Journal of Petrology | Volume 44 | Number 8 | Pages 1349-1374 | 2003
© Oxford University Press 2003
Geochemical Evolution of the Soufrière Hills Volcano, Montserrat, Lesser Antilles Volcanic Arc
1 DEPARTMENT OF EARTH SCIENCES, UNIVERSITY OF BRISTOL, WILLS MEMORIAL BUILDING, QUEENS ROAD, BRISTOL BS8 1RJ, UK
2 DEPARTMENT OF EARTH SCIENCES, THE OPEN UNIVERSITY, WALTON HALL, MILTON KEYNES MK7 6AA, UK
3 GEOLOGY DEPARTMENT, UNIVERSITY OF LEICESTER, UNIVERSITY ROAD, LEICESTER LE1 7RH, UK
4 BRITISH GEOLOGICAL SURVEY, WEST MAINS ROAD, EDINBURGH EH9 3LA, UK
* Corresponding author. Present address: Lamont–Doherty Earth Observatory, 61 Route 9W, Palisades, NY 10964, USA. Tel.: +1-845-365-8907. Fax: +44-845-365-8155. E-mail: gzellmer{at}ldeo.columbia.edu
RECEIVED MAY 28, 2002; ACCEPTED JANUARY 29, 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
The geochemical evolution of Montserrat provides an important background to understanding the current activity of this island arc volcano. Here we present major and trace element, and U-, Th- and O-isotope data for rocks generated in the last 300 kyr that provide constraints on the magmatic processes occurring beneath the volcano. Samples range from low- to medium-K calc-alkaline basalts to dacites. Three suites can be distinguished on the basis of major and trace element compositions: the South Soufrière Hills suite; the Soufrière Hills suite, including the lava from the current eruption; and the mafic inclusions. Magmatic differentiation of the magma that crystallized to form the mafic inclusions appears to have been governed by closed-system processes, modelled by fractional crystallization (F
0·32), whereas the mafic South Soufrière Hills suite evolved in an open system, modelled by continuous magma recharge into a crystallizing reservoir (F 0·7). The Soufrière Hills andesite compositions are attributed to crystal fractionation of the South Soufrière Hills magmas; however, matrix glass compositions fall on a different trend, consistent with partial melting before eruption. Whole-rock 
18O values range from 7·0 to 7·4
, and are, therefore, slightly enriched compared with primitive arc lavas. This might be due to magmatic fractionation, or the assimilation of up to 20% hydrothermally altered arc crust. Extremely low Nb/Th ratios and low (230Th/232Th) ratios compared with depleted mantle, and relatively high but constant 143Nd/144Nd ratios indicate that the magma source beneath Montserrat is enriched by small (
1·2%) amounts of sediment, which was added from the subducting slab probably as a partial melt. High U/Th ratios and large ion lithophile element abundances relative to local sediments suggest that fluid-mobile elements from the dehydrating slab were also added to the wedge, and that the fluid signature in the South Soufrière Hills samples is stronger than in the mafic inclusions. U–Th isotopes are close to secular equilibrium, suggesting that the transfer time of the fluid signature from source to surface is 350 ka. In conjunction with evidence for magma remobilization at Montserrat, much of this time may represent crustal residence, suggesting long time scales of deep-level differentiation relative to the inferred rapid crystallization at shallower levels. KEY WORDS: U-series isotopes; O isotopes; petrogenesis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
The volcanic island of Montserrat is situated in the northern part of the
750 km long Lesser Antilles island arc, which formed by westward subduction of the Atlantic oceanic plate beneath the Caribbean plate. Since the beginning of the current eruption of the Soufrière Hills volcano on Montserrat in 1995 (Young et al., 1998b) there has been much interest in the causes of the renewed activity and the petrogenetic processes operating beneath the volcano. However, recent petrological research (Barclay et al., 1998; Devine et al., 1998; Murphy et al., 1998, 2000; Couch et al., 2001; Harford & Sparks, 2001; Stewart & Fowler, 2001) has largely focused on the products of the current eruption, and few attempts have been made to obtain information on the petrogenetic processes related to the long-term geochemical evolution of the volcano (MacGregor, 1938; Nockolds & Allen, 1953; Rea, 1974; Baker, 1984).
A geological map of Montserrat is given in Fig. 1. Based on the stratigraphy and 39Ar/40Ar dating of Harford et al. (2002), rocks can be divided into six groups, depending on sample location and age. The Silver Hills (>1000 ka) represent the oldest deposits in the northern part of the island, and the Centre Hills (500 ka to 1000 ka) make up the central part of Montserrat. The Soufrière Hills (300 ka to 350 a) and the dome formed by the current eruption (since 1995; here referred to as ‘new lava’), form the highest elevation in the south of the island, reaching >1000 m at the new dome. The flanks of the Soufrière Hills volcano are composed of pyroclastic deposits, and its core is formed by five andesitic lava domes; namely, Gages Mountain, Chances Peak, Galways Mountain, Perches Mountain and the new lava that has overgrown the previous dome, which was known as Castle Peak. The South Soufrière Hills are 130 kyr old and form the southernmost part of the volcano.
|
Whereas eruptive products at Montserrat are dominantly andesitic in composition, the South Soufrière Hills are formed by basalts and basaltic andesites. Mafic lavas are also found as inclusions within the andesites. In this paper, new U-, Th-, and O-isotope data and high precision inductively coupled plasma mass spectrometry (ICP-MS) trace element data are used in conjunction with major and trace element analyses of almost 200 samples to elucidate the long-term petrogenetic processes occurring beneath Montserrat. Sufficient geochemical data are available for the more recent eruptive products to address the following key questions:
- What is the petrogenetic relationship between the various rock groups that make up the South Soufrière Hills–Soufrière Hills volcanic centres? Are the volumetrically dominant andesites produced by fractional crystallization from mafic parental magmas, or do they represent partial melts of previously generated arc crust? What are the relative roles of fractional crystallization and magma recharge during magma differentiation?
- What can be learned from trace element and isotopic constraints about the magma source region in this subduction-related setting? Is there any evidence for enrichment by contributions from subducted sediments and fluids from the dehydrating subducting slab?
- Using U-series isotope data, the ages of individual crystals, and whole-rock and mineral-glass chemical variations, can the time scales of magma transfer processes from source to surface be constrained?
|
PETROLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
The petrology of samples from Montserrat has previously been described in some detail (Rea, 1974; Baker, 1984; Devine et al., 1998; Murphy et al., 2000), and is summarized in the following section. The petrological features of the Silver Hills, Centre Hills and Soufrière Hills suites are broadly similar. The andesites and dacites are typically highly porphyritic, consisting of 30–55 wt % phenocrysts set in a microlite-rich groundmass. The andesite and dacite phenocryst phases are plagioclase, oxides, orthopyroxene, ± clinopyroxene, ± hornblende, ± quartz, ± olivine (rare). The basaltic andesites lack hornblende and quartz, but contain olivine and commonly clinopyroxene as the dominant pyroxene. The Soufrière Hills suite, including the new lava, comprises mainly hornblende–hypersthene-phyric lavas, although Gages dome is a two-pyroxene andesite. The groundmass mineralogy is similar, except that amphibole is absent. Most phenocrysts show disequilibrium textures, including dusty sieve-textured plagioclase, high-temperature overgrowth rims on pyroxenes and plagioclase feldspars, amphibole reaction textures, and pseudomorphs of amphibole replaced by coarse-grained pyroxene–feldspar–oxide reaction products.
The new lava commonly contains mafic inclusions (1% by volume), thought to represent hot material injected into the magma chamber and rapidly quenched in response to the temperature contrast (Murphy et al., 2000). Mafic inclusions are phenocryst (>100 µm) poor, and contain sparse phenocrysts of plagioclase only. Their groundmass consists of plagioclase, abundant amphibole, clinopyroxene, orthopyroxene, magnetite and glass. Very few resorbed xenocrysts of plagioclase, amphibole, orthopyroxene and quartz from the host andesites have been observed (Murphy et al., 2000). Mafic inclusions are also commonly found in older Soufrière Hills lavas.
