More than two kilometers beneath the boreal forests of northern Ontario, water has been seeping through billion-year-old rock for longer than complex life has existed on Earth. As it moves through iron-rich fractures in the Canadian Shield, that water is slowly split apart by chemical reactions and natural radiation, releasing hydrogen gas molecule by molecule into the deep subsurface. Scientists have known about this process in theory for years. Now, a decade of continuous monitoring has confirmed it is real, persistent, and larger than most geologists assumed.
A study published in May 2025 in the Proceedings of the National Academy of Sciences presents what its authors call a decadal record of subsurface hydrogen generation from Precambrian crystalline rock in the Canadian Shield. Led by geochemist Barbara Sherwood Lollar and research scientist Oliver Warr at the University of Toronto, the study tracked hydrogen concentrations across multiple deep boreholes, including sites at the Kidd Creek Mine near Timmins, Ontario, where fracture waters more than 2.4 kilometers deep have yielded some of the highest dissolved hydrogen concentrations ever recorded in continental crust. The gas kept flowing the entire time, not as a brief pulse but as a sustained discharge, distinguishing this work from earlier one-off measurements and establishing that the underlying chemistry operates on timescales relevant to energy planning.
“The hydrogen is present, it continues to be generated, and the volumes recorded over a decade suggest the resource is not trivial,” Sherwood Lollar said in an institutional press statement tied to the PNAS publication.
The finding has landed at a moment when governments and energy companies worldwide are racing to secure sources of low-carbon hydrogen. Most current plans rely on electrolysis, which uses electricity to split water, or on reforming natural gas with carbon capture. If ancient rocks can produce hydrogen on their own, continuously and without external energy inputs, the economics of the hydrogen transition could look very different.
The chemistry driving the process
Two well-understood reactions account for the hydrogen flowing out of the Canadian Shield’s deep fractures.
The first is serpentinization. When water contacts iron-rich minerals in mafic and ultramafic rock, the iron oxidizes and the water molecules break apart, liberating hydrogen gas. The reaction is thermodynamically favorable at the temperatures and pressures found several kilometers underground and can proceed for as long as fresh mineral surfaces and water remain in contact.
The second is radiolysis. Uranium and thorium naturally present in Precambrian rock emit radiation that splits water molecules in surrounding pore spaces and fractures. Unlike serpentinization, radiolysis does not require specific mineral chemistry; it needs only water and radioactive decay, both of which are abundant in old continental crust.
Earlier research by the same group reported exceptionally high dissolved hydrogen concentrations in deep fracture waters of the Canadian Shield and used isotopic signatures to link the gas directly to these two mechanisms. The new PNAS paper builds on that foundation by showing the reactions are not just detectable in snapshots but measurable and ongoing across a full decade of observation.
Why the scale matters
A separate analysis published in Nature concluded that hydrogen generation from ancient continental crust has been systematically underestimated in global models. If Precambrian shield rocks produce more hydrogen than legacy calculations projected, the total natural supply could be substantially larger than anyone assumed.
The Canadian Shield is one of the largest exposed Precambrian formations on the planet, stretching across roughly 4.8 million square kilometers from the Great Lakes to the Arctic. But it is not unique. Similar geology underlies the Fennoscandian Shield in Scandinavia, the Pilbara and Yilgarn cratons in Australia, the West African Craton, and large portions of Brazil. If the hydrogen-generating reactions documented in Ontario operate at comparable rates in those formations, the phenomenon could have a global footprint.
Sherwood Lollar and Warr also emphasized a parallel scientific significance: the same hydrogen flux that interests energy researchers likely sustains deep microbial ecosystems that have been cut off from the surface for hundreds of millions of years, ecosystems that offer a window into how life might persist on other rocky planets.
The gap between detection and extraction
Confirming that hydrogen exists underground and proving it can be extracted commercially are two very different achievements, and the distance between them remains vast.
The PNAS paper provides summary statistics and methodological descriptions of the borehole monitoring, but comprehensive discharge time-series data and detailed borehole logs have not been deposited in public repositories. Without open access to the raw data, independent researchers cannot fully replicate the analysis or test alternative explanations, such as subtle measurement artifacts or unrecognized mixing with other gas sources, without coordinating directly with the original team.
Scaling the results from a handful of boreholes to a craton-wide resource estimate depends on modeling assumptions about fracture density, rock composition, and fluid flow that have not been validated with new empirical field data beyond the study sites. The U.S. Geological Survey has outlined a source-reservoir-seal framework for evaluating geologic hydrogen systems, stressing that both hydrogen generation and effective trapping structures must be present for gas to accumulate in extractable volumes. Whether the Canadian Shield contains the right combination of sealed reservoirs and accessible fracture networks is an open question.
The study’s working hypothesis is that fracture density in iron-rich mafic units, combined with continuous water access, drives production rates more than overall rock age alone. That idea is testable but unconfirmed. Establishing a reliable spatial pattern would require coordinated geophysical surveys, detailed fracture mapping, and systematic gas sampling across multiple shield localities. As of June 2026, no such large-scale campaign has been reported in the primary literature.
Regulatory and environmental unknowns
Canada has no regulatory framework specifically designed for natural hydrogen extraction. Publicly available documents from federal and provincial authorities do not yet address permitting, environmental review, or land-use rules tailored to this kind of resource. Hydrogen could fall into a gap between existing mineral rights and gas rights regimes, complicating questions of tenure, royalties, and Indigenous consultation obligations. No company has announced a commercial pilot in the Canadian Shield targeting geologic hydrogen, and no government agency has published a formal resource assessment for the gas.
Environmental risks are similarly uncharted. If hydrogen accumulates in significant subsurface pockets, drilling and extraction could disrupt deep microbial ecosystems that depend on the gas as their primary energy source. Surface impacts from drilling pads and infrastructure would likely resemble those of other subsurface industries, but the specific consequences of hydrogen leakage, changes to groundwater chemistry, and potential induced seismicity in shield rock have not been rigorously evaluated.
Where geologic hydrogen stands in the clean-energy race
The strongest evidence in this story sits in the peer-reviewed PNAS paper and the earlier Nature analysis. Together, they support two conclusions with reasonable confidence: hydrogen is being generated continuously in Canadian Shield rocks, and the global scientific community has likely underestimated how much hydrogen ancient continental formations can produce over geologic timescales.
What the evidence does not yet support is any claim about commercial viability. The economic and engineering questions remain entirely unresolved: whether geologic hydrogen can be located reliably across a craton, gathered without excessive water or energy inputs, purified to fuel-cell or industrial standards, and delivered at a cost that competes with electrolysis or steam methane reforming with carbon capture.
For the clean-energy sector, the practical significance is specific rather than sweeping. The Canadian Shield findings justify further exploration drilling, targeted geophysical surveys, and the development of regulatory pathways for a new kind of subsurface resource. They do not, on their own, guarantee a new hydrogen industry. Until independent field confirmations at other shield sites, transparent economic analyses, and formal resource assessments emerge, geologic hydrogen from ancient rocks should be understood as a scientifically grounded but commercially unproven addition to the broader portfolio of low-carbon energy options.
The rocks, at least, are patient. They have been making hydrogen for a billion years. The question is whether humans can figure out how to collect it before the energy transition moves on without them.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
