The tale of the hockey stick

Or as an alternate title: “Why we find it difficult to trust certain climate scientists.”

This posting by Bishop Hill, telling the tale of the nefarious temperature reconstruction known as the Michael Mann hockey stick, from start to present, is an excellent summation for the layman reader struggling to understand the entire affair and why it is such an amazing pox on the conduct of science and practice of peer review. This sums it up quite well:

That the statistical foundations on which they had built this paleoclimate castle were a swamp of misrepresentation, deceit and malfeasance was, to Wahl and Amman, an irrelevance.  

I highly recommend reading it, and Bishop Hill deserves thanks for condensing this affair into a readable story.

Get notified when a new post is published.
Subscribe today!
0 0 votes
Article Rating
158 Comments
Inline Feedbacks
View all comments
Christopher Hanley
August 14, 2008 3:19 am

I love all the squirming by the AGW crowd.
Look, as a layman, the ‘Hockey Stick’ diagram superimposed on the rise of human induced ‘greenhouse’ gases was sold as cast-iron proof of the validity of the AGW hypothesis, not withstanding the flawed logic:
http://maybeitsnotwarming.com/wp-content/uploads/2008/02/hockey-stick-temperature.jpg
Without the ‘Hockey Stick’, the AGW hypothesis is just that, an interesting hypothesis which under no circumstances, should be considered a basis of public policy.

trevor
August 14, 2008 3:27 am

Err Steven Talbot. Just a lay observer, but I notice that neither Hadley CRU or GISS is exactly forthcoming with detailed information about the make up of their GMT series, the stations they use or don’t use, the adjustments that they make etc. Au contraire: They hide, obfuscate, confuse, obstruct, waffle.
They say UHI is not a big issue. They say that closure of many temperature stations in Siberia is not an issue. They say that there is a need to continually adjust past temperature records. However, how can you blame us amateurs when all the adjustments (barely explained if at all) are all in the direction that best exaggerates the GMT warming. Just have a look at how much of the claimed warming last century would disappear if the adjustments were reversed. As one of my science teachers said in 1962: “It is in its nature so to do”. Or more accurately in this context, “it is in its nature NOT to so do”.
Seems to me that CRU is just as bad as GISS, and comparing one with the other doesn’t get us there.
You can waffle on. I can waffle on. What we need is facts, honestly presented. I have to say that the ‘real climate scientist’ have dramatically lost cred in the eyes of all but the most committed AGW proponents.

MarkW
August 14, 2008 4:13 am

Steven,
All those other topics have been discussed on this site. Extensively.
This topic is about Hansen and the misuse of data.
Why should we drag in every subject under the sun, in every thread, regardless of topic, just to keep you happy?

JP
August 14, 2008 5:06 am

“Warming Oceans – This entire allegation stems from a single paper published by Lyman et al. two or so years ago. But guess what? The paper had huge sources of bias, and the authors retracted and subsequently submitted a revised paper which has seriously less of a cooling signal. Furthermore, the actual observation is that of a warming ocean, not a cooling one, so the point is moot.”
Counters,
The JPL and Layman issued a new paper this year using the improved Argo network. From 2003-2007, again they found no warming in global SSTs; as a matter of fact, the detected a slight cooling.

JP
August 14, 2008 5:10 am

“Climate Models – Model’s don’t “predict.” Models are used to establish a statistical measure of where we expect the climate to trend. Experiments involving models don’t predict what the weather or climate will be 100 years from now; the develop trends”
They why do researchers like Gavin Schmidt consistently use model representations instead of raw data? Why can’t the models detect with any precision changes in things like the Walker Cell and ENO? How can models predict what the climate will be in 50 years if they cannot model basic atmospheric circulations with any accuracy? Both Hadley and Hansen predicted a Super El Nino to occur in 2007 -the exact opposite occured.

