Specic gravity Specic gravity AB0 132 0 132 110° 30′W 2 Figure 1. A: Geologic map u B N

T of sample locations in ′ Welded B Yellowstone National Park, western United States, and 45° 00 surrounding region; modi­

Lava Creek fied from Christiansen (2001). The extent of Lava Creek deposits are

of meters } N ’s highlighted in green. Gray domains with curved lines Purple Mountain Y308 Y310 Y301 are post­ rhyo­ lites with pressure ridges

u A shown schematically T Section continues 1 Basal fall (Christiansen 2001). Blue Tu Cli deposit domains are lakes and

Depth (m) rivers. The remaining white

Lava Creek areas are undifferentiated recent alluvial sediments or pre­Lava Creek Tuff de­ Y305 posits. B: Schematic mea­ Caldera sured section shows rela­ tive stratigraphic position Section continues of pumice samples. Pum­ ice, lapilli, and gray bed­

ness varies across caldera from 100’s to 10 ness varies across caldera from 100’s ding planes are illustrated 0 schematically. Colors were

Thick chosen to best represent Flagg Ranch 10 km Talus the color of observed Lava Creek Tu B outcrop variations, but Y306 Lava Creek Tu A should also be considered Y307 Welded A Sample locations schematic. Talus

as from the unwelded basal fall layer of Lava the quartz crystals, and >80% of those are hol- tire crystals, creating hollow pathways that con- Creek Tuff B (Fig. 1). Basal falls would have low. Reentrants become slightly less common nect opposite faces (Fig. 2; Videos DR4–DR8). been the preferred material to collect for Lava upsection, appearing in ~20% of quartz crystals Many reentrants discordantly cut across primary Creek Tuff A, but outcrops remain unknown. from pumices in the middle and upper portions growth bands in quartz, which appear as alter- The pumices were crushed, sieved, and hand- of the ignimbrite. Reentrant filling is variable in nating light and dark bands in CL (Fig. 2). CL picked to make quartz mineral separates. Quartz those samples, with some primarily filled with images constrain the relative timing of quartz occur as euhedral bipyramids, or partial crystals glass whereas they are almost all empty in other growth and the reentrants. with some faceted faces. Almost every crystal samples. Quartz from the basal fall of Lava Creek Empty reentrants sometimes contain magne- contains glass-filled, enclosed melt inclusions. Tuff B contains reentrants in ~18% of the crystals. tite crystals up to 50 μm in diameter, as well as The crystals are also embayed with one or more Empty reentrants account for 0.02–1.5 vol% small pockets of glass adhered to reentrant walls reentrants. Many of these reentrants are empty of their host quartz crystals. They appear as em- (Fig. 2). Reentrant shape does not correlate with void space, not filled with glass. bayments or tubes ranging from a few microns its emptiness. In a given sample, both tortuous To establish the statistical significance of the to 400 μm wide that extend into the center of and simple-shaped reentrants may be hollow or empty reentrants across the Lava Creek Tuff, we the crystals (Fig. DR1 in the Data Repository). filled with glass. surveyed a few thousand quartz crystals and their They are locally bulbous with bulging interiors reentrants from pumice samples (Table 1). We that narrow to necks at the crystal surface. “Pen- DISCUSSION selected characteristic reentrant-bearing quartz etration” depth is highly variable, extending as To infer the significance of the empty re- crystals for synchrotron X-ray microtomogra- far as 1600 μm, although most range from 50 to entrants, we must consider how they came to be. phy (μXRT) and cathodoluminescence (CL) 500 μm. Rare embayments tunnel through en- Previous work regarding reentrants describes analyses to establish the relationship between the reentrants and crystallization processes (see TABLE 1. REENTRANT AND INCLUSION ABUNDANCES FOR LAVA the GSA Data Repository1). Reentrants are most CREEK TUFF A AND B, YELLOWSTONE, WESTERN USA common in the earliest-erupted Lava Creek Percent with melt Percent Percent empty inclusions reentrants reentrants Tuff A (Table 1). They occur in one-quarter of Eruption Sample (%) (%) (%) n Y306 95 26 87 728 1GSA Data Repository item 2019256, Figures Lava Creek Y307 93 23 84 1405 Tuff A Y308 97 21 11 831 DR1 and DR2 (additional quartz surface and reentrant Y310 96 20 94 498 textures), Figure DR3 (describing the effect of gas on magma compressibility), and Videos DR4–DR8 (3-D Lava Creek Y301 93 18 62 608 Tuff B Y305 92 18 16 630 X-ray scans of the quartz and reentrants), is available online at http:// www .geosociety .org /datarepository Note:n is the total number of counted quartz crystals in each sample, with ~4700 crystals examined across all /2019/, or on request from editing@ geosociety .org. samples. Repeated counts on the same population produced counting errors of <1%.

