Mt. Ruapehu is an andesite volcano located at the southwest termination of the Taupo Volcanic Zone (TVZ) in the North Island of New Zealand where the Pacific Plate subducts beneath the Australian Plate. Volcanism has migrated south with time as rifting from the Kermadec arc has propagated into the North Island lithosphere making Ruapehu the most recent addition to the TVZ. A detailed study of volcanic rocks and crustal xenoliths collected from Mt. Ruapehu and nearby satellite vents was conducted to establish how differentiation processes in arc magmas change with time in response to tectonic movements in the arc lithosphere. The thickness and composition of the arc crust in the southern TVZ directly controls the differentiation histories of magmas from Mt. Ruapehu and satellite vents. The crustal column beneath Ruapehu is composed of two main lithologies; a localised 5-10 km thick lower crustal meta-igneous granulite overlain by 20-25 km of meta-sediment. While fractional crystallisation is an important process in modifying the major and trace element concentrations, crustal assimilation plays a major role in modifying the isotopic composition of magmas from the southern TVZ. New trace element and isotope data for lavas and crustal xenoliths confirm a distinct change in assimilants between the oldest (Te Herenga - TH) and younger formations (Post Te Herenga - PTH) at Ruapehu. TH lavas are a product of lower crustal differentiation involving assimilation of meta-igneous granulite. PTH lavas display evidence for interaction with the same meta-igneous granulite followed by contamination with meta-sedimentary crust. This change in assimilant coincides with increased rates of extension at the southern tip of the Taupo Volcanic Zone (TVZ). We propose that the change in melt-crust interaction reflected in major geochemical changes from TH to PTH lavas is a response to adjustments in differentiation depth. Peripheral satellite vents near Ruapehu have produced some of the most primitive lavas in the southern TVZ with compositions resembling high-Mg andesite. This contrasts with the central-northern TVZ where the most primitive lavas are high-alumina basalt. The spatial distribution of these two magma series has developed as a result of the evolution of TVZ lithosphere. High-Mg andesites appear to be generated from depleted mantle during the very earliest stages of arc development when the lithosphere is thicker causing fractionation at relatively deep levels. High-alumina basalts however, represent a more mature phase of arc development and are derived from relatively fertile mantle following lithospheric thinning causing fractionation at relatively shallow levels. Finally, new osmium isotope data from meta-igneous xenoliths indicate the lower crust beneath Ruapehu is not formed of altered Jurassic oceanic crust. The relatively unradiogenic 187Os/188Os ratios of the xenoliths instead suggest they represent localised underplated material, which has only developed beneath the graben axis where the magma flux is greatest.

This thesis has two aims: i) an investigation of the effect of acid treatment on C/N, δ13C and δ15N of organic materials and; ii) the reconstruction of palaeoenvironmental change from a lake sediment sequence from south China, using geochemical proxies. To address these aims, a systematic comparative study was developed to examine the three most common acid treatment methods (capsule method; rinse method; fumigation method) and a range of acid reagents (HCl; H2SO3; H3PO4), on a range of terrestrial and aquatic, modern and geological sample materials. Acid treatment is a necessary step in the analysis of organic matter (OM) due to the distinct δ13C of inorganic carbon (IC) relative to the δ13C of OM; however, there is no consensus on “best practice”. I find that C/N, δ13C and δ15N values of OM are not just dependent upon environmental processes but also on acid treatment method, which adds significant non-linear biasing to the OM signal several orders of magnitude above instrument precision. This biasing is caused by the inefficient removal of IC from sample materials and the alteration of OM by the acid treatment process. Consequently, this can significantly alter the environmental interpretation of these proxies, for example, in determining OM provenance, indicating that the assumption that the effect of acid treatment on sample OM is either negligible or systematic, is flawed, with biasing in C/N ~ 1 – 100; δ13C ~ 0.2 – 7.2 ‰ and; δ15N ~ 0.2 – 1.7 ‰ in my range of samples. In addition, a long lake sediment sequence was cored from Lake Tianyang, Leizhou Peninsula, south China (20o31’1.11” N, 110o18’43.02” E) to reconstruct palaeoenvironmental variability using a suite of geochemical proxies (δ13C of bulk OM; XRF elemental ratios, magnetic susceptibility). The Tianyang δ13C record and elemental ratios (MIS 9 through MIS 6) show a strong glacial – interglacial imprint, providing evidence for periods of aridity during interglacial/interstadial periods in south China. The elemental ratios lose this imprint during MIS 5, likely due to an increase in catchment erosion (captured in the La, Nb, Ni, Th and Au records). Additionally, I tested the hypothesis that Ti concentrations reflect winter AM variability and showed that, whilst this may be valid, a robust interpretation is not possible due to the dilution of sedimentary Ti concentrations by local sources. A regional comparison of the Tianyang records appears to suggest a change in the strength of the land-sea thermal contrast in south China, implying a shift in mosiutre source region.

