The Goddess Gaia admires Earth’s clouds in a moment of relax. Clouds are not only an important factor influencing Earth’s climate, they are beautiful, elegant, and everchanging.
You may have heard that clouds are a weak spot in our understanding of Earth’s climate. That’s true, and it is a good illustration of how we Earthlings find it difficult to understand things remote from our everyday experience. It is so difficult that even intelligent people, such as Nobel Prize winner John Clauser, can misunderstand it completely, thinking that clouds act as a thermostat that regulates Earth’s temperature. For a review of the main points of Clauser’s proposal, see Sabine Hossenfelder’s video. But, although the clip nicely demolishes Clauser’s arguments, it doesn’t do a very good job of explaining the interaction of clouds with climate.
You know that Nature is a bitch as a teacher (ahem… sorry Gaia). First you have to pass the exam, then she’ll teach you the right answers. But making mistakes is a good way to learn a lesson. So, let’s see if we can disentangle the matter of what clouds do and don’t do to Earth’s climate.
Clouds — what are they?
We all recognize a cloud when we see one, and when we are inside a cloud, we call it “fog.” But is a cloud a thing? In a sense, no.
A cloud is an assembly of water droplets that collect in a region of the atmosphere when a rising mass of wet air becomes cold enough that the water in it condenses into droplets. These droplets are heavier than air, but they are so small that they fall slowly because of viscous drag (it is called “Stokes' law”). Whenever you see a cloud in the sky, you are seeing a mass of warm air rising.
But clouds can also fall either because droplets became large — in that case we have “rain” — or because the warm air column that created them fades out. In that case, a cloud can majestically roll down a slope while becoming fog. Here is an image of clouds rolling down from the Twin Peaks toward San Francisco’s Bay. A wonderful spectacle.
Clouds and climate
We know that clouds partially block sunlight, and that must have a cooling effect. But the story is much more complicated than that.
First of all, the interaction of sunlight with clouds involves a mechanism unlike anything you normally interact with in your everyday experience. You are surrounded by opaque solid objects that absorb light at their surface and re-emit part of it. Clouds, instead, absorb sunlight through another mechanism called “Mie scattering.” It means that photons change direction when they interact with water droplets, but they are not absorbed — it is called “elastic scattering.” Photons are scattered in all directions, mainly forward, but a fraction is scattered back upward. So, clouds are not opaque; it is just that less light arrives at the bottom.
Remarkably, clouds scatter visible light more or less in the same way at all wavelengths; so they appear pure white. By the way, there is another mechanism called Rayleigh scattering, which operates at the molecular level, and it is why you see the cloudless sky as blue. If you ever looked at clouds from the window of a plane, you saw how white and fluffy they are. (image by Diego Martins)
How large the fraction of light that clouds send back into space depends on how dense and how thick the cloud is. There are many kinds of clouds; meteorologists distinguish ten basic kinds but with many more subdivisions. In her clip, Sabine Hossenfelder has some fun with the “Mammatus” features, which look like mammalian breasts.
But that’s a special case. Here is an image (from NASA) of the most common kinds of clouds.
You can see how a “Cumulonimbus” is a very thick cloud; it may reach 10 km. As you would expect, a thick cloud scatters more solar light, and it appears darker from below. Nevertheless, even the darkest thunderstorms never are as dark as night.
Radiative Thermal Effects of Clouds
So, now we arrived at something that has to do with temperature: high clouds are less dense because there is less water at high heights. And they are also thinner for the same reason. So, they scatter upward less light than thick and dense low clouds. So, you could say that high clouds have a lower albedo than low clouds, and therefore a smaller cooling effect.
Fine, but there is more. You may have read that high clouds warm the atmosphere while low clouds cool it. Now, that’s a tough question: granted that high clouds reflect less light than low ones, why should their thermal balance tilt to warming? You can’t solve the riddle until you get to the other part of Earth’s radiative balance: infrared radiation. Look at this image from NASA.
First, note how the effect of clouds on infrared (longwave) radiation is similar that of greenhouse gases: they warm the atmosphere by scattering back part of the emitted radiation. That explains why the overall radiative forcing effect of clouds is not just cooling (as Clauser mistakenly assumed) but also warming. Typically, the two effects balance each other, and the global result seems to be a slight cooling. But there is no such thing as a “Cloud Thermostat” — as Clauser proposed — generated by clouds reflecting sunlight into space.
But we haven’t yet answered the question of why high clouds warm Earth while low clouds cool it. For that, you have to go more in-depth into the mechanisms of heat transfer in the atmosphere.
