Uncertain Clouds

Guest Post by Willis Eschenbach

I read an interesting quote in the latest Climanifesto from the Intergovernmental Panel on Climate Change, their Sixth Assessment Report, known as “IPCC AR6” to the initiates:

Climate Feedbacks and Sensitivity IPCC AR6 WGI Section 7.7

The net effect of changes in clouds in response to global warming is to amplify human-induced warming, that is, the net cloud feedback is positive (high confidence).

Compared to AR5, major advances in the understanding of cloud processes have increased the level of confidence and decreased the uncertainty range in the cloud feedback by about 50%. An assessment of the low-altitude cloud feedback over the subtropical oceans, which was previously the major source of uncertainty in the net cloud feedback, is improved owing to a combined use of climate model simulations, satellite observations, and explicit simulations of clouds, altogether leading to strong evidence that this type of cloud amplifies global warming.

The net cloud feedback, obtained by summing the cloud feedbacks assessed for individual regimes, is 0.42 [–0.10 to 0.94] W m-2 °C–1. A net negative cloud feedback is very unlikely. (high confidence)

The idea of global net positive cloud feedback always seemed very unlikely to me. In part this is because I lived for years in the tropics, spending much time outdoors. When it gets warmer in the tropics, cumulus clouds form, reflect lots of strong tropical sunshine back to space, reduce the incoming energy, and thus cool the surface. And if warming still continues, thunderstorms form, which cool the surface in a host of ways. So I’ve watched cloud feedback happen day after day, seeing more warming leading to more cloud-based cooling, not to amplified cloud warming. And the tropics is a large chunk of the planet.

Now, clouds have two opposite radiative effects on the surface. They reflect solar shortwave radiation back out to space, cooling the surface. And clouds also both absorb and emit longwave (thermal) infrared radiation, leaving the surface warmer than when there are no clouds.

This is not theoretical. You can feel the shortwave effect on a clear summer day when a cloud comes over and leaves the surface cooler than with no cloud. You can also feel the longwave effect on a clear winter night when a cloud comes over and leaves the surface warmer than with no cloud.

[For those objecting that downwelling longwave radiation from cold clouds can’t leave the surface warmer than if there are no clouds, please see my post “Can A Cold Object Warm A Hot Object“.]

The sum of these two radiative effects, shortwave cooling and longwave warming, is called the “net cloud radiative effect” or “net CRE”. If it is positive, the clouds leave the surface warmer, and if it is negative, the clouds leave the surface cooler, than in their absence.

So I turned once again to the CERES satellite-based dataset to see what I could learn. It contains data on the net surface net cloud radiative effect. Figure 1 shows the net radiative effects of clouds (“net CRE”) around the world.

Figure 1a. Net surface CRE, Atlantic view
Figure 1b. Net cloud radiative effect, Pacific view

There are several interesting things regarding the effects of the clouds shown above. First, on average they cool the surface by about twenty watts per square meter (W/m2). Next, clouds warm the poles by about the same amount, twenty W/m2 or so. And clouds cool the ocean about three times as much as they cool the land.

How is the net CRE related to the temperature? We can look at that in a couple of ways. Figure 2 below shows a scatterplot of average net CRE versus average surface temperature.

Because it is using 21-year averages, this type of analysis has the great benefit of including all feedbacks and slow-acting processes. These are the gridcell temperatures that each gridcell has equilibrated to over decades, after the net water vapor feedback and the cloud feedback and any other feedback have had their effect. Thus, this gives us a good idea of the long-term net cloud feedback at various temperatures.

Figure 2. Scatterplot of 1° latitude by 1° longitude gridcell average temperatures and net cloud radiative effect.

In Figure 2, the trend at any temperature is given by the slope of the yellow/black line. It shows on average how much the CRE changes for a given change in temperature. From this, we can see several things. First, at the coldest temperatures the slope is positive—Antarctic clouds lead to warming. But above temperatures of about -20°C, the general trend of the net cloud feedback is negative.

Next, above about 26°C, which is about 30% of the planet, the net cloud feedback is extremely negative. For each additional degree of warming, the net cloud radiative effect decreases by tens of watts.

