Live link – you need to register to view.
This link for the Mann trial works every day of the trial, so bookmark it if you want to follow along every day.
Stream is live here:
https://dccourts.webex.com/webappng/sites/dccourts/dashboard/pmr/ctb518
Update:
Courtroom 132
https://dccourts.webex.com/webappng/sites/dccourts/dashboard/pmr/ctb132
Meeting number/passcode: 2343 119 3793
To see Mark Steyn’s POWERFUL opening statement, read this: https://www.steynonline.com/14039/opening-statement
My favorite part:
This case is about corruption, terrible, appalling corruption at the heart of a famous institution, Pennsylvania State University. Mr. Mann was the beneficiary of that corruption, as was Jerry Sandusky, as were others.
Addendum 1/20/24
Daily coverage here.
I’m not watching, but did read part of the transcript of Mark questioning of Mann. On paper and in black and white, It did a good job of exposing the pettiness of Mann and his ego.
As for the assertion of Mann;s “fraudulent hockey stick,” I do hope Steyn or Simberg’s excellent attorney note that everyone understood the opinion piece is not the same as an investigative piece where a police department states it. Of course, they will then make a strong case for intentional fraud, given the willful manipulations of data and criticisms of colleagues.
Sad to hear Mark’s health is failing. Mann should not get to slink off like he did not paying Tim Ball’s attorneys as ordered by a Canadian court which Mann himself turned to. Absolute low lifes like Mann and Fauci get way, way too much influence and deference – that they never earned in the first place.
Does anyone recall the 2006 Wegman report on the hockey stick?
How did this report go unnoticed? Did the mainstream media simply not cover it?
Of course the Wegman report was noticed. It was commissioned by Joe Barton, chairman of a house committee, who used it as his chief bludgeon in his 2006 inquiry into MBH. It took a few years for the evidence of how flaky it was to emerge:
1. It was full of plagiarism, especially the social network stuff. Apparently Wegman and co didn’t know anything about SNA, so they delegated to an uncredited author, a student, because she had been to a one week course. She just copied the whole thing from Wikipedia.
2. Their most famous bit was the claim that using Mann’s methods even red noise input would generate a hockey stick, and they showed 12 examples of this happening. But behind the scenes, what they had done was run 10000 red noise instances, and selected the 100 that looked most like hockey sticks. Their 12 were chosen from thta set.
3. The most obvious question, with all their claims of statistical error, was, what if you do the calculation without those claimed errors. Wegman was asked that, and said, no, they didn’t have the funding to do that. That was a lie. McIntyre had already done that. It was Fig 1 in the 2005 paper that was supposed to be Wegman’s chief exhibit. And in fact it made very little difference.
What Wegman had done, closely following McIntyre, as he did, was to focus only on an intermediate construct, P1, in the principal components analysis. But that just means stuff s was shifted to other components. Whne you add it all together, McI gets the same as MBH.
Not sure about point 1 .. I feel like that is acceptable for such a report, it´s a expert weihted summary after all
Point 2, “Mann’s methods even red noise input would generate a hockey stick” is correct adn very bad for Mann
Point 3 is confused! you call a financial statement a lie, so I would expert a financial counter argument, which is missing.
>> it made very little difference.
I am not sure what “it”, but McIntyre and other at his blog spend years to detail various factors and their effect.
The combination of the questionable decentered PCA analysis together with faulty proxies (Mann´s bristle cone pines are not temperature proxies) produces hockey stick graphs.
It seems that all you are saying here is that if the proxies are bad, the scientific method does not matter much, which might be true, but you should probably realize that you support shady science with your statement!
“acceptable for such a report”
Barton et al promoted the report invoking the authority of Prof Wegman. Looks dishonest if it is just pasted from Wikipedia.
“is correct adn very bad for Mann”
HS in what? The big bamboozle in Wegman is that they are talking only about P1. No-one is interested in P1. They want to know about the reconstructed NH temperature. And as McIntyre showed, varying the centering made no difference to the HS of that.
But the cheat of the unreported prior selection of 100 out of 10000 by HS appearance prior to making that claim about P1 is very bad.
“you call a financial statement a lie”
No. The lie is that they hadn’t computed a corrected NH reconstruction. Rep Stupak specifically asked if they had done that, and Wegman said no. But they had, and it made nonsense of all the confected fuss about P1. Congress was never told about this, despite two days of Wegman, McIntyre et al before the committee. It’s a key point.
point 1
“Barton et al promoted the report invoking the authority of Prof Wegman.
Looks dishonest if it is just pasted from Wikipedia.”
Not sure if anybody cares how it looks in your opinion!
The claim that these paleo reconstruction is done by a rather small group of people is true and Wegmann as an expert evaluated how this might spell trouble for independent reviews.
point 2
>> The big bamboozle ..
No it´s not. Like I said,
“Mann’s methods even red noise input would generate a hockey stick” is correct!
You should consider asking a mathematician if the first principle component in a principle component analysis is the most important, not a Mann-related climate scientist, but all your “wiggling” aside, the algorithm does not work as it should but generates spurious trends, when the underlying data has none!
Trying to make a thing more complicated that needed seems to show that you either don’t understand it or work on an agenda not involving the truth.
point 3
Me:“you call a financial statement a lie”
You: “No. The lie is that they hadn’t computed a corrected NH reconstruction.”
You wrote:
“Wegman was asked that, and said, no, they didn’t have the funding to do that. That was a lie.”
I am guessing wildly here, but the reason Wegmann didn´t do it was the missing funding, just like he says! Which means you say wrong things.
“But they had, and it made nonsense of all the confected fuss about P1.”
Well as it happens that is where the main information is found in any PCA, always!
“But that just means stuff s was shifted to other components. Whne you add it all together, McI gets the same as MBH.”
Yes, McIntyre is the bright one here, he can understand what Mann did decentering the PCA AND has many points of criticism of Mann´s work!
And again, because you seem to have missed it:
The combination of the questionable decentered PCA analysis together with faulty proxies (Mann´s bristle cone pines are not temperature proxies) produces hockey stick graphs. It seems that all you are saying here is that if the proxies are bad, the scientific method does not matter much, which might be true,..
but it is still mean that two times bad unscientific behavior in a paper produces nothing but BS!
