13 August 2009

1,264 to 1: Correction 27 to 3!

Update: A few folks properly criticized this post as a bit sloppy. It was. Sorry about that. The real ratio is 27 to 3! Thanks for the eagle eyes and keeping me honest.

1,264 27 = the number of news stories covering Michael Mann and colleagues' new paper claiming that Atlantic hurricanes are at a 1,00o-year high.

1 3 = the number of news stories covering Chris Landsea and colleagues' new paper claiming that there is no trend in Atlantic hurricanes over a century.

Bias? You be the judge.

22 comments:

  1. It is a combination of quantified bias and sensationalism.

    As the song sez:
    “Newspapers sell disaster and sin [and not much else]”

    Either way, NOAA has some “inconvenient” hurricane data and I presented it here.

    ReplyDelete
  2. Landsea has published similar papers before; that might be one reason this one gets little attention; It's fairly old news.

    The real significance (when taken with Mann's) is that El Ninos tend to suppress tropical storm formation in the tropical Atlantic, whilst tropical storm numbers tend to increase during La Nina years. (El Ninos increase wind shear that helps suppress hurricane formation, to put it simply.)

    I would expect to have seen a rise in the actual numbers in periods when La Ninas were more frequent than El Ninos; that wasn't the case during the last century; El Ninos predominated. (I know there has been discussion as to whether warmer SSTs cause more intense tropical storms; I've never thought there would be a simple correlation such as that myself; there's African dust to consider too for example.)

    The significance of Mann's paper is that it is yet another piece of evidence supporting the hypothesis that the eastern Pacific entered into a predominantly la Nina state prior to and during the so-called Medieval Warm Period; a period of prolonged, severe droughts across the Americas.

    Landsea is looking at whether there has been an actual, or simply a reported, rise in tropical storms over the 20th century, not at whether there has been a rise over the past ten centuries. In its own way Landsea's papers supports the hypothesis that ocean variability conditions were different during the MWP. I don't see these two papers contradicting each other.

    I've started writing up an account (from published research) of both the American Medieval droughts and tropical ocean variability conditions at the time on a site here. It's a work in progress (and I have more to write up when I get time) but the first half of this page should give an overview.
    http://sites.google.com/site/medievalwarmperiod/

    If anything, both Mann's and Landsea's papers could be seen as supporting the hypothesis that ocean variability conditions were not the same during the 20th century as during the MWP. Therefore, we are not in a repeat of the conditions that caused the warming of the MWP. I'm not sure is that is news, I've never thought we were.

    ReplyDelete
  3. Not so fast ...

    1. Are the release dates the same? You'd have to normalize for diffusion time, and perhaps also visibility of the journal, and breadth of press release distribution?
    2. Your search on Mann includes all kinds of spurious stuff. "Juice Rallies on Crop-Size Fears; Traders Put on Storm Watch" ?
    3. Some outlets spin Mann the other way. "Hurricanes more common in the Middle Ages" (USA Today)

    I'm not saying there's no bias, just that your data is flawed.

    ReplyDelete
  4. -3-Tom

    All fair points:

    1. NOAA's press release came out 8/11, mann's paper had press releases issued 8/12

    2. I just redid the search as

    hurricane "michael mann" nature

    and came up with 1,383

    3. That USA today story is the only one I've seen with that slant (which is contrary to Mann's own statements).

    I'll accept some uncertainties in the data, but I also think the qualitative point stands strong;-)

    ReplyDelete
  5. Even though I agree with you on this, you missed one (admittedly at a small Florida paper):
    http://www.heraldtribune.com/article/20090517/ARTICLE/905171028/-1/NEWSSITEMAP#

    ReplyDelete
  6. Roger, Do you realise that there's also a film director called "Michael Mann"?
    I just scrolled down the list of results from your Google news search criteria and found lots of references to the new Johnny Depp movie, 'Public Enemies' - directed by Michael Mann!
    Run the Google seach as "Michael Mann" +hurricanes and I only got 27 returns.

    ReplyDelete
  7. Roger,

    When I click through on you google search, I get around 10 news items covering proxy reconstructions of hurricane frequency.

