Sunday, December 23, 2012

Can the Last Glacial Maximum constrain climate sensitivity?

You've had the appetiser from me, now for the main course from jules...

For many years now, a lot of researchers (ourselves included) have been working under the assumption - or at least hope - that models which better simulate the past are likely be more reliable for the future.

This assumption was dealt a bit of a blow by Michel Crucifix's paper in 2006, when he showed that, among the 4 GCMs for which outputs were available, there was no relationship between their sensitivity to the negative forcings of the LGM state, and their response to increased CO2. This suggests that any attempt to constrain the models directly according to their LGM simulations seems doomed to failure. However, Michel only had 4 models available to him at that time, and it would be hard to show significant results (the r-squared of a correlation would have to be 0.95 or higher to be statistically significant). Anyway, research continued, since paleoclimate simulations remain the only real opportunity to test the models' ability to simulate the large changes in climate that arise from large changes in forcing.

At the workshop we attended in Hawai'i early this year, there was talk of trying to write a review - or more accurately a "preview" - paper about the potential for using past modelled climates to constrain uncertainty in future climate in the context of the upcoming new multi-model climate ensemble (PMIP3/CMIP5). It occurred to me that people hadn't looked much at the spatial pattern of past and future correlation in the PMIP2/CMIP3 ensemble. We did this for our little MIROC ensemble a few years ago, and had seen quite a strong latitudinal variation, but I had been put off from doing something similar in PMIP2 by the slight inconsistencies between the model versions used for the past and future in PMIP2/CMIP3. I originally planned to wait until we had results from PMIP3/CMIP5 which should be a much larger and more consistent ensemble. However, for a preview paper I thought we could give it a shot with the old models, just to see what it looked like. There was the added incentive that writing the preview based on those models meant that in the future, when PMIP3 is complete, we will be able to see how totally wrong we were.

So, my approach was to bin the outputs of the PMIP2 simulations for the Last Glacial Maximum onto a 10x10 degree grid, correlate these local temperature anomalies with the models' climate sensitivties and hope to see a nice big correlation in the tropics where the dominant forcing for the LGM is due to GHG changes. And that's what we got. Only we also got a weird correlation the other way round in the southern ocean! That we don't understand, but at least it explains why there is still no correlation on the global scale despite the larger ensemble (7 models). In fact, when we looked again more carefully at Michel's paper, we were reminded that he had also looked at the tropics. He had compared past to future tropical changes which isn't quite the same as our analysis (we don't have spatial maps of future temperature for all models), but although his results do look weakly positive, his ensemble was too small for the result to be significant.

I've not really been so interested in climate sensitivity since we settled the matter to our satisfaction several years ago, but the rest of the world has been slow to catch up. So, it is natural to use this relationship, together with James' new LGM temperature estimate, to generate a sensitivity analysis. There are two fairly standard methods that people have used to do this.

One approach is to use the linear regression (and its associated predictive uncertainty) to directly map the observationally-derived temperature estimate (for tropical temperature) to climate sensitivity. This is basically what Boé et al did for sea ice. An alternative approach is to weight the GCMs according to how well they match the data. Both methods have their strengths and weaknesses, but in practice it doesn't seem to matter that much in this case. Our analyses point towards a moderate climate sensitivity of about 2.5C with a 90% range of about 1-4C, though there are some significant caveats in these results, which we hope the paper makes clear enough.

At some point in the summer, one of the IPCC authors contacted us to ask if we had any new papers on climate sensitivity. Meanwhile, the "preview" paper seemed slow in progressing, as these things tend to be. So, although our analysis was originally intended just as a modest constribution to that larger piece of work, we wrote it up as a stand-alone paper, submitted it to GRL, and somewhat to our surprise it sailed through review with only minor revision.

It's just been published on line, and is open access (which means everyone who is connected to the interwebs can read and download it). Alas, this is not due to any policy about-turn from the AGU, it's just that we had budget to burn.


Steve Bloom said...

Appetizer twice the size of the main course? What sort of restaurant is this anyway?!

James Annan said...

We don't expect anyone to actually finish them....cross-pond influences creeping in perhaps :-)

EliRabett said...

Could the mechanism for breaking out of the LGM set forth in Shakun, et al explain the result about the southern oceans, e.g. that blocking the flow of energy from the tropics to the Arctic by slowing the AMOC increases warming of the Southern Ocean, etc...

Steve Bloom said...

