Monday, June 18, 2012

5283.txt

date: Fri Mar 14 16:15:09 2008
from: Phil Jones <p.jonesatXYZxyz.ac.uk>
subject: Re: Fwd: Re: Pielke et al
to: Thomas C Peterson <Thomas.C.PetersonatXYZxyza.gov>, David Parker <david.parkeratXYZxyzoffice.gov.uk>

David,
If I didn't know Tom better, I might have surmised that he had a beer
or two over lunch !
There is one issue about a full paper and a comment. For the comment
Roger gets the last say!
I agree with all Tom's points. If you want to keep the comment format, then
you'll have to reduce a bit and concentrate on the more important points. I think
you'll have to do much more to get a fuller paper.
Of the 7 that Tom has listed, I think we could omit #2 and 7.
On 2 we agree with Roger, but it wouldn't make much difference, as we assess
homogeneity with tests. I don't think a series of pictures will help that much.

On 7 we're agreeing anyway, so omit it.
The key ones to my mind are - in importance are
5 - the plot John will do will demolish this
4 - this makes little difference at large scales provided things like TOBS
have been taken out
As an aside the bias adjustments in the SST data are far more important
than anything up with the land. All you have to do a moderately reasonable job
with land and you'll be fine. Someday, one of these, will realise this. Roger
can't see this, as he's full of Tom's baloney. Why does everyone go on about the
land?
3. Willett et al show specific humidity goes up in line C-C relationship.
Paper has gone back today to J. Climate. 2 of the mildest reviews I've
ever seen, so I think it will be accepted pretty soon. The editor is making
work for themselves if this goes back for further review.
1 As there is little difference between Tx and Tn warming, if you can't
use Tn then you have to use Tx. Globally the DTR trend to lower values
has to be reduced to one third as the Tx/Tn concept have little meaning over
the ocean.
and then 6 - this is discussed elsewhere, so could be omitted.
He should be referring to Simmons et al (2004) and the CCSP report.
Have a good weekend.
It's been sunny here today and felt warm! It could be land use change outside
our building, higher vapour pressure, but I don't care - it feels warm and that
is what matters!
Cheers
Phil

At 18:12 13/03/2008, Thomas C Peterson wrote:

Hi, David,
My first thought is well, we'll just have to cut it way back. Then I pulled out
Pielke's paper and saw that mountain of baloney and thought where do we draw the line?
There is so much there that should be refuted.
To be pithy, we could just hit the central points with little elaboration:
1. Definition of global temperature (a) Roger gives a definition related heat content
and climate feedback. We give this definition: the average temperature of the earth.
(B) Roger says we shouldn't use minimum temperatures because they can be impacted by
wind. We say temperatures in the nocturnal boundary layer are temperatures that the
world, including plants, animals and us, experience and are therefore can not be left
out of global average surface temperature or it is no longer global average surface
temperature.
2. Lack of photographic metadata. Roger says this is a major omission because, if we
had them over time, they might document local changes unrelated to larger-scale climate
signals. We say they would be nice but they don't exist world wide and particularly
back through time, therefore we've developed statistical tests that identify
undocumented changes in the local environment and adjusts the data to account for them.
3. Surface water vapor. Roger says "ignoring concurrent trends in surface air absolute
humidity therefore introduces a bias in the analysis of surface air temperature
trends". We say baloney. Paying attention to them would introduce a bias. Like clouds
and solar energy, water vapor can impact the temperature. But the temperature is the
temperature no matter what the cause so do anything other than ignoring water vapor
would bias the record.
4. Uncertainties in homogeneity adjustments. Roger says there are uncertainties and
potential improvements that could be made in homogeneity adjustments. We agree, which
is why homogeneity research continues (reference, e.g., the Hungarian series of
conferences). But we should also note that the same is true with magnetic resonance
imaging in doctors' offices but we still rely on those data because the current
processing is the best that is currently available and gives reliable results.
5. Degree of interdependence. Roger quotes an off the top of his head answer to the
question rather than conducting any real assessment of the interdependence of climate
data to point out that of course they give the same answer. We should note that (a)
studies of subsets of the data have revealed essentially the same signal and (b) MSU
data are 100% the same but different groups come up with different results. So
processing can make big differences. Therefore, the fact that different sfc temp
analyses show the same thing supports the view that the signal is robust.
6. Relationship between obs and reanalysis. Roger says obs are wrong because they don't
agree with reanalysis for trends. However, a body of experts (ccsp 1.1) says it doesn't
trust reanalysis trends for many valid reasons.
7. Influence of land cover change. Roger says land cover changes can impact
temperature. We agree. If they are major regional changes, land cover produced changes
in temperature would be part of teh signal we want to capture. If they are local, then
the latest homogeneity adjustment methodology has been shown to remove them (Menne &
Williams I believe).
Conclusion: Roger is full of baloney.
There you go, David. Add in a few references and we have a paper!
Regards,
Tom
David Parker said the following on 3/13/2008 12:25 PM:

