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'iearly 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.
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.
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 Boéet 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 paperson 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.
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
ago, when we first got involved in paleoclimate research and were
thinking about using the Last Glacial Maximum to constrain models, jules asked a respected researcher for an estimate of how cold
the LGM was, in terms of a global annual mean temperature anomaly
relative to the modern pre-industrial climate. Their surprising response
was, "I haven't a clue". Of course this wasn't quite true, as the
person in question was surely confident that the LGM was at least a
degree colder overall than the preindustrial state, but not as much as
(say) 20C colder. But they weren't prepared to specify a global average
value, because the available proxy data were few and far between, and
could not be considered to give good global coverage.
we looked at the literature, we found lots of analyses which looked at
proxy data on a regional basis (such as SST in the tropics, and various
polar ice cores) but not much relating to a global average value. Some
people (including Jim Hansen) have given rough estimates of about 5-6C, and we
used similar arguments to give an estimate of 6C when we needed one for
But it was a bit hand-wavy. As for modelling results, in 2007, the IPCC
AR4 gave an estimate of 4-7C colder than pre-industrial, which was
based more-or-less directly on GCM simulations from the PMIP2 project.
Around the same time, various people (including us here) started more formally combining models with proxy data, by running model ensembles with different parameter values, and came up with similar estimates (e.g. hereand here). So, a consensus view of around 5-6C as a best estimate seemed well established.
Since then, several major compilations of proxy data have been published (most notably, MARGO for the ocean, and Bartlein et al
for land), and although the MARGO authors did not generate a global
mean estimate, there were strong hints that their data set was a bit
warmer than the previously prevailing interpretation of ocean proxy
data, especially in the tropics. When Andreas Schmittner and colleagues
published theirScience paper last year, their climate
sensitivity estimate made the headlines, but it was actually their LGM
reconstruction that was more immediately eye-catching to us. They fitted
a simple(ish) climate model to these most recent and comprehensive
proxy syntheses, and came up with a global mean temperature anomaly of
only 3C (with an uncertainty range of 1.7-3.7C), which is far milder
than most previous estimates. Irrespective of the resulting sensitivity
estimate (which we'll return to later), such a warm LGM would be hard to
reconcile with GCM simulations. Therefore, we thought it would be worth
considering the LGM temperature field as a stand-alone problem. One
obvious weakness of Schmittner et al's paper is that due to
computational limitations, they only used a fairly simple climate model,
which didn't seem to fit the data all that well.
Our main idea was to see if we could do a better job using the state of
the art GCMs. That requires a rather different approach, as we can't
run large ensembles of these models.
Our resulting paper is still under review at CP, and available here. It has had useful and positive reviews, and I'm now revising it, but I don't expect the overall answer (which is an LGM cooling of 4.0+-0.8C) to change.
Our main result is based on a multi-model
pattern scaling approach, based on a method I'd seen in the numerical
weather prediction literature. It's actually just a simple multiple
linear regression. The idea is that each model (from the PMIP2 database)
is assigned a scaling factor in order that their weighted sum
optimally matches the data (and hopefully, interpolates meaningfully
between these points). This gave substantially better results than
either of the other two methods (which I'll describe below for the
interested reader), and also worked well for a number of validation
tests. The resulting reconstruction fits the data rather better than the results that
Schmittner et al reported. The resulting estimated temperature anomaly
is a nice round 4.0+-0.8C, with this uncertainty representing a 95%
confidence interval. So that is a bit colder than Schmittner et al
estimated, but also substantially warmer than most previous estimates.
It's towards the low end of the range of GCM results, though with
significant overlap. Our formal uncertainty range doesn't take account
of the possibility that someone might come along in a few years and
decide that these proxy data are all misinterpreted or hugely biased. For
example, some people argue that the MARGO ocean data are too warm,
especially in the tropics. We don't have a particular position on that,
but intend to report some more sensitivity analyses in the final
paper. Our analysis is just based on
the need to interpolate into data voids, and the fact that different
models suggest somewhat different interpolations.
