This thesis evaluates potential pathways to constrain the evolution of metamorphic CO₂, using the observation that significant CO₂ fluxes have been noted from hot springs above active metamorphic belts. Theoretical constraints on CO₂ behaviour are assessed using available observations and experimental data literature. Metamorphic decarbonation, fluid immiscibility, groundwater mixing, boiling, and phase separation (CO_2(aq) = CO_2(g)) are modelled theoretically, chiefly using water chemistry and isotopic constrains on CO₂ evolution. Measured carbon isotope compositions for spring fluids in the Marsyandi and Trisuli Valley in Nepal reach +13‰, while coexisting free gas-phase compositions are close to -4‰. The considerable variation between gas and liquid isotopic compositions constrains the possible fractionation pathways by the necessity to satisfy mass balance. The large isotopic variation between liquid and complementary gas phase suggests that degassing from some springs is quantitative and requires up to 99% CO₂ loss from solution. Noble gas systematics from effervescing springs in the Marsyandi provide evidence consistent with quantitative degassing, although a lack of data precludes a more detailed appraisal. No mantle noble gas signature is detected in Marsyandi or Trisuli springs attesting to a dominant metamorphic origin of CO₂ (atmospheric and organic CO₂ excepted). Existing petrological methods for the orogen-wide determination of metamorphic CO₂ liberation from the Himalaya are examined and advanced using the study of calc-silicates. Extrapolation of geochemically derived hot spring fluxes, assuming a minimum of ~ 99% degassing of CO₂ from hot spring waters, implies a CO₂ flux between 0.6 and 1.5 x 10¹² mol yr^-1, consistent with estimations made on the basis of petrology. This represents a significant contribution to the global carbon cycle and may potentially force climate. A paradox still remains, however, between climatic cooling during the Cenozoic and the large solid-Earth CO₂ degassing fluxes (which would warm climate).

The Iron Hypothesis suggests that an increased flux of aerosol iron to the Southern Ocean (and other Fe limited regions of the ocean) during glacial intervals may have stimulated primary productivity. This could have resulted in more storage of carbon in the deep ocean and less in the atmosphere. This study was designed to increase our understanding of the iron hypothesis, by increasing constraints on several aspects of the theory. The results show that the mean aerosol Fe solubility at Dome C during the Last Glacial Maximum was relatively high (10%), compared to the solubility of typical crustal materials and assumptions used in previous models of ocean biogeochemistry, but also very variable (1-42%). Measurements of other major elemental concentrations in ice were made to assess their suitability for use as proxies of Fe concentrations in dust and to discriminate changes in dust composition and mineralogy. This study shows that ice core Ca concentration is a good proxy for total, but not seawater-soluble, aerosol Fe concentration. These results suggest that atmospheric dust which was deposited at Dome C was enriched in K-rich clay minerals over the Patagonian source regions. New methods were developed to sublimate ice-core material to extract dry dust for use in biological incubation experiments. This study has been the first to incubate siliceous Antarctic phytoplankton (diatoms) with dust of last glacial maximum age, extracted from the EDC ice core, under realistic conditions. The results show that dissolved Fe released from dust had a direct fertilising effect.

