Physics of the atmospheric greenhouse(?) effect

by Judith Curry

The skeptics thread has shown that it is plausible to be skeptical of a number of issues regarding the findings of  IPCC WG1.  However, whether atmospheric gases such as CO2 (and H20, CH4, and others) warm the planet is not an issue where skepticism is plausible.

Nevertheless, we have seen serious arguments (and even publications) against this theory:

The authors of these arguments believe that they should be convincing to other scientists.  In that sense, these are serious arguments.  While these arguments use physics, they are nevertheless erroneous.  The G&T paper has been rebutted by Halpern (aka Eli Rabett), Colose et al., with a subsequent retort by G&T (both groups seem to be talking past each other).  Are there any others?  I am also unaware of rebuttals to Johnson?

The fact that such papers are being written by scientists who take themselves seriously and are being published implies to me that scientists have done a poor job of explaining and making the case for warming of the planet by gases such as CO2.  Its easy to roll our eyes and mutter “cranks” when we see something crazy such as the sophistry in the little pamphlets put out by various anti-AGW advocacy groups.  But these arguments refuting atmospheric warming by CO2 are being made by scientists that take themselves seriously on this issue.

So what have we been doing wrong in terms of explaining this, to other scientists, the technically savvy public, and the broader public?  Our general argument consists of the following elements:

  1. a narrative with history, usually starting with Fourier, Tyndall, Arrhenius (see for example WeartWikipedia)
  2. the analogy to a greenhouse
  3. evidence for the existence of the planetary greenhouse effect (e.g. effective black body temperature, spectral IR measurements)
  4. the IPCC consensus

Well, #2 and #4 hinder rather than help.   The greenhouse analogy has not served the scientific analogy very well, made painfully obvious by the G & T paper; John Nielsen-Gammon has a lucid new post on this, which looks like the start of an interesting series.  The IPCC reports never actually explain the physics of the greenhouse gas mechanism.  The explanations you find in popular books and undergraduate texts mainly address the issue from some combination of points #1, #2, #2.

There are graduate level texts on atmospheric radiative transfer, including:

Once you digest one of these books, you will have no trouble understanding how this works.  However, these texts are pretty heavy going in terms of physics and maths.

There is a big gap between the simple explanations and the radiative transfer texts.  The blogosphere has stepped in to fill the gap.  Good explanations that I have come across are:

However, a gap remains in terms of explaining the actual physical mechanisms.  Yes, these sites give good explanations of the basic physics of radiative transfer and the Earth’s radiative energy balance, and provide empirical evidence for the existence of the greenhouse effect.  But a good mechanistic explanation of the physical processes occurring seems absent, including an explanation of how local thermodynamic equilibrium is established in response to the absorption of infrared radiation by a small number of molecules.  I don’t have a full understanding of what the actual issues are with the greenhouse effect skeptics (I suspect that Roy Spencer is painfully aware), but I have just received a copy of Slaying the Greenhouse Dragon, which I will read this weekend.

I don’t think the issue of not or mis- understanding the greenhouse effect is salient just for the public and a few seemingly confused scientists.  I have to wonder how many scientists on the PNAS list that supports the consensus (including the biologists and economics) actually have a good understanding of the physical processes and have taken a graduate course in atmospheric radiative transfer.

We need to raise the level of our game in terms of explaining the planetary warming by infrared absorption of CO2 etc.   The missing area of understanding seems to be the actual physical mechanism.  Lets target an explanation at an audience that has taken 1 year each  of undergraduate physics and chemistry, plus calculus.  Once we have something that is convincing at this level, we can work on how to communicate this to the interested public (i.e. those that hang out in the climate blogosphere).  Willis Eschenbach’s help is needed in translating this for the WUWT crowd.

Thoughts on how to approach this?  An excellent start was made on this thread. See Chris Colose’s take here, which explains it in a way that I haven’t seen before.

Moderation note: this is a technical thread and comments will be moderated for relevance.

698 responses to “Physics of the atmospheric greenhouse(?) effect

  1. Judith,

    I find many people in science have a hard time to grasp the concept of mechanics in motion on a molecular level. Compression and storage of energy is an extremely complex system as it can employ more than one energy or mechanical response.
    A piece of plastic is flexible yet will shatter at a certain temperature below freezing as it carries molecules of water in the process of making the product.
    Objects have a center of balance and the more complex objects then have an overall center of balance with all the different components inside having different densities having their own center of balances. These shift in motion, compressing and expanding. Flexible on a molecular level.
    Gases, are effected by temperature, pressure and motional energy to change their densities and compress into a whole new form of liquid storing vast amounts of energy.

  2. When physicists discuss the use and abuse of the Second Law of Thermodynamics as applied to climate science I listen closely. The G&T09 article (Falsification of the CO2 Greenhouse Effect), the Halpern (2010) rebuttal and the subsequent retort by G&T leave me “uncertain” as it appears that each author does not speak directly to each objection. Earlier Dr. Curry, you objected to a cartoon depiction of climate mechanics still available on a NASA web site as a way of dismissing an article that argued against CO2’s major role in a greenhouse effect. My experience with turbulence at the boundary of the air:water interface, and some familiarity with the limits of the Reynold’s Number to describe turbulent flow (uniform density for one), in part has me “not so sure” in accepting each author’s arguement wholely. I do not have a unified replacement climate hypothesis for the CO2 driver one in use today. I am intrigued by the moon’s tidal influence upon our atmospheric sea, the sun’s energy output spectrum, the photons that seem to rattle around in clouds, alien cosmic rays helping to make clouds, orbital excursions of planets and their gravitational pull on the sun as well as the limits (equilibrium condition assumptions) and boundaries (tail end of equations which are assumed to be small and therefore can be ignored) of the laws of physics including the conservation of energy. I remain in a “muddle” so I keep plodding through each arguement and sometimes, paywalls. Thank you for your patients. I may sit in the back of the classroom, yet I am still listening.

    • “I do not have a unified replacement climate hypothesis for the CO2 driver one in use today.”

      RiH008, simplistic as my mind is – and perhaps hubristic at the same time – I’d suggest this, and I’d propose that each be taken and studied with an attempt to falsify it:

      1. What warming there is is local, having to do with urban heat islands, and this because of land use changes as opposed to atmospheric chemistry. We know there is some UHI, so the questions become how much and does it affect the rest of the globe (i.e., the non-urban stations’ records).
      2. It is local because many, many rural stations show no warming at all in the raw data. Global means widespread and more or less ubiquitous, therefore if the heat from urban heat islands is not affecting locations 50-100 miles away, then how truly global is it?
      3. The apparent overall warming in some studies is due to incorrect adjustments at various points in the methodology, leaving artifacts of the processing to give the impression of warming when there is no intrinsic warming. True determination of warming can be accomplished by the inclusion of the many thousands of stations currently left out of the GHCN database, meaning that the coverage of the planet is more uniform.
      4. The effect of station near-environments has not adequately been taken into account as to micro heat islands, and when that is factored in the warming at those stations will be less than it is currently be thought to be. What the amount is we cannot know now; I predict (along with many others) that this effect will substantially erase the apparent warming trend in the existing studies.

      MANY skeptics do not doubt that we have done things that affect the readings on our thermometers. However, this group argues that CO2 is not the culprit, but land use due to population growth. How much that effect is, overall, has not yet been studied – in big part because the demonization of CO2 has swallowed up the vast majority of study grants. We argue that it will dwarf the CO2 effect.

      The cure for this warming is threefold: To site met stations properly to correct the erroneous readings – in almost all cases toward the cool end of the spectrum. To also arrive at accurate site-specific UHI values and use them. And to reduce the population of the planet, especially in regards to centralizing population in urban areas.

      When all is said and done, CO2 will be found to have been an insignificant player, and all the money saved by not slaying a dragon that was harming no one can be used to improve land use.

      And finally, since those rural stations aren’t warming – meaning little to no heat is making its way to them before being dispersed up off the ground – we will find that even the land use changes will have only modest effects.

  3. My hope is that Dr. Nielsen-Gammon will post on this here at more length.
    Dr. N-G has officially declared the greenhouse effect dead as a way to describe the behavior of gasses in the atmosphere.
    Credit to the good Dr. for having the strength of commitment to take this position.
    I like the name he has chosen, by the way.

    • Dr. Curry,
      I should have finished my coffee prior to posting the above post. You clearly saw Dr. N-G’s post prior to my post. In my coffee deprived state, I skimmed right past your reference to it.
      I still hope he will comment here on his clarification.
      I would point out that the various terms using ‘greenhouse’ and now apparently being discarded were all brought to the public square by the climate science community.
      If that family of terms are now agreed to be inaccurate, why is that a problem for the skeptical community?

      • If I understand this fully, I am gratified that the greenhouse effect is being laid aside, finally.

        But lest we forget:

        It is not just A greenhouse effect we are talking about, but a “Runaway Greenhouse Effect,” that one invented by Carl Sagan to explain the 800-900°F surface temperature on Venus (with its 96% carbon dioxide atmosphere) and then applied for well over a decade by people who claimed it also applied to the Earth (with its 0.03% carbon dioxide). Sagan invented it to spare astronomers the anguish of having Immanuel Velikovsky be right in his exactly correct prediction of that surface temp of Venus. It was, in fact, nothing more than a speculation, but it was seized upon by science – because they had no other aces up their sleeves. It would not do to let the infamous Velikovsky have the last word.

        So, the speculation of Sagan about one planet was applied to a planet with a massively different atmosphere. All of the Nitrogen and all of the Oxygen (not to mention the Argon) in Earth’s atmosphere was conveniently overlooked.

        In the haste to scare people into seeing CO2 as an imminent threat, we we the public were bombarded with images of melting glaciers and icecaps, and everything up to boiling oceans. And, the global warming community may or may not currently use what they now call merely “the greenhouse effect,” but we all remember it as “the runaway greenhouse effect.” The climate warming community still uses the greenhouse arguments when they talk about climate “sensitivity,” though without calling on the tainted term “greenhouse effect” so much anymore. “Sensitivity” as they use it still means “runaway,” whether they run from “greenhouse effect” or not.

        It was a speculation, not a fact, when Sagan conjured it up out of the mists in the 1970s and it remains a speculation. In fact, no one knows why Venus’ surface temperature is so hot. There are hypotheses and little more. In the 1970s there weren’t even good hypotheses, only a guess. When it was used in the 1980s and 1990s as a foundation for the threatened (“runaway”) global warming, it was still a speculation. Being revisionists about the AGW history will not let the (runaway) global warming community soft-pedal their past; we can remember what they tried to sell the world on. And the biggest term was “greenhouse effect”, preceded by “runaway.”

        I do recall a couple of years ago on WUWT when this greenhouse was discussed and it was agreed on that “greenhouse effect” was an unfortunate and incorrect choice of terms. Since then – amazingly, and maybe merely coincidentally – the AGW folks stopped using it, forsaking it for “re-radiation.”

        I just wonder if we will hear of “runaway re-radiation ” sometimes soon. If so, they don’t have Sagan to be their Galahad. It would have been his kind of exaggeration. (No, I am not a Sagan fan. I thought he was the most boring, most posturing scientist on the face of the planet…)

      • “It was, in fact, nothing more than a speculation, but it was seized upon by science – because they had no other aces up their sleeves.” should have been:
        “Sagan’s speculation was, in fact, nothing more than a speculation, but it was seized upon by science – because they had no other aces up their sleeves.”

      • feet2thefire, you say “If I understand this fully, I am gratified that the greenhouse effect is being laid aside, finally” but it has not been laid aside. The “greenhouse effect” is still a reality but the IPCC’s speculation about how significant any increase in atmospheric CO2 is seriously exaggerated. Roger Taguchi has provided excellent anayses here, particularly in his comments of Feb. 22, Feb. 7 & Feb. 9 (the last 3 major posts) which I recommend that you read carefully.

        The IPCC’s claim that our continuing use of fossil fuels is leading to a “Runaway Greenhouse Effect” causing catastrophic changes in the different global climates (CACC) is nothing but politically motivated propaganda having nothing to do with trying to take over Nature’s job of controlling global climates. There is speculation that the UN’s real agenda include the redistribution of wealth from taxpayers in developed economies, some possibly to go to give aid to underdeveloped economies but the bulk to go towards:
        – establishment of a framework for global government,
        – enhancing the finances of a privileged few.
        Such speculation may not be far from the truth. There are plenty of power-hungry individuals involved in this scam.

        Best regards, Pete Ridley

    • It takes no ‘strength of commitment’ to say what everyone already knows, the ‘greenhouse’ effect is a poor choice of name or analogy. But analogies, by their nature, are usually deficient in many ways. An atom, as taught to me in early science classes, is like a ball. Well, it is, but that’s also a highly deficient analogy, but it still gives you an initial idea to hang your understanding on.

  4. Good Morning,
    I have seen no news announcements of a revolution in climate thinking, and a withdrawal of the IPCC-sponsored consensus, so it is clear that what anyone has written so far has not been accepted by the other side. I suggest that any hypothesized explanation on the molecular level should not be considered fundamental — Joe LaLonde’s post above sounds like Descartes on vortices in the ether, explaining everything and nothing. We physicists will always insist upon explanations that work on the direct observational level, like “heat rises”, or “you can’t get more out than you put in” (and pretending you can, if you only imagine two competing large flows, that mostly cancel each other, is intellectual fraud — you are deluding yourself, and anyone you insist upon “educating”).
    I consider it proven that Venus’s heat is not due to the greenhouse effect as promulgated by the IPCC scientists, by direct comparison of the Earth and Venus atmospheres over corresponding pressure levels in each. I have done it here, and I afterwards read in Gerlich and Tscheuschner’s recent articles that they specifically recommend doing this.) Not only is Venus’ s atmosphere largely opaque to incoming light (visible freq. range), so the surface is cut out as a source of heat for the atmosphere, but the temperature at corresponding pressure levels of the two atmospheres is very closely in direct (fourth root) ratio with the incident solar power in the two cases. This directly indicates the two atmospheres are heated in the same way, and that must be: directly, by the IR portion of the Sun’s incident energy. Ah, this is why it gets colder as I drive up into the mountains, and why the snow stays, above a certain altitude — the surface is not warming the atmosphere, the atmosphere at altitude is cooling the surface. But don’t forget the ocean, as a heat sink/source, and El Nino et al, as boundary conditions. A pretty mystery to unravel, and you won’t unravel it by quantum mechanics or isolated examples of molecular interactions. Heat transfer is a bulk effect, and essentially, maddeningly diffuse, like water flowing through your fingers, or any sieve you try to hold it with — you can’t direct it, back to the surface for example. Heat does rise (except in a temperature inversion, localized and transient). You can’t trap or slow down IR radiation with air, or keep it from giving up its energy to the air, if it can, as soon as it can. It is just another path for the heat to take, not an obstacle to it. But I sit in an empty room; no one is listening — are they? And I am, thankfully or sadly, not a climate scientist, just an older, independent physicist. No doubt, I have overreached. You’re going to have to figure it out for yourselves, won’t you?

    • Sounds about right to me Harry.

      Given that net LW cools the surface at sea level by ~70W/m^2 we should only be interested in the height in the atmosphere the radiation to space takes place at. The rise in Co2 has theoretically raised that by 1-1.5% or 100 to 150 meters.

      Is this a big deal or small potatoes?

    • I’m sorry, an error in my link (I did it from memory, wrongly), it should be:

      Venus: No Greenhouse Effect

      • Someone needs to tell Hansen and others, since he has been selling the idea of a runaway greenhouse effect ‘like Venus’ happening in on Earth if we do not do what he wants.
        The more I think about this, I think the abandonment of the ‘greenhouse’ is not simply a rose by any other name. This was a major selling point of the AGW movement.

      • No, tallbloke, I can’t agree (though I wish everyone could), the very close Earth/Venus comparison tells me there is no measurable greenhouse effect at all on Venus, or on Earth (which has much less carbon dioxide in the air). At a lapse rate of -6.5K/km, the effect you mention would mean a temperature rise of nearly a full degree, easily measurable and thus wrong, based on what the clear Earth/Venus data tells me. The fundamental disagreement between climate scientists and physicists like myself is the trust the former put into simplistic radiation transfer arguments, which we see has led them into darkness. The simple way I look at it, the temperature lapse rate depends on the bulk specific heat of the atmosphere, and at 400 ppm, carbon dioxide cannot affect that. No doubt climate radiationists would insist that it can, somehow, through its IR absorption, but I would insist that it cannot, because that absorption (and emission) cannot hold back (as they believe it can) the relevant heat transfer demanded by the lapse rate. The bulk properties of the whole atmosphere cannot be affected by a negligible fraction of it — which is what a basic physical intuition would suggest even without hard data. I consider changes in the IR-absorbing constituents of the atmosphere rather as homeostatic controls against global temperature changes, a negative feedback against runaway temperatures, either up or down. As I see it, when the temperature goes up, the oceans release carbon dioxide, which provides an improved heat transfer upwards, and out to space, thus mitigating the temperature change. Rather than speculating on the degree to which the “greenhouse effect” can effect such a change, I think it’s a matter of how big is the homeostatic effect of added co2 (and water vapor) against the change. As a physicist, I suspect it is a large fraction of the unobstructed change itself. There is no equilibrium, of course, so this simple view is certainly not all there is to know.

      • James Belanger

        “The bulk properties of the whole atmosphere cannot be affected by a negligible fraction of it — which is what a basic physical intuition would suggest even without hard data.”

        A statement like the above is easy to disprove: Phase changes of water vapor has a tremendous impact on the properties of the atmosphere yet makes up only 1-4% of the total atmosphere (a negligible fraction). Another example where heuristic intuition leads to an incorrect conclusion.

      • Phase changes of water vapor are localized, and don’t change the lapse rate overall, otherwise we wouldn’t even be talking about a lapse rate as a general condition of the atmosphere. You haven’t disproved anything.

      • I agree, minute amounts can make a huge difference.

        If I leave a bath full of water but with a tap dripping overnight, each drip might be a negligible fraction of the total in the bath but it will still overflow. I can make it an even smaller fraction by using a swimming pool as my example, the drip will still cause an overflow.

        Seemingly minute changes in the concentration of CO2 in the atmosphere could have a similar effect.

      • Louise, your swimming pool example shows exactly the tiny local effect that Harry mentions, as evaporation from the pool surface would greatly exceed the amount of incoming water from the dripping tap. Now, if you are talking about an indoor swimming pool, that is another matter.

      • Isn’t the whole issue of sensitivity analogous to ‘we don’t know the rate of evaporation’ in this example.

        We do know that we have a dripping tap and the drip is getting faster (ie we’re increasing the rate at which we (humanity)release CO2 into the atmosphere) but we don’t know if the evaporation will keep up with this increase.

      • Latimer Alder

        And when it does overflow it will do so at exactly the same rate as your dripping tap (ignoring any effects due to evaporation). You will end up with a small puddle on the floor, just as if there was no swimming pool in the system at all.

        I fail to see the point you are trying to make.

      • The analysis on your website has serious problems.
        The pressure range of your analysis corresponds to an altitude of ca 45-60 km. The Venusian cloud tops that reflect 70% of the incident light are at an altitude of up to 60 km. The ratio of the albedos of Venus and Earth near cancel out the 1.91 incident radiation, so the net radiation is in fact about the same as on the earth, which demolishes your calculation and hence your conclusion.
        So all you are doing is comparing the adiabatic lapse rates of the two planets and saying they are the same. Surprise surprise.

  5. For me the Georgia Tech Link goes to the same destination as the pnas-Link.

  6. The link to the publication by Gerlich and Tscheuschner does not seem to work.

  7. others?

    Miskolzci seems to be a glaring omission here.

    • thx, i’ve added it

      • David L. Hagen


        However, whether atmospheric gases such as CO2 (and H20, CH4, and others) warm the planet is not an issue where skepticism is plausible.

        In so stating the problem, you are close to denying the scientific method. Skepticism is ALWAYS possible. e.g. the issue of the magnitude of the effect vs “if it exists”. Correct modeling etc. Compensation by water vapor/precipitation/clouds etc.

        Physicists continue to to test gravity, relativity etc as measurement capabilities increase.

        Ferenc Miskolczi is one of a handful of world class experts who have mastered the Line by Line quantitative radiation transfer. See HARTCODE – it has 3000 line radiation models with ppm resolution calculations!

        See Zagoni The Saturated Greenhouse Effect of Ferenc Miskolczi
        He dig into Milne’s 1922 original greenhouse equations and discovered and fixed a major flaw in the boundary conditions – the assumption of infinite thickness. That corrected a major step change in temperature at the earth’s surface etc. See Zagoni p 41-45 etc.

        He reviewed radiosonde data and discovered major errors in Kiehl & Trenberth’s 1997 etc See Zagoni p 68, 69

        Using HARTCODE, he has discovered that the upward and downward radiation are almost exactly equal.
        The major controversy is over his modeling up/down radiation up as about equal and the consequences for the 1D average.

        The next step is to quantify the small differences that become significant primarily at the outer atmosphere.

        He knows more about greenhouse gas radiation than 99% of your readers/posters. Take his results seriously for the insights they provide and see how to correct/build on them.

      • The greenhouse effect and its magnitude is one thing, debating over whether CO2 warms the planet through infrared emission and absorption is another; I don’t see any point to debating the latter.

      • Re: Miskolczi:

        If you know a little calculus, you can find the mathematical internal inconsistency in the derivation of his ‘new solution’ to the Milne problem.

        If you understand a little classical radiative transfer, you can discern the fallacy of his ‘new solution’, even without doing the math.

        If you understand some basics about atmospheric structure (see Nullius in Verba, this thread) you can appreciate the irrelevance of his ‘new solution’ in any case.

        If you know some basic physics, you can readily spot the phony energy conservation equation he invokes.

        If you understand some aspects of atmospheric radiative transfer, you will see why the data M presents in support of his theory, really does nothing of the kind.

        You can do this for yourself.

      • David L. Hagen

        I recall enough high school calculus to know that a finite atmosphere boundary condition will give different results from an infinite atmosphere. Furthermore, Milne 1922

        did not realize that the classic Eddington solution is not the general solution of the bounded atmosphere problem and he did not re-compute the appropriate integration constant. This is the reason why scientists have problems with a mysterious surface temperature discontinuity and unphysical solutions, as in Lorenz and McKay (2003).

        Perhaps you will honor us by providing the corrected solution for radiation absorption in a finite atmosphere compared to Milne’s 1922 infinite thickness assumption based on Eddington. Please detail the errors in Miskolczi’s 2007 derivation.

        Then perhaps you could enlighten us as to the differences between conservation of energy, minimization of entropy, and maximization of entropy of climate.

      • David – The Milne problem admits only two boundary conditions, generally applied at TOA and taken to be that (1) the outward LW radiation be equal to the total flux generated within the atmosphere (in the case of stars) or the impinging SW radiation (in the case of a planetarty atmosphere), and (2) the inward LW radiation is zero (no LW radiation from space). Imposing a further arbitrary boundary condition at the base of the atmosphere, as M wishes to do, generally causes one of the TOA boundary conditions to be violated (number 2, above, in M’s case), which renders any such solution unphysical. This property reflects a limitation of the Milne problem and the approximations used to derive it. In many atmospheres, this limitation is resolved by the realization that strict radiative energy balance does not apply in the underlying convective region; one can readliy imagine other conditions under which the Milne formalism fails to be a good approximation. Thus, the application of a BOA boundary condition is neither necessary nor correct.

        Furthermore, M’s derivation (in Appendix B) is internally inconsistent. To find a functional extremum in his integration constant Bo (as required by his ‘energy minimization’ hypothesis), he sets a derivative with respect to Tau_a (total optical depth) equal to zero, holding Bg constant (eqns B9, B10), and so derives a solution for Bg which explicitly depends on Tau_a, in self-contradiction. In fact, using the ‘solution’ for Bg (eqn B11) in his expression for Bo (eqn B7), one finds that there is no extremum in Bo.

        M’s eqn 7 is the phony energy conservation equation; it has no physical basis.

        (I should say that, upon first contemplating these apparent flaws, I contacted Dr. Miskolczi by email, to which he graciously replied. Unfortunately, his explanations made no more sense to me than his paper did.)

        Further criticisms of Miskolczi’s work, of which you are probably aware, can be found at:
        Roy Spencer
        Nick Stokes
        Paper by Rob van Dorland1 and Piers M. Forster (sorry, can’t find the link right now)
        Also, comments by Arthur Smith on various threads, and Neal J. King on Miskolczi’s mysterious application of the virial theorem:
        NJK on virial theorem

        Finally, I see that my comment above has a condescending tone that I did not intend; for this I apologize. I simply wanted to say that many of the flaws in Miskolczi (2007) are readily apparent to anyone who takes the trouble to carefully examine the paper.

      • David L. Hagen

        Thanks Pat.
        Accepted. Apologies for responding in kind.
        I will look at that those derivations more carefully.

        Appreciate your clarification for the assumptions SW/LW division at Top of Atmosphere.
        Doesn’t Miskolczi also change from an infinite to a finite atmosphere?


        “Imposing a further arbitrary boundary condition at the base of the atmosphere, as M wishes to do, generally causes one of the TOA boundary conditions to be violated (number 2, above, in M’s case), which renders any such solution unphysical.”

        I don’t follow. Why don’t the planetary surface conditions matter? Radiation, conduction, convection at the surface should impact the temperature and SW/LW near the surface.
        e.g. Miskolczi assumes 3) a black body interface
        (SW Absorptivity/ LW emissivity = 1)
        He also assumes
        4) that the gas temperature at the surface and the surface temperature are equal.
        How do 3) and 4) cause problems to your 2) of inward LW at TOA?

        At the extreme, won’t a planetary surface A/E = Zero affect the solution. i.e., no SW surface absorption. Similarly with emissivity of 0.01 instead of 1, reducing LW radiation?

        (I presume a quantitative generic case would combine 1+2 and use a line by line radiation model. e.g. assume the total outward radiation be equal to the total flux generated within the atmosphere (in the case of stars) or that impinging radiation (in the case of a planetary atmosphere), plus any planetary internal heat generation.

        Radiation, gas composition & planetary mass all affect the atmospheric profile or lapse rate. For an almost complete thermodynamic solution to the atmospheric profile. See:

        Prediction of the Standard Atmosphere Profiles of Temperature, Pressure, and Density with Height for the Lower Atmosphere by Solution of the (S−S) Integral Equations of Transfer and Evaluation of the Potential for Profile Perturbation by Combustion Emissions
        Robert H. Essenhigh, Energy Fuels, 2006, 20 (3), 1057-1067 • DOI: 10.1021/ef050276y)

      • David — please see reply below (to avoid margin scrunching)

      • David L. Hagen

        Pat thanks for the links. I have pulled them and am reading them.

  8. Leonard Weinstein

    If you would like to see a couple of write ups on the atmospheric greenhouse effect by a couple of skeptics, please read the following (and some of the comment discussions). Just because some skeptics don’t accept the valid physics (and many supporters of CAGW assume large feedback and tipping points that are not supported by anything real), don’t tend to lump them in one batch. I think many of the points made by G&T and others are valid but incomplete (although there may also be errors, but I don’t specifically know that). I think both sides are tending to talk past each other, and misunderstanding what the other claims.

  9. Judith Curry says:
    whether atmospheric gases such as CO2 (and H20, CH4, and others) warm the planet is not an issue where skepticism is plausible

    For Miskolczi, this is not the issue. The issue is whether additional ghg’s are going to warm the atmosphere and ground further.

    And I’m still awaiting your explanation of how LW radiation can heat the oceans, as your bare assertion of the ‘fact’ doesn’t satisfy my need for falsifiable content.

    • tallbloke, i wrote a long response on the ocean heating, if you can find your original query, it should be there, it was a lengthy discussion on ocean surface skin temperature and ocean mixing.

      • Judith, I can’t find your response. All you siad in response to my exposition about the inability of back radiation to heat the ocean on the ‘best case’ thread was:
        “I and some colleagues are working on a new high resolution satellite data set of ocean surface latent heat (evaporative), that works under conditions of extreme fluxes found in the high latitude oceans, warm currents, and hurricanes. More on this soon.”

        Gissa clue. :)

      • Judith has a good response but there’s an even more important to make than the fact that back infrared radiation can heat the ocean.

        That is, increasing the Co2 content of the atmosphere does not primarily increase the greenhouse effect by creating more back radiation to the surface. Rather, increasing CO2 primarily affects what’s happening higher at altitude, and the whole troposphere-surface system pretty much warms in concert as the planet becomes a less efficient emitter to space. In fact, this increase in temperature will generally cause more of the increase in downwelling IR flux than the direct increase in CO2 will– this is particularly true if the boundary layer is already a near-perfect emitter at its temperature (such as in the moist tropics). The additional downwelling IR flux is just one of many terms in the surface energy budget.

      • The emitter issue seems to be what the sample chapter of Greenhouse Dragon is about. I take it the boundary layer is not a “near-perfect” emitter elsewhere.

      • You are talking about the TOA equilibrium with the sun and space I take it. The down-welling flux into the ocean will also increase, which if I understand the mechanism correctly, will create more water vapor from the skin of the ocean. All else being equal, this in turn should produce more clouds I think. The clouds could then reflect more incoming SW back into space. This would have the effect of lowering the TOA equilibrium height. I don’t understand why the cloud feedbacks are better understood.

      • I don’t understand why the cloud feedbacks are NOT better understood.

      • Jim–

        Cloud feedbacks are complicated because there is no simple relation between the amount of water vapor in the air, how much evaporates, etc and cloud cover. Furthermore, once you actually figure out if the cloud forms, its radiative effects dependent on its altitude, optical properties, and has differential day time and night time effects.

      • I was aware that different clouds = different effects. It just seems what with CERES data, it should have been apparent by now.

      • “That is, increasing the CO2 content of the atmosphere does not primarily increase the greenhouse effect by creating more back radiation to the surface.”
        I don’t entirely agree with that. Just as more CO2 raises the effective height of TOA emission, it lowers the effective height of the back radiation source. These are dual, linked, effects.

        Surface temperature is determined as that needed to emit (or advect) the received flux, The received flux is basically insolation (not much dependent on GHG) and back radiation. Back IR has to increase as part of a surface warming mechanism.

        The TOA argument is valid too. They are connected.

      • Nick Stokes–

        I didn’t say increasing CO2 won’t increase the downward infrared radiation, just it’s an incomplete argument in all its forms.

        First off, if you keep the atmospheric temperature fixed, and put in some CO2 to an atmosphere where the low levels are radiating like an ideal blackbody, how are you going to increase the downward IR flux? The emission will be solely forced by the layers temperature in accordance with the Planck function. If the low levels are sufficiently opaque, the increase in CO2 higher up won’t increase the IR to the surface either, as it will be absorbed before it gets there. In this idealized case, the CO2 still warms the surface, but the only way to think about the problem is by bringing the TOA budget into play and allowing the whole troposphere to warm.

        Then, to actually close the surface energy budget, you also need to know how evaporation and sensible heating fluxes respond, not just the radiative ones. If you moisten the Sahara desert and give it an evaporation loss term, the surface will cool even if you bump up the CO2 a bit.

      • Chris and Nick,
        I think the missing link between your arguments and in every argument about the greenhouse effect is the different quality of the two budgets with respect to thermodynamics and the 1st law.
        A flux across TOA is able to change the energy content of the whole earth system. So, TOA rules the game. A flux across the earth surface is not able to change the energy content of the earth system. Therefore, the greenhouse effect and the effect of CO2 can be explained without backradiation using the concept of emission height.
        The greenhouse effect and the effect of CO2 cannot be explained without the TOA budget and the temperature gradient in the troposphere.
        So, I think Chris is right, explaining the greenhouse effect only with backradiation leaves the door open and is incomplete.
        Best regards

      • Leonard Weinstein

        Back radiation does not heat anything (on the average), much less the ocean. Solar radiation heats the ocean (short wave), and heats the Earth and to some extent the atmosphere. The long wave absorbing greenhouse gases and water droplets move some of the outgoing radiation upward through a series of absorptions and emissions, until they reach a height where they can emit to space from high in the atmosphere (plus some emits directly through transmission “windows”). The effective emission height sets the atmospheric temperature at that height, which then sets the temperature by way of the adiabatic lapse rate. The adiabatic lapse rate results in the ground temperature, including the ocean, to be warmer than without the greenhouse gas. The increased temperature at ground level results in more outgoing radiation than without, and the warmer atmosphere along with greenhouse gases causes back radiation, but the net difference in forward radiation and back radiation is small, with forward radiation always dominating (on the average). The outgoing convective heat transfer, including evaporation are the main source of energy transfer, and the sum of the small net outgoing radiation energy transfer and the convective heat transfer balances the solar input (unless the system is not in balance, and some storage action going on).

        In other words, there is back radiation, but it is not a source of average heating, but rather a result of the combination of greenhouse gas and lapse rate as I stated. There is back radiation, but NO back heat transfer.

      • Thanks Jim. I see that neither Andy Lacis nor Judith Curry actually answered my points and questions. I maintain the view that back radiation is incapable of heating the ocean to any significant degree compared to solar insolation and I’ll be happy to debate it with anyone if they are interested.

        The nutshell is:

        1) Back radiation from the atmosphere doesn’t penetrate the surface beyond its own wavelength. A lot of its energy goes into evaporating water molecules on the surface, a cause of cooling of the surface. That energy which is mixed into the surface layer soon leaves the ocean again as freely convecting water molecules head upwards due to their lower density. The wave action which mixes solar energy down to lower levels takes place substantially further down than the level back radiation will be mixed in by light breezes.

        “Under low wind speed, the sea surface can be cooler or warmer than the subsurface due to overlying thermal conditions, and the skin layer can be neutral to the transient process between them.”
        Low wind speed is the predominant mode of the marine atmosphere.

        3) The ocean is on average 3C warmer than the atmosphere. The second law of thermodynamics is on my side rather than the side of those who most often invoke it to dismiss the arguments of others. Heat predominantly flows from the ocean to the atmosphere, not the other way. Atmospheric temperatures lag sea surface temperatures by 3-6 months, clearly demonstrating the direction of causality.

        4) The warming of the tropical ocean by the suns energy which penetrates it to 10o metres and more during the lowered cloud albedo period 1980-1998 sufficiently explains global warming without the need to torture the data and ask the atmospheric tail to wag the oceanic dog. The rise of ocean heat content in the 1993-2003 decade calculated from the steric component of sea level rise indicates the surface forcing was ~4W/m^2 and this is much more than co2 could ever hope to achieve.

        5) The net longwave radiation results in ~70W/m^2 cooling due largely to the latent heat of evaporation and the free convection of water vapour.

      • I, too, can’t see how the air can change the temp of the oceans much. It is easy to understand how the oceans can change air temps though. Chris pointed out that clouds/not clouds depends on factors other than water vapor in the air. That seems to leave the door open for GCRs.

      • My understanding is that the oceans, at depths lower than about 100mtrs is substantially cooler than the atmosphere and quite uniform from the tropics to the poles.
        If the atmosphere was capable of heating the oceans, by whatever mechanism, it would have done so long ago since it’s been about 10,000 yrs since the last ice age.

        Am I wrong in the above assumption?

      • My response was some text from my thermodynamics book. I’ll try to look for it, it is a very long post, do you remember what thread this was discussed on? I’m losing track

      • This is where the local search engine helps. Google provides one free but I don’t know what format the content has to be in. I am surprised that WordPress does not provide something.

      • Judith, Jim linked it above and I’ve answered there.


      • Hi Judith: Contradicting models and theory, the NODC Ocean Heat Content data presented by Levitus et al (2009) (and as recently updated) does not indicate that there is a noticeable anthropogenic component in the warming of the oceans since 1955. To view this, simply divide the oceans into tropical and extratropical subsets.

        The tropical Pacific OHC rises in “steps” in response to the increased DSR associated with the 1973/74/75/76, the 1995/96, and the 1998/99/00/01 La Niña events:

        Curiously, tropical Indian Ocean OHC mimics part but not all of the variability of ENSO until the mid-1990s. It then shifts and becomes much more variable:

        But if we combine the tropical Pacific and Indian Ocean OHC, the long-term relationship with ENSO returns:

        The major increases in Tropical Atlantic OHC appear as lagged responses to the 1973/74/75/76 and 1998/99/00/01 La Niña events. The trend is the influence of the North Atlantic which represents more than 30% of the rise in OHC since 1975:

        The trend of the extratropical South Pacific OHC is reasonably flat from the early 1970s until the 1997/98 El Niño, when it rises in a step. Note the earlier spikes during La Niña events:

        The vast majority of the rise in extratropical North Pacific OHC occurs the late 1980s to early 1990s:

        This coincides with a shift in SLP.

        The South Indian Ocean rises in response to the 1973 thru 76 La Niña, dips and rebounds in the early 1990s (curiously the only noticeable reaction to the eruption of Mount Pinatubo), and then shifts drastically higher during the 1995/96 La Niña and 1997/98 El Niño:

        The North Atlantic OHC is governed by SLP, AMO, and ENSO:

        I discussed and illustrated the preceding in three posts:

      • Thanks Bob, I am sending this one to Peter Webster also

    • The issue is whether they warm by absorbing incident solar radiation, or by absorbing secondary LW radiation from the surface. I say the former, and there is no bouncing and amplifying of heat between the surface and the atmosphere.

    • OT – tallbloke, if you find Judith’s response, could you please leave a map?

    • I can’t help being bothered by the idea that anything is beyond skepticism. For evidence, I invoke the memory of Heisenberg. As nasty as the AGW situation has gotten, I don’t know that it approaches what went on with Heisenberg and Schroedinger. Imagine the necessity of convening all the great physicists of today in one place to have a intellectual slugfest over one issue, neuron to neuron, synapse and dendrite to synapse and dendrite. And then having the resolution be …… you’re both right! Hmmmmm, could be lol.

  10. Try this for the GT paper:

    G&T Paper

  11. “We need to raise the level of our game in terms of explaining the planetary warming by infrared absorption of CO2 etc. “

    Do you consider people as ever being responsible for the opinions they hold? I presume you’re aware there are a large number of people who hold anti-scientific viewpoints on things such as vaccinations and evolution.

    At what point (if any) does the responsibility shift from the scientist to communicate and for the skeptic to listen? Even going one-to-one do you think you could actually change the minds of people posting on this and the other thread who say the greenhouse effect is impossible?

    • Sharpen up and read the Miskolczi paper. This is peer reviewed science.

      • That’s not relevant to the question I posed, see the part about listening.

      • Sharper’s is the fundamental question laid out by this post: is the problem with these “erroneous” papers a something that could be remedied by a lesson accessible to by “an audience that has taken 1 year each of undergraduate physics and chemistry, plus calculus,” or is it ideological in nature?

      • The premise of this is as flawed as the notion of “Global”.

        The climate system is Not static nor will it ever be homogeneous due to all the non-linear aspects and dynamics.

        The idea, to understand the issues requires Indoctrination is a fallacy.

        Climate Science has failed to communicate because it attempts to define “Global” without proper diligence to the regional aspects of the system and without clearly stated deference to its current limitations.

    • I presume you’re aware there are a large number of people who hold anti-scientific viewpoints on things such as vaccinations and evolution.

      Obviously I’m not Dr Curry, but I think it’s relevant to note that this large number of people with an anti-scientific POV is a minority of the US population; I think the far right anti-evolution “evangelicals” comprise under 20% of the GOP (don’t have the cite for this handy, sorry, but this came from Time or Newsweek some time back), which puts them at some 10% of the US population at best. And even then, many are not anti-science since they object to only that which challenges beliefs, which then reduces this number where it concerns the subject at hand. The GHE doesn’t directly challenge biblical belief the way evolution does.

  12. The Miskolczi 2010 abstract:

    by Ferenc Miskolczi
    VOLUME 21 No. 4 2010, AUGUST
    By the line-by-line method, a computer program is used to analyze Earth atmospheric radiosonde data from hundreds of weather balloon observations. In terms of a quasi-all-sky protocol, fundamental infrared atmospheric radiative flux components are calculated: at the top boundary, the outgoing long wave radiation, the surface transmitted radiation, and the upward atmospheric emittance; at the bottom boundary, the downward atmospheric emittance. The partition of the outgoing long wave radiation into upward atmospheric emittance and surface transmitted radiation components is based on the accurate computation of the true greenhouse-gas optical thickness for the radiosonde data. New relationships among the flux components have been found and are used to construct a quasi-all-sky model of the earth’s atmospheric energy transfer process. In the 1948-2008 time period the global average annual mean true greenhouse-gas optical thickness is found to be time-stationary. Simulated radiative no-feedback effects of measured actual CO2 change over the 61 years were calculated and found to be of magnitude easily detectable by the empirical data and analytical methods used. The data negate increase in CO2 in the atmosphere as a hypothetical cause for the apparently observed global warming. A hypothesis of significant positive feedback by water vapor effect on atmospheric infrared absorption is also negated by the observed measurements. Apparently major revision of the physics underlying the greenhouse effect is needed.

    • Is that the same Energy and Environment journal that is thought by some to be rather biased in its approach to publishing? I believe the editor said “I’m following my political agenda — a bit, anyway. But isn’t that the right of the editor?” and “Roger Pielke Jr said in a post answering a question on Nature’s blog in 2007 about peer-reviewed references and why he published in E&E: “…had we known then how that outlet would evolve beyond 1999 we certainly wouldn’t have published there. The journal is not carried in the ISI and thus its papers rarely cited. (Then we thought it soon would be.)”


      There are peer reviewed papers and then there are peer reviewed papers.

      • Judith commented on E&E previously here
        “Re E&E. The reputation of a journal rests in the number of citations it receives, the scientific stature of its editorial board, ts reputation for fair and critical reviews of papers, and its selectivity. E&E scores very low on all of these (with possible exception of the review process, i simply don’t know). Further, based upon their public statements, many of the editors of E&E have an overt agenda with respect to energy and environmental politics. Hence choosing to publish a scientific paper in this journal taints the paper and the author as being driven by a political rather than a scientific agenda.”

      • Are you sure you want to start a debate on the bias of journal editors?

        Nature didn’t publish a solar paper for 5 years.
        Steigs paper on the melting antarctic?? Lol.

      • Science Journal bias is a great target of opportunity for skeptics. Those pesky e-mails and the documented behavior of many AGW promotion journals is well documented.

      • But does any of that matter? Does it really matter who funded the research or who published it or for what reason it was published? Let’s look at the content. Is it solid or not and why?

      • On two fronts, it does matter who funded the research and where it is published to an extent. On the other, the science itself is pretty unbelievable on first glance.

        The basic premise to Miskolzci’s theory is that the atmosphere ‘knows’ how and when to condense water in order to preserve a particular value of its IR optical depth. There is no physical process that he posits or observes in reality to tell him this other than what he presumes is solid humidity data from several decades ago. All of that despite several leading scientists have told him this data is flawed.

        More than that, from a molecular point of view, his theory goes against the laws of thermodynamics. Because water is at a much higher concentration in the atmosphere than other IR absorbing gases that can condense, say acetone, it will take a tremendous amount of water condensation to make a small change in the IR transmission through the atmosphere. Since processes that use the least amount of energy are favored throughout physics, it would seem that the atmosphere ‘would want’ to condense greenhouse gases that are at a much lower concentration than water and absorb IR light inside the water ‘window’. That would maintain the IR optical depth at a specific level while spending less energy overcoming inter-molecular forces in the condensation process.

        But none of this is explored nor explained by Miskolczi.

        Moreover, because his theory is so vague that it doesn’t define which regions of the IR spectrum are important in these processes, there isn’t a real way to falsify it in the lab, where nothing like what he describes has been observed. Maybe you could convince the guys at JPL molecular spectroscopy lab to take up this cause, but I’d suspect you’d have a hard time getting anywhere with that.

        Based on all this I’d say the fact that his work is published in Energy and Environment means it’s garbage.

      • Miskolczi could not even recognize the difference between emission (a flux of energy) and emissivity (a ratio), at least in one of his original papers a couple of years ago. There’s just so much wrong with his idea, but observations themselves show it to be wrong (just like with Claes Johnson’s paper which states that back radiation is implausible…yet it’s observed everyday). Just stupidity.

      • The guy was in an $80K job with Nasa as a physicist. His Boss logged into his computer with his credentials and pulled his paper from JGR. That’s why he resigned and published in E&E

        I see all the usual prejudice being trotted out here.

      • “what he presumes is solid humidity data from several decades ago. All of that despite several leading scientists have told him this data is flawed.”

        If the data is so bad why does this neat correlation appear between specific humidity at the tropopause and solar activity levels?

        It seems that all the data which mitigates against the AGW hypothesis (radiosonde, ISCCP) is flawed, while treemometers and dodgy stats algorithms like Manns reworking of PCA rule the roost.

        Zero credibility.

      • ‘If the data is so bad why does this neat correlation appear between specific humidity at the tropopause and solar activity levels?’

        Most likely coincidence. Without your proposing a physical model that would explain such a correlation naturally, why should we believe it to be anything but a coincidence?

        I think it’s also unfortunate that you are conflating a serious critique of the physical model that Miskolczi presents with an acceptance of any and all other techniques and models. No one mentioned anything about paleoclimate or thermometers or any other measuring devices before the satellite era. You’re simply inferring that because people are not wholesale accepting what you claim.

        Make a better argument then maybe you can start making remarks about others’ ‘credibility’.

      • “No one mentioned anything about paleoclimate or thermometers or any other measuring devices before the satellite era. ”

        You cast doubt on the radiosonde data from 1948. No satellites around then.

        “Without your proposing a physical model that would explain such a correlation naturally, why should we believe it to be anything but a coincidence?”

        I could say the same about co2 and temperature. But then, they don’t correlate as well as solar activity and specific humidity at the tropopause do they?

        And I do have a physical model for that correlation, but this is not the thread for it.

      • bloke,

        ‘You cast doubt on the radiosonde data from 1948.’

        Yes, Miskolczi’s theory necessitates a discussion of the utility of radiosonde data. What does that have to do with other measurement techniques that deal with your claims about others’ ability or inability to accept what you have to say?

        ‘I could say the same about co2 and temperature.’

        Huh? The atmospheric greenhouse effect is a testable and proven model for a causal relationship between temperature and CO2 concentrations in the atmosphere. It’s based on quantum mechanics, arguably the most well tested scientific theory we have come up with yet. So I don’t understand where you get the impression that there isn’t a meaningful physical model relating CO2 concentrations and temperature. It’s called the Beer-Lambert law.

        I do agree that an increased greenhouse effect doesn’t explain ALL the variation we’ve seen in surface temps, but that’s hardly a reason to doubt whether the atmospheric greenhouse effect is a physical model.

      • My doubt concerns the application of theoretical physics to the real atmosphere, the quantities derived, and the assumption that a significant amount of extra IR energy makes it’s way into the deep ocean from the atmosphere.

        Ever tried warming up a cold coffee with a hairdryer?

      • David L. Hagen

        “Miskolczi’s theory necessitates a discussion of the utility of radiosonde data.”
        Any references on how to quantitatively correct the radiosonde data?
        Otherwise Miskolczi used it as the only 60 year data for humidity available.

      • Are you sure you want to start a debate on the bias of journal editors? :)

        And Wikipedias less than even handed approach to climate change??

  13. Two things that bother me about the way the greenhouse effect is explained have to do with description.

    Many people, including NASA and NOAA, claim that greenhouse gases ‘trap’ heat in the lower parts of the atmosphere. That is an oversimplification, however. Because CO2 or CH4 can absorb earthlight and re-emit it back toward the surface, the amount of time that energy stays in the lower portions of the atmosphere increases. But it doesn’t increase to infinity, as is implied by the word ‘trapped’. As a molecular physicist, I think it’s imperative to make sure that the dynamics of each molecule come through in these mechanistic explanations. The lifetime of a vibrationally excited state of a gas phase molecule is on the order of a millionth of a second, if not shorter. That’s hardly trapping.

    Also, I agree with a couple other physicists here that there is a bit too much oversimplification in the picturing of these processes. The idea that the climate started at ‘equilibrium’ may be useful as a pedagogical tool, but it does little to help explain the observed data we have. That is, there is a ~1% difference between the incoming and outgoing radiation at the top of the atmosphere as measured by satellites. Most of that 1% is purportedly due to the increased greenhouse effect due to human influence. However, such a conclusion is based on the assumption that before such satellite based measurements were made this difference in incoming and outgoing radiation was exactly 0%! I’m not sure how good on an assumption that is.

    From my perspective, the best analogy I’ve read so far came from Roy Spencer who compared the atmospheric greenhouse effect to a blanket. Putting a blanket over your body impedes heat from being emitted out to the surrounding air. That ‘excess’ heat causes your clothes and the blanket to heat up until the rate of the energy in (from your body) equals the energy out (from the blanket to the air). By adding CO2, we make the ‘blanket’ more insulating such that the temperature has to increase to increase the rate of energy loss to ‘the air’ (space).

    Is that the only thing at play in climate? Of course not, but it’s hardly disputable physics.

    As far as Venus is concerned, the greenhouse effect is very large there, but more importantly, the Coriolis force on Venus is very small. There are only two days per year on Venus. Since the Coriolis force is responsible for a great deal of the atmospheric dynamics on Earth, the large difference in day length makes comparisons between the two planets difficult at best.

    Thanks for the technical thread.

    • If there is a ~1% difference between the incoming and outgoing radiation at the top of the atmosphere as measured by satellites wouldn’t we be able to measure the differences over the last 25 years and correlate the difference to the changes in atmospheric GHG’s?

      • Rob,

        Sure. People have done that. Trenberth was the first (?) in the late 1990’s, but the noise in the measurements is pretty big. Almost the size of the actual measurement itself. So it’s hard to say definitely whether it’s +1%, +0.5% or some slightly smaller or larger number.

        But even if we had a real concrete measurement, how do we know what the difference between the incoming and outgoing radiation ‘should’ be? From my perspective, this is where the assumption of climate ‘equilibrium’ comes in. It seems implicitly assumed that ‘equilibrium’ means that this difference must be zero. But surely there are fluctuations in this difference due to internal variability (ENSO, Arctic oscillation, etc.) or other factors (people). So what time scale ‘should’ we have to integrate over to get zero explicitly in an ‘equilibrium’ situation?

        I don’t know the answer to that question and I would posit that no one does for sure right now.

        But I think as time goes on, we’ll have more and more data to say how that number (radiation difference) is changing and get a better idea of how a human forced increase to the greenhouse effect is changing the energy landscape of the planet.

      • I would not think the equilibrium issue is really key, since it is a notional concept at best. I would have thought that being able to determine if the difference between incoming and outgoing radiation changed at a rate consistent with the changes in GHG’s, it would potentially confirm or refute positions such as those made by Huffman in an earlier post.

      • The data isn’t good enough yet to know at what rate the net radiation flux of the top of the atmosphere is changing. As I said before, even if we did have good enough data, it would still be a hard sell to claim that the data confirmed anything at this point because the physical models of the climate do not make specific enough predictions to make meaningful use of such observations. Such models would have specifically predict what ‘equilibrium’ meant in an observational sense.

        Also, how many different ways can our current physical climate models give a +1% difference between the incoming and outgoing radiation fluxes? I’d bet it’s more that one.

      • I wonder if the climate researchers have considered that the difference between the incoming & outgoing radiation has always been there, & is due to the Earth’s internal heat being transferred to the atmosphere? Or did they just jump to the conclusion that our CO2 emissions were the cause?

      • “So what time scale ‘should’ we have to integrate over to get zero explicitly in an ‘equilibrium’ situation?”

        About as long as it takes the solar system to traverse one of the galaxy’s spiral arms and the space between that and the next one.

        Click to access shaviv-n-the-milky-waygalaxys-spiral-arms-and-ice-age-epochs-and-the-cosmic-rayconnection.pdf

      • bloke,

        ‘About as long as it takes the solar system to traverse one of the galaxy’s spiral arms and the space between that and the next one.’

        That’s ridiculous. You’re saying that we have to integrate over millions of years to get a meaningful understanding of the internal and naturally variation in climate on earth to the first few orders?

        The very nature of the problem is set in time scales that are less than a ten of a percent of that time period. How can variations with time scales on the order of millions to ten of millions of years substantially affect processes that occur on decade to century time scales? That’s basically the same as saying that the behavior of an excited molecule will depend on which side of the lab I do a measurement.

        It doesn’t.

        If the period of interest were that long, integrating over 1000 years would provide a constant background to the galactic variations, meaning that it would still be useful to us on a day to day basis. More than that though, what you’re proposing is impossible given the fact that the chances civilization lasts that long are close to zero.

        So I would conclude that looking at variations that occur on time scales less that hundreds of millennia does not necessitate accounting for galactic variations. Our galactic background is not fluctuating fast enough to matter on ‘interesting’ time scales.

      • OK, I was taking the longest climatic variation we can identify to make a point. However, there is evidence, and some glimmering s of mechanism to elucidate climate cycles which occur on timescales which do affect our interpretation of climate trends happening now.

        For example, there is an apparent warming and cooling taking place on a millenial timescale, evidenced by archaeological finds on Alpine passes in Europe, where Roman and medieval artifacts have been discovered recently at locations impassible between these periods and between medieval and now. From medieval times there was a descent in temperature to the little ice age, and then rising temperatures to today. This rising trend coincides with an increasingly active Sun from 1670 to 2003. We now seem to be entering a period of solar grand minimum potentially as deep as the Maunder minimum in the C17th. Other solar minima intersperse the deeper ones, with a periodicity in the ensemble of around 180 years.

        The work of Charvatova links these periods with the motion of the Sun wrt the centre of mass of the solar system. The mechanism may involve the disruption of the solar dynamo by these irregular motions.

    • A way to explain the basic effects to a lay audience might be to do a thought experiment starting with an atmosphere with no GHG and then to add just 1 test molecule of a GHG and give a detailed explanation of each additional process which that one molecule can participate in, showing how each process is consistent with Quantum Mechanics, Newton’s Laws and the 1st and 2nd Laws of Thermodynamics, dealing with the various popular objections as you go, while increasing the concentration of the GHG from 1 to 2 to many.

      • JT,

        that’s a great idea. Unfortunately, once you begin mentioning vibrational degrees of freedom of GHG molecules, eyes begin to glaze over. That might even be true in the climate science crowd as much as it is considering a lay audience.

        I think the larger point is that those who posit ‘alternative theories’ to the atmospheric greenhouse effect will do whatever they can to convince themselves (and possibly others) that something makes the greenhouse effect wrong. I don’t think that Dr. Curry’s point about poor communication is valid in this case. People just don’t want to believe that it could be true.

