Covariation in macrostratigraphic and macroevolutionary patterns in the marine record of North America

Students of Earth history have long recognized the correlation between the quantity of preserved sedimentary rock and the diversity of life recorded as fossils. But paleontologists have yet to determine whether this pattern reflects a causal relationship or a unidirectional sampling bias in fossil data imposed by preserved rock quantity. Distinguishing between these two alternatives has been complicated by the fact that many of the basic patterns of paleontologic and lithologic covariation have yet to be quantified rigorously. Here we present the first analyses of the covariation between the macrostratigraphic and macroevolutionary histories of North America based on geographically and temporally explicit co-occurrences of rocks and fossils. The analyses use independent quantitative summaries of the stratigraphic and fossil records by integrating the Paleobiology Database (PaleoDB) and Macrostrat, a macrostratigraphy database for North America, which allows a more direct comparison of the stratigraphic and biological histories of the continent than has heretofore been possible. Within the Macrostrat database, the rock record is divided into discrete packages of sediment that are bound by hiatuses resolvable at the stage-level. Using per interval, per package rates of sediment package initiation and truncation, and genus first and last appearances (herein regional origination and extinction), we find a substantially stronger positive correlation between sediments and biology for extinction-like parameters than we do for origination-like parameters. Four of the largest coincident pulses of regional extinction and sediment truncation occur during the widely recognized end-Ordovician, late Permian, end-Triassic, and end-Cretaceous mass extinction intervals. A further comparison of the global ranges of North American genera to North American macrostratigraphy indicates that the regional and global extinction of genera are more likely to occur in the same stage than are global and regional originations. Together, these results suggest that our general understanding of biodiversity dynamics from the fossil record may not be strongly biased by the preservation of sediments and leaves open the possibility that certain large perturbations to the Earth system are responsible for major changes of state in both the sedimentary and biological systems.