Sunday, April 19, 2009

End of the road

UPDATE: G&Trebuttal-2-6 is now posted. Pls use this one. Lots of small formatting changes to Figures and References. Should have all correction receive until 4/23@10 PM EDT. Adobe Acrobat version also available

Thanks to everyone who participated, the Gerlich and Tscheuschner rebuttal is pretty much finished (IEHO). It has been uploaded to
Rabett Run Labs. The list of authors is incomplete, and some who are listed may want not to be listed, others who are not listed but contributed may want I. Eli tries to please. It would also be good to have affiliations for the authors, although I think not totally necessary.

After waiting a week time for further comments (and yes our Welsh mouse who bravely volunteered, for proof reading), I will send the paper to IJMP for publication. We will see what their response is (sure to be amusing:) and proceed accordingly. At the least we will place this in arXiv and perhaps at some other sites.

To give everyone an idea of the flavor of the thing, which has really changed for the better, here is the first couple of pages of the introduction.

Comment on “Falsification of the atmospheric CO2 greenhouse effects within the frame of physics”

By: E. Rabett, and the Bunnies (add or subtract your name here) Christopher M. Colose, Chris Ho-Stuart, Joel D. Shore, Arthur P. Smith, Jörg Zimmermann

Abstract: Gerhard Gerlich and Ralf D. Tscheuschner claim to have falsified the existence of an atmospheric greenhouse effect. Here, we show that their methods, logic, and conclusions are in error. Their most significant errors include trying to apply the Clausius statement of the second law of thermodynamics to only one side of a heat transfer process rather than the entire process, and systematically ignoring most non-radiative heat flows applicable to Earth’s surface and atmosphere. They claim that radiative heat transfer from a colder atmosphere to a warmer surface is forbidden, ignoring the larger transfer in the other direction which makes the complete process allowed. Further, by ignoring heat capacity and non-radiative heat flows they claim that radiative balance requires that the surface cool by 100 K or more at night, an obvious absurdity induced by an unphysical assumption. This comment concentrates on these two major points, while also making note of some of Gerlich and Tscheuschner’s other errors and misunderstandings.

1. Introduction

In a paper that recently appeared in the International Journal of Modern Physics Gerhard Gerlich and Ralf D. Tscheuschner (2009, hereafter GT09)claim to have falsified the existence of an atmospheric greenhouse effect. Here, we show that their methods, logic, and conclusions are in error. The authors describe “problems” that are not really problems. They are either not related to the greenhouse effect, or well known and understood minor issues such as the differences between the mechanisms by which a glass greenhouse warms and that by which the greenhouse effect leads to a warmer surface. GT09 do not come to grips with how the greenhouse effect emerges at levels of analysis typical of the modern state-of-the-art, such as from line by line calculations of atmospheric radiative transfer, global climate models (GCMs) or even on the level of advanced textbooks (Pierrehumbert 2009), but rather criticize simple, didactic models for not being complete. They make elementary mistakes in doing so. Their analysis of GCMs is limited to quoting from unrefereed sources and pronouncing on what they consider to be the requirements for a scientific theory. Further, they suggest that a good theory of radiative transfer would require use of inappropriate theoretical tools, such as Feynman diagrams for calculating vibrational emission and absorption probabilities in the atmosphere or magneto-hydrodynamics for analysis of flow in the lower atmosphere. In the first case quantum theory provides the theoretical background (for example, such textbooks as Hollas, 2004, Bernath 2005), and spectroscopic measurements the data base for line positions and cross-sections (Rothman, et al., 2008), in the second, conservation laws and discretized versions of the Navier Stokes equations for fluid flow in an unionized medium (Washington and Parkinson, 2005, Imskeller and von Storch, 2001 are book length treatments of these issues) allow us to successfully model the major circulation patterns in the atmosphere.

