Oxygen isotope ratios of phytoliths as a proxy for past climates and environments

Abstract

Past climate and vegetation records are essential to understand the natural range of environmental variability over timescales beyond the instrumental era. Tracers of past climate change are especially needed for arid regions, which are disproportionately sensitive to global climate warming, yet poorly represented in palaeoclimate archives. The oxygen isotope ratios (18O/16O) of phytoliths (silica microfossils produced by plants) hold the potential to serve as a new proxy for past climates in dry environments, namely for temperature and relative humidity. To use this novel proxy, it is necessary to extract phytoliths of sufficient purity for oxygen isotope analysis and to have a deep understanding of the processes controlling the oxygen isotope signature of the phytoliths in sediments. The aims of this thesis are to improve the methods for extraction and analysis of oxygen isotope ratios of phytoliths, and to quantify the relationship between climate and the oxygen isotope ratios to interpret variations. The oxygen isotope analysis of biogenic silica requires high purity samples. To improve and optimise phytolith extraction from plants and soils for oxygen isotope analysis, several methods were tested and compared to a novel approach using non thermal plasma ashing generated by radio frequency excitation of air under vacuum. Residual contamination was assessed using microscopy and Fourier Transform infrared-spectrometry modelling. Plasma ashing was found to have potential for organic material removal from plants, however the procedure is currently time consuming and requires further improvements to become a routine procedure. For extracting soil phytoliths, an added step of sulphuric acid and hydrogen peroxide (Piranha solution), normally reserved for extracting phytoliths from plants, produced samples with the required purity. Biogenic silica (e.g., diatoms and phytoliths) contains exchangeable oxygen that can affect the accuracy of isotopic analysis. The processes to account for this exchangeable oxygen require specialised equipment that is not always available. A novel approach for silica dehydroxylation was developed, involving sample heating and fluorination in an Elemental Analyser, a common piece of isotope analysis equipment. In situ high temperature dehydroxylation was found to adequately replicate established oxygen isotope ratios for international silicate standards and can be used for silica samples as small as 600 μg. To be able to use phytolith oxygen isotope ratios as a palaeoclimate proxy, the impact of climate needs to be understood. To do this, measured phytolith oxygen isotope ratios from plants and soils from a transect across Australia were analysed. Modern plant phytolith oxygen isotope ratios did not exhibit the expected strong correlations with climate, but oxygen isotopes in soil phytoliths did, correlating strongly with mean warmest period relative humidity at the maximum temperature and the mean annual temperature. The oxygen isotope ratios of phytoliths from plants and soils could be successfully predicted using Craig Gordon leaf water modelling and temperature-dependant silica fractionation, indicating that the key environmental processes are well characterised. The oxygen isotope composition of fossilised plant phytoliths provides the potential to make qualitative inference of climate change. However, the influence of multiple climate variables means that the method would be best employed within a multi-proxy framework.