“Don’t be afraid to think outside of the box”

An interview with Gerard Schuster

Coordinated by: Satinder Chopra | Photos courtesy: Penny Colton
Gerard Schuster

Gerard Schuster is a Professor of Geophysics at King Abdullah University of Science and Technology (KAUST) and an adjunct Professor of Geophysics at University of Utah, where he was a professor from 1985 to 2009. He was the founder and director of the Utah Tomography and Modeling / Migration consortium from 1987 to 2009 and has extensive experience in developing innovative migration and inversion methods for both exploration and earthquake seismology. He is now the codirector and founder of the Center for Fluid Modeling and Seismic Imaging at KAUST.

Jerry, as his friends call him, helped pioneer seismic interferometry and its practical applications in applied geophysics, through his active research program and extensive publications, including his book “Seismic Interferometry”. He received a number of teaching and research awards while at University of Utah. He was editor of GEOPHYSICS from 2004-2005 and was awarded SEG’s Virgil Kauffman gold medal in 2010 for his work in seismic interferometry.

Jerry was in Calgary as the 2013 Spring SEG Distinguished Lecturer for the February 2013 CSEG Luncheon, and so the RECORDER requested an interview with him, which he sportingly agreed to. Following are excerpts from the interview.

Jerry, I would like you to speak a little about your educational background and your work experience.

I majored in both physics and chemistry at Portland State University. I switched from physics in graduate school, mainly because of the job opportunities, and received my Master’s degree from the University of Houston. Immediately after that I worked for Arco as a data processor in Dallas and an interpreter in Houston for about 1.6 years. This gave me an opportunity to learn some of the wisdom and practical skills for exploration geophysics, and also fired my thirst to deeply understand imaging and deconvolution algorithms.

Fortunately, I was accepted by the perfect school for me, Columbia University, and studied under a great advisor, John Kuo. He gave me almost complete freedom to fill the gaps in my knowledge, and he also was in charge of a strong geophysical research consortium supported by more than a dozen oil companies. After my Columbia Ph.D. in geophysics, I stayed an additional 1.5 years as a postdoc and then accepted an assistant professorship at the University of Utah in 1985.

I had an extremely rewarding career at the University of Utah from 1985 to 2009, and successfully developed the UTAM industrial consortium; its goal was to develop innovative modeling and imaging algorithms. I very much enjoyed the collegiality at UU, the wonderful support and opportunities for creative research, and the birth of my second family, my current and former graduate students.

In 2009 I accepted a professorship at King Abdullah University of Science and Technology (KAUST) in Saudi Arabia.

How come after getting your Masters from the University of Houston, you decided to join Columbia University for your Ph.D.?

I tried to get into Stanford as my first choice for Ph.D. schools, but was rejected. I then applied to Columbia in geophysics and MIT in electrical engineering. I chose Columbia over MIT when the acceptance letters came in because I wanted to deepen my understanding of the physics of wave propagation rather than do homomorphic deconvolution. Columbia was a great choice – a great city, with a great advisor and fellow students, and I met my beautiful wife at Columbia.

What kind of research work did you do for your Ph.D.?

I developed a generalized Born series approach for the boundary integral equation method. It was a new and interesting twist on BIE methods, but not terribly useful in geophysics. I learned a lot about Green’s theorem and Green’s functions, Morse and Feschbach, numerical analysis, and learned how to clumsily speak the language of functional analysis. Most importantly, I learned to be independent and scientifically confident in myself. I recall that an applied math professor’s first remark at my thesis defense was something like, “Everything you did is mathematically wrong…” I had done my homework and was sure he was wrong. I carefully dissected his argument, and in the end I easily won. Or at least that is the way I like to remember it! He gave me a pass at the end of the defense.

You worked at the University of Utah for more than 20 years. Then how come you decided to move to KAUST?

In 2009, two perfect storms collided, the world financial crisis and my daughter’s impending entrance into an expensive private college, which prompted my move to King Abdullah University of Science and Technology (KAUST) in Saudi Arabia.

Tell us about KAUST. What is the focus of research here?