Pofarmer
August 14, 2008 5:16 am

So, where’s the evidence of the supposed human bias? That, surely, is the first question that any genuine sceptic would ask…..
Part of it may be the fact that they use different reference periods for their anomolies.

Roger Carr
August 14, 2008 5:37 am

Pamela Gray (13:50:00) : “We-think” mentality can be researched and even experimentally demonstrated…
Have you read this, Pamela? It fits well with your “we-think”; and the informational cascades seem to be the same thing.
Diet and Fat: A Severe Case of Mistaken Consensus – JOHN TIERNEY
http://www.nytimes.com/2007/10/09/science/09tier.html?_r=1&ei=5087&em=&en=53bfd0df7f448ca4&ex=1192161600&pagewanted=all&oref=slogin

August 14, 2008 6:05 am

Something is missing in the debate.
We know that the hockey stick graph is erroneous and not a good representative of the global climate. MWP and LIA are both missing and the 20 th century temperature increase is not diverging significant from that of any previous centuries.
But the question is seldom asked here if temperature reconstruction from tree rings and corals give good representations of temperature variations.
Well, it might. That would be the case if temperature were the only parameter that changed during the investigated period,
But, that is not the case.
In the case of Mann’s hockey stick, things get worse because of its heavy reliance on bristlecone pine tress which grows in dry regions in California and Utah.
So in my view if you collect data if things that grow which are sensitive to changes of the CO2 level, those results are not representative of temperature variations.
In the case if the pine trees, the growth have increased during the last century because of the fertilization effect from elevated levels of CO2 and because of better drought resistance cased by reduced water vapor loss This is because changes in the “stomata”, this is the openings in plants that are used for photosynthesis. They become more closed with higher CO2 levels and the plants don’t need to use so much water to absorb CO2.
Both these effects are well documented and have now led to more greening of the planet, especially in drier regions.
So if you normalize the tree ring growth relation to the temperature, by using temperature changes from the last century, but the tree ring growth have more to do with changes in CO2 levels than with temperatures, you are bound to get wrong relations for tree ring growth and temperature.
Add to that, that the temperature values used for this normalization to some extend are contaminated from the Urban Heat Island Effect.
This is what they have done. What they get is a hockey stick graph, with a high temperature increase at the end of the graph and an almost flat graph for the time before 1900.
You don’t need statistical mismanagement to create a faulty “hockey stick graph”. Just use CO2 sensitive biological proxy data and match it up with the temperature from the last 100 years and you are bound to get a hockey stick graph.
Am I missing something here?
Have Mann or any of the other AGW people tried to compensate for the fertilization effect and from the improved drought resistance from elevated CO2 levels?
I’ve never see that!
In my view it is more constructive to criticize the questionable biological proxy than to look at the questionable statistical data handling of the data to debunk the hockey stick graph.

August 14, 2008 6:40 am

Bob Tisdale – You should finish reading to the conclusion of the link, where you’ll be referenced to this paper.
SpecialEd – Well, I’d bet his models were wrong too. They’re two freaking decades old! Do you have any idea of the leaps and bounds made in modeling techniques and what we actual couple in the models nowadays? Hell, just considering the computational power available at the time is enough to laugh at this insinuation. For giggles, though, Wikipedia has the ETA10 as the fastest supercomputer of the late 1980’s; it had a top throughput of 10 giga-flops. By comparison, the IBM Roadrunner of today has a top throughput of… 1 peta-flop. Modeling of two decades ago far lagged in computational power (obviously), as well as sophistication of the techniques being used to build the model. Today’s models can incorporate far more than the surface of teh ocean and the atmosphere; they have many vertical levels to resolve, and they couple many geological, photo-chemical, and bio-chemical reactions to further resolve the system.
I’d respond to more but I simply don’t have the time this morning. I’ll try to get back on a bit later.

