Journal of Petrology Advance Access published online on January 4, 2008
Journal of Petrology, doi:10.1093/petrology/egm080
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Boninites and Adakites from the Northern Termination of the Tonga Trench: Implications for Adakite Petrogenesis
1Arc Centre of Excellence in Ore Deposits and School of Earth Sciences, University of Tasmania, Private Bag 79, Hobart, Tasmania 7001, Australia
2School of Earth Sciences, University of Melbourne, Melbourne, Victoria 3010, Australia
3Department of Geosciences, Oregon State University, Corvallis, OR 97331-5506, USA
Received March 15, 2007; Revised typescript accepted November 19, 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND GEOCHRONOLOGYANALYTICAL METHODSGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Adakitic rocks were recovered by dredging from the northern termination of the Tonga Trench during the 1984 voyage of the R.V. Natsushima and the 1996 voyage of the R.V. Melville. These contain magmatic zircons that have been dated at 2·5 Ma by U–Pb methods, indicating that they are contemporaneous with boninite magmatism previously described from this area. This is the first time adakites and high-Ca boninites have been reported from the same active tectonic setting. The Tonga adakites are classified as high-SiO2 adakites, and are compositionally consistent with an origin as partial melts of subducted Pacific oceanic crust and sediment. Geochemical modelling suggests that the adakites are not involved in the petrogenesis of the Tongan high-Ca boninites. However, the recovery of adakites and boninites from the termination of the northern Tonga Trench suggests that both magma types are related to the unique tectonic setting of this region, where a transition from steep subduction to a transform fault plate boundary has created a slab window with an associated slab edge. Boninites are generated as a result of hot Samoan plume mantle moving through the slab window and subsequently being fluxed by H2O-rich fluids from the subducting Pacific oceanic crust. The Tonga adakites, in contrast, result from the direct melting of the slab edge as a result of the juxtaposition of the subducting slab against hot mantle derived from the Samoan plume.
KEY WORDS: adakites; boninites; Tonga; slab melting; Samoa; slab edge
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND GEOCHRONOLOGYANALYTICAL METHODSGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The petrogenesis of subduction-related magmas is a long-standing problem in igneous petrology (Kuno, 1960
; Green et al., 1967; Green, 1976; Gill, 1981; Crawford et al., 1987; Tatsumi & Eggins, 1993). Many different magmatic rock suites have been documented from both active volcanoes and older exposed rock sequences associated with the subduction process. One of the key goals of research on subduction-related magmas is to produce a coherent geodynamic model of the subduction process that can explain this wide variety of magma compositions. To do this, we need to carefully describe and define the particular tectonic and structural setting, including both spatial and time constraints, of each subduction-related magmatic suite. Two particularly distinctive ‘end-member’ subduction-related magmatic suites are boninites and adakites. Both have been attributed to anomalously high temperatures in the mantle wedge and/or subducted slab, either during arc initiation or during hot subduction (Crawford et al., 1989; Defant & Drummond, 1990).
Boninites are an important ‘end-member’ of supra-subduction zone magmatic suites as they have the highest H2O contents and require the most depleted mantle wedge sources. They are characterized by relatively high SiO2 and H2O contents, and low TiO2 contents compared with tholeiitic suites (Crawford et al., 1989; Le Bas, 2000; SiO2 > 52 wt %, MgO >8 wt % and TiO2 < 0·5 wt %). They lack plagioclase in rocks more mafic than andesite, and contain very magnesian olivine phenocrysts (up to Fo94, Crawford et al., 1989). Two broad types of boninite magmas are recognized, the high-Ca and low-Ca boninite series (Crawford et al., 1989). Both boninite types are a rare but important component of many supra-subduction zone ophiolites and fore-arcs (Rogers et al., 1989; Bloomer et al., 1995; Falloon et al., 1997). As boninite petrogenesis requires a unique combination of high temperatures at shallow depths in the mantle wedge (Crawford et al., 1989), the presence of boninites has important implications for the tectonic setting and/or initiation of subduction zones (Crawford et al., 1989; Bloomer et al., 1995).
Adakite is another ‘end-member’ supra-subduction zone magma type, believed to be related to the melting of subducting oceanic crust (Defant & Drummond, 1990; Martin, 1999; Martin et al., 2005). The process of hot subduction is considered by many workers to be one of the key mechanisms for the formation of continental crust early in the Earth's history (Yogodzinski & Kelemen, 1998; Kelemen et al., 2003; Martin et al., 2005). The adakite magmatic suite includes andesites, dacites and sodic rhyolites, and is not usually associated with parental basaltic magmas (Defant & Drummond, 1990). The trace-element geochemistry of adakites is consistent with their derivation by partial melting of basalt transformed into garnet-bearing amphibolite or into eclogite within a warm subducted slab (Defant & Drummond, 1990; Martin, 1999; Rapp et al., 1999; Martin et al., 2005). Their strong depletion in heavy rare earth elements such as Yb, is, therefore, due to the presence of garnet as a residual phase during magma genesis.
Although boninites and adakites have been reported in close association from Precambrian (Polat & Kerrich, 2004) and Palaeozoic terranes (Niu et al., 2006) and from the Eocene Bonin fore-arc (Ishiwatari et al., 2006), no modern-day association has been reported. Boninites and adakites recovered from ancient terranes clearly support a subduction-related tectonic setting; however, as a result of post-emplacement tectonism and metamorphism it is difficult to determine a precise tectonic setting for these rocks within the general subduction geodynamic environment. One important uncertainty is whether boninites are formed only at the initiation of subduction zones (Bloomer et al., 1995), or alternatively, may also be generated during early arc rifting leading to formation of a back-arc basin (Crawford et al., 1981).
We report for the first time the occurrence of adakites and boninites from the northern termination of the Tonga Trench. Boninites from this area have been extensively studied and so far are the only known occurrence of boninites from an active arc setting (Falloon & Crawford, 1991; Sobolev & Danyushevsky, 1994; Danyushevsky et al., 1995; Falloon et al., 2007). Therefore, the recovery of adakites, indicative of slab melting, along with boninites from an active arc setting, could be helpful in understanding the tectonic setting of ancient terranes where both boninites and adakites have been observed. In this study we present the geochemistry of the Tonga adakites and discuss their potential role in boninite petrogenesis.
|
GEOLOGICAL SETTING |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND GEOCHRONOLOGYANALYTICAL METHODSGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The location of the study area and general tectonic features are shown schematically in Fig. 1. The Tonga intra-oceanic arc is recognized as a type example of an extension-dominated, non-accretionary convergent margin (Tappin, 1994
; Tappin et al., 1994; MacLeod, 1996), with active extensional tectonism throughout the fore-arc and landward trench slopes. The Tonga Trench is the site of westward subduction of the Pacific Plate beneath the northeastern corner of the Australian Plate. Zellmer & Taylor (2001), on the basis of a detailed study of acoustic reflectivity and morphology, have identified three microplates to explain the kinematics of the Lau Basin–Tonga Trench system in the northern part of the Lau Basin. The three plates identified (Fig. 1) are named the Niua fo’ou Plate, the Tonga Plate and the Australian Plate itself, in the southern part of the Lau Basin (Zellmer & Taylor, 2001).