In rivers draining the western Southern Alps, the organic carbon concentration (C_org) and stable isotopes of organic carbon (<i>δ</i>¹³C_org) are used to determine the source of riverine particulate organic carbon (POC). The rivers contain a mixture of POC derived from the terrestrial biosphere (vegetation and soils) and fossil POC in bedrock described by C_organd <i>δ</i>¹³C_org. The fraction of riverine POC derived recently from atmospheric CO₂ (fraction non-fossil POC, F_nf) is quantified using this mixing relationship. F_nf, C_org suspended sediment transfer and water discharge (Q_w) are used to calculate the erosion of POC from the terrestrial biosphere in these active mountain belts. The high yields (34±5 tC km^-2 yr^-1 in western Southern Alps, and 31±7 tC km^-2 yr^-1 in Taiwan) are presently sustained by the net primary productivity of the forest ecosystem in both mountain belts. With detailed knowledge of the source of riverine POC the processes responsible for its mobilisation and transfer are investigated. The primary tool is the detailed sample set collected from 15 rivers in Taiwan that combine the geochemistry of riverine suspended load POC with records of water discharge (Q_w), SSC and precipitation. Remote sensing is used to provide additional insight. Three main routing processes are identified: 1. Landsliding – mobilising standing biomass and soils; 2. Overland flow – during precipitation organic materials are washed from hillslopes to the channel; and 3. Grinding – breakdown or coarse organic fragments during transport in turbid river waters. These processes cause the observed enrichment in non-fossil POC concentration in rivers (POC_nf, mg L^-1) during floods, which conflicts with a conventional view of dilution at high Q_w and SSC. These processes lead to positive relationships between POC_nf and Q_w in Taiwanese rivers and this may be common throughout vegetated mountain catchments. The POC routing processes identified here lead to the high non-fossil POC yields in these mountain belts sustained by both tectonic uplift and climatically-driven erosion and sustenance of the terrestrial biosphere. Because of the observed power law relationship between POC_nfand Q_w, large floods caused by storms are responsible for the bulk of the riverine non-fossil POC transfer. It is recognised for the first time that storm-triggered floods optimise the delivery of non-fossil POC to depositional environments at times when it is likely to be sequestered in sediment. In Taiwan, the mobilisation of non-fossil POC from hillslopes significantly impacts the terrestrial biosphere. A move toward N-limitation on steep slopes (observed in <i>δ</i>¹⁵N of vegetation and soil) is driven by the physical removal of organic matter during rainfall and may influence the net primary productivity of the ecosystem. If common throughout mountains in tropical Oceania, this climatically driven phenomenon affects the ability of these forests to sequester atmospheric CO₂ on decadal time-scales.

We developed a joint interpretation method to allow us to combine the geophysical constraints from seismic and controlled source electromagnetic (CSEM) experiments. We used effective medium techniques to represent the bulk elastic and electric properties of a two-phase composite material, by describing the microstructure and the relative volumes of each phase. We used the misfit, from a calculated value of rock property to a geophysically observed value. The misfit is calculated from both elastic and electric properties of the rock, providing a common variable with which to combine the seismic and controlled source electromagnetic (CSEM) constraints. We applied the joint interpretation technique to geophysical results from the East Pacific Rise, Reykjanes Ridge and the Valu Fa Ridge, providing estimates for fluid fraction, degree of interconnectivity of the fluid phase and pore geometry. Prior to developing the joint interpretation method we revisited the East Pacific Rise CSEM dataset, using more developed modelling methods with fresh information regarding the phase drift of the electromagnetic source. We concluded that the presence of a magma chamber was not required by the dataset, as did the original interpretation. However, we accept a small magma chamber (tens of metres) would not be resolved by the CSEM technique. Our results contrast with original seismic results suggesting the presence of a large magma chamber (>1km wide). However, later reworking of the seismic dataset has suggested the presence of a much smaller magma chamber. We believe that with careful assessment of the limitations of each technique the results are mutually satisfactory.

No description available.

No description available.

