If you read many articles and comments in the blogosphere you would think that “skeptics” have discovered something hidden. Or highlighted an important truth that climate science is trying to hide.
Water vapor is actually the dominant “greenhouse” gas
This is true.
If only climate science actually realized it and stopped pretending that CO2 was the most important “greenhouse” gas..
If Only They Wrote it Larger..
For terrestrial radiation, water vapor is the most important single constituent of the lower atmosphere, although carbon dioxide is always significant..
Atmospheric Radiation: Theoretical Basis, Goody & Yung, Oxford University Press (1989, 2nd edition)
Water vapor is the most important atmospheric greenhouse gas.. Carbon dioxide is the second most important greenhouse gas..
Radiation and Climate, Vardavas & Taylor, Oxford University Press (2007)
Generally speaking, water vapor is the single most important atmospheric absorber in the IR band..
No other atmospheric constituent is better known to the general public as a “greenhouse gas” than CO2. In actuality, water vapor has a larger overall impact on the radiative energy budget of the atmosphere..
A First Course in Atmospheric Radiation, Grant Petty, Sundog Publishing (2006)
Water vapor is the most important gas for the transfer of radiation in the atmosphere..
Global Physical Climatology, Hartmann, Academic Press (1994)
Table 6 shows the relative contributions of H2O, CO2 and O3 to reducing the outgoing longwave flux, from which it is seen that the longwave effect of H2O is significantly larger than the effects of CO2 and O3..
Climate Modeling through Radiative-Convective Models, Ramanathan & Coakley, Reviews of Geophysics and Space Physics (1978)
The importance of water vapor in regulating climate is undisputed. It is the dominant greenhouse gas, trapping more of Earth’s heat than any other gaseous constituent..
The Radiative Signature of Upper Tropospheric Moistening, Soden, Jackson, Ramaswamy, Schwarzkopf & Huang, Science (2005)
The dominant role of water vapor as a greenhouse gas has long been noted..
The Importance and Nature of the Water Vapor Budget in Nature and Models, Lindzen, Climate Sensitivity to Radiative Perturbations: Physical Mechanisms and Their Validation (1996)
The authors find that for the clear sky case the contribution due to water vapor to the total longwave radiative forcing is 75 W/m², while for carbon dioxide it is 32 W/m²..
Earth’s Annual Global MeanEnergy Budget, Kiehl & Trenberth, Bulletin of the American Meteorological Society (1997)
Water vapor is the dominant greenhouse gas, the most important gaseous source of infrared opacity in the atmosphere..
Water Vapor Feedback and Global Warming, Held & Soden, Annual Review Energy Environment (2000)
In fact, it’s so well-known that most times in papers it isn’t repeated. No one involved in atmospheric physics is confused about the subject.
Why the focus on CO2 in that case?
Water vapor arguably lies at the heart of all key terrestrial atmospheric processes. Humidity is essential for the development of disturbed weather, influences (directly and indirectly through cloud formation) the planetary radiative balance, and influences surface fluxes and soil moisture. Water vapor is the only radiatively important atmospheric constituent that is sufficiently short‐lived and abundant in the atmosphere so as to be essentially under purely natural control..
Tropospheric Water Vapor, Convection & Climate, Sherwood, Roca, Weckwerth & Andronova, Review of Geophysics (2010)
Emphasis added.
The point is that water vapor responds to climate – and therefore influences climate as a feedback. The concern is that humans adding CO2 to the atmosphere will cause a change to the climate and water vapor will have a feedback effect (see, for example, Clouds and Water Vapor – Part One and subsequent articles).
It is quite difficult for humans to add water vapor to the atmosphere. The oceans are a vast source of water, and just above the surface of the ocean the atmosphere is saturated
Why isn’t there more focus on water vapor?
There is a huge focus on water vapor because it is a difficult subject. CO2 is well-mixed in the atmosphere so the application of radiative transfer theory to CO2 is not in debate (in scientific circles).
The Science of Doom 
I’m signed up with SpectralCalc for a month. Let me know if you need any spectra calculated.
After watching a couple of the movies at this site where CO2 is mostly a NH phenomenon and is to my eye I not well mixed.
http://www.esrl.noaa.gov/gmd/ccgg/carbontracker/weather_movie.html
Could you clarify the ‘well mixed’ claim for CO2. Maybe my interpretation of well mixed is off. Stir some sugar in a glass of tea take the spoon out. Now while the mixture is still swirling around the sugar while dissolved is generally thought to be well mixed, that is evenly distributed through out the tea. I understand it might not be evenly distributed as one might initially think but certainly more so than the CO2 in those movies.
Stephen, well mixed means that the variation is negligible.
Note the values for each color represented. At the moment the lowest (purple) is at 375 ppm, and the highest (dark red) is at 390 ppm. And the far majority of the map shows between light blue and yellow, which is a difference of about 5 ppm. That’s about 1% difference for the majority of the atmosphere.
I would say that is negligible for our purposes.
Stephen Singer has a bit of a point:
On the surface of the Earth, CO2 can vary in a single day at a single location from 250 to 600 ppmv (Phoenix, AZ is a good example where urban effects create severe variability in boundary layer CO2 levels. Boreal forests can vary from less than 300ppm to day to over 450 ppm at night.) Once you get above the surface layer, the variability is smaller, but there is a clear annual variability, and the variability increases as you go polewards:
See e.g., here.
This web-site discusses the question of variability in more depth as well as the origins of the increase in CO2 over time.
