The reconstruction in Hegerl et al (J Clim) was previously used in Hegerl et al (Nature) which provided some information on the number of series and some other particulars, but did not identify the series. In order to show how farcical the Hockey Team claims of “independence” were, I made guesses last spring about the proxies that would be used in the Hegerl et al reconstruction using a principle of least independence. I made an extremely accurate emulation of the Hegerl et al reconstruction as published in Nature simply using the proxies in Osborn and Briffa 2006, describing the process as follows:
I emulated the CH-long blend using the predictions in my earlier post as follows. All of the 12 predictions are in the 14-series Osborn and Briffa  data set. I removed 2 series from the smoothed Osborn and Briffa data set (the Foxtail series and the Chesapeake Mg-Ca series) , took the average of the 10 series available in 1251 (that’s one more than CH so there’s an adjustment to come) and then scaled the average to the CH-long blend. I’ve obviously been able to replicate the CH-blend pretty accurately without them even having to say what proxies they used. Their weighting methodology is not an unweighted average of the proxies. So it’s hard to tell whether the remaining differences relate to weighting systems or different proxies. There’s at least one proxy that I’ve not matched. Also, I’d be surprised if Hegerl used the Alberta version from Luckman and Wilson  – they probably used an older version.
A few days later, I made the following prediction about what proxies would be in Hegerl et al:
1. Yang composite
3. Polar Urals – on balance, I expect the Briffa MXD version, but it could be the Yamal version. Either qualifies.
6. van Engeln
7. Greenland dO18
8. Jasper – on balance I expect the Luckman version, rather than the Luckman-Wilson version which is too recent, but either qualifies.
9. MBH PC1
There are 3 series in my previous guess, which were simply taken over from Osborn and Briffa without fully weighing maximum overlap. I am making three substitutions to be more consistent with usual Hockey Team choices.
10. Yakutia instead of Mangazeja. Mangazeja is a bit unusual; Yakutia is more consistent with maximal overlap and is used in DWJ06.
11. Quebec – I’m going to go with the Jacoby version rather than the Schweingruber version (which isn’t used outside of Esper/OB). either. One point for the Jacoby version; half point for Schweingruber version.
12. Tirol isn’t used outside of Esper/OB and isn’t maximum overlap. So something from the Jacoby treeline series, it could be the composite or it could be something like TTHH. One point for TTHH, half-point for the composite or half-point for another Jacoby treeline site. I’ll be mad if it’s Tirol.
I made a bet with TCO that I’d get at least 6 predictions right.
How Did I Do?
I think that even the most ungenerous reader would acknowledge that I nailed my predictions, far exceeding the terms of my bet – I pretty much got every location correct, I got most of the series versions exactly correct and, even where I did not get the version exactly right, I was nearly always substantially right. Here’s the report card.
this is the high resolution record (10-yr average) from Yang et al. (2002).
Taimyr Peninsula: this is from Naurzbayev et al. (2002) by way of Esper.
- w. Siberia: in order to avoid any heavy biases of the mean composite by a number of sites from one region, the west Siberia time series is a composite of three/four time series from this region: two “polar Urals” records east of the Urals -Yamal (Briffa et al. 1995) and Mangazeja (Hantemirov and Shiyatov 2002 – both by way of Esper et al.) and two records from west of the Urals (Hantemirov and Shiyatov 2002). The records from each side of the Urals were first averaged and then combined for the w.Siberia.short composite; the w.Siberia.long composite involved Yamal and the west.Urals composite. The sites from Esper have been RCS processed.
I haven’t figured out exactly what they’ve done, but I got the Polar Urals correct and I’m awarding myself a full point.
Mongolia: this is from the D’Arrigo et al. (2001) study. However, the full composite illustrated in this paper is not available. We reconstructed the composite from nine records from tree ring sites sent to the NGDC sites. The early growth part of the treering series from overlapping records was removed without further removal of low-frequency variability.
Again I haven’t figured out exactly what they’ve done, but I got the location bang on and I’m awarding myself a full point. Note the lack of availability of data to Hegerl. There’s something weird here as even Science was unable to get the Mongolia data as used by Esper, although they got the other 13 sites. I think that Esper must have lost the Mongolia version that he used.
- n. Sweden: this is from Grudd et al. (2002) by way of Esper.
No surprises here. Full point.
European historical: this composite was kindly provided by J. Luterbacher et al. (2004). I got slightly wrongfooted here.