The South Soufrière Hills basaltic lavas are porphyritic, and comprise an anhydrous mineral assemblage of plagioclase, clinopyroxene, ± orthopyroxene, ± olivine, ± magnetite (Rea, 1974). The groundmass consists of plagioclase, clinopyroxene and magnetite microlites, and glass. Rea (1974) described cognate xenoliths from the pyroclastic deposits of the South Soufrière Hills, some of which are ultramafic and consist of plagioclase, amphibole and minor magnetite. In addition to these phases, pyroxene and quartz are found in basic to intermediate xenoliths.
|
ANALYTICAL TECHNIQUES |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
X-ray fluorescence (XRF) analyses were performed at Leicester University using standard techniques described by Harvey et al. (1996). Analysis of international reference materials indicates that accuracy and precision are better than 0·5% for major elements and better than 3% for trace elements. ICP-MS sample preparation was performed at Bristol University. Sample powders were dissolved by acid digestion in Teflon beakers. Samples (0·05 g) were digested overnight at 90°C on a hotplate in 3 ml of 8N HNO3 and 1 ml of HF. Dried samples were then redissolved in 3 ml of 8N HNO3 and 3 ml of Milli-Q H2O, transferred to HDPE bottles, diluted by weight with de-ionized water to 2000 times the original powder weight, and sonicated for 30 min. Trace elements were measured using the Boston University VG PQ ExCell quadrupole ICP-MS system. Raw ICP-MS data were blank-subtracted, corrected for drift using an external drift correcting solution (analysed every five samples), corrected for dilution weight, and calibrated using internal laboratory standards. Means and standard deviations of four repeat analyses of an internal laboratory rock standard (MAS1722 from Masaya volcano, Nicaragua) are given in Table 1 and indicate precisions of better than 3% for most elements.
|
|
Oxygen isotope analysis was performed at the Scottish Universities Environmental Research Centre, East Kilbride. Separates of plagioclase, quartz, amphibole, orthopyroxene, magnetite and groundmass were hand picked from five andesite samples for oxygen isotope analysis. Three fine-grained mafic inclusion samples were also selected for bulk analysis. Oxygen was extracted from
1·5 mg of mineral separate using a laser fluorination system based on that of Sharp (1990), converted to CO2, and analysed on a VG PRISM III mass spectrometer. Generally, three or four splits of each mineral separate were analysed to assess sample heterogeneity. Oxygen isotope values are reported in the standard permil notation (18O) relative to Vienna Standard Mean Ocean Water (V-SMOW). The SES quartz internal standard gave values of 10·32 ± 0·44 (2
) for 16 repeats during the course of the analyses.
U-series isotope analyses were performed at the Open University. The dissolution and chemical separation procedures for U and Th were the same as those described by Turner et al. (1997b). U and Th concentrations were determined using a mixed 229Th–236U spike. Isotope ratios were determined by thermal ionization mass spectrometry on a high abundance sensitivity Finnigan MAT 262 system equipped with an RPQ-II energy filter (van Calsteren & Schwieters, 1995). Typical abundance sensitivity was 0·05 ppm at 1 atomic mass unit from the main peaks for 238U and 232Th, and so any tail contributions on 230Th and 234U were insignificant. The dark noise of the multiplier (<0·1 c.p.s.) was negligible relative to the typical 230Th beam intensities during analysis (>20 c.p.s.). Where sample quantities were small, (230Th/232Th) was measured on a Nu Instruments® plasma mass spectrometer, using an Aridus® micro-concentric, desolvating-nebulizer sample introduction system. The abundance sensitivity was <0·1 ppm at 1 atomic mass unit below the peak (Turner et al., 2001). A number of (230Th/232Th) measurements were duplicated using both techniques and were found to be within error of each other. During the period of this study, multiple determinations of the Open University internal laboratory standard Th solution, the Th‘U’ standard (van Calsteren & Schwieters, 1995), gave 230Th/232Th = (6·104 ± 0·097) x 10-6 (n = 175), yielding a 2 standard deviation of 1·6%, and five repeat determinations of the AThO standard gave (230Th/232Th) = 1·024 ± 0·011 (2) and (238U/232Th) = 0·939 ± 0·008 (2). Chemical preparation blanks were 50 pg for Th and 100 pg for U, and are negligible compared with the 100 ng of sample generally loaded. The decay constants used in the calculation of activities were those compiled by Goldstein et al. (1989).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Major and trace element data
Major and trace element data were compiled from new analyses and the literature (Baker, 1984; Davidson, 1987; Devine et al., 1998; Murphy et al., 1998, 2000; Devine, unpublished data, 2003), and selected data are given in Table 1. The complete dataset may be downloaded from the Journal of Petrology website at http://www.petrology.oupjournals.org/. Early major and trace element data (MacGregor, 1938; Rea, 1974) were not included as these were obtained using wet chemical methods that do not compare well with more recent XRF data (see Rea, 1970) and ICP-MS results. All major element oxides are presented as volatile-free analyses, normalized to 100 wt %, with total iron calculated as Fe2O3. Loss on ignition (LOI) data are included where available.
This study concentrates on the geochemistry of the more recent deposits (<300 ka); namely, the South Soufrière Hills and Soufrière Hills suites, and the new lava with its mafic inclusions. Samples from Montserrat are calc-alkaline, typical of destructive plate margin magmatism, and range from low- to medium-K basalts to dacites (Fig. 2). All samples have Mg-number <0·6, and with Ni generally <10 ppm and Cr generally <30 ppm even the least evolved samples have undergone extensive differentiation from more primitive magmas. The South Soufrière Hills are dominated by medium-K basalts and basaltic andesites, and are chemically different from the mafic inclusions, which tend to have higher Fe2O3T, MnO and Zr contents, and lower MgO, K2O, Sr, Sc and V contents at the same SiO2 or La contents (see Figs 2 and 3). Samples from the new lava lie within the field of Soufrière Hills rocks. They are andesitic and have relatively restricted major element compositions.
|
|
To assess the degree of geochemical variability introduced by variations in sample size and by XRF analytical errors, 12 splits of a large (25 cm in diameter) single sample of the new lava (MVO 1227) have been analysed individually. Standard deviations are <1% relative for SiO2 and Al2O3, <3% for most other major element oxides and Sr, and <5% for MgO and most trace elements. The observed range in composition is indicated by the size of the grey box in Fig. 2. The variations in the new lava are greater than can be accounted for by differences in sample size.
Rare earth element data
The rare earth element (REE) data from Montserrat are presented in Table 1, and in chondrite-normalized REE diagrams in Fig. 4. The Soufrière Hills and new lava samples have very similar REE patterns and the former are presented as a field (Fig. 4a). The light rare earth elements (LREE; La to Nd) are enriched relative to the middle rare earth elements (MREE; Sm to Ho) and heavy rare earth elements (HREE; Er to Lu). The MREE and HREE display distinctive trough-shaped patterns, which are thought to reflect the involvement of amphibole, either as a crystallizing phase or in the residue during melting (e.g. Tiepolo et al., 2000, and references therein). The more evolved mafic inclusions display similar trough-shaped patterns (Fig. 4b). In contrast, the less evolved mafic inclusions and the South Soufrière Hills (Fig. 4c) rocks have straighter MREE to HREE patterns, and they tend to have lower and more variable LREE abundances. In addition, many samples display a distinct convex-upward pattern in the LREE and MREE.
|
Oxygen isotope data
Oxygen isotope data, presented in Table 2 and Fig. 5, were obtained for four samples of the new lava, three samples of its mafic inclusions, and a hydrothermally altered lava sample with red to yellow groundmass coloration. The last sample (MVO 959) is an andesite collected from the 26 December 1997 debris avalanche deposit and originates from the old Soufrière Hills edifice. It shows the most marked isotopic variation, with splits of the same phase giving values differing by >2
. 18O values in this sample are generally significantly higher than in the other samples. In particular, the groundmass is up to 4 heavier than the average groundmass in the unaltered samples.
|
|
Phenocrysts from the new lava show less marked
18O heterogeneity and groundmass separates are more homogeneous, with 18O 7·4, indicating that these samples have not been affected significantly by post-eruption low-temperature hydrothermal interaction. Estimated whole-rock values also give very similar results of 18O 7·3, and the oxygen isotope values of the mafic inclusions cannot be distinguished from those of the andesite whole rocks. The variation in 18O values for several phases is greater than the analytical precision, represented by a 2 variation for standard SES quartz of 0·44, indicating 18O heterogeneity in these phases. Mafic inclusion whole-rock samples also show significant variation (1 0·7, Table 2), but this may be due to differences in the proportions of the various minerals in each whole-rock split.