Phil, Tom
Thanks for your comments.
I have incorporated many of these including quite a bit of the "Menne
and Peterson" simulation of Hale et al. 2006 using SST. Should Matt
Menne become an author?
John Kennedy will create time series by sub-sampling alternate grid
boxes of the surface temperature fields. The results will affect the
wording so only when this is done will I will send you a copy.
The text, which now includes summaries of Pielke et al's main points, is
now nearly 4000 words plus references. This plus 2 diagrams is virtually
certain not to fit in 4 JGR pages which is the limit for "Comments" (see
[1]http://www.agu.org/pubs/comments_guidelines.html)
So we may need to reconsider what we include, or submit it as a paper in
its own right. The Hale et al simulation and the global sub-sampling
exercise may go beyond AGU's stipulation: "The Comment addresses
significant aspects of the original paper without becoming essentially a
new paper".
Anyway, I'll keep you posted on progress.
Regards
David
On Tue, 2008-03-11 at 12:08 +0000, Phil Jones wrote:

David,
Don't refer to the Chinese paper that I have submitted. I'm just sending this
for a few more places where the cancelling of homogeneity adjustments has been
shown.
The best of these is the French one. Figure 9 of the Caussinus/Mestre pdf.
I asked Olivier Mestre to produce a histogram of the adjustment factors
for the French Tx, Tn homogeneity. This is the png file, showing the bimodal
distribution. I took this to be the counts of adjustments of certain values.
What I'm suggesting you do is refer to Olivier's paper and maybe Menne and Williams,
as well as Brohan et al.
I got a bit carried away attaching all the things I have!
Cheers
Phil

Date: Mon, 10 Mar 2008 12:41:47 -0400
From: Thomas C Peterson <Thomas.C.PetersonatXYZxyza.gov>
Subject: Re: Pielke et al
To: David Parker <david.parkeratXYZxyzoffice.gov.uk>
Cc: "Jones, Phil" <p.jonesatXYZxyz.ac.uk>
User-Agent: Thunderbird 2.0.0.12 (Windows/20080213)
X-UEA-Spam-Score: 0.0
X-UEA-Spam-Level: /
X-UEA-Spam-Flag: NO
Dear David & Phil,
I think you have a great start on the rebuttal, David. While I have many minor
comments or edits in the attached version, there is one systematic change that I think
we need to make:
Each section should start out with a 1 or two sentence summary of Pielke et al.'s key
point. Then we add our stuff (which you've written). But then we need a 1 sentence
summary each time where we say, therefore Roger's point is (choose one or several):
irrelevant, not supported by the evidence or refuted by the evidence. I'd prefer some
stronger and less technical language, but I know you're too polite to write any such
thing.
Does that sound reasonable?
Also, Phil, I have two questions for you in my comments.
We do need to expect that Roger will want to pick any nits we have showing, so (a) we
should only state the barest and clearest of cases and (b) be ready for an onslaught of
babble.
Regards,
Tom
David Parker said the following on 3/7/2008 11:34 AM:

Tom
Thanks for your comments on Pielke et al. JGR 2007. I have incorporated
your thoughts into my draft response - please see attached.
Regards
David


--
Thomas C. Peterson, Ph.D.
NOAA's National Climatic Data Center
151 Patton Avenue
Asheville, NC 28801
Voice: +1-828-271-4287
Fax: +1-828-271-4328


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
----------------------------------------------------------------------------



--
Thomas C. Peterson, Ph.D.
NOAA's National Climatic Data Center
151 Patton Avenue
Asheville, NC 28801
Voice: +1-828-271-4287
Fax: +1-828-271-4328

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