The Cloud Radiative Effect, Take Two

Guest Post by Willis Eschenbach

Well, in my last post I took a first cut at figuring the cloud radiative “feedback” from the CERES dataset. However, an alert commenter pointed out that I hadn’t controlled for the changes in solar radiation. The problem is that even if the clouds stay exactly the same, if the solar radiation increases, the net cloud radiative effect (CRE) increases due to increased reflection … and I hadn’t thought about that, had I? Dang … so my post was wrong.

So, to control for solar radiation, I did a multiple linear regression. The dependent variable was the net CRE, and the independent variables were the surface temperature and the solar radiation. As you might expect, this gave smaller results than my first analysis. I believe that this method is correct, but I’m always willing to be shown wrong. Not happy to be … but willing to be.

change in cloud radiative effect per one degree goodFigure 1. Net CRE as a function of surface temperature, after controlling for solar radiation. The gray lines are contour lines at zero W/m2 per °C. I suspect that the blue around Antarctica is an artifact due to the presence of the sea ice edge. 

Note that there are several areas in the tropical oceans which have a strong negative change in radiation with respect to temperature. These are the areas of the Inter-Tropical Convergence Zones, about ten degrees both north and south of the Equator. It is in these areas that much of the regulation of global temperature takes place, by means of the combined effect of cumulus clouds and thunderstorms.

In addition, there is a large area of the Southern Ocean where the clouds oppose the temperature rise.

The area of clouds off of the coast of California and northern Mexico is an area of persistent stratus that also strongly opposes warming. (See here for a discussion of this location in the literature).

Finally, I note that the global average change in net cloud radiation for each degree of surface warming is positive, at 0.7 W/m2 per degree. On reflection, it seems to me that we need to compare that to how much we’d expect the cloud radiation to increase if the surface temperature goes up by 1°C.

And I don’t know the answer to that … still pondering on that one.

Finally, it’s worth bearing in mind that the radiative effect of clouds is only the beginning of a long list of ways that clouds cool the surface. These include:

• Physically transporting heat from the surface directly to the upper troposphere where it radiates easily to space. Since the heat is transported either as latent heat, or as sensible heat inside the thunderstorm tower, it doesn’t interact with the large amount of water vapor, CO2, and other GHGs in the lower atmosphere.

• Wind driven evaporative cooling. Once the thunderstorm starts, it creates its own wind around the base. This self-generated wind increases evaporation in several ways, particularly over the ocean.

a) Evaporation rises linearly with wind speed. At a typical squall wind speed of 10 mps (20 knots), evaporation is about ten times higher than at “calm” conditions (conventionally taken as 1 mps).

b) The wind increases evaporation by creating spray and foam, and by blowing water off of trees and leaves. These greatly increase the evaporative surface area, because the total surface area of the millions of droplets is evaporating as well as the actual surface itself.

c) To a lesser extent, surface area is also increased by wind-created waves (a wavy surface has larger evaporative area than a flat surface).

d) Wind created waves in turn greatly increase turbulence in the boundary layer. This increases evaporation by mixing dry air down to the surface and moist air upwards.

e) Because the spray rapidly warms to air temperature, which in the tropics is often warmer than ocean temperature, evaporation also rises above the sea surface evaporation rate.

• Wind driven albedo increase. The white spray, foam, spindrift, changing angles of incidence, and white breaking wave tops greatly increase the albedo of the sea surface. This reduces the energy absorbed by the ocean.

• Cold rain and cold wind. As the moist air rises inside the thunderstorm’s heat pipe, water condenses and falls. Since the water is originating from condensing or freezing temperatures aloft, it cools the lower atmosphere it falls through. It also cools the surface when it hits. In addition, the falling rain entrains a cold wind. This cold wind blows radially outwards from the center of the falling rain, cooling the surrounding area.

• Increased reflective area. White fluffy cumulus clouds are not tall, so basically they only reflect from the tops. On the other hand, the vertical pipe of the thunderstorm reflects sunlight along its entire length. This means that thunderstorms shade an area of the ocean out of proportion to their footprint, particularly in the late afternoon.

• Modification of upper tropospheric ice crystal cloud amounts (Lindzen 2001, Spencer 2007). These clouds form from the tiny ice particles that come out of the smokestack of the thunderstorm heat engines. It appears that the regulation of these clouds has a large effect, as they are thought to warm (through IR absorption) more than they cool (through reflection).

• Enhanced nighttime radiation. Unlike long-lived stratus clouds, cumulus and cumulonimbus generally 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.

• Delivery of dry air to the surface. The air being sucked from the surface and lifted to altitude is counterbalanced by a descending flow of replacement air emitted from the top of the thunderstorm. This descending air has had the majority of the water vapor stripped out of it inside the thunderstorm, so it is relatively dry. The dryer the air, the more moisture it can pick up for the next trip to the sky. This increases the evaporative cooling of the surface.

Finally, since they are emergent phenomena that only arise where the surface is warmer than its surroundings, clouds and thunderstorms preferentially cool mainly the warmer areas in a way which is not well represented in bulk averages. In other words, the averages of the bulk measurements of say temperature and relative humidity in a gridcell containing thunderstorms gives little idea of the high-speed movements of massive amounts of energy which are taking place.

Anyhow, that’s take two on the CRE … I’m still ruminating on what I can learn from the CERES data, it’s far from mined out.

Best to all,

w.

 

Advertisements

  Subscribe  
newest oldest most voted
Notify of
Fred Souder

Willis,
Your list of mechanisms that cool the surface (such as Cold rain and Cold wind) add energy to the atmosphere. Condensation in clouds adds a LOT of energy to the air. Of course, you maintain that this energy is more easily transmitted to space, which I believe is true. Just a semantic use of the term “surface”. I guess the question would be how much more efficient is the radiation of energy from the upper atmosphere as a result of this added ‘heat’ as compared to the decrease in radiation from the surface due to its loss of ‘heat’. Oh, and sorry for the use of the term ‘heat’ as a synonym for temperature increase or energy.

