Talk:Hockey stick controversy

Ian Joliffe rejects Mann's "decentered PCA"

This is significant for two reasons. 1. Joliffe is a recognized expert on principal components analysis. 2. Joliffe has been misquoted by some of Mann's defenders who said Joliffe supported Mann's approach. Joliffe wrote up a nice rejection of decentering and asked it to be published both on Tamino's blog and on Climate Audit.[1] If I remember correctly, decentering was the cause of the Artificial Hockey Stick (the fact Mann's method would produce a hockey stick even if trendless red noise was used as data). RonCram (talk) 23:03, 8 September 2008 (UTC)

If our page said "Joliffe supports Mann" then that should definitely be removed. But it doesn't. It doesn't mention J at all. We don't seem to have considered his supposed support to be notable. You have misrepresented his comments: he has disclaimed *support* but saying he "rejects" M is wrong. So I don't see why he should be mentionned now William M. Connolley (talk) 20:45, 9 September 2008 (UTC)

Oh, and it isn't true that the method produces a HS from red noise (else why the ho-ha over the Bristlecones) William M. Connolley (talk) 20:48, 9 September 2008 (UTC)

William, not true. MBH had multiple problems. Von Storch and Zorita agreed with M&M regarding the Artificial Hockey Stick but said it didn't "matter." I emailed Zorita to ask why it didn't matter and he said it was because MBH had so many other problems the Artificial Hockey Stick was just not that important. One odd feature of the hockey stick produced by decentered PCA on red noise is the fact the hockey stick may angle down instead of up. RonCram (talk) 23:28, 12 September 2008 (UTC)

[responding to WMC] IMS, much of the substance of M&M's criticism of MBH was that Mann's "bent" PCA mines for "hockey sticks" from the multiproxy data they used, ie overweights the bristlecone data. This is nicely explained by McKitrick at [2]. — Preceding unsigned comment added by Tillman (talk • contribs)

Your link is broken. But if its by McK, its probably junk: McI did all the work William M. Connolley (talk) 21:38, 11 September 2008 (UTC)

Sorry -- try [3] . McK has a nice, clear writing style, which McI often doesn't have. --Pete Tillman (talk) 21:49, 11 September 2008 (UTC)

Thanks. Read it before, skimmed it again. First lies I noticed are around his fig 3 - see Description of the Medieval Warm Period and Little Ice Age in IPCC reports for something more reality based. Boreholes: more twaddle, but there is an underlying mystery: see http://scienceblogs.com/stoat/2007/03/the_borehole_mystery.php. Centering: I still don't believe McK, no matter how glibly he may write: http://www.wmconnolley.org.uk/sci/mbh/ William M. Connolley (talk) 22:29, 11 September 2008 (UTC)

And [4] has just come out, about the boreholes. Splendid timing. McI loses again William M. Connolley (talk) 20:55, 15 September 2008 (UTC)

Just wanted to weigh in. Regardless, this is an update, with a notable participant weighing in on the appropriate use of PCA other than standard. In case it becomes something to put up, I've whipped up a short bit on it. Needs much work of course.

Ian Jolliffe, an authority on Principal Components Analysis and author of one of the standard books on the subject was recently referenced as supporting the use of non-centered PCA in MBH98. His response was that "...there is a strong implication that I have endorsed ‘decentred PCA’. This is ‘just plain wrong’." Specifically about MBH98, Dr. Jolliffe said "...given that the data appear to be non-stationary, it’s arguable whether you should be using any type of PCA." and in conclusion "It is possible that there are good reasons for decentred PCA to be the technique of choice for some types of analyses and that it has some virtues that I have so far failed to grasp, but I remain sceptical."

---

To the supporting/rejecting argument, there is this quote:

http://www.secamlocal.ex.ac.uk/people/staff/itj201/RecentTalks.html It certainly does not endorse decentred PCA. Indeed I had not understood what MBH had done until a few months ago. Furthermore, the talk is distinctly cool about anything other than the usual column-centred version of PCA. It gives situations where uncentred or doubly-centred versions might conceivably be of use, but especially for uncentred analyses, these are fairly restricted special cases. It is said that for all these different centrings ‘it’s less clear what we are optimising and how to interpret the results’.

