On the Relative Contribution of Carbon Dioxide to the Earth’s Greenhouse Effect

September 10th, 2010 by Roy W. Spencer, Ph. D.

[NOTE: What follows assumes the direct (no-feedback) infrared radiative effects of greenhouse gases (water vapor, CO2, methane, etc.) on the Earth’s radiative budget are reasonably well understood. If you want to challenge that assumption, your time might be better spent here.]

I was recently asked by a reader to comment on a new paper by Schmidt et al. which put some numbers behind the common question, What fraction of the Earth’s greenhouse effect is due to carbon dioxide?

There are a wide variety of answers to this question depending on how you define “greenhouse effect”, what your assumed baseline is, etc. Conceptually, in any greenhouse atmosphere, greenhouse gases warm the lower layers and cool the upper layers compared to if those gases were not present. That never changes. It’s the way you compute the relative magnitude (say, in percent) of that warming that depends greatly upon your assumptions.

Note that the greenhouse effect can only be calculated based upon theory. The greenhouse effect isn’t a physical variable like temperature that you can measure. It is a radiative process that affects the atmosphere’s energy budget at all altitudes, warms the surface, and whose components must be calculated based upon radiative transfer theory and the IR absorption characteristics of greenhouse gases (and clouds).


The Wrong Question

I will argue that if what we are REALLY interested in is how much the Earth’s greenhouse effect will be enhanced by adding CO2 to the atmosphere (the only reason we are interested in the CO2 issue anyway, right?), then the above question is not very relevant.

In fact, the answer to it can totally mislead us. This is easy to show with 2 simple examples.

First, assume there was NO naturally occurring carbon dioxide in the atmosphere, and we added 300 ppm. In that case, the natural influence of CO2 on the Earth’s greenhouse effect would be zero, but the influence of adding 300 ppm would be quite significant.

Now, as the second example let’s assume the natural CO2 concentration is high, say 1,000 ppm, and THEN we added 300 ppm. In this second case, the natural role of CO2 in the Earth’s greenhouse effect would be very significant, but our addition of 300 ppm more would have a relatively small direct warming influence.

This is because the more CO2 there is in the atmosphere, the more “saturated” the CO2-portion of the greenhouse effect becomes, a well known feature that has a standard simplified, logarithmic formula for its computation.

Everyone already knows about this mostly saturated condition relative to the radiative effect of carbon dioxide – even the IPCC. Adding more and more CO2 causes incrementally less and less warming (again, assuming no feedback, which is a separate issue)….but the radiative effect of CO2 in the atmosphere is not totally saturated.

And it never can be, for the same reason that you can keep dividing a number by two forever, and the resulting number will get extremely small…but it will never reach zero.

So what do these two examples tell us? If the natural contribution of CO2 to the greenhouse effect was ZERO, then the warming effect of our addition of 300 ppm would be relatively large. But if the natural contribution of CO2 to the greenhouse effect was already large, then the incremental warming effect of adding more will be small.

An extreme example would be Venus, which has 230,000 times as much CO2 in its atmosphere as Earth does. Our addition of CO2 to that atmosphere would have essentially no effect.

The point is that knowing what percentage of the Earth’s natural greenhouse effect comes from carbon dioxide alone tells us little of use in determining how much warming might result from adding more CO2 to the atmosphere.

How Much is the Earth’s Greenhouse Effect Enhanced by Adding More CO2?
This is the question we should be asking, and it can be easily answered with a couple of numbers quoted in the Schmidt et al. article.

Schmidt et al. assumes the commonly quoted 33 deg. C as the amount of surface warming due to the Earth’s greenhouse effect, and for the time being I will assume the same. (In my next blog post, I will explain why this number is NOT a good measure of the Earth’s greenhouse effect.)

Thirteen years ago, Danny Braswell and I did our own calculations to explore the greenhouse effect with a built-from-scratch radiative transfer model, incorporating the IR radiative code developed by Ming Dah Chou at NASA Goddard. The Chou code has also been used in some global climate models.

We calculated, as others have, a direct (no feedback) surface warming of about 1 deg. C as a result of doubling CO2 (“2XCO2”).