“You should consider asking a mathematician if the first principle component in a principle component analysis is the most important”
I am a mathematician, and I understand PCA very well. It simply expresses a multidimensional vector in terms of different axes. In those axes it is still the same vector. The point of what Mann did is that you can recombine the top few components (five, in his case) and get much the same result with a lot of noise gone. It may shift patterns between the new axis components, but they don’t go away.
And they don’t. Here is McIntyre’s reconstruction with centered PCA, in the third panel. Top panel is MBH. He also illegitimately removed Gaspe cedar data, which affects 1400-1450. But the modern HS behaviour is exactly the same.
Please correct me if I’m not remembering correctly (I know you will), but wasn’t there mention in the Climategate emails where paleo proxies expert and MBH98 supplier Keith Briffa (RIP) advised Mann and “The Cause” group that the proxies they chose for their hockey stick hypothesis were not appropriate for that purpose?
“Climategate emails”
One of many examples where conspiracists trumpet something they found in an email when that thing was actually the subject of a published and much discussed paper (which is what they are discussing). Incidentally, co-authors on that 1998 paper include Phil Jones and Tim Osborn. They are describing just one more thing that they have to disentangle. Briffa published his own reconstructions.
Do you think that ‘short centered’ or de-centered PCA is a legitimate statistical technique?
Is it one which has been published and approved by the peer review process and statistical authorities and associations?
Why did Ian Joliffe say of it:
I can’t claim to have read more than a tiny fraction of the vast amount written on the controversy surrounding decentred PCA (life is too short), but from what I’ve seen, this quote is entirely appropriate for that technique. There are an awful lot of red herrings, and a fair amount of bluster, out there in the discussion I’ve seen, but my main concern is that I don’t know how to interpret the results when such a strange centring is used? Does anyone? What are you optimising? A peculiar mixture of means and variances? An argument I’ve seen is that the standard PCA and decentred PCA are simply different ways of describing/decomposing the data, so decentring is OK. But equally, if both are OK, why be perverse and choose the technique whose results are hard to interpret? Of course, given that the data appear to be non-stationary, it’s arguable whether you should be using any type of PCA.
I am by no means a climate change denier. My strong impressive is that the evidence rests on much much more than the hockey stick. It therefore seems crazy that the MBH hockey stick has been given such prominence and that a group of influential climate scientists have doggedly defended a piece of dubious statistics. Misrepresenting the views of an independent scientist does little for their case either. It gives ammunition to those who wish to discredit climate change research more generally. 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.
“An argument I’ve seen is that the standard PCA and decentred PCA are simply different ways of describing/decomposing the data, so decentring is OK. But equally, if both are OK, why be perverse and choose the technique whose results are hard to interpret?”
That is a key part. That argument is correct. He complains that it requires extra effort to interpret. But Mann is using it as a sophisticated method of noise removal. His result does not depend on any interpretation, as McIntyre’s centered version shows. You get the same reconstruction, which is the point of the exercise.
The virtue that Joliffe has failed to grasp is that the short period is one for which you actually have an average over that period for every proxy. Truly centered PCA would require an average back to 1400 for every proxy, and they don’t have that.
The contentious PCs were calculated from 70 tree ring series, all going back to 1400 or earlier.
Completely wrong. They were not all tree rings, and rather few went back to 1400. Here is MBH discussing how they diminish:
Don’t be stupid. The North American PCs going back to 1400 were calculated from 70 tree ring chronologies.
This seems to go on forever beside not being really that complicated!
IL: “why be perverse and choose the technique whose results are hard to interpret?” and
“ why be perverse and choose the technique whose results are hard to interpret?”
Wegman and McIntyre answered this: Using “Mann’s methods even red noise input would generate a hockey stick” (cited from you if recall correctly)
We therefore know that Mann´s method produces spurious trends when used with good data, that should have been the end of this discussion right there! A wrong algorithm is indefensible.
And the argument that this wrong algorithm produces a similar result than a good one when used on unscientific data is technically correct, but what does that prove beside you should not work like that!
Nick Stokes, you should NEVER work like that!
This is decades old information!
“We therefore know that Mann´s method produces spurious trends”
Trends in what? That is what you never answer. Wegman only showed, augmented by cheating, that it made a HS appearance in PC1, which incidentally could be up or down. That is not spurious. PC1 is just a calcultion intermediate. It is an artificial glimpse of the data. It doesn’t change the reconstruction, which is what matters. McIntyre himself showed that quite clearly, although it suited him to stop talking about it.
I dont need to answer, I have you answering it for me:
Wegmann reports correctly
“””that using Mann’s methods even red noise input would generate a hockey stick””” (me quoting you)
>> That is not spurious.
Yes it is! That is a classic example of creating spurions!
The data does not have, the PC´s show it => it´s spurious, the algorithm is bad!
IL already commented on the uselessness of Mann´s decentered PCA,
“I don’t know how to interpret the results when such a strange centring is used? Does anyone?“
You’re claiming that ordinary people who have read and are mortified by the Climategate cabal emails are the “conspiracists“?
Is the Antarctic continent in the far north on your upside-down planet, Nick?
In 2008, Kevin Trenberth argued that the absence of observed warming was a travesty. Wouldn’t the contrary be true and considered positive news? This would suggest that the theory might be incorrect, which is encouraging if one believes that a consistent increase in CO2 is causing an environmental catastrophe.
“In 2008, Kevin Trenberth argued that the absence of observed warming was a travesty.”
You really should try to cite and quote properly. What he said was:
“The fact is that we can’t account for the lack of warming at the moment and it is a travesty that we can’t.”
He is complainng about the lack of instrumentation, not the actual climate.
Incidentally this was another case where “sceptics” get excited about finding something in someone’s email, when the email is actually talking about a soon to be published paper. Same with Briffa above. Of course they never quote or read the actual paper. It is much more enticing to think a conspiracy has been uncovered.
