V51C-0705
Record of Hybridization Preserved in Zircon, Aztec Wash Pluton, NV
The mid-Miocene Aztec Wash pluton comprises a smaller granite zone and a larger, highly heterogeneous zone in which evidence for interaction between basaltic and granitic magmas is ubiquitous. Granitic rocks in both zones show textural and compositional evidence for crystal accumulation and melt fractionation. In the heterogeneous zone, basalts have chilled, crenulate margins against granitic rocks, and there is widespread evidence for mechanical contamination of each lithology (coarse resorbed alkali feldspar in fine-grained mafic rock; mafic enclaves in granite). "Grey rocks" of intermediate composition are exposed on dm to 100's of m-scale as enclaves, pods, and initially subhorizontal sheets. They are variable texturally, but most are dominantly fine- grained and equigranular. Textures of grey rocks are consistent with rapid solidification from melt-rich magma, and, in combination with isotopic compositions intermediate between felsic and mafic rocks of the pluton, suggest an origin by near-complete homogenization of a hybrid melt (Bleick et al. 2005; Ericksen 2005). The elemental chemistry of zircon preserves information about the evolving magmatic environment in which it was hosted (Claiborne et al., 2006). Owing to its slow dissolution rate, it has the potential to survive periods of undersaturation with only partial resorption. Thus, it may record drastic shifts in T and melt chemistry that would accompany mafic-felsic hybridization. We are investigating zircon zoning patterns by cathodoluminescence (CL) and elemental compositions by SHRIMP-RG to evaluate the record of processes that they preserve. Temperatures of zircon growth are estimated using Ti-in-zircon thermometry (Watson et al. 2006), assuming a(TiO2) of ca. 0.7 (sphene +/-ilmenite are ubiquitous). Zircons from the granite zone yield estimated T's of 700-860 C, whereas those from grey rocks range from 710- 910 C. While both granite and grey zircon populations show dramatic T variations among and within grains, fluctuation tends to be more common and extreme in grains from the grey rocks, where variations exceed 100°C. Such grains appear to have both survived and recorded a substantial heating event, as would be expected if either mixing or thorough mingling occurred. Zircons from the granitic samples almost invariably have euhedral, concentric, oscillatory zoning; some have truncations in zoning indicating resorption, and many have thin, CL-dark rims. Grey rock zircons typically display thick, concentric zoning or, less commonly, banding, are rarely oscillatory zoned, and many exhibit indistinct or anhedral internal zoning (corresponding to high calculated T). Many have thick CL-bright rims and resorbed surfaces. These rims apparently represent relatively cool (less than 800 C) late stage growth following the heating event recorded in grain interiors. In both granitic and grey samples, Hf (6500-13000 ppm) shows a strong negative correlation with calculated T. In all granite samples, U (100-3000 ppm) & Th (150-3500 ppm) concentrations likewise show a strong negative correlation with calculated T, indicating that both acted as incompatible elements in these magmas. U (50-2000 ppm) & Th (50-2000 ppm) compositions show equally dramatic, but far less systematic, variation in grey samples, with high concentrations at elevated T. Growth of "hot" U, Th-rich zircon at Aztec Wash appears to be limited to the mixing environment.
V51C-0706
Zircons From the Aucanquilcha Volcanic Cluster: Volcanic Sampling of an Evolving Batholith
The Aucanquilcha Volcanic Cluster (AVC), northern Chile, is a complex of about 19 volcanoes which has experienced episodic volcanism from ~11 Ma to present. Its spatial and chemical evolution through time suggests that the AVC is the result of repeated sampling of a growing batholith at depth. Though compositionally diverse in its beginning stages, spanning from basaltic andesite to rhyodacite, protracted AVC magmatism rapidly gave way to a relatively voluminous and homogeneous (dacite to rhyodacite) output after about 6 m.y. of activity. Four main pulses of volcanism have been identified: Alconcha (11-8 Ma), Gordo (6-5 Ma), Polán (4.5-2 Ma), and Aucanquilcha (1- 0.2 Ma). New zircon U/Pb ages and trace element data (SHRIMP) from six lavas from the AVC further constrain the chemical and temporal evolution of its magmatic underpinnings during progressive stages of development. We analyzed between 16 and 24 spots from each sample. Samples from different stages of AVC volcanism show varying crystal residence times as indicated by multiple peaks in the U/Pb age spectra. Only small populations of zircon ages coincide with the 40Ar39Ar eruption age of each sample, and three samples have dominant age populations of zircon grains that are a few hundred thousand years older than the 40Ar39Ar age. The greatest age spectrum observed (~2.5 my) was in the oldest sample analyzed (8.49 ± 0.09 Ma, error is 2σ hereafter), from the Alconcha group. Minor populations (8.97 ± 0.32, 9.44 ± 0.44, and 11.09 ± 0.62, all Ma) demonstrate widespread inheritance of presumably antecrystic grains. Four samples are from the Polán group (with ages of 3.86 ± 0.08, 3.53 ± 0.25, 2.92 ± 0.14, and 2.56 ± 0.10, all Ma), which corresponds to the sudden increase in AVC output. These samples show more restriced age spectra within samples (300 to 500 Ma), indicating that the time of homogenization and increased volumetric output rate corresponds to a time of substantial resetting of the zircon ages, presumably through pervasive dissolution and recrystallization of new populations. Because the Polán samples come from a time of peak AVC volcanism, it is plausible that they reflect a series of hot magma chambers whose crystal cargo had largely been equilibrated with more recent magmatic events. Preliminary trace element data (relatively low Yb/Gd relative to Th/U) from zircons support the interpretation that a more limited spectrum of age populations reflects a hotter differentiation history. A sample from Volcán Aucanquilcha (recent AVC activity when volumetric output rates were again low) yielded a dominant age population at 1.22 ± 0.05 Ma, but with a complex age spectrum (~2 my) ripe with antecrystic grains. Three Mesozoic grains and one early Tertiary grain from this sample are interpreted to be xenocrystic. The age data suggest that crystal entrainment was common in the AVC magmatic system and that most entrainment involved recycling of magmatic phases only a few hundred thousand years older than the eruption age. The diversity of recycled material appears greatest at times of high thermal contrast, that is during beginning and waning phases of magmatism.
