Climate Audit Comedy of Errors

Eli Rabett chronicles the Climate Audit comedy of errors.

One consequence of the error in the RSS satellite data is that the global warming skeptics switched to using RSS, and now they can’t switch back without making it look realy obvious what they are up to.

Update: McIntyre has a new post where he claims:

In the same post that Rabett criticized here, as originally written, I had incorrectly missed a comment in Hansen et al 1988 saying that Scenario B was the “most plausible”, an error which I picked up about 8 hours after the original posting (about 9 am EST) and immediately corrected it when I noticed it. So there was an actual incorrect statement at CA for about 8 hours. Imagine that. I didn’t post up notice of the change until about 9 hours later (I was playing in a squash tournament and had to do some chores and went out after making the correction and posted the notice when I returned.) Meanwhile, a few hours after I made the correction, Lambert wrote a post on this error without mentioning that the error had already been corrected as at the time of his post.

McIntyre, of course, does not link to my post so his readers can check to see what I actually wrote.

It seems rather unlikely that McIntyre corrected the post at 9 am. At 11am he posted a lengthy comment where he continued to argue that A was Hansen’s primary scenario. How did he do this after he supposedly went out for the day? At 11:29 am Roger Pielke Jr posted a comment where he pointed out that Hansen’s paper did say that B was the most plausible. Why we he do this if the posted stated this?

It is possible that McIntyre made the correction in the hour between when Pielke posted his comment and when my post went up and because McIntyre did not mark it as a correction I did not notice that he had changed the text. But if I had noticed such as a unmarked correction, I certainly would have commented on it in my post.

Comments

#1 Lance
January 30, 2008

John Mashey,

Arguing endlessly about lines with big error bars, for 10-year-old papers, is just a waste of time and a diversion

Well, if that paper is seriously flawed and the centerpiece of an argument for changing the entire world’s energy budget, maybe not so much.
#2 luminous beauty
January 30, 2008

Lance,

The paper isn’t seriously flawed and it isn’t the centerpiece of AGW. Paleoclimate reconstructions are useful as an empirical estimator of climate sensitivity and internal variability. If past unforced variability is greater than our best estimates (Loehle is interesting, but a long, long way from becoming a reasonable, much less best, estimate) it suggests we should expect future AGW effects to be even stronger than we now do. AGW isn’t an argument for changing the entire world’s energy budget, but changing human energy resources over to renewable and sustainable resources.
#3 Hank Roberts
January 30, 2008

Excellent post.

[Email sent to webmaster asking how to repost the link from the timestamp without the blog software mangling it!]

— that one is worth referring people to. Someone should elevate it to a main item on a climate blog. John Mashey’s made it very clear.
#4 Lance
January 30, 2008

luminous beauty,

As a scientist the behavior of many supporters of AGW I find most disturbing is the ability to enlist any possible datum as proof of dangerous climate change.

Your statement…

If past unforced variability is greater than our best estimates (Loehle is interesting, but a long, long way from becoming a reasonable, much less best, estimate) it suggests we should expect future AGW effects to be even stronger than we now do.

…employs a twist I must admit I hadn’t anticipated. It presupposes that even if it can be demonstrated that increases in temperature of the same magnitude or greater than that of the last hundred years are consistent with natural variability it is actually worse news for the planet rather than evidence that nothing unusual is really happening.

Wow, I didn’t see that coming. I took the more conventional view that if the natural record showed that similar increases in temperature had occurred without man made CO2 that maybe the manmade CO2 might just be noise in comparison to the much larger natural signal.
Your strategy is pure genius!

Hey, it occurs to me that even a decline in temperature will not be enough to offset your bias, since you could always claim that it would be a whole lot colder if the natural underling variability were the only signal.

It’s the old heads I win tails you lose gambit. It’s hard to falsify this kind of theory. Good political strategy but lousy science.
#5 dhogaza
January 30, 2008

As a scientist

You’re not a scientist, you’re a drop-out. A very dishonest one, to boot.
#6 dhogaza
January 30, 2008

Lance, as a self-proclaimed scientist, you actually think that E&E is a credible source for anything other than politically-driven screeds? Despite the editor’s stated publication policy?

As I said. Dishonest.
#7 elspi
January 30, 2008

Lance
See if you can follow this reasoning.

