Saturday Night Ice

Its the middle of July, temperatures should soar into the 90's tomorrow, so naturally we want to talk about hockey, or anyway, hockey sticks.  And global warming, natch.

Following our "Friday Night Ice" from last February, we have more on the Mann "hockey stick", which purports to show Earth's temperature as stable for most of the last millenium and soaring in the last century (a visual on p. 3).

Critics questioned the statistical methodology; now the House Energy Committee has released the long-awaited Wegman Report.  Some excerpts from their factsheet:

Questions about the reliability of the Mann studies were of interest because they raised policy-relevant questions concerning the objectivity of the IPCC and its reliance upon and “promotional” use of the studies’ ‘hockey stick’ shaped historical temperature reconstruction.

Following receipt of the letter responses, committee staff informally sought advice from independent statisticians to determine how best to assess the statistical information submitted. Dr. Edward Wegman, a prominent statistics professor at George Mason University who is chair of the National Academy of Sciences’ (NAS) Committee on
Applied and Theoretical Statistics, agreed to independently assess the data on a pro bono basis. Wegman is also a board member of the American Statistical Association.

About the Wegman committee: Dr. Wegman assembled a committee of statisticians, including Dr. David Scott of Rice University and Dr. Yasmin Said of The Johns Hopkins University. Also contributing were Denise Reeves of MITRE Corp. and John T. Rigsby of the Naval Surface Warfare Center. All worked independent of the committee, pro bono, at the direction of Wegman. In the course of Wegman’s work, he also discussed and presented to other statisticians on aspects of his analysis, including the Board of the American Statistical Association.

Among the panel’s findings and recommendations:
• Mann et al., misused certain statistical methods in their studies, which inappropriately produce hockey stick shapes in the temperature history. Wegman’s analysis concludes that Mann’s work cannot support claim that the1990s were the warmest decade of the millennium.

So, fill me in - is Wegman credible, did the House Reps take this committee into the tank, what?

John Quiggin at Crooked Timber looks at the other aspect of the report, which dealt with a possible breakdown in the peer review process.

Comments

[SteveMG]

Well, Mr. Quiggin dismisses the report as another example of the "Republican war on science" and yet nowhere does he identify the political affiliation of any of the authors of the study.

Or whether their affiliation (again unknown) affected their scholarship.

And how.

Apparently, science is getting hammered from folks on both sides.

SMG

[Lew Clark]

Well sweating my ass off in 100+ temps in North Texas, I am convinced that global warming is a fact. Of course, when the first "blue norther" hits about late November, I will swing to global cooling and imminent ice age. Because I'm not only a fair weather friend, but, am also a seasonally adjusted climitological theorist.

BTW, spell check says “climitological” is not a word, but, this is a blog. You get to make up words on blogs!

[Charlie (Colorado)]

Tom, the easy takeaway is in the short factsheet, which appears to me to be pretty well done. The gist of it is this:

(1) there is still a case to be made for global warming

(2) the statistcal methods used by Mann et al are very much suspect

(2a) they are based on fairly complex multistage inferences

(2b) the conclusions are sensitive to the assumptions made in those inferences

(2c) at least some of their methods have been shown to lead to similar conclusions, even with input of random numbers --- which means that the statistical methods bias the results.

(3) these flaws dramatically weaken the case for anthropogenic (human-caused) global warming, or more strictly for anthropogenic causes being the dominant issue.

(4) A social network analysis of the authors, data sources, and reviewers of global-warming papers revelas a very small, and closed, clique (for once, used in the strict graph-theoretic sense) which makes the "peer review" and "independent verification" aspects of the publications difficult to credit.

[ordi]

You get to make up words on blogs!

The enviro weenies must all be bloggers because IMO their theories are all made up.

[larwyn]

Thank PoliPundit for the link but
a Mr."Bill" Smith for the LEAK Hmmmmm Bill??? - troll
didn't want to use John - it begins with the telltale "J".

Now we have to keep it really quiet that K/S/S is also a "man".
Shhhhhhh.....
Mr Smith exposes TAC with "Rove Secretly Runs
The New York Times"
July 14, 2006

Rove Secretly Runs The New York Times

By Bill Smith
New York, New York (SatireNewsService) - In a stunning development that would appear to have broad implications for the independence of America's newspaper industry, New York Times Publisher, Edwin 'Pinch' Sulzberger today revealed that longtime President Bush advisor Karl Rove has been secretly running the Times' news and editorial operation for almost four years. According to well-placed insiders on the Times' Board of Directors, a shaken Sulzberger made that announcement in a hastily convened meeting of the Board of the Times' parent company, The New York Times Corp. Sulzberger reportedly told the board that the discovery was made last week.var tcdacmd="sa=a;sz=3;ad";document.write('');
"During an internal investigation, we reached the regrettable conclusion that Karl Rove has been running this newspaper since at least August, 2002," Sulzberger reportedly stated. "His intention is clear - to ruin the reputation of the newspaper and the party that our editorial policy supports."Sulzberger reportedly continued: "I ordered an investigation to determine how the Times had come to publish detailed information about a top-secret government monitoring operation of the international financial transactions of al-Qaeda and other terrorist groups. The publication of this information clearly helps an enemy that killed thousands of people just a few blocks from here. Endangering Americans is something the Times would never intentionally do. Unfortunately this story fits a pattern of publication that has almost ruined the Times' reputation for probity and journalistic honesty as well as causing incalculable damage to the Democratic party that our editorial policy supports."Edith Steingehirn, the Times' internal investigator who made the Rove discovery, told the board: "Our investigation into the publication of the terror financing story quickly led us to discover other frightening actions taken by our news and editorial divisions during the past four years."....

