Seems that there’s some excitement about a new paper A Statistical Analysis of Multiple Temperature Proxies: Are Reconstructions of Surface Temperatures Over the Last 1000 Years Reliable? to be published in Annals of Applied Statistics. Their reconstruction appears to be closest match to a hockey stick shape yet seen:

i-1d91f371ed0bb1448203df8697493ae7-mcshanewynerhockey.png

Also:

Using our model, we calculate that there is a 36% posterior probability that 1998 was the warmest year over the past thousand. If we consider rolling decades, 1997-2006 is the warmest on record; our model gives an 80% chance that it was the warmest in the past thousand years.

Discuss.

Update: Deep Climate on M&W is worth a read.

[Update 2: Eduardo Zorita.

Comments

  1. #1 rms
    August 16, 2010

    That same paper, in the conclusions say: “Natural climate variability is not well understood and is probably quite large. It is not clear that the proxies currently used to predict temperature are even predictive of it at the scale of several decades let alone over many centuries.”

  2. #2 pough
    August 16, 2010

    The hockey stick is dead! Long live the hockey stick!

  3. #3 wag
    August 16, 2010

    This falls into the category of deniers not having the analytical chops to tell the difference between “different” and “wrong.”

  4. #4 jre
    August 16, 2010

    John Mashey, consistently one of the best commenters here (or anywhere), has dropped a hint that he has something to add to this topic. I hope that he will not keep us waiting long.

  5. #5 Martin Vermeer
    August 16, 2010

    The funny thing is that this paper actually replicates Mann et al. 2008 without even noticing it…

    To partake in this dirty little secret, see their Figure 14 on page 30: the blue curve is wiggle-identical and practically a photocopy of Mann’s corresponding EIV NH land curve. As it should be. The higher (green) curve they canonize and which is shown above is the result of an error: they calibrate their proxies against hemispherical mean temperature, which is a poor measure of forced variability. The instrumental PC1 which the blue curve is based on, is a much better measure; its EOF contains the polar amplification effect. What it means is that high-latitude proxies, in order to be made representative for global temperatures, should be downweighted. The green curve fails to do this. Thus, high latitudes are overrepresented in this reconstruction, which is why the “shaft” is at such an angle, due to the Earth axis’s changing tilt effect on the latitudinal temperature dependence described in Kaufman et al. 2009.

    The authors have no way of detecting such an error as their RMSE goodness-of-fit seems to be also based around the hemispherical average…

  6. #6 t_p_hamilton
    August 16, 2010

    It looks like McShane and Wyner heard somebody say to check your work, and made their work look like a check mark.

    An absurdly linear and sloping temperature graph from 1000 to 1850. The answer to the question in their title, is

    Yes, but not ours.

  7. #7 Marco
    August 16, 2010

    Martin, serious comment:
    Perhaps someone who has some knowledge in this field should warn the authors of their fatal flaw?

    Gives them a chance to prevent major embarrassment. Of course, thanks to the enormous cabale from the Wattsians and climateauditors, they’ve already been embarrassed, but at least they can prevent embarrassment amongst their peers.

    If this does not happen, I sincerely hope someone will start writing a comment soon. Sad for McShane and Wyner, but in this hotly contested area there is no room for papers with fundamental flaws.

  8. #8 Bob O'H
    August 16, 2010

    Boy, it’s a good job these analyses weren’t done by Brits or Aussies – can you imagine the Cricket Bat Graph?

  9. #9 SteveWW
    August 16, 2010

    That same paper, in the conclusions say: “Natural climate variability is not well understood and is probably quite large. It is not clear that the proxies currently used to predict temperature are even predictive of it at the scale of several decades let alone over many centuries.”

    Can somebody help me out here – this is using proxies to predict *past* temperatures, right? I’ve never heard of anyone using proxies to predict future temperatures… it’s not really clear to me what these two sentences mean.

  10. #10 Paul UK
    August 16, 2010

    So does this explain why the Romans apparently made excellent wine in the UK!

    OK, so they reckon we should be heading for an ice age, but fossil fuels have reversed that trend.

  11. #11 John Mashey
    August 16, 2010

    Others are taking care of stats… but I have some other info:

    1) This is ~ Wegman Report, the Sequel …

    Among other things, even some oddities of wording seem to derive from the WR.

    People may recall from Deep Climate, 07/29/10, that ~10 pages of Wegman Report is plagiarized from various sources, including Bradley(1999).

    Start with Bradley(1999), p.1, first sentence:

    “Paleoclimatology is the study of climate prior to the period of instrumental measurements…”

    See Wegman Report, specifically p.10, which essentially/reasonably summarizes Bradley(1999) pp.1-10, i.e., not part of the plagiarism. It starts:

    “Paleoclimatology focuses on climate, principally temperature, prior to the era when instrumentation was available to measure climate artifacts.”

    *artifacts* is odd terminology that I don’t recall seeing elsewhere.

    McShane&Wyner have:
    “Paleoclimatology is the study of climate and climate
    change over the scale of the entire history of earth. A particular area of focus
    is temperature….
    The key idea is to use various artifacts of historical periods which
    were strongly influenced by temperature and which survive to the present.”

    *artifacts* again.

    2) Although not as bad as the WR, no black helicopters guys, the Bibliography includes:

    – BBC

    - AIT

    - Green, K.C. Armstrong, J.S., and Soon… (recall that Armstrong is at Wharton)

    - 3 WSJ OpEds on climategate

    - Lamb(1990)

    - Matthes (1939)

    - Rothstein, NY Times article cited as alarming both populace and policy makers.

    3) They cite Wegman often. I believe they will come to seriously regret that in the near future.

    4) The results will stand or be utterly demolished on their own merits, but as background, who are these guys that Judith Curry labeled leading statisticians?

    McShane is a May 2010 PhD from Wharton, Wyner is his Dissertation director.

    WHYNER
    Pubs, through 2003.
    Pubs, Google Scholar, including 3 on baseball (or Bayesball).

    For several years, he contributed (as “Adi” to a group blog, now dormant Politically Incorrect Statistics, occasionally touching upon climate. These might be used to calibrate his (and colleague’s) level of climate expertise.

    http://picstat.blogspot.com/2008/08/is-it-really-so-simple.html August 27, 2008

    http://picstat.blogspot.com/2008/07/vanishing-temperature-trends.html July 29, 2008

    http://picstat.blogspot.com/2008/05/sea-ice-continued.html May 04, 2008

    “Back in 1975, when we are at the end of a 30 year period of declining global temperatures, the consensus among the climate scientists was a coming ice age…”

    http://picstat.blogspot.com/2008/05/sea-ice-continued.html may 04, 2008

    Sea ice variations are just normal, and besides Antarctic ice growing…

    http://picstat.blogspot.com/2008/05/southern-hemisphere-sea-ice.html May 01, NSIDC must be wrong…

    http://picstat.blogspot.com/2006_08_01_archive.html August 07, 2006

    not by Wyner, but by his colleague Dean Foster)

    http://picstat.blogspot.com/2005/11/greenhouse-gases-increasing-but-still.html November 25, 2006

    (again, not by Wyner, but by Dean Foster)

    Like Wegman, who often gave talks to audiences unlikely to have much climate expertise, we have:

    Wyner gave talk March 2010:
    http://stat.wharton.upenn.edu/~gadam/seminar_files/Abraham%20Wyner%20-%20Title%20and%20Abstract.pdf

    People may wish to read this, but put down coffee first. I’m sure all will be pleased to see:

    “The relationship between proxies and temperature is weak”

    MCSHANE:

    Ph.D. in Statistics, May 2010

    Thesis: Integrating Machine Learning Methods with Hidden Markov Models: A New Approach to Categorical Time Series Analysis with Application to Sleep Data

    Thesis Advisor: Abraham Wyner, Department of Statistics

    Marketing Advisor: Eric Bradlow, Department of Marketing

    His C.V is eclectic, including modeling sleep in mice and several baseball papers. It is impressive that he managed to become a paleolclimate expert also.

    Again, one finds a talk to a non-climate audience:

    “Are Reconstructions of Surface Temperatures Over the Last 1000 Years Reliable?”

    Presented
    February 2009 at Information Theory and Applications Workshop, San Diego, CA.” that’s:

    http://ita.ucsd.edu/workshop.php?submitted=1

    http://ita.ucsd.edu/workshop/09/talks/

    Wyner organized session, McShane presented…

    This conference covers a vast range of topics.

    As one last tidbit, I’ve had reason to recently consult American Statistical Association Ethical Guidelines, a good document.

    While “How to Lie with Statistics” is famous, most statisticians do not lie, in my experience….

    What might we find, examining that document:

    “A. Professionalism

    1. Strive for relevance in statistical analyses. Typically, each study should be based on a competent understanding of the subject-matter issues…”

    Oh.

  12. #12 TrueSceptic
    August 16, 2010

    I hope Tim doesn’t mind me repeating my post from the previous thread …

    Tim,
    That’s what I thought. In fact, the “stick” is merely tilted and the “shaft” is actually straighter than MBH’s. Amusingly, the MWP and LIA have pretty much disappeared, having been replaced by a steady downward trend from 1000 to the early 1800s, with a slight dip (LIA?) centred on around 1500. So, when the authors say

    On the one hand, we conclude unequivocally that the evidence for a ”long-handled” hockey stick (where the shaft of the hockey stick extends to the year 1000 AD) is lacking in the data.

    I have to ask: “What, are you blind!?”.

    They also say

    M&M observed that the original Mann et al. (1998) study (i) used only one principal component of the proxy record and (ii) calculated the principal components in a ”skew”-centered fashion such that they were centered by the mean of the proxy data over the instrumental pe- riod (instead of the more standard technique of centering by the mean of the entire data record). Given that the proxy series is itself auto-correlated, this scaling has the effect of producing a first principal component which is hockey-stick shaped (McIntyre and McKitrick, 2003) and, thus, hockey- stick shaped temperature reconstructions. That is, the very method used in Mann et al. (1998) guarantees the shape of Figure 1.

