RealClimate logo


Technical Note: Sorry for any recent performance issues. We are working on it.

Green and Armstrong’s scientific forecast

Filed under: — gavin @ 20 July 2007

There is a new critique of IPCC climate projections doing the rounds of the blogosphere from two ‘scientific forecasters’, Kesten Green and Scott Armstrong, who claim that since the IPCC projections are not ‘scientific forecasts’ they must perforce be wrong and that a naive model of no change in future is likely to be more accurate that any IPCC conclusion. This ignores the fact that IPCC projections have already proved themselves better than such a naive model, but their critique is novel enough to be worth a mention.

The authors of this paper actually have a much larger agenda, and that is to improve the quality of forecasting used in public policy and business everywhere – by the use of ‘scientific forecasting principles’ (of which they have enumerated 140). Most of these principles seem commonsensicial (don’t overfit a statistical model, test models on out of sample data etc.) and are listed on one of their many websites. Basically, you just assign a subjective numerical score for reflecting how well you match a particular principle and at the end you get a ‘scientific’ number that says how well you are doing.

Armstrong helped set up a journal dedicated to this goal, as well as running yearly meetings for scientific forecasters. However, in a recent review of progress he notes: “the diffusion of useful forecasting methods has been disappointing”, and that “forecasting meets resistance from academics and practitioners”. This seems surprising – why wouldn’t people want better forecasts?

G+A’s recent foray into climate science might therefore be a good case study for why their principles have not won wide acceptance. In the spirit of their technique, we’ll use a scientific methodology – let’s call it ‘the principles of cross-disciplinary acceptance’ (TM pending). For each principle, we assign a numerical score between -2 and 2, and the average will be our ‘scientific’ conclusion…

Principle 1: When moving into a new field, don’t assume you know everything about it because you read a review and none of the primary literature.

Score: -2
G+A appear to have only read one chapter of the IPCC report (Chap 8), and an un-peer reviewed hatchet job on the Stern report. Not a very good start…

Principle 2: Talk to people who are doing what you are concerned about.

Score: -2
Of the roughly 20 climate modelling groups in the world, and hundreds of associated researchers, G+A appear to have talked to none of them. Strike 2.

Principle 3: Be humble. If something initially doesn’t make sense, it is more likely that you’ve mis-understood than the entire field is wrong.

Score: -2
For instance, G+A appear to think that climate models are not tested on ‘out of sample’ data (they gave that a ‘-2′). On the contrary, the models are used for many situations that they were not tuned for, paleo-climate changes (mid Holocene, last glacial maximum, 8.2 kyr event) being a good example. Similarly, model projections for the future have been matched with actual data – for instance, forecasting the effects of Pinatubo ahead of time, or Hansen’s early projections. The amount of ‘out of sample’ testing is actually huge, but the confusion stems from G+A not being aware of what the ‘sample’ data actually consists of (mainly present day climatology). Another example is that G+A appear to think that GCMs use the history of temperature changes to make their projections since they suggest leaving some of it out as a validation. But this is just not so, as we discussed more thoroughly in a recent thread.

Principle 4: Do not ally yourself with rejectionist rumps with clear political agendas if you want to be taken seriously by the rest of the field.

Score: -2
The principle climatologist that G+A appear to have talked to is Bob ‘global warming stopped in 1998′ Carter, who doesn’t appear to think that the current CO2 rise is even anthropogenic. Not terribly representative…

Principle 5: Submit your paper to a reputable journal whose editors and peer reviewers will help improve your text and point out some of these subtle misconceptions.

Score: -2
Energy and Environment. Need we say more?

Principle 6: You can ignore all the above principles if you are only interested in gaining publicity for a book.

Score: +2
Ah-ha!

In summary, G+A get a rather disappointing (but scientific!) score of -1.66. This probably means that the prospects for a greater acceptance of forecasting principles within the climate community are not good. Kevin Trenberth feels the same way. Which raises the question of whether they are really serious or simply looking for a little public controversy. It may well be that there is something worth learning from the academic discipline of scientific forecasting (though they don’t seem to have come across the concept of physically-based modelling), but this kind of amateur blundering does their cause nothing but harm.

In association with their critique, G+A have also launched a very poorly thought out ‘climate challenge‘ that is essentially a bet on year to year weather noise. No one is likely to take them up on that, and they don’t seem to be interested in the rather better thought through bets on offer from James Annan and Brian Schmidt. Thus again, the conclusion must be that they are not serious about their stated goals. That’s a shame.

Shorter Armstrong and Green: If our publications are not cited, climate sensitivity is zero.

‘Shorter’ concept by Daniel Davies and Elton Beard


149 Responses to “Green and Armstrong’s scientific forecast”

  1. 1
    danny bee says:

    I am toying here with some gallows humor about global warming, and invite anyone else to join in with comments and feedBack. A new way to look at climate change, via humor. My blog is called “Gallows Humor for Global Warming”……..harmless stuff.