Here is Trenberth’s paper. And here is how he starts out:
” From an energy standpoint, there should be an explanation that accounts for where the radiative forcing has gone. Was it compensated for temporarily by changes in clouds or aerosols, or other changes in atmospheric circulation that allowed more radiation to escape to space? Was it because a lot of heat went into melting Arctic sea ice or parts of Greenland and Antarctica, and other glaciers? Was it because the heat was buried in the ocean and sequestered, perhaps well below the surface? Was it because the La Niña led to a change in tropical ocean currents and rearranged the configuration of ocean heat? Perhaps all of these things are going on? But surely we have an adequate system to track whether this is the case or not, don’t we?
Well, it seems that the answer is no, we do not. But we should!”
Nick,
A mature scientist knowing what you have quoted from Trenberth has a responsibility to speak up when there is a problem with his/her specialty, to avoid dangerous action being taken when there is incomplete understanding.
I my ideal world, Dr Trenberth would have gone public to say words like “Colleagues, we know that there are unresolved problems because we do not fully understand this topic. So please issue caveats when you write papers, to say that your conclusions are NOT to be used to inform new policies.”
Correct me if I am wrong, but there does not seem to have been any attempt to rein in the policies that are now, one country after another, on the road to creating a penurious citizenry.
Debate quickly became an ugly, snarling dog fight while we senior scientists mostly looked on and ate another vegemite sandwich.
What I quoted was a published paper by Trenberth.
>> He is complainng about the lack of instrumentation, not the actual climate.
Is not correct, KT is talking about the models of that time being unbalanced, the measurement grid was good enough for his argument!
KT says nothing about GCM’s or grids. He is complaining about measurement inadequacy, specifically at TOA. Here is part of his conclusion:
“He is complaining about the lack of instrumentation, not the actual climate.”
That’s absurd. He’s terribly upset that he can’t account for where the “radiative forcing has gone” without considering that maybe it doesn’t exist.
Did this come from the same emails chain where one of the contributors wrote –
“What if we’re all wrong? They’ll probably kill us.”
I originally mentioned a different quote. I thought I had Trenberth in mind, but it turns out I was actually recalling a conversation between Mick Kelly and Phil Jones:
“October 26, 2008
Hi Phil
I just updated my global temperature trend graph for a public talk, and noticed that the level has really been quite stable since 2000 or so, and 2008 doesn’t look too hot.
Anticipating the skeptics latching on to this soon, if they haven’t done so already…
It would be awkward if we went through another early-1940s-type swing!
Phil Jones:
Mick, They have noticed for years-mostly with respect to the warm year of 1998. The recent coolish years we put down to La Niña. When I get this question I have 1991-2000 and 2001-2007/8 averages to hand. Last time I did this they were about 0.2 degrees different, which is what you’d expect.
Kelly:
Yeah, it wasn’t so much 1998 and all that that I was concerned about, I’m used to dealing with that, but the possibility that we might be going through a longer-10 year-period of relatively stable temperatures beyond what you might expect from La Niña, etc.
This is speculation, but if I see this as a possibility then others might also. Anyway, I’ll maybe cut the last few points off the graph before I give the talk again, as that’s trending down as a result of the end effects and the recent cold-ish years.“
Proper scientists would have remarked –
“Hhmm, our observations aren’t matching our predictions. We must review and revise our hypothesis. ”
But of course these characters aren’t subscribers to the scientific method.
They’re all apostles to “The Cause” (as M. Mann refers to his “science”.
Unfortunately, too few people have read the “Climategate cabal emails”. I bet I could stop 100 people at random, and I’d bet that not one ever heard of them.
I fear you’re right.
Nobody reads the details of anything any more.
They just skim headlines.
And the msm know this, so they put up ooga-booga headlines to convince folks that climate armageddon is imminent.
(aided and abetted by university media departments who write the press releases of course)
…something you will see is that Niky doesn’t directly respond or address the question asked. He deflects and throws a related but oblique link that he represents as countering the initial question, but if you pursue it you find something like this from the Briffa team in the Paper Nick sent you too.
‘On annual, decadal, and probably even
centennial time-scales, tree-ring data are demonstrably
reliable palaeoclimate indicators, but where the focus of
attention shifts to inferences on century and longer time-
scales, the veracity of inferred change is di§cult to
establish’
which means the answer to your question is “Yes, Briffa did warn the InCrowd that they should be careful using his results”
>> The point of what Mann did is
using a bad proxy and a bad method.
Since you are talking methods here, you have to test the method with good data, and Wegmann reports correctly that the method
“””that using Mann’s methods even red noise input would generate a hockey stick””” (me quoting you)
You don´t seem to debate this (which would be silly)!
Instead you wiggle into what happens when a bad method is used with bad proxies.. nothing interesting, it´s just a bad idea! (seems equally silly to me, are you just trying to be silly here, it sure looks like it!)
Being a mathematician that should not be too difficult for you!
Can you understand that your silliness costs billions of real money?
You harm people by writing this nonsense!
Like I said earlier, I smell a different agenda maybe personal enrichment!? In which case: Shame on you!
Do we really need to debate, why bristle cone pines are bad proxies? Even a mathematician like you should have gotten that memo by now!
Are you trying to defend Mann´s use of that proxy in any way?
Or his use of a bad method (which you did not seem to debate, right? The hockey stick generation by his algorithm on simulated red noise data seems well documented and published, right?)
So what are you really saying here? Beside “look what happens if Mann or McIntyre or whoever uses a bad method on bad proxies”?
We know the method is bad, we know the proxies are bad, move along please! You seem to waste everybody´s time! Why is that?
“Do we really need to debate, why bristle cone pines are bad proxies?”
Please explain why they are. But this is shifting the goalposts. Wegman was called in as a statistician. He doesn’t know anything about bristlecone pines, except what he parrots from McIntyre. Nor do you.
“We know the method is bad”
The method is not bad. You can use totally uncentered PCA if you want to. The penalty is that you have to calculate a PC corresponding to difference in means that you could anticipate.
But to get that gain, you have to subtract the mean for the whole period, and that is unavailable for most proxies. So you can subtract the data for the part periods that you have, or as Mann did, subtract a fixed part period that you always have. Either way, you come out between fully centered and uncentered, both sound methods.
Uncentered PCA is fine if we have reason to believe that the data already has zero mean. By construction, tree ring chronologies are nonnegative with mean around 1, so uncentered PCA is bad.
“Decentered” (Mannian) PCA is much, much worse. It’s not “between fully centered and uncentered.” Jolliffe explains this.
