From mid-January 2025, satellites tracked at least 38,200 tonnes of methane and 2.9 million tonnes of carbon dioxide leaving Fentale’s caldera in Ethiopia, at a pace matching big gas-well blowouts, though no magma reached the surface.
Peer reviewed · Satellite observations · Sediment source proposed, not sampled
Fentale, a volcano in Ethiopia’s Main Ethiopian Rift that has not erupted since about 1810, released at least 38.2 kilotonnes (kt) of methane over three months in 2025, a Nature Geoscience study published on 19 November 2026 reports. Edna Dualeh and 23 co-authors, led from the University of Bristol, measured it from space. It is the first methane plume detected by satellite from a volcano, and a hint of how magma heating buried sediments may once have warmed the planet. The catch: nobody sampled the gas, so its sedimentary origin is inferred.
What you need to know
- Finding: After magma forced its way underground near Fentale in December 2024, the caldera vented at least 38.2 ± 3.9 kt of methane between 19 January and 9 April 2025, peaking at 157 ± 41 tonnes per hour.
- Method: Three gas-sensing satellite instruments (TROPOMI, GHGSat and Tanager-1) measured the plumes, while thermal, optical and radar satellites tracked heat, a ground-hugging cloud and sinking ground inside the caldera.
- Source: Dualeh, Biggs, Sharma et al., Nature Geoscience, 19 November 2026 (open access), with the published peer-review reports.
- Why it matters: Volcanoes typically emit very little methane; this one matched the daily rate of major industrial gas leaks and offers a small, modern example of a process linked to ancient warming events.
- Limitation: There were no ground samples or carbon-isotope measurements at Fentale, so the claim that the methane came from cooked sedimentary rock rests on regional data and reasoning, and the totals are minimums.
Evidence at a glance
| Item | Assessment |
|---|---|
| Evidence status | Peer reviewed. Remote-sensing observations of one event at one volcano. The gas source and the trapping mechanism are the authors’ interpretation. |
| Primary source | E. W. Dualeh, J. Biggs, S. Sharma et al., “Methane outburst driven by magma–sediment interactions at Fentale Volcano detected by satellite”, Nature Geoscience (2026), doi:10.1038/s41561-026-02137-4. Received 13 February 2026, accepted 28 September 2026, published 19 November 2026. CC BY 4.0. Derived data on Zenodo. |
| Evidence strength | Strong that a large methane and CO₂ outburst happened, because independent instruments on different satellites saw it and one peer reviewer called the methane detections “unambiguous”. Moderate for the size, since the authors call the totals minimums limited by satellite overpasses. Limited for the cause, which rests on isotope data from other rift volcanoes and on the gas amounts. |
| Principal limitation | No gas was collected at Fentale, so the thermogenic (sediment-derived) origin and the “sealed cap” story are proposed, not measured. |
| Coverage at a glance | 1 independent credible publisher found as of 21 November 2026, 08:30 UTC (Nature World News), from SciQuest web searches for “Fentale”, “methane” and the paper title. Releases from the University of Bristol and SRON were read and are listed separately; a Newswise reprint and two aggregator rewrites were not counted. Coverage is not evidence that a finding is correct. |
Sources: Dualeh et al., Nature Geoscience (2026), main text, Methods and Peer Review File; University of Bristol and SRON releases (19 November 2026); SciQuest coverage search at the time stated.
Is this a big deal?
For volcano science, yes, a solid step. Since 2018, TROPOMI has spotted methane from oil and gas sites, landfills and mud volcanoes, but never from a volcano. The rate was roughly four orders of magnitude (about 10,000 times) above methane rates measured at other volcanic systems. For today’s climate it is small: the paper puts it below 0.1% of the oil and gas industry’s yearly methane emissions.
What scientists found
On 17 December 2024, magma began pushing sideways underground from near Fentale, about 100 km from Addis Ababa. The intrusion, a vertical sheet of magma called a dyke, ran about 50 km north by 29 December and kept opening until mid-March 2025. Roughly 1.4 cubic kilometres of magma stayed trapped below ground.
The gas came weeks later. TROPOMI, an instrument on the European Space Agency’s Sentinel-5P satellite, first saw a strong methane plume on 19 January 2025, at 157 ± 41 tonnes per hour. About 90% of the total, 33.8 ± 3.5 kt, escaped in the first month. That is an average of about 1,090 tonnes a day.
Carbon dioxide came too. Over a 14-day window, the satellites measured at least 2.9 ± 1.0 million tonnes (Mt) of CO₂, the largest CO₂ release yet measured from a volcano that was not erupting.
How they found it
Bristol’s team was already monitoring Fentale with Ethiopian colleagues during the unrest when SRON, the Dutch space research institute that flags methane plumes for the EU’s Copernicus Atmosphere Monitoring Service, and the satellite company GHGSat spotted the plume and contacted them. “The observation of a methane outburst from a volcano was a complete surprise,” Dualeh said in the university’s release.
TROPOMI’s pixels are 5.5 by 7 km, too coarse to pin the source. So GHGSat pointed a satellite with 25 m pixels at the area on 31 January and traced the methane, then at 62.8 ± 36.0 tonnes per hour, back to Fentale’s caldera, the collapsed bowl at its summit.
Other satellites filled in the story. A thermal hotspot appeared at 22:48 UTC on 14 January. About nine hours later, a white cloud of water droplets sat low in a hollow of the caldera; its tendency to hug the ground pointed to a heavy, CO₂-rich gas. Radar showed the caldera floor sinking 21.6 ± 1.2 cm between 15 and 23 January. On 14 February, part of the caldera collapsed by about 40 m, around the time of a magnitude 6.0 earthquake about 20 km down.
| Date (UTC) | What satellites recorded |
|---|---|
| 17 Dec 2024 | Dyke intrusion starts near Fentale; about 50 km long by 29 December |
| 14 Jan 2025 | First thermal anomaly in the caldera (22:48) |
| 15 Jan 2025 | Low-lying white cloud in the caldera (07:28); ground starts sinking |
| 19 Jan 2025 | First methane plume detected: 157 ± 41 t per hour |
| 31 Jan 2025 | GHGSat confirms the caldera as the source: 62.8 ± 36.0 t per hour |
| 5 Feb 2025 onward | CO₂ detected in the caldera; plumes through February total at least 2.9 ± 1.0 Mt over 14 days |
| 14 Feb 2025 | Caldera collapse of about 40 m with a deep M 6.0 earthquake; methane 20.3 ± 12.0 t per hour |
| Mar–Apr 2025 | Methane falls to 0.2–5.0 t per day; last methane measurement 9 April; low cloud last seen 17 April |
Source: Dualeh et al., Nature Geoscience (2026), main text and Figs 1–3. Uncertainties are one standard deviation.
Why it matters to you
This won’t change your weather or energy bill. Methane traps about 81 times as much heat as CO₂ over 20 years, but the paper calls this single outburst small compared with yearly natural and human emissions, and the Bristol release says it would not noticeably affect today’s climate.
It matters in two other ways. Uncertainty about natural methane sources is still large, and short bursts from rift volcanoes are not fully accounted for in the global methane budget. The gases are also a hazard: methane and CO₂ are colourless and odourless, and the paper notes that CO₂ at about 10% of the air can cause unconsciousness and death. Fentale’s caldera is uninhabited, but other calderas along the rift are home to large populations.
What the evidence does not show
The study does not show where the methane formed. Methane made by magma itself carries a heavier carbon signature than methane cooked out of buried organic matter. No isotope measurements exist for Fentale. Values from other East African Rift volcanoes, −31‰ to −18‰, fit a sedimentary source, and the authors add that magma alone would have had to release vastly more CO₂ than was measured to make this much methane.
The totals are floors. TROPOMI could not get valid readings directly over the caldera, and satellite passes were limited, so the authors call the CO₂ rate likely underestimated. Emissions smaller than about 8 tonnes per hour can slip below TROPOMI’s detection limit.
How this fits what we already knew
Volcanoes are usually poor methane sources. Well-studied European volcanoes emit less than 0.2 tonnes a day each. The East African Rift was already odd: the authors reanalysed older ground measurements from rift volcanoes there and found 0.2 to 10.1 tonnes a day each, with gas up to 45% methane by volume.
The likely reason is buried rock. The rift formed 20–30 million years ago across Mesozoic sedimentary basins whose shales, limestones and sandstones, 2–5 km down, hold oil and gas and draw petroleum exploration. The team proposes that magma heat slowly turned organic matter there into methane, which collected under an impermeable seal until the 2025 dyke cracked it.
Think of a fizzy drink bottle that has been building pressure for decades: crack the cap and most of the gas rushes out at once, then the flow tails off. The paper’s methane record has that shape, a fast release followed by an exponential decline. The analogy has limits. The gas was stored in rock pores, not dissolved in liquid, and the collapse on 14 February briefly revived the flow.
The authors’ arithmetic supports the seal idea. At the slow rates seen along the rift, 0.5 to 1 tonne a day, it would take 100–200 years to build up 38 kt, close to the time since the last eruption around 1810. On a much larger scale, the same magma-meets-sediment process is linked to extreme ancient warm spells such as the Palaeocene–Eocene Thermal Maximum, and to large igneous provinces such as the Siberian Traps, linked to the end-Permian extinction.
Show the maths
First-month rate: 33.8 kt ÷ 31 days ≈ 1.09 kt (1,090 t) per day, matching the paper.
Peak as a daily rate: 157 t/h × 24 h ≈ 3,770 t per day.