A recurring element in Gerlich and Tscheuschner (2009) is that others neglect the thermal conductivity of the atmosphere, presented in the introductory (“Problem background”) section covering the first 4 pages of the paper and also Section 3.8 as well as elsewhere in passing. However, they fail to place this in quantitative context. In fact the heat flows associated with conductivity are tiny, and hardly different from the value of zero they repeatedly criticize in the work of others. For example, for a typical lapse rate temperature gradient of 5 K/km, the resulting heat flow from atmospheric conductivity is about 1.5x10-4 W/m2, (according to Table 5 of GT09) far less than the 10’s or 100’s of W/m2 associated with other atmospheric heat flow processes (Trenberth, Fasulo and Kieho, 2009). Motionless air is a fairly good thermal insulator. GT09’s repeated reference to thermal conductivity is indicative of a lack of basic understanding of the relative scales of the important processes in Earth’s atmosphere.

The second section of GT09 (“The greenhouse effect hypothesis”) presents some information about atmospheric radiative processes, but in comparison to their detailed treatment of thermal conductivity in the first section, the only quantitative value here is an estimate of the (large) number of CO2 molecules an infrared photon might see in one wavelength of travel. This, they claim, requires a complex microscopic theory in contrast to the macroscopic thermodynamic properties associated with thermal conductivity (raised yet again). GT09 fail to acknowledge that spectroscopic properties of gases are themselves well-defined macroscopic properties that have been experimentally studied in extensive detail (Rothman, et al, 2008). The absorption and emission of any given wavelength from a given collection of gas molecules can be accurately quantified, and has been verified in both laboratory experiments (Flekl, et al., 2002 and measurements of radiation passing through Earth’s atmosphere (Bianchini, et al., 2008)

The next 18 pages of GT09 are devoted to showing that the atmospheric greenhouse effect relies on different physical processes than the warming in a glass greenhouse. Their short concluding paragraph (2.6) would have sufficed. Concisely, greenhouses work by restricting the outward flow of thermal energy to the surrounding atmosphere by convection, while the atmospheric greenhouse effect works by restricting the outward flow of thermal energy to space by radiation. In both cases restricting outward energy flow causes warming, so the analogy is not as irrelevant as GT09 claim.

The problems with these three sections are characteristic of the remainder of the paper. First, the authors’ lack of quantitative familiarity with the field they are criticizing, second their claims of complexity or invalidity, impossibility, and occasionally fraud regarding well-established quantitatively verified analyses of atmospheric processes, and third their extensive diversions on topics that do nothing to further their own argument or a reader’s understanding. In the remainder of GT09 we could find only two physically based criticisms of the greenhouse effect that actually require any substantive response.

The first of these claims in GT09 is the assertion that Clausius' statement of the second law of thermodynamics forbids transfer of heat from a colder atmosphere to a warmer surface. However, the second law requires consideration of all heat flows in a process, so one must simultaneously include the larger transfer of thermal energy from the surface to the atmosphere. Clausius’ statement also does not impede temperature changes resulting from a change in system properties such as adding glass to a greenhouse, or adding infrared-absorbing gases to the atmosphere. Any warming under such a change comes from the gradual build-up of energy and heat flows to a new steady-state, not the transfer of energy from cold to warm regions.

The second physically-based criticism that we could find in GT09 is in their model of a hypothetical Earth with a non-uniform temperature distribution, and their resulting arguments regarding average temperature and the magnitude of the greenhouse effect as a whole. Following Smith (2008) which was a response to an earlier version of GT09 (Gerlich and Tscheuschner, 2007) we find that Gerlich and Tscheuschner obtain an absurd result by using a very unphysical assumption, that each part of the planet's surface immediately cools or heats to reach an equilibrium with the locally impinging solar radiation, thereby neglecting the thermal inertia of the oceans, atmosphere and ground and all other heat transfer processes within the atmosphere and surface. Were this to be the case, all parts of the Earth would immediately drop to almost absolute zero at night, and the discrepancy between Earth’s observed average temperature and the average on this hypothetical Earth would be very large, over 100 K. It is shown that a uniform surface temperature model gives a more realistic bound on the greenhouse effect, the commonly quoted 33 K. This value is a lower bound on the magnitude of the greenhouse effect and even Gerlich and Tscheuschner’s result for their unphysical case obeys this bound.

