Basic Truths We Should All Be Able to Agree To

There are some basic truths everyone should be able to agree to. In discussions of climate science, one of those truths is the greenhouse effect is real. I’d like to discuss what I believe should be another.

A few days ago I came across a post at the (formerly known as) wottsupwiththat blog. I was disturbed to see this:

So, I’m no longer interested in discussions about aspects that are largely indisputable. The rise in atmospheric CO2 is anthropogenic. There is virtually no doubt. Our current overall warming is also anthropogenic. Again, there is virtually no doubt. Michael Mann’s Hockey Stick has not been debunked, despite what people might say (McIntyre & McKitrick 2005 has numerous easily explained issues). Exactly where to draw the line with respect to the science is trickier, as there aspects that are uncertain and I certainly don’t even want to prevent discussion about those aspects that are certain. I just don’t want comment threads diverted into lengthy exchanges about something that is largely undisputed.

The first notion is relatively indisputable. I’m not sure what the author means by “overall warming is also anthropogenic” for the second notion. I’m going to assume author means humans have contributed some amount of warming as that is indisputable. The third notion is where my problem lies.

Michael Mann’s hockey stick has been the subject of a great deal of discussion, and it seems the author is simply dismissing all of that with a wave of his hand. Even worse, he says “McIntyre & McKitrick 2005 has numerous easily explained issues,” but he doesn’t say what those issues are. He doesn’t provide a link to a discussion of them. He doesn’t provide a reference for the claim. Even worse, he doesn’t explain why we should only care about that one paper when multiple papers were published criticizing Mann’s hockey stick.

Worse than that? McIntyre & McKitrick published two such papers in 2005. The author of that post refers to one but doesn’t distinguish which. That flawed reference, combined with his failure to mention any other paper, seems to suggest he doesn’t know multiple papers exist. I’m at a loss to explain the paragraph otherwise. Regardless, I thought it’d be worthwhile to seek out the author’s previous refutations of McIntyre and McKitrick’s work. A search of his site pulled up this post from five months ago. I read it, saw the author had little knowledge of the material he discussed in it, and today I wrote up a response. Unfortunately, when I went to submit it, I was given an error by his blog software. It seems the post cannot be commented on now (perhaps due to its age). As such, I’m posting what I wrote here:

This post unfortunately doesn’t examine anything in depth. To address that, I’ll attempt to. To begin, I’ll answer the question asked of Steve McIntyre and/or Ross McKitrick:

did you properly remove any underlying hockey stick profile from the data you used to produce the red-noise that you used in your 2005 paper

This question is poorly phrased. Whether or not the NAOMER series ought to have been detrended prior to being used to calculate red noise profiles is a matter one can dispute. Consider, for example, this post at Real Climate. It discusses the use of pseudoproxies created via examining the red noise profiles of the NAOMER series used by McIntyre and McKitrick. Not only do the authors not say it is necessary to detrend the series first, Michael Mann himself specifically promotes calculations as being done without such a step (in several inline responses). If the lead author of a paper promotes a process, it is difficult to see why critics of the paper should have their criticisms disputed based on the use of that process.

With that caveat in place, I can answer the question as, “No.” However, I can go beyond that. Critics of the hockey stick are often faulted for not examining the impact of the supposed flaws they find. To address this, we can examine the impact using a different red noise profile would have. Fortunately, it’s already been done. Take a look at this comment from Nick Stokes on his post criticizing the Wegman Report for how it critized Mann’s hockey stick. Nick Stokes is a frequent defender of the hockey stick, the post was defending the hockey stick, and he still acknowledges “the MBH tendency to make PC1 HS-shaped.” Another user posted:

I’ve put up plots of runs for AR1 rho=0.1 and AR1 rho=0.0001 (at the same googlsites page linked to by Nick). These are for 2000 rather than 10,000 PC1’s, to save time.