U-series data
U-series isotope data are presented in Table 3 and plotted in an equiline diagram in Fig. 6a. Whole-rock samples from the Soufrière Hills, including the new lava and the mafic inclusions, form a cluster around the equiline at (238U/232Th) and (230Th/232Th) activity ratios of 0·9, and do not display any coherent trends. There are no significant differences in (238U/232Th) or (230Th/232Th) ratios of the Soufrière Hills suite, the new lava and the mafic inclusions. Their U–Th composition is similar to that of samples from Martinique, and contrasts with rocks from many other Lesser Antilles arc volcanoes that display variable U excesses, previously attributed to U addition to the mantle source by fluids from the dehydrating slab (e.g. Turner et al., 1996). One sample of the South Soufrière Hills suite was analysed; it also plots close to the equiline, but at a higher (238U/232Th) ratio than the Soufrière Hills samples.
|
|
|
The Soufrière Hills samples plot between estimates of the isotope composition of Lesser Antilles sediments and depleted mantle, which is inferred to have the U–Th composition of the MORB source of Stolper & Newman (1994). Along-arc variations in the Sr-, Nd- and Pb-isotopic composition of the subducting sediments have been documented previously (White et al., 1985; White & Dupré, 1986). The (238U/232Th) ratios of the subducting sediments also vary along the arc, with low (238U/232Th) in the south and increasingly higher (238U/232Th) ratios towards the north (see inset to Fig. 1). A representative sediment composition subducted beneath Montserrat has a (238U/232Th) ratio of
0·8 (see EN20-18, Fig. 6 and inset to Fig. 1), slightly lower than the analysed Soufrière Hills samples and significantly lower than the South Soufrière Hills samples. In addition, the subducted sediments have higher Nb/Th ratios than observed in samples from Montserrat (Fig. 6b). The implications of these data will be discussed below. |
MAGMATIC DIFFERENTIATION IN THE CRUST |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
With the exception of a brief period of basaltic volcanism in the South Soufrière Hills centre at about 130 ka, the predominant compositions erupted on Montserrat for
1 Myr are andesites (Fig. 7) that frequently host mafic inclusions. The following sections explore whether the andesites are fractionation products of parental basaltic magmas, or remelts of differentiated material in the crust. The mafic inclusions clearly have a key position in that they might represent the parental magmas from which the andesites were generated, or be samples of the magma that simply provided heat to the andesitic body, or some combination of the two. The differentiation trends observed in the individual sample suites are central to the discussion of these issues, and La, the most incompatible REE (see Fig. 3a), will be used as a differentiation index in the following sections. However, before differentiation trends are considered in detail, the role of crustal assimilation needs to be evaluated.
|
Upper-crustal assimilation
Harford & Sparks (2001) invoked remobilization of upper-crustal material (by influx of hot mafic magma) from observations of heavy hydrogen isotope compositions of amphibole. They interpreted their data as indicating assimilation into the new andesite of hydrothermally altered andesite of similar petrological characteristics. The importance of this process can also be evaluated using oxygen isotope data. The high
18O values of sample MVO 959 can be attributed to low-temperature hydrothermal interaction, which resulted in strong 18O enrichment. In the other samples of the new lava, phenocrysts show less marked 18O heterogeneity and groundmass separates are homogeneous in 18O (Fig. 5). The estimated whole-rock values for the non-altered andesites are around 7·3, and the mafic inclusions give similar values, with an average of 7·2. These values are higher than those typical of non-altered olivine-phyric oceanic island arc samples (5–6, Eiler et al., 2000). This may partly be due to fractional crystallization of mafic phases that generally have lower 18O values (Taylor & Sheppard, 1986), such as olivine and magnetite (see Fig. 5). However, heavy hydrogen isotope compositions of amphiboles from early samples of the new lava (Harford & Sparks, 2001) indicate assimilation of hydrothermally altered previous intrusions. The oxygen isotope data presented here can easily accommodate 10–20% assimilation of hydrothermally altered material with a 18O value of 10, similar to the groundmass of sample MVO 959 (see Fig. 5). However, the data suggest that assimilation of >30% of altered crust is unlikely. Assimilation of hydrothermally altered material may potentially affect fluid-mobile element concentrations. These include the large ion lithophile elements (LILE) and uranium, as U6+ is soluble in aqueous fluids (Brenan et al., 1995). Therefore, assimilation needs to be discussed when considering these elements in the following sections.
Differentiation of the mafic inclusions
In the mafic inclusions, there is some fractionation of LREE and MREE with increasing REE abundances, i.e. during differentiation (Fig. 4b). Sm increases with increasing La (Fig. 8a). In addition, other incompatible element ratios remain constant, including U/Th (Fig. 8b), indicating that there is little evidence for increasing involvement of assimilating crustal material during differentiation. Closed-system crystallization may account for the observed differentiation trend. The variations in La/Gd and Lu/Gd (Fig. 9) suggest that amphibole is a key phase in the differentiation of the magma that formed the mafic inclusions, as other phases have little effect on these ratios.
|
|
The fractional crystallization equation (see Cox et al., 1979) was solved for the bulk partition coefficient D of the fractionating assemblage. Relevant partition coefficients, and references, are summarized in Table 4. Thus,
 | (1) |
70% fractional crystallization results in partition coefficients that compare very well with partitioning data from the literature (Table 4). Ratios of partition coefficients are similar to those expected from fractional crystallization of 70% amphibole, required to account for the high HREE/MREE ratios, and 30% plagioclase, the only phenocryst phase observed in the mafic inclusions [using partition coefficients of Halliday et al. (1995) and LaTourrette et al. (1995)]. Experimental phase equilibria constraints (Moore & Carmichael, 1998) indicate that under water-saturated conditions and at pressures above 3 kbar, amphibole is on the liquidus and the crystallization of other mafic phases is depressed. Furthermore, using the most mafic inclusion, mass balance calculations suggest that 70% fractional crystallization of amphibole and plagioclase in the above proportions is consistent with the observed range in major element composition from basalts to basaltic andesites, given the compositional variability of those phases (Murphy et al., 2000). Crystallization of amphibole constrains temperatures to 1050°C, dissolved water contents to >5 wt %, and the depth of the fractionating magma to >7 km (Moore & Carmichael, 1998; Pichavant et al., 2002).
Differentiation of the South Soufrière Hills suite
Compared with the mafic inclusions, the South Soufrière Hills suite shows greater LREE/MREE, and lesser HREE/MREE fractionation (Fig. 4c). In detail, its early evolution is dominated by a moderate increase of Lu/Gd, followed by a strong increase in La/Gd at relatively constant Lu/Gd in the more evolved samples (Fig. 9). Relatively constant MREE concentrations are evident in Fig. 8a, where Sm remains roughly constant with increasing La. Although fractionation of LREE from MREE in the mafic inclusions was ascribed to the involvement of amphibole during magmatic differentiation, this explanation seems unlikely for the evolved South Soufrière Hills samples, which do not show any significant fractionation between MREE and HREE. Another feature of the South Soufrière Hills samples is the sharp drop in U/Th ratios with increasing differentiation (Fig. 8b). This range of ratios indicates that the evolution of the South Soufrière Hills suite is affected by open-system processes, such as magma recharge, mixing or assimilation.
The geochemical evolution of periodically replenished (i.e. open system), periodically tapped, continuously fractionating magma chambers has been modelled by O'Hara (1977) and O'Hara & Mathews (1981), and the case of continuous replenishment was described by DePaolo (1981). Solving his equation for CA, the ‘open-system component’, i.e. the composition of a liquid magma that continuously replenishes the fractionating magma chamber, or a wallrock assimilant, or a combination of both, yields
 | (2) |
 | (3) |
Ultramafic cumulate xenoliths composed of predominantly amphibole and plagioclase occur in some of the South Soufrière Hills deposits (Rea, 1974), and the bulk partition coefficients derived in the previous section (see Table 4) may therefore be applicable in this case also. Then, using one of the least evolved (MVO 1148) and the most evolved (MVO 136) South Soufrière Hills compositions for C0 and CL, respectively, CA is dependent on F and r only. Although a number of combinations of F and r will reproduce the observed differentiation trend, any solution requires the involvement of a geologically meaningful composition CA, and forward modelling indicates that few such solutions exist for r < 1. The values preferred here are F = 0·7 and r = 0·9, resulting in an open-system component with trace element ratios similar to those of the mafic inclusions, but with significantly lower trace element abundances. The primitive trace element composition of this open-system component (Table 4) suggests that it is dominated by an unevolved, replenishing magma, and that significant contributions from crustal assimilants are again unlikely.