I was continuing to root through the AGU FM abstracts and came across this from Christina Ravelo et al. (paragraphed for easier digestion by dyspeptic elderly bunnies):

"The response of climate to past changes in atmospheric greenhouse gas composition can be used to assess Earth System sensitivity. Unlike Charney climate sensitivity, which is related to the strength of feedbacks involving short timescale climate processes such as those involving clouds and water vapor, Earth System sensitivity also integrates feedbacks involving long timescale changes in the cryosphere, terrestrial vegetation, and deep ocean circulation.

"We show that paleoclimate data from the Miocene, Pliocene and Pleistocene epochs provide evidence for changing sensitivity through time, probably due to changing boundary conditions due to tectonics. In the middle Miocene (~18-12 Ma), major climate change trends appear to be coupled to pCO2 changes; however, in the late Miocene (~12-5 Ma), climate was warmer than today even while pCO2 was similar to today, indicating a decoupling between long-term climate evolution and pCO2 change.

"In the Pliocene and Pleistocene, there appears to have been strong coupling between climate and pCO2 changes; there is a wide range of Earth System sensitivity values, all of which exceed or are at the high end of Charney and Earth System sensitivity estimates derived from climate models. In the early Pliocene warm period, pCO2 was 350-400 ppm, implying an Earth System sensitivity (temperature change for a doubling of pCO2) of 7-9C. Subsequently, large Northern Hemisphere glaciations began to occur when pCO2 dropped below about 300 ppm in the middle Pliocene. For the Late Pleistocene ice ages, estimates of tropical sensitivity are approximately 3-4C for a doubling of pCO2, which we suggest represents a minimum value for global sensitivity during the last ~500,000 years.

"Overall, paleoclimate data have important implications.

"First, tectonic changes in basin shape impacted ocean circulation and climate sensitivity through the Miocene and possibly the earliest Pliocene, indicating that the initial conditions of the mean oceanic state plays a role in the climate response to pCO2 change.

"Second, to explain why early Pliocene estimates of Earth System sensitivity are so high, it may be important to improve how the large observed changes in upper ocean circulation are simulated by models.

"Third, in a period when ocean basins were similar to modern, ice age climate sensitivity to pCO2 changes is underestimated by climate models even when long term changes in solar forcing and ice sheet size and distribution are taken into account, implying that internal positive feedbacks are stronger than previously thought."

The first implication (boundary conditions are important) AFAIK is uncontroversial, although from reading elsewhere there seems to be an interesting idea, no entirely worked out as yet, about a sticky thermocline having kept the Miocene from responding to the decrease in CO2.

The second implication seems to be that a step change in Earth system sensitivity (about double), consequent to ocean circulation changes, can be expected pretty much any time now, or perhaps has already happened if there has been an irreversible commitment to the circulation changes. A big, big deal, I would say.

The third implication speaks for itself, pretty much. Are all models indeed wrong?

What say the proprietors?

James Annan said...

The proprietors are a bit puzzled, because they saw Ana Ravelo give a very similar presentation :-)

It was what I was referring to when I mentioned `a rather silly presentation on "changing climate sensitivity" which IMO just pointed to the inappropriateness of trying to analyse all historical climate changes as if they were a response to an imposed CO2 forcing, which of course they were not.'

Or to put it another way, there is no reason to expect global mean temp to be linearly related to CO2 forcing, when all sorts of other things have changed in nonlinear fashion over these long time scales.

And none of it points to any sort of "step change" in the sense that you seem to mean it. These things happened over literally millions of years.

Alex Harvey said...

Dear Jules and James,

Thank you for making these papers publicly available.

I am interested to know the reason for not using the 'paleocalibration' method of estimating climate sensitivity. (I.e. just compute climate sensitivity directly from G = delta T / delta Q.)

Sorry if you have said this somewhere and I have missed it.

Usually it is said the problem with paleocalibration is that it is hard to know the true value of the LGM cooling.

But given that these new papers have claimed a newer and more accurate value of the LGM cooling, isn't paleocalibration the ideal method to use here?

Best regards,
Alex Harvey

James Annan said...

HI Alex,

It's a reasonable question, but the straightforward answer is that we know that the sensitivity to a range of different positive and negative forcings varies (it certainly does in models, therfore presumably also does in the real system) and so cannot be so easily summarised in a single number. Specifically in this case, the effects of increase ice sheets and decreased GHG changes do not combine in a purely additive manner. So while the approach you present (which has of course been widely used) is not a wholly ridiculous approximation, neither is it wholly adequate.

James Annan said...

In fact I should expand a little, to say that the main point of Michel Crucifix's paper was that there was no identifiable relationship between past cooling and future warming across the models that he considered. So our paper (which is certainly not meant as a criticism of his work, more of an update and hypothesis for the CMIP5 models) really has to address that point head on in order to be taken seriously.

Steve Bloom said...

