Extending the record of lacustrine phases beyond the last interglacial for Lake Eyre in central Australia using luminescence dating

Abstract

We show with multiple luminescence dating techniques that the sedimentary record for Lake Eyre, Australia's largest lake, extends beyond 200 thousand years (ka) to Marine Isotope Stage (MIS) 7. Transgressive clayey sand and finely laminated clays overlying the Miocene Etadunna Formation in Lake Eyre North document the deep-lake phases of central South Australia in the past. Until now, unresolved chronology has hampered our ability to interpret these sedimentary records, which are important for understanding the timing of the wettest phase of central Australia's late Quaternary climate. In this study, we apply quartz optically stimulated luminescence (OSL) dating, thermally-transferred OSL (TT-OSL) dating and K-feldspar post infrared infrared stimulated luminescence (pIRIR) dating to lake-floor sediments near Williams Point in Madigan Gulf to provide new age constraint for the lacustrine sediments of Lake Eyre. Methodological studies on quartz and K-feldspar demonstrate that these luminescence dating procedures are suitable for the Lake Eyre lacustrine samples and produce consistent replicate ages. A Bayesian model applied to the new dating results provides a chronological model of lacustrine deposition and shows that the transgressive clayey sand were deposited 221 ± 19 ka to 201 ± 10 ka and that the deep-water sediments were laid down in early MIS 6 (191 ± 9 ka to 181 ± 9 ka). We also find evidence for a potential depositional hiatus in mid MIS 6 and the likely formation of a palaeo-playa later in MIS 6 from 158 ± 11 ka to 143 ± 15 ka. In contrast, the MIS 5 sediments are characterised by oscillating deep-and shallow-water lacustrine units deposited 130 ± 16 ka to 113 ± 20 ka. This study is the first of its kind to provide evidence for a wet desert interior in Australia beyond the last glacial cycle using comprehensive numerical dating. Our results show that past deep-lake episodes of central South Australia, which were previously thought to represent peak interglacial conditions, are actually associated with both warm interglacial and cold glacial periods, with all the wettest episodes generally coinciding with the intervening periods between the glacial and interglacial maximums. We assume from these results that orbital forcing is not a first order control for the long-term dynamics of the Lake Eyre basin and the Indo-Australian monsoon. The high lake-level events of Lake Eyre are well correlated with millennial-scale cooling events and stadials of the North Atlantic, and coincide with weakened episodes/events for the East Asia summer monsoon. This may imply an important role for the northern high latitudes in influencing the Indo-Australian monsoon, which may be associated with a southward migration of the Intertropical Convergence Zone (ITCZ) during cooling periods in the North Atlantic.

Single-grain TT-OSL bleaching characteristics: Insights from modern analogues and OSL dating comparisons

“…Multigrain residual D e values of 5-19 Gy have been reported for several modern aeolian sediments from Eurasia and South Africa (see Duller and Wintle, 2012). Arnold et al (2014) reported a similarly sized multigrain D e of 7.3 ± 0.8 Gy for a modern slopewash and aeolian deposit from north-central Spain, while multi-grain residual D e values of several tens of Gy have been obtained for coastal and lacustrine shoreline deposits from South Africa and Australia (Jacobs et al, 2011;Fu et al, 2017). In contrast, very large multi-grain TT-OSL residual doses of 250-300 Gy have been reported for modern suspended sediments and overbank deposits from the Yellow River (Hu et al, 2010), potentially cautioning against the suitability of TT-OSL dating in turbid and UVdepleted fluvial settings.…”

“…Thirteen samples were collected from actively accumulating, or very recently accumulated, surface sediment deposits that were expected to yield burial doses close to, or consistent with, 0 Gy (assuming adequate signal bleaching during transportation). These samples represent modern analogues for associated archaeological, palaeontological and palaeoenvironmental dating samples being studied as part of recent or ongoing TT-OSL dating projects (e.g., Arnold et al, 2014;Demuro et al, 2014;Fu et al, 2017). Two shallow cave infill samples from the middle Pleistocene palaeoanthropological sites of Galería and Sima del Elefante, Atapuerca, (ATG10-3, ATE10-13) have been chosen for the daylight bleaching experiments, owing to their relatively high and comparable mean burial doses, and uniformly bleached single-grain TT-OSL D e distributions (e.g., Demuro et al, 2014;.…”

“…While these various TT-OSL bleaching assessments have proved insightful, they are based on a relatively modest number of observations (n = < 20 samples) and further work is needed to better characterise TT-OSL signal resetting across a broader range of natural contexts using complementary types of experimental procedures. Additionally, all existing assessments of TT-OSL bleaching characteristics, with the exception of one study (Fu et al, 2017), have been performed at the multi-grain scale of D e analysis. It remains unclear, therefore, whether TT-OSL residual doses reported in existing modern analogue studies partly reflect averaging effects arising from simultaneously measuring grains with different bleaching histories, signal compositions or TT-OSL source trap properties.…”

“…Single-grain TT-OSL has recently been applied at several independently dated archaeological sites from Spain and Australia (e.g., Demuro et al, 2014;Hamm et al, 2016). These single-grain studies have also revealed that multi-grain TT-OSL signals may be dominated by grains with unfavourable TT-OSL behaviours (e.g., and that apparent multi-grain TT-OSL residual doses of several tens of Gy may result from the inclusion of grain types that are routinely rejected by single-grain quality assurance criteria (Fu et al, 2017). Such complications require further examination, and additional single-grain bleaching assessments are needed to better characterise TT-OSL signal resetting at the most fundamental scale of D e analysis.…”

The first successful application of Optically Stimulated Luminescence dating to a colonial era (<0.25 ka) archaeological site in Australia

“…The abundance of quartz dominated landscapes and the poor preservation of organic remains for other radiometric techniques has resulted in a dominance of the approach in recent literature (e.g. ; Fitzsimmons et al, 2014;Fu et al, 2017;Hughes et al, 2014;Williams et al, 2014Williams et al, , 2017. To date, however, the technique has largely focussed on Indigenous archaeological sites, often of considerable age (Dortch et al, 2016;Veth et al, 2017).…”