Sln3412 (talk) 18:06, 11 September 2008 (UTC)

Bias

75.30.129.205 (talk) 04:21, 26 October 2008 (UTC)Wow. Any Wiki article about environmental issues that gives any weight to Joe Barton is seriously biased. The entire article is presented from a skeptical viewpoint regarding the possibility of human impact on climate change, despite and overwhelming consensus amongst scientists. —Preceding unsigned comment added by 75.30.129.205 (talk • contribs)

The criteria for inclusion in Wikipedia verifiability, not truth. Barton is as far removed from reality as you can be. But we do not present his opinion as fact. We discuss the controversy. And please note that the hockey stick, while impressive, is not a major piece of evidence for anthropogenic global warming. Scientists understood the mechanism (more green house gases -> warmer) long before the paper was published. --Stephan Schulz (talk) 08:10, 26 October 2008 (UTC)

Ms. Schakowsky described Mann's seminal paper as:"....this false or inaccurate Dr. Mann study"

Why are you guys teaming up to prevent the Public from viewing the Democrat Congressional Member Schakowsy's considered opinion that the Mann paper was false or inaccurate? Unless you have another source, then these remain her actual words, and they were not "ironic" unless there is evidence to show so.MarkR1717 (talk) 23:44, 19 February 2009 (UTC)

I suggest you read the original source. There is plenty of evidence to show so. If you don't agree, that shows the problem with using primary sources. The consensus is obviously against you. --Stephan Schulz (talk) 23:48, 19 February 2009 (UTC)

If there is plenty of evidence of irony, then produce some, otherwise that is an empty claim.MarkR1717 (talk) 12:46, 20 February 2009 (UTC)

The burden of evidence lies with the editor who adds or restores material. At least three people disagree with you and recognize the irony. So far, not one supports you. --Stephan Schulz (talk) 22:19, 20 February 2009 (UTC)

No, you said the words were ironic, but cannot show any evidence. The burden is on you to establish irony.MarkR1717 (talk) 22:34, 20 February 2009 (UTC)

Mann et al. (1998) was neither false nor inaccurate; the United States Congress does not get to decide that, and particularly one member does not. It's now 10 years later, and Mann is still publishing the same story.[5] You can read that as either that he was basically right 10 years ago, or that there is a multi-national, scientific, Communist conspiracy intent on world domination which hinges on one made-up graph. Adding this quote to the article gives it undue weight. -Atmoz (talk) 22:55, 20 February 2009 (UTC)

I am citing a verifiable, authoritative source to give balance on controversial topic.MarkR1717 (talk) 23:17, 20 February 2009 (UTC)

No, what you are doing is cherry-picking. From the entire sentence[6] (without interuptions) its quite clear that she is paraphrasing Stearns. It's his question - not hers. As Stephan said above - if you can't tell the difference, then you've just discovered why primary sources are bad sources. --Kim D. Petersen (talk) 23:43, 20 February 2009 (UTC)

It's her statement, and she is characterising in her own words her verdict on Mann, "your question wanted to reinforce the notion that this was based on this false or inaccurate Dr. Mann study". So let's put the entire sentence in and then the readers can make up their minds.MarkR1717 (talk) 01:06, 21 February 2009 (UTC)

You as a group are obviously prepared to go to any lengths to prevent, without good reason, readers from seeing a fair and true comment made by a Congressional Member, whereby Ms. Schakowsky described Mann's seminal paper as: "....this false or inaccurate Dr. Mann study". You and your group are "ganging up" to prevent other points of view. You claim a majority. I say Wikipedia is not a Democracy. Therefor I will change the text again. If you revert it I will take it up the complaint chain, as it is clear you have abused the edit process.MarkR1717 (talk) 02:59, 21 February 2009 (UTC)

Re Stephan Shulz comment on reversion: "No again. Primary source, contentious, unreliable." Aren't you supposed to discuss this here before your repeated reversions. How is this unreliable? It is a quote from the Congressional record. It may be contentious to you, but it is a factually correct quote from an authoritative source. What is wrong with quoting from a Primary Source?MarkR1717 (talk) 16:30, 24 February 2009 (UTC)

She was clarifying the comment of another Congresscritter, as shown by reading the full quote. They say lots of things, almost all of them will not be suitable for inclusion into Wikipedia. Congress is not a reliable source for scientific information. -Atmoz (talk) 16:37, 24 February 2009 (UTC)

No, she was clarifying (wrongly as it turns out) that the Gore Graph was not based on the Mann Hockey Stick. Her comment about the nature of the Mann Hockey Stick stands alone, and was fully informed by the evidence before the Committee, which was specifically set up to examine the matter.MarkR1717 (talk) 19:37, 24 February 2009 (UTC)

I've said everything that needs saying above. Everyone but you (MarkR1717) apparently agrees, including the crew at WP:ANEW. If you don't get by now why this addition is not useful and very likely misleading, I cannot help you. --Stephan Schulz (talk) 16:43, 24 February 2009 (UTC)

OK. If you have no further reasoned objection, I will edit to include the quote.

Regulation of comments

Is it reasonable that discussion relating to this subject gets removed from the discussion page, even when it is clearly on topic, polite and accurate ? In particular for example, the comment from last night that William M. Connolley's statement "McI loses again" in reference to the borehole paper is quite plainly wrong. Figure 2 in [7] referenced by Connolley himself shows an extremely clear MWP (check it yourselves - there is no excuse for not doing so); something that was missing on the original hockey stick of MBH which instead had a long rather flattish shaft where the MWP should have been. The idea that one cannot point out even on the discussion page that this supports McI's criticisms of MBH is lunacy. Particularly when the claim is made that it explicitly contradicts McI and that claim is allowed to stay.

