Master of Science / Department of Geology / Matthew E. Brueseke / Low-[delta][superscript]1[superscript]8O silicic magmas are found in many volcanic provinces throughout the world, including the Snake River Plain-Yellowstone volcanic province (SRPY). The origin of SRPY low-[delta][superscript]1[superscript]8O silicic magmas is controversial, and centers on two disputed models: [1] a caldera collapse model that proposes reworking of the hydrothermally altered intra-caldera fill into the underlying silicic magma body, where each successive eruption lowers the [delta][superscript]1[superscript]8O of the magma eventually producing a low-[delta][superscript]1[superscript]8O magma and [2] melting previously hydrothermally altered mid-upper crust to form low-[delta][superscript]1[superscript]8O magmas. The mid-Miocene Santa Rosa-Calico volcanic field (SC) lies in northern Nevada. Brueseke and Hart (2008) described the geology and petrology of the SC, but did not deal with the [superscript]1[superscript]8O compositions of any locally sourced silicic magma. In the existing geological framework of the SC, this project aims to evaluate the two disputed models for low-[delta][superscript]1[superscript]8O silicic magma generation by analyzing the [delta][superscript]1[superscript]8O values of SC silicic eruptive products. Fifteen representative samples of locally erupted silicic units (e.g. ash-flow tuffs and lavas) were chosen for [superscript]1[superscript]8O analyses based on Sr-Nd-Pb isotope compositions, whole rock geochemistry, and field/temporal relationships. Each sample was crushed, sieved, and quartz and feldspar crystals were handpicked, described, and analyzed for their [superscript]1[superscript]8O compositions. Our results show that low-[delta][superscript]1[superscript]8O values exist in the SC and are limited to the youngest erupted silicic unit, the 15.8 to 15.4 Ma Cold Springs tuff, which was also the only unit erupted from a caldera. Cold Springs tuff [delta][superscript]1[superscript]8O feldspar values range from 2.36 to 4.05[per mil]; the unit is not quartz-bearing. Older silicic lavas that are not petrogenetically related to the Cold Springs tuff are characterized by normal [delta][superscript]1[superscript]8O feldspar values that range from 7.19 to 10.04[per mil]. Magma mixing models indicate that the source of the Cold Springs is a mixture of hydrothermally altered Granite Peak-Sawtooth granitoid and local mid-Miocene basalt, with an approximate range of [delta][superscript]1[superscript]8O values of 2-4[per mil], by fluids (with [delta][superscript]1[superscript]8O values ranging from -12[per mil] to + 7[per mil]) from the nearby hydrothermal system at Buckskin Mountain. This result follows the model by Boroughs et al. (2005) of prior alteration and melting, forming low-[delta][superscript]1[superscript]8O silicic magmas.

Subvolcanically emplaced granitic, miarolitic pegmatites associated with the White Mountain Igneous Province (WMIP), New Hampshire, were sampled and analyzed using modern analytical techniques including X-ray fluorescence, electron microprobe, scanning electron microscopy, and direct-coupled plasma spectrophotometry. Analytical results suggest that all the sampled miarolitic pegmatites from this study are petrogenetically related to the same intrusive suite, the WMIP. Based on the geochemical data, all the samples formed in an anorogenic tectonic setting and are rift-related. They are classified as NYF-type and plot in the “within plate granite” field on tectonic discrimination diagrams. The majority of the samples are peraluminous, A1-type granites. The trace element abundances on the spider diagram and chondrite-normalized diagram, which include a pronounced negative Eu anomaly and REE enrichments, are consistent with these miarolitic pegmatites resulting from a strongly fractionated granitic parental melts, but less fractionated than the classic NYF-systems such as South Platte (Simmons et al. 1987) and the Wausau Syenite Complex (Meyers et al. 1984).