The random samples from the results of the MBH method contain many hockeysticks (over 50%). The hockey-sticks are more pronounced for rho=0.1, but also present for rho=0.0001.

The point this user was making is no matter what red noise profiles one uses, the result is the same – the MBH PCA methodology mines for hockey sticks. That is true independent of whatever noise MM may have used.

I also wrote a follow-up comment which addresses the issue of Mann’s hockey stick more generally. I feel the position it advances is unremarkable, and it should be something everyone can agree to. I think it is a basic truth people need to accept in order to advance discussions:

Now then, MBH’s PCA methodology did not create the hockey stick out of nothing. When it mines for a hockey stick, it merely looks for data which has something of a hockey stick within it. It then gives that data extreme weight, causing the results to be a hockey stick. What this means is whether it be red noise or the NAOMER tree ring series, finding a hockey stick requires there be a hockey stick somewhere in the data.

The issue is not whether or not some data shows a hockey stick. The issue is whether or not the results are representative of the data being examined. In the case of MBH’s hockey stick, they are not. Mann himself has acknowledged this in regard to MBH98 in his recent book (page 51):

The tests revealed that not all of the records were playing an equal role in our reconstructions. Certain proxy data appeared to be of critical importance in establishing the reliability of the reconstruction–in particular, one set of tree ring records spanning the boreal tree line of North America published by dendroclimatologists Gordon Jacoby and Rosanne D’Arrigo.

The records he refers to are the NAOMER tree ring series he used. These are the same series MM claim are responsible for the hockey stick. MM claim Mann’s hockey stick wouldn’t have existed without those series, contrary to MBH’s paper which says:

On the other hand, the long-term trend in NH is relatively robust to the inclusion of dendroclimatic indicators in the network

Not only does is the MBH hockey stick not relatively robust to the inclusion of tree ring data, it is entirely dependent upon a small amount of tree ring data – a fact Mann himself has basically acknowledged.

The original hockey stick is dependent entirely upon a small amount of tree ring data. Of the 70 NAOMER series which extended back to the period in question, approximately 20 contribute to the hockey stick. Remove those 20 series, and the hockey stick vanishes. Put them back in, and it reappears. Those ~20 series are responsible for the iconic hockey stick, and the hundreds of others were mostly irrelevant.

One can ignore the topic of red noise if one wants. That demonstration was used merely to show how MBH’s PCA methodology used a small amount of series to create results not representative of the data in general. You can accomplish the same by just removing the data MBH gave undue weight to. The MBH authors did this themselves, and they found the same thing.

That just leaves the question, was it right to base our understanding of northern hemispheric temperatures on basically one set of tree ring data? Was it right for the IPCC to heavily promote such an understanding? I say no.

If people could agree to this, we could try to move on. We could try to discuss what paleoclimatology has done since Mann’s hockey stick. We could discuss criticisms of subsequent temperature reconstructions and whether or not those criticisms are sufficient to call the mainstream position into question. We could make progress and come to understandings.

What I’ve said about MBH’s methodology is beyond dispute. It is easier to verify than the radiative physics behind the fact CO2 emissions cause warming. There is not a single technical, or even coherent, argument which disputes the fact Mann’s hockey stick was dependent entirely upon a small amount of tree ring data.

Advertisements

19 comments

  1. This may not go well, but I’ll give it a try

    Our current overall warming is also anthropogenic

    What I meant by this is that the rise in total energy in the climate system is almost entirely anthropogenic. Internal variability can change the rate of warming in the various parts of the climate system, but the overall warming (or overall increase in energy) is anthropogenic. If it was natural, we’d expect the overall energy to (on average) remain constant. It might vary with time, but the long-term rise cannot be explained via some natural process (or, at least, there is no known natural process).