One thing that is ubiquitous of the measurements is, once you go above the atmospheric boundary layer, the daily variations in CO2 are negligible (See here,) and the farther you are away from CO2 sources or sinks, the smaller the daily variability (See here.)
So a reframing of Singer’s question is “In the context of radiative/atmospheric physics, does the variability in the surface layer matter?”
The thing that seems to be missed by a huge number of lay commenter’s is that Earth is the right temperature for water to easily change state.
I don’t see much solid or liquid CO2 around 🙂
I don’t think anyone would like living here if they did occur naturally on Earth. To me those difference are the key points about why one is the controller.
Well-mixed is a relative term, of course. CO2 is globally less well-mixed than say, N2, but much, much more well-mixed than water vapor, aerosols, SO2, volatile organic carbons, CO, etc.
“Water vapor is the only radiatively important atmospheric constituent that is sufficiently short‐lived and abundant in the atmosphere so as to be essentially under purely natural control”
What about clouds? They are certainly short-lived and abundant. Do clouds count as a state for water vapor”? If so, indirect aerosol effects on clouds are estimated by the IPCC to range from -0.3 to -1.8 W/m2 (at the high end negating CO2 +1.7 W/m2). Do aerosols “dry out” the atmosphere by inducing rain or simply change the nature of clouds?
Water vapor short lived? Tell that to the folks who live in the southern U.S. If you are talking a specific molecule well yes but not water vapor in toto. CO2 lasts what 5 yrs.
As for not making any water vapor we make water vapor all the time in burning fuels. ex. CH4 + O2 -> 2H2O + CO2 We make more molecules of water than CO2.
Frank,
The aerosol indirect effect is a kludge used by modelers to make their models fit the twentieth century global temperature series regardless of the climate sensitivity of the particular model. There was a post at Pielke.Sr.’s some time ago that I can’t find again that recommended that the aerosol indirect effect shouldn’t be included in models at all as its existence at all is questionable.
The other thing is that the aerosol indirect effect is mostly a local phenomenon, not global.
Carrick,
In terms of radiative forcing, changes in the CO2 concentration in the boundary layer has little effect on forcing at the surface. The increase in DLR is closely matched by a decrease in DSR from absorption by CO2 in the NIR at ~4 μm.
“The point is that water vapor responds to climate – and therefore influences climate as a feedback. The concern is that humans adding CO2 to the atmosphere will cause a change to the climate and water vapor will have a feedback effect.” The specific concern, of course, is that the the warming caused by CO2 will increase the amount of water vapor in the air and thereby amplify CO2 warming. In fact, positive water vapor feedback now seems to be accepted pretty much as a “given”.
Yet climate models show no sign of it. An example is GISS Model E (http://data.giss.nasa.gov/modelE/transient/climsim_table1.pdf). According to GISS a forcing of 1.75 w/sq m between 1880 and 2003 causes only 0.53C of warming in Model E. This gives a climate sensitivity of 1.1C, which is about what we would expect from CO2 warming in the absence of any feedbacks. Other models seem to give the same results.
(Ah, I found the reply button this time.)
Roger, those figures are from the transient simulation runs.
They are used to figure the short term variation from a change in forcing. If the CO2 concentration was frozen at 2003 levels, there would have been still more warming since there would still be a radiative imbalance. This means the equilibrium sensitivity would be higher than that is shown in the transient simulations.
Refried:
Yes, but water vapor feedback is a transient response, and the model runs for 120 years.
But the sensitivity is not a transient parameter.
Like I said, if the CO2 (and really any of the non-feedback forcings) were fixed at 2003 levels and the simulation were to continue to run, there would still have been more warming since it had not reached equilibrium yet.
This means that the transient model only measures transient sensitivity, not the equilibrium sensitivity.
DeWitt, regarding CO2 I recognize this of course… I just thought it was an interesting issue to raise. Here’s a good reference:
Balling, GRL 28, 4599 (2001). “Does the urban CO2 dome of Phoenix, Arizona contribute to its heat island?”
Summary from CO2Science:
Their final conclusion, therefore, is that the warming induced by the urban CO2 dome of Phoenix is possibly two orders of magnitude smaller than that produced by other sources of the city’s urban heat island (lower soil moisture levels and enhanced absorption of solar energy by urban surface materials, for example). Hence, although the presence of man and his alteration of the local environment are indeed responsible for high urban air temperatures (which can sometimes rise as much as 10°C above those of surrounding rural areas) and high urban near-surface atmospheric CO2 concentrations (which can sometimes top 600 ppm), the high urban near-surface air temperatures are not the result of a local CO2-enhanced greenhouse effect.
The effect on plant growth was also interesting, if a bit at odds with some other studies I’ve seen:
With respect to its influence on vegetation, the urban CO2 dome likely enhances the robustness of urban vegetation, given the well-documented fact that atmospheric CO2 enrichment tends to enhance plant growth rates and increase the efficiencies with which plants utilize water to produce organic matter. In addition, elevated urban CO2 concentrations should reduce the deleterious effects of airborne pollutants on plant health (Allen, 1990) by reducing the apertures of the stomatal openings by which pollutants gain entry into plant leaves (Pallas, 1965; Kimball and Idso, 1983). Within an urban CO2 dome, therefore, some of the positive effects of one of the major end products (CO2) of urban combustion processes tend to counteract one of the negative effects of some of the minor by-products (air pollutants) of those same processes.
And this is the same place that Idso [Atmospheric Environment 36, 1655 (2002) ] reports CO2 measurements of over 1000 ppmv.