In der vorliegenden Studie wurden kinetische Prozesse beim experimentell simulierten Aufstieg dazitischer Magmen, den Eruptionsprodukten vom Vulkan Taapaca nachempfunden, untersucht. Der Schwerpunkt lag auf der Entgasung der wassergesättigten Proben und der mit dem resultierenden Anstieg der Liquidustemperaturen einhergehenden Kristallisation von Plagioklas. Desweiteren wurden enthaltene Amphibole auf eine mögliche Reaktion als Folge der Dekompression untersucht. Die Entgasung während der Dekompression erfolgte bei einer Temperatur von 850°C in isothermen Dekompressionsraten von 6,3 bar/h bis 450 bar/h stets nahe an der Gleichgewichtskonzentration von Wasser, welche unabhängig durch Gleichgewichtsexperimente bestimmte wurde. So nehmen die Wassergehalte bei einer Dekompressionsrate von 50 bar/h von 4,93 Gew.-% bei 1550 bar zu niedrigeren Drücken kontinuierlich ab. Eine systematische Abhängigkeit von der Dekompressionsrate konnte für die Unterschiede in den jeweiligen Wassergehalten nicht gefunden werden. Die Anpassung der residualen Glaszusammensetzung während der Dekompression wird vor allem von der für die Reaktionen zur Verfügung stehenden Zeit kontrolliert, wie die in dieser Studie eingeführten Größen “Reequilibrationsindex” (REI) und “-geschwindigkeit” (RES) zeigen: Experimente mit Dekompressionsraten von 21,4 bar/h erreichen eine weitgehendere Reequilibration als die schneller dekomprimierten Proben, obwohl die letztgenannten RES aufweisen, die auf eine bis zu 10-fach schnellere Reaktionskinetik hinweisen. Dies kompensiert nicht die wesentlich kürzere, für die Reaktionen zur Verfügung stehenden Zeiten. Die Kristallisation von Plagioklas wurde (neben der Entgasung) als wichtigster Prozess während der Dekompression identifiziert. Weitere signifikante Reaktionen unter Beteiligung von Mineralen konnten nicht gefunden werden. Die Kristallrößenverteilungen (CSD) von Plagioklasen bei unterschiedlichen Dekompressionsraten unterscheiden sich signifikant bei 500 bar, wo langsamer dekomprimierte Proben eine steile CSD mit höheren Populationsdichten bei kleinen Kristallgrößenklassen erreichen. Bei 50 bar hingegen tritt dieser Unterschied nicht mehr auf. Wenngleich unterschiedliche Dekompressionsraten zur sehr ähnlichen CSDs bei geringen Drücken führen, so unterscheidet sich dennoch die Gewichtung mit welcher Keimbildung und Wachstum zur Kristallisation bei höheren Drücken im jeweiligen Experiment beitragen. Die Zusammensetzungen der Plagioklase aus Dekompressionsexperimenten sind im Schnitt An-reicher als die jeweiligen Kristalle aus Gleichgewichtsexperimenten. Während in Gleichgewichtsexperimenten bei niedrigen Drücken ein Anstieg der K2O-Komponente beobachtet werden kann, so tritt dies in den Dekompressionsexperimenten nicht auf. Die Kristallzusammensetzungen von Dekompressions- und Gleichgewichtsexperimenten unterscheiden sich demnach signifikant. In Amphibolen konnten keinerlei Reaktionen auf die Dekompression beobachtet werden. Die Zusammensetzungen der Amphibole bleibt konstant und es bilden sich keine Konzentrationsprofile von Kernen zu Rändern. Auch wurde keine Zerfallsreaktion  vorgefunden, obwohl während der Dekompressionspexperimente das Stabilitätsfeld von Amphibol früh verlassen wurde. Die Aufstiegsraten am natürlichen System des Vulkans Taapaca können anhand der vorliegenden Daten zur Entgasung, der Kristallisation von Plagioklas, sowie den Ergebnissen zu Amphibolen nicht eingegrenzt werden. Die Beobachtungen anhand von simulierten Dekompressionsraten von 6,3 bar/h bis 450 bar/h sind mit den natürlichen Proben vereinbar. Allerdings kann ein systematischer Fehler hier nicht ausgeschlossen werden, da sich das natürliche System vor der Eruption in einem Ungleichgewichtszustand befindet, der mittels der Experimente nicht rekonstruiert werden konnte. Darüber hinaus wurden Wasserlöslichkeiten in der dazitischen Schmelze bei 1000 bis 1250°C und 250 bis 2500 bar bestimmt. Bei 1200°C steigen die Wasserlöslichkeiten von 1,6 Gew.-% bei 250 bar auf 6,5 Gew.-% bei 2500 bar an. Mit steigenden Temperaturen nimmt die Wasserlöslichkeit bei konstantem Druck ab. Bei 1 kbar von 4,3 Gew.-% bei 1000°C auf 3,3 Gew.-% bei 1250 C, bei 2 kbar von 6 Gew.-% auf 5,5 Gew.-% im gleichen Temperaturintervall.