Willis says: “On reflection, it seems to me that we need to compare that to how much we’d expect the cloud radiation to increase if the surface temperature goes up by 1°C.”
Hi Willis. Haven’t you answered your own question in the preceding sentence?
“Finally, I note that the global average change in net cloud radiation for each degree of surface warming is positive, at 0.7 W/m2 per degree. “
So comparing the 0.7W/m^2 to the 0.7W/m^2 seems to be a bit of a “Tada” moment.

Chip Javert

This is getting to be funny. Every now and then (but amazingly frequently…), Willis sits down for a few hours and actually does effective science on a CAGW (or what ever we’re calling it this week) topics of interest.
Geez – if I was one of the hoard of IPCC enthusiastic fools, this would be downright embarrassing. I sure wouldn’t want my mom to know I was so ineffective that some guy named Willis sits in his den and does more and better work than my entire IPCC crowd of hundreds of scientists, economists, psychologists, train engineers, tree surgeons, etc does in 4-5 years.
Keep up the good & interesting work, Willis.

DonV

Willis, I believe you might have missed one important radiative effect that I believe dominates the short term balancing act that water in its many forms contributes as the planets heat governor. You came close when you mentioned the effect of “Physically transporting heat from the surface directly to the upper troposphere . . . ” in latent heat and sensible heat, where it can be radiated out to space . . . but the pertinent question is how? How does water carry and then very rapidly radiate a disproportionate amount of energy back to space as compared to other gases?
I believe you are right on the money concerning latent heat. Water does have an enormous heat capacity, but when you take into account the massive energy consumed during phase transitions, the enthalpy of vaporization, ” 40.65 kJ/mol, is more than five times the energy required to heat the same quantity of water from 0 °C to 100 °C (cp = 75.3 J K−1 mol−1) (WIKI). Likewise, the enthalpy of fusion of water and enthalpy of sublimation are far greater for water than for any other naturally circulating gas or liquid, (except perhaps ammonia). For water these numbers tell the whole story:
Melting: -79.7 cal/g, -330,000 J/kg
Evaporation: -597.3 cal/g, -2,500,000 J/kg
Sublimation: -677.0 cal/g, -2,830,000 J/kg
Freezing: +79.7 cal/g, +330,000 J/kg
Condensation: +597.3 cal/g, +2,500,000 J/kg
Deposition: +677.0 cal/g, +2,830,000 J/kg
Those enormous amounts of energy are absorbed by water at the sea surface, then water physically transports the energy up to the first cloud layer where condensation occurs.
So what did you miss? Well, in addition to cooling the gases around it, I believe the very process of condensing gives off IR radiation. Likewise when sufficient energy is still contained in the liquid water droplets that are billowing higher and higher into the cloud tower – when these water droplets hit the second layer of the atmospheric ocean and freezing occurs, then a second massive round of energy release occurs. So I believe thunderheads must glow even brighter at their pinnacle because of all the IR radiation that is given off by these phase transitions.
I first suspect that IR energy is directly given off by the phase transitions when I saw IR images of Sandy from space at night. The fact that there was “detail” in the images suggested to me that some clouds were actively irradiating (condensing and freezing) while others not so much. The second clue came when I watched the lecture by Gerald Pollack explaining his theory of liquid crystal water. During that explanation he mentions that water droplets take in IR energy to form the thin liquid crystal layer on their surface. So my thought was, if the process works one way, why not the other. Once the water is transported to a layer where the structure that it finds itself in can no longer support its current physical state, it must give off the energy to change state the same way it got it.
I believe the bottom layer of clouds (condensation) and the layer that forms the “anvil” in thunder clouds (freezing) are powerful IR radiation sources and glow quite brightly. In addition all of the electrical energy that the liquid crystal layer creates is rapidly released in ENORMOUS amounts in the form of lightning. . . visible light . . . brighter than the brightest sunlight.
What do you think?

phlogiston

Your mention that clouds are emergent phenomena leads to one more way clouds lose heat from the climate system. Systems with nonlinear emergent pattern export entropy. With cloud this means heat loss to space. The system’s own entropy decreases on account of the emergent structure. This is one aspect of Ilya Prigogine’s nonlinear thermodynamics.

Robert

Don’t trees have the same effect on land surface temperatures as clouds have on the oceans. When land was covered by enormous woods most of the solar radiation did not reach the surface because it was absorbed by the leaves what makes day temperatures colder and keeps night temperatures warmer.Clouds also form more easily over forest what makes the effect even bigger. Cutting down these big extensions of forest ( ship building) must have had a some influence on temperature. Great Britain had nearly no forest left because of this.

What is measured by the CERES dataset?
What does it not measured?
You have shown us (I think) the gridded slope of the best fit through 120 months of CERES, land, and SST data. How good it the fit for each grid cell?
Could we see a sample of cross plots at a selection of grid cells on a North-South traverse, say along longitude 180 or 150W or 150E. every 10, 12, or 15 deg latitude?

DonV

I also forgot to mention the additional theory that perhaps once cloud dropets form they become IR ‘sinks’. That is that they absorb a disproportionate amount of IR energy to form the “liquid crystal” layer that also creates their surface charge. This surface charge is what keeps clouds “organized” and billowing. So not only do clouds have a higher visible light albedo because they are natural white light prisms, but they are also IR “scavengers” and prevent IR from penetrating to the surface. What sort of make sense is that a disproportionate amount of the radiation that they give off during condensation is either radiated disporportionately out to space (because of the high concentration of water vapor ascending vertically below the condensation event) or is absorbed to form the liquid crystal layer by the droplets immediately adjacent to them. Only the droplets on the top surface of the billowing cloud actually have their IR irradiance propagate unrestricted to space. Any IR irradiation given off in any direction other than UP is “scavenged” by water’s HUGE energy capacity in the IR portion of the spectrum.

otsar

DONV,
I have a pet theory that the dust paricles in the water droplets are actually the IR radiators, which is why they promote condensation. I believe the dust particles function very much like the doping elements in plastic scintillators.

otsar

Willis,
I thank you again for your good work and open mind.

I just came across your website AND the website regarding global cooling!!
http://gnosis474.blogspot.com/2009/12/fiction-of-climate-science.html

When the USAF flew planes above and below the cloud deck taking simultaneous measurements,they found that clouds were absorbing different amount of energy than theory said they should. And I’m not talking about a few tenths of a W/m^2 here or there. More like a couple of dozen W/m^2
Anyone interested in this subject really needs to read this paper.
http://tallbloke.files.wordpress.com/2012/11/cess.pdf

I find the discussion of this, and Willis’s first take: http://wattsupwiththat.com/2013/10/03/the-cloud-radiative-effect-cre/#more-95082 fascinating and informative, and I cannot help but compare this against the story that Popular Science has closed its comments because “debating science is dangerous”: http://joannenova.com.au/2013/10/popular-science-stop-comments-because-debating-science-is-dangerous-readers-are-dumb/#more-30899
Here this delightful discussion is speeding development and ideas along, and we’re all learning something, including Willis, who is giving us a prime example of what science is all about… while over there in consensus-land… well, there’s just nothing. No thought, no open-mindedness, no science in any form.
I mean, heck, if you want SCIENCE, it’s clear where you have to be.
A huge thank you, Willis, for demonstrating so clearly how a true scientist thinks and behaves and advances, and also why debate and discussion is so very important, and fun, and educational, and exciting, etc. I’m loving this. 🙂

otsar

Thank you tallbloke for the link.
1. I wonder if anyone has taken an IR spectrum of the top of large cumulonimbus in the tropics. This should give some credence or falsify my pet theory that dust particles are acting as radiators..
2. In my pet theory I would expect dust in water droplets, heated by condensation of water vapor and conduction, to act like black body radiators, allowing some wavelenghs to bypass the many adsorption bands of water, especially in the longer wavelengths.

Otsar: You’re welcome. Some more discussion of Mie scattering and cloud optical oddities here:
http://tallbloke.wordpress.com/2010/11/14/alistairmcd-aerosols-cause-warming/
“Mie solved Maxwell’s equations for a plane wave so the assumption of constant ‘Mie asymmetry factor’ is correct only when light first enters a cloud. Also, substantial direct backscattering at the upper cloud boundary is ignored yet it has an opposite dependence on droplet size than diffuse scattering. Therefore, above a threshold ‘optical depth’, pollution causes a reduction of albedo, another form of AGW. So, at the very least, the IPCC’s predictions of CO2-AGW should be reduced by a factor of about three, possibly much more if ‘cloud albedo effect’ heating explains most recent warming.”
Along with Roy Spencer’s new post we can be sure clouds will e keeping us guessing for some time to come.

David Shaw

If you wish someone to look over your theories, stats viewpoint, I’m happy to collaborate.
David

David L. Hagen

Willis
Compliments on your explorations.
To separate out cause and effect, may I suggest exploring the time lag between solar insolation and surface temperature. This should vary annually with latitude from south to north. See
Key Evidence for the Accumulative Model of High Solar Influence on Global Temperature David R.B. Stockwell
Stockwell finds Pi/2 (90 deg) or 2.75 year lag between the 11 year solar cycle and surface temperatures. I expect similar lags from annual variations varying with latitude with opposite phase between northern and southern hemispheres. That will also give you N-S graduations in the lag magnitude. Are there corresponding variations in your cloud radiation effects with phase lags during the year?

Could greater cloud cover in summer make it cooler as opposed to greater cloud cover in winter make it warmer?

Konrad

Willis,
I believe there may be some benefit in expanding your examination of “emergent phenomena”. It is not just clouds that can be viewed in this way but also the breakaway of surface airmasses as their Raleigh number is exceeded each dawn and intermittently during the day.
The Raleigh Bernard circulation in the Hadley, Ferrel and Polar tropospheric convection cells could be compared to film running through an old 35mm projector. It runs smoothly past the sound pick up (near the tropopause) but stutters past the projection gate frame by frame (air masses breaking away from the surface during the diurnal cycle).
This circulation, including the transport of water vapour, is the primary mechanism for transporting energy high above the level of maximum IR opacity, where it is then radiated to space.
Increasing the concentration of radiative gases in the atmosphere will decrease the Raleigh number for airmass breakaway and increase the general speed of tropospheric convective circulation.
Increasing the concentration of radiative gases in the atmosphere will increase radiative cooling at altitude, and decrease the time to airmass subsidence.
Increasing the speed of tropospheric convective circulation will increase the speed of mechanical energy transport from the surface.
PS. – You mentioned the high altitude ice clouds causing warming claim. It is worth re-examining those papers. The proposed warming mechanism was ice crystals reflecting IR back to the troposphere, reducing radiative cooling of airmasses at altitude, increasing the time to subsidence, slowing the speed of tropospheric convective circulation and decreasing the speed of mechanical energy transport from the surface. I think you may be able to see why those papers had to be quietly forgotten 😉

Bill 2

I admire your efforts, but the following isn’t found in any meteorology text book.
“The air being sucked from the surface and lifted to altitude is counterbalanced by a descending flow of replacement air emitted from the top of the thunderstorm. This descending air has had the majority of the water vapor stripped out of it inside the thunderstorm, so it is relatively dry.”

(In addition to my last question)
I understand we have more cloud cover both summer and winter from cosmic rays as proven by the CERN Cloud experiment.

DonV

The real problem with all this completely logical conjecture is that it will never penetrate the unreasoning dogma of the climate CAGW religion . . . primarily because this theory begins with the assumption that the system is constantly striving to “balance” the energy imbalance. And that therefore, when you integrate underneath the instantaneous energy transfer curve on a second by second basis you don’t come up with a “temperature average”. Even CERES data set is an “average”. The energy released in any given thunderstorm, heck even during one lightning strike is never factored into the “average”. The bottom line is “average air surface temperature” is a terrible indicator of the earths net energy balance act, even when measured, then averaged, and homgenized and averaged again. A better measure would be to integrate the area under the curve on minute by minute basis of the relative humidity at each of the three layers of the atmospheric ocean for any given reporting period.
Look again at the picture at the top of WUWT! See the clouds? See the very very bright reflection off the clouds? See the huge thunder head? See the shadow it casts? What would this picture look like in the IR? Can you get any meaningful measure off of such a picture when you consider that it is actually a motion picture and that the reason for that thundercloud is that the atmospheric ocean in that particular location is very much “unbalanced” and is in the process of rebalancing.
I believe that unfortunately all Willis has done here is create a map, through careful billiant mathematical analysis that show the average distribution of thunder storms across the globe.

