Comparison of soil derived tetraether membrane lipid distributions and plant-wax δD compositions for reconstruction of Canadian Arctic temperatures

Abstract

Polar amplification of climate warming has received much attention as these rapidly rising temperatures have the potential to alter ecosystem function and biogeochemical cycles. In particular carbon preserved in Arctic tundra soil and permafrost may be especially vulnerable resulting in carbon cycle perturbations providing an additional positive feedback to climate change. Reliable methods for reconstructing past temperature changes in polar regions have been established from ice cores and marine sediments; however techniques for the continental terrestrial environments are lacking, but are imperative to examine polar amplification of climate warming. Here we compare two molecular methods for reconstructing continental annual mean air temperature (MAT) for the Canadian Arctic based on the distribution of soil bacterial-derived glycerol dialkyl glycerol tetraether (GDGT) membrane lipids (MBT-CBT proxy) and the hydrogen isotopic composition (δD) of plant wax-derived n-alkanes. These two proxies were applied to both modern soil and paleosols collected from the Yukon Territory, Canada, to evaluate both the accuracy of the reconstructed absolute temperatures as well as the relative change in temperature between the Last Glacial and the Holocene. Branched GDGT-based estimates using the recently revised MBT′-CBT calibration are overall higher by ca. 6 °C compared to the original calibration. MAT estimates for modern soils based on the original MBT-CBT calibration are comparable with those based on the δD of extracted C29n-alkanes and instrumental data, however produced a 6 °C higher temperature signal for the glacial paleosols. Therefore, branched GDGT based temperature reconstructions for glacial soils in the high Arctic may represent the higher temperatures at the time of soil formation when bacterial activity was optimal whereas δD of C29n-alkane plant lipids appear to integrate an average annual signal. When used in tandem, these geochemical proxies may provide a more comprehensive method for reconstructing Arctic paleoclimate.