    As far as your criticism of MBH98 is concerned, I don’t dispute the issues. It may also be true the using the MBH98 data to produce the red noise is largely irrelevant. It does, however, seem odd – as a physicist – to see people claim to produce independent random red noise, but to do so using the data they’re trying to compare to. Maybe that illustrates my ignorance with respect to what actually happens here, but it still seems a little odd. What seems indisputable, though, is that the 10 hockey sticks presented in MM05 (one of the papers, you probably know which one) were not selected randomly from their sample of 10000. They were chosen to be most hockey-stick like. People, however, clearly interpret the results of MM05 as implying that random red noise typically produces hockey sticks, rather than random red noise sometimes (probably quite rarely) produces hockey sticks.

    So, the issue really is how one should judge the issues with MBH98. There clearly are statistical method had flaws. That seems accepted. It does seem, however, that these flaws don’t significantly influence the result (maybe I’m wrong about that, but I’ve seen it said that this is the case). It may also be that it relied on only a few of the data sets. You seem to conclude that the IPCC was wrong to base so much of our understanding of millenial NH temperatures on this one piece of work. I don’t know if they did this, but what seems clear is that the broad picture in MBH98 has been reproduced many times by many different authors, including some who criticise MBH98. So, let’s even assume that the IPCC were wrong to highlight that paper when they did. So what? There’s nothing today that particularly contradicts the basic conclusions. You can read the summary in the recent IPCC document, and very little seems to have changed. Details, certainly, but the broad picture, not really.

    If people could agree to this, we could try to move on. We could try to discuss what paleoclimatology has done since Mann’s hockey stick. We could discuss criticisms of subsequent temperature reconstructions and whether or not those criticisms are sufficient to call the mainstream position into question. We could make progress and come to understandings.

    But we have. There has been numerous work since MBH98. MBH98 is probably largely irrelevant today. Why does it matter whether – today – we agree or not about the merits of the work. It really doesn’t matter as far as I can tell. Ignoring MBH98 from now one will not change our understanding of millenial NH reconstructions one bit (I think).

    So, here’s where I find these kind of discussions frustrating. This one may turn out to be the exception though. It seems as though people want others to acknowledge both the issues with the paper (for example) and their judgement of these issues. It’s not clear why that’s reasonable. I can agree that a paper had some issues, but disagree with what we should conclude from that.

  2. What I meant by this is that the rise in total energy in the climate system is almost entirely anthropogenic. Internal variability can change the rate of warming in the various parts of the climate system, but the overall warming (or overall increase in energy) is anthropogenic.

    I’m not sure I understand this argument, but I also don’t want to focus on it. I’ll just say if natural variability can impact measured temperatures to a significant degree, it’s difficult to discuss “overall warming.” If you pick a period and say there was X warming over it, I don’t see how you assign a % of that warming to man.

    As far as your criticism of MBH98 is concerned, I don’t dispute the issues. It may also be true the using the MBH98 data to produce the red noise is largely irrelevant. It does, however, seem odd – as a physicist – to see people claim to produce independent random red noise, but to do so using the data they’re trying to compare to. Maybe that illustrates my ignorance with respect to what actually happens here, but it still seems a little odd. What seems indisputable, though, is that the 10 hockey sticks presented in MM05 (one of the papers, you probably know which one) were not selected randomly from their sample of 10000. They were chosen to be most hockey-stick like. People, however, clearly interpret the results of MM05 as implying that random red noise typically produces hockey sticks, rather than random red noise sometimes (probably quite rarely) produces hockey sticks.

    I’d be willing to discuss why what was done was done, but first we need to get past the fundamentals. Namely, I have no idea which paper you are referring to. I’ve read both the GRL and E&E 2005 MM papers, and neither does what you describe. You can check the links I provided if you don’t believe me.

    What you say “seems indisputable” is actually completely false. You seem to be attributing to McIntyre and McKitrick something true only of the Wegman Report. You then go on from this completely wrong statement to conclude MBH’s methodology “probably quite rarely” produces hockey sticks from red noise. I have no idea what makes you say this, but code to see how often hockey sticks arise from red noise is freely available. Nick Stokes generously provided his in his post I linked to, and results with several different levels of red noise are linked to in the same post.

    There clearly are statistical method had flaws. That seems accepted. It does seem, however, that these flaws don’t significantly influence the result (maybe I’m wrong about that, but I’ve seen it said that this is the case). It may also be that it relied on only a few of the data sets.

    You are wrong on the first count, and it is indisputable what I said is true on the second.

    You seem to conclude that the IPCC was wrong to base so much of our understanding of millenial NH temperatures on this one piece of work. I don’t know if they did this

    The IPCC Third Assessment Report made Mann’s hockeystick a key element of its PR campaign. It was the most iconic image in global warming discussions.

    but what seems clear is that the broad picture in MBH98 has been reproduced many times by many different authors, including some who criticise MBH98. So, let’s even assume that the IPCC were wrong to highlight that paper when they did. So what? There’s nothing today that particularly contradicts the basic conclusions. You can read the summary in the recent IPCC document, and very little seems to have changed. Details, certainly, but the broad picture, not really.

    There are two major problems with this. First, the reconstructions since Mann’s hockey stick have been routinely criticized, often for problems related to MBH’s. For example, MBH’s NAOMER PC1 (which I discussed in this post) was used in numerous other temperature reconstructions. If reconstructions go so far as to use outputs from MBH, it’s not surprising they might get similar results to MBH. The fact using similar methodology and/or data can produce similar results is meaningless.

    Second, phrases like “broad picture” are uninformative. Just what are the “basic conclusions” nothing contradicts? I have no idea. This is what happens when we compare Mann’s original temeperature reconstruction with several more recent recontrusctions. They aren’t close.

    But we have. There has been numerous work since MBH98. MBH98 is probably largely irrelevant today. Why does it matter whether – today – we agree or not about the merits of the work. It really doesn’t matter as far as I can tell. Ignoring MBH98 from now one will not change our understanding of millenial NH reconstructions one bit (I think).

    The primary reason is MBH introduced problems that are present in practically every major reconstruction since. It exemplifies what is wrong with the field. If people agree MBH was wrong to do certain things, they’ll have to agree it was wrong for other groups to do those things as well.

    Suppose you agree NAOMER PC1 was calculated via inappropriate methodology as I claim in this post. What do you do when I then tell you it was used in half a dozen other temperature reconstructions? If it’s a problem for MBH, it’s a problem for them as well. That means we’d have to examine what effect its use had on them prior to accepting their results.

    Aside from that, MBH is the most analyzed and well-understood temperature reconstruction. There’s no way we can expect people to agree about other reconstructions if they can’t agree on MBH. MBH is the first step to resolving disagreements – start with simple points everyone can agree to and move on.

    So, here’s where I find these kind of discussions frustrating.

    What I find frustrating is people draw conclusions based on little knowledge while ignoring the large amount of knowledge available to them. It’s even worse when they use faulty knowledge such as your description of MM2005. How is one supposed to respond to that, especially if those people refuse to discuss matters?

    I have a standing offer to discuss any temperature reconstruction in any amount of detail with anyone. Nobody ever takes me up on it. It seems like the sheer number of reconstructions is enough to convince them.

  3. Okay, Brandon, this is kind of going as I expected. You regard much of what you say as “indisputable” and say “What I find frustrating is people draw conclusions based on little knowledge while ignoring the large amount of knowledge available to them”. Technically, I haven’t drawn any conclusions. I’ve presented some arguments that you claim to have shot down. I’m not really interested in such a discussion. You clearly believe your conclusions very strongly and that is your right. I reserve the right to draw my own.

  4. You certainly have the right to draw your own conclusions. You also have the right not to participate in any particular discussion. However, you’ve made at least one indisputably wrong statement of fact in your first comment here. You do not have the right to do that and leave without addressing it. As such, I must request you address what you said about McIntyre and McKitrick’s work and my response. Specifically, please explain why you believed something “seems indisputable” when simply glancing at the referenced work would show it was false.

