Guest Post by Willis Eschenbach
Following up on a suggestion made to me by one of my long-time scientific heroes, Dr. Fred Singer, I’ve been looking at the rainfall dataset from the Tropical Rainfall Measuring Mission (TRMM) satellite. Here’s s the TRMM average rainfall data for the entire mission to date:
Figure 1. Average annual rainfall, metres per year, as measured by the TRMM satellite. The TRMM satellite only measures from 40°N to 40°S, hence the “Tropical” in the name. Data source: KNMI: Click on “Monthly Observations”, click the TRMM data checkbox and then click “Select field” at the top of the page. When the page comes up, the NetCDF file link is at the very bottom of the page
Note the horizontal yellow/red area generally girdling the planet above or around the equator. This is the average position of the intertropical convergence zone (ITCZ). The ITCZ is the location of the energetic deep tropical circulation that powers the great atmospheric Hadley circulation. As you can see, the ITCZ is the wettest large area of the planet, with over 4 metres (13 feet) of rain in some areas. Although this is the average position, it moves during the year. You can see in the Pacific south of the Equator near South America the position it takes over part of the year, as a “ghost” of the average position parallel to the Equator.
Now, the TRMM dataset is fascinating in and of itself, but I was interested in it for a specific reason.
My hypothesis is that the earth has a thermoregulatory system keeping the global temperature within narrow bounds (e.g. ±0.3°C over the 20th century). A major part of this thermoregulatory system is that the tropical cumulus and thunderstorms act to limit the tropical temperatures on both the warm and cool ends. Generally in the tropics, when a morning is cooler than usual, cumulus clouds form later in the day and they are weaker. The same is true of the thunderstorms. On cool days, thunderstorms form later than average or not at all. As a result of the reduction of clouds and thunderstorms, the surface is strongly warmed by the sun, and there is reduced loss of surface energy via the various thunderstorm mechanisms.
On days that are warmer than average, the reverse is true. There is an early and strong development of the tropical cumulus field. In addition, the cumulus formation is also earlier and stronger. Both of these act to cool the surface, with both the cumulus and the thunderstorms able to not only slow the warming, but actually cool the surface below their initiation temperatures. I describe the entire daily cycle in my post Emergent Climate Phenomena.
A hypothesis requires observational support, of course. Now, with such a system, according to my hypothesis as the tropical temperature rises, the albedo should go up, and the thunderstorms should also increase. Do the observations support this?
As I’ve discussed before, the CERES satellite radiation dataset lets us test the first of these consequences of my ideas. If my hypothesis is true, in the tropics, we should see a positive correlation of the albedo and the temperature. Here is that relationship on a 1°x1° gridcell basis:
As is predicted by my hypothesis, in the tropics and particularly in the areas just north and south of the Equator where the ITCZ wanders around, there is a strong positive correlation of albedo with temperature.
However, I’d been unable to get any global handle on the effects of the thunderstorms until I started looking at the TRMM. Across the tropics in general and particularly in the ITCZ, the rainfall is from thunderstorms, towering storms that drive the deep tropical convection. Having the rainfall data allows me to do the same thing I did with the albedo—see how the rainfall varies with the temperature. IF my hypothesis is correct, tropical rainfall should increase with temperature, particularly in the ITCZ areas. Figure 3 shows the correlation of rainfall with temperature:
Figure 3. As in Figure 2, correlations. This shows the correlation of rainfall and temperature. A positive correlation means that when temperature increases, so does rainfall, and vice versa.
Now, in the yellow to red sections, as the temperature increases the rainfall increases. As always, there are mysterious and interesting things revealed by any new observational dataset. In this case, the rainfall increases with temperature in the ocean and in the drier parts of the land. But in the wetter parts of the land such as tropical Africa and the Amazon, the rainfall is about neutral or actually decreases with respect to temperature .. go figure.
In any case, over the tropics in general (shown by dashed lines at 23.5°N/S of the Equator) the correlation is generally positive. So for the tropics, my hypothesis is indeed verified—increasing temperature leads to increasing thunderstorms. You can also see how the extra-tropical areas in general are more negatively correlated than is the tropics.