Master of Science / Geology / Matthew E. Brueseke / The Upper Wind River Basin (UWRB) is located in north-central Wyoming, to the south of the Yellowstone National Park boundary and east of Jackson Hole. Both Lava Mountain and Spring Mountain are Quaternary volcanoes in the UWRB. Lava Mountain is a shield volcano composed of 26 separate lavas capped by a scoria cone. Spring Mountain is located about ~36 km east of Lava Mountain, north of Dubois, WY, where eruptions of basalt cut through Paleocene and Eocene strata. The goal of this study aims to reconstruct the petrogenesis of magmas erupted at both volcanoes using geochemical, petrographic, and isotopic analyses. Important local events in geologic history played a large role in the development of the UWRB. This includes a long history of ancient and Cenozoic subduction, regional extension, and also the migration of the North American plate over the Yellowstone hotspot. The few previous studies on Lava Mountain claim the rocks are mafic in composition, however this was based solely on reconnaissance geological mapping. Geochemical evidence presented in this thesis show Lava Mountain rocks range from basaltic andesite to dacite. Basaltic andesite and dacite are interstratified at the base until approximately 2774 m; the rest of the volcano is andesite. All Lava Mountain samples are largely aphanitic and crystal-poor. Conversely, at Spring Mountain, localized normal faulting controls the location of eruptions of olivine-rich basalt. Petrographic analysis for both Lava Mountain and Spring Mountain display a range of evidence for open system processes, including sieved and/or resorbed pyroxenes, olivines and feldspars, as well as xenocrysts that suggest an influence from crustal assimilation. A petrogenetic model is introduced that discusses how Lava Mountain magma production occurred via fractional crystallization of basalt to dacite, then magma mixing of basaltic andesite and dacite, coupled with small amounts of crustal assimilation, to form the locally erupted andesites. All samples, including Spring Mountain basalts, have ⁸⁷Sr/⁸⁶Sr isotopes of 0.70608 and 0.70751, with ¹⁴³Nd/¹⁴⁴Nd isotopes of 0.51149 and 0.51157 and εNd values of -18 to -22. Pb isotopes plot to the left of the Geochron and directly on to slightly above the Stacey-Kramers curve. Strontium, neodymium, and lead isotope data suggest that Spring Mountain basalts are melts of ancient (e.g., 2.8 Ga Beartooth province) lithospheric mantle. The high ⁸⁷Sr/⁸⁶Sr values and exceptionally low εNd values separate the UWRB rocks from both Yellowstone and Snake River Plain volcanics, and suggest they originated from a different magma source. Finally, thermal evidence suggests melting genesis for UWRB rocks may not be Yellowstone plume related; rather it is more likely linked to Cenozoic extension.

Master of Science / Department of Geology / Matthew E. Brueseke / The Sonya Creek volcanic field (SCVF) contains the oldest in situ magmatic products in the ~29 Ma–modern Wrangell arc (WA) in south-central Alaska. The WA is located within a transition zone between Aleutian subduction to the west and dextral strike-slip tectonics along the Queen Charlotte-Fairweather and Denali-Duke River fault systems to the east. WA magmatism is due to the shallow subduction (11–16°) of the Yakutat microplate. New ⁴⁰Ar/³⁹Ar and U-Pb geochronology of bedrock and modern river sediments shows that SCVF magmatism occurred from ~29–18 Ma. Volcanic units are divided based on field mapping, physical characteristics, geochronology, and new major and trace element geochemistry. A dacite dome yields a ~29 Ma ⁴⁰Ar/³⁹Ar age and was followed by eruptions of basaltic-andesite to dacite lavas and domes (~28–23 Ma Rocker Creek lavas and domes) that record hydrous, subduction-related, calc-alkaline magmatism with an apparent adakite-like component. This was followed by a westward shift to continued subduction-related magmatism without the adakite-like component (e.g., mantle wedge melting), represented by ~23–21 Ma basaltic-andesite to dacite domes and associated diorites (“intermediate domes”). These eruptions were followed by a westward shift in volcanism to anhydrous, transitional, basaltic-andesite to rhyolite lavas of the ~23–18 Ma Sonya Creek shield volcano (Cabin Creek lavas), including a rhyolite ignimbrite unit (~19 Ma Flat Top tuff), recording the influence of local intra-arc extension. The end of SCVF activity was marked by a southward shift in volcanism back to hydrous calc-alkaline lavas at ~22–19 Ma (Young Creek rocks and Border Lavas). SCVF geochemical types are very similar to those from the <5 WA, and no alkaline lavas that characterize the ~18–10 Ma Yukon WA are present. Sr-Nd-Pb-Hf radiogenic isotope data suggest the SCVF data were generated by contamination of a depleted mantle wedge by ~0.2–4% subducted terrigenous sediment, agreeing with geologic evidence from many places along the southern Alaskan margin. Our combined dataset reveals geochemical and spatial transitions through the lifetime of the SCVF, which record changing tectonic processes during the early evolution of the WA. The earliest SCVF phases suggest the initiation of Yakutat microplate subduction. Early SCVF igneous rocks are also chemically similar to hypabyssal intrusive rocks of similar ages that crop out to the west; together these ~29–20 Ma rocks imply that WA initiation occurred over a <100 km belt, ~50–60 km inboard from the modern WA and current loci of arc magmatism that extends from Mt. Drum to Mt. Churchill.