CO2 absorbs radiation in the colder infrared but not visible radiation (100 year old physics)

Thus adding CO2 to the atmosphere will insolate the earth from the cold that is outer space without decreasing the amount of heat coming in from the sun.
(like putting another blanket on the bed). The amount of warming from the CO2 released will be 3 C direct (without feedbacks)

It is know that the historical swings from glacial to intra-glacial periods were cause by slight changes is the wobble of the earth on its axis. Such changes had a direct affect of about 1.5 C. It is the feedbacks that gave us the significant changes.

You are arguing that the feedbacks are much larger than previously thought. That means we are screwed, and your best bet is praying for the rapture.

In order to disprove Global warming, you have to disprove the 100 year old physics. You might as well try to disprove the law of gravity.

Perhaps, Lance, you should give up on the whole PhD in Physics thing and practice your
“would you like fries with that”
#8 dhogaza
January 30, 2008

Try to keep current, and maybe read something not prescreened by the climate priests at RC. The link you post to is out of date. Loehle took the criticism of Gavin, and others, and updated his study.

“prescreened by the climate priests …”

This insulting description of scientists with a huge number of published papers (unlike our resident drop-out physicist-wannabee) is being applied to folks who Loehle admitted had uncovered substantial errors in his first effort.

Hilarious.

It’s interesting how denialists like our wannabee-physicist will deny the work of thousands of scientists while eagerly touting a single paper which has already had to be corrected once as being gospel.

Why does he do so? Because he likes the political implications of the single paper he thinks is “right”.
#9 dhogaza
January 30, 2008

Also, Lance dishonestly claims that

Loehle took the criticism of Gavin, and others, and updated his study.

The truth, as usual, is not quite as Lance claims…

Loehle has issued a correction that fixes the more obvious dating and data treatment issues, but does not change the inappropriate data selection, or the calibration and validation issues.

I did a quick skim and didn’t even see any acknowledgement of other problems having been identified.
#10 Rattus Norvegicus
January 30, 2008

And the interesting thing is that Loehle’s corrected reconstruction only extends to 1935. Why is that a problem? There has been about .5C warming since then which puts current temps above his MWP (which he claims is about .3C above 20th century values).
#11 Lance
January 31, 2008

dhogaza,

The only reason you know that I have not finished my PhD is because I told you. How nice of you to constantly throw it in my face.

What scientific credentials do you posses? Of course this is playing a meaningless game. Appeals to authority are illogical.
#12 Hank Roberts
January 31, 2008

dhogaza, check the posting over at ‘in it for the gold’ a while back on not getting provoked. You lose if you start speaking to the unchanging posters instead of to tomorrow’s readers who find the thread.

Or as a friend put it long ago:

Don’t get flushed even if they yank your chain really hard. Once you do that they will just keep using you.
#13 Lance
January 31, 2008

So Hank, it’s my behavior not dhogaza’s that is the problem on this thread? Have I called anyone a liar lately or used foul language or impugned anyones character? Do I continuosly substitute personal invective for rational discussion?

You may want to step back and take a good look at the company you are keeping Mr. Roberts, or do you condone reprehensible behavior so long as you are in agreement with the politics of the offender?
#14 TSK
January 31, 2008

Thanks John.
——-
Arguing endlessly about lines with big error bars, for 10-year-old papers, is just a waste of time and a diversion …
——-
I do not question AGW, so we don’t need that. What really counts is that Tim Lambert started this thread with an attack on CA. Steve McIntyre is widely vilified and painted as a crank. The reason is obvious: He attacked Manns “hockey stick graph” and raised a blog in which he pursues his agenda.
The question is what technical arguments can be raised against him. During this whole time on this blog (many,many posts) *noone* was able to come up with a proven lie or data fabrication. TCO said he and others “pushed him where Steve evaded” but was not able to give links.
Feeble at all.
Now lets get back to the graphic in IPCC TAR 2001. I searched for calibration points, severe deviations (“volcanic winter”) which can be used to compare historical records with the graph. But to my disappointment the severity of the climatic influence can be diminished by multiproxies, historical exaggeration etc. etc.
The winter does show up on the different climatic graphs, but in different severities. So no cigar on my part.

Lets look at the graph 2.20.
You not only see Manns reconstruction, but also Briffa and Jones. Mann added a linear trend curve from 1000 to 1900, right ?
This trend curve is declining, indicating that until 1910 the temperature drops.
The standard error area is not (as in physics) simple deviation (68%), but double deviation (95%), meaning that Mann is very cautiously and contrary evidence should still be inside that area.