Broad changes in atmospheric oxygen have been identified for the last 4 Ga whereas changes in ocean oxygenation are not as well constrained. Mo isotope ratios have already been suggested as a potential global ocean palaeoredox proxy. However, a shorter oceanic residence time, theoretically greater isotope shifts, and no reliance on free H2S to alter its isotope composition, suggest Se isotope ratios could be a better global ocean palaeoredox proxy. The Cariaco Basin, on the Venezuela margin, has been the subject of numerous palaeoceanographic studies. Alternating laminated and non-laminated sediments indicate cyclical changes in water column oxygenation. It is therefore an ideal natural laboratory in which to study these effects on Se isotopes. A comparison with the Bermuda Rise, in the western North Atlantic, over the same time period, allows an evaluation of the spatial variations of sedimentary Se isotope signatures. Over the last interglacial-glacial transition, laminated (interglacial) sediments from the Cariaco Basin are characterised by high Se (~10 ppm) and high δ82/76Se (~0.03‰). In contrast, sediments from non-laminated (glacial) sediments have low Se (~0.5 ppm) and negative δ82/76Se (~-0.97‰). These end members result from two modes of Se cycling. Laminated sediments are dominated by organically-bound Se and low Se/Corg. Bioturbated sediments have a higher proportion of authigenic Se and higher Se/Corg. Over older interglacial-glacial cycles, the relationship between the two end-members and interglacial-glacial states becomes less clear. Similar end-members to those in the Cariaco Basin exist in the Bermuda Rise except the low δ82/76Se end-member has lower Se concentrations. As in the Cariaco Basin, variations are controlled by local climate rather than overall interglacial-glacial state. A comparison of the δ82/76Se values for the two sites show similarities which suggest that Se isotope ratios might not be just recording local changes.

Although investigations to find evidence for Quaternary environmental changes is conducted in Africa and the south Arabian Peninsula, in Saudi Arabia there are also vast areas covered with deposits that preserve detailed Quaternary histories of past climatic events. Unfortunately to date much of Saudi Arabia has not been explored in terms of past climatic changes, nor have in any sites been accurately dated using numerical dating techniques. As a result there is a scarcity of Quaternary palaeoclimatic archives set in a reliable temporal framework. This research aims to help remedy this situation through the examination and reconstruction of environmental changes in the area around Burydah, central Saudi Arabia. The findings of this research have shown evidence for significant wet phases in Marine Isotope Stage (MIS) 7, as well as the MIS 5d, 5b and 5a, which coincide with wetter phases elsewhere in the Arabian Peninsula and Africa. A later wet phase is represented by palaeolake deposits formed in MIS4. Similar to most areas in the Arabian Peninsula and North Africa, there is no evidence of climatic conditions during the Last Glacial Maximum (LGM), perhaps suggesting that conditions were hyper-arid with an environment that was erosive rather than depositional. Moving into the Holocene, interdune lake tufa deposits formed at MIS 1 c. 12-7ka. This research shows that dunes were forming during the early Holocene, as there were significant wet events in the Burydah, such as active wadis, and tufa deposits which formed at lake sites. Moving through to the mid-Holocene and towards present day, there is evidence for a drying out of the wadi systems and the establishment and growth of sand sheets covering the landscape. From c. 5ka onwards the Burydah area had become arid, with similar conditions to those that persist today.

