RealClimate logo

Unforced variations 3

Filed under: — group @ 19 March 2010

Another open thread. OT comments from the Amazon drying thread have been moved over. As usual, substantive comments only please and no abuse.

844 Responses to “Unforced variations 3”

  1. 751

    Barton (#747), I await the reference list with interest.

  2. 752

    Since there is no OT on this thread, I suppose it would be the best place to plug my new web article on Guy Callendar. He’s underappreciated–not as well-remembered as Plass (to whom he was a bit of an epistolary mentor), and yet it was Callendar who transformed the state of the art of CO2 theory from nineteenth-century norms to those of the twentieth century.


    Hope all are having a happy Easter!

  3. 753
    Septic Matthew says:

    Happy Easter everyone. As I write I am playing the Good Friday music from Parsifal in the background.

  4. 754
    Hank Roberts says:

    Free webcams may be as useful as satellites in tracking changes in phenology. Clearly the Amazon is lacking, but for North America, it looks promising.

    (hat tip to MotherJones)

  5. 755
    Completely Fed Up says:

    “Since Beer’s law applies in the case of monochromatic radiation it seems that you are not invoking anything like it?”

    Beers Law applies when you have a medium that doesn’t change the radiation balance.

    A thermally isotropic medium will obey Beers Law because there’s no differentiation between the obscuring medium at the emitter end and the obscuring medium at the receiver.

    A medium with a thermal lapse rate will not obey that requirement. At the VERY LEAST, the occupation of the absorption bands will be changed as the medium is traversed.

  6. 756
    John Peter says:


    Cleaning up a loose end. Some here may not realize that a GE researcher, S.D.Silverstein, fixed Wood’s “process error”. It seems that both the glass hothouse effect and the atmospheric greenhouse effect are related. “…The atmospheric blanket provides a radiation shieid, which is in
    principle the same as the glass window. The details are different as the atmosphere is diffuse rather than concentrated, and convective mixing occurs within the atmosphere rather than at a fixed outer boundary…”

    Thanks to Hank Roberts

  7. 757

    A belated Happy Easter to all! Christos anestay!

  8. 758
    Patrick 027 says:

    Re 748 John E. Pearson


    Basically right.

    At a given wavelength, adding some atmospheric absorption where previously there was none will initially result in an approximately linearly proportional radiative forcing, but the radiative forcing (at the tropopause level) can only increase to the point where the net LW flux is zero; getting near that point, each doubling of optical thickness halves the distances over which photons can reach and thus tends to halve (in the approximation of blackbody intensity being linearly proportional to temperature, which will always be true for sufficiently small temperature variations, but more easily the case at longer wavelengths)) the remaining room for farther forcing. (Of course, Beer’s Law, or actually, a more general version called Schwarzchild’s equation, does form part of the physical basis for all this, but it is the shape of the CO2 spectrum that leads to the logarithmic relationship.)


    [ Schwarzchild’s equation; for a given frequency and direction:

    Where I is intensity (flux per unit area per unit solid angle), and x is distance along the direction, so that I varies over x as dI/dx:

    dI = emission – absorption

    Assuming LTE (actually, ‘quasi-LTE’ (see future comment on solar cells) – There is some thermodynamic disequilibrium (an abundance of unoxidized CH4, for example), but in a form that is not generally reacting fast enough to significantly affect LTE for the given chemical and physical conditions as it relates to photon fluxes):

    dI = Ibb*ecsv*dx – I*acsv*dx

    Ibb is the blackbody intensity for that location
    ecsv is the emission cross section per unit volume
    absv is the absorption cross section per unit volume
    and at LTE, ecsv = absv (at least when the optical properties are isotropic or at least symmetrical; deviation can occur if there were a multitude of particles that were mirrored on one side (requiring an additional term in the equation for scattering) and blackbodies on the other and they were preferentially aligned; in that case, acsv would be equal to ecsv for the opposite direction, but that is (so far as absorption and emission are concerned) sufficient to satisfy the second law of thermodynamics, so long as the effect of scattering is also included)

    A still more general form (excluding reflection at surfaces):
    For a given frequency and polarization:

    dI# = Ibb#*ecsv*dx – I#*acsv*dx – I#*scsv*dx + IO#*scsvO*dx

    (PS not a standard notation)

    where the # symbol indicates that the intensity has been appropriately scaled to the index of refraction (specifically, at least for isotropic refraction (same index in all directions at a given location), intensity is proportional to the square of the real component of the index of refraction at the location it is evaluated.)

    and scsv is the scattering cross section per unit volume
    (All of these cross sections are proportional to the amount of material or particles which provide them).

    and the last term, which depends on the intensity of radiation in other directions (IO#), represents radiation scattered into the path. (there is a relationship between scsv and scsvO, which among other things satisfies the second law of thermodynamics (when I# is initially isotropic, it won’t spontaneously be reduced in some directions and increased in others by scattering alone (such an increase in anisotropy would be analogous to net heat flowing from cold to hot)).

