'Correcting' Trenberth et al.

(See the note below before taking this post seriously – Anthony)

Guest essay by Steven Wilde

clip_image002

Here we see the classic energy budget analysis supporting the hypothesis that the surface of the Earth is warmer than the S-B equation would predict due to 324 Wm2 of ‘Back Radiation’ from the atmosphere to the surface.

It is proposed that it is Back Radiation that lifts the surface temperature from 255K, as predicted by S-B, to the 288K actually observed because the 324 Back Radiation exceeds the surface radiation to the air of 222 Wm2 ( 390 Wm2 less 168 Wm2) by 102 Wm2. It is suggested that there is a net radiative flow from atmosphere to surface of 102 Wm2.

I now discuss an alternative possibility.

The portions I wish to focus on are:

i) 390 Wm2 Surface Radiation to atmosphere

ii) 78 Wm2 Evapo-transpiration surface to atmosphere

iii) 24 Thermals surface to atmosphere

iv) 324 Back Radiation atmosphere to surface

The budget needs to be amended as follows:

The 78 Wm2 needs to be corrected to zero because the moist adiabatic lapse rate during ascent is less than the dry lapse rate on adiabatic descent which ensures that after the first convective cycle there is as much energy back at the surface as before Evapo-transpiration began.

The 24 Wm2 for thermals needs to be corrected to zero because dry air that rises in thermals then warms back up to the original temperature on descent.

Therefore neither ii) nor iii) should be included in the radiative budget at all. They involve purely non radiative means of energy transfer and have no place in the radiative budget since, being net zero, they do not cool the surface. AGW theory and the Trenberth diagram incorrectly include them as a net surface cooling influence.

Furthermore, they cannot reduce Earth’s surface temperature below 255K because both conduction and convection are slower methods of energy transmission than radiation. To reduce the surface temperature below 255K they would have to work faster than radiation which is obviously not so.

They can only raise a surface temperature above the S-B expectation and for Earth that is 33K.

Once the first convective overturning cycle has been completed neither Thermals nor Evapo-transpiration can have any additional warming effect at the surface provided mass, gravity and insolation remain constant.

As regards iv) the correct figure for the radiative flux from atmosphere to surface should be 222 Wm2 because items ii) and iii) should not have been included.

That also leaves the surface to atmosphere radiative flux at 222 Wm2 which taken with the 168 Wm2 absorbed directly by the surface comes to the 390 Wm2 required for radiation from the surface.

The rest of the energy budget diagram appears to be correct.

So, how to decide whether my interpretation is accurate?

I think it is generally accepted that the lapse rate slope marks the points in the atmosphere where there is energy balance within molecules that are at the correct height for their temperature.

Since the lapse rate slope intersects with the surface it follows that DWIR equals UWIR for a zero net radiative balance if a molecule at the surface is at the correct temperature for its height. If it is not at the correct surface temperature it will simply move towards the correct height by virtue of density variations in the horizontal plane (convection).

Thus, 222 UWIR at the surface should equal 222 DWIR at the surface AND 222 plus 168 should add up to 390 and, of course, it does.

AGW theory erroneously assumes that Thermals and Evapo-transpiration have a net cooling effect on the surface and so they have to uplift the radiative exchange at the surface from 222 Wm2 to 324 Wm2 and additionally they assume that the extra 102 Wm2 is attributable to a net radiative flux towards the surface from the atmosphere.

The truth is that there is no net flow of radiation in any direction at the surface once the air at the surface is at its correct temperature for its height, which is 288K and not 255K. The lapse rate intersecting at the surface tells us that there can be no net radiative flux at the surface when surface temperature is at 288K.

A rise in surface temperature above the S-B prediction is inevitable for an atmosphere capable of conducting and convection because those two processes introduce a delay in the transmission of radiative energy through the system. Conduction and convection are a function of mass held within a gravity field.

Energy being used to hold up the weight of an atmosphere via conduction and convection is no longer available for radiation to space since energy cannot be in two places at once.

The greenhouse effect is therefore a product of atmospheric mass rather than radiative characteristics of constituent molecules as is clearly seen when the Trenberth diagram is corrected and the lapse rate considered.