Finally, there are two areas where the net cloud feedback is positive—where the average temperature is below -20°C (mostly the Antarctic peninsula) and where it is between 15°C and 25°C (temperate zone)

There’s another way that we can look at the long-term net cloud feedback. It also involves the looking at the same gridcell averages of temperature and net CRE, but in a different way. The method involves looking at the area around each gridcell, to see what the trend is at that gridcell.

The logic behind the method is that if we look at some given gridcell, it has an average temperature and an average cloud radiative effect. And if we want to see what happens if the average temperature is 1°C higher or lower, we can look at the surrounding gridcells to see what’s happening at different temperatures in that local area.

For example, here are a couple of typical patches of Pacific Ocean area, each one measuring 9° latitude x 9° longitude.

Figure 3a. Temperature of a 9° square patch of the Pacific.
Figure 3b. Net cloud radiative effect of a 9° square patch of the Pacific.

As you can see, in that part of the Pacific the correspondence between ~ steady-state average temperatures and ~ steady-state average net CRE is strongly negative. I calculate the trend, and assign it to the central gridcell of the block. I repeat the process for each of the world’s 64,800 gridcells, examining what’s happening in the local area, and that gives me the global map shown in Figure 4 below.

Figure 4a. Change in cloud radiative effect from a 1°C increase in surface temperature, Pacific View. White/black lines show the global average, about -2 W/m2.
Figure 4b. Change in cloud radiative effect from a 1°C increase in surface temperature, Pacific View. White/black lines show global average, about -2 W/m2.

Some things of note. The net cloud feedback is positive over the land, and negative over the ocean. In agreement with slope of the yellow/black line in Figure 2, the Antarctic plateau and the temperature zones are the positive areas, while the tropics are negative. And as indicated in Figure 2, some parts of the warmest tropical ocean is strongly negative..

Finally, as a global area-weighted average, this analysis gives a global negative net cloud feedback of -1.9 W/m2 per degree of surface warming. Negative.

Yes, I understand that this is the exact opposite of what the climate model simulations, satellite observations, and explicit simulations of clouds” referenced by the IPCC say… but then, the IPCC is a political body, not a scientific body.

And more to the point, this analysis is based on what the earth is actually doing, not on “climate model simulations” which even the IPCC agrees are greatly flawed and uncertain on clouds.

And yes, it disagrees with the “scientific consensus” … when I was a kid we had asbestos ceiling tiles in our grade school because at that time the scientific consensus was that asbestos ceiling tile was 100% totally harmless. So you’ll excuse me if consensi don’t impress me much.

My very best wishes to all,


Endnote: This is only the radiative effects of clouds. In addition to the cloud radiative effects, clouds cool the surface in a variety of other ways.

• They increase the wind, which increases evaporation, which increases surface cooling.

• Wind also increases sensible heat loss from the surface.

• Wind over the ocean leads to an increase in surface albedo through the effects of white breaking waves, spray, and spume.

• Via both spray and waves, wind increases the surface area of the ocean, which leads to increased evaporative and sensible heat loss.

• Clouds lead to rain and snow, both of which have a strong cooling effect on the surface.

• Thermally driven condensing clouds are surrounded by slowly descending dryer air, which allows more radiation to escape to space.

• Wind increases oceanic surface overturning, which brings cooler deep water to the surface.

So the radiative cooling shown in the graphics above greatly underestimates both the total cooling effect and the total net negative feedback of the clouds.

PS: Can I say how bored I am with personal attacks, and with uncited claims that I’m wrong in some unspecified statement that I purportedly made somewhere, and with people repeating the opposing case without evidence … let me recommend to everyone my post “Agreeing To Disagree”, which contains the following graphic:

Graham’s Hierarchy of Disagreement

If you are commenting, please take a look at that hierarchy and determine where your comment fits … then see if you can move it up a level or two …

Most critically, the author of that graphic said:


The most convincing form of disagreement is refutation. It’s also the rarest, because it’s the most work. Indeed, the disagreement hierarchy forms a kind of pyramid, in the sense that the higher you go the fewer instances you find.

To refute someone you probably have to quote them. You have to find a “smoking gun,” a passage in whatever you disagree with that you feel is mistaken, and then explain why it’s mistaken. If you can’t find an actual quote to disagree with, you may be arguing with a straw man.


That, in part, is why I always ask people to quote the exact words you are discussing.

Like this:

Like Loading…


via Watts Up With That?


September 17, 2021 at 12:39PM

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s