Upside Down Tiljander is BAD — Who knew?
Posted by Jeff Id on November 23, 2011
Wow, my whole post vanished. I’ll try again but this is very disappointing.
I’ve read about 5 percent of the emails so far and at this point the main conclusion I come to is that the scientists are far more skeptical in private than they are in the public. They don’t say dammit, we know for certain that today’s warming is the greatest ever, they say that it appears to at least equal the historic level. They say that they are unsure of the results and often comment that the data isn’t good enough. This is all in private of course. What we get in public is the certianty that they are correct. We get comments from Mann that the finnish varve sediments can be used either way – upside right or down but the emails discuss corrections to published papers for the same problems. Below are a couple of graphs taken from climate audit. My blog is acting up so you will see them as I am writing rather than after I’ve finished.
This is what Mann had to say about the matter in reply to an official comment by McIntyre and I believe McKitrick on the use of the data in the upside down orientation shown above:
The claim that ‘‘upside down’ data were used is bizarre. Multivariate regression methods are insensitive to the sign of predictors. Screening, when used, employed one-sided tests only when a definite sign could be a priori reasoned on physical grounds. Potential nonclimatic influences on the Tiljander and other proxies were discussed in the SI, which showed that none of our central conclusions relied on their use.
Of course we get an entirely different and correctly skeptical view of using a temperature series upside down from the behind-the-scenes emails below. My bold.
date: Wed, 23 Sep 2009 16:49:47 +0100 (BST)
subject: Re: more from McIntyre
to: “Darrell Kaufman” <Darrell.Kaufman@nau.edu>
sorry to be out of loop – [[[redacted: health]]]
- today I went in to university but will not be
officially back til 1st Nov. I will try to respond in detail tomorrow -
but I do not know the basic crux of the Yamal complaint – I suppose I
must spend time on the Climate Audit website – but have been loathe to do
so as many criticisms that have come to my attention seem crass or
misguided. Please prompt again if you can point me to the specifics of the
problem with Yamal – as for other long chronologies in Eurasia , I am
aware of one further East than the area upon which we focussed in our
Phil. Trans. paper – published I believe in a conference abstract only -
the raw data are not , I believe, available until Malcolm Hughes releases
them . The chronology that exists (produced by Russian colleagues) might
be available and I will try to find out – you could ask Anders Moberg for
the data directly – this would help me in the meantime – I need to know
more specifics re Yamal. I do not believe there are long series other than
those you used but will also have to check the 60 north criterion to be
> Hi Keith:
> I realize that you are out of the office and I’m sorry to bother you
> about this again, but the recent Arctic warming paper seems to have
> had a major impact. Because it’s in the lime light, I think it’s
> important that we publish an erratum to correct my error in reversing
> the Finnish lake-varve series. Two other authors also found minor
> errors in their proxy time series and we will include these revisions
> as well. None of the revisions change the conclusions, but we want to
> be as accurate as possible.
> As we are preparing for to publish the erratum, it would be helpful if
> you could double check that we have included all of the most current
> tree-ring time series that meet the criteria of the study:
> – north of 60N
> – at least 1000 years long
> In addition the the Gulf of Alaska, we included the three from your
> 2008 Phil Trans paper: Fennoscandia, Yamal, and Avam-Timyr
> Did we miss any? It’s important because we claim that our study
> includes all of the published proxy data that meet these criteria. Is
> there a published record from the Indigirka River region that we
> neglected to include?
> Regarding the criticism that the new Yamal series looks so much
> different than the earlier Polar Urals series, I assume that the RCS
> approach that you used for the 2008 study supersedes and is superior
> to any previous analyses. Is that a fair statement?
> On Sep 8, 2009, at 10:12 AM, email@example.com wrote:
>> I do not read his blog – if YOU ask him for more detail of what he
>> (it can not be Grudd’s work) then I will think about it. Will be
>> back in
>> touch before end of week on other stuff.Keith
>>> Hi Keith:
>>> this from McIntyre:
>>> Briffa’s Yamal series, which has been a staple of these sorts of
>>> studies for many years. It would be highly desirable for someone
>>> do a detailed reconciliation of why the updated version of Polar
>>> yields such different results to Briffa’s Yamal series.
>>> I assume he’s referring to Grud’s (sp) recent work. Let me know if
>>> have a quick explanation.
What we skeptics (is that what we are?) know though is that upside down data does reverse the conclusions of far more popular hockey sticks.