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

Paper No. 257-14
Presentation Time: 4:30 PM

MASS EXTINCTIONS AS DRIVERS OF PREDICTABILITY IN THE HISTORY OF MARINE LIFE


BUSH, Andrew M., Geosciences & Ecology and Evolutionary Biology, University of Connecticut, 354 Mansfield Road - Unit 1045, Storrs, CT 06269 and PRUSS, Sara B., Department of Geosciences, Smith College, Northampton, MA 01063

Mass extinctions are generally considered to be sources of unpredictability and contingency in the history of life: they disrupt potential evolutionary trends, kill off successful lineages, and eliminate traits that favor success in “normal” times. The extinction of the non-avian dinosaurs is a classic example.

For the marine fauna, we suggest that some mass extinctions had the opposite effect on the history of life: they were, in no small measure, a source of predictability. High-level taxonomic turnover in marine animals was driven in large part by mass extinctions, and the dominant organisms today are those whose ancestors survived repeated catastrophes. To an extent not appreciated until recently, many of these mass extinctions had a similar pattern of selectivity: taxa with a high ratio of CaCO3 shell to biomass suffered preferentially. The end-Permian extinction was the most severe of these events, but similar selectivity has been observed in the early Cambrian, Late Devonian, Guadalupian, Toarcian, latest Triassic, Aptian, Paleocene, and perhaps others. Geological and geochemical evidence implicate environmental disturbances like volcanism, ocean acidification, anoxia, and climate change during these events. Some events were too small to substantially affect metazoan diversity trends; however, the frequency of these selective events suggests they are a predictable and recurrent component of life’s history.

Organisms that were less susceptible to these extinctions (e.g., arthropods, vertebrates, mollusks) dominate the modern ocean, whereas susceptible groups have suffered repeated catastrophic insult. Similar patterns of turnover are to be expected if one could “rewind the tape of life” to the early Paleozoic. The timing of these changes might be difficult to predict, because a massive, end-Permian-sized event—while likely given enough time—could have taken place earlier or later. Numerous paleoecologic trends may be partially predictable as well due to functional correlations. For example, heavily calcified animals are more likely to be sedentary and epifaunal than animals with lighter shells; paleoecologic shifts in motility and tiering may be, in part, a reflection of predictability in extinction.