RealClimate logo


Technical Note: Sorry for any recent performance issues. We are working on it.

Aerosols, Chemistry and Climate

Filed under: — gavin @ 12 July 2008

Everyone can probably agree that the climate system is complex. Not only do the vagaries of weather patterns and ocean currents make it hard to see climate changes, but the variability in what are often termed the Earth System components complicates the picture enormously. These components – specifically aerosols (particulates in the air – dust, soot, sulphates, nitrates, pollen etc.) and atmospheric chemistry (ozone, methane) – are both affected by climate and affect climate, since aerosols and ozone can interact, absorb, reflect or scatter solar and thermal radiation. This makes for a rich research environment, but can befuddle the unwary.

I occasionally marvel at the amount of nonsense that is written about climate change in the more excitable parts of the web, and most of the time, I don’t bother to comment. But in relation to the issue of aerosols, chemistry and climate, I read yesterday (h/t Atmoz) probably the most boneheaded article that I have seen in ages (and that’s saying a lot).

The hook for this piece of foolishness were two interesting articles published this week by Ruckstuhl and colleagues and a draft EPA report on the impacts of climate on air quality. First, Ruckstuhl et al found that as aerosols have decreased in Europe over the last few decades (as a result of environmental standards legislation), the amount of solar radiation at the ground has increased while the amount reflected to space has decreased. They hypothesize that this may have helped Europe warm faster in the last few decades than it would have otherwise done. Or equivalently, since the aerosols are anthropogenic, that European temperatures had been subdued due to the cooling effects of the aerosols – and since they are now decreasing, the full effects of the greenhouse gases are starting to be felt. This is just an update to the ‘global brightening‘ story we have touched on before. The EPA report is concerned with the impacts that climate change can have on atmospheric chemistry, and in particular the summertime peaks in urban ground-level ozone which are a well-known and serious health hazard. These are affected by local temperatures, cloudiness, temperature sensitive biogenic emissions and patterns of weather variability. Again, it is a story we have discussed before.

But the NewsBusters article succeeded in getting almost every aspect of these stories wrong. How do I correct thee? Let me count the ways.

  1. Aerosols are not smog:

    First they confuse aerosols with photochemical smog. Both are pollutants, but the first is dominated by sulphate emissions from coal burning power plants, the second from ozone precursors such as NOx, volatile organic compounds, and carbon monoxide mainly emitted from vehicles. (Note that ozone is not directly emitted, but is created by chemical reactions from the precursors with the addition of a bit of photolysis – i.e. sunlight-driven chemistry). The effects on climate are very different: ozone is a greenhouse gas, so increases cause a warming, while sulphate aerosols are reflective, and so increases cause a cooling. The air quality issues in the EPA are almost all focused on ozone.

  2. Europe is not the Globe:

    The next error is to equate changes in temperatures in Europe to the globe. While it would be true that if global aerosol levels declined it would lead to increased global warming, aerosol trends in Asia are increasing strongly, even while those in the US and Europe are dropping. The net effect is possibly a slight drop, but the impact on global temperature is as yet unclear. This regionality matters in both the sulphates case and for ozone. The relevant chemistry is sensitive to water vapour and temperature in varying ways as a function of the pollution level. In remote ocean areas, surface ozone will likely decrease as the globe warms for instance (due to increasing water vapour). In polluted environments increased temperatures and larger temperature-sensitive emissions of isoprene cause enhanced ozone levels.

  3. Surface ozone is not in the stratosphere:

    Next, NewsBusters asserts that the ozone story is confusing because of the

    .. treaty called the Montreal Protocol. This was designed to reduce and eventually eliminate the production and release of a number of substances thought at the time to be depleting ozone.

    Ummm…. those substances (chiefly chlorofluorocarbons – CFCs) are still thought to be depleting the ozone layer – which is in the stratosphere, some 30km above the ground-level ozone that people shouldn’t be breathing. CFCs have no impact on ground-level ozone at all (since their reactive chlorine is only released in the stratosphere).

  4. The final inanity:

    Wouldn’t it be fascinating if such efforts [such as the Montreal Protocol] lead to cleaner air around the world which ended up warming the planet, and that additional warmth is now breaking down the very ozone we thought we could save?

    Every part of this sentence is wrong. The Montreal Protocol had no impact on cleaning the air, it stopped the growth of CFCs which are powerful greenhouse gases (in addition to their role in depleting stratospheric ozone), therefore it slowed global warming, rather than increasing it, and we aren’t trying to save ground-level ozone. Had any of this been true it would indeed have been fascinating.