While CO2 and H2O are emissive gasses there’s a big difference in how they force. CO2 emits to space from the stratosphere while H2O emits to space from the troposphere. Further, increases in CO2 mean at the margins of the CO2 band, energy will emit to space from a higher level than it would otherwise. Increases in H2O, on the other hand, will continue to emit from the troposphere. So CO2 forcing appears certain ( since it takes place above the constraining troposphere ) while water vapor feedback appears constrained ( since it is subject the dynamic motions of the troposphere ). Paradoxically, since the cooling of the upper troposphere is due to H2O, there is a tendency for subsidence driven by the presence of H2O. But that subsidence also dries the “cap” and makes the base of the “cap” more emissive to space ( since there is less H2O above the base of the subsidence inversion ). Thus more H2O leads to less H2O – a negative feedback to the water vapor feedback.
SOD says:
“CO2 is well-mixed in the atmosphere… ”
That is just flat out wrong! A high pressure cell has more regional mass and less humidity than does a low pressure cell. Humidity alters the conc of CO2 in real air.
As I said in my last post:
If a volume of air contains clouds, you have no way of knowing the MASS of CO2 in the gas phase. A slight change in pressure or temperaure wiil affect the amount of CO2 as well as water vapor added to or removed from the gas phase.
Go check a recent satellite image of the earth. This image shows there is no uniform distribution or shapes of clouds in the atmosphere. Also the density of water droplets in clouds is quite variable. For example, fluffy cotton-ball cumulus vs menacing cumulonimbus.
For CO2, how big (in terms of %) and how extensive are the natural variations? Are these variations important compared to the GLOBAL 100% increase in CO2 that seems inevitable without massive intervention? If not, isn’t it reasonable to call CO2 well-mixed?
Water vapor, clouds, and aerosols are a very different matter.
“Water vapor is the only radiatively important atmospheric constituent that is sufficiently short‐lived and abundant in the atmosphere so as to be essentially under purely natural control..”
I’ve never liked this idea. Its not lifespans in the atmosphere of individual H2O molecules that matters, its the average global humidity and that can change in the longer term through changes in the ocean conditions.
It all comes down to this fundamental “AGW” belief that the idea of internal forcing is irrelevent or even non-existent.
Tim: I may have a intuitive understanding of the concept of natural variability in a chaotic system, but I have no idea of what “internal forcing” means. Any explanations or references?
“I have no idea of what “internal forcing” means. Any explanations or references?”
In a general sense, an internal forcing is simply internal variability that applies over a long enough timeframe to push the earth’s climate away from equilibrium.
A completely hypothetical illustrative example of this might be that 1000 years ago when the NH was warmer (during the MWP) the water temperature that sank in the Arctic and began making its way back to the tropics in the deep ocean was a little warmer. Now we’re seeing it emerge and that warmth has effected the climate and raised global temperatures.
That would make it a “1000 year” effect.
I want to make it perfectly clear that I’m not say this IS happening but I also want to make it perfectly clear that these kinds of effect IMO could be happening and we haven’t found them yet.
RE: Positive Water Vapor Feedback Hypothesis (PWVFBH)
The PWVFBH goes like this: Emission of GHG’s by humans causes a slight global warming. This causes a slight increase in the evaporation of water. Since water is the main GHG, the air heats up causing more water to evaporate which cause more global warming. And round and round this goes.
What wrong with the PWVFBH? The wind is main mechanism the transports water vapor into the atmosphere. In a strong wind, nitrogen (FW=28) and oxygen (FW=32) molecules and argon atoms (AW=39) are like sand blasters that just blow water molecules (FW=18) right out of the water into air even when the water is cold such as occurs during a nor’easter.
To see the force of the wind, get a bowl water and blow on the surface using a straw. Note the depression due to the force of the wind. Then set a glass of crushed ice about ca 6″ down wind, blow on water and note the condensation of water vapor on the cold glass. Now stop blowing on the water, wipe off the glass, put it back at the same spot and note the time it takes for water to condense in still air. It takes a lot longer.
Air pressure changes influences water evaporation. A aneroid barometer shows for various pressures ( inches Hg): >30, fair; 29-30, change; <29, rain.
Suppose air is at 14 deg C, 50% RH and one atm pressure. If the temperature increases by 1 deg, the additional water vapor for a cubic meter is ca 0.4g or 8 drops from a eye dropper.
Harold, As I recall, the net effect of water evaporation is cooling. This occurs below the Knudsen layer. Above this layer, sunlight, not IR radiative effects, warms the evaporative layer of water. So, while the air is warmer, the water is cooler. That’s why chillers work so well. Doesn’t this eventually neutralize the warm air effect you describe and will not lead to more evaporation?
[…] and the limitation to absolute humidity at a given temperature for saturated air. Science of Doom covers this rather well. Pointing out that water vapor is Earth’s dominant greenhouse gas does not […]
It’s worth keeping in mind that the varying distribution of the radiative forcing over the globe by CO2 is not so much due to differences in atmospheric concentration (say between a polluted site and a pristine site like Mauna Loa), but rather differences in the temperature profile between different regions, or variations in overlapping opacity with other gases.
It’s often forgotten that the infrared opacity is just one part of the story for the greenhouse effect. The vertical temperature structure matters too, and so differences in the tropopause height between the tropics and the poles, or inversions over Antarctica, etc lead to differences in the top of the atmosphere forcing. The traditional metrics of radiative forcing (defined as a reduction in the OLR at the TOA (or tropopause)) tends to be maximized in the tropics. The distribution of water vapor over the globe also leads to heterogenity in the applied downward radiative forcing at the surface by excess CO2, as in Fig. 2 here
http://www.springerlink.com/content/6677gr5lx8421105/fulltext.pdf . When you have lots of water vapor (as in the tropical boundary layer), you would exhibit a minimum in direct increases in the downward longwave flux to the surface.