Section 2 below disposes of GT09’s arguments about the second law of thermodynamics, principally laid out in their Section 3.9. In Section 3, simple models are introduced and then used to demonstrate Gerlich and Tscheuschner's aphysical picture of heat flow on a rotating planet (their Section 3.7). These two points clearly demonstrate that the radiative greenhouse effect significantly increases the surface temperature of the Earth, contrary to the conclusions of GT09.

The rest of GT09 essentially follows their initial pattern of poor physical intuition, wild claims, and irrelevant diversion. One example of their wild claims is the denigration of a simple net energy flow schematic for the earth such as is shown in Fig 1 below and Fig. 23 in GT09, because such diagrams (pp 322)

Go RTFR at Rabett Run Labs


Ian said...

Eli, nice draft intro.

Did you know that you're quoted as a mistaken but "respected authority" by a G&T apologist? He is Hans Schreuder, a retired "analytical chemist." If you can stand it, take a look at ; your description of the general greenhouse effect is on the 2nd from last page. Surely some congratulations are in order...

Between Herr Schreuder and your new friend Inferno Jones, my cup runneth over with amusement.

Arun said...

Thank you! (for doing what was necessary).

Hank Roberts said...

Is your illustration from the update to Kiehl, rather than from the original 1997 paper cited?

All I have is this mention, not the actual copy, of the newer paper:

Hank Roberts said...

Can you take Open Office file format with change marks? I'd like to send a test file back showing a few suggestions. Tell me if you'd like it left on a page somewhere from which you can download it, or emailed.

Throwaway contact info for me, for the moment:

EliRabett said...

Open Office is fine. The files can be emailed to me at elirabett2003 at yahoo or left on the Rabett Run Labs page.


Arthur said...

Nice job Eli! I wonder if folks have any opinions about condensing the "non-physics" arguments a bit, or leave them as they are? As it is they're a bit spread out (section 1 and sections 4 and 5)? I think it reads ok, but it could be a bit more coherent on all that I think...

Horatio Algeranon said...

Well done Eli and Distinguished collaborators!

Maybe some time I'll write a ditty about the exploits of Tscheuschner and Gerlich (including the debunking/dunking they got from Eli and Friends, of course)

It's got all the ingredients of an epic tragicomedy.

..though to do it justice would take a considerable effort (and I'm not sure I'm willing to devote even a small fraction of the time you have to this)

chriscolose said...

There are still some typos.... I do not want to nit-pick too much but at this stage of development it is probably appropriate. Figure 1 should be "Kiehl" and not "Kieho" (and yes Hank, it is from the newer paper, which is now fully accessible in one of the comments provided under that posting on my site).

In the paragraph prior to the conclusion you should have "latter." and not "later."

These are based on just a very quick overview, so other commenters are welcomed to review in much more detail to pick out all of the minor issues.

EliRabett said...

Picking nits is an appropriate and social activity in your average burrow, so please, do so. I will make the changes on the fly, replacing the file at Rabett Run Labs as done.

EliRabett said...

I had the same thought as Arthur, but on balance decided to keep at least some of the non-physics arguments at the front, so the average reader would see them before the article got too much into the thermo woods. It establishes a tone. On the other hand moving all of them to the front would make is seem that there were no non-physics arguments. So on balance, although not perfect the ABA structure works for Eli.

Arthur said...

Figures 2, 4, and 8 seem to be missing (and some of the figure captions messed up) in version 2-4 that I just downloaded. On the other hand, I'm looking at this in Word 2004 on a Mac, so who knows what the problem really is...

Is there a more standard way of handling figures in Word?

EliRabett said...

Yeah, I had a lot of problems with the figures between the versions I received back. The text seemed ok. Maybe the best way of handling it would be to put up an Adobe Acrobat version? Tonight

Anonymous said...

Don't just leave it on your blog, get it peer reviewed - and if successful of course - it will be published. Just like Gerlich and Tscheuschner! That's what science is about.

Joel said...


That is indeed our plan, hence Eli's statement in the 2nd paragraph of this post: "After waiting a week time for further comments (and yes our Welsh mouse who bravely volunteered, for proof reading), I will send the paper to IJMP for publication."

Of course, if the journal editors of IJMPB had been doing their job, this all never would have been necessary...but given that they failed miserably, it is.

Hank Roberts said...

