U14A-01 16:00h
Simulated Trends in African Glacial and Interglacial Vegetation: Implications for Late-Pleistocene Hominid-Plant Interactions
Most theories of human evolution in south, central and eastern Africa are predicated on the assumption that savannas and grasslands almost exclusively dominated Pleistocene (glacial) landscapes. It was our aim to evaluate this assumption using a state-of-the-art fully-coupled earth system model (HadCM3LC), which we used to predict potential palaeovegetation following representative glacial and interglacial climate-forcing. Our glacial simulations indicate that tropical broadleaf forest was not severely displaced by grassland expanding into central Africa, although the outer extent of closed forest decreases, particularly in the north. Our vegetation-climate simulations also indicate that the extent of closed tropical forest during typical interglacials is not represented by today's observed vegetation distributions. Simulated interglacial climate results in expansion of tropical forest from coast-to-coast across much of central Africa. Our modelling experiments have implications for interpreting biogeography and phylogenies of various African plant and animal species, including the evolution of our own species, Homo sapiens sapiens.
U14A-02 INVITED 16:20h
Biomarkers Reveal Late Neogene Changes in East African Vegetation
Multiple paleoenvironmental records incorporating diverse proxies and sites are needed to build a complete picture of African environments during the late Neogene period of hominid evolution. Augmenting the picture of progressive aridification of Neogene environments indicated by the progressive enrichment of soil carbonates (Cerling and Hay, 1988; Cerling, 1992; Wynn, 2000), Hominid environment-evolution hypotheses have drawn on evidence for high frequency environmental variability to emphasize the potential importance of climate variability as an evolutionary selection factor (Potts, 1996). Dramatic evidence for Neogene climatic variability is seen in high benthic foraminiferal oxygen isotope records of high latitude climate (Shackleton, 1990, Mix et al., 1995). This `global' signal has also been detected in low latitudes as shown by orbital frequency climate variability in dust records from marine cores around the African continent (deMenocal, 1995). Terrestrial records may provide a more direct record of local habitat change than distal dust deposits, but are unable to resolve with continuity the orbital frequency climatic variability seen in marine records. We therefore turn to molecule-specific carbon isotopic analyses of plant biomarkers in marine sediments to indicate regionally integrative C3-C4 vegetation changes. The validity of this approach has already been demonstrated in marine cores off West Africa (Schefuss et al, 2003). In order to bridge the gap between large scale climatic conditions and local rift valley conditions, we select a marine core in close proximity to hominid sites in the East African rift valley to generate a high resolution, regional integrative record of vegetation change during the Neogene. We take the approach of high resolution (3ka) sampling, capable of resolving precessional variability, in `windows' downcore to provide a measure of the mean and variability of vegetation for timesclices during the last 10Ma. Our results show that the long term trend towards carbon isotopic enrichment is consistent with paleosol data. Biomarker results further demonstrate that orbital frequency isotopic variability in the Pliocene and Pleistocene was of equal or greater magnitude than the long term trend. High frequency environmental variability is also seen at ~9Ma, where although carbon isotopic compositions indicate exclusively C3 vegetation, biomarker fluxes indicate significant environmental variability on precessional timescales. These initial biomarker results suggest the importance of high frequency environmental variability in hominid habitats in excess of long term change.
U14A-03 16:40h
Stepping Out: Investigating Hominin Dispersals Within a Palaeoclimatic Framework.
At present the number of well dated early hominin sites (Homo ergaster, Homo erectus etc.) is insufficient to reconstruct the patterns of dispersal and occupation within the Old World. Many questions still remain therefore about the dominant factors influencing the patterns of arrival and their timing. In order to understand these dispersal patterns a modelling framework has been developed, for the late Pliocene, early Pleistocene period (2Ma to 0.5Ma). This time period corresponds to what is believed to be the first expansion of hominin out of Africa. The framework integrates data sets of vegetation patterns, land mass configuration, climate, and orography. In order to estimate vegetation patterns during the simulation a simple method has been developed to combine benthic foraminifera ocean core data with snapshot climate model simulations. The climate estimate produced can then be used to simulate global vegetation patterns using a global vegetation model. The dispersal of hominin is then simulated and the predicted dispersal patterns compared to existing hominin finds.
