
We are fortunate to have worked closely with Dan during our careers. Beyond the incredible breadth and depth of Dan’s theoretical and practical knowledge, what we appreciated the most about working with Dan was his willingness to work with young and experienced scientists alike. There is now a large crop of geophysicists that have enjoyed his mentoring and gentle personality in various phases of their careers.
- Gray, S. H., et al. (2001 Seismic migration problems and solutions, GEOPHYSICS 66(5):1622.