Tuesday, 12 August, 2025 / Edition 71

An operator had an awkward problem. They had an idea that there was a vertical well in an area where they were going to drill horizontal wells and there was a possibility of a frac hit on the vertical well. SOP was to pressure up the vertical well, at least for frac. There was one problem: they didn’t know exactly where the well was, AND there was a likelihood it was either in someone’s backyard or under their basement. Yikes. I’d tell you how to story ends, but I honestly don’t know.

Knowing where old wells are is a safety and environmental issue, but so many were drilled before good records were kept that we can’t find them. A couple of methods, including AI, might help change that. As we know, AI requires a lot of data and power, and Google is on the hunt for better ways to power its algorithms. We learn about both this week. Let’s dig in!

Sarah Compton

Editor, Enspired

Improving Abandoned Well Discovery with AI

J.J. Gouin/Shutterstock

Swiss cheese: Potentially hundreds of thousands of undocumented orphan wells (UOWs) exist, and while many could be harmless, most were “plugged and abandoned” with little more than walnut shells and gravel.

That means leaks of environmentally harmful liquids and gases are possible.

Here, kitty kitty: The solution seems clear: find them and plug them with modern methods, but only about 50% of that equation is easy because we don’t know where most of these UOWs are.

Where’s Waldo: Older wells often have no better location information than “three tall man’s paces left of Bob’s tree,” and Bob’s tree was cut down 80 years ago.

Modern tools such as drones, LIDAR, and an array of sensors make locating wells easy (if a house isn’t on top of the well, of course), but knowing where to start is an issue.

Old school plus new school (again): Late last year, scientists at Lawrence Berkeley National Lab sought to change that using old data and new AI.

Plethora of data: Since 2011, the USGS has uploaded 190,000 scans of historical (between 1884 and 2006) USGS geotagged topographic maps.

For maps between 1947 and 1992, a hollow black circle consistently represents an oil and gas well, which is useful, but the US covers more than 3 million square miles.

That’s a lot of maps to cover, even for an intern fueled by Celsius and meat sticks.

Many of today’s “older” AI algorithms were built for almost exactly this, however.

The algorithm can be “trained” on what a well would look like on maps — using that symbol as a guide with some refinement to discern between cul-de-sacs or just a stain on the map — and then set to run amok across the rest of the data set.

X marks the spot: Charuleka Varadharajan’s team at Berkeley used AI on data from four counties of interest — LA and Kern counties in California, and Osage and Oklahoma counties in Oklahoma — and found 1,301 potential undocumented orphaned wells.

Geocaching skills transfer: The real work begins once the team IDs a potential location, because it’s been learned the algorithm tends to put the team within about 10 meters (30 feet) of the actual location.

Fine-toothed combs: A combination of on-the-ground footwork, drones to detect local methane leaks, LIDAR to search for near-surface structures, and magnetometers to look for any indication of casing or wellheads can help locate the well.

It seems like more work is needed to refine the results even further (though it’s possible there really were 1,300 undocumented wells in those four counties given their oil and gas history), but there’s a running theme here of using new tools with old data to tackle these problems.

Geoscience scout card skills: As folks familiar with poring over scout cards and other resources to squeeze every last bit of information about potential wells, geoscientists are best poised to help with this sort of work.

To read the paper, go here.

Sponsored

Harnessing Machine Learning for Enhanced Seismic Interpretation

Discover how ML-driven fault and horizon interpretation is delivering measurable improvements in subsurface imaging across complex geological settings including the North Sea, Gulf of Mexico, and offshore Brazil.

Next Gen Batteries Are … Carbon Dioxide Powered?

KM Stock/Shutterstock

She needs more power, Captain! It’s no secret data centers require a lot of power, and with some cities voting against data centers in their towns, it’s crucial for tech giants to solve their utility demands.

Sourcing their own: There hasn’t been a one-size fits all approach: Microsoft is commissioning a rebuild of Three Mile Island in a bid for rights to the electricity generated there. Google partnered with a company called Energy Dome 25 July 2025 to use their CO₂ battery technology to power their operations.

Zero carbon goal: Google would like to run 24/7 on carbon-free power by 2030. It’s heavily invested in wind and solar, but also in long-term energy storage to be called upon when the wind isn’t blowing and sunshine’s nowhere to be found.

Enter, Energy Dome, whose CO₂ battery can release energy for 8 to 24 hours, is modular and site-independent, and suffers no supply chain bottlenecks due to off-the-shelf equipment.

Energy Dome’s CO₂ battery uses CO₂ much the same way most turbines use water: the CO₂ is heated, evaporates and expands, and the expansion turns the turbine.

Zero emission power: The system leaks no CO₂ to the outside and reduces the costs typically associated with CO₂ because it’s stored at ambient temperatures in its liquid phase.

CO₂ has some advantage over other battery options such as Lithium.

Maybe easier to work with: CO₂ is both abundant and something we’re actively trying to find either a use or storage for vs lithium, which is not the simplest raw material to collect in useful amounts.

The round trip efficiency is lower for Li (75% for CO₂ battery vs 85% for Li-ion) but there’s little to no degradation of the CO₂ battery over time.

The storage tanks for the liquid CO₂ don’t leave much of a footprint, but their CO₂ “dome,” which is where the gaseous CO₂ is stored after it’s done its work turning the turbine, appears to leave quite a footprint (I couldn’t find exact dimensions).

Growing kiddo: This is likely because CO₂ expands to nearly 535 times its liquid volume as it changes to a gas, a fact the company is using to move those turbines, but also requires a large area to store the gaseous phase in.

Although Energy Dome’s explanation of the process is easy enough to understand, a basic knowledge of the phases of CO₂ tells you it’s not the temperature that needs to change but the pressure.

At ambient temperatures and pressures, CO₂ is a gas. You need about 5x atmospheric pressure to get it into a liquid phase and less than 88 degrees Fahrenheit temperature.

Their site explicitly mentions “no cryogenic temperatures,” so I’m having a hard time reconciling “heating up” when it seems like the process really would be more of a “depressurizing.”

Exact mechanisms aside, they have caught Google’s attention in the company’s pursuit of its lofty 2030 goal, and the battery has one more potential reservoir for all the CO₂ we’re poised to catch in the future.

To learn more from Energy Dome themselves, go here.

👍 If you enjoyed this edition of Enspired, consider supporting AAPG's brand of newsletters by forwarding to a friend or colleague and signing up for our other newsletters here.

➡️ Was this email forwarded to you? Sign up for Enspired here.

✉️ To get in touch with Sarah or suggest a topic for Enspired, send an email to editorial@aapg.org.