RealClimate logo

Friday round-up

Filed under: — group @ 24 July 2009

Two items of interest this week. First, there is an atrocious paper that has just been published in JGR by McLean, de Freitas and Carter that is doing the rounds of the denialosphere. These authors make the completely unsurprising point that that there is a correlation between ENSO indices and global mean temperature – something that has been well known for decades – and then go on to claim that that all trends are explained by this correlation as well. This is somewhat surprising since their method of analysis (which involves taking the first derivative of any changes) eliminates the influence of any trends in the correlation. Tamino has an excellent demonstration of the fatuity of the statements in their hyped press-release and Michael Tobis deconstructs the details. For reference, we showed last year that the long term trends are still basically the same after you account for ENSO. Nevermore let it be said that you can’t get any old rubbish published in a peer-reviewed journal!

Second (and much more interestingly) there is an open call for anyone interested to contribute to setting the agenda for Earth System Science for the next couple of decades at the Visioning Earth Science website of the International Council for Science (ICS). This is one of the umbrella organisations that runs a network of committees and programs that prioritise research directions and international programs and they are looking for ideas. Let them know what your priorities are.

533 Responses to “Friday round-up”

  1. 251
    wmanny says:

    John Mashey
    7/09, 11:21 PM


    Since you believe that reputation matters, what’s your take on the four NAS signers, including the Nobel laureate Giaever? More generally, do you believe we need to listen to any physicists at all (Dyson probably the most notable example this year) if they are not actively publishing in the climate field? By extension, should we even care what APS thinks about AGW one way or the other?


  2. 252
    simon abingdon says:

    Martin #246 I’ll try again. Ray #193 says “90% confidence does not equate to a 10% chance of being wrong”. Therefore 90% confidence does not equate to a 90% chance of being right. What then is the relationship (if any) between probability and confidence expressed as a percentage (not confidence interval or confidence limit, just plain “confidence”)?
    To repeat, am I wrong to equate confidence with probability by saying (for example) that my confidence is 50% that the next coin toss will come down heads?

  3. 253
    Mark says:

    “More generally, do you believe we need to listen to any physicists at all (Dyson probably the most notable example this year) if they are not actively publishing in the climate field?”

    Ah, the lovely summer straw!

    Listen to them.

    No problem.

    But if you have an expert on one side and a non-expert on the other, unless you’re going to investigate on your own, take the experts advice.

    Just because there are no lawyers on slashdot, does that mean you shouldn’t listen to their advice when you’re in legal trouble? Citizens Advice have no legal professionals on their team, but you should listen to their advice. But if you already have a lawyer, listen to their advice and follow it if you just want to take advice.

    NOTE: a lawyer is quite often wrong too. They often tell you to settle because you haven’t got much chance of winning. When in fact there is a large chance of winning. But if you want to ignore the lawyers advice, you had better get some work done in understanding law first.

  4. 254
    Hank Roberts says:

    Fred, you’re once again trying the “founder” notion — yes, Millikan was a founder. Yes, his value was wrong. Yes, it was a good paper, by the criteria of scientific papers. It led other researchers to publish good work, and over time, the science improved.

    Science isn’t a religion. There’s no original foundation on which everything else is based, that can be tipped or cracked to collapse all built on it.

    Science is not like the mighty oak or towering redwood, a single great idea pointing in a clear direction. Science is like kudzu, the work is done at the growing edge, and it grows wherever the conditions are better, and the “direction” is something we decide on after the fact.

    But you know this. Old scientists are like old kudzu, part of the mass but not holding up what’s happening now.

  5. 255
    Hank Roberts says:

    Oh, for any kids writing papers, the Millikan story is worth understanding — it makes the point clearly that early work leads to later work, and early work isn’t valued because it’s perfect but because it leads to interesting work later. You know how to look this stuff up by now — or your librarian can help you with it. Here’s a start:

  6. 256
    Mark says:

    “Therefore 90% confidence does not equate to a 90% chance of being right.
    Comment by simon abingdon ”

    The IPCC report isn’t saying just “CO2 is the main cause of warming”.

    It’s saying “the rate of warming is between 2 and 4.5C per doubling”.

    Now, lets say the rate of warming was 4.6.

    Are the IPCC wrong?

    In saying that CO2 is a major cause: no.
    In saying that CO2 doubling will produce a 2-4.5C warming? Yes.

