Given the growing interest in secondary carbonate deposits from polar regions as paleoclimatic proxies, this thesis evaluated if they could be effectively be used in paleoclimatic reconstructions. To facilitate comparison between studies, the cold-climate carbonate precipitates were classified into three categories: powders, crusts and speleothems. The carbonate powders include those that precipitated in relation to aufeis aggradation (cryogenic aufeis calcite) and in relation to the growth of annual and perennial ice formations in caves (cryogenic cave calcite). The carbonate crusts were further subdivided according to their lithic environment; those that precipitated on the upper surface of bedrock/clasts (i.e. subglacially precipitated calcite and evaporative calcite crusts); those that are located on the underside of clasts (i.e. active layer carbonates); and those that precipitated in rock outcrop fissures (i.e. endostromatolites). The speleothems consist of a group on their own because they are not restricted to polar regions and most are currently inactive due to the presence of permafrost. To determine if the secondary carbonates in polar regions can be used as reliable paleoclimatic proxies, the chemical and isotopic (18O/ 16O; 13C/12C) partitioning that is occurring prior to and during the precipitation of carbonates and the water from which they precipitated was examined. The cold-climate carbonate precipitates, which were collected in distinct geological settings (carbonated vs. non-carbonated), have a delta18O composition between -6.5 and 36‰ and delta 13C values in the -5 to 17‰ range. The measurement of the difference (Delta) in stable C-O isotope composition of actively forming carbonate deposits and of the water from which they precipitated provided valuable insights into the formative mechanism that led to their precipitation. It was found that carbonates that precipitated under equilibrium physico-chemical conditions (i.e. cryogenic aufeis calcite powders, cryogenic cave calcite pearls, subglacially precipitated calcite) had a delta13C value that is in equilibrium with that of the parent water, while their delta18O compositions were more variable, as it is in part controlled by the temperature of reaction, by the delta18O and calcite saturation state of the parent water and formative mechanism. By contrast, carbonate deposits that precipitated under non-equilibrium physico-chemical conditions (i.e. cryogenic cave calcite powders, evaporative calcite crusts), their delta18O and delta 13C values are highly enriched relative to that of the parent water due to the faster rate of reactions which precludes isotopic equilibrium to be reached. In the case of biologically precipitated carbonate deposits (i.e. endostromatolites), their delta18O composition reflects that of the parent water, while its delta13C composition was enriched over that of the parent water because bacteria prefer to metabolize the light C (12C) in the DIC pool. These findings have significant implications regarding the use of cold-climate carbonate precipitates in paleoclimate studies as the delta18O signature preserved in most carbonates have been modified by freezing or other processes prior to their precipitation, which will modify delta18O composition of the carbonates.

