Journal of Petrology | Volume 45 | Number 1 | Pages 27-58 | 2004
© Oxford University Press 2004; all rights reserved
The Cougar Point Tuff: Implications for Thermochemical Zonation and Longevity of High-Temperature, Large-Volume Silicic Magmas of the Miocene Yellowstone Hotspot
DEPARTMENT OF GEOLOGY AND GEOPHYSICS, UNIVERSITY OF UTAH, SALT LAKE CITY, UT 84112, USA
* Corresponding author. Telephone: 801-585-9168. Fax: 801-581-7065. E-mail: hecathey{at}mines.utah.edu or bpnash{at}mines.utah.edu
RECEIVED DECEMBER 18, 2001; ACCEPTED JUNE 30, 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
The 12·7–10·5 Ma Cougar Point Tuff in southern Idaho, USA, consists of 10 large-volume (>102–103 km3 each), high-temperature (800–1000°C), rhyolitic ash-flow tuffs erupted from the Bruneau–Jarbidge volcanic center of the Yellowstone hotspot. These tuffs provide evidence for compositional and thermal zonation in pre-eruptive rhyolite magma, and suggest the presence of a long-lived reservoir that was tapped by numerous large explosive eruptions. Pyroxene compositions exhibit discrete compositional modes with respect to Fe and Mg that define a linear spectrum punctuated by conspicuous gaps. Airfall glass compositions also cluster into modes, and the presence of multiple modes indicates tapping of different magma volumes during early phases of eruption. Equilibrium assemblages of pigeonite and augite are used to reconstruct compositional and thermal gradients in the pre-eruptive reservoir. The recurrence of identical compositional modes and of mineral pairs equilibrated at high temperatures in successive eruptive units is consistent with the persistence of their respective liquids in the magma reservoir. Recurrence intervals of identical modes range from 0·3 to 0·9 Myr and suggest possible magma residence times of similar duration. Eruption ages, magma temperatures, Nd isotopes, and pyroxene and glass compositions are consistent with a long-lived, dynamically evolving magma reservoir that was chemically and thermally zoned and composed of multiple discrete magma volumes.
KEY WORDS: ash-flow tuff; Bruneau–Jarbidge; rhyolite; Yellowstone hotspot; residence time
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
For several decades, detailed studies of the mineralogy and chemical composition of rhyolite tuffs and lava have been employed to deduce the nature of physical and chemical processes in silicic magma reservoirs. Gradients in temperature, crystallinity and melt composition are common (e.g. Hildreth, 1981
; Fridrich & Mahood, 1987; Vogel et al., 1989; Huysken et al., 1994). Chemical analyses of eruptive products including glass, minerals, pumice, fiamme and bulk tuff have documented compositional gaps in several tuffs and provided evidence for stepwise gradients in pre-eruptive reservoirs; physical (volcanological) evidence is consistent with simultaneous withdrawal of multiple discrete magma volumes during eruption (e.g. Fridrich & Mahood, 1987; Boden, 1989; Schuraytz, 1989; Streck & Grunder, 1997). Reversals in zoning and cyclical behavior have implications for tapping mechanisms and eruption progress (Byrd & Nash, 1993; Brown et al., 1998). More recently, silicic magma residence times exceeding 200–250 kyr in large silicic systems have been suggested by direct dating of individual magmatic zircons by ion microprobe (Reid et al., 1997; Brown & Fletcher, 1999). Detailed Ar/Ar and Rb/Sr isotopic studies on melt inclusions in quartz in the Bishop Tuff (Van Den Bogaard & Schirnick, 1995; Christensen & Halliday, 1996), as well as Rb–Sr isochrons defined by lava and glass–mineral isotope ages (Halliday et al., 1989; Christensen & DePaolo, 1993; Davies & Halliday, 1998) suggest that discrete magma volumes that differentiated at different times may persist in a single reservoir for 
0·3 to 1·0 Myr. Magma longevity of 105–106 years for the Bishop Tuff is inferred from 
18O homogeneity (Bindeman & Valley, 2002). Wolff et al. (1999), however, have demonstrated contamination by wall rock of glass included in quartz of the Bandelier Tuff, and cautioned against the interpretation of long magma residence on the basis of ‘aged’ melt inclusions. Evidence of shorter residence times in large silicic systems also comes from U–Pb isotope analyses of zircons in the Bishop Tuff (Reid & Coath, 2000). Short residence times are also inferred for rhyolites erupted in the last 27 kyr at the Taupo volcanic zone in New Zealand (Houghton et al., 1995; Sutton et al., 2000). In this paper we provide evidence for sustained stepwise thermal and compositional gradients in pre-eruptive magma and demonstrate the compositional relationships among the multiple units of the 12·7–10·5 Ma Cougar Point Tuff. These relationships have bearing on the question of the persistence of long-lived magma reservoirs for silicic magmas. |
BACKGROUND |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
The Cougar Point Tuff (CPT) consists of 10 ash-flow tuffs that are exposed along the Idaho–Nevada border, from approximately 115° to 116°W longitude, at the southern margin of the central Snake River Plain. This sequence of outflow ash-flow sheets reaches a maximum thickness of
475 m in the Bruneau River canyon along the Black Rock escarpment in southern Idaho (Bonnichsen & Citron, 1982). The Bruneau–Jarbidge eruptive center is inferred to be the source for these tuffs, based on their areal distribution and the regional distribution of directional structures in some units such as stretched and lineated gas pockets and parallel grooves on sheeting joint planes (Bonnichsen & Citron, 1982). The southern and western margins of the eruptive center itself have been suggested by geologic mapping, and the buried northern and eastern margins by indirect evidence such as regional aeromagnetic anomalies (Bonnichsen, 1982). The center is situated east of the 87Sr/86Sr(i) = 0·706 line (Fig. 1), and is the first major silicic volcanic center along the axis of the Eastern Snake River Plain to be developed in continental crust with Precambrian basement. The eruptive center forms a structural and topographic depression as would a caldera or system of nested calderas, but direct surface evidence for structures and deposits associated with caldera collapse is obscured by late Miocene basalt and voluminous rhyolite lava flows that postdate the CPT, and by Quaternary basalts and intercalated fluvial and lacustrine sediments. The history of explosive volcanism at the Bruneau–Jarbidge eruptive center is characterized by higher magma temperatures and eruptive frequency than in later silicic centers developed eastward along the track of the Yellowstone hotspot (Perkins et al., 1995; Perkins & Nash, 2002).
|
Nomenclature for the units of the Cougar Point Tuff
The various units of the CPT have been designated by Roman numerals (Bonnichsen, 1981
; Bonnichsen & Citron, 1982). From the base of the sequence to the top, these are Units III, V, VII, IX, X, XI, XII, XIII and XV. Perkins et al. (1995) subsequently identified distinct airfall glass compositions at two different locations of Unit XV and therefore they used XVj to refer to exposures in the Jarbidge River drainage at Murphy Hot Spring and XVb for exposures in the Bruneau River drainage. In addition to their different airfall glass compositions, these exposures have distinct mineral and Nd isotopic compositions (Perkins et al., 1995; this paper). Both exposures are younger than Unit XIII, but it remains unclear whether they are different cooling units of a composite ash-flow sheet (Bernt, 1982) or if they represent two distinct ash flows. Because this numbering system is discontinuous and includes gaps, it is somewhat cumbersome for discussions involving several units; therefore, for the purposes of this paper, we use a continuous number scheme (Table 1) in the hope of clarifying the eruption sequence of the various units of the Cougar Point Tuff.
|
Physical features of the Cougar Point Tuff
The typical exposure of a unit of the CPT from base to top consists of: 0·5–5·0 m of gray, very thinly bedded fine ash fallout [with five of the CPT eruptions depositing >150 cm of fallout ash in the Trapper Creek area of southern Idaho,
100 km east of the Bruneau–Jarbidge eruptive center (Perkins et al., 1995)]; fallout ash layers may be overlain by a thin crystal-rich layer and cross-laminated ground layer deposit; a densely welded gray to black basal vitrophyre grading upwards to a spherulitic zone where diameters of spherulites reach 0·5 m in some units; a pervasively devitrified interior with central massive columnar jointing with or without marginal base-parallel platy jointing; and a capping vitrophyre that varies in texture from black, densely welded glass to porous and frothy (Units 6 and 7) to unconsolidated ash. Pumice clasts are extremely rare in the exposures of the Cougar Point Tuff. In some tuffs, much of the basal airfall deposit has been fused by the overlying ash flow. In such cases, vitrophyres were collected from the base of the pyroclastic flow deposit overlying the fused fallout tuff. Flow foliations in the massive interiors are common, as are high-temperature devitrification textures such as lithophysae with open star-shaped vugs. Paleosols underlie each of the airfall tuff units. Descriptive characterizations of the individual units in outcrop and in thin section, as well as chemical and mineralogical data, have been presented by Citron (1976) and Bonnichsen & Citron (1982).
Eruptive volumes and areal extents
The units of the CPT represent large-volume eruptions. Fallout ash from Units 1, 4, 6, 8 and 9j occurs from eastern California and western Nevada to the High Plains of the central USA. These eruptions have areal extents comparable with or greater than the airfall deposit of the 500 km3 Bishop Tuff erupted from Long Valley caldera. Airfall tuffs from CPT Units 2, 3, 7 and 9b are less widely distributed, but are found over much of the western interior of the USA (Nash et al., 1999; Perkins & Nash, 2002). Perkins & Nash (2002) recorded 14 fallout tuffs that were erupted from the Bruneau–Jarbidge center, and they estimated the average volume of ash-fall tuff to be 350 km3 with an average discharge rate of 2300 km3/Ma over the interval from 12·67 to 10·5 Ma. Estimates of the ratio of ash-fall tuff to ash-flow tuff for large Quaternary eruptions indicate that the volume of ash-fall tuff is 
0·5 times the ash-flow volume (Izett et al., 1988), suggesting ash-flow volumes 700 km3 for the larger eruptions of the CPT.