V51C-0707
Magmatic construction of the Searchlight magmatic system (Eldorado-Newberry Mountains, Nevada) as revealed through zircon geochemistry and Ti-in-zircon geothermometry
The Miocene Searchlight pluton and coeval volcanic rocks in the Colorado River Extensional Corridor, southern Nevada (USA) provide an excellent record of the growth of a large subvolcanic magmatic system. Mapping, geochronology, geochemistry, and isotopes have established a clear petrogenetic relationship between volcanism and growth and solidification of the Searchlight pluton. The main quartz monzonite pluton grew by repeated intrusion of trachydacite/trachyandesite magma from about 17.0 Ma to 16.2 Ma, with periodic venting of crystal-rich magma as flows and domes. The pluton then underwent fractionation to granite, with eruption of rhyolite as domes, flows, and tuffs (16.2 to 15.8 Ma). This final period of magmatism was also accompanied by new mafic input. Trace elements in zircons from 3 pluton samples (felsic quartz monzonite to high-silica granite), 1 trachydacite dike, and 2 volcanic rocks (a trachydacite and rhyolite that mark the transition from intermediate to silicic volcanism) show patterns typical of igneous fractionation, including deepening of Eu anomalies, higher Hf, and lower Th/U with decreasing temperature as estimated by Ti-in-zircon thermometry. Zoning observed in CL images and Ti-in-zircon temperatures indicate that zircons had diverse and often disparate histories. Zircons from the plutonic samples and the rhyolite show the greatest range of temperature distribution among the analyzed samples, ~150-200°C. Conversely, zircons from the trachydacite dike and trachydacite lava flow show a narrower temperature range ~90-100°C, and for the trachydacite nearly all zircons give temperatures within ~50°C (excluding one analysis). Zircons from the pluton have lower temperature rims surrounding higher temperature interiors and show mainly oscillatory zoning but the range of variation between grain interiors and rims varies dramatically; in some cases <20°C (within uncertainty), and in other cases ~100°C. In contrast, zircons with rounded, resorbed cores surrounded by thinner, euhedral rims are common in the two volcanic samples, and temperatures of the rims are invariably 50- 100°C higher than core temperatures for these zircons. We interpret the data above to indicate that zircons grew in and were transferred between diverse thermal and chemical environments during the growth and solidification of the pluton. Resorption and pronounced temperature increase recorded in rims of zircons from the volcanic rocks likely indicates thermal rejuvenation of the magma chamber accompanied eruption.
V51C-0708
Pervasive and Persistent Large-Volume, low delta 18O Silicic Magma Generation at the Yellowstone Hotspot, 12.7-10.5 Ma: Ion Microprobe Analyses of Zircon in the Cougar Point Tuff
High precision analyses of oxygen isotopes measured by ion microprobe in cores and rims of 111 zircon crystals from the voluminous Cougar Point Tuff (CPT) of the Bruneau-Jarbidge eruptive center in southern Idaho, USA, confirm the presence of the largest low δ18O silicic volcanic province yet discovered (c.f. Boroughs, 2005). These data provide an unprecedented, high resolution view into details of large-volume magma generation and source materials at a major eruptive center of the Yellowstone hotspot. The CPT consists of at least ten members of high-temperature (800-1000 oC) ignimbrites erupted from 12.7-10.5 Ma with a collective eruptive volume estimated at ≥ 7000 km3. Analyses of zircon were performed on the University of Wisconsin Cameca IMS-1280 using a spot diameter of 10 microns, with an average analytical precision for δ18O on the KIM5 standard of 0.3‰ (2 s.d.). Core and rim spot locations were carefully selected using cathodoluminescence and secondary electron imaging to evaluate zoning patterns, identify core locations and avoid inclusions. All but two zircon grains display rims that are depleted in 18O, and ~60 % of crystals display core-rim zonation of 0.5 to 7.4‰. The total range of δ18OVSMOW in cores of the 109 zircons with low δ18O rims is from -1.8 to 6.4 ‰; rims of these zircons range from -3.1 to 3.3 ‰. Within individual units, cores of zoned zircons are generally more variable than rims and vary over a minimum of 1.5 ‰ (Unit IX) and up to 8.2 ‰ (Unit VII). Rims vary 0.4 (Unit IX) to 5.2‰ (Unit XII). Our data indicate that the vast majority of zircons (96%) grew in low δ18O crustal melts, and that multiple discrete lowδ18O sources were involved in magma generation for most eruptive units. Whereas zircon cores display the most variable compositions, rims in several units converge to a more restricted range of δ18O values, although all units but IX show rim variability > 0.8 ‰. The presence of multiple discrete magma volumes at the time of eruption has been documented previously by polymodal assemblages of glass and pyroxene compositions in several units and is consistent with the zircon δ18O data. Temporal patterns of variation in δ18O over the sequence of eruptions are consistent with zircon core inheritance from unerupted residual magmas and/or remelting of solidified remains of earlier CPT magma reservoirs that were depleted in 18O by hydrothermal alteration. Oxygen isotopes in bulk quartz separates measured by laser fluorination range in δ18O from 1.3 to 4.5‰, and temporal variations in quartz track those of average δ18O in zircon rims in each unit, and thus are consistent with a comagmatic origin with zircon. From the fourth eruption in the sequence to the tenth, average δ18O in quartz (=3.4, sd=0.4) and zircon rims (=0.9, sd=0.5) does not change appreciably from one eruption to the next. Measured Delta (Qz-Zrc) values using δ18O in zircon rims range from 1.8-2.6, and most fractionations are consistent with predictions based on mineral thermometry. First order estimates of the averageδ18O of the total eruptive volume (i.e. 7000 km3) range from 2.4 to 3.4‰, making the Cougar Point Tuff the largest volume of low δ18O magma known.
