Introduction

Chief Editor’s Note

As many of you might be aware, the RECORDER team brought out a special print edition commemorating the 75th anniversary of the CSEG. In one section of this special edition, we asked professionals across the globe about their vision of geophysics for the next 25 years. Our past Chief Editor, Satinder Chopra, coordinated this compilation and made it happen. Due to some unforeseen constraints, we could print only summarized versions of those responses. However, we are now ready to publish the full unabridged commentaries on our online portal. In this edition, we are posting the final installment of our six-part series.

Cheers.
Nilanjan Ganguly

Where do you see geophysics in the next 25 years?

This question was shared with me by Nilanjan Ganguly, Chief Editor of RECORDER, who requested me to invite contributions for the RECORDER special issue commemorating the CSEG’s 75th anniversary this May. It’s truly a significant milestone for the CSEG, one that brings great pride to us as Canadian geophysicists.

I reached out to several prominent figures in our industry, asking them to share their vision for the future of geophysics. While 25 years is a long time to predict, current trends provide valuable insights for making educated forecasts. Given the open-ended nature of the question, some respondents initially expressed concerns about addressing it within the stipulated word limit of 500-1000 words. They were assured, however, that they could use their discretion and flexibility in crafting their responses.

I received responses from eighteen individuals, which have been included in this article. The order of the responses reflects the order in which I received them.

Satinder Chopra

Responses

LAURIE WESTON

PRESIDENT, SOUND QI SOLUTIONS LTD., CALGARY

LAURIE@SOUND-QI.COM

I am fortunate enough to have witnessed the evolution of geophysics over the last 25 years – from rulers, paper, coloured pencils, and early computing to massive computational power enabling previously unimaginable data processes, stunning 3D visualization, and futuristic AI-driven interpretation. Not to mention all the upheaval by external economic, political, and environmental influences causing complete business pivots. What could possibly be next?

There are two avenues to what the next 25 years have in store for us: what I would like to see happen, and what is likely to happen. I have a utopian vision of the ultimate balance between human and machine, both contributing optimal and perfectly proportional power and expertise to a completely integrated and interactive workflow providing nothing but good for the planet and its inhabitants. I described this vision in some detail in Chapter 13 of the recent publication: “Geophysics and the Energy Transition” (Elsevier, 2024).

The crucial words in my vision statement above are: “balance”, “integrated”, and “interactive”. These are also the potentially problematic concepts that may derail my vision to varying degrees. Over the past 25 years, we have seen significant changes in technology and science, but people are the same. People evolve much more slowly than technology, meaning that human needs, motivations, emotions, insecurities and blind spots, failings and follies, resilience and resentment, wonder and imagination are identical to what they were 25 years ago – perhaps hundreds, or even thousands, of years ago – and will not change in the next 25 years.

Regardless of the tools we have at our disposal, most of us are going to perceive them through the lens of our own individual combination of that same list of human attributes, most probably using them to further our own imperfect aspirations. Cynical? Maybe.

Realistically, what is likely to happen is that emphasis will swing too far in the AI direction and then experience a very human backlash that throws the baby out with the bathwater (similar to what happened to the concept of AVO in the 1980s; it overpromised, underdelivered and even the strong defensible aspects of it were in the doghouse for at least 10 years). An ideal machine/human balance will not be approached until the last decade of the next 25 years.

Integration is a concept that no one can fault, yet few actually accomplish. There is sometimes an illusion of integration when images from engineers and images from geoscientists show up in the same presentation slide deck. But true integration at the data level remains elusive. Those ubiquitous human qualities are the reason; the “stay in your lane” mentality that makes people defensive if someone is perceived to encroach on their area of expertise. Even with the best of intentions, genuine integration might not happen until robots are in charge.

Finally, the most likely of my crucial concepts to come to pass is “interactive”. There has been exponential progress in the technology that enables it, but the benefits of this capability will be limited by the lack of integration inclination just mentioned. So, the promise enabled by technology will be restricted by the human components in the equation.

If there is a renaissance of enlightenment that inspires the human race to achieve an age of selfless logical decisions for the common good – or androids take over – there is hope for my utopian vision.

Fingers crossed.

PER AVSETH

CTO AND CO-FOUNDER, DIG TECHNOLOGY, OSLO, NORWAY

PER@DIGITECH.NO

We are living in transitional times. Amidst the chaos, a voice shouts, “Drill baby, drill!”, but beyond the climate debate, the steadily increasing energy demand in a world, driven by a growing population, means that fossil fuels will continue to play an important, yet diminishing role in the global energy mix in the next 20-30 years. This transition to renewable energy will also have a significant impact on the field of geophysics. On top of that comes the digital revolution that is reshaping our industry, as well. So, where are we headed?

Despite the rapid advances in renewable energy (predominantly solar and wind), fossil fuels will play an important role in the energy mix for several decades. The decline in their use may not be driven by demand, but by supply, as many oil provinces around the world are reaching maturity. In Norway, the Norwegian Offshore

Directorate (NOD) projects that oil and gas production will begin to decline before 2030, largely due to the production decline in the “last giant” Johan Sverdrup field, expected to have peaked in late 2024. NOD is urging the industry to act swiftly, hoping to make new discoveries that could delay the imminent decline. A key word in their message is “technology”.

