Unthreaded #38

I’ve lost track of the number of “unthreadeds” on CA, but the last one at 800 + posts is getting a bit unwieldy. So please continue here. – Anthony


793 Comments

  1. Boudu Moon
    Posted Mar 19, 2009 at 12:48 PM | Permalink

    Could I be the first to congratulate you on a fresh new thread !

  2. OldUnixHead
    Posted Mar 19, 2009 at 12:59 PM | Permalink

    FWIW, this one would be #38. I’m still trying to catch up with the last one!

  3. Craig Loehle
    Posted Mar 19, 2009 at 1:28 PM | Permalink

    Re Chaos on old thread: The problem with determining if the climate system is chaotic is that you need both forcing data and climate data over very long time periods if the chaotic behavior plays out over decades/centuries. There is no possible way to estimate the characteristics of a chaotic attractor from only 50 years of good data (even granting it is “good”). Re: Dan Hughes and “trends”: if there are multiple cycle lengths in the behavior (let us say a 70 year cycle and a 1000 yr cycle for concreteness) you can have a “trend” from the 1000 yr cycle (looks linear, too short of a series to estimate this cycle).

  4. pj
    Posted Mar 19, 2009 at 1:46 PM | Permalink

    This is a request for a recommendation for a good book or books on data analysis. I think a lot of us on this site have some level of knowledge about statistics and can kind of follow what going on, but I for one would be incapable of doing the type of analysis that Steve and others are doing here. I would really like to strengthen my abilities in this area for a couple of reasons. One to have a better grasp on what Steve is doing and to perhaps toy around with doing some of my own work. Second, for career purposes. I’m pretty confident in my ability to teach myself if I have good material to study. Any suggestions?

    • Mark T
      Posted Mar 19, 2009 at 2:25 PM | Permalink

      Re: pj (#4), If you do a search on “component analysis” over at Amazon you’ll get quite a few hits that are worthwhile. I would expect UC, Jean S, and RomanM can provide some details on the better ones of the lot (and a few others in here). The only books I have directly related to multivariate analysis are Independent Component Analysis by Hyvarinen and Oja and Adaptive Filter Theory by Haykin (which is only roughly related to the topics here). Given that Joliffe was widely used in support of the Team, and the subsequent fallout when he deflated their bubble, it might be worth picking up his PCA book for spite. :) I’ve never read it but the Team sure thought he was the authority… for a while at least.

      Mark

  5. Alan Bates
    Posted Mar 19, 2009 at 2:00 PM | Permalink

    I wondered what the significance of the picture was. Then I moved the cursor over it …

  6. Robert Freerks
    Posted Mar 19, 2009 at 2:20 PM | Permalink

    For a stastical book for beginners,I have found Box, Hunter and Hunter, “Statistics for Experimenters–An Introduction to Design, Data Analysis, and Model Building” published by Wiley & Sons in 1978 to be very useful. Statisticians may have additional references, but George E. P. Box is a classic statistician. I believe his most famous quote is: “all models are wrong, but some are useful.” Appropriate to climate modeling as well as everything else.

    • Curt
      Posted Mar 20, 2009 at 12:13 AM | Permalink

      Re: Robert Freerks (#6),

      Robert, I agree that the Box, Hunter, & Hunter text is an excellent starting point for general statistics. However, it is important to remember that the title of the book is “Statistics for Experimenters”, so the emphasis is on well-designed experiments, of which there are (and unfortunately, can be) precious few in climatology.

      You brought to mind a section of the book (14.7, starting p.487) entitled “Hazards of Fitting Regression Equations to Happenstance Data”. It starts:

      “The reader must be warned that, should he admit to some knowledge of least squares and multiple regression, he may be expected to extract information from historical records or “happenstance” data. This risk is especially great should he have the temerity to suggest that a designed experiment is needed to solve a particular problem…”

      It then lists a set of problems in using historical data:

      1. Inconsistent data
      2. Range of Variables Limited
      3. Semiconfounding of Effects
      4. Nonsense Correlation – Beware the Lurking Variable!
      5. Serially Correlated Errors
      6. Dynamic Relationships
      7. Feedback

      I think of this section often when I look at paleo-climatological regression analyses. Obviously, there is no choice in this field but to use historical data. However, I usually get the sense that the authors of these studies act as if they are running a well designed, pre-planned experiment, and do not worry about any of these caveats.

  7. Ian
    Posted Mar 19, 2009 at 3:08 PM | Permalink

    Can I just thank Tom Vonk for a beautifully crafted introduction to chaos at the end of the old unthreaded, it deserves re-posting in this thread, and indeed worthy of it’s own thread. I’ve understood the math for a while but that post glues the bits together for me.

  8. MJW
    Posted Mar 19, 2009 at 4:09 PM | Permalink

    I have a (perhaps dumb) question on “R” that I hope someone can answer.

    I’m having a discussion with someone on the AccuWeather blog about the validity of using correlation to estimate the contribution of CO2 to global warming. I think I’ve mostly convinced him that using ordinary linear regression won’t do, by showing that the correlation between atmospheric CO2 and the Consumer Price Index over 95 years is a rather amazing 0.986. Now the discussion has shifted to correcting for autocorrelation. I want to use the function rls to look at some data, but I don’t quite understand the output. Here’s an example, done with an AR(1) model:

    ———-
    >mod.glsCO2 = gls(Anom ~ logCO2, data=TempVsLogCO2, correlation=corARMA(p=1))
    > summary(mod.glsCO2)
    Generalized least squares fit by REML
    Model: Anom ~ logCO2
    Data: TempVsLogCO2
    AIC BIC logLik
    1004.166 1015.511 -498.0832

    Correlation Structure: AR(1)
    Formula: ~1
    Parameter estimate(s):
    Phi
    0.4216385

    Coefficients:
    Value Std.Error t-value p-value
    (Intercept) -1885.5737 144.14300 -13.08127 0
    logCO2 327.4367 25.01217 13.09110 0

    Correlation:
    (Intr)
    logCO2 -1

    Standardized residuals:
    Min Q1 Med Q3 Max
    -2.64484809 -0.77052068 0.02319532 0.70883718 2.22051057

    Residual standard error: 13.75999
    Degrees of freedom: 128 total; 126 residual
    ———-

    My basic question is whether there’s a way to get the autocorrelation-corrected version of r or r^2 from this output, or to compute it from the glsObject. There’s a section:

    Correlation:
    (Intr)
    logCO2 -1

    But I’m not sure what it means, since neither the sign nor the magnitude make sense for a correlation between temperature and CO2. (For ordinary least-squares, r = 0.874.)

    • bender
      Posted Mar 19, 2009 at 8:46 PM | Permalink

      Re: MJW (#10),
      1. Your question is not an R question. It’s a statistics question.
      2. You engaged in a debate using a methdology you don’t understand. You deserve to lose your argument.
      3. You don’t understand the output because you don’t understand REML.
      4. You don’t see R2 in the output because you are not using OLS.
      5. You don’t understand goodness-of-fit using gls. Regale your friends and demolish your opponents with this enjoyable freshman intro on how to interpret AIC/BIC:

      http://www.cs.cmu.edu/~zhuxj/courseproject/aicbic/sld001.htm

      (Brings a smile to your face, this great thing that Al Gore helped to invent.)
      6. Forget R2. What do you make of your t-values and p-values?
      7. There are no statistics than can save you from spurious correlation between two variables. When a model is mis-specified it’s not autocorrelation that is the main problem.
      8. Sometimes “correcting” for autocorrelation (i.e. trend) leaves nothing worth explaining. That does not necessarily mean there is no causation. It could mean you’ve not got enough data to answer the question.
      9. Is this a regular habit – getting in so deeply in over your head? Doesn’t it hurt?
      10. Send a url. It will be fun to watch you roast.
      11. What IPCC-cited paper are you auditing?

      • Geoff Sherrington
        Posted Mar 19, 2009 at 9:45 PM | Permalink

        Re: bender (#17),
        Bender, you made me read the whole reply to find the sting in the tail at 11. Bully.Re: Andrew (#11),

      • stephen richards
        Posted Mar 20, 2009 at 7:04 AM | Permalink

        Re: bender (#16),

        OOH you an be so so cruel!!!

  9. Andrew
    Posted Mar 19, 2009 at 4:16 PM | Permalink

    Steve really should put all the unthreadeds into the unthreaded category. I accidentally comment on #31. And I would have thought this would be #33. But apparently, I’m wrong.

    Anyway, this is my comment from the wrong unthreaded:
    Anyone feel up to updating this graphic now that more stations have been classified?

    • Geoff Sherrington
      Posted Mar 19, 2009 at 9:51 PM | Permalink

      Re: Andrew (#11),

      Hit the button too soon in 20. Sorry.

      Andrew, if I were able, I’d do an upgrade of that TOBS graph, but I’d do it only to show that it is of little purpose. I do not see much value in adding a modern ending to a fairy tale like “Snow White and the 7 Albedos”, which has adequate charm as it stands.

      • Andrew
        Posted Mar 20, 2009 at 8:26 AM | Permalink

        Re: Geoff Sherrington (#20), perhaps those seven albedos might be better served if Snow White could only find a handsome prince-I can’t think of anyone in particular, though. Perhaps one of the Pielkes?

        Re: bender (#34), I think I beginning to better understand your antagonism toward me (yes, I’m that Andrew) about various forms of internal variability (especially with regards to the PDO-what name had you suggested? Pacific Eignenthingy or something…). It is far to difficult to separate these phenomenon from external (and sometimes technically internal) sources of forcing. I would agree with this although add the caveat that I suspect that any effect they have on one another is small-going both ways. Its just speculation, though.

        • bender
          Posted Mar 20, 2009 at 8:51 AM | Permalink

          Re: Andrew (#35),
          The antagonism was regarding your specific statements on that subject. You don’t exist for me. Only your statements do.

        • bender
          Posted Mar 20, 2009 at 9:02 AM | Permalink

          Re: Andrew (#35),

          It is far too difficult to separate these phenomenon from external (and sometimes technically internal) sources of forcing.

          1. Is PDO really “O” – oscillating predictably, stably? Or is it a chaotic fluctuation? If it itself is an example of an internal chaotic fluctuation then it is truly nothing more than a Pacific EOF that Happened to Oscillate Decadally During Some Limited Period of Observation. Until we know its “cause”, we may call it PEHODDSLPO. Or Pacific eigenthingy for short.

          2. Now add on top of that your favorite forcing. How does PEHODDSLPO respond to forcing? Linearly? Independently? Well, then you are going to have an argument with Tom Vonk, aren’t you?
          .
          You see? I was trying to spare you from the wrath of Tom & Jerry. How is that “antagonistic”? Looks altruistic to me.

        • Andrew
          Posted Mar 20, 2009 at 11:32 AM | Permalink

          Re: bender (#38), As I remember it you were rather frustrated because I didn’t “get” what you were saying, and at the time, I felt antagonized. Fortunately for me, I never did run afoul of Vonk. But thanks for the effort anyway. :)

  10. Andrew
    Posted Mar 19, 2009 at 4:22 PM | Permalink

    MJW-don’t forget to correct for “aerosols” ;)

  11. Howard S.
    Posted Mar 19, 2009 at 7:15 PM | Permalink

    Judging by the heads, those appear to be unthreaded screws, not bolts.
    And are they straight, phillips, hex or square drive heads?

    What data was used to determine they are bolts?
    Were they scanned, measured or modeled?

    Is there any consensus on the boltage status?

    This really appears to be heavily biased in favor of bolts.
    What’s the agenda?

    • Posted Mar 19, 2009 at 7:23 PM | Permalink

      Re: Howard S. (#14),

      They are proxies for threaded bolts.

    • Dave Dardinger
      Posted Mar 19, 2009 at 9:16 PM | Permalink

      Re: Howard S. (#13),

      Judging by the heads, those appear to be unthreaded screws, not bolts.

      Heads don’t determine if something is a bolt or a screw. it’s a bit hard to see whether the heads are actually angled or just flanged, but in either case they’d be sunk in. But the body of the objects don’t have a taper which means they are bolts rather than screws. Most such bolts I’ve used are internal hex bolts. They’re generally used to attach machine parts together.

      • Mike Lorrey
        Posted Mar 23, 2009 at 9:04 PM | Permalink

        Re: Dave Dardinger (#16), screws vs bolts are determined by whether you are screwing them in with a screw driver or a wrench/socket, but mainly by the usage, as defined in the Machinists Handbook:

        Differentiation between Bolt and Screw –
        A bolt is an externally threaded fastener designed for insertion through holes in assembled parts, and is normally intended to be tightened or released by torquing a nut.
        A screw is an externally threaded fastener capable of being inserted into holes in assembled parts, of mating with a preformed internal thread or forming its own thread of being tightened or released by torquing the head.
        An externally threaded fastener which is prevented from being turned during assembly and which can be tightened or released only by torquing a nut is a bolt. (Example: round head bolts, track bolts, plow bolts.)
        An externally threaded fastener that has thread form which prohibits assembly with a nut having a straight thread of multiple pitch length is a screw. (Example: wood screws, tapping screws.)

        The objects in the photograph are neither screws nor bolts, they are pins, or, depending on usage, keys. Calling something an ‘unthreaded bolt’ is akin to calling something, like, say CO2, which lags temperature rises, “anthropogenic global warming”.

        • Dave Dardinger
          Posted Mar 23, 2009 at 10:44 PM | Permalink

          Re: Mike Lorrey (#110),

          All I know is that we used to use a lot of what we called bolts which were like those shown and then inserted into a tapping die and the threads put on them. They would either be inserted into a top piece with a hole slightly larger than the “unthreaded? portion and then screwed into the lower piece, or in some cases both pieces would be threaded manually and the bolt would have the upper part of the shaft machined down so it would only be in the lower portion when tightened down. Perhaps we were using the wrong terminology, but for me screws are tapered and bolts aren’t.

  12. rephelan
    Posted Mar 19, 2009 at 9:52 PM | Permalink

    bender:
    March 19th, 2009 at 8:46 pm
    Re: MJW (#10),

    Good Lord! A Lot of us are learning new stuff here. I haven’t cracked a math book in thirty years.. my fault, not yours, but please be gentle, I may not be cherry, but it hasn’t been used a lot.

  13. Mark T
    Posted Mar 19, 2009 at 10:29 PM | Permalink

    I think bender’s point is that when learning, don’t pretend to be an expert – things that may seem “simple” on the surface are often much deeper, and much more complex than someone learning may be able to grasp in the beginning. It is common enough in the climate science world, that should be enough.

    Mark

    • MJW
      Posted Mar 20, 2009 at 1:17 AM | Permalink

      Re: Mark T (#22), Perhaps you would be so kind as to quote the section where I pretended to be an expert. Was it where I said I was asking a “perhaps dumb” question?

  14. VG
    Posted Mar 19, 2009 at 11:11 PM | Permalink

    I posted this last week (somewere else in CA)

    http://wxmaps.org/pix/temp8.html

    ….remarking that I have observed this map daily for two years and that it hasn’t changed significantly (except for 2-3 days at most). This is a visual assessment of course. We could surmise that cloud cover is to blame and my bet is that it could support Svensmark’s cosmic ray (or muons if you like) effect on lower rain cloud formation (cumulus nimbus mainly due to recent solar activity changes (quiet). Any ideas or feedback?

  15. Posted Mar 20, 2009 at 1:19 AM | Permalink

    Now there’s more Anthony Watts on Climate Audit and more Steve McIntyre on WhatsUpWithThat. ;-)

  16. TomVonk
    Posted Mar 20, 2009 at 4:39 AM | Permalink

    Bender from previous unthreaded

    I think the two – chaos + external forcing trend – can co-exist in a data series, that it is not “contradictory” to suppose this is the case, eiither in a model or in the real climate system. IMV the problem is quantitative: efficient separation/estimation of the two independent components. In a model it is a no-brainer because you specify the deterministic forcing trend and you can study the deterministic chaotic noise in the absence of external forcing.

    I cannot really understand that .
    Or more precisely I can’t bring it in relation with what is known about chaotic systems .
    You seem to consider that a dynamical system (the Earth&water&atmosphere&etc) behaves like a “sum” (?) of 2 independent “components” (?) .
    One would be classicaly deterministic f.ex some function f(t) with no sensibility to initial conditions .
    Another would be deterministic chaos f.ex some function g(t)
    The system would behave like f(t) + g(t) (?)
    Well this is not possible for many reasons .
    .
    First is that when a system is chaotic it is COMPLETELY chaotic – there is no particular equation or coefficient that is “cause” of the chaotic behaviour , it is the whole that is responsible .
    .
    The second is what D.Hughes wrote .
    If what you call “external forcing” is the energy supply to the system (aka the sun) and you switch it off , then the system immediately begins to spiral to a 1 point attractor – an equilibrium where it stops being chaotic . This particular spiral solution gives no information about what happens when the energy supply is on . It may even be non chaotic .
    .
    If what you call “external forcing” is the CO2 concentration despite the fact that it is an internal degree of freedom of the system then you may of course add or substract artificial CO2 sources and look at the 2 behaviours .
    The 2 behaviours will be chaotic as expected because the system is chaotic .
    There is a reason for that and that is precisely because the CO2 concentration is one of the internal degrees of freedom and not some independent variable that can be tinkered with .
    If you add/substract CO2 terms to the equations you will get 2 different chaotic solutions whose difference will mean nothing and certainly not some “trend” (because of the Lyapounov coefficients of course :))
    It is like getting 2 solutions of the Lorentz system by taking 2 different values of a control parameter , both leading to a chaotic regime .
    There will be no meaningful (statistically or otherwise) relationship between the 2 solutions .

    • Craig Loehle
      Posted Mar 20, 2009 at 6:39 AM | Permalink

      Re: TomVonk (#27), Bender is right. You are assuming “forcing” to be constant. what if the sun oscillates its output on a period of X years? Just for argument sake. Even small oscillations in the external forcing can kick the earth climate into different parts of its attractor (assuming there is one) or change the control parameters, as you say, to give a different dynamic. On all time scales the solar forcing has been shown to fluctuate due to orbital shifts and internal sun variability. The earth system could be responding in a simple linear fashion to this external forcing, or be responding as a perturbed nonlinear oscillator itself. How can we tell? We would need long long series of very good data and a better handle on the internal oscillation (e.g., how ocean currents run and switch).

      • Ian
        Posted Mar 20, 2009 at 12:50 PM | Permalink

        Re: Craig Loehle (#28), I think even I can answer that, Tom hasn’t anywhere said that he’s assuming a constant for the external forcing. Indeed I thought it quite clear that this forcing is expressed as a subset of the N equations in the INTERNAL (where INTERNAL refers to the earth chaotic system). Not appearing to be chirlish but I think a few posters in this thread need to go back and read carefully what Tom said elsewhere, e.g. I now accept as Tom says a chaotic system by definition is TOTALLY chaotic even if you convince yourself that there are over some periods something that may look non chaotic.

        • Ryan O
          Posted Mar 20, 2009 at 1:08 PM | Permalink

          Re: Ian (#46),
          .

          Tom says a chaotic system by definition is TOTALLY chaotic even if you convince yourself that there are over some periods something that may look non chaotic.

          .
          That was my understanding of Tom’s points as well. However, extending this to mean that the system has zero predictability and that all variability within the system is indistinguishable from chaotic variability does not seem to Tom’s point. He even explicitly mentions a chaotic system (the earth’s orbit) that can both be predicted quite well within a given window and has variability that is discernible from chaotic variability.
          .
          The conceptual difficulty I have with Tom’s points mainly relate to his comments on the use of statistics. It would seem to me that the variability due to chaotic behavior over a finite time period is bounded. If the bounds are small compared to the effect you are looking for, then statistical analysis would be valid. If you do not know the bounds, then you do not know if your inferences are valid. However, I do not see anything in Tom’s comments that directly support this.

        • Ian
          Posted Mar 20, 2009 at 2:18 PM | Permalink

          Re: Ryan O (#47), I don’t see a contradiction, and I don’t see where the behaviour is bounded over a time period you choose, given that you can’t know how some of the control parameters (with high gain) over that period may suddenly “kick in”. I’d suggest given our learning of the tragectory of the earth over recorded history we can infer a predictability which allows us to determine our likely path out a few thousand or million years, but that merely suggests that we’re using experience rather than math (given we don’t know the equations). On the climate scale there is no reason to suspect it wouldn’t have even more volatile attractors as surely the earths climate is going to incorporate the planet tragetory equations, to which it will add it’s own set of shorter timescale variance. In effect going back to my original point that the energy forcing from the sun is just another set of equations to add to the tragectory of the planet, to which we add what some would call the internal equations which are the NS flows of atmospehere etc, etc. On this basis I can probably produce a model for the next 100 years based on my historical perspective, or I could just guess. In a chaotic system either approach would be as good, but the better bet would be to go with the historical perspective, assume interia in the equations, and hope. I would imagine Tom would agree that for something like climate we end up with climatic chaos equations + sun energy equations + galactic equations, but that the timescale for variability decreases as we move from the galactic to the climatic environment.

        • Ryan O
          Posted Mar 20, 2009 at 2:59 PM | Permalink

          Re: Ian (#49), I am not saying that we know or can infer the bounds with respect to climate (that is a whole separate question); I am saying that bounds must exist for any arbitrary, finite length of time. These bounds could be as generic as basic physical limitations – such as conservation of energy or speed of light to limit the rate of propogation or rate of growth of an effect.
          .
          More importantly, I am saying that while the system may be chaotic, it may have properties that are predictable and to which statistical tools can be accurately applied. If this were not true, then I would be at a loss to explain the predictability of macroscopic fluid dynamic properties. A related question would be how would someone distinguish between predictable and unpredictable properties if only partial information is available.

    • bender
      Posted Mar 20, 2009 at 8:49 AM | Permalink

      Re: TomVonk (#27),

      You seem to consider that a dynamical system (the Earth&water&atmosphere&etc) behaves like a “sum” (?) of 2 independent “components” (?) .
      One would be classicaly deterministic f.ex some function f(t) with no sensibility to initial conditions .
      Another would be deterministic chaos f.ex some function g(t)
      The system would behave like f(t) + g(t) (?)

      Replace “like” with “somewhat like/somewhat unlike” and yes, that’s my presumption (which could be wrong). Here’s why.

      when a system is chaotic it is COMPLETELY chaotic

      There are grades of chaos. It’s the rate of exponential growth that matters. My presumption is that the climate system is an in-between system that is not “COMPLETELY” chaotic. Recall we are talking about spatiotemporal chaos here, not your garden variety type. I am happy to be proven wrong in this or any of my presumptions.

      If you add/substract CO2 terms to the equations you will get 2 different chaotic solutions whose difference will mean nothing and certainly not some “trend” (because of the Lyapounov coefficients of course

      If we are talking weak spatiotemporal chaos, not COMPLETE temporal chaos, then a global mean difference between forced and unforced runs might mean something.

      I am happy that Tom chooses to argue specific points. It’s how we learn.
      .
      Speaking of learning:
      MJW’s rejoinder, if he wanted to zing me, would be this:

      “ok, smartass, why can’t I use the residual sum of squares (RSS) and the total sum of squares (TSS) from the gls object and use that to compute a pseudo R2?”

      And from me he would hear: crickets. Because that’s a good question. Just because the software doesn’t calculate it for you doesn’t mean it isn’t worth calculating for oneself. And just because I’ve never seen that done in the literature doesn’t mean it shouldn’t be done. It shouldn’t be done for nonlinear models (and gls is in package nlme); however MJW’s model is linear, so …

      So inside his “dumb” question there is actually a good question. [It's just really bad form to get into an argument you're not prepared to fight yourself. This is a lesson Dr. Browning taught me once. And it still smarts. Pass it on.]

      • MJW
        Posted Mar 20, 2009 at 2:03 PM | Permalink

        Re: bender (#36), I appreciate your more temperate tone. I replied to your previous comment last night, but it was rejected by the spam filter, and I just thought, “to heck with it.”

        I mentioned several things related to your more recent comment. First, that R^2 is basically an estimate of the amount of explained variance, and that ought to be applicable to GLS. Second, that in his book “A Guide to Econometrics,” Peter Kennedy says on page 144:

        The OLS estimator, by definition, maximizes R^2. The GLS estimator can be used to produce estimates of the dependent variables that can be used to calculate an R^2 that must be less than the R^2 from OLS.

        I reread my original comment, and I still don’t see why you reacted so strongly. Essentially, I said: 1) I don’t believe an OLS estimate of the attribution of global warming to CO2 is reliable, since time-series often have spurious correlations. 2) I was interested in using rls to see what effect correcting for autocorrelation would have. 3) I would like to know if there’s a way to get R^2 for the gls estimator. I didn’t say: 1) I would accept an autocorrelation-corrected estimate as meaningful. 2) Such a correlation would prove or disprove a causal relationship between CO2 and global warming. 3) I don’t understand the meaning of the t-values and p-values; I do.

        Regarding, “It’s just really bad form to get into an argument you’re not prepared to fight yourself.” I’m not sure who I’m fighting. If it’s you, I had no intention of picking a fight. If it’s the other poster on AccuWeather, the original disagreement was on spurious OLS correlations — something I know enough about to handle myself. Once the issue of correcting for autocorrelation arose, I thought I might like to understand the issues better by reading a few articles and experimenting with R.

        • bender
          Posted Mar 20, 2009 at 7:01 PM | Permalink

          Re: MJW (#48),
          What can I say. When I channel TCO I can get to be a real jerk.
          Let’s compute a pseudo R2 and see what it looks like.
          Can you post your data and script?

        • MJW
          Posted Mar 21, 2009 at 2:02 AM | Permalink

          Re: bender (#57)
          Following your earlier suggestion, here’s what I did:

          > mod.lmCO2 = lm(TempVsLogCO2$Anom ~ TempVsLogCO2$logCO2)
          > mod.glsCO2 = gls(Anom ~ logCO2, data=TempVsLogCO2, correlation=corARMA(p=1))
          > tss.CO2 = sum((TempVsLogCO2$Anom - mean(TempVsLogCO2$Anom))^2)
          > rss.lmCO2 = sum(mod.lmCO2$residuals^2)
          > rss.glsCO2 = sum(mod.glsCO2$residuals^2)
          > r2.lmCO2 = 1 - rss.lmCO2/tss.CO2
          > r2.glsCO2 = 1 - rss.glsCO2/tss.CO2
          >
          > tss.CO2
          [1] 98832.05
          > rss.lmCO2
          [1] 23313.96
          > rss.glsCO2
          [1] 23318.91
          > r2.lmCO2
          [1] 0.7641052 (which, of course, equals the lm r2)
          > r2.glsCO2
          [1] 0.7640552

          So it made very little difference. I'm surprised correcting for autocorrelation had so little effect.

          I computed the Durbin-Watson statistics for mod.lmCO2, which seem to show it's not really an AR(1) process.
          > durbin.watson(mod.lmCO2, max.lag=10)
          lag Autocorrelation D-W Statistic p-value
          1 0.39517835 1.192171 0.000
          2 0.11266619 1.745344 0.178
          3 0.10393177 1.720665 0.130
          4 0.21687309 1.474126 0.004
          5 0.10071407 1.688347 0.124
          6 0.15393559 1.572172 0.034
          7 0.18854371 1.501515 0.014
          8 0.18684536 1.483421 0.018
          9 0.06451483 1.727710 0.350
          10 0.12263674 1.545294 0.064

          I tried it as an AR(8) process, and the pseudo-r2 was 0.7541054.

          I plan later to try modifying a script Steve McIntyre wrote for Lucia to do the Cochrane-Orcutt method so I can compare the results.

        • bender
          Posted Mar 21, 2009 at 8:11 PM | Permalink

          Re: MJW (#60),
          Now you know why I said what I said about autocorrelation not being your problem. The shared trend is the problem. The autocorrelations in a trended data series do not damp quickly. They extend way out. Your process is not AR(8); it’s non-stationary in mean. Your effective degrees of freedom is very, very low. You don’t have a lot of data to go on.

        • MJW
          Posted Mar 22, 2009 at 12:25 AM | Permalink

          Re: bender (#71), Bender, the main reason I was interested in the effect of accounting for autocorrelation is to make sure it wouldn’t undermine my counterexample to the CO2-temperature correlation, so I’m actually pleased (though still somewhat surprised) at how little it matters.

          It may be kind of silly and pointless, but I’ll explain the discussion on AccuWeather, if for no other reason than to show I’m probably on the same page as you, even though I know considerably less about the page.

          A person mentioned a blog posting by someone else that supposedly proved that 76% of the 20th century global warming was caused by CO2, based on an (OLS) correlation. I replied that time-series regression was a dubious basis for such a claim, especially when both series are basically monotonically increasing, and gave the almost perfect CO2 vs. CPI correlation as a counterexample. The other person cross-posted part of my comment to Tamino’s website, asking for an opinion. Tamino replied that my comment made no sense, and that time-series correlation statistics are “valid” as long as autocorrelation is accounted for. (I’m not certain what “valid” implies in that context.) The person who originally did the correlation redid it using Cochrane-Orcutt to correct for autocorrelation, and now says CO2 accounts for 60% of the variance. That’s what got me interested in the effect of autocorrelation on the variance.

          Given the difference between the effect of correcting for autocorrelation that I saw and the effect he saw, I think that the person who did the correlation may have made a mistake in applying the Cochrane-Orcutt method, but until I try it myself, I can’t be sure.

          In any case, the residuals for the CO2-CPI regression are highly autocorrelated: they’re are basically a smooth function. However the correlation between temperature and the CPI is almost identical to that between temperature and CO2, and the behavior of the residuals is very similar. My point, by the way, isn’t that the relationship between temperature and CO2 is spurious just because it’s similar to a nonsensical correlation; it’s just that I doubt that regression, even if corrected for autocorrelation, can be used to determine whatever relationship there is.

        • bender
          Posted Mar 26, 2009 at 1:26 AM | Permalink

          Re: MJW (#79),
          The “explained” variance estimated in this case is nonsense. Nothing is “explained” when all you have is a correlation between two similarly non-stationary time-series. If the time-series are stationary, Tamino has a point. As they are not, he does not. He is flat wrong. Detrend the series by taking first differences and correlate the first differences. What is your new correlation? What does Tamino make of this new result?

        • MJW
          Posted Mar 27, 2009 at 2:42 AM | Permalink

          Re: bender (#136), I tried using differencing, and the results were interesting (at least to me).

          For the 128 year temperature vs. CO2, R2 was only 0.03384, but p was still significant at 0.0384.

          For the 95 year temperature vs. CPI, R2 was a tiny 0.002076, and p was a decidedly insignificant 0.663.

          For the 95 year CO2 vs. CPI, R2 was a still high 0.4418, and p was an outlandishly significant 2.77*10^-13. So either CO2 causes inflation (or vice versa) or the somewhat exponential shape of both original sets of data made first differencing less effective.

        • Posted Mar 26, 2009 at 2:53 AM | Permalink

          Re: MJW (#79),

          Tamino replied that my comment made no sense, and that time-series correlation statistics are “valid” as long as autocorrelation is accounted for

          There are some opposing views; see for example McGuirk&Spanos 2004

          The practice of adopting the alternative model when the data reject the no autocorrelation assumption is often inappropriate. The problem is that the presence of residual autocorrelation is interpreted as evidence for the ACLRM. This is an example of the classic fallacy of rejection : ‘evidence against the null is interpreted as evidence for the alternative’. The ACLRM is presumed to be ‘the’ appropriate model, even though the residual autocorrelation could have arisen in numerous alternative ways, one of which is that the error follows an AR(1) process. It goes without saying that if the appropriate model is not the ACLRM, the OLS estimator is no longer unbiased or consistent, and the ACLRM simply constitutes another misspecified model; see Spanos (1986).

          ..

          Mizon (1995), in a paper entitled “A Simple Message to Autocorrelation Correctors: Don’t,” elaborated on these sufficient conditions and recommended that the traditional way of ‘correcting for serial correlation’ is a bad idea. Unfortunately, that advice is largely ignored by the recent applied econometrics literature. As we will demonstrate, the consequences are very serious in terms of the reliability of inference based on such models.

        • MJW
          Posted Mar 27, 2009 at 3:14 AM | Permalink

          Re: UC (#137), Thanks for that very informative comment and for the linked-to paper. It certainly supports what Bender’s said and what I would intuitively believe. I can think of lots of reasons why the residuals would be autocorrelated that have nothing to do with ARMA processes.

          When I began looking into this (lo those many days ago), I was surprised that almost every thing I read had to do with correcting for ARMA type processes. I sort of wonder if the emphasis on correcting for AR(n) processes is just because it’s pretty easy to do. Like the joke about the guy who looks under the streetlight for keys he’d lost elsewhere, because the light’s better.

  17. Posted Mar 20, 2009 at 7:35 AM | Permalink

    Yes, Bender’s comment seem quite reasonable to me. For example, the parameter r in the Lorenz equations is a forcing parameter. You could solve the equations with r=r(t) for example r = a + b t and then you would see chaotic oscillations around a trend. As Bender says, the difficulty would then be extracting the linear trend from the chaos if you were presented with the data but not the rules for how it was generated. Of course Tom is also right that the output is not simply the sum of the two effects.

    This may well be what’s going on in the climate. We are trying to determine whether there is a long-term trend due to CO2 or solar variation but it’s very difficult because of the chaotic fluctuations.

    Can I remind people about the Message Board. It’s much better than unthreaded because you can keep topics separated and you can go back and edit your posts! Why not start a chaos thread there if people want to discuss this further?

    • Posted Mar 20, 2009 at 7:51 AM | Permalink

      Re: PaulM (#30),

      I already have some results for one form of r(t). Who will decide if we need to go the the Message Board?

      Generally, the energy source and the dissipation are in very (extremely) delicate balance for all ODE systems that exhibit chaotic response. The motion is very easily shut down, or diverge. And if the ‘climate system’ is the subject, the forcings are not usually monotonic functions of time; more like oscillatory. The parameter r is the energy addition in the Lorenz system.

    • TAG
      Posted Mar 20, 2009 at 8:02 AM | Permalink

      Re: PaulM (#30),

      Like the path of a pendulum located in a rising elevator? Is this what Bender means?

    • Posted Mar 21, 2009 at 1:36 PM | Permalink

      Re: PaulM (#29), and a few others
      If the Ra parameter is taken to be a function of time, or any parameter in any ODE system that exhibits chaotic response in the absence of dependency on time, then you can’t say anything general about what the response of the modified system will or will not be.

      The theoretical basis of chaotic response of systems of non-linear ODEs is completely based on autonomous equation systems. That is, the right-hand side functions are not functions of time. For one aspect, the generality of arbitrary functions means that nothing in general can be established relative to the numerical values and behavior of the solutions. If Ra = A+Bt, for example, if B is taken to be negative, eventually the motions will cease.

      If the Lorenz system is taken as some kind of fuzzy template for any real-world physical system, it seems to me that the basic characteristics of the Lorenz system relative to the physical system is an important consideration. The parameters in the Lorenz system are constants. The Ra parameter accounts for constant energy input into the Lorenz system, and is absolutely essential for maintaining the motions of the system. The Lorenz system does not exhibit chaotic response for arbitrary values of the parameters.

      Additionally, it is well-established that ill-suited numerical solution methods when applied to ODEs and even PDEs that cannot exhibit chaotic response will in fact calculate chaotic response (or at least numbers that ‘look like’ chaotic response. And finally, discontinuities in any aspect of the discrete approximations to the continuous equations, easily produces response that ‘looks like’ chaotic response. And I do mean any aspect at all; the discrete grid(s), any parameterizations, coding bugs, the continuous equations, … everything.

      None of these hold if the real-world physical system is the weather or the climate. There is not constant energy input to the system. At times there is energy loss from the system. The phenomena and processes in the real-world correspond to the parameters in the Lorenz system are not constants. They are in fact functions of time.

      I have the Lorenz system, and several other chaotic ODEs, coded for solution by several numerical methods. I have calculated the Lorenz system with the Ra parameter being taken to be sine and cose. Let me know if you want to see the results of these calculations or any others in which the parameters are taken to be functions of time.

      All corrections will be appreciated.

  18. David L. Hagen
    Posted Mar 20, 2009 at 8:00 AM | Permalink

    For Freedom of Information Act (FOIA) requests for code and data, see new guidelines by Attorney General Holder:

    Attorney General Issues New FOIA Guidelines to Favor Disclosure and Transparency

    http://www.usdoj.gov/opa/pr/2009/March/09-ag-253.html

    Memorandum for Heads of Executive Departments and Agencies Re The Freedom of Information Act (FOIA)

    http://www.usdoj.gov/ag/foia-memo-march2009.pdf

  19. bender
    Posted Mar 20, 2009 at 8:16 AM | Permalink

    I never said the two components, external forcing & internal chaos, could be “summed” independently. In fact, I was thinking of explicitly stating that that is NOT what I think happens, figuring that Tom might protest. The forcing gets folded into the trajectory and becomes part of the chaos. That’s why I say statistical partitioning of the convolution would not be easy.

  20. bender
    Posted Mar 20, 2009 at 9:04 AM | Permalink

    We need to close “unthreaded” for comments, otherwise people will keep posting there.

  21. Howard S.
    Posted Mar 20, 2009 at 9:43 AM | Permalink

    How typical. So they can be screws or bolts? Perfect.
    Warming or cooling, screws or bolts, I guess we can’t reach any scientific conclusions.

    Obviouosly those could be bolts or screws. It’s hard to tell.
    But the fact remains the owners refuse to release any proof of either.

    Not to mention the complete absence of any concern for nuts, washers and related fasteners.
    As if they don’t matter?

  22. Harry Eagar
    Posted Mar 20, 2009 at 12:13 PM | Permalink

    Lorenz, in his Danz Lectures, said he didn’t know whether climate is chaotic or not.

    I would say that, over the long term, it is antichaotic. If Lorenzian chaos is the situation in which unobservably small inputs result in unpredictably large outputs, then Earth climate is a system in which huge changes in inputs result in minimal changes in outputs.

    Within the minimal change regime, the variations seem to be pretty hard to predict, but I don’t see that as mathematically chaotic. It just means we don’t have 1) enough data; and 2) a good theory.

    • bender
      Posted Mar 20, 2009 at 9:35 PM | Permalink

      Re: Harry Eagar (#43),

      I would say that, over the long term, it is antichaotic.

      Re: Ian (#46),

      I now accept as Tom says a chaotic system by definition is TOTALLY chaotic even if you convince yourself that there are over some periods something that may look non chaotic.

      Personal opinion aside (contrast Harry Eagar’s opinion vs. Ian’s), I think the question is what does Hansen mean by the “cascade” of chaotic variability “across all time scales” and what do IPCC reports have to say about this “cascade” process? People say “climate” is non-chaotic due to some bounding that happens over multi-decadal to secular time scales. What is the proof? Opinions are fine, but IPCC reports are supposed to summarize known facts and assess the uncertainties. So where’s the discussion in the 4AR about chaotic uncertainty and how that figures into the statistics of attribution of trends over decadal to secular time scales? Am I the only one who thinks something is missing here?
      .
      Gavin, comments?

    • Jaye
      Posted Mar 20, 2009 at 11:18 PM | Permalink

      Re: Harry Eagar (#43),

      You don’t understand chaos then. Attactors or fixed points describe “predictable” end states.

  23. Ryan O
    Posted Mar 20, 2009 at 12:21 PM | Permalink

    This is in reply to Tom Vonk here: http://www.climateaudit.org/?p=4804#comment-333518
    .
    I hate looking foolish, but I hate not understanding even more. This took awhile to digest. Even Hamiltonians and Jacobians are a distant memory, and I have never done any math related to chaos theory. So at the risk of looking foolish I will present my understanding of what you are saying and allow all comers to tear it apart. :)
    .

    The relationship between chaotic systems and statistics. An easy answer would be that there is none and leave it at that. Indeed chaotic systems and especially those where the control parameters vary have no means, no standard deviations, no trends and no probability densities.

    .
    Unless I am not understanding what you mean, this seems very counterintuitive as a general property. The first example is airflow over a wing.
    .
    Even “laminar” flow is chaotic. I may not be able to predict the path of a given molecule as it transitions over the wing, but I can predict macroscopic properties, such as lift. I can predict these properties as I vary parameters – such as speed and angle of attack – and I can place confidence intervals on the results of experiments. I cannot predict the exact behavior of vortex shedding, but I can predict the average properties of the vortices and the corresponding average contribution to drag.
    .
    If I were to increase the angle of attack such that the flow becomes turbulent, however, I can no longer predict these quantities. My layman’s interpretation of this type of behavior has always that the reason is due to the magnitude of the contribution of chaotic behavior to the overall variability of the system. In laminar flow, the magnitude of the contribution of the chaotic behavior to system variability is small compared to the properties I am investigating. In turbulent flow, it is not.
    .
    Now for the second example, which is the flow of smoke from a cigarette in still air. This flow, too, is chaotic. However, I still have a window in which I can describe the macroscopic behavior of the smoke plume. I can measure the width of the plume, the flow rate, the direction . . . and I can even assign confidence intervals for a limited distance (or time). At some point, the turbulence in the flow will grow large enough to make these characterizations impossible. But there is a limited window in which I can make those characterizations and I can predict those aspects of its behavior.
    .
    In other words (in my layman’s interpretation), for any finite time period, the range of possible states (range of variability) due to chaotic behavior is bounded. The system behavior cannot be predicted with an uncertainty less than those bounds, but some predictions may still be made and statistical inferences drawn.
    .
    From this, I would conclude that it is not categorically impossible to predict certain properties of the climate, although which properties could be predicted and what the timeframe would be is not known. If we were to find the set of equations that describe the climate, however, we could (in principle) determine which properties we could predict and the window of time during which we could make valid statistical calculations about those properties.
    .
    .
    .
    Shifting gears a bit, I would think that the biggest problem isn’t the predictive ability of the models. This would be important, yes, but it would seem the far more fundamental problem is the problem of attribution of cause. We neither know for certain whether climate is chaotic nor what the bounds of the chaotic behavior are because we do not have a full set of equations to describe it. Because of that, there is an unquantifiable possibility that attributing all or part of the current temperature change to anthropogenic sources is in error.

    • Craig Loehle
      Posted Mar 20, 2009 at 12:39 PM | Permalink

      Re: Ryan O (#44), Pretty astute analysis Ryan. In contrast to mathematical chaos, in the climate system energy dissipation means that the effect of a small perturbation winds down over time. Even hurricanes dissipate after a week or so. The problem Lorenz was addressing was small scale climate predictability. There is general agreement that this problem is fundamentally unsolvable beyond some modest horizon like a week or 2. This does NOT prove that the entire climate system (i.e., global temperature) is chaotic. Neither has anyone proved that there is not SOME chaos in it (e.g., El Ninos, PDO, etc). The climate models assume that Lorenzian weather uncertainty averages out over time in the same way that the ripples in a stream do not affect the larger scale average velocity down the mountain of the water. This is not an outrageous position, but the question of the magnitude of the true macroscale PDO-type effects is completely left out by this assumption.

  24. Terry
    Posted Mar 20, 2009 at 2:57 PM | Permalink

    Hi y’all. This is off the current topic but I wanted to put this out there. I’d like to see someone with a satistical and/or scientific background evaluate the value of this “scientific” expedition.

    http://www.catlinarcticsurvey.com/

    It is funded by Prince Charles (the UKs version of Al Gore), the Wprld Wildlife Federation and the United Nations Environment Programme. UNEP is the group that formed the IPCC with its preconceptions and assumptions of agw.

    You’ll see on the main page of their website that this is the goal of the mission:

    “A pioneering scientific expedition to help determine the lifespan of the Arctic Ocean’s sea ice cover”

    I think it is a joke myself. Some of the stuff on the website is just hilarious, considering that they are trying to present this as a serious scientific enterprise.

    You may want to start a seperate thread on this.

    • romanm
      Posted Mar 20, 2009 at 3:09 PM | Permalink

      Re: Terry (#50),

      It’s already “out there”. Try searching the previous threaded for “Catlin” to see several comments about the expedition. It’s often a good idea to do this before posting a “new” older topic.

  25. TommyS
    Posted Mar 20, 2009 at 3:02 PM | Permalink

    I once watched a facinating chaotic system. We were climbing near the sea, in a small canyon with a dead end. The wind speed were variable, and a cross section of the canyon had a slight V profile.

    What happened was that I lost a very thin and elastic plasticbag while having a lunch at a small ledge. There were four main players in the following hour:
    _
    a. The gravity pulling the bag to the ground.
    b. The bag’s variable size.
    c. The local scale wind turbulence.
    d. The larger scale wind turbulence. (Felt as variable wind speed in the canyon)
    _
    I have absolutely no problem with accepting a chaotic system that will stay within bounds.

  26. Tim
    Posted Mar 20, 2009 at 3:31 PM | Permalink

    I HAVE NO CLUE WHERE TO PUT THIS:

    UNIVERSITY OF MONTANA APPROVES CLIMATE CHANGE MINOR

    http://hosted.ap.org/dynamic/stories/M/MT_UM_CLIMATE_CHANGE_MINOR_MTOL-?SITE=MTBOZ&SECTION=HOME&TEMPLATE=DEFAULT

    *gag*

    • Andrew
      Posted Mar 20, 2009 at 7:23 PM | Permalink

      Re: Tim (#54), In India, I recall reading that you can Major in Global Warming. Potential Required reading: An Inconvenient (Pack of Lies) Truth.

      Hm, maybe I can major in GW and then use it to get a job “denying”. I have to find some way to cash in, now that the oil money’s dried up. ;)

    • AnonyMoose
      Posted Mar 21, 2009 at 10:38 PM | Permalink

      Re: Tim (#53), “UNIVERSITY OF MONTANA APPROVES CLIMATE CHANGE MINOR”

      There are things a minor should not do. :-)

  27. Andrew
    Posted Mar 20, 2009 at 9:52 PM | Permalink

    Any comment Re: this study?

    http://www.geo.lsa.umich.edu/~shaopeng/2008GL034187.pdf

    someone gave me the link suggesting it verifies the spaghetti graph picture of climate. I’m dubious. Has this paper been dissected yet? Will it be?

  28. rephelan
    Posted Mar 20, 2009 at 11:46 PM | Permalink

    Andrew:
    March 20th, 2009 at 9:52 pm

    There’s a discussion going on right now over WUWT


    The science in the paper is way over my head but I did submit the following comment over there:

    Thank you. I actually had that one from another source already, so now I’ve read it twice. The discussion of heat flux is really quite beyond me but there were a few points in the paper that stuck out:

    1. It is not presenting any new data or clarification of methods;
    2. It was written specifically to reconcile their earlier publications with the IPCC 2007 report;
    3. They appear to be claiming that their results are consistent with the instrumental record, which Anthony Watts’ Surface Station Project seems to be showing is badly flawed, and various proxy records which have been criticized in great detail on their own merits (e.g. bristlecone pines).

    I’d have to learn far more about Prof. Huang’s science that I really want to address the adequacy of his temperature reconstructions, but it sure looks like another set of proxies that correlate well with another set of semi-discredited proxies. Not a smoking gun.

    William Connelly also has a delightfully vicious commentary on the paper as well. It is probably a topic that should be addressed all on its own.

    • Andrew
      Posted Mar 21, 2009 at 7:11 AM | Permalink

      Re: rephelan (#63), That’s were I got it! Hm, I wonder if Steve will comment-perhaps he has in the past?

  29. Jan
    Posted Mar 21, 2009 at 8:58 AM | Permalink

    [snip - Steve has deleted the Gerlich and Tscheuschner paper a number of times from CA since it often provokes some vicious arguments. I'm in agreement with a number of folks that it is mostly unsubstantiated- The CA Zamboni]

    • Dave Dardinger
      Posted Mar 21, 2009 at 11:02 AM | Permalink

      Re: Jan (#67),

      Well, they sure don’t pull any punches! I expect Steve won’t allow much discussion of this article here as it’s not in his area of expertise. I’d be interested in knowing where it may be being discussed, however, as it’s taking on directly the basic physics or lack of it for the GHG theory.

      More after I’ve read / skimmed the entire article.

  30. Bill Illis
    Posted Mar 21, 2009 at 1:46 PM | Permalink

    Craig Loehle’s new paper published in Energy and Environment on ocean heat content from the Argo buoy data is going to generate a flurry.

    When he told us to stay tuned for the paper a few weeks ago, he didn’t hint it was going to be such a blockbuster.

    The ocean heat content (at least in the upper 700M) is continuing the decline that started in 2003.

    Chart from Jennifer Marohasy.

    Dr. Loehle is going to face a lot of criticism from the pro-AGW crowd now (since the undertaker has put away the hammer after the final nail went into the coffin).

    The modelers desperately needed the oceans to be absorbing heat content away from the surface (absorbing the forcing which would have normally heated the surface) since none of the other “cooling” forcings seem capable of explaining what is happening to the climate right now.

    Aerosols can’t work since there is greater warming in the areas which are affected the most by Aerosols (see Los Angeles and Beijing temps), apparently the Sun doesn’t vary enough to cause such cooling, there is no volcanic forcing right now, and Phil Jones has resurrected the Urban Heat Island effect from the days of effectively Zero.

    The only thing they had left was the oceans absorbing a large amount of heat from the surface (in fact, more than the GHG forcing itself to actually cause a decline in surface temperatures). Nope (see undertaker comment above).

    [I guess they could now go to the very deep ocean is absorbing some of the GHG forcing but there is no known mechanism for the deep ocean to warm while the upper third is cooling and any such warming of the deep ocean is supposed to take many, many decades and even hundreds of years].

    Thanks Craig. Will provide whatever support I can when the AGW crowd gets going.

    • bender
      Posted Mar 21, 2009 at 8:33 PM | Permalink

      Re: Bill Illis (#66),

      The only thing they had left was …

      Wrong. They still have “radiative imbalance” and its corollary: committed warming “in the pipe”.

      • Andrew
        Posted Mar 21, 2009 at 9:36 PM | Permalink

        Re: bender (#72), warming in the pipe is dead, it would seem:

        http://climatesci.org/2009/03/05/is-there-climate-heating-in-the-pipeline/

      • Ron Cram
        Posted Mar 22, 2009 at 12:42 PM | Permalink

        Re: bender (#72),

        Radiative imbalance shows up as warmer oceans. If ocean heat content is not increasing, there is no warming in the pipe and therefore no radiative imbalance.

        • bender
          Posted Mar 26, 2009 at 1:18 AM | Permalink

          Re: Ron Cram (#90),
          Wrong. Heat lost to space can lead to radiative imbalance, independent of ocean heat content change. To equate OHC and radiative imbalance is to ignore heat loss to space, which some think may be significant – for example during major El Ninos such as 1998.

        • Neil Fisher
          Posted Mar 26, 2009 at 3:28 PM | Permalink

          Re: bender (#135),

          Wrong. Heat lost to space can lead to radiative imbalance, independent of ocean heat content change. To equate OHC and radiative imbalance is to ignore heat loss to space, which some think may be significant – for example during major El Ninos such as 1998.

          I must be missing something. I thought a radiative imbalance was:
          TOA incoming – TOA outgoing 0.
          And further that, over time, eveything below TOA must accumulate (or shed, as appropriate) Joules – that this was the “source” of global warming (or cooling, as the case may be). Surely heat lost to space adds to TOA outgoing, thus affecting total (im)balance? There may be other reasons to suggest that OHC does not directly correlate with TOA radiative imbalance (such as melting ice, mechanical work etc). Please bender, tell this oaf what he’s missed, because it seems to me that you are suggesting that heat lost to space is independent of radiative imbalance, which is, well, non-sensical to me – but hey, I’m not a climate scientist.

        • Ron Cram
          Posted Mar 26, 2009 at 11:05 PM | Permalink

          Re: Neil Fisher (#154),

          You’ve got it right.

      • Ron Cram
        Posted Mar 22, 2009 at 12:43 PM | Permalink

        Re: bender (#72),

        Oops. I see Andrew already commented. I did not mean to pile on.

  31. Anthony Watts
    Posted Mar 21, 2009 at 2:29 PM | Permalink

    I just did a writeup on this, I’ve found a similar trend as demonstrated by Loehle in Mauna Loa CO2 data

    http://wattsupwiththat.com/2009/03/21/recent-ocean-heat-and-mlo-co2-trends/

  32. Raven
    Posted Mar 21, 2009 at 3:26 PM | Permalink

    Jennifer Marohasy has a thread on the G&T paper.

    http://jennifermarohasy.com/blog/2009/03/atmospheric-greenhouse-effect-falsified-again

  33. VG
    Posted Mar 21, 2009 at 9:33 PM | Permalink

    This is a MAJOR paper. Surprised when the first version came out some attacked is as quackery so now it has definitely been published in a major Physics Journal http://arxiv.org/abs/0707.1161. This debunks the WHOLE concept of AGW and the physics behind it!

  34. Raven
    Posted Mar 21, 2009 at 9:51 PM | Permalink

    A warmer claimed that the PDO is caused by Rossby Waves.

    http://www.noc.soton.ac.uk/JRD/SAT/Rossby/Rossbyintro.html

    http://ams.allenpress.com/perlserv/?request=get-abstract&doi=10.1175%2F2008JCLI2593.1

    http://adsabs.harvard.edu/abs/2008AGUFMGC33A0748S

    I don’t have access to the papers behind the paywalls but I was wondering if anyone has more thoughts on this?

    • Ron Cram
      Posted Mar 22, 2009 at 12:53 PM | Permalink

      Re: Raven (#74),

      What causes the PDO to switch from warm to cool is probably a less important question than: what does the PDO cause? If Spencer is correct that the PDO is involved in the formation of clouds, then the cool phase of the PDO could prevent much of the sun’s warmth from reaching the planet. This would explain how the PDO could be involved in a 30 year trend of warming or cooling. It may be the best explanation submitted so far for explaining natural climate variability. But I still think other factors are at work we have not even considered yet. If the MWP was several degrees warmer than today, as I think is clear from the ecological data such as Steve presented at ICCC, we simply do not yet understand the mechanisms of natural climate variability between the extremes of MWP and LIA.

      • Andrew
        Posted Mar 22, 2009 at 3:10 PM | Permalink

        Re: Ron Cram (#92), Surely you don’t mean several degrees globally? If that were the case we’d already be long on our way to a glaciation! Personally, I think global peak MWP was maybe a degree, probably less warmer than the present. But it does seem to be a very understated oscillation.

        • Ron Cram
          Posted Mar 22, 2009 at 5:53 PM | Permalink

          Re: Andrew (#93),

          I think the amount of natural variability the earth has seen is very poorly understood. I think we have been cooling for several years but the MWP was probably two or three degrees warmer at high latitudes than at our warmest. How that might compute to global temps, I am not sure and I am not sure anyone is. I also think the earth was warmer in the 1930s and early 1940s than in the 1990s. It is difficult for me to look at the dust bowl years of the 1930s and say they were cooler than what we have seen in the 1990s. It is easy to blame the unwarranted adjustments of GISTEMP and HadCRUT3 for the misconception.

          Re: Alexander Harvey (#94),

          I have not had a chance to read the paper you linked yet, but I think the trend approach you are using may be off-track. When Pinatubo erupted, the oceans cooled almost immediately and it took about five years for the oceans to warm up to former levels. It also took the atmosphere about five years to clear all of the particles responsible for blocking the Sun. Once the particles were clear, the oceans were back to their old temps. If the earth were out of radiative balance, we would see a steady year over year increase in OHC.

        • Alexander Harvey
          Posted Mar 23, 2009 at 9:24 AM | Permalink

          Re: Ron Cram (#95),

          Ron & curious,

          I have done my best and using the trend from 1997-2003 (1997 is effectively after pinatubo according to the GISS forcings and 2003 is the last year of the GISS forcings) I can get the flux up to 0.57W/m^2 (whole earth surface). Taking the average forcing (1997-2003) I get 1.80 W/m^2. The difference being 1.23 W/m^2. I have had to eyeball the figures to get values but I think I have been fair.

          Now it gets tricky, how much did the temperature go up. To get a temp for 1880 (0 forcing in the GISS data) I have taken the average HadCRUT3 for the 10 years centred on mid 1880 and I get -0.26C and again the aveage for the period 1997-2003 for which I get 0.40C the difference being 0.66C.

          This gives a sensitivity of 0.54C/(W/m^2) or (if you take the forcing due to a doubling of CO2 as 3.7W/m^2) 2.00C/doubling.

          In doing this I think I have steered away from many pitfalls and I have tried to get a high value for the sensitivity. I can do a little more if I took a higher value of temperature increase by using the average temp for 2003 but that would still only get me to 2.17C/doubling. I realise that all of this is going to be end-point sensitive but I have done what I can.

          Now working the other way I find that 3C/doubling would have required the oceans to be averaging an uptake of 0.96W/m^2 or a trend of 1.57*10^22J/yr.

          Now the record does exceed that briefly for the 2 yr interval 2001-2003 at ~2.75*10^22J/yr). But on the other hand it is almost flat since 2003.