    Oh, really?

  13. #13 Doug
    August 16, 2010

    One of the criticisms put by the deniers is that climate scientists don’t work close enough with statisticians. There is some validity to this comment and Mike Mann and Gavin Schmidt have commented that efforts are underway to improve this (Not that I think climate scientists have done anything significantly remiss in this area).

    But from the comment of Martin’s above I might conclude that this is an example of two statisticians that needed to work closely with climate scientists, sepcifficaly some Paleoclimatologists.

  14. #14 dhogaza
    August 16, 2010

    The results will stand or be utterly demolished on their own merits, but as background, who are these guys that Judith Curry labeled leading statisticians?

    She also claims this paper “looks like the real deal” in the same comment.

    Anyone really doubt Judith is anything other than a vanilla denier at this point in time?

  15. #15 TrueSceptic
    August 16, 2010

    11 John Mashey,

    Excellent, as always.

    The use of “artifact” really is odd, isn’t it, suggesting that the misuse was copied. The word’s meaning doesn’t really allow its use in that way.
    —-
    1 an object made by a human being, typically an item of cultural or historical interest : gold and silver artifacts.
    • Archaeology such an object as distinguished from a similar object naturally produced.
    2 something observed in a scientific investigation or experiment that is not naturally present but occurs as a result of the preparative or investigative procedure : widespread tissue infection may be a technical artifact.
    —-

  16. #16 Jeremy C
    August 16, 2010

    Tim,

    You say “discuss”! Way no! I am laughing too hard at the graph! I showed the cat the graph and its laughing so hard its already brought up 3 big fur balls.

    Good grief!

  17. #17 chek
    August 16, 2010

    Yeah, but these error bars go up to eleven.

  18. #18 Hank Roberts
    August 16, 2010

    Well, the version linked at
    http://www.imstat.org/aoas/next_issue.html
    still reads like a draft. If it is, the authors, or the referees, or the journal editors may have time to address the problems as they are being pointed out.

    Agreeing with Marco above — Martin, you might be someone who they’d listen to if you’re going to point out the problem you found before they go to publication.

    If “blog review supplementing peer review” is going to become a trend, might as well make it a positively useful one that authors can learn to take note of.

  19. #19 Jeremy C
    August 16, 2010

    But seriously folks.

    Who will join with me in funding for the next Heartland anti-AGW conference a team of US cheerleaders to gate crash the keynoter running down the aisle all carrying hockeysticks above their heads! We’ll have them running onto the stage behind the keynoter where they will chant, “McShane!!…. Wyner!!!….AND WEGMANNNNNN!!!!!!!”. On the last they’ll toss the hockey sticks into the air.
    |
    |
    |
    |
    All in the best possible taste.

  20. #20 SteveC
    August 16, 2010

    Thanks for posting this Tim. I saw that Watts had posted on this a few days ago, but a lot of the technical stuff is a bit dense and thus over my head, so I left it to pore over another day.

    For anyone with the stomach for it, the McShane and Wyner paper is headline news at WTFUWT, but make sure you pinch your nose and block your ears before diving into the swamp. As a preview, Watts says:

    Commenters on WUWT report that Tamino and Romm are deleting comments even mentioning this paper on their blog comment forum. Their refusal to even acknowledge it tells you it has squarely hit the target, and the fat lady has sung – loudly.

  21. #21 Steve Reuland
    August 16, 2010

    Someone correct me if I’m wrong, but in my cursory look at the paper the authors seem to be saying the following: 1) We get basically the same result as past studies. 2) But it’s tilted slightly! 3) And our error bars are really big, meaning that there’s some small probability that past temperatures were, like, far off from the graphed line.

    Which boils down to saying that there is a bit of extra uncertainty in temperatures for the past 1000 years compared to other studies. Okay. But apparently the uncertainty is not enough such that we can no longer say that recent temperatures are likely warmer than the past:

    For example, 1998 is generally considered to be the warmest year on record in the Northern Hemisphere. Using our model, we calculate that there is a 36% posterior probability that 1998 was the warmest year over the past thousand. If we consider rolling decades, 1997-2006 is the warmest on record; our model gives an 80% chance that it was the warmest in the past thousand years. Finally, if we look at rolling thirty-year blocks, the posterior probability that the last thirty years(again, the warmest on record) were the warmest over the past thousand is 38%.
    [...]
    For k = 10, k = 30, and k = 60, we estimate a zero posterior probability that the past thousand years contained run-ups larger than those we have experienced over the past ten, thirty, and sixty years (again, the largest such run-ups on record). This suggests that the temperature derivatives encountered over recent history are unprecedented in the millennium.

    I must say, this does not exactly convince me that we are not currently in an anomalously warm period, or that the current rise in temperatures is normal.

    I’m not sure why the denialists are happy with this result, other than as a desperate salvo in their bizarre and ill-conceived War on Michael Mann.

  22. #22 chek
    August 16, 2010

    [Steve C said:](http://scienceblogs.com/deltoid/2010/08/a_new_hockey_stick_mcshane_and.php#comment-2730330) “(Watts says) it tells you it has squarely hit the target, and the fat lady has sung – loudly”.

    Strangely enough, I rather think it tells me that a desperate and ridiculous mustachioed denialist twit has jumped the gun on something he hasn’t understood (Pielke must be away on vacation).

  23. #23 Dan
    August 16, 2010

    Okay, hoping this doesn’t come across as concern-trolling. It’s not meant to be, but who knows…

    I read the paper pretty closely last night. Some parts of it (the statistics) I understand pretty well, but I really don’t know anything about climate science, so there’s some things that puzzle me and I wanted to ask about. The parts of the paper that seemed to present worries were:

    1. When you account for uncertainty about parameter estimates as well as noise in the data, the error bars widen a lot. That seems to be the main message behind Figure 16 (which Tim has posted) and the somewhat melodramatic Figure 17. To my mind (as someone who knows stats but not climate science) this seems pretty obvious, but nevertheless worth pointing out (unless someone else has already done so in the relevant literature).

    2. The near-linearity of the backcasts (I love that word) may be somewhat illusory. I don’t think the paper goes very far on this point, but to take a simple example, the Bayesian posterior predictive distribution shown in Figure 16 is the aggregate across all possible backward projections of the temperature series, weighted by how closely they overlap with the instrumental record. Each individual backcast can be quite nonlinear, but because we don’t know which one to trust, we average them, and the average looks quite linear. If the proxy data were pure autoregressive noise, then I *think* I could show that the average would necessarily be linear, but the real data are neither simple autoregressive processes, nor (we hope!) are they unrelated to the true temperatures. Again, to me at least this point is not highly surprising: lots of real world data sets seem to have this “averaging looks linear” character. Not sure whether it tells you much about the strength of the temperature signal in the proxy data.

    3. A bit more worryingly, they find some evidence (not strong, but not trivial as far as I can tell) that the proxies aren’t able to capture a very sharp change in temperatures. This is their lower panel of Figure 18. Training up the model on most of the data, and using the most recent 30 years as hold out data shows that the model does not predict the current sharp warming trend. I think they do a pretty good job of trying to say that this doesn’t necessarily mean that the proxies are no good; what I think they’re trying to emphasise is the fact that the real data (black lines in that plot) actually go outside the error bars, meaning that if the true historical temperature record included very sharp rises just like the ones we’re seeing now, the model may be unable to detect them. This did surprise me a bit… I’d be curious to hear whether other people (esp. people who – unlike me – might know some climate science) find that to be a problem.

    4. Saving the thing that really did worry me for last… when fitting the model to the instrumental record, using real proxies as the predictors outperforms very weak null models, but is consistently slightly worse than using autoregressive random noise as the predictors (e.g., Fig 9 and Table 1). In one sense, I can see that this isn’t a problem: because of the autocorrelation, AR1 noise models do capture some of the short range temporal structure in the data. So they’re not purely “noise” models. On the other hand, by definition they don’t have the ability to capture any long-range signal. By extension, the concern it raises is whether the proxies actually have a strong enough long-range signal that would allow large scale temperature reconstructions work. That is, over the time period for which we do have data, the proxies perform no better than other predictors that only capture short-range structure. As such, it’s hard to see how the instrumental record can serve as a method for verifying that the proxies carry a long range signal. I actually can’t see a flaw in this argument, if it’s construed narrowly. For instance, it doesn’t rule out the possibility of long-range a signal in the proxies: it just seems to suggest that this isn’t the way to find it. Again though, I worry that my near-total lack of knowledge about climate science means I might be missing something.

    In any case, that’s what I got out of the paper from a stats perspective. But obviously there’s got to be specifics to this that depend a lot on substantive knowledge of, say, climate science. For instance, Martin Vermeer’s comment above about the latitudinal issue is something I wouldn’t have known (as an aside, at least one of their analyses aimed to model the local temperature data individually rather than fitting the average – see section 3.6 – but I wouldn’t have a clue if this addresses the concern!) So I’m wondering what it is that I’m missing. I’d be really interested to hear from people who actually know the subject if I’m missing something important in summarising the paper this way.

    One last thing: in all of the above I’ve ignored the points that affirm my previously held beliefs, that the recent temperatures are very high when compared to the historical record, since (a) I already believed them, and (b) Tim’s post summarised them nicely.

  24. #24 Chris O'Neill
    August 16, 2010

    Steve Reuland:

    I’m not sure why the denialists are happy with this result

    I think it’s known as “proof by contradictory citation”.