    I started this blog just to have a place to post things in case I needed to link them for someone. It wasn’t intended for discussions, and I doubt there will be many. However, one thing I will require of anyone posting is any challenge to a direct statement of fact must be answered.

  5. I actually have to go to a meeting so I’m going to not be responding anyway for a while. I really think you should read Nick Stoke’s post more carefully. I also, may have been wrong about it being 10, but here – I believe – is the bit of M&M05’s code in which the select the 100 most hockeystick like from their sample

    #SAVE A SELECTION OF HOCKEY STICK SERIES IN ASCII FORMAT
    order.stat<-order(stat2,decreasing=TRUE)[1:100]
    order.stat<-sort(order.stat)

    hockeysticks<-NULL
    for (nn in 1:NN) {
    load(file.path(temp.directory,paste("arfima.sim",nn,"tab",sep=".")))
    index<-order.stat[!is.na(match(order.stat,(1:1000)+(nn-1)*1000))]
    index<-index-(nn-1)*1000
    hockeysticks<-cbind(hockeysticks,Eigen0[[3]][,index])
    } #nn-iteration

    dimnames(hockeysticks)[[2]]<-paste("X",order.stat,sep="")
    write.table(hockeysticks,file=file.path(url.source,"hockeysticks.txt"),sep="\t",quote=FALSE,row.names=FALSE)

  6. This is my blog, and while you are posting on it, I am perfectly entitled to insist you behave in any way I want.

    As for what you said, it was “technically” not a statement of fact (thought it was certainly a statement). If you wish to weasel out of this by saying you used the word “seems,” you’re welcome to try. In the meantime, I suggest you consider the possibility when a person quotes you while referring to what you’ve said, it is likely that quotation is a reference.

  7. Please reread this comment of yours and my first comment responding to you about what you said “seems indisputable” about McIntyre and McKitrick’s paper (which you still have not bothered to reference). Once you have done so, please correct any errors you note in your comments here.

  8. Sorry, but that’s not how my moderation works. I will not delete things for being wrong. Your comments will stay up so people can see them.

    For example, people will now see you feel perfectly free to make bold accusations about sources without referencing those sources. They’ll see you feel free to make bold accusations that have no bearing in reality – that you’ll say things which are obviously false to anyone who has even glanced at your sources. They’ll see when you are challenged regarding your accusations, you feel free to refuse to check your sources. And they’ll see much more.

    For example, I went through the trouble of providing links to your own source despite you refusing to specify which it was. All it would take to find the figure you claim existed within your source is 30 seconds with each link. Anyone who does so will see the figure does not exist within your source. When I pointed this out, rather than even glance at your own source, you went to a totally different source and falsely portrayed it as supporting your earlier claim:

    I actually have to go to a meeting so I’m going to not be responding anyway for a while. I really think you should read Nick Stoke’s post more carefully. I also, may have been wrong about it being 10, but here – I believe – is the bit of M&M05′s code in which the select the 100 most hockeystick like from their sample

    The code you referenced was not used in any McIntyre and McKitrick paper. It was code they wrote but did not publish the results of. As such, the results of that code were not included within the paper you referenced (or rather, failed to reference). Had you bothered to check your own sources, you’d have known this. You did not. Instead, you had the audacity to rudely tell me I should read a source I provided more carefully as though it somehow proved me wrong when the post does nothing to support the claim you had made. Not only did you refuse to look at your own source, you provided a new source which you promptly misrepresented. And now, rather than address your multiple misrepresentations, you’re running away.

    So no, I am not going to delete your comments. They are going to stay on the record demonstrating the standards you hold yourself to.

  9. Brandon, I wasn’t going to comment again as this, to me, is a classic sign of a bad faith discussion

    So no, I am not going to delete your comments. They are going to stay on the record demonstrating the standards you hold yourself to.

