RealClimate logo

Technical Note: Sorry for the recent unanticipated down-time, we had to perform some necessary updates. Please let us know if you have any problems.

Water vapour: feedback or forcing?

Filed under: — gavin @ 6 April 2005 - (Deutsch)

Whenever three or more contrarians are gathered together, one will inevitably claim that water vapour is being unjustly neglected by ‘IPCC’ scientists. “Why isn’t water vapour acknowledged as a greenhouse gas?”, “Why does anyone even care about the other greenhouse gases since water vapour is 98% of the effect?”, “Why isn’t water vapour included in climate models?”, “Why isn’t included on the forcings bar charts?” etc. Any mainstream scientist present will trot out the standard response that water vapour is indeed an important greenhouse gas, it is included in all climate models, but it is a feedback and not a forcing. From personal experience, I am aware that these distinctions are not clear to many, and so here is a more in-depth response (see also this other attempt).

First some basics. Long-wave (or thermal) radiation is emitted from the surface of the planet and is largely absorbed in the atmosphere. Water vapour is the principle absorber of this radiation (and acknowledged as such by everybody). But exactly how important is it? In terms of mass, water vapour is much more prevalent (about 0.3% of atmospheric mass, compared to about 0.06% for CO2), and so is ~80% of all greenhouse gases by mass (~90% by volume). However, the radiative importance is less (since all molecules are not created equal). One way to quantify this is to take a radiation model and remove each long-wave absorber (principally the greenhouse gases, but also clouds and aerosols) and see what difference it makes to the amount of long-wave absorbed. This gives the minimum effect from each component. The complementary calculation, using only each particular absorber in turn, gives the maximum effect. Generally these will not be equal because of overlaps in the absorbing spectra (i.e. radiation at a particular frequency can either be absorbed by water vapour or CO2).

Removed absorbers Fraction LW Rad. Forcing
absorbed Tropo. (W/m2)
None 100% 0
H2O 64 (64, RC78) -56
Clouds 84 (86, RC78)
CO2 91 (88, RC78) -23
O3 97 (97, RC78)
Other GHG 98 -3
H2O+Clouds 34
H2O+CO2 47 -89
All except H2O+Clouds 85
All except H2O 66 (60-70, IPCC90)
All except CO2 26 (25, IPCC90)
All except O3 7
All except Other GHG 8
All 0%
Instant calculation, global mean, Jan. 1, 1979 RC78=Ramanathan and Coakley (1978)
‘All’ includes aerosols, O3 and other minor gases as additional absorbers.

The table shows the instantaneous change in long-wave aborption when each component or combination of components is removed using the radiation code from the GISS GCM. (The source code is available for those who have the patience to get it to work). This isn’t a perfect calculation but it’s quick and easy and is close enough to the right answer for our purposes. (N.B. This is very similar to what was done by Ramanathan and Coakley (1978) using a single column model – their numbers are in the table for reference). [Update Oct 2010:The numbers in this post have been somewhat updated and published in Schmidt et al (2010). ]. Because of the overlaps, the combined changes are larger than the changes due to each individual component. Another calculation is the instantaneous radiative forcing at the tropopause, but that is complicated for clouds, O3 and Aerosols which have impacts on solar radiation as well as the long wave, so I only give that value for the ‘pure’ greenhouse gases.

The overlaps complicate things, but it’s clear that water vapour is the single most important absorber (between 36% and 66% of the greenhouse effect), and together with clouds makes up between 66% and 85%. CO2 alone makes up between 9 and 26%, while the O3 and the other minor GHG absorbers consist of up to 7 and 8% of the effect, respectively. The remainders and uncertainties are associated with the overlaps which could be attributed in various ways that I’m not going to bother with here. Making some allowance (+/-5%) for the crudeness of my calculation, the maximum supportable number for the importance of water vapour alone is about 60-70% and for water plus clouds 80-90% of the present day greenhouse effect. (Of course, using the same approach, the maximum supportable number for CO2 is 20-30%, and since that adds up to more than 100%, there is a slight problem with such estimates!).

Since we are looking at the whole of the present-day greenhouse effect (around 33 C), it is not surprising that the radiative forcings are very large compared to those calculated for the changes in the forcing. The factor of ~2 greater importance for water vapour compared to CO2 is consistent with the first calculation.

So where does the oft quoted “98%” number come from? This proves to be a little difficult to track down. Richard Lindzen quoted it from the IPCC (1990) report in a 1991 QJRMS review* as being the effect of water vapour and stratiform clouds alone, with CO2 being less than 2%. However, after some fruitless searching I cannot find anything in the report to justify that (anyone?). The calculations here (and from other investigators) do not support such a large number and I find it particularly odd that Lindzen’s estimate does not appear to allow for any overlap.

While water vapour is indeed the most important greenhouse gas, the issue that makes it a feedback (rather than a forcing) is the relatively short residence time for water in the atmosphere (around 10 days). To demonstrate how quickly water reacts, I did a GCM experiment where I removed all the water in the atmosphere and waited to see how quickly it would fill up again (through evaporation from the ocean) . The result is shown in the figure. It’s not a very exciting graph because the atmosphere fills up very quickly. At Day 0 there is zero water, but after only 14 days, the water is back to 90% of its normal value, and after 50 days it’s back to within 1%. That’s less than 3 months. Compared to the residence time for perturbations to CO2 (decades to centuries) or CH4 (a decade), this is a really short time.

Only the stratosphere is dry enough and with a long enough residence time (a few years) for the small anthropogenic inputs to be important. In this case (and in this case only) those additions can be considered a forcing. Oxidation of anthropogenic methane (which is a major source of stratospheric water) and, conceviably, direct deposition of water from increases in aircraft in the lower stratosphere, can increase stratospheric water and since that gives a radiative forcing effect, they do appear on the forcings bar chart (under “H2O from CH4“). Some scientists have argued that changes to irrigation and other land use changes (which effect evaporation) are also direct forcings to water vapour amounts, but I think it’s cleaner to think of that as an indirect water vapour response to the change.

When surface temperatures change (whether from CO2 or solar forcing or volcanos etc.), you can therefore expect water vapour to adjust quickly to reflect that. To first approximation, the water vapour adjusts to maintain constant relative humidity. It’s important to point out that this is a result of the models, not a built-in assumption. Since approximately constant relative humidity implies an increase in specific humidity for an increase in air temperatures, the total amount of water vapour will increase adding to the greenhouse trapping of long-wave radiation. This is the famed ‘water vapour feedback’. A closer look reveals that for a warming (in the GISS model at least) relative humidity increases slightly in the tropics, and decreases at mid latitudes.

How do we know that the magnitude of this feedback is correctly simulated? A good test case is the response to the Pinatubo eruption. This caused cooling for up to 3 years after the eruption – plenty of time for water vapour to equilibriate to the cooler sea surface temperatures. Thus if models can simulate the observed decrease of water vapour at this time, it would be a good sign that they are basically correct. A good paper that demonstrated this was Soden et al (2002) (and the accompanying comment by Tony DelGenio). They found that using the observed volcanic aerosols as forcing the model produced very similar cooling to that observed. Moreover, the water vapour in the total column and in the upper troposphere decreased in line with satellite observations, and helped to increase the cooling by about 60% – in line with projections for increasing greenhouse gases.

To be sure there are still some lingering uncertainties. Some recent data indicates that tropical upper tropopsheric water vapour does not quite keep up with constant relative humidity (Minschwaner and Dessler, 2004) (though they still found that the feedback was positive). Moist convection schemes in models are constantly being refined, and it’s possible that newer schemes will change things . However, given the Pinatubo results, the models are probably getting the broader picture reasonably correct.

*R.S. Lindzen, 1991. Quart. J. Roy. Met. Soc., 117, pp. 651-652

111 Responses to “Water vapour: feedback or forcing?”

  1. 51
    Stephen Berg says:

    In response to #48, this paper seems to suggest that SSTs are a factor in atmospheric CO2 concentration:

    From the article’s conclusions section:

    “…[atmospheric] CO2 concentration fields, calculated in the recent years over an area including Europe, northern Africa and the Boreal Atlantic Ocean, display a seasonal cycle that seems well related with the seasonal variations of SST.”

    “…during Spring, the fields exhibit
    medium and low values, but during Summer the increase
    of the SST (stronger near the Tropic of Cancer) seems in
    good agreement with the appearance of intense maxima
    of atmospheric CO2 concentration over the southern
    part of the extratropical Atlantic Ocean.”

    With oceans warming as a result of climate change, you would assume that the same would be true, not just on a seasonal basis, but annually and decadally, as well. In other words, as the global mean temperatures increase (which should and does result in an increase of SSTs), so should atmospheric CO2 concentrations.