  25. #25 Eli Rabett
    August 16, 2010

    Eli explains it all Well, at least some of it

  26. #26 Dean Morrison
    August 16, 2010

    Seems to me that the denialists shouldn’t be too happy about a paper that blows any ‘Medieval Warming Period’ out of the window. If there’s isn’t enough data for a hockey stick handle, there certainly isn’t enough evidence for a convenient hump.

    “As mentioned earlier, scientists have collected a large body of evidence which suggests that there was a Medieval Warm Period (MWP) at least in portions of the Northern Hemisphere. The MWP is believed to have occurred from c. 800-1300 AD (it was followed by the Little Ice Age). It is
    widely hoped that multi-proxy models have the power to detect (i) howwarm the Medieval Warm Period was, (ii) how sharply temperatures increased during it, and (iii) to compare these two features to the past decade’s high temperatures and sharp run-up. Since our model cannot detect the recent temperature change, detection of dramatic changes hundreds of years ago seems out of the question.”

    I’m also puzzled about their worries about the lack of ‘predictive’ power of proxies. If we had a perfect instrumental record for the last thousand years, that wouldn’t predict anything either.

  27. #27 apeescape
    August 16, 2010

    Correct me if I’m wrong, but Mann et al. (2008) and Li et al. (2007) seems to smooth their proxies (decadal-scale) first, then does the modeling? OTOH, McShane and Wyner (p26, 2010) doesn’t seem to. Can this add (the interannual variability) to the uncertainty in MW2010’s reconstructions?

  28. #28 Susann
    August 16, 2010

    I openly acknowledge that I am not up to snuff when it comes to analyzing the paper in terms of the validity of the methods and conclusions based on them. I am in the basic stats knowledge category, not much beyond simple stats, variance, regression analysis, etc. However, the authors left a big gaping hole open into which they might retreat at some point — they state quite clearly that they have not addressed any data quality issues and have accepted the data at face value. “Skeptics” can always fall back on claims about data quality if the paper’s methodology and conclusions are criticized and/or found lacking — you know, all those unjustified adjustments, smoothing, filling in missing data, bad proxies, etc. In fact, if you can stomach it, the comments at WTFIUWT already are pointing to these issues when the fact of the missing MWP and LIA is brought up, retreating to criticisms of proxy choices and data quality.

  29. #29 Dan
    August 16, 2010

    Ah. Eli’s post and the links therein helped, and makes more sense out of Fig 9 in the paper. What they’re calling the grid-proxy reconstruction is where they’re predicting the full Z matrix of local instrumental records, rather than the averaged time series Y. The point that several commenters seem to have made in various places is that the real proxies are highly sensitive to local variation, so you’d expect better performance when predicting Z, not Y.

    Okay, that makes sense, and that method does seem to perform much better. It’s still not better than the random autoregressive model, but maybe that just means that in order to capture the structure of the existing instrumental data your model only actually need to have short-range structure built in? Especially if your error measure includes both RMSE over interpolation holdout blocks as well as extrapolation holdout blocks. Under those circumstances a model that has short-range only (e.g., any of their AR1 pseudo-proxies) might be expected to perform just as well as a model that has both short-range structure and long-range structure (which is what we believe the real proxies have). I think I buy that (my ignorance makes me uncertain), but that means that we’re still in the position of needing to rely on other reasons to trust that the proxies do encode something about the temperature record over the longer time scales, right, since the instrumental record (i.e., the matrix Z) doesn’t disambiguate between the real and pseudo proxies. But I assume that’s not a serious worry, and that there are in fact substantive reasons to believe this. I can’t imagine that climate scientists go around collecting arbitrarily chosen data and trying to correlate any- and every- thing to the temperature record. Rather, it must be the case that the proxies are selected on other substantive grounds, meaning that there’s a priori reasons to believe that a long-range signal exists. As such, the fact that the model fitting can’t actually verify that this signal exists is somewhat irritating, but not as damaging as I thought when I first read the paper. (Yes?)

    So the logic here is that fitting a model to the instrumental data isn’t meant to serve as evidence that the proxies actually contain a long-range signal; rather, we believe that for other reasons entirely. Instead, what we’re trying to do is parameter estimation, not model validation; that is, figuring out exactly how to calibrate the proxies against the temperature record. Is that right?

  30. #30 Eli Rabett
    August 16, 2010

    Well, and figure out which proxies have a (significant) temperature response over any period. . . .

  31. #31 sod
    August 16, 2010

    thanks to Tim for tackling this one early. (in contrast to the usual denialist conspiracy theory about information being supressed..)

    and special thanks to John Mashey, for his analysis. as always very good work!

    on collide a scape Judith Curry said this:

    leading statisticians, to be published (in press) by a leading statistics journal.

    as John has pointed out, the first claim is at best a serious stretch and the second half is also rather weak, as the magazin is a rather young spin off from another magazin.

    she of course continued to say:

    This paper looks like the real deal to me

    which seems to be a rather hasty assessment of a paper, that hasn t even been printed yet.

    she alos swallowed the blogs are blocking comments on the paper conspiracy theory:

    To be expected from Romm. I would have expected better from Tamino (but not anymore). Will be interesting to watch RC.

    why does Judith Curry constantly make such strong claims about stuf that she obviously does not understand or even know?

    why is she constantly making claims that are obviously false?

    and why have we, (as always) not seen her take back any of these false claims so far?

  32. #32 Dan
    August 16, 2010

    and figure out which proxies have a (significant) temperature response over any period

    Hm. I think I’m confused again. It’s the “over any period” phrase that I’m unsure about… Sorry if I’m being obtuse, but do you that mean that the strength of the correlation between a particular proxy and the temperature record is expected to fluctuate? That would imply that the estimated regression-weights in the Lasso-regression model would themselves be time-dependent. Unless I missed something, their model doesn’t do that. And given the nature of the data I can’t imagine how you could even do that safely.

    Or do you just mean that part of the calibration process involves determining which proxies (or principal components thereof, etc etc) should be assigned non-zero regression weights on the basis of the known instrumental record? That makes perfect sense to me. The reason I’m confused is that for the Lasso regression model that tends to be folded into the parameter estimation procedure. That is, the penalty term in the Lasso regression model (p13) tends to solve the “variable subset selection” problem by pushing a lot of the regression weights to zero, and so estimating the model parameters also involves removing proxies that don’t correlate very well with the data.

  33. #33 sod
    August 16, 2010

    Dan, i am not a specialist in proxy records and i haven t had the time to really read this paper.

    but i might have some minor answers to your questions.

    it is clear without any statistical knowledge, that individual proxies might not be accuate at an annual or decadal level. pests can hamper tree growth, wind blow away accumulating snow and a fish hiding eggs could destroy layers of sea sediments.

    i also was following the Loehle paper closely, and there was a detailed look at the majority of the proxies over at climateaudit. many proxies show really erratic changes over short time periods ( think they were called “high frequency proxies”) which clearly do not track short term variations well.

  34. #34 Paddy Constantine
    August 16, 2010

    I must confess I always dreaded playing hockey when I was at school. Seeing this updated model of Mann et al.’s reconstructed millennial temperature graph, fills me with the same cold, anxious feeling of another 3 hour class on a dreary, wet, winters morning in rural Staffordshire! Standing there caped to the high heels in rotten mud I would always hear “Paddy, stop standing still, and get stuck in” as I shivered in the rain, with drips falling down my cheeks!

    So, back to fact of the matter, Tim Lambert’s version of Fig. 14 in McShane and Wyner claims to clear the original MBH “hockey-stick”, with its clever modification of a downward sloping handle on the failure of McS&W. Correct me If im wrong but the MBH hockey stick had a horizontal (no slope) handle? and the McS&W downsloping handle clearly shown in Tim’s version actually restores the MWP that was abolished by MBH.

    To me there is a small problem with the upsloping blade being used to signify the instrumental record since 1850 in both MBH and McS&W, which is that the CRU dataset they use is, i think anyway is a little fictitious until well into the 20th century. From the bits of research i have under-taken I do believe that there was in reality no instrumental coverage of more than half of the globe until the 1950s, with the non-covered areas almost wholly in the tropics. Therefore this to me could be used to create a false cold “global” anomaly in 1850-1950 relative to the warmer anomaly base period of 1960-1990 which does include the tropics?

    It is indeed surprising that while McS&W checked the origin of the proxy data in MBH etc, they might of failed to check the CRU “time series”? Had they done so the upsloping blade might largely disappear?

    Sorry for throwing a spanner in the work folks, its just in my eyes the Hockey stick brings back bad, bad memories! Good day to you…

  35. #35 Chris O'Neill
    August 16, 2010

    Paddy Constantine:

    “Paddy, stop standing still, and get stuck in” as I shivered in the rain, with drips falling down my cheeks!

    So your classes were held out in the rain. No wonder you didn’t learn anything.

  36. #36 Dave Springer
    August 16, 2010

    Bravo to: Dan | August 16, 2010 6:33 PM

    You got it right. The gist of the paper is that the proxy record can’t pick up temperature increases as rapid as what occurred in the latter half of the 20th century.

    This is where the infamous Climategate “trick” to “hide the decline” comes into play.

    The proxy record failed to show the temperature increase in recent decades that was indisputably measured across the globe by satellites that became operational in 1979.

    So what Mann and his cohorts in crime (just an expression) did was drop the disagreeable portion of the tree ring proxy record from 1960 onward and stitch in the satellite temperature record in its place.

    The long and the short of it is that the blade portion of the hockey stick could have occurred once or even many times in the past 1000 years and the tree ring proxy would not have shown it since it failed to show it in the single instance where we know it occurred.

    Mann and company knew about this problem with the proxy data all along and purposely tried to hide it.

