Robert Vestrum works as Chief Geophysicist at Thrust Belt Imaging, Calgary and is known for his work on elastic anisotropy. Following three years of work in complex-structure imaging at Shell Canada, Rob went on to the University of Calgary in 1997 to begin Ph.D. studies and to work on the problem of imaging below dipping anisotropic strata.
He worked with Kelman on developing their 2D anisotropic depth migration and model-building software before moving on to Veritas GeoServices to pursue 3D anisotropic depth migration. Rob has processed seismic data from the foothills of the Canadian Rockies through time and depth migration, investigated migration algorithms for both depth and time, and developed anisotropic depth migration workflow and velocity-model-building methodology.
Rob's work on anisotropic depth migration was recognized with the CSEG Best Paper, Honorable Mention for Best Paper in Geophysics, two awards for papers at the CSEG convention, and the SEG's J. Clarence Karcher Award for significant contributions to science by a young Geophysicist. His thesis on elastic anisotropy also won the best M.Sc. thesis in Geology and Geophysics for his graduating year. It was a pleasure engaging Rob in a chat and following are the excerpts from that meeting.
Rob, let's begin by asking you to tell us about your educational qualifications and your work experience.
Well, I started my interest in geophysics when I went to Mount Royal College here in Calgary for my first undergraduate year. I was not sure what I was going to do. I thought about forestry or something like that and Wayne Haglund, Tor Haglund's dad, who was a geology instructor at Mount Royal, talked me into thinking about geophysics. Since I was a math and physics enthusiast, I went to U of A to do my undergraduate degree there. That worked out well, and I followed it up with a Masters Degree at U of C in '94. My first job was at Shell Canada, where I worked for three years in foothills processing before going back to school again to do a Ph.D. in '97. I didn't finish until in 2003 – it was a bit of an epic journey. I worked at Kelman, who helped me get through school, and then moved on to Veritas.
And then – uncertain about how and what direction my career was going – I was looking at different options, oil companies, universities, and Jon Gittins suggested that big oil or government burocracy wasn't going to be for me, and we should start our own company. It appealed to my vanity, I suppose, and I said, "Okay, we will give it a try," and it has worked out really well.
So because you were always interested in science that's why you decided to pursue a career in geoscience?
Yeah, yeah, I'd say so. I used to think that I wanted a job that involved the mountains somehow, which is why I started thinking about forestry. I was really fortunate to get off on this path where I could combine my interests in math and physics with the anisotropy and my interests in the mountains: my geophysical destiny.
Tell us, for the benefit of some of the curious readers, how did you decide to come to the University of Calgary for the Masters and Ph.D.?
Well, University of Calgary is a good school, no doubt about it, and we are closely tied to the Industry here and with a strong research reputation. What attracted me, though, were the people, the individuals. I came initially for the Master's Degree because Doug Schmitt at the University of Alberta had infected me with the seismic anisotropy bug during my undergraduate thesis. Jim Brown and Scott Cheadle had a very interesting paper, which explained a lot of the fundamentals of the math and physics behind orthorhombic velocity symmetry. I thought, wow, these guys really know what they are doing, so I came to U of C to work with Jim Brown. When I was thinking about going back for the Ph.D., it was Don Lawton and the Foothills Research Project that attracted me. It was the place where I could combine anisotropy with the mountains. It seemed like a dream come true. So, in both cases it was the draw of the Professor.
Very nice. Tell us, who were some of your mentors?
Doug Schmitt at the University of Alberta got me into anisotropy. Jim Brown helped me to work independently because he was a rather handsoff supervisor – much to the dismay of some of his Grad Students, who called him "the elusive Doctor Brown" – but that style worked out well for me. Jim was really fussy about the writing, so I learned to write under Jim Brown's tutelage, which was great. Then Kai Muenser was my depth-imaging mentor at Shell Canada, who got me keen about depth imaging in the Foothills.
Brian Link at Kelman helped me adjust to – and learn to love – the service side of the Industry. Brian has the skill of listening to people's problems and really understanding, empathizing with their problem and figuring out how to solve it. Still working on those skills.