RTWT

[SPQR]

One thing that isn't getting enough emphasis is that Mann et al have not merely made some errors in statistical analysis but they have actively sabotaged efforts by others to reproduce and analyse their results by refusing to share details of their statistical methods, and by hiding raw data and intermediate data.

This is not a breakdown in peer review but the text book example of the corruption of the scientific process.

[larwyn]

Only semi OT as NYT is great promoter of Gore's Global Warming scare tactics.

This is beyond even our K/S/S' mindrays:

A sniper loyal to Shiite cleric Moqtada al Sadr fires towards U.S. positions in the cemetery in Najaf, Iraq.Michele McNally: “Right there with the Mahdi army. Incredible courage.”
LGF:
The Media Are the Enemy
New York Times photographer Joao Silva was right there in the room as a member of Muqtada al-Sadr’s “Mahdi Army” tried to kill American troops: The New York Times - New York Region - Slide Show - Slide Show: Memorable Photographs.

Assistant Managing Editor for Photography Michele McNally comments on this one:
A sniper loyal to Shiite cleric Moqtada al Sadr fires towards U.S. positions in the cemetery in Najaf, Iraq.Michele McNally: “Right there with the Mahdi army. Incredible courage.”

UPDATE at 7/15/06 8:11:33 pm:Here’s a site where New York Times photographer Joao Silva is hawking a book......

.....UPDATE at 7/15/06 9:02:08 pm:Jeff Goldstein comments: Sleeping with the Enemy.
Incredible courage? Well, far be it for me to question such self-congratulatory enthusiasm, but it seems to me that actual “incredible courage” would have entailed, say, Joao Silva getting word to US troops, or his bumrushing the sniper and beating him unconscious with a heavy telephoto lens.Whereas what we’ve witnessed here is the product of dangerous opportunism in the service of plaudits and cocktail party invites.
7:27 PM PDT

Please use link to see this despicable NYT (redundant)photograph.

[blubi]

I always thought the hockey stick was "too good to be true". It was the perfect result to prove AGW: no climate change in 1000 years and a sudden temp. rise. It also fit the ecodoomer´s dogma that a stable, idyllic climate is being threatened by man-made emissions.
In statistical analysis, it is all too easy to get the results you want, whether intentionally or due to lack of objectivity. However, Mann´s response towards those who have questioned his results puts his scientific honesty in question.
The hockey stick may not be dead, as there will always be believers, but it is certainly no longer part of that "consensus" we keep hearing about.

[Tim Lambert]

Tom, your post in February was about how the National Academy of Sciences had been asked to look at the hockey stick. They did and concluded that there were some technical issues where Mann et al should have done things differently but that these did not make much difference to his results.

Wegman is credible, but he was just asked to look at those technical issues and not what effect these had on the hockey stick graph. So Barton's committee gets a report they can use to argue that the hockey stick is broken. And it might have worked if it weren't that meddling National Academy of Science.

[Pofarmer]

I wish I could remember where I saw it, but I read an article a little while back that seemed pretty credible. It stated that human activity releases less than 5% of the gases attributed to global warming. So, even if there is a hockey stick, what are we going to do about it? That result could be much more attributable to urbanization around collection points, than anything else. Lot's of measurements taken in areas that remain rural don't come up with these results.

[Pofarmer]

I wish I could remember where I saw it, but I read an article a little while back that seemed pretty credible. It stated that human activity releases less than 5% of the gases attributed to global warming. So, even if there is a hockey stick, what are we going to do about it? That result could be much more attributable to urbanization around collection points, than anything else. Lot's of measurements taken in areas that remain rural don't come up with these results.

[Pofarmer]

I wish I could remember where I saw it, but I read an article a little while back that seemed pretty credible. It stated that human activity releases less than 5% of the gases attributed to global warming. So, even if there is a hockey stick, what are we going to do about it? That result could be much more attributable to urbanization around collection points, than anything else. Lot's of measurements taken in areas that remain rural don't come up with these results.

[Pofarmer]

I wish I could remember where I saw it, but I read an article a little while back that seemed pretty credible. It stated that human activity releases less than 5% of the gases attributed to global warming. So, even if there is a hockey stick, what are we going to do about it? That result could be much more attributable to urbanization around collection points, than anything else. Lot's of measurements taken in areas that remain rural don't come up with these results.

[Pofarmer]

Wow, sorry about that.

[Rick Ballard]

The NAS Report that said things warmed up a bit as the Little Ice Age ended? It's pesky all right. Not quite as pesky as Mann not coughing up the code that allows him to draw hockey sticks to order, though.

Very nice appeal to authority without addressing the validity of Wegman's assertions - which the NAS also did not address to any extent. If random numbers can be manipulated using Mann's "Secret Statistical Formula" to generate hockey stick graphs on demand, then how much "weight" should given to his research?

I'll go with Motl's take on the report.

[TCO]

TM: Thanks for blogging this. Aren't you happy to have something simpler then Plamegate! :)

Tim: The NAS panel broadened their terms charter to a general look at climate predictions and eliminated questions on frisking Mann's work. They ducked. Wegman did not. W's report actually redid calculations (see especially figures 4.6 and 4.7) of the Mann method.

The worst thing about the NAS panel "duck" was on data availability, which reflects scientific ethics and philosophy. When a reporter asked them about whether Mann shared data adequately, they said that was too "big a question for them"! This about the guy (Mann) who said releasing the algorithm for the hockey stick would be "giving in to intimidation"! We just have to trust him, you see. That's not the science that TCO learned. Homie don't do no "trust me". Homie wanta check the fine print...

Also, the NAS report is a bit schizophrenic. It can be concluded as chopping the stick down to 400 years or not (all depending on what the meaning of "plausible" is), but in any case, they certainly agreed that they could not quantify error beyond 400 years out (a rebuke to Mann, who said he could).