  37. #37 Joel Shore
    August 16, 2010

    Paddy: I think your statement about a lack of coverage in the tropics (independent of the merits of how true that is or not) reveals some misunderstanding of how graphing temperature anomalies works. One important reason to use anomalies rather than absolute temperatures is that, with anomalies you should not have biases introduced by whether you have better coverage in warmer or colder places. (Or, at least, what biases exist because of this should be much weaker…and much less obvious which direction they would even be in.)

  38. #38 Dave Springer
    August 16, 2010

    I have a question for Dan.

    You picked up on the contents of the paper nicely but why did it “worry” you?

    I should think you’d be worried that Mann is right and the earth is warming at an unprecedented rate. This paper reveals that Mann showed no such thing and that the earth could have warmed at this rate many times in the past.

    Do you -want- a climate catastrophe and you’re worried that you might not get what you want?

    I mean this appears to be the mental state of most of catastrophic global warmists. They seem to have some sort of messiah complex where they envision themselves as saving the world. If there’s no catastrophic warming actually happening then there’s nothing to save and hence they can’t be heroes and instead look like a bunch of chicken littles. That’s the real problem here. Where there should be dispassionate scientists there are instead wannabe super-heroes.

  39. #39 Chris O'Neill
    August 16, 2010

    Dave Springer:

    the tree ring proxy would not have shown it since it failed to show it in the single instance where we know it occurred.

    Not true. The proxy record showed the increase from 1910 to 1945.

  40. #40 t_p_hamilton
    August 16, 2010

    Dave Springer said:”The long and the short of it is that the blade portion of the hockey stick could have occurred once or even many times in the past 1000 years and the tree ring proxy would not have shown it since it failed to show it in the single instance where we know it occurred.”

    The blade starts well before 1960, and the proxy did “show it” in that “instance”.

    The logic fails because of an incorrect premise, based on a lack of knowledge.

  41. #41 jakerman
    August 16, 2010

    >*The long and the short of it is that the blade portion of the hockey stick could have occurred once or even many times in the past 1000 years and the tree ring proxy would not have shown it since it failed to show it in the single instance where we know it occurred*

    Even with the errors pointed out [by Martin](http://scienceblogs.com/deltoid/2010/08/a_new_hockey_stick_mcshane_and.php#comment-2729979), McShane and Wyner still find that:

    >*”Using our model, we calculate that there is a 36% posterior probability that 1998 was the warmest year over the past thousand. **If we consider rolling decades, 1997-2006 is the warmest on record; our model gives an 80% chance that it was the warmest in the past thousand years”**.*

  42. #42 Vorlath
    August 16, 2010

    True that the graph at the top of the article is the correct reconstruction using Mann’s data. But what the paper specifically says is that graph is 100% useless. It’s no good for determining past temperatures because random data actually correlates to instrumental temperatures better than that graph seen at the top of the article.

    Get it? RANDOM DATA IS BETTER THAN THAT GRAPH!

  43. #43 Stephen Gloor (Ender)
    August 16, 2010

    Dave Springer – “The long and the short of it is that the blade portion of the hockey stick could have occurred once or even many times in the past 1000 years and the tree ring proxy would not have shown it since it failed to show it in the single instance where we know it occurred.”

    Then how do the proxy reconstructions show the upswings and downswings of the MWP and LIA that you people crow about. If the proxy record cannot show these things then perhaps the MWP and LIA did not occur and/or where not as extreme as the patchy only anecdotal evidence for them suggests.

  44. #44 Susann
    August 16, 2010

    So what Mann and his cohorts in crime (just an expression) did was drop the disagreeable portion of the tree ring proxy record from 1960 onward and stitch in the satellite temperature record in its place.
    The long and the short of it is that the blade portion of the hockey stick could have occurred once or even many times in the past 1000 years and the tree ring proxy would not have shown it since it failed to show it in the single instance where we know it occurred.

    First, yeah I really believe the use of “cohorts in crime” is just an expression…

    Second, as to the issue of divergence, two main points: not all trees failed to track recent warming; some track the recent warming in the instrumental record just fine. As well, trees that have diverged and those that don’t track earlier temps together well. Taken together, these two pieces of evidence suggest that there is something unique about the recent ‘divergence’ observed in some trees; See Skeptical Science for a discussion of this.

    From SS:

    The divergence problem has been discussed in the peer reviewed literature since the mid 1990s when it was noticed that Alaskan trees were showing a weakened temperature signal in recent decades (Jacoby 1995). This work was broadened in 1998 using a network of over 300 tree-ring records across high northern latitudes (Briffa 1998). From 1880 to 1960, there is a high correlation between the instrumental record and tree growth. Over this period, tree-rings are an accurate proxy for climate. However, the correlation drops sharply after 1960. At high latitudes, there has been a major, wide-scale change in tree-growth over the past few decades.

    Note that according to SS, trees in the south did reflect recent warming. It was trees in the northern / high latitudes that showed divergence. Cook (2004) showed that trees in the northern and southern hemispheres track each other and temps well back to the medieval period, suggesting that the cause is something unique to the modern post-1960 era. Warming induced drought, decline in ozone layer/increased UV light at surface, microsite issues. In other words, anthropogenic causes or regional causes.

    Divergence was known to paleoclimate types and dendros, so it wasn’t a secret.

  45. #45 Vorlath
    August 17, 2010

    @42:

    Then how do the proxy reconstructions show the upswings and downswings of the MWP and LIA that you people crow about.

    They don’t. The proxy reconstruction was done to show what it’s supposed to actually look like using correct methods. However, that’s not enough. They also tested both the graph as well as random data to see how well they correlate to instrumental data. The random data outperformed the reconstruction.

    So the reconstruction is useless. It doesn’t show anything other than the hockey stick is bogus (aka worse than random data). All assertions about temperatures are done WITHIN the model. So those assertions are also bogus since they are detached from physical reality. They made all of this quite clear in the paper. They repeat it over and over.

  46. #46 John Mashey
    August 17, 2010

    re: Paddy: #33:
    “‘Correct me If im wrong but the MBH hockey stick had a horizontal (no slope) handle”

    Yes, you are wrong … as is clearly seen on p.135 of the IPCC TAR WG I, Figure 2.20, which shows a NH linear trend line (red dashes) from ~-0.2C to ~-0.35C from 1000AD to ~1850AD.
    [If anything, many other reconstructions have even less MWP, albeit even lower LIA's ... which of course would make the distinct bend in the hockey stick even stronger.]

    Of all the reconstruction charts I’ve seen, *exactly one* has a horizontal handle … and that one is the one on the cover of my copy of The Hockey Stick Illusion…
    where Montford draws a horizontal gray bar. Have I missed one somewhere else is that where you got it? And do you believe that book?

  47. #47 MFS
    August 17, 2010

    Paddy, Paddy, Paddy…

    Coming from someone who thinks [Tim Curtin is the 'bees knees'...](http://scienceblogs.com/deltoid/2010/04/tim_curtin_thread_now_a_live_s.php#comment-2710892) why does this not surprise me.

    [Your quote:](http://scienceblogs.com/deltoid/2010/08/a_new_hockey_stick_mcshane_and.php#comment-2730941)
    >From the bits of research i have under-taken I do believe that there was in reality no instrumental coverage of more than half of the globe until the 1950s, with the non-covered areas almost wholly in the tropics. Therefore this to me could be used to create a false cold “global” anomaly in 1850-1950 relative to the warmer anomaly base period of 1960-1990 which does include the tropics?

    Shows, above all, that you have no idea how global temperatures are calculated, and about anomalies. For starters global anomalies are calculated from local anomalies, not the global average temperature. For seconds, the effect of global warming is more extreme in the temperate and subpolar latitudes than the tropis, so you shoot yourself in the foot there. A dataset compased predominantly of temperate latitude measurements would show more warming than a tropic-slanted one, not less.

  48. #48 Dan
    August 17, 2010

    @Dave Springer:

    I have a question for Dan. You picked up on the contents of the paper nicely but why did it “worry” you? I should think you’d be worried that Mann is right and the earth is warming at an unprecedented rate. This paper reveals that Mann showed no such thing and that the earth could have warmed at this rate many times in the past. Do you -want- a climate catastrophe and you’re worried that you might not get what you want?

    Fair question, if somewhat tendentiously put. I don’t want a climate catastrophe. I’ve never met anyone who does. The “worry” is a certainty-of-belief thing: I worry about the possibility that I believe the wrong thing. Don’t you?

    In my own work I’m always worried that I’m missing something. Most scientists will confess to similar anxieties if you get enough beers into them: worrying about data analysis is an occupational hazard. Obviously, it’s not my research being discussed here, but it’s hard not to feel “worry-by-proxy”, if you’ll forgive the expression. Of course, I don’t know if McShane & Wyner are correct in their scientific conclusions, even though I can follow the statistics in considerable detail. But even the possibility that they might be right on the substantive questions is nerve-wracking from a pure science perspective. Because if they are, it opens up the possibility that scientists are providing inaccurate policy advice. This is something that *no* scientist wants, and we all go to great lengths to avoid regardless of which scientific field we’re in.

    Hence, “worrying”: this is important, and no honest person wants to be wrong on an important topic. But I think you already knew that’s what I meant, didn’t you?

  49. #49 Steve Reuland
    August 17, 2010

    But even the possibility that they might be right on the substantive questions is nerve-wracking from a pure science perspective. Because if they are, it opens up the possibility that scientists are providing inaccurate policy advice.

    For what it’s worth, even if it turned out that we couldn’t say anything one way or another about paleoclimate over the last 1000 years (which would be a pretty extreme interpretation of their results, as I understand them), this would not have any significant effect on policy advice. It would not, for example, mean that greenhouse gases haven’t caused the recent run-up in temperatures, or that they will continue to do so in the future. Paleoclimate isn’t really relevant to that question at all. It only seems that way because some people obsess over it. If they had a direct case against the greenhouse effect, they’d go with that, but since they don’t, paleoclimate (and Mann specifically) have become a, um, proxy for the whole of climate science.