  2. 52

    For additional information on water vapor feedback, I’d like to point out the article I wrote on water vapor for the CalTech General Circulation Symposium, “On the relative humidity of the Earth’s Atmosphere” It’s posted on my web site ( under “publications.”
    Eventually, it will come out in the general circulation book published by Princeton University Press

    This article has some calculations which bear on the old “98%” myth. By my calculations, in the Tropics, CO2 is more like 1/3 of the greenhouse effect, and it’s pretty clear that if you took it out the Earth would fall into a snowball state.

    Some other things in that article that may be of interest include:

    * A calculation of how much the climate would change if water vapor were completely removed, if water vapor were forced to be saturated, and if water vapor content were half that produced by standard parameterizations,

    *An analytic argument saying why relative humidity should stay roughly constant as climate changes, together with some GCM studies shedding light on the way the argument works.



  3. 53

    As one of the authors of the TAR chapter on “Physical Processes and Feedbacks,” I can state definitively that water vapor feedback was not ignored in TAR. We read and considered several dozen key articles on observation and modelling, and I can assure you the discussions in our drafting sessions were lively — particularly in view of the fact that Dick Lindzen was also an author on this chapter. I think we were quite clear about the nature of the possible uncertainties (notably regarding the role of microphysics in transport), but also about the reasons for believing that models are probably not too badly wrong with regard to water vapor feedback.

    By the way, the characterization of Minschwaner’s paper as saying that observed water vapor feedback is less than that in GCM’s is a gross oversimplification and over-interpretation of their result. Among other things, their observations apply only to the very upper part of the troposphere, which accounts for only a small part of the total water vapor feedback.

  4. 54
    Mike Doran says:

    From the above article:

    “How do we know that the magnitude of this feedback is correctly simulated? A good test case is the response to the Pinatubo eruption.”

    Again the example is confounded with the complexity of large scale electrical features on cloud microphysics, and then back to the reverse, the affect of the varying chemistry of microphysics on large scale cloud dyanmics. Pinatubo is perfect example because not only does the SOx drop the phase change temperature of heat trapping cirrus, for instance, but the ion content of forming clouds is altered or was altered and then in DC fields the microphysics affect of large scale DC fields is altered per the China paper. Think of it this way–the ions in the super cooled droplet will migrate in the DC field to the opposite pole charge–negative ions toward relatively positive ionosphere and negative ions toward relatively negative oceans–and you get asymmetries of cloud microphysics. If the microphysics becomes relatively too asymmetrical, the cloud mass in the relatively strong DC field has no chance of competing against forming clouds where the asymmetries do not exist. The result is that if the ion content is greater from, say, SOx disolved to sulfur acid, these asymmetries become more extreme.

    I am a student of hurricanes, and Pinatubo is particularly important with respect to Hurricane Andrew for this vary reason, in terms of cloud and wind levels, as well as intensities.

    But what is even more interesting following Pinatubo is the 1997-8 El Nino. Conductivity patterns, largely biological, flow from a nutrients and upwellings in relation to both the cooling that followed the eruption as well as the SOx consumed by sulfur loving microves and relativel bloom activity in relation to that nutrient. Pinatubo gives a sense of both the input and the modulation by the living earth.

    “They found that using the observed volcanic aerosols as forcing the model produced very similar cooling to that observed. Moreover, the water vapour in the total column and in the upper troposphere decreased in line with satellite observations, and helped to increase the cooling by about 60% – in line with projections for increasing greenhouse gases.”

    But again this is confounded by the microphysics impact by large scale electrical features. Okay. How to explain? Appreciate that these DC fields are basically virtual capaciter behaviors between the ionosphere and the oceans–those as the two ‘plates’. The static fields in the ionosphere are held in pattern by the earth’s magentic field . . . and, of course, the oceans are conductive. The microphysics of clouds–they are inbetween ionosphere and ocean where the DC field exists. However, also inbetween is WATER. Water, in droplet or vapor or ice form, has an impact on how the magnetic fields flow. This impact, mathematically, is expressed by dielectric constant. It turns out that water is about 80 times the dielectric constant to air. What that then means is, like Pinatubo, increases in CO2 and other human inputs have changed the global electrical circuit, which then impacts cloud microphysics behaviors, and changes water content, which in turn has its own electrical significance. Climate change.

    “Moist convection schemes in models are constantly being refined, and it’s possible that newer schemes will change things . However, given the Pinatubo results, the models are probably getting the broader picture reasonably correct.”

    The models change because the context of CO2 as a gas exchange and water content is changing. There are, of course, other natural electrical changes that occur as well. For instance, in the hurricane season there is a pattern called the QBO which is a wind in the tropics–caused by induction not unlike an electric engine on the light upper atmosphere. Dr. William Gray has made a living using that as a factor for his hurricane forecasts, but, like the Nirvana song about the man with a gun, but he don’t know what it means, Dr. Gray sees a factor but does not know why it impacts the hurricanes the way it does. That impact is really a switch in the global electrical circuit, which then impacts the ionosphere and hence how these electrical couplings and microphysics patterns occur.

    [Response: You appear to be confusing a number of issues here. The first order impact of Pinatubo was the direct effect of increased reflective aerosols in the lower stratosphere. This was very clearly observed and has been shown to explain very well the subsequent cooling of the planet. There were of course secondary effects (for instance on stratospheric ozone) that are also of great interest. Increases in upper tropospheric aerosols due to settling may have affected cloud microphysical properties too (through the standard aerosol indirect effects). Impacts on the global electric circuit due to the increased number of ions is an intriguing possibility, but a demonstrated (rather than hypothesised) connection to cloud formation and climate remains elusive. Regardless of whether such a mechanism exists, it was not the dominant effect after Pinatubo. With respect to the QBO, you should read the classic papers by Lindzen (and others) demonstrating very clearly that the mechanism is related to gravity wave-mean flow interaction and has nothing to do with electrical induction. -gavin]

  5. 55
    John Finn says:

    Gavin, in your response to John A, you say

    Given a system like the Earth’s climate, the best one can hope for is a reasonable match to observations

    From the Minschwaner and Dessler paper (referenced by you), observations suggest the following increases in water vapour in the upper troposphere

    3 ppm/K +/- 1.2 ppm/K
    1.5 ppm/K +/- 1.7 ppm/K (excluding 1997-98 El Nino data)

    What do the IPCC models predict – 30 ppm/K? Whatever – it’s likely to be of that order for a constant RH.

    So let’s say that some, possibly hypothetical, scientist had developed a model several years previously which predicted an increase in water vapour of, say, around 1.7 ppm/K. Would he be dismissed as just another crank contrarian?

    [Response: I checked both the GISS model response to interannual SST variations (1990-1999) and the response to 2xCO2, and your estimate of what the models predict for the tropical upper troposphere mean humidity is about right: 20-28 ppm/K at 180mb (smaller above that, larger, below). I haven't done the calculations for any other model, so there may well be differences among them. I strongly doubt that any of the modelling groups will be described as 'cranks' if their numbers are closer to the MD04 estimates. Upper tropospheric processes are a known source of uncertainty, as acknowledged very clearly in IPCC, and getting more data and improving the models for this region are active areas of reasearch. They are however only a very small part of the overall water vapour feedback. - gavin]

  6. 56
    DrMaggie says:

    Re #45: “It is strange that online hourly or live CO2 measurements seem to be extremely rare (and/or difficult to find), whereas data with daily and monthly CO2 means are plenty….”

    For those interested in looking at flux data (including CO2) from studies of various ecosystems, there is actually quite a lot of data available for downloading on the web. One example is the data archive of the EUROFLUX and MEDEFLU projects: Depending a bit on the measurement site, data is available with a number of different time resolutions, from half-hourly to yearly. Datasets can be downloaded freely (e.g. as formatted ascii files), and there are links to documentation about the various parameters. (We used data from this site in an applied ecosystems modeling course I attended last year, with great success.)

  7. 57
    Mike Doran says:


    There indeed was an albedo impact from the eruption. The assumption you make in ‘determination’ of what is primary and what is secondary is in relation to how long an effect takes place. Volcanic inputs are actually quite short termed given the microphysics ongoing impact by the ocean.

    The oceans are huge. A volcano eruption in relation to the oceans is tiny. Within a short time the SOx emissions from the volcano are modulated away, fall out of the atmosphere. Even short term microphysics properties are more a function of ocean saltiness than the SOx in the air. The best example of that is Hurricane Andrew–which is the most proximate and recognizable storm in relation to the Pinatubo eruption. Then the 1997-8 El Nino is a reflection of upwelling from cooling leading to algae blooms from upwelling and increases in nutrients to the surface of the oceans and the Sox as a nutrient leading to sulfur loving blooms–and the living earth feedbacks which caused a warming event. Water vapor is about control, modulation, dampenings–from the perspective of a living earth. Your analysis that volcanic inputs are a primary forcing, from the standpoint of inputs only–in this case, SOx, then, misses the context of human activities.

