PLEISTOCENE PERMAFROST THAWING HISTORY OF THE NORTH AMERICAN ARCTIC AND CORDILLERA FROM U-TH DATING OF CAVE SPELEOTHEMS

Permafrost, or permanently frozen ground, is widespread in the Arctic, sub-Arctic and northern cordillera of North America. It is estimated to contain twice as much carbon as the atmosphere in the form of frozen organic matter, which can be released to the atmosphere as greenhouse gases under a warming climate. However, because permafrost can be slow to respond to warming, the short instrumental record does not adequately capture long-term temperature trends. The modest temperature changes of the past few millennia provide poor analogues for understanding the substantial warming projected for the next century or beyond unless the response of permafrost can be accurately predicted.

One way to address this problem is to assess the stability of permafrost during previous interglacial periods, some warmer than today. Speleothems in caves in the Arctic and other regions that are currently permafrost are relicts of past periods of thaw that enabled meteoric waters to seep into caves and deposit calcite (e.g., Lauriol et al., 1997). We employed uranium-thorium (U-Th) dating to constrain the chronology and extent of permafrost thaw in these North American regions during the past 600,000 years. We sampled caves from a range of permafrost zones (continuous, discontinuous, and isolated permafrost) and latitudes (66.5°N to 49°N), in the Yukon, Alaska, Northwest Territories, and Rocky Mountains of Alberta/British Columbia. Thirty-nine samples from 30 speleothems have been analyzed at this point in time; 23 samples lie beyond the U-Th dating range, and finite ages tend to cluster near the Marine Isotope Stage (MIS) 11 interglacial 400,000 years ago, as well as perhaps MIS 9, 13, and 15. This preliminary data set, coupled with a similar study in Siberia (Vaks et al., 2013), thus suggests an episode of widespread thawing during MIS 11 when global temperature was perhaps only ~1°C warmer than known pre-industrial temperatures. We anticipate dating more speleothems in the months prior to NEGSA, and exploring the use of U-Pb for samples extending beyond the analytical limit of U-Th dating.