Well, of course it’s finished in the temporal sense. But not in terms of scientific interest. This is perhaps the main conclusion of our latest review paper, which was published in QSR a few days ago. All followers of @geschictenpost on twitter (never miss an interesting paper!) will already be aware of this, but they might not realise that the full paper is freely available via this link until midnight on the 23rd Dec at which point it will turn into a pumpkin (ok, hide behind a paywall).
The review was invited back at the end of 2012, but at the time we were in the process of publishing some relevant work so we didn’t actually get round to writing anything until half way through 2013. We had fun over the summer chasing up some old papers that we had not previously seen, and even got a paper print of a CLIMAP reconstruction – which we were not allowed to remove from the library, or even copy! Conversations with several more senior scientists were helpful, including a few emails exchanged with Tom Crowley. And a reviewer added another perspective which was very helpful.
In the paper, we trace the evolution of model-data comparison for the Last Glacial Maximum, at least for surface temperature, talking about what has been achieved (the models and data agree to some extent on the broadest scales) and what issues remain (regional differences are large, not only between models and data, but also amongst different models and different interpretations of data). The LGM remains the gold standard for testing the ability of models to reproduce a climate state very different to the present day, thanks to the relatively abundant data and large, well-understood forcing. But as well as suggesting that models are basically on the right lines, the results of the simulations also suggest there is plenty of room for improvement.
9 comments:
During LGM were Skies Blue?
Well it was certainly thought to be drier. I think the jury might be out on clouds though.
Do I sense amused smirks on your faces when you note, "These estimates have remained close to 3 °C throughout changes in estimates of both components", or am I simply inserting subtext where none exists?
I think that says more about you than us. But well done for reading that far!
I am probably searching badly, have you commented on Hansen et al 2013?
http://rsta.royalsocietypublishing.org/content/371/2001/20120294.full
Perhaps particularly
"However, we suggest that an even more fruitful approach would be a focused effort to define the glacial-to-interglacial climate change of the Eemian period (MIS-5e). The Eemian avoids the possibility of significant human-made effects, which may be a factor in the Holocene. Ruddiman [127] suggests that deforestation and agricultural activities affected CO2 and CH4 in the Holocene, and Hansen et al. [91] argue that human-made aerosols were probably important. Given the level of Eemian warmth, approximately +1.8°C relative to 1880–1920, with a climate forcing similar to that for LGM–Holocene (figure 5), we conclude that this relatively clean empirical assessment yields a fast-feedback climate sensitivity in the upper part of the range suggested by the LGM–Holocene climate change, i.e. a sensitivity of 3–4°C for 2×CO2. Detailed study is especially warranted because Eemian warmth is anticipated to recur in the near term."
3–4°C is not outside the range you suggest but is 'glacial-to-interglacial climate change of the Eemian period' data so good as to narrow the range that much?
Is that where you would like to see effort devoted rather than 'diminishing' returns from looking at LGM-Holocene?
Don't think I have commented on that specific paper, but it is probably worth pointing out that our most recent LGM temperature estimate is a bit more mild than his estimate. Also, there are pretty big assumptions (or one could say, approximations) underlying the wort of energy balance arguments he relies on. Basically, Delta T / Delta F is not really a constant parameter of the earth system, especially when the type of forcing changes.
The Eemian has a significant regional (solar) forcing, which is a complication. We were recently surprised to learn that hippo remains had been found in a cave near our house in Yorkshire...high latitudes were much warmer than now.
Kelly and Tan 2011
http://www.artsci.wustl.edu/~gradconf/ZhuoTan.pdf
concludes that
"Learning occurs more slowly if true climate sensitivity is high."
Just looking at fig 4 vs fig 5, that seems the case. However the explanatory text says
"A hight(sic) climate (sic) has two effects on the learning dynamics. First, as temperature is higher, signals are stronger, hence reduce the learning time. Second, since our prior locates further away from the true value, it takes longer to learn. The results shown here indicates that the second effect overweigh (sic) the first."
This seems to suggest to me that if climate sensitivity is lower than expected then learning would/could also be slower than if sensitivity is close to the expected level.
Am I being thick or pedantic or ... ?
My first reaction is that it's a shame that this silly fat tail stuff persists in the economics literature years after it was debunked in climate science ;-)
Thanks for your comments. I don't think my arguments have had much effect on someone claiming "The fact that the PDF for ECS has been slow to converge towards a single true value, provides support for the position that the current effective ECS maybe appreciably higher than 3C."
FWIW
The effect of learning on climate policy under fat-tailed uncertainty
In Chang Hwang and Frederic Reynes and Richard Tol
http://mpra.ub.uni-muenchen.de/53681/1/MPRA_paper_53681.pdf
looks like a better paper to me.
Although the fat tail persists much longer, it also takes some time to get the range small and the range doesn't get far too small.
Post a Comment