Evaluation and application of palaeoceanographic proxies at the East Antarctic margin

Palaeoceanographic studies based on geochemical proxies allow the reconstruction of ocean conditions under different climate regimes, which aids our understanding of how the climate system may respond to current anthropogenic disturbances. However, gaps remain in our knowledge of the biogeochemical cycles of some paleo-proxies in the marine environment, especially at geochemically complex interfaces such as ice-ocean margins. The aim of this thesis is to assess the reliability of several proxies of past ocean circulation and productivity in a region particularly sensitive to climate change: Wilkes Land margin. To do this, new geochemical datasets were generated through measurements of trace metal and isotope concentrations in six sediment cores and seawater samples collected from the East Antarctic margin.
This project firstly assessed the ability of neodymium isotopes (εNd) to track past bottom water mass circulation at the East Antarctic margin (Chapter 2). The authigenic ε_Nd signal leached from bulk marine sediment has been used as a tracer of deep ocean circulation, but in some locations, the leached ε_Nd signal disagrees with bottom water signatures, raising concerns about its reliability as a proxy. To address this, a short and weak reductive leaching procedure was applied to surface sediment samples and compared to the isotopic signature of the overlying bottom water. The sediment leaches were analysed for ε_Nd, rare earth elements (REE) and trace metal concentrations to constrain the origin of the leached isotopic composition. Results showed that the leached εNd signal does not reflect a bottom water ε_Nd signature, but rather originates from a minor, reactive, and isotopically unique mineral phase present in the sediment. Elemental data suggests phosphates are important REE sedimentary host-phases and that the authigenic εNd record is controlled by diagenesis in the studied sediment. These findings suggest that authigenic ε_Nd extracted via sediment leaching is not a suitable approach for reconstructing ocean circulation on the East Antarctic margin and highlights the importance of testing proxies’ reliability on a site-by-site basis.
The following paleo-proxy assessment focused on components related to ocean biological productivity and carbon export (Chapter 3), which play an essential role in the regulation of the global climate. Components such as organic carbon, opal and barite are known paleo-productivity tracers, however, two main issues may compromise their use in paleo-reconstructions: sediment resuspension and poor preservation. Firstly, vertical profiles of dissolved 230-thorium (²³⁰Th) were examined to determine the particle dynamics in this region, allowing a discussion of the robustness of the ²³⁰Th-normalisation technique to correct for sediment resuspension. Secondly, a multi-proxy record, (including chlorin, organic carbon, opal, excess barium ‘xsBa’ and excess cadmium ‘xsCd’) was examined to assess the proxies preservation in the surface sediment. The results show consistent variations between the Th-normalised fluxes of most proxies, indicating that the method used is a robust way to reconstruct paleo-productivity in the region. One exception was the preserved vertical fluxes of xsBa which appeared to be controlled by different processes than the other proxies. It seems that xsBa may not reflect marine barite (i.e., the barium fraction associated with productivity), but likely another mineral phase, implying that xsBa may not be a robust paleo-productivity proxy in sediments from the East Antarctic margin. This multi-proxy investigation provides new insights into the sedimentary geochemical cycle of Ba in Wilkes Land margin sediment and recommendations for future paleo-productivity reconstructions.
Finally, the most robust tracers of ocean productivity determined in the previous chapter were applied on a Kasten core from the same region, to reconstruct productivity variations since the last deglaciation (~ 22 thousand years; Chapter 4). The ²³⁰Th-normalised productivity proxies displayed enhanced fluxes during the deglaciation (~ 22-18 kyr) and Holocene Optimum (HO; ~9 kyr), while low values were found during the Antarctic Cold Reversal (~14-12 kyr) and the late Holocene. Measurements of the ratio of stable nitrogen isotopes (δ ¹⁵N) in the bulk organic sediment fraction showed a high nutrient availability during the HO, suggesting the high productivity observed at this period resulted from an increase in upwelling intensity of warm deep water, which generated substantial ice melt and provided light for phytoplankton photosynthesis. This study helps our understanding of productivity drivers during warmer climate and their implication for the oceanic carbon pump along the Antarctic margin.
This thesis contributes to a better understanding of the marine geochemical sedimentary cycle of various paleo-proxies at the East Antarctic margin and provides recommendations for future palaeoceanographic reconstructions in this area. This work notably emphasises the unique behaviour of authigenic Nd and Ba in sediments collected off Wilkes Land, which needs further investigation. Results also showed that the use of ²³⁰Th-normalised fluxes of biogenic components such as organic carbon and opal is a robust method to reconstruct paleo-productivity in the studied area, and that other less conventional proxies (such as chlorin or excess cadmium) should be used more widely. Overall, this thesis confirms the importance of assessing the reliability of paleo-proxies either by comparison with modern observations or by using multi-proxy records. More geochemical data from the Southern Ocean, in co-located seawater, sediment, particle and porewater samples would help to further increase our understanding of trace metal and isotope marine geochemical cycles. Further work is needed on the Antarctic margin to enable better calibration and accurate use of paleo-proxies, essential to understanding past and future ocean processes.