Since one can never have more than 390 Wm2 at the surface without increasing conduction and convection via changes in mass, gravity or insolation a change in the quantity of GHGs cannot make any difference. All they can do is redistribute energy within the atmosphere.

There is a climate effect from the air circulation changes but, due to the tiny proportion of Earth’s atmospheric mass comprised of GHGs, too small to measure compared to natural variability.

What Happens When Radiative Gases Increase Or Decrease?

Applying the above correction to the Trenberth figures we can now see that 222 Wm2 radiation from the surface to the atmosphere is simply balanced by 222 Wm2 radiation from the atmosphere to the surface. That is the energy being constantly expended by the surface via conduction and convection to keep the weight of the atmosphere off the surface. We must ignore it for the purpose of energy transmission to space since the same energy cannot be in two places at once.

We then have 168 Wm2 left over at the surface which represents energy absorbed by the surface after 30 Wm2 has been reflected from the surface , 77 Wm2 has been reflected by the atmosphere and 67 Wm2 has been absorbed by the atmosphere before it reaches the surface.

That 168 Wm2 is then transferred to the atmosphere by conduction and convection leaving a total of 235 Wm2 in the atmosphere (168 plus 67).

It is that 235 Wm2 that must escape to space if radiative balance is to be maintained.

Now, remember that the lapse rate slope represents the positions in the atmosphere where molecules are at their correct temperature for their height.

At any given moment convection arranges that half the mass of the atmosphere is too warm for its height and half the mass is too cold for its height.

The reason for that is that the convective process runs out of energy to lift the atmosphere any higher against gravity when the two halves equalise.

It must follow that at any given time half of the GHGs must be too warm for their height and the other half too cold for their height.

That results in density differentials that cause the warm molecules to rise and the cold molecules to fall.

If a GHG molecule is too warm for its height then DWIR back to the surface dominates but the molecule rises away from the surface and cools until DWIR again equals UWIR.

If a GHG molecule is too cold for its height then UWIR to space dominates but the molecule then falls until DWIR again equals UWIR.

The net effect is that any potential for GHGs to warm or cool the surface is negated by the height changes relative to the slope of the adiabatic lapse rate.

Let’s now look at how that outgoing 235 Wm2 is dealt with if radiative gas concentrations change.

It is recognised that radiative gases tend to reduce the size of the Atmospheric Window (40 Wm2) so we will assume a reduction from 40 Wm2 to 35 Wm2 by way of example.

If that happens then DWIR for molecules that are too warm for their height will increase but the subsequent rise in height will cause the molecule to rise above its correct position along the lapse rate slope with UWIR to space increasing at the expense of DWIR back to the surface and rising will only stop when DWIR again equals UWIR.

Since UWIR to space increases to compensate for the shrinking of the atmospheric window (from 40 Wm2 to 35 Wm2) the figure for radiative emission from the atmosphere will increase from 165 to 170 which keeps the system in balance with 235 Wm2 still outgoing.

If the atmosphere had no radiative capability at all then radiative emission from the atmosphere would be zero but the Atmospheric Window would release 235 Wm2 from the surface.

If the atmosphere were 100% radiative then the Atmospheric Window from the surface would be zero and the atmosphere would radiate the entire 235 Wm2.

==============================================================

Note: I’m glad to see a number of people pointing out how flawed the argument is. Every once in awhile we need to take a look at the ‘Slayer’ mentality of thinking about radiative balance, just to keep sharp on the topic. At first I thought this should go straight into the hopper, and then I thought it might make some good target practice, so I published it without any caveat.

Readers did not disappoint.

Now you can watch the fun as they react over at PSI.  – Anthony

P.S. Readers might also enjoy my experiment on debunking the PSI light bulb experiment, and note the reactions in comments, entirely opposite to this one. New WUWT-TV segment: Slaying the ‘slayers’ with Watts

Update: Let me add that the author assuredly should have included a link to the underlying document, Earth’s Global Energy Budget by Kiehl and Trenberth …

0 0 votes
Article Rating

Discover more from Watts Up With That?

Subscribe to get the latest posts sent to your email.

417 Comments
Inline Feedbacks
View all comments
Dave Worley
April 8, 2014 4:44 am

Convection carries water vapor above the densest layer of Greenhouse gases, where it is more likely to radiate into space. Hadley cells circulate a large percentage of the entire atmosphere in an up and down cycle.