    What is the chance that the warming is outside 2-4.5? Maybe 1%. Maybe 10%.

  7. 257
    Doug Bostrom says:

    John Mashey 30 July 2009 at 12:11 AM

    I think it would indeed be crazy not to exploit modern technologies as much as possible!

    Just the availability of abstracts online (leaving aside the battle over embargo of full articles) has been so beneficial to everybody. I’ll hazard a guess that extensive casual access to journals that might not be in the local university library roster translates directly into better science. This is also to the benefit of the general public as well; think of how the discussion here is promoted and bettered by that means.

    I’d say that support for low latency online discussion would require tight access controls, if discussion were not to degrade to, say, the level found here, which is generally of remarkable quality but nonetheless attracts its fair share of crackpots and folks insufficiently informed (me, for instance) to make useful contribution to particular areas of scientific inquiry. Open access means relatively intensive work for a moderator; Gavin for instance has an incredibly sure eye but he’s doing much more work than is strictly necessary or desirable for the purpose of advancing any given path of investigation.

    Yet some mechanism would need to be in place to handle authentication of participants beyond simply a roster of print subscribers or membership lists of the professional bodies associated with the journal’s field of coverage. It’s often the case that cross-disciplinary communications are beneficial, AGW itself being an excellent example.

    Such a vehicle would require another form of reviewer as well, somebody reasonably well versed in a given field yet with a schedule permitting frequent moderation duty. Come to think of it, maybe this could become some kind of socializing service for grad students to perform. Perhaps I’m exaggerating that requirement; if access control is well handled then moderation might not require much expertise.

    Forward-thinking journals ought to be laying tracks to run these systems, but at least some little amount of resources will be necessary. Seems to me the access control is the biggest nut to crack.

  8. 258
    Mark says:

    “To repeat, am I wrong to equate confidence with probability by saying (for example) that my confidence is 50% that the next coin toss will come down heads?

    Comment by simon abingdon”

    No. But the IPCC report isn’t that.

    Your coin-tossing is more like you have tossed it 10 times and got 4:6 heads:tails.

    What is your confidence that the chance of the next coin being thrown as a head is 40%?


  9. 259
    Rod B says:

    Doug Bostrom (233), Partially disagreeing with you in the first of your post, I think the signatories are claiming the current science is partially incorrect; they are not making a claim of corruption (in the normal common accepted meaning of that term.) IMO the current situation is somewhat incorrect — “somewhat” being a bit less than “partially” in my arcane mind. I do not think it “really stinks” — not even close to. Nor do I think a revolution is in order to fix the current problems — again not even close to. I don’t think the signatories are saying that either and I agree with what they say according to my interpretation. If your interpretation is correct (and I don’t think it is even close…, but…) then I do not agree with them.

  10. 260
    Rod B says:

    Ray (242), I did not say evidence becomes dogma. I said evidence can become (be turned into) dogma — and sometimes does. Dogma is not necessarily religious in nature; it can be an inflexible principle or set of principles laid down in any group by some authority. If you think nobody has ever taken a bit (or alot) of evidence and turned it into dogma, you need to get out more. ;-)

    If you’re interested in why I think much of climate science has morphed into dogma, ask — though it will probably just make you mad.

  11. 261
    Rod B says:

    Mark, pretty good (250). Except I think they’ll eat us, not enslave us; clearly not help us.

  12. 262
    Ray Ladbury says:

    Simon Abingdon,
    Confidence is inherently connected to intervals–in this case, probably those of climate sensitivity. A sensitivity above 2 degrees per doubling equates to anthropogenic causation. The evidence is sufficient to establish that at the 90% CL–that doesn’t mean there is any evidence to the contrary.

  13. 263

    To John Mashey:

    In #234 you asked me some questions about Moorad.

    I will forward them to him and suggested he might see fit to respond here.


  14. 264
    Ray Ladbury says:

    Walter Manny,
    If you listen to what Dyson says, it is clear that 1)he is out of his depth when discussing climate; and 2)he is a technological optimist who believes we’ll be saved by our inventiveness. I would also contend that while Dyson is a smart guy, he has little understanding of how technology progresses. It takes time, and thanks to 20 years of delay, we’ve already spent all the time we had to spend.

  15. 265

    #251 Wmanny.

