Monday, April 30, 2007

IPCC AR4 comments

At last, the IPCC AR4 is out - at least, most of it is (there is still supplementary material to come). Since I was unhappy with some of what was written in the previous (2nd) draft of Chapter 9, I looked at that first.

At first glance, I'm pleased to see that it has been significantly improved. The drafts were never meant to be a polished final version, and indeed were only released on condition that they were kept private (although the 2nd draft can easily be found on the web). So I'll restrict my comments to what they have agreed on for the final version itself.

Section 9.6 "Observational Constraints on Climate Sensitivity", contains the following:
"Note that uniform prior distributions for ECS [equilibrium climate sensitivity], which only require an expert assessment of possible range, generally assign a higher prior belief to high sensitivity than, for example, non-uniform prior distributions that depend more heavily on expert assessments (e.g., Forest et al., 2006)."
Many people may think this statement is too trivial to be worth making much of, but when I made essentially the same point about a uniform prior implying high prior belief in high sensitivity, Allen and Frame dismissed it as "just a rhetorical flourish". This statement from the IPCC also appears to directly contradict much of the peer-reviewed literature, which claims that uniform priors represent ignorance. It is encouraging to see that it is now the consensus of 2,500 climate scientists that this is not the case :-) Another significant aspect is the comment that even uniform priors "require an expert assessment of possible range", which at least takes a baby step towards acknowledging our point that the choice of upper bound can have a dramatic influence on the result. As far as I know, this critical detail (which undermines the whole rationale for uniform priors) does not appear anywhere in the peer-reviewed literature, although one reviewer did single it out as a particularly interesting point in one of our submissions. It could also conceivably be called trivial were it not for the fact that so many people have apparently been oblivious to it (or else deliberately deceptive in failing to mention it) for several years.

The defence the IPCC authors provide for the use of the uniform distribution is that it "enables comparison of constraints obtained from the data in different approaches". Of course this is not the same thing as generating a pdf which credibly represents the opinion of an intelligent researcher, but they don't actually go so far as to explicitly state this rather embarassing fact (which leads inescapably to the conclusion that these "pdfs" cannot be considered policy-relevant and used in decision support, eg economic analyses such as the Stern report etc). Most of the results they quote are based on uniform priors, but they hardly had a choice since this approach dominates the recent literature.

The section chapter also makes extensive reference to the "multiple constraints" argument (a significant feature of Hegerl et al's Nature paper, as well as our GRL paper), which is great. As I said more than a year ago, our calculation was rather simplistic and anyone who doesn't like it is welcome to generate their own answer, taking account of the arguments we have presented. Interestingly, I'm still waiting...

So in summary it might not be exactly what I would have written myself, but it's clearly a step in the right direction and it seems like the IPCC comment/review system has had some effect. We'll have to wait a little while longer to see what else they wrote about Bayesian estimation in the Appendix, since this is still not published. Whether this means Frame and Allen will now have the sense to slink away and pretend the whole sorry mess about uniform priors never really happened, remains to be seen.

JAMSTEC - the movie!

This film has been fairly popular in Japan, it seems - at least, our local video rental has been pushing it - so we succumbed to their temptations yesterday and rented a copy. It's called "Nippon Chinbotsu" (日本沈没) which means "Japan sinks" and is actually a remake of a 1973 film of the same name. The plot is something silly about plate tectonics dragging Japan under the sea, and I won't spoil it for those who might want to watch other than to say there are lots of volcanoes, earthquakes and tsunami. I'm not sure if and when it might be released internationally, this version was Japanese only but simple enough for us to follow most of what passed for a plot.

JAMSTEC had a substantial starring role, mostly in the form of the research vessels that have to be used to investigate the problem and try to save Japan from its fiery and watery fate. For some reason the Shinkai ("Depths of the ocean") submersibles were renamed Wadatsumi (which may mean "sea" or "Poseidon", although my dictionary only has watatsumi for those), but the massive drilling ship Chikyuu ("Earth") kept its name. I'm tempted to say something cynical about spending science budget on expensive props for trashy films but the filming probably didn't disrupt research much and the publicity probably made it well worth their while. Needless to say, it doesn't exactly represent my typical day...

