For those who don’t know, Ross McKitrick of the University of Guelph co-authored the first paper with Steve McIntyre debunking Michael Mann’s first Hockey Stick paper, MBH98. Ross wrote this essay in today’s Financial Post, excerpts are below. Please visit the story in that context here and patronize their advertisers. – Anthony
Flawed climate data
Only by playing with data can scientists come up with the infamous ‘hockey stick’ graph of global warming
Ross McKitrick, Financial Post
Friday, October 2, 2009
Beginning in 2003, I worked with Stephen McIntyre to replicate a famous result in paleoclimatology known as the Hockey Stick graph. Developed by a U.S. climatologist named Michael Mann, it was a statistical compilation of tree ring data supposedly proving that air temperatures had been stable for 900 years, then soared off the charts in the 20th century. Prior to the publication of the Hockey Stick, scientists had held that the medieval-era was warmer than the present, making the scale of 20th century global warming seem relatively unimportant. The dramatic revision to this view occasioned by the Hockey Stick’s publication made it the poster child of the global warming movement. It was featured prominently in a 2001 report of the U.N. Intergovernmental Panel on Climate Change (IPCC), as well as government websites and countless review reports.
Steve and I showed that the mathematics behind the Mann Hockey Stick were badly flawed, such that its shape was determined by suspect bristlecone tree ring data. Controversies quickly piled up: Two expert panels involving the U.S. National Academy of Sciences were asked to investigate, the U.S. Congress held a hearing, and the media followed the story around the world.
The expert reports upheld all of our criticisms of the Mann Hockey Stick, both of the mathematics and of its reliance on flawed bristlecone pine data.
…
Most of the proxy data does not show anything unusual about the 20th century. But two data series have reappeared over and over that do have a hockey stick shape. One was the flawed bristlecone data that the National Academy of Sciences panel said should not be used, so the studies using it can be set aside. The second was a tree ring curve from the Yamal Peninsula in Siberia, compiled by UK scientist Keith Briffa.
…
But an even more disquieting discovery soon came to light. Steve searched a paleoclimate data archive to see if there were other tree ring cores from at or near the Yamal site that could have been used to increase the sample size. He quickly found a large set of 34 up-to-date core samples, taken from living trees in Yamal by none other than Schweingruber himself!Had these been added to Briffa’s small group the 20th century would simply be flat. It would appear completely unexceptional compared to the rest of the millennium.
Combining data from different samples would not have been an unusual step. Briffa added data from another Schweingruber site to a different composite, from the Taimyr Peninsula. The additional data were gathered more than 400 km away from the primary site. And in that case the primary site had three or four times as many cores to begin with as the Yamal site. Why did he not fill out the Yamal data with the readily-available data from his own coauthor? Why did Briffa seek out additional data for the already well-represented Taimyr site and not for the inadequate Yamal site?
Thus the key ingredient in most of the studies that have been invoked to support the Hockey Stick, namely the Briffa Yamal series, depends on the influence of a woefully thin subsample of trees and the exclusion of readily-available data for the same area. Whatever is going on here, it is not science.
Read the complete story at the Financial Post
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
RE: Joel Shore (09:22:21) : It seems to me that Ross McKitrick’s reading of the National Academy of Sciences report on temperature reconstructions is somewhat selective.
Joel Shore, the question is whether the temperatures in the late 20th century are higher than those in the last 1000 years or more.
Please note that the NAS report you quoted found Mann’s “hockey stick” claim to be merely “plausible.” Not proven beyond a reasonable doubt, nor proven to a reasonable certainty, nor proven to be probable (that is, more likely true than not).
Before deciding that someone else is being “selective,” read each word and try really hard to accept that the words you quote may not support your faith.
Eliminating the “Medieval Warm Period” has been an obvious goal of the people who have come up with the “hockey sticks.” The reason is just as obvious–they want to be able to make the claims that they have been making, i.e., the recent warming is unprecedented and therefore man, not natural variation, is responsible for it.
So far, they have claimed (as you accept on faith) to have proven that recent temperatures exceed the MWP, but their claims have not risen above the “plausible” level among people who rely on science rather than faith.