What should we make of this? Unfortunately one must conclude that no mistake is too dumb for someone, somewhere to make if they think they can spin it into supporting their anti-science agenda. For them complexity is something to be abused rather than a challenge to be understood, underlining quite clearly (again) the difference between science and propaganda.


356 Responses to “Aerosols, Chemistry and Climate”

  1. 301
    Hank Roberts says:

    > knew exactly
    Sock puppet, mind reader, or humorist? Yawn. Don’t care.

  2. 302
    Poptech says:

    Hank, sorry to ruin your fantasies but I am no socket puppet. Though I will take any checks \big oil\ wants to send me.

  3. 303
    spilgard says:

    Re 300:
    Granted, as a kid I recall proclaiming — after repeatedly shooting myself in the foot — that it was all really just a joke, but even then I knew enough to let it go. Why am I reminded of Good Morning, Vietnam?:

    “Lieutenant, you don’t know if you’re shot, screwed, powder-burnt or snake-bit”.

  4. 304
    Ray Ladbury says:

    We seem to be enduring a period of what Roy Schwitters referred to as “The Revenge of the C Students,” though in some cases on here, that would amount to grade inflation. While the annoyances are obvious, there is also an upside. Occasionally, one of the more intelligent denizens of the giggle factories like Newsbusters will respond to the call to come over to “show them scientists a thing er two,” and notice that what we have here is a congenial atmosphere for learning and exchange of ideas, and where warming is not attributed to some evil UN heat ray. notice that it never happens the other way. What I saw of Newsbusters on a short perusal was more than sufficient for a lifetime.

  5. 305
    Arch Stanton says:

    Poptech (300),

    Sarcasm and seriousness are not mutually exclusive. Obviously there is also sarcasm here. The difference in this regard IMHO in the 2 sites is that RC does not bait serious folk and then revel with remarks like “this is delicious” when they respond.

    We don’t expect you to defend NB. Come here if you have questions about the science of climate change. Go there if you want to make jokes about it – but don’t be surprised when someone calls you on them.

    Why don’t we go over there? You might ask: Because it is obviously not a place for a serious discussion about climate change. The funny (peculiar) thing is that so many folks seem to think that it is…

  6. 306
    Hank Roberts says:

    Yup. Enthuse, but with focus.
    http://www.amazon.com/review/R1FPCOU91OQ0BV

  7. 307
    Jim Galasyn says:

    Why don’t we go over to NB?

    Okay, I finally visited NB. Searched on “RealClimate” to find the comments for the Gavin hit-piece. Clear Thinker appends the following link to each of his posts:

    Communist Goals for American Takeover, apparently from the 1963 Congressional Record.

    How fun is that? I wonder why he doesn’t append this to his RC posts?

  8. 308
    dhogaza says:

    How fun is that? I wonder why he doesn’t append this to his RC posts?

    The Communist Goals for taking over the United States probably involved a fair amount of science denialism, given Stalin’s history in the USSR …

  9. 309
    Ray Ladbury says:

    Jim, I toooold yooouuu! Not for the faint of heart.

  10. 310
    Guenter Hess says:

    #286 Mark,
    Don’t worry I enjoy the energizing discussion.
    I agree that this might be only a simple model. I concentrated on one line for clarity, but the principle should work for each wavelength.
    Line by Line techniques to my understanding compute one wavenumber at a time and sum over all lines throughout the spectrum to calculate the absorption coefficient. The model takes the line strength and line width from databases like HITRAN and use Voigt or Lorentz-profiles as beam shapes. You would do the same thing in infrared spectroscopy. Usually the line profiles are adjusted to agree better with laboratory measurements.
    Therefore using infrared spectroscopy as analogy seemed to me at least on equal footing with the single layer model that is used in my textbook about global warming to calculate the greenhouse effect of 33°C.
    Of course modern computers might enable more sophisticated techniques to calculate absorption and emission throughout the simulation. But also to my experience in spectroscopy, if you have broad bands with overlapping lines, the gain from sophisticated calculations or fancy line profiles is limited.

  11. 311
    Hank Roberts says:

    Now you sound like you’re wanting to continue the digression. Sure?

    Captcha says: agreement ify
    I swear there’s an oracle in there somewhere.

  12. 312
    spilgard says:

    Re 307:
    I suspect it’s something required by the NB site to establish street credibility, similar to fantasy RPG sites where it’s obligatory that your sig line include a banner featuring a dragon or a buxom elven-lass. If RealClimate ever adds a sig-line feature, I’d like to display Maxwell’s equations but I’m torn between the mathematically-tractable differential form and the integral form which more clearly illustrates physical meaning.

