Further development and application of the speleothem U-Pb geochronometer

© 2020 John Richard EngelSpeleothems are valuable archives recording information on past climates and processes of landscape evolution. In fact, the high preservation potential of speleothems and their ubiquity across the globe are key attributes and speleothems are now extensively used in palaeoclimate modelling as complements to the more traditional marine and ice core climate records. The real ’engine’ of speleothems science, however, lies in the ability to reliably and robustly place these archives into chronological context. Historically this has been facilitated by the highly successful U-Th geochronometer, which can accurately date speleothems that formed within the last 650 thousand years. However, many speleothems – and by extension their proxy archives – are far older than this 650 ka limit and so require a significantly longer-lived radiometric methodology: the U-Pb geochronometer has recently filled this void. The U-Pb geochronometer is well known in the geological sciences, but its application to the analytically-challenging medium of speleothems has been relatively recent. Modern mass spectrometry methods now allow the accurate and precise analysis of the extremely low-levels of U and Pb within speleothems, thus providing access to such ’deep-time’ (i.e. multi-million years and further) palaeoclimatic and landscape evolution archives. Despite these significant advances, there remain fundamental impediments to the whole-scale production of speleothem U-Pb ages. This thesis investigates some of these shortfalls with the ultimate aim of advancing the speleothem U-Pb methodology and application such that future speleothem-based studies may more rapidly and more reliably place their reconstruction models into a more robust chronology. The first half of this thesis addresses two fundamental limitations of the speleothem U-Pb method: the large expenditure of time and effort required for producing calcite U-Pb ages, and the potential limitations in the correction for initial isotopic disequilibrium effects in the U-Pb decay chain. To address the former, I provide new software for the rapid production of publication-quality U-Pb isochron figures and ages. Additionally, a new chemical separation (i.e. chromatography) procedure for generating high-purity U and Pb fractions is developed. This new ’stacked resin’ protocol can produce mass spectrometry-ready U and Pb sample aliquots in a single working day – roughly half the time of the previous method – and so accelerates the acquisition of U-Pb ages. In terms of subsequent data deconvolution, the largest hurdle to obtaining accurate speleothem U-Pb ages arguably relates to the uncertainty added from having to estimate an initial isotopic disequilibrium value after the system has returned to equilibrium. As a potential remedy to this problem, this research program develops a framework in which to calculate appropriate disequilibria values for the production of disequilibrium-corrected U-Pb ages. The second part of this thesis is designed to highlight new applications of the calcite U-Pb chronometer made possible by the methodological advances documented in the first half. This research program investigates the utility of speleothems as proxies for regional uplift rates in karst terranes. This is accomplished by producing chronologies for 120 speleothems from the Buchan karst along southeast Australia’s passive margin. The results of this study indicate that SE Australia experienced a renewed period of uplift occurring at a maximum rate of 76 +/- 7 m / Ma beginning at least 3.5 million years ago. This speleothem-derived uplift rate is consistent with other independent regional uplift estimates thus adding credibility to the utility of speleothems as proxies for uplift. The relatively straight-forward method developed here is likely applicable in many future karst studies. Additionally, this research program also investigated the potential applicability of U-Pb geochronology to another form of carbonate archive – stromatolites. The methodological advances described in this thesis were applied to stromatolites from Lake Turkana, Kenya, and the palaeo-lake Bungunnia in southern Australia. The results of this experiment suggest that stromatolites contain too much inherited Pb that simply overwhelms the radiogenic ’age signal’ and so are unlikely to be successful candidates for U-Pb dating, at least for stromatolites formed during much of the Cenozoic period