    The relationship between emission and absorption that satisfies the second law of thermodynamics, is for LTE, ecsv in in direction is equal to acsv in the opposite direction. But ror randomly oriented (gas molecules, usually) or spherically symmetric (small cloud droplets) particles, all the cross sections should be the same in all directions. Even when that is not the case (sometimes cirrus ice crystals will float with some tendency toward a prefered orientation, giving rise to some optical phenomena), if the particles are symmetrical for any 180 degree rotation, then (unless I forgot something here), ecsv and acsv, and scsv, will each be the same in opposite directions and thus ecsv = acsv in the same direction. More generally, it is possible to imagine particles which could cause ecsv to be different from acsv in one direction, but if I’m not mistaken, the same extinction cross section per unit volume (scsv + acsv) must be present both in forward and reverse directions in the same path at the same location, and so an assymetric particle may scatter radiation from one direction Q and scatter radiation in other directions back toward that direction Q (with a relationship that satisfies the second law of thermodynamics – it doesn’t decrease the entropy of I# – I suspect (I haven’t studied scattering in such detail) that the relationship is that the fraction of radiant intensity from Q that is scattered toward direction S must be the same as the fraction of radiant intensity from S that is scattered toward Q for the same polarization in forward and reverse direction in Q and the same polarization in forward and reverse direction in S, but allowing that the polarization may be different between S and Q (?) – the same relationship applies to partially reflecting interfaces), while absorbing radiation from direction -Q and emitting radiation radiation back toward direction -Q with (at LTE) ascv for the radiation from -Q = escv for radiation toward -Q (at a given frequency and polarization). These relationships are sufficient to satisfy the second law of thermodynamics for emission and absorption, scattering, and reflection.


    That that all presumes scattering preserves frequency; types of scattering called Raman scattering and Compton scattering also occur wherein there is partial absorption or emission of a photon (it may gain or lose energy in a collision). These, as well as fluorescence and phosphorescence, are processes wherein emission of photon energy (and absorption ?) does not occur at (quasi-)LTE, but these still obeys the second law because the if the photon gas is hotter or colder than the non-photon material, there will tend to be a net transfer of heat from one to the other, or in the absence of such exchange, the reverse flow of heat will not occur spontaneously. The same should be true with fluorescence, etc. But all this would require adding terms to the formula for dI, or else allowing escv to not be equal to ascv in the opposite direction as described in the last paragraph).

    A greenhouse effect could in principle be based not just on absorption/emission or just scattering (or selective reflection, for that matter, though a naturally occuring sheet or sharp interface in a gaseous atmosphere would be a bit odd…) or a combination, but also could be based on Raman/Compton scattering or fluorescence or phosphorescence (with the interesting complexity that phosphorescence would cause radiant energy to, in effect, be transported by convection) – but those are rather exotic atmospheres.


    (PS all this assumes optical properties and temperature, etc, don’t change, which is a good approximation if they change by only small amounts in the time between a photon’s emission and it’s absorption within the system considered. Relativistic effects are also set aside. Though I’d assume any corrections related to these things don’t break the second law of thermodynamics (doppler shifting can do work on a photon gas; is the same true of gravity? – is the spectrum of a material shifted by the same amount as gravitational redshifting of a photon?).)

    (Also assuming linear optics – I haven’t studied nonlinear optics, but I know that generally, any waves (including sound, seismic, inertio-gravity and Kelvin, Rossby, … and presumably Alfven??) in a medium can alter the medium so as to alter the wave propagation, so that waves may interact nonlinearly with other waves or themselves. This effect is weak if amplitudes are small.)

    The phase matters when interference patterns are set up, such as in the interaction of reflected and incident waves, which is important in determining the overall reflection and transmission of multiple partially-reflecting interfaces (and maybe other wavelength-scale structures, as in Mie scattering). However, the phases of individual photons relative to some set time frame are not important; what is important is the spacing of objects relative to wavelength; it will have the same effect on photons of all phases for a given frequency and polarization and direction.

    Because refraction changes the wavelength as a photon propagates, it is more readily applicable to categorize photons by frequency, or by the equivalent wavelength in a vacuum.

    The scsv (scattering cross section density) by randomly oriented particles (such as gas molecules, usually) or spherically-symmetric particles should be independent of the polarization of incident radiation; however, the direction of the scattered radiation does generally depend on the polarization. (So the reflection of sunlight depends on the angle of the sun both through reflection of direct sunlight and through the resulting polarization of diffuse sunlight at various angles.) In the absence of other effects, multiple scatterings tend to drive the distribution of I# towards an isotropic state, and so (I think) a sufficient frequency of scattering events should tend to reduce any polarization caused by an initial scattering (?).

    (For all the radiant intensity * volume emitted from a volume V1 at some polarization (and frequency and direction) that reaches a volume V2, if there is scattering and/or partial reflection along the way, that radiation could reach volume V2 from various different directions and with various polarizations.)

    Diffraction around a particle adds to it’s scattering effect.


    See also my comment:
    (and the link within it, and also possibly a followup comment there about interfaces)




    At, I was describing the instantaneous forcing, before stratospheric adjustment. The forcing most directly relevant to tropospheric and surface climate is the tropopause level forcing after stratospheric equilibration (this is an intermediate equilibration; the stratosphere must adjust again to tropospheric and surface respones).

