No description available.

No description available.

The geochemical evolution of subducting oceanic crust (including both mafic and sedimentary components) has important implications for our understanding of the origin of mantle heterogeneity and the geochemistry of volcanic-arc basalts (VAB). However, knowledge of the actual geochemical evolution of the subducting oceanic crust is largely based on indirect evidence, such as the composition of VAB and our understanding of the geochemical properties of key trace elements. Certain kinds of blueschist belt are thought to consist of exhumed portions of subducted oceanic crust. It is therefore possible to derive direct evidence of the geochemical evolution of subducted oceanic crust by studying the metamorphic rocks (i.e. greenschists, blueschists and eclogites) of blueschist belts. The bulk rock chemical analysis of greenschists, blueschists and eclogites from the western Tian Shan blueschist belt, Xingjiang Province, People's Republic of China (PRC), has enabled the geochemical effects of subduction zone metamorphism to be determined. Some eclogites from Tian Shan are thought to have undergone "ultra-high pressure metamorphism" (UHPM) at pressures >2.5GPa, and to have passed through the dehydration reactions associated with subduction zone processes. However, it is shown in this thesis that the eclogites have not been subjected to UHPM. Novel approaches have been developed to interpret the geochemistry of metamorphic rocks, including a new method for identifying "immobile elements" and a method for identifying fresh (i.e. un-metamorphosed) chemically-similar samples from large published datasets. Protoliths of meta-basaltic rocks from Tian Shan are shown largely to originate from seamounts, continental volcanic arcs and basalts associated with attenuation of continental crust. Comparison of the chemistry of the metamorphosed rocks with likely protolith compositions has not revealed any systematic changes associated with subduction zone metamorphism. To augment the information obtained from the Tian Shan rocks, samples were also analysed from the blueschist belts of the Qilian Mountains, Gansu/Qinhai Province, PRC. Protoliths of the meta-basaltic rocks were shown to be back-arc basin basalts and basalts associated with back-arc rifting. Comparison of the geochemistry between likely protolith and metamorphic rock compositions revealed no systematic changes. This indicates that no geochemical effects associated with subduction zone metamorphism. It is shown, by comparing the Loss on Ignition (LOI) of carbonate-bearing meta-basaltic rocks with the H2O content of H2O-saturated MORB, that the blueschists and eclogites have not significantly dehydrated. Thus, no mobile phase with the capacity to transport fluid-mobile elements was present during subduction zone metamorphism. The identification of no chemical changes associated with subduction zone metamorphism in this work is consistent with similar investigations that have recently been published. Ostensibly, such findings have important implications for our understanding of the origin of mantle heterogeneity and VAB compositions. However, the assumption that rocks from blueschist belts, such as those of the western Tian Shan blueschist belt, are analogous to actual deeply subducted oceanic crust is shown not to be valid. This may particularly be true with respect to differences between initial H2O content of blueschists and eclogites compared with actual subducting altered oceanic crust. Consequently, any interpretations of the geochemical evolution of such rocks should not be extended to cover the evolution of actual subducting oceanic crust.

In north-west Zambia, Kalahari Sand of aeolian origin forms a cover of varying thickness on residual latosols. Sedimentological and mineralogical characteristics indicate that the Sand is similar to Kalahari Sand occurring elsewhere in southern Africa, and to aeolian sands in general. Geochemical sampling at 9 inch depth on residual soil defines Cu anomalies overlying Cu mineralisation in bedrock. At two localities, the surface Gu anomalies have been extended into Kalahari Sand overburden by the use of the silt fraction for analyses. In both areas, Cu anomalous Sand overlies poorly mineralised bedrock. The Cu, Pe, Mn and Zn content of Kalahari Sand is considerably lower than in soil. Metal patterns in the Sand appear to be epigenetic and reflect variation in bedrock content. Lateritic horizons are widely developed in the residual soils, Kalahari Sand and davnbo sediment, and result in concentrations of Pe, Gu, iin, Zn and other trace elements. In the absence of laterite development, the distribution of metals in Kalahari Sand and some soil profiles displays an upward increase in content. This, together with the epigenetic anomalous patterns, is believed to result from seasonal upward migration of metal-bearing moisture during the dry season. Analyses of grain size fractions of residual soil, Kalahari Sand and dambo sediment indicate that metal content increases with decreasing grain size. Maximum Cu contrast occurs in the silt fraction. Metal values in humus are generally similar to overburden values but greater than the vegetation content. Metal values of termitaria are similar to A horizon values of the overburden. A geobotanical survey reveals no significant variation in tree frequency related to Cu anomalous overburden. Randomly selected common tree species reveal a Gu correlation with surrounding Kalahari Sand values, with maximum content and contrast in bark.

