You are a geologist

I began my geoscience education in standard fashion, sketching line drawings of outcrop bed geometries from road-cuts, scrubby creek bank-cuts, and the occasional quarry. Eventually the class moved on to exercises in stereographic aerial photography interpretation, again utilizing at its core simple lines on a page to represent geologic interpretation. The act of tracing lines on paper is almost insultingly literal, yet the geologic implications of the lines, the hypothesis that is laid bare, can be a thing of reasoned beauty.

Later I spent several seasons creating geologic maps from disparate pieces of spatial and temporal information, but looking back, I was just practicing how to meaningfully draw the lines. I was learning how to hypothesize within the constraints of the available evidence, guided by classical geologic reasoning. One summer I had a chance to work with 3D seismic data and things have never been the same. The amount of geologic information available in one place absolutely staggered me. I have been a proud seismic interpreter (a.k.a. geologist) ever since.

Philosophy and tips

Seismic reflection interpretation is one method of geologic interpretation. In order to get better at it you need to practice constructing and destructing testable geologic stories or hypotheses. Some may label this approach as ‘model driven’; I suggest geoscientists are required to drive a testable model to be debunked, improved, or expanded. You will need to grasp all the options and weigh their merits before being able to meaningfully assign risk or resource to your geologic assessments. The following are a set of simple tips for crafting an interpretation from seismic reflection data:

• Obtain and integrate with your interpretation every piece of geologic information available.
• Remember, the top of a geologic package must have a bottom.
• Always look for and map packages of geometries, usually called sequences.
• Avoid mapping just one surface at a time.
• As you pick seismic markers, think about what those surfaces represent geologically.
• Is that bright marker you were drawn to a downlap feature, an unconformity, or just a bright marker that you can map on a few lines?
• What are the implications of your current pick for the upcoming strike line? What should you see if you are right with your current model?
• Visualize what the map will look like before hitting the Grid button.
• When working with 2D data, interpret between the lines. What are you predicting to exist between data points?
• If you don’t hold the working hypothesis in the forefront of your mind while interpreting then you are just an expensive autotracker; be a geologist.

The more familiar you become with interpreting while hypothesizing, the quicker you will be at calculating options and understanding as many geologic arguments fitting the data as possible. People may see this insight as a form of blind intuition, when in fact it is an attainable- through-practice science reasoning skill. Seek to habitually sharpen your geologic reasoning skills while interpreting. When done fluently, the results will be a clear and comprehensive understanding of the possible subsurface models, appropriately scaled to the available data quality and coverage.

Never forget to assess the confidence you have in the evidence that is feeding your hypothesis. With new data comes a recalibration of the geologic story, don’t be afraid to change. It is distressing, but sometimes interpretations need to die.

The more familiar you become with interpreting while hypothesizing, the quicker you will be at calculating options and understanding as many geologic arguments fitting the data as possible. People may see this insight as a form of blind intuition, when in fact it is an attainable- through-practice science reasoning skill. Seek to habitually sharpen your geologic reasoning skills while interpreting. When done fluently, the results will be a clear and comprehensive understanding of the possible subsurface models, appropriately scaled to the available data quality and coverage.

Never forget to assess the confidence you have in the evidence that is feeding your hypothesis. With new data comes a recalibration of the geologic story, don’t be afraid to change. It is distressing, but sometimes interpretations need to die.

Q&A:

I like the statement that you make in your article, ‘seismic reflection interpretation is one method of geologic interpretation’. It is very true when you are interpreting with a working geologic hypothesis in mind. Do you think some individuals don’t do this?

There are times when my mind wanders off into other thoughts while I’m tracking some event – moments when I’m not really mindful of the interpretation and I’m being mechanical. All of sudden I’m picking on a diverging event and I’ve forgotten to think why I went up or down because I was thinking about getting the map made and the well tie that I need to check on, and maybe what I wanted for lunch... What I was getting at was the need for us to stay thinking about the implications to exercise focus and stay ‘in the zone’ when you are really conscious of the options and thinking ahead to the next lines or visualizing the structure/depositional element/XXX you are trying to verify or debunk, and by this active mindful process you can hold multiple options in the forefront of your mind, kind of like playing chess, you can see multiple paths and seek to test them each with data. This allows you to understand what critical datasets you could acquire or what specific observations could be used to resolve multiple important questions you have found. It’s about finding the right questions to ask, and understanding what data would resolve them.

Another interesting statement you make is, ‘seek to habitually sharpen your geologic reasoning skills while interpreting’. This would be an ongoing process in that you think of all the possible models that would have a seismic response as is being interpreted. Please elaborate.