So, this immediately gives us numbers we can use to compute a percentage increase in the greenhouse effect: Doubling of atmospheric CO2 (which will probably happen by late in this century) enhances the Earth’s greenhouse effect by about (1/33=) 3%.

This value (3%) for the enhancement of the Earth’s greenhouse effect from our addition of CO2 is much smaller than the 20% value that Schmidt et al. get…but remember that we are addressing two different issues. I claim what we should be interested in is the relative size of our enhancement of the greenhouse effect, rather than how much of the Earth’s natural greenhouse effect is due to CO2. The latter question really proves nothing about how much effect adding MORE CO2 to the atmosphere will have.

Next Time: Why 33 deg. C is a Misleading Number

In my next post, I will discuss why the use of 33 deg. C for surface warming due to the greenhouse effect is very misleading. The issue is not new, as it has been known since the 1960s. I wasn’t aware of its central importance to the global warming debate until Dick Lindzen published his 1990 paper, Some Coolness Concerning Global Warming.

WeatherShop.com Gifts, gadgets, weather stations, software and more...click here!


23 Responses to “On the Relative Contribution of Carbon Dioxide to the Earth’s Greenhouse Effect”

Toggle Trackbacks

  1. Harold Pierce Jr says:

    The link to D. Lindzen’s paper does not work: “.pd” should be “.pdf”.

    Venus is hot because it is closer to the sun, and the thick atmosphere of CO2 has a very high R value which greatly reduces the flux of heat from the surface to outer space. The clouds also reduce the flow heat to outer space.

  2. wayne says:

    “… say 1,000 ppm, and THEN we added 300 ppm. In this second case, the natural role of CO2 in the Earth’s greenhouse effect would ^ be very significant,

    Did you mistakenly leave out “not” at the caret?

    • Anonymous says:

      Dr. Spencer,

      Sorry, seems I misread that paragraph. Ignore question.

      Here’s my problem and you are speaking right in that area. I know there is a GHG effect, I know of “back radiation” to depth. Please don’t take this too lightly (or too heavily).

      First, thank you so very much for your openness. Much appreciated. I have taken your simple GCM Excel spreadsheet you gave to everyone many posts ago, labeled the constants by name and that allowed me to learn greatly from that gesture. I even extended it to am actual program so I could easily change the parameters and more importantly the length of years by any factor so different scales can be viewed. (yours on just asking) And that brings me exactly to what you are speaking of in this post.

      You said: “First, assume there was NO naturally occurring carbon dioxide in the atmosphere, and we added 300 ppm.” But right there you didn’t remove water’s already logarithmic decrease in the dT of 33 ºC and then calculated later on just CO2. Why, ‘o why? Is that in the next article? People want to know this fine but very critical point.

      However, I wish you would not have left out in your initial conditions the 1-2% water vapor that is always in our atmosphere since near day one and then spoke of CO2 from zero to today’s 400 ppmv. Why 300 ppmv, are you going to address this 100 ppmv in the next article? I think many people reading here know physics to an adequate level to realize this is important, very very important. That seems why someone evidently asked your opinion on Schmidt et al. paper. The 20% is roughly what I have always come up with and that is why the 33 ºC is so misleading, some try to put the 33ºC “greenhouse effect” totally on CO2’s back, will wait for next installment. Water vapor has the same logarithmic effect as CO2 and it’s doubling and doubling and doubling from 1 ppmv to water vapors ~10-20,000 ppmv current concentration must be subtracted from that 33 ºC first if we are going to only speak of CO2’s influence, not a unreal Earth with only O2 & N2 with absolutely no GHG’s and then speak only of CO2. That’s completely meaningless.

      I hope you don’t make this series (it sounds like your are beginning a series) so simple that once again all real and critical factors are left out and initial conditions so simple that once again we advance little, just a bunch of commenters complaining of all of those missing factors left out and we go nowhere.

      Now you may be writing this series to address readers who know little of physics and this is an attempt to bring them very roughly along to the basics involved and if so, please say something in a little “que” so many deep into the physics involved can just skip it and not confuse the flow of comments.

      BTW: I and many others keep coming up with ~1.67% since logarithmic, not the linear 1/33 or 3%. Were you just being very rough or am I off-track? Of course, ignoring any feedbacks.