    • Maxwell,
      As a molecular physicist perhaps you could help me to understand how the energy from molecular H2O condensation (latent heat) is transferred away from the H2O cluster?

      There are only 2 mechanisms – radiation and conduction. Radiation I can understand, although I would like to know the most probable wavelength this would occur at (I could see it being a range of frequencies, depending on the nature of the H bond formed and the cluster size).

      What I am stumped by is how conduction could be a direct transfer process. My thinking is that the energy must be released immediately in order for the H bond to form. If it does need to be released immediately, then I cannot see how conduction would work. Is it possible that the latent heat energy is stored in the molecular cluster – perhaps distributed throughout momentarily until such time as a collision with another molecule occurs, for example N2 or O2, to transfer it?

      The mechanics of evaporation makes sense to me, but condensation does not.

      • Kan,

        I’d think it would have to be collisions with other gas phase molecules while the water molecules themselves were near other water molecules (condensation is not a single molecule process) or collisions with clusters of ‘stuff’.

        My guess is that the second process plays a larger role in condensation just based on contrails from airplanes. For example, the ‘large’ material particles in the jet exhaust immediately cause the highly concentrated water vapor to condense. This process is known as nucleation and is at the heart of many processes like condensation, i.e. micro/nanofabrication growth, crystal formation, etc.

        I’d also guess that since these ‘large’ material particles have many, many degrees of freedom, it’s not hard for them to absorb the energy necessary for the condensation of some water molecules. I don’t know how many ‘some’ is, however. But then again, we’re now talking about cloud formation, which may be the most uncertain physical process in climate science.

        Hope that helps.

  14. Judith,

    I come to this very much as an outsider, so this is really two questions.

    Surely it makes no sense to claim the greenhouse effect exists, but place no limits on its magnitude – the size of the effect could be infinitely small, or is the greenhouse certain to have a value of at least (say) 2 deg per doubling in the absence of feedback effects?

    I mentioned in an earlier comment the analogy with Venus’ atmosphere, and I wonder if the temperature at 1atm pressure in that atmosphere (far more moderate that the surface temperature at ~92 atm) could be used to estimate the actual size of the greenhouse effect.

    • David, you appear to be talking about the so-called enhanced greenhouse effect due to increasing greenhouse gases such as doubling CO2. This thread is actually about the primary greenhouse effect, which supposedly makes earth habitable in the first place.

      • Well surely from a scientific point of view, the basic greenhouse effect and any enhancement due to extra CO2 are the same thing – we are just altering one parameter.

      • No, these would be the same thing is you assume that the effect dependence on the parameter is monotonic with altering its “concentration”. It is quite possible that the dependence reaches a plateau or has other kind of inflection point. That’s why conflating the primary effect with “sensitivity” is wrong and confusing to many.

  15. Good explanation and discussion here too

  16. There are three fundamental problems with the greenhouse model as an adequate model.
    1) The pure physics only gives a coherent answer in a pure gas without clouds. On top of this, how the surface albedo varies with space and time is maddenly complex. Thus when IPCC supporters say “it is radiation physics you dolt” they are either ignorant or disingenuous. Pure physics does not allow calculations in such a spatially complex system. As I have mentioned before, strain and fracture in pure substances is well-characterized and used by engineers all the time–so why can’t we predict earthquakes? It is the heterogeneity and scale that prevent us and this applies to the atmosphere also.
    2) The “pure physics” even if adequate only give 1 deg C or so of warming for a doubling of CO2 from historic levels. The rest is the famous unproven (and assumed) water vapor feedback. If your entire argument of doom is based on a hand-waving assumption, don’t expect anyone to join you in your hysteria.
    3) The problem of figuring out heat transfer in a complex multifluid is quite intractable. We know the GCMs don’t handle thunderstorms well as heat pumps. In addition, let us say that part of the process of ENSO type systems is concentrating or dispersing surface warm/cold waters. If warm waters concentrate at the surface, more heat might radiate to space, and conversely when heat mixes rather than concentrating at the surface. How do we handle/compute this? Not “pure physics” at all.
    Thus the problem is not just physics but that the heterogeneity and turbulence prevent nice analytic solutions and may create unexpected outcomes.

    • strain and fracture in pure substances is well-characterized and used by engineers all the time–so why can’t we predict earthquakes?


      • Very funny–that’s a prediction? How about when and where closer than “california within 30 yrs”. Thanks for making my point.

      • That’s your point? Climate predictions are vague in the same way earthquake predictions are vague, for the same reason: the complexity, which you correctly pointed to in your comment.

        The complexity doesn’t stop scientists from saying “The likelihood of a major quake of magnitude 7.5 or greater in the next 30 years is 46%” and shouldn’t stop scientists from saying “that climate sensitivity is likely to be in the range of 2 to 4.5°C with a best estimate of about 3°C.”

      • PDA,

        Of course both the climate predictions and the earthquake predictions are right; they both are based upon models. Everyone knows that computer models can’t be wrong!!

        Good own goal there, PDA!

      • Haw haw haw AllenC, ya got me. Yes, computer models can be wrong, as can everything else.

        I agree that the way forward is to leave aside any idea how to quantify the correctness of simulations, and how to incorporate additional evidence from direct observation. Rather, let’s just continue to guffaw “COMPUITR MODOLZ R TEH ST00PID!!1!”

        This is citizen science at its finest.

      • If you think the unverified predictions of earthquakes (47% blah blah) are good enough and such vague predictions are good enough for climate change, this is NOT what the alarmists are claiming, they are claiming certainty of disaster. It is possible to MAKE earthquake predictions, but their track record of being right is vanishingly small.

      • I don’t live in California, but don’t they have more restrictive building codes there because earthquakes are more frequent, and shouldn’t we apply the same thinking to the damage that could be caused by warming?

      • Damage by earthquakes has been observed for centuries. Damage by CO2 induced heating is still only a hypothesis.

    • Craig, none of these issues are ‘problems with the greenhouse model’

      No one argues that cloud aren’t complex, but they are still physics problems. There are many issues at the interface of dynamics (e.g., what is forcing the air to rise or sink), thermodynamics (e.g., the availability of water vapor), and microphysics, and a fundamental issue is that it’s computationally expensive to model everything relevant at the scales necessary to resolve the finer details. But the fundamental ways the greenhouse effect works is still at play with clouds. They interact with radiation in understood ways, and just because they have an albedo component doesn’t mean the ‘greenhouse’ component is invalid.

      Your statement about water vapor feedback just reflects you are ignorant of observations, theory, or how we have come to learn about the water vapor feedback (hint, it’s not “assumed”). Similarly, water vapor distribution is also a ‘physics problem’ (what else would it be, astrology?). Just because something is a feedback, or not part of the “CO2-only” response, doesn’t make it any less valid.

      You’re just making things up.

      • Craig, you write ” The “pure physics” even if adequate only give 1 deg C or so of warming for a doubling of CO2 from historic levels.”

        I have no quarrel with what you have written, except for this sentence. I object to the alleged 1C rise for doubling CO2. This number has never been measured. This number can never be measured, since any attempt to do so would be coufounded by any feedback effect. I queried the physics used by the IPCC to estimate this number, and I believe Judith agreed with me that this physics leaves a lot to be desired. There seem to be a number of oversimplistic assumtions made in order to arrive at the number.

        But, if the number can never be measured, it must always be subject to the Kelvin Fallacy, and we simply can never know for sure what the number is. And this will never change.

        The number of 1C for a doubling of CO2 is much more science fiction than it is physics.

      • I agree that the no-feedback sensitivity can never be observed (unless one day we make a laboratory Earth) but it can be worked out on the back of an envelope and there are really no big assumptions that go into it. It is, for instance, not at all dependent on the land surface physics and is almost entirely constrained by the radiative forcing for CO2 (which can be measured). It’s also a value of virtually no spread amongst climate models which do more than back of the envelope. It’s one of the easiest pieces of physics in this discussion and can amount to a simple derivative of the Planck function, so I don’t see a reason not to believe it.

      • Chris Colose writes “I agree that the no-feedback sensitivity can never be observed (unless one day we make a laboratory Earth) but it can be worked out on the back of an envelope and there are really no big assumptions that go into it”

        Lord Kelvin worked out that the age of the earth was 10,000,000 years. He used a similar sort of method as you described for estimating global temperatures. He claimed he was right because he had taken into account “all known factors”. In fact the earth is 4.2 billion years old. That is why we call it the Kelvin Fallacy.

        I am sorry. If you cannot measure a number, then you have no idea what that number is. Period. That was the physics that I had hammered into my head 65 years ago when I studied Physics 101. It was true then, and it is still true today.

      • Since when has climate science ever bothered to worry about observations?

        Any data that doesn’t agree with the latest theory is ‘adjusted’ until it does. I believe that guys in New Zealand are especially good at this, followed closely by UEA/CRU.

      • They must be brassed about coming second, since they trained Jim Salinger in the first place.

      • In fact the earth is 4.2 billion years old.

        But I thought “If you cannot measure a number, then you have no idea what that number is.” No idea. “Period.”

        It would probably have been better if your Physics professors had not used a hammer.

      • Sorry, PDA, I have no idea what you are talking about. Ernest Rutherford, as a research student from New Zealand, provided the measured data that proved that the earth was 4.2 billion years old.

      • Ernest Rutherford, as a research student from New Zealand, provided the measured data that proved that the earth was 4.2 billion years old.

        I have no idea what you are talking about. Rutherford was at McGill (which – at least on the Earth I am writing from – is in Canada) when he made the discoveries about radioactive decay that led to the practice of radiometric dating.

        In 1929, Rutherford estimated the Earth’s age, in “Origin of Actinium and Age of the Earth,” as 3.4 billion years. It wasn’t until 1953 – 16 years after the death of Rutherford – that scientists began to approach within a billion years of today’s best estimate of 4.55 ± 0.02 billion years.

        These are estimates of the Earth’s age, based on analyses of the lead isotope ratio in very old rocks like meteorites. They are not “measurements.”

      • Rutherford was born in New Zealand. The original sentence was ambiguous. You are both right.

      • Sure, if you want to interpret the sentence as “Ernest Rutherford, as (a full professor in Canada who was once) a research student from New Zealand, provided (an essential discovery about radioactive decay that when combined, decades after his death, with) the measured data that (was collected by another researcher) proved that the earth was 4.2 billion years old” then yeah. We’re both right.

      • Latimer Alder

        Is it coffee time or cocktail time in your time zone? Suggest that you have a stiff one anyway, and chill a bit.

        Rutherford was a native of New Zealand who was working in Canada at the time he performed the work described. It is not wrong to say ‘a researcher from New Zealand’. Nor is it wrong to add ‘working in Canada’

        You are both right. Cool it (if that is still possible in these globally warmed days).

        In snowy Surrey UK winter has come at its earliest for about twenty years. And we are all laughing our butts off watching the Loonies in Cancunies screaming at us that we’re forever doomed to perpetually warmer winters. Bring them on!!

      • “Student” was wrong, as was the attribution of the estimate of Earth’s age to Rutherford, as was the estimate itself.

        I’m perfectly cool. It doesn’t upset me one bit to have these sorts of disagreements with strangers, but if it upsets you, I certainly think your Rx is a good one to self-administer.

      • If this is meant to imply that the geological age of the earth cannot be measured (with error bars) then you are plain wrong – try half-life timespans from various elements found in meteorites

        And actually, Professors of Geology use geo hammers :)

      • PDA Fair enough. However, it does not take away from the Kelvin Fallacy. Kelvin did the calculations, and he was wrong. That is the key point. That is why the estimations for a change in global temperatures may be subject to the Kelvin Fallacy, and are, therefore, totally unreliable.

      • Rutherford did the calculations, and he was wrong. Does that mean radiometric dating was “totally unreliable?” Of course it doesn’t.

        The “point” that scientists have been wildly wrong in the past and so could be wildly wrong about global warming now is both trivial and irrelevant. Of course they could be wrong. Everything we think we know is subject to revision.

        Unless you point out the specific errors, though, just repeating “well, they could be wrong” ad nauseam is mere handwaving.

      • PDA, you are missing the point. If you have replicated measured data, there is no question of who is right or wrong. The measured, replicated data is right. That is what the “scientific method” is based on.

        What I am pointing out is that the proponents of CAGW have little replicated measured data to support the hypothesis of CAGW. Craig put up this myth that doubling CO2 causes global temperatures to rise 1 C with no feedbacks. This number can never be measured, and so is completely measingless. As are any other numbers which cannot be measured.

      • I am not “missing” the point. I am disagreeing with it.

        The age of the Earth is a physical property that can not be “measured” in any realistic way. We have no ‘timers’ dating from the moment of the cooling of the molten planetary mass. Its age can, however, be estimated with a fair degree of accuracy thanks to our understanding of radioactive decay.

        Likewise, the effect on mean global temperatures of a doubling in carbon dioxide concentrations in the atmosphere is a physical property that cannot be measured. It can, however, be estimated. I’d say that the accuracy is roughly commensurate with that of geology at the same stage of development. Most of his contemporaries thought Kelvin was way underestimating the age of the Earth, by the way.

        Reasoning by analogy breaks down at a certain point. If you have criticisms of how climate sensitivity is calculated, be specific about them. But leave poor William Thomson’s ghost out of it, please.

      • Back of the envelope climate sensitivity.
        Change in no feedback temp of CO2 = 5.35xln(540/270) = 1.2C

        Change in temperature = change in no feedback temp of co2/(1 – feedbacks)

        Feedbacks = (water vapor (+/-) albedo (+/-) lapse rate (+/-) clouds)

        Best empirical observations = WV = +0.6, albedo = +0.1, lapse rate = -0.3 clouds = +.15

        Change in temp = 1.2/ (1 – .55) = 2.7 C

        Climate sensitivity per doubling of CO2 is 2.7. When error bars are included for the feedbacks, the range is likely between 2C – 4.5C with 2.7 being the most likely.

      • Gryposaurus, where did you get the 5.35xLN{}=1.2C from?

      • 5.35 is the radiative forcing of CO2 based of thousands of lines of atmospheric satellite data calculated by Myhre 1998. LN is the natural logarithm. And 540/270 is the CO2 doubled and original CO2.

        540/270 = 2
        the LN of 2 = 0.69
        5.35 x .69 = 3.7 watts per square metered
        = 1.1C +/- .1 with error bars

      • Chris Colose, please do us skeptics a favor and derive for us the 3.7W/m2 forcing from 2xCO2 on the back of an envelope. I would be also tremendously interested how do you propose to measure it. Please.

      • This isn’t quite a back of the envelope calculation. You need something along the lines of line-by-line radiative transfer calculations that go through the absorber and temperature profiles and the spectral selectivity of the gases in question. I don’t do this! but Andy Lacis could probably go through this process with you, he did so in some degree in this comment on my blog …also see the Myhre et al 1998 results. Note that this is the forcing, not the temperature change.

      • “Modeling radiative transfer in the Earth’s atmosphere is a bit messy and complicated, if not outright murky and opaque.” – A. Lacis

        “… three vertical profiles (a tropical profile and northern and southern hemisphere extratropical profiles) can represent global calculations sufficiently. These profiles are used for all calculations in this study” – Myhre et al. GRL(1998)

        Thank you Chris Colose. Whole three near-tropical profiles, I understand. This must be definitely representative, especially for polar areas. Thanks.

      • Am I being super-dumb here or could you not conduct an experiment something along the lines of taking a large volume of air w/o any GHGs apart from CO2. Measure its properties. Double the proportion of CO2 and measure again. That will give you a base understanding of the warming just due to CO2. Then introduce various amounts of different other GHGs and you can measure the ‘feedback’ due to each…and in any combination you deem useful.

        Not an ideal experiment, I concede, but it probably doesn’t require anything nearly as expensive or difficult as the LHC…and it would give some raw numbers that seem to be so sadly lacking in the whole field of climatology.

        It would require some passing facility with experimental techniques rather than just sitting in a computer suite like Harry tearing his hair out, but maybe some passing chemist or experimental physicist could help out in their tea breaks. And they could borrow a statistician to help them interpret the results.

      • It would be very easy to demonstrate that CO2 emits at, for example, 15 microns in proportion to its amount and temperature. This is also easily seen with the IR spectrometer looking up at the sky. I don’t think the skeptics believe in experimental evidence like this, however, or perhaps they just don’t know about it.

      • Well I used to do quite a bit of IR spectrometry as part of chemistry. Its a useful way of helping to find out something about what that funny white organic powder you have just made. So I’m tolerably familiar with the principles, thanks for asking.

        If you tell me that CO2 emits in the IR at 15 microns, I’m more than happy to believe you. You could even show this on national TV to convince people by tuning an IR camera to that wavelength and wave it around a bit.

        But its a waaay long stretch from that to positive feedbacks of 4x or more and hence frying tonight. And very little measured experimental data around. That tells me that theory is in a very primitive state when stood against the norms of ‘proof’ needed in hard science.

        My suggestion is a way to focus in on the key issues and do some freaking measurements. This would avoid the Kelvin fallacy. And restore some scientific credibility to the tattered field of climatology.

      • Would it satisfy you to see clear sky emitting 250 W/m2? For feedback, it is hard to imagine any better experiment than warming some water by a degree and seeing if the vapor above it increases in an amount that maintains its relative humidity, which is saturated at the surface. This is what we are doing to the oceans.

      • Hi Jim D

        It would be a useful starting point. But by itself it proves only exactly what it measures…a point the learned Jim C was trying to make above. Once you feed that number into a theory (I guess the overall radiation balance), then you have to look at the theory as a whole..not just the individual bits and come up with ways to prove the overall theory.

        I had a lovely theory once upon a time when the world was young, when summers were hot and winters weren’t as frigging cold as this one (ie in the days before global warming) of a little piece of high atmosphere Chemistry. All the computer simulations gave great was super. But sadly, the experiments showed that it was fundamentally wrong. We canned the theory.

        You have to show every bit of the logic. And where the logic chain is weak, invent experiments to test it. Make them reproducible. Show the working. Publish the code. Test each bit individually, then a few bits collectively. (Simple integration engineering this). Record your findings in an auditable way – like in pharma or nuclear or safety-critical engineering.

        Then when you have both the theory and the test data to back it up, and have withstood the best objections that the best sceptics can devise, my scepticsm will largely go away. You will have done a good job of the science.

        So far…climatology scores very low on all those counts and more. The terrifying thing is that climatologists seem to be so insulated from the rest of science that they don’t realise just how poor a job they have done.

      • The science is a lot more solid than you think. It was mostly established prior to the AGW theory which just put those established pieces together with the newest CO2 and temperature data. No one complained about the fundamental science until the issue became politicized in the 90’s. It is clear to me that skeptic arguments are aimed at the public, not at the scientists, who they know they won’t sway.

      • “warming some water by a degree and seeing if the vapor above it increases in an amount that maintains its relative humidity, which is saturated at the surface. This is what we are doing to the oceans.”

        But Jim, the oceans have been cooling since 2003. How does extra co2 do that?

      • Is CO2 expected to explain every wiggle in sub-decade periods? I don’t think so. Wait a few more years before claiming this is climate, or compare this decade with previous ones.

      • Is CO2 expected to explain every wiggle in sub-decade periods? I don’t think so. Wait a few more years before claiming this is climate, or compare this decade with previous ones.

        Granted ENSO is a subdecadal wiggle, with an amplitude of at most 0.08 °C. However the AMO is a pretty major 60-year wiggle with an amplitude of 0.1 °C (i.e. a range of 0.2 °C). I see no sign of either kind of wiggle dying out in the next century, though AMO may well disappear for the duration of the 23rd century before returning.

        Feel free to quote me if it doesn’t. ;)

        Of course if our descendants are all fried by then it’s moot. My hope is that we’ll have avoided this by being mostly be on nuclear fusion power by 2060, after which the AMO warming should re-emerge from the cancelled CO2 warming which is currently almost completely masking it (but you can see it clearly if you know the trick for removing the mask).

      • Jim D

        A parabolic mirror pointed to the clear night sky has been known to freeze water at its focal point.
        When the ambient temperature arounf the dish was well above zero C

        What does that tell you?

      • I am not sure what you think this proves. The sky’s radiative temperature is below freezing? 250 W/m2 equivalent to a cold black body, so this would be expected in that case.

      • Well every IPCC proponent up to now has claimed that the radiative effects of the “greenhouse effect” heats up the near Earth atmosphere.
        The fact that the radiation from the sky can actually reduce the near Earth atmosphere at night has not featured in any of their explanations that I have come across so far.

      • The radiation from the sky does not freeze water. The water freezes, because the mirror prevents radiation from surrounding ground to warm the water. The parabolic form of the mirror allows for choosing the dominant direction of the radiation that reaches the foxal point to be one of minimal incoming intensity. The water at the focal point radiates at a higher intensity than the incoming radiation and cools.

        During a clear night the amount of infrared radiation coming from the atmosphere is reduced and sufaces open to the northern sky (on the northern hemisphere) cool well below the temperature of the surrounding air. With inceased CO2 this effect is a little weaker than before, but it is still there.

      • Good answer Pekka the dish transmits as well as receives, but perhaps there’s more heat transfer than radiative.
        The dish is still in thermal contact from the ground so local air contact and ground conduction supports the waters temperature.
        However this is not sufficient to overcome the radiative interactive losses to the night sky.
        In the post below as well as giving strong support to Woods experiment they report on a strange almost counter-intuitive effect (which they say has also featured in other reports).
        On some cold winter nights the temperature inside the polytunnel greenhouses actually falls below the outside ambient temperature.
        Its bulk features like these that make me a little sceptical of the greenhouse effect sustaining a 33C increase in the Earth surface temperature

        Click to access penn_state_plastic_study.pdf

      • Jim, for your experiment heating the water, make the water body with a depth to width ratio equivalent to that of the average ocean depth versus the height of the atmosphere above it. Then heat the air volume by a degree & see how much it warms the water body. My bet is the result would be a tiny warming of the water.

      • You are assuming the ocean mixes the heating in a very much deeper layer than it actually does. The deep parts to play much of a role on time scales of years. A shallow ocean is a better approximation in this case.

      • My whole point is that I think a warming of the air will not produce much warming of either the ocean or the land if the sun didn’t shime on them.

      • That’s “shine” above. The atmosphere has much less capacity to carry & transfer heat than water or land. Why don’t we hear about the heating of the oceans from the Earth’s interior? After all, the ocean basins are where the crust is thinnest, & there are all the volcanics along the mid-ocean ridges. All climate science papers seem to assume the oceans are only heated from above. They surely would be colder without the heat from below. Ditto the land surface.

      • Yes, the sun keeps the ocean warm. If the air has more CO2 it keeps it even warmer.
        Heating from underneath is not a big factor. The coldest water is at the bottom, and it stays there. We don’t see convection coming up.

      • Or maybe we could invent a whole new field of ‘experimental climatology’.

        That would be very interesting … and could help to settle many of the fundamental disputes between the ivory-towered academics (warmists by and large) and practical engineering types (sceptics mostly) that we see on this blog.

      • There already is a lot of experimental atmospheric science measuring radiation spectra and such from earth and space. No surprises yet regarding the theory of those spectra. It is well established, and again skeptics don’t even know these field measurements are happening all the time.

      • Fine. super duper. Triffic.

        Now put all that together with some air and do the measurements of what actually happens when you apply that bit of physics onto the chosen subject.

        You build a chain of logic step-by-step and test each link by experiment. So far the first link – that gases absorb and emit radiation has been proved. Great. No surprises there – astronomers have been doing that for 100 years. Chemists for as long.

        Only about a hundred other links to go.

      • You have a very odd idea of what ‘sceptics’ do and don’t believe. And why.

        Read the ‘denizens’ thread on this blog. You will find that the majority of sceptics posting here are practical experimental engineering types…with lots and lots of real world experience.

        And their scepticism is mostly rooted in that background. They have a healthy regard for theories…but and even healthier understanding of how far removed theory can be from practice when it gets out of the computer science lab and into real-world applications…like weather and oceans and clouds and rain and climate and stuff.

        My own scepticism started when I tried to find the real experimental data behind ‘The Science is Settled’. And I was appalled to discover that there is hardly any. Just a lot of ‘adjusted’ historical data, some very dodgy statistics and a terrifyingly naive belief in unverified computer models as a substitute for real experiment.

        So please don’t jump to conclusions about what ‘sceptics’ believe.

      • You have a strange idea of what atmospheric scientists do. The modelers, experimentalists and theorists all talk to each other to establish the building blocks. A model that did not match the observations would not last long. I see a lot of skeptics asking for observational proof, and that was what I was answering about. I can’t make you believe data, whether it is Spencer’s UAH, or Jones’s CRU, but they agree with each other independently.

      • In which case this entire thread is redundant. You can satisfactorily explain the raw ‘greenhouse effect’ with existing experiments.

        Go ahead. Since I’ve never had a problem believing in such an effect anyway, I’m prepared to be convinced.

      • Like I mentioned above, measurements of 250 W/m2 from clear sky (at night) are the raw evidence for the greenhouse effect. Counter-theories fall at this observational hurdle, but still they don’t go away, but just get perpetuated on the Web and even in books.

      • Latimer Alder

        Lets assume I want to explain to my mate Joe Sixpack exactly why your observation is a ‘proof’ of the ‘greenhouse’ effect. How would I do so? Assume Joe is an intelligent layman. work from 250 W/m2 –> therefore ‘greenhouse’ effect.

        I;m not trying to be clever here…pesrsonally I don’t need convincing that there is an effect…but Joe SP might,

        (We can cover climate feedbacks, ‘unprecedented’ warming, runaway warming and catastrophies…and their experimental proofs later. Let’s just start with the easy stuff).

      • Regarding Joe SP, this tells him that people who don’t believe in the greenhouse effect have predicted a measurement would be zero, when in fact it is 250. I think that is all he needs to know to form an opinion on those people.

      • PS Hell will freeze over before I believe anything at all that comes from CRU.

        I read Harry_Read_Me, and as a professional IT guy it is terrifyingly amateur.

      • Latimer Alder

        Like expecting to hear from an F1 racing team and instead meeting a kiddie with a Dinky toy and big ideas. That far away from ‘world class’. Shocking and disturbing that said kiddies are in charge of the xrawx – sorry ‘adjusted’ data on which all of AGW theory rests. And none of you guys even raise an eyebrow at their sloppiness.

        Do you not understand? Not care? Or just believe that it must all be a made-up story by evil Big Oil- funded deniers?

      • It is no sufficient that things work in understood ways if you can’t compute it properly and have no way to test your understanding of how to handle the complexity. You say clouds interact with radiation in “understood ways” but 1) it is impossible to simulate this accurately and 2) the GCMs are at such a coarse scale that they use parameterizations of clouds, not cloud microphysics. Just because you think you understand something does not mean you can make predictions.

      • Craig, you specifically said in your last comment that “There are three fundamental problems with the greenhouse model as an adequate model.”

        Now you are jumping to the argument that we don’t know everything about clouds or how they will evolve in the future. These two ideas don’t logically follow and is just a revision of the “we don’t know everything, therefore we know nothing fallacy.”

      • Chris Colose | November 30, 2010 at 5:45 pm
        just a revision of the “we don’t know everything, therefore we know nothing fallacy.”

        And this is the ‘all or nothing, one extreme or the other’ fallacy.

        The point is we don’t know enough to model the climate in a way that enables them to be made with enough certainty to be useful as policy making tools.

      • Chris Colose | November 30, 2010 at 1:06 pm | Reply
        Craig, none of these issues are ‘problems with the greenhouse model’

        No one argues that cloud aren’t complex, but they are still physics problems.

        Yes, currently intractable physics problems. Which means 90% certainty is out of the window, off the bell curve, down the plughole.

  17. Judith
    From the sceptic viewpoint looking in, it appears that climate scientists over empathise the radiative transfer of energy, almost to the exclusion of the other means of heat transfer.

    Yet when I try to follow their logic there seems (to me) to be large gaps.
    For instance I am led to believe that at night that radiative transfer from the atmosphere is all that stops the Earth surface from temperatures of the order of -100C

    Fair enough, so I pick 15um radiation for a calculation of probabilities.
    Picking an atmospheric temperature of 243K, I do a probability calculation using Maxwell Boltzmann statistics.
    Absorption by CO2 of 15um from the Earth surface is very likely since virtually all CO2 molecules will be in translational mode only.
    The extra energy absorbed is shared out by collision with mainly N2 and O2.
    However emission due to subsequent collision is highly unlikely compared to absorption.

    1% for CO2 stopping ” dead”
    0.25% for CO2 to be left with average KE
    either I’m making a mistake or CO2 in this situation seems to be cooling the surface

    • David L. Hagen

      See Roy Spencer: Why 33 deg. C for the Earth’s Greenhouse Effect is Misleading September 13th, 2010
      But what many people don’t realize is that the 33 deg. C of surface warming is not actually a measure of the greenhouse warming – it represents the balance between TWO competing effects: a greenhouse warming effect of about 60 deg. C (the so-called “pure radiative equilibrium” case), and a convective cooling effect of about 30 deg. C. When these two are combined, we get the real-world observed “radiative-convective equilibrium” case.

      • David from Roy Spencer
        …..”The value of 33 deg. C represents the difference between the observed average surface temperature of the Earth, and the estimated surface temperature if there was no atmosphere.”…..

        This is a fairly artificial construct.
        What about the storage capacity of the Oceans?
        What about an atmosphere without CO2 and H2O radiating in the IR.
        To come up with the worst case scenario- a bare rock Earth and then speculate that without IR active CO2 and H2O average Earth surface temperatures would be around 255K is claiming to much for the so called “greenhouse effect”.

      • David L. Hagen

        What are you trying to say?
        Ocean storage has very little impact on average temperature.
        (only day-day – season-season averaging etc.)
        No CO2 & H2O – no IR absorption / radiation
        = no conventional planetary “greenhouse”.

      • These “what ifs” are a bit artificial.
        A more instructive comparison to test the true extent of the Greenhouse Effect would be;
        1. Bare rock Earth temperature 255K .
        2. Earth with Oceans and Atmosphere butwith no IR radiative gases.
        I would guess that 2 would be pretty close to our own climate a bit warmer by day and colder by night.
        But nothing like the difference of 33c claimed for the Greenhouse Effect.

      • The post may be misinterpreted the CO2 and H2O would be there but not IR active.

  18. If we want to convince people of the reality of the greenhouse effect, we should first stop calling it the greenhouse effect – I guess this is the reason for the “?” in the title of the thread. I assume we are all agreed that the “greenhouse effect” is not what keeps greenhouses warm? Although there is much nonsense in the G&T paper, they are correct about this – largely by quoting extensively the experiments of Wood (1909). Greenhouses work by preventing the hot air rising. There are other obvious simple experiments to confirm this – eg open a roof window a bit and most of the heat is lost, which wouldn’t happen if absorption and re-emission of radiation was the main process.
    This highlights a major flaw of climate science – it seems to only consider radiative heat transfer and neglect heat transfer by convection.

    • I agree, the word greenhouse is a misnomer, but I suspect we are stuck with the terminology.

      • Irreversibly stuck … something about hoists and petards, I think

      • I think that is too bad. It is difficult to sell the idea that something is an egg, when in fact it is not.
        Frankly I think as people get more and more frustrated with a science community that is obsessed with what is in effect marketing, the drift of names, from global warming to global climate change to global climate disruption added ot the idea that what we are told is the fundamental driving mechanism is in fact not at all properly named to the point of being misleading, is not going to do the climate science community any good at all.
        Add to that the string of dubious claims and failed predictions, and you have yourselves painted into a corner.

    • PaulM
      R W Wood showed that;

      1. A glasshouse only worked by stopping convection.

      2. The remaining real residual radiative effect was so small that it could be almost ignored.
      Its this second part that most people seem to miss.
      Here is a paper that more people should read.
      Especially as it comes from a source with no “spin” on the AGW debate.

      The way I read the paper is it gives massive support for the conclusions of the famous Woods experiment.

      Basically the project was to find if it made any sense to add Infra Red absorbers to polyethylene plastic for use in agricultural plastic greenhouses.

      Polyethylene is IR transparent like the Rocksalt used in Woods Experiment.

      The addition of IR absorbers to the plastic made it equivalent to “glass”

      The results of the study show that( Page2 )

      …”IR blocking films may occasionally raise night temperatures” (by less than 1.5C) “the trend does not seem to be consistent over time”

      Thus inside a large real greenhouse where all the IR is trapped it made little or no difference to the temperature.
      Its this kind of simple bulk effect that makes sceptics convinced that the greenhouse effect is so small that it cant be responsible for the claimed 33C increase in atmospheric temperature

      • Bryan, do you have a citation or even better a link to this paper?

      • David Wojick

        Sorry all I forgot to post the link.

        Click to access penn_state_plastic_study.pdf

      • I may comment more later since I only skimmed the short paper.

        First, did you notice this comment in the results section: “but again the unreplicated nature of that comparison make firm conclusions difficult to make.”

        Second, an important part of blocking radiation is the thickness of the material. That paper is unclear about these values (though I suppose I could research the Dupont materials used). If you have something very thin, adding an even thinner coat of anything may not do much. Common glass is much thicker than Seran wrap, for example. If we have significant thinness, then getting any extra heating suggests there may be something very significant there (contrary to what you are suggesting). Note, that our atmosphere is very thick, allowing IR to be absorbed not just over microns, but repeatedly over miles.

        Third, a professor Pratt performed a recent experiment that strongly suggests a greenhouse effect. One point he noticed is that it takes extra time for extra warmth of many degrees C (rise in temp in his experiment setup) to accumulate under various covering scenarios (with some being better greenhouses than others). Waiting an hour, I think, was not sufficient to differentiate the different qualities of coverings, with the better material actually going a little slower in heating up to its final significantly higher temp. In the case of the current experiment, they were interested simply in slowing down heat loss, not in seeing how hot it can get, and they appear to have covered the vegetables in the evening after most of the sunlight strength was already diminished. [Again, they give few experimental details — just like the meager details given by Woods.]

        In short, the authors don’t claim any firm conclusion, and the paper lacks much detail and doesn’t explore the problem too seriously (they state this). They certainly make no claims that greenhouse effect is negligible (as you appear to want to suggest from their paper) or that our thick earth atmosphere somehow doesn’t improve the heat retention ability of the environment during the night when all we have out there is super cold space with almost no direct radiation.

  19. Judy – Let me put a plug in for Raymond Pierrehumbert’s forthcoming opus, “Principles of Planetary Climate” (Univ. Chicago), due out in January (I have a draft copy). Like some of your other references, it addresses radiative transfer and greenhouse effect mechanisms in quantitative detail, but also deals with many other climate-related topics. Unlike what seems to be the focus of some of the texts you’ve listed, this one addresses the effects of greenhouse gases not only from the perspective of radiative transfer, but also with a discussion of radiative/convective equilibrium as well as other processes relevant at various levels – surface, boundary layer, free troposphere, stratosphere, etc. I believe that anyone who makes the (sometimes arduous) effort to read through the book will come away with a good understanding of the greenhouse effect as well as the role of specific molecular species (CO2, methane, water, etc.) in mediating the effect.

  20. Perhaps things might be easier to discuss if we simply dump the term ‘Back Radiation.’ Scattering would probably be a better term. Scattering is something we see with our own eyes quite often. That glow we see along the beam of our automobile headlights is due to scattering of light. That glow along the beam of other automobiles is also visible and even provides a little illumination of objects at the side of the road. While not being particularly helpful for driving at night, it does provide an example scattering we can all relate to.

    Next, can we also dispense with the idea that cooler things cannot add heat to warmer things. All things, including atmospheric gasses radiate photons at wavelength’s and quantities determined by the object’s absolute temperature. Photon’s are not particularly smart. They really have no clue whether they are heading toward a hot or cold place. Whatever absorbs them will receive their energy.

    Whatever happens after a photon is absorbed is determined by the specific materials characteristics. It may simply get warmer. It may quickly re-radiate a new photon of the same wavelength in either the direction the original photon was traveling or some other direction (scattered). It may re-radiate a photon of lower energy level and keep some of the energy as heat.

    Those two fairly simple concepts allow us to conclude that thermal radiation from atmospheric gasses, or at least part of it, can occur in the direction of the ground and will have a heating effect. The issue should not be whether those effects exist but what their total quantitative impact is.

    Does that satisfy non technical folks without violating any scientific principles in describing a ‘green house’ effect? (By the way, I am one of those folks labeled as a skeptic.)

    • No, it isn’t scattering. It is thermal emission of IR.

      • Point taken. I do suppose we can say that convection can move heated air upward which can raise the energy level of CO2 molecules to occasionally reach a state where they release a photon to return to a lower energy state. Also, I suppose absorption of an infrared photon could result in a simple kinetic energy increase which gets transferred to nearby gas molecules. Then again, the CO2 molecule could just remain at that higher energy level for a while and then shoot a photon out at a later time. All are fun reactions to contemplate. I’m curious what the higher probability thermal activity of CO2 molecules is at high altitudes.

      • Gary,

        ‘Then again, the CO2 molecule could just remain at that higher energy level for a while and then shoot a photon out at a later time.’

        That is the dominant process that leads to the greenhouse effect.

        A CO2 molecule absorbs an IR photon giving off by the thermally excited surface of the earth (earthlight). The energy in that photon gets redistributed by non-radiative relaxation processes (collisions with other molecules mostly) and then emits a lower energy IR photon in a random direction. A collection of excited CO2 molecules will act like a point source, emitted IR radiation in all directions. Some of that light is directed back at the surface of the earth where it is absorbed and the whole thing happens over again.

        All of this is very well understood, though in the context of the CO2 laser. If you’re interested in these dynamics, there is a great literature on the relaxation processes (radiative and otherwise) that occur in an atmosphere-like gas.

      • Maxell:
        That sounds correct to me. I guess it was a little simplistic to label it scattering. I was shooting for a non-technical image and took a liberty with terminology.

      • Gary,

        I don’t think it’s simplistic to label it scattering. Inelastic light scattering does occur in the atmosphere from CO2 molecules. It’s known as Raman scattering in fact.

        It’s just that the probability of a scattering event is much, much smaller than the probability of an absorption and then subsequent emission. So I think that’s why some took note of your using the word ‘scattering’ in this context. Because that word is used to denote an entirely different physical process.

        Keep thinking though. We need it.

  21. Dr. Curry, I have no scientific dog in this fight so I watch it merely as an issue analyst. I am struck by the fact that you first say greenhouse skepticism is erroneous and not plausible, then add “I don’t have a full understanding of what the actual issues are with the greenhouse effect skeptics.”

    This reminds me of Thomas Reid’s famous response to Hume regarding the problem of induction. Reid is said to have said something like “If a man goeth down a road and by and by he falleth into a coal pit, it taketh no great wit to see that he hath made a wrong turn.” There is of course no rebuttal in this colorful response. “You must be wrong” is not an argument.

    Thus it might be that in addition to a deeper understanding of the greenhouse effect it might be of some use if we actually understood the skeptics. Just a thought.

    For example, so far as I can tell one basic problem is that in addition to radiative transfer there is a lot more going on in the atmosphere. Perhaps we are lacking a good account of the mechanism (as you suggest) just because that mechanism is both complex and ill defined. Then too there is this puzzling “talking past” in which the different sides do not seem to be responding to specific points made by the other. As Kuhn pointed out this problem is often a symptom of deep conceptual and theoretical differences.

    Taken together these two problems may help explain the complex confusion that seems to pervade this issue of the greenhouse effect.

    • When somebody says the greenhouse effect doesn’t exist, i do not regard this as plausible. I have not delved deeply into the arguments made by Miskolczi and others, but on the surface they make no sense.

      • “I have not delved deeply into the arguments made by Miskolczi and others, but on the surface they make no sense.”

        Please take a look at the thread on my blog. There are a couple of links to non-technical summaries which may help with orientation on the technical paper.

      • The Miskolczi makes sense to me and I barely understand this stuff. His is an argument from observation, not from theory. I am not saying he is right just that his is a real scientific argument, the kind that one has to delve deeply into. In fact I know of no serious scientific issue that can be resolved without delving. Science is hard. Perhaps you should just say you are skeptical, not that they can’t be right, given that you have not done the work.

      • “His is an argument from observation, not from theory”
        This is an argument that Miskolczi supporters resort to when the flaws in the theory become too obvious. But it just isn’t true. M used a whole lot of radiosonde data from a database. But it doesn’t include any radiative measurements, or other data that would substantiate his theory. It’s just pressure and gas concentrations, which he then feeds into his computer program for radiation modelling. The “observations” in no way validate his claims.

      • Nick, as Nullias in Verba points out below, the noddy greenhouse explained by most physicists is for an Earth with no convection in it’s atmosphere. Miskolczi uses gas pressure, adiabatic lapse, and concentrations because these are what matter over most of the atmospheric column. You guys really need to get a handle on the relative scale of effects.

      • Miskolczi refers to no convection data in his 2010 paper, and it doesn’t seem to appear in his theory. His calculations seem to be entirely radiative, based on his LBL program.

      • You just said:

        “M used a whole lot of radiosonde data from a database. But it doesn’t include any radiative measurements, or other data that would substantiate his theory”

        Now you say:
        “His calculations seem to be entirely radiative, based on his LBL program.”

        So which is it?

        If you are going to say he isn’t using empirical radiative measurements, I’m sure you’ll find he is in good company…

      • Exactly as stated. He has no radiative measurements – his main activity is doing radiative calculations using Hartcode. This needs as input data gas concentrations and pressures, which he gets from the radiosonde database.

      • He’s not the only person doing theoretical work with models of radiative transfer and empirical data is he? Why do you think this constitutes a fatal criticism? It’s good that a variety of different approaches are used, utilising those longer term datasets we have, even if they are not perfect.

        What he has discovered is that there has been a balancing out of effects from co2 and water vapour, and this leaves the field open for other causes of warming. i.e. the empirically measured decrease in tropical cloud cover 1980-1998 which no warmist ever wants to discuss. I calculated that this led to a 4W/m^2 forcing on the surface in the 1993-2003 decade. That was mostly the sun’s energy, not increased back radiation Nick.

  22. There are two textbooks I can think of offhand that give very good accounts of what happens when you enhance the greenhouse effect. Houghton’s ‘Physics of Atmospheres,’ and the best description I’ve seen is not yet available to the public, but within a month or so look for Ray Pierrehumbert’s upcoming text on ‘Principles of Planetary Climate.’ These texts both involve math and physics a bit beyond the typical blogosphere article, but it’s not so technical as, say, Goody and Yung and so I think they are accessible to an educated audience. Pierrehumbert, in particular, makes you think of the greenhouse effect in the context of other atmospheres as well, such as on Mars/Venus, during snowball Earths, and he delves into the runaway greenhouse issue.

    Another great book is Grant Petty’s ‘Introduction to Atmospheric Radiation.’ If you’re interested in all the molecular-stuff at a respectable but not overwhelming level of technical detail, this is a must-read (definitely know algebra, just a bit of calculus and physics). It’s probably the best radiation book for an audience making the transition from a few calculus and physics courses to their first atmospheric science exposure. Petty is a satellite expert and so he focuses a lot on spectra from space and you’ll be very good at interpreting them after reading the book, but he doesn’t actually complete the connection to the greenhouse effect, although he could have easily connected the dots to one of the better ways of describing how it actually works.

    Any review of these texts will allow the reader to dissect the problems in G&T, Miskolczi, or ‘slaying the dragon’ type reads. None of them even provide respectable objections to the current physics, they are just ridiculous. Some people will say they reflect poor understanding, but IMO they are so bad in their understanding that they reflect an intent to mislead. Our paper, Halpern et al (2010), was a bit off in that wingnut theories typically won’t get the detailed ‘rebuttal’ that we provided, but we (or at least I) felt it was necessary only because of the importance of the topic to the non-scientific audience, but I was unaware of how this article was allowed to get into IJMPB. It only shows if you are persistent enough you can publish anything, especially in obscure journals (like Miskolczi’s original piece in a odd Hungarian journal or E&E)

    Part of the issue with the whole ‘blanket analogy’ is that it tells you the greenhouse primarily impacts the “energy out” side of the budget rather the “energy in” part, but that’s it– it doesn’t actually tell you how this works. It doesn’t allow you to appreciate the importance of the thermal structure of the atmosphere, the spectral selectivity in absorption, etc. All of these things can be learned without understanding Partial Differential equations and other sorts of crazy math, but allow you to actually understand what’s going on and connect it to other situations.

    • Depending on the level you are looking for, I would also again recommend chapters 2-5 of David Archer’s “Understanding the Forecast” (2. Blackbody radiation, 3. The layer model, 4. Greenhouse gases, 5. The temperature structure of the atmosphere), which comes with video lectures and online models related to learning assignments in the text.

      This is taught at the first year university level, but does not require much prior work in physics or chem. He also does a pretty good job of discussing the physics at the molecular level, as I recall.

    • Chris, since the net longwave gives a cooling of ~70W/m^2, mostly due to the concentration of back radiation on the sea surface causing evaporation, please could you explain why more back radiation wouldn’t lead to more evaporation.


  23. As a simple lay-person, I like to keep things simple. Of course, in the ‘climate debate’ hardly anything is simple but try this for size…

    As a skeptic (or at least one who is developing that view!), I certainly do not ‘deny’ that the greenhouse gasses have a property which can lead to, er… a warming effect commonly known as ‘greenhouse’. (How anyone got away with coining that term just shows how science can be corrupted by soundbites.)

    Anyway, it is the scale of the warming effect of ghgs that most skeptics can’t accept. Please correct me if I’m wrong in that assertion.

    All the greenhouse gasses put together amount to 0.04% of the dry atmosphere. This means that 99.96% (Nitrogen, Oxygen and Argon) CANNOT be ‘heated’ by radiation (because radiation is not heat per se) and therefore must be warmed by conduction.

    If you add H2O to the mix (I will use an average of 2%), you effectively have, out of a total of 102%, 2.04% being able to absorb radiation, 0.04% being able to re-emit radiation and 99.96% which can only be warmed by conduction. (I discount convection because, in reality, convection merely re-distributes any heat.)

    The (C)AGW theory appears to be based on a belief that an increase in the 0.04% (actually 0.028% back in 1850) is likely to cause a significant, nay, catastrophic rise in global temperatures.

    A debate on the core physical processes of the Greenhouse Effect is probably long overdue.

  24. BlueIce2HotSea

    So what have we been doing wrong in terms of explaining this…

    It is what you suspect; your target audience has not been targeted. Most information is directed at either school children or climate scientists.

    Here is the skeleton of a simple Newtonian approach, no need to start off with radiative heat transfer and atmospheric physics:

    1. The rate of incoming heat and outgoing heat at the Earth’s surface is convergent.
    2. The rate of heat flow between two bodies is proportional to the difference in temperatures.
    3. The direction of heat flow is from the higher temperature body to the cooler. (Sun to earth, earth to space)
    4. If (perhaps a big if) we can choose one body as the surface of the earth, and the second body as a spherical shell consisting of the atmosphere at a given height from the surface, then:
    5. Any increase in temperature in the atmospheric body will decrease the rate of heat transfer from the surface.
    6. The resultant increased surface heat will cause surface temperatures to increase until the out-going rate of heat flow converges to the incoming rate.

    What remains is making the case for atmospheric temp increase (#5) and quantifying the magnitude of the effect.

    Too simplistic for physicists, yes. But not necessarily unacceptable to the intended audience.

    • BlueIce2HotSea

      To put some sinew on the skeleton, if the shells are arbitrarily thin, then the higher order terms of the Stefan-Boltzmann equation drop out. Radiative heat transfer becomes proportional to the difference in temperature – similar to convective cooling in a Newtonian ‘body’. Also, arbitrarily thin shells could represent both a Newtonian ‘body’ and an S-B surface. We end up with different constants for convective and radiative heat transfer and a model analogous to an electric parallel circuit.

      This seems intuitively sensible, qualitatively correct and all we need are 1st order differential equations to make it work.

      Isn’t this better than a greenhouse model?

      • BlueIce2HotSea

        Of the several problems I see with this approach, the biggest is that it only works when looking at heat transfer between adjacent temperature shells. Otherwise the errors caused by dropping the higher order S-B terms accumulate.

  25. Here is an item I wrote up to try and crunch the whole GHG idea into a paragraph.

    The Greenhouse Effect 101 . To balance incoming short-wave solar energy, the earth constantly radiates long-wave heat energy (infrared photons) back out to space. This is how the earth maintains a fairly constant temperature. Some of the outgoing radiation is absorbed & re-radiated back to earth by various trace gases like carbon dioxide, water vapor, methane, & ozone. This gives us the greenhouse effect, & keeps the earth from becoming a ball of ice. The warming effect works according to the laws of quantum physics. When an infrared heat photon of a specific energy collides with a greenhouse gas molecule, that photon will be absorbed & an excited electron will jump to a higher orbit. However, that new electron orbit is not stable & will soon fall back to its “rest” energy level, & emit a photon of the same energy level (frequency) that was absorbed. While the original photons were on their way out to space when they were absorbed, the new photons will be radiated in a random direction. Some will continue up, some will return toward earth. It is this back-to-earth radiant energy that causes the underlying earth to warm to a new equilibrium temperature. Because of its complexity, only computer models can adequately quantify the greenhouse effect. The exact magnitude of this effect caused by gases humans are adding to the air is not yet precisely understood, but all models to date have shown various amounts of warming. None show the earth cooling or remaining the same.

  26. I am Dr Curry’s assumed audience – I took the right courses, plus a few more in grad school, but I am not prepared to do a degree in physics or climate science. As a luke-warmer, I’d like to make a distinction: Not all ‘deniers’ deny the existence of a greenhouse effect. I’ll grant a greenhouse effect to CO2, to what ever degree the relevant scientific community assign it.

    What I’m skeptical of, however, is the degree to which adding CO2 to the atmosphere affects planetary heat content and temperatures. We’re not dealing with a sealed flask of gases on a lab bench here. While I do not know the physics of the various and sundry forcings and constraints of planetary temperatures, I do know that in spite of the fact that CO2 levels have gone up and down over billions of years, there has never been a thermal runaway that burned the water off the planet. Why not?

    Obviously, there is some kind of thermostat operating – and operating against the effects of greenhouse gases. If Hansen’s tipping point exists now – at the end of Obama’s first term – then why hasn’t it existed in the past? This is where my skepticism is located – not in the existence of the greenhouse effect – which is well-demonstrated – but in the chaotic mix of causes that sum together to give us our planetary climate. I don’t need convincing on the greenhouse effect – I need convincing on the application of the greenhouse effect to measured decadal predictions of planetary climate.

  27. Another very simple analog is a layer of heat insulation. Of course diffusion of heat in an insulating layer is in many ways different from the psysics of radiative transfer of energy in the atmosphere, but the basic idea is rather similar.

    When the incoming flux of energy is kept the same, better insulator leads to a higher temperature at the source. In the equilibrium the outgoing heat flux equals the incoming flux, but the temperature gradient is higher to compensate for the reduced conductivity of heat.

    • Simple, yet false analogs are not what we need. As Dr. Curry indicated in her initial post, the debate is over detailed mechanisms.

      • Perhaps I should expand a little on the analogy.

        In the diffusion of heat the macroscopic diffusion equation is very simple, but on the molecular level the situation is much more complicated. On the molecular level, energy is transmitted in all directions exactly as absorption and emission proceed in all directions. If absorption is strong at all wavelengths the radiative energy transmission proceeds very closely as heat conduction in solid insulator. The principal difference comes from those wavelengths where radiation can get through a large fraction of the atmosphere and even the whole atmosphere without being absorbed. Convection adds to the differences, but even so the dominant effect is very similar.

        I think it is very much debatable whether this is a false analog. Certainly it is not complete, but is it really false?

  28. Nullius in Verba

    A great deal of confusion is caused in this debate by the fact that there are two distinct explanations for the greenhouse effect: one based on that developed by Fourier, Tyndall, etc. which works for purely radiative atmospheres (i.e. no convection), and the radiative-convective explanation developed by Manabe and Wetherald around the 1970s, I think. (It may be earlier, but I don’t know of any other references.)

    Climate scientists do know how the basic greenhouse physics works, and they model it using the Manabe and Wetherald approach. But almost universally, when they try to explain it, they all use the purely radiative approach, which is incorrect, misleading, contrary to observation, and results in a variety of inconsistencies when people try to plug real atmospheric physics into a bad model. It is actually internally consistent, and it would happen like that if convection could somehow be prevented, but it isn’t how the real atmosphere works.

    This leads to a tremendous amount of wasted effort and confusion. The G&T paper in particular got led down the garden path by picking up several ‘popular’ explanations of the greenhouse effect and pursuing them ad absurdam. A tremendous amount of debate is expended on questions of the second law of thermodynamics, and whether back radiation from a cold sky can warm the surface. It can, but whether it can or not is actually completely irrelevant to the real greenhouse effect. It doesn’t matter, because back radiation isn’t what causes the surface to be as warm as it is anyway.

    The greenhouse effect requires the understanding of two effects: first, the temperature of a heated object in a vacuum, and second, the adiabatic lapse rate in a convective atmosphere.

    For the first, you need to know that the hotter the surface of an object is, the faster it radiates heat. This acts as a sort of feedback control, so that if the temperature falls below the equilibrium level it radiates less heat than it absorbs and hence heats up, and if the temperature rises above the equilibrium it radiates more heat than it is absorbing and hence cools down. The average radiative temperature for the Earth is easily calculated to be about -20 C, which is close enough although a proper calculation taking non-uniformities into account would be more complicated.

    However, the critical point of the above is the question of what “surface” we are talking about. The surface that radiates heat to space is not the solid surface of the Earth. If you could see in infra-red, the atmosphere would be a fuzzy opaque mist, and the surface you could see would actually be high up in the atmosphere. It is this surface that approaches the equilibrium temperature by radiation to space. Emission occurs from all altitudes from the ground up to about 10 km, but the average is at about 5 km.

    The second thing you need to know doesn’t involve radiation or greenhouse gases at all. It is a simply physical property of gases, that if you compress them they get hot, and if you allow them to expand they cool down. As air rises in the atmosphere due to convection the pressure drops and hence so does its temperature. As it descends again it is compressed and its temperature rises. The temperature changes are not due to the flow of heat in to or out of the air; they are due to the conversion of potential energy as air rises and falls in a gravitational field.