It’s also worth keeping in mind that when comparing CO2 vs. water vapor, it’s not just because water vapor is a “feedback” that makes CO2 a competitive gas. After all, the atmosphere always has moisture in it, regardless of whether an individual parcel of air is going through the cycle of condensation. The reason CO2 can still be competitive is because it has some absorption features where water vapor is not very strong. See fig. 2 here http://geosci.uchicago.edu/~rtp1/papers/PhysTodayRT2011.pdf
The structure of the absorption co-efficient vs. wavelength here limits some of the competition between the gases, but keep in mind again that in the high atmosphere where it is cold and dry, CO2 has an important role since gases in the high atmosphere have a profound impact on the OLR. This is also why small changes in the water vapor at high levels under global warming dominate the water vapor feedback. It’s certainly possible to have a very warm and moist situation, especially in cases where the stratosphere becomes quite wet, in which case CO2 becomes effectively negligible (this is what happens during a runaway greenhouse process), but Earth is quite far from any such regime.
As far as what greenhouse gas is “most important”, this of course depends on what you mean by important. WV is of course “most important” if you ask what gas contributes most to the infrared opacity in the modern atmosphere. But as Pierrehumbert et al (2007), Voigt and Marotzke (2009), Lacis et al (2010), and others have shown, CO2 keeps the atmosphere warm enough to have a substantial water vapor effect in the first place. If you could take all the water vapor out of the atmosphere (and keep it out somehow) you could trigger a snowball Earth, but the same is true with CO2, because you make it a lot easier for the outgoing radiation to escape but also collapse much of the water vapor greenhouse as it becomes colder, in addition to increasing the surface albedo. And because CO2 is what principally governs the ability for the greenhouse effect to change on climate timescales, one could just as well argue that it is the “most important.”
Great comments, Chris. Thanks for the input.
Harold,
your argument on the water vapor feedback falls apart right at the start, by assuming the increase in evaporation is what causes a positive water vapor feedback. In fact the water vapor inventory can increases independently of evaporation changes, and actually does so at a much faster rate in global warming simulations. This fallacy is still a source of confusion even sometimes in the literature.
…I discussed this all further in my RC post by the way
http://www.realclimate.org/index.php/archives/2010/11/more-on-feedbacks/
refriedscience:
“Like I said, if the CO2 (and really any of the non-feedback forcings) were fixed at 2003 levels and the simulation were to continue to run, there would still have been more warming since it had not reached equilibrium yet.”
You miss my point. What I’m saying is this. GISS Model E shows temperatures increasing in response to increasing CO2 since 1880, and there has apparently been an increase in atmospheric water vapor content in response to the temperature increase. This means we should see a positive water vapor feedback. But we don’t. Model E returns a 1.1C zero-feedback climate sensitivity all the way from start to finish. It has to, because if the sensitivity were any higher it wouldn’t match observations.
Roger,
you’re misunderstanding the table you’re looking at. In fact all models have a water vapor feedback , see e.g., Table 1
Click to access bjs0601.pdf
The GISS model from the Efficacy 2005 paper has a climate sensitivity of about 0.75 degrees C per W/m2 forcing, in line with IPCC estimates of climate sensitivity. Please look at the paper again
Click to access 2005_Hansen_etal_2.pdf
By the way, CO2 does not double from 1880-2003. That’s where the no-feedback 1 degree response and 2-4.5 degree C (with feedback, at equilibrium) response comes from.
If I have the math right, the actual multiplier is around 0.45 or so currently (log(380/280)/log(2)).
So given the uncertainty in climate sensitivity (while remembering that some of the warming from previous GHG emissions is still unrealized), this would suggest a total range of warming from anthropogenic forcing mid-1800s to now of (1.5-4.0) x 0.45 = 0.7-1.8 °C.
Things get a lot more interesting if we hit 750 ppmv, as DeWitt’s modeling suggests we might:
log(750/280)/log(2) = 1.4 so (1.5-4.0) x 1.4 = 2.1-5.6°C.
Chris:
Thank you for your comments. However, I still have a problem. You say the GISS model has a climate sensitivity of about 0.75 degrees C per W/m2 forcing. Effective forcings in the model are 1.75 W/m2. I multiply 0.75 by 1.75 and get 1.3C. The model shows only 0.53C. What am I missing?
Roger,
You can’t get anything about the equilibrium climate response from the particular table you’re looking at. The GISS efficacy site has a lot more plots which I would explore (I don’t have all of their information memorized) but the Hansen et al 2005 paper clearly shows the equilibrium climate response we’re talking about. If you think about it the warming we have actually seen so far over the century is about 0.8 C, much in line with models, but keep in mind we don’t actually know what the total forcing is (because of aerosols).
Also, look at Table 4.1 which shows some different model responses in transient and equilibrium states
Click to access sap3-1-final-ch4.pdf
as well as section 4.2.2 on water vapor feedbacks in the models.
DeWitt Payne,
Let me turn back to a comment of you made earlier and elsewhere (http://judithcurry.com/2010/12/21/radiative-transfer-discussion-thread/#comment-24168).
You said there:
“Clouds absorb 100% of upward LW radiation. There is no window for clouds, no forward scattering as there is for visible light. The 40 W/m2 in K&T and KT&F is the 99 W/m2 clear sky emission reduced by 60% because only 40 % of the surface has clear sky. There’s an additional 30 W/m2 from colder than the surface cloud tops directly to space. I think that’s too low, but it might be caused by multiple layers of clouds.”
