| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Oxygen Isotope Geochemistry of the Lassen Volcanic Center, California: Resolving Crustal and Mantle Contributions to Continental Arc Magmatism
1Department of Earth Sciences, Montana State University, Bozeman, MT 59717, USA
2United States Geological Survey, 345 Middlefield Road, Mail Stop 910, Menlo Park, CA 94025, USA
RECEIVED SEPTEMBER 14, 2006; ACCEPTED FEBRUARY 18, 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL AND PETROLOGICAL...SAMPLE COLLECTION AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This study reports oxygen isotope ratios determined by laser fluorination of mineral separates (mainly plagioclase) from basaltic andesitic to rhyolitic composition volcanic rocks erupted from the Lassen Volcanic Center (LVC), northern California. Plagioclase separates from nearly all rocks have
18O values (6·1–8·4
) higher than expected for production of the magmas by partial melting of little evolved basaltic lavas erupted in the arc front and back-arc regions of the southernmost Cascades during the late Cenozoic. Most LVC magmas must therefore contain high 18O crustal material. In this regard, the 18O values of the volcanic rocks show strong spatial patterns, particularly for young rhyodacitic rocks that best represent unmodified partial melts of the continental crust. Rhyodacitic magmas erupted from vents located within 3·5 km of the inferred center of the LVC have consistently lower 18O values (average 6·3 ± 0·1) at given SiO2 contents relative to rocks erupted from distal vents (>7·0 km; average 7·1 ± 0.1). Further, magmas erupted from vents situated at transitional distances have intermediate values and span a larger range (average 6·8 ± 0·2). Basaltic andesitic to andesitic composition rocks show similar spatial variations, although as a group the 18O values of these rocks are more variable and extend to higher values than the rhyodacitic rocks. These features are interpreted to reflect assimilation of heterogeneous lower continental crust by mafic magmas, followed by mixing or mingling with silicic magmas formed by partial melting of initially high 18O continental crust (
9·0) increasingly hybridized by lower 18O (6·0) mantle-derived basaltic magmas toward the center of the system. Mixing calculations using estimated endmember source 18O values imply that LVC magmas contain on a molar oxygen basis approximately 42 to 4% isotopically heavy continental crust, with proportions declining in a broadly regular fashion toward the center of the LVC. Conversely, the 18O values of the rhyodacitic rocks suggest that the continental crust in the melt generation zones beneath the LVC has been substantially modified by intrusion of mantle-derived basaltic magmas, with the degree of hybridization ranging on a molar oxygen basis from approximately 60% at distances up to 12 km from the center of the system to 97% directly beneath the focus region. These results demonstrate on a relatively small scale the strong influence that intrusion of mantle-derived mafic magmas can have on modifying the composition of pre-existing continental crust in regions of melt production. Given this result, similar, but larger-scale, regional trends in magma compositions may reflect an analogous but more extensive process wherein the continental crust becomes progressively hybridized beneath frontal arc localities as a result of protracted intrusion of subduction-related basaltic magmas. KEY WORDS: oxygen isotopes; phenocrysts; continental arc magmatism; Cascades; Lassen
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL AND PETROLOGICAL...SAMPLE COLLECTION AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
In the past several decades considerable effort has been devoted to better characterizing the physical and chemical processes affecting magmas in open-system crustal magma chambers (e.g. Bergantz, 1995
; Wilson, 1995; Grove, 2000; Spera & Bohrson, 2001; Dufek & Bergantz, 2005; Annen et al., 2006). A large part of this effort has been motivated by the goal of constraining the sources of igneous rocks, as this has important implications for problems such as the mass balance between mantle and crustal contributions to magmas. A key question in this regard is by what means can mantle and crustal contributions to continental arc magmas be resolved? Assessment of magma sources using commonly employed radiogenic isotopic systems (e.g. Sr and Nd) is often difficult in arc settings owing to small compositional contrasts between the magmas and the young arc-related rocks they intrude (e.g. Gill, 1981; Bullen & Clynne, 1990; Davidson et al., 1991; Feeley, 1993; Borg & Clynne, 1998; Hart et al., 2002; Lackey et al., 2005). In contrast, careful examination of oxygen isotope data for rocks at some volcanic centers has allowed identification of shallow and deep crustal differentiation processes and imposed constraints on magma sources and their evolution with development of the centers (e.g. Grunder, 1987; Bacon et al., 1989, 1994; Grunder & Wickham, 1991; Feeley & Sharp, 1995; Donnelly-Nolan, 1998; Bindeman et al., 2001; Wolff et al., 2002; Boroughs et al., 2005).
The advantage of oxygen isotope data to constrain magma source rocks arises from: (1) the homogeneity of most primary or near primary mantle-derived basalts (5·7 ± 0·3; e.g. Eiler et al., 2000; Valley et al., 1998; Eiler, 2001); (2) the small isotopic fractionations between silicate minerals and melt at high magmatic temperatures; (3) relative to mantle sources, enrichment or depletion in 18O in crustal rocks as a result of geological processes that are independent of age; (4) simplifications in mass-balance relationships among endmember sources as a result of small differences in the abundance of oxygen among diverse rock types relative to the large and often poorly constrained variations in other elemental abundances. For these reasons, oxygen isotope values for fresh volcanic rocks can be particularly useful for quantifying shallow-level crustal contamination processes (as summarized by Taylor, 1968; Muehlenbachs et al., 1974; Taylor & Silver, 1978; Macdonald et al., 1987) and are also useful for detecting cryptic processes such as deep mafic underplating, crustal hybridization, and contemporaneous hydrothermal alteration of the upper continental crust (Grunder & Wickham, 1991; Feeley & Sharp, 1995; Wolff et al., 2000; Bindeman et al., 2001, 2004).
Despite the potential to elucidate magma source rocks, oxygen isotopes have not been widely used in detailed studies of composite continental arc volcanoes. In large part this probably reflects the ease by which primary magmatic 18O values may be modified by secondary processes such as hydrothermal alteration or low-temperature hydration of glassy groundmass. This susceptibility of oxygen isotope values to post-eruptive modification can preclude the use of whole-rock values in evaluating crustal source rocks. As a result, many previous studies of arc-related rocks are either descriptions of regional isotopic trends or evaluations of limited datasets from single centers (e.g. Matsuhisa et al., 1973; Blattner & Reid, 1982; Davidson & Harmon, 1989; Singer et al., 1992; Barragan et al., 1998; Pineau et al., 1999). In contrast to whole-rock values, it is now well established that oxygen isotope ratios derived from laser fluorination of separated phenocrysts can avoid problems arising from the susceptibility of glassy volcanic rocks to post-eruptive alteration (Baker et al., 2000; Eiler, 2001; Bindeman et al., 2004).
In this paper we build on existing age, petrological, and compositional studies of volcanic rocks erupted from the Lassen Volcanic Center (LVC; Fig. 1), California, by using new oxygen isotope determinations by laser fluorination of mineral separates (mainly plagioclase) to evaluate the proportions of mantle and crustal sources in a continental arc composite volcano as a function of space and time. The LVC is ideally suited for an oxygen isotope investigation because the center has been the subject of detailed studies for well over half a century. As a result, a wealth of data exists on the field relations, compositions, and petrogenesis of the rocks (e.g. Williams, 1932; Eichelberger, 1978; Heiken & Eichelberger, 1980; Bullen & Clynne, 1990; Clynne, 1990, 1999; Tepley et al., 1999; Christiansen et al., 2002; Clynne & Muffler, 2008). In addition, rocks associated with the LVC erupted across an area extending 15 km to the north and NE of the central focus of the magmatic system, and rocks with comparatively high 18O values (relative to mantle values) are likely to be present in the lower to middle continental crust beneath the center (Guffanti et al., 1996; Bacon et al., 1997). Collectively, these features allow the LVC to serve as a case study for how oxygen isotope values can be used to assess mantle and crustal contributions to a composite continental arc volcanic center across a wide region, including areas distant from the voluminous central focus of the volcanic system.
|
|
|
|
GEOLOGICAL AND PETROLOGICAL BACKGROUND |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL AND PETROLOGICAL...SAMPLE COLLECTION AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Geological background
The LVC is a large Pleistocene to Holocene stratovolcano and dome field in northern California. It is the southernmost active volcanic center of the Cascades volcanic arc, where it is associated with easterly subduction of the Gorda Plate segment of the Juan de Fuca plate system (Fig. 1). The volcanic center has a composite volume of approximately 215 km3, over a third of which are volcanic rocks with >60% SiO2 (Sherrod & Smith, 1990
; Guffanti et al., 1996). It is therefore one of the most voluminous Quaternary magma systems in the Cascades arc. Geological and seismic-refraction studies indicate that the continental crust beneath the center is 38 ± 4 km thick and probably consists of 1–4 km of late Cenozoic volcanic rocks underlain by Mesozoic and perhaps Paleozoic Klamath and Sierra Nevada Mountains type ultramafic to granitoid rocks intruded in the late Cenozoic (< 7 Ma) by basaltic arc magmas (Blakely et al., 1985; Mooney & Weaver, 1989; Blakely & Jachens, 1990; Guffanti et al., 1990; Stanley et al., 1990; Benz et al., 1992).
Abundant volcanic activity in the Lassen area of the southernmost Cascade Range began at 7 Ma and generally occurs on two scales (Clynne, 1990; Clynne & Muffler, 2008; Fig. 2). Regional volcanic activity is predominantly mafic in composition (<60% SiO2). These magmas erupted from hundreds of coalescing small- to medium-sized volcanoes with volumes of up to a few cubic kilometers and relatively short lifetimes of one to a few thousand years. Local volcanism is characterized by at least five long-lived (up to a million years) and much larger (hundreds of cubic kilometers) composite volcanic centers that erupted magmas ranging from basaltic andesitic to rhyolitic in composition (Guffanti et al., 1990; Bacon et al., 1997; Borg & Clynne, 1998).
The LVC is the youngest of the long-lived centers and forms the basis for interpreting the evolution of the older, more deeply eroded composite volcanic centers. The eruptive history of the LVC is divided into three stages designated as the Rockland, Brokeoff and Lassen eruptive stages (Fig. 2; Clynne & Muffler, 2008). The Rockland stage is composed of the 50 km3 Rockland tuff and several correlative dacitic and rhyolitic domes, lava flows, and pyroclastic units with a cumulative volume of 75 km3. It began at about 825 ka and lasted until about 610 ka, culminating with eruption of the Rockland tuff (Lanphere et al., 2004; Clynne & Muffler, 2008). Eruption of the Rockland tuff probably produced a small caldera, 5–8 km in diameter, although this has not been recognized in the Lassen area. As such, the caldera and some Rockland units have been completely buried by younger volcanism of the LVC, especially the deposits of the Brokeoff volcano (see below). Because events of later volcanic stages have obliterated many features of the Rockland stage, this study focuses exclusively on rocks associated with these younger stages.
The second stage is the Brokeoff stage, in which an 100 km3 stratovolcano (Brokeoff volcano) was active between 600 and 385 ka. The stratigraphy of Brokeoff volcano is divided into two sequences: the Mill Canyon sequence and the Diller sequence. During the Mill Canyon sequence, numerous thin flows of olivine–augite basaltic andesitic and two-pyroxene ± olivine mafic andesitic lava flows erupted from a central vent region between 600 and 470 ka. The end of this stage is marked by eruption of a group of several distinctive hornblende dacitic lava flows (dacite of Twin Meadows). The Diller sequence represents continued growth of Brokeoff volcano by the eruption from 470 to 385 ka of thick lava flows of silicic two-pyroxene andesitic to mafic dacitic magmas from vents high on the flanks of the edifice constructed during the Mill Canyon stage. The core of Brokeoff volcano is now deeply eroded and hosts a 10 km2 zone of intense hydrothermal alteration. Whole-rock 18O values of rocks in the alteration zone vary from 9·8 to 0·6, although the vast majority of altered samples have 18O < 6·5 (Rose et al., 1994). In detail, low 18O (e.g. < 5·0) hydrothermally altered volcanic rocks are widely exposed in two concentric areas within the eroded core of Brokeoff volcano, recording high-temperature (250°C) isotopic exchange with circulating waters during the Brokeoff volcanic stage (Rose et al., 1994). Solfatarically altered rocks with high 18O values (e.g. >8·0) are sparsely present in the core of Brokeoff volcano, although these occur only in localized, discontinuous areas typically associated with the modern hydrothermal system (John et al., 2006).