|
|
Global Positioning System (GPS) measurements indicate that the instantaneous plate convergence across the northern Tonga Trench is 24 cm/year, the fastest recorded convergence velocity on the modern Earth (Bevis et al., 1995
). At the northern terminus of the trench near 15°S (Figs 1 and 2), plate convergence gives way to complex strike-slip motion along a transform fault and back-arc extension in the northern Lau Basin. The trench–transform fault transition forms a slab-edge, the geometric consequences of which are the continuing rupture and rifting of oceanic lithosphere (Millen & Hamburger, 1998; Govers & Wortel, 2005). A consequence of the slab edge is the presence of a ‘slab window’, which has allowed the flow of hot sub-Pacific mantle into the northern Lau Basin, above the subducting Pacific Plate (Danyushevsky et al., 1995; Millen & Hamberger, 1998; Smith et al., 2001; Wiens et al., 2006). The presence of hot, depleted plume mantle [residual to Samoan ocean island basalt (OIB) magmatism] as a component of this inflowing mantle, fluxed by slab-derived fluids, is considered the primary cause for the presence of boninitic magmas at the termination of the Tonga Trench and northern Lau Basin (Danyushevsky et al., 1995; Falloon et al., 2007). Multibeam mapping along this boundary and the termination of the trench (Fig. 2) reveals a tectonically complex terrain, including the transition from subduction to strike-slip motion. The terrain contains: (1) new seafloor generated by back-arc spreading along the North Eastern Lau Spreading Centre; (2) a deep, well-defined graben structure cutting across the north Tonga Ridge arc crust; (3) extensional rift zones associated with large caldera-like features; (4) young volcanic seamounts located within these extensional and graben structures (Fig. 2). Boninites and adakites together have been recovered by dredging at station D113 during the 1996 voyage of the R.V. Melville and adakites were recovered at station 22 during the 1984 voyage of the R.V. Natsushima. Details of the dredge locations are given in Table 1 and Fig. 2.
|
|
PETROGRAPHY AND GEOCHRONOLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND GEOCHRONOLOGYANALYTICAL METHODSGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The boninites recovered by dredging at station D113 are olivine + orthoypyroxene ± clinopyroxene ± plagioclase-phyric vesicular lavas. All have experienced varying degrees of seafloor alteration. Orthopyroxene phenocrysts are pseudomorphed by chlorite, and in some case by microcrystalline quartz. Olivine phenocrysts have been mostly replaced by carbonate. The groundmass (formerly glass + orthopyroxene and clinopyroxene microlites) has undergone smectite/chlorite ± zeolite alteration, with common micro Fe–Ti oxides. Ignoring the effects of alteration, the boninites from dredge 113 are petrographically identical to the fresh boninites recovered from station 21 by the R.V. Natsushima (Fig. 2; Falloon et al., 1987
; Falloon & Crawford, 1991).
In contrast to the boninites, the adakites are relatively fresh [loss on ignition (LOI) <1·13 wt %] medium- to fine-grained holocrystalline rocks with a microgranitic texture. They contain a mineral assemblage typical of adakites (Defant & Drummond, 1990; Martin et al., 2005) consisting of approximately equal amounts of quartz and plagioclase (
An60 in sample 4-14), plus 5–15 modal % hornblende, with accessory titanite, zircon, apatite and titanomagnetite. The texture of the Tonga adakites is consistent with their crystallization as dykes in a hypabyssal environment.
The presence of accessory zircon allows the adakites to be precisely dated using the U–Pb decay system and laser ablation inductively coupled plasma mass spectrometry (ICP-MS; see below). The results of our dating are given in Table 2 and Fig. 3. Both the adakite samples we have dated have well-constrained ages of 2·5 Ma (Table 2), contemporaneous with the recent to active boninite magmatism at the northern termination of the Tonga Trench (< 3 Ma; Acland, 1996; Falloon et al., 2007). This is the first time that a contemporaneous association of adakites and boninite lavas has been documented from an active arc setting.
|
|
|
ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND GEOCHRONOLOGYANALYTICAL METHODSGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The major and trace element and isotopic compositions of the north Tongan boninites and adakites are presented in Tables 3 and 4. Samples were ground in an agate mill. Major and trace element analyses were performed at the School of Earth Sciences (UTAS) by X-ray fluorescence (XRF) spectrometry, using the methods of Robinson (2003
). ICP-MS trace element analyses at UTAS (Table 4) were obtained using the methods of Robinson et al. (1999) and Yu et al. (2000). Isotope analysis were performed at the University of Melbourne using the methods of Woodhead (2002) and Maas et al. (2005). Samples were hand-picked rock chips washed in hot 6N HCl for 15 min and then dissolved in HF–HNO3. After extraction of Pb using conventional anion exchange using HBr–HCl media, Sr and light rare earth elements (LREE) were extracted on EICHROM Sr. resin and RE.resin, respectively, followed by Nd purification on EICHROM LN resin. All isotopic analyses were carried out on a Nu Instruments multi-collector ICP-MS system coupled to a CETAC Aridus desolvating nebulizer. Full details of all our analytical techniques, including analyses of international standards, have been given by Falloon et al. (2007).
|
|
For U–Pb dating, zircons were separated from the samples by crushing in a ring mill (3–4 s) to <400 µm and panning to concentrate the heavy minerals. Magnetic minerals were first removed using a hand magnet, and zircons were then hand picked under cross-polarized transmitted light and mounted in a 2·5 cm epoxy block. After grinding and polishing to expose the top of the crystals, they were analysed using a 213 nm New Wave solid-state laser and a Hewlett Packard 4500 quadrupole ICP-MS system. The zircons were ablated in a He atmosphere in a custom-made chamber with the laser pulsing at 5 Hz and a beam diameter of 30 µm delivering
12 J/cm2 and drilling at 1 µm/s. A total of 10 masses were counted (96Zr, 146Nd, 178Hf, 202Hg, 204Pb, 206Pb, 207Pb, 208Pb, 232Th, 238U) with longer counting times on Pb isotopes giving a total quadrupole cycling rate of 0·2 s. Each analysis began with a 30 s analysis of background gas followed by 30 s with the laser switched on. Four analyses of the primary standard (Temora zircons of Black et al. (2003) and 2 analyses of the secondary standard (91500 zircon of Wiedenbeck et al. (1995) were analysed both before and after every 12 zircons to correct for mass bias, machine drift and down-hole fractionation. Trace element concentrations were calculated using the 91500 zircon assuming stoichiometric Zr and Hf. Further details of the method used have been given by Kosler & Sylvester (2003), Black et al. (2004) and Jackson et al. (2004). |
GEOCHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND GEOCHRONOLOGYANALYTICAL METHODSGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Boninites
The Tongan boninites have been divided into two broad groups by Sobolev & Danyushevsky (1994
) based on geographical location, which is also reflected in differences in FeOT contents (Fig. 4d). The eastern group, associated with rifting of the north Tonga Ridge, has generally higher FeOT contents (stations 21, 23, D115, D118; Fig. 4d) compared with the western group. The western group boninites are associated with rifting in the northern Lau Basin immediately adjacent to the north Tonga Ridge (Sobolev & Danyushevsky, 1994; Falloon et al., 2007). Figure 4 shows that the major element chemistry of the D113 boninites, taking into account the effects of seafloor alteration (e.g. elevated K2O abundances, Fig. 4g), is identical to that of the eastern group boninites from north Tonga, including the U-shaped, LREE-enriched chondrite-normalized REE patterns (Fig. 5). They also have similar normalized trace element abundance patterns (NAP) to the eastern group boninites, apart from elevated Rb and K abundances, which are possibly due to alteration (Fig. 6). The Sr, Nd and Pb isotopic composition (Figs 7 and 8) of boninite sample 113-1-12 (Table 2) is also very similar to those of the eastern group boninites (Falloon et al., 1989, 2007). In summary, the D113 boninites have essentially identical whole-rock geochemistry (apart from K and Rb) to the recent, fresh boninites associated with rifting of the northern Tonga Ridge, at the northern termination of the Tonga Trench.