Is the Earth still recovering from the “Little Ice Age”?

We were honoured by a visit from this curious chap shortly before I went off to the EGU. Some might recognise him as one of the "stars" of Durkin's swindle. Apparently he is Alaska's most famous sceptic, which sounds like praising with faint damnation to me, but never mind about that. I'm not sure why he was invited to speak, my guess is that our institute feels somewhat obliged due to our close relationship with IARC (a joint USA-Japan enterprise) which he helped to establish. He is certainly a deservedly well-respected scientist in his own field (auroral storms) but why anyone thinks this means he has much interesting to say about climate change is anyone's guess.

His argument goes something like this:
  • The IPCC says the planet has warmed by 0.6C, mostly due to anthropogenic GHGs.
  • The observed warming hasn't been exactly linear.
  • Therefore, the non-linear component must be due to other things that the IPCC is ignoring.
  • Therefore, the IPCC is wrong and GHGs have a rather small effect
Really, I haven't made this up - I don't think I could. It isn't even coherent, let alone correct. The full manuscript is on the page I linked to at the top.

He was actually very reticent in his presentation - always prefacing his comments with disclaimers about how he wasn't an expert on climate science, he was just putting forward some ideas. It's disappointing that he assumed that all climate scientists (even collectively) are as ignorant as he appears to be. He also spent a substantial portion of his time criticising the press for exaggeration - a point that the audience was keen to emphasise their agreement over, in what seemed like a transparent attempt to maintain the wa in a potentially embarrassing situation.

He had two specific criticisms of the AR4 SPM - firstly, that it didn't define "most" (in "most of the observed warming") and secondly, that it didn't show its working. In the question time afterwards, I explained that "most" was a common English word that meant "more than 50%" (at a conservative minimum) and pointed out that the SPM was merely a summary (that's what the "S" stands for): the details to justify that claim will be discussed in the soon-to-be-published (just out today, in fact) full report. All in all, it was two hours I could have more usefully spent in a number of other ways. It is often said that shame is a powerful motivating force in Japan, so hopefully his ramble will inspire some of my colleagues to displace him from his pedestal as one of Japan's most famous scientists!

Friday, April 27, 2007

Too good to be true?

Heard this on 5live this morning:
THOUSANDS of rich women were conned by a firm into believing LAMBS were valuable miniature POODLES
Entire flocks were imported to Japan from the UK and Australia then sold by the internet company as the latest “must have” pet.

The bizarre scam was rumbled when Japanese movie star Maiko Kawakami complained on a talk show that her new poodle refused to bark or eat dog food.

She showed photos of the animal and was devastated when told that it was a lamb.

Hundreds of women contacted police to say that they had also been sold lambs instead of pedigree pups by the tricksters based in Sapporo, Japan.

However, this story is nowhere to be found in the Japanese press (and The Sun is hardly one of the UK's most reliable sources), and a lot of these Japanese stories gain something in translation (eg this "hibernation" story), so I wouldn't be at all surprised if reality turns out to have been rather different. Still, it is good enough that one wishes that it might be true (like the wingnuttery I referred to yesterday).

Update
As someone mentioned in the comments, Snopes was on to it pretty quickly. However, this story has been picked up and parrotted uncritically by major media outlets all around the world, such as CNN (apparently), the Sydney Morning Herald, Ninemsn, Fox, Daily Express. Don't believe everything you read in the papers!

Thursday, April 26, 2007

Wingnuttery or satire? You decide...

