TNR: Stuck On Stupid

It's not just in Baghdad they fall for false reporting. It's global, baby.

Before getting to this fresh idiocy, check out Mark Steyn's column on "warm-mongers," which also takes a shot at TNR for its fake Beauchamp reporting. I note TNR doesn't mention that part of the article.

On to the stupidity.

But how big an error was it? Well, 1998 went from being listed as 0.01 degrees warmer than 1934 to being listed as 0.02 degrees cooler. That means 1934 is back to being the "official" hottest U.S. year on record, although it's still a statistical tie.

Nothing's really changed....

The two years were, as he admitted, in a statistical tie back in 2001, and they're still in a tie today.

As I think McIntyre himself pointed out, large arguments occur over changes in temperature of only .1 degree C; indeed, the entire case for global warming is predicated on a mere .6 degree C rise over the last century. When a yearly temperature average is claimed to have "shattered records" by a mere .1 degree C, the warm-mongers trumpet this as quite significant indeed, and a threat to our very existence. So it's all a bit too much for this TNR crank to now begin claiming that teeny-tiny adjustment in a year well before the earth's atmosphere was poisoned by a doubling of the Invisible Killer CO2 is completely meaningless.

I don't remember hearing anything about this "statistical tie" business when 1998 was being trumpeted as the undisputed, unchallenged warmest year in human history. Boy, you guys suuuuuure seemed to keep a lid on the "statistical tie" thing back then, huh? It's only now -- when 1934 becomes the warmest year -- we're hearing about "statistical ties."

(He does quote Hansen himself noting it was a statistical tie, sometime. Once. Obviously the global warming cultists must have missed that, or else I'm sure they would have modified their cant to say "1998 is the warmest year on record, but you should note, in fairness, it's statistically tied with 1934 for that honor.)

And I find it odd that this nitwit is continuing to pimp the claim that seven (whatever) of the last ten years are among the warmest on record.

Really? Let me ask... I'm just guessing here... would it be the case that many of these years are in fact in "statistical ties" with years in the 20s, 30s, and 40s for that honor? Why is it that once again you seem to forget all about "statistical ties" when it suits you?

He also claims no fraud occurred. In what other branch of science are scientists permitted to hide their data and the algorithms and "adjustments" made to the actual data? Hansen hid his, refused to share it with skeptics. This is TNR's idea of peer-reviewed, relentlessly verified science? When a scientist says "Hey, can I have your non-proprietary algorithms, which, by the way, were paid for by the state and therefore belong to taxpayers," that scientist is allowed to say FOAD?

Really?

Another claim is made that global temperature records haven't changed... much. Well, "global temperature records" are something of a myth. Only two areas of the world -- the US and Europe, of course -- bothered to keep extensive, systematic temperature records for the past 100 years, and of those two, we now know that at least one now shows absolutely no upward trend line in temperature at all.

Absolutely none? Well, I'm not quite sure about that. But I don't see a major obvious trendline here:

The global temperature "records" -- constructed partly by interpolation and guesstimation -- still show a marked upward trendline, of course. But it's rather easy to get such a trendline when you're free to, ahem, adjust and guesstimate the data-set you're analyzing.

I'm sorry, TNR. But the fact that one of the two real data-sets (not interpolated, estimated, guessed at based on "temperature proxies") for global temperatures now shows the current era no different than the 1930s-1940s period, which of course ended with the great global cooling that threatened a worldwide ice age.

Perhaps the stupidest moment comes here:

Finally, some bloggers seemed to be confused and think that this error somehow debunks scientific climate projections. But the models used to predict future warming aren't constructed using surface temperature records--they're physical models, not statistical models.

A-ha! I see! The actual data may have changed, but the computer models-- quite independent of such debased things as empirical data -- live on! Well! Thanks for letting me know!

This is even stupider than it sounds, and either deliberately deceptive or exposing the utter ignorance of this poster. When he says the models are "physical," not "statistical," I take that to mean the models are based upon a certain change in the CO2 fraction of the atmosphere causing a certain change in temperature, which in turn causes a certain fraction of ice to melt, which in turn completely destroys the world.

Well, Joker, check this out: The very first assumption in those "physical models" regards how much temperature change will be forced by how much a change in the carbon dioxide fraction of the atmosphere. See if you can grasp this: If the world's best temperature records show no appreciable increase in temperature after carbon dioxide has doubled or tripled, then those "physical models" will, presumably, have to be modified accordingly. They will have to be changed to model (physically!) a world in which temperature is not quite as sensitive to increases in carbon dioxide as previously proven guessed completely out of one's ass.

Of course they won't be modified, but they should be.

Finally, there's this:

Here's a paper laying out the differences between the warm weather of the '30s--which was largely confined to the high latitudes--and the more serious warming of today (see p. 5). Among other things, the Arctic ice wasn't disappearing as rapidly in the 1930s. Which is just to say that Mark Steyn's column on the subject is, as you'd expect, total nonsense.