|
|
|
|
|
Adakites
Recently, Martin et al. (2005
) suggested that adakites from modern arc settings can be assigned to either of two broad compositional groups, termed high-SiO2 adakites (HSA) and low-SiO2 adakites (LSA). The HSA are considered by many workers to be modern analogues for the Archaean TTG (tonalite–trondjhemite–granitoid) crustal suite (Defant & Drummond 1990; Martin et al., 2005), whereas the LSA are considered to be analogues for the Archaean sanukitoid suites (Shirey & Hanson, 1984).
The whole-rock geochemistry of the Tonga adakites is presented in Figs 4, 5 and 7–9
, and compared with that of the Tongan boninites and modern adakites (HSA and LSA, Martin et al., 2005). Based on their MgO, SiO2, CaO and FeOT contents, the Tonga adakites are most appropriately classified as HSA. Compared with the majority of HSA, the Tonga adakites have relatively lower TiO2, Al2O3, Na2O, K2O and P2O5 contents, and they also have overall lower abundances of incompatible elements (Fig. 9) compared with average HSA and LSA of Martin et al. (2005). The Tonga adakites have LREE-enriched REE patterns similar to the Tongan boninites (Fig. 5). The range of REE patterns shown by the eastern group boninites reflects fractional crystallization of parental boninites (MgO 20 wt %) to rhyolites (MgO 1 wt %). REE abundances increase with decreasing MgO, and become slightly more enriched in LREE. However, boninite REE patterns retain their broad U-shaped profiles with progressive differentiation. The adakites, in contrast, have REE patterns that are discordant to those of the boninites, with very low heavy REE (HREE) and relatively high LREE contents (Fig. 5). This suggests that the adakites are unlikely to be the result of advanced degrees of crystal fractionation from boninite parental magmas, a conclusion supported by the isotopic composition of adakite 4-14, which has significantly lower 143Nd/144Nd, higher 87Sr/86Sr and lower 206Pb/204Pb than the boninites (Figs 7 and 8).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND GEOCHRONOLOGYANALYTICAL METHODSGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Comparisons with adakites
In Figure 9 we compare the primitive mantle normalized trace element abundance patterns of the Tonga adakites with the average HSA and LSA adakites of Martin et al. (2005
). Also shown is a representative recent adakite from Kadavu Island, Fiji (Verbeeten, 1996). The Tonga adakites have typical adakite trace element patterns, with a distinctive depletion in Ti and a slight enrichment in Zr, relative to the middle REE (MREE) Eu and Gd, as well as a relative enrichment of Sr compared with the REE and high field strength elements (HFSE). The Tonga adakites, however, have two features that distinguish them from the majority of adakites. First, the Tonga adakites have lower abundances of trace elements (Fig. 9), as reflected in very low Rb, K (Figs 9 and 10d) and TiO2 abundances (Figs 9 and 10e). Second, the Tonga adakites are relatively rich in Nb, resulting in a lack of a typical subduction-related negative Nb anomaly relative to La (Fig. 9). LaN/NbN values for the Tonga adakites are 1·05 compared with >3 for most adakites. The Tonga adakites have significantly higher Nb contents relative to the HSA adakites at a given SiO2 content (Fig. 10f). In Fig. 10, the geochemistry of the Tonga adakites is compared in detail with the compositions of HSA and LSA adakites based on the key geochemical indices chosen by Martin et al. (2005) to highlight the differences between HSA and LSA adakites. The Tonga adakites plot with the HSA (Fig. 10) but are distinctive in their low Rb, K and TiO2, and high Nb values. The Tonga adakites in terms of their K, Rb, TiO2 and Cr/Ni values plot in the domain where the HAS field overlaps the LSA field (Fig. 10d and e).
|
Possible role in boninite petrogenesis?
A particularly interesting aspect of Tongan boninite petrogenesis is the requirement of temperatures as high as
1480°C at depths of 45 km in the mantle wedge (Sobolev & Danyushevsky, 1994; Falloon & Danyushevsky, 2000). To account for these high temperatures, Danyushevsky et al. (1995) proposed a model involving the melting of Samoan plume mantle. In this model, depleted or refractory Samoan plume mantle [component D1 in the model of Danyushevsky et al. (1995)], which had moved across the ‘slab window’, a geodynamic consequence of the transform fault–trench intersection above the subducting Pacific plate, underwent melting as a result of contact with H2O-rich fluids derived from dehydration processes occurring within subducted Pacific oceanic crust [component E1 in the model of Danyushevsky et al. (1995)]. This model also includes a separate low-degree, plume-derived silicate melt fraction as a distinct component in the petrogenesis of some boninites [component E2 in the model of Danyushevsky et al. (1995)]. Thus components D1 and E2 are, in general terms, identified with the Samoan plume, and component E1 with a ‘subduction zone component’. Danyushevsky et al. (1995) successfully modelled the geochemistry of these north Tongan boninites by mixing between the components D1, E1 and E2.
One geodynamic consequence of the transform fault–trench intersection at the northern termination of the Tonga Trench is the juxtaposition of the edge of the subducting Pacific oceanic crust against hot Samoan plume mantle. Given the high temperatures proposed for boninite petrogenesis (1480°C), slab melting and the consequent generation of adakite magmas could be an expected consequence of the presence of a ‘slab edge’, as has been proposed for some scenarios of hot subduction (Martin et al., 2005; Thorkelson & Breitsprecher, 2005). The model of Danyushevsky et al. (1995) attributed the E1 component to H2O-rich fluids, based on the geochemistry of the boninites, and did not consider the possibility of slab melting and mixing between slab-derived melts and boninite magmas. In this section, we explore the possibility of adakite as an enriching agent in boninite petrogenesis.
