Search

Magmatism has played an integral part in the evolution of the Earth through geologic time. Mantle-derived mafic magmatism is a primary mechanism by which the bulk silicate Earth segregated into mantle and crust, and the continental crust has further differentiated through metamorphism, anatexis, and assimilation. The study of magmatic processes has given rise to our understanding of how the Earth has evolved.Central to this understanding are chemical, isotopic, and geochronological studies. Chemical and isotopic studies provide unparalleled richness of information; this is especially true now, as technological advances extend our analytical and computational capabilities. The difficulties presented in geochemical field studies of magmatic processes are primarily twofold: we are mostly limited to surface samples at the end of their evolutionary story (while the processes of interest are those occurring from source to surface), and it can be difficult to disentangle the myriad causes that produce the chemical and isotopic effects observed. These difficulties can be addressed by combined geochemical-geophysical-geospatial studies, by computational modeling based on an extensive (and growing) body of thermodynamic data, and by application of novel geochemical techniques.As with chemical data, geochronological data are fundamental descriptors in their own right, but they are also crucial because they anchor rates. Almost all geologic processes or features are rate-dependent, from growth of crystals on an atomic scale, to morphologies of volcanoes, to the evolution of the crust. An understanding of chronologies and rates is therefore fundamental to understanding the processes of magmatic evolution.The diverse chapters in this thesis reflect my diverse interests in the Earth sciences, but all are motivated by desire to understand both fundamental aspects of the chemical evolution of magmas, and how that evolution informs our broader understanding of volcanoes, volcanic fields, and continental crust evolution.The first chapter is a geochemical study of the San Francisco Volcanic Field in central Arizona. In this study, I combine geochemical, geochronological, geospatial, and isotopic data with the computational thermodynamic modeling package alphaMELTS to understand the chemical evolution of the volcanic magmas from source to surface. I show that the isotopic and chemical characteristics of the SFVF can be largely explained by assimilation of the deep crust by parent magma derived from fertilized asthenospheric mantle. The computational treatment with alphaMELTS allows for quantitative limits to be placed on the mineralogy, chemistry, and dynamics of the source of the volcanic field.The second chapter is a geochronological study of magmas from the Hawaii Scientific Drilling Project phase 2 core. This study works to extend the U--Th/He dating technique to olivine phenocrysts in relatively undegassed submarine basalts, rocks which are normally difficult to date by other chronometers.The third chapter investigates the evolution of felsic magma chemistries and the changing nature of continental crust source rocks during the Himalayan collision. In this study, I apply relatively new (Mg, Ca) and established (Sr, Nd) isotopic techniques to a sample suite of diverse crustal lithologies (primarily granitoids and felsic volcanics) from the Lhasa, Tibet region. I show that Ca and Mg isotopes are sensitive to specific mineralogies and processes, including carbonate-bearing shale assimilation, and that the Ca and Mg isotopic systems are promising tools in studies of granite petrogenesis, metamorphism, assimilation, and subduction recycling.

Chemical speciation in molten silicates is of broad interest for understanding thermodynamic and transport properties. However, it is difficult to elicit the nature of dissolved species in silicate liquids, let alone the role that speciation plays in the diffusion process. This dissertation is about chemical diffusion in silicate melts; it combines experiments with isotopic measurements to infer the physical mechanisms of cation diffusion and the role that diffusion plays in producing isotope effects in nature. Diffusion of calcium (Ca) between natural volcanic liquids in diffusion-couple experiments leads to 44Ca/40Ca variations of ca. 5‰ due to a mass dependence of Ca diffusion coefficients. The efficiency of Ca isotope separation by diffusion varies with bulk liquid composition and depends also on the magnitude and direction of aluminum (Al) gradients. Some Ca isotopic fractionations seem to arise solely from gradients in Al. These observations indicate that isotopic discrimination by diffusion reflects the mass difference between isotopically-substituted molecular species, and there is evidence for at least one Ca-bearing and one Ca-Al-bearing diffusing species in the volcanic liquids that were studied. The inferred existence of a Ca-Al complex is consistent with the additional observation that Ca diffuses slowly since it is inferred that Ca atoms interact strongly with their nearby Si and Al "solvent" molecules in the liquid.The third part of this thesis describes Ca diffusion between silicate liquids of simplified chemical composition. Results from these experiments indicate that the efficiency of diffusive separation of Ca isotopes is systematically related to the normalized diffusivity - the ratio of the mobility of the cation DCa to the mobility of the liquid matrix DSi. A similar dependence is observed for Fe, Li, and Mg isotopes, and in aqueous solutions. This empirical result provides a predictive tool that can be used to understand diffusive isotopic effects in a wide variety of geologic environments and a basis for a more comprehensive theory of isotope separation in liquid solutions. A conceptual model is presented for the relationship between diffusivity and liquid structure that is consistent with available data.