COS 98-4 - Timing and drivers of woolly mammoth extinction on St. Paul Island in the late Quaternary

Friday, August 12, 2016: 9:00 AM
Palm A, Ft Lauderdale Convention Center
Yue Wang1, Pete Heintzman2, Soumaya Belmecheri3, Brendan J. Culleton4, Lee Newsom4,5, Warren P. Porter6, Beth Shapiro7,8, John W. (Jack) Williams1, Matthew Wooller9 and Russ Graham3, (1)Geography, University of Wisconsin-Madison, Madison, WI, (2)University of California, Santa Cruz, (3)Earth and Mineral Sciences Museum, Pennsylvania State University, University Park, PA, (4)Anthropology, Pennsylvania State University, University Park, PA, (5)Anthropology, Penn State University, (6)Integrative Biology, University of Wisconsin - Madison, Madison, WI, (7)Ecology and Evolutionary Biology, University of California, Santa Cruz, (8)Ecology and Evolutionary Biology, University of California, Santa Cruz, Santa Cruz, CA, (9)Institute of Marine Science, University of Alaska Fairbanks, Fairbanks, AK
Background/Question/Methods

Understanding the causes and effects of the late Quaternary megafaunal extinctions requires precise estimates of extinction timing and associated environmental factors. On St. Paul Island, an isolated remnant of the Bering Land Bridge, a late-surviving population of woolly mammoth (Mammuthus primigenius) persisted into the middle Holocene. Here we present an improved chronology of the timing of St. Paul woolly mammoth extinction based on spore abundances of three coprophilous fungi: Sporormiella, Sordaria, and Podospora from lake sediment cores retrieved from Lake Hill, in comparison with direct dating of newly recovered mammoth fossils and ancient DNA. We then tested hypothesized mechanisms for the extinction event, using vegetation reconstructions based on fossil pollen data and a physiological model (Niche Mapper) driven by simulated climates and vegetation since the last glacial maximum and trait data for woolly mammoths. This work helps resolve the timing and drivers of extinction for a small island population that was the second-latest surviving population of woolly mammoths worldwide.

Results/Conclusions

The five independent indicators indicate a timing of mammoth extinction at 5,650 ±100 cal BP. Podospora drops to zero at 7,020±170 cal BP, but Sporormiella and Sordaria drop to zero at 5,680±80 and 5,650±80 cal BP, corresponding our last record of mammoth aDNA at 5,650±80 cal BP and latest radiocarbon date of mammoth bone at 5,530±50 cal BP from St Paul Island. All coprophilous spores have low abundances (<620 grains/cm3) below 1,050 cm, hypothesized to indicate a small mammoth population around Lake Hill prior to island isolation ca. 13,500 BP by sea level rising. Grass-herb tundra is the major vegetation type throughout the record, with more Poaceae in the late glacial and more herbs in the Holocene. Pollen, macrobotanical and ancient DNA data suggest that St. Paul and coastal Beringia were not refugia for woody plant species except for dwarf willow (Salix herbacea), although spruce pollen in low concentrations is found during the Younger Dryas but interpreted as long-distance transport. Preliminary simulations by Niche Mapper point to the importance of net primary productivity and freshwater availability as limits on the carrying capacity of St. Paul for mammoth populations, with a low carrying capacity making this population highly vulnerable to extinction.