Extending the record of coastal aeolian landscape change into the Middle and Early Pleistocene: a multi-method comparison SAR-OSL, TT-OSL and post-IRIR, Walker Bay, South Africa

The dynamic Cape south coast of South Africa is widely recognised for its extensive occurrence of aeolian coastal dunes and older (cemented) aeolianites; the latter are thought to preserve records of dune formation spanning multiple glacial-interglacial cycles. However, existing records are dominated by Marine Isotope Stages (MIS) 1 and 5e ages. Extensive and (assumed) much older deposits have been identified, but are largely unstudied. By reconstructing their chronologies, we aim to generate new insights into coastal change during the Early to Middle Pleistocene and examine the factors that control long-term preservation of aeolianite deposits on a tectonically stable coastline. We anticipate preservation to be largely influenced by local topography and underlying geology.Here we focus on the geomorphic history of Walker Bay, southwest of Cape Town. Our aim is to integrate a suite of geophysical (ground penetrating radar-GPR), geochemical (ICP-MS, SEM) and luminescence (TT-OSL (quartz) and post-IR-IRSL (K-feldspar)) dating methods to unravel the chronology, structure and provenance of dune sands within the embayment. Initial results indicate an age range of ~1 million years to ~600,000 years. The applied methods TT-OSL, IR225, and IR290 show some results that are close to each other, while others vary outside the error range. We also observe several unusual aeolian deposits, some of which are far from the modern coastline and reach elevations of more than 250m above sea level (amsl). Other locations with closely juxtaposed aeolianites (dating to >600ka), substantially greater than any yet published for this coastline, and uncemented sands (late Pleistocene) dated to MIS-3, a period with fewer records & sea-levels were significantly lower than present. The results challenge existing models, which suggest that pulses of dune formation occurred primarily during the MIS-5 and MIS-1 highstands. Several questions arise as to the mechanisms of aeolianite formation/preservation at such heights and distances relative to the modern coast, and the results present further questions: Is Walker Bay Unique? Or are such complex suite deposits much more widespread? On this basis, we consider whether methodological and sampling limitations have led to a spatio-temporal biased record of long-term dune formation in this region. Or is it a completely different system than what has been observed on the coast before?