The fractionation between different stable isotopic species of the elements carbon and oxygen provides a powerful tool for investigation of various physical, chemical and biological processes in nature. Complex systoms involving the interaction of several such processes can be studied by combining isotope ratio measurements with other suitable data. The oceans constitute one of the most important of these systems; the distribution of carbon and oxygen in the sea is governed by a balance amongst physical, chemical and biological forces so that information on the isotopic abundances can be profitably used in oceanographic description. In this work, experimental techniques for determination of 6180 of seawater, and extraction and measurement of total dissolved inorganic carbon (EC02) from seawater and its 613C -assay are described. The assumptions involved in these methods are investigated and the associated errors discussed, in particular with regard to the mass spectrometry. The storage of oceanographic water samples between collection and analysis may allow biological activity to alter the dissolved oxygen/ carbon system. Inconsistencies in previously published work are pointed out and experiments are detailed which illustrate that temporal rCO2 and ý13C variations mirror the changes known to occur in the bacterial population of seawater stored without the addition of poison. Different types of storage vessel are investigated for preserving sample integrity and it is concluded that the importance of immediate poisoning and proper storage has been seriously underestimated in much previous work. Vater samples have been collected from areas of the Pacific and Antarctic Oceans, and analysed for 5180 of the water, >C02 and its 613C and, in some cases, pH and dissolved oxygen. The 6180 results have been taken in conjunction with conventional, hydrographic data to provide descriptions of the physical oceanography of the regions; water mass structure, and mixing has been elucidated and discussed. In the light of this, the ECO2,613C, pH and dissolved. oxygen measurements have been used to investigate the chemical and biological regimes. Comparison of the results of several different investigators in the Pacific has suggested that systematic errors may exist between directly and indirectly determined. ECO2 values. The regional oceanography deduced from this research has been in general accord with commonly accepted views derived from other studies.

This thesis investigates the role of submarine hydrothermal vents in the global marine Fe budget. While debate continues over the sources of dissolved Fe to the global deep-ocean dissolved Fe budget, it had been presumed, until recently, that all the Fe emitted from hydrothermal vents precipitates and sinks to the seafloor close to the vent source. However, in the open ocean, dissolved Fe exists at concentrations greater than the predicted solubility because of the presence of organically complexed Fe. If similar complexes were formed in the hydrothermal systems then there would be the potential for dissolved Fe export via hydrothermal plumes to the deep-ocean. To investigate the fate of hydrothemally sourced Fe, samples were collected from hightemperature hydrothermal vent-field plumes at 9°N on the East Pacific Rise and at 5°S on the Mid-Atlantic Ridge. The samples from the East Pacific Rise were analysed for Fe and dissolved and particulate organic carbon. Although hydrothermal systems are presumed to be inorganically dominated, elevated concentrations of dissolved organic carbon compared to background seawater were detected in near-field buoyant plumes and the concentration of organic carbon appeared to relate to the total Fe concentration, consistent with the presence of some organic-Fe interaction. Non-buoyant plume samples from the Mid-Atlantic Ridge were analysed for total dissolvable and dissolved Fe and Mn as well as speciation studies on a subset of the dissolved Fe samples using Competitive Ligand Exchange – Cathodic Stripping Voltammetry. The dissolved Fe concentrations in the dispersing plume were higher than predicted from dissolved Fe(II) oxidation rates alone. Further investigation into the speciation of the dissolved Fe revealed the presence of stable Fe-ligand complexes, similar to those detected in the open ocean, but with higher concentrations. If these Fe-ligand complexes were representative of all hydrothermal systems, submarine venting could potentially provide between 11 to 22% of the global deep-ocean dissolved Fe budget. Buoyant plume samples from the same vent site were analysed for total dissolvable and dissolved Fe and Mn as well as particulate Fe, Mn, P, V, Cu, Zn and the rare earth elements. Fe isotopes were also analysed in the particulate fraction, as a potential tool for tracing the biogeochemical cycle of Fe in the ocean. The forms of particulate Fe were elucidated using the particulate trace element data, enabling the isotope fractionation caused by Fe sulfide precipitation to be determined. A diagnostic isotope signature for a potential stabilised dissolved Fe fraction was predicted to be isotopically heavier than the original vent fluid, potentially enabling Fe inputs from hydrothermal vents to be traced throughout the ocean.