    “With respect to the QBO, you should read the classic papers by Lindzen
    (and others) demonstrating very clearly that the mechanism is related to
    gravity wave-mean flow interaction and has nothing to do with electrical

    I wouldn’t trust Lindzen’s view of atmospherics without electrical cloud microphysics properties. He, and other scholars in the field, lack the ken required to analyze the problem. See:

    The QBO is entirely about such an electrical switch which impacts cloud microphysics, or water vapor. The QBO
    reverses from the TOP down. IOWs, the upper part of that wind slows and starts to reverse first. Lower atmospheric wave conditions would merely cause bottom down reversals . . . The reason this top down reversal occurs as such is that there is an ELECTRICAL FIELD switch FIRST which is followed by movement first of the least dense aspect of the atmosphere by induction, because that part of the atmosphere has the least amount of momentum to mass and therefore the force exacted can move on it. The changes in microphysics globally as a result then causes a feedback in the global electrical circuit which then eventually results in a switch, and the process reverses. Rossby waves, all these features which Lindzen and others address are forced by these switches in the global electrical circuit. They confuse condition with sign.

    Presently the oceans are too salty for nucleotide complex sortings in clouds to have a significant forcing impact per the China paper. Life has grown . . . has been sorted to become so complex that chemistries are now contained in cells. Cells that in cumulation have significant enough conductivity meaning over chemical diffusion that there can be electrical modulation without the nucleotide complex modulating inside the cloud parasols, and the aid of saltwater spray as what was previously â??noiseâ?? against a signal of these complexes has become an aid for a living earth, made of cells, in its modulations. The earth has been designed to have saltier oceans! Of course, there are diatoms and other cellular life that find themselves as cloud nucleatation particles, and that is part of the modulation, but purely from a chemical standpoint, the oceans are now too salty to allow direct cloud microphysics control or design by nucleotide complex.

    Modulation without cells cannot take place if the cloud has too much salinity in it blown up from the ocean per the China paper. Membranes evolved because they contain chemistries over diffusion, and that then has a significant ELECTRICAL meaning in cumulation. Yet, in the cloud parasol a membrane will prevent the sorting by size, shape, mass and charge of the nucleotide complex as the parasol super cools and forms or not into a cloud droplet/particle. This led to sexual reproduction, which is a complexity that connected symbiotically nucleotide complex (male) to membraned life (female egg). The complexity of a living earth in relation to water content in the air and volcanoes are the sulfur loving microbes. These are amoung the first life forms, in the catagory extremophiles or archae.

  8. 58
    John Dodds says:

    Re #15, 30 & 56.
    Wow, neat data about how daily CO2 actually rises from the baseline ~390ppm to over 500ppm on many nights, along with the relative humidity, and then falls back to baseline as the sun warms the air up. It is just like seeing the results of an actual run of a simulation program but with a daily solar forcing function.

    Seems funny how the daily CO2 concentration in air goes INVERSELY with temperature or solar radiation, when on the yearly ice core data CO2 temp and water vapor all go in the same direction. Also how a daily fluctuation can be 50 times larger than the annual increases which form the basis for the Global warming projections. Makes you wonder if the Global simulation programs might have forgotten or over/underestimated something small, or are missing a process we do not know about.

    I understand that the accepted explanation for CO2 cycling is primarily normal plant respiration, but on some warmer nights we only got 5% changes instead of 25% changes in CO2. Also you can see rain effects, and Francis says you can even see the afternoon rush hour commute. (pollution spike). Nice job Francis, thanks for the info.

    Open Question: do the ocean plants/algae do the same thing? Is there any evidence of a daily CO2 cycle over water. or does it all get absorbed & never get to the air?
    What does this daily cycling do to the residence time for CO2? Or is residence time really a computer simulation term that refers to the time it takes for a perturbation to return to the preexisting condition, and not the actual residence time of a molecule in a particular location?

    I assume that since this is presumably a cyclic phenomenon then it zeros out, as does the yearly ~5ppm CO2 cycle that is measured in Hawaii and is ignored in the annual averaged values used in the 200 year global warming computer models.
    Francis, what is the annual variation in your baseline CO2 levels? Is it also on the order of 5ppm summer to winter like Hawaii (which would have a larger ocean influence)

    Does this wide variation in CO2 impact the accepted CO2 in air measurements? Is the daily cycling accounted for? My first guess would be yes, since the Luxembourg data baseline at ~390ppm is comparable to the accepted global average of 380+.

  9. 59
    Dave Dardinger says:

    re Gavin’s response to comment #35

    Sorry I didn’t respond to this before, but it’s hard, when a thread gets so long, to notice that a response has been added to a message you posted 20 responses before.

    While I appreciate the problems with determing cloud feedbacks, your response is still rather a brush-off than an answer. It amounts to saying, nobody knows what the true answer is, but we’re sure it’s small. An actual answer would consist of explaining why it must be small, or if that’s too complicated, at least a link to a site which does go into details of the complications. BTW, this means something other than what you’ll find in the IPCC site. They list a lot of things, but not the details, which are needed to understand whether or not the findings pass the smell test. E.g. if you simply state that high level clouds give positive forcing, as you did, you need to say what the reason is. If it’s “They’re thin and let a lot of visible light through but trap almost all the outgoing IR” then you need to reconsider this in light of the increased absolute humidity, which would presumably make such clouds thicker and less able to let light through. I’m not saying that this is an actual proposed mechanism, but I hope it illustrates what I was hoping for in an answer.

    [Response: Sorry about that, but I do have a real job, and so responses here sometimes have to wait. As I stated above, cloud feedbacks are complex and I cannot do justice to them in a response here. A good review is available from the National Academies though. There is of course a connection between it being thought of as small and it also being uncertain - if it were a really large effect it would be more obvious in the data and we wouldn't be arguing about it. Water vapour feedback for instance is clearly seen in the data (Soden et al, 2000). -gavin]

  10. 60
    David H says:

    Despite its somewhat unprepossessing title this thread has been one of the more informative though the original question is rather semantic. Jeff Severinghaus, discussing ice core studies in December, said “In other words, CO2 does not initiate the warmings, but acts as an amplifier once they are underway. From model estimates, CO2 (along with other greenhouse gases CH4 and N2O) causes about half of the full glacial-to-interglacial warming.”

    So CO2 is also a feedback. For different reasons a warmer world, however caused, means more CO2 and more water vapour. What is hard to dispute is that mankind can and probably has artificially increased CO2 but can’t do much about water vapour. The question between proponents and those Gavin calls contrarians is whether the likely increase in CO2 will be a problem for us.

    This thread and everything I have read so far makes me think we are a long way from having sufficient certainty to warrant the actions being proposed. It is promising to see that John A’s comment was published and Gavin should realise that the price of freedom is not only eternal vigilance but having to put up with the tiresome business of responding to those who challenge his views. We do not need a scientific Taliban

    But if I might say so, contrarians do not have a monopoly on folly and obfuscation. The British House of Lords is conducting an enquiry into global warming and have heard arguments on both sides from creditable witnesses. Notable and directly relevant to this thread was the statement (according to the uncorrected draft transcript) from Kyoto proponent and one time head of the Confederation of British Industry, Adair Turner, who said “We know that that CO2 plays a fundamental role in the climate. Basically, if there was not CO2, we would be as cold as Mars or somewhere like that, and we would not have human life”

    Statements similarly economic with the actuality were made by the previous and current chairmen of the IPPC. Sir John Houghton describing 20th century warming said “Then, if you get to the middle of the century, you find the temperature rise stops somewhat!??”. When asked how much of 20th century warming took place in each half of the century Dr Pachauri said “The bulk of it took place in the second half of the century.”

  11. 61

    Some more comment on water feedback…

    I have the impression that the Pinatubo eruption is not a good (anti)surrogate for global warming caused by GHGs. The main effects of the eruption were in the lower stratosphere and the higher troposphere, while CO2 acts mainly in the lower troposphere. There were changes in ozone concentration, temperature (the change in temperature in the troposphere was larger measured by balloon data and satellites than at the surface) and in jet stream position (see: Rutgers).