Look at figure 2 from the link, can you see a large bulge at 0.5 to 0.7 ? If the answer is yes, then it supports McI's claim that there was a MWP, something that was conspicuously absent from the hockey stick: [8]. Btw, the response that the borehole reconstructions do not have enough spatial coverage to be more than local climate variations and are therefore NOT proof of a MWP also implies that the statement "McI loses again" must be wrong. Clearly if it is only showing local climate variations it can't disprove McI's claim that MBH did not accurately represent the historic global temperature trend. 87.61.237.244 (talk) 16:21, 17 March 2009 (UTC)David

And here is the authors conclusion on fig. 2:

None of the reconstructions show MWP peak temperatures as high as late 20th century temperatures, consistent with the conclusions of both National Research Council [2006] and Intergovernmental Panel on Climate Change [2007] about the warmth of the MWP. The LIA temperature minimum shows an amplitude about 1.2 K below the MWP maximum, and about 1.7 K below present-day temperatures.

Which is shows us that its important to read and comprehend - rather than glance and assume. (notice the wording: consistent with) --Kim D. Petersen (talk) 18:23, 17 March 2009 (UTC)

Thank you for replying. I did read and (to a degree) comprehend. The point of contention was that MBH98 had NO noticable MWP, whereas the borehole study DID have a significant MWP. This reinforces McI's criticism that MBH98 inaccurately showed historic temperature variability. Being consistent with IPCC 2006/2007 is very different from being consistent with MBH98 - see http:/upwiki/wikipedia/commons/thumb/b/bb/1000_Year_Temperature_Comparison.png/300px-1000_Year_Temperature_Comparison.png - MWP temperatures are rather comparable with the reconstructed temperatures to the middle of the 1900's. 87.61.237.244 (talk) 00:33, 18 March 2009 (UTC)David

Sorry but your assertion is wrong. The Spaghetti graph (which originated in the NRC report, and is included in the IPCC) includes the MBH reconstruction. And as you can see in that one, all reconstructions are pretty much in agreement. Its btw. incorrect to state that the MBH graph doesn't have an MWP. Btw. you are making exactly the mistake that McI did earlier - its specified in the Borehole reconstruction, that you cannot compare it to the instrumental record. The claim that the current warming episode is outside the variations in the last 1K years - still holds, just as the claim that the MWP was colder than now. (just as a sidenote: in the Spaghetti graph, the MBH (blue) reconstruction of the MWP is actually one of the warmer ones) --Kim D. Petersen (talk) 01:23, 18 March 2009 (UTC)

I don't think you understand what I mean by MBH not having a MWP. The MWP is regarded as a time of significantly increased warmth. For there to be a time of significantly increased warmth, it must have on one side or another a reasonably large region of significantly cooler temperature. The MBH hockey stick is criticised so much not because of the uptick in recent times but because of the lack of temperature variability on the "shaft". Saying that it was "warm" in the MWP in MBH, and therefore it has a MWP is incorrect. If you look at the MBH98 paper, figure 5b, between 1400 and 1800 there is maximum temperature variation of 0.3 degrees. Is a 0.3 degree change in temperature indicative of a significant warm period and change in climate ? The HPS2008 borehole paper referenced above shows temperature drops of over 1 degree on either side of the MWP - that is a real MWP.

Your trouble comes from your interpretation of "significantly cooler"/"warmer". 0.4-0.8°C is significant on the global/nh scale. Thats in all of the reconstructions. The borehole data shows a bit more amplitude - and it has exactly the same fast warming in the 20th century. I think you miss McK's point which argues for a warmer MWP than 20th century - something which is entirely unsupported in all reconstructions. --Kim D. Petersen (talk) 23:42, 18 March 2009 (UTC)

I was aware the MBH graph was in the graph I linked to, anyone looking at the graphs can see that there is a MUCH larger degree of temperature variability than MBH shows for probably all of the other reconstructions. They all however roughly agree on the temperature around 1000 years ago as being only a couple of tenths of one degree below temperatures given by the proxy reconstructions for the mid/late 20th century.