>> Please explain why they are.
Does that mean you claim not to know this? How about your argument falling apart if that statement would be true?
NS: “The method is not bad. ”
Is not correct, and endlessly repeating it will not someheow make it true!
As you already said:
“using Mann’s methods even red noise input would generate a hockey stick”
So here you have already agreed on that the method is bad!
(Or at least it´s doing something which it NEVER should be doing and thus distorts data in a reproducible way, which is NOT GOOD!)
I concede that for bad data like the one which was used by Mann, that difference does not matter, but it is also well known what happens when you remove the “Graybill bristlecone pines” from the dataset:
All of the sudden there is a very significant difference between Mann´s faulty method and a centered PCA analysis!
Here the faultiness clearly matters, there is not wiggle room about this!
For new readers, here is the information about Mann´s faulty proxies, again known since decades, Nick Stokes (and also Mann of course) knows this too, at least he has been told for decades :
https://www.geo.arizona.edu/Antevs/Theses/AbabnehDissertation.pdf
“”” Climatic factors include temperature, precipitation, soil moisture, and prevailing winds (Fritts 1976), while other factors include soil moisture,
fire history, nutrient availability, and parent-rock material (Fritts 1976; LaMarche 1975; Taiz and Zeiger 1996). The degree to which these parameters affect tree growth varies mainly with microsite characteristics, elevation, and whether an area is mesic, arid, or semiarid. “””
and (about stripbark pines in her study)
“””Unfortunately, it was not possible to unequivocally identify a clear climate signal with a linear relationship to radial growth, and so no reconstruction was developed.”””
So your authority is a PhD student, writing in partial fulfilment of the degree requirements. The first quote is just standard. That is why they do the calibration and testing. The second is from the second (of 3) substudy of her thesis. She just had a small sample from one location, so indeed a full reconstruction was going to be difficult (whatever the trees).
But in the third study, ahe says:
“In this research, I use a paleoclimatic model that is based on temperature inferences and precipitation reconstructions from tree-ring widths of bristlecone pine in the White Mountains of California to help explain aboriginal subsistence-settlement in alpine villages of the White Mountains. Temperature inferences and precipitation reconstructions are compared visually and statistically with calibrated archaeological 14C dates.”
IOW, a reconstruction based on bristlecone pines.
>> But in the third study, a he says:
Are you claiming here that Ababneh published a temperature reconstruction of stripbark pines?
That is wrong!
Or are you talking about some other trees?
That would make your defense of Mann´s paper weirder then it currently looks like..
“Mann´s method is wrong, but there are trees” does not really cut it!
So, it remains established that Mann´s method is wrong and Mann used a bad proxy! If you dont like the newer PhD thesis, Graybill was very clear about it too https://agupubs.onlinelibrary.wiley.com/doi/epdf/10.1029/92GB02533 :
” It is notable that trends of the magnitude observed in 20th century ringwidth growth are conspicuously lacking in all of the time series of instrumented climatic variables that might reasonably be considered growth-forcing in nature”
The 2006 NAS panel stated that stripbark chronologies (i.e. the Graybill bristlecone chronologies) should be “avoided” in temperature reconstructions.
It is instructive so look at the data
https://climateaudit.org/2018/10/24/pages2k-north-american-tree-ring-proxies/
(that is also one of McIntyre´s many articles showing what happens when Mann omits the faulty proxy the hockey stick disappears. Mann knew this and did not disclose it, data torture indeed, I hope McIntyre takes Mann apart in court!)
Nick, training the PCA as was done was faulty statistics. It is an error commonly made in the social sciences. It seems logical but it generates faulty results (spurious correlations).
Just to clear here, this was not some training accident, but is a fundamental problem of Mann´s method!
“Selecting on the dependent variable” is one of the most common, bonehead statistical errors. It shows up repeatedly in climate science and all the social sciences as “amplification” or “calibration”. It looks completely logical but it is an error. The result is spurious correlations (false cause and effect).
But if Mann and co. had simply collected as many ancient tree core samples as possible and selected entirely at random they would not have got their hockey stick 🤔.
Agreed random selection would not have has the statistical problem of selection bias.
“Decentered PCA”
That is a red herring. Decentering is not the issue. The issue is using the CO2 period as a filter for PCA. The so called calibration is selecting on the temperature dependency on CO2 which is a bogus method.
The training interval was 1902-1980. But then it has to pass the verification interval 1854-1901.
Thank you, Nick. And there are several publications to which you can link to show the mathematical results of such training and verification?. You and I know that these procedures have subjective elements that are seldom described as subjective and hence capable of error.
An anecdote. In my early CSIRO job, just graduated, I was handed several sheets of results of a factorial experiment, told to find Ronald Fisher’s text book and conduct a manual Analysis of Variance. These days peope can have a program to do the job on a computer somewhere, a process that usually lacks the learning of why the calculation was done in a particular way and what its limitations are. Same token, we encouraged field explorers to know how to navigate the wilderness without GPS in case of flat batteries, we encouraged hand-drawn and coloured maps for years after computer mapping aids were easy.
Remember those times?
Geoff S
He does not need to, we already know that the algorithm is bad!
It has been tested and published!
Just stop using it already!
Oh and of course withdraw/correct any publication using it!
Geoff,
Mann’s Nature paper (MBH98) is a good starting point, especially the methods section.
Followed by the Wegman report showing how the method is wrong!
As you (Nick Stokes) already posted:
“using Mann’s methods even red noise input would generate a hockey stick”
Asked and answered!
You keep deliberately misrepresenting. My full para describing Wegman was:
“Their most famous bit was the claim that using Mann’s methods even red noise input would generate a hockey stick, and they showed 12 examples of this happening. But behind the scenes, what they had done was run 10000 red noise instances, and selected the 100 that looked most like hockey sticks. Their 12 were chosen from thta set.”
Plus what you deliberately and persistently overlook – none of this describes the actual reconstruction, which is not affected. Only the artificial intermediate PC1. Wegman insisted that he couldn’t be expected to do a recon, being too thick to notice that McI had already done it.
Like I wrote: Asked and answered!
>> You keep deliberately misrepresenting.
Is a lie! I was even quoting you!
