School of Earth Sciences - Theses

Permanent URI for this collection

Search Results

Now showing 1 - 10 of 18
  • Item
    Thumbnail Image
    The geology, petrology and geochemistry of the granitic rocks of Victoria
    Rossiter, Allan G. (University of Melbourne, 1973)
    In the field the granitic rocks of Victoria show a wide variety of characteristics but chemically all are related and are the result of similar processes that occurred cyclically during the Palaeozoic. The granites may be divided into two groups on the basis of whether or not primary hornblende is present in the more basic members of the suite. Rocks of the hornblende-free series concentrate mainly in central Victoria and usually contain biotite that is red-brown in colour. In the granites of the hornblende-bearing group green-brown biotite is generally present. The two different colours appear to indicate unlike oxygen fugacities in the magmas from which the micas crystallized. The rocks of the hornblende-bearing and hornblende-free provinces also vary slightly in chemistry - the most important difference being that the basic members of the first group contain slightly more Ca than those of the second having comparable silica content. The compositions of the Victorian granites is consistent with their derivation by anatexis of a mixture of basic igneous rocks and sedimentary material. The conclusion is drawn that a combination of Cambrian basalts and dolerites and Cambro-Ordovician sediments constitute the source. The explanation of the fact that two chemical trends are observed in the granites may lie in the possible existence of two chemically distinct groups in the Cambrian igneous rocks - the one having lower Ca predominating under central Victoria.
  • Item
    Thumbnail Image
    Investigation of air-sea fluxes over the Southern Ocean using an eddy-covariance technique and parameterization using stability functions
    Chen Reddy, Sushma Reddy ( 2019)
    The Earth is an integrated system that consists of sub-systems that interact and influence each other. These interactions have an important influence on the understanding of weather and climate of the earth system. Air-sea interactions are one such interaction that affects the Earth's system — thus making it essential to understand the physical processes that affect the prediction and forecast of the weather and climate. The present state of art climate and numerical weather prediction models use bulk models which are based on Monin-Obukhov similarity theory and Charnock's relations to determine the fluxes across the air-sea interface. The COARE 3.5 model is the best performing model available, and it is seen that the model underestimates the fluxes at higher wind speeds. Hence, to avoid any assumptions and circular dependencies, we need to build a simple parameterization of coefficients of fluxes to determine fluxes. Eddy Covariance, the purest form of flux calculation, is used to develop the parameterization. Eddy covariance relies on high-frequency 3-D winds, which, on ships, are contaminated by platform motions. However, in the absence of reliable accelerometer data, or a failed collocated accelerometer, calculating these motions is difficult. Here, in this study, we studied if the ship's motion reference data can replace external collocated accelerometer data. We have characterized that for the anemometer mounted on the foremast of the R/V Investigator, and there is a lag of 1.4 sec in the ship's motion reference unit data. Hence, we can correct the wind speeds for platform motions using the ships' motion data after adjusting to the lag. The spectral speak due to the platform motions observed in the measured raw data by anemometer is removed after the corrections performed by the ship's data. Hence, achieving the redundancy of the external collocated accelerometer, GPS receiver, and heading sensors. The fluxes computed from the eddy covariance technique are used to get a simple parameterization to estimate fluxes. Here, we have developed the coefficients of drag, latent heat fluxes in terms of simple functions of Reynolds and bulk Richardson number, which are physically dependent on velocity and stability of the atmospheric boundary layer. The model proposed does not depend on any assumptions or does not have any circular dependencies. The coefficient of sensible heat flux could not be parameterized as we observed that there is no dependence on Reynolds number in the neutral, stable region. The proposed model is performing better compared to that of the COARE 3.5 model at higher wind speeds. Gas transfer across the air-sea interface is challenging to measure, and the existing relationships for the gas transfer velocity with wind speeds have a high variance at high wind speeds. It is essential to measure gas transfer velocities in the Southern Ocean as it is least sampled with the rough environment and high surface waves. It is estimated that the Southern Ocean is the largest sink of anthropogenic carbon dioxide, with about 40% of the total world ocean sink. Gas transfer velocities of CO_2 in the Southern Ocean are measured, and it is found that the results obtained are within the range that is reported by the previous researchers. However, there are no sufficient data points, and the variance in the data is high to get any conclusions from the results obtained.