    This sets up a constant temperature gradient in the atmosphere. The surface is at about 15 C on average, and as you climb the temperature drops at a constant rate until you reach the top of the troposphere where it has dropped to a chilly -54 C. Anyone who flies planes will know this as the standard atmosphere.

    Basic properties of gases would mean that dry air would change temperature by about 10 C/km change in altitude. This is modified somewhat by the latent heat of water vapour, which reduces it to about 6 C/km.

    And if you multiply 6 C/km by 5 km between the layer at equilibrium temperature and the surface, you get the 30 C greenhouse effect.

    It really is that simple, and this really is what the peer-reviewed technical literature actually uses for calculation. (See for example Soden and Held 2000, the discussion just below figure 1.) It’s just that when it comes to explaining what’s going on, this other version with back radiation getting “trapped” gets dragged out again and set up in its place.

    If an increase in back radiation tried to exceed this temperature gradient near the surface, convection would simply increase until the constant gradient was achieved again. Back radiation exists, and is very large compared to other heat flows, but it does not control the surface temperature.

    Increasing CO2 in the atmosphere makes the fuzzy layer thicker, increases the altitude of the emitting layer, and hence its distance from the ground. The surface temperature is controlled by this height and the gradient, and the gradient (called the adiabatic lapse rate) is affected only by humidity.

    I should mention for completeness that there are a couple of complications. One is that if convection stops, as happens on windless nights, and during the polar winters, you can get a temperature inversion and the back radiation can once again become important. The other is that the above calculation uses averages as being representative, and that’s not valid when the physics is non-linear. The heat input varies by latitude and time of day. The water vapour content varies widely. There are clouds. There are great convection cycles in air and ocean that carry heat horizontally. I don’t claim this to be the entire story. But it’s a better place to start from.

    Of course, given the number of competing claims and assertions in this area, all this just looks to most participants like another crank version and gets ignored, even by sceptics. I am not hopeful. It would require more ‘official’ climate scientists to stop using their simplistic version and be a bit more careful, and would take a while to percolate into the public consciousness. Probably they don’t because they know what people would say.

    My apologies for the length of this comment.

    • (This was supposed to be a reply so I am reposting it as one, to maintain continuity.) Very nice, Nullius (may I call you Nullius?). This at least gives the flavor of the complexities, and helps explain the confusion that is obvious wherever this topic arises. I am particularly interested in the idea that some calculations “use averages as being representative, and that’s not valid when the physics is non-linear.” The challenge of non-linearity pervades climate science, so it is not surprising to find it here too.

    • While this is one way of looking at the effect and while this may well be the way atmosphere is analyzed, this does not explain the effect. The basic physical process that leads to warming is the absorption and emission of radiation. Any explanation must start from that fact.

      You stated that purely radiative calculations are self-consistent. They give also quite reasonable results. A simplified atmosphere with convection is likely to lead essentially to the same result. This is hardly a coincidence. I think that there is more truth in the purely radiative calculation than you indicate. It is just another way of looking at the same effect. It has the great advantage that it is based more directly on the basic physical effect.

      (I am not an atmospheric scientist, but I am a physicist. These comments are largely based on my intuition as a physicist.)

      • Again I continue my argument.

        My understanding is that convection is not essential for understanding the greenhouse effect. The lapse rate is determined by the stability condition of the atmosphere. If the lapse rate would not be correct strong instabilities would be created, but when it is correct they are not essential for explaining the effect. Therefore it is correct to explain the greenhouse effect looking only at the radiative energy transfer. Minor corrections are needed for precise results but not for explaining the effect.

      • No, convection is not a ‘minor correction’ and a convective atmosphere has a completely different temperature profile and heat flux balance from a purely radiative one. The purely radiative argument – as described repeatedly by climate scientists – is simply not what happens in the Earth’s atmosphere.

      • I was not clear enough on what I ment. My idea was that the temperature profile remains essentially the same and changes in the profile are “the minor correction”.

        Perhaps my later posts (from Dec 1) lower in this chain explain better, what I have in mind.

      • Nullius in Verba

        It’s not quite that straightforward, it is like asking why the temperature of a pan of boiling water on the stove is 100 C. The answer given is that it is because the gas is set to 2, which liberates so many Watts of heat, which enter the bottom of the pan and increase its temperature. So if I turn up the gas to 4, will the temperature of the water increase to 200 C?

        Instead, what I am saying is that the reason the temperature of water is 100 C is that it is the boiling point of water, and that more heat only increases convection, which carries the heat away faster.

        As an explanation for why the water is hot, the gas is a good starting point. As an explanation for why the temperature is 100 C, the gas argument misses the point entirely. It doesn’t control the temperature.

        The radiative calculation gives the wrong result. The temperature profile of the atmosphere would be exponential, not linear, and the average surface temperature would be around 60 C (IIRC). Both are contradicted by the evidence. The controlling mechanism is not back radiation.

      • I am not sure, whether you read my later message.

        It is clear that one cannot explain the temperature profile of the atmosphere or even the change in the temperature profile by radiation only.

        Still one may be able to explain the warming of the earth surface through the analysis of the radiative processes with a minor complementary addition from the stability condition.

        (Here I naturally consider only the basic effect forgetting many types of feedback.)

      • Nullius in Verba

        Yes, this is exactly what the climate scientists do. They explain only the pure radiative process, and then describe convection as a “minor correction”. Except that from the point of view of the dynamics, it isn’t minor – it actually dominates the effect.

        With convection, the basic mechanism is as above, and the radiation is a minor correction. (It just reduces the amount of convection necessary to maintain the adiabatic lapse rate).

        You don’t actually need to involve radiation internal to the atmosphere, and given all the fuss and argument and scepticism it raises, it’s positively counter-productive to use the radiation argument. So why use it?

      • Why use the radiation?

        Because CO2 affects the radiation and (in practice) only the radiation.

      • Nullius in Verba

        “Because CO2 affects the radiation and (in practice) only the radiation.”

        Yes. But that’s already incorporated into my model. More CO2, being opaque to longwave IR, increases the thickness of the “fuzzy layer” and hence the average altitude of emission to space. The relevant radiative property of CO2 is indeed included. Irrelevant and distracting properties are ignored.

        When I say “the radiation argument” I mean the pure radiation, non-convective argument.

      • In any case the whole effect of CO2 is radiative. That is the only direct effect. Everything else is adaptation to the changes that this basic effect creates.

        If one wishes to convince scientists with sufficient background to understand the explanations, one must base the whole argument on this. Whatever else is needed about the atmospheric physics, is addtional.

        In a stationary situation the net energy flux through any surface is zero. This fact guarantees that there are many different equally correct ways of describing the same effect. The question is, which of these ways is easiest to explain to educated people who are not atmospheric scientists.

      • Nullius in Verba

        Please, I’m not saying that the direct effect of CO2 is not radiative. But just because the first step in the chain is radiative, you can’t then say that all the rest of the chain of effects in the mechanism are purely radiative too.

        If you want to convince scientists with sufficient background to understand, you have to use the correct mechanism. If you use the wrong mechanism, and then some sceptic comes along and points out the flaws and contradictions in it, they’ll become sceptical too. They’ll start asking why they’ve been fed this line of hooey for all these years, and why none of their climatology colleagues said anything.

        I should know.

      • As I stated below, if I understand the argument you two are having: Pekka is saying that increasing the thermal resistance for a given heat flux will require higher temperatures and that analogously explains the greenhouse effect. Nullius is saying that the additional piece that is missing without taking into account the physics behind the nearly constant lapse rate is that the resistance increase comes from a thicker layer, not from a change in the insulator’s conductivity.

      • May I ask, if more CO2 increases the thickness of the “fuzzy layer”, does it actually increase temperature at say 1mtr height?

      • I believe the answer should be yes – because the radiating height is at 255K and if that point moves higher, it increases temperatures of all points between that height and earth’s surface.

      • Yes, it turns out CO2 will warm all layers of the atmosphere below the tropopause

      • Hi Chris, my question is: has it warmed all levels of the atmosphere below the tropopause? I’m only asking because I don’t know and I was just wondering if measurements had confirmed this.

      • And the predicted hotspot?

      • Nullius in Verba

        For data, see figure 1 here. The answer is that the measured trends are almost all positive, but I think that the error bars are such that this probably cannot be relied upon above 2 km.

        The tropical hotspot is not a consequence of greenhouse gases, but of water vapour changing the lapse rate. The theory of water vapour feedback says that warming from CO2 alone should increase the amount of water vapour in the atmosphere, which reduces the lapse rate (the rate at which temperature falls with altitude). The temperature at the 5 km emission level acts like the pivot on a see-saw. The temperature there rises due to CO2, the temperature below it rises but not as much, due to the see-saw levelling out, and the temperature above it rises even faster than CO2 alone requires.

        The absence of a hotspot is evidence that the feedbacks are probably smaller than the climate models suggest. It doesn’t say much one way or the other about whether GHGs are responsible for the late 20th century rise.

        The hot spot is another reason why a proper understanding of the greenhouse effect requires discussion of lapse rates, not just radiation.

      • For a recent discussion of the hotspot, see this

        Perhaps hunter has already read it.

      • Nullius in Verba

        And the follow up here, of course. Although that one does get rather technical and doesn’t present the data by altitude as clearly.

      • Pat – That is an exceptionally detailed and informative analysis of the tropospheric amplification (“hotspot”) issue. What emerges is that the main problem in accurate assessment is probably not statistics but instrumentation bias due to instrument changes over time. The most direct measurements (but also most vulnerable to instrumentation errors) come from radiosondes. All other sources are indirect, and tend to disagree with each other. Tropospheric amplification has been well documented over short intervals but remains problematic over longer ones, altough this discrepancy lends itself better to a measurement trend bias problem than a physical explanation.

        The “hotspot” is not specific to GHG-driven warming, but is an expected feature of surface warming from any cause, and does not originate in climate models but in the basic equations of climate physics, including the quasi-exponential Clausius-Clapeyron equation. It is not a reflection of feedbacks in general, but of a particular negative feedback, the lapse-rate feedback. In that sense, one could argue that if the discrepancies are real, negative feedback is weaker (and hence positive feedbacks more dominant) than is ordinarily estimated. I don’t consider that argument to be plausible, because I expect the eventual resolution of the issue will reside in improved methods for quantifying long term trends. It is also possible that a true physical process underlies a small part of the discrepancy, but since the evolution of instrumentation has progressively reduced the discrepancy over time, measurement issues probably dominate. Interestingly, although some indirect measurement techniques (e.g., MSU data) still show some distance between recorded and predicted data, other techniques show none – including thermal wind equation methods and recent observations on SST warming-related changes in convective theshhold.

    • Excellent summary, thank you. These complexities of the convective sitution is a lot of the reason for my scepticism about the ability of the current models to give worthwhile results. It is also the reason why I take any sensitivity calculation with a big dose of salt. The Earth’s climate system is a heat engine, and if some variable like the latitudinal position of the jet streams alters a small amount, the effect will make a 30% change in the level of co2 causing a 1.7W/m^2 forcing seem pretty tiny in comparison.

      • Leonard Weinstein

        The discussions by Nullius, which are basically the same as mine, tell why there is an atmospheric greenhouse effect. They do not tell anything about feedback, which is the real issue of the big debate. The main issue of feedback is whether it is large positive, small positive or negative. The result is that a doubling of CO2 may cause several degrees C per doubling, or a fraction of a degree C per doubling. I think it is a fraction of a degree C. I don’t know what Nullius thinks.

      • Leonard, we might listen if the ‘counter-arguments’ were not ones from ignorance, straw man attacks ,or other such fallacious lines of thought.

      • Leonard Weinstein

        Please give examples of what you are talking about. If it is my opinion on the amount of feedback, this is the opinion of Roy Spencer and others, and agrees with actual heating rates measured, without having to balance two large effects (heating from CO2 plus feedback with aerosols) to get a small result. If it my analysis of how the atmospheric greenhouse works, I will elaborate as much as you wish. You attack my statements with no support at all.

      • Based on your history elsewhere of what you post, I find no use in discussing this with you.

      • Leonard Weinstein

        Sounds like you are afraid to discuss facts. I had a nice discussion with scienceofdoom, and I think he is closer to my position on this model now. I am presently debating Tom Curtis on climate clash (on more general issues) and he seems to be a very smart person. What is your problem?

      • This is what always happens when we get close to the heart of the issue. Closing of ranks, and minds. Aerosol feedbacks are a figleaf to cover a failed theory. Some aerosols warm, some cool. There is no tropospheric hotspot , which would have to be there for the atmosphere to be responsible for ocean temperature increase, and instead it’s the ocean which drives the atmospheric temperature. This is easily demonstrated by observing the lag between ocean temp change and the following atmospheric response.

        It’s the Sun heating the ocean through lowered tropical cloud albedo 1980-1998 which caused the warming. The empirical data tells us this, but some people just pull their blinkers a bit tighter when confronted by the fact.

      • “Aerosol feedbacks are a figleaf ”
        Someone here, I forget who, introduced me to the term “theory-saving” to describe this kind of argument, and I rather like it. In strict terms, I suppose it’s “counter-parsimonious” :-)

      • Chris,
        Inevitably at this point, on other blogs, someone would be posting links to the Song of Sir Robin to highlight your choice.

    • Leonard Weinstein

      You got it exactly right. This is what I keep telling Chris and others, but you told it more clearly, so they may pay more attention.

    • I think another way to look at it analogously is that when you have a constant current flow (radiated heat) at steady state through a resistive material (atmosphere) with a certain resistance per unit length (equivalently related to the lapse rate), adding more CO2 is equivalent to making the resistor longer resulting in a higher potential difference (equivalently temperature difference between earth’s surface and the radiating height) across the resistor. It is not however equivalent to making the resistance per unit length higher (i.e. increasing the lapse rate) for a certain fixed length of resistor. For Pekka’s example above, it is equivalent to making the insulating layer thicker not changing the material conductivity itself.

    • Nullius, Thank you very much for your comment. I appreciate its tone and I feel like I learned from it. Your explanation does make a great deal more sense to me than the greenhouse explanations I have heard before. I take it from your later comments that you have been burned at some point regarding these issues and I would like to thank you for going once more into the breech and contributing.

    • Can I just say what a pleasure it is as a nonscientist to read such a lucid explication? Whether they agree with your science or not, Believers have much to learn from your prosaic style.

    • The prize for the best comment on the thread goes to Nullius in Verba (3:13pm).
      This morning I was going to post more on convection and its apparent neglect but I don’t need to because NV has done it much more clearly and thoroughly than I could.
      The bottom line is that you do NOT need absorption of LW radiation by GHGs to explain the surface temperature of the earth.
      Everyone should read his comment, then read it again.

      • Tomorrow I am going to open a new greenhouse thread, pulling what I regard to be the most lucid explanations of the various positions in this.

      • Nullius in Verba


        We need to be careful on this point. The greenhouse gases (mainly H2O) are opaque to IR, which is why the IR-visible surface emitting to space is not the solid surface, but higher up. The fact that they are GHGs is important. It’s not purely a pressure effect, either.

        A completely transparent atmosphere would give an average surface temperature of -20 C, and convection (which would still occur given uneven illumination) would result in an atmosphere that got even colder than this with altitude.

        There are thought experiments that I’ve gone through in previous debates in which it is possible to get a greenhouse effect without greenhouse gases – substituting various sorts of high-level clouds instead. There is also a variant on Willis Eschenbach’s ‘Steel Greenhouse’ in which the planet is surrounded by a totally opaque steel shell with a convective atmosphere underneath it. (Willis usually posits a vacuum beneath it in his version.) But I hadn’t discussed them, and they’re a bit more complicated to analyse.

    • Latimer Alder

      Please expand on ‘increasing CO2 makes the fuzzy layer thicker’. CO2 is still measured in a few hundred parts per million, and some worry about whether the few is three or four. How does even an increase from 300 ppm to 600ppm provide the heating that is claimed? The increase in CO2 of the order of one ten thousandth only.

      • Leonard Weinstein

        Even though CO2 (and water vapor and methane) is present in small amounts, it is an effective absorber of certain wavelengths of long wave radiation, with the amount of absorption related to the concentration. There is a distance near the ground that results in 50% absorption (of those wavelengths) for CO2 that is a few hundred meters. This is an exponential type absorption. Twice the distance and 50% of the remainder is absorbed, and so on. The gases that absorb also emit. If you went 10 path distances, less than 0.1% gets through, so the energy is pretty much absorbed. However, as you go higher in the atmosphere, the lower gas density results in that absorption distance increasing, until at some height, the radiation is not absorbed and goes to space. This is not a thin layer, but spread out. Adding more CO2 shortens the absorption path, and slightly raises the effective height of outgoing radiation. That is what is meant by a fuzzy layer of outgoing radiation.

      • Nullius in Verba

        The fat content of skimmed milk is about 0.3%, or 3,000 ppm, and yet it is effectively opaque to the eye. If you dilute it by a factor of ten, do you believe intuitively that you will be able to see through 10 km thickness of the stuff?

        It’s a common question, and quite reasonable, but hopefully the milk example should help the intuition.

      • Latimer Alder


        Fat molecules are much bigger than CO2 molecules…and form large clusters. I think that milk is effectively an emulsion or a suspension, not a solution of fat molecules.

        But we are talking about (effectively) a gaseous solution of N2, O2, VH4 H2O etc as well as CO2. I don’t think that your analogy stands.

      • Latimer Alder

        Sorry CH4. I don’t know of a vanadium tetrahydride. But I had an undergraduate tutor who would have revelled in trying to make it – and probably have blown up his lab in the process.

      • Nullius in Verba

        Fat molecules are bigger, yes, which will increase their area by a factor of ~50 (order of magnitude). Feel free to dilute further to account for that.

        It is also true that it forms an emulsion of droplets, but this only means that the gaps between the droplets are even bigger.

        It’s only meant to be a simple example to aid intuition.

      • OK -I think what you are trying to say is that if i looked at the atmosphere at some particular IR wavelengths, then a bit of CO2 would act like a dye, and would make the whole planet look black.

        Fair enough…I understand that bit. But what about the water vapour that is present in vastly greater concentration and is also a GHG? Or the methane that is (apparently) a far more potent (even more blackerer) GHG than CO2. Once all teh wavelength at a particular frequency is being absorbed by one of the gases then adding more of another gas doesn’t get to absorb more radiation. there aren’t deeper and deeper shades of black.

        I’d add that like others, I very much like your clear style of explanation, but in this case I think a few diagrams would help immeasurably.

      • The pictures and related explanations of John Nielsen-Gammon might help even more.

        This link has been given in this discussion at least twice before, but it is worth it. Unfortunately the third part of the presentation is still missing and a major part of the argument should be there.

        Each of the greenhouse gases (or Tyndall gases if J. N-G. gets his wishes fullfilled) is influencial exactly because they are effective at different wavelenghts. Water wapor covers most, but leaves a major gap for the others.

    • Nullius in Verba:

      Why does “Increasing CO2 in the atmosphere make the fuzzy layer thicker, increases the altitude of the emitting layer, and hence its distance from the ground.”

  29. Very nice! This at least gives the flavor of the complexities, and helps explain the confusion that is obvious wherever this topic arises. I am particularly interested in the idea that some calculations “use averages as being representative, and that’s not valid when the physics is non-linear.” The challenge of non-linearity pervade climate science, so it is not surprising to find it here too.

  30. ” Our general argument consists of the following elements:

    2. the analogy to a greenhouse”

    Can you quote any scientific source in the last 50 years or so that has based the argument on the analogy with a greenhouse?

    • Nullius in Verba

      Does NASA count as a scientific source?

      “Certain gases in the atmosphere behave like the glass on a greenhouse, allowing sunlight to enter, but blocking heat from escaping.”

      • less (IMO) on the issue of climate than I would like. lol

      • Well, it is badly put by whoever wrote it (although it’s true that glass has those properties). But there are any number of patient explanations that greenhouses and the GHE are different. For example, contrary to this post’s claim that
        “The IPCC reports never actually explain the physics of the greenhouse gas mechanism.”
        the AR4 does have a FAQ 1.3 on that, and they say:
        “The glass walls in a greenhouse reduce airflow and increase the temperature of the air inside. Analogously, but through a different physical process, the Earth’s greenhouse effect warms the surface of the planet. “

      • I was not sure, were you making a joke about the quality of the IPCC report or are you suggesting it is a good explanation?

      • The quote is not the explanation – it’s an intro remark. The FAQ itself is brief, and not a completely detailed explanation. But it isn’t nothing.

      • Nullius in Verba

        Yes, I agree that “Analogously, but through a different physical process,” is not quite “based on […] the analogy with a greenhouse”. But it’s not an explanation, is it?

        It’s also noticeable that the “like a greenhouse” argument doesn’t evoke the same level of vituperation from climate scientists as the typical sceptic argument.

      • Not when communicating to the public on their public outreach sites, they have made some real boner mistakes in this regard. I have no idea who they have writing this stuff.

      • Nullius in Verba

        True. You and I can look at it and immediately see it as nonsense. It has evidently churned out by some poor PR type who’s been told to “knock up some outreach pages on climate change”. But Joe Public is only going to see the name “NASA” at the top of the page and he’s going to assume his tax dollars have been well spent. I mean, they launch space rockets to Saturn, don’t they?

        More, he’s going to assume that the thousands of scientists scouring the web for inaccuracy (it’s amazing how fast they pop up when some new sceptic paper or argument hits the net) would have seen it and let NASA know and insisted that it be corrected. After all, that’s the sort of diligent fact-checking and public peer review by which scientists assure us of the quality of their work.

        The poor PR hack didn’t just make it up. These explanations abound for a reason. Some scientist told them that. Many others shrugged and let it pass. Such carelessness and lack of principle is why we are where we are today.

      • Not peer-reviewed, is it ?

      • Simple, document properties lists authors.

        Randal Jackson, Cecelia Lawshe, Justin Moore, Joshua Rodriguez

      • google each name with the word nasa, they are web developers

      • I doubt web developers write the technical content.
        Nasa is dumbing down a simplistic analogy to convince the masses.

      • After reading Vaughan Pratt’s little history, I sort of think we should keep GHE.

  31. I didn’t went through all the comment, but it seems that there are two main points:

    1-) Co2 is the main driver of climate.

    2-) Water vapor is the main driver .

    The biggest problem to solve which one is true, is that there exist no experiment that can be repeated time and time again showing the same or similar result.

    The use of computer model thus becomes necessary, but since there is a lot we don’t understand about the atmosphere, they require some assumption about these unknowns.

    So everyone agree that Co2 and water vapor are GHGs. But no one agree about these assumptions that cannot be proven wrong or right by anyone.

    What doesn’t help to understand the physics is the uncertainty associated with all the data set and mainly the surface temp, sea and land.

    The question is, how can we know anything if we can’t be sure of that the data used is good enough. Well other than Co2 and water vapor are Ghgs.

    • Leonard Weinstein

      It is the Sun that is the overall driver of climate. Greenhouse gases cause a slightly higher overall level in ground temperature, and water vapor is the main greenhouse gas. Clouds are also very important factor. Other greenhouse gases like CO2 and methane, and aerosols, also affect the details. The ocean currents (which have long period cycles) and location of land are also major factors (land shifts over long enough times). It gets even more complicated, since many of the processes involved are non linear chaotic process, and result in unknowable long range results. Did you ever hear of the butterfly effect? Computer models are basically a joke for long range forecasts of even trends.

  32. Part of being a scientist is knowing how much weight to place on ideas as they progress from hypothesis, to experimental testing, to provisional theory, and eventually established theory. Though I am skeptical of the IPCC consensus, most of the alternative theories you cite aren’t worth mentioning.

    The appropriate path forward for Claes Johnson’s hypothesis is to identify a physical situation where the standard theory of atmospheric radiation makes different predictions than his theory. Then do the experiment (or review the existing evidence). Claes Johnson hasn’t even proposed such an experiment. Until then, his theory isn’t worth mentioning because there is plenty of experimental evidence supporting of the standard theory (particularly observation of downward long wavelength radiation, DLR, expected to be emitted by GHGs).

    In their reply to Halpern, G&T agree that DLR from the colder atmosphere to a warmer earth is compatible with the 2nd Law, as long as more radiative energy is flowing in the opposite direction. Therefore, they do not object to the fundamental mechanism behind “greenhouse theory”. (The nature of their remaining objections is not clear to me.)

    As for Slaying the Dragon, I got as far as: “Redirecting radiant energy back to the source cannot increase its temperature”. If that energy can be absorbed (as DLR can), this statement is incompatible with the law of conservation of energy! Do the authors confront this dilemma?

    • Frank

      …….”In their reply to Halpern, G&T agree that DLR from the colder atmosphere to a warmer earth is compatible with the 2nd Law, as long as more radiative energy is flowing in the opposite direction. Therefore, they do not object to the fundamental mechanism behind “greenhouse theory”….

      This radiative interaction however does not HEAT the colder Earth surface.
      The Halpern group rebuttal mistook this for “the photons not arriving at the warmer surface.

      Hence their rebuttal completely missed the target and so G&T paper has not had a serious challenge.
      This is unfortunate since science only progresses through constructive criticism.

    • Frank, can you please give an example in the real world of when reflecting radiant energy back to the source can increase temperature?

    • Frank – you say you are “skeptical of the IPCC’s consensus.” If you get a moment, please could you briefly state the nature of that skepticism. Thanks

    • Frank,
      I asked a specific question and am looking for an answer.

  33. Was there a Doctor (of physics) in the House?!?
    During the question and discussion portion of the House Science & Tech…Hearing on Climate Change held November 17th, some fleeting comments on atmospheric physics seemed alarming. Lindzen had a different understanding of responses around saturation vapour pressure than Cicerone (55:55-58:30). Lindzen also had an understanding of how molecules reacted to infrared radiation that differed from his colleagues and that left him wondering aloud if the testimony of his panel-mates had much worth (discussion leading up to 1:20:47).
    (N.B. The host of this blog was part of a subsequent panel, but I would have liked to have heard her thoughts.)

    • D Gemmell

      Lindzen was correct, but the smear at his colleagues was hardly justified. First of all, bringing up things like “dipole moments” is hardly necessary in a congressional testimony. In fact, it’s only the 3-atom gases that are vitally important as greenhouse gases on Earth, as the 2-atom molecules he mentioned are decidedly higher-order effects. When we think about the terrestrial greenhouse effect, in addition to clouds, nearly 100% of the effect can be approximated by thinking about water vapor, CO2, ozone, methane, and nitrous oxide. Humans have made some CFC contributions as well but it’s pretty small. O2 and N2, which is the bulk of the atmosphere, don’t have the properties to make a dipole moment.

      There’s always some caveats to the 1-sentence descriptions of what makes up greenhouse gases, and so forth. For instance, there are opacity features, especially for water vapor, that don’t come from ‘lines’ but from ‘continuum absorption’. There are also situations where the diatomic gases can in fact behave as greenhouse gases. This is particularly the case in very dense atmosphere like Titan, and is important as well on the gaseous planets in the outer solar system. Hydrogen is the predominant greenhouse gas in the context of the solar system. For solar-like stars, the dominant source of continuum opacity in a stellar spectrum (at optical wavelengths) comes from H- (Hydrogen with a second electron in the bound state)

      • I think if Dr. Lindzen matched smear-for-smear his schedule would be blocked out for the next several years.

  34. If you can give me quotes from their written testimony or oral testimony, I can respond, but not sure what you are referring to?

    • I could not find a transcript so I only offer, with apologies, this highly edited report:

      Inglis inquired about how much warming would result from a doubling of CO2, which in the course of their responses and explanations about CO2 and additional forcings from water you heard this:
      Cicerone (56:00-): “…disproportionate amount of evaporation increase as we warm a body of water…A fact of physics…which Dr. Lindzen denies…”

      Lindzen(56:41-): “…Clausius–Clapeyron relation…the saturation vapour pressure…for water as a function of temperature…[but] the atmosphere…is almost never saturated…[water bottle and cup prop example]”

      Cicerone (57:21-):”…I know the relation he’s speaking of [but] I don’t understand what he’s saying.”

      (One would have to watch the video to see Lindzen’s bottle and cup demo to see what confused Cicerone.)

      Lindzen stated that because the atmosphere isn’t always saturated the Clausius-Clapeyron relation was not applicable. Cicerone replied that an approximation could be used and despite Lindzen’s protestations one couldn’t hold back the vapour pressure of a liquid despite the saturation of what it is evaporating into. (58:30 the subject is dropped, but not necessarily resolved)


      Baird (1:19-): “How does a relatively small trace element impact…temperature?”
      Lindzen: “…there is no simple relation between the amount of a constituent and its ability to absorb radiation…CO2 is a significant absorber, I differ with my colleagues about the reason why, it’s the permanent dipole moment that is important…[other explanations about why CO2 absorbed radiation, involving numbers of atoms, had been offered previously by other panel members] I don’t know, it makes me wonder about their testimony. But still, it is possible for a trace gas to be important…”

  35. “Crimes against humanity”


  36. Judith, if you’d like to convince this skeptic of the fundamental correctness of AGW, first offer a proof that the adiabatic lapse rate is an equilibrium configuration with zero convective flux so that +/- deviations lead to +/- convective fluxes. My own math indicate that, for an ideal gas in a gravitational field, the configuration of maximum entropy is isothermal and the adiabatic profile requires a significant convective flux for its support.

    The assertion of ‘convective equilibrium’ is a convenient AGW ‘trick’ to keep convective flux at the zero level and leave radiative transport dominant. I’ve no quibbles with the basic Schwarzschild flux calculations. Of course, when I calculate the net radiative flux at the surface and the tropopause (MODTRAN), the latter’s twice the former. Radiative surface cooling to be sure, but how do I then restore this steady energy deficit without a steady convective flux and still claim the profile to be adiabatic? And then, if convection can balance a substantial radiative cooling flux, why should it not be able to also influence much smaller GHG increments?

    Richard Lindzen has repeatedly stated the essential question of skeptics is climate sensitivity. (I’d add a phrase on the effects of convection on sensitivity.) A clever youngster, once told there’s 250W/m2 coming out of a tropopause 65K cooler than the surface, came up with an Ohm’s Law sensitivity of 3.8W/m2/K and, then given a 5W/m2 forcing for CO2 doubling, deduced a temperature increase of 1.3K. Anyone think they’re smarter than a 5th grader?

    • Leonard Weinstein

      I started out thinking the equilibrium configuration with zero convective flux would be adiabatic, but found the approximation I was using was not correct. You are correct it would be isothermal based on maximum entropy.

      However, the day/night and latitude variation of solar heating, along with planetary rotation and evaporation and condensation, generate strong mixing. The solar heating is mostly absorbed at the ground (and seas), so heated air at ground level causes buoyancy, which causes large convection currents upward. These currents carry most of the energy upward. The result is that for the Earth, there is no chance for isothermal to dominate. In fact, if you initially had an adiabatic lapse rate and zero radiation and convection occur (and assuming rotation did not matter), it would take MILLIONS of years to convert the adiabatic profile to isothermal due to the very low thermal conductivity of air. It doesn’t take much mixing at all to maintain an adiabatic lapse rate. Your argument does not show anything about AGW.

      • Leonard,
        My own estimate of relaxation by thermal conductivity is about 10^4 years, but that assumes uniform temperature gradient configurations. Thermodynamics indicates nature likes to get things done as quickly as possible (max. entropy, etc.) and if that entails inhomogeneous configurations with regions of higher than average lapse rates, sobeit. Suppose we have an pure N2 atmosphere. Surface heating of the atmosphere will heat it and turbulence spreads it around, but once in the air, that energy has no where to go – no IR absorption, no IR emission – and the entire atmosphere warms to the surface temperature ca. 255K. Roy Spencer blogged on this point a year or so ago. The questions I ask: If we step the surface temperature, how long does it take for the change to reach the 10Km altitude? Is there a difference between plus and minus steps?

      • Quondam, there would be a huge difference. With a plus step, warm rising thermals would take the change up into the atmosphere very quickly, but with a minus step there would be no convection so it would take ages.

      • Leonard Weinstein

        Excuse my comment on millions.That was for Venus. Earth is more than an order of magnitude less. Also the exact level of residual gradient enters the calculation.

    • “My own math indicate that, for an ideal gas in a gravitational field, the configuration of maximum entropy is isothermal and the adiabatic profile requires a significant convective flux for its support.”

      I had a little trouble with this concept a while back also. How I resolved it was to think about what constituted a significant convective flux. I imagined smaller and smaller altitude differences until reaching molecular thermal motion range. Convective flux, though small, is still in effect at that level. Adiabatic temperature profile is self maintaining at the atmospheric molecular level. It is not necessary to have big updrafts and downdrafts.

      Anyway, that is how I have it worked out.

  37. Leonard Weinstein

    If an atmosphere was transmitting to sunlight but a near perfect optical absorber for Earth’s thermal radiation, but still had the same location of outgoing radiation to space as the Earth, the average ground temperature would be the same as present. The radiation up would exactly balance the back radiation at the surface and near surface, and all of the energy from solar heating would be transported up to radiate to space by convection only. This is the case of what it would be like with much more CO2 and other greenhouse gases, but note the critical assumption of where the radiation to space occurs.

    Note that back radiation and even amount of greenhouse gases did not matter. Only the LOCATION of outgoing radiation and the fact of the adiabatic lapse rate. THE BACK RADIATION DOES NO HEATING ON THE AVERAGE. It is a result of, not cause of the atmospheric greenhouse effect.

    The only important effect of adding some CO2 to the present Earth’s atmosphere is to raise the location of outgoing radiation a small amount.

    There may be feedback effects, but even they only change the effective outgoing height a small amount more or less.

    • So, the CO2 and water vapor feedback will warm. The more CO2, the more warming, the more feedback, until a new equilibrium is reached. But then, the oceans cool, which in turn cools the atmosphere, and also absorbs extra CO2. This reduces the CO2-related warming. The oceans have to be the proximate driver of air temperature because air has such a minuscule heat capacity compared to the ocean. So then the question becomes what determines the temperature of the various parts of the oceans. That probably would be the large ocean currents and solar SW radiation. So the currents cause delayed effects from changes in insolation. Then the question becomes what controls the incoming SW? GCRs? Clouds? Something else?

      • Your chain of logic broke apart after sentence 2. Why do the oceans have to cool?

        Nothing ‘controls’ the incoming SW except for the sun and the distance to the sun. The degree of planetary reflectivity (albedo) modulates the fraction of that incoming energy that is absorbed, but there’s no reason the albedo is going to act to stabilize the climate.

      • I was thinking about Joe Bastardi’s prediction that we would undergo an era of cooling due to negative AMO and PDO. If there are volumes of ocean water surfacing that are less warm than we have experienced lately, then we could cool due to the ocean cooling the air. (I really don’t see how the air could cool the oceans except over a very long span of time, do you?). Galactic cosmic rays could modulate cloud cover, which would the modulate the heat absorbed by the oceans. More or less heat absorbed would determine the water temperature in various ocean currents, which could submerge and re-emerge to heat or cool again.

      • Leonard Weinstein

        Your comment that “there’s no reason the albedo is going to act to stabilize the climate” is interesting. You are positive cloud formation due to some warming is not going to act as a negative feedback by changing albedo? What do you base this certainty on? Roy Spencer and Lindzen seem to think otherwise.

      • I’m glad your entire position is based on what two scientists think in the face of the whole community, but there’s just no evidence for it (and generally they think on the OLR side of the equation anyway, not the albedo). The papers they have certainly have not convinced anyone, and for good reason. There’s also some papers showing low clouds thinning in a warmer climate. Another example is the Clement paper, but mostly the observational and paleo record just doesn’t work with your implied sensitivity. It’s not a settled topic but the magic stabilizing ideas are just wishful thinking

      • Leonard Weinstein

        Take a look at this. His statement of CO2 rise is off (15% not 20%), but this is not important for the remainder:
        He also has a slide show at:

      • Judith – I have questions on this but they would be off topic. Could we have a thread on what we know/agree/disagree about clouds please?

      • This is coming eventually. Will have a series on climate sensitivity, then will dig into water vapor/lapse rate and cloud feedback processes. There will be a long series on clouds and interactions with aerosols. In the new year.

      • If we get an increase in cloud cover equivalent to permanently whitening the area of the US, we would get a one degree C cooling out of it (1% albedo increase). I think we would notice if this was happening, but I won’t say it is impossible, because increased haze did this to cause the, maybe, 0.4 C cooling up to the 70’s, and new areas are industrializing now. This effect could offset the 3-4 degree CO2 effect by maybe another half degree, in my view.

      • The problem you are having communicating the back radiation component comes from your using the idea that less cooling = warming,
        and less warming=cooling…

        With the addition of more CO2, the back radiation slows the exit of surface heat 24/7 like insulation in a house wall, during the night the minimum temp drops slower so is higher at sunrise, since the dew point is the factor to consider for the specific heat content, the less the near surface air cools over night the higher is the resultant dew point the next day (that can result on average) as more moisture is left in the air, (was not precipitated out as fog, dew, mist or rain, snow).

        The increase in latent heat does not show up on a thermometer, as the sun increases the temperature (thermal input) the next day the higher specific heat from what ever increase in dew point is left over (above the initial lower CO2 sample) allows the air to absorb more Jules for the same increase in temperature you had the day before (results in greater latent heat, and higher dew point), which allows more convection to form as the increased moisture content is + feedback for convective forcings.

        The slight increase in convection leads to a greater probability for more overall cloud cover than the initial conditions, which is an immediate (next day) feedback that increases the albedo, slowing the heating from less SW absorbed, just enough to keep the dew point and resultant specific heat content of the air mass regulated.

        If the night dew point increase causes less night time low level clouds and fogs, it allows more time for radiant heat to be lost before the sun rises, then increasing the SST in the day before clouds return in the afternoon.

        Colder ocean surface temperatures from up welling, do not cool the air above it, it only warms up more slowly, so the air ends up cooler from less heating, and slightly less outgoing radiation. IF the dew point is higher than the upwelling SST then condensation from the air mass will form on the surface, leaving the heat of condensation in the skin of slightly less saline concentration.

        The only way volumes of sea water down wells, is from the formation of sea ice, forcing out the extra salt that increases the density of the 2 to 4C degree (more saline) denser down flows that drive the polar cold ends of the global sea currents, Convection stops the less dense/saline skin of warmer (due to heat of condensation of water vapor out of the air) sea water from down welling.

        It is when the ocean surface temperature starts to climb above the dew point, that the vapor pressure drives moisture into the atmosphere carrying the increase in latent heat and some temperature increase.

        In the models there is no compensation for the ion content of the near surface air containing the sea spray moisture droplets, and aerosols from plant respiration, which can shift the balance in the distribution of particle size of droplets in the fogs, and clouds that form, the greater the mutual static repulsion between drops as the overall static/ion charge builds the smaller the droplet size, the more transparent the fog or cloud is to solar SW passage.

        When the solar wind density increases, the increased magnetic flux increases the homopolar generator fields in the geomagnetic fields, which ends up increasing the positive ion content along the equator. Increasing the resultant pole – to Equator + standing charge gradient on the Earth which reduces the cloud density in the mid latitudes, allowing more solar SW to reach the surface.

        By this mechanism the increased solar wind activity (with no or little change in TSI) can cause resultant warmer SST’s, driving short term increases in severe weather tornadoes and hurricanes. The North / South declinational tides of the moon shift the global atmospheric meridional flows in phase with the magnetic pole rotation of the sun and hence the polarity of the solar wind shifts in phase with the declination peaking around culmination, which can be shown to account for the increased production of tornadoes at 3 days either side of culmination.

        Ion production and discharge globally by homopolar generator processes give an ion assist additional to the thermal gradient across frontal boundaries that form in the mixed air masses that combine to form the tidal bulges, to drive the local precipitation rates above usual, generating narrow frontal boundaries with tornadoes, drechieos, flash floods, and other severe weather types.

      • Very interesting analysis Richard, and given support by Dst measurements of the horizontal magnetic field. The positive component correlates with solar wind speed, and the negative component correlates with solar activity.

  38. I’m just a layman lurker, but I get tired of the entire “CO2 absorbs infrared” bit. Of course that’s true. What’s important is the net effect, and that’s far more complex.

    For example, increased CO2 increases plant growth. Does it increase it enough to offset the slight warming effect of the CO2 as plants absorb more solar energy in the production of biomass? Does anybody care? As an erstwhile motorcyclist, I know the temperature difference between green fields and bare dirt can be pretty dramatic.

    • You get tired? Imagine how tired scientists get every time someone thinks they’re the first one these things have occurred to. No, Pops, nobody’s considered the net effect, except maybe (Solomon 2009), (Peng 2004, (Tao 2008), (Morgan 2007), (McCabe 2007), (Cai 2008)…

      • Solomon 2009: asserts irreversible changes due to increases in atmospheric CO2.
        Peng 2004: simulation models show decrease in rice yields due to increased night-time lows.
        Tao 2008: incorrect URL provided.
        Morgan 2007: changes in plant varieties due to warming.
        McCabe 2007: warming may create water supply shortages.
        Cai 2008: impact of rising temperatures on inflows in Darling-Murray Basin.
        At this point I begin to wonder what question you thought you were addressing with these references. So I’ll repeat: what is the net effect on global climate of increasing atmospheric CO2? Does it get hotter or cooler, or no change?
        I also get tired of chasing irrelevant references.

      • On what basis?

    • It looks like I’m not the only one who wondered about this – check out this paper from NASA…

      • David L. Hagen

        NASA quantifies Negative biomass feedback

        A new NASA computer modeling effort has found that additional growth of plants and trees in a world with doubled atmospheric carbon dioxide levels would create a new negative feedback – a cooling effect – in the Earth’s climate system that could work to reduce future global warming.

        The cooling effect would be -0.3 degrees Celsius (C) (-0.5 Fahrenheit (F)) globally and -0.6 degrees C (-1.1 F) over land, compared to simulations where the feedback was not included, said Lahouari Bounoua, of Goddard Space Flight Center, Greenbelt, Md.

        See also:
        Bounoua, L., G. J. Collatz, S. O. Los, P. J. Sellers, D. A. Dazlich, C. J. Tucker, D. A. Randall, 2000: Sensitivity of Climate to Changes in NDVI. J. Climate, 13, 2277–2292. doi: 10.1175/1520-0442(2000)0132.0.CO;2

        . . . The difference between the maximum and minimum vegetation scenarios resulted in a 46% increase in absorbed visible solar radiation and a similar increase in gross photosynthetic CO2 uptake on a global annual basis. . . .
        Important effects of increased vegetation on climate are:
        *a cooling of about 1.8 K in the northern latitudes during the growing season and a slight warming during the winter, which is primarily due to the masking of high albedo of snow by a denser canopy; and
        * a year-round cooling of 0.8 K in the Tropics.

  39. Please pardon my ignorance.
    My laymans understanding so far, (having read numerous articles and literally thousands of comments on both sides of the debate) is as follows..
    There are only 2 ways a greenhouse warms earth (earth includes it’s atmosphere, which is part of the system, only less dense than the surface).
    1-) Reduction of the atmospheric window, where radiation escapes directly from the surface to space.
    2-) GHGs delaying the escape of radiation to space due to the so called trapping or back radiation.

    So some questions.

    Has there been any observations that demonstrate the atmospheric window has reduced during this latest warming period?
    If so, by how much has it reduced and how does this reconcile with the measured rise in global Ts and the Keel Trenberth energy budget?

    Considering radiation travels at or about the speed of light, how long is the above mentioned delay due to back radiation in an atmosphere of a few kilometres. Seconds? minutes? weeks or years?
    If the answer to the above is any longer than hours, why do Ts plunge so dramatically during clear sky nights, and why is the ground always warmer than the atmosphere?

    As an addendum, do the radiative properties fo GHGs behave differently during daytime to nightime? If not, why would an EGHE manifest itself mainly in the rise of minimum Ts i.e. mainly at night? Why wouldn’t they “do their work” at all hours of the day?

    thankyou in advance

  40. I would ask the skeptics whether they agree with the following statement: A rotating sphere at the earth’s distance from the sun with an albedo of 0.3 will have a radiative equivalent temperature of 255 K. Yes or no?
    Hopefully, if they know basic physics they will say yes.
    OK, so why doesn’t the earth’s surface temperature average -18 C, but nearer +15 C?
    Explain this your own way.
    Now in reality, the radiative temperature of the earth is 255 K, but it comes from the atmosphere. Note that radiation can only come from an atmosphere with IR-active gases, such as H2O and CO2, not from pure N2 or O2 which are transparent to IR. On their own O2 and N2 would leave the surface to radiate at 255 K.
    The rest of the explanation is that since the earth has to radiate at 255 K, and there is an atmosphere helping to do that, and we know the atmosphere gets warmer as you go deeper (actually because of convection and compression), the surface will be warmer with this type of atmosphere than without.

    • Leonard Weinstein

      Jim D,
      You seem to have not been reading the comments. Several skeptics have discussed the atmospheric greenhouse effect, which is real, but the actual issue is the nature of the feedback, not if is there a CO2 effect. There are some skeptics that think there is not a greenhouse gas effect, but most who are scientists agree there is one. Locking in on outliers of either side of the issue is chasing straw men.

      • Unfortunately I read most of the 200 comments, but the item at the top and most of the links listed there were related to complete denial. I expect JC was going to have a separate thread on feedbacks once we get this and maybe other basic building blocks established first, and I will contribute on it when that time comes.

      • Jim D, OK, it can be hard to find the quality comments among the dross, but please read the comments by Leonard Weinstein and by Nullius in Verba. You do NOT need IR-absorbing gases. You just need an atmosphere. NV’s argument is in terms of convective lapse rates. If you prefer a radiative argument, think of the atmosphere ‘back-radiating’ to the earth (warm N2 radiates as much heat as warm CO2!), and the heat getting back into the atmosphere from the earth by thermal convection. In other words, think of the standard GHG argument but with the earth -> atmosphere heat transfer by convection instead of IR absorption.

      • So you look at radiation codes like MODTRAN (find the U Chicago site where you can play with it), and you think they are part of the big plot, despite these theories being developed decades ago? These adequately quantify the greenhouse effect. It is well understood at the molecular level. Why look for something else? First you have to show that MODTRAN is wrong despite matching spectroscopic measurements. Then you can come up with an alternative theory.

      • N2 doesn’t radiate anything. Where do you get that from?

    • At the expense of embarassing myself I’ll give it a shot.

      yes or no?….Yes
      Why 15C and not -18C?….Conduction???

      That was a very interesting question.
      I have one that is also (to me at least) just as interesting.

      A rotating sphere etc etc but with a non-GHG gaseous atmosphere. (Say N2 and O2 only) Given enough time, what will the temperature be?

      • ooops, just to clarify that question, “what will the Temperature of the atmosphere be?”

      • Latimer Alder

        The temperature will be ideal. No humans. No CO2. Paradise as envisioned by Hansen et al.

        Apart from the plants – notable by their absence. Or most any other form of life.

      • Haha ha Latimer, I like the humour.
        Still, it’d be nice if someone could help me with the answer. I assumed (yes assuming is dangerous and I barely have high school education) the physics involved would be much simpler than when GHGs are involved.

        Anyone willing to help with an answer would be much appreciated.

      • The global temperature will be constrained to be no greater than 255 K (with modern albedo). There’s been modeling experiments done that remove most of the greenhouse effect (like the Lacis et al. results) and the temperature drops even further than that.

      • Thnx CC. Your answer surprises me, -18C.
        How does the atmosphere lose it’s heat if it can’t radiate to space, at least nowhere near as efficiently as if it contained GHGs?
        (I probably shouldn’t be taking up time and space on a technical thread).
        I would have assumed that the GHGless atmosphere would accumulate heat until it reached the same T as the surface.
        I’ll see if I can access Lacis et al

      • Baa,

        To explore this further,
        You can see all of this by starting with the most basic energy balance equation:


        where S is the solar constant (roughly 1370 W/m2), a is the albedo (~0.3), 0.25 is the cross-section to surface area geometric ratio that must be accounted for (this will always be this value unless the object of interest is not spherical), σ is a constant, and T is the emission temperature. The left hand side is the incoming absorbed sunlight, and the right hand side of the equation is the outward power output (energy per unit time) per unit area. Solving for T yields ~255 K. Note that since we are dealing with no opacity, the emission temperature is the surface temperature.

        I’ll note that both the atmosphere and the surface become very cold in the limit of no infrared opacity. If the atmospheric opacity is very small, the ratio of the temperature at the top of the atmosphere (TOA) to the temperature at the bottom of the atmosphere (BOA) will be about one. As the atmospheric opacity increases, the atmospheric temperature gradient will increase, and TOA:BOA temperature ratio will decreases, going to zero in the limit as the opacity approaches infinity.

        As opacity increases, the emission temperature will stay the same (~255 K) except that it is no longer the surface temperature, but rather some temperature encountered high in the atmosphere. In the real world where we have to deal with varying degrees of opacity depending on the wavelength of interest, the primary source for low opacity (highly transparent) wavelengths (the so-called atmospheric window) is near the surface. The primary source of flux for high opacity (low transparency) wavelengths is in the high atmosphere, ranging from perhaps the middle troposphere near the wings of the spectral lines to the stratosphere at the center (at least for CO2, ozone works a bit different). In a simplified “grey” atmosphere where we sort of average over the the spectrum, this average emission height is about 5 km (call this height H) in the modern atmosphere. You can thus extrapolate down the adiabatic lapse rate (L) to achieve the surface temperature.

        T_s = T_e + (L*H)

        and so T_s > T_e in the presence of an IR absorbing atmosphere. Physically, the decreased flux from the opacity is causing a total decrease in the outward energy term of the planet. If there’s no greenhouse effect, this doesn’t happen and the surface must equilibriate with the only incoming energy term, the absorbed shortwave radiation.

      • Thankyou for so much detail, and I apologise for the late reply.
        ok, so my planet with a GHGless atmosphere would have a uniform T from ground to TOA of 255K?
        {You said above “If the atmospheric opacity is very small, the ratio of the temperature at the top of the atmosphere (TOA) to the temperature at the bottom of the atmosphere (BOA) will be about one.”}
        Whilst researching the constant (SB Constant) I came across this in Wiki…
        “With the average emissivity set to unity, the effective temperature of the Earth is:
        TE = 254.356 K or -18.8 °C.
        This is the temperature of the Earth if it radiated as a perfect black body in the infrared, ignoring greenhouse effects.”

        Which agrees with you. But then this, from the same paragraph…
        “If we wish to estimate what the temperature of the Earth would be if it had no atmosphere, then we could take the albedo and emissivity of the moon as a good estimate. The albedo and emissivity of the moon are about 0.1054 and 0.95 respectively, yielding an estimated temperature of about 1.36 °C.”

        Do I have this right Chris, a virtual earth with an atmosphere of N2 and O2 only yields T=255K, but one with no atmosphere at all yields T=274.36K? meaning an atmosphere of N2 and O2 is cooler?

        How can this be? Surely the N2 and O2 molecules receive some energy via conduction, with no way to emit that energy (non-GHG) therefore accumulating from the ground up until it’s T is uniform from BOA to TOA, and also receive some energy via direct sunlight (diurnal bulge?)

        Chris all this may be quite tedious for you, if so, I fully understand.

      • Ooops looks like I overdid the bolding (he says sheepishly) sorry JC

      • “but one with no atmosphere at all yields T=274.36K?”
        The second calc used a different albedo – the moon’s – which is lower. Also emissivity=0.95. Both changes make it warmer.

      • Baa Humbug,

        Not tedious at all. I like that you are phrasing your queries as questions, rather than bold assertions about the status of climatology and the morality of the scientists who study it.

        The albedo is an intrinsic property of the planet/moon (depending on cloud cover, land surface characteristics, and particles/molecules in the atmosphere itself)– it has nothing to do with the distance to the sun, so I’m not sure why we would assume the moon is a good analog to Earth’s no-GHE albedo parameters. It could be, but that would just be by coincidence. Albedo is also a function of the climate itself. Snow, ice, and clouds have humidity and temperature thresholds at which they form and dissipate, making them potential feedbacks to climate change, effects which are largely not operable on the moon.

        The atmosphere affects the planetary albedo in a variety of ways. First, you get direct reflection from aerosol and clouds. The albedo on Earth is dominated by cloud cover (as you’d see just by looking at a visible satellite image from space. The clouds are very reflective). Some ~5% of the planetary albedo is also due to Rayleigh scattering of air molecules. Then, there are also the affects that atmospheric opacity to downwelling shortwave radiation limits the amount of downwelling shortwave radiation reaching the surface and, finally, atmospheric opacity to shortwave radiation upwelling from the surface that limits the amount of shortwave radiation reflected by the surface that escapes to space (the globally averaged shortwave absorption in the atmosphere is about 20% of the incoming stellar flux).

        The land surface (especially snow and ice) are also very reflective, but the land surface contributes only a small fraction of the total planetary albedo. In fact, even in these ice-covered regions, the local albedo is generally dominated by clouds.

        In simulations of a greenhouse free atmosphere by Lacis (as well as other papers), the albedo of the planet actually increases substantially. This is, in part, due to the prospect of a runaway icehouse planet, with most or all of the oceans freezing over. This increases the total planetary albedo to over 30%, resulting in temperature even colder than the 255 K value above.

      • I’m sorry, it looks like I didn’t explain my thought process too clearly. p.s. I wear white T-Shirts in summer.

        My virtual planet with an amosphere of non-GHGs would necessarily be one without ice, vegetation etc A similar surface of rock and dust as per the Moon, hence I can see why Wiki uses the Moon as an example of albedo and emissivity properties.

        If Wiki is correct (with the moon essentially being the same distance from the sun as is Earth) then a T profile of 1.6DegC would be close to the mark, but probably a little high due to Rayleigh scattering you mentioned.

        So, being God, I’ve created a new Earth consisting of rock and dust, an atmosphere of N2 O2 (we’ll disregard Argon etc for this purpose).
        My Earth has a T of about…say 0DegC? uniform throughout the atmosphere and surface, meaning no wind, no weather.
        I’m now ready to add some GHGs to the atmosphere, but before I do, have I gone wrong somewhere along the way so far?

      • Baa: I don’t seem to be able to directly reply to your last comment, so hopefully you spot this.

        Your underlying premises are not correct. Certainly if the Earth were a bare rock, the albedo would probably be similar to the moon or Mercury in this regard, in terms of albedo. But the land surface can still be different. The albedo on Mars for example is more than double that of the moon, even with a thin atmosphere. Dustiness helps. The albedo of tundra, granite, limesone, desert sand surfaces, etc are all generally higher than ~0.1.