Matthew R Marler

Good post. Thank you.
I liked your list of mechanisms whose quantitative effects are poorly characterized. Skeptics are accused on not knowing “the science” (or “the physics”) but you are adept at highlighting physical processes known to exist that are poorly known.

DonV

Otsar: Intersting theory. I agree that in order for condensation to occur in a “supersaturated system” some sort of seed is necessary to trigger droplet formation, but I think that dust is not as large a player as energetic particles bombarding the planet from outer space are. Look at Willis’s map. The Sahara generates more atmospheric “dust” than any other large land mass IMHO. Yet just to it’s west you see a large “blue” area? What’s going on there? Yet lower dust generated clouds represented by the thunderstorms coming off of the west of South America have much higher IR. Why the disparity?

BarryW

You analysis relates indirectly to another point that has been troubling me and that’s the use of anomalies. The affect of a temperature change from -41 to -39 is extremely different from say -1 to 1 deg. Same hold true for everything else. Raise the temps, raise the evaporation. Raise the winds, raise the evaporation. Dry downdrafts, and so on. As you pointed out the Inter-Tropical Convergence Zones have a different effect than say the Arctic. Not too many thunderstorms there.
They tried to compare the amount of energy they claim is absorbed by the ocean to what the temperature effect would be if released to the atmosphere. How about the reverse? How much would the ocean temps rise if it absorbed the present atmospheric temp increase”? Negligible I bet.
Low tropical cyclone activity in the Atlantic? Dust from Africa causing it? Wonder if that is accounted for in the models? Jet Stream, upper air jets. TS Karen is getting blown apart. So many interacting pieces that aren’t being accounted for.

Leo Morgan

Willis,
I just saw this at Bishop Hill’s website. http://www.bishop-hill.net/blog/2013/10/2/new-blogs-on-the-block.html
He refers the reader to the work of another blogger who has done similar work to your own.
I mention it in case you haven’t already seen it.

AndyG55

Willis has gone where no “climate scientist” has been before.
Admitted he goofed… and tried to fix it !
Well done Willis. 🙂
Other climate scientists should learn this from you, at the very least.

Bill 2 says:
“I admire your efforts, but the following isn’t found in any meteorology text book…”
I wouldn’t know, since I have not read every meteorology text book like you have. But the explanation makes perfect sense to me.
A thunderstorm produces rain. That is the majority of the water vapor that is stripped out.
If there is a problem with the description given, please identify the problem. Don’t just appeal to a vague authority that cannot be easily checked by readers.

The sun’s local noon zenith angle at any location of the earth can change as much as 47 degrees over parts of the year. Suppose we break the dataset up into months of similar noon zenith angle and calculate the CRE vs Temp effect
If the CERES dataset is binned into months, then, based upon the Analemma
November, December, January, belong to the Southern set,
May, June, July, belong to the Northern set,
August, September, October belong to the Autumnal set
February, March, April belong to the Vernal set.
With these four sets, you should get 30 points (3 months/yr X 10 years) for each grid cell to calculate the Figure 1 map based upon solar inclination. It would be interesting if the R^2 got better or worse than using the whole year for the least squares fit. You are getting more consistent solar illumination on each grid cell, but you are also getting less of a temperature range to calculate the slope.
Autumnal Set and Vernal Set ought to be similar for the tropics and maybe even the temporate zones. But not necessarily. A contra indicator is that we get hurricanes in the Autumnal set but not the Vernal set.

DonV says:
October 5, 2013 at 4:58 pm
Don I discuss some of the energy transfer here:
http://wattsupwiththat.com/2013/10/03/the-cloud-radiative-effect-cre/#comment-1435512
Heavy rain at 1″ per hour is 15,924 W/m^2 of surface area. Kind of a lot…
I was thinking about how much more surface area the cloud top must have to radiate away enough energy to balance the 15,924 W/m^2 of heat released during heavy (1″ per hour) rain. I only assumed condensation to a liquid, used a typical water droplet diameter of 0.02mm, and cumulonimbus water content of 2g/m^3. Given that, the cloud top “surface” should radiate at around 31 W/m^2. This indicates the surface area needs to be about 515x the size of the 1 m^2 that is raining at 1″ per hour to dissipate that much heat. So if it is raining that fast on 1m^2, the cloud radius at the top would have to be 12.8m so support that much energy dissipation (assuming no GHG above that). Actually, if the storm was this big, freezing would be occurring, and at 50,000 feet the temperature would be more like -57°C, and radiation would only be around 12W/m^2 of cloud “surface” and the storm diameter ratio would go to 20x. At that level, there is only about 10% of GHG’s above the storm, I didn’t account for this either.
droplet dia 0.02 mm
droplet radius 0.01 mm
droplet area 0.001256637 mm^2
droplet area 1.25664E-09 m^2
droplet volume 4.18879E-06 mm^3
droplet volume 4.18879E-15 m^3
droplet volume 4.18879E-09 cm^3
Cumulonimbus g/m^3 2
n droplets/m^3 477464829.3 per m^3
surface area all droplets 0.6 m^2
W/m^2 at 0°C H2O IR emission 309.3445884
Watts/m^3 185.6067531
surfaces on m^3 6
watts per surface 30.93445884
w/m^2 raining 1″/hr 15924
surface area of cloud top 514.765753 m^2 per m^2 rain
cloud top radius 12.80058703 m per m rain
Just a first shot at estimating radiation from cloud tops. It depends very much on a lot of factors – I just used some averages and assumed 0°C.