Blowout totals (rate × days): Louisiana 2019, 1,289 × 38 ≈ 49 kt; Aliso Canyon 2015, 866 × 112 ≈ 97 kt; Kazakhstan 2023, 639 × 205 ≈ 131 kt. Fentale’s 38 kt total is smaller than all three, even though its first-month rate falls between them.
CO₂ rate: 2.9 Mt ÷ 14 days ≈ 207,000 t per day (paper: 2.1 × 10⁵ t per day).
Build-up time: 38,000 t ÷ 1 t per day = 38,000 days ≈ 104 years; at 0.5 t per day, about 208 years.
Warming equivalent (SciQuest calculation): 38.2 kt × 81 ≈ 3.1 Mt of CO₂-equivalent over 20 years, similar to the CO₂ the volcano released directly. This uses the paper’s rounded factor and ignores uncertainty.
What the peer reviewers said
No independent researcher had commented publicly by the time of writing. The journal published the two anonymous peer reviews; Simon Carn is named as one reviewer, but no comment is attributed to him. Reviewer 1 wrote that “the methane emissions observed by TROPOMI and GHGSat are unambiguous” but found the comparison with ancient lava provinces “somewhat tenuous”, since one dyke is tiny next to their huge sheets of magma.
The same reviewer first judged that “the origin of the observed methane emissions remains ambiguous” and asked for magma-made methane to be considered. The authors added that analysis, and the reviewer then recommended publication. Reviewer 2 called the idea “a really intriguing proposition” and asked whether the dyke could have generated fresh methane rather than releasing old gas.
How other headlines framed it
With one independent outlet so far, this compares it with the two institutional releases.
- University of Bristol framed it as a solved puzzle with lessons for past climate change. It dated the observations December 2024 to March 2025; the paper’s methane measurements run from 19 January to 9 April 2025, with the dyke starting on 17 December.
- SRON led on a record, the largest methane outburst measured from a volcano, and on its own role in spotting the plume. The record claim matches the paper, which calls it the largest ever detected from a natural point source.
- Nature World News also led on the record and stressed that today’s climate is unaffected, but it did not link the paper.
What happens next
The authors call for ground-based gas measurements along the rift to test how the methane forms and escapes. Fentale is still active: high ground temperatures persisted through 2025 and were still visible in mid-2026, and parts of the collapse margin are sinking about 400 mm a year. Any similar outbursts before 2018, when TROPOMI began global coverage, probably went unseen, so continued satellite watch on rift volcanoes is the main way to learn how often this happens. For background on what drives magma upward, see what triggers volcanic eruptions.
In one sentence
A non-erupting Ethiopian volcano vented methane at the pace of a major gas-well blowout after magma cracked the ground beneath it, and the likely culprit, gas cooked out of buried sediments, still needs to be confirmed on the ground.
Quick check: did it click?
- Did Fentale erupt in 2025? No. The magma, about 1.4 km³, stayed underground in a dyke; the gas escaped through the caldera.
- Why do the authors think the methane came from sediments rather than magma? Isotope values at other rift volcanoes fit a sedimentary source, the area sits on hydrocarbon-rich basins, and magma alone would have needed far more CO₂ than was measured.
- Will this outburst change today’s climate? No. It is below 0.1% of the oil and gas industry’s yearly methane emissions.
Read next
- What triggers volcanic eruptions: the pressure, gas and magma behind an eruption.
- End-Permian extinction hit North China’s land and sea in step: the giant-scale version of magma meeting buried carbon.
- From crust to core: the layers of Earth: where rifts, dykes and sedimentary basins sit.
Sources and disclosure
- Dualeh, E. W., Biggs, J., Sharma, S. et al. Methane outburst driven by magma–sediment interactions at Fentale Volcano detected by satellite. Nature Geoscience (2026). Open access, CC BY 4.0.
- Peer Review File for the paper, Nature Geoscience (2026), CC BY 4.0.
- Derived datasets, Zenodo, doi:10.5281/zenodo.21361500.
- University of Bristol, “Scientists uncover secret of mysterious methane outburst from Ethiopian volcano”, 19 November 2026.
- SRON, “Satellites capture largest methane outburst ever measured from a volcano”, 19 November 2026.
- Nature World News, “Satellites Catch the Biggest Methane Burst Ever Measured From a Volcano, in Ethiopia”, 20 November 2026.
Disclosure: This article was drafted with AI assistance and checked against the sources above, including the full paper and its peer-review reports. The rate totals, build-up times and CO₂-equivalent figure in “Show the maths” are SciQuest calculations from the paper’s numbers; the graphic redraws the paper’s mechanism figure in simplified form. No conflicts of interest. Spotted an error? See our corrections policy.
Discover more from SciQuest
Subscribe to get the latest posts sent to your email.

Leave a comment