December 8, 2006

A Stickier Handle on the “Hockey Stick”

Many of the documents generated by the Intergovernmental Panel on Climate Change (IPCC) make repeated reference to the famous “Hockey Stick” - the depiction of reconstructed Northern Hemisphere temperature for the past 1,000 years (Figure 1). We at the World Climate Report have maintained an anxious curiosity about the temperature series. If you will remember, at issue are the absences within the “Hockey Stick” of the so-called “Medieval Warm Period” of 1,000 years ago and the “Little Ice Age” that began 450 years ago and ended around 1900. Lacking evidence of the “Medieval Warm Period,” the “Hockey Stick” can be wielded to characterize the warmth of the last of the 20th century as beyond anything of the past millennium. Therefore, that pronounced warmth is a key piece of evidence in the climate change debate. Unfortunately for those swinging the “Hockey Stick,” the handle portion of the stick has become increasingly sticky from a stream of articles that show evidence of the warm period in the climate records of locations sprinkled across Earth. Yet another such piece appeared earlier this year in the Journal of Geophysical Research.

Figure 1. Reconstruction of Northern Hemisphere air temperature over the past 1,000 years from the IPCC.

The research outlined in the article “Svalbard summer melting, continentality, and sea ice extent from the Lomonosovfonna ice core” was led by Aslak Grinsted of the Arctic Centre at the University of Lapland in Finland. Grinsted and four colleagues developed a proxy for continentality, or distance to open water, from oxygen isotopes within an ice core taken from Lomonosovfonna, the highest ice field in Svalbard (78o51’53”N, 17o25’30”E). The isotope that they used has been shown to be related to the surface temperature in modern snow, and especially so on Lomonosovfonna. Continental interiors are associated with greater seasonal temperature variations than are coastal regions or islands, which are subject to the modifying effect of a large water body with a larger heat capacity, and therefore smaller variation in temperature. The principal idea that was followed in the work of Grinsted et al. is that an oxygen isotope reflective of temperature, if within an ice core taken from an island near the seasonal sea ice margin, would highly reflect variability in the distance to open water, or the degree of continentality of the island location. The seasonality in the amplitude of the oxygen isotope within the ice core could therefore be used to create a continentality index that could represent distance to open water through time, or relative size of an ice field that advances/retreats toward/away from the island.

In addition to the oxygen isotope amplitude reconstruction, or “continentality index,” Grinsted et al. developed a “chemical washout index.” The washout index stems from the concept that particular ions are “preferentially washed out of the ice during seasonal melt” and that it can be assumed that “ions are removed fractionally as the melt progresses.” Two washout indices were formed using two different chemical species.

The researchers tested the new indices by comparison with the instrumental records for nearby sites. The oxygen isotopes significantly correlated with mean, summer season, and winter season temperatures, annual range in temperature, and nearby (Barents Sea) sea ice extent. The continentality index most significantly correlated with annual temperature range, while the washout indices significantly correlated with sea ice extent.

Reconstructing the climate of Lomonosovfonna, the data confirm that the “Little Ice Age” ended at around year 1900 (Figure 2), which is consistent with previous literature. The beginning of the cold period is less distinct, but the researchers report that a clear signal of the period is evident by 1600 “probably indicating that the sea ice margin had advanced close enough to Svalbard to affect local climate.” The researchers also report that “continentality and washout indices show more gradual changes,” but that “the coldest and most continental period seen in the core is the nineteenth century.” (Figure 2).

As for the “Medieval Warm Period…Grinsted and his colleagues report: “In the oldest part of the core (1130–1200), the washout indices are more than 4 times as high as those seen during the last century, indicating a high degree of runoff. Since 1997 we have performed regular snow pit studies, and the very warm 2001 summer resulted in similar loss of ions and washout ratios as the earliest part of the core. This suggests that the Medieval Warm Period in Svalbard summer conditions were as warm (or warmer) as present-day, consistent with the Northern Hemisphere temperature reconstruction of Moberg et al. [2005].”

Figure 2. Fifteen-year moving averages of Lomonosovfonna ice core data. (a) Oxygen isotopes, (b) continentality proxy [A], (c) stratigraphic melt indices [SMI], and (d) washout indices. (Taken from Grinsted et al. 2006).

The work of Grinsted and his colleagues represents the latest headache for defenders of IPCC’s heralded “Hockey Stick”. The researchers’ representation of runoff from melt in the washout indices for Lomonosovfonna (Figure 2, bottom graph) is startling. We can see a hockey stick alright, but it seems to be swinging at puck located in 1000 AD!


Grinsted, A., Moore, J.C., Pohjola, V., Martma, T. and Isaksson, E. 2006. Svalbard summer melting, continentality, and sea ice extent from the Lomonosovfonna ice core. Journal of Geophysical Research, 111, 10.1029/2005JD006494.

Moberg, A., Sonechkin, D.M., Holmgren, K., Datsenko, N.M., and Karlénm, W. 2005. Highly variable Northern Hemisphere temperatures reconstructed from low- and high-resolution proxy data, Nature, 433, 613– 617.

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress