Study posits how carbon dioxide cools the upper atmosphere, and warms earth below

Researchers have supposedly solved a long-standing atmospheric puzzle: How rising carbon dioxide cools the stratosphere even as it warms Earth’s surface and lower atmosphere.

Even as temperatures rise on Earth’s surface and in the lower atmosphere, the planet’s upper atmosphere has cooled dramatically. This paradoxical pattern is a well-known sign of humanity’s climate impacts—but until now, the underlying physics has remained a mystery.

In a new study, researchers from Columbia University describe the phenomenon’s mechanics, illuminating how it is largely determined by the way carbon dioxide (CO2) interacts with different wavelengths of light.

“It explains a phenomenon that’s a fingerprint of climate change, has been known to occur for decades, and has not been understood,” says Robert Pincus, a research professor of ocean and climate physics at Lamont-Doherty Earth Observatory, which is part of the Columbia Climate School, and co-author of the study published in Nature Geoscience.

In the lower atmosphere, CO2 molecules trap heat that would otherwise escape into space. Higher in the atmosphere, though, the dynamics change. In the stratosphere—the atmospheric layer that extends from about 11km to 50 km above Earth’s surface—CO2 molecules function almost like a radiator, absorbing infrared energy from below and emitting some of that energy into space. When more CO2 is added, the stratosphere radiates heat away more efficiently and it cools.

This was predicted in the 1960s by climatologist Syukuro Manabe’s Nobel Prize-winning models of Earth’s climate and CO2-induced global warming. The stratosphere has cooled by roughly 2 degrees Celsius since the mid-1980s. That’s estimated to be more than 10 times the amount of cooling that would have occurred in the absence of human-caused CO2 emissions.

However, though the basic principles of stratospheric cooling are understood, the specifics have remained cloudy. “The existing theory was incredibly insightful, but at the moment we lack a quantitative theory for CO2-induced stratospheric cooling,” says Sean Cohen, a postdoctoral research scientist at Lamont-Doherty Earth Observatory, which is part of the Columbia Climate School, and the study’s lead author.

Cohen, Pincus, and Lorenzo Polvani, a geophysicist in Columbia Engineering’s Department of Applied Physics and Applied Mathematics, developed their theory through an iterative method of identifying the key processes involved in stratospheric cooling, assigning mathematical values to them, comparing the results of their pen-and-paper models to comprehensive simulations and real-world data, tweaking their equations and repeating. Over several months they deduced the equations that best fit.

The researchers arrived at a central factor: how CO2 molecules interact with light, and in particular infrared—also known as longwave—light. Not every infrared wavelength passes through them in the same way. Some wavelengths contribute to cooling more than others, and the team determined that wavelengths in a certain “Goldilocks zone” are especially efficient. As CO2 accumulates in the atmosphere, that zone expands.

“It’s those changes in efficiency that are going to ultimately be what’s driving stratospheric cooling,” says Cohen.

The researchers also quantified the roles played by ozone and water vapor. These are implicated in similar processes as CO2—they too can trap heat in the lower atmosphere but contribute to cooling in the stratosphere by radiating heat—but turn out to have little influence compared with CO2.

The researchers’ equations fit with three well-described phenomena: How stratospheric cooling varies by altitude, with the least cooling occurring at its lowest level and the most at its highest level; how each doubling of CO2 translates to a cooling of 8 degrees Celsius at the stratopause, or the stratosphere’s upper reaches; and how a cooler stratosphere lets less infrared energy escape to space, increasing CO2’s heat-trapping effect. In other words: CO2 makes the stratosphere better at radiating, which cools it—but because it becomes colder, the Earth system ends up losing less heat to space overall, strengthening warming below.

“Here’s this process that we’ve known about for 50-plus years, and we had a pretty decent qualitative understanding of how it worked. However, we didn’t understand the details of what actually drove that process mechanistically,” says Cohen.

Cohen and Pincus say the implications of the work are less about adding one more piece of evidence to support global warming—that reality is already clear—than developing a better understanding of the mechanisms involved in stratospheric cooling. “This is really telling us what is essential,” says Pincus, and it can inform future research on the process. The findings may also help scientists studying conditions outside of Earth.

“Maybe we can better understand what’s going on in the stratospheres of other planets in our solar system or exoplanets,” says Cohen.


Journal Nature Geoscience DOI 10.1038/s41561-026-01965-8 

The climate data they don't want you to find — free, to your inbox.
Join readers who get 5–8 new articles daily — no algorithms, no shadow bans.
1.8 8 votes
Article Rating
Subscribe
Notify of
31 Comments
Inline Feedbacks
View all comments
Bruce Cobb
May 13, 2026 2:09 pm

What a load of cobblers.

Reply to  Bruce Cobb
May 13, 2026 4:32 pm

Yup cherry cobbler…
There’s a lot of room for cherry picking in the satellite record back to the 80’s…..
see the known problems bit…
https://www.nature.com/articles/s41597-024-03147-w

Chris Hanley
May 13, 2026 2:31 pm

The stratosphere has cooled by roughly 2 degrees Celsius since the mid-1980s. That’s estimated to be more than 10 times the amount of cooling that would have occurred in the absence of human-caused CO2 emissions

Natural factors, El Chichón eruption (1982) and Mt Pinatubo eruption (1991) were far more significant.

http://www.euanmearns.com/wp-content/uploads/2016/01/uahratpacstratosphere.png

Reply to  Chris Hanley
May 13, 2026 2:54 pm

The stratosphere has a positive temperature gradient… where to they measure the temperature ??

As the atmosphere warms from natural solar influence, the tropopause will increase in altitude…

… that means that measuring the stratosphere at a set altitude will have a cooling trend.

Nothing to do with CO2.

jshotsky
May 13, 2026 2:47 pm

Any paper that states that CO2 ‘traps’ heat is not worth reading, at all. Why? Because CO2 radiates CONTINUOUSLY as long as it is above 0 Kelvin. That means it loses energy constantly. That means it COOLS constantly, (inappropriate term for a molecule, but if the paper says CO2 can TRAP heat, I can say it LOSES heat.) It is all the INERT gases that trap heat. They can’t radiate it away. Inert molecules can ONLY lose heat energy through collisions.

Reply to  MyUsernameReloaded
May 13, 2026 4:10 pm

LOL..

CO2 DOES NOT TRAP HEAT.

The Earth’s is NOT an enclosed glass bottle.

Everything is your link is basically a highlight reel of the ANTI-SCIENCE behind the CO2 warming scam.

David Dibble has consistently shown WITH ACTUAL DATA that the bulk movement of energy is several magnitude larger than any possible theoretical CO2 effect.

CO2 is nothing but the equivalent of a flea bite on an elephants posterior.

roywspencer
Reply to  jshotsky
May 13, 2026 4:22 pm

But any gas that emits IR must also absorb IR. So, how much of each?… what is the balance between absorption and emission? The answer is GHGs absorb IR basically independent of temperature, but their emission of IR is strongly dependent upon temperature. For this reason, every layer of the troposphere is almost always in a state of IR radiative imbalance… the lower troposphere absorbs much more IR than it emits, while the upper troposphere emits much more IR than it absorbs The difference is made up by convective heat transport.

jshotsky
Reply to  roywspencer
May 13, 2026 4:59 pm

Wrong. If you could ‘see’ the radiation from CO2, it would look like a flashlight. It is ALWAYS radiating if above absolute zero. A photon can have any of three results with a collision with CO2 or any other greenhouse gas. It can be deflected, it can be reflected, and it can be absorbed. Those are the only choices – it has nothing whatsoever to do with temperature, given that it is above zero K.

May 13, 2026 2:50 pm

Using “climate models” by Manabe and Wetherald …… OOPS !!

Mathematical artefacts from using a “not-of-this-planet” atmosphere model.

