#### Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Options

# "Evidence" for "water vapor feedback is strongly negative"

I'd like to get some background information on the blog post New paper finds water vapor feedback is strongly negative from Clive Best.

My current knowledge is very slim, it's just that water feedback is hard to determine, that the IPPC scenarios assume a positive feedback, and that clouds can work both ways. References to background information would also be welcome.

I hope we can have a discussion far away from the battle fields and shielded from all the noise.

• Options
1.

It's a bit charitable to call it a "paper" since it was rejected for publication ...

By the way, it's not really correct to say that the IPCC models assume a positive feedback. They calculate a positive feedback.

The blog entry / manuscript seems to have two components: (1) presenting a "solution" to the Faint Young Sun paradox (despite the author later claiming in the comments that he's not trying to present a solution), and (2) arguing against a positive water vapor feedback.

(1): I'm not too interested in picking apart "novel" theories, but I can't see how a waterworld solves anything. (His assumptions don't appear to depend on a waterworld, and they aren't physically justified from anything as far as I can see either.)

(2): It splits into two sub-arguments: (2a) that a positive water vapor feedback contradicts paleoclimate evidence, and (2b) that a positive water vapor feedback contradicts historical temperature data.

(2a): I don't see where the nonlinear curve in his first figure comes from. He starts from the energy balance equation $DT = (DS + F DT) G_0$. (This is Equation 4 of Roe (2009) with some redefinition of terms.) Solving for the temperature change, $DT = G_0 DS/(1-FG_0)$. Assuming the solar forcing ($DS$) changes linearly with time, and everything else in the equation is constant, that should give a linear temperature change in time.

That aside, the broader question is what do climate feedbacks have to do with the FYS paradox? His argument doesn't really say anything about water vapor feedbacks specifically, just the total feedback. Regardless of the shape of the curve, it is true that a stronger climate feedback exacerbates the FYS paradox. To recap, the paradox is that incoming solar radiation was supposedly weaker in the past, and therefore the climate should have been much colder than it really was. (One of the blog comments cites a recent review article.) If you increase climate feedbacks, you increase the cooling in response to a reduced forcing, which makes the prediction even colder and worsens the paradox.

That being said, it's naive to claim that this tells you much about water vapor feedbacks (or even much about total feedbacks). There are other proposed solutions to the FYS paradox, such as an enhanced greenhouse effect. You can't just attribute all climate to feedbacks acting on changes in incoming solar radiation. There are problems with getting a strong enough greenhouse effect with current estimates of paleo greenhouse levels, so this explanation doesn't solve everything. But neither does any other explanation, really. You're not going to solve it just by introducing a low climate sensitivity (and that would introduce many more problems in explaining other paleoclimates). So until we know what is going on, how are we going to get an estimate of climate feedbacks from such a distant time? And there is likely a lot of state-dependence involved too with such dramatically different physical situations, so extrapolating to modern-day climate feedbacks is problematic. Anyway, you can't just ignore things like greenhouse gas changes completely in Figure 1, even if the author doesn't think they solve the paradox themselves. They'll certainly change the curves in the figure.

(2b): Here he claims to estimate water vapor feedbacks by comparing warming at arid vs. humid locations, to factor out the temperature-dependence. I don't think this estimation method can work. There is a lot of confounding. For example, some arid regions are polar and have large albedo feedbacks, regardless of the water vapor feedback, so you'll see polar amplification which the author's method would misattribute to a negative water vapor feedback; there is also spatial variation in cloud feedbacks, etc. The author also dismisses confounding from thermal inertia effects specifically and teleconnections with the oceans generally, which I don't find persuasive. What would be a lot more persuasive is a perfect model experiment, showing that this method correctly diagnoses the known water vapor feedbacks in the multimodel ensemble of 20th-century GCM hindcasts. I really doubt that it can.

This method's findings also contradict the existing literature that finds a positive water vapor feedback in the observational data. (See references in Chapter 9 of the IPCC AR4 WG1 report.) No explanation is given for the discrepancy or why this method should be superior.