Okay, so in '94 you joined Shell as you said, then went on to Kelman in '97, to Veritas in 2001, and then you started your own company Thrust Belt Imaging (TBI). Tell us about each of these moves and how they affected your personal development? Every company has a different culture so tell us about that.
Right. Well, it was funny, I loved Shell and it was a great company to work for and but after a few years I definitely wanted more, I wanted more research type work and those types of things. I enjoyed working the Foothills and the Foothills Team and field trips and stuff like that, but I definitely wanted to do some research, so I started talking to Don Lawton about going back to school. My wife and I had a few children and we had to try to figure out how to make ends meet, so I shopped around town looking for some work to support the research and make a living. Brian Link and Ron Schmid already identified an anisotropic depth problem on the Benjamin Creek Data Set. They felt they couldn't converge to a single velocity model without correcting for anisotropy, so I worked there on contract, programming the anisotropic depth migration and upgrading the model builder to create TTI models.
By the time I left Kelman, I was manager of R&D, but the direction of the company was towards marine processing, to get into the Gulf of Mexico market, which was not the direction my career was heading. At that time, Jon Gittins had already made a lot of headway in getting Veritas into the Foothills for time imaging and they wanted to grow their depth imaging team. They had a team of four people – they were building it up – and they asked me to come and help move that forward. So I did and it was great. They had some momentum and I enjoyed working with the research team there. After a number of years, I found again that the Gulf of Mexico was hanging over my head. A big global company needs to focus on the big markets in the North Sea and Gulf of Mexico, so I was told that we couldn't have software-development resources dedicated to foothills depth migration.
After history repeating itself at a second service company, I didn't know really where to go, which is when Jon said that the only way that this is ever going to work for me is if we can have control over our own destiny, so let's do our own company thing. I thought it was crazy, to be perfectly frank about it, but it turned out to be quite the exciting opportunity.
Very nice. All right, so tell us about your personal qualities that you think enabled you to achieve the professional status that you enjoy today. Was it just hard work or was it ambition or anything else, and do you also think that a firm grounding in mathematical concepts, your zeal for physics and mathematics, did help you and kept you in good stead?
Definitely, my enthusiasm for math and physics has been key to moving my career forward. I find the expression "hard work" kind of funny because I don't consider myself to be a super hard working guy. I work long hours and I am enthusiastic about making things happen, but a lot of people have the discipline to work hard at things that they maybe don't enjoy fully and I don't have that discipline. I admire that in other people. For me I have to love what I am doing and so I am constantly looking for something that I love to do. When I get caught up in it, I lose track of time, I lose track of what I am doing and I am focused on solving a problem or at least digging further into it.
So I am following my passion and definitely focusing on professional activities that I enjoy has made it easy for me to work hard.
Around town, once people have gained enough experience, some of our colleagues they like to go out on their own, they start their own consultancy or these kinds of things, so I wanted to get your views on this. Do you think it was a good decision? Of course you mentioned that it worked out very well for you, but tell us a bit more about it.
Well there are some plusses and minuses definitely. I really miss being part of a big organization with you lots of resources, with different experts doing different things. One of the things I enjoyed doing was being in an organization where I didn't have a lot of authority. I enjoyed employing guerrilla guerrilla tactics to make change. I liked the challenge of going against convention and fighting for change. It was a great opportunity to refine my ideas and to develop influencing.
One thing I miss about being in a large organization is being part of a larger team of technical colleagues to share ideas and challenge each other. How we've worked to overcome that is by working closely with other companies. We have a strategic alliance with WesternGeco, for example. I was teaching three/four times a year in their Depth Imaging School down in Houston, which involved a lot of interaction with other depth imaging experts, which was really fun.
Companies are getting so specialized now, which gives further opportunities to collaborate. You look at Techco Geophysical, they've got the MASTT statics, but also their velocity analysis software and they have been great to work with, doing customization of their software to meet our needs. The same goes for dGB, the developers of OpendTect, who are very open to input on their depth-migration velocity model builder. On the algorithm side, there is Tsunami, with their reverse time migration, they are working with Acceleware, and I have had interesting interactions with both companies as they have sought my input on TTI anisotropy. Anyway, there are all these specialty companies that are developing algorithms, moving technology forward, and I feel like they satisfy my need to be in a larger technical community.