  50. #50 John Mashey
    August 17, 2010

    re: #47
    1) I have reason to believe that your statistical questions well-answered soon by various competent folks.

    2) But, to pick a different example. 2 statisticians, with zero experience in the medical field, publish a paper that shows that medical science really has no basis for thinking there so a link between smoking and disease. In doing so, they reference 3 Wall Street Journal OpEds, a BBC piece, and multiple decades-old references, all of which I pointed out in #11. They also heavily reference a report infamous for its poor quality [Deep Climate has shown 10 pages of plagiarism already (there is more), I mentioned the reference to a black-helicopters conspiracy-theorist, and there is much, much badly wrong.]

    So, is your response to say:

    OK, these guys may be on to something, my kids can smoke.

    OR

    Pretty unlikely, I’ll wait a bit and see what experts think.
    Lots of papers are just wrong.

    I RARELY OFFER PREDICTIONS, BUT HERE’S ONE: when the dust settles on this, within a month or two, the authors, Wharton, and the journal will be *very* embarrassed by this.

  51. #51 himThere
    August 17, 2010

    No doubt some will be touting this paper as the final nail in the AGW coffin.

    It seems to me that this particular casket must be so full of ironmongery that there will be no room left for a corpse.

  52. #52 Stephen Gloor (Ender)
    August 17, 2010

    Vorlath – “They don’t. The proxy reconstruction was done to show what it’s supposed to actually look like using correct methods.”

    So you are now saying that the MWP and LIA did not happen? Remember MBH99 was first criticised by M&M for not showing these two events and now here you are saying that a reconstruction using ‘correct’ methods does not show them and cannot show them – hmmmmm interesting.

    As to the other parts of your post I really do not think you understand the implications of this paper other than the rantings on CA and WUWT. Perhaps you should read some more on it before deciding this.

    BTW what about the other proxy reconstructions? The ones that use neither tree rings or PCA analysis that show essentially the same thing – are they also bogus?

  53. #53 wag
    August 17, 2010

    The key to the hockey stick debate is that it doesn’t matter. We care about the future increase in temperature, not the 1 degree “blade” on any existing hockey stick that we’ve already experienced. I’ve drawn a rough new graph showing what a hockey stick would look like with a modest 3.75 degree C increase over the 21st century (it’s pretty terrifying):

    http://akwag.blogspot.com/2010/08/enough-ado-about-hockey-sticks-worst-is.html

  54. #54 Marco
    August 17, 2010

    I’ve just waded through the climateaudit thread on this (loads of “vindication!” type of messages).

    But most interestingly, Abraham Wyner made a short comment, indicating there would be comments and discussion added. Curious to see who gets to comment/discuss.

  55. #55 David Horton
    August 17, 2010

    I’m always puzzled by this stuff. I take it the authors, and some of the deniatariat responding here, believe that it is just a sheer coincidence that the “blade” corresponds to the massive upswing in CO2 emissions. A very useful coincidence because you can just give the green light to more and more emissions, nothing to worry about, any moment now the blade will just swing around and resume the downward slope, and in another 500 years or so we won’t even see a blip corresponding to the last three decades. Is that what these people believe?

  56. #56 MFS
    August 17, 2010

    [John Mashey](http://scienceblogs.com/deltoid/2010/08/a_new_hockey_stick_mcshane_and.php#comment-2731122),

    May I take the liberty of extending your quote
    >”Pretty unlikely, I’ll wait a bit and see what experts think. Lots of papers are just wrong.”

    To:
    >”Pretty unlikely, I’ll wait a bit and see what experts think. Lots of papers are just wrong, and this one hasn’t even gone through peer-review and been accepted for publication yet.”

  57. #57 harryrsnape
    August 17, 2010

    Stephen Gloor wrote:
    “Then how do the proxy reconstructions show the upswings and downswings of the MWP and LIA that you people crow about. If the proxy record cannot show these things then perhaps the MWP and LIA did not occur and/or where not as extreme as the patchy only anecdotal evidence for them suggests.”

    That’s what one expects. They are saying proxies don’t do a good job of tracking fast steep swings. So as Dave Springer said, similar fast warming events as recorded at airports in the last 60 years could have occurred many times.

    All you need to do is show that the MIA and LIA only lasted 50-60 years and you’d have a point. Comprehension is a skill best practiced without religious zeal.

  58. #58 pete
    August 17, 2010

    and this one hasn’t even gone through peer-review and been accepted for publication yet

    Is this true? Annals of Applied Statistics has the paper on its “Next Issues” page.

    On the other hand, it looks very much like a draft, it’s longer than the journal 20-page policy, and it’s full of errors.

    Maybe it’s not too late for AAS to avoid embarrassment?

  59. #59 Hank Roberts
    August 17, 2010

    If this hasn’t yet been reviewed, why is it included in the articles linked under “next issues” at the journal?

    http://www.e-publications.org/ims/submission/index.php/AOAS/user/submissionFile/6695?confirm=63ebfddf

    If it has been reviewed, why does the title at the magazine refer to “Surgace” temperatures? and why is the error range amost all on the low side for the year 2000? zoom in on their black line.

  60. #60 cohenite
    August 17, 2010

    Dan at 29 says: ” I can’t imagine that climate scientists go around collecting arbitrarily chosen data and trying to correlate any- and every- thing to the temperature record.”

    Makes the thread worthwhile; now Dan, google: YAD061

  61. #61 Wow
    August 17, 2010

    Of course, McI, Right S Fred, Monkey and Watts *never* do this.

    Anyhoo, it seems to be the new denialist meme: The Science of Global Warming is settled. The next time you hear this, reply “YAD061″.

    Which is bollocks:

    1) “The science is settled” is a strawman that denialists make up. Define “The science” and then you can say whether it’s “settled” or not. The atomic theory of matter is pretty settled, don’t you think?

    2) It’s odd that in a complaint about cherry picking, this one tree is cherry picked by denialists as THE ONE AND ONLY TREE Mann used. Oddly enough, the other proxies (many other trees included) don’t show a global MWP either. In fact, how can ONE TREE show a global trend? All it CAN show is that MWP WAS NOT global, since it didn’t happen where the tree was.

    So I think cohen’s post is actually more of what makes this thread worthwhile: a close insight into the denialist meme.

  62. #62 Dave Springer
    August 17, 2010

    I’m always puzzled by this stuff. I take it the authors, and some of the deniatariat responding here, believe that it is just a sheer coincidence that the “blade” corresponds to the massive upswing in CO2 emissions. A very useful coincidence because you can just give the green light to more and more emissions, nothing to worry about, any moment now the blade will just swing around and resume the downward slope, and in another 500 years or so we won’t even see a blip corresponding to the last three decades. Is that what these people believe?

    Posted by: David Horton | August 17, 2010 2:50 AM

    Can’t speak for other members of the deniatariat but I can’t find anything theoretically or empirically inconsistent with the IPCC position that each doubling of atmospheric CO2 causes a 1.1C increase in surface temperatures absent feedbacks. Where I take issue is this silliness about a positive feedback that amplifies that 1 degree change into a 3 degree change. There’s not a shred of physical evidence to support it. CO2 concentration in atmosphere indicated by proxy in the geologic column going back many millions of years has CO2 levels up to 20 times what they are today and temperature at the same time was only 7-8 degrees higher. This is consistent with a 1 degree rise per doubling and no feedback. If there were positive feedbacks we’d get a runaway greenhouse and that simply has never happened. If it had we wouldn’t be here to talk about it.

    So then, acknowledging that a CO2 doubling causes a temp rise of 1 degree I next ask if there’s enough fossil fuel on the planet to more than quadruple current levels. No way. We might be able to quadruple but it isn’t sustainable.

    So we’re looking at a 2.2C temporary rise in temperature over then next century or two. Next I have to look back at the geologic column again and have a look-see at what the earth was like under those conditions. Without exception when the earth is warmer and richer in CO2 the biosphere is more productive. So I’m supposed to prefer more rocks and ice to more plants and animals? I don’t think so. I’m rather looking forward to a more productive biosphere. Who wouldn’t?

  63. #63 John Mason
    August 17, 2010

    Copy of my post over at Tamino’s blog:

    I did try posting on WUWT last night, not something I do often, suggesting that people waited until the paper was published and other specialists in the relevant field had formally responded before arriving at a considered opinion – as per standard academic procedure.

    It didn’t make me that popular, although some of the membership did seem to broadly agree.

    I feel that this practice of circulating a draft MS around the blogosphere prior to publication – for any reason – is an unwelcome development in this or any other scientific discipline. It encourages the exact opposite to objectivity.

    Cheers – John

  64. #64 Wow
    August 17, 2010

    “Where I take issue is this silliness about a positive feedback that amplifies that 1 degree change into a 3 degree change. There’s not a shred of physical evidence to support it.”

    Yes there is. The current temperature of the Earth.

    33C warmer than it would be without ANY GHGs. CO2 is either more than 33% of that to have a less than 3:1 feedback effect or CO2 is a much bigger forcer of global temperatures than expected (or any denialist has EVER stated is ever feasible: it’s always “H2O is a MUCH bigger forcing!!!”).

    Since it’s the same calculation that gives 1.1C per doubling that says ~10C of that 33C warming is from CO2, to take a stance that feedbacks don’t make it 3:1 warming would be internally inconsistent.

    This is absent all the papers on physical measurement constraints on the sensitivity done by people like Annan et al.