KAUST is a new graduate school in Saudi Arabia patterned after Cal Tech. It opened its doors in 2009 and only accepts graduate students. The goal is to have about 75% of the student body from outside KSA, and more than 95% of the professors are from outside the Kingdom of Saudi Arabia (KSA) and educated at western universities. A goal is to help KSA transit from an oilbased economy to a knowledge-based economy; they are hoping that KAUST will play the same role as Stanford did with the emergence of Silicon Valley in the 50s and 60s. My fellow professors are quite talented and driven.

Overall, it has been quite an experience at KAUST, and the research support is unprecedented. The result is that the quality of my current students/postdocs and their output is at least equal to the best groups I worked with at UU. UTAM might have ended in 2010, but our new CSIM consortium easily picked up from where UTAM ended, and even more so. The main focus in seismic imaging is on developing improved methods for full waveform inversion, MVA, least squares migration, and interferometry. My research Professor Sherif Hanafy is in charge of our million dollar geophysical lab, and we are doing collaborative seismicresistivity- EM-gravity work with the KAUST water resources group on characterizing wadis for their potential use as water reservoirs. This is a unique collaboration and might have important implications for sustainability in water poor areas.

Looking back now what do you have to say about how your career shaped up?

For someone like myself with modest talents, things turned out much better than I could have hoped. The main reason for success is that I have had so many talented graduate students and postdocs, and they have produced so many interesting results. I sometimes get blamed for their good work, but it is mostly their doing.

What personal and professional vision did you work towards all through your career?

Do good and honest science. I also think it is a good idea to underpromise and over-deliver. I try hard to keep to this philosophy, but don’t always succeed.

What personal qualities do you think helped you achieve all that you have achieved?

Passion, curiosity, and a good work ethic. I am passionate, amazed, and in awe about the wonders of science, especially physics and applied math. I am very curious, I know how to dig deeply when I am very interested in something, and I can sometimes be relentless in getting to the bottom of things.

Please share with us one or two of your most exciting successes.

Understanding how interferometry worked while I was at Stanford in 2000 was very rewarding. I didn’t intuitively understand how it worked until I applied stationary phase analysis to the interferometric imaging equations, and then it all became clear to me. This allowed our group at UU to develop many applications almost immediately. Jianhua Yu was the hero here, he was able to trial many of these ideas on synthetic data and field data as well. He is a field data processing genius, he seems to know and understand all imaging technologies.

Satinder and Gerard

What are your aspirations for the future?

My 5-year contract with KAUST ends June 2014, so I am considering a number of options, including coming back to the states as well as extending the KAUST contract. Stay tuned.

Tell us about your teaching – what courses did you teach at University of Utah over the years?

I taught at least 10 different courses: Advanced seismic inversion I and II, signal and image processing, seismic interferometry, volcanoes and earthquakes, finite-difference modeling methods, exploration seismology, saddle point integration methods, tomography for computer scientists, and seismic interpretation. I came to the UU not really knowing much about the above subjects, but I had a good foundation in classical physics and math, so I could teach myself as I went along. I became a better instructor by accumulating knowledge and adopting active learning principles. I never was a very good teacher for Volcanoes and Earthquakes. I simply couldn’t sustain my enthusiasm past the midterm, and I also ran out of interesting material because I didn’t know enough. But I learned a lot of geology by teaching the course about 5 times, and believe it or not, I liked teaching it most of the time.

How did you get into interferometry?

Around 1996, someone at DOE contacted me to see if I wanted to be a co-PI in a DOE contract, where the original co-PI unexpectedly dropped out. The subject of using 1D autocorrelations for characterizing the 1D subsurface seemed interesting, so I agreed to partner with Fred Followill and Lew Katz on this grant. I played around with the theory, and was able to extend it to higher dimensions so we could image 2D and 3D reflectivity distributions by migrating VSP autocorrelograms; autocorrelation migration might be considered a light-weight precursor to interferometric imaging of cross-correlated data. The turning point for me, however, was listening to the helioseismology daylight imaging talk in 1999 by James Rickett and Jon Claerbout. I was amazed! How could they turn passive seismic data, i.e., noise, into signal? I was hooked, and I had to figure out how the heck they came up with those shot gathers on the sun. For some reason, Jon generously asked me if I wanted to visit for sabbatical in 2000. I went to Stanford for sabbatical in early 2000, and James Rickkett explained to me what he was doing, but I still didn’t get it. I need to understand equations and physics in an intuitive-pictorial manner and I could not understand the intuition behind what they were doing. So, I sat down for a few weeks with pencil and paper, going down this wrong path and that wrong path in deriving analytical formulas. In desperation, I used a stationary phase analysis of the interferometric migration formulas, and it all became clear to me. This was a eureka moment which lead to the simplified Feynmann-like ray diagrams that clearly (to me) explained the basis of seismic interferometry. I strongly suspect that Claerbout had this intuition all along, but did not express it in the Feynmann-like diagrams that made so much sense to me.

