Thursday, August 19, 2010
The Mann "hockey stick": RIP?
Oh, my. There is a new and important study on temperature proxy reconstructions (McShane and Wyner 2010) submitted into the Annals of Applied Statistics and is listed to be published in the next issue. According to Steve McIntyre, this is one of the “top statistical journals”. This paper is a direct and serious rebuttal to the proxy reconstructions of Mann. It seems watertight on the surface, because instead of trying to attack the proxy data quality issues, they assumed the proxy data was accurate for their purpose, then created a bayesian backcast method. Then, using the proxy data, they demonstrate it fails to reproduce the sharp 20th century uptick.
Now, there’s a new look to the familiar “hockey stick”.
It would be a wonderful thing if the actual scientists and statisticians in the TigerHawk readership weighed in, pro or con.
13 Comments:
By 
, at
Thu Aug 19, 06:35:00 PM:
Off Topic.
Essilor optical buying Luxotica giving Essilor > 6000 optical locations.
Where is anti-trust investigation???
By 
JPMcT, at
Thu Aug 19, 07:24:00 PM:
Back on Topic:
I guess that's what happens when you try to extrapolate a few tree rings to world climate predicitons. It ends badly.
What a bad year for Al Gore...I bet he's glad he's already soaked the Progressive movement for a couple of hundred million. Glad he didn't get to soak the rest of us.
Still think we will get a Cap & Trade volley by the lame duck House this winter??
In my own back yard (medicine), we tend to at least try to follow the data. There used to be quite a few Extracranial to Intracranial arterial bypasses done back in the 80's to attempt to rebuild cerebral circulation in the presence of blocked internal carotid arteries. It made perfect sense and everybody had the occasional patient who "did well". Then cam along a single paper with a careful, statistical analysis of the procedure. Alas the damn thing just didn't make a statistical difference in survival or quality of life. The operation went away overnight.
It's called intellectual honesty.
I wonder if we will see any of it from the Mann-Made Global Warming crowd.
I doubt it. it's, after all, a religion and not a science.
By , at
Thu Aug 19, 08:21:00 PM:
I just returned from a trip to San Diego and while at the zoo, visiting the polar bear exhibit, I meandered past a very large "hockey stick" billboard-like display in an area nearby that explained to all the little children that our world is doomed from global warming (and by implication those poor polar bears were gonners).
Guess they'll need to rework that propaganda piece, eh?
By , at
Thu Aug 19, 08:31:00 PM:
1) Don't forget ... "they assumed the proxy data was accurate for their purpose." We know there's problems with the data.
2) CO2 was pinned as the culprit responsible for the "hockey stick" and given assumed magical thermodynamic properties necessary to have caused it. Models were then built that fitted to the hockey stick with CO2 as the driver. These models were then projected forward to show that the world would boil over. But no "hockey stick," no problem.
3) "after all, a religion and not a science." Leon Festinger studied what happens "When Prophecy Fails" (1956). Cult members often make rationalizations and only grow more fervent. They often turn to proselytizing others. Festinger's study gave us the term "cognitive dissonance."
So when we look back 20 years from now, which Nobel Prize will be the bigger joke ... Al Gore's or Obama's ?
By , at
Thu Aug 19, 08:46:00 PM:
I just had an acid trip flashback.
In the month after Climategate, Michael Mann got a $560,000 grant from Stimulus. I suppose because his was a "shovel ready project."
I could never have imagined such a thing. George Orwell, meet Philip K. Dick.
By Georg Felis, at
Fri Aug 20, 01:15:00 AM:
Just need to follow a couple simple rules in climate papers.
1. If you have a model that you feed your data into, and it produces an output that matches your theory, try feeding the model random numbers. If it still validates your theory, there's something wrong.
2.If you don't want to release your raw data, there's something wrong.
3.If your model predicts something horrible will happen now (or really soon), that has not happened in the last hundred-thousand years+, there's something wrong.
4. If you have to adjust your raw data in the same direction all the time, there's something wrong.
Disclaimer: I'm not a scientist. I got B's in Science in College. But somebody should have read the Evil Overlord List when they were putting this together. Somewhere in the middle of that should be a line like "My Scientists will not simply gather data points that agree with my Evil Scheme, but will be rewarded for finding critical errors in same, which will then be fixed."
By Don Cox, at
Fri Aug 20, 05:55:00 AM:
I can't see any point in commenting on this article when it has not yet been published. Let's wait until we have a chance to read it carefully.
If it turns out to be behind a paywall, then it is not possible to discuss it.
As for the "magical" properties of CO2, there is no magic involved. It is ordinary, well known physics which has been understood sine the 19C. The problem would be to explain how the concentation of CO2 could increase sharply without an increase in global temperature.
By TOF, at
Fri Aug 20, 10:13:00 AM:
Georgfelis is partly correct with the random number check; however, the random numbers have to be made to have a small correlation with previous numbers (autocorrelation). Economic series, for example are always correlated to some degree; weather patterns imply that temperature data too would be correlated. In any case, if the past is unable to predict the future something is wrong -- or you're taking the wrong approach.
Don Cox: The paper has already been read by at least several umpires. However, unlike the climate change community, the statisticians have professional pride in their objectivity.
By Don Cox, at
Fri Aug 20, 12:16:00 PM:
"The paper has already been read by at least several umpires. "
But I can't read it yet, you can't read it yet. So how can we comment on it?
By , at
Fri Aug 20, 06:00:00 PM:
If reading the actual paper is your problem, Don Cox, here it is. I understand it's straight forward enough to be read by most laymen.
http://www.e-publications.org/ims/submission/index.php/AOAS/user/submissionFile/6695?confirm=63ebfddf
By , at
Fri Aug 20, 10:52:00 PM:
As someone already posted CO2 is not a factor in climate change sooo logically if your base premise is wrong, your conclusions are also wrong. End of story. There is no AGW.
By TOF, at
Sat Aug 21, 09:56:00 AM:
Hey, Don Cox: If you have the qualifications of the people who wrote the paper I'll concede your point. If you don't it's highly unlikely you'll understand its arguments anyway.
By , at
Sun Aug 22, 09:06:00 AM:
If by "scientist" you mean to include soft sciences like history and anthropology, or in my case the use of obsidian hydration to date sites (in Africa)
I was perfectly willing to believe in man-made global warming until I saw Mann's hockey stick. He abolished the well established Medieval Warm Period and Little Ice Age (the evidence for both is overwhelming, the later even made it into 30s and 40s slushy movies (Hudson's Bay and Forever Amber)-- aside from the historical records, the architecture, memoirs recalling dates of the first freezes, etc., there are now numerous modern scientific studies listed on the CO2 Science website. Mann and Co have never explained how 13th century farmers in Britain could get two harvests in a year using medieval technology, so they simply ignored it.n For a far better explanation of climate change I would recommend a perusal of the work of Hans Svensmark and Nir Shaviv.