  13. 313
    Clear Thinker says:

    Mr. Galasyn,

    Every site that I post to has the very same sig line, so what? I do not include it here because I understood that this thread was not about politics, so out of respect I did not include. Something wrong with that?

    And that leads me to another question…. what’s wrong with what I link too? It’s important, so why not?

    I totally, and absolutely do not understand your consternation with my link. In fact, most people would be concerned about what’s in the link and the lessons to be learned by it.

  14. 314
    Ray says:

    Great words from Peter Huber.

    “So does the climate computer have a real audience, or is it really just another bag lady muttering away to herself in a lonely corner of the intellectual park? That the computer is heard in Hollywood, Stockholm, Brussels and even some parts of Washington is quite beside the point–they have far less global power and influence than they vainly imagine. Vinod Dar is right: “Contingency planning should entail strategic responses to a warming globe, a cooling globe and a globe whose climate reverberates with laughter at human hubris.”

  15. 315
    David Donovan says:

    Poptech..

    I read the article myself…The context seemed very clear to me. Also,r,call me humorless but I am afraid I missed the “intended” sarcastic bent of Mr. Sheppards writing. As a matter of fact I find it downright knee slapping hilarious that you seem to be defending the piece now as a work of sarcasm !

  16. 316
    Hank Roberts says:

    Hmmmm. Sometimes it’s hard to know who’s who without an IP lookup.

    http://blog.matthewmiller.net/2007/09/revisiting-firefox-myths-part-2-tangent.html?showComment=1190271180000#c4130730304092334427

    Captcha, oracularly: Fellowes Enrolled

  17. 317
    Jim Galasyn says:

    Clear Thinker: Purity of Essence, baby.

  18. 318
    spilgard says:

    Every site that I post to has the very same sig line, so what? I do not include it here…

    Weird. It’s included in every site that you post to, except for the the site that’s not included in every site that you post to?

    In fact, most people would be concerned about what’s in the link and the lessons to be learned by it.

    I doubt that we’re thinking of the same thing, but I’m inclined to agree with you.

  19. 319
    David Klar says:

    Inhabitants of the denialsphere like clearthinker and poptech have no interest in learning about climate science. They are trying to attack legitimate climate scientists like Dr. Schmidt using rhetorical tricks. They have not presented a single scientific idea that refutes or even brings AGW into question. Dr. Schmidt has been dishonestly smeared by both them and Noel Sheppard on NB, using smear tactics, innuendo, and unsupported claims. Sheppard is a climate hack, who states that he has written many climate science articles but doesn’t come to Realclimate with all his knowledge. It’s pretty obvious that any real contrarian science ideas could simply be presented here for critique. Sheppard’s real purpose is to slander legitimate climate scientists, calling them “alarmists.” He has implied that Dr Schmidt may have been dishonest with some climate science data, a really disgusting and despicable tactic. He Never brings his claims to realclimate.

  20. 320
    Mark says:

    Guenter, # 310. Two things.

    1) Most of the energy is NOT at 15um. So why do you keep on about it?

    2) kinetic energy you posted was 2.5 times lower than the 15um energy but you’ve forgotten that the velocity is an average, so some are going faster and that there are TWO CO2 molecules involved, so the energy is doubled. Getting pretty close to allowing motion to excite the absorbtion band, isn’t it. Now add on that it isn’t a 1um wide band….

    If you want to keep with the spectroscopy, do the integral over the entire emission spectrum.

  21. 321
    DavidONE says:

    Phew. Just finished reading the comments, which are at least as enlightening as the article. Thanks to all who contributed science to the discussion.

    Clear Thinker, your forced politeness appears disingenuous. If someone signs with their first name, they do so for a reason. Use it.

    If you really have some notion that formal salutation is required, use the correct one, which for Gavin is ‘Doctor’, not ‘Mister’ – as has already been pointed out to you. [edit]

    [Response: Just to make it clear, I perfectly happy with 'Gavin' and use of 'Dr' (and certainly 'Mr') is discouraged. - gavin]

  22. 322
    a.c. says:

    I thought the part about discrediting the american founding fathers was particularly disturbing, and it seems that those damned Reds have their tentacles wrapped all around the internet, too….

    ——
    http://www.whitehouse.gov/history/presidents/gw1.html

    “As the first of every thing, in our situation will serve to establish a Precedent,” he wrote James Madison, “it is devoutly wished on my part, that these precedents may be fixed on true principles.”