I was right about a European documentary series, but they used Luterbacher starting in 1500 instead of van Engeln. I was sure that van Engeln was in because the CH-version in Nature started in 1251 which was the date that Luterbacher started. The plot in Figure A1 starts in 1500. Do I deserve part of a point here – it doesn’t matter.
west Greenland: this composite is from Fisher et al. (1996).
Again no surprise. Full point.
Alberta, Canada: this time series is also a composite of two different analyses of the 1997 reconstruction of Luckman (1997) – one is unchanged from Luckman’s paper, and the other (Athabasca) has been RCS processed by Esper et al. (2002). The correlation between these analyses is unimpressive (0.11); the records were simply averaged. Note that although the correlation with the decadally smoothed 30-90N instrument (land) temperatures varies greatly between the two records (0.14 for Athabasca, 0.82 for Jasper), the composite correlation is 0.84.
Full point, although I didn’t anticipate the average of the Luckman and Esper versions.
western U.S.: this time series uses an RCS processed treering composite used in Mann et al. (1999), and kindly provided by Malcolm Hughes, and two sites generated by Lloyd and Graumlich (1997), analyzed by Esper et al. (Boreal and Upper Wright), and provided by E. Cook. The Esper analyses were first averaged. Although there are a number of broad similarities between the Esper and Hughes reconstructions, the correlation is only 0.66. The two composites were averaged.
I didn’t anticipate the averaging of the PC1 and the foxtails (which were used in Osborn and Briffa and certainly within the Maximal Overlap concept. I think that I deserve a full point.
e.Siberia: the Esper et al. (2002) composite used the Zhaschiviresk time series from Schweingruber. However, this composite only went to 1708. We combined it with a ring width (by Schweingruber, available NGDC) series from the nearby Ayandina River site after removing the obvious growth overprint in the early part of the younger record. Ayandina River is at 6825-14310 from 1553 1991 russ063w. Zhaschieviersk RUS; 67N,142E; Schweingruber,F.H russ053w .
Got the right idea, but not the exact series. If I was scratching for points, I’d argue for part of a point. In terms of my equivocation about whether to go with Mangazeja or Yakutia (leaving Mangazeja off), it looks like they rolled Mangazeja in with Yamal somehow and it’s in after all.
Quebec: The situation with Esper’s Quebec reconstruction is somewhat similar to what was experienced for the Mackenzie Delta time series; the correlation between their Quebec record and the 30-90N average is only 0.25, partly because the time series ends in ~1930. Examination of the NGDC data base indicates that the original Esper et al. reconstruction appears to be from the Boniface site. A record from nearby St. Anne also shows many similarities to Boniface (r=0.66), extends closer in time to the present, but is also slightly shorter (the Boniface/St. Anne correlation is 0.70). Although the Boniface/St Anne composite has a very high correlation with the 30-90N (land) record (0.88), inspection of shorter records from Fort Chimo and No Name Lake showed a different 20th century response – earlier warming and late cooling. In order to preclude a Quebec composite from indicating a potentially unrealistic magnitude of late 20th century warmth for the whole region, we created a shorter composite of the four sites that averages records from a Fort Chino and No Name Lake composite after 1806. The new composite still shows significant warming in the 20th century, but not as extreme as the Boniface/St. Anne sites alone. It is not claimed that this solution represents the best possible way to deal with the conflicting evidence from Quebec; the problem can best be resolved if more long records from other regions of Quebec can be uniformly stacked together without any late-century adjustments.
The Ste Anne series is an alter ego for Gaspé. I think that I deserve a full point.
- Mackenzie Delta: The original time series (Szeicz and MacDonald 1995) provided by Esper et al. only had a 0.04 correlation with the 1880-1960 decadal average of NH temperature, which yields a very small weight if used for the hemispheric composite. We experimented with various other data from the National Geophysical Data Center (NGDC, http://www.ncdc.noaa.gov/paleo/ftp-search.html) to determine if other reconstructions for that area would yield more information for a hemispheric reconstruction. We found generally that proxy data for that region show little correlation with hemispheric mean temperature. We nevertheless included this site for the sake of completeness and in order to include as many long sites as possible.
I didn’t expect the Szeicz-MacDonald series, but it’s only used in Esper et al 2002. I was obviously uncertain about this series, but directionally got the right idea.
Anyhow, I think that my predictions of Hegerl proxy selection based on the principle of “least independence” have been massively vindicated and show how farcical the claims of proxy “independence” are.