I was under the impression that Anthony wanted this site to be taken seriously. Of course he owns it and can do as he likes but to me it seems a pity.
“What does Monbiot’s face look like right now? It’s bad enough at other times he looks like his raging buddies Galloway, Chavez and Saddam (now a sex slave for Satan)”
The above quote is a previous post in its entirety. I predict that if that sort of trash is allowed to remain
1) The number of real scientist posting here will fall, perhaps precipitously.
2) This site will lose all credibility.
This is a very sad day for me.
RE: Jason S (08:34:58) : Real Climate’s response to the claim that we have been waiting 10 years from Briffa’s data:
“[Response: The russ035w data has been ‘lying around’ on the web since 2001….”
Jason S, is that really Real Climate’s response to the criticism of Briffa for refusing to reveal his Yamal data? If so, it seems really lame–the “russ035w data” is not Briffa’s data, it is the data used by McIntyre in his sensitivity test of Briffa’s analysis. It is the nearby tree core data that McIntyre substituted for Briffa’s recently revealed data.
So, who cares that it has been “lying around” for years? It isn’t the data requested from Briffa which Briffa refused for years to disclose.
McKitrik claims:
Two expert panels involving the U.S. National Academy of Sciences were asked to investigate, the U.S. Congress held a hearing, and the media followed the story around the world. The expert reports upheld all of our criticisms of the Mann Hockey Stick, both of the mathematics and of its reliance on flawed bristlecone pine data.
We all know that this is not a true statement. At the very best it might be considered “very misleading.”
That is why I posted the link of Wahl and Ammann
http://en.wikipedia.org/wiki/Hockey_stick_controversy
Why do I need to comment on Bishop Hill?
We should be grateful that Ross has outlined the situation so well.
After a few days people tend to lose the big picture and argue about details, motives, and personalities. It leads to a rather useless “he said, she said” quagmire.
Ross took us back to the issue.
Briffa will have more to say about this and he may have a good explanation. That looks doubtful at the moment but far stranger things have happened.
My guess is that Briffa erred in method and just didn’t realize the weakness of his choices. Sometime after publication he saw the problems but tried the Move On stall and hoped the matter would eventually go away without consequences.
Making mistakes is not that big a deal in itself. It is embarassing, perhaps very much so. OTOH deliberately selecting data which supports a desired conclusion is another matter.
So I wish him the best of health and expect he will reply as promised.
Too bad this audit couldn’t have been done years ago. We have all gotten a look at how poor peer review can be and how lax publications often are about enforcing their own standards.
The article is also in the National Post
http://network.nationalpost.com/np/blogs/fullcomment/archive/2009/10/01/ross-mckitrick-defects-in-key-climate-data-are-uncovered.aspx
Joel Shore
“the GCMs do not incorporate the hockey stick into them in any way”
My question what more to do with how they have calibrated climate sensitivity parameters. Has the hockey stick had any part to play in this?
“We have a variety of different sources (including, but not limited to the instrumental temperature record) for the blade.”
It needs a contiguous data series (same data and methods for both shaft and blade) to argue that the shaft and blade both exist. Is that not what Yamal was trying to do? You refer to a variety of different sources for the blade and that just says something isn’t right about the shaft and/or the blade.
“The question regarding the blade is really a question about how well these temperature proxies are good proxies for temperature”
Quite – if the proxies fail to follow recent purported rise of temperature, we need to be ready to conclude that proxies are bad measurements.
“the GCM’s climate sensitivity comes out of the physics that goes into them.”
Is that limited to the S-B response, or does it rely on amplification by positive feedback. I find the amplification argument extremely unconvincing.
“determined by empirical data”
Back to my question – have the hockey sticks been part of that empirical data?
PR Guy (10:12:06) :
RC’s post shows up in the Google News feed along side Ross’s article. Yet another indicator that RC has some serious PR flacks behind the scenes (presumably Fenton Communications) who know how to get a simple blog post to show up as a news story of equal weight to a Financial Post article.
Those of you who get frustrated by RC are under the mistaken impression that its a science blog. In fact, it’s a professional, highly sophisticated PR production.