Ice core records show that the atmospheric concentration of methane (CH₄) during the Last Glacial Maximum (LGM) was 40-50% lower than during the preindustrial Holocene. To understand this natural variation it is important to know how the sources and sinks of CH₄ change over time. Natural wetlands were the single largest contributor of CH₄ to the atmosphere in glacial times, yet models used to estimate their behaviour and CH₄ flux are largely based around relationships derived under modem day conditions. This thesis responds to this issue by exposing wetland mesocosms with contrasting nutrient availability, to the atmospheric concentration of carbon dioxide (CO₂) present at the LGM for 2 years. At the end of this experiment, total CH₄ flux was suppressed by an average of 29% in the nutrient rich fen (P < 0.05). In contrast, the nutrient poor bog showed no response to the treatment (P > 0.05). Further exploring the effects of CO₂ starvation showed that the fen ecosystem exhibited notable reductions in dissolved organic carbon, dissolved CH₄ and a change in the response of CH₄ flux to changing temperature, variables and relationships which all remained unchanged in the bog. The contrasting response of the two ecosystems to CO₂ starvation could be largely explained by differences in nutrient status, species composition and dominant CH₄ production pathways. In particular, it is hypothesised that bog plants under LGM CO₂ concentrations supplemented photosynthesis through the use of subsurface derived CO₂, thus counteracting the treatment effect. The results from this thesis suggest that the CH₄ source strength of late-glacial and early Holocene wetlands may currently be over-estimated because fen ecosystems are a far smaller CH₄ source under low atmospheric [CO₂] than they are today. Furthermore, the results provide new insights into the role of glacial atmospheric CO₂ concentrations in influencing CH₄ emissions from terrestrial ecosystems and provide empirical evidence for a connection between glacial-interglacial changes in atmospheric CH₄ and CO₂ concentrations observed in ice cores.

In this thesis a wide range of analytical techniques were used to characterise several petrophysical properties of Berea sandstone, including the mineral distribution at the pore surface and its pore structure, both playing a crucial role in determining its response to low salinity enhanced oil recovery (EOR) investigations. In addition, the role of different cations in affecting the wettability state of pure quartz and Berea sandstone was experimentally investigated in order to gain an insight on the behaviour of sandstone reservoirs during low salinity waterflooding EOR. Results from the multi-technique, multi-scale characterisation of Berea indicate that the mineralogy exposed to the pore spaces is highly heterogeneous across different length scales, going down to the often-neglected nanoscale were significant amounts of phases identified as grain coatings. In addition, analysis of the porosity and pore-connectivity also requires a multi-length approach for its full characterisation to be realised. Both aspects are crucial to understand the role of mineral surface chemistry in determining oil/water and oil/minerals interactions in both experiments and field conditions. Investigations on wettability alteration using environmental scanning electron microscopy (ESEM) and contact angle measurements on ideal quartz surfaces showed that reduced salinity leads to a more water-wet state. These measurements were complemented with atomic force microscopy adhesion measurements on quartz surfaces, the results giving further insight into the role of nano-scale roughness on quartz surfaces in wettability alteration by increasing the amount of oil retained on the surface. Finally, similar wettability alteration experiments were performed on Berea sections. The effect of brine was consistent, reproducible and reversible and again showed a low salinity effect, i.e. a change to more water-wet conditions with lower salinity. The results also demonstrate that quartz surfaces always contributes at least in part to the low salinity effect, decreasing oil wettability when salinity is low. In addition, we demonstrate that the ESEM can be an essential tool in studying the wettability alteration of rocks and minerals.

The development of a rapid analytical technique for the determination of the platinumgroup elements (PGE) and gold in geological samples is described. The technique is based on selective aqua regia acid leach followed by a selective extraction (using diphenylthiourea and 1,2-dichloroethane) to separate the PGE and Au as a group from concomitant matrix elements. Organic extracts and aqueous raffinates were analysed by graphite furnace atomic absorption spectrometry, for which a comprehensive assessment of matrix interference effects was undertaken. Direct analysis of the aqua regia acid leachates by inductively coupled plasma-mass spectrometry (ICP-MS) was also evaluated. A number of other techniques were used to evaluate recoveries following acid attack, including x-ray fluorescence spectrometry and instrumental neutron activation analysis of solid residues and beta-autoradiography of thin sections to characterise PGE mineral solubility. The results of these investigations identified optimum conditions for aqua regia extraction (20 ml of normal (3: 1) aqua regia for a 10 g sample, stirring for two hours at room temperature) and that quantitative recoveries can be expected for Au and semiquantitative for Pd with lower, but variable recoveries for Pt, Rh, Ru and Os and very low recovery of Ir. The solvent extraction procedure was effective in selectively extracting Au and Rh but was not quantitative in the extraction of Pt, Pd and Ru, the extraction of which appeared to be influenced by the sample matrix. Further studies on the solubility in aqua regia of individual PGE minerals indicated that the main control of aqua regia extraction efficiency was sample mineralogy. The aqua regia leach procedure was applied to a range of samples that had been independently analysed by NiS fire assay as well as appropriate reference materials. This extraction procedure, with direct analysis of leachates by ICP-MS, was also used as the primary technique for characterising the homogeneity of two new chromitite reference materials, CHR-Pt+ and CHR-Bkg.