After reading the first post on this subject (The Cloud Radiative Effect (CRE)) and now the follow-up, I must comment on some things I read. First, one thing from the original, first post.
“…they [clouds] warm the surface by increasing the downwelling longwave radiation.”
For the most part, except for fog, clouds in the sky are colder than the surface. The cloud base is usually several thousand feet above the surface at the LCL (Lifted Condensation Level) – the level at which the temperature has cooled due to lift to reach saturation. A colder surface cannot warm a warmer surface. Now, cloud cover will prevent cooling (and it takes very little [thin] cloud cover to stop the surface IR from radiating out to space) but it cannot ‘add’ warmth to the surface below it when it is colder than that surface.
Now, on to The Cloud Radiative Effect, Take Two. In it, Willis states:
“[clouds] Physically transporting heat from the surface directly to the upper troposphere where it radiates easily to space.”
This seems to state that (for example) an 85 deg. parcel of air at the surface, when lifted to the upper troposphere (let’s just presume a cumulonimbus & say the tropopause), will radiate out to space as still an 85 deg. parcel of air which, of course, is entirely untrue. Remember a few *very* important things;
1) As parcels are lifted by whatever reason, they cool (loose IR energy) due to expansion. At first, they cool at the dry adiabatic rate until they reach the LCL (see above), then they cool at the slower moist adiabatic rate…but they are still cooling as long as they are warmer than their surroundings which, in the tropics, is usually only a few degrees.
2) The tropopause is very, very, very cold (especially in the tropics).
3) The tropopause is where those surface parcels reach equilibrium with their environment (the same temperature) so they stop lifting & spread out horizontally.
Once the cloud parcels are at equilibrium with their environment, there is no ‘excess’ energy to radiate out to space. That is why on the NOAA OLR (Outgoing Longwave Radiation) web pages from satellite observations, the largest amount of OLR is in clear sky (ground is warmest) & the least is where there are thunderstorms (tops are coldest).
Clouds, in whatever shape or form, emit IR at whatever temperature the level of the cloud is at. This is especially evident with cumulus clouds. If you were to observe the vertical tower with an IR imager, you would see a thermal gradient up the side of the cloud with warmest at the base & coldest at the top and the top growing colder as it grows higher.
Now, another statement from Willis:
“Since the heat is transported either as latent heat, or as sensible heat inside the thunderstorm tower, it doesn’t interact with the large amount of water vapor…”
What?!?!? Latent heat doesn’t interact with water vapor??? Latent heat is *because* of water vapors transition to either a liquid or solid. How can it not…interact…with water vapor?? Heat is transported vertically as sensible heat with latent heat added on as condensation occurs.
“Delivery of dry air to the surface.”
All of this is basically true but what was left out was that the descending air is warming at the dry adiabatic rate as it falls which counteracts the cooling issue.
Now, some statements made by DonV:
“Well, in addition to cooling the gases around it, I believe the very process of condensing gives off IR radiation…”
Right – that is the latent heating from condensation.
“Likewise when sufficient energy is still contained in the liquid water droplets that are billowing higher and higher into the cloud tower – when these water droplets hit the second layer of the atmospheric ocean and freezing occurs, then a second massive round of energy release occurs…”
No, the process of condensation from lift, as well as the associated release of latent heat, is a continuous process regardless if the process is going from gas to liquid, liquid to solid or gas to solid. It’s all as seamless, continuous process in a towering cumulus cloud.
‘So I believe thunderheads must glow even brighter at their pinnacle because of all the IR radiation that is given off by these phase transitions.”
No, as the parcels are lifted higher into the atmosphere, they cool & get ‘dimmer’ (less IR energy) not warmer & ‘brighter’ (more IR energy).
“I first suspect that IR energy is directly given off by the phase transitions when I saw IR images of Sandy from space at night. The fact that there was “detail” in the images suggested to me that some clouds were actively irradiating (condensing and freezing) while others not so much.”
Well, yes, the lower (warmer) clouds *do* radiate more IR than the higher (colder) cloud tops and, yes, there will be ‘detail’ in the imagery. Active convective clouds do not have flat tops. Even the so-called ‘flat-tops’ of thunderstorms can have convective detail from strong, persistent updrafts & overshooting tops which, due to inertia, overshoot the tropopause but then fall back down to the equilibrium level with the rest of the cloud mass…which is very, very, very cold with very, very, very little IR energy to radiate out to space compared to all that is below it.
I believe you are getting confused by the color enhancement that is sometimes used in the IR satellite imagery. Remember, digital satellite imagery are just numbers from 0-1024 (0 = coldest, 1024 = warmest). They can make any number be any color and usually, if they want to ‘enhance’ a certain feature (thunderstorms, for example), they make the warmest (lowest) clouds a grey scale (dark to light) then, at some level, shift to different colors, usually transition from blue to green to yellow to red with red being coldest because that associates the colder cloud tops with the stronger storms *HOWEVER* do not try to associate any specific color with an actual temperature unless you have the look-up table to go by. You can get fooled to thinking the ‘reds’ are warmest when in fact they are the coldest.
Cheers,
Jeff
NOAA/NESDIS

Manfred

Hi Willis,
temperature change may affect cloud cover, but cloud cover change also effects temerature.
How do you separate cause and effect ?

Not being a scientist, most of this discussion is beyond my ability to comprehend. But one thing is very clear, this is REAL peer reviewed science, not the crap that they pass off at the IPCC. Open and honest give and take without too much ego getting in the way.

Mike M

JKrob says:” All of this is basically true but what was left out was that the descending air is warming at the dry adiabatic rate as it falls which counteracts the cooling issue.”
It’s always colder after a mid-afternoon summer/tropical thunderstorm. A lot of heat goes up in the form of warm moist air and the only thing that comes back down is drier cooler air plus cold rain so what is the counteraction?

otsar

JKrob,
Has anyone actually scanned a cumulonibus from top to bottom, from the side, with a wideband IR spectrometer? Mapped a cumulonimbus from the top and from the bottom for the wavelengths present?
I would appreciate the link.

