The Island of Doubt

1934 warmest year on record?

That portion of the blogosphere that takes no shame in including Ann Coulter in their blogrolls is all atwitter with the news that NASA has “silently” released adjusted temperature records showing that 1934 is the warmest year on record, not 1998 or 2005 or 2006. How will Al Gore, James Hansen and all the other “enviromoonbats” recover from this embarrassing revelation? Probably without breaking a sweat, I would think.

Many of the blogs make no distinction between “warmest year in American history” and “warmest year in world history.” And the difference, as you might expect, is more than a little significant. The revised list, from NASA, does indeed put 1934 as the warmest year — in the lower 48 contiguous members of the United State of America.

But the warmest year globally remains 2005, followed by 1998, 2002 and 2003 and 2004. And the of the 12 hottest years on record, only one — 1990 — does not occur in the last 12 years. (Thank you Mount Pinatubo).

Many a right-wing blogger (such as this one, this one and this one, (in)conveniently glosses over or ignores the distinction entirely and spews out lines the likes of

Don’t expect any press releases from NASA or NOAA about this change nor much coverage on the networks or major newspapers.

Which, to be fair, is a pretty good prediction. But only because the revised list, reportedly due to some kind of Y2K recalculation (see here for real origin), will not affect global averages significantly and is more properly relegated to the footnotes of obscure journals. The revised calculations after all, may have changed the rankings of the top warmest years in the U.S., but only by a wee bit, +0.02 degrees C in the case of 1934. Even in Fahrenheit, that’s only 4/100th of a degree. To put it all in perspective, we’re already 0.7 degrees C above pre-industrial levels globally, with another full degree in the inevitably pipeline due to climate inertia.

But even those who probably do understand the math are happy to make a mountain out of this molehill. Daily Tech’s Michael Asher, who was up front about the fact that we’re dealing with US temperatures only, couldn’t restrain his glee (emphasis mine):

The effect of the correction on global temperatures is minor (some 1-2% less warming than originally thought), but the effect on the U.S. global warming propaganda machine could be huge. Then again — maybe not. I strongly suspect this story will receive little to no attention from the mainstream media.

It would seem this pseudo-scandal started with the help of veteran climate change denialist Steve McIntyre, but my attempts to link to his work at only generated a WordPress error.

Ah well. Nothing to see here, folks. Move along now.

Update: Check out this headline from KXMB, a CBS television station in North Dakota considers bloggers the equivalent of the wire services: “NASA Drastically Revises Global Temperature Numbers.” Nothing like letting the facts get in the way… oh you know the rest.


  1. #1 Hume's Ghost
    August 10, 2007

    I first heard this on the Rush Limbaugh show. Whenever I hear Rush make some kind of claim like this you (e.g nicotine isn’t addictive, Africans are getting fat, etc.) can assume with about 99.9% certainty he doesn’t know what he’s talking about.

    This also confirms an informal rule of thumb I’ve developed about Limbaugh. You can flip to his radio program randomly and if he’s not lying he will be lying within 30 seconds. When I got in my car and turned the radio to his program this story was what he was talking about.

  2. #2 natural cynic
    August 11, 2007

    You mean to say that the rest of the world exists? And isn’t just like us of want to be us? Shocking!

  3. #3 GiggleStick
    August 11, 2007

    While an important omission, that this just affected the American temperature data, I think it’s still worth noting that James Hansen from NASA won’t open up their data, and simple errors like a Y2K bug have to be inferred through secondary means.

    That doesn’t portend much confidence in all the data as a whole. And if 1%-2% changed the data it affected (the American) so much that only one of the hottest years was still recent, then I guess it did matter some.

    I really don’t think you should just wave your hand about incorrect and unreviewed data. Does that work in other sciences?

  4. #4 Skeptic8
    August 11, 2007

    Thanks for your timely analysis!
    The twisted data by the likes of Rush Limbo & co. isn.t unexpected. The story will serve as a “marker” for the recipients of the “Corporate Whore” award. How many are eligible on this one story?

  5. #5 Andy B
    August 11, 2007

    Greenhouse gases from the burning of fossil fuels are the cause of global warming. America deserves much of the blame for the warming of the planet. Screw 1934…

  6. #6 Ignorant Jerk
    August 11, 2007

    Did you see the other article?

    So the US inland data is suspect too. The data is unreliable. The analysis was corrected.

    What about the data for the rest of the world? How was that collected? Are there bugs in that analysis too?

    Is there really a theory? Were the hottest years REALLY in the past 10? Can we trust the data? the analysis? Is there a real theory here?

    This blog is called the island of doubt, but there doesn’t seem to be much skepticism here.

  7. #7 bill
    August 12, 2007

    This misses the point. If the data were open for review this would have been picked up early on. The only reason this was discovered was that the data were reverse-engineered and the error discovered, THEN the “corrected” data was published.

    The same thing happened with the “hockey stick”. The initial analysis was faulty. When the Principal Components are properly normalized, the “hockey stick” goes away.

    Having seen how “creative” scientists are with their data, I prefer to see open data and open models underly science and policy decisions. There is a reason peer review in Medical journals usually involve statisticians.

  8. #8 Brian
    August 12, 2007

    to see the error and correction graphically, check out tamino’s post here

  9. #9 Skeptic8
    August 13, 2007

    Thanx for the tamino link, Brian.
    Is there any source that averages ice & glacier cover from satellite images? This sort of bulk effect should be revealing.

  10. #10 Brian
    August 13, 2007

    Skeptic8 says: “Is there any source that averages ice & glacier cover from satellite images? This sort of bulk effect should be revealing.”

    I don’t know offhand. But tamino is rather open to suggestions about integrating other data…the post I linked to already has four updates as a result of commenter inquiries. I point to tamino on this, because he rarely makes errors when it comes to quantitative assessments.

  11. #11 caerbannog
    August 13, 2007

    The same thing happened with the “hockey stick”. The initial analysis was faulty. When the Principal Components are properly normalized, the “hockey stick” goes away.

    When the principal components are properly normalized and a number of them sufficient to capture the majority of the information in the data are retained, the hockey-stick reappears.

    When the principal-component step is skipped entirely and the data are used directly, the hockey-stick reappears.

  12. #12 Lance
    August 13, 2007


    Do you have any background in statistics? If you do and you would invest two or three hours at reading the various archives of the methodology and data used by Mann I think you would change your mind.

    Then there is the question of Mann’s inappropriate use of the bristlecone pine data that further skews the study.

    The evidence is quite clear and the “hockey stick” is quite dead. Those that defend it are either ignorant of the important details of the matter or deliberately trying mislead others. The NAS review panel was mostly in the former category. Mann is in the latter.

    From which group are you commenting?

New comments have been temporarily disabled. Please check back soon.