    The other thousand or so appear to cover Michael Mann's moonlighting job as a Hollywood director:
    http://www.nytimes.com/2009/08/14/movies/14movies.html

    ReplyDelete
  8. Dr Pielke:

    I have to disagree with your accusation of "bias" in this case. News media report news, not science.

    'The End of the World is Nigh' is news.
    'The End of the World is Not Nigh' is not news.

    The 'Mann reconstruction' suggests increasing disaster in the immediate future and, therefore, it is news. But the Landsea analysis reports no discernible change and, therefore, it is not news.

    Hence, I respectfully suggest that the comparison you present is not evidence of media bias concerning climate science.

    However, I agree that there is much biased reporting of climate science by the news media. For example, I know of two formal complaints to the BBC concerning the BBC's blatant bias on the subject.

    In my opinion, unfounded accusations of such bias hinder attempts to correct the real problem of biased reporting of climate science by some important media organisations.

    Richard

    ReplyDelete
  9. -6- and -7-

    Not sure what you are looking at but if you click this link:

    http://news.google.com/news/more?pz=1&ned=us&cf=all&ncl=d4twtYgd2gpNK2Mn2DBwSIbiJ_0pM

    You'll come up with well over 1,000 articles covering the Mann paper in Nature. Google even has a graph of the coverage in the upper right.

    ReplyDelete
  10. Even a 26:3 ratio indicates a large bias. -- John M Reynolds

    ReplyDelete
  11. Concerning bias from so-called “journalists”, there is no better proof than that offered by the Columbia Journalism Review.

    As I noted in this post:

    When Slate.com sees a problem with so-called ‘Liberals’ propagandizing on behalf of a political religion, you’ve got to know that it is a big, BIG, BIG problem!”

    ReplyDelete
  12. "2. I just redid the search as

    hurricane "michael mann" nature

    and came up with 1,383"

    hurricane landsea noaa

    returns 1,392 results.

    In both cases, very few articles actually cover the studies in question. Your post is misleading.

    ReplyDelete
  13. -New York-

    Try restricting your search to Aug 13 and 14 2009.

    hurricane "michael mann" nature = 1,412

    hurricane landsea noaa = 0

    http://news.google.com/news/search?pz=1&ned=us&hl=en&as_q=hurricane+nature&as_epq=michael+mann&as_oq=&as_eq=&as_scoring=r&btnG=Search&as_qdr=m&as_drrb=b&as_minm=8&as_mind=13&as_maxm=8&as_maxd=15&as_nsrc=&as_nloc=&geo=&as_author=&as_occt=any

    http://news.google.com/news/search?pz=1&ned=us&hl=en&q=hurricane+landsea+noaa&as_drrb=b&as_minm=8&as_mind=13&as_maxm=8&as_maxd=14

    Try again ;-)

    ReplyDelete
  14. Click here for more proof of the breathtaking climate change bias from so-called “journalists”.

    Columbia Journalism Review describes that link as “Everything You Wanted to Know About Climate Change - Some key Web sites for journalists”.

    Does anybody see even ONE skeptic site on that entire list?

    Christ! They even omitted the #1 science blog for 2008 - WattsUpWithThat (which, as it happens, features Roger’s father in the most recent post)

    The Blog List in my blog (see the sidebar on the right) includes a wide assortment of very credible skeptic sites -- for those who prefer not to be dragged around by the nose by self-described propagandists posing as “journalists”.

    ReplyDelete
  15. Why restrict it to just those days, when the NOAA release came out on the 11th? If I set the range from the 11th to the 12th (the NOAA report was out on the 11th), I get the 1,392 result. Biased? You decide.

    Regardless, you are missing the broader point, that the vast majority of the search results in both cases don't cover studies in question, rendering your statement:

    "1,264 = the number of news stories covering Michael Mann and colleagues' new paper claiming that Atlantic hurricanes are at a 1,00o-year high."

    grossly inaccurate and misleading.