Maybe she should stick with A.C. to avoid confusion. :)

I want to think about this more, and I'll read Michel's paper (and the PALEOSENS paper on paleoclimate sensitivity) before saying much else, but I'll say now that part of my confusion has been trying to think of sensitivity within the traditional units of X degrees per CO2 doubling. It sounds as if that can't be right since sensitivity at the start of a doubling (or halving for that matter) will have changed long before climate can equilibriate with the doubling or halving. To that I had added the idea that Charney and Earth System sensitivity (the latter being a fairly new concept in the literature, right?) could be thought of as on somewhat independent scales, but that's not right either. I think I now have a clear picture of the concept, but I suspect confusion over it is widespread.

Re my suggestion of a step change, I was indeed thinking of an ocean circulation organization happening fairly quickly, potentially faster than ice sheet melt, and nothing like millions of years. That too I will think about more.

A question to close: Has anyone attempted to plot a sensitivity curve on a long time scale (i.e. from LGM to an equilibriated ice-free climate)? It would be interesting to see what that looks like.

Thanks for the help.

crandles said...

LGM was colder and there can be more ice retained over land. North has lots of land where ice can expand so albedo changes is a large forcing and this can cool oceans in northern hemisphere which is both limited in extent and between cold land.

In the southern hemisphere by contrast there is little land the ice can extend onto - Antarctica is already fully covered. There can be some expansion on southern tip of South America but this is a tiny bit of land surrounded by lots of water. So it is hard to see much albedo change and without a large forcing, there isn't much reason for big change in size of ice shelves.

There are other forcings - dust orbital cycles etc. What should these be expected to do to areas around Antarctica?

I guess that just asserting that high sensitivity models being associated with a large temperature rise in southern ocean is just these high sensitivity models showing themselves to be unlikely is rather too hand wavy.

Alex Harvey said...

Dear Jules and James,

Fair enough. I also read Michel's very interesting paper and understood the problem (well, roughly).

But a recent paper that used paleocalibration was Koehler et al. (2010, QSR). In this paper the issue identified in Crucifix, 2006 is handled by adding a 'scaling factor'. I guess you would argue that this solution isn't adequate either?

The authors write,

"A recent study (Hargreaves et al.,
2007) with a general circulation model indicates that the LGM
sensitivity to CO2 is likely to be smaller than the sensitivity of the
current climate state, although the model spread is large (i.e. in
some model versions the LGM sensitivity was smaller and in some
versions it was found to be larger). The results by Hargreaves et al.
(2007) indicate that the LGM sensitivity is on average about 15%
smaller than for 2 x CO2 climate (see their Fig. 5), and we therefore
use a best guess of 0.85 and a standard deviation of 0.2 for the
scaling factor."

They use 5.8 +/- 1.4 K based on Schneider von Deimling et al., 2006 for the LGM cooling. And in this way they find an ECS best guess of 2.4 K - very close to the value you've obtained using a validated models method.

However, the difference between the LGM cooling used in Koehler and in the present study is, of course, large. That's primarily why I am puzzled.

(Is the value of the radiative forcing you have used the same as that obtained by Koehler et al.?)

If so, is it therefore implied by your study that the LGM sensitivity is actually even *more* than 15% smaller than the 2xCO2 sensitivity of the present climate, as Koehler assumed?

Kind regards,

James Annan said...

Alex, well at first thought, one could interpret our analysis as meaning that the overall sensitivity is lower than previously thought, or that the ratio between past and future sensitivity is different to what was previously thought.

I think you can do the LGM energy balance argument (in fact our GRL paper contains this calculation), so long as you realise that this number doesn't directly indicate the response to a CO2 rise. The Kohler et al approach is one possible way to address it.

James Annan said...

Chris, the Schmittner et al paper had an analysis of the dust forcing, or perhaps cited someone else...I don't think there is a big effect over Antarctica, dust could darken the surface a bit but there is not a lot of sun and no melting anyway...

Steve Bloom said...

I'm still reading other stuff as mentioned above, but just wanted to note that Matt Huber has an open-access climate sensitivity perspective piece in the latest issue of Nature Geoscience.

Timothy Chase said...

From the main essay:

"It's just been published on line, and is open access (which means everyone who is connected to the interwebs can read and download it). Alas, this is not due to any policy about-turn from the AGU, it's just that we had budget to burn."

You might want to ask for your money back. At least for me, AGU is insisting on an email and password.

James Annan said...

Thanks, I suspect a snafu due to the Wiley take-over. (It disappeared completely for a day or two, as did some other papers.) Jules will email them to give them a kick up the arse.

There's a copy on my work web page for now...

Timothy Chase said...

Thank you!

James Annan said...

Open access seems to be working again now.