Thus, in the preferred model the South Soufrière Hills data are consistent with up to 30% fractional crystallization of the observed cumulate xenolith phases (amphibole and plagioclase) in a magma reservoir continuously recharged by influx of mafic magma. The mafic magma was characterized by lower LILE abundances and U/Th ratios than the primitive South Soufrière Hills compositions. In contrast, the amphibole-free phenocryst assemblage of the South Soufrière Hills suite may indicate that the final depth of magma storage was shallow, and hence outside the amphibole stability field at the magmatic temperatures of the South Soufrière Hills magmas.
Petrogenesis of the Soufrière Hills andesite suite and the new lava
There is a significant amount of geochemical variability in the new lava, compared with the small sample-scale heterogeneities indicated by the grey box in Fig. 2. This variability may in part be due to differences in modal mineral proportions in the strongly crystalline andesite samples, which could, for example, be the result of small and variable degrees of crystal–liquid segregation, implying real heterogeneities within the erupted new lava on a larger scale. Modal mineral proportions estimated from point counting of 15 thin sections of samples from the new lava are indeed very variable for phenocrysts of plagioclase (25–40 wt %), amphibole (2–12 wt %), orthopyroxene (1–4·5 wt %), clinopyroxene (0–0·5 wt %) and oxides (1–4 wt %), and with 28–36 vol. % groundmass crystals. Although these estimates of the variability of crystal content may not be truly representative because of the large size of some phenocrysts, particularly amphibole, compared with the size of a thin section, variations in the proportion of titanomagnetite phenocrysts, for example, may be the reason for up to 0·4 wt % range in TiO2 at any given SiO2 content (Fig. 2b). However, the 7 wt % range in SiO2 observed in the Soufrière Hills and new lava samples is too large to be a result of variations in phenocryst content, and the coherent trends displayed by Fe2O3, MgO and CaO also suggest that sample compositions reflect magmatic processes.
The oxygen isotope data provide further evidence for compositional heterogeneity. A range of 18O values is observed for phenocrysts of a single phase even within individual samples (Fig. 5). In conjunction with results from studies of hydrogen isotope heterogeneity in amphibole (Harford & Sparks, 2001), this supports the notion that the crystals of the Soufrière Hills andesite experienced a variety of different magmatic conditions during their growth histories. Diffusion modelling of Sr distribution in plagioclase crystals (Zellmer et al., 2003) indicates that crystal residence times at magmatic temperatures were short (typically <350 years). This is consistent with the evidence for reheating of previously intruded material from reverse zoned and dusty sieve-textured plagioclase crystals, reverse zoned orthopyroxenes with high-temperature rims overgrowing low-temperature cores, and a variety of other mineral disequilibrium textures (Devine et al., 1998; Murphy et al., 1998, 2000; Couch et al., 2001). It has been proposed that the remobilization of older andesitic intrusions occurred in response to the influx of mafic magma (see Murphy et al., 2000) and that the new lava and older Soufrière Hills andesites were generated by reheating, large degrees of partial melting and amalgamation of previously intruded small andesitic magma bodies within the upper crust (see Zellmer et al., 2003). It is striking that the matrix glass compositions of the andesites plot on a very different trend compared with the whole-rock samples (Fig. 10). This highlights the fact that crystal–liquid equilibria during the magmatic differentiation that produced the range of whole-rock compositions were very different from those observed on the scale of a hand specimen. The glass might reflect partial melting of andesitic material as a result of reheating or during decompression, or residual glass in equilibrium with the observed phenocryst phases.
|
The petrogenetic history of the intruded andesitic material before its solidification is more difficult to constrain. Both the mafic inclusions and the South Soufrière Hills lavas could represent parental magmas to the Soufrière Hills andesites. However, the mafic inclusion-type differentiation products have LREE/MREE ratios significantly lower than those of the Soufrière Hills andesites (Fig. 4b), best seen on a plot of Sm vs La, where the most differentiated mafic inclusions have evolved to higher Sm concentrations than the Soufrière Hills andesites (Fig. 8a). The South Soufrière Hills differentiation products also have lower HREE/MREE than the Soufrière Hills suite (Fig. 4c). It is inferred that the South Soufrière Hills magmas did not crystallize as much amphibole as the magma that formed the mafic inclusions, and accordingly did not attain significantly elevated Lu/Gd ratios (Fig. 9). In contrast, the Soufrière Hills andesites display the highest La/Gd and Lu/Gd ratios of all Montserrat samples and have trough-shaped MREE–HREE patterns (Fig. 4), pointing to a significant involvement of amphibole during differentiation. The abundance of amphibole phenocrysts in Soufrière Hills andesites (see Murphy et al., 2000) is also consistent with amphibole being part of the crystallizing assemblage.
Thus, the Soufrière Hills andesites do not lie on a simple extrapolation of the within-suite trends for the mafic inclusions and the South Soufrière Hills samples. However, generation of the intruded andesitic material by differentiation from a composition similar to that of the most evolved South Soufrière Hills sample remains feasible, involving crystallization of amphibole to account for high La/Gd and Lu/Gd ratios (Fig. 9) and low V/La and Sc/La ratios (Fig. 3c and d). Grey dashed arrows in Figs 3b–d and 9a show the effect of 50% fractional crystallization of equal amounts of amphibole and plagioclase; typical Soufrière Hills andesite compositions are consistently obtained. Mass balance calculations using appropriate mineral compositions (Rea, 1974) suggest that fractionation from one of the more evolved South Soufrière Hills andesites (MVO 136) is also consistent with major element constraints, although the projection to low Fe and Ti contents in the Soufrière Hills andesites and the South Soufrière Hills xenolith petrography indicate that magnetite crystallization also occurs.
In summary, the Soufrière Hills andesites are inferred to have been produced in five stages, outlined in Table 5: (1) partial melting in the mantle wedge generated basaltic magmas; (2) fractional crystallization at depth, with amphibole and plagioclase as fractionating phases, produced the observed whole-rock trends; (3) intrusion of discrete magma batches to upper-crustal levels, 5 km as inferred from phase equilibria constraints (Barclay et al., 1998; Devine et al., 1998), resulted in crystallization and solidification owing to rapid cooling as suggested by the short residence times of the crystals at magmatic temperatures (Zellmer et al., 2003); (4) influx of mafic magma from depth, now sampled as mafic inclusions in the andesites; (5) reheating and partial melting initiated the current eruption, and it might have been responsible for the observed differences between the whole rock and matrix glass compositional trends. Significantly, the heat was supplied by mafic magma with a different differentiation history, now preserved as rounded, quench-textured mafic inclusions in the new lava (Murphy et al., 2000; Harford & Sparks, 2001).
|
|
CONTRIBUTIONS FROM MULTIPLE SOURCE COMPONENTS |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
As the compositions erupted at Montserrat are evolved (Mg-number <0·60) compared with primary melts from the mantle, the following arguments are based on isotope ratios and on trace element ratios that are minimally affected by shallow-level differentiation processes. Melt generation in island arcs requires contributions from the subducting slab, and the role of sediment and fluid addition to the source region beneath Montserrat is discussed below. Causes for the differences between the mafic inclusions and the South Soufrière Hills samples will be explored. The section concludes by considering the time scales of magma transfer from source to surface and the potential for remobilization of previously generated arc crust.