The volcanic rocks of the Afro-Arabian volcanic province chart the nature and evolution of the mantle plume source responsible for a major continental flood basalt event and the manner in which it interacted with the lithosphere over the course of continental separation. The flood basalts of the Ethiopian Plateau constitute the larger part of this province and are associated with melting in the head of the Afar Plume. High-Ti picrites from the eastern part of the NW Plateau are among the earliest uncontaminated outpourings of flood volcanism; their compositions are therefore considered to be close to that of the primary melt. These are exposed in a stratigraphically coherent sequence with variably porphyritic high-Ti olivine-rich basalts and ankaramites, which together provide a continuous record of volcanism reflecting a transition from high- to low-flux magmatism. Phase-relations differ fromthose in other tholeiitic flood basalts in that clinopyroxene and magnetite crystallisebefore plagioclase. This is typical of ferropicrites and reflects a high-Fe-Ti-, and low AI-source. The olivine phenocrysts exhibit a range of morphologies and compositions which reflect fractionation and magma-mixing at varying depths prior to eruption, and the introduction of water into the magmas at shallow levels is evident from pervasive iddingsitization. The lavas have similar major element chemistry to MORB but show higher concentrations of incompatible elements inherited from a HIMU-type lithospheric component. and enrichments in LREE relative to HREE which reflect the presence of residual garnet in the source. The picrites and ankaramites have lower AI₂O₃ and higher K₂O, P₂O₅, TiO₂ and Fe₂O₃^(tot) contents than any reported lavas with equivalent MgO content from other flood basalt provinces. These least evolved lavas were derived from oxidised (QFM+1), deep-seated magmas generated by different degrees of partial melting of the primary melt, whereas the olivine basalt magmas evolved by crystal fractionation at a shallow level from a melt similar in composition to the ankaramites. The MgO content of the primary melt calculated from the most Mg-rich olivines (Fo₉₀) is between 17 and 18 %, and it is likely that this ponded at the base of the crust where it evolved at pressures near to 1.2 GPa. Primary melt fractions of 3 - 6 % were generated in the plume at pressures of 4 - 5 GPa and temperatures in excess of 1600 °C, implying that melting occurred beneath thick (120- 150 km) lithosphere prior to extension. Sub-chondritic initial ¹⁸⁷Os/¹⁸⁸Os ratios (0.125 - 0.126) indicate derivation from a depleted mantle source, and high ³He/⁴He (18.6 R_a) and solar-like Ne isotopic ratios imply an origin in undegassed mantle.

