Abstract
The redox status of the Earth’s ocean-atmosphere system and the extent to which it is enriched or depleted in oxygen controls the oxidation state of the metallic elements, dictating essential chemical reactions for life, such as photosynthesis and respiration. Therefore, the fluctuating oxygenation conditions of Earth’s biosphere is intrinsically linked to the evolution and diversification of complex eukaryotic life. Throughout much of the Earth’s 4.6 billion year history, the ocean-atmosphere system has been severely depleted in free oxygen, resulting in globally widespread anoxic, ferruginous (anoxic and iron-rich), and euxinic (anoxic and sulphidic) conditions within the ocean. It is only in the last ~600 million years that free oxygen levels in the oceans and atmosphere reached near modern levels, resulting in less than 1 % of the modern global ocean being anoxic.
Because of the interconnected relationship between Earth’s redox state and the evolution of life, it is vital to be able to accurately and reliably reconstruct past ocean redox changes, to better aid in our understanding of the evolution of the Earth’s system. This has been achieved in recent years, in large part, due to the advent of advanced analytical techniques, such as multiple-collector inductively coupled plasma-mass spectrometry (MC-ICPMS), which has allowed for the high precision analysis of metal isotope variations in sedimentary archives. The analysis of redox-sensitive element concentrations and isotopic variations in marine sediments has led to the development of palaeo-redox proxies, which are capable of reconstructing oxygenation fluctuations during geological intervals of major climatic and ecological change. However, a detailed understanding of the mechanisms which control the biogeochemical cycling of redox-sensitive elements and their isotopes across a range of modern ocean redox settings is necessary for the reliable deployment of these tracers.
This study investigates the biogeochemical cycling of redox-sensitive elements and the uranium ‘stable’ isotope system within Framvaren Fjord, a permanently euxinic fjord in Norway, aiding further calibration efforts. Framvaren Fjord represents an ideal natural laboratory for studying the behaviour of redox-sensitive elements and their isotopes, as it possesses the highest levels of hydrogen sulphide of any modern anoxic marine basin. As such, Framvaren Fjord is often considered to be the main ‘type location’ for modern euxinic conditions analogous to much of the Earth’s geological past. This study is initially focussed around a high-resolution depth profile of dissolved waters and particulates sampled from within the water column of Framvaren Fjord allowing for the simultaneous comparison of redox-sensitive elements within the dissolved and particulate phases, with emphasis over the redoxcline, representing the boundary between oxic and anoxic waters and uniquely located at 18 ± 1 m depth, in close association with the biological activity of the photic zone. The results show that the primary mechanisms controlling the distributions of redox-sensitive elements within the water column of Framvaren Fjord include water mass mixing induced by inflowing seawater, biological uptake and utilisation, redox reactions resulting in solubility changes, and the formation of insoluble metal-sulphides. Based on the unique behaviour of individual redox-sensitive elements in the water column, the robustness of a variety of palaeo-redox proxies was assessed using the seawater-derived, authigenic sediment concentrations in four sediment cores extracted from different chemically zoned regions of the overlying Framvaren Fjord water column. The distributions further confirmed the reliability of some palaeo-redox proxies when applied to the severely de-oxygenated and sulphidic conditions of the fjord, but demonstrated that proxies relying on the sedimentary concentrations of nickel and cobalt could not accurately predict the depositional environment of Framvaren Fjord. This was attributed to the unique biogeochemical cycling of nickel and cobalt observed within the overlying water column that differed from their distributions in other anoxic basins.
Additionally, the systematics of the uranium (U) isotope system (monitored as the 238U/235U ratio; reformulated as δ238U) was investigated within the overlying water column and underlying sediment pile of Framvaren Fjord, for the first time in this study, to provide further constraints on the utility of the δ238U palaeo-redox proxy. The U isotope mass balance was investigated within the water column of Framvaren Fjord, through the parallel characterisation of the dissolved and particulate U phases. Removal of up to 21 % of U from the water column first occurs at the onset of reducing conditions at ~17.4 m depth, due to the reduction of soluble U(VI) to insoluble and particle-reactive U(IV). The reduction and removal of U coincides with an isotopic excursion towards lower δ238U values in the dissolved phase and higher δ238U values in the particulate phase, displaying a maximum isotopic offset of 1.16 ‰. The U isotope systematics describing U reduction and removal were investigated using a closed- and open-system Rayleigh fractionation model, producing U isotope fractionation factors of ε = -0.99 ± 0.05 ‰ and ε = -1.08 ± 0.05 ‰ (1SE), respectively. These values are the largest observed for any modern anoxic marine basin, and overlap with the full ‘intrinsic’ ε-values describing U(VI)-U(IV) reduction derived from experimental and ab initio molecular orbital studies and microbial-mediated U reduction experiments. This provides strong evidence that U(VI)-U(IV) reduction is occurring in the water column of Framvaren Fjord, contradicting the predominantly held assumption that U(VI)-U(IV) reduction occurs primarily within the interstitial pore-waters of underlying anoxic sediments.
The uranium isotope systematics within the sediment pile of Framvaren Fjord were simultaneously investigated within the interstitial pore-water and authigenic sediment phases, through the distributions of U concentrations and δ238U compositions. This is the first time paired measurements of these two phases had been investigated in a modern anoxic marine basin. The results show that the sediment cores extracted beneath euxinic conditions within the overlying water column, do not display the expected pore-water and sediment δ238U compositions for a restricted anoxic basin overlain by a permanently euxinic water column, nor agree with dynamic reactive transport models. The unique behaviour of δ238U within the sediment pile of Framvaren Fjord has been attributed to several potential processes, including U(VI)-U(IV) reduction occurring within the overlying water column and sediment, resulting in the two-step isotopic fractionation of U, identified for the first time within a natural anoxic marine environment, the complexation of U with organic matter, or that the U isotope system in the sediment pile of Framvaren Fjord is not yet at steady-state due to the influx of brackish water into the fjord ~170 years ago. The identification of U(VI)-U(IV) reduction within the overlying water column and sediment pile provided crucial new constraints on the utilisation of δ238U as a palaeo-redox proxy and the reliable interpretation of δ238U compositions within sedimentary archives.