        But why does a greenhouse gas-less atmosphere imply no ice? The Earth is some 2/3 ocean, which have plenty of room to freeze over. The moon isn’t. So there’s no reason to take your GHE-free virtual Earth to be similar to that of the moon offhand, and as I stated before, the albedo doesn’t really care that there’s a similar distance to the sun between the two objects. What’s more, if we want to remain in the realm of some degree of realism instead of a complete virtual world, there will also be some hydrologic cycle at work if you took away the non-condensable GHG’s (CO2, CH4, etc) and clouds. You can’t tell the water vapor and clouds to go away in real life, even if you can shut them off in a model. In a wide variety of papers (Pierrehumbert, the Lacis paper, Voigt and Marotzke) all get a snowball in a CO2-less atmosphere, for example, and the albedo increases.

        There’s another point to be made too. For airless bodies like Mercury or the moon, the radiative balance equation S*(1-a)/4=σ*T^4, doesn’t hold a lot of meaning. This is just because the right hand side is non-linear and there’s very large diurnal temperature gradient (of several hundred K). In this case, talking about an ‘effective temperature’ doesn’t really give as much insight, as compared to a hemispheric temperature, or even the temperature at a specific point, depending on the problem at hand.

      • Thankyou Chris, I very much appreciate the time and effort you’ve put in to help me.
        I’ve learned a little more.

  41. Leonard Weinstein

    I read your writeup, and want you to consider the following. If the altitude of outgoing radiation is higher, the temperature of the gas at that higher altitude (and it is actually spread out, but we are both using an average) has to be the same as it was at lower altitude or the amount of outgoing radiation would be lower than absorbed solar radiation! The adiabatic lapse rate is a GRADIENT, not a level of temperature, and fixing the temperature at one location sets the levels. The balance of incoming and out going solar caused heating sets the temperature at the effective outgoing level (255 K). However, we are only talking about 150m or so increase in altitude for outgoing radiation. The same temperature at the higher altitude plus the lapse rate times the higher altitude is the cause of the higher ground temperature. An example of why this is so can be seen in an ideal model:
    If an atmosphere was transmitting the same amount of sunlight through an ideal atmosphere to the surface, but the atmosphere was a perfect optical absorber for Earth’s thermal outgoing radiation, but still had the same location of outgoing radiation to space as the present average location on Earth, the average ground temperature would be the same as present! The radiation up would exactly balance the back radiation at the surface and near surface, and all of the energy from solar heating would be transported up to the altitude, to radiate to space, by convection only. This is the case of what it would be like with much more CO2 and other greenhouse gases to completely fill the absorption window, and cut the absorption path to essentially zero. However, note the critical assumption of where the radiation to space occurs. I know this is an unreal idealized model, but correct with the assumptions, and shows the point I was making.

    Note that back radiation and even amount of greenhouse gases did not matter. Only the LOCATION of outgoing radiation and the fact of the adiabatic lapse rate (I did have to assume the same lapse rate, which is radiation independent. It is -g/Cp but modified by water vapor condensation). THE BACK RADIATION DOES NO HEATING ITSELF ON THE AVERAGE. It is a result of, not cause of the atmospheric greenhouse effect. You can look at the back radiation as a cause of heating, but that is confusing cause and effect.

    • We are considering interlinked phenomena. In many cases it is just semantics to argue on what is cause and what effect.

      For a surface the backradiation is one of the causes for its temperature. In that sense it is a cause.

      The change in the altitude of tropopause may be considered as a cause or explanation for the warming, but it is not the basic cause, but must be explained by something else. For the warming from an increase of CO2 the change in radiative energy transfer is a much more basic cause.

      My hope is that a proper way of handling the radiative transfer explaines the change of the altitude of tropopause (and also a simultaneous small change in the temperature at tropopause) so that one can finally come back to offering the radiative processes as the correct approximate explanation of the whole effect (excluding feedbacks) without the need of discussing the temperature profile of the atmosphere.

      • Leonard Weinstein

        Back radiation does no heating (on the average). The outgoing radiation is always equal or larger (on the average) than back radiation. Back radiation is an energy transfer, but it is the NET radiation transfer that heats, and the net radiation is always out (on the average). Convection plus the net radiation together transfer the absorbed solar energy to the TOA to radiate to space. That is the mechanism, and claiming the back radiation HEATS the ground is simply wrong.

      • Leonard,

        I made the statement for a surface, not an more extended layer (but similar issues apply to more extended layers as well).

        The energy balance determines the temperature of the surface. The back radiation is part of the incoming radiative energy coming to the surface and thus contributes to its temperature. The other components of the balance include emission of radiation by the surfaceas as well as conductive and convective transfer of heat.

        The incoming radiation including back radiation are components not affected directly by the temperarture of the surface, all other components are affected byt its temperature. Back radiation is affected by the temperature of the gas.

  42. “the temperature of the gas at that higher altitude (and it is actually spread out, but we are both using an average) has to be the same as it was at lower altitude or the amount of outgoing radiation would be lower than absorbed solar radiation!”
    No, it doesn’t have to be the same. There is the atmospheric window. As the emission altitude rises, the whole atmosphere and surface gets somewhat warmer. More energy escapes via the AW. Somewhat less escapes via GHG emission. So the temperature at the new emission point (after GHG increase) is higher at that altitude but lower than it was at the previous emission level.

    The surface warming that increases outward IR via the AW comes from down IR. This has increased because the emission level for down IR has been lowered.

    • Leonard Weinstein

      As greenhouse gas concentration increases, the atmospheric window DECREASES slightly (band broadening). Also the EFFECTIVE height of outgoing radiation is not based on a single height, but an average, consisting of broad areas of different height for each gas, and the atmospheric window. Obviously this is a gross simplification to call it a single height, but since both the reduced window and increased CO2 effective height would increase, my statement (as limited by the crude averaging) is essentially correct. Including all the complex details might cause some temperature change, but not much.

  43. Based on an earlier post by Nullius in Verba (Nov 30, 3:13 pm) and the following posts by him and myself, I try to formulate better how this non-atmospheric physicist understands the situation. I hope this helps in finding more widely understandable explanations.

    There are several related issues:
    1) Ultimately we need to know the relationship between the surface temperature and outgoing radiation to the space.
    2) The bulk of the radiation comes finally from the upper parts of the atmosphere as the troposphere is opaque to most infrared wavelengths.
    3) This is the basis for Nullius in Verbas statement that the radiating surface is high in the atmosphere.
    4) Some wavelengths penetrate better through troposphere. Therefore there is also direct radiation to the space from the surface of the earth and lower troposphere.
    5) The stratosphere is not convectively mixed. Thus the radiative energy transfer dominates in the stratosphere.
    6) The effective radiative temperature of the earth as seen from space is about -20 C. The earth is radiating less than it would at -20 C at wavelengths dominated by the highest troposphere and more at wavelengths penetrating well through the troposphere.
    7) The temperature of the tropopause is determined by the energy balance of the earth.
    8) The altitude of the tropopause and the temperature difference between the surface of the earth and the tropopause are interlinked through the lapse rate, which in turn is determined by the stability condition of the atmosphere and the physical properties of atmospheric gases.
    9) Additional CO2 influences the transfer of radiation at those wavelengts where CO2 has absorption, but the absorption is not fully saturated at all relevant distances. Thus it influences the direct radiation from the surface of the earth and the lower troposphere reducing its share in the total outgoing radiation. Stratospheric CO2 influences also the radiation from the top of troposphere and this is effective at wavelenghts where the absorption of CO2 is too strong for the radiation to penetrate far in the troposphere.
    10) Any explanation of global warming must specify through which mechanisms the changes in the absorption of infrared radiation influence the parameters of that basic explanation. If the altitude of the tropopause is an essential parameter, it must be linked to the changes in the radiative transfer, and furthermore, possible changes in the temperature of the tropopause and the lapse rate must be analysed.

    Explaining global warming may require going through the whole “basic greenhouse effect” with many complications, but it may also be possible to explain the warming as a change in the energy balance without a full description of the atmospheric temperature profile. I hope this more straightforward approach can be formulated in a way that may also be considered correct, while not comprehensive.

  44. The sunny face at point 8 is not intentional, but created automatically by the web software from 8 and ).

  45. Sun only provides the energy for heating the Earth. Energy is stored in the oceans and taken back and forth by ocean currents, and this process takes decades and even centauries. Ocean currents are not constant, and are affected by number of factors too. There also volcanoes, geo-tectonics and astronomy (Milankovic).
    Some of these come together in the North Atlantic as:

    and central Pacific as:
    Do not be mislead: part of the heat affecting today’s temperatures have arrived here from the sun, long before any of us were born.
    In long term (within Milankovic formula) the energy from the sun is probably nearly constant , it is just its distribution in time and space that varies, it is all down to the ‘mother Earth’ what she does with it.
    Humans and CO2 are nothing but a flee in the elephants tail.

    • The long term storage of solar energy in the ocean is exactly the reason why equilibrium based calculations won’t give us accurate sensitivity deductions from observation.

  46. Latimer Alder

    Would it be impolite to observe that after 24 hours and 250+ comments from some of the finest brains on the planet – and with the greatest knowledge of this subject available – that we are no nearer coming up with an explanation at any of the levels that Judith asked for:

    ‘… At an audience that has taken 1 year each of undergraduate physics and chemistry, plus calculus. Once we have something that is convincing at this level, we can work on how to communicate this to the interested public (i.e. those that hang out in the climate blogosphere). Willis Eschenbach’s help is needed in translating this for the WUWT crowd.’

    I started out on this one thinking ‘yeah OK – basic greenhouse effect – no probs with that – well understood stuff’. But now I’m less, not more, convinced.

    Surely it can’t be that difficult? To explain the absolute basic 100% foundation of everything about global warming because of CO2. The sine qua non. (And don’t call me Shirley :-) )

    • Nullias in Verba has the best so far IMO.

      • But Nullias in Verba is not explaining at all the basic first step: How does the CO2 cause the effect. Stating that it makes the fuzzy layer thicker is not a correct explanation. The correct explantion must involve the energy balance.

        The thicker fuzzy layer is not the cause, it is an effect of the change in the radiative energy balance from reduced direct radiation from the surface and the lower atmosphere.

    • I’m predicting that it will take another thread to get this focused. It has been illuminating to hear the skeptical arguments laid out and debated. It will take a little more cogitating to get where we want to be :)

      • steven mosher

        judith given the way skeptic “arguments” spin in a thousand different directions, it might make sense to do a single thread at a time.
        Like putting G&T to rest. and moving on to the others. It quite a struggle to
        get people to focus and follow an argument through. Perhaps a guest post from Lindzen, spencer etc on RTE.. or working engineers explaining how they use that physics.

      • tomorrow i am collecting some of the main arguments into a new thread, to try to get some focus. creating the new skeptics discussion thread at least sent the noise to a different location.

      • It would be good have a thread on each of the major “topics” that either completely disbelieve that GHG’s have a “warming effect” (vs greenhouse) as well as separate threads on those that show the degree of impact that GHG’s have vs other potential causes. It might then be really interesting to seeing “polling” , both of readers and your associates as to their acceptance of various positions. Just a thought

      • Steven

        Refocus please.

        The problem here is not that there are a zillion different sceptical arguments. It is that there seem to be a zillion different basic POVs about how the ‘greenhouse’ effect works.

        In other words, nobody seems to have a really convincing and defendable position of the basic physics. Some look at complex radiative transfer models, others at the ideal gas law, energy balances etc etc.

        But none of them have yet been able to come up with a reasonably simple explanation that we can all agree on about how and why it works.

        I don’t doubt that such an effect exists in some shape or form, but it is very disturbing that something so absolutely basic to the whole global warming argument is so poorly understood in detail.

        It is not really a tenable scientific position to say ‘we have studied all other factors in detail and can reject them all as being important factors in the climate change we believe we have data to confirm. So (by lack of anything else) it must all come down to this thing we call ‘The Greenhouse Effect’. But when we come to look in detail at the Greenhouse Effect, we find we don’t actually understand it at all. But we did understand all the other factors in absolutely enough detail to reject them as causes. Honest!’

        It is just not very convincing.

        Don’t blame sceptics for the very poor quality of the basic scientific understanding. The alarmists job is to prove their case. Our job is to point out flaws in the alarmist argument. And this is a big one.

    • Agreed.

      It just goes to show the intransigence with which some hold their preferred beliefs and the futility of trying to fashion a broadly understandable explanation when the responses go back and forth between a rejection on the basis of over-simplifying and rejection on the basis of being too complex to understand without doing an UG course in physics.

      As someone noted earlier – when do we hold individuals responsible for their own errorneous beliefs?

      • It just goes to show the intransigence with which some hold their preferred beliefs and the futility of trying to fashion a broadly understandable explanation when the responses go back and forth between a rejection on the basis of over-simplifying and rejection on the basis of being too complex to understand without doing an UG course in physics.

        There is never going to be one “correct” explanation which is at the right level for the public – some people are capable of grasping a more complicated explanation than others, some people are simply more inclined to take the time to try to understand the subject in greater detail. The most valuable thing for a layman such as myself is to have explanations at various levels available so I can educate myself to a level I feel comfortable with – but that’s because I find it interesting, not because I think I neccessarily need to have a more detailed explanation in order to understand the essential principles of the GH effect and how they relate to the case for AGW in general.

        As someone noted earlier – when do we hold individuals responsible for their own errorneous beliefs?

        Exactly – Judith’s statement

        “The fact that such papers are being written by scientists who take themselves seriously and are being published implies to me that scientists have done a poor job of explaining and making the case for warming of the planet by gases such as CO2. ”

        seems to me to be not just unduly harsh on herself and other scientists who accept that the GH effect is real, but also incredibly naive.

      • Tosh.

        The problem here is not that of ‘trying to fashion a broadly understandable explanation’. It is that nobody seems to have a defendable explanation at all.

        Do not blame sceptics for pointing out that just about everything we’ve heard on this thread so far has big holes in it. If the explanation of any theory doesn’t hold water and can’t be defended against simple criticism by non-experts, then the problem lies with the explanation, not with the nature of the critic!

        Focus your attention on the science, not on the sceptics. If you guys really can’t come up with a proper explanation of the greenhouse effect your case will take yet another damaging blow.

      • Latimer,

        The fact that no one on this thread has presented an explanation that is to your liking does not mean that no explanations available. Go and read one of the standard text books on the subject if you are really interested. Also read Science of Doom’s comment below (or visit his website – I’m sure there is something on there about it).

        It seems to me the experts here are doing just fine at defending the criticisms by non-experts. The problem is they can do so until they are blue in the face and people will still not change their minds.

    • Leonard Weinstein

      Nullius has clearly explained it. The explanation had to be simplified, because there are many local complexities (day/night, clouds, rotating planet, storage sited like the oceans, long period cycles, etc.), but his (and my) explanation is the basics, and correct. He is also correct saying that people on both sides of the issue get it basically wrong most of the time. When I read over many of the comments I feel I am in grade school, with people with short attention spans.

      • Well I’m glad that you have spoken and have self-certified that you are completely right. And I’m sorry if some of us are too ill-educated to still have a few questions that we’d like answered. And that you find this an inconvenience.

        Strangely enough I was under the impression that this was part of the purpose of this come up with an explanation of the greenhouse effect accessible to reasonably scientifically literate laymen who are not professional scientists.

        Obviously I was mistaken in this understanding … maybe I should go back to ‘grade school’ (whatever that is – is it an American expression?) and work harder on my attention span. Toodle pip

  47. I never see the heat generated in the earth’s core discussed as a parameter for climate modeling, or for accounting in energy equations. Is this because I’ve been unobservant, (and if so can anyone point me to a useful link?) or because the effect of the earth’s own supply of energy upon the atmosphere is generally agreed to be negligible?

    I keep coming up with a mental model of an equilibrium-seeking dialogue between two originators of energy – on the one hand a very great one, the sun, separated from the earth’s surface by a great distance, most of it nearly void and (pace the cosmic winds) capable of only one of the 3 forms of heat transfer – radiation – and on other hand a much lesser one, the earth’s core, separated from the earth’s fluid integuments by a smaller distance largely consisting of material capable of conducting and, in its molten phase, conveying heat, but not of radiative transfer.

    The “fluid integuments” – ocean and atmosphere, are capable of all three forms of heat transfer but, in the case of radiative transfer, frequency-specific “lacunae” impair this capability, interrupting the energy flux and causing it to reside as sensible heat. The complexity and nonlinearity of the resulting system is owed entirely to its fluid components, of which there are three – ocean, atmosphere and the molten portion of the earth. Yet I see occasional references to submarine vulcanism, a lot more to supermarine vulcanism, but not to other ways in which the earth’s core’s energy product is delivered to the surface. So my mental model has a missing pole, as it were.

    I’m sure this is a naive model, but I hope it is sufficiently coherent for me to ask the question above?

    • It’s not a bad question, just an energy source term that is many orders of magnitude smaller than the incoming stellar energy, and even an order of magnitude smaller than the forcing terms we talk about with respect to climate change. This isn’t true on the outer planets but once a solid crust forms, the flow of heat from the interior of the Earth to the surface is really insignificant (heat diffuses very slowly through solid rock). For example, this source puts it at ~0.075 W/m2. Pollack et al 1993 did a study of this over about 20,000 sites and found a similar value. Actually, even Fourier came to this realization in his early work.

  48. Consider the polar night, no sunlight. Conditions become statically stable (little to no turbulence, absolutely no convenction). Radiation and horizontal advection are the only things operating. The temperature profile is much closer to isothermal, no sign of dry or saturated adiabatic lapse rates.

    See Curry 1983 (this was actually part of my Ph.D. thesis) to understand how the atmospheric temperature profile and the surface temperature evolve under these conditions. In fact i ignore horizontal advection, so this is just pure infrared radiation. It would be really interesting to rerun these experiments using doubled CO2 (which I unfortunately don’t have the time to do.)

    For some observed wintertime profiles in the arctic, go to the University of Wyoming atmospheric soundings site, click on PABR (Barrow, Alaska) (for type of plot I suggest GIF: Stuve), the current surface temperature is about -25C, and the tropospheric temperature is about -55C. This is a relatively boring wintertime sounding (and it isn’t real cold yet). Click on YRB, Resolute, you see the temperature profile at 800 mb is warmer than the surface, but overall close to isothermal. Check out Cambridge Bay (YCB), the temperature at 650 mb is the same as the surface temperature. Note, these are early winter temperature profiles, gets much more interesting in January (and even more interesting in Siberia). Often the vertical profiles get very jagged and irregular, arising from horizontal advection and residuals from previous cloud layers, but these profiles are pretty smooth and look to be dominated by IR.

    So these cases represent pure greenhouse effect (infrared radiative transfer) in action.

    • “It would be really interesting to rerun these experiments using doubled CO2 (which I unfortunately don’t have the time to do.)”

      Well there might be people, who find the time. It looks like stuff, Fraedrich and Lucarini and others in their vicinity should be interested in.

    • Leonard Weinstein

      If you wish to see the effect of CO2 increase in the near polar cases, be sure to use the local insolation rather than the global average. In winter I suspect the Solar insolation is so low (near zero at the poles) that the entire atmosphere and ground would be much colder unless horizontal advection replaced the lost energy. An isothermal (or even inverted) lapse rate would mainly be due to lost ground radiation through the long wave “window”, without solar replacement heating

    • Judith

      Your link to the University of Wyoming atmospheric sounding page site is broken.

      I was able to track down their site. I wrote this post about how to generate your own DIY temperature sounding profile for citizen scientist.

  49. Apologies for not having read all the comments on this, so this may have been said already :-
    Judith, you say “these arguments refuting atmospheric warming by CO2 are being made by scientists that take themselves seriously on this issue” and things like “The G&T paper has been rebutted by Halpern“.

    You talk about a rebuttal as if it disproved the paper it rebuts. Well, the way I understand science, it doesn’t. It is perfectly normal for scientists to put up opposing positions, and for the disagreement to continue for years. How do these disagreements get resolved? In the end only the actual science – ie. the evidence – can do it. You have to keep going back to the underlying science until incontrovertible evidence has built up.

    And even then new evidence can show up that overturns it again…..

  50. I don’t take issue with mainstream physics on the basic effect, understanding it roughly, enough to sort of “park” there, still ready to move away if needed.

    To the extent I see it as provisional, it’s more with the on-lookers curiosity as to what modellers may need to do in order to accomodate proxies/geo-sciences and solar-cosmos-science empiricism if necessary.

    Still I’m not sure any of these known listed physics-maverick opinions would be vindicated necessarily by whatever that might turn out to be. Parts of someone’s perhaps. Also not about basic-physics-explanations as being key to what overall listening to skeptics should amount to. But this I get it will come later.

    There is even still some confusion around about the so common acronym AGW, whether it means any part or most of the modern warming coming from raised CO2.

  51. Tallbloke, I am trying to think of other ways to explain the ocean heating issue.

    Take a single column ocean mixed layer model (I recommend Clayson’s model that includes the details of the skin layer also)

    Force the model with observed solar radiation; surface air (10 m) wind, air temperature, humidity (the model calculates the sensible and latent heat flux). Then run two experiments where you use the observe downwelling infrared radiation flux, and then you perturb the downwelling IR flux for more and less CO2.

    If I am correct, then the surface temperature and upper ocean heat content will warm/cool by adding/subracting CO2 (more/less downwelling IR). If you are correct, the surface temperature will remain unchanged since the IR heating simply causes increased evaporation and latent heat flux. Again, I do not have time to do this (I am already way overstretched by this blog), but I am highly confident of the answer based upon all of the simulations using the mixed layer model that Clayson did in her Ph.D. thesis (I am her advisor). I realize that this is an appeal to experience and inside info, but all I have time for. But this is testable.

    Re your idea that that increased surface heating results in increased evaporative/latent heat flux. The latent heat flux is driven by surface wind and the difference between atmospheric surface specific humidity and saturation specific humidity of the ocean surface (which is a function of the skin surface temperature). There is no particularly direct relationship between surface temperature and latent heat flux, and latent heat flux is driven more by wind and atmospheric humidity. Take a look at some global climatologies of ocean surface latent heat flux.

    Click to access grodsky.pdf

    The largest values are actually over the western boundary currents like the Gulf Stream. Instantaneous (daily) maps show that storms dominate the latent heat fluxes.

    So here are some ways to test your ideas. Based upon my background knowledge (which is quite extensive in this particular area), I don’t think your idea on this is correct. Like I said previously, unfortunately I don’t have time to do the calculations.

    • I have a couple of questions on that.
      1. Do you remember what value or values of the ratio of ocean heating by SW vs LW?
      2. You mentioned models. Has any of that been confirmed by ARGO floats or other real measurements? And on ARGO floats, I keep reading that the raw data isn’t published. Is that true and, if so, why?

      • The Kantha and Clayson ocean mixed layer model has been validated extensively by observations, including research quality observations conducted during ship campaigns. It is the best one on the “market.”

        Re the ratio of SW to LW, that depends on the location. The KC model has been applied extensively in tropical latitudes (high SW) and high (north) latitudes (low SW).

        If you want more documentation on this, let me know and I will provide further refs.

        Regarding the ARGO floats, I don’t know anything about the details of this situation. I agree that the raw data should be made available, along with clear documentation of the processing that went into the processed version.

      • I can’t get to it until the weekend, but I would be interested in more information on the model. I can’t afford to get papers behind paywalls, so it that is the case, just say so and I’ll move along.

      • Oh my, everything is behind paywall. Kantha and Clayson have written two textbooks. The only paper not behind paywall is one on my web site:

        Click to access Webster_JC9.pdf

        Anyone who has a serious interest in these papers (or any others that i mention) and you don’t have free access to them, send me an email and i will email a copy to you.

      • Re the ratio of SW to LW, that depends on the location. The KC model has been applied extensively in tropical latitudes (high SW) and high (north) latitudes (low SW).

        Seamen tell us vast areas of ocean around the tropics (about 8deg north and south from memory) are called “the Doldrums” because the seas are usually very light with little or no wind.
        This would confirm the KC model would it not?

    • Judith, thanks for this more detailed response.

      I’m surprised the ‘experiment’ you propose hasn’t been done already. Would it be appropriate to contact Carol Anne Clayson to ask?

      My knowledge and ideas are evolving as I deepen my study into oceanology. Of primary interest to me is that the reduction in humidity since 1948 as measured by radiosonde balloons may well have offset any increase in radiative flux from co2. Miskolczi’s work with HARTCODE, leaving aside his ‘saturated greenhouse’ theory confirms this. This renders the co2 perturbation moot in the real atmosphere it seems to me, and I don’t understand why the radiosonde measurements are ignored by mainstream climate science. They can’t be all that bad, or the interesting correlation betwen specific humidity at the 300mb level and solar activity I discovered wouldn’t have such a good fit.

      The question then becomes, “what caused the warming?”. The obvious answer is the Sun, and the reduced albedo empirically measured by the ISCCP cloud project. It is much easier for the sun’s shortwave radiation to warm the ocean than it is for downwelling longwave radiation.

      At night, when there is no shortwave, the skin temperature is 0.3C lower than the subsurface according to theory. (See Science of Doom’s latest post in comments ) Doesn’t this indicate which way the flow of energy is going? He quotes extensively from Kawai and Wada 2007.

      • Tallbloke, think of this, get rid of all the CO2, what will happen to the ocean temperature then? will it stay the same, with reduced surface evaporation? Nope, it will cool.

        Re the skin temperature, i discussed this at length in my original reply to you. I agree that running some experiments with kantha-clayson ocean mixed layer model would help demonstrate this to you, but clayson is swamped and we pretty much already know the answer based the sensitivity studies that have already been conducted using the model.

        Explaining the warming in the ocean for the past 50 years is one issue; there could be multiple causes. Saying that adding more CO2 to the atmosphere won’t warm the ocean is altogether another issue. So it seems that you are conflating two issues here.

      • Judith,
        thanks for the followup reply. The effect of co2 in the atmosphere is logarithmic. Remove all of it, and it will for sure affect the altitude at which the majority of radiation to space occurs. However, what I have been saying is that the increased ‘back radiation’ from post industrialisation co2 level increase isn’t going to cause the ocean to cool more slowly to any significant degree. I’m not saying there will be zero effect, I’m saying it won’t account for the 0.15C the top 1000m of the ocean must have warmed in the 1993-2003 decade to account for the steric component of sea level rise measured by satellite altimetry after the estimated melt volume and other minor factors are subtracted out. The figure I got for that was more in accord with the sort of ocean heat content figures Levitus et al were proposing in 2000 from XBT readings. The figures from Levitus et al 2007 and other oceanography teams are inconsistent with the satellite altimetry. I suspect they may have been fudged to bring the surface forcing into line with the 1.7W/m^2 co2 forcing promulgated by the IPCC and to sideline the solar contribution. Maybe you can shed some light on the issue?

  52. Judith,
    I have noticed not a single person has addressed the oceans salinity changes.
    This is not caused by AGW but by pressure as the changes are on the oceans surface.
    It would effect how much solar radiation penetration in the oceans and change the surface reflecting of solar energy.

    Through known experimentation, this process is impossible through the evaporation process as ALL the salt would be effect and not just the surface salt.

  53. I am having trouble with the “nesting” of replies. PDA on November 30, 2010 at 8:32 pm | writes “Likewise, the effect on mean global temperatures of a doubling in carbon dioxide concentrations in the atmosphere is a physical property that cannot be measured. It can, however, be estimated. I’d say that the accuracy is roughly commensurate with that of geology at the same stage of development.”

    Fair enough. Would you give me the logic as to how you arrive at the accuray with which the estimate of rise of global temperatures as a result of doubling CO2, without feedbacks, has been made. If you have no experimental data, how on earth can you estimate how accurate the number is? How do you know that the way the estimate has been made does not have unknown factors which make the estimate just plain wrong? I am sorry, but William Thompson cannot be excluded from this discussion.

    • The real atmosphere has feedbacks. The concept of otherwise real atmosphere but no feedbacks is an artificial construct, which can be defined in many different ways by choosing which physical effects are included as feedbacks and which in the no-feedback part.

      Because it is an artifical concept its users can choose the definition. It is defined in such a way that calculation is based on known physical principles well understood and reliable. All the significant uncertainties are put into the feedbacks.

      This approach is perfectly legitimate and may be useful. This chain has indicated clearly that there are many different ways of interpreting the same knowledge on this no-feedback atmosphere. I am not the only one involved in an argument while possibly agreeing fully on the final result of the calculation. In these cases the argument concerns the way of interpreting the logic of the calculation: What is cause and what effect or how the physical effects should be presented to make them understandable to as many as possible?

      • Pekka Pirila writes “Because it is an artifical concept its users can choose the definition. It is defined in such a way that calculation is based on known physical principles well understood and reliable. All the significant uncertainties are put into the feedbacks.”

        I have never read such a load of scientific nonsense in my life. I thought this was Judith Curry’s blog where we talked proper science.

  54. Judith,
    If CO2 is the cause of the “greenhouse effect”, why are we currently having record breaking low temperatures in Canada and Europe?

    • Joe – ‘cos measurements of temperature over small time periods (days, months, years) are weather, not climate. Climate trends should be looked at over at least 30 year periods

      • Louise,
        If the planet was warming, then we should NOT be having any record lows at all.

      • Joe,

        That’s a complete misunderstanding of AGW.

        There is no basis for your assertion. If you can find any research suggesting this, than please share it.

      • Joe – imagine three groups of children:

        100 6 year olds
        100 7 year olds
        100 14 year olds

        (all of these children share the same day of the year they were born).

        It is very likely that the average height of the six year olds is less than that of the seven year olds but it is also very likely that some of the six year olds will be taller than some of the seven year olds. However, it is also very unlikely that there will be any six year olds that are taller than any of the fourteen year olds (but not impossible).

        In the same way, an occasional record low (and I’m not aware of any recently) does not disprove global warming.

      • Or alternativelyl – we know that the population is getting taller, therefore there should be no very short individuals?

      • Joe,
        The high/low distraction does not prove or disprove AGW.

      • Including time period of thirty years.

    • The warming claimed for CO2 is way less than 1 deg C over the whole 20thC. Record lows and highs are at individual points, where temperatures go up and down by many degrees from year to year. Over a long period of time, one could expect (if the AGW hypothesis is correct) the frequency of record low temperatures to decline slightly, and of record high temperatures to increase slightly. Regardless, both will likely continue to occur for many years yet.

      And talking of “long period of time”, don’t take the 30-year suggestion seriously. Given that the Pacific oscillation (PDO) has been taking about 60 years for a complete cycle, and has taken 100+ years in the past, a 30-year period is totally inadequate for most climate considerations. In fact, it has the potential to be the most misleading period that you could possibly select (witness the last 30 years of the 20thC).

  55. If I understand it correcly the basic explanation is that incoming short wavelength radiation penetrates the CO2 in the atmosphere while outgoing long wavelength radiation gets blocked by the CO2.

    However clouds (and water vapour) also have a significant affect on both the incoming and the outgoing radiation.

    Regarding ModelE climate experiment simulations I have read that ; “there is really nothing that is being assumed about cloud and water vapor feedbacks, other than clouds and water vapor behave according to established physics.” Andy Lacis

    I’m sorry to go all technical but can a turbulence model be described as established physics? Can an empiricial model for wall heat transfer be described as established physics? I have a strong feeling that all the elements understood within the term “established physics” are in fact subject to large amounts of uncertainty.

    I think it is fair to say that cloud behaviour is certainly not fully understood and is linked to many things including turbulence.

    Because turbulence is involved does this mean that all cloud effects can disregarded as merely fluctuations around a mean value set by the CO2 concentration? This seems a somewhat gross oversimplification.

    “Tennekes, more than any other individual, challenged the models that climate scientists were constructing, saying models could never replicate the complexity of the real world. ”

  56. I know this reply isn’t specifically scientific but I do feel that it relates directly to the premise of this post ie: that you feel it is important to explain the mechanics of the so-called “Greenhouse Effect” to as wide an audience as possible and to explain the confidence in said mechanics in the face of alternative explanations.
    Honestly, I think the “Greenhouse Effect” is a premise that is (almost) universally accepted by science and layman alike. As one of the latter – at least in terms of science – I’m aware of the ‘nuance’ that seems to exist between atmospheric physicists and Engineers – especially in relation to the 2nd Law of Thermodynamics – on this subject, but philosophically this is not an argument that is even necessary to the debate concerning the the main area of contention in the AGW debate, even amongst scientists, I would contend.

    It is possible, for instance, to argue that space time is essentially “flat” and that “curvature”, even spheres, whilst perceptually the way we interpret our universe are perhaps a more of an interpretational anomaly we see and measure simply because of our limited three dimensional senses. However interesting these concepts are – and I think they’re fascinating! – I, along with most people, am quite happy to discuss the future of space exploration, accepting that the World – and other celestial bodies are spherical. Every logical argument requires a premise of some sort!

    In a similar way, it is quite possible – and perfectly acceptable I believe – for any relatively informed and critically equipped person to both understand and discuss the evidence for positive feedback mechanisms within the climate system, or the possibility of external cosmic radiation affecting reflective cloud cover, or albedo, or enso, or the effects of phytoplankton on CO2, or any other variable within our climate which may have a bearing on the relative “temperature” of our planet…..without any deeper understanding of the primary “Greenhouse Effect” than the same broad acceptance that the work of Galileo enjoys within Astronomy. The fact that a lack of intimate understanding of the primary physics of CO2 doesn’t preclude an informed opinion within this debate is often – quite deliberately in my opinion – obfuscated by those who try to exclude those without a physics degree from the discussion.
    In short, I’m quite happy to accept the premise of the “Greenhouse Effect” and move on.

  57. @Saaad

    ‘I think the “Greenhouse Effect” is a premise that is (almost) universally accepted by science and layman alike’

    Prior to this thread, even I, buried deep in my lair at Sceptic Central (*), would probably have agreed with you.

    But the more people begin to argue about it, the less it seems is actually understood. I don’t believe in ‘consensus’ as being a valid way of making scientific progress, but I am aghast that 100 years after Arrhenius the mechanisms of the effect can still not be clearly and unambiguously described.

    And yet it is the absolute bedrock of all ‘warming theory’. It is the only game in town. All other things have been eliminated so the only possible thing that can explain warming is the ‘greenhouse’ effect because of CO2. Or so we have been told for over thirty years. TINA – There Is No Alternative.

    One of the common errors causing planes to crash is that when an in-flight emergency occurs the crew get too busy doing their own little bit of emergency handling (eg radio, engines, navigation etc) that they are distracted. The crash does not occur because of the emergency but because everybody forgot to fly the the d…d aircraft.

    Can it be that for the last hundred years, belief in the greenhouse theory has been so prevalent that everyone has taken it as read. But nobody was actually flying it? It begins to look a little like it.

    (*) No – there isn’t really a place called Sceptic Central, and I am not buried in it. It is a figure of speech. I am actually in Surrey in my study watching the snow come down and worrying about how to pay (not be paid by) Big Gas when the winter bill comes through the door. Do not leap to unwarranted conclusions.

    • I take your point Latimer. I was not meaning to imply blind acceptance of the premise of GHG theory as per Arrhenius. Rather, I was suggesting that to have a meaningful debate about the current perceived points of “uncertainty” within the science at the policy end of things, we must at least allow for some sort of agreed premise from which to proceed. Despite the many threads of discussion above, I still think Arrhenius is a sound first principle, even accepting your caveats.

    • Between this thread and the post by climatologist Dr. Nielsen-Gammon over at his Climate Abyss blog, my acceptance of the greenhouse effect is substantially reduced as well. It is also interesting that, like the melting glaciers and the very name of the crisis, when pressed hard by skeptics, it turns out that things climate have been mis-named, mis-represented or over hyped. This list is growing, is only growing under skeptical pressure, and is significant: it deals with the basic claims of the AGW community.
      What else needs to be closely scrutinized?

      • Steven Mosher

        hunter his post says nothing. I’d read some more basic texts first.

      • Steven,
        I have come to respect Dr. N-G quite a bit over the two years or so I have been reading him.
        I am not certain how you can say he said ‘nothing’. The two posts on what he calls the Tyndall effect seem like good primers and help get past the (mis)use of greenhouse.

  58. Alan Siddons writes in ‘Slaying the Dragon’ that the sun nevers sets in the climate models. Is this true?

    • ??? each location has the appropriate diurnal cycle, but the sun is always shining somewhere on earth.

      • That is true, but,

        What Alan has claimed in the book is: If we take a point source (sun) and shine it on the sphere (earth), half of it is recieving energy at a given point. This energy is simply divided by 4 and spread out over the entire sphere (the earth) in the models.

        To quote:
        “It is important to understand that radiant energy models don’t deal with sun and earth conditions as they actually exist. If a somewhat realistic model were used, the earth would naturally be hottest at the noon equator, coldest at the poles, but beyond that what – wouldn’t it be close to absolute zero on the shadow side?

        Such a problem is hard to solve, especially considering that the earth also rotates, thereby adding the complication of exposure duration vs heat-retention. Modelers therefore find it much easier to avoid these difficulties by imagining that sunlight has equal strength all over the planet. They do this by diminishing sunlight’s power to a quarter of its actual value.

        342 is what a modeler takes as the energy impingin on every square meter of the planet at once.

        All at once. Keep that in mind. Like the summer sun in the Arctic, a modeler’s sun never sets”

      • This is only a small excerpt from a sample chapter. Siddons develops his concepts further.

        I just wanted clarification on this little point about radiant energy models.

      • Ohmigod! This insight by Siddons will pretty much mean that the coffin of CAGW no longer even exists – it’s nothing but nails now. Devastating.

        How was this egregious blunder overlooked for so long?

        I’m convinced. “Slay the Dragon” is now on my must read list.

      • Sounds like he read the blogs on how to calculate T=255K using the solar constant divided by 4 and got confused about what a climate model means versus back of the envelope calculations. I don’t think he even knows why you’d have to divide by 4.

      • This is completely incorrect. Siddons is talking about simple energy balance climate models (the stuff where you estimate the equivalent black body temperature). This has nothing to do with how all this is treated in numerical global climate models.

      • Like I said, Dr C, Siddons does not tell that this is how the sun in ‘modeled’ in GCMs. (only in ‘radiant energy models’).

        The chapter title is “The weakness of a constant irradiance model’. He cites Kiehl-Trenberth as an example of this kind of model and points out one of its weakness, derived starting from the passage quoted above.

      • So,
        Is this how the earth is heated in radiant energy models? Is Alan Siddons correct in saying this?

      • I’m not an expert, but ….
        “Earth is heated by a sun that never sets in radiant energy models …” sounds like a bad and erroneous way of stating that K&T energy budget diagrams try to match the average incoming energy to the average outgoing energy at equilibrium and that the average incoming heat flux amounts to a quarter of the solar constant. In any case, the phraseology of somebody who states that the night side of the planet should be at absolute zero is not worth analyzing at depth.

      • This is quite interesting.

        You havent even read what Siddons wrote and yet you, and others too, have managed to say funny things.

      • “Tyndall needed no equations, but only simple logic, to see what many since him overlooked: it is at night that the gases are most important in blocking heat radiation from escape, so it is night-time temperatures that the greenhouse effect raises the most.”


      • Shub,
        Nobody is arguing that the greenhouse effect raises night temperatures – it is the reason why day/night swings on the moon are 250 degrees and much smaller on earth, I don’t know why you are posting this. But one can still look at the heat flux averaged over day and night and argue for it to be conserved. Let’s put this another way. Assume that you have a resistor in parallel with a capacitor. Let’s assume that you have a current source that pulses on and off periodically with exactly the same “ON” and “OFF” time over each period. As you increase the value of the resistor, the voltage swings on the capacitor become smaller. But if you were to average the incoming current, it would be exactly equal to the mean voltage across the capacitor divided by the value of the resistor. In other words, the average is conserved whether or not you assume that the incoming current source was a pulsing source or whether it was a current source that was always “ON” with a value equal to the average current over the averaging period. For the earth system, the sun’s heat flux is equivalent to the current source for the electrical system.

      • Tyndall followed with rich Victorian prose, arguing that water vapor “is a blanket more necessary to the vegetable life of England than clothing is to man. Remove for a single summer-night the aqueous vapour from the air… and the sun would rise upon an island held fast in the iron grip of frost.”


        In summa,
        Please read what Siddons has to say, instead of just going after the excerpts I quoted, quite bereft of their context.

      • Is this meant serious? I honestly can’t tell. Do you honestly believe this?

      • Leonard Weinstein

        But Curryja,
        Haven’t you herd of people being told to stick their head where the Sun doesn’t shine? (Please excuse, I couldn’t help this).

    • Steven Mosher

      Not sure what he means by the sun not setting.

      SUBROUTINE DAILY(end_of_day) 3,9
      !@sum DAILY performs daily tasks at end-of-day and maybe at (re)starts
      !@auth Original Development Team
      !@ver 1.0
      !@calls constant:orbit, calc_ampk, getdte
      USE MODEL_COM, only : im,jm,lm,ls1,ptop,psf,p,q
      * ,itime,itimei,iyear1,nday,jdpery,jdendofm
      * ,jyear,jmon,jday,jdate,jhour,aMON,aMONTH,ftype
      USE GEOM, only : areag,dxyp,imaxj
      USE DYNAMICS, only : byAM
      USE RADPAR, only : ghgam,ghgyr2,ghgyr1
      USE RADNCB, only : RSDIST,COSD,SIND, dh2o,H2ObyCH4,ghg_yr,
      * omegt,obliq,eccn

      USE DAGCOM, only : aj,j_h2och4
      INTEGER i,j,l,iy
      LOGICAL, INTENT(IN) :: end_of_day

      C**** Tasks to be done at end of day and at each start or restart
      call getdte(Itime,Nday,iyear1,Jyear,Jmon,Jday,Jdate,Jhour,amon)

      C**** This is for noon (GMT) for new day.
      * SIND,COSD,LAM)

  59. Steven Mosher


    This is a good resource:

    Click to access PhysMetLectNotes.pdf

    a fairly lucid description of the “greenhouse” effect.

    Te is known as the effective emission temperature. It is determined solely by the insolation and the planetary albedo. On Earth, Te is much colder than the observed global-mean surface temperature of 15◦ C or 288 K. The difference must be due to the atmosphere. The warming effect of the atmosphere, known as the greenhouse effect, is best understood as
    follows. The atmosphere is opaque in the infrared, which means that the mean emission level is lifted off the ground. The mean temperature at the emission level (i.e. the mean brightness temperature) must be Te in order for emission to match absorbed insolation. But the atmosphere has a positive lapse rate, and so the temperature at the ground must be
    greater than Te .

    In general, an atmosphere must satisfy 2 conditions in order to provide a greenhouse effect: it must absorb radiation radiation in the spectral range associated with black body radiation at temperature Te , and it must have a positive lapse rate. An atmosphere with a negative
    lapse rate (temperature increasing with height) will have a surface temperature colder than Te . This in known as the anti-greenhouse effect, and actually occurs on Earth in the polar regions during winter, as evidenced by Fig. 5.17b.

    Basically, the very first step is to get people to recognize and accpet that the earth is warmer than it would be with no atmosphere.

    • Leonard Weinstein

      Did you not read the very clear discussion by Nullius in Verba Nov 30 at 3:13 pm, and my similar but less well stated version on Nov 30 at 6:30 pm (this is out of order as it was a reply). I also gave links to more detailed discussions (including mine) at Nov 30 10:46 am.

      These all clearly state what you are trying to say.

  60. Dr Curry,

    Thank you again for putting up with this blog. While obviously a burden to keep up, it is a wonderful resource for those of us who are trying to get a better grasp of the relevant factors, It is a public service of the best kind.

    One housekeeping item.
    The very wide margins that are used in this blog make it more difficult to follow the thread of a discussion. This topic for instance when cut and pasted into Word takes 123 pages. Would it be possible to change the format to something akin to that used in Climate Audit, so that the full width of the page is employed?

  61. For what it’s worth, I’ve got my illustrated explanation of the Tyndall gas effect (more generally sort of known as the greenhouse effect) written up and posted. Since I divided my time today between helping students and writing this entry, some of this may be redundant, but oh well.

    • i like it (unfortunately i’m not your target audience), lets see what the reaction is.

    • I’m not sure if I’m a “target,” but I like the “Tyndall gas effect” better than greenhouse gas or greenhouse effect.

    • I hesitate to consider myself a ‘target’ for Dr. N=G, since he lives in Texas and I have questioned him on more than a few occasions. ;^) But his two posts on what should now be called the ‘Tyndall Gas Effect’ are very good primers and clear up the difference between what is happening in the atmosphere and a greenhouse. And since I am probably a good example of a non-scientist interested in climate and other science issues, I guess I am a target. Should I duck?

  62. Peter Milford

    You say :- “However, whether atmospheric gases such as CO2 (and H20, CH4, and others) warm the planet is not an issue where skepticism is plausible.”
    It seems to me that, within certain limits, skepticism of this statement is perfectly plausible. I tend to agree that a doubling of CO2 would likely cause a small increase in temperature, if all else is perfectly unchanged. But, a doubling of CO2 should cause other changes, relating to plant growth, transpiration, evaporation, clouds, animal growth, etc, etc.
    Why is it so plausible or logical to consider some sort of false future where CO2 is increased, but all else remains exacly the same? Isn’t it plausible that an increase in CO2 may possibly cause a small change in atmosheric water vapour, or clouds, which couteracts the theoretical laboratory effect of CO2 increase on temperature ?

  63. How about the Martian galactic rays too?

  64. Physics of the atmospheric greenhouse effect: What exactly is the question mark about?

    Some have complained that the greenhouse effect is inappropriately named. But they don’t insist that a butterfly be named something else, given that it is neither a fly, nor is it made of butter. English is English for whatever that is worth. You arrive at the words and their conventional meanings because the words get defined by those who use them.

    But “serious” arguments against the (greenhouse) theory by the likes of
    Gerlich and Tscheuschner, Claes Johnson, and Miskolczi? Perhaps the word “serious” was meant to suggest that the authors of these papers were being sincere in what they were writing, and therefore “serious”.

    I could also entertain the possibility that they might be writing their stuff as a “spoof”, just to see who might be taken in. You use equations, make the language sound authoritative by including the right shop talk and buzz words, then come up with some outrageous conclusion. I have seen this happen before in the social sciences,winding up on the Op-Ed pages of the New York Times.

    Actually, the Gerlich and Tscheuschner, Claes Johnson, and Miskolczi papers are a good test to evaluate one’s understanding of radiative transfer. If you looked through these papers and did not immediately realize that they were nonsense, then it is very likely that you are simply not up to speed on radiative transfer. You should then go and check the Georgia Tech’s radiative transfer course that was recommended by Judy, or check the discussion of the greenhouse effect on Real Climate or Chris Colose science blogs.

    The technical books by Goody and Yung, Thomas and Stamnese, and Liou are a bit heavy going. I would recommend the more readable paperback text by Grant Petty (A First Course in Atmospheric Radiation, available from sundog publishing for $36). It does not answer all questions, but does have a lot of good and useful material.

    If the authors Gerlich and Tscheuschner, Claes Johnson, and Miskolczi are not “spoofing”, then I am rather curious about their thought processes as to just how it is that they arrive at such erroneous conclusions. Did they bother to check their results against the published literature? Radiative transfer has been around for a long time, so there are thousands of relevant papers to compare their results against. If they found differences with published results, they should have checked to see what errors they may have made. Perhaps it is that they somehow feel that they have arrived at some new understanding that has been missed by everybody else.

    The notion by Gerlich and Tscheuschner that the second law of thermodynamics forbids the operation of a greenhouse effect is nonsense. The notion by Claes Johnson that “backradiation is unphysical because it is unstable and serves no role” is beyond bizarre. A versatile LW spectrometer used at the DoE ARM site in Oklahoma sees downwelling “backradiation” (water vapor lines in emission) when pointed upward. When looking downward from an airplane it sees upwelling thermal radiation (water vapor lines in absorption). When looking horizontally it sees a continuum spectrum since the water vapor and background light source are both at the same temperature. Miskolczi, on the other hand, acknowledges and includes downwelling backradiation in his calculations, but he then goes and imposes an unphysical constraint to maintain a constant atmospheric optical depth such that if CO2 increases water vapor must decrease, a constraint that is not supported by observations.

    A useful starting point to get a better feel for how thermal radiation works is to consider the classical isothermal cavity maintained at a fixed temperature T. Radiation emitted through a small pinhole from this cavity is always going to be isotropic Planck radiation at temperature T whether the cavity is empty, or if there is absorbing or reflecting material within the cavity. From this concept, it is straightforward to develop equations for how much an absorbing layer within the cavity of specified optical depth and temperature T will emit, since the sum of the layers emission and transmission must be equal to the Planck radiation coming through the pinhole. The same emission and transmission equations apply for the absorbing layer in question if removed from the isothermal cavity and placed in atmospheric context that is in local thermodynamic equilibrium.

    • Mr Lacis, I take issue with your 2nd paragraph only.

      True, that insect is neither a fly nor made of butter. Nor is a pineapple an apple or the fruit of a pine tree. However, being thus named has no consequences on geopolitics or global economies.

      The average person in the street who isn’t interested in studying the science hears the term “Greenhouse”, recalls their knowledge of or visit to a botanical greenhouse and says “yeah, it gets bloody hot in those places”. And coupled with statements such as “we can’t keep pumping this POLLUTION into the atmosphere” decides that there IS a problem and reflects this at the polling booth. (Recent elections in OZ with substantial increases in Green party support).

      Words are bullets, especially in a field like science. I for one expect, nay demand, true and accurate descriptors.
      And when very very highly educated people shortcut their descriptors, I want to know whats on their agenda. I want to know if they are impartial seekers of truth or advocates.

      No, The Greenhouse Effect is not a good descriptor. I know what makes a marketgardeners greenhouse hot and it ain’t CO2.

    • …..”The notion by Gerlich and Tscheuschner that the second law of thermodynamics forbids the operation of a greenhouse effect is nonsense.”…

      In their paper G&T make clear that only some of the “greenhouse theories” fall foul of the second law.

      Some other fall foul of the first law and multiple other mistakes.
      In fact one of G&Ts complaints is they cannot find a version of the “Orthodox Greenhouse Theory”.

      You have recommended a number of textbooks relating to climate physics.
      Do these text books (and in general UG courses in climate science) include the Carnot Cycle and the Second Law of Thermodynamics?
      I’m certainly not “spoofing” when I say that Ive found an incredible ignorance about the direction of heat flow in the atmosphere amongst supporters of the IPCC position .
      Statements such as “backradiation heating the Ocean” are not hard to find.

      • Bryan, FYI, basic thermodynamics is part of the undergraduate curriculum in climate science (and in the earth and atmospheric science departments that teach climate science.) I have written a textbook on Thermodynamics of Atmospheres and Oceans, and there are about a half dozen similar texts on the market. Most meteorology and atmospheric science departments teach a full course in thermodynamics (applied to the atmosphere) at the undergraduate level, which is a required course in the curriculum.

      • Thanks for the reply.
        I’m reassured that thermodynamics is being taught.

        However I asked specifically about the Carnot Cycle.
        It forms the core of analysis into heat engines and refrigerators.
        Its conceivable that since neither are found naturally in the atmosphere that it might be omitted.

        However this would be a mistake as it also is also the introduction to analysis of the Second Law.
        I have come across a number of proponents of IPCC position who have interpreted the Second Law as;
        “its alright for heat to flow from low to higher temperature as long as more heat flows from high to low temperatures at the same time.”
        Which of course is wrong.

    • I used to agree with you, that the name is not important. But when the AGW community has a pattern of shifting names under pressure and also indulges in the entire ‘denialist’ and ‘conspiracy’ schtick in dealing with skeptical challenges, I think names are a bit more significant than just place markers.
      Also, as we watch the list of failed predictions about the global warming crisis grow, it is good to take time to revisit what the AGW community has told us over the years and decades.

    • David L. Hagen

      A Lacis

      Miskolczi, on the other hand, acknowledges and includes downwelling backradiation in his calculations, but he then goes and imposes an unphysical constraint to maintain a constant atmospheric optical depth such that if CO2 increases water vapor must decrease, a constraint that is not supported by observations.

      I think you have misread Miskolczi. He starts with the data, calculates each of the radiative parameters across an atmosphere 150 layers deep in 9 different directions. He calculates a full Planck weighted optical depth.
      From that he fits the data to curves to come up with simplifying relationships. That he turn interprets in light of entropy maximization principles. From what I understand, he does NOT a priori impose “an unphysical constraint to maintain a constant atmospheric optical depth”.

      Rather he develops subsequent theory to model and explain the atmosphere behaving effectively in that way to explain the very stable global optical depth actually observed over the last 61 years including all available moisture, temperature and CO2 data.

      See my detailed response to your posts in Judith’s best of the greenhouse summary.
      See Essenhigh’s thermodynamic model of the atmospheric column

      Comment on “unphysical constraint”

      comment on order of calculations
      I think you read his process backwards – He starts with the data and radiative evaluations, not the simplifying assumptions and entropic explanation. Look forward to your feedback – over at the summary page.

      • Its good to see more discussion on Miskolczi’s paper. It is not easily understood, and ranks IMO as the one skeptical analysis that is worth some detailed discussion. I’ve read the paper, felt that it didn’t make much sense, but I didn’t spend much time on it.

  65. I have not seen this book discussed

    Global Warming and Global Cooling, Volume 5: Evolution of Climate on Earth (Developments in Earth and Environmental Sciences) , O.G. Sorokhtin, Leonid F. Khilyuk Ph.D. Ph.D. , G.V. Chilingarian

  66. Probably somewhat late to comment.

    Nullius in Verba said on November 30, 2010 at 3:13 pm:

    A great deal of confusion is caused in this debate by the fact that there are two distinct explanations for the greenhouse effect: one based on that developed by Fourier, Tyndall, etc. which works for purely radiative atmospheres (i.e. no convection), and the radiative-convective explanation developed by Manabe and Wetherald around the 1970s, I think. (It may be earlier, but I don’t know of any other references.)

    Climate scientists do know how the basic greenhouse physics works, and they model it using the Manabe and Wetherald approach. But almost universally, when they try to explain it, they all use the purely radiative approach, which is incorrect, misleading, contrary to observation, and results in a variety of inconsistencies when people try to plug real atmospheric physics into a bad model. It is actually internally consistent, and it would happen like that if convection could somehow be prevented, but it isn’t how the real atmosphere works.