>> My full para describing Wegman was:
Yes, for everybody to see, the faulty algorithm does something it should not do!
>> Plus what you deliberately and persistently overlook
Is another lie, I wrote:
I quote one of my answersn:
“””
Since you are talking methods here, you have to test the method with good data, [..] You don´t seem to debate this (which would be silly)!
Instead you wiggle into what happens when a bad method is used with bad proxies.. nothing interesting, it´s just a bad idea!
“””
I really think only very silly people defend an analysis with faulty data! You should know better, but the evidence I presented is
It is interesting that you somehow try to change the burden of proof! Mann did analyzes the data without that faulty proxy and the hockey stick disappears. It is unethical that he did suppress that fact in his article!
Nitpick Nick is another closet marxist for whom the ends justify any means.
No amount of nitpicking can help supporting this view!!
And one of the first comments to this post is one of mine quoting McShane and Wyner:
“””
[..]Consequently, the application of ad hoc methods to screen and exclude data increases model uncertainty in ways that are ummeasurable and uncorrectable.[..]
“””
So with that we have a faulty algorithm and (Nick Stockes agreed to that!), bad data and even if both of that were not a problem there is the question if the selected data is representative for the question.
At this point you can ask, what did they do right with that publication? They wrote their names on it so you can held them accountable!
Bradley is already trying to blame Mann for this analysis, but he is as accountable for the numerous mistakes made in the publication as is Mann! He should be questioned for his admission that he didn´t check the mathematics of an article he is co-author! He lent his fame, but of course now suffers the blame too!
And Hughes was Ababneh´s PhD supervisor and knew fully well about the problem´s of stripbark proxies not representing the temperature!
There is no believable way that not ALL three of them knew that this analysis is Bullshit! (Mann hid one without those trees and as a result without hockey stick shape in his censored directory, I am guessing they will talk about that tomorrow)
“Nick Stockes agreed to that!”
Lie!
Poor reasoning Nick: whatever happened “behind the scenes” does not invalidate the fact that 12 examples of red noise generating a “hockey stick” were indeed found and presented . . . just as was claimed.
Wait, maybe we can just say the algorithm is “just a little bit wrong”!
This way Nick Stokes has invented a new mathematical thing!
Before him there were only right and wrong algorithms, but this one is “just a little bit wrong” which is a completely novel way to look at things!
Before him only scammers and button feeders used terms like that, but now it can enter the world stage! Possible he gets invited to talk and prestigious societies about this and maybe he wins a Nobel price! How hard can that be with ideas like that?
Or of course it is just bullshit and Nick Stokes knows it well!
After all he also claims to be a mathematician!
Uh, here is a serious question not to post empty thins like “Nick Stockes agreed to that!”
Lie!”
and waste everybody’s time here!
Does anybody ever seen a more condemning verdict about a method by Ian Joliffe than this one? (It might be worthwhile to know who that man is to understand how bad that really is!)
“and waste everybody’s time here!”
To save time, don’t lie.
“It might be worthwhile to know who that man is to understand how bad that really is!”
It might. Then you’d know that his name is actually Ian Jolliffe. You can tell who is getting their stuff direct from McIntyre from that mis-spelling.
But trhere is no evidence that Jolliffe actually looked at the effects of Mann’s algorithm. He just dismisses it as something that he doesn’t teach in his class.
To save time, don’t lie.
Good advice. BTW, are you still claiming that wind and solar are cheap because wind and sun are free?
You have spelled his name with one l in this very comment section, Nick.
And your claim is still nonsensical. The contentious North American ITRDB PCs going back to 1400 were calculated from 70 tree ring chronologies, all of which have data back to 1400 or earlier. Nothing stopped Mann from doing “truly centered PCA.”
“You have spelled his name with one l “
I wasn’t the one claiming “to know who that man is”
“ The contentious North American ITRDB PCs going back to 1400 were calculated from 70 tree ring chronologies, all of which have data back to 1400 or earlier. Nothing stopped Mann from doing “truly centered PCA.””
The ITRDB PCs were only a small part in the whole reconstruction. The other indicators diminished as in the section I displayed from MBH99.
They were not a small part, but that’s irrelevant. It’s not like ice core data from the Andes or whatever stopped Mann from using conventional PCA on the tree ring data.
If you take 10000 instances of red noise, without any MBH processing, and select the top 100 according to McIntyre’s HS index, then you’ll get hockey sticks too.
It is not the question what I can do, but what Mann did!
And there we have
the Wegman report wrote that ” using Mann’s methods even red noise input would generate a hockey stick” (your words)Graybill stripbark pines are not temperature proxies and shouldn’t be used in that matter according to the NAS 2006 reportand a spurious proxy selection procedureAs Mc Shane and Wyner write :
“””
[..]Consequently, the application of ad hoc methods to screen and exclude data increases model uncertainty in ways that are ummeasurable and uncorrectable.[..]”””
Altogether solid published criticism, which makes that publicaiton indefensible, as Nick Stokes keeps on showing!
You got it all wrong, Nick Stokes! Mann´s decentered PCA method is uniquely bad! And “just a little bit wrong” as you may call it..
And I am NOT the one who needs to lie baout it!
Please provide proof of that statement.
OK. Here is a plot from here of 12 instances of PC1’s of random noise, calculated with centered differencing. They should not have HS look; that is the point of Wegman’s 12 tableau with short-centered mean. But they have been subjected to McI’s HS 1% filter, so yes, hockey sticks is what you see:
“OK. Here is a plot from . . . so yes, hockey sticks is what you see:”
So, you seriously want to assert in front of the discerning WUWT audience that any single one of the 12 “instances” that you have given is in any sense similar in features to the infamous Mann “hockey stick” graph?
Seriously???
For your information (and it’s quite obvious you need the lecture), a hockey stick is characterized by a relatively flat linear “handle”, a fairly sharp break point (i.e., radius) and <90° turning to a “blade” extension, and the blade itself being relatively linear. In this sense, Mann’s graph appears to resemble a hockey stick laying somewhat flat on its handle on the left side with the blade turned upward on the right side.
To help you out in seeing how badly you missed presenting any “hockey stick”-like graph in your 12 cherry-picked examples, I’ve attached the original Mann, Bradley & Hughes [1999] hockey stick-like graph that is central to the above article and Mann’s trashing of scientific data and ethics (graph extracted from https://en.wikipedia.org/wiki/Hockey_stick_graph_(global_temperature) ).
