A MULTI-PROXY RECONSTRUCTION OF YOUNGER DRYAS ENVIRONMENTAL CONDITIONS IN NORTHERN MAINE AND THE VALIDITY OF CHIRONOMID-INFERRED TEMPERATURES

Previously published chironomid studies from New Brunswick and Maine suggest that the difference between modern and Younger Dryas (YD) summer temperatures were as great as 8-11 ºC. Our primary objective was to validate the YD temperature depression inferred from chironomid remains using a variety of proxies including plant macrofossils and pollen. Pollen transfer function and analog models for mean July temperature and mean annual precipitation were generated from a training set of 1100 lacustrine surface samples and applied to pollen counts from two late-glacial-age sediment sequences from Aroostook County. Mean July temperature models have prediction coefficients (r2boot) ranging from 0.99 to 0.94, and root mean square error of prediction (RMSEPboot) ranges from 1.1 to 1.4 ºC.

Both pollen and chironomid transfer functions indicate northern Maine YD temperatures were around 6 °C lower than today. This is less than the 8-11°C differences between modern mean July air temperature and chironomid-inferred YD mean summer surface water temperatures for sites in New Brunswick and southeastern Maine, but is consistent with the steep north-south YD temperature gradient proposed for this region. Macrofossil content of lake sediment remained fairly stable through the late-glacial period until the beginning of the Holocene. Conifer needles were absent until after 10,000 14C yrs BP. Pollen evidence indicates a YD increase in grass and other herbs suggesting a change from shrub to herb dominated tundra during the Younger Dryas chronozone.

Pollen-reconstructed mean annual precipitation estimates and an earlier multi-proxy lake-level study indicate that the region was fairly dry during the YD, but not drier than during the pre-YD warm period. Increased precipitation resulted in no more than a 1-1.5 m rise in lake level by 10,000 14C yrs BP. To produce unrealistically cold chironomid-based temperature estimates, YD lake levels would have needed to increase substantially over pre-YD levels. The absence of such changes for this time period and the consistency of temperature reconstructions from our two proxies lead us to conclude that Younger Dryas chironomid-temperature reconstructions from this region are not strongly influenced by lake level changes.