In Fe-limited or Fe-stressed waters Fe(II) cycling is an important feature of Fe biogeochemistry because processes favouring Fe(II) production or stability always expand the pool of Fe available to aquatic microorganisms. However almost nothing is known about Fe(II) organic speciation in natural waters and this limits our understanding of how this metal ion may interact with biological Fe uptake systems. Therefore I investigate how organic material affects Fe(II) biogeochemistry in rainwater, riverwater and in suspensions of aged glacial particulates. We compare Fe(II) concentrations (12-3600 nM) and, by adapting a ferrozine based reverse titration technique, ligand binding constants (logKFe(II) <5.5-11) in 5 temperate and sub-tropical river-estuary systems with varying dissolved organic carbon (DOC 200-1300 μM), labile dissolved Fe (Fe <0.2μm available to ferrozine after reduction with ascorbic acid 100 nM-20 μM) and pH (5.5-8.5). In riverwater we identify a natural class of humic Fe(II) ligands with weak Fe(II) binding constants (logKFe(II) <8) and concentrations in excess of Fe(II). Stronger ligands (logKFe(II) 11) were found only in the Itchen (Hampshire, UK) and may be associated with anthropogenic waste inputs. Similarly, the previously identified stabilisation of Fe(II) within North Carolina rainwater may be associated with anthropogenic emissions. Fe(II) concentrations in Wilmington (North Carolina, USA) rainwater have decreased from a high of 52 nM (and a 1:1 ratio of Fe(II):Fe(III)) in summer 2000 to a record low of 3 nM (and a 0.23:1 ratio of Fe(II):Fe(III)) in summer 2013 concurrently with improving air quality. The weak (logKFe(II) <8), terrestrially derived ligands we report in 4 temperate/sub-tropical river systems are likely to be found in most surface freshwater systems, but their impact in higher salinity coastal seawater will be minimal. Stronger Fe(II) ligands (logKFe(II) 11) associated with effluent emissions however may exert a stabilising effect on Fe(II) concentrations in the natural environment. This may result in localised increases in bioavailable Fe concentrations and have implications for aquatic ecosystems in industrialised areas.

No description available.

The marine carbonate chemistry system is the reactions and dynamic equilibria in seawater that involve dissolved carbon dioxide (CO2) and the deprotonated forms of carbonic acid, which are bicarbonate ions and carbonate ions. This system affects other marine biogeochemical cycles and ecosystems through its strong influence on seawater pH. Currently, a rapid chemical perturbation is being driven by the accumulation of anthropogenic CO2 in the ocean, which has taken up a significant fraction of the CO2 emitted to the atmosphere by fossil fuel burning, cement manufacture and land-use change since pre-industrial times. To predict the future consequences, it is first essential to be able to accurately measure the present state of the system, and to understand how it operates. To contribute to these goals, I have firstly made new developments to the measurement techniques for two key system variables, namely total alkalinity and the stable isotopic composition of dissolved inorganic carbon. Secondly, I have combined results from recent research cruises along a hydrographic transect in the Northeast Atlantic with historical datasets in order to quantify the rate at which anthropogenic CO2 has accumulated in the interior ocean over the past three decades, demonstrating that the regional accumulation rate is greater than the global average. I have used model output to assess uncertainties in these results that are caused by spatiotemporal heterogeneity in the distribution of observations. Thirdly, I have carried out a theoretical investigation into the influence of marine calcifying phytoplankton, like coccolithophores, on air-sea CO2 exchange. I have shown that these organisms can behave either as CO2 sources or sinks depending upon their relative rates of calcification and autotrophic production, their nutrient uptake stoichiometry, and local seawater conditions. Finally, I provide suggestions for developments to the work presented in this thesis that might help to overcome some challenges that are likely to face this field of research in the future.

In the first part of this thesis an overview is given of methods available for the analysis of nanomolar nitrate and phosphate in seawater before going on to describe in more detail a system built in our laboratory comprising liquid waveguide capillary cells connected to a conventional segmented-flow autoanalyser. This approach is suitable for routine field measurements of nitrate and phosphate and achieves detection limits of < 1 nM phosphate and nitrate. Investigations were conducted into interferences of silicate and arsenate with the analysis of nanomolar concentrations of SRP, the effect of sample filtration on the measurement of nanomolar nitrate + nitrite and SRP concentrations, and the stability of samples during storage are described. Arsenate interference scaled linearly with phosphate concentrations of up to 50 nM, resulting in an overestimation of SRP concentrations of 4.6 ± 1.4% for an assumed arsenate concentration of 20 nM. The interference effect of added Si(OH)4 on the measured SRP signal is small at the dissolved silicon concentrations typically found in oligotrophic waters. Filtration of surface seawater samples resulted in a decrease in concentration of 1.7 – 2.7 nM (±0.5 nM) SRP, and a small decrease in nitrate concentrations which was within the precision of the method (±0.6 nM). A stability study indicated that storage of very low concentration nutrient samples in the dark at 4°C for less than 24 h resulted in no statistically significant changes in nutrient concentrations. The second half of this thesis presents a dataset from aerosols collected at the Cape Verde Atmospheric Observatory (CVAO) between July 2007 and July 2008 and collected during a research cruise in the (sub-) tropical North Atlantic Ocean in January 2008. Total acid digestion followed by ICP-MS analysis reveals that the total elemental composition of the dust is close to average crustal composition and shows a high degree of consistency. Based on elemental composition data alone, dust collected on the cruise appears similar to dust collected at the CVAO. Zn and Pb are elevated above crustal values indicating an anthropogenic source, but show an association with periods of high mineral dust concentration. Ultrapure water leaches of dust samples combined with analysis for nutrients and trace metals show a picture of atmospheric concentrations of soluble trace metals and nutrients throughout the year in the study region. Estimated dry deposition fluxes for Fe, and inorganic N and P show a marked difference between summer and winter, with higher Fe and P deposition during winter when mineral dust concentrations are at their greatest, while N inputs are more constant throughout the year. Relative to Redfield ratios, atmospheric inputs are greatly enriched in Fe relative to N and P.

The ecophysiology of calcified macroalgal species of the genera Corallina (C. officinalis and C. caespitosa) and Ellisolandia (E. elongata) (Corallinales, Rhodophyta) was examined in intertidal rock pools of the NE Atlantic, to facilitate predictions of ocean-acidification and warming impacts on these ecosystem engineers. An initial phylogenetic study highlighted significant cryptic diversity within the genus Corallina, and demonstrated that C. officinalis is restricted predominantly to the North Atlantic, while the recently established C. caespitosa shows a cosmopolitan distribution. Three subsequent studies were performed across the NE Atlantic (Iceland to northern Spain) to examine (i) the production, respiration, calcification and growth of Corallina in relation to irradiance, water temperature, and carbonate chemistry; (ii) the photoacclimation and photoregulation strategies of Corallina and Ellisolandia; and (iii) the recent-past (1850 – 2010) and present-day skeletal mineralogy (Mg/Ca ratios) of Corallina and Ellisolandia and its relationship to sea surface temperature. Data demonstrated that species currently experience significant seasonal and tidal fluctuations in abiotic conditions that may be important when considering future responses to ocean-acidification and climate-change. Seasonality in production, calcification and growth were demonstrated, with decreasing growth observed with increasing latitude. Photoacclimation to allow maximal light utilisation during winter periods, and photoregulation via nonphotochemical quenching were highlighted as important in allowing Corallina and Ellisolandia to maintain maximal productivity while controlling for photo-stress. Seasonal cycles in skeletal Mg incorporation were demonstrated with strong relation to sea surface temperature, though no significant change in skeletal mineralogy was evident since pre-industrial times. Taken together, data indicated that Corallina and Ellisolandia have the potential to survive under future ocean-acidification and warming conditions, though loss of species at high latitudes and shifts in the relative abundances of species across the region is likely to be evident, with overall range contraction predicted for C. officinalis due to both warming and ocean-acidification impacts.