V51C-0709
Late Pliocene-Holocene explosive volcanism and glacial history in Kamchatka: major calderas, oxygen isotope depletions, and deep sea sediment record
We present results of a study of explosive silicic volcanism from Late Pliocene to Holocene caldera-forming eruptions in Kamchatka arc, northwest Pacific. Mid-Pleistocene to Holocene calderas are well expressed in the topography and are present both in the front and rear volcanic arc. Due to uplifts and erosion, calderas older than 1My are not preserved morphologically but thick intracaldera ignimbrite fills and reconstructed caldera outlines provide rough estimates on the large, >>100km3 eruptive volumes. Major tephra and ignimbrite deposits were dated by a variety of techniques: 14C, K-Ar, and Ar-Ar. Reconstruction of space images and extensive long-term field observations permitted identification of moraines belonging to at least two last glacial stages that are morphologically preserved and are intercalated with many caldera deposits. Older moraines may be also present but are more difficult to relate or date. Kamchatka was not covered by a peninsula-wide glaciation, but the majority of volcanic edifices and calderas served as centers of glaciation. Investigation of marine record of ash deposits from sites in the northern Pacific display increased explosiveness starting from 2.65 Ma that is close in time to the onset of the Northern Hemisphere glaciation. Thus, it appears that glacial unloading plays little role in causing explosive silicic volcanic eruptions, but the availability of meteoric water seems to promote hydration of the upper crust by low-d18O synglacial waters in calderas. Detailed ash record in the Sea of Okhotsk demonstrates more explosiveness in interglacial cycles, the record in the Pacific does not show this trend. Radiocarbon-constrained on-land record of the Holocene explosive eruptions suggests retardation of major caldera-forming eruptions by several k.y. after the onset of the deglaciation. Based on the Iceland's case, we suggest that deglaciation promoted basaltic volcanism that remelted hydrothermally-altered crust to form large silicic magma chambers, and that it took several k.y. to form large silicic magma chambers. Oxygen isotopic values of phenocrysts within major caldera- forming eruptions and intracaldera products and reconstructed magma d18O values are variable from 7 permil to 3.4 permil, but low values predominate in Pleistocene. Holocene magmas are both normal and low-d18O that may reflect memory effects of the last glaciation. This suggests that silicic volcanic rocks are derived from the upper hydro thermaly-altered crust, altered by synglacial meteoric waters.
V51C-0710
Using Oxygen Isotopes of Zircon to Evaluate Magmatic Evolution and Crustal Contamination in the Halifax Pluton, Nova Scotia
Oxygen isotope analysis of zircon (Zrc) is well suited for parsing out the magmatic history in granitoids. The Halifax pluton is the largest pluton (1060 km2) in the peraluminous South Mountain batholith. The Halifax pluton is mapped as a concentrically zoned body, with outer units comprising granodiorite, monzogranite and a mafic porphyry; these units are locally rich in metasedimentary xenoliths and magmatic enclaves. The exterior units surround a more felsic core of leucogranite [1]. Previous oxygen isotope studies of the pluton report high whole rock δ18O values that range from 10.7-11.7‰ [2], and indicate a significant supracrustal component in the source of the pluton. We report the first δ18O(Zrc) values from the Peggy's Cove monzogranite and an associated mafic porphyry. Samples were collected across 30 km of discontinuous exposures of the monzogranite. Values of δ18O(Zrc) vary from 7.71-8.26‰ (average = 8.15±±0.32‰(2 S.D.); n = 10). Small but systematic E-W regional variation in δ18O(Zrc) values suggests heterogeneous magmatic contamination within the monzogranite. Meter-scale magmatic enclaves, observed in close association with pods of diverse xenoliths and smaller enclaves at the western Cranberry Head locality, are slightly enriched in δ18O relative to the host monzogranite. These data combined support a model of magma mingling and heterogeneous mixing at the rim of the pluton, with contamination by high-δ18O rocks. Additional high-δ18O(Zrc) data from granodiorites on the northern margin of the Halifax pluton concur with these observations [3]. Typically, closed magmatic systems show increasing δ18O with SiO2 because more felsic magmas have a greater percentage of high-δ18O minerals such as quartz and feldspar. Thus, the Halifax pluton appears to exhibit an enrichment trend opposite of what would be expected of a closed evolving system. Emplacement mechanisms for the Halifax pluton proposed by previous workers suggest that the outer units intruded first, followed by the more felsic luecogranites at the core of the pluton [1]. Based on δ18O(Zrc) data, we propose a model in which early magmas were already enriched in δ18O from a metasedimentary source; during emplacement, the magmas mixed with and variably assimilated high δ18O(10-13‰; [2]) Meguma Group metasedimentary wallrocks and melts thereof, which increased the δ18O of the magma. The first magmas emplaced cleared the way for later, more evolved magmas to intrude without significant contact with country rock. Additionally, δ18O (Zrc) values are in disequilibrium with published whole rock and quartz δ18O values from the pluton, indicating that progressive contamination or subsolidus isotopic exchange elevated whole rock δ18O of the monzogranite after zircon crystallization. 1. M. A. MacDonald, R. J. Horne, Maritime Seds Atlantic Geol 24, 33 (1988). 2. F. J. Longstaffe, T. E. Smith, K. Muehlenbachs, Can J Earth Sci 17, 132 (1980). 3. R. M. Nowak, J. S. Lackey, J. W. Valley, GSA Abs (2007).