First, cloud infrared scattering is a very different story in comparison to the greenhouse effect. Greenhouse gases absorb light only in some specific regions of the IR spectrum. Here, you see Earth’s emission spectrum. (image from Wikipedia, by Robert Wentworth CC BY-SA 4.0).
You probably saw this image already. Note how the main greenhouse gases absorb IR radiation in a narrow region of the spectrum. Note also the “Infrared window” on the right side of the CO2 band in the figure. In that region, there is almost no greenhouse absorption, and Earth can happily scatter excess heat into space.
But the image is for a clear sky; No clouds are involved. What happens if we add clouds? Well, clouds interact with IR radiation very much like they do with solar light. They scatter it back, in part, by the Mie effect. Now, remember what we said: the Mie effect doesn’t depend so much on wavelength; it is about the same across the whole spectrum. So, the effect of clouds on the infrared emission spectrum is simply to bring down the emission intensity in the whole infrared spectrum. It means that Earth emits less, therefore it warms. That’s another way to understand the warming effect of clouds.
Now, one more effort to explain why high clouds warm the atmosphere. Pay attention because this is a tricky point.
First, take into account how air density changes with height (image from Wikipedia)
Here, pressure is proportional to density, the concentration of CO2 varies approximately in the same way. At 5 km of height, the CO2 density in terms of molecules per unit volume is about one-half of that at the ground level, and this halves its capability to absorb IR radiation. It means that at low height practically all the IR in the CO2 absorption band is absorbed and then turned into vibrational energy; almost nothing is emitted into space. But at higher heights, a progressively larger fraction of this IR radiation manages to reach space and disappear, cooling the atmosphere.
Now, consider how a low cloud blocks IR in the region of absorption of CO2. That changes very little to the outgoing IR spectrum because the IR is practically all absorbed by CO2. So, the effect on the radiative balance (and hence on temperature) is small.
Now, consider a high cloud. It does the same, but at a height when the CO2 concentration has become small enough that it can’t absorb so much radiation of the wavelengths corresponding to its absorption band. It means that the cloud prevents some IR radiation from escaping into space, as it would otherwise do without the cloud. Hence it has a warming effect that low clouds do not have.
Non-radiative Cooling
The above is not the whole story. It describes only how clouds affect climate through radiative transfer. However, heat moves in the atmosphere through convection and phase transitions. Convection is the movement of air masses that transfer heat to the high atmosphere, where it can be dissipated more efficiently than from Earth’s surface. The same effect occurs for phase transitions generated by clouds. They act as “heat pumps." I described this phenomenon in a previous post.
Does that mean that, after all, clouds do cool the Earth, and that by chance Clauser got the right answer by the wrong reasoning? Could it be that a large fraction of the warming we are seeing is due to deforestation, and hence less low height clouds, rather than to the radiative effect of greenhouse gases? This is the thesis proposed by Anastassia Makarieva, who, unlike Clauser, arrived at this idea not by chance but by the right kind of reasoning. You can read Anastassia’s proposal in her super-interesting blog.
So, how important is the convective/phase transition cooling effect of clouds? It is a general problem in complex systems. When two different factors push the system in the same direction, it is hard to disentangle the respective contributions. How much of the current global warming is due to deforestation, and how much to CO2 emissions? For the time being, we have to be very cautious. What we can say is that there is little paleoclimatic evidence that links forests to cold periods, especially in relatively recent times such as the Miocene. So, it would seem that, overall, forests are a minor factor in determining Earth’s climate. But, as for everything that has to do with climate, always expect surprises. Nothing can change so quickly as a complex system, a statement that also holds for our perception of it.
Great blog, would like to emphasize once more that the changes in temperature globally have been in large part caused by deforestation in the tropics which have a far more detrimental effect than those at higher latitudes because of their higher capacity to cool to offset the far more intense amount of energy coming in at the equator. We have done the calculations in this paper: https://medcraveonline.com/IJBSBE/IJBSBE-09-00237.pdf, based on which we wrote the book Cooling the Climate. https://ethicspress.com/products/cooling-the-climate (happy to send a PDF to anyone interested) as the fastest way to stabilize weather and climate and avert the tipping points which are closing in.
Excellent blog. On low level clouds, the current suggestion is that the more than expected recent warming is due to reduced low level clouds.
https://insideclimatenews.org/news/05122024/reflective-low-clouds-decline-may-contribute-to-record-heat/
On deforestation, there has been a suggestion that the Little Ice Age was at least partially caused by the depopulation of Latin America after the measles epidemic introduced by los conquistadores. The steep population decline led to spontaneous reforestation thereby reducing atmospheric CO2. I doubt it's true but it's a great theory.
Now too, since microplastics have been detected in clouds, it's suggested this could be a factor in climate change too.
Joni Mitchell was right, we really don't know clouds at all.