In the comments to a recent post, someone pointed me to this delightful letter, as featured on Tamino's blog:




There follows much discussion as to whether it was serious, or meant as a joke. The problem is that there is so much of this wingnuttery around, it is no longer possible to tell whether people mean it seriously. For example, Deltoid finds this example of cluelessness. And Conservapedia ("a conservative encyclopedia you can trust") has surely been infiltrated by members of the reality-based community who are helping the genuine contributors to make themselves look silly.

Frequent commenter and occasional guest columnist on Prometheus, Benny Peiser (see, I can do the "framing" thing too) is another occasional culprit. When he froths that "Plato's dream of the academic elites controlling and running the world is close to being realised as a nightmarish reality", others are left genuinely confused as to whether he really does sleep in a tinfoil hat or is merely parodying himself. Being some sort of an academic, one might expect that he takes himself seriously at least half the time, but I know of no effective way to discriminate between the two cases.

I suspect that some people just throw their spaghetti at the wall, see if it sticks, and then say "oh, that bit was a joke" for any that falls. Perhaps my comment to Fergus (as quoted by Stoat, with no link or attribution...and he didn't even buy me a beer at the EGU) has some relevance here. But what about the people who do not only "not value reason" but who are actually incapable of recognising it at all?

(PS According to Snopes, the letter really is a joke. Or so the author claims now...)

Wednesday, April 25, 2007

Brian has a bet!

Check it out here. He's not got such a good deal as I did, that's for sure. But a set of high-low bets at evens centred on 0.15C/decade, and 2:1 bets centred on 0.1C/decade (with large volcanoes voiding the bets) seems sound enough to me. Making it a series of bets, with the stakes ramping up over time, is a nice way to build up trust and reduce risk.

Well done to both Brian and his opponent David for negotiating through to a satisfactory conclusion.

Tuesday, April 24, 2007

We want summer time!

I've been saying it for years, and the Japanese Keidanren (business lobby) has finally come out in support of me. I'm not convinced it will make a huge difference to CO2 emissions but it would make our summer evenings a bit longer (an hour, to be precise).

Acquitted!

This will certainly make big headlines in the UK and the English-speaking Japanese press. It's a truly remarkable verdict given the evidence about what he did, including two dead bodies, his own home videos, telephone records linking him to the deceased, etc etc.

The finger-pointing seems most appropriately aimed at a judicial system that relies almost entirely on confessions (about 98% of cases, wrung out of suspects during lengthy interrogations with no effective right to a legal defence). The effect of this is that in the rare cases in which a confession is not forthcoming, the police don't seem to have any idea how to proceed competently. I suppose we should be grateful that he got convicted of the stuff that was actually on tape.

Monday, April 23, 2007

The EGU review

I've just spent a week in Vienna at the EGU. I enjoyed Vienna more than on my previous visit two years ago, partly due to jules finding a slightly better hotel and also some good restaurants.

Austria was its usual efficient and clean self. Since many of my foreign trips are to the UK, it's good to see that some parts of Europe can put on a show to rival Japan. On arrival, the first thing we did was find jules' Dad (who was also attending the meeting and who arrived at about the same time) and wander in to town for a look around and a massive Wiener schnitzel at the famous Figlmuller restaurant. In Japan, I'm usually one of the broader customers but here I felt positively skinny. However, after a few days of Viennese meals, I was well on my way to fitting in... I doubt many of the Japanese tourists coming here develop "Vienna syndrome", but rapid-onset diabetic syndrome must be a distinct possibility.

Unlike that slacker Connolley, I didn't have time to blog during the week, so here follows some highlights from my notes. For those who are interested in more details, abstracts can easily be found by searching the authors on the EGU site.