    1. I would like to comment on “when they try to explain it“.
    In atmospheric physics text books and papers on the subject, when climate scientists “try to explain it” the explanation is correct.
    Perhaps someone did it incorrectly once, but I haven’t found it yet in any technical publications discussing this subject.

    Many websites set up to explain to the general public in media-friendly sound-bites probably do explain it badly. Or non-technically. Or completely wrongly.

    But I believe it is important to differentiate between the two worlds.

    To criticize how climate scientists describe atmospheric physics based on what NASA or the Met Office explain it via their marketing departments on their public websites seems somehow unfair to me.

    Or even to criticize how someone technically competent tries to explain a complex technical subject to a non-technical audience by saying “you made it too simple”.. Equally you could say “you made it too complex”..

    Wonderful if Moving on.

    2. Another comment that can, and has, caused confusion is the somewhat false dichotomy of “good” vs “bad” explanations of how the “greenhouse” effect works.

    The “real mechanism” as Nullius and Leonard Weinstein describe is correct – as more “greenhouse” gases are present, the effective height of radiative cooling to space increases. And because temperature decreases with height due to adiabatic expansion, a higher altitude for this radiation = less radiation = less ability to move heat out of the climate system and therefore = “a heating”.

    This explanation is the one you find in the atmospheric physics textbooks, and the papers.

    However, the idea that “back radiation” is irrelevant is not really correct. It is essentially a complementary effect.

    How is it possible for surface temperatures to increase under this model? Surface fluxes must balance. Yet solar radiation is still the same. A higher surface temperature will cause a higher convective flux and a higher radiative upwards flux. (Increase in energy from the surface).

    How is this possible? There must be a balancing surface item. The balancing item is the back radiation. More “greenhouse” gases will cause more downward radiation (lowering effective altitude of downward radiation).

    It isn’t one vs the other. One is a consequence of the other. The “controlling mechanism” is the radiative cooling to space which is determined by the effective height of the radiation. Downward surface radiation from the atmosphere increases as a result. All are linked.

    (Note these explanations are all prior to any feedbacks).

    Thermal Equilibrium of the Atmosphere with a Given Distribution of Relative Humidity, by Manabe and Wetherald was published in 1967.

    • scienceofdoom
      Good post and the outlines of a testable proposition emerges.
      The consensus of this thread seems to be that as you say;
      ………” The “real mechanism” as Nullius and Leonard Weinstein describe is correct”……….
      However you would like to include the effects of the postulated increased CO2 as increasing the greenhouse effect at near surface lifting the layered structure of the atmosphere up.
      If you can supply evidence that the magnitude of the CO2 increase causes energy density at near surface atmosphere to increase correspondingly then you have a case.
      The proof would need to be in appropriate units such as Joules per cubic metre rather than a mere assertion.

      • Leonard Weinstein

        The adiabatic lapse rate is a gradient not a specific temperature. Thus if the temperature at the “effective” altitude of outgoing radiation that matches the absorbed solar radiation is raised, the ground temperature will increase from the adiabatic lapse rate times altitude plus outgoing gas effective temperature. Near the ground, that higher resulting temperature will result in more radiation from the ground and more back radiation, but the only radiation HEAT transfer is due to the difference of outgoing radiation and back radiation. Thus on the average there is no back radiation HEATING, but there is back and forth energy transfer. Confusion of the difference of energy transfer and heating is the cause of most of this confusion. The increased back and forth radiation is an EFFECT, not cause of the increased heating.

      • Leonard Weinstein

        in my sentence “Thus if the temperature at the “effective” altitude of outgoing radiation that matches the absorbed solar radiation is raised, the ground temperature will increase from the adiabatic lapse rate times altitude plus outgoing gas effective temperature”, I meant to say: Thus if the location where the temperature at the “effective” altitude of outgoing radiation that matches the absorbed solar radiation is raised, the ground temperature will increase from the adiabatic lapse rate times altitude plus outgoing gas effective temperature

      • Leonard

        I fully appreciate that the lapse rate is a gradient.
        Its no coincidence that the dry lapse rate involves the constant 9.81 the magnitude of the acceleration due to gravity.
        The lifting of the tropopause would be because the lapse rate was further reduced by condensation and convection mainly.
        However scienceofdoom (if I understand him properly) was proposing that because of increased CO2 we would have a stronger “greenhouse effect” thus increasing surface temperatures.
        Increased surface temperatures while not changing the lapse rate would increase the height of the tropopause.
        This is because the molecular KE of gases at surface boundary had increased.
        Now this is a perfectly rational proposition and further it seems testable.
        It seems that an analysis of past records of the height of the tropopause and how it varies with;

        1. Surface temperature
        2. Humididy
        3. CO2 levels in the atmosphere

        There are a number of other factors I’m sure.
        However it does seem possible that the relevant driving factors could be teased out from data we already hold.

    • Nullius in Verba

      Hi there, scienceofdoom,

      We’ve had this discussion before, of course. I still respectfully disagree.
      (Respectfully because I’ve learned a great deal from your excellent site.)

      “a higher altitude for this radiation = less radiation = less ability to move heat out of the climate system and therefore = “a heating”.”

      The ability to move heat out of the climate system is exactly the same, the heat being radiated out of the system is exactly the same: it’s the heat that is being absorbed by the system and is fixed by insolation and albedo. The only difference is that it is at a higher altitude. The higher it is, the thicker the sandwich of air between there and the ground, which when multiplied by the lapse rate gives the surface temperature.

      “A higher surface temperature will cause a higher convective flux and a higher radiative upwards flux. (Increase in energy from the surface).”

      I’m not entirely sure, but by emphasising the word “from”, I suspect that you’re still making the assumption that convection can only transport heat upwards, never downwards. Convection (in the sense of a bulk air movement induced by temperature differences) can transport heat both ways. “Convection” refers to the entire cycle.

      “The “controlling mechanism” is the radiative cooling to space which is determined by the effective height of the radiation.”

      The controlling mechanism for the temperature of air at the average emission altitude (the IR-visible ‘surface’ of the planet) is as you say. The controlling mechanism for the temperature at the solid surface of the planet is the altitude of the average emission level, and the lapse rate between that point and the surface.

      So long as the temperature profile of the atmosphere is adiabatic, and hence linear, the temperature at any point is related to the temperature at any other point as a simple function of the gradient of the line and the difference in heights and nothing else. Unless the back radiation can affect one or the other of these, it cannot possibly have any effect.

      The one part of my model that I have left deliberately vague is the determination of the average emission height, which involves not just the concentrations of the greenhouse gases but their temperature and density profiles. If the level of back radiation affected the tropopause height, say, and hence the thickness of the moist turbulent layer, it could affect the heights of emission from water vapour. Or maybe convection across the full range of altitudes would be impossible without it. But I don’t see any other way for it to have a critical effect. A linear function has only two parameters. dT = MALR*dz. Which of these does back radiation affect?

      • I like the way scienceofdoom puts it and I think the difference lies in transient versus steady state behavior. I think the point is that if backradiation did not warm up the surface more, you would not have energy balance at the top of the atmosphere and therefore, if you assume a system where you suddenly turned the sun on, you would have less energy leaving the system than was entering it until you reach steady state.

      • In other words, I believe that the derivation of a constant lapse rate implicitly assumes steady-state behavior i.e. it doesn’t take into account the start-up energy buildup within the earth system to achieve a mean “stored energy” state.

      • Hot air rises, cold air descends. Which direction is energy transported in each case? I suggest the directions of mass and energy fluxes need not be the same.

      • Nullius in Verba

        Hot/cold relative to what?

        Consider a completely transparent atmosphere of pure Nitrogen/Oxygen on a planet with a hot surface at the equator and a cold surface at the poles. Radiation to space all occurs from the surface. The air immediately in contact with the surface is at the same temperature as it by conduction.

        What happens? Does it convect?

        If it is static, what force keeps the dense cold air at the poles from flowing towards the hot light air at the equator? (And the air at altitude from flowing out towards the poles?) If convecting – rising at the equator and descending at the poles, from which it will be radiated, which way is the heat flowing at the poles? Or does all the air rise and never descend again?

      • Nullius,
        By Wikipedia definitions, I am referring to natural convection while you have proposed a forced convection pump.

        The steady-state solution to your posed problem is more than adequately given by thermal conductivity. Gradients are several orders of magnitude below those required for convection. Thermal profiles are isothermal vertically with a slight warming at higher altitudes increasing with latitude. An atmosphere ca. 10Km thick is assumed for an earth-sized planet and rotational effects are ignored.

      • Nullius in Verba

        “Convection” versus”forced convection pump” is a question of differing definitions. I was trying to keep things simple.

        I’m interested to know more about your steady state solution. Can you say specifically: what force keeps the dense cold air at the poles from flowing towards the hot light air at the equator? And the air at altitude from flowing out towards the poles?

        And what is the gradient required for convection?

      • Nullius,
        Just a back of the envelope calculation. First, visualize a film 1 unit thick and 1000 units long on a level hot plate with a horizontal temperature differential of 100K. With that aspect ratio, the obvious first guess is a uniform vertical temperature. To refine the picture, as you indicate, cooler air of higher density will try to slip under warmer air adjacent, but it will also be warmed by the underlying hotplate and thus we have a competition between thermal and viscous diffusion. The dimensionless Prandtl number, for air about 0.8, indicates these processes are closely matched and consequently horizontal and vertical temperature gradients will be similar. The given surface gradient 10^-2 K/Km, applied vertically, yields a 0.1K vertical temperature difference and a lapse rate well below that needed for turbulence.

      • Nullius in Verba

        Are you saying the force that keeps the cold air from slipping under the warm is viscosity? Or that it does slip under, but then gets warmed up? (i.e. the temperature field is static, but the air is not?)

        If the latter, isn’t that a convective cycle rather than a static solution?

      • I’m saying we’ve discussed a problem equivalent to energy and mass transfer through a capillary layer (L/W = 1000) connecting two reservoirs of different temperatures and ill-defined pressures and for which viscosity limits mass transport.

        My ‘research’ indicates that definitions of convection can cover a range from molecular diffusion to turbulence. Personally, I favor a more restrictive definition which would not include capillary flow. If I used the words ‘static solution’, I intended steady-state or stationary.

        As to the hypothetical GHG-free planet, I’d first look for solutions involving a thin fluid layer on the surface of a rotating sphere.

      • Nullius in Verba

        Capillary flow? You’re saying that we would have a 10 km thick boundary layer? I think I must be misunderstanding in some fundamental way what you’re intending to say.

        Your definition of “capillary” makes no sense to me. The dimensions of the atmosphere (10 km x 10,000 km) are the same whether the atmosphere contains GHGs or not, so if the transparent atmosphere is a capillary layer, then so is Earth’s actual atmosphere. Personally, I’d define ‘capillary’ as one in which capillary forces had a significant effect on the flow. Nor do I understand why you’re restricting the definition of convection. For the purposes of this discussion, it’s the bulk motions of air caused by temperature-induced density changes.

        You haven’t said why you think viscosity limits mass transport, and I’m still not clear on whether you think the air would move or not.

        If the air is stationary then pressure at the surface must be constant, or there would be unbalanced horizontal forces. Hence the mass of air over each square metre must be the same. If temperatures are different over hot and cold areas, then their densities will be different, and you will need a much taller air column over the hot areas to balance the pressure over the cold areas. But with nothing but the vacuum of space to hold it up, any such column would spill out on top of the cold air, raising the pressure there and inducing circulation. You can’t get the same pressure at the bottom and the same air column height at the top with different densities.

        Saying “I’d first look for solutions…” gives the impression that you don’t yet have a solution. (Incidentally, if the air is not moving, then the rotation has no effect.) But you said above “The steady-state solution to your posed problem is more than adequately given by thermal conductivity.” Do you know of a solution or not?

      • Nullius:

        Your last paragraph is crucial. The whole disagreement is removed, when its meaning is properly understood. There are different ways of looking at the same phenomena and they meet in understanding your last paragraph.

    • OK, I’m a bit of a simpleton, I guess, but I see it this way. The visible radiation from the Sun heats the ground which in turn radiates infrared radiation upwards. Much of this radiation is captured by CO2. This heats the CO2. The pressure in the lower atmosphere is such that the CO2 almost immediately bangs into another molecule and heats it. So the “heat” captured by CO2 rapidly heats the air around it. Now, the only way this cannot heat the atmosphere is if the heat were instantaneously dissipated into space. Since this cannot happen instantaneously, the CO2 in fact heats the atmosphere and causes warming. The details of how this heat gets spread around from this point don’t really matter.

  67. Question

    1) Does a physical climate model exist, or has anybody built one? I’ve tried to search on Google.

  68. Climate models are not being specifically labeled as being “physical” climate models. But typical climate GCMs would all be “physical” climate models since their whole setup is formulated to model explicitly the hydrodynamic and thermodynamic processes of climate. One such model is the GISS ModelE for which the FORTRAN code is available from the GISS webpage at

    While we speak of the greenhouse effect primarily in radiative transfer terms, the key component is the temperature profile that has to be defined in order to perform the radiative transfer calculations. So, it is the Manabe-Moller concept that is being used. In 1-D model calculations, such as those by Manabe-Moller, the temperature profile is prescribed with the imposition of a “critical” lapse rate that represents convective energy transport in the troposphere when the radiative lapse rate becomes too steep to be stable. In 3-D climate GCMs no such assumption is made. The temperature profile is determined directly as the result of numerically solving the atmospheric hydrodynamic and thermodynamic behavior. Radiative transfer calculations are then performed for each (instantaneous) temperature profile at each grid box.

    It is these radiative transfer calculations that give the 33 K (or 150 W/m2) measure of the terrestrial greenhouse effect. If radiative equilibrium was calculated without the convective/advective temperature profile input (radiative energy transport only), the radiative only greenhouse effect would be about 66 K (for the same atmospheric composition), instead of the current climate value of 33 K.

    • I wonder if you have a response to the comment I made above regarding ModelE? -> Richard, December 1, 2010 at 10:27 am


  69. Dr Lacis
    Thanks for your reply. I was wondering about the Fultz and Rossby kind of stuff

  70. orbital wobble

    At last

    A site where different perpectives can be sensibly discussed without rancour, insult or deletion. Judith, if this keeps up we might send you to sort out the middle east…..

  71. David –
    “Why don’t the planetary surface conditions matter?” Yes, some do, of course. For instance, I should have said “…the outward LW radiation [must] be equal to …that portion of the the impinging SW radiation not reflected by the surface as outward SW radiation.

    But Miskolczi thinks that it is important to eliminate the discontinuity in temperature at the surface that the Milne formalism imposes, and this cannot be done without introducing physical mechanisms beyond those described by the purely radiative equations (and attendant approximations). Specifically, the Eddington approximation divides the radiation field into upward and downward fluxes (each semi-hemispherically isotropic). The Milne equations describe the spatial variation of these fluxes as a function of optical depth, given their values at a single altitude. The only place where the values are known a priori is at TOA. (It may help to think of these fluxes as ‘wave-like’: you can launch a wave at a point where you specify its properties, but how those properties change as it propagates through space are determined by the medium, in this case the optical depth as a function of altitude.) So one cannot simultaneously impose conditions at two altitudes on the downward LW flux, independent of the properties of the intervening medium. This characteristic is manifested mathematically by the fact that the governing differential equations are second order, requiring two boundary conditions, one (and only one) for each component of the flux.

    A consequence is that there is a discontinuity between the temperature at the base of the atmosphere and the solid surface, the surface being warmer because it receives both the downward SW radiation and the downward LW radiation from the atmosphere (the so-called ‘back radiation’). The magnitude of the discontinuity diminishes as total optical depth increases (the surface and air become more strongly coupled, as one might expect), but radiation alone does not erase the discontinuity. Miskolczi has a problem with this, but Nature does not, as convection, evaporation and other modes of heat transfer are adequate to the task of handling the discontinuity. As we know, in the terrestrial atmosphere the usual situation is that convection extends far up in the troposphere, so the formal discontinuity issue does not even arise. A convective profile (e.g., moist adiabat) is matched to a rigorously calculated radiative profile, as you indicate. Why Miskolczi tried to base an energy minimization scheme on the purely radiative equations all the way to the surface is a mystery to me.

    I hope this helps.

    • David L. Hagen

      Pat Cassen
      Thanks Pat for the clarification of your previous comments.

      Miskolczi is beginning with a 1D radiation problem to quantitatively model a planetary “greenhouse”, and working out from that. By finding what radiation can or cannot do, that exposes what must be done by convection, precipitation, clouds etc.

      I understand your point of 2 fluxes, 2 boundary conditions.

      You note: “So one cannot simultaneously impose conditions at two altitudes on the downward LW flux, independent of the properties of the intervening medium. ”
      “The only place where the values are known a priori is at TOA”

      However, with a gas above a solid planet, the radiative properties of the surface affect the surface absorption & radiation of both SW and LW and the surface temperature. This gives more parameters necessitating more boundary conditions. Those will affect the surface temperature and consequently upward SW reflection as well as LW radiation. I don’t see how you can model a planetary atmosphere with a solid surface without that. (With a totally gaseous planet this is not required.)
      The portion of SW absorption & reflection requires a SW absorptivity parameter. Miskolczi assumes a black body.
      The portion of LW absorption & reflection requires a LW absorptivity parameter.
      Miskolczi again assumes a black body.
      Internal planetary heat radiation approximated to zero for the initial model.

      (A future refinement would be use average planetary absorptivity and emissitivity.)

      Miskolczi starts with the actual data on the atmospheric profile for 11 geographic regions. He fits this to obtain the empirical atmospheric profiles.
      See Zagoni Slide 68 where he shows the actual global average temperature lapse rate profile) from the radiosonde data. Then he shows his fit to that with his “semi-transparent” model. That fit appears to be good. Note that it is more accurate than the USST-76. Note particularly the major difference in water column between the USST-76 and actual global average based on the available data.

      Contrast Milne’s “semi-infinite” model which seriously diverges from the global average – being warmer at higher altitudes and cooler near the surface. Note the very large surface discontinuity in the Milne model.

      You note: “but radiation alone does not erase the discontinuity. Miskolczi has a problem with this, but Nature does not, as convection, evaporation and other modes of heat transfer are adequate to the task of handling the discontinuity.”

      Setting the surface boundary temperature equal to the atmospheric temperature, appears to be equivalent to setting no fluid boundary resistance, equivalent to high convection.
      The resulting error compared to the global average temperature profile appears to be small. See Zagoni Slide 68. I agree that this should be relaxed in future modeling to provide a more accurate local surface temperature and absorptivity emissivity.

      From that he conducts a detailed calculation for all radiatively active species (“greenhouse gases”), for 150 layers, in 9 different directions. He applies a solar (Planck) weighting, to then find the global Plank weighted optical depth. I believe this assumes constant solar radiation, zero LW input, and a blackbody lower surface.

      He does this both for the actual TIGR data and for the NOAA reconstruction. See Zagoni Slides 69 and 79.

      From that he calculates the optical depth from annual data for the last 61 years, including all variations in water, CO2 and temperature with depth. See Zagoni Slides 58-60.
      Note the very small variations in total global optical depth over the last 61 years.

      You note: “Why Miskolczi tried to base an energy minimization scheme on the purely radiative equations all the way to the surface is a mystery to me.”
      Separate out Miskolczi’s global optical depth evaluation vs subsequent theoretical modeling. With multiple atmospheric components, the gas proportions can change. e.g. CO2 increasing. More importantly, H2O can increase or decrease depending on temperature. i.e. precipitation and ice formation. The equilibrium properties of water provide another constraint. Then the gravitational field & pressure lapse rate come into play. The atmospheric height or lapse rate also changes with temperature and Miskolczi accounts for this change.

      For a future full thermodynamic development see Essenhigh:
      Prediction of the Standard Atmosphere Profiles of Temperature, Pressure, and Density with Height for the Lower Atmosphere by Solution of the (S−S) Integral Equations of Transfer and Evaluation of the Potential for Profile Perturbation by Combustion Emissions Robert H. Essenhigh Energy Fuels, 2006, 20 (3), 1057-1067 • DOI: 10.1021/ef050276y

      So may I encourage you to review Miskolczi’s optical depth evaluations vs his theoretical modeling in more depth.

      • David –
        Thanks for your response, which I just now found.

        You state: “…the radiative properties of the surface affect the surface absorption & radiation of both SW and LW and the surface temperature. This gives more parameters necessitating more boundary conditions.”

        More parameters, but not more boundary conditions. (Mathematically speaking, adding parameters at the surface does not change the order of the differential equations.)
        You can demonstrate by direct calculation that Miskolczi’s ‘solution’ violates the boundary condition at TOA; it requires a non-zero downward LW flux from space, which is unphysical. His ‘solution’ is invalid; it is not a correction to Milne. I emphasize that you can, with pencil and paper, show this for yourself.

        “Setting the surface boundary temperature equal to the atmospheric temperature, appears to be equivalent to setting no fluid boundary resistance, equivalent to high convection.”

        Yes, but high convection then negates the validity of Milne’s (and Miskolczi’s) purely radiative equations. His prediction of a preferred optical depth depends on a minimization of radiative transport all the way to the surface, where, in fact, transport is dominated by convection at the surface. So, even if he had done the math right (he hasn’t), the analysis is irrelevant.

        You (and tallbloke, I guess) are impressed by the fact that Miskolczi derives an average atmospheric optical depth from a theory (which is demonstrably flawed in concept, methodology and mathematical derivation), which matches that derived (to three significant figures!) from a deficient data set. I am only suspicious.

        I urge you to take seriously the other critiques of Miskolczi that I previously linked. (Here is a link to the Dorland and Forster critique.)

        Thanks for the discussion. I offer you the last word on this.

      • Pat, thanks for the link, I will read and try to learn. I am not an expert in radiative physics, my training is in using others knowledge to help me assess theories and the operational processes which build data validity, an art in itself, not an exact science.

        I do wonder how badly flawed the radiosonde dataset is though. I have a couple of reasons for thinking it might not be as bad as has been made out. And I can see some motivations for doubt being cast on its validity which might not be directly driven by fully objective criteria.

      • David – In the above, I should have said “…negates the applicability…”, not “…negates the validity….”

  72. From my comment:

    A higher surface temperature will cause a higher convective flux and a higher radiative upwards flux. (Increase in energy from the surface)..

    Nullius said on December 2, 2010 at 12:10 pm:

    I’m not entirely sure, but by emphasising the word “from”, I suspect that you’re still making the assumption that convection can only transport heat upwards, never downwards. Convection (in the sense of a bulk air movement induced by temperature differences) can transport heat both ways. “Convection” refers to the entire cycle../i>”

    I’m not making the assumption that convection can NOT transport heat downwards.

    Generally the surface is at a higher temperature than the atmosphere and so convective heat flux is upward.

    However, I don’t think I gave a good explanation before – at least, it was much too brief. Let me try again.

    In the case where “greenhouse” gases increase:
    1. the radiative cooling to space is at higher altitude
    2. therefore, from a colder temperature
    3. therefore, less radiative cooling.
    (As noted before and no disagreement between us).

    What happens now?

    Less cooling to space means that the climate system progressively stores more heat (until a new steady state is reached).

    This is how the climate system warms in a general sense.

    But how does the surface temperature change specifically?

    a) If the atmosphere first increases in temperature then the atmosphere will radiate more to the surface. = More back-radiation which will increase the temperature of the surface.

    b) If the surface first increases in temperature then convective and radiative flux will then increase the temperature of the atmosphere above. = More back-radiation which will increase the temperature of the surface.

    There is another effect – the more opaque the atmosphere the lower the effective radiative altitude of the downward radiation. (If the atmosphere was transparent it would not radiate at all).

    Nullius said:
    ..So long as the temperature profile of the atmosphere is adiabatic, and hence linear, the temperature at any point is related to the temperature at any other point as a simple function of the gradient of the line and the difference in heights and nothing else. Unless the back radiation can affect one or the other of these, it cannot possibly have any effect..

    It’s a confused statement.

    The surface temperature is determined specifically by the balance of fluxes at the surface.

    It’s quite simple and should be uncontroversial. If back radiation was unimportant then if it dropped the surface temperature would be unaffected.

    At 15’C, a 1’C increase in temperature – for a surface like the ocean – increases the radiative flux by 6 W/m^2.

    Yet with a 1’C increase in temperature – if the back radiation has not increased and the solar radiation has not increased and the convective flux has not changed – then the surface will cool. Because an extra 6 W/m^2 is being lost.

    Very simple and surely no one disagrees with this fact?

    So the only way an extra 6W/m^2 can be radiated is if an extra 6W/m^2 is received at the surface.

    This is what I mean by the fact that “back radiation” is a complementary effect.

    Saying back radiation “has no effect” seems like such a strange statement I think I haven’t understood the point of the statement maker..

    • Leonard Weinstein

      Please keep in mind that heat transfer is not due to the level of energy exchange, but only due to the net difference of energy exchange. If you keep increasing back radiation due to a hotter lower atmosphere, but at the same time increase forward radiation due to a hotter ground (both on the average), any heat transfer would only depend on the difference, and which transfer direction was larger. On the average the upward radiation is slightly higher than back radiation, so all net radiation heat transfer is up. The main heat transfer over the average Earth is convection driven up to the altitude where radiation out occurs. Since there is no net heat transfer down (on the average) from back radiation, there is no heating from back radiation, and back radiation is a result of, not cause of atmospheric greenhouse effect. It is not complementary, it is an effect. Back radiation has no effect means it does no heating.

    • Nullius in Verba

      “Saying back radiation “has no effect” seems like such a strange statement I think I haven’t understood the point of the statement maker..”

      I think that’s quite possible. It’s also possible that I’ve made a mistake and am missing something. So I’d like to pursue this a little longer.

      I’ll pick up my pan of water analogy again. A pan of water is boiling on the stove – what is the explanation for its temperature being 100 C? If we turn up the gas, what happens to the temperature?

      On the one hand, it is true to say that the temperature is determined specifically by the balance of heat fluxes in and out of the water. There is the heat conducted in through the base from the gas, the heat output from radiation from the sides of the pan, from the water’s surface, and via evaporation as steam escapes. We can say that the temperature is 100 C because of the precise balance of conduction, radiation, and evaporation. Clearly, the gas being on “has an effect” – and if we turn the gas up from 2 to 4, the detailed heat fluxes will change. Similarly, if we replace the matt black pan with a shiny silver one, the radiation out from the sides of the pan will change considerably, again changing the heat fluxes.

      So if you want to say that the temperature of the water is determined by the heat fluxes in an out of it, then the reason that the water is at 100 C requires that you include gas setting and the pan colour. They are a part of the specific reason for the temperature.

      What I am trying to say is that they don’t actually matter, because the convection/evaporation term will always increase or decrease in such a way as to maintain the temperature at exactly 100 C. The temperature is controlled by convection/evaporation, which ‘dominates’ the other effects in a dynamic sense (as opposed to in magnitude).

      Likewise, with the atmosphere. The convection will always increase or decrease to maintain the adiabatic lapse rate, and the surface temperature is simply the effective radiative temperature (-20 C) plus the lapse rate (6 C/km) times the difference in altitude of the average emission altitude and the surface (about 5-6 km).

      To get any other value, you have to change one of the inputs (e.g. by claiming that back radiation changes the adiabatic lapse rate), or falsify one of the basic assumptions (e.g. by saying that convection is impossible and the thermal profile of the static atmosphere is stratified). I’m open to the possibility, but I can’t figure out which one you’re saying is wrong.

      I get the impression that you’re trying to say that without back radiation, convection would be impossible. Like saying that if you turn the gas off, the water would stop boiling. But that would be to misunderstand the point I intended to make.

      • Your discussion contains four basic values: temperatures at the earth surface and at the tropopause, the altitude of the tropopause and the lapse rate. One equation connects these three values. The lapse rate is determined by the properties of the atmospheric gas. We have still two independent values that must be determined from additional conditions.

        One of these remaining conditions is related to the factors that determine where the convection stops. The last one is determined by the energy balance, i.e. by the condition that there is no net flow of energy to the troposphere. The amounts of CO2, H2O and other greenhouse gases affect this energy balance through changes in the radiative energy transfer within the troposphere. These radiative energy transfers include back radiation.

      • I wonder whether disagreements about back radiation are more about semantics than physics. Individual IR photons back radiated to the surface will impart kinetic energy to the molecules of water or terrestrial components, thereby warming them. On a global average, however, the net flux is from surface to atmosphere, and so one can describe the net effect as a reduction in cooling rather than as a warming.

        Also, averaged globally, convection is quantitatively more important than radiation in moving sensible and latent heat upward to where it will be radiated to space. However, this varies with latitude. In the warm tropical oceans, convection strongly dominates, whereas at cooler latitudes with less atmospheric water, radiation plays a larger role. Radiation is also a necessary background for convection. Increased latent heat rising from the surface requires surface warming (or reduced cooling) via down-dwelling radiation. In addition, the air overlying ocean surfaces is generally saturated with water and can’t sustain increased evaporation unless it is warmed, which occurs mainly via greenhouse gas radiative transfer.

        As far as I know, current theory and models base their conclusions on a combined radiative/convective equilibrium paradigm.

      • I’m not sure whether these are all purely semantic differences, but it sounds to me like NiV says that backradiation does not matter, only the lapse rate matters. As Nick Stokes writes here – an N2 only atmosphere will still have the dry adiabatic lapse rate, but the surface will be at 255K and emit 235 W/square meter and the atmosphere will progressively cool as you go higher into the atmosphere based on the dry adiabatic lapse rate. Effective radiation height is the surface of earth. Backradiation will however change the surface flux and hence the surface temperature. Using the lapse rate, you can determine where the effective radiation height lies. In the case of the N2-only atmosphere, it lies at the surface of the earth and when you add greenhouse gases, it lies somewhere higher.

      • In other words, we still need a blackbody emitting at 255K for energy balance with the sun, however, the earth’s surface will be at a higher temperature because of the backradiation. Using the lapse rate, you can determine at what height energy balance was achieved with the sun.

      • Nullius in Verba

        “In other words, we still need a blackbody emitting at 255K for energy balance with the sun, however, the earth’s surface will be at a higher temperature because of the backradiation.”

        It will be at a higher temperature because air descending from the emitting black body altitude to the surface will be compressed.

      • Convection is not the process that creates the actual temperature difference. Its role is to prevent the difference from being twice as large. You need the back radiation to reach the actual lapse rate in the atmosphere.

      • Nullius in Verba

        Would you agree that convection is the process that controls the magnitude of the temperature difference?

      • I agree that convection is an essential controlling factor in the tempertature difference. The power that feeds the effect has also an important role. The more back radiation, the more heating power and the higher temperature difference after conduction has done its part.

        Conduction keeps the result essentially linear in power.

      • Nullius in Verba

        Conduction? What part is played by conduction?

      • Error. I meant convection.

      • Nullius in Verba

        OK, good.

        Given the length of this thread, I don’t think anything is to be gained by dragging the debate out endlessly here. I’m sure the subject will arise again, here and elsewhere.

        The essential point I had hoped to make was that convection was the mechanism that controlled the magnitude of the warming. And I certainly agree that back radiation constitutes a large part of the surface heat flux in practice. We’ll leave the hypothetical thought experiments for another time.

      • I add only, that I have been learning a lot during the last couple of days. I have a strong background in physics, but on many details of atmosphere I had only rather vague knowledge. Reading these threads including your messages has helped me in getting things sorted.

        Now I join many others in thinking, how one could include all essential features in a description, which is essentially correct in all respects, but still sinple enough to present in a relatively brief space. I do not know, whether it is possible, but I have some ideas.

      • RB – I admit I sometimes have difficulty interpreting what is intended in certain comments. In any case, back radiation is an essential element of the greenhouse effect. The surface radiates upward towards space. Without greenhouse gases, that radiation would escape unimpeded, and the Earth’s temperature would be much colder. In the absence of back radiation, the same thing would happen even in the presence of greenhouse gases – these would simply be acting as “relay stations” for IR photons headed upwards. My point about semantics was simply that if back radiation makes the surface warmer than it would be otherwise (which it does), the disinction between “warming” and “reduced cooling” is a disagreement more about descriptive terminology than about what is happening to the temperature.

      • Fred,
        There is some semantics involved. I think L. Weinstein is saying you cannot say “backradiation warms” because net heat transfer is only in one direction. If that is the terminology, I agree. I however think NiV is wrong and scienceofdoom is correct. My position is that:
        1. For energy balance, we need ablackbody radiating at 255K.
        2. Backradiation results in a higher flux and higher resulting surface temperature than 255K
        2. The physics behind a constant lapse rate determines that the black body at 255K is obtained at a greater height as you increase the amount of GHGs. That is, to conform with the physics behind the constant lapse rate, as you add more GHGs, the atmospheric layer has to become thicker.

      • Stated differently, if you have x-y=k (a constant), as you increase y (the backradiation), the value x (surface radiation) has to increase.

      • Nullius in Verba

        Can you expand on point 2, please? Why does a higher flux result in a higher surface temperature? Why does it not result in more vigorous convection, and the same surface temperature?

      • Nullius,
        Assuming that earth is surrounded by a series of concentric shells, the shell closest to earth is radiating in both directions and the last absorber and re-radiator is the earth’s surface. Therefore, the surface cannot stay at the same temperature as without the GHGs.
        If you have a fixed electrical current that you have to force through a resistor, as you increase the resistance, the potential gradient that develops across the resistor has to increase. Back-radiation essentially is a resistance to the flow of heat from earth to space. As the back-radiation increases, you need to establish a greater temperature difference between the surface and the effective radiation height to force the heat flux through the atmosphere.

      • “the potential difference that develops across the resistor …”

      • Nullius in Verba

        The shell model would result in an exponential temperature profile and an average surface temperature of 45 C. This is contradicted by observation.

        The surface does not get that hot because it is “short-circuited” by convection, that kicks in as soon as the gradient exceeds a set value. So long as the surface stays below the threshold, convection slows to a stop and you are left with the radiative/conductive resistance, but the moment it goes over, the “circuit-breaker trips” and convection drops the resistance to upward heat flow dramatically, cooling the surface.
        (For those who know some electronics, a reverse-biased Zener diode does something like this.)

        Or to take another example, if you boil water on gas setting 2, it’s temperature is 100 C. If you turn the gas up to 4, a greater heat flux is entering the water, but it’s temperature remains at 100 C. How is this possible?

        My question is, why does a higher flux result in a higher temperature? I don’t see how one logically follows from the other.

      • Nullius,
        The shell model is to illustrate that surface temperatures have to be higher than without the GHGs because a blackbody absorber is also a blackbody emitter and the shell radiates in the IR regime in both directions and that every IR emitter (and absorber) is in equilibrium with the IR emitters surrounding it. Back-radiation is therefore a boundary condition to be applied in solving for the vertical temperature profile. Convection places a constraint on the temperature gradients permissible to less than models that used radiative equilibrium alone and has to be used in conjunction with radiation models to explain tropospheric temperatures. Convection-imposed lapse rate is therefore a limiter that will also have to be satisfied. Therefore, convection has to be used in conjunction with the backradiation to explain surface temperatures. As stated here , Radiative-convective equilibrium is equilibrium of radiative+ convective fluxes .
        I will have to continue this discussion later.

    • Scienceofdoom,
      (If the atmosphere was transparent it would not radiate at all)
      This is just not true! You seem to have a fixed belief that the atmosphere cannot radiate any heat unless it is absorbing greenhouse gases (this also appears in places on your blog).

      Also, I wonder if you and NiV are talking past eachother because you are using different conditions for determining surface temp, (a) NiV using lapse rate and implicitly assuming surface temp=lower atmosphere temp,
      (b) SoD thinking of surface flux balance.
      In reality there is a surface atmospheric boundary layer.

      • Leonard Weinstein

        An ideal optically transparent gas cannot absorb or radiate at the realistic temperatures in the atmosphere of Earth (they can when they reach temperatures where some electron excitation can occur, but this is very hot). Real diatomic gases like N2 and O2 do have some modes of molecular excitation, but they do not have significant absorption or emission at Earth temperatures (or incoming sunlight), but they do have a very small amount. Dilute gases are not like blackbody absorbers or emitters, and the narrow spectral bands of interaction make them effectively non absorbing or emitting. This means the ground and seas, heated by incoming sunlight, would radiate right through the atmosphere if there were no greenhouse gases (water vapor is a greenhouse gas, and assuming it is not there also removes clouds). The presence of the greenhouse gases, with absorption and radiation acts like a radiation insulator, but convection still transfers heat and thus the greenhouse effect. I get the impression you consider air to be like a black body, but except for water drops (clouds) it is not.

      • Even in a 100% N2 atmosphere, nitrogen would heat due to contact with the surface of the Earth. So it would heat up, just more slowly and by a non-radiative mechanism.

      • I tend to agree with Jim, the Sun shining by day will produce a certain surface temperature.
        The molecules of air hitting the surface will leave with this characteristic temperature.
        The higher the surface temperature the higher the tropopause.
        The dry adiabatic lapse rate is set by the Gravitational Field Strength.
        This will reduce the molecular KE (hence temperature) in the vertical direction.

        If you work out a table of points for a molecule(Temperature,Height) moving vertically as it gains PE at the expense of KE you have the basic picture.
        Convection and condensation effects modify the dry adiabatic lapse rate forcing the troposphere height up.
        Radiative effects will also modify the picture but in a secondary role.

      • Leonard Weinstein

        If you read what I have been saying, you will see that I said convection will occur with or with out greenhouse gases, and the adiabatic lapse rate will still happen, at least initially. However, the lapse rate is a GRADIENT, not a level. There would be no radiation to space from the upper atmosphere, and no extra heating at ground level without the greenhouse gases. The tropopause and stratosphere are from different causes, where convection stops mixing, and where low density gases interact with solar UV and solar ions. After a long while the lapse rate would even decrease to below the adiabatic lapse rate due to thermal conduction.

      • So you are saying a, for example, 100% nitrogen atmosphere would continue to heat until, I suppose, it got so hot it could radiate IR? An almost infinite warming? So by your lights CO2 cools? Is that your scenario?

      • That’s wrong, it wouldn’t heat indefinitely, the atmosphere would eventually be the same temperature as the black body equilibrium temperature between the Earth/Sun-Space.

      • And the surface of the Earth would radiate freely into space.

      • Correction: The (LWR) from the Earth’s surface would radiate freely into space.

      • Leonard

        Sorry if I’m not making myself clear

        …”If you read what I have been saying, you will see that I said convection will occur with or with out greenhouse gases, and the adiabatic lapse rate will still happen, at least initially. However, the lapse rate is a GRADIENT, not a level.”

        …” There would be no radiation to space from the upper atmosphere, and no extra heating at ground level without the greenhouse gases.”
        Agreed with the extra being GHG contribution.
        There would still be radiation from surface and even more if CO2 and H2O did not radiate in the IR.

        …”The tropopause and stratosphere are from different causes, where convection stops mixing, and where low density gases interact with solar UV and solar ions. “…


      • Convection transfers heat and that leads to a reduction in greenhouse effect.

  73. David L. Hagen

    “If the atmosphere was transparent it would not radiate at all.”
    Please be precise in terminology eg “transparent” – to what?
    You can have perfect transparency in the SW visible, while there is still radiation in the LW long wave/infrared.

  74. Pekka Pirilä at December 3, 2010 at 4:28 pm nicely sums up the subject.

    Nullius in Verba at December 3, 2010 at 3:48 pm asks an interesting question:

    Why does a higher flux result in a higher surface temperature? Why does it not result in more vigorous convection, and the same surface temperature?

    and later (December 3, 2010 at 4:23 pm) comments:

    Or to take another example, if you boil water on gas setting 2, it’s temperature is 100 C. If you turn the gas up to 4, a greater heat flux is entering the water, but it’s temperature remains at 100 C. How is this possible?

    My question is, why does a higher flux result in a higher temperature? I don’t see how one logically follows from the other..

    I’m not sure that Nullius and myself have a substantive difference, although it seems that Nullius really thinks we do (as he earlier stated).

    The following comments are not to try to claim any inaccuracy on Nullius’ part, and as he has offered analogies I am unsure of the essential point of the analogy..

    1. If there was more back radiation then the surface temperature would rise (1st law of thermodynamics)
    2. A higher surface temperature would induce more radiation (inevitable result of the Stefan Boltzmann law) and more convection to the atmosphere
    3. These would increase the temperature of the atmosphere:
    i) thus reducing convective transfer from the surface to the atmosphere (convection is governed by temperature differences)
    ii) thus increasing back radiation, therefore reducing net radiation between the atmosphere and the surface
    iii) thus increasing atmospheric radiation to space
    4. Increasing atmospheric radiation to space causes the climate system to cool down

    If this is what Nullius means by the boiling pan analogy then it seems like a good analogy.

    If the boiling pan analogy is meant to mean that any increase in surface temperature simply causes more convection then this is not really correct, as convective heat transfer is only one part of how heat moves from the surface to the atmosphere.

    • Nullius in Verba

      “If the boiling pan analogy is meant to mean that any increase in surface temperature simply causes more convection then this is not really correct, as convective heat transfer is only one part of how heat moves from the surface to the atmosphere.”

      It was meant to give an example where any increase in heat supply to the surface simply causes more convection even though convection is only one part of how heat moves from the water to the surroundings, as a counter-example to the simple assertion that more heat input from one source implied a higher temperature. It’s only meant to demonstrate that it is physically possible – it breaks none of the laws of thermodynamics.

      The way that convection suddenly turns on at a threshold temperature makes it a bit like a thermostat. In the case of water, the threshold is the boiling point. In the case of the atmosphere (at the surface), it is -20 C plus the average emission-to-space height times the adiabatic lapse rate. What I’m trying to say is that none of these three controlling parameters – the S-B temperature, the emission to space, or the adiabatic lapse rate – involves the back radiation.

      I am really unsure whether the difference is substantive. I am hoping that we agree on the physics (of the Earth’s atmosphere, at least), and that this is an argument only about interpretation – whether the adiabatic thermostat or the detailed balance of fluxes that add up to it constitutes the ‘better’ explanation. But I don’t know.

  75. With apologies for the brevity of my earlier parenthetical comment:

    “..There is another effect – the more opaque the atmosphere the lower the effective radiative altitude of the downward radiation. (If the atmosphere was transparent it would not radiate at all)..”

    My slightly expanded statement: “If the atmosphere was transparent to terrestrial radiation then it would not radiate at all

    PaulM said, on December 3, 2010 at 1:12 pm:

    This is just not true! You seem to have a fixed belief that the atmosphere cannot radiate any heat unless it is absorbing greenhouse gases (this also appears in places on your blog).

    You are correct that I do have a fixed belief, an unfortunate effect of the indoctrination from many atmospheric physics and radiation textbooks. My fixed belief is this:

    a) if a gas can absorb at a given wavelength it can emit at that wavelength
    b) if a gas CANNOT absorb at a given wavelength it CANNOT emit at that wavelength
    c) absorptivity = emissivity; inherent material properties that are functions of wavelength (and in the case of non-diffuse surfaces, also direction)

    You can read more about these crazy ideas at Planck, Stefan-Boltzmann, Kirchhoff and LTE.

    Therefore, if the atmosphere was unable to absorb terrestrial radiation (4-100um) it would be unable to radiate in the 4-100um band.

    Why not have a read of the above article and then demonstrate your claim: “this is just not true.

    If it would help I can scan a few pages of a “non-cherry picked” textbook.

    • scienceofdoom

      ……”a) if a gas can absorb at a given wavelength it can emit at that wavelength”…….

      You would need to be careful that this did not imply that for every 15um photon absorbed by a CO2 molecule we expect a 15um photon to be emitted.

      Earlier I wrote

      …..”Pick 15um radiation for a calculation of probabilities.
      Picking an atmospheric temperature of 243K, I do a probability calculation using Maxwell Boltzmann statistics.
      Absorption by CO2 of 15um from the Earth surface is very likely since virtually all CO2 molecules will be in translational mode only.
      The extra energy absorbed is shared out by collision with mainly N2 and O2.
      However emission due to subsequent collision is highly unlikely compared to absorption.

      1% for CO2 stopping ” dead”
      0.25% for CO2 to be left with average KE
      The other N2 and O2 molecules now have the energy and its highly unlikely to result in another photon.
      CO2 is much more likely to emit at a longer wavelength at these temperatures.
      Eli Rabitt has come to much the same conclusion in his sites most recent posting

      • The wavelength 15 um is close to the maximum point of black body radiation at the lowest tropospheric temperatures. In lower atmosphere the maximum is around 10 um. Practically all infrared radiation from CO2 comes at approximatetly 15 um. The emissivity is very low at all other wavelengths of the infrared region.

        The absorption and later emission are separate processes. Sometimes the emission, however, follows absorption with very little delay based directly on the exitation created by the absorption. When the emission is a separate process, the intensity is determined by the temperature.

      • Pekka

        The CO2 molecule that absorbs a 15um photon will find in all probability that the new energy appears as rotational and vibrational modes.
        This molecule then experiences 10 to the power of 10 collisions per second.
        Virtually all the increased KE of the CO2 is shared out mainly with N2 and O2 molecules.
        In other words the energy is thermalised
        I have compared an average molecule at 247K with one at the same temperature which then absorbs a 15um photon.
        The emission of a 15um photon is possible but much less likely.
        The CO2 molecule would have to acquire by collision quite a high value of KE
        Using MB statistics I work it out at one emission for 100 absorptions and this is when the CO2 molecule gives up all translational KE.
        I conclude from this that the magnitude of backradiation of 15um wavelengths is much smaller that the Earth surface radiation.
        Of course longer radiative wavelengths are much more likely and probably more likely to come from H2O moleclues.

      • Of course the CO2 is thermalized and energy divided to all molecules. The temperature of the troposphere is such that thermalized CO2 emits efficiently at 15 um, because this is very close to the most favoured wavelength at those temperatures and because CO2 has a high emissivity at that wavelength.

      • I think you are comparing a CO2 molecule to a radiating “black body”.
        The real situation is the probability of molecules colliding with enough extra energy to release a 15um photon when the temperature is 243K (a typical mid troposphere temperature)

        Maxwell Boltzmann statistics gives Np/Nt = e^-(Ep – Et)/kT

        Np =Number of particles witj enough energy to release 15umphoton.
        Nt = Number of particles with average translational KE;

        Ep=Translational KE energy of molecule + 15umPhoton.
        Ep = 3kT + hf
        Ep = 18.05×10^-21
        Et =Translational KE = 3/2kT = 5.02 x 10^-21J
        Nt/Nt = e^-3.89 = 0.02 or 2%
        Thus we have only 2 emissions of 15um photon for every 100 absorptions.

      • Ep = 3/2kT + hf

        The 2 was missing above but just a copying error the calculation is still correct I think!

      • The correct interpretation for the 2% is that 2% of the molecules are in the 15 micron vibrational state, and this balance is maintained by continual emissions and absorptions at 15 microns that maintain that population. Collisions don’t necessarily have to occur. Emissions exactly balance absorptions in the special case of radiative equilibrium which is not the case for the atmosphere, where emissions slightly outweigh absorptions leading to net radiative cooling (that is balanced by convective heating).

      • I disagree the statistics show that it would be highly unusual for a molecule at this temperature to have such a high level of energy without having absorbed a photon .
        When absorbed the energy is spread out by collision to N2 and O2 molecules which don’t radiate in the IR.
        Further the spectrum at TOA from satellite looking down shows a large bite missing at around 15um which is exactly what I would expect.

      • It means that statistically 2% are in the vibrational state, and emitting at the temperature of the molecule which, yes, is determined by collisions and radiation. The bite is actually emission at colder temperatures, such as 220 K, compared to the warmer background wavelengths where CO2 has let the radiation pass through. When you look down on the atmosphere you see the 15 micron radiation from higher up due to the opacity of CO2 there.

      • ….”It means that statistically 2% are in the vibrational state, and emitting at the temperature of the molecule which, yes, is determined by collisions and radiation.”……

        With 10^10 collisions per second any of the 2% wont hold it for long.
        The other 98% however in the translational mode can accept a 15um photon from the Earth surface most likely (especially at night).
        Thus we have thermalisation of the surface radiation which causes local heating.
        My explanation of the 15um bite at TOA seems simpler so Ill go with Occam on that one.
        WUWT in a current post have a nice picture of the bite which backs me up.

      • Bryan,

        I referred to the spectrum of black body radiation as simple evidence that shows 15 um to be close to the optimal wavelength for emission at the tropospheric temperatures for molecules in thermal equilibrium with the environment. It is advantageous for the emission that the spectral line coincides with the maximum of the black body spectrum.

      • Pekka
        Yes I agree that the 15um wavelength is readily absorbed and have previuosly said so.
        All sources I have found say this photon energy has been transferred to rotational and vibrational modes.
        However the With 10^10 collisions the energy will not be confined to the absorbing CO2.
        I don’t think that anything I have said on this question is particularly from a “sceptical” viewpoint.
        I thought it was pretty much accepted as the way it worked otherwise how on Earth do you get a “greenhouse effect”.
        To make emission more likely than my 2% calculation you would need to show that probability of emission>than the probability of thermalisation.

      • Bryan,
        You are correct in that the energy is not confined to the absorbing CO2, but the collisions with other molecules maintain the occupation of these exitation levels of all CO2 molecules at such a level that the emission is strong.

        The blackbody radiation at 15 um would not be strong unless all exited levels capable of emitting at 15 um would have a high occupation. This is the connection to the shape of the black body radiation spectrum.

        Another way of expressing this same fact is that the actual emission strength is the product of emissivity and the strength of black body radiation at that particular wave length. For tropospheric CO2 both factors are large at 15 um.

      • Pekka
        CO2 molecules at such a level that the emission is strong.
        How strong?
        You will need to quantify.
        I have given you hard numbers from my calculations for that wavelength at that temperature.
        Are my calculations wrong?
        If so point out the mistakes.
        For every 100 absorptions how many emissions?

      • Bryan,
        The number of absorptions is proportional to the intensity of radiation. The number of emissions is determined by the temperature and emissivity of CO2-molecules. Looking in this way on cannot tell how many emissions are for 100 absorptions.

        For telling the ratio of emissions to absorptions one needs to know the intensity of radiation and the temperature. These values are, however, linked because the level of radiation is determined by the temperature of CO2 in a relatively small neighbourhood. If the temperature and CO2 concentration are equal throughout this neighbourhood, there will be exactly 100 emissions for 100 absorptions, although there is (almost) no connection between any particular pair of absorption and emission.

      • I just add that 2% is a high occupation level for an exited state. There is no need to show that it would be any higher.

      • Pekka

        My point is that the de-excitation is most likely to follow thermalisation rather than much more unlikely emission.

      • Bryan,
        De-exitation is indeed very likely, but so is also re-exitation, which leads with certain probability to emission.

        When 2% of CO2 -molecules are always in an exited state than can release its energy as radiation, there will be a lot of emission.

      • Pekka

        ……” there will be exactly 100 emissions for 100 absorptions”….

        Do you realise that if this were the case there would be no thermalisation of the atmosphere and hence no “greenhouse effect”

      • Bryan,
        I realize that very well. Therefore I wrote to my message 0f 6:55 am that the number would be the same, if the temperature would be constant throughout the neighbourhood. In the real atmosphere the temperature is not constant but decreasing with altitude. This is the way all this is connected to the greenhouse effect.

      • Pekka

        Here’s a nice graph from the top of the tropopause looking down.
        It shows the large “bite” taken from the blackbody envelope of the themalisation around 15um

      • Bryan,
        That graph shows that the CO2 emission corresponds to the thermal equilibrium at the temperature of 220K at the top of troposphere while the radiation at 10-13 um corresponds to almost 290 K of the earth surface.

        All this is in full agreement with everything I have written in this thread. From the space one can see only to the top of troposphere at 15 um, but all the way to the surface at 11-13 um.

      • What I see around the 260K line( nearest to the temperature of 243K) is a large part of the radiation around 15um “missing” i.e. thermalised.
        This is is exactly my point.
        However I think you will agree that we have been going round in circles not moving the topic on.
        It would be a good point to leave the discussion as it stands.

      • Bryan,
        I agree that this goes to detail and it is time to stop.

        Without going further to any of the mechanisms I just add that there are three important temperatures occuring in these discussions:

        The average temperature of the earth surface: about 288 K.

        The effective temperature of the earth as seen from the space: about 255 K

        The temperature at tropopause: about 220 K.

        The temperature of the troposphere spans the whole range from 220 K to 290 K.

    • “If the atmosphere was transparent to terrestrial radiation then it would not radiate at all”

      It’s revealing that one would even think to question such a statement. But then I’ve also see a corollary oft repeated that such an atmosphere would still have an adiabatic lapse rate sans radiative energy removal from the top of the atmosphere chilling off all those excited molecules.

      Photon emission by CO2 in an excited level is a classic example of spontaneous emission. Whether a molecule has been in this level 1 yr or 10^-12 sec does not alter the probability of its emission at any point in time. This was described early on by Einstein with his A coefficient and bothered Uncle Albert no end. Feynman’s QED provided an explanation and unified electromagnetism (CO2 decay) and the weak force (radioactive decay), or so I’ve been told. Just a mention here as the basics of the spontaneous emission involved in the “greenhouse” effect gets scant mention.

  76. In response to my point –
    ……”a) if a gas can absorb at a given wavelength it can emit at that wavelength”…….

    Bryan said, on December 4, 2010 at 5:08 am:

    You would need to be careful that this did not imply that for every 15um photon absorbed by a CO2 molecule we expect a 15um photon to be emitted.

    Why do I need to be careful?
    You need to be careful.
    You are still confusing emissivity and emission.

    Emissivity is a material property.
    Emission is determined by
    –a) Planck’s law (radiant intensity as a function of wavelength and temperature of the body) in conjunction with
    –b) emissivity (also a function of wavelength)

    • …”a) if a gas can absorb at a given wavelength it can emit at that wavelength”……

      Thanks for the clarification.
      Some people interpret the above “can emit” as “must emit”

  77. Leonard Weinstein

    Science of doom and others,
    I keep seeing a big oversight in some comments. The fact that the ground is hotter due to atmospheric greenhouse gas effects DOES NOT mean that there has to be more heat transfer by convection or any other means. On the average, only solar radiation heats the ground (obviously at night and at the winter poles this can reverse, but I am talking about on the average). This is ALL the heat that has to be removed. When the greenhouse gases result in both an increase in ground temperature and lower atmosphere temperature, the increase ground radiation is balanced by an increased back radiation, but heat transfer is only due to the DIFFERENCE. There is some radiation heat transfer out through the transmission window, and some net radiation flux up if the absorption distances are reasonably large (due to the lapse rate). The radiation UP plus convection heat transfer carry all of the solar energy to be radiated to space. The only way increasing the atmospheric greenhouse effect causes a need for increased convective heat transfer is by REDUCING the radiation contribution. Making the transmission window smaller, or making the DIFFERENCE between outgoing and back radiation smaller causes a slight increase in convective transfer.

