How cleaning the air changed the methane budget
A Nature paper from May 2025 shows that air pollution has been quietly modulating the global methane sink for two decades, partly buffering the gas’s rise. Why cleaning up soot and carbon monoxide turns out to have complicated climate side effects.
Methane is the greenhouse gas everyone is suddenly serious about. It warms the atmosphere far faster than carbon dioxide over a twenty-year horizon, and its concentration has climbed irregularly for two decades without a fully agreed explanation. Emissions get most of the attention, but a paper published in Nature on 28 May 2025 argues that the other side of the ledger, the sink that removes methane from the air, has been moving too, and that the movement is driven by air pollution.
The lead author is Bo Zheng of Tsinghua Shenzhen International Graduate School, working with a French and German international team. Their analysis covers 2005 to 2021, a period when China and others aggressively cut sulfur dioxide, carbon monoxide and particulate pollution. The standard expectation was that cleaner air would help. What the modelling shows is that cleaner air also changed the chemistry that scrubs methane out of the atmosphere.
The argument runs through the hydroxyl radical, OH. Hydroxyl is the atmosphere’s detergent: it reacts with methane and turns it into water and carbon dioxide, and the size of the hydroxyl pool sets how long methane lingers. Hydroxyl is itself controlled by a chemistry of competing reactions, and that chemistry responds to pollution. Over the period studied, rising tropospheric ozone, increasing water vapour and falling carbon monoxide together pushed the methane sink upward by somewhere between 1.3 and 2.0 teragrams per year, every year, accumulating across the interval. The effect was concentrated in the tropics, where the warm, humid air drives the fastest chemistry.
Why does falling carbon monoxide matter? Carbon monoxide scavenges hydroxyl as well, so when cleaner skies lower carbon monoxide, more hydroxyl becomes available to break down methane. Reducing one pollutant therefore freed the detergent to attack the other. The authors’ point is not that air pollution control is bad; it is that the sink has been quietly buffering the atmospheric methane growth rate, absorbing a share of the emissions that would otherwise have hit the air harder.
That buffer has an uncomfortable implication. The reason methane has not risen even faster these past two decades is partly luck of chemistry riding on pollution control. If emissions continue to fall on their own as economies decarbonize, or if the remaining pollution is cleaned up further, the sink can shrink again and the same emissions leave a larger warming imprint. The climate system is more coupled than the emission-led accounts suggest: what you do to the air affects how long the next greenhouse gas survives in it.
The paper also bears on a long-running argument about methane’s recent surge. Part of the year-to-year wobble in methane growth tracks the sink as much as the source, and the interannual swings that look like emissions surprises may partly be hydroxyl surprises. Pinning down which is which matters, because the policy lever for emissions is very different from the lever for atmospheric chemistry.
The result is a reminder that mitigation is not a set of separate dials. Cutting carbon monoxide and soot saves lives immediately, and it remains the right thing to do, but it cannot be assumed to be free for the climate. The sink that pollution control has been lending us is not permanent. Treating methane as a numbers problem for oil and gas alone misses a system in which the cleaning of the air and the lifetime of the gas are quietly coupled.