I started off on Monday morning in the CL40 session on model intercomparison (mostly technical stuff with no major insights), before moving on to CL12: monthly to decadal prediction. The most notable presentation here was Mojib Latif explaining that the recent unexpectedly strong warming in western Europe was due to a strengthening of the MOC, which he expected to decline and therefore somewhat ameliorate coming climate change in that region on the decadal time scale. After lunch, there was the session of most direct interest to me, CL20 (probabilistic climate prediction). There were a couple of interesting talks on principles and generalities. However, I was disappointed by some of the the specific applications. Were I in a more optimistic mood I'd probably portray it as a vigrous "zoo" of ideas but from where I was sitting it looked in one or two cases more like people floundering around not really knowing what to do, and as a result just doing "something". Of course my conclusion from this is that I should pursue _my_ ideas because they are clearly the most interesting and valuable ones around :-) Hey, I can dream.

I then went off to drown my sorrows in the free wine at the poster session. This year the CL (climate) division had decided to have all poster sessions in the early evening, with no talks scheduled to compete with them. This scheduling, combined with the wine, made the posters very well attended and the policy must be considered a success. However it also made it difficult for those presenting the posters to go and see other ones, especially as jules also had her poster some distance away at the same time so could not cover for me. Anyway, plenty of people had a look, a large poportion seemed to understand it and no-one tried to argue seriously for uniform priors. Unfortunately the main protagonists were not there so it's possible they will continue with their fingers-in-ears "la-la-la I can't hear you and anyway everyone else agrees with us" approach. Time will tell.

Tuesday morning started off with CL23 (solar and aerosols). There wasn't any "it's the sun wot dun it" stuff, fortunately. Palle said that planetary albedo had changed a lot (in both directions), but all the data seem pretty inconsistent and inconclusive. Someone (Philipona I think) gave another competing (or perhaps complementary) explanation for European warming, that the decline in aerosols had had a significant effect in this area. This "brightening" effect is now running out so the end result (less future warming) might be hard to discriminate fom Latif's theory.

CL28 was hockey stick stuff, in fact one made an appearance in Ray Bradley's excellent Hans Oeschger Medal Lecture. He talked about the various problems of proxy data and reconstructions pretty bluntly, I thought - including the divergence problem and dependence on a very few proxies with limited spatial coverage - and also pointed to some new data sources that might help in the future.

The SSP paleo session on Wed am was very interesting, with some people looking more at model simulations, and others analysing data. Didier Roche thinks that the oceanic dO18 distribution in warm paleoclimates might explain a significant part of the "gradient problem" - that (according to conventional analyses) the data suggest a strong polar warming but little or nothing in the tropics, whereas models give a much more uniform warming. Correcting for his model's estimated dO18 distibution made for a more even warming, more in line with what the models (and perhaps common sense) would suggest. There's plenty more work to be done in bringing models and data closer together in this field.

There was a CL division business meeting at lunchtime, so I rolled up for the promised free rolls (which were better than the stuff on sale, not that I have any complaints about the on-site catering). In amongst the minutiae that my reader won't be interested in, the subject of corporate sponsorship was raised. I didn't comment (others got there first) but think it would be hard to convince everyone that the benefits would outweigh the potential loss of independence.

In NL4.01 (nonlinear time series analysis) Anastasios Tsonis gave a fascinating talk about synchronicity and coupling in networks. He suggested that the sharp changes in global temperature trend were coincident with changes in behaviour of the climate system. He found this behaviour both in 20th century data, and also in a 21st century (A1B?) run of the NCAR model. However, before any sceptics get too excited, this latter result only explains changes in the small drift (perhaps 0.05C/decade) above and below the linear forced trend, not the entire warming itself, so it would be hard to claim that it explains the entire historical record. Nevertheless, it might help to explain the oddly sharp piecewise-linear behaviour observed over the 20th century. It's certainly always looked to me like natural variability probably played a bit of a role here (not that the IPCC have ever denied this, of course).

