Waves and swell incident on a continuous cover of sea ice can become coupled to the ice and propagate as flexural-gravity waves. These ice-coupled waves are determined by a dispersion equation and attenuation coefficient which depend critically on the material properties of the sea ice. Two models of sea ice are proposed: - a linearly viscoelastic thin plate with uniform properties throughout; and a thermorheologically simple, viscoelastic thin plate whose properties vary through the ice sheet due to the temperature gradient across its thickness. Using the former model, an approximate solution consisting of the superposition of three ice-coupled waves is found for deep water waves impinging on ice at normal incidence. The characteristic properties of these waves are discussed in detail. The latter, more complicated model is developed and solved assuming that the viscoelastic Poisson's ratio is constant and the viscoelastic Young's modulus at a particular reference temperature satisfies some sea ice model proposed in the literature. Flexural-gravity wave solutions are discussed with emphasis laid on their temperature dependence and the effects of rotatory inertia and transverse shear. Obliquely incident waves are also considered. It is shown that the presence of significant attenuation within the ice cover leads to inhomogeneous flexural-gravity waves near the ice edge but the degree of inhomogeneity and angular spread decreases with distance. The existence of a critical angle of incidence enhances this effect considerably. Experiments carried out using strain measuring instruments on fast ice in Newfoundland are discussed and the data obtained are compared with theoretical predictions. It is shown that the direction and magnitude of wave components at some locations may be found using a rosette of three instruments but that large errors may occur in the subsequent data processing. A method avoiding these errors is proposed and implemented. Additional observations of ice-coupled waves recorded at Strathcona Sound are reported and possible mechanisms for their generation are suggested.

Early work relating to the Ilfracombe Beds of North Devon is reviewed and it is shown that many problems remain unsolved, for example, the age of the beds, the stratigraphical succession, and the structure. The coastline between Combe Martin and Ilfracombe has been mapped on the scale of 50 ins = 1 mile, and it is found that there are only two important limestone developments, each about 30 feet thick; the lower is the Jenny Start Limestone, which is of Upper Givetian age, and the higher is the David's Stone Limestone of Lower Frasnian age. There are two other thin, but mappable, limestone horizons, each between 2-5 feet thick; these are the Rillage Limestone (about 200 feet below the Jenny Start Limestone), and the Combe Martin Beach Limestone (lying between the two main ones). N.W.-S.E. dextral tear faults, probably of Tertiary age, and E-W normal and reverse strike faults, probably related to an early orogeny, displace the strata. The faults, combined with the very intense folding, have resulted in numerous repetitions and occasional omissions in the local successions. The full succession between the Hangman Beds and the top of the Ilfracombe Beds is summarized in tabular form. The second part of this thesis is a detailed study of the coral fauna, including a complete morphological study and revision of the genus Metriophyllum. 51 species of corals belonging to 23 genera are figured and described. Of these 20 are new species and many are recorded for the first time in North Devon. The coral fauna establishes the age of the main limestones with moderate certainty, but the intervening beds are unfossiliferous and it is, therefore, impossible to draw absolute boundaries.

Problems with the identification of tubular fossils from ancient hydrothermal vent and cold seep deposits have hindered understanding of the evolutionary history of vent and seep communities. This thesis aims to (1) improve knowledge of lesser-studied tubicolous annelids occupying vents and seeps, (2) study the diversity of tubes at vents and seeps, (3) investigate the fossilisation of tubes within modern vents and seeps, (4) better interpret the fossil record, and (5) provide insights into the palaeoecology these environments. Results presented here on investigations of Sclerolinum contortum, a species belonging to a little- studied genus of the major vent/seep dwelling annelid family Siboglinidae, demonstrate that it exhibits tube morphological plasticity, wide habitat preferences and a global distribution spanning the Arctic to the Southern Ocean. These results also suggest that this species has dispersed throughout this range using chemosynthetic habitats as stepping stones. A detailed investigation of the full mineralisation process of Alvinella (Alvinellidae) tubes at modern hydrothermal vents shows that these tubes are fossilised by pyrite and silica that template organic tube layers, and that microorganisms living on tube surfaces are also exceptionally well-preserved alongside the tubes. No known ancient vent tube fossils resemble mineralised Alvinella tubes. A major morphological and compositional comparison of both modern and fossil tubes from vents and seeps revealed that two fossil tube types from the Mesozoic were likely made by vestimentiferans (Siboglinidae), suggesting that this major vent and seep annelid lineage has a longer history within vents and seeps than proposed by molecular clock age estimates. This analysis also demonstrates the need for greater caution in assigning affinities to fossil vent and seep tubes. Finally, this thesis reports the remarkable preservation of filamentous microorganisms on the walls of Silurian vent tube fossils, giving the first insights into ecological associations between microbes and metazoans within the oldest known hydrothermal vent community.