    The change in SST during the peak Pinatubo event is even smaller than the global surface temperature drop (see: NOAA page 41), and not distinguishable from the (solar/ENSO induced) noise in SST trends. It seems that the change in water vapour in the total air column is not correlated to the sea surface temperature, the origin of most of the water vapour…

    At the other hand, sea air temperatures in the (sub)tropics have increased in the last decades (see: Chen, Carlson, Del Genio) with 0.085 K/decade. No matter which was the cause of the warming, this caused faster Hadley cell circulation, leading to a negative trend in clouds, upper troposphere humidity and more loss of heat (2.8 W/m2 net loss to space over the whole 30N-30S band). This could mean that GHG warming of the oceans, at least over the tropics, has a negative feedback from drying the upper troposphere (leading to less clouds), instead of a positive one from more water vapour…

    More interesting reading:

    [Response: There are no perfect surrogates. If there were we'd not be arguing about this! Each different kind of variability (seasonal, ENSO, volcanic. solar etc....) can be used to test different aspects of the model's responses and feedbacks. Pinatubo is useful for the water vapour feedback because it was a global cooling (as opposed to compensating warm/cool patterns) and there were clear measurements of the effect. There were reductions in tropical SST because of the eruption, but it is clearer if you look at the monthly values - the peak cooling of about 0.25 deg C in the 20 S-20 N band was towards the end of 1992. Surface air temperatures cooled more (0.5 deg C) because of the increased cooling over land (which does not have the heat capacity of the ocean). Pinatubo is in fact much more useful than the trends you mention because we know exactly what the driver was. - gavin]

  12. 62
    John Dodds says:

    Re: Why consider Water vapor… in the opening paragraph, and #29 Dave who has not heard the Water Vapor arguments:

    The “honorable” State of California in Code section 42801.1. defines Grennhouse Gases as :
    “42801.1. For purposes of this chapter, the following terms have the following meanings:…
    (h) “Greenhouse gases” include all of the following gases: carbon dioxide, methane, nitrous oxide, hydrofluorocarbons, perfluorocarbons, and sulfur hexafluoride.”


    Californians are prohibited by law from considering water vapor as a greenhouse gas, and this is the basis used to recently pass the Vehicle Emissions Regulations that mandated increased vehicle efficiency in order to reduce CO2 and global warming. AND is being implemented in several other states and Canada.

    I am not sure how the California University researchers get around this law when doing climate research.

    PS this is NOT a misplaced item from the Doubts about the advent of Spring, April 1st post.

    [Response: Since this is not from a science textbook, but from the law regulating greenhouse gas emissions, it makes sense that they do not include water vapour as one of the emissions they are interested in controlling. It would have been much worse to include it for these purposes. Remember context is everything. - gavin]

  13. 63
    Mike Doran says:

    58 wrote:

    Q: “Open Question: do the ocean plants/algae do the same thing? Is there any evidence of a daily CO2 cycle over water. or does it all get absorbed & never get to the air?”


  14. 64
    Dave Dardinger says:

    re response to #59

    Thanks for the link, I’ll see what I can learn there. But the very first sentence in the summary on the first page of the chapter rather undercuts your position: “Cloud feedback and its association with water vapor feedback and lapse rate feedback appear to be the largest contributors to uncertainty in climate sensitivity….”

    If cloud feedback is small and at the same time the largest contributer to uncertainty, then no wonder you warmers feel so cocky! But I think it needs more discussion.

  15. 65
    John Dodds says:

    Gavin, Re 62 response
    I can’t believe that you think it makes sense that greenhouse gases are legally defined to not include water vapor in the context of greenhouse gas emissions.
    What could be “much worse” than reality?
    What if the entire rule making and the subsequent regulations are thrown out because they illegally include the greenhouse gas effects of water vapor feedback in calculating effects and cost benefit analyses etc?
    The law does not believe in context- just what the law says.

    I can understand making a practical law that says do not monitor the greenhouse gas water vapor.

    Did you also support the Indiana Legislature when they defined Pi as 3.0 to make calculations easier?

  16. 66
    Brian Jackson says:

    Re #60

    Statements similarly economic with the actuality were made by the previous and current chairmen of the IPPC. Sir John Houghton describing 20th century warming said “Then, if you get to the middle of the century, you find the temperature rise stops somewhat!??”. When asked how much of 20th century warming took place in each half of the century Dr Pachauri said “The bulk of it took place in the second half of the century.”

    I’m not quite sure why you think they are being “economic with the actuality”, since their comments seem to be in line with the data.

  17. 67
    Peter Hearnden says:

    Re 65, well, John, we’re 66 post into a thread (and excellent initial post – thanks Gavin) and surely you’d need ‘selecto vision’ specs on not to know one of the crucial point is WV is a feedback not a forcing? Or maybe you can name some significant anthropogenic WV emissions for me – with estimated forcing?

    Your last comment is what I’d call tiresome, just what are you trying to imply?

  18. 68
    Michael Jankowski says:

    RE #62&65, there are also proposed legislations (and maybe even some passed ones) which classify CO2 as a “pollutant.” Obviously, this could also have serious ramifications if taken literally and applied to a number of different things.

    Note also the fine-print so prevalent in carefully-worded legal docs (my caps): ***FOR THE PURPOSES OF THIS CHAPTER, the following terms have the following meanings:…
    (h) “Greenhouse gases” include all of the following gases: carbon dioxide, methane, nitrous oxide, hydrofluorocarbons, perfluorocarbons, and sulfur hexafluoride.***

    Additionally, it doesn’t state that water vapor is NOT a greenhouse gas. It simply states that the term “greenhouse gases” includes CO2, CH4, etc. It doesn’t state that those are the ONLY greenhouse gases recognized by the State of Cal. And coupled with the context of “for the purposes of this chapter,” it is certainly not illegal for Californians to consider water vapor to be a greenhouse gas.

  19. 69
    Tom Huntington says:

    Open Question regarding the distinction between water vapor as a feedback versus a forcing:

    How do climatologists view atmospheric water vapor derived from evapotranspiration of irrigation water? By one estimate (below) global irrigation is about 3,500 cubic km of water per year on 2.5 million sq km of croplands. I realize that water vapor has a very short residence time, but this irrigation-induced increase in ET (a human alteration of the water cycle) is not something that is likely to go away any time soon, in fact, it is increasing, albeit at a decreasing rate. IF irrigation contributes to a short-term (but, effectively continuous) significant increase in water vapor, would this consitute a measurable forcing in its own right?

    [Response: This is a very good question. Personally, I would consider this a land surface change (similar to the chopping down of a forest to be replaced by cropland) that has an indirect effect on water vapour. Others have different opinions. These kinds of indirect forcings are however difficult to characterise on the standard 'forcings' bar chart, since they don't necessarily have a significant 'instantaneous' radiative forcing. One way to get around that is to use more complicated calculations for the forcings, for instance seeing how the radiation reacts to the change if you allow the whole column to adjust (while keeping surface temperatures constant), but this becomes more model dependent than the standard definition. -gavin]

  20. 70
    Dave Dardinger says:

    Follow-up to #64

    I’ve now read the chapter gavin suggested and have a couple of remarks and one quote.

    1. The chapter has the same flaw as the IPCC documents available in terms of discussing the subject under discussion. It’s a very high-level discussion, listing subject of interest and needing further research, but it only lists other papers to be looked at and doesn’t give any details. Anyone wanting to poke under the hood, as it were, is forced to find many articles and read them. This means it’s only of real use to a full-time researcher and not the layman (no matter how well trained scientifically). What I’m looking for is essentially a set of text-books which let a person look at actual data, equations and even simplified models if necessary. Not that anyone has time to look at everything, but it’d be useful if a person could dig as deep as desired in one limited area without having to spend great amounts of time and money to prepare for it.

    2. The chapter is well worth reading for anyone who thinks climate scientists have things well in hand when it comes to understanding clouds et. al. The chapter is practically an unending series of apologies for how much is not known. Nor does it state a la Gavin that these are minor details which won’t have much effect on our understanding of climate change if and when they’re addressed. (A lot of the problem involves needing many years of data to test models and observe natural variations).

    To illustrate, here’s one striking quote from the middle of page 26:

    “If the structure or area coverage of clouds change with the climate, they have the potential to provide a very large feedback and either greatly increase or decrease the response of the climate to human-caused forcing. At this time both the magnitude and sign of cloud feedback effects on the global mean response to human forcing are uncertain.”

  21. 71
    Joseph O'Sullivan says:

    The California law

    You have to read the entire chapter to understand the greenhouse gas clause.
    The scientists at Realclimate and scientists who comment at the site know that non-scientist sometimes get the science wrong. Lawyers and people in the regulatory field see the same things with laws and regulations.

    Reading laws and regulation is like reading a contract or a credit card statement (lawyers write these too). You have to read the entire thing, including the fine print, even though it is truly tedious. The chapter is titled California Climate Action Registry (Chapter 6, sections 42800-42870). The section cited by John Dodds gives the definitions of greenhouse gases which Michael Jankowski notes are for the purposes of the chapter. The next section is 42810, the Purposes of the California Climate Action Registry
    ( )
    To summarize, the purpose of the chapter is to encourage and record voluntary reductions of the greenhouse gases as defined in the earlier section.

    An environmental law takes science into consideration but also has to consider what is politically, technologically and economically feasible. This could account for the exclusion of water vapor in the definition.
    Science and the law is a messy area and like climate change has been increasing politicized. The Center for Progressive Regulation has a lot of info on this ( ), and in the interest of full disclosure, one of the legal scholar/attorneys in the Center was one of my law school professors.