Sorry but as far as I can see your flux numbers do not match with observations. TOA clear sky emission in the window region, according to CERES instruments, is 79 W/m2, while its all-sky amount is 65 W/m2.
Calculating the “lost” radiation, KT97 and TFK2009 give 40 and not 40+30 W/m2. Their atmospheric LW absorption is surface minus window (396–40), not surface minus window minus cloud top.
Further you said:
“If you treat the cloud tops as part of the surface than the average temperature of the planet must go down because cloud tops are a lot colder than the surface. Tau at the cloud tops is also a lot lower so the average tau for the planet goes down as well. Tau from the surface for cloud covered sky is effectively infinite. No matter how you slice it, an average τ of 1.8 for the planet can only be constructed by making fallacious assumptions like ignoring clouds”
The measured global average all-sky TOA window flux, related to the global average surface flux, 65/396, gives you tau = 1.8 .
I was asked in an email, why I brought this here, into the Water vapor vs CO2 greenhouse gas topic.
Well, just because tau=1.87 is a GHG-invariant constant.
Why do you have to bring your nonsensical jibberish about this here? The Miskolczi nonsense has been thoroughly debunked and completely discounted by anyone serious in the scientific community.
Chris:
I appreciate that short-term model runs don’t tell us much about equilibrium climate sensitivity (although a 123-year run should contain at least some equilibrium response). But that isn’t the issue. What I’m concerned with is water vapor feedback amplification, which is a transient effect, and my question is why GISS E shows no sign of it, or at least none that I can find.
I’m also intrigued by your comment that “we don’t actually know what the total forcing is”. But existing estimates (IPCC/Hansen, Crowley, Lean, Model E) are all in the same range, and GISS (http://data.giss.nasa.gov/modelforce/RadF.txt), goes do far as to give estimates for ten different categories of forcing down to the nearest 0.0001 W/m2. I’ve always assumed that these estimates were at least somewhere close to correct. Was I wrong?
Roger,
No, we really don’t know what the radiative forcing is with high confidence. See this figure from the IPCC AR4 report
You can see the large spread in the estimated forcing, most of which is due to aerosols. This is one reason most people don’t like using the 20th century to evaluate climate sensitivity, instead turning to paleoclimate evidence, modeling approaches, etc. Knutti and Hegerl 2008 provide a review.
I’m not sure I can make much sense of your objections to the water vapor feedback. Of course it’s acting now. There’s evidence the water vapor inventory went down after the Pinatubo cooling too. Without feedbacks, you don’t explain the modern warming, but the uncertainty in ocean heat uptake and the total forcing is such that we can’t narrow the total sensitivity with much constraint, as in the top part (“Instrumental Period”) of Fig. 3 here
Click to access knutti08natgeo.pdf
Scienceofdoom:
“The point is that water vapor responds to climate – and therefore influences climate as a feedback. The concern is that humans adding CO2 to the atmosphere will cause a change to the climate and water vapor will have a feedback effect.”
“Water vapor is the only radiatively important atmospheric constituent that is sufficiently short‐lived and abundant in the atmosphere so as to be essentially under purely natural control”.
Three points:
1. mkelly’s chemical reaction showing man making water vapour at twice the rate he makes carbon dioxide.
2. What has the stock-life of a radiatively active molecule to do with its capacity to absorb and emit radiation and directly affect forcing?
3. Professor Trenberth, co-author of the global mean energy budget, the principal instrument of the AGW hypothesis, lamented in a leaked email the inability of science to explain where all the heat has gone since 2003. His underlying analysis summarises human influences on the climate system and the system’s responses to them. Trenberth assesses the direct effect of accumulating CO2 in the atmosphere as equivalent to 0.7% of the energy coursing through the climate system. Burning fossil fuels produces other heating effects as well as cooling ones which cancel out. Thus, according to Trenberth, the net human effect is about the same as the CO2 effect alone.
He also finds that the climate system’s net response, or feedback, reduces the human effect to 0.4% of total energy in the system. That is, the negative radiative feedback exceeds the sum of the two positive, non-radiative ones, water vapour and ice-albedo.
The strengths of both the radiative and the water vapour feedbacks are determined by temperature. Actual observed temperature, after all emissions and feedbacks, determines the negative radiative feedback. The positive water vapour feedback, on the other hand, is computed from a notional, higher temperature that is consistent with total direct warming effects exclusive of offsetting direct cooling ones. It seems likely, therefore, that the overall net effect of burning fossil fuels is smaller than the 0.4% of energy coursing through the climate system estimated by Professor Trenberth. This would help explain how the current cooling wrong-footed the models.
John:
1. Well, if we consider that there is about 25 times more water vapor in the atmosphere than CO2, then we have a much less of a proportional effect on water vapor compared to CO2.
2. It has to do with the rate it leaves the atmosphere; the shorter residence time, the faster it leaves. Therefore our small contribution to the water vapor content is almost completely overpowered by the lower lifetime, which means that the variation in water vapor is essentially under the control of natural processes.
3. Trenberth’s “lack of warming” was referring to not being able to account for short term inter-annual variability. And Trenberth’s 0.4% net response is after taking the TOA radiative imbalance of 3.6 W/m^2 (1.5%) and subtracting 2.8 W/m^2 (1.1%) cause by an increase in surface radiation from a 0.75C rise in temperature. This give us our current radiative imbalance of 0.9 W/m^2 (0.4%). The 0.9 W/m^2 is our current net radiative imbalance, not the overall effect of fossil fuels.
