Tuesday, October 30, 2012

Karolygate II

Read Karolygate I first
 __________________

AFTER deficiencies in their paper were pointed out and with efforts to stonewall outside examination of the data not going well, David Karoly writes to two colleagues about what might be done to save the rotten fruit of three years' work and God only knows how much money. The email was sent in early June:
Hi Raphi and Joelle,
Following some email discussions with Mike Mann and helpful discussions with you both last week, there appear to be several different approach es that we can take with revising the Australasian temp recon paper. I am going to go through some of them briefly, and then raise some suggestions for further data analysis that might be needed.

1. Amend the manuscript so that it states the actual way that the proxy selection was done, based on correls that included trends and were significant at the 5% level. The calibration was also done using the full data variations, incuding trends, over the calibration period. As Mike Mann says below and in the attached papers, this is a common approach. Don't seriously address the proxy selection for detrended data

2. Revise the manuscript to present results for reconstructions based on both proxy selections for full correls and proxy selections for detrended correls. Expand the paper to show both sets of results and explain why the full correls are better.

3. Re-do the analysis for proxy selection based on what the manuscript says, proxy selection based on detrended carrels, which gives only about 9 selected proxies and only one prior to 1400. No reliable reconstruction prior to 1400.

4 . Redo the analysis based on proxy correlations with local/regional temps at interannual and decadal timescales, not the Australasian area average; select proxies that have strong local temperature signals, then average the proxies to get the area average temperature. This approach is like what Raphi is doing for the SH paper, I think.

My preference is now for 1. or 2. above, and not for 3.....

Now re-read proposition #3 -- the one that concedes that, if the study had been done as advertised -- only nine proxies would have been relevant, with just one of those purporting to define climate prior to 1400AD. Obviously, that's not going to save the research and get the paper published.

Karoly's preference is to .... ignore that option!

Now look at option #2, one of Karoly's two favoured options, and don't forget that the paper originally sold the notion that Australian temperatures were at a 1000-year high on the strength of allegedly "detrended" data. That detrending wasn't done, which is why the paper had to be spiked, so now Karoly's idea is for everyone concerned to insist un-detrended numbers are the better option.

Just coincidentally, that approach saves the headline-grabbing claim that Australia is hot as Hades and poised to get a whole lot worse.

Sheesh.

9 comments:

  1. And in option 2 this sentence is a doozy,

    "Expand the paper to show both sets of results and explain why the full correls are better."

    Why would full now be better? Their first attempt was intended to use detrended - for a good reason you would expect - I mean, we're not just making this up as we go are we? That produced a hockey stick and they were happy with it.

    Then they discover that hockey stick can't be supported IF the originally preferred detrended method was used - which it wasn't, as it turned out.

    So what we do? Stick to the original method, the on e we thought was more robust, and produce a non hockey stick result? Or, change the preferred method so we keep the hockey stick?!

    You see, the preferred method is always the one which produces the hockey stick, but you can bet when the explanation was provided for why full was better it will be some other invention which at the time that email was written hadn't even been thought of yet.

    ReplyDelete
    Replies
    1. Actually Dave, even using their dodgy, hand-picked data they can't come up with a hockey stick. Indeed, using their methodology, they found that the temperatures around 1250 AD were 0.09 degrees C less than current temps. Yep, they claim that using this methodology they can calculate temps 750 yrs ago to a precision of two decimal places.

      It'd be funny (hilarious) if it wasn't for the fact that on such BS analysis we are spending the future.

      Delete
    2. Yes hazym, I should say their 1000 year unprecedented warming in the 20th century. Though it's still insisted that later reconstructions have reconfirmed Mann's earlier work even where they look nothing like it.

      Delete
  2. No 4 is a dodgy option as well.

    ReplyDelete
  3. The Old and Unimproved DaveOctober 30, 2012 at 7:40 AM

    The claim of being...or rather, of having been...a climate scientist is going to be death on future career prospects in the latter half of this decade.

    Which makes the term 'Curriculum Vitae' more than a little ironic.....

    ReplyDelete
  4. The release of the first Climategate emails in November 2009 was the disaster the alarmists predicted it to be. It simply laid bare the tactics used to further their shoddy work (self-selecting reviewers, pressurising critical journal editors, threats, PR, and all of the rest of it).
    They also recognised the huge threat FOI legislation had to their modus operandi and that has been their latest two-step focus – plan to reduce chances of FOI applying to their government-subsidised work by using private emails and trying to have FOI not apply to their work.
    With precise relevance to this latest release, Climategate also laid out quite clearly the tactics used in this case of the Karoly paper: try at all costs to stop Steve McIntyre getting any credit for exposing their “mistakes”. Climategate even had a farcical example email where a certain (litigation-ready) “scientist” proposed in answer to a McIntyre expose to say it was all wrong and that they would then have time to look at it and get something looking more complete to answer it. Not bad at predictions eh?
    No, this is more of the same. Astounding results blaring out around the world by their usual announcement by press release and when its fatally flawed nature is exposed, try and stop the critics from gaining the credit for the exposure.
    Climategate laid down all their schemes and the world will be forever grateful to RC (‘A miracle has happened’) poster onto Climate Audit in November 2009.

    ReplyDelete
  5. It was posted a few years ago but is still a very good guide on just how weak the statistical evidence is for the 20th Century having a rise in average temperature that can only be put down to a man made cause. It shows you how proxy data is dealt with and just how risky it is to make assumptions based on it.

    http://iowahawk.typepad.com/iowahawk/2009/12/fables-of-the-reconstruction.html

    The author as well as being an excellent blogger has worked professionally in the statistical area and if you have some time to spare, are reasonably strong at mathematics and can drive a spreadsheet well, you will gain considerable insight into the issue at heart of the Climate Change issue.

    ReplyDelete
  6. 'Normally', I am down to nature's covering only from dusk to dawn but the 'climate' this year has been so damned cold that I'm still clothed and submerged under a stack of blankets in order to feel warm during my slumber. WTF do these guys know.

    ReplyDelete
  7. I kind of like the fact that Option 4 was even considered by honest, serious scientists such as these obviously are:
    'select proxies that have strong local temperature signals, then average the proxies to get the area average temperature.'
    Upon the 'settled science' of these turkeys humanity is investing exactly how many billions???

    ReplyDelete