Paper No. 6
Presentation Time: 2:35 PM

ANALYSIS OF EARLY PALEOZOIC ZOOPLANKTON (GRAPTOLOIDA) EXTINCTION RISK USING MODEL-CHOICE


BOYLE, James T.1, SHEETS, H. David2, WU, Shuang-Ye3, GOLDMAN, Daniel3, MELCHIN, Michael4, COOPER, Roger A.5, SADLER, Peter6 and MITCHELL, Charles E.7, (1)Evolution, Ecology and Behavior, University at Buffalo, SUNY, 126 Cooke Hall, North Campus, Buffalo, NY 14260, (2)Dept. of Geology, SUNY at Buffalo, 411 Cooke Hall, Buffalo, NY 14260, (3)Department of Geology, University of Dayton, 300 College Park, Dayton, OH 45469, (4)Department of Earth Sciences, St. Francis Xavier University, Antigonish, NS B2G 2V5, Canada, (5)GNS Science, P.O. Box 30-368, Lower Hutt, 5010, New Zealand, (6)Department of Earth Sciences, University of California, Riverside, CA 92521, (7)Geology, University at Buffalo, SUNY, Buffalo, NY 14260, jamesboy@buffalo.edu

The factors driving extinction risk in planktonic organisms are poorly understood in comparison to benthic groups but the exceptionally well-sampled graptolites, early Paleozoic zooplankton, provide an opportunity to fill this gap. Graptolites are also a particularly interesting test-clade because two distinct sets of species, biofacies, are identified by their facies distribution and have significantly different extinction risks. Moreover, the biofacies are commonly interpreted to represent vertical habitat zones: the oceanic biofacies taxa, which may have preferentially inhabited the mesopelagic oxygen-minimum zone, have relatively short species durations on average. In contrast the less-facies limited taxon set, which appear to have inhabited the shallow epipelagic biotope and had ranges that extend farther on to continental shelves, have a longer mean species duration. We employed a global dataset of Ordovician graptolites and an analysis based on general linear models and AIC model ranking to examine the contributions of biofacies, sampling, endemicity, region and eight different measures of geographic range as predictors of extinction risk. Partial least squares regression analysis was used to account for covariation between measures and reduce model complexity. The results showed that overall commonness, region, biofacies, geographic range, and sampling, in order of decreasing strength, were the main determinants of extinction risk in the taxa analyzed. The strong contributions of biofacies and region, i.e., the part of the globe in which species occur, suggest that oceanographic properties such as the occurrence of upwelling zones and water mass shifts were likely contributors to graptolite extinction. Geographic range and sampling were relatively weak contributors to extinction risk compared to the contributions for these factors previously documented among benthic organisms. Latitudinal extent, often taken as a proxy for climate tolerance, was one of the lowest ranked models, counter to what might be expected by the latitudinal zonation of modern plankton.