    Speaking of media bias, remember the media storm over the following GISS correction, which had a negligible impact on global trends?

    http://data.giss.nasa.gov/gistemp/2007/Fig4_correction.gif

    However, how many media reports covered upward corrections to the satellite record (particularly UAH) with actual serious implications for the global trend and climate science? There have been several significant ones over the last decade (i.e. diurnal drift).

    http://www.nsstc.uah.edu/data/msu/t2lt/readme.18Jul2009

    ReplyDelete
  16. Roger -

    Like New York, I still disagree with your search technique, and 2nd NewYork's point in 14.

    If you news.google hurricane "michael mann" the first hit is a group of 1394 articles, many of which are totally irrelevant. If you substitute "chris landsea" you get the same 1394 articles. Clearly the search triggers some kind of topic grouping on google, that may or may not involve the author. If you restrict the dates, you still get the same group for mann but not for landsea, but I'm not sure what that actually means. In any case, your searches in 13 still aren't conceptually the same.

    I think you'd have to go source by source and see who has covered each paper (and then correct for base rate visibility) to get a solid claim of bias.

    Here's a fun one
    http://www.google.com/trends?q=hurricane&ctab=183661152&geo=all&date=all

    In 3 you argue that the USA Today slant (more medieval hurricanes) is contrary to Mann's statements, but Mann says things like "...our key conclusion (that levels of activity during the Medieval era might have equaled or even exceed current levels of activity) is actually strengthened, not weakened."* Maybe that's not all Mann says, but it would seem that USA Today got it right. Ironically, they now have an article titled, "Study: Hurricanes more frequent than in 1,000 years"
    * http://holocene.meteo.psu.edu/~mann/Nature09/responses.htm

    Tom

    ReplyDelete
  17. Roger, you've Peisered yourself.

    Please retract.

    http://norvig.com/oreskes.html

    ReplyDelete
  18. > 1,264 = the number of news stories covering Michael Mann...

    The fact that you thought searching for just a name (shared by a film director along with, possibly, other people) demonstrates a level of myopia and incompetence that is hard to comprehend.

    > hurricane "michael mann" nature = 1,412

    Nonsense. That search returns "about 13" hits: : http://imgur.com/ZhD7K.png Restricting the dates produces a small anomaly and reports "about 27" hits. "Chris Landsea" produces "about 14" hits for the same date range. Woo hoo! Conspiracy!

    What an embarrassing display of incompetence and delusion....

    ReplyDelete
  19. Here are some of the highly-relevant (NOT!) links from your restricted Mann search:

    Atlantic may soon see first named storm - NHC Reuters India - ‎Aug 14, 2009‎ NEW YORK, Aug 14 (Reuters) - The Atlantic Ocean could see its first named storm of the hurricane season in a day or two as a low pressure system off the ...

    Atlantic Weather System May Become Hurricane, Planalytics Says Bloomberg - Brian K. Sullivan - ‎Aug 13, 2009‎ Aug. 13 (Bloomberg) -- A system of thunderstorms in the Atlantic Ocean off the coast of Africa may develop . . .

    Guillermo could strengthen to hurricane in Pacific The Associated Press - ‎Aug 13, 2009‎ MIAMI — Forecasters say they're expecting Tropical Storm Guillermo to strengthen to a hurricane as it moves farther out to sea in the Pacific. ...

    Apparently there are 933 more just like this. Ain't point and click so much fun? You don't even have to use your brain!

    ReplyDelete
  20. "Your post is misleading."

    It's not misleading, it's plain stupid, but not half as stupid as the people who will defend it.

    Sorry to be so blunt about it. I'm also stupid, so I feel I'm allowed to say it. It takes one to know one.

    ReplyDelete
  21. Roger: If you click "sort by date with duplicates included" and you'll reduce those thousands of results to 16 results. I get 5 results for a similar search for stories between August 12 and August 15 on: hurricane noaa landsea.

    A 3-fold difference can, I think it's fair to say, be explained by the differing prominence of Nature and the Journal of Climate.

    ReplyDelete
  22. Looks like "27 to 3" isn't right either, but it's a step in the right direction. More accurate is not accurate, however. Perhaps in a few weeks, we'll have the correct figures, long after the political blogs have eagerly regurgitated "1,264 to 1" to the knee-jerk crowd.

    ReplyDelete