Periodic pressure fluctuations beneath ocean waves deform the seabed beneath them and can generate seismic waves that propagate around the globe. These pressure fluctuations are recorded on seafloor pressure gauges and the associated ground displacements on seismometers, where they contribute to ambient seismic noise. The signals offer an opportunity to study or monitor ocean waves that are traditionally difficult to measure because of their low wave heights (deep water infragravity waves) or their remote locations (deep water swell waves). However, the link between ocean waves and the oscillations of the pressure and seismic wavefields has remained unclear. The aim of this thesis was to increase our understanding of ocean sources of ambient seismic noise, including their location, how well they can be located, and their relationship with ocean wave parameters. Using cross-correlated pairs of seafloor pressure records, infragravity waves offshore California/Oregon were found to originate mostly from local coastlines during northern-hemisphere winter and from the south during summer. A first attempt to estimate the coastal reflection coefficient of remote arrivals was made and found to be 0.49-0.74, which has implications for infragravity energy in the deep ocean and may be important for models in which infragravity waves are propagated across ocean basins. P-wave sources in the North Pacific during winter were located using seismometers in California and found to be concentrated around 40-50◦N 160-180◦E. Observed source locations were within 10◦ of the modelled source locations. Significant wave height in the deep ocean was estimated from P-waves and correlated with modelled wave height with a correlation coefficient of 0.63. Previous work only attempted to estimate coastal wave heights. Combining additional records from Japan and Europe improved source location, including imaging of multiple sources. Accuracy in source location and amplitude estimation are essential if microseisms are to be used to monitor wave activity in the deep ocean.

The perceived supply risk for essential materials used in the development of green energy and other state-of-the art technologies creates the need for investigation of new sources for these raw materials. Many of these raw materials are characterized as “critical” given supply risks posed by geographic location, the economic and political stability of producing countries, potential substitution and opportunities for recycling [European Commission, 2014]. At present, 20 raw materials are listed by the EU as critical and this inventory is likely to grow in the coming years as the world population increases, driven by the development of India, China, Africa, Brazil and others. Among these critical elements, the rare earth elements and yttrium (REY) form a group of 15 metals essential for the development of wind turbines, cell phones and batteries among other applications and their production has been under Chinese domination for the last three decades. More than 95 % of the consumed REY worldwide originated in China during the last thirty years, a monopole that reflects economical constrains rather than the unequal distribution of REY resources across the world. Indeed, important proven reserves are known outside China though their extraction is expensive and energy consuming. In addition, most REY-rich deposits possess important concentrations of actinides (U and Th) problematic for waste disposal. This study therefore investigates the potential recovery of REY from umbers, metalliferous sediments of the Troodos massif in Cyprus, as an alternative to the dominant magmatic-related REY deposits. Field evidence and geochemical characterisation of umbers show strong similarities with high-temperature plume fall-out deposits observed in most mid-oceanic ridge settings. Umbers constitute fine-grained brown Fe-Mn-rich mudstones with an amorphous oxyhydroxides dominated mineralogy and total rare earth oxide contents of ≈0.05 wt. %. REY fractionation trends show excellent comparison with signatures of hydrothermal particles settling around active vents. The umbers display a negative Ce anomaly in a convex upward REE trend when normalized to chondrite, characteristic of a hydrothermal signal overprinted by seawater. From an economic perspective, although the REE content is low, the absence of mineralogical control on the distribution of these elements in umbers and the extremely low radioactive content (Th + U < 5 ppm) makes their potential extraction attractive. A protocol for the leaching of umbers is presented testing a variety of lixiviants used in the REY extractive industry. Results show a strong mobilisation of the lanthanides in the solution in comparison with non-targeted elements. Most importantly, the results presented highlight that 80 to 90 % of the initial REY content of umbers is leached out using weak acid concentration in a matter of hours at low temperature. Fractionation along the REY series during leaching usually favours the release of the middle and light REE with a decreasing trend towards the heavy REE, except for Yttrium. Ce recovery is minimal as a result of its tetravalent oxidation state allowing formation of acid-resistant Ce oxides. Furthermore, a process of selective precipitation is presented for the purification of the leach solution and extraction of a solid REY phase using ammonium oxalate as a complexing and chelating agent. Precipitation experiments show the precipitation efficiency is a function of pH, between pH values ranging from 0.7 to 3.2, with more than 96 % of REY precipitated at pH > 1.1. Purity of the precipitate is adjusted using precise pH buffering to avoid Ca-oxalate formation as the major impurity. Indeed, mass balance calculations and direct EDS measurement of the oxalate precipitate by SEM show maximal purity at pH 1.1 (66 – 94 % REY) while increasing Ca precipitation decrease purity below 10 % at pH > 1.5. The fractionation observed along the lanthanide series during the precipitation experiments was successfully reproduced via numerical modelling using PHREEQC software. REE distribution within the precipitate therefore reflects the interplay of aqueous and solid REY-oxalate complexes stability constants as well as incorporation of REY within the structure of co-precipitating Ca- and Na-oxalates. This study demonstrates the feasibility of extracting efficiently REY from Fe-Mn oxide-rich metalliferous sediments. These deposits constitute interesting alternatives to high-grade deposits and their processing for REY production could be valuated as a by-product of pigment production. Alternatively, the process presented here could be applied to other oxide-based formations including marine ferromanganese deposits, or industrial wastes containing comparable high-tech metals concentration and enrichment process.

