2014 GSA Annual Meeting in Vancouver, British Columbia (19–22 October 2014)

Paper No. 76-1
Presentation Time: 1:00 PM

CLIMATE, CO2 AND THE HISTORY OF C3 AND C4 GRASSES FROM THE LAST GLACIAL MAXIMUM TO PRESENT ACROSS NORTH AMERICA


COTTON, Jennifer M., Department of Geology and Geophysics, University of Utah, 115 South 1460 East, Salt Lake City, UT 84112, CERLING, Thure E., Department of Geology and Geophysics, University of Utah, Salt Lake City, UT 84112, HOPPE, Kathryn A., Geology Dept, Green River College, 12401 SE 320th St, Auburn, WA 98092, MOSIER, Thomas M., School of Mechanical, Industrial and Manufacturing Engineering, Oregon State University, 204 Rogers Hall, Corvallis, OR 97330 and STILL, Christopher J., Department of Forest Ecosystems and Society, Oregon State University, 321 Richardson Hall, Corvallis, OR 97331

C4 grasses make up approximately 20% of Earth’s primary productivity, yet their potential responses to increasing temperature and precipitation patterns, as well as to increasing atmospheric CO2 concentrations, are not well constrained relative to vegetation using the more common C3 pathway. Bison (Bison bison) and Mammoth (Mammuthus columbi, Mammuthus primigenius) tissue records the isotopic composition of their diet (δ13C), and can be used as a proxy for the average abundance of C3 and C4 vegetation in the areas in which they lived. Here, we use the δ13C of bison hair, and bison and mammoth tooth enamel and collagen coupled with advanced statistical methods to model changes in the abundance of C3 and C4 grasses through the last glacial/interglacial transition in North America. Using the Random Forest multiple regression tree algorithm and eight climatic and soil predictor variables, the model explains 91% of the spatial variability in bison tissue δ13C values and determines that summer average temperature is the most important predictor of all climate variables. We apply this model to high-resolution gridded climate data and CMIP5 ensemble paleoclimate model outputs to produce a predictive isotope landscape model (‘isoscape’) for the current, Last Glacial Maximum (LGM) and Mid-Holocene Climatic Optimum (MHCO) average δ13C of vegetation across western North America. From the LGM to the modern, we find a large increase in the abundance of C4 grasses in the central Great Plains, along with a decrease in the abundance of C4 grasses along the Gulf coast. This model also allows us to directly quantify the effects of atmospheric CO2, and predict a threshold of ~270 ppm, below which CO2 concentrations begin to exert an influence on the distribution of C3 and C4 grasses. However, at low atmospheric CO2 values, the importance of summer average temperature on grass abundance is three to four times greater than it is for atmospheric CO2.