Two flameless atomic absorption methods are described for the determination of Hg content and for the identification of Hg forms in rocks, soils and ores; the Ag-trap and the step-heating methods, respectively. In the first method the sample is rapidly heated to 800 °C in 3 min and in the second method the sample is gradually heated to around 900 °C in 30-35 min. In both methods the sample peaks are compared with standard peaks. The method was applied to three mineralization areas - Abergairn, Strontian and Tyndrum. 315 samples were collected at Tyndrum and primary and secondary dispersion patterns of Hg, Pb, Zn and Cu were investigated. The primary dispersion patterns are confined to the fracture zone. Hg is enriched in sphalerite and occurs as HgS and as a lattice-held component. Galena and chalcopyrite are poor in Hg. A study of the geochemistry of mercury in different soil horizons has revealed that widest anomalies are obtained with samples of the -0.063 mm fraction of soils from B horizon. HgS is the predominant phase of mercury in the soils with an increasing importance over the ore zone. Organic matter is the chief adsorbing medium whilst clay minerals are unimportant. 1316 rock samples were collected over an area of 230 km2 in the Cairngorm granite. The importance of sample size and in-situ variance are discussed. The mercury patterns are presented as a contour map. The granite is remarkably low in Hg (average 6 p.p.b.) which is distributed uniformly in most areas. Micas contain the highest concentration of Hg followed by feldspars and quartz. Mercury is not associated with pegmatite veins. In all the samples Hg is predominantly held in the lattices of rock forming minerals but small amounts of HgCl2 and HgS are found in the anomalous samples. The granite is also low in Zn and Cu. Although the studied areas occur in a Hg-poor province mercury appears to be a useful pathfinder element in the search of sulphide mineralization. The ease of analysis and also the potential of the thermal release technique together with chemical manipulation of samples are the obvious advantages of mercury.

Direct evidence for the response of Earth’s largest continental ice mass, the East Antarctic ice sheet (EAIS), to climatic warmth is extremely limited. The primary aim of this thesis is to improve understanding of the behaviour of the EAIS during the warmer-than-present Pliocene Epoch (2.58 to 5.33 million years ago). To this end, I analysed the radiogenic neodymium and strontium isotopic provenance of fine-grained (<63μm) Pliocene detrital marine sediments deposited offshore of the East Antarctic continent, which can provide information on source bedrock characteristics, continental erosional patterns and marine sediment depositional processes. In addition, I also analysed argon isotopic ages of ice-rafted hornblende grains (>150μm), to infer sites of major iceberg production events through time. Within this thesis, I present Pliocene marine sediment data from various cores drilled from the East Antarctic margin, thereby developing a detailed framework for linking provenance variability to ice sheet behaviour. My key findings have been collated into five distinct chapters, providing: i) the first evidence for significant retreat of the EAIS in the low-lying Wilkes Subglacial Basin in response to the earliest Pliocene climatic warmth; ii) insights into the benefits and pitfalls associated with utilising different tools in glaciomarine sediment provenance studies; iii) constraints on the behaviour of the EAIS and West Antarctic ice sheet during the warmth of Pleistocene super-interglacial, Marine Isotope Stage 31; iv) insights into the role of declining sea surface termperatures during the Pliocene on the flux and provenance of distally sourced ice-rafted detritus, along with evidence for potential ice sheet destabilisation events in the Aurora Subglacial Basin during Pliocene interglacials; and v) advances in understanding of the evolution of the EAIS during the Late Pliocene climatic transition, and its role in global Pliocene climate change. Hence, the findings presented within this thesis provide new and significant evidence for the behaviour of the EAIS during the Pliocene, and suggest it has in the past been more sensitive to climatic change than previously realised.

The aim of the thesis was to test the isotope systematics of Cu, Zn and Pb in atmospheric particulate matter and potential source material in order to critically discuss its implications and applications to studies of global biogeochemical cycles and anthropogenic impact. Four specific objectives were addressed: First, we tailored existing analytical methods to enable for the first time accurate and precise Cu and Zn isotope ratio measurements in dust and aerosol samples with low element content. Second, we studied the Pb, Cu and Zn isotope composition in aerosols collected in London to test their potential to trace anthropogenic sources in the urban atmosphere. Third, we characterized the Cu and Zn isotope systematics of major Asian and African dust sources and constrained possible controls and implications on the marine isotope signature. Finally, we investigated the isotope fractionation of Zn during dissolution experiments with acidic solution of natural dust to test the possible effect of atmospheric processing on the isotope signature of deposited Zn. The conclusions from this work are: First, stable isotopes confirm that brake and tyre emissions are the likely dominant sources of Zn and Cu in urban aerosols and that recycled Pb deposited originally from leaded gasoline still contributes significantly to the burden of atmospheric Pb in London. Second, we show that the Cu and Zn isotope signatures vary significantly between the major Asian and African dust sources and we find a significant correlation between Cu isotope ratio and illite content. Mass balance calculation suggest that bulk samples record faithfully the isotope signature of the <4 μm fraction. Third, significant isotope fractionation of Zn is observed during dust dissolution suggesting that the isotope signature of Zn deposited in the surface ocean is not only controlled by source mixing but also by atmospheric processing during long range transportation. The work clearly supports the initial hypothesis that metal isotopes are powerful sources to study global and regional biogeochemical cycles.