Reply to  bnice2000
May 13, 2026 4:52 pm

Roy Clark is a physical chemist.

mh
May 13, 2026 3:07 pm

I see two problems with this article. Firstly, the theory of CAGW claims Earth warms because rising CO2 reduces energy loss to space so energy in > energy out. But we have been measuring energy loss to space (outgoing longwave radiation or OLR) since he start of the satellite era and it is steadily rising not falling directly conflicting with the core tenant of the CAGW thesis. Secondly, we can tell exactly where there CO2 radiation is coming from by the emission temperature (CO2 concentration is high enough to make it emit like a black body at the central emission wavelength). It is only the last 1-2 absorbance that can radiate to space, any emission from below that is reabsorbed by the CO2 above. This shows that emission to space from the two side bands comes from low down in the stratosphere – near the tropopause and is by far the greatest fraction of all CO2 emission to space. Because the absorptivity of the central wavelength peak is much stronger than the two side lobes the emission at this wavelength comes from a region where CO2 concentration is lower ie: higher in the stratosphere. But from the emission temperature it is clearly well below the stratopause. According to this experimental observation there is no significant CO2 at the stratopause. What there is,is ozone which emits strongly at 10 microns. Maybe its time to look at the contribution of ozone.

Nick Stokes
Reply to  mh
May 13, 2026 4:00 pm

“directly conflicting with the core tenant of the CAGW thesis”

A common error – that is not a tenet of AGW. Heat fluxes are governed by something simpler – conservation of energy. Heat flux out = heat flux in, minus any that was diverted to warming the oceans. AGW says that GHGs place a resistance in the path of outgoing flux, and so a larger temperature difference is required to get it through – hence surface warming. The flux need not change.

Here is a plot (from here) of TOA imbalance vs measured heat uptake by the ocean. They match pretty well. The fluctuations are mostly ENSO plus seasonal – the gradient represents the underlying warming of the ocean.

comment image

Reply to  Nick Stokes
May 13, 2026 4:13 pm

So the increase in absorbed solar radiation is causing the Earth to warm up a bit..

Hardly surprising !

Absorbed-solar-radiation
real bob boder
Reply to  Nick Stokes
May 13, 2026 5:31 pm

Nick

BS, there would by necessity be a lag between the 2. You have just provided more proof that the warming is not related to GHGs.

Bob
May 13, 2026 3:37 pm

This doesn’t make any sense. My understanding is that CO2 doesn’t trap heat rather it slows down its escape from the troposphere, heat rises, heat moves to cold, the temperature of the lower stratosphere is minus 60 degrees Fahrenheit, the upper stratosphere is minus 5 degrees Fahrenheit primarily due to ozone’s interaction with energy from the sun. If the heat/energy isn’t trapped and the temperature of the lower stratosphere is minus 60 degrees Fahrenheit it would seem to me that the extra heat has been handled.

Reply to  Bob
May 13, 2026 4:16 pm

Actual measurements show that any slight increase in absorption in the CO2 band is more than compensated by an increase in radiation out through the adjacent bands.

CO2 does not trap or slow anything !

radiative-change-2
May 13, 2026 3:40 pm

Once again, the magic molecule changes it’s physical behavior based on where it resides. In the ocean, it causes increased acidity. In the lower atmosphere, it traps heat by reflecting downward. A little higher, it releases heat by radiating upward. It is truly a wonder of modern physics.

jshotsky
Reply to  doonman
May 13, 2026 5:13 pm

Traps heat by reflecting downward??? Dude, when a photon is emitted (not reemitted) it would look like a flashbulb, not a ‘direction’. It is an omnidirectional ‘flash’. Some is earth directed, some is space directed. Lasers are designed to focus the radiation in one direction. Without that, they are flashbulbs.

May 13, 2026 3:51 pm

From the article: “However, though the basic principles of stratospheric cooling are understood, the specifics have remained cloudy. “The existing theory was incredibly insightful, but at the moment we lack a quantitative theory for CO2-induced stratospheric cooling,” says Sean Cohen, a postdoctoral research scientist at Lamont-Doherty Earth Observatory, which is part of the Columbia Climate School, and the study’s lead author.”