    ….

    He pursued two intertwined interests: military arts and western expansion.

  23. 323
    Rod B says:

    Mark (320), I initially raised my eyebrows at your description, but maybe…. See if I pass the test (assuming I understand the question :-) ). Two molecules (only one need be CO2…???) are tooling around at just a bit over the gases’ kinetic energy rms average — 1.25 times, in the example. These two molecules collide head on, both come to a complete halt with now zero kinetic energy, and must transfer the 2.5 times the average energy — exactly what the CO2′s vibration accepts. The N2 (say) transfers its 1.25E to CO2 vibration while simultaneously CO2 is transferring its own kinetic energy (again at 1.25E) to its own vibration. Is this how vibration of a molecule, whose vibrationally excited energy level is much greater than the average kinetic energy of any molecule in the neighborhood, gets excited via collision (versus absorption)? Does this “tendency” differ with different collider molecules (N2, O2, CO2, etc.)? I assume the 1.25 times is the minimum; anything above that would also work though the colliding molecules would retain some kinetic energy. Correct?

    Is the probability or incidence of such a collision then reduced by somehow adjusting the mean free path based on the population of the molecules at 1.25 times or greater the average rms kinetic energy of the gas ala Boltzmann’s distribution (or something similar)?

  24. 324
    Guenter Hess says:

    Mark, #320
    I keep on going with spectroscopy, because I want to understand the question I posed. But you can choose not to answer.
    The question is:
    If I integrate the infrared emission spectrum from CO2 in air at atmospheric pressure and divide it by the integrated absorption spectrum I get a number well below 1 (the quantum yield at atmospheric pressure is well below one across the whole range), whereas in radiative transfer models the number is usually set to 1 via the assumption of Kirchhoff’s law.
    This means that most likely as you pointed out the emission spectrum does not represent the emitted energy in the atmosphere. However, I did not understand so far and want to know why. Neither did the answers I got so far convince me.
    Moreover, if the emission spectrum is not representative, why can the absorption spectra be used to “tune” the radiative transfer models.

  25. 325
    Ray Ladbury says:

    Rod B., Careful with language. The phrase: “…whose vibrationally excited energy level is much greater than the average kinetic energy of any molecule in the neighborhood…” doesn’t make sense. Individual molecules do not have an average energy. However, essentially, you are correct–roughly 4% of molecules have enough energy to excite the vibrational band.

  26. 326
    Hank Roberts says:

    Guenter, are you having problems because you’re using “atmospheric pressure” numbers? Have you gone through the published work, and Weart’s history of how this point has been reached?

    I know very little about this. But I can’t tell what your starting point is or what you might be missing.

    http://www.aip.org/history/climate/Radmath.htm
    http://scholar.google.com/scholar?q=emission+infrared+%22top+of+atmosphere%22

  27. 327
    Phil. Felton says:

    Re #324

    Guenter, you can consider a CO2 molecule as being surrounded by a ‘bath’ of N2 and O2 molecules which are continually colliding with each other and exchanging energy, the kinetic energy distribution of these molecules will be a Boltzmann distribution. The Boltzmann distribution at room temperature will have about 5% that have an energy greater than the first vibrational energy level of CO2.

    Regarding the quantum yield you would expect it to be considerably less than 1 because most of the molecules activated by the IR will be collisionally deactivated by the surrounding molecules (average collision time less than 1 nanosec whereas emission time of the order of millisec).

  28. 328
    Timothy Chase says:

    Guenter Hess wrote in 324:

    If I integrate the infrared emission spectrum from CO2 in air at atmospheric pressure and divide it by the integrated absorption spectrum I get a number well below 1 (the quantum yield at atmospheric pressure is well below one across the whole range), whereas in radiative transfer models the number is usually set to 1 via the assumption of Kirchhoff’s law.

    With the integration you are speaking of, it sounds like you are interpretting Kirchoff’s law as stating that the total emission is equal to the total absorption. But it isn’t stating this. Kirchoff’s law is expressed in terms of emissivity and absorptivity, which under both thermodynamic equilibrium and local thermodynamic equilibrium conditions are equal. And the emissivity of a body is determined by comparison of its material with a black body at the same temperature.

