Bunn et al.  have an interesting discussion of 20th century tree growth (especially foxtail pines) in the Sierra Nevada and White Mountains, in the current Holocene, which, needless to say, was interesting to me. The extraordinary and uncritical embedding of MBH98-99 in paleoclimate mentality recurs here in a curious way. Roger Pielke wondered whether the emphasis on deconstructing the hockeystick is warranted in policy terms ( a question that I mean to discuss some time); here is a case where the embedding of the hockeystick leads to problems in a scientific paper relying on Mann et al., without fully understanding its assumptions and defects.
Bunn et al. report on a network consisting of 13 tree ring site – 7 collected by co-author Graumlich and her students (including Bunn), 6 are Graybill series archived at WDCP. The Graybill series include some familiar sites to readers of this site: Sheep Mountain and Campito Mountain (both bristlecones) and 4 foxtails. The 6 sites include the strongest hockey stick shaped sites in the MBH98 NOAMER roster. Although the Bunn et al. network overlaps the MBH98 network, they describe it as significantly augmenting" the network.
Our 13 sites and five species significantly augment the existing North American tree-ring proxy indicator series (Mann et al., 1998, 1999) in two ways that enhance the interpretation of the record. First, our high elevation chronologies include three species not represented in previous work. In addition to foxtail pine (Pinus balfouriana) and bristlecone pine (P. longaeva) we contributed western juniper (Juniperus occidentalis), limber pine (P. flexilis) and whitebark pine (P. albicaulis).
Here is a listing of the Bunn et al.  sites. The 7 new sites differ from sites reported in other articles by the authors, but no explanation is given as to the selection here.
|Spillway Lake||PIAL||3305||37 50||119 13||N|
|MP||Mammoth Peak||PIAL||3350||37 52||119 17||N|
|BP||Boreal Plateau||PIBA||3420||36 27||118 20||N|
|UWL||Upper Wright Lakes||PIBA||3510||36 37||118 22||N|
|HAM||Hamilton||JUOC||2630||36 34||118 39||N|
|CRA||Crabtree||PIBA||3378||36 34||118 22||N|
|LU||Lundy Lake||PIFL||2925||38 3||119 14||Y|
|CIR||Cirque Peak||PIBA||3505||36 27||118 13||N|
|FLO||Flower Lake||PIBA||3291||36 46||118 22||Y|
|SHP||Sheep Mountain||PILO||3475||37 22||118 13||Y|
|CAM||Campito Mountain||PILO||3400||37 30||118 13||Y|
|TGL||Timber Gap (Upper)||PIBA||3216||36 22||118 37||Y|
|TGU||Timber Gap (Lower)||PIBA||3017||36 22||118 37||Y|
The above comments have to be taken with a grain of salt. There are actually 8 JUOC and 8 PIFL sites in MBH98 (although none in the 1000-1399 MBH99 addition). It’s hard to see how the PIAL sites (SL and MP in the first row below) show anything unique about the 20th century – so the addition of these sites and this species is simply window-dressing. The juniper site (HAM) is unambiguously precipitation-controlled. It shows an uptick during the 20th century; Knapp et al. , cited by Bunn et al., argued that higher CO2 levels increased water use efficiency in junipers – so this would not seem to be independent evidence of climatic change. An earlier work by Bunn et al. [Bunn et al, 2003] had argued that higher CO2 levels also improved water use efficiency in strip-bark forms. The new PIFL site (Lundy Lake) is strip-bark – so again this is not new unambiguous evidence (20th century levels are similar to MWP levels anyway at this site.) However, there remain 3 new foxtail sites (BP, UWL, CRA) which are not strip-bark (also Cirque Peak is a Graybill foxtail site, which is shown as not strip-bark.)
Figure 1. Figure 2 from Bunn et al. .