What is bothering me and which is not so explainable by “jiggles” is that with the exception of MBH99 other curves
go consistently *upward*. 1850 is the last time cooling occurs there before the 1950s. While Briffa scrapes the (2dev !) error area in 2.21, Jones even drop outside it.
And Jones is not alone:
Esper 2002 outside
Moberg 2005 outside
Darrigo 2006 outside
Hegerl 2006 outside

Oerl, Juckes and Huang are climbing steadily in contrast to Mann. I know that Hadcru indicates a low in 1910, but it is not or very hard discernible by the proxy data.

Are you still ready to put your head on the line that Mann has committed only neglible errors which don’t affect *his* outcome ?
#15 TSK
January 31, 2008

Oops.
The first two lines were cited with “—” above and beyond,
they are deleted. The part between “Arguing” and “diversion…” is from John.
#16 Hank Roberts
January 31, 2008

A meta moment on formatting links, just in case the problem’s just in this thread.

Reposting a link I tried earlier to test whether it still gets broken when pasted in:

http://scienceblogs.com/stoat/2007/05/the_significance_of_5_year_tre.php

(Trying to demonstrate the problem for the software’s help people, who aren’t seeing it)

Hmmm, looking back, the place I tried to post a URL, and another place where Barton tried to post a URL, both of us thought our links were broken and rewrote them.

And both originals work now — both show as the word “link” in the text; there’s HTML formatting behind the page visible with ‘View Source’that adds ‘nofollow’ — is the software coming along after us and cleaning up, with a time lag? I understand the reason, I’d do it myself, but it seems to be happening automagically. Sufficiently advanced blognology?
#17 TCO
January 31, 2008

TSK:

I can supply links.
#18 Hank Roberts
January 31, 2008

> Mann has committed only neglible errors
> which don’t affect his outcome ?

That’s what the people the House Energy Subcommittee brought in, both the independents and the National Academy of Science, said. It’s time to move on, they also said.

You appear to be stuck in, er, neutral. Metaphorically.
Why is it so hard to look past the earliest work?

Do you have complaints about the early work in any other science subject area? Ocean pH? Hydrology?

Are you saying that the first paper with Mann’s name on it either had to be so good that he should have stopped doing research and publishing, or done something besides continuing to improve on his subject?

Have you looked at the sequence of publications? At all?
By him or any other scientist you care to talk about?

Tell us why this matters so much to you. If you must.

Do you understand how science works?
#19 bi
January 31, 2008

“…the depths to which people will sink in an effort to protect their fragile worldview from unwelcome scrutiny.” — Lance

“Have I called anyone a liar lately or used foul language or impugned anyones character?” — Lance

Enough said.

Oh, and I thought the denialists (um, “skeptics”) keep saying that climate models can be made to fit any data. So why’s it that the same denialists are saying they can’t get Hansen’s model to fit the historical climate?

Maybe it’s because the denialists are full of junk, as they’ve always been?
#20 Rattus Norvegicus
February 1, 2008

Lance, I see you do not have an answer to my comment about the Loehle’s corrected reconstruction. Is this because it does not fit with your preconcieved notions? If this is the best that the contraian reconstructions can supply aren’t your arguments about paleo recons fairly weak? Not to mention the contemporary evidence all around you; melting glaciers, procession of the seasons, movement of ecosystems to a more northerly (or southerly, depending on your frame of refernece) climes. These and many other symptoms of anthropogenic climate change are pretty much undeniable. There are more disturbing observations too, would you deny all of the changes which are being observed?
#21 dhogaza
February 1, 2008

Lance sayeth:

As a scientist the behavior of many supporters of AGW I find most disturbing

Then whines:

The only reason you know that I have not finished my PhD is because I told you. How nice of you to constantly throw it in my face.

Lance, if you quit claiming to be a scientist, I’ll quit telling people you are NOT a scientist. Because I only do so to make clear to people that you have a tendency to misrepresent your credentials.

Let’s face it, honesty is not your strong point.

Deal?
#22 Lance
February 1, 2008

dhogaza,

I have completed my BS and have enough credits to qualify for a masters degree, if I chose to apply. I teach at university, all be it part time at the present, and have been collaborating on research that will probably be published in the next year.