The Hoyle Pond mine is located 15 km northeast of the city of Timmins, Ontario, in the Porcupine Gold Camp. It is hosted in deformed mafic to ultramafic volcanic rocks of the Tisdale assemblage. Geochemical data and structural analyses show interesting relations between host rock control and mineralization. The Au-bearing quartz-carbonate veins of the Hoyle Pond mine are hosted in a south-facing homoclinal sequence of stacked meta-volcanic rocks that are concordant with the Hershey Lake and Central formations of the Tisdale assemblage. The rocks were subjected to numerous sub-greenschist metamorphic events; however, detailed mapping shows that the hydrothermal mineralization took place during D3 and D4 events, and was associated with isoclinal folding, shearing, and thrusting. The alteration associated with the gold mineralization consists of a sericite alteration, which is immediately enveloped by an albite alteration zone. The inner alteration zone is composed of sericite, Cr-mica "fuchsite", quartz, Fe-dolomite, arsenopyrite, pyrite, tourmaline and graphite, and has a concomitant enrichment in K2O, Cr, SiO2, CO2 , As, and S. The albite alteration zone is composed of albite, quartz and Fe-dolomite, and has an enrichment in Na2O, CO2 and SiO2. In addition, carbon enrichment is also associated with mineralization. A pervasive graphite alteration envelopes the mineralization, and is the result of the oxidation of organic matter associated with Au mineralization. Rare Earth elements, Al2O3, Zr, TiO2 and Y were relatively immobile with respect to alteration and can be used to identify rock type. However, K2O, Na2O, Cr2O 3, Rb, CO2, CaO, Eu, FeO, MgO and to a lesser extent La (LREE), were mobile during alteration. Fluid mixing occurred between oxidizing fluids with soluble Cr 6+ and other elements (e.g. K and Na), and reducing fluids carrying gold as a thiol-organic complex (Au-HSCxHx), and other species, (e.g. B and As). The most plausible source of Au, As and B were the meta-sedimentary rocks of the surrounding Porcupine assemblage, whereas the Cr was likely derived from the ultramafic and mafic volcanic rocks. Two generations of fault-fill veins are present at Hoyle Pond: those hosted in D3 structures at lithological contacts, and those associated with D4 structures and located at flow contacts and also on both limbs of isoclinal folds. Due to the intensity of alteration and deformation of the Tisdale assemblage rocks at Hoyle Pond, it remains an open question as to whether the permeability for the mineralization at flow contacts was primary in nature, the result of mechanical contrasts during deformation, or both. The V10b or "chickenfeed" unit of the Tisdale Assemblage hosted significant gold mineralization, and was formerly thought to be composed of pillow basalt. Detailed mapping and geochemical work demonstrate that it is a pillow-lobe dacite of the tholeiitic suite associated with abundant hyaloclastite. The unit had a high intrinsic permeability, competency and Fe/Mg ratio malting it an ideal host for gold. At a regional scale there are two types of Au deposits associated with tholeiitic volcanic rocks: Those hosted in mafic rocks in which C may be important in precipitating gold from hydrothermal solution, and those hosted in more felsic rocks wherein the high Fe/Mg stabilizes the formation of Au scavenging pyrite.

The aim of this dissertation, based on three papers, is to utilize novel analytical techniques to examine the potential of non-traditional geochemical proxies, such as calcium and boron isotopes, to address questions related to earth system science. The first paper presents analytical protocols for high-precision calcium isotope measurements using thermal (TIMS) and plasma ionization mass spectrometry (MC-ICP-MS). These are applied to determine calcium isotope abundances of the Late Mesozoic belemnites that, in turn, are used to reconstruct the isotope composition of paleo-seawater and the evolution of the oceanic calcium cycle. The second paper discusses in details the history of the oceanic calcium isotope budget during the Phanerozoic (last ∼500 Ma), which is based on the analysis of calcium isotope compositions of several hundred marine skeletal carbonates, mostly brachiopods, measured by the TIMS technique. The observed experimental record is simulated using a numerical model of coupled calcium/carbon/magnesium global cycles and the results are discussed in the context of changing fluxes of major cations to the oceans. The third paper aims to utilize boron isotopes to constrain temporal changes in pH of the Jurassic surface oceans, with the ultimate goal to evaluate the effect of seawater pH on the oxygen isotope (delta18O) composition of coeval belemnites, hence on the delta18O-based paleo-temperature estimates. Overall, the results of this thesis demonstrate the usefulness and limitations of calcium and boron isotope proxies for studies concerning the evolution of earth system, specifically oceans and the atmosphere, over geological time scales.