    “what’s your take on the four NAS signers, including the Nobel laureate Giaever? More generally, do you believe we need to listen to any physicists at all (Dyson probably the most notable example this year) if they are not actively publishing in the climate field? ”

    THey are good scientists ignorant about climate science. Unless they predicted something which has transpired they are inconsequential people in this field merely speculating. Dyson, as far as I know, predicted square trees, but has not made a prediction based on his understanding of climate. I never read Einstein making comments on brain surgery. If these NAS scientists are right about climate, then they must contribute to its understanding by passing the ultimate peer test, not from a school, nor a journal, the ultimate peer in this science is the future. Not trying to forecast it, is a sign of ignorance in this domain. Most modellers, I dare say, are fearless, succeed and fail with computer projections.
    They have earned my respect and admiration. Most contrarians, coward and fear the future.

  16. 266
    Mark says:

    “I think the signatories are claiming the current science is partially incorrect; ”

    OK, so which bits are they saying is partially incorrect and where’s their proof?

    And why isn’t the current system working on an partially incorrect results?

  17. 267
    simon abingdon says:

    Mark #258 I read your post a few times but couldn’t make sense of it. The point I have been trying to make is that ascribing a numerical value to “confidence” is spurious. Where are the calculations that enable Ray to say “Our evidence is sufficient to claim 90% confidence that we are warming the planet”? The use of a figure suggests a rigour of derivation that just isn’t there. I’ve asked for a definition of “confidence” but no-one has come up with one; only hand-waving. BTW you may be amused to google “astronomically small” (#250).

  18. 268
    Doug Bostrom says:

    Rod B 30 July 2009 at 10:08 AM

    Fair enough. Perhaps it hinges on how picky (neurotic?) one (me, for instance) is about the implied content of the letter in question. To me it smacks of an indictment but it’s possible the authors did not intend it quite that way.

    Understand, you’re treating with a person who can become quite passionate about such things as the exact optimal way to load a dishwasher.

    In this case I’m inclined to become highly incensed with the mere suggestion that a plurality or significant minority of authors from multiple disciplines producing articles related to the AGW phenomenon are incompetent or motivated by something other than a search for improved awareness.

    Perhaps the authors are too focused on the pure physics component, ignoring the rather large body of indirect evidence from other disciplines lending support to the interpretation and prediction direct physical processes contributing to climate change? In any case they’re jumping the gun by suggesting a response to a problem they can’t show as factual.

    I only wish they’d apply the same scrupulous attention to detail and proof in this matter that they would in their core professional activities.

  19. 269
    David Watt says:

    Having dealt with the 50 APS Society members who wrote to Nature, what is your take on the American Chemical Society who are threatening to throw out their editor for being too partisan in his support for AGW?
    Do you think there is any risk of the samr thing happening on Realclimate?

  20. 270
    Jim Prall says:

    I’ve been updating my list of climate authors and signers of statements about climate – I was nearly caught up, but now here comes another declaration! My big project recently has been to collect stats on how many papers each author or signer has published that mentions the word “climate” – a useful proxy for being involved in climate research, I’d venture.

    Anyway, after a few keystrokes in linux with sed and tr, I’ve got a text file list of the signers from their declaration website (currently showing 61 names “as of July 22”). Of these, linux “comm -12” tells me at least 13 already showed up in my list as signers of prior skeptics’ statements. I’ll need a bit of time to process the rest of the names.

    For now, you can see the publication status of the 13 already listed in my
    table of climate skeptic declaration signers. This table is sorted by number of Google Scholar matches on the word “climate” for each author.

    Quick summary: of the 13 past signers I’ve got listed already, only Singer (110), Knox (23) and Douglass (16) are in double digits for the number of GS papers mentioning “climate.” Next comes Battaglia with 5. The rest have zero to two:
    Cohen: 2
    Friedman: 2
    Giaever: 0
    Happer: 1
    Hayden: 0
    Monce: 0
    Sheahen: 1
    Stilbs: 1
    Tipler: 2

    (Sheahen signed the Cato letter to Obama, except they mis-spelled his last name!)

    So while many of the names on this new open letter have impressive credentials in whatever area of physics they have worked in, the selection I’ve seen before have next to no publication record *on climate*. This is consistent with the larger list of climate skeptic signers, for which the median number of papers mentioning ‘climate’ is … two.

  21. 271
    Ray Ladbury says:

    Simon Abingdon says, “The use of a figure suggests a rigour of derivation that just isn’t there. I’ve asked for a definition of “confidence” but no-one has come up with one; only hand-waving.”

    WRONG! Confidence is well defined in statistics–whether the statistics be frequentist or Bayesian. In point of fact there are many ways we could establish more than 90% confidence in the conclusions of climate science wrt anthropogenic causation. First would be to look at the confidence intervals for climate sensitivity. All lines of evidence favor a value around 3 degrees per doubling, and most confidence intervals drop like a shot when you get around 2 degrees.
    We can look at the various GCM–none of which produces anything remotely Earthlike with a sensitivity below 2 degrees per doubling–23 samples, 23 successes…

    And so on. The fact of the matter is that the 90% confidence level for the conclusions is in my opinion conservative, precisely because there are no alternate theories out there that come close to explaining even a tiny fraction of the observations/evidence. To contend otherwise is simply ignorant.

  22. 272
    Martin Vermeer says:

    #252 Simon:

    To repeat, am I wrong to equate confidence with probability by saying (for example) that my confidence is 50% that the next coin toss will come down heads?

    The problem here, Simon, is that you are using here “confidence” and (subjective) “probability” as synonyms. You may do so casually, don’t sweat it, but scientific usage, as in “confidence interval” etc., is quite different.

    It is related to probability in the sense that confidence or significance level is one minus the probability that an observed result could have been produced by mere chance, in the absence of the phenomenon hypothesised. It relates specifically to hypothesis testing.

  23. 273
    Mark says:

    “Mark #258 I read your post a few times but couldn’t make sense of it. The point I have been trying to make is that ascribing a numerical value to “confidence” is spurious. ”

    And I answered it.

    Toss a coin 10 times.

    4 heads. 6 tails.

    Is the coin loaded or not?


  24. 274
    Mark says:

    “I said evidence can become (be turned into) dogma — and sometimes does.”

    [citation needed]

  25. 275
    Chris Dudley says:

    simon abingdon (#252),

    Confidence intervals do result from an understanding or probability but they are a little more complex than your example that you have 50% confidence that the next coin toss will be heads.

    After a series of trials, you may construct confidence intervals on the likelihood that the coin is unbiased with statements like I have confidence greater than 90% that the coin is unbiased within 2% (in 10,000 tosses it is unlikely to give 5,400 heads or higher).

    And, this is why we are hearing the statement that 90% confidence does not mean 10% of the opposite. The coin may be perfectly unbiased. You just have not studied it enough to know that. Confidence statements are statements about how well you have studied the problem and they may improve with further study. With more trials you may be able to say that you have 95% confidence that the coin is unbiased within 2% or tighten the 2% limit to 1% or less.

    So, confidence intervals are about the state of our knowledge as well as the state the the object of our knowledge. The climate up to 2007 is the climate up to 2007 but with more detailed simulations, we may understand better how it was changing. We may end up saying that at least 80% of the warming up to 2007 is human induced with 99.9% confidence as some point since our hindsight can be improved. 90% is not a stopping point, it can be improve upon.

  26. 276
    Doug Bostrom says:

    Oh, boy, another firebrand hurled on the roof by the Guardian’s Monbiot!

    “One of the allegations made repeatedly by climate change deniers is that they are being censored. There’s just one problem with this claim: they have yet to produce a single valid example. On the other hand, there are hundreds of examples of direct attempts to censor climate scientists.

    Where, on the other hand, is a single verifiable instance of a climate denier being silenced by the authorities? They have yet to produce one. But it suits them to cry wolf. They love to imagine that they are important enough to censor. The claim chimes with their paranoid invocation of a great conspiracy – involving most of the world’s scientists, most of the world’s governments, most of the world’s media and a few hundred million others – to suppress the truth about global warming.

    Now we have another marvellous instance of this hypocrisy. Anthony Watts spends much of his time maligning climate scientists and environmentalists on his blog Wattsupwiththat. But while he can dole it out, he can’t take it. As Kevin Grandia of desmogblog shows, Watts has just used US copyright laws to take down a YouTube video which exposes his claims. Grandia has since reposted the video (see above) so you can see for yourself what all the fuss is about.”

    The rest of the story:

  27. 277
    Lynn Vincentnathan says:

    RodB (#215 re #210), you didn’t seem to grasp my point. I was speaking about the long time it takes for an application of a mechanism that causes or traps heat to actually produce enough heat to be perceptible when applied to something really large, like a big pot of water, or the earth.

    I think it’s probably easier for us housewives to understand this than scientists, at least those 50 physicists, and also for us to understand that if GHGs are emitted and if they do act as a heat-trapping mechanism (which is what most scientists say about the natural greenhouse effect and about AGW), then it would not be surprising at all that it takes some time to heat up our atmosphere and the oceans enough to be perceptible. Even if everything else were equal (and we know there are other factors enhancing and suppressing this heating process, like ENSO and solar cycles, etc) there would still be some time delay before things get really hot.

    So I’m not very surprised that climate has not heated up enough to kill off, say, half of life on earth….but I wouldn’t want to keep stoking the fires just because that hasn’t happened yet. Eventually even the watched pot does boil, as you say.

    My suggestion for those 50 physicists is to take their eyes off the pot, get back to whatever work they do, then come back in a few years or a decade or two, and see if there’s been any perceptible (by them) warming — unexplainable by other non-AGW factors like el ninos, solar irradiation cycles, and cosmic rays.

    For them at least, the watched pot never boils; and that’s because they just don’t know how to look at it the way real climate scientists look at it, who can see those teeny tiny bubbles on the pot’s bottom and know the pot will eventually get to a rip roaring boil if we don’t take away the heat-trapping mechanism. You’ve got to be as smart as a housewife to understand this.

  28. 278
    SecularAnimist says:

    Rod B wrote: “If you’re interested in why I think much of climate science has morphed into dogma, ask — though it will probably just make you mad.”

    Interested in WHY you think something has happened, when you have offered NO evidence that it has actually happened, despite repeated requests for such evidence? Well, I’m about as interested in that as I would be in hearing your opinions about “why” NASA faked the 1969 Moon landing. And somehow I suspect that both opinions would have something to do with “liberals” and “big government”.

    At this point you are not making me “mad”, you are making me laugh.

  29. 279
    wmanny says:

    To Simon, Ray, Chris and others chiming in on confidence levels via examples such as marbles and coins, I would refer you once again to the IPCC guidelines which are, after all, more pertinent to the subject of climate research, in the hope that you can clarify what the IPCC intended to convey to the public. I doubt that the guideline authors had the arcana of Bayesian and Neyman-Pearson usages of the term “confidence” in mind, but perhaps they did. In any event,

    In Table 3, we see:
    Very High confidence – At least 9 out of 10 chance of being correct
    High confidence – About 8 out of 10 chance, etc.
    In Table 4, it’s:
    Virtually certain > 99% probability of occurrence
    Very likely > 90% probability
    Likely > 66% probability, etc.

    For those of us who do not have the background to understand the science line by line, a complete read of the report, or merely the Summary for Policymakers, is an exercise in trying to sort out what is significant about all the ‘High’s and ‘Very Likely’s. It would appear that the IPCC is using “confidence” to indicate probability, and for example implying if not stating that Very High Confidence refers to a 1 out of 10 chance of a well defined outcome not having occurred or occurring in the future. Perhaps the authors have subsequently made it clear that they do actually intend the strict, statistician’s interpretation – I have no idea.

  30. 280
    Arthur Smith says:

    John Mashey (#234) – I’ve also been in touch with Bob Austin as a friend and colleague to try to figure out where he’s coming from – do you mind if I send an adapted version of your list of questions to him (or perhaps you have a list you’d like to ask specific to his background in biological physics)? So far he hasn’t found the time to respond to any of my questions about specific science claims he’s made that I find lacking (mainly in a critique of a NAS analysis), so I’m not sure we’ll get much out of him being antagonistic about it. But he’s been friendly enough otherwise.

  31. 281
    Jacob Mack says:

    You know folks, it equally frightens me when people over and under-state AGW and potential future consequences.

  32. 282
    MarkB says:

    Re: #270,

    Thanks, Jim Prall. Your site that lists climate scientists is very useful. I often see climate contrarians cite some petition of “prominent” scientists or what not, then often go on to claim the IPCC contributors are politicians or have few credentials. Your site is an excellent resource to completely refute such claims, rather than having to painstakingly do research on each scientist.

    It’s also a useful site in putting a human face on scientists. Many view the scientific consensus on global warming as coming from elite academies and organizations or buried among the scientific literature, helping feed the false notion that scientists are aloof and the consensus on global warming is a product of a few “political” elites in various organizations. Your site helps dispel such notions and provides an opportunity for anyone interested to examine the work and credentials of any climate scientist and in many cases, allows them to contact them with questions. While ideally this would disarm those who believe global warming is a vast government conspiracy, I think those in this camp would probably not have the patience or inclination to spend much time on your site. There are those who still want to believe in the Easter Bunny.

  33. 283
    Doug Bostrom says:

    David Watt 30 July 2009 at 11:54 AM

    “Having dealt with the 50 APS Society members who wrote to Nature, what is your take on the American Chemical Society who are threatening to throw out their editor for being too partisan in his support for AGW?”

    Is he partisan? Do you think he is? How so? Are the governors of ACS proposing to fire him, as opposed to “dozens of letter writers”?

    Here’s the editorial:

    Here are the letters:

    They seem to be a mixed bag, I think you’re exaggerating the situation just a wee bit.

  34. 284
    simon abingdon says:

    Mark #273 I have done as you asked, tossed a coin 10 times and (satisfyingly) got your predicted result. No the coin is not loaded, although your question probably is. I am very confident indeed that I have never handled a loaded coin. I would be foolish to put a figure on it because I could have no way of justifying it.

  35. 285
    simon abingdon says:

    Chris Dudley #275 Thank you for your clear and helpful explanation. But I am still troubled by “And, this is why we are hearing the statement that 90% confidence does not mean 10% of the opposite” But, so long as confidence remains at “only” 90%, does that not leave the the door open for the opposite to be true with 10% confidence? The 90% confidence in AGW has to coexist with a 10% chance of it being found to be false, otherwise the confidence in it would be greater.

  36. 286
    simon abingdon says:

    Years ago I was in court as witness to a traffic accident when the following memorable exchange took place:
    Defendant: 50% of what the constable says is lies.
    Magistrate: And what about the other 50%?
    Defendant: Some of that’s lies as well.

  37. 287
    Hank Roberts says:

    Simon, no, you’re not getting it. It’s a Statistics 101 thing, and it really does change how you view what you know, once it clicks.

    Robert Grumbine’s site might be a good place to ask, or Tamino’s.
    It really is an issue that could deserve a whole website — if any statisticians are up for teaching the basics to all who might want to come learn.

    There’s a well reviewed introduction to statistics here, in comic format:

  38. 288
    simon abingdon says:

    Hank #287 Thanks for the advice and link which I appreciate. I will soon apply myself to diligent study.
    Two questions for now:
    (1) In the light of #279 do you think the IPCC themselves “get it”?
    (2) Can you help my simple understanding by saying something useful about the “missing” 10% when a 90% confidence is claimed?

  39. 289
    MarkB says:

    David Watt writes:

    David Watt: “what is your take on the American Chemical Society who are threatening to throw out their editor for being too partisan in his support for AGW?”

    I didn’t realize parents of subscribers to a magazine who write a letter had the ability to “throw out” an editor of that magazine.

    “I happened to be browsing through my daughter’s June 8 issue of C&EN…”

    or those who agree with the editor…

    “The more people try to trivialize global warming, the more we and our descendants will suffer the results, some of which have already been quantified (for example, glacier melting and polar ice disappearing).”

  40. 290
    Rod B says:

    David Watt (269) says, “…Do you think there is any risk of the same thing happening on Realclimate?”

    As a resident skeptic, I’ll cast a vote: I certainly hope not. But more significant there isn’t anyone or anything who has the where-with-all to do that, other than the moderators themselves (who unlike the members of ACS own RC hook, line, and sinker) which is an inane idea — until maybe one day they get tired and say fug it (a takeoff from a very old Timothy Leary (??) song ;-) )

  41. 291
    Rod B says:

    Ray (271), I’ll briefly weigh in. You still didn’t answer the question. The question was about a credible mathematical derivation for the number associated with confidence level. You still described a simple judgment call — looking at all of the evidence and subjectively concluding with a high level of personal confidence that it must be highly correct.

    Or maybe I misread your post…

  42. 292
    Hank Roberts says:

    1) yes
    2) no; it really takes a little bit of study to get an idea of what a confidence interval means. I’d say work on understanding the difference between a one-tail and a two-tail test, and why you need more observations for the latter than the former to have the same confidence in the results. That was what made this click for me, long ago.

  43. 293
    Hank Roberts says:

    PS, confidence in statistics has to do with specific observations made over time (how many, over how much time, and how much they vary).

    The IPCC policy summary is not talking about any _one_ study and not reporting on a statistical analysis they did. They’re a review organization — charged with reading all the research and summing up.

    This ought to be obvious. I’d recommend not paying much attention to anyone who’s trying to confuse you about what the IPCC is saying. Look at any annual review publication in any field for similar assessments of the state of the work.

  44. 294
    wmanny says:


    There was a recent NYT article in which Norman Mailer was reported to have used ‘fug’ in “The Naked and the Dead” to get around the Supreme Court. Fuggin’ worked, too. Now to get it past the commentators…


  45. 295
    sidd says:

    I notice a paper by Siddall et al., Nature Geoscience, 26 July 2009.Their model predicts between 7 and 82 cm of sea level rise by 2100. Equilibrium sea level rise is expressed as having an inverse sinh relationship to temperature change over the last 22KYr, and instantaneous sea level rise is governed by a constant, tau, which gives a best fit result between 2.4 and 3.4 Kyr. Can this 3KYr timescale to be reconciled with Andrill data indicating WAIS collapse on the timescale of 1KYr ?

    They point out that the albedo, altitude, and isostatic rebound feedbacks accelerate deglagiation. They fit the data over 22KYr before present. Could it be as Hansen argues, they are modelling the external forcing timescales rather than the internal timescales for ice sheet disintegration ? Or is there something like the Weertman instability shortening the timescale for WAIS collapse in the Andrill data, which is not present in the periods they model?

    I hope the moderators forgive me if I post this on the both the sea ice thread and this one.

  46. 296
    wmanny says:

    Hank, since I’m the one who keeps referring to the IPCC in an attempt to get us away from coins and such, I assume you are referring to me when you mention “anyone who’s trying to confuse you.” If not, never mind, but my intent is just the opposite, actually – I am trying to get to the question about how the IPCC intended its confidence levels and likelihoods to be interpreted, either in the Policymaker summary or in the chapters themselves. My interpretation is that they are not using the strict statistical version of “confidence”, but I am no expert, only a math teacher, and so I am asking. Assuming you have read through the Guidance piece, what is your take? I find it Very Likely if not Virtually Certain that you have one.


  47. 297
    Chris Dudley says:

    simon abingdon (#285),

    Yes, the door is open to the statement being incorrect and the chance of that being the case could be as large as 10% given present knowledge. But the irreducible uncertainty, where future knowledge ends, may arrive at showing the possibility that the statement is incorrect is very much less than 10%. That is why it is safer not to try to turn those statements around without plenty of explanation.

  48. 298
    Deep Climate says:

    Cross-posted at Tamino’s Open Mind and Deltoid as well …

    Here is my take on McLean et al. and associated PR campaign.

    Here is the summary:

    * The McLean, de Freitas and Carter paper presented unsubstantiated conclusions that are contradicted by a cursory analysis of the very data presented.

    * There is widespread agreement among climate scientists that this paper should not have passed review and should not have been accepted for publication.

    * The authors actively participated in a deceptive public relations campaign that trumpeted and exaggerated the paper’s claims, a campaign that even substituted a press release headline for the true title of the paper.

    * The authors permitted an egregious breach of copyright in the dissemination of the paper in its published form at the New Zealand Climate Science Coalition website.

    What more is needed to prod the AGU and the Journal of Geophysical Research to do the right thing? The paper should be withdrawn, and the editor responsible disciplined. Now.

    In explanation of the third point (for those who won’t wade through the whole post):

    That’s right – according to the International Climate Science Coalition website, the title of the paper was – wait for it:

    “Nature not Man responsible for recent global warming”.

    Well, that’s one way to sneak a preferred title past reviewers.

  49. 299
    Mark says:

    “(2) Can you help my simple understanding by saying something useful about the “missing” 10% when a 90% confidence is claimed?

    Comment by simon abingdon”

    Apparently nobody can help.

    Maybe a therapist.

  50. 300
    Mark says:

    “I am very confident indeed that I have never handled a loaded coin. I would be foolish to put a figure on it because I could have no way of justifying it.

    Comment by simon abingdon”

    Then you have no clue about statistics.

    And there are plenty of adult learning colleges where you can find out.

    There you will find enlightenment.

    If you look for it.