Also, NAS panel says that there are general problems with the tree-ring based studies and then says, "but it's not just Mann there are a bunch of studies. Say whuuuu? BTW, Wegman NAILS this by showing how the "different studies" share proxies, thus being confounded, not independent.

All: This is a big victory for Steve McIntyre. He has gotten a lot of ad hominem abuse from (a likely scared) Mike Mann for not having a Ph.D. or for having been a mining guy. But when Steve digs in and starts doing computer-based statistics or even linear algebra, stand the f** back! The old man can bring it. He's obviously got more math smarts then Mann. Just talk to each of them. You can always tell who's really smart by if they avoid specifics or try to snow you.

Now, we have a true heavy from traditional statistics showing that Steve's critiques of Mike Mann are valid. Wegman has 170 papers and is head of the NAS Applied Stats panel. He is from the heart of the academic study of hard core theoretical and applied statistics. He's got way more background then either Mike Mann or Steve M. He is the REAL DEAL. See his CV:

http://www.galaxy.gmu.edu/stats/faculty/wegman.resume2.htm

At this point, even the Mann Hockey Team "side" are starting to realize that he did make method errors and the argument is (properly) over amount of the effects and over why Mann refuses to admit it, why he wrote so opaquely to disguise it, why he resisted and resists having his work checked. Only Mann remains adamantly truculent. He still refuses to address the off-centering issue and censors questions about it on his Real Climate board.

It's a big victory for the little tailor!
READ ALL ABOUT IT! READ ALL ABOUT IT!
http://www.climateaudit.org/

[TCO]

BTW, the MSM other then WSJ and NR is hiding this. I think they are flumoxxed. Don't know what to do, are waiting for a good liberal spin to come out. Need the pajamahedin to ram this down their faces.

[noah]

I've seen intermittent double postings but I have never seen a uninterupted quad before...congrats Pofarmer!

I will take GW seriously when somebody offers a plausible plan for dealing with it that has a snowball's chance of being implemented!

[ajacksonian]

This report only looks at statistical validity of data analysis, not at actual theoretical underpinnings. They do point out that the models are insufficient in a number of areas, especially trying to account for over-sampling in the last 100 years or so. Of particular interest is this from page 49, point 10: "We note here that we are statisticians/mathematicians who were asked to comment on the correctness of the methodology found in MBH98/99. In this report we have focused on answering this question and not on whether or not the global climate is changing."

And then go on to talk about things that may be worrying. But since no one has a decent model for the Earth's climate nor can replicate the last few hundred million years of same, there is no way of knowing just *what* the contribution of carbon dioxide is on global temperature. Such a model would, of course, need to take a look at: solar output over time and its variations, continental drift rate over time and its variations, changes in the Earth's rotation due to Lunar drag and changes in tide due to overall Lunar distance, volcanism and its contribution, extraterrestrial impacts of bolides of various origins, density and types of plant life, carbon extraction and deposition via plants, coral and terrestrial organisms that do *not* leave a good fossil record... dear, me! Quite a list, isn't it? Not something so easily summed up in: 'human activity bad, must go back to caves to prevent global warming'. And this is *before* adding in epicyclic phenomena such as orbital libration and where the solar system has moved with respect to other parts of the galaxy including dust clouds.

Be that as it may, this report does nothing to address those things and still leaves me just where I always have been. We are coming out of the Little Ice Age. Temperatures are rising. This is news?

I will go back to the things that *will* happen, sooner or later, that no one is preparing for.

[Seixon]

Think Progress hasn't jumped all over this to attack it either. I think that says a thing or two, as they always jump all over anything that would seem to compromise their shilling for Al Gore.

In other words, their silence is golden.

[richard mcenroe]

If that's Australia's John Quiggin, he never met a tree he didn't want to hug. He's been an extremely uncritical backer of global warming, er, climate change, er, AGW, er, global warming... dammit, which is it this week?

[Charlie (Colorado)]

Wegman is credible, but he was just asked to look at those technical issues and not what effect these had on the hockey stick graph. So Barton's committee gets a report they can use to argue that the hockey stick is broken. And it might have worked if it weren't that meddling National Academy of Science.

Tim, your first sentence is completely false. Wegman et al were asked to look at the technical issues, and showed that the technical issues were such that Mann et al's results including the 'hockey stick' were invalid; the conditioning Mann et al used results in a 'hockey stick' when applied to random numbers (see Figure 4.1).

Again, this doesn't mean that there is no global warming; it just means that the case for anthropogenic global warming is very much weakened.

[Bob]

And what does the earth's magnetic field have to do with all of this?

http://news.bbc.co.uk/2/hi/science/nature/3359555.stm?ls>Earth loses its magnetism

We don't even know how it works, but hey Al Gore knows everything... just ask him!

[Charlie (Colorado)]

Richard:

Global climate change: the observation that climate world wide is changing, especially on kiloyear time scales. Not controversial.

Global warming (GW): the observation that global climate change has been in the direction of warmer for about 400 years (since the LIttle Ice Age.) Again, not controversial.

(Aside: I can't figure out why this doesn't cause people to collectively slap their foreheads and say "well, duh!" It amounts to the assertion that the climate has gotten warmer since the most recent time it was coldest.)

Anthropogenic global warming (AGW): the inference that global warming has been exacerbated radically by human-generated climate forcings, in particular by CO2 emissions. Very controversial, but at this point very much in question.

[David Stockwell]

"Need the pajamahedin to ram this down their faces." good one TCO.

The main implications of these revelations of higher MWP not generally realized is that the current warmth falls within the natural climatic variation. So whether it is man made or not is irrelevant so far as impacts. The environment has experienced it before, quite recently actually, and can deal with it.

[Tim Lambert]

Rubbish, TCO.