Geological Society of America | GEOLOGY | Volume 47 | Number 8 | www.gsapubs.org 711

Downloaded from https://pubs.geoscienceworld.org/gsa/geology/article-pdf/47/8/710/4793701/710.pdf by University of California Berkeley Library user on 24 July 2019 Magnetite that pumiceous, inflating melt responding to de- A 100 µm in red compression would expand efficiently through tortuous and irregular pathways to produce evacuated, clean reentrants. Instead, formation could occur via dissolution or by synchronous growth, if the fluid exsolved at a similar rate to quartz growth (Gutmann, 1974). The dissolution model is supported by tex- tures preserved within the crystals. Many re- entrants crosscut CL bands (Fig. 2). Most of Empty Plane of the reentrants are interpreted to have formed by reentrants CL slice 200 µm local, accelerated dissolution of quartz by pre- B eruptive bubbles (Busby and Barker, 1966; Gut- mann, 1974; Donaldson and Henderson, 1988). Accelerated dissolution has been documented experimentally and is inferred to be driven by local thermodynamic disequilibrium and en- hanced molecular transport along the fluid-melt interface (Busby and Barker, 1966) (Fig. DR2). This process of dissolution in the presence of bubbles of exsolved fluid was ongoing during the crystallization of quartz. We see evidence for bubble departure, as some crystals contain 200 µm 100 µm early-generation reentrants in their interiors that crosscut CL bands from earlier growth, but were C Glass- lled in green later filled with quartz (Fig. 2). Quartz-hosted, glass-filled reentrants cut across CL bands in eruptions from other cal- deras, including the Toba Tuff (Indonesia), the Oruanui Tuff (New Zealand), and the Central Plateau Member rhyolite lavas () (Liu et al., 2006; Vazquez et al., 2009; Girard and Stix, 2010; Matthews et al., 2012; Loewen and Bindeman, 2015). In these ex- amples, melt filled the reentrants. The dissolu- 200 µm 100 µm tion to produce those reentrants may provide D evidence that those magmas contained ex- solved fluid bubbles at one time. Indeed, vola- tile saturation is proposed for those and many other large silicic eruptions. For example, the Bishop Tuff (California) is estimated to have been volatile-stratified, with its upper portions containing at least 3 wt% exsolved gas, equiva- lent to 5–20 vol% exsolved fluid (Wallace et al., 1995). Primary bubbles in reentrants from the 150 µm 200 µm Bishop Tuff provide additional physical evi- dence that the magma contained an exsolved Figure 2. Synchrotron X­ray microtomography (μXRT) and cathodoluminescence (CL) images fluid (Anderson, 1991). of quartz from Lava Creek Tuff, Yellowstone, western United States. μXRT images (left) are Primary, magmatic fluid inclusions com- shown both with opaque and transparent surfaces to show surface morphology and interior posed of H O-CO mixtures are sometimes distribution of reentrants (blue pathways). Magnetite and glass are shown in red and green, 2 2 respectively. CL images (right) display crystal interior along slice plane shown in gray on preserved in phenocrysts from volcanic erup- transparent surface in μXRT images. Growth bands display prominent grayscale differences. tions (e.g., Lowenstern, 2003; Kamenetsky and White arrows highlight quartz partially to completely filling earlier reentrants. Kamenetsky, 2010; Audétat and Lowenstern, 2014). These “voids” within crystals preserve a record of pre-eruptive exsolved volatiles. Mag- them being filled with dense glass that preserves Empty reentrants require a different mecha- matic fluid inclusions are commonly used as volatile diffusion profiles that are enriched in nism of formation. They were either initially observational data sets to understand the fluids the interior and decrease systematically toward filled with melt that was subsequently expelled supplied to magmatic ore deposits (e.g., Hein- the outlet (e.g., Liu et al., 2007; Humphreys by vesiculation, or primarily filled with exsolved rich et al., 1999). They are less commonly dis- et al., 2008; Lloyd et al., 2014). Commonly, a fluid. During decompression, the expanding cussed in the volcanic context, but are similar to single, outsized bubble occurs at the crystal- fluid would force out melt, but that mechanism established intrusive features such as miarolitic melt interface, which acts as the sink for the is not preferred because when glass is preserved, cavities (e.g., Kamenetsky et al. 2002). Volcanic, degassing volatiles. it is not highly vesiculated. Further, it is unlikely primary fluid inclusions have been described in

712 www.gsapubs.org | Volume 47 | Number 8 | GEOLOGY | Geological Society of America

Downloaded from https://pubs.geoscienceworld.org/gsa/geology/article-pdf/47/8/710/4793701/710.pdf by University of California Berkeley Library user on 24 July 2019 different mineral phases crystallized from a di- modern Yellowstone magmatic system is likely Christiansen, R. L., 2001, The Quaternary and Pliocene verse array of mafic to silicic eruptions ( olivine: gas saturated (Lowenstern and Hurwitz, 2008). Yellowstone Plateau volcanic field of , , and : U. S. Geological Survey Roedder, 1965; plagioclase: Gutmann, 1974; Gassy magma has different physical properties Professional Paper 729-G, 120 p., https:// pubs Naumov et al., 1996; quartz: Davidson and than volatile-undersaturated melt. Of particu- .usgs .gov /pp /pp729g /pp729g .pdf . Kamenetsky, 2007; Pasteris et al., 1996). Indeed, lar importance is the compressibility of magma Cloos, M., 2001, Bubbling magma chambers, cupolas, Gutmann (1974) described some fluid inclu- because it affects the evolution of magma pres- and porphyry copper deposits: International Geol- ogy Review, v. 43, p. 285–311, https://doi .org /10 sions as “tubular voids” that are open at the sure and the volume of magma that will erupt .1080 /00206810109465015 . surface of the crystal, similar to the reentrants (Edmonds and Woods, 2018). Cluzel, N., Laporte, D., Provost, A., and Kanne- described here. Our results have bearing on what may make wischer, I., 2008, Kinetics of heterogeneous The empty, quartz-hosted reentrants are a eruptions super. There are two key questions. bubble nucleation in rhyolitic melts: Implications new example of magmatic fluid inclusions that First, why do large volumes of magma accu- for the number density of bubbles in volcanic conduits and for pumice textures: Contributions provides a physical record of exsolved volatiles mulate before eruption? Second, why are the to Mineralogy and Petrology, v. 156, p. 745–763, in the Lava Creek Tuff reservoir. We theorize that eruptions themselves so large? The presence https:// doi .org /10 .1007 /s00410 -008 -0313 -1 . the hollow reentrants were filled with exsolved of exsolved volatiles greatly increases magma Davidson, P., and Kamenetsky, V.S., 2007, Primary volatiles in the pre-eruptive magma, and can be compressibility (Fig. DR3). In turn, high com- aqueous fluids in rhyolitic magmas: Melt inclu- sion evidence for pre-and post-trapping exso- considered quartz-hosted “bubbles.” Bubbles pressibility buffers magma reservoirs from large lution: Chemical Geology, v. 237, p. 372–383, demonstrate that the magma was saturated prior changes in overpressure during recharge events https:// doi .org /10 .1016 /j .chemgeo .2006 .07 .009 . to eruption. But how bubbly was the magma? (Townsend et al., 2019). High compressibility Degruyter, W., and Huber, C., 2014, A model for erup- Empty reentrants occupy 0.02–1.5 vol% of the also promotes eruptions that are long lived tion frequency of upper crustal silicic magma host quartz. Reentrants from the basal ignimbrite and discharge a greater proportion of stored chambers: Earth and Planetary Science Letters, v. 403, p. 117–130, https://doi .org /10 .1016 /j .epsl of the earliest-erupted Lava Creek Tuff occupy magma (Huppert and Woods, 2002). Our mea- .2014 .06 .047 . the most volume, accounting for 0.4 ± 0.4 vol% surements directly document the existence of Donaldson, C.H., and Henderson, C.M.B., 1988, A of the quartz. Throughout the Lava Creek Tuff, exsolved volatiles in the reservoirs that supplied new interpretation of round embayments quartz ~15%–25% of quartz crystals contain embay- supereruptions. crystals: Mineralogical Magazine, v. 52, p. 27–33, https://doi .org /10 .1180 /minmag .1988 .052 .364 .02 . ments. Quartz is a common phenocryst, but ac- Edmonds, M., 2008, New geochemical insights into counts for only ~10 vol% of the dense rock. ACKNOWLEDGMENTS volcanic degassing: Philosophical Transactions Reentrants thus represent a tiny percentage of We thank Rachel Bruyere for assistance in the field of the Royal Society of London A: Mathemati- magma, <0.03 vol%. But, the number and vol- and sample preparation. We also thank Madison cal, Physical and Engineering Sciences, v. 366, ume of quartz-hosted reentrants is a strict mini- Myers, Jacob Lowenstern, and Eric Christiansen for p. 4559–4579, https:// doi .org /10 .1098 /rsta .2008 .0185 . mum for the pre-eruptive bubble content; the thorough reviews. CL imagery was collected with the assistance of James Maner at University of Texas at Edmonds, M., and Woods, A.W., 2018, Exsolved vola- total exsolved fluid must have been much higher. Austin (USA). Samples were collected under U.S. tiles in magma reservoirs: Journal of Volcanol- Bubbles preferentially nucleate on crystal National Park Service Permit #YELL-07072. This ogy and Geothermal Research, v. 368, p. 13–30, surfaces because reduced surface tension low- research was supported by National Science Founda- https://doi .org /10 .1016 /j .jvolgeores .2018 .10 .018 . Ferguson, D.J., Gonnermann, H.M., Ruprecht, P., ers the required supersaturation (Hurwitz and tion grants EAR-1724429 to Befus and EAR-1724469 to Manga. This research used beamline 8.3.2 of the Plank, T., Hauri, E.H., Houghton, B.F., and Swan- Navon, 1994). Magnetite crystals significantly Advanced Light Source (Lawrence Berkeley National son, D.A., 2016, Magma decompression rates drop the required supersaturation, making them Laboratory, Berkeley, California), which is a Depart- during explosive eruptions of Kīlauea volcano, an excellent heterogeneous nucleation surface ment of Energy Office of Science User Facility under Hawaii, recorded by melt embayments: Bulletin for bubbles (Gardner, 2007). Quartz surfaces contract DE-AC02–05CH11231. We thank Dula Par- of Volcanology, v. 78, 71, https://doi .org /10 .1007 kinson for help with the μXRT imaging and analysis. /s00445 -016 -1064 -x . do not strongly encourage bubble nucleation Gardner, J.E., 2007, Heterogeneous bubble nucleation (Cluzel et al., 2008). Texturally, we know that in highly viscous silicate melts during instanta- bubbles must have attached to quartz in order to REFERENCES CITED neous decompression from high pressure: Chemi- drill the reentrants via dissolution. These were Anderson, A.T., Jr., 1991, Hourglass inclusions: cal Geology, v. 236, p. 1–12, https://doi .org /10 Theory and application to the Bishop Rhyolitic .1016 /j .chemgeo .2006 .08 .006 . large bubbles whose size likely correlates with Tuff: American Mineralogist, v. 76, p. 530–547. Gardner, J.E., 2009, The impact of pre-existing gas on reentrant diameter (10–400 μm wide). Smaller Audétat, A., and Lowenstern, J.B., 2014, Melt inclu- the ascent of explosively erupted magma: Bulletin bubbles may have been established on oxide sions, in Holland, H.D., and Turekian, K.K., eds., of Volcanology, v. 71, p. 835–844, https://doi .org crystals. This creates a picture of the pre-erup- Treatise on Geochemistry (second edition): Ox- /10 .1007 /s00445 -009 -0276 -8 . tive magma: large, free-floating bubbles were ford, UK, Elsevier, p. 143–173, https://doi .org /10 Gerlach, T.M., and McGee, K.A., 1994, Total sulfur .1016 /B978 -0 -08 -095975 -7 .01106 -2 . dioxide emissions and pre‐eruption vapor-satu- distributed throughout the Lava Creek magma, Busby, T.S., and Barker, J., 1966, Simulative stud- rated magma at Mount St. Helens, 1980–88: Geo- some preferentially attaching to quartz and other ies of upward drilling: Journal of the American physical Research Letters, v. 21, p. 2833–2836, phenocryst phases. Ceramic Society, v. 49, p. 441–446, https:// doi https:// doi .org /10 .1029 /94GL02761 . We cannot constrain the absolute abundance .org /10.1111 /j .1151 -2916 .1966 .tb15413 .x . Gerlach, T.M., Westrich, H.R., Casadevall, T.J., and Caricchi, L., Annen, C., Blundy, J., Simpson, G., Finnegan, D.L., 1994, Vapor saturation and ac- of exsolved magmatic volatiles in the Lava and Pinel, V., 2014, Frequency and magnitude cumulation in magmas of the 1989–1990 eruption Creek Tuff. The reentrants demonstrate that of volcanic eruptions controlled by magma in- of Redoubt Volcano, Alaska: Journal of Volcanol- the magma was saturated, with more exsolved jection and buoyancy: Nature Geoscience, v. 7, ogy and Geothermal Research, v. 62, p. 317–337, fluid likely in the upper portions of the reservoir, p. 126–130, https:// doi .org /10 .1038 /ngeo2041 . https:// doi .org /10 .1016 /0377 -0273 (94)90039 -6 . but the empty reentrants provide only a lower Caricchi, L., Sheldrake, T.E., and Blundy, J., 2018, Gerlach, T.M., Westrich, H.R., and Symonds, R.B., Modulation of magmatic processes by CO2 flush- 1996, Preeruption vapor in magma of the climac- bound. Nevertheless, the textures record retained ing: Earth and Planetary Science Letters, v. 491, tic Mount Pinatubo eruption: Source of the giant exsolved volatiles in the magma, complement- p. 160–171, https:// doi .org /10 .1016 /j .epsl .2018 stratospheric sulfur dioxide cloud, in Newhall, ing both geodetic measurements that indicate .03 .042 . C.G., and Punongbayan, R.S., eds., Fire and Mud: that exsolved volatiles are present in magmatic Cassidy, M., Manga, M., Cashman, K., and Bach- Eruptions and Lahars of Mount Pinatubo, Philip- mann, O., 2018, Controls on explosive-effusive pines: Seattle, University of Washington Press, reservoirs, and the geochemical evidence for volcanic eruption styles: Nature Communica- p. 415–433. volatile exsolution (Kilbride et al., 2016). Sur- tions, v. 9, 2839, https:// doi .org /10 .1038 /s41467 Ghiorso, M.S., and Evans, B.W., 2008, Thermody- face gas flux measurements demonstrate that the -018 -05293 -3 . namics of rhombohedral oxide solid solutions and

Geological Society of America | GEOLOGY | Volume 47 | Number 8 | www.gsapubs.org 713

Downloaded from https://pubs.geoscienceworld.org/gsa/geology/article-pdf/47/8/710/4793701/710.pdf by University of California Berkeley Library user on 24 July 2019 a revision of the Fe-Ti two-oxide geothermom- Zealand: Evidence from volatiles and trace ele- Naumov, V.B., Tolstykh, M.L., Kovalenker, V.A., and eter and oxygen-barometer: American Journal of ments in melt inclusions: Contributions to Min- Kononkova, N.N., 1996, Fluid overpressure in Science, v. 308, p. 957–1039, https:// doi .org /10 eralogy and Petrology, v. 151, p. 71–87, https:// andesite melts from central Slovakia: Evidence .2475 /09 .2008 .01 . doi .org /10 .1007 /s00410 -005 -0046 -3 . from inclusions in minerals: Petrology, v. 4, Giordano, D., Russell, J.K., and Dingwell, D.B., 2008, Liu, Y., Anderson, A.T., and Wilson, C.J., 2007, Melt p. 265–276. Viscosity of magmatic liquids: A model: Earth pockets in phenocrysts and decompression rates Pasteris, J.D., Wopenka, B., Wang, A., and Harris, and Planetary Science Letters, v. 271, p. 123–134, of silicic magmas before fragmentation: Journal T.N., 1996, Relative timing of fluid and anhy- https:// doi .org /10 .1016 /j .epsl .2008 .03 .038 . of Geophysical Research, v. 112, B06204, https:// drite saturation: Another consideration in the Girard, G., and Stix, J., 2010, Rapid extraction of dis- doi .org /10 .1029 /2006JB004500 . sulfur budget of the Mount Pinatubo eruption, crete magma batches from a large differentiating Lloyd, A.S., Ruprecht, P., Hauri, E.H., Rose, W., Gon- in Newhall, C.G., and Punongbayan, R.S., eds., magma chamber: The Central Plateau Member nermann, H.M., and Plank, T., 2014, NanoSIMS Fire and Mud: Eruptions and Lahars of Mount rhyolites, Yellowstone Caldera, Wyoming: Con- results from olivine-hosted melt embayments: Pinatubo, Philippines: Seattle, University of tributions to Mineralogy and Petrology, v. 160, Magma ascent rate during explosive basaltic Washington Press, p. 875–891. p. 441–465, https:// doi .org /10 .1007 /s00410 -009 eruptions: Journal of Volcanology and Geother- Putirka, K.D., 2008, Thermometers and barometers -0487 -1 . mal Research, v. 283, p. 1–18, https:// doi .org /10 for volcanic systems: Reviews in Mineralogy and Gonnermann, H.M., and Manga, M., 2007, The fluid .1016 /j .jvolgeores .2014 .06 .002 . Geochemistry, v. 69, p. 61–120, https://doi .org /10 mechanics inside a volcano: Annual Review of Loewen, M.W., and Bindeman, I.N., 2015, Oxygen .2138 /rmg .2008 .69 .3 .

Fluid Mechanics, v. 39, p. 321–356, https:// doi isotope and trace element evidence for three-stage Roedder, E., 1965, Liquid CO2 inclusions in olivine- .org /10 .1146 /annurev .fluid .39 .050905 .110207 . petrogenesis of the youngest episode (260–79 ka) bearing nodules and phenocrysts from basalts: Gualda, G.A., Ghiorso, M.S., Lemons, R.V., and of Yellowstone rhyolitic volcanism: Contributions American Mineralogist, v. 50, p. 1746–1782. Carley, T.L., 2012, Rhyolite-MELTS: A modi- to Mineralogy and Petrology, v. 170, 39, https:// Rust, A.C., Cashman, K.V., and Wallace, P.J., 2004, fied calibration of MELTS optimized for silica- doi .org /10 .1007 /s00410 -015 -1189 -5 . Magma degassing buffered by vapor flow rich, fluid-bearing magmatic systems: Journal of Lowenstern, J.B., 1995, Applications of silicate-melt through brecciated conduit margins: Geol- Petrology, v. 53, p. 875–890, https:// doi .org /10 inclusions to the study of magmatic volatiles, in ogy, v. 32, p. 349–352, https:// doi .org /10 .1130 .1093 /petrology /egr080 . Thompson, J.F.H., ed., Magmas, Fluids, and Ore /G20388 .2 . Gutmann, J.T., 1974, Tubular voids within labradorite Deposits: Mineralogical Association of Canada Shamloo, H.I., and Till, C.B., 2019, Decadal transi- phenocrysts from Sonora, Mexico: American Short Course 23, p. 71–99. tion from quiescence to supereruption: Petrologic Mineralogist, v. 59, p. 666–672. Lowenstern, J.B., 2003, Melt inclusions come of age: investigation of the Lava Creek Tuff, Yellowstone Heinrich, C.A., Günther, D., Audétat, A., Ulrich, T., Volatiles, volcanoes, and Sorby’s legacy, in De Caldera, WY: Contributions to Mineralogy and and Frischknecht, R., 1999, Metal fractionation Vivo, B., and Bodnar, R.J., eds., Melt Inclusions Petrology, v. 174, 32, https:// doi .org /10 .1007 between magmatic brine and vapor, determined in Volcanic Systems: Methods, Applications and /s00410 -019 -1570 -x . by microanalysis of fluid inclusions: Geology, Problems: Developments in Volcanology, v. 5, Tait, S., 1992, Selective preservation of melt inclusions v. 27, p. 755–758, https:// doi .org /10 .1130 /0091 p. 1–21, https:// doi .org /10 .1016 /S1871 -644X in igneous phenocrysts: American Mineralogist, -7613 (1999)027 <0755: MFBMBA>2 .3 .CO;2 . (03)80021 -9 . v. 77, p. 146–155. Humphreys, M.C.S., Menand, T., Blundy, J.D., and Lowenstern, J.B., and Hurwitz, S., 2008, Monitoring a Townsend, M., Huber, C., Degruyter, W., and Bach- Klimm, K., 2008, Magma ascent rates in explo- in repose: Heat and volatile flux at mann, O., 2019, Magma chamber growth dur-

sive eruptions: Constraints from H2O diffusion the Yellowstone Caldera: Elements, v. 4, p. 35–40, ing intercaldera periods: Insights from thermo‐ in melt inclusions: Earth and Planetary Science https:// doi .org /10 .2113 /GSELEMENTS .4 .1 .35 . mechanical modeling with applications to Letters, v. 270, p. 25–40, https:// doi .org /10 .1016 Luhr, J.F., 1990, Experimental phase relations of Laguna del Maule, Campi Flegrei, Santorini, and /j .epsl .2008 .02 .041 . water- and sulfur-saturated arc magmas and the Aso: Geochemistry, Geophysics, Geosystems, Huppert, H.E., and Woods, A.W., 2002, The role 1982 eruptions of El Chichón volcano: Journal v. 20, p. 1574–1591, https:// doi .org /10 .1029 of volatiles in magma chamber dynamics: Na- of Petrology, v. 31, p. 1071–1114, https:// doi .org /2018GC008103 . ture, v. 420, p. 493–495, https:// doi .org /10 .1038 /10 .1093 /petrology /31 .5 .1071 . Vazquez, J.A., Kyriazis, S.F., Reid, M.R., Sehler, R.C., /nature01211 . Malfait, W.J., Seifert, R., Petitgirard, S., Perrillat, J.-P., and Ramos, F.C., 2009, Thermochemical evolu- Hurwitz, S., and Navon, O., 1994, Bubble nucleation Mezouar, M., Ota, T., Nakamura, E., Lerch, P., tion of young rhyolites at Yellowstone: Evidence in rhyolitic melts: Experiments at high pressure, and Sanchez-Valle, C., 2014, Supervolcano erup- for a cooling but periodically replenished post- temperature, and water content: Earth and Plan- tions driven by melt buoyancy in large silicic caldera magma reservoir: Journal of Volcanology etary Science Letters, v. 122, p. 267–280, https:// magma chambers: Nature Geoscience, v. 7, and Geothermal Research, v. 188, p. 186–196, doi .org /10 .1016 /0012 -821X (94)90001 -9 . p. 122–125, https:// doi .org /10 .1038 /ngeo2042 . https://doi .org /10 .1016 /j .jvolgeores .2008 .11 .030 . Kamenetsky, V.S., van Achterbergh, E., Ryan, C.G., Matthews, N.E., Smith, V.C., Costa, A., Durant, A.J., Vigneresse, J.-L., Truche, L., and Richard, A., 2019, Naumov, V.B., Mernagh, T.P., and Davidson, P., Pyle, D.M., and Pearce, N.J.G., 2012, Ultra-distal How do metals escape from magmas to form 2002, Extreme chemical heterogeneity of granite- tephra deposits from super-eruptions: Examples porphyry-type ore deposits?: Ore Geology Re- derived hydrothermal fluids: An example from from Toba, Indonesia and Taupo Volcanic Zone, views, v. 105, p. 310–336, https://doi .org /10 .1016 inclusions in a single crystal of miarolitic quartz: New Zealand: Quaternary International, v. 258, /j .oregeorev .2018 .12 .016 . Geology, v. 30, p. 459–462, https:// doi .org /10 p. 54–79, https:// doi .org /10 .1016 /j .quaint .2011 Waite, K.A., Keith, J.D., Christiansen, E.H., Whitney, .1130 /0091 -7613 (2002)030 <0459: ECHOGD>2 .07 .010 . J.A., Hattori, K., Tingey, D.G., and Hook, C.J., .0 .CO;2 . Myers, M.L., Wallace, P.J., Wilson, C.J., Morter, B.K., 1997, Petrogenesis of the volcanic and intrusive Kamenetsky, V.S., and Kamenetsky, M.B., 2010, Mag- and Swallow, E.J., 2016, Prolonged ascent and rocks associated with the Bingham Canyon por- matic fluids immiscible with silicate melts: Ex- episodic venting of discrete magma batches at the phyry Cu-Au-Mo deposit, Utah, in John, D.A., amples from inclusions in phenocrysts and glasses, onset of the Huckleberry Ridge supereruption, and Ballantyne, G.H., eds., Geology and Ore De- and implications for magma evolution and metal Yellowstone: Earth and Planetary Science Let- posits of the Oquirrh and Wasatch Mountains, transport: Geofluids, v. 10, p. 293–311,https:// doi ters, v. 451, p. 285–297, https:// doi .org /10 .1016 Utah: Society of Economic Geologists Guide- .org /10 .1111 /j .1468 -8123 .2009 .00272 .x . /j .epsl .2016 .07 .023 . book 29, p. 69–90. Kilbride, B.M., Edmonds, M., and Biggs, J., 2016, Ob- Myers, M.L., Wallace, P.J., and Wilson, C.J., 2018, Wallace, P.J., Anderson, A.T., Jr., and Davis, A.M., serving eruptions of gas-rich compressible mag- Inferring magma ascent timescales and recon- 1995, Quantification of pre-eruptive exsolved mas from space: Nature Communications, v. 7, structing conduit processes in explosive rhyolitic gas contents in silicic magmas: Nature, v. 377, 13744, https:// doi .org /10 .1038 /ncomms13744 . eruptions using diffusive losses of hydrogen from p. 612–616, https:// doi .org /10 .1038 /377612a0 . Liu, Y., Anderson, A.T., Wilson, C.J., Davis, A.M., melt inclusions: Journal of Volcanology and Geo- and Steele, I.M., 2006, Mixing and differentia- thermal Research, v. 369, p. 95–112, https:// doi tion in the Oruanui rhyolitic magma, Taupo, New .org /10.1016 /j .jvolgeores .2018 .11 .009 . Printed in USA

714 www.gsapubs.org | Volume 47 | Number 8 | GEOLOGY | Geological Society of America

Downloaded from https://pubs.geoscienceworld.org/gsa/geology/article-pdf/47/8/710/4793701/710.pdf by University of California Berkeley Library user on 24 July 2019