  • Item
    Thumbnail Image
    Inverse modeling of regional CO2 surface fluxes in the Latrobe Valley/ Otway basin
    Taneja, Ritu ( 2019)
    Carbon dioxide (CO2) is the most important greenhouse gas (GHG) attributable to human activity. Many natural components are responsible for the carbon emissions in the atmosphere. Therefore, large uncertainties exist in the current carbon budget. Using the atmospheric transport model and inverse techniques, we can estimate the carbon sources and sinks from the atmospheric CO2 measurements. The current global network is quite efficient in constraining the global carbon emissions. However, to obtain detailed knowledge about the controlling processes and better understanding of the carbon cycle, estimates of regional fluxes are important. To constrain the regional fluxes, it is important to capture the right signals at the right time. Therefore, continuous monitoring of atmospheric CO2 has been an essential part of the project. An optimal network design indicates the best location for the atmospheric monitoring units to be placed in the given domain. For our network design study, we have used the Lagrangian Particle Dispersion Model (LPDM), driven by mean wind velocity, turbulent kinetic energy and potential temperature. The necessary fields to drive LPDM are taken from Weather Research Forecasting (WRF) Model. Since, the input fields for LPDM are dependant on the output of the WRF model. On that account, it is important to choose an appropriate planetary boundary layer (PBL) scheme, which can produce all the required fields to drive LPDM. Scarcity of measurements and inaccurate representation of vertical transport within the planetary boundary layer in atmospheric transport can also lead to large uncertainties. Therefore, selection of an appropriate planetary boundary layer scheme to represent realistic atmospheric transport is indispensable. This study covers three main aspects: 1. Comparison between different planetary boundary layer schemes available in the Weather Research Forecasting (WRF) model. 2. Obtain an optimal network to constrain the greenhouse gas emissions in the Latrobe Valley. 3. Constraining the carbon sources and sinks in the Otway basin. We have used Bayesian synthesis inversion and incremental optimization to obtain an optimal network. The stations towards the central Latrobe Valley and east of the given domain would be able to constrain emissions from the power plants and major towns in the Latrobe Valley. As a proof of concept, we demomstrate the methodology using observations. We constrain emissions in the Otway Basin. For inversion, we adapt Bayesian synthesis approach. We further assessed the inverted fluxes with the flux tower data. Once we have data for the Latrobe Valley, we can use the same approach to estimate carbon fluxes.
  • Item
    Thumbnail Image
    Isotope and trace element systematics in subduction systems and the concept of ambient mantle
    Ikei, Anzu ( 2018)
    This project employs Hf - Nd isotope ratio and trace element concentration data to investigate aspects of magmatism in subduction systems. The primary goal is to use the compositions of back-arc basin lavas as a means to investigate the nature of the pre-subduction mantle wedge ( ‘ambient mantle’). New Sr, Nd, Pb and Hf isotope data for the Lau Basin back arc spreading centres and Hf – Nd data for the Manus Basin back arc are reported. Both areas have been extensively studied using major, trace element and Sr-Pb isotope data but many of these elements are highly mobile in aqueous fluids released from the subducting slab and so are unfavourable for studying the mantle wedge itself; here the behaviour of the less mobile elements Hf and Nd can provide additional insights. Hf is a high field strength element (HFSE) and believed to be relatively immobile in subduction zone fluids until slab melting occurs whereas Nd is slightly mobile in aqueous fluids, and also mobile in melts. The new Hf and Nd data from both the Lau and Manus back-arc basins plot broadly in the Indian Mid-Ocean Ridge Basalt(MORB) domain with the exception of two lavas from the Eastern Lau Spreading Centre (ELSC). Two geochemical filters (Ba/Nb < 7 and Th/Ta < 3), used in a previous study of the back arc region to the Mariana arc, were modified slightly for each studied site and applied to the new data to remove major influences from subduction components in the mantle wedge. These filtered data for the Lau and Manus Basins confirm that the true mantle affinity in these regions is indeed Indian MORB. When compared to existing data from the Mariana Trough and Scotia Sea, it is clear that each mantle wedge has its own unique ambient mantle character and that these site-specific compositions should be used in preference to global average MORB compositions for the purpose of modelling. Despite the geochemical filtering methods employed, it is clear that all of the samples still retain a small subduction input in terms of their LILE contents. A subsidiary part of this project re-evaluates existing HFSE concentration data for lavas of the Tongan arc, previously determined by laser ablation methods. New solution ICP-MS data suggest that the older data are likely influenced by analytical artifacts, a conclusion that dramatically changes the genetic model required for their interpretation.