John F. Hultquist

Bill 2 says:
October 5, 2013 at 6:55 pm

As dbstealey says at 7:59 pm, “If there is a problem with the description given, please identify the problem. Don’t just appeal to a vague authority that cannot be easily checked by readers.
Because at least 2 of us don’t see in your comment at 6:55 what you seem to think is clear.
From the following link:
http://www.exploringnature.org/db/detail.php?dbID=112&detID=2633
“As the clouds rise into the colder atmosphere, the water droplets come together and get bigger and heavier. Finally they will begin to fall. As they fall, they pull air with them. This is called a downdraft. ”
and . . .
“The thunderstorm downdraft can bring a gust of cool air to the Earth’s surface. This is called a gust front.”
Drawing here:
http://www.srh.noaa.gov/jetstream/tstorms/wind.htm

Jon

“Cutting down these big extensions of forest ( ship building) must have had a some influence on temperature. Great Britain had nearly no forest left because of this.”
A lot was burned down much earlier, from the Bronze Age, to be used as grassland for sheep, goats, cows and farm land. Not long ago in Norway 90% lived in the districts living of this land. Today 90% is living in the towns and we have today have 100% more wood mass than we had aprox 100 years ago.

Willis,
Have you ever thought about the hygroscopic effects of exposed salt on the ground, and cloud formation of (e.g., lack of). The areas of exposed salt are: Lake Uyuni, Bolivia; Great Salt Lake, Death Valley (+other areas in Nevada and Utah); Lakes Eyre, Frome, and Torrens, in Australia; The Dry Valleys of Antarctica (are probably dry because any snow falling there dries out (melts) because of ground salt); The Kavir of Iran; northern Algeria; parts of Marokko; East African Rift lakes (Magadi,etc), Quattara Depression, Egypt; Baluchistan in Pakistan, etc, etc.
You see, these dry and very salt locations are also those areas of the terrestrial earth that are lowest in elevation (probably due to mineral corrosion – i.e., dust formation by the aggressive salt). I guess some of the fine salt particles find their way into the above atmosphere and prevents droplet formation – very little rain…?
All the best MH

DougS

Good health to you Willis. My hope this night is for people all over the world to know how science is practiced. It’s mastered in the way that you have demonstrated with this post. Once we learn to say “I may have made an error on my previous calculation” and then invite our colleague’s to comment as you have done here, knowledge progresses.
Gods speed and may the wind remain at your back.
Doug Sherman

Crispin in Waterloo but really in Yogyakarta

@AndyG55
I think he made a model and had it peer reviewed and improved it based on the feedback received. I call it ‘science’.
@Fred Souder says:
“Condensation in clouds adds a LOT of energy to the air. Of course, you maintain that this energy is more easily transmitted to space, which I believe is true.”
Well, it is all transmitted to space eventually. Every last Joule. So we are really investigating the retention of energy at some point in the vertical scale until it is effectively radiated into space. That is why they use of the concept of a ‘height’ at which the heat effectively leaves us forever. A change in GHG’s like water vapour changes the ‘effective height’ which is an average, and thus may miss the ‘punch through’ of a rising thunderhead that creates instead a very bright spot of IR far above the supposed average altitude. Does thunderstorm punch-through dominate heat loss, or GHG’s providing insulation?
It is worth considering the comments on this subject made by Prof Adrian Bejan, the author of numerous heat transfer textbooks. His comment with regard to the efficiency of rising heat columns (which form naturally if there is spot heating below) is that the vertical transport of heat energy takes place at the maximum efficiency possible, never going faster upwards at a speed that induces turbulent flow. Thus the centre rises faster than the space immediately around it which in turn is faster than the space around that, etc, until the toroidal flow is a cell large enough (governed by the viscosity of the air) to break into concentric cells with opposing flows. That happens in a hurricane which is why as it passes over, it rains like hell for a while then stops, only to be repeated in waves hour after hour. Each rainfall is a cell connected to the others on its outer surface.
Bejan’s point is that discrete cells form naturally and operate at their most efficient possible heat transport rate. If the surface warms slightly, the column is slightly larger in diameter and it dumps the heat upward more efficiently because everything within the system is moving vertically as efficiently as possible. Where this not the case, large thunderstorms would not cool more efficiently than small ones.
The model of this is two horizontal plates, the lower one being heated from below, and a convective medium moving heat from the lower to the top, colder plate. When the plates are near each other, as is the case when looking at the atmosphere in cross section, cells develop automatically with connected toroidal flows, and they dump heat as efficiently as possible to the cold plate. Heating the bottom plate slightly more does not result is a much warmer region immediately above the bottom plate. It just increases the heat flux rate to the plate above. If the temperature is varied a lot, the cell structure adjusts into square, hexagonal and other patterns. It is amazing and he explains the math involved. The fact that the top plate is cooled by IR radiation into space is irrelevant. It is just how things work on the case of a planet.
Bejan said that the problem was so simple to solve it was not even interesting.
Anything that adds heat to the bottom increases the efficiency of heat transport. Because clouds are a complication, they strongly affect several aspects how the heat gets to the surface or not. Myself, I am holding out that Svensmark is correct about clouds and Bejan is correct about self governing heat transport mechanisms. Together they show the Earth has a strong self-governing, damped temperature regulating mechanism that is happy to stabilize for long periods at any average temperatures from a balmy Arctic to a frozen Virginia. The real controls are outside the system.

wayne

JKrob:
Jeff, thank you, you just made my day right at the top of your comment. Sure appreciate all of the details and a better ways to word such subjects (my weakness).