Constraints on sediment contributions to the source region
Sediments have previously been identified as one of the source components of arc magmas, both in the Lesser Antilles (White & Dupré, 1986) and in other island arcs (e.g. Armstrong, 1971; Hawkesworth, 1982; Pearce, 1982; Elliott et al., 1997; Turner et al., 1997a; Zellmer et al., 2000). Samples from Montserrat have (238U/232Th) ratios intermediate between depleted mantle (DM; see Table 4) and the subducting sediments (White et al., 1985) (Fig. 6a). Sediment addition will not only decrease the U/Th ratio of the source, but in addition will strongly affect both its Nb/Th and its 143Nd/144Nd ratios (e.g. Elliott et al., 1997). This is illustrated in Fig. 6b, where mixing of DM with small amounts of local sediments (EN20-18) is modelled. At typical degrees of melting (15–25% in the Lesser Antilles; Pearce & Parkinson, 1993), neither Nb/Th nor U/Th ratios change significantly, and mantle melting therefore has no implications on the conclusions drawn from this diagram. As evident from the inset, the restricted 143Nd/144Nd ratios observed during >1 Myr of eruptive history (Davidson, 1987) are slightly lower than DM and require a small amount of Nd addition from subducted sediments to the melting region. This would be less than 1·2% in the case of bulk sediment addition (see also White & Dupré, 1986), but considerably less in the case of addition of partial melts of the sediments.
The Nb/Th ratios of Montserrat samples appear to be lower than those of the local bulk sediment (Fig. 6b), although the Nb concentration used for the sediment is the relatively high value from the Northern Antilles average [taken from Plank & Langmuir (1998)]. Although the presence of amphibole during crustal differentiation may decrease Nb/Th ratios slightly, the REE patterns of the least evolved samples do not show evidence for significant amphibole involvement (Fig. 4b and c). Hence, a partial sedimentary melt with lower Nb/Th ratios than bulk sediment may have enriched the mantle wedge in incompatible trace elements before mantle melting. Both the Nb/Th and 143Nd/144Nd ratios of the least evolved mafic inclusions and the South Soufrière Hills samples are reproduced if a 1·5% partial melt of local sediment with a small amount of residual rutile forms a proportion of 0·1% of the mantle wedge before melting (see Fig. 6 caption for details). As these estimates are also dependent on the type and amount of mineral phases present in the sediment, they should be taken as indicative only. However, a robust constraint from the available data is the maximum permissible amount of sediment in the mantle (1·2%), and sediment melting will considerably decrease that maximum amount. Finally, the limited range of 143Nd/144Nd ratios and the consistently low Nb/Th ratios indicate that the amount of sediment contribution to the mantle source has remained relatively unchanged for >1 Myr.
Fluid addition from the subducting slab
Another feature of Fig. 6b is the slightly elevated U/Th ratios of the mafic inclusions and the considerably elevated U/Th ratios of the South Soufrière Hills samples with respect to mixtures of DM with small amounts of sediments. Samples are also generally enriched in the fluid-mobile LILE, particularly Ba, compared with the subducting sediments. Although assimilation of hydrothermally altered material may potentially affect the fluid-mobile element concentrations, it was shown above that there is no indication of increasing involvement of assimilating material during differentiation. An alternative explanation for the elevated U/Th ratios and LILE abundances is therefore required. The release of fluids from the dehydrating subducted slab into the overlying mantle wedge would result in an enrichment of the mantle source (e.g. Gill & Williams, 1990; McDermott & Hawkesworth, 1991; Elliott et al., 1997; Hawkesworth et al., 1997; Zellmer et al., 2000). Hence, the addition of slab-derived fluid rich in uranium to the source region of the Montserrat samples is likely to be the cause of their elevated U/Th ratios, as widely inferred for other island arcs. This interpretation requires a relatively greater fluid contribution to the source of the South Soufrière Hills compared with the mafic inclusions, as the South Soufrière Hills have the highest U/Th ratios of all analysed samples (Fig. 6b). Therefore, other significant geochemical differences in the geochemistry of these rock suites need to be checked for consistency with this interpretation.
The mafic inclusions have higher Fe2O3T contents, and lower K2O and Sr contents than the South Soufrière Hills suite at any given SiO2 content (see Fig. 2). As the Al2O3 contents of the mafic inclusions and the South Soufrière Hills samples follow the same trend (Fig. 2f), the lower Sr content of the mafic inclusions cannot be attributed to greater amounts of plagioclase crystallization during differentiation in the crust. Instead, this is likely to be a source effect, and as both Sr and K are fluid-mobile LILE, the enrichment of the South Soufrière Hills samples in these elements, like the enrichment in U, is consistent with a stronger signature from fluids from the dehydrating subducting slab. The evolved nature of the Montserrat magmas makes it difficult to distinguish between increased fluid flux or greater wedge depletion as a reason for the relative U and LILE enrichment of the South Soufrière Hills basalts compared with the mafic inclusion magmas. However, the overall similarity of the REE patterns of the least evolved magmas (Fig. 4b and c) would argue against greater wedge depletion during the generation of the South Soufrière Hills samples.
Finally, variable major element contents can be used to infer the relative depths of the source regions, their degree of depletion, the degree of melting, and to assess the effects of water during melting (Turner & Hawkesworth, 1995, and references therein). When extrapolated back to an MgO content of 8 wt %, lower Fe8, slightly lower Ti8 and Na8, and higher Si8 concentrations of the South Soufrière Hills samples compared with the mafic inclusions are all compatible with higher water contents in the South Soufrière Hills source, and therefore with larger amounts of slab-derived fluids.
Time scales of magma transfer
The elevated U/Th ratios of all Montserrat samples relative to mixtures of DM with local sediments are consistent with fluid addition from the subducting slab to the mantle source region. Thus, the range in (230Th/232Th) activities at constant 143Nd/144Nd ratios at Montserrat (inset to Fig. 6b) may then represent ingrowth of 230Th by decay of excess U with time. Returning to Fig. 6a, the whole-rock samples from Montserrat lie on or very close to the equiline, and this suggests that any initial U excesses have decayed back to secular equilibrium, indicating that 350 kyr have passed since preferential mobilization of U in the slab-derived fluids.
Many of the andesite samples erupted at Montserrat show evidence for reheating of previously intruded material by influx of undegassed lava from depth (Devine et al., 1998; Murphy et al., 1998, 2000; Young et al., 1998a, 1998b). Thus, differentiation of the initial magmas, solidification and remobilization all occurred since the U-rich fluids left the subducted crust, which is here constrained to be at least 350 kyr before eruption. Oxygen isotope constraints (see above) indicate that 10–20% hydrothermally altered crust is involved in the petrogenesis of the Soufrière Hills andesites and the mafic inclusions, and remobilization of unaltered arc crust at depth is consistent with the notion of lengthy crustal residence times.
The rates of differentiation at depth are difficult to assess. The matrix glass and the phenocrysts in the erupted andesite have formed in the final stage (see Zellmer et al., 2003), and so they are younger than the whole-rock differentiation trends. At Montserrat, differentiation has successfully been modelled by open- and closed-system crystallization processes, which from the thermal viewpoint are likely to be slower at depth than at shallow levels. The data presented in this contribution are thus consistent with long time scales of deep-level differentiation followed by rapid crystallization in the shallow crust, and finally by reheating, partial melting and amalgamation of separate intrusive bodies immediately before eruption.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
(1) At Montserrat, the andesites and the mafic inclusions have
18O values slightly higher than those of primitive arc magmas, suggesting that up to 20% assimilation of hydrothermally altered arc crust may potentially be involved in the petrogenesis of these volcanic rocks. However, differentiation modelling indicates that assimilation has not significantly affected the trace element trends.
(2) Differentiation of the magma that formed the mafic inclusions is dominated by closed-system processes and can be modelled by fractional crystallization (F 0·32) of amphibole (70%) and plagioclase (30%). In contrast, differentiation of the South Soufrière Hills suite is dominated by open-system processes, as indicated by variable incompatible trace element ratios. Their petrogenesis can be modelled by a combination of crystallization and magma recharge using DePaolo's (1981) model of a continuously replenished fractionating magma chamber, with F = 0·7 and r = 0·9. The replenishing magma has trace element ratios similar to that parental to the most mafic inclusions, but lower trace element abundances.
(3) The andesites of the Soufrière Hills suite, including the new lava, and the mafic inclusions have different incompatible trace element ratios. The Soufrière Hills andesites cannot be produced by differentiation of a magma similar to their mafic inclusions. However, their compositions can be modelled by crystallization of equal proportions of amphibole and plagioclase from compositions similar to the more evolved samples of the South Soufrière Hills suite.