So, a little more work needs to be done.

Editor
May 13, 2026 4:05 pm

This reads like BS. “When more CO2 is added, the stratosphere radiates heat away more efficiently and it cools.”. If you feed a cooler stratosphere into a model whose only variable is CO2, then the model is incapable of finding any cause for the coolness other than CO2. So you tweak and tweak until you get the required result, but you have to studiously avoid any kind of reasonableness test if you want to publish. No problem, though, your reviewers are all in the same boat.

roywspencer
May 13, 2026 4:15 pm

Manabe and Strickler (1964) demonstrated this effect even with their less-thorough understanding of the IR emission/absorption characteristics of various gases, and all climate models with a stratosphere produce it. It’s no mystery. I don’t understand what they have done that is fundamentally new. In fact, without convection, the upper troposphere would also undergo massive cooling due to this effect. This has all been well understood for over 60 years.

May 13, 2026 4:23 pm

‘In the lower atmosphere, CO2 molecules trap heat that would otherwise escape into space. Higher in the atmosphere, though, the dynamics change. In the stratosphere—the atmospheric layer that extends from about 11km to 50 km above Earth’s surface—CO2 molecules function almost like a radiator, absorbing infrared energy from below and emitting some of that energy into space. When more CO2 is added, the stratosphere radiates heat away more efficiently and it cools.’

These guys are hopeless. For one thing, CO2 doesn’t spontaneously emit to space below the top of the Mesosphere, i.e., ~84 km above the Earth’s surface. Just another example of how climate ‘scientists’ continue to tweak the phenomenological physics of radiative transfer theory in a futile attempt to maintain their narrative that CO2 is the ‘control knob’ of the Earth’s climate system.

Curious George
Reply to  Frank from NoVA
May 13, 2026 5:49 pm

Higher in the atmosphere, CO2 molecules function almost like a radiator, absorbing infrared energy from below and emitting some of that energy into space.
I love this elementary explanation. Instead of letting the infrared energy simply pass through, the evil CO2 absorbs it, and then emits some of it into space. Absolutely devilish. Please, no mathematical formulas.

May 13, 2026 4:34 pm

Gemini says:
This study attempts to provide a neat mathematical bridge for a phenomenon predicted by Syukuro Manabe’s models in the 1960s. However, by tweaking equations until they fit pre-existing assumptions, minimizing the profound impacts of ozone and solar UV variability, and relying on a paradoxical thermodynamic explanation, the authors have not solved a mystery. Instead, they have merely manufactured a complex mathematical justification to keep the politically expedient “CO2 control knob” theory alive and guilt trip humans. 

May 13, 2026 4:39 pm

Harold The Organic Chemist Says:
CO2 Does Not Cause Warming Of Air!

Shown in the chart (See below) is a plot of the annual mean temperature in Adelaide from 1857 to 1999. In 1857 the concentration of CO2 in air was ca. 280 ppmv (0.55 g CO2/cu. m. of air),and by 1999 it had increased to ca. 368 ppmv (0.72 g CO2/cu. m.) but there was no increase in air temperature in this port city. Instead there was a slight cooling. In 1857 Tav was 17.2° C, and by 1999 it had declined to 16.7° C. Darwin had a similar cooling. Note how little CO2 there is in the air.

To obtain recent Adelaide temperature data, I went to:

https://www.extremeweatherwatch.com/cities/adelaide/average-temperature-by-year.

The Thi and Tlo temperature data from 1887 to 2025 was displayed in a long table. The computed Tav was 17.4° C. Since temperature measurement accuracy is +/- 0.1° C., it is concluded that there has been no warming of air in Adelaide for 168 years. This empirical data falsifies the claim by the IPCC that CO2 cause global warming. In 2025 the concentration of CO2 in was 427 ppm (0.84 g CO2/cu. m. of air).

The chart was taken from the late John L. Daly’s website:
“Still Waiting For Greenhouse” available at http://www.john-daly.com. From home page, go to end click on “Station Temperature Data”. On the “World Map” click on “Australia”. There is displayed a list of weather stations. Click on “Adelaide”. John Daly found over 200 weather stations that showed no warming up to 2002.

