Reflections on the Challenge

November 21st, 2005

Posted by: Roger Pielke, Jr.

A few weeks ago we posed a challenge to both parties involved in the so-called “hockey stick” debate to explain why the rest of us ought to care about the debate. We asked, “so what?” We received responses from Steve McIntyre and Ross McKitrick while everyone on the other side declined to participate, though a few showed up in the comments. Here I’d like to offer a few assorted reflections on the responses and the subsequent discussion.

1. First, thanks to Steve McIntyre (SM) and Ross McKitrick (RM) for providing thoughtful responses. The responses motivated a healthy discussion and for me provided some greater insight into the dynamics of the ongoing debate within the climate community not just over the hockey stick, but broader issues as well.

2. Interestingly enough, the response from SM is completely in agreement with RealClimate contributors Stefan Rahmsdorf (SR) and William Connelley (WC) that the “hockey stick” debate is pretty much irrelevant to the scientific question of whether or not greenhouse gases will affect the future climate. Consider:

SR: “The discussions about the past millennium are not discussions about whether humans are changing climate; neither do they affect our projections for the future.”

WC: “Why is this fight important to the rest of us? the answer is: you shouldn’t. It isn’t..”

SM: “I’m inclined to agree that, for the most part, the Hockey Stick does not matter to the great issue of the impact of 2xCO2.”

This agreement is interesting because it means we can move beyond the often invoked assertion that the hockey stick is the keystone supporting the entire scientific basis of climate science. Others may assert that the hockey stick is a scientific keystone, but apparently not the principals involved in this debate.

3. But the agreement among the parties raises a very interesting set of questions that have much more to do with climate science policy than climate policy. First among these questions is a very good point raised by both SM and RM, if the hockey stick doesn’t matter to the case for greenhouse gas effects on climate, why was it included and featured in the Summary for Policy Makers in the Second Assessment Report of the Intergovernmental Panel on Climate Change? SM provides a compelling answer to this question, “So even if the Hockey Stick did not “matter” to the scientific case, it mattered to the promotion of the scientific case.” The role of the SPM and the “promotion” of science by IPCC officials is a fair subject for discussion, independent of the answers to the technical questions M&M are debating with Mann et al.. Ultimately, the hockey stick debate is relevant to policy questions, but these questions have more to do with how we think about and organize science for policy, than any particular questions of climate policy itself.

4. RM suggests that the hockey stick is a symbol of the fidelity of journal peer review, the credibility and legitimacy of the IPCC, and how governments (in this case Canada) use scientific symbols to promote particular policies. It seems to me that MM sure could have made these points a lot more prominently early on. I am more convinced about the importance of the IPCC and journal peer review than the argument about the influence of the hockey stick on Canadian climate policies, but I am open to the case being made. I see that MM briefly raise the issue of journal peer review in their 2005 publication in Energy and Environment (PDF) and nothing comes up in their GRL publication (PDF). SM did raise some of these issues on his blog last February and in a related op-ed. RM raises some of these issues in this conference paper (PDF). But these are pretty hard-to-find nooks and crannies. My unsolicited advice to SM and RM is to spend more time (a lot more time) talking about the “so what?” questions as they pursue the obscure technical details. (By all means pursue the obscure and technical, but if indeed you care about the broader issues, then it is the broader issues that matter most.) They might find themselves with some allies if they talk more about peer review in science and international assessments, for which there are many people with interests and concerns. Pretty much all I do is “so what?” related to climate science, and I did not understand the positions of M&M until they wrote these essays. All of us are more likely a tool of those seeking to use our work for their purposes if we do not clearly and repeatedly (ad nausem) stake out your claim to the “so what?” ground. Perhaps, one reason that the folks on the other side chose not to participate may be their desire to leave the “so what?” ground open for occupation.

5. The concerns raised by SM and RM about the IPCC are part of a much broader set of experiences that raise questions about the credibility, legitimacy and salience of the IPCC. SM is perfectly justified in asking questions about the IPCC. We are all stakeholders in the IPCC process, and there appears to be no independent venue for raising issues about the IPCC process. The reception of the paleo climate community to M&M, regardless of the merits of their claims, is a good reason why such an independent venue makes sense. Certainly M&M could have been more tactful and diplomatically astute in their efforts, but still, the IPCC is the international organization responsible for bringing climate science and economics to policy makers, it can’t afford to be petty or aloof. I’d contrast the reception that Wentz and Mears received upon bursting on the scene with evidence that the Spencer/Christy satellite data was flawed. The dynamics here are easy to understand, and they can be found in all sorts of places, but it is the job of the IPCC to treat science and scientists fairly, not to protect a consensus or political symbols.

6. Finally a few comments about the discussion that followed. I continue to be amazed at the degree of tribal behavior that the climate community generates. Different camps give themselves and their opponents cute names — “hockey team,” “skeptics,” “contrarians, “mainstream”. They meet in club houses like RealClimate and ClimateAudit where they talk amongst themselves. A telling comment appeared early in the exchange when some one asked me if I was “embarrassed” to be providing a forum for RM. This tells me that some folks are less interested in resolving the climate debate than perpetuating it. I suppose the fight is good sport. But if progress is ever going to be made on the issue then people on different sides will have to meet, discuss and compromise. If I were a proprietor of RealClimate or ClimateAudit I would have some very real concerns about creating an “echo chamber“. Sometimes I wonder if these sites do less to educate their self-selected visitors than make their proprietors more strident and extreme in their own views, a la Cass Sunstein.

7. Finally, I found it amusing to find myself being attacked simultaneously on both the RealClimate website and the ClimateAudit website for being in the camp of the other. As one post said, “if you are not with us you are against us”. This perspective, which is held not only among anonymous blog commentators, but some scientists, issue advocates and politicians helps to explain why the climate debate is locked in stalemate, and everyone chooses to fight about science instead.

Thanks all for participating. If you have any suggestions for topics and contributors that we might invite in the future to engage one another, please send them along.