You repeatedly assert this. I do not doubt you, but do you have any evidence of this, or is it just gut feel (since you are by your handle in the same game)?
Scott A. Mandia (14:40:27) :
Glad you have so much faith in Wikipedia. Since we’re using web sites, I guess the uninitiated can go here too. I don’t guess the Wiki site quoted what North said under oath.
http://www.climateaudit.org/?p=2322
And the Bishop Hill summary really is useful for the truly intellectually curious as well, despite your lack of interest in it.
BTW, JohnM (08:50:26) is not me.
If Steve McIntyre’s analysis fails to show the signal of recent warming that we know about from many other kinds of data, that rather suggests that he’s doing something wrong, doesn’t it? How can you draw valid conclusions from an analysis that doesn’t reflect real-world observations?
Scott Mandia
I can’t believe you could have posted a link to the HS controversy on Wiki and kept a straight face. I suggest you look at the history and discussion behind the article and recognise that:
1) wIki editorial policy requires information to be ‘verifiable’ not necessarily correct.
2) That the gatekeeper of this section is an associate of Michael Mann and has represented Real Climate at conferences. This is hardly an objective summary.
tonyb
Well, b’golly, the Wiki page does take a strange sort of shot at being balanced. Must have happened while Connelly was rowing his boat.
Seems the good doctor presents a different picture in the press than he does when under oath. And yet, he clings to the “could be right even with the wrong method” premise.
Or maybe it means treerings aren’t thermometers.
Kuhnkat, Jeff Id has done a great job of listing all those faux hockeysticks in the RC museum showcase.
RE: Icarus (15:57:24) :
If Steve McIntyre’s analysis fails to show the signal of recent warming that we know about….
What signal do you know about, and how would it be displayed?
The point in McIntyre’s analysis is to show that Briffa’s analysis depends for its “blade” on the its “hockey stick” entirely on a very small set of data from the Yamal peninsula.
The “hockey sticks” are attempts to put the temperature records of roughly the past 150 years into context with the past 1000 (or more) years. The more recent history includes temperature measurements, but the more distant past has to be estimated based on temperature proxies. Tree rings are offered as proxies for temperatures in that more distant past.
You apparently believe that the more recent past should be displayed as a steeply rising line on the graph compared to the more distant past.
The hockey stick makers apparently have the same belief.
Ironically, many people wonder if the hockey stick makers’ belief has led them to construct biased analyses that display something which is not correct; and your apparent belief is that any analysis which doesn’t show the “blade” on the “hockey stick” must be mistaken.
Could it be that, compared to the past 1000 years or so, the graph should not have a “blade,” but should instead show that our more recent temperatures are no higher than, and perhaps lower than, the Medieval Warm Period?
That is, after all, the question–whether recent temperatures are “unprecedented” and thus perhaps manmade. You assume the answer you want and disparage any analysis that doesn’t agree with your assumption.
John M (16:30:21) : …maybe it means treerings aren’t thermometers.
I think that would contradict what we know about biology – no-one actually disputes that trees will *tend* to grow more when there is warmer weather and a longer growing season, all other things being equal. A careful and thoughtful analysis of tree ring data would have to include ways of isolating the effects of climate change from other influences… agreed? If McIntyre’s analysis isn’t able to do this, perhaps this suggests it is not a useful contribution to the science.
To those saying McKitricks representation of Wegmann is misleading.
I suggest read the entire report (ie lets let the data do the talking):
In general, we found MBH98 and MBH99 to be somewhat obscure and incomplete and
the criticisms of MM03/05a/05b to be valid and compelling. We also comment that they
were attempting to draw attention to the discrepancies in MBH98 and MBH99, and not to
do paleoclimatic temperature reconstruction. Normally, one would try to select a
calibration dataset that is representative of the entire dataset. The 1902-1995 data is not
fully appropriate for calibration and leads to a misuse in principal component analysis.
However, the reasons for setting 1902-1995 as the calibration point presented in the
narrative of MBH98 sounds reasonable, and the error may be easily overlooked by
someone not trained in statistical methodology. We note that there is no evidence that Dr.
Mann or any of the other authors in paleoclimatology studies have had significant
interactions with mainstream statisticians.