The ulttamafic and related lithologies of the Swat valley ophiolite lie along the Main Mantle Thrust (MMT) between the Kohistan island arc and Indo-Pakistan plate. The textural and mineralogical characteristics of the ultramafic rocks seem to be the result of a complex interplay of a number of processes including partial melting, serpentinization, metamorphism, carbonation and late-stage hydrothermal alteration. The modal mineralogy of the least modified varieties (non-carbonated serpentinized) of the ultramafic rocks indicate that the majority of them them are harzburgites. The chemical composition of the primaiy phases (chrome spinel and pyroxenes) and major, minor as well as trace element contents of the non-carbonated ultramafic rocks suggest that the major bulk of the rocks originated as the upper mantle residue left after an intermediate degree of partial melting. Pervasive serpentinization of the rocks, evident from the almost common occurrence of serpentine and orthopyroxene bastites, was followed by a low to medium grade regional metamorphism converting the original low temperature serpentine (lizardite) to antigorite, forming olivine and, rarely, diopside. The trace amounts of Ni-rich opaque phases in the rocks largely formed by the modification of primary sulphides during serpentinization. Locally, alteration by a CO2-bearing fluid resulted in the carbonation of the formerly serpentinized ultramafic rocks forming quartz- and/ or talc-carbonate lithologies. The oxygen and carbon isotope composition (delta18Ocalculated = +13.6 1.2 � SMOW; delta13Ccalculated= -1.8 0.7 � PDB) indicates that the fluids responsible for carbonation were probably derived from the subduction-related metamorphism of the carbonate-bearing sedimentary rocks of the Indo-Pakistan plate. Late-stage hydrothermal activity by Be, B and K-bearing fluids, apparently related to quartz veins and stockworks, has affected the ultramafic rocks mainly the carbonated types because of their highly fissile and permeable nature. This process has produced emerald, Cr-tourmaline and fuchsite. The chromium in these phases was extracted from the host ultramafic rocks. The oxygen isotope composition of parental waters (delta18Ocalculated = +14 1.7 � SMOW), calculated from the measured values of emerald, quartz, tourmaline and fuchsite, is high in the delta18O component. This high delta18Ocalculated well as the range of 5D of channel waters from emerald (-42 6.6� vs SMOW) and that of fluid calculated (-47 7.1 � vs SMOW) are consistent with both metamorphic and magmatic origin. However, the close similarity between the measured 5D values of the hydroxyl waters in fuchsite (-74 to -61� vs SMOW) and tourmaline (-84 and -69 %o vs SMOW) with pegmatitic muscovite and tourmaline suggests that the mineralization was probably caused by modified (18O-enriched) hydrothermal solutions derived from an S-type granitic magma.

No description available.

The aim of this study is to attain a better understanding of the fate of carbon and the degree of preservation of organic matter in impacts craters. This topic is relevant to prebiotic chemistry, planetary exploration and panspermia. Three natural impact craters were investigated: The Eocene Haughton impact structure on Devon Island in the Canadian High Arctic, where the target is composed of a thick sedimentary sequence with pre-impact hydrocarbons, the Ordovician Lockne impact structure in central Sweden, where organic- and uranium-rich Alum Shale is present in the target, and the Pre-Caledonian Gardnos impact structure in south-central Norway, also with carbon-rich shales in the target. An analogue example of rapid heating of organic matter by igneous intrusion in carbon-rich rock, and an experimental hypervelocity impact in a carbon-rich rock were also studied. In the Haughton impact crater, fossil biological markers are preserved within lithic clasts in the impact melt breccia. Fossil biological markers are also preserved in the experimental impact crater, and in the material excavated during the experimental impact. Instantaneous melting and quenching in the Gardnos suevite (impact melt rock) has caused the incorporation of disordered carbon in the suevite melt. In the Haughton impact melt breccias, around 20% total organic carbon is preserved after impact, and in the Gardnos impact crater, at least 38% of the original carbon is preserved in a disordered form. In the Lockne impact crater, carbon is particularly preserved by polymerisation around radioactive minerals. This study shows that organic matter does not necessarily oxidise or change in to highly crystalline forms during an impact event, but that substantial amounts can be preserved in a disordered form.