  2. 2
    Jim Cross says:

    Gavin, I assume you don’t object to applying some kind of methodology
    to forecasting but only to the way the authors have applied to their methodology
    to the IPCC forecasts.

    Since their web site says the following:

    Peer review and independent audits are welcome, and will be posted if the site guidelines are followed. In addition, all peer review must include the authors’ names, positions, emails, and any relationship that might be construed as a source of bias.

    You could provide an alternate assessment.

  3. 3
    Charles Raguse says:

    This article should be permanently enshrined in the “primary literature” required for close reading by anyone studying to become “a scientist”. It deals with common sense, which, it is said is called “common” because it is so rare!

  4. 4
    ChrisC says:

    The paper seemed to rely heavily on citations from the un-refereed publications and non-scientific literature. For example, several stories from the New York Times got mentioned as examples of climatologists predictions. Several non-refereed websites also found their way into the references. The “global cooling” myth got a run as an “expert”, which William Connolley might have something to say about.

    Call me crazy, but in my field of meteorology, we would never head to popular literature, much less the figgin internet, in order to evaluate the state of the art in science. You head to the scientific literature first and foremost. Since meteorology and climatology are not that different, I would struggle to see why it would be any different.

    The authors also seem to put a large weight on “forecasting principles” developed in different fields. While there may be some valuable advice, and cross-field cooperation is to be encouraged, one should not assume that techniques developed in say, econometrics, port directly into climate science.

    The authors also make much of a wild goose chase on google for sites matching their specific phrases, such as “global warming” AND “forecast principles”. I’m not sure what a lack of web sites would prove. They also seem to have skiped most of the literature cited in AR4 ch. 8 on model validation and climatology predictions.

    In short, the authors do not seem to understand the actual practice of climatology, or the physical sciences in general. A poorly thought out critque.

  5. 5
    AlZ says:

    When I see statements like: “Based on our Google searches, those forecasting long-term climate change have no apparent knowledge of evidence-based forecasting methods, so we expect that the same conclusions would apply to the other three necessary parts of the forecasting problem” and I just cringe.

    A “Google search” sounds like something I’d see on a second-rate undergraduate research paper; how about using the Science Citation Index for example and show me you’ve really done a thorough lit. review. This is just lazy.

  6. 6
    Gene Hawkridge says:

    Kesten Green and Scott Armstrong are right about one thing: the IPCC forecast is wrong, beyond any doubt. That is the nature of forecasts for chaotic systems with so much uncertainty in so many of the assumptions that must be made, and lack of complete understanding in how a myriad of factors, known and unknown, will influence outcomes. I stongly suspect that politics has had too much to do with the IPCC forecasts, and that reality will be worse: sea level rise will exceed the forecast, just as it did with the 1995 NOAA study for the “worst case” scenario. No evidence of an accelerating trend? That’s not what my analysis of the data I’ve seen tells me. When Russia, China, and the United States all put pressure on the IPCC to alter its report (to not make things seem as bad as they likely are) I knew we were not going to get a realistic assessment of what’s coming. G&A positing no change being more accurate? It’s wishful thinking, not science.

  7. 7
    steven mosher says:

    Thanks Gavin,

    Some of us climate agnostics found the G&A piece underwhelming. There was a lot of heat from both sides on the differences between forecasts, simulations, and projections. ( Dr. Curry had an odd sense of the necessary and sufficient conditions for forecasthood, for example) A good portion of the discussion ( ok the messy portion) centered on the issue of forecasts versus projections, which seemed a semantical side issue that would advance nothing but engender plenty of food fights. Fun while they last, but not very nourishing.

    One thing that came up was the issue of hindcast and the issue of confirming past “forecasts, projections simulations” against the contemporaneous record.

    So thanks for the link to the 2007 paper. Unfortunately my subscription doesn’t allow me to get it. This is exactly the kind of paper that would make for a great discussion on the web ( and the ancilliary food fights of course) but unless everyone has access dialogue is kinda thwarted.

    Oh well.

  8. 8
    John E Pearson says:

    President Bush turns up number 1 in google searches for certain naughty words. That Green and Armstrong’s website turns up number 1 in google searches for “forecasting” is equally significant.

  9. 9
    Tim McDermott says:

    It seems to me that much of the failure of the G&A article comes from the fact that they are economists. Economics doesn’t have anything resembling physics or thermodynamics, it only has models. For a long time, they thought the velocity of money was stable. Then it changed. For a long time, the P/E ratio of most stocks stayed in the range of 10 to 20. Then the range changed.

    In this kind of world, non-linear models are simply an exercise in curve fitting. So it isn’t surprising that they don’t trust them.

    It is also interesting that they use the word “forecast,” which isn’t what I think of scientific models as doing. Scientific modeling is part of the conversation Science has with Nature. Models are an executable representation of the thing being studied. They are (in the age of computers) a way to collect experimental results, and to make predictions. I doubt that G&A understand this. They live in a world where a the “quants” make forecasts of GDP, or the value of a particular secutity two weeks from now.

    I find that stupidity is more often the cause of egregious behavior than malice. But in the current AGW politics, it is hard to say.

  10. 10
    Lawrence Brown says:

    On principle 3. ‘Be Humble’. They sound like people who think they used to be conceited but now don’t have any faults. Their attack on climate models seems to be based on prejudged notions, and not very much research on the reliability of the models. In addition to the successes that Gavin mentions. Models run from the past have successfully reproduced present day conditions.Also the vertical distribution of the temperature of the atmosphere has been accurately reproduced, as well as the geographic distribution of precipition.
    There are enough hits to put good reliance of their projections of possible future outcomes.They want perfection? Get real! Ted Williams,of the Boston Red Sox, the last player to hit 400(meaning he was out 6 times out 10) is considered one of the greatest hitters of all time.

  11. 11
    Dan says:

    re: Principle 6. Sadly, it will undoubtedly work. Expect to see G&A’s “To Heck With The Scientific Method Tour” trumpeted and appearing in person on Fox News (?) any day now. :-P

  12. 12
    Goedel says:

    Well, they do have the good intentions of making climate modeling every bit as accurate as economics. What I wouldn’t give to have 70 degree highs in summer in Florida…

  13. 13
    steven mosher says:

    Gavin,

    Thanks again for some of the pointers. I found this:

    Climate simulations for 1880-2003 with GISS modelE

    The data page is quite nice so Kudos to Hansen et al.

    One question. we can use the page to draw plots. Is there a way to access the data behind the plots?

  14. 14
    Dave Arthurson says:

    So the paper hasn’t been submitted to a peer reviewed journal yet and they’re looking for peer review on the internet.

    Seems like a good place to start. I Googled “Peer Review”, it seems pretty popular.

  15. 15
    Walt Bennett says:

    The debunkerati thoroughly misunderstand the essence and nature of climate models. This type of ‘analysis’ is utterly typical of their efforts.

    I have long ago concluded that their tactics are simple and, in the short term anyway, probably more effective than any other approach they might take:

    1. Discern the uncertainties and unknowns of climate science, and exploit them.

    2. Develop pop theories that might be plausibly true today and declare them “alternative” theories, and then go the next step and declare them of equal or greater weight than current MCS understandings.

    3. As soon as the ink dries on one alternative theory, quickly come up with another, to keep MCS on defense. If MCS are always responding to the latest Debunkerati “theory”, they never have time to give the public (a) a thorough explanation as to why the alt theory is bogus or at least flawed, and (b) they are keeping debunkerati theories alive by giving them play.

    4. Once a debunked debunkerati theory has been out of mind for several months, repackage and re-present it. Keep the cycle going.

    As ad hominem as it sounds, we are certainly arriving at a point where it is quite enough to know the source of the ‘theory’ in order to know – KNOW! – that it is bunk.

  16. 16
    Hank Roberts says:

    http://environment.newscientist.com/article/dn12318-influence-of-global-warming-seen-in-changing-rains.html

    (For those with paid subscriptions, the article is at Nature):
    http://www.nature.com/nature/journal/vaop/ncurrent/full/nature06025.html

    New Scientist wrote about this:

    “… Detecting the effects of climate change on rainfall patterns has proved much more elusive than temperature changes because of the much greater natural variability of precipitation.

    “The key was to take results from 92 computer simulations, using 14 different global circulation models, and to compare the average of these with actual rainfall data over wide bands of latitude around the world.

    “The results show a clear agreement with the observed trends in global rainfall data over the past century. In fact, although they agree in direction, the observed changes were much stronger than the predictions.

    “‘Over the 20th century, we now detect the signal [in rainfall changes] that is predicted by climate models,’ says Francis Zwiers, one of the research team. ‘If you’re able to reproduce the past, you also have greater confidence for predictions of the future.’”

  17. 17
    Bob Sell says:

    Gavin,

    Thank you again for making Real Climate available to me.

    The other day my daughter, after listening to me talk about Real Climate, asked me if I thought a worldwide transition to Wind Power would affect weather and climate. I posited a no. I suspect there are local effects but unlike the butterfly in the Amazon, I don’t see a major effect.

    Bob

  18. 18
    Steve Horstmeyer says:

    Of course there is a (USD) $95.00 paperback for sale on the website link included in this post. (USD $166 if you want a hard back).

    A look at the list of authors and reviewers is enlightening. While a smattering are from information sciences or mathematics the majority are from business schools or business related institutions. Psychology is also represented. Unless I missed someone there is no one from the physical sciences as either an author or a reviewer.

    Steve Horstmeyer

  19. 19
    Chuck Booth says:

    The first “paper” cited by G & A is titled, “FIRE AND ICE Journalists have warned of climate change for 100 years, but can’t decide weather we face an ice age or warming.” Is the use of “weather” supposed to be a pun? Or is it a misspelling? If the former, it makes no sense grammatically (i.e., to use a noun as a conjunction). If it is the latter, it would be, I suppose, even worse, though consistent with the quality of the actual reference.

  20. 20
    B Buckner says:

    Hank #16
    This paper seems like nonsense to me. They say in the article that the detection of 20th century rainfall changes seems (seems? Is this a scientific term?) “barely discernible from the noise right now.” They then take results from 92 computer simulations, using 14 different global circulation models, and to compare the average of these with actual rainfall data. How does it make any sense to average these results? How does this represent anything? They then say the results show a clear agreement with the observed trends in global rainfall data over the past century. Clear agreement with the same observed trends that are barely discernable? They then conclude that they demonstrate that the pattern of rainfall around the world is being changed by greenhouse-gas emissions from human activities. Really?

  21. 21
    Richard says:

    Gavin,

    A pity you chose to pen a petulant, silly and smug response to their paper. Real scientists interested in putting forward a lasting scientific contribution would have reviewed this paper carefully and responded appropriately. These authors are serious heavyweights in their field. They have a legitimate role to play to improve forecasts such as used in the IPCC paper. They have opened the paper up for discussion, so why not be serious about it. One can only conclude that you haven’t got the guts or the science behind you to respond in proper scientific terms to their paper.

  22. 22
    Dan says:

    re: 21. For goodness sake. Assuming you are not just another drive-by poster showing up as if on cue and if you’ve read anything else on this web site, we are talking about climate *science*. Science that has been extensively peer reviewed and followed the scientific method (both are cornerstones to science). So two laymen who are not climate science modelers (G&A) magically know more than literally thousands of expert climate scientists with decades of research and knowledge from around the world. And who magically know something that every major scientific atmospheric science society in the world does not, huh? And you suddenly appear to attack a leading climate science peer-reviewed researcher for “not having the science behind” him? Please! This is the 21st century, not some anti-science Medieval period. And please let everyone know G&A’s climate science peer reviewed credentials. Or one can only conclude that you haven’t “got the guts” (now there’s a scientific phrase…not!) to stand on the hard science of global warming research.

  23. 23
    steven mosher says:

    re 21. Richard

    Gavin is a reasonable fellow. As you have seen I have asked him for access to the data for a recent paper from Goddard, an institution funded by taxpayers. He will produce it, I have no doubt. As for G&A, they really didn’t do all their homework. Even bright folks make mistakes. This is why folks like Gavin will open up to people checking his work. He is bright, but I don’t think he considers himself omniscient.

    So, Gavin. Can I get a data dump from the paper mentioned above? I have a big hard disk.

    [Response: Data for each figure you make is available at the bottom of the page. For more extensive data, more diagnostics and in a more convenient format are available through the PCMDI data server (see also updates and corrections at http://www.giss.nasa.gov/tools/modelE/modelE_AR4_issues.html ). - gavin]

  24. 24
    Dano says:

    RE 21 (Richard):

    The point is that they are not in their field. Reading their CVs shows no expertise in climate.

    It is like you looking over the plumber’s shoulder telling him what to do because you saw a show on the HG network and you work in nanotubes (what?!? They’re tubes, right?).

    Best,

    D

  25. 25
    Hank Roberts says:

    B. Buckner wrote: “This paper seems like nonsense”

    You aren’t reading the paper. You’re reading an excerpt I took from an article in an entertainment magazine, New Scientist.

    You can perhaps read the paper:
    (For those with paid subscriptions, the article is at Nature:
    http://www.nature.com/nature/journal/vaop/ncurrent/full/nature06025.html

  26. 26
    Hank Roberts says:

    This is not the article, this much is available to nonsubscribers; it’s at least amusing to compare this by climate scientists to the marketing department’s article on how to do forecasting.
    the first page
    http://www.nature.com/nature/journal/vaop/ncurrent/abs/nature06025.html
    and the supplemental material
    http://www.nature.com/nature/journal/vaop/ncurrent/extref/nature06025-s1.pdf

  27. 27
    James Killen says:

    Re: #17

    By “a worldwide transition to Wind Power,” do you mean, replacing the use of fossil fuels entirely with wind power? While you posit no major effect, I wonder how this could not have a significant effect on climate systems. Has this scenario been modelled?

  28. 28
    Lawrence Brown says:

    Regarding Richard’s comment No. 21- If a marketing professor and an economics and business forcaster can critique a climatologist’s field, then turnabout is fair play. Granted Gavin’s ‘rating system’ is done with tongue in cheek and some lampooning,yet the essence of his points are clear. The science of economics and the science of climatology are two very different breeds of cat. Actually I find the assumptions of some economists to be a kind of baffling. There is or was a school of thought that considered 4% of 4.5% employment as ‘full employment’(Unless,I imagine you happen to be an employed economist).

  29. 29
    Joseph O'Sullivan says:

    This was a good post to demonstrate how to judge the reliability of a review paper. Its also a good demonstration on how the science is done out in the field.

    I think when people don’t accept climate science some of it is not because they don’t understand the details of the climate science. Its because they don’t understand how science itself works. An effort to show people not just the scientific method but how scientific research is done could be helpful.

    Does anyone know anything about Kesten Green and Scott Armstrong’s work? Are they legitimately trying to improve forecasting in regulations? Or are they free-market types who are disingenuously casting doubt on regulation. This issue reminds me of the way the business community pushes cost-benefit analysis to subvert regulation, and it might be that G&A are in the second category.

  30. 30
    SCP says:

    So let me see, this ivy league guy writes books and papers and gets recognized, all in the ’80s and ’90s and having nothing to do with climate science. Then, these authors apply the established principles to IPCC forecasting, upon which we’re supposed to be willing to bet our economy in order to save our children.

    Then… the authors put their own money and reputation where their mouth is by offering a bet and publicizing it.

    The realclimate response is to ridicule them with a sarcastic and irrelevant parody and to pretty much neglect to defend the IPCC against the serious claims in their paper. I was keeping an open mind, but your post convinced me that you see the IPCC’s position as indefensible. If not, I think you would have defended it, or at least tried to.

    Your strongest reproach seems to be that they’ve got profit motives (BTW, I think you might have forgotten
    to accuse them of being funded by the oil companies ;-).
    They’re selling a book you know. Well I suppose everyone’s got a profit motive. If you can’t show me how their profit motive influenced their conclusions, then why bring it up? Again, I’m surmising it’s because it’s the strongest defense that the IPCC has.

    According to these authors, the IPCC violated 72 forecasting principles out of 89 which were evaluated. Another 51 couldn’t be evaluated for whatever reason. What does RealClimate say about this? Do you dispute the principles that they claim were violated? Did the authors get it wrong and the IPCC didn’t really violate the principles? I guess maybe I overlooked the informative part of your post where you discussed stuff like that.

    If you’ve got something useful and on topic, I sure wish you’d say it. This Real Climate post struck me as being a whole lot of words just to say “la-la-la-la I can’t hear you”.

    [Response: Maybe I was being too subtle. The critique is worthless because the authors are not in any position to be able to make fair evaluations of their criteria. They neither bothered to read the more relevant parts of IPCC, nor engage anyone who would have been able to steer them to the appropriate literature. By performing their audit in a state of blissful ignorance, they themselves demonstrate the failing they accuse IPCC of - they used their methodology to give a scientific sheen to their prior prejudices. Clear enough? If you'd like more concrete examples, take any 10 of the principles they think were violated and I'll point out why their scores were wrong. - gavin]

  31. 31
    Aaron Lewis says:

    In fact, G&A, ask an excellent question, “Are current global climate models good enough to use as the basis for public policy decisions?” Moreover, while G&A did not frame their argument in terms that RC likes, the conclusion that current climate models are not adequate to support public policy is reasonable.

    Yes, the models support the concept AGW is real and that we should do something about it. However current models do not suggest how much should invest this week, this month, this year, or this decade, given the realities of economic market places, discounting of future events, and the cost of money.

    Consider my home state of California. The models say our water infrastructure will be impacted by AGW. The models say that we can expect real drought. So, we should invest, and build for drought? Maybe not. The models also say that given construction schedules and cycles, by the time we have gotten ready for real drought, our problem may be too much rain. Why spend $80 billion on drought infrastructure if by the time we get it built, we need to spend $80 billion on flood infrastructure? Why not just spend $160 billion now and build for both? Because , this year in California both rain and capital are in short supply.

    Gavin’s models do not give us enough information to make these kinds of economic decisions. Sorry.

    [Response: I've never suggested that the science of climate change should determine public policy exclusively. Economic decisions are rightly within the realm of politics. One could agree with every word in IPCC WG1 and still not think it worth doing anything. Personally, I would disagree, but action is not determined by the scientific understanding of the problem. However, G+A's critique has nothing to do policy actions - it is solely concerned with what climate science has to say. On that, they are woefully (and I think deliberately) ignorant. - gavin]

  32. 32
    Ike Solem says:

    G&A quote Roger Pielke Sr:

    “Weather is very difficult to predict; climate involves weather plus all these other components of the climate system, ice, oceans, vegetation, soil etc. Why should we think we can do better with climate prediction than with weather prediction? To me it’s obvious, we can’t!”

    It’s one thing for economists with little knowledge of science to make silly claims, but scientists should know better.

    The response and mixing times of the atmosphere and the oceans are very different, for starters. This is also true for soils, vegetation and ice sheets. The ‘memory’ of the these components of the climate is far longer than that of the atmosphere.

    As one example, look at the ability of large volcanic eruptions to inject a long-lived cooling signal into the oceans. There are a lot of primary literature discussions of this. A good news article on this topic is at Volcanic eruptions cool ocean (Krakatoa’s 50-yr cooling signal)).

    At most, a large (stratospheric) volcanic eruption has a strong direct atmospheric effect for only a few years. The ocean signal lasts far longer. That’s one reason why long-term climate is far more predictable than short-term weather.

    Take precipitation as another important climate issue. Warming oceans mean more evaporation and more precipitation. Warmer ocean waters also seem to lead to changes in atmospheric circulation which tend to keep moisture out of the continental interiors.

    As far as the economic arguments go, there’s a discussion of the Stern report on the economic damages due to global warming in the july 13 2007 science issue (Stern and Nordhaus)

    Meanwhile, we have unusually heavy rains and record flooding in Britain, while interior Europe is baking in yet another record heat wave – but both regions are experiencing severe crop losses and large economic damages.

  33. 33
    Pekka Kostamo says:

    Hurrah! The old and obsolete GIGO is dead!

    Now you can buy the brand new GIJO computer model! “Garbage In, Jewels-of-wisdom Out”!

    Famous University marketing professors G&A did it! Evaluate yourself any solution or proposal with reliable results! It only takes one hour! Beware of the “experts”! All life’s problems are about forecasting the future!

    Seriously, the old snake oil peddlers are at it again. The business forecasting problem is different. Major future facts there are truly unknown or indeed unknowable. A new invention, a new competitor strategy, a new fad of customers, a new government regulation, a new war or a major natural disaster. They do not exist yet and are therefore unknowable. Some real world miracles there, revolutionary new ideas emerging from literally nothing. No model can reliably evaluate these impacts. Claims to the contrary are unfounded.

    The idea of “unknowns” is being peddled in the physics as well, like the magical disappearance of human-produced carbon dioxide or its effects. In physics, forecasting is possible as the major facts are knowable and for the most part known. Miracles are extremely unlikely.

  34. 34
    steven mosher says:

    RE 31,

    RCP. G&A ask a good question, but their attack on AGW science is fluff. You need to distinguish between AGW science and AGW “politics and economics”.

    To be sure the AGW community has not done the best job of conducting open source science, But as Gavin pointed out G&A didnt dig very deep to understand their subject.
    They didnt talk to anybody, read only one chapter of the IPCC and to my knowledge never asked for data or methods. You cannot audit without access to the data
    and methods.

    Now, Gavin has been kind enough to link to a some articles that address the only substantive questions G&A raise, questions about “projection skill”. Further,
    I requested above that he give me a pointer to the data. Plots are nice, but data is better. G&A did a lousy audit. In my mind they didnt even look for the data. And Gavin is rightly calling them out.
    So, seems only fair to actually ask for the data. His criticsim is spot on. They don’t appear to have read any of the supporting papers. I Just finished one authored by Hansen and asked for the data.
    We shall see.

  35. 35
    steven mosher says:

    RE 23.

    Thanks gavin. I wont bug you anymore. you can delete
    my previous if you like.

  36. 36
    MoZ says:

    I enjoyed your send-up of their methodology. I read their paper and wrote a piece about the “Climate Bet” and their paper about a month ago. Although I’m no scientist, I reached about the same conclusions that you did: the forecasting audit “instrument” operates (certainly in their application of it to the IPCC chapter) almost laughably as a device to hide their desired conclusions behind numbers; the guys are desperate for publicity; their thesis can be reduced to the assertion that no prediction is valid unless first run through their omniscient Prediction Validator(tm) software.

    My piece also contains a certain amount of needlessly snotty commentary about Armstrong’s relentless self-promotion and an old paper he wrote called “Bafflegab.” I think the juvenile Climate Bet stunt just ticked me off. I mean, come on, a live counter on their website????

  37. 37
    MoZ says:

    RE # 29: I don’t think they have a clear ideological axe to grind. I think they are mainly interested in promoting scientific principles in forecasting and – as luck would have it! – their own work along the way. If there is an underlying “bias” it might be that Armstrong has long promoted a skeptical view of expert consensus. He argues that expert consensus often is a better indication of laziness or groupthink than of accuracy. He spends several entertaining paragraphs in the paper providing examples of big embarrassing expert faux pas of the past. He has authored several papers wherein he propounds the thesis that naive predictors are equal to or superior than experts at predicting the results of complex phenomenon. Given this work, it seems likely that the frequently-cited consensus on global warming made him feel the area was ripe for his own special brand of debunking.

  38. 38
    John Monro says:

    The egos and ignorance of the vast majority of economists knows no bounds, not content with making absurd predictions that our great grandchildren will be ten times better off than we are, so why do anything about global warming when they’ll be able to fix it before breakfast, and when the have carefully worked out that saving the planet in 2100 is not worth the present discounted cost of doing so, the same economists, about who it was said, according to one of the few economists that I ever cared a whit about, Kenneth Galbraith, “The only function of economic forecasting is to make astrology look respectable”, and who seem to find no contradiction between an infinitely expanding economy and a finitely constrained planet, now they have to dabble in spheres in which they have absolutely no expertise and rubbish the conclusion of thousands of experts who have. Just contrast Gavin’s reasoned reponse to Aaron’s chiding that his models can’t help us make considered economic decisions “Economic decisions are rightly within the realm of politics. One could agree with every word in IPCC WG1 and still not think it worth doing anything. “, the scientist’s understanding of his own expertise, with the arrogance of such as Green and Armstrong and so many others of his profession. It’s bad enough that the whole world suffers under the delusional society that modern economists have helped us construct, to think that our science is going to be tainted with the same nonsense is appalling! (By the way, I have this thing about economists!)

  39. 39
    Adam says:

    “in a recent review of progress he notes: “the diffusion of useful forecasting methods has been disappointing”, and that “forecasting meets resistance from academics and practitioners”.”

    But they were able to forecast that this would be the case, right? Or were their forecasts of how well the take-up would be a little awry?

  40. 40

    [[A pity you chose to pen a petulant, silly and smug response to their paper. Real scientists interested in putting forward a lasting scientific contribution would have reviewed this paper carefully and responded appropriately.]]

    Right, and the place to get such feedback is in a peer-reviewed science journal. But they didn’t submit their work to such a journal. Are you wondering why not?

    [[ These authors are serious heavyweights in their field. ]]

    They are not serious heavyweights in climatology, which is the field they need to be serious heavyweights in if they’re going to critique climate models. The fact that they are serious heavyweights in their OWN field means nothing. William Shockley was a serious heavyweight in solid state physics, but his opinion on race and IQ was still worthless because he was talking out of his field.

    [[They have a legitimate role to play to improve forecasts such as used in the IPCC paper.]]

    No, they do not. They have made no effort to understand what they’re critiquing, as Gavin’s review showed.

    [[ They have opened the paper up for discussion, so why not be serious about it.]]

    Because they themselves are not being serious about it. If they want to “open it up for discussion,” they can submit it to a peer-reviewed climatology journal like the Journal of Geophysical Research, Geophys. Res. Letters, Journal of Atmospheric Sciences, Climate, or for that matter, Science or Nature. Opening it up for discussion by random amateurs on the internet won’t result in any improvement in the paper at all, and they know it. But you’re too gullible to see through what they’re doing.

    [[ One can only conclude that you haven’t got the guts or the science behind you to respond in proper scientific terms to their paper.]]

    One can only conclude you haven’t got [edit] the discipline behind you to learn a little bit about the science involved yourself so you can see why the G&A paper is a load of crap.

  41. 41
    UC says:

    Gore should take the bet. Easy money, that naive model is a random walk predictor, and it is easy to see that local temperatures are not random walk (closer to white, actually).

  42. 42
    Hank Roberts says:

    UC from “signals.auditblogs.com” — that’s bad advice. I’d guess you know better since you seem to be a self-appointed auditor?
    If you do believe what you’re saying, you haven’t read James Annan, who explains what’s wrong in simple clear language. A lot of us are waiting to see if the marketing prof ever answers the challenge issued weeks ago by Annan. But so far, nothing.

    I would guess the marketing principle involved here is “there’s no such thing as bad publicity.”

  43. 43
    Vernon says:

    I see more of the “if your not a climatologist, then you cannot speak about climatology” mind set is still good and well. Though you attacked the individuals I fail to see where you attacked their work. Armstrong is one of the leading experts in forecasting and has been for years. He points out where the IPCC is not following the established guidelines for forecasting that were established well be fore the IPCC came into existence. You say he does not understand climatology and how perfect the models are, but, if I read the IPCC report correctly, all he is doing is pointing out how the IPCC is making forecasts and not following the accepted guidelines.

    From what I have read, nothing has be presented to show that the forecasting guidelines are wrong. The fact that climate models are constantly changing would indicate that one the major points they raise;

    Complex models (those involving nonlinearities and interactions) harm accuracy because their errors multiply. Ascher (1978), refers to the Club of Rome’s 1972 forecasts where, unaware of the research on forecasting, the developers proudly proclaimed, “in our model about 100,000 relationships are stored in the computer. Complex models also tend to fit random variations in historical data well, with the consequence that they forecast poorly and provide misleading conclusions about the uncertainty of the outcome. Finally, when complex models are developed there are many opportunities for errors and the complexity means the errors are difficult to find. Craig, Gadgil, and Koomey (2002) came to similar conclusions in their review of long-term energy forecasts for the US made between 1950 and 1980.

    You have also failed to show where another major point is wrong, namely:

    Agreement among experts is weakly related to accuracy. This is especially true when the experts communicate with one another and when they work together to solve problems, as is the case with the IPCC process.

    I do not see them making claims about the science but about the forecasts. Please point out where the they are wrong about what they are experts in, forcasting methodology.

  44. 44
    Tim McDermott says:

    Vernon (43):

    Is forecasting a skill that can be separated from what is being forecast? I don’t think so. Do these “established guidelines” apply to forcasting, say, eclipses of the sun? No, astronomical forecasts involve involve nonlinearities and interactions. Astronomers agree on how to forecast eclipses, so they are not to be trusted!

    And Vernon, why are you willing to accept economists as authority figures but not accept climatologists?

  45. 45
    Dan says:

    re: 43. We see more of the “complete failure to understand the significance and importance of scientific peer review within the realm of climate science” anti-science mind set is alive and well. There is an excuse for not understanding the scientific method (which the science of global warming and climate modeling follows). There is, however, very little excuse for the failure to try to learn about it especially since it has been explained many times here. The idea that laymen know more than literally thousands of climate researchers throughout the world whose work has been rigorously peer-reviewed and every major atmospheric science society is the height of (purposeful?) arrogance. If the idea that laymen know more than the peer-reviewed science is truly beleived, do not go to a doctor the next time you become ill as they follow peer-reviewed science for evaluating and providing treatment as well; go to perhaps a lawyer or a baseball player. Surely the later two occupations know more about treating an illness than a trained doctor. ;-)

  46. 46
    Chuck Booth says:

    Re # 43 “Agreement among experts is weakly related to accuracy. This is especially true when the experts communicate with one another and when they work together to solve problems, as is the case with the IPCC process.”

    And where is the evidence to support this assertion that you find so compelling?

  47. 47
    Neil B. says:

    Once again, in addition to other errors, the G&A team avoid the theoretical basis for CO2 warming. They hide behind a pure empiricism (AFAICT, not having time to read it all, frankly) typical of AGW skeptics.

    BTW, where is the best place to find the actual calculation of direct CO2 effect, in calories per m^2, and then temperature change, from the CO2 and other GH gases?

  48. 48
    guthrie says:

    Vernon, the problem is that the problems you have quoted are all potential ones, and a scan through the “paper” shows that the authors have not seriously linked any problems to the actual IPPC forecasts. Using Google scholar is not enough. Comments by non-climatologists that are wrong, such as Carter:

    “the slope and magnitude of temperature trends inferred from time-series data depend upon the choice of data end points. Drawing trend lines through highly variable, cyclic temperature data or proxy data is therefore a dubious exercise. Accurate direct measurements of tropospheric global average temperature have only been available since 1979, and they show no evidence for greenhouse warming. Surface thermometer data, though flawed, also show temperature stasis since 1998.”

    are given prominence. Nobody but a mathematical iiliterate would claim that temperature shows stasis since 1998, and the Troposphere is actually warming.
    Thus, this “paper” is a hatchet job.

  49. 49
    Neil B. says:

    Gene Hawkridge #7: I have an intuitive sense from my own experience that you are right, but even that things have already changed more than we’re told. I remember it snowing a lot more here in SE VA in the 60s, as a boy. I hear also from others, things have changed more over the last few decades than a degrees average. What’s up with that?

    PS, I hope some of you got to look at the infamous WSJ “Laffer Curve” drawn around corporate tax data, whereas the collection really went mostly up with increasing tax rates up to a point.

  50. 50
    Jim Bouldin says:

    The very first paragraph, 2nd sentence, of the abstract of the G + A paper is by itself quite revealing:

    “In 2007, the Intergovernmental Panel on Climate Change’s Working Group One…issued its updated, Fourth Assessment Report, forecasts. The Report was commissioned at great cost in order to provide policy recommendations to governments.”

    From the IPCC’s “Principles Governing IPCC Work” website (http://www.ipcc.ch/about/princ.pdf) is the following official statement of the IPCC’s role:

    “2. The role of the IPCC is to assess on a comprehensive, objective, open and transparent basis the scientific, technical and socio-economic information relevant to understanding the scientific basis of risk of human-induced climate change, its potential impacts and options for adaptation and mitigation. IPCC reports should be neutral with respect to policy, although they may need to deal objectively with scientific, technical and socio-economic factors relevant to the application of particular policies.”

    The IPCC DOES NOT MAKE POLICY. G + A don’t even understand this fundamental point. And as mentioned several times above, they’re not climate scientists, meteorologists, physicists, mathematicians, or chemists. How in the world can they expect to be taken in any way seriously?


Switch to our mobile site