THE APPLICATION OF THE BRGDGT PALEOTHERMOMETER TO GLACIAL LAKE HITCHCOCK - WHAT CAN WE LEARN FROM SEASONALLY RESOLVED VARVED SEDIMENTS?

Within sediment archives, the presence of lipid biomarkers has proven widely useful in the analysis and reconstruction of pre-anthropogenic paleoclimate settings, particularly in terrestrial environments where other proxies are not readily available. BrGDGTs (branched glycerol dialkyl glycerol tetraethers) are a suite of bacterial membrane lipids that have been found in a wide variety of environmental archives across the globe. Distributions of brGDGTs have been shown to relate to temperature in modern settings, and a variety of calibrations have been developed to allow for the reconstruction of paleotemperatures back through time in lacustrine sedimentary records. However, numerous issues remain with the proxy, such as the influence of other environmental parameters (e.g. dissolved oxygen, pH, conductivity) as well as the seasonality of the recorded signal.

Here we present brGDGT data from seasonally varved sediments of Glacial Lake Hitchcock, which filled the Connecticut River Valley during the period of deglaciation ~15,000–12,000 years before present. We collected samples from Lake Hitchcock deposits near Mt. Sugarloaf in South Deerfield, MA. The varves at this location consist of alternating layers of light gray, coarser grained silt and clay from the ice-free season and darker gray, finer-grained silt and clay from the winter season when ice was present on the lake surface. The seasonally resolved layers are on the order of cm-scale at this site, allowing for detailed sampling of individual summer and winter sediments from Lake Hitchcock to test assumptions regarding the seasonality of brGDGT production in lacustrine systems. For example, available evidence from high latitude/cold environments suggests that brGDGTs primarily record a “months-above-freezing” or warm season temperature. Our preliminary results from Lake Hitchcock show that the majority of winter layers exhibit lower MBT’_5Me Index values and therefore generally colder reconstructed temperatures than summer layers, though this pattern is not universal in our data. In this study we analyze possible causes for this discrepancy, including differential seasonal production of brGDGTs and potentially variable sources of these biomarkers throughout the year (e.g. soil vs. lake-derived).