U14A-04 17:00h
LGM to Holocene African climate variability and the Sr, Nd, and Pb isotope ratios of eolian deposits off northwest Africa
Abrupt changes in African climate from the last glacial period through the Holocene have been documented in a core located off Cap Blanc, northwest Africa, ODP Hole 658C (deMenocal et al, 2000). Large changes in terrigenous sedimentation to the ocean, due in large part to humid-arid shifts in terrestrial vegetation, are observed in the percent terrignous and Th-230 normalized terrigenous flux measurements. While there is overall high biogenic and eolian terrigenous sedimentation, terrigenous input decreased during the African Humid Period (AHP, 14.8-5.5 kyr). Additionally sea surface temperatures were lower during that time compared to the later Holocene. Others have shown that radiogenic isotope isotope ratios of terrigenous deposits from marine and ice cores vary over time and provide provenance information. Alternatively, it has been suggested that changes in Sr isotope ratios of dust may relate to the degree of chemical weathering. We sampled the <63 micron detrital fraction to measure the Sr, Nd and Pb isotope ratios in order to quantify the eolian dust proportion and potential source regions. Nd and Pb measurements record source area contributions. However, Sr isotope data indicate a clear shift toward less radiogenic values during the African Humid Period, where as there is no corresponding change in Nb and Pb values. Additionally, we observe an abrupt termination of the AHP and other rapid events. This research seeks to enhance our understanding of African and low-latitude responses to global climate forcings over the last deglaciation.
U14A-05 17:20h
A new Application of Ar-40/Ar-39 Dating: A Provenance Study of Middle Stone Age Obsidian Tools from Ethiopia
The identification and geographic origin of potential sources for obsidian tools manufactured by MSA (Middle Stone Age) hominid groups is important to reconstruct source utilization and trade routes. Obsidian sourcing is done by geochemical investigation of obsidian artifacts and potential source materials and the identification of possible matches between them [e.g., 1]. Due to uncertainties arising, e.g., from intrasource inhomogeneities [2] additional methods are desirable. As a part of a pilot study, we dated by Ar-40/Ar-39 stepwise heating numerous debitage pieces of obsidian artifacts from an MSA excavation site and several potential source rock samples from a nearby obsidian outcrop in the Gademotta-Kulkuletti area near Ziway lake, Ethiopia. The ages were used, along with chemical data, to trace possible source rocks used to manufacture the artifacts. Most of the debitage samples show ages around 1.29 Ma and exhibit flat, well behaved spectra. Less well behaved spectra are obtained for two debitage samples with ages of about 1.26 Ma, and two with ages of about 0.62 Ma. We point out that despite this significant spread in the ages found, all debitage samples as well as the potential source rocks show homogeneous major and trace elemental compositions generally indicating a positive match in archeological sourcing. The ages of two potential source rocks are about 1.26 and 0.87 Ma. Thus, while there is a match with the two debitage pieces dated at 1.26 Ma, the exact outcrop of the major source for the 1.29 Ma old artifacts has not been identified yet. A second set of samples with further possible source materials for the Kulkuletti artifacts as well as another potential debitage-source material pair from a second excavation site in the Ziway lake area, Porc Epic, are currently being dated and will be presented at the conference. Already this first set of data shows not only the general applicability of Ar-40/Ar-39 dating to archeological obsidian sourcing, but also its great potential as a complementary tool revealing more detailed information than the geochemical data alone. [1] Merrick and Brown (1984) African Archaeological Review 2:129-152. [2] Shackley (1998) Journal of Archaeological Science 25: 1073-1082.