Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

CO2 cooling?

  1. Aug 26, 2008 #1

    G. V. CHILINGAR, L. F. KHILYUK, and O. G. SOROKHTIN, 2008, Cooling of Atmosphere Due to CO2 Emission, Energy Sources, Part A, 30:1–9, 2008 ISSN: 1556-7036 print/1556-7230 online DOI: 10.1080/15567030701568727

    Heads and tails of the study:
    I can follow that logic. Discussion?
    Last edited: Aug 26, 2008
  2. jcsd
  3. Aug 26, 2008 #2
    I don't follow the logic at all. You need warming due to absorption of infrared radiation leading to expansion and more convection and then you get a net cooling? But if you get a net cooling you don't have the warming you need to explain the expansion and the extra convection you need. It is self contradictory.
  4. Aug 27, 2008 #3
    are not the upper levels of the atmosphere near the tropopause cooler then expected
    where you get the .2 atm
    while the lower air near surface air is warmer

    so hot or cooler may depend on how high up you are
  5. Aug 27, 2008 #4
    Okay, Indeed, it's not happening on the same levels. Let me elaborate.

    First some understanding about convection:


    Note that convection is visible most of the time as it produces cumulus clouds, thunderstorms, tornadoes, hurricanes.

    Hence, convection is a means of transporting and distributing heat-energy within the atmosphere, causing the higher levels to be warmer than the natural http://www.bbc.co.uk/weather/features/understanding/lapse_rates.shtml [Broken]. Why? because convection due to solar heat only works at day time, not at night, when the Earth cools due to radiation. This only makes the air in contact with the surface cooler and hence more dense, which prevents convection backwards.

    However, the warmer air that has convected to higher atmospheric layers also radiates out heat, basically 50% to the Earth and 50% to the universe, also cooling in the process, not only at daytime but also at night!

    Now the article argues that more greenhouse gas enhances the transfer of heat energy from the surface to the lower layers of the atmosphere, which obviously gets warmer faster but this amplifies the rate of convection and also the rate of cooler air from aloft replacing the convecting air, neutralizing the warming effect at the surface.

    At the higher altitudes however, the increase of concentration of greenhouse gasses makes the energy radiation out of the atmosphere more effective, which increases the cooling rate, which is supposed to balance the increase of energy at the surface.

    What they don't even consider, is that the overal effect of the 24/7 day night operation of enhanced radiative cooling in the upper atmosphere should compensate more than for the daytime only increase of thermal energy convection. Hence increase of greenhouse gasses seems to cause a net stronger radiative cooling.

    Does that help?

    Notice again that the greenhouse effect hypotheses attributes the difference between blackbody temperature and actual temperature attributes to radiation, whereas in reality this difference is mainly caused due to the one way convection of energy, only up, not down, just like a vertical conveyor belt..
    Last edited by a moderator: May 3, 2017
  6. Aug 27, 2008 #5
    I could see inscreased co2 in the atmosphere eventually having a cooling effect. According to global warming increased co2 in the atmosphere inscreases the greenhouse effect and increases temperature. If it is true that temperature is increasing will that not increase evaporation resulting in heavier cloud cover and rain. Also melting of ice caps should cool the ocean or slow down the currents which carry heat away from the equator. I suppose this could also have a negative impace on regions near the equator.

    Anyways I live in south east ontario and this summer was not very summerish at all. It was not as hot as it should have been and it rained a hell of alot. We broke 100 year old rainfall records in some areas. There was hardly any days with clear blue skys. Even the nicest days were still filled with cumulus clouds. Shady or even rainy periods. I have been wondering if its just a freak summer or if it could have something to do with climate change.
    Also the farmers almanac calls for a colder then normal winter. Just awesome.
    Last edited: Aug 27, 2008
  7. Aug 27, 2008 #6


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I think qualitative dynamics like this doesn't mean much, especially when different dynamics work in opposite directions: one should put them in a computational model and see how all this behaves. In fact, as long as one doesn't have a computational model that is 1) based upon sound physical basis (and not on empirical parameter fitting black box dynamics) and 2) corresponds to genuine experimental data, I think speculation is running loose.
  8. Aug 27, 2008 #7
    But it's a start, since classic greenhouse effect ideas might underestimate the energy transport by convection, it's unlikely that the current models do justice to the effect of convection.
  9. Aug 27, 2008 #8


    User Avatar
    Gold Member

    This may be totally naive, but why can not scientists fill a chamber with gasses in some proportions and measure it properties to reflect, absorb whatever light/heat.
  10. Aug 27, 2008 #9


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Because the physics is not scale invariant. Nor is there any kind of easy scaling theory for any of this.
  11. Aug 27, 2008 #10
    Indeed, there isn't, moreover the processes are not linear. You could compare the double effect of heating and cooling with a tap opening a bit more (heating) to fill a bucket with a hole in the bottom, which we make a bit bigger (cooling). there is no way to tell what the water level is going to do, without quantification.