Pofarmer
August 14, 2008 7:18 am

cooling troposphere discussed on ClimateAudit here
http://www.climateaudit.org/?p=3048

Pofarmer
August 14, 2008 7:29 am

Have Mann or any of the other AGW people tried to compensate for the fertilization effect and from the improved drought resistance from elevated CO2 levels?
Obviously you missed the memo. CO2 is a pollutant, not a fertilizer.
For giggles, though, Wikipedia has the ETA10 as the fastest supercomputer of the late 1980’s; it had a top throughput of 10 giga-flops. By comparison, the IBM Roadrunner of today has a top throughput of… 1 peta-flop. Modeling of two decades ago far lagged in computational power
Well, if the assumptions are wrong, your just gonna get wrong results faster.

DAV
August 14, 2008 7:30 am

counters (17:27:07) :Climate Models – Model’s don’t “predict.” Models are used to establish a statistical measure of where we expect the climate to trend
ummm… “establishing a statistical measure” to get a trend IS a prediction. IOW: expected trend = prediction.
Hockey Stick – It still hasn’t been shown to be “debunked.” Debunked implies it is incorrect; …There are many independent paleo-climate reconstructions used to corroborate the “big picture” of the past.
and later: Too much emphasis is being placed on the Hockey Stick. Please realize, there are many other paleo-climate reconstructions out there
The Hockey Stick was shown to be a mathematically flawed to the point it was worthless. I call that incorrect. Are you trying to say: “got the right answer but for the wrong reasons?”
I hope you aren’t including Thompson’s ice core chart that appears (a tad doctored) in AIT. Mann’s Hockey Stick is shown on the same page in Thompson’s paper right along side Thompson’s graph to show the similarity. Never mind that O18 measurements are better precipitation analogs instead of temperature. (BTW: precipitation may be the very thing that drives the bristle-cone data Mann so heavily weighted). What other paleoclimate reconstructions did you have in mind?
The problem the AGW folk are having is that there is no definitive proof of AGW. The Hockey Stick was the closest thing to proof. If it was not then why was it trumpeted so far and wide without proper vetting? If the HS has been vindicated by subsequent research why is that research not being trumpeted as the HS was? Maybe, just maybe, because it isn’t all it’s been made to be?
All of the rest is obfuscation of the central issue: Anthropomorphic Global Warming. That is being accomplished through the deliberate confusion of GW, which is undoubtedly occurring (or at least did from the 70’s-90’s), and AGW , of which there is no compelling proof.
Even if we were to accept without question all of your points, they do nothing to address the issue of AGW.

An Inquirer
August 14, 2008 8:14 am

Well, Counters (and Steve Talbort), one can get the impression that a multitude of emotionally-charged individuals are ganging up on you. Therefore, I hesitate to post my thoughts, but there are a couple of issues which I have not seen yet. I tend to be overly generous in assuming sincerity of others, and I do not call Hansen dishonest. However, I would say this – he acts like I would expect a dishonest person to act. Nevertheless, perhaps a more likely explanation is that he liked the results that his procedures produced and did not scrutinize the details for biases or mistakes. Moreover, he has acted in ways inconsistent with my perceptions of academic and scientific integrity. Only in climatology would it be acceptable for a forecaster be allowed to validate his model results with his own estimate of observed values – especially when his estimates of observed values are not reproducible nor verifiable. (Yes, Steve Talbort, in one sense, his estimates of observed values are verified by similarity to HadCrut estimate values, but HadCrut also has issues with UHI which apparently have been examined even less than in GISS. And we can get into comparison with satellite data which would make this post excessively long.) It appears that Hansen gets pretty much a free pass in the main stream media (MSM) because the MSM is dominated by people whose ideology line up with Hansen’s. Otherwise, it is not conceivable that he could get away with what he gets away with.
The other issue which I have not seen well addressed is the number of studies (there are several dozens) which show the hockey stick to be a flawed representation of temperatures experienced over the past several hundred years. For starters, look at http://www.co2science.org/data/mwp/mwpp.php.