And here is what you have all been waiting for- pictures of our reconstruction of the LGM anomaly, which are probably worth more than the 1000 words of this blog post. First, the surface air temperature anomaly, and then the sea surface temperature anomaly. These match closely over the open ocean, but at high latitudes the layer of sea ice insulates the ocean and allows them to diverge.
The dots are the proxy data points. Overall, the main features are very much as expected. Some of the smaller details are probably just noise, like the apparent "hole" in the top plot in the Southern Ocean at around 70S 30W, and also the N Pacific. The method generates uncertainty estimates which are particularly large in these regions. The slight warming in the Arctic SST (under thick sea ice) is in accordance with the proxy
reconstructions, that in the south is more speculative as there are no data there
What this means for climate sensitivity, is left for a future post (some readers may be ahead of me at this point)...
Appendix: For anyone who is interested, here's a quick summary of two other methods we investigated. The paper has a fuller description.
first tested the simplest method we could think of: direct smoothing of
the data, as is commonly performed for modern temperature data by the
likes of GISS and NCDC. However, the paleoclimate data are much more
sparse and noisy than modern data, and the results looked pretty messy.
Worse, when we tried sampling "pseudoproxy data" from GCM results (i.e.,
taking data from them in the same locations as the real proxy data) and
smoothing into fields to see how well we could recover the whole field
of GCM output, the results from this process had a huge overall bias.
The basic reason for this is that there are no proxy data available in
the areas which were covered by huge ice sheets back then but which are
now bare (particularly North America and Scandinavia), and it is these
regions which exhibit the largest anomalies for the LGM. So although we
could smooth the real proxy data and obtain an estimated global mean
temperature anomaly of 3.2C, we also knew this could easily be biased by
a degree (or more, or less) compared to the real temperature.
Bias-correcting this (based on the results
generated by the pseudo-proxy experiments) gives an estimate of
4.1+-0.9C cooling. An ugly feature of the smoothing is that the spatial
pattern generated is completely unreasonable, not showing any of the
extreme cooling that must have occurred (simply due to direct albedo and
altitude effects) over the large ice sheets that were in existence
So, we quickly concluded that we needed to use a model of some sort to
interpolate between the data points. As an alternative to running an
ensemble of simple climate models, we decided to use the state of the
art PMIP2 model results which were generated by essentially the same
full complexity GCMs which also contributed to the IPCC AR4 (CMIP3
project). We obviously
weren't able to re-run ensembles of these models with different
parameter values, so instead, we just used a simple pattern-scaling
approach to fit them to the data. We think that this is pretty much
comparable to changing the sensitivity parameter of a simple model -
opinions seem to differ on to what extent this is true, but anyway it's
all we could reasonably do with these models.
The results were markedly better than we got from the smoothing, but still rather disappointing. One minor curiosity, which we had not anticipated but which
is obvious with hindsight, is that this pattern-scaling approach tends
to lead to an underestimate of the overall temperature change. This is
essentially due to regression attenuation which interested readers can
google for more info (or look at our manuscript). Anyway, after bias correction this gives an overall estimate of about 4.5C of cooling, but the uncertainty is actually a bit higher than for smoothing, because different model patterns interpolate into the data voids differently.
So these two methods generate results which are compatible with our main result, but which are somewhat inferior.
Oh OK. I give in. But most of them weren't in Yellowstone... and see how few of them are bears!
Elk (there are plenty in Yellowstone, but this one was just north of the park entrance):
Mountain goat (Glacier Park):
Mummy rare black-tailed prairie dog (Devil's Tower):
Kiddiwink prairie dog:
Big Horn Sheep (The Badlands):
Mummy deer (Boulder garden variety):
Kiddiwink deer (it actually did have 4 legs):
Tourist and moose (Lake MacDonald Lodge, Glacier Park):
We saw a live moose in the distance just outside Glacier Park. Unlike the wolf, it was easily visible to the naked eye, but again there was someone there with a scope who kindly let us get a closer look. Pronghorn deer are everywhere in Wyoming, but they are small and springy and I didn't try to photograph them.