Source: Trenberth, “An imperative for climate change planning: tracking Earth’s global energy.” Current Opinion in Environmental Sustainability 2009.
“the TOA radiative imbalance of 3.6 W/m^2 (1.5%)”
Let me rephrase that:
“the TOA radiative imbalance of 3.6 W/m^2 (1.5%) (a combination of GH gasses minus aerosol forcing plus water vapor and ice-albedo feedbacks)”
chriscolose on February 27, 2011 at 5:27 am
Thank you for your kind answer. May I ask you to provide us your calculations of the window radiation, compared to KT97, TFK2009, and CERES?
chriscolose on February 27, 2011 at 5:27 am:
Please respect the Etiquette.
There is scope for unkind remarks:
But usually after a small effort in explaining the flaws in the allegedly bad science.
Scienceofdoom:
This whole issue is not as complicated as it seems. If you want to know how water vapor responds to climate, or how water vapor and climate responds to CO2, you have to have numbers about the infrared absorption of the atmosphere.
My first point is that NASA CERES measurements do not verify the 40 W/m2 atmospheric window radiation of TFK2009. Their all-sky value is 65 W/m2.
The second point is that this change will alter also the atmospheric LW absorption of that figure, resulting 396-65=331 W/m2.
The third point is: these quantities can be calculated by radiative transfer (RT) methods on several individual observed vertical atmospheric profiles, and also on their global average. Hence you get a chance to compare your calculations to measurements. The first and only accurate line-by-line RT computation of the Atmospheric Window, the atmospheric LW absorption, and DLR, as far as I know, confirms the values given by the CERES measurements, and not of TFK2009.
The fourth step, analyzing decadal trends in several reanalyses, may confirm or reject your actual annual values.
And the fifth one is, when you can put some relationships behind your calculated flux results.
I do not want to go further here, speculating on the physical background and interpretation of those relationships, but that will be necessary if one really wants to understand how the whole energy distributive system works.
My last sentence above, making chriscolose so snappish, that tau=1.87 is a GHG-invariant constant, is only the terminus of this journey.
If you or anyone is interested, I would suggest starting at the beginnings: meticulous dissection of the data, measured and calculated.
Miklos,
The all sky window contains emission from cloud tops as well as from the surface. Radiation from the surface does not penetrate clouds at all. That’s the source of K&T’s 40W/m2 window. The clear sky window is ~100W/m2. Cloud cover is ~60%. Therefore radiation from the surface through the window is 40W/m2. Cloud tops emit too. Most of that emission is absorbed by the atmosphere and not radiated directly to space. ~30 W/m2 is radiated directly to space from cloud tops. 40 + 30 is approximately equal to 65 W/m2.
Thank you DeWitt Payne for your answer.
If one looks at the CERES measurements data, one can find that the clear sky window emission is ~ 79W/m2. So, if calculated on your way it would result only ~ 32 W/m2.
~ 100 W/m2 is the clear-sky radiation of the USST76 atmosphere (with appendix B of Liou 92), which contains only about the half water vapor as the realistic global average.
The same is true for your cloud-top calculation. KT97 says that in the cloudy case 99 W/m2 decreases to 80 W/m2. With an average cloud layer at ~ 1.8 km, with a cloud-top temperature 276 K, one will have 80 W/m2 on the USST76 atmosphere in the 8-12 um region. Applying 60 % cloud cover, that would result 48 W/m2, not 30.
So, with your method, you should have 40+48 W/m2 allsky, very far from the realistic 65 and also from the KT97-TFK2009 values.
Further, when you want to know the greenhouse effect, you must have a precise knowledge of the atmosphere’s longwave absorption. That depends directly on the ‘lost’ (transmitted, unattenuated) emission, not on the window+clouds radiation.
Sorry, I still do not know what is your (or the ‘official’) suggestion for the atmospheric LW GHG absorption, 396-40, 396-32, 396-(40+30), 396-(40+48), or how much.
At the moment, IPCC AR4 WG1 Chapter 1 Fig 1 (=KT97) and TFK2009 is tought on the universities in hundreds of courses to thousands of students, and it seems to be in contradiction with observations.
DeWitt Payne,
Let me add:
According to your last sentence, you seem to accept the CERES 65 W/m2 as the realistic all-sky (clear + average cloudy) TOA window flux.
Now you can relate it to the corresponding 396 W/m2 surface radiation, to get the global average all-sky transmittance:
Ta = “Atmospheric Window” / “Surface Radiation” = 0.1641.
With this, you will get by definition
tau = – ln Ta = – ln (65/396) = 1.8 .
This seems to contradict to your earlier point where you have said:
“an average τ of 1.8 for the planet can only be constructed by making fallacious assumptions like ignoring clouds”
You’re still comparing apples and oranges and ignoring clouds. The surface radiates an average of 396 W/m2. But 60% of the surface is obscured by clouds. No surface radiation escapes directly to space. That means tau is infinite. For low level clouds and a surface temperature of 288 K, the cloud top temperature is about 265 K and lower for higher level clouds. But let’s say that it’s all lower level clouds. 265 K is ~280 W/m2. So if we average together the 60% at 280 and the 40% at 396 we get 326 W/m2. -ln(65/326) = 1.6 maximum, less if there are higher level clouds, and there are. And it’s still a meaningless number that has no general applicability.
