cc: Malcolm Hughes <mhughesatXYZxyzr.arizona.edu>, Tim Osborn <t.osbornatXYZxyz.ac.uk>, Keith Briffa <k.briffaatXYZxyz.ac.uk>, "Raymond s.bradley" <rbradleyatXYZxyz.umass.edu>, Phil Jones <p.jonesatXYZxyz.ac.uk>, Scott Rutherford <srutherfordatXYZxyz.edu>
date: Wed, 12 Nov 2003 13:19:12 -0700
from: Tom Wigley <wigleyatXYZxyz.ucar.edu>
subject: Re: fine
to: "Michael E. Mann" <mannatXYZxyzginia.edu>
I will check out possibilities here. My thinking is that the only way to truly squash M&M
is to have an independent third party come along and say ... I used exactly the same data
and method as MBH and got exactly the same results, and, furthermore, I endorse the
method. I will read your paper with interest -- this will be a good putdown, but M&M
may still say that you are a mutual admiration society.
Michael E. Mann wrote:
Fair enough on all counts. You know how this works--hard to get every single nitty
gritty detail in the short Nature space, then someone comes along, obviously w/ hostile
intent, and your inclination to help them is limited--then they turn around and say you
didn't disclose the data, methods, etc. (which is at least partly an outright lie,
though there is a minor kernal to the claim that they can try to grab on to, because the
methodological description was terse).
Actually, Tim, Keith, Phil, Ray, Malcolm, Scott and I are all planning to pursue a much
more careful intercomparsion of results, methods, etc. We have a paper, the draft of
which I'm forwarding separately as it is pretty big (in review J. Climate) which should
go a long way in this regard. It controls for different datesets and methodologies, and
shows that the results are basically robust, with the conclusion that spatial and
seasonal sampling seems to matter (as we would expect) the most, but results seem pretty
robust with respect to statistical methodology (if you've done it right!). Would have
been nice if this were in the press right now, but alas its still in review...
Nonetheless, wouldn't be a bad idea to have some graduate students, or some NCAR
postdocs(?) try this--I'd be happy to help out where I can, but be hands off too to keep
the effort independent.
Let me know what you think...
At 12:36 PM 11/12/2003 -0700, Tom Wigley wrote:
OK, Mike. So you are choosing my option 2 (rightly so).
But there are broader issues, and it may still come down to option 3. Perhaps
a middle ground would be to try to get one of the people I named to get
the data and do an honest and informed version of what M&M tried to do?
It would be a nice student's warm up exercise, at the beginning grad student
level in a stats dept.
>From the flurry of emails, there may still be some things about the method that
you would have to pass on. I must admit that, having read the papers, I don't
think there is enough information for *me* to reproduce what you have done.
I could certainly do something similar, and I might discover the nuances as I
proceded. But it would still be tough.
I still don't think that hard-earned data needs to be made freely available.
Professor Michael E. Mann
Department of Environmental Sciences, Clark Hall
University of Virginia
Charlottesville, VA 22903
e-mail: firstname.lastname@example.org Phone: (434) 924-7770 FAX: (434) 982-2137