-- Posted By Blogger to jules' pics at 12/21/2012 10:00:00 AM
James has promised his parents a trip to Yellowstone. Weirdly this has resulted in them sending us lots of bears for Christmas and birthdays. It seems important, therefore to point out that, apart from buffalo, wildlife can be elusive. For example, see the wolf, 4th photo down, here.
We had better luck with the mountain goats further north:
(Those prairie dogs in the distance are rare black tailed ones!)
When you do see a bear, usually it is too quick to hardly even notice, but sometimes it looks like this:
-- Posted By Blogger to jules' pics at 12/20/2012 10:00:00 AM
Gingerbread men weren't morbidly obese when I were a lass (in the 1970s). We shared him, of course.
I don't understand US food portions. I suppose the idea is that people support the economy by eating lots of food, and then they support it even more paying for the resulting treatment to lose the weight gained. In Japan the portions are all much smaller, which means that even if you go to a restaurant alone you can sample a pleasant variety of food for your meal, whereas in the US one is in danger of exploding before finishing half a portion. Because it is not necessary in Japan to pay someone to make you thin again, it makes sense economically that the food is much more expensive.
-- Posted By Blogger to jules' pics at 12/19/2012 08:24:00 PM
From the department of complete lack of surprises: the LDP has apparently won the Japanese election. It seems that 3 years of DPJ rule was more than enough, and the people want the real Govt back again. The new PM is one of the dismal failures who briefly led the country about 6 years ago, back for another bite at the cherry. But don't worry, if you don't like this leader, there'll be another one along soon enough.
Main manifesto promises are: even more concreting over the countryside, and increased patriotism. Let's beautiful Japan!
Sorry for the delay, but I was too busy having fun at the AGU to blog about it. For some reason, we had no real jet lag at all on our trip, either while over there or following our return to Japan. I don't really have much of a clue why it sometimes works out like that, but it's nice when it does. Also, we've really sorted out the food situation in SF. There are lots of good sandwich shops for lunch, which is a much better option than trying to fight your way in and out of the overcrowded restaurants during the lunch break. Of course the huge range of restaurants is good for dinner, even though one unexpected side-effect of discovering possibly the best pizza in Tokyo is that the best pizza in SF is not so special any more. San Francisco is a good city for a huge conference like the AGU. The weather was mostly nice this year, apart from the fun run which happened to be scheduled just perfectly for the few hours in the week when it absolutely tipped down.
The scientific program seemed better than usual, too. First day, I started off with some last millennium stuff. The convenors had generously invited me to talk about this work, but I thought it would be appropriate to decline on the grounds that I hadn't done anything new since I talked about it at a very similar session (albeit with different convenors) the previous year! There was actually someone doing rather similar data-assimilation-y things, which was interesting. I'm not sure if and when I'll have time to work on this again. I think jules was learning about ice sheets at the same time.
I had a walk round the posters after lunch. This is still a bit of a disaster area as far as I'm concerned. The "author in attendance" period is supposed to be a self-selected hour within a 4h:20 time window, so it's almost impossible to meet anyone without prior arrangement. At the similarly vast (well, half as big) annual EGU meeting, there is a 90 minute early evening slot where - at least within the Climate division - there are no oral presentations scheduled at all, so all the poster authors are present together and it's an excellent opportunity for (wine-enhanced) discussion and mixing.
Last session on Monday was on atmospheric feedbacks, which in practice included climate sensitivity stuff. There was 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. The atmos feedbacks continued into the following morning, after which I went to the "communicating climate science" session. That was the expected knockabout stuff, though unfortunately Dan Kahan (who would probably have had the most scholarly perspective) withdrew. Naomi Oreskes talked about her theory of "erring on the side of least drama" which I didn't find completely convincing. Now I've seen the paper, her analyses of scientific underestimation seem rather less compelling than she made it sound in the talk. I should come back to this in a later post. Jules tried to get into the Charney Lecture, but the room was too full. However, it is available on-line. I left a bit early in the afternoon, as we had an important dinner date with some lobster, quail, lamb, steak and cheese. There may be Tokyo restaurants to match Gary Danko, but not at that price.