    The law covering vehicle emissions is another law. The UCS has a good summary of this issue at:
    There are ongoing lawsuits about this issue and other efforts to curb greenhouse gases. Generally some states are proposing regulation trying to curb CO2 and other pollutants and they are being opposed by the EPA, other states and industry primarily on the grounds that the governing law, the Clean Air Act (CAA), does not include CO2 as a regulated pollutant or that only the federal government and not state governments can regulate these areas. It’s too early to tell how these suits will come out and how the efforts to reduce CO2 and other emissions will be effected

    The National Academy of Science has reviewed air pollution regulation (Air Quality Management in the United States (2004) online at: ), and stated “economic assessments of the overall costs and benefits of AQM in the United States indicate, despite uncertainties, that implementation of the CAA has had and will probably continue to have substantial net economic benefits.” The NAS also stated that a key challenge facing air quality management (AQM) will be “adapting the AQM system to a changing (and most likely warmer) climate”.

  22. 72
    John Dodds says:

    Re 68
    Michael & Gavin,
    My position is that I heartily support & approve of and agree with Gavins topic here on water vapor (see comment 15.)
    However I was pointing out to Dave #29,and supporting Gavin’s opening statement, that not everyone believes that water vapor is a greenhouse gas.
    The example being the State of California , which in Sec 42801 defines Greenhouse gases (for the purpose of the 42801 chapter) as to NOT include Water vapor. The 42801 chapter was originally written (I think in the 90s or maybe 2000) for the purpose of monitoring Greenhouse Gas Emissions – for which obviously monitoring water vapor does not make sense.
    HOWEVER, California also issued the law AB1493 in 2002. This is the law that required that the Vehicle emissions be evaluated to see if there was a greenhouse gas and global warming impact, and if so then they should be controlled – hence the basis for California to regulate CO2 auto emissions (as a pollutant) by mandating what the auto efficiency must be, when the US EPA claims that what they are really doing is controlling auto efficiency which the courts have ruled is a Federal mandate. Ab1493 invoked the 42801 definition of Greenhouse gases – ie Greenhouse gases do not include water vapor, for the purposes of the Law AB1493 (controlling vehicle emissions).

    To me this is stupid illogical and just plain wrong. It is as wrong as defining Pi to be 3.0. Regulating Greenhouse gases IS the Legislatures perrogative. Defining the reality that water vapor is not a greenhouse gas is not.
    If you are going to look at a subject and then force the Ca residents to spend billions of dollars to implement it, then you need to look at ALL of it. and honestly.
    Water vapor, as Gavin pointed out, is a greenhouse gas, AND terribly important as a feedback. It is also an anthropogenic emission (eg you eat hydrocarbons & breathout CO2 and water vapor, auto and airplane exhaust includes water vapor, electric power plants belch out water vapor and CO2 etc etc, BUT Natures feedback mechanism makes it ridiculous to try to control it.

    My statement was that if, by law, you do not consider water vapor a greenhouse gas, then how can you count the water vapor feedback as a greenhouse gas effect?

    My response in 65, was the total incredulity that Gavin in 62, could justify the scientificly invalid definition of greenhouse gases, when in the his opening paragraph he indicates that it is a greenhouse gas.

    Up until this point , I had thought that Gavin et al at RC were reputable, reasonable logical ethical scientists. But to admit reality and then ignore it in another context makes me wonder. Maybe Michael Crichton does have a point or two.

    John Dodds

  23. 73
    Tom Huntington says:

    Regarding Comment 69
    Gavin- Thank you for your response and for this very informative topic. If anyone knows of a bar chart similar to the forcings bar chart in the IPCC TAR WG1 Figure 6.6 (or updated in Figure 28 in Hansen et al. “Efficacy of Climate Forcings” submitted to JGR in January -
    that would include the various feedbacks (rather than forcings) please pass that information on.

    I think such a chart would be a useful summary in this discussion even if the levels of uncertainty are very high. I suggest this in the spirit of comparing apples to apples, as it appears myself and others in this thread are trying to do. As I understand it the major feedbacks are (1) water vapor in both troposphere and stratosphere, (2) changes in the albedo, of snow, ice, and sea ice, and (3) changes in the properties of clouds.

  24. 74
    Eli Rabett says:

    I think John Dobbs is completely missing Gavin Schmidt’s point. The rapid interchange of water vapor and water through evaporation and condensation means that on the time scale of decades and centuries, water vapor concentrations are responses to other forcings rather than being drivers.

    In that context, what the California legislature did was perfectly logical They listed greenhouse gas emissions which were to be studied and perhaps controlled if they were shown to have a potential impact on global warming. You might try and fault it for not distinguishing between greenhouse gases and greenhouse gases that were to be evaluated and perhaps controlled, but that would be splitting hairs. To use John Dodd’s example AB 1493 required evaluation of vehicle emissions to see if there was a greenhouse gas and global warming impact. Water vapor concentrations cannot be controlled by limiting vehicle emissions or increasing vehicle emissions (you could plant or cut down a whole lot of trees….???). This is obvious even to members of a state legislature, so one wonders what the point of bringing this up was. Kind of the kind of point Michael Crichton makes as far as I can see.

  25. 75
    Dave Dardinger says:

    I hate to rain on your parade, John, but I think reason for the distinction between H2O and CO2 is quite clear. CO2 sticks around a good while while H2O rains out quickly. Therefore how much H20 is released is meaningless unless it is both constantly being released and the amount released is an appreciable % of the total H2O in the atmosphere. Autos and power plant emissions of H2O don’t pass this test. I’ve worked through the math before and it’s tiny.

    This is not to say that I share the warmer’s fear of increasing CO2, but the debate should be primarily aimed at the things which matter, not trivialities and semantics.

  26. 76

    I am not that sure that water is not a primary driver for temperatures, at least on land, as result of irrigation. See the work of Christy and Norris at:

    The summer minimum temperatures in the Californian Central Valley increased with 1-2 K since the mid-seventies, while stations at the bording foothills (200-1000 m high) nor mountain stations (>1000 m) show such a trend. Maximum temperatures in the valley show a small cooling trend. This all may be due to irrigation.

    It seems that some stations surrounded by irrigated land need corrections for this phenomenon, as good as is the case for the UHI effect of growing towns…

  27. 77
    David H says:

    Re #66
    The choice of the phrase ‘economic with the actuality’ was to avoid suggesting either of the IPCC men deliberately misled. Both the questions and the answers have to understood in the context of the assertion that increasing CO2 will cause increasing temperature. If you take the mid century year of 1950 the trend is definitely down not up on both the graphs to which you link. If you take the trend from 1939 to 1975 it is also a decrease. Describing it as “the temperature rise stops somewhat” is just spin. CO2 rose steadily but temperature most definitely did not. However, for the counter argument, through that period the solar cycle lengthened. The correlation with solar cycle length as shown at is far more convincing than the CO2 theory. Apart from that mid century period the solar cycle has steadily shortened from 1890, which is the coldest period shown on your graphs, to the present. The solar cycle length has never again been as long as it was in 1890 and I don’t need science to tell me we now have a more active sun than I have seen in my lifetime. Just think about skin cancer.

    My quarrel with Dr Pachauri is the word “bulk”. The rises in temperature either side of the mid century point on your graphs are similar and I do not think an objective scientist would use the word bulk. This, however, is not true of the rise in CO2 in the same period.

    The quarrel sceptics and contrarians have is that every piece of data is spun to point in the direction of CO2 to such an extent that even adroit operators like Adair Turner come to think and pronounce publicly that CO2 is only thing that determines how warm our planet is. Almost daily until the start of our election campaign we have been told that CO2 is the “principle greenhouse gas”. Significantly, global warming is not in the electors top 10 interests and no major party is proposing anything serious. Anyone want to predict the UK Green vote?

  28. 78
    Joseph O'Sullivan says:

    John Dodds does bring up a very good point and is absolutely right that the best and most comprehensive science should be the basis for laws and regulations, especially expensive ones. However there are many other things that influence the law making process. In addition to science issues there are economic, political, technical issues and even timing (i.e. we want to pass a law now, but the science is not fully developed). There are also legal concepts like the role of state vs. federal government and property rights. Environmental regulation involves many complex issues and makes a lot of work for lawyers. In a naked bit of self-interest, I think it is a great system that improves my employment prospects.

    In the current issue, AB 1493 (its at ) as a matter of law declares that greenhouse gases would have negative effects on California and should be regulated. The California legislature instructed state agencies to set greenhouse gas emissions for autos, examining the technology available and economic effects. The definition of greenhouse gas excluded water vapor for, and only for, auto emissions.
    The issue should be if water vapor emissions from autos should be regulated, not if water vapor should be considered a greenhouse gas to determine possible climate effects.