  • Item
    Thumbnail Image
    The Flying Electric Generator: evaluating the claims of a largely ignored proposal for generating electricity from high-altitude winds
    Kambouris, Steven ( 2015)
    This thesis concerns the Flying Electric Generator (FEG), a technology proposed to generate electricity from winds several kilometres high in the sky. Airborne Wind Energy (the generation of electricity from high altitude winds) is an emerging field of research, with several technological approaches under development. High altitude winds are attractive for this purpose because they are generally much faster than surface level winds, and because power is a cubic function of wind velocity. Winds are fastest within the subtropical jet stream, located about 25–30 degrees north and south of the equator, at an altitude of 10–12 km. The FEG is a device consisting of multiple rotors attached to a frame, which is tethered to the ground. The rotors work as autogyros to provide lift; additional energy extracted from the wind is converted to electricity and conducted to the ground via the tether. The FEG would operate kilometres high in the atmosphere, up to jet stream levels. Papers about the FEG were first published in 1979, and in 2002 a company was founded to commercialise the FEG. So far, this has not happened, and many details of how the technology would operate remain uncertain, despite three decades of research literature. Only small test craft (rotors of up to 24 feet in diameter) have flown at low altitudes (up to 100 feet). Many of the claims in the literature, which are optimistic about the FEGs performance at high altitude, are experimentally untested. FEGs have never operated at the altitudes described in the corresponding literature, and the project has not been commercialised or attracted much if any recent research funding. Other, newer entrants to the Airborne Wind Energy field have seen success in research funding and commercialisation. This thesis addresses two problems: first, it tests some of the claims in the FEG literature and second, it attempts to fill in details not provided. The particular claims concern the power density available in high altitude winds over Australia and its seasonal variation, the amount of time a hypothetical FEG setup would be "grounded" due to insufficient wind speeds to keep it aloft, and expected capacity factors of a hypothetical FEG setup. Claims about the magnitude of the wind power resource were tested using reanalysis data (the ERA-40 dataset was used, and was validated against Bureau of Meteorology upper air statistics). Power density and wind speeds at different altitudes above Australia were calculated and analysed. The reanalysis wind data in conjunction with a model of FEG operation (based on lifting rotor theory detailed in the FEG literature) were used to calculate downtime and capacity factors. The results showed a clear seasonal variation in power density over Australia, which was most pronounced at 30 degrees south of the equator (although winds above Tasmania showed much less variation). Winter had the strongest winds, and summer the weakest. The highly skewed distribution of power density meant that median power densities (unreported in the FEG literature) were more appropriate than means. Downtime calculations showed that a particular FEG setup rated at 240 kW operating at a pressure level of 600 hPa would be landed for at least 20% of the year at all locations in Australia, and for at least 40% of the year north of 20°S. Annual capacity factors for the same FEG setup were calculated to vary between 0.1–0.4 over Australia, no different from conventional ground-based wind turbines. Capacity factors for the summer months were substantially lower than the annual values. These results support the main contention of the thesis, that the FEG is far more limited in its potential as source of energy from the wind than the literature claims.
