from: Tim Osborn <t.osbornatXYZxyz.ac.uk>
subject: Re: Fwd: right data?
to: Keith Briffa <email@example.com>,Jan Esper <esperatXYZxyz.ch>
Keith forwarded your message and excel file to me to take a look at. I
have not had a chance to do so yet, but one thought occurred to me...
The "interannual" correlations listed in our Table 1 were (see footnote of
Table) computed between "timeseries of residuals from decadally smoothed
curves". In other words, we used high-pass filtered series to compute
them. The filter we used had gaussian-shaped weights, chosen to remove
time scales longer than about 10 years.
If you didn't filter the MXD and temperature, then that might explain the
>>Date: Tue, 16 Mar 2004 15:47:03 +0100
>>To: Briffa Keith <k.briffaatXYZxyz.ac.uk>
>>From: Jan Esper <esperatXYZxyz.ch>
>>Subject: right data?
>>I attached an Excel file with some correlation numbers that might deviate
>>from the ones provided in your 98 Nature paper.
>>The file contains two spreadsheets (Source, MXD_Eurasia).
>>"Source" is the original data sent by Tim.
>>I copied then some relevant data to the sheet "MXD_Eurasia"
>>At the bottom of the columns, I calculated correlations over the period
>>1881-1981. These are 0.56 for CSIB, 0.75 for WSIB, and 0.85 for NEUR (MXD
>>against Apr-Sep temperatures).
>>Table 1 in your paper, however, says 0.51 for CSIB, 0.83 for WSIB, and
>>0.87 for NEUR (Column "Intrannual" Period 1881-1981).
>>Since this (and other numbers for TRW) are not SYSTEMATICALLY off, I am a
>>bit worried that I got the right data. Of course, it could also be that I
>>did a stupid mistake. In this case, I will get a beer for you on Monday...
>>I just need some confidence to continue with the data Tim sent me.
Dr Timothy J Osborn
Climatic Research Unit
School of Environmental Sciences, University of East Anglia
Norwich NR4 7TJ, UK