Jeff Norman
August 14, 2008 8:16 am

Pieter Folkens (09:11:37) :
When I first met up with the Mann Hockey stick I was made skeptical by the error bars presented with the trend. I knew I could not sustain that level of precision over a ten year period using the thermometers installed to measure steam temperatures in my turbines (*). I was amazed by their claims for 600/1000 years.
I assumed they had used some statistical methodolgy unknown to me so I read the paper in hopes of learning something. The paper was less than illuminating. I recall expressing these doubts to John Daly. He agreed.
I am just and engineer, so what do I know. I thought all those serious scientists would examine the MBH claims and correct any misconceptions. Not one climate scientist publicly challenged the MBH conclusions.
This at the very least is a sin of omission. At the very worst it is a criminal conspiracy. Somewhere in between lies the truth. This makes the entire climate science community suspect. Steve McIntyre’s experiences only reinforces this perception and makes them appear to be at the very least foolish.
(*) calibration drift, no reference temperature, broken thermometers, new thermometers, thermocouples, fouled thermowells, etc.

BraudRP
August 14, 2008 9:23 am

Have Mann or any of the other AGW people tried to compensate for the fertilization effect…
As I understand that issue, the AGW people are in denial that the effect is other than minimal.

Ken G
August 14, 2008 9:44 am

Counters,
Hockey Stick – It still hasn’t been shown to be “debunked.”
From the Wegman committee report:
“Overall, our committee believes that the MBH99
assessment that the decade of the 1990s was the likely the hottest
decade of the millennium and that 1998 was likely the hottest year of
the millennium cannot be supported by their analysis.”
http://energycommerce.house.gov/reparchives/108/Hearings/07192006hearing1987/Wegman.pdf
“There are many independent paleo-climate reconstructions used to corroborate the “big picture” of the past.”
Such as?
“Warming Troposphere – Just because a dude writes an op-ed claiming it’s not there doesn’t mean that’s the case.”
I don’t know what op-ed you’re referring to, but I can’t help but notice you countered with an mere article of your own. There have been actual published studies on this, such as:
“A comparison of tropical temperature trends with model
predictions”
ABSTRACT: We examine tropospheric temperature trends of 67 runs from 22 ‘Climate of the 20th Century’ model
simulations and try to reconcile them with the best available updated observations (in the tropics during the satellite era).
Model results and observed temperature trends are in disagreement in most of the tropical troposphere, being separated by
more than twice the uncertainty of the model mean. In layers near 5 km, the modelled trend is 100 to 300% higher than
observed, and, above 8 km, modelled and observed trends have opposite signs. These conclusions contrast strongly with
those of recent publications based on essentially the same data.
http://icecap.us/images/uploads/DOUGLASPAPER.pdf
“Warming Oceans – This entire allegation stems from a single paper published by Lyman et al. two or so years ago…. Furthermore, the actual observation is that of a warming ocean, not a cooling one, so the point is moot.”
Can you point me to the “actual observation” that says the oceans are warming? According to the Argo floats network, that simply isn’t true.
“GISS – It’s not under “serious scrutiny.” I’m sorry, but a couple of blogs haranguing Hansen does not equate with legitimate investigation.”
That’s how the debunking of the hockey got started too. It is unfortunate that it takes an independent third party to point out serious errors that make it past “peer review”. With the blatant lack of oversight, it’s got to start somewhere.
We understand the physics” – The thread underneath this is about solar cycles. No skeptic on the blogosphere has been able to demonstrate where the physics behind AGW is incorrect. Period.
Well the complete lack of the key signal as shown above, as necessary for a ghg forcing scenario, sure tells us something is off. Of course, nobody has been able to show it is correct either, so you’re point is rather moot.
Climate Models – “there is a great deal of misinformation about what climate models are, what they do, how they work, and what they’re used for.”
No kidding.

