Luminescence dating and geochemistry constraints on late Pleistocene hydrological and ecological change for north-eastern Australia

Abstract

Extensive debate surrounds the timing and cause of late Pleistocene megafauna extinction across the continent of Sahul (Australia and New Guinea). Central to these debates is uncertainty in the timing and impact of key events during the last glacial cycle (125 – 12 thousand years ago; ka) including the arrival of humans on the continent, the effect climate change had on environments, and the geographical and temporal patterns of megafauna extinction. This uncertainty can be attributed to the lack of well-constrained palaeoenvironmental records, limited spatial coverage and dating reliability. To address these knowledge gaps, this thesis applies geochronological and geochemical techniques to a series of new palaeoenvironmental archives to investigate megafauna extinction in the context of climate-induced environmental change. Chapters 2 and Chapter 3 of the thesis present optically stimulated luminescence (OSL) and radiocarbon (14C) dating results on palaeoenvironmental archives (sediment cores) from two subtropical wetlands (Welsby Lagoon, Chapter 2; Brown Lake, Chapter 3). Analysis of twenty-one OSL ages and seven 14C ages from Welsby Lagoon (Chapter 2) highlights the importance of carefully considering the material being dated, especially where contamination is likely. In general, 14C ages derived from identifiable terrestrial plant macrofossils agree with OSL ages, while 14C results derived from unidentifiable plant remains and charcoal are scattered and produce stratigraphic inversions. The need to consider different sediment accumulation histories on a site-by-site basis is also demonstrated in Chapter 2 and Chapter 3 using Bayesian modelling. Further geochronological results are reported in Chapters 4 and Chapter 5, which combine OSL, ESR and U-series dating to establish megafauna last appearance at the new tropical megafauna fossil locality of South Walker Creek, Queensland (Chapter 4 & Chapter 5). Hierarchical Bayesian modelling – which incorporates 66 new ages from eight sites – is used to provide new evidence for megafauna disappearance in the Fitzroy River catchment between 41.7 and 35.8 ka (95 % C.I.; Chapter 5). Additionally, hierarchical Bayesian modelling of fluvial and fossil deposits at the South Walker Creek fossil locality indicates a decrease in hydrological activity in the area during mid- to late- Marine Isotope Stage 3 (MIS3) (Chapter 5). The comprehensive dating approach presented in each study (Chapters 2 – 5) enables improved examination of the spatiotemporal pattern of climate-induced environmental change taking place across eastern Australia immediately prior to, during, and after, megafauna disappearance (Chapter 5). Alignment of XRF and age data from Welsby Lagoon shows two phases of increased dust deposition at 71 – 67 ka and 58 – 48 ka (Chapter 2) – the latter coinciding with the wetland transitioning from a lake to swamp. Brown Lake also experienced an increase in dust deposition through late MIS3, in addition to entering a phase of shoreline regression lasting until the Holocene (Chapter 3). Notably, when these sites are included in a regional spatiotemporal synthesis of hydrological activity (compiling data from other wetlands and regional catchments) and are considered in the context of local megafauna records (Chapter 4), there is evidence to suggest megafauna disappearance in the Fitzroy River Basin catchment was coincident with broader environmental change during mid- to late- MIS3 (Chapter 5). This thesis employs a robust, geochronologically driven approach to examine ecological and environmental change during significant and understudied periods of Australian prehistory (e.g., pre- LGM). The overarching conclusion is that there is little support for rapid or synchronous human-mediated, continent-wide megafauna extinction during the late Pleistocene. Rather, the data suggest extinction coincided with a period of prolonged hydroclimate deterioration initiated during the mid- to late-MIS3.