V51C-0711
Insights From in-situ U-Pb, Trace Elements and Hf Isotopic Analyses in Zircon From a Young, Incrementally-Filled, Monzogranitic Pluton
Ion microprobe U-Pb zircon analyses of the Miocene Monte Capanne pluton (Tuscan Magmatic Province, Elba, Italy) - one of the youngest granitic plutons exposed on Earth - reveal ages consistent with incremental intrusion of the three petrographically defined facies of the pluton (Dini et al. 2002). A positive correlation between U-Pb age and high uranium contents (up to 3%) compromises the apparently large age range (7-10 Ma) for the duration of magmatism and is attributed to an unconstrained matrix effect during sputtering. Selecting only zircon data with normal U concentrations, suggests timescales in the order of ca. 1 Ma (from 8.08 ± 0.09 Ma to 7.04 ± 0.29 Ma), i.e., of similar range to magmatic activity in the coeval Capraia volcano (from 7.80 ± 0.08 Ma to 7.10 ± 0.09 Ma), North of Elba. Importantly, zircon ages for the components of the pluton (mafic enclaves, dykes, granite porphyries) partially overlap, suggesting that zircon was transferred and/or recycled throughout the lifetime of the whole plutonic system. Much older inherited zircons also occur (from Archaean to Mesozoic) and are interpreted as being derived mostly from the wall rocks. Detailed BSE imaging has revealed a variety of textures (including patchy-zoning, small and large scale oscillatory zoning etc), associated with an extremely large range of trace to minor elements compositions (e.g., Hf: 7159-21284 ppm, U: 67-65319 ppm and Th: 0-46225 ppm). Intra-grain variations (zoning) in most elements (P, Th, U, Y and HREE) are interpreted to reflect changes in melt chemistry due to iterative mixing and recharge, where growth/resorption of zircon was also associated with reactions between other accessory minerals (particularly monazite, allanite and apatite; Dini et al., 2004). εHf(t) ranges from -14.95 to -4.25 and two populations are found in almost every component of the pluton, which concur with the magma mixing model. A positive correlation between Th and εHf(t) may be related to some wall-rock contamination (in agreement with the occurrence of xenocrystic zircon in most of magma products) and/or derivation from a strongly metasomatised mantle source during mantle-crust mixing (in agreement with previous works on the Tuscan Magmatic Province). Such a model involving mixing, crustal contamination and crystal fractionation is consistent with both chemical and isotopic zoning in K-feldspar megacrysts (Gagnevin et al., 2005a, b) and whole-rock geochemistry (Gagnevin et al., 2004). References: Dini et al. (2002), Geological Magazine 139, 257-279. Dini A. et al. (2004), Lithos 78, 101-118. Gagnevin et al. (2004), Lithos 78, 157-195. Gagnevin et al. (2005a), Journal of Petrology 46, 1689-1724. Gagnevin et al. (2005b), Geochimica Cosmochimica Acta 69, 1899-1915.
V51C-0712
Evaluating the Timing of Volcanism at Baitoushan Volcano (North Korea/China) in the Context of Open-system Effects: Insights from Sr, Nd, and Pb Isotopes at the Single Grain Scale
Baitoushan volcano, located along the North Korean-China border, is responsible for one of the largest caldera- forming rhyolitic eruptions in the northern hemisphere in the last 2000 years. In addition to an ~1000AD eruption, additional activity occurred at ~0AD and ~2000BC. These eruptions ejected large volumes of comenditic airfall and pyroclastic materials in addition to a small pantelleritic airfall deposit. We are at the initial stage of evaluating the sources of isotope variations, including whether these systems are open or closed, in order to evaluate the timing and residence of highly alkaline rhyolitic magma systems. Initial whole-rock isotope ratios suggest that young (<10ka) Baitoushan satellite basalts generally have slightly lower 87Sr/86Sr and higher 143Nd/144Nd than similarly young comendites, while the pantellerite has 143Nd/144Nd ratios similar to basalts. Sr isotopes in basalt hosted plagioclase crystals are generally higher than both whole rock hosts and accompanying clinopyroxene crystals, suggesting isotopic modification of basalts at crustal depths. These plagioclase 87Sr/86Sr ratios are similar to those in comendites and are consistent with early crustal inputs of either a limited amount of regional Archean basement or more extensive assimilation of Baitoushan "root" rocks with similar isotopic ratios. If limited, elevated Rb/Sr ratios in these rocks and resulting 87Sr/86Sr ratios in the comendites should be able to date magmatic residence while Nd and Pb isotopes remain constant. Such evaluations will be undertaken at the single grain scale where any potential of "open-ness" of the system can be constrained. Results will aid in determining the extent to which chronometric information is degraded by open-system processes leading up to large volume rhyolite eruptions.
V51C-0713
Integrating Trace Element Abundances and Derived Temperatures for Zircon and Titanite to Elucidate Petrogenesis of a Cretaceous Granodiorite, Southeastern California
Zircon and titanite are among the more important accessory phases useful for geochronology and geochemistry. With the advent, calibrations, and subsequent refinements of the Ti-in-zircon and Zr-in-titanite geothermometers, two powerful new methods have become available to track and integrate the compositional variations of these minerals during magmatic processes. Using the SHRIMP-RG, we measured REE, U, Th and other trace elements in co-existing zircon and titanite from a Cretaceous granodiorite from Joshua Tree National Park, California. Concurrent measurements of Ti in zircon and Zr in titanite, coupled with independent assessments of aTiO2, aSiO2} and pressure, allowed the estimation of derived crystallization temperatures TTi and TZr in zircon and titanite, respectively. Data indicate that crystallization of zircon and titanite occurred largely simultaneously, from close to 800 °C (the calculated zircon saturation temperature) down to ~700 °C, with continuing zircon crystallization down to ~650 °C. In zircon, concentrations of both Th and U vary somewhat irregularly with TTi, but with cooling the Th/U ratio decreases steadily from 4 to 0.2. In contrast, Th concentrations in titanite decrease sharply with decreasing TZr, while U contents show two separate trends that intersect at the lowest TZr (~700 °C). Above TZr = 750±10 °C, titanite Th/U decreases steadily from 14 to 4 with decreasing TZr, but below this temperature there is an abrupt change in Th/U to nearly constant values of 0.1-0.5. Total REE concentrations decrease in both zircon and titanite with decreasing temperature, but the observed europium anomalies (Eu/Eu*) illustrate complex, dissimilar trends for each mineral. In zircon, interior zones show constant or decreasing Eu/Eu* (at values of Eu/Eu* <1) with decreasing TTi while outer rims show an increase, but still never attaining Eu/Eu* >1. In titanite, hotter zones have Eu/Eu* <1, but Eu anomalies decrease in magnitude and become increasingly positive (i.e. to values of Eu/Eu* >1) in cooler zones. In addition, an abrupt change in slope observed in Eu/Eu* vs. TZr occurs with cooling at 750±10 °C, roughly the same temperature at which the zircon interior-to-rim Eu/Eu* transition is also observed. These data suggest a major physiochemical change occurred in the magma around 750 °C which affected multiple compositional parameters and was manifested in both zircon and titanite. Additional changes in chemical trends in zircon occur below ~700 °C, likely due in part to the absence of competing titanite crystallization. The evaluation of additional geochemical data from the other accessory and major minerals will help elucidate the nature of these and other magmatic processes.