Terrestrial moisture fluxes associated with plant growth represent one of the largest movements of mass and energy in the Earth's outer spheres, yet the relative contributions of abiotic water vapor fluxes and those that are regulated solely by the physiology of plants remains contentious and poorly constrained. Herein, based on the interpretation of the stable isotopes of water as tracers of evaporation, a methodology for partitioning total evaporation (ET) into its constituent fluxes is developed. This methodology enabled the separation of evaporation from soils and water bodies (Ed), canopy evaporation (I n), and plant transpiration (T) and was developed for the Fly River watershed (∼76,000 km2), situated in a rainforest-dominated region of central New Guinea and then subsequently applied to thirteen other watersheds in North America, South America, Africa, and Australia. Estimates of the annual water balances for the selected watersheds suggest that T is the dominant water vapor flux in 'water-limited' regions, comprising approximately 67% of ET and 55% of P. Moreover, these regional estimates of T co-vary with P in a manner similar to that observed for independent estimates of net primary productivity (NPP), suggesting that similar to small-scale measurements of plant water-use, water vapor and carbon dioxide fluxes are inherently coupled and the flux of water vapor to the atmosphere during photosynthesis is several orders of magnitude larger than the corresponding sequestration of atmospheric carbon. Similar estimates of a partitioned terrestrial water vapor flux are scarce beyond this study and hence the results are informative, emphasizing the inter-dependency between solar radiation and terrestrial processes and the considerable uncertainty that is currently associated with these aspects of the Earth system. Although admittedly first-order in scale, the estimates offer a broadly conceptual perspective on the dynamics of energy exchange between terrestrial systems and the atmosphere, whereby the terrestrial carbon cycle is essentially driven by solar energy via the water cycle intermediary. In order to constrain the sources of dissolved constituents in the Fly River watershed, additional data pertaining to the carbon isotope composition of dissolved carbon in river was collected. Monthly samples were collected for a period of two years and higher-resolution data was collected over a period of several weeks in 2007. These data suggested that the dissolution of carbonate minerals in the watershed were the predominant source of dissolved constituents and that constituents related to the dissolution of silicate minerals were negligible by comparison.

Shallow groundwaters were examined in the Attawapiskat kimberlite region, James Bay Lowlands, Ontario, Canada in order to assess buried kimberlite impacts on shallow groundwater geochemistry. Kimberlite pathfinder metals consisting of light rare earth elements (LREEs), Ni, Cr, Ti, Ba, Rb, Cs, and the ratio of Mg/Ca are commonly elevated in peat groundwaters along sampling transects as far as 200 m outside of kimberlite margins. The principal mechanism in transporting these metals upwards from kimberlites, through up to 21 m of TSS and up to 4 m of peat, is upward movement of deep groundwater. This mechanism is inferred based on hydrogeology and geochemistry of groundwaters in the TSS and peat. Fractures along the boundaries between kimberlites and limestone are suggested to provide pathways for upward groundwater movement. Metal -- dissolved organic matter (DOM) speciation calculations using Visual MINTEQ 3.0 and the NICA-Donnan database were conducted for Nd, Ni, Ba, and K. These pathfinder metals were used, as they are analogs for trivalent and divalent transition metals, alkaline earth metals, and alkali metals, respectively. The model predicts almost 100% of soluble Nd, Ni, and Ba form complexes with DOM at sampling sites with little to no contribution from upwelling (minerotrophic) groundwater (i.e., dissolved organic carbon (DOC) concentrations = 40 to 132 mg/L, pH = 3.9 to 5.5, and log ionic strength = < -3). Comparatively, as little as 10% of Nd and Ni, and 0% K and Ba are predicted to complex with DOM under conditions with large contributions from upwelling groundwaters (i.e., DOC concentrations = < 40 mg/L, pH = 5.5 to 6.5, and log ionic strength = -3 to -2). Based on modeling calculations, competitive availability of DOM binding sites where ionic strength is elevated (upward migration of groundwater) is the dominant control on metal-DOM complexation. MINTEQ modeling calculations indicate soluble Ni and Nd are likely scavenged from solution under strong upwelling conditions (i.e. high ionic strength). Metal scavenging is likely due to increased adsorption onto precipitating ferrihydrite (based on logSIferr close to or greater than 0) coupled with decreases in metal fractions complexed with DOM. Total Ni and Nd concentrations are typically 5 times lower than waters with little to no upwelling and ferrihydrite saturation indices (logSIferr) strongly indicate precipitation (up to 5). Comparatively, the concentrations of dissolved Ni and Nd in peat groundwaters are elevated where they are predicted to dominantly complex with DOM, and ferrihydrite is highly under saturated (logSIferr -18 to 5). Metal complexation with DOM may effectively inhibits metal scavenging from solution. Groundwaters from TSS over limestone, from boreholes in limestone, and kimberlites have lower CH4 concentrations and variable Delta 13CDIC-CH4 values (55 to 70‰) consistent with biological methane oxidation. All but one CH4 sample in TSS groundwater over kimberlite indicates in situ formation. Less negative delta 2HH2O values (greater than the local meteoric water line) are typically those samples from TSS over kimberlites and suggest the production of CH4, as biological DIC reduction requires H+ ions from H2O to form CH4. The ratio of CH4 to the concentrations of Fe3+, SO42- and O2(aq) are greatest in the majority of TSS groundwater samples collected over kimberlites, and suggests kimberlites may indirectly influence the CO2-CH4 system by consuming oxidized ions in the overlying TSS. The results of this study suggest that buried kimberlites can be identified through the detection of high Ni, Cr, Ti, LREEs, K, Rb, Ba concentrations, and the ratio of Mg/Ca in peat groundwaters over or near the kimberlite margin. However, geochemical processes such as metal -- DOM complexation, adsorption, and incorporation of metals into precipitated minerals are important considerations when identifying the spatial distribution of pathfinder metals along transects. Variations in CH4 emissions occur in soils and groundwaters overlying kimberlites could be due to the release of reduced ions during serpentinization reactions. Methane gas coupled with C isotopic variations in CH4 and DIC may represent a possible pathfinder geochemical tool to use in conjunction with pathfinder metals. (Abstract shortened by UMI.)