  • Item
    Thumbnail Image
    Investigating potential pathways to remediate thiocyanate-contaminated groundwater and wastewater at a Victorian gold mine
    Bosnjak, Angelina ( 2014)
    Thiocyanate (SCN-) is toxic to higher organisms, affecting the central nervous system, to cause irritability, nervousness, hallucinations, psychosis, mania, delirium and convulsions. This contaminant is commonly associated with gold mining activities and forms when cyanide, used for extraction of gold, reacts with reduced sulphur species in wastewater. Significant concentrations of thiocyanate have been detected in groundwater bores around the No. 2 Tailings Dam at Stawell Gold Mine in Victoria, with concentrations steadily increasing over time. Thiocyanate can be eliminated from groundwater and wastewater by oxidation to less harmful products such as ammonia and sulphate through chemical oxidation and biodegradation. Certain bacterial strains indigenous to thiocyanate-contaminated sites have demonstrated the ability to degrade thiocyanate and can be used to remediate contaminated land and water. Chemical and biological oxidation of thiocyanate was explored in this study through synthetic abiotic laboratory redox optimisation experiments and field-based injection experiments to determine the controls on thiocyanate degradation and potential pathways which could be implemented to remediate thiocyanate-contaminated groundwater and wastewater at the site. Analysis of site groundwater chemistry and aquifer properties revealed the presence of a plume of thiocyanate in acidic to near-neutral groundwater outside the tailings dam wall at monitoring bores SE12 and SE14 confined to the upper unconsolidated aquifer with low hydraulic conductivity (0.001-0.004 m/d) and low transmissivity (0.01-0.05 m2/d). The source of thiocyanate at the site was determined as the tailings dam. Inconclusive results from the nitrate reduction-thiocyanate oxidation experiments in synthetic and actual groundwater and wastewater suggests that nitrate may not form a redox couple with thiocyanate. However, in acidic solutions (pH 2) of Fe-EDTA and thiocyanate heated to 80°C, the Fe2+ was generated from thermal degradation of the Fe-EDTA compound, while thiocyanate was completely hydrolysed within 22 days. Catalysis of thiocyanate hydrolysis by iron reduction was not determined. However, solution pH and temperature were important factors, as thiocyanate hydrolysis did not proceed at pH 5.5 and 80°C. At pH 2, the hydrolysis of thiocyanate was faster at 80°C compared to 70°C. Thiocyanate-degrading microorganisms were not successfully stimulated in simulated injection (push-pull) experiments, as no appreciable decrease in thiocyanate concentrations was observed in groundwater or wastewater replicates.
  • Item
    Thumbnail Image
    The palaeovegetation, palaeoclimate and biostratigraphy of South-eastern Australia during the Eocene Greenhouse to Oligocene Icehouse transition
    Nguyen, Ngoc ( 2014)
    The shift from the Eocene 'Greenhouse' to Oligocene 'Icehouse' conditions is a globally significant event. The Eocene – Oligocene boundary (33.9 million years ago) marks the onset of permanent large-scale glaciation in Antarctica and the onset of the current global climatic 'Icehouse‘ state. In the Austral-Antarctic region, the transition is well recorded in deep water marine carbon and oxygen isotopes, however in the terrestrial realm little is known of the nature of environmental change associated with this event. In Australia, this is partly due to a lack of precise methods of dating Paleogene terrestrial strata. The most common practice of dating terrestrial sediments is the application of Gippsland Basin‘s spore-pollen biostratigraphy. This research assessed the applicability of the Gippsland Basin‘s spore-pollen biostratigraphy to the Otway Basin. Results demonstrated Gippsland Basin‘s spore-pollen biostratigraphy provides inaccurate ages when applied to samples from the Otway Basin. In addition, results from pollen analyses combined with foraminifera, dinoflagellate, and calcareous nannofossils and other stratigraphic data were used to create a detailed climate history of this high palaeolatitude region (~60°S). The spore-pollen assemblage indicates the palaeovegetation of the Otway Basin was dominated by Nothofagidites spp., with an abundance of species from the families Myrtaceae, Casuarinaceae and Proteaceae. The Nothofagus-dominated mesothermal rainforest probably had an emergent layer of Podocarpaceae and Araucariaceae species and a diverse understorey of ferns. The palynological record of the Otway Basin shows no vegetation change occurred during the time prior to and across the Eocene-Oligocene boundary even though there is evidence of permanent continental–scale glaciation on Antarctica.