April 8, 2014 4:47 am

This is not correct. Evaporation and convection are the main surface cooling fluxes. Most of the energy radiated to space by the atmosphere, got there by non-radiative means.

Nylo
April 8, 2014 4:50 am

Convection does have a cooling effect in the lower levels of the atmosphere which needs to be accounted for. Convection causes a mass of hot air near the surface to be replaced by a mass of cold air. This increases the loss of heat of the surface due to conduction to the air immediately on top. If the air didn’t move, loss due to this conduction would be lower (conduction losses depend on diference of temperature). It is true that, while descending later, the air gains as much energy as it lost while ascending. However, between the two moments, the air has lost additional energy due to radiation that took place while it was in the upper layers of the atmosphere. So the air returns colder than it left. What convection does is increase the temperature of the upper layers of the atmosphere with heat coming from lower layers. This both reduces the temperature of the lower layers and increases the outward radiation of the upper levels which means that it is a way to radiate the same but with a lower temperature in the lower levels. Which means it does cool the lower levels.

johnmarshall
April 8, 2014 4:58 am

There is so much wrong with Trenberth’s idea but let us start with his flat earth idea with 24/7 sunshine. Hardly realistic. He spreads solar energy over the whole planet’s surface but reality spreads it over half the ROTATING planet.
Evapouration is far too low. Every cloud formed includes latent heat so every cloud is evidence of heat being LOST from the surface as well as increasing albedo on formation.
Reality has 960W/m2 on the subsun point decreasing to 0 at the poles. The average is 480W/m2 which relates to a temperature of 33C not the -49C from Trenberth’s 167W/m2.
If you want a realistic model see Postma’s paper ”A Discussion on the Absence of a Measurable Greenhouse Effect. His model is realistic and simple to understand

JPS
April 8, 2014 5:02 am

Sorry but this post is nearly completely incorrect and extremely confused.

Nick Stokes
April 8, 2014 5:11 am

“Therefore neither ii) nor iii) should be included in the radiative budget at all.”
Who said it’s a radiative budget? The article you have taken it from (if it’s T&F2008) is titled “Earth’s Global Energy Budget”. Above the diagram, in big red letters there it says: Global Energy Flows Wm-2.
But the 78 W/m2 latent heat flow is hard to argue with. It is simply calculated from precipitation. The water that condensed left that amount of heat behind in the atmosphere.
The thermals flux is a nett upward flux. It is heat transport.
“I think it is generally accepted that the lapse rate slope marks the points in the atmosphere where there is energy balance within molecules that are at the correct height for their temperature.”
Reference?

hunter
April 8, 2014 5:13 am

Interesting conjecture. Do you have the calculations and the physics to support it?

Box of Rocks
April 8, 2014 5:14 am

This post is a start.
Thanks for starting a conversation on the diagram.
I have a sens that the original idea from Trenberth is incorrect, I at this point in time can’t put my finger on it.
The whole idea of 324 watts/m^2 of back radiation needs a good look. Just because it exist does not mean it does any work to warm the atmosphere.

April 8, 2014 5:17 am

trenberth’s, wilde’s, posma’s….which of the theoretical models contrasts better with empirical data or measurements…or is it we have not yet enough data to assess them, as prof. Freeman Dysson explained …

MikeB
April 8, 2014 5:25 am

Here we see the classic energy budget analysis supporting the hypothesis that the surface of the Earth is warmer than the S-B equation would predict.

This diagram is not intended to support any hypothesis whatsoever. It is a simple attempt to allocate numbers to various heat transport mechanisms. It is no more than that.

Neither ii) nor iii) should be included in the radiative budget at all?

The diagram doesn’t purport to be a radiative budget…it’s an energy budget….back to the drawing board for you
The S-B expectation for Earth is 33K? What does that mean?
Convection doesn’t cool the surface? Isn’t that obviously wong?
Evaporation doesn’t cool the surface either?
And finally, the surface of the Earth is not warmer than the S-B equation would predict. The amount of radiation from the surface precisely accords with the temperature of that surface as determined by S-B. It can do no other. You need to clarify that what you mean is when ‘seen from space’ the Earth system appears to be at 255K (and we know that the surface is much warmer).
Sorry, couldn’t read the rest.

