Research Training Program

Smithsonian Institution
National Museum of Natural History

PROJECT SUMMARY
1997


Surangi Punyasena
Yale University
New Haven, Connecticut
Anna K. Behrensmeyer, Ph.D.
Supervising Scientist
Department of Paleobiology

"The RTP offered me the chance to work with, and learn from, scientists that I had formerly only read, to discover the people behind the papers."

Measuring Historical Diversity: Diversity Indices and the Plio-Pleistocene Mammals of Northern Lake Turkana

ABSTRACT

Characterizing the diversity of an ecosystem, difficult and disputed in modern communities, becomes further complicated when working with paleontological data. The different biases introduced into the fossil record by the process of preservation and collection distort visions of the past. Complete information on species richness and abundance is never available, although the same could be said of modern ecosystems. Despite these concerns, however, the record provides a sole glimpse into the ecology of past environments. Sometimes, the biology can be separated from the taphonomy. East Africa's Turkana Basin, in particular, has extensively sampled and well-dated deposits for the Plio-Pleistocene. Though mostly noted for their hominid specimens, over 246 mammalian species have also been identified from the basin's Koobi Fora, Nachokni, and Shungura formations. With over 425 localities and 10,092 published specimens, this region has one of the richest and best sampled mammalian faunas, making it an excellent candidate for attempts at paleoecological analysis.

The study had two overall aims. The first was to evaluate nine diversity indices, including Margalef, Shannon and Alpha diversity, to determine if there were patterns of consistent differentiation between the diversity found at the locality, member and regional level, as well as discover the robustness of the data available. Preliminary analysis indicates that the larger members, i.e. those containing a relatively large number of taxa and specimens, do contain enough information to generate stable results, indicating that the index calculated would not be greatly changed by additional specimen information. To anchor these numbers, comparisons to modern ecosystems were also made. The second focus was on the patterns of diversity generated by these indices, between localities, members, and regions, geographically and temporally, to generate a macroscopic view of the changing ecology of the region.

This research was made possible by a grant from the Office of the Director.

Letter of Gratitude