AI helps pinpoint hidden sources of underground 'gold hydrogen'

Researchers at Ohio State University have developed a deep learning model to scan the Earth for surface expressions of subsurface reservoirs of naturally occurring free hydrogen.

Green hydrogen production concept
Green hydrogen production concept - AdobeStock

Researchers used the algorithm to help narrow down the potential locations of ovoids or semicircular depressions (SCDs) in the ground that form near areas associated with natural or ‘gold hydrogen’ deposits.

Though these circular patterns often appear in areas of low elevation and can be hidden by agriculture or other vegetation. Recent discoveries of these circles in the US, Mali, Namibia, Brazil, France and Russia have revealed that they exist in greater numbers than previously thought.

Lead researchers Sam Herreid and Saurabh Kaushik, both postdoctoral scholars at the Byrd Polar and Climate Research Centre, Ohio State, combined their model with global satellite imagery data to identify SCDs.

Research teams compiled a list of known SCD locations to train the search algorithm. After using remote sensing data to analyse what these sites look like from above, they drew on geomorphic and spectral patterns to determine which sites around the world are most likely to be associated with SCDs related to geologic hydrogen.

The project found that AI demonstrates a ‘unique ability’ to map out surface expressions of potential subsurface hydrogen reservoirs around the world, as well as establish a baseline for further investigation of hydrogen-associated sites.

In a statement, Joachim Moortgat, the project’s principal investigator and an associate professor of earth sciences at Ohio State, said that scientists have long been aware of hydrogen’s potential as one of nature’s cleanest and most efficient energy sources.

“Hydrogen in general is a very attractive energy source,” Moortgat said. “If you burn it, its only by-product is water, and unlike wind or solar energy, hydrogen can be stored and transported, so there are all kinds of industries trying hard to make the switch.”

Since gold hydrogen is also produced continuously within Earth’s crust, some believe that access to a low-carbon energy resource that is nearly devoid of greenhouse gas emissions could reshape the global energy landscape. Yet, researchers have found that locating hydrogen requires the development of new exploration tools, said Moortgat. 

“One reason [hydrogen sites] are difficult to find is that they probably occur in different kinds of geologies and locations than where you would find oil or gas,” he said. “But with the AI tools we develop, we map everything that could potentially be an SCD.”

Researchers predict that, despite how fast the field seems to be moving, it will take ‘at least a few more years’ before natural hydrogen reservoirs are successfully integrated as a reliable source of clean energy.

“The biggest challenge is that we need to find more SCDs and then really investigate how these things form,” said Moortgat.

“Once we discover a lot more, we will be in a better position to again use AI tools to find similar worldwide.”

The research was presented last week at the annual meeting of the American Geophysical Union.