    For solar forcing, the tropopause level forcing should be a bit larger after stratospheric equilibration, while for greenhouse gas forcing (at least well-mixed greenhouse gas forcing), the opposite should be true (hence, tropopause level forcing with stratospheric equilibration is more readily compared for forcing by different mechanisms, though there is still room for variations in climate sensitivity from different forcing mechanisms via spatial-temporal patterns in forcing and feedbacks (different forcing efficacies)).

    The change is due to the change in downward LW flux at the tropopause caused by a change in stratospheric temperature. The total tropopause level forcing can be divided into two contributions; one from the instantaneous forcing, and the other from stratospheric equilibration.


    For CO2 forcing, the instantaneous forcing is still concentrated into two intervals of the spectrum where the CO2 opacity is significant but not close to saturated; the tropopause forcing from stratospheric adjustment is also not found where tropopause level LW forcing is already saturated, it will be found in the same intervals where the instantaneous forcing comes from, but also in other parts of the spectrum where there is downward LW radiation from stratospheric water vapor and ozone.

    The effect of stratospheric adjustment on outgoing radiation to space is found in those intervals as well as where tropopause LW radiative forcing is saturated.

    My understanding is that, because the stratosphere has only intermediate optical thickness or less for most wavelengths where there is a significant optical thickness, to a first approximate the change in LW emission out of the stratosphere, which is equal to the instantaneous stratospheric forcing (the difference between the instantaneous forcings at the top of the atmosphere and at the tropopause level), should be about the same at the top and bottom of the stratosphere; it may be a little less for the downward LW radiation at the tropopause because the temperature change (for a well-mixed GHG, and also, I think, for solar forcing) tends to occur in the upper stratosphere, and because the temperature generally (some regional/latitudinal/vertical exceptions) increases upward in the stratosphere, making radiant flux emitted from the upper stratosphere more sensitive to a unit temperature change.

    Thus, for (hypothetical – I’m not saying the ratios for CO2 are exactly like this) example, a forcing that instantaneously is 5 W/m2 at the tropopause level (a reduction in net outgoing LW flux plus increase in absorbed SW radiation below) accompanied with a 3 W/m2 forcing at TOA (top of atmosphere) imlies a forcing on the stratosphere of (3-5) W/m2 = -2 W/m2, and the combination will lead to an approximately [5 + (3-5)/2] W/m2 = (5 – 2/2) W/m2 = 4 W/m2 forcing at the tropopause level after stratospheric equilibration, give or, in some possible cases, take.

    For CO2, The LW TOA instantaneous forcing will be, after saturation in the center of the band (where saturation means that the outgoing LW flux has gone as low as it can before rising back up following increased emission from the warmer parts of the stratosphere) approximately proportional to changes in the logarithm of CO2 concentration, for the same reason as the tropopause level forcing is, but the difference in net LW flux in changing monochromatic CO2 optical thickness from zero to infinity is smaller at TOA than at the tropopause, and that will make the TOA forcing smaller.

    There is another effect, though; in the strongest part of the band, the outgoing LW flux can increase, adding some additional amount to stratospheric cooling. (On the other hand, there is also a SW CO2 effect – relatively small compared to the LW effect, but it will reduce the stratospheric cooling – although it will also have an instantaneous negative forcing at the tropopause level via shading from above, although it might also have some positive forcing via reducing the albedo.)


    For the greenhouse effect in general, it should be noted that while some wavelengths can approach saturation (zero net LW flux at the tropopause level), there (almost) always has to be some range of wavelengths for which there is remaining room for farther increased forcing; this is because there (almost) has to be a net LW flux up through the tropopause level to balance the solar heating below, since the convective heat flux through the tropopause is (approximately) zero.

    If there were some combination of greenhouse gases wherein an amount A saturated the forcing at the tropopause level for all waveleng (and ths and some greater amount were added, that greater amount could still generally be expected to cause additional warming, because among the feedbacks, the tropopause level would then have to rise high enough to ‘unsaturate’ at least some portion of the spectrum (and for a given lapse rate, this would mean higher surface temperature).


    (This gets into how radiative forcings and climate sensitivity to radiative forcing can be different between a forward and reverse change, if the change is large enough, assuming equilibrium is attained after each change, and even without hysteresis.

    For example, with some CO2, if all CO2 were removed, this would have a radiative forcing of X; the feedback would include a large drop in water vapor, with radiative feedback Y. After this colder equilibrium, if the same amount of CO2 were added back, then (setting all other feedbacks aside, as this is just an illustrative example) the forcing would be greater than X because of the lack of overlap with water vapor, and the water vapor feedback would be less than Y by the same amount due to the overlap with CO2. But even this doesn’t take into account lapse rate and tropopause-height changes, among other things. In general, the forcing by a change in CO2, or CH4 or the sun or aerosols, etc, will depend on the climatic state, as will the feedbacks, but in a complementary way so that, assuming the forcing mechanism is held fixed (no CO2 feedbacks, etc.), the equilibrium climate sensitivity to the change in CO2 or CH4, etc, will be the same in both directions between two states, assuming no hysteresis. But this isn’t so important for relatively small changes (relative to totals, not to consequences for life, etc.), including AGW so far as I know; it’s just interesting to note.)

  9. 759

    Climate skeptic Roy Spencer runs into another extremely hot month: Calls 2nd hottest in satellite record ‘quite warm’

  10. 760

    Thanks for the update, Kees. (#759.)