  65. #65 AS
    August 17, 2010

    cohenite’s stock in trade are red-herrings .. so no need to address [this](http://scienceblogs.com/deltoid/2010/08/a_new_hockey_stick_mcshane_and.php#comment-2729979) error. To paraphrase your lovely quote – just run to mummy McIntyre.

    [Here](http://img515.imageshack.us/i/oyad06.pdf/) is the chronology without YAD061

    Or read a full report [here](http://scienceblogs.com/deltoid/2009/10/mcintyre_misunderstood_somehow.php)

  66. #66 Dave Springer
    August 17, 2010

    @Posted by: Dan | August 17, 2010 12:36 AM

    re; worry

    God, grant me the serenity
    To accept the things I cannot change;
    Courage to change the things I can;
    And wisdom to know the difference.

    I can’t change how the climate reacts to CO2 and I can’t change how much fossil fuel the world burns. I work very hard at lowering my own energy consumption but it’s only because I’m an engineer and constantly look for more efficient ways of getting things done rather than out of any altruistic desire. I’m under no delusion that anything I do will make any difference in the grand scheme of things.

    The point is in some instances I’m happy to be wrong. Say I’m on a lonely highway at night and I believe that I don’t have enough gas to make it to the next gas station. I worry that I’m right not worried that I’m wrong. If you’re worried that we’re heading towards a climate catastrophe you should be glad to be wrong, shouldn’t you?

  67. #67 Lotharsson
    August 17, 2010

    > CO2 concentration in atmosphere indicated by proxy in the geologic column going back many millions of years has CO2 levels up to 20 times what they are today and temperature at the same time was only 7-8 degrees higher. This is consistent with a 1 degree rise per doubling and no feedback.

    Not so. 20 times is barely more than 4 doublings which is in the 4.5-5 degrees range. And implicit in your claim is the assumption that the feedback effect is constant all the way up to 20 times current levels, which I don’t think many climate scientists would agree with. Many positive feedbacks (e.g. Arctic ice albedo effects) operate over a limited range and then cease – i.e. once the ice has all melted.

    > If there were positive feedbacks we’d get a runaway greenhouse…

    Not so. This is a common fallacy based on a lack of understanding about positive feedbacks which leads many people into drawing seriously flawed conclusions.

    Imagine you have a system at equilibrium, and then you change something which changes the key variable you are measuring (e.g. you add some greenhouse gases which – sans feedback – changes the temperature). Call the magnitude of this change “delta”.

    Now imagine you have a feedback gain of “g”. What happens?

    The initial temperature change is +delta, and the feedback adds delta*g more.

    But that extra temperature change due to the feedback is itself subject to feedback, so you get another (delta*g)*g change.

    And that is subject to feedback, so you get (delta*g*g)*g change.

    And so on, which leads to a total change at equilibrium which is the sum of a geometric series delta * SumOf(g^k) for all k between 0 and infinity. If you remember high school maths, this sum is finite for g in the range (-1..1) (non-inclusive) – and then the sum equates to delta / (1-g).

    So you **only** get a runaway situation if g is >= 1 (or <= -1 which isn’t relevant here). And given the result delta/(1-g) you can get **any** positive amplification you like at equilibrium with a gain less than one. For example when g = 2/3, delta / (1-g) = 3*delta.

  68. #68 Lotharsson
    August 17, 2010

    Argh, forgot that markdown would mess with my asterisks. Try “x” for multiplication instead of “*” in the middle section:

    Now imagine you have a feedback gain of “g”. What happens?

    The initial temperature change is +delta, and the feedback adds delta x g more.

    But that extra temperature change due to the feedback is itself subject to feedback, so you get another (delta x g) x g change.

    And that is subject to feedback, so you get (delta x g x g) x g change.

  69. #69 Dave Springer
    August 17, 2010

    33C warmer than it would be without ANY GHGs. CO2 is either more than 33% of that to have a less than 3:1 feedback effect or CO2 is a much bigger forcer of global temperatures than expected (or any denialist has EVER stated is ever feasible: it’s always “H2O is a MUCH bigger forcing!!!”).

    Since it’s the same calculation that gives 1.1C per doubling that says ~10C of that 33C warming is from CO2, to take a stance that feedbacks don’t make it 3:1 warming would be internally inconsistent.

    This is absent all the papers on physical measurement constraints on the sensitivity done by people like Annan et al.

    Posted by: Wow | August 17, 2010 4:12 AM

    Wow. Just wow. There has been no runaway greenhouse in the earth’s history. CO2 level has been up to 20 times higher in the past yet temperatures were never more than 7-8 degrees higher.

    If you won’t let facts get in the way of your beliefs then I really don’t have anything more to say to you because facts are all I have to offer.

  70. #70 cohenite
    August 17, 2010

    Red herrings AS? Unlike your plump gooses. I prefer this to reflect the influence of “the most important tree in the world”;

    http://network.nationalpost.com/np/blogs/fpcomment/00-YAMAL.eps.jpg

    I note you haven’t responded to the relative contributions of H2O and CO2 to the greenhouse temperature: this will refresh your memory:

    http://scienceofdoom.files.wordpress.com/2010/02/ramanathan-coakley-1978-role-of-co2.png

  71. #71 Wow
    August 17, 2010

    “Wow. Just wow. There has been no runaway greenhouse in the earth’s history.”

    Aaargh! A blinding flash of the Obvious!

    Never said there was. Where do you think it exists in that quote?

    “CO2 level has been up to 20 times higher in the past yet temperatures were never more than 7-8 degrees higher.”

    And the Sun was 10% cooler.

    Duh.

    “If you won’t let facts get in the way of your beliefs”

    If you knew the facts instead of your beliefs…

  72. #72 Wow
    August 17, 2010

    “The point is in some instances I’m happy to be wrong.”

    And what if you’re wrong about AGW not being a problem?

    Catastrophe. Because there’s a ~50 year lead time on what you do now to the end result so you’re 50 years too late to avoid the crash.

    Oddly enough, you don’t seem to be worried about being wrong there…

  73. #73 Jeff Harvey
    August 17, 2010

    *Without exception when the earth is warmer and richer in CO2 the biosphere is more productive*

    Totally and utterly incorrect, and this kind of statement reflects a poor understanding of complex adaptive systems. Dave Springer has in one single remark excluded trillions of important biotic interactions which play an important role in determining system productivity and stability, as well as relative contributions made by other elements in regulating biogeochemical cycles. The fact is, and I have said this a million times before but hopefully this time with the likes of Dave Springer it will resonate:

    Our planet evolved more biological diversity at any time in it’s history when atmospheric levels of C02 were comparatively low (e.g. about 8,000 years ago). Repeat that over and over again until it sinks in. Diversity and productivity are driven by a wealth of factors that include temperature, much less so concentrations of C02 in the atmosphere. Yet this old myth seems to resonate with the denialati, who appear to use it to draw solace on the fact that humans are driving much of the planet’s systems to hell and beyond. Rapid climate warming and puimping more and more C02 into the atmsophere are not recipes for creating a green utopia in which biodiversity thrives, in fact quite the contrary. The key is the rate of change, and not the absolute values. Natural systems and in particular the species and populations that make them up are not, for the most part genetically equipped to respond to a suite of anthropogenic assaults, including the great experiment our species is conducting on the atmosphere and climate. And certainly not temporally, given the fact that these changes are occurring in the blink of an evolutionary eye. If we were talking about several thousands of years, and we were to remove other anthropogenic threats such as overharvesting, habitat fragmentation and destruction, other forms of pollution etc. from the picture, then I would be a tad more hopeful, but in combination and given the short time scales involved, its folly to talk about being optimistic on the basis of a single study as well as on erroneous ideas of correlation and causation between C02 and systemic health. A 2.2. degree rise in the space of less than 100 years will be a disaster for many of the world’s ecosystems. There. It is now said.

    And, for the record, I am not even bothering to discuss the effects of C02 on primary and secondary plant metabolites and their broader ecological consequences. This has been comprehensively dealt with on the TC thread. I exhausted my breath countering Tim Curtin’s simplistic C02-related nonsense on that thread, and then Dave Springer pops up here saying effectively the same thing. Sorry if I appear frustrated at this kind of grade-school connect-the-dots level of understanding, but it is always shocking seeing how many people peddle the same story.

  74. #74 Dave Springer
    August 17, 2010

    @Posted by: Lotharsson Author Profile Page | August 17, 2010 4:37 AM

    Let’s take the temp record and CO2 record from 1880 to 2000 at face value (I do). 1880 is when CO2 level began rising.

    We can see two obvious complete cycles of 60 years each which happens to correspond with SST multi-decadal oscillations with roughly 30 years of warming followed by 30 years of no warming or slight cooling.

    1880-1910 cooling
    1910-1940 warming
    1940-1970 cooling
    1970-2000 warming

    The warming exceeded the cooling by 0.4C in each iteration.

    1880-1940 CO2 increased by 25ppm
    1940-2000 CO2 increased by 50ppm

    This is perfectly consistent with the way a GHG gas, or any insulator, works. Additional insulation has diminishing effectiveness. Compare to adding a first blanket over yourself on a cold night to adding a 100th blanket. The 100th blanket will have very little effect compared to the first one.

    This is also perfectly consistent with a 1.1C rise in temperature with every CO2 doubling and no feedback. Where’s the feedback? It ain’t in the actual data that’s for sure.

    Right on schedule, the warming stopped in 2000. Even Phil Jones has admitted there has been no statistically significant warming for the past 10 years. Trenberth in the Climategate emails called it (privately of course because there can be no public questioning of the dogma) a travesty that they couldn’t explain the lack of warming.

    It’s pretty easy to explain, actually. All they have to do is admit that the positive feedback scenario is bullshit and everything then falls neatly into place. A 1.1C calculated and actual surface temp increase per CO2 doubling with a multi-decadal SST oscillation riding on top of it.