    However, you very clearly state,

    The code you referenced was not used in any McIntyre and McKitrick paper. It was code they wrote but did not publish the results of

    The link for the code is here. If you consider the URL it is the online Wiley library and is directly related to McIntyre and McKitrick’s 2005 GRL paper. This is what is provided (unless I’m mistaken, which I’m always willing to consider) by the publisher as the code used in their paper. The bit of code I provided in my earlier comment is directly from the code linked to here. As far as I’m concerned this is a pretty solid source.

    I have absolutely no issue with you leaving my comments. I own everything I say, even when I may be wrong. I was offering you the option of deleting my comments if you wished to do so. My statement without prejudice from me was an indication that I would not play the game you clearly are choosing to play by saying demonstrate the standards you hold yourself to.

    At the beginning of my first comment I suggested that this wouldn’t go well. It clearly has not. You commented earlier

    I have a standing offer to discuss any temperature reconstruction in any amount of detail with anyone. Nobody ever takes me up on it. It seems like the sheer number of reconstructions is enough to convince them.

    Maybe you should give some thought as to why that might be. I think I know the answer. I don’t think you’ll like the answer. This is definitely my last comment.

  10. Brandon, I wasn’t going to comment again as this, to me, is a classic sign of a bad faith discussion

    I described my moderation policy. I described what you did. I then said your comments would stand as a testament to the standards you hold yourself to. There is nothing about that which indicates bad faith. If you want to talk about bad faith, this is a sign of it:

    The link for the code is here. If you consider the URL it is the online Wiley library and is directly related to McIntyre and McKitrick’s 2005 GRL paper. This is what is provided (unless I’m mistaken, which I’m always willing to consider) by the publisher as the code used in their paper. The bit of code I provided in my earlier comment is directly from the code linked to here. As far as I’m concerned this is a pretty solid source.

    You specifically said:

    What seems indisputable, though, is that the 10 hockey sticks presented in MM05 (one of the papers, you probably know which one) were not selected randomly from their sample of 10000.

    After it had been pointed out to you MM05 was a non-specific reference. You’ve refused to cooperate on even the simplest of conversational matters – stating what you’re talking about.

    Moreover, you specifically referenced “10 hockey sticks” being presented within a paper. I directly told you they were not within the paper. Your response was not to look at the source you discussed, but instead to go to a different source all together. Rather than point to where in a paper these “10 hockey sticks” were presented, you cited code. When I informed you that particular code was not used for the paper as McIntyre and McKitrick never presented the “10 hockey sticks” you mentioned, you merely repeated that its in the code for the paper – completely failing to address what I said.

    You have been directly told, multiple times, what you claim is in the paper is not in it, and you adamantly refuse to do so much as look at the paper. That is acting in bad faith.

    My statement without prejudice from me was an indication that I would not play the game you clearly are choosing to play by saying demonstrate the standards you hold yourself to.

    Nothing I’ve said indicates I’m playing a game. You are just making things up about what I’ve said like you did about what McIntyre and McKitrick said. That is, you are fabricating things to attack people you dislike. That is a classic sign of bad faith.

    Maybe you should give some thought as to why that might be. I think I know the answer. I don’t think you’ll like the answer. This is definitely my last comment.

    I’ve given plenty of thought to it. What I know is every person who’s refused to engage in such discussions has, like you, completely fabricated things in order to attack people. And, like you, it was only after I pointed their fabrications that they quit the conversation. Based on that, I’ve concluded you guys have no interest in actually discussing these subjects. It’s the last thing you’d ever want to do. You just need talking points to promote whatever views you hold.

    In other words, you’re trolls.

  11. I find it amazing someone would claim deceptive graphs were published in a paper when the graphs didn’t even appear in the paper. I find it more incredibly amazing when that person is directed to the paper, told the paper doesn’t include those graphs, and they still insist the paper published those deceptive graphs.