    Although you could quantify both warming and cooling rates logaritmically proportional to the density of the greenhouse gas, but quantifying the increase in convection rate would be rather complicated especially when the role of water evaporation and condensation is taken into consideration.

    It may be noted that problems in the basic mechanisms of the greenhouse idea have been signalled by Miskolsci here as well as Douglass et al here.

    Bottom line however from the study remains that it makes a case that the atmosphere is primarily heated by convection instead of radiative effects. It appears that this would reduce the effect of radiative forcing heating or greenhouse effect considerably.
    Last edited: Aug 27, 2008
  12. Aug 28, 2008 #11
    It may be recalled that also our cbacba had some issues with his model here.

    It seems that Chilingar et al 2008 pretty much have answered that question, convection. Curiously enough I learned that already some four decades ago, going for my glider flying permit. However, there is very little mentioning of convection in the IPCC 4AR WG1 climate issues. Why?
    Last edited: Aug 28, 2008
  13. Aug 28, 2008 #12
    But surely if CO2 becomes more important to heating at higher altitudes, then this would make the atmosphere more buoyantly stable. In other words, wouldn't increasing CO2 have the opposite effect to that proposed, ie, it would tend to inhibit convection??
  14. Aug 28, 2008 #13
    The idea is, that it becomes more important at cooling the higher altitudes by radiating out the convected energy.

    Alternately consider the null hypothesis, suppose that there was no greenhouse gas in the atmosphere. It would still be heated by conduction at the lowermost levels, which would still generate convection.

    But since the convection is one way traffic only, without the possibility to loose that heat energy by radiation, the atmosphere would continue to heat up by convection until the lapse rate was to become stable, preventing further convection, then the atmosphere would be a lot warmer than without the radiative cooling at higher levels.

    hence, in this simplified setting, it appears to be the greenhouse effect, the capability to radiate out the heat in the higher atmosphere, that keeps convection going.
    Last edited: Aug 28, 2008
  15. Aug 28, 2008 #14
    Why is cooling by radiation more important at higher altitudes? I can see how the inverse would be true: that heating by absorption would be more important as a mechanism of heat transfer at higher altitudes, simply because there is no surface to conduct heat to the air at high altitudes. I guess what we really need to know is the net effect, in terms of relative importance, as to the effect of greenhouse gases to heat transfer as a function of altitude. If it turns out that air at high altitude emits radiation at a higher rate than it absorbs then this is unstable and convection may ensue.

    Of course, we have convection, and we can explain it without the need to resort to this effect. Then we may reasonably ask, how important is this effect? What is the relative order of magnitude of this effect in relation to the standard model of atmospheric convection: hot air rises, expands and cools, advects, and sinks. Afterall, the concept of the Hadley cell, which fits global observations of convection patterns neatly, does not rely on this effect. This might suggest that this effect, if it exists at all, is negligible.
  16. Aug 29, 2008 #15
    See the OP. Jack:

  17. Aug 29, 2008 #16
    Another way of looking at it is measuring the temperature/altitude of the outgoing radiation emitted by CO2 and H2O

    http://www.bom.gov.au/weather/satellite/about_satpix.shtml#wV [Broken]

    So if these are the normal atmospheric levels that radiate IR energy out, as a consequence, these levels should logically be cooling.
    Last edited by a moderator: May 3, 2017
  18. Aug 31, 2008 #17
    Andre, the quote is meaningless in terms of my objections. I'm not objecting to the assertion that convection is important, and that if there were more convection we would lose heat quicker and thus cool, I agree with that point. I was objecting to whether the presence of CO2 actually had a sensible impact on atmospheric convection.

    I'm sorry, you're going to have to spell this one out to me. I really don't follow your point.
    Last edited by a moderator: May 3, 2017
  19. Aug 31, 2008 #18


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Also, in this context, perhaps someone might like to comment on the following paper.

    Lunt et al, Nature 454, 1102-1105 (2008)
  20. Sep 1, 2008 #19
    That would well be worth its own thread, Gokul, as one post would not do justice to tons of publications about this kind of material.

    Furthermore, from the abstract only, there is little chance to comment on the scientific merit. But we see 'model' and then they "find", etc. Now, recall the scientific method, what is the role of a model? Merely to work out a hypothesis. A hypothesis can be considered only substantiated (not proven), if the forthcoming predictions (i.e. from models) are confirmed with solid evidence and we can't really see any of that in the abstract.