          Alex

        • Ron Cram
          Posted Mar 23, 2009 at 10:22 PM | Permalink

          Re: Alexander Harvey (#105),

          I do not claim your math is wrong, it is probably correct. I think a trend analysis is the wrong way to analyze the data. Perhaps you read my post in #103 and found it poorly worded. I will try to explain my thoughts differently. A radiative imbalance must cause a warmer ocean in the same year the imbalance occurred.

          Accountants understand there are limits to trend analyses of a business. A trend analysis of the type you are doing would show all of the big banks in perfectly fine financial shape. If you really want to know the profit for a period, you have to match revenues and expenses. Accountants are forced to use Generally Accepted Accounting Principles – GAAP – rules which are determined by the Federal Accounting Standards Advisory Board. Scientists should be required to operate under and present their findings under Generally Accepted Scientific Principles – GASP – rules. Such a practice would allow consumers of the research to understand if a physical relationship actually existed and what the relationship meant.

          Heat can be said to be “in the pipeline” when the oceans are warmer than usual. We would expect this heat to be released into the atmosphere eventually. But radiative imbalance cannot be “in the pipeline” and show up in a warmer ocean years later. If the earth’s energy budget is out of balance, the oceans will warm the same year. This is the important fact trend analysis obscures.

          When scientists are looking for a relationship between radiative imbalance and an increase in ocean heat, they must state their findings in relation to the finite period in which they may be seen. The difference in ocean heat content between 2002 and 2003 would tell you the radiative imbalance for 2003.

          Try this. Take a look at the difference in ocean heat content between 1998 and 1999 and then tell me if you think the earth’s energy budget was out of balance for 1999. Then do the same for 2003, 2004, 2005, 2006, 2007 and 2008. Then tell me, on the basis of this analysis, how many of those years were out of balance?

          The type of trend analysis you are doing is similar to what Judith Curry did in her presentation on the relationship between warmer oceans and stronger hurricanes. She prepared a trend analysis that obscures the relation between warmer oceans and stronger hurricanes. I asked for a graphic that would show the relation on a year by year comparison and David Smith came up with this graphic. As you can see, there is no relationship between warmer oceans and stronger hurricanes. But if you look at a trend analysis, it will appear that a relationship is there.

        • Posted Mar 24, 2009 at 12:28 AM | Permalink

          Re: Ron Cram (#112),

          Scientists should be required to operate under and present their findings under Generally Accepted Scientific Principles – GASP – rules. Such a practice would allow consumers of the research to understand if a physical relationship actually existed and what the relationship meant.

          Science by regulation is no answer. No one is qualified to tell me how to do my work or what statistics I should use, as I am not qualified to tell you how to do yours. Everyone is qualified to review our collective work and reality will determine its accuracy.

          While regulation for the release of data for verification of its reproducibility makes sense, science is not accounting. The outcome is rarely exact. Regulation of science will invariably result in ever greater abuse and is Pandora’s elixir for the politician/polyscientician controlling the rules.

          I understand that you may want limited regulation, I might want it too but I promise it won’t stop where we think it should. That just isn’t how the world works. I vote to leave that box closed…

        • Ron Cram
          Posted Mar 24, 2009 at 4:57 AM | Permalink

          Re: Jeff Id (#115),

          The suggestion was somewhat tongue-in-cheek in that I love the GASP acronym. In accounting, the American Institute of Certified Public Accountants (AICPA) have a Financial Accounting Standards Board who issue standards for the private sector. The AICPA designated the Federal Accounting Standards Advisory Board to issue accounting standards for federal entities. So, I was proposing proposing self-regulation by scientists, not federal regulation of science.

          Also, my proposal had more to do with how research should be presented to research consumers, not how the basic science was done. I am still not certain my point is entirely clear.

          Let’s take the example of a scientist studying the relationship of SST to stronger hurricanes, as Judith Curry has done. For the sake of clarity, let’s make up some data. Let’s say from 1980-2008, we had alternating years of warm and cool SSTs. Let’s say the cool years were all about the same, but the warm years grew warmer by a small but significant amount each warm year. Then let’s say all of the warm years had five cat 4/5 hurricanes, but the cool years started with ten strong hurricanes and incremented by two with each passing cool year. If the scientist gathers all the data and presents it using a trend analysis, you will see an increasing number of stronger hurricanes as the oceans warm. Consumers of the researchers will assume the suspected causal relationship was confirmed. This is exactly what Judith Curry did. When you actually look at the data, matching the stronger hurricanes with the proper SSTs, the causal relationship is thrown into doubt.

          I still think it would be efficient, save time and end a lot of bickering if someone could look at Judith’s presentation and say “Hey, you violated scientific principle 11.2 in that you are doing trend analysis when you should be matching data from the same years.” Does that make sense now?

        • Posted Mar 24, 2009 at 8:50 AM | Permalink

          Re: Ron Cram (#118),

          Sorry Ron, I’ve been a little grumpy about our world governments lately and am a bit sick of the rules. Put’s me in a bad mood.

        • Ron Cram
          Posted Mar 24, 2009 at 9:37 AM | Permalink

          Re: Jeff Id (#119),

          No problem, Jeff. I am right there with you when it comes to governments outgrowing their own britches.

        • Alexander Harvey
          Posted Mar 25, 2009 at 8:04 AM | Permalink

          Re: Ron Cram (#112),

          There seems to be little in common with how we define or think about the “pipeline”.

          You wrote:

          Heat can be said to be “in the pipeline” when the oceans are warmer than usual. We would expect this heat to be released into the atmosphere eventually. But radiative imbalance cannot be “in the pipeline” and show up in a warmer ocean years later.

          It is not “heat” but the potential for further heating of the oceans that is in the pipeline. The effect occurs when the oceans are “cooler” than one would expect them to be at equilibrium. There is no net heat that will be released into the atmosphere eventually. There is an imbalance that would tend to zero at equilibrium which does not mean that flux reversals do not occur but they are not the essence of the pipeline effect. The expected temperature rise in the pipeline is the simple product of the imbalance (W/m^2) and the climate sensitivity (C/(W/m^2)).

          I think that this must be at odds with your point of view.

          The Hansen paper:
          Earth’s Energy Imbalance: Confirmation and Implications
          Basked in the oceanic warming that occured between 1993 and 2003. Interestingly he used a CS of 0.67 (2.47C/doubling was quite a low figure) should he try this again using data upto 2008 (I can’t as I do not have GISS forcing data after 2003) it is likely that he would have to look at a much lower figure for the CS as the average oceanic flux has decreased markedly.

          I agree that the average imbalance in a year has to be accounted for in that year but so must the average imbalance in a decade be accounted for in that decade. That was his approach and mine.

          The point is that the OHC is not rising as quickly as the models would seem to imply and any values for the CS above about 0.68 (2.5C/doubling) seem to be unlikely. Also his temperature increase in the pipeline might have to be reduced from 0.6 to ~0.3C.

          Alex

        • Ron Cram
          Posted Mar 25, 2009 at 2:49 PM | Permalink

          Re: Alexander Harvey (#127),

          I agree that our points of view seem to have little in common. I see Andrew has already directed you to Roger Pielke’s post which explains my point of view. Do you have a link that explains yours in more detail?

          Regarding the focus on yearly results: The oceans are the largest energy storage unit in the world. As such, the oceans are the best subject to study for understanding if a radiative imbalance is happening. My point in focusing on the yearly changes and therefore the yearly imbalances, is because those years when the oceans are not heating makes it clear no radiative imbalance was present during that year. The way AGW theory is currently explained, a radiative imbalance is present each and every year. If no radiative imbalance is present in a particular year, then it is necessary to modify the theory to allow for such an observation and that means a greater role for natural climate variation.

        • Posted Mar 26, 2009 at 5:09 AM | Permalink

          Re: Alexander Harvey (#127),

          It is not “heat” but the potential for further heating of the oceans that is in the pipeline.

          That’s has no sense to me, neither your following reasoning.

          If the incoming radiation is greater than the outgoing radiation, the climate system has to increase immediately its heat content, if it is lower then the climate system has immediately to cool.
          There is no room for “potential heating”!
          So, as far as the climate system is concerned (Re: bender (#135)), one can equate heat content change and radiative imbalance.

        • Alexander Harvey
          Posted Mar 26, 2009 at 8:58 AM | Permalink

          Re: Paolo M. (#140),

          Paulo, to clarify:

          I said nothing about “potential heating” I would not know what that would mean either.

          During a period of an increasing trend in forcing (e.g increasing CO2) the OHC will tend to be lower than its equilibrium value. So when the forcing becomes constant (constant level of CO2) the OHC will be below its equilibrium value so there is a “potential for further heating of the oceans” until equilibrium is reached. While the OHC moves towards equilibrium the downward flux from the surface would tend to reduce the upwards flux from the surface to space thus tending to supress surface temperatures.

          I do not think I disagree with anything else you wrote.

          The problem is one of thermal lag. Yes, energy is conserved at every moment, but the process of increasing OHC by warming the surface causes the oceans to move away from thermal equilibrium.

          For example: to achieve thermal equilibrium, (after the surface heating process has ceased and the surface temperature is held constant), will require further increases in OHC. The imbalance will still be present after the surface temperature is held constant, but decay away until equilibrium is achieved.

          Interesting, as far as the oceans are concerned equilibrium does not mean constant temperature throughout. The temperature gradient (hot at surface, cold in the abyss) is the equilibrium condition. It is believed (it is in the text books) that the tendency for the bulk of the oceans to rise (about 1.3 cm/day), and the tendency for the heat to diffuse downwards, produces an equilibium with a temperature gradient.
          Now raising the surface temperature increases the magnitude of the gradient near the surface, if one then holds the surface temperature constant, the gradient will move towards a new equilibrium condition, this requires additional heat.

          I hope I have made myself more clear.

          Alex

        • Craig Loehle
          Posted Mar 26, 2009 at 10:00 AM | Permalink

          Re: Alexander Harvey (#142), I agree with your explanation. If there is radiative imbalance, as Hansen claims, then the earth system must on net be absorbing energy (heat) but the ocean could NOT be cooling at the same time unless in some bizarre way the cold sinking water was warmer than it used to be and we aren’t measuring deep water heat content. The magnitude of the imbalance would seem to me to preclude this possibility.

        • Ron Cram
          Posted Mar 26, 2009 at 10:40 AM | Permalink

          Re: Craig Loehle (#144),

          Wow, Craig. This is getting weirder and weirder. You say you agree with Alexander but then seem to disagree with him. I think maybe we need to use more precise terms for clarity of thought.

          Surface sea temps are measured in degrees and are used in the calculation of ocean heat content. Ocean heat content is measured in joules and is calculated by using surface sea temps, deeper water temps down to 700m from the Argo network and satellite sea level measurements to calculate thermal expansion.

          If there is a radiative imbalance, the earth system must be getting warmer which means ocean heat content is going up, and we have warmer sea surface temps, warmer deeper ocean temps and thermal expansion causing a slight but measurable increase in sea level. There is no way such a condition can be considered “cooling.”

          Alexander’s explanation:

          During a period of an increasing trend in forcing (e.g increasing CO2) the OHC will tend to be lower than its equilibrium value. So when the forcing becomes constant (constant level of CO2) the OHC will be below its equilibrium value so there is a “potential for further heating of the oceans” until equilibrium is reached. While the OHC moves towards equilibrium the downward flux from the surface would tend to reduce the upwards flux from the surface to space thus tending to supress surface temperatures.

          While this explanation is logically sound, it is in complete disagreement with the warming earth system we observed in the 1990s. This is just not how it works.

          The theory of AGW says increased atmospheric CO2 will create a radiative imbalance because less heat is emitted to space. A radiative imbalance will trap heat in the oceans causing the oceans to warm as measured in all categories (except possibly the lowest of oceans depths which are rarely sampled). The heat trapped in the oceans will be released into the atmosphere at the rate of about 40% in 5 years and the other 60% over a period of decades. (I am sorry I cannot remember the exact rates of release but I think I am roughly correct.)

          But the main point here is one cannot say a radiative imbalance exists if ocean heat content is not going up on an annual basis. Do you agree or disagree?

        • Craig Loehle
          Posted Mar 26, 2009 at 10:48 AM | Permalink

          Re: Ron Cram (#145), one cannot say a radiative imbalance exists if ocean heat content is not going up on an annual basis–but the Hansen paper does not measure radiative imbalance from ocean temperatures, they measure it at the top of the atmosphere. The failure of the ocean to warm in the presence of what they calculated means they made a mistake. That is, the radiative imbalance is either erroneously calculated or it is presumed by theory rather than from showing that heat is being accumulated. If this is not what Alexander meant then I need to change my coffee brands…or maybe not try to do multiple things at once…

        • Andrew
          Posted Mar 26, 2009 at 11:28 AM | Permalink

          Re: Craig Loehle (#146), My understanding was that Hansen et al. were estimating the TOA imbalance from the OHC data, not measuring it. AFAIK there were no measurements of TOA imbalance until very recently (and even now maybe?).

        • Craig Loehle
          Posted Mar 26, 2009 at 11:37 AM | Permalink

          Re: Andrew (#147), The Hansen paper:
          http://pubs.giss.nasa.gov/docs/2005/2005_Hansen_etal_1.pdf which is a 2005 pub, calculates the imbalance from his models, then uses increasing ocean heat content from the 10 years ending in early 2003 to support the claims. AFTER their data, the OHC turn downward. From 2003 to present, the data do NOT support his model.

        • Ron Cram
          Posted Mar 26, 2009 at 11:54 AM | Permalink

          Re: Craig Loehle (#146),

          Okay, we are on the same page then. I was beginning to wonder about myself.

        • Posted Mar 26, 2009 at 11:36 AM | Permalink

          Re: Alexander Harvey (#142),
          sorry, but I continue to not understand.

          During a period of an increasing trend in forcing (e.g increasing CO2) the OHC will tend to be lower than its equilibrium value.

          No, if there is a increase in forcing, the heat content of the climate system will rise immediately.

          For example: to achieve thermal equilibrium, (after the surface heating process has ceased and the surface temperature is held constant), will require further increases in OHC.

          Sorry?

          The energy source is only one: the Sun.
          The Sun immediately makes the climate system heat content increase every day with pulses of energy. If the system becomes less prone to let heat escape to the outer space, that heat is already inside the system. What else do you need?

        • Craig Loehle
          Posted Mar 26, 2009 at 12:12 PM | Permalink

          Re: Paolo M. (#148), If you turn on the burner under a pot of water, the system is out of equilibrium. The new heat is immediately in the system but the increased forcing is NOT immediately reflected in the water temperature, which takes time to reach a level at which losses due to steam escaping and heat loss by radiative means equals the inputs at the bottom of the pot.

        • Posted Mar 26, 2009 at 2:16 PM | Permalink

          Re: Craig Loehle (#151),
          the discussion here is whether there is heat in the pipeline.
          Nobody here disputes that if you provide energy to the system, the system would or would not get warmer.

          In your analogy of a pot of water, the increased forcing is not equivalent to the joules produced by the burner, but by the joules that go through the wall of the pot base. Well, be sure that those joules immediately reflects in the water heat content, i.e. water temperature.
          If you are good enough, you can transform all that water in vapour at once!
          Of course that water emits more as it gets warmer, but, as I said, it’s another question.

        • Posted Mar 26, 2009 at 2:37 PM | Permalink

          Re: Paolo M. (#152),
          Craig, if your burner goes on to burn, well that doesn’t mean that you have heat in the pipeline.
          It means that yuo have a continuing pulse of energy, period.

          If CO2 continues to increase, that could mean that a greater amount of ocean heat (sun energy, the never ending suppler) will not go to the outer space. Where the heat in the pipeline is?

        • Craig Loehle
          Posted Mar 26, 2009 at 4:02 PM | Permalink

          Re: Paolo M. (#153), It seems to me that “in the pipeline” has been used in two ways. It is used to mean that a given level of elevated (CO2) forcing will cause future warming, because the system is not yet in equilibrium. Second, it has been used informally to wave away the recent atmospheric cooling by asserting (without data) that the extra heat has been taken up by the ocean. The first sense is correct but we can now say that (probably) the second is not.

        • Mark T
          Posted Mar 26, 2009 at 4:28 PM | Permalink

          Re: Craig Loehle (#156), Unfortunately, for alarmists, the “pipeline” for elevated CO2 “forcing” (sigh) is rather short.

          Mark

        • Posted Mar 26, 2009 at 5:19 PM | Permalink

          Re: Craig Loehle (#156),
          whatever “in the pipeline” has been used, it was a wrong way.
          As I tried to say, a given level of forcing (CO2 or whatever) causes a warming now, never in the future. You cannot preserve a pulse of energy (the Sun) for the future. You have to transform it into heat content of the climate system..immediately…or lose it into the outer space. Tertium non datur.

          The ocean is “the climate system”! At a first approximation.
          We hadn’t to wait for this recent cooling to say that the statement “heat has been taken by the ocean” is a climatic nonsense.

        • Craig Loehle
          Posted Mar 26, 2009 at 5:42 PM | Permalink

          Re: Paolo M. (#158), Sorry Paolo, but not quite. It is true that the HEAT (energy) from a step-increase in forcing must be realized now but the temperature could continue to rise for a long time because the oceans must warm up to the new equilibrium. Detecting the non-equilibrium change in heat over short intervals in the noisy climate system if very difficult: the signal is less than the noise.

        • Raven
          Posted Mar 26, 2009 at 7:02 PM | Permalink

          Re: Paolo M. (#158)
          Think about a simple capacitor-resistor circuit. A step change in voltage will not immediately increase the volatage across the capacitor because it takes time to build up the charge stored in the capacitor. IOW, the system is not preserving the pulse – it is simply limiting the rate at which the energy can flow into the system.

          The oceans work like the capacitor because they also limit the rate at which energy can flow into the system. However, one can still diagnose a non-equilibrium situation by looking at the rate of energy flow into the system. If the rate of energy flow is close to 0 then there cannot be an imbalance.

          The only caveat is a noisy system could produce periods where there rate of energy flow drops to 0 even when there is an imbalance.

        • Posted Mar 27, 2009 at 4:21 AM | Permalink

          Re: Raven (#160),

          The oceans…limit the rate at which energy can flow into the system

          What?
          The ocean is the system!!!
          The ocean limits nothing! All the energy that gets into the ocean, it will be immediately inside the ocean. What flow has the ocean to limit?

          Re: Craig Loehle (#159), Re: Raven (#160),

          How difficult is to use the appropriate terminology?

          The climate system has no heat stored somewhere that it will realize sometime in the future. No pipeline there!
          I think we all agree.

          What you are saying is that the system is not in equilibrium with an eventual pulse of energy that will arrive tomorrow.
          How could it be otherwise? How could the system account for somenthing that has still to happen?
          Of course that doesn’t mean that there is heat in the pipeline. It’s a wrong and deprecable use of terminology.
          If some big personality of the climate science uses this wrong expression, that doesn’t mean that we have to be addicted to this use.
          The attempt to forecast the state of the system, if in the future a greater amount of Sun energy is not lost in the outer space, has nothing to do with whatever heat in the pipeline.

        • Craig Loehle
          Posted Mar 27, 2009 at 6:34 AM | Permalink

          Re: Paolo M. (#165), The ocean limits the rate at which the earth system warms because it is NOT well-mixed. Heat applied to the surface may take hundreds of years to warm the deep ocean. Thus it will tend to absorb heat applied to the system until the system comes to equilibrium. “in the pipeline” is a colloquial expression, not a scientific one. It means that all the warming we expect to see from an increase in forcing has not yet happened. Like the pot of heated water or the capacitor. Yes, it is an awkward expression, but we are here responding to what the big cheeses of climate science are saying.
          Re: Ron Cram (#162), I believe this excuse was from Gavin or Hansen, but I did not print out the reference. It was an attempt to explain away the atmosphere cooling recently by saying the energy imbalance is going into the ocean. Anybody remember where this was?

        • Dave Dardinger
          Posted Mar 27, 2009 at 7:37 AM | Permalink

          Re: Craig Loehle (#167),

          Heat applied to the surface may take hundreds of years to warm the deep ocean.

          This is where I have a problem. Statements like this imply that there’s a real case for heat working its way down to the depths. I’d suggest there’s little of this. There are pools of warm water on the surface consisting of the mixed layers. Then there are the areas at the poles where surface water is cooled and either evaporation or ice formation happens to the point where it’s denser than the deeper water and a body of water sinks to the deep ocean. Now there will be some minor differences in the composition / temperature of this water, but on the whole it’ll be pretty consistant. The sinking of this water will have to be offset by rising cold water elsewhere around the world. But the point is that the cold dense water at depth will not be replaced with heated water as long as there are polar areas which can produce such cold dense water.

          There may some intermediate waters which can pick up heat from the mixed layer, but they’ll only do so extremely slowly and won’t make much difference. But it seems to me that the mixed layer is prone to going through periods of warming and periods of cooling (PDO etc) and doesn’t just heat forever. I think if the oceans are properly modeled they will indeed look something like a capacitor which periodically is shorted out and discharges some of it’s energy. This process may look like atmospheric thunderstorms with a very long time period.

        • Posted Mar 27, 2009 at 11:25 AM | Permalink

          Re: Dave Dardinger (#172),

          a possible mechanism for mixing is friction among different density layers of water triggered by tides. Of course, mixing is enhanced in proximity to the continental escarpment or islands.
          It seems scientists are not sure about the magnitude of this phenomenon.

          Re: Craig Loehle (#167),
          sorry, I think I was too harsh in #176.

        • Ron Cram
          Posted Mar 27, 2009 at 7:40 AM | Permalink

          Re: Craig Loehle (#167),

          If that is what they said, they should be called on it. The oceans were cooling first.

        • Posted Mar 27, 2009 at 10:04 AM | Permalink

          Re: Craig Loehle (#167),

          The ocean limits the rate at which the earth system warms because it is NOT well-mixed.

          Craig, what you write is a nonsense and your following reasoning doesn’t improve your position.

          You have first to answer to the question: what is the climate system?

          What is the sense of this statement (more or less) of you: “the upper ocean warms but the system does not”?

          What does the distribution of heat inside the system have to do with your claim that the ocean limits the rate of warming of the system?

          No offense is intended, but you are so atmosphere-centric, so addicted to wrong terminology that you then can write statement as that I quoted above.

        • Craig Loehle
          Posted Mar 27, 2009 at 12:41 PM | Permalink

          Re: Paolo M. (#176), I should have said that the rate of mixing of warm surface water downward limits the rate of warming of the “atmosphere”–I am not atmosphere-centric, but that is what everyone is basing this whole debate on and we don’t even have measurements of the deep ocean.

        • Ron Cram
          Posted Mar 27, 2009 at 7:36 AM | Permalink

          Re: Paolo M. (#165),

          Because most people look to the surface temp record as the metric for monitoring global warming and so the atmosphere is the only aspect being observed. The heat stored in the ocean is “in the pipeline” because it is not yet in the atmosphere. I know. It is a silly and simplistic way to look at it, but it was their way of explaining why the global surface record was not breaking records year after year.

        • Ron Cram
          Posted Mar 26, 2009 at 11:10 PM | Permalink

          Re: Craig Loehle (#156),

          I’ve never seen “in the pipeline” used in the second sense (that the atmosphere is cooling because the oceans are warming). Besides being nonsensical, that view becomes more ludicrous since the oceans stopped warming before the atmosphere did.

        • Andrew
          Posted Mar 27, 2009 at 6:25 AM | Permalink

          Re: Ron Cram (#162), I’m pretty sure you meant that the other way around.

        • Ron Cram
          Posted Mar 27, 2009 at 7:30 AM | Permalink

          Re: Andrew (#166),

          The oceans stopped warming in 2003. According to GISTEMP, the warmest surface temps were in 2005. Of course, if you use HadCRUT, it says 1998 was the warmest. But then 1998 and 2005 were very, very close even in HadCRUT and 2005 was not as strong an El Nino year so the alarmists look at 2005 as the year with the greatest AGW warming. I guess that was my point.

        • Andrew
          Posted Mar 27, 2009 at 7:37 AM | Permalink

          Re: Ron Cram (#169), Given the serious problems with the sruface data, I prefer not to go by either, especially since neither represents “the atmosphere”. If we go by satellites, the date is more like 2001 (1998 to, but, well, that’s basically because of the El Nino). GISTEMP is especially iffy, I think.

        • Ron Cram
          Posted Mar 27, 2009 at 7:51 AM | Permalink

          Re: Andrew (#171),

          I agree the surface temp record is pretty close to worthless. I believe the satellite record also shows 1998 the warmest by far. At least the graphic on Wikipedia looks that way. I am just saying the alarmists who might make the claim Craig is talking about will view 2005 as the warmest on record. So they are making a claim contrary to their view. It does not make sense.

        • Andrew
          Posted Mar 27, 2009 at 9:55 AM | Permalink

          Re: Ron Cram (#174), Good point! Hadn’t thought of it that way.

        • Alexander Harvey
          Posted Mar 31, 2009 at 6:27 AM | Permalink

          Re: Paolo M. (#148),

          Re: Alexander Harvey (#142),
          sorry, but I continue to not understand.

          I wrote:

          During a period of an increasing trend in forcing (e.g increasing CO2) the OHC will tend to be lower than its equilibrium value.

          You replied:

          No, if there is a increase in forcing, the heat content of the climate system will rise immediately.

          Yes, but not to its equilibrium level, that is the point. There is no contradiction.

          I suggest that you read: Climate Response Times: Dependence on Climate Sensitivity and Ocean Mixing (Hansen et al, Science 229,1985)

          It is worth a read and it does contain an early reference to “unrealized warming” with a diagram to illustrate what was meant (Fig 2).

          Alex

        • Posted Mar 31, 2009 at 11:33 AM | Permalink

          Re: Alexander Harvey (#209),
          you are a bit late.
          After my comment 148, there was a long excange:

          Craig Loehle 151
          Paolo M 152
          PM 153
          CL 156
          PM 158
          CL 159
          Raven 160
          PM 165
          CL 167
          Ron Cram 170
          PM 176
          CL 181

          I suggest you read all those comments. What you write is still incorrect.

        • Alexander Harvey
          Posted Mar 31, 2009 at 1:53 PM | Permalink

          Re: Paolo M. (#217),

          Have you had time to read the paper I mentioned? The term “unrealized warming” has been around a long time and the Hansen 1985 paper gives a good overview.

          Do I like the term? No, I feel it to be misleading but it is the term that has come done to us.

          I suggest you read that paper if you have not. I am not sure that either of us is wrong, it is a matter of definitions. If you read that paper I expect you will find that I am correct in the context of that paper and the others that I have mentioned.

          Alex

          Alex

        • Posted Mar 31, 2009 at 3:45 PM | Permalink

          Re: Alexander Harvey (#221),
          as I said in comment 165:

          If some big personality of the climate science uses this wrong expression, that doesn’t mean that we have to be addicted to this use.

          Roger Pielke has been complaining about this misleding use of the concept by the head of GISS for 5 years now.
          So, don’t mind if I make you notice now and then.

        • Alexander Harvey
          Posted Mar 31, 2009 at 4:20 PM | Permalink

          Re: Paolo M. (#225),

          The term “unrealized warming” is in a 1985 Hansen paper and I think that I am using the term in the way that it was accepted for the next ~20 years. If Pielke wants the effect go away and chooses to aid his cause by redefining the meaning of the term that is his affair.

          Have you read the 1985 paper, I think it makes things clear as to the nature of the effect?

          Can you comment on how Pielke infers what Urban & Keller meant but failed to say?

          Alex

        • Andrew
          Posted Mar 31, 2009 at 6:16 PM | Permalink

          Re: Alexander Harvey (#226), The problem is that given a choice between Hansen and Pielke Senior, most of us would choose the one with the best record of being correct and least biased. And that ain’t Hansen.

        • Alexander Harvey
          Posted Mar 31, 2009 at 7:00 PM | Permalink

          Re: Andrew (#227),

          You are probably correct but why change the nature of the debate just when Hansen’s definitions and analysis seems to be arguing against extreme global warming.

          The Urban & Keller paper combined with the emerging OHC record seems to mitigate against the conclusions for 0.6C of “commited warming” and more importantly against all but the lower end of the IPCC CO2 doubling temperature rise.

          I can not see why one should abandon a battleground that is fit for purpose just when the tide appears to be turning. To do so seems to me to encourage dismissal of valid arguments by the mainstream community.

          Alex

        • Andrew
          Posted Mar 31, 2009 at 7:47 PM | Permalink

          Re: Alexander Harvey (#228), It shouldn’t matter-either Hansen’s definition is right or it isn’t. I’m open minded about it but I won’t accept it as correct just because it supports what I tend to believe on climate sensitivity etc. That’s a bad argument-its not scientific at all.

        • Alexander Harvey
          Posted Mar 31, 2009 at 8:57 PM | Permalink

          Re: Andrew (#230),

          I have no problem with either of the Hansen papers I have mentioned, (1985 & 2005). I consider them to contain valid arguments. I think that he was fortunate (2005) in that the 1993-2003 period was atypical of the 1955-2008 period so it argued for a higher CS than would be the case if it was republished now. I can not see what is un-scientific about a valid argument for higher values of CS, becoming a valid argument for lower CS when the emerging data dictates it should be so.

          When I read the 2005 paper I agreed with his logic and accepted that his conclusions followed. I still agree but given more data the conclusions that follow are somewhat different in scale.

          BTW Can you, or anyone else, tell me what in the Urban & Keller paper led Pielke to deduce what they meant by “unrealized Warming” as I cannot see anything like his interpretation in their paper. Perhaps there is nothing. Perhaps they inferred it in another paper.

          As far as I know Hansen may have coined “unrealized warming”, his 1985 reference is certainly an early one. But whatever it is called the arguments in the 1985 paper a pretty solid and I am yet to hear anyone here contest them. They are text book stuff regarding thermal processes.

          One cannot wish away the effect he illustrates any more than one can expect local noon to be the hotest time of the day or expect the summer solstice to be the hotest day of the year. These types of delay of temperature following forcing are a feature whenever thermal mass is invovled.

          Alex

        • Ron Cram
          Posted Mar 31, 2009 at 9:27 PM | Permalink

          Re: Alexander Harvey (#234),

          I do not see where Hansen claims a radiative imbalance causes the oceans to cool, or however you phrased it. In
          Alexander Harvey (#127), you wrote:

          It is not, “heat” but the potential for further heating of the oceans that is in the pipeline. The effect occurs when the oceans are “cooler” than one would expect them to be at equilibrium.

          In his 2005 paper, he writes:

          Our climate model, driven mainly by increasing human-made greenhouse gases and aerosols, among other forcings, calculates that Earth is now absorbing 0.85 ± 0.15 watts per square meter more energy from the Sun than it is emitting to space. This imbalance is confirmed by precise measurements of increasing ocean heat content over the past 10 years.

          These measurements of OHC show a warming ocean, not a cooling ocean nor one that is “cooler than one would expect them to be.” You misunderstand Hansen and blame Pielke for misunderstanding. Believe me, Pielke understands. He is an ISI highly cited researcher who is carefully read by Hansen and most everyone in climate science. If Pielke had made a mistake, he would have been notified and corrected it by now.

        • bender
          Posted Mar 31, 2009 at 9:41 PM | Permalink

          Re: Ron Cram (#235),
          You are still – despite your quotations – making the mistake that I corrected you on earlier, in #130. Heat not gained by the oceans could have been lost to space. Ergo “imbalance” (=temporary departure from equilibrium). Lack of OHC rise – especially in surface waters (recalling the depths are unsampled) – is virtually uninterpretible.

        • Ron Cram
          Posted Mar 31, 2009 at 9:49 PM | Permalink

          Re: bender (#240), Re: bender (#241),

          Thank you for the crosspost notice. We probably have another coming up.

          I did not notice your comment about heat lost to space until much later but it seems nonsensical to me. The “imbalance” has to do with the difference between incoming and outgoing heat. If incoming heat exceeds outgoing heat, the imbalance causes the climate system to warm. The warming will be most noticeable in the oceans. If outgoing heat exceeds incoming heat, the climate system cools and will be most noticeable in cooler oceans.

          Contrary to your statement, oceans depths have been routinely sampled to 700m depth since 2000 (esp. 2003 when ARGO achieved global coverage). Deeper samplings do occur but are more rare. These rare samplings have shown no ocean warming at the lower depths.

          None of this has anything to do with Hansen’s clear words… a post I assume you are responding to now.

        • bender
          Posted Mar 31, 2009 at 9:52 PM | Permalink

          Re: Ron Cram (#242),

          oceans depths have been routinely sampled to 700m depth since 2000 (esp. 2003 when ARGO achieved global coverage). Deeper samplings do occur but are more rare.

          This is not the opposite of what I said. It’s a paraphrase of what I said.

        • Ron Cram
          Posted Mar 31, 2009 at 10:25 PM | Permalink

          Re: bender (#244), Re: Bill Illis (#246),

          Hansen explains his concept

          e. Planetary energy imbalance. The planetary energy imbalance, which had been
          predicted in earlier papers [Reference 5], can be measured accurately in the indirect fashion that
          we employed for our present paper [Reference 6], i.e., via measurements of internal ocean
          temperature supplemented by precise satellite altimetry of the ocean surface. It is desirable that
          the internal ocean measurements be extended to greater depths of the ocean. It also will be
          necessary to account for the heat used to melt ice, if ice sheet disintegration and sea level rise
          accelerate, but this term can be obtained accurately from the measurements described below
          (items h and i). Planetary radiation imbalance can in principle be measured by satellite, but in
          practice the needed absolute accuracy is not yet attainable. This is because the measurements
          must include the solar radiation reflected by the Earth in all directions, as well as the thermal
          radiation emitted in all directions, the measurements must be made to an absolute accuracy of
          ~0.1 W/m2, and there must be no temporal gaps in the measurements. Ocean heat storage is an ideal proxy measurement, because temperature can be measured accurately and the ocean
          naturally integrates the energy imbalance over time. Although satellite radiation budget
          measurements do not meet the accuracy requirements, they are a valuable complement to the
          ocean measurements. The satellite data can be calibrated against the ocean data, and the satellite
          can then provide greater spatial and temporal information that is useful for many climate
          analyses and other purposes.

          I think it is pretty clear Hansen considers OHC as a proxy for a direct measurement of radiative imbalance. While lucia is a very smart woman, her concept does not agrees with Hansen or Pielke. I do not think there is anyway lucia’s concept could allow OHC to be a direct proxy for radiative imbalance.

        • bender
          Posted Apr 1, 2009 at 2:03 AM | Permalink

          Re: Ron Cram (#250),
          It appears you are playing word games trying to belittle Hansen, whereas I am trying to explain what you say is “ridiculous”, and “nonsensical”. Perhaps it is you that is ridiculous and not able to make sense of things? There is a difference between an accounting “imbalance” and a dynamic “imbalance”. You appear not to understand this difference. If oceans were warming then there would be an accounting imbalance that Hansen says in that excerpt would prove GHG theory. In OTHER exceprts, which you are now ignoring, Hansen makes reference to heat lost to space. This is the sort of thing that could lead to a dynamic imbalance – the turkey that awaits inevitable cooking. It’s easy to make anyone appear horribly wrong, including Hansen, if you insist on turning his arguments into a set of unrelated strawmen.

        • Ron Cram
          Posted Apr 1, 2009 at 7:26 AM | Permalink

          Re: bender (#258),

          Give me a quote from Hansen. I have been asking for one for some time now and no one has produced any quote for anything like what Hansen is claimed to have said. I have provided quotes for my understanding of Hansen’s argument and for Pielke’s understanding.

        • Alexander Harvey
          Posted Apr 1, 2009 at 5:59 AM | Permalink

          Re: Ron Cram (#235),

          I do not see where Hansen claims a radiative imbalance causes the oceans to cool, or however you phrased it.

          I doubt that he did, I certainly didn’t.

          I wrote:

          The effect occurs when the oceans are “cooler” than one would expect them to be at equilibrium.

          In the 1985 paper Hansen wrote:

          The results (Fig. 2) show that a large part of the equilibrium CO2 warming is not yet realized, … If climate sensitivity is 3C or greater for doubled CO2, most of the the expected warming due to the CO2 increase since 1850 has probably not yet occurred;

          (Please go to the paper to see figure 2 and the rest of the description.)

          I think what I did write was consistent with Hansen. I think it is clear that “unrealized warming” is “warming not yet realized”, as in “has not happened yet”. I can not read this or any other part of this and more importantly the Urban & Keller paper as implying:

          By “unrealized warming in the pipeline”, they mean heat that is being stored within the ocean, which can subsequently be released into the ocean atmosphere.

          From Hansen we have that the heating has not yet occurred (yet to be realized). Given the Hansen interpretation, Pielke appears to say that Urban & Keller mean that “heat that has not yet entered the ocean is being stored in the ocean” I am sure that is not what they meant, so he cannot be using the same interpretation as Hansen but he is using the same language which is I feel unfortunate.

          Urban & Keller write in their paper:

          Observations of changes in surface air temperatures result
          in positively correlated estimates of climate sensitivity
          and ocean diffusivity (Fig. 2, panel A). This is because the
          climate sensitivity signal in the observed warming is confounded
          by the amount of warming “in the pipeline” due to
          the lagged response of the oceans [Hansen et al., 1985, 1997].

          They are, I believe, using the Hansen interpretation. It is all about lagged responses.

          Alex

        • DaveR
          Posted Apr 1, 2009 at 6:48 AM | Permalink

          Re: Alexander Harvey (#264)

          I think it is clear that “unrealized warming” is “warming not yet realized”, as in “has not happened yet”.

          The phrase is equally as applicable to an instance where warming has occurred, but it hasn’t been recorded, so people don’t realize it’s occurred.

        • Alexander Harvey
          Posted Apr 1, 2009 at 7:26 AM | Permalink

          Re: DaveR (#267),

          If you think that your alternative interpretation accords with the process described in the Hansen 1985 paper, I can only suggest that you read the paper and see if you still do.

          Alex

        • Ron Cram
          Posted Apr 1, 2009 at 7:29 AM | Permalink

          Re: Alexander Harvey (#264),

          Okay, finally you have provided a quote from Hansen’s 1985 paper where he wrote:

          The results (Fig. 2) show that a large part of the equilibrium CO2 warming is not yet realized, … If climate sensitivity is 3C or greater for doubled CO2, most of the the expected warming due to the CO2 increase since 1850 has probably not yet occurred;

          Let’s understand exactly what Hansen is doing here. He is trying to explain why GISTEMP has not risen as quickly as he predicted. He is thinking about the atmosphere and near surface temps. So where is the heat going if not into the atmosphere? Into the oceans.

          This quote supports my understanding of Hansen’s view. Do you have another?

        • Alexander Harvey
          Posted Apr 1, 2009 at 7:52 AM | Permalink

          Re: Ron Cram (#270),

          Let’s understand exactly what Hansen is doing here. He is trying to explain why GISTEMP has not risen as quickly as he predicted. He is thinking about the atmosphere and near surface temps. So where is the heat going if not into the atmosphere? Into the oceans.

          I think you are guilty of doing to Hansen what Pielke seems to be doing to Urban & Keller: putting words into over peoples mouths. Your conclusion does not follow from the 1985 paper. Please read the paper and bear in mind when it was written. If you are referring to his 1988 prediction given before Congress you should notice that the paper in question was published 3 years earlier. He may be a clever man but that would be a miracle. If not that prediction, which one prior to 1985 are you referring to. As it stands it is not at all clear how he could be explaining something that would not be an issue for many years to come. Perhaps you simply failed to notice the data of the quote (I did give it) and jumped to a conclusion.

          Alex

        • Ron Cram
          Posted Apr 1, 2009 at 8:13 AM | Permalink

          Re: Alexander Harvey (#271),

          I’m sorry I used an imprecise term. I did not have his testimony before Congress in mind. What I had in mind was his “prediction” in the loose meaning of the term (I should have used “public or common expectation”) based on the 3C or greater climate sensitivity on doubled CO2.

          Believe me, whenever you see Hansen talk about “cooler than expected” or “warming that has not been realized” he is referring to the atmosphere. He is not referring to the ocean.

          I have explained how the Hansen quote you provided fits my understanding. I have provide for you a number of Hansen quotes in which it is quite clear he identifies the radiative imbalance with an increase in ocean heat content and uses OHC as a proxy for radiative imbalance. You have not addressed how I might have misunderstood these quotes or how these quotes fit your understanding.

        • Alexander Harvey
          Posted Apr 1, 2009 at 11:27 AM | Permalink

          Re: Ron Cram (#272),

          I am not sure that we disagree about much or anything unless you hold the Pielke inference from the Urban & Keller paper to be valid.

          I do not simply accept but indorse the concept that the first derivative of the OHC by time represents the majority of the radiative imbalance.

          To be certain I will bore you all by being a bit long-winded.

          I beg your pardons for using unconventional notation but here goes.

          RI = FI – FO (states a radiative imbalance due to a difference between the inbound and outbound fluxes at the TOA.

          This can be equated to the OHC (ignoring all other heat sinks which are considered to be minor).

          dQ/dt = RI (Q being the OHC).

          giving dQ/dt = FI – FO or FO = FI – dQ/dt

          This holds at each and every instant.

          Now dQ/dt is not a constant. According to the data it ranges widely between both +ve and -ve values year to year and probably moment to moment.

          Interestingly its yearly average varies by more as +/- 2 W/m^2 when averaged over the Earth’s surface. So we have forces at work at least as big as the effects of GHGs and these forces seem to be poorly understood.

          Now FO is commonly held to be a function of both temp (T) and GHG concentrations (GH) and albedo (A) amongst other things some of which I would say must vary with time.

          So lets say

          FO(T(t),GH(t),A(T,t),UN(t)) were UN is a placeholder for unknown processes, albedo is shown varying with temp and with time which is a reasonable assumption.

          So dQ/dt = FI(t) – FO(T(t),GH(t),A(T,t),UN(t)) assuming a FI that can vary in time.

          Let us see what happens if we vary GH, Please assume partial differentiation where appropriate in all that follows:

          (dFO/dGH)*deltaGH gives us the “forcing” due to a change of deltaGH.

          dFO/dT gives us (1/CS) where CS is the climate sensitivity

          finally (dFO/dUN)*deltaUN gives us an unknown forcing at least as big as the forcings due to deltaGH (since say 1880) and highly variable on periods as long as a year. (I will come to A, the albedo later, for now assume that it is bundled into UN.

          Now if we seek to find CS we evaluate dFO/dT.

          Now we do not seem to know FO sufficiently well to do this directly but we can make assumptions.

          We have a fair handle on FI (TSI) and we have some data from which we can derive values for dQ/dT, so we can estimate FI from the equation above:

          FO = FI -dQ/dt

          Now we cannot perform the partial derivitive dFO/dT as we do not know the function FO in any detail.

          But we can regrees FO against T to get an estimate. Now the problem is the magnitude of dFO/dUN but one could be tempted to assume that given enough years of data we can estimate dFO/dT and hence CS and use a statistical technique to give us some error bars.

          Finding CS is more or less the Holy Grael of climate science and it is not going to be given up easily. I have given it a go and low values for CS are favoured but the uncertainties are great.

          In Hansen 2005 he only used the 1993-2005 data and he found it to be consistent with CS = 0.67W/m^2. Now as we know, since 2004 the OHC has not changed much so dQ/dt = ~0. Also prior to 1993 the dQ/dt was highly variable but when aaveraged lower than its peak years 2002/2003 so a re-evaluation along the lines of his 2005 paper or the Urban & Keller paper would I think give us a lower value for the CS.

          I must emphasise again just how big the variation in dFO/dUN is. Currently this “forcing” is -ve (or balancing all of the other forcings after the increase in FO due to the increase in T is subtracted) back in 2002/2003 it was big and +ve, indeed greater than all the other forcings. The proviso being that the OHC record is robust enough to stand the prouction of interannual differences to give us our oceanic surface yearly averaged flux values.

          Now I come back to A(T,t) the albedo, well I expect that this may be a large part of the forcing I bundled into “unknown” but I do not know how much. Certainly A varies from moment to moment (clouds) and from season to season (land and sea ice changes) and from year to year (melting glaciers). But importantly it does not really matter whether we understand the nature of any “unknown” forcings as we can use the OHC or rather its first differential with time to calculate a value. Strictly speaking corrections have to be made for minor matters, heat absorbed through the land surface and taken up by the melting of ice if one wants a more carefully considered estimate for CS.

          End of boring bit.

          FWIW you will notice that I make no reference to pipelines or unrealized anything. They are not good phrases and should probalby have been better left uncoined.

          Please excuse typos etc. I am doing this in a hurry without notes so be tollerant but please point out any errors politely.

          Alex

        • Alexander Harvey
          Posted Apr 1, 2009 at 11:31 AM | Permalink

          Re: Alexander Harvey (#281),

          To get the ball rolling I neglected to state that the FO for each year has to be corrected for the assumed chang in GHG forcing for that year prior to performing a regression against temps.

          Alex

        • Alexander Harvey
          Posted Apr 1, 2009 at 11:48 AM | Permalink

          Re: Alexander Harvey (#281),

          Erratum:

          (dFO/dGH)*deltaGH gives us the “forcing” due to a change of deltaGH.

          should read:

          -(dFO/dGH)*deltaGH gives us the “forcing” due to a change of deltaGH.

          In practice as dFO/dGH is not linear we use logarithmic variant of dGH (dGH)/GH instead but it makes the presentation very messy, so please take it as read.

          I told you I was in a hurry.

          If I may digress I also hate the “pseudo” “forcings” like we are given for the GHGs etc. It makes it look like there is extra energy in the system which is not the case.

          Alex

        • Ron Cram
          Posted Apr 1, 2009 at 10:22 PM | Permalink

          Re: Alexander Harvey (#281),

          Perhaps we are not as far apart as it sometimes seems. You write:

          I do not simply accept but indorse the concept that the first derivative of the OHC by time represents the majority of the radiative imbalance.

          I wholeheartedly agree. The remainder of the radiative imbalance stored in land or melted ice is almost trivial.

          Regarding your notations, you do not identify all of them and so I cannot follow completely. I agree with the portions I understand.

          You write:

          Finding CS is more or less the Holy Grael of climate science and it is not going to be given up easily. I have given it a go and low values for CS are favoured but the uncertainties are great.

          I can certainly agree with this. One of my problems with estimating CS is the answer is always given in changes to surface temperature. This is a terrible metric and leads us to the whole problem of “heat in the pipeline.” If you estimate CS as a change in OHC, the concept of heat in the pipeline goes away because there is no heat in the pipeline for the climate system (only for the atmosphere).

          For AGW theory has to be valid, it has to be able to explain under what circumstances the oceans may not warm for five straight years. It currently fails this test. Perhaps it has to do with the Sun being at a minimum. But this cannot explain all of the years without warming. Perhaps it has to do with other natural climate variables such as the PDO. But again, this cannot explain it all since the PDO did not switch to its cool phase until late 2007. We really have no clue why the oceans did not warm from 2003-2007. Whatever the explanation may turn out to be, natural climate variability has to be much greater than the IPCC estimates.

          Josh Wills from JPL claims the models allow for the oceans not to warm for a few years in a row. I cannot help but wonder which models those are? I should email him and ask.

        • Andrew
          Posted Apr 2, 2009 at 12:15 PM | Permalink

          Re: Ron Cram (#290), But how do you relate OHC to impacts?

        • Ron Cram
          Posted Apr 2, 2009 at 1:24 PM | Permalink

          Re: Andrew (#292),

          Good question, Andrew. Theoretically you could translate increased OHC to higher surface temp numbers if the ocean heat release rates to the atmosphere actually work as advertised. To be honest, I am not sure it works exactly like that. One of the issues that confounds the theory is how fast the oceans cooled after Pinatubo. Blocking the Sun just a little bit seems to have a huge effect so it is hard to see how you can predict climate 100 years into the future. Perhaps will a solar minimum, you might not even need volcanic eruptions to dramatically cool ocean heat.

          But the ocean is still the critical component of the climate system. It is where the excess heat would be stored, so it is the obvious place to look for global warming. Now with ARGO, our observation system for the ocean is far superior to any other observing network.

        • Ron Cram
          Posted Apr 1, 2009 at 8:26 AM | Permalink

          Re: Alexander Harvey (#271),

          By the way, Alexander, it should be obvious that the Hansen quote you provided was referring to the atmosphere. Estimates for climate sensitivity for doubled CO2 are always given in terms of increases to surface air temperature. The answer is never given in terms of increases to OHC (which would be a much better metric).

        • Posted Apr 1, 2009 at 11:33 AM | Permalink

          Re: Alexander Harvey (#271),
          I don’t think Pielke was wrong on Urban and Keller. They used the misleding terms “heat in the pipeline” that can only be interpreted as “water in the tube ready to be used as you wash your hands”.
          Regarding how Hansen intented his “heat in the pipe” or “unrealized heat”, I think you are right, i.e. if today I suddenly have an increase of 4 Wm-2 because of a CO2 doubling, tomorrow I still have an extra forcing of 4-epsilon, so that an extra amount of the 270 Wm-2 incoming Sun energy will not go to the space.
          Anyway, if I understood well the AGW hypothesis, a lack of rising upper OHC is not allowed.

        • Alexander Harvey
          Posted Apr 1, 2009 at 12:06 PM | Permalink

          Re: Paolo M. (#283),

          I think it best that we beg to differ over Pielke,

          Anyway, if I understood well the AGW hypothesis, a lack of rising upper OHC is not allowed.

          Well the AGW hypothesis is either sadly wrong or has been over simplified to an stupid degree.

          Sometimes I think it is simply the reluctance for “experts” to “confuse” the public by making more truthful statements like:
          “In the short term (decadal) the temps could go either way but upwards is the better bet if you like a wager. In the long term GHG increases will tend to increase temps but we do not know by how much. We are confidient that this will make important differences to the human population and the planet in general but we do not know precisely when or where or in what way.”

          I really wished that they had made it much more clear that a decade or more of no warming was a distinct possibility but a low probability back in 1998 when it was hot, but AFAIAA they didn’t. If they had they would not appear to be playing catchup.

          Perhaps they will argue that we are still warming but we simply do not realise it. As pointed out above the phrase “unrealized warming” stands that interpretation see DaveR above.

          Alex

        • bender
          Posted Apr 2, 2009 at 7:28 AM | Permalink

          Re: Alexander Harvey (#286),

          I really wished that they had made it much more clear that a decade or more of no warming was a distinct possibility but a low probability back in 1998 when it was hot, but AFAIAA they didn’t. If they had they would not appear to be playing catchup.

          I could not agree more. I contend that there was a systematic effort at all levels to suppress uncertainty on the trend. They were more than happy to pretend they were addresssing it, with mock confidence envelopes. But they could only admit so much uncertainty, no more. Like the MWP that needed to be disappeared, they needed those envelopes to appear narrower than they really are.

        • Dale S
          Posted Apr 1, 2009 at 8:22 AM | Permalink

          Re: Ron Cram (#270),

          Perhaps I’m off-base, but just on the basis of that Hansen quote, it seems as if the argument is as follows:

          1) Suppose doubled CO2 increases temperature by 3C before reaching equilibrium
          2) The increase in CO2 since 1850 hasn’t produced near the temperature increase we’d expect from #1
          3) Therefore, we haven’t reached equilibrium from the increased CO2, and temperature must go up further even without further CO2 increases.

          If you grant point #1, I think #3 must logically follow, even without any casual mechanism being identified to do it. Being a layman, my first instinct would be to say if #2 is true, #1 probably is NOT true without some very compelling reasons to think otherwise.

        • Ron Cram
          Posted Apr 1, 2009 at 8:29 AM | Permalink

          Re: Dale S (#273),

          You are correct. Hansen’s view is that the oceans are storing much of the excess heat. The oceans will release this heat into the atmosphere over time. The approximate release rates are known. This is part of the ocean-atmosphere interface.

        • DeWitt Payne
          Posted Apr 1, 2009 at 10:05 AM | Permalink

          Re: Ron Cram (#275),

          The oceans will release this heat into the atmosphere over time. The approximate release rates are known. This is part of the ocean-atmosphere interface.

          All this should be phrased in terms of an energy balance model. Your heat source is the radiative imbalance. The reservoirs are the atmosphere, the cryosphere and the world ocean (aquasphere?). Because the the ocean and possibly the cryosphere have much larger heat capacities than the atmosphere, the atmosphere will not attain equilibrium temperature and radiative imbalance will be maintained because the energy required to heat the atmosphere is flowing into the other reservoirs and their higher heat capacity means a slower increase in temperature. As those reservoirs are filled, more energy becomes available to heat the atmosphere. The heat stored in the ocean is never released into the atmosphere unless the imbalance goes in the other direction and the system cools. Think of a series of parallel capacitors connected through a single resistor to a common constant voltage source and apply a step change in the voltage. As the capacitors charge, there is never a flow of current from one capacitor to another.

        • Ron Cram
          Posted Apr 1, 2009 at 10:22 AM | Permalink

          Re: DeWitt Payne (#277),

          I agree with much of what you said. When you write: “The heat stored in the ocean is never released into the atmosphere unless the imbalance goes in the other direction and the system cools.” This may or may not be correct. The heat could be released more in the wintertime. At least that is my understanding.

          Re: Gerald Machnee (#278),

          The oceans warmed significantly in the 1990s and supported Hansen’s view. But the oceans have not warmed since 2003. According to Pielke, if a radiative imbalance existed then it would show up in warmer oceans year after year. The fact the oceans are not warming means AGW theory should be modified to agree with these observations.

        • Posted Apr 1, 2009 at 11:47 AM | Permalink

          Re: DeWitt Payne (#277),
          I don’t like your reservoir analogy.
          The state of the atmosphere depends always on the underlying ocean. To state that, as the ocean is filled, then the atmosphere can act as the new unfilled reservoir is misleading.
          The climate system is made of different units but its behaviour must be regarded as a unity.

        • DeWitt Payne
          Posted Apr 1, 2009 at 2:05 PM | Permalink

          Re: Paolo M. (#284),

          It’s far from a perfect analogy as the equilibrium temperatures for each reservoir (voltage drop across the capacitors) will not be the same. But it is unitary in the sense that all capacitors charge at the same time from the same source. Each capacitor would also have to have its own resistor as well so there is an associated time constant. In addition, the hydrosphere probably has multiple time constants. It’s a toy model of a proper energy balance model, which itself would be a simplified model of the system. I think it gives a better feel for what might be happening than saying there’s heat in the pipeline.

          The oven analogy can be improved as well by saying the oven heater is set to constant power and the equilibrium temperature, where heat lost through the insulated walls is balanced by the power to the heater, is 300 F. Now boost the heater power by a few percent. The oven warms to a new temperature over a given time span. But lets say there was a 50 pound block of steel in the oven that was also at equilibrium initially. Now boost the power by the same amount. The final equilibrium temperature will be the same, but it will take longer to get there than with an empty oven. Now lets say that the power level is increasing linearly with time up to a certain point and then will remain constant. The oven temperature will continue to increase long after the power level stops increasing. That is my interpretation of what Hansen means by the rather infelicitous phrase ‘heat in the pipeline’.

        • Gerald Machnee
          Posted Apr 1, 2009 at 10:12 AM | Permalink

          Re: Ron Cram (#275),
          Sure, Hansen may say that the oceans are storing the excess heat. It is “plausible” that the oceans “may be” storing excess heat. They are “capable” of storing more heat than the air or ground surface.
          But is there really excess heat left there. The cooling surface or near surface must indicate that there is cooling below. Has anyone found pockets of significantly above normal temperatures down there? How long have they been measuring all the depths?

        • Ron Cram
          Posted Apr 1, 2009 at 10:27 AM | Permalink

          Re: Gerald Machnee (#278),

          Regarding your question on oceans depths, ARGO has been routinely sampling ocean depths to 700m since 2000. ARGO reached global coverage in 2003. Sampling of lower ocean depths is rare but so far has not indicated any warming at the lower depths, but then you would not really expect the lowest depths to be warming if the near surface was not warming. If ARGO and lower depth sampling had been in place during the 1990s, we would know more about the mechanisms of heat movement in the oceans.

        • Andrew
          Posted Mar 31, 2009 at 10:53 PM | Permalink

          Re: Alexander Harvey (#233), Your exact argument, as I understood it, was that I should accept Hansen’s arguments because they fit my beliefs now. I will decide whether I agree with Hansen’s explanation only on the basis of whether it makes sense.

          Re: Raven (#251), To bad for them there is no basis for the idea that there was still volcanic cooling “in the pipe” so to speak:

          http://arxiv.org/pdf/physics/0509166

        • Raven
          Posted Mar 31, 2009 at 11:07 PM | Permalink

          Re: Andrew (#254)
          Tamino does the math here: http://tamino.wordpress.com/2008/10/19/volcanic-lull/
          It was in response to people poking him about LTP.

        • Andrew
          Posted Mar 31, 2009 at 11:11 PM | Permalink

          Re: Raven (#255), Its definitely, er, interesting, but not convincing.

        • Raven
          Posted Mar 31, 2009 at 11:24 PM | Permalink

          Re: Andrew (#256)
          Well, he acknowledged the gross error during the war years and implies the models would do better. But if you look at the AR4 graph I linked above you will see the models don’t do any better.

        • bender
          Posted Apr 1, 2009 at 3:09 AM | Permalink

          Re: Raven (#255),

          It was in response to people poking him about LTP.

          LTP is not a competing theory to explain 20th c. warming. It’s a statistical problem for anyone wishing to estimate forcings based on the assumption that weather noise (internal climate variability) is white or red or pink noise. A 1/f noise model competes for “signal” and ends up drawing down the forcing coefficients.
          .
          Seems none of the wise asses at Open Mind understood this. Maybe because they don’t understand how these estimates are made in the first place?

        • Raven
          Posted Apr 1, 2009 at 3:22 AM | Permalink

          Re: bender (#261)
          Do you any comments on the Linzden paper on Watts? The more I look at the less I like it. Way too much significance attached to a signal uncertain measurement from one part of the globe. The fact that he has not updated his paper with the latest (albeit well adjusted) datasets raises huge questions in my mind.

        • Alexander Harvey
          Posted Apr 1, 2009 at 6:32 AM | Permalink

          Re: Andrew (#254),

          Your exact argument, as I understood it, was that I should accept Hansen’s arguments because they fit my beliefs now. I will decide whether I agree with Hansen’s explanation only on the basis of whether it makes sense.

          No, I do not think I mentioned mentioned your beliefs only mine. I understood the Hansen argument in 2005 and I understand it now. I thought it was a valid argument then and I do now. I will still think it valid if the OHC takes a sharp turn upward.

          I actually wrote:

          I can not see why one should abandon a battleground that is fit for purpose just when the tide appears to be turning. To do so seems to me to encourage dismissal of valid arguments by the mainstream community.

          I do not think that moving the goal posts (redefining what the terms mean) is very helpful in general and perticularly not now. The current OHC figures will argue against high end CS values. I think that this is an argument well worth making. I read the Hansen 2005 paper back then and I accepted the conclusion given the data available. Now that there is more data available I think that one would draw slightly different conclusions. At no point did I suggest that one should accept the paper as valid in order to draw such conclusions.

          I also feel that redefining the terms as Pielke seems to have done is only likely to lead to ideas, that are couched in the new definitions, being dismissed by people who use the Hansen definitions. I think that Pielke’s emphasis on OHC is very important as do Urban & Keller. It imposes a constraint on the high en of CS values that might otherwise be lacking. I think that drawing attention to their paper is helpful. But the inference that they use a different interpretation of “unrealized warming” to Hansen’s was not helpful.

          I hope that I have made myself clear.

          Alex

  35. Andrew
    Posted Mar 21, 2009 at 9:57 PM | Permalink

    Raven-interesting. I know that Lindzen has done a lot of work on Rossby Waves-I’m not sure if would know how they relate to the PDO though. But what exactly is their point?

    • Raven
      Posted Mar 21, 2009 at 10:09 PM | Permalink

      Re: Andrew (#76)
      The point was Rossby Waves cause the PDO which means we know why the PDO occurs and know that there are not any longer term oscillations.

      • Andrew
        Posted Mar 21, 2009 at 10:46 PM | Permalink

        Re: Raven (#78), In what? The PDO? Who ever suggested that? Proxies seem to show PDO going up and down and up and down and up and down. Is this pattern predicted by the Rossby wave idea?

        • Raven
          Posted Mar 22, 2009 at 12:43 AM | Permalink

          Re: Andrew (#81)
          My understanding so far is the Rossby waves are the things that shove the ocean currents around and cause the PDO/AMO switches. The waves themselves are likely still chaotic phenomena which means it would hard to say that they ‘predicted’ any given pattern. But it could mean that the mechanism for the PDO switches is known and therefore quantifiable. This could, in turn, allow one to rule out certain patterns such as a 1000 year pattern of significant amplitude.

        • Andrew
          Posted Mar 22, 2009 at 1:03 AM | Permalink

          Re: Raven (#83), Which would perhaps mean that PDO is unrelated to, say, the MWP. I’m still not getting the “clinch” factor that makes this fact crucial for defending AGW.

        • Raven
          Posted Mar 22, 2009 at 1:16 AM | Permalink

          Re: Andrew (#84)
          The best alternate hypothesis on the table is the the climate just changes chaotically. I felt this was a reasonable assumption since we have these unexplained events like the PDO so it was plausible to think of cycles that happen on longer scales. Obviously, one can still postulate longer cycles but it is less clear that we evidence that they exist.

          OTOH, your graph is interesting because asymmetry in the PDO could give rise to secular trends. i.e. the area under the graph is larger in the 1600s than in the 1800s.

        • Andrew
          Posted Mar 22, 2009 at 2:43 AM | Permalink

          Re: Raven (#85), I’m really bad at visually integrating, so I hadn’t noticed. Nice catch! That is interesting. Well, after some intense effort, I was able to produce a similar graph for AMO from these data sources:
          ftp://ftp.ncdc.noaa.gov/pub/data/paleo/treering/reconstructions/amo-gray2004.txt

          http://www.cdc.noaa.gov/tmp/ca55cfa2582cad6a710e5750fecb6fe4/data.76.110.165.189.80.2.39.42

          I probably screwed up at some point, so just ignore the y axis

        • Raven
          Posted Mar 22, 2009 at 2:53 AM | Permalink

          Re: Andrew (#86)
          That is cool – it is consistent with the PDO.

          But we still have the chicken and egg question. i.e. are the AMO/PDO indexes lower because something else lowered the temperatures or are the AMO/PDO indexes causing the temperatures to decrease?

        • Andrew
          Posted Mar 22, 2009 at 3:18 AM | Permalink

          Re: Raven (#87), Careful, talking of such indices as causes can get you into trouble! Personally, I think they are just indicators of some sort of underlying shift of climatic “regimes”-They couple and decouple and behave chaotically. All very interesting but it still doesn’t quite tell us what we want to know. As it stands, these modes we know at least are well reproduced by GCMs.

        • Raven
          Posted Mar 22, 2009 at 3:33 AM | Permalink

          Re: Andrew (#88)
          The issue is how to quantify the amount of internal variabilty in the system. For example, we can safely say that the system is bounded enough to ensure that unforced 5 degC shifts are not going happen randomly. OTOH, 1 degC shifts do not seem be that out of the question.

        • Craig Loehle
          Posted Mar 23, 2009 at 6:52 AM | Permalink

          Re: Andrew (#85), Andrew: you are right to say we must be careful about causation re AMO etc. but I believe it is NOT true that the models are able to reproduce AMO, PDO etc.

        • Andrew
          Posted Mar 23, 2009 at 8:05 AM | Permalink

          Re: Craig Loehle (#101), Yes, I meant to say “are not” my bad! :)

  36. rephelan
    Posted Mar 21, 2009 at 10:06 PM | Permalink

    Zamboni:

    looks like that paper just won’t stay snipped.

  37. AnonyMoose
    Posted Mar 21, 2009 at 10:33 PM | Permalink

    “Are we training pit bulls to review our manuscripts?”: blogger worried that reviewers might be too harsh. We can’t be overly strict in our scientific publications, can we?

  38. Solomon
    Posted Mar 22, 2009 at 7:03 AM | Permalink

    Re MJW #9 and #82
    Has anyone yet proved that the CO2 – temperature correlation is not spurious? We assume that the CO2 – CPI correlation is spurious, but is there any proof (other than “common sense”)?

    When I studied “random variables”, as statistics was known in the dim distant past, it was esssential to consider the validity of the assumptions which were to be used in creating the tests to verify the hypothesis.

    It strikes me that often when using more and more advanced mathematical models with which to test our hypotheses we may lose sight of the need to assure ourselves that the assumptions on which the mathematics are based are still valid.

    • MJW
      Posted Mar 22, 2009 at 7:26 PM | Permalink

      Re: Solomon (#87), My opinion is that I tend to accept the consensus view that increased CO2 should cause some warming, but that linear regression is useless in determining how much CO2 affects temperature.

      If I had to search night and day to find a spurious correlation, it could be argued that since it’s matter of probability, one can always find an occasional spurious correlation, but that doesn’t mean correlation is meaningless. The problem with that argument in this case is, first, that the CPI example was the first one I tried for which I could find an adequate amount of historical data, and second, that the p-value isn’t just 0.05 or 0.01, it’s 2*10^-16 (which seems to be the minimum p-value for R’s lm function), so if there weren’t some underlying problem, my chance of finding such a good counterexample would be small, to say the least.

      Bender mentioned how few effective degrees of freedom there are in this type of situation where the regression is between two rising trends such as temperature and CO2 or the CPI. That’s certainly more what I had in mind when I originally dismissed the regression than all the stuff about autocorrelation of the residuals and ARMA processes. I don’t know if there’s some theoretical way to determine the effective degrees of freedom to account for this (perhaps Bender does), but it seems obvious that if two functions are repeatedly moving up and down together for many years that that’s a lot better indication that they’re somehow connected than if two functions are basically both going up for the same period. It seems — off the top of my head — that it’s related to information theory.

      • MJW
        Posted Mar 23, 2009 at 2:14 AM | Permalink

        Re: MJW (#97), To be clear, the consensus view is that there should be some warming; it’s my own view that regression isn’t useful for estimating the amount of warming. As far as I’m aware, there’s no consensus one way or the other on the second point.

  39. Bill Illis
    Posted Mar 22, 2009 at 7:07 AM | Permalink

    Craig’s paper just indicates that the warming from GHGs must be far less than predicted by the models and the theory.

    This result would be consistent with a warming potential from GHGs of around one-third of that predicted with natural ocean cycle and solar variation accounting for the up and down swings in the climate.

    The modelers were riding the wave up from the major climate shift of 1976 to 1998 (or 2006) ascribing that increase to GHGs when it was really caused by the AMO and a slight increase in El Ninos.

    [The PDO is just an extension of the last two or three or four ENSO events or you could look at them as being part of the same system. At the end of the ENSO regions, the ocean currents turn to the north toward Japan and the Kuroshio current (the Gulf Stream of the Pacific) and the prevailing winds turn to the northwest toward California and Alaska and the whole northern Pacific gets impacted by an accumlation of ENSO events through these two drivers. The actual PDO index is very hard to work with and it doesn't give very good correlations with temperature and the climate. The ENSO alone works better.]

  40. AnonyMoose
    Posted Mar 22, 2009 at 7:34 AM | Permalink

    Or what the index is measuring has side effects. In the case of the PDO, the “cool” phase places cool water next to North America, and that might encourage the Arctic air flowing along the Rocky Mountains, thus dragging down temperatures across the continent. Whatever causes the PDO may be completely unrelated to some of the effects.

  41. Alexander Harvey
    Posted Mar 22, 2009 at 3:47 PM | Permalink

    From a NOAA paper Global Ocean Heat Content 1955-2008 in light of recently revealed instrumentation

    I glean that from ~1970-~2008 (a local minimum to a current maximum) the trend in OHC is .4*10^22 J/yr.

    Now that is (I think, so please check) ~.25 Watts/m^2 globally (~.36 Watts/m^2 Ocean only).

    Are such values not rather low when seen in the light of a 3C/CO2 doubling?

    GISS gives the increase in total forcing (1880-2003) at around 1.9 Watts/m^2

    I think that such a low oceanic uptake would imply a sensitivity closer to ~1.5C/CO2 doubling assuming a temperature increase of ~0.7C.

    I stress that the trend was from a local minimum (See Figure 1 in the paper). I do not know what it would be from the local maximum around 1958 except it would be a lot less.

    Perhaps someone here would look at the paper, do some sums and give me their opinion. I may simply have done something very wrong but the paper in Table T1 (last page) gives a figure of 0.364 Watts/m^2 and states that this is “per unit area of ocean surface” so I think I am in the ballpark.

    All I can say is that ~.25Watts/m^2 imbalance does not equate to a lot of warming in the “pipeline”.

    Does anyone know if it is possible to get OHC data from the CMIP2(+) archived model outputs?

    Alex

  42. curious
    Posted Mar 22, 2009 at 7:11 PM | Permalink

    Re: Ron and Alex above: FWIW – agree the arithmetic on the bulk figure. As an observation Page 33 of the paper suggests a factor of 8 difference between highest (North Atlantic) and lowest (North Indian Ocean) equivalent fluxes for 1969-2008 for 0-700m layer. Page 29 S11 shows regional time series for 1955-2008 – by eye the heat content levels/falls from 2002ish. Levitus’ earlier papers show as being previously discussed on CA.

    • Ron Cram
      Posted Mar 23, 2009 at 7:40 AM | Permalink

      Re: curious (#96),

      I do not doubt the math, I doubt the physical theory underlying it. Trying to determine a radiative imbalance over a multi-year trend in ocean heat content would only make sense if the trend was consistent. The relationship between radiative imbalance and ocean heat content is more like an accounting problem in which you are required to match revenues and expenses in the same time period. There is no physical theory that says a radiative imbalance in 2003 will cause the oceans to warm in 2010 or even in 2004. If a radiative imbalance exists in 2003, the oceans will be warmer in 2003 than in 2002.

  43. AnonyMoose
    Posted Mar 22, 2009 at 9:20 PM | Permalink

    “Dispelling the Global Warming Myth” has nice collection of graphs from Heartland conference.

  44. pj
    Posted Mar 23, 2009 at 5:36 AM | Permalink

    Thanks Robert Freerks (#6) and Mark T. (#7) for the book recommendations.

  45. TomVonk
    Posted Mar 23, 2009 at 7:26 AM | Permalink

    So many posts during the week end when I am not here .
    I’ll try to sum up .

    The general gist of those (Craig , Bender , PaulM , RyanO , Harry Eagar) who hope that :
    a) the climate is not REALLY chaotic and/or that there are “grades” of chaos
    or
    b) if it is , then there are SOME statistics that one could make
    or
    c) a small perturbation will/could/should elicit linear answer
    consists to actually say the same thing .
    The climate is not chaotic .
    At worst pseudo periodical but not chaotic .
    I have no problem with that assumption but all my arguments are based precisely on the opposite hypothesis – the climate is chaotic .
    For those who don’t remember why I think that , it is because weather is chaotic and the climate is a time average of the weather .
    That’s why it is rather hard to use chaos theory for arguments where the assumption is that it doesn’t apply to the problem at hand :)
    .
    But I will try to add some comments despite everything , bearing in mind that what I say , per definition , applies to chaotic systems only .
    .
    First Craig :

    You are assuming “forcing” to be constant. what if the sun oscillates its output on a period of X years? Just for argument sake. Even small oscillations in the external forcing can kick the earth climate into different parts of its attractor (assuming there is one) or change the control parameters, as you say, to give a different dynamic.

    I only assume that the system is chaotic . I tried , apparently without success , to explain why I will not use terms and concepts like “external” and “forcing” .
    Dan Hughes already explained that issue remarkably well and with mathematical rigor .
    This bad term of “external forcing” actually means “energy supply” .
    You will all recall that there is NO chaotic system without energy supply and energy dissipation .
    In the most general case the energy supply is some finite sum over i of
    [Ai(t).Cos(Wi(t).t + Pi(t))]
    So this term will figure in those N ODE that describe the dynamics .
    You notice all those t ?
    That means that the control parameters in this general case vary with time .
    The attractors do the same and there will appear regimes where the systems stops being chaotic .
    All this are details . The important thing is that WHEN the system is in a chaotic regime it is not predictible , answers non linearily on perturbations and presents no statistical properties . Equally important is that if I take out one or several terms of the sum defining energy supply and am still in a chaotic regime , you will be unable to say what terms I have taken out by only looking at the 2 solutions .

    .

    There are grades of chaos. It’s the rate of exponential growth that matters. My presumption is that the climate system is an in-between system that is not “COMPLETELY” chaotic. Recall we are talking about spatiotemporal chaos here, not your garden variety type. I am happy to be proven wrong in this or any of my presumptions.

    This is a powerful remark and I would be the last to forget that we are talking about spatio-temporal chaos and I finished on this issue in my “Chaos Q&A” in the previous unthreaded .
    Of course the easy way out by saying that we consider local variables is not available because it would mean that variables separate and they , by definition , don’t .
    However I mentionned that one can define a function (current/flow function) that is formally equivalent to the Hamiltonian in the purely temporal chaos case .
    So it is possible to go from spatio-temporal to temporal only . This is technical and really poorly understood . I will try to make a post specifically on this issue later . At that stage I can’t prove you wrong Bender , especially as the “in-between chaos” is not terribly well defined .
    .

    More importantly, I am saying that while the system may be chaotic, it may have properties that are predictable and to which statistical tools can be accurately applied. If this were not true, then I would be at a loss to explain the predictability of macroscopic fluid dynamic properties. A related question would be how would someone distinguish between predictable and unpredictable properties if only partial information is available.

    RyanO I choose this quote among your long and well written posts because it pretty well sums everything up .
    Actually what it shows is that most people are not really comfortable with the true chaos and the tools that analyse it .
    You ask what can be predicted and what cannot ?
    Well the velocity field can’t and that is pretty damning .
    The vorticity neither and that is even more damning as Jerry Browning would immediately confirm because enstrophy is not conserved in 3D flows .
    So what are you thinking about when you mention “predictability of macroscopic fluid dynamic properties” ?
    Pressure ? But that is energy . Temperature ? Energy too .
    Do you want to say that something can be said about energy in chaotic systems ?
    Well that would be a rather trivial statement because chaotic systems are conservative . More technically they live on an hypersurface of dimension 2N-1 given by a constant Hamiltonian . Constant Hamiltonian ? Constant energy . That’s all ergodic theory is about . Doesn’t teach us much about the dynamics of the system , does it ?
    .
    But let’s get more specific .
    Energy is velocity squared . Even if I perfectly knew energy , could I reconstruct the velocity field ? Of course not because you can’t reconstruct 3 vector components from 1 scalar . That’s why Navier Stokes can’t be solved by Reynolds averaging without additionnal more or less empirical closure equations .
    Without getting too technical that’s the philosophy of Kolmogorov statistical turbulence theory too .
    To tackle the problem , he supposed that the vortex distribution was HOMOGENEOUS and ISOTROPIC .
    The former gets rid of the time and the latter of space directions .
    So now provided that you are only interested in squared velocity components (aka energy) you can develop a statistical turbulence theory (very similar to statistical mechanics) which allows to derive the energy spectrum function of wave number (spatial scales) .
    It works more or less for very high Reynolds numbers and fails completely for low ones .
    The easily understandable rule you can use is that if a flow is obviously non isotropic and homogeneous at the scale of the problem , you can say nothing about its dynamics and very little about the energy .
    Cigarette smoke is non isotropic with low Reynolds , Kolmogorov fails and we have a true spatio-temporal chaos . No 2 cigarettes smokes are the same even if they all share a general “cigarette smokiness” shape because of similar control parameters and therefore similar attractors . Velocity field is totally unpredictible with the exception of the very small space immediately above the cigarette where the Reynolds are highest and the density fluctuations smallest .
    In conclusion you see that all of the above has almost nothing to do with chaotic systems – we are in a steady state (the time disappeared) with isotropy (space disappeared) .
    Not much is left only a few scalars connected to energy …
    But as it is energy that makes planes fly , it is useful for this task because we are blessed by a high Reynolds environment :)

    • Ryan O
      Posted Mar 23, 2009 at 10:44 AM | Permalink

      Re: TomVonk (#102),
      .

      Do you want to say that something can be said about energy in chaotic systems? Well that would be a rather trivial statement because chaotic systems are conservative. More technically they live on an hypersurface of dimension 2N-1 given by a constant Hamiltonian. Constant Hamiltonian? Constant energy. That’s all ergodic theory is about. Doesn’t teach us much about the dynamics of the system, does it?

      .
      Thank you for that paragraph. The light bulbs are beginning to turn on. All of the “macroscopic properties” I can think of relevant to fluid flow are fundamentally energy quantities.
      .
      This means my “bounds” with respect to internal variability are nothing more than misguided restatements that the energy within the system must be conserved. The system cannot reach states that violate the conservation of energy, but all other states are fair game. Furthermore, there is no way to predict when the system will reach any particular state.
      .

      In conclusion you see that all of the above has almost nothing to do with chaotic systems – we are in a steady state (the time disappeared) with isotropy (space disappeared).

      .
      Again, perfect sense. I had drawn a qualitative analogy in my head between the short-term predictability of cigarette smoke and possible short-term (~100 years or so) predictability of the climate. However, the important condition is not some qualitative “short term”; the important condition is to limit the consideration of the system to scales such that the time and space dependence are small enough to be neglected. Doing this would not be permitted for climate. By the definition of climate being a time-averaged representation of weather the time dependence cannot be removed unless you restrict your averaging to such small time scales that you end up with the trivial result that climate at time A ~ climate at time B.
      .
      Extending this to what can be predicted, just as in the fluid case, we would be limited to quantities that are conserved. This would mean that, while using surface air temperature is not categorically ruled out, we would have to show that changes in SAT provide a good approximation of the changes in the system energy – a relationship that I feel is rather suspect.
      .
      It would also seem to prohibit predictions of micro-scale climate. While system energy will be conserved, energy is not conserved in an individual cell in the grid (cells will experience net influxes and outfluxes of energy for every time step). While you might be able to predict some energy flow using conduction and radiation, because you cannot (for example) predict the velocity field for wind, you cannot predict any net influx/outflux from any particular cell.
      .
      The reason models arrive at (somewhat) similar results for microclimate, then, is more a confirmation that they parameterize similarly (like using local energy balances to represent viscous dissipation) than a confirmation that independent methods corroborate the result.

      • TomVonk
        Posted Mar 24, 2009 at 3:15 AM | Permalink

        Re: Ryan O (#106),

        This means my “bounds” with respect to internal variability are nothing more than misguided restatements that the energy within the system must be conserved. The system cannot reach states that violate the conservation of energy, but all other states are fair game. Furthermore, there is no way to predict when the system will reach any particular state.

        Yes you got it !
        That is exactly what happens and you are now ripe to go deeper in the chaos theory :)
        A chaotic system has an infinity of dynamical states what is equivalent to say that a surface in the phase space (attractor) has an infinity of points on it .
        But they are all on a surface of constant energy .
        This is actually not exactly true for a system as complex as the climate because its energy is not constant but you got the meaning which is that for a given energy there is an infinity of possible states and the system will chaotically (this doesn’t mean “randomly” !!) switch between them .

  46. Posted Mar 23, 2009 at 3:38 PM | Permalink

    NASA Simulation Envisions World without Ozone

    March 20, 2009

    http://www.laboratoryequipment.com/news-NASA-ozone-hole-simulation-032009.aspx

    Another of those “Thank us for having averted Apocalypse, and don’t forget that you need us in the future to save you again” posts.

    An alternate title would be “the future will be bleak, you will die with your skin burned if you don’t listen to us”

    I just started to dawn on me that the politicos, badly in need of an impending apocalypse to save us from, especially considering that their ratings are flying like a brick with no engine (Atlas Shrugged, by Ayn Rand, briefly sold better than the latest B. Obama’s book last week), are trying to line-up some reasons why we need them despite the economic scourges they are inflicting upon us. Ozone hole and global warming are amazing occasions to regulate, to justify their warm and cozy payroll spot.

    • Alexander Harvey
      Posted Mar 23, 2009 at 4:52 PM | Permalink

      Re: JFA in Montreal (#107),

      I freely confess that I am not well versed on ozone depletion but at the end of the article it states:

      “Stratospheric ozone has been depleted by 5 to 6 percent at middle latitudes, but has somewhat rebounded in recent years. The largest recorded Antarctic ozone hole was recorded in 2006.”

      I can only ask whether that was to be expected?

      Alex

      • Andrew
        Posted Mar 23, 2009 at 7:18 PM | Permalink

        Re: Alexander Harvey (#108), I recall hearing that CFC’s were supposed to keep on eating away at ozone for 70 years, so I would imagine that it wasn’t. The thing about the ozone depletion scare is that UV exposure varies a lot by latitude for obvious reasons already, meaning “depletion” could be compensated for by a tiny move North or South. Anyway UVA, which is supposed to be implicated in melanoma, isn’t blocked by Ozone-UVB, which is not so important for cancer, is. Oh, and then there’s this:

        http://www.nature.com/news/2007/070924/full/449382a.html

  47. cba
    Posted Mar 23, 2009 at 9:14 PM | Permalink

    1998 was the year that albedo plummeted around 10%. That is about the equivalent forcing wise of 3 doublings of co2 around 10w/m^2. It was undoubtedly clouds that caused such a large change – which recovered substantially afterwards.

    There is excellent correlation between mean surface T and albedo measurements too.

  48. crosspatch
    Posted Mar 23, 2009 at 10:49 PM | Permalink

    “An externally threaded fastener that has thread form which prohibits assembly with a nut having a straight thread of multiple pitch length is a screw. (Example: wood screws, tapping screws.)”

    Hmm, I use machine screws all the time that are not tapered and are torqued with a nut.

    “has a cylindrical shaft and fits into a nut or a tapped hole, a small bolt.”

  49. crosspatch
    Posted Mar 24, 2009 at 1:43 AM | Permalink

    “reality will determine”

    Novel concept these days. Reality seems to be what 51% of the people believe. The “democratic process” and all. (World According to Hansen c.2009)

  50. R Keene
    Posted Mar 24, 2009 at 10:29 AM | Permalink

    This article here describes new standards for what constitutes “Dangerous Anthropogenic Interference”. The proposal is to define it down so less change is still a crisis. So let me get this straight, since global warming now appears to be less than anticipated, lets change the definitions so less is still bad?

    • Andrew
      Posted Mar 24, 2009 at 12:17 PM | Permalink

      Re: R Keene (#121), That’s an eye roller right there…

    • John Baltutis
      Posted Mar 24, 2009 at 2:12 PM | Permalink

      Re: R Keene (#121),

      Same list of fear mongers that have been spouting the IPCC nonsense for over two decades.

  51. EddieO
    Posted Mar 24, 2009 at 11:43 AM | Permalink

    The extent to which the IPCC have changed our lives with their poorlty justified claims of imminent global catastrophe was illustrated to me on the way home from work last night. A reporter was interviewing a nice lady from our local ambulance service who have just invested heavily in new ambulances. She listed all of the modern lifesaving features they have fitted and was obviously delighted with the new vehicles. Then she let it slip that they had chosen the four wheel drive model because of the expected future problems with “global warming”. (Surely there is some irony in this logic!)

    This got me to thinking about my own job in an engineering faculty in a University. Most of my academic colleagues consider it essential to justify all of their new research applications by including a reference to the imminent catastrophe of “global warming”. To my knowledge almost none of them have bothered to criticise or even look at the published evidence for or against AGW, but they are more than happy to quote it as an undisputed fact when applying for research funds. So much for the free thinkers in Academia.

    In their defence they are neither climatologists nor physicists.

  52. Hu McCulloch
    Posted Mar 24, 2009 at 4:08 PM | Permalink

    Tonight …

    Next on NOVA: “Extreme Ice”

    http://www.pbs.org/wgbh/nova/extremeice/

    Tuesday, March 24 at 8 p.m. (Check your local listings as dates and times may vary. Broadcast in high definition where available.)

    Remarkable time-lapse footage by James Balog, one of the world’s foremost nature photographers, reveals massive glaciers and ice sheets splitting apart, collapsing, and disappearing at a rate that has more and more scientists alarmed. This NOVA-National Geographic Television special investigates the latest evidence of a radically warming planet.

  53. Hu McCulloch
    Posted Mar 25, 2009 at 9:12 AM | Permalink

    RE #125, 126,
    It was an interesting show. Even if one disagrees with its presuppositions, it will be widely discussed and influential, so it’s worth watching and thinking about.
    It is intriguing that one of the co-sponsors was Exxon-Mobil. Will this discredit the show in the eyes of warming alarmists?
    A lot of the time-lapse photography was of ice shelves that were calving, then moving forward, and then calving again, with essentially no net movement of the calving face. However, all that snow that keeps burying Antarctic AWS’s — see Anthony’s thread Antarctica — Digging out the Data” doesn’t just build up, it slowly oozes downhill toward the sea. So the calving depicted does not necessarily mean that Antarctica is losing ice, just that the ice it has is turning over.
    A similar fallacy would be to assume that N. America is dessicating because all its water is running into the ocean in rivers.
    The show did at least demonstrate that Balog himself is clinically insane — anyone who would peer down a bottomless ice chasm without a rope tied to him is clearly nuts!

    • Andrew
      Posted Mar 25, 2009 at 12:12 PM | Permalink

      Re: Hu McCulloch (#128), Ah, see, its oil funded. We can safely ignore it. ;) And I to would worry about the sanity of a man who would peer down a deep hole without securing himself…Re: Alexander Harvey (#127), Hansen’s paper was highly misleading, kinda jumping to unwarranted conclusions, as has been hashed out before (I could, of course, dig up some references) however, maybe you could explain to me why your definition of heat “in the pipeline” is so different from RP Sr’s? Because he would appear to disagree with your claims.

      • Alexander Harvey
        Posted Mar 26, 2009 at 4:00 AM | Permalink

        Re: Andrew (#129), Re: Ron Cram (#130),

        Yes I am aware of Pielke’s point of view but I have a problem with it.

        He quotes from Complementary observational constraints on climate sensitivity and then says:

        By “unrealized warming in the pipeline”, they mean heat that is being stored within the ocean, which can subsequently be released into the ocean atmosphere.

        I can not see where in their paper they say anything to support his interpretation.

        Strangely the papers I would give that seems to intrepret the term in the way that I do would also be: Complementary observational constraints on climate sensitivity
        and the Hansen paper I mentioned above.

        I do not disagree with Pielke’s approach in that knowing the enthalpy is as important as knowing the temperature which is precisely what Urban & Keller state in the paper that he quotes.

        Alex

        • Alexander Harvey
          Posted Mar 26, 2009 at 4:27 AM | Permalink

          Re: Alexander Harvey (#138),

          If anyone reads the Urban & Keller paper be warned that they use the term “warming” to quantify either the increase in temperature or the increase in heat content depending on the context. There are places where this could cause ambiquity and even a reversal of meaning. They are by no means unique in using the term “warming” where clearer terms would serve better.

          Alex

        • Ron Cram
          Posted Mar 26, 2009 at 8:14 AM | Permalink

          Re: Alexander Harvey (#138),
          Thank you for citing these papers, but could you supply a quote?

          Re: Alexander Harvey (#139),

          If anyone reads the Urban & Keller paper be warned that they use the term “warming” to quantify either the increase in temperature or the increase in heat content depending on the context. There are places where this could cause ambiquity and even a reversal of meaning. They are by no means unique in using the term “warming” where clearer terms would serve better.

          This is a baffling comment, Alexander. The use of the term warming for an increase in surface temps or an increase in OHC is to be expected. While I can see it may cause ambiguity at times, I cannot see how it could ever cause a “reversal of meaning.” Warmer always means warmer. It never means cooler.

          Alexander, are you familiar with the ocean-atmosphere interface? I am sorry if I sound snobbish or anything but I cannot help but think there may be a gap in your conceptual understanding. Perhaps I am the one who has a gap in understanding.

  54. Mark
    Posted Mar 25, 2009 at 4:35 PM | Permalink

    Interesting to note:

    Last year there was a big fuss about the Hadley Centre changing their graphs/data for the annual temperature series to remove the current year YTD number because it “distorted” the true picture as to what was really happening (in other words it was dragging down the graph line).

    http://www.climateaudit.org/?p=2955

    Interesting to note that now that temperatures have crept up a bit, they are more than happy to show the YTD number in the annual unsmoothed data (but not in the smoothed data). These guys have an agenda or something?

  55. Andrew
    Posted Mar 25, 2009 at 7:41 PM | Permalink

    Behold, I sayeth to Boris, the surface data hath issues, he doth respond that if I doth not think there hath been any surface warming, he doth not know what to say. Non sequitor I do verily say!

    http://rankexploits.com/musings/2009/with-all-due-respect-aint-gavin-a-hoot/#comment-12530

  56. Hemst 101
    Posted Mar 25, 2009 at 8:55 PM | Permalink

    Re Extreme Ice

    The photography was excellent and the time lapse material interesting. However, when they repeated the ice core/CO2 graph and left out the very important fact that CO2 lags temperature, thus implying once again that CO2 drives temperature change, I had had enough and switched off the show.

  57. Geoff Sherrington
    Posted Mar 25, 2009 at 10:13 PM | Permalink

    An item out of the blue again for Unthreaded. Or rather, going into the blue. “Unprecedented” is used, so it’s climate related.

    In a couple of years, all going well, we might learn a bit about the gravity of the earth. It’s outlined below.

    New Launch: 2009 March 17, 1421 UTC
    Site: Plesetsk Missile and Space Complex, Russia
    Launcher: Rockot
    International Designators(s): 2009-013A

    SSC Name Owner
    34602 GOCE ESA

    “Europe launched a slender, winged ion-driven satellite Tuesday to glide through the upper atmosphere for nearly two years measuring Earth’s gravity field with unprecedented precision.

    “The arrow-shaped 2,319-pound [1,052-kg] satellite blasted off at 1421 GMT (10:21 a.m. EDT) from the Plesetsk Cosmodrome in northern Russia. The 95-foot-tall [29-m-tall] Rockot launcher, a modified SS-19 ballistic missile, roared away from the snowy, forested spaceport on 420,000 pounds [1,868,000 N] of thrust.”

    “The European Space Agency commissioned the GOCE mission to create a highly detailed, exceptionally accurate map of the planet’s geoid, a global model illustrating subtle variations in the gravity field if oceans were motionless.

    “But oceans aren’t stationary. Ocean currents and changing sea levels can have drastic consequences on Earth’s climate.”

    “Scientists use the geoid’s reference surface to weigh against measurements of ocean activity. The comparisons allow scientists to more accurately study ocean circulation and sea level changes.”

    “GOCE data will also provide a new understanding of tectonic activity that could lead to better forecasts of earthquakes and volcanoes. Scientists expect changes in Earth’s interior to show up in the satellite’s gravity measurements.” (GHS comment …dah)

    “GOCE, which stands for the Gravity field and steady-state Ocean Circulation Explorer, is the first of six ESA Earth Explorer satellites built to focus on aspects of processes affecting the planet’s climate.”

  58. Craig Loehle
    Posted Mar 26, 2009 at 9:53 AM | Permalink

    My paper on ocean cooling since 2003 is now posted:

    http://www.ncasi.org/publications/Detail.aspx?id=3152

  59. Neil Fisher
    Posted Mar 26, 2009 at 3:30 PM | Permalink

    Oops – damn wordpress! should have been a “not equal” before the 0 in my last post. I used greaterthan/lessthan and it swallowed it as a an HTML tag I guess.

  60. Mike Davis
    Posted Mar 27, 2009 at 7:15 AM | Permalink

    Craig:
    I saw the same comment regarding the globe cooling because the ocean was storeing the heat for later. I do not remember it was last year some time.

  61. DeWitt Payne
    Posted Mar 27, 2009 at 11:14 AM | Permalink

    Can someone explain to me how the ocean heat content can be constant but the sea level is still rising? Maybe I’m missing something, but I don’t see how that can happen.

    • Posted Mar 27, 2009 at 11:32 AM | Permalink

      Re: DeWitt Payne (#177),
      do you mean in absence of ice cap melting?

      Mixing of warmer water with deep cold water, i.e. heat flowing downwards.
      Of course you have to be very sure of your altimeter measurements…

      • DeWitt Payne
        Posted Mar 27, 2009 at 4:50 PM | Permalink

        Re: Paolo M. (#179),

        Heat content to a first approximation is temperature times total mass times heat capacity. Sea level can increase by an increase in total mass at constant temperature or an increase in volume caused by an increase in temperature. In either case alone or some combination of these effects should result in an increase in total heat content. A loss in volume of sea ice would not change the sea level, at least to a first approximation. It would count as an increase in syatem heat content, but would it affect ocean heat content as measured by ARGO? ARGO has only been in operation since 2003 and there has been one significant correction already. I think we should be careful of confirmation bias and take the current ARGO results cum grana salis.

        Re: Craig Loehle (#183),

        Ocean heat content requires measurement to millidegree precision and accuracy in a free floating automated system, is that any more difficult than altimetry? Altimetry has been in operation long enough to find and work out a lot of bugs and has several independent means of measurement for confirmation. That doesn’t mean the ARGO data is wrong, but it does cause me at least to not accept it unquestioningly.

        I ran a quick calculation and if I did my sums correctly, a change of 0.014 degrees/year in the average temperature of the top 700 meters of ocean is equivalent to a radiative imbalance of 0.85 W/m2 over the entire planet.

    • Craig Loehle
      Posted Mar 27, 2009 at 12:43 PM | Permalink

      Re: DeWitt Payne (#177), Some of the sea-level rise charts I have seen look like the rate of rise has slowed in the past few years. I suspect also that this data is very difficult to get right year-to-year at the very slight level of rise we currently have.

  62. curious
    Posted Mar 27, 2009 at 11:53 AM | Permalink

    Interesting paper on the relationship of temps in the top 5m layer and SST here:

    http://www.scielo.cl/scielo.php?pid=S0717-65382004000200055&script=sci_arttext

    WEAKENING OF HORIZONTAL SEA SURFACE TEMPERATURE GRADIENTS AT LOW WIND SPEEDS WITH STRONG INSOLATION

    Kristina B. Katsaros

    Not sure about fig4 but all else seems reasonable. Might be relevant to the discussion above.

  63. Andrew
    Posted Mar 27, 2009 at 12:42 PM | Permalink

    Playstation® Climatology strikes again!

  64. DeWitt Payne
    Posted Mar 27, 2009 at 6:53 PM | Permalink

    Yet more calculations:

    If the sea level is increasing due to land based ice melting, the increase in ocean heat content is quite small. A change in sea level of 3 mm out of the top 700 m has a very small effect on total heat content, which should have been obvious. It’s also not very much ice, about 1,000 km3, compared to the 33,000,000 km3 or so of land based ice. Thermal expansion of that magnitude, OTOH, requires a fairly large temperature change. How large is rather complicated since the thermal expansion coefficient of water varies a lot with temperature and most of the ocean is at maximum density and minimum expansion coefficient. So the ARGO data is not inconsistent with sea level increase and would seem to makes thermal expansion very unlikely and loss of land based ice volume much more likely as the cause of sea level increase, assuming both it and the sea level measurements are correct.

    • Posted Mar 28, 2009 at 3:37 AM | Permalink

      Re: DeWitt Payne (#185),
      since it seems that actual ice cap melting rate is too low and upper ocean content is stationary (from ARGO), it is inferred (not measured) that is heat flowing to the bottom of ocean to contribute to the sea rise.
      There are not (as Craig said) adeguate deep ocean measurements, except some local transect recently reported.
      Anyway, the eventual flow of heat towards the deep ocean is nothing extraordinary: it has been occurring since the end of the last glaciation and I’m not aware of a recent acceleration.

      I agree that all that has to be regarded cum grano salis.

  65. Andrew
    Posted Mar 27, 2009 at 7:01 PM | Permalink

    I just noticed that NOAA has change the graphic on their paleo-page from the Hockey stick to this:

    Apparently models are now preferred to botched data…

  66. David Wright
    Posted Mar 28, 2009 at 1:50 AM | Permalink

    I thought the CA community might be interested in this example, from non-climate science, of how correlation vodoo can drive spurious results. This effect has a long and sad history among many well-meaning scientists.

  67. Bill Illis
    Posted Mar 28, 2009 at 8:34 AM | Permalink

    All the paleo charts used in IPPC FAR can be seen on this page.

    http://www.ncdc.noaa.gov/paleo/pubs/ipcc2007/

    And the data is located here.

    ftp://ftp.ncdc.noaa.gov/pub/data/paleo/ipcc2007/

  68. Solomon
    Posted Mar 28, 2009 at 2:57 PM | Permalink

    I am fascinated. In all the experiments which we did at school we learnt that heat rises. When diving I have always found that, excpet for the immediate surface, the sea is colder the lower one goes. Why are oceans different? Is it because of the pressure which might affect the density?

    • DeWitt Payne
      Posted Mar 29, 2009 at 11:13 AM | Permalink

      Re: Solomon (#192),

      Air that is warmer and less dense than the surrounding air rises, but as it rises, it expands and cools. The apparent loss in energy is compensated by a gain in gravitational potential energy. The result is that the temperature drops with altitude. The rate varies with humidity. For dry air, temperature decreases at a rate (the adiabatic lapse rate) of 10 degrees C/km. Moist air cools more slowly with altitude (lower lapse rate) because of the latent heat of vaporization of water. A lapse rate greater than the adiabatic rate is unstable because the air below is less dense than the air above. Solar heating of the surface will increase the lapse rate and cause convection. Unstable lapse rates can also be caused by a cold air mass riding up over a warmer air mass.

      Water, OTOH, is essentially incompressible. The density is mainly a function of temperature and salinity. The thermal coefficient of expansion of water also changes sign and water becomes less dense as it approaches the freezing point. The deep ocean water is at about the temperature of maximum density at 4 C. This wasn’t always true. According to fossil oxygen isotope data, the deep oceans were about 13 C at the Eocene optimum about 50 million years ago when there were no permanent polar ice caps.

  69. AnonyMoose
    Posted Mar 29, 2009 at 10:07 AM | Permalink

    “Eye-opening Access” looks at the refutation of a biology study, with the second study appearing in an open journal which includes reviewer comments.

  70. Posted Mar 29, 2009 at 1:37 PM | Permalink

    Water, OTOH, is essentially incompressible. The density is mainly a function of temperature and salinity. The thermal coefficient of expansion of water also changes sign and water becomes less dense as it approaches the freezing point. The deep ocean water is at about the temperature of maximum density at 4 C.

    This is true of freshwater not saltwater which has a maximum density at the freezing point.

    Curves are lines of constant density.

    • Andrew
      Posted Mar 29, 2009 at 4:03 PM | Permalink

      Re: Phil. (#195), Yes it has a lower melting point, to. Doesn’t even have to get all that hot.

    • DeWitt Payne
      Posted Mar 30, 2009 at 9:48 AM | Permalink

      Re: Phil. (#195),

      Thanks for the correction and the link. Learn something new every day. I checked my numbers again with the new data and it still seems that the ARGO data implies that the increase in sea level must be mostly due to an increase in volume from loss in land ice volume rather than thermal expansion. A very large change in temperature over a small depth (1C/10m @ 20C) or a very small change over a larger depth (0.014C/700m @ 4C) that produces a sea level change of about 3 mm also implies a change in OHC of 1.4E22 Joules/year, much larger than the ARGO data. OTOH, melting 1000 km3 of ice only requires ~4E20 Joules.

  71. See - owe to Rich
    Posted Mar 29, 2009 at 1:50 PM | Permalink

    Is it possible that the overall ocean heat content is rising, hence rising sea levels, but the ocean heat content measured by Craig Loehle is decreasing? That is, the upper layers could be cooling, but the lower layers warming. Do data on that exist, or is it pretty much guesswork? (Excuse me if I’ve missed this being covered earlier.)

    Rich.

  72. DaveR
    Posted Mar 29, 2009 at 2:42 PM | Permalink

    Is there any compensation made for the rising and falling of continents, not to mention the creation of “new land” by volcanic activity, made in sea level measurements and if so, how?

  73. curious
    Posted Mar 29, 2009 at 7:16 PM | Permalink

    11 March 09 Hansen paper on forcings inc. aerosols here:

    http://www.columbia.edu/~jeh1/2009/Copenhagen_20090311.pdf

    Not sure where else to post it or if it’s been covered elsewhere and I’ve missed it. Feel free to move/snip if req.

    • Andrew
      Posted Mar 29, 2009 at 7:26 PM | Permalink

      Re: curious (#199), Aerosols “pulled out of a hat” eh? Well, he seems to include the same stuff everybody does in attribution studies, but I’m sure many here, including me, could elucidate many things he neglects to cover…

  74. Andrew
    Posted Mar 30, 2009 at 9:25 AM | Permalink

    Okay Steve, you got me. I could eat the snark and political over-the-top-ness of that post with a spoon.

    Anybody care to bet some quatloos on the March temp anomalies, up or down?

  75. Knut Witberg, Norway,
    Posted Mar 30, 2009 at 9:45 AM | Permalink

    Phil, thanks! I will try with modified wording. They will not get too many comments about the definition of a climate specialist. I did send the comment with an email also, and it has not appeared so far.

  76. Howard S.
    Posted Mar 30, 2009 at 7:32 PM | Permalink

    OT
    Sorry but I had to flag this.

    Gavid at RC spread the AGW=more C5 Hurricanes nonsense.

    …radio talk show host Nancy Skinner earlier today said that AGW is heating up the oceans and causing more category 5 Hurricanes.

    Gavin Schmidt

    http://www.realclimate.org/index.php/archives/2009/03/a-potentially-useful-book-lies-damn-lies-science/

    Posted
    “We’ve been told that AGW will lead to more frequent & destructive hurricanes.”

    Gavin response.

    [Response: It may well do. The magnitude of such an effect is still difficult to discern. - gavin]

    “may well do”?

    I may well win the lottery too.

    This is science?

    Gavin A. Schmidt is a central figure in the IPCC as a climatologist and climate modeller at the NASA Goddard Institute for Space Studies (GISS).

    Gavin Schmidt is misrepresenting IPCC science.

    “The IPCC assessments in 1995 and 2001 also concluded that there was no global warming signal found in the hurricane record.”

    http://sciencepolicy.colorado.edu/prometheus/archives/science_policy_general/000318chris_landsea_leaves.html

    • Andrew
      Posted Mar 30, 2009 at 8:58 PM | Permalink

      Re: Howard S. (#205), Nothing is OT in Unthreaded! Well, almost nothing…

      RC has been pretty damn biased on the hurricane issue. Considering that this is one of the weakest and most easily and commonly disputed parts of the alarmist thesis, the fact that they actually try to hold it up belies there advocacy. Science it is not.

  77. Raven
    Posted Mar 31, 2009 at 12:38 AM | Permalink

    Watts has posted a paper from Linzden: http://wattsupwiththat.com/2009/03/30/lindzen-on-negative-climate-feedback/

    Obviously it would be nice if the results were true but something does not seem right.

    In particular:

    1) Why doesn’t he update it with the latest models and the latest data? If would be worth the effort given how signicant this finding is.

    2) Why does he assume that a strong negative feedback in the tropics won’t be offset by strong positive feedback elsewhere?

    3) If there was such as strong negative feedback to a strong CO2 forcing then how could the GMST change at all? It would take huge forcings to cause any change to climate yet we have observed previous changes which may be random or may be related to relatively small solar changes. The only way it could make sense is if the feedback mechanism was specific to CO2 but he did not offer any explaination for that.

    Anyone hear have any ideas or a way to ask Linzden if he would answer questions on his post?

    • Andrew
      Posted Mar 31, 2009 at 9:00 AM | Permalink

      Re: Raven (#207), Lindzen has often dealt with the issue raised against negative feedback that it would appear to prevent any significant past changes. The answer is “Not exactly”. Say you have a very small change in net radiation from Milankovitch cycles. How can such small changes cause big climate changes even in the mild positive feedback paradigm? The answer is that the forcing is spatially heterogeneous-in the presence of a strong negative feedback in the tropics, alterations of horizontal heat fluxes, as by Milankovitch forcing, will lead to large mean temperature changes. However, in other cases, a large forcing still appears to be necessary. Well, what if, as Roy Spencer has been suggesting, internal variability leads to small changes in cloud cover? Even a small change would be a big change in the radiation budget.

      • Raven
        Posted Mar 31, 2009 at 9:41 AM | Permalink

        Re: Andrew (#211)
        The problem is Linzden and Spencer contradict each other. If Linzden is right warming caused by random changes in clouds would quickly be suppressed by the negative feedback. Spencer’s mechanism requires that cloud feedback be neutral or slightly positive.

        That said, I am not certain. The net response is still position in the case of Linzden so a change in cloud forcing could still cause net warming. A lot depends on how the random walk math works out.

        • Andrew
          Posted Mar 31, 2009 at 9:56 AM | Permalink

          Re: Raven (#212), Spencer would strongly disagree with you there. He believes cloud feedback is negative. Spencer mechanism is that essentially random cloud changes, which are not to be interpreted as feedback, cause large forcings which cause climate changes, even in the presence of negative cloud feedback-indeed, the cloud forcing changes will look like positive feedback but that gets cause and effect backwards.

        • Raven
          Posted Mar 31, 2009 at 10:17 AM | Permalink

          Re: Andrew (#213)
          Spencer is saying that the clouds are changing independently of the CO2 and are not strongly affected by the warming/cooling. His models suggest that the recent warming could be entirely explained by clouds. He is arguing that CO2 sensitivity is over estimated because co-indicidental cloud changes are being attributed to CO2 feedback.

          Linzden is saying that clouds respond strongly to warming/cooling in way that counter acts the warming/cooling. He is arguing that CO2 sensitivity because cloud feedback is strongly negative.

          They are different arguments. Perhaps Linzdens data analysis can be re-interpreted in a way that supports Spencer. As it stands today they are contradicting each other.

        • Andrew
          Posted Mar 31, 2009 at 10:23 AM | Permalink

          Re: Raven (#214), Spencer’s view is that the response of clouds acting against warming is not visible long term because it is masked by clouds acting to cause warming. See:

          http://www.drroyspencer.com/research-articles/satellite-and-climate-model-evidence/

          He clearly states that the “real” feedback is negative.

        • Raven
          Posted Mar 31, 2009 at 10:30 AM | Permalink

          Re: Andrew (#215)
          In other words, Spencer is treating them as different arguments but saying both effects are active at once.

        • Andrew
          Posted Mar 31, 2009 at 11:41 AM | Permalink

          Re: Raven (#216), Yes, exactly.

  78. Alan Wilkinson
    Posted Mar 31, 2009 at 1:30 AM | Permalink

    Raven, wouldn’t radiation from the tropics outweigh everything else given the fourth power relationship of radiation to temperature?

    And since the temperature has been flat since 2000 you would expect that latest data would not provide an equivalent test to that covering the decades when temperature was increasing?

  79. cba
    Posted Mar 31, 2009 at 6:36 AM | Permalink

    #207

    1. Not sure but it could be that all necessary data isn’t available. Perhaps too it really isn’t that significant or unique
    2. tropics is where most incoming solar is. However, with a cloud formation negative feedback, then why would not it be present to a lesser extent further out? And why should there be a positive feedback elsewhere?
    3. The feedback is not to co2 forcing but to temperature change. And yes, huge forcings are behind climate change and just because they were ignored doesn’t mean they don’t exist. Albedo has the same impact as solar forcing and it has been assumed rather constant. When measured over 20 years, it proved to vary by 10%, equivalent to multiple co2 doublings worth of forcing. It’s a monkey wrench tossed into the mechanism. That means it’s an uncertainty far greater than any variations that were attempted to be used as a determination of climate sensitivity.

  80. Andrew
    Posted Mar 31, 2009 at 12:31 PM | Permalink

    We have all heard the claim of “accelerating warming”-so why is the rate of warming from 1978 to 2008 almost exactly the same as the rate from 1911 to 1941?

    • Raven
      Posted Mar 31, 2009 at 3:06 PM | Permalink

      Re: Andrew (#219)
      You are comparing apples to oranges since the first period is volcano free and the latter had 2 major eruptions. That said, I find it interesting the slopes are identical despite the fact that periods are very different in terms of the forcings applied.

      • Andrew
        Posted Mar 31, 2009 at 3:15 PM | Permalink

        Re: Raven (#222), The response and subsequent recovery from the effects of volcanoes is pretty rapid. As long as neither begins or ends during a volcanic eruption (should end at about 5 years later-Pinatubo was 18, so we should be okay. If anything, El Chicon may exaggerate the later trend). The thing that should be pointed out is that I didn’t find the earlier trend by chance but through careful cherry picking of endpoints-no better than cherry picking end points to “prove” the contrary though.

      • Bill Illis
        Posted Mar 31, 2009 at 7:35 PM | Permalink

        Re: Raven (#222),

        You are comparing apples to oranges since the first period is volcano free and the latter had 2 major eruptions. That said, I find it interesting the slopes are identical despite the fact that periods are very different in terms of the forcings applied.

        There was a major eruption in the first time period as well (Novarupta in June 1912 which was a VEI6, the same as Pinatubo) and the El Chichon volcano in 1982 did not have any real impact on temperatures (temps even spiked 0.5C in the months after El Chichon which is as high as the temperature record changes on such short timescales over the record).

        So the two time periods contain the same volcanic impacts.

        What is the same in the two different time periods is this – the AMO.

        • Andrew
          Posted Mar 31, 2009 at 7:52 PM | Permalink

          Re: Bill Illis (#229), Two points-El Chicon ~did~ lower temperatures relative to where they would have been given the El Nino at that time. It actually does look like you’re about right about those volcanoes though:

        • Bill Illis
          Posted Mar 31, 2009 at 8:26 PM | Permalink

          Re: Andrew (#231),

          You’d have to assume that the El Nino of 1982-83 would have increased temperatures by 0.8C for that to be true – that would be triple the impact El Ninos are estimated to have.

          I looked very closely at all these volcanoes and the only one which seems to meet expectations was Pinatubo.

          I think the optical thickness data (measured in the newest volcanoes but “estimated” by Hansen and Sato in the older ones) vastly over-estimate the impact of volcanoes – I think the optical thickness data is distorted by the impact volcanoes have on the stratosphere where it is clear volcanoes have a very large impact but they just don’t have much impact on surface temperatures – that and Hansen needed big negative impacts from volcanoes to make his models work.

          I could show you a ten year temperature history without the dates on it and you could not pick out the dates of volcanoes.

        • Andrew
          Posted Mar 31, 2009 at 8:38 PM | Permalink

          Re: Bill Illis (#232), See:

          Its quite clear that Chicon suppressed the temperatures compared to where the El Nino said they should have been.

      • Hemst 101
        Posted Mar 31, 2009 at 9:27 PM | Permalink

        Re: Raven (#222),

        What were those different forcings? The volcanoes? Bill and Andrew seem to disagree. The sun? Then you will butt heads with Leif. The fact that the ~1912/45
        and ~1975/1998 increases in temperature are statistically indistinguishable (Micheals -Climate of Extremes – ~page 12) has to be adequately explained. To date I haven’t seen a convincing explanation. (Michaels thinks it’s the sun by the way.

        • Ron Cram
          Posted Mar 31, 2009 at 9:39 PM | Permalink

          Re: Hemst 101 (#236),

          You are correct that the two separate global temp increases are statistically indistinguishable using HadCRUT3. Some have argued that atmospheric CO2 has an algorithmic warming effect so that the first parts per million of CO2 contributed more warming than later CO2. But I doubt that explanation holds water mathematically, but I am not the one to defend or critique the viewpoint. I think it is strong evidence for greater natural climate variability than most climate scientists would like to admit, no matter the cause. Michaels could be correct that it is the Sun or Spencer could be correct that it is PDO not creating clouds during its warm phase. Or it could be a combination of those two or something we have not even considered.

        • Raven
          Posted Mar 31, 2009 at 10:27 PM | Permalink

          Re: Hemst 101 (#236)
          The current alarmist thinking appears to blame the MWP on volcanos, as you can see from these model runs.

          The argument is volcanos have a long term effects on the climate in addition to the short term response which fades away quickly.

          I did check the volcano history and I found this.

          It does appear that volcano density in the 1100s was less that then 1600s but I wonder how accurate the historical numbers.

          The attributions from AR4 seem to suggest that the El Chichon and Pinatubo volcanos would have caused slight cooling over the last 30 years.

  81. Alexander Harvey
    Posted Mar 31, 2009 at 3:41 PM | Permalink

    To all,

    Regarding the way that heat is believed to diffuse down through the ocean.

    There is an outline of the nature and and a description of Walter Munk’s insight here:

    Introduction to Physical Oceanography

    FWIW simple models based on the diffusion approach when forced by SST records correlate well with the OHC as depicted here:Global Ocean Heat Content 1955-2008 in light of recently revealed instrumentation problems, for a lower values of vertical diffusivity than is commonly assumed.

    If the effective diffusivity is low, as suggested by the OHC record, then the amount of “unrealized warming” (as defined by Hansen) is lower than he suggested in his 2005 paper which relied on 1993-2003 period which may be seen as atypical in the context of the 1955-2008 record.

    Re my 138:

    Has anyone checked the Urban & Keller paper quoted by Pielke, to ascertain where they implied his:

    By “unrealized warming in the pipeline”, they mean heat that is being stored within the ocean, which can subsequently be released into the ocean atmosphere.

    He can have his point of view, but he did seem to be putting words into Urban’s & Keller’s mouths. Perhaps he has had some other conversation with them where they explained what they meant to say. They do not seem to support his inference in their paper.