Wednesday morning was the not-so-much-fun run. Actually, it wasn't that bad running in the rain for once - but I suspect the novelty would quickly wear off - and least it wasn't all that cold. Drying off back at the hotel caused us to miss the start of the two morning Uncertainty Quantification sessions, which was a bit of a shame as this had several interesting-looking talks and is not available on video. After lunch I went to Stephan Lewandowsky's "Construing Uncertainty" which was more social than hard science, including Oreskes again. Then jules was talking in the last UQ session, on the use of paleoclimate simulations for testing and validating models. Lenny Smith also gave a very good and thought-provoking talk as always. This meant I missed the Lorenz lecture, given by Eugenia Kalnay, but I knew it was being videoed, and have watched it subsequently. I highly recommend it to anyone interested in the theory and practice of how prediction is done in highly complex and nonlinear systems such as weather forecasting. (The preceding lecture included in the same video may not be so widely interesting.)
Charles Jackson also talked in one of the UQ sessions, and that night we went out to dinner with him (and two Japanese colleagues) afterwards at the Thirsty Bear. Once the shouty early-evening crowd have moved on, it's a surprisingly civilised place for good pub grub.
Thursday morning was a bit thin in the program, and I believe we might have had a bit of a lie-in after the night before :-) The afternoon had a big paleoclimate session, with several people each talking about their own latest core and its interpretation (or so it seemed to me), and some broader syntheses including my presentation on the Last Glacial Maximum. I will also blog about this in more detail, after doing what I hope will be some final revision to the paper.
After a fairly packed week, (and contrary to the previous year) Friday was almost a complete blank. Or at least it would have been, had Lewandowsky et al not conveniently arranged an off-site discussion meeting which followed up on their "construing uncertainty" session. I'm always a bit dubious of these talking shops, and I'm not sure exactly how much real progress was made, but it was surprisingly fun and interesting, and it was good to have the chance to discuss and expand on ideas in a much more flexible and informal format than the tightly scheduled AGU sessions. Attendees included a large number of the usual blogging suspects (Rabett, Nielsen-Gammon, Sceptical Science, Easterbrook, Baer, and probably one or two more I've forgotten), though Michael Tobis was a surprising absentee. One point of interest was the broad consensus among the climate scientists present that things like Hurricane Sandy and other extreme events were not the most sound basis for convincing people about the reality and/or impact of climate change. Although it's plausible that many extremes will get more extreme, the evidence (both theoretical and observational) for many specific examples (such as, changes in the frequency and size of storms hitting NY) is not really that solid yet.
It was noticeable how much quieter the main conference centre was by the time we returned to it after lunch. Not many of the participants stay to the bitter end. We spent the afternoon in a cryosphere session. I suppose it is well-known within the field, but it was interesting to see to what extent the ice sheets vary even within the supposedly "quasi-equilibrium" state of the Last Glacial Maximum. I doubt it has major effects on broad scales, but certainly helps explain how some people can find historical evidence of vegetation in locations where the models have big fat ice sheets :-) (This issue only arises close to the the edge of the big fat ice sheets, to be fair.)
And this year, after the weekend of travel back home, we did make it back in time for the carols in Yokohama, followed with some mince pies that I made earlier.
There's a lot of good stuff available as video on demand. We are slowly working through it in idle moments. Maybe it's not quite as good as being there, but it's a lot cheaper and involves a lot fewer air miles. Jules' plan of world domination through the VGU (Virtual...) is getting closer year by year.
Frankly, I'm surprised it took so long. And since the review period is over and the next round of revision is probably already underway, it doesn't seem particularly useful or important. For that matter, the drafts were freely available to anyone who wanted to sign up to review them anyway. Revkin think this leak means some "new process" is needed, but I don't see any real arguments presented.