Couldn't get changemarks to survive back and forth between OpenOffice Mac and Word, so, dirty laundry airing time, starting nitpickery run:

Same question in text as Fig 1 caption:
(Trenberth, Fasulo and Kieho, 2009). Should it read
(Trenberth, Fasulo and Kiehl, 2009)?
(not found in Scholar, can't be sure

Is 'aphysical' a word?

"diagrams (pp 322)"
(p or p., if one page, but could be missing page numbers in a range?)

Inconsistent form:

"Using the Stefan- Boltzmann law we can calculate ..."
(hyphenated; extra space after hyphen; "nn" in "Boltzmann)

but elsewhere one 'n' no hyphen:
"Stefan Boltzman law."

"Petitit (1940)" could be a typo for Petit?
Thermal profiles of the eclipsed moon refers to Petit (1940) ...

Do the numbers cited to (Kiehl and Trenberth 1997) need to be updated from the new paper?

That's just the spellcheck runthrough; more to come.

Eeek! Eeek! comb...pick...crunch...

Hank Roberts said...

Inconsistent: period sometimes used after "Fig " and "p "

Missing period if there should be one after "This is scientific misconduct as well" -- but not clear why that whole paragraph indented at left but not right? If it's a block quote, indent both left and right (means the same as putting the whole long thing inside quotation marks) and follow with the cite or footnote.

The phrase:
"... misconstrue an argument or a figure and then present long arguments to show something is wrong that has not been claimed to be true."

--> means 'arguments against something they misattribute (mistakenly attribute) to the paper they criticize, that its authors do not say or claim therein'?

I could go back and gnaw on the difference between "that" and ", which" but the Europeans ignore it so I won't. But this needs help:

"found in Fig. 32 on page 340 (Fig. 3 above here. Note that the figure which refers to the ground and stratosphere and text which refers to the stratosphere and atmosphere ..."

Could it say:

"found in their Fig. 32 on page 340 (Fig. 3 above here. Note that their Fig. 32 (which refers to the ground and stratosphere) and their text (which refers to the stratosphere and atmosphere) ...)"

Or if the nesting parentheses bid fair to hatch gargoyles, perhaps break differently

"found in their Fig. 32 on page 340 (Fig. 3 above here). Note that their Fig. 32 (which refers to the ground and stratosphere) and their text (which refers to the stratosphere and atmosphere) ...."

Hank Roberts said...

"If the Clausius statement referred to any flow of heat when the two disks were placed opposite each other B would have to stop radiating towards A because if it did not, heat would be transferred between a body at lower temperature to a body at higher temperature."

Does "between" not imply both directions?


If the Clausius statement referred to any flow of heat -- rather than to net flow -- that would imply that when the two disks were placed opposite each other B, a body at lower temperature, would not radiate toward A, a body at higher temperature.

well, enough, I'm getting well beyond my education in attempting to clarify the poetry here. I'll try again with changemarks tomorrow, if time permits and more jumps off the page at me.

EliRabett said...

Thanks Hank. I've made (I think) the corrections you suggested with the exception of the that which and the misconstrue sentence. Have to think about rephrasing that

I think I have also fixed the pictures, but to be sure I have put up an Adobe Acrobat version.

Also, on the open office/MS Word issue, if the corrections don't carry over, print an Adobe Acrobat version from open office and send Eli that.

Current version is 2.5 at Rabett Run Labs

Joel said...

The PDF version 2.5 looks good. Here are a few more minor corrections that I had after a brief scan through it:

(1) There are two figures labeled "9" and none labeled "8", so presumably the first of those should be "8". Similarly, there are two figures labeled "2" and none labeled "3".

(2) Some of the figure captions are lacking a period at the end of the last sentence.

(3) Two of the figures that appear to be taken from other works (Figs. 7 and 9) still don't say where they are taken from.

Hank Roberts said...

> 2.5
change the note at top of thread, which still says 2.4, to catch attention

Anonymous said...

Note -- it should be "Kiehl" rather than "Kieho."

Anonymous said...

Oops. I see a couple of people already caught that one. Sorry.

Hank Roberts said...