Going back to KT97, the cloud model is 49% low level clouds, 6% mid-level and 20% high level. In addition, the high level clouds have an emissivity of 0.6 rather than 1. At 220 K, the total emission from high level clouds is 80 W/m2 and emission in the 8-12 μm window is 12.6 W/m2. Twenty percent of that is 2.5 W/m2. For the mid-level clouds, the temperature is ~255 K, emission in the window is 51 W/m2. ~20% of the mid level clouds will be obscured by high level clouds so 51 * 0.048 = 2.4 W/m2. Adding up the areas we have 20% high cloud, 4.8% mid-level cloud, 38% clear sky leaving 37.2% low cloud. If we take 278 as the cloud top temperature that gives 82.5 W/m2 in the window times 0.372 = 30.7 W/m2. For a surface temperature of 288.2 and an emissivity of 0.98, 97.3 W/m2 emission in the window times 0.38 = 37 W/m2. Summing I get 72.6 W/m2 in the window which is in the ballpark of the CERES 65 W/m2. Those high level clouds are going to significantly reduce the average radiated power in the tau calculation. For worst case you have 396 * 0.38 + 339*0.372+79.7*0.20+240*.048 = 304 W/m2. That give tau = 1.5 restricting the definition of tau to the window.
But of course that’s wrong. The higher you go, the wider the window and the more radiation that’s actually transmitted directly to space. So essentially all the radiation from the high clouds will escape to space. That raises the value of the numerator and makes tau even smaller. It’s a meaningless number and it certainly isn’t a constant.
Refriedscience:
Macroscopic stock turnover rates in the atmosphere of its radiatively-active constituents can have no effect, that I can see, on the very high frequency microscopic interactions between the energy field and the gaseous matter which together make up the atmosphere and give rise to the greenhouse effect.
My reading of the paper you reference goes as follows. Trenberth, applying the same methodology, was not able to achieve in the period 2004 to mid-2008 what he had achieved for the period 1993-2003 – an accounting of incremental energy source and application. He couldn’t account for 20% to 70% of “observed” human-sourced energy (Table 1). Why the inverted commas? I suspect they mean “inferred from observed emissions”. He blames inadequate means to measure where the energy goes – the notorious travesty. The apocryphal alien visiting the issue might easily ask why the travesty could not equally encompass the theory.
Trenberth’s theoretical accounting goes (Fig 4; units in Wm-2):
Net anthropological forcing 1.6
Add/subtract natural responses:
Subtract radiative response 2.8
Add water vapour/ice albedo response 2.1
Net imbalance at TOA 0.9
The net natural response to anthropological forcing constitutes negative feedback, a unit of forcing resulting in a net 0.56 imbalance at TOA . Of these four numbers only the last one is a measure. The second one is a calculation based on a measured temperature, after all responses. The third one is based on a notional temperature, related to the net anthropological forcing, before any responses. The before-response temperature would exceed the after-response one. But, given contemporaneous natural responses this difference is artificial. Accordingly, the natural positive feedback and the net imbalance are overstated and the climate is less affected by anthropological forcing than the theory posits.
This would help explain the models’ failure to pick the current cooling. As to explaining the different outcomes between the two periods examined by Trenberth, logic suggests an alignment of natural and human influences in the earlier period and their divergence in the later one.
DeWitt Payne:
If you are interested in how much vitamin was eaten in fruits, you can well compare apples and oranges. If you are interested in the greenhouse effect, you must compute how much IR radiation is being “eaten” by the atmosphere, cloudy and free, and how much was let to escape into space. One should be careful with the averaging, as vast part of the 33 K greenhouse temperature comes from the clear sky effect (~28 K) and only the remaining 5 K from the clouds. So the “clear-sky tau” might be much more dominating as it seems.
But you are right in saying that there are other types or physical forms of the tau with much lower average values then 1.8, playing central role in the dynamics of the whole system. While the individual air column tau-s around the world scatter widely, they show a specific structure, and there is a particular tau about 1.4 that, in a certain relationship, assures the global average allsky 1.87.
I think radiative transfer and optical depth are good tools to analyze the IR absorption characteristics of the atmosphere. “Meaningless”, is not a solution. Whether it is kept constant by the system or not, it should be decided on measurements.
What really counts for the surface temperature is back radiation. It depends strictly on LW atmospheric absorption. For DLR, TFK2009 gives 333 W/m2 as the average of three reanalyses. A publication in JGR two years ago gave 337 as an average of hundreds of surface station observations. Tau=1.87, with 396 W/m2 surface radiation, gives 335 W/m2. The physical meaning of that tau cannot be so wrong.
But at that level of tau, you don’t really constrain the surface temperature. If I take a surface temperature of 289.1 for 396 W/m2 emission, then tau for 66 W/m2 through the window is 1.791 and Teff for the atmosphere is 276.2 K. If I raise the surface temperature by 5 degrees to 294.1, then tau only needs to be 1.86 to have 66 W/m2 through the window and Teff is 281.9 or 5.7 degrees higher. Those seem like reasonable numbers too and well within the experimental error of the radiosonde measurements.
Have you read the IPCC AR4 Summary for policymakers?
Where does it say that water vapour is the dominant greenhouse gas?
Oh dear, it doesn’t. Instead it makes the deliberately misleading remark:
“Carbon dioxide is the most important anthropogenic greenhouse gas”
Paul, I fail to see why that is a misleading remark. I would suggest it is a very clear and accurate remark. The fact is, Co2 is the most important anthropogenic greenhouse gas.
So it is, but the IPCC report is about climate change isn’t it? Then one would have expected that to give our policy makers a more complete understanding of the issues it would be natural to brief them as to what the climate forcings are. It would have taken one sentence and a diagram to give a simple overview of the climate system.
