Congratulations to James Jolles who defended his dissertation on Friday, June 5, 2020
Advisor: Rebecca Lange
High-silica rhyolites (>75 wt% SiO2) are the most differentiated magmas on Earth and are known for explosively erupting in large (100-1000’s km3) quantities. Understanding the formation of these magmas is crucial, as the processes involved fundamentally reconstitute continental crust, and the western U.S. has been the site of two super-volcano eruptions of high-silica rhyolite in the past 800,000 years. This thesis focuses on the extensively studied—yet controversial—origin of the Bishop Tuff in California, which formed when >600km3 of zoned high-silica rhyolite was erupted in less than a week, forming the Long Valley caldera. Key to understanding the origin of this voluminous magma are tight constraints on its temperature and volatile content. Various phenocryst phases in the Bishop Tuff can be used for thermometry and hygrometry and, in all cases, requires accurate analyses of mineral phases, as well as valid tests of mineral-mineral (and mineral-melt) equilibrium.
Chapter 2 addresses a long-standing controversy in the literature over the application of Fe–Ti oxide thermometry to the Bishop Tuff. Analyses of ilmenite and titanomagnetite in each single pumice clast span a wide compositional range, which had been interpreted as evidence of disequilibrium. In this study, quantification of analytical error allows: (1) accurate identification of disequilibrium phases (rare); and (2) determination of the minimum number of analyses required for high-precision thermometry (< ±10 °C). The resulting Fe–Ti oxide temperatures (~700-800°C) show excellent agreement with two-feldspar temperatures. Application of plagioclase-liquid hygrometry provides a high-resolution record of decreasing H2O content (~7.2-4.1 wt%) in the Bishop rhyolite with increasing temperature. Documentation of linear compositional variations in phenocrysts with temperature confirm that their growth occurred after thermal and compositional gradients in the Bishop rhyolite melt were developed.
Chapter 3 examines the origin of the strong variations in different trace element concentrations with temperature in the high-silica rhyolite portion (>400 km3) of the Bishop Tuff. It is hypothesized that these gradients arose from the segregation of variable melt fractions from a granitic crystalline mush. This hypothesis is tested by constraining the stoichiometry of the crystallization/melting reaction in the crystalline mush by determining the relative compatibility value of each trace element and combining it with their respective partition coefficients (between all mineral phases in the granitic mush and high-SiO2 rhyolite). Constraints are additionally placed on the pressure (350-500 MPa) and average activity of H2O (~0.5) during melt segregation. Several unresolved paradoxes in the literature are now explained, including: why fO2 increases with temperature across the Bishop magma; why H2O decreases with temperature; and why plagioclase phenocrysts are subordinate to quartz and sanidine.
In Chapter 4, focus shifts to the source of heat and H2O required to form the Bishop rhyolite, namely the Long Valley basalts, which erupted in close spatial and temporally proximity. It is established that phenocryst growth in the basalts was rapid, and that the most Mg-rich olivine analyzed in most samples (eight of nine) passes the liquidus test when paired with a liquid composition that matches that of the whole rock. When olivine-melt thermometry/hygrometry is applied to this pairing, resulting temperatures range from 1170–1060˚C and the average minimum water content is ≥2.9 wt% H2O. These new data are necessary to accurately model how the influx of basalt into the crust beneath Long Valley caldera led to the formation of the Bishop rhyolite through the transfer of their heat and water.