BTW, thank you for providing today’s comedy . . . I knew you had it in you!
“I’ve attached the original Mann, Bradley & Hughes [1999] hockey stick-like graph that is central to the above article”
That is not the original. The MBH1999 paper was in black and white. PAGES2K, as marked on the right axis, was not even thought of until ten years later.
My examples were cherry picked; that is the point. So were Wegman’s, by HS index, in the same way. My examples were not intended to emulate MBH98. They were intented to emulate what Wegman showed as supposed proof that MBH methods would produce a hockey stick. He said it, not me. And here is what he showed:
“a fairly sharp break point (i.e., radius) and <90° turning to a “blade” extension”
In fact, Mann’s proxy results did not show that, nor would you expect them to. The recon finished in 1980, and a lot of proxies didn’t get that far. The proxy method was never intended to tell you about recent times; it was to tell about the blade – how it was before we had instrumental data. The blade itself is much better attested by modern instruments.
Here again are Mann’s results, as plotted and verified by Steve McIntyre
Blade?
“ it was to tell about the blade”
I mean, of course, handle.
Nick,
if you had bothered to consult the Wiki link I provided, you would see the that title of the associated graph I attached to my post states the following, verbatim:
“The original northern hemisphere hockey stick graph of Mann, Bradley & Hughes 1999, smoothed curve shown in blue with its uncertainty range in light blue, overlaid with green dots showing the 30-year global average of the PAGES 2k Consortium 2013 reconstruction. The red curve shows measured global mean temperature, according to HadCRUT4 data from 1850 to 2013″
(my bold emphasis added)
If you have a problem with this title, please take it up with Wikipedia . . . I’m sure they will give it all the attention that it deserves.
The hyperlink associated with “Mann, Bradley & Hughes 1999” in this graph’s title is:
Mann, Michael E.; Bradley, Raymond S.; Hughes, Malcolm K. (1999), “Northern hemisphere temperatures during the past millennium: Inferences, uncertainties, and limitations”, Geophysical Research Letters, 26 (6): 759–762, Bibcode:1999GeoRL..26..759M, doi:10.1029/1999GL900070
but you probably didn’t bother to notice this.
I intentionally referenced the Wikipedia article containing that graph instead of the MBH99 paper as I do not believe in propagating the scientific malfeasance and garbage presented by the MBH paper (here I’m making a exception to emphasize my point), but moreover because the Wikipedia article contains a discussion (albeit biased) of the details and history of the Mann “hockey stick” saga, as a reference for all WUWT readers.
However, since the Wiki graph (with its extra data overlay) bothers you —and in preparation for what’s to be said below—I’ve extracted directly from the above-cited MBH99 paper the infamous hockey stick graph as it was published as Figure 3(a) in that paper, and it is attached. Please note that this graph covers the years 1000 to 1998 AD.
And, yes, for your benefit, the Wiki reference specifically refers to the “blade” portion of the hockey stick in its second paragraph. And contrary to what you assert, I do believe the blade end of the attached MBH99 graph is quite easy to discern.
However, I notice that your posting of “Figure 1. NH Temperature Index” “as plotted and verified by Steve McIntyre“, which you claim “are Mann’s results” has a x-axis spanning only 1400–1980 AD. The result of your posted “Figure 1” graph having a greatly expanded x-axis (time axis) compared to that in the attached MHB99 graph is to make the break and rise in temperature anomaly appear much more gradual. Also, the result of your posted “Figure 1” graph terminating at 1980 compared to the MHB99 graph continuing to 1998 is reduce the apparent magnitude of rise in the “blade” portion of graph. Thus, your final question “Blade?” is see to be nothing more than lack of comprehension on your part or a sophomoric attempt at misdirection.
Finally, as to your back-peddling statement:
I need only quote from your above post of January 29, 2024 10:23 pm:
(my bold emphasis added)
I fail to see your point of this line of arguments!
Are you saying that Mann´s method does not data mine?
That would be wrong!
Are you trying to say there are other ways to data mine?
That is correct, but irrelevant (until we start looking into Mann´s later publications)
How long are we supposed to look at each of those pictures using other mehtods, before we can conclude:
Yeah Mann´s method is bad, just Wegman said?
Probably. But if you a select them with correlated recent “warming” then they’ll all be hockey sticks.
Nick, how would you go with this approach –
instead of using your expertise to defend all aspects MBH98’s science from all critics, try applying your expertise to investigating the flaws in MBH98, as the Scientific Method requires.
Remember Nick, as a scientist, you’re a referee, not a quarterback playing on the home-side team.
I investigated Mann’s technique. By so did McIntyre. He ran with Mann’s, and with centered differencing (or the best approximation to it that he could). And he got the same answer. The paper is here. The key graph showing that it makes no difference, is agaih:
I don’t thinks those charts show what you claim they show, given there is a significant difference in starting point and trend prior to 1700.
The difference in starting point is due to McIntyre erasing the Gaspe cedars 1400-1450, as the second bar shows. But “prior to 1700”? The argument has been about MBH creating hockey sticks, ie the modern warming. But McI has that exactly the same.
No nick, the discussion is about the suitability of the methodology used by Mann et al. and the scientific integrity of the authors (you know, the reason for the OP and all of this discussion). And, specifically the ability of that model to…IN YOUR OWN WORDS… ” know about the reconstructed NH temperature.”
Abstract
The differences between the results of McIntyre and McKitrick [2003] and Mann et al. [1998] can be reconciled by only two series: The Gaspé cedar ring width series and the first principal component (PC1) from the North American tree ring network. We show that in each case MBH98 methodology differed from what was stated in print and the differences resulted in lower early 15th century index values. In the case of the North American PC1, MBH98 modified the PC algorithm so that the calculation was no longer centered, but claimed that the calculation was “conventional”. The modification caused the PC1 to be dominated by a subset of bristlecone pine ring width series which are widely doubted to be reliable temperature proxies. In the case of the Gaspé cedars, MBH98 did not use archived data, but made an extrapolation, unique within the corpus of over 350 series, and misrepresented the start date of the series. The recent Corrigendum by Mann et al. denied that these differences between the stated methods and actual methods have any effect, a claim we show is false.
