2004 Denver Annual Meeting (November 7–10, 2004)

Paper No. 6
Presentation Time: 9:35 AM

INTEGRATING DEPOSITIONAL ENVIRONMENTS, PALEOGEOGRAPHY, AND FAUNAL ASSEMBLAGES: A NEW VIEW OF HADAR PALEOENVIRONMENTS ACROSS SPACE AND TIME


CAMPISANO, Christopher J., Anthropology, Rutgers Univ, 131 George St, New Brunswick, NJ 08901-1414, cjcampi@rci.rutgers.edu

For almost three decades, the site of Hadar, Ethiopia has remained one of the most important paleoanthropological sites in the world. In addition to its extraordinary fossil hominin record, the Hadar Formation also contains an enormous amount of geological and paleontological information pertaining to mid-Pliocene East African paleoenvironments and mammalian evolution. Traditionally, faunal assemblages are used to reconstruct the “average” paleohabitat that existed. While such approaches can elucidate general paleoenvironments and their change over time, they often homogenize the diversity of habitats that existed across a landscape at any one particular time. This study takes an alternative, high-resolution approach to compare catalogued faunal assemblages from Hadar using tightly constrained depositional units and geography as controls.

Using this methodology, distinct faunal and paleoenvironmental communities can be distinguished within sub-members that are usually treated as one unit. For example, two distinct paleoenvironments are recognized within the KH-2 sub-member (~ 3 Ma). The eastern region is comprised primarily of a forest dwelling faunal assemblage with taphonomic and geologic data suggesting a low-energy proximal floodplain. This is interpreted to be a gallery forest dominated environment between meanders in the ancestral-Awash River. A very different pattern exists in the western region where the faunal assemblage shows a mosaic habitat of forest and more open environments while the geology indicates proximal to distal floodplains leading outward from the edge of a fluvial cut-bank. Together this indicates a limited gallery forest giving way to more open environs away from the river. In this particular scenario, A. afarensis appears more common in the mosaic environment, or is at least not tethered to the forested ones.

Additional analyses from other sub-members also display degrees of spatial variation, primarily amongst habitat-specific bovid tribes used to distinguish forest, wet-grassland, and arid habitats. Temporal variation between sub-members for the most part agrees with earlier analyses. This combined geo-paleontological approach may provide an additional level of resolution in identifying early hominin habitat preference across paleolandscapes.