The volcanic Island of Fogo that is situated near the end of the southern island chain in the archipelago of Cape Verde is one of the most active oceanic volcanoes in the world. The purpose of this study was to investigate temporal variations in lavas from Fogo by using whole rock major and trace elements and describe trends in magmatic differentiation, melting systematics and mantle source compositions. Further, for the first time Fogo lavas have been investigated by using 18O isotopes, to test for crustal assimilation and source variations. The time span for this study includes one sample of unknown prehistoric age and then from 1799 and onward all eruptions including the most recent in 2014-2015 are covered. The Fogo lavas are mafic, alkali rich, porphyritic, basanite-tephrites dominated by clinopyroxene phenocrysts followed by olivine and opaque phases situated within glass and microcrystalline groundmass.  Fractional crystallization and phenocryst accumulation is important mechanisms for magmatic differentiation, and the best result from fractional crystallization modelling gave a crystallizing assemblage of 20.7 % clinopyroxene, 5.4 % olivine, 4.2 % Fe-Ti oxides and 0.8 % apatite. Temporal variations in MgO content reveal a trend that change from more evolved lavas in older historic eruptions, to more primitive compositions in lavas from the middle of the 19th century, back to more evolved compositions in recent eruptions. Oxygen isotopes of glass and microcrystalline groundmass vary within a normal upper mantle range (δ18O = 5.7±0.3 ‰) which rule out any significant crustal assimilation, and the variations observed for the oxygen isotopes are probably related to source variations as there are tendencies towards covariation with source lithology proxies. Trace element ratios together with high TiO2 contents indicate that pyroxenite is an important source lithology. The Fogo lavas are highly enriched in incompatible elements and utilizing a simple batch melting model suggests ~2-4 % degree of melt for most sampled eruptions. In addition, melting is occurring at high pressures in the presence of residual garnet as the REE are fractionated. Trace element ratios of Ce/Pb and Zr/Y in Fogo lavas are consistent with mixing of a HIMU-like and an EM 1-like mantle source. There are also indications of an increasing involvement of the EM 1-like source through time. Magma mixing is indicated by petrographic observations such as complex zonation patterns and the occurrence of rounded cores in euhedral clinopyroxenes, as well as inferred from a large scatter in certain trace element ratios. / Vulkanön Fogo som ligger nära änden av den sydvästra ökedjan i Kap Verdes skärgård är en av de mest aktiva vulkanerna i världen. Syften med denna studie var att undersöka den kemiska variationen i lava från Fogo över tid och koppla variationen till olika magmatiska processer i vulkanen. Lava från Fogo har dessutom i denna studie för första gången undersökts med hjälp av syreisotoper för att testa för inblandning av material från jordskorpan samt för variationer i manteln i jordens inre där lavan bildas. Denna studie omfattar en tid som sträcker sig från ett förhistoriskt icke tidsbestämt vulkanutbrott, samt från 1799 till och med det senaste utbrottet 2014-2015. Lava från Fogo i denna studie är basisk och relativt magnesiumrik samt rik på alkaliska ämnen och har en porfyrisk textur som innebär att större mineralkorn sitter inbäddade i en finkornig grundmassa. Detta är typiskt för ytliga vulkaniska bergarter som svalnat snabbt. Mineralsammansättningen i lavan domineras av silikatmineralen pyroxen följt av olivin tillsammans med vad som troligen är järn-titan oxider. Dessa mineraler sitter i en grundmassa som består av mikroskopiska mineralkristaller och av glas. Lavan från Fogo har spårämnen som tillsammans med en hög koncentration av titandioxid påvisar att pyroxenit smält i manteln där lavan producerats, till skillnad från peridotit som är vanligare förekommande. Lavan har producerats genom smältning i manteln under relativt högt tryck vilket är typiskt för den här typen av vulkanöar och är dessutom berikad i vissa spårämnen som beror på att bara en liten andel av manteln smält när lavan bildats. En enkel smältningsmodell beräknade att ca 2-4 % smältning av manteln kunde ge de observerade koncentrationerna av spårämnen. Öarna i Kap Verdes skärgård har en geografiskt varierande geokemisk karaktär som beror på att den underliggande manteln består av skilda mantelkomponenter med olika kemisk sammansättning. I lava från den södra ökedjan dit Fogo tillhör ingår en mantelkomponent som saknas i den övriga delen av skärgården. Spårämnen i lava från denna studie överensstämmer med en blandning av dessa mantelkomponenter, dessutom verkar det förekomma en tidsmässig variation av inblandning från respektive komponent. Det finns även bevis för att olika typer av magma blandats från t.ex. mikroskopistudier av mineraltexturer. Resultatet från de analyserade syreisotoperna visar på en variation inom vad som är normalt för övre delen av manteln (5,7 ± 0,3 ‰), vilket utesluter inblandning av större mängder material från jordskorpan. Den variation av syreisotoper som förekommer i lavan beror därför förmodligen på att manteln inte är enhetlig.