Now as far as the plusses, I am definitely enjoying our autonomy.
Tell us about Rob the person, your habits, your likes, your dislikes, and anything else you would like to share with us.
Well, I don't know, I am a golf enthusiast – a fun game – maybe I get a little bit get carried away at times. I'm also a keen gadget guy. I love my Mac computers and my iPhone. My teenaged sons bought and hacked the first iPhone for Christmas for me the year before the iPhone came to Canada.
Relating to my family, we've got the six kids as you know, and they are a lot of fun. They take a lot of time, a lot of energy and, really, it has been nice. I acknowledged my wife and kids in both M.Sc. and Ph.D. thesis acknowledgements, referring to them as "stress relieving distractions." Right, because when you have a family, you have no choice, they require a certain amount of focused energy. They provide a chance to kind of think about something different for a while and then I can go back and have a fresh view on my work problem tomorrow.
Oh, that works very well. That's absolutely good. How about likes and dislikes?
I guess I like social people, I like interacting with people. I find that kind of snobby or snooty people or people who are into the hierarchy of things, I find that a bit frustrating to deal with some times. My pet peeve I guess.
That's a good example. Okay, tell us for the record, your contribution to seismic imaging and I believe that is what is close to your heart and you have also mentioned it so many times.
Yeah, well the thing with my contribution is in raising awareness of the anisotropy problem in the foothills. Sure, I started out writing codes for doing anisotropic depth migration and correcting for dipping anisotropy and all that, but there are a lot of good programmers that have done similar work on that and likely much better than I did. I think why I received recognition for my work in anisotropic depth migration in the Foothills is not so much that I did the earliest coding or developed the most sophisticated algorithm, but that I went out into the industry, learned how to understand the problem, developed an overall workflow to try to solve the problem, and them put it into practice.
Those days at Kelman were exciting. I started off writing code, then putting it in production, then training the staff, listening to the problems of the Foothills interpreters, and then further refining the process. Learning how to make a practical difference with my research was probably where my biggest contribution came from.
Very good. Okay, so what has been the most memorable moment in your professional life and also tell us about the successful landmarks in your career?
I have a lot of memorable moments but one of the most exciting, which is kind of funny, came at a turning point in my career. Attitudes towards anisotropic depth migration were changing, as people started thinking, "Okay, it's not just a crazy idea, it might actually help reduce exploration risk."
I received a call from Scott Jamieson, who was on the CSEG Convention committee the year they had the Roman theme. They had this fundamentals day with key note speakers and so I got a call from the always gracious Scott, who explains that they are planning this keynote session and they are bringing industry experts from around the world to talk on the specific topic and it is going to be a plenary session, with everybody there, and the speakers get an hour to develop the theoretical background of the topic in question. Scott is going on about how important this is as a corner stone of that year's conference and what the key people are going to have. This was pretty early in my career and the whole thing was going to my head, when Scott breaks into my selfcongratulatory train of thought by adding, "And well, we had a cancellation, so I am not sure if you can help us out?" I am just thankful that he was calling to ask me to speak and not to ask if I could suggest a speaker. But the story got more interesting from there.
So we get to the keynote session, and Brian Russell is the first speaker, and he goes into his talk and as the questions for Brian start to wind down, Larry Lines stands up and says, "I don't have a question, I just wanted to tell everybody that it's Brian Russell's birthday today." And you know Larry is a beautiful singer, so he says, "I would like everyone to join me in a rousing rendition of 'Happy Birthday' for Brian." So everyone laughs their heads off, and of course Larry launches his beautiful singing voice and the huge crowd sings along with gusto. Brian, of course, blushes and the whole thing is so funny. I had the good fortune of having the very next talk slot. Well, of course the audience is in such a great mood that any little joke I made resulted in raucous laughter, because they've already been laughing so hard and in such a good mood after this Brian Russell birthday moment.
I love making technical presentations, but nothing could ever compare to that experience.
So Rob, what are your aspirations for the future now?
I would like to see our little company grow for sure. We are not sure what direction we want to take with that. It would be nice to have some remote offices in other parts of the world. With so much work in South America, it would be nice to have an office in Bogotá.