    I find it so remarkable I’ve decided to highlight it a bit. andthentheresphysics, formerly known as wottsupwiththat, adamantly defends his claim this paper presented 10 hockey sticks “chosen to be most hockey-stick like.” That paper has three figures. You can see all three in this image:

    http://wp.me/a4609C-13

    There are only six graphs. Anyone who glances at the five page paper would know this. All they’d have to do is open the paper and look for ten graphs. When they found only six, they’d know there definitely weren’t 10 hockey sticks presented within the paper.

    While andthentheresphysics failed to cite the paper in question, I still managed to provide him a link to it. I gave him this link while telling him there weren’t 10 hockey sticks within it. Somehow, despite having been provided the link and needing only to scroll through five pages to see I was right, he managed to double down on his position. Rather than click on a link to the source he was discussing, he went to an entirely different source.

    It is dumbfounding. It is insane. How are you supposed to respond when someone fabricates things so blatantly then refuses to even look at the source they’re discussing?

  12. He is a very curious individual. He claims to be a research active scientist, yet shows none of the characteristics of a scientist. I followed his blog for a while with a kind of morbid fascination, reading post after post saying “I don’t know much about this subject but I read this and I think I agree with it”, and then gave up. When questioned on the details, he either has to go to a meeting, as here, or writes another paragraph of vague waffle and opinion. Well done on picking up on the “numerous easily explained issues” – that is a nice example. It reminds me a bit of Fermat’s comment about his proof of this last theorem. I wonder what was going through his mind when he wrote of the 10 hockeysticks in MM05. There aren’t 10 hockeysticks in either of the MM05 papers. Fig 2 of the GRL paper clearly shows the hockeystickness of the Mann algorithm for all 10,000 runs.

    There’s also the climategate emails of course. Wottsit probably hasn’t even read them. They show quite clearly that the climate science community knew that Mann’s work was junk and that the M&M criticism was correct-

    “There has been criticism by Macintyre of Mann’s sole reliance on RE, and I am now starting to believe the accusations.”

    “Of course he and other members of the MBH camp have a fundamental dislike for the very concept of the MWP, so I tend to view their evaluations as starting out from a somewhat biased perspective,”

    “I have just read the M&M stuff critcizing MBH. A lot of it seems valid to me. At the very least MBH is a very sloppy piece of work”

    “maybe someone(s) ought to have another look at Mann’s paper. His statistics were suspect as i remember”

    “I think, that “our” reaction on the errors found in Mike Mann’s work were not especially honest”

    Rob Wilson even did tests himself-
    “I first generated 1000 random time-series in Excel … The reconstructions clearly show a ‘hockey-stick’ trend. I guess this is precisely the phenomenon that Macintyre has been going on about.”

  13. I doesn’t seem hard to figure out where he’s coming from. The criticism he raises is basically that leveled against the Wegman Report. There were twelve hockey sticks shown in it, but two were picked via a different method so that’d be why he said 10. The confusion probably arose because the code used for the figure in the Wegman Report was written by Steve McIntyre. McIntyre just didn’t use it in any of his papers. He wrote the code, but he didn’t publish the results of it.

    A person who hadn’t read the paper could easily be misled by the fact the code for this was included with the code for the paper. It’s not hard to imagine how one could mistake that as indicating the code was used in the paper. However, anyone who even looked at the paper should know better. That, combined with the fact he failed (more than once) to provide an actual reference to his source, suggests he simply hasn’t bothered to look into things. He read what he took as criticisms of McIntyre and McKitrick, liked the conclusions and decided they were gospel truth.

    On a side note, I find it remarkable people like him keep claiming the selection process used in the Wegman Report is horrible. The Wegman Report never claimed it was providing a random sample. When showing an effect, it is understandable one might choose to pick strong examples of the effect rather than weak ones. That makes it clearer and easier for the audience to understand the effect. The worst you can fairly say is one should have clearly stated it wasn’t using a representative sample.