54 Responses to “Reflections on the Challenge”

  1. Gavin Says:

    Roger, a reasonable summary, but I think you leave an incorrect impression that somehow RealClimate exists merely to defend the HS from M&M and Climateaudit. By referring to Stefan and William as if they protaganists (they are not) as opposed to just being interested observers, you leave the impression that RealClimate is somehow synoymous with Mann, Bradley and Hughes. Mike and Ray B. are part of the RealClimate team and thus we have more than a passing interest in the questions raised by this issue. But neither Stefan, William or myself (or Eric, David, Rasmus, Thibault or Ray P.) have any direct involvement in this issue despite what you might read elsewhere. We created RealClimate to talk about all aspects of climate science. This topic has probably already taken up more than it’s fair share of bandwidth.

  2. 2
  3. Roger Pielke Jr. Says:

    Thanks Gavin for this clarification. But just so you know how things might look to an observer, on the RC WWW the top 2 scientific “Highlights” are about the hockey stick debate.

  4. 3
  5. Roger Pielke Jr. Says:

    Gavin- I’d also point you to Michael Mann’s response to us, which pointed us to RC and Stefan R.’s post specifically:

  6. 4
  7. William Says:

    Roger, you’re still getting it wrong. Your very first “interesting question” isn’t. You say:

    “But the agreement among the parties raises a very interesting set of questions that have much more to do with climate science policy than climate policy. First among these questions is a very good point raised by both SM and RM, if the hockey stick doesn’t matter to the case for greenhouse gas effects on climate, why was it included and featured in the Summary for Policy Makers in the Second Assessment Report of the Intergovernmental Panel on Climate Change?”

    You have your tenses wrong (and you’ve confused the SAR with the TAR). At the time of the TAR, the hockey stick was of some importance – it was one of the many threads that pointed to the attribution. Subsequent to the TAR, a pile of other reconstructions confirmed the basic findings of MBH, and the importance of the hockey stick is therefore much less. Hence, its correct to say that it *doesn’t* matter; it would have been incorrect to say that it *didn’t* matter. But because subsequent studies *have* reconfirmed the important points (ie, everything the TAR SPM said about MBH is true for all the others, as far as I can see), the TARs featuring of MBH is, in rettrocpect, fully justified.

    Now: since SM and RM, and indeed you, have got this so badly wrong, what is left?

  8. 5
  9. Roger Pielke Jr. Says:


    Thanks for your comments. I did write SAR when I meant TAR. So your position now is that the hockey stick was in 2001 a key study in making the case for attribution. That is, that without the hockey stick the case for attribution in 2001 would have been somewhat weaker? I disagree.

    My assertion is that the claims of attribution in 2001 were largely insensitive to the presence of the hockey stick. The SAR (yes, SAR) would seem to bear this out as it pre-dated the hockey stick and made a strong claim for attribution. The evidence for attribution is fairly broad as I understand it, and is not built upon the paleo studies.

    If now you are saying that the hockey stick is irrelevant because other proxy studies say the same thing, then this would give one even more cause to endorse M&M’s call for an independent evaluation of data archiving, methods, etc. of all such relevant paleo work.

    You simply can’t have it both ways. Either (a) the hockey stick and other paleo resonstructions are important to the question of attribution and hence climate policy, in which case they deserve great external scrutiny, or (b) the paleo-reconstructions have been oversold as evidence of attribution, given the vast other non-paleo literature that makes the case as well, in which case the IPCC improperly used the HS as an icon in the TAR. It looks like (b) is the case from where I sit, are willing to argue for (a)? If so then you may have more in commmon with the commenters at ClimateAudit than you’d care to admit!

  10. 6
  11. Tom Rees Says:

    Roger, I think that the truth is half way between your two caricatures of what WMC is saying. It’s clear that the hockey stick was part of the scientific case in the TAR. I don’t think it was a key part – that’s clear from a perusal of the SPC (it’s 5% of the text, and 1 on 15 or so figures).

    Since the TAR, there have been a number of other reconstructions of millenial temperatures (not only climate), as well as new reconstructions of past climate forcing (especially solar and volcanic). There have been many new studies of climate sensitivity related to the past millenium, most of which base their conclusions only partly on MBH99, and some of which are based on other material entirely.

    So, in AR4, millenial climate and what it can tell us about climate sensitivity will still feature, but it will still be only a relatively small part of the whole. Furthermore, the contribution of MBH99 to that small part will be smaller still, because there has been so much additional work conducted.

    However, the essential conclusions about what millenial climate reconstructions can tell us about climate sensitivity will, I believe, be pretty much the same in AR4 as they were in the TAR. The reason for this, as WMC points out, is that subsequent work is largely in accord with the seminal work by Crowley et al 2000.

    Therefore, if MBH was to suddenly disappear, the (expected) conclusions of AR4 would not change greatly (although the absence of MBH would have had a somewhat larger, although still small, effect on the TAR).

  12. 7
  13. Hans Erren Says:

    Here’s more on “the seminal work by Crowley et al 2000″ :-D

  14. 8
  15. Steve McIntyre Says:

    Folks, you are ignoring the fact that the other sudies are not “independent” of the Mann HS. We have never argued that the Mann HS is “simply” an artifact of the flawed PC method, as people have sometimes mischaracterized our work. The flawed method interacts with flawed proxies – especially bristlecones. We have emphasized both data and methodology, although the total weirdness of the erroneous Mann PC method has obviously atracted a lot of interest – not least because it is interesting that something that contained this bad an error should have survided what was advertised as the most rigorous review in humnan history.

    Aside from the supposedly independent studies being done by non-independent coauthors, they are not independent in critical proxies. Bristlecones/foxtails occur prominently in every multiproxy study. So does Briffa’s flawed Polar Urals dataset which shows an exceptionally “cold” eleventh century (no wonder it is beloved of multiproxy authors), but the eleventh century portion is based on 3 incorrectly cores. If you take away only a couple of flawed proxies, the rest of the studies start to fall apart as well.

    I’m not especially familiar with the studies on forcing. What I would look for if I immersed myself in them is the extent to which they are tuned intentionally or unintentionally on HS-type results. If the HS is wrong, as granted for the sake of the discussion, it may affect other studies – a point which we insufficiently discussed in our head posts.

  16. 9
  17. Tom Rees Says:

    Steve, suppose, for the sake of argument, that every single attempt to reconstruct the climate of the past 1000 years is wrong. Therefore, every attempt to use such reconstructions to provide information about how climate responds to forcing would have to be discarded. What we would be left with is all the other studies. What these other studies tell us is pretty much the same thing that the studies looking at the paleoclimate of the past 1000 years tells us. Therefore, the overall conclusions would not change – although the strength of the conclusions would be weakened to some extent (because one of the many strings, would be gone).

    The only way an important consequence to the climate sensitivity debate would follow would be if you could replace the current reconstructions with something radically different, and show that this has important implcations for our understanding of climate sensitivity.

    This doesn’t mean, of course, that showing (as you believe you have) that paleo proxies are unreliable or wrongly employed is unhelpful. A more accurate record of the past 1000 years could, feasibly, help to constrain climate sensitivity. The problem with the current reconstructions is that, although they provide some constraint, provide no more constraint than other avenues of research (and considerably less than some).

    The reason, incidentally, that millenial reconstructions do not provide much constraint is that the currently available ones differ in some crucial details. If we could sort out which were reliable and which were not, it would be helpful.

  18. 10
  19. Dano Says:

    Roger says:

    “the paleo-reconstructions have been oversold as evidence of attribution, given the vast other non-paleo literature that makes the case as well, in which case the IPCC improperly used the HS as an icon in the TAR. It looks like (b) is the case from where I sit, ”

    Here we go again.


    Roger, can you define “proper”, and tell us what would have been a “proper” optic for the IPCC?

    Just an answer to this, please, not a link to a paper you wrote.

    Thank you.



  20. 11
  21. Dano Says:


    Tom’s third para just above gets at the essence of what I mean – personified in a lil’ comment on your site* about how you question why I think you should go out and core & do some work. Of course, Tom says it better than I do.

    You have no alternative to your negative narrative.

    It’s OK to say ‘this doesn’t work’, but you haven’t shown what is true [this is the case for all septics, which makes your position problematic].

    In order to look credible, you have to make an effort to say ‘this is not true, but I think this is’, else you get lumped in with the Singers and Dalys.

    This is esp. so when your website – at a cursory and even a second glance (OK, third glance too) – looks like it was set up to assassinate a character*. The testosterone smell over there doesn’t help, either, but let’s close by saying you have a tin ear for impression management.



    * Enabling HTML would allow me to link multiple times in comments here, making for a stronger comment [even if I'm the only one to say so]

  22. 12
  23. Steve McIntyre Says:

    Tom, it’s hard to capture the right nuance in this discussion and I’ve not caught the nuance that I’m seeking so far. If you had (1) a proven model of climate sensitivity to 2xCO2 and (2) screwed-up multiproxy studies, then, for policy purposes, any screw-up in the Mann HS (and any screw-ups by fellow travellers) could be worked around. However, I can see a couple of potential problems:

    (1) are any of the nmodels “proven” in the sense that they can accurately replicate past climate history on all scales? I don’t know the model literature particularly well, but I get the sense that different models tend to be tuned to different situations. So that Ice Age models hypothesize the existence of ice sheets, which become a “forcing” How are the key models used for IPCC projections tuned? How do they perform in Ice Age situations? Can they replicate MWP phenomena even if they are held to be “local” e.g. higher treelines in Siberia and the Sierra Nevadas and glacier retreats in the Alps? I haven’t seen any reports of this: mostly they seem to be tuned to HS aggregate outcomes. So if the HS and its cousins were wrong, you’d have to check carefully on what this affected. Also if it doesn’t “matter” – you have to check it anyway to ensure that your representation that it “doesn’t matter” is true. Anyway I don’t get the impression that any of the models are “proven” on all scales. Could the model defects lead to incorrect understanding of the effect of 2xCO2 – either overestimate or undeerestimate?

    (2) how do we know that we can rely on the climate models which are the “real” argument, if supposedly the most rigorous review in human history failed to identify problems with its posterchild. I don’t see much external due diligence on climate models, which we are being asked to rely on for estimates of sensitivity. For example, the proponents of the various models publish articles about their models in peer-reviewed literature. These articles tend to be self-serving. My guess is that they do not include a careful listing of all the potential problems known to the model authors, as few climate scientists adhere to (or even understand) standards of “full true and plain disclosure” as understood in prospectuses.

    A peer reviewer of an article about the GISS or GFDL models is hardly going to check the model itself. So if there was a defect in the model, it would not be identified by a journal peer reviewer. The IPCC has made it abundantly clear that they don’t check anything and merely transliterate journal articles. So there’s no checking there either. Where is the independent verification of any of these models by someone who is not a fellow traveller? (As distinct from intercomparisons, which are useful, but different). People say that this would cost a lot of money. Well, bank audits cost a lot of money and they are still done. There’s as much money riding on these policies as on bank financial statements. Yet no one ever checks them. I don’t have the resources to check a climate model. However, the results from checking a smaller topic – multiproxy studies – if construed as a spot sample – can give no confidence whatever in the larger topics. Maybe they are done better; I hope that they are. But, as far as I can tell, no one can say that they have carried out on independent external review.

  24. 13
  25. Paul Dougherty Says:

    One dictionary definition of proper is, “according to what is correct or prescribed for a particular situation or thing”, another is “according to or respecting recognized social standards or conventions; respectable,”

    An example of “proper” is a passing on of knowledge or the expression of a reasonable point of view by a visitor to a website that is attempting to promote understanding and civilized dialogue.

    An example of “improper” is the disrespect, ad hominem attacks, and just plain vitriol of visitors to the same website who obviously have no understanding of the meaning of the word, “dialog”.

  26. 14
  27. Roger Pielke Jr. Says:


    In an IPCC Summary for Policy Makers, proper means of relevance to decision makers. Improper thus means advanced as relevant when in reality it is either irrelevant or redundant, it does not matter.

  28. 15
  29. William Connolley Says:

    Roger – your content filters are getting ratty again and rejecting my posts. Hopefully this will get through: you’re mischaracterising what I’ve said & you’re still wrong about the significance of MBH. See

  30. 16
  31. Roger Pielke Jr. Says:

    William- Yes they are. Found out this AM that was being rejected. If anyone has troubles with spam blocking just send me an email and we’ll fix it. Thanks.

  32. 17
  33. Roger Pielke Jr. Says:

    William- You are splitting small differences now. The IPCC SPM clearly identified the HS as a critical piece of evidence on attribution. Not “somewhat” important as you have suggested. Reasonable people can disagree about whether its importance was overstated in that report, that is a matter of interpretation and judgment. Now that I have learned quite a bit more about the issue, I believe that its importance to the attribution case was dramatically overstated because it made for a powerful symbol. We can agree to disagree on this point. No need to take it further than that. Thanks.

  34. 18
  35. Dano Says:

    Thank you for your input, Paul. I’m sure Roger will come along shortly and give us his definition.

    But you forgot the second part of my question, Paul: what would have been a “proper” optic for the IPCC to use to illustrate the anthropogenic contribution to climate change? That is: what is the result of our industrialized society on earth systems, specifically the atmosphere?

    Would you have “properly” used temperature, or some other “proper” thing like human’s allocation of Net Primary Productivity? What about a proper optic capturing ecosystem stresses from human encroachment? What about the DMSP sat pix of the nighttime lights of the world? Sure, these aren’t climate-related and thus are problematic wrt climate, but what other optics are there to capture…oh, wait: maybe CO2 increases.

    Anthro CO2 increases. _That’s_ an optic. But how can the public understand, Paul, what CO2 increases mean? Should we have kept the anthro-caused temp increases to the last 150 years? 50? Then there’s the follow-up question of what optic to use now…but I’ll ask that of Roger.



  36. 19
  37. Dano Says:

    Thank you for your reply, Roger.

    Initially you called the Hockey Stick improper, because it ‘oversold’ the attribution, but when defining “improper”, you claimed it was information deemed relevant when it was actually irrelevant. Is the hockey stick irrelevant because it conveyed too much or partial information?

    Thus the optic used by the IPCC was not relevant to decision makers, if I understand your comments correctly, because it didn’t convey the information well enough?

    If so, then should information be conveyed only by text to decision-makers? Are we to assume most or all decision-makers are not visual people thus not apt to gain clarity from visuals?

    Do you, when advising and consenting decision-makers use any optics to convey information or inform decision-making? If so, what are your rules for visual presentation? Do you reject any optics that can be construed as symbology? Are there hard and fast rules to reject potential symbology yet convey information? Are there times when words can be used as symbols or to oversell information?

    I think you can see where I’m going with this. I’m troubled by your apparent position, which seemingly retroactively denounces the optic chosen by the IPCC because they failed to consider that it may be used as a symbol [or totem, as I would say], yet you have given no ground rules created by yourself or others that govern some Tufte-esque applicability of visual aids to inform decision-making.



  38. 20
  39. Roger Pielke Jr. Says:


    Good questions. It is a fair and difficult question to ask how scientists might symbolize climate change in a way that is both useful to decision makers and faithfully represents science. It is not so easy or straightforward. I’m pretty sure this was not a question addressed by the IPCC TAR, and judging from reactions in the various comments here, not a topic that is to be met with a warm reception among many with a stake in the HS debate.

    You can find my discussion of the hockey stick as symbol here:

    The questions that you raise would seem to be worth discussing in some depth as the IPCC gears up to present its AR4 results. Symbols matter.

  40. 21
  41. Dano Says:

    I agree Roger, and thank you for linking to your previous post, which is both a good reminder of the issue, precursor to the recent discussion, and indicator of astroturf energy.

    I hope you will write a post in the future on acceptable symbology and how you, personally, choose visual aids when presenting: 1. so we can have guidance for ourselves when we go forth and behold an optic, and 2. to obviate disagreements over what is an acceptable visual and what is not.



  42. 22
  43. William Connolley Says:

    Roger asserts “The IPCC SPM clearly identified the HS as a critical piece of evidence on attribution.”. He offers no evidence for this. The TAR SPM dies discuss attribution: see: There are 7 bullet points there. MBH is 1/2 of one of those. Incidentally, those bullet points come direct from chapter 12:

  44. 23
  45. Roger Pielke Jr. Says:

    William- Thanks for your comment and sorry about the spam-blocker (Thanks GM!). I did address this issue of HS as symbol of climate change in the SPM in this long post:

  46. 24
  47. William Connolley Says:

    Roger – we’ve now wandered back and forth and between blogs and rather lost sight of my original point: which was that your first “interesting question” isn’t, because of its tense error: although MBH was of some importance in the TAR (though *not* key: you have put words in my mouth there) its less important now, being only one reconstruction amongst many. Your assertion is that “the claims of attribution in 2001 were largely insensitive to the presence of the hockey stick.”. I agree.

  48. 25
  49. Jim Clarke Says:

    Through all of this, I can not help but notice that there is no mention of the effect the hockey stick image had and continues to have on the general public; the 6 billion or so people who really don’t have the time or knowledge to grapple with the details of climate change.

    While one may argue correctly about the insignificance of the study to the scientific debate, to the world of global opinion, no single image is more important or more persuasive. Politicians will usually decide policy in part on the desire of the masses. If the masses have been mislead, poor policy results, and the impact can and will be huge.

  50. 26
  51. Roger Pielke Jr. Says:

    Jim Clarke-

    Thanks for your post. Good evidence for an effect on public opinion would be a change in opinion pre- to post- hockey stick. The polls I am aware of, e.g.,

    show no such change. In any case I am skeptical of such a tight relationship between science and public opinion, and also public pinion and climate policy. I should do a full post on this, as I just paricipate in an exchange in the comments over at RealClimate on this subject, and there clearly a lot of interesting dimensions here.