In terms of my personal career, I would like to see us, as we grow, do a lot more of our own software development. We have done some things, we have our own 2D anisotropic depth migration, we have a few different methods for calculating travel times through anisotropic material, and we have our own interpretation software for building TTI models. It is pretty hands-on, almost like a paint program to draw whatever velocity-model interpretation the subsurface model requires. I would like to see a lot more development. I miss the research parts of my past career. That development may take the form of working with a company like Tsunami or Acceleware, or it may take the form of us increasing inhouse development, but I definitely want to plug in more with that.
Good, so tell us where you see research in seismic imaging being pursued?
Right. Where is the interesting thing because you could think of it in terms of – is it academia, is it Industry, is it –
Exactly. It's funny, it seems like a good mix these days between academics and Industry. Sven Treitel, years ago when he came to Calgary, expressed his concerns about the nature of industry research, as oilcompany research was in decline and research consortiums at Universities were trying to fill that gap.
I remember I had interviewed him and in that also I had asked that question –
Okay, yes, if you think back, it was true at the time. And I think now as time goes on, the big service companies are, and even the mid-size service companies, are able to have their own software development teams. With computing hardware being so much cheaper now, it's a lot easier for a mid-size or small company or custom software companies to be able to do some interesting research. And I think we are going to see a lot more of that coming out of the service companies.
Rob I was interested if you could tell our readers where these research centres are? Is it in the Universities? Where is this research more focused then the rest? I am sure you have –
Yeah, I mean in terms of the Foothills and that sort of stuff I think you know the University of Calgary still has very strong research in seismic imaging and imaging development, algorithms. A lot of it is in Houston in terms of both the academics and in terms of the Industry – Houston is still a big centre but it is mostly focused on sub-salt imaging and marine problems. So in terms of land seismic imaging, I think we are still seeing Calgary as a strong centre for that both in the service companies that are here and the University.
How about University of Alberta?
The University of Alberta is probably less focused on imaging – they have strengths in signal processing and rock physics. Multi-domain interpolation, seismic signal analysis and the rock physics stuff are definitely a huge thing. But in terms of seismic imaging, maybe I am just unaware of it ...
One thing that comes to mind is that least squared migration that came out of U of A.
Oh, right, sure.
Outside of North America would it be the Delft Consortium, are they doing anything, because that's again, mostly attenuation...
Right. There's also Israel – not only some of the research at Paradigm but also Geomage, you know, that Company that does the multi-focused imaging of seismic events. I think that is an interesting method and, like many methods in geophysics, it's a bit dangerous to let it loose on its own, but I think with the right hand at the helm and the right constraints on the problem I think that shows a lot of potential.
Could be useful. In 2002 you received the prestigious SEG Clarence Karcher Award for significant contributions to geosciences by a young geophysicist. How did it feel at the time?
It was really, really exciting. I remember when they announced it at the CSEG Luncheon, Sam Grey said, "No one ever told me that Rob Vestrum was a young geophysicist!" Anyway, it was nice to have all that attention and such substantial recognition for all the work. Like you say there was no real cataclysmic event in terms of my research. It was a slow progression in the algorithm and the workflow.
Culminating into something useful.
Culminating into something useful, and then to have this sudden recognition, made it feel like, now this was a significant accomplishment.
Especially because of the fact that others hadn't done that and you came up at the time and demonstrated it. I think that paved the way for more development in that area for others to catch up.
It was a lot of work and I felt that I made a contribution, so to have my work acknowledged in such an international way was really exciting.
We have some finest geoscience programs in the world here in Canada, yet we rarely see young Canadian geophysicists recognised for their work in this way. Why is that?
You know, I wonder if there could be two problems, and I think that they are both things we could fix if it's either or both. One of them is we need to give opportunities for our young people to make a difference. We've got high caliber young geophysicists and so maybe they need more opportunities to make a contribution.
The second one is that maybe contributions are not acknowledged enough. We are pretty humble people as Canadians and maybe we need to be more ready to shout the world that we've got awesome young geophysicists here.
Yeah,—it's funny, when I thanked Larry Lines and Don Lawton for nominating me for the award Larry said, "No, don't thank us, we didn't do it for you, whenever you win an award it makes us look good." He was kidding around, humbly deflecting my gratitude, but there is a little truth there and that's maybe part of the hesitation. Maybe we are shy about saying we've got the best students in the world. We have some great young talent and we should tell the world.