    Secondly, working out a single hypothesis to explain correlation is inferior to comparing two hypotheses, exchanging cause and effect. Such a competing hypothesis could be: variation in temperatures may cause variations in atmospheric CO2 due to the variation in oceanic capacity with temperature to store CO2. So what is cause and what is effect?

    If we agree that cause comes before the effect, then the ice cores produce no support for the idea that CO2 changes lead to temperature changes. As apparently temperature changes lead CO2 with some 600 years in this case of the last glacial termination:




    Yes there is something about positive feedback but before using that argument it may be an idea to find the publication first that demonstrates the physical feasibility of such a mechanism using these ice core data. I'm not aware of its existence.

    Furthermore there are examples of higher temperatures coupled with moderate CO2 concentration, in the Paleocene for instance: http://www.sciencemag.org/cgi/content/abstract/sci;292/5525/2310

    Incidentily, a "needle" was found under the ice sheet of Greenland when drilling the NGRIP ice core.


    Later it proved to be willow bark (not published, communication with the authors) What would that say about the age of the Greenland ice sheet?
    Last edited: Sep 1, 2008
  21. Sep 1, 2008 #20
    it appears that we have a misunderstanding here, Jack I'm sorry. What is sensible? Chilingar et al attribute 8% to radiation (greenhouse effect). Estimates of the CO2 contribution to radiation, run from 5 to 25%, the rest mainly being water vapour. This would seem to get you at a maximum of 2%, disregarding feedbacks.

    trying again. As the satellites measure the IR frequency finger print of the outgoing radiation it is clear that it is radiated from these middle trophospheric levels and mostly by water vapor. So, if this is the out radiating source, then it is also place where the atmosphere cools effectively, since net out radiation means losing energy. Obviously this cooling is in some dynamic equilibrium with heat sources from below, in which convection seems to dominate. Hope it helps
  22. Sep 1, 2008 #21
    Query for "convection" in the IPCC report,


    See also fig 2 FAQ 2.1. pp 136, no convection mentioned.


    So chap 3 acknowledges convection as a predominant process of something, even energy mentioned, however chap2 does not seem to take convection into account as a heating mechanism of the atmosphere like Chilingar et al 2008. One would at least expect it to be a "forcing" factor in chap2.

    http://ipcc-wg1.ucar.edu/wg1/Report/AR4WG1_Print_Ch03.pdf Convection not mentioned
    Last edited by a moderator: May 3, 2017
  23. Sep 6, 2008 #22
    The paper by Chilingar, Khilyuk, and Sorokhtin 2008 entitled "Cooling of atmosphere due to CO2 emission" needs to be read carefully in order to see its logic, which has several steps. If the logic of the paper is not followed carefully, then its message is not conveyed.

    Step 1 is to say that convection is most important, and should be the backbone of a heuristic argument.

    Step 2 is to say that CO2 has an immediate effect on convection by virtue of its molecular weight. This immediate direct effect makes the atmosphere more dense. Taking into account the concurrent immediate radiative effect through the coefficient Cr in a model based on convection leading to near adiabatic gas law conditions, this leads to a warming of less than 0.01 K for a doubling of the present CO2 level.

    Step 3 is to note that the mere immediate effect of addition of CO2 is not the whole story. There are feedbacks to CO2 addition: in particular, added CO2 is partly absorbed by the ocean and deposited as insoluble carbonates; the particular carbonates involved have the effect of taking oxygen from the atmosphere because of the reductive capacity of FeO. It is also partly replaced by oxygen when plants use it as food. Unlike the "feedback on radiative forcing by CO2" doctrine of the Arrhenius devotees, these are true feedbacks. These feedback effects have the side-effect of longterm reduction of atmospheric density.

    Step 4 is the consequent longterm climate cooling, by a little less than 0.1 K.

    Though the long term side-feedback effects are greater than the immediate direct effects, both immediate direct and side-feedback effects are miniscule and not practically important from a physical or geographical viewpoint. Mistaken impressions on this subject have, however, a mighty emotional and political impact.
    Last edited: Sep 6, 2008
  24. Sep 7, 2008 #23
    Count Iblis:

    In terms of the Arrhenius-IPCC doctrine of "feedback to radiative forcing by CO2", your concern is logical.

    But the doctrine is unphysical, and so its logical consequences are in general also unphysical.