This work reports the use of the vibrating gold microwire electrode, with new methods developed for the speciation of Zn, Cu, Mn and As at natural levels in waters of neutral pH. Trace metals and metalloids can be distributed as different species in the environment, which can control mobility, toxicity and bioavailability and in turn depends on many complex factors. Analysis of this distribution (speciation) can provide an understanding of the relationship between such elements and their relationship with organisms in marine environments and humans through contaminated drinking water supplies. Such speciation can be analysed using a vibrating gold microwire electrode (VGME), which is easily prepared and maintained at minimal cost. High sensitivity is found for trace metals: Mn; Cu and Zn; and the metalloid As, resulting from a very low diffusion layer (~0.8 urn for a Sum gold wire) means that they can be measured at trace levels in natural waters. This combined with the VGME's portability, reliability and stability for long term measurement (repeated measurements over several days) and its capability to distinguish between distinct forms ofthe above trace elements mean that speciation methods could be successfully developed and validated in natural waters, with no, or minimal sample preparation. Such methods made it possible to analyse speciation on-site, which decreases the potential problems inherent in maintaining sample speciation during storage. Contamination of groundwater with As is a major health risk through contamination of drinking and irrigation water supplies. In geochemically reducing conditions As is mostly present as ASIII, which is why a method that uses cathodic stripping voltammetry (CSV) to determine reactive AS"I was developed. The ASIII is detected after adsorptive deposition of As(OHho, followed by a potential scan to measure the reduction current from AsIII to Aso. The method is suitable for waters of pH 7-12. The CSV method was successfully applied to groundwaters from Severn Trent, UK, however speciation using this method is severely hampered by high levels of iron and manganese. Experiments showed that the interference is eliminated by addition of EDTA, making it possible to determine the arsenic speciation on-site by CSV in West Bengal, India. The VGME is also used to detect nanomolar levels of dissolved Mn by anodic stripping chronopotentiometry (ASC) and sub-nanomolar levels of dissolved Zn by anodic stripping voltammetry (ASV) in neutral pH seawater. The limits of detection for Mn (1.4 nM) and for Zn (0.3 nM) in seawater with a 300 s plating time, are better than achieved using other non-mercury based electrodes and nearly as good as a mercury film electrode for Zn. Deposition of Mn at the VGME was further utilised to catalyse the reduction of Asv to ASIII, enabling for the first time the direct electrochemical determination of Asv in natural waters of neutral pH including seawater by ASV using a manganese coated gold microwire electrode. Direct electrochemical determination of Asv in neutral pH waters is impossible due to its electro-inactivity. Therefore Mn is added to excess (~1 JlM Mn) to the water leading to a Mn coating during the deposition of Asv on the electrode, when depositing at -1.3 V. The detection limit was 0.2 nM Asv using a deposition time of 180 s. Speciation of Cu is determined without the need for sample preparation, using scanned stripping techniques for the first time at natural levels in seawater. A desorption potential (-1.2 V) and a conditioning interval between scans make the VG ME suitable for on-site and potentially in-situ copper speciation. The resulting pseudopolarograms are analysed using an experimentally constructed 'chelate scale' to determine the strength of copper ligand interactions in real seawater samples.

No description available.

No description available.

The focus of this thesis is on biogeochemical cycling in terrestrial ecosystems, and the effects on soil and surface water quality at a range of spatial scales throughout Scotland. The fundamental challenge of this research was to integrate the dynamic Model of Acidification of Groundwater In Catchments (MAGIC), within a spatial framework, by extrapolating knowledge of biogeochemical processes at the catchment scale to larger (i.e. regional and national) spatial units. The MAGIC model was applied to 9 Scottish sites in the U.K. Acid Waters Monitoring Network (UKAWMN), 59 sites in the region of Galloway and 733 sites throughout Scotland. Reduction in sulphur (S) emissions associated with the Second S Protocol and different forestry (land use) scenarios were modelled at these different scales to predict the existing and likely future extent of soil and surface water acidification in Scotland. The sensitivity of MAGIC to soil input data derived from two different methodologies was tested at the national scale. Anticipated reductions in S emissions are predicted to have a marginal beneficial effect on the reversibility of soil acidification at all spatial scales throughout Scotland, irrespective of the methodology used to determine the soil input parameters. With the exception of the most acid sensitive parts of Scotland, surface water Acid Neutralising Capacity (ANC) modelled at a national scale presents a picture of improving ANC in response to the Second S Protocol. From a policy perspective however, these results are potentially misleading. It is important that European legislation targets the most acid sensitive soils and surface waters as it is these ecosystems that require protection.