I believe that I can safely call myself a scientist.

I believe I could call you many things, but honest is not one of them.
#23 Lance
February 1, 2008

Rattus,

I would be glad to discuss Loehle’s paper with you if you had taken the time to actually read it and understand it.

It is clear that you have done neither.
#24 dhogaza
February 1, 2008

Well, my academic background’s pretty much identical, though I never entered a PhD program, but did take every graduate CS course offered at the time by my university (it was a long time ago, that would be hard to do today).

I chose to found my own compiler technology company instead.

But I’d never represent myself as a “computer scientist”. Lambert’s a computer scientist, I’m a grunt. By choice.

And, if I were in your shoes, I certainly wouldn’t represent myself as a “scientist”. Though come to think of it, I’ve done plenty of field work and have reviewed a couple of published papers in ornithology, being named as a result.

Maybe I should puff myself up, just as you do.
#25 Tim Lambert
February 1, 2008

Hank Roberts:

>And both originals work now — both show as the word “link” in the text; there’s HTML formatting behind the page visible with ‘View Source’that adds ‘nofollow’ — is the software coming along after us and cleaning up, with a time lag? I understand the reason, I’d do it myself, but it seems to be happening automagically. Sufficiently advanced blognology?

Alas, no. I fixed them manually. Please do not put bare URLs in comments. Instead, turn them into proper links. Eg, type

Use [markdown](http://daringfireball.net/projects/markdown/dingus)

which will give you:

Use [markdown](http://daringfireball.net/projects/markdown/dingus)
#26 Hank Roberts
February 1, 2008

Ah, thanks Tim. I was reading “You may use … for style”
(as an option to prettify plain text).

I now read that as “Thou shalt use …. so thy link worketh without Tim’s cleaning up after thee” — and glad to be told.

“You may” use markdown to avoid screwing up posting links.

Got it. Lesson learned. Clue appreciated.
#27 lucia
February 2, 2008

Lance said:

Well, if that paper is seriously flawed and the centerpiece of an argument for changing the entire world’s energy budget, maybe not so much.

No Lance. Paleo-reconstructions of temperature, can, at most, be supporting planks used to test a theory. Moreover, statistical test related to individual hypothesis, so:

1) Whether right or wrong, Mann’s results would only test hypothesis about the historical temperature record, not specifically the cause for the record.
2) The centerpeice of for the theory of Anthropogenically induced Gobal Warming (AGW), an argument based on physics , and physical chemistry (and I suspect partly biology. But I know nothing about biology.)
3) There are many individual sub-theories and hypotheses that are thought to be correlaries of the over all theory. These get tested against data.
4) No test of an individual hypothesis becomes the centerpiece. (Though some convince more people.)

The issue at any point in science, whether climate science or engineering science or social science is generally decided by the balance of the evidence. Currently, the balance supports the theory of AGW, and rather strongly.

Could this change? Possibly. This isn’t quite as strong as Newton’s laws of mechanics that, though modified by Einstein, still remain intact for most purposes. But, the balance of the evidence is that recent warming is caused by human activity.

I have a gripe in the various blog discussions is that, collectively, (not necessarily individually) both sides appear embrace error bars when its convenient and deny the other side similar use of the error bars. But this behavior happens in all fields– not just climate science. (I also have a gripe with labels like “climate priest”. It would be best if you cut that out. )

Martin: Yes. There is something dishonest about cherry-picking. It should never be done intentionally, and one should scrutinize ones data choices to be sure it’s not done unintentionally.
#28 Martin
February 3, 2008

Lucia, of course it is always wrong for a scientist undertaking research to “cherry-pick” from their results so that they get the conclusions they want. That is serious misconduct.

On the other hand the same expression can also be used to describe the normal and legitimate practice in which an advocate for a policy in a film or a letter or a speech will choose examples and data that best illustrate the point they wish to make. It was in the context of advocacy – in the letter to the UN, and An Inconvenient Truth – that I said that cherry-picking is not inherently dishonest. I am surprised that you do not see the point I was clearly making – others appear to be being obtuse in order to make a silly debating point.
#29 Hank Roberts
February 3, 2008

Martin, not credible.

Claiming your work is “advocacy science” does nothing to legitimize lying. You’re cherrypicking to try to justify cherrypicking.