  75. #75 Lotharsson
    August 17, 2010

    > CO2 level has been up to 20 times higher in the past yet temperatures were never more than 7-8 degrees higher.

    This cannot lead to the conclusion you draw if other non-trivial temperature influences are at different levels now compared to then, or more generally if the sum of all such temperature influences are.

    Are you *really* making either of those claims, and if so on what basis?

  76. #76 Lotharsson
    August 17, 2010

    > Where’s the feedback?

    Total feedback is defined at equilibrium, which we are nowhere near at.

    > It ain’t in the actual data that’s for sure.

    As I implied in my previous statement, you can ONLY make that argument if you account for the variation of all other temperature influences over the time period in question. You are not doing so.

    > Even Phil Jones has admitted there has been no statistically significant warming for the past 10 years.

    Oh, please. That’s a complete denialist red herring, and if you’re seriously arguing that is relevant you’re not interested in science. (Phil Jones also noted that there *was* statistically significant warming over about 15 years – so if you quote him as an authority, or use statistically significant warming *over climatic time-scales* as a point of evidence, then in both cases your illustrative “point” is refuted.)

    > Trenberth in the Climategate emails called it (privately of course because there can be no public questioning of the dogma) a travesty that they couldn’t explain the lack of warming.

    Right on cue – another denialist red herring. Try finding out what Trenberth was talking about rather than what you think he was saying. Hint: he was NOT saying that it has stopped warming (if only because that’s an easily and robustly refuted claim).

    > It’s pretty easy to explain, actually. All they have to do is admit that the positive feedback scenario is bullshit and everything then falls neatly into place.

    Except that as pointed out previously, then via fairly straightforward physics the earth would be 20 degrees C colder than it is – which is precisely the opposite of “pretty easy to explain, actually”.

    So far you’ve based your arguments almost entirely on fairly simple fallacies…

  77. #77 JamesA
    August 17, 2010

    > Where I take issue is this silliness about a positive feedback that amplifies that 1 degree change into a 3 degree change. There’s not a shred of physical evidence to support it.

    Try reading the IPCC report sometime. WG1 has an entire subsection devoted to exactly that (9.6: Observational Constraints on Climate Sensitivity). And then there’s some inconvenient but well-established physical principles like the thermodynamic properties of water and stuff.

    And please, the whole ’20 times more’ meme is getting old; CO2 concentrations weren’t the only thing that was different back then, you know. You are aware that the ending of that period illustrates the role of ice albedo as a climate feedback, right?

    > If there were positive feedbacks we’d get a runaway greenhouse and that simply has never happened. If it had we wouldn’t be here to talk about it.

    Errrr…. no. Positive feedbacks in climate amplify a forcing such that the equilibrium shifts, but they don’t necessarily create a runaway situation.

    > I’m rather looking forward to a more productive biosphere. Who wouldn’t?

    Oh, I don’t know. All the people living in areas that are going to be affected by desertification or sea level rises maybe? Just guessing here.

  78. #78 Chris O'Neill
    August 17, 2010

    Dave Springer:

    Where I take issue is this silliness about a positive feedback that amplifies that 1 degree change into a 3 degree change. There’s not a shred of physical evidence to support it.

    Garbage. How is it that there has been at least 0.7 deg C of warming with half a doubling of CO2 which you think can only produce 0.5 deg C of warming? And this is ignoring the heat that’s going into the oceans and the effect of increase in aerosols.

    No physical evidence? Pull the other one.

  79. #79 Wow
    August 17, 2010

    “God, grant me the serenity To accept the things I cannot change; Courage to change the things I can; And wisdom to know the difference.”

    I’m afraid he’s failed you.

    Or you failed him.

  80. #81 Dave Springer
    August 17, 2010

    A 2.2. degree rise in the space of less than 100 years will be a disaster for many of the world’s ecosystems. There. It is now said.

    And it’ll be a be a boon for many more of them. Plants are the primary producers in the food chain. Wherego they go the rest of the food chain including us. Plants don’t grow well grow well in ice and snow. Even more to the point the warming we’re seeing is predominantly in the higher latitudes of the northern hemisphere right where all the ice and snow is concentrated. If we could order some beneficial warming that’s right where we’d ask it to be delivered. The ecological catastrophe is nothing but shrill handwaving with no empirical support whatsoever.

  81. #82 Wow
    August 17, 2010

    “Even Phil Jones has admitted there has been no statistically significant warming for the past 10 years”

    Our survey says:

    Nr Nrrrr.

    Phil Jones has admitted that the 0.12C per decade warming seen in the past 15 years is not yet statistically significant. A year ago. Before 2010 and 2009 which were both really hot years.

    In other news: Dave Springer has never denied beating his wife and says he loves it.

  82. #83 Wow
    August 17, 2010

    “And it’ll be a be a boon for many more of them.”

    A boon for weeds. For most food plants for humans, the useful edible content is reduced in a higher CO2. Additionally the toxixcity of the natural insecticides produced by corn leaves to deter beetles is reduced in high CO2 concentrations, making their pest species pick them over much more readily.

    Dave, if you won’t let facts get in the way of your beliefs, you won’t learn anything.

    PS anyone want to check Nasif’s IP address against this one?

  83. #84 Chris O'Neill
    August 17, 2010

    Dave Springer:

    CO2 levels up to 20 times what they are today and temperature at the same time was only 7-8 degrees higher. This is consistent with a 1 degree rise per doubling and no feedback.

    Even with the above figures and assuming nothing else has changed, this gives a sensitivity of 7-8 degrees/(4.3 doublings of CO2) = 1.6-1.9 degrees/doubling.

    Even your understanding of arithmetic is wrong Dave.

  84. #85 Dave Springer
    August 17, 2010

    82

    “Even Phil Jones has admitted there has been no statistically significant warming for the past 10 years”

    Our survey says:

    Nr Nrrrr.

    Phil Jones has admitted that the 0.12C per decade warming seen in the past 15 years is not yet statistically significant. A year ago. Before 2010 and 2009 which were both really hot years.

    In other news: Dave Springer has never denied beating his wife and says he loves it.

    Posted by: Wow | August 17, 2010 5:25 AM

    Jones was interviewed early this year.

    Transcript: http://news.bbc.co.uk/2/hi/science/nature/8511670.stm

    Excerpt:

    C – Do you agree that from January 2002 to the present there has been statistically significant global cooling?

    No. This period is even shorter than 1995-2009. The trend this time is negative (-0.12C per decade), but this trend is not statistically significant.

    This is perfectly consistent with what I wrote about a 60 year cycle where the latest 30-year warming episode ended right on schedule in 2000 and that there would be no significant warming for the next 20 years followed by 30 years of warming commensurate with whatever fraction of CO2 doubling actually occurred from 2000-2060. So far that is spot on although I’m not going to argue that proves I’m right. But it certainly looks good so far.

  85. #86 SteveC
    August 17, 2010

    Dave Springer @81:

    A 2.2. degree rise in the space of less than 100 years will be a disaster for many of the world’s ecosystems.

    And it’ll be a be a boon for many more of them.

    Evidence please.

    Plants are the primary producers in the food chain. Wherego they go the rest of the food chain including us

    So all 9bn of us will be off to the balmy shores of Norway huh? Along with all the other plants, the primary consumers, the tertiary consumers, the water catchments, the fertile soils, the soil fauna, the…

    Plants don’t grow well grow well in ice and snow.

    They’re not that flash in deserts either.

    Even more to the point the warming we’re seeing is predominantly in the higher latitudes of the northern hemisphere right where all the ice and snow is concentrated.

    Ah yes, of course. The entire world only consists of the northern hemisphere. All that stuff about record droughts, record periods of record high temperatures, bushfires yada yada yada in the southern hemisphere is just media hype, right?

    The ecological catastrophe is nothing but shrill handwaving with no empirical support whatsoever

    ROFL :-) I laughed so hard I nearly got my wallet out…

  86. #87 Dave R
    August 17, 2010

    Dave Springer
    >ended right on schedule in 2000

    [No it didn't](http://www.metoffice.gov.uk/corporate/pressoffice/2009/pr20091208b.html).

  87. #88 Deech56
    August 17, 2010

    Good catch, Marco @54. Here’s the quote from Wyner: “The paper has been accepted, but publication is still a bit into the future as it is likely to be accompanied by invited discussants and comment. Stay tuned…” The journal may be trying to do the right thing. It will be interesting to see who the discussants are.

  88. #89 adelady
    August 17, 2010

    Dave,
    The plants absorb more carbon so everything’s OK is very appealing – except for the poor old oceans. Unfortunately drought (and flood) make the notion of a burgeoning green saviour of the planet more and more unlikely.

    According to this article on the mathematician Inez Fung –
    “One major finding: droughts have already diminished the carbon absorbing capacity of the land and will continue to do so. Previous greenhouse experiments suggested that elevated CO2 levels caused plants to grow bigger and faster, an effect known as CO2 fertilization. The implication was that land-based carbon sinks — that is, plants — might be able to keep pace with higher CO2 levels. Fung’s modeling shows that on a global scale, regional droughts are likely to curtail this effect.”
    and
    “Her model also projects that the tropics are likely to become hotter and drier in summer months. As that happens, plants will absorb less carbon dioxide as a way to avoid water loss. In fact, atmospheric measurements over the past decade have already confirmed this effect.”