I believe that. All right, tell us about other awards you have won.
I have won scholarships and a few honours along the way. Probably the most exiting academic award was the John Kendall Doctoral Thesis Award for top Ph.D. Thesis in the Faculty of Science. That was really exciting. I previously had no idea there were so many departments in the Faculty of Science so I was thrilled to score one for our department.
I put in this next question just for the benefit of the readers so that they gain some clarification. You know that seismic imaging methods fall into two types – the integral methods where we have the Kirchhoff type algorithms, and the wave equation base methods. The integral methods are very easy to understand, and developers seem to master them quickly. The wave-field continuation methods yield more accurate images of the complex subsurface structures. I would request you to tell us from the most practical standpoint which method is suited for which purpose.
Right, your Kirchhoff approach and even, to extend it further, Gaussian Beam which is kind of a hybrid but really still an integral method in a way, they have theoretical limitations.
From a practical standpoint, the interesting thing is that Kirchhoff has had decades of really smart people working to overcome the various limitations in traveltime calculations, antialiasing, wave-front healing and all these types of Kirchhoff migration limitations. Those decades of refinement now make it difficult for these new methods to emerge, yet from a theoretical standpoint, these new methods are so much better because they can handle so many more wave-propagation effects, and we can get a more accurate picture of the subsurface and improved amplitude distribution.
In terms of the practical application I think that's where we are going to see the breakthroughs now. They may be incremental improvements, but they are going to look like breakthroughs because they are going to get more practical.
It looks like you have answered the next question partly – what are the challenges that face us in the world of 3D imaging?
In terms of 3D imaging a big factor is just illumination of the sub-surface, so better resolution and wave-field sampling is a big challenge. Once we see a lot more geophones being laid out, once we can more cost effectively put more receivers on the ground, that's going to be a a big improvement.
In terms of using multi-component data, even with the Foothills, John Behr presented a paper a few years ago at the CSEG showing several factors that resulted in improved imaging of just the P-wave data using multi-component results. One factor is that the emerging energy at the surface can often be far away from the vertical because you've got very high velocity on the surface so when we only take the vertical component, we are losing a lot of signal.
All right. It is generally said you have done a lot of work and demonstrated anisotropic migration should be done all the time for better results. Now the challenge is not to run anisotropic migration codes, it is rather to estimate the parameters that go into the migration. Is this correct?
I would like to say that we have given up on inverting for the anisotropic depth-migration velocity model. It can't be done. So what we do, we treat the imaging problem as an inversemodeling exercise. Our first depth-migration velocity model is anisotropic because if we ignore the anisotropy, we are going to have to use much higher velocity than the true vertical velocity to flatten the image gathers and get a reasonable image. So then if we introduce anisotropy later, we would need to decrease velocity and start over again with the model analysis.
So, we start off with an anisotropic model, we have an interpretation from the structural geologist for the major velocity boundaries and the interpretation of the TTI model dips. Then we basically layer-strip down from the surface with the velocity analysis. If we find along the way that the wells do not tie to the correlated reflector depth, then we can look at tweaking the anisotropic parameters to refine the depth tie.
Okay, any other areas in geophysics that fascinate you in particular but you have not yet had the chance to work on?
Well, I must say AVO versus azimuth fascinates me. Maybe the grass is greener on the other side of the fence. It looks to me like it's so powerful and elegant in its simplicity. The AVO response gives you ratio between P and shear, and then the azimuthal variation of the AVO response gives you the directional orientation of your sheer wave anisotropy. From just the P-wave reflectivity, they estimate fracture orientation and relative changes in fracture density. Whenever I see an opportunity to watch a case history on AVO vs. azimuth, I definitely jump on it.
All right. What directions do you see R & D worldwide going in the future, and can we expect any major advances such as we had with the introduction of 3D seismic?