No description available.

Southeast Iceland exhibits a granophyre pluton called Reyðarártindur, which has never been described in detail before. The Reyðarártindur magma chamber formed 7.30 ± 0.06 Ma ago (Padilla, 2015). Glacial and coastal erosion expose the pluton, and a river cuts through the pluton roof and walls, revealing interaction between different blob-like structures of magma. The formation of magma chambers can take a very long time, it is therefore likely for several different magmas to interact. Incremental formation of different magma batches give rise to mixing and mingling in magma chambers. To understand when the magma mixing was initiated and the mechanisms controlling it, descriptive analysis were made to obtain textural properties of collected rock samples from the field. The purpose for this thesis study was to examine if there is a frequency size and shape distribution of the magma blobs and if the different magma blobs are systematically distributed across the river. Previous studies have inferred conduit locations and magma mixing processes through similar methods. Extensive field studies have provided all samples for this thesis. Five distinct, magma types were described and found to be interacting. There was one ‘host magma’ which the other four different magma types are exposed as ‘blobs’ within. The statistical analysis involved mapping the blob-like structures from photos taken with an Unmanned Aerial System (URA; drone), using the software Inkscape. The data and measurements for the blobs was collected and summarized in ImageJ. The data was then statistically analyzed in Excel, illustrating the frequency of the magma blob’s size and shape distribution in selected parts of the river. The results of the statistical analysis of the magma blobs showed that ~80 % of the blobs existed in a size interval between 0 – 0.1 m2 . This thesis provides a discussion about the implications of the blob distributions for magma chamber recharge and processes within this section of the magma chamber. The shape distribution analysis showed an indication for all the blobs to be more rounded and equant. This suggest that the magma mixing event probably happened at the same time, during a liquid phase. / På sydöstra Island återfinns en granofyrisk pluton kallad Reyðarártindur. Det är en magmakammare som aldrig tidigare blivit beskriven i detalj. Magmakammaren bildades för 7,30 ± 0,06 Ma sedan (Padilla, 2015). Plutonen har blivit exponerad genom glacial- och fluvial erosion samt att en flod skär igenom plutonens tak och väggar. Den eroderande floden exponerar olika fläckliknande strukturer av magmainteraktioner. Ett gradvist bildande av olika magmasatser har över lång tid gett upphov till en blandning i magmakammaren. För att förstå när magmablandningen initierades och mekanismerna bakom fenomenet, har en beskrivande analys gjorts för att ta reda på texturella egenskaper av insamlade bergartsprover från fält. Syftet med denna studie var att undersöka om det fanns en storleks- och formdistribuering av magmafläckarna samt om de olika magmafläckarna är systematiskt distribuerade i flodbädden. Tidigare studier har antytt var undersökta magmakanalsystemen kan finnas samt hur blandningsprocesser för magma går till med hjälp av liknande metoder. Redan genomförda fältstudier har samlat in allt råmaterial som ligger till grund för denna studie. Fem olika magmatyper har beskrivits och påvisades interagera. Det fanns en ”värdmagma” som de andra fyra magmatyperna var exponerade som fläckar i. De statistiska analyserna inkluderade kartering av de fläckliknande magmaformerna baserat på foton tagna med hjälp av en drönare. Tre olika mjukvaror användes för att samla in, mäta och analysera data; Inkscape, Image J och Excel. Resultaten från den statistiska analysen visade att ungefär 80 % av alla fläckar existerade inom ett areaintervall mellan 0 – 0,1 m2 . Denna studie innehåller en diskussion om implikationerna av magmafläckarnas distribution med avseende på magmaomladdning och -processer inuti denna sektion av magmakammaren. Den generella formdistributionen visade en indikation för att alla fläckar tenderar att vara mer rundade och kvadratiska. Detta antyder att magmans blandningsförlopp troligtvis inträffade vid ungefär samma tidpunkt, under en flytande fas.