Chris @NJSnowFan
April 8, 2014 5:30 am

If a normal everday person looked at this they would be lost.
I even find it confusing.

Editor
April 8, 2014 5:32 am

> the surface of the Earth is warmer than the S-B equation would predict due to 324 Wm2 of ‘Back Radiation’ from the atmosphere to the surface.
I’d be more inclined to say it “retards surface cooling” rather than imply it warms the surface. You really don’t want to wake up the Slayers….
Though it’s probably too late.

Martin A
April 8, 2014 5:37 am

If you work out what happens, starting with a cold planet and allowing its temperature to rise until there is equilibrium between incoming energy (primarily in the visible wavelengths) and outgoing energy (in the long infra red), you find that all the warming is done by the incoming sunlight. Back radiation is there, but all the warming was caused by the incoming light. So no need to cause confusion by talking about back radiation warming things

RobertInAz
April 8, 2014 5:40 am

Agree with the prior comments – the analysis needs a lot of work. It appears to me the 24 + 78 are “absorbed” by the atmosphere to be returned as part of the 324 back radiation or last as part of the 165 emitted. So, looking at the atmosphere we have
67+24+78 + 350 – 165 – 324 – 30 = 0.

Editor
April 8, 2014 5:41 am

The 78 Wm2 needs to be corrected to zero because the moist adiabatic lapse rate during ascent is less than the dry lapse rate on adiabatic descent which ensures that after the first convective cycle there is as much energy back at the surface as before Evapo-transpiration began.

Like other commenters have noted, IR radiation is more efficient at altitude than lower thanks to the bypassing a lot of the greenhouse effect. I don’t have a good sense for the difference between the radiation that escapes from the warm ground vs. the cold gas higher up. Also, in wet adiabatic conditions there are clouds and IR radiation from the ground is reflected and reradiated from the cloud base.
Thanks to latent heat release, wet adiabatic convection, think thunderstorms, gets a lot of IR radiating material higher in the atmosphere so it radiates better than dry convection. Also, rainfall cools the surface as it undergoes no adiabatic compression on the way down.

Crashex
April 8, 2014 5:41 am

As a long time reader and fan of WUWT, I just want to note that this has to rank as one of the worst posts ever. It is wrong on so many levels. This is the type of post that will be ridiculed by many for its lack of understanding of fundamental science in an effort to discredit everything else this site has ever posted.
To claim that evapotranspiration cooling should be omitted from an assessment of the heat transfer budget at the surface because it operates at a lower rate than radiation is ridiculous.
REPLY: I don’t disagree, but see my note below about the real reason I published this. +1 for your comment – Anthony

DirkH
April 8, 2014 5:46 am

“If a GHG molecule is too warm for its height then DWIR back to the surface dominates but the molecule rises away from the surface and cools until DWIR again equals UWIR.
If a GHG molecule is too cold for its height then UWIR to space dominates but the molecule then falls until DWIR again equals UWIR.”
Why? Let’s say mean free path length for an IR photon at 15 micrometer, in the CO2 absorption / re emission band is 25 m at 1 atmosphere. Meaning it gets emitted and re absorbed and re emitted multiple times on its way until it reaches either surface or open space (or a water droplet in the atmosphere, which acts as a blackbody). Each re-emission happening in a random direction.
In all cases this should result in the atmosphere being an opaque fog on this frequency , “shining” roughly the same amounts of IR on this frequency back to Earth and the other half to outer space.
I am assuming Local thermodynamic equilibrium, allowing for the application of Kirchhoff’s Law. As a GHG molecule travels in a parcel of air with the same temperature (to find its correct height), this should hold most of the time.
The climate modelers seem to think that GHG molecules swallow IR photons, not re-emitting them, leading them to call them “heat-trapping gases” and modelling a tropospheric hotspot that has not been observed in reality. At least I think that’s one of their mistakes.