    The error of the doctrine is that it assumes that the only direct effect of CO2 is to enhance the radiative processes. It ignores the physical fact that CO2 also directly and immediately affects convection, which is the main mechanism of heat transport from the surface to the upper atmosphere. The Arrhenius-IPCC doctrine demotes the convective effect to a "feedback" status; this is not even correct logic: even, for the sake of argument temporarily granting the doctrine's putting the radiative mechanism as the only direct effect, the convective effect should properly be called a compensatory effect not a "feedback" effect. A doctrine that ignores or demotes the main physical effect cannot be relied upon; that is the reason for the Chilingar Khilyuk and Sorokhtin paper.

    The physics of the Chilingar Khilyuk and Sorokhtin paper "Cooling of atmosphere due to CO2 emission" is that the main immediate direct effect of CO2 is not the "radiative forcing" that is the only admitted immediate direct effect of the Arrhenius-IPCC doctrine. Physically, the main direct effect is to change the threshold for convection. The result is that at the same temperature, the convection is affected, without need for any radiative effect, putative or actual. The threshold for convection is changed because of the molecular weight of CO2. The most sensitive threshold for convection is deep tropical convection that goes right to the tropopause, prevalent in the intertropic convergence zone. The effect is to weaken convection slightly, and, taking into account also the radiative properties of CO2, this slightly warms the system, by less than 0.01K indeed for a doubling of CO2.

    But there are delayed feedback effects which have side effects. These side effects are the main long-term influence. They continue the long-term cooling of the earth. They are chemical in nature, removing some oxygen from the atmosphere, and eventually they make the atmosphere less dense, and this leads to long-term cooling, a bit less than 0.1K. This happens because the lesser density enhances convection at the same temperature, without need for any radiative effect, putative or actual.
  25. Oct 26, 2008 #24
    So I toyed a bit with MS Excel today, I was wondering what the chance is for a single photon (IR), emitted from the Earth surface, to escape into space. This would largely be dependent on two factors.
    First: the maximum length of the path (optical depth?) it follows until it is absorbed and assuming that it is re-emitted in a random direction and assumingly for 100%
    Second: the vertical thickness of the atmosphere in that same unit, the number of vertical path lengths/optical depths between Earth surface and the 'top' of the atmosphere.

    So in the gross little model I defined that maximum path length as one unit and made one dimensional random walk simulating a photon emitted from the surface each time making a random step between -1 and +1 under various optical depths of the atmosphere. I used 20,000 cells for the random walk and would reset for the next photon each time, it hit the extreme values either <0 for re-absorption on the earth surface or when bigger than the preset number of optical depths. For each optical depth values I averaged the result of 10 runs and got these results:

    Optical depth: 1 unit; escaped percentage of photons 24.7%
    Optical depth: 2 units; escaped percentage of photons 15.8%
    Optical depth: 4 units; escaped percentage of photons 8.6%
    Optical depth: 20 units; escaped percentage of photons 2.0%
    Optical depth: 40 units; escaped percentage of photons 1.0%
    Optical depth: 100 units; escaped percentage of photons 0.1%

    Now, I have no idea what the optical depth is for a IR photon under various atmospheric conditions, maybe millimeters or hundreds of meters and the number of optical depths of the Earths atmosphere probably orders of magnitudes more. So it seems I don’t get the photons out of the Earth Surface, they drop back in almost all the time. Am I doing something wrong?

    Next, I wondered about a photon starting at a certain altitude, with X optical depths below and Y optical depths above the starting point. But when I modified the program I began to realize what the outcome would be: the chance of escape of a photon versus absorption on the earth surface would be X/(X+Y) and that was indeed supported by the simulation runs.

    So this means that the only way to get photons/ IR radiation out of the atmosphere and balance the incoming solar radiation is by emission from a higher altitude, and the most effective way to get the required energy at higher altitudes is convection.

    Note that the role of the concentration of greenhouse gasses is undefined here, if the concentration increases it affects both the number of optical depths below and above the initial emission altitude, and if that ratio doesn’t change, so would the number of emitted photons not change.

    If there is no convection in a thick atmosphere, then it seems that the energy can not be emitted out. Would that help explaining the difference between Earth and Venus where there is no convection in the lower levels of the atmosphere?
  26. Oct 26, 2008 #25


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The problem with your approach is that you have an almost identical physical setting as the diffusion of particles in a material - like neutrons in a moderator. That's well-described by the diffusion equation, but I think you miss the essence of the physics then. The essence is that photons can be absorbed, that this heats the material locally, or excites molecules, and that photons of different wavelengths can be emitted (through specific de-excitations, or through general black body radiation, which is a kind of statistical average of many low-energy emission processes which are too complicated to follow up individually).
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook