A new study by Alexander Stine and Peter Huybers offers strong evidence that reduced light availability (“global dimming”) explains the apparent lack of tree-ring evidence in many Arctic regions for the recent warming observed in instrumental records there. Dendroclimatology uses measurements of the yearly growth width and density of tree rings to reconstruct changes in past climate. The creation of long tree-ring chronologies from overlapping measurements of living and dead trees has allowed annual precipitation and temperature to be reconstructed far into the past, extending our knowledge of climate beyond the relatively short instrumental period. This approach works with most confidence when using trees that grow (or grew) near their environmental limits, and when a single limiting factor such as precipitation or temperature dominates.
Trees growing in cool Arctic climates are often highly sensitive to the temperature of past growing seasons, with higher temperatures promoting wider rings and denser post-spring “latewood” growth. Many summer temperature reconstructions have consequently been based upon tree-rings from these regions. Up to the 1960s, comparison to instrumental records showed that both the widths and densities of Arctic tree-rings largely followed variations in average temperature. But since then, while they have continued to mirror year-to-year variation, tree-ring records have often failed to reproduce the more gradual longer-term increase seen in Arctic temperatures. This is known as the “divergence problem”, and the possibility that Arctic trees may not have fully tracked average temperature at other times has been seized upon by those wishing to discredit tree-ring temperature reconstructions that show the recent twentieth century warmth to be effectively unprecedented.
Proponents of the “global dimming” hypothesis have suggested that reduced average tree-ring widths and densities are linked to reductions in the light reaching the Earth’s surface, which may stem in part from increased anthropogenic particulate emissions. But this has proven hard to test. Stine and Huybers now show that divergence is most notable in Arctic regions that are already light-limited by cloud cover, but effectively vanishes in less cloudy regions, even when these experience similar temperatures. They also show that sudden additional decreases in light availability occurring for several years after large volcanic eruptions have the most impact on already-light-limited trees. Thus, for some Arctic regions, light availability may matter as much as temperature to tree-growth. As well as being a compelling explanation for the divergence problem, the study has implications for the regional selection of trees in temperature reconstructions, and highlights the potential of trees to contribute to reconstructions of past surface light intensity.
Citation: Stine AR, Huybers P, 2014. Arctic tree rings as recorders of variations in light availability. Nature Communications, doi: 10.1038/ncomms4836