      OTOH, maybe I’m getting ahead of you.

      Thank again from the many willing to check and sometimes learn the proper physics involved.
      -wayne

  3. Leonard Weinstein says:

    Roy,
    The statement “Conceptually, in any greenhouse atmosphere, greenhouse gases warm the lower layers and cool the upper layers compared to if those gases were not present” is only true if the ground level temperature is still rising, not if it has settled to a new high level due to the greenhouse gas. The greenhouse gas will warm the ground and lower atmosphere, and will raise the effective level in the atmosphere where the outgoing radiation leaves from, but when a new balance condition is obtained, the outgoing has to equal incoming. I assume you were referring to Earth, where the temperature was assumed to still be rising due to continual increases in greenhouse gas, so the upper level outgoing was lagging incoming, but you need to point this out.

  4. Eric Anderson says:

    Good point that the question often asked is not really the right question to ask with respect to AGW. Nonetheless, I’m highly suspicious about the amount of the natural effect assigned to CO2 in the 33 degree average, so hopefully that is somewhat addressed in your next post . . .

  5. Juraj V. says:

    Fact is, that those “radiative forcings” are either minuscule or swallowed by negative feedback, because models including them are way out of reality. But models without them are much better.

    http://www.esrl.noaa.gov/psd/research/review/2010/presentations/1_sardeshmukh.pdf

  6. Joletaxi says:

    dear dr Spencer few weeks ago, you proposed an article in which you wanted to persuade us of the greenhouse effect.Really, this topic was fascinating, and some friends and I were on the subject of much discussion, and I must admit, it has occupied our evenings and we even prevented Us from sleeping.
    Now, you seem to have the project too tell us that after all this has little importance, and the famous greenhouse effect would be ridiculous?
    Pielke’s paper is very interesting and joined the commentary I made on your article.
    Finally, where are You intended Us too go?.That water is the most important element in the regulation of flows?
    We French, we already knew that for years: always 4 volumes of water in one volume of Pastis,that is the rule.

  7. chris y says:

    Roy, thanks for the link to Lindzen’s paper. I had not seen this before now.

    I look forward to your next posting on the 33 degree greenhouse effect.

    I like to start with an atmosphere-free Earth with albedo =0.1 (no clouds) to get a baseline surface temperature of about 1 C. Greenhouse gases in our standard atmosphere without convection gives a surface temperature of 65 C. Convection, weather and clouds reduce this to 15 C. It doesn’t take much of a change in convection, clouds or heat transport to counter small increases in CO2 forcing. In fact, I think it is probably drowned out by natural fluctuations.

  8. Loodt Pretorius says:

    Hi Roy,

    A question I have been too shy ask anybody else.

    When the CO2 level increases from say 290 to 390ppm, it logically implies that the additional 100ppm is replacing some of the other gases in the air, because you can only have a million parts in ppm.

    The additional CO2 must push-out or reduce the concentration of some of the other gases, what gases are we talking about?

  9. Martin C says:

    Dr. Spencer,
    Thank you VERY much for the commentary as a reponse to my question. As I am trying to understand the whole CO2 issue better (I was first trying to get my arms around the issue of what percent the atmospheric constituents contribute to the greenhouse effect, and then consider increases), it is obvious to me now that it is the change in greenhouse effect (and IF it changes ‘feedbacks’) with changes in CO2 that is the key issue.
    From other websites and information I have seen, it sure seems to me that the dynamics of the atmosphere (heated air rising, latent heat of water evaporation from the oceans and condensing at higher altitudes giving up the heat where the heat can be reradiated into space)can sure take care of any increase in temperature from CO2.
    And I look forward to your next post.

  10. Hello Loodt,

    Dr. Spencer can correct me if I am wrong, but carbon burns and combines with diatomic oxygen to produce CO2. So 100 O2 molecules will combine with carbon to produce 100 CO2 molecules. The answer to your question is that oxygen is decreased. According to Wikipedia, the oxygen content of dry air is 20.95% and CO2 is 0.039%. So presumably when CO2 was 0.029%, the oxygen was 20.96%.

  11. Loodt Pretorius says:

    Hi Werner, thank you for your reply.