Danyushevsky et al. (1995) recognized that the compositions of Tongan boninites could be explained by magma mixing between relatively ‘depleted’ and ‘enriched’ boninite compositions. Recently, Falloon et al. (2007) have identified three end-member boninite magmas, termed types 1–3, based on new sampling of boninites at the northern termination of the Tongan Trench. In Fig. 11 the normalized trace element patterns of the three boninite end-members are compared with those of the Tonga adakites. The type 1 boninite end-member in the model of Danyushevsky et al. (1995) is considered to be the result of a subduction component (E1), probably a slab-derived fluid, fluxing and causing melting of depleted OIB source mantle (D1, the Samoan plume). The type 1 end-member has a gently sloping, LREE-depleted REE pattern (LaN/SmN = 0·73, Fig. 12) and very low abundances of HFSE and HREE (e.g. Nb 0·28 ppm, Yb 0·61; Falloon et al., 2007). These features of the type 1 end-member reflect the depleted D1 source component. The influence of the E1 component (slab-derived fluid) is reflected in the typical subduction zone enrichment in large ion lithophile elements (LILE; Rb, Ba, K, and Sr) seen in the type 1 end-member (Fig. 11). The type 1 boninite end-member has 143Nd/144Nd values within the range shown by Pacific mid-ocean ridge basalt (MORB) and modern volcanoes of the Tofua volcanic arc (TVA), but displaced to significantly higher 87Sr/86Sr values (Fig. 7). The type 1 boninite end-member also has Pb isotope values that are close to those of Pacific MORB and TVA lavas (Fig. 8).
|
|
The type 2 boninite end-member in the model of Danyushevsky et al. (1995
) is considered to be the result of the addition of the E2 (OIB) component to the type 1 end-member. The type 2 boninite end-member has a LREE-enriched REE pattern (LaN/SmN = 3·5, Fig. 12), with low HREE abundances (Yb 0·85 ppm, Falloon et al., 2007), and enrichment in HFSE (LaN/NbN = 1·0; except TiO2, which remains relatively low) and LILE. The type 2 end-member has low 143Nd/144Nd within the range of that of the Samoan lavas, but with lower 87Sr/86Sr (Fig. 7). However, the type 2 end-member has Pb isotope compositions that fall within the range displayed by Samoan lavas (e.g. Fig. 8).
The type 3 boninite end-member is intermediate in character between the type 1 and type 2 boninite end-members. It has a primitive mantle normalized trace element abundance pattern that shows enrichment in LREE and HFSE relative to the type 1 boninite (resulting in a U-shaped REE pattern; Fig. 12), but still retains a subduction zone-type LILE-enriched normalized trace element abundance pattern (Fig. 11). The type 3 boninite end-member has Nd, Sr and Pb isotopic values within the range displayed by lavas from Samoa (Figs 7 and 8). Mixing between these three boninite end-members can account for the geochemistry of all the enriched (in terms of LREE and 143Nd/144Nd values) boninites recovered from the north Tonga Ridge and northern Lau Basin (Falloon et al., 2007).
In Figs 12–14 we present the results of quantitative bulk mixing models to explore the possibility that the addition of an adakite melt to a depleted type 1 boninite end-member could explain the geochemistry of the more enriched boninite end-members. Figure 12 demonstrates that a 30:70 mix (Model A, Fig. 12) of adakite 113-2-12 and type 1 boninite produces a very close match to the REE pattern of the type 3 boninite end-member. Bulk mixing models, however, cannot explain the REE abundances of the Type 2 boninite, as they are richer in all REE compared with the Tonga adakites (Fig. 12). Figure 13 demonstrates that bulk mixing of a Tongan adakite component cannot successfully explain the trace element geochemistry of the enriched boninite end-members. A 30:70 mix (Model A), although giving a good approximation to the type 3 boninite normalized trace element abundance pattern, would impose significant positive Sr and Zr anomalies, not present in the Tongan boninites. This conclusion would hold firm if other adakitic compositions were arbitrarily chosen as end-members in mixing calculations. The addition of an adakitic component into the type 1 boninite end-member would produce significant Zr enrichment over Sm and positive Sr anomalies, which are not observed in the Tongan high-Ca boninites.
|
|
|
In Fig. 14a we illustrate mixing calculations between the Tongan adakite 4-14 and the Tongan type 1 boninite end-member in Sr and Nd isotope space (Model B). Although the Sr and Nd isotopic compositions of adakite 4-14 are appropriate for it to be an end-member mixing component to explain the isotopic composition of the type 3 boninite (as well as the D113 boninite 113-1-12), the mixing proportions required (8:92, Model B) are inconsistent with proportions required for REE (30:70, Model A, Fig. 12). In addition, the Pb isotopic composition of adakite 4-14 indicates that a third component with distinctive Pb isotopic composition would be required for such a mixing model to be feasible. In summary, quantitative mixing models suggest that it is unlikely that adakite magmas are directly involved in the petrogenesis of the high-Ca boninite magmas at the northern termination of the Tonga Trench.
Tongan adakite petrogenesis
Although the Tonga adakites are unlikely to be involved in high-Ca boninite petrogenesis, their recovery at the northern termination of the Tonga Trench suggests that melting of the slab edge is occurring. Strong geochemical and experimental evidence shows that most true adakite magmas are the direct result of melting of the slab. Significant reaction of these HSA adakitic slab melts with the overlying mantle could eventually lead to the generation of LSA (Kelemen et al., 2003; Martin et al., 2005), or it could alternatively produce fertilized mantle zones, which, if subsequently remelted, could also produce LSA adakites (Rogers & Saunders, 1989; Martin et al., 2005). However, some workers have stressed that not all arc magmas with adakitic features necessarily indicate slab melting. They have argued, for example, that high-pressure fractionation of garnet from a ‘normal’ arc magma may also be an important process in generating magmas with a ‘slab melting’ signature (Castillo et al., 1999; Macpherson et al., 2006).
The Tonga adakites have several features consistent with an origin linked to slab melting and limited reaction with the overlying mantle wedge. For example, their Nd and Sr isotopic compositions are clearly distinct from those of the Tongan boninites and volcanic rocks of the TVA, which indicates that they cannot be the result of either high- or low-pressure crystal fractionation of such magmas (Fig. 7). The Nd and Sr isotopic compositions are, however, entirely consistent with a source composition that is a mixture between Pacific MORB and the average composition of Tongan pelagic sediments (Fig. 14b). In Figs 12 and 14b and Tables 5–7 we present the results of quantitative modelling of the Tonga adakite Sr and Nd compositions and REE abundances.
Cretaceous-age Pacific oceanic crust currently being subducted at the northern Tonga Trench was generated at the present-day East Pacific Rise (EPR) Spreading Centre. We have, therefore, used the average composition of present-day EPR MORB to model the composition of unaltered subducted Pacific oceanic crust. The composition of sediment currently being subducted at the northern Tonga Trench has been sampled at Deep Sea Drilling Project (DSDP) site 595/6 and the average sediment composition for this crustal section has been calculated by Plank & Langmuir (1998). As shown in Fig. 14b, a mixing line between average EPR MORB and the average DSDP site 595/6 sediment passes below the Tonga adakite in Sr and Nd isotopic space (Model C, Fig. 14b). However, subducting Pacific oceanic crust is very likely to have experienced seafloor alteration, which would raise its 87Sr/86Sr value. The Sr isotopic composition of altered oceanic crust from DSDP–ODP (Ocean Drilling Program) sites 417/418 ranges from 0·70364 to 0·70744 with an average value of 0·70475 (Staudigel et al., 1995). If we assume this average value to be representative of altered Pacific oceanic crust, then a mixing line between average altered MORB and the DSDP site 595/6 sediment passes above the Tonga adakites in Sr and Nd isotopic space (Model D, Fig. 14b). If the Tonga adakite is indeed a melt of a mixture of altered Pacific oceanic crust and sediment, then the composition of the altered Pacific oceanic crust must lie between the 87Sr/86Sr values of unaltered EPR MORB and the average 87Sr/86Sr determined by Staudigel et al. (1995) for MORB from DSDP–ODP sites 417/418 in the Atlantic Ocean. If we make the reasonable assumption that the Pacific oceanic crust has had its 87Sr/86Sr raised by 0·001 units by seafloor alteration (Leat et al., 2004), then the average 87Sr/86Sr value of the altered Pacific Oceanic crust is 0·703563. A mixing line using this 87Sr/86Sr value passes directly through the Tonga adakite (Model E). Model E (Fig. 14b) suggests that the source for the Tonga adakite consists of 0·76 altered Pacific oceanic crust and 0·24 Tonga sediment.
We can demonstrate that this source composition is consistent with the REE and Sr abundances in the Tonga adakites by making the following series of assumptions: (1) Sr is completely incompatible during partial melting and therefore, along with the assumption of batch partial melting, can fix the amount of partial melting of the source (F); (2) the residue after adakite melting is eclogite; (3) we can constrain the proportions of clinopyroxene and garnet in the residual eclogite by using the high-pressure normative composition (Yaxley & Green, 1998) of the bulk source; (4) low-pressure REE crystal/liquid Kd values determined for clinopyroxene and garnet for dacitic and rhyolitic compositions at low pressure can be applied at higher pressures.
The bulk composition of the source for the Tonga adakite is calculated using the mass balance constraints derived from the isotopic modelling (Fig. 14b), and this composition is presented in Tables 5–7 (Bulk Slab). The Bulk Slab has a Sr content of 233 ppm which, based on our assumptions, means the degree of partial melting F is 0·3. This amount of melting lies within the experimentally determined range for adakite production (F 0·1–0·4, Martin et al., 2005). In Table 6, we present the major element composition and high-pressure normative composition of the Bulk Slab. The model Bulk Slab has a normative ratio of Garnet/Jadeite + Diopside of 14/23. We therefore make the reasonable assumption that the garnet/clinopyroxene ratio in the eclogite residue (0·38 garnet and 0·62 clinopyroxene) will also be approximately the same as that in the source composition, because the melt component will be dominated by normative Quartz. If we use this modal constraint on our residue, combined with appropriate Kd values (Table 7), melting of the Bulk Slab (F = 0·3) produces an close match to the REE pattern of the Tonga adakites (Model F, Fig. 12).
|
|
In summary, the Sr and Nd isotopic and REE compositions of the Tonga adakites are consistent with slab melting where the Bulk Slab comprises a mixture of altered oceanic crust and sediment. The Pb isotopic composition of the adakite is not consistent with mixing between a Pacific MORB composition and the average Tongan pelagic sediment. However, as sedimentary Pb isotopic compositions are highly variable, this may imply that the Pb isotopic composition of the Tonga adakites sampled a subset of this range, or that a third component is required to fully explain the Pb isotopic compositions of the Tongan adakites.
The relatively high Nb contents (Fig. 10f) of the North Tonga adakites are also consistent with an origin linked to slab melting. The data of Zamora (2000) demonstrated that experimental melts of basalt in the range of 850–1150°C and 0·7–3·5 GPa have 8–16 ppm Nb at SiO2 contents >70 wt % (see Martin et al., 2005, fig. 6). There are at least three factors that may explain the presence or absence of a Nb anomaly in adakites. First, the pressure of melting may be important in stabilizing rutile as residual phase. Experiments reported by Xiong et al. (2006) suggested that pressures >1·5 GPa are necessary to stabilize rutile. Second, oxygen fugacity may be an important control on whether rutile is a stable phase in the slab or wedge (Reagan & Gill, 1989). Previously, Green et al. (1987) have suggested that the transform fault–trench intersection at the northern termination of the Tongan trench may allow deep degassing of reduced CH4–H2-rich fluids, which can subsequently interact with the more oxidized mantle and slab, causing redox melting. In such a scenario it may be possible for rutile to be stabilized in zones dominated by more oxidized fluids, in contrast to the absence of rutile in zones dominated by more reduced fluids (Reagan & Gill, 1989). Third, the presence or absence of residual rutile may simply be related to the degree of partial melting of the slab. The relatively high Nb contents of the Tonga adakites supports models that explain the enriched geochemistry of LSA adakites as the result of remelting of a mantle wedge source previously fertilized by HSA adakites (Rogers et al., 1985; Martin et al., 2005).
In general, the Tonga adakites have major element geochemistry consistent with an origin by slab melting (Rapp et al., 1999; Martin et al., 2005). In Fig. 15, the composition of the Tonga adakites is compared with that of experimental slab melts and the compositions of HSA and LSA adakites. Also plotted is the parental eastern group boninite composition determined by Sobolev & Danyushevsky (1994). Clearly, the Tonga adakites have SiO2 contents and Mg-numbers overlapping the relatively high SiO2, high Mg-number part of the experimental melt field and the HSA adakite array. Interestingly, the Tonga adakites lie on a linear extension from the eastern group boninite parental composition, which passes through the experimental compositions of reacted slab melts from the experimental study of Rapp et al. (1999). The slightly lower SiO2, higher Mg-number, and higher Cr and Ni abundances in adakite 113-2-12 compared with adakite 4-14 are consistent with sample 113-2-12 reacting with the mantle wedge before eruption. The parental eastern boninite composition has a lower SiO2 content and has been interpreted as a fully reacted subduction component with refractory harzburgite at 1·5 GPa and 1480°C (Falloon & Danyushevsky, 2000).
|
Favourable conditions for slab melting and adakite magma generation have been summarized by Thorkelson & Breitsprecher (2005
). Adakite generation is mostly likely to occur between the garnet-in and amphibole-out phase boundaries between 0·7 and 2·6 GPa. At 1·5 GPa, which is the inferred pressure of boninite generation, the adakites would be formed in the temperature range 650–1050°C (see Thorkelson & Breitsprecher, 2005, fig. 2). Using the software PETROLOG (Danyushevsky, 2001), adakite 4-14 would have a clinopyroxene liquidus temperature of <1037°C, at the higher end of the range of favourable conditions assuming the melt contains >2 wt % H2O. Adakite 113-2-12 with 2 wt % H2O has a clinopyroxene liquidus of 1155°C, outside of this range. However, the higher calculated liquidus temperature for adakite 113-2-12 would be consistent with some reaction with a higher-temperature mantle wedge, as suggested by the SiO2 vs Mg-number relationships shown in Fig. 15.
Defant & Drummond (1990) proposed that on the modern Earth, special tectonic circumstances are required to produce adakites; in particular, the subduction of young (< 25 Ma) relatively hot lithosphere appears to be a requirement for slab melting to occur. The recovery of adakites from the northern termination of the Tonga Trench supports geodynamic models involving melting of the slab edge in the transition from subduction to a transform plate boundary (Thorkelson & Breitsprecher, 2005). In the case of Tonga, the slab edge is melted as a result of the juxtaposition of a cold and old subducted slab against hot mantle from the Samoan mantle plume.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND GEOCHRONOLOGYANALYTICAL METHODSGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The recovery of adakites and boninites from the northern termination of the Tonga Trench indicates that both magma types are related to this unique tectonic setting. The transition from a subduction to a transform fault plate boundary has created a slab window with an associated slab edge. Boninites are generated as a result of hot, refractory Samoan plume mantle moving across the slab window and subsequently being fluxed by H2O-rich fluids from the subducting Pacific oceanic crust. The Tonga adakites, in contrast, result from the direct melting of the slab edge heated by the hot Samoan plume. The geochemistry of the Tonga adakites is consistent with melting of a mixture of subducted Pacific oceanic crust and sediment. Geochemical modelling shows that the adakites are not involved in the petrogenesis of the Tongan high-Ca boninites.
|
ACKNOWLEDGEMENTS |
|---|
We are grateful to Hervé Martin for letting us use his adakite database (Martin et al., 2005
). This research was supported by the Australian Research Council and by NSF grants OCE-9521023 and OCE-9521039. We wish to thank all the scientists, officers and crew who participated in the 1994 R.V. Natsushima and the 1996 R.V. Melville voyages. We thank Phil Robinson, Katie McGoldrick and Sarah Gilbert for assistance with geochemical analyses. We thank Andrew Stacey for help with Fig. 2. We thank editor Majorie Wilson and reviewers Hervé Martin, Paul Leat, Richard Arculus and Steve Parman for their constructive reviews and comments.