    Alex

  82. Bill Illis
    Posted Mar 31, 2009 at 9:23 PM | Permalink

    The best estimate of the ENSO’s impact on global temperatures is:

    [0.08 * Nino 3.4 Anomaly (three months later) = Global Temp Anomaly C]

    The El Chichon eruptions ocurred in March and April of 1982. The Nino 3.4 anomaly started out at neutral at this point and started rising immediately after reaching an anomaly of +2.85C by January of 1983.

    The satellite temps increased from -0.2C at the time of the eruption to +0.1C by April 1983, a little higher than expected with the ENSO formula.

    I’m not sure where the chart above got a +0.7C impact from for the 1982-83 El Nino. The formula would require an El Nino at +8.75C anomaly.

  83. bender
    Posted Mar 31, 2009 at 9:31 PM | Permalink

    This is what I thought when I first read Pielke’s take: he bungled the meaning of “committed/unrealized warming” “in the pipe” as proposed by Hansen. It’s not strictly about OHC. There is no physical “pipe”. It’s about temporary radiative imbalance as the equilibrium state is in the process of being achieved. This is, I think, lucia’s take as well. This seems to be (part of) what Alexander Harvey is arguing. I was surprised at Pielke’s apparent misconception, but there it is. This was in response to an article written by Ramanathan.
    .
    It sure would help if IPCC were not so ambiguous about such basic things.

    • Ron Cram
      Posted Mar 31, 2009 at 9:40 PM | Permalink

      Re: bender (#237),

      bender, how then do you explain Hansen’s plain words that I quoted in comment #235?

      • bender
        Posted Mar 31, 2009 at 9:42 PM | Permalink

        Re: Ron Cram (#239), #240 was a crosspost.

      • bender
        Posted Mar 31, 2009 at 9:49 PM | Permalink

        Re: Ron Cram (#239),
        Perhaps there are two “imbalances” – the radiative imbalance and a temperature imbalance. Look carefully at the quote. Hansen here seems to be referring to a temperature imbalance: the heat’s stored in the oceans, not the atmosphere. This is not the same thing as a radiative imbalance, which refers to energy absorbed but then potentially lost to space.
        .
        I think it is dangerous to try to overinterpret one man’s words. I would prefer to focus on IPCC rather than Hansen.

        • Ron Cram
          Posted Mar 31, 2009 at 9:57 PM | Permalink

          Re: bender (#243),

          According to Hansen and Pielke, a radiative imbalance creates a temperature imbalance. I do not see how the IPCC can disagree with that. I am focusing on Hansen because Hansen is the author Alexander put forward for his rather unorthodox view of how a radiative imbalance would effect the ocean temperature. As we can see, Hansen disagrees with Alexander.

          Give me a quote and a link where the IPCC says more energy coming in than going out would lead to cooler oceans and I will get all of their funding pulled. It is a ridiculous concept.

        • Neil Fisher
          Posted Apr 2, 2009 at 6:25 PM | Permalink

          Re: bender (#243),

          Perhaps there are two “imbalances” – the radiative imbalance and a temperature imbalance. Look carefully at the quote. Hansen here seems to be referring to a temperature imbalance: the heat’s stored in the oceans, not the atmosphere. This is not the same thing as a radiative imbalance, which refers to energy absorbed but then potentially lost to space.

          OK, I think I am beginning to understand your point, but there are still a few things that I am not quite “up” on. If I read it right, you are saying that a radiative imbalance is created, which leads to a temperature imbalance, in that the new equilibrium temperature (that is, the surface temperature required to re-establish radiative balance) has not yet been reached. That’s fine. The part I still don’t understand is how, if the heat gained from the radiative imbalance is subsequently lost to space, this does not create an imbalance in the opposite direction from the first (although they may be seperated in time). This would seem to me to indicate – if this is indeed where the heat has gone – that our understanding of the radiative transfer process is badly specified; that we have “missed” something. Yes?

        • bender
          Posted Apr 2, 2009 at 9:25 PM | Permalink

          Re: Neil Fisher (#294),
          I refuse to theorize. I’m just trying to understand what Hansen thinks and says. My point – and I’ve already shown it’s something Hansen has referred to – is that there is more to “imbalance” than just oceans and atmosphere. Space is part of the equation. That’s not me talking; it’s Hansen. Ron Cram and Bill Illis are oversimplifying what Hansen says by picking on selected quotes and ignoring others. That’s what Pielke seems to have done as well, and Alexander Harvey picked up on it.
          .
          When you open the oven door on that turkey as the oven is heating, you lose some heat. But it’s temporary. That heat will be realized when the system equilibrates, as it must. One could say that unrealized warming is “in the pipe”. It’s not right there in the oven. It’s outside the system, waiting to be brought in.

        • DaveR
          Posted Apr 3, 2009 at 3:22 AM | Permalink

          Re: bender (#300)

          When you open the oven door on that turkey as the oven is heating, you lose some heat. But it’s temporary. That heat will be realized when the system equilibrates, as it must. One could say that unrealized warming is “in the pipe”. It’s not right there in the oven. It’s outside the system, waiting to be brought in.

          Except when you’re hypothesising that more heat is being lost to space, what is the mechanism by which you “shut the oven door” to allow extra warming to be “in the pipe”? I would consider the situation more akin to a defective oven. As more heat is added, the oven door progressively buckles, allowing more rapid heat loss, so a balance is maintained below what would otherwise be attained in a fully functional oven. Unless you “fix” the oven door, there is no heat “in the pipe”.

        • Neil Fisher
          Posted Apr 4, 2009 at 5:08 PM | Permalink

          Re: DaveR (#302),

          what is the mechanism by which you “shut the oven door” to allow extra warming to be “in the pipe”?

          Or perhaps more to the point, we don’t know what caused the door to open, we don’t know how much it’s been opened, we don’t know if it’s still in the process of opening or is in the process of closing, and we don’t know when or even if it will close all by itself. Perhaps the hot air escaping will melt the seal on the door and it will continue to “leak” heat, or perhaps the melted seal will fuse the door shut! And I’m beginning to suspect that we don’t even know how much the turkey weighs and can’t remember whether we stuffed it or not…

          Well, that’s probably about as far as that analogy can go. ;-)

        • Ron Cram
          Posted Apr 3, 2009 at 7:16 AM | Permalink

          Re: bender (#300),

          You still have not provided a quote from Hansen saying what you claim he says about space. Or, at least I have not seen it. I still think you are misunderstanding what Alexander has been saying. Alexander agrees any radiative imbalance will show up in the oceans because other storage units are almost trivial.

  84. Bill Illis
    Posted Mar 31, 2009 at 10:06 PM | Permalink

    The “warming in the pipeline”, the “unrealized warming” explanation is best explained by Lucia’s example of a turkey cooking in the oven.

    You set the oven temperature to 400F. While the turkey is cooking (while the GHGs are trapping additional IR energy that would have escaped to space – the GHG forcing), the temperature in the oven is only 395F because the cold turkey (the cold oceans) are absorbing some of the energy.

    When the turkey (when the oceans) reach 395F (when the oceans reach equilibrium), the temperature in the oven finally starts rising since the turkey (the oceans) is no longer absorbing the heat energy (the GHG forcing) from the oven.

    The oven temperature then rises to 400F – the same temperature as the heating element (the GHG forcing) is providing.

    While the 400F turkey is completely burnt (the oceans will not be) and the surface temperatures then rises to the level that the GHG forcing says it should be since the ocean are no longer absorbing the energy from the surface.

    But the ocean absorbing heat energy (in the form of photons) from the surface has two time elements to it. The surface ocean temperatures reach equilibrium within just a few years. The deep ocean (at 3C to 0C) takes hundreds or even thousands of years to reach equilbrium with the surface.

    The warming in the pipeline is just a delay caused by the cold oceans between when the surface temperatures catch up to the forcing that the GHGs provide.

    • Ron Cram
      Posted Mar 31, 2009 at 10:11 PM | Permalink

      Re: Bill Illis (#246),

      Really? Lucia said that? So sea surface temps are going to reach equilibrium with air temperature? When exactly is this going to happen?

      • Bill Illis
        Posted Mar 31, 2009 at 10:24 PM | Permalink

        Re: Ron Cram (#247),

        Lucia didn’t say exactly what I wrote but she did start with the turkey in the oven analogy and I just extended it a little and it is the best analogy in my mind to what Hansen really means by warming in the pipeline.

        You could extend the turkey analogy to the glacier and ice feedback by adding a 6 inch ice block to the bottom of the oven and then you are getting closer to what Hansen is promoting recently.

        The 400F oven (the doubled CO2 forcing) cooks the turkey and melts the ice block (melts Greenland and most of Antarctica) and the temperature in the oven eventually reaches 410F (or it only starts at 390F (0.6C global temperature anomaly) while the turkey is cooking and the iceblock is melting and then it eventually reaches 400F after the ice has melted and the turkey is burnt (and global temperatures have increased +6.0C)).

        • Ron Cram
          Posted Mar 31, 2009 at 10:32 PM | Permalink

          Re: Bill Illis (#249),

          The fact heat from the oven is being stored in the turkey is understandable and analogous to ocean warming but after that I am not sure of the value of the analogy. I don’t know exactly what lucia said but unless her concept allows for OHC to be used as a direct proxy for radiative imbalance, then she does not exactly agree with Hansen or Pielke.

          Where the analogy breaks down I think is the fact the oceans can give back some of the OHC to the atmosphere. That really is the “heat in the pipeline.”

        • Ron Cram
          Posted Mar 31, 2009 at 10:50 PM | Permalink

          Re: Bill Illis (#249),

          lucia’s analogy did get me thinking. Kind of a crazy idea. I wonder if it is possible to calculate the number of joules in a turkey before putting it in the oven and then calculate the number of joules about half way through cooking it and from that information estimate the temperature in the oven over the time period of cooking? That appears to be what Hansen is doing in calculating the radiative imbalance of the earth. What do you think?

      • bender
        Posted Apr 1, 2009 at 2:25 AM | Permalink

        Re: Ron Cram (#247),
        I think you’re better off discussing this directly with lucia at her Blackboard than at CA on “unthreaded”.
        .
        I think Hansen’s logic is as follows:
        If the oceans are heating beyond what the models predict then GHG theory is proved. If they are not heating it’s because the heat that should be there has been lost to space, an example of radiative imbalance = committed warming in the pipe..
        .
        There are obvious empirical dangers in that style of argument, but it’s not logically flawed. Yes, there is an appearannce of Hansen arguing both ways at different times to suit his one purpose – but that does not mean the argument taken as a whole is flawed. My advice is to discuss it with lucia.
        .
        I go back to Bill Illis’s #65:

        The only thing they had left was the oceans absorbing a large amount of heat from the surface

        Wrong. There is still the idea of a dynamic imbalance caused by loss of heat to space. Thisis not about Hansen’s text. It’s about “them”.

        • Raven
          Posted Apr 1, 2009 at 3:04 AM | Permalink

          Re: bender (#259)
          If the dynamic imbalance exsists then there must be a continuous net energy flow into the system. One can test for the existence of this imbalance by measuring the *rate* at which energy accumulates. If energy is not accumulating then there can be no dynamic imbalance. If the OHC is not increasing (after accounting for uncertainties and noise) then one can conclude the dynamic imbalance does not exist.

          This is my understanding of Pielke Sr’s argument and why I don’t think he is wrong even if the pipeline stuff makes no sense.

          For that reason this makes no sense:

          There is still the idea of a dynamic imbalance caused by loss of heat to space.

          If the heat is not entering the system then there is no imbalance unless one understands the reason for the heat loss. For example, one could argue that the coulds temporarily shielded the system but I see this as a rediculous semantic argument because one could also argue that the clouds are providing negative feedback in response to warming. Go luck proving that one way or another.

        • DaveR
          Posted Apr 1, 2009 at 3:35 AM | Permalink

          Re: bender (#259), That makes no sense to me whatsoever (as an outsider looking in, but who is very familiar with equilibrium systems). I thought the whole point of there being an imbalance was that CO2 was producing “excess” heat. If the heat isn’t in the atmosphere, it has possibly been “stored” in the oceans (which it doesn’t appear to have been). If this “excess” heat is being lost to space, then why is that an imbalance? Surely it’s a “return to equilibrium”? Heat lost to space has gone, it’s not coming back, so how can it be “warming in the pipe”? *totally confused* :/

        • bender
          Posted Apr 3, 2009 at 7:56 AM | Permalink

          Re: DaveR (#263),
          You are confused by your own presumptions. You assume that heat that has been convected off to space is a continuous, reliable process and that the amount lost will always balance the amount gained from GHGs. Whereas I have made no such assumption. If the amount lost to space ON AVERAGE is less than the amount gained then you have a persistent imbalance. Futhermore, if the amount lost in any one instance is greater than that gained over the same time then you have a temporary imbalance = “in the pipe”.

          Re: Ron Cram (#305),

          You still have not provided a quote from Hansen saying what you claim he says about space.

          I thought I had, but looking over the thread, it seems you’re right. I haven’t. I ran across it in my readings. I think it was not in a published paper. But I suspect any statement of that nature should be in the published Hansen literature. Will get back to you.

        • DeWitt Payne
          Posted Apr 3, 2009 at 8:45 PM | Permalink

          Re: bender (#306),

          You get the same effect as a decrease in long wave optical depth (increase in radiative heat loss) if albedo increases and the energy gets reflected away. However, we don’t know why the albedo or the long wave optical depth changes, or even if that is the cause of the current apparent radiative balance. But is anything like this being modeled or even investigated?

  85. Bill Illis
    Posted Mar 31, 2009 at 10:11 PM | Permalink

    And I guess I should add that the oceans are not absorbing any forcing over the last 5 years. The Ocean Heat Content down to 900M (which is already getting into the deep ocean territory) has not increased at all in the last 5 years so the turkey is already burnt and Hansen is still trying to cook it (whatever books are still available to cook that is).

  86. Bill Illis
    Posted Apr 1, 2009 at 6:45 AM | Permalink

    If the oceans and the ice sheets are not absorbing the increased GHG forcing currently, then there is no radiative imbalance.

    Temps have increased +0.6C to date and there is no warming left to come from the current level of GHGs.

    At the surface and in the atmosphere, the greenhouse effect operates at the speed of light and at the speed of quantum physics. It is only a delay measured in “hours” of how long a photon from the Sun stays around on Earth, skipping from molecule to molecule, before it is lost to space.

    The temperature drops overnight by an average 10C. In just 12 hours, one third of the greenhouse effect is lost.

    If the Sun stopped working for two days, the Earth’s average temperature would fall below -18C by the end of the second night and the entire greenhouse effect would be gone.

    If the Sun stopped working for three days, rivers and lakes would freeze to the bottom and polar ice would start freezing down several metres. Equatorial oceans would cool off slower and provide some relief to land next to the oceans, and winds would still try redistribute the remaining heat around the planet, but the centre of the continents, even at the equator, would be frozen solid in three days.

    I haven’t seen the math completely worked out, but the greenhouse effect is just a delay of 36 hours in the average time it takes for a photon from the Sun to enter the Earth system before it eventually escapes into space.

    So, if the oceans are not absorbing any heat from the GHG forcing right now, then we are already at radiative balance. All the increased temperature which can be provided by the increased GHG forcing has already occurred in the last 36 hours.

  87. stephen richards
    Posted Apr 1, 2009 at 9:56 AM | Permalink

    I think Roy Spencer is getting in the April fols mood.

    http://www.drroyspencer.com/

    Gore announces his retirement from Climate Change activities

  88. Andrew
    Posted Apr 1, 2009 at 2:02 PM | Permalink

    There is an interesting paper here:

    http://www.atypon-link.com/IAHS/doi/pdf/10.1623/hysj.54.2.394

    And some Consensus supporter’s “rebuttal” here:

    http://www.atypon-link.com/IAHS/doi/pdf/10.1623/hysj.54.2.406

  89. Posted Apr 1, 2009 at 7:46 PM | Permalink

    Gavin has presumed to lecture about advocacy on RC. He’s currently bashing WUWT for a recent guest post. Since I can’t even post small questions over there I took the opportunity to reply.

    http://noconsensus.wordpress.com/2009/04/02/ten-replies-to-gavin-advocacy-vs-science/

  90. AJ
    Posted Apr 2, 2009 at 7:09 PM | Permalink

    I posted this OT on WUWT. Thought it might be better suited here.

    Regarding:

    http://www.nationalpost.com/news/story.html?id=1453831&p=2

    Using the reporting of NOAA’s study, I’ve done a back of the envelope estimate that the climate sensitivity to a doubling of CO2 is ~1.65C.

    Here’s the relevant bits from the National Post story:

    Some of the changes in North America’s warming trend of the past half-century have been due to shifting ocean currents, the NOAA team found. It estimates the “natural” change is substantial and could be close to half of all warming in North America (though it is still less than the amount caused by greenhouse gases.)

    The study found:

    – The 56-year trend of annual surface temperature showed a rise of 0.9C, plus or minus one-tenth of a degree.

    – Seven of the warmest 10 years since 1951 occurred in the decade from 1997 to 2006. The data in the study cover only to the end of 2007.

    So a bit more than half of the 0.9C increase is due to GW. I’ll use 0.9C * 55% = ~0.5

    I’ll also use the Mauna Loa CO2 measurements for the study start and end dates:

    1951: 311 ppm
    2007: 383 ppm

    If I understand it correctly, the change in temp can be calculated as:

    tempchange = forcefactor * ln(CO2[end]/CO2[start])
    so:
    forcefactor = tempchange / ln(CO2[end]/CO2[start])

    Plugging in the above values we get:

    forcefactor = 0.5 / ln(383/311)
    forcefactor = 2.4

    And sensitivity to doubling of CO2 as:

    sensitivity = forcefactor * ln(2)
    sensitivity = 1.65C

    This is significantly less than the 3.0C estimated by the model ensemble used by the IPCC.

    Also interesting, this is consistent with Bill Illis’s analysis from a while back:

    http://wattsupwiththat.com/2008/11/25/adjusting-temperatures-for-the-enso-and-the-amo/

    AJ

    • Andrew
      Posted Apr 2, 2009 at 8:00 PM | Permalink

      Re: AJ (#295), The claim that the natural warming is “less than that from Greenhouse Gases” is baseless. They don’t know that. Its pure speculation. That aside, your calculations have many short comings, some of which would serve to lower your estimate, and others raise it. As interesting as it is, it ain’t a smoking gun. Cheers.

  91. Bill Illis
    Posted Apr 2, 2009 at 7:56 PM | Permalink

    To AJ,

    Good stuff, we are on the same wavelength here.

    Note that if the oceans are no longer absorbing any increased GHG forcing right now, then these calculations ARE the actual impact from CO2/GHGs.

    The numbers would indicate there is no substantial water vapour feedback (the different temp series produce results which indicate there is slight negative feedback to a slight positive feedback).

    Taking these results with the actual water vapour numbers we do have, the balance of the empirical evidence shows there is NO water vapour feedback at all. It is a Constant entity.

    So we are back to the basic physics calculations which provide for a 1.2C per doubling impact.

    My newest numbers are 1.62C using the Hadcrut3 temp series to 0.725C using the RSS temp series.

    Aerosols are the biggest uncertainty remaining. Are they really reducing global temps right now so that the temp-to-date calculations are lower than they really should be (let’s assume we clean up the air, do temps then go up?).

    Well, the areas which should be the most affected by Aerosols (Asia, southern California etc.) have had the biggest temperature increase of any region. Europe’s temps should now be rising rapidly given there is reduced Aerosols there now (but Europe’s temps are stable or falling). The empirical evidence says that the sign is opposite for Aerosols – China didn’t rise 1.8C in the last 20 years because Aerosols reduced its temperatures.

    So, we are back to 1.2C per doubling once again – and we have already seen half (0.6C) of this warming already.

  92. Bill Illis
    Posted Apr 2, 2009 at 8:25 PM | Permalink

    To Andrew,

    I have exactly emulated GISS Model E’s GHG forcing using a similar formula to AJs and the same process emulates the IPCC’s mean predictions as well. It is just a LN (CO2/GHGs) calculation.

    • Andrew
      Posted Apr 2, 2009 at 9:46 PM | Permalink

      Re: Bill Illis (#299), My beef isn’t with the math itself but some of the assumptions. They probably over all don’t impact the result to much, but I’m a pedant SOB.

  93. AJ
    Posted Apr 3, 2009 at 7:08 AM | Permalink

    Just a few points about my sensitivity analysis:

    – I haven’t read the actual NOAA study. It’s based solely off the news article.
    – I couldn’t determine if the reported trend was for North America only or global.
    – The article did not mention which temperature record was used, but being a NOAA study, I can guess.

    I simply was interested in seeing if this simple sensitivity calculation was consistent with Bill’s. Assuming a global trend and GISSTemp, then it is. Kudo’s to Bill… NOAA agrees with you ;> Maybe you should do a write up on the study? I wonder if you were cited?

    As for what’s included in the forcing factor, I suspect it is both CO2 and UHI. My personal hunch is that UHI also has a logarithmic pattern over time. For example, if a temperature sensor was well sited on a wide open grassy area and then a 100 foot radius circle of asphalt was paved underneath, you would expect to see a significant increase in measured temperature. If we then doubled the area of that paved circle by adding 41 feet to the radius (sqrt(2)=1.41), I wouldn’t expect the impact to be as great.

    As for whether the satellite temp records are accurate, I believe they are. Both UAH and RSS are both very highly correlated with GISSTemp for the USA48 over the period 1980-2008. The trend lines are also very closely matched. Despite GISSTemp’s issues, I believe they do a good job in the USA48. IMO, there’s enough high quality data to make the necessary adjustments. Given the close match, I conclude that the satellites are at least good over land at the mid-latitudes. The question is, who are you going to trust over high latitudes, the tropics, and over the oceans? This is where the surface record is most spotty, so my vote is with the satellites.

    Being a lukewarmer by faith, I believe in neutral feedbacks and a 1.2C sensitivity. If forced to go either higher or lower, however, I would go lower.

    And that’s all I’ve got to say about that.

    Have fun!

    AJ

  94. Andrew
    Posted Apr 4, 2009 at 12:46 AM | Permalink

    Bender, get ready to get out your foam finger (:D) to wag at Mann and pals-they have something on the North Atlantic Eigenthingy:

    http://www.newscientist.com/article/dn16892-natural-mechanism-for-medieval-warming-discovered.html

    Ah adventures in alarm land. What fun!

    • bender
      Posted Apr 4, 2009 at 8:54 AM | Permalink

      Re: Andrew (#308),
      What do you mean? They prove my point: short-term pattern-based eigenthingies break down if you look at them over a long enough time frame. “Not inconsistent” with spatiotemporal chaos, but inconsistent with the GCMs. This is what I’ve been telling you from day one.

      • Andrew
        Posted Apr 4, 2009 at 11:33 AM | Permalink

        Re: bender (#309), The credit MWP to NAO. I seem to recall you having a problem with such arguments…

        • bender
          Posted Apr 4, 2009 at 6:39 PM | Permalink

          Re: Andrew (#310),
          Reread what I wrote. You’re missing the point. Your precious AM”O” ain’t oscillating if tends to get stuck in one configuation for hundreds of years at a time. Ergo my point about eigenthingies vs “oscillations”. And what caused it to switch? The flap of a butterfly’s wing? That’s some robust oscillator.

        • Andrew
          Posted Apr 4, 2009 at 6:56 PM | Permalink

          Re: bender (#313), Let’s move for a second away for a second from our old argument-I’ve “moved on” and now agree with you. My comment’s snark was probably the wrong approach because I confused the point I wanted you to note, which was that, as I recall, you liked to harp on about how it was wrong of some skeptics to try and attribute changes to eigenthingies. You are probably (definitely?) right on that point. But now Mann and pals want to attribute MWP to an eigenthingy. Not the AM”O” either but the NA”O”.

        • bender
          Posted Apr 5, 2009 at 5:38 PM | Permalink

          Re: Andrew (#314),

          now Mann and pals want to attribute MWP to an eigenthingy

          But there is nothing logically wrong with doing that. So I really don’t understand your criticism.

        • bender
          Posted Apr 5, 2009 at 5:46 PM | Permalink

          Re: Andrew (#314),
          Also, there is nothing new here. Mann has been suggesting MWP was caused by a MCA for a long time now. This paper simply attempts to better characterize that MCA, by linking it to a present-day eigenthingy that we recognize as NAO.
          .
          Also, you seem to want to split hairs over AMO vs NAO. They are just limbs of the NAM. I don’t seem them as causally distinct. What distinguishes them is arbitrary human choices over the spatial and temporal extents and parameters used to define eighenthingies and subthingies and so on. Your faith in the global circulation pattern is remarkable.

        • Andrew
          Posted Apr 5, 2009 at 6:13 PM | Permalink

          Re: bender (#317), You don’t remember excoriating skeptics over arguments like this at all? How is that different from what Mann and pals are claiming?

          I am splitting hairs, yes, and your point about there being essentially no difference but arbitrary things said by human beings is well taken. I have no “faith in the global circulation pattern”. If you could stop trying to make this about things you think I’ve said in the past, well, we might get somewhere. Okay, how about this, I will say this one last time, now and forever, and I don’t want you acting like I haven’t ever done this again. I Recant. Happy? Now can you excoriate the Team for making the same error you seem to want to pin exclusively on me?

        • bender
          Posted Apr 5, 2009 at 8:51 PM | Permalink

          Re: Andrew (#318),
          What error are they making? You still have not explained the basis of your complaint.

        • Andrew
          Posted Apr 5, 2009 at 9:09 PM | Permalink

          Re: bender (#319), They attribute a climate change to an eigenthingy-just as Spencer tries to do. Are they both wrong, or is there some difference between their arguments?

  95. Posted Apr 4, 2009 at 3:26 PM | Permalink

    One instrumental temperature record is available that is unaffected by CO2 or other climatic changes; it is apparently fairly close to a pure measurement of direct and indirect insolation.

    It’s on the Moon. I don’t know where the data is, but this research paper is by someone involved in that data’s storage, and the document states that there has been a warming trend in sunlight reflected from the Earth since the 1970s.

    He attributes this to “global dimming” but this strikes me as one of multiple possible interpretations.

    http://www.geo.lsa.umich.edu/~shaopeng/ResStatement.pdf

    I wonder what the data itself shows.

    ===|==============/ Level Head

  96. Posted Apr 5, 2009 at 8:26 AM | Permalink

    Didn’t the Team just criticize that Cato petition for cherry-picking information to support their contention that AGW is not a looming disaster – from scientific papers that ultimately did not support Cato’s stated conclusion?

    I’m just asking.

  97. Mark T
    Posted Apr 6, 2009 at 10:15 PM | Permalink

    Sorry, bender, but I’m to the point where all I have left for Team (et al.) methodology is sarcasm.

    Mark

    • bender
      Posted Apr 6, 2009 at 10:45 PM | Permalink

      Re: Mark T (#12),
      I understand. I was trying to pull a Kenneth Fritsch in his absence. Remarkable resemblance, wouldn’t you say?

  98. Mark T
    Posted Apr 6, 2009 at 11:26 PM | Permalink

    You did well. I was quite impressed.

    The real question I have is pretty simple, actually. If ANY naturally occurring pattern is capable of driving regional or global temperatures/climate to this level, how can we ever be sure NOW is not just another recurrence? This applies to everything, IMO. If the current downturn is due to natural variability, doesn’t that sort of shoot a hole in the foot of “natural variability cannot explain the past 100 years.”?

    Wouldn’t this be another case for your growing special pleading file, bender?

    Mark

    • Jonathan Schafer
      Posted Apr 7, 2009 at 7:01 AM | Permalink

      Re: Mark T (#14),

      I thought this applicable…

      In 1934, Sir Karl Popper published The Logic of Scientific Discovery, generally considered to be one of the two most significant works on the philosophy of science in the past hundred years. Here he is in Part 67 of that book:

      [S]ince probability estimates are not falsifiable; it must always be possible in this way to ‘explain’ by probability estimates, any regularity we please. Take, for example, the law of gravity. We may contrive hypothetical probability estimates to ‘explain’ this law in the following way. We select events of some kind to serve as elementary or atomic events; for instance the movement of a small particle. We then assume that these events show a chance-like distribution. Finally we calculate the probability that all the particles within a certain finite spatial region, and during a certain finite period of time – a certain ‘cosmic period’ – will with a specified accuracy move, accidentally, in the way required by the law of gravity. The probability calculated will, of course, be very small; negligibly small, in fact, but still not equal to zero. Thus we can raise the question of how long an n-segment of the sequence would have to be, or in other words, how long a duration must be assumed for the whole process, in order that we may expect with a probability close to 1 (or deviating from 1 by not more than an arbitrarily small value) the occurrence of one such cosmic period in which, as the result of an accumulation of accidents, our observations will agree with the law of gravity. For any value as close to 1 as we chose, we obtain a definite, though extremely large, finite number. We can then say: if we assume that the segment of this sequence has this very great length – or in other words that the ‘world’ lasts long enough – then our assumption of randomness entitles us to expect that the occurrence of a cosmic period in which the laws of gravity will seem to hold good, though ‘in reality’ nothing ever occurs but random scattering. This type of ‘explanation’ by means of the assumption of randomness is applicable to any regularity we choose. In fact we can in this way ‘explain’ our whole world, with all its observed regularities, as a phase in a random chaos – as an accumulation of purely accidental coincidences. [Italics in original]

      To me, this describes much of the agw scenario. The models, the proxies, etc, appear to be dealt with in such a way as to make any man-made global warming scenario possible, via cooling, warming, regional/global actions, teleconnections, etc. Anything and everything point to global warming and they use any and all of these probability estimates to explain agw.

    • bender
      Posted Apr 7, 2009 at 7:40 AM | Permalink

      Re: Mark T (#14),
      Normally I do not reply to such hand-waving skepticism.

      If ANY naturally occurring pattern is capable of driving regional or global temperatures/climate to this level, how can we ever be sure NOW is not just another recurrence?

      1. Circulation is by definition regional redistribution of haat. It can not cause global-scale warming.
      2. If you do not agree that the globe is warming, then please explain your evidence. Other than Antarctica, the globe was warming (up until 2001). (Hence the relevance of the Steig article.)

      If the current downturn is due to natural variability, doesn’t that sort of shoot a hole in the foot of “natural variability cannot explain the past 100 years.”?

      1. No one knows the source of the flatline (it’s not a downturn) – whether it’s natural, semi-natural, or anthropogenic in origin. (Some are arguing it’s poorly measured aerosol effects during the last decade.)
      2. Even “mike” at RC admits natural variability *could* explain the past 100 years. The consensus is that this is highly improbable, given the known facts about GHG physics and their visible effects when incorporated into computer models. (Hence Steve M’s request for an engineering-quality exposition, and my plea for a statistically robust estimate of the uncertainty on the GHG/AGW effect.)

      Wouldn’t this be another case for your growing special pleading file, bender?

      Apparently not.
      .
      There is such a thing as unhealthy skepticism. As I keep telling Andrew, “natural variability” is not an alternative theory for GW. It’s just a simple fact that must be contended with, can not be ignored, but is often ignored (or over-simplified) when it comes to statistical attribution exercises.

      • Posted Apr 7, 2009 at 8:53 AM | Permalink

        Re: bender (#325),

        1. Circulation is by definition regional redistribution of haat. It can not cause global-scale warming.

        Hand waving, but I don’t agree with your certanty “It can not…”.
        My feeling is that we don’t know, but it can.
        What effect could have an atmospheric circulation rearrangement, I don’t know, for istance on cloud cover? On eddy transport of heat? On land cover change? etc…Too chaotic and complex, don’t you think?

        Of course a rearrangement of heat in the ocean, no OHC change but with the surface involved, has surely an impact on global atmospheric heat content.

        • bender
          Posted Apr 7, 2009 at 9:32 AM | Permalink

          Re: Paolo M. (#326),
          I’ll agree that some global circulatory configurations might be more effective at dissipating heat than others, if that’s what you mean. But is there any proof of this?
          .
          (Perhaps this is what Tsonis has in mind with his theory (and I’m paraphrasing very liberally here) that the various modes can’t all be running “synchronously” in warm-phase at the same time – that something has to give as ocean surface waters most everywhere start warming? Either a change in circulation, or a convective expulsion of heat to the upper atmosphere (which is then radiated to space), some sort of Lindzenesque negative feedback effect (iris effect) that could account for the 2001-2009 flatline? Questions that come to the open mind.)

        • Posted Apr 7, 2009 at 9:54 AM | Permalink

          Re: bender (#327),

          But is there any proof of this?

          I don’t know for sure. But, surely, we need more model runs that are not forced to agree with AGW hyphotesis.
          Open minds and open models!

          By the way, as a first guess, if ocean surface starts to warm (or cool), I would think about albedo, then a second guess, a third one, etc…

        • Ron Cram
          Posted Apr 7, 2009 at 11:10 AM | Permalink

          Re: bender (#327),

          Are you completely unfamiliar with Spencer’s theory?

        • bender
          Posted Apr 7, 2009 at 11:49 AM | Permalink

          Re: Ron Cram (#332),
          In what peer-reviewed journal article or IPCC chapter was this “Spencer theory” published? Full citation, please.

        • Andrew
          Posted Apr 7, 2009 at 11:54 AM | Permalink

          Re: bender (#334), The work is not yet published, so, I am sorry to say, your request is not going to be satisfied. However, I expect that you know it isn’t published and that if and when it does get published, it won’t change your mind.

        • bender
          Posted Apr 7, 2009 at 12:03 PM | Permalink

          Re: Andrew (#335),

          The work is not yet published

          Oh? You don’t think he’s referring to his new hobby-horse: Spencer et al. 2007. Cloud and radiation budget changes associated with tropical intraseasonal oscillations GRL 34?

        • Andrew
          Posted Apr 7, 2009 at 12:10 PM | Permalink

          Re: bender (#336), No, because that paper has nothing to do with what he is talking about. Spencer and Braswell 2008 is slightly more related, but it still isn’t Spencer’s PDO idea (which I refuse to defend, just so you know). If you followed Spencer’s site, you would know…

        • bender
          Posted Apr 7, 2009 at 12:36 PM | Permalink

          Re: Andrew (#337),

          that paper has nothing to do with what he is talking about

          I’m not surprised. I don’t know what he’s talking about. Because he hasn’t told me yet.

          If you followed Spencer’s site, you would know…

          I’ve never heard of “Spencer’s site”. And I’m not much of a follower. More a reader of the primary literature (except where Steve M and lucia are concerned because they don’t publish). So now that I know you’re not talking about Spencer Weart, tell me in your own words: what’s “Spencer’s theory”?

        • Andrew
          Posted Apr 7, 2009 at 12:42 PM | Permalink

          Re: bender (#338), You thought we were speaking of Weart? :D ROFL. Er I mean, no. We we’re speaking of Roy Spencer. His website is, so you know, drroyspencer.com

          They specific theory was linked by me above from our “fight” about Mann and Pals…

        • bender
          Posted Apr 7, 2009 at 12:51 PM | Permalink

          Re: Andrew (#339),
          1. And the theory is … ?
          2. Theories are pretty easy to come by. I hope this one is supported by data.
          3. The existence of negative feedback in the moist convection would bid down GHG sensitivity, but it would not cut it to zero.
          4. That a feedback is expected to kick in in the future does not imply it was operating in the past, where circumstance might have been quite different.

          Looking forward to your essay with Ron’s co-authorship.

        • Andrew
          Posted Apr 7, 2009 at 1:46 PM | Permalink

          Re: bender (#344),

          1. And the theory is … ?

          http://www.drroyspencer.com/research-articles/global-warming-as-a-natural-response/

          2. Theories are pretty easy to come by. I hope this one is supported by data.

          Roy thinks so. I’m agnostic, and a little doubting.

          3. The existence of negative feedback in the moist convection would bid down GHG sensitivity, but it would not cut it to zero.

          Strawman. No one, including Spencer, argues that negative feedback would do this.

          4. That a feedback is expected to kick in in the future does not imply it was operating in the past, where circumstance might have been quite different.

          Why would the underlying processes change? I grant you there are slow feedbacks (outgassing, lag and what not) but why should existing feedbacks suddenly disappear? Kicking in at a future time is also something no one has argued.

          Looking forward to your essay with Ron’s co-authorship.

          Your derision aside…I won’t be authoring any “essay” to defend a theory I don’t have strong beliefs about.

        • bender
          Posted Apr 7, 2009 at 1:57 PM | Permalink

          Re: Andrew (#345),

          why should existing feedbacks suddenly disappear?

          Who said anything about “sudden disappearance”? You are familiar with the theory and fact of nonlinear dynamics?

      • Mark T
        Posted Apr 7, 2009 at 10:46 AM | Permalink

        Re: bender (#325),

        Normally I do not reply to such hand-waving skepticism.

        It wasn’t handwaving about anything, I asked some questions.

        1. Circulation is by definition regional redistribution of haat. It can not cause global-scale warming.

        Who said anything about circulation? Is there some solid evidence that these patterns have only internal causes? It seems to me that there is ample evidence that some of these cycles may have external causes. In other words, saying “it was the AMO” or “it was ENSO” still leaves open the question: what causes these (if anything, of course)?
        .
        And, btw, I recall that the 1998 did cause the global temperature to spike… why is this so if circulation patterns are not global (really, this isn’t a rhetorical question)?

        2. If you do not agree that the globe is warming, then please explain your evidence. Other than Antarctica, the globe was warming (up until 2001). (Hence the relevance of the Steig article.)

        When did I ever say anything contrary to a warming planet? Besides the strawman here, the equator has not (to my knowledge) shown any warming. Since the measure of “global warming” is really an integral of all the planet’s temperature, it is quite conceivable that it is not global, but constrained to certain regions which are warming, e.g., the arctic, and that dominates the integral. This is why, IMO, the whole single temperature metric is displaced, i.e., there is no “global climate,” since it varies radically over the planet (you cannot average two temperatures from different mass compositions: as an extreme, does it make sense to average the temperature of the air over the ocean with the ocean itself? Of course not).

        1. No one knows the source of the flatline (it’s not a downturn) – whether it’s natural, semi-natural, or anthropogenic in origin. (Some are arguing it’s poorly measured aerosol effects during the last decade.)

        By definition, since it went from rising to flat, it is a downturn.

        2. Even “mike” at RC admits natural variability *could* explain the past 100 years. The consensus is that this is highly improbable, given the known facts about GHG physics and their visible effects when incorporated into computer models. (Hence Steve M’s request for an engineering-quality exposition, and my plea for a statistically robust estimate of the uncertainty on the GHG/AGW effect.)

        He admits that grudgingly, and certainly we never hear any of this in the media. Your second part that the consensus thinks it is improbable is what I was getting at, i.e., they don’t think nature can overcome CO2 – yet here we are with repeated periods of no temperature rise and continually rising CO2. Ultimately, they simply don’t want to admit that nature has a larger role than previously thought (well, nature was to blame originally, until the environmental movement found the CO2 hobby horse). And, for the record, the known facts about GHG physics are well-known at a molecular level, but poorly-known in a complex dynamical system such as the earth.

        Mark

        • Mark T
          Posted Apr 7, 2009 at 11:03 AM | Permalink

          Re: Mark T (#329),

          why is this so if circulation patterns are not global (really, this isn’t a rhetorical question)?

          Should have read “…patterns do not cause global changes…” which makes a bit more sense. Really, what the question boils down to is whether or not the oscillatory patterns are the cause of the climate change, or just another effect from some other underlying cause. If the latter is true, than the assertion that these patterns cannot cause the global phenomena we are witnessing is only true in the sense that they are not the cause in the first place, but evidence of the true cause (as would be the climate change in general). Hehe, which came first, the chicken or the egg?

          Mark

      • Andrew
        Posted Apr 7, 2009 at 11:41 AM | Permalink

        Re: bender (#325),

        As I keep telling Andrew, “natural variability” is not an alternative theory for GW. It’s just a simple fact that must be contended with, can not be ignored, but is often ignored (or over-simplified) when it comes to statistical attribution exercises.

        Oh get off your damn hobby horse and leave me alone. I don’t need lecturing from you. The last part I completely agree with. Why are you determined to make me into a strawman to beat the sh*t out off? Leave me be!

  99. Mark T
    Posted Apr 7, 2009 at 10:55 AM | Permalink

    Ooops, since I cut the last bit above…

    Re: bender (#325),

    There is such a thing as unhealthy skepticism. As I keep telling Andrew, “natural variability” is not an alternative theory for GW. It’s just a simple fact that must be contended with, can not be ignored, but is often ignored (or over-simplified) when it comes to statistical attribution exercises.

    Whoa, here, bender. Take a step back on your assumptions of what I believe and don’t believe. I’ve NEVER implied that there is not some impact from anthropogenic forces and that natural variability is 100% the cause of our climate. I think natural variability is far too underestimated by the alarmist camp, and I think there is a good reason for that underestimation, but I won’t go into that here. Yes, it must be contended with and no, it is not being properly considered by the major players. I think more of the climate community does understand this than the media and near sighted folks such as Schmidt and Mann would choose to believe, but these are the outlets we have to deal with, and thus, whom I criticize as “the climate community.” If more of the fence sitters would open their freaking mouths, it wouldn’t be such a lopsided story we hear daily.
    .
    Quite frankly, I’m simply tired of the shrill voices forcing their version of morality down our throats.
    .
    Mark

  100. Posted Apr 7, 2009 at 12:00 PM | Permalink

    Just how good would the speleothem and tree ring data have to be for any mathematical analysis to trump the work of archeologists who put shovels in the soil of Greenland and found farms dated to the MWP?

  101. Mark T
    Posted Apr 7, 2009 at 3:31 PM | Permalink

    It would seem adaptive systems could easily be cast into a phase-space context, too, Ron. Hmm, perhaps I have read about such an approach in the past. I’ve done all I can to forget everything I knew up through last May, however, so it may be difficult to find relevant literature. :)

    Mark

    • Mark T
      Posted Apr 7, 2009 at 7:01 PM | Permalink

      Re: Mark T (#348), What on earth did you clip my comment on non-linear behavoir for, Steve? I can understand the bit about bender (which was said half jokingly anyway), but the rest was completely relevant to what can happen in non-linear systems. C’mon man.

      Mark

      • Andrew
        Posted Apr 7, 2009 at 7:04 PM | Permalink

        Re: Mark T (#353), It happens. Reiterate your points and cut out the offending remarks, and you’ll be okay.

  102. bender
    Posted Apr 7, 2009 at 3:45 PM | Permalink

    A process that becomes increasingly strong as a system changes state is “nonlinear”. It does not require a “sudden” appearance or disappearance of anything. Negative feedbacks from moist convection might, for example, be more powerful at warm GMTs than cool GMTs. The idea that cloud feedbacks may cap future warming does not say anything about their role in 20th c. warming. And the topic here at CA is, 99% of the time, climate of the PAST.
    .
    ///nts.

    • Andrew
      Posted Apr 7, 2009 at 5:18 PM | Permalink

      Re: bender (#349), Bender, Spencer’s work is not about hypothetical “future” feedbacks but instead about observed past feedbacks. There is some work, however, which may interest you because it seems to be saying what you are about nonlinear feedbacks:

      http://www.climatesci.org/publications/pdf/R-260.pdf

      I find generally when you and I “disagree” it is because I don’t understand what you are saying (it can’t be hard to articulate when you are busy antagonizing).

  103. Andrew
    Posted Apr 7, 2009 at 5:20 PM | Permalink

    That should be “can be”.

  104. Andrew
    Posted Apr 7, 2009 at 5:48 PM | Permalink

    The Idsos demolish “its worse than we thought” claims:

    http://www.co2science.org/education/truthalerts/v12/copenhagen.php

  105. Posted Apr 7, 2009 at 8:18 PM | Permalink

    Diagramme du jour:

    This shows the linear trends in the monthly UAH lower tropospheric temperatures. I’d like to understand why they look the way they look.

    Why is there a May warming minimum in both hemispheres?

    The Northern Hemisphere maximum warming in February looks consistent with AGW (greatest warming in the cold months), but the Southern Hemisphere’s coldest months are June, July and August, so why is the SH peak warming in October to January?

    Are these historical patterns consistent with GCM hindcasts?

    • Andrew
      Posted Apr 7, 2009 at 9:08 PM | Permalink

      Re: David Smith (#355), You know, I’ve been thinking of doing a thorough analysis of all the individual aspects of the satellite data sets, and this one is interesting. However, according to “Deep Climate” the monthly trends (globally) are very different for RSS and UAH, so it might be a good idea to repeat the analysis:

    • bender
      Posted Apr 7, 2009 at 9:47 PM | Permalink

      Re: David Smith (#354),

      Are these historical patterns consistent with GCM hindcasts?

      My bet is lucia can’t leave a question like that alone.

    • Mark T
      Posted Apr 8, 2009 at 9:43 AM | Permalink

      Re: David Smith (#354), Yeah, I know, but I can still gripe a bit. ;)

      This was the contended, but legitimate, statement:

      His (bender’s) point is valid, however; non-linear dynamics allow for the possibility that feedbacks may come and go (as well as other effects). Of course, this also implies a lack of stationarity, which causes problems with statistical analyses.

      Mark

  106. Posted Apr 8, 2009 at 5:44 AM | Permalink

    I’ll plot the RSS data this evening.

  107. Mark T
    Posted Apr 8, 2009 at 9:44 AM | Permalink

    Oops, that should have been Re: Andrew (#353). It’s early, only 9:45 a.m. and I’m still a bit sleepy.

    Mark

  108. Mark T
    Posted Apr 8, 2009 at 9:46 AM | Permalink

    Oh, wow, how relevant. My daily issue of PlanetAnalog arrived in the inbox today with this gem:

    I just found a new word: saw this is a commentary column this week (sorry, I neglected to note which one, then lost the page): “anecdata”, meaning when an anecdote is used as a data point for extrapolating to a trend, or used in place of unavailable or unobtainable actual data to support a position. And there is way too much of it in today’s so-called journalism and reporting, IMO.

    Mark

  109. Andrew
    Posted Apr 8, 2009 at 11:45 AM | Permalink

    Anybody else see the glitch looking thing here?

    • Andrew
      Posted Apr 8, 2009 at 3:40 PM | Permalink

      Re: Andrew (#363), I emailed Cryosphere Today, and Bill Chapman agrees there is probably a glitch here, they are looking into it.

  110. Andrew
    Posted Apr 8, 2009 at 4:07 PM | Permalink

    I’ve been working on comparing Arctic temperatures with sea ice data. Here’s what I’ve got so far. For reference, I am using the sea ice data from here:
    ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/

    And CO2 science’s convenient GHCN data base temperature data here:

    http://www.co2science.org/data/temperatures/ghcn.php

    I think, based upon eyeballing my home globe and sea ice maps, that it would be best to compare sea ice with temperatures from between 70 degrees North to 90 and 75 degrees North to 90. Because GHCN data is only available in 5 degrees latitude increments, I will use both. The sea ice area values have been corrected for the area which is outside the scope covered by the satellites-without this, both the values and trend are slightly less. So here is what I found:

    The dark blue curve is temps from 75 North to 90, the light blue is from 70 North to 90

    Thirteen month smooths (not moving averages) of sea ice extent (dark blue) and area (light blue)

    Now look closely. If sea ice trends are really linked to changing temperatures, then is it unreasonable to suppose that there were similarly low levels of sea ice in the late thirties compared to today?

  111. curious
    Posted Apr 8, 2009 at 5:18 PM | Permalink

    I guess that depends what you mean by “linked”…:) I’d also say the temp plot makes the 30s data look interesting – warming from 70 north but cooling from 75 north?

    • Andrew
      Posted Apr 8, 2009 at 5:31 PM | Permalink

      Re: curious (#366), Yes, and if anything, warming should be more dramatic the closer you get to 90 North. Curious indeed.

  112. curious
    Posted Apr 8, 2009 at 5:47 PM | Permalink

    Hmmm – that’s interesting too – what do you think happened in the extra 5deg band during the 30s that didn’t persist?

    • Andrew
      Posted Apr 8, 2009 at 6:00 PM | Permalink

      Re: curious (#368), I have no hypothesis on this, really. So I’ll say something you probably don’t hear often enough in climate discussions. I don’t know. Part of the journey of understand climate variability for me has been to piece the observational puzzle together. I work together explanatory hypotheses later. So your guess is as good as mine at this point.

      In the food fight me and bender were having earlier, which has mostly been snipped (rightly so) we were discussing Roy Spencer’s theory about attribution of GW-he has just further elaborated on his points:

      http://www.drroyspencer.com/2009/04/the-800-pound-gorilla-in-the-climate-system/

      And bender will be pleased to here is actually going to try to get it published in JGR!

  113. curious
    Posted Apr 8, 2009 at 6:20 PM | Permalink

    No, I don’t know either – would need to check out the source for the figures for both latitudes and how it was averaged around 360deg. Paper here that has 130year temp record at approx. 70N latitude 50W longitude:

    http://rdgs.dk/djg/pdfs/106/1/04.pdf

    • Andrew
      Posted Apr 8, 2009 at 7:06 PM | Permalink

      Re: curious (#370), Its a pretty interesting paper, and I have seen similar ones about Greenland. Can’t think of the references at the moment, though.

  114. Roy Spencer
    Posted Apr 8, 2009 at 6:45 PM | Permalink

    I think you are splitting hairs…there’s so little data that goes into the averages that far north that there’s going to be a lot of noise. The important thing is that Arctic warming in the 20’s and 30’s looks almost identical to the warming in the 80’s and 90’s.

    If sea ice continues to recover, and temperatures continue to fall…maybe the polar bears will be saved after all!

    • bender
      Posted Apr 9, 2009 at 10:28 AM | Permalink

      Re: Roy Spencer (#371),
      captain lecture pants the second telling Andrew what for

      • Andrew
        Posted Apr 9, 2009 at 10:33 AM | Permalink

        Re: bender (#382), read it again, I’m almost positive he was talking about curious and his pondering the difference between 70 North on up from 75 North on up.

        • Andrew
          Posted Apr 9, 2009 at 10:36 AM | Permalink

          Re: Andrew (#383), Plus, he didn’t have your disagreeable attitude…

        • bender
          Posted Apr 9, 2009 at 10:45 AM | Permalink

          Re: Andrew (#383),

          …there’s so little data that goes into the averages that far north that there’s going to be a lot of noise

        • Andrew
          Posted Apr 9, 2009 at 11:02 AM | Permalink

          Re: bender (#385), Two points:

          1. curious started the discussion about the difference between 70N-90N and 75N-90N, and this is what I think Roy meant. We completely agree that

          The important thing is that Arctic warming in the 20’s and 30’s looks almost identical to the warming in the 80’s and 90’s.

          and

          If sea ice continues to recover, and temperatures continue to fall…maybe the polar bears will be saved after all!

          2. Re: Andrew (#384), As I said, you have a nasty disposition towards me, and Roy gives no indication of disliking me.

  115. curious
    Posted Apr 8, 2009 at 7:17 PM | Permalink

    I agree re: the quality of the records and their average/extrapolation.

    As far as sea ice extent goes I’d be grateful for any leads from anyone on histories of sea ice extents from whaling expeditions(ship logs or other – I’ve seen ref. to catch related analysis) – I think this was in its heyday in the 1930’s so I’d expect there to be some good info. on sea conditions at both Arctic and Antarctic latitudes.

    Re: Antarctic peninsula ice sheet loss I’m unsure how these small (un!)attached ice shelves which have caught recent news would recharge? I don’t think they are recharged from a “land” mass feed? Do they need a sustained cold spell over a few years? Any info/refs. appreciated. Apologies if covering known/irrelevant ground.

    • Ron Cram
      Posted Apr 8, 2009 at 10:23 PM | Permalink

      Re: curious (#373),

      Regarding leads on the history of sea ice extent, I do not have any real data but I may have a few leads.

      Roald Amundsen was an explorer who first made the voyage across the Northwest Passage from 1903 -1905. From all accounts it was a rough trip and he used a kind of ice breaker ship but I understand there was not much ice by 1905. Although I could be wrong. Perhaps his logs have information about the amount of sea ice.

      The Northwest Passage was also made in 1944.

      Evidently, there were also reports of “unprecedented sea ice melting” in 1922. I hope you find these leads helpful.

      • Ron Cram
        Posted Apr 8, 2009 at 10:29 PM | Permalink

        Re: Ron Cram (#377),

        curious, I just found that Anthony Watts has blogged on the ice melting of 1922. It has some additional info that may help your search.

      • Ron Cram
        Posted Apr 8, 2009 at 10:35 PM | Permalink

        Re: Ron Cram (#377),

        Evidently there was relative warming and ice melt way back in 1817 as well. The rapidity of the change is, I think, what alarmed these people the most. And the fact they assumed the change would be ongoing. In most cases, I think the ice returned pretty much in the next year or two. I have not seen much in the way of multi-year alarmism in these situations… only in recent decades.

      • curious
        Posted Apr 9, 2009 at 2:26 PM | Permalink

        Re: Ron Cram (#377), Hi Ron – thanks for taking the trouble to post the links. I’ve seen refs to wide historical variation in Arctic sea ice cover and it seems fairly clear this is not a situation unique to the late 20c. The reasons I was asking for 30s whaling data were twofold: i) this is the period Andrew was querying in post 365 and ii) I’ve seen this ref. (and others):

        http://www.nature.com/nature/journal/v389/n6646/full/389057a0.html

        which looks promising but is not public access. For the Arctic it seems that navigation possibilities are valuable here, so there is evidence of passage when available. However this doesn’t apply to the Antarctic and it seems explorers and whalers are the best sources of info (?). As there was a whaling station on South Georgia:

        http://www.coolantarctica.com/Antarctica%20fact%20file/wildlife/whales/whaling1.htm

        my thinking was that this could offer insights into the Antarctic peninsular conditions from either physical obervations by those sailing out of there or, possibly, (perish the thought!) correlated observations. Posting on CA I’d hope that if any of the experts can come in on this they would. I’m sure with enough googling these things are “discoverable” but input from those who know is a lot quicker and more valuable!

        In the context of the small Antarctic ice sheets which are currently getting the news attention it would be good to have a decent “long view” opinion. Thanks in advance for any info.

  116. Jim Arndt
    Posted Apr 8, 2009 at 7:52 PM | Permalink

    Andrew and Mark,

    How many data sources? What are your SD’s? I don’t think there are very many station for that and if you use GISS what is the resolution?

    • Andrew
      Posted Apr 8, 2009 at 8:24 PM | Permalink

      Re: Jim Arndt (#374), All great questions which I am not prepared for answers to. Sorry. I’m kind of an arm chair analyst…

  117. Mark T
    Posted Apr 8, 2009 at 10:04 PM | Permalink

    Great, yes, but why, oh why, was I included? :)

    Mark

  118. Jim Arndt
    Posted Apr 9, 2009 at 9:26 AM | Permalink

    Mark T: #376

    Sorry a glitch in my browser thought your name had a graph to go along.

  119. Mark T
    Posted Apr 9, 2009 at 9:42 AM | Permalink

    Gotcha. I was being facetious anyway. ;)

    Mark

  120. Hemst 101
    Posted Apr 9, 2009 at 6:14 PM | Permalink

    This seems somewhat on topic because of the recent discussion.

    Why do the Ch 04 GLOBAL temperatures follow the temperatures expected for the Northern Hemisphere (highest in July)? http://discover.itsc.uah.edu/amsutemps/

    Wouldn’t a flatter line be more expected? Does this mean that the NH has far more influence on global temperatures than the SH? If this is true, what are the implications? This may have an obvious explanation and be a dumb question but just curious.

    • Andrew
      Posted Apr 9, 2009 at 6:36 PM | Permalink

      Re: Hemst 101 (#388), The Southern Hemisphere is mostly ocean, so its annual cycles are damped.

      • Raven
        Posted Apr 9, 2009 at 8:51 PM | Permalink

        Re: Andrew (#389)
        Yet the SH receives more solar isolation than the NH because the earth is closer to the sun during the SH summer. Where does this extra energy go? The ocean? Back into space?
        Open water has low albedo so if it goes back into space then it is not simply because the SH has more ocean.
        If it is absorbed then where does the heat go from there?

        • Andrew
          Posted Apr 9, 2009 at 9:13 PM | Permalink

          Re: Raven (#390), I have no idea. :)

        • DeWitt Payne
          Posted Apr 10, 2009 at 9:35 AM | Permalink

          Re: Raven (#390),

          Where does this extra energy go? The ocean? Back into space?

          The ocean does not warm or cool as fast as dry land. The ocean has a much higher effective heat capacity than dry land for several reasons. The specific heat capacity of water is higher and the thermal conductivity, including mixing, is orders of magnitude higher. Also, the solar energy in the visible and UV wavelengths penetrates up to 100 meters or so into the ocean. The end result is that the daily surface temperature variation of the ocean is a few tenths of a degree compared to tens of degrees for dry land. The seasonal variability of SST would also be smaller than for dry land.