In geochemical applications, data classification commonly involves 'mapping' continuous variables into discrete descriptive categories, and often is achieved using thresholds to define specific ranges of data as separate groups which then can be compared with other categorical variables. This study compares several classification methods used in applied geochemistry to select thresholds and discriminate between populations or to recognize anomalous observations. The comparisons were made using monte carlo simulation to evaluate how well different techniques perform using different data set structures. A comparison of maximum likelihood parameter estimates of a mixture of normal distributions using class interval frequencies versus raw data was undertaken to study the quality of the corresponding results. The more time consuming raw data approach produces optimal parameter estimates while the more rapid class interval approach is the approach in common use. Results show that provided there are greater than 50 observations per distribution and (on average) 10 observations per class interval, the maximum likelihood parameter estimates by the two methods are practically indistinguishable. Univariate classification techniques evaluated in this study include the 'mean plus 2 standard deviations', the '95th percentile', the gap statistic and probability plots. Results show that the 'mean plus 2 standard deviations' and '95th percentile' approaches are inappropriate for most geochemical data sets. The probability plot technique classifies mixtures of normal distributions better than the gap statistic; however, the gap statistic may be used as a discordancy test to reveal the presence of outliers. Multivariate classification using the background characterization approach was simulated using several different functions to describe the variation in the background distribution. Comparisons of principal components, ordinary least squares regression and reduced major axis regression indicate that reduced major axis regression and principal components are not only consistent with assumptions about geochemical data, but are less sensitive to varying degrees of data set truncation than is ordinary least squares regression. Furthermore, correcting the descriptive statistics of a truncated data set and calculating the background functions using these statistics produces residuals and scores which are predictable and thus can be distinguished easily from residuals and scores calculated for data from another distribution. / Science, Faculty of / Earth, Ocean and Atmospheric Sciences, Department of / Graduate