As remaining prospects are getting smaller and more subtle, often with stratigraphic components, the role of geophysics is becoming more important. The use of rock physics, quantitative interpretation and seismic inversion integrated with other geoscience disciplines (basin analysis, sequence stratigraphy) has become crucial in the mapping and derisking of subtle traps. The paradox, however, is that, even with the anticipated decline in oil and gas production, the demand for geophysical technology may actually increase as we try to delay this decline. This also applies to improved oil recovery (IOR), where the use of 4D seismic monitoring will be crucial in mapping and extracting as much moveable hydrocarbons as possible from reservoirs during their later stages of production.

One significant challenge facing the Norwegian oil and gas industry is the imminent retirement wave among technical employees, including geophysicists. Many of these professionals were hired in the 1980s and ‘90s and will soon be retiring from their careers. One might expect a new generation of young employees to step into these roles right out of universities. However, there is a problem: young students are not particularly interested in petroleum-related geoscience anymore. The key reason for this is not the fact that the Norwegian continental shelf is mature, but rather due to the growing political debate and heightened awareness about the environmental impact of oil and gas on the climate.

From 2008 to 2023, I taught Reservoir Geophysics at the Norwegian University of Science and Technology (NTNU) as an adjunct professor and guest lecturer. Last year, the course was terminated due to declining interest among students, compounded by the lack of foreign students, who stopped coming to Norway after the government imposed relatively high tuition fees. So, the big question remains: who will step in to fill the roles of the soon-to-retire geoscientists? And how will this affect the imminent decline of oil and gas production? One could argue that this is good for the renewable energy sector, as talented young students will likely choose to pursue careers in renewable industries, focusing on long-term solutions for future challenges, rather than working in an industry that some consider a “sinking ship”. Many politicians even claim that the oil and gas sector has become a “sleeping pillow,” delaying the transition to renewables.

In my view, however, this transition will take decades, and we will still need young professionals who are willing to work in geophysics, not only for oil and gas, but also for renewable energy applications. Geophysics, after all, is a highly transferable field. What students learn in a course such as “Reservoir Geophysics” is highly applicable to not only oil and gas, but also to geothermal, CCS, and subsurface characterization for offshore wind farms.

However, future geophysicists may not necessarily pursue a Ph.D. in Geophysics. Instead, they will likely possess broader skills in fields like mathematics, computer science, and earth science, as technological solutions become increasingly interdisciplinary and data- and AI-driven. There will also be a stronger focus on cost efficiency, especially in industries with tighter margins than the oil and gas sector. We are already seeing the rapid development of distributed acoustic sensing (DAS) technology, which in many cases can replace more expensive seismic surveys. The seismic industry has seen significant decline over the last 10-15 years, and it is unlikely that we will witness technical breakthroughs that can rival the advances made with broadband and full waveform inversion (FWI) technology.

Perhaps the greatest opportunities in the coming decades will not lie in improving data acquisition, but in how we utilize and analyze the data with machine learning and quantum computing technology. The best-performing machine learning algorithms for geophysical challenges today are those that integrate physics and geoscience domain knowledge into their algorithms/workflows. We are already witnessing how machine learning (ML) and deep learning (DL) are enabling geophysicists to analyze massive datasets more effectively and identify patterns that were previously difficult to discern. And we have likely only seen the beginning of this trend.

To conclude, I think that geophysics will become even more important in the future. We stand at the forefront of addressing some of the pressing global challenges, particularly related to climate, environment, and energy. However, I foresee that the future geophysicist will be a more cross-disciplinary scientist. My own area of expertise, rock physics, may not experience significant growth, but will remain highly relevant, particularly in relation to deep learning inversion algorithms.

While the search for oil and gas may diminish, the demand for geophysical technology will persist, especially as we explore for resources such as rare earth minerals and fresh water – fields where geophysical tools and techniques developed for oil and gas will be adapted and further advanced. The continuing search for oil and gas will require sustainable solutions for CO2 sequestration, which will require good subsurface understanding and effective geophysical monitoring. In this way, the future of oil and gas will likely serve as a springboard toward an increasingly renewable future, and geophysics will be an important part of this transition. We will find ourselves on both sides of the “equation,” navigating both the challenges of the past and the opportunities for a sustainable future.