        • Mark T
          Posted Apr 10, 2009 at 12:38 PM | Permalink

          Re: Raven (#390),

          Open water has low albedo so if it goes back into space then it is not simply because the SH has more ocean.

          Doesn’t the albedo from open water change significantly (increase) with angle of incidence, moreso for calmer water as well? Just curious.

          Mark

        • BarryW
          Posted Apr 10, 2009 at 2:00 PM | Permalink

          Re: Raven (#390),

          Here’s a definition

          Albedo, is the related to the reflectance of the light. If not reflected then the light and the heat is released though various other paths

      • Hemst 101
        Posted Apr 9, 2009 at 9:15 PM | Permalink

        Re: Andrew (#389),

        Thanks Andrew. Another piece of the puzzle. So what might be the implications of this? Wider variations in temperature in the NH and Arctic on longer term periods? Would this not cause an “imbalance” between the NH and SH? Or nothing of consequence?

      • Hemst 101
        Posted Apr 9, 2009 at 9:37 PM | Permalink

        Re: Andrew (#389),

        One more thing. Is the data for that graph available to the public? I get “you do not have rights to this server” or some such message when I use the URL on the graph.

  121. Andrew
    Posted Apr 10, 2009 at 12:25 PM | Permalink

    There is an interesting note on horizontal (as opposed to vertical) energy balance here. I don’t completely agree with it, but it is at least intuitively obvious to me that Milankovitch cycles would significantly alter the horizontal heat balance versus little for the vertical balance, and that their effects from that would be bigger.

    • DeWitt Payne
      Posted Apr 10, 2009 at 1:46 PM | Permalink

      Re: Andrew (#396),

      The basic facts in your link are correct AFAIK. The annual radiative surplus/deficit graph can also be found in A First Course in Atmospheric Radiation by Grant W. Petty. I hadn’t seen the seasonal variation charts before, but they logically follow from the seasonal insolation data. We also know from fossil evidence that a boreal forest existed in Antarctica about 50 MYa. The northern limit of the NH boreal forest was significantly farther north than present only 5 kYa during the Holocene Optimum. That would imply that in a warmer planet, the maximum latitude of the energy balance point during hemispheric summer would increase, possibly up to and including the pole. This would be balanced by an increased surplus of outgoing radiation at the other pole during hemispheric winter because the average temperature would be higher, i.e. polar amplification.

      IIRC, the GCM’s don’t get the position or velocity profiles of the circumpolar vortices right either.

      • jc-at-play
        Posted Apr 10, 2009 at 2:23 PM | Permalink

        Re: DeWitt Payne (#398),

        We also know from fossil evidence that a boreal forest existed in Antarctica about 50 MYa. … That would imply that in a warmer planet, the maximum latitude of the energy balance point during hemispheric summer would increase, possibly up to and including the pole.

        On the other hand, Antarctica didn’t cover the South Pole until about 2 million years ago. [At least as far as I can tell, looking at http://www.exploratorium.edu/origins/antarctica/ideas/gondwana2.html .]

        On the other other hand, in support of your point, even 50 million years ago it looks like Antarctica was still pretty far south, about 85% contained within the Antarctic Circle.

        • Andrew
          Posted Apr 10, 2009 at 6:02 PM | Permalink

          Re: jc-at-play (#400), Wait, how can you be mostly in the Antarctic circle but not cover the pole when you are a huge landmass?

        • Craig Loehle
          Posted Apr 11, 2009 at 6:34 AM | Permalink

          Re: jc-at-play (#400), When Antarctica reached the south pole, it set up a new circulation pattern, with the southernmost ocean circling the continent–my guess is that this has a huge effect because the cold water stays far south instead of mixing with warm water. A similar effect occured when North America got far enough west to block (at least slow down) circulation to the arctic with the Alaska string of islands (Aleutians). Also about 2 million years ago the Isthmus of Panama joined N. & S. America, cutting off an ocean current (and causing extinctions in the S. as animals migrated down). Without these geographic factors there would have been no ice ages.

  122. Hemst 101
    Posted Apr 10, 2009 at 4:23 PM | Permalink

    Also, I forgot , but I got interested in this by the discussion by Tom Vonk here:

    http://www.climateaudit.org/phpBB3/viewtopic.php?f=4&t=693&start=190

  123. cba
    Posted Apr 11, 2009 at 7:04 AM | Permalink

    Earth’s orbit puts it closest to the sun in Jan and at a maximum distance in july. Peak solar incoming in Jan. is about 95 w/m^2 greater than minimum in July. It’s a sine wave. That means though that the SH gets more rms power than the NH.

    Alebdo for water is quite low (0.04 might even be generous) at zenith incoming and increases with angle and probably turbulence as well. Of course, the incoming solar is much less at increased angles due to geometry and due to increased path through the atmosphere. Penetration of the ocean to greater depths is in the blue and perhaps somewhat uV realm which is a small amount of total sunlight yet it is the most variable portion since there is very little variation of the whole.

  124. Steve McIntyre
    Posted Apr 11, 2009 at 9:25 AM | Permalink

    I’m in Colorado Springs today for another wedding. Snowing outside my window. Might as well be in Canada.

    • Steve Reynolds
      Posted Apr 11, 2009 at 11:58 AM | Permalink

      Re: Steve McIntyre (#406),
      Mr. McIntyre,
      In the unlikely event that you have enough time while here in Colorado Springs to meet with Climate Audit readers, I would like to do that. Possibly at a local restaurant?

    • Mark T
      Posted Apr 11, 2009 at 12:52 PM | Permalink

      Re: Steve McIntyre (#406),

      I’m in Colorado Springs today for another wedding. Snowing outside my window. Might as well be in Canada.

      Yup, it was snowing when I got up, too, but it seems to have stopped now. It’s not really cold enough to sustain snowflakes. The misty crud, however, needs to go. We haven’t had much snow at all this year, btw. It’s been drought conditions again.

      Congrats/condolences to the newlyweds!

      Mark

  125. Max
    Posted Apr 11, 2009 at 9:26 AM | Permalink

    Totally unrelated to the topics at hand, but…
    Has anyone ever figured out how much wood is required to build a house to offset its carbon production from heating and lighting, or how much carbon has been locked up by using wood as a main construction material?.. This just occurred to me while researching new green materials like ICF’s and precast walls, while it may be “greener” on one level, it has zero carbon lockup when compared to wood.

    • D. Patterson
      Posted Apr 11, 2009 at 10:08 AM | Permalink

      Re: Max (#407),

      The production of concrete is one the principal anthropogenic sources of carbon dioxide emissions. Regrowth of wood serves as an important sink for carbon dioxide, wile it also encourages carbon dioxide emissions from the production of new soils.

      • Posted Apr 11, 2009 at 2:25 PM | Permalink

        Re: D. Patterson (#408),
        Yes, I figured that as well from seeing the train loads of coal lined up at our nearest cement plant, along with the monster gas line. I was just thinking that when people complain about deforestation and loss of a carbon sink, they are ignoring that the harvested wood most likely will lock that carbon up for some time in finished wood products, while the forest is replanted to again sequester more carbon.
        It would just be an interesting statistic to know how much is locked yearly by building with wood. The forest industry is always painted as evil doers, howerver they may be the oldest man made carbons sinkers in history.

        • Craig Loehle
          Posted Apr 11, 2009 at 2:35 PM | Permalink

          Re: Max (#412), Ah, finally a question that I am paid to answer! Use of wood can have several carbon sink benefits. Using wood offsets the fossil fuels required to make/process other materials. A good portion of the energy used to process wood to make paper, for example, comes from burning the scrap/bark. The trees then grow back. In a more complex analysis, carbon put into buildings stays out of the atmosphere for a long time (100+ yrs) and then the demolished home goes into a landfill where most of the wood never decays. Those worried about global warming, however, do not want to give credit for these ways of sequestering carbon because they simply do not want anyone to cut trees for any reason.

        • TAG
          Posted Apr 11, 2009 at 7:21 PM | Permalink

          Re: Craig Loehle (#413),

          then the demolished home goes into a landfill where most of the wood never decays.

          In Canada, bottled water has become the new green cause with many municipalities banning it for sale in schools etc. One repeated claim is that the plastic water bottle will not decay for 1000 years in a landfill.

          From a AGW point of view, wouldn’t biodegradability be a bad thing. Plastic does not decay and so must be an effective mehtod of carbon sequestration. Isn’t this true??

          As an aside, I used to work for a nylon manufacturer. Large quantities of nylon are used for home furnishings (carpets, upholstery etc. ) and in women’s clothing. Nylon is inert and does not biodegrade. Discarded clothing is going to last forever in landfills. This is a much more serious issue than water bottles. However, I do not foresee that green’s will be requiring women to sue something else rather than nylon in their clothing. AGW is serious but fashion is fashion.

      • Geoff Sherrington
        Posted Apr 12, 2009 at 12:21 AM | Permalink

        Re: D. Patterson (#408),

        Can you please expand on the “production of new soils?” Over much of Australia we are used to very old soils so lacking in carbon that we don’t even think in terms of carbon sinks or sources, more in terms of equilibria over millennia.

        Re: Craig Loehle (#413),

        My intuition is that we should ignore short-term C sequestration schemes like landfills amd wood buildings because they are a tiny man-made blip on the C cycle and they do not last very long compared to human history. Dendros know this from the difficulty of finding old wood to count rings upon. Overall, what goes into trees as CO2 mostly comes out of trees as CO2 with a miniscule portion hanging around for a few centuries, unless it goes through a coal making type of process which ain’t happening just now to any significant degree.

        So, if you want to avoid get-rich-quick snake oil schemes, don’t invest in new tree plantations. Unless you (a) increase the bound mass of C per given hectare over previous and (b) preserve that increased mass FOREVER, you are not doing anything to reduce CO2 in the air over time lines that matter.

        Great analogy with Ross McKitrick’s pay as you go emissions imposts and rewards. Tree planting companies seeking credits should be accounted every 5 years and paid (or made to pay) for the change in sequestered carbon they have managed. Big repay if you burn it down or turn it into fuel or anything else that oxidises at other than geological time scales.

        • MrPete
          Posted Apr 12, 2009 at 5:21 AM | Permalink

          Re: Geoff Sherrington (#419), interesting thoughts.
          Where this takes me is acidification, which AFAIK is a question of near-term “hits” to the carbon balance. If the issue is rate of carbon release, perhaps there are actions that would provide a temporary buffer against the rapid release. Or not. (Depending on whether it is even possible for human action to significantly influence the picture in any useful timeframe.)
          An interesting way to reflect on all this.

        • Craig Loehle
          Posted Apr 12, 2009 at 3:50 PM | Permalink

          Re: Geoff Sherrington (#419), Permanence is a big problem for carbon sequestration projects like new plantations. Any plantation can burn up or be killed by pests, so it is not guaranteed to be removed from the atmosphere. Plus, the acres of carbon projects are infintessimal compared to total emissions–it is strictly feel-good stuff.

        • D. Patterson
          Posted Apr 13, 2009 at 8:13 AM | Permalink

          Re: Geoff Sherrington (#419),

          Can you please expand on the “production of new soils?” Over much of Australia we are used to very old soils so lacking in carbon that we don’t even think in terms of carbon sinks or sources, more in terms of equilibria over millennia.

          Are you thinking mostly in terms of the arid and semi-arid regions of Australia, or all of Australia?

  126. Posted Apr 11, 2009 at 6:52 PM | Permalink

    Thanks Craig, could you come with a tonnage of carbon lock, based on Annual forest industry numbers, you would have to factor in cutting waste, de-limbing and debarking, then again much of the cutting waste now becomes particle board.
    I know its been totally ignored by the green crowd, I should call them overboard greens, as referring to them as just greens makes the rest of sound like anti greens, which we are not, we are just trying to live our lives in a rational manner inside our own economical boundaries.

  127. Posted Apr 11, 2009 at 7:16 PM | Permalink

    Here’s a plot of temperature trends, by month, for the Northern hemisphere:

    There are apples-to-oranges aspects to the graph, as satellites measure lower-troposphere values while Hadley uses surface measurements. Also, the satellites have limitations/gaps at high latitudes.

    And, the trend values are calculated as if the trends are linear whereas the true trends may be anything but. And, there is random variation.

    Despite these warts I think there are still a few points worth mentioning:

    One, it seems like CO2’s greenhouse effect should be strongest in the cold, dry winter, when there is minimal water vapor. The converse should be true in the warm, humid summer.

    What I see in the plots are few indications of such seasonality. There seems to be a hint of a minimum in December/January followed by a peak around February/March. Perhaps the late-winter peak is the wintertime CO2 effect shifted by several months, but why? And, why an early-winter minimum?

    The humid summer should be a time of minimum CO2 effect, one would think, but RSS and Hadley don’t suggest a minimum. UAH does show such a minimum. Interestingly, the UAH plot suggests that the effect is stronger in the early, less-humid part of summer, which is odd.

    Two, the patterns seem roughly consistent with the “amplification aloft” idea, with the lower-troposphere trends being greater than the surface trend, at least in winter. UAH hints at amplification aloft in the cool months but not in the warm months, which seems odd to me since my understanding is that the expected amplification is largely convection-related and convection (and amplification) should be stronger in summer.

    This data is obviously neither conclusive nor persuasive and barely rises to the weakly-suggestive level, but it is something to ponder for a minute or two.

    • DeWitt Payne
      Posted Apr 12, 2009 at 8:15 PM | Permalink

      Re: David Smith (#415),

      One, it seems like CO2’s greenhouse effect should be strongest in the cold, dry winter, when there is minimal water vapor. The converse should be true in the warm, humid summer.

      CO2 has the greatest forcing in the warm humid tropics and the least at the poles because the difference in temperature between the surface and the altitude where the atmosphere becomes transparent to CO2 emission at 15 micrometers is greatest in the tropics. I don’t know for sure that this is true on a seasonal basis in mid-latitudes, but I suspect that it works the same way there as well. One could look at the seasonal variation of temperature and altitude at the 200 Pa pressure level to see.

      • Andrew
        Posted Apr 13, 2009 at 7:51 AM | Permalink

        Re: DeWitt Payne (#423), At first this didn’t sound right to me, but it turns out you are correct:

        however, the gradient of temperature response is expected to be quite the opposite. Must just be the ice albedo feedback.

        • DeWitt Payne
          Posted Apr 13, 2009 at 10:01 AM | Permalink

          Re: Andrew (#426),

          I think, and I’m speculating here because nobody who knows better has confirmed or denied my hypothesis, that the increased temperature response at high latitudes (polar amplification) does have something to do with low humidity at the poles. We know that massive amounts of heat are transferred from the tropics to the poles (your post above, e.g.) driven by the temperature difference. Increased tropical forcing won’t raise the temperature of the tropics much because the high humidity means high heat capacity. But even if the mass transfer rate of warm air from the tropics to the poles doesn’t change (and I suspect it would increase), the amount of heat transferred would increase and the lower heat capacity of the drier air at the poles would mean a higher temperature change for a given heat input. Ice/albedo feedback could also contribute, but given the high percentage of cloud cover over Antarctica anyway, I’m not sure how much difference that would actually make.

          Anybody know what the average cloud cover is in the Arctic?

        • Andrew
          Posted Apr 13, 2009 at 11:11 AM | Permalink

          Re: DeWitt Payne (#428), The black line is observed total cloudiness percent by latitude:

        • curious
          Posted Apr 13, 2009 at 11:15 AM | Permalink

          Re: DeWitt Payne (#428), Hi DeWitt – this is OTTOMH but do you really think “polar amplification” is happening? I’ve been looking at Jeff IDs recon. of temps from surface station data and I don’t see it. Over the last 50yrs the trends are very small.

          From following at a distance I’m probably missing something but FWIW I have a lot more confidence in Jeff IDs result than the Steig paper which asserted the whole of West Antarctic has been experiencing a strong warming trend:

          An outstanding question in Antarctic climatology has been
          whether the strong warming of the peninsula has also occurred in
          continental West Antarctica(19). Our results indicate that this is indeed
          the case, at least over the last 50 years.

          Anthony Watts has posted a suggestion on Jeffs work of treating the peninsular as a separate climatic region and this seems sensible – the coastline to land area ratio for the peninsular will be way higher than for the other 65-70deg S latitude locations. Steve M has also highlighted a mining technique for “debiasing” the influence of “nuggets”.

          Which dataset(s) are you using to support the polar amplification theory for Antarctica?

        • Andrew
          Posted Apr 13, 2009 at 11:29 AM | Permalink

          Re: curious (#431), We are really discussing Polar Amplification in theoretical terms, but there are a few things to keep in mind about Antarctica. First of all, it is a huge mass of ice, giving it a year round very high level of reflectance. In Antartica, therefore, when there are fewer clouds, it actually its hotter-instead of the reverse. And it is not very cloudy to begin with (see above). The second issue is the unusual thinness of the Ozone layer there-there have been model studies done suggesting that Ozone depletion would cause cooling there. Not saying I agree but, well, the argument is there. So, we probably expect the signal to be clearer in the Arctic (and the theory of Polar amplification is pretty well supported in the proxy record (for any kind of change, actually)) but then we see that though the changes in the Arctic are big, they don’t seem to be 1. Unprecedented or 2. Monotonic (rather, pretty multidecadal).

        • Andrew
          Posted Apr 13, 2009 at 11:30 AM | Permalink

          Re: Andrew (#432), Er sorry:

          In Antartica, therefore, when there are fewer clouds, it actually its hotter-instead of the reverse.

          That’s wrong. I meant it gets colder

        • DeWitt Payne
          Posted Apr 13, 2009 at 10:27 AM | Permalink

          Re: Andrew (#426),

          Using the Archer MODTRAN interface, the forcing from doubling CO2 (280 to 560) in mid-latitude summer conditions (100 km looking down) is 2.81 W/m2. In mid-latitude winter conditions, the forcing is 2.29 W/m2. At 12 km altitude, the summer pressure used in the MODTRAN calculation is 209 millibar while in winter, the pressure at the same altitude is 188 millibar. The temperature difference between the surface and 12 km is 72 degrees in the summer and 53 degrees in the winter.

  128. Andrew
    Posted Apr 11, 2009 at 7:40 PM | Permalink

    Everybody have fun with this:

    http://wattsupwiththat.files.wordpress.com/2009/04/making-holocene-spaghetti-sauce-by-proxy.pdf

    Caio for now.

  129. Posted Apr 11, 2009 at 8:01 PM | Permalink

    Here is a plot of the relationship between UAH’s and RSS’ estimates of lower troposphere temperature:

    As expected, the two are highly-correlated.

    It is interesting, though, that the NH relationship seems to be stronger than the SH. And the relationship seems to vary through the year, with the strongest correlations in local summers.

    It’d be interesting to learn if these apparent differences are indeed real and if so, why they exist.

  130. Mark T
    Posted Apr 12, 2009 at 4:46 PM | Permalink

    I should add the weather update in CO Springs: we got several inches of snow this morning. Yay! It’s pure slush, however, and it ended up raining as we were coming back from our holiday lunc at Edelweiss, a nice German restaurant on the other side of town. :) You’d think people in CO would know how to drive in this not very unusual spring crud. They do not.

    Mark

  131. nevket240
    Posted Apr 13, 2009 at 12:20 AM | Permalink

    http://www.bloomberg.com/apps/news?pid=20601124&sid=aeYV68u.Fhp4&refer=home

    we can only hope, I suppose.

    regards

    • Craig Loehle
      Posted Apr 13, 2009 at 6:39 AM | Permalink

      Re: nevket240 (#424), How about the conflicts of interest when scientists serve on boards of advocacy groups? Even if your political cause is virtuous it can distort your science. University types seem to imagine themselves purer than the driven snow, but in fact may be so into the latest fad or so deep into politics that there is no way they can be objective in their research. This doesn’t just apply to climate science, of course, but here the stakes are much higher than in many other cases.

  132. curious
    Posted Apr 13, 2009 at 11:46 AM | Permalink

    Re: Andrew above – my question was/is:

    Which dataset(s) are being used to support the polar amplification theory in the Antarctic?

  133. curious
    Posted Apr 13, 2009 at 1:00 PM | Permalink

    Thanks Andrew – nice numbers! How do they support polar amplification theory for Antarctica?

    • Andrew
      Posted Apr 13, 2009 at 2:12 PM | Permalink

      Re: curious (#436), During the glaciation cycles, Antarctic temperature appears to have swung around in a ~ten degree range versus (someone point out if I’m wrong here) something like 5 degrees globally.

    • DeWitt Payne
      Posted Apr 13, 2009 at 2:22 PM | Permalink

      Re: curious (#436),

      Compare the Vostok temperature change of 7 to 8C from 25,000 years ago to a change of about 4C in tropical Africa over the same time period here. As noted in the linked abstract, tropical sea surface temperatures change even less.

      The problem right now is that we don’t know the size of the underlying trend in Arctic temperatures, if any, because they appear to be driven by variations in the AMO as well. If the AMO index continues to drop to the levels observed during the 1980’s and holds there for a few years, we should have a better idea of any longer term trend. The satellite high northern latitude anomalies do appear to be following the AMO index, (see graph) but it’s way too soon to tell if that’s a coincidence or not.

  134. curious
    Posted Apr 13, 2009 at 3:49 PM | Permalink

    Andrew and DeWitt above – I thought the Polar Amplification theory was postulated in relation to AGW theory and it’s recent effect on the Arctic:

    Evidence of polar amplification depends on the timescale
    of examination. Over the past 100 years, it is possible
    that there has been polar amplification, however,
    over the past 50 years it is probable that polar amplification
    has occurred.

    p22 (and 23): http://www.acia.uaf.edu/PDFs/ACIA_Science_Chapters_Final/ACIA_Ch02_Final.pdf

    Thanks for the pointer to the longer view – is there a reference on Polar Amplification?

    • Andrew
      Posted Apr 13, 2009 at 4:07 PM | Permalink

      Re: curious (#439), Actually, its been known for some time now that all climate changes seem to be more dramatic in the polar regions. Now if one believes that all past climate changes are due to CO2, you might think that this is an exclusive signal of AGW/GHG induced change, but it seems to just be what you would logically expect from an ice albedo feedback and possibly the idea stated above that humid air has a higher heat capacity.

      • Posted Apr 15, 2009 at 10:42 AM | Permalink

        Re: Andrew (#440),
        I cannot understand why temperature changes at the Poles are to be linked to air water content.
        Polar air is drier. That’s a fact. And it has no sense to me to ask how polar areas would be if the atmosphere over there were moister or to compare tropical and polar water content.
        Poles are colder than tropics for geographical reasons and an effect of this is that specific humidity MUST be less. Period.
        But Poles are as cold as they are not only for geography but also because water freezes at temperatures below zero.
        It’s a common experience to observe that two close areas, one with snow and the other snow free, have very different T2m in event of tranquill weather.
        Moreover, snow or ice covered surfaces are active meteorological elements as they stabilize the atmosphere so that low level radiative cooling can act with few disturbs, with further stabilization . Meanwhile air becomes drier and drier.
        If air holds less competitors (water molecules), CO2 can show all its efficiency in IR absorption.
        In this sense, CO2 has a more direct effect at the Poles than in moister climates.

        Anyway, because the most effective polar feedback is the presence of frozen water, remove ice or snow and you have the polar amplification. No air water content implication is required.

        And all this regards climate change time scales.

        If weather is concerned, year by year (as well as seasonal or monthly) variability is bigger where gradients and advection are important players, i.e. at midlatitudes and poles. The Equator is the weather machine engine and has two big allies that will defeat any attempt from higher latitudes: the Sun and atmospheric circulaton!
        It has also an enemy: the underlying sst and its changes.

        Andrew, I know that is not an idea of you.

        • Andrew
          Posted Apr 15, 2009 at 11:05 AM | Permalink

          Re: Paolo M. (#449), It has more to do with the air over the tropics being harder to heat up (takes longer), due to its relative moistness. But you make some good points, so please remember this was not my suggestion originally but DeWitt Payne’s (#428).

        • Posted Apr 15, 2009 at 3:02 PM | Permalink

          Re: Andrew (#450),
          my last sentence in post #449 clearly stated that this was not a suggestion of you, don’t you agree?
          Now, your answer #450 makes it clear that you endorse DeWitt Payne’s opinion on a polar amplification leaded, at least in part (and I restate “in part”), by water air content.
          I tried to explain you that simple meterological considerations make it clear that this can’t be the case.

          Poles are heat sinks and the primary way to change polar air heat content is by changing heat advection from the tropics. Of course if summer insolation changes through astronomical or albedo variations, that also would have an effect.
          I really would not be concerned by how long it takes to heat tropics, since tropics heat poles!
          Radiation at poles has the only effect to modulate the rate of cooling.
          No enough room is left for significant influences by air water content.

        • DeWitt Payne
          Posted Apr 15, 2009 at 3:02 PM | Permalink

          Re: Paolo M. (#449),

          If air holds less competitors (water molecules), CO2 can show all its efficiency in IR absorption.
          In this sense, CO2 has a more direct effect at the Poles than in moister climates.

          No it doesn’t as I’ve pointed out several times, most recently here. There is a competing effect which dominates.

          On reflection I see I haven’t been using correct terminology to explain what I mean by humidity causing polar amplification. When I said heat capacity, I should have said heat content or heat capacity at constant relative humidity. The heat capacity of air in the absence of liquid water is not very different than dry air. The heat content of moist air, OTOH, can be much higher than dry air and increases exponentially once specific humidity becomes significant. For example, according to this table, the heat content (enthalpy) of air saturated with water vapor at 25C is 94.4 kJ/kg. At 30C, the heat content is 117.8 kJ/kg, or a difference of 23.4 kJ/kg or ~4.7 kJ/degree kg. At -40C, the heat content is -22.2 kJ/kg and at -30C it’s -11.7 kJ/kg for a difference of 10.5 or 1.05 kJ/degree kg, not very different from the heat capacity of dry air at 1.004 kJ/degree kg. So the same amount of energy required to raise the temperature of 1 kg of saturated air at 25C by 1 degree while maintaining saturation would raise the temperature of -40C air by 4.5C. The effect is smaller at lower relative humidity and nearly vanishes at constant water vapor pressure, a condition that seems unlikely in the presence of liquid water.

          Hence higher radiative forcing in the tropics doesn’t raise the temperature enough to achieve radiative balance (it’s out of balance already), so the excess energy must be transported to higher latitudes where it causes a larger temperature increase in cooler air with lower specific humidity and a lower rate of change of water vapor pressure with temperature.

        • Posted Apr 15, 2009 at 3:31 PM | Permalink

          Re: DeWitt Payne (#454),
          here is late so I’m going to sleep, but many sentences of you are quite obscure.
          What does “the excess energy must be transported to higher latitudes” mean? I can’t find out a sense.

          It seems you are talking of “power advection” (watts) and that does’t exist.
          In atmospheric physics we deal with “temperature advection” and the system has already accounted for all your consideration on tropical moisture.
          I really don’t understand what you are dealing with. Maybe it’s the night….
          good night

  135. See - owe to Rich
    Posted Apr 14, 2009 at 1:31 AM | Permalink

    I just got back from ski-ing in Sestriere, Italy. The ski instructors were miffed that the resort was closing yesterday when the snow was so good – new snows of 70cm on April 3rd and 20cm on April 12th.

    I don’t know what the closing dates were for previous seasons, but I’m wondering if they had been fed global warming warnings about ever-shortening ski seasons.

    Rich.

  136. Christian
    Posted Apr 15, 2009 at 4:48 AM | Permalink

    Sorry. i know this is besides the topic, but I dont know where else to ask. I hope you will, nevertheless, either direct me to a proper place for my question, or simply answer them. Thank you in advance.

    1) What are the two or three major points that suggest that AGW is unsupported by science??
    2) What are the two or three major scientific facts that actually support AGW??

    Thank you for a great site. I am in molecular biology, so I am acquainted with science but not climatology or physics, and I find this site one (if not thé) most serious.

    Christian
    Denmark

    • Scott Brim
      Posted Apr 15, 2009 at 5:55 AM | Permalink

      Re: Christian (#3)
      Christian, I don’t have good answers for you myself.

      However, I do find it somewhat interesting that in comparing the two graphs shown above, the US Government’s graph indicates a steady upward trend in temperature anomaly between 1980 and 2008, while Steve’s graph indicates that, visually at least, the overall trend was flat — more or less, and subject to interpretation.

      I would find it even more interesting if an objective analysis were to be performed of the data that went into each graph, and also of the processes and the procedures that were used to produce each graph, and that the two methodologies then be compared side by side from both an accounting and a statistical perspective.

      After performing this comparison, is it possible that we might then discover, as in the words of Shakespeare, that “something is rotten in Denmark.”

    • Steve McIntyre
      Posted Apr 15, 2009 at 7:04 AM | Permalink

      Re: Christian (#442), Christian, much to the frustration of many readers, I discourage efforts to discuss AGW in one-paragraph bites and normally disallow it. The effort here is to verify mainstream studies. I’ve long sought an engineering-quality A-to-B exposition of how doubled CO2 leads to say 3 deg C global warming; I’ve asked many critics of this site and so far none have been provided. If such is identified, we will discuss it. In the mean time, I discourage bite-sized opinions.

      Many readers have far more rigid views than I do.

  137. Allen63
    Posted Apr 15, 2009 at 5:22 AM | Permalink

    Christian,

    I’ll leave it to others more qualified to give the long and best answer. However, perhaps like you, I was not current with AGW theory a year ago. Now, I have studied and even done my own models and calculations.

    Regarding AGW facts: Even the simplest pro-AGW “facts” that I believed a year ago seem “questionable” now.

    The argument for AGW seems to boil down to this:

    1. It is plausible that CO2 increases could “warm” the atmosphere — lab experiments have shown this since the early 1800s.

    2. CO2 has increased over the last 100 years. And, temperature has increased over the last 100 years.

    3. The “Climate Scientists” can find no other reason than CO2 that the temperature has increased.

    Of course, a different responder would give a different list — you have to study this yourself to really understand.

    Now, against AGW:

    Too many to list. For example, as shown in this article, its not even certain that the temperature is rising anymore. The chart from 1980 to now could be explained as basically no-change with dips where there were serious volcanic eruptions and a high where there was a strong el nino.

    And, Scientists have found “other reasons than Co2″ for what heating has occurred and what cooling is occurring now.

    I have become very skeptical. But, it is up to each individual to do their own studies and reach their own conclusions. For what its worth.

  138. Harry Eagar
    Posted Apr 15, 2009 at 6:31 AM | Permalink

    Christian, the chief reason to regard claims about runaway global surface warming as unsupported, as far as I am concerned, is that there are no global surface temperature observations before the 21st century.

    If you want to make claims, you are therefore making claims against proxies for past climate. As Mr. McIntyre keeps showing, the proxies are problematic.

  139. Allen63
    Posted Apr 15, 2009 at 7:24 AM | Permalink

    Cristian,

    I agree with Steve that short answers cannot move the “debate” forward. That is why I emphasize that individuals must do their own diligent studies — and form their own point of view.

    Worth remembering that the “burden of proof” lies with those who profess AGW “is a catastrophe in the making” — not with those who are “skeptical” of such claims.

    BTW, “skeptical” does not mean one has proven the opposite. In my case, it merely means that I am unconvinced by the so-called proofs provided by the pro-AGW camp.

  140. Posted Apr 15, 2009 at 7:27 AM | Permalink

    Christian, you could start by looking at Jeff Id’s Ten reasons to be a global warming skeptic.

  141. compy
    Posted Apr 15, 2009 at 11:38 AM | Permalink

    Christian,

    I think your question is a little too simplistic. IMHO, the critical debate is not whether there is an AGW effect, but how significant it is. The IPCC says the sensitivity of temperature to C02 doubling is 2-4 degrees celcius. There is an initial one degree of warming that is increased by feedbacks. Many of us readers of this blog accept AGW, but are skeptical of this range. If the feedback effects are neutral or negative, then AGW is not a crisis.

    I am not a meteorologist, climate scientist or physicist. My background is financial / Wall Street quantitative analysis. My knowledge of radiation, Stefan–Boltzmann’s law and related items is pretty close to zero. So I am in no position to comment at all about the Greenhouse Effect. I accept that it exists as there appears no credible evidence from reputable studies suggesting otherwise. I don’t believe that I have any other option but to follow authority on this point. However, I have a quite reasonable expertise in stochastic processes and statistics. And I feel quite qualified to read the Wegman report, NAS findings and the host of blog entries on the Hockey Stick here and on other lesser sites (Tamino, that means you). It is clear to me from spending a horrible number of hours on reading, that the hockey stick just cannot be supported by good analysis. Not that this in anyway disproves an AGW effect, just that the data has nowhere near the resolution to allow one to pronounce authoritatively on global temperature changes on a millennial scale. The reaction of the pro-AGW response to these challenges was instructive. Instead of accepting legitimate criticism, there was a circling of the wagons characterized by a widespread (but not universal) refusal to archive data and reluctance to concede any errors. It may appear that this is all just a sideshow, but evidently the AGW mainstream was partially relying on the paleo evidence to validate their computer models and associated sensitivity estimates. Not to mention the PR benefits of being able to show a graphical hockey stick. It is up to each individual to reach their own conclusions as to what this means for the credibility of the 2-4 degree range (and other pronouncements of the H-S defenders).

    • Andrew
      Posted Apr 15, 2009 at 11:58 AM | Permalink

      Re: compy (#451), GASP! An evil Wall Street exec! I know this is totally off topic, but I have to ask, hodo you feel about the current political situation around you? There are many who are calling for execs heads…

      More on topic, I completely agree with the sentiment that the real issue is not “is there any AGW” but “is there an alarming amount of AGW”-which I would answer in the negative.

      • compy
        Posted Apr 15, 2009 at 12:38 PM | Permalink

        Re: Andrew (#452), Andrew, I wouldn’t go so far as to call myself an exec – leave my head alone!

        • Andrew
          Posted Apr 15, 2009 at 3:36 PM | Permalink

          Re: compy (#453), No worries, I’m a capitalist, I don’t believe in beheading the rich.
          Re: Paolo M. (#455), I don’t endorse it, I really haven’t had time to mull over yours and Dewitt Payne’s arguments and decide who is right. I was just trying to explain my understanding of his arguments.

  142. nevket240
    Posted Apr 16, 2009 at 7:43 AM | Permalink

    http://paul.kedrosky.com/archives/2009/04/the_end_of_chea.html

    apologies for only a link. I cannot get the graph to transfer.
    Look at the frequency and duration of drought. Huge.

    regards

  143. Willem Kernkamp
    Posted Apr 17, 2009 at 7:06 PM | Permalink

    I did not mean to say that future climate states do not depend on those in the past. To the contrary. I am arguing that climate development can be studied by calculation with Global Climate Models even when their accuracy for weather prediction breaks down after a number of days. Even if a model does not perfectly track the actual succession of weather patterns it still can have an equivalent succession of weather patterns (i.e. the same climate).

    I am an aeronautical engineer with experience in Numerical Methods for aerodynamic calculations. These methods have benefited from the increase in speed and associated grid refinement to make them practical design tools. There continue to be issues, but measurements have issues also. Readers of climate audit are well aware of that fact as they regularly examine the difficulty in assessing global temperature trends. The key is to make an intelligent use of both.

    • Andrew
      Posted Apr 17, 2009 at 9:10 PM | Permalink

      Re: Willem Kernkamp (#18), What a coincidence, I’m studying to become an aeronautical engineer! Stick around, I’m sure you and I may have many interesting conversations about the perspective we have! :)Re: Ron Cram (#20), You say:

      Also, your belief that climate is “inherently stable” is a bit misleading. I think this is pretty well understood by most of the regular readers of this blog. If you believe the story about “We’ve got to get rid of the MWP!” is true, then it is seems part of Mann’s goal with MBH98/9 was to prove climate changed within a very narrow band prior to the industrial age.

      Er, a Sine wave is stable, no?

      • henry
        Posted Apr 17, 2009 at 10:45 PM | Permalink

        Re: Andrew (#23), I’ll agree, that a sine wave of a specific frequency could be considered “stable”.

        But if you understand the principles of heterodyning, the mixing of two sine waves of different frequencies results in four frequencies (the original two, and the sum and difference frequencies).

        It could be, that all of these different “oscillations” (NAO, PDO, ENSO, etc) are creating other “mixed” signals and that these new signals are what we see as climate (or as climate “noise”).

        • Andrew
          Posted Apr 17, 2009 at 11:12 PM | Permalink

          Re: henry (#26), I basically agree with what you are saying. But one needs to be careful as to how one defines “mixing”-to the extent that signals are not independent, they can’t just be added. And then we get into those nasty non-linearities. This is why one should be extremely skeptical of claims of signal detection. There are a lot of periodic, aperiodic and quasi-periodic phenomenon superimposed on one another which not only combine by interact in mathematically interesting ways.

          Re: Willem Kernkamp (#25), In defense of models, which believe me is not something I often do, the problem with the MWP may not necessarily be with the parameters so much as the input. How do they know that their paleo solar forcing and volcanic forcings are correct? Well, the really don’t, especially in the solar case. And how do they know that brightness changes and volcanoes are the only natural factors they need to account for? Again, they don’t. The fact that possible garbage goes in means that probable garbage comes out, regardless of how good the model itself is.

      • Ron Cram
        Posted Apr 18, 2009 at 9:45 AM | Permalink

        Re: Andrew (#461),
        Of course, you are correct. But I do not think climate operates as a sine wave exactly. It certainly does not operate as the alarmists would like us to think, within very narrow bounds. While certain aspects of climate may be predictable, such as the PDO, the climate as a whole is not.

        Re: Willem Kernkamp (#462),

        My advice is to forget global temp data. It’s useless, confounded with so many problems as to be meaningless. There are UHI effects, microsite issues and the fact the data is subject to mischief by the keepers of the data (some of whom refuse to share much info about their methods). More and more people are coming to this conclusion. The only reliable metric to monitor radiative imbalance is ocean heat content calculated from Argo data. Argo data is freely accessible to everyone and so there is less chance of abuse. If the planet is heating up, it will show in the oceans first and be impossible to hide.

        Yes, there are features of nature that yield themselves to arithmetic but not a complex process like climate. We do not even know all of the inputs that affect climate. Just recently Roy Spencer proposed the PDO phases could affect cloud formation. I think he is on to something, but if he is right how many other inputs are out there we do not know about? Lots, IMO. Plus, the computer modelers have not even incorporated the negative feedback Spencer found over the tropics. I am not saying GCMs have no value. The main value is they teach us what we do not know about climate. In other words, they can tell us where to look to learn more – so there is value there. However, regional climate models over short time frames are about as good as flipping a coin. What makes you think global climate models over 100 year time period are going to be any better? Climate models have no predictive value… and I predict they never will.

        • bender
          Posted Apr 21, 2009 at 12:30 AM | Permalink

          Re: Ron Cram (#472),

          While certain aspects of climate may be predictable, such as the PDO

          PDO is wholly unpredictable. Show me the proof otherwise.

        • Ron Cram
          Posted Apr 21, 2009 at 5:52 AM | Permalink

          Re: bender (#493),

          Based on observations in the south Pacific, Bratcher and Giese published a paper in 2002 predicting the PDO would switch to the cool phase “in about four years.” They were slightly off, since it switched in late 2007. But it was not a bad call. Each PDO phase generally lasts 30-35 years. This is not a secret. Perhaps you were thinking I meant prediction in a much more precise way?

        • bender
          Posted Apr 21, 2009 at 7:16 AM | Permalink

          Re: Ron Cram (#494),

          Bratcher and Giese published a paper in 2002

          What is the citation?

          the PDO would switch to the cool phase “in about four years.” They were slightly off

          One correct guess does not imply predictability.

          it switched in late 2007

          Says who?

          Each PDO phase generally lasts 30-35 years.

          Again, what is your evidence of 30-year periodicity? Stop talking and start showing.

        • Ron Cram
          Posted Apr 21, 2009 at 10:14 AM | Permalink

          Re: bender (#495),

          Google Scholar is very easy to use to find the Bratcher, Giese paper. This paper, and its prediction the PDO would turn in about four years, never got the attention it deserved.

          A study of PDO variability (using treerings) back to 1661 showed the average length of to be about 23 years. However, the length has grown longer as it got closer to the 20th century.

          According to JISAO:

          Two main characteristics distinguish PDO from El Niño/Southern Oscillation (ENSO): first, 20th century PDO “events” persisted for 20-to-30 years, while typical ENSO events persisted for 6 to 18 months; second, the climatic fingerprints of the PDO are most visible in the North Pacific/North American sector, while secondary signatures exist in the tropics – the opposite is true for ENSO. Several independent studies find evidence for just two full PDO cycles in the past century: “cool” PDO regimes prevailed from 1890-1924 and again from 1947-1976, while “warm” PDO regimes dominated from 1925-1946 and from 1977 through (at least) the mid-1990’s.

          The description of the graphic, while similar to others I have seen, is significantly different from other descriptions of the same graphic. Most descriptions say the graphic shows a warm phase beginning in 1905 and ending in 1945, then a cool phase from 1946 to 1976, then warm from 1977 to 2007. One confounding factor is the fact it is possible to have short-term counter trends to the overall PDO phase, as you see around 1920 and 2000. Based on the graphic, it appears the first warm phase started around 1900.

          Regarding the usefulness of this information of past behavior to “predict” when the PDO will turn, authors of one study “suggest that long-term fire planning using the PDO may be possible in the Pacific Northwest, potentially allowing decadal-scale management of fire regimes.” So, I am not alone.

          What makes you think the PDO is entirely unpredictable?

        • bender
          Posted Apr 21, 2009 at 10:57 AM | Permalink

          Re: Ron Cram (#498),

          Google Scholar is very easy to use

          Gosh. Thanks, Dr. Ladbury.

          A study of PDO variability (using treerings) back to 1661 showed the average length of to be about 23 years

          And so you accept that tree-ring based climate reconstructions are sufficiently precise to yield that kind of, errr, estimate. Proof please. (This oughta be fun.)

        • Mark T
          Posted Apr 21, 2009 at 11:13 AM | Permalink

          Re: bender (#501),

          And so you accept that tree-ring based climate reconstructions are sufficiently precise to yield that kind of, errr, estimate. Proof please.

          You asked for a paper, he provided one. Why not dissect the paper first prior to asking for further proof?

          Mark

        • bender
          Posted Apr 21, 2009 at 11:24 AM | Permalink

          Re: Mark T (#503),
          Oh please. He provided one piece of evidence, which I question. The burden of proof is on the proponent, not the skeptic. You know that.

        • Ron Cram
          Posted Apr 21, 2009 at 2:46 PM | Permalink

          Re: bender (#503),

          I did not provide one piece of evidence, I provided several. The Bratcher and Giese paper, the treering paper (I understand from our host that treerings have some viable uses even if temperature reconstructions are not among them) and the chart from JISAO.

          As I recall, you never did respond with a Hansen paper describing radiative imbalance in the sense you understood it.

          Steve wants a limit on food fights. I have a feeling this is degenerating into one.

        • Ron Cram
          Posted Apr 21, 2009 at 2:47 PM | Permalink

          Re: Ron Cram (#510),

          Oh, and I also gave you the forestry paper describing the proposal to use PDO for long range forestry planning.

        • bender
          Posted Apr 21, 2009 at 11:26 AM | Permalink

          Re: Mark T (#503),

          Why not dissect the paper

          IIRC I was among the first to discuss it here. Maybe Ron can figure out how to use the search tool to see where? It’s oh-so-helpful.

        • bender
          Posted Apr 21, 2009 at 11:28 AM | Permalink

          Re: Mark T (#503),

          You asked for a paper

          I did not. I asked for “evidence”. As in: a coherent and credible argument based on robust analysis of verifiable facts.

        • bender
          Posted Apr 23, 2009 at 7:44 AM | Permalink

          Re: Ron Cram (#498),
          Ron Cram, you suggest we are using different notions of the term “predictability”. However I dispute that. It is quite clear from your remark:

          This paper, and its prediction the PDO would turn in about four years, never got the attention it deserved.

          that you think that switches in PDO phase are predictable within a 4-year time horizon. That is what I disagree with, so stop pretending we’re using different dictionaries. We’re not.

          What makes you think the PDO is entirely unpredictable?

          You try to shift the burden of proof onto me to prove that PDO is not predictable. The burden of proof is on you who says it is “predictable”. I agree that it is not a white noise process. That’s as far as I’ll go.

          You ask me to refute a “forestry article” on fire prediction that is 100% irrelevant to the issue of weather prediction. Please. Stop jumping around and either prove your assertion that PDO is periodic and predictable or retract it. PDO is not predictable. You are brainwashing yourself, and it is very entertaining. What people will believe when they want to believe it.

          As for Bratcher & Giese, why don’t you tell us what you find so compelling about their prediction methodology? Why do you say this paper did not get the attention it deserved? Just because [you assert that] they made a correct guess?

          Out of a hundred meteorologists you are guaranteed by random chance alone that one of them is going to correctly “predict” an extreme, unlikely event. Cherry picking post-hoc, based on performance, not methodology, will give you a new weather guru every time. Sport punditry uses this method and is full of false prophets.

          Tell me Ron: how many years does it take before a phase-switch can be inferred? Two? Twenty? What does your paper say? I don’t think you understand the difference between hindsight and foresight.

        • Posted Apr 23, 2009 at 12:24 PM | Permalink

          Re: bender (#529),

          Really? I don’t understand the difference between hindsight and foresight? I am the one who is constantly drawing attention to the peer-reviewed papers on scientific forecasting, including the journal Foresight. Climatologists haven’t a clue when it comes to the principles of scientific forecasting.

          Bratcher and Giese’s paper does not show foresight? Based on their observations, they predicted the PDO would turn in “about four years.” They were off by a year and a half. Not too bad and certainly not hindsight. It has been several years since I read the B&G paper so I cannot recall even what the observations were that led to their conclusion, but I remember being impressed with the article. I chalked up the fact it did not get more attention because the paper was less than alarmist. By the way, did the B&G paper follow the principles of scientific forecasting? No. But the outcome was close enough to be respectable.

          To answer some of your questions: The PDO has been fairly consistent over the last 100 years, lasting between 30-40 years. If I were to guess the current cool phase will last 35 years, I will probably be correct within five years. How many climate processes can you say that about? Evidently, it is possible to predict more precisely after certain events have occurred in the south Pacific. I would have to reread the B&G paper to relearn those events.

          One of the confounding factors is when the PDO shifts temporarily. Kind of like Braxton Hicks contractions, it is a false signal. These false signals seem to be brought on by particular strong El Nino (during warm PDO) or La Nina (during cool PDO) events, either during or just after. I consider this to be a kind of negative feedback to ENSO.

          The thing that irritates me the most about this exchange, bender, is the way you want to make everything so personal. You act as if I am a complete idiot to think the PDO is predictable. I am not alone. That was the point of the forestry paper. This is an example of people planning forestry efforts based on predictions about the PDO. So my thinking is not completely uninformed. You may or may not agree with the position I have stated, but there is no reason to attack my intelligence. I have read and I have provided you sources and citations.

          If you think I am so far off the wall, provide me some sources that say the opposite of my statements. Give me the citations where people say the PDO is not predictable.

          Now you ask me to find citations for your arguments. I won’t waste my time because I do not think such a Hansen article exists. You must have read him incorrectly. I will continue to think so until you provide the citation that proves me wrong.

          Nothing personal, bender. If you want to discuss the science and learn from one another, great. But stop attacking people.

        • bender
          Posted Apr 23, 2009 at 12:37 PM | Permalink

          Re: Ron Cram (#543),

          Based on their observations, they predicted the PDO would turn in “about four years.”

          Quote, please.

      • bender
        Posted Apr 19, 2009 at 8:06 PM | Permalink

        Re: Andrew (#461),
        All evidence suggests climate does not behave as a sine wave.

        • BarryW
          Posted Apr 19, 2009 at 8:13 PM | Permalink

          Re: bender (#483),

          Then the PDO is weather not climate?

        • bender
          Posted Apr 19, 2009 at 9:07 PM | Permalink

          Re: BarryW (#484),
          Ask Andrew, he’ll set you straight.
          1. PDO is not “Earth climate”. It is a tiny piece of it.
          2. There is no strong evidence that even PDO behaves as a sinusoid.
          3. There is no evidence the other modules behave as sinusoids.
          As for “weather vs. climate” – a great question – it may help to consider the gray between that silly black and white dichotomy. Circulation IMO is neither weather nor climate. It’s in between.

        • Andrew
          Posted Apr 19, 2009 at 9:53 PM | Permalink

          Re: bender (#483), Bender, a sine wave is just an example of a function which is stable but changes. I make no claim that climate behaves that way. See here for evidence that I say just the opposite.

          Re: bender (#485), I totally agree that the black/white view of climate versus weather is silly and that there is no clear boundary but huge swaths of gray. Still, what are reasonable constraints on the gray area? Obviously 1 day is weather, and a century is climate, no?

        • bender
          Posted Apr 20, 2009 at 7:08 AM | Permalink

          Re: Andrew (#486),
          A sine wave is an example of something that is irrelevant to the question. To argue that climate is “stable” requires a much more convincing example.

        • Andrew
          Posted Apr 20, 2009 at 7:43 AM | Permalink

          Re: bender (#487), Bender, it is enough to show that the dichtomy that stable=no change is false. I’m sure there is a better example than a sine wave, but that doesn’t mean it is “irrelevant”-You and I just have different ideas of what thepoint I am trying to make is. Not surprisingly, I am the one who knows what I am trying to say, and not you.

        • bender
          Posted Apr 21, 2009 at 12:27 AM | Permalink

          Re: Andrew (#488),

          I am the one who knows what I am trying to say, and not you

          If you were clear about what you were saying then no one would have to guess. As long as we agree that there is no evidence whatsoever that ‘climate is stable, like a sine wave’ then let’s just quit while we’re ahead. There are folks – perhaps not you – who are trying to make that very argument. You think it’s always about you. It’s not. It’s about propagandists peddling half-truths about sinusoidal background global climate variability. If you’re not among them, bully for you.

        • Posted Apr 20, 2009 at 10:05 PM | Permalink

          Re: Andrew (#486),

          What you’re describing is a stable limit cycle, however the PDO doesn’t look like one.

        • Andrew
          Posted Apr 21, 2009 at 7:31 AM | Permalink

          Re: Phil. (#491), I don’t recall mentioning the PDO at all.

          Re: bender (#492), I get a bit edgy when you respond to my comments. There is a history there, after all.

        • Posted Apr 21, 2009 at 9:18 AM | Permalink

          Re: Andrew (#496),

          Re: Phil. (#491), I don’t recall mentioning the PDO at all.

          Others responding to your posts did, this is a discussion board not your personal email.

          Re: Ron Cram (#494),

          I see this ~30 year PDO periodicity mentioned frequently here and elsewhere, I agree with bender (that will no doubt shake him to the core ;) ) where’s the evidence?

          The monthly data for the last 109 years is available here for anyone who wants to do some analysis.

        • Ron Cram
          Posted Apr 21, 2009 at 10:20 AM | Permalink

          Re: Phil. (#497),

          Without seeing your comment, the graphic you posted is the same as being discussed in my comment above. It is hard for me to see the cool phase from 1890 to 1924. It sure looks like a warm phase to me. But I certainly see a cool phase from ~1945-1975, then a warm phase from 1976-2007.

        • Andrew
          Posted Apr 21, 2009 at 10:59 AM | Permalink

          Re: Phil. (#497), Well address them, not me. Otherwise you are just wasting my time-which is something I expect more from my personal email than a discussion forum…

        • bender
          Posted Apr 21, 2009 at 11:47 AM | Permalink

          Re: Andrew (#502),
          Phil linked to your comment in addressing the forum. He did nothing wrong. He simply asserted that PDO did not behave as a stable limit cycle. He never said that you said this, or that you were wrong. He helpfully pointed out that there are others out there making this assertion, even if you are not. He could have achieved the exact same effect, not addressing by saying:

          What Andrew is describing is a stable limit cycle, however the PDO doesn’t look like one.

  144. Hank
    Posted Apr 17, 2009 at 8:49 PM | Permalink

    Willem Kernkamp

    I don’t see where your statement that, “weather is inherently unstable, while climate is inherently stable” comes from. As a statement it has a nice symmetry, but what justifies anyone saying it? I would say that climate is just as unstable as the underlying weather it represents.

    I don’t think creating averages and putting things into a range makes it stable. My personal example of how unstable something like rainfall patterns can be would be represented by the water levels of Devil’s Lake, North Dakota. In 1940 it was almost dry, but weather patterns changed and now water is almost fifty feet higher. The lake is instructive of long term patterns because it is in an enclosed basin with little enough rainfall to over-top the natural outlet – in “typical” circumstances.

    Devil’s Lake water levels in historic times:

    http://nd.water.usgs.gov/devilslake/data/dlelevation.html

    Prehistoric water levels of Devil’s lake:

    http://www.swc.state.nd.us/4dlink9/4dcgi/GetContentPhoto/PB-42/640/480

    Models are only as good as our understandings of the processes being represented. If things are overlooked or estimates are off the models will be be wrong. I’m not an engineer but I’m pretty confident that most any model that an engineer constructs needs to be “put to the test” so that it can be assessed. I’m sure the trick to creating good models will mostly hinge on good detailed understanding of the kinds of things William Gray studies.

  145. Joseph Temple
    Posted Apr 17, 2009 at 11:08 PM | Permalink

    Further to “weather is unstable/climate is stable”: I have thought for some time now that there is some very questionable logic behind the assumed distinction of and consequent relationship between climate and weather. It seems that some folks regard climate as a real object which exists apart from weather rather than as a name for the history of weather. But what is real is what happens and what happens is weather. Over time, in some region, one kind of weather may be commonplace and usual and when it is we call that the local climate. Then we make an illogical inversion of cause and effect and offer that “climate” as an explanation of the day’s weather in the sense of proposing it as the cause of said weather. But it is not a cause, it is a summary. After all, how do we know when there has been change in the climate but by observing change in the weather? When a bear walks through the mud it leaves tracks and when we see those tracks we rightly postulate the existence of the bear and infer something about its size and weight and species. It does not follow that when we see weather we can postulate the existence of a “climate” and start inferring things about it from smoothed averages of various features of the weather. After all, what does it add to the discussion to disguise the fact that we are talking about measures made on the weather, and statistics computed on those measures, by pretending that we are giving a description of the climate? It just allows us to delude ourselves that weather is one thing and climate another and that we can still be right about the climate when we are utterly wrong about the weather.

  146. Stevo
    Posted Apr 18, 2009 at 4:11 AM | Permalink

    What do people mean by ‘stable’ here?

    We have:
    – Lyapunov stability – that all the trajectories close to a point of equilibrium stay close forever.
    – Exponential stability – the poles of the input-output function have strictly negative real parts. (Inside the unit circle for discrete-time systems.)
    – Structural stability – that all trajectories close to a point of equilibrium have the same topology.
    – Having a stable probability distribution – adding independent variables with this distribution gives an example of the same distribution (up to scale and location parameters).
    – Any others?

    Or do you mean something else, like the range of variation being bounded, or the time averages being a deterministic function of the inputs, or variance being zero below a particular frequency threshold?

    I’ve often heard this claim that if you average things over 30 years (why 30?) then any variation remaining must be deterministic and evidence of external forcing. I have often been told that you can’t determine anything from trends over a few years, but that you can from trends over a few decades. But nobody has ever managed to tell me how they know that decades are sufficient. Is this number derived somehow from the physics, or what?

    If NAO, PDO, and so on operate on timescales of decades, do they therefore have to be deterministic, and is there any possibility of further longer-term oscillations beyond our data horizon?

    • Andrew
      Posted Apr 18, 2009 at 4:42 AM | Permalink

      Re: Stevo (#30), 30 years is tradition! That is way climate science was done in old country, that is way we do it forever!

      (No, really, it is mere convention)

  147. David Cauthen
    Posted Apr 18, 2009 at 6:00 AM | Permalink

    -“weather is inherently unstable, while climate is inherently stable”-

    True on a human time scale. Not true on a geological time scale.

    • Craig Loehle
      Posted Apr 18, 2009 at 7:28 AM | Permalink

      Re: David Cauthen (#468), It is interesting that RC uses the argument that a few years of cooling is not “climate”, only 30 plus years would be, to dismiss recent cooling trends, but neglect the fact that the warming post-1950 (when IPCC says human influence is non-trivial) was mostly concentrated in about a 20 year period. Ah, to have one’s cake and eat it too.

    • Andrew
      Posted Apr 18, 2009 at 9:10 AM | Permalink

      Re: David Cauthen (#468), What? What on Earth makes you think that climate is unstable on a geological timescale? Big changes does not mean unstable! Jeez, sine wave anybody?

      • BarryW
        Posted Apr 18, 2009 at 9:32 AM | Permalink

        Re: Andrew (#470),

        If 30 years implies a stable climate, how do they explain where this roller coaster come from?