(4) Between-sample variations in the new lava and older Soufrière Hills rocks may partly be the result of variable phenocryst proportions. However, within-sample O-isotope heterogeneities also suggest that these rocks represent a mixture of crystals with different growth histories, in line with recent results from Sr-diffusion studies in plagioclase phenocrysts (Zellmer et al., 2003). This implies reheating and amalgamation of a number of discrete intrusive bodies within the crust prior to eruption. Partial melting of previously intruded material is consistent with a matrix glass compositional trend different from that of the whole rocks.
(5) The trace element compositions of the least evolved samples indicate that the mantle source beneath Montserrat was enriched by addition of small amounts (1·2%) of local sediments derived from the subducting slab. The sediment component may have been added as a partial melt, in which case the volume of sediment involved would be less. Slab fluids are required to account for elevated U/Th ratios and elevated LILE abundances relative to the sediment-enriched mantle wedge. There is evidence for a greater slab fluid signature in the South Soufrière Hills suite compared with that in the younger magmas represented by mafic inclusions.
(6) All analysed samples are close to 238U–232Th secular equilibrium, suggesting that initial U excesses as a result of slab fluid contributions have decayed. This implies that the transfer time scale of the fluid signature through mantle wedge and overlying arc crust is 350 kyr. Much of that time may represent crustal residence, indicating that deep-level differentiation trends were produced over significantly longer time periods than the rapid shallow-level crystallization and reheating before eruption.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
For supplementary data, please refer to Journal of Petrology Online.
|
ACKNOWLEDGEMENTS |
|---|
We would like to thank the Montserrat Volcano Observatory for fieldwork support. The assistance of Pete Evans, Yiming Huang and Rob Hughes at the Open University U-series facility, and the comments of Jon Blundy and Simon Turner are gratefully acknowledged. Tony Fallick provided assistance during oxygen isotope analysis. Many thanks go to Terry Plank for organizing ICP-MS work at Boston University. We thank David Peate and Michel Pichavant for their constructive reviews, and Philip Leat for a more critical review, all of which greatly improved the manuscript, as did the firm editorial hand of Ray Macdonald. The authors acknowledge a grant (IP/633/0999) from NERC Services and Facilities for collaboration with the Open University Uranium Series Facility. G.F.Z. was funded by a Leverhulme Research Grant (F/182/BI), R.S.J.S. by an NERC fellowship, and C.L.H. by an NERC studentship.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGYANALYTICAL TECHNIQUESRESULTSMAGMATIC DIFFERENTIATION IN THE...CONTRIBUTIONS FROM MULTIPLE...CONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Armstrong, R. L. (1971). Isotopic and chemical constraints on models of magma genesis in volcanic arcs. Earth and Planetary Science Letters 12, 137–142.[CrossRef][Web of Science]
Baker, P. E. (1984). Geochemical evolution of St. Kitts and Montserrat, Lesser Antilles. Journal of the Geological Society, London 141, 401–411.
Barclay, J., Rutherford, M. J., Carroll, M. R., Murphy, M. D., Devine, J. D., Gardner, J. & Sparks, R. S. J. (1998). Experimental phase equilibria constraints on pre-eruptive storage conditions of the Soufrière Hills magma. Geophysical Research Letters 25, 3437–3440.[CrossRef][Web of Science]
Brenan, J. M., Shaw, H. F., Ryerson, F. J. & Phinney, D. L. (1995). Mineral–aqueous fluid partitioning of trace elements at 900°C and 2·0 GPa: constraints on the trace element chemistry of mantle and deep crustal fluids. Geochimica et Cosmochimica Acta 59, 3331–3350.[CrossRef][Web of Science]
Chabaux, F., Hémond, C. & Allègre, C. J. (1999). 238U–230Th–226Ra disequilibria in the Lesser Antilles arc: implications for mantle metasomatism. Chemical Geology 153, 171–185.[CrossRef][Web of Science]
Couch, S., Sparks, R. S. J. & Carroll, M. R. (2001). Mineral disequilibrium in lavas explained by convective self-mixing in open magma chambers. Nature 411, 1037–1039.[CrossRef][Medline]
Cox, K. G., Bell, J. D. & Pankhurst, R. J. (1979). The Interpretation of Igneous Rocks. London: Chapman and Hall.
Davidson, J. P. (1985). Mechanisms of contamination in Lesser Antilles island arc magmas from radiogenic and oxygen isotope relationships. Earth and Planetary Science Letters 72, 163–174.[CrossRef][Web of Science]
Davidson, J. P. (1987). Crustal contamination versus subduction zone enrichment: examples from the Lesser Antilles and implications for mantle source compositions of island arc volcanic rocks. Geochimica et Cosmochimica Acta 51, 2185–2198.[CrossRef][Web of Science]
DePaolo, D. (1981). Trace element and isotopic effects of combined wallrock assimilation and fractional crystallisation. Earth and Planetary Science Letters 53, 189–202.[CrossRef][Web of Science]
Devine, J. D., Murphy, M. D., Rutherford, M. J., Barclay, J., Sparks, R. S. J., Carroll, M. R., Young, S. R. & Gardner, J. E. (1998). Petrologic evidence for pre-eruptive pressure–temperature conditions, and recent reheating, of andesitic magma erupting at the Soufriere Hills Volcano, Montserrat, W. I. Geophysical Research Letters 25, 3669–3672.[CrossRef][Web of Science]
Eiler, J. M., Crawford, A., Elliott, T., Farley, K. A., Valley, J. W. & Stolper, E. M. (2000). Oxygen isotope geochemistry of oceanic-arc lavas. Journal of Petrology 41, 229–256.
Elliott, T., Plank, T., Zindler, A., White, W. & Bourdon, B. (1997). Element transport from slab to volcanic front at the Mariana arc. Journal of Geophysical Research 102, 14991–15019.[CrossRef]
Ewart, A. & Griffin, W. L. (1994). Application of proton-microprobe data to trace-element partitioning in volcanic rocks. Chemical Geology 117, 251–284.[CrossRef][Web of Science]
Forsythe, L. M., Nielsen, R. L. & Fisk, M. R. (1994). High-field-strength element partitioning between pyroxene and basaltic to dacitic magmas. Chemical Geology 117, 107–126.[CrossRef][Web of Science]
Gill, J. B. & Williams, R. W. (1990). Th isotope and U-series studies of subduction-related volcanic rocks. Geochimica et Cosmochimica Acta 54, 1427–1442.[CrossRef][Web of Science]
Goldstein, S. J., Murrel, M. T. & Janecky, D. R. (1989). Th and U isotopic systematics of basalts from the Juan de Fuca and Gorda Ridges by mass spectrometry. Earth and Planetary Science Letters 96, 134–146.[CrossRef][Web of Science]
Hack, P. J., Nielsen, R. L. & Johnston, A. D. (1994). Experimentally determined rare-earth element and Y partitioning behaviour between clinopyroxene and basaltic liquids at pressures up to 20 kbar. Chemical Geology 117, 89–105.[CrossRef][Web of Science]
Halliday, A. N., Lee, D. C., Tommasini, S., Davies, G. R., Paslick, C. R., Fitton, J. G. & James, D. E. (1995). Incompatible trace elements in OIB and MORB and source enrichment in the sub-oceanic mantle. Earth and Planetary Science Letters 133, 379–395.[CrossRef][Web of Science]
Harford, C. (2001). The volcanic evolution of Montserrat. Ph.D. thesis, University of Bristol, 195 pp.
Harford, C. L. & Sparks, R. S. J. (2001). Recent remobilization of shallow-level intrusions on Montserrat revealed by hydrogen isotope composition of amphiboles. Earth and Planetary Science Letters 185, 285–297.[CrossRef][Web of Science]
Harford, C. L., Pringle, M. S., Sparks, R. S. J. & Young, S. R. (2002). The volcanic evolution of Montserrat using 40Ar/39Ar geochronology. In: Druitt, T. H. & Kokelaar, B. P. (eds) The Eruption of the Soufriere Hills Volcano, Montserrat, from 1995 to 1999. Geological Society, London, Memoir 21, pp. 93–113.