    It’s been interesting to watch the UAH persistently hang about at these elevated anomalies, and more interesting still to watch them be so persistently ignored by the denialosphere. Suddenly sea ice extent is so much more interesting!

  11. 761
    Patrick 027 says:

    PS in 758 it may not have been clear:

    1. Earthly conditions:

    SW radiation: scattering (clouds, air, surface) and absorption (surface material, water vapor, clouds, ozone, others) dominate, and reflection at the surface (sometimes diffuse reflection (scattering by a surface), sometimes specular reflection (calm water)).

    LW radiation: emission and absorption dominate, scattering and reflection are minor

    Raman and Compton scattering are minor (so far as I know).

    2. Alternative terrestrial planet, with dry-ice clouds (possibility for early Mars):

    scattering becomes more important for LW radiation; scattering is still mostly not Raman or Compton (so far as I know).

    3. For comparison: Photosphere of sun:

    (I actually am not sure whether Compton scattering or complete photon emission and absorption is more important in getting the photon intensity to near the blackbody values for the temperatures of the non-photon plasma, but I think it’s one of those.)

  12. 762
    Patrick 027 says:

    “1. Earthly conditions:”…

    All radiation: “Raman and Compton scattering are minor (so far as I know).”

  13. 763
    Gilles says:

    This is a general attitude in all climate debates, in my opinion. In the “warming sphere”, western Europe temperatures were very interesting in 2003, Hurricanes very interesting in 2005, and sea ice extent very interesting in 2007, but not so much the other years ….

  14. 764
    Completely Fed Up says:

    You forgot 1998 which, amongst denialists was, in 1999, an outlier indicating NOTHING. Though later on, 1998 became the date that was the END OF WARMING.

  15. 765
    Completely Fed Up says:

    “and sea ice extent very interesting in 2007, but not so much the other years”

    This too is a denialist meme: that 2007 is the date that the Arctic Ice Began To Recover ™.

  16. 766
    Completely Fed Up says:

    “(I actually am not sure whether Compton scattering or complete photon emission and absorption is more important in getting the photon intensity to near the blackbody values for the temperatures of the non-photon plasma, but I think it’s one of those.)”

    I don’t think you realise how deep the sun’s atmosphere is.

    It’s really, REALLY deep.

    The photosphere is 6000K. This doesn’t strip H and He of their electrons. But there are an awful lot of them.

    NOTE: if the sun were not optically thick at ~6000K, the photosphere would not “appear” to be at 6000K.

  17. 767
    Deep Climate says:

    Frank O’Dwyer raised the supposed “contamination” of surface temperature record by “economic factors”, as promulgated by Ross McKitrick and Patrick Michaels.

    McKitrick is at it again.

    The latest climate contrarian meme appears to be (baseless) accusations of scientific “gatekeeping” and “censorship”. Ross McKitrick provides an example of this unmistakable trend, with a blow-by-blow account of difficulties encountered in publication of his upcoming paper on the supposed contamination of the surface temperature record. The new paper purports to debunk a single statement in the 2007 IPCC Fourth Assessment Report, one denigrating the conclusions of a previous paper by McKitrick and Patrick Michaels.

    McKitrick criticizes the IPCC assertion that “locations of greatest socioeconomic development are also those that have been most warmed by atmospheric circulation”. He claims that other sections cited to support that statement do no such thing. But it turns out that McKitrick himself has it completely wrong, as he cites a passage concerning regional warming over the 21st century, instead of the actual relevant passage concerning the period 1975-2005.

    Moreover a review of the relevant scientific literature reveals substantial flaws in the previous analyses of McKitrick and Michaels. That, rather than any close-mindedness or “censorship”, is the real reason why McKitrick’s analyses have become increasingly marginalized in the scientific literature, if not in the right-wing press.

  18. 768
    Hank Roberts says:

    > gatekeeping

    Upwards of 80 hits at CA for ‘gatekeeper’ and twice that for ‘gatekeeping’ — this isn’t new, they’ve been throwing that label at anyone (including ordinary readers like me) who points out trolling or intentional misinformation.

    They don’t object to gates.
    They object to _thresholds_.

    It’s a much lower bar they’re trying to get rid of.

  19. 769
    Completely Fed Up says:

    “with a blow-by-blow account of difficulties encountered in publication of his upcoming paper”

    There can be difficulties in publishing a paper when it’s carp…

    “That, rather than any close-mindedness or “censorship”, is the real reason why McKitrick’s analyses have become increasingly marginalized in the scientific literature,”

    Or, yet again, just that it’s crap?

    And isn’t continually proclaiming the record is fixed and there’s a conspiracy an example of closed-mindedness?

  20. 770
    Deep Climate says:


    Well, I did charactrize McKitrick’s work as “substantially flawed” and cited that as the real reason for difficulties in publication. Other equivalent descriptions are possible, however.