*Corresponding author. E-mail: trevor.falloon{at}utas.edu.au
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY AND GEOCHRONOLOGYANALYTICAL METHODSGEOCHEMISTRYDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Acland AS. Magma genesis in the northern Lau Basin, S. W. Pacific. In: Ph.D. thesis (1996) University of Durham. 260.
Bevis M, Taylor W, Schultz BE, Recy J, Isacks BL, Helu S, Singh R, Kendrick E, Stowell J, Taylor B, Calmant S. Geodetic observations of very rapid convergence and back-arc extension at the Tonga Arc. Nature (1995) 374:249–251.[CrossRef]
Black LP, Kamo SL, Allen CM, Aleinikoff JN, Davis DW, Korsch RJ, Foudoulis C. TEMORA 1: a new zircon standard for Phanerozoic U–Pb geochronology. Chemical Geology (2003) 200:155–170.[CrossRef][Web of Science]
Black LP, Kamo SL, et al. Improved 206Pb/238U microprobe geochronology by the monitoring of a trace-element related matrix effect; SHRIMP, ID-TIMS, ELA-ICP-MS, and oxygen isotope documentation for a series of zircon standards. Chemical Geology (2004) 205:115–140.[CrossRef][Web of Science]
Bloomer SH, Taylor B, MacLeod CJ, Stern RJ, Fryer P, Hawkins JW. Early arc volcanism and the ophiolite problem: a perspective from drilling in the Western Pacific. Active Margins and Marginal Basins of the Western Pacific. American Geophysical Union, Geophysical Monograph—Taylor B, Natland J, eds. (1995) 88:1–39.
Castillo PR, Janney PE, Solidum RU. Petrology and geochemistry of Camiguin Island, southern Philippines: insights to the source of adakites and other lavas in a complex arc setting. Contributions to Mineralogy and Petrology (1999) 134:33–51.[CrossRef][Web of Science]
Crawford AJ, Beccaluva L, Serri G. Tectonomagmatic evolution of the West Philippine–Mariana region and the origin of boninites. Earth and Planetary Science Letters (1981) 54:346–356.[CrossRef][Web of Science]
Crawford AJ, Falloon TJ, Eggins S. The origin of island arc high-alumina basalts. Contributions to Mineralogy and Petrology (1987) 97:417–430.[CrossRef][Web of Science]
Crawford AJ, Falloon TJ, Green DH. Classification, petrogenesis and tectonic setting of boninites. In: Boninites and Related Rocks—Crawford AJ, ed. (1989) London: Unwin Hyman. 1–49.
Danyushevsky LV. The effect of small amounts of H2O on crystallization of mid-ocean ridge and backarc basin magmas. Journal of Volcanology and Geothermal Research (2001) 110:265–280.[CrossRef][Web of Science]
Danyushevsky LV, Sobolev AV, Falloon TJ. North Tongan high-Ca boninite petrogenesis: The role of Samoan plume and subduction zone–transform fault transition. Journal of Geodynamics (1995) 20:219–241.[CrossRef][Web of Science]
Defant MJ, Drummond MS. Derivation of some modern arc magmas by melting of young subducted lithosphere. Nature (1990) 347:662–665.[CrossRef]
Falloon TJ, Crawford AJ. The petrogenesis of high-calcium boninite lavas dredged from the northern Tonga ridge. Earth and Planetary Science Letters (1991) 102:375–394.[CrossRef][Web of Science]
Falloon TJ, Danyushevsky LV. Melting of refractory mantle at 1·5, 2 and 2·5 GPa under anhydrous and H2O-undersaturated conditions: implications for the petrogenesis of high-Ca boninites and the influence of subduction components on mantle melting. Journal of Petrology (2000) 41:257–283.
Falloon TJ, Green DH, Crawford A. Dredged igneous rocks from the northern termination of the Tofua magmatic arc, Tonga and adjacent Lau Basin. Australian Journal of Earth Sciences (1987) 34:487–506.[Web of Science]
Falloon TJ, Green DH, Crawford AJ. Petrogenesis of high-Mg and associated lavas from the north Tonga trench. In: Boninites and Related Rocks—Crawford AJ, ed. (1989) London: Unwin Hyman. 357–395.
Falloon TJ, Bloomer SH, Crawford AJ. Boninites recovered from the Tonga Forearc. EOS Transactions, American Geophysical Union, Fall Meeting, San Francisco (1997) 78(46):815.
Falloon TJ, Danyushevsky LV, Crawford AJ, Maas R, Woodhead JD, Eggins SM, Bloomer SH, Wright DJ, Zlobin SK, Stacey AR. Multiple mantle plume components involved in the petrogenesis of subduction related lavas from the northern termination of the Tonga Arc and northern Lau Basin: evidence from the geochemistry of arc and backarc submarine volcanics. Geochemistry, Geophysics, Geosystems (2007) 8(1). doi:10.1029/2007GC001619.
Gill JB. Orogenic Andesites and Plate Tectonics (1981) Berlin: Springer. 358.
Govers R, Wortel MJR. Lithosphere tearing at STEPS faults: response to edges of subduction zones. Earth and Planetary Science Letters (2005) 236:505–523.[CrossRef][Web of Science]
Green DH. Experimental testing of ‘equilibrium’ partial melting of peridotite under water-saturated, high pressure conditions. Canadian Mineralogist (1976) 14:255–268.
Green DH, Falloon TJ, Taylor WR. Mantle derived magmas—role of variable source peridotite and depleted peridotite from 2 to 30 kbars and application to basalt petrogenesis. Journal of Petrology (1987) 29:1257–1282.
Green TH, Green DH, Ringwood RE. The origin of high-alumina basalts and their relationships to quartz tholeiites and alkali basalts. Earth and Planetary Science Letters (1967) 2:41–51.[CrossRef][Web of Science]
Irving AJ, Frey FA. Effect of composition of rare-earth elements, Hf, Sc and Co between garnet and liquid—experimental and natural evidence. EOS Transactions, American Geophysical Union (1976) 57(4):339.