Extra carriage return at "HERE" in the middle of the Santer cite:

... D. W. Pierce, "HERE"
Smith, A. P., 2008. Proof of the atmospheric greenhouse effect. arXiv

Only one space at end of sentence in that cite, before "arXiv" -- two is your standard.

Hank Roberts said...

Still not getting trackchanges to work, so I'm abusing the comments with nits:

In the references, first author:

L.S. Rothman,
should read
Rothman, L.S.,

In alpha order, Schmidt must follow Santer.

Hank Roberts said...

and there should be one more fairly long nitpick in the bin somewhere that didn't appear yet from this morning.

Hank Roberts said...

Still sez "UPDATE: G&Trebuttal-2-4 is now posted. Pls use this one"

But it would be wrong.

Looks like the long nitpick didn't survive. Look for capital letters followed by commas -- likely should be a period there, for initials. I forget what else; maybe tonight if time permits.

I found 2.5, but no 2.6, so assume 2.5 is still current?

EliRabett said...

Hank, look for 2.6 after midnight.

Hank Roberts said...

It's there.
At the top of the thread where it says "UPDATE: G&Trebuttal-2-6 "
the link actually still goes to 2-4
but editing the url gets 2-6.

EliRabett said...


Hank Roberts said...

Email sent with interim markup. More will follow. Lots of little trivial nits, crunch crunch eek eek, enough to be interesting.

Recommend inviting/encouraging someone with math/symbol/physics competence, ideally a fresh reader if you have a volunteer, slowly and carefully nitpick, character by character, through the non-English sections, looking for the same kind of odd spacing/font/not-quite-right/off-by-one kind of typo that are easy to miss when you know what should be on the page. That's what nitpickery is for.

Hank Roberts said...

Drat. I'm good with Word's wildcards but don't own it at home. Anyone have a pointer to equivalent regular expressions for NeoOffice on Mac?

Anyhow: inconsistency, sometimes a space between digit and capital C and capital K, when used for temperatures. Global for "any digit space C" and "K" to lose the spaces.

Horatio Algeranon said...

I realize it's kind of late for making suggestions, but here goes anyway:

Gerhard Gerlich and Ralf D. Tscheuschner claim to have falsified the existence of an atmospheric greenhouse effect.It looks like you have addressed T&G's main arguments (eg, about the 2nd law), but I wonder if it might be appropriate to put in a brief description of what it means to "falsify" something in the scientific sense -- ie, essentially what T&G must show (and failed to show) to make their case that there is no greenhouse effect:


1) experimental evidence that shows the opposite of what an atmospheric greenhouse effect would necessarily produce and/or

2) evidence that the greenhouse effect would actually violate some physical law (eg, 2nd law of thermo)

The pot on the stove example is obviously an attempt to show that you get a colder temp with the water than without, but I think it's worthwhile explicitly stating that "because T&G failed to demonstrate that the pot on the stove example is a valid analogy for the earth, they failed to falsify the atmospheric greenhouse effect"

And you could also add a sentence stating that "because T&G failed to show that the greenhouse effect would require a violation of the 2nd law [because their arguments were incorrect], they also failed to falsify"

Hank Roberts said...

Just left final email with attachments; done unless you want to put up another round and ask us to look for anything, or something in particular.

If you really want to challenge us ask us to cut the length by a third without taking out anything important.

fred staples said...

Horatio Algeranon makes a late but appropriate point.

"because T&G failed to demonstrate that the pot on the stove example is a valid analogy for the earth, they failed to falsify the atmospheric greenhouse effect"

G and T, as they are more commonly known, make the obvious point that a (massive) increase in energy absorption will cool the pot, not heat it. To rebut the argument Eli must show that it is not a fair analogy (earth the pot, CO2 and water vapour the absorptive materials) for the earth/atmosphere system. How does he do this?

“once you remove a large heat flow, for example by letting all the water boil away, the surface heats up”

So, Eli is saying, if absorptive material is added (to the atmosphere) the surface cools; if absorptive material is removed, the surface warms. Something wrong here, surely.

fred staples said...

Hank suggests reducing the length by a third without losing anything important. Can I suggest reducing the length to a third by removing everything which is doubtful, misleading, or plainly wrong?