Anyway, I generalized his daylight imaging formula so that its use could be extended from passive seismic data to also include deterministic data that we collect in the field. Therefore, I came up with the name “interferometric seismic imaging” and “interferometric migration”, which I used in my 2001 EAGE abstract. I had also used this terminology in my 1999 and 2000 UTAM papers. I was a bit hesitant in coming up with the name “seismic interferometry” out of respect for Claerbout’s beautifully descriptive term “Daylight Imaging” but I had no choice since exploration data is not daylight. As a rank-amateur astronomer, I was familiar with microwave interferometry and thought that interferometry was a better descriptor.

What is its promise?

Interferometry has significant importance in the earthquake community in extracting surface waves from passive seismic data, where any geophone becomes a virtual shot point. They use the surface waves to infer the shear velocity distribution all the way down to the mantle. This helps explain the tectonic history of plates and continents. Exploration seismology has not yet seen an interferometry killer app for surface seismic data, but I believe that some of the recent work in imaging multiples is showing real promise. VSP interferometry definitely has a useful place in seismic imaging, and I believe that it will become a standard processing tool for VSP data. But the VSP seismic processing budget is less than 1% of the surface seismic budget, so the prize is interferometric imaging of surface seismic data.

Figure 1 shows an RTM image of a Gulf of Mexico data set, where the bottom of the salt is not clearly imaged because a time shift statics was not properly identified in the data. In contrast the least squares migration result obtained by imaging the freesurface related multiples shows a much superior image of the salt bottom because it automatically accounts for the source static. This is a teaser, because this is the only data case we have trialed. We have to wait until this method is applied to many other data sets to pronounce victory.

If I said your most important contribution to geophysics is in seismic imaging and the development of interferometry, would I be right, or there is more to it?

Yes, there is more to it. Jianhua Yu was the pioneering developer of deterministic seismic interferometry, and if it wasn’t for him much of the early work would not have emerged. The theory for wave equation traveltime inversion was all Yi Luo. His ideas were often thunderous bolts from the blue, and I went along for the fun ride. So, I am a partner with the many students and postdocs that I worked with over the years, no one contribution can be solely attributed to me. My name pops up all the time because my student/postdocs move on every 4 years or so while I stick around.

You have written and published many, many refereed research papers. You have also written a book on interferometry. Could you share with us your writing experiences?

Know who your audience is, and write in a style and language that they easily comprehend. Same advice for a good talk. Economy of words and equations, intuitive punchline diagrams that explain the method, complicated math in an appendix. This is my ideal, still trying to achieve it. If you think only 4 or so editing iterations are needed for writing a good paper, then you are probably wrong. To paraphrase a famous American writer: “If you really want to be boring, write everything you know.” To honor Mark’s words, I’d better stop here.

What are the directions in which future R & D worldwide is focused in our industry?

I am mostly in academia, so my answers are only educated guesses. Standard seismic research is often prioritized according to where the big money paybacks can be found. Therefore, if your money comes from GOM then wider azimuth, longer offset data are the keys to accurately peering under salt; the associated research goes into finding better velocity models and into designing fancy new experiments to give such data at a cost-effective price. Land acquisition is also following a similar path. If land data and unconventional plays give you the big money, then going beyond VTI imaging seems to be a must, especially with vertical or sub-vertical fractures that need to be honored in designing optimum placement of wells and drill paths. Full wave inversion seems to be enjoying a resurgence, but is still a challenge at deep depths. Attributes (e.g., Marfurt’s group) and extracting diffractions (e.g., Landa, Fomel, et al.) are and will become more useful for unconventional oil+gas land plays, and reservoir analysis. For any large integrated oil company, there is a large financial payback in making microseismic work better in monitoring drilling and well placement for more efficient reservoir operations. As far as I can tell, no killer app yet in interferometry for surface seismic yet. Some people are trying to make it work well for passive reservoir monitoring or hazard analysis (e.g., Joe Dellinger et al at BP).