  29. 79
    Andrew Watkins says:

    Re: #34


    Im not sure where the reduction in rainfall in SW Western Australia came into this, but i think people are a little off the track with timing. The greatest reductions are in the May-July period (See:, laregly driven by the late arrival of the Autumn rains. You can see this trend in the simple trend maps for Autumn at the BoM site, e.g:

    I think the BoM web site does a fine job in showing this, but agree it would be nice to be able to see any period you wanted.

    The IOCI reports suggest that based upon climate modelling and analysis of past synoptic conditions, the most likely cause for this change in precip is a global warming induced shift in the atmospheric circulation patterns. Land clearing has also been considered, however it cant realistically explain the broad scale changes in the circulation at this stage.

    Whichever way you look at it, SW W.A has a disasterous drying trend.

  30. 80
    Michael Jankowski says:

    Re#79, I will try to keep this brief as it probably is a little off-topic.

    With regard to your first link (IOCI report), I have a MAJOR problem with how they presented/interpretted the data. There is certainly an overall downward trend from 1925-2003 in Fig 2. When you have a downward trend, the mean at a later period within that timeframe is generally going to be lower than the mean at the beginning of the period. As an oversimplistic example, draw a downward trendline, and split the line into two segments (anywhere along the line). The mean across the first segment will be higher than the mean across the second segment, even though there was no change in trend. This is similar to what is presented in Fig 2 of the link, comparing the 1925-1975 mean vs 1976-2003 mean. This does not itself imply that there was some sort of drastic change in 1975/1976. Even in the case where you have a steeply declining trendline for the early period and then a flatter trend for the later period (even possibly going up!), you can have the same situation where the mean for the later period is much lower than that of the early period. Anyhow, I find it absurd to draw and/or present the conclusions they did using such methodology. I’m curious if Gavin or one of the more “expert” people feels as strongly as I do about this. I also think this method of comparing means for two time segments heavily skews the conclusions on page 2.

    Page 2 does refer to “a global change in the atmospheric circulation” for 1975/1976, so that does support why they selected those two timeframes. However, from eyeballing Fig 2, you could get similar results stopping the first period in the early or late 60s, the early 50s, the mid 40s, etc, so it doesn’t seem like the 1976-2003 timeframe really is exceptional in the context of the 20th century. Maybe there are more statistical elements behind the scenes that would show such a thing, and maybe my eyeballing is way-off, but I think this note ignores analysis which could be legitimate and relevant and instead uses oversimplistic methodology which is flat-out insufficient, quite misleading, and could produce incorrect conclusions.

    With regard to your 2nd link, the timeseries charts for “western Australia” don’t seem to show a declining trend for winter rainfall since 1975, and the autumn rainfall trend seems to be consistent with pre-1975 autumns and actually going upwards ( Annual rainfall is also on the increase. Granted, these are not specifically for only the “south west of Australia,” so it’s possible decreases in the south west of Australia are compensated for by increases in rain elsewhere in western Australia. So on to the trend maps…

    In looking at the 1960-2004 trend map you linked to, there clearly is a decline in autumn rainfall in the south west of Australia, up to 15-20mm/decade. I don’t dispute that rainfall in autumn (or winter) in the south west of Australia has been decreasing over time. But what I found was that the 1970-2004 trend map shows a decline of only up to 10-15mm/decade, which indicates that the decreasing rainfall trend in autumn has actually slowed in the south west of Australia post-1970. Isn’t that a good thing? The rainfall is still decreasing, yes, but that rate of decrease slowed…yet your first link (IOCI report) implied everything was pretty much fine until 1975/1976 and then fell apart. When you look at the other maps, you’ll find even more information to dispute the position of the IOCI report. The 1950-2004 map looks almost identical to the 1970-2004 map in the south west. Ditto for 1940-2004. Using only the trend maps as a source of info, it appears that autumn rainfall in the southwest of Australia declined from 1900-2004. However, this rate of decline peaked in the 1960s, and the decline rate of 1970-2004 is quite similar to that of 1940-2004 and not much worse in highest magnitude than that of 1920-2004. So it’s tough to see from those links the basis for arguing that there’s been a drastic change in the 1970s.

    I must maintain the caveat that not being able to select specific start-and-end points with the trend maps can produce some misleading conclusions. But I think they (along with the IOCI data) show enough to suggest that the decline in autumn rainfall in the south west of Australia over the past 3 decades is more of a continued trend over much of the century than a sudden change.

  31. 81
    Jeff Norman says:

    All this talk about evaporation, water vapour and precipitation and not one mention of enthalpy, heat of vapourization and heat transfer.

    Evaporated water rises up in the atmosphere where it achieves a heat balance with the molecules that make up the surrounding air. If there is an imbalance the water will change phase from the vapour state to the liquid state releasing the heat of vapourization which energizes the air molecules, increasing their temperature.

    Warmer air “retains” more water vapour because it is cannot absorb the incremental heat energy from the incremental water vapour.

    According to the CRU, the Earth’s surface temperature has increased. According to the discussion above this means more moisture should evaporated at the Earth’s surface. This vapour has to go somewhere. It doesn’t appear to be accumulating in the atmosphere so it must be precipitating out and falling back to Earth, leaving the heat energy in the atmosphere.

    Where is all this heat energy going? According to the MSU data (even the RSS interpretation) it is not staying in the atmosphere. Does this mean the atmosphere is better at shedding heat into space than the models suggest?

  32. 82

    answer to #58:
    here are the yearly averages I measured the last 3 years on the same location:

    year /annual___/JanFeb___/JunJul___/delta
    year /mean_____/mean_____/mean____(summer-winter)
    2002 398.0___381.7___384.1___2.4
    2003 402.0___404.5___394.2___-10.3___heatwave summer
    2004 407.5___393.1___401.4___8.3

    I used the MIR9000 sensor from Environnement SA (a French company).
    I have a problem with the 2001 data (the sensor was down during a hefty 10% of the time, so the yearly mean is questionable). The years 2002 to 2004 (sensor uptime nearly 100%) show an increase of about 4 ppm/year, which seems more or less double the Mauna Loa means). I would have preferred a negative delta (summer – winter) for all years, but the data don’t respect this wishful thinking…

  33. 83
    Dan Allan says:

    Debunking contrarians sure gets tiresome. You never seem to get anywhere. I don’t really get this crisis over a California law that says that, for the purposes of regulation, WV should not be considered a GHG. What is the big conspiracy theory? Are they proposing that there is not enough regulation, and we should begn immediately to regulate WV emissions? Do hey think scientists a purposely surpressing critical data that WV really is a GHG? I don’t get it.

    Regarding post 77, how many times does one have to repeat: Nobody has ever said that CO2 is the only forcing, so nobody ever expected a temperature chart to follow a trendline of CO2 concentration in the atmosphere without incorporating other forcings. In fact, if historical temperatures tracked in perfect parallel to CO2 in the atmosphere, that would be a serious PROBLEM for most GCMs. What is incontrovertible is that when you add the CO2 forcing you wind up with a temperature trendline that is closer to historical reality than without it.

    I would also like to know which part of the CO2 / Global warming argument is being disputed:

    1) CO2 is accumulating in the atmosphere due to human activity – this is basically proven fact. I know of nobody who seriously disputes it.
    2) CO2 is a greenhouse gas that will cause the atmosphere to warm as its concentration increases. Does anyone seriously question this?? This seems to be pretty basic science.

    There is plenty of room for debate about positive and negative feedbacks, how much or how little the atmosphere is likely to warm in the future, etc. But it would be nice if wasn’t necessary to continually go back to square one to persuade those who see a giant cabal that there is no grand conspiracy.

  34. 84
    Dave Dardinger says:

    But what good does it do to try talking about feedbacks when I get no responses worth discussing? Gavin claims the feedback from clouds are known to be small and then cites a link which basically contradicts what he just claimed.

    So what’s the next subject up for discussion? Clouds? Of course not. It’s Ozone. Like skeptics are going to get fired up about that subject?

  35. 85
    Dan Dorritie says:

    This is not a comment; it’s a question (or more).

    In discussing the global warming potential (GWP) of various greenhouse gases (GHGs), climate scientists frequently employ compilations of “integrated time horizons.” (Example, the 1990 IPCC report, “Climate Change”, table 2.8, p. 60.) GWP compares the warming potential of other GHGs to that of carbon dioxide, but they are compared according to time horizons of 20, 100, and 500 years. (I do recognize that these particular time horizons have been used because they are the most useful, and that, depending on one’s purpose, other time horizons could also be used.) Let’s take methane. Methane (including its indirect effects, which are irrelevant to my question) has a GWP of 63 for a time horizon of 20 years. Methane also has an atmospheric residence time of less than 10 years. So what does the 63 represent? Does it mean that if we start with equal amounts of carbon dioxide and methane, the amount of warming produced by that methane will be 63 times that of carbon dioxide, despite the fact that it will all be gone in less than half (under 10 years) of that 20 year time horizon (whereas much of the carbon dioxide input will still be around)? Does the time horizon therefore reflect both the intrinsic balance of solar radiation input, thermal radiation output, as well as the length of time that each gas will remain in the atmosphere?