I would suggest getting a 300 year temperature decline wrong is enough to hint that the model is not very good and the obfuscation and evasion by the authors is damning.
You are quoting the Abstract of McIntyre’s paper. But all it says is something about the intermediate PC1, and the issue of Gaspe cedars. The latter is not an issue of method; it is typical of McIntyre’s auditor approach. He said that 50-year data set to 1450 had 4 missing values (which Mann interpolated), so it has to go. Nothing else wrong with the data. But this is the opposite of science:
“ihe scientific integrity of the authors”
How about the scientific integrity of McIntyre and Wegman. In the Wegman report, a main feature was this plot:
One of the most compellinbg illustrations… he says, and the narrative about Mann’s method creating hockey sticks out of red noise has been repeated endlessly. What he doesn’t tell you is that they did 10000 runs with red noise, but picked from that the 100 that look most like hockey stick (HS index). Then they select the 12 to show from that 1%.
Of all 400+ proxy records, this outlier was the only one Mann decided to manipulate in this way. McIntyre spotted the cherry picking and undid it.
“ this outlier was the only one Mann decided to manipulate in this way“
Nonsense again. Like all datasets, there are many missing values, which are infilled. Gaspe ws not an outlier. McIntyre just invented a rule about when you can do it and when you can’t. And discarded a whole lot of good information for no reason. He isn’t interested in making use of what the data is telling us. He is only interested in the auditor’s gotcha.
It was the most hockey-stick shaped series with data back to 1400 (except that it didn’t go back 1400), so yes, it was an outlier. And again, it was the only proxy record that was extrapolated back in time. McIntyre correctly undid the Mannipulation.
“It was the most hockey-stick shaped series”
Hardly relevant, if true. McI erased the years 1400-1450.
“And again, it was the only proxy record that was extrapolated back in time.”
Again, if true, hardly relevant. The years 1400-1403 were missing. That doesn’t justify discarding 1404-1450.
As Mann said, even if they posted a recon back to 1404 rather than 1400, it wouldn’t make nearly the difference that McI’s botch made.
It is true, and you would have known this if you had looked at the data.
Again, Mann “deleted” proxy data the same way McIntyre did, except for Gaspé. McIntyre correctly excluded the one proxy record that Mann cherry-picked.
CHAIRMAN BARTON: Dr. North, do you dispute the conclusions or the methodology of Dr. Wegman’s report?
DR NORTH: No, we don’t. We don’t disagree with their criticism. In fact, pretty much the same thing is said in our report. But again, just because the claims are made, doesn’t mean they are false.
From the North 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 both additional large-scale surface temperature reconstructions and pro- nounced changes in a variety of local proxy indicators, such as melting on ice caps and the retreat of glaciers around the world, which in many cases appear to be unprec- edented during at least the last 2,000 years. Not all individual proxy records indicate that the recent warmth is unprecedented, although a larger fraction of geographically diverse sites experienced exceptional warmth during the late 20th century than during any other extended period from A.D. 900 onward. “
In 2006, of course, neither Dr North nor congress knew about the shenanigans of plagiarism and the hidden preselection of allegedly random hockey stick shapes, nor had they been told about the MM2005 paper which showed that changing the centering made no difference.
“He said that 50-year data set to 1450 had 4 missing values (which Mann interpolated)”
The data was filled in at the beginning of a dataset, outside of the known data range….that is extrapolation.
you know this whole line of “what aboutism” started sounding really familiar,
and now I know why. “One trick nick” at it again….
https://wattsupwiththat.com/2023/05/01/mcintyre-reverse-engineering-a-hockey-stick-shows-bogus-methodology/
The thing is, people here greatly condemn Mann’s science and statistics, about which they don’t really have a clue. If you ask why they think that, they say, well, Mcintyre said…
So it’s relevant.
Nick,
Thankyou, but do you think that I have not read MBH several times over already?
Geoff S
So Mann used thermometer data to confirm thermometer data. Same problem. Selecting on the dependent variable. Thermometers being dependent on CO2.
Good fortune allowed me to conduct science (successfully) in the period 1960-1990, a time in which science flourished in a spirit of cooperation, achievement and mutual respect between scientists. We saw Man on the Moon and more.
In those times, a legal case like this would have been unlikely to happen. In those golden times, if there was a problem with some science, the more usual response would be offers from others
to help. Sadly, that attitude has been swept aside since the late 1990s by conflict, antagonism and growing disrespect of science by the public. I can even see “hate”, a troublesome trait, expressed by some of the trial participants.
The present case of Mann vs Stein has opportunity to reverse the present decline, to put respect back into the social picture. It is, in my opinion, unlikely to do this because in a trial by
jury difficult concepts tend not to be understood adequately. How can a jury be
wise about matters like the weighting of components in a PCA, when mathematicians
and statisticians have not settled their differences?
Thankfully, the Judge appears to have informed himself of matters like the Daubert Standard that provides a systematic framework for a trial court judge to
assess the reliability and relevance of expert witness testimony before it is
presented to a jury. This might be reflected in his final directions to the jury, before they gather to decide.
It is not for me as a distant outsider the try to tell the Judge what to do. All that I can contribute is some years of experience (including management of some past Australian cases before our highest court) and a good memory of the “Golden Years” and their benefits.
If I can be allowed to state a few preferences, they would include at least these three:
The Judge might consider asking the jury to distinguish between evidence that is vague, even misleading, as opposed to crisp, direct and relevant. An example would be to go beyond the Mann “I wash my hands” evidence that co-authors selected the data for the MBH hockey stick, when it was Mann’s deletion of parts of the selected data that was promptly criticised by others.
The Judge might consider that the evidence that some science was fraudulent is more important than who claimed fraud and how it was asserted. Fraud can approach a yes/no state like pregnancy; it need not matter who the father was. Long time readers of Climate Audit might like me, be convinced that Stephen McIntyre has shown many instances of fraudulent science in such unequivocal detail that the Daubert Standard is satisfied.