Take a lie, say it’s advocacy; say advocacy is puffery; say puffery is okay because nobody expects it to be credible; present puffery as scientific information.

Wrong. Get it?

You may not realize this. You’re parroting an old, discredited, failed industrial PR idea about how to manage science in court cases.

Look it up, please. This is much written about, and different court systems in different countries approach the issue in different ways. And looking at the outcomes you can see which approach works in practice.
#30 dhogaza
February 3, 2008

Earlier, Martin said…

It is one thing to say that a scientist is cherry-picking, or is unreasonably reluctant to share data – but neither of those accusations are anywhere near fraud.

Now martin says, after being beat up frequently, that …

Lucia, of course it is always wrong for a scientist undertaking research to “cherry-pick” from their results so that they get the conclusions they want. That is serious misconduct.

A totally different story, which he justifies by saying …

It was in the context of advocacy – in the letter to the UN, and An Inconvenient Truth – that I said that cherry-picking is not inherently dishonest.

which just proves his inherent dishonesty.
#31 Martin
February 4, 2008

Dhogaza said about Climate Audit “that they continuously say that the NASA climate scientists are guilty of fraud”. Despite being unable to find a single example of Climate Audit even once accusing someone of fraud, he has failed to admit that he was wrong. In his defence he has tried to argue that “misconduct” is identical to “fraud”, but this is not going to convince anyone with either a knowledge of the English language or a dictionary.

Dhogaza then argued that the statement “there has been no net warming since 1998″ is equivalent to “global warming stopped in 1998″. Since a true statement cannot be equivalent to a false statement that was another elementary error in logic; but again he has yet to acknowledge his error and move on.

Now he is tying himself in knots in an attempt to blur the difference – which is obvious to anyone – between a scientist deliberately cherry-picking data in a piece of research, and an advocate for AGW choosing facts that best illustrate his case in a letter, film or speech. Why does he bother? How can he imagine that anyone takes his posts seriously after his refusal to admit errors?
#32 Hank Roberts
February 4, 2008

Bye, Martin.
#33 Lee
February 4, 2008

Martin,

Dhogaza is often over the top, and sometimes pisses me off with his somewhat shrill responses. But he is not being dishonest – you are.

McIntyre has made overt allegations of misconduct. He insinuates fraud continually – he posts stories about proven frauds, and throws in side comments in those same threads about ‘the team’ over and over. You have been shown this. In return you keep offering this absurd parsing of whether it is fraud or misconduct being alleged. Stop making a fool of yourself.

There has been net warming since 1998. 2005 was hotter. Even absent that, though, the choice of 1998 is cherry picking, the intent is to hide the fact that regression, even including the cherry-picked start point, shows continuing warming. That statement by that person is an overt and intentional attempt to make people believe something that the speaker knows not to be true. That is dishonest. As are you, for defending it.

Cherry picking is dishonest whether for ‘scientific’ or ‘advocacy’ purposes. What you are arguing in effect is that it is ok to be dishonest for advocacy, and that doing so is not dishonest. Crap. People ARE often dishonest when doing advocacy – that doesn’t make them less dishonest, it only makes the dishonesty relatively common.

Martin, please stop defending dishonesty.
#34 Chris O'Neill
February 4, 2008

Martin:

How can he imagine that anyone takes his posts seriously after his refusal to admit errors?

What a hypocrite.
#35 TSK
February 7, 2008

[Mann has committed only neglible errors which don't affect his outcome ?]

That’s what the people the House Energy Subcommittee brought in, both the independents and the National Academy of Science, said.

I was seriously peeved, so I waited until my wrath cooled down before answering. Please read the NRC report, chapter 9 pp.90 ff. and chapter 11, p.112 ff.
The NRC report does not vindicate Mann, it actually acknowledged (!) McIntyres criticism. They confirmed AGW because other people got similar independent curves, not because Mann was correct. If I use an ill-advised method, but get accidentally a correct result, is that an, erm,
*vindication* ?
Who of you pseudoskeptics actually read the NRC report ?!

You appear to be stuck in, er, neutral. Metaphorically. Why is it so hard to look past the earliest work?

Because I have other reasons to ask here than informing myself about the latest progress and cognitions in climate science which are a whole different area. This article don’t discuss the newest findings, it specifically attacks the ideas of a specific person.

Do you have complaints about the early work in any other science subject area? Ocean pH? Hydrology?