    There’s more of this discouraging stuff in the rest of the article. http://www.grist.org/article/2010-07-30-carbon-hunter/

  89. #90 Jeff Harvey
    August 17, 2010

    Dave Springer,

    As SteveC said. Please tell me, in detail, how complex adaptive systems will benefit by warming that occurs at rates faster than in many thousands of years, against a wealth of other anthropogenic stresses? Do you honestly believe that we could crank up the global thermostat by 2.2 C in no time and that all and that natural systems would suddenly blossom as they have not done in many millenia? What about local adaptation? Genetic constraints? Interaction network webs and phenological processes? Your posts show me, speaking as a population ecologist, that you have no basic knowledge about the factors that influence the rules governing the assembly and functioning of communities, ecosystems and biomes. Nor do you appear to know anything about the importance of scale. Are you a scientist, and have you perused any of the relevant literature on these topics? My guess is that the answers to these questions are a big whopping NO. Am I Correct?

    Moreover, as you said, temperatures will rise (and already are) much faster at higher latitudes. How will plants and consumers adapted to low temperature regimes respond to a sudden, dramatic local warming? How will soil biota respond under the same scenario? Will boreal acid soils suddenly become more alkaline and support deciduous vegetation? Soil chemistry and biology plys a significant role in the functioning of ecosystems through positive and negative feedbacks with above ground vegetation, herbivores and natural enemies in complex food webs. Pushing up the temperature rapidly in combination with C02 regimes, as well as various other human-induced effects will result (correction: is also resulting) in the unraveling of food webs and critical ecological interactions. This will in turn reduce the capacity of these systems to support themselves and, ultimately, us.

    Youer problem, Dave (another Dunning-Kruger disciple) is that you think natural systems function in a simple, linear way. That we tweak (or force) one parameter – in this case temperature – and everything else follows along nicely. I hate to spoil your party, but things wiull not work out this way. Humans have already simplified nature quite dramtically, reducing systemic resilience and hence stability, and also by reducing genetic variability in populations, a pre-requisite for adaptation to rapid change. I wish that you were correct, but you are not. You are wrong, and you are camouflaging what little you know about nature with frankly vacuous arguments (e.g. your ‘hand-waving’ argument).

    I suppose I spoke over your head with respect to my stoichimetry argument and when I discussed the effects of warming and increased atmospheric C02 on primary and secondary plant metabolites. But these are vitally imporant, in that they will cascade up through food webs and then outwards over communities. Ignoring these critical ecophysiological processes as you did does not make them go away.

    Until I get some form of substantial argument from you, and not a kind of tooth-fairy approach characteristic of the denial lobby (heck, most of them are not scientists, hence it is no small wonder that they spew out the simple stuff that they do), I might as well be speaking to a wall.

  90. #91 jakerman
    August 17, 2010

    Dave Springer writes:

    >*acknowledging that a CO2 doubling causes a temp rise of 1 degree I next*

    That would be argument by bogus assumption. Given large feedbacks are consistent with, and necessary to explain warming events such as the PETM and [inter-glacial cycles](http://en.wikipedia.org/wiki/File:Vostok_Petit_data.svg).

    >*I can’t change how much fossil fuel the world burns.*

    Really, then why are Fossil Industry groups spending so much money to lobby [against a price on carbon](http://www.desmogblog.com/climate-cover-up)?

  91. #92 Andrew Dodds
    August 17, 2010

    Pete:

    Here’s the problem.. if the paper ends up being rejected or heavily modified, then the denialosphere will merrily claim that they they are being censored/silenced. And if it isn’t corrected, it will be used as ‘Final Proof That the hockey stick is broken’ for ever more.

    A suspicious mind would think that there is a deliberate PR campaign going on here to publicize the paper to the max before any rewrite or retraction can happen. Still, I’m sure that McIntyre and Watts will open their inboxes to the public to demonstrate that no such shenanigans have been arranged.

  92. #93 Martin Vermeer
    August 17, 2010

    Oops.

    My explanation at link for why the M&W reconstruction is erroneous, was a little too simple. It’s the Wabett who gets it completely right: the fundamental error is calibration only against a hemispheric average, when local data — the 5×5 degree grid cells of instrumental data as used by Mann et al. — provide a so much richer source of variability — i.e., signal — to calibrate against.

    It is this poor signal/noise ratio that helps the calibration pick up spurious low vs. high latitude temp difference “signal”, which in the reconstruction interacts with the Earth axis tilt change effect.

    What stands is the observation that doing the calibration against the instrumental PC1 (instead of hemispheric average) will give you back pretty exactly the genuine Mann stick(TM) even in spite of this.

    Congrats Eli!

  93. #94 jakerman
    August 17, 2010

    >*How is it that there has been at least 0.7 deg C of warming with half a doubling of CO2 which you think can only produce 0.5 deg C of warming? And this is ignoring the heat that’s going into the oceans and the effect of increase in aerosols.*

    And it also ignores the delay required to reach radiative equilibrium.

  94. #95 jakerman
    August 17, 2010

    >*CO2 levels up to 20 times what they are today and temperature at the same time was only 7-8 degrees higher.*

    And Dave suddenly wants to pretend CO2 was the only thing different.

    Dave CO2 is not the only forcing in Earth’s history. Time for [Richard Alley](http://www.agu.org/meetings/fm09/lectures/lecture_videos/A23A.shtml).

  95. #96 SteveC
    August 17, 2010

    Oh Dave (81), that was the other thing:

    Even more to the point the warming we’re seeing is predominantly in the higher latitudes of the northern hemisphere right where all the ice and snow is concentrated. If we could order some beneficial warming that’s right where we’d ask it to be
    delivered

    See Dave there’s all this prolonged cold and what have you way way up there in the northern hemisphere Dave, and so the ground gets frozen quite a bit – it’s called permafrost. Well anyway, apparently some scientist bods found these things called methane clathrates in the permafrost, and… you still with me Dave?… well anyway it turns out that when a warming climate causes these areas of permafrost to thaw … Dave, you following me here?… the methane (which is a greenhouse gas, apparently) will be released, which could contribute to greatly accelerated climate ch…oh. Where did Dave go?

  96. #97 Dave Springer
    August 17, 2010

    84

    Dave Springer:

    CO2 levels up to 20 times what they are today and temperature at the same time was only 7-8 degrees higher. This is consistent with a 1 degree rise per doubling and no feedback.

    Even with the above figures and assuming nothing else has changed, this gives a sensitivity of 7-8 degrees/(4.3 doublings of CO2) = 1.6-1.9 degrees/doubling.

    Even your understanding of arithmetic is wrong Dave.

    Posted by: Chris O’Neill | August 17, 2010 5:41 AM

    I said it was consistent, not exact, with the distant past. The error bars for temp vs. CO2 millions of years ago are rather large. The only exacting numbers we have are very recent and must embody two complete multidecadal SST oscillations so we measure from trough to peak to eliminate the cyclic component.

    25ppm increase 1880-1940 caused 0.4C rise

    50ppm increase 1940-2000 caused 0.4C rise

    Projecting forward:

    100ppm increase 2000-2060 causes 0.4c rise

    100ppm increase 2060-2090 causes 0.2C rise

    That brings us to an approximate doubling (280ppm to 555ppm) for a grand total of 1.4C rise per doubling.

    IPCC estimates 1.1C per doubling. Actual looks like it will come in at 1.4. If there’s any feedback it appears to be about 27% at most (1.4/1.1=1.27). Contrast this with a feedback that drives the doubling to 3C increase (3.0/1.1=2.72) or 172% positive feedback.

    Reality = 27% positive feedback
    Imagination = 172% positive feedback

    Which is the greater error from the IPCC calculated 1.1C rise per doubling with no feedbacks at all?

  97. #98 jakerman
    August 17, 2010

    Dave Springer:

    >*We can see two obvious complete cycles of 60 years each which happens to correspond with SST multi-decadal oscillations with roughly 30 years of warming followed by 30 years of no warming or slight cooling.*

    The big problem that knocks this narrative off its perch, is that is requires us to ignore [SO2 and other aerosol forcing](http://scienceblogs.com/illconsidered/2006/03/what-about-mid-century-cooling.php). Dave how did you disappear these forcing factors?

  98. #99 Dave Springer
    August 17, 2010

    96

    Oh Dave (81), that was the other thing:

    Even more to the point the warming we’re seeing is predominantly in the higher latitudes of the northern hemisphere right where all the ice and snow is concentrated. If we could order some beneficial warming that’s right where we’d ask it to be delivered

    See Dave there’s all this prolonged cold and what have you way way up there in the northern hemisphere Dave, and so the ground gets frozen quite a bit – it’s called permafrost. Well anyway, apparently some scientist bods found these things called methane clathrates in the permafrost, and… you still with me Dave?… well anyway it turns out that when a warming climate causes these areas of permafrost to thaw … Dave, you following me here?… the methane (which is a greenhouse gas, apparently) will be released, which could contribute to greatly accelerated climate ch…oh. Where did Dave go?

    Dave is here reminding you that nothing in the past, including far more CO2 than we can ever pump into the atmosphere, has ever increased the temperature of the earth more than 7-8C from where it is now. There is no such thing as a runaway greenhouse caused by CO2, or methane, or anything else in the history of the earth.

    There have been runaway glaciations though. If there’s a tipping point its a tipping point to a snowball earth.

    The average temperature of the ocean is 4C. The only way it could be that low (it has 1000 times the heat capacity of the atmosphere) is that is the average temperature of the surface over a 120,000 year complete recent cycle of glaciation and interglacial period. Interglacials last about a tenth as long as glacial periods.

    Yeah boy, let’s see what we can do to help this interglacial come to a quicker ending and get back to that average global temperature near freezing. A glacier rolling over Washington D.C. isn’t such a bad idea. It might solve more problems than it creates.

  99. #100 MartinM
    August 17, 2010

    IPCC estimates 1.1C per doubling. Actual looks like it will come in at 1.4. If there’s any feedback it appears to be about 27% at most (1.4/1.1=1.27).

    We’re still not at equilibrium. Your figures include only those feedbacks which operate on short timescales.

1 2 3 6

Current ye@r *