  21. 771
    Patrick 027 says:

    Re 766 Completely Fed Up – Yes, I agree/accept that the photosphere (~ 500 km deep, although less than 0.5 kg per square meter; temperature goes from 6400 K to 4400 K through the layer: ) is opaque enough; I just wasn’t sure which specific form of photon-non photon interaction produced the opacity and allowed the photon gas temperature to be near that of the material. (Earlier I had suggested a dense plasma might have energy levels like a molten metal, but the photosphere is much less dense than air; I think Ray Ladbury mentioned scattering (Compton, presumably) as a mechanism but I don’t know if he specifically was refering to the photosphere…)

  22. 772
    Greg says:

    So I know that the radiative forcing concept is estimated using logarithmic equations because it’s only at the tips of GHG’s absorption spectra that increases make a significant difference. However, emissivity along all wavelengths increases with temperature, so as temperatures increase, won’t the greenhouse effect become stronger, at least in terms of W/m^2 at the tropopause? I’ve read about H2O as a feedback, but I haven’t heard/seen much of anything about this. Is it a relatively insignificant effect? Would this impact the temperature at which incoming/outgoing radiation would be at equilibrium?

  23. 773
    Completely Fed Up says:

    “I just wasn’t sure which specific form of photon-non photon interaction produced the opacity and allowed the photon gas temperature to be near that of the material.”

    The number of collisions helps a lot.

    It has also been spectrum-spread already by plasma collisions, so the need to thermalise is from one temperature (hundreds of thousands of degrees) to another (6000K) is easier.

    The pressure broadening and so on help too. As does a significant number of metallic atoms in the photosphere. Not forgetting a still somewhat present excited state (Halpha/Hbeta) on the general medium.

    IIRC the thermalisation is the same deal here: just takes a greater depth to manage it is all. The proportion of compton and electron shell really depends a lot on the microscale processes. But if it was just excitation then there’s still plenty of atmosphere to make it happen.

    At the corona (millions of degrees, but so thin that “temperature” doesn’t really mean much) and deeper inside the sun there are much more interesting phenomena.

  24. 774
    David B. Benson says:

    Greg (772) — Water vapor feedback more than doubles the direct effect of CO2, which is about 1 K for 2xCO2:

  25. 775
    Doug Bostrom says:

    Hank Roberts says: 6 April 2010 at 10:13 AM

    It’s a much lower bar they’re trying to get rid of.

    For McKitrick et al the ideal outcome would be to lower the bar of merit for everybody, making it impossible for the public and policymakers to distinguish between facts and fiction. When I lift a corner of the tinfoil tightly covering my cranium, I have to wonder if his obsessively detailed essay is all about establishing the case for adding peer review to the ever-growing list of “debatable” topics, such as has methodically been done to nearly every single collection of data lending confirmation of anthropogenic climate change.

    If there’s no method to McKitrick’s extended whine it remains pretty unflattering to him but with no conceivable benefit.

  26. 776
    John E. Pearson says:

    I just received this Dear john letter from the Union of Concerned Scientists:

    Dear John,

    UCS Climate Scientist, Dr. Brenda Ekwurzel will go head to head with a skeptical
    meteorologist during “couples counseling” tonight on the Colbert Report. It should
    be funny and informative. We hope you tune in, tonight, April 6 on Comedy Central at
    11:30 p.m. EDT.

    If you miss the program tonight, it should be available on the Comedy Central
    website over the next few days. Be sure to check it out!

  27. 777
    Greg says:

    David Benson (774):
    I understand that water vapor as a feedback strengthens other warming factors, but what I’m wondering is how much of an impact already-saturated wavelengths will have, since as temperature increases, emissivity will increase along all wavelengths, meaning that more heat will be captured by the same greenhouse gas concentrations.

    I don’t know if I’m making myself clear.

    On this site, there were two pages discussing the question of whether the CO2 effect was already saturated:

    So it’s clear that increasing GHG concentrations will decrease the amount of radiation escaping along certain wavelengths. What I’m wondering is, as temperatures increase and more radiation is emitted along already-saturated wavelengths, how will this impact the strength of the greenhouse effect? Is this accounted for in calculations of climate sensitivity?

  28. 778
    Hank Roberts says:

    Greg, don’t forget the stratosphere _cools_ as the troposphere and surface warm.

  29. 779
    John Peter says:


    Atmospheric CO2is not saturated according to atmospheric measurements.

    “…If the CO2 effect was saturated, adding more CO2 should add no additional greenhouse effect. However, satellite and surface measurements observe an enhanced greenhouse effect at the wavelengths that CO2 absorb energy. This is empirical proof that the CO2 effect is not saturated…”

  30. 780
    Patrick 027 says:

    From that NASA website (see previous comment) “The average density of the photosphere is less than one-millionth of a gram per cubic centimeter. ”

    1e-6 g/cm3 = 1e-6 kg/L = 1e-3 kg/m3 = 1 kg/(m2 km)

    I forgot a factor of a thousand; the photosphere has less than 500 kg per m2.

    But just to be clear, I was never arguing that the photosphere couldn’t be emitting approximately as a blackbody (would be a blackbody if isothermal) (PS until reading that NASA website I wouldn’t have guessed that the photosphere was made of as little as 0.5 kg/m2 – which turned out to be wrong, anyway) , with photons emitted over a broad range of frequencies with an intensity nearly in thermodynamic equilibrium with the material’s temperature; I was only saying I wasn’t sure which mechanism dominated the photon – non-photon energy transfers under such conditions.

  31. 781
    John Peter says:


    According to Ram, in the four isotopes of CO2, there are several tens of strong fundamental and weaker isotopic and excited state absorption bands. The strong bands, logarithmic with concentration, provide 80% of the absorption. The weak bands, linear with concentration, provide the rest.

