PP31C-1151:
Carbon and hydrogen isotope composition of plant biomarkers as proxies for precipitation changes across Heinrich Events in the subtropics
Wednesday, 17 December 2014
Thomas Elliott Arnold1, Katherine Freeman2, Mark Brenner1 and Aaron F Diefendorf3, (1)University of Florida, Ft Walton Beach, FL, United States, (2)Pennsylvania State University Main Campus, Geosciences, University Park, PA, United States, (3)University of Cincinnati, Cincinnati, OH, United States
Abstract:
Lake Tulane is a relatively deep (~23 m) solution lake in south-central Florida. Its depth and location on a structural high, the Lake Wales Ridge, enabled continuous lacustrine sediment accumulation over the past >60,000 years. Pollen in the lake sediments indicate repeated major shifts in the vegetation community, with six peaks in Pinus (pine) abundance that coincide with the most intense cold phases of Dansgaard–Oeschger cycles and the Heinrich events that terminate them. Alternating with Pinus peaks are zones with high relative percentages of Quercus (oak), Ambrosia (ragweed), Lyonia (staggerbush) and Ceratiola (rosemary) pollen, genera that today occupy the most xeric sites on the Florida landscape. This suggests the pollen record indicates the Pinus phases, and therefore Heinrich Events, were wetter than the intervening Quercus phases. To test the connection between Heinrich Events and precipitation in Florida, we analyzed the carbon (δ13C) and hydrogen (δD) isotope signatures of plant biomarkers extracted from the Lake Tulane sediment core as proxies of paleohydrology. The δ13C of plant biomarkers, such as n-alkanes and terpenoids, are determined, in part, by changes in water-use efficiency (WUE = Assimilation/Transpiration) in plant communities, which changes in response to shifts in mean annual precipitation. Plant δ13C values can, therefore, provide a rough indication of precipitation changes when other factors, such as plant community, are relatively stable throughout time. Paleohydrology is also recorded in the δD of plant leaf waxes, which are strongly controlled by precipitation δD. In this region, precipitation δD is negatively correlated with rainfall amount (i.e. the “amount” effect) and positively correlated with aridity. Thus, the δ13C and δD signatures of molecular plant biomarkers provide relative indicators of precipitation change, and when combined, provide a test of our hypothesis that vegetation changes in this region are driven by changes in aridity.