Tell us about some of the new technology ideas you are experimenting with?

I find imaging of multiples very interesting, which I think will eventually be used as a useful complement to imaging primaries, not just an afterthought. Designing experiments that utilize both multiples and primaries will likely give you more bang for the buck, in some cases. Dongliang Zhang at KAUST and Yike Liu in China (and others) are doing some very interesting work in the area.

Sometimes it is interesting to ask bold new questions. What do you think are the three most important unsolved problems in geophysics?

Your questions get harder, as I get more worn down! My three guesses: 1). Extracting permeability from seismic data has not yet been convincingly demonstrated as far as I know. Reliably determining saturation, porosity, crack density+orientation from seismic data are still challenges. Such parameters are what the reservoir engineers need to optimize production and EOR. 2). Reliable subsalt velocity and images, 3). Optimal+useful ways to combine different types of data such as seismic, gravity, and EM.

In 2005, the SEG celebrated 75 years of its existence. During these 75 years, what particular technical papers do you think had a real impact on the advancements in geophysics?

For seismic, the most important breakthroughs include stacking (Harry Mayne), predictive decon (Robinson, Treitel and MIT group), statics, NMO velocity analysis (Taner and Koehler), Kirchhoff (Schneider et al.), wave equation (Claerbout), and reverse time (Whitmore, Lowenthal et al., McMechan) migration, rock physics (Nur’s group and others), 3D seismic imaging and acquisition (French and many others), anisotropy (Winterstein, Thomsen, Tsvankin and others), traveltime tomography (Gulf group with Langan, Bube, et al.), FWI (Tarantola, Lailly, Mora and others), SRME (Delft and Shell groups), AVO (Ostrander and others), MVA, attributes (Amoco group and Marfurt group), wide azimuth acquisition (BP and other groups).

Could you comment on the news that we heard some time back (Twilight in the desert) that Saudi Arabian oil may dry up sooner than anticipated?

I’m too low and on a different totem pole to shed new light on this issue, which was raised by a banker from the same street that gave us the crash of 2008. The Red Sea is a frontier area that Aramco is starting to explore, and it is also discussing the possibility of exploring for unconventional gas in the Northwest of KSA, and new technology can sometimes surprise even “Twilighters” and bankers.

Yours is a well known name in the industry and with the DL tour you will be able to reach out to many distant places. Any comments on that?

If I can reveal new things to an audience that can eventually improve the effectiveness of their work, then I feel satisfied.

How do you feel about all the travelling that comes with the DL?

I like to travel, but the DL flight schedule is much more than I would choose for a vacation. I look upon it as something I need to do for many reasons, partly paying back the society and the profession that have given me so much. Also, I kind of like an unexpected pat on the back, even while mumbling, “Aw shucks.”

Of the many different awards that you have received, which one is the most dear to you and why?

I like the Kauffman Gold Medal the most. The surprising combination of two heavy medals and a plaque are beautiful. But I also like it the most because it is one of the highest SEG awards for achievement. Until I got it, I didn’t think much about achievement or recognition. Like most scientists and engineers, we do science to have fun and learn deeply about new things.

After having achieved so much, what motivates you now?

Have fun doing a good job and keep moving forward. My latest batch of students, postdocs, and research professor is really, really fun to work with.

What other interests do you have?

Reading world history, military history, science, some archaeology and politics. Motorcycling. Adventure stuff, such as my recent 7-day trek to Annapurna or my December camping experience in Wadi Rum, Jordan, site of the Lawrence of Arabia film.

Do you have any words of advice or inspiration for young people considering a career in Geophysics?

Follow your passions, with some bias to the passions that will support you financially. Don’t be afraid to think outside of the box. I use a rule of thumb, 70% straightforward and interesting stuff to pay the bills and 30% blue-sky stuff to possibly have some real fun. Don’t work 7 days/week in graduate school – you need a day off/week to recharge the batteries. Take at least one refreshing vacation per year. Try to achieve balance in your life: career, family, physical and emotional health.


Share This Interview