    There should always be “room at the inn”, if not with the fundamental, then with a weaker isotopic band (actually a mess of lines)

  32. 782
    BobFJ says:

    David B. Benson Reur 726/p15:

    BobFJ (724) — Yes, one needs net of all forcings plus some index of internal variabiliy to remconstruct the instrumental record. On a decadal scale, the MAO acts as such an index, but imperfectly so since the net of the nonlinear portion of forcings will also effect the AMO.

    I’ve had a quick Google on such internal variabilities, and there do indeed appear to be some good correlations. The combination of AMO + PDO in particular surprised me, because it appears to correlate very well with the 1940 through 1975 cooling period. (although its coefficient WRT global surface T is ? )

    I’ve made a composite graph with the HADCRUT NH and SH here, to look at hemispherical bias. The southern oscillations, including the famous 1998 ENSO seem to have a strong influence on the NH.

    This science article is rather good I think: Pacific Decadal Oscillation (PDO) + Atlantic Multidecadal Oscillation (AMO)

    This one has some useful supplementary references and figures:
    Multidecadal Ocean Cycles and Greenland and the Arctic
    BTW, I’m still hoping Tamino will respond to my Email to help simplify and consolidate my concerns on his “Volcanic Lull” article…. Perhaps he has taken an Easter break?

  33. 783
    David B. Benson says:

    BobFJ (782) — Thanks, but I’ll stick with just the AMO since it seems the more important and picks up some of the PDO variation in any case.

    Tamino is moving his household and otherwise professionally very busy.

  34. 784
    Hank Roberts says:

    Don’t forget to page down for the really wacko stuff.
    Bob, that bag of hammers award? It’s the sources you use, as well as what you do with what you find there.

  35. 785
    BobFJ says:

    Completely Fed Up,
    You wrote in full (764):

    You forgot 1998 which, amongst denialists was, in 1999, an outlier indicating NOTHING. Though later on, 1998 became the date that was the END OF WARMING.

    Where did that come from? Are you able to advise of any sceptics that claim that the El Nino of 1998 occurred in 1999? You could perhaps study this NOAA graph and note that it is an irregular oscillation which BTW feeds into the PDO. (See my 782 to David B Benson)

    You might also try to comprehend this logical consideration:
    Although it may contain some noise, it is probably unfair to treat 1998 as an outlier because it is a recognised oscillation which is proven to have consequences on global temperatures, including in the NH. Furthermore, it is typical that when there is a sharp positive spike, it is usually followed by a sharp negative “correction”. (which makes sense in terms of ocean turn-over) Thus the mean of the high 1998 temperature when married with the low 1999 + 2000 temperatures is not all that special. If for instance there is an attempt to correct for the 1998 high, then there should also be a correction for the 1999 + 2000 lows. Thus, although controversial, there is a logical consideration that the current warming plateau which is best illustrated in the HADCRUT3 global average; can be sensibly considered as starting in 1998.

  36. 786
    Patrick 027 says:

    Re 772,777 Greg –

    1. Clarification: Emissivity is an optical property; it can be a function of temperature, but so far as I know that is not a strong feedback (though it may be significant or substantial, along with differences in line broadenning, to how the optical thickness of 1 g/m2 near the surface differs from 1 g/m2 near the tropopause or 1 g/m2 within the stratosphere, etc, for the same substance.)

    I got the sense that you may have been thinking of the increase in LW emission with increasing temperature as described by the Planck function for blackbody radiation. With optical properties held constant, an increase in temperature will increase the flux of emitted LW radiation.

    Some portion of the increase in LW emissions within the troposphere and from the surface are absorbed within other parts of the troposphere+surface. Some escapes, changing the net flux across the tropopause; the source of this radiation is distributed among the surface and layers of troposphere; this is what must increase (via changes in temperature) to balance the tropopause level forcing (with equilibrated stratosphere) + other feedbacks (water vapor (LW, some SW), clouds (SW and LW feedback, surface changes (SW, maybe some LW), other compositional changes, and the stratospheric response to tropospheric warming). Because the tropopause, which shifts position as part of the climate response, doesn’t have a temperature discontinuity, the net flux where the greenhouse effect is saturated at the tropopause level remains zero. Thus, the increase in upward LW radiation at the tropopause occurs at other wavelengths; surface temperature has the biggest effect on the tropopause LW radiative flux between about 8 and 12 microns.

    3. A radiative forcing caused by a change in optical properties or amount or distribution of something with some optical properties depends on the climate state (temperature dependence of LW radiation, overlaps and saturation of optical effects); thus the same such change can have a different magnitude of radiative forcing than the opposite change if climate is allowed to reach equilibrium after each step (even assuming no hysteresis). However, the same is true of feedbacks, and the effects will be complementary, so that without hysteresis, the same magnitude of equilibrium climate response will occur in reverse as in forward. These differences are small for small (relative to totals) changes (so far as I know, doubling CO2 to 600 ppm would have about the same magnitude of forcing as halving CO2 from 600 ppm, though I haven’t actually studied the issue in that detail). They are more important for large changes (taking out all CO2 or adding it all back, for example).

