Interviews

It is a great choice

An interview with Kris Innanen

Kris Innanen

Kris Innanen works as an Associate Professor at the Department of Geoscience, University of Calgary. After earning a Ph.D. from the University of British Columbia (UBC), Kris worked as an intern at BP Amoco, a postdoctoral fellow at UBC, as an Assistant Professor at University of Houston, and then came to Calgary in 2009.

Being strong in theoretical geophysics, Kris has worked on advancing the concept and algorithmic solutions on problems such as elimination of multiples from seismic data and to locate and determine subsurface structure in the absence of a velocity model. His work has fetched him 9 awards so far which also includes the J. Clarence Karcher Award that he received from the Society of Exploration Geophysicists (SEG) in 2006.

The RECORDER requested Kris for an interview, which he sportingly agreed to. His impressions and opinions on different aspects are contained in the following excerpts from the interview.

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

My earliest and most meaningful experiences with science came from my parents. My father was a professor of astrophysics, and my mother was educated as an astronomer as well. My sister and brother and I grew up surrounded by the most fun and interesting aspects of science; you know, we went on a vacation to Key West in 1986, specifically because it was a good place from which to see Halley’s comet. I could go on, but you get the idea. Science was in the air as I grew up.

What about your schooling?

I did my early university schooling in Toronto, at York University. I knocked around for a while as an undergraduate in the Space Science program, the Earth Science program, and the Physics program. After a few years I settled on Physics & Geophysics and (eventually!) graduated in that program. One of the large number of excellent professors at York, the outstanding John Miller, took me on as an MSc student in the Physics department, where I learned about optical remote sensing. To this day I consider John a key role model – a true gentleman scientist.

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

UBC used to have a great program where they would fly prospective students to Vancouver for a few days. By the time the students had gotten me to Kits beach I was just about sold. What cinched it was the Department of Geophysics and Astronomy, its people in general, and Tad Ulrych particularly. I’m sure you know what I mean, Satinder, the kind of warmth Tad was capable of showing, right away I knew I wanted to spend as much time with him as I possibly could.

But you must also have been very interested in signal analysis to join Tad’s group?

Interested and terrified. In my undergraduate degree I didn’t do well at all in my signal analysis course. Didn’t understand it at all. So going to work with a world champion signal analysis expert was a little nervewracking to say the least, but I guess I figured somehow it would all work out OK. People in their early twenties think like that. Good thing too. People in their forties like me become quite cowardly.

Anyway, in my first semester at UBC I sat in on Tad’s undergraduate signals course, and for whatever reason things seemed a lot clearer. I’ve often tried to think what kind of life lesson that is. Would I recommend to a student that they move 3000km to work on a subject they’re ostensibly terrible at? It worked out well for me but that doesn’t mean it wasn’t a dumb thing to do. I think what it probably means is that who you work with is very important, ranking right up there with what you work on.

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

Tad taught me about seismic Q, and that became the main theme of my thesis research. Tad had a great imaginative vision of how to mix signal processing, inverse theory, and information theory, and I was trying to apply that to Q compensation. I was also learning a tremendous amount from Art Weglein, who by a few years into my degree had become an unofficial co-supervisor for me. I was trying to apply some of the inverse scattering methods his group was developing to Q compensation, and inversion for Q. It all massed together into a sort of patchwork thesis. I still work on Q related problems whenever I can. A great topic – strange and in a lot of ways poorly understood. Best kind of problem.

Satinder and Kris

Tad and Art have quite different technical approaches to seismic problems.

True. It made for a mixed-bag of a thesis, that’s for sure. But overall, hugely positive. You kind of inherit the DNA of your supervisor. Some students in fact wind up taking on the speaking style as well as the research style of their supervisors. So if you have two very different people act as co-supervisors, you get a mixed up DNA inheritance, and that can be really good. Good in helping you find your own unique way of coming at research problems.

You worked at the University of Houston for some time.

Four years in the Department of Physics. What an amazing set of professors there. Art was the reason I had that job. In fact, any job as an academic. My daughters were born pretty soon after I graduated from UBC, so from a purely financial standpoint I wouldn’t have been able to patiently wait through several postdocs for a University slot to come available. But Art spearheaded efforts to have a junior professor position open up, and was incredibly supportive of my application, so I was very fortunate to slip into a faculty position.

Then how come you decided to move to Calgary?

