So why would two experienced geoscientists, specializing in seismic processing and interactive interpretation, who have each published hundreds of articles, conducted numerous short courses, and received countless awards from top geological organizations decide to interrupt their well-earned retirements to write a sequel to a book released 17 years ago?
“To be honest,” said Satinder Chopra, founder and president of SamiGeo Consulting in Calgary, “because we are getting older.”
Of course, “retirement” is relative, as readers will recognize Chopra as active in the pages of this magazine. He is the editor of the Geophysical Corner section of the EXPLORER, and his co-author is a frequent contributor.
The new book in question, “Essentials of Seismic Attributes and Impedance Inversion,” being released this month, which Chopra co-authored with Kurt J. Marfurt, former professor at the Frank and Henrietta Schultz professorship of geophysics at the ConocoPhillips School of Geology and Geophysics at the University of Oklahoma, was written, the authors said, to reintroduce the field of seismic to a world that, to the authors’ minds, needs the reintroduction. The book is published by the Society of Exploration Geophysicists.
This is important to both of them because they believe the success of oil and gas industry exploration and development depends to a large extent, on the seismic method, which in turn is dependent on seismic interpretation.
“This success,” they believe, “is because seismic data provide a detailed image of the subsurface, providing not only an image of the current reflector geometries, but through the geoscientist’s understanding of geologic processes, of the geologic past.
To that point, what kind of seismic features are best mapped by seismic attributes, what kind by machine learning, and what kind by attributes used as input to machine learning?
Passing the Baton
“After having taught continuing education courses for the past 25 years, we realized,” said Chopra, “that many of the younger participants were not familiar with the physical and mathematical bases of many of the more popular seismic attributes.”
Marfurt believes this has resulted in inappropriate assumptions as to what they measure and therefore as to how they can best be used.
“For example, many interpreters incorrectly assume that attributes like coherence and curvature measure fractures, where in fact they measure strain, although with appropriate calibration to well logs and production data they may be used as a fracture proxy,” he said.
Students entering the field have an even steeper learning curve, said Chopra.
“Their interpretation workstation software presents them many dozens of attributes, usually (for reasons of propriety) with highly limited documentation.”
He said much of this work is now dated and has been superseded by more effective attributes and interpretation workflows.
“And that includes ours.”
Chopra said adding to these challenges is the introduction of machine learning in seismic interpretation.
The two, who have read and digested most of the published literature on seismic attributes, say their journeys through the decades have helped many of the questions that are out there.
“Common to most scientists,” said Marfurt, “as they approach or pass the ends of their careers, we wanted to pass along effective workflows and best practices we have learned from software developers, seismic processors, structural geologists, stratigraphers and skilled interpreters to the next generation.”
Successful application of seismic attributes requires an understanding of not only their physical basis but also an understanding of seismic data quality and geologic processes. There are numerous obstacles to studying seismic potential, including the incorrect loading and scaling of the original seismic data.
But that’s not all.
“There is also,” said Chopra, “the more subtle pitfalls of interpreting noise patterns and attribute artifacts in the image as geologic features of interest.”
Longtime, Long-Distance Collaborators
While the two have never worked for the same employer, much less lived in the same country, they have been long-time collaborators. Marfurt, in addition to his work in academia, has written software for cutting-edge technologies, while Chopra, being in the service industry catering to the oil and gas sector, had access to lots of seismic and well data. Together their areas of expertise would complement each other, with Chopra providing feedback on the software usage that Marfurt authored.
“We worked ‘distance’ before distance was cool,” laughed Chopra, “using something called a land line, snail mail and faxes.”
One of Chopra’s early tasks when he first joined CTC Pulsonic, which is now part of the Core Laboratories Group, was to generate coherence cube volumes from seismic data.
“I was told that if he had questions about the algorithms and their parameterization, to call Marfurt, for all the software that I was using was written by him.”
This started off phone conversations, which Chopra said Marfurt very patiently explained and answered all the questions. This led to joint publications and presentations.
When Marfurt moved to the University of Houston, he started a seismic attribute consortium funded by the oil and gas and services companies. By this time, Chopra had moved to Arcis Corporation in Calgary, which became a sponsor of the consortium and got access to the software. Through the years and their career moves their working relationship continued, and the fruits of that collaboration are regularly featured in the EXPLORER’s Geophysical Corner.
After collaborating on that first book on seismic attributes in 2007, “Seismic Attributes for Prospect ID and Reservoir Characterization,” which was jointly published by the SEG and EAGE, both authors made several false starts to begin work on the sequel, but careers, as they are wont to do, got in the way.
Finally, in 2020, Chopra started work on this latest effort.
“Marfurt procrastinated,” Chopra said, smiling, “remaining busy with his teaching, research and traveling responsibilities until he retired in 2021. At that point, his excuses evaporated.”
The pace of their effort then picked up, culminating in the current effort, which, even though it looks the part, is not the end of the story.
Actually, it’s about midway through the story.
“During the past five years,” they write in the preface, “we have worked hard in collating information on the implementation and use of different seismic attributes in a manner that is easy to grasp. We are still not there. Consequently, we have decided to publish the first half of our work on seismic attributes in this book, which runs close to 800 pages. The remaining half of our collated work on seismic attributes will be completed and published soon.”
Clearly, their retirement plans will have to wait.