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

Paper No. 289-11
Presentation Time: 11:00 AM


KELLER, Jonathan Stanford, Department of Ecology, Evolution, and Behavior, University of Minnesota, 987 Upper Buford Cir, Falcon Heights, MN 55108, FOX, David L., Department of Earth Sciences, University of Minnesota, 310 Pillsbury Drive SE, Minneapolis, MN 55455-0219 and MCNULTY, Kieran P., Department of Anthropology, University of Minnesota, 395 Hubert H. Humphrey Center, 301 19th Avenue South, Minneapolis, MN 55455, kell1077@umn.edu

Diet is a primary mode by which species interact with their landscapes, and the trophic diversity of consumer communities reflects food resource availability, habitat structure, and climate. Community composition of modern rodents, which are abundant, widespread, morphologically disparate, and ecologically diverse, is related to both ecosystem precipitation and productivity, suggesting the ecological structure of fossil rodent assemblages may be useful for characterizing landscape change. Conventional proxies for diet such as hypsodonty and carbon isotopes indicate aspects of diet (e.g. consumption of C4 grasses) but are not specific enough to estimate trophic category across rodent diversity. Here, we show that dental ecometrics calculated from cheek teeth of extant North American rodents can reliably identify five distinct trophic categories: insectivore (I), frugivore (Fr), folivore (F), rootivore (R), and granivore (G), but with current sample sizes cannot yet distinguish omnivores (O) from granivores. At present, our dataset includes µCT scans of crania and mandibles from 136 species at 4-31µm resolution. Using both enamel and dentin surfaces of lower tooth rows, we replicated several published methods (e.g. Dirichlet normal energy; orientation patch count, OPC; relief index; and volumetric hypsodonty index) and developed numerous novel metrics. We have complete data for 75 species capturing much of the taxonomic, morphological, and trophic variation of modern North American rodents. A discriminant function analysis (DFA) using six metrics can delineate I, R, and Fr from the other three categories as a group. A second DFA using only species in the latter group can distinguish F, O, and G using six different metrics. Our analyses correctly classified 65 of 75 species (87%) by cross-validation. I, R, and Fr were best identified by a volumetric enamel/dentin ratio, OPC, and occlusal surface area ratios, while F, O, and G were demarcated by m3 concavity and occlusal perimeter ratios. We are expanding the dataset to encompass most modern North American rodents, and we are applying the DFAs to a series of Neogene and Quaternary fossil assemblages from the Great Plains to examine how trophic structure of rodent communities evolved in the context of regional and global environmental change.
  • JonathanKeller_GSA2014.pptx (23.8 MB)
  • FinalGSAScript.docx (29.8 kB)