V51C-0714
Magmatic Lensing: A Mechanism for the Growth of Large Crustal Magma Chambers
We present a model for the growth and thermal evolution of large crustal magma reservoirs, based upon the focusing of rising dikes towards a pressurized and buoyant melted region (the magma chamber). Dikes propagate orthogonally to the least compressive (local) principle stress, which in turn is determined by the regional stress trajectories created by the magma chamber and far field stresses. Therefore, there exists a capture radius at depth that is potentially much larger than the dimensions of the chamber, forming a system that we call a "magmatic lens." This phenomenon allows influx of magma to compete with the processes of cooling, crystallization and viscous relaxation of stresses to form large chambers. In time, these magma chambers may either 1) grow through melting and influx until excess overpressure causes eruption, 2) exist stably, or 3) freeze and shrink until magma lensing no longer occurs. We find analytical solutions for the stress fields of two-dimensional, viscoelastic, pressurized, and buoyant inclusions of circular and elliptic geometry, with and without a free surface, to model a lower or mid-crustal magma chamber. We then couple an algorithm for dike propagation with a thermal code that takes into account melting, crystallization, and a nonlinear melt fraction curve for both chamber and country rock to simulate magma lensing dynamically. This allows us to map the relationship between eruption-forming chamber overpressures, lower crustal estimates for melt flux, and the effects of regional extension.
V51C-0715
2D Lattice Boltzmann conduction-advection model with phase-change: Developing a new tool to assess melting/crystallization dynamics in magma chamber.
As the crystallinity of erupted volcanic deposits never exceeds ~50 volume %, a fundamental parameter for the prediction of volcanic risks is the evolution of the crystal fraction in a given magma reservoir. In such reservoirs, crystallinity varies in a non-linear way as a function of (1) exchange of heat with its surrounding (both addition of heat by magma recharge and loss through wall rocks), (2) convection-advection dynamics (silicate melt and gas flow in the reservoir) and (3) phase changes (melting/crystallization exsolution/dissolution). To better understand magmatic systems, the effects of all these parameters should be taken into account (multi- physics, multi-scale problem). Our goal is to develop a flexible numerical tool based on the Lattice Boltzmann method that is able to deal with complex interactions of the above cited phenomena. In this poster, we present the step by step development of our code and show tests of its efficiency on different volcanological cases. We are particularly interested in large silicic systems leading to caldera-forming eruptions.
V51C-0716
Rigidity anisotropy as a control for the formation of sills and the growth of laccoliths and other plutons
Despite a wealth of available field data, the mechanisms by which magmas stall in the crust and the physical processes leading to the formation of magma chambers and plutons remain unresolved. Field measurements of the geometry and dimensions of sills, laccoliths, plutons and batholiths suggest the existence of different growth processes related to the size of the intrusions considered. Many field and geochronological data also indicate that laccoliths, plutons and magma chambers develop and grow by amalgamation of numerous sills, and suggest a link between the time-scale associated with their growth and their size. In many cases sills appear to be the building blocks of larger laccoliths and plutons. Yet, the mechanics and dynamics of sill formation remain poorly understood. Different hypotheses for the emplacement of sills, proposed decades ago, have only recently been tested seriously. Analogue experiments involving the injection of fluid into a solid of gelatine reveal that under hydrostatic conditions the formation of sills requires the presence of layers of different rigidity, with sills forming only when their feeder dyke intersects an interface between an upper more rigid, stronger layer and a lower less rigid, weaker layer. That lithological discontinuities and rigidity contrasts can control sill formation provides a mechanism for the growth of laccoliths and plutons. Solidified sills provide favourable rigidity anisotropy for the emplacement of subsequent sills so that laccoliths and plutons can grow by over-accretion, under-accretion or even mid-accretion of successive sills. Supported by field data, this model predicts that laccoliths and plutons grow mainly by vertical expansion, representing the cumulative thickness of their internal sills, while maintaining a comparatively constant lateral extend. This model also predicts that the time-scale over which laccoliths and plutons form is essentially determined by the cumulative time between successive sill intrusions. The experiments also show that sill dynamics are controlled by viscous dissipation of the fluid along their length, which have consequences for their size and shape, enabling sills to propagate further and grow thicker than dykes of similar magmas.
V51C-0717
Thermochronology and Cooling Histories of Intrusive Suites: Implications for Incremental Pluton Assembly
Debate persists about the timescales and mechanisms of pluton emplacement and batholith formation. An understanding of whether plutons accumulate as large masses of magma or as an incremental series of pulses, in which the active magma body is small relative to the final pluton, is important for understanding the relationship between volcanoes and plutons. If volcanic eruptions < 1000 km3 are the most common size and large ignimbrites are rare, it follows that most plutons should record accumulations of small volumes of melt that were emplaced over long time intervals (millions to tens of millions of years) and therefore preserve predictable, protracted thermal histories. Modeling predicts observable differences in thermal histories of plutons and their aureoles that can be tested through thermochronology. Application of multiple chronometers (zircon and titanite U-Pb TIMS; hornblende, biotite, and K-feldspar 40Ar/39Ar; zircon and titanite (U-Th)/He) combined with K-feldspar multiple diffusion domain (MDD) modeling were used to determine the thermal history and to calibrate thermal models of two eastern California, Sierra Nevada batholith intrusive suites; the dike-like John Muir Intrusive Suite (JMIS) and the laccolithic Mt. Whitney Intrusive Suite (MWIS), and their wall rocks. Preliminary results of (U-Th)/He zircon data from the JMIS and its wall rock (the Tinemaha granodiorite) show a tight cluster of dates ranging from 75.6 to 70.4 Ma. The JMIS is thought to be mesozonal (8 to 11 km) and these data are interpreted as resulting from exhumation and additional 40Ar/39Ar data are required to determine if the thermal history reflects incremental intrusions. In contrast to the JMIS, preliminary (U-Th)/He zircon data from the MWIS and its wall rock (the Bullfrog pluton) show a wide range in dates ranging from 91.4 to 74.6 Ma that are interpreted to reflect reheating events. Amphibole 40Ar/39Ar inverse isochron dates, K- feldspar age spectra, and (U-Th)/He zircon data for two MWIS units indicate very rapid cooling through 500 ° C followed by protracted cooling of 3 to 10 million years through 180 ° C. Preliminary thermal modeling suggests that this pattern of cooling can be reproduced by top-down incremental emplacement of sheet-like bodies consistent with field relations. This sequence generates a pattern characterized by initial rapid cooling followed by subsequent heating thereby protracting low-temperature mineral closure in the host rock until the incremental intrusion stops. Two additional general insights can be derived from our initial results. 1) Single grain (U-Th)/He titanite analyses yielded unreasonably young dates relative to single grain (U-Th)/He zircons, suggesting the titanite diffusion parameters for individual samples and alpha-recoil corrections for titanite must be reexamined. 2) Not all samples show agreement between K-feldspar MDD modeling and (U-Th)/He zircon analyses suggesting a need for more data in order to correlate these two techniques.