  37. 787
    BobFJ says:

    Hank Roberts Reur 784:
    Bag of hammers award”? I have not heard of that one. Is that like that “Nobel Award” thingy given to Al Gore et al?

    BTW, the figures and references that I was interested in at “” seemed to be well sourced to me upon a first quick Google, and they still do upon a second look. You should try to be more constructive perhaps, regardless of what you opine on the website itself?

    Oh, and incidentally, you seem to have missed an important point in my 782. One of the sceptic arguments is that you guys cannot explain the cooling period between 1940 and 1975. However, I was very surprised to see that a graphical combination of the AMO and PDO indexes, (subject to whatever calibration or weighting may have been used), appears to have a good correlation with that cooling. Thus, if it is correct, that sceptical criticism would be defeated, but nevertheless, you argue against it! Golly Gosh!

  38. 788
    Completely Fed Up says:

    “Thus, if it is correct, that sceptical criticism would be defeated, but nevertheless, you argue against it! ”

    Isn;t pointing out the PDO etc effects arguing against it?

    So why the “golly gosh!”?

  39. 789
    Completely Fed Up says:

    “1. Clarification: Emissivity is an optical property; it can be a function of temperature, but so far as I know that is not a strong feedback”

    Except where the temperature is such that the bulk radiation of the thermal medium introduces significant or overwhelming flux in the frequencies you are investigating.

    I.e. at the IR range, the temperature of the bulk atmosphere also radiates, so it introduces a significant element in the flux. Temperature changes ~300K make a lot of difference to the IR fluxes. For the Sun at 6000K, optical wavelengths (such as OII, Ha/Hb, etc absorption wavelengths) are emitted and so the bulk medium contribute to the visible fluxes.

    This isn’t an effect of emissivity changes but of the invalidity of assuming the bulk has no effect on the flux that Beers Law, for example, assumes.

  40. 790
    Completely Fed Up says:

    “Where did that come from?”

    et al.

    “It’s been cooling for the last 12 years!”.

    2 years ago (2008), “It’s been cooling for the last 10 years!”.

    And so on.

    YOU have made that accusation, BJ.

  41. 791
    Completely Fed Up says:

    “(PS until reading that NASA website I wouldn’t have guessed that the photosphere was made of as little as 0.5 kg/m2”

    Do you want to check your calculations?

    Are you talking of the *pressure* at the nominal photosphere surface?

  42. 792
    Completely Fed Up says:

    “Is that like that “Nobel Award” thingy given to Al Gore et al?”

    No, because that Nobel Award is the Nobel Peace Prize. It’s genuine. Hence no scare quotes.

    Whereas you get a bag ‘o hammers to see if you can manufacture a clue.

  43. 793
    Hank Roberts says:

    Pretty pictures. There are lots of them.

    Push them around til they line up.
    Some do.
    Have you explained something?

    Anything new?


    Some temperature variation is heat moving around.

    We knew that. It’s called weather. It’s always happening. It happens in the atmosphere, and it happens in the ocean.

    Temperature goes ’round and ’round.

    Can temperature moving around explain a recent longterm trend?

    It’s a very strong signal — that goes up and down; it has no trend.

    Can we sort the weather out and see if there is any way to conclude there is no sign of the predicted trend from greenhouse gases?

    Let’s look:

    You have nothing, Bob. You’re just pushing lines around without any explanation, and your pictures don’t

  44. 794
    John E. Pearson says:

    A ten page article on the economics of climate change by Paul Krugman: Building a Green Economy

  45. 795
    Patrick 027 says:

    Re 789 Completely Fed Up (Re my 786):

    I was refering to the emissivity for the composition as is, being a function of temperature but I don’t think that’s a significant climate feedback (ie a particular line strength for 1 g/m2 of gas at sea level may be (?)significantly(?) different from that of 1 g/m2 of the same gas at the tropopause level due to temperature; for that matter, the line broadenning varies with height (function of pressure, temperature, and also, composition of the medium), so the optical thickness …

    (which is, aside from these complexities, linearly proportional to amount of substance per unit area or number of some category of particles per unit area (and different contributions to optical thickness add linearly); transmission in a particular direction being exp(-optical thickness in that direction from point A to point B), which pertains to, of photons of a given frequency and polarization which are emitted or scattered into a path that reach point A, the fraction that reach point B without subsequent absorption or scattering; a relationship that holds regardless of what is emitted into the path between points A and B; to find the total intensity at point B, one must integrate from B back towards A and beyond, the emission + scattering into the path per unit distance * the transmission from that location to point B)

    … contributed by 1 g/m2 will not be the same as that contributed by 1 g/m2 at a different location, particularly a different vertical location; however, the temperature changes (and pressure changes such as from the addition of mass to the atmosphere from water vapor feedback) that are part of a significant climate change cause are considerably smaller than the variations over height, so I’d guess that the dependence of optical properties (for a given composition and physical phase) on temperature (and pressure) is not a significant source of radiative feedback compared to the Planck response (the temperature dependence of LW emission for a given set of optical properties and the spatial and temporal arrangment of such) and feedbacks involving changes or rearrangments of composition and physical phase (water vapor, clouds, snow and ice, vegetation, and acting through those, circulation changes…), etc.