Overall, the NP (nonlinear processes) stuff seemed a little disappointing this year, with a higher proportion of fluff than usual. It's a bit sad when someone presents "novel method Y", a member of the audience asks "how does novel method Y compare to (well-known and widely used) method X", and the answer is "I don't know about X, so I can't answer that". There were also some boilerplate applications of routine ideas that had probably already been done to death a decade ago. Fortunately there were also some interesting ideas, but perhaps fewer than I've come to expect from these sessions. On the other hand it's important to balance the established and excellent speakers with opportunities for newcomers and the less eminent, and IMO the EGU generally does a pretty good job at that.

The relatively new ERE (Energy, Resources and the Environment) division had some interesting work. Faust from Munich Re was trying to tease out any relationship between SST and hurricanes, and seemed to think he had managed. I was just leaving to dive into to a parallel NP session as someone in the audience started ranting about insurance companies making too much money. Later I returned to hear about some attempts to calculate optimal or tolerable emissions pathways using simple coupled climate/economic models (Bruckner, Held). It turns out that "tolerable" is very much dependent on a large number of rather subjective decisions.

Thursday: CL18 Detection, modelling, impact. There wasn't much here that engaged me other than Douville saying that we still didn't know much about how the hydrological cycle would change, other than on the global scale. He seemed to say that the models don't really agree better than for the TAR, despite the AR4 wording. Right at the end of his talk he slipped in the comment that because demand is forecast to rise sharply, "reliable projections are urgently needed". I would have thought that a strong increase in demand implies that the need to to manage supply and demand more effectively can be confidently predicted (the real question is to what extent it should be centrally planned response versus left up to the magic of the market) irrespective of whether accurate projections of rainfall can ever be made. Indeed some people have been known to argue that the science is adequately settled inasmuch as its relevance to policy decisions is concerned, but I don't expect many climate scientists say that on their grant applications :-)

I didn't stay to hear Fred Singer waste 15 minutes of everyone else's time, but by all accounts he only made himself look like a fraud, so perhaps it was worth letting him have some rope. I also missed the Gregory-Rahmstorf cage fight, so rather than make it up I'll just refer to you Stoat.

At some point Geert Jan van Oldenborg gave his view on the European warmth, which I missed, but I found him in a poster session and asked him about it. His view is that the models are all wrong in having too thick a mixed layer in this region, and therefore in contrast to Latif and Philipona the strong warming will continue. The disagreement between them may not be quite as strong as I've made it sound, as the areas and time scales they considered are not all identical, but it still sounds like there could be the opportunity for a bet here!

Friday's program was pretty thin and so I'd arranged to spend much of it chatting to various people about some admin and science-related ideas. The poor Friday afternoon speakers drew the short straw (although perhaps not quite as short at those who didn't get to talk at all). In any case, the convenor's party was amazing and well worth staying for...

The EGU is back in Vienna again next year, at the same time. I wouldn't mind a change of venue but OTOH I can see why they keep going back.

At the airport, jules was a little ahead of me getting to the gate and asked if the exit seats were available. No, they were already full, she was told. As I approached, looming over the crowd of Japanese midgets and average-height Austrians she cheekily added "are they already full for him too?" 5 minutes later we had our new seat numbers :-) I'll fly Austrian Airlines again.

Friday, April 13, 2007

Gambling on Global Warming Goes Mainstream?

Here's an interesting article about betting on global warming.

Based on the bets offered, I think the question mark I have added to the subject line is appropriate! I'm relieved to see that Gavin broadly agrees with me :-)

Wednesday, April 11, 2007

EGU poster

It's just the same old same old for next week's EGU meeting in Vienna. If I'd managed to get it published anywhere I'd have moved on to something else, but although this seems to have been dragging on pretty much for ever, only a handful of people have actually seen it so far (mostly at last summer's workshop) and it's far from clear how widely accepted it is. So I'll give it one last airing and see if it generates any reaction...

Jules tells me I should mention her poster, and also this one, which was originally submitted as a rather speculative offer to describe some not-yet-done work, and which has basically degenerated into a glorified advert (see also here) for someone to come and do it - anyone want to visit to Japan for a few years?