        Hadley trends

        • Andrew
          Posted Apr 18, 2009 at 9:57 AM | Permalink

          Re: BarryW (#471), Ugh! I don’t mean stable like “stable as a table” stable! I think what I mean is more like “rates of change never get arbitrarily close to infinite” but I definitely don’t mean flat.

        • BarryW
          Posted Apr 18, 2009 at 10:20 AM | Permalink

          Re: Andrew (#473),

          I agree, but my point is that if I started 30 yrs ago and kept measuring 30 yr trends I would be convinced that the trends were positive and increasing, but if I’d started about 1940 I’d be convinced that the trends were decreasing (at least until about 1970)! So a 30 yr trend doesn’t cut it (IMO) in terms of identifying where things are going.

        • Andrew
          Posted Apr 18, 2009 at 10:24 AM | Permalink

          Re: BarryW (#474), I agree to, which is why I have said 30 years ’tis nothing more than mere tradition.

        • BarryW
          Posted Apr 18, 2009 at 2:51 PM | Permalink

          Re: Andrew (#475),

          ’tis nothing more than mere tradition.

          And a flawed one. My original comment wasn’t directed at you, but in agreement with you (I asked how do “they” explain it not you). Sorry I was a little vague.

        • Andrew
          Posted Apr 18, 2009 at 3:45 PM | Permalink

          Re: BarryW (#476), Well, my suspicion is that it is something like how Ronald Fisher made 95% the gold standard for hypothesis testing-there is no real basis for it being the way to look at things, but everyone does it anyway. Sometimes it makes you want to scream “Just report your p values! Let the world judge.”-I am not sure what a similar cry for transparency and separation of reaching “conclusions” and data analysis would sound like, but I wager the stats guys here could come up with something.

      • David Cauthen
        Posted Apr 20, 2009 at 3:34 PM | Permalink

        Re: Andrew (#470),
        I was just referring to the fact that, on geological time scales, the earth’s climate ratchets back and forth between varying degrees of warmth warmth and frozen tundra. But I guess that could be some kind of sine wave, huh?

        • Andrew
          Posted Apr 20, 2009 at 3:45 PM | Permalink

          Re: David Cauthen (#489), Well, because that is probably controlled by Milankovitch cycles, it is sort of like a nonlinear combination of different “waves”.

  148. Ivan
    Posted Apr 19, 2009 at 10:13 AM | Permalink

    Steve, sorry for this off topic post (feel free to snip it or remove on another tread if you wish), but I hope it will not be quite useless.

    You said:

    I’ve therefore taken the position that this blog will deal with finite issues, focussing on verification and statistical analysis, and do not permit this sort of “big question” to enter specific threads, while having sympathy for people who wonder about it. I wonder about the “big questions” myself and regret that IPCC has not provided an engineering-quality A-to-B exposition of how doubled CO2 leads to 3 deg C.

    But, actually IPCC already did a couple of very interesting, controversial, “finite” and having in mind your auditing and investigating skills, quite verifiable, assertions. One of those assertions that easily could become subject to “verification and statistical analysis” is that ice core records show that today’s Co2 concentration is “unprecedented”‘ in the last 650 000 years and that it is a proof that Co2 buildup in XX century is man-made. This is main founding block of IPCC “consensus science”. But, many early (before 1985) ice-core measurements showed significantly higher values than contemporary, and “IPCC consensus view” of pre-Industrial 280 ppmv concentration came up only as a result of very doubtfull process of selection of the data, where higher values than 280ppmv were routinely discarded to fit political narative of man-made increase of CO2 (Calender, who is a Saint of global warming fraternity and main IPCC reference, actually rejected 16 out of 25 XIX century measurements in order to obtain politically correct answer of XIX century CO2 concentration. If he were not, average XIX century concentration would be 335 ppmv, and not 285-290 ppmv!). Before 1985 vast majority of ice core measurements of ancient Co2 content of atmosphere (Norway, Greenland, Antarctica) showed singificantly higher values that today, sometimes much, much higher. It was only after 1985 when ice cores began to show convinient picture of “no more than 280ppmv”. There is a huge peer reviewed literature mainly from until 1985 that documents main IPCC narative of pre-Industrial 280ppmv is simply false. Wouldn’t be interesting to explore that a little bit, maybe to invite eg Segalstad and Jaworowski, very prominent scientists who published a lot about the topic as a guest writers, and soem form IPCC “camp” as well (if you have no time enough to do all the work by yourself, just like you invited Leif Svalgard to handle Sun climate connection issue on CA last year)? This is not a “big picture” philosophizing, but very specific and “finite” question that happen to have very large big picture consequences.

    Sorry once more for being of topic but I try to understand your motivation to not discuss the topic which, according to your own many times repeated criteria is highly, supremely relevant for CA auditing. It is on one hand part of the official IPCC science, actually its main founding block (much more than Mann’s Hockey Stick) and on the other hand, very “finite” and “statistically and empirically” verifable issue.

    • Steve McIntyre
      Posted Apr 19, 2009 at 10:42 AM | Permalink

      Re: Ivan (#478),

      There are many relevant topics that editorially I discourage for a variety of reasons. I focus on studies relied upon by IPCC. I don’t permit discussion of these papers here. I’d prefer that you do so elsewhere.

  149. Ivan
    Posted Apr 19, 2009 at 11:07 AM | Permalink

    Steve, you haven’t read my post carefully. I am not talking about some “sceptic theory”, but about official IPCC theory. Ice core measurements are exactly studies “relied upon by IPCC”. I offered reasons why studying those issues is supremely important exactly according YOUR OWN criteria of a) theory being part of IPCC science, b) can be statistically and empirically verifiable (and not some “big picture” philosophizing). You give zero evidence that I am wrong, i.e. that I misrepresented your general policy or criteria.

    If you want to discuss hard core “skeptic” causes that I regard as being dubious, I’d prefer that you do so elsewhere.

    So, am I correct to conclude that your reason to not touch that hot potato is political or “political”, and not compliance with your own proclaimed criteria for relevance? You simply don’t want to be associated with “those guys”, so you are dissmissing any evidence in advance before actually examining it, despite you should examine it according to your own stated policy.

    Steve:
    If I were worried about being “political”, this blog wouldn’t exist. I’m not interested in discussing these articles here and am not interested in arguing about it. If you feel like discussing them, please do so elsewhere. No more on this please.

  150. Ivan
    Posted Apr 19, 2009 at 12:55 PM | Permalink

    Steve, it is not my intent to argue or even to try to convince you what to do about this issue. I was just curious why you didn’t explore this problem, although it perfecty fits YOUR OWN criteria for something to be relevant for CA. You didn’t have any plausible explanation exept that you didn’t want to explore this, what we knew ex ante. Your dismissive response (actually lack of any response) and ban to comment further on this (even on unthreade) is little disappointing for me.

    Maybe you think those theories are noth worth even disproving, but I doubt you could be so self-confident (with your famous Popperian falibility and e.g. reluctance to give any statement about whether MWP was warmer thatn today) or confident in what IPCC says, contrary to what majority of scientists before 1985 said. If you believed in such a way to IPCC, concerning paleo-climate, you never would question HS. I don’t know what is the basis to believe to IPCC in advance any more in this particular case, without actually exploring evidence?

  151. Ivan
    Posted Apr 19, 2009 at 12:56 PM | Permalink

    ‘Popperian falibilism” :)

  152. Posted Apr 22, 2009 at 9:25 PM | Permalink

    I’m just wondering why, out of the four resources I mentioned in my response to bender, this one paper was the only one to be discussed? Perhaps because it was the only one bender addressed. The Bratcher and Giese paper is also very interesting.

  153. bender
    Posted Apr 22, 2009 at 11:09 PM | Permalink

    PDO is not sufficiently oscillatory to be predictable. You say it is. I say prove it. You cite papers. I want arguments. You want me to review papers for you. I want you to review them for yourself. That two guys made a guess in 2002 that you say turned out to be correct in 2007 is not a proof of predictability. But you just don’t understand that, do you?

    I explained radiative imbalance. You simply don’t get it. You persist in this idea that if it’s not in the atmosphere it must be in the oceans. Nonsense. True: I said I would find the Hansen paper; but I haven’t had time. But I definitely recall it being cited somewhere. While you’re waiting, why not find it yourself?

    I provided the link you asked for. You just didn’t see it.

    • Raven
      Posted Apr 23, 2009 at 3:05 AM | Permalink

      Re: bender (#523)
      I am been doing some thinking about this “pipeline” issue.

      It sounds like there are two “conceptual models” that could explain the observations.

      The Jim Hansen conceptual model is like a house with a single heater where the heat output is rising constantly. If nothing else changes the house will warm until the heat loss balances the heat input. If someone opens a window the temperature will stop rising and may actually cool but once the window is closed the house will continue to warm at an accelerated rate because the heat output from the heater was rising even if the window was open.

      The Roy Spencer conceptual model is like a house with two heaters where one heater turns on and off at unknown intervals and one heater has constantly rising heat output. When both heaters are on the house will warm rapidly but when one heater is off it only warms at a rate depending on the heat output of the other heater.

      In the Jim Hansen conceptual model the output of the heater is assumed to be a known quantity and if the temperaure does not rise as expected then there must be a window open somewhere. In the Roy Spencer conceptual model the output of both heaters is an unknown quantity and if temperature does not rise as expected then that implies the output of the constantly rising heater must be lower.

      I think these models illustrate the point that Bender was making on the pipeline issue but Bender may (probably?) disagree.

      • bender
        Posted Apr 23, 2009 at 7:14 AM | Permalink

        Re: Raven (#527),
        This is not inconsistent with my understanding of their views. What it is incnosistent with is the interpretation proposed by Ron Cram.

    • Mark T
      Posted Apr 23, 2009 at 10:20 AM | Permalink

      Re: bender (#523),

      PDO is not sufficiently oscillatory to be predictable.

      Define “sufficiently.” Is that some standard that is relevant only to your views?

      For the record, there are several distinct peaks in an FFT of the raw PDO data (1308 points) that Phil. provided (the link is actually to the graphic). That qualifies as oscillatory and at the very least, your broad brush stroke is debatable if not outright incorrect.

      The first (largest) is in bin 2 representing approximately 55 years, the second (at -3 dB, or half power) is in bin 19 representing approximately 5.7 years, the third (at -9 dB, or 1/8th power) is in bin 110 representing 1 year (expected peak), and the fourth (at -13 dB, around 1/20th power) is in bin 219 representing approximately 6 months. There are others, but they are much smaller.

      Mark

      • bender
        Posted Apr 23, 2009 at 11:37 AM | Permalink

        Re: Mark T (#533),

        For the record, there are several distinct peaks in an FFT of the raw PDO data (1308 points) that Phil. provided

        Show me. You well know that “1608 data points” does not mean anything. It’s the number of “oscillations” that matters.

        • Mark T
          Posted Apr 23, 2009 at 11:56 AM | Permalink

          Re: bender (#535), 1308 points, and I already stated which bins the peaks reside in the resultant FFT. If you cannot understand Fourier theory, it is not my job to explain it to you.

          Re: bender (#536), Drop the arrogance. On more than one occasion I have legitimately refuted your arguments and yet you continue. So far your entire argument is 1) a paper is not evidence, 2) all tree ring reconstructions are necessarily wrong, and 3) there is no “sufficient” oscillatory behavior in the PDO for prediction. Argument 1 is simply splitting hairs, 2 is flat out incorrect, and 3 is also incorrect as I have demonstrated.

          Mark

        • bender
          Posted Apr 23, 2009 at 12:06 PM | Permalink

          Re: Mark T (#537),

          Drop the arrogance

          Good advice again. What makes you think I don’t know what FFT is?

          Have you read the Bratcher & Giese paper? It has NOTHING TO DO WHATOSOVER with predictability of PDO. That’s why Ron’s all crickets.

          I pre-empted your verbal description of the FFT of PDO by admitting the spectrum is not white. So your additional notes add nothing.

          What makes you think your FFT is stable? How many oscillation in your time-series?

          Is it being an ass to ask for proof of assertions? I think not.

        • Mark T
          Posted Apr 23, 2009 at 12:16 PM | Permalink

          Re: bender (#539),

          What makes you think I don’t know what FFT is?

          ? You said:

          Show me. You well know that “1608 data points” does not mean anything. It’s the number of “oscillations” that matters.

          after I said:

          The first (largest) is in bin 2 representing approximately 55 years, the second (at -3 dB, or half power) is in bin 19 representing approximately 5.7 years, the third (at -9 dB, or 1/8th power) is in bin 110 representing 1 year (expected peak), and the fourth (at -13 dB, around 1/20th power) is in bin 219 representing approximately 6 months. There are others, but they are much smaller.

          I clearly stated which bin the FFT was in then pointed out what period that appeared as. Technically, I provided redundant information. Knowing the FFT was 1308 points, the data are monthly (Phil.’s data link), and what bin each peak is in is sufficient information to know how many cycles appear in the record. Again, it is not my job to explain to you why this is. Perhaps if you had not been so antagonistic, I would oblige you.

          I pre-empted your verbal description of the FFT of PDO by admitting the spectrum is not white. So your additional notes add nothing.

          That it is not white is immaterial. That there are clear peaks in the record is. You simply do not understand what “my additional notes” add because you do not properly understand the theory. Nothing wrong with that, but do not arrogantly proclaim you do when you cannot even decipher the basic parts.

          What makes you think your FFT is stable? How many oscillation in your time-series?

          Figure it out for yourself. Twice now, I’ve provided you with all the information you need.

          Is it being an ass to ask for proof of assertions? I think not.

          No, it is the manner in which you are “asking.” Demanding, actually, is a better word, and your “refutations” are nothing other than declarations of your opinion, based on nothing factual.

          Mark

        • bender
          Posted Apr 23, 2009 at 12:31 PM | Permalink

          Re: Mark T (#541),
          How many oscillations in the time-series, Mark T? I want to hear you say it. I already know what the answer is. Just as I already knew what the spectrum looked like before getting into this issue.
          .
          Algebra is not statistics. Engineers hate that.

        • Mark T
          Posted Apr 23, 2009 at 12:38 PM | Permalink

          Re: bender (#545),

          How many oscillations in the time-series, Mark T? I want to hear you say it. I already know what the answer is. Just as I already knew what the spectrum looked like before getting into this issue.

          I don’t know, bender, ~55 year cycle and 109 years. ~5.7 year cycle and 109 years. What do you think, bender?

          Here’s a test, take the significant peaks, which are meaningful even if you don’t wish to believe so, and inverse FFT. It’s an interesting view.

          Algebra is not statistics. Engineers hate that.

          I’m a statistical signal processing engineer, actually. Maybe you just hate that someone else knows how to analyze data, too?

          Mark

        • bender
          Posted Apr 23, 2009 at 12:41 PM | Permalink

          Re: Mark T (#547),
          Is the number smaller than 1608?

        • Mark T
          Posted Apr 23, 2009 at 12:49 PM | Permalink

          Re: bender (#548), First of all, the number of points is 1308. Twice I’ve made this distinction. Second of all, a 1308-point time series cannot have 1308 oscillations, it can only have half that. If you understood the theory, you would understand why.

          Re: bender (#550), I’ve already provided the proof. You don’t understand it. You keep proving your point about the donkey that I made earlier.

          Mark

        • bender
          Posted Apr 23, 2009 at 12:58 PM | Permalink

          Re: Mark T (#551),

          a 1308-point time series cannot have 1308 oscillations

          The strawest of all men, as I never said this.
          .
          Tell us what you think the significance of that number is. Why did you cite it, rather than the far more relevant number – the number of low-frequency “oscillations” in the series? It’s a curious choice. Convenient. But deceiving to some.

        • Mark T
          Posted Apr 23, 2009 at 1:01 PM | Permalink

          Re: bender (#554),

          The strawest of all men, as I never said this.

          No, you said:

          Is the number smaller than 1608?

          Of course the number is smaller than 1308 because it MUST be smaller than 654 by definition.

          Mark

        • Mark T
          Posted Apr 23, 2009 at 1:03 PM | Permalink

          Re: bender (#554),

          Tell us what you think the significance of that number is. Why did you cite it, rather than the far more relevant number – the number of low-frequency “oscillations” in the series? It’s a curious choice. Convenient. But deceiving to some.

          It is not “curious” at all, it is the largest, which I stated. The peaks I chose to mention were merely the ones that stood out from the surrounding area, something I also explained.

          Do you have a legitimate argument to make? Really, do you?

          Mark

        • bender
          Posted Apr 23, 2009 at 1:08 PM | Permalink

          Re: Mark T (#558),

          Do you have a legitimate argument to make? Really, do you?

          You’re right. I have nothing to say. Carry on. PDO is as sinusoidal as Mark T’s wobbling bridges.

        • Mark T
          Posted Apr 23, 2009 at 1:13 PM | Permalink

          Re: bender (#559),

          You’re right. I have nothing to say. Carry on. PDO is as sinusoidal as Mark T’s wobbling bridges.

          If you actually read what I said you would understand, but apparently you have not. I clearly said that your position that it is not “sufficiently” oscillatory (whatever that means) is debatable at best and outright incorrect at worst. There are clear oscillations. Whether that continues over all time I cannot say, nor did I say. But it did continue for 109 years well enough to be the dominant component in the entire series.

          Stick to what I said, bender, not what you think I said.

          Mark

        • bender
          Posted Apr 23, 2009 at 1:21 PM | Permalink

          Re: Mark T (#562),

          Stick to what I said, bender, not what you think I said.

          Except that this is not about you. It is about “predictability” of PDO, as asserted by several here. Because I reference your posts does not mean I am strictly addressing you. (This is not your inbox. It is a discussion forum. Oops, I’m being an ass.)
          .
          PDO phase “switches” – if “phases” even exist (if PDO even exists!) – are not predictable. If you think otherwise, prove it. Ron says he had a citable proof. I showed where he was wrong. Moreover I showed where he was seriously distorting what the authors said. I welcome any rebuttal he can come up with.
          .
          If the instrumental data are sufficient to prove PDO is robustly oscillatory, then why did the dendroclimatologists feel compelled to get in on the act?

        • Mark T
          Posted Apr 23, 2009 at 1:39 PM | Permalink

          Re: bender (#564),

          Because I reference your posts does not mean I am strictly addressing you.

          But you were addressing me, bender, and my points were very limited in scope. Namely, is there an oscillation in the data: yes, there is. I have not answered anything regarding predictability intentionally. Had you at least acknowledged that, I would not have replied the way I did. Same with Phil. and his reply to Andrew earlier which clearly implied he was not addressing a larger audience and he took specific exception with something Andrew had said.

          PDO phase “switches” – if “phases” even exist (if PDO even exists!) – are not predictable.

          They may not be, hence my comment on “debatable” earlier, i.e., there is an oscillation, but there is no guarantee that it will continue along that path. No, I do not think 2 cycles that represent 17% of the signal power* will allow you to say it will any certainty (quantifiable), not without some other information. The 5.7 year cycle has many more (obviously), but they only represent a few percent of the total power in the signal, making them even harder to make claims about, i.e, they may only exist in any quantifiable form for a small portion of the record.

          I never defended their positions, either way, just pointed out flaws in your criticism, which appears to be more personal than it should.

          Mark

          *17% if you take the neighbor bins on either side of the peak to determine the power contribution since the signal is not quite centered in the bin, 11% otherwise.

        • bender
          Posted Apr 23, 2009 at 1:47 PM | Permalink

          Re: Mark T (#566),

          there is an oscillation, but there is no guarantee that it will continue along that path

          Define “is”? :)
          .
          There WAS a fluctuation that resembled an oscillation. That does not mean there WAS an oscillation or that the pattern will continue. If the process is ergodic there is hope. Alas … proof?

        • Mark T
          Posted Apr 23, 2009 at 1:56 PM | Permalink

          Re: bender (#568),

          That does not mean there WAS an oscillation or that the pattern will continue.

          Actually, it means there was an oscillation, but it does not mean it will continue. An oscillation does not need to continue for all time to be considered an oscillation. That was sort of the point of using wavelets for analysis purposes when they gained popularity in the 90s. Indeed, there is really no such thing as a “pure sinusoid” since that would require infinite energy and is thus impossible. But I digress…

          That’s me putting the “sufficiency” question back to you.

          I think a better way to describe your analogy would be to ask “if you only observed it for 10 seconds and there were two clear cycles (oscillations)…” ;)

          The “is” joke was baaaaad, bender, ahem…

          Mark

        • bender
          Posted Apr 23, 2009 at 2:05 PM | Permalink

          Re: Mark T (#570),
          If there was an “oscillation”, explain its cause. In the case of the bridge in the crosswind, the cause is clear. In the case of some arbitrarily specified eigenthingy in the ocean … ??? Raven’s “echoes of the past”? Sorry – that’s a bit too mysticotelconnective for me.

        • Mark T
          Posted Apr 23, 2009 at 2:17 PM | Permalink

          Re: bender (#571),

          If there was an “oscillation”, explain its cause.

          Don’t know, and have made it clear that I don’t care*. Nowhere in the definition of “oscillation” is it required that I, or you, or anyone, know what the cause is for it to exist. The principal of causality says there is a cause (even for random processes), but doesn’t provide specific hints as to what they are.

          Mark

          *I don’t care for purposes of my original point. In the long run, I do care. I don’t buy the “PDO caused it” argument since I tend to view the PDO as an effect of something else.

        • Posted Apr 24, 2009 at 6:10 AM | Permalink

          Re: Mark T (#574),

          It is fine to look at the PDO as being caused by something else, even if unknown. However, that does not mean the PDO is not the more immediate cause of the climate regime shift. In the medical field, we talk about biochemical cascades. At the point of trauma, one event happens which causes another which causes another and so on. The cascade may be interrupted by a medical intervention at any point along the process.

          In a chaotic system like climate, natural processes can and do intervene and cause a PDO phase to lengthen or shorten. But the fact the PDO is going to shift back to its opposite phase at some point is or should be obvious to all. When that shift is going to happen and whether it is predictable with any helpful degree precision is an important topic. Most likely, the “cause” of the PDO is a combination of factors involving the solar cycle, resulting ocean heat content, cloud formation and other oceanic oscillations. But the fact is we do not know neither enough about our climate system to begin to understand how these factors work together to lengthen or shorten the PDO’s cycle. This is one of the reasons the concept of using a GCM to predict climate 100 years from now is completely ludicrous.

        • bender
          Posted Apr 24, 2009 at 7:47 AM | Permalink

          Re: Ron Cram (#585),

          the fact the PDO is going to shift back to its opposite phase at some point is or should be obvious to all

          More bull. PDO is an abitrarily defined EOF (see unthreaded #33). It is not defined by physics. Under your assertion that climate is chaotic, what guarantee is there that these EOFs aren’t constantly spatially re-organizing themselves? “None”. That is the correct answer.
          .
          Things that you say “should be obvious to all” are not at all obvious to me.

        • Mark T
          Posted Apr 24, 2009 at 10:54 AM | Permalink

          Re: Ron Cram (#585),

          But the fact the PDO is going to shift back to its opposite phase at some point is or should be obvious to all.

          Only if the process is stationary and only if there is not some other, dominant modulation that we do not have enough data to discern. For the former point, a lack of stationarity implies that the PCs will change over time, which means any oscillatory behavior may not always be present, or at least, it may change from one oscillation to another. Can you say for certainty the 55-year peak in the FFT will remain? The latter point is trivially understood.

          Mark

        • Posted Apr 24, 2009 at 6:31 AM | Permalink

          Re: Mark T (#574),
          Re: bender (#580),

          Regarding the origin of the PDO, you might find this abstract interesting:

          Abstract The Pacific decadal and interdecadal oscillation (PDO) has been extensively explored in recent decades because of its profound impact on global climate systems. It is a long-lived ENSO-like pattern of Pacific climate variability with a period of 10–30 years. The general picture is that the anomalously warm (cool) SSTs in the central North Pacific are always accompanied by the anomalously cool (warm) SSTs along the west coast of America and in the central east tropical Pacific with comparable amplitude. In general, there are two classes of opinions on the origin of this low-frequency climate variability, one thinking that it results from deterministically coupled modes of the Pacific ocean-atmosphere system, and the other, from stochastic atmospheric forcing. The deterministic origin emphasizes that the internal physical processes in an air-sea system can provide a positive feedback mechanism to amplify an initial perturbation, and a negative feedback mechanism to reverse the phase of oscillation. The dynamic evolution of ocean circulation determines the timescale of the oscillation. The stochastic origin, however, emphasizes that because the atmospheric activities can be thought as having no preferred timescale and are associated with an essentially white noise spectrum, the ocean response can manifest a red peak in a certain low frequency range with a decadal to interdecadal timescale. In this paper, the authors try to systematically understand the state of the art of observational, theoretical and numerical studies on the PDO and hope to provide a useful background reference for current research.

          On the decadal and interdecadal variability in the Pacific Ocean by Yang Haijun1 and Zhang Qiong.

        • bender
          Posted Apr 24, 2009 at 7:35 AM | Permalink

          Re: Ron Cram (#586),
          Like I told you, the causal mechanism is not currently known.

        • bender
          Posted Apr 23, 2009 at 2:15 PM | Permalink

          Re: Mark T (#566),

          there is an oscillation, but there is no guarantee that it will continue along that path. No, I do not think 2 cycles that represent 17% of the signal power* will allow you to say it will any certainty

          Then I’ve made my point and we agree on everything that matters. Now it is up to others on this thread to state whether they’ve changed their view any in light of your analysis.

        • Mark T
          Posted Apr 23, 2009 at 2:28 PM | Permalink

          Re: bender (#573),

          Then I’ve made my point and we agree on everything that matters.

          Maybe. I still don’t buy your one paragraph dismissal of the Biondi paper. I’d like to say I don’t have time to read it and thus make up my own mind, but here I sit typing away… ahem. I did read the intro and it seemed they weren’t approaching the tree rings in the same manner as the Team has.

          Mark

        • bender
          Posted Apr 23, 2009 at 2:47 PM | Permalink

          Re: Mark T (#575),
          I still don’t buy your one paragraph dismissal of the Biondi paper.
          Fair enough. That can be diucussed. I was prepared to discuss it then … more than a year ago.

        • Mark T
          Posted Apr 23, 2009 at 3:05 PM | Permalink

          Re: bender (#576),

          Fair enough. That can be diucussed. I was prepared to discuss it then … more than a year ago.

          I stopped posting in that thread after post 258 and the Biondi paper came up much later. In fact, your post (340) on April 18 was the day of my defense. I was a bit scatterbrained at that time, to put it mildly.

          Mark

        • bender
          Posted Apr 23, 2009 at 12:44 PM | Permalink

          Re: Mark T (#547),

          I’m a statistical signal processing engineer

          Great. You’re a somebody and Ron knows how to cite papers. Can we get past credentials and appeals to authority and start talking about proof?

        • Mark T
          Posted Apr 23, 2009 at 12:51 PM | Permalink

          Re: bender (#550), And, for the record, my reply was in response to your clear argumentum ad-hominem. It’s looking like I’m actually better at dissecting fallacious arguments. Indeed, why don’t we get past them and see what you can actually add that is relevant?

          Mark

        • bender
          Posted Apr 23, 2009 at 1:02 PM | Permalink

          Re: Mark T (#552),

          your clear argumentum ad-hominem

          It was not an attempt to dismiss or discredit you. I like you. It was a challenge to you to admit that a mere data transformation does not answer the question. A 55-year cycle in a 109-year series? What would your boss say to that?

        • Mark T
          Posted Apr 23, 2009 at 1:09 PM | Permalink

          Re: bender (#557),

          It was not an attempt to dismiss or discredit you.

          The implication was that I do not understand statistics. I do. This has nothing to do with statistics only “is it oscillatory.” Over the 109 year period, yes, it is. In fact, there are two that show up across the whole series, drowned in red noise centered around DC.

          I like you.

          Normally I would say the same about you, but lately it seems you’ve turned from critical, to jaded, and have taken on crusades that seem more personal than investigatory.

          A 55-year cycle in a 109-year series? What would your boss say to that?

          He’d say: “it looks like you have 2 cycles there: what do you think caused them?” And I’d reply: “Don’t know as I have not looked into it!” Quite frankly, at the moment, I do not care.

          All I’ve claimed is that the cycles are there, and they stand out. This one in particular is the largest component of the whole series (and is even visible in the time series).

          Mark

        • bender
          Posted Apr 23, 2009 at 1:13 PM | Permalink

          Re: Mark T (#560),

          This has nothing to do with statistics

          This is where your eyes may need some opening.

        • Mark T
          Posted Apr 23, 2009 at 1:14 PM | Permalink

          Re: bender (#561),

          This is where your eyes may need some opening.

          You still fail to understand my point, bender.

          Mark

        • bender
          Posted Apr 23, 2009 at 1:43 PM | Permalink

          Re: Mark T (#563),
          You have a 109-year long sample from a huge, possibly non-ergodic, ensemble of possibilities. Let’s start there …
          .
          Suppose you have a model of a suspension bridge amd you subject it to a 100mph crosswind. How long do you need to observe that simulation experiment before you know if your bridge will blow apart from harmonic oscillations? 10 seconds? 10 minutes? 10 hours? That’s me putting the “sufficiency” question back to you.

        • bender
          Posted Apr 23, 2009 at 1:55 PM | Permalink

          Re: Mark T (#560),

          lately it seems you’ve turned from critical, to jaded, and have taken on crusades that seem more personal than investigatory

          You have a point. And the reason is because I see a bunch of propaganda being circulated about how Earth’s background climate variability is “oscillatory” and “predictable” and “stable” with a “30-year cycle”, that this patterns stems from the “PDO”, that it means “global warming has abated” for 20+ years etc. A lot of the talk – coincidentally – that I am seeing repeated on this thread. Hmmm. All I ask for is proof. Show me. And don’t give me links to papers unless you are willing and able to discuss those papers in full.

        • Mark T
          Posted Apr 23, 2009 at 2:10 PM | Permalink

          Re: bender (#569), Understood, but you don’t need to be as antagonistic about it as you have been. Of course, I suppose I’ve been equally, if not more so, antagonistic in my responses. I guess I’m just looking out for the well-meaning, but thin-skinned of us?

          Part of the problem with analyzing what is arguably a chaotic system is that there can also be underlying oscillatory behavior that really is there. A quick analysis on what data we do have suggests some oscillation is there, but we can’t say more than that (and at 17% power, it does not dominate, though it is the single largest component).

          Mark

        • bender
          Posted Apr 23, 2009 at 12:53 PM | Permalink

          Re: Mark T (#547),

          the significant peaks, which are meaningful even if you don’t wish to believe so

          That you state the 55-year peak to be “meangingful” does not make it so. Significance is a matter of statistics. Meaningfulness is a matter of physics.
          .
          That you can algebraically transform data as well I can does not mean you are any closer to understanding what causes the pattern than I am. You can’t tell me the pattern is robust until you can explain to me what causes it. Certainly a couple cycles worth’ of data is not a sufficient basis for inference. Hence my earlier claim about “sufficiency” – which you called into question.
          .
          Your exaggerations on the effective size of the data set are noted.

        • Mark T
          Posted Apr 23, 2009 at 1:00 PM | Permalink

          Re: bender (#553),

          Significance is a matter of statistics. Meaningfulness is a matter of physics.

          The 55-year peak is the largest component in the entire record by more than 3 dB. It is not only significant, it is dominant throughout the entire series.

          You can’t tell me the pattern is robust until you can explain to me what causes it.

          I don’t need to. All I said is that is a clear oscillation, and it is. As is the 5.7 year oscillation though it represents half the power. I have not made any attempts at inferring what causes it, or what it means, just that it is there.

          Your exaggerations on the effective size of the data set are noted.

          Exaggerations? The data record I analyzed is 1308 monthly points. There is no “exaggeration,” just a simple statement of fact.

          Mark

  154. nevket240
    Posted Apr 23, 2009 at 1:02 AM | Permalink

    http://www.reuters.com/article/environmentNews/idUSTRE53M00N20090423?pageNumber=1&virtualBrandChannel=0

    any comments on this article. from my understanding of the subject, limited, they may have it A-about. They are essentially saying that climate change will cause, climate change. Or have they got it backwards??
    regards

  155. Geoff Sherrington
    Posted Apr 23, 2009 at 2:25 AM | Permalink

    Hard to know where to post this.

    Report from Down Under, April 23, 2009.

    1. P.J. O’Rourke of the Cato Institute etc, visited Canberra and reported that –

    “The killjoys are back in charge. The mopes, the fusstails, the glumpots. You listen to them, you know: the world is experiencing a financial and polar icecap meltdown causing sea levels to rise, sending cold water flooding down Wall Street, where the rapidly acidifying ocean currents are corroding our retirement accounts and releasing collateral debt obligations full of hot air into the atmosphere until every breath we take is full of CO2. Especially when we exhale, and exhaling should be banned when children are present lest their uninsured healthcare be damaged by second-hand greenhouse gases that are causing endangerment of plants and animal species – America’s Republicans are already extinct – and a shortage of green leafy vegetables vital to the fight against growing epidemics of obese hunger and housing foreclosure on the homeless.” And more. “The Australian”, national newspaper, reported by Christopher Kerr. Page 2.

    2. Skeptic Prof Ian Plimer’s book “Heaven and Earth: Global Warming and the Missing Science” released 2 days ago has gone from first printing of 5,000 to third printing of 5,000, very high for here. Plimer said “The average punter out there feels helpless and disenfranchised”. Ian took several hundred pages to say much the same as P.J. above, but with more scientific terms. Hi Ian. His debating AGW opponent claimed Ian “had only used selective evidence when quoting more than 200 scientists from peer-reviewed papers”. I understand that the Tooth Fairy was not mentioned.

    3. “The Australian” newspaper again, page 3, by Greg Roberts. “Sea ice around Antarctica has been increasing at a rate of 100,000 sq km a decade since the 1970s”. Ref to GRL, British Antarctic Survey, who seem anally fixated with the ozone hole. “Cooling has been recorded at the Australian bases in and elsewhere in east Antarctica.” In reply, BAS project leader John Turner avoided excommunication by quoting that “Satellite images indicated the ozone layer had strengthened surface winds around Antarctica, deepening storms in the South Pacific area of the Southern Ocean. This had resulted in a greater flow of cold air over the Ross Sea, leading to more ice production”.

    4. The third point is really quite funny when analyzed logically. You see, it is not the Antarctic that is cooling, it is the winds that blow over it that are getting cooler. It is not easily explained how, if regional cooling is not happening, these coolest of all winds on earth actually get cooler under Global Warming. The logical obverse is that the Earth is not heating up under Global Warming, it’s just the air that blows over the surface that is warming. Then, to retain membership of the Club, Turner concludes “We expect ozone levels to recover by the end of the century and by then there is likely to be around one-third less Antarctic sea ice.” How on Earth can such a prediction be made? Remember that the Vostok core for 600,000 years reports no period of land ice melting to produce what in sedimentary geology is called an “unconformity”. Ergo, no sea level change from this mechanism.

    5. David Stockwell and Miklós Zágoni gave lectures at Newcastle Uni on 15 April. Each produced strong and logical arguments which can be read at http://landshape.org:80/enm/newcastle-lecture-update/

  156. Mark T
    Posted Apr 23, 2009 at 9:52 AM | Permalink

    Give it a rest and stop being an ass. Providing a paper is about the only sort of “evidence” one can really do on message board thread. There is inherent data included in the paper, too. If your only legitimate argument is the distinction between “evidence” and a paper you’re really reaching. I could equally claim that your argument about why the paper should be dismissed is specious at best. Simply saying “I’ve already declared the paper bad” is pretty thing.

    Mark

    • bender
      Posted Apr 23, 2009 at 11:33 AM | Permalink

      Re: Mark T (#531),

      Give it a rest and stop being an ass.

      sound advice. try following it.

    • bender
      Posted Apr 23, 2009 at 11:46 AM | Permalink

      Re: Mark T (#531),

      Providing a paper is about the only sort of “evidence” one can really do on message board thread.

      Ass talk more characteristic of real climate than climate audit. It is very easy to recapitulate someone’s argument if you’ve read and understood the paper.

      This is the second time in this thread you’ve tried to start a food fight by accusing someone of being an ass.

  157. Mark T
    Posted Apr 23, 2009 at 9:53 AM | Permalink

    Thin should be the last word, not thing.

    Mark

  158. Mark T
    Posted Apr 23, 2009 at 12:01 PM | Permalink

    And, for the record, I also stated what the periods were in (#533). So that makes point 4) of your argument “what is the number of oscillations” incorrect as well. No, it is not I that is behaving arrogantly.

    Mark

  159. bender
    Posted Apr 23, 2009 at 12:11 PM | Permalink

    Mark T: How many times did you encounter the phrase “PDO” when you read Bratcher & Giese?
    Ron Cram: Same question.

    • Mark T
      Posted Apr 23, 2009 at 12:18 PM | Permalink

      Re: bender (#540), I have not read it, nor have I commented on it. For that matter, nor have I defended any argument of predictability nor any of Ron’s or Andrew’s specific points. I merely pointed out the problems with your argument, which has been, up till now, specious. Your comments regarding this particular paper, in regards to me, are nothing but a strawman.

      Mark

  160. bender
    Posted Apr 23, 2009 at 12:26 PM | Permalink

    From the paper:

    Several possibilities exist as to why the 1976 climate regime shift occurred and what could trigger a shift back to pre-1976 conditions. These include, but are not limited to, a multi-equilibria system, and the synchronous phase of ENSO, decadal variability, and interdecadal variability [Luo and Yamagata, 2001]. The purpose of this paper is to bring attention to the similar, but opposite in sign, pattern as that seen prior to the 1976 climate shift and emphasize the importance of understanding the separation between natural variability and anthropogenic forcing in the climate system.

    i.e. They did not analyse PDO pattern, did even dare to make a prediction, let alone provide a mechanism. So citing this paper as proof that PDO is “sinusoidal”, “oscillatory”, “predictable”, “bistable”, or “stable” is as fanciful as the team saying that MWP was caused by an NAO-like MCA.

  161. DeWitt Payne
    Posted Apr 23, 2009 at 12:41 PM | Permalink

    AMO index data in tabular form from 1856 to 2009 is here if anyone’s interested.

  162. bender
    Posted Apr 23, 2009 at 1:33 PM | Permalink

    I clearly said that your position that it is not “sufficiently” oscillatory (whatever that means) is debatable at best and outright incorrect at worst

    I heard you the first time. It is fair to ask me this question. It is fair to ask the same question to those who suggest that prediction of phase switches is possible four years in advance. If they think their predicitons are better than a coin toss, let them state why.
    .
    The future is not the past.

  163. Andrew
    Posted Apr 23, 2009 at 5:05 PM | Permalink

    Jesus, guys, cut it out, you are ruining the thread for the rest of us. Take it to the forum.

  164. Mark T
    Posted Apr 23, 2009 at 6:15 PM | Permalink

    It’s kinda done if you hadn’t noticed.

    Mark

    • bender
      Posted Apr 23, 2009 at 8:03 PM | Permalink

      Re: Mark T (#579),

      It’s kinda done if you hadn’t noticed.

      Exactly. And a happy ending I might add.

      Re: Andrew (#578),

      you are ruining the thread

      1. Ruining what “thread”? Unthreaded? Get real.
      2. Your idea of ruination is bizarre. It is not allowed to ask for proof and quotes and full-formed arguments when someone starts spouting questionable statements backed by nothing but linkies? That’s teamstyle.
      3. If anyone is ruining anything it is your butting in and adding nothing.
      .
      Apologies, all, but this *will* pick up again once we get a rational response from Ron Cram. Who should have taken his initial nonsense (“PDO is predictable”) to the bulletin board, according to Andrew’s policy.
      .
      Let the record show that Ron’s paper does not use the phrase PDO once. That’s why we’re hearing more crickets after I asked for a direct quote instead of his partial, decontextualized snippet of a quote.

      • Posted Apr 23, 2009 at 9:55 PM | Permalink

        Re: bender (#580),

        Okay, bender, I just got back to the computer. I do have a life, ya know. I saw your request above for a quote about the PDO switching phases “in about four years.” It has been three years since I read the paper and my memory was faulty.

        Since subsurface temperature in the south tropical Pacific started to decrease in the early 1990’s, and if negative anomalies follow a pattern similar to the positive anomalies in the mid–1970’s, then a cool tropical Pacific SST anomaly may soon weaken the global warming signal.

        How soon?

        The analysis also suggests that low frequency changes in the tropical Pacific lead changes in global
        surface air temperature by about 4 years.

        You will argue the four year time lag is not the PDO and you are correct. I misremembered the paper. But they did predict a climate shift “soon” and they were correct.

        You act as if the paper is not about the PDO at all. Well, the paper mentions “oscillating” twice, uses “variability” as a synonym repeatedly, and uses the term “climate shift” or “climate regime shift” for the same time period when the PDO last shifted. The time scale is decadal, so this is not ENSO which has a much shorter oscillation. The subject of the paper is the southern tropical pacific sea and subsurface temperature. So it is either talking about the PDO or the IPO. Perhaps the authors see the two as connected and did not attempt to separate them.

        The retrospective analysis indicates that the conditions present in the southern tropical Pacific resemble those prior to the 1976 climate shift, except with the opposite sign. If tropical Pacific SST responds to these subsurface changes in a similar way, then it could be an indication of a climate regime shift to pre-1976 conditions. Given the considerable effect that tropical Pacific SST has on global atmospheric circulation, a climate shift to pre-1976 conditions could lessen the warming trend that has existed since 1976. A similar situation existed in the early 1940’s when SST records show an equatorial Pacific cooling with the period from 1942–1976 generally cooler than the period following the 1976 climate shift [Zhang et al., 1997].

        This “climate regime shift” took place in 1976 when the PDO shifted from the cool phase to its warm phase. The authors were also aware 1942-1976 was a cooler climate regime, a time when the PDO was in its cool phase. When the paper is discussing Oscillations that happen in the Pacific on Decadal time scales, it seems silly to me to claim the paper is not about the PDO.

        • bender
          Posted Apr 24, 2009 at 7:39 AM | Permalink

          Re: Ron Cram (#581),

          You will argue the four year time lag is not the PDO and you are correct. I misremembered the paper.

          I’m glad for the admission of error. Hopefully you will be more careful not to spread misinformation in the future.

      • Posted Apr 23, 2009 at 10:30 PM | Permalink

        Re: bender (#580),

        The last time you asked for evidence, I just gave you a few things off the top of my head. But your stubbornness has caused me to go to Google Scholar for more. I am not saying I have read all of these papers, but I did scan through looking for quotes which may bear on this discussion.

        Recent advances in understanding mechanisms for persistence and slow changes in extratropical SST anomalies offer improved confidence for PDV predictability at lead times of one to a few years. Accurate PDO monitoring and prediction may have practical benefits in both seasonal and longer term climate forecasts for select regions. Gershunov and Barnett (1998), for example, argued that combining PDO and ENSO information may enhance the skill of empirical North American climate forecasts.

        From The Pacific Decadal Oscillation

        Thus, the similarity between the theoretical and the actual forecast errors provide a measure of the
        underlying dynamical assumptions used in the prediction model, i.e. that SST anomalies in the Pacific can be well described as a stable linear process driven by additive Gaussian white noise (Penland and Matrosova 2001, Newman et al. 2003a).

        From Forecasting Pacific SSTs: Linear Inverse Model Predictions of the PDO

        One of the leaders in PDO prediction seems to be Michael Alexander. His publication record is here.

        This is offtopic but while searching I found this paper on how the PDO and ENSO interact in east Asia. The fact they interact is central to my understanding of oceanic oscillations. El Ninos are way stronger when the PDO is in its warm phase. La Ninas will be much cooler when the PDO is in its cool phase.

        Interdecadal modulation of PDO on the impact of ENSO on the east Asian winter monsoon

        • Geoff Sherrington
          Posted Apr 24, 2009 at 2:09 AM | Permalink

          Re: Ron Cram (#582),

          These arguments often deal mostly with northern hemisphere topics when there’s a bloody great lump of ice in the Antarctic.

          Also, having laboured throught the last 50 or so posts, I agree with bender that one should not fixate on climate change candidates unless the offering is both necessary and sufficient.

          There is a comical list floating around that lists several hundred disasters attributed to Global Warming. Above, we have a shorter list of candidates that might cause climate change. Both ought to pass a threshhold of causation before speculation. Separately, I have derived some observations of past local climate that seem to follow a pattern, but I am not about to publicise them – they do not reveal a cause.

        • Posted Apr 24, 2009 at 5:47 AM | Permalink

          Re: Geoff Sherrington (#583),

          Actually Geoff, they deal with south tropical pacific ocean oscillations and the climate regime shifts that happen at the same time. Whether these events and the resulting climate regime shifts are cause and effect or something else is a subject of discussion. Also, whether the PDO shifts can be predicted is a subject of discussion. You can either read the papers and join the discussion or you can pick a side without adding anything to the discussion, your choice.

        • Geoff Sherrington
          Posted Apr 24, 2009 at 6:46 PM | Permalink

          Re: Ron Cram (#584),

          You would have to agree that the geographic coverage of the PDO topic in publications is weighted heavily to the NH and a bit S of the Equator. There is a huge amount of ocean between that and the Antarctic. Even the albatross does not cross over – though it now hangs around the neck of many people who were too confident too soon.
          (This is a literary reference to Coleridge, “The Rime of the Ancyent Marinere”)

          I’m not noted for picking sides. I’m an old-fasioned scientist who works on the hypothesis/experiment/interpret/cause and effect type of approach. When I do pick, it is to reject weak correlations unless or until they can be shown to be causative beyond reasonable doubt and useful to expand real future knowledge. This often implies a need to show prediction (not the same as projection) and/or replication.

          A point of mentioning the comic list of tragedies caused by Global Warming is to show that there is huge internal inconsistency. Diminishing returns apply to each additional speculation.

          That’s why I like Steve’s objective Climate Audit.

        • bender
          Posted Apr 24, 2009 at 7:29 AM | Permalink

          Re: Ron Cram (#582),
          Ron, the discussion is over. Yes, PDO predictability would be fantastic news. Same with ENSO. As you are learning, there is currently no predictive accuracy. Certainly not 4 years ahead of time, as you were pretending was the case. Not even one year. Please read the papers you are quoting and try to understand them. Otherwise you are nothing but a source of misinformation. Which is a Bad Thing.

      • Barclay E. MacDonald
        Posted Apr 26, 2009 at 2:14 AM | Permalink

        Re: bender (#580),

        “2. Your idea of ruination is bizarre. It is not allowed to ask for proof and quotes and full-formed arguments when someone starts spouting questionable statements backed by nothing but linkies? That’s teamstyle.”

        Agreed! We’ve had extensive experience on this blog with misdirection and misinformation generated by just such conduct.

        • bender
          Posted Apr 26, 2009 at 7:56 AM | Permalink

          Re: Barclay E. MacDonald (#619),
          It will be noted that my strong insistence on full quotes and complete arguments – which was supposedly “ruining the thread” – eventually led the proponent to retract his statement and ulitimately to a happy consensus with Mark T. PDO phase shifts – if they even exist – are not predictable years in advance.

        • KevinUK
          Posted Apr 26, 2009 at 3:03 PM | Permalink

          Re: bender (#620),

          Just like climate climate change bender? Predicting the unpredictable is something which self-deluded climate modellers seem to think is perfectly possible. Indeed a certain Dr Hansen seems to think that he can do this with 99% certainty even though he has been repeatedly demonstrably wrong in all his predictions since 1988.

          KevinUK

        • bender
          Posted Apr 26, 2009 at 9:36 PM | Permalink

          Re: KevinUK (#621),
          My hunch is there is larger uncertainty on the GHG forcing component than the modelers are willing to admit. I think lucia has done a pretty good job holding the modelers feet to the fire on that one. What is the actual estimate? Don’t ask me. The science is truly too opaque for me to judge. AS soon as that engineering-quality report is ready … :)

  165. Jonathan Schafer
    Posted Apr 24, 2009 at 6:46 AM | Permalink

    I see the Democrats have invited Al Gore to testify before the House Energy subcommittee. Gore will be doing his usual song and dance as Congress prepares some sort of cap and trade bill of the worst sort. The Republicans in turn, invited Lord Monckton to testify opposite Gore, but when Democrats found out who the Republicans invited, the refused to allow him to testify. So once again, Gore’s testimony will go unrebutted.

    Too bad we can’t get Ross in there to talk about his T3 tax. Not that they would ever go for it, even though if you were going to do something, it’s the best plan I’ve seen put forward from an economic standpoint.

  166. bender
    Posted Apr 24, 2009 at 7:53 AM | Permalink

    It is worth noting that the GCMs do not generate any of the kind of low-frequency patterning observed in these EOF “circulatory modes”. No PDO. No ENSO. Yes ITCZ. (Global mean temperature is off, and sometimes badly.) If the these modal behaviors have a physical cause then it is due to something not currently in the GCMs. IOW their cause is a mystery. IOW their persistence should not be taken for granted.

    • Mark T
      Posted Apr 24, 2009 at 10:58 AM | Permalink

      Re: bender (#592),

      It is worth noting that the GCMs do not generate any of the kind of low-frequency patterning observed in these EOF “circulatory modes”. No PDO. No ENSO.

      Interesting. That is a problem for the modelers using these as predictive… oh wait, i’m sorry, what is the word (or phrase) they use? Possible outcomes?

      Mark

      • KevinUK
        Posted Apr 24, 2009 at 11:18 AM | Permalink

        Re: Mark T (#597),

        Projections?

        KevinUK

        • Mark T
          Posted Apr 24, 2009 at 11:31 AM | Permalink

          Re: KevinUK (#599), That was it… yes, projections.

          Mark

        • KevinUK
          Posted Apr 24, 2009 at 12:01 PM | Permalink

          Re: Mark T (#603),

          See the difference a change in a word makes Mark (they are projections not predictions so when they deviate from reality at some point in the future you can’t accuse them of being wrong because they were projections not predictions)? As a US tax payer you are paying for professors in decision science at Columbia University to assess the effectiveness of these subtle word changes. Sad, very sad!

          KevinUK

        • KevinUK
          Posted Apr 24, 2009 at 12:09 PM | Permalink

          Re: KevinUK (#604),

          Propaganda cloaked as decision science

          KevinUK

        • Geoff Sherrington
          Posted Apr 25, 2009 at 5:29 AM | Permalink

          Re: KevinUK (#606),

          Yes, propaganda cloaked as decision science it is. Note that, for example, the McKitrick tax option was not mentioned as an alternative.

          Unhappily, the material as presented in the article has been done over many years ago in the study of property rights, see Nobel Laureate Coase and successors at Chicago in particluar. The Chicago school tended to present results acedemically. The Columbia effort is non-scientific, with bare-faced admissions of subjectivity and working to an agenda.

        • Mark T
          Posted Apr 24, 2009 at 12:10 PM | Permalink

          Re: KevinUK (#604), Oh, I know, that’s why I brought the point up. Them’s called “weasel words” by folks like you and me. It’s an easy way to deny culpability when your words can be interpreted in multiple, if not conflicting, ways.

          Mark

    • DeWitt Payne
      Posted May 4, 2009 at 4:10 PM | Permalink

      Re: bender (#592),

      It is worth noting that the GCMs do not generate any of the kind of low-frequency patterning observed in these EOF “circulatory modes”. No PDO. No ENSO. Yes ITCZ. (Global mean temperature is off, and sometimes badly.) If the these modal behaviors have a physical cause then it is due to something not currently in the GCMs. IOW their cause is a mystery. IOW their persistence should not be taken for granted.

      According to McGuffie and Henderson-Sellers, A Climate Modeling Primer, 3rd ed., 2005, p205, AOGCM air and ocean modules are coupled asynchronously so it should not be at all surprising that low-frequency patterning is not observed in the model results. There are all sorts of other tricks used to improve computational efficiency that are non-physical like reducing the heat capacity of the deep ocean by up to an order of magnitude to reduce spin up time.

  167. bender
    Posted Apr 24, 2009 at 8:22 AM | Permalink

    HFL, from DrK#2:

    Demetris [Koutsoyiannis] argues that by admitting that uncertainty is an intrinsic property of nature; that causality implies dependence of natural processes in time, thus suggesting predictability; and that even the tiniest uncertainty (e.g., in initial conditions) may result in unpredictability after a certain time horizon, it is possible to shape a stochastic representation of natural processes. In such a representation, predictability (suggested by deterministic laws) and unpredictability (randomness) coexist and are not separable or additive components. Deciding which of the two dominates is simply a matter of specifying the time horizon of the prediction.

    One wonders if these circulatory modes (PDO, ENSO etc) – which vary on timescales in between that of weather vs. climate – are not a manifestation of Dr. K’s theory: so tantalizingly close to predictability, but never quite there. Rivers in oceans – a slow-flowing brew of near chaos.

  168. Wansbeck
    Posted Apr 24, 2009 at 10:45 AM | Permalink

    You Canadian guys have taken this teleconnection thing too far:

    I might have a stroll along there tomorrow.

  169. Gunnar
    Posted Apr 24, 2009 at 10:48 AM | Permalink

    >> Like I told you, the causal mechanism is not currently known

    I would think that ENSO events are caused by ocean currents. A conveyor belt that brings up water from deep in some areas, and sinks in other areas. It’s not predictable, because we don’t have enough deep ocean thermetrics. The underlying cause would be solar cycle pulsing, along with heating and cooling by the crust, along with chaotic like current dynamics.

  170. Wansbeck
    Posted Apr 24, 2009 at 11:13 AM | Permalink

    Teleconnection once more with link:
    Beach

  171. Wansbeck
    Posted Apr 24, 2009 at 11:25 AM | Permalink

    It’s hard to say because in true BBC fashion they have used a generic beach photo in their article rather than the real thing.

  172. Wansbeck
    Posted Apr 24, 2009 at 11:28 AM | Permalink

    Wow, I’ll have to withdraw my last comment; they have just changed the picture.

    A bit like climate data!

    • KevinUK
      Posted Apr 24, 2009 at 12:03 PM | Permalink

      Re: Wansbeck (#602),

      Bet they didn’t replace it with that oil sands image though did they?

      KevinUK

  173. Ivan
    Posted Apr 25, 2009 at 8:26 AM | Permalink

    It seems that Academy of Science of Poland agrees with Segalstad and Jaworowski that rising CO2 today is not of anthropogenic origin whatosever, and that ice core measurements show that in the last 400 000 years concentration of CO2 was often similar or higher than today’s. See here.

  174. Ivan
    Posted Apr 25, 2009 at 8:44 AM | Permalink

    Oops, I wrote they said rising Co2 has nothing to do with humans. I wasn’t correct,because they said, some portion of CO2 is due to the man. But on the other hand, they said concentrations of all greenhouse gasses are rising like in all previous episodes of warming (specifically they single out CO2 release from oceans). Similarly, they go directly against IPCC “solid rock” consensus science of “unprecedented concentrations” of CO2 today. Steve, it seems that Polish Academy of Science is joining extreme deniers among CA readers, in their wierd concpiracy theories that real climate scientists intentianally skrewed up ice core measurements in order to fit political narative of “unprecedented concentration caused by fossil fuels”. :)

    • Andrew
      Posted Apr 25, 2009 at 12:43 PM | Permalink

      Re: Ivan (#612), Um, I don’t see them say that at all.

      • Ivan
        Posted Apr 25, 2009 at 1:13 PM | Permalink

        Re: Andrew (#613),

        Please, read their article 6.

        6. Over the past 400 thousand years – even without human intervention – the level of CO2 in the air, based on the Antarctic ice cores, has already been similar 4 times, and even higher than the current value.

        It seems that Polish Academy of Sciences doesn’t beleive “IPCC Consensus Science” that highest CO2 concentration in pervious 400 000 or 600 000 years was 280 ppmv. “Current value” is about 380 ppmv, so they actually explicitly say in last 400 000 years CO2 concentration was at least four times “similar or higher” than 380 ppmv. That’s quite interesting – either Academy of Science of Poland is made of extreme right-wing climate deniers nuts, or IPCC Consensus at this particular point is wrong. Tertium non Datur.

  175. Ivan
    Posted Apr 25, 2009 at 1:15 PM | Permalink

    Or maybe Polish Academy is wrong and IPCC is right, but that could be established only if we examine actual evidence.

    • Andrew
      Posted Apr 25, 2009 at 2:37 PM | Permalink

      Re: Ivan (#615), There is another possibility. They may have simply made an innocent error in their interpretation of the ice core data. But attempting to assign motives-“extreme right-wing climate deniers nuts”-is just totally inappropriate reaction-something like shooting the messenger.

