Paper No. 4
Presentation Time: 8:45 AM

PERMIAN CLIMATE AND PALEOBOTANICAL SAMPLING METHODS


RHODES, Kristopher J.S., Geological Sciences, University of Michigan, Ann Arbor, 2508 Ruthven Musems, University of Michigan, Ann Arbor, MI 48109 and SYVERSON, Valerie J., Museum of Paleontology, University of Michigan, 1109 Geddes Ave, Ann Arbor, MI 48109, krisrhodes@gmail.com

Here we provide evidence that a sample of fossil plants from the Permian of West Texas are drawn from different populations, which supports a climatically driven change in local flora. These populations draw from mixed xeric and hydric floras, and have been interpreted by DiMichele et al. (2006) as representing a regional oscillation of climate during Permian time. Our data support this interpretation. We describe this pattern as "time mixing", in which a sample is drawn from multiple temporally discrete intervals, as distinct from time averaging, in which the sample is drawn from a continuous population of individuals living over a period of time.
Paleobotanical sampling using mixed quadrat sizes has been used since its description by Pfeffercorn et al. (1975). This method records presence/absence data for each taxon on single rock specimens of varying size. We tested several possible models for the relationship between specimen size and the presence of different taxa, using simulation and real world data. Herein we describe a general model that relates specimen size to discovery probability for each taxon. This allows different components of the plant community to be separated based on their overall discovery probability; in this case we interpret these components as temporally distinct floras in a time-mixed sample.