Please don't find any typos. Or if you do, don't tell me :-)

Tuesday, April 10, 2007

Wag TV threatens to sue Wunsch

According to George Monbiot, Wunsch
says he has received a legal letter from Durkin's production company, Wag TV, threatening to sue him for defamation unless he agrees to make a public statement that he was neither misrepresented nor misled.
This is of course with reference to Durkin's swindle on global warming, which Wunsch complained vigorously about, saying he was "duped" and that the program was "close to fraud", and threatening to take further action.

Personally I don't think that lawsuits either way are the best way forward. Even if Wag TV were to win, it's unlikely that this would outweigh the bad publicity they would certainly receive, and I'm sure that Wunsch would do better to forget the sorry episode than try to drag them though the Independent Television Commission complaints procedure. As I said at the time, it didn't seem like he was really seriously misrepresented except perhaps by association (there are certainly lots of worse manglings of science in the media, deliberate or not). And the email he was sent prior to his appearance should have been ringing alarm bells:
The aim of the film is to examine critically the notion that recent global warming is primarily caused by industrial emissions of CO2. It explores the scientific evidence which jars with this hypothesis and explores alternative theories such as solar induced climate change. Given the seemingly inconclusive nature of the evidence, it examines the background to the apparent consensus on this issue, and highlights the dangers involved, especially to developing nations, of policies aimed at limiting industrial growth.
OK, it's not absolutely explicit, but there were some warning signs there...and a google on Durkin would have thrown up his history too.

Surely it's best to let this die and not give it any more publicity...oops :-)

Saturday, April 07, 2007

Data analysis: Frequently Bayesian

Despite not being a US citizen and not even having ever gone to one of their meetings, I joined the AGU recently for the princely sum of $20. The main reason for this is to stop them from spamming me with invitations to join :-) No, really it's largely because the discount on the upcoming EGU meeting registration fee is more than the AGU subscription, plus I get some freebies such as a subscription to Physics Today. Actually I can claim the fee for the EGU meeting back off my employer but not the AGU sub, so it does actually cost me money, but it still seems like too good an offer to pass up.

This month they have a nice short article discussing Bayesian and frequentist probability, which is only available to subscribers, so tough luck :-)
Hypothesis tests and the method of maximum likelihood are among the most widely used tools in the analysis of experimental data. But notice that frequentists only talk about probabilities of data, not the probability of a hypothesis or a parameter. The somewhat contorted phrasing that their methods necessitate seems to avoid the questions one really wants to ask, namely, "What is the probability that the parameter is in a given range?" or "What is the probability that my theory is correct?"
(the latter questions are precisely those which Bayesian methods can address).

Of course it's precisely this problem that leads to the contorted and easily misinterpreted language in climate science in general and IPCC SPM in particular. Not that it's particular to climate science - here's a paper complaining about the same situation in high energy physics (and for those who want more, the same author has written a lot of fairly easy reading on the same subject).

Election time

There's a bunch of election campaigns going on in Japan right now, most prominently for Governor of many prefectures including Tokyo. The incumbent (who will almost certainly win) is a racist old windbag like most of Japan's politicians. More amusingly, this is the election broadcast of one of the other candidates (subtitled in English courtesy of Trans-Pacific Radio):



There is a range of eccentrics and celebrities standing, and it would be obvious that this is also a joke were it not for the fact that the candidate actually has served prison time for violence in the past.

With apparently no sense of irony, the Japanese govt has asked Youtube to take down the video in the interests of democracy. Fortunately Youtube has shown no signs of compliance and it has been one of their most-watched clips recently. Japan has some bizarre and archaic laws concerning elections, which severely restrict the ability of candidates to communicate with voters (eg candidates cannot send email or update their web sites during the campaign). However, there are no restrictions on noise pollution so obnoxious "sound trucks" drive round residential area blaring out slogans and candidates' names. I think voting day is Sunday. I'll be glad when it's over...