Yes - the MBH also has the MWP a "couple of tenths below the mid-20th century. So whats your point. (not one of the reconstructions (including the borehole one) put the MWP much more than a couple of hundreds of a degree from MBH....). If i had to guess, the Moberg and the borehole reconstruction are pretty much the same for hi/lo extremes (albeit not reconstruction)--Kim D. Petersen (talk) 23:42, 18 March 2009 (UTC)

You accuse me of "making the same mistake as McI". I have no idea what you are talking about - I did not compare the instrumental record to the borehole reconstructions, I explicitly said "MWP temperatures are rather comparable with the reconstructed temperatures to the middle of the 1900's" (referring to the multi reconstructions image I had directly previously linked to). Additionally, unless you can point me to a reference I'm unaware of, McI has never said MWP was warmer than present. What he has done is point out, as a matter of analytical interest, that removing the dubious bristlecone and gaspe series from MBH's own calculations resulted in a reconstructed MWP higher than the reconstructed later 20th century temperatures (this was something MBH had done, but not reported). 93.161.57.174 (talk) 20:48, 18 March 2009 (UTC)David

Sorry - i did mean McK. But McI doesn't actually do a reconstruction - so the rest of your points are moot. And no reconstruction agrees with your interpretation of McI (that the MWP should have been warmer).

But i suggest that we stop discussing - because this section is already in breach of WP:TALK and WP:FORUM. --Kim D. Petersen (talk) 23:42, 18 March 2009 (UTC)

I checked the guidelines on editing others comments on talk pages, and have reinserted three portiojs of text. See bullet point 3 under editing other user's comments, which says inappropriate behaviour includes "Deleting material not relevant to improving the article". ^rossnixon 03:21, 18 March 2009 (UTC)

Actually that's the third bullet point under Some examples of appropriately editing others' comments: Short Brigade Harvester Boris (talk) 03:53, 18 March 2009 (UTC)

Oops, you are right! I misread. Sorry. I don't know if I agree with the policy, as it seems open to abuse/censorship. ^rossnixon 02:15, 19 March 2009 (UTC)

Side remark: Question is wether one should focus on the actual graph - what has been presented so far momentary is not much more than a diagram on a blog. Better to look on Pielkes sturdy pointing out that global temperature models and values are void and useless for regional risk assessments. A statement like ""McI loses again" is preposterous - McI has been essential and globally important in enhancing the role of blogging in scientific discussions, among the results among others - the likes of realclimate. I personally miss this aspect in the article more than any quick shot on the newest content on climateaudit --Polentario (talk) 22:28, 30 September 2009 (UTC)

43 coincidence?

Anyone noted that the section footnoted 43 is about a group of 43 scientists who have close ties to Mann? This just a coincidence, or a plot to destroy us all? —Preceding unsigned comment added by 203.202.43.53 (talk) 22:28, 14 July 2009 (UTC)

Yes. --Stephan Schulz (talk) 22:36, 14 July 2009 (UTC)

Explanation of Tony Sidaway's revert of an edit by 70.234.173.145

Somebody editing from IP number 70.234.173.145 added the following statement to the lead:

The hockey stick controversy is a typical example of cargo cult science where unverifiable results are used in lieu of experimental validation to support a deceptive presentation of data under the appearance of scientific objectivity.

The source given was Lying With Statistics, an article on the website of the Physics Department of Southern Methodist University. The article itself introduces the basic principles of how to recognise the misuse of statistics as presented in How to Lie with Statistics, the famous 1954 book by Darrell Huff. As such it doesn't discuss the Hockey Stick graph, and the most charitable interpretation I can find for this edit is that the statement above is 70.234.173.145's own personal opinion. In other words, the edit is Wikipedia:Original research.

I've reverted it as such. --TS 20:21, 30 September 2009 (UTC)

"Hockey Stick" now proven not to exist

Just passing along some news for those watching this article. I'm actually surprised it's not already in the article, as it's causing some waves in the media.

It's already been mentioned in the article that the original authors of the Hockey Stick were withholding the data used to generate their graph. Since this data has been released at last, the graph has been recreated with a fuller, less-biased set of data. It appears to be definitive proof that the results of the "hockey stick" are incorrect, using the broadest tree ring data available. The methodology is well documented in lots of technical detail. Here is a link to the report: http://www.climateaudit.org/?p=7168

Original Research Alert: I personally would go so far to say that the resulting graph seems to demonstrate that tree ring temperature extrapolation looks like a noisy signal. Here is some background on why using tree ring data is problematic.

Cnadolski (talk) 15:42, 2 October 2009 (UTC)

You want [9]. Or for a less serious view [10]. Or Talk:Global_warming#End_of_the_Hockey_stick (no, don't go there). Or... William M. Connolley (talk) 15:48, 2 October 2009 (UTC)

Denial Depot looks like a WUWT parody. The RealClimate post was mostly snark and sarcasm, along with comparisons of the tree ring regression to unrelated charts such as borehole temperature regressions. The only rational rebuttal was from Briffa's response, though I would have liked a more thorough explanation why Briffa's choice was superior to McIntyre's choice aimed at people who are not dendrochronologists. The lack of respect and rational debate is what drives me further and further from the AGW camp to the skeptical camp.