No description available.

Information on the concentration and distribution of dissolved methane (CH4), together with other geochemical tracers, in real time is of great value in detecting, monitoring, and understanding the functioning of hydrothermal plumes. Water column anomalies of light transmission, dissolved CH4, manganese (Mn), and iron (Fe) were located over segments 15 and 16 of the Central Indian Ridge (CIR 20ºS), in December 2006. Along segment 15, a hydrothermal plume was present at 19°33’S/65°50’E. The source might be located north of that position and dispersed along the western flank by NW-SE currents. Methane to manganese ratios suggest that methane is produced by magmatic processes. On Segment 16, evidence for 1 or 2 hydrothermal plumes were detected over a lava plain (18°20’S/65°18’E). These data suffered from uncertainties due to sampling issues, which demonstrate the need for a reliable in-situ methane sensing technology. Current in-situ methane sensing technology is based on gas partitioning across gas permeable membranes, which are poorly characterised and variable in terms of permeability and environmental pressures. Two optical techniques were laboratory tested for the measurement of dissolved methane; Near Infrared Fibre-optic Evanescent Wave Spectroscopy (FEWS) and Surface Plasmon Resonance (SPR). No detection (at the µM level) was possible with FEWS, but the second technique using SPR sensors associated with a methane specific binding chemically showed great promise. A limit of detection of 0.2 nM and a linear concentration range from 1 to 300 nM was demonstrated, under a range of temperature and salinity. In-situ deployments confirmed the suitability of the method for in-situ measurements. Values given by the sensor correlated well with the concentrations measured by traditional techniques. Future work is needed to decrease instrumental noise and to reduce the response time, and associated hysteresis effect.

This study investigates the preserved fluid contents of different generations of mineralised and unmineralised vein minerals from multiple different deposits across the Zambian Copperbelt to define the fluid physicochemical characteristics of the fluids and investigate the importance of fluid processes through the basin history. An investigation into REE and trace element and sulphur isotope signatures of sulphides was conducted to investigate sulphide formation mechanisms. Re-Os dating of sulphides from the Domes Region to provide temporal constraints on mineralisation processes in this area. A textural study on uranium mineralisation at the Lumwana deposit was undertaken to understand the relationships between uranium and sulphide mineralisation. Variable Cl/Br of fluids indicates they were derived from evaporation of seawater, deposition of evaporite sequences and the subsequent dissolution of these evaporite sequences, with bittern brines dominant earlier in the basin history, and dissolution of halite more important during compression and orogenesis. Cation contents of fluids record the development of alteration assemblages caused by the movement of these brines at temperatures typically >200C and salinities >30 wt% NaCl equiv. Stable isotope data records the role of organic reductants and fluid-host rock equilibration through basin history. Rare earth element signatures of sulphides further record the importance of the development of the alteration assemblage on the changing chemistry of the hydrothermal fluids, whilst trace element concentrations suggest the metal budget in the fluids reflects source rock variation. Sulphur isotope data suggests the main mechanism of sulphide formation was thermochemical sulphate reduction of seawater sulphate, consistent with the temperatures of the fluids reported here. Re-Os dating records the importance of orogenesis on the timing of mineralisation in the Domes Region, and are consistent with published Re-Os dating of sulphides and U-Pb dating of uranium minerals. Textural relationships indicate initial uranium mineralisation at the Lumwana deposit pre dates sulphide mineralisation, and final movement on the shear zone, but has undergone a remobilisation event later in the deposit history.