The TRMM dataset also allows us to see not only the correlation, but how much actual change in rainfall we are talking about. Figure 4 shows the change in rainfall per degree C of warming.
Now, this is indeed interesting … across the tropics, on average we get 22 mm/year more rain for each degree of surface temperature increase. And since the trends have the same signs as the correlations, as with the correlations the areas north and south of the tropics generally show falling rainfall with increasing temperatures.
Here’s the beauty part. I realized that we can use these TRMM rainfall figures to estimate the amount of energy involved. The main cooling mechanism of thunderstorms is evaporative cooling. We can calculate the energy involved in that evaporative cooling by noting that to reverse the old saying, with rainfall it’s “what comes down must go up” … meaning that whatever water rains down, it had to be evaporated first. To evaporate a cubic metre of seawater in one year takes a constant energy flux of about 80 W/m2. This works out to about 0.08 W/m2 to evaporate one mm of rain. So let me use that conversion, 0.08 W/m2 of evaporative cooling per millimetre of rain, to show the same TRMM data from Figure 1 in terms of the energy needed to annually evaporate the amount of water of the gridcell annual rainfall.
I was pleasantly surprised by the very large amount of energy being moved constantly by thunderstorms in the ITCZ and elsewhere. In parts of the ITCZ, the evaporative cooling effect is well over 300 W/m2 … we can compare that to the cooling effect of the earths variable albedo, in W/m2, shown below in Figure 6.
Figure 6. Surface cooling from the reflection of the solar energy by the earth’s albedo. This includes both surface and cloud reflections. I note in passing the odd equality of the mean hemispheric reflections.
Taken together these last two graphs show something interesting. In the tropics, the average surface cooling from evaporation is about the same as the cooling from albedo reflection of solar energy. Both are about 90 W/m2. However, the variation in thunderstorm evaporative cooling (Fig. 5, from 0 – 375 W/m2) is much larger than the variation in reflected energy (Fig. 6, from 50 -175 W/m2).
To assess the instantaneous strength of these cloud and thunderstorm thermoregulatory mechanisms, we must bear in mind that these are annual averages. Even in the ITCZ it doesn’t rain all the time. So when it is raining, the effect would be much larger. How much larger? Well, a lot. My guess from living in the tropics near and in the ITCZ is that you might be under a thunderstorm maybe 5%-10% of the time on an annual basis … and if that’s the case, then the instantaneous evaporative cooling effect of individual thunderstorms will be about 10 to 20 times larger than the annual averages shown in Figure 5.
Let me move on to the question of what happens to the cloud reflective cooling and the evaporative cooling as the surface warms. This can be calculated as the trend of evaporative cooling in W/m2 for each additional degree C of warming. This is the same rainfall trend data shown in Figure 4, but expressed as the energy needed to evaporate that amount of rain. Figure 7 shows the change in surface evaporative cooling in watts/m2 per degree C of warming (W/m2/°C):
Figure 7. Change in thunderstorm evaporative cooling with surface warming, in W/m2 per degree celsius of surface warming. Negative values indicate a reduction in evaporative cooling with increasing temperatures.[NOTE: Figure updated.]
It is important to note that this average of the surface-cooling effect of the clouds and thunderstorms hides the fact that the effects are not applied evenly across an area. Instead, the clouds and thunderstorms form only over the warmer areas, and move huge amounts of energy out of those warmer areas and up into the troposphere. As a result, their efforts are concentrated exactly where and when they are needed. Cooling is applied only when and where the surface is warmer, and warming is applied only when and where it is cooler.
To close the circle, we can compare the amount of change in evaporation (Figure 7) per degree of warming with the change in reflective cooling per degree of surface warming (Figure 8 below). Figure 8 shows the change in albedo per degree of warming times the gridcell annual average downwelling solar.
Near the poles there is a strong negative correlation between albedo and temperature, meaning that there is reflective warming (negative values in Figure 8). However, because the sun is so weak in those areas the additional warming in actual watts per degree of temperature rise is not that large.