Wegman's results are basically a subset of the NAS/NRC report. In particular, both reports find that "off-centre" method used in Mann Bradley and Hughes' 1998 paper (MBH98) tended to produce hockey stick shapes in the first principal component (PC1). Unfortunately, Wegman stop there and do not address the question of what difference this makes to the reconstruction (which is not the same as PC1). The NRC panel did address this question and found that it made little difference.

Over at my blog you tried to claim credit for making the distinction between PC1 and the reconstruction at climateaudit, but it seems that over here you're happy to spread confusion on the issue. What's your excuse?

And don't believe TCO's spinning of the findings of the NAS/NRC. Read the press release.

[TCO]

Tim, I agree that quantitative measurement of the effect on the reconstruction needs to be done (not just the PC1). I have pushed Steve on this. The NAS Panel did not do such a quantitative calculation either. If they did, what's the number!?

The Wegman report finds other problems with the MBH work. The trending series of data (autocorrelated) prevents significance of the validation. Therefore the work is not statistically significant.

In addition, the Wegman report points out the inadequacies of tree ring data and that the "other reconstructions" using tree rings, share the same proxies. They are not independant. That is an improvement over the vague and schizophrenic remarks of the NAS Panel.

In addition, Wegman faults Mann for not adeqautely describing methods and for refusal to share data.

[TCO]

Of course, if the PC1 is rubish, why do we use PCA anyhow? What a snail's shell that MBH peice of rubish is. The NAS showed what they believe in. That's the last 400 years (which btw have a lot of instrumental data and a lot of non-tree ring proxies). They're not going to go out on a limb with Mann for the Midevil Warm Period!

[TCO]

BTW, I disagree with Stockwell's comment about a warmer MWP. Basically we don't know what happened then. Mann claimed to know...and he's had his "hockey stick" cut off. Now we just don't know.

[Charlie (Colorado)]

Tim, they say there's "high confidence" the planet is the warmest it's been in 400 years --- in other words, since the Little Ice Age.

As I said above, "well, duh!" It's the warmest it's been since it was most recently coldest. It's 84°F here, the warmest it's been since today's low. Without other information, that doesn't really tell us much of anything about AGW.

The NAS say that Mann et al's conclusion about the warming being "unprecidented" in the last 1000 years to be "plausible". That's all. Wegman et al have shown that the reconstruction method Mann et al used is flawed, and leads to an upward inflection --- a "hockey stick" --- in the absence of any signal at all. That doesn't mean that Mann et al's notion is not plausible. It just means that it can't be considered any more than just plausible, ie, apparently reasonable. Not "correct", or "likely", just apparntly reasonable.

But Wegman et al is not a "subset" of the NAS report: they go rather beyond it in noting: that the MBH methods are statistically flawed; that the flaws are such as to entirely account for the MBH results; that "Manns assessments that the decade of the 1990's was the hottest decade of the millennium and that 1998 was the hottest year of the millennium cannot be supported by his analysis"; that "it's not clear that Mann and associates realized the error in their methodology at the time of publication"; and that the criticisms of McIntyre and McKittrick are "valid and compelling".

Mann's response to this criticism was entirely ad hominem.

[Charlie (Colorado)]

TCO, remember that the tree ring proxies from some sites (from Niwot Ridge, as I recall, which I can see from here), as interpreted by the dendrochronological methods, show a error of -1° to -2° to actual recorded temps. That error alone is sufficient to account for the data suggesting this is the warmest period in 1000--2000 years.

(Again, this doesn't mean MBH is implausible -- just that their position can't be statistically distingusihed from the "null hypothesis", which is that current temperature variations are just part of the normal cyclic variation.)

[richard mcenroe]

Tim Lambert, John Quiggin and Global Warming/AGW/Climate Change have been much discussed

Here

Here

Here

and Here

There's plenty more there if you want to search the site for climate change, global warming, polar bears and fish

[Tim Lambert]

TCO, quantitative measures of the effect of centred vs non-centred analysis have been done. The hockey stick signal ends up in PC5 instead of PC1. The reconstruction still looks the same. How come you didn't know this? Oh right, you're getting all your information from climateaudit.

[Charlie (Colorado)]

On plausibility

Here's why "plausible" is not strong enough a reason to accept MBH's arguments: we say something is "plausible" if it's "reasonable". It means there is no obvious flaw so that we could say that there is no possible way MBH could be correct. It's possible. It's not necessarily not so.

It's also plausible that anthropogenic factors have no particular effect, or as Roger Pielke suggests, only a small effect, and that the warming since 1500CE is just natural variation. (The null hypothesis.)

But then, it's also plausible that we've reached the top of the temperature variation, and we're starting back toward a "Little Ice Age". There are data that could be interpreted that way. (And if so, don't panic, this is a kiloyear cycle; the people in 2500CE can deal with it then.)

What this means is that its "plausible that we should:

(1) take steps to reduce the CO2 loading to prevent further anthropogenic warming;

(2) take no particular action, because we're having no particular effect, or;

(3) take steps to increase the CO2 loading in order to prevent the slide back into a Little Ice Age.

[Charlie (Colorado)]

Tim, Wegman, M&M, and Stockwell have shown pretty strongly that the "hockey stick" is an artifact of a flawed statistical treatment. That's the whole point.

[TCO]

Tim, have you looked at what a straight average of the proxies looks like compared to the Mannian reconstruction?

The PC5, great! What does that do to the TEXT of the discussion in MBH, where they say that the PC1 pattern shows the "dominant mode of variance". I guess the dominant mode of variance is different, with proper PCA!

On topic: no, that is NOT a quantitative study. I want to know HS index with and without proper PCA (Mann's method not proper...although he is too much of a moral coward and not a real scientist to admit it.) I want a number!

And of course, you have made no defense of the use of autocorellated statistics as if it were independent.