No description available.

The thesis here presented is based on three main parts. The first part of the thes is based on modelling the water circulation of the Clyde Sea, in order to understand the dynamics of the dispersion of the Neprophs larvae. Previous researches in this area of Scotland highlighted the importance of the temperature- and saline-driven circulation in the Clyde Sea. However, few researches were focused on the surge dynamics, that are governing the dynamics of the water level in winter. A three-dimensional finite-volume model was used to simulate the surge propagation, while a historical re-analysis was applied to understand the pattern and the propagation of the surge wave in the Clyde. The results highlighted that the largest storms that hit the Clyde in the past 30 years were mostly generated in the North Atlantic. Most interestingly, the results also suggest that severe surges are not only caused by extreme surge events, but also by the coupling of spring high tides with moderate surges. In the second part the coupled dynamics of waves, tides and wind-driven circulation in the east coast of Scotland are studied. Wave-Current Interactions (WCI) are particularly relevant close to the coastline, where the effect of the spectral dispersion and wave breaking are more important and where the currents are stronger. The results showed that the coupling of strong currents with large waves travelling inopposite direction could enhance in east coast of Scotland the significant wave height (Hs) up to 3 m, threatening potentially infrastructures and ships near the coastline. The last part of the thesis was dedicated to an experimental study of rogue waves in crossing sea. Crossing sea is one of the most common state in world seas, and occurs when a wind-generated wave train interact with another train of waves, that can be swell waves or another wind-generated wave train, maybe caused by a rapidly turning wind direction. Some numerical studies showed that this interaction can lead to the mechanism of modulation instability and, consequently, to the formation of rogue waves. A water tank experiment was carried out to confirm this theory. The results shows that the angle of the interaction is a fundamental variable that can decrease or increase the instability of the wave train. However, most interestingly, in the sameconditions, the monochromatic sea state was showing a larger number of rogue waves than the crossing sea.

Microbial biogeography is being increasingly more studied, both in terms of genetic divisions and 'ecotype' variation. This thesis investigates the regional (100-1000s km) and local (10s m) distribution and diversity of microorganisms around hydrothermal vents and seamounts at the East Scotia Ridge and the Southwest Indian Ridge. Microbial communities were characterized using Illumina dye sequencing to de- termine taxon richness and diversity and flow cytometry to obtain cell counts. In addition I investigated the physicochemical environment (nutrients, organic carbon, salinity and temperature) in which the microorganisms persist. Typical deep-sea microorganisms were abundant at vents and below the euphotic zone on the seamounts. The surface layer of the seamounts contained typical open-ocean photoautotrophic organisms. Microbial communities were correlated to or- ganic carbon on both hydrothermal vents and seamounts. With microorganisms possi- bly having a large influence on carbon sequestration into the deep-sea from hydrothermal vents. On a local scale the hydrothermal vents had a relatively higher abundance of chemosynthetic Epsilonproteobacteria and the Gammaproteobacteria family SUP05, which were closely correlated to the redox potential in the vent effluent. This was prominent in both plumes from the East Scotia Ridge and the Southwest Indian Ridge, which also had distinct microbial community structures. Across the Southwest Indian Ocean the microbial communities were firstly segregated by depth. However, on a regional scale their physical environment primarily divided the microbial communities into three biological regimes the sub-tropical, con- vergence zone and the sub-Antarctic. The microbial community structure and biogeography is influenced by steep environmental gradients, displaying a distance-decay relationship between sampling locations. With environmental conditions persisting at different scales, from local (10s m) around the hydrothermal vents to regional (100-1000 km) between the seamounts, driving the microbial community patterns. The same horizontal and vertical patterns for microorganisms and metazoans points to fundamental differences throughout all parts of the ecosystem/food web.

