Something happened on the seabed that was not supposed to happen
In the Clarion–Clipperton Zone between Mexico and Hawaii, researchers lowered chambers to the seafloor to measure how quickly organisms and sediments consume oxygen. Inside a sealed chamber, its concentration should fall.
Instead, oxygen rose in some experiments — within two days to more than three times the background level. The authors of a 2024 paper in Nature Geoscience called the phenomenon dark oxygen production.
~4 kmdeep, with no sunlight
>3×the background level in some chambers
0.95 Vhighest reported surface potential
The hypothesis: rocks behaving like natural batteries
The seabed is covered with polymetallic nodules — irregular objects that accumulate layers of manganese, nickel, cobalt and copper over millions of years. They are valuable to industry, but also form part of a poorly understood ecosystem.
The original team measured electrical potential differences on some nodules and proposed a bold explanation: clusters of minerals might act as “geobatteries”, supporting reactions that split water and release oxygen.
The crucial word is hypothesis. The study reported an unusual oxygen signal and voltages on nodules, but it did not directly demonstrate a complete electrolysis mechanism.
Why would this be revolutionary?
Photosynthesis is not the only known route to oxygen in darkness — rare microbial processes and radiolysis also exist. But oxygen production on the abyssal seafloor at the reported scale would be something new.
It could mean that local “oxygen islands” arise in dark environments and support metabolism without a direct supply from sunlit surface waters.
That raises a provocative question about life’s history: could small oxygenated niches have existed before photosynthetic organisms oxygenated the wider planet?
Thermodynamics says: show the evidence
In December 2025, another group published a detailed critique. Its central objection was simple: splitting water requires energy, and the measured voltages did not reveal a source capable of sustaining the process.
- Ideal water electrolysis requires at least 1.23 V, and more in practical systems.
- Hydrogen, which should accompany oxygen during water splitting, was not measured.
- Comparable seabed studies usually reported oxygen consumption, not production.
The most ordinary suspect: trapped air
Critics argue that initial oxygen concentrations in some chambers were unusually high and variable. The instruments may not have been completely flushed with bottom water, allowing trapped air to dissolve gradually inside them.
Particularly awkward for the “geobattery” hypothesis are control runs in which oxygen reportedly rose even in chambers without nodules. If those data accurately represent the experiment, the apparatus becomes a stronger suspect than the minerals.
On April 8, 2026, Nature Geoscience added an editor’s note saying that concerns about the paper were being examined. The article has not been retracted, but its central conclusions remain unresolved.
The dispute has another layer
This is not a simple contest between discoverers and completely detached sceptics. Some authors of the critical paper work for companies interested in deep-sea mining, while the original study also received partial funding connected to the sector.
A conflict of interest does not decide who is right. It means the evidence, experimental controls and independent replication deserve more weight than institutional authority.
The decisive question is whether different instruments, in another location and with stronger controls, will detect the same signal.
Two machines are meant to settle the argument
The Dark Oxygen Research Initiative built two specialist deep-sea landers: Alisa and Kaia. They are designed to measure oxygen, hydrogen, conductivity, pressure, turbidity and markers that can distinguish electrolysis from biological activity.
An Aquatic Eddy Covariance system will also measure oxygen exchange above the seafloor without enclosing it in a chamber — essential for testing whether the container itself created the signal.
Project information said the instruments were due to be deployed in the Clarion–Clipperton Zone in 2026. As of this article’s publication on July 18, 2026, final results were not yet publicly available.
What does this mean for deep-sea mining?
Nodules contain metals used in batteries and electronics. Companies want to collect them with large machines moving across the seabed.
If nodules really participate in oxygen production, removing them could destroy an ecosystem function missing from current models. If dark oxygen proves to be an artefact, mining does not automatically become safe: physical habitat loss, sediment plumes and damage to poorly known species remain.
Three possible endings
- Breakthrough: independent measurements confirm a new oxygen source.
- Surprise: oxygen is real, but microbes or another reaction produce it.
- Artefact: the signal comes from chambers, calibration or trapped air.
Every outcome would be useful. Even disproving the spectacular hypothesis would improve the tools used to study one of Earth’s least understood environments.
The central idea
Science does not lose credibility when it detects an error. It would lose credibility if it stopped trying to detect one.
There may be an unknown source of oxygen on the Pacific floor. There may only be a poorly vented measurement chamber. The inspiring part is that researchers are not being asked to choose the more attractive story: they are building better instruments and returning for an answer.
Sources and further reading
This article describes an active scientific dispute. Oxygen production by polymetallic nodules has not been independently confirmed.
← Back to Science room