    And, practically speaking, why choose to examine GWPs on the basis of one time horizon versus another? Should we therefore worry about the immediate warming effects of methane (because they are “big”) but not so much about the longer-term effects (because they’re less)?
    Despite some investment of personal time on attempting to answer this question, I’ve not located any adequate discussion of the matter. . . and article after article presumes that that the answer is either self-evident or common-sensical. It’s not.

    Many thanks,

    [Response: Lifetimes are more complex that a simple number might indicate. There is some discussion at A lifetime is more an e-folding time that a time for complete disappearence, so I think that saying there would be no methane left after 10 years is wrong (disclaimer: not my area of expertise). The exact defn is given in the TAR: The resason for looking at different time horizons is that the "total" warming effect is different for long and short lived gases. The "instantaneous" effect of a certain amount of methane is more that the same amount of CO2; but if you integrate the warming over some time horizon - perhaps 100 years - then the balance shifts. This is useful is you need to balance methane/CO2 emissions and their effects - William]

    [Response: With respect to methane in particular, there are a number of issues. Firstly, the 'residence time' (total atmospheric content divided by total sources (or sinks) ) for methane is around 8.4 years. However, because of a non-linearity in the CH4-OH chemistry, the time scale for a perturbation to decay away is around 10 years. This is the time for a perturbation to decay to 1/e of it's initial value, and so after 20 years, there will still be around 13% of the initial pulse. Thus the integrated forcing due to methane needs to take the molecule per molecule greater impact into account, and also the changes in abundance. - gavin]

  36. 86

    reply to #83:
    my mixed sympathy to Dan Allen who gets upset debunking contrarians. BUT: when you read media on GW, you (almost) never hear anything else than CO2 bashing (even here in Kyoto friendly Europe, the 5 other gases are almost never spoken upon: nearly all political forced strategies go into reducing CO2, THAT BIG AND ONLY CULPRIT!). No wonder that the fine points of climate science get lost, and that someone who does no agree to this very biased view rightly claims his opposition!

  37. 87
    Stephen Berg says:

    Francis, if you looked at this link, you would see why CO2 is the primary target by policymakers in Europe:

    It is the most prominent greenhouse gas that is causing the human-induced climate change problem.

  38. 88
    Eli Rabett says:

    I hate (OK, I love) to make more work for others, but there is an important article in the Journal of Chemical Physics 122 (11) 114309 (2005) by Cormier,m Hodges and Drummond on measuring the water vapor continuum absorption. It shows (or makes a real good attempt at showing) that the mid IR continuum absorption for water vapor is a couple of orders of magnitude lower that a commonly used empirical model (CKD 2 4). So, my question is how would that affect Gavin Schmidt’s table?

    This looks like a really good experiment.

    [Response: Indeed. This is a bit more complicated than just putting in a new number and testing it though, but I will consult further and report back at some point! - gavin]

    [Update: After some discussion, the consensus appears to be that the uncertainties in the continuum are on the order of a few W/m2 of absorbtion (this is about 2% in the numbers in the table). The data (even including this new paper) are pretty diverse and hard to measure. Thus independent replication of any new result is necessary before the codes will be changed. The GISS radiation scheme uses a theroretical basis (Ma-Tipping) for the continuum absorption and this is at least as good as any other scheme out there compared to the data. However, this is a definite source of significant uncertainty, although it's impact on the sensitivity or water vapour feedback is less important. -gavin]

  39. 89
    John Dodds says:

    Re Forcings chart (#87?)
    It is precisely because of the unquestioning acceptance of the “Forcings Chart” that Gavin had to write this Water Vapor topic in the first place.
    It is a chart of apparently arbitrarily (but logically) defined forcings in order to make the computer models work.
    It is NOT a chart of the the causes of global warming, but many people seem to use it that way.
    There is also something called feedback, (see above) which is apparently primarily water vapor which will result in more warming as a result of increasing temperature from all of the factors in the chart. There is also a feedback greenhouse effect from existing CO2 responding to increased solar energy etc which doesn’t seem to get mentioned often. etc etc.
    In fact if you make a similar “Causes of Warming chart” in W/m2 then the water vapor feedback dwarfs the CO2 column. Hence as Gavin pointed out, when you get 3 sceptics together, they always bring up water vapor as being neglected.
    The forcings chart is IMHO unfortunately biased towards making CO2 the biggest reddest box on the chart. It is highly misleading.
    A more fair Causes chart should show water vapor and all feedbacks, and then indicate that there is no practival way to control water vapor, just like there is no practical way to control the sun, and no practical way to control the natural part of the ~380ppm of CO2.

    I have several questions about the Forcing chart, which I hope Gavin can address when/if he does Forcings vs Feedback as a discussion topic. (Good luck in explaining that one!!- hope you do as good a job as you did for water vapor!)
    1. It seems to me that a forcing is defined as an external change to the system. eg the sun adds heat, man adds CO2 to the air, etc. My basic question is what is external? what is the system?
    eg1 The sun causes all of the greenhouse effects due to its energy input and trapped reflected energy. Are all of the GH gas “forcings” therefore really feedbacks? Why not?
    eg2 Manmade CO2 is produced from burning fossil fuels previously caused by the sun which are already, or should be, included in the Earth total energy balance, just like CO2 is included in limestone and corals and dissolved in water, and (negative or cold) energy is stored in ice and glaciers etc Are we therefore defining the “system” to be only the atmosphere, so forcings are then external changes to the atmosphere? In which case why is melting ice not a forcing? (my answer – because it is already accounted for in the computer model & you don’t want to double count it)
    A New thought just struck me, is “the system” the computer model and not the earth or air ecosystem? If this is so, then a chart of forcings (without feedbacks) makes no sense whatsoever, since it is so arbitrary and depends on how the programmer accounted for things.

    2, Does the CO2 in the Forcings chart include the natural feedback of the CO2 increases due to solar warming? – ie the sun warms, ice melts, more land/sea is available for plants to produce CO2, which is the normal mechanism that has resulted in all of the previous warmings & CO2 increases from the previous ice ages?
    I assume that the CO2 column does NOT include the water vapor feedback from increased temp due the CO2 forcing. – does it?
    Does the CO2 forcings column include the CO2 feedback from increased greenhouse effect (from the 378 ppm of old CO2) from the increased and varying solar warming?

    Now considering these questions, I begin to wonder, just what good is the current forcings chart if it does NOT include all the relevant feedback effects?

    [Response: You ask some very good questions, and in truth the answers are sometimes a little fuzzy. In a zeroth-order practical sense, you can define as a forcing as a change in anything that isn't calculated prognostically in your system. So for instance, O3 can be a forcing in a simple model, but in a model that calculated atmospheric chemistry, it would be a feedback (while the forcing would be the emission of precursor molecules). In common parlance, forcings are assumed to apply to the basic dynamical GCM system (i.e. that includes winds, temperature changes, water etc.) but not including the bio-geo-chemistry or interactive vegetation that many of the newer Earth System models come with.
    You also need to remember what forcing 'bar charts' were designed to do. The idea was to have a metric to compare different effects and be able to assess how the climate will react when given combinations of the forcings without having to do the calcuations every time. Since water vapour (and ice albedo, and clouds and long-wave radiation and evaporative cooling....) will change as a result of all the forcings, it doesn't make sense to associated a water vapour feedback say only with CO2. If one makes the assumption that each forcing really is equal (which it isn't really), then including the water vapour term would simply multiply all the lines by whatever the gain is. It is a cleaner calculation not to include that for any of them.
    Now with respect to any of the constituents that have an anthropogenic component and a climate related feedback (i.e. CO2, CH4, etc.) it is a little difficult to seperate the two. The standard method is to take the observed concentrations and simply use that (which assumes (correctly in most cases) that the natural feedback term is small). Alternatively, you use the emissions data and the bio-geo-chemical models (which can calculate the natural feedback term) and re-apportion the forcings (which must add up to the same number as in the first case) between the different emissions. I was a co-author on a recent paper that tried to do this for methane (Shindell et al, 2005) which actually demonstrates that methane is actually undervalued on the standard chart! (but that is compensated for by a negative forcing from NOx).
    You may be right in thinking that a more expansive discussion of this is required..... - gavin]

  40. 90
    graham dungworth says:

    The original work done on heat transport by water vapour in the atmosphere is by Tyndall.

