Water Vapor Feedback

Climate wonks with a key to get through Science’s gate should check out Andy Dessler’s excellent discussion in today’s issue of the water vapor feedback. For those who don’t, here’s the nut graph:

[A]lthough there continues to be some uncertainty about its exact magnitude, the water vapor feedback is virtually certain to be strongly positive, with most evidence supporting a magnitude of 1.5 to 2.0 W/m2/K, sufficient to roughly double the warming that would otherwise occur. To date, observational records are too short to pin down the exact size of the water vapor feedback in response to long-term warming from anthropogenic greenhouse gases. However, it seems unlikely that the water vapor feedback in response to long-term warming would behave differently from that observed in response to shorter-time scale climate variations. There remain many uncertainties in our simulations of the climate, but evidence for the water vapor feedback–and the large future climate warming it implies–is now strong.


  1. If the water vapor feedback is so positive and the earth has been warm and wet in the past, long before humans, why are we here? Something must have stopped the warming or we would not exist. What was it?

  2. (Andy Dessler does good stuff.)

    It’s the Stefan-Boltzmann Law, http://en.wikipedia.org/wiki/Stefan-Boltzmann_law,

    which relates the energy related per area, j* to the absolute temperature T:

    j* ~ T^4

    Hence, if you add heat to the Earth, i.e., via increase in solar irradiance, or these days, increase in GHGs that retard the heat dissipation, the positive feedback pushed the temperature up faster and higher than it would without.

    Sooner or later, the T^4 effect radiates energy faster and overpowers the feedbacks, and we’d just settle at a higher equilibrium, not turn into Venus in practice.

    On the other hand, we could get to +6C, see Mark Lynas “Six Degrees”, which would not be good.

    Separate topic: does John have any journalism insights on this Washington Post / George Will brouhaha?

  3. John,
    Thanks. I understand black body radiation. Water effects and black body radiation would seem to predict a permanent equilibrium at some higher temperature. Given this, where do ice ages come from? Are they random events initiated by dust clouds from large volcanic explosions or do they come from some other underlying cause?

    The reason for these questions is that my intuition and research says that a number of climate studies are sort of like the stock pickers that Lehman brothers used. They focus on a few variables that the author understands and ignore bigger or more complicated ones. In the stock picking programs, the quants were told to suppress variables that displayed greater risk in the results. We see how well that turned out.

    In climate projections, I worry much less about suppression of variables and much more about scientists being trained to be very narrow and missing the bigger picture or, in the case of a number of simulations, leaving out important variables because to include them would take too much computational time. The benzene study that I talked about before left out most of the important variables.

    For either narrowness or lack of compuation, the predictions are right within the set of variables that the authors chose, but when all the variables are included we find that the authors’ error bars should have been hundreds or thousands of percent.

    Does my worry make sense to you?

  4. Your worry makes sense … if you think that climate models are done like stock-picking or other statistical series analyses, but they aren’t, they’re physics models.

    1) Ice ages are pretty well understood to happen:

    a) With the Earth having low enough CO2 and continents in theright places.

    b) Driven by Milankovitch orbital cycles, with the size of the temperature jiggles driven by amplification {CO2, water vapor, and some CH4}.


    2) But, what you really want to do, to get a coherent framework is read:

    (general, quick reads):
    a) David Archer, The Long Thaw, 2008.
    b) WIlliam Ruddiman, Plows, Plagues, and Petroleum, 2005.

    c) and for more detail, David Archer, Global Warming – Understanding the forecast, 2007.

    Archer & Ruddiman are both serious scientists, and I have reviews of a) and b) on Amazon.

    3) Models

    and my piece about the specific reasons why people used to one kind of models get confused about others:

  5. At the bottom of the Milankovitch cycle article, deep in the caveats, it say that the earth will either cool for 25,000 years or stay warm for 50,000 years.

    What is an inquiring mind supposed to do with these huge error bars?

  6. Well, it’s Wikipedia, not directly authoritative.
    On my iPhone (ohh, the spelling corrector automatically capitalizes it right), it’s not handy toblook up the refernces.

    But, given the imprecision of language, and the fact that interglacials all cool bck into glacials, but vary in the rate and depth, it is quite possible for the climate to slowly cool and still be a long time before glaciers come back to new York city.

    In any case it’s irrelevant, because we’ve put enough CO2 out that we’re not likely to go back to deep ice age during the like future of human civilization.
    That’s part of Archer’s book.

    If we keep a tech civilization and the glaciers start coming again, it’s actually relatively easy to emit longlived fluorines that vastly exceed CO2 on strength of effect.

  7. I found a reference I was look for, a response to:

    in which Ray Pierrehumbert speculates that we could use a very high radiative-efficiency, long-lived manufactured molecule like SF6, if we ever get in danger of an unwanted ice age.

    Anyway, back to big error bars. The studies referenced are about 20 years apart, the second references the first, and if I were messing with that Wikipedia page, I think I’d just reference the (second) Berger/Loutre paper [can you access that? It’s a nice short perspectives piece that covers the issues.]

    Part of the issue is language ambiguity, in that terms like cooling need a rate to be meaningful, and “ice age” includes everything from “glaciation starts to occur on Baffin Island” to “new Ork is covered in ice”. Also, paleoscientists’ use of the word “soon” often needs qualification for us laypeople. 🙂

    Ruddiman has substantial discussion, although tehre has been additional interesting work since.

  8. The anticipated forcing from the Milankovitch cycles is relatively small compared to that in past ice ages because of the way that the various cycles combine. Anthropic forcing is strong. RTFR Eric

    “The present-day CO2 concentration of 370 ppmv is already well above typical interglacial values of ~290 ppmv. Taking into account anthropogenic perturbations, we have studied further in which the CO2 concentration increases to up to 750 ppmv over the next 200 years, returning to natural levels by 1000 years from now (13, 15). The results suggest that, under very small insolation variations, there is a threshold value of CO2 above which the Greenland Ice Sheet disappears (see the bottom panel of the figure). The climate system may take 50,000 years to assimilate the impacts of human activities during the early third millennium.

    In this case, an “irreversible greenhouse effect” could become the most likely future climate. If the Greenland and west Antarctic Ice Sheets disappear completely, then today’s “Anthropocene” (17) may only be a transition between the Quaternary and the next geological period. J. Murray Mitchell Jr. already predicted in 1972 that “The net impact of human activities on the climate of the future decades and centuries is quite likely to be one of warming and therefore favorable to the perpetuation of the present interglacial” [(1), p. 436].”

  9. Eli,

    I clearly have some work to do in order to understand the details of modeling.

    At some levels, I am still trying to work out mass balance considerations and random event considerations.

    For instance, by mass balance, all the carbon that is currently in coal and oil used to be on the surface of the earth. When it was on the earth’s surface, as far as I know, the continents consisted of lots of tropical forests of ferns. So man, by burning oi and gas is not creating some new earth, in terms of carbon dioxide, only restoring an old one by bringing carbon to the surface that used to be on the surface.

    Since this old earth was a precursor of the current earth and no runaway occurred between then and now, I get suspicious of predictions of runaway events. The predictions may be right but, to me, they need to predict not only the future but also the transition from the tropical forest earth to the current earth.

    As to the butterfly effect, I had people at LANL predict that LANL pensions were completely solid. They weren’t. These people would not consider certain variables. The quants in Wall Street also were not allowed to consider certain destabilizing variables. They were told to have a low risk fairly linear model that would not predict stock market collapses.

    So, my question is do the climate models include destabilizing variables, such as tsunamis, volcanic explosions, breaking off of large glaciers from Antartica, or coupling of land, air, water, and biology or do they assume continous derivatives, no coupling, and no random events.

    The models that I have seen, across a number of fields, that do not include the random/chaotic events tend to get the predictions wrong.

    If I am going to bet my daughter’s future on a model, I would like to know that the model did not leave out important events, for instance random events or events that were know to occur.


  10. Eric: it’s generally safe to bet that if you’re reading “worst thing that could happen” and “what could possibly go wrong” someone’s filtering.
    Read what the scientists are saying directly.

    Worst case if the sun cools slightly for a few thousand years but we’ve stabilized CO2, is we have a vastly improved more efficient power grid, and can use SF6 or CFCs to warm the planet back up and save the carbon compounds to build things, and save the oceans.

