cc: k.briffa@uea.ac.uk, "Phil Jones" <p.jones@uea.ac.uk>, "raymond s.bradley" <rbradley@geo.umass.edu>, mhughes@ltrr.arizona.edu
date: Thu, 30 Oct 2003 06:22:50 -0500
from: "Michael E. Mann" <mann@virginia.edu>
subject: Re: Fwd: op ed for USA Today
to: Tim Osborn <t.osborn@uea.ac.uk>

   Thanks a bunch Tim,
   Well, we didn't add your name because we weren't sure, but USA Today probably won't publish
   it--if not we may try to distribute it.
   But more importantly, as we speak, I am drafting a long description of what they done
   wrong. Just over the last 24 hours I've discovered something extremely dishonest that it
   appears they did. In their reconstruction based on their 'redo' of the MBH98 proxy network,
   the one that shows the ridiculous warming in the early centuries, it appears that they
   eliminated all of our ITRDB Western North American (and Stahle max latewood chronologies)
   from our network. As you guys know, the ITRDB WNA data are fairly important to our
   reconstruction. Based on Table 7.5 in their paper, if you read the fine details, it looks
   like they've just eradicated the earlier data because they claim they couldn't find it on
   the NGDC website--even though we all know the data are there. And more importantly, all of
   those data were on our public ftp site on holocene.
   So in one extremely dishonest stroke of data eradication, they removed the most important
   indicators from our network from 1400-1600--and I'm pretty sure that's how they get their
   spike. Would be interesting to see what cross-validation they get using *their* network
   available from 1400-present. I bet we're talking REs approaching negative infinity...
   So I think that is what they did! Do you guys have the paper--does anyone mind
   double-checking, and assuring that I'm correct about this. If I am, this is really
   scandalous, and it should be as broadcast as widely as possible. Note that they don't even
   report how many proxy data were available in their network back in time, they only show the
   # of reported/found proxies in the Mann et al network (apparently our data site was missing
   a few of the series). This is probably intentional as well--they didn't want to show how
   many series they had actually eliminated from the set. And of course, if they're using a
   completely different set of proxies, then the would have to reapply the selection rules,
   they can't just use the basis set that we had determined, based on application of the
   selection rules to the data at hand...
   So its looking increasingly dishonest, deceptive, and intentionally so. I've identified
   other problems, they used an incorrect version of the the Mann et al proxy dataset that
   Scott had put into excel format, so the early PC proxy series were overprinted w/ later
   ones kept in the same column. And they used inconsistent CRU surface temperature datasets
   and inconsistent normalization conventions to un-normalize the Mann et al EOFs, etc.  And
   all of this could lead to significant differences. But I think its the dropping of the key
   predictors w/ barely a mention, that gives them the AD 1400-1600 spike
   Second opinions--am I imagining this?
   Thanks,
   mike
   At 10:44 AM 10/30/2003 +0000, Tim Osborn wrote:

     At 17:45 29/10/2003, you wrote:

     We need to submit within the next hour or so, so its really do-or-die time!

     Mike,
     was away yesterday, so I missed all the fun-and-games!  If you went ahead and submitted
     it with my name on anyway, then that's fine because I would have agreed had I been
     here.  If you dropped me in my absence, then fine too - you had enough co-signees, I'm
     sure.
     Going back to an earlier email when you were asking whether anyone had reviewed the E&E
     piece by M&M (have I got the initials correct? have to avoid confusion with M&M sweets -
     do you get them in the US? some are nuts, which seems appropriate!).  Anyway, just
     wanted to confirm that I did not review it.
     Despite the hard and time consuming work that it evidently took you to get to the bottom
     of their work's problems, I think it was essential to get this cleared up so soon.  It's
     important to get this information out as publicly as possible, so that nobody who wants
     to push the M&M conclusions can do so while claiming ignorance of the fact that data
     problems make their conclusions baseless and wrong.  If you want to avoid the
     climatesceptics list then perhaps one of us (or all of us?) here in CRU could circulate
     a note to that list, hence the cc to Keith and Phil.  Let us know.  Do you ever use the
     CLIMLIST mailing list?  It's not generally a debating type list, but I'm sure it would
     be relevant to post something there that makes clear the M&M conclusions are invalid -
     as a public information service?
     Cheers
     Tim
     Dr Timothy J Osborn
     Climatic Research Unit
     School of Environmental Sciences, University of East Anglia
     Norwich  NR4 7TJ, UK
     e-mail:   t.osborn@uea.ac.uk
     phone:    +44 1603 592089
     fax:      +44 1603 507784
     web:      [1]http://www.cru.uea.ac.uk/~timo/
     sunclock: [2]http://www.cru.uea.ac.uk/~timo/sunclock.htm

   ______________________________________________________________
                       Professor Michael E. Mann
              Department of Environmental Sciences, Clark Hall
                         University of Virginia
                        Charlottesville, VA 22903
   _______________________________________________________________________
   e-mail: mann@virginia.edu   Phone: (434) 924-7770   FAX: (434) 982-2137
            [3]http://www.evsc.virginia.edu/faculty/people/mann.shtml