Finally, the Judge might consider recommending future actions that will reduce some of the harm being done to science by this case. He might recommend that the validity of the hockey stick findings be formally tested after 25 years of argument on many platforms. Simply, there are recorded criticisms of defined aspects of the hockey stick science. Many criticisms describe what “should have” been done to avoid the criticism. Surely, it is not beyond the wit of top US science administrators to assemble a team in the neutral corner to repeat the Mann, Bradley & Hughes work with the “should haves” included; then to show the results. A by-product might be the writing and publishing of better official guidelines for the required standards of conduct and reporting of science, such as requirements to report validation tests, to provide copies of original data, to fully detail estimation of uncertainty of measurement, to report sub-experiments in the work that failed to produce the expected result as well as those that did; and so on. There is too much sloppy science, of the types that people like Nobel Laureate Richard Feynman has made the subject of books.
Geoff S
Surely, you overestimate/misplace the objectivity, capabilities and motivations of “top US science” administrators.
The supporting evidence for this misfeasance and malfeasance of office has been amassed over at least the last 25 years, but unlike Mann they will never be brought to trial.
My apologies. Steyn, not Stein.
Auto correct again. Of course I know the correct spelling.
Geoff S
Crap science like Mann’s would sink without trace if it didn’t have the “vested interests” political activism backing.
Cleaning the academic stables of ideological / political influences is the Hurculean task that must be undertaken.
Steyn is valiantly trying to get the hockey stick put on trial in this case but it isn’t the central part of the case at all. Mann cannot prove that Steyn/Simbergs articles caused him any damage alone as the media was abuzz with many similar articles saying the same thing at the time and Mann cannot prove that his grants were affected and not other departments. Whether the hockey stick was fraudulent may not be settled here.
Google: “selecting on the dependent variable”. It sounds like such a good idea. Only look at sick people to “amplify” or “calibrate your study of some medicine.
Only look at thermometer data to “calibrate” or “amplify” your tree rings.
It is bogus math. It is the same mistake that said the countries that eat the least fat have the lowest heart disease. It is the bogus math that is causing huge amounts of diabetes.
“ Only look at sick people to “amplify” or “calibrate your study of some medicine.”
Which is, of course, exactly what they do. They don’t test a chemotherapy med, say, on random people. They test it on those they think might benefit from it. In the same way, they don’t look at trees in England. They go to places like Siberia and White Mountains, where temperature is likely to be a limiting factor. Th at practice long predates Mann.
You have to calibrate to get the proportionality constant between tree rings and temperature. If you can’t establish one, you can’t use the tree. No choice.
Nitpick Nick tilts at the windmills again, trying to prop up the tree thermometers.
What’s the real measurement uncertainty intervals of these, El Nicko?
Whether or not Mann used “Selecting on the dependent variable” on purpose is an interesting question.
If you have statistics training you should know about the problem and how hard it is for the lay person to understand because it defies common sense. I expect it is the sort of thing that would not be caught by peer review.
Human beings are easily fooled and the easiest person of all to fool is yourself.
The training interval was 1902-1980. But then it has to pass the verification interval 1854-1901.
=======
Nick,
I note you have not disputed that “calibration is a form of selection bias. Likely to lead to spurious correlations.
I contend that it is a form of selection on the dependent variable, wearing lipstick and high heels.
Others have called it circular reasoning, because it is a form of feedback. You are using the thermometer data to selectively amplify the signal.
But your feedback data is displaced in time, scaled, and trimmed, leading to distortion of the underlying signal. This distortion invalidates the result.
Then after all this “training”, the models are nothing more than linear extrapolations along whichever CO2 growth curve the operators prefer. This is another reason why averaging the outputs of diverse models is bogus.
Plus the historic LIG temperature data before 1900 isn’t reliable for these purposes because of innate systematic errors they cannot “adjust” away.
Both of these problems were revealed by Pat Frank, BTW.
Yes, it’s like that grouse hunter in the woods in grizzly country who comes across a steaming 1-foot high pile of berry-laden dung,
and looks UP expecting to find his prey . . .
Years ago I read that 25% of the population have an aptitude for math. The other 85% do not.
Very much akin to the old joke about how 10 out of 9 dentists prefer […].
Seems about right.
The updated webex link is below. The courtroom changed.
https://www.dccourts.gov/sites/default/files/Public-Access-to-Remote-Court-Hearings.pdf
https://dccourts.webex.com/meet/ctb132
This is part of the fraud of the Climate Change meme. Stokes continues to defend the MBH89 graph, which was flagrantly fraudulent as it purported to show tree ring data, but had thermometer records grafted in somewhere in the 1960’s, so a tree ring proxy graph which was not a tree ring proxy graph. BEST annilihated this , as did McIntyre.
Why do we re-litigate this?
Stokes is paid to lie.
Moon
Nick Stokes tries to play the role of the “temperature data police” on this site. Whenever there’s a post about USCRN, hockey stick, uncertainty, UAH, etc., he’s here in a heartbeat!
I thought old mate Nick would be in a DC courtroom called as an “expert witness” by Mann!
Michael Mann took the stand for the third day at the trial of his own making to start Week 3. Mark’s cross-examination continued in full force today. By the end of the day, Mann’s lawyers were clearly flustered as illustrated by their attempts to rebut Mann’s own testimony (in vain). C.S. Lewis (yes, that C.S. Lewis) once said, “Hell begins with a grumbling mood, always complaining, always blaming others.” We witnessed a lot of blame in Room 132 of the DC Superior Court today, so hell must be near.
https://www.steynonline.com/14056/the-emperor-has-no-clothes
Thanks for link.
It reports Mann as saying “[Judith Curry] is what I would a call a serial misinformer when it comes to science.”
Make of that what you will!
It also showed that Mann, whose statistical analysis of the hockey stick is supposedly beyond reproach, is unable to do a statistical analysis on his own grant funding or applications, getting 7 out of 12 completely wrong (one by over $3 million) between 2020 and 2023. Mann may end up getting sentenced for perjury at this rate.
Anthony, the internet finds things.. maybe you can help!
I bet that stare ha nothing to do with Steyn´s article, but a very personal reason related to one of Mann´s insulting emails, a brother of a Nobel price winner maybe?
What a wordsmith and orator Mark Steyn is…such wit, such dripping sarcasm. I am in complete awe of his mastery of the written word and his ability to deliver same.