    Note that you are assuming that all the CO2 is created by combustion, burning. However, due to global warming the oceans are heating up and in the the process are venting or degassing a lot of dissolved CO2. (I’ll stay away from volcanos that are releasing gas from deep down where the earth is millions of degrees a la Gore)

    I still pose the question, what gases are being replaced with the addition of CO2 to the atmosphere?

    • kuhnkat says:

      Loodt,

      I knew a chemist I was discussing this issue with. I was talking about the oceans emitting CO2 as they warmed and he pointed out that the Partial Pressure of the gasses were a large determinant of how much CO2 would be emitted. At the time I hadn’t thought through what he was trying to tell me and I simply asked him for a rough calculation as to the differences. He didn’t help me.

      At this time I realize that if he had answered he would have partially undermined his pro AGW stance. As we emit CO2 it increases the partial pressure and REDUCES the amount of CO2 that will be released from the oceans. If any smart guys are reading this it would be a good time to support or attack AGW with some realistic calculations of the differences in ocean CO2 emissions with and without human fossil fuel CO2 emissions!!

    • Anonymous says:

      “I still pose the question, what gases are being replaced with the addition of CO2 to the atmosphere?”

      Why do you think that CO2 needs to replace another gas? It seems to me that one could add CO2 and the atmosphere would just contain more gas. I am not aware that there is a practical constraint on the amount of gas in the atmosphere.

      To the extent any replacement does in fact take place, the combustion of carbon consumes oxygen.

      • Anonymous says:

        Beacuse the unit of measure is parts per million — a finite number. If co2 increases one ppm, something else must decrease by one ppm. Otherwise, the measure would be parts per million-and-one.

  12. You are welcome Loodt!

    “Note that you are assuming that all the CO2 is created by combustion, burning.” I am assuming that just the EXTRA CO2 since 1750, which raised the level from about 280 ppm to 390 ppm, is due to combustion by man. My first answer was an initial approximation. However I did not consider that any extra CO2 got dissolved in the ocean. This is happening however.

    “However, due to global warming the oceans are heating up and in the the process are venting or degassing a lot of dissolved CO2.” This was the reason CO2 went up before 1750 at different times. While it is true that warm water dissolves less CO2 than cold water, the oceans have not warmed up that much since 1750 to cause a net degassing. The oceans are still a net sink of CO2 due our contribution and absorb a large amount of the extra CO2 that man puts in. (See http://www.telegraph.co.uk/earth/earthnews/6538300/Climate-change-study-shows-Earth-is-still-absorbing-carbon-dioxide.html
    One sentence in this site says: According to the study, the Earth has continued to absorb more than half of the carbon dioxide pumped out by humans over the last 160 years.) So with this adjustment, the O2 in 1750 would have been slightly higher than 20.96% or 209,600 ppm.

    “I still pose the question, what gases are being replaced with the addition of CO2 to the atmosphere?” IF the extra CO2 is due to combustion, just oxygen is being replaced.

    I am not sure if this is what you are getting at, but let us assume Earth had just 80.00% nitrogen and 20.00% oxygen. Then let us assume a huge asteroid containing dry ice (solid CO2) hit Earth and caused the concentration of CO2 to go from 0.00 ppm to 100 ppm. In that case, using round numbers, nitrogen would go down 80 ppm from 800,000 ppm to 799,920 ppm and oxygen would go down 20 ppm from 200,000 ppm to 199,980 ppm. In this scenario, nothing is being replaced or lost in that sense. Nitrogen and oxygen are just getting more diluted.

  13. Ken says:

    What water vapour profile did you use in your radiative transfer model in which you calculated a 1 degree Celsius warming due to a CO2 doubling? Did you use the USST76 standard atmosphere or a realistic water vapour profile?

    The NOAA and the TIGR radiosonde databases show almost identical global average water vapour profiles, of 2.6 precipitable cm (prcm) while the USST76 atmosphere contains only about half of this amount.

  14. Alexander Davidson says:

    I’ve also been thinking about CO2, triggered by a figure in the article to which you link: http://en.wikipedia.org/wiki/File:Radiative-forcings.svg . 1.6 W/m^2 ‘Total net anthropogenic’ is offset mainly by -0.5 W/m^2 ‘direct effect’, -0.7 W/m^2 ‘cloud albedo effect’.

