Diversity in the Middle Pliocene: A multi-proxy analysis of climate's effect on vegetation and mammalian community structure with implications for human evolution Open Access
Downloadable ContentDownload PDF
Three-and-a-half million years ago, as many as four different hominin species are found across Africa. One, Australopithecus afarensis, is associated with 3.4 million year old cut-marked bones and is anatomically distinct from a contemporaneous northern Kenyan hominin, Kenyanthropus platyops, that is associated with the oldest stone tools at 3.3 million year old. Isotopes in their teeth show that both species incorporated food resources from grasslands and, further, the stone tools from Kenya and the cut-marked bones in Ethiopia indicate that the mid-Pliocene hominins likely ate the animals that lived alongside them in the expanding savannahs. The discovery of contemporaneous tool use and similarity in isotopic niche space prove difficult to interpret given that previous research hinted at ecological differences between basins. Why would two separate hominin genera exhibit simultaneous changes in behavioral ecology if they inhabited different environments? To provide insight into this quandary, this dissertation examines the link between vegetation, climate, mammalian community structure, and hominin distributions in the African mid-Pliocene (approximately 3.5 million years ago). Novel methods of quantifying paleo-habitat variation, using modern environments as a baseline, are used to understand variation in both vegetation and mammalian communities dated between 3.8—3.0 Ma and to illustrate ecological distinctions between contemporaneous hominin environments. Comparisons reveal persistent rarity of hominins in the Shungura region over a 400-thousand-year time and relatively higher abundance in sites such as Hadar and West Turkana, where environments had relatively even proportions of mixed-woodland as well as grassland environments. There is a correlation in extant communities between climate parameters, woody cover, and species richness, which is maintained even when the data are aggregated to reflect biases in the fossil record. However, a comparison of six independent fossil communities also provides new information on both Pliocene mammal community composition and vegetation, which did not show a similar correlation. This suggests that Pliocene rift valley woody cover is not as strong a predictor of productivity and richness as it is today in Kenya. This may be because woody cover was found to be higher overall throughout the middle Pliocene at the sites compared to today.