old construction worker

As I have said before, we live in a big swamp cooler

Willis – (1) You ask “On reflection, it seems to me that we need to compare that to how much we’d expect the cloud radiation to increase if the surface temperature goes up by 1°C.“. Maybe the figure given by Wentz may help:
Wentz et al, Science 13 July 2007: Vol. 317 no. 5835 pp. 233-235 DOI: 10.1126/science.1140746
https://www.sciencemag.org/content/317/5835/233.abstract
Climate models and satellite observations both indicate that the total amount of water in the atmosphere will increase at a rate of 7% per kelvin of surface warming. However, the climate models predict that global precipitation will increase at a much slower rate of 1 to 3% per kelvin. A recent analysis of satellite observations does not support this prediction of a muted response of precipitation to global warming. Rather, the observations suggest that precipitation and total atmospheric water have increased at about the same rate over the past two decades. “.
There was confirmation of the 7% figure on ABC(Australian Broadcasting Corporation)’s Catalyst program on 4 July 2013: “By studying over 8,000 rain gauges across the world, Australian scientists have confirmed that extreme rainfall events have also been intensifying. That means we’re getting more water from a big storm than we would have 30 or 40 years ago. Around 7 percent more per degree rise in temperature.“.
[I don’t have a link, the text was emailed to me, but Susan Wijffels was on the program, so the figure is reliable. See https://www.llnl.gov/news/newsreleases/2012/Apr/NR-12-04-02.html. There should be a paper by Durack and Wijffels somewhere.].
(2) I still think you are very wrong in your basic assumptions (see my comments on your previous thread). You say “ I suspect that the blue around Antarctica is an artifact due to the presence of the sea ice edge.“. It was that blue around Antarctica that caught my attention then. I think it is a symptom of your error, and that you are probably finding a seasonal effect not a feedback.

Roy

This week’s New Scientist magazine vol. 220, no. 2937, 5 October 2013, has a special report on climate outcomes in the wake of the IPCC report, plus an editorial and a couple of very short articles that are also on the climate. One of those short articles (on page 17) is about the large fall in CO2 emissions in what was the USSR after the collapse of communism. The article does not actually say that all we need now is a collapse of capitalism and the global warming problem will be solved!
The other short article, Climate Science Tweaked, on page 6 is more relevant to Willis’s post on clouds – but comes to the opposite conclusion. The final paragraph states:
Clouds are also a menace. In 2007, it was uncertain whether clouds cooled or warmed the planet overall. “We now believe that they are a positive feedback on temperature,” says Piers Forster of the University of Leeds, UK. “Their warming effect will intensify with global warming.”
I wonder what Willis Eschenbach makes of the claim that clouds have a positive feedback effect on temperature?

thingadonta

“…problem is that even if the clouds stay exactly the same, if the solar radiation increases, the net cloud radiative effect (CRE) increases due to increased reflection”
What about if the oceans warm or cool over long periods (e.g. decades) due to their slow equilibrium response to forcing? Wouldn’t this also mean the net cloud radiative effect would change during this time over the oceans, due to increased/decreased heat exchange between the ocean and atmosphere?.

DonV

JKrob says:
Jeff, thank you for the feedback. I learned a lot from your discussion and it was very informative and welcomed. Thanks for the education.
However, I noted several things in your discussion that do not jibe with my own observations what I see when I watch a storm, and my own knowledge of what I know happens in cooling towers and water based evaporators.
First things first:
“A colder surface cannot warm a warmer surface.” This statement by itself is essentially true. But I believe you are missing something important. . . . while the laws of thermodynamics state that NET energy balance between a cold radiator and a warmer radiator will have the warmer radiator transferring net energy to the colder – 1) it does NOT mean that radiation does not OCCUR in the opposite direction. The cooling of one black body radiator at a lower level to a cooler state MUST create an IR signature that indicates its black body temperature. You are correct that as you ascend into a cloud this IR signature changes as the temperature changes. But at each incremental foot up, the droplets of water that are cooling and radiating this energy DO NOT care whether they are radiating towards a warmer body or the colder layer of water droplets above them or the very very cold of outer space. They simply radiate! What is REALLY important is whether these radiating water droplets are warmer than the AIR around them. They MUST BE IR brighter if they can be SEEN with an IR imager. So 2) the NET transport of energy between clear dry air, and water vapor, droplet, ice saturated air is orders of magnitude greater in favor of the later. So yes at 1000 feet ASL a thunder cloud is transferring energy out at that level because the water at that level is initially warmer, but gives up an enormous amount of energy as it phase transitions. 3) Any radiative energy that shines back down from just cooled water droplets to warmer water vapor saturated air coming up must be simply absorbed and added to the upwelling water vapor energy content. Any that radiates up throught the cloud does also, but instead into water droplets. And any that are close to the clear edge radiate to outer space or, if we had one, our tuneable IR camera. And finally 4) I believe the “continous process” you speak of is the loss of sensible heat, not latent heat. Latent heat loss occurs at one temperature/pressure transition. It is discontinuous and often quite sudden. Sensible heat loss, the cooling of water droplets or water vapor or ice crystals AFTER phase transition can be continuous and does most likely occur throughout the transport of water up through a cloud, but it is orders of magnitude lower than latent heat loss.
Now you also state:
“Clouds, in whatever shape or form, emit IR at whatever temperature the level of the cloud is at. This is especially evident with cumulus clouds. If you were to observe the vertical tower with an IR imager, you would see a thermal gradient up the side of the cloud with warmest at the base & coldest at the top and the top growing colder as it grows higher.” Again, parts of this statement are essentially true, but they are also combined with pure conjecture. More importantly the combined statement it ignores the important salient point I raised in my statement of theory. At each elevation the standard black body radiation curve must match the local temperature of whatever molecules are present. I posit that when water, 1) first as vapor, then 2) as water droplets, then 3) as ice crystals are present at any given elevation, the local black body temperature that creates the local IR signature will ALWAYS be greater than when compared to clear air. Hence water is serving as an energy transport medium. Second, it is not the IR signature spectrum that is important (the IR signature spectrum is what indicates the local black body temp), rather it is the INTENSITY of RADIATION at that signature that tells whether a significant energy loss is occurring. You mention “No, as the parcels are lifted higher into the atmosphere, they cool & get ‘dimmer’ (less IR energy) not warmer & ‘brighter’ (more IR energy).” By “dimmer” and “brighter” what I believe you mean is that the whole black body radiation curve shifts to longer and longer wavelength, where the outer tropopause has the coolest IR signature. But again, relative to what? If ice crystalizaion or sublimation is occuring then it MUST have a greater INTENSITY at that local IR signature when compared to normal cooling of the surrounding nitrogen or oxygen gases, (very little CO2 at this level, too heavy). The evidence for this would be a comparative IR picture, comparing the normal IR signature INTENSITY of the normal lapse rate of a low relative humidity clear air column side by side to the same IR signature INTENSITY of a storm cloud. My theory is that at the level at which condensation is occuring and then again at the elevation where most of the freezing is occuring you will see an very large difference in the IR radiation INTENSITY at that particular IR black body signature spectrum. You will also see a somewhat less intense IR radiation signal that is still significantly greater than clear air that is due to sensible heat loss, because water has a higher heat capacity than air.
And finally you state:
“No, the process of condensation from lift, as well as the associated release of latent heat, is a continuous process regardless if the process is going from gas to liquid, liquid to solid or gas to solid. It’s all as seamless, continuous process in a towering cumulus cloud.” This statement does not match what I SEE from the side of a towering thunder cloud. I SEE a discontinuous process. I SEE at least two distinct layers. I SEE a layer where water vapor has formed into water droplets. Therefore the evidence from my eyes tells me that at that layer condensation is rapidly occuring, Therefore, I presume that the local temp and pressure at that layer must have created the conditions for supersaturation, and therefore at that layer and NOT BELOW IT one might expect to see greater IR intensity at whatever local black body radiation temperature might be. Next, I only SEE cloud billowing up from below. What I don’t SEE is a uniform and sudden appearance of cloud all the way from the first layer up to the second! Therefore I SEE condensation only where water vapor has “punched” through the already saturated cloud, (where no further condensation is occurring) but where local conditions still favor the liquid phase of water not the gaseous phase or the solid phase. The P and T conditions that favor condensation occur all the way up to where I SEE what I believe to be the next “layer”, which is where now the conditions favor freezing and ice crystal formation – the anvil “head” of the thundercloud. When that occurs, again compared to what energy is contained in the black body radiator surround each droplet to ice transition, the NET IR radiation HAS to be from the phase transition event OUT to the local air and hence its IR radiative INTENSITY at that local black body radiation spectrum MUST be higher than when compared to clear dry air.
At this point, I will repeat what “otsar” asked: “Has anyone actually scanned a cumulonibus from top to bottom, from the side, with a wideband IR spectrometer? Mapped a cumulonimbus from the top and from the bottom for the wavelengths present?”
And I will add, How does that spectrum with intensity variation compare to clear dry air? If such a “picture” exists, I would like to see it. That kind of “picture” would show me the data and give me a clearer understanding of the processes.
Finally, I am NOT “confused by the different color enhancements made”. The picture I saw had no such enhancements (at least that is what the description claimed). And what I noticed about the picture was the very blackness of the ocean around the storm. This to me indicated that the INTENSITY of the OLR signature for the “balancing act” is significantly greater for clouds and the water cycle than for non-water GHG processes.
Again, respect to you Jeff, Tallbloke, otsar, etc. and Willis for this discussion. Educate me some more. I love this discussion.

A. Scott

Bill 2 … excerpts from a number of “meteorology” articles below. But is it even necessary to have their confirmation?Simple common sense tells us air generally dries with altitude, and that cooler drier air is pulled into the storm and drawn thru the downdraft to the surface. Anyone who has experienced a thunderstorm has seen and felt the cooling and significant reduction in humidity as the storm collapses and passes :
“Eventually the water droplets and ice crystals in the clouds become so large that they can no longer be supported by the uprising air mass, and they begin to fall forming rapid downdrafts on the leading edge of the cloud. In the mature stage of thunderstorm development updrafts and downdrafts operate side by side within the cloud. This is the most dangerous stage of a thunderstorm because of the high winds accompanying the downdrafts, the heavy rain, as well as thunder, lightning, and possible hail and tornado development. Eventually the cloud reaches the dissipating stage as the downdrafts drag in so much cool dry air that it prevents further updrafts of warm moist air. With lack of updrafts of warm moist air, the cloud begins to dissipate and eventually it stops raining. ”
“The second stage is the mature stage of development. During the mature stage warm, moist updrafts continue to feed the thunderstorm while cold downdrafts begin to form. The downdrafts are a product of the entrainment of cool, dry air into the cloud by the falling rain. As rain falls through the air it drags the cool, dry air that surrounds the cloud into it. As dry air comes in contact with cloud and rain droplets they evaporate cooling the cloud. The falling rain drags this cool air to the surface as a cold downdraft”
“Recall that as precipitation builds within a cumulonimbus cloud, it eventually generates a downdraft. Well, as the downdraft travels downward and exits the base of the cloud, the precipitation is released. A rush of rain-cooled dry air accompanies it. When this air reaches the Earth’s surface, it spreads out ahead of the thunderstorm cloud–an event known as the gust front. The gust front is the reason why cool, breezy conditions are often felt at the onset of a downpour.”

Crispin in Waterloo but really in Yogyakarta

@DonV
You commented to JKrob
>First things first:
>>“A colder surface cannot warm a warmer surface.” This statement by itself is essentially true.
There is frequently confusion about the difference between a cold object heating a warmer object by conduction and energy exchange by radiation. It is frequently raised in debates about heat transfer and the root problem is how science is taught in high school (badly and incompletely and with poor retention).
Radiation from all objects transfers heat to all other bodies all the time whether they are they same temperature or not. Conduction is another matter entirely. It may surprise people to know that their skin is cooled by the walls of their house by conduction through air, though the energy moved is tiny.
Cold objects cannot warming a hot object by conduction, but they certainly radiate energy towards them in an unequal fashion, eventually gaining more from the hot object than they lose in return.
What you said is correct – plus some sensible things far above.

DonV:
(very little CO2 at this level, too heavy) What makes you say that?