NotAGolfer
April 8, 2014 5:47 am

They’re making it more complicated than it needs to be. The Beer-Lambert equation is used to determine the amount of heat absorbed by various levels of CO2 at various lengths. You need to integrate across the changing pressure profile and gas-mixtures as they change with altitude, which makes it complicated, but it is still much more straight forward than trying to determine what the actual temperature of the earth is, was and should be. Those are a fool’s game.
The Beer-Lambert can be used to accurately predict the expected change in temperature with any change in concentration, whereas the Steffan-Boltzman requires estimating emissivity and such. Using SB is like trying to determine the amount of solids suspended in a tank of water by using 2 different pressure transducers at top and bottom of the tank, then hoping they are calibrated, then subtracting… When you could just insert one end of one transducer into the bottom of the tank and the other end of the same transducer into the top to read the difference directly. Well, actually, using the Steffan-Boltzmann is much more complicating than this example shows.

Doubting Rich
April 8, 2014 5:47 am

“The 24 Wm2 for thermals needs to be corrected to zero because dry air that rises in thermals then warms back up to the original temperature on descent.”
I am afraid you are not correct here. This does have a net transfer of energy, as in thermal circulation the air warms to be warmer than the surrounding air then rises. The cooling in rising is adiabatic, so there is no energy transfer, as is the warming during sinking. So the warm air rising is taking energy it has absorbed from the ground. Once aloft it moves towards the cooler air column (which has lower pressure at altitude) and cool, thus sinking again.

Doubting Rich
April 8, 2014 5:53 am

How can the back radiation to the Earth from the atmosphere be approximately double the outward radiation from the atmosphere? Surely radiation is not directional, and while the atmospheric density and temperature fall with altitude so the lowest levels will radiate most, the upper levels allow much of that through and add their own, and the lower levels also absorb some of the downwelling radiation from higher parts of the atmosphere. Overall these should roughly balance out.
What am I missing here?

Doubting Rich
April 8, 2014 5:55 am

Just realised that al I needed for the last comment was “what DirkH says”.

Leonard Weinstein
April 8, 2014 5:55 am

This post shows the author does not understand the actual processes of the so called atmospheric greenhouse effect. While Trenberth may be wrong in some of his levels, his basic model is correct. The back radiation does not HEAT the surface, it is effectively a net radiation resistance, reducing the net surface radiation up to well below black body level. The numbers to consider are the 390 up minus 324 back radiation for a net radiation up of only 66 w/m2. The difference of absorbed radiation of 168 minus this 66 net radiation up gives an excess of 102 w/m2. This 102 is then carried up by the conduction, convection and evapo-transporation processes. Increasing back radiation by adding more greenhouse gases would result in the NET radiation up from the surface decreasing, and other processes increasing to keep the balance. this also would further increase the altitude of outgoing radiation and increase the surface temperature. It is the raising of the average altitude of radiation to space (by radiating up from the clouds and greenhouse gases) that results in the net average surface temperature increase. The lapse rate is a critical part of this process, since the average temperature is lower at the higher altitude where radiation to space occurs.

Roy Spencer
April 8, 2014 5:57 am

Sigh. 🙁

tty
April 8, 2014 5:57 am

“The 78 Wm2 needs to be corrected to zero because the moist adiabatic lapse rate during ascent is less than the dry lapse rate on adiabatic descent which ensures that after the first convective cycle there is as much energy back at the surface as before Evapo-transpiration began.”
This is nonsensical. The reason the dry adiabatic lapse rate is steeper is because dry air has a smaller heat capacity than dry air, so it will heat more for a given amount of energy. The 78 W/m is simply due to water condensing and precipitating, and leaving the heat of evaphoation behind.
“The 24 Wm2 for thermals needs to be corrected to zero because dry air that rises in thermals then warms back up to the original temperature on descent.”
If that was true it couldn’t get back down. It can do that only because it has lost energy at altitude.

Truthseeker
April 8, 2014 5:59 am

I know, let’s start by not having a flat Earth and have a spherical one, then we can add … I know, I know … a day/night cycle with planetary rotation … yes and then we could have … maybe just maybe … a Sun that actually provides 1370 W/m2 of energy to the top of atmosphere on the day side of the planet. Let’s try that and see how we go.
This is a cartoon, not a diagram and Trenberth’s most important piece of equipment seems to have been crayons. It does not matter what you do to it, you are still not going to end up anywhere useful.
Try this for a realistic diagram:
http://tinypic.com/usermedia.php?uo=fNkd6hpTbcMU7xs0vRLosYh4l5k2TGxc#.U0PyR156PRo

1 2 3 17