    When the modellers bet the farm on ‘global dimming’, the ‘direct effect’ was proven but the ‘indirect effect’ was an assumption presumably based on Twomey’s work, but he had warned it didn’t work for thick clouds. Unsurprisingly, experiment couldn’t prove ‘cloud albedo cooling’.

    Twomey was given a prize: http://geo.arc.nasa.gov/sgg/singh/winners4.html . Look carefully at the claimed physics, plausible but wrong. Also: http://terra.nasa.gov/FactSheets/Aerosols/ claims up to 90% albedo, as observed. Yet Twomey’s theory can’t give >50%. Many believe in greater reflection from more, smaller droplets, but it’s nonsense.

    Consider a thick [optical depth ~1], non-absorbing cloud with 70% spherical albedo. 30% diffuse light exits the base, the same from the top and superimposed on this is 40%, highly directional, from a different effect. I’ve got some ideas about it based on the details of Mie scattering: there’s a strong dependence on droplet size. It’s probably why clouds about to rain become darker. As an exercise deconstruct 90% spherical albedo [clue: a very dark cloud].

    So, apart from thin clouds, there never was a cooling ‘cloud albedo effect’. If anything, reduction of droplet size lowers albedo, a heating effect. So AGW prediction in AR4 is unjustified. Firstly near halve it then revamp the models. I’ll be surprised if it ends up more than a third.

    As a final point, if NASA’s intention through claiming false optical physics was to mislead the discipline into believing that despite no experimental proof there was sufficiently good theoretical justification to include it in AR4, that was scientific fraud, a ‘bait and switch’.

  15. Christopher Game says:

    Christopher Game asking a few questions.

    Dear Dr Spencer,

    You present blog article reads:

    “The greenhouse effect isn’t a physical variable like temperature that you can measure. It is a radiative process that affects the atmosphere’s energy budget at all altitudes, warms the surface, and whose components must be calculated based upon radiative transfer theory and the IR absorption characteristics of greenhouse gases (and clouds).”

    and

    “Schmidt et al. assumes the commonly quoted 33 deg. C as the amount of surface warming due to the Earth’s greenhouse effect, and for the time being I will assume the same. (In my next blog post, I will explain why this number is NOT a good measure of the Earth’s greenhouse effect.)”

    According to J.R. Holton,An Introduction to Dynamic Meteorology, 4th edition, Elsevier, Amsterdam, 2004, ISBN 9780123540157, at page 51 in his Section 2.7.2 The Adiabatic Lapse Rate, the adiabatic lapse rate can be calculated simply by use of the hydrostatic equation and the ideal gas law. Adiabatic means that there is conceived for this calculation to be no transfer of heat; all putative energy transfer is of potential energy or of work. Thermal radiation is a form of transfer of heat and is therefore apparently excluded by this term ‘adiabatic’.

    Holton is thus calculating a virtual or hypothetical temperature lapse rate in the troposphere entirely without reference to radiation.

    Holton of course goes on to show that this virtual lapse rate is not usually actualized in the real atmosphere.

    Your above quoted remarks seem to attribute the tropospheric temperature lapse rate principally to radiative transfer.

    I do not find in textbooks a comfortable resolution of this apparent paradox. For example, Holton’s Section 2.7.2 does not actually mention radiation and does not explicitly say that his idea of adiabatic process excludes radiation.

    I don’t think this one is easy to understand. I am not suggesting there is a real paradox or contradiction. I am just saying the theory is not easy to understand as presented in textbooks. What do you think?

    While we are about it, what do you think of the quantitative definitions of the greenhouse effect proposed by Inamdar and Ramanathan (1997) ‘On monitoring the atmospheric greenhouse effect from space’, Tellus 49B: 216-230, and by Stephens, Slingo, and Webb (1993) ‘On measuring the greenhouse effect of Earth’, NATO ASI Series, vol. 19, pages 395-417?

    I&R97 define the greenhouse effect G by G = Sg – OLR, where Sg denotes the upwards thermal radiation flux density entering the atmosphere from the ground (land-sea body).

    SS&W93 define it by G` = Sg / OLR.