Stable isotope ratio analyses offer a unique opportunity to obtain information about ecosystem dynamics, patterns and processes. The anthropogenic contribution to the global atmospheric CO2 rise through fossil fuel combustion, deforestation and other related human activities has changed the stable carbon isotope composition (δ13C) of the atmosphere over the past 200 years. Changes in the carbon isotopic patterns of terrestrial biosphere, lithosphere and oceans are also expected. The global ocean has been acting as a net sink for CO2 emissions and although it moderates the climate, it is currently in a critical state of health. While the physico-chemical consequences (ocean acidification) of the increasing CO2 uptake by the ocean are fairly well known, the perturbation to marine ecosystems and the related effects on biota still entail large uncertainties. This thesis investigates the feasibility of using measurements of δ13C of seawater samples to increase our understanding of the biogeochemical responses of marine ecosystems to human CO2 perturbation. The isotopic composition of all the individual inorganic and organic carbon species from three long term mesocosm experiments (Sweden 2013, Gran Canaria 2014, Norway 2015) was determined. To have accurate and precise isotopic measurements, mass spectrometry instrument calibrations and method validation procedures were performed. Universal and inter-laboratory accuracy of the analysis was assessed by running standard materials provided by the International Atomic Energy Agency (IAEA, Vienna) and by the Scottish Universities Environmental Research Centre (SUERC) stable isotope laboratory, respectively. Precision and internal consistency was assessed from isotopic measurements of seawater reference materials from A.G. Dickson and D. Hansell for dissolved inorganic and organic carbon, respectively. A novel accurate, precise and rapid method, coupling a Shimadzu 5000A total organic carbon (TOC) analyser to an isotope ratio mass spectrometer (Thermo Scientific Delta V Advantage IRMS), was successfully developed in order to determine the δ13C of dissolved organic carbon in seawater samples which, due to analytical challenges, is currently not a widespread technique. The combination of isotopic and non-isotopic measurements from mesocosm experiments provided a holistic view of the biogeochemical mechanisms that affect carbon dynamics under different CO2 conditions (up to 2000 ppm). A clear CO2 response was detected in the isotopic datasets, but increased CO2 levels had only a subtle effect on the concentrations of the dissolved and particulate organic carbon pools. Distinctive δ13C signatures of the particulate carbon pool both in the water column and the sediments were detectable for the different CO2 treatments and they were strongly correlated with the δ13C signatures of the inorganic carbon but not with the δ13C of the dissolved organic pools. Phytoplankton fractionation was positively affected by high CO2 either because of the higher CO2 availability or because of a shift in phytoplankton community composition, however, phytoplankton bloom intensity and evolution was independent of CO2 concentrations and higher CO2 levels had no significant effect on inorganic nutrient uptake or carbon production/consumption. Overall this study proved the stable carbon isotope approach to be an effective tool for the assessment of the major biogeochemical interactions among individual compartments within the marine system opening the door to new interpretations for past, present and future changes of the global carbon cycle.

Observation of ice-stream and ice-shelf flow has revealed interaction with the ocean tide, in some cases far upstream of the grounding line. Ice-stream velocities can be greatly affected by local ocean tides at a variety of timescales but of particular interest is a fortnightly modulation in flow that has been observed on the Rutford Ice Stream. This frequency is absent in the forcing, far stronger than the dominant tidal constituent signals and has been observed far inland. Understanding what mechanism can produce this signal can provide a window into mechanisms at the base of the ice stream that affect its flow. A consequence of this nonlinearity is that the mean flow of the Rutford Ice Stream is increased in comparison to a situation where no tidal forcing is present, implying that changes in tides could alter the long term flow of ice from Antarctica. Since local tides are sensitive to bathymetry which can change drastically due to changes in ice-shelf geometry and grounding line position, a feedback exists between tides and the ice sheet whereby changes in the flow of ice could alter tides which in turn would change the ice velocity. This thesis presents results from both ice-stream/shelf and tidal models to investigate these two mechanisms. Firstly a nonlinear 3D viscoelastic full-Stokes finite element model explores possible sources of nonlinearity that can produce fortnightly modulation in ice-stream flow far upstream of the grounding line. This model looks at stress transmission, grounding zone flexure, hydrology and grounding line migration and the conclusion is that tidally-induced changes in the subglacial water pressure play the largest role in producing this effect. Then, the OTIS tidal model is used to demonstrate that reduction in the thickness and/or extent of ice-shelves, in particular the Filchner-Ronne ice shelf, could greatly impact both tidal amplitudes and dissipation rates in the region. The research in this thesis clearly shows the importance of tidal interactions with the Antarctic ice sheet and that these are not limited to short temporal timescales.