Now that sounds for me that climate science wasn’t invented before 199x. But it builds on relative stable areas of knowledge like meteorology, geology, hydrology, statistics
and dendrochronology which are much older. The problem is not errors; errors have been rampant since the invention of science and noone couldn’t be hold responsible for small or
big errors.
But the way how scientists handle criticism and how they acknowledge errors is a whole different area. And here I notice unfortunate shortcomings.

Are you saying that the first paper with Mann’s name on it either had to be so good that he should have stopped doing research and publishing, or done something besides continuing to improve on his subject?

Have you looked at the sequence of publications? At all? By him or any other scientist you care to talk about?

Do I sense hurted feelings that I dare to decry the work of a scientist which gave so much valuable work to its field ? Are you aware how much self-restraint and reluctance many
colleagues of you displayed while badmouthing and ridiculing criticism ? For me this reproach sounds hollow.
#36 Barton Paul Levenson
February 7, 2008

TSK, it wasn’t a question of his being wrong and accidentally getting the right answer, as you appear to believe. His method had some mistakes in it but was sound overall, which is exactly what the NAS report said. Other people, using his data and improved methods, get essentially the same results. It’s not a question of being peeved; it’s a question of anti-AGW political partisans ganging up on a scientist McCarthy-style because they didn’t like what he was saying. Your side has the narrow-minded suppressors of the truth, not our side.
#37 dhogaza
February 7, 2008

They confirmed AGW because other people got similar independent curves, not because Mann was correct. If I use an ill-advised method, but get accidentally a correct result, is that an, erm, vindication ? Who of you pseudoskeptics actually read the NRC report ?!

All of us, I suspect, and the report and follow-on comments by the authors who make it clear that you are grossly misrepresenting the report.

The point about the statistical analysis was that it wasn’t the BEST way to analyze, but not “wrong” in any material sense. And this was in view of hindsight. Denialists like yourself take “not the best approach” to mean “junk”, “fucked up”, “broken”, etc, which is not the case here.

Mann’s PCA approach gave reasonable results. The “best” approach gave – essentially the same results.

Case closed.

It’s a fucking bitch that you continue with this misrepresentation that’s so egregiously wrong that members of the NRC committee felt compelled to point out that their conclusions were being MISREPRESENTED by the denialist community.

And still are. By the likes of you.
#38 z
February 7, 2008

“Oh, and I thought the denialists (um, “skeptics”) keep saying that climate models can be made to fit any data. So why’s it that the same denialists are saying they can’t get Hansen’s model to fit the historical climate?”

Good point. Right up there with “There is no warming, it’s an artifact of sloppy measurement techniques, and it’s caused by the sun”.
#39 hengav
February 8, 2008

“Mann’s PCA approach gave reasonable results. The “best” approach gave – essentially the same results.
Case closed.”

C.O.S.

M.W.P. N.I.C.
S.M.D.
QED
#40 TSK
February 8, 2008

Before anyone here further embarasses yourself, READ the report. READ it NOW. Don’t give a damn about what press reports, academics or whoever claimed what is standing there.
Read the page numbers I cited before:
The link is [here](http://books.nap.edu/openbook.php?record_id=11676&page=R1)

If you can’t understand what it says, then don’t argue about it. If you still convinced that the report reports
a “overall sound methodology”, “neglible errors” etc. cite the relevant passages directly from the report. Thank you.
#41 Jeff Harvey
February 8, 2008

Before TSK further embarasses him/herself, I suggest he looks at the lecture presented on Tim’s thread by Naomi Oreskes. It pretty well demolishes the arguments of the sceptics by showing how they have massively politicized the issue by twisting and mangling the underlying science to promote their own agendas, and shows that concern over the burning of fossil fuels in driving climate change go back more than half a century. She highlights research by the likes of Revelle and Keeling in the 1950s and how they forewarned of the consequences of pumping huge quantities C0 2 into the atmosphere. This warning was repeated many times in following years, well before James Hansen spoke up in the late 1980′s.