Russ R.
August 14, 2008 9:51 am

Here’s a nice little data point from an urban setting, that is perilessly close to the tipping point, and should be baking under UHI and increased GHG’s, yet has somehow managed to keep it’s cool. I will vouch for the varacity of this source. I have not had my AC on all summer, and I was only tempted a few times due to the humidity.
http://www.chicagotribune.com/news/local/chi-tom-skilling-explainer-13aug13,0,918946.story
Those responsible for this “hockey stick” abomination should know better. They have had plenty of warning that this was junkscience of the worst kind.
It passed the threshold for “mistake” long ago. This sordid example says more about human nature, than it does about the future climate.

J. Peden
August 14, 2008 11:32 am

(As for the second charge, just for now, for those who’ve picked up the idea that the IPCC has ‘dropped the hockey stick’, look at pages 466-7 of the 4thAR WG1, Chapter 6 on Palaeoclimate . It’s the first page of the section dealing with the last 2,000 years).
Steven Talbot
No, apparently the ipcc did not completely drop the Mann HS Figure, as I had incorrectly implied. Still, the ipcc’s Fourth Assessment Report [4AR] Physical Science Basis, Summary for Policy Makers [SPM4] did not include the Mann Hockey Stick Figure in this very important presentation to the “Policy Makers” of the World, whereas the previous Summary For Policy Makers in the ipcc’s Third Assessment Report [TAR] “Scientific Basis” section did.
And I think many are beginning to see just why this Figure has not been carried forward from Summary to Summary by the ipcc, and why and how it was still retained at all by the ipcc, while in fact being known to be at least scientifically moribund.
Regardless, mea culpa: the Mann HS apparently is “carried forward” from the TAR to the remaining main body of the 4AR, contrary to what I implied in my comment to the NCDC. So I apologize to all for this mistaken blanket claim, which I made on the NCDC thread.
Which still brings us back to the critically important question of whether the Mann Hockey Stick Figure and the rather devious, misguided attempts thus far to continue to present it as credible should be carried forward only as classical examples of perversions of Science.

Pofarmer
August 14, 2008 11:34 am

About those climate models.
This outta leave a mark.
http://www.jennifermarohasy.com/blog/archives/003327.html

It is clear from the above formulation of the surface temperature rise and the associated amplification gain that each is sensitive to the specification of evaporation increase with temperature. Substitution of the average evaporation specification of computer models into the formulation will boost the projected temperature rise from the above expected value of 0.5oC to 1.5oC, the lower end of IPCC projections. When the specification of evaporation increase with temperature is very low, as in the more extreme models, then the feedback amplification gain increases to a value of about ten; the temperature sensitivity of the computer model becomes highly exaggerated and model would likely simulate the behaviour of runaway global warming. The behaviour, of course, is false and arises only because of the significant under-specification of evaporation.

And the guy would seem to have the chops to back it up.
William Kininmonth is a former head of Australia’s National Climate Centre; a consultant to the World Meteorological Organization; and author of Climate Change: A Natural Hazard (2004, Multi-Science Publishing)

Steven Talbot
August 14, 2008 11:58 am

Ken G,
You quote Wegman, well I’ll quote the NAS report:

The basic conclusion of Mann et al. (1998, 1999) was that the late 20th century warmth in the Northern Hemisphere was unprecedented during at least the last 1,000 years. This conclusion has subsequently been supported by an array of evidence that includes the additional large-scale surface temperature reconstructions and documentation of the spatial coherence of recent warming described above (Cook et al. 2004, Moberg et al. 2005b, Rutherford et al. 2005, D’Arrigo et al. 2006, Osborn and Briffa 2006, Wahl and Ammann in press)

http://books.nap.edu/openbook.php?isbn=0309102251&page=115
So where does that get us? It certainly doesn’t look to me like agreement that a hockey-stick-style pattern to palaeo reconstructions has been debunked, but it does to you, I guess? (Oh, and you asked for some other examples of confirmatory studies – there’s a start).
There have been actual published studies on this, such as:
“A comparison of tropical temperature trends with model
predictions”