  52. 27
  53. Tom Rees Says:

    Steve, you ask some involved questions that I’m not really qualified to answer. Now we’ve got that out of the way :) here goes:

    The models used in paleo work are essentially the same models as used to make future predictions. In my opinion, the ones that match climate observations from other periods (including modern) don’t recreate MBH99 – as I’ve argued in this thread on UKWW I think that you need to distinguish between energy-balance models (EBM, such as that used by Crowley et al 2000) and general circulation models (GCMs). There’s no simple way of tuning a GCM to get the desired result, but EBMs can be tuned to give a specific climate sensitivity. Crowley chose an EBM with low sensitivity, and managed to recreate MBH99. I don’t know of a GCM that has managed to find something similar to MBH99 without using values for solar variability at the low end of feasibility. As to whether GCMs can reproduce all climate on all timescales, the short answer is yes and no, depending on how tolerant you’re willing to be.

    As to whether the alleged failure of MBH99 has implications for the rigour with which climate models are tested, I would say that they are different fields with different standards. Furthermore, it seems to me that central to your argument is a belief that there is an lack of criticism within the millenial paleoclimate field. No doubt this is true to at least some extent, but I don’t believe it’s true to the extent you imply. It’s not as if MBH99 was published, and then everybody accepted it without question. Certain groups (e.g. Esper et al, Huang et al, Moberg et al) were explicit in making the point that their findings contradict MBH99. Now, you might believe that they are also wrong, but there’s no doubt in my mind that they believed that MBH99 was wrong, and that the differences are important. If you were right, then MBH99 would have been the end of the debate. However, looking back over the past 6 years, it’s clear that it was more like the beginning (even without your contribution).

    As an aside, I think that the consensus between the two camps in the current debate (you vs Mann) that all the millenial reconstructions are essentially the same is, I think, bogus and based more on politics than science. If Moberg is right and MBH99 is wrong, then it makes a big difference (but not necessarily to projections of future climate…).

    As to whether anybody checks climate models apart from modellers, I doubt very much this is the case. You have to have a fair amount of skill and computer resource, and I suspect that the modern models are so complex that no one person understands any model completely anyway. The code for many of them is freely available, and is used by teams with modelling expertise who are not ‘allied’ with the code originators. However, here is a real (and acknowledged) danger that parameter selection leads to ‘curve fitting’. Previously, there was not much that could be done about this. However, in recent years there have been concerted efforts to test model senstivity to assumptions made in parameterisation (most famously, The results from these studies tend to suggest that modellers have actually parameterized their models in such a way as to underpredict future climate change (compared with what the models potentially allow).

    Finally, there are organized efforts to understand the differences between models. Partly this is because climate modelling is a much, much bigger enterprise than paleoclimate reconstruction (which, as you are aware, has only a few full-time individuals applied to it). See the Program for Climate Model IntercomparisonPaleoclimate Model Intercomparison Project.

  54. 28
  55. Gavin Says:

    Tom, A couple of points. GCMs have been used sucessfully to test whether hypothesised forcings are consistent with paleo-climate reconstructions. However, the NH mean temperature record is not the most interesting feature – as Crowley showed you can get that with an EBM. A higher sensitivity is as consistent with the data given the uncertainties in the forcing reconstrutions (volcanic and solar). No, the actual part of the reconstruction that is most interesting are the *spatial* patterns of change. This is the cutting edge for the GCMs – whether modes of variability like ENSO or the NAO have forced (and potentially predictable) components. Our own work has used the MBH reconstruction to look for patterns related to solar and volcanic forcing with some success (Schmidt et al, 2004; Shindell et al; 2001, 2004) as have others (Adams et al, 2004 for instance).

    It is one of those odd ironies that the figure that everyone keeps talking about is actually the least important from the point of view of understanding climate responses to forcing.

    Secondly, coupled model output generated for the IPCC AR4 is available to any interested scientists for comparison and analysis and there are over 300 indpendent teams currently combing over this output very finely. Some of these teams are from the modelling centers themselves, some are complete outsiders. Since many of the analyses being done have never previously been looked at many of the centers, there is little possibility of inadvertent ‘curve fitting’. The runs were done, the analyses are being finalised, and the chips will fall where they will. The impression that you have that only modellers can look at their own models is not true – though it may be true that other scientists in related fields don’t necessarily feel comfortable analysing model data. Hopefully that is changing though.

  56. 29
  57. Dano Says:

    >>Secondly, coupled model output generated for the IPCC AR4 is available to any interested scientists for comparison and analysis and there are over 300 indpendent teams currently combing over this output very finely. <<

    See, this is the bankruptcy of the IPCC: these data should be made available for _amateurs_ to audit, not just the high dendro priests…um…because it’s been amply demonstrated that MBH do something that somebody doesn’t like…or…something…hmmm…shoot, how does that go…

    John A, other C.A. expert commenters, can you take over for me? Get my back, so to speak.



  58. 30
  59. Roger Pielke, Jr. Says:


    Thanks. You write, “coupled model output generated for the IPCC AR4 is available to any interested scientists for comparison and analysis and there are over 300 indpendent teams currently combing over this output very finely.”

    Really? Is there a list of these teams? Seems like a great potential for considerable redundancy of effort here.

    I wonder how many teams of researchers are working on developing post-Kyoto climate policies. I’d bet way less than 300!

  60. 31
  61. Steve McIntyre Says:

    It is nice to hear that “coupled model output generated for the IPCC AR4 is available to any interested scientists for comparison and analysis”.

    This is obviously a very different state of affairs than in paleoclimate where I am very knowledgeable about data availability. I have listed the gaps in data availability in paleoclimate on a number of occasions.

    IPCC specifically does not require paleoclimate scientists to archive data on the grounds that this would be interfering with journals. This has been made clear to me in correspondence by IPCC WG1 and at the US CCSP workshop. So I am surprised to learn that policies in respect to model data and results differ so remarkably. Does IPCC have policies in this respect or is this just a differing trade practice?