Irving AJ, Frey FA. Distribution of trace-elements between garnet megacrysts and host volcanic liquids of kimberlite to rhyolitic composition. Geochimica et Cosmochimica Acta (1978) 42(NA6):771–787. doi:10.1016/0016-7037(78)90092-3.[CrossRef][Web of Science]
Ishiwatari A, Yanagida Y, Li Y.-B, Ishii T, Haraguchi S, Koizumi K, Ichiyama Y, Umeka M. Dredge petrology of the boninite- and adakite-bearing Hahjima Seamount of the Ogasawara (Bonin) forearc: An ophiolite or a serpentinite seamount. Island Arc (2006) 15:102–118. doi:10.1111/j.1440-1738.2006.00512.x.[CrossRef]
Jackson SE, Pearson NJ, Griffin WL, Belousova EA. The application of laser ablation-inductively coupled plasma-mass spectrometry to in situ U–Pb zircon geochronology. Chemical Geology (2004) 211:47–69.[CrossRef][Web of Science]
Eiler J. Along-strike variation in the Aleutian Island Arc: genesis of high Mg-number andesite and implications for continental crust. Inside the Subduction Factory. American Geophysical Union, Geophysical Monograph (2003) 138:223–276.
Kosler J, Sylvester PJ. Present trends and the future of zircon in geochronology; laser ablation ICPMS. Reviews in Mineralogy and Geochemistry (2003) 53:243–275.
Kuno H. High-alumina basalt. Journal of Petrology (1960) 1:121.
Leat PT, Pearce JA, Barker PF, Millar IL, Barry TL, Larter RD. Magma genesis and mantle flow at a subducting slab edge: the South Sandwich arc–basin system. Earth and Planetary Science Letters (2004) 227:17–35.[CrossRef][Web of Science]
Le Bas MJ. IUGS reclassification of the high-Mg and picritic volcanic rocks. Journal of Petrology (2000) 41:1467–1470.
Ludwig KR. User's Manual for Isoplot 3.0. Berkeley Geochronology Center Special Publication (2003) 4.
Maas R, Kamenetsky MB, Sobolev AV, Kamenetsky VS, Sobolev NV. Sr–Nd–Pb isotopic evidence for a mantle origin of alkali chlorides and carbonates in the Udachnaya kimberlite, Siberia. Geology (2005) 35:549–552.
MacLeod CJ. Geodynamics of the Tonga subduction zone and Lau Backarc basins. EOS Transactions, American Geophysical Union (1996) 77:F325.[CrossRef]
Macpherson CG, Dreher ST, Thirlwall MF. Adakites without slab melting: high pressure differentiation of island arc magma, Mindanao, the Philippines. Earth and Planetary Science Letters (2006) 243:581–593.[CrossRef][Web of Science]
Mahood GA, Hildreth EW. Large partition coefficients for trace elements in high-silica rhyolites. Geochimica et Cosmochimica Acta (1983) 47:11–30. doi:10.1016/0016-7037(83)90087-X.[CrossRef][Web of Science]
Martin H. The adakitic magmas: modern analogues of Archaean granitoids. Lithos (1999) 46:411–429.[CrossRef][Web of Science]
Martin H, Smithies RH, Rapp R, Moyen J.-F, Champion D. An overview of adakite, tonalite–trondhjemite–granodiorite (TTG), and sanukitoid: relationships and some implications for crustal evolution. Lithos (2005) 79:1–24.[CrossRef][Web of Science]
Melson WG, Vallier TL, Wright TL, Byerly G, Nelen J. Chemical diversity of abyssal volcanic glass erupted along Pacific, Atlantic, and Indian Ocean sea-floor spreading centers. In: The Geophysics of the Pacific Ocean Basin and Its Margin. Geophysical Monograph Series—Sutton GH, Manghnani MH, Moberly R, McAfee E, eds. (1976) vol. 19. Washington, DC: American Geophysical Union. 351–367.
Michael PJ. Partition coefficients for rare earth elements in mafic minerals of high silica rhyolites. The importance of accessory mineral inclusions. Geochimica et Cosmochimica Acta (1988) 52:275–282. doi:10.1061/0016-7037(88)90083-X.[CrossRef][Web of Science]
Millen DW, Hamburger MW. Seismological evidence for tearing of the Pacific Plate at the northern termination of the Tonga subduction zone. Geology (1998) 26:659–662.
Nash WP, Crecraft HR. Partition coefficients for trace elements in silicic magmas. Geochimica et Cosmochimica Acta (1985) 49:2309–2322. doi:10.1016/0016-7037(85)90231-5.[CrossRef][Web of Science]
Nicholls IA, Harris KL. Experimental rare earth element partition coefficients for garnet, clinopyroxene and amphibole coexisting with andesitic and basaltic liquids. Geochimica et Cosmochimica Acta (1980) 44:287–308. doi:10.1016/0016-7037(80)90138-6.[CrossRef][Web of Science]
Niu H, Sato H, Zhang H, Ito J, Yu X, Nagao T, Terada K, Zhang Q. Juxtaposition of adakite, boninite, high-TiO2 and low-TiO2 basalts in the Devonian southern Altay, Xinjiang, NW China. Journal of Asian Earth Sciences (2006) 28:439–456.[CrossRef][Web of Science]
Plank T, Langmuir CH. The chemical composition of subducting sediment and its consequences for the crust and mantle. Chemical Geology (1998) 145:325–394.[CrossRef][Web of Science]
Polat A, Kerrich R. Precambrian arc associations: boninites, adakites, magnesian andesites and Nb-enriched basalts. In: Precambrian Ophiolites and Related Rocks. Developments in Precambrian Geology. vol. 13—Kusky TM, ed. (2004) Amsterdam: Elsevier. 567–597.
Rapp RP, Shimizu N, Norman MD, Applegate GS. Reaction between slab-derived melts and peridotite in the mantle wedge: experimental constraints at 3·8 GPa. Chemical Geology (1999) 160:335–356.[CrossRef][Web of Science]
Reagan MK, Gill JB. Coexisting calcalkaline and high-niobium basalts from Turriala Volcano, Costa Rica: Implications for residual titanates in arc magma sources. Journal of Geophysical Research (1989) 94:4619–4633.
Robinson P. XRF analysis of flux-fused discs, Geoanalysis 2003. In: The 5th International Conference on the Analysis of Geological and Environmental Materials. Geological Survey of Finland and International Association of Geoanalysis—Carignan, et al, eds. (2003) Finland: Rovaniemi. Abstracts, p. 90.
Robinson P, Townsend AT, Yu Z, Munker C. Determination of scandium, yttrium and rare earth elements in rocks by high resolution inductively coupled plasma-mass spectrometry. Geostandards Newsletter (1999) 23:31–46.[Web of Science]
Rogers G, Saunders AD. Magnesian andesites from Mexico, Chile and the Aleutian Islands: implications for magmatism associated with ridge-trench collision. In: Boninites and Related Rocks—Crawford AJ, ed. (1989) London: Unwin Hyman. 416–445.
Rogers G, Saunders AD, Terrell DJ, Verma SP, Marriner GF. Geochemistry of Holocene volcanic rocks associated with ridge subduction in Baja California, Mexico. Nature (1985) 315:389–392.[CrossRef]
Rogers NW, MacLeod CJ, Murton BJ. Petrogenesis of boninitic lavas from the Limassol Forest Complex, Cyprus. In: Boninites and Related Rocks—Crawford AJ, ed. (1989) London: Unwin and Hyman. 288–313.
