Intrataxonomic trends in herbivore enamel δ13C are decoupled from ecosystem woody cover.

Analysis of enamel stable carbon isotopes (δ13Cenamel) of fossil herbivores is an important tool for making inferences about Plio-Pleistocene vegetation structure in Africa and the environmental context of hominin evolution. Many palaeoecological studies implicitly or explicitly assume that individual variation in C3-C4 plant consumption among fossil herbivores directly reflects the abundance of C3 (trees, shrubs) or C4 (low-altitude tropical grasses) vegetation. However, a strong link between δ13Cenamel of herbivores and ecosystem vegetation structure has not been rigorously established. Here we combine δ13Cenamel data from a large dataset (n = 1,643) with multidecadal Landsat estimates of C3 woody cover across 30 African ecosystems to show that there is little relationship between intrataxonomic variation in δ13Cenamel and vegetation structure. This is especially true when removing forested ecosystems (>80% woody cover)-which numerous lines of evidence suggest are rare in the Plio-Pleistocene fossil record of eastern Africa-from our analyses. Our findings stand in contrast with the common assumption that variation in herbivore δ13Cenamel values reflects changes in the relative abundance of C3-C4 vegetation. We conclude that analyses using herbivore δ13Cenamel data to shed light on the environmental context of hominin evolution should look to explicitly community-level approaches for making vegetation inferences.

Late Pliocene environmental change during the transition from Australopithecus to Homo.

It has long been hypothesized that the transition from Australopithecus to Homo in eastern Africa was linked to the spread of open and arid environments near the Plio-Pleistocene boundary, but data for the latest Pliocene are scarce. Here we present new stable carbon isotope data from the late Pliocene mammalian fauna from Ledi-Geraru, in the lower Awash Valley (LAV), Ethiopia, and mammalian community analyses from the LAV and Turkana Basin. These data, combined with pedogenic carbonate stable isotopes, indicate that the two regions were largely similar through the Plio-Pleistocene, but that important environmental differences existed during the emergence of Homo around 2.8 million years ago. The mid-Pliocene to late Pliocene interval in the LAV was characterized by increasingly C4-dominated, arid and seasonal environments. The early Homo mandible LD 350-1 has a carbon isotope value similar to that of earlier Australopithecus from the LAV, possibly indicating that the emergence of Homo from Australopithecus did not involve a dietary shift. Late Pliocene LAV environments contrast with contemporaneous environments in the Turkana Basin, which were more woody and mesic. These findings have important implications for the environmental conditions surrounding the emergence of Homo, as well as recent hypotheses regarding Plio-Pleistocene environmental change in eastern Africa.

Thinking locally: Environmental reconstruction of Middle and Later Stone Age archaeological sites in Ethiopia, Kenya, and Zambia based on ungulate stable isotopes.

Our knowledge of the Pleistocene environments of Africa consists primarily of data at a scale too coarse to capture the full habitat variation important to hominins 'on the ground.' These environments are complex, highly variable, and poorly understood. As such, data from individual sites are a needed addition to our current paleoenvironmental reconstructions. This study offers a site-based approach focusing on stable isotope analyses of fossil faunal tooth enamel from three archaeological sites in tropical Africa. Carbon and oxygen stable isotope data are reported from the sites of Porc Epic, Ethiopia, Lukenya Hill, Kenya, and Kalemba Rockshelter, Zambia. Stable isotope data from tooth enamel are used to measure two environmental variables: (1) aridity based on oxygen isotope composition and (2) dietary reconstructions of fossil ungulates based on the relative proportions of C3 browse and C4 graze in the diet. These data allow for a preliminary assessment of existing models that attempt to explain the behavioral and technological variation characteristic of the transition between the Middle and Later Stone Ages. Results indicate spatial and temporal variation in aridity and phytogeography in tropical Africa during the Pleistocene, suggesting that no single model is likely to provide an explanation for the transition at all sites across Africa.