The second law of thermodynamics looks simple, but is, in fact, extremely subtle. It is (as G and T point out) a statement about the difference between heat, work, and energy. Entropy is a measure of the quality of energy – its ability to do work or raise other temperatures.

Heat is not simply energy, and energy is certainly, and demonstrably, not heat. Think about the waste heat from a power station, in complete isolation from the rest of the processes involved. Why do we waste that massive amount of energy in the form of water vapour (the principal greenhouse gas, we are told) through the cooling towers into the atmosphere?

To use the waste energy productively (to warm your living room, for example) we have to find a sink at a lower temperature to which it can be transferred.

Heat is a process. By definition and everyday experience, it is the transfer of energy from a higher to a lower temperature. It (heat, not energy) will not go the other way, either from plate B to plate A or from the atmosphere to the earth.

In the atmospheric case, the sun warms the surface and the surface warms the atmosphere. The surface is the source and the atmosphere is the sink. It is impossible for heat (not energy, heat) to pass from the sink to the source. (The sink can insulate the source, of course, allowing the sun to raise its temperature, but that is a different story).

Grasp this, and Eli’s example of the three atmospheric layers radiating against each other becomes absurd.
It looks plausible. The results seem sensible, and, as G and T say, people think they understand it. It depends on each layer receiving and absorbing radiation from above and below, and re-radiating half up and half down as heat energy, capable of raising temperature.
The idea can be reduced to absurdity by demonstrating that the answer depends on the number of layers into which the atmosphere is divided, which is arbitrary.

To simplify the equations, assume that the net incoming radiation is W, and eliminate the constants by calculating the ratio of the top layer temperature, 255 degrees K, to the surface.
First consider an atmosphere of just one of layer, perfectly absorbing and emitting, half up and half back to the surface. If the solar radiation is W, the surface will receive 2W, (W from the sun and W from the atmosphere), and must therefore emit 2W. The atmosphere will receive 2W from the surface, return W, and emit W to space.
The temperature ratio will be the fourth root of 2W/W, or the fourth root of 2, which is 1.19. The surface temperature will be 1.19 multiplied by 255, which is 303 degrees K
Now assume two layers, as in Eli’s paper.
The top layer receives 2W, and emits W to space and W back to the first layer.
The first layer sends 2W up and down, and so must receive 4W, 3W from the surface and W from the top layer. The surface receives W from the sun, and 2W from the first layer, emitting 3W.
Our temperature ratio is now the fourth root of 3, or 1.315.and the surface temperature increases to 335 degrees K, which is Eli’s result. Three layers increases the ratio to the fourth root of 4, and the temperature to 381 degrees K, not bad for the stroke of a pencil on a diagram.

If n is the number of layers into which you divide the atmosphere, the ratio of Tsurface to T1 (the top) is the fourth root of (n+1). It is easy to prove, and is set as a problem in Grant Petty’s book on Atmospheric Radiation, Page 144. (Hank)
This reduces to absurdity the notion of back-warming radiation, as G and T (and R W Woods with his famous terrestrial greenhouse experiment ) point out.

Anonymous said...

Not that Fred Staples will be back, but from the EPA Endangerment Response to Comments document:

"For example, as a disproof of the greenhouse effect, the paper by Gerlich and Tscheuschner presents the example of a pot of water, noting that the bottom of the pot will be cooler if it is filled with water than if it is empty. Contrary to the assertion in the paper, the primary thermal effect of adding water to the pot is not a reduction in heat transfer, but rather an increase of thermal mass. We assert that a more appropriate example for the paper to have examined would have been the addition of a lid to a pot of
water, which reduces the rate of heat loss, and leads to an increase of heating of the water compared to a case with no lid"


Anonymous said...

Also, Fred's key error in the "layer" example is not changing the absorption of the layers when you divide them in half: eg, what he is doing is effectively adding a 2nd atmosphere on top of the 1st when he adds a layer. Instead, he should determine what the total direct transmission of IR from the ground to space is (say, 10%). So with one layer, it transmits 10%, but with two, each layer transmits 32%. (eg, sqrt(.1)). And so on... this leads to an asymptotic solution as N approaches infinity, rather than an infinite solution.