On May 1, I posted the above comment on Gavin’s website “RealClimate”, and the Real Climateers attacked me like a pack of
African hyenas. They called me a troll and a fool, and said I did not know what is was talking about. I shrugged off this vicious attack.

I ask everyone: Is my comment about CO2 not causing warming of air correct?

NB: If you click on the chart, it will become expand and become clear. Click on the “X” in the circle to contract the chart and return to Comments.

adelaide
May 13, 2026 5:03 pm

They need to collaborate with these guys to determine if and where there is cooling and heating…since one blindfolded person is feeling the trunk and the other the tail of the same elephant…
https://www.nature.com/articles/s41598-023-28222-x

E. Schaffer
May 13, 2026 5:15 pm

It comes as a surprise to me, this simple, self-evident mechanism was a “yet to be solved long-standing atmospheric puzzle”..?!

“Some wavelengths contribute to cooling more than others, and the team determined that wavelengths in a certain “Goldilocks zone” are especially efficient” – they found CO2 has an absorption band? Not so new!

“In the stratosphere.. CO2 molecules function almost like a radiator, absorbing infrared energy from below and emitting some of that energy into space. When more CO2 is added, the stratosphere radiates heat away more efficiently and it cools.”

Ouch! As CO2 increases the optical thickness of the stratosphere, it emits AND absorbs more IR. If one believes it was warmed by absorbing more terrestrial IR from below, while emitting more into space, then why would it cool? Should that not be zero-sum game? The problem is already evident in the terminology, calling it “infrared energy”. It is just infrared radiation. What they fail to comprehend is, though with more CO2 the stratosphere absorbs more IR from below, it also emits more IR downward – and that is the zero sum game.

CO2 cools the stratosphere because it emits IR into space, but absorbs no (or barely) SW-solar radiation. Add something that does absorb SW-solar, and the stratosphere warms..

May 13, 2026 5:43 pm

I did not purchase the paper, so who knows, there may be something in it that explains or supports it.

The first statement of the abstract makes this work highly suspicious:

“The cooling of the stratosphere in response to increasing carbon dioxide concentration is a fingerprint of human effects on climate.”

That sounds like a presupposed conclusion to which the paper will dutifully create some mathematics.

May 13, 2026 6:32 pm

Earth is cooler w atmos/water vapor/30% albedo not warmer.
Ubiquitous GHE balance graphics don’t + violate GAAP & LoT.
Kinetic heat transfer processes of contiguous atmos molecules render “extra” GHE energy from a BB surface impossible.
GHE = bogus & CAGW = scam.

Chuck Higley
May 13, 2026 7:53 pm

As the upper tropical troposphere is where this heat as supposed to be sourcing (according to the IPCC), there is a big problem. This region is -17 deg C and the surface 15 deg C. The upper tropical troposphere simply CANNOT warm the surface, being colder than the surface. Thermodynamically impossible. Any IR radiation from the upper troposphere will be reflected (rejected) by the surface as the energy levels for -17 deg C are already full.

For that matter, CO2 has three IR absorption bands, equivalent to -80, 400, and 800 deg C blackbody radiation. The later two are not used anywhere on Earth as CO2 never naturally becomes this how and thus not used in the troposphere. The first (-80 deg C) is colder than everything on the planet and thus rejected by everything. In fact, CO2 works 24/7 at cooling the planet by sending -80 deg C IR out to space. Physicists have not been able to get CO2 to warm anything and discovered that it is the world’s best refrigerant, being abundant, cheap, stable, and nontoxic. It is now being used as the primary refrigerantt in new ice rinks and Mercedes car A/C units. Basically, this paper has the science completely wrong, particularly as it is based on models.

Furthermore, CO2 in the upper tropical troposphere is just not abundant enough to have the effect claimed. It’s just a nonstarter, like using a picket fence with slats one foot apart to corral mice.