    Worst case if the sun doesn’t vary significantly, and we don’t do the cheap and easy energy efficiency and stabilize CO2, is we lose the oceans in the short term and get a warming excursion in the slightly longer term.

    There was nowhere you could have put your pension money that would have been safe, because the constraints on speculation that had been in place for a century were thrown off in 1999. They told you it couldn’t be a problem, what could possibly go wrong? Ignoring the history of every other time that was done; a few got rich, the whole financial system took a big hit. It’s a ratchet-racket.

    This paper from back then sums up the history of previous excursions into speculation and the on-off nature of controls through the last century or more up to just before that event occurred, if you’re curious. Good reading:

    SSRN-Why the Law Hates Speculators
    Duke Law Journal, Vol. 48, p. 701, 1999

  11. oh, that was the abstract; here’s your full text link:


    What’s interesting about this as an analog to the climate change argument is the way a large amount of information was simply being written off because of an economic argument that it couldn’t possibly be as much of a problem as history suggested. This paper summed up the history, sort of a last caution before they threw over the regulations. After that, well, eight years of speculative boom happened, then the inevitable.

    We’re well into the speculative boom from overdosing on coal.

    “… economic reasoning itself cautions against simply dismissing antispeculation laws as the mistaken products of ignorance and envy. An important strain of the law and economics literature argues that the common law generally favors efficient rules.23 Many contemporary antispeculation laws, including the CEA, can trace their roots back to the earliest days of the common law.24 These origins hint that [*pg 708] antispeculation rules actually may serve an important, if generally unrecognized, efficiency function.”

  12. To everyone

    Huge thanks.

    We have invented and are building the engine that gets the high efficiency and low carbon emissions that are desired.

    But, I keep getting questions about the broader picture. The most recent questions were about the validity of sequestering carbon through growing algae.

    The answers here are helping me to understand the broader picture in scientific terms, with error bars, so that I can talk humbly but correctly.

  13. I talk to algae folks on occasion, although the primary interest there is in biofuels, sometimes as backends for fossil-fuel plants, as per http://www.greenfuelonline.com/ .

    Of course, those are recycling things, not sequestration.

    So far, the most interesting *real* sequestration effort I’ve seen is {CO2 => {cement, concrete}}, one of which I described here:


    I don’t know if it scales, but if it does, it may be one of the few things that might actually work.

  14. John,
    I think that I have algae sorted out, but I will look at the URL you listed.

    As to the ‘real’ effort of Calera, to me it seems that there are a number of apparent gotchas.

    Here are some of them.

    1. To get the Ca in order to make CaCo3 for concrete, you normally have to calcine CaCO3 to drive off the CO2 and produce CaO. If this is true, Calera’s process is net CO2 neutral in a full cycle because you emit to the atmosphere the same CO2 that you absorb later.
    2. I think that the organisms that live in the sea, kelp, abalone, etc. all need Mg and Ca in order to survive. So large amounts of Calera’s softened output water would poison the indigenous sea life by removing the ions that they need to live thus making scaling of the process nearly impossible.
    3. I read that normal concrete made with Portland cement absorbs CO2 naturally once the concrete is poured. If this is true, all the normal concrete will be absorbing CO2 without a need for Calera’s process. In addition, as the son of an architect, I would be worried that adding CaCO3 to concrete during hardening in place of the CaO in Portland cement would lower the tensile and compressive strength of the concrete. If this is true, then buildings using Calera concrete would have to be redesigned to use this weaker concrete.

    I do not yet know whether any or all of the above caveats are true for Calera. I do not have the relevant technical information about their process. So this is a work in progress on my part.

    A possible worry if Calera’s process is scaled up, say in northeastern Australia or of the coast of Florida, is that the Ca necessary in the process will be replenished in the ocean by dissolving it out of the local source for Ca–coral reefs. So a possible consequence of wide adoption of the Calera process would be the destruction of coral reefs. The destruction of coral reefs by man is viewed as a bad thing.

    In addition, as you point out, the process is not so useful where the coal plants are–far from the ocean. Power companies that have asked me about this process are all far from the ocean.

    More later.

  15. I suggest continuing that at Deltoid to avoid diverting this one here. I think most of those concerns were already discussed there.

Comments are closed.