    Re me 786 Re Greg:

    On point 2:
    “Some portion of the increase in LW emissions within the troposphere and from the surface are absorbed within other parts of the troposphere+surface.”

    That of course doesn’t directly result in any net heating or cooling of the whole of the surface+troposphere.

    (It is analogous to looking inside a block of red-hot iron. Inside, red radiation is being emitted and absorbed over rather short distances. If there were a temperature gradient within the material, this radiative energy flux would carry heat from hot to cold; if the material were made more opaque, this internal net flux would decrease because the shorter distances of radiative exchange would be between more similar temperatures – if it were made sufficiently more transparent, and/or if enough of the opacity came from scattering, then this net flux would decrease because the regions with different temperatures would not be able to absorb and emit as much radiation; however, the nonlinearity in blackbody intensity relative to temperature…

    (which is, relative to the peak wavelength or frequency for blackbody radiation for the given temperature, greater at shorter wavelengths/higher frequencies; at sufficiently long wavelengths/low frequencies, blackbody intensity is approximatley linearly proportional to temperature)

    … means that if the block’s average temperature were increased with the same temperature gradients, then the net fluxes from hotter to colder parts would increase.

    (of course, unlike the atmosphere, thermal conduction is a very important pathway for heat fluxes within a block of iron).

    But the radiation emitted and absorbed by the block to and from the outside is what is responsible for radiational energy gain or loss.

    The greenhouse effect reduces the net LW heat loss from the surface+troposphere by concentration the source of emission of outgoing radiation into the colder parts (by reducing the distances that photons can travel between emission and absorption) while also increasing downward LW radiation from the stratosphere. To reach equilibrium, the surface+troposphere have to warm up enough to increase the outgoing LW radiation to balance the initial forcing plus any other feedbacks (LW and SW).

    The forcing and feedbacks and the temperature change also affect internal energy fluxes within the surface+troposphere, but convection generally tends to respond (with latitudinal and region/seasonal/diurnal etc. variations in the tendency) in the opposite way (a decrease in net LW cooling of the surface or increase in SW heating of the surface would tend to cause an increase in convective cooling of the surface), so (to a first approximation, with deviations in space and time from this tendency – see an earlier comment somewhere in this thread) the temperatures of all parts of the surface and troposphere tend to respond to the fluxes across the tropopause. Of course, the changes in convection (and latent heating/cooling) that respond to radiative heating/cooling changes (which are reshaped by the former) are a part of the climate (it’s not just about temperature, of course).

  46. 796
    Completely Fed Up says:

    Patrick, I figred that half way through, but I suspect that among those who didn’t do degree level physics, emission and emissivity aren’t all that different.

    Especially the way the other guy was using it.

    Emission in IR DOES depend on temperature.

  47. 797
    Completely Fed Up says:

    “for that matter, the line broadenning varies with height (function of pressure, temperature, and also, composition of the medium),”

    It depends on velocity of the constituents and the pressure *if* the energies involved are delicately balanced: OIII is a transition that doesn’t exist here because the pressure we can obtain (well in the days when OIII was thought to be the signature of a new element, can’t remember the name off the top of my head, but something along the lines of “plasmium”) removes the transition from the available states. That’s nothing to do with line broadening, however.

  48. 798
    Completely Fed Up says:

    PPS (I’ll get around to finishing soon…), line broadening hasn’t got anything to do with emissivity (inelastic collisions do, if they occur while emission is going on), unless you’re talking emissivity per unit wavelength, in which case the emissivity changes on wavelength because more energies are acceptable to the system.

  49. 799
    Completely Fed Up says:

    “but what I’m wondering is how much of an impact already-saturated wavelengths will have,”

    At some level up, (since we have a vacuum of space at one end, the atmosphere HAS to go to nothing at some point going up) those saturated wavelengths are no longer saturated.

    Adding more then makes them saturated.

    And since the temperature goes down as you go up, IR flux to space goes down.

    [Response: And for those still confused about the ‘saturated wavelengths’ idea, there is a good discussion by Ray Pierrehumbert in an earlier RealClimate post, here–eric]

  50. 800
    CTG says:

    Re #785 BobFJ

    We have to remember that BobFJ mainly comes round here looking for insults so he can go and whine to his friends how mean we all are at RC. But hey, let’s not disappoint him.

    Are you able to advise of any sceptics that claim that the El Nino of 1998 occurred in 1999?

    So that would be a fail on reading comprehension, then. CFU is pointing out that in 1999 the denialist meme de jour was that 1998 should be ignored as an outlier. Once 2008 came along, though, suddenly 1998 became the “starting point of the plateau”. But then came 2009, and suddenly 11 year trends went out of fashion, and 12 became the new 11. Whoops, there go the goalposts again.

    If for instance there is an attempt to correct for the 1998 high, then there should also be a correction for the 1999 + 2000 lows.

    Gosh, if only there were some sort of statistical technique that could do that sort of correction. Not just for 1998/1999/2000, but for all years. Sort of like averaging out the noise or something. Like this maybe. Now where’s your plateau?

    Oh, and by the way – most of your denialist friends have already given up on 1998 as the “turning point”, and talk about 2001 instead, as it avoids those inconvenient 1999/2000 low points. You guys really ought to get your story straight – all these contradictions just make it look as though you just don’t know what you’re talking about.