As a result of refusals by Moberg, Sonchkin and Lauritzen, I’ve filed a Materials Complaint with Nature, which will hopefully result in the delivery of the data in less than geological time. In an email to me concerning a possible one year misdating of U.S. bristlecones, Moberg said that their "reconstruction does not contain any information on timescales shorter than 4 years". As I point out below, the SI contains an annual reconstruction, so this statement is not exactly correct. Here’s the complaint.
Sept. 9, 2005
Physical Sciences Editor
We haven’t corresponded for a while. You may be aware that Ross McKitrick and I continued our research on the paper by Mann et al. , attracting a considerable amount of attention, even getting mentioned in Nature. Some of the issues, which we discussed before (e.g. Mann’s refusal to provide source code) have attracted considerable interest outside the immediate climate science community, as I’m sure that you’ve noticed. I believe that there are substantial issues involved and I hope that Nature re-considers some of its policies in this matter. However, as I’ve mentioned before, while I disagreed with your conclusions and decisions, I appreciated the cordiality of your correspondence.
I am writing today with a Materials Complaint in a different matter, which I hope can be resolved more expeditiously than the Mann et al. case. I have been examining the paper by Moberg et al.: Highly variable Northern Hemisphere temperatures reconstructed from low- and high-resolution proxy data Nature, Vol. 433, No. 7026, pp. 613 – 617, 10 February 2005. I have consulted the Nature on-line Supplementary Information and corresponded with both of the two senior co-authors without resolving the matters reported here.
1) The Norway stalagmite series from Lauritzen et al  appears to end c.1938 in the Nature SI, whereas Figure 11 of Lauritzen et al  indicates an end c.1865. I have requested a digital version of the series as used, both from Moberg and Lauritzen. It has been refused by both. Moberg referred me to Lauritzen, who said that "These are unpublished data, and they come with co-authorship." Obviously, the data is not unpublished if they have been used in Moberg et al . Moberg used the data under Nature policies which require that the data be made available to researchers. Moberg had an obligation to obtain consent to such usage prior to using the Lauritzen data and cannot rely on Lauritzen’s present refusal as justification for failing to provide the data.
2) The Indigirka series is not archived anywhere. Moberg said that his “Russian colleague Dmitry Sonechkin got this series from one of his Russian colleagues.” Sonechkin then said that he cannot send the Indigirka series used in Moberg et al  because the "series developers do not want to disseminate it. They say this series will be re-calculated soon to reject some errors in it (a general trend etc.)." Again, in order to use the Indigirka series in a Nature article where there is an obligation to provide data, Moberg and Sonechkin had an obligation to obtain consent from the series developers prior to using the data. If the series developers did not want to disseminate it, then Moberg and Sonechkin appear to have breached their understanding with the series developers by including the series in their Nature article. Further, if the series developers now change the series, this will be of no assistance in replicating the data used in Moberg et al .
3) Moberg et al  provide separate listings for Methuselah Walk and White Mountain Master. These are two different versions of the Methuselah Walk chronology ( the White Mountain Master (ca506) is a 1962 version, while the Methuselah Walk version (ca535) is a new edition made in the 1980s. Moberg says that he was aware of this and thought, but did not know, that different trees were used in the two studies. It seems to me that Moberg et al  should have clearly stated that these two series both refer to the same site and justified this duplicate use.
4) Moberg said: "from discussions with Dmitry, that we had some problems with deciding to which year we should assign the WUS-tree data. It is not impossible that, for example, the year 1962 in my file actually corresponds to 1963 in the programs that the Russian colleagues used to make the reconstruction. Honestly, I am not quite sure about this." Moberg goes on to point out that their "reconstruction does not contain any information on timescales shorter than 4 years" and that “It would also be nearly impossible to see by eye any difference between two versions of the reconstruction with the WUS trees shifted one year back or forwards." The first part of this statement is incorrect. Moberg et al. archived a reconstruction expressed on an annual basis at http://www.nature.com/nature/journal/v433/n7026/extref/nature03265-s6.doc. Sonechkin said that the series were located at the ITRDB data bank. I am aware of this, but this does not with the problem identified by Moberg.
Thank you for your consideration.