TSK, you are doing exactly what the sceptics want you to do: focus your wrath on one small piece of evidence in a gigantic scientific puzzle in which the evidence for AGW is massive. Concern over AGW pre-dates the proxy reconstructions of Mann et al. by many years. But if the think tank brigade and their foot soldiers can convince the public that the ‘science isn’t settled’, and can do so by getting the media to hammer away at this one small area, then they will say, “See?! The science isn’t clear, and thus AGW isn’t necessarily happening or even if it is it isn’t serious or due to human activities, so we don’t need to do anything about it!”. You know what? I think they stink. They and their scientific distortions.
#42 stewart
February 8, 2008

For TSK, Martin, and others, who seem to think that less than optimal choices doom a paper, I offer a quite from Andrew Vickers, who wrote a recent Medscape letter addressing their viewpoint. ‘Just like any other
science, what you want to know about any statistical technique is the degree to which it might give you a wrong answer, and whether there other methods around that give you a better chance of getting things right. There aren’t rules, laws, and commandments about this; you just have to know the latest research data.’
#43 Tim Lambert
February 8, 2008

TSK, I *have* read the NRC report, and you are misrepresenting it. Mann’s “off-centre” PCA makes a difference to the first principle component in one step of his method, but does not make a difference to his reconstruction. McIntyre likes to give you the impression that it does, in order to make it sound like he broke the hockey stick. This is the same approach that he used to make the error he found in GISS temp made a big difference to the GISS temperature record (“1998 is no longer the warmest!”). Look at the sequence of errors this post was about and notice which way they all went.
#44 cce
February 8, 2008

RE: PCA

“As part of their statistical methods, Mann et al. used a type of principal component analysis that tends to bias the shape of the reconstructions. A description of this effect is given in Chapter 9. In practice, this method, though not recommended, does not appear to unduly influence reconstructions of hemispheric mean temperature; reconstructions performed without using principal component analysis are qualitatively similar to the original curves presented by Mann et al.”
http://books.nap.edu/openbook.php?record_id=11676&page=113

Translation: “negligible error,” and that’s if you stretch “not recommended” into “error.”
#45 TCO
February 9, 2008

Tim: I (basically) agree with your assessment of McIntyre and the acentering. I think it has been very shameful how he has tried to resist answering questions about overall reconstruction versus first moment, how he has conflated acentricity with covariance/correlataion. I have tried pinning him down on those issues and he has danced like a tingling school girl. That’s shameful.

Note: Although I am a denialist skeptic, I think it’s shameful for my side to use sophistry. I also find it wrong that Mann refuses to answer basic questions on the acentering itself (if it was deliberate or a mistake, etc.) It’s very minor individual item…but it shows me a tendancy of Mann. In a weird way, they are sort of similar.

I could indulge in some trollish comments about what I’d like to do to each of them, but you would get all prissy and disemvowel me.
#46 TSK
February 15, 2008

Sorry for the long delay, but I wanted to avoid to be long-tongued and keep it short.
Under “Spurious Components” p.90 you will find that the authors acknowledge that the usage of principal components likely introduce artefacts (The experiments by McIntyre & McKitrick reproduce the peak artefact in the population eigenvector almost always).
So we know that there may be a problem.

The main statement is on p. 95:

Mann et al. (1999), gives CE values ranging from 0.103 to -0.215, depending on how far back in time the reconstruction is carried.

A CE of 0.1 and lower without a given r^2 (I couldn’t find any in the report) is garbage and that was the passage which caused my anger. So you don’t need to focus on the PCA method, Tim. The authors did know that because
they are writhing:

Reconstructions that have poor validation statistics (i.e., low CE) will have correspondingly wide uncertainty bounds, and so can be seen to be unreliable in an objective way. Alas ! Really ??!
Even a low CE value may still provide prediction intervals that are useful for drawing particular scientific conclusions. What did *they* mean with “may” ? Most likely, likely, perhaps or snowball chance in hell ?! If a scientific text is extraordinarily vague in this context, you may assume the worst case.

So what to make out of cce’s citation ?
The point is that the argumentation “reconstructions performed without using principal component analysis are qualitatively similar to the original curves presented by Mann et al” does *not* address the quality of Mann’s work.
This is easy to see if you imagine that a pupil tells his teacher: Sir, my calculations on the math problem are right because it matches the correct result.
No, it isn’t a valid argumentation. Errors may cancel out or you get a right result by pure luck. This is especially
the case if you are demanding “qualitatively similar” results. So it’s ok to throw darts on a paper as long as they resemble a hockey stick ? Obviously not.

And why do the panel writhe again in such vague and unscientific statements ? There *is* an objective value how similar a curve is to other curves: cross-correlation. And you may draw your own conclusions why they didn’t applied it.