[maryrose]

Wake me up when it is actually happening...
until then I'm going to enjoy those lazy hazy crazy days of summer...

[The Earth]

I'm fine.

Cheers,

Earth

[Tim Lambert]

TCO, if you had been paying attention, way back in Feb 2005 you would have seen some nice graphs that show what the reconstruction looks with a centred analysis. Hey! It looks the same! Not a good idea to get all your info from ClimateAudit.

[Seixon]

Still nothing at Think Progress. I know what that means. Do you Lambert?

PS. Hey Tim, how you doing these days? Still defending the Lancet study to the death (no pun intended)?

[Charlie (Colorado)]

Tim, the point is that the graphs look the same even if all you do is feed MBH's method random numbers. (See Wegman's Figure 4.1.)

This has been replicated multiple times.

Mann's methods are wrong. These methods can't be used as evidence confirming anthropogenicity.

What's more, Wegman et al strongly supported McIntyre&McKittrick ("we found ... the criticisms of [McIntyre and McKittrick] to be valid and compelling.") Which is to say, the criticisms on Climate Audit.

Whatever your criticisms of TCO, what Tom's looking for is insight into the AGW issue; have you got any?

[richard mcenroe]

Charlie (Colorado) You seem to think you are having a rational discussion with a rational partner. This assumes several facts not in evidence. You might want to go look at that first link I posted.

[peter]

Hey has anybody seen my copy of the Bush Middle East Roadmap? We seem to be headed straight to hell.

[TN]

Tim,

Why doesn't Mann release exactly what he did, so we can see the effects of centered and non-centered analysis for ourselves?

[Charlie (Colorado)]

Charlie (Colorado) You seem to think you are having a rational discussion with a rational partner. This assumes several facts not in evidence. You might want to go look at that first link I posted.

No, I'm figuring that if I keep pressing him to make a rational argument, he'll eventually start screaming at me.

[SteveMG]

Hey has anybody seen my copy of the Bush Middle East Roadmap? We seem to be headed straight to hell.

That's right Professor Pete, our Fletcher School of Diplomacy graduate. He's been in office a whole six years and he hasn't settled the entire Middle East problem.

Hell, six years? Six and one-half years.

Do you make a concerted effort on saying silly things or does this just come naturally?

SMG

[Charlie (Colorado)]

Why doesn't Mann release exactly what he did, so we can see the effects of centered and non-centered analysis for ourselves?

Why didn't he release it to Wegman?

[Rick Ballard]

"Why didn't he release it to Wegman?"

Random number generating algorithms can only be released to qualified personnel?

Coke doesn't release its "secret formula". Why should Mann?

The formula is so complex that Mann is the only person in the world smart enough to understand it?

David Copperfield won't agree to reveal the "secret"?

[Al Gore]

You kids stop making fun of me, or I am going to turn around and shut down the internet. So help me I will

[peter]

SteveMG

If you think there is something the least bit silly about Bush's lighting of a fuse that has lead to this explosion in the middle east then you need a emergency empathy transplant.

[SteveMG]

If you think there is something the least bit silly about Bush's lighting of a fuse that has lead to this explosion

If you think Bush set off this fuse all by himself you've ignored about 2,000 years of history.

You do realize, I hope, that people around the world - other countries, over movements, other leaders - have their own free will? They act on their own motives and plans. They are not simply pawns that Bush (or the US) can totally control or command?

But if you want to exonerate Hezbollah and Syria and Iran, be my guest. They only acted the way they did because Bush did something or other. Or didn't act. Or something.

Which ignores the fact, of course, that, among other things, before Bush became president, these various groups were waging conflict on one another as well as Israel. I guess, for example, the Intifada was Bush's fault too?

Nah, why study history. Why consider the complexity of the situation, the history, the politics, the economics, the religious elements. Just blame Bush. It's all so much easier than having to actually think.

Make sure your candidate three years from now has that as a major plank in his (or her) platform. Bush is to blame for the Middle East problems.

Not Syria. Not Iran. Not radical Islam and dysfunctional governments.

SMG

[Extraneus]

Although you have to admit, things aren't necessarily as simple as we might like to think.

For example, if Bush had put a plan in place on 9/11/2001 to take out the Axis of Evil by 1/20/2009, and he was on schedule to succeed, what would the world look like at this particular point in time (and where would American military forces be best positioned)?

I know it's said that Bush doesn't have a "roadmap," but I'm not sure that's actually true.

[Pofarmer]

I'd like to think someone as intelligent as peter could stay on topic, guess not.

I'm sure that Belmont club, has lots on the Israel situation.

[cathyf]

Pofarmer, shut yo mouth. The last thing Belmont needs is idot trolls.

cathy :-)

[cathyf]

Oops, hit the "post" rather than "preview" button. The absolutely most embarrassing error possible in a blog post or comment -- misspelling "idiot" when one is attempting to apply it to someone else!

cathy :-)

[PeterUK]

A simple rule of thumb with quasi-scientific theories,Marxism,Freudian analysis and Anthropomorphic Global Warming,is to examine who amangst the ranks of the Great and the Good hold them to be gospel.The one knows for sure it is balderdash.

Judging by a current example,Intelligence Measuring can safely be added to the list,well at least .0001% certain.

[Joe P.]

I've looked much further into the issue of global warming abd the ipcc process since I last posted here regarding gw. I looked mostly at the ipcc 2001 report itself. There is no doubt that the whole ipcc process, and those of its fellow travellers, is biased - and the ipcc admits it, as I previously found and quoted from its 2001 report:

"12.8.1.... However, it must be said that potential gains [benefits of Global Warming] have not been well documented, in part because of lack of stakeholder concern in such cases and consequent lack of special funding."

This bias is seen throughout the body of the report. And it's so perfuse and varied in form that the ipcc report is literally preposterous and laughable.

