Thompson #1: Dunde Ice Core Versions

 

The Dunde (Himalayas) ice core is a staple of multiproxy studies (MBH98, Crowley and Lowery  [2000], Yang et al [2002] used in Jones and Mann [2003],  Jones and Mann [2004]). In our consideration of Crowley and Lowery [2000], it was one of 4 series that contributed the only hockey-stick- ness to the results.

 

Although the core was taken in 1987, it has not been archived at WDCP. I have located 3 different and inconsistent versions of this dataset in "grey" versions:

  • The most recent version is a decadally smoothed version from 1000-1987, shown in Climatic Change 2003, Figure 5. As a result of my requests to Climatic Change, the decadally smoothed data was archived at Thompson's website here. The same data version was used in Jones and Mann [2004]; in the data set provided to me by Philip Jones (pers. comm., July 2004), the 3 Thompson Himalayan cores were shown with Andean data (and vice versa). This does not appear to have affected the graphs in Jones and Mann [2004], but I think that some of the calculations in Table xx are reversed.

  • a "grey" version is archived in connection with MBH98 at Mann's FTP site; this is annual from 1600-1987.

  • the Dunde ice core was used in Yang et al. [2002], which in turn was used as a proxy in the multiproxy study Jones and Mann [2003]. Yang sent me the version which he used - this is a 50-year average from 200-1987.

The series has been published in print form, but I have not yet tried to collate the print versions to the digital versions. I have attempted to have the unprocessed data archived (for reasons that will be shown below), but these attempts have been unsuccessful so far.

 

Figure 1 below compares a smoothed version of the CC2003 version to the Yang version in their period of overlap. The discrepancy is obvious.

 

 

Figure 2 below compares a smoothed version of the annual Dunde data used in MBH98 (11-year smoothing) with the CC2003 decadal version, again with obvious differences.

I have been trying for nearly 18 months to get an accurate archived version of Dunde (as well as Guliya and Dasuopu). It seems to me that, with the proliferation of inconsistent versions, there is an obligation on Thompson to archive the original sample information, together with a reconciliation of the different results.

 

 

 

 

Regards, Stephen McIntyre