I would say you must consider models/solutions that may not have seismic response while you are interpreting seismic responses....if you don’t, it’s probably the fastest way to be excluding viable geologic solutions. I would want to emphasize that somewhere when you move from acquiring/processing into interpreting data, you move into geologic interpretation so you have to practice and improve geologic reasoning skills to be the best interpreter you can be. I don’t believe all interpretations are equal or deserve equal consideration (blasphemy!) – why don’t we just grab somebody off the street to read your X-ray? A good interpreter is not constrained within the data at hand, only by the limits of their reasoning skills.

What do you think are the challenges for a seismic interpreter, considering that seismic data contain massive amounts of information which has to be extracted using the right tools and knowhow?

Geologists long ago, and on to this moment have been developing/ advancing logical ways to make sense of even more massive amounts of data than seismic contains: from all earth process related observations. But I take your point. Seismic data to me is a playground – it’s a chance to see something new. The biggest challenge is not to get complacent and keep being thrilled by the joy of discovery even in the smallest spaces.

An important task that the seismic interpreter is faced with is the prediction about the location of hydrocarbons in the subsurface. Once the geologic model is clear in the mind, how do you think this can be made more convincing and presented to the management?

Be credible. Make good maps; structure maps with impossible faults or sloppy gridding really hurts technical credibility upfront. Seismic profiles with unexplained thickness changes, poor correlation.... Understand the geology of the basin and what events you mapped on mean. Don’t be a sales person – be a scientist, but also allow yourself to be genuinely excited when you find something of interest. If you are correct, it will sell itself.

To be credible, providing multiple hypotheses and outlining their relative merits or detriments to management is key. You want to show observations and how you used them to build the predictions. Show your worst line and most difficult piece of the interpretation along with the cleanest lines/examples. Articulate the key stepping stones on your logic train – bring them along with you so that they can see how you came to your conclusions.

Doing detailed analysis for your interpretation could be a time-consuming exercise. Usually interpreters are faced with completing their work in a limited time, which may not result in favorable results. Do you think some kind of accountability would help ease such a pressure on the minds of the interpreters or it would add to it?

Seems like time pressures are only increasing, and it does take some quiet space and concentration for me to get into my ideal interpretation zone. Part of the answer is to be able to discern quickly the key products you need to focus critical time on, a sort of project triage. I would argue that if you are in good interpretation shape, that is, your reasoning is honed and you have thinking stamina, you can create quality interpretation products that capture the key points more quickly. I was once asked to make a gross depositional element map of the Jurassic east Indonesian Bintuni-Australian NW shelf margin in one week, and tell whether a certain block had a good chance for reservoir development or not, having never worked the area before. Was it a great detailed product, a masterpiece to stand the test of time? No. Was it useful? Yes, the decision makers were able to understand the outcomes and risk.

Would the tips that you mention in your article be valid for unconventional reservoirs also, where the analysis is more demanding?

Fundamentally, unconventional reservoirs are just unusually (for the time being) low permeability reservoirs that need new engineering solutions to extract HC’s from this lower permeability media. Most of the game-changing commercially important types are source rock reservoirs, like Eagle Ford and Bakken. But what makes them work? The right depositional setting, burial history/maturity, the right mineralogy, these are geologic problems. Seismic geometries, responses and attributes will have to relate to geologic realities that help high grade regions – I could safely say this same statement regarding any hydrocarbon play type.

One of the big challenges the seismic interpreters face is that of scalability. The information sought from well log data and cores etc. is on a fine scale, not close to what is interpreted on the scale of the seismic data. What is your take on reconciling the two effectively?

Good outcrops stand in that apparent void. But really all datasets in totality are greater than the sum of the parts. “Reconciled” seems to say that they had a fight and need to forgive one another – I think they can all get along and add dimensions to each other from the start. I was standing at an outcrop in Java where simple grains of volcanic quartz in the sandstone – tiny things – proved that there were volcanos present at a time when many models from geophysical reconstructions though otherwise. It’s about solving problems, sometimes the big picture helps, sometimes the small one, always they need each other to make the best prediction.

Derik, on a lighter note, as someone said, life is not measured by the number of breaths we take but by the moments that take our breath away. Tell us about some of these moments from your professional life?

Some of my favorite geology-related memories come from pre-professional graduate field work and the joy of discovering new outcrops and therefore new data that had not been documented before in the forests of southwest Washington. I also remember the first time I saw and used a 3D seismic volume. It was from the Anadarko basin and it captivated me. I would have worked for free that summer to come in and visualize with that data – it helped me tremendously in understanding geology and training/clarifying my thinking around many outstanding field observations. I have also had an amazing project recently uncovering new plays in Central Borneo, integrating geologic field data and subsurface datasets that has been incredibly rewarding. But what really stands out are the creative and motivated people I have met, the experiences, the simple little joys of discovery I get occasionally when an idea works – these are golden.