Figure 8 also shows that as the surface warms, the change in W/m2 of tropical evaporative cooling per degree C of warming is about twice that of the change in W/m2 of reflective cooling (Fig. 8, 1.2 W/m2 tropical increase in solar reflections per °C of warming, versus Fig. 7, 2.2 W/m2/°C increased evaporative cooling)
Finally, note that albedo changes and evaporative cooling are only two of the ways that clouds and thunderstorms cool the surface. As a result, the effect will be slightly larger than the numbers above indicate. I append a more complete list in the notes.
So that’s why I wanted to look at the TRMM data. I wanted to determine if my hypothesis about thunderstorms increasing with temperature is correct. And in the event, it appears that my hypothesis has been totally supported by the results.
• This is a significant addition to the variety of evidence that I’ve amassed showing that the earth indeed has strong thermoregulating mechanisms (see links below). It gives us an idea of the size of the cooling effect due to the tropical thunderstorms, along with actual values for the increase in thunderstorm cooling with increasing temperature.
• As is consonant with my hypothesis, both the tropical albedo and evaporative cooling increase with temperature, especially around the ITCZ.
• The evaporative cooling effect in the ITCZ is on the order of hundreds of watts per square metre. This is evidence of the strength of my hypothesized thermoregulatory mechanism.
• The change in evaporative cooling in the ITCZ is on the order of 10-20 W/m2 more evaporative cooling per degree celsius. This is evidence of the thermally responsive nature of the thermoregulatory mechanism.
• The thermoregulation from tropical clouds and thunderstorms occurs on a daily and hourly basis, not on the yearly basis shown in the Figures. As a result, we know that the instantaneous changes from clouds and thunderstorms are many times larger than the averages shown above. In addition the clouds and thunderstorms only emerge in response to local high temperatures, so their effect is not averaged across space and time as is shown in the Figures. The result is that the thermoregulatory system is applying cooling on the order of hundreds of watts/m2, but not blindly—the cooling is focused only where and when it is needed, towards the warmer sections of the local areas.
My best to you all at one am of a foggy night,
To Avoid Misunderstandings: Let me request that if you disagree with someone, you quote the exact words you disagree with. This lets all of us understand just what you are objecting to.
TRMM Data: For convenience I’ve placed the KNMI TRMM netCDF file here
Ways Other Than Albedo and Evaporation That Thunderstorms Cool The Surface
Cold rain and cold wind. As the moist air rises inside the thunderstorm’s heat pipe, water condenses and falls. The water starts out at the very cold or freezing temperatures aloft. As a result, it cools the lower atmosphere it falls through, and it cools the surface when it hits. The falling rain also entrains a downwards wind which is strongly cooled by the evaporation of the falling raindrops. When it strikes the ground, this cold wind blows radially outwards from the center of the falling rain. Because it is much cooler than the surrounding air, this radial wind runs along the ground, cooling the surrounding area. When I lived in the tropics, at night this wind was often the first indication of a nearby thunderstorm, as it outpaces the rain. It smells wonderful, crisp like the pure upper air … and best of all, on a muggy tropical night it blows through the open windows of all the houses and chills the entire area surrounding the rain.
This combination of cold rain and cold wind could be a shocking change, particularly when I’d be running an open skiff across the ocean at night. The temperature would go from a warm tropical night before I’d hit the thunderstorm, to cold and shivering under the storm, and back into the warmth once I’d run clear of the storm. Not fun. Well, yeah, fun, but cold fun …
Modification of upper tropospheric ice crystal cloud amounts (Lindzen 2001, Spencer 2007) . These clouds form from the tiny ice particles that come out the top of the smokestack of the thunderstorm heat engines. It appears that the varying amounts of this type of clouds has a large radiative effect, as they are thought to warm (through IR absorption) more than they cool (through reflection).
Enhanced night-time radiation. Unlike long-lived stratus clouds, tropical cumulus and cumulonimbus often die out and vanish as the night cools, leading to the typically clear skies at dawn. This allows greatly increased nighttime surface radiative cooling to space.
Drying of the bulk atmosphere. Thunderstorms move huge amounts of air vertically at a rapid rate. During the ascent, almost all the water vapor is stripped from the rising air column and falls as rain. After exiting the top of the thunderstorms, the now-dry air descends in the area around and between the thunderstorms. And because this air is dryer than it would be without the thunderstorms, the reduced levels of water vapor allow for increased longwave radiative surface cooling in the bulk of the atmosphere between the storms.
I haven’t even attempted a back-of-the-envelope calculation of the global average size of those effects. And of course, having mentioned it, I now have to give it a shot. Rats, I thought I was almost done with this post … here we go.
The effect of the cold rain, well, in the tropics if the temperature is say 26°C and the rain is maybe at 10°C when it hits the ground, for each cubic metre of rain that’s about 2 W/m2 of cooling effect. (I use my rule of thumb, that 1 W/m2 over 1 year heats 1 cubic metre of water by 8°C.)
I couldn’t even guess the amount of change from the Iris Effect per degree of surface warming. I’ll leave that to the good Drs. Lindzen and Spencer.
The night-time radiation … the night-time cloud radiative effect is entirely longwave, and is about 26 W/m2 of warming. So if there’s say a 10% decrease in night-time clouds that would also be one or two watts/m2.
Finally, the drying of the bulk atmosphere. This is a tough one, in part because the maximum daytime drying will likely occur around the afternoon peak in the temperature, and will be at a minimum around dawn when it’s cool. Hang on, I’ve got an idea … ok, MODTRAN says that in the tropics, if I set the water vapor to zero it lets an additional 58 W/m2 through the atmosphere.
So if the drying of the bulk atmosphere is on the order of 10%, it would be a cooling effect of about 6 W/m2.
Between these three, then, we have a total cooling effect of somewhere around 10 W/m2 on a 24/7 basis. Compare this to the thunderstorm evaporative effect, which Figure 5 says in the tropics averages about 90 W/m2. Thus, it appears that these secondary effects increase the total thunderstorm evaporative effects by on the order of 10%.
There is one more factor that increases thunderstorm cooling, but generally is not occurring on the above Figures. This is when a storm is delivering freezing rain and snow. In that case, the latent heat of fusion also needs to be considered. This is the energy needed to melt the ice at the surface. Energy to melt ice is about an eighth of the energy needed to evaporate the same amount of water. So for polar storms with snow, sleet, hail, or graupel, they would have a total cooling effect about 10% greater than a storm delivering the same amount of rain.
Further Reading About My Thermoregulatory Hypothesis
The Thermostat Hypothesis 2009-06-14
Abstract: The Thermostat Hypothesis is that tropical clouds and thunderstorms actively regulate the temperature of the earth. This keeps the earth at an equilibrium temperature.
