Friday, May 18, 2012


cc:,, Malcolm Hughes <>,,,
date: Thu, 01 Mar 2001 12:08:02 +0000
from: Phil Jones <>
subject: Re: letter to Science
to: "Michael E. Mann" <>, "Henry N. Pollack" <>, "Michael E. Mann" <>

Dear Mike et al,
Here's my input. I'll be here tomorrow till about 11am British time then I'm away all
next week. Maybe you'll have time to do something today and email another copy
at the end.
First a minor thing. Science's new endnote style means that you can have only
one reference in each and a wordy endnote with a reference has to have this referenced
as a separate endnote. So you'll need to be very careful which 15 or so are chosen.
Two other small things. First we should not stoop to the skeptic level and use the
phrase 'hockey stick' in print. Just say recent. Also I think the references you do use
should be mostly earlier ones than 1999 or 2000 (except for our few recent compilation
papers) so they are clearly references that have been around sometime and if Broecker
is going to write such an article he should have been aware of.
Specific points.
First, I think it is a bit long - I'll suggest reductions.
1. Sentence 2 'To reach this conclusion ...' . Add in here that these records are
calibrated against instrumental records. Just say ... that are calibrated against
instrumental records and are the foundation ...
Could add and verified as well after calibrated. This will make it clear in the
second sentence that Broecker is rejecting the standard practice in much of
2. Just reference Hughes and Diaz (possibly Lamb as well) re MWP.
After this sentence could say:
The two warming periods (~1920-45 and since 1975) in the 20th century show patterns
of warming (which predominate) but there are a few regions that show cooling. Even
during these two strong warming periods (~0.4C over the NH) few of temperature
trends are locally significant ( Jones et al., 1999 Rev Geophys.) .
3. The final sentence of (1) is the most important of all. Needs to be emphasised
and expanded upon. Thompson estimates that some of these ice caps/glaciers
(which have ice layers in the Middle Ages) will disappear in the next 20 years.
This is our best piece of evidence. I know we don't use them in many of the compilations,
but if these areas had layers in the Middle Ages and are disappearing arguments that
globally it was warmer in the MWP are shot to pieces.
4) Remove the whole of the middle of section (2) from Cook to the sentence begining
'Several.. ' . This material was in the papers but it didn't convince Broecker then, so
it's not worth repeating here. Just reference Cook - not the seg length curse, and
possibly add Briffa et al (2001) if this is out in JGR. Don't say that
always take these trends into account in interpretation because they clearly don't.
Try and keep the text short, pointed and don't leave ends that can be used against us.
I'd make the points after several into a different bullet point. This just needs to
say that even the LIA wasn't ubiquitous cold from the 15th to the 19th centuries.
The 'near instrumental' period - trying to think of name like the Middle Ages for the MWP
to use instead of LIA - is characterised in many regions by a cool 17th and 19th centuries
and a milder 18th century. It wasn't ubiquitously cold. Long European temperature
records clearly show this - and they agree with borehole evidence back to 1659.
5. The glacier point could perhaps be a separate bullet. Broecker here thinks you
can develop an NH glacial record which he says can be related to temperatures.
Glaciers respond mainly to summer temperatures. Need to compare single glaciers
or a regional average if they respond similarly. They can't be lumped. The reference
is for northern Fennoscandia :
Raper, SCB, Briffa, KR and Wigley, TML, 1996 J. Glaciology 42, 341-351.
6. (3) is mostly OK. Need to say decadal-scale variance. Also we know from the
instrumental period that the two main factors explaining much of the variability are solar
and volcanoes. Why bring in something else before you've exhausted these two.
Solar is the important one here. The record isn't great but it is the best we have. It is
good job the 14C or 10Be records didn't have a 1500 yr cycle as Broecker would have
had the cause of his supposed 'cycle' . He has looked for this, but can't find anything.
7 (4) I think the first part of this is week and like the arm-waiving we are accusing
Broecker of. Hopefully Tom Delworth can help here. I don't know much about this,
but what is there doesn't sound at all convincing.
8 (4) Second part. Isn't that convincing either. Final para is fine.

At 13:35 28/02/01 -0500, Michael E. Mann wrote:

Hi Henry,
None of us are disputing that conditions were relatively warmer in the earlier centuries
(ie, 11th-14th) than in the later centuries (15th-19th), but the main point is that they
were not as warm as the latter 20th century. At the face of it, your GRL article does
certainly seem to suggest otherwise, and this how Wally has interpreted it. My point was
that there is a sensitivity of that conclusion to a priori constraints. What we need
then is a clear and defensible statement from you as to why that paper isn't in conflict
with what we're saying. If you can suggest a specific rephrasing to replace what we
have, I would appreciate that.
I thought you and Tom had already consulted on this, but perhaps I'm mistaken.
Thanks in advance,
At 01:22 PM 2/28/01 -0500, Henry N. Pollack wrote:

Hello everyone!
On Wed, 28 Feb 2001, Michael E. Mann wrote:
> ........................................... Although one analysis of
> heat flow measurements suggests warmer temperatures than the surface
> proxies during the Middle Ages (Huang and Pollack, GRL. 1997), the
> considerable sensitivity of the resulting trends to a priori statistical
> assumptions has lead borehole researchers to restrict their attention to
> the more reliably interpretable temperature fluctuations during the past
> five centuries (Huang and Pollack, Nature). .........................
Henry Pollack comments on the above statement:
Linking the two geothermal studies (1997, 2000) is not quite kosher. We
did not later "restrict our attention" to the last five centuries because
of the considerable sensitivity to the a priori assumptions in the 1997
GRL paper. Throughout the range of the a priori assumptions there is an
indication of warmth in the middle ages. But as I mentioned in my comments
to Tom Crowley yesterday, whether that warmth exceeded the end-of-20th
century temperatures is perhaps debatable. I said to Tom that the hockey
stick will not rise or fall on the basis of the 1997 GRL paper that
analyzed heat flow variations with depth.
Our later emphasis on the past five centuries was intended to bring into
sharper focus the late pre-industrial and industrial eras. The five
century study reported in Nature used much higher quality data, actual
temperature vs. depth data rather than inferred heat flow vs. depth data.
It is no secret that the results of this five century study show some
disagreement with the hockey stick also, in the magnitude and timing of
the LIA minimum. That is another issue, not one directly addressing Wally
Broecker's discussion of the MWP.
p.s. I will be traveling as of this afternoon (2/28), returning on 3/5. I
would be grateful to receive a final copy of what you submit, but
respectfully decline to sign on as a co-author.

Professor Michael E. Mann
Department of Environmental Sciences, Clark Hall
University of Virginia
Charlottesville, VA 22903
e-mail: Phone: (804) 924-7770 FAX: (804) 982-2137

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

No comments:

Post a Comment