Comment Source:It's a bit charitable to call it a "paper" since it was rejected for publication ... By the way, it's not really correct to say that the IPCC models assume a positive feedback. They _calculate_ a positive feedback. The blog entry / manuscript seems to have two components: (1) presenting a "solution" to the Faint Young Sun paradox (despite the author later claiming in the comments that he's not trying to present a solution), and (2) arguing against a positive water vapor feedback. (1): I'm not too interested in picking apart "novel" theories, but I can't see how a waterworld solves anything. (His assumptions don't appear to depend on a waterworld, and they aren't physically justified from anything as far as I can see either.) (2): It splits into two sub-arguments: (2a) that a positive water vapor feedback contradicts paleoclimate evidence, and (2b) that a positive water vapor feedback contradicts historical temperature data. (2a): I don't see where the nonlinear curve in his first figure comes from. He starts from the energy balance equation $DT = (DS + F DT) G_0$. (This is Equation 4 of [Roe (2009)](http://www.annualreviews.org/doi/abs/10.1146/annurev.earth.061008.134734) with some redefinition of terms.) Solving for the temperature change, $DT = G_0 DS/(1-FG_0)$. Assuming the solar forcing ($DS$) changes linearly with time, and everything else in the equation is constant, that should give a linear temperature change in time. That aside, the broader question is what do climate feedbacks have to do with the FYS paradox? His argument doesn't really say anything about water vapor feedbacks specifically, just the total feedback. Regardless of the shape of the curve, it is true that a stronger climate feedback exacerbates the FYS paradox. To recap, the paradox is that incoming solar radiation was supposedly weaker in the past, and therefore the climate should have been much colder than it really was. (One of the blog comments cites a recent [review article](http://www.agu.org/journals/rg/rg1202/2011RG000375/).) If you increase climate feedbacks, you increase the cooling in response to a reduced forcing, which makes the prediction even colder and worsens the paradox. That being said, it's naive to claim that this tells you much about water vapor feedbacks (or even much about total feedbacks). There are other proposed solutions to the FYS paradox, such as an enhanced greenhouse effect. You can't just attribute all climate to feedbacks acting on changes in incoming solar radiation. There are problems with getting a strong enough greenhouse effect with current estimates of paleo greenhouse levels, so this explanation doesn't solve everything. But neither does any other explanation, really. You're not going to solve it just by introducing a low climate sensitivity (and that would introduce many more problems in explaining other paleoclimates). So until we know what is going on, how are we going to get an estimate of climate feedbacks from such a distant time? And there is likely a lot of state-dependence involved too with such dramatically different physical situations, so extrapolating to modern-day climate feedbacks is problematic. Anyway, you can't just ignore things like greenhouse gas changes completely in Figure 1, even if the author doesn't think they solve the paradox themselves. They'll certainly change the curves in the figure. (2b): Here he claims to estimate water vapor feedbacks by comparing warming at arid vs. humid locations, to factor out the temperature-dependence. I don't think this estimation method can work. There is a lot of confounding. For example, some arid regions are polar and have large albedo feedbacks, regardless of the water vapor feedback, so you'll see polar amplification which the author's method would misattribute to a negative water vapor feedback; there is also spatial variation in cloud feedbacks, etc. The author also dismisses confounding from thermal inertia effects specifically and teleconnections with the oceans generally, which I don't find persuasive. What would be a lot more persuasive is a perfect model experiment, showing that this method correctly diagnoses the known water vapor feedbacks in the multimodel ensemble of 20th-century GCM hindcasts. I really doubt that it can. This method's findings also contradict the existing literature that finds a positive water vapor feedback in the observational data. (See references in Chapter 9 of the IPCC AR4 WG1 report.) No explanation is given for the discrepancy or why this method should be superior.
• Options
2.

Thanks. One reason why I asked is that this article was mentioned on the forum of a main stream German newspaper (Die Zeit). I have started to comment on this one and on the forum of the other big main stream weekyl newspaper in Germany on climate change, mostly to counter rather basic misconceptions.