  78. Dear Pat Cassen,

    Some late reflections:

    — ‘energy minimization’, I think, is not a hypothesis, but one of the most basic principles of physics.

    — Miskolczi’s 2007 Eqn 7 is not a ‘phony’ energy conservation equation but a very clear expression of the limit of available incoming energy. You may not see its physical basis but this does not mean it does not have. Furthermore, empirical evidences sharply support it (g=1/3, see my website for context).

    On the constancy of tau : There is of course a dependence of the greenhouse (and transfer) functions on tau; it is not ‘prefixed’ in the theory. Its constant value comes from further considerations and experiments. My favourite example is the constant body temperature of warm-blooded animals, e.g., of humans (36.7 C). In theory, it depends on a whole lot of things, from the outer temperature to the humidity of the body, etc. But as an empirical fact, it turns out to be stationary. One may then speculate what physical principles make it so stable, fluctuating exactly around its theoretically predicted constant …..

    • Miklos – Thanks for your comment.

      Regarding eqn (7) of Miskolczi (2007): I stand by my original statement. One cannot derive this expression from any rigorous consideration of net energy flow. Readers will have to decide for themselves.

      I understand that tau is not ‘prefixed’ in Miskolczi’s theory. I challenge the derivation of its constancy.

      • Dear Pat:

        Miskolczi’s 2007 Eqn (7) is one of the most delicate issues in his work. I would suggest to try this approach:

        All starts with his Aa=Ed equation.
        Substitute this into (1) and (2) of M2007, which are the well-known energy balance equations for the atmosphere and the surface.

        As a result, the atmospheric balance equation becomes Eu=K+F +P,
        and the surface equation now is Su-OLR=Ed-Eu (=G),

        where G is Ramanathan’s 1989 greenhouse factor (G=Su-OLR). Now we have a new epression for it on the right-hand side too.

        In this equation, definining the greenhouse factor, we have now two nett IR fluxes. The first is the ‘old’ one (Su-OLR), it is a downwelling one. This turned out to be equal to another one, Ed-Eu, an upwelling one. The first heats the ground with a nett energy of G, the other cools it (or heats the atmosphere by the same amout). Always have in mind the Eu=F+K+P relationship too.

        The presence of these two nett IR fluxes in the air is a result of the IR-acitve gases. Without them, they would disappear. The only source of these two fluxes is the incoming available energy, Fo+Po (=OLR). So their sum cannot be higher than that. The equality expresses the maximum. This is shown by the (Su-OLR)+(Ed-Eu)=OLR relationship.

        The equation leads directly to g=G/Su=1/3, which equals to the observed empirical greenhouse factor.

        But this is only one-quarter of the whole story–or, I would say, only the ‘first-order approximation’…

      • Errata:
        I wrote:
        “In this equation, definining the greenhouse factor, we have now two nett IR fluxes. The first is the ‘old’ one (Su-OLR), it is a downwelling one. This turned out to be equal to another one, Ed-Eu, an upwelling one. The first heats the ground with a nett energy of G, the other cools it (or heats the atmosphere by the same amout). ”

        I meant:
        In this equation, definining the greenhouse factor, we have now two nett IR fluxes. The first is the ‘old’ one (Su-OLR), it is an upwelling one. This turned out to be equal to another one, Ed-Eu, a downwelling one. The first heats the atmosphere with a nett energy of G, the other cools it (or heats the ground by the same amout).”


    • Miklos, i would like to thank you very much for your participation in this discussion

  79. My last sentence above (“One may then speculate what physical principles make it so stable, fluctuating exactly around its theoretically predicted constant …..”) is of course not about the 36.7C, but about tau=1.87 … :-)

    • Miklos,
      would you agree that the constancy of the relationship speaks of the validity and accuracy of the radiosonde data too?


      • tallbloke,

        The 61-year NOAA NCEP/NCAR reanalysis has a global average water vapor about 2.6 prcm. This fits pretty well with other reanalysis data, and also with the TIGR2 weather balloon data set. On the contrary, USST76 with appendix B of Liou 1992 contains only about 1.4 prcm. The latter is the basis of KT97 (=IPCC 2007) energy budget. So I think the NCEP/NCAR archive, contrary to its known early inconsistencies, is still much better than the USST76–which was good enough for the IPCC.

        Further, Ferenc repeated his calculations on several different time periods in the 1948-2008 range, and, as far as I konw, they all fit very well to the long-term constant — which, in fact, equals to Ferenc’s theoretically predicted stationary value.

        So I must say again: although I read interestingly here all the views, beliefs, hypotheses, interpretations, explanations or even speculations on the greenhouse effect, I think since Galileo natural science is about numbers. That’s why I try to encourage every radiative experts here to present their own numbers on the longwave aborption, downward radiation and window radiation — the basic components of atmospheric greenhouse effect. That would be the ground when a really meaningful discussion could start.

        One more note, please. It is often said that with increasing CO2 the water vapor amount must decrease to support Ferenc’s constant. No, the stability of tau can be maintained by changes in the distribution of water vapor and temperatures also. If there is really a physical contraint on it (as Ferenc claims to prove) then the system has just enough degree of freedom to accomodate itself to this limit, and will ‘know’ which one is the energetically more ‘reasonable’ (satisfying all the necessary minimum and maximum principles) to choose.


  80. Wrangler Wayane

    I read that the Earth’s surface is warmed just by the incoming sunlight and
    the air. But isn’t the heat from within the Earth also providing quite a bit of warmth also? As the air’s molecules absorb the radiation from the Earth’s surface, doesn’t this energy then cease to be heat, just molecular vibrations?
    At this point, how can it contribute to any moderated greenhouse effect?Also, isn’t the magnitude of the convection transport of heat far greater than
    any moderated greenhouse effect?

  81. Lets target an explanation at an audience that has taken 1 year each of undergraduate physics and chemistry, plus calculus.

    I’ve been trying for months to put together a simple computer simulation here and there. And there too. I wasn’t trying to model a whole planet. Just a patch of ground with some atmosphere above it. No need for calculus. Didn’t get very far though. I’m astonished that there aren’t dozens of such little DIY simulation models around, to demonstrate the elementary facts of heat transfer, in this our modern computing age. Nobody seems to do it.

    Maybe I’m a bit odd….

  82. cant find where i posted my last question, your kind answers to which i have finally understood.

    Another question:

    When i look at the graph that started all this,

    it is apparent that Mother Nature has ample power to control CO2.
    She does it every growing season, observe the the curve year to year.
    She lets it build up in winter then bashes it back down every spring.

    Look at [i]slope[/i] of that red curve, maybe one of you folks can get data from the site and plot the curve’s first derivative… then compare width of positive and negative half-cycles.

    Looks to me like a slight widening of the seasonal negative slope intervals would correct the slight upslope of the average curve(black one).
    That would just be shorter winters and longer spring/fall seasons. Anecdotal evidence suggests she’s doing just that already.
    How much longer, at current rate, before she’s arrested the increase?

    Now – these climate model programs are obviously complex and beyond my comprehension. But i am not hearing any control system theory discussion from the support side , just arguments over what’s feedback and what’s not. I am skeptical of just how well the basic ‘modelling’aligns with reality.
    Having worked in simulation, i assure you it is easy to build a huge math model that is just wrong. Equations are mostly linear, while Mother Nature throws at us things like phase change which are not. Generally she’s arranged for the nonlinearities to provide system stability, eg ice floats otherwise the oceans would be shallow with a floor of sunken ice.

    Mother Earth being an equilibrium seeking system subject to math of closed loop systems , where’s the bode plots??? . Looks to me like she has about 45 degree phase shift (solar insolation to CO2 slope) which at first glance looks like a pretty stable system…

    when i add this to my non-scientific impression of the despicable vermin (Goldman, Soros, Gore, Congress) who are setting up to cash in on a carbon trading market, i get outraged at the blatant and blasphemous misuse of science.

    so your point, try to put out something for the people who passed freshman physics, i applaud. We outnumber the PHD’s and can carry the conversations to the lunchrooms.

    If Al Gore ever passed a physics course it’s news to me. His degree was in political science, and in these times i dont know if that term is more oxymoron or irony.

    thanks, and sorry for the rant. old jim .

  83. i finally read, and understood, a couple links that somebody here referenced.

    had to get myself up to the point could handle them. Thanks for your patience.

    To folks like me, who had a physics course and had real doubts about the integrity of the AGW crowd, i recommend those links, repeated below.

    Appreciating the history helped me. I’d thought Asimov was the first.

    old jim

  84. Dr Curry, I have read the “Sky Dragon” and as a layman found it fascinating.
    On several points I now have trouble accepting standard Greenhouse theory and am wondering wether you can give satisfactory explanation.
    The surface of the moon is on average about 100 C, “Sunnyside up” and -150 C on the “Dark Side”. Both the earth and moon would receive similar watts per sq meter from the Sun. So its obvious our atmosphere moderates these extremes, greenhouses do not, so how do we have our atmosphere as a greenhouse when it is far cooler when sunlit and not so cool when dark ?
    It appears our atmosphere ‘refrigerates’ more than acting as greenhouse.
    (I am not satisfied with ‘averaging’ as this may be useful for programers not natural systems.)

    In regard to the so called greenhouse affects of water vapour, why is it cooler in the humid tropics than equatorial deserts if H2O is such a strong greenhouse gas ? (and yes it delays cooling at night but it doesn’t ‘create heat’ )

    In regard to the so called greenhouse effect, how can radiation being absorbed then re- radiated create heat? Surely it can only delay cooling not make things hotter. ie, the hours of warm temperature should be extended but not increased.
    After all blankets (that is a constant AGW analogy) don’t create heat they only slow cooling and if the reradiating heat makes things hotter, why doesn’t this happen in a thermos ?
    If this reradiated heat can actually make atmosphere warmer, why isn’t this principle being used in engineering ?

    Thanks for your site and consideration.

    • I am as skeptic as it goes, but really, attacking GH effect (very badly named) with those types of arguments is counter-productive…Sure, GH effect does not “create heat”, it only slow the cooling down. Now which maximum T do you think is attainable on earth thanks to good old sun, if you “jut slow cooling down”? A hint: stay in your car on a sunny day, stop the engine and the airco, and wait a few hours…Well,maybe not too long, if you want to survive the experiment.
      It’s not the same insulation mechanism, but it prove that insulating can be a very efficient way to increase T, when the absolute T_max that can not be exceeded is the emission temperature of the sun (+- 4000K iirc). Basic GH effect is quite simple, even if computing the value is tricky (line by line computation due to IR freq dependence, unsteady regime due to earth rotation). It’s the feedbacks, the treatment of the troposphere, and maybe the estimation of CO2 half-life, that are the shaky parts of AGW…oh yes, and the c in cAGW is not at all obvious either…

      • kai

        “…Sure, GH effect does not “create heat”, it only slow the cooling down”

        I agree with that, but the 33C effect claimed by IPCC advocates is totally attributed by them to radiation.
        The other insulating properties of the Ocean and Atmosphere are ignored.

      • The problem is that the description of the GH effect is really poor, even the one that is used for vulgarisation by mainstream climate science: it is based on a simple layer shell which concentrate all the radiative effect of the atmosphere and forget about other heat transport mechanisms. I have devoloped another “simple” model (see Pierrehumbert discussions) which do not completely forget about convective/latent heat transport, but still, it is an approximation. Only one I think is a little bit closer to what happen, while still manageable as a simple mental model.

        This inadequacy of the most popular explanation of the GH effect is sad, because it becomes suspicious as soon as we start thinking about it. But still, it is right on one thing imho: the GH effect is really responsible for the fact that average ground temperature is above what we should expect from a simple perfectly conductive grey-body sphere model. There really is a a radiative insulation effect that is the main driver, because above troposphere, the main effect is radiative heat exchange and the rest can mostly be ignored. This radiative layer act as a boundary condition to the more complex effects at play in the lower layers. On can not extend the radiative model up to the surface, so it is not sufficient to predict T (contrary to what simple greenhouse models pretend sometimes, but the wavelenght-differential radiative insolation above troposphere is really there, and thus a T increase from CO2 increase should be expected. How much ground temperature will be affected is dependent on much more (feedbacks in the troposphere level, in H2O aboce troposphere, and albedo changes, mainly), and thus not predictable by simple models, but still, as a skeptic, I do not doubt that if CO2 increase, T will increase. Enough to be measurable, I doubt, enough to be main factor, I doubt more, and enough to be detrimental, I doubt even more. Climate science has not relaxed my doubts, on the contrary ;-)

  85. Kai, you state, “I am as skeptic as it goes, but really, attacking GH effect (very badly named) with those types of arguments is counter-productive…Sure, GH effect does not “create heat”, it only slow the cooling down. ”

    I don’t care about politics so am not interested in “productive or counter-productive.” I seek to understand and your analogy as a response ?
    Arrrggghhhh, what is it with AGW? If the explanation is not complicated enough to be impossible to understand, its rubbish, yet simplistic 2 dimensional analogies abound like duck poo around a pond !
    The car analogy is the poorest I have heard since I had to endure listening to Stephen Schneider compare our atmosphere to a bath.
    Cars heat from lack of conduction and convection due to heated air being trapped, GHG’s in a “glasshouse” have no real bearing on temperature at all.
    Its not trapped radiation that heats the car, it’s simply heated air that is trapped.

    My analogy would be, “sit in a convertible automobile and place a chain link fence 20 feet above it and measure the increase in temperature of the “chain link insulator.” A hint; stay in the convertible and see if you can survive the boredom.

    But as for your insular analogy, we already have one, the near vacuum of space is an excellent insulator, as it thwarts conductive heat loss, and we don’t cook like a car, so how do GHG’s provide this miraculous insulation for radiative transfer? How does it ‘blanket’ radiative loss ? ( I can accept some but 33 degrees ??? Pull the other one ! )

    More critically how does re-radiated heat make its source hotter ?
    If I put a reflector against a heater, will it increase the temperature ?
    Of corse not, so how do GHG’s ? Remember its not hot air being insulated as you assert with a car, its IR radiation that is absorbed with a smaller percentage reflecting back to its already warm not cold source.

    You have given no satisfaction to my query at all. The notion the GHG’s keep the earth from freezing by providing a staggering 33 degrees extra does not wash.

    If the earth had just an oxygen/nitrogen atmosphere devoid of so called GHG’s, it would quickly warm the “Sunny side”. The Sun would still warm the earth and the earth would still warm the atmosphere.
    Sure the lack of H2O as a GHG would allow quicker cooling but then the lack of clouds would also cease their cooling effect in reflecting radiation back to space.
    I am not saying it would be quid pro quo but to ascribe all of the 33degrees to CHG’s is only possible to a theory that is obsessed and blinded by the alleged power of a trace gas.
    (This theory would be rightly obscure except for its political currency in demonising CO2.)

    In other words, surely a climate without GHG’s would be like deserts, hotter during day, colder at night but any notion of a ‘snowball earth’ effect that is implied for a lack of GHG is risible.
    As a moderator of temperature however the action of GHG’s holds weight, as in my example of the tropics v’s desert .

    • OK, Peter, I will try again….
      “Cars heat from lack of conduction and convection due to heated air being trapped, GHG’s in a “glasshouse” have no real bearing on temperature at all.
      Its not trapped radiation that heats the car, it’s simply heated air that is trapped.”

      No. Cars heat because they absorb the radiation emitted by a much hotter body: the sun (at least when engine is stopped, it is the main source of heating on a sunny day – by far). Like the earth.
      Then, once heated up (please don’t play on words, I know it is a simultaneous process), they cool down. By various processes. Conduction, radiation, and mainly, convection (natural, or forced). lower the efficiency of any of those mechanisms, and the car will cool down less. As we are discussing a steady state, it will reach a higher equilibrium T, the one at which the reduced cooling mechanism will again remove as much heat as sun radiation bring in.

      Now, maybe you have a problem with cars, I dont’t know (personally, I like them, that’s maybe why I have a problem with many current cAGW proponents). No problem.

      If you live in a currently cold area, a very simple experiment: just bring your hand in front of a wall, and move it in front of a closed window. Feel the quick cold sensation? I guess so, at least if it is cold enough outside. Why? conduction? nope, too fast for air conduction to really quick in. Convection? Nope, ditto. Radiation heating? Maybe, but both the wall and the window are colder than your hand, so they both should not warm you, according to the “Backradiation is unphysical/ a cold body can not heat a warm body” school. But classic interpretation explain it perfectly: at all time, your palm radiate (thus loose) heat. In front of the wall, the lost heat is partially remade by the (lower) radiation coming from the lukewarm surface. In front of the window, less so, because the radiation is less (coming from a colder outside).

      Another thought experiment: Imagine a heater in a hermetic fully transparent sphere, where conduction/convection/your prefered heat conduction mechanism is so high than for all practical aspects T is uniform. We inject 1W of energy. Outside the sphere there is vacuum, and then a bb surface of uniform temperature T_out. Will T_in be lower, higher or equal to T_out?

      Now imagine doing this for T_out1 and T_out2, and measuring T_in1 and T_in2. If T_out1>T_out2, will T_in1 be lower, equal or higher than T_in2?

      If you have aswered T_in>T_out for 1), and T_in1>T_in2 for 2), we agree…and you agree to basic radiative transfer following S-B law (there is either a back radiation from the outside to the inside…or the inside body somehow know the Temperature from the outside body through vacuum…by some kind of EM wave, what else?..coming from the outside to the inside…yes, back to backradiation.

      Now, if you agree that, till now, we do not have specified how the 1W of heating power is provided, and that it could as well come from a hotter body radiating in shorter wavelength (the sun), you are dangerously close to classical GH effect description.

      I do not agree to the 33K greenhouse, the treatment of the troposphere, and feedbacks evaluation, not to say anything about the relevance of GCM (mainly because I do not really knwo what kind of PDO – if any- they solve, only thing I know for sure is that it is not Navier-Stokes with added radiative integral equations). But that a greenhouse effect should increase the ground temperature if an IR absorbing medium is added between sun-heated ground and space, that do not doubt…

      • Kai, you say “ .. But that a greenhouse effect should increase the ground temperature if an IR absorbing medium is added between sun-heated ground and space, that do not doubt… ”. Please would you clarify how the “greenhouse effect” can ADD (create?) the energy necessary to “increase the ground temperature” above what it would otherwise reach without the presence of that “greenhouse effect”. I was under the impression that “energy can neither be created nor destroyed” – something to do with an ancient Law of Conservation of Energy – but that is something that I was taught in the 60’s. Perhaps the laws of physics have changed since then.

        Best regards, Pete Ridley

      • Pete,
        Have you ever heard that additional insulation makes a house warmer? There is nothing more controversial in the the fact that greenhouse gases can make the earth warmer – or at least there should not be.

      • Pete,

        It’s not adding energy, just slowing down the rate at which energy leaves the surface.
        It’s just like a pot with a lid gets hotter on the same stove than the same pot without a lid.

      • Yes, I have tried to tell that many time with different examples and words…but still no effect :(

        There is no added energy, no real heating from GH gas and induced downward radiation.
        The real heating is from the sun, wich is much hotter and thus have no entropy trouble transfering all the heat you want to earth.
        The cooling is from space, wich at 4K is again perfectly able to absorb a lot of heat from earth by radiative transfer.
        The GH gas are an insulator w.r.t radiative transfer.
        Pete, if you do not get how an insulator inserted between a heated object (earth, heated by the sun) and a cooling medium (space) can increase the T of the heated object, compared to the T it will get with the same heating, same cold object, and no insulator, I see no hope of mutual comprehension…but I see a lot of commercial opportunities: Pete, do you have some good insulator (winter clothes) to sell on Ebay? Useless to keep you warm, so can I expect a good price? ;-p

      • But it does not runaway or burn up becuase of the lid, does it?

      • Kai, its still so unsatisfactory.

        I stated in my first post that I am well aware that all a GHG’s can do is slow the rate of cooling.

        I am still stumped as how that causes warming.

        It would indicate that a daily top temperature would “hold” for a slightly longer time but “increase temperature” ???? It can’t raise temperature and it certainly cant drive warming.
        Analogies of cars or housing insulation don’t mimimic atmosphere at all, unless as I have stated it’s a convertible or an insulated house roof with half the roof missing.

      • peter (laux), I was struggling with the same concept too, around 1st Feb. but after reading an E-mail from Roger Taguchi late last evening I had another read of relevant comments here. One by RB used an analogy that I, as a retired electronics engineer, could relate to. Here are relevant extracts from the response that I sent to Roger a few hours ago and maybe it will be of use to you – until an “expert” like Roger comes along and points out that I have misunderstood the physics or overlooked something else.

        Most of the exchanges on Judith’s “Physics of the Atmospheric Greenhouse (?) effect .. ” thread use scientific explanations that are way outside my experience but I can usually follow the reasoning if not the detail, however, there is one issue that I could not get my head around. There were numerous contributors claiming that as a result of the “greenhouse effect” the earth must get warmer than it would without that back-radiation from those gases. I don’t recall whether or not Roger made a direct comment about that but I could not visuallise the “greenhouse effect” (back radiation?) doing anything more than simply slowing down the earth’s cooling rate by hindering the radiation back into space some of the energy received from the sun (I talked about this on 1st February at 6:01 am).

        On 3rd December at 4:05 pm. RB uses as an analogy the current (radiated energy) through a resistor (the atmosphere) requiring a higher voltage (temperature difference between earth and space) to maintain current flow if the resistor increases in value (more greenhouse gases). I saw this as being a flawed analogy because it assumed that the flow through the resistor (earth to space) is from a constant current generator. The correct analogy is that of a charged capacitor (the earth stores energy received from solar radiation) discharging through a resistor (the atmosphere) to space. Increasing the resistor (adding greenhouse gasses) simply reduces the discharge rate. It does not increase the voltage across the capacitor (higher earth-space temperature difference).

        I initially represented this in simple symbolic form, assuming that the Earth has a capacitance C (unchanging), the atmosphere a resistance R0 without GHGs and R1 with GHGs. I then represented the sun as a voltage E and space as the “ground return”. I inserted a c/o switch to represent night and day, discharging C through R0 or R1 during night-time and charging C during daytime.

        At that point I recognised that my analogy had an incorrect component, the switch, because E is never switched out of circuit (the sun continuously sends energy to the earth). Also, I had missed a component, R3, between E and the capacitor. I then understood the point of the argument. The combination E and R3 can be represented as a constant current source (constant energy from the sun impinging upon the earth. The circuit is now simply a constant current source (the sun) charging a capacitor (the earth) across which is a series arrangement of a fixed resistor R0 (the N2 & O2 components of the atmosphere) and a (very slightly) variable resistor R1 (the greenhouse gases).

        I’m no longer puzzled about that part of the “greenhouse effect” argument which claims that increased greenhouse gases increases the earth’s temperature (until someone comes back and shoots down my electronics engineer’s analogy) but it doesn’t help all that much because its only a very simple model and ignores all those other nasty complicating passive components like conduction, convection, latent heat, etc. (resistors) helping to discharge the capacitor. On top of that there are those equally nasty reactive components evaporation and precipitation (capacitors) and the movement of that charge from one part to another (inductance L), as well as the fluctuations in the “constant current” source itself. Oh dear, my head’s starting to hurt again! – and I’m keeping it simple. You “experts” must get terrible migraine trying to sort this all out, then you have to explain it to us lay people, who are mainly reponsible for voting those even less knowledgeable politicians into power.

        Best regards, Pete Ridley

      • Thanks Pete but I still cannot grasp that warming occurs due to GHG’s, as you and I have said, only delayed.

        I have spent nights in the Australian desert in blistering heat and insignificant humidity and have then frozen at night, the 385 ppm of CO2 didn’t warm me at all, I have then gone to the tropics (Darwin) and the humidity DID stop me being roasted and kept the temperature to around 30degrees C but then that assured me a restless night in similar temperature and near 100% humidity. The CO2 “blanket” in the desert sure didn’t do that.
        The H2O as a greenhouse gas didn’t heat Darwin to hell, it kept the atmosphere cooler than the atmosphere in the desert and stopped it being 45+ degrees , then kept it warm at night, unlike the desert freeze.
        So it appears H2O is profound in its action, in keeping atmosphere cool by day and keeping it warm at night but CO2 ?
        I keep thinking of the analogy of a “bath of coffee with one grain of sugar – you now increase the sugar by 100% , what changes ?”
        As a layman this whole CO2 driver thing seems like an extra grain of sugar trying to sweeten a bath of coffee. Does it really sweeten it in the first place let alone provide extra sweetness? In other words, yes we know the sugar is sweet but it cannot influence the bath .

        Now with yourself as an electronics engineer and with all the talk of “2nd law of thermodynamics etc”, I was wondering “how does under 400 parts per million or even worse the 4% attributed to industry have the “energy” to drive a vast chaotic climate system ?
        I was wondering if you know of any work that has calculated the “energy” (in terms of watts or sweetness as I would analogise) that 400ppm of CO2 could contribute, or of all “greenhouse gases” for that matter ?

        p.s. pete, for your entertainment ! Its on how little CO2 we produce industrially as opposed to our and livestocks breath.

      • Hi Peter!
        I’m delighted that someone with personal experience can attest to the fact that deserts can be blistering hot during the daytime, and freezing cold at night. In my article “Net Feedback in Global Warming Calculations” (which Pete has a pdf copy of, sent in Dec. 2009), I calculate just these facts in Appendix 1.
        The reason is that water vapour is a powerful greenhouse gas, in fact contributing more to the greenhouse effect than CO2. The difference is that water vapour can be varied (drastically reduced over desert areas and the poles) whereas CO2 cannot condense out (sublimate, if you want to be picky). From the areas of water vapour absorption “bites” taken from the 288 K black body spectrum emitted by a warm Earth, the power/m^2 absorbed by water vapour is 71.2 W/m^2 (compared to the 3.7 W/m^2 everyone is haggling over when CO2 increases from 300 to 600 ppm). I calculate that at latitude 25 degrees from the Equator, this means a temperature swing of the average from 10 Celsius at night to 38 Celsius in the daytime (from a minimum of 2 Celsius to a maximum of 46 Celsius). Because CO2 absorbs 56 W/m^2, the temperature extremes would be even worse on the Earth without the CO2. Hence CO2 is a greenhouse gas.

        You are right about high humidity moderating temperature extremes as well: during the daytime, a lot of heat from the Sun is used to convert liquid water to water vapor (water in the form of individual gas molecules, not droplets). This involves the very large Latent Heat of Vaporization. At night, this heat is liberated back to the air and the Earth’s surface on condensation (formation of dew or fog). In addition, cloudy nights are warmer than clear nights, as fewer IR photons emitted by the warm surface of the Earth can escape directly in straight lines to outer space.

        IMO it is the job of scientists to answer as clearly as possible questions asked in good faith. I hope I have at least partially succeeded.

      • Roger T,

        I’m curious about your co2 contribution number of 57 w/m^2, That is about twice what I’ve calculated for an atmospheric column. The h2o vapor value is only slightly off from my calculation which could be explainable by a difference in overall wavelengths as mine is limited to shorter than about 65 microns rather than a more typical value near 100 microns. As for a co2 doubling, my calculation is a 3.6 w/m^2 at the tropopause for an increase of 330 to 660 ppm so again, it is quite close to accepted calculations. These are done using a 1-d model.

      • Roger many thanks for your reply.
        I was hoping you could help with my original query which I have had no satisfactory answer,

        “The surface of the moon is on average about 100 C, “Sunnyside up” and -150 C on the “Dark Side”. Both the earth and moon would receive similar watts per sq meter from the Sun. So its obvious our atmosphere moderates these extremes, ‘greenhouses’ do not, so how do we have our atmosphere as a greenhouse when it is far cooler when sunlit and not so cool when dark ?
        It appears our atmosphere ‘refrigerates’ more than acting as greenhouse.”

        Perhaps ‘moderates’ is a better term than refrigerates but what intrigues me is how its cooler at the earths surface than the moons.

      • Hi Peter, cba and Pete Ridley!
        For Peter, the Stefan-Boltzmann Law says outward flux (in W/m^2) equals emissivity times 5.67 x 10^-8 times T^4, so using your values for the Moon, the Moon’s surface during the daytime would emit 1098 W/m^2 using emissivity = 1, and only 13 W/m^2 during the night (the factor of 85 difference comes from the fact that T^4 is involved: (373/123)^4 = 84.6). [ There’s something not quite right about the daytime and nighttime temperatures given, as the Solar insolation at the Earth’s distance from the Sun is 1371 W/m^2, but this is for a circular cross-section. When this insolation is spread out over a sunlit hemisphere, this is halved to 686 W/m^2. So maybe the daytime high temperature on the Moon is too high, considering that the albedo is not zero, so only part of the insolation should be involved in heating the surface during the daytime.]
        In any case, can we say that the Moon’s average (mean) temperature is about 248 K = -25 Celsius?

        We’d expect the Earth to be at this temperature if it were like the Moon. The Earth’s temperature range is moderated by the large heat capacity of the oceans, and by its more rapid rotation.
        Because N2 and O2 (the main components of dry air) are transparent to IR radiation, they provide no “greenhouse effect”.

        [To cba: I got my figures for CO2, H2O and ozone absorption by tracing the spectrum in the MODTRAN simulation available at and measuring the areas of the relevant peaks taken as “bites” out of the 288 K black body curve which is the smooth upper envelope. The CO2 absorption at 667 cm^-1 corresponded to 42 W/m^2 (when the entire area under the 288 K black body curve corresponds to 383 W/m^2). Because the IPCC and others incorrectly interpreted the truncation as “220 K black body radiation”, I calculated the P-branch absorption profile for CO2 absorption for a series of different CO2 concentrations, and deduced that the IPCC chopped off another 33% = 14 W/m^2, which when added to the 42 gives 56 W/m^2. I concede that my decision as to how the CO2 absorption profile overlaps with the nearby H2O pure rotation (no change in vibrational quantum number) spectrum could have introduced some error, but if my estimate for CO2 absorption is too high, then my other results are even better in refuting the IPCC estimate for AGW. Please read my Feb. 22 posting (at the end of this thread) regarding the IPCC truncation error.]

        Back to Peter’s question: A 288.2 K black body of emissivity 0.98 emits 383.3 W/m^2. This is emitted from the solid and liquid Earth, and would all escape to outer space during the nighttime and the daytime unless some of it were absorbed by greenhouse gases such as CO2 and H2O AND THE ENERGY TRANSFERRED DURING RADIATIONLESS COLLISIONS TO N2 and O2 molecules THAT CANNOT RE-EMIT IR (infrared) RADIATION, since they are homonuclear diatomic molecules with no permanent electric dipole moment that can interact with the changing electric field of an electromagnetic wave (photon). According to the MODTRAN graph, 260 W/m^2 escapes at altitude to outer space. This means 383 – 260 = 123 W/m^2 has been absorbed by greenhouse gases and transferred to the air (mainly N2 and O2).

        The greenhouse effect is the temperature difference that this 123 W/m^2 would correspond to. In my Feb. 7 posting (near the end of this thread), I show that the temperature sensitivity is 1/4 of that used by the IPCC. The general rule is that the %temperature change of the emitting surface is 1/4 the % change in the flux (W/m^2 emitted outward). Therefore if we substitute T=288.2 K, “delta j” = 123 W/m^2, and j = 383.3 W/m^2, we get “delta T” = 23 degrees. That is, without the greenhouse H2O, CO2 and ozone, the Earth would be 23 degrees Celsius cooler than 288 K.

        We can readily check this value for the “greenhouse effect”: Substitute T = 265 K, emissivity = 0.98 (unchanged) into the Stefan-Boltzmann Law, and we get j = 274 W/m^2, close to the 260 W/m^2 escaping to outer space at altitude. [The rule for % changes is strictly speaking accurate only for small differences approaching differentials.]

        To cba: I contend that the CO2 absorption is 14 W/m^2 more, meaning an even greater greenhouse effect. This means, however, that the mean temperature of the Earth is closer to 285 K than to 288 K, and this appears to be the case when you look more closely at the MODTRAN graph: a 285 K black body curve would fit more snugly to the observed spectrum as an envelope than the 288 K black body curve, which leaves a bit of a gap (and may have contributed to my deriving a slightly higher base value for CO2 absorption).
        To cba: you seem knowledgeable about CO2 absorption, so I’d appreciate it if you would check my Feb. 7 posting on Temperature Sensitivity for any mistakes. It’s the third-last major posting in this thread.

        To Peter and Pete Ridley: I hope this helps answer your original questions, which perhaps others were unable to do.

      • Further to the Moon’s daytime output at 100 Celsius = 373 K:
        the 1098 W/m^2 is 80% of the Solar insolation of 1371 W/m^2, so I assume it means an albedo of 0.20 for the Sun directly overhead. Thus this high temperature can be achieved on the Moon’s surface, since it is a local maximum, not the mean daytime temperature for the whole hemisphere.

      • Ta Roger, you have ensured me a mental cluster**** for a day or two why I process your response !!!!!

        It is greatly appreciated.

      • Roger Taguchi

        Hi Peter and others!
        I contend that due to IPCC truncation error, there is 14 W/m^2 more absorbed by CO2 in the Earth’s atmosphere, meaning that the flux escaping to outer space is only 246 W/m^2 instead of 260 W/m^2.

        The temperature of a black body emitting 246 W/m^2 that ALL escapes to outer space (i.e. with no greenhouse gasses) would be the 4th root of
        246/[0.98(5.67 x 10^-8)] = 257.9 K = -15.3 Celsius.

        If we add 246 + 123 + 14 to get 383 W/m^2 (using 137 W/m^2 as the total flux absorbed by greenhouse gases AND TRANSFERRED BY COLLISION TO N2 and O2 THAT CANNOT RE-EMIT ir TO OUTER SPACE), then the Earth’s surface temperature would be the 4th root of
        383/[0.98 (5.67 x 10^-8)] = 288.1 K = 14.9 Celsius = 15 Celsius (approx.), the value stated on the MODTRAN graph. The greenhouse effect would then be 14.9 + 15.3 = 30.2 Celsius = 30 degrees (approx.).

        The envelope of the actual spectrum shown on the MODTRAN graph is a lot closer to 280 K than to 300 K, so 288 K leaves a bit of a gap. It also implies an albedo of 0.24, a little on the low side compared to literature values close to 0.30. If we shave 14 W/m^2 off the total emitted from the Earth’s surface (the amount of the IPCC’s truncation error), then only 246 + 123 = 369 W/m^2 is the total emitted from the surface, and the temperature would be the 4th root of
        369/[0.98(5.67 x 10^-8)] = 285.5 K = 12.3 Celsius, and the greenhouse effect would be 15.3 + 12.3 = 27.6 Celsius = 28 degrees (approx.). This slightly lower value would correspond to an albedo of 0.28, much closer to the literature values of approx. 0.30. So everything now fits with a total greenhouse effect of 28 degrees, and a mean temperature for the Earth’s surface of 12.3 – 27.6 = -15 Celsius (approx.) without any greenhouse gases.

        How does this compare with the Moon’s mean temperature?
        “The average temperature on the moon is about -5 degrees Celsius.” This would actually be warmer than our calculation of -15 Celsius for the Earth w/o greenhouse gases! So I don’t know where they got this.

        The article also says that the average temperature during the Day period is 107 Celsius (with a peak at 123 Celsius = 396 K, which would correspond to a flux of 1366 W/m^2, essentially the entire Solar insolation being absorbed by the Moon’s surface and re-radiated outward as 396 K black body radiation). The average temperature during the Night period is -153 Celsius (I don’t know how they got this figure; maybe from a nighttime IR spectrum or a bolometer reading of the black body radiation). The average of 107 and -153 Celsius is -23 Celsius, now LOWER than our calculation of -15 Celsius for the Earth without greenhouse gases. If we average the absolute peak daytime temperature possible on the Moon with the (minimum?) nighttime temperature , the average of 123 and -153 is -15 Celsius!

        So, given the range of possible values for the Moon’s temperature, we can say that the Moon and the Earth would indeed have the same mean temperature in the absence of greenhouse gases (as the physics of radiation balance must give). Therefore the greenhouse effect is real and profound for the Earth. In addition to raising the mean temperature, the greenhouse gas water vapor (H2O) also moderates the temperature swing from daytime to nighttime, as desert experience shows.

        So Peter, you show all the hallmarks of a good scientist: (1) keen observation of actual real-world conditions (facts) that require theoretical understanding: large temperature swings in desert areas, and the moderation of extremes in areas of high humidity, (2) a good, logical question re the Moon and the Earth (which others have not answered), and (3) the willingness to actually listen to a proposed answer, instead of arrogantly dismissing anything that does not conform to a really bad paradigm memorized without question during your previous training.

      • Roger Taguchi

        The albedo of the Earth is 0.28 (28%) in my best calculations. Because of the bright clouds, you’d expect the albedo for the Earth to be higher than that of the Moon, hence less Solar power/m^2 hitting the Earth’s surface, and therefore a lower mean temperature in the absence of greenhouse gases. So the Moon’s mean temperature could actually be -5 Celsius, higher than that of the Earth w/o greenhouse gases. Assuming emissivity = 0.98, the Stefan-Boltzmann Law gives the outward flux for a -5 Celsius = 268.2 K black body surface as
        0.98(5.67 x 10^-8)(268.2)^4 = 287.5 W/m^w. Multiplying this by a factor of 4 to include emission during the Lunar daytime and nighttime, and for the compression from over a hemisphere to a circular cross-section, the power/m^2 absorbed during the daytime = 1150 W/m^2. Since the Solar insolation over a circular cross-section is 1371 W/m^2, this means a Lunar albedo of (1371 – 1150) x 100%/1371 = 16.1%. [Compared to 28% for the Earth.]

      • I ended up above using the analogy of inductive components to account for the movement of charge (energy) within the capacitivey earth. I’ve just had a quick re-read of Dr. Jeffrey Glassman’s response to my comment on his September 2009 Rocket Science article “IPCC’S Fatal Errors: Internal Modelling Mistakes BY IPCC are Sufficient to Reject its Anthropogenic Global Warming Conjecture” ( Considering that he said “But among these fields, heat is unique. Like other forms of energy, it has no mass, and consequently has no kinetic energy. It cannot oscillate. Heat has no inertia.” He would probably reject my inclusion of an L term in my analogy. Dammit, another puzzle to get my head around!


  86. Roger Taguchi

    Here is the mechanism for the “greenhouse effect”: incoming Solar energy, primarily in the visible and IR not absorbed by CO2 and H2O, is absorbed by the the solid and liquid Earth, which warms up. As a black body at 288 K, the solid and liquid Earth emits IR peaking at around 15 microns. This happens to be the resonant wavelength for CO2 bond bending vibration (frequency 667 cm^-1). Water vapour and ozone molecules also absorb IR emitted from the Earth at different frequencies. All the molecules that absorb IR photons can either (a) re-emit IR photons (which results in scattering, but no warming of the atmosphere) or (b) transfer energy during radiationless collisions with N2 and O2 molecules, the main components of air. Because the vibrational energy levels of molecules are quantized, with the first excited level (which has vibrational quantum number v = 1) having an energy several times the average kinetic energy of air molecules, there will be an increase in the translational and rotational energies of N2 and O2 molecules after collision, i.e. an increase in the air temperature. Because N2 and O2 are homonuclear diatomic molecules, they have no permanent electric dipole moment, and therefore cannot absorb or re-emit IR photons and are therefore not greenhouse gases.
    Because N2 and O2 cannot emit IR, and CO2 , H2O and O3 can absorb/emit only at certain frequencies, the atmosphere cannot emit “black body radiation”, which consists of a continuous frequency spectrum over a wide range. Thus discussion of a 220 K black body emitting layer at 10 or 20 km shows lack of understanding of black body radiation. If you want to see what a true 210 K black body radiation curve looks like, go to the spectrum looking down on a Thunderstorm Anvil over the Tropical Western Pacific at The 210 K temperature is that of the top of the cloud level, which is made up of solid ice crystals (or perhaps supercooled liquid droplets). Condensed states can emit a continuous black body spectrum because there are weak van der Waals forces between molecules, as well as strong covalent bonds within gas molecules. The weak intermolecular forces act like weak springs, with lower vibrational frequencies. Combinations of these zillions of lower frequencies with the resonant vibrational frequencies and their overtones produce all the frequencies of the continuous black body spectrum.
    The spectrum looking down on the Thunderstorm Anvil shows excess IR emission at CO2 and O3 resonant frequencies, which cannot come from the 210 K cloud. Because the cloud has blocked 288 K black body radiation from the warm Earth, it cannot come from the solid or liquid Earth either.
    This excess IR emission at CO2 and O3 resonant frequencies also appears in spectra looking down on Antarctica. It cannot come from the cold Earth
    (the Second Law of Thermodynamics says that net heat cannot flow from a cold surface to a warm one). Invoking collisions to excite molecules to the v=1 state would lead to a cooling of the air as IR photons escape to outer space, contrary to a steady temperature profile. The obvious power source for the emission is incoming Solar radiation, which contains many IR frequencies which can excite CO2 molecules high in the upper atmosphere (as a guess at 80-100 km) to many vibrational states. To first approximation, the vibrational energy levels in a harmonic oscillator are evenly spaced, so that transitions from v=4 to v=3, from v=3 to v=2, from v=2 to v=1, and from v=1 to v=0 will all produce photons at 667 cm^-1 for CO2 bond bending. There is enough energy in incoming Solar 5780 K black body radiation to explain the observed emission. Note that the CO2 emission seen over the Thunderstorm Anvil exactly matches the bottom of the CO2 absorption profile seen over a warm Earth. The explanation is that CO2 is such a powerful greenhouse gas that the central frequencies emitted from the 288 K Earth are almost totally absorbed within tens or hundreds of metres of the surface, meaning essentially zero transmission by 20 km, including IR black body frequencies from the Thunderstorm Anvil. The very high altitude emission superimposed on this zero transmission results in an apparently truncated absorption valley down from a 288 K black body curve, showing approx. 33% transmission at central CO2 frequencies. This has fooled everyone else into believing in a 220 K black body layer at 20 km to explain the truncation. Because the angle of the incoming Solar radiation over Antarctica is low, the signal-to-noise ratio is half that compared to the spectra over warm Earth, so that there is only one prominent Q-branch|upward spike at 667 cm^-1. This is produced in between the usual P- and R-branches because the bond-bending mode of CO2 vibration produces a changing electric dipole moment perpendicular to the O=C=O axis, so “delta J = 0” transitions are possible. [Note that the interpretation of the heights of IR absorption peaks as “temperature probes” measuring black body “temperatures” is laughably wrong to any competent organic chemist.]
    Because of anharmonicity, the Q-branch spike for the v=3 to v=2 transition occurs 19 cm^-1 to the left of the main Q-branch spike (which includes both v=2 to v=1 and v=1 to v=0 transitions). This smaller Q-branch spike is clearly seen in all the satellite spectra obtained looking down on a warm Earth or Thunderstorm anvil. This is absolute proof that the emission must come from extremely high altitudes, where the probability of collisional deactivation is small compared to the probability of spontaneous emission of an IR photon. The existence of the v=3 state also means that it cannot possibly be produced by radiative exchange of IR photons emitted from the 288 K Earth, because the population of v=3 states at 288 K is vanishingly small.
    Even if we assume that climate change from 1850 to 2010 is real, and produced by CO2, the IPCC projections of future warming are a factor of 3 too large. If a CO2 increase from 300 to 600 ppm results in a 3 degree rise, then an increase from 300 to 400 ppm should result in a 1 degree rise, assuming a linear relation. But since CO2 absorption is highly saturated
    (doubling the CO2 will not result in a 100% increase in energy absorbed, but only 8-9% more), the early increase should be more-than-linear, resulting in 1.4 degrees. What is the historic record for climate change when CO2 increased from 280 to 380 ppm (which ought to be slightly greater than from 300 to 400 ppm, due to less saturation)? 0.7 degrees plus/minus 0.1 degrees. The IPCC projection is a factor of 2 too large, way outside the error bars, for change that has already occurred. Because of saturation effects, projections of future change will be even worse, too large by a factor of 3. The net feedback is essentially zero, not the assumed 2 degrees to be added to the 1 degree increase due to doubling CO2 alone.

    Detailed calculations of IR absorption by CO2 and explanations of the molecular physics involved can be forwarded to anyone who contacts me at

  87. Hi Judith, as one of the “..the technically savvy public .. ” I was puzzled by a couple of comments of yours on your article at .

    You started by saying “whether atmospheric gases such as CO2 (and H20, CH4, and others) warm the planet is not an issue where skepticism is plausible”. Later (December 1, 2010 at 12:54 pm) you said “The greenhouse effect and its magnitude is one thing, debating over whether CO2 warms the planet through infrared emission and absorption is another; I don’t see any point to debating the latter”. Then further on (December 1, 2010 at 12:54 pm) you said “The greenhouse effect and its magnitude is one thing, debating over whether CO2 warms the planet through infrared emission and absorption is another; I don’t see any point to debating the latter.”

    This may be just a matter of semantics but I see it as fundamentally misleading for non-scientists like me. Exchanges that I’ve had recently with John O’Sullivan over his repeated claim the “the Slayers” have killed off the theory of the “greenhouse effect” relates to this.

    To keep it simple for John (he’s not a scientist but seems to think that because he’s co-authored one book and is rubbing shoulders with some he is now an expert) I commented on his blog “Well John, this is going to be interesting, seeing “the Slayers” prove that putting a douvet on the bed does not keep us warm because it doesn’t generate heat. Forgive my simplistic explanation of a complicated scientific principle but, like you John, I’m not a scientist” (

    John’s E-mailed response was “I suggest you answer your own question by imagining putting a duvet over a corpse and measuring how much heat is generated- there’s your answer to that strawman.”

    That response suggests to me the that John has heard comments like yours above and taken them literally to be claiming that CO2 in the atmosphere generates heat which warms the earth beneath (and the atmosphere around it). As I responded to John “Why does the corpse cool down, even with a duvet over it, whereas the hot live body heats up the duvet? (hint – think in terms of energy absorption and
    emission rather than about heat or temperature, but to keep it simple ignore any energy transformation due to decomposition).”

    My rider in that E-mail was “Maybe we should take this debate over to the “Top Scientists in Heated Debate over ‘Slaying of Greenhouse Gas Theory” ( or to Judith’s thread when it starts up.

    John is ignoring my prompting on his blog thread for him to join in the discussion so Judith, please please please get a thread going on your blog to debunk the claim by “the Slayers” that they have brought about the “Death of the Sky Dragon”, because if this isn’t done soon there’ll be another book out (due in the Spring) which, if it becomes a “best seller” as is being claimed for that first one) we may have people falling for this nonsense, especially non-scientists like me.

    For me (again emphasising that I am not a scientist) “greenhouse gases” do not heat up the earth, they simply stop it from cooling so quickly when the source of energy (E/M emissions from the sun) are cut off). A bit like my duvet keeping a live person warm and why a corpse fails to keep warm under one.

    Come on you scientists, explain to me in lay language how my simple analogy fails to adequately describe the process. After all, as some of the disciples and followers of the doctrine that our continuing use of fossil fuels is leading to catastrophic changes in those different global climates argue, I’m just a nasty, senile old retired engineer.

    Best regards, Pete Ridley


    I’ve just had a read of Roger Taguchi’s comment. Once I had stripped out all oof the stuff that is way over my head it seems to fit my own understanding. “Here is the mechanism for the “greenhouse effect”: incoming Solar energy, primarily in the visible and IR not absorbed by CO2 and H2O, is absorbed by the solid and liquid Earth, which warms up. .. the solid and liquid Earth emits IR .. All the molecules that absorb IR photons can either (a) re-emit IR .. or (b) transfer energy .. with N2 and O2 molecules .. i.e. an increase in the air temperature. .. net heat cannot flow from a cold surface to a warm one). .. The obvious power source for the emission is incoming Solar radiation .. ”.

    That appears to me to fit nicely with my simple analogy of the duvet (the atmosphere), the live being and the corpse. Thanks Roger.

    • Pete, re your statement:

      You started by saying “whether atmospheric gases such as CO2 (and H20, CH4, and others) warm the planet is not an issue where skepticism is plausible”. Later (December 1, 2010 at 12:54 pm) you said “The greenhouse effect and its magnitude is one thing, debating over whether CO2 warms the planet through infrared emission and absorption is another; I don’t see any point to debating the latter”. Then further on (December 1, 2010 at 12:54 pm) you said “The greenhouse effect and its magnitude is one thing, debating over whether CO2 warms the planet through infrared emission and absorption is another; I don’t see any point to debating the latter.”

      In the context of the thread where that statement was made, the term “greenhouse effect” was being used by different people to mean different things. The fundamental aspects of infrared radiative transfer of a gaseous atmosphere are well known and IMO not particularly worth a further debate. The complexities of heat transfer in the atmosphere (involving convection, advection, water vapor feedback, etc. in addition to radiative transfer in an atmosphere with clouds and aerosol ) are where the scientific debate should be. An individual using the term “greenhouse effect” may be referring only to the infrared radiative transfer in gases (the so-called Tyndall effect) or the entire atmospheric heat transfer processes in response to an addition IR absorbing gases. So I have been trying to clarify this muddle of communication by referring to the infrared radiative transfer in gas issue to the Tyndall effect (following a suggestion of John Nielsen-Gammon).

      I am trying to put to rest the debate about the infrared radiative transfer in gases (the Tyndall effect). Claes Johnson has challenged this with his chapter in Computational Radiation. I don’t find his argument convincing at all. I am hoping to focus the discussion on this particular argument, so that we can put to rest. Whether or not something is called “back radiation” seems irrelevant to me, since people seem to be using the word in different ways. IR emission is isotropic, it radiates in all directions. So I am asking CJ to clarify and defend his arguments, and for others to critique them. It is not clear that we are getting far, but I am certainly not seeing anything here that moves me from my original assessment “I don’t find his argument convincing at all.”

      • So I am asking CJ to clarify and defend his arguments, and for others to critique them. It is not clear that we are getting far, but I am certainly not seeing anything here that moves me from my original assessment “I don’t find his argument convincing at all.”

        Judith, CJ’s ideas are not restricted to the ‘greenhouse effect’ but if he’s right would invalidate the standard methods used by engineers daily to calculate such things as the rate of radiational heat loss from fluids flowing through pipes etc.. Those methods are tested daily and found to be successful, either his methods yield the same equations in which case the difference is semantic or they are different and he’s proven to be wrong! I’ve asked him to apply his methods to some standard textbook problems but he ignores it. To test his theory it’s not necessary to apply it to a poorly defined system like the atmosphere it can be applied to a well defined system in a lab where the results can be accurately determined. The problem is that this is done every day and the results agree with the standard methods, which gives us the confidence to apply those same methods to the atmosphere and hence arrive at the ~33ºC warming.
        Merely asserting that ‘back radiation’ would lead to instability doesn’t cut it!

      • Judith,

        Whether or not something is called “back radiation” seems irrelevant to me, since people seem to be using the word in different ways

        Yes, that thread does seem to have been hijacked by arguments over back-radiation.
        The irony is, the opponents seem to be all saying the same thing, but disagreeing because of the wording – or something. I can’t really understand why they’re disagreeing so much. They seem to be talking past each other.

  88. Hi Judith, thanks for getting back so promptly. Only yesterday I E-mailed Roger saying how helpful you are like that, which can’t be said of all disciples and supporters of the doctrine that our continuing use of fossil fuels is leading to catastrophic changes in those different global climates (otherwise known as CACC). I think that you may have missed the point of my quoting from your comments here. The significant thing that I thought I was drawing attention to is the impression that some appear to have that “greenhouse gases” actually ” .. warm the planet .. “. Several before me here have quoted those words and I believe that many of us non-scientists (like John O’Sullivan) understand that to be what “the greenhouse effect” is trying to say, hence the claim made by “the Slayers” to have “killed the dragon”.

    My lay understanding is that the “greenhouse effect” (ignoring all of the other enormous complications arising from evaporation, condensation, ocean and atmospheric transport of energy) does not say anything like that but simply says that it reduces the amount of cooling that the earth experiences before another boost of E/M energy is poured into its surface when the sun comes out again.

    Regarding Claes’s hypothesis, when and where do you intend to start refuting it. I’d appreciate a short E-mail from you with a link when you have started that debate Also, can you afford the time to go beyond just Claes and refute any other hypotheses offered by those “experts” who authored “Slaying the Sky Dragon”.

    Best regards, Pete Ridley

    • Yes, without the greenhouse gases, the surface of the earth would be colder. some of the IR emitted from the surface is absorbed by gases (CO2, H20) and clouds. These gases and the clouds also emit IR radiation, some of which is radiated back in the direction of the earth’s surface, and so in effect you are returning some of energy that was lost.

      Regarding taking on other chapters in the book. Well Claes’ arguments were the most “scholarly” among those in the book (albeit unconvincing and incorrect IMO). I’m not sure I could get anyone to take the other chapters very seriously. But if there is interest, we can tackle another chapter, but I doubt that the publishers want to keep making more chapters publicly available.

      • After your long trip to the land of the Portugesse , are you up early or going to bed late?

        Best wishes in either case.

      • Yes, without the greenhouse gases, the surface of the earth would be colder.

        I wonder if that’s true Dr Curry. I have a different hypothesis and would welcome feedback.

        Without GHGs (including H2O) the atmosphere (of N2 and O2) would warm during the day, but would be unable to shed that warmth via radiation. It could only shed the warmth via conduction with the surface once that surface cools to a level lower than that of the atmosphere immediately above it. (at some stage in the night)
        At this time, a temperature inversion would be created as the cooler lower atmosphere could not be displaced by the warmer air above.

        The next day, as the surface starts to warm, it would in turn warm the air above. This warming of the atmosphere (given enough days) would necessarily reach a level equal to the warmest part of the day.

        The upshot being, an atmosphere that is substantially warmer than that of an earth with GHGs.

        If the above is true, it leaves us in a dillemma.

      • The troposphere in that scenario would have an upper bound at the surface temperature (it is heated from below not above).

      • I need you to expand on that Phil. I don’t understand what “upper bound at the surface t” means.

      • The surface would heat the troposphere so the troposphere will not be able to exceed the surface temperature and so would be colder than an atmosphere with GHGs.