What the report says, is effectively: “Mann did commit severe errors, but as the main point is the sharp increase of temperature which is validated by other scientists we won’t indulge too deep in it.”

@Jeff: Your lecture of political usage is not new. Any skepticism will attract deniers; find a valid point of criticism of neo-darwinism and you will be quote-mined by creationists. The problem which *many* pseudo-skeptics have is that they must acknowledge the quality of their opponents and react accordingly. If valid criticism (even supported by a tendency to deny) is painted as cranky and the public find out that it is *not* cranky, the credibility
of the scientists is damaged. Knee-jerk excoriates from scientists who might be able to see the validity with a keener look are self-damaging (Remember this ludicrous rat-race of “best science blog” with totally incompetent statements of Myers ?)

So dismiss something if it is foreseeable that it can easily be refuted, but be careful if there are valid points.

(Again so long….sorry)
#47 bi
February 15, 2008

TSK:

The main statement is on p. 95: Mann et al. (1999), gives CE values ranging from 0.103 to -0.215, depending on how far back in time the reconstruction is carried.

On what grounds do you say this is the “main” statement? Because you want it to be the main statement?

This is easy to see if you imagine that a pupil tells his teacher: Sir, my calculations on the math problem are right because it matches the correct result. No, it isn’t a valid argumentation. Errors may cancel out or you get a right result by pure luck.

Except what our scepticos are doing is to try to say, “Hey, this pupil used a totally wrong method to find the hypotenuse of this right-angled triangle, therefore Pythagoras’s Theorem is false! And… Pythagoras was a Saddam sympathizer!”

Any skepticism will attract deniers; find a valid point of criticism of neo-darwinism and you will be quote-mined by creationists.

Oh yes, and any charge of “denialism” will immediately attract suggestions that there’s an invisible, ethereal, unstated Valid Criticism somewhere out there, except somehow it’s suddenly the AGW/evolution’s duty (rather than — Heaven forbid! — the duty of the critics) to find this Valid Criticism.
#48 TSK
February 17, 2008

On what grounds do you say this is the “main” statement? Because you want it to be the main statement?

It is the only *quantative* evaluation of the article in question which I have found there; therefore it is the “main” statement.

Except what our scepticos are doing is to try to say, “Hey, this pupil used a totally wrong method to find the hypotenuse of this right-angled triangle, therefore Pythagoras’s Theorem is false! And… Pythagoras was a Saddam sympathizer!”

Both finding a fault or starting personal attacks don’t influence the truth value of a purported statement and you are free to point that out. Neither does a possible reconstruction fault influences the steady rise of global temperatures and CO2 levels nor the worrisome temperature anomaly in the vicinity of the poles.

Oh yes, and any charge of “denialism” will immediately attract suggestions that there’s an invisible, ethereal, unstated Valid Criticism somewhere out there,

In the timeframe between 19th and 20th century there were “valid criticisms” of atomic theory by Mach and Ostwald which are sounding ludicrous today, but which were bitterly defended by scientific authorities. So, being aware that neither I nor you are aware of our blind spots, how can you be convinced that you can spot a “Valid criticism” if it walks right in front of your eyes ?

But it does not matter. Simply say: “Fine, it may be out there, but as long as we haven’t found out we stick with our current knowledge.”

except somehow it’s suddenly the AGW/evolution’s duty (rather than — Heaven forbid! — the duty of the critics) to find this Valid Criticism.

Says who ? I surely not. Let us leave AGW/evolution as examples and speak more generally: The problem is less that criticisms does not exist, the problem is that you must find possible pearls under the garbage and all of them are judged by default as invalid. :->
Having not enough data to decide, not able to see the implications, being convinced by a compelling counterargument which is unfortunately invalid…the reasons that possible “valid criticisms” are overlooked are hundredfold. We are exactly as stupid as all our predecessors.
#49 z
February 18, 2008

“how can you be convinced that you can spot a “Valid criticism” if it walks right in front of your eyes ?”

yes, but on the other hand, one can identify invalid criticisms; those which are internally contradictory, those which contradict known facts, those which contradict things which are not facts but are taken as axiomatic (e.g., conservation of energy), those which appear to be completely irrelevant to the thing which they are supposed to criticize, etc. narrows the field down quite a bit.