To date, research on the fate of methane in marine settings has mainly focused on anaerobic microbial processes. An alternative fate for methane is aerobic methane oxidation (AMO) by methanotrophic bacteria which takes place in aerobic surface sediments and the overlying water column. Tracing methanotroph activity in past environments can be achieved via analysis of a distinctive suite of biomarkers called bacteriohopanepolyols (BHPs). BHPs are membrane lipids produced by many prokaryotes comprising a pentacyclic triterpenoid structure with an extended polyfunctionalised side chain. Although, there is much debate about the role of BHPs, studies suggest they regulate cell membrane fluidity, however, the factors controlling their expression are poorly constrained. They have a wide range of structural variation which varies between bacterial phyla and species. The major BHPs produced by methanotrophs are collectively known as the 35-aminoBHPs, most commonly including 35-aminobacteriohopane-32,33,34-triol (aminotriol), 35-aminobacteriohopane-31,32,33,34-tetrol (aminotetrol) and 35-aminobacteriohopane-30,31,32,33,34-pentol (aminopentol), with aminopentol seen as a diagnostic marker for Type I methanotrophs from the phylum Gammaproteobacteria. The changes in methanotroph community composition in estuarine sediments under a range of environmental perturbations and the effect this had on BHP composition, namely the 35-aminoBHPs, was assessed. Aerobic microcosms inoculated with River Tyne (UK) estuarine sediment with a 5% methane amended headspace (unless otherwise stated), were subjected to a range of environmental perturbations; methane concentration (0.1-5%), temperature (4-60°C), pH (4-9) and salinity (1-150 g/L NaCl). Methane oxidation rates were monitored and methanotroph diversity was determined by targeting the particulate methane monooxygenase gene (pmoA). Methane oxidation was observed between 4 and 50°C, at all tested pH values and up to salinities of 70 g/L; however, methanotroph community composition varied with temperature, pH and salinity and these changes were reflected in the 35-aminoBHP signatures quantified by LC-MS analysis. For example, aminopentol was not enriched at pH 9 when the unusual Type I Methylomicrobium spp. were dominant, whilst the maximal production of C-3 methylated aminopentol was witnessed at 50°C when a Methylocaldum sp. was enriched. The hpnR gene, required for the methylation of BHPs at the C-3 position, was also identified in sediments at the aforementioned temperature. Novel iv compounds, identified after the analysis of six previously untested Type I marine methanotrophs within this study, were also found in microcosm sediments in varying abundances. The effect that of growth stage on 35-aminoBHP abundance was determined by analysing aerobic microcosms inoculated with River Tyne estuarine sediment over a 28 day period at times before and after methane oxidation. It revealed the continued production of aminopentol at mesophilic temperatures after methane oxidation was complete. This may have implications for the interpretation of the sedimentary record where aminopentol witnessed in marine settings may not represent periods of significant methane oxidation but rather a response to methane limiting conditions. Anaerobic producers of BHPs were investigated and the preservation and/or degradation of individual compounds was assessed in long-term studies. Microcosms inoculated with anoxic River Tyne estuarine sediment were subjected to sulphate-reducing and methanogenic conditions over a period of 706 and 665 days respectively. Changes in BHP composition over time were quantified by LC-MS with compounds including bacteriohopane-32,33,34,35-tetrol (BHT) and adenosylhopane found to be more resistant to degradation over the course of the study compared to bacteriohopane-32,33,34,35-tetrol cyclitol ether (BHT cyclitol ether). This indicates that some compounds are more resistant to degradation over time compared with others.

Methane is a climatically active gas and is a potential source of rapid global warming. Future climate change scenarios predict increased global temperatures, which, could destabilise large reservoirs of organic carbon currently locked in sediments and soils and further accelerate global warming. Comparable to the Arctic, the tropics store a large proportion of sedimentary carbon that is potentially highly vulnerable to climate change. However, the significance of tropical methane sinks in modifying methane emissions during past climate warm periods remains unresolved. This study focussed on determining the importance of aerobic methane oxidation (AMO) within the Congo during modern conditions and the Pleistocene. Bacteriohopanepolyols (BHPs), specifically aminobacteriohopane-31,32,33,34-tetrol (aminotetrol) and 35-aminobacteriohopane-30,31,32,33,34-pentol (aminopentol) are diagnostic molecular markers preserved in soils and sediments that can be used to trace AMO and, therefore, CH4 cycling within both modern and ancient systems (hereafter termed CH4 oxidation markers). In this project, BHP distributions were determined within modern samples from the Congo catchment, including; 22 soils, 6 Malebo pool wetlands and an estuarine sediment. To complement this work on modern systems, BHPs were also analysed within ancient sediments from the Congo fan (ODP 1075) dated to 2.5 Ma, including high resolution studies of marine isotope stages (MIS) 5, 11 and 13. Within ODP 1075, high concentrations of CH4 oxidation markers are observed with no strong down core degradation signature. The study presents the oldest reported occurrence of CH4 oxidation markers, to date, with these biomarkers detected in sediments dated to 2.5 Ma (226 meters composite depth). Similarly, high concentrations of aminotetrol and aminopentol are also observed within the modern Malebo pool wetland and estuarine sediment, suggesting these sites as likely sources of CH4 oxidation markers to the Congo Fan. High concentrations of CH4 oxidation markers were found during MIS 5, 11 and 13 within ODP 1075 sediments. The CH4 oxidation marker signature during MIS 5 and MIS 11 coincides with high global CH4 concentrations (EPICA Dome C), whereas MIS 13 was characterised by low atmospheric CH4 concentrations. The strong similarities in CH4 oxidation marker concentrations during all three interglacial suggests a non-linear response in BHP production/burial and global CH4. This iii disparity could be due to the displacement of main sediment supply to the Congo fan and/or the relative similarity of CH4 sources during these time intervals. Alternatively, these differences could also be due to a threshold behaviour between methanogenesis – methanotrophy and the synthesis of diagnostic BHPs. A long term reduction in the mean concentration of CH4 oxidation markers is observed between 1.7 Ma and 0.3 Ma, suggesting a long term trend towards greater continental aridity within the Congo, consistent with changes in vegetation zones. Significant uncertainty still remain about the response of tropical CH4 sources and sinks during global climate perturbations, however, this study emphasises the large potential of BHPs as powerful novel tracers for methane cycling, both on land and in the ocean and across all climate zones.