Wilmking et al. carried out a comprehensive sampling of 1558 white spruce trees at 13 treeline sites in the Brooks Range and Alaska Range. The date of the sampling is not stated but seems to be between 2000 and 2003 and includes the warm 1990s. These findings cover the same species as were primarily used in Jacoby and d’Arrigo’s northern treeline temperature reconstruction, used in multiproxy reconstructions Jones et al , Jones and Mann  as well as MBH98. They found that 40% of the trees had a negative growth response to increased temperatures, 36% had a positive response and the balance had no significant response. They state:
Without accounting for these opposite responses and temperature thresholds, climate reconstructions based on ring width will miscalibrate past climate.
Jacoby and d’Arrigo had hypothesized a linear relationship between ring width and temperature, an assumption embedded in MBH98 as a condition for a valid proxy. Instead, Wilmking et al. report:
Our findings of both positive and negative growth responses to climate warming at treeline challenge the widespread assumption that arctic treeline trees grow better with warming climate. High mean temperatures in July decreased the growth of 40% of white spruce at treeline areas in Alaska, whereas warm springs enhance growth of additional 36% of trees and 24% show no significant correlation with climate.
They go on to state:
The finding of highly significant negative relationships between July temperature and radial growth as the most common climate signal present at treeline and near treeline sites in Alaska is quite surprising, and apparently not consistent with much published literature… Here we report that the most frequent growth response to increasing temperatures is negative at the northern and altitudinal tree limit in Alaska. Not only single trees, but large populations of trees at every site show reduced growth with warming temperatures in North America (Garfinkel & Brubaker, 1980; Jacoby & D’Arrigo, 1989; Jacoby et al., 1996; Overpeck et al., 1997; Lloyd & Fastie, 2003).
As I mentioned in my earlier posting on D’Arrigo et al. re the TTHH site, the breakdown or non-existence of a linear relationship between temperature and ring width poses serious problems for climate reconstructions, as it is impossible to tell whether a narrow ring represents a cold or warm leg of the upside-down U relationship.
Update Mar 2007:
ITRDB identification for all Brooks Range sites but one are as follows:
BRFR 11001 ak047 Firth River PCGL 68.06 -141.06 100 1676 2002 Martin Wilmking
BRSJ 111007 ak053 Sheenjek River PCGL 68.05 -143.08 100 1715 2000 Janet Joergenson Martin Wilmking
BRNC 11004 ak050 Nutirwik Creek PCGL 67.09 -149.08 100 1731 2000 Martin Wilmking
BRNF 11005 ak051 Nutirwik Creek Floodplain PCGL 67.09 -149.08 100 1828 2001 Martin Wilmking
BRCL 11002 ak048 Chimney Lake PCGL 67.07 -150.05 100 1725 2001 Martin Wilmking
11003 ak049 North Fork Koyukuk PCGL 67.09 -150.05 100 1715 2001 Martin Wilmking
BRHF 1006 ak052 Hunt Fork of John River PCGL 67.08 -152.04 100 1730 2002 Martin Wilmking
Identifications for 2 Alaska Range sites are:
ak054 – Cabin Creek
ak049 – Slime Creek
Reference: Wilmking, Martin , Glenn P. Juday, Valerie A. Barber and Harold S. J. Zald, 2004. Recent Climate Warming Forces Contrasting Growth Responses Of White Spruce At Treeline In Alaska Through Temperature Thresholds, Global Change Biology 10, 1724–1736, doi: 10.1111/j.1365-2486.2004.00826.