V51C-0718
A Pb and Hf Isotopic Study for the Unzen Volcano
The Unzen volcano is one of the many active volcanoes in Japan. It has been active since at least 0.5 Ma, and the most recent eruption was in December of 1990. The Unzen volcano lies on the Shimabara peninsula of western Kyushu in southwestern Japan, and is within the Unzen graben that forms the northern tip of the Okinawa Trough. Two different models, mixing between an aphyric mafic magma and a felsic magma rich in crystals in the magma chamber, and assimilations of crustal materials during the ascend of a rhyodacitic magma, have been proposed to explain the evolution of the magmas for the Unzen volcano. In this study, Pb and Hf isotopic compositions of the lavas, erupted over the past 4000 years, from the Mt. Fugen of Unzen volcano have been analyzed, together with the existing Sr, Nd, and O isotopic data, in order to study the generation and evolution of the magmas for the Unzen volcano, and also to constrain the assimilation and fractionation crystallization (AFC) model proposed by previous studies. Six samples covering the eruptions from 1993 A.D. to 4000 years ago are included in this study. In general, the Pb isotopic compositions are consistent within analytical errors for all the samples, except for the systematic lower 6/4, 7/4, and 8/4 ratios for the two 1663 A.D. lavas, which also have the lowest Si contents among these samples. Similar variations between the Si content and Sr and Nd isotopic data have also been previously reported. Although the differences are smaller, Hf data also exhibit the similar variations as the Pb data for these lavas. The Pb isotopic data for the Unzen volcano are plotted very close to the MORB field in the Pb-Pb diagrams, except for the slightly elevated 7/4 and 8/4 ratios. Furthermore, the Hf and Nd data of these samples lie within the field of the mantle array in a Hf-Nd isotopic diagram. The coupled variations between Si contents and Sr and Nd isotopic data have been explained with an AFC type model. While the Pb and Hf data obtained in this study are generally consistent with such an AFC scenario, more data, especially the Pb and Hf isotopic data of the local crust in Kyushu and the older erupted lavas from the Unzen volcano are necessary in order to better constraint the generation and evolution in the magma chambers of the Unzen volcano.
V51C-0719
History of the Magmatic Feeding System of the Campi Flegrei Caldera
The definition of the magmatic feeding system of active volcanoes, in terms of composition, time-scale of crystallization, relation between composition of the erupted magma and structural position of vents, magma chamber processes and architecture, is of extreme importance for the hazard evaluation. The studies that are carried out for the definition of the magmatic systems include detailed mineralogical, geochemical and isotopic analyses (Sr, Nd, Pb). The Campi Flegrei caldera magmatic structure is characterized by deep and shallow magma chambers. In the deep reservoir (20-10 km depth) mantle derived magmas differentiate and are contaminated with continental crust. In the shallow reservoirs isotopically distinct magmas further differentiate, mix and mingle before the eruptions. These processes generated isotopically distinct components that were variably involved along different structures of the Campi Flegrei caldera during time. At Campi Flegrei caldera the relation between the structural position of the eruptive vent, for the last 14 ka of activity, and the isotopic composition of the emitted magma allow us to reconstruct the architecture of the magmatic feeding system and to infer the chemical and isotopic composition, and the magma chamber location and processes, of the future eruption, according to the position of the vent
V51C-0720
High-silica rhyolite magmatism in the Big Pine volcanic field, eastern California
The Quaternary Big Pine volcanic field (BPVF) located in the Owens Valley of eastern California is dominated by basaltic cinder cones and associated lava flows, but contains a single rhyolite lava erupted at circa 1 Ma. Despite its uniqueness, the petrogenesis of this rhyolite is poorly known. At nearby Coso volcanic field, an abundance of rhyolite relative to basalt suggests crustal melting by mafic magmas stalled in mid to upper crustal reservoirs, whereas the paucity of rhyolite relative to basalt at BPVF suggests only brief crustal residence of ascending mafic magmas (Mordick and Glazner, 2006). In order to determine the origin of rhyolite magmatism at BPVF (e.g., crustal melting versus extreme fractionation), we have examined the geochemical and petrographic characteristics of the Fish Springs high-silica rhyolite. The Fish Springs rhyolite comprises a single thick coulee with a volume of at least 0.05 km3 (DRE) of highly evolved (~76 wt.% SiO2) magma. The outer portions of the coulee are composed of autobrecciated and felsitic rhyolite, and internal portions, as exposed by quarrying, are pumiceous perlite with local obsidian. Fish Springs rhyolite is crystal poor (~1%), with small (<0.5 mm) phenocrysts of generally euhedral to subhedral plagioclase, sanidine, quartz, orthopyroxene, clinopyroxene, biotite, hornblende, Fe-Ti oxides, apatite, pyrrhotite, and zircon, as well as apparent xenoliths and xenocrysts of metamorphic and igneous wallrocks. Orthopyroxene phenocrysts show compositional zoning, with rims that contain higher Mg and lower Fe concentrations than cores. Trace element concentrations in Fish Springs rhyolite are characterized by very low concentrations of typically compatible elements such as Ba (~15 ppm), Sr (~8 ppm), La (~10 ppm) and Zr (~80 pm), as well as a pronounced europium anomaly, comparable to other high-silica rhyolites elsewhere in the Owens Valley, and suggesting high degrees of feldspar and accessory mineral fractionation. Samples taken from different portions of the coulee are essentially identical in composition, suggesting effusion of relatively homogenous rhyolite. Zircon saturation thermometry yields temperatures of approximately 730° C, comparable to highly evolved rhyolites at nearby Coso and Long Valley. Taken together, the relatively low temperature, multi-phase saturation, and low crystallinity of the Fish Springs rhyolite suggest an origin via fractionation of highly evolved melt from granitic mush, similar to high-silica rhyolites from voluminous caldera systems, rather than simply by localized crustal melting. Nevertheless, scattered xenoliths and xenocrysts record at least minor contamination of the rhyolite. Near-rim reverse zoning in orthopyroxenes suggests that heating and/or magma mixing affected the late history of the Fish Springs rhyolite, perhaps serving to catalyze magma ascent and eruption. Ref: Mordick and Glazner, 2006, CMP 152: 111