The sheeted complex of the ~92 Ma Tenpeak pluton, in the Northern Washington Cascades crystalline core, forms a <1.5-km wide zone with a moderate, NE-dip at the SW margin of the pluton. Sheeted magmatic complexes, such as the one in the Tenpeak pluton, are common in plutons and represent examples of incremental growth of plutons. Though it is widely accepted that plutons are constructed incrementally over prolonged timescales of several million years, it is not clear if and to what degree individual batches of magma interact, the timing and size of each magma pulse, and the role, timing, and location of magmatic differentiation. This project uses a combination of field evidence, bulk rock chemistry, and mineral geochemistry to address the (1) role of magma mixing and fractionation, (2) constraints on the relative timing of magma differentiation, (3) diversity of mixing styles preserved, and (4) physical properties that dictate how individual batches of magma interact within this sheeted complex. Rock samples were collected throughout the complex from mafic, felsic, dioritic, thinly-banded, and gradational sheets. Field evidence shows a range of sheet contacts that vary from sharp to diffuse, strong prevalence of mafic enclaves, and localized cases of mechanical mixing in which plagioclase feldspars from a felsic sheet are incorporated into a mafic sheet. In general, sheet thickness increases farther from the contact with the White River shear zone. The bulk rock and mineral chemistry suggests that the felsic magmas in sheets formed independently from the more mafic and hybridized sheets. The composition of the felsic sheets cannot be modeling by binary mixing processes involving mafic and felsic magmas or result from fractionating the most mafic magmas. However, mass-balance calculations using a linear least-squares mass balance calculation and Rayleigh fractionation models indicate that it is possible to explain the range of felsic compositions by internal, crystal fractionation driven mostly by plagioclase crystallization (~40-58%). Negative Eu anomalies in amphiboles from the felsic sheets imply that plagioclase fractionation commenced prior to the onset of amphibole crystallization. With the exception of the most primitive mafic sheet sampled, the mafic and hybridized sheets represent variable proportions of the mafic parental magma and the range of felsic differentiated magmas. Efficient mixing that resulted in these mafic to hybridized magmas must also have occurred prior to mineral growth as the mineral chemistry reflects intermediate, mixed compositions. The bulk rock and mineral chemistry of the most primitive, mafic sheet suggest that it did not mix with any felsic magmas. However there is evidence that the mafic sheet underwent plagioclase fractionation prior to emplacement. This is evident by lower bulk rock Sr/Ba relative to calculated Sr/Bamelt of plagioclase that cannot be reconciled without removing ~40-58% plagioclase. In contrast to the felsic sheets, the amphiboles from this mafic sheet lack Eu anomalies implying that amphibole crystallization occurred prior to major plagioclase fractionation. Chemical evidence reveals that magma mixing played an important role in controlling the chemical composition of individual sheets and field observations suggesting that there was a range of mixing styles. Throughout the sheeted complex, there are localized sites of mechanical mixing where plagioclase phenocrysts from adjacent felsic sheets are mechanically mixed into mafic sheets. Evidence for mechanical mixing is present across both sharp and gradational contacts. This implies varying rheological and viscosity contrasts between different sheets, though in both cases crystallinity and viscosity appears sufficiently low to allow crystals to migrate across sheet contacts. Variability in sheet thickness and contact type suggests that the physical parameters (i.e. temperature, viscosity, rheology, and magma flux) of the system continue to evolve throughout the formation of the sheeted complex. Near the White River, sheets are thin and more heterogeneous but become progressively thicker (>302 m) and more felsic in composition up-section. The composition of plagioclase and amphibole is remarkably uniform in all of the felsic sheets suggesting that each sheet formed from an array of felsic parental magmas. Thicker, felsic sheets most likely reflect hotter conditions where larger magma fluxes could be accommodated or viscosity-temperature contrasts that were low enough to allow for efficient mixing between two adjacent sheets and therefore erase sheet contacts. / Graduation Date: 2013

No description available.