    Please see:

    The emissivity of a material (usually written η) is the ratio of energy radiated by a particular material to energy radiated by a black body at the same temperature. It is a measure of a material’s ability to radiate absorbed energy. A true black body would have an η = 1 while any real object would have η < 1. Emissivity is a dimensionless quantity (does not have units).

    http://en.wikipedia.org/wiki/Emissivity

    Now here is a technical derivation of Kirchoff’s law demonstrating that spectral emissivity (emissivity at a given wavelength) and spectral absorptivity are equal under thermodynamic equilibrium conditions — that is then generalized to local thermodynamic equilibrium conditions:

    Radiative Transfer
    Interaction of Radiation with Matter
    http://www.cv.nrao.edu/course/astr534/Radxfer.html

    … and as noted in the above text:

    …most of the Earth’s atmosphere is in LTE . It has a well-defined gas temperature T~300 K) measureable with an ordinary thermometer, but it is in a nonequilibrium radiation field (T~5800 K sunlight during the day, the cold dark sky at night, plus anisotropic emission from the ground) distinctly different from a black body at the gas temperature.

    ibid.
    http://www.cv.nrao.edu/course/astr534/Radxfer.html

    *

    But where exactly does Kirchoff’s law break down?

    Basically where the atmospheric pressure drops below 20 mb. We are at approximately 1000 mb at sea level. At 20 mb, collisions are too infrequent, and for example, different quantized states of molecular excitation (bending, rotating, vibrating and rovibrational states) will begin to assume different brightness temperatures, and in the extreme, stimulated emission may occur. And at that point you need to drag out your Einstein coefficients. But above 20 mb, molecules undergo roughly a million or more collisions over the half-life of any of the relevant states of molecular excitation.

    Oh, and incidentally, the general circulation models are taking into account non-local thermodynamic equilibria where Kirchoff’s law no longer applies. There is still a fair amount of work to be done with aerosols and ocean circulation, but radiation? Not so much.

  29. 329
    Guenter Hess says:

    Re:#327 Phil. Felton
    Yes, you are right, that is why I thought that most of the energy that is absorbed by a slab of atmosphere (that is under LTE) is transferred to the next layer via collisions. This should be valid up to about 75 km.
    But
    #328 Timothy Chase and Mark #320: thanks for the discussion and the interesting references
    have stated that under LTE conditions the spectral emissivity and spectral absorptivity are equal.
    This means that most of the energy is transferred to the next layer by radiation.
    I seem to be alone with my headaches and have of course to reconsider.

  30. 330
    Timothy Chase says:

    PS to 328

    The eta η should have been an epsilon ε… Should have previewed.

  31. 331
    Mark says:

    Rod B, #323, why do they have to collide with EXACTLY the right velocity? If they hit, they can transfer more energy, if one is in an excited state then it can be an inelastic collision, they can connect with enough energy to move the excitation UP another level, they can collide with no change.

    Also note: a 5% chance is a virtual certainty when attempted millions of times.

    Tim in #328 has given a lot better response than I have, even including links. I notice neither you nor Guenter have provided links to what make you think these are important and unmodelled (and also, incorrectly parameterised, else not modelling doesn’t change the result if your parameterisations are good).

  32. 332
    Adam says:

    “However, global mean surface temperature has not risen since 1998 and may have fallen since late 2001.”

    This statement at the very begining of the paper would tend to cast doubt on the credibility.

  33. 333
    Guenter Hess says:

    Mark (#331),
    Mark, I actually agree that Timothy has given an excellent comprehension with links about how radiative transfer is modeled and parameterized. I do not question that this is done correctly.
    My source for this is the book “Radiative transfer in the atmosphere and the ocean” by Gary E. Thomas and Knut Stammnes. They state:”Radiative emittance is governed by Kirchhoff’s law, according to which the directional emittance equals the directional absorbance in the case of surfaces; and in the case of extended media the emission coefficient is proportional to the absorption coefficient.” I might have confused emissivity and emittance in my texts, but I do not question this either.
    However, Phil. Felton (#327) has described the situation I was referring to in his contribution #327. And that is also my headache. Following the assumptions and derivations of Thomas and Stamnes the radiative transfer models imply that the main part of the absorbed energy is transferred on by radiation respectively via emission.
    However, my spectroscopy book by Wolfgang Demtroeder clearly states that the quantum yield for infrared emission at atmospheric pressure (the condition described by Phil. Felton) is very low, meaning that the main part of the absorbed energy is transferred to translational modes of nitrogen or oxygen molecules. Of course there is a finite probability to transfer the energy back to CO2. But all in all this argument implies that the energy transfer is also dominated by collisions and not by radiation. I just can’t bring the two things together. I feel that I miss an argument beyond the simple statement that the spectroscopy case cannot be applied. That was my small plight, find the missing argument that radiation is the main energy transfer mode compared to collisions.