      • Ivan
        Posted Apr 25, 2009 at 3:06 PM | Permalink

        Re: Andrew (#616),

        Andrew I was ironical “concerning righ wing nuts”. People who usually express similar attitudes as Polish Academy in this statement, are routinely labeled as crackpots of right-wing deniers nuts by, well, real climate scientists. :)

        I actually believe they (Polish Academy) didn’t make an error in interpretation of ice core data, but expressed their “revisionist” interpretation of those data (most ice core data before 1985 show much higher pre-Industrial values than canonical IPCC 280 ppmv). I arrived to Canada from Serbia just couple of months ago. I can tell you that vast majority of climate scientists and meteorologists in Serbia are also worst climate deniers you can imagine. I know personally 8 or 10 most prominent professors of climatology and geophysics on University of Belgrade, and official meteorologists with government weather service and other institutions. ALL they are “Holocaust deniers” according to Mr Hansen’s criteria, most of them even strong believers in Sun-climate connection, and also many openly questioning even ice core interpretation of IPCC. Consensus science in my home country is that “IPCC sucks”. The same holds true, as you can see, for Poland, probably Czech Republic, Croatia, Slovakia, Russia, everywhere in (Eastern) Europe where big government money didn’t already create strong vested interests among climate scientists (the only promoters og AGW in Serbia are NGO crackpots connected with government money). A couple of years ago Russian academy of science declared that Kyoto protocol “has no scientific merit”. Can you imagine NAS declaring anything remotely similar? And newspapers from West (otherwise, highly critical of Mr.Putin’s crackdown on freedom of speach) called Russian president to “keep his scientists in line”.

  176. nevket240
    Posted Apr 26, 2009 at 1:18 AM | Permalink

    Have we reached Hansens ‘tipping point”??

    Very cool in Victoria,Oz, overnight with more to come. Just watching TV news before venturing off to work. Large snowfalls in the Oz snowfields, third year running. Third year in a row and NO mention by the tabloid press about “global cooling”.
    Snowfield operators are understandably happy.
    So. Have we reached the “tipping point”??
    regards.

  177. Knut Witberg, Norway,
    Posted Apr 27, 2009 at 1:00 AM | Permalink

    Arctic Sea Ice
    Just when the Arctic sea ice seemed to exceed the old average, the page

    stopped working :-(. What’s up?
    I also use

    but that page does not contain the average and seems to use slightly different data.
    Anyone who can enlighten me? Does it?

  178. Knut Witberg, Norway,
    Posted Apr 27, 2009 at 1:07 AM | Permalink

    Arctic Sea Ice (Hope it gets right this time)
    Just when the Arctic sea ice seemed to exceed the old average, the page

    http://nsidc.org/arcticseaicenews/index.html
    stopped working. What’s up?

    I also use

    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    But the data seem to be slightly different. Are they?

  179. Geckko
    Posted Apr 27, 2009 at 8:52 AM | Permalink

    An interesting article in the NYT.

    Industry Ignored Its Scientists on Climate

    An ironic quote from someone well known to Steve. Mc.:

    “I’m amazed and astonished,” Dr. Santer said, “that the Global Climate Coalition had in their possession scientific information that substantiated our cautious findings and then chose to suppress that information.”

    Heaven forbid someone should suppress scientific information in their possession!!!

    • Andrew
      Posted Apr 27, 2009 at 5:44 PM | Permalink

      Re: Geckko (#627),

      suppress scientific information in their possession

      Well Santer would know…

      http://www.worldclimatereport.com/archive/previous_issues/vol1/v1n21/feature1.htm

    • MJW
      Posted Apr 28, 2009 at 1:07 AM | Permalink

      Re: Geckko (#627),
      To repeat the substance of a comment I made on the AccuWeather blog, the NYT article, “Industry Ignored Its Scientists on Climate,” seems to me to be misleading.

      The supposedly ignored statement from the Dec. 1995 report is, “The scientific basis for the Greenhouse Effect and the potential impact of human emissions of greenhouse gases such as CO2 on climate is well established and cannot be denied….”

      As far as I can find, the Global Climate Coalition, didn’t claim otherwise, at least not after 1995. The article mentions, as if it were a contradiction, the GCC’s statement in an early 1990s “backgrounder” that “[t]he role of greenhouse gases in climate change is not well understood…” That’s clearly isn’t an actual contradiction. The 1995 report supports the claim, saying, “it is still not possible to accurately predict the magnitude (if any), timing or impact of climate change as a result of the increase of global greenhouse gas concentrations.”

      The online version of the NYT story links to the report, and to a copy of the backgrounder. The backgrounder says in the previously mentioned quotation, “But scientists differ on whether the concentrations of these gasses will cause an ‘enhanced greenhouse effect,’ or warming of the planet, because the role of greenhouse gases in climate change is not well understood.” That could be taken as contrary to the scientific report. The problem for the story’s premise is, the backgrounder is from the early 1990s, while the report was made in late 1995. The GCC can hardly be accused of ignoring a report that hadn’t yet been written.

      I found an updated version of the GCC backgrounder on the Greenpeace website. It was dated Dec. 1995 — the same month as the scientific report. The sentence had been revised to say, “But scientists differ on the rate and magnitude of the “enhanced greenhouse effect” (warming) that will result due to the increase in the concentrations of these gases or warming of the planet, because the role of greenhouse gases in climate change is not well understood.” So it appears the GCC modified the backgrounder to take into account the report’s conclusions.

      • kim
        Posted Apr 28, 2009 at 2:37 AM | Permalink

        Re: MJW (#636),

        MJW, I hope you point this out to Andy Revkin.
        ============================

        • MJW
          Posted Apr 29, 2009 at 1:39 PM | Permalink

          Re: kim (#637), I hope you point this out to Andy Revkin.

          After seeing your comment, I decided that would be a good idea, so I emailed a slightly modified version of my comment to Andy Revkin. He sent a polite reply, saying he wasn’t aware of the change of wording in the backgrounder when he wrote the story, and mentioning that he was doing some fresh reporting, and would soon write a piece that would clarify the situation. I look forward to reading it.

        • kim
          Posted May 3, 2009 at 12:01 AM | Permalink

          Re: MJW (#647),

          Woohoo, MJW has provoked a correction by Andy Revkin at the New York Times.
          ================================================

  180. kim
    Posted Apr 27, 2009 at 11:04 AM | Permalink

    I’m amazed and astonished that such a big deal is being made of Revkin’s find. Whether or not there was skepticism back then about the basic science, and Bob Carter says that of course there was, there certainly is now. This whole foofaraw misses the forest for the trees; skepticism about the CO2=AGW paradigm is a lot more advanced now than it was then.
    ===========================================

  181. Mark T
    Posted Apr 27, 2009 at 1:27 PM | Permalink

    Well, it snowed in CO Springs again today. I got nearly three inches of global warming in my yard. Good thing, too, because it’s still too cold at night to turn on my sprinklers and my grass looks like fire starter. I’m beginning to wonder about our planned camping trip over Memorial weekend. Could be chilly, particularly since the campground is at nearly 10000 feet.

    Mark

  182. See - owe to Rich
    Posted Apr 27, 2009 at 3:14 PM | Permalink

    Hello, please can you help?

    I was recently directed to a site grist.org which supposedly has arguments to use against global warming sceptics. However, when I try to follow the links, I just get an error. Does the same happen to other people, and is the information available elsewhere. I am especially interested in the “climate sensitivity is not very high” one, as that is my belief, and I’d like to know what the arguments against it are. Or do I just Google “pipeline” here on CA?

    Here are those topics:

    It’s cold today in Wagga Wagga
    Antarctic ice is growing
    The satellites show cooling
    What about mid-century cooling?
    Global warming stopped in 1998
    But the glaciers are not melting
    Antarctic sea ice is increasing
    Observations show climate sensitivity is not very high
    Sea level in the Arctic is falling
    Some sites show cooling

    TIA, Rich.

    • Craig Loehle
      Posted Apr 27, 2009 at 3:51 PM | Permalink

      Re: See – owe to Rich (#630), The argument for high sensitivity is basicly that the models assume constant relative humidity, and when you do the calculations, warming gets amplified. Note I said “assume” and in fact the humidity data (fraught with even more issues than tree rings) show conflicting results on relative humidity, with perhaps even declining RH with the past few decades. See Spencer’s site also

      http://www.drroyspencer.com/

    • DeWitt Payne
      Posted Apr 27, 2009 at 5:11 PM | Permalink

      Re: See – owe to Rich (#630),

      You need to turn on your sarcasm tag or at least put in a smiley face.

  183. Steve McIntyre
    Posted Apr 27, 2009 at 5:51 PM | Permalink

    Apologies for my radio silence. Taxes in Canada are due in April 30. I’ve been self-employed for years and always do my accounting for the past year in one binge. Every year I resolve to do my accounting in an orderly way the following year and every year I forget the resolution right away. However, I lost over 25 pounds last year and have kept it off through almost constant attention, so maybe this year I’ll keep my accounting resolution. Because of the market decline, I’ve sort of been in denial in terms of dealing with my own affairs and so I’ve done even less accounting than usual. It’s not that I’ve got that much income, but I’m trying to locate expense claims and things like that – and send them out and see what I can collect.

    I also spent some time over the past couple of weeks on a gold mining project. One of my friends wants me to go back to business at least on a part-time basis. Unlike most businesses, there are some gold projects that look pretty good. So I’m mulling that over.

    Be back on line in a few days.

  184. Posted Apr 27, 2009 at 7:22 PM | Permalink

    Mark T:
    April 27th, 2009 at 1:27 pm
    Well, it snowed in CO Springs again today. I got nearly three inches of global warming in my yard. Good thing, too, because it’s still too cold at night to turn on my sprinklers and my grass looks like fire starter. I’m beginning to wonder about our planned camping trip over Memorial weekend. Could be chilly, particularly since the campground is at nearly 10000 feet.

    In Pa/NJ it’s been at the opposite end of the spectrum, today was the third day over 90ºF, somewhat unusual for April.

  185. Larry T
    Posted Apr 28, 2009 at 3:47 AM | Permalink

    The 90ºF that I have here in PA is related to the same reason that you got 3 inches of snow in CO. There is a large dip in the jet stream that brings cold down in the dip and brings heat up in front of the dip.

    • Mark T
      Posted Apr 29, 2009 at 9:53 AM | Permalink

      Re: Larry T (#638), Yup. S’posed to be 75 here today, with more crud moving in this weekend. I’m getting tired of not being able to enjoy the weekends. Warm on a weekday doesn’t help much for outdoor activities (well, today is soccer, aka “a big beehive,” day for the squirt).

      Mark

  186. Andrew
    Posted Apr 28, 2009 at 5:20 PM | Permalink

    Scotland is going to try “centrally planned” adaptation:

    http://www.scotland.gov.uk/Publications/2009/04/23145206/0

    There is a word for this which comes to us from biology, but I can’t say it. Just know that it gets some biologists riled if you talk about it…

  187. See - owe to Rich
    Posted Apr 29, 2009 at 12:59 AM | Permalink

    Re Craig #631, I wasn’t asking about why the models have high sensitivity, but about why any climatologist can believe that the temperature record is consistent with the models. I believe the “answer” on the web page I can’t get to talks about oceans buffering heat. I know there are places on CA I could look for recent results about that.

    Still, I’d be interested to know whether anyone else can read all the stuff at http://www.grist.org/article/series/skeptics .

    Re STeve #634, regarding gold mines many of us think you have one right here. But I can understand you might prefer the real thing!

    Rich.

    • Andrew
      Posted Apr 29, 2009 at 12:02 PM | Permalink

      Re: See – owe to Rich (#640), Actually, it will probably be along the lines of “How do you explain Ice Ages then?” Bull crap. The non spatially/temporally homogenous change in insolation from glacial to interglacial cannot be compared to more uniform changes in radiative effects of well mixed gases.

    • Craig Loehle
      Posted May 1, 2009 at 7:15 AM | Permalink

      Re: See – owe to Rich (#640), regarding Grist’s rebuttals of skeptics: Take for example their rebuttal to “The satellites show cooling”: it refers to a long-ago correction to the satellite data, dismisses the satellite data as not accurate, and then says flat out that “it has been many years since the satellite data showed cooling” which is simply false. The satallite data show cooling of the past 8 to 12 years, depending on how you analyze it. They further conflate the upper troposphere data with the surface data. It is an example of hand waving, superior attitute, and pea-under-the-thimble con game.

  188. Vernon
    Posted Apr 29, 2009 at 12:04 PM | Permalink

    Hello, I have been looking for an answer to, how we know that CO2 is a strong climate driver with positive feed backs, when the historical record shows just the opposite? The ice core record shows, which is the only high resolution CO2/temperature record I am aware of, goes back for at least 480,000 years but it does not show that CO2 has ever caused warming. In fact, what it does show is that CO2 lags temperature rise by 800-2500 years and does the same when temperatures cool. Now I know that has been addressed lots of times but the answer I always hear is that once the CO2 starts rising, it raise the temperature. What I do not see when I look at the ice records, is any time that the peak of CO2 was before or at the same time as the temperature peak. The temperature peak always precedes the CO2 peak. This indicates to me that CO2 did not have a major impact on the warming, and that there was no latent temperature rise after the CO2 started rising. Where does the ice core record support the argument that CO2 is a powerful driver with strong feedbacks?

    Does anyone know why?

    • Andrew
      Posted Apr 29, 2009 at 12:20 PM | Permalink

      Re: Vernon (#643), Yes, there is a lag, but it doesn’t mean there isn’t an effect after (clearly it isn’t 100% of the change, of course) so the question then becomes, how big of an amplifier is it really? Probably not very, even under generous assumptions.

  189. Vernon
    Posted Apr 29, 2009 at 12:31 PM | Permalink

    That is my whole point. If CO2 was a strong driver the temperature peak should be when CO2 peaked. If there was lag, the CO2 peak should happen before the temperature peak. Neither of those happen. This indicates to me that CO2 is not much of a climate driver. It also indicates that there is either no lag between CO2 increase and increased temperature or CO2 is weak that temperature does not go up after CO2 does peak.

  190. bender
    Posted Apr 29, 2009 at 9:24 PM | Permalink

    Vernon, reread what Andrew wrote. He got what you meant the first time. But … Steve M is not interested in people discussing the so-called “lead-lag” problem. Andrew’s one-line reply is a propos.

    • Andrew
      Posted Apr 30, 2009 at 7:41 AM | Permalink

      Re: bender (#648), I think that’s the first complement you’ve given me. A sign that you think I’ve “matured”? :)

      • See - owe to Rich
        Posted Apr 30, 2009 at 2:32 PM | Permalink

        Re: Andrew (#650), I’m afraid I can’t compliment you on your spelling :-).

        Apropos of nothing, I was thinking of how much I like watching ice melt (or not), sunspots form (or not), Fourier series converge (OK, it’s my one weakness…). Anyway, perhaps not in such a daily fashion, I’d like to watch CO2 concentrations rise and sea levels rise (or not). I bemoan the fact that the sources of these I am aware of seem to have huge lag times (and probably not due to ice albedo effects :-)). Does anyone know of relatively up to date versions? Also, are there independent versions, to give us more confidence about the reported values?

        Happy May Day. Actually it’s still April where I sit, but I just posted on landscheidt.auditblogs.com, and it’s May there, so I suppose it must be in Australia or some such oriental place.

        Rich.

        • Andrew
          Posted Apr 30, 2009 at 2:47 PM | Permalink

          Re: See – owe to Rich (#652), :blush: I guess I need a reminder note taped to my laptop-“think before you type” or something like that! I to have been puzzled as to why the sources from which I usually get CO2 data seem to be more than a few years behind in getting the latest concentrations in. But I can’t seem to find sea level data at all (I haven’t looked to hard and it isn’t so important to me) just graphs. :shrug:

  191. Vernon
    Posted Apr 30, 2009 at 4:25 AM | Permalink

    Ok, sorry.

  192. Doug
    Posted May 1, 2009 at 8:10 AM | Permalink

    Re: Vernon (#645),

    Co2 is highly soluble in water, and unlike solids, the solubility is inversely porportional to temperature. When the climate warms, the oceans expel CO2, when it cools, they absorb it.. The 500-1000 year lag is a good number for how long it takes deep oceanic water to turn over and re-establish equilibrium with the atmosphereafter a climate shift.

  193. See - owe to Rich
    Posted May 1, 2009 at 10:29 AM | Permalink

    Re sea level, I apologize: the graph at http://www.aviso.oceanobs.com/fileadmin/images/news/indic/msl/MSL_Serie_J1_Global_NoIB_RWT_PGR_NoAdjust.png seems reasonably up to date. Is there any independent measurement from this, though?

    Re CO2, from http://www.esrl.noaa.gov/gmd/ccgg/carbontracker/co2timeseries.php#imagetable it looks like we have to wait until October or November to see 2008’s data graphed.

    Rich.

    • DeWitt Payne
      Posted May 1, 2009 at 1:38 PM | Permalink

      Re: See – owe to Rich (#656),

      That sea level graph is what makes me doubt the accuracy of the recent OHC measurements and the claim that OHC hasn’t increased since 2003. If that’s indeed the case then there has been a lot more melting of land based ice than has been thought. Assuming 100% of the increase in sea level is expansion then you need an annual increase in OHC of about 1E22 Joules which corresponds to a planetary radiative imbalance of about 0.7 W/m2. Even if you assume that half the increase in sea level/year comes from melting land based ice, which is high compared to the estimate of ~0.8 mm/year before 2003, the increase in OHC is still much higher than reported by the ARGO system. Of course it’s quite possible that I didn’t do the math correctly, but the simpler explanation is that all the bugs haven’t been worked out of the ARGO experiment yet.

      • BarryW
        Posted May 1, 2009 at 2:57 PM | Permalink

        Re: DeWitt Payne (#657),

        Global temps have been flat. I’m wondering if the supposed melt would account for some or all the lack of temperature increase?

      • Andrew
        Posted May 1, 2009 at 3:00 PM | Permalink

        Re: DeWitt Payne (#657), Well, its possible that the Ocean floor itself is rising, but that seems unlikely I think. It’s a puzzle. Either one or both datasets is probably at least a little wrong.

      • Andrew
        Posted May 1, 2009 at 3:04 PM | Permalink

        Re: DeWitt Payne (#657), Re: Andrew (#659), Actually, another possibility just occurred to me. Generally, people seem to forget that ARGO does not measure total OHC, but upper ocean heat content. The heat may be hidden deeper down (which isn’t predicted by models either).

        • DeWitt Payne
          Posted May 1, 2009 at 7:45 PM | Permalink

          Re: Andrew (#661),

          The heat may be hidden deeper down (which isn’t predicted by models either).

          Maybe, but that only makes the thermal expansion problem worse. The thermal coefficient of expansion of sea water decreases with temperature and temperature decreases with depth, so if the heat is going into the deeps, then it takes even more heat to cause the same increase in sea level. The structure breaking properties of the dissolved salt means that sea water density increases all the way to the freezing point, unlike pure water. The other problem is that it’s usually thought that heat transfer from the surface to the deep ocean is slow with a time constant on the order of 1,000 years.

          Re: BarryW (#658),

          I’m wondering if the supposed melt would account for some or all the lack of temperature increase?

          The amount of heat required to melt the amount of ice/year needed to increase sea level is orders of magnitude less than is required for a similar increase in sea level caused by thermal expansion, if I did my sums right. For about a 3 mm rise you need to melt ~1E12 kg or about 1,000 km3 of ice. That seems like a lot, but it’s a tiny fraction of the total amount of land based ice. The Greenland ice cap may be losing as much as 150 km3/year, but that’s still only about 15% of what would be needed. The contribution from the Antarctic ice cap is even less, about ~25 km3/ year or even less depending on whose data you believe.

        • Andrew
          Posted May 1, 2009 at 8:06 PM | Permalink

          Re: DeWitt Payne (#665), Well, I don’t know, maybe there is a delay in the time it takes for the Oceans to thermostatically respond to heat content changes? I don’t know but you seem to have given the issue some serious thought, so I’ll defer to you and call the situation a wash.

        • jae
          Posted May 1, 2009 at 8:40 PM | Permalink

          Re: DeWitt Payne (#665),

          How about some references to all this stuff, DeWitt? Trust, but verify, eh?

          The Greenland ice cap may be losing as much as 150 km3/year,

          I’m getting real sick of “may-bes.”

          Oh, and does anyone know what happened to McIntyre? No posts for a long time. Is he snowed in?

        • jae
          Posted May 1, 2009 at 9:04 PM | Permalink

          Re: DeWitt Payne (#665),

          The structure breaking properties of the dissolved salt means that sea water density increases all the way to the freezing point, unlike pure water.

          What are you saying here, DeWitt? Are you saying that salt water sinks when it freezes, unlike pure water? WTF? If that were to happen, we would have oceans of ice, not water.

          If water were like virtually all other substances, this world we know could not exist. Water is a “miracle” substance, arguably allowing the Earth to be a singular event in the Universe.

        • MJW
          Posted May 1, 2009 at 9:45 PM | Permalink

          Re: jae (#668), Sea water continues to contract as the temperature decreases, but sea ice floats because the salinity is lower. I tried to provide a link, but my comment was rejected as spam. Not doubt Spam Karma will brag about another spam eaten.

        • MJW
          Posted May 1, 2009 at 10:05 PM | Permalink

          Re: MJW (#669), I should have at least provided a reference and quotes. The book is Marine Geology by H. Kuenen, which can be read on Google Books.

          On contracting:

          But it is much less widely known that sea water of normal salinity does not show this remarkable property and continues to contract right down to the freezing point. Theoretically, there is a temperature of maximum density, but it lies far below the freezing point and therefore is of no consequence in nature.

          On lower salinity:

          The ice from sea water has lower salinity than the water itself because part of the dissolved salts, particularly the chlorides, escape as brine during the freezing process. This happens especially when the freezing is slow. In the course of time more brine leaks out of the ice, and finally a composition is reached rendering the sea ice potable on melting.

        • curious
          Posted May 2, 2009 at 4:30 AM | Permalink

          Re: DeWitt Payne (#665), Hi DeWitt – I make 1000km3 = 1000*(1000^3)m3 = 1e12m3 = 1e12 tonnes (at 1tonne/m3) = 1e15kg?

        • DeWitt Payne
          Posted May 2, 2009 at 8:40 AM | Permalink

          Re: curious (#674),

          That was a mistake. I meant 1E12 m3 not kg which is indeed 1E15 kg. Heat of fusion is 3.34E5 Joules/kg so the energy required is 3E20 Joules compared to 3E22 Joules to expand the top 700 m of sea water by 3 mm assuming an average temperature of 10 C.

          stupid spam filter. Let’s see if it takes it in pieces.

      • See - owe to Rich
        Posted May 4, 2009 at 1:49 AM | Permalink

        Re: DeWitt Payne (#656), you didn’t address my request for an independent measurement, though. IIRC, JASON was taken offline for several months in 2008, at a time when sea level seemed to be flattening out. I would hate to draw an erroneous conclusion as to why this was done, but I’d feel a lot happier if there was an independent group measuring sea level.

        After all, problems with UAH and MSU satellite measurement probably wouldn’t have been discovered if there had only been one of them.

        Rich.

        • DeWitt Payne
          Posted May 5, 2009 at 9:27 AM | Permalink

          Re: See – owe to Rich (#729),

          I don’t know how independent this is, but it does address my concerns about the sea level and OHC numbers not adding up. According to this paper by Cazenave, et.al., 80% of the ~2.5 mm/yr sea level increase from 2003 to 2008 was caused by an increase in ocean mass from melting ice, half from Greenland (135 GT/yr, 0.38mm/yr) and Antarctica (198 GT/yr, 0.56mm/yr) and the other half from other land based ice like mountain glaciers. Thermal expansion was small, ~0.3mm/yr, much more in line with the ARGO Ocean Heat Content measurements but still positive amounting to about 0.3E22 J/yr.

          There’s another interesting post about OHC at Pielke, Sr.’s site.

  194. Scott Brim
    Posted May 1, 2009 at 3:02 PM | Permalink

    .
    In his Climate Audit blog article “Tropical Troposphere – March 2009″ posted on April 14th, Steve McIntyre graphically plots data for the tropical troposphere anomaly from 1980 through March, 2009, as retrieved from UAH T2LT, RSS TLT, CRU, NOAA, GISS, HadAT 850 hPa radiosonde, and as recentered on 1979-1997 with Troposphere anomalies divided by 1.2 (per Christy). This is Steve’s graphic:
    .

    .
    As previously noted in the commentary section for that thread, NOAA’s National Climate Data Center has published a visual graphic on page 17 of the January 2009 review draft of their Unified Synthesis Product (USP) entitled Global Climate Change in the United States, as follows:
    .

    .
    NOAA’s graphic from page 17 of their USP is consciously designed to do three things: 1) to convey a powerful visual message that global warming is primarily the result of rising CO2 concentrations worldwide; 2) to act as prima facie evidence that CO2 is the culprit in the global warming upward trend; and 3) to give a strong sense of credibility to the overall CO2-is-guilty message by wrapping it in an aura of science-based provenance.
    .
    Just out of curiosity, and for purposes of encouraging further discussion in Steve McIntyre’s Climate Audit discussion thread, I rescaled his tropical troposphere temperature anomaly plot to degrees Fahrenheit, and then I visually overlaid it on top of the 1980-2007 segment of the NOAA global mean temperature anomaly plot taken from their USP.
    .
    This was done by graphically adjusting the two plot image files to the same horizontal and vertical scales for Year and for Temperature Anomaly (F), respectively, and then by merging the two image files together. Not having access to the raw data, and not currently having the software plotting skills needed to work with it, I did not recenter the two basis data sets to a common datum. Had I done so, the differences would likely be visually imperceptible. Here was the result, as posted in the earlier thread:
    .

    .
    One of the comments made by Climate Audit reader “curious” in that original discussion thread was that NOAA’s plot of CO2 concentration hadn’t been included in the merged graphic. The CO2 plot is now shown on this latest enhanced version, which also includes the entire original NOAA graphic decomposed into its various components.
    .

    .
    A higher resolution version of this latest graphic is available here:
    .
    High Resolution GIF Image, Enhanced Merged Graphic
    .
    As illustrated on the new graphic, I have added my own visual trace of NOAA’s global temperature anomaly plot. The visual trace is shown as a green line broken into four segments, with the begin/end points of each segment labeled 1 through 5 respectively.
    .
    These four segments represent my own visual interpretation of the temperature anomaly pattern contained in NOAA’s plot. Other CA readers might have different interpretations, particularly around the trace point which I label as “4” at approximately the year 1965. In placing only four segments, my justification is that I assume that over multi-decadal time frames, global mean temperature either generally falls or generally rises, it never stands still.
    .
    In addition, I have consolidated the multiple colors of the original McIntyre tropical troposphere anomaly plot into a single color. This effectively creates a visual composite of the temperature anomaly information Steve McIntyre had previously taken from UAH T2LT, RSS TLT, CRU, NOAA, GISS, and HadAT 850 hPa radiosonde.
    .
    I do not attempt to locate a visual trace line on Steve’s plot of tropical temperature anomaly. However, I do note that the overall trend for the period of 1980 through March 2009 is generally a rising trend, but at a slower rate of increase than NOAA’s plotted trend of global mean temperature anomaly.
    .
    Let us note at this juncture that NOAA, in their original graphic from page 17, has employed their own version of a visual trace line by using the CO2 concentration plot as the interpretive trend indicator for global mean temperature anomaly. They accomplish this by conveniently adjusting the respective vertical scales of their two data plots to give the appearance of a very close correlation among CO2 concentration and temperature rise.
    .
    The theory of anthropogenic global warming (AGW) depends heavily on the presumption that rising CO2 concentrations influence global mean temperature directly, significantly, and predictably — doing so well outside the bounds of natural variation, and in a more-or-less mechanical fashion through water vapor feedback mechanisms. In objective numerical terms, the temperature impact predicted by current AGW theory can be stated very succinctly as “a doubling of CO2 concentration yields 3 degrees Centigrade global warming.”
    .
    A central prediction of AGW theory is that a significant upward trend ought to be observed in the tropical troposphere temperature anomaly. When I look at the above merged data plot, and if I take what the plot seems to indicate strictly at face value – something which NOAA/NCDC also expects us to do with their own graphical plot from page 17 of their USP – certain features stand out, at least to me:
    .
    — While CO2 concentrations are rising at a more or less steady rate between 1880 and 1965, two significant episodes of general global cooling covering decadal timeframes occur within that time span.
    .
    — The rate of increase in global mean temperature anomaly between and 1908 and 1940 is roughly comparable to the rate of its increase between 1965 and 2005, while the rate of increase in CO2 concentration between and 1908 and 1940 is clearly less than the rate of its increase between 1965 and 2005.
    .
    — Wide variations, both up and down, occur in tropical temperature anomaly between 1980 and March 2009.
    .
    — Although the rate of increase in tropical temperature anomaly between 1980 and 2005 is generally up, its overall rate of increase is significantly less than the overall rate of increase in global mean temperature anomaly between 1980 and 2005.
    .
    — The plot of tropical temperature anomaly indicates that an apparent downward tendency is now developing for that anomaly, possibly from as early 2003.
    .
    .
    From my perspective as an old mechanical engineer, a close look at this latest graphic raises obvious questions concerning the true sensitivity of the earth’s global mean temperature to a steady rise in CO2 concentration, and it also raises obvious questions concerning the supposed lack of influence of natural variation.
    .
    In his various commentaries as posted on his Climate Audit blog, Steve McIntyre often emphasizes the current lack of an engineering-quality exposition which adequately and professionally explains the full theoretical basis for the generalized AGW prediction that a doubling of CO2 concentration yields 3 degrees Centigrade global warming. The need for such an exposition is surely self-evident, is it not?
    .

    • bender
      Posted May 1, 2009 at 3:37 PM | Permalink

      Re: Scott Brim (#660),
      You have just graphically proven to yourself what many here have been saying for awhile: the divergence between predicted and observed temperatures has been diverging for some time and is becoming harder and harder to ignore. Ny take is similar to that of lucia’s: model uncertainty has been WAY under-represented. Gavin?

    • curious
      Posted May 1, 2009 at 5:57 PM | Permalink

      Re: Scott Brim (#660), Hi Scott – nice graphic! The whole thing of “climate sensitivity” is what led me to CA a little while ago when trying to find a definitive explanation of the derivation of the CO2 forcing and thus the impact it has on “global temp.”. This was a question Steve and others had posed previously and it doesn’t seem to have been answered. To my engineer mind this seems unacceptable but I’m only following at a distance.

      The “CO2 climate sensitivity” seems to be little more than an empirical fit to recent temp. data – so what? There could be lots of short term parameters which could have sensitivities assigned to them (GM share price 1980-2000 with a suitable vertical scale?!) but it doesn’t make a causal relationship. Lots of other things are not clear – you touch on the tropical hotspot issue, I wonder about the polar amplification idea, influence of winds, proper integration of temp. vs time, instrumental limits etc etc. All stuff others have noted before and put work into. Hence my reference to “red herrings” on the other thread – from what I see the science is a long way from settled and following CA and other blogs shows a lot of smart and well informed people think so too. Did you read the presentation by Demetris Koutsoyiannis flagged up by HFL on the “Kousoyiannis et al 2008 #2″thread?

      One thing which would be good is if CA had an index system – so much has been covered with some great (often long) previous posts that I’ve stumbled across by chance. FWIW my view is that the science will come good on this because in the long run it can’t do otherwise – even if the situation becomes one of “we don’t know”. The posts that come in with new stuff put together in a way that makes sense are IMO all playing a part in advancing the arguments. Whether or not the best policy decisions get made in the meantime is another matter.

      I know the above repeats stuff already covered so apologies. And if anyone has a public access version of the BAS paper I’d like to read it:

      http://www.antarctica.ac.uk/press/press_releases/press_release.php?id=838

  195. MJW
    Posted May 1, 2009 at 6:25 PM | Permalink

    One slightly notable thing about the NOAA graph of CO2 concentration and global temperature is that they used the linear CO2 concentration instead of the logarithm. As we all know, the response is supposed to be logarithmic. The reason it’s only slightly notable is that if the log CO2 graph is scaled and offset to best match the linear CO2 graph, there’s very little difference.

  196. Hu McCulloch
    Posted May 1, 2009 at 9:53 PM | Permalink

    RE royfOMR at Comment #127 on the Irreproducible Results thread, I concede that Lord Monckton did an excellent job in his recent talk at Texas A&M. Evidently this is the talk he was not allowed to deliver to Congress.

    His relatively disappointing talk at the Heartland conference may have just been due to after-dinner-itis.

    • royfOMR
      Posted May 3, 2009 at 9:47 AM | Permalink

      Re: Hu McCulloch (#670),

      “His relatively disappointing talk at the Heartland conference may have just been due to after-dinner-itis”

      Dyspepsia – I like your diagnosis.

      AFAIK, there isn’t a video of a debate, involving the Viscount, that, recently, took place at ST Andrews University in Scotland (porridge and toast probably!). There is, however, a short report written by a participant that you may find of interest here:

      http://www.grumpyoldsod.com/global%20garbage%2016.asp

      I think that the refusal by the Democrats to allow Mr Gore to ‘indulge’ in debate with Lord Monckton was spot on.
      (1) On grounds of simple human decency and compassion; the phrase involving’Christians’ and ‘Lions’ springs to mind.
      (2) As is evidenced by the link above, Lord Monckton was able to reverse the opinion of a young audience, influenced and exposed to a life-long barrage of unquestioned, AGW propaganda. Putting him and his ‘pesky’ facts into quarantine, clearly defends the belief-systems of vulnerable politicians from a dangerous heresy!
      (3)Debate should never be undertaken when defeat is inevitable.

  197. thefordprefect
    Posted May 1, 2009 at 10:10 PM | Permalink

    An interesting plot???

    This seems to show a very reasonable increasing temperature with increasing CO2 PPM (0.01degC per co2 ppm)

    Data for CO2 before 1952 is linearly interpolated between yearly averages to get monthly figures.
    Data is from 1853 to present.

    CO2 and hadcrut3nh temps are then sorted on increasing CO2 and plotted – no other data processing

    • Raven
      Posted May 2, 2009 at 12:17 AM | Permalink

      Re: thefordprefect (#672)
      1) CO2 has a logarithmic relationship with temperature (i.e. a 1PPM increase in CO2 100 years ago would have 40% larger effect on temperatures than a 1PPM increase today).
      2) If one assumes the entire increase is due to CO2 then your data suggests a CO2 sensitivity of <2degC/doubling (i.e. no more than 1.2degC of additional warming likely which translates to ‘nothing to worry about’).

      • thefordprefect
        Posted May 2, 2009 at 7:48 AM | Permalink

        Re: Raven (#673), over the ppm interval displayed the straight line fit is very similar to the log.
        Current ppm = 388 double = 776
        Current temp deviation from the 1961 to 1990 average is 0.5C but @776ppm deviation is 2.5C
        additional warming over current temps is 2degC

        All this is assuming log CO2 effects.
        but below a certain level the CO2 absoption spectrum becomes less saturated and presumably this log proportionality will be less relevant.
        Also as more absoption spectru reaches saturation at higher ppm the log effect will be again changed.

        Log fit and extension of CO2 concentrations up to 800ppm shown here (note I freely admit that this s very inaccurate considering the very small ppm range figures available!):

        I find it interesting that there is such an obvious relationship between temperature and CO2.

        • Andrew
          Posted May 2, 2009 at 9:33 AM | Permalink

          Re: thefordprefect (#675), Obvious? Back to stats 101 with you…

        • Scott Brim
          Posted May 2, 2009 at 9:45 AM | Permalink

          Re: thefordprefect (#675)

          Fordie, just in the interests of further educating CA’s readers, could I ask you to give us some further background information on the CO2 log graphic you have posted in comment #675 — who generated it, what kinds of information it is intended to illustrate, what theories it describes or references, what sorts of analytical or graphical processes were used to produce it.

          Now, if it was you who generated it, I’m not expecting you to say who it is you actually are, just that it was you who generated it, if that is the case. But some background information as requested above would be useful for most everyone who reads this blog in terms of understanding the theory and process behind the graphic.

        • thefordprefect
          Posted May 2, 2009 at 10:27 AM | Permalink

          Re: Scott Brim (#679) RomanM (#677),

          Fist apologies: the first plot does have more points in the right hand pane. Unfortunately neither the blog nor image shack allow corrections Here is the corrected plot – both panes with the same data:

          The plots are mine
          CO2 1958-03 to 2009-03 (monthly series) is from

          http://co2now.org/index.php/Current-CO2/CO2-Now/current-data-atmospheric-co2.html

          Data prior to 1958-03 (yearly)is from
          http://data.giss.nasa.gov/modelforce/ghgases/Fig1A.ext.txt (a conglomeration of data from all sorts of sources!)
          Data is linearly interpolated between years to provide monthly data.
          Temperature is from this source

          http://hadobs.metoffice.com/hadcrut3/diagnostics/

          The statements made by many are that there is no relationship between CO2 and temperature on this sort of scale. These plots seem to visually show that there is. It may not be a simple linear/log/sine/etc but there is something.

          For comparison some have said the TSI influences temperature so again I plotted temperature vs SunSpot Number (approx equiv to TSI) with this result

          (please ignore the CO2 plots the one above is more recent) So I ask which has the effect on temperature?

          I readily admit plotting any rising variable with time and CO2 will produce this result. But isn’t this what it is all about – temperature is rising?
          Mike

        • curious
          Posted May 2, 2009 at 12:42 PM | Permalink

          Re: thefordprefect (#680),

          The statements made by many are that there is no relationship between CO2 and temperature on this sort of scale. These plots seem to visually show that there is. It may not be a simple linear/log/sine/etc but there is something.

          “Seem to” is the point. Lets say you are sitting on the dock of the bay one morning watching the scene as boats come and go etc etc. As it moves on towards midday you notice it is getting warmer. And, hey!, you also notice that the guys aren’t having to lift the boxes so high from the decks onto the quay. Whats going on? The sea level rise must be driving the temperature! A few hasty measures with a borrowed thermometer of unknown heritage and a tape measure down from the quay and bingo – the chart shows correlation! Local radio come over when you give them a breathless call and a crowd gathers as you explain the theory to the watching wonderers – there will be big implications of this for sure!…

          How long do you think the theory would stand up? Day after day? Week after week? etc etc? Just because two variables increase over some period of time does not mean they are a) related or that b) one causes the other.

        • Andrew
          Posted May 2, 2009 at 12:49 PM | Permalink

          Re: thefordprefect (#680),

          I readily admit plotting any rising variable with time and CO2 will produce this result.

          And yet he still does it and waves it triumphantly. Astounding.

          The statements made by many are that there is no relationship between CO2 and temperature on this sort of scale.

          Huh? Who says that? Me thinks I smell a strawman.

          For comparison some have said the TSI influences temperature so again I plotted temperature vs SunSpot Number (approx equiv to TSI) with this result

          So I ask which has the effect on temperature?

          Go back to stats 101, you are so dense it is unbelievable.

          http://en.wikipedia.org/wiki/Normally_distributed_and_uncorrelated_does_not_imply_independent

        • Andrew
          Posted May 2, 2009 at 1:27 PM | Permalink

          Re: Andrew (#686), Oh and a false dilemma to.

        • Scott Brim
          Posted May 2, 2009 at 11:30 AM | Permalink

          Re: thefordprefect (#675)

          Mike (a.k.a. thefordprefect) …. Just so the CA readership fully understands what has been done here, could you further explain the genesis and meaning of the blue dotted line shown in the graphic posted in comment #675. Here is your graphic:


          Mike’s Temperature versus CO2 Plot

    • RomanM
      Posted May 2, 2009 at 9:32 AM | Permalink

      Re: thefordprefect (#672),

      An interesting plot???

      Yes, but not for the reasons that you seem to think.

      The labels on the two plots purport that they present the same data, yet a simple sort is capable of discovering 100s of new points (with CO2 lessthan 310 ppm)! Amazing!

      Secondly, connecting the plotted points with a line in the second plot is both inappropriate and misleading. Connecting points makes sense if there is a natural progression from one point to another based on other variables (e.g. time, order of data collection, etc.). In your case it serves to smudge the data points hiding characteristics which may be otherwise informative. The number of point in your plots decreases as CO2 increases, but you can’t tell that from a bunch of jagged lines – it just looks like the strength of the relationship is more variable at the beginning because you can’t tell that the variability is likely due to the presence of a greater number of data points.

      Does it look like a roughly linear relationship? Not particularly. The lower CO2 level part has several stretches (with a lot of points) where the relationship seems to be reversed. You could get virtually the same results plotting CO2 versus any variable which increased (or decreased) over the same time period: positively correlated population statistics for the number of people or for their expected lifespans, or negatively with death rates from diseases such as cancer.

  198. DeWitt Payne
    Posted May 2, 2009 at 8:58 AM | Permalink

    For some reason it won’t let me respond to jae.

  199. thefordprefect
    Posted May 2, 2009 at 10:33 AM | Permalink

    Wrong again!! I think this is what I meant to plot:

    • RomanM
      Posted May 2, 2009 at 11:56 AM | Permalink

      Re: thefordprefect (#681),

      But isn’t this what it is all about – temperature is rising?

      HAS risen. I would be a lot more impressed if you had predicted the situation before it had actually occurred and then plotted the data. We already knew there has been some increase in the temperature over the past century. This plot does not in any way add any evidence that the increase is caused by CO2 and it suppresses a lot of the information of how the fluctuation relates to time. In particular, the left half of your graph indicates pretty much a lack of any relationship (decreasing initially?) until the CO2 level reaches about 330 or 340 ppm and from 370 up the temperature flattens and even comes down a bit. Your adding the superfluous connecting lines does try to fool the eye into adding a continuity to those increases.

      The linear fit is yet another misleading aspect because it implies that this increase supposedly took place at a constant rate. By the way, what’s the R-squared for the fit?

      • thefordprefect
        Posted May 2, 2009 at 2:28 PM | Permalink

        Re: RomanM (#683), for what it’s worth various curve fits:
        = 3.371E-05x^2 – 1.323E-02x + 6.844E-01
        R² = 4.611E-01

        y = 8.907E-03x – 2.920E+00
        R² = 4.564E-01

        y = 2.8934E+00ln(x) – 1.6749E+01
        R² = 4.5196E-01

        Re: Scott Brim (#682), the blue dotted line is the log curve fit equation as above

        Re: Andrew (#686), Why attack me, why not tell me where I am wrong. While you’re doing that draw and analyse an electronic schematic for electric power steering with feedback from torque sensor and output to 1kW 3 phase motor taking into account EMC and fail-safe safety requirements to operate fro -40 to 105C – I can, or are you too dense?

        Re: David Cauthen (#684), This is the real question here, everyone else is too keen to attack me rather than the data. Unfortunately I cannot see how this can be answered with 150 years of data

        Perhaps another factor that need including here is what has happened to the other GHGs over this 150 years? Anyone know of a source for the data?
        Mike

        • Andrew
          Posted May 2, 2009 at 2:58 PM | Permalink

          Re: thefordprefect (#688), I attack you because you are apparently totally ignorant of the relationship between correlation or lack thereof and causality or the presence of a relationship between variables. But you are just bragging going on about your electrical engineering skillz. The issue is, you seem to not be able to see why your analyses are meaningless. That’s pretty dense.

          Here’s a good one for you:

          or

          Do you get it yet? You haven’t proved a damn thing, so stop acting like an infantile triumphalist and live up to your alleged credentials. All I have seen from you so far is a claim of intelligence. Prove it by realiving why you are wrong.

        • Raven
          Posted May 2, 2009 at 3:56 PM | Permalink

          Re: Andrew (#689)
          Andrew you should plot temps vs. pirates until 2009 the correlation still holds!

        • curious
          Posted May 2, 2009 at 5:02 PM | Permalink

          Re: Andrew (#689), Any data available for the medieval period?

        • Ryan O
          Posted May 2, 2009 at 3:04 PM | Permalink

          Re: thefordprefect (#688), Dude, there’s nothing wrong with your data, which is why no one is attacking it. What is wrong is that you found a correlation between CO2 and temperature and then assume that there is a physical relationship. Not only that, but you go further and show that there is no pointwise correlation between sunspot number and temperature and then assume that this disproves a relationship between TSI and temperature.
          .
          You may be a whiz at fixing cars, but that skill has not translated into statistical ability. So no, you’re not stupid, but it does not seem that statistics is your game. You clearly do not understand what you are trying to show.

        • Jonathan Schafer
          Posted May 2, 2009 at 3:20 PM | Permalink

          Re: Ryan O (#691),

          Perhaps he has a future as a climate scientist then.

        • TAG
          Posted May 2, 2009 at 3:34 PM | Permalink

          Re: thefordprefect (#688),

          An extended verion of your graph was shown by Al Gore in “An Inconvenient truth”. It showed Mann’s hockey stick and compared it to carbon dioxide concentration. Gore remarked on the similarity and noted the same causation that you do.

          However Mann’s hockey stick does not show the Medieval Warm Period. If it did then the relationship between CO@ concentration and global temperature would seem to be questionable. This is why one aim of climate scientists is to “get rid of the medieval warm period.” It is also why RomanM is unimpressed with your chart.

        • curious
          Posted May 2, 2009 at 5:57 PM | Permalink

          Re: thefordprefect (#688), Try “trace gas climate change” in Google scholar for some pointers. Not looked far but I think aerosols are speculatively discussed with a backcast but less on CH4 etc.

          Also re: 696 Google on “pirates climate change” pulled this up: http://www.globalwarmingisreal.com/blog/2006/11/17/become-a-pirate-and-help-stop-global-warming/

          The same in Scholar pulls this one: http://geus.dk/publications/review-greenland-98/gsb183p61-67.pdf along with others :)

        • thefordprefect
          Posted May 3, 2009 at 4:39 AM | Permalink

          Re: curious (#698), Thanks, not found long enough records yet many stop in 2001 – why?!
          Re: Ryan O (#699),

          Now you’ve proven you don’t understand your own argument. Monckton’s graph (of which that quote is the caption) is shows both CO2 concentration and temperature as functions of time over the past 8 years. The two diverge.
          This speaks to Roman’s criticism of your plots. By presenting them as simple scatterplots, you have removed the time dependence. You cannot tell how well the two variables track with time. You cannot use your plots to “disprove” Monckton’s statement because your plots have nothing to do with Monckton’s.

          Please would you tell me where time comes into my arguement. in simplistic terms I am saying that if I surround a 100watt lamp (yes-even a low energy lamp) with thermal insulation today it will reach equilibrium at a certain air temperature depending on the power input (100watts – non varying) and the efficiency of the thermal insulation (unvarying). If I reduce the thermal insulation the temperature will fall – if i increase the power input the temperature will rise (etc. I’m sure you get my drift).

          If I did did this experiment now, or at any other time, the equilibrium temperature would be the same – or am I missing some fundamental physics here?

          There are many types of inputs to the global temperature equation Solar output (insignificant variation over the periods plotted. Other GHGs. Dust. Albedo. I’m sure you know them all.
          This will lead to variability of the temp vs co2 plot. I am in no way suggesting CO2 is the only input to thermal equilibrium.

          As to cherry picking plots of temperature and co2 plotted against time I am 100% certain I could pick a similar time sequence that would show an exact proportionality between temp and CO2.

          A question
          If you wanted to see how 2 variables compared would you plot these against a third non dependant variable (you want number of pirates (NOP) variability with sunspots so would you plot NOP vs CO2 and Sunspots vs CO2 and then visually/statistically compare the data? In my book plotting NOP vs Sunspots is the easy way.

          As I said above There are many inputs to the global temperature. How would you determine which is the major effect on temperature?

        • curious
          Posted May 3, 2009 at 5:55 AM | Permalink

          Re: thefordprefect (#704)

          As to cherry picking plots of temperature and co2 plotted against time I am 100% certain I could pick a similar time sequence that would show an exact proportionality between temp and CO2.

          Look at Scott’s fourth graphic in 660 – if you focused on 1880 to 1905 and/or 1940 to 1965 you could probably have a stab at claiming an inverse relationship: What would it prove?

          Re: your question – plotting two variables against each other would also identify “outliers” – e.g. the MWP for anthropogenic CO2 vs temp. It should highlight areas that don’t fit any postulated relationship and, for the theory to advance, you’d need to provide consistent arguments to explain these mismatches. At the moment IMO it seems the AGW theory lacks these supporting arguments.

        • Michael Smith
          Posted May 3, 2009 at 5:57 AM | Permalink

          Re: thefordprefect (#704),

          Please would you tell me where time comes into my arguement. in simplistic terms I am saying that if I surround a 100watt lamp (yes-even a low energy lamp) with thermal insulation today it will reach equilibrium at a certain air temperature depending on the power input (100watts – non varying) and the efficiency of the thermal insulation (unvarying). If I reduce the thermal insulation the temperature will fall – if i increase the power input the temperature will rise (etc. I’m sure you get my drift).

          I don’t think anyone here denies the basic physics of CO2-longwave radiation interaction. What is troubling — at least to me — is your effort to draw a conclusion based on only part of the facts.

          Yes, we’ve seen an increase in the surface temperature record over the last century, and yes, we’ve seen an increase in atmospheric CO2 levels. But we also have the following facts:

          1) The surface temperature record apparently suffers from UHI/citing problems, station drop-outs, etc. Supposedly, adjustments have been made for these factors — but as far as I know, no one can find out exactly what the adjustments have done. So we don’t know with much confidence the actual magnitude of the rise that has taken place at the surface.

          2) We have numerous global climate models that predict — given the observed increase in CO2 — an amplified temperature increase — amplified relative to the surface — in the tropical troposphere.

          3) However, we have other global climate models, admittedly a small minority, that predict the increase in CO2 concentration should cause a relatively small, virtually trivial heating of the tropical troposphere.

          4) We have a tropical troposphere temperature record since 1979 that shows essentially no heating.

          5) We have other historical records showing that global-scale warming of at least the magnitude seen over the last century has occurred in the absence of any significant human contribution to CO2.

          These certainly aren’t all the facts that should be considered, but they are sufficient — at least for me — to conclude that we cannot yet assign a cause to the (apparent) increase in surface temperatures seen over the last century. Do you agree?

        • Ryan O
          Posted May 3, 2009 at 7:49 AM | Permalink

          Re: thefordprefect (#704),

          As to cherry picking plots of temperature and co2 plotted against time I am 100% certain I could pick a similar time sequence that would show an exact proportionality between temp and CO2.

          .
          Yes. I am quite certain you could. So if you want to show Monckton’s error, that is what you must do. A scatterplot does not show that Monckton is wrong. Comparing both CO2 and temperature as a function of time would at least show that Monckton is cherry-picking dates to get the answer he wants. But what you’ve presented so far does not do that.
          .
          Re: Phil. (#702),

          Well that’s what happens when the noble lord ‘adjusts’ the data as far as I can tell from the ISAM data.
          Perhaps someone would like to ‘audit’ his letter?

          .
          That would be too easy. ;)
          .
          <–not a fan of Monckton, but then again, I didn’t bring him up.

        • curious
          Posted May 6, 2009 at 5:16 PM | Permalink

          Re: thefordprefect (#704), Hi Ford – you might have seen this by now but it could help with what you were looking for:

          http://www-ramanathan.ucsd.edu/publications/Ram%20JGR%2090%20D3%205547-5566%201985.pdf

          Ramanthan has other papers on this topic too.

        • thefordprefect
          Posted May 7, 2009 at 7:06 AM | Permalink

          Re: curious (#777), thanks for the reference will read in depth when not at work!

          http://www-ramanathan.ucsd.edu/publications/Ram%20JGR%2090%20D3%205547-5566%201985.pdf

          A quick scan seems to suggest that CFCs ozone,NxO CH4 etc will have more effect than CO2 if the predicted increases happened.

          Seems to sugest a increase in T of 1-2C

          Presumably there are more modern papers that disprove this one.

          How much effect has the ban on CFC etc had. I seem to recall that the levels had peaked?
          Mike

        • curious
          Posted May 7, 2009 at 7:34 AM | Permalink

          Re: thefordprefect (#778), I flagged it up as I remembered it from a while ago and I took a hardcopy (not to hand). Sorry it is a while since I looked at it so I can’t recall the implications but it seemed relevant as you wanted older trace gas figures and, despite its age, I doubt any more historic source data on this has been “found”(?). I think there is another one by Ramanathan et al which precceded this one which may give more detail on the sources of his data but I haven’t got a copy of that.

        • John M
          Posted May 9, 2009 at 8:25 AM | Permalink

          Re: curious (#780),

          (and previous)

          Have you seen these?

          http://www.climateaudit.org/?p=2638

          http://www.pnas.org/content/101/46/16109.full.pdf+html

          The first is a discussion about Hansen’s 1988 scenario’s with a link to data at Real Climate. I’m not absolutely sure, but I think the data at RC is a combination of “real” and projected GHG concentrations. There is also a data table where concentrations have been converted to forcings.

          The second paper indicates “we” might be able to buy time by concentrating on GHGs other than CO2 (co-authored by, of all people, you-know-who, in his pre-death-train days).

  200. David Cauthen
    Posted May 2, 2009 at 12:30 PM | Permalink

    So which is it? Is temp rising with CO2, or is CO2 rising with temp?

  201. Andrew
    Posted May 2, 2009 at 3:00 PM | Permalink

    Agh, here is your second ref:

    http://upload.wikimedia.org/wikipedia/commons/f/fd/Mierscheid_Law_Graph.svg

  202. thefordprefect
    Posted May 2, 2009 at 5:32 PM | Permalink

    Quoting from page 6 of Monckton’s document: http://scienceandpublicpolicy.org/images/stories/papers/monckton/temperature_co2_change_scientific_briefing.pdf

    No correlation, so no causation: Neither the global-temperature trend (red line) nor the global-CO2
    trend (cyan line) falls within the regions that encompass the IPCC’s projected intervals. Furthermore, there is a startling absence of correlation between the CO2-concentration trend and the temperature trend, necessarily implying that – at least in the short term – there is little or no causative link between the two.

    Do not my plots show this is not the case?

    http://en.wikipedia.org/wiki/Global_warming_controversy

    Climate sensitivity
    Equilibrium climate sensitivity refers to the equilibrium change in global mean surface temperature following a doubling of the atmospheric (equivalent) CO2 concentration. This value is estimated by the IPCC Fourth Assessment Report as “likely to be in the range 2 to 4.5 °C with a best estimate of about 3 °C.”
    Using a combination of surface temperature history and ocean heat content, Stephen E. Schwartz has proposed an estimate of climate sensitivity of 1.9 ± 1.0 K for doubled CO2.[80] , revised upwards from 1.1 ± 0.5 K.[81] Grant Foster, James Annan, Gavin Schmidt, and Michael E. Mann[82][83] argue that there are errors in both versions of Schwartz’s analysis. Astronomer Nir Shaviv also has proposed a low value for climate sensitivity.[84][85] Petr Chylek and co-authors have also proposed low climate sensitivity to doubled CO2, estimated to be 1.6 K ± 0.4 K.[86]

    There is science behind these predictions of temperature rise due to CO2. In my post (#675) the simple log curve fit gives a temperature rise 0f 2degC for a doubling of CO2 – not far off some estimates.

    Re: Andrew (#689), I assume that your graph of temperature vs pirates can be justified from scientific reserch? What are the latest predictions for a doubling of pirates?
    Mike

    • Dave Dardinger
      Posted May 2, 2009 at 5:51 PM | Permalink

      Re: thefordprefect (#696),

      Well as long we’re all being silly,

      I assume that your graph of temperature vs pirates can be justified from scientific reserch? What are the latest predictions for a doubling of pirates?

      As you may have noticed, there’s has been a recent upsurge in pirate numbers (see many headlines from the middle-east). And we’ve also seen a cessation in global warming (some say a downtrend). So what more do you want? Peer review is now passed, so let’s publish!

    • Ryan O
      Posted May 2, 2009 at 7:40 PM | Permalink

      Re: thefordprefect (#696),

      No correlation, so no causation: Neither the global-temperature trend (red line) nor the global-CO2
      trend (cyan line) falls within the regions that encompass the IPCC’s projected intervals. Furthermore, there is a startling absence of correlation between the CO2-concentration trend and the temperature trend, necessarily implying that – at least in the short term – there is little or no causative link between the two.

      .
      Now you’ve proven you don’t understand your own argument. Monckton’s graph (of which that quote is the caption) is shows both CO2 concentration and temperature as functions of time over the past 8 years. The two diverge.
      .
      This speaks to Roman’s criticism of your plots. By presenting them as simple scatterplots, you have removed the time dependence. You cannot tell how well the two variables track with time. You cannot use your plots to “disprove” Monckton’s statement because your plots have nothing to do with Monckton’s.

      • Posted May 2, 2009 at 11:41 PM | Permalink

        Re: Ryan O (#699),

        Now you’ve proven you don’t understand your own argument. Monckton’s graph (of which that quote is the caption) is shows both CO2 concentration and temperature as functions of time over the past 8 years. The two diverge.

        Well that’s what happens when the noble lord ‘adjusts’ the data as far as I can tell from the ISAM data.
        Perhaps someone would like to ‘audit’ his letter?

        • Andrew
          Posted May 3, 2009 at 8:38 AM | Permalink

          Re: Phil. (#702), Lucia already did that.
          Re: Mark T (#701), His words were “No correlation, so no causation”, no use of “implies” was made.
          Re: thefordprefect (#704),

          Solar output (insignificant variation over the periods plotted

          I wasn’t aware that Queen Victoria had such an advanced space program! Tell me, those satellites, how accurate were they? By the way, you also suggest that you can be sure we know all the inputs into the “climate equation”. That is downright false. You don’t know all the inputs, we don’t know all the inputs, no one does.

        • Mark T
          Posted May 3, 2009 at 11:34 AM | Permalink

          Re: Andrew (#711),

          His words were “No correlation, so no causation”, no use of “implies” was made.

          No, he did not. According to the quote above, he said:

          necessarily implying that – at least in the short term – there is little or no causative link between the two

          which cannot be taken as a fallacious argument by itself.
          Mark

  203. Andrew
    Posted May 2, 2009 at 11:13 PM | Permalink

    Guys, the origin of the pirate plot is the guy who came up with Pastafarianism.

    By the way, I am totally okay with Monckton being wrong (his suggestion that lack of correlation implies lack of causation is fallacious to) because I am not Monckton-and he is kinda out there sometimes.

    • Mark T
      Posted May 2, 2009 at 11:29 PM | Permalink

      Re: Andrew (#700),

      By the way, I am totally okay with Monckton being wrong (his suggestion that lack of correlation implies lack of causation is fallacious to) because I am not Monckton-and he is kinda out there sometimes.

      Only if you’re using the statistical/logical definition of “implies.” In statistics, “implies” is essentially the same thing as “causes,” or similarly, “A implies B” is the same as “if A, then B.” However, the standard use of “implies” (by normal humans) is much less rigorous and actually a bit more like “A suggests B.” Given what I know about Monckton, he’s probably (statistical word?) using the latter definition, though someone should point the distinction out to him.

      Statistically, even orthogonality (zero correlation) does not imply (logical) a lack of causality. Input a sinusoid into a 90 degree phase shifter, then calculate the correlation between the input and output: it is zero, but clearly they are causally linked. :)

      Mark

  204. jae
    Posted May 3, 2009 at 7:22 AM | Permalink

    Did McIntyre get swine flu or something? Long time no post!

    • David Cauthen
      Posted May 3, 2009 at 7:36 AM | Permalink

      Re: jae (#707),
      He commented in another post that he was busy doing his taxes, and that someone had offered him an opportunity in some kind of gold mining venture. Guessing his tax bill was kinda high. : )

  205. David Cauthen
    Posted May 3, 2009 at 8:06 AM | Permalink

    Basically, when you don’t remove time from climate, over the last 130 years (or 1000 years for that matter) there is a correlation between CO2 and temp… except when there’s not.

  206. Earle Williams
    Posted May 3, 2009 at 9:19 AM | Permalink

    Phil.,

    Excellent idea! I look forward to your report with great enthusiasm!

  207. thefordprefect
    Posted May 3, 2009 at 10:33 AM | Permalink

    So what we have is:
    global temperatures – no idea what is happening and never will

    Even if temperatures were known then 130 year record is still not long enough

    Even if temperatures were known then it is not allowable to have points off the curve especially if momentarily has the wrong slope.

    Seems that metrologists are onto a looser all the way as no data is valid. We will never be able to say the world is worming or cooling until it is too late. So lets continue as we always have!

    Re: Andrew (#711),

    I wasn’t aware that Queen Victoria had such an advanced space program! Tell me, those satellites, how accurate were they? By the way, you also suggest that you can be sure we know all the inputs into the “climate equation”. That is downright false. You don’t know all the inputs, we don’t know all the inputs, no one does.

    TSI has a good proxy in Sun Spot Number. Leif Svalgaard suggests SSN/TSI is valid back to the 1700s http://www.leif.org/research/

    http://www.leif.org/research/TSI%20(Reconstructions).xls

    I am totally unaware that I had stated that I knew all the inputs to the climate model. I suggested a few that is all.

    Re: curious (#705), Surely my scatter plot also shows outliers. It just does not allocate a date to them?

    I still do not see why date is important to the relationship between CO2 and temperature. However perhaps a 3d plot would be possible?

    • Mark T
      Posted May 3, 2009 at 11:29 AM | Permalink

      Re: thefordprefect (#714),

      I still do not see why date is important to the relationship between CO2 and temperature. However perhaps a 3d plot would be possible?

      Because you’re attempting to find a relationship between two time series? Really, that’s a trivial one.

      Mark

      • thefordprefect
        Posted May 3, 2009 at 1:57 PM | Permalink

        Re: Mark T (#716),

        Because you’re attempting to find a relationship between two time series? Really, that’s a trivial one.

        No, Time is not a major factor unless you are suggesting that the temperature increase is not reaching equilibrium before the cause is increased.
        If one assumes that Temperature is in equilibrium then time plays no part in global temperature. if an insulated box in space has exactly 20 watts dissipated inside then it will be at the same equilibrium temperature now as in 10^6 years time.

        Much more important are the effects not included. For example volcanic erruption throwing massive amounts of dust into the upper atmosphere producing cooling. Or different heat movement eg. collapse of the THC.
        Mike

        • Mark T
          Posted May 3, 2009 at 5:06 PM | Permalink

          Re: thefordprefect (#719),

          No, Time is not a major factor unless you are suggesting that the temperature increase is not reaching equilibrium before the cause is increased.

          I’m sorry, but you really don’t know what you’re talking about. They are two time series and thus, time matters. You cannot ignore this.

          Mark

        • thefordprefect
          Posted May 3, 2009 at 7:00 PM | Permalink

          Re: Mark T (#722), OK I give in, I am ignorant – please tell me why time matters, but keep it simple.

          My thoughts are – a simplistic case.
          I want to establish a relationship betweeen x and y.
          I measure y. At the same time I take an x reading.
          I change y and then measure x.
          I take thousands of readings and plot them x vs y
          Is this not the usual way of doing it.
          y does not even have to be done in increasing values.
          Each measurement is taken at a time and oviouslty a time series could be drawn but this is not what is being investigated. In an earth system CO2 cannot be changed at will, so time has to be allowed to pass for CO2 to vary.
          If other variables are recorded “simultaneously” then these can also be plotted x vs a/b/c… some of these could have the same effect on temperature. It is then a question of which is pushing and which is being pushed.

          The reason for my plot was simply that MANY have said that there is no relationship between co2 and temp, but the plots show a resonable relationship. I have never claimed it proves cause and effect.

        • ianl
          Posted May 3, 2009 at 8:07 PM | Permalink

          Re: thefordprefect (#723),
          So, your point is … ??

        • Mark T
          Posted May 3, 2009 at 10:35 PM | Permalink

          Re: thefordprefect (#722), Both of your “variables” are functions of time, period. RomanM already pointed out why this matters. If there is a relationship, it will hold over time, i.e., at time t = T1, what is the relationship between x and y, again at time t = T2?

          The reason for my plot was simply that MANY have said that there is no relationship between co2 and temp, but the plots show a resonable relationship.

          Nobody says that – a strawman at best. The plot does not show a relationship, contrary to your opinion, it shows similar behaviors – behaviors that are not really all that similar over even your plots (let alone time series comparisons).

          I have never claimed it proves cause and effect.

          If there is a relationship, there is cause and effect. So, which is your position?

          Mark

        • See - owe to Rich
          Posted May 4, 2009 at 1:37 AM | Permalink

          Re: Mark T (#725), I agree with both sides of the argument (but not equally). If temperature really did immediately reach equilibrium then its time evolution would not be important. However, in respect of any forcing, e.g. solar or CO2, or in respect of internal variability, e.g. ocean temperatures at different depths, it is certainly not in equilibrium. Therefore the time series element is vital to the understanding, and most especially why it warmed so much from 1910 to 1940. Solar effects were important there, and IMHO have had some influence since.

          Rich.

        • thefordprefect
          Posted May 4, 2009 at 8:40 AM | Permalink

          Re: See – owe to Rich (#728), Using the hadcrut3NH series the max rate of change over a long period seems to be 0.0229 deg C per year from 1977 to 2005 and 0.0191degC/year from 1918 to 1840:

          The temperature reaction time of the earth seems to be at least 0.0229 degC per year. If these in fact were limits then the apparent RAPID downturn from 2005 to present would be smoother – Without a few years of new figures this will not be known. I would therefore suggest that the measured temperature is not constrained by the time constant of the earth – for the rates of change seen so far. (if this were not the case then the temperature vs CO2 slope could be much steeper)
          Heres a few random CO2 and Temperature vs time plots.

        • curious
          Posted May 4, 2009 at 4:46 AM | Permalink

          Re: Mark T (#725),

          If there is a relationship, there is cause and effect. So, which is your position?

          If you have a neighbour who works for the same employer, at the same location as you, on the same shifts as you and you plot your departure time for work against his they are going to be pretty strongly related. Sickness and holidays will probably throw up some outliers which could be investigated and understood. If you now get promoted to head office which is 9-5 and in another town what will happen to the relationship between your morning leaving times? Where is/was the cause and effect?

        • Mark T
          Posted May 4, 2009 at 11:13 AM | Permalink

          Re: curious (#731),

          If you have a neighbour who works for the same employer,

          Cause and effect – the “relationship” is that they work for the same employer.

          ^Phil. – In addition to all your other expertise, you’re also an expert on British legal precedent now? Thought not. Until you can offer up some legal declaration that he is not allowed to make the claims he has made, you are committing libel. I’ve caught you making claims about things you did not understand before, so this is hardly a surprise.

          Mark

        • Posted May 4, 2009 at 11:48 AM | Permalink

          Re: Mark T (#735),

          ^Phil. – In addition to all your other expertise, you’re also an expert on British legal precedent now? Thought not.

          And you are such an expert? I at least have the clear statement of the Statute in my support that he is not a “Member of the House of Lords” and the explanatory document that he is not entitled to “the privileges of membership of the House of Lords”. Pretty clear

          Until you can offer up some legal declaration that he is not allowed to make the claims he has made, you are committing libel.

          I have done so, it’s the 1999 Act itself unless it has been repealed recently. I think it would be up to Lord Monckton to demonstrate that he has been given special privileges by the House of Lords, until then the presumption is that he is subject to the terms of the Act.

          I’ve caught you making claims about things you did not understand before, so this is hardly a surprise.

          I think not.

        • Mark T
          Posted May 4, 2009 at 12:34 PM | Permalink

          Re: Phil. (#736),

          I think not.

          Excuse me? Recall the magical mirror that created an increase in stored energy without an imbalance? Recall that you failed to understand that the step response resulted in an imbalance and resulted in the increase in energy I caught you red-handed making claims about something you clearly had zero knowledge of. Oh no, Phil., I demonstrated mathematically how wrong you were. Shall I direct readers to your nonsense?

          Your claim amounted to a violation of the law of conservation of energy, which, by your logic, should dismiss any and all of your future claims about anything scientific as nonsense.

          Re: curious (#737),

          Yes they do, but prior to identifying and investigating the outliers one could have speculated the relationship of time of departure was based on proximity of residence?

          You asked what the cause-effect was and I showed you what it was. What more is to explain? Speculations on what it could have been are immaterial. Question asked and answered.

          Mark

        • curious
          Posted May 4, 2009 at 1:05 PM | Permalink

          Re: Mark T (#738),

          You asked what the cause-effect was and I showed you what it was. What more is to explain? Speculations on what it could have been are immaterial. Question asked and answered.

          I gave you an example which developed to have exhibited no correlation to your postulated cause (same employer) back to correlation but with a different cause (new employer). I thought that was the context of the discussion with Ford starting from Scott’s post. Apologies if it was trivial or I misunderstood or didn’t clarify properly.

        • Mark T
          Posted May 4, 2009 at 1:34 PM | Permalink

          Re: curious (#740), The problem with thefordprefect’s “example” is two-fold. First, he was attempting to knock down a strawman. Nobody really claims there is no causal link between CO2 and temperature. The simple fact that it has mass and can store energy immediately refutes that idea, as does the simple fact that a warming ocean releases more CO2 (or absorbs less, same net change in flux). The second problem was that his example was intended to show this causal link, even though he hid it with wordsmithing, but it did nothing of the sort. He merely showed two variables with similar behavior. Plotting the time series shows that the implied “relationship” isn’t nearly as nice as he wanted it to be.

          I do understand your point a bit better, however, and I don’t think we’re really in disagreement with anything.

          Also, looking at your example, there’s still a causal link even after the neighbor moves to another job, though it is more subtle. ;)

          Mark

        • Posted May 4, 2009 at 8:26 PM | Permalink

          Re: Mark T (#738),

          Excuse me? Recall the magical mirror that created an increase in stored energy without an imbalance? Recall that you failed to understand that the step response resulted in an imbalance and resulted in the increase in energy I caught you red-handed making claims about something you clearly had zero knowledge of. Oh no, Phil., I demonstrated mathematically how wrong you were. Shall I direct readers to your nonsense? Your claim amounted to a violation of the law of conservation of energy, which, by your logic, should dismiss any and all of your future claims about anything scientific as nonsense.

          Not quite, it was a dichroic mirror, I perfectly understand how that system works, I’ve even built a few related devices.
          I may have not explained it well enough for you to understand it but that’s a different matter.
          There are even commercial light bulbs based on the same principle. There is no violation of the conservation of energy. As I recall the difference was that I was talking about the steady state operation and you were talking about the transient response.

        • Mark T
          Posted May 5, 2009 at 12:37 PM | Permalink

          Re: Phil. (#745),

          Not quite, it was a dichroic mirror, I perfectly understand how that system works, I’ve even built a few related devices.

          I know exactly what it was. What you did not understand is that your step response resulted in a temporary imbalance which is where the extra energy came from. You acknowledged that there was no violation of conservation of energy, but you were wholly incapable of explaining where the extra energy came from. I knew, explained it perfectly well, and you persisted in arguing the point.

          I may have not explained it well enough for you to understand it but that’s a different matter.

          I understood it perfectly well. That was never an issue except perhaps in your mind.

          There is no violation of the conservation of energy. As I recall the difference was that I was talking about the steady state operation and you were talking about the transient response.

          Um, no. You said there was NO imbalance, but the temperature rose any way. Your exact statement was:

          Here’s an example where there’s no imbalance between in and out but a significant increase in temperature.

          I don’t see any mention of equilibrium and this statement indicates a lack of understanding of where the increase originated (a first semester control theory student would have caught this). Should that imbalance go the other way (more out than in), the temperature would drop. Even after I pointed out the distinction, you failed to acknowledge it.

          So, either you were willfully spreading false information, OR, you were ignorant of how feedback really works in the system you claim to understand so well. Which was it, Phil.? What does it say about your credibility when you fail to demonstrate an understanding of such a simple point?

          Btw, I can demonstrate this graphically if you choose to remain ignorant and find out for yourself (very easy to do in Excel). For the record, the increased storage is the integral of the transient period delta between incoming and outgoing energy. It is exactly the infinite series that converges to 1/(1-a) where a is the fraction of reflected energy.

          This is completely different than any active amplifier in which the gain is actually gain, not just increased storage capacity. That is the problem with the silly microphone example that some people like to use.

          Mark

        • Mark T
          Posted May 5, 2009 at 12:41 PM | Permalink

          Re: Mark T (#748),

          It is exactly the infinite series that converges to 1/(1-a) where a is the fraction of reflected energy.

          Multiplied by your input energy, of course.

          Mark

        • curious
          Posted May 4, 2009 at 11:51 AM | Permalink

          Re: Mark T (#735),

          Cause and effect – the “relationship” is that they work for the same employer.

          Yes they do, but prior to identifying and investigating the outliers one could have speculated the relationship of time of departure was based on proximity of residence? Lets say following promotion management doesn’t work out so you decide to go back to shop floor work, but your old employer has filled your post. So you get a new job with a different employer running the same shift patterns as your neighbour and, hey ho, the proximity of residence once again appears to correlate to time of departure.

          Sorry if this comes across as labouring the point and in a non mathematical way (I came in on Scott’s 660 now 559) but its how I see the anthropogenic CO2 driving temp argument. Yes, maybe there is something there but on the story so far there could just as well not be?

        • Mark T
          Posted May 4, 2009 at 12:56 PM | Permalink

          Re: curious (#737),

          Yes, maybe there is something there but on the story so far there could just as well not be?

          Yes! That’s the whole point. The time series comparison simply provides a better story than the scatter plot.

          Your last paragraph in #720 was a good analogy to the problem, in my opinion.

          Mark

        • bender
          Posted May 4, 2009 at 6:31 PM | Permalink

          Re: thefordprefect (#722),

          I want to establish a relationship betweeen x and y.
          I measure y. At the same time I take an x reading.
          I change y and then measure x.
          I take thousands of readings and plot them x vs y
          Is this not the usual way of doing it.

          Your problem is u,v,w and z – which, like x, are not fixed, but are instead free to vary. They are known to be causal too, but they are excluded in your model, which violates the most important rule of regression analysis – that you have a correctly specified model. IOW your way of doing it is not at all “usual”.

        • Mark T
          Posted May 5, 2009 at 12:46 PM | Permalink

          Re: bender (#743), Plus, he has not considered a case in which the variables are related non-linearly, which may be effected by the others, u, v, w, and z. By plotting the time series, you gain the potential to at least see that there is a non-linearity (or, potentially, a lack of true correlation).

          Mark

    • Andrew
      Posted May 3, 2009 at 12:21 PM | Permalink

      Re: thefordprefect (#714), He suggests, he does not know. Because nobody knows, and they have little justification for confident pronouncements. No measurements means no certainty, not even close.

      Also:

      So what we have is:
      global temperatures – no idea what is happening and never will

      Even if temperatures were known then 130 year record is still not long enough

      Even if temperatures were known then it is not allowable to have points off the curve especially if momentarily has the wrong slope.

      That rather depends on your definition of “global temperatures”, but strictly speak it probably is impossible to “know” as such. This does not mean that our best efforts are invalid-no one has suggested that. Moreover the problem of “long enough-strictly speaking it is damn hard-near impossible-to show a causative relationship no matter how long the data series. This does not mean a relationship can’t be shown some other way, or that the record that exists can’t constrain possibilities. But now ask yourself, what do deviations from your curve actually mean? They mean, by your logic, that there are other forces at work because-per your SSN v Temp graph-you believe lack of correlation implies lack of causation. “Other forces” might be called by the other name of “confounding variables”-it is just as probable that these forces exaggerate the “true relationship” between CO2 and temperature as obscure it. It does not make sense to hand wave away instances of temp behaving not as expected while ignoring the implication that the relationship itself is a spurious result of those confounding variables.

      Re: Mark T (#717), That quote is indeed not so bad, but I was thinking of

      No correlation, so no causation: Neither the global-temperature trend (red line) nor the global-CO2
      trend (cyan line) falls within the regions that encompass the IPCC’s projected intervals.

      He later explains more accurately what he means, but his initial statement that is supposed to summarize it is not the same and is fallacious.

    • curious
      Posted May 3, 2009 at 3:01 PM | Permalink

      Re: thefordprefect (#714), and 719 above. Hi Ford – apologies, maybe my wording was unclear – your scatter plot is plotting one variable against another and it would identify outliers. So the next question is “what are those outliers?”. Going back to the data would then give you date references and hence things like the “MCA” would be identifiable and so explanations could be sought or wrought depending on your POV!

      Re: your 714 – no don’t continue as we are – IMO this would mean continuing to prop up a theory which has been over stretched like bubble gum, and, despite the holes appearing, some factions are still blowing. The result will probably be similar: sticky mess on face! Better IMO to acknowledge the need to understand as much as poss. re: longterm climatic behaviour and set about the science with that objective instead of trying to prove a pre established position.

      Lets say you design a power steering system and half way through its transition from LH to RH lock it does a few skips back to the left. How long would you keep perservering say “no, no, its all fine honestly”? Especially given there would be a major product liability issue? Similarly, would you say “Oh well, this new fangled thing ain’t workin’ so lets just stick with what we have”? Or would you say “We need to crack this – time for a rethink”?

  208. thefordprefect
    Posted May 3, 2009 at 10:35 AM | Permalink

    Using the term worming as in the Dune sense i.e. hot and arrid!!:o)

  209. TonyS
    Posted May 4, 2009 at 10:31 AM | Permalink

    Sorry if somebody has already posted this, but this study has hit my inbox with a “OMG THE WORLD IS GOING TO END!!!11!” comment attached to it. Maybe something for you to look into?

    http://www.nature.com/nature/journal/v458/n7242/full/nature08017.html

    Greenhouse-gas emission targets for limiting global warming to 2 °C

    Malte Meinshausen1, Nicolai Meinshausen2, William Hare1,3, Sarah C. B. Raper4, Katja Frieler1, Reto Knutti5, David J. Frame6,7 & Myles R. Allen7

    1. Potsdam Institute for Climate Impact Research, Telegraphenberg, 14412 Potsdam, Germany
    2. Department of Statistics, University of Oxford, South Parks Road, Oxford OX1 3TG, UK
    3. Climate Analytics, Telegraphenberg, 14412 Potsdam, Germany
    4. Centre for Air Transport and the Environment, Manchester Metropolitan University, Chester Street, Manchester M1 5GD, UK
    5. Institute for Atmospheric and Climate Science, ETH Zurich, 8092 Zurich, Switzerland
    6. Smith School of Enterprise and the Environment, University of Oxford, Oxford OX1 2BQ, UK
    7. Department of Physics, University of Oxford, Parks Road, Oxford OX1 3PU, UK

    Abstract

    More than 100 countries have adopted a global warming limit of 2 °C or below (relative to pre-industrial levels) as a guiding principle for mitigation efforts to reduce climate change risks, impacts and damages1, 2. However, the greenhouse gas (GHG) emissions corresponding to a specified maximum warming are poorly known owing to uncertainties in the carbon cycle and the climate response. Here we provide a comprehensive probabilistic analysis aimed at quantifying GHG emission budgets for the 2000–50 period that would limit warming throughout the twenty-first century to below 2 °C, based on a combination of published distributions of climate system properties and observational constraints. We show that, for the chosen class of emission scenarios, both cumulative emissions up to 2050 and emission levels in 2050 are robust indicators of the probability that twenty-first century warming will not exceed 2 °C relative to pre-industrial temperatures. Limiting cumulative CO2 emissions over 2000–50 to 1,000 Gt CO2 yields a 25% probability of warming exceeding 2 °C—and a limit of 1,440 Gt CO2 yields a 50% probability—given a representative estimate of the distribution of climate system properties. As known 2000–06 CO2 emissions3 were approx234 Gt CO2, less than half the proven economically recoverable oil, gas and coal reserves4, 5, 6 can still be emitted up to 2050 to achieve such a goal. Recent G8 Communiqués7 envisage halved global GHG emissions by 2050, for which we estimate a 12–45% probability of exceeding 2 °C—assuming 1990 as emission base year and a range of published climate sensitivity distributions. Emissions levels in 2020 are a less robust indicator, but for the scenarios considered, the probability of exceeding 2 °C rises to 53–87% if global GHG emissions are still more than 25% above 2000 levels in 2020.

  210. John Baltutis
    Posted May 4, 2009 at 8:11 PM | Permalink

    R for Mac OS X 2.9 is a Mac OS X implementation of the R language and environment for statistical computing and graphics. It’s free for Mac OS X 10.4.4 and up (Universal Binary).

  211. Andrew
    Posted May 5, 2009 at 9:16 AM | Permalink

    William DiPuccio on OHC and AGW:

    http://climatesci.org/2009/05/05/have-changes-in-ocean-heat-falsified-the-global-warming-hypothesis-a-guest-weblog-by-william-dipuccio/

    Plug away.

    • bender
      Posted May 5, 2009 at 12:56 PM | Permalink

      Re: Andrew (#746),
      Plugging …
      At first glance I don’t see any problems with this argument.
      [Steve M: can we please have a new "unthreaded"? This one is overloaded and performing poorly.]

      • Andrew
        Posted May 5, 2009 at 1:01 PM | Permalink

        Re: bender (#751), Someone is bound to have quibbles with it. And I second that call-this baby is getting unwieldy again…

      • Mark T
        Posted May 5, 2009 at 1:19 PM | Permalink

        Re: bender (#751), Agreed. I just read it myself. If true, it is damning to the alarmist position.

        What I find interesting is the end section where the explanations for the missing heat are discussed. Climate scientists are not looking the first place they should be looking, i.e., they are not looking at their original hypothesis. Instead, they are trying to “find” the missing heat because it is a foregone conclusion that their hypothesis is correct. That’s not science, that’s voodoo.

        Mark

        • bender
          Posted May 5, 2009 at 5:06 PM | Permalink

          Re: Mark T (#753),
          The irony, Mark T, is that a couple of years ago I was asking at RC if they could describe how heat accumulates in oceans and if it is possible that heat could be transferred deep. I was mocked as a denier by know-it-alls Ladbury and Roberts.

        • Andrew
          Posted May 5, 2009 at 5:31 PM | Permalink

          Re: bender (#756), How rude! Jeez, the way RC treats people just looking for information never ceases to amaze-no wonder they are frustrated at their difficulty convincing people-the act like total a$$h@!#$ about it…Not that I’m just realizing this now.

        • Mark T
          Posted May 6, 2009 at 9:07 AM | Permalink

          Re: bender (#756), Yes, they are 100% correct about everything because they are “believers in the consensus.” Hence, anyone that asks a question that may present a challenge to the consensus must be labeled and marginalized as a denier. They cannot admit anything that may expose a flaw or crack in the consensus position for fear of upsetting the consensus position. Even admitting points that the deniers and “believers in the consensus” agree upon is not allowed. There are many parallels going on around here, including the Chu situation (not so much the marginalization, but the refusal to admit the obvious for fear of exposing a crack in the consensus position).

          It is nothing more than pathetic, actually.

          Mark

        • DaveR
          Posted May 6, 2009 at 10:05 AM | Permalink

          Re: Mark T (#762),

          They cannot admit anything that may expose a flaw or crack in the consensus position for fear of upsetting the consensus position.

          Of course they can’t because “the consensus” is all that stands in the way of global apocalypse. People must be convinced that GW is real and that won’t happen if those with “authority” express doubts and those who question or disagree aren’t squashed. They must present a united public face of absolute certainty in their convictions at all times.

  212. See - owe to Rich
    Posted May 5, 2009 at 2:41 PM | Permalink

    I find it interewsting that DiPuccio claims that sea level has been dropping slightly since 2005. This contradicts the JASON graphs. Will the real sea level data please stand up?

    Rich.

    P.S. New thread please, I can’t type correctly if I can’t see what I’m typing.

    • Andrew
      Posted May 5, 2009 at 2:49 PM | Permalink

      Re: See – owe to Rich (#754), If you cherry pick from the very end of 05, its basically flat:

      Can’t find a more recent graph, this one is a couple months old.

      • BarryW
        Posted May 5, 2009 at 7:15 PM | Permalink

        Re: Andrew (#755),

        How often does U of Col update the sealevel data on their website? Doesn’t seem to have been updated since Jan.

        • Andrew
          Posted May 5, 2009 at 7:39 PM | Permalink

          Re: BarryW (#758), I’ve lost the URL, so I have no clue, unfortunately.

  213. Andrew
    Posted May 5, 2009 at 7:57 PM | Permalink

    D’oh! Just found it:

    http://sealevel.colorado.edu/

    Now for the fun part. According to this page:

    http://sealevel.colorado.edu/steric.php

    The steric contribution to sea level rise over this period should be about 1.54+-.23 mm per year. 3.2+-.4 mm per year was observed by the altimeters, so the non steric contribution must be somewhere around 1.66 mm/year. Now my question is, what does steric mean? And where does the other more than half of the rise, er, arise from?

    • DeWitt Payne
      Posted May 6, 2009 at 2:44 PM | Permalink

      Re: Andrew (#760),

      what does steric mean? And where does the other more than half of the rise, er, arise from?

      Steric is short for thermosteric or thermal expansion. The rest of the increase comes from melting land based ice. See for example here. Thermal expansion is an indirect measure of heat content. Changes in salinity have an effect as well.

      • Andrew
        Posted May 6, 2009 at 2:58 PM | Permalink

        Re: DeWitt Payne (#769), What kind of effect do changes in salinity have? Anything significant enough to worry about it confounding our use of sea level to indirectly look at OHC? You’ve been pretty helpful so far, so I hope you can answer these questions to! Thanks.

        • Mark T
          Posted May 6, 2009 at 3:06 PM | Permalink

          Re: Andrew (#770), It changes the temperature/density curve. Pure water actually peaks around 4 C as I recall, so pure water near freezing would actually compress as it warmed. Sea water doesn’t have a peak until right before it freezes (someone posted the curve, so I’m guessing a bit on this).

          Mark

        • Andrew
          Posted May 6, 2009 at 3:17 PM | Permalink

          Re: Mark T (#771), Why question was why, but more “how much”.

        • Andrew
          Posted May 6, 2009 at 3:24 PM | Permalink

          Re: Andrew (#772), Er, should be not why.

        • DeWitt Payne
          Posted May 6, 2009 at 3:48 PM | Permalink

          Re: Andrew (#770),

          This page has a density vs temp and salinity calculator down at the bottom. The Argo floats measure salinity, temperature and depth. I don’t know how much the salinity varies, though.

        • DeWitt Payne
          Posted May 6, 2009 at 3:52 PM | Permalink

          Re: Andrew (#770),

          There are density vs temp and salinity calculators on the web. I tried to link to one, but it seems to trip the spam filter for some reason. The Argo floats measure salinity, temperature and depth. I don’t know how much the salinity varies, though.

        • Andrew
          Posted May 6, 2009 at 4:28 PM | Permalink

          Re: DeWitt Payne (#775), Based on this chart of fresh water equivalent volume anomalies in different parts of the Atlantic, I would say “Some”

  214. mondo
    Posted May 6, 2009 at 5:39 AM | Permalink

    Terrific discussion (debate even) between Boris and some obviously knowledgeable people at Jo Nova’s site on CO2 sensitivity – http://joannenova.com.au/2009/04/12/climate-bull-or-bear-trend/ Scroll down a bit to see the fun.

  215. Patrick M.
    Posted May 6, 2009 at 11:07 AM | Permalink

    There is an interesting paragraph in the latest RC post by rasmus here:

    I found insufficient detailed description in SW09 of the methodology used in their analysis to be able to judge the real merit of their work. The paper provides a link to auxiliary material that does not work. However, the figures in the paper don’t really convince when I don’t know how they were made.

    I’m sure regular CA readers will appreciate the irony.

  216. Patrick M.
    Posted May 6, 2009 at 11:10 AM | Permalink

    Wow, post 764. I suppose I should put on an oxygen mask at this altitude.

    • Stan Palmer
      Posted May 6, 2009 at 12:11 PM | Permalink

      Re: Patrick M. (#765),

      Wow, post 764. I suppose I should put on an oxygen mask at this altitude.

      However that degree of armwaving should develop sufficient lift to support a 747 at any altitude.

  217. Stan Palmer
    Posted May 6, 2009 at 12:08 PM | Permalink

    Accepted Network Internet Models are Wrong

    http://www.ams.org/notices/200905/rtx090500586p.pdf

    This is a paper from AMS. The previously accepted network models of tbe Internet are wrong. They were based on poor data and giove a false picture of the true properties of the Internet.

    Who would have thought that models derived from poor data derived from poorly constructed experiments could have been accepted in the peer-reviewed literature.

    See the reference to Yule whou should be familar to all readers of this blog.

  218. Mark T
    Posted May 6, 2009 at 3:34 PM | Permalink

    I don’t know the actual answer though I would guess you should probably account for it even if it might be small.

    Mark

  219. David L. Hagen
    Posted May 8, 2009 at 1:15 PM | Permalink

    Steve
    Per your “maximum likelihood method”, may I refer you to Nicola Scafetta‘s work. He demonstrates that “Diffusion Entropic Analysis” can identify physical phenomena underlying complex time series, including non-Gaussian Levy and other series.

    See: Nicola Scafetta, Dissertation: An entropic approach to the analysis of time series.

    Abstract

    Statistical analysis of time series. With compelling arguments we show that the Diffusion Entropy Analysis (DEA) is the only method of the literature of the Science of Complexity that correctly determines the scaling hidden within a time series reflecting a Complex Process.

    The time series is thought of as a source of fluctuations, and the DEA is based on the Shannon entropy of the diffusion process generated by these fluctuations. All traditional methods of scaling analysis, instead, are based on the variance of this diffusion process. The variance methods detect the real scaling only if the Gaussian assumption holds true. We call H the scaling exponent detected by the variance methods and d the real scaling exponent. If the time series is characterized by Fractional Brownian Motion, we have H¹d and the scaling can be safely determined, in this case, by using the variance methods. If, on the contrary, the time series is characterized, for example, by Lévy statistics, H ¹ d and the variance methods cannot be used to detect the true scaling. Lévy walk yields the relation d=1/(3-2H). In the case of Lévy flights, the variance diverges and the exponent H cannot be determined, whereas the scaling d exists and can be established by using the DEA. Therefore, only the joint use of two different scaling analysis methods, the variance scaling analysis and the DEA, can assess the real nature, Gauss or Lévy or something else, of a time series. Moreover, the DEA determines the information content, under the form of Shannon entropy, or of any other convenient entopic indicator, at each time step of the process that, given a sufficiently large number of data, is expected to become diffusion with scaling. This makes it possible to study the regime of transition from dynamics to thermodynamics, non-stationary regimes, and the saturation regime as well.

    First of all, the efficiency of the DEA is proved with theoretical arguments and with numerical work on artificial sequences. Then we apply the DEA to three different sets of real data, Genome sequences, hard x-ray solar flare waiting times and sequences of sociological interest. In all these cases the DEA makes new properties, overlooked by the standard method of analysis, emerge.

    Thesis PDF University of North Texas, 2001. (Emphasis added.)

    The power of this method is shown in Scafetta’s application of this DEA method to detecting complex correlations between solar activity and earth’s climate. See: Nicola Scafetta and Bruce J. West, “Is climate sensitive to solar variability?” Physics Today, 3 50-51 (2008).

    See also his 2009 EPA presentation: “Climate Change and Its causes: A Discussion about Some Key Issues” N. Scafetta. Invited author at the U. S. Environmental Protection Agency, DC USA, February 26, 2009.
    Scafetta’s Presentation Slides

    From the sun’s rotation around the solar system’s center of mass, Scafetta has developed a model that appears to identify the solar signature in climate. See Slide 66 of his Presentation slides.

    This explains the 1910-1945 warming and implies that about 70% of the observed warming from 1975 to 2002 was part of this natural climate cycle during its warm phase.

  220. Craig Loehle
    Posted May 9, 2009 at 6:59 AM | Permalink

    A journal has asked me for some names for reviewers for my paper. I need some people who are statistically literate, such as RomanM, UC, Jean S, none of whom I have contact info for. If any of these or others able to act as reviewers could contact me craigloehl at aol.com I would appreciate it. The paper has been stuck for 10 months in review. Thanks.

  221. Ron Cram
    Posted May 9, 2009 at 7:37 PM | Permalink

    This morning I stumbled across a blog I had not visited before called Climate Progress and, as fate would have it, the proprietor had just posted on a new effort to indoctrinate children in climate alarmism. I tried to interject a little realism into the discourse and was soon informed that I did not know what I was talking about. A lively little debate was just beginning when I was called away for life. I returned to the blog this evening and attempted to provide readers of the blog with some additional information. However, comments to the blog are now moderated and my comment did not make it through. It certainly was not because I had written anything disrespectful, but the readers of the blog had expressed their opinion I should not be allowed to post there. Read it for yourself. Of course, your comments might not be welcome there either, but I think the episode is instructive. I think it is sad when people stick there head in the sand.

    • John Baltutis
      Posted May 10, 2009 at 12:02 AM | Permalink

      Re: Ron Cram (#784),

      Hmmm!!! I see at least three of your posts there.

      • Ron Cram
        Posted May 10, 2009 at 7:37 AM | Permalink

        Re: John Baltutis (#785),

        Yes, there was no moderation at all initially and I posted five times. All five posts still show. As I mentioned above, it was developing into a lively debate. After I left and came back, they initiated “moderation” which turned out to mean anybody who agreed with the proprietor could comment and I could not. A number of misstatements were made which I would like to correct, including the moderator making some blanket statements disagreeing with me but not offering any evidence to back up his claims. It never ceases to amaze me how people who supposedly understand the science are afraid of any debate. That alone should convince people to be wary and to understand the alarmist’s arguments will not hold up to scrutiny.

      • Ron Cram
        Posted May 10, 2009 at 9:03 AM | Permalink

        Re: John Baltutis (#785),

        At one point, Joe Romm at Climate Progress mentioned two blogs he likes – RealClimate and SkepticalScience. Of course, I know all about RealClimate and the way they censor comments they do not have an answer for but I had not heard of SkepticalScience. I visited and decided to comment. It seems they have a policy against “trolling” which leaves plenty of room for interpretation. But they may actually interpret it in a way that would allow discussion and debate as long as it is civil and informed. We shall see. It is really disappointing when people are not willing to have their preconceived notions challenged.

    • Andrew
      Posted May 10, 2009 at 11:20 AM | Permalink

      Re: Ron Cram (#784), Joe Romm(el) is dishonest scum. I wouldn’t waste your time with him. And no, there really is no nicer way I can describe trash like him.

  222. Ron Cram
    Posted May 10, 2009 at 10:54 PM | Permalink

    Andrew, thank you for this link. I had never heard of Joe Romm before today. The link was very interesting background on him.

  223. oakwood
    Posted May 11, 2009 at 12:37 AM | Permalink

    What’s happened at the BBC?
    Last night, I saw an excellent programme on the wonders of South Pacific Islands, and the precarious lives of some of the human inhabitants (BBC2, 9th May 20:30 hrs ‘ South Pacific’). But, I didn’t hear a single reference to global warming or the threats of sea level rise! How could this be?

    • Andrew
      Posted May 11, 2009 at 1:28 PM | Permalink

      Re: oakwood (#790), They forgot.

      (Okay Steve, I can see where my previous comment might have warranted a snip. I think I will take some lessons on toning down the snark)

  224. jae
    Posted May 11, 2009 at 4:49 PM | Permalink

    Wow. The bristlecones are no longer the oldest living things. Swedish spruces about TWICE as old! Reverse hockey stick, too :)

  225. Greg F
    Posted Mar 19, 2009 at 8:09 PM | Permalink

    Re: RAcookPE1978 (#15),
    This discussion is fastenating.

  226. Posted Mar 30, 2009 at 6:46 AM | Permalink

    Re: Knut Witberg, Norway, (#201),

    Try removing the word ‘specialist’ from your post, that triggers the drug name detector.

  227. Pat Frank
    Posted Apr 7, 2009 at 10:44 AM | Permalink

    Re: bender (#8), “The only species-site combinations eligible for temperature reconstruction are the ones where temperature is far more limiting to growth than moisture.

    bender, do you have any idea how one can turn a qualitative judgment about temperature limited growth into a quantitative and physically valid measure of temperature in Kelvins?

  228. bender
    Posted Apr 7, 2009 at 11:43 AM | Permalink

    Re: Pat Frank (#16),
    Quantitative, yes. Qualitative, no. Please don’t waste time trying to bait me into silly rhetorical arguments. Can’t we stay on topic?

  229. Raven
    Posted Apr 7, 2009 at 2:39 PM | Permalink

    Re: bender (#347)
    Spencer’s feedback analyses have apparently turned out to be a form of phase space anaylsis: http://en.wikipedia.org/wiki/Phase_space
    I think he is on the right track when it comes to understanding and quantifying the mechanisms that will lead to chaotic variations in the climate.

    He obviously does not have all of the answers but it definitely worth looking at if you have an interest in climate and chaos.

  230. KevinUK
    Posted Apr 8, 2009 at 4:52 AM | Permalink

    Re: Steve McIntyre (#7),

    And of course Steve there are also several dendros who believe it is OK to add certain inappropriate (by NAS) proxies inorder to enhance the signal they are looking for.

    Listened to your presentation on the Heartland 2009 Conference over the weekend. It was great to hear you recap and summarise a lot the findings you’ve presented on many of the threads I’ve read here on CA over the last few years. In particular it was great to hear Christopher Booker and Lord Monckton (who both gave hilarious presentations) give yourself and Anthony Watts such well deserved praise.

    Heartland 2009 conference proceedings

    KevinUK

  231. Willem Kernkamp
    Posted Apr 17, 2009 at 10:03 PM | Permalink

    Re: Ron Cram (#20), and Re: Hank (#21),
    Each year considerable effort needs to be expended to determine whether global temperature has gone up or down by a fraction of a degree. This, thankfully, is the stability of the climate. Ten days from now, the weather can be ten degrees warmer or ten degrees colder than the average for the month. The outcome depends on small disturbances, present today, that grow exponentially and that are very hard to track. This is the instability of the weather.

    Arithmetic is not good at tracking things that grow exponentially, because round-off error also grows exponentially. That is the signature of chaotic systems. The stock market is an example of such a system and attempts to capture it with arithmetic have lead to disaster.

    There are features of nature that lend themselves well to arithmetic. For example, the tide charts in a nautical almanac predict daily sea level variations far into the future and with accuracy.

    I am well aware that certain properties must be tuned. In fact, I proposed to “tune” the models to Professor Gray’s observed atmospheric behavior when I proposed:

    to run the models based on a water vapor distribution that better matches the observations.

    So GCM’s are a tool to investigate “what if” scenarios. They are not gospel. If they don’t show a MWP they are missing something.

  232. Mark T
    Posted Apr 21, 2009 at 12:24 PM | Permalink

    Re: Mark T (#510),

    Perhaps if you had originally stated that “this paper has been discussed and found lacking”

    adding the “originally” makes a difference in the language.

    Mark

  233. bender
    Posted Apr 21, 2009 at 2:10 PM | Permalink

    Re: Mark T (#510),
    Baloney. You can’t both love and hate tree ring based climate recons at the same time. If you love ‘em, you are obliged to say why. The burden of proof is not on the skeptic. Sorry, it is that simple.

  234. bender
    Posted Apr 21, 2009 at 2:38 PM | Permalink

    Re: Mark T (#510),

    Simply stating “IIRC I was among the first to discuss it here. Maybe Ron can figure out how to use the search tool to see where?” after the fact is disingenuous.

    Here you go:
    bender’s beef about Biondi’s PDO recon

    Andrew should recall this argument very well. I’m surprised he hasn’t leaped to my defence.

  235. Mark T
    Posted Apr 21, 2009 at 2:52 PM | Permalink

    Re: bender (#508), Bullpuckey. It depends upon the context and what is actually being used and how/why it is being used. Tree rings may well be good for something, I don’t know, but to simply dismiss them because “you can’t both love and hate tree ring based climate recons at the same time” is a ridiculous argument.

    I put no burden of proof on you whatsoever other than asking for a clear statement of WHY you ignored his response after asking for “evidence” which he supplied in the form of a paper. You simply provided an arrogant reply requiring him proving the point about tree rings, failing to mention that you’ve even discussed his specific evidence in the past till I brought it up. All you had to do was acknowledge that you had discussed it before, nothing more.

    I understand your frustration about the lack of rigor in climate science, but summarily dismissing every argument put in front of you only offering cryptic responses and expecting everyone to assume that you’ve already convincingly covered something in the vast volume of discussions you’ve had (or anyone else has had) is a bit over the top.

    Mark

  236. Mark T
    Posted Apr 21, 2009 at 3:11 PM | Permalink

    Re: bender (#509), For the record, I was not offering any defense of that particular article.

    Mark

  237. Andrew
    Posted Apr 21, 2009 at 5:49 PM | Permalink

    Re: Mark T (#513), I actually understand bender on this point. If we are going to be weary/cautious of tree rings, one should apply such a thing across the board. If it can be show that a unique and truly linear relationship exists to something, then tree rings might allow one to reconstruct past variability. Even considering that the premise thus stated is probably impossible, there is still the possibility that the connection established as valid does not apply going back in time. Still, I’m not sure I’d throw out trees as having no possible useful information just yet. Even though I once stated on this site that “trees cannot be trusted (IMHO)”.

    See bender, I can grow! ;)

  238. bender
    Posted Apr 23, 2009 at 8:35 AM | Permalink

    Re: Mark T (#512),
    More baloney from you. Citing a paper is not “providing evidence”. Let Ron tell me what’s so great about Bratcher & Fiese and then I’ll tell him what’s wrong with it. Let the proponent begin to state his case.

  239. Andrew
    Posted Apr 21, 2009 at 5:52 PM | Permalink

    Re: Andrew (#515), I should add that, at present, I see tree rings as only conveying one thing clearly-whether conditions at that place and at the time you want to look at, conditions were either more or less favorable to tree growth. I have strong agnosticism that they convey information which is anything more specific than that. But I could be wrong.

  240. Ron Cram
    Posted Apr 21, 2009 at 8:02 PM | Permalink

    Re: Andrew (#515),

    Steve has looked into the use of tree rings for other purposes and found them reasonable in areas other than producing temperature reconstructions. I have used the search feature to find the exact exchange but to no avail. But Steve did commend the work of at least one dendrochronologist on this site. I cannot recall his name but his work had more to do with monitoring precipitation or humidity, I believe.

  241. Mark T
    Posted Apr 22, 2009 at 12:15 AM | Permalink

    Re: Andrew (#514),

    I actually understand bender on this point. If we are going to be weary/cautious of tree rings, one should apply such a thing across the board.

    You miss the point. bender’s clearly stated premise was that if you dismiss tree ring reconstructions in any one case, you dismiss them in all. That’s simply not true, statistically, scientifically, or otherwise – there may be legitimate uses (Mannian PCA with BCPs certainly ain’t one of ‘em). That in no way defends the use of the Biondi method or conclusions, but neither should we allow them to be summarily dismissed simply because the Mann et al. usage of tree rings is (deeply) flawed. His objection to the Biondi paper might be a little deeper than that, but it still seems to be rooted in this premise.

    Mark

  242. Andrew
    Posted Apr 21, 2009 at 8:17 PM | Permalink

    Re: Ron Cram (#516), Let me know if you find it. Sounds enlightening! :)

  243. Ron Cram
    Posted Apr 21, 2009 at 11:23 PM | Permalink

    Re: Andrew (#517),

    Okay, I found the comment I was looking for finally. As you can see, Steve has some familiarity with Cook’s work and found it sensible. Steve’s comment opened my eyes not to throw out all the work by dendros. I am not trying to support the paper I cited to Bender (it may be deeply flawed), just show that it is not unreasonable for someone who does not like tree ring thermometry to cite a paper by a dendroclimatologist.

    The more important point is that I provided multiple strands of evidence, from actual predictions to scientists planning actions based on PDO predictions.

  244. Andrew
    Posted Apr 22, 2009 at 7:49 AM | Permalink

    Re: Ron Cram (#518), I’ll look at Cook’s work, but I suspect that it is subject to similar assumptions that temperature recons are. Connections to drought/precipitation may be more robust, but I haven’t looked into them. I’m agnostic about the utility of tree rings to establish paleoclimate, but I do think there is one thing which dendrochronology definitely can tell us-whether conditions at any given time favored tree growth or not. So I’ll look into Cook’s work, keeping an open mind. As I understand it MBH might be considered a case in which my stipulations that the actually show a robust, unique relationship to temperature are not met, so it is appropraite to dismiss them. The PDO work I haven’t looked at too critically (or at all, really) and at the moment I am ambivalent about it. Could be good, could be not, so I give them benefit of the doubt-As I’d have given even to MBH, although I believe in that old Reagan idea “Trust, but verify” and MBH, I think, does not stand up to scrutiny.

  245. Mark T
    Posted Apr 22, 2009 at 11:09 AM | Permalink

    Re: Andrew (#520),

    As I understand it MBH might be considered a case in which my stipulations that the actually show a robust, unique relationship to temperature are not met, so it is appropraite to dismiss them.

    In a bad way, actually. In fact, MBH98 violates it’s own stated assumptions in the very beginning of the paper, i.e., the sorts of “if these conditions are not met…” declarations that are never shown to be met and have since been shown to not be met. Forget even the algorithmic problems, their hypothesis couldn’t even live up to the tests the authors set out for it, let alone those applied by outside auditors.

    The PDO work I haven’t looked at too critically (or at all, really) and at the moment I am ambivalent about it. Could be good, could be not, so I give them benefit of the doubt-As I’d have given even to MBH, although I believe in that old Reagan idea “Trust, but verify” and MBH, I think, does not stand up to scrutiny.

    Neither have I, which is why I did not argue the work should be instantly believed any more than should it be instantly dismissed. All scientific efforts need to be evaluated on their own merits. Take a different application, different data, and a different method, and it might be a legitimate effort. It might not as well.

    Mark

  246. Geoff Sherrington
    Posted Apr 23, 2009 at 2:38 AM | Permalink

    Re: Andrew (#520),

    You might like to look at the rather reasonable points made about dendro work Cook, see

    http://www.john-daly.com/huonpine.htm

    Having visited the region many times, I concur with the weak link between Tasmanian climate on the west coast and recording stations to the East. The Huon Pine is our equivalent of the Bristlecone in some ways, the solitary plank on which a lot od papers rest precariously.

  247. Pat Keating
    Posted Apr 26, 2009 at 7:30 PM | Permalink

    Re: Bob Koss (#609),
    You are of course forgetting (deliberately?) the mindset.

    I once chided friend of mine, after he claimed to be a strong environmentalist, for driving the largest SUV that GM made (before the Hummer). He replied, airily, “Oh, it’s not what you do personally, it’s what you can do politically”. IOW, everyone else must be made to conserve, but not him.

  248. Bob Koss
    Posted Apr 27, 2009 at 8:37 AM | Permalink

    Re: Pat Keating (#622),
    Heh. Yeah, it was deliberate. I know some of those people you described.

    Those who truly believe it to be a worthy purpose would be willing to pay the extra amount. Those who talk the talk, but don’t walk the talk will still do what they want. But a subtle reminder of their hypocrisy every time they buy fuel can’t be a bad thing. Occasionally they may even have a twinge of conscience and furtively glance around to see if anyone is watching while they make their selection. :)

    People would be voting with their wallet. Can’t get more accurate than that for gauging public opinion.

  249. Posted May 3, 2009 at 10:20 PM | Permalink

    Re: Andrew (#727),

    Andrew:
    May 3rd, 2009 at 9:33 pm
    Re: Phil. (#725), That’s a serious accusation. Can you show where he has intentionally (important distinction) told untruths (things he knew to be false) and created false information while knowing perfectly well it was inaccurate? My understand of his “faked” graphs is that he simply used a bad method, with flawed reasoning, but whether he knew he was doing things wrong is another beast entirely. And impostor? Who is he pretending to be?

    He pretends to be “a member of the Upper House of the United Kingdom legislature” in a letter to Olympia Snowe which is untrue, and also uses the badge of the Houses of Parliament on his graphs (which is copyrighted and which he has no right to use).

  250. Mark T
    Posted May 3, 2009 at 10:46 PM | Permalink

    Re: Phil. (#724), You should spend less time hanging out in liberal blogs. From Wiki (bold is mine):

    Lord Monckton has never said he is a sitting member of the House of Lords: he is, however, a member of the Upper House by succession (hence his title), is registered as such on the list of Peers entitled to be elected by his fellow hereditary peers, and, as a member of the House in good standing, is entitled to use its facilities, though not to speak or vote in the Chamber, for it is in this sense alone that the House of Lords Act 1999 removes the right of membership from hereditary Peers. Proposed correction: Preferably, delete this damaging libel altogether. Otherwise, replace by “He is a hereditary peer, but his father’s automatic right to sit and vote, like that of most hereditary Peers, was terminated by the Peerage Act 1999”.

    So, when you said:

    He pretends to be “a member of the Upper House of the United Kingdom legislature” in a letter to Olympia Snowe which is untrue, and also uses the badge of the Houses of Parliament on his graphs (which is copyrighted and which he has no right to use).

    which is essentially libelous, can we then infer that you were knowingly providing false information?

    Oh, and, all other things aside, why am I not surprised that YOU of all people would resort to argumentum ad-hominem rather than simply finding a legitimate argument. Gee…

    Mark

  251. Posted May 4, 2009 at 12:23 AM | Permalink

    Re: Mark T (#726),

    Mark T:
    May 3rd, 2009 at 10:46 pm
    Re: Phil. (#724), You should spend less time hanging out in liberal blogs. From Wiki (bold is mine):

    I don’t hang out in liberal blogs, whatever they are.

    So, when you said:
    He pretends to be “a member of the Upper House of the United Kingdom legislature” in a letter to Olympia Snowe which is untrue, and also uses the badge of the Houses of Parliament on his graphs (which is copyrighted and which he has no right to use).
    which is essentially libelous, can we then infer that you were knowingly providing false information?

    No I was quoting him, see: http://www.ff.org/centers/csspp/pdf/20061212_monckton.pdf

    The only way he can become a Member of The House of Lords is to be elected, from the Act you quoted above: “No-one shall be a member of the House of Lords by virtue of a hereditary peerage.” Additionally, the House symbol is primarily used to authenticate communications from Members, Monckton is not a Member of the House and his use of it on his material is inappropriate.

    Oh, and, all other things aside, why am I not surprised that YOU of all people would resort to argumentum ad-hominem rather than simply finding a legitimate argument. Gee…

    His credibility is in question, it’s not ad hominem, on the other hand he throws ad hominem attacks around without hesitation. As to legitimate argument I already posted on here yesterday that the data used in his graphs did not match the data presented in the IPCC report.

  252. Mark T
    Posted May 4, 2009 at 2:03 AM | Permalink

    Re: Phil. (#727),

    No I was quoting him

    I never said he didn’t make the claim, I said his claim was legitimate, as I have shown you. You, on the contrary, made a libelous statement:

    He pretends to be “a member of the Upper House of the United Kingdom legislature” in a letter to Olympia Snowe which is untrue, and also uses the badge of the Houses of Parliament on his graphs (which is copyrighted and which he has no right to use).

    Bold mine. You’re wrong. He did not misrepresent himself, which is why the entry was stated as I quoted (and the accuastion removed from the original page). As he is a member by title, he is also entitled to use the seal as a result. So, again, I ask, did you willfully lie, or did you just not do enough research?

    His credibility is in question, it’s not ad hominem, on the other hand he throws ad hominem attacks around without hesitation.

    So, he’s wrong because of his credibility? Um, an argumentum ad-hominem, in spite of your weak pleas to the contrary.

    As to legitimate argument I already posted on here yesterday that the data used in his graphs did not match the data presented in the IPCC report.

    Interesting that you chose a lie in the quote I referenced. Hmmm… And, what evidence do you have that he “willfully” posted graphs that did not match? Willful intent was the original question posed to you, was it not?

    Mark

  253. Posted May 4, 2009 at 9:18 AM | Permalink

    Re: Mark T (#730),

    Re: Phil. (#727),
    “No I was quoting him”
    I never said he didn’t make the claim, I said his claim was legitimate, as I have shown you. You, on the contrary, made a libelous statement:
    “He pretends to be “a member of the Upper House of the United Kingdom legislature” in a letter to Olympia Snowe which is untrue, and also uses the badge of the Houses of Parliament on his graphs (which is copyrighted and which he has no right to use).”
    Bold mine. You’re wrong. He did not misrepresent himself, which is why the entry was stated as I quoted (and the accuastion removed from the original page). As he is a member by title, he is also entitled to use the seal as a result. So, again, I ask, did you willfully lie, or did you just not do enough research?

    Being allowed to use the Dining room at the House of Lords does not make one a member of the legislature! In fact in the direct quote from the appropriate statute (which you deleted) it explicitly states that he can not be a Member by virtue of his title.
    Here it is again: “No-one shall be a member of the House of Lords by virtue of a hereditary peerage.”
    Further: “The Act deprives excluded hereditary peers of all the privileges of membership of the House of Lords, including the privileges they enjoyed as members of Parliament. Parliamentary privileges cover various matters, many of which relate to the House of Lords as a whole (such as punishing improper conduct within the House itself), but include some that are personal to individual peers. One of the most important personal privileges is that no action can be taken against a peer for what he or she may say in Parliament. Hereditary peers excluded by the Act also lose the right to be paid allowances and to use the facilities of the House that are available to members, such as its library, research and restaurant facilities. The removal of these rights does not prevent the House from deciding to grant some rights to use the facilities of the House to a hereditary peer under the exercise of its own authority.” So unless he’s been granted special permission he can’t even use the Dining room!
    http://www.opsi.gov.uk/acts/acts1999/en/ukpgaen_19990034_en_1

    Also if he wishes to use a seal on his graphs he should use his own, by using the badge of the Houses of Parliament he might mislead readers into thinking that it’