Well, one could argue that 3D was an incremental improvement, with the breakthrough coming when we developed the hardware capacity to record many more receiver channels. I think we are going to see breakthroughs coming again on the computational side, with improvements in visualization and in GPU cluster computing. Many advanced algorithms will become practical and we will continue to develop new ways to interact with older approaches. Again, reverse time migration is one of the oldest migration algorithms we have, and it is elegant in its simplicity, but it has taken us this long to get the computing power to make it work in a practical sense.
Visualization is such a big factor in making good decisions about seismic imaging. We are still seeing improvements. When we walked in here and I saw your view of the all the big displays in your boardroom, I thought, yeah, that's just going to keep getting better.
But now, you know these displays you have in this room probably have three megapixels each, so you have six or seven million pixels right here, which is double the resolution that we used to have on those big video walls, and that is only going to get better.
Rob you have presented and published many research papers. I would like to know your perception about writing; is it sharing of information, is it sharing of experience, knowledge, or something else and how can we encourage technical communication amongst the CSEG members?
Well, writing is a lot of work and where the real pay-off comes in, each step of the way, everything you have to do to have to write: sit down, make a plan for what you are going to write about, and get your figures and organize your thoughts. Well all that stuff is building a logical sequence in your own head. So for me it helps sort out what I am thinking about. Before I can explain anything to somebody else, I need to better understand it myself.
Put your ducks in a row.
Exactly and that is hugely valuable. Then you have to write, and while you are doing the writing, the ducks you thought were in a row, well, they are not, so you go back and reorganize and keep on keeping on.
So then you are done with the writing and you have a solid draft. Your coauthors are on board. And then there is the review process. In the early days I dreaded the review process. One of the reviewers says she rejects it, one of the reviewers says it need major revision, one of the reviewers says it has some potential. Okay. I submitted a paper in 1998 that received reviews like that. I did not think the paper was as bad as all that, but I will admit that the revisions made the paper much better. After the required changes, it was published in 1999 and I received Honorable Mention for Best Paper in GEOPHYSICS.
Now, I find the review process to be a very practical part of my writing process, not only in terms of making a better final paper, but also it helps me solidify my arguments and to see the problem from someone else's perspective. Improvements from reviewer comments have also trickled down to improve slides for my talks and even how I would draw waves on the back of a napkin in a pub conversation about imaging.
That is important. I think you just mentioned some of the valued points here – very good. Rob, tell us about your other interests, besides the science that you practice.
Well you know I am pretty caught up with my family, and I love TV. We have a little home theater in our basement and I cuddle up with my girls on the couch and we watch funny shows or whatever and it's kind of a family thing. I don't have a lot of time to spend watching TV, but we have a PVR, so we can watch our favorite shows when we can. So that's pretty cool.
I have learned to share my golf enthusiasm with my wife, Tammy, and our kids so two of my daughters and my son, RJ, are also golf enthusiasts.
I am pretty keen about photography, especially photos of my kids and the mountains….and my kids in the mountains!
If it weren't for business travel, I likely would not travel. I like plugging in with local scientists and local offices and I see how they do things and I learn different perspectives, different attitudes towards solving problems. I really get into that and I find myself doing a lot of research on kind of the local countries and the local geological setting as well as the history and culture of the region.
Rob what would be your message to young entrants who have just taken geophysics as a profession?
Well, you know a huge lever to success in our industry, which you will hear from many people, is networking. Networking is such a big factor in career development and we have so many opportunities to network, whether at an industry function, or walking along the Plus 15 on a cold day.
You have your area of interest and I have mine. We found that somewhere along the way that we have kind of fallen into our geophysics destiny. The the way to find one's success is to keep your head up and look around and while you are meeting with people, talking with people, you may find that you may prefer some involvement with acquisition or instrument design or you have a skill that is just waiting for a problem to solve. We have such a diverse community, not only the people, but also the types of jobs. There are so many types of geophysics jobs that it's crazy. To network would be the only way to really see that cross section of our society. Along the way, you will make some good friends, you will have people you can call on when you have a particular problem, and people will call you and give you an opportunity to make a difference in our little community.
Very good. One last question Rob, do you think I missed out on anything that you expected me to ask and I didn't?
No, is the short answer!
Thank you very much for giving us this opportunity to sit down and chat with you and I am sure the members will benefit from all the information you have given us and so I thank you for your generosity.
Well thank you – it was a lot of fun!