ABSTRACT
The Lower Pleistocene fossil site of Swartkrans has been excavated periodically since 1948 by Robert Broom and John Robinson and, most recently, by C.K. Brain. Brain’s last major effort ended more than fourteen years ago. This cave site has yielded the single largest sample of the early hominid species Paranthropus robustus, and several specimens referred to as the human ancestor, Homo erectus. The final seven year period of excavation, initiated in 1979, was conducted in a meticulous fashion, and has produced an abundance of carefully recorded geological, fossil, and artifact data. More than 20,000 fossils, along with a number of bone and stone tools, were recovered from Members (strata) 1, 2, and 3 alone. The excavation was carried out largely in 10 cm elevation increments with a metal grid erected over the site to allow more accurate provenancing.
Fossil remains and tools found at
the site of Swartkrans have been attributed to a variety of taphonomic stimuli
such as hominid activity, carnivore activity, alluvial deposition and
gravitation. In fact, accumulations
at this and other Plio-Pleistocene cave sites in South Africa have probably
resulted from a combination of these factors.
This webpage presents information on a method developed to allow
researchers to archive and visualize fossil, artifact, and geological data in
their spatial contexts, and to further explore taphonomic factors responsible
for such accumulations.
A 3D Geographic Information
System (GIS) was built with data from a variety of sources: 1) a 1999 survey
using a laser theodolite to record spatial coordinates for remaining geological
features; 2) the digitization of CK Brain’s field diagrams from the final
seven years of excavation to reconstruct the original geology of the site; and
3) a relational database including information on more than 20,000 vertebrate
fossils (macrofauna) and artifacts from Members 1, 2, and 3.
Mapping and 3D reconstruction of sites such as Swartkrans present challenges when using traditional GIS approaches because they cannot interpolate data from overhang features (there must be only one z-value for each x-y coordinate). This study uses a new 3D approach combining Intergraph’s Voxel Analyst and Esri’s Arcview 3D Analyst to overcome this limitation. Results present an accurate three-dimensional model of the site and its contents for data storage and analysis.