Use of Trace Element Abundances in Augite and Hornblende to Determine the Size, Connectivity, Timing, and Evolution of Magma Batches in a Tilted Batholith

Home , Hornblende

Use of trace element abundances in augite and hornblende to determine the size, connectivity, timing, and evolution of magma batches in a tilted batholith

N. Coint, C.G. Barnes, A.S. Yoshinobu, M.A. Barnes, and S. Buck Department of Geosciences, Texas Tech University, Lubbock, Texas 79409-1053, USA

ABSTRACT The occurrence of large volumes of volcanic record the evolution of the melt composition in rocks ejected during single eruptive events (e.g., magmatic systems and thus have great potential to The tilted Wooley Creek batholith (Klam- Ritchey, 1980; Bacon and Druitt, 1988; Lipman provide information that allows reconstruction of ath Mountains, California, USA) consists of et al., 1997; Bachmann et al., 2002, 2005; Chris- the history of intrusive and extrusive complexes. three main zones. Field and textural relation- tiansen, 2005; Hildreth, 2004; Christiansen and In the case of calc-alkaline magmas, augite and ships in the older lower zone suggest batch- McCurry, 2008) suggests that large volumes of hornblende are of particular interest because they wise emplacement. However, compositions of magma are stored in the middle to upper crust. can crystallize at or close to the liquidus, depend- augite from individual samples plot along indi- Recent high-precision U-Pb dating of zircon ing on the amount of water present in the system vidually distinct fractionation trends, confi rm- indicates that it is not uncommon for intrusive (Piwinskii, 1968; Eggler and Burnham, 1973), ing emplacement as magma batches that did systems to grow over several millions of years and because they occur throughout a wide range not interact extensively. The younger upper (e.g., Glazner et al., 2004; Matzel et al., 2006; of compositions, from basaltic andesite to rhyo- zone is upwardly zoned from tonalite to gran- Grunder et al., 2008; Memeti et al., 2010). Ther- lite. By studying the compositions of these min- ite. Major and trace element compositions mal modeling, however, indicates that even large erals, we can also track the chemical evolution of of hornblende show similar variations from batches of magma should not remain above their the melt, which is generally not accessible when sample to sample, indicating growth from a solidus for >1 m.y. (Glazner et al., 2004; Paterson working with bulk-rock data on intrusive rocks single magma batch that was homogenized et al., 2011). These modeling results have led to because many plutonic rocks are partial cumu- by convection and then evolved via upward emplacement models in which plutons form in lates (e.g., Deering and Bachmann, 2010). percolation of interstitial melt. Highly porphy- increments that do not interact extensively with In this study we utilize major and trace ele- ritic dacitic roof dikes, the hornblende com- one another, and have also been interpreted to ment abundances in augite and hornblende to positions of which match those of upper zone indicate that plutons are not necessarily related to reconstruct the assembly of a tilted calc-alka- rocks, demonstrate that the upper zone mush volcanic rocks (Coleman et al., 2004; Mills and line pluton, the Wooley Creek batholith (WCb; was eruptible. The central zone contains rocks Coleman, 2010). These emplacement models also Barnes et al., 1986b; Barnes, 1987). Tilting and of both lower and upper zone age, although imply that important processes in the chemical subsequent erosion have exposed ~9 km of in most samples hornblende compositions evolution of magmatic suites such as fractional structural relief through the intrusion, making it match those of the upper zone. The zone is crystallization, assimilation, and magma mixing a good candidate for study of the organization rich in synplutonic dikes and mafi c mag- (e.g., De Paolo, 1981; Bohrson and Spera, 2007; of intrusive bodies. Furthermore, the presence of matic enclaves. These features indicate that Ohba et al., 2007; Claiborne et al., 2010; McLeod roof dikes in the structurally highest part of the the central zone was a broad transition zone et al., 2010; du Bray et al., 2011) are restricted to system provides samples of magma that escaped between upper and lower parts of the batho- the lower part of the crust (Annen, 2009; Mills the system (Barnes et al., 1986a), enabling us to lith and preserves part of the feeder system to and Coleman, 2010; Tappa et al., 2011). Incre- address the problem of the connection between the upper zone. Homogenization of the upper mental emplacement of batholith-scale intrusions volcanic and plutonic rocks. zone was probably triggered by the arrival of has been demonstrated through high-precision mafi c magma in the central zone. Continued geochronology (Glazner et al., 2004; Matzel GEOLOGICAL SETTING emplacement of mafi c magmas may have pro- et al., 2006; Walker et al., 2007; Grunder et al., vided heat that permitted differentiation of the 2008; Memeti et al., 2010; Miller et al., 2011) in Klamath Mountains upper zone magma by upward melt percola- cases where emplacement spanned hundreds of tion. This study illustrates the potential for use thousands to millions of years. However, some The WCb is situated in northern California, of trace element compositions and variation in batholith-scale plutons may be emplaced in time USA, in the Klamath Mountains geologic prov- rock-forming minerals to identify individual scales of <1 m.y. (Miller et al., 2011; Coint et al., ince. The province consists of a series of north- magma batches, assess interactions between 2013). In such cases, alternative methods must be south–oriented accreted terranes bounded by them, and characterize magmatic processes. used to identify magma batches and assess their regional east-dipping thrust faults, resulting in size and evolution. preservation of a record of more than 400 m.y. INTRODUCTION High-temperature mafi c minerals crystallize of active subduction along the western North early in magmatic systems and are capable of American margin (Snoke and Barnes, 2006). The ways in which plutons are assembled incorporating trace elements in abundances large The WCb is one of a series of plutons (Wooley are currently the subject of vigorous debate. enough to be measured in situ. These minerals Creek suite) emplaced from Middle to Late

Geosphere; December 2013; v. 9; no. 6; p. 1747–1765; doi:10.1130/GES00931.1; 9 ﬁ gures; 5 supplemental ﬁ les. Received 13 March 2013 ♦ Revision received 17 July 2013 ♦ Accepted 29 August 2013 ♦ Published online 23 October 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Coint et al.

Jurassic time (167–156 Ma) into rocks of the western Paleozoic and Triassic belt, central Legend metamorphic belt, and eastern Klamath belt Late Jurassic 123°20’00”W Ji (Allen and Barnes, 2006). The Slinkard plu- plutons ton, which crops out northeast of the WCb, Western is related to WCb magmatism (Barnes et al., Klamath terrane 1986a). However, extensive subsolidus recrys- c. 152 Ma tallization prevented us from including the Orleans thrust Slinkard pluton in this study. At the time of the wHt Western/eastern plutonic event, the area was undergoing exten- eHt Hayfork terranes Lower zone sion associated with formation of the backarc Josephine ophiolite ca. 159–164 Ma (Harper 351 RCt Rattlesnake Creek et al., 1994; Hacker et al., 1995). The Wooley terrane Creek suite was emplaced east (modern coordi- Ukonom Lake nates) of the Josephine ophiolite. At the same Central zone time, subduction-related magmatism was active west of the Josephine ophiolite and is now rep- 6209 resented by the Chetco plutonic complex and 5209 100 Pigeon Rogue Formation. The Chetco-Rogue arc, the 6309 132 7109 Roost Josephine ophiolite, and its cover sequence, the 594 Galice Formation, form the western Jurassic 133 belt, which was thrust underneath the western 30 Paleozoic and Triassic belt along the Orleans Ten 5709b 6809 thrust (Harper et al., 1994) during the Neva- Bear Cuddihy dan orogeny (153–150 Ma; Allen and Barnes, 397 Mtn. Lakes basin 128a 777b 4809 208 2006). This thrusting truncated the base of at 4909 least some Wooley Creek suite plutons (Barnes, Deadman Z5 Lake 1982, 1983; Jachens et al., 1986). Exhumation 317 of high-pressure rocks of the Condrey Mountain Schist through a structural window tilted the

833 41°30’30” N WCb ~15°–30° toward the southwest (Barnes et 394 al., 1986b), and subsequent erosion has resulted in ~9 km of structural relief through the pluton. 687 Medicine Mtn. WCb Upper zone

The WCb (Fig. 1) was emplaced between Gradational 159 and 155 Ma (Coint et al., 2013). The batho- compositional lith intrudes three host terranes of the western boundary Orleans thrust fault Paleozoic and Triassic belt; from structurally (teeth on HW) higher to lower, these are (1) the eastern Hay- Thrust fault fork terrane, a chert-argillite mélange, (2) the (teeth on HW) western Hayfork terrane, a volcaniclastic sand- 590 Field area stone and argillite unit, and (3) the Rattlesnake Klamath Creek terrane, a serpentine matrix to block-on- 471 N Mts. block ophiolitic mélange overlain by a coher- CA ent cover sequence (Wright and Wyld, 1994). 377b The WCb can be divided into three zones, all of which display gradational contacts (Fig. 1). The region of abundant lower zone is composed primarily of biotite ± 548 “Roof zone” dikes hornblende two-pyroxene diorite and/or gabbro 551 164b and biotite ± hornblende two-pyroxene tonalite. 236a Two U-Pb (zircon, chemical abrasion–thermal ionization mass spectrometry, CA-TIMS) 2408 ages from this unit are identical with analytical 257a 0 5 km uncertainty at 158.99 ± 0.17 Ma and 159.22 ± Ji 0.10 Ma (Coint et al., 2013). Lower zone rocks are heterogeneous at the scale of the outcrop and Figure 1. Simpliﬁ ed geologic map of the Wooley Creek batholith dis- locally display sheet-like organization with vari- playing the location of samples analyzed for this study (modiﬁ ed from able amounts of pyroxene. Numerous pyroxene- Barnes, 1987).

1748 Geosphere, December 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Pluton assembly using trace elements in augite and hornblende

rich blocks and dikes (pyroxenite and melagab- strike of foliation is nearly perpendicular to the Hornblende is present in some samples as a late bro), showing variably angular and gradational intrusive contact (Coint et al., 2013). The south- magmatic phase, growing in continuity with contacts with the host magmatic rock, as well as ern part of the upper zone is intruded by a small pyroxenes (Fig. 2A). Quartz, where present, sparse mafi c magmatic enclaves (MME), mafi c (2 km × 0.7 km) body of hornblende biotite is interstitial. Accessory minerals are apatite, dikes, and appinite dikes and pods, are pres- granite (Fig. 1) with a U-Pb (zircon) age of ca. zircon, allanite, and rarely tourmaline. Actino- ent in the lower part of the intrusion. Foliation 156 Ma (Coint et al., 2013). The majority of this lite, chlorite, epidote, and sericite are second- is primarily magmatic, is oriented north-south, unit is fi ner grained than the surrounding upper ary minerals, which are variable in abundance. and dips steeply to the east or west (Coint et al., zone rocks, making the sharp intrusive contacts Where epidote displays sharp contacts with 2013) except along the northeastern contact, readily apparent in the fi eld. hornblende, it is interpreted as being magmatic, where tonalitic rocks display a protoclastic foli- Selvages of two-pyroxene ± hornblende dio- whereas where present as an anhedral phase ation that dips northeast, subparallel to the con- rite and quartz diorite occur locally along the associated with sericitized plagioclase and chlo- tact. Because deformation of these latter rocks western and southern contacts of the upper zone ritized biotite, it is secondary. was subsolidus, it is unlikely that the minerals (Fig. 1), and have U-Pb (zircon) ages of 159.28 Pyroxenite and melagabbro blocks and dike- preserve their igneous composition; therefore, ± 0.17 Ma and 158.32 ± 0.32 Ma, respectively like sheets are coarse grained and have an ortho- these rocks were excluded from this study. (Coint et al., 2013). The contact between the cumulate texture (Fig. 2B). These rocks consist The central zone (Fig. 1) is mainly composed southern mafi c selvage and the granodiorite is of subhedral exsolved augite and orthopyrox- of biotite hornblende quartz diorite and tonalite; lobate (fi g. 7 in Coint et al., 2013). No sharp ene, with or without olivine or olivine pseudo- some samples contain scraps of pyroxene sur- contact between the western mafi c selvages and morphs. Pale green to brown amphibole is pres- rounded by hornblende. Detailed mapping in the granitic to granodioritic rocks of the upper ent as poikilitic crystals and in some samples is the Cuddihy Lakes basin (Fig. 1) shows that the WCb was observed. optically continuous with pyroxene. Hercynite zone consists of multiple decimeter- to meter- Along the western and southwestern contact is present as an accessory mineral. Talc, serpen- scale sheets of tonalite and quartz diorite (with zone, a series of porphyritic dikes intrudes the tine, and green actinolite are secondary miner-

or without MME) along with variably deformed host rocks and the structurally highest parts of als. Plagioclase (An77 to An50) and quartz are and disrupted mafi c synplutonic dikes, includ- the upper zone. Andesitic dikes generally con- interstitial and occur in veins that brecciate the ing appinites (Barnes et al., 1986a; Leopold and tain phenocrysts of augite, enstatite (herein pyroxenite blocks. Yoshinobu, 2010; fi g. 3F in Coint et al., 2013). orthopyroxene), and plagioclase, but some con- In outcrop some sheets display characteristics of tain augite and hornblende. Dacitic dikes have Central Zone upper zone rocks and others of lower zone phenocrysts of plagioclase and hornblende ± rocks. These distinctions are reinforced by U-Pb biotite ± quartz. Crosscutting relationships sug- Rocks in the central zone range from biotite (zircon) ages of two central zone samples (Coint gest that some andesitic dikes were emplaced hornblende quartz diorite to biotite hornblende et al., 2013). The age of one sample (sample Z5; before the emplacement of the upper WCb tonalite. Although these samples contain simi- 159.01 ± 0.20 Ma) is identical within uncertainty magma, because the dikes are cut by the upper lar mineral assemblages, they display a range to lower zone samples, whereas the age of the granodiorite, whereas some andesitic dikes of grain size and magmatic foliation intensity. other (sample WCB-4909; 158.30 ± 0.16 Ma) intrude the upper WCb (Barnes et al., 1986a). Pyroxenes are rarely preserved as scrappy inclu- is identical to samples from the upper zone (see In contrast, dacitic dikes that cut the upper zone sions in hornblende (Fig. 2D) and some relict following). The presence of samples with both have not been observed. In Barnes (1987) and grains are evident as intergrown actinolitic horn- upper and lower zone characteristics is refl ected Barnes et al. (1990), the andesitic dikes were blende and quartz blebs surrounded by mag- in the boundaries of the central zone with the interpreted to be derived from the lower zone, matic hornblende. Hornblende habits vary from two adjacent zones. Ridge traverses indicate whereas the dacitic dikes were interpreted to be euhedral prismatic to poikilitic (Fig. 2C). The that these boundaries are gradational and diffuse from the upper zone. These interpretations are hornblende crystals are optically zoned, with over distances of at least 500 m. Foliation in the tested (see following) on the basis of mineral brown cores and green rims. Plagioclase (An central zone is magmatic and has variable strike compositions. 60 to An ) is subhedral to euhedral and displays and dip (Coint et al., 2013). 35 oscillatory normal and oscillatory reverse zon- The upper zone ranges from structurally PETROGRAPHY ing. Subhedral biotite, plagioclase, and amphi- lower biotite hornblende quartz diorite and/ bole defi ne a weak magmatic foliation. Quartz or tonalite to structurally highest biotite horn- Lower Zone and K-feldspar are interstitial. Chlorite, epidote, blende granite (Fig. 1), and three samples have sericite, minor sphene, and actinolite are sec- U-Pb zircon ages ranging from 158.21 ± 0.17 Other than the pyroxenites and melagabbros, ondary minerals and can locally be abundant. Ma to 158.25 ± 0.46 Ma (Coint et al., 2013). rocks from the lower zone range from biotite The association of epidote with hornblende sug- Outcrops in the upper zone are relatively homo- two-pyroxene diorite and/or gabbro to biotite gests that some epidote is magmatic. geneous and repeated traverses along ridges and hornblende quartz diorite and tonalite. They in cirque basins over a 20-yr period have failed display hypidiomorphic granular texture and are to identify internal intrusive contacts. Swarms more or less porphyritic, with subhedral augite Upper Zone of MME are locally present and are most com- phenocrysts (Fig. 2A). Augite, orthopyroxene, mon near the contact with the central zone and and plagioclase are euhedral and defi ne a weak Tonalite through granite rocks of the upper along the southwestern intrusive contact. Mag- magmatic foliation. Plagioclase compositions zone are hypidiomorphic granular (Fig. 2E).

matic foliation is typically weak and forms a vary between An78 and An33, with normal to Euhedral to subhedral hornblende and plagio- broadly concentric pattern that is subparallel oscillatory normal zoning (Barnes, 1987; this clase deﬁ ne a weak magmatic foliation. Pla- to intrusive contacts except east of Ten Bear paper). Biotite is subhedral and commonly gioclase is weakly normally zoned in samples Mountain, where the approximate east-west shows a reaction relationship with pyroxene. collected near the transition with the central

Geosphere, December 2013 1749

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Coint et al.

MMB-351 A Lower zone quartz diorite MMB-351 B Pyroxene-rich blocks MMB-100

Hbl Cpx Bt Pl Pl Act Cpx Srp Cpx Px Pl Hbl Pl Act Pl Cpx Px Px Opx Px Hbl 2 mm 1 mm

Act WCB-6209 Central zone tonalite E Upper zone granite WCB-2308 C Hbl Biot Hbl Cpx D 0.5mm Hbl Qz K-spar WCB-7109 Pl Bt Hbl Bt Hbl Hbl Hbl Pl Pl

Qz Qz 2mm Pl 2 mm Pl

F Andesite roof dike MMB-553 G Dacite roof dike MMB-548 Pl Pl Act Hbl Pl

Opx Hbl Qz Hbl Cpx Cpx Pl Cpx

Bt Pl 2mm 1 mm Plag

Figure 2. Thin section scans and photomicrographs. Abbre- MME Granite H viations: Cpx—clinopyroxene, Opx—orthopyroxene, Bt— Hbl Qtz biotite, Srp—serpentine, Pl—plagioclase, Hbl—hornblende, Pl Act—actinolite, Qz—quartz, K-spar—K-feldspar, Px— pyroxene. (A) Biotite hornblende two-pyroxene quartz diorite Bt from the lower zone. White dash lines outline pyroxene crys- Hbl tals. (B) Pyroxene-rich block. (C) Biotite hornblende tonal- Pl ite. (D) Pyroxene surrounded by actinolite and green horn- Bt Qz blende. (E) Biotite hornblende granodiorite from the upper K-spar Pl zone. (F) Andesitic roof-zone dike containing phenocrysts of Hbl orthopyroxene, augite, and plagioclase. (G) Dacitic roof dike Pl with quartz, hornblende, plagioclase, and biotite phenocrysts. Hbl (H) Contact between a maﬁ c magmatic enclave (MME) and the host granodiorite. The groundmass in the enclave is mainly 2 mm composed of poikilitic K-feldspar, optically continuous across the boundary between the granite and the enclave.

1750 Geosphere, December 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Pluton assembly using trace elements in augite and hornblende

zone (An56 to An42) and becomes strongly clase (An91 to An81); the foliation deﬁ ned by the crystals in the host granodiorite to granite and

normally zoned (An50 to An19) in samples col- latter wraps around the pyroxene glomerocrysts. contain numerous inclusions of hornblende, pla- lected from the structurally higher (southwest- Sample MMB-594 from the western selvage gioclase, and biotite (Fig. 2H). Maﬁ c magmatic ern) part of the unit (Barnes, 1987). Subhedral (Fig. 1) is a quartz-bearing biotite two-pyroxene enclaves from the upper zone are porphyritic,

biotite crystals are randomly oriented. Quartz diorite. Plagioclase is euhedral and zoned (An79 with phenocrysts of zoned plagioclase, horn-

is interstitial in tonalite but has euhedral faces to An33) and augite and orthopyroxene are euhe- blende, and subhedral biotite. Hornblende clots against K-feldspar in granodiorite and granite. dral, whereas biotite is poikilitic to interstitial. are common in MME; some display actinolitic Orthoclase is interstitial to poikilitic and is vari- Relict olivine has been replaced by orthopyrox- cores with small opaque mineral inclusions. Zir- ably perthitic. Zircon, apatite, and allanite are ene and opaque minerals. Zircon and apatite are con and apatite are accessory minerals. accessory phases. Pyroxene is rarely preserved accessory minerals. This mineral assemblage is as remnant cores in hornblende. However, the identical to that common in the lower zone, and ANALYTICAL METHODS presence of rounded quartz inclusions in horn- the bulk compositions of samples from the west- blende suggests that pyroxene was present ern selvage are similar to lower zone composi- Major and Trace Element Analyses as a high-temperature phase. Whether these tions (Coint et al., 2013). crystals grew from the upper zone magma or Methods used to analyze bulk-rock major were inherited is not possible to determine. Roof Dikes and trace element compositions were presented The upper zone is petrographically distinct in Coint et al. (2013). Major element data for from both the central and lower zones in hav- Andesitic roof dikes are porphyritic and con- minerals come from a variety of sources. Some ing euhedral hornblende phenocrysts as much tain 28%–41% phenocrysts. Optically zoned hornblende data are available in the literature as 1 cm long, a seriate distribution of euhedral plagioclase, augite, and orthopyroxene occur as (Barnes, 1982, 1983, 1987). New microprobe to subhedral hornblende, and a general lack phenocrysts and glomerocrysts (Fig. 2F). Pla- data collected for this project are presented of evidence for augite to hornblende reaction. gioclase displays oscillatory normal to oscilla- in Supplemental File 11. Some of the data Chlorite, epidote, actinolite, sericite, and minor tory reverse zoning with compositions varying published here were collected on the JEOL

sphene are secondary minerals. between An77 and An50 (Barnes, 1987). Olivine JXA8900 electron microprobe at the Univer- is thought to have been present in a few samples sity of Wyoming, with operating conditions Late Granite in which vermicular opaque minerals are sur- of 15 kV accelerating voltage, 20 nA current, rounded by orthopyroxene. Orthopyroxene cores and a beam diameter of 1–2 µm when focused. The late granite that intrudes the southern part surrounded by augite rims are present in a sev- Matrix-matched natural and synthetic stan- of the upper zone (Fig. 1) is texturally distinct, eral samples (e.g., MMB-557 and MMB-164a). dards were used for calibration. The precision

with euhedral, strongly zoned plagioclase (An31 Pyroxene is more or less altered to actinolitic for individual major component was <0.1 wt%

to An11), acicular hornblende, and acicular to amphibole, and plagioclase is partially albitized. (Li, 2008). The remaining data, mainly obtained prismatic biotite. Perthitic orthoclase is present Dacitic roof dikes are porphyritic (Fig. 2G), on rocks from the central zone, were collected as poikilitic crystals that reach 1 cm in diameter. with 23%–54% phenocrysts of euhedral horn- at the University of Oklahoma using a Cameca Quartz is subhedral to interstitial and displays blende, plagioclase ± biotite, and quartz. Relict SX50 electron microprobe equipped with ﬁ ve slight undulose extinction. Euhedral sphene is pyroxenes are rare as cores in amphibole phe- asynchronous wavelength-dispersive spectrom- a primary mineral and is commonly associated nocrysts and as inclusions in plagioclase phe- eters and PGT PRISM 200 energy-dispersive with magnetite. The late granite is the only part nocrysts. Plagioclase is oscillatory normally X-ray analyzer. Analysis was done by wave-

of the WCb in which sphene is a primary phase. zoned with compositions varying between An53 length-dispersive spectrometry using 20 kV

Apatite and zircon are accessory minerals. to An29 (Barnes, 1987). Glomerocrysts are com- acceleration, 20 nA beam current (measured at posed of large zoned plagioclase and euhedral the Faraday cup), and 2 µm spot size. The PAP Mafi c Selvages hornblende, with interstitial quartz and minor algorithm (Pouchou and Pichoir, 1985) was K-feldspar and biotite. Accessory minerals are used to correct from matrix effects, with oxygen Two main mafi c selvages rim the intrusion zircon, apatite, and allanite. Chlorite, actinolite, content calculated by stoichiometry. along the southern and western contacts. The epidote, and sphene are secondary minerals and Hornblende, augite, and orthopyroxene trace textural complexity observed in the southern partially replace hornblende and biotite. element data were collected in situ, in polished mafi c selvage suggests a complex history. For sections, by laser ablation (LA) ICP-MS analy- example, samples MMB-236a and WCB-2408 Mafi c Magmatic Enclaves and Synplutonic sis using a NewWave 213 nm solid-state laser are slightly porphyritic, with optically zoned Dikes and Agilent 7500CS ICP-MS at Texas Tech Uni- euhedral augite microphenocrysts. Relict oliv- versity. Spot diameter was 40 µm with a laser ine is surrounded by orthopyroxene and sec- The synplutonic dikes that are characteristic pulse rate of 5 Hz and fl uence of 11–12 J cm–2. ondary oxides. The groundmass consists of of the central zone are slightly porphyritic to For each analysis, 25 s of background (laser

euhedral to subhedral plagioclase (An72 to An46) aphanitic. Phenocrysts and groundmass min- off) and 60 s of signal were recorded. NIST 612 deﬁ ning a weak magmatic foliation. Apatite is erals are mainly hornblende and plagioclase glass was used as the standard and was analyzed present as acicular needles included in plagio- with lesser subhedral to anhedral biotite. The every 5–7 unknown mineral analyses. Precision clase. In contrast, MMB-257a is distinct from MME are porphyritic, with plagioclase and the other pyroxene-bearing rocks in the selvage. hornblende phenocrysts (Barnes, 1987; Buck et Pyroxenes occur as aggregates associated with a al., 2010). The groundmass of the MME con- 1Supplemental File 1. New microprobe data of hornblende and pyroxene. If you are viewing the PDF relatively large proportion of opaque minerals, sists of 200–300 µm subhedral plagioclase and of this paper or reading it ofﬂ ine, please visit http:// and the pyroxene-oxide clusters are surrounded hornblende. In the upper zone, poikilitic quartz dx.doi.org/10.1130/GES00931.S1 or the full-text arti- by a groundmass of subhedral, aligned plagio- and/or K-feldspar are optically continuous with cle on www.gsapubs.org to view Supplemental File 1.

Geosphere, December 2013 1751

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Coint et al.

and accuracy of the LA-ICP-MS system were Mg/(Mg+Fe) Sr (ppm) 0.9 800 determined by repeated analysis of basaltic synplutonic dikes glass BHVO-2 and are given in Supplemental ABandesitic roof-zone dikes File 22. The contents of SiO or CaO for horn- dacitic roof-zone dikes 2 Mg-rich roof-zone dikes blende and orthopyroxene and CaO for augite, 0.7 600 as determined by electron microprobe, were lower zone tonalite used as internal standards to correct for variabil- lower zone ity in ablation efﬁ ciency. Reduction of the horn- tonalite

blende data using either CaO or SiO2 as internal 0.5 400 standards gave results within the uncertainty of the counting statistics. The LA-ICP-MS data set is presented in Supplemental File 33. 0.3 200 lower zone Geochemistry l.z. pyroxenite/melagabbro central zone upper zone A discussion of the bulk-rock major and trace late-stage upper zone granite 0.1 0 element data was presented in Coint (2012) and 45 50 55 60 65 70 75 80 5 10 15 20 25 MgO Coint et al. (2013), in which we show that the SiO2 lower zone has scattered compositions that do 100 not defi ne a clear-cut compositional trend. The CD data also show that mafi c selvages along the central zone upper zone western contact are compositionally identical to increasing SiO2 the lower zone. In contrast, rocks of the central in central zone samples and upper zone were interpreted to defi ne a com- lower zone lower zone increasing SiO positional array with infl ections in the trends of 2 a number of trace and minor elements that are indicative of fractional crystallization. The data 10 also support a cogenetic relationship between rocks of the upper zone and dacitic roof dikes. These relationships are illustrated and expanded upon in Figures 3A and 3B, in which lower zone the following is evident. (1) The lower zone is pyroxenite & clearly distinct from the central and upper zones melagabbro in terms of Mg/(Mg + Fe) and Sr contents (Figs. 1 3A, 3B). (2) The compositions of lower zone La Ce Pr Nd SmEuGdTb Dy Ho Er TmYb Lu La Ce Pr Nd SmEuGdTb DyHo Er TmYb Lu tonalites do not plot on the crude trend defi ned by more mafi c lower zone rocks (Fig. 3A); this 100 indicates that fractional crystallization is not EF roof-zone dikes synplutonic dikes an adequate explanation for the compositional 2- variation in the lower zone. (3) The central and pyroxene andesite upper zones defi ne an array of compositions increasing SiO2 that overlaps those of dacitic roof-zone dikes. in dacitic dikes (4) Andesitic roof zone dike compositions tend to cluster in the low-SiO part of the central and 2 10 upper zone trends. (5) The late-stage hornblende biotite granite is compositionally distinct from the rest of the batholith. Rare earth element (REE) patterns of most WCb samples are broadly similar across rock

2Supplemental File 2. BHVO statistics obtained 1 on the laser ablation system at Texas Tech Univer- La Ce Pr Nd SmEuGdTb Dy Ho Er TmYb Lu La Ce Pr Nd SmEuGdTb DyHo Er TmYb Lu sity. If you are viewing the PDF of this paper or reading it ofﬂ ine, please visit http://dx.doi.org/10.1130 /GES00931.S2 or the full-text article on www.gsapubs Figure 3. Bulk-rock major element diagrams. (A) Mg/(Mg + Fe) versus SiO2 (l.z.—lower .org to view Supplemental File 2. zone). (B) Sr versus MgO. (C) Rare earth element (REE) patterns normalized to chon- 3Supplemental File 3. LA-ICP-MS data of augite drite (Boynton, 1984) of the central and lower zone rocks. (D) REE patterns normalized to and hornblende. If you are viewing the PDF of this paper or reading it ofﬂ ine, please visit http://dx.doi.org chondrite (Boynton, 1984) of upper zone rocks. (E) REE patterns normalized to chondrite /10.1130/GES00931.S3 or the full-text article on (Boynton, 1984) of the roof-zone dikes; both dacite and andesite. (F) REE patterns normal- www.gsapubs.org to view Supplemental File 3. ized to chondrite (Boynton, 1984) of synplutonic dikes.

1752 Geosphere, December 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Pluton assembly using trace elements in augite and hornblende

types, with moderate negative slopes and small ponents vary between 2.3–11.1 mol%, with Al vage. The Zr and REETOT contents of augite or no Eu anomalies (Figs. 3C–3E). Typical gab- being the most abundant nonquadrilateral cation from the southern maﬁ c selvage are similar broic through tonalitic lower zone rocks are (Barnes, 1987). Trace element abundances are to augite from the two-pyroxene andesite roof shown as a ﬁ eld in Figure 3C because these compared to Zr contents in Figure 4 because dikes (Figs. 4E, 4F). samples have parallel slopes. As shown in Coint Zr is incompatible in pyroxene and behaved Augite from the two-pyroxene andesitic roof

et al. (2013), the REE abundances of these rocks incompatibly in rocks of the lower zone (e.g., dikes has higher TiO2 and Al2O3 and lower CaO

are positively correlated with SiO2 contents. In Barnes, 1983). Therefore, increasing Zr in a contents than augite from the lower part of the contrast, pyroxenite and melagabbro blocks and single pyroxene grain is reﬂ ective of increasing intrusion (Barnes, 1987). Chromium contents intrusive sheets in the lower zone have lower Zr in the melts from which the pyroxene grew. are relatively constant and low (<1000 ppm) REE abundances and ﬂ atter patterns, and dis- Augite from the lower zone varies between compared to augite from the lower WCb (Fig. play both positive and negative Eu anomalies wollastonite, Wo 0.42–0.47, enstatite, En 0.40– 4C). A few spots analyzed in the intermediate (Fig. 3C). Although samples from the central 0.47, ferrosilite, Fs 0.14–0.18. In terms of trace parts of the crystals display higher Cr contents,

and upper zones have total REE (REETOT)abun- elements, augite crystals from the lower zone especially in augite from samples WCB-164b dances broadly similar to lower zone samples have compositions that vary widely and can be and MMB-557, where Cr concentrations reach

(Figs. 3C, 3D), with increasing SiO2 the abun- divided in two groups based on their Cr con- 3060 ppm and 2700 ppm, respectively. The dances of middle and heavy REEs decrease and tent, with one group containing >3700 ppm Cr augite from the roof dikes has higher Sr (>22 the REE patterns show more pronounced con- and the other containing <2700 ppm Cr. With ppm) than augite from the lower zone, except cave-upward shapes (Fig. 3C; Barnes, 1983). increasing Zr, Cr decreases exponentially from in sample MMB-590, in which augite Sr con-

The REE patterns of andesitic and dacitic roof 5995 to 167 ppm (Fig. 4A). REETOT contents centrations overlap those of lower zone augite zone dikes are shown in Figure 3E. The andes- range from 5 to 156 ppm. Individual crystals (Fig. 4D).

itic dike patterns are parallel to those of the least tend to show increasing REETOT from core to

evolved (lowest SiO2) dacite dikes. As with rim (Fig. 4B). Strontium contents are nearly Hornblende upper zone samples, the REE abundances of the constant with increasing Zr and vary between In the following we use the amphibole

dacitic dikes decrease with increasing SiO2 and 12.2 and 25.6 ppm (Fig. 4D). The augite crys- nomenclature of Leake et al. (1997). We refer the REE patterns show greater upward concav- tals displaying the highest Sr content are part of to brown-green amphibole as hornblende and to ity (Fig. 3E). the high Cr group. pale green amphibole as actinolite. Actinolitic Maﬁ c synplutonic dikes, which character- Augite crystals in pyroxene-rich blocks have hornblende is the transition between the two. ize the central zone, show wide compositional compositions slightly more magnesian (Wo Amphibole compositions in the lower zone scatter, but essentially overlap compositions of 0.44–0.49, En 0.42–0.46, Fs 0.07–0.18) than the range from magnesiohornblende to actinolite lower zone samples (Figs. 3A, 3B). The REE augite from the common rock types in the lower (ﬁ g. 6 in Barnes, 1987). Chromium concentra- patterns of these dikes are similar to those of zone. Augite from the pyroxene-rich blocks tions vary from 2023 to 139 ppm, with most lower zone rocks, but show a greater range of plots in two distinct groups in the Zr versus Cr spots below 1200 ppm (Fig. 5A); Cr is not cor-

REE abundances (Fig. 3F). and REETOT diagrams (Figs. 4A, 4B). The ﬁ rst related with Ti contents. Hornblende contains group, mainly composed of cores and inter- variable amounts of REEs (Fig. 5C); the light Mineral Chemistry mediate zones, overlaps with compositions of (L) REEs vary from 10× to 100× chondritic augite from the common rock types in the lower values, whereas heavy (H) REEs vary from

The major element compositions of miner- zone, with high Cr and low REETOT. The second 6× to 50× chondrites. Hornblende REE pat- als from the WCb are summarized here from group, composed of intermediate zones, rims terns are not necessarily parallel to each other,

previously published work (Barnes, 1987; for and one core, has REETOT values higher than especially for the LREEs, where the shape of additional analyses, see Supplemental File 1 typical augite from the lower zone. the patterns is quite variable (Fig. 5C). The [see footnote 1]; for LA- ICP-MS analyses for In the upper zone, relict augite was analyzed size of the negative Eu anomaly increases with augite, orthopyroxene, and hornblende, see Sup- in only one sample, MMB-397. The few augite increasing REE abundance. plemental File 3 [see footnote 3]). Attributions crystals preserved have low Cr (92–471 ppm), Amphiboles in the pyroxene-rich blocks are of core, intermediate, or rim labels for the loca- intermediate Zr (14.8–19.3 ppm; see Supple- magnesiohornblende and these amphiboles tions of the laser spots are based on geometri- mental File 3 [see footnote 3]; Figs. 4E, 4F), and have higher Mg# (0.8–0.9) than does horn- cal criteria, because most augite crystals in the Sr concentrations that overlap with augite from blende in other lower zone samples. In contrast, batholith lack optical zoning. In contrast, augite the lower zone. Cr contents overlap with hornblende from the in the roof dikes displays sharp growth zoning. Compositions of augite in the western sel- lower zone (Fig. 5A). Titanium and Zr are pres- Hornblende is optically zoned, with olive-brown vage overlap those from the lower zone and the ent in relatively low concentrations (<5000 ppm cores and green rims. Intermediate zones (man- pyroxene-rich blocks (Figs. 4E, 4F). In contrast, and <50 ppm, respectively; Fig. 5B). The REE tles) in hornblende are typically brown. Subsoli- augite from the southern selvage is distinct in concentrations in magnesiohornblende in the dus amphibole, mainly actinolitic hornblende having a broader range of Mg# (0.62–0.78 ver- pyroxenites are lower than those from the other and actinolite, is not considered in this study. sus 0.65–0.68 from the western selvage augite; rocks of the lower zone, with LREE concen- Barnes, 1987). In terms of trace element abun- trations between 7× and 50× chondrites, and Augite dances, augite from the western selvage has Cr HREE concentrations between 2.5× and 12× In the following section, pyroxenes are classi- and Sr contents similar to those from the lower chondrites. Both slightly positive and slightly fi ed following the offi cial nomenclature (Morim- zone (Figs. 4E, 4F), but has higher concentra- negative Eu anomalies are present. oto et al., 1988). Both augite and diopside are tions of REEs. Augite from the southern mafi c Amphiboles in the central zone are mag- present; however, for simplicity clinopyroxene selvage has higher Sr (>22 ppm) and lower Cr nesiohornblende. In these amphiboles, Ti, P, is referred to as augite. Nonquadrilateral com- compared to the augite from the western sel- and Zr contents decrease from core to rim in

Geosphere, December 2013 1753

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Coint et al.

Lower zone Two-pyroxene andesitic roof dikes 6000 Cr 6000 A # Core Int Rim Cr C # Core Int Rim 5209 5000 5000 557 augite from 133 553 All lower zone pyroxenite and 30 4000 4000 augite melagabbro 164b 6809 but one blocks 128 704 3000 3000 833 590 555 2000 2000

1000 F 1000

0 0 REE Sr 150 TOT B 35 D

30 100 augite from 25 pyroxenite and melagabbro 20 blocks All lower zone augite but one 50 F 15 10

0 0 1020304050 0 10 20 30 40 50 Zr Zr Mafic selvages

Sr Pyroxenite and melagabbro blocks augite E Mafic selvage augite 35 two-pyroxene andesitic core int rim roof dike augite but 3 236a 30 594 25 257a 20 Upper zone augite

15 397

10 MMB-590 Lower zone augite 200

REETOT Lower zone augite F Figure 4. Diagrams displaying trace element concentra- 150 tions in augite crystals. Int—intermediate parts of the augite. (A, B) Trace element composition of augite from the lower zone and from the pyroxenite-melagabbro blocks. Expected fractionation trends (F) are displayed 100 in a schematic diagram in the lower right corners. The gray arrows in B display variable fractionation trends recorded by the augite in different samples. REE tot— 50 total rare earth elements. (C, D) Trace element compo- two-pyroxene andesitic sition of augite from the roof-zone dikes. (E, F) Trace roof dike augite but 3 element composition of augite from the maﬁ c selvages 0 and upper zone. 01020304050 Zr

1754 Geosphere, December 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Pluton assembly using trace elements in augite and hornblende

2550 ppm Cr Lower zone 2000 5209 6809 Figure 5. (A, B) Trace element plots of hornblende from 1600 128a 351 the lower zone and pyroxenite-melagabbro blocks. The striped pattern shows the range of composition found 833 133 in the upper zone hornblende. In A, the black arrows 1200 Pyroxene-rich blocks indicate analysis of hornblende close to the pyroxene 100 132 core, where the reaction is potentially incomplete and 800 Cr is inherited from the augite (see the text for more explanations). (C) Rare earth element patterns of hornblende (Hbl) from the lower zone, normalized to chon- 400 drite (Sun and McDonough, 1989). A 0 Zr B Hbl/Chondrites 120 upper zone Hornblende C

100 100

40 10

20 0 0 5000 10000 15000 20000 25000 Ce Nd Sm Gd Dy Er Yb Ti La Pr Pm Eu Tb Ho Tm Lu

every sample (Fig. 6A). Chromium concentra- found in the groundmass. Cr contents are anti- Hornblende phenocrysts in the dacitic roof tions are between 42 and 484 ppm, but only correlated to Zr and Ti and increase in concentra- dikes are magnesiohornblende to actinolitic samples from the Cuddihy basin and Dead- tion from core to rim (Fig. 7A). Small, euhedral magnesiohornblende (Barnes et al., 1987). man Lake (Fig. 1) contain hornblende with groundmass hornblende interpreted as growing These hornblende samples are similar in major Cr > 200 ppm (Fig. 6A). Hornblende crystals late in the crystallization history of the enclaves and trace element concentrations (Mg#, Ti, from both Granite Lakes and Ukonom Lake has Cr concentrations that are highly variable, Cr, Zr, V, Sc, REEs) to hornblende from the (Fig. 1) have lower Cr contents than hornblende from 60 to 800 ppm (Fig. 7A). REE concentra- upper zone (Fig. 8H) and from several samples from the other samples collected in the central tions are correlated with Ti and Zr concentrations from the central zone. This similarity includes zone (black dashed line in Fig. 6A). The horn- and decrease from core to rim. The negative Eu decreasing concentrations of Zr, Ti, and Cr from blende in the central zone has the tendency to anomalies in the cores of hornblende pheno- crystal cores to rims. be reversely zoned, with higher REE concentra- crysts are larger than in rims and in groundmass tions in the cores associated with larger negative hornblende (Fig. 7D). Hornblende crystals in DISCUSSION Eu anomalies, and lower REE concentrations in synplutonic dikes contain higher amounts of Cr, the rim with slightly negative to no Eu anoma- Zr, and Ti than are found in MME. The infl uence of orthopyroxene exsolution on lies (Figs. 6C–6H). LREE concentrations vary Hornblende from the upper zone is magne- augite composition is discussed in Supplemental between 35× and 100× chondrites, and HREE siohornblende to ferrohornblende in which Ti File 44. A simple experiment in which exsolved concentrations vary between 12× and 60× chon- and AlIV decrease from core to rim as Mg# (fi gs. cores and rims were analyzed demonstrates that drites. Two crystals, in samples WCB-7109 and 6A and 7A in Barnes, 1987) and Sc abundances the REE patterns of augite from which orthopy- WCB-6209, display lower REE concentrations (this work) remain approximately constant. The roxene has exsolved are not affected by the (10× chondrites for LREEs and HREEs; Figs. trace elements P, Hf, Y, Sr, V, and Zr all decrease amount of orthopyroxene exsolution, because 6F, 6G). Hornblende from sample WCB-5709b toward the rims (not shown). The REE abun- the proportion of the REEs that orthopyrox- (Fig. 1) has the lowest REE abundances of all dances are between 40× and 100× chondrites for ene can accommodate is small. Therefore, we samples from the central zone and displays small the LREEs and 20–60× chondrites for HREEs negative to positive Eu anomalies (Fig. 6H). (Figs. 8D–8F). Most of the REE patterns are Hornblende from MME and synplutonic subparallel to each other. Hornblende cores 4Supplemental File 4. The role of exsolution on augite REE content. If you are viewing the PDF of this dikes is magnesiohornblende (Barnes, 1987). display medium-sized negative Eu anomalies, paper or reading it offl ine, please visit http://dx.doi Phenocryst cores contain higher concentrations which decrease in the intermediate part of the .org/10.1130/GES00931.S4 or the full-text article on of Zr and Ti than in the rims or in late hornblende crystal and increase toward rims (Fig. 8C). www.gsapubs.org to view Supplemental File 4.

Geosphere, December 2013 1755

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Coint et al.

Cuddihy Lakes basin Deadman Lake Hbl/Chondrites Cuddihy Lakes basin MMB-777b 777b 4809 4909 5709b D Core Core Core Core Rim Rim Rim Rim 100 Late Late Late Granite Lake bassin Ukonom Lake 7109 6309 6209 Core Core Core Rim Rim Rim 10 Late Late Late Cuddihy Lakes basin WCB-4809 E Cr A WCB-4909

400 100

300

200 10 Ukonom Lake WCB-6209 F 100

100 0 Zr B 120 100 10 80 60 Granite Lakes WCB-6309 G 40 WCB-7109 20 100 0 0 5000 10000 15000 20000 25000 Ti Eu* C 10 Rim 1 Deadman Lake WCB-5709b H

100

0.5 Eu

Core 10 0.1 0 20 40 60 80 100 120 Ce Nd Sm Gd Dy Er Yb Zr La Pr Pm Eu Tb Ho Tm Lu Figure 6. Trace element plots from the hornblende of the central zone. (A) Cr versus Ti binary diagram. (B) Zr versus Ti diagram. The dashed line contours the hornblende analysis from the Granite Lakes basin. (C) Eu* versus Zr in hornblende. The back arrows points out the trend existing from core to rim. (D–H) Hornblende (Hbl) rare earth element (REE) patterns normalized to chondrite (Sun and McDonough, 1989) shown by location compared to the REE patterns of the hornblende from the upper zone (striped ﬁ eld) and three late hornblendes from synplutonic dikes (gray ﬁ eld). Sample numbers are indicated in italics.

1756 Geosphere, December 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Pluton assembly using trace elements in augite and hornblende

1000 Cr MME Dikes Hbl/Chondrites Syn-plutonic dikes 900 A Core Core C 800 Rim Rim Late Late 100 700 600 500 400 10 300 MME 200 100 D 100 0 1 Zr B 100 10 80 Ce Nd Sm Gd Dy Er Yb La Pr Pm Eu Tb Ho Tm Lu 60 Figure 7. (A, B) Binary diagrams displaying chemical variations in 40 hornblende from maﬁ c magmatic enclaves (MME; MMB-681b and MMB-686b) and synplutonic dikes (MMB-775, MMB776, and MMB- 20 860) compared to those found in the upper zone (striped pattern). (In the legend, “Dikes” refers to synplutonic dikes.) (C, D) Hornblende 0 (Hbl) rare earth element patterns normalized to chondrite (Sun and 0 5000 10000 15000 20000 McDonough, 1989) from maﬁ c magmatic enclaves and synplutonic Ti dikes, normalized to chondrite (Sun and McDonough, 1989).

assume that the compositions obtained by LA- out physical removal of the newly grown phase in this study because their intracrystalline varia- ICP-MS refl ect magmatic augite compositions. from the system, and to fractionation when the tions are broadly correlated with the REE data, process responsible for the chemical variation is and we therefore conclude that compositional Pluton Assembly Based on the Mineral not identifi ed. variation in the minerals is primarily a function Chemistry The amount of a trace element incorporated of the magmatic record. into a mineral lattice is governed by the parti- Mineral compositions record information tion coeffi cient between the melt and the min- Magmatic Evolution of the Lower WCb about the composition of the melt from which eral, which varies with pressure, temperature, Based on Augite Compositions the mineral grew. The melt composition is, in and magma composition. Inasmuch as both turn, infl uenced by the minerals that precipi- augite and hornblende were high-temperature High bulk-rock concentrations of MgO and tate from it. These minerals need not be physi- phases in the various parts of the WCb, the two Sc in the lower zone indicate that some of cally fractionated (i.e., to be removed from the minerals recorded the evolution of the melt from these rocks are cumulates, and therefore that system) to induce compositional variations in which they crystallized, and therefore allow us fractional crystallization affected lower zone newly formed growth zones of existing miner- to identify and map the extent of the distinct magmas. Augite was one of the fi rst minerals als. Therefore, core to rim compositional varia- magma batches in the batholith. Distinct magma to crystallize, along with plagioclase (Fig. 2A), tions in minerals can result from a variety of batches should produce minerals with distinct and may be used to track the evolution of the magmatic processes such as fractional crystalli- compositional and zoning patterns and magma magma compositions. Hornblende, which is zation, mixing, and assimilation–fractional crys- mixing should result in discernible composi- present as a rimming phase around pyroxene, tallization, assuming that intracrystal diffusion tional reversals. grew at lower temperature conditions and is a is slow enough for the zoning patterns to be pre- In this study, the core to rim variations in less valuable indicator of changes in melt com- served (see following). Bulk-rock compositions, trace element abundances are used to under- position in this zone (Fig. 2A). unlike mineral ones, will also record processes stand the evolution of the melt composition The abrupt core to rim decrease in Cr and the

such as accumulation. Therefore, both data sets through time. Because intracrystalline diffusion proportional increase in REETOT and associated need to be combined to understand in detail the of REEs in diopside is very slow under crustal increasing Zr concentrations in augite (Figs. effects of these different processes. Here we conditions (Van Orman et al., 2001), the REE 4A, 4B) are consistent with evolution of mag- refer to fractional crystallization when there is record preserved in both hornblende and augite mas in the lower zone by fractional crystalliza- chemical evidence for separation between the is used as a proxy for the melt evolution. Ele- tion. However, the extent of fractional crystal- minerals and the residual melt, to in situ frac- ments such as Ti (Cherniak and Liang, 2012), lization is not the same in every sample. In some tionation when chemical variation was induced Sr, and Cr are more sensitive to intracrystalline samples, augite displays wide compositional by precipitation of minerals from the melt with- diffusion. However, these element data are used variations (e.g., WCB-5209), whereas in others

Geosphere, December 2013 1757

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Coint et al.

Zr Hbl/Chondrites 100 A Qtz monzodiorite MMB-317 D MMB-208 80 100

10 20

0 Cr 300 B Tonalite MMB-687 E

100 200

100

10 0 0 5000 10000 15000 20000 25000 MMB-394 Granodiorite-granite Ti MMB-471 F

100 1.1 Eu/Eu* MMB-397 C

0.9

10 0.7

Tonalite MMB-397 0.5 G

100 0.3 Low LREE Hbl patterns 0.1 050100 Zr 10 Dacitic/rhyodacitic roof dikes MMB-548 Roof dike MMB-397 Upper zone H MMB-551 Hbl cores MMB-693 100 Hbl intermediate Hbl rims Smaller Hbl

Upper zone Hbl Late granite Hbl RZD Hbl 10 Ce Nd Sm Gd Dy Er Yb La Pr Pm Eu Tb Ho Tm Lu

Figure 8. Trace elements in hornblende (Hbl) from the upper zone. (A–C) Binary diagrams displaying chemical variations in hornblende from the upper zone compared to hornblende from the late granite and hornblende from the dacitic roof dikes (RZD). In C, the red arrow points out the evolution trend for hornblende from the upper zone, from core to rim, whereas the black arrow follows the trend from core to rim described by hornblende in sample MMB-397. LREE—light rare earth elements. (D–H) Hornblende REE patterns normalized to chondrite (Sun and McDonough, 1989). Black arrows show the LREE–depleted patterns (see the text for explanations) interpreted as late rims crystallizing from strongly fractionated residual melt. Qtz—quartz.

1758 Geosphere, December 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Pluton assembly using trace elements in augite and hornblende

(e.g., WCB-128) variations are limited (Figs. formed hornblende and the melt might not be Maﬁ c Selvage 4A, 4B). Furthermore, compositional trends for reached. Other cations with 3+ and 4+ valences, both augite and hornblende are distinct from one such as Al, Ti and REEs, could also be inherited The augite crystals found in the western sample to the next, as is evident from the vari- from preexisting augite. However, Supplemen- selvage (e.g., sample MMB-594) display geo-

able slopes in plots of REETOT versus Zr (Figs. tal File 1 (see footnote 1) shows that the abun- chemical characteristics similar to those from 4A, 4B, 5A, and 5B). These different slopes are dances of these elements are generally higher the lower zone (Figs. 4E, 4F). This similarity interpreted to indicate that each sample crystal- in hornblende than in augite, which indicates and the fact that the western selvage was coeval lized from a distinct magma batch and that each that they are more readily partitioned into horn- with the lower zone are interpreted to indicate batch had a distinct crystallization history. This blende. Therefore, it is probable that the concen- that the western selvage formed from magma variation can in part be explained by differences trations of these elements in the latter represent batches similar to those that formed the lower in initial magma composition, and can also be equilibrium between the growing hornblende zone of the pluton. related to differences in the extent of fractional and melt, whereas Cr concentrations reflect Rocks from the southern maﬁ c selvage show crystallization. Such differences could result inheritance from precursor augite. This example much more diversity in texture, bulk-rock com- from variable cooling rates or from the effective- shows the importance of combining textural positions (Fig. 3), and mineral compositions ness of crystal-liquid separation. What is clear observations with the geochemistry of mineral (Figs. 4E, 4F). Augite analyzed in samples is that compositional variations of individual phases when interpreting the nature of peritectic MMB-257a and MMB-236a overlap with magma batches can result in the distinct augite reactions. augite from the two-pyroxene andesite roof compositional trends. We conclude that variable dikes in terms of Sr concentrations and display

augite compositions in the lower zone refl ect Pyroxene-Rich Blocks and Their similar trends in REETOT versus Zr (Figs. 4E, emplacement of multiple magma batches, the Relationships with the Lower Zone 4F). Therefore, we interpret the southern mafi c compositions of which ranged from gabbroic to selvage to be related to the two-pyroxene andes- tonalitic. These compositional variations com- Pyroxene-rich blocks and intrusive bodies itic roof dikes. bined with differences in the extent of fractional in the lower zone can be either (1) cumulates crystallization resulted in the observed range of resulting from disruption of dikes, or (2) frag- Upper Zone augite trace element abundances and patterns. ments of ultramafi c host rocks that reacted with the host magma, resulting in the crystallization Bulk-rock compositions from the upper zone Lower Zone Hornblende: Magmatic or of pyroxenes. The composition of the olivine of the WCb display narrow trends (Fig. 3).

Inherited Compositions? found in sample MMB-100 (forsterite, Fo71) is Therefore, fractional crystallization and asso- not appropriate for mantle rocks (as pointed out ciated crystal accumulation were interpreted On the basis of textural relationships, horn- in Barnes, 1983, 1987), leaving the fi rst hypoth- to be the main processes responsible for bulk blende crystals in samples from the lower zone esis as the most likely explanation. The compo- compositional variation (Coint et al., 2013). grew late in the magmatic history (Fig. 2A), and sition of the augite in the lower zone overlaps In the upper zone, allanite, apatite, and zircon so are more likely to have recorded processes with augite compositions from the pyroxenite are accessory phases. Zircon and apatite occur that occurred at lower temperatures relative to and melagabbro bodies (Figs. 4A, 4B), and this as inclusions in biotite, hornblende, and pla- those preserved in augite. The Cr contents of similarity indicates that augite in these bodies gioclase, suggesting that these minerals were some of the hornblende (Fig. 5A) are anoma- crystallized from magmas similar to those that stable at temperatures above the solidus and lously high for crystals that formed from a low- formed the lower zone. The fi eld relationships were likely to have fractionated during crystal- temperature, evolved melt, and are much higher and the textural observations suggest that this lization. Fractionation of hornblende along with than the Cr concentrations found in the upper overlap is mostly the result of crystal exchange the accessory minerals will result in decreasing zone hornblende. These Cr concentrations between the pyroxene-rich bodies and the host concentrations of the REEs in the residual melt, partly overlap with the Cr contents of the lower magmas. For example, sample WCB-133 dis- and therefore in the minerals that grow from the zone augite crystals (cf. Figs. 4A and 5A). A plays a gradational contact with a large block progressively fractionated melt. One of the con- few high-Cr hornblende crystals are also found of pyroxene melagabbro. All analyzed augite sequences is that incompatible elements whose in the central zone (Fig. 6A) and in one rock cores and mantles in sample WCB-133 have enrichment can be used to unequivocally track from the upper zone (Fig. 8B). These high Cr high Cr concentrations and are interpreted as the evolution of the system are absent. However, contents may result from incomplete reaction being inherited from the adjacent melagabbro, the amount of Ti incorporated into hornblende between melt and pyroxene to form hornblende whereas rims overlap with augite compositions depends on temperature (Otten, 1984; Ernst and or from direct crystallization of hornblende typical of lower zone augite (Fig. 4A). These Liu, 1998; Femenias et al., 2006), and therefore from a Cr-rich mafi c melt. In the lower zone, rims are interpreted to form after augite from is used as a proxy to track the down-temperature the Cr-rich hornblende forms clusters or is asso- the pyroxene-rich block was transferred to the variations of trace elements (Fig. 8). ciated with quartz beads. Both of these textural magma that hosted the block. The high Cr con- Hornblende from the upper zone shows a features result from pyroxene + melt reactions tents and high Mg# (~0.8 instead of values of limited variation in Mg# (Barnes, 1987), REE (Castro and Stephens, 1992; Stephens, 2001), 0.65–0.75 typical of most lower zone rocks; concentrations, shape of the REE patterns suggesting that the high Cr content of the horn- Barnes, 1987) of the augite in pyroxenite and (Figs. 8D–8H), and Cr contents (Fig. 8B). blende is inherited from the original pyroxene. melagabbro bodies and the high Sr contents of The abundance of REE elements and the size The extent of this inheritance is not well con- these augite crystals compared to the rest of the of the Eu anomaly tend to decrease from core strained. Cr is present in larger concentrations lower zone augite (Figs. 4A, 4E) suggest that to mantle (Fig. 8C), but hornblende rims tend in augite than in hornblende; therefore if diffu- parent magmas of the pyroxene-rich bodies to display larger Eu anomalies than the man- sion is slow, or if the reaction takes place rap- were relatively primitive and water rich, consis- tle. This core to rim decrease in trace element idly, complete equilibration between the newly tent with the paucity of olivine in these rocks. abundances can be explained in terms of the

Geosphere, December 2013 1759

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Coint et al.

compatible behavior of the REEs, although the variations in REE compositions than the rest of blende from the upper zone, with moderate Cr slight decrease in the Eu anomaly from core to the upper zone hornblende (cf. Figs. 8D, 8E, concentrations and decreasing Ti and Zr from mantle (Figs. 8D–8H) is counterintuitive. In the 8F). This rock is also distinct from other ana- core to rim (Figs. 6A, 6B). In samples WCB- upper zone, fractionation of a small amount of lyzed samples because of the presence of relict 4809 and WCB-4909, hornblende REE patterns apatite and allanite is thought to counteract the pyroxene as inclusions in poikilitic hornblende overlap perfectly with the REE patterns of horn- expected increase in the size of the negative Eu and plagioclase with reverse zoning (Barnes, blende from the upper zone (Fig. 6E). These anomaly. We present simple models to illustrate 1987). These features suggest a distinct mag- data suggest that hornblende in WCB-4809 and this effect in Supplemental File 55. The presence matic history compared to the rest of the upper WCB−4909 crystallized from magmas similar of large (1 mm diameter) allanite crystals in the zone. The composition of the relict pyroxene in composition to the upper zone magma and granodioritic and granitic rocks suggests that in (Figs. 4E, 4F) indicates that the pyroxene are consistent with the age of zircon from WCB- some cases the proportion of allanite that frac- grew from a magma similar to magmas that 4909, which is identical to upper zone samples. tionated may be greater than the value used in formed the lower zone and the western mafi c These two samples were collected from the the model (see Supplemental File 5 [see foot- selvage (e.g., MMB-594; Fig. 1). Hornblende structurally lower part of the central zone (Fig. note 5]). In such a case, allanite fractionation cores and mantles have compositions similar 1) and they are organized in a series of alternat- will deplete the residual melt in LREEs relative to hornblende from the central and the upper ing comagmatic sheets (fi g. 3F in Coint et al., to the HREEs. This depletion effect can explain zones, and are interpreted to have grown from 2013). They are, therefore, interpreted be part of the presence of a few hornblende rims that have upper zone magma. In contrast, the rims of the feeder system to the upper zone. lower LREE concentrations than typical (Figs. these hornblendes display steeper REE pat- Five other central zone samples contain horn- 8D, 8F). The larger negative Eu anomaly associ- terns with enrichment in LREEs but without blende having REE patterns that show more ated with some of the LREE-depleted patterns, Eu anomalies. The REE patterns of hornblende variability. For example, hornblende cores in compared to the rest of the hornblende patterns rims are interpreted as resulting from crystal- these fi ve samples commonly overlap composi- from the upper zone, cannot be accounted for by lization from a hybrid magma that formed by tions of hornblende from the upper zone; how- allanite fractionation, but could be the result of mixing of upper zone magma with a crystal- ever, their rim compositions have lower REE rare instances of in situ K-feldspar fractionation rich lower zone–type magma batch from the contents and display smaller Eu anomalies than (e.g., Bachmann et al., 2005). This possibility is adjacent mafi c selvage. Most of the textural in upper zone samples (Figs. 6D, 6F–6H). One consistent with the fact that most of the horn- characteristics of upper zone rocks are present sample, MMB-777b, shows a bimodal distribu- blende rims with relatively large Eu anomalies in this rock: medium grain size (2 mm), poi- tion of the REE patterns (Fig. 6D). The com- are from granitic or granodioritic samples. kilitic to interstitial K-feldspar, large (at least positional overlap between hornblende cores The abundances of elements such as Zr and 100 µm long), subequant zircons, euhedral from the central zone with hornblende from Ti in upper zone hornblende are well correlated hornblende, and glomerocrysts of zoned pla- the upper zone is strongly suggestive that these and defi ne a single trend (Fig. 8A). By com- gioclase. All of these features indicate that the hornblende cores grew from upper zone mag- parison with the lower zone (Fig. 4B), which upper zone magma accounted for much of the mas. However, unlike WCB-4809 and WCB- represents multiple magma batches with limited mass of this rock. The presence of pyroxene 4909 and most of the hornblende in the upper chemical communication, this single trend of crystals similar in composition to ones found zone, the rims of central zone hornblende have upper zone hornblende compositions indicates in the western selvage (MMB-594; Figs. 4E, lower REE contents and small negative to no Eu that upper zone hornblende crystallized from (1) 4F) as inclusions in hornblende, plus abundant anomalies (Fig. 6C). These zoning features can a single batch of magma or (2) multiple magma hornblende grains that contain actinolite ± be explained in several ways. batches that were compositionally identical and oxide inclusions are interpreted to be the result had identical crystallization histories. The well- of reaction between magma of the upper zone Fractional Crystallization defi ned bulk-rock trends in Harker diagrams with augite from the western selvage. If this is (fi gs. 3 and 8 in Coint et al., 2013) coupled the case, it would suggest that the upper zone It has already been demonstrated that in with the decreasing An content of plagioclase magma invaded mushy magma of the western the upper zone REEs were compatible and

(An46 to An12; Barnes, 1987) with increasing selvage and incorporated some of its minerals, concentrations generally decreased with frac-

bulk-rock SiO2 concentrations are consistent such as augite. This interpretation is consis- tional crystallization. In the central zone, min- with either interpretation. However, the lack of tent with hornblende compositional variations erals such as K-feldspar and allanite, which intrusive contacts in the upper zone compared from core to rim. Hornblende cores grew from were involved in the evolution of upper zone with the abundance of intrusive contacts in the upper zone magma, whereas the rims crystal- magma, are scant. Therefore, the variations central and lower zones, and the gradual upward lized from a more maﬁ c, hybrid magma, which in the rims of the hornblende in the central change from tonalitic to granitic compositions resulted in variations in the LREE pattern part cannot be associated with fractionation among upper zone rocks support crystallization shape and smaller negative Eu anomalies than of such minerals. Textures of MMB-777b, from a large homogeneous magma body. found in most upper zone hornblende. WCB-6209, WCB-6309, and WCB-7109 are Hornblende from one upper zone sample, also distinct from the upper zone samples. In MMB-397 (Figs. 1 and 8G), displays wider Central Zone the latter samples, hornblende and biotite are poikilitic and contain numerous inclusions of The central zone of the intrusion plays a key plagioclase, hornblende, and apatite, and relict 5Supplemental File 5. Fractional crystallization role in understanding the link between the upper pyroxene cores are common. These features model of apatite and allanite and the consequence on and lower zones because, as indicated above, it are rarely observed in the upper zone rocks, the size of the Eu anomaly. If you are viewing the PDF of this paper or reading it ofﬂ ine, please visit http:// constitutes a transition between the two. Horn- so it is unlikely that hornblende compositions dx.doi.org/10.1130/GES00931.S5 or the full-text arti- blende in the central zone displays major and from these samples result from simple frac- cle on www.gsapubs.org to view Supplemental File 5. trace element concentrations similar to horn- tional crystallization.

1760 Geosphere, December 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Pluton assembly using trace elements in augite and hornblende

Mixing between MME, Synplutonic Dikes, tant in the central zone, then the resulting horn- 8F, 8G). We therefore conclude that the rocks and the Upper Zone Magma blende compositions should be between those from the Ukonom Lake and Granite Lakes areas of the groundmass hornblende from the synplu- of the central zone formed by mixing between A second possibility is that hornblende rims tonic dikes and the hornblende from the upper a pyroxene-bearing magma and upper zone in samples MMB-777b, WCB-6209, WCB- zone (Fig. 7A). However, such mixing does not magma. The nature of the pyroxene-bearing end 6309, and WCB-7109 grew from hybrid magma. explain the low Cr contents of hornblende rims, member is diffi cult to constrain as no obvious Ample evidence of mixing and mingling is pres- the wide range of Cr contents, and the lack of mixing trends are observed in Harker diagrams ent in the central zone, such as disrupted syn- Ti increase in phenocryst rims and groundmass (fi g. 8 in Coint et al., 2013) or in Figures 6, 7, plutonic dikes and swarms of MME (Barnes et hornblende (Fig. 6A). These features indicate and 8. However, these samples were collected al., 1986a). The presence of pyroxene cores sur- that the potential mixing end member was not in proximity to the gradational contact with the rounded by hornblende (Fig. 2D) would suggest primitive (in the sense of containing high Cr and lower zone, which is a good candidate for a mix- that mixing could have occurred between lower Ti contents) and therefore not a synplutonic dike ing end member. zone and upper zone magmas. magma. Instead, the mixing end member was In summary, most of the central zone crystal- The extent of the interactions between parent probably of intermediate composition. lized from upper zone magma, and we interpret mafi c magmas of MME and synplutonic dikes Hornblende in central zone sample WCB- the sheeted part of the central zone preserved in (mainly basaltic andesite) and central zone mag- 5709b is unlikely to be the result of mixing the Cuddihy basin to be part of the feeding system mas is diffi cult to constrain for the following of upper zone magma with synplutonic dike to the upper zone (see Coint et al., 2013). Some reasons. The cores of hornblende phenocrysts magma because hornblende in this sample has samples, particularly ones collected near the in the MME have compositions similar to horn- higher LREE and lower middle REE and HREE boundary with the lower zone (samples MMB- blende cores from central and upper zone rocks concentrations than the hornblende from the 777b, WCB-7109, WCB-6309, and WCB-6209), (Fig. 7). This similarity is interpreted to indi- synplutonic dikes (Fig. 6H). The low middle are primarily the result of mixing between lower cate that hornblende phenocrysts in MME are and heavy REE concentrations (Fig. 6H) and the and upper zone magmas, with local complica- xenocrysts from the surrounding magma that small Eu anomaly suggest that the hornblende tions due to further mingling and mixing with were mechanically incorporated into the MME in WCB-5709b crystallized from a relative magmas parental to the MME and synplutonic magmas. One of the MME from the upper part mafi c magma. Relict pyroxene, now completely mafi c dikes. Among the analyzed samples, WCB- of the pluton has a groundmass of centimeter- transformed into hornblende, are abundant in 4809 and WCB-4909 are the only ones that show scale, poikilitic K-feldspar crystals that are opti- the sample, and we interpret this feature to indi- no evidence of input from mafi c magmas. cally continuous with K-feldspar crystals in the cate that sample WCB-5709b is a fragment of host granodiorite (Fig. 2H). This relationship the lower zone now surrounded by central zone Roof Dikes suggests that many MME are complex hybrids rocks, an interpretation consistent with the age and that mixing involved not only exchange of of central zone sample Z5, which is identical to Two-Pyroxene Andesitic Roof Dikes large crystals, but also fl ow of interstitial melt ages of lower zone samples. Some of the two-pyroxene andesitic dikes are between enclave and host magma. For these rea- composite, with older external two-pyroxene sons, the MME are not viewed as appropriate Mixing between the Lower and Upper Zone andesite and inner biotite granodiorite related to mixing end members. Magmas the upper zone (Barnes et al., 1986a; Coint et al., In contrast, hornblende in the synplutonic 2013). Other two-pyroxene andesitic dikes cut dikes is more likely to refl ect compositions of A second mixing model for the central zone the uppermost part of the upper zone (Coint et al., the mafi c magmas because there was evidently involves mixing between magmas of the upper 2013). These mutual crosscutting relationships much less interaction between the dikes and and lower zones. Texturally, the central zone indicate that the two-pyroxene andesitic mag- the surrounding magma, as is indicated by the rocks are possibly hybrids between the lower mas were coeval with the magmas of the upper sharp contacts between dikes and their host zone, which contains pyroxene partially reacted zone. The previous interpretation of the origin of rocks (fi g. 3E in Coint et al., 2013). As is the to actinolitic hornblende or hornblende (Fig. the two-pyroxene andesitic dikes (Barnes et al., case in the MME, the cores of hornblende phe- 2D), and the upper zone, which contains coarse 1986a) was that they represented magmas that nocrysts found in synplutonic dikes have REE phenocrysts of euhedral zoned plagioclase and were derived from the lower zone. This interpreta- compositions similar to those of hornblende in hornblende (Fig. 2E). The gradational contacts tion was based on the broad overlap of bulk-rock the host magma (Fig. 7C), suggesting that the between the different zones of the intrusion and compositions of the dikes with the lower zone hornblende phenocrysts in the dikes were inher- abundant evidence for mixing in the central zone (Barnes et al., 1986a). However, augite crystals ited from their hosts. However, phenocryst rims indicate that this hypothesis needs to be tested. in the andesitic roof dikes have compositions that and groundmass hornblende in the synplutonic Unfortunately, most of the relict pyroxenes are are distinct from augite in the lower zone, with dikes have high Cr contents (Fig. 7A) yet lack partially transformed into actinolite and do not higher Sr and lower Cr contents (Figs. 4C, 4D). evidence for reaction from pyroxene. This lack preserve magmatic compositions. Augite from the andesitic dikes is also distinct indicates that the Cr concentrations observed Sample MMB-397 was interpreted to result in showing zoning reversals (e.g., MMB-164b; in the groundmass hornblende and hornblende from interaction between the mafi c magma Figs. 4C, 4D), whereas most lower zone augite rims are a refl ection of the concentration of Cr that formed the western selvage (composition- crystals are normally zoned (Figs. 4A, 4B). The in the dike magma. Therefore, we consider the ally and temporally related to the lower zone; zoning reversals indicate that many of the two- compositions of hornblende rims and ground- see Fig. 1) and upper zone magma. Hornblende pyroxene andesitic roof dike magmas underwent mass hornblende to be representative of a pos- from sample MMB-397 has compositions and a mixing event, a conclusion that is consistent sible mixing end member. zoning patterns similar to hornblende in sam- with the common occurrence of orthopyroxene If mixing between the central zone magma ples from Ukonom Lake and Granite Lakes phenocrysts rimmed by augite and zoning rever- and the synplutonic dike magmas was impor- (WCB-6209, WCB-6309, WCB-7109; Figs. sals in plagioclase phenocrysts (Barnes, 1987).

Geosphere, December 2013 1761

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Coint et al.

The absence of high-Al orthopyroxene in these (Figs. 4A, 4B) can be explained by variation in over a range of rock types and structural levels. dikes (Barnes, 1983, 1987) suggests that mixing the compositions of individual magma batches, These similarities suggest that the broadly tonal- was not a deep-crustal event and was more likely different cooling rates of the individual magma itic magmas initially emplaced in the central and to occur in a shallower reservoir. batches, and variable segregation of interstitial upper zones were very similar in bulk composi- Although the compositions and zoning pat- melts. Local gradational contacts between tex- tion. Convective mixing inferred for the upper terns in the andesitic roof dikes rule out an ori- turally distinct rocks indicate that some of the zone did not signifi cantly affect the central zone, gin from lower zone magmas, it is possible that magma batches were signifi cantly above their because the heterogeneous nature of the central the dike magmas came from the central zone, solidus when adjacent batches were emplaced, zone (e.g., deformed synplutonic dikes, MME the only part of the pluton with abundant evi- and this relationship is exemplifi ed by inheri- swarms) is preserved. Evidently, these similar dence for mixing and mingling. Pyroxene in the tance of augite from pyroxenitic blocks in adja- tonalitic magmas in the central and upper zones central zone is rarely preserved, which makes cent quartz diorite and tonalite. gained their compositional features in structur- comparison with pyroxene in the roof dikes dif- The andesitic to dacitic upper zone magmas ally deeper levels of the magmatic system. fi cult. Zoned augite crystals in roof-zone andes- were more buoyant than the rocks and mushes Convection in the upper zone permitted growth ite (e.g., MMB-164b) show an abrupt increase in the lower zone and so were emplaced above of hornblende and plagioclase phenocrysts from in Cr contents that suggests mixing with rela- the lower zone, into rocks of the eastern Hay- a broadly homogeneous magma body. However, tively primitive mafi c magma. No evidence of fork and western Hayfork terranes (Figs. 9B, once large-scale convection stopped (Fig. 9E), such mafi c input has been recorded by horn- 9C). Local mixing between the lower zone and either due to increased crystallinity or loss of blende in the central part of the intrusion. We upper zone magmas occurred, such as along the mafi c heat input, upward percolation of residual therefore conclude that the source of the two- western mafi c selvage and in the central zone melt through the mush resulted in the broad pyroxene andesitic magmas was a part of the (Fig. 9C). Mixing created hybrid rocks with upward zoning of the upper zone, in a manner batholith that is not exposed. This zone of the relict pyroxene from the lower zone surrounded similar to that proposed for monotonous ignim- batholith most probably underlies the southern by euhedral hornblende crystals, the composi- brites (Christiansen, 2005). This process resulted and southwestern portion of the upper zone. tions of which indicate an affi nity with the upper in the lower part of the upper zone becoming a zone. In the meantime, hot basaltic andesitic was partial cumulate, upward increase in the abun- Dacitic and Rhyodacitic Roof Dikes emplaced into the central zone, where it locally dances of quartz and K-feldspar, and develop- Dacitic and rhyodacitic roof dikes were inter- mingled with the host magma and reheated the ment of euhedral quartz. Percolation of residual preted as originating from the upper zone magma system (Fig. 9C). Ages obtained on the central melt through a mush can also explain why the (Barnes et al., 1986a; Coint et al., 2013) on the zone are in agreement with the geochemical upward zoning is not systematic (fi g. 2 in Coint et basis of their bulk-rock compositions, which data presented here, in that they overlap with al., 2013), because melt percolation would depend overlap compositions of the upper zone rocks ages for both the upper and lower zone (Fig. 9 on local conditions of porosity and permeability. (fi g. 8 in Coint et al., 2013). In addition, horn- and Coint et al., 2013). In contrast, upward zoning due to emplacement blende phenocrysts in the dacitic and rhyodacitic Several studies based on numerical modeling of successively more evolved magmas in subhori- roof dikes are compositionally identical to horn- and natural examples have suggested that mix- zontal layers should result in much more system- blende from the upper zone in terms of major ing due to injection of mafi c magma into silicic atic zonation (e.g., Bartley et al., 2008). Moreover, and trace element abundances, and display the magma is unlikely to be extensive, leaving the emplacement of variably evolved magma batches same zoning relationships (Fig. 8H). The pres- mafi c magma as a source of heat and fl uid (Rob- would not be expected to result in crystallization ence of these dikes structurally above the upper inson and Miller, 1999; Bachmann et al., 2002; of hornblende with uniform composition through- zone indicates that the upper zone magma was Huber et al., 2009; Ruprecht et al., 2012). Bur- out the thickness of the upper zone. eruptible. Presumably, the compositional range gisser and Bergantz (2011) demonstrated that We therefore conclude that the upper zone of dacitic to rhyodacitic roof dikes refl ects a the heat generated by a small volume of mafi c was once a large, convecting body of chemi- temporal change in composition of differentiated material emplaced at the base of a crystal-rich cally and physically interconnected magma that magmas in the uppermost part of the upper zone. silicic magma can cause rapid remobilization of crystallized compositionally very similar horn- the overlying mush (unzipping), in a few days to blende and plagioclase (Fig. 9D). Upward zon- IMPLICATIONS FOR THE PLUTON several months after the intrusion of the mafi c ing occurred after initial crystallization of horn- ASSEMBLY AND COMPARISON WITH material. This remobilization results in homog- blende and plagioclase phenocrysts by upward OTHER SYSTEMS enization of the felsic magma due to convec- percolation of residual melt (Fig. 9F). tion (Robinson and Miller, 1999). If this type of Our results indicate that the lower zone mobilization occurred in the upper zone magma, CONSEQUENCES FOR VOLCANIC resulted from emplacement of many individual the scant evidence for resorption in plagioclase, SYSTEMS: DID WCb MAGMAS ERUPT? magma batches of broadly andesitic composi- hornblende, and quartz suggests that heating did tion. The CA-TIMS ages cited here (presented not rejuvenate a locked mush such as in some The abundance of mafi c through felsic dikes in detail in Coint et al., 2013) show that this large dacitic eruptions (Bachmann et al., 2002), in the roof zone (Figs. 2E, 2F) of the batholith zone was emplaced through a short period of but instead initiated the homogenization of leads to the conclusion that some WCb mag- time, from 159.22 ± 0.10 Ma to 158.99 ± 0.17 magma having crystallinity low enough to permit mas escaped the system and probably erupted. Ma. The predominantly north-south strike and free fl ow (Fig. 9D; Robinson and Miller, 1999). The fi rst magmas to erupt (i.e., roof dikes) are steep dips of internal contacts and foliation in Hornblende in the central zone, the inferred comparable to the lower part of the system, and the lower zone are suggestive of batch-wise feeder system of the upper zone, is composi- were mainly basaltic andesitic in composition. magma emplacement in subvertical sheets tionally nearly identical to hornblende in the Few clear-cut examples exist because many (Fig. 9A; Coint et al., 2013). The range of rock upper zone, and hornblende in the upper zone is of the oldest mafi c roof dikes were intensely compositions and texture from sample to sample essentially identical from one sample to the next altered by thermal effects of emplacement of the

1762 Geosphere, December 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Pluton assembly using trace elements in augite and hornblende

Depth (km) P (kbar) 159.32 to 158.82 MaMMB-590 A 159.32 Ma B SSW NNE SSW NNE

2 2 10 10 4 4 20 20 6 6 Figure 9. Emplacement model for the Wooley Creek system (WCb—Wooley 30 8 30 8 Creek batholith). Ages reported here are chemical abrasion–thermal ionization mass 10 10 spectrometry (CA-TIMS) U-Pb zircon ages 40 40 from Coint et al. (2013). They are presented 12 12 as the oldest to youngest ages for each step andesite dacite of the pluton assembly accounting for the 2σ to 158.82 Ma 158.71 Ma error. P—pressure. (A) Emplacement of the C D lower zone as individual magma batches. SSW NNE SSW NNE The black lines between the different Western magma batches indicate that the interaction selvage 2 2 between these particular batches with the 10 MMB-594 10 surrounding ones were limited; however, 4 4 they do not indicate the presence of a sharp 20 20 contact in the ﬁ eld. (B) Emplacement of the 6 6 central zone as individual magma batches. (C) Emplacement of the upper zone and 30 8 30 8 interaction between the upper and lower zone magmas in the western maﬁ c selvage 10 10 followed by the arrival of the basaltic ande- 40 40 sitic magma in the central zone and emplace- Andesitic magma andesitic basaltic ment of the two-pyroxene andesitic batches 12 12 basaltic andesite magma andesite at depth. (D) Basaltic andesite emplacement in the central zone triggers convec- to 158.71 Ma 156.71 Ma to 154.79 Ma E domes G tion and homogenization in the upper zone. SSW Fig. 2e NNE SSW NNE (E) Development of the upper zone as a large batch of magma of relatively homogeneous Fig. 2c & d eHt composition. (F) Evolution of the upper zone wHt 2 2 by fractional crystallization with develop- 10 10 ment of a broad upward zoning with more Dacite Fig. 2f & g RCt 4 4 felsic rocks occurring toward the top of the 2 mica 20 20 intrusion. Eruption of more two-pyroxene granite 6 6 andesitic dikes and emplacement of the late RCt and two-mica granite occurred later. The 30 30 8 8 dashed line is the current level of exposure. RCt—Rattlesnake Creek terrane; eHt— 10 eastern Hayfork terrane, wHt—western 40 10 40 late andesitic Hayfork terrane. andesitic basaltic basaltic granite magma 12 magma andesite 12 andesite

Basaltic andesite Mafic magmatic enclaves

Lower zone and Upper WCb Two- Slinkard pluton Rhyolite/Granite Dacite Pyroxene Andesite Andesite Pyroclastic flow

Geosphere, December 2013 1763

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Coint et al.

upper zone. However, dike samples such as fi ne- emplacement of multiple batches of magma in Annen, C., 2009, From plutons to magma chambers: Thermal constraints on the accumulation of eruptible silicic magma grained gabbro MMB-590 (Barnes et al., 1990) the middle crust, magma batches that under- in the upper crust: Earth and Planetary Science Letters, indicate that lower zone magmas reached high went very limited homogenization. As a result, v. 284, p. 409–416, doi:10.1016/j.epsl.2009.05.006. levels of the system and were capable of erupt- individual magma batches can be identifi ed Bachmann, O., Dungan, M., and Lipman, P., 2002, The Fish Canyon magma body, San Juan Volcanic Field, Colo- ing. In typical volcanic systems, such magmas on the basis of augite compositional variation. rado: Rejuvenation and eruption of an upper-crustal would have formed scoria cones and lava fl ows; The central zone of the pluton corresponds to batholith: Journal of Petrology, v. 43, p. 1469–1503, however, the volume ejected would have been a complex mixing and transition zone, where doi:10.1093/petrology/43.8.1469. Bachmann, O., Dungan, M.A., and Bussy, F., 2005, Insights limited (Figs. 9A, 9B). The lateral extent of the magmas from the lower zone interacted with into shallow magmatic processes in large silicic lower zone suggests that volcanism was unlikely the upper zone magmas. It is also a part of the magma bodies: The trace element record in the Fish Canyon magma body, Colorado: Contributions to Min- to be focused at one volcano, but rather formed a intrusion where replenishment of mafi c magmas eralogy and Petrology, v. 149, p. 338–349, doi:10.1007 volcanic fi eld with several eruptive vents, releas- occurred. The mafi c magmas provided heat (and /s00410-005-0653-z. ing magmas of variable compositions. possibly fl uids) to the upper part of the system, Bacon, C.R., and Druitt, T.H., 1988, Compositional evolution of the zoned calc-alkaline magma chamber of Mount As the dacitic magma batches that formed the but interaction of the mafi c magmas with the Mazama, Crater Lake, Oregon: Contributions to Min- upper zone arrived in the system, it is possible host dacitic (upper zone) magmas was limited. eralogy and Petrology, v. 98, p. 224–256, doi:10.1007 that these batches erupted, forming part of the The upper zone crystallized from convecting, /BF00402114. Barnes, C.G., 1982, Geology and petrology of the Wooley roof dike system observed today. The emplace- relatively homogeneous, dacitic magma. Once Creek batholith, Klamath Mountains, northern Califor- ment of several batches of dacitic magma would convection stopped, fractionation occurred by nia [Ph.D. thesis]: Eugene, University of Oregon, 214 p. Barnes, C.G., 1983, Petrology and upward zonation of the suggest that eruptions would not necessarily be melt percolation through the mush pile, resulting Wooley Creek batholith, Klamath Mountains, Cali- focused in a single edifi ce, but probably at several in a broad, nonsystematic upward zoning of the fornia: Journal of Petrology, v. 24, p. 495–537, doi: vents. At the same time, batches of two-pyroxene unit with more mafi c, partly cumulative rocks in 10.1093/petrology/24.4.495. Barnes, C.G., 1987, Mineralogy of the Wooley Creek batho- andesite were emplaced underneath the southern structurally lower parts of the zone and progres- lith, Slinkard pluton, and related dikes, Klamath Moun- part of the intrusion; these magmas rose into the sively more evolved granodiorite and granite tains, northern California: American Mineralogist, v. 72, uppermost part of the upper zone and the roof toward the roof. Roof dikes of dacitic composi- p. 879–901. Barnes, C.G., Allen, C.M., and Saleeby, J.B., 1986a, Open- zone to form the two-pyroxene andesite dikes. tion indicate that the upper zone magmas were and closed-system characteristics of a tilted plutonic Arrival of these porphyritic magmas at the sur- eruptible and escaped the underlying magma system, Klamath Mountains, California: Journal of Geo- face could have formed thick lava fl ows (Fig. 9C). chamber. Unlike many models for incremental physical Research, v. 91, p. 6073–6090, doi:10.1029 /JB091iB06p06073. In this model, emplacement of basaltic andes- assembly of plutons, we can demonstrate that Barnes, C.G., Rice, J.M., and Gribble, R.F., 1986b, Tilted ite in the central WCb resulted in convection fractional crystallization was a viable process in plutons in the Klamath Mountains of California and Oregon: Journal of Geophysical Research, v. 91, and homogenization of the upper zone magmas the middle to upper parts of the crust and that p. 6059–6071, doi:10.1029/JB091iB06p06059. (Figs. 9C–9E). There are numerous examples of evolved volcanic rocks need not acquire all their Barnes, C.G., Allen, C.M., and Brigham, R.H., 1987, Isoto- dacitic eruptions triggered by the arrival of mafi c chemical characteristics in the lower crust. pic heterogeneity in a tilted plutonic system, Klamath Mountains, California: Geology, v. 15, p. 523–527, doi: magma at the base of silicic systems (e.g., Bach- This study demonstrates that mineral trace 10.1130/0091-7613(1987)15<523:IHIATP>2.0.CO;2. mann et al., 2002; Pallister et al., 1992). If the element chemistry provides an important com- Barnes, C.G., Allen, C.M., Hoover, J.D., and Brigham, R.H., dacitic roof dikes are appropriate examples of plement to geochronology in order to better 1990, Magmatic components of a tilted plutonic system, Klamath Mountains, California, in Anderson, J.L., the state of the upper zone at the time of eruption, understand the history and pace of batholith ed., The nature and origin of Cordilleran Magmatism: then the percentage of crystals, ~40%–50%, is emplacement. In particular, the mineral composi- Geological Society of America Memoir 174, p. 331– 346, doi:10.1130/MEM174-p331. similar to crystal proportions in homogeneous tions provide a tool to map the extent of ancient Bartley, J.M., Coleman, D.S., and Glazner, A.F., 2008, Incre- dacitic ignimbrite deposits described throughout magma reservoirs, identify magma batches, and mental pluton emplacement by magmatic crack-seal: the southwestern United States (Hildreth, 1981; understand their chemical and physical evolution Royal Society of Edinburgh Transactions, Earth Sciences, v. 97, p. 383–396, doi:10.1017/S0263593300001528. Bachman et al., 2002; Christiansen, 2005). The and interactions. Bohrson, W.A., and Spera, F.J., 2007, Energy-constrained size of the WCb upper zone is not comparable to recharge, assimilation, and fractional crystallization ACKNOWLEDGMENTS these large ignimbrite deposits, which represent (EC-RAcFC): A Visual Basic computer code for cal- 3 culating trace element and isotope variations of open 1000–5000 km of erupted magma, but it could We thank Monika Leopold, Samantha Buck, and system magmatic systems: Geochemistry Geophysics easily have fed eruptions as large as or larger Brendan Hargrove for assistance in the fi eld and Geosystems, v. 8, no. 11, doi:10.1029/2007GC001781. than the 7000 ka culminating eruption of Crater George Morgan (University of Oklahoma) and Susan Boynton, W.V., 1984, Geochemistry of the rare-earth elements: Swapp (University of Wyoming) for their help with meteorite studies, in Henderson, P., ed., Rare earth ele- Lake (Bacon and Druitt, 1988) (Figs. 9E, 9F). microprobe analyses. We thank Calvin Miller, an ment geochemistry: Amsterdam, Elsevier, p. 63–114. As volcanism waned, the upper zone magma anonymous reviewer, and associate editor Mike Wil- Buck, S., Barnes, C.G., Yoshinobu, A.S., Barnes, M.A., Coint, liams for their helpful comments and advice. This N., Hargrove, B., and Leopold, M., 2010, Magma min- differentiated by fractional crystallization, gling and mixing: Mafi c enclaves and dikes in the Wooley developing a felsic cap that could have erupted work was supported by National Science Foundation Creek batholith, N. California: Geological Society of grant EAR-0838342 to Yoshinobu and C. Barnes, a as rhyolitic domes, capable of producing minor America Abstracts with Programs, v. 42, no. 5, p. 102. 2009 Geological Society of America Penrose grant Burgisser, A., and Bergantz, G.W., 2011, A rapid mecha- pyroclastic fl ows. to Coint, and the Texas Tech University Department nism to remobilize and homogenize highly crys- of Geosciences. talline magma bodies: Nature, v. 471, p. 212–215, CONCLUSIONS doi:10.1038/nature09799. REFERENCES CITED Castro, A., and Stephens, E.W., 1992, Amphibole-rich polycrystalline clots in calc-alkaline granitic rocks and their On the basis of analogies with modern arc Allen, C.M., and Barnes, C.G., 2006, Ages and some cryp- enclaves: Lithos, v. 30, p. 1093–1112. volcanoes, such as in the Cascade Range, the tic sources of Mesozoic plutonic rocks in the Klamath Cherniak, D.J., and Liang, Y., 2012, Ti diffusion in natural Mountains, California and Oregon, in Snoke, A.W., pyroxene: Geochimica et Cosmochimica Acta, v. 98, WCb corresponds to the type of magmatic res- and Barnes, C.G., eds., Geological studies in the p. 31–47, doi:10.1016/j.gca.2012.09.021. ervoir expected to be beneath typical arc-related Klamath Mountains Province, California and Oregon: Christiansen, E.H., 2005, Contrasting processes in silicic A volume in honor of William P. Irwin: Geological magma chambers: Evidence from very large volume composite volcanoes. The fi rst magmatism to Society of America Special Paper 410, p. 223–245, ignimbrites: Geological Magazine, v. 142, p. 669–681, occur was mafi c to intermediate and represented doi:10.1130/2006.2410(11). doi:10.1017/S0016756805001445.

1764 Geosphere, December 2013

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021 Pluton assembly using trace elements in augite and hornblende

Christiansen, E.H., and McCurry, M., 2008, Contrasting ori- Hildreth, W., 1981, Gradients in silicic magma chambers: Otten, M., 1984, The origin of brown hornblende in the gins of Cenozoic silicic volcanic rocks from the western Implications for lithospheric magmatism: Journal of Artfjallet gabbro and dolerites: Contributions to Min- Cordillera of the United States: Bulletin of Volcanology, Geophysical Research, v. 86, p. 10153–10192, doi: eralogy and Petrology, v. 86, p. 189–199, doi:10.1007 v. 70, p. 251–267, doi:10.1007/s00445-007-0138-1. 10.1029/JB086iB11p10153. /BF00381846. Claiborne, L.L., Miller, C.F., Flanagan, D.M., Clynne, M.A., Hildreth, W.S., 2004, Volcanological perspectives on Long Pallister, J.S., Hoblitt, R.P., and Reyes, A.G., 1992, A basalt and Wooden, J.L., 2010, Trace element composition of Valley, Mammoth Mountain, and Mono Craters: Sev- trigger for the 1991 eruptions of Pinatubo volcano?: igneous zircon: A thermal and compositional record of eral contiguous but discrete systems: Journal of Volca- Nature, v. 356, p. 426–428, doi:10.1038/356426a0. the accumulation and evolution of a large silicic batho- nology and Geothermal Research, v. 136, p. 169–198, Paterson, S.R., Okaya, D., Memeti, V., Economos, R., and lith, Spirit Mountain, Nevada: Contributions to Miner- doi:10.1016/j.jvolgeores.2004.05.019. Miller, R., 2011, Magma addition and fl ux calcula- alogy and Petrology, v. 160, p. 511–531, doi:10.1007 Huber, C., Bachmann, O., and Manga, M., 2009, Homogeni- tions of incrementally constructed magma chambers in /s00410-010-0491-5. zation processes in silicic magma chambers by stirring continental margin arcs: Combined fi eld, geochrono- Coint, N., 2012, Assembly and evolution of the Wooley and mushifi cation (latent heat buffering): Earth and logic, and thermal modeling studies: Geosphere, v. 7, Creek batholith: Evidence from mineral compositions Planetary Science Letters, v. 283, p. 38–47, doi:10.1016/j p. 1439–1468, doi:10.1130/GES00696.1. and U-Pb geochronology [Ph.D. thesis]: Lubbock, .epsl.2009.03.029. Piwinskii, A.J., 1968, Experimental studies of igneous rocks Texas Tech University, 479 p. Jachens, R.C., Barnes, C.G., and Donato, M.M., 1986, Sub- series central Sierra Nevada batholith, California: Jour- Coint, N., Barnes, C.G., Yoshinobu, A.S., Chamberlain, surface confi guration of the Orleans fault: Implications nal of Geology, v. 76, p. 548–570, doi:10.1086/627359. K.R., and Barnes, M.A., 2013, Batch-wise assembly for deformation in the western Klamath Mountains, Pouchou, J.L., and Pichoir, F., 1985, “PAP: φ(ρz) correction and zoning of a tilted calc-alkaline batholith: Field California: Geological Society of America Bulletin, procedure for improved quantitative microanalysis, in relations, timing, and compositional variation: Geo- v. 97, p. 388–395, doi:10.1130/0016-7606(1986)97<388: Amstrong, J.T., ed., Microbeam analysis: San Fran- sphere, doi:10.1130/GES00930.1. SCOTOF>2.0.CO;2. cisco, California, San Francisco Press, p. 104–106. Coleman, D.S., Gray, W., and Glazner, A.F., 2004, Rethink- Leake, B.E., and 21 others, 1997, Nomenclature of amphi- Ritchey, J.L., 1980, Divergent magmas at Crater Lake, Ore- ing the emplacement and evolution of zoned plutons: boles: Report of the Subcommittee on Amphiboles of gon: Products of fractional crystallization and vertical Geochronologic evidence for incremental assembly the International Mineralogical Association: Canadian zoning in a shallow, water undersaturated chamber: of the Tuolumne Intrusive Suite, California: Geology, Mineralogist, v. 35, p. 219–246. Journal of Volcanology and Geothermal Research, v. 7, v. 32, p. 433–436, doi:10.1130/G20220.1. Leopold, M., and Yoshinobu, A.S., 2010, Diachronous p. 373–386, doi:10.1016/0377-0273(80)90039-6. Deering, C.D., and Bachmann, O., 2010, Trace element indi- hypersolidus deformation during repeated magma Robinson, D.M., and Miller, C.F., 1999, Record of magma cators of crystal accumulation in silicic igneous rocks: injection: Field observation from the tilted Wooley chamber processes preserved in accessory mineral Earth and Planetary Science Letters, v. 297, p. 324– Creek batholith (WCb): Geological Society of America assemblages, Aztec Wash pluton, Nevada: American 331, doi:10.1016/j.epsl.2010.06.034. Abstracts with Programs, v. 42, no. 5, p. 102. Mineralogist, v. 84, p. 1346–1353. De Paolo, D.J., 1981, Trace element and isotopic effects of Li, Y., 2008, Petrologic and geochronological investigations: Ruprecht, P., Bergantz, G.W., Cooper, K.M., and Hildreth, combined wallrock assimilation and fractional crystal- Age and petrology of Grenville-age rocks in west W., 2012, The crustal magma storage system of vol- lization: Earth and Planetary Science Letters, v. 53, Texas and New Mexico and mechanism of carbonate cano Quizapu, Chile, and the effects of magma mix- p. 189–202, doi:10.1016/0012-821X(81)90153-9. assimilation at Hortavær Igneous Complex, Norway ing on magma diversity: Journal of Petrology, v. 53, Du Bray, E.A., Bacon, C., John, D.A., Wooden, J.L., and [Ph.D. thesis]: Lubbock, Texas Tech University, 275 p. p. 801–840, doi:10.1093/petrology/egs002. Mazdab, F.K., 2011, Episodic intrusion, internal differ- Lipman, P., Dungan, M., and Bachmann, O., 1997, Comag- Snoke, A.W., and Barnes, C.G., 2006, The development of entiation, and hydrothermal alteration of the Miocene matic granophyric granite in the Fish Canyon Tuff, tectonic concepts for the Klamath Mountains prov- Tatoosh intrusive suite south of Mount Rainier, Wash- Colorado: Implications for magma-chamber processes ince, California and Oregon, in Snoke, A.W., and ington: Geological Society of America Bulletin, v. 123, during a large ash-fl ow eruption: Geology, v. 25, p. 915– Barnes, C.G., eds., Geological studies in the Klam- p. 534–561, doi:10.1130/B30095.1. 918, doi:10.1130/0091-7613(1997)025<0915:CGGITF ath Mountains province, California and Oregon: Eggler, D.H., and Burnham, W.C., 1973, Crystallization and >2.3.CO;2. A volume in honor of William P. Irwin: Geologi-

fractionation trends in the system andesite-H2O-CO2- Matzel, J.E.P., Bowring, S.A., and Miller, R.B., 2006, Time cal Society of America Special Paper 410, p. 1–29, O2 at pressures to 10 kb: Geological Society of Amer- scale of pluton construction at differing crustal levels: doi:10.1130/2006.2410(01). ica Bulletin, v. 84, p. 2517–2532, doi:10.1130/0016 Examples from the Mont Stuart and Tenpeak intrusions, Stephens, W.E., 2001, Polycrystalline amphibole aggre- -7606(1973)84<2517:CAFTIT>2.0.CO;2. North Cascades, Washington: Geological Society of gates (clots) in granites as potential I-type restite: An Ernst, G.W., and Liu, J., 1998, Experimental phase-equilib- America Bulletin, v. 118, p. 1412–1430, doi:10.1130 ion microprobe study of rare-earth distributions: Aus- rium study of Al- and Ti-contents of calcic amphibole /B25923.1. tralian Journal of Earth Sciences, v. 48, p. 591–601, in MORB—A semiquantitative thermobarometer: McLeod, G.W., Dempster, T.J., and Faithfull, J.W., 2010, Deci- doi:10.1046/j.1440-0952.2001.00880.x. American Mineralogist, v. 83, p. 952–969. phering magma-mixing processes using zoned titanite Sun, S.S., and McDonough, W.F., 1989, Chemical and iso- Femenias, O., Mercier, J.C., Nkono, C., Diot, H., Berza, T., from the Ross of Mull granite, Scotland: Journal of topic systematics of oceanic basalts: Implications for Tatu, M., and Demaiffe, D., 2006, Calcic amphibole Petrology, v. 28, p. 55–82, doi:10.1093/petrology/egq071. mantle composition and processes: Geological Society growth and compositions in calc-alkaline magmas: Memeti, V., Paterson, S., Matzel, J., Mundil, J., and Okaya, of London, Special Publications, v.42, p. 313–345. Evidence from the Motru Dike Swarm (Southern Car- D., 2010, Magmatic lobes as “snapshots” of magma Tappa, M.J., Coleman, D.S., Mills, R.D., and Samperton, pathians, Romania): American Mineralogist, v. 91, chamber growth and evolution in large, composite batho- K.M., 2011, The plutonic record of a silicic ignim- p. 73–81, doi:10.2138/am.2006.1869. liths: An example from the Tuolumne intrusion, Sierra brite from the Latir volcanic ﬁ eld, New Mexico: Geo- Glazner, A.F., Bartley, J.M., Coleman, D.S., Gray, W., and Nevada, California: Geological Society of America Bul- chemistry Geophysics Geosystems, v. 12, no. 10, doi: Taylor, G.K., 2004, Are plutons assembled over millions letin, v. 122, p. 1912–1931, doi:10.1130/B30004.1. 10.1029/2011GC003700. of years by amalgamation from small magma cham- Miller, C.F., Furbish, D.J., Walker, B.A., Claiborne, L.L., Van Orman, J.A., Grove, T.L., and Shimizu, N., 2001, Rare bers?: GSA Today, v. 14, p. 4–11, doi:10.1130/1052 Koteas, G.C., Bleick, H.A., and Miller, J.S., 2011, earth element diffusion in diopside: Inﬂ uence of tem- -5173(2004)014<0004:APAOMO>2.0.CO;2. Growth of plutons by incremental emplacement of sheets perature, pressure, and ionic radius, and an elastic Grunder, A.L., Klemetti, E.W., Feeley, T.C., and McKee, C.L., in crystal-rich host: Evidence from Miocene intrusions of model for diffusion in silicates: Contributions to Min- 2008, Eleven million years of arc volcanism at the Aucan- the Colorado River region, Nevada, USA: Tectonophys- eralogy and Petrology, v. 141, p. 687–703, doi:10.1007 quilcha Volcanic Cluster, northern Chilean Andes: Impli- ics, v. 500, p. 65–77, doi:10.1016/j.tecto.2009.07.011. /s004100100269. cations for the life span and emplacement of plutons: Mills, R.D., and Coleman, D.S., 2010, Modeling large- Walker, B.A., Jr., Miller, C.F., Claiborne, L.L., Wooden, Royal Society of Edinburgh Transactions, Earth Sciences, volume felsic eruptions from trace-element geochem- J.L., and Miller, J.S., 2007, Geology and geochronol- v. 97, p. 415–436, doi:10.1017/S0263593300001541. istry: Geological Society of America Abstracts with ogy of the Spirit Mountain batholith, southern Nevada: Hacker, B.R., Donato, M.M., Barnes, C.G., McWilliams, Programs, v. 42, no. 4, p. 52. Implications for time scales and physical processes of M.O., and Ernst, W.G., 1995, Time scales of orogeny: Morimoto, N., Fabries, J., Ferguson, A.K., Ginzburg, I.V., batholith construction: Journal of Volcanology and Geo- Jurassic construction of the Klamath Mountains: Tec- Ross, M., Seifert, F.A., Zussman, J., Aoki, K., and Got- thermal Research, v. 167, p. 239–262, doi:10.1016/j tonics, v. 14, p. 677–703, doi:10.1029/94TC02454. tardi, G., 1988, Nomenclature of pyroxenes: American .jvolgeores.2006.12.008. Harper, G.D., Saleeby, J.B., and Heizler, M., 1994, Forma- Mineralogist, v. 73, p. 1123–1133. Wright, J.E., and Wyld, S.J., 1994, The Rattlesnake Creek tion of the Josephine ophiolite and the Nevadan orog- Ohba, T., Kimura, Y., and Fujimaki, H., 2007, High-magnesian terrane, Klamath Mountains, California: An early eny in the Klamath Mountains, California, Oregon: andesite produced by two-stage magma mixing: A case Mesozoic volcanic arc and its basement of tectonically U/Pb zircon and 40Ar/39Ar geochronology: Jour- study from Hachimantai, northern Honshu, Japan: Jour- disrupted oceanic crust: Geological Society of Amer- nal of Geophysical Research, v. 99, p. 4293–4321, nal of Petrology, v. 48, p. 627–645, doi:10.1093/petrology ica Bulletin, v. 106, p. 1033–1056, doi:10.1130/0016 doi:10.1029/93JB02061. /egl075. -7606(1994)106<1033:TRCTKM>2.3.CO;2.

Geosphere, December 2013 1765

Downloaded from http://pubs.geoscienceworld.org/gsa/geosphere/article-pdf/9/6/1747/3346579/1747.pdf by guest on 25 September 2021