The "job" of the ipcc from the start was to demonize global warming and blame fossil fuel use for it. The ipcc "science" is therefore totally biased, and not science even in so far as the biasedly projected "disasters" and "blames" themselves - the body of the ipcc report admits multiple serious black holes in its data base. Etc.. It was not the ipcc's job to be scientific in any way.

[Therefore there cannot possibly be any scientific "consensus" about it, and none has been found, except for the positive negative kinds - like "cannot rule out that x might be the case".]

The whole effort in fact was designed to produce a "perception" only, as a massive p.r. effort only, and right from the start. Take that to the bank, get some real "change", then get yourself some "Enerex" instead to ensure male potency, The money will be better spent.

The "war on science" was conducted quite scandalously by the ipcc and all its fellow "stakeholders". There is really no doubt about it. Mann's hockey stick is only another example of this "science is only political science" mentality. Heads should roll. Perception is not reality anymore than hallucinations, fantasies, or wishes are. Sorry about that.

[Neo]

For years, probably decades, I've heard complaints that teacher don't get respect anymore.

In the 1950's version of War of the Worlds, I recall the line .. "he's a scientist" said with that glowing respect at all those teachers have been complaining about losing.

Then we have Mann and the "hockey stick" that no one could reproduce. It arrived virtually all of the "he's a scientist" respect intact, brought to a zenith by all those rocket sciences that took us to the moon and beyond.

It leaves with the review by Wegman et al, shown to be a poor (to be generous) or bad piece of science, which has now been wrapped in enough politics to get a statewide candidate elected.

Worst of all, the scientists are now lumped with the teachers. My congratulations to all the fathers.

[topsecretk9]

Novak on meet the press today:

MR. NOVAK: A third party close to the primary source called me after the investigation was launched and said, and said that he believed that it was—he believed he had given me inadvertent—inadvertently given me information—this information.

Doesn't it just sound like Novak is saying the 3rd party called Novak on his own, not at the instruction of UGO?

"he believed"

Either UGO knew he inadvertently talked or didn't, and sounds like 3rd party panicked and witness tampered.

Because Novak, on H&C's, said that 3rd called to say this inadvertent bit...and Novak said he didn't know if it was inadvertent but also didn't think it was a deliberate smear.

[topsecretk9]

oops, wrong thread...

[peter]

Wow

Unbelievable. You simply can't justify what Bush has done in the middle east so, as usual for the republican spin machine, you blame everyone and everything else.

"Free will"? What about Bush's free will? A will that drove him to bomb, invade and occupy Iraq. What about your free will to reject this monster in 2004? I hope you have an once of empathy left for those Iraqis that YOU have sent to their deaths for
nothing.

"I know it's said that Bush doesn't have a "roadmap," but I'm not sure that's actually true."

So your plan is to do nothing and saying nothing as the incompetent Bush admin foments more chaos because your not sure he doesn't have a plan. Good plan. Just trust me huh?

SMG. Your argument is nowhere to be found... just ramblings of somebody who does not think but simply parrots.

In case you have not noticed the topic of interest to our citizens is the mess that Bush has made in Iraq. Of course it's not my blog so go ahead keep pretending that you know a damn thing about science in general and the science of global warming in particular. Sometimes I can't tell if you are arguing about claims of global warming or claims of
cigarettes as a risk factor for lung CA, COPD, heart disease etc. You sound and act just like Phillip Moorse 1960.

[Tim Lambert]

Charlie, the point is that the reconstruction does NOT look different whether you use centred or uncentred PCA. I already posted a link that shows the two results drawn on top of each other.

[TCO]

Tim, Tim, Tim:

That RC post does not show the effect of centering on the overall reconstruction (through the Mannomatic so to speak). In addition it is not for the entire network of proxies. In addition, it introduces an UNDOCUMENTED in the paper and UNCODED in the code, assertion of "Preisendorfer's n" as a criteria for deciding what PCs to keep.

[MJW]

As I understand TCO's comment, Lambert admits without the incorrect centering, the hockey stick shape is moved from PC1 to PC5. As long as Mann can retain this PC in the analysis, the result will look similar; however, retaining up to PC5 is pretty questionable. Mann's justification for retaining this PC supposedly relies on a statistical criteria called "Preisendorfer's n," which doesn't actually apply to the situation.

I must say, if Lambert's view that Mann's results are independent of the incorrect centering is correct, it's surprising Professor Wegman didn't point that out.

[PeterUK]

"In case you have not noticed the topic of interest to our citizens is the mess that Bush has made in Iraq."

Mr .0001,a very simple system obtains here,Mr Maguire posts on a subject and commenters are free to comment on that subject.
As far as is observable, there is no sampling or polls taken to evaluate the amount of interest various posts will generate,that can be judged,roughly, by the number of comments a post attracts.
This can be skewed by a monomaniacal troll posting off topic,probably something he read in the bottom of his litter tray which is exercising his fevered mind at any one point in time.Rather than contribute his great intellect to the subject in hand,the troll will rattle his tin cup against the bars,scream and defecate, in order to be the focus of attention.
There are plenty of sites discussing the Middle East,like this one for example,where your comments will be responded to with courtesy and consideration.
Horses for Courses, Sweet Prince,Horses for Courses.

[Tim Lambert]

TCO claims "That RC post does not show the effect of centering on the overall reconstruction"

The RC Post states "This figure shows the difference in the final result whether you use the original convention and 2 PCs (blue) and the MM05 convention with 5 PCs (red). The MM05-based reconstruction is slightly less skillful when judged over the 19th century validation period but is otherwise very similar. In fact any calibration convention will lead to approximately the same answer as long as the PC decomposition is done properly and one determines how many PCs are needed to retain the primary information in the original data."

Waiting for your retraction, TCO.