“If you read many articles and comments in the blogosphere you would think that “skeptics” have discovered something hidden. Or highlighted an important truth that climate science is trying to hide.
Water vapor is actually the dominant “greenhouse” gas”
I’m not sure where you’re writing this from, but here on planet Earth it is, and has been, well known for a very long time that water vapour is the dominant greenhouse gas. Do you have any reason for assuming that sceptics didn’t know this and have only just found this out?
geronimo:
I think you misunderstood what I wrote.
It is often claimed that climate science doesn’t know this, or is trying to hide it. I’m just trying to make it clear that it’s very basic fundamental knowledge in climate science, it’s in the textbooks and everyone knows it.
[…] as possibly causing a much higher than 1°C increase is water vapor feedback. Water vapor is the dominant “greenhouse” gas, but is variable in space and time as it responds to climate […]
[…] and the limitation to absolute humidity at a given temperature for saturated air. Science of Doomcovers this rather well. Pointing out that water vapor is Earth’s dominant greenhouse gas does not […]
[…] vapor is the most important atmospheric “greenhouse” gas. But its effect on the surface radiation and TOA radiation are quite […]
Hi, could you tell me what the radiative efficiency of water vapor is? That is, the radiative forcing given a change in concentration. For instance, the radiative efficiency of CO2 is 1.4 x 10^(–5) (W/m^2) / ppb.
http://www.ipcc.ch/publications_and_data/ar4/wg1/en/tssts-2-5.html
To some extent, and probably a large extent, this depends on the absolute concentration because it is ultimately the slope of the line. I fully recognize that detail, so if I could get a simple graph of radiative forcing per concentration, that would be even better (either way, I don’t care). So, a graph of (W/m^2) on the y-axis and the ppm or ppm concentration, or partial pressure, or number density of water vapor on the x-axis. Yes, I know that the concentration changes with temperature. That doesn’t prevent the existence of this data.
This entire discussion seems like a non-starter, because it is predicated entirely on the fact that water vapor gives positive radiative forcing… but I can’t find numbers for this anywhere.
The most recent posting linked above your comment gives the answer:
Such a single number is not meaningful for water vapor. That’s the case because water vapor cannot increase uniformly throughout the whole atmosphere as the non-condensing GHG’s do. Therefore the answer depends heavily on the way the increase is distributed.
The posting considers two alternatives where the addition occurs either near surface in the Boundary Layer or in the rest of the atmosphere. The results are quite different. Neither of these alternatives is fully realistic, and defining a single right way is not possible.
The effect is positive in all reasonable cases, but its strength varies greatly between alternatives.
[…] are a vast source of water, and just above the surface of the ocean the atmosphere is saturated Water Vapor vs CO2 as a ?Greenhouse? Gas | The Science of Doom So add more co2 into the atmosphere=more evaporation=more warming. lol But without co2 the water […]
Science:
I could not find table 6 in Hartmann’s book as you indicated here:
Global Physical Climatology, Hartmann, Academic Press (1994)
“Table 6 shows the relative contributions of H2O, CO2 and O3 to reducing the outgoing longwave flux, from which it is seen that the longwave effect of H2O is significantly larger than the effects of CO2 and O3..”
A similar statement can be found on page 71:
“From the results of radiative-convective equilibrium calculations presented in Figs. 3.17 and 3.18, we conclude that water vapor is by far the preeminent greenhouse gas in the natural atmosphere.”
Dave,
Sources are given below the quotes. Thus that quote is from Climate Modeling through Radiative-Convective Models, Ramanathan & Coakley, Reviews of Geophysics and Space Physics (1978) p. 481.
Dave,
Unfortunately the WordPress formatting creates gaps that “match” up (by eye) the quote with the reference above. But the reference is below each quote.
The quote from Hartmann is:
This is on p. 67 of his book.
And the quote referencing table 6 is from Ramanathan & Coakley.
Thank you very much guys. I got it.
Is it true that if there is more H2O in the atmosphere there should be less CO2?
Hameed,
No.
There isn’t a law of “conservation of mass of CO2 + H2O” in the atmosphere.
There is, however, a very difficult to disentangle relationship – where:
– a) more CO2 results (all other things being equal) in higher surface temperatures
– b) higher temperatures result in more water vapor in the planetary boundary layer
– c) more water vapor in the planetary boundary layer may mean more water vapor in the troposophere (lower atmosphere) as a whole
Before elaborating further on the standard theories and limitations of standard theories of climate – what made you suggest that idea?
Thank you for your reply.
What made me suggest this idea is some findings from my model. If you can see my email, please email so that I can share my results (plots and analysis) with you privately because I am still working on it before I publish it.
If you can not email me, we can discuss here without showing the results.
Cheers
Hameed,
It sounds like you’re channeling Miskolczi’s argument that the global annual average optical density of the atmosphere must be constant. It isn’t.
Hameed,
I’m happy for you to share your results here when you are ready.
If water vapor increases and precipitation increases, then the mass of CO2 scrubbed from the atmosphere is increased. Probably not significant…
Howard,
Exchange to and from the surface of the world’s oceans should dwarf the effect of rainfall. Annual rainfall amounts to a volume of 1 m³ for every square meter of surface. The well mixed layer of the ocean is a lot deeper than that.
[…] due to human activity – especially methane (CH4) and nitrous oxide (N2O). And of course, the most important GHG is water vapor, but changes in water vapor concentration are a climate feedback – that is, changes in water […]
[…] an earth-centered universe.(12) CO2 is not the most important greenhouse gas, not by a long shot. Water vapor is. At any given time, water vapor may be up to fifty times more abundant than CO2, yet has a […]