Doyle (1977, 1979), working on the Southern Canadian Interior Plain evaluated regional geochemical patterns based on parent material. He found that regional variation in the total concentration of copper, iron, manganese and zinc arose largely from differences among parent materials, rather than from more local (within parent material) differences caused by pedological factors. To extend Doyle's work, a more detailed local investigation of the within parent material variation of copper, iron, manganese and zinc on four parent materials was carried out in the Rosetown area, Saskatchewan. Principal objectives of this study were to find the causes and magnitude of downslope catenary changes in the geochemical pattern on each of four soil parent materials, utilizing DTPA extraction to indicate the availability of these micronutrients to crops. Soil samples were collected from soils developed on lacustrine clay (Regina Soil Series), lacustrine silt (Elstow Soil Series), glacial till (Weyburn Soil Series) and aeolian sand (Dune Sand Soil Series). Five sites were selected for each parent material and at each site, five pits were dug at intervals downslope. Laboratory analysis for the entire sample set included the determination of pH, organic matter, copper, iron, manganese and zinc concentrations for soil digested with 4:1 HNO₃/HClO₄ mixture and the same metals extracted with DTPA solution. Further analysis comprised sequential extraction of copper, iron, manganese and zinc, particle size separation and X-ray diffraction. Highest total elemental concentrations are found in the Ap horizon of the Rego Dark Brown Chernozems developed on lacustrine clay, followed by lacustrine silt and glacial till soils, with soils on aeolian sands having the lowest values. The A horizons of lacustrine silt soils contain the highest DTPA extractable concentrations of iron, manganese and zinc, whereas, maximum extractable concentrations in the C horizons are associated with lacustrine clays. DTPA extractable copper in both A and C horizons is at a maximum in lacustrine clay soils. For the majority of sites, the highest total and DTPA elemental concentrations occur at the base of the slope, this being most marked for lacustrine silt soils. Total elemental concentrations for the four parent materials exhibit a relatively greater uniformity when considering both trends downslope and down profile than DTPA concentrations. A much greater proportion of DTPA extractable manganese and zinc occurs in the organic rich surface horizons compared to the more alkaline C horizons. This is also found for copper and iron but to a lesser extent. Analysis of variance shows that the compositional variation among parent materials for total elemental data, accounts for well over 50% of the overall data variability. Duncan's New Multiple Range test results further substantiate these textural groupings into lacustrine clay, lacustrine silt and glacial till and aeolian sand. Results are less conclusive for DTPA elemental data. However, even though soil copper, iron, manganese and zinc are influenced by many pediological factors operating separately and jointly, a large percentage of the total variability when predicting DTPA elemental concentrations can be accounted for by the variables included in the regression equations. The Index of Determination (I ) shows that the variability in DTPA elemental concentrations is best accounted for by the regression equations for the lacustrine clay soils with 93-98% of the total variability explained. / Land and Food Systems, Faculty of / Graduate

This study involves an examination of geochemical data from the Tatara-San Pedro Volcanic Complex (TSPVC), an approximately 250 km2 Quaternary frontal volcanic arc center located at 36° S, 70.5° W in the Southern Volcanic Zone (SVZ) of the Chilean Andes. Every lava flow in almost thirty, mostly vertical, stratigraphic sections was sampled, and used in conjunction with field observations, K-Ar and 40Ar/39Ar ages, and paleomagnetic data, to construct generalized chemo-stratigraphic columns. These columns not only demonstrate changes in lava composition with time, but were used to reveal the sequence of petrogenetic processes that formed those lavas. Geochemical variation both within a single lava flow and amongst members of an eruptive package of lavas was established, using lava flows which had erupted at different times, and were characterized by different mineralogical and geochemical features. Paleomagnetic and geochemical data were also used to correlate lava flows and packages of flows erupted onto different geographic regions of the complex. Petrogenetic modelling suggests that magma supply rates may be the dominant parameter controlling lava compositions; high rates of supply produce compositionally monotonous sequences of mixed basalt and basaltic andesite, which largely comprise the main volcanic edifices. The final-erupted lavas in many of these sequences are the least contaminated and may represent the best proxies for recharge magma compositions. During periods of intermediate magma supply rates, increased cooling and crystallization produces greater compositional range between mixing endmembers, and mixed andesites are the result. Low supply rates are associated with either single, short-lived flank eruptions of mafic lava, or more evolved compositions derived independently from differentiation of magma within small volume chambers. The relationship between fractionated, erupted, and recharge volumes determines the nature of erupted sequences. The range of lava compositions at TSPVC, as the result of multiple sources, different crustal contaminants, and different petrogenetic processes, is almost as great as that exhibited by all volcanoes from the Southern Volcanic Zone. This suggests that the dominant control on SVZ compositions may not be crust:mantle ratio, but different petrogenetic processes and changing crustal contaminants with time.

No description available.

<p>Godkänd; 1994; 20080401 (ysko)</p>