The most recent stage of magmatism (the Lassen stage) represents a fundamental change in the character of volcanism. During the Lassen stage the locus of volcanic activity shifted to the north and NE flanks of Brokeoff volcano, where a field of silicic lava flows and domes was constructed during two episodes that are distinguished on the basis of age (Fig. 2). These are: (1) the Bumpass sequence, consisting of hornblende–biotite and hornblende–two-pyroxene rhyolitic and dacitic domes, flows, and pyroclastic rocks that erupted following extinction of Brokeoff volcano between 300 and 190 ka; (2) the Eagle Peak sequence, consisting of hornblende–biotite dacitic and rhyolitic domes, lava flows, and pyroclastic flows erupted during the past 65 ka. Notably, vents for the Eagle Peak sequence form a north-striking linear group that extends as far as 12 km north of Diamond Peak, the inferred center of Brokeoff volcano (Fig. 2; Clynne, 1990; Rose et al., 1994; Clynne & Muffler, 2008). As discussed by Guffanti et al. (1990), recent north–south to NW–SE alignment of volcanic vents in the southernmost Cascades to some extent reflects westward encroachment of the extensional Basin and Range Province on the subduction-related arc. Notable Holocene eruptions of the Eagle Peak sequence include the 1100-year-old Chaos Crags domes and associated pyroclastic rocks (Heiken & Eichelberger, 1980; Clynne et al., 2002).
Collectively, the silicic lava flow- and dome-field represents eruption of 30–50 km3 of magma. Petrographically, the rocks are distinct in that nearly all contain undercooled inclusions (i.e. Bacon, 1986) of medium-K basaltic andesitic and mafic andesitic magma compositionally similar to lava flows erupted from Brokeoff volcano (Williams, 1931; Heiken & Eichelberger, 1980; Clynne, 1990).
Also included in the Lassen stage are rocks of the Twin Lakes sequence, which forms a petrological and magmatic transition between the LVC and the regional mafic volcanism. Magmas of the Twin Lakes sequence mainly erupted in the northeastern region of the LVC, where they form the Central Plateau of Lassen Volcanic National Park (Fig. 2). Here, 10 km3 of basaltic andesitic and mafic andesitic magmas that contain distinctive hybrid phenocryst assemblages (ol + cpx + qtz) and partially melted crustal xenoliths erupted from 310 ka to the present. The youngest Twin Lakes sequence rocks on the Central Plateau are the AD 1650 eruptive products of Cinder Cone (Fig. 2; Clynne et al., 2000). A few rocks assigned to the Twin Lakes sequence also erupted in the vicinity of the silicic dome field on the north flank of Brokeoff volcano. These include the inclusion-bearing May 1915 andesitic to dacitic eruptions of Lassen Peak (Clynne, 1999).
Petrological background
Fundamental to interpretation of the oxygen isotope compositions of the volcanic rocks is our contention that the most silicic magmas erupted in the LVC are largely bulk crustal melts, as opposed to liquids derived from less evolved parental magmas by crystal–liquid fractionation processes. We adopt this position because detailed studies conducted over several decades utilizing diverse petrological and geochemical datasets have found fractional crystallization models from basaltic andesite to rhyolite with or without assimilation of continental crust to be untenable (e.g. Eichelberger, 1978; Bullen & Clynne, 1990; Borg et al., 1997; Hart et al., 2002). In contrast, chemical trends and textural disequilibrium features in LVC rocks clearly attest to the dominant role of mixing and mingling between relatively well-homogenized silicic crustal melts and heterogeneous basaltic andesitic to andesitic magmas to produce petrographically diverse hybrid magmas (Eichelberger, 1978; Heiken & Eichelberger, 1980; Bullen & Clynne, 1990; Clynne, 1990, 1999; Guffanti et al., 1996; Borg & Clynne, 1998; Tepley et al., 1999; Hart et al., 2002). In fact, it is this profusion of diverse investigations that all arrive at the same conclusion that makes the LVC an attractive target for a detailed oxygen isotope study of source rocks at a composite continental arc volcano. Nearly identical models invoking lower crustal partial melting and mixing have recently been invoked to explain the origin of similar composition, subduction-related silicic magmas in the NE Japan arc and the Mesozoic Sierra Nevada batholith of California (e.g. Sisson et al., 1996; Ratajeski et al., 2001, 2005; Yamamoto, 2007).
That the LVC suite does not represent a continuous liquid line of descent from basaltic andesitic to rhyolitic compositions is illustrated on chemical variation diagrams, where trends are strikingly linear and a small discontinuity (i.e. a ‘chemical gap’; Eichelberger et al., 2006) exists at intermediate compositions (62 wt % SiO2; Fig. 3). Importantly, on mixing lines drawn from the most mafic to the most silicic rocks erupted during the Brokeoff and Lassen eruptive stages, this discontinuity falls precisely on points where magmas should theoretically contain equal proportions (i.e. 50%) of both endmembers (Fig. 3). As discussed by numerous workers (e.g. Eichelberger, 1975; Sparks & Marshall, 1986; Feeley & Grunder, 1991; Feeley et al., 1998; Wilcox, 1999; Eichelberger et al., 2006; Shukuno et al., 2006), these relationships can be interpreted to indicate that the discontinuity represents a chemical and thermal boundary separating hybrid magmas formed by (1) liquid–liquid mixing where the proportion of the hot mafic endmember exceeds the proportion of the cooler silicic endmember and those formed by (2) liquid–solid mingling involving disaggregation of largely solidified basaltic andesitic to andesitic inclusions (which themselves are hybrids) in silicic magmas just prior to or during eruption from shallow reservoirs. This interpretation readily explains the diversity of textural and chemical disequilibrium features of LVC rocks summarized below.
|
Previous studies documented complex petrogenetic histories for LVC rocks principally involving magma mixing and mingling processes (e.g. Heiken & Eichelberger, 1980
; Bullen & Clynne, 1990; Clynne, 1990, 1999; Tepley et al., 1999; Wilson et al., 2005). Derivation of Brokeoff volcano rocks involved mixing between relatively well-homogenized silicic crustal melts and heterogeneous, contaminated mafic magmas to produce petrographically diverse basaltic andesitic to andesitic hybrid magmas (Bullen & Clynne, 1990; Clynne, 1990). The rocks contain variable phenocryst contents (10–30 vol. %) dominated by normally to oscillatory zoned plagioclase (An50–60) often containing sieved mantles and thin clear rims (Clynne, 1990). Twin Lakes sequence rocks on the Central Plateau are distinguished by disequilibrium phenocryst assemblages of ubiquitous olivine phenocrysts and clinopyroxene armored quartz xenocrysts (‘quartz basalts’; Finch & Anderson, 1930; Carlson & Wheeler, 1980), and in many rocks partially melted, coarse-grained quartz- and plagioclase-bearing crustal xenoliths (up to 10 cm across) devoid of pristine hydrous mafic phases. The xenoliths at Cinder Cone were studied by Borg & Clynne (1998), who interpreted them as residues of partially melted granitic country rocks. These features almost certainly reflect incorporation, partial melting, assimilation and disaggregation of felsic basement lithologies, perhaps Mesozoic intrusive rocks, in relatively mafic magmas originally similar to those erupted throughout the southernmost Cascades (Borg & Clynne, 1998).
Production of compositional diversity in Lassen stage inclusion-bearing domes and lavas reflects the combined effects of a more complex set of processes. These include: (1) mixing during injection of compositionally diverse mafic magmas into shallow chambers containing dacitic to rhyolitic crustal melts; (2) formation of hybrid magma layers beneath the silicic magmas; (3) buoyant ascent of hybrid magma blobs to form undercooled basaltic andesitic to mafic andesitic inclusions within the silicic magmas; (4) mechanical disaggregation of some of the inclusions within the silicic magmas to produce a broad range (andesitic to rhyolitic) of host lava compositions (Bullen & Clynne, 1990; Thomas & Tait, 1997; Clynne, 1999; Tepley et al., 1999; Wilson et al., 2005).
The petrogenetic processes associated with production of the inclusion-bearing rocks are manifest by the presence in host lavas and inclusions of phenocrysts with distinctive textures and compositions [see photomicrographs in the papers by Clynne (1999) and Tepley et al. (1999)]. Most notably, inclusions contain embayed quartz crystals and plagioclase xenocrysts with finely sieved cores (‘reacted’ phenocrysts of Clynne, 1999; An25–45; Clynne, 1990; Tepley et al., 1999). These crystal types were undoubtedly derived from mixing during injection of mafic magmas into shallow silicic chambers, as host lavas and domes contain identical composition, albeit texturally equilibrated phenocrysts (‘unreacted’ phenocrysts of Clynne, 1999). Quartz and plagioclase xenocrysts in the inclusions are mantled by clinopyroxene-bearing reaction corona and clear euhedral rims, respectively, which precipitated following entrainment of host crystals by the inclusion-forming magmas. In the case of the clear plagioclase rims, these are similar in composition (An70–50) to acicular groundmass plagioclase microlites that formed during undercooling of inclusion-forming magmas (Clynne, 1999). The inclusions are therefore hybrids formed by magma–magma mixing (see Bacon, 1986; Feeley & Dungan, 1996; Clynne, 1999; Wilson et al., 2005). Host rocks of silicic composition contain texturally and chemically similar reacted plagioclase and quartz phenocrysts and groundmass microlites derived from mechanical disaggregation of the inclusions (Clynne, 1999). Importantly, they also contain a substantially larger population of unreacted plagioclase and quartz phenocrysts (Clynne, 1999; Wilson et al., 2005). The host rocks are therefore also hybrids, although mixing involved disaggregation of largely solidified inclusions in silicic magmas (i.e. mingling; see Bacon, 1986) just prior to or during eruption from shallow chambers (Clynne, 1999; Tepley et al., 1999; Wilson et al., 2005).
Previous studies demonstrate that the ranges in the Sr, Nd and Pb isotopic compositions of intermediate and silicic volcanic rocks in the LVC overlap those of regional mafic volcanic rocks, although they consistently fall near the more crust-like extremes of the ranges (i.e. high 87Sr/86Sr and low 143Nd/144Nd; Bullen & Clynne, 1990; Borg & Clynne, 1998). These features have been interpreted to indicate that the silicic magmas are well-homogenized partial melts (20–25%) of lower continental crust compositionally similar to regional calc-alkaline mafic lavas erupted during the late Cenozoic in the southernmost Cascades (Bullen & Clynne, 1990; Guffanti et al., 1996; Borg & Clynne, 1998). Furthermore, the lack of correlations between the major-element compositions and radiogenic-isotope ratios of the silicic magmatic rocks has also been interpreted to indicate that assimilation of Mesozoic plutonic rocks was not important (Bullen & Clynne, 1990; Borg & Clynne, 1998). Likewise, Hart et al. (2002) argued that the Os isotopic compositions of LVC intermediate and silicic composition rocks also reflect partial melting of mafic lower continental crust similar in composition to uncontaminated, but fractionated, late Cenozoic basaltic lavas. However, because the Os isotopic compositions of the evolved rocks are significantly more radiogenic (
Os = +23 to +224) than typical oceanic mantle values, Hart et al. (2002) suggested that the mafic component must be older than the majority of regional lavas exposed at present and remained isolated in the lower crust for a period of 5–10 Myr prior to melting to allow for in situ decay of 187Re to radiogenic 187Os.
|
SAMPLE COLLECTION AND ANALYTICAL TECHNIQUES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL AND PETROLOGICAL...SAMPLE COLLECTION AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
All plagioclase and quartz mineral separates analyzed in this study were separated from fresh lava flows or pumice clasts found in rapidly cooled, nonwelded pyroclastic flow deposits. Although most rocks analyzed in this study were previously investigated by Bullen & Clynne (1990
), Clynne (1990, 1999) and Borg & Clynne (1998), we re-collected new material at the same sites when necessary. For samples where sufficient material was available, we used the same rocks as analyzed in the previous studies. For Brokeoff stage lavas we collected fresh samples from three principal areas: Mill Creek on the south flank and Bailey and Blue Lake Canyons on the west flank of Brokeoff volcano (see Clynne, 1990; sample localities are shown in Fig. 2). These areas lie well outside the central region of extensive hydrothermal alteration. Oxygen isotope results for all samples are presented in Table 1, along with geographic locations, bulk-rock SiO2 contents, and previously determined whole-rock 87Sr/86Sr data (Bullen & Clynne, 1990; Borg et al., 1998). Many of the bulk-rock major-element analyses were determined prior to this study (Bullen & Clynne, 1990; M. A. Clynne, unpublished data). For several samples, however, including all undercooled inclusions, we obtained new major-element analyses by wavelength-dispersive X-ray fluorescence (XRF) spectrometry at the GeoAnalytical Laboratory at Washington State University by the technique described by Johnson et al. (1999), at the University of Massachusetts following the methods of Rhodes (1988), and at the USGS Analytical Laboratory in Lakewood, Colorado, by the method of Taggart et al. (1987). Whole-rock oxygen isotope values were measured at USGS laboratories by the conventional fluorination method reported by Bacon et al. (1989). Based upon standard and duplicate unknown analyses, the average uncertainty of the values is considered to be ±0·2.