Is that true? Depends on how one defines "the 30s."

D.C. resident John Lockwood was conducting research at the Library of Congress and came across an intriguing Page 2 headline in the Nov. 2, 1922 edition of The Washington Post: "Arctic Ocean Getting Warm; Seals Vanish and Icebergs Melt."

The 1922 article, obtained by Inside the Beltway, goes on to mention "great masses of ice have now been replaced by moraines of earth and stones," and "at many points well-known glaciers have entirely disappeared."

"This was one of several such articles I have found at the Library of Congress for the 1920s and 1930s," says Mr. Lockwood. "I had read of the just-released NASA estimates, that four of the 10 hottest years in the U.S. were actually in the 1930s, with 1934 the hottest of all."

1922. The year after 1921, which, if TNR is keeping up, is now the third warmest year on record according to the best-kept temperature records available.

Correction: I incorrectly stated the adjustment to 1934 was by .3 degrees C. The correct figure is .03 degrees C. Quite a bit less.

Thanks to brad.

The Flexible Definition of "Meaningless: From the NCDC:

Major Highlights

GLOBAL AVERAGE TEMPERATURE FOR JANUARY HIGHEST ON RECORD,
U.S. TEMPERATURE NEAR AVERAGE FOR MONTH

Global Temperatures
The combined global land and ocean surface temperature was 1.53°F (0.85°C) warmer than the 20th century average of 53.6°F (12.0°C) for January based on preliminary data, surpassing the previous record set in 2002 at 1.28°F (0.71°C) above the average.

Note the big headline. January 2007 smashed the old record holder of January 2002 by -- let me get my calculator out -- 0.14 degrees C.

0.14 degrees is enough to warrant a scare headline, whereas .03 degrees C is completely "meaningless"? Certainly .14 is a larger number than .03. But the former grabs a big headline and the latter is utterly without any import whatsoever?

Really? What exactly is the cut-off point here? I just want to know so that in the future you guys are on record. What's meaningful? .05? .07? You tell me; let's get the threshold for significance stated right now.

It seems only upward adjustments or high readings are "meaningful." Downward adjustments or low readings are "meaningless."

Incidentally... Brad Plumer claims this paper proves that the 1930s warm period observed in the US (by temperature records, not proxies or interpolations) was restricted to "higher latitudes."

With all due respect, I've scanned through it, and I'd like Mr. Plumer to point out specifically where such a thing is proven.

I should note the paper doesn't even seem to address what he claims it to address (it discusses high-latitude warmth, but does not seem to claim the 1930s warmth was restricted only to the higher latitudes). What the paper seems to be about is sea-ice thickness in the Arctic since the 1970s.

Again, given that my PDF reader is having trouble with this article, perhaps I just missed this key proffer of evidence. In which case I invite Mr. Plumer to tell me where the article actually establishes what he implies it establishes.

Found It: As I figured, the claim seems largely based on modelsand interpolated estimates of unrecorded temperatures, not actual temperature readings. If Plumer can point out the data set that actually establishes lower-than-60-degree-N global temperatures, I'd be appreciative.

I hardly need to remind everyone that this is what is always claimed by the global warming cultists. When there is strong evidence from those parts of the world keeping good records that in fact the world was warmer in the past, they always claim -- always -- that the parts of the world which were conveniently not keeping very good records were colder, thus offsetting the warmth of the north.

They did the same thing with the Medieval Warm Period, of course, claiming for a long time (a claim I know think they've abandoned) that the very well documented Medieval Warm Period was restricted only to Europe and upper latitudes. (I should note the Medieval Warm Period is not established by temperature readings, but by the historical facts that at that time England was rowing wine-grapes, Greenland could support wheat farming, and European wheat production soared, resulting in depressed prices for wheat and other food.)

They assured us that temperature proxies in the southern hemisphere demonstrated that the warmth was occurring only in Europe, and that the rest of the world was experiencing lower than average temperatures. They've now, it seems, walked back from this previously-"proven" claim, now saying that the earth is warmer than it has been in 400 years, rather than in all of human history. Because what was "proven" before -- while the upper latitudes baked, the rest of the world was freezing! -- is now not quite so proven at all.

How the hell did it manage to be "proven" for ten years, I wonder.

This just seems awfully convenient. The best data-set available (US temperature records) shows no real upward trend in temperature -- at least it shows nothing more than a similar cycle to that which occurred in the thirties and forties -- but they insist that their models and proxies and interpolations and, um, estimated data-sets (which are also known as "not actually data sets at all") demonstrate that such warmth occurred only where it can be conclusively proven to have happened. Where the evidence is decidedly shakier, proxies and the like "prove" it was cold.

posted by Ace at 12:25 PM

| Access Comments