DOUG SCHMITT

PROFESSOR, PURDUE UNIVERSITY, WEST LAFAYETTE

DOUG.SCHMITT@GMAIL.COM

Before attempting to predict where geophysics will go in the next 25 years, I first reflected on my introduction to the discipline during the energy crisis of the late 1970s and early 1980s. Back then, I interpreted seismic reflection profiles using a prescribed set of expensive color pencils and learned how to hand-pick and hand-contour structural maps. Concepts such as 3D seismic acquisition, 4D monitoring, and computer-based interpretation and quantitative interpretation using full waveform inversion were not even envisaged. As we all know today, there have been incredible advancements in the volumes of data that we can handle, as well as in the technologies that produce high fidelity data. However, I expect many of my colleagues will be better equipped to project the future of applied geophysics in this area. Instead, I will focus on where I think our understanding of complex Earth materials may evolve. This is, of course, biased by my work in scientific drilling and laboratory-based rock physics, and tempered by some thoughts based on what I have seen coming across my desk these last few years in my editorial duties at a geophysical journal. While I am not sure I can predict anything with certainty, I can outline some trends that I anticipate will be important.

Some of my comments are driven by the concern that, in many ways, the field of rock physics, particularly as it is applied to the interpretation of seismic data, has become somewhat stagnant. There are likely many reasons for this. Over the last 15 years, support from both industry and public research for riskier, more innovative research has largely disappeared. Geophysical interpretation software is now loaded with standard (but aging) theoretical approaches to modelling, and there seems to be little incentive to expand past these established methods. Additionally, based on my experiences as an editor at a major geophysical journal, much of the ‘new’ theoretical rock physics research appears to be nothing more than daisy chaining of a number of long-standing theories into increasingly complex equations, requiring an ever-growing number of variables that can never be empirically validated. It seems we are plateauing on the S-curve, and I am not sure we are making as much progress as we should be.

OK, I needed to get that off my chest. But as an optimistic person, I am ready to move past the negativity! That said, there are still incredibly exciting directions for rock physics over the next 25 years, both fundamentally and in terms of practical applications. While I am not going to make specific predictions, I will offer my perspective on where I think new trends will lead us.

First, our ability to better characterize the Earth materials we study has improved remarkably. The promise of these tools to help us gain a more fundamental understanding of fluid flow, seismic wave propagation, electrical conductivity, and (put your own favorite property here) is large. Techniques such as microCT tomography reveal the arrangement of minerals and the complexities of pore spaces at the micron scale. 3D numerical models derived from these scans, allow for prediction of physical properties on small samples. Such tools can now even be used in rapid time-lapse mode to track fluid-motion in partially saturated rocks as well as the initiation of failure. We are also developing refinements to methods that help determine at the same time the mineralogical composition and even alignment of the mineral crystals. Other tools, such as neutron scattering, allow us to probe even smaller pores within the incredibly complicated and heterogeneous ‘shales’ that drive much of hydrocarbon exploration today. Having access to such detailed characterizations will allow us to better understand how pore scale phenomena affect geophysical observations.

Despite these advances, much of our interpretation of geophysical seismic data still assumes that the earth is both isotropic and perfectly elastic (i.e., does not attenuate). These are both areas where significant advances are expected to occur in the coming years.

Let us consider anisotropy first. A propagating seismic wave is influenced by several factors, including the texture of the rock, the scale of layering and structure, and the presence of preferentially oriented fractures. These factors collectively make most rock masses anisotropic. As is well known, anisotropy provides valuable clues that are important to understand subsurface structure and in situ permeability. Future research may rely on combinations of core samples, geophysical logs, and borehole seismic studies, perhaps tied to electrical probes, that will allow for more informed and useful interpretations.

Additionally, experimental validation (or rejection) of existing theories for fracture induced anisotropy, which include fracture interactions, scale, and stress dependence, remain an important area of focus.

Second, it has been well known for decades that attenuation is partially related to fluid motion induced by the passing seismic waves. Several theories (again coming back to my daisy chaining comment above) have been developed to explain this. Over the last two decades, experimental techniques have evolved rapidly to measure the variations in elastic properties with frequency. These methods examine the effects of wave driven fluid oscillations within pores, cracks, and fractures. However, there remains a great deal to accomplish here with regard to making these measurements conform to the strains induced by passing waves, by studying partial saturations and associated capillary effects, including both dynamic (shear) and bulk viscosity of the fluids. Field measurements of attenuation from borehole observations remain complicated, and there is room for advancements that will improve calibration and accuracy.

At this point in history, I am expected to comment on how machine learning (ML) and artificial intelligence (AI) will solve all our problems. Certainly, there are many areas where ML/AI will be incredibly useful. However, I am skeptical of the idea that AI will provide a complete solution – especially if we are content to rely on black box ‘average’ models trained solely on existing knowledge, with the risk that over time, the AI ‘data’ will simply recycle its own previous outputs, without providing novel insights. There is still a great deal we do not know about the properties of Earth materials. They are probably the most complex materials that humans deal with, as we have no control over their formation. As such, it would be shortsighted to stop trying to find and explain new fundamental discoveries. I believe that empirically based geophysics will remain a core component of the future although ML methods may help us find phenomena we did not expect.

One final thought is that, while fundamental understanding is important, the largest challenge will remain in applying this knowledge fruitfully to the interpretation/inversion of geophysical data. We can easily fit any theory to any observation, and so, given that we are constrained by incomplete and band-limited observations, how can this be done to best reveal reality? This issue will undoubtedly keep geophysicists busy for decades, long past 2050!