My wife Kate and I had come to think of Western Canada as home, so from that point of view coming to Calgary was not at all a difficult decision, even though of course we now would have yet another set of great friends and colleagues to miss in Houston. Moving around has that side effect. We guessed right, though. Life here is great. Professionally, the biggest motivation was that I knew I’d be able to learn about field and experimental seismology if I came and worked with CREWES. I always admired researchers who were as adept in the field as they were at the blackboard. Someday I would like to be like that. Gary Margrave, Don Lawton, Laura Baird and the incomparable field experts at CREWES like Kevin Hall, Malcolm Bertram and Kevin Bertram will be responsible for it if I ever become less of an amateur.

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

I can’t believe how well it has turned out, considering how unplanned it all was. Probably everybody feels that way.

What personal qualities do you think helped you achieve all that you have achieved? Please share with us one or two of your most exciting successes?

It certainly isn’t raw talent in any particular area. I think my best attribute is that I don’t mind hard work. It also doesn’t bother me when the 53rd thing I try to solve a problem doesn’t work and I have to think of a 54th. If I wind up solving something important one day, I’m guessing it will be for that reason more than anything else.

Tell us about your teaching. What courses have you taught at University of Houston and Calgary over the last 10 years?

Pedagogical techniques and innovation are a big deal in Universities these days, but I can’t say I’ve been too active in that kind of thing, admirable though it is. In lecture courses I like it best when it is just me, a blackboard, and the students. I taught graduate level math methods, and introductory level physics in Houston. Here at the University of Calgary, I teach introductory geophysics and the senior level Geophysics Field School, and each year I alternate graduate courses in seismic inversion and seismic signal processing. The grad courses are fun because they mix students with researchers from downtown Calgary, who can enroll as open studies students even if they’re not doing degrees. The projects coming from this mix of students and researchers are fascinating to me.

How about undergraduate teaching?

We’re particularly proud of the undergraduate Geophysics Field School course. In addition to the non-seismic methods Adam Pidlisecky does a great job teaching, we give the students the chance to lay out several kilometres of seismic lines, and shoot into them with our vibe. The model, which was developed over many years by Don Lawton, makes for a unique opportunity for our undergraduates, who get to do quick velocity and resolution calculations on shot records that are coming hot off the printer in the recording truck. Meantime they are keeping track of the line crew and vibe, and overseeing troubleshooting. It is an intense and memorable experience.

Some years ago, a presentation was made at the SEG titled ‘Is there anything left to be done in AVO?’ How would you answer that question?

There is lots left to be done. I think of AVO in very general terms – to me, it is any quantitative use of changes in the strength of a wave after scattering. The change could be with angle, offset, wavenumber, even frequency. If you take a view that broad, there is an unlimited amount of work left to be done. For instance, if we could reliably measure the AVF signatures (F for frequency) of the tops of low-Q zones, we could in principle do Q compensation within and below the zones with data-driven operators constructed very similarly to multiple prediction operators. Wouldn’t that be something? AVO is also very closely tied to the big theories for seismic inversion, full waveform inversion and inverse scattering, and developing those relationships is important. Lots to do. Even in more standard AVO or AVAZ settings it seems to me there is lots to do, finding robust ways of inverting for the tougher parameters with near-critical angles, and so forth.

Looking at many of your publications, I notice that they are based on hard-core math for the theory, which, though very important, many people would shy away from. Tell us how you took a liking for hard-core math.

The theorist in me likes this aspect of geophysics, because it connects with a grand problem in the physical sciences, which is to understand the qualitative content of the great equations. In our case, the wave equation. I think all seismic people have felt this from time to time: we look at the wealth of phenomena in a seismic data set – refractions, surface waves, dispersion, diffractions, scattering, microscopic phenomena impacting macroscopic attenuation, and feel wonder at the fact that you can write in one line an equation containing all this. Seeing the connections between a simple equation and the results of it acting in complex circumstances, that is an important problem. Studying basic aspects of waves leads in very interesting directions. We’re practical people, geophysicists, but we shouldn’t forget that in seismology we’re not that far from really big ideas. John Wheeler first convinced himself of the plausibility of black holes by thinking about the velocity of seismic waves.

So to get back to your question, I do like a quantitative approach. But I should add that the main objective is to get the information out of the data, by hook or by crook. That is true in all of my papers, even the ones that have a lot of mathematical- looking development. Decoding the data, making tools to do that, those come out of the mathematical approach and that is very satisfying.

