Geochemists Directly Measure Natural Hydrogen Seeping From Billion-Year-Old Canadian Shield Rocks, Establishing a First Large-Scale Resource Estimate
A PNAS study by University of Toronto and University of Ottawa researchers provides the first direct, decade-long measurements of white hydrogen discharging from an Ontario mine, extrapolating to 140+ tonnes per year at a single site.
Overview
Geochemists at the University of Toronto and the University of Ottawa have published the first direct, long-term measurements of naturally occurring hydrogen gas discharging from the Canadian Shield — the ancient Precambrian rock formation that underlies much of northern Canada. The study, led by University Professor Barbara Sherwood Lollar and co-authored by Assistant Professor Oliver Warr, appeared in the Proceedings of the National Academy of Sciences on May 18, 2026, and draws on more than a decade of continuous underground monitoring at an operating mine near Timmins, Ontario, according to Phys.org.
The result is the first documented large-scale measurement of sustained natural hydrogen discharges anywhere, providing empirical grounding for what had previously been a resource assessed largely through models and indirect inference.
What We Know
The measurements
The study site contains approximately 15,000 boreholes. According to measurements reported by Phys.org, each borehole discharged an average of 0.008 metric tons of hydrogen annually — roughly 8 kilograms, or about the weight of a car battery — and that discharge was sustained continuously for 10 years or more. Extrapolated across all boreholes at the site, the researchers estimate that the mine leaks more than 140 tonnes of hydrogen per year, generating the energy equivalent of approximately 4.7 million kilowatts annually, enough to cover the yearly needs of more than 400 households, as reported by the University of Toronto Faculty of Arts & Science.
What ‘white hydrogen’ is
“White hydrogen” — the term used for naturally occurring geological hydrogen — is defined by its origin: it forms underground through chemical reactions between rocks and groundwater rather than through industrial synthesis. As Autonocion notes, it “is not refined and not manufactured.” Several mechanisms can drive the process. Nature World News identifies three: water-iron reactions, in which “water comes into contact with iron-rich minerals underground” and the process “can split water molecules and release hydrogen gas”; serpentinization, in which “certain rock types react chemically with water under underground heat and pressure”; and radiolysis, in which naturally radioactive elements break apart water molecules over geological timescales. According to Sciencing, scientists studying the Canadian Shield believe serpentinization may play a major role in producing hydrogen beneath these ancient formations.
Geological setting
The Canadian Shield is Precambrian rock — Sciencing places its formation between 500 million and 3 billion years ago — extending across millions of square miles. The same iron-rich geology that produces hydrogen is associated with the mineral deposits already being actively mined. As Oliver Warr noted in the University of Toronto press release: “Natural hydrogen is produced in the same rocks where Canada’s nickel, copper and diamond deposits are found, and that are currently under exploration for critical minerals such as lithium, helium, chromium and cobalt.”
According to SciTechDaily, formations with high hydrogen-production potential have been identified across Northern Ontario, Quebec, Nunavut, and the Northwest Territories. The Canadian Shield also extends across the border into northern Minnesota, Wisconsin, Michigan, and New York, as Autonocion reports, though the paper does not claim that identical flow rates exist beneath U.S. territory.
Novelty of the research
The University of Toronto press release characterizes this as the first documented measurement of large, sustained hydrogen discharges. Prior research in natural hydrogen was based on models or conducted by microbiologists studying subsurface ecosystems and astrobiology, rather than by geochemists making direct long-term flow measurements, according to the University of Toronto Faculty of Arts & Science.
Sherwood Lollar framed the broader context: “There is a global race to increase hydrogen availability in order to decarbonize and reduce the costs of the existing hydrogen economy. We now have a better understanding of the economic viability of this resource that can be mapped to hydrogen deposits around the world that are both already known and yet to be discovered.”
What We Don’t Know
Several important questions remain open. The paper makes no claims about production timelines for commercial viability, costs at the point of distribution, or the economics of hydrogen-powered applications, as Autonocion notes. Extracting the resource would also require purification to remove mixed natural gases — Sciencing describes this as a significant technical challenge that currently limits viability. Infrastructure for transporting and processing geological hydrogen at scale does not yet exist.
The question of whether co-located mining activity is a prerequisite — providing the already-drilled boreholes needed to measure and access the gas — or whether standalone hydrogen extraction would be economically justifiable on its own is not addressed by the paper.
Analysis
The study’s significance lies less in the scale of any single site than in demonstrating that sustained, measurable natural hydrogen discharge is real and quantifiable. The 140-tonne annual estimate for one mine is modest relative to industrial hydrogen demand — SciTechDaily places the current global hydrogen industry at approximately $135 billion, with most production coming from fossil fuels that emit carbon monoxide and carbon dioxide. But Sherwood Lollar’s point about mapping the resource globally is the more consequential one: a measurement framework that works in Timmins can, in principle, be applied wherever similar geology occurs.
Warr’s observation about co-location also matters for the economics. If natural hydrogen systematically appears in the same places as the critical minerals that mining operations are already accessing, the marginal cost of measuring and potentially capturing it could be far lower than developing greenfield hydrogen infrastructure from scratch. “The co-location of mining resources and hydrogen production and use mitigates the need for long transportation routes to market, for hydrogen storage and major hydrogen infrastructure development,” he said, as quoted in the University of Toronto press release.
For Sherwood Lollar, the bottom line is geological: “Canada is blessed that vast amounts of its territories, especially on the Canadian Shield, contain the right rocks and minerals to create this natural hydrogen.”