    Moreover, nobody seems to notice, or at least mind, that Nick Stokes’s post is highly misleading. Because of the selection process it used, the Wegman Report only showed positively oriented hockey sticks. That makes sense because the orientation of the calculated PCs is irrelevant. Upside down PCs just get flipped. For all intents and purposes, a negatively oriented PC is the same as a positively oriented one.

    Nick Stokes claims to show what happens when one doesn’t pick out strong hockey sticks, but the reality is he does more than that. He shows a random sample of positively and negatively oriented PCs. That creates a far stronger visual effect than if he showed a random sample of positively oritened PCs.

    If Stokes wanted to make an issue of the Wegman Report not showing negatively oriented PCs, he could have. Had he argued negatively oriented PCs should be shown, people would have known why he found such a large visual disparity. They would have known his display depends on saying it’s imperative to show negatively oriented PCs that will just be flipped in a later step.

    Instead, he didn’t even mention the issue. That left readers to believe the effect he found was entirely from not cherry-picking, something he knew to be untrue.

  14. I previously paid only superficial attention to the hockey stick issue, preferring to concentrate on my scientific strengths and not my statistical weaknesses with regard to climate science issues. However I identified strongly with andthentheresphysics’ statement that

    “People clearly interpret the results of MM05 as implying that random red noise typically produces hockey sticks, rather than random red noise sometimes (probably quite rarely) produces hockey sticks.”

    Possibly I’ve been reading too many blogs like andthentheresphysics.

    You wrote the above some time ago and I’m glad it has resurfaced now. Whether it was intended or not, your summary of the HS issue crystalizes the HS issue well. If MBH98 was based on erroneous methodology and later papers followed suit, can any hockey-stick-like millenian reconstruction be trusted without a thorough re-evaluation of the statistical methodology used?

  15. Chic Bowdrie, you quote Anders as saying:

    “People clearly interpret the results of MM05 as implying that random red noise typically produces hockey sticks, rather than random red noise sometimes (probably quite rarely) produces hockey sticks.”

    This quote is funny to me as I’ve followed the hockey stick debate since before these issues were first brought up. Steve McIntyre and Ross McKitrick were clear enough in their argument I had no trouble understanding it at the time. I was not confused as Anders says so many people are. I’ve always attributed this to one simple thing: I read what McIntyre and McKitrick had to say.

    The misleading comments about their argument mostly stemmed from their critics. Their critics had a tendency to misrepresent their arguments to create strawmen which they felt were easier to knock down. This caused a lot of confusion and caused quite a few people to misunderstand the argument. Beyond what Anders says, a lot of people came to believe the PCA issue was all that mattered even though McIntyre and McKitrick made it clear there was more to the topic than that.

    In other words, I believe the primary reason people were confused about McIntyre and McKitrick’s arguments is a lot of people didn’t read them, choosing to instead go off second-hand information which was inaccurate, and now Anders acts as though that is McIntyre’s fault.

    You wrote the above some time ago and I’m glad it has resurfaced now. Whether it was intended or not, your summary of the HS issue crystalizes the HS issue well. If MBH98 was based on erroneous methodology and later papers followed suit, can any hockey-stick-like millenian reconstruction be trusted without a thorough re-evaluation of the statistical methodology used?

    I’m glad to hear it helps! In case it isn’t clear though, I should point out something. Subsequent papers didn’t use the same methodology as MBH. They used methodologies which had a number of similarities to MBH’s,* but they all had some differences. Similarly, I don’t think any subsequent reconstruction used the exact same data as MBH, though many used some of the same key proxies. Amazingly, some even used NOAMER PC1, the key proxy created by MBH’s screwed-up implementation of PCA.

    Sadly, I’ve seen a couple papers which I think used methodologies which could avoid the problems of MBH and the like. This is sad because those papers all used data series which have been cherry-picked at one point or another. None used fair or representative samples.

    *Nearly every temperature reconstruction has been susceptible to the screening fallacy, or something very similar.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s