2003 Seattle Annual Meeting (November 2–5, 2003)

Paper No. 22
Presentation Time: 8:00 AM-12:00 PM

ASSESSING MAMMALIAN PALEOFAUNAL DIVERSITY: DISCREPANCIES BETWEEN PUBLISHED AND MUSEUM COLLECTION DATA FOR THE MIOCENE OF NORTHWESTERN NEVADA, USA


DAVIS, Edward and PYENSON, Nicholas, Integrative Biology, Univ of California, Berkeley, 3060 Valley Life Sciences Building, Berkeley, CA 94720, daviseb@socrates.berkeley.edu

Many paleobiological database projects, e.g. MIOMAP and the Paleobiology Database, allow researchers to investigate large-scale ecosystem dynamics through meta-analyses of published fossil records, addressing questions concerning the interplay between organismal and environmental change. Published records are the foundations of these projects; however, in many cases they may not adequately reflect the composition of fossil deposits, possibly confounding meta-analyses. Here we address the gap between the published record and the fossil record by examining the vertebrate fossil collections at the University of California Museum of Paleontology (UCMP), focusing on the Miocene mammal fossils from the Virgin Valley and Thousand Creek regions of northwestern Nevada. The fossil collections are compared to published data to assess similarity of diversity, relative abundance, and evenness, using rarefaction analyses, cenograms, and taxonomic similarity comparisons. Initial results indicate a discrepancy between the published data and actual museum collections. For example, published data for a sample Thousand Creek locality indicate the presence of eight genera, while a survey of the museum collections indicates the presence of 13 genera. Similarly, a sample Virgin Valley locality has 11 published genera and 15 genera represented in collections. These differences in generic richness are reversed when comparable numbers of specimens are compared through rarefaction analysis, probably because the original researchers were interested only in the best-preserved specimens for taxonomic purposes, leading to a much steeper rarefaction curve. These results show that data published by early workers are not comparable to the more intensive studies published today, which lends credence to an additional layer of bias in paleontological records. Database projects would benefit from research programs with an emphasis on investigation and publication of all collected fossil material.