The quotes you make do not supports Manns paper or methods but indicate that other studies have provided evidence of warming.
The evidence of warming can however only be confidantlt asserted in the last 400 years as the earth went tinto the little ice age and later came out of it.
Micajah says:
Please don’t put words in my mouth. I have never claimed that anything was proven. (And, in fact, I have previously noted that the notion of “proof” is reserved for a deductive enterprise like mathematics; science is necessarily inductive and can never prove anything.) In the case of the temperature reconstructions, the point that the NAS made, and which I tend to agree with, is that the available evidence seems to point mainly in the direction that the late 20th century was warmer than any comparable period during the MWP; however, given the considerable uncertainties in these temperature proxies and the difficulty in even quantifying the uncertainties in the reconstructions using them, it is not possible to put a likelihood estimate on that conclusion.
Of course, since that NAS report was reached, some additional work has been done, such as Mann et al. (2008) that seem to provide further support for this conclusion. But, obviously, considerable uncertainties still remain.
John M:
It is not at all unusual in science for the first pioneering work in a field to be done using methods that are later found to be problematic in certain ways and thus superceded by better methods. It is not all black-and-white.
[thanks but since you made the first ever copyrighted comment here, I’m forced to delete it]
©2009 Dave Stephens
Further information and links to the reports can be found here
http://www.climateaudit.org/?p=2322
Jordan says:
No. The sensitivity comes out of the models based on the physics that is put into them (some, such as that involving clouds, that does have to be parametrized). The hockey stick played no role in this.
Well, the evidence for the water vapor feedback operating about as expected is quite strong (see http://www.sciencemag.org/cgi/content/summary/sci;323/5917/1020 ). And, it is hard to argue against the ice-albedo feedback being positive. The only feedback that seems like it could possibly save you from having amplification is to have a significantly negative cloud feedback.
However, besides the fact that none of the various independent modeling groups who have built their models using different sorts of cloud parametrizations find this to be the case (and that the climateprediction.net simulations couldn’t even create them by varying parameters within plausible ranges), it becomes very difficult to explain past climate events, such as the glacial – interglacial cycles with a small climate sensitivity.
No. I think there may have been a study or two that tried to look at what the reconstruction had to say about the climate sensitivity…and I think one actually concluded that a reconstruction with somewhat more temperature variability (such as that of Moberg et al.) was more compatible with estimated climate sensitivity ranges. However, such studies are hampered by the fact that I don’t think the estimates of forcings from solar and volcanic activity are really known well enough over that time period to really have that much confidence in such a conclusion.
And, at any rate, I don’t think these studies had any effect on either the estimates of climate sensitivity from empirical observations or estimates of climate sensitivity produced by the GCMs.
“Or maybe it means treerings aren’t thermometers.”
I was going to say the same thing. Actually, Steve has been talking about the “divergence problem” for years now. “Divergence” is just a fancy way of saying that maybe trees aren’t such good thermometers.
Anyway, I think Icarus’ question touches on a subtle fallacy which is common among warmists.
I think the assumption in Icarus’ thinking is that this is some kind of competition to see who is better at predicting future temperatures or estimating past temperatures. But skeptics (and deniers) don’t have to estimate or predict anything. We need merely point out that the warmists are not as good as they claim (or pretend) to be.
But the mainstream is… warming to the cool – George Will yesterday:
http://www.washingtonpost.com/wp-dyn/content/article/2009/09/30/AR2009093003569.html?sub=AR
He may even be the champion to follow this story to publication.
Will’s call for a national commission to examine the evidence is something we skeptics would like now. It’s something politicians might well seize upon in order to get rid of a hot potato and give them “cover” for a reversal of their position. And it will be accepted by warmies, if cap-and-trade gets stalled and they can see no other way to dislodge the obstructionists. Our side should echo Will’s proposal.
Icarus says:
Are you talking about the Team’s analyses? There seems to have been no warming in the last 10 years …
Glad you have so much faith in Wikipedia. Since we’re using web sites, I guess the uninitiated can go here too. I don’t guess the Wiki site quoted what North said under oath.
http://www.climateaudit.org/?p=2322
Hit page-down 11 times to get to the quote from North.