Which way to the feedback? 2010-12-11
There is an interesting new study by Lauer et al. entitled “The Impact of Global Warming on Marine Boundary Layer Clouds over the Eastern Pacific—A Regional Model Study” [hereinafter Lauer10]. Anthony Watts has discussed some early issues with the paper here. The Lauer10 study has been controversial because it found that…
The Details Are In The Devil 2010-12-13
I love thought experiments. They allow us to understand complex systems that don’t fit into the laboratory. They have been an invaluable tool in the scientific inventory for centuries. Here’s my thought experiment for today. Imagine a room. In a room dirt collects, as you might imagine. In my household…
For some time now I’ve been wondering what kind of new evidence I could come up with to add support to my Thunderstorm Thermostat hypothesis (q.v.). This is the idea that cumulus clouds and thunderstorms combine to cap the rise of tropical temperatures. In particular, thunderstorms are able to drive…
It’s Not About Feedback 2011-08-14
The current climate paradigm believed by most scientists in the field can be likened to the movement of balls on a pool table. Figure 1. Pool balls on a level table. Response is directly proportional to applied force (double the force, double the distance). There are no “preferred” positions—every position…
Well, after my brief digression to some other topics, I’ve finally been able to get back to the reason that I got the CERES albedo and radiation data in the first place. This was to look at the relationship between the top of atmosphere (TOA) radiation imbalance and the surface…
The Tao of El Nino 2013-01-28
I was wandering through the graphics section of the TAO buoy data this evening. I noted that they have an outstanding animation of the most recent sixty months of tropical sea temperatures and surface heights. Go to their graphics page, click on “Animation”. Then click on “Animate”. When the new…
Emergent Climate Phenomena 2013-02-07
In a recent post, I described how the El Nino/La Nina alteration operates as a giant pump. Whenever the Pacific Ocean gets too warm across its surface, the Nino/Nina pump kicks in and removes the warm water from the Pacific, pumping it first west and thence poleward. I also wrote…
In my last post, “Emergent Climate Phenomena“, I gave a different paradigm for the climate. The current paradigm is that climate is a system in which temperature slavishly follows the changes in inputs. Under my paradigm, on the other hand, natural thermoregulatory systems constrain the temperature to vary within a…
I’ve mentioned before that a thunderstorm functions as a natural refrigeration system. I’d like to explain in a bit more detail what I mean by that. However, let me start by explaining my credentials as regards my knowledge of refrigeration. The simplest explanation of my refrigeration credentials is that I…
Dehumidifying the Tropics 2013-04-21
I once had the good fortune to fly over an amazing spectacle, where I saw all of the various stages of emergent phenomena involving thunderstorms. It happened on a flight over the Coral Sea from the Solomon Islands, which are near the Equator, south to Brisbane. Brisbane is at 27°…
Decadal Oscillations Of The Pacific Kind 2013-06-08
The recent post here on WUWT about the Pacific Decadal Oscillation (PDO) has a lot of folks claiming that the PDO is useful for predicting the future of the climate … I don’t think so myself, and this post is about why I don’t think the PDO predicts the climate…
Stalking the Rogue Hotspot 2013-08-21
[I’m making this excellent essay a top sticky post for a day or two, I urge sharing it far and wide. New stories will appear below this one. – Anthony] Dr. Kevin Trenberth is a mainstream climate scientist, best known for inadvertently telling the world the truth about the parlous…
The Magnificent Climate Heat Engine 2013-12-21
I’ve been reflecting over the last few days about how the climate system of the earth functions as a giant natural heat engine. A “heat engine”, whether natural or man-made, is a mechanism that converts heat into mechanical energy of some kind. In the case of the climate system, the…
The Thermostatic Throttle 2013-12-28
I have theorized that the reflective nature of the tropical clouds, in particular those of the inter-tropical convergence zone (ITCZ) just above the equator, functions as the “throttle” on the global climate engine. We’re all familiar with what a throttle does, because the gas pedal on your car controls the…
The CERES data has its problems, because the three datasets (incoming solar, outgoing longwave, and reflected shortwave) don’t add up to anything near zero. So the keepers of the keys adjusted them to an artificial imbalance of +0.85 W/m2 (warming). Despite that lack of accuracy, however, the CERES data is…
Dust In My Eyes 2014-02-13
I was thinking about “dust devils”, the little whirlwinds of dust that you see on a hot day, and they reminded me that we get dulled by familiarity with the wonders of our planet. Suppose, for example, you that “back in the olden days” your family lived for generations in…
The Power Stroke 2014-02-27
I got to thinking about the well-known correlation of El Ninos and global temperature. I knew that the Pacific temperatures lead the global temperatures, and the tropics lead the Pacific, but I’d never looked at the actual physical distribution of the correlation. So I went to the CERES dataset, and…
Albedic Meanderings 2015-06-03
I’ve been considering the nature of the relationship between the albedo and temperature. I have hypothesized elsewhere that variations in tropical cloud albedo are one of the main mechanisms that maintain the global surface temperature within a fairly narrow range (e.g. within ± 0.3°C during the entire 20th Century). To…