GSA Annual Meeting in Seattle, Washington, USA - 2017

Paper No. 220-7
Presentation Time: 3:00 PM

RELATIONSHIPS BETWEEN EXTINCTION SELECTIVITY AND REGIONAL CLIMATE CHANGE INTENSITY IN MIO-PLIOCENE PLANTS OF NORTH AMERICA


SIMPSON, Andrew G., Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, DC 20560, andy.g.simpson@gmail.com

Increasingly, scientists are looking at the reciprocal influence of ecological factors and macroevolutionary trends. Within plants, multiple studies now link modes of pollination and seed dispersal to trends in diversification and extinction. While patterns of diversification and extinction must, in turn, influence ecological parameters such as community composition through time, relatively little work has been done investigating the nature if this interplay.

We here compare the rates of survival and patterns range contraction of genera in Miocene and Pliocene floras in the eastern and western United States. Climatic changes endured by floras of eastern North America (ENA) have been different in character from those endured by those of western North American (WNA). While climate in both areas has cooled from the Miocene to the present, ENA retained a wet temperate environment while much of WNA transitioned to steppe or desert during the same interval.

Plant genera present in WNA floras (Clarkia, Fingerrock, Latah, Mascall, Purple Mountain, Stewart Valley, and others) are markedly less prone to extinction through this climatic change if they possess animal-mediated seed dispersal (p = 0.002), and are less strongly protected against extinction by animal-mediated pollination (p = 0.01). We were unable to detect any impacts of either (dispersal p = 0.66; pollination p = 0.22) on survivorship in ENA floras (Alum Bluff, Brandon, Brandywine, Citronelle). In the case of pollination, this non-significance is plausibly due to smaller sample size in ENA. The data hint that animal dispersal may convey protection in ENA as well, but if so then the effect is much weaker and is non-significant even when adjusted for sample size.

We suggest that the abundance, contiguity, and size of refugia in ENA relaxed selection for effective long-distance dispersal, while more severe climate change in WNA resulted in fewer refugia located farther apart, with animal dispersal thus being more valuable in enabling colonization and, ultimately, survivorship. Insect pollination may have enabled survival of small populations in refugia, obviating the need for recolonization.