  34. 334
    Ray Ladbury says:

    Guenter,
    You have to consider several things here. First, the excited vibrational state of the CO2 molecule involved is very long-lived. Second, the pressure assures that the cross section for collisional relaxation is not negligible. Third, the gas is not really in equilibrium with the photon field of the IR radiation around it. What you have here is a competing set of reactions–CO2 + photon–>CO2* and CO2*+N2–>CO2 +N2(fast). Because there are many more photons than would be the case for the gas at equilibrium, you have more excited CO2 molecules (CO2*). Because you have a high density of N2, you favor relaxation via collision. Or have I misunderstood your predicament.

  35. 335
    Hank Roberts says:

    > at atmospheric pressure

    Seriously, I think this may well be your problem.

    Only sea level is at “atmospheric pressure” (one standard atmosphere). Everything higher is at lower pressure.

    Have you read Spencer Weart’s history yet?

  36. 336
    Phil. Felton says:

    Re #333
    Guenter your problem is that you are confusing ‘emittance’ and ‘absorbance’ (which are both coefficients and have a value between 0 and 1) and the total emission and absorption.
    Consider a perfect black body ( emittance = absorbance = 1) in a vacuum with constant radiation incident on it, it will achieve a temperature at which emission = absorption. Replace the vacuum with a gas at high pressure and in addition to emission the black body will lose heat via conduction and convection so the temperature will be lower, the emittance & absorbance will be unchanged however.
    HTH

  37. 337
    Guenter Hess says:

    #334 Ray,
    Yes, you misunderstand me. Your three statements are correct.
    But my question is: what is the dominant energy transfer mechanism at pressures where LTE holds, usually pressures up to a height of 60-75km. First: Absorptionby CO2/Emission by CO2 or Second: Absorption by CO2/Deexcitation via collision and transfer of energy to a translational mode of a groundstate molecule
    Hank Robert (#335),
    Yes I read Spencer Weart. Yes it is my problem, because I like to understand it. Sorry for my bad communication. With atmospheric pressure, I meant a pressure range where LTE holds, usually pressures up to height of 60-75km.

    Phil. Felton (#336):
    But at steady state at lower temperature, emittance & absorbance would be lower or not?

  38. 338
    Mark says:

    Guenter, you can also answer my query to Rob B: How does your queries resolve to errors in GCMs?

    NOTE: Absorbtion==Emmission cannot be true when the system is out of equilibrium so your continued demand that this be true is incongruent.

  39. 339

    Guenter Hess,

    I think I may see the source of your confusion. You’re saying — correct me if I’m wrong — that the incoming radiation, say 10 watts per square meter in a given narrow wavelength range, excites the CO2 in a layer of atmosphere, but since only 4% of the CO2 is excited, the outgoing radiation should be only 0.4 watts per square meter, and Kirchhoff’s Law should fail. Do I have it right?

    Yes, the incoming energy gets distributed to all the non-greenhouse gases by collision. But in order to get out again it all has to go through the greenhouse gases, since nitrogen and oxygen don’t radiate significantly. The collisions transfer the energy back to the CO2, which radiates it all out again. Collisions explain mainly why all the gases in a layer, not just the greenhouse gases, are heated by incoming radiation. The energy bounces around a lot before it gets out again. But the income and outgo balance.

  40. 340
    Ray Ladbury says:

    Guenter, Absorption by CO2 followed by collisional relaxation is the dominant transfer mechanism in the troposphere. Again, however, remember that the gas is not in equilibrium with the radiation field, which comes from warmer regions below. The difference from eqilibrium is not sufficiently large that you have to use nonequilibrium thermodynamics.

  41. 341
    Timothy Chase says:

    Guenter Hess wrote in 337:

    But my question is: what is the dominant energy transfer mechanism at pressures where LTE holds, usually pressures up to a height of 60-75km. First: Absorptionby CO2/Emission by CO2 or Second: Absorption by CO2/Deexcitation via collision and transfer of energy to a translational mode of a groundstate molecule
    Hank Robert (#335).

    Well, it depends upon scale, really.

    If you are speaking at the molecular level, then the dominant energy transfer mechanism would certainly be collisions rather than photonic emission. Afterall, Kirchoff’s law depends on this — as this is what results in the radiation which is being emitted by matter being in local thermodynamic equilibrium with the matter that is emitting it. However, if you are speaking at the level where energy finally escapes the atmosphere, then photonic emission isn’t simply the dominant energy transfer mechanism — it is effectively the only energy transfer mechanism. And it is worth pointing out that this mechanism operates primarily under LTE conditions.