Hart, S. R. & Dunn, T. (1993). Experimental cpx/melt partitioning of 24 trace elements. Contributions to Mineralogy and Petrology 113, 1–8.[CrossRef][Web of Science]
Harvey, P. K., Lovell, M. A., Brewer, T. S., Locke, J. & Mansley, E. (1996). Measurement of thermal neutron absorption cross section in selected geochemical reference materials. Geostandards Newsletter 20, 79–85.[Web of Science]
Hauri, E. H., Wagner, T. P. & Grove, T. L. (1994). Experimental and natural partitioning of Th, U, Pb and other trace elements between garnet, clinopyroxene and basaltic melts. Chemical Geology 117, 149–166.[CrossRef][Web of Science]
Hawkesworth, C. J. (1982). Isotope characteristics of magmas erupted along destructive plate margins. In: Thorpe, R. S. (ed.) Andesites: Orogenic Andesites and Related Rocks. New York: John Wiley, pp. 549–571.
Hawkesworth, C., Turner, S., Peate, D., McDermott, F. & van Calsteren, P. (1997). Elemental U and Th variations in island arc rocks: implications for U-series isotopes. Chemical Geology 139, 207–221.[CrossRef][Web of Science]
Heath, E., Turner, S., Macdonald, R., Hawkesworth, C. J. & van Calsteren, P. (1998). Long magma residence times at an island arc volcano (Soufriere, St. Vincent) in the Lesser Antilles: evidence from 238U–230Th isochron dating. Earth and Planetary Science Letters 160, 49–63.[CrossRef][Web of Science]
Higuchi, H. & Nagasawa, H. (1969). Partition of trace elements between rock-forming minerals and the host volcanic rocks. Earth and Planetary Science Letters 7, 281–287.[CrossRef][Web of Science]
Jenner, G. A., Foley, S. F., Jackson, S. E., Green, T. H., Fryer, B. J. & Longerich, H. P. (1994). Determination of partition-coefficients for trace elements in high pressure temperature experimental run products by laser-ablation microprobe inductively-coupled plasma-mass spectrometry (LAM-ICP-MS). Geochimica et Cosmochimica Acta 57, 5099–5103.
Johnson, K. T. M. (1994). Experimental cpx/ and garnet/ melt partitioning of REE and other trace elements at high pressures; petrogenetic implications. Mineralogical Magazine 58A, 454–455.
LaTourrette, T., Hervig, R. L. & Holloway, J. R. (1995). Trace element partitioning between amphibole, phlogopite, and basanite melt. Earth and Planetary Science Letters 135, 13–30.[CrossRef][Web of Science]
Luhr, J. F. & Carmichael, I. S. E. (1980). The Colima Volcanic Complex, Mexico I. Post-caldera andesites from Volcan Colima. Contributions to Mineralogy and Petrology 71, 343–372.[CrossRef][Web of Science]
Lundstrom, C. C., Shaw, H. F., Ryerson, F. J., Phinney, D. L., Gill, J. B. & Williams, Q. (1994). Compositional controls on the partitioning of U, Th, Ba, Pb, Sr and Zr between clinopyroxene and haplobasaltic melts; implications for uranium series disequilibria in basalts. Earth and Planetary Science Letters 128, 407–423.[CrossRef][Web of Science]
MacGregor, A. G. (1938). The Royal Society Expedition to Montserrat, B. W. I. The volcanic history and petrology of Montserrat with observations on Mt Pelée, in Martinique. Philosophical Transactions of the Royal Society of London, Series B 229, 1–90.
Matsui, Y., Onuma, N., Nagasawa, H., Higuchi, H. & Banno, S. (1977). Crystal structure control in trace element partition between crystal and magma. Bulletin de la Société Française de Minéralogie et de Cristallographie 100, 324–328.
McDermott, F. & Hawkesworth, C. J. (1991). Th, Pb, and Sr isotope variations in young island arc volcanics and oceanic sediments. Earth and Planetary Science Letters 104, 1–15.[CrossRef][Web of Science]
McKay, G., Le, L., Wagstaff, J. & Crozaz, G. (1994). Experimental partitioning of rare earth elements and strontium; constraints on petrogenesis and redox conditions during crystallization of Antarctic angrite Lewis Cliff 86010. Geochimica et Cosmochimica Acta 58, 2911–2919.[CrossRef][Web of Science]
McKenzie, D. P. & O'Nions, R. K. (1991). Partial melt distributions from inversion of rare earth element concentrations. Journal of Petrology 32, 1021–1091.
Moore, G. & Carmichael, I. S. E. (1998). The hydrous phase equilibria (to 3 kbar) of an andesite and basaltic andesite from western Mexico: constraints on water content and conditions of phenocryst growth. Contributions to Mineralogy and Petrology 130, 304–319.[CrossRef][Web of Science]
Murphy, M. D., Sparks, R. S. J., Barclay, J., Carroll, M. R., Lejeune, A.-M., Brewer, T. S., Macdonald, R., Black, S. & Young, S. (1998). The role of magma mixing in triggering the current eruption at the Soufriere Hills volcano, Montserrat, West Indies. Geophysical Research Letters 25, 3433–3436.[CrossRef][Web of Science]
Murphy, M. D., Sparks, R. S. J., Barclay, J., Carroll, M. R. & Brewer, T. S. (2000). Remobilization of andesite magma by intrusion of mafic magma at the Soufriere Hills volcano, Montserrat, West Indies. Journal of Petrology 41, 21–42.
Nash, W. P. & Crecroft, H. R. (1985). Partition coefficients for trace elements in silicic magmas. Geochimica et Cosmochimica Acta 49, 2309–2322.[CrossRef][Web of Science]
Nockolds, S. R. & Allen, R. (1953). The geochemistry of some igneous rock series. Geochimica et Cosmochimica Acta 4, 105–142.[CrossRef][Web of Science]
O'Hara, M. J. (1977). Geochemical evolution during fractional crystallisation of a periodically refilled magma chamber. Nature 266, 503–507.[CrossRef]
O'Hara, M. J. & Mathews, R. E. (1981). Geochemical evolution in an advancing, periodically replenished, periodically tapped, continuously fractionated magma chamber. Journal of the Geological Society, London 138, 237–277.
Pearce, J. A. (1982). Trace element characteristics of lavas from destructive plate boundaries. In: Thorpe, R. S. (ed.) Andesites: Orogenic Andesites and Related Rocks. New York: Wiley, pp. 230–249.
Pearce, J. A. & Parkinson, I. J. (1993). Trace element models for mantle melting: application to volcanic arc petrogenesis. In: Prichard, H. M., Alabaster, T., Harris, N. B. W. & Neary, C. R. (eds) Magmatic Processes and Plate Tectonics. Geological Society, London, Special Publications 76, 373–403.
Peccerillo, A. & Taylor, S. R. (1976). Geochemistry of Eocene calc-alkaline volcanic rocks from the Kastamonu area, northern Turkey. Contributions to Mineralogy and Petrology 58, 63–81.[CrossRef][Web of Science]
Perret, F. A. (1939). The volcano-seismic crisis at Montserrat, 1933–1937. Publications of the Carnegie Institution, Washington 512, 1–76.
Pichavant, M., Martel, C., Bourdier, J.-L. & Scaillet, B. (2002). Physical conditions, structure, and dynamics of a zoned magma chamber: Mount Pelée (Martinique, Lesser Antilles Arc). Journal of Geophysical Research 107(ECV1), 1–28.
Plank, T. & Langmuir, C. H. (1998). The geochemical composition of subducting sediments and its consequence for the crust and mantle. Chemical Geology 145, 325–394.[CrossRef][Web of Science]
Rea, W. J. (1970). The geology of Montserrat, British West Indies. Ph.D. thesis, University of Oxford, 196 pp.
Rea, W. J. (1974). The volcanic geology and petrology of Montserrat, West Indies. Journal of the Geological Society, London 130, 341–366.