[SteveMG]

Pete:
SMG. Your argument is nowhere to be found... just ramblings of somebody who does not think but simply parrots.

Sorry, I have no idea as to what you're talking about. You've mixed up comments by others and attributed them to me.

And someone who thinks that the Palestinians and Israelis (and the rest of the peoples in the ME) were just living in peaceful co-existence until Bush invaded Iraw is really lost.

TM allows threads to go off on different directions but I think discussing the Middle East situation - a problem that pre-dated Bush by about 2,000 or so (if not more years) - on a thread about global warming (another problem that Bush created all by himself I'm sure) and the science behind it is too much of a diversion.

You're on you're own little "post hoc propter hoc" view of history on this one.

Good luck with this one, Pete.

SMG

[richard mcenroe]

Tim Lambert is maybe not the scientific authority he might wish...

[Jean S]

Tim:

quantitative measures of the effect of centred vs non-centred analysis have been done. The hockey stick signal ends up in PC5 instead of PC1. The reconstruction still looks the same

This is complete BS. PCA is nothing but orthogonal projections of data. The whole idea is to reproduce as faithfully (in MSE sense) as you can the data with lower dimensionality data, or, in other words, find "the meaningful" projections of data that summarize the data best. If you actually do PCA on data Tim is refering to, you don't choose the 5th PC according to any reasonable (objective) PC selection measure (such as Akaike, MDL, BIC etc.). Only by ad-hoc measure used by Mann et al the 5th PC gets chosen (in this case). In fact, this actually shows the inappropriateness of the selection method used (and/or "Mannian PCA"): it is impossible to have a situation such that with true PCA one is supposed to select more PCs than with another projection ("Mannian PCA").

Mann's "PCA" is an ad-hoc procedure that tends to find "hockey-stick" projections from data, and it has nothing to do with the original meaning of PCA.
This all is well explained in Wegman's report, and even Tim should be capable of understanding it.

Tim, please do a favor and stop posting about things you clearly do not understand.

[Charlie (Colorado)]

I must say, if Lambert's view that Mann's results are independent of the incorrect centering is correct, it's surprising Professor Wegman didn't point that out.

I haven't pursued the recentering arguments specifically in any detail, but with Wegman, Stockwell, and M&M all showing that Mann's methods result in a "hockey stick" using random data ("red noise"), it wouldn't be a surprise to see that Mann's results are independent of the centering issue. I think this is the stake through the heart of the argument: if Mann is detecting signal, his methods applied to a data source with signal in it would produce a different result than when they're applied to a data source with no signal, ie random noise.

Since Mann's results are statistically indistinguishable from random noise, we can conclude that the methods have no "skill" --- they don't tell us anything. Any conclusions based on those results ar invalidated.

[Charlie (Colorado)]

Tim, please do a favor and stop posting about things you clearly do not understand.

Death of Internet predicted.

[paulv]

Are Iran and Syria caught in quaymire in Lebanon. Hezzbollah is getting kicked around by IDF and Syria is too scared of IDF to give support and Iran distance makes it safer by unable to aid the terrorists in fight with IDF. Lebanon alreadt hates Syria and Hezzbollah and now can take action against weakened Hezzbollah,

[Hans Erren]

pofarmer:

consider the following:

You put 6 dollars in the petty cashbox, the estimated cash flow of other people is 150 dollars. At the end of the day the net amount in the petty cas has risen by 3 dollars.

somebody nicked half of your contribution.

[Slartibartfast]

Not being a statistician or climatologist, I'd nonetheless be appreciative of seeing this discussion continuing without the ad homs. Lambert's opinions are either true or false independent of whether he's a genius or an idiot, a Democrat or a Republican, black, white or purple.

In other words, I'm much more interested in getting at the meat of this issue than pointing out the various flecks of gristle and bone tissue.

[RogerA]

I think Charlie pretty well listed the implications of the study, whether right or wrong--The really big question for me is EVEN IF AGW is happening, how much does it contribute to overall GW and how much would it cost to "fix" it--Those studies havent been done to my knowledge. A subsequent question would be what would the earth look like if temps went back up--this is beyond the gloom and doom crap out there: presumably more areas will be brought under cultivation; what other benefits would accrue from global warming?

[Slartibartfast]

Oh, and furthermore: the Tim Blair fun-making should not be construed as being equivalent to debate. The bit about the earth radiating energy to the rest of the universe, while being phrased in a way that...well, I wouldn't say it that way, is accurate. The earth actually radiates energy back to the Sun.

Me, I think of the "universe" as a collection of physical objects, so to phrase it that way is, to me, wrong. However, one can think of the universe as being EVERYTHING: mass and empty space, so in that sense the phrasing is dead right. More neutrally, though, the Earth radiates energy outward. Where that energy actually falls is irrelevant.

I've got much less use for Al Gore than the average American, but that doesn't mean that everything he says is wrong.

[SteveMG]

You're on you're own little "post hoc propter hoc" view of history on this one.

Make that, of course, post host ergo propter hoc.

Although for some folks, it probably should be post Bush ergo propter Bush.

SMG

[Forbes]

It would seem that McKitrick and McIntyre have also provided a thorough debunking of Mann's "hockey stick" by demonstrating that a random set of trendless numbers would generate a hockey stick outcome.

[Neo]

From reading here, many have raised policy-relevant questions concerning the objectivity of the IPCC and the appearance by the IPCC that the intelligence and facts were being fixed around the policy.

Whoa .. I've heard that sort thing before somewhere.

[Tim Lambert]

One more time for the people who didn't get it before. PC1 is NOT repeat NOT the same as the reconstruction. Contrary to Jean's misrepresentations, Wegman did not report what happened to the reconstruction, just to PC1. The reconstruction has a hockey stick shape WHETHER OR NOT you do a centred or uncentred PCA.

