In a famous exchange from the movie “The Graduate,” Mr. McGuire says to the young Benjamin Braddock, “I want to say one word to you. Just one word:
“Plastics.”
If Ben Braddock happened to be a geology student today, Mr. McGuire would no doubt say two words:
“Data science.”
It is almost astounding how much a knowledge of computing and data applications can strengthen the geoscientist’s toolbox. The good news is, no one has to become a data expert.
But a little software coding experience can help.
Zane Jobe is research professor in the Department of Geology and Geological Engineering at the Colorado School of Mines and director of the school’s Chevron Center of Research Excellence. Jobe said data skills were not emphasized when he was taking graduate geoscience courses – a handicap after he entered the oil industry to work for Shell.
“As a result, there were just a lot of things I couldn’t do or didn’t know how to do, or just didn’t have the tools to do,” he recalled.
Developing data expertise and learning how to code “made me a much better geologist,” Jobe said.
A Deluge of Data
Adding computing and data skills to geoscience expertise can create a powerful combination, with both areas contributing to successful outcomes. Communication between geologists and data scientists is key.
“Historically, indeed, geology has been qualitative and descriptive, which is great,” said Jef Caers, professor of geological sciences at Stanford University and faculty affiliate in the school’s Institute for Human-Centered Artificial Intelligence.
“More and more, we are facing a deluge of data. The ability to process huge amounts of data, that’s where machines have to come in. However, that doesn’t exclude the expert,” he said.
“Humans sense through five senses,” noted Matthew Bauer, vice president of geoscience and spatial analytics for Sabata Energy Consultants in Golden, Colo.
“When you start increasing the dimensionality of data sets, beyond five, it gets hard for us to wrap our heads around. That’s where machine learning comes in,” he said.
Learning the Language
In addition to coping with the vast quantities of data now involved in geoscience, data skills can help the geologist understand the value – or lack of value – in input from data scientists.
All too often, Jobe said, a new computing expert or data specialist without prior experience in oil and gas will enter the industry and issue some innovative, sweeping recommendation. And it can turn out to be “mostly nonsense, because it’s not rooted in the business world or the geoscience realm,” he said.
“There are no magical solutions. Knowing what the limitations are of what’s being presented to you and having a healthy skepticism is still very important,” Caers observed.
So it’s increasingly important for geoscientists to know what data scientists are saying, to understand the math behind data concepts, to learn the relevant tools – to “be able to talk the talk,” Jobe said.
“Find something you enjoy and pick up data skills. It doesn’t have to be geoscience. If you’re interested in it, you’re going to do it. To summarize, find something you’re passionate about and learn to code with that,” he suggested.
Then keep practicing your data skills, because a programming language “is a language. It’s like Spanish or Mandarin or anything else. You have to practice it or it’s not going to stick,” Jobe noted.
Teach Yourself to Code
Data specialists emphasize self-education as a start for the geoscientist who wants to learn or improve computing and data skills. Fortunately, an abundance of information resources is available, especially online.
“Science is constantly evolving. The information sciences have evolved exponentially more than the geosciences in the past decade. I think education is the key here, and that’s true for any career,” Caers noted.
“There are lots and lots of resources online. There are countless YouTube tutorials. With COVID, those have proliferated,” Jobe said.
“Another one is the Software Underground. It is a community, so people talk about social coding,” he added.
Online, softwareunderground.org describes itself as “the place for scientists and engineers that love rocks and computers. The Software Underground is a grass-roots community of digital subsurface professionals. We are academic and applied geologists, geophysicists, engineers and others.”
Bauer recommended two free, general courses in the Python language, probably the leading general-purpose coding language for geoscience data today:
• “Dr. Chuck’s Python for Everybody,” a two-course, open-enrollment online resource that grew out of a course at the University of Michigan
• Massachusetts Institute of Technology’s “Intro to Computer Science and Programming in Python,” essentially an online course with video lectures and course materials available
Bauer also presents short courses in Python for the Rocky Mountain Association of Geologists and others, and has taught “Practical Python for Earth Scientists” for AAPG student chapters.
“I think there’s a lot of material out there for free, and there’s no real point in charging a lot for that,” Bauer commented. “There’s a lot of stuff available. If you want to start learning it, you just need to learn the syntax.”
Another frequently recommended activity is the “hackathon,” a collaborative event usually involving both experienced and novice coders working on one project or multiple projects as a learning and sharing experience.
“It speeds up your learning process. You can see how other people are doing things,” Jobe said.
An Integrated Field
Caers said it’s also helpful for geoscientists to learn some basic data science concepts and to understand the role of data in making informed decisions.
“We have long characterized the profession into pieces,” he noted, separating geology from geochemistry and geophysics, for instance.
“It’s becoming an integrated field. Now things are more decision-focused, identifying what data can be used to make decisions. And that means reducing risk,” he said.
The oil and gas industry continues to make gains in data applications, “particularly in how quantitative models are being used to make decisions. That’s where the rubber meets the road,” Caers said.
For geoscience students today, learning computing and data application tools and developing quantitative skills looks like essential preparation for entering the industry of tomorrow.
“The challenge lies often in the lack of quantitative background that geoscience students come in with. It’s important that you don’t become a one-trick pony, so to speak – that you don’t only know your own area well, but you know about other areas, also,” Caers observed.
In giving advice to geoscience students now, “I would say, ‘If you don’t know how to code, you’d better learn,’” Jobe said.
“It’s not going to get any less important through a student’s career. It’s only going to get more important as time goes by,” he observed.
In the future, routine data tasks in geoscience will become increasingly automated, Bauer predicted: “Things that don’t require a lot of creative thought, just automate them,” he said.
Bauer noted that petroleum geoscientists seem to spend 75 percent of their time chasing and developing data, compiling data readings, “pulling data, picking tops.”
By drawing on improved data skills, “I think we can flip that percentage around and spend a lot more time on interpretation,” he said.