    I might add that I was asked to be a reviewer for IPCC 4AR and accepted the invitation. In that capacity, I asked for data relating to two unpublished studies from IPCC, the authors and the receiving journals. Both IPCC and the authors refused to make the data available to me. In fact, after being directed to inquire of the authors for data by the IPCC WG1 TSU on the grounds that they would not perform “Secretarial services” for me, when I proceeded to make such inquiries, IPCC later and inconsistently took the position that my inquiring for data from the authors constituted a breach of confidentiality, since my knowledge of the existence of the articles derived from the IPCC review process. Their position was that the role of the reviewers was not to carry out independent review of the articles for the IPCC, but merely to ensure that the IPCC characterization of the articles was consistent with journal representation. They argued that I did not need to see any data to discharge that duty and accordingly was not entitled to examine the data if the authors and the journal chose not to make it available. It seems too strange to believe, but it’s so.

    Since realclimate has inaccurately characterized the availability of data in paleoclimate, where I have intimate knowledge of data availability, and refuses to accept any posts to the contrary, you will pardon me if I do not have a great deal of confidence in your assurances that the situation is so much better in modelling.

  62. 32
  63. Gavin Says:


  64. 33
  65. TCO Says:

    I have seen (in action) the tendentious habit of RC to censor quite temperate posts that argue on the side opposite them. And I say that as an honor code following Naval Academy graduate.

  66. 34
  67. Tom Rees Says:

    Gavin, you make a good point about spatial patterns. This highlights another reason why MBH98 was so important – it’s the only reconstruction to give spatial information with the necessary detail. I’ve read the GISS papers you mention, but will look up the Adams one.

    Another point worth making is that, although we always talk about temperature, it’s only one part of the climate system. The PCA technique has been used to recreate other climate indices, such as the NAO (e.g. Lutterbacher) and SOI. Models can and have been tested against these also, with varying degrees of success.

  68. 35
  69. Tom Rees Says:

    Gavin, another thought. Given that biological models of tree growth exist, has anyone tried to put the output from a GCM (temperature, soil moisture content, surface radiation etc) and CO2 etc into such a model to compare with actual recorded tree rings. This would be an additional diagnostic tool that would not be senstive to criticisms about whether tree rings are adequate temperature proxies.

  70. 36
  71. Steve McIntyre Says:

    Gavin, why such pain. Look that’s a commendable archiving of model information. One can also say , sigh, and point to and you’ll see a vast archive of paleoclimatic data. But what isn’t there is data sufficient to replicate the key multiproxy studies because:

    sigh – Briffa has not disclosed the sites used in Briffa et al [2001];
    sigh – Esper uses sites that are not archived
    sigh – Crowley used grey data and has “misplaced” the original data and doesn’t remember where he got the data;
    sigh – Moberg used some grey data, which he has either been unable or unwilling to provide;
    sigh – Mann said that he used data that he didn’t; used grey data; edited data without disclosure; said that he used methods that he didn’t; and after a lengthy striptease archived a complete dataset only in July 2004.

    The problems go back to underlying proxy archiving:

    sigh – Thompson archived decadal dO18 data on Himalaya sites for the first time in 2004 after a series of complaints by mem but – sigh- sample information pre-averaging remains unavailable as does the rest of the core information.

    sigh – Jacoby has archived less than 10% of his results from the last decade, choosing to archive only results that “tell” a story. sigh – he’s not reported updated results from Gaspe since , sigh- it “doesn’t tell” as good a story. sigh – he hasn’t archived the location of the Gaspe sample and now says it can’t be located because it was , sigh, done pre-GPS.

    So – sigh- it’s wonderful that modelers are well-behaved. But this proves nothing about paleoclimate which is where we started on this.

    So save your pretentious sighing and ask your paleoclimate cousins to archive their data and methods, rather than – sigh – falsely telling readers at realclimate that the data has been there all along.

  72. 37
  73. Steve McIntyre Says:

    BTW when I checked the link and checked the first entry, all I found was the following information:

    “Changing Modes of Variability and Their Impact on Extreme Weather Events
    PI: John Abatzoglou
    Institution: University of California, Irvine
    Additional Investigators:

    I am interested in studying the changing nature of preferred modes of variability (e.g., ENSO, NAO, PDO, MJO) and their impact on the occurance of extreme weather events particularly over the Americas.”

    Is that what you mean by a proper archive of data and methods? But don’t let me divert you from answering about paleoclimate.

  74. 38
  75. Hans von Storch Says:

    The debate about the hockeystick is techically not really relevant. We have achived our main goal, namely that the premature claims that the issue of millennial temperature reconstructions was mostly solved have been broadly rejected. One or two years ago it was hard to publish results which were inconsistent with the MBH reconstruction; now everybody agrees that there may be more to it. The jury is still out and I expect that consensus will settle on something with significant larger variations in the shaft of the hockeystick.

    Having said this – the debate about the hockeystick is most significant when it comes to the culture of our science. Posting the hockeystick as key evidence in the SPM and Synthesis Report of the IPCC was simply stupid and evidence for what Bray calls post-sensible science – as science which is encroached by moral entrepreneurship. Or post-normal science. We have more cases of this type of claims-making, which is usually a mix of “good” political intentions and personal drive for the limelight. Have we, as a community, become better in rejecting such claims? I am afraid, we have not.

  76. 39
  77. John A Says:

    It seems that William Connelley can’t resist the temptation of all political propagandists to rewrite history by careful deletion of unwanted facts.

    The SPM features prominently the HS
    and makes only one reference to a reconstruction of climate for the last 1000 years

    Yes, its that Hockey Stick again.

    The rest are climate models, which as we know have zero predictive value, but fortunately for Bill Connelley that isn’t their purpose. Their purpose is startlingly clear – to rewrite the past to conform to the prejudices of the authors. They provide scary headlines, lots of employment for people who otherwise would struggle to stay employed at a whelk stall, and lots and lots of free time to delete other people’s comments on weblogs, delete other people’s words on wikipedia, and a startling amount of time to spend producing materially false statements on other people’s websites.