  • Item
    Thumbnail Image
    Investigations into diamond drilling and rock drillability: microscale and fullscale impregnated bits
    Siribumrungsukha, Boonsom ( 1980)
    Drill bits having cutting edges formed of a matrix of sintered metal powder impregnated with diamond bort (the so called "impregnated" diamond bits) have the potential to greatly increase the economy of diamond drilling, due to low production costs and long bit life resulting from reconditioning of the cutting edges. However, little detailed research into the performance of impregnated bits has been conducted, and the capacity to predict their field behaviour is limited. As a contribution to the knowledge of impregnated bit performance, this project has studied basic relationships between thrust, rotational speed and the penetration of diamond bits into rock, with particular attention paid to the development of techniques suitable for laboratory testing using miniature impregnated bits. A conventional bench drill and a radial arm drill were modified and instrumented to enable operating conditions to be controlled and drilling parameters to be measured and monitored, for microscale and fullscale drilling using impregnated drill bits. Bit performance and drilling characteristics were studied for four rock types, and rock drillability studies were carried out on seven rock types. Statistical relationships between penetration rate, specific energy, torque and drilling distance that were determined enable projection of drilling data from a standard "sharp" condition. A wear measuring device was developed to assess matrix wear of the impregnated microbits. Reconditioning was done by drilling medium-strength, abrasive Stawell sandstone. Initial penetration rate increases linearly with increases in thrust and/or rotational speed, but only within a limited range, depending on rock type and the other operating parameters. Matrix contact with the rock surface, "clogging" of the diamonds, time-dependency, and strength of diamonds and the diamond-matrix bond play important roles in these phenomena. Comparison of rock drillability between that of microscale and fullscale bits shows a straight-line relationship indicating the possibility to predict drilling performance in the field. Uniaxial compressive strength, tensile strength, and Sklerograf hardness can be used as a preliminary, but not reliable guide, to predict drillability. Petrographic characteristics and the relative scale of diamond size to grain size of minerals affect rock drillability.
  • Item
    Thumbnail Image
    The major, trace and precious metal geochemistry of some Permian layered intrusions, Central Queensland
    Reeves, Shane John ( 1990)
    The Bucknalla Complex, previously known as the Westwood Layered Intrusion, is a small, 10 km2, layered, tholeiitic, mafic-ultramafic intrusion located 50 km southwest of Rockhampton that was emplaced into an active continental margin environment in the Permian. The complex comprises clinopyroxenites, olivine clinopyroxenites, wehrlites, troctolites, hornblende gabbros, gabbros, anorthosites, leucogabbros and dolerites. It is a saucer-shaped lopolith (2200 m X 6 km at maximum stratigraphic intersection) which intruded Lower Permian spilitic pillow lavas, cherts and tuffs of the Rookwood Volcanics during the Lower Permian. It has subsequently been tilted vertically and in a northeast direction. It consists of over 15 laterally discontinuous igneous units ranging in thickness from 1-50 m. Plagioclase is a cumulus phase throughout the intrusion while orthopyroxene is absent until the very uppermost levels of the stratigraphy. The chromium composition of magnetite analysed by electron microprobe has been found to mimic whole-rock mg# and is a good measure of the degree of fractionation of the rocks. Electron microprobe analyses of samples from two traverses perpendicular to layering reveal cryptic variation in the primary phases (olivine: Fo69-83; plagioclase: An54-97; clinopyroxene mg#: Cpx67-87) which is not a simple function of stratigraphic height. Background PPGE (Pd & Pt), Au, S and Cu values for the intrusion are high while IPGE (Ir & Ru) are low. A total of 120 analyses has produced the following range of values: Pd, 2-70 ppb; Pt, 3-40 ppb; Au, 1-20 ppb; Ir, 0.01-0.07 ppb; Ru, 0.2-0.6 ppb; S, 150-400 ppm and Cu, 40-600 ppm. Platinum, Pd and Au display good correlations with Cu, particularly at more elevated levels, while Ir and Ru are better correlated with whole rock Ni and Cr. Palladium, Pt, Au, Cu & S are elevated in rocks which have intermediate whole-rock mg# (47-60). These trends suggest that the PGE are, to some extent, controlled by fractionation and that the high melting point PGEs (Ir, Ru) were precipitated with the early crystallising phases, such as olivine and clinopyroxene, whereas Pt, Pd and Au were removed from the magma by sulphides. Mantle normalized metal plots for both the mineralized and unmineralized rocks of the Bucknalla Complex display similar trends. Both plots display the anomalous low Ir content, PPGE enrichment and the clear control of sulphides on the distribution of the PGEs and Au. The ratio Pd/Ir is extremely high (1800-9300) indicating extreme fractionation of the PGEs. These trends may, in part, reflect PGE abundances inherited from the source (i.e. relatively low degrees of partial melting) but were exaggerated by the extraction of the IPGE during the early stages of fractional crystallization and by the precipitation of a PPGE-enriched sulphide component. The Complex is known to host minor Pd-Pt-Au-Cu mineralization, disseminated throughout the intrusion. The mineralization consists of chalcopyrite and bornite and their alteration products digenite and covellite, electrum (Au-Ag alloy), Pd-As, Pd-Sb, Pd-S, michenerite (PdBiTe2) and sperrylite (PtAs2). A common host rock is olivine gabbro and the silicate minerals are generally fresh. The mineralization is considered to be primary magmatic for a number of reasons, foremost of which are (i) the clear association of the PGMs with intercumulus (magmatic-textured) fresh or relict Cu-sulphides and (ii) a continuum in Pd/Pt, Cu/Pd, and Cu/Pt ratios from background to mineralized samples which strongly suggests that the processes responsible for the enhanced PGE content of the Bucknalla Complex were also responsible for mineralization. In as much as the former must have been produced by magmatic processes it is concluded that the higher grade PGE-Cu-S mineralization was also caused by primary magmatic processes. A model is proposed in which mineralization is sporadically generated by influxes of small batches of PGE-rich S-undersaturated magma into a magma chamber in which the resident magma has reached S-saturation due to fractional crystallization processes. Other intrusions in the region, namely; the Eulogie Park Complex, the Fred Creek intrusion and the Boogargan intrusion, are not considered prospective for stratiform PGE mineralization due to their low background PGE tenor, low Pd/S and Pd/Se ratios and high S contents.
  • Item
    Thumbnail Image
    Leachate chemistry of two modern municipal waste landfills in Melbourne, Victoria
    Strudwick, Darryl Grant ( 1999-11)
    This study investigates the occurrence and chemical composition of leachate at Clayton South and Brooklyn Municipal waste landfills in Melbourne, Victoria. Both are ‘modern’ municipal waste landfills, being engineered and managed consistent with current regulatory requirements. These landfills accept only putrescible and solid inert waste, but not soluble chemical, hazardous, liquid or prescribed industrial waste. (For complete abstract open document) Analyses of an extensive range of chemical parameters reveals a complex mixture of inorganic and organic compounds, similar to those of international authors researching leachate chemistry. Dominant ions in these leachates are NH4+, Na+, HC03- and Cl-. Except for Fe, heavy metals are not present in significant concentrations (mostly