Robertson, R. E. A., Aspinall, W. P., Herd, R. A., Norton, G. E., Sparks, R. S. J. & Young, S. R. (2000). The 1995–1998 eruption of the Soufriere Hills volcano, Montserrat, WI. Philosophical Transactions of the Royal Society of London, Series A 358, 1619– 1637.[CrossRef]
Roobol, M. J. & Smith, A. L. (1998). Pyroclastic stratigraphy of the Soufriere Hills volcano, Montserrat—implications for the present eruption. Geophysical Research Letters 25, 3393–3396.[CrossRef][Web of Science]
Sharp, Z. D. (1990). A laser-based microanalytical method for the in situ determination of oxygen isotope ratios of silicates and oxides. Geochimica et Cosmochimica Acta 54, 1353–1357.[CrossRef][Web of Science]
Shepherd, J. B., Tomblin, J. F. & Woo, D. A. (1971). Volcano-seismic crisis in Montserrat, West Indies, 1966–67. Bulletin of Volcanology 35, 143–163.
Skulski, T., Minarik, W. G. & Watson, E. B. (1994). High-pressure experimental trace-element partitioning between clinopyroxene and basaltic melts. Chemical Geology 117, 127–147.[CrossRef][Web of Science]
Stewart, M. L. & Fowler, A. D. (2001). The nature and occurrence of discrete zoning in plagioclase from recently erupted andesitic volcanic rocks, Montserrat. Journal of Volcanology and Geothermal Research 106, 243–253.[CrossRef][Web of Science]
Stolper, E. & Newman, S. (1994). The role of water in the petrogenesis of Mariana trough magmas. Earth and Planetary Science Letters 121, 293–325.[CrossRef][Web of Science]
Sun, S. S. & McDonough, W. F. (1989). Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: Saunders, A. D. & Norry, M. J. (eds) Magmatism in Ocean Basins. Geological Society, London, Special Publications 42, 313–345.
Taylor, H. P. & Sheppard, S. M. F. (1986). Igneous rocks; I, Processes of isotopic fractionation and isotope systematics. In: Valley, J. W., Taylor, H. P. and O'Neil, J. R. (eds) Stable Isotopes in High Temperature Geological Processes. Mineralogical Society of America, Reviews in Mineralogy 16, 227–271.
Tiepolo, M., Vannucci, R., Bottazzi, P., Oberti, R., Zanetti, A. & Foley, S. (2000). Partitioning of rare earth elements, Y, Th, U, and Pb between pargasite, kaersutite, and basanite to trachyte melts: implications for percolated and veined mantle. Geochemistry, Geophysics, Geosystems 1, Paper number 2000GC000064.
Turner, S. & Hawkesworth, C. (1995). The nature of the sub-continental mantle: contraints from the major-element composition of continental flood basalts. Chemical Geology 120, 295–314.[CrossRef][Web of Science]
Turner, S., Hawkesworth, C., van Calsteren, P., Heath, E., Macdonald, R. & Black, S. (1996). U-series isotopes and destructive plate margin magma genesis in the Lesser Antilles. Earth and Planetary Science Letters 142, 191–207.[CrossRef][Web of Science]
Turner, S., Hawkesworth, C., Rogers, N., Bartlett, J., Worthington, T., Hergt, J., Pearce, J. & Smith, I. (1997a). 238U–230Th disequilibria, magma petrogenesis, and flux rates beneath the depleted Tonga–Kermadec island arc. Geochimica et Cosmochimica Acta 61, 4855–4884.[CrossRef][Web of Science]
Turner, S., Hawkesworth, C., Rogers, N. & King, P. (1997b). U–Th isotope disequilibria and ocean island basalt generation in the Azores. Chemical Geology 139, 145–164.[CrossRef][Web of Science]
Turner, S. P., van Calsteren, P., Vigier, N. & Thomas, L. E. (2001). Determination of thorium and uranium isotope ratios in low-concentration geological materials using a fixed multi-collector-ICP-MS. Journal of Analytical Atomic Spectrometry 16, 612–615.[CrossRef]
van Calsteren, P. & Schwieters, J. B. (1995). Performance of a thermal ionisation mass spectrometer with a deceleration lens system and post-deceleration detector selection. International Journal of Mass Spectrometry and Ion Processes 146–147, 119–129.[CrossRef]
White, W. M. & Dupré, B. (1986). Sediment subduction and magma genesis in the Lesser Antilles—isotopic and trace-element constraints. Journal of Geophysical Research 91, 5927–5941.
White, W. M., Dupré, B. & Vidal, P. (1985). Isotope and trace element geochemistry of sediments from the Barbados Ridge–Demerara Plain region, Atlantic Ocean. Geochimica et Cosmochimica Acta 49, 1875–1886.[CrossRef][Web of Science]
Young, S. R., Francis, P. W., Barclay, J., Casadevall, T. J., Gardner, C. A., Darroux, B., Davies, M. A., Delmelle, P., Norton, G. E., Maciejewski, A. J. H., Oppenheimer, C. M. M., Stix, J. & Watson, I. M. (1998a). Monitoring SO2 emission at the Soufriere Hills volcano: implications for changes in eruptive conditions. Geophysical Research Letters 25, 3681–3684.
Young, S. R., Sparks, R. S. J., Aspinall, W. P., Lynch, L. L., Miller, A. D., Robertson, R. E. A. & Shepherd, J. B. (1998b). Overview of the eruption of Soufriere Hills volcano, Montserrat, 18 July 1995 to December 1997. Geophysical Research Letters 25, 3389–3392.[CrossRef][Web of Science]
Zellmer, G. F., Turner, S. P. & Hawkesworth, C. J. (2000). Timescales of destructive plate margin magmatism: new insights from Santorini, Aegean volcanic arc. Earth and Planetary Science Letters 174, 265–282.[CrossRef][Web of Science]
Zellmer, G. F., Sparks, R. S. J., Hawkesworth, C. J. & Wiedenbeck, M. (2003). Magma emplacement and remobilization timescales beneath Montserrat: insights from Sr and Ba zonation in plagioclase phenocrysts. Journal of Petrology 44, 1413–1431.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. C. Feeley, M. A. Clynne, G. S. Winer, and W. C. Grice Oxygen Isotope Geochemistry of the Lassen Volcanic Center, California: Resolving Crustal and Mantle Contributions to Continental Arc Magmatism J. Petrology, May 1, 2008; 49(5): 971 - 997. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Le Friant, E.J. Lock, M.B. Hart, G. Boudon, R.S.J. Sparks, M.J. Leng, C.W. Smart, J.C. Komorowski, C. Deplus, and J.K. Fisher Late Pleistocene tephrochronology of marine sediments adjacent to Montserrat, Lesser Antilles volcanic arc Journal of the Geological Society, January 1, 2008; 165(1): 279 - 289. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Huang and C. C. Lundstrom 231Pa excesses in arc volcanic rocks: Constraint on melting rates at convergent margins Geology, November 1, 2007; 35(11): 1007 - 1010. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. GARRIDO, J.-L. BODINIER, J.-P. BURG, G. ZEILINGER, S. S. HUSSAIN, H. DAWOOD, M. N. CHAUDHRY, and F. GERVILLA Petrogenesis of Mafic Garnet Granulite in the Lower Crust of the Kohistan Paleo-arc Complex (Northern Pakistan): Implications for Intra-crustal Differentiation of Island Arcs and Generation of Continental Crust J. Petrology, October 1, 2006; 47(10): 1873 - 1914. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. ANNEN, J. D. BLUNDY, and R. S. J. SPARKS The Genesis of Intermediate and Silicic Magmas in Deep Crustal Hot Zones J. Petrology, March 1, 2006; 47(3): 505 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. GURENKO, R. B. TRUMBULL, R. THOMAS, and J. M. LINDSAY A Melt Inclusion Record of Volatiles, Trace Elements and Li-B Isotope Variations in a Single Magma System from the Plat Pays Volcanic Complex, Dominica, Lesser Antilles J. Petrology, December 1, 2005; 46(12): 2495 - 2526. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. DUFEK and G. W. BERGANTZ Lower Crustal Magma Genesis and Preservation: a Stochastic Framework for the Evaluation of Basalt-Crust Interaction J. Petrology, November 1, 2005; 46(11): 2167 - 2195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. F. ZELLMER, R. S. J. SPARKS, C. J. HAWKESWORTH, and M. WIEDENBECK Magma Emplacement and Remobilization Timescales Beneath Montserrat: Insights from Sr and Ba Zonation in Plagioclase Phenocrysts J. Petrology, August 1, 2003; 44(8): 1413 - 1431. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||