    Tyndall, J., 1859, Note on the transmission of heat through gaseous bodies, Proceedings of the Royal Society, London, 10, 155�158.

    Quote “acts more energetically upon the terrestrial rays than upon the solar rays; hence, its tendency is to preserve to the earth a portion of heat which would otherwise be radiated into space”.Tyndall’s work predates Arrhenius’work by almost two generations and he estimated in his 1859 paper that water vapour absorbed 16000 fold more infrared heat than dry air. Despite its low residence time in the atmosphere he regarded it as the most important “GHG”.

    The 88% also comes from a book reference
    Peixoto, J. P. and A.H. Oort, 1992, Physics of Climate. American Institute of Physics, 520pp.

    Perhaps the 98% ref was a ” back of an envelope calculation” combination of this 16000 fold estimate and the ratio of respective residence times!

  41. 91
  42. 92
    Andrew Boada says:

    The extent of my knowledge of climate science comes from the 101 course I took and the text book I read, so my depth of knowledge isn’t very great. However, I was wondering: couldn’t it be said that water vapour isn’t a forcing agent simply because the amount of water vapour in the atmosphere is dependent on the temperature of the air, not vice versa? Also (from my anecdotal observations) humid air seems to require more energy to raise its temperature (is this correct?) If true, then even though humid air would cool more slowly at night, it would also heat more slowly during the day. Water vapour seems just to stabilize air teperatures.

  43. 93
    Mike Doran says:


    [Response:Well obviously I think so! I will do a summary of the results soon... - gavin]

  44. 94
    John Dodds says:

    Gavin, a follow-up to #89 about the Forcings Bar chart, in the water vapor feedback, section of, (I am using the 1750-2000 chart @ ):
    I can see why a chart of the “forcings” (or causes) should not include the feedback “effects” of water vapor. BUT I think you need to be able to address ALL the effects somehow if “The idea was to have a metric to compare different effects and be able to assess how the climate will react when given combinations of the forcings without having to do the calcuations every time.”

    Now I have a question: Is the 1.5W/m2 attributed to CO2, strictly from the forcing part of the CO2 (ie the addition between 1750 and 2000 – or nominally the ~90ppm difference between ~290ppm of CO2 and the current 380ppm)? OR does it include the change in W/m2 due to the increased CO2 “feedback” as a result of the positive Solar irradiance forcing acting on the preexisting ~290ppm of CO2 also? ie Is the 1.5W/m2 strictly due to Anthropogenic causes or is there a piece of it (290/380=75%)due to the solar forcing on natural CO2 also? (because if the natural part is included then the use of this box to guesstimate effect from anthropogenic (which #87 is trying to do) over estimates the effect). Same question applies to all the greenhouse gases. And the other forcings?
    Just what are the magnitudes of the natural GHG feedbacks from the solar forcing? (so that I can accurately determine how much of the temperature rise (or total W/m2) was due to the solar forcing and how much was due to the anthropogenic forcing.)

    [Response: The 1.5 is purely from measured CO2 changes. I don't understand why your think there should be solar influence mixed in it - W]

    Then questions on the Sun box in the Forcings bar chart.
    The solar irradiance actually oscillates over time. Is the 0.3W/m2 a net value difference between 1750 and 2000?
    Would the number be significantly different if you had gone from solar trough to peak (~1820 to 1998)? ie is the choice of time interval a hidden consideration in this chart so that you can’t use this history snapshot to project the solar part (& its feedbacks) into the future?
    If the chart had gone for example from 1942 to 1970, when the net sun W/m2 would have been negative, during cooling, what would the CO2 box(es) have been – positive or negative?

    [Response: Firstly, remember its not *measured* solar - its reconstructed from proxies with more or less certainty. I presume its the avg over the 11y cycle. From 42-70 the CO2 forcing was +ve, of course (you have remembered sulphate aerosol, haven't you?) - W]

    Similar questions apply to the Forcings vs Time chart. .
    ie Does the light green CO2/GHG line include the result of solar forcing of the natural CO2 component? or is that an effect that is also missing like the water vapor feedback, and needs to be added to the solar forcing in order to get the total solar + feedback effects. Does this then result in a more intuitively logical conclusion that the solar forcing with feedbacks dominates the Anthropogenic CO2 forcing effects, just like the sun dictates shorter and longer (daily, yearly, little ice age, and presumably ice age) temperature cycles?

    I think it would clarify the whole picture if you created “Forcings And Feedbacks” Bar (and Time) charts, with a water vapor box above or below each of the other CO2, CH4 , sun, clouds etc components. tying the WV box to each component indicates its dependence on that component (ie its a feedback from that component, not a forcing) BUT it allows the user/public to add all the values up to get the total W/m2 impact from all the other forcings and feedbacks. Also each of the GHGs would be broken out to show anthropogenic or natural and the natural components should be associated with their forcing, which is the sun. (My guesstimate of this chart would show the solar forcing to be responsible for 75-80% of all the 1750-2000 W/m2 effects.)
    Right now the forcings bar and time charts give a very incomplete picture- lots of pieces are missing or not expressly broken out, so the public can not accurately see the relative magnitude of the components. I do not see how you could use them to accurately assess the future impacts

    Finally a very subtle but important question: The Forcings bar chart, or the proposed Forcings and Feedbacks charts are a historical explanation of what caused what etc. However a chart trying to project the future, does not care if the existing GHGs were anthropogenic or natural. Any estimate of the future impact depends on the existing conditions (eg CO2 @380ppm) and what the future forcings will be. ie If the sun cools off next year (a negative forcing), but CO2 increases 2ppm (similar to what it did in 1999 and 2000), what will be the dominant “forcing/cause” and the effect on temperature? Won’t the negative sun forcing on all 380ppm of CO2 dictate what the temperature does since it far outweighs (380/382=99.5%) the 2ppm CO2 forcing increase? – as it did in 1999. Are we not then at the mercy of the sun, and the only thing that anthropogenic CO2 etc does is magnify (by GHG feedback) whatever the sun is forcing? Therefore, a future 2x rise in CO2 over 50 years has the potential to change the temperature BUT only as a multiplier/feedback on the magnitude, and direction, dictated by the orbitally modified solar irradiance forcing. ie Raising the CO2 levels for the next 50 years ALSO raises the potential that the temperature will go a few degrees lower WHEN the next solar cooling happens. Increasing GHG levels has only widened the band for the temperature range by a fraction of one or two degrees, similar to the way that the Earth precession tilt and eccentricity (see Zachos 2001 fig 1C, Oceans/GES206/papers/Zachos2001.pdf – or google: Zachos 2001), have narrowed the range by several to ten degrees over the last few thousand years.

    [Response:This all seems a bit strange to me - you're implying a dominant role to the sun, when the evidence seems to be that solar changes are rather small. If CO2 rises, that doesn't imply (as you seem to believe) that a subsequent reduction of solar would cause a greater cooling than if CO2 was lower - W]

    Doesn’t this then dictate that the majority of the results of any future temp projection are dictated by the estimate of what the sun will do in the future? (Pay attention Drs Lean and Solanski etc!) And by what the unknown and not evaluated longer term solar irradiance cycles, and orbital and Milankovitch multiplying factors and their associated feedbacks, do to the solar forcing? (A Strange concept that the GHG increases result in major feedbacks to the solar + Milankovitch factors such that they may really explain all of the ice age cycles in spite of being so small!)

    It can’t be this simple. Have I missed something?
    John Dodds

    PS If all the above is valid, then have I succeeded in unifying the valid (with uncertainties) IPCC Science Data ( but not the political position) with the skeptics who do not believe that CO2 is the major cause of global warming ? If so, then as an unaffiliated, unemployed amateur scientist (actually an engineer) I will gratefully accept all accolades rewards & remunerances. :) – BIG smile! :).

  45. 95

    [...] effect. This then leads us to the conclusion that excludes water vapor as an option because it is a feedback, not a forcing: While water vapour is indeed the most important greenhouse gas, the issue that [...]

  46. 96

    [...] absorbance of between 9 to 26%. We’ve already covered the importance and relevance of water vapour before but of course, you never actually learn from these articles even though you claim to be someone [...]

  47. 97

    [...] those interested how climate scientists deal with water vapor, I recommend the RealClimate website. It explains very clearly that water vapor is a “feedback” rather than a [...]

  48. 98

    [...] Original von Gavin Schmidt (NASA Goddard Institute for Space Studies in New York) bei RealClimate [...]

  49. 99

    [...] link goes into pretty good detail, and also helps raise your understanding on the issue quickly: RealClimate Water vapour: feedback or forcing? Sorry to hear about your hand. Maybe you can learn to type with your toes? All the best. [...]

  50. 100

    Re: Clinton and Obama: Failures on War and Global Warming…

    " Water? "

    Water vapor is a feedback mechanism in climate change..
    If you……

Switch to our mobile site