Linking paleo- and modern species-area assessments as a tool for understanding biodiversity response to global change in the western USA

Background/Question/Methods

By defining paleospecies-area relationships (PSAR), previous work has shown that the preanthropogenic (>11,000 years ago) diversity baseline for mammals was substantially higher than for the period when pre-industrial humans were common in the western U.S. (~11,000 to 500 years ago).  It is unclear, however, how the paleospecies-area curves compare with modern ones derived from sampling mammals on the present-day landscape, because the sampling of the modern, versus fossil taxa, greatly differs.

We show that straightforward comparisons between the two cannot be made because of the differing sampling biases involved with acquiring modern and fossil data. However, with sample standardizing techniques, we attempt to develop a methodology that will allow for parallel comparison of the paleo- and modern SARs.  Our technique emphasizes subsampling the modern record to make it look like the fossil record.  This is accomplished by obtaining modern faunal samples that were collected by similar vectors as those that resulted in fossil samples, primarily bone-laden raptor pellets and carnivore scats in rock shelters and caves. 

Results/Conclusions

Modern faunal list per locality were built based on the sampling procedures previously noted and used to construct species-area curves. The modern curves constructed in such a manner were more similar to fossil Holocene curves than to late Pleistocene ones in reflecting depressed diversity relative to the pre-anthropogenic condition. Analyses of the modern curves compared to paleospecies-area curves constructed from Holocene sites in the same area revealed biologically meaningful differences in local and regional biodiversity in the past few decades with respect to most of the Holocene. The modern species-area curves based on pellet/scat samples were also compared to modern species-area curves constructed from specimen records existing in the Smithsonian Natural History Museum collections of North American mammals. The curves constructed from museum records and range maps overestimated both alpha (higher species richness) and beta (steeper slop) diversity with respect to the modern pellet/scat curves. By analyzing the taxonomic list from which the various curves were built we have identified taxonomic biases that provide insights into appropriate ways to subsample more complete modern data to make it resemble and therefore be more directly comparable to fossil data. This in turn will allow us to identify whether the present-day biodiversity baseline (i.e., that of the past two centuries) still resembles that when humans were present in pre-industrial times (11,000-500 years ago), or if it has declined even further.