Biodiversity can be measured on multiple scales, ranging from the sample-level (e.g. a bulk sample), to the local and regional (e.g. the Cincinnati Arch), provincial (e.g. Baltoscandia) and global-level. At each of these spatial scales, components of biodiversity, such as richness, evenness and the number of fossil individuals influence one another and affect the composite biodiversity signal. As a precursor to the present study, I investigated the relationship among these properties at the sample-level with respect to lithology and position along an environmental gradient. Within assemblages, richness was influenced by sample size and evenness, while evenness and the number of individuals in a sample were significantly correlated only when considering samples from disparate communities. While the interactions of these variables are understood at the sample-level, it is not clear how the components of biodiversity influence one another when aggregating sample-by-sample and locality-by-locality. The purpose of this study is to determine how variables such as richness, evenness and sample size as well as paleoenvironmental variation and lithology can influence assessments of local and regional biodiversity.

To accomplish this goal, a total of 140 bulk samples yielding over 5,000 individuals were collected from the highly fossiliferous Fulton submember of the Kope Formation (type Cincinnatian, Upper Ordovician) of Ohio and Kentucky. These data were aggregated sample-by-sample to generate a series of local and regional-level biodiversity assessments. At each spatial scale, I generated rarefaction curves, dominance diversity assessments, collector's curves and assessed evenness. Additionally within the Fulton samples, the rare taxa (defined here as taxa with less than 1% of total taxonomic abundance) comprise only 3.2% of total fossil abundance, yet yield over 60% of the richness. This “rare” component of fossil communities, often overlooked, may also influence the way that biodiversity assembles. Since paleontologists seek to build biodiversity assessments and trends from composite data sets, this investigation into the quantitative relationships among sample-level and locality-level properties is an important step towards understanding how biodiversity assembles on multiple scales.