Friday, March 31, 2006

RealClimate article on climate sensitivity

There's a nicely-written puff piece by Gavin about our recent GRL paper here on RealClimate. I've added a couple of brief comments here and here. Virtually none of the readers' comments actually touched on the content of our paper...


coby said...

Well, one fellow ask this:

"I am a little unclear as to what exactly is included in terms of feedbacks. I would assume that a prediction of 3oC warming for 2x CO2 includes at least H2O feedbacks. Maybe it includes ice albedo feedbacks. It probably does not include methane from melting permafrost. Or I suppose it takes the view that it does not matter where the GHG's come from, such feedbacks of CO2 or methane only mean we get to 2x CO2 faster.

In short is it the conclusion that if one snapped one's fingers and doubled CO2, at the end of a few decades it would be 3oC+/-? What about the longer timeframe for ice sheet response and CO2 ouitgassing from the oceans etc?"


and got no answer...that was about the paper.

James Annan said...

Sorry Coby,

I'll reply here cos no-one else will be interested (actually, I thought Gavin more-or-less dealt with it somewhere) :-)

Your "in short" is basically correct. The standard calculation includes water vapour and sea ice feedbacks, but not major ice sheets (too slow). If we are going to take account of CH4, vegetation changes and ocean uptake then we've really got to include a full carbon cycle and then the CO2 level becomes a function of emissions rather than a prescribed boundary condition - so "sensitivity to doubled CO2" ceases to have much useful meaning. But on the global 100 year time scale, all these effects are thought likely to be small - the carbon cycle can be persuaded to be significant (not dominant) if you try hard enough, but even the sign is not clear. So our working definition is a reasonable approximation to what we think are the dominant effects on the multidecadal time scale.

Alastair said...

Well I'm interested, and I expect there were several lurkers out there who would have enjoyed hearing it from the horse's mouth :-)

BTW, you should count yourself lucky that you can choose where you post! My replies to that thread do not appear. Not only does that mean that David Donavan's erroneous view is the only one braodcast, it also seems as if I have no answer to his arguments. No doubt this further "defeat" of my science will be cast up against me if I ever dare to raise the matter again :-(

But back to your paper, and climate sensitivity. I can understand that it is the forcing for a doubling of CO2, and since the forcing is believed to be logarithmetically related to CO2 concentration then a base point is irrelevant. However, how do you get from, say, the concentration in aerosol from an eruption such as Mt. Pinatobu to concentration in CO2?

The water vapour forcing from Mt Pinatubo was investigated by Soden et al. in Science 2002, but in the accopanying perspective;296/5568/665 Del Genio writes:

"A few cautionary notes are in order. Although the agreement shown by Soden et al. is good, volcanic eruptions are not perfect reverse proxies for greenhouse gas climate change. Volcanic aerosols affect incoming solar energy more than they do Earth's thermal radiation, whereas the reverse is true for greenhouse gases. Volcanic forcing decreases more from equator to pole than does greenhouse gas forcing. And both types of climate change reduce the rate at which temperature decreases with height from the surface to the upper troposphere, even though one is a global warming and the other a global cooling."

How is it that you can calculate the CO2, H2O, and sea ice forcing when they are having trouble with H2O alone?

Cheers, Alastair.

James Annan said...


The volcanic effect relies quite heavily on models, and of course there is some uncertainty associated with it. However, the whole point of our argument is that you don't need a highly precise estimate from one single line of evidence - rather, by combining different estimates (ie data sets) we get a better answer than any individual data set could provide.

Alastair said...

I also understand the idea behind Bayesian logic, that the more evidence you collect, the more accurate is your result. But I don't see how the effects of volcanic aerosols can tell you about the effects of greenhouse gases. If you were combining the effect of the Eemian to LGM change in temperature ppm of CO2 with the effect of the Younger Dryas to Holocene change in temperature ppm of CO2 then that might have some validity. But how can you use aerosols to tell you about CO2. I presume you convert the aerosols to forcing via climate models, then convert the equivalent CO2 forcing via models to a rise in temperature. Is that correct?

If so, can I point out to you that yet another problem has been found with the models. See I suppose you can take comfort in the fact that William has managed to put a spin on the story so that it does not appear to be a bug. Just a feature of the science!

Cheers, Alastair.

coby said...

Thanks for that James. This is definately one of those things that is too subtle for media sound bites.

"But on the global 100 year time scale, all these effects [carbon cycle feedbacks] are thought likely to be small"

I suspect this aspect is poorly understood and clumsily modeled rather like ice sheet dynamics. If we rely on the ice core history, shouldn't we expect significant GHG feedbacks with even only a couple of degrees C warming? Though I suppose the issue of latency is quite hard to judge.

James Annan said...


"significant" feedback is certainly possible. There are competing effects and no-one is quite sure how they will play out. But remember that 100ppm of CO2 has a substantial forcing effect going from 180 to 280, but a much lesser effect going from 550 to 650 (and note further that the 100ppm feedback is associated with a 6C warming from LGM to present day).


Please go away and read the references in the paper.

coby said...

James, I confess I need that reminder from time to time. It does seem likely that whatever mechanisms were in play in the carbon feedbacks of the glacial cycles could kick in full force and not change the picture all that much (3oC becomes...4?).

My impressions of the effect of serious clathrate releases is that they are potentially much worse, but, another confession, that is entirely from journalism not journals.

James Annan said...

It's not even that much - only 0.7C extra for the numbers I quoted. So perhaps not entirely negligible, but definitely modest.

Clathrates are another matter - but it seems like all the informed opinion is that they are unlikely to generate a disaster (even though the potential may in principle be there). RC did an article on that - .

EliRabett said...

Wanna bet on that:) or maybe :(

Anyhow, it is an extreme downside risk

James Annan said...

In case anyone missed the linked full stop(!), here it is again.