Graumlich  was an earlier discussion of foxtails by one of the co-authors, which has itself been widely cited in discussions of CO2 fertilization. The sites reported in Graumlich  were also foxtail sites from the same area (See below). One (Crabtree) appears to overlap directly; Upper Wright Lakes site (36o37;118o22,3510) is obviously very close to the previous Bighorn Plateau site; the previous Boreal Plateau (36o27; 118o20, 3420) is a little SW of Crabtree.
|WTF||West Tyndall F||PIBA||36 39||118 23||3450|
|WTL||West Tyndall L||PIBA||36 39||118 23||3450|
|BP||Bighorn Plateau||PIBA||36 36||118 22||3430|
|CR||Crabtree||PIBA||36 35||118 22||3350|
|KP||Kaiser Pass||JUOC||37 17||119 5||2700|
The tree-ring chronologies in the earlier study did not have enhanced 20th century growth, as shown in the former Figure 2. In fact, the absence of enhanced 20th century growth in the sites studied by Graumlich  was considered by her to be strong evidence against the hypothesis of CO2 fertilization.
These previous arguments are not re-visited and reconciled in Bunn et al. . Now for the connections to MBH98. Bunn et al. state:
The possibility that tree growth (e.g., annual rings) in natural environments has been fertilized by elevated atmospheric CO2 in the last several decades of the twentieth century has been discussed in several dendroecological studies (e.g., LaMarche et al., 1984; Graumlich, 1991; Briffa, 1992; Graybill and Idso, 1993; Nicolussi et al., 1995; Knapp et al., 2001). … There is no consensus, however, on the presence or impact of CO2 on growth; a review article on the subject states that there is inconclusive evidence to support CO2 enhancement on tree growth in natural environments (Jacoby and D’Arrigo, 1997). Given the potential for multiple interpretation of the growth signal in this data set, however, we feel compelled to address this issue further.
This is not a very imposing survey of recent literature on the impact of CO2 on high-altitude growth. They don’t mention the important studies from Kàƒner or Hattenschwiler or the many FACE (free air CO2-enhanced) studies. Jacoby and d’Arrigo can hardly be considered unbiased.
They go on to say:
The dramatic trend that is seen in the late twentieth-century growth rates of the trees in this study is similar to that seen in other palaeoclimate research, particularly the Mann-/Bradley-Hughes (MBH) multiproxy reconstruction of Northern Hemisphere annual temperatures (Mann et al., 1999).
Notice the embedding: the MBH98 reconstruction on a millennial scale is NOT robust to the presence/absence of bristlecone/foxtails, so it is hardly surprising that a network which is dominated by foxtails/bristlecones should look like MBH98. They are not independent networks. Bunn et al. then discuss MBH98 as follows:
By combining instrumental temperature data with multiple temperature-sensitive proxy records, the MBH reconstruction indicates that the twentieth-century warming is abrupt and truly exceptional in the context of the last millennium. Because of concerns about CO2 fertilizations affecting the results of that reconstruction, the reconstruction was rebuilt with a sparser data set without tree-ring data (instead relying on historical instrumental, coral and ice-core records), and the authors verified that there was no bias from non-climatic influences on their tree-ring data sets (Mann et al., 2000).
This non-robustness claim of Mann et al. seems especially objectionable to me. The calculations in their CENSORED directory show that they were well aware that their 15th century reconstruction was not robust to the presence/absence of bristlecones. Without the bristlecones, there is no MBH98-99 evidence that 20th century warming is "abrupt and truly exceptional". If one wanted evidence for just how misleading the Mann et al.  discussion of robustness is, the reliance by Bunn et al. right here is surely an example. Bunn et al. then say:
Comprised mostly of high-elevation tree-ring data, the MBH tree-ring data are similar to the data used in this study. As the MBH data show no evidence of CO2 fertilization in their high-elevation tree growth data, we have no reason to suspect that the growth trends seen in this work are related to atmospheric CO2 fertilization. Instead it is our contention that the signals seen in these data are the result of climatic influence.
The sleight-of-hand, relying on the misrepresentation by Mann et al., is completed. Mann et al. cannot argue anything about the 15th century or earlier without the bristlecones. So Bunn et al. cannot rely on MBH conclusions (using virtually the same sites) to show that their sites are climatic.
Bunn, Andrew G., Lisa J. Graumlich and Dean L. Urban, (2005). Trends in twentieth-century tree growth at high elevations in the Sierra Nevada and White Mountains, USA. Holocene 15, 481-488. Graumlich, L. 1991. Subalpine tree growth, climate and increasing CO2: an assessment of recent growth trends. Ecology 72, 1-11.