    Yours sincerely,

    Christopher Game

  16. Christopher Game says:

    Christopher Game commenting on Dr Spencer’s blog article On the Relative Contribution of Carbon Dioxide to the Earth’s Greenhouse Effect posted on September 10th, 2010.

    A way of thinking about the relative contribution of CO2 and H2O to the greenhouse effect is to ask:

    What percentage change in water vapour column amount, uniformly distributed, would have the same effect on the infrared opacity of the atmosphere as a doubling of CO2 ?

    Answer: using the global average Planck-weighted greenhouse-gas optical thickness as the measure of the infrared opacity of the atmosphere, a 2.77% increase in water vapour column amount, uniformly distributed, would have the same virtual direct no-feedback effect as a doubling of CO2. This figure is stated on page 257 of Miskolczi’s 2010 paper.

    Also stated there is that the opacity as defined above is directly (no-feedback) affected by temperature. This effect is not envisaged in the IPCC doctrine and is not representable in the standard version of the IPCC formalism. The virtual direct no-feedback increase in opacity due to CO2 doubling would be just cancelled by the virtual direct no-feedback decrease in opacity due to a 2.65K uniform increase of temperature. If the IPCC formalism were somehow modified to enable it to represent this effect, it would count as a negative feedback not currently envisaged in the IPCC doctrine.

    The above figures are virtual direct no-feedback contributory effects, not eventual combined direct and indirect actual after-feedback effects.

  17. Dave Kelly says:

    First question, even with increasing levels of CO2 resulting in less effect on the climate system, with roughly 14 times the CO2 as that of Earth, shouldn’t Mars demonstrate a more significant greenhouse effect? Next question; when referring to greenhouse gas levels approaching saturation, is the effect cumulative regardless of the distribution of greenhouse gasses or does each gas have an independant level of saturation? Last question; if an increase in CO2 results in warming that results in more water vapor in the atmosphere, wouldn’t the same thing happen with water vapor alone? More water vapor results in more warming which results in more water vapor. That feedback loop should result in complete humidity that would endure as long as there was liquid water. That conundrum alone demonstrates at least some certain degree of negative feedback.

  18. RW says:

    I’m hoping someone can help me out here. I’m just a lay person that has been researching this issue for about a year. Maybe I’m missing something obvious, but as best I can tell, the whole CO2 forcing theory seems to be one big “the emperor has no clothes.” There are many things that just don’t seem to add up.

    Firstly, I don’t quite understand how the so-called “intrinsic” effect from a doubling of CO2 results in about a 1 degree C increase of the surface temperature. Is this assuming no other substance in the atmosphere? No convection either? Wouldn’t about half of this effect be radiated out to space (half goes up, half goes down to the surface)- making the “intrinsic” response only about 0.5 degrees C? Also, doesn’t water vapor overlap the main CO2 absorbing LW bands (around 15u) and exist in much higher concentrations than CO2? I mean it seems to me that even without factoring in clouds (and their potential feedbacks – negative or positive), it seems highly questionable to me that a doubling of CO2 would cause any increase in the surface temperature – let alone the 1 degree C that is supposed to cause all the warming via positive water wapor and cloud feedbacks (according to the IPCC models).

    I mean isn’t this the crux of the whole thing? I mean without the 1 degree of warming, there can’t be any amplified warming from positive water vapor and cloud feedbacks, right?

    I also see that according to the proxy geological evidence, CO2 levels have no causal relationship to temperature going back thousands, even hundreds of millions of years. Even when CO2 levels were like 10+ times higher than present, we had colder temperatures than now. In addition, we have more recent history like the Roman Warming and the Medieval Warming – both likely warmer globally than now. If the current CO2 AGW theory says that that the little increase above current temperatures from a doubling of CO2 is supposed to trigger runaway warming via positive water vapor and cloud feedbacks, why didn’t that happen during the Medieval or Roman warmings? Instead the exact opposite occurred – it got colder. How can these things be ignored?

    Maybe I’m wrong, but it looks highly probable to me that this whole thing is a fairytale. The weight of the evidence seems to strongly suggest the effect of increasing CO2 is – if not infinitesimal, extremely small. Or am I wrong?