FWIW, I thought the second order draft (at least, the bits I looked at) was mostly pretty reasonable, certainly improved a lot on the first draft. One thing I do find a bit unsatisfactory and unecessarily obstructive about the process is that there is no way of seeing replies to the first set of review comments, so the system is rather crippled compared to a normal peer-review process where reviewers and authors can have a proper exchange of views. This has a real impact when comments appear to be ignored - the reviewer has no way of knowing if the authors' counter-argument is strong, weak, or whether the lack of action was simply an accident. The latter happened this time, in fact, but it took a private email to a relevant IPCC author to find that out. They assured me that the small oversight would be fixed for the final version.
San Francisco has a modern art museum. It is important not to view modern art on an empty stomach. Luckily, art-priced luncheon is available on the top floor. Note the brilliant Mondrian cake.
Like a multi-floored Japanese department store, starting at the top and working down is a good tactic, as the whole experience is then a slow run for the door. But after a while one may start to confuse up and down...
Apparently this one is about uncertainty, and since this is wot we work on, I had to take a picture.
Some staved off madness by retreating into familiar worlds.
James' trousers were admired several times!
-- Posted By Blogger to jules' pics at 12/13/2012 06:22:00 PM
Unfortunately I did not remain at the chocolate sundae shop all week, while James went to hang out with the spods at the AGU. Although... perhaps I could become one of those ladies who lunch eat chocolate...
Instead we left Ghiradelli behind...
and visited some old boats from the motherland...
Artistic misty golden gate bridge with posing gull was the reward for making it to the end of the pier...
The walk back to our hotel was quite interesting. Really ought to have taken more pictures. We did, however, have the honour of being flashed by a street photographer. Is this really an advert for Dropbox?
Dinner was in Chinatown. I didn't previoulsy know that the temples are built in the upper stories.
Sorry if these pictures are awful. Work has given me a new MacBookPro and it makes all my photos look phenomenal. Hopefully some discernment will return with practice.
-- Posted By Blogger to jules' pics at 12/11/2012 03:47:00 PM
I've been remiss in blogging, the AGU has been lots of fun so far but I've not had spare time or energy to write about it. It seems to improve the more we attend, as we get better accustomed to the sessions and the city. Still not convinced it is quite as fun as the EGU, but it's well worth attending. Especially assuming Ben Sanderson's calculations are correct, the energy cost of the flights over here is only equivalent to a handful of model runs, so by coming here instead of working we are helping to save the planet :-)
Unfortunately it was too wet this morning to take cameras, so you'lll have to take our word for it, but we did the AGU 5k fun run this morning. The weather was not exactly monsoon quality, but still quite wet enough. My garmin had trouble seeing the satellites and recorded an extra 140m on the 5k course - I don't know if that is correct or not, but my time (and also jules') was certainly slow enough, at least a minute outside our rough goals though jules was still second in her age group. Mind you, we were carrying a few extra pounds of rainwater in our shoes. And delicious though it was, the previous night's dinner in Restaurant Gary Danko might not have been the ideal preparation for a dawn race. Hopefully I'll find time in the next couple of days to enjoy a sunnier morning jog along the Embarcadero. (Ah, based on a quick glance at the previous year's results, it seems that several people who did both events were about a minute slower this time too, so we needn't be too disappointed.)
Rumour has it there's an exciting talk at 4pm tomorrow in one of the paleoclimate sessions. I'll try to be there...
Following the spate of falling-off-the-seat incidents and associated legal action which has tied up the San Francisco court system for months, all dangerous restaurant seating now comes equipped with safety gear and warning signs. Fortunately, I got through the meal unscathed without so much as a wobble let alone a full-scale tumble.
(This post was brought to you by the Anchor Steam brewery.)
Not as loud as cicada, but much louder than the same number of Japanese people in the same space. Of course being able to hear others is unimportant to Americans psyche as they only need hear their own voices in order to feel normal - hearing yourself shout is just about still possible with 90 decibels background.