Thursday, January 19, 2012

2064.txt

date: Fri, 20 Jun 2003 16:42:25 +0100
from: Phil Jones <p.jonesatXYZxyz.ac.uk>
subject: Re: Sceptics' discourse
to: Timothy Carter <tim.carteratXYZxyzaristo.fi>,k.briffaatXYZxyz.ac.uk

Tim,
Thanks for this. I've been in touch with this guy (Steve McIntyre) before. I think he
works
in the US. He asked me a few things about the instrumental data, then more, then more
and asked for more data. I eventually gave up but he is quite able.
The Finn is Timo Hameranta (or something like that) and is right of right field !
Cheers
Phil
At 09:57 18/06/03 +0300, Timothy Carter wrote:

Dear Phil/Keith,
It was good to see you both last week.
I am copying you the English language part of an email I just received from a local
sceptic (a lawyer with no climate training at all!) which refers to some of your work. I
usually dispose of these mails as they come in, but this one seems to be calling into
question some published science. McIntyre is probably well known to you, but just in
case here it is. Sorry, but the figures referred to in this mail did not get through our
firewall.
Best regards,
Tim
From: [1]Steve McIntyre
To: [2]Climate Sceptics
Sent: Sunday, June 15, 2003 9:27 PM
Subject: [Climate Sceptics] More on Mann and Jones Datasets - Fennoscandia
Dear all,
More on the underlying datasets in the millenial datasets of Mann; Jones etc.
One of the prominent data compilation datasets is Fennoscandia, used in both
Bradley-Jones 1992 and Mann 1998, derived from Briffa et al. In fact, Briffa makes
several essays at temperature reconstruction for northern Fennoscandia. There seem to be
three main variants: (1) BETA1 is reported in Briffa et al 1990 (Nature) and is based on
cubic splines; (2) a RCS version, both uncorrected and corrected , is reported in
Climate Dynamics 7 (1992), where it is compared to the cubic spline version in a useful
way (clearly showing the inappropriateness of cubic splines for long-term data
analysis). Figure 8 in Clim Dyn 7 matches Fig 2(2) in Bradley-Jones 1992 and my graph
produced from the Mann 1998 proxy 67 which all seem therefore to be the same dataset.
(3) a third reconstruction is reported on Fennoscandia by Briffa and Schweingruber in
Climate since 1500AD (1992) data on which is at [3]www.ngdc.noaa.gov. In C1500, Briffa
and Schweingruber do not reconcile to the prior discussions. I have graphed this dataset
together with the other one below. The correlation between the two datasets for the
overlapping period 1587-1975 is 0.032 an interestingly low correlation for what are
reconstructions produced from relate data. I have shown the two
The Climate Dynamics 7 reconstruction contains a fudge by Briffa et al, described as
follows:
The density chronology shows a low-frequency decline over the last century which appears
anomalous in comparison with both the TRW data and the instrumental data over the 19^th
and 20^th centuries. These facts suggest that the density-coefficients in the regression
equation may be biased as would be the case if the density decline were not climate
related (CO2 increases and/or the potential effects of increasing nitrogen input from
remote sources may be implicated here.) &The residual MXD data (actual estimated) are
plotted in Fig. 7. A systematic decline is apparent after 1750. By fitting a straight
line through these residuals (1750-1980) and adding the straight-line values (with the4
sign reversed) to the RCS density curve, the anomalous post-1750 decline was removed.
This corrected RCS curve was then used along with the RCS ring-width curve in a final
reconstruction of the April-August temperatures.

This hardly seems like justifiable statistical procedure.Without the fudge, the
"reconstruction" shows declining temperatures in the 20th century. A very similar
decline in residuals occurs from 1100 to 1250 and one wonders whether a similar
adjustment would be allowable then.

The Climate Dynamics article does not contain a description of the regression
methodology and I have not yet consulted the predecessor article describing the
regressions. Suffice it to say that the tree ring data is highly autocorrelated, as is
(to a lesser extent) the temperature data. The meaning of such correlations is not
clear. The reconstructions end up being a weighted sum of the tree ring widths over two
summers and MXD s over two summers. The coefficients are very unstable under different
reconstruction methodologies.

Regards, Steve McIntyre

Prof. Phil Jones
Climatic Research Unit Telephone +44 (0) 1603 592090
School of Environmental Sciences Fax +44 (0) 1603 507784
University of East Anglia
Norwich Email p.jonesatXYZxyz.ac.uk
NR4 7TJ
UK
----------------------------------------------------------------------------

No comments:

Post a Comment