      • I see. But the surface continually warms and cools. The atmosphere on the other hand must accumulate some warmth as it can’t radiate. it can only lose it’s warmth the same way it got it, via conduction.

        During the day, the atmosphere close to the surface will be warmed by the surface. It will then rise and be replaced by cooler air which is also warmed etc until the sun starts to go down.

        During the cooling phase sometime after noon, the atmosphere near the surface will cool via conduction, but it can’t rise and can’t be replaced by the warmer air above it. i.e. the atmosphere will accumulate heat. And if it’s accumulating heat, it will do so until it reaches a T close to (but possibly not quite) the surface T at day time.

        that’s warmer than an atmosphere with GHGs is it not?

      • Baa Humbug (please let’s have a real name), just a bit puzzled by your response to Phil. Felton “ .. the surface continually warms and cools. The atmosphere on the other hand must accumulate some warmth as it can’t radiate .. ”. Let’s take the situation that you described in ??? where you postulated “ .. the atmosphere (of N2 and O2) .. ” without GHGs (including H2O). Although this hypothetical atmosphere may be unable to emit or absorb radiated energy wouldn’t it just lose any energy that it picked up by returning it to the earth as that cooled down by radiating at IR, i.e. the reverse of how it obtained the energy in the first place (just as Phil Felton said)? You argued that “ .. the atmosphere near the surface will cool via conduction, but it can’t rise and can’t be replaced by the warmer air above it. i.e. the atmosphere will accumulate heat. .. ” but surely energy would be transferred from higher to lower levels as a result of molecular collisions between warmer upper and cooler lower levels, i.e. the energy is transferred from atmosphere to the cooling earth to keep it a little warmer for a little longer than if there was no atmosphere at all ?

        Best regards, Pete Ridley

      • Pete,

        If I understand this correctly:
        Because of convection – the fact that hot(ter) air rises, the atmosphere would tend to accumulate heat from the top downwards, so the losses by conduction back to a colder surface would likely be minimal. The atmosphere would, depending on the size of the losses, gradually heat up to something approaching the highest daytime surface temperatures.

      • Phil,

        Which surface temperature would that be? Surely the daytime surface temperature in the tropics wouldn’t differ greatly to what it is now? Possibly even hotter, given the (implicit) absence of clouds?

      • Another related question, maybe simpler because it is steady state, would be “what is the equilibrium T profile of a perfectly transparent (ie convection/conduction only) atmosphere if T ground is fixed?”.

        Originaly, I though it will rapidly go go to a lapse rate profile by convection, and then, more slowly, to a constant T=Tground once conduction kick in.

        After more thinking, I do not think that conduction would lead to uniform T, I think that classic Fourier law can be derived from kinetic gas theory only outside of a gravity well. In presence of gravitational potential energy, it probably have to be modified because the top layer should have an average kinetic energy lower than the ground level. It would mean that equilibrium T profile would not vanish but go to a hot dow / cold up profile. Possibly the lapse rate itself???

        Imho this case has to be resolved before going to a non-transparent atmosphere.

        Next problem would be a semi-transparent grey atmosphere with fixed T ground and perfectly reflecting mirror at TOA (or above), so that a zero heat flux is imposed at TOA. Again, what would be the profile in this case? Different from the fully transparent one? Maybe, but I am not so sure anymore….

        I am somewhat surprise I have not seen such treatment as GH effect investigation, they are simple limit cases that should provide a more solid physical foundation of the coupled gravity well+conduction/convection/radiation-absorption problem…

      • Looked a little bit on the net about what I thought to be a simple problem: “what is the T profile in a column of the simplest gas one can imagine (pefectly transparent (and thus non-emitting), monoatomic, dilute…as close as posible to kinetic theory hypothesis as possible) subjected to gravity (or any kind of external force field)?”.

        Sounded so simple as to be part of classical text books.

        It turn out it was (and still is, afaik, because it has not be completely resolved) one of the big dispute of the early statistical thermodynamics, between Boltzmann (who believed in constant T) and Loschmidt (who believed in an equilibrium T gradient, T higher at the bottom than the top).

        Anybody know if more modern physics has definitively solved the puzzle? Some source I found think Loschmidt may have been right, even if Boltzmann has win at the time because Loschmidt seems to allow for perpetual heat engine. Given the link with perpetual engine, some of the source tend to crackpotism, so I wondered if this dispute has be solved in mainstream.

        This particular probelm seems highly relevant to GH discussion, because it could fundamentaly change some heat transfer law in the vertical direction and/or reference lapse rates….

      • Kai, you ask good questions.
        Perhaps this might help: Walter J. Moore in the 3rd edition of his classic college textbook “Physical Chemistry” (Prentice-Hall, 1962) on pp.230-231 derives the Barometric Formula, showing how atmospheric pressure is a decreasing exponential function of altitude, assuming constant temperature, T. This assumption allows for use of the Ideal Gas Equation, showing density varies directly as P for constant T. This is intuitively obvious.
        However, it takes work to compress a gas (you have to force a gas into a smaller volume when you use a bicycle tire pump), and the added energy shows up as an increase in gas temperature. The reverse is also consistent with the Law of Conservation of Total Energy: when a gas expands against an external pressure, it does work on something else, and so decreases in energy itself. This shows up as a decrease in temperature of the gas. Therefore as you go up in altitude, the air temperature drops as a result of this work of expansion against the pressure of the rest of the atmosphere pressing down on it. Thus the temperature profile in the troposphere is a decreasing temperature with increasing altitude (complicated on hot summer days by convection currents which move bulk air packets upward against gravity). Convection ends at the level of cumulus cloud formation, I think, so ignore this for now (the math of convection is too fiendishly hard for me to give you a pencil-and-paper calculation).
        In the stratosphere, the temperature starts to increase with increasing altitude. The molecular explanation is that the formation of ozone molecules from O + O2 = O3 releases energy (is exothermic), as O3 is in a lower potential energy state relative to separate O atom and O2 molecule. This energy released gets spread out by collision, resulting in an increased translational and rotational energy of air molecules (i.e. as an increase in temperature). The O atoms are formed when a high energy UV photon from incoming Solar black body radiation blasts apart O2 molecules. So the heating of the stratosphere with altitude ultimately results from incoming Solar energy.
        Above the stratosphere, temperature once again decreases with altitude due to the work of expansion against pressure from above. Way up in the thermosphere, the temperature profile again increases with altitude once the really hard UV photons from the Sun start to rip apart all molecules into atoms and ions/electrons. So the heating there is once again due to incoming Solar energy.
        The derivation of the Barometric Formula is made easy by ASSUMING a constant temperature. This is not, IMO, the same thing as Boltzmann saying that the temperature IS a constant. Anyone climbing or driving up a mountain knows that this is not true.
        I hope this off-the-cuff explanation helps.

      • On reading my reply above, I realized that the temperature rise in the stratosphere with altitude could as well be due to absorption of visible and UV photons by ozone (it is a bright blue molecule, so absorbs some visible light). If the ozone is then in an excited vibrational or electronic state, it could then transfer energy on collision to other air molecules, resulting in an increased air temperature. You’ll have to Google a website on the temperature profile of the atmosphere to see which of these two possibilities is the accepted one (or if there is a combination of these two:
        heat of recombination when forming ozone, or absorption of visible and UV photons to form vibrationally and/or electronically excited states).

      • Thanks Roger, this is indeed the classical way to explain T profile in real atmosphere…

        My question was more theoretical than that though, but, as with all theoretical question, may lead to question the classical derivation of real case by reevaluation of certain hypotheses, or even some laws/equation classically used.

        The Boltzmann/Loschmidt dispute was about such a theoretical case that is newer encountered in nature, and is quite difficult to reproduce experimentally (hence the dispute, when experiment is easy, theoricians usually accept the experimentalist verdict…it is a good definition of science, after all ;-) ).

        The idea is to derive the equilibrium state of a column of transparent perfect gas isolated but in a gravity well (or other king of external volume force leading to potential energy).
        -Equilibrium, so, no net heat flux is allowed, and all average value/ thermodynamical value (T, p, rho should be enough for a perfect gas) should be constant in time
        -Isolated, so no heat flux at boundaries either
        -perfectly transparent, so radiative transfer is non-existent. Convection and conduction can exist in transient, but we are looking at steady state so they should vanish and we should have a dynamically stable profile
        -Isolated, so no heat flux at boundaries either.

        Boltzmann though that equilibrium T must be constant with height (based on 2nd law arguments to prevent perpetual heat engine)
        Loschmidt though that equilibrium T should decrease with height (based on kinetic gas mechanical model).

        I am somewhat reluctant to dismiss Loschmidt, because this conclusion seems remarkably similar to virial theorem for a transparent perfect gas submitted to self-gravity. On the other hand, I do not believe in perpetual engine, of the first or second kind…
        But I think that maybe the two may somewhat reconciliated by including potential energy in entropy, and modifying fourier law in presence of gravity/external forces to allow for no heat flow even with a T gradient aligned with the external force.

        This seems like a fundamental problem to really understand lapse rates in atmosphere, and I am somewhat surprised a definitive solution to the dispute is not more available….But maybe I am wrong, and the attempt at experimental validation of Loschmidt I have found are fully in crackpot territory (they seem to confirm L.).
        Anyway, after giving a very rapid response based on a similar question posed here, I am quite excited the solution seems less obvious than I expected, I hope some could provide good insight in this problem (which is sufficiently different from real atmosphere to avoid deniers/catastrophist name calling ;-) )

      • Judith, maybe you could consider this question (or a similar one, better phrased, which should be an improvement as I am not a native english speaker nor I am used to write articles about this specific domain – I have written quite a few reviewed paper but in acoustics) in the greenhouse theory series?

        It is quite strongly related to Miskolczi ideas imho, and I was not aware the stuff go back to early kinetic gas theories and to Boltzmann…

      • Roger Taguchi

        OK, Kai, I now get that this is a theoretical model, not the real atmosphere you are interested in. As I have never seen Loschmidt’s argument before, here are some comments which could be just bs.
        (1) In the real atmosphere we have spherical shells of equal density, so it is not made up of cylinders or rectangular volumes. Does this make a difference in the model? In particular, I assume that the Ideal Gas does not expand outward as you go up in height. So far then it would be like the derivation of the Barometric Formula, using a cylinder or rectangular volume.

        (2) From Statistical Mechanics, Boltzmann’s solution is for the problem of “What is the most probable way of distributing a finite amount of energy among a finite number of molecules whose energy state depends only on their gravitational potential energy?” Here are some highly unlikely ways of distributing the energy:
        (a) have a subset of the molecules jammed at the top of the enclosure until they have used up all the available energy, with all the other molecules piled up at the bottom of the enclosure. This would be somewhat like rolling a pair of dice and getting only 2’s and 12’s after a zillion rolls. Logically possible, but highly unlikely.
        (b) Calculate the average energy for all the molecules, and place all the molecules in a narrow band or volume centered on one height somewhere in the middle of the enclosure. This would be like getting only 7’s after rolling a pair of dice a zillion times. Logically possible, but highly unlikely.
        The statistical mechanical solution to this problem is the most probable distribution being a decreasing exponential function of height: exp [-cmgh], where m= mass of molecule, g = acceleration due to gravity, h = height, and c = some constant. mgh is of course the gravitational potential energy. c = 1/kT, where k = Boltzmann’s constant and T = absolute temperature, which at equilbrium is another constant. The average kinetic energy of a gas molecule is mv^2/2 = 3kT/2, where v is the root mean square (r.m.s.) speed. What could possibly be wrong with this analysis?

        Well, the decreasing exponential function given so far uses only potential energy, assumed to be total energy. This would be true only if kinetic energy is zero (i.e. in the impossible state of T = 0 K). Since total energy for a molecule is equal to the potential energy + kinetic energy, the molecules in the picture so far at the top have higher total energy on average than those below (as before), but the total energy is now a linear function of T as well as a linear function of h. Therefore a plot of the natural logarithm of the distribution function vs h will not be a straight line. Since a straight line on such a graph would mean a constant temperature, a curved line would mean a changing value of T with changing height , h. Is this the source of Loschmidt’s argument? Or did I mess up in the argument so far (it’s really possible!)?

        Statistical mechanics also allows for fluctuations from the mean. Would small pockets or bubbles of hotter or colder molecules move upward or downward, resulting in a net creation of a temperature gradient with height, or would they always move in such a direction to damp out any such separation (my initial bet)?

        If there is a temperature gradient with height, then what about heat transferring via emission and absorption of photons? For instance there would be a net flow of photons from the hot layer to the cold layer until the two layers are at the same temperature, when the rates of energy flow are equal in both directions. A complication could be the fact that a photon falling downward in a gravitational field gains energy (has its frequency increased) by Einstein’s General Theory of Relativity, and a photon moving upward loses energy. This effect has been confirmed with gamma rays and the Mossbauer Effect in nuclear physics.

        Now my brain really hurts!!!

      • Yes, a very interesting subject: question is deceptively simple, but the more you look at it, the more your brain hurts ;-)

        First, a simplification so at leat this aspect will not bother us. Let’s assume transparent gas, no absorption of EM radiation at any freq. At, per kirchoff, no emission either. That way, we can completely evacuate discussion about radiative heat transfer, it is simple non-existent…

        Now that this issue is sidestepped (although it is important in the earth atmosphere), let’s evacuate another difficulty:
        I assume constant gravity force. Curvature is neglected, so everything will be in cartesian coordinates, not spherical. This is imho valid as a first approximation for earth atmosphere, because it is quite thin anyway. So we can consider a 1D problem, velocities can have horizontal components, but no average quantity could vary except in the vertical direction.

        Now that it si properly simplified, the first answer that comes to mind is constant T, exponential p. As you have shown, Boltzmann derived it from statistical arguments, but you can also get it from more “classical” physics: fourier law (T HAS to be constant, else heat exchange would take place which we want to avoid for steady state), and then derive p from perfect gas and hydrostatic equilibrium, and voilà ;-)

        Well, except that both this classical derivation and Boltzmann (it seems, also in your explanation) have subrepticely introduced what is in fact an hypothesis: T is a constant (explicit with Boltzmann, implicit with classical approach where fourier law is assumed to be valid even if potential energy gradient is of the same order as T (kinetic energy) gradient, while I am open to a fourier law that incorporate somehow potential energy).
        Loschmidt’s argument is that T=constant is not necessarily a valid hypothese, and can not be introduced a priori. On the contrary, he use the purely mechnical argument that each independent particle (kinetic gas theory, remember) is subejcted to gravity during its free flight between collisions (with walls or other particles). hence, its cinetic energy should be higher the lower it goes. Add averaging and it makes sense that the average kinetic energy of particles is higher where potential energy is lower. in other words, T is higher down, lower up. As I said, this is remarkably similar to virial theory, that have the same conclusion about the average kinetic energy of any set of particles whose interact by a power law potential (typically gravity). Typical example is the derivation of a T profile in a cloud of perfect gaz in space hold together by its gravitational self-attraction, and, afaik, it is considered valid newtonian astrophysics…

        Really interresting problem, now I think u understand why I am really not sure about what kind of T profile would be the equilibrum solution…:-/

      • Here you have a paper on this issue

        Click to access jas04.pdf

        This and many other papers confirm that under the standard assumptions the column is isothermal, but the paper discusses also, how the result turns to adiabatic lapse rate or to something between these extremes when the setting is modified.

      • Thanks a lot Pekka, exactly to the point and not behind a paywall, the authors are even in y neighborhood! I will read that carefully :)

      • This paper is also interesting to read.

        Click to access 43.pdf

        While it does not answer directly the question on the temperature dependence on the altitude, it demonstrates, how the fact that particles lose kinetic energy when going up does not contradict the constancy of the temperature. The explanation is in the fact that the particles with lowest energy never get high up. It turns out that the energy loss and the selection gain compensate each other exactly, if the distribution of velocities is in accordance with Boltzmann’s distribution. The mathematical proof is in the Appendix C.

      • After a first read, and concentrating on the constant T derivation, I have a nagging question: the authors speek of net heat flow in the column, which should be 0 (I agree). However, imho, it seems not enough, because one can add heat on top and remove it on bottom without violating constant enthalpy. It would correspond to a constant heat flow, that is maybe what is needed to obtain the constant T (if no heat flow leads to T gradient). I am not sure that I do not make a stupid mistake (thermodynamic is not my prefered physic topic ;-) ), but….this ring warning bells in my mind ;-) Well, if still have this nagging question and without additional hint from this blog, I should be able to contact on of the author – he is almost next door ;-)

      • aaah, second paper is more like classical kinetic theory…but first model assume non interacting particles and feels somewhat ad hoc and partly circular. The one with colliding particles neglect gravitational fall between collision…not good for the problem at hand, since it is precisely what pulled Loschmidt towards T gradient in the first place.

        Some people proposed to investigate possible T gradient experimentally, using gas as close as possible to perfect gaz ( some argon, I guess ) in very powerful centrifugator. If they manage to get good insulating walls and a fluid at rest in the rotating frame, they expected to have very strong T gradient easy to measure. But to get an at-rest fluid, good luck!
        static column suffer from the same problem: the gradient expected is quite small and it is not easy to have a resting gas on a high column without any heat flow from the walls (for a long time, to ensure equilibrium)…

      • Roger Taguchi

        Thanks, Pekka, for the great references, and to Kai for stimulus. I had forgotten about the Virial Theorem which combines the potential and the kinetic energies, so I was just floundering in the dark.

      • And if somebody likes to go to the origins, he may search Gibbs Scientific papers from Google books. There is a scanned pdf available, but the scanning is not perfect. A few letters are left out from the beginning of each line on page 145, where this issue is discussed.

        It may also be difficult to valuate fully what Gibbs has done.

      • Roger Taguchi

        Hi Pekka and Kai!
        I’ve had second thoughts about the Virial Theorem, which applies to a bound state. E.g. an electron in the hydrogen atom, where the average kinetic energy is 1/2 the size of the potential energy. But the average kinetic energies of the molecules in an Ideal Gas are not related to the gravitational potential energy by the Virial Theorem. For example, in the field-free state, the molecules still possess kinetic energy, although U would be zero.
        So the total energy of the molecule is the sum of the gravitational potential + the kinetic energy it would gain in falling down through the potential + the random gas kinetic energy. So maybe the argument of the exponent cannot be simplified into one term proportional to h, and so the plot of the log of the distribution function vs h may not be a straight line, meaning there is a temperature gradient. Or maybe it can. Perhaps this was all covered in the literature references which I only briefly skimmed, but my head hurts so much I’m gonna give it a break.

      • I have thought along the Virial theorem too, as it is central to Mikolski (???spelling?) GH “heresy” ;-)

        People attack it telling it is only valid when potential is between any pair of particles, and it get much more complicated when an external potential field (like earth gravity) is used. I wonder though: imagine a collection of elastic particles under self-gravity. This is the typical application of Virial therorem (one classic of astrophysics). It also happen to be classic kinetic theory model of a perfect gas cloud in space. Whe should thus be able to use prediction of Virial to derive mean kinetic energy there, and from this and kinetic gas theory, a temperature.

        Now, pick, at radius R far away from the center, a control volume such as a thin column of height H << R. Provided the solid angle is very small, and h << R, whe have a control volume remarkably similar to my gas column. Only thing is, for now, we only have a control volume, with virtual boundaries transparent to energy and momentum exchange.
        This is, we have an equilibrium. I think that, from equilibrium, it is valid to repplace virtual boundary py a perfectly reflecting boundary (with respect to energy and momentum) without changing any internal state, and keep equilibrium. I.E a perfectly insulating rigid wall. This slab then verify all my hypothesis (under the previous geometrical assumptions), but can get T profile from Virial theorem. I fail to see where I made my mistake, but I think that T profile from Virial is not constant…Either I made a error somewhere (The impermebilisation of virtual boundary? The Virial theorem in fact predict average kinetic energy which IS equivalent to constant T within kinetic gas theory framework, given mean density profile? That's the 2 possible errors I see), or Boltzmann was wrong, and Loschmidt right (in principle, if not in quantitative prediction of T gradient)

      • Virial theorem is not in general a powerful tool in understanding physics. It may be very useful in some situations, about which we have very little detailed knowledge as is the case in some astrophysical problems. It is a general theorem and valid, when applied correctly, which is not always so easy. Even when there are no problems in getting it right, much more is usually known through other more detailed approaches. When this is done, checking that the virial theorem is satisfied is usually easy, but useless. In doing this one notices, how little the virial theorem tells compared to other knowledge.

      • Roger Taguchi

        Sorry, Kai and Pekka for not thinking clearly in my previous postings: I had been excited about another posting I had finished about 1 am, and got only about an hour of sleep and so my brain was fried yesterday.

        Here are some other points to ponder:
        (1) U = mgh is only an approximation for U =
        – GMm/r, the gravitational potential energy of a molecule of mass m at distance r from the center of a spherical Earth of mass M. For short distances h from the surface of the Earth, the inverse function
        -1/r is approximated by a straight line segment of positive slope, so U is approximately linearly proportional to h. Assuming a constant temperature T, the barometric formula (a decreasing exponential function of h) follows.
        But you can immediately see all sorts of complications when we examine the assumptions listed above.

        (2) First, does the Virial Theorem apply to objects bound by gravity to the surface of the Earth? For the electron of charge -e bound to a proton of charge +e in the hydrogen atom, the potential energy is proportional to -e^2/r (a form similar to -GMm/r), and the Virial Theorem does apply to its orbit. Yet if we are at the Equator we are rotating 25,000 miles in 24 hours (one day), a speed more than 1,000 mi/h (more than Mach 1) but we still fall back down if we jump up in the air. The reason is that we are not in orbit (which would require something like Mach 25). So the Virial Theorem doesn’t apply for us stuck on the surface of the Earth. Nor for gas molecules at 0 K resting on the surface of the Earth. The average speed of a gas molecule at normal temperatures is about Mach 1 (since sound travels as a wave when gas molecules collide with each other), so they’re not in orbit either, so does the Virial Theorem apply to them?

        (3) Why are we as massive objects stuck to the Earth’s surface while air molecules are not? Because our molecules are stuck together by chemical (electrical) forces, and there’s only one way we want our molecules to be arranged, whereas there are zillions of ways energy can be distributed among zillions of air molecules as they move independently to various heights h. Thus the entropy can be calculated in statistical mechanics, leading to the decreasing exponential Boltzmann function for a given equilibrium temperature T. Kai has noted that these calculations are circular: we assume a constant temperature and derive a decreasing exponential function which contains a constant temperature.

        (4) It is possible to have non-Boltzmann distribution functions, however, in the course of chemical reactions. For example, the reaction F + H2 = HF’ + H is exothermic (releases energy to the surroundings). HF’ represents a product hydrogen fluoride molecule in a highly excited vibrational and rotational state. Because there are more ways of distributing energy democratically among many molecules than keeping it hoarded in one molecule, after 1-10 collisions, rotational energies will have been transferred between molecules to give a Boltzmann distribution funcion, with a weighting factor of (2J+1) for the density of states with the same rotational energy, where J is the rotational quantum number. Similarly, after hundreds or thousands of collisions, energy in vibrationally excited states will be distributed among colliding molecules, giving a Boltzmann function which is a decreasing exponential function of v, the vibrational quantum number, since vibrational energy is proportional to (v + 1/2). At normal temperatures and pressures, there are about 10^10 collisions per second, so excited rotational and vibrational states are rapidly quenched, and the heat of reaction spread throughout the surroundings. This is not true in the upper atmosphere, at 100 km, where the number of quenching collisions is so low that excited vibrational states formed by incoming Solar radiation can live long enough to emit IR (infrared) photons. This fluorescence observed by the NIMBUS satellites has fooled climate scientists into thinking that these IR photons came from the Earth’s surface by radiative transfer all the way up to a hypothetical 220 K layer at 20 km, from which they were able to escape. This non-physical theory is immediately falsified by the excess IR emitted at CO2 frequencies over Antarctica and over a 210 K Thunderstorm Anvil.
        Experimentally, if vibrationally excited product HF’ molecules are observed before they are quenched by collision, a highly non-Boltzmann distribution is seen. A plot of the log of the distribution function vs v (which is a measure of the energy) does not show a straight line, but a Gaussian-like peak. The slope of the line segment joining adjacent points gives one value for the “temperature”, but if the entire graph does not form one straight line, the “temperature” changes from point to point. If there is a population inversion (when the population of a higher vibrational level is actually higher than that of a lower one), then the slope of the log plot would be in the opposite direction, meaning a negative absolute “temperature”. That’s why I have put the word “temperature” in quotation marks, since technically it ought to be used only for equilibrium situations which follow the Boltzmann distribution with a constant absolute temperature (which is positive). However, it’s a handy word to use to describe the relation between any 2 points on a distribution curve, even if it is non-Boltzmann. There is a use, of course, for population inversions, in lasers. This was the basis for President Reagan’s “Star Wars” program. Huge tanks of H2 and F2 shuttled into orbit could power HF chemical lasers, with the IR beams used to destroy any Commie missiles. This panicked non-scientist political and military leaders into an arms race that eventually bankrupted the Evil Empire, which is no more (just as Reagan planned all along?).
        So in general, any deviation from a straight line plot on a graph of the log of the distribution function vs energy would constitute a change in “temperature”.
        This does occur in non-equilibrium situations such as the course of chemical kinetics. The question is, can it occur in the hypothetical model under consideration?

        (5) When an Ideal Gas expands through a tiny orifice into a vacuum, no work is being done, and since there are no attractive forces between molecules, there is no change in temperature. The Joule-Thomson coefficient is zero for an Ideal Gas.
        Real gases, however, do have attractive forces between molecules, resulting in a temperature change on expansion into a vacuum. This is the basis for refrigerators and air conditioners.

        (6) Even an Ideal Gas follows the rule that when work is done on it (when it is compressed), it warms up, and when it does work on expansion against an external pressure, it cools down. So sure, in the hypothetical problem, the temperature of the column of Ideal Gas will be cooler with increasing height.
        But cooling a gas at constant pressure means a contraction of volume, which means the density increases. So the density is higher than you would expect at constant temperature, i.e. higher than the predicted Boltzmann distribution using the higher temperature. Since total energy is the product of the number of molecules times the energy of each molecule, does the Boltzmann distribution still work when energy is the independent variable, not the height, so that the increasing density compensates for the decreasing temperature, giving the same (or nearly the same) result as assuming constant temperature? If so, then a temperature gradient is consistent with a constant temperature!

        (7) Some may wonder why I even brought up the possibility of photon exchange, when in another posting I stated and proved that gas molecules in the atmosphere cannot emit or absorb black body radiation. But the Earth can. So can outer space, which is filled with 3 K microwave photons. So by exchange of energy during inelastic collisions, photons (radiation) can allow for heat flow from hot to cold. That’s why it’s impossible to achieve a region of 0 K, since it will always receive net energy flow coming in as photons from the walls of a hotter container, or at the least from the 3 K background of outer space, if there is no container.
        At this stage, I’m not sure if this mechanism is necessary or sufficient to equilibrate temperatures in the hypothetical column even if there is a temperature gradient established in a metastable state (like supersaturated solutions or supercooled liquids which can exist for a long time, but are not in equilibrium with pure crystals).
        Kai, it’s been fun, but I am more interested in models that approach reality, so please forgive my laziness in not reading every word in your postings or in the literature forwarded by Pekka. I’m sure I’ve screwed up along the way, but hey, this is only a hypothetical problem!

      • To review, the Boltzmann function for barometric pressure says density/pressure is a decreasing exponential function with height, and is derived assuming a constant temperature. But due to energy lost as work of expansion against an external pressure, the temperature of even an Ideal Gas decreases with altitude. This means the average kinetic energy of the molecules is lower at altitude, and the Boltzmann function would predict a lower density at altitude compared to that using the warmer temperature.

        If a gas at constant pressure is cooled, it shrinks, increasing the density. So if this factor counteracts the previous change, then the density could follow the density distribution curve predicted using a constant temperature. This means the plot of log P vs h is a straight line. Therefore the apparent contradiction comes from two different ways of defining the temperature. The more fundamental way is from the slope of the log P vs h graph; if it is a straight line, there is only one temperature, and the system is at thermodynamic equilibrium. Note that this is a bulk concept: it makes no sense to talk about a temperature and calculate a slope if there is only one point on the entire graph! The other way of defining temperature is to have the average translational kinetic energy of the molecule equal to 3kT/2. The word “average” implies average over bulk material, but everyone can visualize ONE molecule moving at the r.m.s. speed corresponding to this energy. THIS temperature decreases as altitude increases, giving rise to the paradoxes, because everyone is taught that heat flows from hot to cold until equilibrium is reached at one value for the temperature.

        As I see it now, Boltzmann was right, but you have to understand temperature as measured from the slope of the straight line graph of log P vs h. From the perspective of the Gibbs Free Energy function,
        delta G = delta H – S, at equilibrium delta G = 0. For a gas molecule moving upward against gravity, it’s endothermic, so delta H is positive. For delta G to equal zero, delta S must be positive (there must be an increase in entropy, which does occur when gas molecules have more freedom of movement upward and downward in the atmosphere). As T approaches zero, the second term in the Gibbs Free Energy expression approaches zero, so at equilibiiurm, delta H must approach zero too. This means molecules cannot climb up very much in potential energy, so the density distribution becomes confined to a layer near the Earth’s surface.

      • Argl, I though I was on something with Virial theorem….My mistake, it only gives global average of kinetic energies, not time average quantities that can be localised. I will thus forget about using it for the problem at hand.

        Ditto from Boltzmann, from what I get: It use one single T, like a global quantity, probably related to average kinetic energy in the whole system too. And from this, barometric pressure is indeed derivable…or velocity distribution in spatially homogeneous system.

        Problem is that here we are interested in a more local temperature, and no field is really assumed homogeneous along the gravity direction. I start to wonder if there is no fundamental problem in deriving a spatially variable T that is consistent with statistical thermodynamic :-/

        I’ll have to check more the thermodynamical approach sent by Pekka (thanks again), but I’d like to also find a “generalised” kinetic theory approach. It’s a long time I examined kinetic theory in details (during the early years of my mechanical engineering degree), and I remember I was impressed at all you got from such simple initial hypothesis, although the statistical treatment was slightly tricky. I’ll have to find a good detailed explanation of it again, any free detailed derivation on the net?

      • Roger Taguchi

        Forget the first 3 sentences in the second paragraph in the above article; I think they’re garbage (so Kai, don’t feel too bad – everyone screws up when exploring new territory). The rest is OK.

        Here’s how to resolve the paradox, I think: the log P vs h graph is probably not a straight line experimentally. The slope of the best straight line probably gives the average temperature between the bottom and the top of the column. It’s like saying the slope of a chord of a curve is the average of the slopes of the tangents to the curve at the two ends of the chord.
        So both Boltzmann and Loschmidt got something right: Loschmidt is right in that there is a temperature gradient, and Boltzmann in that the best definition of temperature involves the slope of the tangent to the log of the distribution curve. But this tangent might change slope slightly with height, meaning that the decreasing exponential function adjusts its steepness as temperature changes with height FOR ANY REASON. At any given temperature, statistical mechanics gives the most probable way of distributing a finite amount of energy among available energy levels. Boltzmann’s assumption of constant temperature works for small changes (e.g. for 1 km altitude change, the temperature might change by 3 degrees, but this is only 1% of 300 K = 27 Celsius).

        Kai, the value of this hypothetical problem is that it explains the temperature gradient in the troposphere and mesosphere as well as the overall decreasing pressure with altitude. The atmosphere is modelled rather well by Ideal Gases, except where photons from the Sun can excite or smash apart molecules, creating thermal inversions in the stratosphere and thermosphere. For those who worry about the temperature gradient at equilibrium, the contents of the universe are not at equilibrium. The present temperature profiles are for steady state conditions, with the 5780 K Sun trying to warm up the 288 K Earth and the 3 K cosmic background, and the 288 K Earth trying to warm up the 3 K background. All things are tending to approach the heat death of the universe as a limit.
        Then the gradients will disappear. But don’t worry; we won’t be around to freeze in the dark.

      • Roger Taguchi

        Oops, by “garbage” I meant in MY article of 10:18 am, not kai’s article of 1:52 pm.

      • No problem Roger, especially as my own post was not correct ;-)

        I have gone back to the Maxwell-Boltzmann distribution, to check exactly how it is obtained. The argument is quite easy, it is the distribution of energies of highest probality in a discrete system made of a very large number of dof (usually particles in motion), under two constraints: total energy is fixed (no energy exchange between the system and the outside), and total number of dof is constant too (no mass exchange).
        Nowhere is the T explicitely required at this stage: even if it is usually introduced early in the derivation, it is more suitable for the problem at hand to wait a little bit before introducing temp, ang just use probability with, for now, unknow constants).
        The result is that the probability to find an energy e_i is equal to g_i/exp(a+b*e_i), with g_i the “multiplicity of energy e_i”, i.e. the number of arangements of dof leading to the same energy e_i (which then will be indistiguishable from an energetic point of view). a is a normalisation constant so that the sum of probabilities is 1 (mass normalisation), b is another constant that can be adjusted to reach the imposed (energy normalisation) total energy. No need at this stage to relate b to T :-)

        For the problem at hand (column of perfect gas in a constant gravity field (aligned with first coordinate), isolated from the environment), e_i is m*(0.5*sum_(j=1->ndim)(v²_i )+g*x_1), multiplicity is 4.
        we thus have p(v_1,v_2,v_3,x_1)=4/exp(a+b*m*(0.5*sum_(j=1->ndim)(v²_i )+g*x_1))
        =[4/exp(a)*exp(-0.5*b*m*g*x_1)]*exp(-0.5*b*m*sum_(j=1->ndim)(v²_i ))

        On the other hand, the velocity distribution p(v_i) proportional to exp(-0.5*b*m*sum_(j=1->ndim)(v²_i )) is the well known Boltzmann distribution for T=1/(kb). It follows, as b is the same for all height x_1 in the colmun, that at all height, T is also the same …Problem solved!

        After that, the classical equilibrium can derive p and rho profile, or we can go further with Boltzmann distribution and use the m*g*x_1 term.

        I now do not have any doubt that Boltzmann was correct, even if I should still try to find a more “mechanical” explanation about why acceleration do not lead to higher mean velocity the further down you go. As Pikka have said, it is probably a subtle cancelling, and that’s the problem of the statistical approach: it is incredibly powerfull, but it hides the details that may help to intuitively grasp what is happening.

        Well, that ends it for me, B. wins against L. ;-)

      • We can also say that, for any potential field depending of particle position, The M-B distribution of velocities will the be the same at any point, hence a constant T. only density and pressure will change, at equilibrium. This is interesseting too.
        And to have non-constant T…we would need something that kinetic energy is not additive for other form of energy, some special coupling. I do not see any, except for drag…which then is not kinetic equilibrium thermodynamic, because it dissipate microscopic motions. I guess some version of it exists, that mix charged particles moving in fields and electromagnetic radiation to define a more general Temperature.

        I think I increased my understanding here, thanks to other participants :-) I am usually not a huge fan on “classical” thermodynamic, it sounded a little bit like recipies…but when you relate it to statistical approach, it indeed becomes fascinating physics…

        Thanks to all for taking time to examine this little though experiment :-)

      • To Kai, Pekka and others: I remember hearing once that when you first study thermodynamics, you have no idea what’s going on. Then after you teach it for a couple years, you think you know what’s going on. Then after many years, you realize you never did understand what’s going on!
        Kai, your derivation involves internal energy only.
        In classical thermodynamics, this corresponds to using C_v, the heat capacity at constant volume (e.g. in a bomb calorimeter). However, if gases are allowed to expand as they gain altitude, since work is done against an external pressure, you have to consider C_p, the heat capacity at constant pressure. From any college textbook on Physical Chemistry, C_p = C_v + R for each mole of particles of an Ideal Gas. Thus for a temperature change delta T, the enthalpy change delta H = C_p delta T = C_v delta T + R delta T. Thus the enthalpy = heat content contains a term = RT per mole due to compression from infinite gas particle separation. So as pressure decreases with altitude, there will be a temperature drop due to expansion, even for an Ideal Gas (the energy for the work done against outside pressure must come from the internal energy of the molecules, so temperature goes down). So far, Loschmidt was right.

        But as you correctly concluded, Boltzmann’s decreasing exponential function is also correct. In statistical mechanics, you can derive the formula as a function of bE, where E is the total energy (let’s for the moment ignore any complication due to degeneracy of any energy levels). Then in a plot of log P vs E, the slope will at any point be the slope of the tangent to the curve, and will equal -1/kT, where k = Boltzmann’s constant = R/Avogadro’s Number. If the entire graph is a perfect straight line, then the slope will be constant, and T will be constant (because the slope is negative, T will always be a positive value at equilibrium). But since even an Ideal Gas will produce a temperature gradient with height, the slope of the tangent must change slightly with height. Boltzmann’s “temperature” is constant as an infinitesimal, since it comes from the slope of a tangent. This is how temperature can be both constant and changing. So Boltzmann’s decreasing exponential function is correct, even if T changes:
        at any altitude, use the actual temperature there. At a higher altitude, T is lower, but just use that value in the formula; this just means that the slope of the tangent to the log P vs h curve will be steeper.
        So both Boltzmann and Loschmidt have been accommodated (you read it here first).

        What about the kinetic energy of the gas molecules, and how it changes as the molecules move up and down in the gravitational field? The Boltzmann formula uses total energy = gravitational potential energy + kinetic energy gained on falling down + kinetic energy of random gas molecular motion.
        The first two terms make up a constant term equal to the gravitational potential energy of a non-moving molecule, so it just gives the simple result already obtained.

        The last term is a function of temperature, which we said changes with altitude, even for an Ideal Gas. For a monatomic gas, this average kinetic energy would be equal to 3RT/2 per mole, or 3kT/2 per molecule [look up the theory of heat capacity for gases at the high temperature limit, where the Principle of the Equipartition of Energy holds]. Since the Boltzmann decreasing exponential function has in its argument the energy divided by kT, then the kinetic energy term, when separated from the potential by using the fact that the exponents are added when powers to the same base are multiplied, gives exp[-3/2]. This is a constant factor to be multiplied times the decreasing exponential function of the potential energy of a non-moving molecule. Since the same constant factor can be factored out of the state sum (partition function) which must appear in the denominator of the distribution function, the kinetic energy constant factors cancel, leaving the simple result already obtained.

        The only complication for an Ideal Gas is that the value of T is a function of h, due to the work of expansion against an external pressure. But Boltzmann’s formula is right.

        For real gases, the heat capacities are not exactly constant, especially for lower temperatures. This is especially important if vibrational energy states of real molecules must be included, since at normal temperatures most molecules are in the lowest quantum vibrational energy state and are stuck there unless they get a whopping jolt of energy. Thus vibrational states do not follow the Equipartition of Energy rule except at really high temperatures.

        Kai, I think this solves the hypothetical problem, and also shows how thermodynamics really hurts your brain!

      • Roger and Kai,
        Thermodynamics is certainly rather difficult. At my university one earlier professor (who understood it well) joked: “There are three people in the world, who really understand thermodynamics – we meet annually in Paris.”

        Perhaps it is not quite so difficult, but it is true that many people have used it extensively for years and still have serious misconceptions.

        The classical thermodynamics is an abstract mathematical theory. All of it and more can be derived from more basic physical principles applying statistical methods to classical or quantum physics. Most results are same for both, but some differ substantially and can be explained correctly only by quantum mechanics. The most obvious case concerns differences between fermions (particles with spin 1/2, 3/2, ..) and bosons (particles with integer spin).

      • Roger Taguchi

        Curse you, Kai! I couldn’t get this problem, though solved, out of my head and realized how I could derive the lapse rate for the atmosphere directly from the results. It would assume no extra heating from either the condensation of water vapour, or the absorption of IR by water vapour and CO2, and no convective mixing, so would work best at the lowest point in temperature on a clear night over a vast desert.

        Before I started my calculation, however, I decided to check by Googling “lapse rate”, which led me to
        (I have never taken a course on climate science, and have never read more than a couple pages of any climate science textbook). I was a little disappointed that the problem had already been solved by others [Click on the links to “lapse rate calculation from first principles” and “Heat capacity ratio”]. However, I enjoy calculating stuff from first principles, and I did learn how to defend Boltzmann’s derivation of the barometric formula , while at the same time showing how Loschmidt was also right.

  89. Hi Judith, you say “ .. I doubt that the publishers want to keep making more chapters publicly available .. ” but
    John O’Sullivan is (and perhaps also his co-authors are) absolutely convinced that “..One of the greatest open debates of modern times concerning the greenhouse gas theory is set to start in earnest. Critics predict an almighty global warming showdown between ‘deniers’ and ‘doomsayers.’ .. ” ( and seem to be depending on this “fact” to attract funds to help this august body of “experts” set up their “not for profit” private company Principia Scientific International (PSI – Of course, John is proposed as the CEO of this august body “ .. at the cutting edge of challenging falsities in science whether it is purveyed by corporations, governments or misguided individuals .. ”.

    I should have thought that knowing that their scientific analyses are rock solidly irrefutable the authors and the publisher would be delighted to support “the greatest open debates of modern times” by making the content of their “best seller” freely available for peer review on your blog. After all, John has already declared QUOTE: Top Scientists in Heated Debate over ‘Slaying’ of Greenhouse Gas Theory” .. O’Sullivan emailed Dr. Curry and challenged her to debate .. both parties finally agreed to an open debate to begin next week on Curry’s blog .. UNQUOTE (

    For “the Slayers” to refuse to make the contents available to debaters they would leave themselves open to (justified?) accusations from supporters of the CACC doctrine of cowardice. I don’t think that they or the publishers would want that. On the contrary, they have another book coming out in the Spring and many others planned so, if their expressed confidence in the validity of their arguments is more than simply marketing hype, in the medium and long term they will benefit enormously from such on-line peer review of this first book.

    Why not drop the “Slayers” and the publishers an E-mail accepting the challenge on the proviso that the contents are made freely available? I have all of their E-mail addresses which I can E-mail to you. I can even draft out the whole E-mail if you don’t have time. My close involvement with them during the past 5 weeks has now virtually ended as a result of what I refer to as my “PSI & Due Diligence” activities. They are all aware of my position regarding their claims, their future plans and the risk of attracting ridicule so it would create no problems on that score.

    If you prefer then send me an E-mail and we can discus it privately.

    Baa Humbug, can you explain to a non-scientist why one part of the earth, having absorbed a broad band of E/M radiation from the sun during daytime and warmed up as a consequence, cannot then start radiating IR into the cold of outer space when the sun is feeding energy to the other side of the globe and heating it up there? Please for the moment keep it simple by ignoring non-radiative energy transfer mechanisms. I know that’s a big simplification step but we can then perhaps build more into the model.

    Best regards, Pete Ridley

  90. Peter317, I think that up to a point we (and Kai) are saying the same thing – no energy added, only its rate of release slowed down. As I see it, without any atmosphere the earth would retain that energy which is otherwise carried away from the surface by convection hence would get hotter than without any atmosphere. All that the atmosphere does is help to keep the earth cooler then return that energy when the source is switched off, keeping the earth warmer than otherwise, but the total energy absorbed and emitted by the earth atmosphere system remains unchanged overall, i.e. mean temperature remains unchanged. You say “The atmosphere would, depending on the size of the losses, gradually heat up to something approaching the highest daytime surface temperatures” (February 1, 2011 at 1:35 pm) but surely this depends upon the rate at which energy is exchanged between earth and atmosphere. I haven’t come across any analysis of that, have you?

    The moon hasn’t much of an atmosphere, does it, and “ .. the temperature of the Moon can dip down to -153°C during the night. .. the temperature of the Moon in the day can rise to 107°C” ( This gives a mean moon surface temperature of –23°C. Hey, add that 33C “greenhouse effect” that Phil (Felton) was talking about (at 9:42 am) and we have 10°C, pretty close to the claimed “ .. average temperature on Earth is about 13°C .. ” ( Maybe there is a “greenhouse effect” after all! Is anyone aware if the hypothetical mean temperature of the earth has been calculated for a hypothetical atmosphere with no “greenhouse gases”?

    Pekka Pirilä, insulating my home hasn’t made it any warmer, it’s simply stopped it cooling down so fast because the insulation hinders the outflow of energy when I switch the heating off overnight, but I think we are using too simple an analogy, don’t you? I get the impression that we can go on exchanging opinions on this without getting anywhere because we need to know a bit more about rates of exchange of energy between this hypothetical atmosphere and the real earth. There could be some sums involved there so come on you experts (Chris Colose, Josh – AKA Eli – , Roger) what fundamental aspects are we overlooking here?

    On second thoughts, I’m happy that there’s a “greenhouse effect” so, as maybe, as Judith said earlier, “ .. I don’t see any point to debating the latter ..” (December 1, 2010 at 12:54 pm).

    Best regards, Pete Ridley

    • Pete – The greenhouse gases raise the mean temperature of the Earth and atmosphere. Without an atmosphere, or without greenhouse gases in the atmosphere, the Earth would be much colder. The effect is greater at night, when the sun isn’t shining, but the warming is also present during the day, adding to the solar contribution. This is apparent in the observational record, which shows rising temperatures during both day and night, but with a narrowing of the diurnal difference.

      I just happened to be browsing, so if you respond, I may not see it. However, standard explanations of the greenhouse effect and radiative transfer should help clarify the point. Of the names that appear above, Judith Curry, Phil. Felton, and Pekka Pirila, among others, would also be well qualified to answer you.

    • Pete,
      I think my analogy is not at all too simple, when the question is whether the added CO2 may lead to increased temperature of the earth surface. At this level the issue is really as simple as noting that the house gets warmer with increased insulation, when outside temperature does not change and heating is kept at the same level.

      It is unbelievable that people continue to argue on this trivial point.

      • Roger Taguchi

        I agree with Pekka on all points.
        Pete, not only do greenhouse gases such as CO2 and H2O slow down the rate of heat loss via IR radiation lost to outer space, they both raise the mean temperature of the Earth, as you stated (sarcastically? I hope not).
        The reason is as follows: only incoming Solar radiation warms the Earth’s surface (not much heat escapes from the molten core any more). Therefore if the rate of heat loss at night is reduced (for example, by increasing greenhouse gas concentrations), this also reduces the rate of heat loss in the daytime. The same incoming Solar radiation would then increase the daytime temperature. But the increase is tempered by the 4-th power Stefan-Boltzmann law of emission (if temperature rises by a large amount, the rate of heat loss by IR radiation increases by a huge amount). Because the slight mean temperature increase carries over to the nighttime (the mean by definition is the average of the daytime and nighttime values), the rate of heat loss at night is now by the 4-th power law higher than before. Since the black body radiation emitted in the daytime also rises, the total of the rate of radiation loss at night and during the daytime must once again balance the incoming Solar radiation (which hasn’t changed). So greenhouse gases both slow the rate of heat loss to outer space and raise the mean surface temperature. To really understand this, calculate the black body radiation lost to space using the Stefan-Boltzmann law from a mean temperature of 288 K, from a nightime temperature of 286 K, and from a daytime temperature of 290 K. Then subtract, say 4 W/m^2, from the heat loss rate at night, and calculate the value of T that would make up for this loss of heat. The rise in nighttime temperature must also be the rise in daytime temperature (for small temperature changes), meaning a rise in the mean temperature. I call this calculation Semi-Global Warming, since the physics of temperature change during a day must be the same as that over the long term, whether caused by CO2 changes or cloud cover changes. Use a value for the albedo of about 0.28.

        I have derived a simple connection between the amount of forcing and the resulting temperature change (i.e. temperature sensitivity) by differentiating the Stefan-Boltzmann Law. For details , contact me at

      • Roger Taguchi

        Oops, lousy wording in the reply above.
        On energy balance, the net loss to outer space at night must not change. The nighttime temperature rises, and the greenhouse gases transfer energy by collision to N2 and O2 molecules in the air, which warm up, but cannot re-radiate IR radiation to outer space. The term “reduces the rate of heat loss by IR radiation” applies only to the “stress applied to the system” (as in Le Chatelier’s Rule in chemistry); after energy balance has been achieved (at a higher mean temperature), the nighttime heat loss to outer space must be the same as before for a given albedo.

      • Roger Taguchi

        Sorry, Pete, Pekka and others, I rushed my previous replies because Tuesday evenings I play ntn trivia at the New Edinburgh Pub here in Ottawa (we compete against 3000 other bars in Canada & the USA), so they may sound like gibberish.

        Here’s a visual explanation of what I mean: first examine a MODTRAN computer calculation of the absorption spectrum such as the one at .
        The Stefan-Boltzmann law for emissivity 0.98 and T = 288.2 K gives a total energy output rate of 383 W/m^2. This would correspond to the total area under the smooth black body curve for 288.2 K. But the graph says that at 300 ppm CO2 the outward IR escaping at altitude to outer space is 260 W/m^2. The difference of 123 W/m^2 is the power absorbed by greenhouse gases such as H2O, CO2 and O3, and doesn’t escape by radiation.
        Therefore the amount of radiation scattered isotropically or as backscatter is irrelevant; the 123 W/m^2 is transferred during inelastic collisons from greenhouse gases to N2 and O2 molecules’ translational and rotational motions (i.e. the lower troposphere warms up). The N2 and O2 molecules have zero electric dipole moment and cannot emit IR photons, either to outer space or to the ground. The 123 W/m^2 corresponds to the total area between the smooth black body curve and the jagged spectrum, which shows absorption peaks going downward.

        What happens when CO2 increases from 300 ppm to 600 ppm?
        The area between the smooth black body curve and the CO2 valley at 667 cm^-1 increases by 3.7 W/m^2. Because the central frequencies are saturated, the increase in area shows up in a widening of the valley. This means a rise in the temperature of the air, after collisions have transferred energy from the excited state (v=1) bond-bending vibration of CO2 to translational and rotational energies of N2 and O2 molecules, which cannot emit IR photons. By conduction and backscatter, the solid and liquid Earth warms up slightly. How much? As T increases, the Stefan-Boltzmann law says a new black body curve will be emitted, at a slightly higher value at each frequency, with a very slight shift in the location of the peak. This increases the total area under the Planck black body curve by 3.7 W/m^2, exactly the right amount to compensate for the extra power absorbed by the increased CO2 and transferred as heat to the atmosphere. So the power eventually escaping at altitude to outer space goes back to 260 W/m^2, as it must, to maintain energy balance (since the amount of incoming Solar insolation hasn’t changed, and we have assumed no change in the albedo). By the way, the graph shows the change is 3.4 W/m^2, not 3.7 W/m^2, perhaps reflecting the range of error possible in this value. In any case, the value of outgoing IR at altitude of 256.72 W/m^2 for 600 ppm CO2 is to be thought of as the “stress” applied to the system. As in Le Chatelier’s Principle in chemistry, the system adjusts to counteract this stress, by increasing the surface temperature of the solid and liquid Earth
        (since the air cannot emit black body radiation, only resonant frequencies of greenhouse gases), until the steady state value of 260 W/m^2 escaping to outer space is reached.

        What is the surface temperature increase? Differentiating the Stefan-Boltzmann law with respect to T gives 1.02 Celsius degrees, an accurate value for “temperature sensitivity” derived from basic physics, not an experimental correlation of inaccurate measured average values. To complicate matters, the apparently truncated absorption valley for CO2 at 667 cm^-1 has been incorrectly interpreted in the MODTRAN simulation as evidence for a 220 K emitting layer at 20 km (which cannot exist if you understand that true black body radiation must come from solid or liquid or plasma which can emit a virtually infinite number of different frequencies to produce a continuous spectrum over the entire range of electromagnetic frequencies). I have explained the apparent truncation as due to the superposition of IR fluorescence powered by incoming Solar radiation at very high altitudes (as a guess, at 80-100 km) on top of a deep CO2 absorption valley at essentially zero transmission at central frequencies. This means that CO2 really does have central frequencies “saturated”, with nearly zero transmission from the ground up to 20 km, instead of 33% escaping after radiative transfer from the ground, a value logically inconsistent with the idea of “saturation”. Therefore I have estimated the temperature sensitivity upward from 1.02 to 1.5 Celsius degrees, to take into account the underestimated amount of CO2 absorption in the lower troposphere. Since there has been a climate change from 1850 to 2010 of 0.7 +/- 0.1 degrees as CO2 increased from 280 to 380 ppm (approx. 300 to 400 ppm), then the predicted future warming as CO2 increases from 400 to 600 ppm is 1.5 – 0.7 = 0.8 degrees. On the other hand, the IPCC projection for the same time period, using the actual climate change that has already occurred, is 3 – 0.7 = 2.3 degrees, a factor of 3 times larger than 0.8 degrees. It would be even worse if their error in interpreting the truncation had not partially cancelled the larger error in assuming a positive feedback of 2 degrees added to a climate change of 1 degree due to increased CO2 alone. Hope this discussion clarifies matters.

        If you want detailed calculations of the above, and explanations of basic Molecular Spectroscopy and Atomic Spectroscopy, I can forward you email seminars if you contact me at

      • Are you implying that all Earth’s energy budget is radiated out by way of CO2 spectra?

      • Roger Taguchi

        Hi hunter!

        I’m not sure what you mean by “CO2 spectra”.
        I do NOT mean that energy budget is due solely to CO2.
        I DO mean that energy budget is due solely to radiation outward to empty space, in the IR region of the spectrum, since conduction and convection in empty space is negligible. In the absence of any gases capable of absorbing resonant IR frequencies (e.g. on the Moon), the radiation emitted would follow the smooth black body curve. Greenhouse gases absorb some of the IR emitted from the solid and liquid Earth, and transfer that energy during inelastic collisions with N2 and O2 molecules which cannot re-radiate that energy as IR. So the air warms up.