Are you seriously not aware of the limitations of this paper? How is that posters here can take such relish in attacking perceived limitations in a ‘team’ paper but be, apparently, entirely unaware of far more obvious limitations in a paper which just happens to support the ‘sceptical’ argument? Is that what you call scepticism? How is it that you attack MBH 98/99 but not Loelhe 2007? Well, guess what, Loehle shows a strong MWP, so then it’s ok to gloss over the comical limitations of the methodology there, eh? How is it that so many here attack the GISS methodology but never say a word about Spencer & Christy’s huge mess-up of the UAH data record, up until 2005, when other scientists had to sort out their mess for them? (They got their plus and minus muddled up, in case anyone doesn’t know).
‘Scepticism’? It looks like double standards to me.

Peter
August 14, 2008 12:03 pm

counters:
“I’d respond to more but I simply don’t have the time this morning.”
Why? Is your supercomputer too slow?
Sorry, couldn’t resist….
🙂

Pofarmer
August 14, 2008 12:20 pm

Talbot.
You probably already know the contentious nature of Briffa, Wahl, and Ammman??? yes? There was just a big write up on it at Climate Audit.
How is it that so many here attack the GISS methodology but never say a word about Spencer & Christy’s huge mess-up of the UAH data record, up until 2005, when other scientists had to sort out their mess for them?
Because it WAS sorted out? Because the raw data was published? Because there weren’t hoardes of scientists backing up the faulty numbers? Because the data was subject to review?

Steven Talbot
August 14, 2008 12:42 pm

Profarmer,
Oh, for sure, I know what is being said on Climate Audit. I look forward to Steve McIntyre publishing his paper on the matter, presuming that he has something to say beyond insinuations on a blog. If he has proof, then let him publish.
“How is it that so many here attack the GISS methodology but never say a word about Spencer & Christy’s huge mess-up of the UAH data record, up until 2005, when other scientists had to sort out their mess for them?”
Because it WAS sorted out?

Only when others got onto the case, and not by S&C themselves (they applied the adjustments but had to be shown the error). Before that a great deal of hay was made out of the UAH trend being lower (including by Spencer & Christy themselves – funnily enough, I can’t find any contrite statements from them regretting what they’d said on the back of their bogus record!).
Because the raw data was published?
Nope.
Because there weren’t hoardes of scientists backing up the faulty numbers?
Nope – just Spencer & Christy “doing their own thing” which, apparently, is something folks here like to criticise Hansen for!
Because the data was subject to review?
Nope, it wasn’t. I hope that answers all your questions.
I don’t have anything against Spencer & Christy on this matter. Anyone can make a mistake, and they’re all dealing with imperfect observations and methodologies that are still being evolved. I’m just pointing out that the obvious ‘bias’ here is in people choosing to attack Hansen!

August 14, 2008 1:08 pm

Counters: I had read the realclimate link to the end. The last paragraph, which includes your referenced Thorne et al paper, reinforces Gavin Schmidt’s opening thoughts (the normal intent of any concluding paragraph) and confirms what I said in the earlier comment. But let me quote Gavin again from that realclimate thread, “If the pictures are very similar despite the different forcings that implies that the pattern really has nothing to do with greenhouse gas changes, but is a more fundamental response to warming (however caused).”
Again, if you’re seeing a warming in the upper tropical troposphere, it’s a result of El Ninos, not a substantial increase in CO2 or TSI.
Regards.

MarkW
August 14, 2008 1:14 pm

“No skeptic on the blogosphere has been able to demonstrate where the physics behind AGW is incorrect. Period.”
No warmist has been able to demonstrate that the physics are correct either.
On the other hand, the models all assume that relative humidity will stay constant as temperatures rise.
No study has ever confirmed this assumption. Most refute it strongly.