Because of the complex petrogenetic histories of the LVC eruptive products, we separated distinct textural varieties of mineral phases. These include plagioclase and quartz phenocrysts from lava flows and pyroclastic rocks, disequilibrium-textured plagioclase and quartz xenocrysts from several undercooled inclusions, groundmass plagioclase crystals from inclusions, and plagioclase and quartz crystals from crustal xenoliths recovered from a few lava flows. We separated phenocrysts and xenocrysts in all samples by crushing rock chips in a disk mill, followed by mechanical sieving and collection of the 0·5–0·6 mm size fraction. For inclusion-bearing host rocks, we initially selected rock chips visibly free of inclusions. Crystals were concentrated from the 0·5–0·6 mm size fraction by standard magnetic techniques followed by hand-picking clear, inclusion-free fragments with a 30x binocular microscope. During hand-picking of phenocrysts from inclusion-bearing host lavas, we focused on excluding crystal fragments with visibly sieved regions. This was not possible for plagioclase xenocrysts in the inclusions, however, as most crystals display some degree of reaction (see discussion below). Groundmass plagioclase separates from several inclusions were prepared by hand selection of 2–4 cm sized rock chips visibly free of phenocrysts, sieving and collecting the 100–75 µm size fraction, concentrating plagioclase by standard magnetic techniques, and carefully hand-picking the concentrates to exclude aggregates and crystals with adhering glass. All separates were briefly washed ultrasonically with dilute HCl, rinsed thoroughly in deionized water, and dried in a drying oven.
For the mineral separates oxygen was extracted at the University of New Mexico stable-isotope laboratory on 1· 0–2·5 mg aliquots using the laser fluorination technique similar to that described by Sharp (1990) with a Merchantek CO2 laser and BrF5 as a reagent. The 18O values were measured on a Finnigan MAT Delta XL mass spectrometer in dual-inlet mode. Mineral pairs used in oxygen isotope thermometry were analyzed in sequence during the same day to avoid possible intra- and inter-run analytical drift. The results are presented in Table 1 where they are expressed in permil using standard delta notation relative to Vienna Standard Mean Ocean Water (VSMOW). Precision and accuracy were determined by duplicate analyses of more than 50 samples examined in this and a concurrent study of phenocryst phases from little evolved regional basaltic lavas (Underwood et al., 2004). The average reproducibility determined for all duplicate analyses in these studies is 0·05. In addition, we also determined oxygen isotope compositions for an average of five standard minerals (UWG-2 garnet 18O = 5·8; Valley et al., 1995, and in-house Gee Whiz quartz, 18O = 12·5 relative to NBS-28 = 9·6) during each analytical session. Because we report in Table 1 only analyses of unknown samples determined when the standard analyses ran within 0·1 of the accepted values, the average uncertainty is considered to be at most 0·1 and all data are reported as raw, uncalibrated values.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL AND PETROLOGICAL...SAMPLE COLLECTION AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Oxygen isotope analyses determined in this study are presented in Table 1. In this study we make the simplifying assumption that the
18O values of the plagioclase separates are representative of magmatic values and make no attempt to account for small 
18Oplag–melt fractionation as a function of melt composition and anorthite content in the rocks studied. For example, data for the andesitic magma (58 wt % SiO2) studied by Zhao & Zheng (2003) predict a 18Oplag–melt value of approximately –0·1 at 900°C for An60 plagioclase, which is the dominant composition of plagioclase phenocrysts in Brokeoff stage lava flows and groundmass microlites in Lassen stage undercooled inclusions. Similarly, interpolation between andesitic and rhyolitic melts (74·0 wt % SiO2) studied by Zhao & Zheng (2003) for An35 plagioclase in dacitic magmas at 850°C (see below) predicts a 18Oplag–melt value of approximately –0·13. Because these fractionations are analytically identical and are at or near the limit of analytical resolution of the oxygen isotope determinations in this study, they do not affect our conclusions.
Figure 4 illustrates 18O values of plagioclase separates from LVC rocks vs whole-rock SiO2 contents. All of the plagioclase separates have 18O values between 6·10 and 8·35, and thus are in the normal range for unaltered silicic igneous rocks as defined by Taylor (1968). Also illustrated in Fig. 4 are calculated magmatic 18O values for little evolved regional basaltic lavas erupted in the arc-front and back-arc regions of the southernmost Cascades during the late Cenozoic, estimated from laser fluorination analyses of olivine phenocryst separates (Underwood et al., 2004; whole-rock 18O values from Borg et al., 1997; Bacon et al., 1997). Following numerous studies (i.e. Muehlenbachs & Kushiro, 1974; Stern et al., 1990; Eiler et al., 2000; Eiler, 2001) we use a 18Omelt–olivine value of 0·4 because these sparely porphyritic lavas (< 5% phenocrysts) have the most primitive characteristics of any volcanic rocks erupted in the southernmost Cascades. These characteristics include: high calculated eruption temperatures (1225–1275°C), moderate to high compatible element concentrations (MgO = 8–11 wt %, Ni > 100 ppm, and Cr > 200 ppm), and FeO*/MgO compositions of ferromagnesian minerals that are in equilibrium with mantle assemblages (Mg-number = 86–91; Borg et al., 1997; Clynne & Borg, 1997; Bacon et al., 1997).
|
Because the petrogenetic histories of the LVC rocks are complex, particularly the inclusion-bearing domes and lava flows, several issues are addressed prior to interpretation of the data. These include: (1) the effect of inclusion disaggregation on the compositions of bulk plagioclase separates from host lavas; (2) whether the mineral separates reliably preserve magmatic oxygen isotope compositions; (3) the oxygen isotope compositions of the mineral separates relative to whole-rock values.
Effect of inclusion disaggregation on bulk plagioclase separates
As described above, plagioclase phenocrysts within inclusion-bearing LVC domes and lava flows are generally of two distinct types: (1) clear phenocrysts with unreacted cores and rims; (2) a smaller population of reacted crystals with finely sieved cores mantled by clear, more calcic rims. As discussed by Clynne (1999), the plagioclase phenocrysts with clear, normally zoned cores and rims represent crystals that precipitated directly from silicic host melt. In theory, these crystals therefore provide the best record of primary oxygen isotope compositions of crustal melts in the LVC system. In contrast, the reacted phenocrysts represent crystals that initially precipitated in silicic melts and that subsequently were mixed into hotter, more mafic magma where they underwent thermal and chemical dissolution followed by precipitation of a more calcic rim in equilibrium with the hybrid liquid (Tsuchiyama, 1985). Some of these crystals were recycled back into the silicic host magmas during mechanical disaggregation of the inclusions upon eruption. Although we strived to exclude reacted phenocrysts during hand-picking of plagioclase crystals from the host lavas, it is therefore possible that our plagioclase separates from inclusion-bearing host rocks contain a small percentage of reacted phenocrysts. On average, reacted phenocrysts make up 25% of the host plagioclase phenocryst populations in the least evolved inclusion-bearing rocks (62 wt % SiO2), with correspondingly smaller proportions in more silicic rocks as a result of reduced amounts of inclusion disaggregation (Clynne, 1999). In contrast to the host lavas, nearly all large plagioclase crystals in the inclusions have cores that are variably reacted and thus are xenocrysts (Clynne, 1999; Tepley et al., 1999; Wilson et al., 2005). Exceptions include a small population of plagioclase phenocrysts in inclusions of a few units that have clear, unreacted calcic cores. These were derived from basaltic or andesitic magmas prior to mixing with silicic magmas.
To evaluate the potential impact of reacted crystals on oxygen isotope analyses of bulk plagioclase separates from the host lavas, we determined 18O values for phenocryst separates from several inclusion–host pairs where the hosts have a wide range in SiO2 contents (63–69 wt % SiO2). These include rocks from the May 1915 eruptions of Lassen Peak, the Chaos Crags domes, and the 27 ka dacite of Lassen Peak (Fig. 2). For the dacite of Lassen Peak and the May 1915 eruption products, we analyzed reacted plagioclase xenocryst separates from multiple inclusions within each unit. In addition, we also analyzed groundmass plagioclase crystals from the inclusions because the textural occurrence of these crystals indicates that they precipitated directly from inclusion-forming magmas. In theory, these latter crystals best record the oxygen isotope compositions of the hybrid inclusion-forming magmas. Data for coexisting inclusion-host pairs are illustrated in Fig. 5 as a function of crystal type. In Fig. 5, tie lines connect data points for plagioclase phenocrysts from host lavas with corresponding xenocrysts and groundmass separates from inclusions.
|
Important features of the data illustrated in Fig. 5 are as follows. First, the range in oxygen isotope compositions of inclusion groundmass crystals is large (6·38–8·35
). This range is similar (albeit shifted to slightly higher values) to the range in whole-rock 18O values determined for 19 inclusions from three Holocene silicic lava flows erupted from Medicine Lake volcano, 125 km north of the LVC (5·4–7·9; Fig. 1; Donnelly-Nolan, 1998). The range in 18O values of LVC inclusions cannot simply reflect mixing with silicic host magmas, which have a more restricted range in composition. The range is also probably not due to fractionation of oxygen isotopes between silicate liquid and aqueous vapor during degassing of volatiles associated with inclusion formation (e.g. Bacon, 1986) because high-temperature melt (glass)–water (aqueous vapor) fractionation factors are small (a few tenths of per mil) and negative (although they are poorly known; Wade et al., 2005). Instead, the range in inclusion 18O values demonstrates that, similar to Medicine Lake volcano (Donnelly-Nolan, 1998), basaltic andesitic- to andesitic-composition magmas intruding shallow chambers beneath the LVC have diverse oxygen isotope compositions, even during discrete eruptive episodes. In a similar sense, Tepley et al. (1999) found that multiple basaltic andesitic inclusions in single Chaos Crags domes show analytically significant variations in 87Sr/87Sr ratios, including some with higher and more variable ratios than their associated rhyodacitic host rocks. Furthermore, Wilson et al. (2005) documented large variations in many other geochemical features of inclusions in the dacitic dome of Mount Helen (Bumpass sequence) that were acquired prior to injection of basaltic andesitic to andesitic magmas into upper crustal silicic chambers. Second, reacted plagioclase xenocrysts in the inclusions have oxygen isotope compositions that are intermediate relative to compositions of host lava phenocrysts and inclusion groundmass crystals. A straightforward interpretation of this relationship is that the xenocrysts have 18O values that reflect crystallization in both host and inclusion-forming magmas, which is consistent with the textural features of the crystals. Tepley et al. (1999) arrived at a similar conclusion in a Sr isotope microanalytical study of reacted plagioclase phenocrysts in Chaos Crags lavas and inclusions. Importantly, the intermediate 18O values of the reacted inclusion xenocrysts also indicate that the range in inclusion microlite 18O values is not an artefact of post-crystallization alteration or contamination of the separates with groundmass glass. Third, relative to inclusion xenocrysts and groundmass crystals, host-rock phenocryst separates have a much more restricted range in oxygen isotope compositions. This suggests that although mafic inputs into shallow magma chambers beneath the LVC may be diverse, oxygen isotope analyses of host lava phenocryst separates may nevertheless be reasonable estimates of the oxygen isotope compositions of crustal melts in the LVC system.
In detail, reacted plagioclase xenocrysts from inclusions within the dacite of Lassen Peak have a relatively wide range in 18O values (6·74–7·25; Table 1; Fig. 5a). Simple mass-balance relationships, assuming conservatively that plagioclase phenocryst separates from the dome (average of seven analyses from different lobes is 6·98 ± 0·04; Table 1) are a mixture of 25% reacted crystals derived from inclusion disaggregation (18O = 7· 00) and 75% unreacted phenocrysts (Clynne, 1999), indicate that a separate consisting exclusively of unreacted phenocrysts should yield an oxygen isotope analysis of 18O = 6·99. This value is identical to the analyses cited above determined for different lobes of the dome. Similar reasoning for the May 1915 dacitic dome and lava flow (average of two plagioclase phenocryst separates is 7· 00 ± 0·02) suggests that the maximum shift resulting from addition of 25% reacted crystals to a pure separate of unreacted phenocrysts is –0·09, which is at the limit of analytical resolution. The May 1915 lava flow and dome are among the most mafic inclusion-bearing rocks in the LVC (62·8–63·4% SiO2) and therefore plausibly contain the highest proportions of reacted plagioclase phenocrysts (Clynne, 1999). As such, the effect of inclusion disaggregation on the oxygen isotope compositions of mineral separates from these rocks should approach the maximum in the LVC. The effect on oxygen isotope compositions of recycling plagioclase phenocrysts from host to inclusion and back to host therefore appears to be negligible for bulk separates from host rocks. Alternatively, our host rock plagioclase separates may be relatively free of reacted phenocrysts as a result of thorough hand picking. In either case, we contend that plagioclase separates from the lavas and domes provide reasonable estimates of the oxygen isotope compositions of crustal melts in the LVC system, especially for high-silica dacitic and low-silica rhyolitic magmas containing relatively small proportions of disaggregated inclusion material.