  91. Roger Taguchi

    Pete Ridley makes an excellent point regarding the Moon, which receives about the same amount of Solar insolation as the Earth (measured in W/m^2, so the difference in cross-sectional areas is irrelevant). During the night, it can lose energy only by radiation of IR to outer space (which is at 3K, the temperature of the cosmic microwave background radiation); this is how net heat flows from a hot object (the Moon) to a cold one. Because the Stefan-Boltzmann relation says heat lost by radiation varies as the 4th power of the absolute temperature, the rate of heat loss at 120 K is (250/120)^4 = 18.8 times faster at the average temperature of 250 K than at the low point of 120 K at night (there might be a lag period). During the daytime, the heat loss rate at the highest point is (380/250)^4 = 5.33 times that at the average temperature. Compared to the lowest point, heat loss at the highest point is 100 times that at the lowest temperature.
    The Earth, however, has two mitigating factors: (1) large bodies of water that store energy during the daytime/summer and can transfer energy via winds and currents during the nighttime/winter, and (2) greenhouse gases such as CO2 and water vapour. As Pete has shown, they can account for our present comfortable conditions here on the Earth. I have calculated that where water vapour is absent (i.e. over large expanses of desert), temperatures can swing from 2 Celsius at night to 46 Celsius in the daytime
    (from freezing cold to blistering hot, something which is in agreement with experience – at altitude in the Atacama Desert, water actually freezes at night). Without the CO2, temperature swings would be even more severe.
    I can send details on request to me at

    • There is one more difference that makes the temperature variations so large on Moon, the length of the day, which gives time for the ground to heat and cool.

      • Roger Taguchi

        Good point, Pekka!
        I wrote my reply to Pete off the cuff, without looking up any reference books, so I missed that. Nice to see someone on the ball, and that we can agree on corrections without rancor.

  92. In response to Pekka’s comments I was initially tempted to say that we shouldn’t labour this point because I think that basically we are all on the same wavelength when discussing our unrealistic model of the earth/atmosphere system (in which the source of energy is the sun and the only energy transport mechanism is radiation). Both the hypothetical atmosphere that Baa H described and the real world one with large amounts of water vapour and other trace “greenhouse gases” act like an insulator by reducing the rate of emission of energy by IR radiation from the earth. On reflection maybe it is worthwhile, at least for me, spending some more time on it, so I hope that no-one is getting bored with giving help to those of us who are struggling with the science. I also hope that the following is not too disjointed.

    Roger, thanks for taking the time to do what appears to be a very useful summary of your Nov. 2009 article “Net Feedback in Global Warming Calculations” and the content of some of your E-mails from early last year. (Has anyone taken you up on your offer to provide a copy of your article?) As I understand it, that article is an update of your August 2009 article “Mistakes in IPCC Global Warming Calculations” ( The revisions followed your discussions with Dr. Jeffrey Glassman ( and less detailed exchanges with science student Chris Colose (see the QUOTE: in my comment of November 1, 2009 at 12:28 pm on (Unfortunately your exchanges with Chris are no longer readily available because Australian ex-Senator Fielding’s blog has been discontinued – unless you saved a copy.)

    I posted those E-mailed comments of yours straight onto Josh Halpern’s “Eli can retire Part VIII – The EPA reads Rabett Run” thread ( The responses from Josh and his supporters were all highly critical of your arguments until I posted the ones that you repeat here, after which the exchanges ended with no further response from any of them. I wonder why.

    (I see that Judith makes reference to Chris and Josh in her article and Chris has contributed several comments above.)

    I am finding these exchanges of opinion very useful and I thank you all for being so patient, because you are helping me sort out a few puzzles. One of the (many) outstanding ones, which I alluded to (February 1, 2011 at 4:05 pm) when I referred to “ .. rate of release .. ” of energy reduced by the insulating effect of the atmosphere, is how rate of energy exchange combines with the different heat capacities of the earth and the atmosphere. My instinctive reaction is that the atmosphere will quickly change temperature with these exchanges of energy, whereas the earth’s temperature will remain relatively constant (suggesting that my point on February 1, 2011 at 4:05 pm about “ .. keeping the earth warmer than otherwise .. ” might have better been stated as “ .. cooling the atmosphere quickly to that of the earth .. ”). NB: This is in contrast with Baa Humbug’s ultra-simple model of the earth/atmosphere system where there are no “greenhouse gases” and the earth is simply dry dust and rock (December 2, 2010 at 1:25 am). BTW, Roger, I wasn’t being sarcastic about the increased mean temperature of the earth/atmosphere system, understanding that if energy cannot escape so readily because of the insulating effect of the atmosphere then mean energy increases therefore mean temperature increases. The only problem I have is with the extent of that increase.

    This is a topic that Dr. Glassman discussed last March in “THE CAUSE OF EARTH’S CLIMATE CHANGE IS THE SUN” ( saying that “ .. Oceans, because of their mass, their heat capacity, and their color, are the dominant mechanism of Earth’s energy balance between the Sun and space. The atmosphere as a reservoir plays a minute role .. ”. During subsequent exchanges he says “ .. I would argue that you need to compute the heat flux, and not just rely on surface temperature differences, on Earth or the Moon .. ”. Any experts among you prepared to comment on this in relation to our exchanges about the insulating effect of the atmosphere, with or without “greenhouse gases”?

    When giving a reasonable answer to “ .. what have we been doing wrong in terms of explaining this .. ” (November 30, 2010 at 1:48 pm) BlueIce2HotSea also referred repeatedly to the rate of heat flow between earth and atmosphere and Phil. Felton (February 1, 2011 at 9:42 am) also made mention. I learned a little about the basic scientific principles required in order to understand these debates properly but despite that, and after decades without having to use what I learned, I struggle, mainly with unfamiliar terminology but also with what are to me strange concepts. One example is the reference to heat flow, rather than energy flow. I have no problem with energy being transferred by radiation, convection or conduction but “heat” leaves me cold (pardon the pun). I’m not in the least surprised that those without any science background (the vast majority? especially of politicians) find it way beyond their ken.

    BTW Roger, how do you react to those comments made by A Lacis on December 2, 2010 at 2:04 am? Also, it would be helpful if the pdf’s that you have available to send to those of us who are interested could be readily searched using the acrobat tool. The two that I have (your 29th Nov 2009 article “Net Feedback in Global Warming Calculations” and “Molecular Spectroscopy 0001”) are scanned copies. Do you have alternative versions?

    Kai. Further to Roger’s comment at 10:50 am, are the profile and associated text from “Atmosphere, Weather and Climate” by Barry, R. and R. Chorley ( the idiot’s guide that I use) of any use to you? (Pages 35 & 36 in )

    Best regards, Pete Ridley

  93. Hi Pete!
    I sent you a private reply, but here’s a public comment.
    My original article “Mistakes in IPCC Global Warming Calculations” was the result of 2 days’ work starting from scratch. It correctly derived an accurate value for “temperature sensitivity” from basic physics, not inaccurate and questionable experimental correlations. This is the 1 degree rise expected for a CO2 increase from 300 ppm to 600 ppm, not including feedback. I did not know at the time how to calculate net feedback.

    I also pointed out the fundamental lack of understanding of “black body radiation” shown by the IPCC report, which seems to assume that the atmosphere can actually emit a continuous black body spectrum over the entire electromagnetic spectrum, peaking in the IR. It can’t. Only a condensed state such as solid or liquid, or a plasma (such as at the surface of the Sun) can emit a true black body spectrum. But there IS outgoing IR detected by the NIMBUS satellite at CO2 frequencies. But this is due to fluorescence at very high altitude, not black body radiation. Ignorantly calling it black body radiation led to the conclusion that the tip of an absorption peak is a “temperature probe” that “measures” the black body temperature, instead of it simply being the value of the net transmission (or absorption); this is laughably wrong to any competent organic chemist. There are zillions of other absorption peaks in the entire spectrum observed by the satellite: do they all measure different “black body temperatures”? If so, what does this mean for the temperature profile of the atmosphere? The 220 K black body emitting layer at 20 km does not exist (it’s at 220 K, but not black body emitting). The entire spectrum observed by the satellite is simply 288 K black body radiation emitted from the warm solid and liquid Earth, except for absorption by greenhouse gases in the lower atmosphere PLUS CO2 and O3 fluorescence powered by incoming Solar radiation at very high altitudes where the pressure is so low that the chance of collisional quenching is small compared to the probability of emission of an IR photon.

    Collisional quenching is the mechanism of energy transfer near the Earth’s surface from “hot” vibrationally excited CO2 molecules to the surrounding “cold” N2 and O2 molecules , which subsequently cannot re-emit energy as IR photons. Since the v=1 state of the CO2 molecule contains several times the average kinetic energy of “cold” air molecules, statistical mechanics says that the most probable state at equilbirum will have that excess energy distributed among all the other modes (translational and rotational) of motion of all the molecules after collision, with each mode taking a slice of that excess. Net result: the air gets warmer. After solving Kai’s problem involving Boltzmann and Loschmidt (see previous postings in this thread), I realized that this way of thinking, obvious to me and Prof. John Nicol who know something about physics and chemistry, may be completely unknown to those whose backgrounds are in biology, journalism, computer science or climate change “science”. Hence the sorry state of the “settled science” in the climate change literature. The “radiative exchange” theory is, as Nicol put it, “non-physical”, akin to magical thinking. For it cannot clearly explain to a layperson HOW the air near the Earth’s surface warms up. Enormous effort has gone into discussing isotropic scattering or backscattering, which cannot by themselves warm the air, for the photon energy remains the same. Backscattering may be observed by a spectrometer looking upward from the ground, but it does not affect the net energy flow outward predicted by the Stefan-Boltzmann law, because backscattered photons are re-emitted on the other side of the constant mean temperature Earth (alternatively, you can say the energy reabsorbed by the Earth as a backscattered photon on energy balance must be immediately re-emitted outward as another photon). The derivation of the Stefan-Boltzmann law is for NET outward flow of energy, so backscatter is irrelevant. See

    At any rate, the theory of radiative exchange to a 220 K black body emitting layer at 20 km is immediately falsified by the observed excess CO2 emission seen over a 210 K thunderstorm anvil and over an even colder Antarctica on a clear day. The “theory” is just wrong, wrong, wrong, and must be abandoned.

    There were two weaknesses in my original article, which were corrected in my article “Net Feedback in Global Warming Calculations”. This took me another 2 days on a separate occasion, and I also figured out how to calculate net feedback (it’s essentially zero). Increasing water vapour (a powerful greenhouse gas) with increasing temperature is apparently balanced by a slight increase in cloud cover (an increase in albedo from 0.28 to 0.29 leads to a decrease in climate of 1 degree). I can send you details if you contact me at
    The first correction involved the Beer-Lambert Law. At low absorbances (up to about 15%), absorption is linear with increasing concentration, but due to saturation effects, this is not exactly true at higher values. In my Triangle Model for absorption profile, I ignored this non-linearity on saturation because I was lazy. I did it right in “Net Feedbacks”.
    The second error was more subtle, but vastly more important conceptually. The Triangle Model (which I made up because it explains “high saturation” conceptually) assumes a constant wheelbase for an absorption profile. But the IR P- and R-branch wings of a CO2 absorption band are decreasing exponential functions. So even a small absorption value like 0.001 (0.1%) will, after 10 doublings of CO2 = 2^10 = 1028, increase to saturation levels! So I had to abandon the Triangle Model and calculate the absorption profile from the rules of molecular spectroscopy (which I had previously avoided out of laziness). This also applies to John Nicol’s paper. He easily calculates that a single CO2 frequency (line) rapidly reaches saturation, so further increases in CO2 will not change the amount of energy absorbed. He claims this occurs within hundreds of metres from the Earth’s surface. But the CO2 molecular absorption band consists of many, many vibration-rotation lines, and it’s the tiny absorptions at the extreme wings that continue to absorb IR energy on increasing CO2 concentration, even if the central frequencies have long since been saturated (at 99.9% absorption, or more). So Nicol’s Atomic Line Absorption Model must be abandoned (molecular absorption bands do not work exactly the same as single atomic spectral lines). On request, I can forward email seminars on Atomic Spectroscopy and Molecular Spectroscopy to those who want to learn more, but shudder to think of plowing through Gerhard Herzberg’s 3-volume masterpiece on Molecular Spectroscopy (Herzberg won the 1971 Nobel Prize for Chemistry). My email address is

    • Roger,
      How can you conclude that IPCC assumes “that the atmosphere can actually emit a continuous black body spectrum over the entire electromagnetic spectrum”. That claim in definitely without basis.

  94. If you talk about 220 K black body radiation, that’s what you are assuming! Think about it. Look at the shape of the entire black body curves. You can actually see 210 K black body emitted by a thunderstorm anvil, which is made up of tiny solid ice crystals (or maybe supercooled liquid droplets). But none of this is added (or needs to be added) to the 288 K black body curve emitted by the solid and liquid Earth to get the observed spectrum over the warm Earth on a clear day. By entire electromagnetic spectrum I mean mainly IR, although visible & UV etc. would theoretically be there in decreasing exponential amounts. So I’m definitely not saying you can detect black body UV emitted from the Earth. Black body emission means a continuous spectrum of radiation. Period. It’s not there at 20 km of clear atmosphere.

    • Roger,
      Sometimes the reports are not written carefully on points that are clear to all scientists of the field. They may refer to the temperature that leads to the same total radiative power or make some other simplifications that are not explained in full, but they are not stupid and all authors of the reports know very well, the general features of the radiation within the atmosphere as well as the properties of the outgoing radiation at the TOA.

      There are hardly any serious errors in the whole 950 page report of the WG1. Some details are perhaps controversial, but not obviously wrong.

  95. Roger Taguchi

    Correction to my email discussion with Kai on Boltzmann vs Loschmidt:

    The Boltzmann barometric formula gives the ratio of the pressure (or density) at height h compared to that at the ground (h=0). This is
    P/P0 = exp[-mgh/kT]/exp 0 = exp[-mgh/kT], since exp 0 = 1.

    When the total energy = gravitational potential energy + kinetic energy of the molecules, we get exp [-(potential energy + kinetic energy)/kT]
    = exp[-mgh/kT].exp[-kinetic energy/kT] in the numerator of the ratio, using the property of multiplication of powers with the same base.
    Thus the potential energy ratio gives the same result as before, but the total ratio involves the ratio of exp[-kinetic energy/kT] at the two different heights, and even for an Ideal Gas, temperature varies with height h!
    Luckily, the kinetic energy of a monatomic gas is 3kT/2, so exp[-kinetic energy/kT] reduces to exp[-3/2], which is a constant, and doesn’t depend on the temperature! Therefore the same expression evaluated at the ground or at any height will be the same, even if the temperature changes with height! Therefore the ratio of the expressions involving kinetic energy reduces to 1, so the overall barometric formula is exactly the same as before! The only complication is that T is a function of h, so the pressure and density do not vary exactly as a decreasing exponential function of h, but varies more-than-exponentially in a decreasing fashion. At any altitude h, the actual value of T must be used in the barometric formula, so Boltzmann’s assumption of “constant T” really means for over an infinitesimal height change, there will be only an infinitesimal temperature change (the slope of the tangent to the log P vs h graph will be constant for an infinitesimal height change).
    In my discussion of Feb. 4, 11:08 am, I mentioned the partition function (state sum), which is in the denominator when you want to know the fraction of molecules with a given total energy; however, in this case we want only the ratio of pressures, so you don’t use the partition function.
    I realized this when I went over the derivation in my head, and would have caught the mistake earlier if I had actually written anything down at the time, instead of just typing off the cuff.
    In my discussion here, I used the internal kinetic energy of a monatomic gas; if you want to be picky and include the work of expansion against constant pressure, then the heat capacity per molecule is 5k/2 instead of 3k/2, so the energy is 5kT/2, and the fraction becomes exp[-5/2] divided by itself, which again gives 1. If you want to use air as a diatomic gas (both N2 and O2 are diatomic gases which are very close in behaviour to Ideal Gases), then the heat capacity per molecule is very close to 7k/2, and again exp[-7/2] divided by itself gives 1. So Kai, this hypothetical problem is actually very useful in modelling the lapse rate (change in T with altitude) in the atmosphere. You can look up the theory of heat capacity of gases at

  96. Roger Taguchi

    For the 288.2 K Earth, assuming emissivity = 0.98, the Stefan-Boltzmann law says that the power output per square metre is
    j = 0.98(5.67 x 10^-8 J/(s.m^2.K^4)) (288.2 K)^4 = 383.3 W/m^2.
    If 260 W/m^2 escapes at night, at energy balance, the total Solar power reaching the Earth’s surface must be 2(260) = 520 W/m^2 in the daytime. The 260 & 520 W/m^2 are measured perpendicular to the Earth’s surface (i.e. up or down at ground level). Since the area of a hemisphere (e.g. the daylit half of the Earth) is 2.pi.r^2, and the area of a circle (e.g. the circular cross-section of the Earth) is pi.r^2, the net incoming Solar radiation to the cross-sectional area of the Earth must be 2(520) = 1040 W/m^2. Since the Solar insolation reaching the circular cross-section of the Earth is 1371 W/m^2, this means 1371-1040 = 331 W/m^2 must be reflected outward, and not reach the Earth’s surface.

    The Solar insolation value of 1371 W/m^2 may be checked at or calculated from the Stefan-Boltzmann law for a 5780 K Sun with emissivity = 1 (it is essentially a perfect black body).

    This means the albedo is 331/1371 = 0.24, a little on the low side compared to literature values of around 0.30. The reason is that the IPCC number of 260 W/m^2 shown on the MODTRAN simulated spectrum at underestimates the power absorbed by CO2 near the Earth’s surface, because they calculated only the truncated area of the CO2 absorption curve in the spectrum. This truncation is actually caused by CO2 fluorescence in the upper atmosphere, powered by incoming IR photons, part of the black body spectrum emitted by the 5780 K Sun.

    Because eyes glaze over and minds tune out whenever there is hard math, I will use only arithmetic in the following sample calculations.
    EXAMPLE 1:
    During the daytime, the Earth warms up (Semi-Global Warming) and at night it cools down. Suppose the average temperature goes up by 2 degrees in the daytime, from 288.2 K to 290.2 K, a relative change of 2 x 100%/288.2 = 0.694%. Then for emissivity = 0.98, the Stefan-Boltzmann law says the emission of IR from the solid and liquid surface of the Earth is
    j= 0.98(5.67 x 10^-8)(290.2)^4 = 394.1 W/m^2.
    Therefore the net IR outflow has increased by 394.1 – 383.3 = 10.8 W/m^2. This is a FORCING, an increase in j by 10.8 x 100%/383.3 = 2.82%. This is 2.82/0.694 = 4.06 times the % increase in temperature.

    During the nighttime, let’s assume the average temperature drops by 2 degrees (the same size as the daytime increase), from 288.2 K to 286.2 K, a relative change of -0.694%.
    The power output from a 286.2 K solid and liquid Earth is
    j = 0.98(5.67 x 10^-8)(286.2)^4 = 372.8 W/m^2. The change in j is therefore 372.8 – 383.3 = -10.5 W/m^2.
    This FORCING is a change in j of -10.5 x 100%/383.3 = -2.74%, which is 2.74/0.694 = 3.96 times the % decrease in temperature.

    The average of the daytime and nighttime powers emitted is (394.1 + 372.8)/2 = 383.5 W/m^2, just slightly higher than the 383.3 W/m^2 emitted at the mean temperature of 288.2 K. The slight discrepancy comes from the fact that output varies as T^4, not as T, so increases are not linear.

    Note that the average of 4.06 & 3.96 is 4.00% for the ratio of the % change in power output to the % change in temperature. Could it be exactly 4?

    Because the average of the power output changes is (10.8 + 10.5)/2 = 10.65 W/m^2, the temperature sensitivity is 2 degrees/10.65 W/m^2 = 0.188 K/(W/m^2), compared to a typical literature value of 0.8 K/(W/m^2). The literature value is a factor of 4.3 times larger.

    Because of saturation effects at central frequencies, CO2 absorption is almost 100% (i.e. transmission at central frequencies is essentially zero, not 33%). I calculate that this means 14 W/m^2 more is absorbed by CO2 compared to the IPCC total of 42.4 W/m^2 which corresponds to the area of the truncated absorption valley. This FORCING of 14 W/m^2 reduces the IR lost to outer space from 260 to 246 W/m^2. This results in a temperature change of 2.6 degrees, so the mean temperature drops from 288.2 to 285.6 K. A black body curve at this 285.6 K more closely fits the envelope of the actual outgoing spectrum published at than does the one at 288.2 K.

    If only 246 W/m^2 escapes to outer space at night, then 2(246) = 492 W/m^2 reaches the Earth’s surface in the daytime. Therefore 2(492) = 984 W/m^2 is the net amount of Solar insolation reaching the Earth’s surface when we compress the power spread out over a hemisphere down to a circular cross-section. Therefore 1371 – 984 = 387 W/m^2 is reflected back to outer space, so 387/1371 = 0.28 is the albedo, much closer to the literature values of around 0.30 than the IPCC value of 0.24.

    EXAMPLE 2:
    Let us repeat the temperature sensitivity calculation for a 2 degree daytime increase from a 285.6 K Earth.
    At 285.6 K, j = 0.98(5.67 x 10^-8)(285.6)^4 = 369.7 W/m^2. A 2 degree temperature rise in the daytime is a % increase of 2 x 100%/285.6 =
    0.700 %, and the power output at 287.6 K is
    j = 0.98(5.67 x 10^-8)(287.6)^4 = 380.2 W/m^2.
    The power output has increased by 380.2 – 369.7 = 10.5 W/m^2, an increase of 10.5 x 100%/369.7 = 2.840%. This is a factor of 2.84/0.700 = 4.06 greater than the % increase in temperature.

    If nighttime mean temperature is 285.6 – 2 = 283.6 K, this is a temperature change of -.700% from the daily mean.
    The power output at night is
    j = 0.98(5.67 x 10^-8)(283.6)^4 = 359.4 W/m^2. The change in power output from the mean is 359.4 – 369.7 = -10.3 W/m^2,
    a % change of 10.3 x 100%/369.7 = 2.786%. This is a factor of 2.786/0.700 = 3.98 greater than the % temperature change.
    Note again that the average of 4.06 & 3.98 is (4.06 + 3.98)/2 = 4.02. Could it be really exactly 4?

    The average of the daytime and nighttime power output changes is (10.5 + 10.3)/2 = 10.4W/m^2.
    Therefore the temperature sensitivity is 2 degrees/10.4 W/m^2 = 0.192 K/(W/m^2).
    Note that this is different (by 2% for a temperature change of only 0.70%) from the value of 0.188 K/(W/m^2) calculated in EXAMPLE 1, so temperature sensitivity is not constant. Therefore values evaluated at really low temperatures (e.g. -25 Celsius = 248 K) will be significantly different from those at 288 K. This could have a bearing on correlations of experimental data used to assess climate sensitivity.

    Both EXAMPLES show that (a) the % change in black body power output is at energy balance about 4 times the % change in the temperature of the solid & liquid Earth emitting the infrared (IR) photons, and (b) the IPCC value for temperature sensitivity is slightly more than 4 times that obtained from basic physics (the Stefan-Boltzmann law). COINCIDENCE????

    [skip this if your eyes are glazing over]:
    The derivative of T^4 with respect to T is 4T^3. Therefore as the Stefan-Boltzmann law is
    j= epsilon.sigma.T^4 (1), where epsilon and sigma are constants,
    then taking the derivative of both sides with respect to T gives
    dj/dT = 4.epsilon.sigma.T^3 (2).
    Now dividing both sides by j = epsilon.sigma.T^4 and
    “cross-multiplying the dT” [something which gives the right answer, although it horrifies Calculus professors] gives
    dj/j = 4 dT/T (3).
    Now replace infinitesimals by differences, giving
    delta j/j = 4. deltaT/T (4)
    This says that the relative (or %) change in power output is exactly 4 times the relative (or %) change in absolute temperature. This is just fundamental physics I derived from scratch, and allows us to calculate the temperature change for any FORCING; we don’t need to look at experimental correlations which are highly inaccurate or possibly non-relevant. It is confirmed by the approximate calculations in my two EXAMPLES.

    Rearranging equation (4) gives
    delta T/delta j = T/4j (5)
    for the functional form of the temperature sensitivity (the rate of change of the temperature of the black body emitting surface with the change in power output). Note the factor of 4 which now appears in the denominator,
    which means that temperature sensitivity is 1/4 the value it would have if power output were a linear function of T rather than a 4th power.

    SPECULATION: Is it possible that the IPCC and climate science literature value for climate sensitivity is a factor of 4 too high simply because someone in the early days assumed incorrectly that the relative (or %) changes in T and j were EQUAL? This assumption shows ignorance of the fact that the Stefan-Boltzmann law varies as T^4, not linearly as T.

    For if we use T = 288.2 K, j = 383.3 W/m^2 and a forcing value for “delta j” of 3.7 W/m^2, then the predicted climate sensitivity is 0.70 degrees. If the error in 3.7 is +/- 0.1 (+/- 2.7%), then the error in 0.70 degrees is +/- 0.02.
    If the error in 3.7 is +/- 0.3 (+/- 8%), a possibility indicated by the quotation of 3.4 on the MODTRAN graph, then the error in 0.70 degrees is +/- 0.06. Thus the IPCC “best value” for climate sensitivity of 3 degrees is a factor of 3/0.7 = 4 larger, way outside the possible error bars. But once a value like 3 degrees is established, it is not surprising that correlation after correlation can be found “supporting” this value of 3 degrees, even though the experimental error bars are huge, large enough to cover everyone’s ass!

    After correction for truncation error in the IPCC report, the true value of climate sensitivity is increased to 1.2 +/- 0.1 degrees on increasing CO2 from 300 ppm to 600 ppm, but does not take into account any possible feedbacks. However, using this value predicts that there would be a measured temperature change when CO2 increases from 300 ppm to 400 ppm of 0.57 +/- 0.14 degrees. The historic record from 1850 to today for a CO2 increase from 280 ppm to 380 ppm (which, due to less saturation, should give a slightly higher temperature change) shows 0.7 +/- 0.1 degrees. Therefore net feedback must be less that 0.13 +/- 0.24 degrees, essentially zero within experimental error.

    Thus I predict that future climate change when CO2 increases from 400 ppm to 600 ppm will be 1.2 – 0.57 = 0.63 degrees, whereas the IPCC prediction would be 3 – 0.7 = 2.3 degrees, a factor of 2.3 /0.63 = 3.7 higher.

  97. Roger Taguchi

    Pete Ridley asked me to explain how the greenhouse effect actually occurs with a mechanical model. In a previous posting on this site on Feb. 2, 2011 at 1:13 am, I gave an explanation in terms of energy balance, which is essentially correct. However, it does not convincingly explain the step-by-step causal chain of events by which the greenhouse effect actually occurs.
    CASE 1:
    It might help if you draw a circle divided vertically in half to represent the Earth, with the Sunlit half on the right. Draw arrows pointing from the Sun to the Earth, at a power/m^2 level of 2000 units. [The value of the Solar insolation is 1371 W/m^2, but let’s round up for easy visualisation.] If the Earth is at a constant mean temperature, it cannot have a net input or output of power/m^w2 to outer space, so on average there must be 1000 units radiating outward on the nighttime side (you can draw smaller arrows pointing to the left) and 1000 units radiating outward on the daylit side (show smaller arrows pointing to the right). Then the 2000 coming in during the daytime would be balanced by the 1000 + 1000 = 2000 radiating outward over the whole Earth. Of course, as I showed in the previous posting on Feb. 7, 2011 at 7:44 pm, the Earth actually warms up during the daytime (Semi-Global Warming) and cools down from the mean at nighttime by the same amount, for small temperature changes. [On the Moon, because of the huge temperature swing during one rotation, the power loss at night is only 1% of the power loss during the daytime peak. This is accomplished by having a large temperature rise in the daytime radiating most of the incoming power immediately back outward to space, so that there is only a small net input that can be balanced by a small output to space at nighttime. The Stefan-Boltzmann 4th power relation makes this possible, so that temperature swings downward at night never reach absolute zero, while temperature rises in the daytime are unbounded.]

    CASE 2:
    Suppose we now add a layer of greenhouse gas near the surface of the Earth in CASE 1. Draw another circle divided in two, and relabel the power/m^2 of 2000 units coming in from the Sun. At this stage it is important to remind ourselves that the 5780 K black body radiation emitted from the surface of the Sun is mostly in the visible and near-infrared (near-IR) regions of the electromagnetic spectrum. Assume a black surface for the Earth (this is obviously not true for a multi-coloured Earth, but the 2000 is the net amount coming in, after reflection is taken into account). That 2000 units of power/m^2 consists of photons of whopping amounts of energy each, and can boost molecules to very high vibrational levels, or even excited electronic levels. When equilibrium has been reached, however, that concentrated energy will have been distributed to give the most probable distribution of energy, the Boltzmann distribution. Heat will have flowed from one “hot” molecule to many modes of motion (vibrational, rotational and translational) of many, many molecules. The net result is a rise in surface temperature. But the Stefan-Boltzmann law says black body power output varies as the 4th power of the absolute temperature T. Thus even small changes in temperature (e.g. 3 %) will result in a large change in power output (in this example, by 4 x3% = 12%). Let’s assume only a small temperature change that results in a power output of IR (where the black body spectrum peaks at temperatures around 300 K) at the Earth’s surface from 1000 to 1200 units. But if both daytime and nighttime surfaces radiate 1200 units outward, that would total 2400 units outward from the Earth against the same 2000 coming in from the Sun. That would lead to a cooling of the Earth. Unless there were something in between the surface of the Earth and outer space that could absorb some of the 1200 units and keep it from leaking out. Let’s call it a greenhouse gas (or gases). In this example, exactly how much power/m^2 would have to be absorbed by the greenhouse gases and transferred via inelastic collisions to N2 and O2 molecules which cannot re-radiate IR, resulting in warming of the atmosphere to a steady state temperature profile? It’s easy! The difference between 1200 and the 1000 that MUST be the net outflow at altitude to balance the same 2000 units coming in from the Sun! That is, 200 units must wind up heating the atmosphere to keep the same 1200 output at the Earth’s surface, as well as the same 1000 output leaking into outer space.
    The increase in output at the Earth’s surface will be 200 units, an increase of 200 x 100%/1000 = 20%. Since I have derived (see previous postings) by differentiating the Stefan-Boltzmann law the result that the % change in temperature is 1/4 the % change in the power output, this means there will be a change in the surface temperature of 20%/4 = 5%. For example, at T = 288 K, this would mean a climate change of 5 x 288/100 = 14.4 degrees for the greenhouse effect. Pete, I trust this finally removes any lingering doubts about understanding the greenhouse effect.
    You read it here first. You might check to see if any climate textbooks have explained the greenhouse effect this way, since I have never read more than a couple pages of any textbook.
    My previous explanations were mathematically equivalent to the above, but were a little harder on he brain.
    CASE 1 was the same as before.
    CASE 2 involved keeping the same initial 1000 units radiating outward from the Earth’s surface both in the daytime and the nighttime. Then greenhouse gas placed around the Earth absorbed 200 units and transferred this energy via inelastic collisions to air molecules, so the air warms up. Thus only 800 units escape at altitude both in the daytime and at night. The total flowing outward would then be 800 + 800 = 1600 units. Since the Solar insolation is still 2000 units, there would be a net power inward of 400 units on the Sunlit side. This would increase the temperature of the sunlit side, and the result would be 200 units more output at the Earth’s surface both at night and in the daytime. 200 units added to the initial 1000 would give 1200 units for the steady state output at the Earth’s surface, and 200 units added to 800 at altitude would once again restore the necessary 1000 units output day and night to balance the 2000 input during the daytime.

    In any previous discussion mentioning backscatter, I goofed. Although backscatter does occur, and can be detected by pointing a spectrometer upward from ground level, it does not affect the Stefan-Boltzmann law, which is derived for net outward flow of photons. The energy of any backscattered photon would increase the energy/temperature of a surface that absorbed it, and result in the immediate release outward of another photon which would take away that added photon energy, for a surface at a steady state temperature.

    I also goofed in my posting of Feb. 2, 2011 at 1:13 am, when I derived a value for temperature sensitivity (not corrected for IPCC truncation error)of 1.02 instead of the correct value of 0.70. I should have used the power output of 343 W/m^2 at the Earth’s surface rather than the value of 260 W/m^2 at altitude. That’s what happens when you blindly substitute into a formula without truly understanding the physics. The final conclusions are basically the same, however. The IPCC values for temperature sensitivity and predicted future climate change are too large by a factor of 4, and the net feedback is essentially zero, within experimental error.

    I thank Pete Ridley for pressuring me to clarify my thoughts, which allowed me to spot mistakes before they became established as “settled science”.

    First, the “radiative transfer” theory is a good one – FOR THE INTERIOR OF THE SUN. Rayleigh scattering is a good theory – to explain why the sky is blue. Both theories, however, have little to do with the greenhouse effect in the Earth’s atmosphere, as I wish to show in this article. The Ideal Gas Theory is a very good one – for gases at normal temperatures and pressures. However, I am not disproving or denying centuries of established science when I say that the Ideal Gas Theory must be abandoned when we deal with solids (which do not follow Boyle’s Law, for example).

    Beneath the convective zone in the Sun’s interior, radiative transfer is the way in which heat is transported outward from the nuclear fusion zone at the Sun’s core. See
    “The average Mean Free Path [of a photon] at the centre of the Sun is about 10 cm. A typical photon produced in the Sun’s core might undergo 10^19 collisions before reaching the photosphere 10 million years later at this rate. In actual fact the mean free path increases as you go outward from the core, so that the real transit time is more like 200,000 years.” Gamma rays produced by nuclear fusion reactions are scattered inelastically by charged particles such as free electrons in the Sun’s interior (Compton scattering). See . This means that gamma rays lose some of their initial energy, which is transferred to the recoiling electrons, and the frequency of the scattered photon becomes lower. By Maxwell’s Equations, charged particles (such as electrons) which are accelerated will emit electromagnetic waves (which we now see as a stream of photons). This phenomenon is exploited in the creation of radio and television waves, and in the production of microwaves in a magnetron vacuum tube (electrons forced to travel in a curved path in a magnetic field emit microwaves whose frequency can be continuously varied). In high-energy particle accelerators, much of the power consumed comes out as synchroton radiation beamed in the forward direction by relativistic effects as charged particles are forced to move in curved paths in strong magnetic fields.
    In the Sun’s interior, any time an electron changes in velocity (in speed or direction) in a close encounter with another charged particle, it is by definition accelerated, and therefore photons in a continuous spectrum will be created and absorbed. At any distance from the Sun’s center, there will be a steady-state temperature and pressure established by numerous collisions, with a temperature gradient from the center toward the photosphere at 5780 K, where the photons can finally escape. This occurs because the atoms run out (for a Sun of finite mass). Although the Sun is a ball of gas, its mean density is 1.41 g/cm^3, a little denser than liquid water at the Earth’s surface. Beneath the Sun’s surface, photons are readily scattered by the high density of charged particles, so the Sun’s interior is opaque to visible light (where the Solar spectrum peaks). Once the density suddenly plummets at the photosphere, the mean free path of the photons becomes essentially infinite, and they can travel in straight lines outward from the Sun to the Earth and beyond. Since the Earth’s atmosphere is transparent to visible light, we can readily see the Sun’s photosphere as a disc. The spectrum of the Sun is essentially that of a 5780 K black body, given by the Planck radiation law. The total power output can be calculated by multiplying the Stefan-Boltzmann flux times the surface area of the Sun.
    There are a few Fraunhofer lines due to absorption of certain frequencies by atoms or ions in between us and the Sun, but this does not change the essential nature of the black body spectrum, which is a CONTINUOUS spectrum over the entire electromagnetic range, peaking in the visible and near-infrared (near-IR). See

    During a total Solar eclipse, however, we can see the Sun’s corona, a tenuous “atmosphere” which is at a temperature of millions of degrees. Because the density of charged particles is so low in the corona, they are not coupled with the photons, and therefore it is wrong to consider the corona as an emitter of BLACK BODY radiation. Thus the power output of the Sun is essentially that of a 5780 K black body alone, and we must not add a “corona black body flux” which would vary as (10^6)^4 = 10^24 = 1,000,000,000,000,000,000,000,000 compared to the photosphere’s output which varies as (5780)^4, which is “only” 1.12 x 10^15.

    Now consider the spectrum of the solid and liquid Earth, which is at a mean temperature of 15 Celsius (288 K). The Planck radiation law gives a continuous spectrum peaking at around 15 micron wavelength, which happens to be the wavelength for the bond-bending fundamental vibration of the CO2 molecule (at 667 cm^-1). At this point, it is important to know that condensed states (solids and liquids) can produce a continuous black body spectrum because there are many, many weak electrical forces between molecules close to each other, and these correspond to low vibrational frequencies. These low frequencies, when combined with the higher bond frequencies WITHIN molecules and their overtones (multiples), produce the many, many frequencies in a complete continuous black body spectrum. In contrast, the electrons in bound states (in electrically neutral atoms and molecules) have quantized energy levels, and can absorb or emit only at certain frequencies. It is important to know that the vibrations and rotations of molecules also follow the rules of quantum mechanics, with only certain energy levels allowed. Therefore the IR vibrational and rotational spectra of gas molecules consist of widely separated frequency BANDS (atomic spectra consist of widely separated LINES). Therefore to talk about “220 K black body emission” from ANY layer in the Earth’s atmosphere (at 10 km or 20 km, wherever) is to show lack of understanding of gas molecules and of CONTINUOUS black body radiation. For the Earth’s troposphere is made up of neutral gas molecules, not free electrons, protons and photons like the interior of the Sun.
    The three gas molecules (N2, O2, Ar) that make up almost 100% of dry air do not possess permanent electric dipole moments, and therefore cannot absorb or emit IR radiation, black body or otherwise. Therefore they cannot impede the flux of IR photons from the 288 K solid and liquid Earth to outer space, both during the daytime and at night. This outward flux (in W/m^2) for a mean steady state temperature must just balance the net incoming flux of Solar radiation during the daytime, after reflection outward from clouds, water surfaces, etc. has been taken into account (the albedo). Trace amounts of greenhouse gases such as CO2 and H2O (water vapour) CAN absorb IR photons emitted from the solid and liquid Earth.
    Although the O=C=O molecule is linear, with a net zero electric dipole moment in the equilibrium configuration, both asymmetric stretch and bond-bending produce changing electric dipole moments that can interact with certain electromagnetic frequencies, and lead to absorption and emission of those frequencies. The H2O molecule is bent, and so is IR-active in symmetrical stretch as well as during asymmetric stretch and bond-bending.

    The IR vibration-rotation absorption spectrum of the diatomic molecule HCl is available at
    The spectrum of the polar HCl molecule shows two broad “wings”, a P-branch at lower frequencies corresponding to changes in the rotational quantum number J of -1 (e.g. from J=1 to J=0, from J=2 to J=1, etc.) and an R-branch at higher frequencies corresponding to changes in J of +1 (e.g. from J=0 to J=1, from J=1 to J=2, etc.). These rules reflect the Law of Conservation of Angular Momentum, since integer values of J measure quantized angular momentum, and the photon as a particle of spin 1 carries 1 unit of angular momentum. The spacing between the individual vibration-rotation lines is roughly constant, and depends on the value of the rotational constant B which is inversely proportional to the moment of inertia. Because there are two common isotopes of chlorine with slightly different masses, the HCl spectrum actually shows two slightly separated spectra, one for the HCl molecule containing the Cl-35 isotope, and the other the Cl-37 isotope. The author of the article is puzzled by the fact that the Cl-35 lines are not 3 times the height of the Cl-37 lines, since Cl-35 is roughly 3 times more abundant than Cl-37. He has failed to understand the onset of “saturation”: when absorbance is greater than about 15%, it is no longer linearly proportional to concentration. I.e. doubling concentration no longer doubles the height of the absorption peak. “Saturation” also explains why the shape of the envelope of the P- and R-branches does not exactly follow the calculated relative populations of the initial rotational energy levels. In order to get a pleasing spectrum, a high HCl concentration was used, but at the cost of loss-of-linearity.

    Most of the CO2 vibration-rotation absorptions consist only of P- and R-branches. However, the 667 cm^-1 band involves a changing angular momentum around the molecular axis (something not possible for a diatomic molecule like HCl). Thus a Q-branch consisting of “delta J” = 0 transitions (J=1 to J=1, J=2 to J=2, etc.) shows up as a narrow, intense spike in the gap between the P- and R-branches.

    Now for the true explanation of the CO2 greenhouse effect: a vibrationally excited CO2 molecule can (a) spontaneously re-emit an IR photon, (b) be stimulated by another IR photon of the same frequency, and therefore re-emit, or (c) transfer its energy in a radiationless collision with other gas molecules (primarily the surrounding N2 and O2 molecules that constitute the bulk of the atmosphere). The first possibility (fluorescence) would simply result in the same as elastic scattering, with no transfer of energy to the air. Because of quantized energy levels, there can be no gradual degradation of the energy/frequency of the IR photons, unlike the photons in the interior of the Sun which CAN undergo INELASTIC scattering. The second possibility (stimulated emission) is important in CO2 IR lasers, but as it requires population inversions, is not important for the Earth’s atmosphere. That leaves the third, and obvious possibility which no one else in climate science seems to emphasize. It is important to know that an excited CO2 molecule (with vibrational quantum number v=1) has more than twice the average kinetic energy of an air molecule, and so we can call it a “hot” molecule. On collision, there is a chance that the “hot” molecule on collision with a “cold” air (N2 or O2, most probably) molecule will be “quenched”, suddenly dropping down to the ground vibrational state (with vibrational quantum number v=0), and the energy difference somehow divided among all the possible motions of all the molecules (rotation and translation, most likely). The vibrational energy gaps for N2 and O2 are much higher than for CO2 bond-bending, and so excitation of N2 or O2 vibration is unlikely. At the bimolecular elastic collision level, the principle of microscopic reversibility holds: we cannot tell if a movie of the collision is going forward or backward in time. However, when a red-hot poker is plunged into a bucket of water, the poker is rapidly quenched, and the water gets slightly warmer. If we saw the reverse occurring, with cold water heating a cold poker until it was red-hot, we would know that we were watching a movie run backwards. So it is with the quenching of excited greenhouse gas molecules; the reverse is unlikely to happen, and so we avoid an endless loop of events which would result in no change. The air does warm up, and the N2, O2 and Ar molecules CANNOT re-emit IR photons which might escape to outer space. The greenhouse effect is real, and significant for the Earth, contrary to what the extreme climate change deniers would have us believe.

    What about Rayleigh scattering, which explains successfully why the daytime sky is blue? First, it is elastic scattering, so the frequency of the scattered wave is the same as the incident wave, and so there is no transfer of energy (the air does not get warmer or cooler). You can’t explain the greenhouse effect (warming of the atmosphere) with ANY form of elastic scattering: Rayleigh, Thomson, whatever. Secondly, it is not energetically significant for most of the day, for the Solar spectrum is unaffected except just at Sunrise and at Sunset when the Sun’s disc is reddened (by scattering of blue preferentially during the long path length through the Earth’s atmosphere at grazing angles). Most crushingly, Rayleigh scattering varies as the 4th power of the frequency, so for visible light with wavelength 5000 Angstroms = 500 nm = 0.50 microns, the scattering would be 30^4 = 810,000 times more important than for 15 micron CO2 IR radiation.

    At this point, it should be noted that Rayleigh scattering of photons from electrons in molecules takes about 10^-15 s, whereas the fluorescent lifetime of an electronically excited state is of the order of 10^-8 s. Therefore scattering of IR photons results not from Rayleigh scattering, but from absorption, followed after a delay, by re-emission. But since the energy of the photon has not been weakened, radiative exchange cannot explain the greenhouse effect in the troposphere.

    When there is a dispute between adherents of two different theories, the matter ought to be settled by examing the experimental evidence. A MODTRAN computer simulation of an actual satellite IR spectrum obtained looking down on a warm cloudless Earth in the daytime is shown at .
    The overall envelope of the actual spectrum is a black body radiation curve for 288 K (actually, 285 K would be even closer). This is the spectrum of IR photons emitted from the solid and liquid Earth travelling unimpeded to outer space, except at certain frequencies which correspond to resonant vibrations and rotations of CO2, H2O and O3 (ozone) molecules. The bites taken out of the 288 K black body curve correspond to the dark Fraunhofer lines that mar the otherwise continuous 5780 K black body spectrum of the Sun’s photosphere. I.e. the Earth’s atmosphere is transparent to IR photons except at the bites. There is absolutely no evidence for any additional 220 K BLACK BODY (i.e. continuous) spectrum escaping to outer space, any more than there is any evidence for 10^6 K black body radiation from the Sun’s transparent corona.

    What about the truncation of the CO2 absorption curve at 667 cm^-1, which corresponds to the height of a 220 K black body radiation curve?
    If the power/m^2 is restricted to the CO2 frequency band, it is due to fluorescence (powered by incoming Solar radiation in the high upper atmosphere); by definition, it CANNOT be black body radiation which would include all the frequencies across the width of the spectrum shown.
    This elementary mistake in interpreting the IR spectrum (laughably incompetent to any good organic chemist) has led to the acceptance of the utterly WRONG interpretation of the peak height as a “temperature probe” which somehow measures the “black body temperature of the emitting layer” of the Earth’s atmosphere. Total garbage. Ask any organic chemists used to seeing IR spectra of complex molecules with zillions of absorption peaks! They know more about IR spectra than any computer scientist or radiation physicist. Even the MODTRAN spectrum shows numerous peaks of differing heights. Can anyone seriously believe that each measures a different temperature at a different layer in the Earth’s atmosphere???? The different peak heights just reflect different absorptive strengths, that’s all.

    The MODTRAN spectrum actually shows two different computer curves for CO2 absorption (you’ll have to look carefully at the two very close, but differently coloured curves). One is for 300 ppm CO2, and the other for 600 ppm CO2. Why so little difference (about 8% of the total area of the CO2 absorption curve)? Because of “saturation”, doubling CO2 does not increase absorption by 100%, but only by about 8%. The increased absorption shows up only in the wings of the absorption curve, because the central frequencies have been completely saturated (have reached nearly 100% absorption, or 0% transmission). A logical person would find it strange, therefore, that the MODTRAN curve shows about 33% transmission at those central frequencies that are supposedly saturated!
    That’s part of the illogic of invoking a 220 K black body emitting layer instead of just asking a molecular spectroscopist who would have quickly deduced the explanation as due to fluorescence superimposed on a 0% transmission trough.

    It’s not just my handwaving say-so. Here’s quantitative proof of the Standard Model’s invalidity: substituting T = 220 K into the Stefan-Boltzmann law, and assuming emissivity = 1 (the maximum possible), the outgoing flux is predicted to be 130 W/m^2. But the actual amount escaping at altitude is 260 W/m^2, according to the MODTRAN graph.
    They can’t even get the thing they were trying to explain right!

    If we use 260 W/m^2 and emissivity = 0.98, the temperature of the so-called emitting layer would be 261.5 K = -12 Celsius. As a guess, this might be at an altitude of 3 or 4 km, within reach of mountaineers, pilots and balloonists. Has anyone ever found any evidence for a layer there where suddenly IR photons could escape to outer space? I’d bet the house that they haven’t, because that sudden discontinuity in the troposphere does not exist. It only exists in computer models which might have aped the concentric layers in the interior of the Sun. There IS a discontinuity at the photosphere of the Sun, because that’s where the atoms run out! Of course, a 261.5 K black body curve would no longer truncate the CO2 absorption curve at the “right” amount to explain the observed spectrum, so either way, the “temperature probe” idea is WRONG!

    If you want to see what 210 K black body radiation emitted from the Earth looks like, go to and go to the spectrum obtained by a satellite looking down on a 210 K Thunderstorm Anvil at 10 km altitude. Because the Thunderstorm Anvil is made of tiny ice crystals (or perhaps supercooled liquid droplets), a true continuous black body emission spectrum is emitted acrosss the entire frequency range. The bottom of the cloud blocks any 288 K or warmer black body radiation from the Earth’s solid and liquid surface from being seen by the satellite. You will note emission peaks at CO2 and ozone frequencies poking above the smooth 210 K black body curve. This excess power/m^2 CANNOT be explained by IR emitted from the ground or from the 210 K cloud (without violating the Law of Conservation of Energy). The “radiative exchange” theory has been falsified (and according to Karl Popper, the “theory” must be abandoned). This excess outward flux at CO2 and ozone frequencies is also seen in the spectra looking down on Antarctica, in another part of the link, and also at .
    How does the Standard Model handle this excess flux? By hypothesizing a “temperature inversion”: the emitting layer must be at a higher “black body temperature”, of course! This “explanation” invokes the ignorantly WRONG interpretation of the peak height as a “temperature probe”.

    Here’s another way of testing theory with experiment. Print out a copy of the Thunderstorm Anvil spectrum, and use a marker to trace out the envelope of the 210 K black body curve, slicing off the CO2 and ozone peaks. Now use a marker of a different colour, and carefully trace the actual spectrum (on the same page) looking down on a warm surface of the Earth, including all the bites. Now step back and see if there is any significant difference between the 210 K black body curve (which is only slightly lower at all frequencies than a 220 K black body curve) and the actual spectrum. Most fair-minded persons would say the match is poor or non-existent. The “220 K black body layer” must be rejected as a theoretical model lacking any real-world value.

    It may be easily seen now that the actual spectrum looking down on a warm surface is essentially a 288 K black body spectrum minus bites taken out at CO2, H2O and ozone frequencies PLUS some emission at central CO2 frequencies, enough to give an appearance of truncation. The exact match of the CO2 fluorescence emission and the truncation of the warm surface spectrum on the same graph means that even the IR emitted from the 210 K Thunderstorm Anvil at CO2 frequencies must be totally absorbed by CO2 molecules above 10 km and below the region of fluorescence, because otherwise the match would not be exact.

    Further evidence for the high altitude (as a guess, at 80 – 100 km) for the region of fluorescence comes from the existence of two Q-branch spikes in all the spectra except over Antarctica (where the signal-to-noise ratio is half, due to the low angle of the incoming Solar radiation). Because of anharmonicity (higher vibrations depart from the simple harmonic oscillator model), the Q-branch spike for the v=3 to v=2 transition occurs 19 cm^-1 to the left of the Q-branch spike for the combined v=2 to v=1 and v=1 to v=0 transitions. Unlike the CO2 absorption of IR emitted from a 288 K solid and liquid Earth, which involves only the v=0 to v=1 transition, the fluorescent emission is the envelope of v=1 to v=0, v=2 to v=1, v=3 to v=2, and v=4 to v=3 transitions (and possibly more). The P- and R-branches for the higher transitions are displaced slightly to lower frequencies than the combined v=1 to v=0 and v=2 to v=1 transitions.
    The two Q-branch spikes may also be seen in the spectrum obtained over Guam, reproduced at .
    Since the population of CO2 molecules in the v=3 vibrationally-excited state is negligibly small at 300 K (at the Earth’s surface), the appearance of the smaller Q-branch spike (a) completely invalidates the importance of excitation by radiation exchange from the Earth’s surface, and (b) shows that excitation must occur at very high altitudes where the pressure is so low that collisional deactivation of excited states is negligible compared to cascade downward in energy by spontaneous emission.

    These arguments should have been enough to invalidate the Standard Model involving radiative exchange of IR from the Earth’s surface to a hypothetical 220 K black body emitting layer at 20 km, from which IR photons can finally escape to outer space. But Thomas Kuhn in his 1962 “The Structure of Scientific Revolutions” explained that working scientists are resistant to “paradigm shifts”, especially from paradigms they learned as they were apprenticing.
    “Looking at the moon, the convert to Copernicanism does not say, ‘I used to see a planet, but now I see a satellite.’ That locution would imply a sense in which the Ptolemaic system had once been correct. Instead, a convert to the new astronomy says, ‘I once took the moon to be (or saw the moon as) a planet, but I was mistaken.’ ”
    This kind of quantum leap in understanding requires emotionally letting go of a once-comfortable notion. Scientists, being only too human, find it difficult to admit mistakes and the possibility that their careers may have involved work that was worthless. Kuhn may have taken some, if not all, of his celebrated ideas from the philosopher and physical chemist Michael Polanyi. You can read more at and .

    • Roger has written far too much to start commenting, what in his writing is correct and what seriously wrong. From the total text most is correct and this also in full agreement with generally accepted views, but where his results differ, they are wrong. Sometimes he also claims that his results differ, although they do not. He has just misinterpreted the general understanding.

      I pick now just one point.

      Roger writes:

      What about the truncation of the CO2 absorption curve at 667 cm^-1, which corresponds to the height of a 220 K black body radiation curve?
      If the power/m^2 is restricted to the CO2 frequency band, it is due to fluorescence (powered by incoming Solar radiation in the high upper atmosphere); by definition, it CANNOT be black body radiation which would include all the frequencies across the width of the spectrum shown.
      This elementary mistake in interpreting the IR spectrum (laughably incompetent to any good organic chemist) has led to the acceptance of the utterly WRONG interpretation of the peak height as a “temperature probe” which somehow measures the “black body temperature of the emitting layer” of the Earth’s atmosphere. Total garbage.

      This accusations are nonsense. Nobody is trying to say that the atmospheric CO2 would be radiating as a blackbody. What happens is that the emissivity of a thick enough layer of CO2 in the 15 um band is very close to one. Therefore the strength of radiation from such a layer of CO2 has the same intensity as blackbody radiation of the same wavelength, when the temperatures of the blackbody and the layer of CO2 are the same. Therefore comparison with blackbody spectra tells indeed, what is the temperature of the atmosphere in those layers, where the outgoing radiation escapes. The layer is not exactly at one temperature, but at wavelengths with the strongest absorption it is thin enough to have a well defined temperature.

      All who have drawn such pictures as the one Roger refers to, understand this fully and correctly, it is Roger, who didn’t understand it.

      There are numerous other errors, but there is no reason to go to all of them. For everybody it is a better choice to just read some of the well established descriptions of these processes. The book of Pierrehumbert is certainly too much for many, but descriptions of variable level of detail can easily be found to select from.

      • Come on Pekka, I think that it is not good enough to pick up on one point and say “Roger has written far too much to start commenting, what in his writing is correct and what seriously wrong. .. but where his results differ, they are wrong. Sometimes he also claims that his results differ, although they do not. He has just misinterpreted the general understanding”.

        Surely the onus is on you to fully define and substantiate your claims. After all, Roger has gone to a lot of trouble to present his argument clearly and fully.

        Best regards, Pete Ridley

      • Pete,
        Everybody must decide, how far it is reasonable to go.

        When Roger has written several so long messages, the effort of going through all of them is too much – and they do depend on each other so much that one should discuss all of them.

        Most of the basics is correct, as I already said, but all these messages are on that part of climate science, where the existing understanding is good and uncertainties small. Roger has presented so much correctly that I am sure, he will find his remaining errors and misunderstandings rapidly, when he decides to look for them instead of claiming that the common knowledge is in error. I may have used the word “numerous” in a place, where “several” would be more appropriate, as he doesn’t really differ from the common thinking in