    Like Connelley’s denial of the Global Cooling Scare of the 1970s, it’s not what he writes in but what he leaves out, that is so fascinating.

  78. 40
  79. Harold Brooks Says:

    The link that Gavin provided is to the listing of projects that have registered using the data. The data are available from the links in the left-hand column of the page.

  80. 41
  81. Steve McIntyre Says:

    I think that the archiving of model results is very commendable. In a quick look at the left frame URLs, I could only see links to certain IPCC models and was unable to see model results for e.g. the Abatzaglu model.

    But even if all the models are archived (which it doesn’t look like), that is still unresponsive to the problems in the paleoclimate area, especially centered around a small subset of authors: Thompson, Jacoby, Esper, Briffa, Jones etc.

  82. 42
  83. Jim Clarke Says:


    I stand corrected on my opinion that the Hockey Stick has had its biggest influence on public opinion. The examples of public opinion polls you provided indicate little change since 1997, a year or two before the Hockey Stick appeared. At most, it may have just reinforced the believers, while the 25% not buying into the hype before, didn’t buy the Hockey Stick either.

    What was most interesting about the polls were the comments provided by the poll takers. These comments revealed a serious lack of understanding of the science and a reiteration of the most stricking ‘doom and gloom’ scenarios. It seems obvious that the large portion of the general public that believes a global warming disaster is right around the corner, have accepted the worst pronouncements of the mainstream media and combined them into some kind of Stephen King nightmare!

    On the other hand, it is interesting that the 75% are not demanding more action. I have a theory that poll-takers actually feel like they are being ‘tested’ and often search to give the ‘right answer’ to show the pollsters how smart they are. I think this is particularly true when asked about science related issues. Do these people really believe the climate is out of control and the future looks bleak, or they just trying to pass the test?

  84. 43
  85. Steve Bloom Says:

    Here are some current poll results, first from California:

    - 86% of California adults believe that global warming will affect current or future generations. 57% believe the effects are already being felt.

    - 62% identify human activities as the primary cause of global warning. Only 22% say naturally occurring increases in temperature are responsible.

    - More residents trust the state government (52%) than the federal government (43%) to provide correct information about the condition of the environment.

    - A majority (54%) believe that the state government, apart from the federal government, should address the issues of global warming.

    - 38% of Californians say they approve of President Bush’s performance in office. Fewer approve of his handling of environmental (32%) and energy (29%) issues.

    These results are rather more tilted toward action than the slightly older ones Roger cited. I included those last figures since they demonstrate that a substantial chunk of the remaining Bush supporters in CA disagree with his global warming stance. See the whole poll at . The oldest similar poll that I could find for California (from 7/2000, available on the same site) seemed to show somewhat weaker support for action on global warming, although the questions weren’t phrased the same. What was interesting is that the “more research” cop-out answer seems to have nearly vanished.

    The fresh Fox News poll at seems fairly consistent, with 77% saying they think global warming is happening, although the order of questions 3, 4 and 5 might have tended to pull the results down a bit.

    The wierdest result in this poll was question 29:

    Do you think the United States is doing more or less to reduce global warming than other industrially developed countries?
    - 34% U.S. is doing more
    - 38% U.S. is doing less
    - 7% U.S. is doing the same
    - 2% don’t believe in global warming
    - 19% don’t know

    On the one hand when put in this context the deniers are reduced to a statistically insignificant 2%, but on the other hand I wonder what planet those 34% have been living on.

    Generally speaking, results from these polls seem to indicate that while there may be a slight trend toward convincing the remaining undecided, with the die-hard deniers down in the 10% range, that part of the game is indeed over. Efforts now need to be directed toward convincing those who already understand global warming is a problem to support sufficient action to deal with it.

  86. 44
  87. Murray Duffin Says:

    Speaking for just one lay person, I was an ardent believer in AGW, to the extent of making serious presentations at industry conferences, prior to reading the TAR, and it was the HS curve plus the SRES that caused me to start questioning the science. However I have never met a lay person face to face that has put as much effort into following the debate as I have. I have encountered about 10 on the internet, about equally split on their positions, although a couple of them are involved in some aspect of climatology and therefore not strictly laypeople. I have discussed AGW with some 10s of Californians who were purely general public, and found that almost all believe in global warming, and a large majority believe in AGW, but none knew anything of the science or the details of the debate.
    As for Steve Bloom’s assertion about the 2%, he misses the point. Most skeptics are not in the camp of not believing in global warming. However a significant portion of the small share of the other 98%, who are also prety well informed, do question the extent of the warming, the degree of contribution made by CO2, and the dire predictions of the likely future. I wish people would stop mischaracterizing the intelligent and informed skeptics. Murray Duffin

  88. 45
  89. Steve Bloom Says:

    Murray, note that I carefully used “denier” rather than “skeptic,” which I believe is the distinction you would like drawn. Of course there are many skeptics in the other response categories in these polls, although I think we both would agree that the level of information most people base their views of climate change on (noting again that 34% result that flies in the face of all reported reality) is so limited that using a term like skeptic to describe anything beyond a very small group of people is probably inappropriate. The same logic applies to global warming “believers.”

  90. 46
  91. Ross McKitrick Says:

    Roger: regarding item #4, we deserve some credit for trying to get the So What? message out. But we had to first get a hearing on the What? question before anyone would listen to us on the So? question. Prior to Feb 2005 we didn’t have much of an audience, though I did get asked to do a talk on the policy angle at a USask conference on science and democracy in fall 2004.

    After GRL in Feb 2005 the invitations began arriving and in April-May we did 11 talks in 7 cities, each time talking not just about decentered PCs, but (at length) about the policy questions we think the episode raises. We get criticised for speaking to Marshall/CEI events, but they’re the ones who showed enough interest to ask us to speak, and if you look at the transcript we did talk about the implications of understanding the inherent limitations of peer-review and assessment reports in the policymaking process. I used an invitation to give the keynote at the International Policy Forum on GHG Management at UVic in April 2005 to talk about bias-proofing the assessment process. That paper just came out in a journal.

    So we’ve been trying to get the discussion going on the So? side, and it does seem to have a lot of traction with audiences. But in terms of engaging a larger audience the fact is there’s been no interest from magazines like SciAm or New Scientist, or TV science reporters, whose only spin on this issue is the valiant scientists hounded by the evil oil industry. For whatever reason it’s the conservative think tanks that invite us to talk about the institutional questions being raised. Pew, to take one example of a non-conservative think tank supposedly interested in climate policy issues, has not contacted us, nor have any government agencies, environmental groups, etc. I take that to mean they’re unaware or uninterested in the issue, or they don’t want to think about it. It’s not that we’re reluctant to get the discussion going about what-it-all-means-for-policy: the demand in many quarters is just not there.

  92. 47
  93. Dano Says:

    While not quite up to an LOL level, Ross, your comment above did elicit a chuckle. Not enough to spray coffee on the keyboard, mind you, but a good ‘un nonetheless.



  94. 48
  95. Paul Dougherty Says:

    Murray Duffin said, …”a significant portion of the small share of the other 98%, who are also pretty well informed, do question the extent of the warming, the degree of contribution made by CO2, and the dire predictions of the likely future. I wish people would stop mischaracterizing the intelligent and informed skeptics”. I believe that Jim Clarke would join me in seconding you on that sentiment!

    In comments above, Hans Von Storch, who is certainly a little more informed than the three of us, effectively praises M and M for opening up a debate on the hockey stick. He goes on to deplore its use by the IPCC. Yet M&M are ridiculed by some on this site and particularly at Real Climate. Why? Do you folks have a corner on climate knowledge or science in general? Condescending comments and ad hominem attacks have no place in scientific debates and usually serve to display the unreliability of the author.

    Like Von Storch I do not argue with the qualitative conclusion of the hockey stick. I am, however, highly disturbed at the (to me) unjustified quantitative implications. And as a scientist I think the stick is useless until the same type of data are used in the clubhead as in the handle. This kind of representation along with such things as purporting to know a future fifty to a hundred years from now does the debate no good.

  96. 49
  97. Dano Says:

    >>And as a scientist I think the stick is useless until the same type of data are used in the clubhead as in the handle. This kind of representation along with such things as purporting to know a future fifty to a hundred years from now does the debate no good.<<

    Please Paul. Really.

    You know full well there is no instrumental record to draw upon to have the same sort of data pre-ca. 1850s. That argument won’t fly.

    And your ‘purporting to _know_ a future 50-100 years from now’…is a hoot. That strawman is so loosely wound it will ignite in a warm breeze. Surely there are better talking points out there, waiting in the weeds to be spotted by intrepid seekers.



  98. 50
  99. Paul Dougherty Says:

    Surely you are aware that the hockey stick handle is paleo data while the club head is instrumental readings.
    As far as knowing a climate future for 50-100 years, just read earlier postings on this site and at RC, etc.
    You are a master of the attack though. Dano, and I am no match for you there. I prefer to treat everyone with respect in the hope of learning something.
    Would you comment on the Von Storch post that I referenced as well as the first two paragraphs of my post? Thank you

  100. 51
  101. Dano Says:

    Paul, you said “I think the stick is useless until the same type of data are used in the clubhead as in the handle.” As a scientist, you know this can’t happen. So I don’t know why you clarified in your comment.

    And your “As far as knowing a climate future for 50-100 years, just read earlier postings on this site and at RC, etc. ” I presume this is a misunderstanding on your part about scenario analysis. Surely scenario analysis is batted about in some circles, even to the point of using ” ” around scenario, as if it wasn’t serious.

    As a scientist, you surely know the increasing use of scenario analysis in projecting future possibilities, as projections better account for surprise, whereas predictions do not. If’n it’s good enough for the military & biness, and increasingly scientific adaptive management (as a scientist, you know about this so forgive me for the obvious) it’s good enough fer me.

    And to your “I prefer to treat everyone with respect in the hope of learning something. ” You’ll excuse me, as I’ve misinterpreted your comments elsewhere and here as being something rather different. Unless there is another PD somewhere who I’m confusing with you. If so, apologies.

    But to your question, can you give me an example, today, of an optic that should be deplored? Knowing what we know in the future, that is, then looking back in hindsight at it. And to the ‘corner on knowledge’ point, it’s about the narrative and hiding the approach, and I believe my comments above frame your question accurately – that is: I’ve already commented on it here and elsewhere on this site beyond the point of prolixity. Namely, the CA framing of the underlying issue is a constructed narrative, and the type of non-response responses to addressing that narrative should be a clue.



  102. 52
  103. Number five is alive Says:

    I think the Dano bot has thrown a cog… “Namely, the CA framing of the underlying issue is a constructed narrative, and the type of non-response responses to addressing that narrative should be a clue.”

    Eliza was better.

  104. 53
  105. Dano Says:

    I’ve docked the pay of my editors and they are attending a continuing ed. class today to ensure I post with clarity, continue to have coffee available, and sleep the requisite number of hours. They are also scanning resumés to find someone to help clear the dead trees off my desk.

    All in order to ensure that I clearly point out the non-response responses on this thread (and others).



  106. 54
  107. TCO Says:


    Not clear if you don’t follow, disagree, or are just being tendentious. Of course, there is no instrumental information pre1850. I’d be the first to have an issue with JohnA or a similar type if they said something like that. (Like you JohnA is one who doesn’t follow a lot of the math.)

    But the point is that you should use an ALL-PROXY curve then. Otherwise, you’re picking and choosing. If the proxies are so damn good that they’re reliable pre-1850, then they should be reliable post-1850. If THEY’RE NOT RELIABLE post 1850 (and you need to use instrument data then), then that casts doubt on their effectivess pre-1850 or even on their status as proxies.