TERRESTRIAL COSMOGENIC NUCLIDE GEOCHRONOLOGY OF STRATIFIED MIDDLE STONE AGE ARTIFACT ASSEMBLAGES, NORTHERN MALAWI RIFT, EAST AFRICA

Northern Malawi contains a rich record of Middle Stone Age (MSA) archaeology in Pleistocene-to-Holocene fluvial and alluvial fan deposits adjacent to Lake Malawi. The sites are situated within a landscape sensitive to documented changes in climate and lake level over the last 500 ky. Work in the 1960s hinted at a diversity of stone artifact manufacture-and-discard strategies that parallel late Pleistocene assemblages in south Africa. However, the chronological, palaeoecological, and depositional contexts within which to contextualize this diversity were unavailable. New work on artifact-bearing fluvial and alluvial fan deposits along the broad coastal plain hangingwall of the Malawi Rift suggest geomorphologic context of these sites influence dating strategies. Chronology, therefore, is built from multiple localities, each with a unique history of postdepositional surface processes.

Large MSA assemblages have been excavated from 7 sites along the hangingwall; additional artifacts and observations have been recovered from 21 trenches within the sedimentary surfaces including sampling for terrestrial cosmogenic nuclide (TCN) dating, optically-stimulated luminescence (OSL) dating, and characterization of sediment composition and morphology.

Two independent methods allow us to date these surfaces: (1) depth profiles, using 10Be and stable 21Ne—the latter extends our dating ability into the Pliocene; (2) burial isochrons, which rely on TCN pair 26Al/10Be—inherited concentration of nuclides in samples recovered at depth (~2-3m) are used to measure burial ages up to 5 Ma.

We use a recent innovation in the field of TCN dating: the isochron approach, used to converge on a best-fit burial date by iteratively correcting for estimates of post-burial production. This approach allows more flexibility than conventional burial dating to account for factors such as post-burial TCN production. Previous estimates of deposition rates and other a priori knowledge of absolute dates within a stratigraphic column allow a more efficient approach to the calculation of a burial date. Our approach uses independent constraints on the geologic setting in concert with reasonable estimates of the depositional environment to more quickly constrain uncertainties in the calculation of an age.