I’d better also add a further word in case an actual mathematician hears us talking like this. The math I make use of is very seat of the pants – it is more like an assistant to geophysical intuition. In this approach one feels free to inject an intuitive idea into the development whenever it is needed. It is effective, I believe, but not rigorous.

Kris Innanen

You have written and published many, refereed research papers. Could you share with us some of your writing experiences? Have you ever thought of writing a book?

I haven’t, no. I’ve written course notes and organized them into something sort of book-like, and the difficulty of doing even that makes writing a book kind of daunting. I might consider it in the future, if, in retrospect, it looks like the things I worked on have a common thread to them.

The only thing I would say about writing is that it involves lots of discomfort and frustration. When I’m writing 90% of the time I’m in a state of dissatisfaction. The 10% satisfaction comes at the end if it looks like I’ve explained myself decently. Just enough satisfaction to get going on the next one!

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

There is a lot of interesting recent progress on the acquisition side, with fibre-optics and nodal systems. Determining what the relationships are between the signals we get from fibre and those we get from standard phones, quantitatively, is an important task right now. On the theoretical and numerical side, there are some interesting trends in which the big inversion theories, full waveform inversion and inverse scattering, are being connected and supported by longstanding practical methods, well-control, AVO, etc. It will be nice to see that continue. Multicomponent is developing rapidly right now again, possibly in response to the growth of full wave methods. I was at a multicomponent forum in Beijing last year, and I expected to be presenting to ten people, but there were more than a hundred, all researchers working full-bore on multicomponent problems. So those are some directions.

Are there any areas we should be focusing on we’re not?

In very general terms I see some things I think we should be cautious about, for sure. As someone who is a proponent of maintaining a significant basic research component in any R & D picture. When I say basic research, I am not excluding applied geophysics, by the way. What I’m referring to is research whose direct scientific or industrial application is an unknown number of budget cycles in the future.

Support for the basic stuff is always in a state of fragility. For obvious reasons! It is harder to answer the question “what is the value?” for it than for shorter term research. I guess what I would want to really emphasize, is that a lot of the short-term research projects being carried out today are only achievable because long-term research was done when it was supposed to be.

Let me give you an example. The basic character of full waveform inversion technology, which is now seriously coming online in industrial applications, was set out in papers in the early 1980s, and gradually developed. So now, in 2015, you could easily see a company carrying out an R & D project where, in a year, a full waveform inversion procedure is implemented for, say, velocity model building. It is tempting to say, hey, look at all the value we get from these one-year projects! That’s what we need to focus on! But we need to remember that that was not a one-year project. It was a 30-year project, funded by people and institutions taking a long view.

Budgets are tight today, and that pressure is translated onto research groups whose portfolios include long-term research. What I want to advocate is that we do not stop doing long-term research, and we do not stop funding it! The 1980s did not supply us with an infinite number of new ideas and nascent technologies.

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

A very important seismic exploration and monitoring problem is the full integration of rock physics and seismic inversion. Another good one is the missing wavelengths in inversion. I think the most important problem of all is the “missing physics” problem, where we try to quantify and manage the gap between the completeness of a convenient chosen physics model and one that adequately explains our data. The problem includes defining what “adequate” means. These are not easy questions. In my opinion we don’t even know how to pose many of them properly yet, let alone solve them.

What motivates you to remain charged up in your work?

Being surrounded by students, colleagues and researchers in a research group who are enthusiastic and dedicated. The students at CREWES right now are absolutely amazing. My most lethargic moment is dissipated immediately around people like that. On top of that, having interesting problems to solve. Which we do right now in geophysics. I have no trouble whatsoever keeping charged up!

What other interests do you have?

Anything that involves spending time with my family. We camp and hike and travel as much as possible, see movies and play together as much as possible. Our daughters are ten years old. Geophysics will be around next year, but my 10-year olds won’t be – they’ll be eleven. My wife Kate and I have become very aware of that aspect of life.

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

It is a great choice. You can ask basic scientific questions, and the answers contribute to the creation of new knowledge and technology. Seismic methods are still really the only way to see significant distances into the Earth without drilling, so if you find the idea of determining the structure of this (or any) planet beguiling, geophysics is your ticket.

There is serious and meaningful work to be done by clever and hard working geophysicists after graduation, in academic domains and in industry. Geophysics is going to play a major role in the world’s energy and environmental problems in the next 50 years. Very few disciplines offer as significant a way of engaging with these issues as geophysics does.

End

Share This Interview