Nearly two and a half kilometers beneath the boreal forests of northern Ontario, water has been trickling through cracks in rock that formed before complex life existed on Earth. As it moves, it reacts with iron- and magnesium-rich minerals, and that slow chemistry does something remarkable: it splits water molecules and releases hydrogen gas. A peer-reviewed study now documents that process happening at volumes researchers had not previously measured directly across Canada’s Precambrian Shield. The paper, published in the Proceedings of the National Academy of Sciences (Warr et al., 2024, “Broadscale hydrogen generation from natural water-rock interactions in Precambrian continental crust”), raises a tantalizing possibility: a clean fuel source that has been quietly generating itself underground for billions of years.
The chemistry beneath the Canadian Shield
Two well-understood geologic reactions drive the process. The first, called serpentinization, occurs when groundwater contacts iron- and magnesium-rich minerals deep in the crust. The chemical interaction strips hydrogen atoms free and releases them as dissolved gas. The second, radiolysis, happens when naturally radioactive elements in the surrounding rock emit enough energy to break apart water molecules over geologic time. Both mechanisms have been operating inside ancient continental crust for billions of years, and the U.S. Geological Survey identifies them as the primary pathways for geologic hydrogen formation.
The Canadian evidence is not theoretical. At the Kidd Creek mine near Timmins, Ontario, researchers tapped into deep fracture fluids roughly 2.4 kilometers below the surface, sealed inside rocks dated to 2.7 billion years old. The gas mixture they recovered contained hydrogen alongside helium, methane, and nitrogen, according to a study published in Nature. Because those fluids had been isolated from the surface for an extraordinarily long time, the hydrogen had accumulated through continuous, slow-burn reactions rather than a single geologic event.
Subsequent hydrogeochemical surveys across Precambrian shield groundwaters, including multiple Canadian sites, recorded dissolved hydrogen at millimolar concentrations and percent-level hydrogen in the free gas phase. Those are not trace amounts. They register clearly on standard field instruments. A separate global synthesis published in Nature in 2014 by Sherwood Lollar and colleagues argued that Precambrian continental lithosphere contributes to worldwide hydrogen production through radiolysis and hydration reactions at rates the authors described as previously “underappreciated” in global hydrogen budgets.
The geographic scope stretches well beyond a single mine. A peer-reviewed review focused on Quebec inventoried rock types across the Canadian Shield that share the mineralogy and fracture networks associated with natural hydrogen generation. That analysis, published in Frontiers in Geochemistry, concluded that similar hydrogen-producing geology could extend across broad stretches of the province, widening the potential footprint considerably.
Why harvesting it is a different problem entirely
Detecting hydrogen underground and collecting it as usable fuel are separated by an enormous engineering gap. No publicly available data yet report daily or annual hydrogen flux rates from specific Canadian Shield sites tied to the PNAS study. Without those numbers, it is difficult to judge whether the gas accumulates fast enough to justify extraction infrastructure or whether it seeps away too slowly to compete with manufactured hydrogen, which today is produced mainly through electrolysis or steam methane reforming.
“We have moved from asking ‘is it there?’ to asking ‘how much and how fast?'” said Oliver Warr, a geochemist at the University of Toronto and lead author of the PNAS study, in a summary accompanying the paper’s release. That shift in framing captures the current state of the field: the existence of the hydrogen is no longer in doubt, but its practical recoverability remains unproven.
To put the challenge in perspective: a single large-scale electrolysis plant can produce tens of thousands of kilograms of hydrogen per day on demand. Whether fracture networks in billion-year-old rock can deliver anything close to that rate, steadily and predictably, is an open question. Field logs that could reveal what triggers episodic hydrogen pulses, such as seasonal groundwater recharge cycles or micro-seismic activity along fracture zones, are referenced in the scientific literature but remain unavailable as open datasets.
A quantitative comparison of Canadian Shield hydrogen output against total global Precambrian lithosphere production appears only in secondary summaries, not in a single authoritative accounting. Translating broad estimates into site-specific reserve figures for Canada requires additional drilling data, pressure measurements, and flow modeling that do not yet exist in the public record.
There is also the question of what extraction would physically look like. Options could range from repurposing existing deep mine shafts to drilling dedicated wells, but no pilot project in Canada has publicly tested either approach for sustained hydrogen recovery as of June 2026.
A global race for natural hydrogen
Canada is not the only place where geologists are chasing naturally occurring hydrogen. In the village of Bourakebougou, Mali, a borehole that accidentally struck a hydrogen reservoir in 1987 has been powering a local generator for years, providing one of the few real-world demonstrations that natural hydrogen can be tapped. Startups and exploration companies in France, Australia, and the United States have launched prospecting campaigns, and the U.S. Department of Energy has funded research into what it calls “geologic hydrogen.”
What sets the Canadian Shield apart is its sheer scale. Precambrian rock underlies vast stretches of the country, from Ontario through Quebec and into the northern territories. If even a fraction of that geology produces hydrogen at the concentrations measured so far, the total resource could be significant. But “significant” in geologic terms does not automatically translate to “economically recoverable,” a distinction the energy industry learned repeatedly during decades of shale gas exploration before hydraulic fracturing made those reserves viable.
What the science has settled and what it has not
The strongest evidence comes from direct field measurements. The Kidd Creek fracture fluids, the dissolved hydrogen concentrations in shield groundwaters, and the PNAS documentation of sustained hydrogen generation all represent primary, peer-reviewed data collected at specific depths and locations. These are chemical analyses of real rock and real water, not modeled projections, and they consistently show hydrogen present in measurable quantities.
The global production estimates carry more uncertainty because they extrapolate from limited sampling sites to an entire class of continental crust. The underlying logic is sound, since Precambrian rocks share broadly similar mineralogy worldwide, but the extrapolation depends on assumptions about fracture density, water availability, and temperature gradients that vary from site to site. Those figures are best understood as informed estimates, not confirmed reserves.
Barbara Sherwood Lollar, a geoscientist at the University of Toronto whose work on deep subsurface fluids underpins much of this research, has noted that the discovery of hydrogen-rich waters in Precambrian settings “opens up new questions about habitability, energy, and resources” in Earth’s deep crust. That framing underscores the dual significance of the findings: they matter for energy prospecting and for understanding how microbial life may persist far below the surface.
For anyone following the clean energy transition, the takeaway is specific: ancient rocks in Canada are generating hydrogen through well-understood chemistry, and the volumes are larger than scientists had previously documented. The unresolved question is whether that hydrogen can be tapped at rates that matter for energy markets and at costs that compete with existing production methods. Answering it will require field-scale pilot projects, regulatory frameworks for subsurface gas rights, and sustained monitoring of flow rates measured in years, not weeks. The science has moved from theory to measurement. The engineering and economics still have a long way to go.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