That is why the National Academy of Scoences panel said that it made little difference which method you used.

[nanny_govt_sucks]

"One more time for the people who didn't get it before. PC1 is NOT repeat NOT the same as the reconstruction. "

Where is that being claimed, Tim?

[nanny_govt_sucks]

"The reconstruction has a hockey stick shape WHETHER OR NOT you do a centred or uncentred PCA."

It sounds like the uncentered PCA done by Mann has other differences as well as the uncentering, like the selection of the 5th PC.

Can you comment on that aspect to help straighten things out?

[David Stockwell]

TCO: BTW, I disagree with Stockwell's comment about a warmer MWP. Basically we don't know what happened then.

The problem with this stance is all the evidence pointing to a MWP. Don't get all denialist on us. Now in addition, I would like to see a quantitative climate field reconstruction by qualified statisticians using reliable proxies, and this should be the response after Wegman.

[nanny_govt_sucks]

Sorry, I think I misstated my last post.

Tim, it sounds like Mann's "centered" PCA had some extra bits that selected the 5th PC. So it wasn't a true apples to apples comparison of an uncentered to centered PCA. Can you comment on this specifically? Do you think it was OK to select the 5th PC?

[Jean S]

Tim keeps distorting things:

The reconstruction has a hockey stick shape WHETHER OR NOT you do a centred or uncentred PCA.

Yes, if you allow an ad-hoc criterion to select 5 PCs which normally used criteria do not allow. But be careful, if you select too many you might not get your beloved HS either.

That is why the National Academy of Scoences panel said that it made little difference which method you used.

Yes, it makes little difference, since in both cases the reconstruction has skill close to zero. I can do better with random noise.

Instead of pointlessly arguing about things you do not understand, why don't you try to read the Appendix in Wegman's report. Or finally reproduce those correlation stats I asked you many times before over CA.

[Jean S]

Tim: One more thing: there is no such a thing as "uncentered PCA", or actually "short centered PCA". It simple does not exists beyond Mannian literature.

[Old Dad]

Can someone help this befuddled layman out?

The global mean temperature has increased about a degree F in the last 100 years? It's a tough calculation, but most scientists agree that it's a pretty good number.

The hockey stick boys claim that the 1990's was the hottest decade (in the Northern Hemisphere) in maybe 1000 years (make it 2000 years just to be safe). Some call bullshit on the hockey stick boys. The NAS weasels out and calls their story "plausible." Criminy, I got away with murder in high school with plausible stories.

But let's say the hockey stick boys are spot on. Moreover, let's say that all the global warming took place in 1998, in July, or spread it out, however you like. And let's grant that it's entirely my fault--I'm running a giant SUV 24/7.

Despite all my efforts, the globe has cooled since 1998.

The globe was warmer in the near past, ca 1600 or ca the first Christmas--you pick. Why should I care?

Shakespeare and Caesar drove horses.

[Neo]

hakespeare and Caesar drove horses, but probably had farting cows.

[Slartibartfast]

There's nonfarting cows?

[Charlie (Colorado)]

Charlie, the point is that the reconstruction does NOT look different whether you use centred or uncentred PCA. I already posted a link that shows the two results drawn on top of each other.

Tim, the point is that the reconstruction doesn't look different even if you feed it random numbers. The "hockey stick" is an artifact of his statistical method.

[TomB]

The bit about the earth radiating energy to the rest of the universe, while being phrased in a way that...well, I wouldn't say it that way, is accurate. The earth actually radiates energy back to the Sun.

Slart, what is it exactly that Gore said? Was it the "altering the balance of energy between our planet and the rest of the universe.”?

If so, I guess, in a pedantic way it is true, but it is unbelievably incoherent.

[Charlie (Colorado)]

Old Dad, I started to write you an answer, and then realized I was writing the survey article on global climate change I've been promising to write forever. I'll link it when I'm done.

[Charlie (Colorado)]

If so, I guess, in a pedantic way it is true, but it is unbelievably incoherent.

"Unbelievably"? Nah, I find it completely believable.

[Old Dad]

Charlie (Colorado):

Thanks.

[Slartibartfast]

If so, I guess, in a pedantic way it is true, but it is unbelievably incoherent.

It is true, if the claims are true. If the claims are false, it's false. So, it's either incredibly awkward (possibly to the point of distraction) phrasing of the I took the initiative in creating the Internet variety and true, or it's all that and false.

It seems as if some folks are getting hung up on phrasing, when all that describes (if poorly and potentially misleadingly) is the phenomenon of black- or grey-body radiation.

[peter]

Mark my words. Your families will remember your support of Bush and hate you for it and I mean hate.
Adios anachronisms.

[Bob]

peter with a little pee shows what drives his thinking... or lack of.

He goes to sleep at night worried if his family is going to hate him in the morning. Well maybe you and the rest of the bed wetters should spend a little time with the facts, instead of your neurotic theories.

This was the same thinking Clinton used for 8 years... worried only about his legacy and what his family of moonbats would think instead of doing the right thing.

Enjoy your world were nobody "hates" you... just don't close your eyes!

[Tim Lambert]

Charles, you are confusing PC1 with the reconstruction. Uncentred PCA gives you a hockey stick shape in PC1 from random numbers. But PC1 is just one step in the reconstruction. Random numbers do not give you a hockey stick in the RECONSTRUCTION.

You've been deliberately misled by the likes of Jean S.

[Jean S]

Tim:

But PC1 is just one step in the reconstruction. Random numbers do not give you a hockey stick in the RECONSTRUCTION.

Again BS. If you have hockey sticks (from whatever sources) and you continue with Mann procedure, you end up with the hockey stick in the reconstruction.

[J F Beck]

Lambert should know misleading when he sees it, he can mislead with the best of them.

See examples here, here and here.