Eruption ages
High-precision ages for three units of the CPT have been obtained by the 40Ar/39Ar laser fusion technique on sanidine separates from fallout tuffs (Perkins et al., 1998). These units include Unit 1 (12·67 ± 0·03 Ma), its successor Unit 2 (12·07 ± 0·04 Ma), and Unit 8 (10·94 ± 0·03 Ma). The remaining ages (Table 2) have been interpolated based on the tephrochronology (40Ar/39Ar dates of reference horizon ash beds) of numerous stratigraphic sections containing CPT fallout tuffs in Miocene sedimentary basins across the northern Basin and Range province. Interpolation provides both an unbiased and relatively precise age estimate. A detailed description of the technique for interpolating ages of airfall tuffs has been provided by Perkins et al. (1998). For example, using dated ash horizons within a single stratigraphic section at Trapper Creek, Idaho (Perkins et al., 1995), the interpolated ages vs 40Ar/39Ar ages for Cougar Point Tuff Units 1 and 2 are 12·69 ± 0·08 vs 12·67 ± 0·03 and 12·10 ± 0·11 vs 12·07 ± 0·04 Ma, respectively. The estimated error for the interpolated ages of various units of the CPT is 0·10 Ma.
|
Isotopic characteristics of regional country rocks
The Bruneau–Jarbidge eruptive center lies SW of the intersection of the central axes of the Western and Eastern Snake River Plains, forming a depression in the topographic lowlands of the Central Snake River Plain (see Fig. 1). Along the southern margins of the Bruneau–Jarbidge eruptive center, the Cougar Point ash flows were emplaced on older volcanic rocks, notably the Miocene Jarbidge Rhyolite and Eocene Bieroth volcanics. Cretaceous and Tertiary quartz dioritic to granitic plutons and late Paleozoic marine sedimentary rocks underlie the Tertiary volcanics (Bonnichsen, 1982
; Honjo, 1990). The presence of ancient continental crust beneath the Bruneau–Jarbidge eruptive center is inferred by its location to the east of the 87Sr/86Sr(i) = 0·706 line. Precambrian crustal material at the southern margin of the central Snake River Plain is indicated by the extremely depleted Nd isotopic signatures of Neogene silicic volcanics in southern Idaho and northern Nevada, including the Jarbidge Rhyolite (
Nd = -26) and the tuff of Arbon Valley (Nd = -20) (Nash et al., 1997). Northeast of the Bruneau–Jarbidge eruptive center, Archean crustal xenoliths with Nd from -22·6 to -52·0 are present in hybrid lavas in the Square Mountain–Magic Reservoir area at the northern margin of the central Snake River Plain (Leeman et al., 1985). The nature of the country rocks hosting the eruptive center is unknown, but it is likely that Cretaceous batholithic rocks were significant crustal components in the vicinity of the Bruneau–Jarbidge eruptive center, based on their distribution north of the central Snake River Plain and their presence at its southern margin. The boundary between Mesozoic accreted terranes and Precambrian North America finds surface expression only far north of the Bruneau–Jarbidge eruptive center and the Snake River Plain, along the Western Idaho suture zone. Precambrian material at this boundary consists of Proterozoic orthogneisses and metasedimentary rocks intruded by late Cretaceous plutons of the Idaho Batholith (Fleck, 1990).
Yellowstone hotspot setting for volcanism at the Bruneau–Jarbidge eruptive center
The Bruneau–Jarbidge volcanic complex is one of several major eruptive centers distributed along the trace of the Yellowstone hotspot. The 700 km track of time-progressive silicic volcanism, containing the Bruneau–Jarbidge center, that is aligned antiparallel to the velocity vector of the North American plate, supports the hypothesis that the heat source driving Yellowstone hotspot volcanism is, like typical oceanic hotspots, a deep-seated mantle feature. A variety of evidence supports a hotspot origin for the Quaternary volcanism of the Yellowstone Plateau, including elevated 3He/4He ratios in fluids and minerals (Craig et al., 1978; Welhan et al., 1983; Kennedy et al., 1985; Craig, 1993), extraordinarily high heat flow, a pronounced 1000 km diameter geoid anomaly, and a parabola of active seismicity and tectonism in the Yellowstone Plateau–Eastern Snake River Plain (Pierce & Morgan, 1992; Smith & Braile, 1994). The pronounced topographic decline to the west of the Yellowstone Plateau that defines the Snake River Plain can be modeled as the response of cooling lithosphere that has additional loading from injection of mantle-derived basalt (Humphreys et al., 2000), consistent with the model for the generation of silicic melts presented in this paper. The inability to image a distinct plume in the upper mantle has led to the suggestion that the Yellowstone source may not extend below 250 km (Smith & Braile, 1994; Humphreys et al., 2000).
|
SAMPLING |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
Samples were collected at five sites (Fig. 1). Fallout tuffs and basal and capping vitrophyres were collected from each unit, with the exception of Unit 5, for which no vitric material is exposed. Unit 5 has been identified only at the Cougar Point location in outcrops flanking the East Fork of the Jarbidge River. Samples of Unit 5 were collected from the base and top of the unit, which consists of a pervasively devitrified, hackly and/or platy jointed flow interior. Also, no capping vitrophyre for Unit 1 was located. The section from Hole in the Ground south along the Bruneau River toward Rowland, NV, offers the most complete exposures of the CPT, with all units except 5 and 9j. The two primary collection sites, Stowell Ranch (Rowland) and Hole in the Ground (5 km south of Black Rock escarpment) are within 7 km of one another, and there is nearly continuous exposure of seven of the eight ash-flow tuff sheets present (Unit 8 is discontinuous in this area). More detailed vertical sampling of all units was achieved, but initial analytical results using basal, interior and capping materials indicated only modest chemical variation in each unit. Therefore detailed analytical work on phenocryst and whole-rock trace element chemistry focused on fresh vitric samples. Assessment of lateral variation in ash-flow chemistry is not permitted by the data, although analyses from individual ash-flow sheets at widely separated locations are presented for Units 2 and 7. Sample numbers, brief descriptions and collection sites for the various CPT units are listed in the Appendix.
|
ANALYTICAL TECHNIQUES |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
Sample preparation and analysis of the airfall tuffs has been described in detail by Perkins et al. (1995)
. Electron microprobe analyses of phenocrysts and glass shards were performed using a Cameca SX-50 system at the University of Utah. Analytical conditions are identical to those described by Nash (1992) for both vitrophyre and airfall glass. Perkins et al. (1995) analyzed an average of 20 shards with one point per shard for the airfall tuffs. Thin sections of the vitrophyres were prepared for analysis of glass compositions, and transmitted light was utilized on the microprobe to locate shard cores for analysis, thereby avoiding sintered shard margins. Approximately 25 shards per sample were analyzed (one point per shard) in the vitrophyres. Mineral separates were obtained from 30 samples in the 125–250 µm grain size fraction (60–120 mesh). Samples were crushed, sieved and cleaned prior to mineral separation using heavy liquids and/or a Frantz Isodynamic SeparatorTM. Sample preparation for all analyses requiring crushed rock involved immersion in 10% (v/v) HNO3 for 15–20 min followed by 15–20 min in an ultrasonic cleaner to dissolve carbonate minerals and remove surficial clay alteration. Samples were then washed at least three times with distilled water and dried at 80°C for 24 h. On average, 20 pyroxenes, 25 feldspars and 25 Fe–Ti oxides were analyzed (one point per grain) per sample where mineral abundance was sufficient. Phenocrysts and point locations were randomly selected during microprobe analysis. Whole-rock samples of vitrophyres were considered appropriate for trace element analyses because xenolithic material is rare, phenocryst abundances are low, and glass is relatively fresh. Cognate pumice fragments are extremely rare in the CPT, and were not utilized because they are intensely flattened, stretched and welded such that they cannot be separated successfully from the matrix. X-ray fluorescence (XRF) analyses were performed at the University of Utah, and analytical conditions have been described by Perkins et al. (1995). Instrumental neutron activation analysis (INAA) was performed by XRAL Activation Services Incorporated in Ann Arbor, MI. Neodymium isotopic ratios were determined by B. P. Nash in the Radiogenic Isotopic Geochemistry Laboratory at the University of Michigan. |
CHEMICAL COMPOSITION |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
The Cougar Point Tuff magmas are metaluminous to slightly peraluminous rhyolites with average glass compositions of 72–76 wt % SiO2, 1·4–2·7 wt % Fe2O3 and 0·4–0·8 wt % CaO (Table 2). The glass shares some characteristics with anorogenic ‘A-type’ granites, including high temperatures (800–1000°C), high wt % FeOT/MgO (10–100), elevated high field strength element (HFSE) concentrations (Zr + Nb + Ce + Y = 500–1000 ppm), relatively high fluorine contents (F = 0·04–0·30 wt %), high wt % TiO2/MgO (1–200 at 73–76 wt % SiO2) and low CaO (<0·8 wt %) (Eby, 1990
; Patiño Douce, 1997). The whole-rock major element compositions of the CPT, reported by Bonnichsen & Citron (1982, table 1) also share many characteristics of anorogenic ‘A-type’ magmas including those produced experimentally. Skjerlie & Johnston (1993) reported on compositions of quenched glass from fluid-absent melting experiments on a hornblende and F-enriched biotite-bearing tonalitic gneiss over a range of temperatures at three pressures (6, 10 and 14 kbar). Patiño Douce (1997) reported results of dehydration melting experiments on a hornblende–biotite tonalite at a fixed temperature of 950°C and pressures of 4 and 8 kbar. The average composition of the CPT is in closest agreement with Patiño Douce's results at 4 kbar; however, it is also consistent with experimental results for compositions of quenched melt at 10 kbar and temperatures in excess of 975°C, based on MgO, CaO, FeOT, TiO2 and F contents [Bonnichsen & Citron (1982, table 1); Patiño Douce (1997, table 1); Skjerlie & Johnston (1993, fig. 5 and table 7)]. Exceptions at 10 kbar are slightly higher SiO2 (73·5 vs 72·1), lower Al2O3 (12·4 vs 14·4) and higher K2O (5·7 vs 4·9) in the CPT. A small degree of feldspar fractionation in the experimental liquids, however, (10%) could produce SiO2 and Al2O3 contents within the range of CPT compositions.
Mineral assemblages are anhydrous, characterized by plagioclase, ± sanidine, augite, ± pigeonite, ± quartz, titanomagnetite, ilmenite, ± fayalite, and accessory zircon ± apatite (Table 2 and Fig. 2). Unit 1 contains orthopyroxene rather than pigeonite. Crystal content in the Cougar Point Tuffs ranges from <5 modal % up to 25% in crystal-rich horizons. Modal analyses on samples collected through the vertical extent of six units show that the vast majority of tuff samples have a crystal content <10% (Citron, 1976; Bonnichsen & Citron, 1982).
|
Glass
The presence of discrete compositional modes in airfall glass is a distinctive feature of the chemical variation in the CPT, as is the recurrence of similar modes in different airfall tuffs throughout the sequence. In the sequence from Unit 2 to Unit 9j, five of the eight airfall samples display two or more modes. Airfall glass compositions fall into three broad groupings according to their Fe and Mg contents, and within these three groups several compositional clusters can be recognized on bivariate plots of Ca vs Fe (Fig. 3). Although electron microprobe analyses were obtained on glass shards in all the welded vitrophyres, the data from unfused airfall tuffs are most useful in the illustration of primary compositional variation in glass. As the earliest erupted material, it is least likely to be homogenized with parcels of magma from other regions of the magma reservoir during eruption. Composite plots of Ca vs Fe (Fig. 3a–c) in airfall from the entire sequence of units show nine distinct modes with respect to Ca vs Fe. Airfall glass in Units 2, 3, 7, 9b and 9j displays multimodal behavior. Similar glass compositions recur in the airfall of different units, including mode G2 (Units 2–4), mode G3 (Units 2, 3, 6 and 7) and mode G5 (Units 3 and 7).
|
A composite plot of Fe-number vs Fe, where Fe-number = Fe/(Fe + Mg) (mol %), illustrates the three compositional evolution trends (Groups VI, V2 and V3) within which the Fe–Ca airfall modes are found (Fig. 3e). In airfall glass, Unit 1 alone forms the Group V1 trend (Fig. 3e). Units 2, 3, 4, 6, 7, 8, 9b and 9j appear in the Group V2 trend, and Units 3, 7, 9b and 9j in Group V3.
The following general features are noted in the vitrophyres: (1) variation in color of shards from gray to brown to dark brown; (2) an expansion of the range of compositions compared with airfall in individual units, including the appearance of V1 compositions in the vitrophyres of most units; (3) the appearance of V3 compositions in Units 2, 3, 6 and 8; (4) the persistence of similarity among certain units (e.g. Units 2, 3 and 7); (5) the restriction of high Fe-number to units late in the eruption sequence (e.g. Units 6, 8 and 9j); (6) the loss of modal definition, particularly in the upper vitrophyres. Figure 4d illustrates the total variation of Fe in airfall and vitrophyre shards as a function of vertical position in each unit, as well as throughout the entire sequence of CPT units. The vitrophyre data represent the variation in shard composition found in single thin sections. We suspect the very low iron compositions measured in the upper vitrophyre of Unit 9b and in the basal vitrophyre of Unit 2 to reflect post-emplacement effects. Several gray shards that were analyzed from these samples are lined with tiny microlites that appear to be iron oxides.
|
Pyroxenes
Euhedral phenocrysts of pyroxene in the 125–250 µm grain size fraction were analyzed (Tables 3 and 4). Ferroan augite is present in all units and pigeonite is present in all units but 1 and 8, with trace amounts in Unit 9j. Fe, Ca and Mg contents of pigeonite and augite in the Cougar Point Tuff define discrete compositional modes (Fig. 4). Three units (4, 8 and 9j) appear to be unimodal with respect to their pyroxene compositions. In the seven units that contain multiple modes of pigeonite and/or augite (Units 1, 2, 3, 5, 6, 7 and 9b), multiple modes are present at the hand-specimen level, i.e. at a single stratigraphic horizon. Pyroxenes in a lava flow between Units 7 and 8 located at Black Rock escarpment were analyzed by electron microprobe in thin section, and examined for possible zonation. Most crystals are euhedral and pyroxene compositions are unimodal. The cores and rims of 18 pyroxene phenocrysts were analyzed (15 pigeonite, three augite) and none exhibit detectable differences between core and rim compositions. The pigeonites are identical to mode P2 in Unit 7, and the augites to A1 in Unit 7.
|
|
Pigeonite
Six pigeonite modes (labeled P1–P6 in Fig. 4) define a linear array with compositional gaps and step function changes in Fe and Mg. Figure 4 illustrates that many of these compositional modes are shared by different eruptive units. Not only are individual modes shared by different units, but similar sets of modes are found in different units. For example: the triplet of modes P1, P2 and P3 is found in two units (2 and 3); the doublet of modes P2 and P3 is present in three units (2, 3 and 7). The recurrence of a single mode rather than pairs or triples in different units is illustrated by modes P4 and P6: mode P4 is found in three units (2, 4 and 9b), and mode P6 is found in three units (2, 5 and 6). Mode P5 appears to be restricted to Unit 9b, although a single analysis comes from Unit 9j. Table 3 lists the median composition of each pigeonite mode on a unit-by-unit basis, and identifies what number of the analyses come from basal vs upper vitrophyres. These proportions are approximate, based on random sampling of an average of 40 pyroxene grains per unit during microprobe analysis of heavy mineral separates from each vitrophyre sample. However, they consistently illustrate stratigraphic variation in the proportions of different modal pyroxene compositions. In units with multimodal pyroxene, the upper vitrophyres host the largest proportion of the most Mg-rich modes in each unit, whereas the basal vitrophyres have the greatest proportion of more Fe-rich modal compositions (Fig. 5).
|
Five of the six pigeonite modes (P1, P2, P3, P4 and P6) that recur in different eruptive units maintain their discrete compositional identities in spite of different eruption ages. Pigeonite compositions in Unit 2 cannot be distinguished from those in Unit 3 (see modes P1, P2 and P3 in Fig. 4). Units 5 and 6 cannot be distinguished from one another in mode P6. Slight differences in Ca content in modes P2 and P3, however, distinguish pigeonites in Unit 7 from those in Units 2 and 3. Unit 7 has slightly higher Ca content in these modes than Units 2 and 3. Similarly, Units 4 and 9b can be distinguished on the basis of Ca content in mode P4. These distinctions suggest small differences in magma temperatures of these units (see below).
Augite
Figure 4 shows clustering of augite compositions and patterns of compositional overlap among different units that are similar to those in pigeonite. There are notable differences compared with patterns in pigeonite, however, including the presence of two discrete modes of augite that are shared by Units 5 and 6, whereas these units contain and share only one mode of pigeonite (mode P6). Also, the majority of augite in Unit 2 overlaps compositionally with Units 5 and 6, reinforcing the similarity among these units suggested in the pigeonite data and in the airfall data, where Units 2 and 6 overlap in glass mode G3 (no glass in Unit 5). The most Fe-rich pyroxene compositions occur in Units 5, 6, 8 and 9j. Table 4 lists median compositions of augite modes on a unit-by-unit basis. Individual augite analyses were assigned to modes A1–A8 on the basis of XEn together with geothermometry results (discussed below). Augite analyses that are determined to be equilibrated with one of the six pigeonite modes are assigned a number from one to six to correlate with the equilibrated pigeonite. Augite modes are not as tightly constrained on plots of Fe vs Mg as pigeonite modes, and assignment of some individual augite analyses to modes is therefore ambiguous. There are six discrete modes of pigeonite, and nine of augite. Modes A7–A9 do not coexist with a low-Ca pyroxene and are the most Fe-rich pyroxene in the CPT. Fayalite coexists with augite in Unit 8.
Fe–Ti oxides
Magnetite and ilmenite are present in all units. Most mineral grains appear to be unoxidized in the welded vitrophyres, although postmagmatic oxidation and/or exsolution textures are evident in some magnetite grains. Upon recalculation of the analyses to obtain ferric iron contents (Stormer, 1983), 40% of ilmenite crystals and >60% of magnetite yield totals below 98·5%. Consequently, few analyses of magnetite are presented for Units 2, 6 and 8, and none for Unit 9j. Units 2 and 9b each show development of at least two modes of ilmenite, illustrated by plots of wt % Mg vs Mn (Fig. 6). Average compositions of magnetite and ilmenite from each unit are available in the Electronic Appendix, Table 1. The Electronic Appendix may be downloaded from the Journal of Petrology website, at http://www.petrology.oupjournals.org.
|
Feldspars
Plagioclase is present in all units of the CPT, and sanidine in all units but Unit 7. Plagioclase compositions are more restricted in some units than in others. For example, the anorthite component varies over An22–28 in Unit 8, An23–40 in Unit 2, and An31–43 in Units 3 and 7. Microprobe analyses of feldspars were performed on mineral separates that included many crystal fragments. Therefore, zonation within individual grains was not evaluated, with the exception of feldspars in thin sections of Unit 9b where core, mid-grain and rim compositions were analyzed on several plagioclase phenocrysts. Rim-to-core variation in anorthite content of single crystals in Unit 9b is small and ranges from 0·2 to 3 mol % An, with some grains showing slight reverse zonation. In contrast to the maximum variation of 3 mol % An within individual crystals, the total variation in Unit 9b spans 8 mol % An. Median compositions of feldspar in each unit are available in the Electronic Appendix, Table 2. Unit 2 shows the largest variation in plagioclase composition, as it does with respect to pyroxene composition. Modal development is not obvious, although compositional overlap among units is reminiscent of patterns in the glass and pyroxene data.
Sanidine in the CPT varies in composition from Ab45An5Or50 to Ab36An2Or62. It dominates the mineral assemblage in Unit 1, where sanidine phenocrysts are larger (reaching several millimeters) than in other units. There is typically 6 mol % variation in the orthoclase component of sanidine within an individual unit, reaching 10 mol % in Unit 5. Sanidine compositions do not appear to be divisible into groups or modes.
Minor phases
Sparse phenocrysts of fayalitic olivine have been identified in Units 2, 5, 6, 7 and 9b, and it is a major phase in Unit 8, where low-Ca pyroxene is absent. Olivines in Units 5, 6 and 7 were found in devitrified samples. Fayalite compositions are plotted in Fig. 2, along the En–Fs join in the pyroxene quadrilateral. Fayalite in Unit 8 defines the most Mg-poor cluster, and another cluster includes fayalite in Units 5, 6 and 7. A single analysis from Unit 9b is the most Mg-rich fayalite composition. Zircon is present in all units, most commonly as inclusions in oxides. Apatite occurs as extremely fine-grained inclusions in oxides.
Xenocrysts and xenoliths
Mafic xenoliths and xenocrysts were found during microprobe analysis of heavy mineral separates. Many were identified in the basal vitrophyre of Unit 7 at two locations, and a few were found in Units 2, 6 and 8. Xenocrysts include forsteritic olivine (Fo80–88 in Units 6 and 7), titaniferous augite and plagioclase. Backscattered electron images of different xenoliths reveal both quench and cumulate textures. The quench texture shows euhedral olivine phenocrysts 150 µm in length in a groundmass of clinopyroxene and plagioclase laths. Olivine ranges in composition from Fo55 (Unit 8) to Fo87 (Unit 7), augite from Wo42En45Fs14 to Wo40En51Fs8, and plagioclase from An57 to An70. The cumulate texture is hypidiomorphic and roughly equigranular compared with the quench texture, and minerals include olivine (Fo65), plagioclase, titanomagnetite and titaniferous augite (Wo41En45Fs14). Because there are no known basalt lava flows underlying or intercalated with the Cougar Point Tuff, these inclusions possibly derive from basaltic magmas fueling the base of the magma chamber or perhaps the vent walls. Xenocryst compositions are available in the Electronic Appendix, Table 3.
Whole-rock trace element composition
Results of INAA and XRF analysis of whole-rock samples of airfall glass and basal and upper vitrophyres of each unit are listed in Table 5. Based on whole-rock compositions, the units of the Cougar Point Tuff show only minor compositional variation from the base to the top of individual units. Concentrations of trace elements in late erupted products (i.e. crystal-bearing upper vitrophyres) rarely differ by more than a factor of ±1·5 from concentrations in early erupted products (i.e. crystal-free airfall glass) (Table 5). The elements that display greater variation are compatible in major phases and include Sr, Ba, Eu and Ce, as well as Co, Sc and Zn. Trace elements incompatible in major phases such as Cs, Hf, Nb, Rb, Th, U, Y and the rare earth elements (REE) show minor compositional variation within individual units. Figure 7 shows elemental variation as a function of stratigraphic position. Units 3, 7, 9j and 9b are consistently higher in Fe2O3, Ti, Mn, Ba, Sr and Zr than the other units. (Unit 5 is not considered here, as it has anomalous whole-rock compositions with respect to REE, crystal content, and has no glassy material.) The three samples from Unit 8 have the highest REE and Y contents, followed by those in Units 6 and 9j, whereas Units 3, 7 and 9b are lowest. Units 6, 8 and 9j also have the highest Fe-number (e.g. Fig 3e) in the Cougar Point Tuff.
|
|
Chondrite-normalized REE abundances are shown in Fig. 8 and are nearly identical in several samples. Minor zonation in most units is indicated (Fig. 7), with the exception of Unit 1, which exhibits greater compositional variation. The largest differences lie in the degree of the Eu anomaly, and in variations among heavy REE (HREE) abundances, which can be explained by varying degrees of fractionation–accumulation of plagioclase and clinopyroxene, respectively. Unit 1 is the only unit to show significant enrichments in REE in early vs late erupted products. Unit 5 shows patterns that are inconsistent with the other units, with a flat to positive Eu anomaly, and a spike in Ce content. The samples for this unit are pervasively devitrified, which may account for the unusual patterns. Alternatively, these anomalies may be due to the higher feldspar crystal content in Unit 5 than in other units. The steep light REE (LREE) enrichment pattern in all units but Unit 5 is consistent with the absence of a fractionating phase that accommodates LREE such as allanite, chevkinite or monazite. The elevated but relatively flat HREE element pattern indicates the absence of garnet or an HREE-retaining phase in the source region.
|
) are unusual compared with the other units. The greatest variation between basal and upper horizons of deposits is seen in Unit 1, where the top of the deposit is depleted in REE relative to the base, and the sample is also devitrified. Unit 5 samples are pervasively devitrified.Neodymium isotopes
Neodymium isotopic analyses on crystal-free glass separates from the airfall tuffs yield a range in
Nd values from -8·5 to -6·6 with a precision of ±0·2 (Table 2). There is an overall trend to less negative values with time, although there is a prominent drop in Nd in Unit 7. Units 1–4 yield values within analytical error of one another (Nd = -8·0 ± 0·2). Units 6 (Nd = -7·4) and 7 (Nd = -8·5) have distinct signatures, and Units 8, 9b and 9j have values similar to one another that are the least negative in the CPT (Nd = -6·7, -6·9 and -6·6, respectively). No isotopic data are available for Unit 5. These values are too high for the rhyolites to be partial melts exclusively of Proterozoic and/or Archean crust, which typically have epsilon values <-15, and reach values <-50 (Leeman et al., 1985; Fleck, 1990). Mantle-derived basalt is a required component in the source materials for the Cougar Point Tuffs. Such a component may be contributed by partial melts of hybridized crustal materials, and/or by direct interaction of crustal- with mantle-derived melts during magma generation, through mixing, mingling or diffusion.
The relative proportion of mantle to crust in the silicic magmas can be represented by the neodymium crustal index (NCI) of DePaolo et al. (1992):
 |
Nd values for the glass, mantle and crustal components. Possible NCI values range from zero for pure mantle component to unity for entirely crustal melts. An average Nd value of -15 is taken for the crustal component, typical of Cretaceous and Proterozoic crust north of the hotspot track immediately east of the 87Sr/86Sr(i) = 0·706 line (Fleck, 1990). We take the mantle component to be that of contemporaneous Snake River Plain basalt with Nd = -3, the mean value for basalt along the hotspot track east of the 0·706 line (Leeman et al., 1992). Values of NCI are provided in Table 2 and range from 0·42 in early erupted magmas (CPT Units 1–4) to 0·31 in the youngest (CPT Units 8, 9b and 9j). Unit 6 has an intermediate value of 0·37, and Unit 7 is anomalously high at 0·46. Except for Unit 7, the overall trend is for an increasing mantle contribution to the magma system over time. |
PRE-ERUPTIVE MAGMA TEMPERATURES |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
Estimates of pre-eruptive magma temperatures for the units of the Cougar Point Tuff were obtained from three geothermometers: Stormer's two-feldspar thermometer (J. C. Stormer, Jr, unpublished computer program, 1994), Ghiorso & Sack's (1991)
Fe–Ti oxide thermometer, and the Ca-QUILF algorithm of Andersen et al. (1993). Pyroxene thermometry results from QUILF are favored because the compositional data are the most internally consistent and abundant.
Pyroxene thermometry
To determine temperatures and evaluate equilibration among pigeonite and augite modes, the projected composition (
En, Wo, Fs) of each pyroxene analysis in every unit of the Cougar Point Tuff was used in QUILF, keeping En, Wo and pressure fixed at 5 kbar and allowing the program to calculate temperature, En and Wo of the coexisting pyroxene. We use the oxide projection scheme supplied with the QUILF program. The choice of 5 kbar is arbitrary, but considered appropriate based on the neodymium isotopic ratios, which show no indication of assimilation of Archean crust and suggest minimal residence time in the upper crust. With pigeonite compositions fixed, calculated compositions of the coexisting augites in the same unit are in good agreement with measured compositions as would be the case for phases in equilibrium (Fig. 9 and Tables 6 and 7). Unit 2 exhibits at least three equilibrium pairs, and Unit 3 at least two pairs. Some pigeonite modes do not have a coexisting augite (e.g. P3 in Unit 3; P2 and P3 in Unit 7; P5 in Unit 9b), which may be a function of sampling density, because of the greater abundance of pigeonite than augite in high-temperature modes. We note, however, that mode A1 in Unit 7 overlaps with many of the calculated equilibrium augite compositions for pigeonite mode P2 (see Fig. 9).
|
) for each pigeonite-bearing unit. Measured compositions of individual pigeonite phenocrysts are used in QUILF to calculate XFs and XEn of the equilibrium high-Ca pyroxene; typically there are more analyses of pigeonite phenocrysts than of augite, so that calculated compositions outnumber measured. Plots are organized such that units that share identical pigeonite modes are grouped together. (Note change of scale for Units 5 and 6.)
|
|
With augite compositions fixed, the agreement between calculated and measured pigeonite compositions is less satisfactory. Measured
Wo in pigeonite is 8% higher than calculated, and En 2% lower. Because there is some uncertainty in the amount of Fe3+ in augite, which influences its projected values of Wo and En, and because of the very restricted compositional range of individual pigeonite modes, measured Wo and En in pigeonite modes are favored as the fixed variables. The average variation for temperatures of individual modes is ±30°C (2
of total variation within a mode). Temperatures calculated from low-Ca pyroxene in the Cougar Point Tuff, based on the median composition for each mode, range from 870°C in Unit 1 to 970°C in Units 2 and 3. Temperatures calculated from augite compositions range from 740°C in Unit 5 to 960°C in Unit 7. Figure 10 illustrates thermometry results for pigeonite and augite, showing systematic variation of En with temperature over the entire CPT dataset. Individual modes are apparent in plots of En vs temperature, with En descending in stepwise fashion with falling temperature.
|
Feldspar thermometry
Estimates of feldspar equilibration temperatures for the Cougar Point Tuff were obtained using Stormer's two-feldspar thermometer. Minimum divergence (±15°C) among TAn, TOr and TAb was achieved using the lowest An plagioclase (oligoclase) in each unit. Reported temperatures are based on the mineral pair compositions in each unit that best satisfy equilibrium conditions for all three components (i.e. TAn = TOr = TAb) in the model formulation, using unadjusted individual mineral analyses. Sanidine grains that optimize the fit have low to mid-range Or component. Greater uncertainty is associated with the feldspar temperature for Unit 3 than for other units. Oligoclase is absent in Unit 3, and sanidine is not in equilibrium with the andesine analyzed. Overall, best fits for CPT feldspars are obtained at P = 1 bar, although several analyses yield good agreement at pressures up to 5 kbar. Results of feldspar geothermometry at P = 1 bar range from
800°C in Unit 1 to 910°C in Unit 9j, and are listed in Table 8 along with best-fit compositions.
|
Iron–titanium oxide thermometry
Average ilmenite and magnetite compositions that fulfill Bacon & Hirschmann's (1988)
Mg–Mn partitioning condition for equilibrium were used in the Fe–Ti oxide geothermometer–oxygen barometer of Ghiorso & Sack (1991). In Unit 9b, two modes of ilmenite are clearly suggested by the data (Fig. 6). The magnetite data for 9b were separated into two groups on the basis of a slight break in Mg content at 0·35 wt %. The average of the high-Mg magnetite group was paired with the average composition of the high-Mg ilmenite mode, and likewise for the low-Mg magnetite and ilmenite. The remainder of the oxide compositions were averaged within each unit and used in the thermometer. This approach yields temperatures that are significantly lower than those determined by previous workers using different geothermometers (e.g. Honjo et al., 1992), but they show trends that are generally consistent with feldspar thermometry, with Units 3, 4, 7 and 9b recording the highest temperatures. A second strategy was also employed for this thermometer, in which all possible pairs of individual phenocryst analyses within a given unit were used if they met Bacon & Hirschmann's equilibrium criterion. Several pairs in some units yield high temperatures that are more consistent with the other thermometers. Maximum temperatures obtained using this method, excluding outliers, are: Unit 1, 855°C; Unit 2, 904°C; Unit 3, 898°C; Unit 4, 908°C; Unit 7, 976°C; Unit 8, 870°C; Unit 9b, 923°C. (Results for average analyses are listed in Table 8.) Because Fe–Ti oxide inclusions are common in CPT pyroxenes, it is expected that oxide equilibration temperatures would overlap those of pyroxenes. There is significant disagreement between pyroxene and oxide thermometry results, however, and the oxides analyzed in this study are suspected to have undergone significant resetting. This conclusion is supported by use of oxide compositions in QUILF, which yields similarly low temperatures. Honjo et al. (1992) similarly found that pyroxene thermometry yielded more consistent and higher temperatures than iron–titanium oxides in rhyolites from the Snake River Plain. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
The multiple eruptions of the Cougar Point Tuff record the explosive silicic phase in the life-cycle of volcanism at the Bruneau–Jarbidge eruptive center, which extended over a 2·2 Myr interval. The collective tuffs thus provide a time-series of windows into the internal dynamics of a major silicic system characterized by several large-volume, high-temperature eruptions that were produced at a frequency more than twice that of the Quaternary Yellowstone Plateau volcanic system. Our objective in this paper is to document and interpret the evolution of the system through time on the basis of the collective observations that emerge from examination of the individual eruptive units. From this perspective, the Cougar Point Tuff presents an interpretative challenge because of the distinctive and repetitive patterns that emerge in the data. In any synthetic view of the system, the following observations are among those that must be accounted for: (1) the recurrence of identical assemblages of pyroxene modes in successive eruptions and in non-sequential eruptions; (2) the recurrence of identical compositional gaps in pyroxene; (3) the recurrence of similar modes of glass in different eruptions; (3) the presence of multiple discrete modes of glass and pyroxenes at single stratigraphic horizons; (4) the higher proportion of high-temperature pyroxene modes in upper vs lower vitrophyres for multimodal units; (5) the repetition of near-liquidus, high-temperature conditions; (6) the Nd isotopic variation. Following discussion of magmatic evolutionary trends and heterogeneities, we assess the merits and deficiencies of various competing hypotheses that may account for the collective observations.
Magmatic evolutionary trends
Although compositional variation within the CPT is limited and the eruptive units are very similar to one another, different degrees of magmatic evolution are apparent in glass, mineral and whole-rock trace element composition. As described above (Fig. 3e), compositions of airfall glass fall into three broad trends—V1, V2 and V3—on the basis of Fe vs Fe-number, where V3 is least evolved (highest Fe) and V1 is most evolved. As discussed below, we interpret these trends to represent distinct evolving magma volumes (hence ‘V’) within the reservoir. In individual eruptive units that contain glass compositions from more than one group, there is systematic variation among Fe, Ca and Si that allows distinction between less evolved and more evolved compositions. Glass modes G5, 8 and 9 (Fig. 3) fall into group V3 and are less evolved (lower Si, higher Fe, Mg and Ca) than their syn-erupted counterparts in V2 (modes G2–4, 6 and 7). Systematic variation of Fe with Si in units with multimodal airfall glass is illustrated in Fig. 11a.
|
Within each of the V groups, it is not obvious which material to consider ‘most’ and ‘least’ evolved, because variation among Fe, Ca and Si is not systematic. However, those units with higher Fe-number (e.g. Units 8 and 9j) also record lower temperatures than other units in the same V group as well as more evolved (higher XFs) pyroxene and feldspar (Ab-rich) compositions. Units that contain airfall glass from V3 (i.e. Units 3, 7, 9b and 9j) have whole-rock compositions that are less evolved than other units with respect to trace elements (see Fig. 7). On a plot of Zr/Rb vs Fe2O3, bulk airfall glass samples that contain exclusively V2 glass are more evolved (e.g. Units 2, 4 and 6) than those containing V3 airfall (e.g. Units 7, 9b and 9j) (Fig. 11b). Unit 1, with airfall in V1, is the most evolved unit of the CPT. Whole-rock vitrophyres from Units 7, 9b and 9j fall among the ‘least evolved’ material on this plot, which is consistent with the designation of V3 glass as the least evolved in the CPT. Samples from these units are also relatively high in compatible trace elements (see Table 5 and Fig. 7). As mentioned above, we exclude discussion of trace element analyses on the pervasively devitrified samples in Unit 5 because they yield anomalous results. Unit 1 can be distinguished from the rest of the units by the presence of orthopyroxene rather than pigeonite in the mineral assemblage, higher crystal content and size, and distinct trace element composition that more closely resembles magmas erupted from the older and more westerly Owyhee–Humboldt volcanic center (Perkins & Nash, 2002
). Therefore, it is not considered to be part of the same magmatic system as the rest of the CPT, and we exclude it from the following discussion.
Summary of compositional and thermal heterogeneity
Compositional heterogeneity is exhibited in each unit of the CPT (Figs 3 and 7), although the variation within a single unit may not be large. Airfall glass in Units 3, 7, 9b and 9j, and possibly Unit 2, displays multimodal behavior. Multiple modes of pigeonite occur in Units 2, 3, 7 and 9b. Multiple modes of augite occur in Units 2, 3, 5 and 6, and equilibrium assemblages of pigeonite and augite are observed in Units 2, 3, 4, 5, 6, 7 and 9b. The presence of multiple discrete modes of both glass and pyroxene at single stratigraphic horizons, the different degrees of magmatic evolution represented by syn-erupted glass modes, and the systematic variation of temperature with XEn in pyroxenes are consistent with stepwise thermal and compositional gradients in pre-eruptive magma where the deeper levels contain the more Mg-rich pyroxene. On the basis of pigeonite thermometry, pre-eruptive thermal gradients in the magma can be reconstructed for Units 2, 3, 7 and 9b. Modes in these units yield temperature gradients of 25°C to 40°C from the deepest level tapped to the top, with a difference of 15–20°C between modes (Table 6 and Fig. 10). For example, the three pigeonite modes (P1, 2 and 3) in Unit 3 record falling temperatures with increasing Fe content in pigeonite from 971°C in P1, to 950°C in P2 and to 934°C in P3, suggesting the more Fe-rich pyroxenes are the more evolved. Low temperatures from augite thermometry for Units 5 and 6 suggest a possible gradient of 150°C; however, we are uncertain as to how to unequivocally interpret these temperatures. The calculated temperature for the most ferric mode of augite (A9) yields a temperature of 760°C in contrast to 900°C calculated for coexisting modes P6 and A6. If real, these temperatures may reflect near-solidus phases incorporated from the wall or roof of the reservoir. Alternatively, the low temperatures may be the result of uncertainty in the ferric iron content of these iron-rich pyroxenes.
The stratigraphic variation observed in the proportions of different modal pyroxene compositions in Units 2, 3, 7 and 9b is also qualitatively consistent with models for evacuation of magma reservoirs (Spera, 1984) involving the withdrawal of progressively deeper levels of magma that is normally zoned with respect to composition and temperature. In the four units mentioned, the upper vitrophyres host the largest proportion of the most Mg-rich modes in each unit, whereas the basal vitrophyres have the greatest proportion of more Fe-rich modes (Fig. 5). The patterns of compositional variation as a function of stratigraphic height within individual tuff units are consistent with the tapping of progressively more extensive regions of the magma reservoir with eruption progress. Because the differences between compositional modes are small, the mixing of different populations of glass and minerals in whole-rock samples serves to dampen the signal of pre-eruptive heterogeneity within the magma reservoir. However, slight normal zonation in pre-eruptive magma is indicated by a decrease in whole-rock LREE content between the bases and tops of Units 2, 3, 6, 9b and 9j (Table 5 and Fig. 7). Several units show slight enrichments in compatible elements such as Sr, Ti, Ba and Mn in upper vitrophyres, which may reflect higher crystal content and/or greater proportions of less evolved material in upper vitrophyres. Slight reversals in zoning may reflect complex eruption dynamics including pauses and relaxation of the magma reservoir.
Of particular interest is the repetition of distinct modes of airfall glass in successive eruptive units. Mode G2 occurs in Units 2, 3 and 4. Mode G3 occurs in Units 2, 3, 6 and 7, and Mode G5 is found in Units 3 and 7 (Fig. 3). Similarly, pyroxene modes recur in successive ash flows. The triplet of modes P1, P2 and P3 is found in two units (2 and 3); the doublet of modes P2 and P3 is present in three units (2, 3 and 7). Mode P4 is found in two units (4 and 9b) and also suggested in Unit 2 by two analyses. Mode P6 is found in two units (5 and 6) and also suggested in Unit 2 by two analyses. The presence of multiple modes of glass in single eruptive units indicates the presence of discrete magma volumes within the reservoir at the time of eruption. The recurrence of distinct glass compositions in subsequent eruptions may indicate that the particular magma volumes have remained in the reservoir throughout the recurrence interval. The repeated eruption of identical modes of pyroxenes is consistent with the persistence of magma volumes indicated by the glass data. Although these relationships prompt us to favor a process in which magma volumes persist and evolve for extended durations in this large, high-temperature system, we first consider several processes that could account for at least some of the observational data.
Generation of heterogeneous magmas
Successive partial melting of a crustal protolith
In this model the advection of mantle-generated basalt magma into the lower crust results in partial melting and mixing producing hybrid magmas that ascend to mid- or upper-crustal reservoirs and are subsequently erupted at the locus of the hotspot. The process is essentially identical to the melting–assimilation–storage–hybridization (MASH) model developed by Hildreth & Moorbath (1988) for the production of voluminous intermediate to silicic melts in the southern Andean subduction zone. At the beginning of such a process, successive melt fractions could be extracted with rising temperature and increasing degree of partial melting in the source region. These magma batches could then arrive subsequently in a mid- to upper-crustal reservoir, each with their characteristic melt compositions and near-liquidus phenocrysts. In this simple model, phenocrysts represent equilibrium assemblages crystallizing from distinct magma volumes.
Partial melting and entrainment of refractory residue
The repetition of mineral compositions in different units of the CPT as well as the coexistence of different populations in individual units may reflect the inheritance of mineral constituents in the CPT from a heterogeneous source region. In the case of CPT units with multiple discrete populations of pyroxene (e.g. Units 2, 3, 5, 6, 7 and 9b), this scenario would entail mobilization of multiple components in the source. Some units have more modes of pyroxene than they do of airfall glass, which would be consistent with a homogeneous liquid that entrained heterogeneous pre-existing minerals. For instance, Unit 6 has at least two modes of augite and only one mode of airfall glass (Figs 3b and 4b). Glomerocrysts have been observed in some thin sections of the CPT, and slight reverse zoning in plagioclase has been observed in Unit 9b; these observations are consistent with a crystal inheritance scenario.
If CPT pigeonite modes represent refractory material or restite from the partially melted source region, such material by definition would reside well below liquidus temperatures. In the case of CPT pyroxenes, their geothermometry requires temperatures in excess of 900°C for modes P1–P6 and A1–A6. The restricted compositional range of each pigeonite mode also requires that little zonation would develop in the minerals as a result of interaction with the liquid. Such conditions might be met if the source region were composed of tonalitic source materials such as those used in the melting experiments of Skjerlie & Johnston (1993) and Patiño Douce (1997). Patiño Douce reported volume melt fractions of 40% at 4 kbar and 30% at 8 kbar, at a constant temperature of 950°C. Skjerlie & Johnston (1993) reported melt fractions of 25 wt % at 975°C and 10 kbar, increasing to 50 wt % at 1075°C. Because the CPT magmas are crystal poor (<10%), effective segregation of crystals from melt in the source region would be required as the magma reservoir developed, but a small amount of restite material could be retained with the liquid. The heterogeneity of glass in individual eruptions, however, and the compositional differences between syn-erupted modes suggest that discrete liquids may accompany the discrete pyroxene compositions. This differs from the classic restite model in which there is a single melt composition, and the variation in bulk composition is simply due to varying proportions of melt and restite.
The presence of both multiple liquids and multiple equilibrium pairs in single eruptions suggests the presence of equilibrium conditions between minerals and melts produced by multiple melting events in the source. Perhaps modest stepwise gradients could be generated if a series of discrete batches of liquid and the restite pyroxenes in equilibrium with those liquids were delivered sequentially to the magma reservoir. For example, in Unit 2, upon delivery of mantle-derived basalt into the crust, partial melting would begin at eutectic temperatures and melt accumulation would proceed until T 900°C (temperature of A6 in Unit 2). At this point melt would be segregated from the source, retaining small amounts of restite pyroxene, and would accumulate in the magma reservoir. This event would be followed by successive batches of magma, each representing equilibrium conditions with observed pyroxene compositions. Mode P1 and its respective liquid would be delivered last, at a temperature of 970°C. Such a scenario could produce a normally zoned reservoir with a modest compositional and thermal gradient. In many respects this process does not differ substantially from the first model except that the phenocryst assemblages are produced during partial melting in the source region. It may be difficult to distinguish inherited pyroxene from magmatic pyroxene under such conditions, as both would be equilibrated with the liquid. We note, however, the euhedral shape and small size of the pyroxene grains that define the modes and the absence of reversely zoned grains that record temperatures more consistent with lower-crustal conditions.
Although we can envision such a process producing the heterogeneity observed in Unit 2, we think it unlikely that the process would be duplicated for the eruption of Unit 3, which evacuates multiple pyroxene modes and glass that are identical to those in Unit 2. A third repetition would be required for Unit 7. It seems that for each appearance of identical modes, a new, fertile location of the same heterogeneous source would be required (i.e. not depleted in material that could again produce the lower-temperature compositions of P3 and P2). In addition, the process of melt segregation is required to be duplicated in detail.
Rejuvenation of pre-existing plutons
In this case, high-temperature melts generated in the lower crust would migrate through a hetereogenous plutonic source and entrain different mineral compositions identical to those observed in the CPT. If the source were heterogeneous and consisted of nested plutons, perhaps repeated duplications of compositions could take place in different eruptive units. However, in such a model, we would expect a homogeneous liquid rather than multiple discrete glass compositions, together with a xenocrystal assemblage that would be randomly distributed in the magma volume. Random distribution of different compositions in the reservoir would not be expected to consistently produce ordered distribution of those compositions in the deposits of multiple eruptions. Instead, we observe patterns in the relative abundances of different modes in basal vs upper stratigraphic horizons of individual units where higher-temperature, less evolved material dominates the upper vitrophyres. These patterns are more consistent with an organized arrangement of crystals in the pre-eruptive magma on the basis of composition and temperature. Xenocrysts would also be expected to exhibit reaction relationships with the liquid, as reported by Bachmann et al. (2002) in the Fish Canyon Tuff.
Remobilization of material from chamber margins
In this model, the walls and carapace of the chamber are the primary source of the observed phenocryst population rather than the existing magma at the time of eruption. The multiple pyroxene populations would be an artifact of eruption rather than a representation of comagmatic phases. A similar scenario has been suggested for the phenocryst assemblage in the Bandelier Tuff (Wolff et al., 1999). Although we do not discount the possibility that some of the minerals (particularly the low-temperature augites) in the CPT may derive from solidified chamber margins (or a near-solidus mush), we think it is unlikely that the observed patterns within and between units can be accounted for satisfactorily by this hypothesis. The compositional gaps would need to be preserved in the margins, which might be possible if the system were stratified, or if a continuous succession of distinct, high-temperature magmas occupied the reservoir allowing high-temperature phases to be preserved in the carapace and margins. However, if entrainment and chamber-wide redistribution of pre-existing pyroxene crystals from the chamber roof or walls prior to eruption were responsible for the repetition of identical modes, it is again difficult to explain the consistently higher proportions of less evolved pyroxene compositions in the upper vitrophyres vs basal vitrophyres. If the unerupted portion of the reservoir were to cool significantly between eruptions, one might expect a considerable range of crystal compositions and sizes in the erupted magma if the crystals were derived from chamber margins as well as zonation within individual crystals, reflecting a protracted growth history.
We believe the multimodal glass compositions in several units, the absence of zoning within individual pyroxenes, the predominance of higher-temperature, less evolved mineral assemblages in upper vitrophyres (Fig. 5), the near-liquidus temperatures, the distinctive compositional gaps between pyroxene modes, and the recurrence of multiple identical compositions in different units argue against scenarios dominated by cycles of reactivation of a partially molten system, successive duplications of heterogeneous magmas that are nearly identical, or repeated sampling of accidental xenocrysts or sidewall material as being primary explanations for the observed repetitions.
Conceptual model of the CPT magmatic system
We favor a model in which a zoned reservoir with compositional gaps develops prior to the eruption of Unit 2, composed of two or more dominant volumes (represented by V groups), that persist and evolve under open-system conditions to eruption of Unit 7. Variations in eruptive volume, vent location and lateral distribution of different layers account for the differences in how many and what modes are represented in each eruptive unit. If the reservoir had a large aspect ratio, sizeable fractions of magma may remain unerupted in each event. Units 8, 9b and 9j may represent a reorganization of the system in response to a major recharge event, suggested by the Nd data, or be genetically unrelated to the preceding units. Fe–Mg modes in pigeonite define an array consistent with a series of discrete differentiation events to more Fe-rich compositions. The origins of such modes and compositional gaps in glass and phenocrysts have not yet been fully explored, but their presence in many silicic tuffs and their absence in plutonic rocks suggest that the gaps reflect a fundamental style of zonation developed in rhyolitic magmatic systems during the explosive (or volcanic) phase of their development. Punctuated reorganization events in response to density and thermal instabilities have been suggested as a mechanism for generating multiple layers (e.g. Streck & Grunder, 1997). Alternatively, V groups in the CPT may have originated through the delivery to the reservoir of successively less evolved magma batches from the same source region, as discussed above, or perhaps through double-diffusive convection. The more subtle differences in composition between glass modes in a single V layer (e.g. G2–4 in V2) represent different degrees of differentiation of magma within that group. The lack of pumice and fiamme prevents us from making unequivocal assignments of airfall glass modes to pyroxene modes. Glass modes are defined on the basis of fallout tuffs, and fallout compositions are not expected to reflect the complete spectrum of compositions tapped by a single eruption. The wider spectrum of glass compositions indicated in most of the vitrophyres compared with airfall (Fig. 3d) illustrates this point.
Table 9 is a schematic representation of the history of the CPT magmatic system, illustrating the persistence of magma volumes V2 and V3, and indicating the range of airfall glass, pyroxene and feldspar compositions present in each eruption. Over the time interval between eruptions of Unit 2 and Unit 7, there are two cases where consecutive eruptions evacuate identical material. Unit 3 duplicates multiple compositions that are found in Unit 2, although it does not duplicate the complete spectrum found in Unit 2. Nd isotopic ratios in airfall for these two units are within analytical error of one another. If the zoned magma of Unit 2 cooled significantly prior to the generation of Unit 3 magma, crystal surfaces would be expected to reequilibrate, and at least to develop zonation as a result of prolonged crystallization. To restore the compositional modes in Unit 3 that were identical to those in Unit 2, crystal memory of the cooling event would need to be perfectly obliterated for each mode, and identical thermal conditions to those that existed for Unit 2 would need to be re-established. A residence time of 0·3 Myr is implied for the magmas in Unit 3. In a similar fashion, the minerals in Unit 6 are indistinguishable from those in Unit 5. Based on its mineralogy, Unit 5 comprises some of the most evolved material in the CPT, indicated by the abundance of sanidine, quartz and plagioclase, and by Fe-rich augite and pigeonite. Two of the most ferroan CPT augite modes appear in Unit 5 (A6 and A9), and are repeated in Unit 6, where they overlap exactly in composition, as does the single pigeonite mode in both of these units (P6). Although we have no age control for Unit 5, there is a paleosol between Unit 5 and Unit 6, indicating a significant time interval. The areal extent of Unit 5 is very limited, suggesting it was a relatively small eruption in comparison with other CPT units, perhaps tapping only the upper reaches of the magma reservoir. Unit 6 is a far more widespread ash-flow tuff with an airfall deposit that is found over much of the western USA from western Nevada to Nebraska (Perkins & Nash, 2002).
|
Non-consecutive eruptions also contain identical materials. The glass and pyroxene modes in Unit 7 reproduce those found in Units 2 and 3, extending possible magma residence times to 0·9 Myr. Unit 7 has the lowest
Nd in the CPT sequence, which may reflect minor assimilation of wall rocks in the hotter regions of the reservoir. The absence of more evolved material in Unit 7 may be accounted for by a vent location where overlying layers were absent (e.g. with more evolved material forming cupolas that pinch out laterally); or perhaps by unusually high discharge rates that enhanced the draw-up height of the layers at the vent. In the latter scenario, the less evolved magma would pinch off overlying layers at the vent, preventing their evacuation (Boden, 1989). Compositions in Units 5 and 6 also duplicate the more evolved material found in Unit 2 (e.g. A6 and P6), but lack the less evolved material common to preceding Units 2 and 3. A residence time of 0·8 Myr is suggested for Unit 6 magma. We interpret the eruptions of Units 5 and 6 to tap the upper reaches of the magma reservoir, perhaps a cupola and carapace material, whereas Units 3 and 7 tap predominantly V3 material. Possibly the eruption of Unit 6 evacuated the uppermost material in the chamber at that time, as these modes do not reappear in subsequent eruptions; new magma may have been generated prior to eruption of Unit 8; or Unit 8 may have been derived from a different cupola.
The compositional range of material in Unit 4 is restricted to single modes of airfall glass and pyroxene, with minor scatter. Glass composition overlaps with G2 in Units 2 and 3, although there is only minor overlap of pyroxene (five analyses in Unit 2 in modes P4 and A4). The value for Nd in airfall glass is not distinguishable from that in Unit 3 airfall. The strong development of the P4 mode in Unit 4 may represent a greater extent of crystallization with the coexisting liquid than was present at the times of eruptions for Units 2 and 3. There is a slight shift in Nd between airfall of Unit 4 and that of Unit 6, which may reflect a minor recharge event prior to eruption of Unit 6 or perhaps isotopic heterogeneity within the reservoir.
With the eruption of Unit 8, similarity to preceding units is lost, except for its V2 affinity and the position of its airfall glass on the Fe–Ca evolutionary trend defined by modes G2–G5. The significant change to higher Nd in airfall glass (–6·7 in Unit 8 vs -8·5 in Unit 7) must reflect a significant recharge event or a different source compared with the preceding units. Unit 8 is volumetrically the most significant of the CPT, with the largest areal distribution of ash-flow tuff and airfall distribution comparable with that of the Huckleberry Ridge and Lava Creek B Tuffs (Perkins & Nash, 2002). It is also cooler than other units of the CPT.
Units 9j and 9b are dissimilar to Unit 8 and other preceding units in that their glass compositions lie off the Fe–Ca trend defined by the other units. Their Nd is similar to that of Unit 8 and the pigeonite in Unit 9b overlaps with that in Unit 4 (mode P4). They also retain a bimodal character and compositions fall into groups V2 and V3. Perhaps a small remnant of the earlier system (e.g. Units 2–7) survived and was co-erupted with Unit 9b.
|
CONCLUDING REMARKS |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
The multiple eruptions of the Cougar Point Tuff record the evolution of a large silicic system at a continental hotspot during an episode of explosive volcanism with more than twice the frequency of eruption as the present-day Yellowstone Plateau volcanic system. Magmas are crystal poor, with anhydrous phenocryst assemblages, consistent with pyroxene thermometry results recording near-liquidus temperatures of 900–1000°C. The multiple discrete populations of glass and pyroxenes (and Fe–Ti oxides in Units 2 and 9b) found at single stratigraphic horizons in several units indicate compositional heterogeneity in pre-eruptive magma, and patterns of abundance between upper and lower vitrophyres in multimodal units are consistent with a stratified magma reservoir. The spectrum of material erupted in Unit 2 presages most compositions observed in the following five eruptions. Duplications of the same doublets and triplets of glass and mineral modes, including equilibrium mineral pairs, in different eruptive units both consecutive and non-consecutive, suggest the persistence of discrete magma volumes at near-liquidus conditions over protracted periods of time. The recurrences of pyroxene modes argue against the possibility that CPT modes are an artifact of vent dynamics during eruption, or preferential sampling of certain compositions. Such a hypothesis would require repeated sampling of identical compositions (and gaps) within a continuously zoned reservoir. The possibility that such overlap among pyroxene compositions could be reproduced from separate batches of magma generated in the same source region is also unlikely. Such a scenario requires successive duplications of isochemical volumes that existed in previous magma reservoirs. If the Cougar Point Tuff units each represent magma volumes generated independently from one another yet similar in composition, one would expect a more continuous range of pigeonite compositions rather than repeated duplications of discrete compositional clusters.
The patterns of duplication observed among different units in the Cougar Point Tuff may be more reasonably accounted for by a steady supply of heat from a mantle source to the base of an evolving large and long-lived mid- or deep-crustal magma reservoir that was tapped by multiple large explosive eruptions. The reservoir represents an open system that was resupplied at least once during its lifetime. Generation of the Cougar Point Tuff is consistent with the existence of a long-lived (0·9 Myr) common magma reservoir for the eruptions of Units 2–7, which maintained a stable thermal and compositional stratification over time despite numerous eruptions. Although residence times suggested for the CPT are of the same order of magnitude (105–106 years) as those that have been estimated for other large silicic systems such as Long Valley, CA, not all silicic magma systems may necessarily behave in this fashion. In this case, however, the tectonic setting of a continental hotspot is in keeping with the maintenance of a large-volume, long-lived high-temperature reservoir as a result of the abundant and consistent supply of basalt magmas to the lower crust from a well-developed and persistent thermal anomaly in the upper mantle.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
Supplementary data for this paper are available on Journal of Petrology online.
|
APPENDIX: SAMPLE COLLECTION LOCATIONS AND DESCRIPTIONS |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
|
|
ACKNOWLEDGEMENTS |
|---|
We thank Mike Perkins for providing data on the airfall tuffs, Vivian Shell and Mike Perkins for their assistance in fieldwork, and Bill Bonnichsen for helpful discussions. We thank J. C. Stormer, Jr, for the feldspar geothermometry computer program. We are grateful to Francis Brown for review of an initial draft. The manuscript was substantially improved by reviews from Ilya Bindeman, Martin Streck and Michael McCurry. We are indebted to George Bergantz for the editorial effort devoted to this paper. This work was supported by NSF grant EAR-9316289 to B.P.N. and GSA Research Grant 5567-95 to H.E.C.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONBACKGROUNDSAMPLINGANALYTICAL TECHNIQUESCHEMICAL COMPOSITIONPRE-ERUPTIVE MAGMA TEMPERATURESDISCUSSIONCONCLUDING REMARKSSUPPLEMENTARY DATAAPPENDIX: SAMPLE COLLECTION...REFERENCES |
|---|
Andersen, D. J., Lindsley, D. H. & Davidson, P. M. (1993). QUILF: a Pascal program to assess equilibria among Fe–Mg–Mn–Ti oxides, pyroxenes, olivine, and quartz. Computers and Geosciences 19, 1333–1350.
Bachmann, O., Dungan, M. A. & Lipman, P. W. (2002). The Fish Canyon magma body, San Juan volcanic field, Colorado: rejuvenation and eruption of an upper-crustal batholith. Journal of Petrology 43, 1469–1503.
Bacon, C. R. & Hirschmann, M. M. (1988). Mg/Mn partitioning as a test for equilibrium between coexisting Fe–Ti oxides. American Mineralogist 73, 57–61.[Abstract]
Bernt, J. (1982). Geology of the southern J-P Desert, Owyhee County, Idaho. M.S. thesis, University of Idaho, Moscow.
Bindeman, I. N. & Valley, J. W. (2002). Oxygen isotope study of the Long Valley magma system, California: isotope thermometry and convection in large silicic magma bodies. Contributions to Mineralogy and Petrology 144, 185–205.[Web of Science]
Boden, D. R. (1989). Evidence for step-function zoning of magma and eruptive dynamics, Toquima caldera complex, Nevada. Journal of Volcanology and Geothermal Research 37, 39–57.[CrossRef][Web of Science]
Bonnichsen, B. (1981). Stratigraphy and measurements of magnetic polarity for volcanic units in the Bruneau–Jarbidge eruptive center, Owyhee County, Idaho. Idaho Bureau of Mines and Geology Open-File Report 81–5.
Bonnichsen, B. (1982). The Bruneau–Jarbidge eruptive center, southwestern Idaho. In: Bonnichsen, B. & Breckenridge, R. M. (eds) Cenozoic Geology of Idaho. Idaho Bureau of Mines and Geology Bulletin 26, 237–254.
Bonnichsen, B. & Citron, G. P. (1982). The Cougar Point Tuff, southwestern Idaho and vicinity. In: Bonnichsen, B. & Breckenridge, R. M. (eds) Cenozoic Geology of Idaho. Idaho Bureau of Mines and Geology Bulletin 26, 255–281.
Brown, S. J. A. & Fletcher, I. R. (1999). SHRIMP U–Pb dating of the preeruption growth history of zircons from the 340 ka Whakamaru Ignimbrite, New Zealand; evidence for >250 k.y. magma residence times. Geology 27, 1035–1038.
Brown, S. J. A., Wooden, J., Wilson, C. J. N. & Cole, J. W. (1998). The Whakamaru group ignimbrites, Taupo Volcanic Zone, New Zealand: evidence for reverse tapping of a zoned silicic magmatic system. Journal of Volcanology and Geothermal Research 84, 1–37.[CrossRef][Web of Science]
Byrd, B. J. & Nash, W. P. (1993). Eruption of rhyolite at the Honeycomb Hills, Utah: cyclical tapping of a zoned silicic magma reservoir. Journal of Geophysical Research 98, 14075–14090.[CrossRef]
Christensen, J. M. & DePaolo, D. J. (1993). Time scales of large volume silicic magma systems: Sr isotopic systematics of phenocrysts and glass from the Bishop Tuff, Long Valley, California. Contributions to Mineralogy and Petrology 113, 100–114.[CrossRef][Web of Science]
Christensen, J. N. & Halliday, A. N. (1996). Rb–Sr ages and Nd isotopic compositions of melt inclusions from the Bishop Tuff and the generation of silicic magma. Earth and Planetary Science Letters 144, 547–561.[CrossRef][Web of Science]
Christiansen, R. L., Yeats, R. S., Graham, S. A., Niem, W. A., Niem, A. R. & Snavely, P. D., Jr (1992). Post-Laramide geology of the U.S. Cordilleran region. In: Burchfiel, B. C., Lipman, P. W. & Zoback, M. L. (eds) The Geology of North America, Volume G-3: The Cordilleran Orogen; Conterminous U.S. Boulder, CO: Geological Society of America, pp. 261–406.
Citron, G. P. (1976). Idavada ash-flows in the Three Creek area, southwestern Idaho, and their regional significance. Masters thesis, Cornell University, Ithaca, NY.
Craig, H. (1993). Yellowstone Hotspot; a continental mantle plume. EOS Transactions, American Geophysical Union 74(43), 602.
Craig, H., Lupton, J. E., Welhan, J. & Poreda, R. (1978). Helium isotope ratios in Yellowstone and Lassen Park volcanic gases. Geophysical Research Letters 5, 897–900.[Web of Science]
Davies, G. T. & Halliday, A. N. (1998). Development of the Long Valley rhyolitic magma system; strontium and neodymium isotope evidence from glasses and individual phenocrysts. Geochimica et Cosmochimica Acta 62, 3561–3574.[CrossRef][Web of Science]
DePaolo, D. J., Perry, F. V. & Baldridge, W. S. (1992). Crustal versus mantle sources of granitic magmas: a two-parameter model based on Nd isotopic studies. Transactions of the Royal Society of Edinburgh: Earth Sciences 83, 439–446.[Web of Science]
Eby, G. N. (1990). The A-type granitoids: a review of their occurrence and chemical characteristics and speculations on their petrogenesis. Lithos 26, 115–134.[CrossRef][Web of Science]
Fleck, R. J. (1990). Neodymium, strontium, and trace-element evidence of crustal anatexis and magma mixing in the Idaho Batholith. In: Anderson, J. L. (ed.) The Nature and Origin of Cordilleran Magmatism. Geological Society of America, Memoir 174, 359–373.
Fridrich, C. J. & Mahood, G. A. (1987). Compositional layers in the zoned magma chamber of the Grizzly Peak Tuff. Geology 15, 299–303.
Ghiorso, M. S. & Sack, R. O. (1991). Fe–Ti oxide geothermometry; thermodynamic formulation and the estimation of intensive variables in silicic magmas. Contributions to Mineralogy and Petrology 108, 485–510.[CrossRef][Web of Science]
Halliday, A. N., Mahood, G. A., Holden, P., Metz, J. M., Dempster, T. J. & Davidson, J. P. (1989). Evidence for long residence times of rhyolitic magma in the Long Valley magmatic system; the isotopic record in precaldera lavas of Glass Mountain. Earth and Planetary Science Letters 94, 274–290.[CrossRef][Web of Science]
Hildreth, W. (1981). Gradients in silicic magma chambers: implications for lithospheric magmatism. Journal of Geophysical Research 8611, 10153–10192.
Hildreth, E. W. & Moorbath, S. (1988). Crustal contributions to arc magmatism in the Andes of central Chile. Contributions to Mineralogy and Petrology 98, 455–489.[CrossRef][Web of Science]
Honjo, N. (1990). Geology and stratigraphy of the Mount Bennett Hills, and the origin of west–central Snake River Plain Rhyolites. Dissertation, Rice Unversity, Houston, TX.
Honjo, N., Bonnichsen, B., Leeman, W. P. & Stormer, J. J. C. (1992). Mineralogy and geothermometry of high-temperature rhyolites from the central and western Snake River Plain. Bulletin of Volcanology 54, 220–237.[Web of Science]
Houghton, B. F., Wilson, C. J. N., McWilliams, M. O., Lanphere, M. A., Weaver, S. D., Briggs, R. M. & Pringle, M. S. (1995). Chronology and dynamics of a large silicic magmatic system; central Taupo volcanic zone, New Zealand. Geology 23, 13–16.
Humphreys, E. D., Dueker, K. G., Schutt, D. L. & Smith, R. B. (2000). Beneath Yellowstone; evaluating plume and nonplume models using teleseismic images of the upper mantle. GSA Today 10, 1–7.
Huysken, K. T., Vogel, T. A. & Layer, P. W. (1994). Incremental growth of a large volume, chemically zoned magma body; a study of the tephra sequence beneath the Rainier Mesa ash flow sheet of the Timber Mountain Tuff. Bulletin of Volcanology 56, 377–385.[Web of Science]
Izett, G. A., Obradovich, J. D. & Mehnert, H. H. (1988). The Bishop ash bed (middle Pleistocene) and some older (Pliocene and Pleistocene) chemically and mineralogically similar ash beds in California, Nevada and Utah. US Geological Survey Bulletin 1675, 1–37.
Kennedy, B. M., Lynch, M. A., Reynolds, J. H. & Smith, S. P. (1985). Intensive sampling of noble gases in fluids at Yellowstone; I, Early overview of the data; regional patterns. Geochimica et Cosmochimica Acta 49, 1251–1261.[CrossRef][Web of Science]
Leeman, W. P., Menzies, M. A., Matty, D. J. & Embree, G. F. (1985). Strontium, neodymium and lead isotopic compositions of deep crustal xenoliths from the Snake River plain; evidence for Archean basement. Earth and Planetary Science Letters 75, 354–368.[CrossRef][Web of Science]
Leeman, W. P., Oldow, J. S. & Hart, W. K. (1992). Lithosphere-scale thrusting in the western US Cordillera as constrained by Sr and Nd isotopic transitions in Neogene volcanic rocks. Geology 20, 63–66.
Nash, W. P. (1992). Analysis of oxygen with the electron microprobe; applications to hydrated glass and minerals. American Mineralogist 77, 453–457.[Abstract]
Nash, W. P., Perkins, M. E., Christensen, J. N., Lee, D. & Halliday, A. N. (1997). Isotopic tracking of the Yellowstone Hotspot; temporal variation in neodymium and hafnium. Geological Society of America, Abstracts with Programs 29(6), 70.
Nash, W. P., Perkins, M. E. & Cathey, H. E. (1999). Identification of Yellowstone hotspot ashes in Miocene sedimentary interbeds of the Columbia Plateau basalts. EOS Transactions, American Geophysical Union 80(46), Supplement, 1188.
Patiño Douce, A. E. (1997). Generation of metaluminous A-type granites by low-pressure melting of calc-alkaline granitoids. Geology 25, 743–746.
Perkins, M. E. & Nash, B. P. (2002). Explosive silicic volcanism of the Yellowstone hotspot: the ash-fall tuff record. Geological Society of America Bulletin 114, 367–389.
Perkins, M. E., Nash, W. P., Brown, F. H. & Fleck, R. J. (1995). Fallout tuffs of Trapper Creek, Idaho—a record of Miocene explosive volcanism in the Snake River Plain volcanic province. Geological Society of America Bulletin 107, 1484–1506.
Perkins, M. E., Williams, S. K., Brown, F. H., Nash, W. P. & McIntosh, W. (1998). Sequence, age, and source of silicic fallout tuffs in middle to late Miocene basins of the northern Basin and Range Province. Geological Society of America Bulletin 110, 344–360.
Pierce, K. L. & Morgan, L. A. (1992). The track of the Yellowstone hot spot; volcanism, faulting, and uplift. In: Link, P. K., Kuntz, M. A. & Platt, L. B. (eds) Regional Geology of Eastern Idaho and Western Wyoming. Geological Society of America, Memoir 179, 1–53.
Reid, M. R. & Coath, C. D. (2000). In situ U–Pb ages of zircons from the Bishop Tuff; no evidence for long crystal residence times. Geology 28, 443–446.
Reid, M. R., Coath, C. D., Harrison, T. M. & McKeegan, K. D. (1997). Prolonged residence times for the youngest rhyolites associated with Long Valley Caldera: 230Th–238 U ion microprobe dating of young zircons. Earth and Planetary Science Letters 150, 27–39.[CrossRef][Web of Science]
Schuraytz, B. C., Vogel, T. A. & Younker, L. W. (1989). Evidence for dynamic withdrawal from a layered magma body; the Topopah Spring Tuff, southwestern Nevada. Journal of Geophysical Research 94, 5925–5942.
Skjerlie, K. P. & Johnston, A. D. (1993). Fluid-absent melting behavior of an F-rich tonalitic gneiss at mid-crustal pressures: implications for the generation of anorogenic granites. Journal of Petrology 34, 785–815.
Smith, R. B. & Braile, L. W. (1994). The Yellowstone hotspot. Journal of Volcanology and Geothermal Research 61, 121–187.[CrossRef][Web of Science]
Spera, F. J. (1984). Some numerical experiments on the withdrawal of magma from crustal reservoirs. Journal of Geophysical Research 89, 8222–8236.
Stormer, J. C., Jr (1983). The effects of recalculation on estimates of temperature and oxygen fugacity from analyses of multicomponent iron–titanium oxides. American Mineralogist 68, 586–594.[Abstract]
Streck, M. J. & Grunder, A. L. (1997). Compositional gradients and gaps in high-silica rhyolites of the Rattlesnake Tuff, Oregon. Journal of Petrology 38, 133–163.[CrossRef][Web of Science]
Sutton, A. N., Blake, S., Wilson, C. J. N. & Charlier, B. L. A. (2000). Late Quaternary evolution of a hyperactive rhyolite magmatic system; Taupo volcanic centre, New Zealand. Journal of the Geological Society, London 157, 537–552.
Van Den Bogaard, P. & Schirnick, C. (1995). 40Ar/39Ar laser probe ages of Bishop tuff quartz phenocrysts substantiate long-lived silicic magma chamber at Long Valley, United States. Geology 23, 759–762.
Vogel, T. A., Noble, D. C. & Younker, L. W. (1989). Evolution of a chemically zoned magma body; Black Mountain volcanic center, southwestern Nevada. Journal of Geophysical Research 94, 6041–6058.
Welhan, J. A., Rison, W., Poreda, R. & Craig, H. (1983). Geothermal gases of the mud volcano area, Yellowstone National Park. EOS Transactions, American Geophysical Union 64(45), 882.
Wolff, J. A., Ramos, F. C. & Davidson, J. P. (1999). Sr isotope disequilibrium during differentiation of the Bandelier Tuff: constraints on the crystallization of a large rhyolitic magma chamber. Geology 27, 495–498.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Girard and J. Stix Magma Recharge and Crystal Mush Rejuvenation Associated with Early Post-collapse Upper Basin Member Rhyolites, Yellowstone Caldera, Wyoming J. Petrology, November 3, 2009; (2009) egp070v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Streck Mineral Textures and Zoning as Evidence for Open System Processes Reviews in Mineralogy and Geochemistry, January 1, 2008; 69(1): 595 - 622. [Full Text] [PDF]
|
|
|
|
 |
|
 W. P. Leeman, C. Annen, and J. Dufek Snake River Plain - Yellowstone silicic volcanism: implications for magma genesis and magma fluxes Geological Society, London, Special Publications, January 1, 2008; 304(1): 235 - 259. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. H. CHRISTIANSEN Contrasting processes in silicic magma chambers: evidence from very large volume ignimbrites Geological Magazine, November 1, 2005; 142(6): 669 - 681. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Boroughs, J. Wolff, B. Bonnichsen, M. Godchaux, and P. Larson Large-volume, low-{delta}18O rhyolites of the central Snake River Plain, Idaho, USA Geology, October 1, 2005; 33(10): 821 - 824. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||