Clumped isotope analysis of central Australian carbonates: a potential palaeoclimate proxy for Australia’s arid interior

Abstract

Quantitative records of past temperature variability in arid environments are crucial for validating climate models and their ability to capture the full range of the Earth’s surface. However, arid region temperature reconstructions are rare, particularly in the Southern Hemisphere, including Australia. The recently developed clumped isotope thermometer provides a novel approach to potentially address this demand by allowing the estimation of carbonate precipitation temperature independent of environmental water isotopic composition. Two types of carbonate materials offer potential for clumped isotope temperature reconstructions in arid central Australia: fossil mollusc shells deposited within the shoreline sediments of now dry lakes, and tufa deposits formed in springs fed by continuous discharge of Great Artesian Basin groundwater. Clumped isotope analysis was conducted for a suite of tufas from central Australian mound spring deposits in addition to a series of lake shells from the same region. Micro-XRF mapping was used to evaluate sample quality and heterogeneity, alongside XRD analysis. In addition to benefitting clumped isotope analyses, these data may improve the data quality and success of U-series dating of mound spring tufas. Air temperatures inferred from tufa Δ47 measurements suggested temperatures ~5°C cooler than modern-day conditions during the early Holocene, which supports previous palaeoclimate models for central Australia. Palaeotemperatures ~17°C cooler than present were inferred from shells during 70-35 ka, with two warmer periods coinciding with lake filling episodes during ~48 and ~33 ka. Carbonate δ18O appeared largely to have been driven by changes in water δ18O for lakes but not mound springs, reflecting different hydrological controls on the two water sources. Agreement between temperatures and palaeoclimate models suggest clumped isotope analysis may function as a valuable quantitative palaeotemperature proxy in central Australia. Analysis of additional tufa and shell samples should be prioritised, in addition to further studies into the genesis of different tufa.