By the way, it's not really correct to say that the IPCC models assume a positive feedback. They calculate a positive feedback.

And there is likely a lot of state-dependence involved too with such dramatically different physical situations, so extrapolating to modern-day climate feedbacks is problematic.

That's been my first thought, too - it seems to be very naive to assume a constant feedback over a range of such diverse states. (And of course it seems to be even more naive to write headlines like "proof of..." based on this article.)

One of the blog comments cites a recent review article.

The link takes me to a login screen with no hint of the authors and the title of the article :-(

Comment Source:Thanks. One reason why I asked is that this article was mentioned on the forum of a main stream German newspaper (Die Zeit). I have started to comment on this one and on the forum of the other big main stream weekyl newspaper in Germany on climate change, mostly to counter rather basic misconceptions. <blockquote> <p> By the way, it's not really correct to say that the IPCC models assume a positive feedback. They calculate a positive feedback. </p> </blockquote> Ok - someday I would like to learn more about how the models do that... <blockquote> <p> And there is likely a lot of state-dependence involved too with such dramatically different physical situations, so extrapolating to modern-day climate feedbacks is problematic. </p> </blockquote> That's been my first thought, too - it seems to be very naive to assume a constant feedback over a range of such diverse states. (And of course it seems to be even more naive to write headlines like "proof of..." based on this article.) <blockquote> <p> One of the blog comments cites a recent review article. </p> </blockquote> The link takes me to a login screen with no hint of the authors and the title of the article :-(
• Options
3.

The link takes me to a login screen with no hint of the authors and the title of the article :-(

Try this: Feulner "The faint young Sun problem" Review of geophysics 50 (2012)

Comment Source:> The link takes me to a login screen with no hint of the authors and the title of the article :-( Try this: Feulner "The faint young Sun problem" [*Review of geophysics*](http://www.agu.org/pubs/crossref/2012/2011RG000375.shtml) **50** (2012)
• Options
4.

Tim,

Briefly, they have thermodynamics and radiative transfer in them. A warmer world changes the evaporation-precipitation balance to keep more water vapor in the atmosphere (thermodynamics), which is a greenhouse gas (radiative transfer). From this you can calculate the water vapor feedback.

That's been my first thought, too - it seems to be very naive to assume a constant feedback over a range of such diverse states.

The author actually considers a forcing-dependent greenhouse effect, but doesn't seem to consider time- or state-dependent feedbacks.

The link takes me to a login screen with no hint of the authors and the title of the article :-(

arXiv preprint

Comment Source:Tim, > Ok - someday I would like to learn more about how the models do that... Briefly, they have thermodynamics and radiative transfer in them. A warmer world changes the evaporation-precipitation balance to keep more water vapor in the atmosphere (thermodynamics), which is a greenhouse gas (radiative transfer). From this you can _calculate_ the water vapor feedback. > That's been my first thought, too - it seems to be very naive to assume a constant feedback over a range of such diverse states. The author actually considers a forcing-dependent greenhouse effect, but doesn't seem to consider time- or state-dependent feedbacks. > The link takes me to a login screen with no hint of the authors and the title of the article :-( [arXiv preprint](http://arxiv.org/abs/1204.4449)
• Options
5.

FWIW, I posted an answer on the forum to an article about the latest CO2 measurement in the arctic, record CO2 concentration (German), partially using my own words and Nathan's. It's the most prestigious German weekly newspaper, and they have very strict moderation rules: Offending comments get deleted quickly.

So I think it is worthwhile to engage from time to time over there.

Comment Source:FWIW, I posted an answer on the forum to an article about the latest CO2 measurement in the arctic, <a href="http://www.zeit.de/wissen/umwelt/2012-06/treibhausgas-konzentration">record CO2 concentration</a> (German), partially using my own words and Nathan's. It's the most prestigious German weekly newspaper, and they have very strict moderation rules: Offending comments get deleted quickly. So I think it is worthwhile to engage from time to time over there.