    The atmosphere at a given altitude will be optically thick to some wavelengths of radiation, meaning that most photons of that wavelength which are emitted at that height will be absorbed before they have the chance to escape to space, but it will be optically thin to other wavelengths, meaning that most at that wavelength do escape to space without being absorbed. And in fact there is a height at which most thermal radiation that is emitted escapes without being absorbed. It is the effective radiating height, at 5-6 km, well below the 75 km where non-LTE conditions gradually become appreciable, and in fact well below the 50 km altitude where non-LTE conditions become significant at any wavelength.

    Please see:

    User’s Manual for SAMM2, SHARC-4 and MODTRAN4 Merged
    AFRL-VS-HA-TR-2004-1145
    Environmental Research Papers, No 1145
    Spectral Sciences / Air Force Research Laboratory
    http://www.dodsbir.net/sitis/view_pdf.asp?id=DothH04.pdf

    For any wavelength that the atmosphere is optically thick to at sea level, there will be a height at which the atmosphere goes from being optically thick to optically thin. This is the principle that underlies our ability to perform infrared imaging of the atmosphere and its constituents at various altitudes. For example, here is carbon dioxide (a wavelength of 15 μm, I believe) at an altitude of 8 km:

    NASA AIRS Mid-Tropospheric (8km) Carbon Dioxide
    http://www-airs.jpl.nasa.gov/Products/CarbonDioxide/

    You will notice the plumes rising off the heavily populated east and west coast of the United States. We are able to image things at that altitude for that wavelength because most of the photons of that wavelength are absorbed below that altitude, but at that altitude or higher, once they are emitted, they will generally escape to space without further absorption.

  42. 342
    Rod B says:

    Timothy (341) a clarification, please. Are you sure that at ~5-6km an emitted photon from a CO2 molecule most likely will escape rather than be absorbed by another CO2, say? At this height (CO2 density) is the escape just barely more than 50%? and get progressively more likely as altitude increases? At this level is relaxation still predominately by collision?

    Is this the altitude where the atmosphere changes (by convention, I assume) from optically thick to thin? Does this altitude differ for different gases or concentration (density)?

    Is there any explicit reason why the 8 km CO2 concentration looks a tad odd, like why does it pile up over the western Atlantic e.g?

    Thanks for the info.

  43. 343
    Timothy Chase says:

    Rod B wrote in 342:

    Timothy (341) a clarification, please. Are you sure that at ~5-6km an emitted photon from a CO2 molecule most likely will escape rather than be absorbed by another CO2, say? At this height (CO2 density) is the escape just barely more than 50%? and get progressively more likely as altitude increases? At this level is relaxation still predominately by collision?

    Is this the altitude where the atmosphere changes (by convention, I assume) from optically thick to thin? Does this altitude differ for different gases or concentration (density)?

    The effective radiating height isn’t defined with respect to carbon dioxide only, but with respect to the climate system as a whole. As such this includes other greenhouse gases (methane, water vapor, ozone, etc.), clouds and even the surface of the planet itself. It is a “mean.” And I tried to suggest as much using the example of CO2 at 15 μm at 8 km:

    For any wavelength that the atmosphere is optically thick to at sea level, there will be a height at which the atmosphere goes from being optically thick to optically thin. This is the principle that underlies our ability to perform infrared imaging of the atmosphere and its constituents at various altitudes. For example, here is carbon dioxide (a wavelength of 15 μm, I believe) at an altitude of 8 km:

    NASA AIRS Mid-Tropospheric (8km) Carbon Dioxide
    http://www-airs.jpl.nasa.gov/Products/CarbonDioxide/

    For photons of that wavelength, the altitude at which they tend to escape is 8 km, not 5.5 km. As I have said, 5.5 km is a mean value — for the climate system as a whole. And typically it wouldn’t make sense to break it out according to a specific gas as there will be different bands that overlap, and what gets emitted by one band will often be absorbed by another.

    Rod B wrote in 342:

    Is there any explicit reason why the 8 km CO2 concentration looks a tad odd, like why does it pile up over the western Atlantic e.g?

    East coast traffic.

    Seriously. Carbon dioxide emissions are rising up from the heavily-populated east coast and being carried over the Atlantic. Consequently the concentration of carbon dioxide is thicker there, making the atmosphere optically thicker to 15 μm at that height — meaning that light at that wavelength tends to escape at a somewhat higher altitude.

    You can learn more about infrared sounding here:

    AIRS – Multimedia: Animations
    http://airs.jpl.nasa.gov/Multimedia/Animations/

    2. Meteorological sensor systems on GOES I-M
    http://goes.gsfc.nasa.gov/ams/meteor.html

    Visualization of the global distribution of greenhouse gases using satellite measurements, by Michael Buchwitz. The Encyclopedia of Earth. Posted July 31, 2007
    http://www.eoearth.org/article/Visualization_of_the_global_distribution_of_greenhouse_gases_using_satellite_measurements

    … and its commercial use in mining and topography, for example, here:

    Figure 20. The 2-μm CO2 absorption strength (A) can be converted to topographic elevation (B). The derived elevations matches the USGS Digital Elevation Model (DEM) (C). The CO2 absorption strength image (A) is brighter for increasing strength. Because the atmospheric path length is smaller with increasing elevation, the absorption strength decreases, becoming darker in the image. The DEMs (B, C) are brighter for increasing elevation, thus are inversely correlated with the CO2 strength in (A).
    http://speclab.cr.usgs.gov/PAPERS/tetracorder/FIGURES/fig20.dem.abc.gif

    Figure 21. The 2-μm CO2 absorption strength versus USGS DEM elevation shows a linear trend with an excellent least squares correlation coefficient.
    http://speclab.cr.usgs.gov/PAPERS/tetracorder/FIGURES/fig21.co2_graph.gif

    Imaging Spectroscopy:
    Earth and Planetary Remote Sensing with the USGS Tetracorder and Expert Systems
    Roger N. Clark, et al
    Journal of Geophysical Research, 2003.
    http://speclab.cr.usgs.gov/PAPERS/tetracorder/

  44. 344
    Rod B says:

    Thanks, Timothy

  45. 345
    Guenter Hess says:

    Barton Paul Levenson(#339),
    I had the impression that the emission spectrum for an atmospheric layer within the Line by Line models is calculated just by setting absorbance equals emittance.
    I realized with the help of Timothy’s references and reading again through the Book by Thomas and Stamnes about Radiative Transfer that in case of atmospheric layers the emission spectrum is obtained by weighing the absorption spectrum with the Planck function. I had overlooked this important difference.

  46. 346
    Guenter Hess says:

    Timothy Chase (#341),
    Thanks for your comments and references.
    Timothy just to get it right for me:
    1. On a microscopic level:
    Do I understand you correctly that within the atmosphere up to the height up to where LTE holds, collisions are the dominant processes for energy transfer compared to absorption/reemission. The lower the pressure the higher is the percentage of absorption/emission processes.
    However, at the escape height which is below the height up to where LTE holds, reemitted photons start to have a significant probability to escape to space.
    2. On a macroscopic level:
    Averaged across the Sphere of the escape height Kirchhoff’s law for extended media holds

  47. 347
    Timothy Chase says:

    Guenter Hess wrote in 345:

    I realized with the help of Timothy’s references and reading again through the Book by Thomas and Stamnes about Radiative Transfer that in case of atmospheric layers the emission spectrum is obtained by weighing the absorption spectrum with the Planck function.

    Under LTE, the emission spectrum would simply be the spectral absorptivity times the Planck function – where assuming a spectral absorptivity of 1 the Planck function would give the black body spectrum at temperature T — if I am not mistaken. Incidentally, you helped me out when you brought up volumes earlier.

  48. 348
    David B. Benson says:

    On topis with regard to chemistry anyway. This article

    http://www.ias.ac.in/currsci/jun252006/1607.pdf

    suggests that 420 ppm CO2 will have human health consequences.

  49. 349
    Timothy Chase says:

    David (#348),

    I had run across that article before, but for some reason I seemed to have put it out of my mind. A bit unusual for me. The good news is that we probably won’t have to worry about hydrogen sulfide — at least not for our own species, anyway.

    Acidosis, headaches, difficulty breathing, developmental abnormalities, … this is beginning to sound like a pollutant by any medical definition of the term. Regulation of carbon dioxide as a pollutant due to its greenhouse effects seems a bit of a stretch under the EPA’s current mandate. But regulation of carbon dioxide as a pollutant due to direct health effects of high levels of exposure where those levels will otherwise be attained later in this century seems well within the EPA’s jurisdiction.

  50. 350
    trrll says:

    I have a rather basic question regarding energetics: To what extent do the relatively short-term year-to-year temperature variations reflect variation in the input-output energy balance of the earth, and to what extent do they reflect variation in how that energy is partitioned here on earth?


Switch to our mobile site