Plagioclase–quartz fractionations
Figure 6 illustrates oxygen isotope thermometry derived from plagioclase and quartz phenocryst separates from the same rocks. Lines in Fig. 6 labeled with temperatures in °C are equilibrium quartz–plagioclase fractionations derived from linear interpolation between values for quartz–albite and quartz–anorthite (An35; Borg & Clynne, 1998; Clynne, 1999; Tepley et al., 1999) from the data of Chiba et al. (1989). The data of Bottinga & Javoy (1975) and Clayton & Kieffer (1991) yield essentially identical results. Quartz–plagioclase fractionations from rocks of the silicic lava and dome field range from +1· 03 to +1· 33 and yield temperature estimates between approximately 800 and 900°C. These temperatures are similar to magmatic temperature estimates for LVC silicic magmas (800–850°C) based on Fe–Ti oxide and two-pyroxene thermometry (Heiken & Eichelberger, 1980; Borg & Clynne, 1998). We therefore consider that these fractionations and temperatures record equilibrium values for dacitic to low-silica rhyolitic magmas.
|
, dacitic and rhyolitic rocks. Typical analytical uncertainty (±0·1In contrast, Twin Lakes sequence quartz-bearing mafic andesites erupted on the Central Plateau have
18Oqtz–plag much greater than +1· 5 and yield temperature estimates well below 700°C. Plagioclase and quartz in these samples are thus clearly not in oxygen isotope equilibrium and the quartz crystals must be xenocrysts, which is consistent with textural features such as rounding, embayment, and clinopyroxene-bearing reaction coronae (Clynne, 1990; Borg & Clynne, 1998). In addition, the elevated value for plagioclase phenocrysts from the andesitic Fantastic Lava Beds flow 2 erupted from Cinder Cone (18O = 8·15; Table 1) suggests that these crystals are also xenocrysts.
Phenocryst separates versus whole-rock data
Figure 7 illustrates whole-rock 18O values versus analyses of plagioclase phenocryst separates from the same rocks. We assume that the plagioclase results closely approach magmatic values based on the discussion above, because the crystals are visibly pristine, and because 18Oplag–melt values are typically small ( –0·1) and near the limit of analytical uncertainty in andesitic to dacitic composition magmas (Taylor, 1968; Chiba et al., 1989; Bindeman et al., 2001). The continuous line in Fig. 7 indicates a fractionation of 18Oplag–melt = 0, and the dashed lines delineate an error envelope of ±0·1. Arrows in the figure schematically indicate the effects of low- and high-temperature alteration on whole-rock values. Many of the whole-rock values are 0·5–2·0 higher than the plagioclase phenocryst values. We therefore conclude that many of the whole-rock 18O values are not pristine and have been increased by low-temperature hydration of glassy groundmass. This appears to be a common process affecting glassy volcanic rocks, even those that are young and appear pristine in hand sample and microscopically (Taylor, 1968; Muehlenbachs & Clayton, 1972; Pineau et al., 1999; Bindeman et al., 2004).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL AND PETROLOGICAL...SAMPLE COLLECTION AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Oxygen isotope constraints on sources of LVC magmas
The shaded field in Fig. 4 is the ‘normal
18O array’ from Bindeman et al. (2004), which depicts modeled oxygen isotope values at a given SiO2 content resulting from closed-system crystal–liquid fractionation of primitive basaltic magma (18O = 5·8 ± 0·2) under a variety of fractionating mineral assemblages and parental magma compositions. The array includes fractionation of subalkaline basaltic magma involving olivine and anorthitic plagioclase on the liquidus, which is typical of arcs (Bindeman et al., 2004). As illustrated in Fig. 4, nearly all rocks from the LVC, with the exception of a subgroup erupted during the Lassen stage, plot well above the closed-system mantle fractionation field. LVC magmas therefore do not represent a continuous liquid line of descent from typical mantle-derived basalts, which is consistent with the results of previous studies that indicate a major role for crustal melting and hybridization processes (Heiken & Eichelberger, 1980; Bullen & Clynne, 1990; Clynne, 1990, 1999; Tepley et al., 1999; Wilson et al., 2005).
Most evolved LVC rocks also have compositions that are significantly enriched in 18O relative to the majority of little evolved basaltic lavas erupted in the arc-front and back-arc regions of the southernmost Cascades during the late Cenozoic. Therefore, although LVC silicic rocks have radiogenic isotope compositions that fall within the ranges for these basaltic lavas, they cannot, as previously argued (Bullen & Clynne, 1990; Borg & Clynne, 1998; Hart et al., 2002) be explained by direct partial melting of intrusive equivalents of these rocks; most must also contain high 18O crustal components. For example, Borg & Clynne (1998) modeled partial melting of mafic lower crust under variable f(H2O) to produce high-silica dacitic to rhyolitic magmas at the LVC and other composite volcanoes in the southernmost Cascades. Because the dacitic to rhyolitic magmas do not exhibit heavy rare earth element depletions (relative to chondrites), modeled restite mineral assemblages are garnet- (and olivine-) free, but plagioclase-rich with variable amounts of pyroxene ± amphibole. These residual assemblages have very small equilibrium bulk mineral–melt fractionations at high temperatures that are unable to produce significant enrichments or depletions in 18O (0·1) relative to their crustal sources at the degrees of melting inferred for southernmost Cascades silicic magmas (20–25%; Chiba et al., 1989; Borg & Clynne, 1998; Bindeman et al., 2005). Furthermore, if crustal sources for the magmas are more silicic and, presumably, more plagioclase-rich than the basaltic compositions used in the calculations of Borg & Clynne (1998; see below), then the partial melts will have 18O values essentially identical to their sources because of smaller bulk mineral–melt fractionations and the requirement of larger degrees of partial melting. 18O values of silicic magmas erupted at the LVC are therefore considered to be representative of their crustal sources and these are, in most cases, significantly higher than those of little evolved basaltic magmas erupted in the arc front and back-arc regions of the southernmost Cascades. The oxygen isotope data for silicic rocks in the LVC therefore provide unequivocal evidence for partial melting of crustal source rocks other than little evolved, late Cenozoic arc-related lower crustal mafic intrusions.
The data illustrated in Fig. 4 are consistent with a three-stage model for magma evolution in the continental crust beneath the LVC that has been elaborated on in detail in numerous studies (e.g. Heiken & Eichelberger, 1980; Bullen & Clynne, 1990; Clynne, 1990, 1999; Tepley et al., 1999; Wilson et al., 2005). The first stage involves production of compositionally diverse parental magmas during assimilation plus fractional crystallization of mantle-derived mafic magmas in heterogeneous lower continental crust (continuous curved line in Fig. 4). The second stage involves production of compositionally more uniform silicic magmas by homogenization of bulk crustal melts. In the third stage, mixing and mingling of endmember magmas in crustal chambers occurs to produce a range of intermediate composition hybrid magmas. In this regard, we infer, based on the relationships illustrated in Fig. 3 and elaborated on in previous studies (e.g. Bullen & Clynne, 1990; Clynne, 1990, 1999), that rocks with SiO2 contents >62% are principally mingled hybrid magmas, whereas those with <62% SiO2 are mixed hybrid magmas (Fig. 4). The new result of the present study, however, is that crustal sources involved in the production of many silicic LVC magmas were not limited to relatively primitive lower crustal mafic intrusions similar to regional basaltic lavas in the southernmost Cascades. For many rocks, a source component derived from relatively high 18O crustal material is also required. Furthermore, the large range in 18O values for the basaltic to basaltic andesitic magmas that extends from the normal mid-ocean ridge basalt (N-MORB) field in Fig. 4 indicates that the crust assimilated by mafic magmas at the LVC was very diverse in oxygen isotope composition and that this diversity was not homogenized prior to injection of the magmas into shallow crustal magma chambers. As noted above, 18O values determined for multiple inclusions in single flows (e.g. Glass Mountain) by Donnelly-Nolan (1998) at Medicine Lake volcano are similar to the range observed in LVC basaltic andesitic to andesitic inclusions and Brokeoff lavas, indirectly corroborating this result.
Oxygen isotope variations with stratigraphic and spatial position
Figure 8a illustrates 18O values of plagioclase from basaltic andesitic to andesitic magmas vs bulk-rock SiO2 contents as a function of stratigraphic group. In general, there is a small increase in 18O values of basaltic andesitic to andesitic magmas in the LVC with decreasing age. 18O values of lavas erupted from Brokeoff volcano average 6·6. This value is slightly lower than the average value (7·1) for younger basaltic andesitic to andesitic composition Lassen stage magmas. These relationships may indicate that differentiation of Lassen stage intermediate composition magmas was more strongly controlled by crustal assimilation relative to Brokeoff stage magmas. An alternative interpretation is that rates of assimilation for mafic magmas in both stages were approximately equal, but that crustal sources for the basaltic andesitic and andesitic magmas varied spatially in O isotope compositions. Specifically, basaltic andesitic to andesitic composition magmas present as inclusions in lavas erupted on the flanks of the LVC may have melted crustal rocks with slightly higher 18O values relative to sources for lavas erupted from Brokeoff volcano at the center of the system. Because, as discussed below, Lassen stage dacitic to rhyolitic rocks show similar spatial trends, we prefer the latter interpretation.
|
Figure 8b illustrates
18O data for plagioclase phenocrysts from Lassen stage inclusion-bearing host-rocks vs whole-rock SiO2 contents by stratigraphic group. The figure demonstrates the lack of a significant relationship between 18O values and stratigraphic age. Host rocks from the youngest eruptive sequence (Eagle Peak sequence) tend to higher 18O values at a given SiO2 content relative to rocks from the Bumpass sequence, although rocks from both sequences have values that essentially span the entire range of the silicic rocks.
In contrast, 18O values for Lassen stage host rocks show broad spatial patterns, particularly for high-silica dacitic to rhyolitic magmas (e.g. 68–71 wt % SiO2; from this point on referred to as ‘rhyodacitic’) that probably contain the smallest amounts of disaggregated inclusion material. For example, Fig. 8b demonstrates that rhyodacitic magmas erupted from vents located within 3·5 km of Diamond Peak, the inferred center of Brokeoff volcano, have consistently lower 18O values (average 6·3 ± 0·1) at given SiO2 contents relative to the same composition magmas erupted from distal vents (e.g. 7–12 km; average 7·1 ± 0·1). Furthermore, rhyodacitic magmas erupted from vents situated at transitional distances from Diamond Peak (e.g. 3·5–7· 0 km; Fig. 2) have intermediate values and span a larger range (average 6·7 ± 0·2). Oxygen isotope data for quartz phenocrysts from Lassen stage domes and lava flows display similar relationships (Fig. 8c), as do data for plagioclase crystals from basaltic andesitic to andesitic composition rocks (see above; Fig. 8a).
We interpret the spatial trends in 18O values of Lassen stage host rocks to reflect variations in crustal source compositions with distance from the center of Brokeoff volcano because numerous petrological studies demonstrate similar differentiation processes involving magma mixing and mingling for all rocks regardless of vent location (e.g. Heiken & Eichelberger, 1980; Clynne, 1990, 1999; Thomas & Tait, 1997; Tepley et al., 1999; Wilson et al., 2005). We discount sub-crustal processes such as addition of high 18O components through slab-derived fluids to mantle melt generation zones (e.g. Borg & Clynne, 1997; Eiler et al., 2000) given the relatively restricted range in oxygen isotope compositions of little evolved basaltic rocks erupted in the arc-front and back-arc regions of the southernmost Cascades as documented from analyses of olivine phenocrysts (Fig. 6; Underwood et al., 2004). Furthermore, there is no evidence to support the idea that subcrustal processes have produced spatial trends in the oxygen isotope compositions of erupted basaltic lavas similar to those documented here for the LVC silicic magmas (Underwood et al., 2004). On the contrary, as described by Underwood et al. (2004), little evolved mafic lavas with the highest olivine 18O values (6·0–6·3) erupted in the fore-arc region several tens of kilometers west of the LVC, whereas those with the lowest 18O values (5·5–6·0) erupted beneath the arc-front (where the LVC is situated) and back-arc regions.
The data suggest that 18O values of crustal source rocks for LVC silicic magmas decrease toward the central focus of the magmatic system. There are two potential origins of this trend. The first is increased melting of low 18O, hydrothermally altered upper crustal rocks toward the center of Brokeoff volcano (e.g. Feeley & Sharp, 1995; Bindeman et al., 2004). This inference is suggested by the presence of a 10 km2 zone of intense hydrothermal alteration within the eroded core of Brokeoff volcano where whole-rock 18O values are as low as 0·6 (Rose et al., 1994; John et al., 2006). The second is melting of initially high 18O continental crust increasingly hybridized by mantle-derived mafic magmas (18O = 5·8–6·0) toward the center of the system. This idea is supported by the observation that Brokeoff volcano represents the focus of the LVC where the largest volume of magma has erupted and also where, by inference, the greatest intrusion of mantle-derived mafic magmas into the crust has occurred (Guffanti et al., 1996). Furthermore, Mesozoic granitoid basement with comparatively high 18O values (relative to mantle values) is likely to be present in the lower to middle continental crust beneath the center where production of the silicic magmas by partial melting is thermodynamically more favorable than at surface or near-surface conditions (Dufek & Bergantz, 2005; Annen et al., 2006). A third possibility is increased melting of solfatarized rocks with 18O >8 with distance from the center of Brokeoff volcano (e.g. Rose et al., 1994; Zellmer et al., 2003). We discount this possibility because, as discussed by Rose et al. (1994) and John et al. (2006), rocks with these features are restricted to small areas near active geothermal systems in the altered core of Brokeoff volcano and there is no evidence for increased abundance of these rocks in areas distant from the central focus of the volcanic system. Finally, a fourth possibility is that the analyzed samples become increasingly altered toward the center of Brokeoff volcano. We discount this possibility because all rocks are fresh, were collected from areas well outside the zone of hydrothermal alteration, and quartz–plagioclase fractionations are reasonable for dacitic to low-silica rhyolitic melts at magmatic temperatures (e.g. Fig. 5).
To distinguish between the two scenarios above, Fig. 9a illustrates 18O values of plagioclase separates from rhyodacitic Lassen stage host rocks vs 87Sr/86Sr ratios for the same rocks (data from Bullen & Clynne, 1990; Table 1) plotted as a function of distance from Diamond Peak. In Fig. 9a, only data for rhyodacitic Lassen stage rocks (e.g. >68 wt % SiO2) are illustrated because the 18O values in this study were determined on plagioclase separates, whereas the 87Sr/86Sr values of LVC rocks were determined in previous studies on whole-rocks. Thus, it is essential to limit our comparison only to those rocks containing the least amounts of disaggregated inclusion material, as this affects the whole-rock 87Sr/86Sr values but may have little effect on 18O values (see discussion in previous section). Also illustrated in Fig. 9a are fields for Sierra Nevada–Klamath Mountains granitoids and partially melted granitic xenoliths in Cinder Cone lavas (data from DePaolo, 1981; Barnes et al., 1992; Borg & Clynne, 1998), Brokeoff volcano lavas (Sr isotope data from Bullen & Clynne, 1990; Table 1), and little evolved basaltic magmas erupted in the arc-front and back-arc regions (Sr isotope data from Borg et al., 1997; oxygen isotope data from S. J. Underwood & T. C. Feeley, unpublished data).
|
As illustrated in Fig. 9a, rhyodacitic rocks erupted from vents located at distances greater than 7 km from Diamond Peak generally have higher 18O/16O and 87Sr/86Sr ratios relative to rocks erupted from proximal vents (e.g. <3·5 km; no Sr isotopic analyses are available for intermediate distance rhyodacitic rocks). We interpret this relationship to indicate that the spatial trends in oxygen isotope compositions of LVC silicic magmas reflect melting of continental crust progressively less hybridized by lower
18O, lower 87Sr/86Sr basaltic magmas away from the center of the volcanic system. This idea is further supported by the following lines of evidence. First, there is a greater proportion of rhyodacitic magma erupted from distal vents relative to magmas erupted from more proximal vents (Clynne, 1990). This is consistent with more silica-rich rocks in the zone of crustal melting beneath distal vents, smaller amounts of mafic recharge into upper crustal chambers and correspondingly smaller amounts of inclusion disaggregation, or, more probably, both. Second, incompatible trace element ratios such as Rb/Nb, Ba/Zr, Ba/Nb, and Rb/Ba become progressively higher and more crustal-like in Lassen stage rhyodacitic magmas erupted at increasing distances from the core of Brokeoff volcano relative to the regional basaltic magmas, which have more restricted ranges in composition (Fig. 9b–e).Significant elevation in incompatible trace element ratios such as Rb/Ba in the rhyodacitic magmas also provides additional evidence for the lack of a crystallization–differentiation relationship with the regional basaltic lavas, as these ratios should remain relatively unchanged during differentiation because of similar bulk mineral–melt distribution coefficients (e.g. Davidson et al., 1988). Finally, slightly increasing CaO/Al2O3 values in rhyodacitic rocks erupted at increasing distance from Diamond Peak are also consistent with a more silicic crustal source component (Fig. 9f; Beard & Lofgren, 1991; Shukuno et al., 2006). For example, the results of dehydration-melting experiments on amphibolites at 6·9 kbar by Beard & Lofgren (1991) showed that rhyodacitic magmas (70 wt % SiO2) produced by melting of andesitic composition rocks (sample 557 of Beard & Lofgren, 1991; 57· 0 wt % SiO2) have CaO/Al2O3 of 0·23, whereas melting of basaltic composition rocks (sample 571; 51·4 wt % SiO2) produces rhyodacitic magmas with CaO/Al2O3 of 0·15. This range is nearly identical to the range and in the same spatial sense as that observed in LVC rhyodacitic magmas (Fig. 9f).
Rhyodacitic magmas erupted from vents located at distances greater than 7 km from Diamond Peak also have 18O values that plot between those of Brokeoff lavas and Sierra Nevada–Klamath granitoids. This demonstrates again that these rocks cannot have formed simply by partial melting of intrusions related to Brokeoff volcano or the regional basaltic lavas, but must include high-18O crustal material. This component plausibly represents Mesozoic to Paleozoic plutonic rocks, as plagioclase xenocrysts from late Holocene Cinder Cone volcanic rocks have 18O values similar to the granitoids. In contrast, rocks erupted from vents located at distances less than 3·5 km from the center could, conceivably, have formed by partial melting of intrusions related to Brokeoff volcano or regional basaltic lavas, as they have isotopic compositions similar to these units. However, most Brokeoff lavas must also contain isotopically heavy crustal material because, with the exception of only a few samples (e.g. Fig. 4), they have higher 18O values than little-evolved basaltic magmas erupted in the arc-front and back-arc regions of the southernmost Cascades (Fig. 9).
Source rock proportions and hybridization of the continental crust
A key advantage in the interpretation of oxygen isotope data is that estimating the proportions of endmember mantle and crustal sources contained within hybrid igneous rocks is relatively straightforward because of the small differences in the concentrations of elemental oxygen between rock types. In this case, the 18O values of the hybrid magmas will fall along a simple linear mixing line between values of the endmembers, and the proportions can be determined directly from the positions of the rocks on the line. These calculations are further simplified by the fact that most primary or near primary mantle-derived basalts have a relatively restricted range in oxygen isotope compositions (18O = 5·7 ± 0·3; e.g. Eiler et al., 2000; Valley et al., 1998). For the mixing calculations described below, we use a 18O value of 6·0 for the mantle endmember, which derives from our olivine phenocryst data for the most MgO-rich late Cenozoic basaltic lava erupted along the modern arc-front in the southernmost Cascades (e.g. Fig. 4; Borg et al., 1997; Underwood et al., 2004).
A larger uncertainty involves estimating the composition of the unhybridized crustal endmember (i.e. juvenile crust prior to recent addition of basaltic magmas), as a multitude of sources are possible, given the complex nature of the crust beneath convergent margins (e.g. Lackey et al., 2005) For the LVC we estimate this composition directly from the 18O values of xenocrysts and xenoliths contained within the rocks, particularly those erupted during the Twin Lakes sequence on the Central Plateau, distant from the central focus of the magmatic system and presumably where the smallest degree of crustal hybridization as a result of injection of mantle-derived basaltic magmas has occurred. A bulk separate of plagioclase xenocrysts from Fantastic Lava Beds flow 2 erupted on the Central Plateau at Cinder Cone (Twin Lakes sequence) 15 km NE of Diamond Peak yielded a 18O value of 8·15. Similarly, a bulk separate of plagioclase xenocrysts in a Brokeoff volcano Mill Canyon sequence lava flow (andesite of Blue Lake Canyon) yielded a 18O value of 8·96 (Table 1). Values of several quartz xenocryst separates from different Cinder Cone eruptive units range from 9·71 to 10·85 (Table 1). These values are nearly identical to those of quartz grains from granitic xenoliths present in Pleistocene and Late Holocene lava flows erupted at Medicine Lake volcano (Donnelly-Nolan, 1998). Similar to Medicine Lake volcano, xenocrystic plagioclase and quartz with high 18O values along with quartz- and plagioclase-bearing, partially melted felsic xenoliths suggest that the juvenile crustal source rocks for LVC magmas may be broadly granitic (sensu lato; Borg & Clynne, 1998). Therefore, taking the average of the xenocryst compositions to reflect the composition of juvenile continental crust in the zone of partial melting, we estimate the 18O value of the unhybridized crustal endmember to be 9·0. This value is within the range of compositions determined for evolved Mesozoic Sierra Nevada–Klamath granitoids discussed by Barnes et al. (1992) and DePaolo (1981), although it is higher than whole-rock 18O values of exposed granodioritic country rocks (7·3–8·3) in the southernmost Cascades sampled by Borg & Clynne (1998). However, it is similar to whole-rock 18O values (7·8–8·8) of partially melted crustal xenoliths in Cinder Cone eruptive units (Borg & Clynne, 1998), the majority of pristine, unmelted granitic xenoliths in Medicine Lake volcano lavas, 125 km to the north of the LVC (8·8–9·8; Donnelly-Nolan, 1998), and average estimated values for the continental crust (8·9 ± 0·7) and granitoids worldwide (9·3 ± 2·4; Simon & Lécuyer, 2005).
Figure 10a illustrates the results of mixing calculations for inclusion-bearing Lassen stage dacitic and rhyolitic rocks using the mantle and crustal endmembers defined above. The 18O values of bulk plagioclase phenocryst separates from these rocks are consistent with the magmas containing on a molar oxygen basis approximately 42 to 4% isotopically heavy continental crust, with proportions declining in broadly regular fashion toward the center of the LVC. As described above, we interpret this trend to reflect melting of high-18O crustal material increasingly hybridized by intrusion of mantle-derived basaltic magmas toward the center of the volcanic system.
|
Also shown in Fig. 10a are the results of mixing calculations for basaltic andesitic to andesitic composition Brokeoff volcano lava flows and Lassen stage inclusions. Noteworthy features of the results include the following. First, the basaltic andesitic to andesitic rocks display a larger calculated range in crustal proportions (generally 4–50% but as high as 78%) relative to the more silicic Lassen stage host rocks. As
18O values of the LVC basaltic andesitic to andesitic rocks show no apparent correlation with bulk-rock composition (e.g. Fig. 4), we interpret this feature to reflect assimilation of heterogeneous lower crustal rocks, some with very high 18O values, prior to mixing with well-homogenized bulk crustal melts. Several workers have noted analogous variations at other continental arc composite volcanoes wherein silicic lavas are typically more uniform in isotopic ratios and other compositional parameters relative to the more mafic rocks (e.g. Tormey et al., 1995; Feeley et al., 1998; Dungan et al., 2001). At the Tatara–San Pedro complex, Chile, Feeley et al. (1998) and Dungan et al. (2001) related these features to assimilation of compositionally heterogeneous wall-rocks during ascent in lower crustal conduits. In this regard, it is intriguing that in the Tehachapi Mountains, southern California, where the deepest parts of the Sierra Nevada batholith are exposed, rocks with the highest and most variable 18O values predominate in deep parts of the batholith, with lower values occurring in shallower parts (Lackey et al., 2005). Second, Lassen stage basaltic andesitic to andesitic inclusions generally have higher 18O values relative to older Brokeoff volcano lava flows. A similar shift was observed at Medicine Lake volcano by Donnelly-Nolan (1998), where late Holocene basaltic andesitic to andesitic inclusions generally have higher and more diverse 18O values relative to Pleistocene andesitic lava flows. Donnelly-Nolan (1998) interpreted the shift at Medicine Lake volcano to indicate a change in the kind of crust involved and depth of magma storage and differentiation with time (i.e. melting of higher 18O, more granitic crust). For LVC basaltic andesitic and andesitic magmas we concur with this interpretation, but suggest that the trend to higher 18O values with time also reflects a lateral shift in the zone of differentiation and crustal assimilation. Specifically, because the inclusions, like their silicic host-rocks, erupted on the flanks of Brokeoff volcano, they probably assimilated heterogeneous continental crust less hybridized by mantle-derived mafic magmas relative to intermediate-composition magmas erupted in the central focus region.
A basic conclusion of the oxygen isotope values for the LVC rocks is that older continental crust is an important source component of the magmas. Although the importance of older crust has been documented in many other studies of continental arc volcanic centers, it has gone previously unrecognized at the LVC, owing to large ranges in radiogenic isotope ratios and bulk elemental compositions of spatially associated mafic rocks (Bullen & Clynne, 1990; Borg & Clynne, 1998; Hart et al., 2002). A more fundamental and intriguing conclusion with respect to the genesis of magmas in arc settings is that oxygen isotope data can provide a powerful means to assess spatial contributions of mantle and crustal sources to magmas on the scale of single, long-lived centers. In the case of the LVC, the oxygen isotope values of the erupted rocks indicate that older crust may be less important as a source in melt production regions characterized by abundant basalt injection, as inferred from erupted magma volumes. Similar to the model for the genesis of chemical diversity in silicic arc magmas recently proposed by Annen et al. (2006), melts produced in deep crustal melt zones (i.e. MASH zones of Hildreth & Moorbath, 1998) are typically generated by two distinct processes: partial crystallization of mantle-derived basaltic sills coupled with assimilation of crustal rocks (AFC curve on Fig. 4) and bulk partial melting of the surrounding crust, which can include pre-existing metasedimentary and meta-igneous basement rocks and earlier basaltic intrusions. Mixing of melts produced by these processes leads to the diversity in isotope and trace element chemistry of silicic rocks. As discussed by Annen et al. (2006), in deep crustal regions subjected to repeated basalt intrusion, the earliest intruded basalts may cool below or remain near their solidus temperatures only to be reheated and partially melted by subsequent basaltic intrusions. In this situation, erupted silicic partial melts will have isotopic compositions similar to the intruding basalts with diminishing contributions from older crust with time. In contrast, in deep melt zones characterized by more limited basalt intrusion, production rates of melt by partial melting of older crustal rocks reach a maximum early and then diminish because of conduction of heat away from basalt to surrounding crustal rocks. In this case, erupted silicic melts will contain isotopic and other compositional features indicating a larger contribution from pre-existing crust. Oxygen isotope values of LVC volcanic rocks appear to provide empirical evidence for these deep-seated processes wherein magmas erupted near the central vent region reflect large-scale melting of mantle-derived mafic magmas emplaced during initial growth of the system. In contrast, magmas erupted from flank vents where less voluminous magmatism occurs appear to contain a greater contribution from pre-existing continental crust and hence represent greater amounts of crustal recycling, as opposed to new crustal growth.
In a manner similar to that above, it is possible to broadly estimate the extent of hybridization of the continental crust using the compositions of plagioclase phenocryst separates from rhyodacitic Lassen stage rocks, which are least likely to be influenced by inclusion disaggregation. In this case the 18O values of the separates are taken to be representative of the oxygen isotope compositions of unmodified crustal melts formed during Lassen stage magmatism and thus of the continental crust in the zone of melt production and homogenization (see discussion above). The results of the calculations are illustrated in Fig. 10b and are essentially the inverse of the results shown in Fig. 10a. The 18O values of the most silicic Lassen stage lavas suggest that on a molar oxygen basis the continental crust in melt production zones beneath the LVC has been substantially modified by intrusion of mantle-derived basalts, with the degree of hybridization ranging from 60% at distances between 7 and 10 km from the center of the system to 97% directly adjacent to and within the focus region (e.g. <3·5 km). At greater distances, for example on the Central Plateau and more regionally in the southernmost Cascades, the continental crust may at present be relatively little modified at the depth of silicic melt generation, as evidenced by the common presence of high 18O xenocrysts and xenoliths in erupted basaltic to basaltic andesitic magmas (this study; Borg & Clynne, 1998; Underwood et al., 2004).
We can compare our results with those of Guffanti et al. (1996), who calculated the volume of basalt input into the lower crust beneath the LVC from petrological models and the mass and heat requirements required to generate the magmas. For LVC rocks younger than Rockland stage units, Guffanti et al. (1996) estimated addition of 143 km3 of basalt to the crust, which includes 112 km3 associated with the evolution of Brokeoff volcano and 31 km3 associated with development of the Lassen stage lava and dome field and Central Plateau. Considering volume estimates for erupted Brokeoff volcano lavas (100 km3) and Lassen stage magmas (60 km3) and average primitive basaltic components in these magmas based on the calculations above (80%), we estimate that 130 km3 of basalt was added to the crust during the past 600 kyr of volcanism at the LVC. This amount agrees well with the volume of basaltic input calculated by Guffanti et al. (1996), especially considering the numerous uncertainties, simplifications, and assumptions inherent in both approaches.
The calculated mass-balance and hybridization values discussed above are, of course, semi-quantitative because they depend strongly upon our choices of endmember compositions. For example, calculated 18O values for Mg-rich arc-front and back-arc basalts in the Lassen region range from 5·9 to 6·4. These values are elevated relative to MORB and may reflect subduction zone enrichment beneath the southernmost Cascades, crustal contamination, or both (Bacon et al., 1997; Borg et al., 1997; Underwood et al., 2004). Nevertheless, if the latter value is used as the composition of the mafic endmember, calculated proportions of isotopically heavy crust in Lassen stage dacites range up to 25% on a molar oxygen basis for lavas erupted on the periphery of the system. It should be noted, however, that this figure represents a minimum value because some of the rhyodacitic rocks have 18O values lower than 6·4 and thus must be derived from partial melting of rocks with lower values. Regarding variations in the crustal endmember, unmelted granitic xenoliths in Medicine Lake volcano lavas are as heavy as 10. Similarly, quartz from xenoliths in Cinder Cone are as heavy as 10·9. Thus, if a crustal endmember value of 10 is used instead of 9, calculated proportions of isotopically heavy crust in Lassen stage rhyodacitic rocks range up to 30% on a molar oxygen basis. On the other hand, using a crustal endmember value of 8 predicts proportions of isotopically heavy crust in Lassen stage rhyodacitic rocks erupted on the periphery of the system of up to 60% on a molar oxygen basis. In sum, using empirical evidence for the O isotope compositions of the endmembers, the maximum proportions of isotopically heavy crust in Lassen stage silicic magmas may range from 25 to 60%. Our estimate of 42% represents the average of these values and is independently supported by heat and mass-balance requirements (Guffanti et al., 1996).
Another issue is that there are no explicit constraints on the timing of basalt intrusion into the crust and associated hybridization. We contend, based on the wide range in 18O values of Brokeoff volcano lavas followed by the regular spatial trends in Lassen stage silicic lavas, that crustal hybridization was probably associated with recent magmatism directly related to development of the LVC. However, hybridization could equally well reflect a more protracted process involving intrusion of basaltic magmas during earlier magmatic activity associated with production of the Rockland tephra and regional mafic volcanism. In either case, the oxygen isotope values of LVC magmas indicate that chemical and isotopic distinctions between primary mantle-derived basaltic magmas and the lower continental crust may decrease with increasing proximity toward the center of long-lived composite volcanoes and probably also with time. Thus, silicic magmas at mature arc volcanic centers (or compositionally similar plutonic bodies; i.e. Ratajeski et al., 2001) with radiogenic isotope compositions similar to assumed mantle-derived values or spatially associated mafic magmas may not provide a priori evidence for a largely crystallization–differentiation origin (Yamamoto, 2007). In these cases other types of evidence must be carefully considered in combination, including petrographic features, trace element data, and oxygen isotope values. Perhaps a more intriguing proposal is that, viewed in light of the results of this work, well-documented across-strike changes in compositional features of continental arc rocks, such as increasing K2O contents or 87Sr/86Sr isotope ratios with distance from the arc front (e.g. Kistler & Peterman, 1973; Tatsumi & Eggins, 1995) may not necessarily represent a fundamental change in crustal composition prior to the onset of magmatism. Oxygen isotope values of LVC rocks illustrate on a relatively small scale the strong influence that intrusion of mantle-derived mafic magmas can have on modifying the composition of pre-existing continental crust in regions of melt production. Given this result, larger-scale regional trends in magma compositions may reflect an analogous but more extensive process wherein the continental crust becomes progressively hybridized beneath frontal arc localities as a result of protracted intrusion of subduction-derived basaltic magmas (Feeley, 1993; Walker et al., 1995; Lucassen et al., 2006; Schnurr et al., 2007). Although the suitability of this idea is at present difficult to demonstrate on a large scale for any single arc, given the paucity of high-precision oxygen isotope data, it is provocative because it links across-strike variations in the compositions of erupted rocks, the composition of the continental crust and magma volumes ultimately back to mantle processes and thus the subduction process itself. In other words, relative to back-arc regions, greater time-integrated primary melt production beneath frontal arc regions may result in production of intermediate to silicic composition magmas with more ‘mantle-like’ chemical and isotopic features as they interact with continental crust increasingly hybridized by young, subduction-related basaltic intrusions.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL AND PETROLOGICAL...SAMPLE COLLECTION AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- The range in oxygen isotope ratios of volcanic rocks erupted at the LVC greatly exceeds that expected for production of the magmas by partial melting of little evolved basaltic lavas erupted in the arc front and back-arc regions of the southernmost Cascades during the late Cenozoic. Therefore, most LVC rocks must contain high 18O crustal material.
- 18O values of erupted rocks show strong spatial patterns. Rhyodacitic rocks erupted from vents located near the inferred center of the LVC have consistently lower 18O values at given SiO2 contents relative to rocks erupted from more distal vents. Basaltic andesitic to andesitic composition rocks show a similar spatial trend, although as a group 18O values of these rocks are more variable and extend to higher values than those for the rhyodacitic rocks.
- Such features are interpreted to reflect assimilation of heterogeneous lower continental crustal by mafic magmas, possibly in conduits, followed by mixing or mingling with silicic magmas formed by partial melting of initially high 18O continental crust increasingly hybridized by lower 18O mantle-derived basaltic magmas toward the center of the system.
- Mixing calculations using estimated endmember source 18O values imply that LVC magmas contain on a molar oxygen basis 42–4% isotopically heavy continental crust, with proportions declining in a broadly regular fashion toward the center of the LVC.
- 18O values of rhyodacitic rocks suggest that the continental crust in melt generation zones beneath the LVC has been substantially modified by intrusion of mantle-derived basalts, with the degree of hybridization ranging on a molar oxygen basis from 60% at distances up to 12 km from the center of the system to 97% directly beneath the focus region.
- Considering volume estimates for LVC eruptive products, these values suggest that 130 km3 of basalt was added to the crust during the past 600 kyr of volcanism at the LVC. This estimate agrees well with previous estimates for the volume of basalt input into the lower crust beneath the LVC from petrological models and the mass and heat requirements required to generate the magmas.
- These results demonstrate on a small scale the strong influence that intrusion of mantle-derived mafic magmas can have in modifying the composition of pre-existing continental crust in regions of melt production. Given this result, larger-scale, regional trends in magma compositions may reflect an analogous but more extensive process wherein the continental crust becomes progressively hybridized beneath frontal arc localities as a result of protracted intrusion of subduction-derived basaltic magmas.
|
ACKNOWLEDGEMENTS |
|---|
This work was supported by US National Science Foundation grant EAR-9983769. The authors wish to thank Jeff Schmidt, Andrea Frangiosa, Sandra Underwood, and Luke Wilson for able and cheerful field assistance; Toti Larson, Zach Sharp, Viorel Atudorei, and Dan Breecker for technical assistance; Peter Larson and Sandra Underwood for helpful discussions; and Patrick Muffler for assistance in locating sites of previously collected samples and commenting on an earlier version of the manuscript. We also thank the staff at Lassen Volcanic National Park for their hospitality and assistance, without which this work would have been immeasurably more difficult. The lead author is grateful to the Institute of Mineralogy and Geochemistry at the University of Lausanne, Switzerland, and especially to Mike Cosca and Lukas Baumgartner, for generously providing logistics, facilities, and hospitality during the period the manuscript was written. The manuscript was substantially improved by constructive and helpful reviews from Ilya Bindeman, Karsten Haase, Julie Donnelly-Nolan, Tom Sisson, Wendy Bohrson, and an anonymous reviewer.
*Corresponding author. E-mail: tfeeley{at}montana.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL AND PETROLOGICAL...SAMPLE COLLECTION AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Annen C, Blundy JD, Sparks R. SJ. The genesis of intermediate and silicic magmas in deep crustal hot zones. Journal of Petrology (2006) 47:505–539.
Baker JA, Macpherson CG, Menzies MA, Thirlwall MF, Al-Kadasi M, Mattey DP. Resolving crustal and mantle contributions to continental flood volcanism, Yemen; constraints from mineral oxygen isotope data. Journal of Petrology (2000) 41:1805–1820.
Bacon CR. Magmatic inclusions in silicic and intermediate volcanic rocks. Journal of Geophysical Research (1986) 91:6091–6112.
Bacon CR, Adami LH, Lanphere MA. Direct evidence for the origin of low-18O silicic magmas; quenched samples of a magma chamber's partially-fused granitoid walls, Crater Lake, Oregon. Earth and Planetary Science Letters (1989) 96:199–208.[CrossRef][Web of Science]
Bacon CR, Gunn SH, Lanphere MA, Wooden JL. Multiple isotopic components in Quaternary volcanic rocks of the Cascade arc near Crater Lake, Oregon. Journal of Petrology (1994) 35:1521–1556.
Bacon CR, Bruggman PE, Christiansen RL, Clynne MA, Donnelly-Nolan JM, Hildreth W. Primitive magmas at five Cascade volcanic fields: melts from hot, heterogeneous, sub-arc mantle. Canadian Mineralogist (1997) 35:397–423.[Web of Science]
Barnes CG, Allen CM, Hoover JD, & Brigham R. Magmatic components of a tilted plutonic system, Klamath Mountains, California. In. The Nature and Origin of Cordillera Magmatism. Geological Society of America Memoirs—Anderson JL, ed. (1990) 174:331–346.
Barnes CG, Petersen SW, Kistler RW, Prestvik T, Sundvoll B. Tectonic implications of isotopic variation among Jurassic and Early Cretaceous plutons, Klamath Mountains. Geological Society of America Bulletin (1992) 104:117–126.
Barragan R, Geist D, Minard H, Larson P, Kurz M. Subduction controls on the compositions of lavas from the Ecuadorian Andes. Earth and Planetary Science Letters (1998) 154:153–166.[CrossRef][Web of Science]
Beard JS, Lofgren GE. Dehydration melting and water-saturated melting of basaltic and andesitic greenstones and amphibolites at 1, 3, and 6.9 kbar. Journal of Petrology (1991) 32:365–402.
Benz HM, Zandt G, Oppenheimer DH. Lithospheric structure of northern California from teleseismic images of the upper mantle. Journal of Geophysical Research (1992) 97:4791–4807.[CrossRef]
Bergantz GW. Changing techniques and paradigms for the evaluation of magmatic processes. Journal of Geophysical Research (1995) 100:17603–17613.[CrossRef]
Bindeman IN, Fournelle JH, Valley JW. Low-18O tephra from a compositionally zoned magma body; Fisher Caldera, Unimak Island, Aleutians. Journal of Volcanology and Geothermal Research (2001) 111:35–53.[CrossRef][Web of Science]
Bindeman IN, Ponomareva VV, Bailey JC, Valley JW. Volcanic arc of Kamchatka: a province with high-18O magma sources and large scale 18O/16O depletion of the upper crust. Geochimica et Cosmochimica Acta (2004) 68:841–865.[CrossRef][Web of Science]
Bindeman IN, Eiler JM, Yogodzinski GM, Tatsumi Y, Stern CR, Grove TL, Portnyagin M, Hoernle K, Danyushevsky LV. Oxygen isotope evidence for slab melting in modern and ancient subduction zones. Earth and Planetary Science Letters (2005) 235:480–496.[CrossRef][Web of Science]
Blakely RJ, Jachens RC. Volcanism, isostatic residual gravity, and regional tectonic setting of the Cascade volcanic province. Journal of Geophysical Research (1990) 95:19439–19451.[CrossRef]
Blakely RJ, Jachens RC, Simpson RW, Couch RW. Tectonic setting of the southern Cascade Range as interpreted from its magnetic and gravity fields. Geological Society of America Bulletin (1985) 96:43–48.
Blattner P, Reid F. The origin of lavas and ignimbrites of the Taupo volcanic zone, New Zealand, in the light of oxygen isotope data. Geochimica et Cosmochimica Acta (1982) 46:1417–1430.[CrossRef][Web of Science]
Borg LE, Clynne MA. The petrogenesis of felsic calc-alkaline magmas from the southernmost Cascades, California: origin by partial melting of basaltic lower crust. Journal of Petrology (1998) 39:1197–1222.[CrossRef][Web of Science]
Borg LE, Clynne MA, Bullen TD. The variable role of slab-derived fluids in the generation of a suite of primitive calc-alkaline lavas from the southernmost Cascades, California. Canadian Mineralogist (1997) 35:425–452.[Web of Science]
Boroughs S, Wolff JA, Bonnichsen B, Godchaux M, Larson PB. Large-volume, low-18O rhyolites of the central Snake River plain, Idaho, USA. Geology (2005) 33:821–824.
Bottinga Y, Javoy M. Oxygen isotope partitioning among minerals in igneous and metamorphic rocks. Reviews in Geophysics and Space Physics (1975) 13:401–418.[CrossRef]
Bullen TD, Clynne MA. Trace element and isotopic constraints on magmatic evolution at Lassen volcanic center. Journal of Geophysical Research (1990) 95:19671–19691.[CrossRef]
Carlson K, Wheeler G. Renaming ‘quartz basalts’ of Cinder Cone, Lassen Volcanic National Park. California Geology (1980) 33:101.
Chiba H, Chacko T, Clayton RN, Goldsmith JR. Oxygen isotope fractionations involving diopside, forsterite, magnetite and calcite; applications to geothermometry. Geochimica et Cosmochimica Acta (1989) 53:2985–2995.[CrossRef][Web of Science]
Christiansen RL, Clynne MA, Muffler L. JP. Geologic map of the Lassen Peak, Chaos Crags, and Upper Hat Creek area, California. US Geological Survey Geologic Investigations Series (2002) I-2723.
Clayton RN, Kieffer SW. Oxygen isotopic thermometer calibrations. In. Stable Isotope Geochemistry; a Tribute to Samuel Epstein. Special Publication of the Geochemical Society—Taylor HP Jr, O’Neil JR, Kaplan IR, eds. (1991) 3:3–10.
Clynne MA. Stratigraphic, lithologic, and major element geochemical constraints on magmatic evolution at Lassen volcanic center, California. Journal of Geophysical Research (1990) 95:19651–19661.[CrossRef]
Clynne MA. A complex magma mixing origin for rocks erupted in 1915, Lassen Peak, California. Journal of Petrology (1999) 40:105–132.[CrossRef][Web of Science]
Clynne MA, Borg LE. Olivine and chromian spinel in primitive calc-alkaline and tholeiitic lavas from the southernmost Cascade Range, California: a reflection of relative fertility of the source. The Canadian Mineralogist (1997) 35:453–472.
Clynne MA, Muffler LJP. Geologic map of Lassen Volcanic National Park and vicinity, California. US Geological Survey Scientific Investigations Map (2008) I-2899. scale 1:50 000, 203 pp. pamphlet (in press).
Clynne MA, Champion DE, Trimble DA, Hendley JW, II & Stauffer P. How old is ‘Cinder Cone’?—Solving a mystery in Lassen Volcanic Park, California. US Geological Survey Fact Sheet (2000) 023-00:4. pp.
Clynne MA, Christiansen RL, Trimble DA, & McGeehin J. Radiocarbon dates from volcanic deposits of the Chaos Crags and Cinder Cone eruptive sequences and other deposits, Lassen Volcanic National Park and vicinity, California. US Geological Survey Open-File Report (2002) 02-290:21. pp.
Davidson JP, Harmon RS. Oxygen isotope constraints on the petrogenesis of volcanic arc magmas from Martinique, Lesser Antilles. Earth and Planetary Science Letters (1989) 95:255–270.[CrossRef][Web of Science]
Davidson JP, Ferguson KM, Colucci MT, Dungan MA. The origin and evolution of magmas from the San Pedro–Pellado Volcanic Complex, S. Chile: multicomponent sources and open system evolution. Contributions to Mineralogy and Petrology (1988) 100:429–445.[CrossRef][Web of Science]
Davidson JP, Harmon RS. The source of Central Andean magmas; some considerations. In:. Andean Magmatism and its Tectonic Setting. Geological Society of America, Special Papers—Harmon RS, Rapela CW, eds. (1991) 265:233–243.
DePaolo DJ. A neodymium and strontium isotopic study of the Mesozoic calc-alkaline granitic batholiths of the Sierra Nevada and Peninsular Ranges, California. Journal of Geophysical Research (1981) 86:10470–10488.
Donnelly-Nolan JM. Abrupt shift in 18O values at Medicine Lake Volcano (California, USA). Bulletin of Volcanology (1998) 59:529–536.[CrossRef][Web of Science]
Dufek J, Bergantz GW. Lower crustal magma genesis and preservation: a stochastic framework for the evaluation of basalt–crust interaction. Journal of Petrology (2005) 46:2167–2195.
Dungan MA, Wulff A, Thompson R. Eruptive stratigraphy of the Tatara–San Pedro complex, 36°S, Southern Volcanic Zone, Chilean Andes: reconstruction method and implications for magma evolution at long-lived arc volcanic centers. Journal of Petrology (2001) 42:555–626.
Eichelberger JC. Origin of andesite and dacite: Evidence of mixing at Glass Mountain in California and at other circum-Pacific volcanoes. Geological Society of America Bulletin (1975) 86:1381–1391.
Eichelberger JC. Andesitic volcanism and crustal evolution. Nature (1978) 275:21–27.[CrossRef]
Eichelberger JC, Izbekov PE, Browne BL. Bulk chemical trends at arc volcanoes are not liquid lines of descent. Lithos (2006) 87:135–154.[CrossRef][Web of Science]
Eiler JM. Oxygen isotope variations of basaltic lavas and upper mantle rocks. In:. Stable Isotope Geochemistry. Mineralogical Society of America, Reviews in Mineralogy and Geochemistry—Valley JM, Cole DR, eds. (2001) 43:319–364.
Eiler JM, Crawford A, Elliott T, Farley KA, Valley JW, Stolper EM. Oxygen isotope geochemistry of oceanic-arc lavas. Journal of Petrology (2000) 41:229–256.
Feeley TC. Crustal modification during subduction-zone magmatism; evidence from the southern Salar de Uyuni region (20°–22°S), Central Andes. Geology (1993) 21:1019–1022.
Feeley TC, Dungan MA. Compositional and dynamic controls on mafic–silicic magma interactions at continental arc volcanoes; evidence from Cordon El Guadal, Tatara–San Pedro Complex, Chile. Journal of Petrology (1996) 37:1547–1577.
Feeley TC, Grunder AL. Mantle contribution to the evolution of Middle Tertiary silicic magmatism during early stages of extension: the Egan Range volcanic complex, east–central Nevada. Contributions to Mineralogy and Petrology (1991) 106:3154–169.
Feeley TC, Sharp ZD. 18O/16O isotope geochemistry of silicic lava flows erupted from Volcan Ollagüe, Andean Central Volcanic Zone. Earth and Planetary Science Letters (1995) 133:239–254.[CrossRef][Web of Science]
Feeley TC, Dungan MA, Frey FA. Geochemical constraints on the origin of mafic and silicic magmas at Cordon El Guadal, Tatara–San Pedro complex, central Chile. Contributions to Mineralogy and Petrology (1998) 131:393–411.[CrossRef][Web of Science]
Finch RH, Anderson CA. The quartz basalt eruptions of Cinder Cone, Lassen Volcanic National Park, California. University of California Publications in Geological Sciences (1930) 19:245–273.
Gill J. Orogenic Andesites and Plate Tectonics (1981) Berlin: Springer. 330.
Grove TL. Origin of magmas. In: Encyclopedia of Volcanoes—Sigurdsson H, Houghton BF, McNutt SR, Rymer H, Stix J, Ballard RD, eds. (2000) New York: Academic Press. 331–343.
Grunder AL. Low 18O silicic volcanic rocks at the Calabozos Caldera Complex, Southern Andes; evidence for upper-crustal contamination. Contributions to Mineralogy and Petrology (1987) 95:71–81.[CrossRef][Web of Science]
Grunder AL, Wickham SM. Homogenization and lowering of 18O/16O in mid-crustal rocks during extension-related magmatism in eastern Nevada. Earth and Planetary Science Letters (1991) 107:416–431.[CrossRef][Web of Science]
Guffanti M, Clynne MA, Smith JG, Muffler L. JP, Bullen TD. Late Cenozoic volcanism, subduction, and extension in the Lassen region of California, southern Cascade Range. Journal of Geophysical Research (1990) 95:19453–19464.[CrossRef]
Guffanti M, Clynne MA, Muffler L. JP. Thermal and mass implications of magmatic evolution in the Lassen volcanic region, California, and minimum constraints on basalt influx to the lower crust. Journal of Geophysical Research (1996) 101:3003–3013.[CrossRef]
Hart GL, Johnson CM, Shirey SB, Clynne MA. Osmium isotope constratints on lower crustal recycling and pluton preservation at Lassen Volcanic Center, CA. Earth and Planetary Science Letters (2002) 199:269–285.[CrossRef][Web of Science]
Heiken G, Eichelberger JC. Eruptions at Chaos Crags, Lassen Volcanic National Park, California. Journal of Volcanology and Geothermal Research (1980) 7:443–481.[CrossRef][Web of Science]
John DA, Breit GN, Lee RG, Dilles JH, Muffler LJP, & Clynne M. Fossil magmatic–hydrothermal systems in Pleistocene Brokeoff volcano, Lassen Volcanic National Park, California. EOS Transactions, American Geophysical Union (2006) 87. Fall Meeting Supplement, Abstract V53A-1745.
Johnson DM, Hooper PR, Conrey RM. XRF analysis of rocks and minerals for major and trace elements on a single low dilution Li-tetraborate fused bead. Advances in XRF Analysis (1999) 41:843–867.
Kistler RW, Peterman ZE. Variations in Sr, Rb, K, Na, and initial 87Sr/86Sr in Mesozoic granitic rocks and intruded wall rocks in central California. Geological Society of America Bulletin (1973) 84:3489–3512.
Lackey JS, Valley JW, Saleeby JB. Supracrustal input to magmas in the deep crust of Sierra Nevada batholith: Evidence from high-18O zircon. Earth and Planetary Science Letters (2005) 235:315–330.[CrossRef][Web of Science]
Lanphere MA, Champion DE, Clynne MA, Lowenstern JB, Sarna-Wojcicki AM, Wooden JL. Age of the Rockland tephra, western USA. Quaternary Research (2004) 62:94–104.
Le Bas MJ, Le Maitre RW, Streckeisen A, Zanettin B. A chemical classification of volcanic rocks based on the total alkali–silica diagram. Journal of Petrology (1986) 27:745–750.
Lucassen F, Dramer W, Bartsch, Wilke H.-G, Romer RI, Dulski P. Nd, Pb, and Sr isotope composition of juvenile magmatism in the Mesozoic large magmatic province of northern Chile (18–27°S): indications for a uniform subarc mantle. Contributions to Mineralogy and Petrology (2006) 152:571–589.[CrossRef][Web of Science]
Macdonald R, Sparks R. SJ, Sigurdsson H, Mattey DP, McGarvie DW, Smith RL. The 1875 eruption of Askja Volcano, Iceland; combined fractional crystallization and selective contamination in the generation of rhyolitic magma. Mineralogical Magazine (1987) 51:183–202.[Web of Science]
Matsuhisa Y, Matsubaya O, Sakai H. Oxygen isotope variations in magmatic differentiation processes of the volcanic rocks in Japan. Contributions to Mineralogy and Petrology (1973) 39:277–288.[CrossRef][Web of Science]
McBirney AR. Petrochemistry of the Cascade andesite volcanoes. Oregon Department of Geology and Mineralogy Industry Bulletin (1968) 62:101–107.
Mooney WD, Weaver CS. Regional crustal structure and tectonics of the Pacific coastal states: California, Oregon, and Washington. In: Geophysical framework of the continental United States. Geological Society of America, Memoirs—Pakiser LC, Mooney WD, eds. (1989) 172:129–162.
Muehlenbachs K, Clayton RN. Oxygen isotope studies of fresh and weathered submarine basalts. Canadian Journal of Earth Sciences (1972) 9:172–184.
Muelenbachs K, Kushiro I. Oxygen isotope exchange and equilibria of silicates with CO2 or O2. Carnegie Institution of Washington Yearbook (1974) 73:232–236.
Muehlenbachs K, Anderson AT, Sigvaldsson GE. Low-18O basalts from Iceland. Geochimica et Cosmochimica Acta (1974) 38:577–588.[CrossRef][Web of Science]
O’Neil JR. Theoretical and experimental aspects of isotopic fractionation. In:. Stable Isotopes in High Temperature Geologic Processes. Mineralogical Society of America, Reviews in Mineralogy—Valley JW, Taylor HP Jr, O’Neil JR, eds. (1986) 16:1–40.
Pineau F, Semet MP, Grassioneau N, Okrugin VM, Javoy M. The genesis of the stable isotope (O, H) record in arc magmas: the Kamchatka's case. Chemical Geology (1999) 135:93–124.
Ratajeski K, Glazener AF, Miller BV. Geology and geochemistry of mafic to felsic plutonic rocks associated with the Cretaceous intrusive suite of Yosemite Valley, California. Geological Society of America Bulletin (2001) 113:1486–1502.
Ratajeski K, Sisson TW, Glazner AF. Experimental and geochemical evidence for derivation of the El Capitan Granite, California, by partial melting of hydrous gabbroic lower crust. Contributions to Mineralogy and Petrology (2005) 149:713–734.[CrossRef][Web of Science]
Rhodes JM. Geochemistry of the 1984 Mauna Loa eruption: implications for magma storage and supply. Journal of Geophysical Research (1988) 93:4453–4466.
Rose TP, Criss RE, Mughannam AJ, Clynne MA. Oxygen isotope evidence for hydrothermal alteration within a Quaternary stratovolcano, Lassen Volcanic National Park, California. Journal of Geophysical Research (1994) 99:21621–21633.[CrossRef]
Schnurr W. BW, Trumbull RB, Clavero J, Hahne K, Siebel W, Gardeweg M. Twenty-five million years of silicic volcanism in the southern central volcanic zone of the Andes: geochemistry and magma genesis of ignimbrites from 25 to 27° S, 67 to 72° W. Journal of Volcanology and Geothermal Research (2007) 166:17–46.[CrossRef][Web of Science]
Sharp ZD. A laser-based microanalytical method for the in situ determination of oxygen isotope ratios of silicates and oxides. Geochimica et Cosmochimica Acta (1990) 54:1353–1357.[CrossRef][Web of Science]
Sherrod DR, Smith JG. Quaternary extrusion rates of the Cascade Range, northwestern United States and southern British Columbia. Journal of Geophysical Research (1990) 95:19465–19474.[CrossRef]
Shukuno H, Tamura Y, Tani K, Change Q, Suzuki T, Fiske RS. Origin of silicic magmas and the compositional gap Sumisu submarine caldera, Izu–Bonin arc, Japan. Journal of Volcanology and Geothermal Research (2006) 156:187–216.[CrossRef][Web of Science]
Simon L, Lécuyer C. Continental recycling: The oxygen isotope point of view. Geochemistry, Geophysices, Geosystems (2005) 6. doi:10.1029/2005GC000958.
Singer BS, O’Neil JR, Brophy JR. Oxygen isotope constraints on the petrogenesis of Aleutian arc magmas. Geology (1992) 20:367–370.
Sisson TW, Grove TL, Coleman RG. Hornblende gabbro sill complex at Onion Valley, California, and a mixing origin for the Sierra Nevada Batholith. Contributions to Mineralogy and Petrology (1996) 126:81–108.[CrossRef][Web of Science]
Sparks R. SJ, Marshall LA. Thermal and mechanical constraints on mixing between mafic and silicic magmas. Journal of Volcanology and Geothermal Research (1986) 29:99–124.[CrossRef][Web of Science]
Spera FJ, Bohrson WA. Energy-constrained open-system magmatic processes I: general model and energy-constrained assimilation and fractional crystallization (EC-AFC) formulation. Journal of Petrology (2001) 42:999–1018.
Stanley WD, Mooney WD, Fuis GS. Deep crustal structure of the Cascade range and surrounding regions from seismic refraction and magnetotelluric data. Journal of Petrology (1990) 95:19419–19438.
Stern CR, Frey FA, Futa K, Zartman RE, Peng Z, Kyser TK. Trace element and Sr, Nd, Pb, and O isotopic composition of Pliocene and Quaternary alkali basalts of the Patagonian plateau lavas of southernmost South America. Contributions to Mineralogy and Petrology (1990) 104:294–308.[CrossRef][Web of Science]
Taggart J. E. Jr, Lindsey JR, Scott BA, Vivit DV, Bartel AJ, Steward KC. Analyses of geologic materials by wavelength-dispersive X-ray fluorescence spectrometry. US Geological Survey Bulletin (1987) 1170:E1–E19.
Tatsumi Y, Eggins S. Subduction Zone Magmatism (1995) Oxford: Blackwell. 211.
Taylor H. P. Jr. Oxygen isotope geochemistry of igneous rocks. Contributions to Mineralogy and Petrology (1968) 19:1–71.[CrossRef]
Taylor HP Jr, & Silver L. Oxygen isotope relationships in plutonic rocks of the Peninsular Ranges batholith, southern and Baja California. In:. Short Papers of 4th International Conference on Geochronology, Cosmochronology, Isotope Geology. United States Geological Survey Open-File Report—Zartman RE, ed. (1978) 78–701. 423–426.
Tepley F. J. III, Davidson JP, Clynne MA. Magmatic interactions as recorded in plagioclase phenocrysts of Chaos Crags, Lassen volcanic center, California. Journal of Petrology (1999) 40:787–806.[CrossRef][Web of Science]
Thomas N, Tait SR. The dimensions of magmatic inclusions as a constraint on the physical mechanism of mixing. Journal of Volcanology and Geothermal Research (1997) 75:167–178.[CrossRef][Web of Science]
Tormey DR, Frey FA, López-Escobar L. Geochemistry of the active Azufre–Planchon–Peteroa volcanic complex, Chile (35°15'S): evidence for multiple sources and processes in a Cordilleran arc magmatic system. Journal of Petrology (1995) 36:265–298.
Tsuchiyama A. Dissolution kinetics of plagioclase in the melt of the system diopside–albite–anorthite, and the origin of dusty plagioclase in andesites. Contributions to Mineralogy and Petrology (1985) 89:1–16.[CrossRef][Web of Science]
Underwood SJ, Feeley TC, Clynne MA. Oxygen isotope geochemistry of primitive calc-alkaline mafic lavas in the southernmost Cascades. Geological Society of America, Abstracts with Programs (2004) 36:359.
Valley JW, Kitchen N, Kohn MJ, Niendorf CR, Spicuzza MJ. UWG-2, a garnet standard for oxygen isotope ratio: strategies for high precision and accuracy with laser heating. Geochimica et Cosmochimica Acta (1995) 59:5223–5231.[CrossRef][Web of Science]
Valley JW, Kinny PD, Schulze DJ, Spicuzza MJ. Zircon megacrysts from kimberlite: oxygen isotope variability among mantle melts. Contributions to Mineralogy and Petrology (1998) 133:1–11.[CrossRef][Web of Science]
Wade JA, Plank T, Stern RJ, Tollstrup DL, Gill JB, O’Leary JC, Eiler JM, Moore RB, Woodhead JD, Trusdell F, Fischer TP, Hilton DR. The May 2003 eruption of Anatahan volcano, Mariana Islands: geochemical evolution of a silicic island-arc volcano. Journal of Volcanology and Geothermal Research (2005) 146:139–170.[CrossRef][Web of Science]
Walker JA, Carr MJ, Patino LC, Johnson CM, Feigenson MD, Ward RL. Abrupt change in magma generation processes across the Central American Arc in southeastern Guatemala; flux-dominated melting near the base of the wedge to decompression melting near the top of the wedge. Contributions to Mineralogy and Petrology (1995) 120:378–390.[CrossRef][Web of Science]
Wilcox RE. The idea of magma mixing: history of a struggle for acceptance. Journal of Geology (1999) 107:421–432.[CrossRef][Web of Science]
Williams H. The dacites of Lassen Peak and vicinity, California, and their basic inclusions. American Journal of Science (1931) 22:385–403.[Web of Science]
Williams H. Geology of Lassen Volcanic National Park. University of California Publications in Geological Sciences (1932) 21:195–385.
Wilson LF, Feeley TC, Underwood SJ. Spatial distribution and compositional variations of magmatic inclusions in the Mount Helen dome, Lassen Volcanic Center: implications for magma chamber dynamics. Geological Society of America, Abstracts with Programs (2005) 37:291.
Wilson M. Milestones in geology. In:. Memoir. Geological Society of London—Le Bas MJ, ed. (1995) 16:205–218.[CrossRef]
Wolff JA, Grandy JS, Larson PB. Interaction of mantle-derived magma with island crust? Trace element and oxygen isotope data from the Diego Hernandez Formation, Las Canadas, Tenerife. Journal of Volcanology and Geothermal Research (2000) 103:343–366.[CrossRef][Web of Science]
Wolff JA, Balsley SD, Gregory RT. Oxygen isotope disequilibrium between quartz and sanidine from the Bandelier Tuff, New Mexico, consistent with a short residence time of phenocrysts in rhyolitic magma. Journal of Volcanology and Geothermal Research (2002) 116:119–135.[CrossRef][Web of Science]
Yamamoto T. A rhyolite to dacite sequence of volcanism directly from the heated lower crust: Late Pleistocene to Holocene Numazawa volcano, NE Japan. Journal of Volcanology and Geothermal Research, (2007) doi:10.1016/j.jvolgeores.2007.05.011.
Zellmer GF, Hawkesworth CJ, Sparks R. SJ, Thomas LE, Harford CL, Brewer TS, Loughlin SC. Geochemical evolution of the Soufrière Hills Volcano, Montserrat, Lesser Antilles volcanic arc. Journal of Petrology (2003) 44:1349–1374.
Zhao Z.-F, Zheng Y.-F. Calculation of oxygen isotope fractionation in magmatic rocks. Chemical Geology (2003) 193:59–80.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. E. Otamendi, M. N. Ducea, A. M. Tibaldi, G. W. Bergantz, J. D. de la Rosa, and G. I. Vujovich Generation of Tonalitic and Dioritic Magmas by Coupled Partial Melting of Gabbroic and Metasedimentary Rocks within the Deep Crust of the Famatinian Magmatic Arc, Argentina J. Petrology, May 8, 2009; (2009) egp022v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||