Schnetzler CC, Philpotts JA. Partition coefficients of rare-earth elements between igneous matrix material and rock-forming mineral phenocrysts; II. Geochimica et Cosmochimica Acta (1970) 34(3):331–340. doi:10.1016/0016-7037(70)90110-9.[CrossRef][Web of Science]
Shirey SB, Hanson GN. Mantle derived Archaean monzodiorites and trachyandesites. Nature (1984) 310:222–224.[CrossRef]
Sisson TW. Hornblende–melt trace element partititoning measured by ion microprobe. Chemical Geology (1994) 117(1–4):331–344. doi:10.1016/0009-2541(94)90135-X.[CrossRef][Web of Science]
Sisson TW, Bacon CR. Garnet high-silica rhyolite trace-element partition-coefficients measured by ion microprobe. Geochimica et Cosmochimica Acta (1992) 56(5):2133–2136. doi:10.1016/0016-7037(92)90336-H.[CrossRef][Web of Science]
Smith WHF, Sandwell DT. Global sea floor topography from satellite altimetry and ship depth soundings. Science (1997) 277:1956–1962.
Smith GP, Wiens DA, Fischer KM, Dorman LM, Webb SC, Hildebrand JA. A complex pattern of mantle flow in the Lau backarc. Science (2001) 292:713–716.
Sobolev AV, Danyushevsky LV. Petrology and geochemistry of boninites from the north termination of the Tonga Trench: constraints on the generation conditions of primary high-Ca boninite magmas. Journal of Petrology (1994) 35:1183–1211.
Stacey JS, Kramers JD. Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters (1975) 26:207–221.[CrossRef][Web of Science]
Staudigel H, Davies GR, Hart SR, Marchant KM, Smith BM. Large scale isotopic Sr, Nd and O isotopic anatomy of altered oceanic crust: DSDP/ODP sites 417/418. Earth and Planetary Science Letters (1995) 130:169–185.[CrossRef][Web of Science]
Sun S-S, McDonough WF. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. Magmatism in the Ocean Basins. Geological Society, London, Special Publications—Saunders AD, Norry MJ, eds. (1989) 42:313–345.
Tappin DR. The Tonga Frontal-arc basin. In: South Pacific Sedimentary Basins—Balance PF, ed. (1994) Amsterdam: Elsevier. 157–176.
Tappin DR, Bruns TR, Geist EL. Rifting of the Tonga/Lau Ridge and formation of the Lau backarc basin: evidence from site 840 on the Tonga Ridge. In: Proceedings of the Ocean Drilling Program, Scientific Results, 135—Hawkins J, Parson L, Allen J, et al, eds. (1994) College Station, TX: Ocean Drilling Program. 367–372.
Tatsumi Y, Eggins SM. Subduction Zone Magmatism (1993) Oxford: Blackwell. 224.
Taylor SR, Gorton MP. Geochemical application of spark source mass spectrography III. Element sensitivity, precision and accuracy. Geochimica et Cosmochimica Acta (1977) 41:1375–1380.[CrossRef][Web of Science]
Thorkelson DJ, Breitsprecher K. Partial melting of slab window margins: genesis of adakitic and non-adakitic magmas. Lithos (2005) 79:25–41.[CrossRef][Web of Science]
Turner SP, Hawkesworth CJ, Rogers NW, Bartlett J, Worthington TJ, Hergt JM, Pearce JA, Smith IEM. 238U–230Th disequilibria magma petrogenesis and flux rates beneath the depleted Tonga–Kermadec island arc. Geochimica et Cosmochimica Acta (1997) 61:4855–4884.[CrossRef][Web of Science]
Verbeeten A. Petrology, geochemistry and tectonic implications of magmatism along the northern Hunter Ridge and Kadavu Island group, Fiji. In: Ph.D. thesis (1996) Hobart: University of Tasmania.
Wiedenbeck M, Alle P, Corfu F, Griffin WL, Meier M, Oberli F, Vonquadt A, Roddick JC, Speigel W. 3 natural zircon standards for U–Th–Pb, Lu–Hf, trace-element and REE analyses. Geostandards Newsletter (1995) 19:1–23.[Web of Science]
Christie DM, Fisher CR, Lee S-M, Givens S. Mantle structure and flow patterns beneath active back-arc basins inferred from passive seismic and electromagnetic methods. Back-Arc Spreading Systems—Geological, Biological, Chemical, and Physical Interactions. American Geophysical Union, Geophysical Monograph (2006) 166:43–62.
Woodhead JD. A simple method for obtaining highly accurate Pb isotope data by MC-ICP-MS. Journal of Analytical Atomic Spectrometry (2002) 17:1–6.[CrossRef][Web of Science]
Workman RK, Hart SR, Jackson M, Regelous M, Farley KA, Blusztajin J, Kurz M, Staudigel H. Recycled metasomatized lithosphere as the origin of the enriched mantle II (EM2) end-member: Evidence from the Samoan Volcanic Chain. Geochemistry, Geophysics, Geosystems (2004) 5. doi:10.1029/2003GC000623.
Wright DJ, Bloomer SH, MacLeod CJ, Taylor B, Goodlife A. Bathymetry of the Tonga Trench and Forearc: a map series. Marine Geophysical Researches (2000) 21:489–511.[CrossRef][Web of Science]
Xiong X, Adam J, Green TH, Nui H, Wu J, Cai Z. Trace element characteristics of partial melts produced by melting of metabasalts at high pressures: constraints on the formation condition of adakitic melts. Science in China, Series D: Earth Sciences (2006) 49:915–925. doi:10.1007/s11430-006-0915-2.[CrossRef][Web of Science]
Yaxley GM, Green DH. Reactions between eclogite and peridotite: mantle refertilization by subduction of oceanic crust. Schweizerische Mineralogische und Petrographische Mitteilungen (1998) 78:243–255.
Yogodzinski GM, Kelemen PB. Slab melting in the Aleutians: implications of an ion probe study of clinopyroxene in primitive adakite and basalt. Earth and Planetary Science Letters (1998) 158:53–65.[CrossRef][Web of Science]
Yu Z, Robinson P, Townsend AT, Munker C, Crawford AJ. Determination of HFSE, Rb, Sr, Mo, Sb, Cs, Tl and Bi at ng g–1 levels in geological reference materials by magnetic sector ICP-MS after HF/HClO4 high pressure digestion. Geostandards Newsletter (2000) 24:39–50.[CrossRef][Web of Science]
Zamora D. Fusion de la croûte océanique subductée: approche expérimentale et géochimique. Université Thesis, Université Blaise Pascal, Clermont-Ferrand. (2000) 314.
Zellmer KE, Taylor B. A three-plate kinematic model for Lau Basin opening. Geochemistry, Geophysics, Geosystems (2001) 2. paper number 2000GC000106.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Schuth, C. Munker, S. Konig, C. Qopoto, S. Basi, D. Garbe-Schonberg, and C. Ballhaus Petrogenesis of Lavas along the Solomon Island Arc, SW Pacific: Coupling of Compositional Variations and Subduction Zone Geometry J. Petrology, May 1, 2009; 50(5): 781 - 811. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. V. Danyushevsky, T. J. Falloon, A. J. Crawford, S. A. Tetroeva, R. L. Leslie, and A. Verbeeten High-Mg adakites from Kadavu Island Group, Fiji, southwest Pacific: Evidence for the mantle origin of adakite parental melts Geology, June 1, 2008; 36(6): 499 - 502. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||