When the Hunga Tonga-Hunga Ha’apai volcano erupted on 15 January 2022, it sent a column of ash, rock, and seawater vapor to the edge of the mesosphere the most violent eruption recorded in modern history. Scientists have spent the years since cataloguing what it put into the atmosphere. A new study in Nature Communications focuses instead on what the eruption quietly took out: hundreds of billions of grams of its own methane, destroyed by chemistry triggered in the volcanic plume itself.
The findings, published by van Herpen and colleagues, also demonstrate something scientists lacked until now a satellite method capable of detecting and measuring atmospheric methane destruction from orbit, even over the oceans where conventional methane sensors go blind.
Methane is a powerful greenhouse gas, responsible for around 0.5°C of current warming. It is roughly 80 times more potent than CO₂ over a 20-year window, and atmospheric concentrations have been rising at their fastest rate in over 40 years. Unlike CO₂, methane breaks down naturally in the atmosphere within about a decade, which means addressing it could reduce global temperatures within years not centuries. The catch is that conventional strategies cannot eliminate all methane emissions. As a result, researchers have begun exploring whether they can deliberately accelerate the atmosphere’s natural methane-destroying chemistry.
Any such approach would require robust monitoring and verification. The National Academy of Sciences identified this capability as a critical gap because scientists currently lack tools to quantify methane removal at scale. The 2022 Hunga Tonga–Hunga Haʻapai eruption provided an unexpected real-world test case.
The research team used data from TROPOMI, a high-resolution instrument aboard Sentinel-5P, to track the eruption’s stratospheric plume during the days after the blast. Rather than measuring methane directly, which is difficult to detect from space over ocean surfaces, the researchers focused on formaldehyde, a short-lived gas produced when methane breaks down. In the absence of local sources such as forest fires, elevated formaldehyde within the plume provides a reliable fingerprint of methane oxidation.
The results were unprecedented. On 16 January, formaldehyde concentrations reached about 12 parts per billion at an altitude of 30 kilometers, more than one hundred times higher than any previous stratospheric measurement. Other instruments detected sulfur dioxide and volcanic aerosols in the same location, confirming that the signal originated deep in the stratosphere rather than in the lower atmosphere, where formaldehyde is common. The enhancement persisted for at least ten days. This persistence ruled out the possibility that the eruption had injected formaldehyde directly, because formaldehyde has a photochemical lifetime of only about 2.5 hours at midday. Any formaldehyde emitted during the eruption would therefore have disappeared within hours.
Instead, the plume was generating formaldehyde continuously.
The researchers calculated the formaldehyde production rate required to maintain the observed concentrations while accounting for its rapid photochemical breakdown. From this analysis, they estimated that methane oxidation proceeded at approximately 900 metric tons per day across the plume on the first day, with peak local rates of around 60 parts per billion per day. Scientists could still detect the signal on 25 January, indicating that the process continued at substantial rates for more than a week. Overall, the analysis suggests that the eruption injected at least 330 billion grams of methane into the stratosphere and then began oxidizing it.
At the same time, satellite observations detected elevated levels of carbon monoxide and hydroperoxyl radicals within the plume. Both species are consistent with active methane oxidation chemistry and independently supported the formaldehyde-based estimate.
In the stratosphere, methane oxidation is driven primarily by hydroxyl radicals and, to a lesser extent, chlorine atoms. The team found that the chlorine concentration needed to explain the observed methane oxidation was roughly 130 times higher than previous models of the Tonga plume had predicted. Their calculations indicate that the eruption generated active chlorine at rates of 2 to 5 billion grams per day. Neither bromine-catalyzed reactions nor chlorine recycling involving water vapor fully explained this large amount.
The authors propose that iron-chloride photochemistry may have supplied the missing chlorine. The eruption likely lofted between 120 and 380 billion grams of fine ash into the stratosphere. Recent studies showed that these particles persisted longer than initially expected and became coated with sulfuric acid, making them chemically similar to sulfate aerosols. According to the authors, iron within these coated ash particles may have used sunlight to catalytically generate chlorine atoms from sea-salt chloride that entered the atmosphere with seawater. Scientists have previously studied this mechanism in the marine boundary layer, but they have never implicated it in stratospheric chemistry.
The estimated production rates are consistent with observations of mineral dust over the North Atlantic Ocean, even after accounting for the suppressive effect of sulfate coatings. The authors emphasize that this mechanism remains plausible rather than proven. Atmospheric modeling and laboratory experiments will be necessary to test it directly.
Several analytical steps in the study introduce uncertainty. TROPOMI was designed primarily for tropospheric measurements, so the team had to apply sensitivity corrections to interpret the stratospheric signal. These corrections carried uncertainties of about 20%. Cloud cover and interference from biomass burning during subsequent days also limited the researchers’ ability to quantify the full plume. Data from the NASA Microwave Limb Sounder could have provided an independent check, but unusually high water vapor concentrations caused parts of those data to be flagged as unreliable.
The interpretation that methane oxidation produced the formaldehyde also depends on an assumption. Other organic compounds released by the volcano could, in principle, have contributed. The authors argue that volcanic non-methane organic emissions are typically at least an order of magnitude lower than methane emissions, making them an unlikely dominant source, although they cannot exclude this possibility completely.
The study does not estimate the eruption’s net climate impact, nor does it suggest that volcanic eruptions represent a meaningful global methane sink. Although the quantities destroyed are large in atmospheric chemistry terms, they account for only a small fraction of annual human-caused methane emissions.
The study’s most lasting contribution may be methodological. This approach detected methane oxidation at rates of 900 metric tons per day even over open ocean, where direct satellite methane measurements are ineffective. According to the authors, this sensitivity is sufficient to monitor large-scale methane removal technologies currently under investigation. One modeled intervention scenario removes methane at rates about 40 times higher than those observed in the Tonga plume. If the TROPOMI-based method can detect the smaller volcanic signal, it should, in principle, be capable of verifying much larger engineered methane-removal efforts.
As atmospheric methane removal moves from conceptual proposal toward field trials, the governance of such approaches will depend critically on whether scientists can confirm that methane removal is actually occurring, at scale, in the open atmosphere. Until now, the answer to that question was: we lack the tools. This paper offers a first, plausible answer to what those tools might look like.
Reference:
van Herpen, M.M., De Smedt, I., Meidan, D. et al. Satellite quantification of enhanced methane oxidation applied to the stratospheric plume following Hunga Tonga-Hunga Ha’apai eruption. Nat Commun 17, 3746 (2026). https://doi.org/10.1038/s41467-026-72191-4