V51C-0721
Pervasive Crustal Melting on a Regional Scale: Sr-Nd Isotopic Evidence from Eocene Intrusions in NE Washington
During the Eocene the Pacific Northwest was the site of a short-lived but voluminous and geochemically diverse magmatic episode, commonly termed the Challis event. To investigate the origins of this event we have measured whole rock Sr and Nd isotopic compositions of 12 plutonic and hypabyssal samples, ranging from basalt to two-mica granite, collected along a 250 km transect across NE Washington. This transect crosses the 0.706 line (Armstrong 1977), the boundary between dominantly Mesozoic crust to the west and older crust to the east. The results reveal a wide spread in isotopic compositions (87Sr/86Srm = 0.7041 - 0.7262; εNdm = +3.8 to -18.5) with no systematic relationship between isotopic composition and bulk composition (e.g., MgO). This decoupling of isotopic composition and bulk chemistry suggests mixing between mantle and crustal melts was of minimal importance, and that these rocks are dominantly of crustal origin. The range in εNdm also indicates melting of crustal sources that varied considerably in age. Samples with εNdm > +2 range from basalt (13 wt.% MgO) to two-mica granite (0.3 wt.% MgO). Such juvenile εNdm in a two-mica granite precludes significant involvement of ancient metasedimentary material and implies rapid intracrustal differentiation of a mantle-derived source, which may have been deep arc crust of Mesozoic age. At the other end of the spectrum, samples with εNdm < -15 come from Springdale (87Sr/86Srm = 0.7071; εNdm = -18.5) and Medical Lake (87Sr/86Srm = 0.7143; εNdm = -15.2) at the eastern side of the study area. Data from these sites, both east of the 0.706 line, are similar to values reported for the nearby Silver Point Quartz Monzonite and attributed to melting of late Archean to Early Proterozoic crust (Whitehouse et al. 1992). Other samples analyzed in this study are broadly similar in isotopic composition (εNdm = +1 to -8; 87Sr/86Srm = 0.706-0.709) to rocks of the Colville Igneous Province and probably formed by melting of Proterozoic arc crust (Morris 2000). Geographic variability in Sr-Nd data indicates that isotopically distinct crustal domains are juxtaposed laterally and/or vertically, in some cases on a small scale. The sample with the highest 87Sr/86Srm (0.7262; εNdm = -13.3) is a dacite porphyry well west of the 0.706 line, while at Porcupine Bay adjacent plutons differ by almost 10 εNd units (-7.4 and 2.2). Ongoing work is designed to further characterize the crustal sources and better understand the nature of the thermotectonic event that drove such widespread crustal melting.
V51C-0722
Early Cretaceous Erlangmiao Metaluminous A-type Granite in the Eastern Qinlin Orogen, central China: Geochronological and Geochemical Constraints
It is possible to study the Mesozoic lithospheric thinning and crust-mantle interaction due to Early Cretaceous intensive magmatism widely developed in the North China Block. As an important part of the Early Cretaceous magmatism in the North China Block, the widespread Early Cretaceous granites in the Dabieshan-Qinlin Orogen, situated in the southern margin of the North China Block, contain hornblende and/or biotite and are I-type granites, but other type granites are less reported. Early Cretaceous Erlangmiao garnet granite (EGG) with garnet and biotite found recently near Erlangmiao in Fangcheng County, Henan Province, China, has provided new constraints for evolvement of the Dabie-Qinlin Orogen. This abstract reports results of zircon U-Pb dating, elemental geochemistry, and Sr-Nd isotopic compositions of the EGG and its wall rocks in an attempt to constrain its petrogenesis and the Early Cretaceous tectonic evolution of the Dabie-Qinlin Orogen. The EGG, located in the eastern Qinlin Orogen, was emplaced in gneissic monzogranites (GG). There are abundant GG xenoliths and surmicaceous enclaves in the eastern margin of the stock. LA-ICPMS zircon U-Pb dating of the EGG suggests its crystallization at 118 ±2 Ma, whole-rock Rb-Sr isochron age, represented the cooling age of the EGG, is 110±1 Ma. The EGGs are syenogranites with almandine-Spessartine garnet and biotite. The EGGs are characterized by high SiO2 and alkali contents, high FeOt/MgO ratio, low Al2O3 and CaO contents, and display high-K calc-alkaline and metaluminous characteristics. They show low total Rare Earth Element content (ΣREE), strong negative Eu anomaly, are generally enriched in Rb, Th, Ta, Nb, Zr, Hf, Y, Yb and depleted in Sr, Ba, Ce, P, Ti, and have high Rb, Rb/Sr and Ga/Al ratios. Al2O3, Fe2O3, K2O, Zr, Rb, Pb of the EGGs decrease with increasing SiO2 content. Initial 87Sr/86Sr ratios of the whole rocks range in 0.706-0.708, while the εNd(110Ma) values vary from -6.6 to -9.0 and Nd depleted mantle model ages are 1.5-1.7 Ga. Geochemistry and modal calculations suggest the EGGs are metaluminous A-type granites, underwent fractional crystallization of plagioclase, allanite and apatite, and formed in extensive background. We suggest the formation of the EGG is related to Early Cretaceous asthenosphere upwelling and lithosphere extension in Eastern China. The GGs-wall rocks of the EGG, crystallized at 128±2 Ma according to LA-ICPMS zircon U-Pb method, display metaluminous and high- K calc-alkaline characteristics. The GGs have high ΣREE, Sr/Y and (La/Yb)N ratios, show middle negative Eu anomaly, are enriched in Light REE, Rb, Sr, Ba and depleted in Ta, Nb, Zr, Hf, Ti, P, Y. The isotope is homogeneous, Initial 87Sr/86Sr ratios of the whole rocks are 0.707, the ε Nd(130Ma) values are -11.2 and Nd depleted mantle model ages are about 1.9 Ga. It suggested that partial melting of crustal rocks may form high-temperature A-type granites. Comparing the characteristics of the EGGs and the GGs, we suggest that partial melting of the GGs formed the EGGs influenced by asthenosphere upwelling.
V51C-0723
Contrasting Sources Of Granites In The Fosdick Migmatite Dome, West Antarctica
The Fosdick Mountains in Marie Byrd Land, West Antarctica comprise a migmatite dome 80x15km composed of paragneiss, orthogneiss and associated granites. The dome represents exposed middle-crust that experienced a polymetamorphic evolution associated with the active margin of Gondwana during the Devonian-Carboniferous and the Cretaceous. Research to date has focused on the metamorphic and structural evolution of the dome, with little attention to the origin and petrogenesis of the granites and migmatite leucosomes. Here, we report new geochemical and Sr-Nd isotope data for granites and migmatite leucosomes from the dome that bear on the source and petrogenesis of the parent melts (see also Saito et al., 2007, isaes.confex.com, 2.PS-38). We have obtained whole-rock major and trace element, and Sr-Nd isotope data for a suite of samples comprising paragneiss and orthogneiss, leucosome from paragneiss, and various types of granite from within the migmatite dome, and for possible protolith lithologies from outside the dome. Possible protoliths are Devonian biotite- and hornblende- bearing Ford Granodiorite and pre-Upper Ordovician Swanson Formation metapelites. Geochemical and isotopic compositions of the paragneiss and orthogneiss are comparable to those of the Swanson Formation metapelites and the Ford Granodiorite, respectively, consistent with previous interpretations regarding the protoliths of the gneisses (Siddoway et al., 2004, GSA SP380). Granites and leucosomes within the dome are silicic (71-78 wt % SiO2) and slightly peraluminous (1.02-1.18 alumina saturation index), and are granite sensu stricto in terms of normative mineralogy. The major element composition of granites and leucosomes is broadly similar to those of high-pressure experimental melts produced from granodiorite sources (Skjerlie and Johnston, 1993, J. Pet.34; Patin~{o}-Douce, 1997, Geol.25) and sedimentary sources (Montel and Vielzeuf, 1997, CMP128; Koester et al., 2002, J. Pet.43). Our data suggest that the granites and leucosomes likely formed by partial melting of the Ford Granodiorite or a sedimentary protolith represented by the Swanson Formation. In terms of trace element compositions, the granites in the dome may be grouped into high-Sr and low-Sr type. Our initial Sr-Nd isotope data suggest that sources of the high-Sr and low-Sr granites are the Ford Granodiorite and the Swanson Formation, respectively. The difference in Sr contents of two granite types may be attributed to different hydrate-breakdown melting reactions in the contrasting sources. Mica-breakdown melting will not contribute a significant amount of Sr to the melt, so we infer that the low-Sr granite likely formed by a mica-breakdown melting reaction in the metasedimentary source. In contrast, hornblende-breakdown melting or hornblende-biotite-breakdown melting of a granodiorite source may contribute Sr to the melt, and we infer the high-Sr granite likely formed by this process. Continuing research will investigate whether these attributes may be used for correlation with contemporaneous magmatism recorded in contiguous parts of the East Gondwana margin, particularly New Zealand (e.g. Tulloch and Kimbrough, 2003, GSA SP374) or Thurston Island (e.g. Bradshaw et al., 1997, Ter. Ant. Pub.).
V51C-0724
Crustal and Tectonic Controls on the Distribution and Composition of Large Silicic Calderas in Arc Settings
The study of silicic magmatism in arc settings is key to understanding felsic plutonism and thus how continental crust forms and persists. We undertook a global compilation of silicic calderas in arc settings in order to: (1) determine whether the distribution and composition of silicic magmatism at convergent margins is related to tectonic and crustal factors suggested by previous studies; and (2) examine the spatial relationship between silicic volcanism in arcs and the associated "main volcanic axis," defined by stratovolcanoes of intermediate composition. Geological and tectonic characteristics were compiled for 91 arc-related silicic calderas younger than 2 Ma and larger than 5 km in diameter. A measure of "caldera density" was calculated for each arc that quantified the number of silicic calderas per kilometer of arc length. In addition, an arc-normal distance was measured between each caldera and the associated main volcanic axis. There is a positive correlation between caldera density and trench-normal convergence rate, a trend likely related to magmatic flux. In addition, silicic calderas tend not to form in volcanic arcs with intense backarc extension. Composition of the caldera forming eruption is related to both the thickness and nature of the underlying crust. Thin, young or oceanic crust yields almost exclusively dacitic calderas. Rhyolitic calderas are dominantly located on continental crust thicker than 25 km and Mesozoic or older in age. Additionally, 77% of rhyolitic calderas formed under local extension, suggesting that this tectonic regime favors magmatic evolution. Calderas in general do not preferentially form at a specific distance behind the arc, but in arcs on young unevolved crust, calderas tend to occur within 10 km of the main volcanic axis; on older continental crust, calderas tend to be distributed over a wide area behind the axis. This trend is likely related to the difference in the overall width of the two types of arcs. The results of this study suggest that rate of magmatic input and the nature and stress regime of the underlying crust fundamentally control the distribution and character of silicic magmatism in arc settings.