| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology Volume 43 Number 2 Pages 315-343 2002
© Oxford University Press 2002
Metasomatic Mantle Xenoliths from the Bismarck Microplate (Papua New Guinea)—Thermal Evolution, Geochemistry and Extent of Slab-induced Metasomatism
1INSTITUT FÜR MINERALOGIE, TU BERGAKADEMIE FREIBERG, BRENNHAUSGASSE 14, D-09596 FREIBERG, GERMANY
2GEOFORSCHUNGSZENTRUM POTSDAM, TELEGRAFENBERG, D-14473 POTSDAM, GERMANY
Received August 18, 2000; Revised typescript accepted August 20, 2001
 |
ABSTRACT |
|---|
A suite of ultramafic mantle xenoliths from the TUBAF and EDISON seamounts in the Bismarck Archipelago NE of Papua New Guinea was sampled by video-guided grab. The xenoliths, which were transported to the sea floor by rift-related, Quaternary trachybasalts, mainly represent part of the oceanic mantle. Mineral zoning in peridotite xenoliths testifies to slow cooling after mantle formation at a mid-ocean ridge. Cooling rates in the range of 1°C/Ma were calculated from zoning of Ca in olivine using the Lasaga algorithm. Subsequent to this cooling, a strong metasomatism affected the mantle peridotites when metasomatic agents emerged from the underlying slab of a subduction zone, which was stalled about 15 my ago. This resulted in the formation of orthopyroxene-, clinopyroxene-, phlogopite- and hornblende-bearing veins crosscutting spinel peridotites and olivine clinopyroxenites, as well as pervasively metasomatized plagioclase lherzolites. The metasomatic xenoliths reveal strong chemical disequilibria between the metasomatic minerals and the adjacent, unaltered host rock minerals, which are especially prominent in the veined samples. Temperatures during the metasomatic overprint, estimated using spinel–olivine thermometry, range between 660 and 950°C. Oxygen barometry reveals an elevated oxygen fugacity, with
log(fo2)FMQ values of 0·4 to >4. A geochemical study of the ultramafic rocks shows that all types of xenoliths have been metasomatized. Pervasively metasomatized plagioclase lherzolites and cumulate olivine clinopyroxenites have high contents of middle and heavy rare earth elements compared with veined peridotites. Cryptic metasomatism, indicated by increased light rare earth elements and Nd concentrations, results from LREE-rich hydrous fluid circulation. The investigated peridotites underwent a three-stage evolution from depleted oceanic ridge residues via repeated depletion to metasomatic imprint within a supra-subduction-zone setting. KEY WORDS: mantle metasomatism; ultramafic xenoliths; thermobarometry; geochemistry; Bismarck Archipelago–Papua New Guinea
|
INTRODUCTION |
|---|
Evidence for metasomatic processes in the Earth’s mantle is provided by trace element and isotope variations in volcanic rocks (e.g. Hawkesworth et al., 1987
), by metasomatized sections of ultramafic massifs (e.g. Bonatti et al., 1981) and by metasomatized mantle xenoliths of mainly peridotitic composition, which are transported to the surface by mafic or ultramafic magmas (e.g. Jagoutz et al., 1979). Two different styles of metasomatism are recorded in these rocks. Some ultramafic rocks show cryptic metasomatism, resulting in trace element enrichment without any visible mineral or textural changes (Dawson, 1984). The second type of metasomatism is characterized by petrographically well-recognizable replacement textures and/or the presence of secondary, hydrous minerals (e.g. phlogopite, hornblende) as well as secondary clinopyroxene, plagioclase, and accessory minerals such as apatite, Ti-magnetite, ilmenite, rutile and sphene. In some cases, the occurrence of glass in intergranular positions or as inclusions in primary mantle minerals is observed. This type of metasomatism is called patent (Dawson, 1984) or modal (Harte, 1983). Mantle xenoliths with modal metasomatism have been described from numerous locations: e.g. San Carlos, Arizona (Frey & Prinz, 1978); West Eifel and Rhön areas in Germany (Jagoutz et al., 1979; Stosch & Seck, 1980; Zinngrebe & Foley, 1995; Franz & Wirth, 1997; Witt-Eickschen & Kramm, 1997); Ahaggar, Central Sahara (Dautria et al., 1987); Mongolia (Ionov et al., 1994); Persani mountains, Romania (Vaselli et al., 1995); La Palma, Canary Islands (Wulff-Pedersen et al., 1996); southern Patagonia, Chile (Kempton et al., 1999a, 1999b). In some cases, modally metasomatized xenoliths contain veins, in which new mineral assemblages have been generated by metasomatic agents. These xenoliths are often referred to as composite types and are commonly characterized by strongly variable compositions of veins and host rock. Modal metasomatism is explained by the reaction of migrating H2O-rich fluids or silicate melts with the peridotite wall rock, whereas the formation of anhydrous, ultramafic veins is assigned to the crystallization of magma in the mantle [for a review of the mechanisms of modal metasomatism, see Kempton (1987) and Foley (1992)]. Although the mechanisms of metasomatism are relatively well understood, the origin of the metasomatizing agent is often enigmatic or controversial. In most cases, mantle fluids or melts are thought to be responsible for the mineralogical changes in the peridotite. Only in a few cases has evidence for mantle metasomatism induced by dehydration of a subducted slab been observed in xenoliths (e.g. Maury et al., 1992; Kepezhinskas et al., 1993, 1995; Szabó et al., 1996; Churikova et al., 1999; Draper et al., 1999; Grégoire et al., 2001; McInnes et al., 2001). We report here data for a suite of metasomatized mantle xenoliths, which originate from the mantle wedge above a stalled Tertiary subduction zone in the Bismarck Archipelago, NE of Papua New Guinea. This paper describes fluid–host rock reactions, thermal effects, the geochemical character of the metasomatic agent as well as its oxygen fugacity, and the extent of the metasomatism, based on xenoliths that were erupted to the sea floor within Quaternary mafic lavas. |
GEOLOGICAL SETTING |
|---|
The Bismarck Archipelago is located NE of the island of Papua New Guinea and includes the islands of New Britain, New Ireland–New Hanover, Bougainville and the Solomons. The plate tectonic situation in this area involves southwestward subduction of the Pacific plate beneath the Bismarck microplate at the Manus–Kilinailau trench (Fig. 1). Concurrent with this subduction, voluminous calc-alkaline arc volcanism occurred during the Oligocene and Miocene. About 15 my ago, the trench was blocked by its collision with the Ontong–Java plateau (Coleman & Kroenke, 1981
). This resulted in plate rotation and stress relocation, and led to subduction reversal with the formation of the currently active NNW- to NNE-dipping New Britain trench. Today the Solomon Sea microplate is being subducted below the Bismarck microplate. Back-arc spreading occurred in the Manus basin contemporaneous with this subduction and separated the Bismarck microplate into a northern and a southern segment (Fig. 1). According to investigations on gabbros (McInnes et al., 1994), the oceanic crust of the Bismarck microplate formed in a mid-ocean ridge tectonic setting. Re–Os data point to a model age of 
120 Ma for the lithosphere of the Bismarck microplate, suggesting that the Cretaceous Pacific plate was subducted below its own detached fragment (McInnes et al., 1999).
|
The Tabar–Lihir–Tanga–Feni volcanic island chain extends for a distance of >260 km NE of New Ireland. About 3·6 my ago, volcanic activity started on Simberi island (Tabar island group) in the New Ireland fore-arc region (McInnes, 1992; Rytuba et al., 1993). This high-K calc-alkaline volcanism seems to be related to lithospheric extension along NE-trending faults, which were generated during the opening of the Manus back-arc basin (Taylor, 1979; Wallace et al., 1983; Steward & Sandy, 1988; Kennedy et al., 1990). The most recent eruption on land occurred 2·3 ky ago on Feni island (Licence et al., 1987). The genesis of these magmas is connected with thinning of the New Ireland basin lithosphere, which led to adiabatic decompression melting of underlying mantle regions. Melt generation was triggered by strong metasomatic influence from the Pacific plate subducted along the Manus–Kilinailau trench (McInnes & Cameron, 1994).
In the course of two cruises with the R.V. Sonne (cruises SO-94 and SO-133), 12 submarine volcanoes were detected near Lihir island with the help of the hydrosweep sonar method. These were sampled using dredge and video-guided grab (Herzig et al., 1994, 1998). The steep morphology of these seamounts, the pristine samples and the absence of covering marine sediments point to a very young age for this volcanism (Herzig & Becker, 1996; Herzig et al., 1998). Samples from two of these volcanoes, TUBAF and EDISON seamounts (Fig. 1), contain abundant xenoliths from the oceanic crust and the underlying mantle. The ultramafic mantle xenoliths in particular show a distinct metasomatic overprint as a consequence of infiltration of fluids from the underlying subducted slab (McInnes et al., 2001).
|
PETROGRAPHY OF THE XENOLITHS |
|---|
The following rock abbreviations are used in the text and the tables: SL, spinel lherzolite (sample 56-2H); SD, spinel dunite (sample 54-2A); SH, spinel harzburgite (samples 56-2P and 56-2T); SHOV, spinel harzburgite with orthopyroxene veins (sample 56-2A); SLOHV, spinel lherzolite with orthopyroxene–hornblende veins (sample 56-2B); SLPCOV, spinel lherzolite with phlogopite-bearing clinopyroxene–olivine veins (sample 54-2D); SLGPV, spinel lherzolite with garnet–phlogopite veins (sample 56-2X); OCPHV, olivine clinopyroxenite with phlogopite–hornblende veins (sample 33-2A); SHIM, spinel harzburgite with irregular metasomatism (sample 54-2H); PLPM, plagioclase lherzolite with pervasive metasomatism (samples 55-2C and 56-2M); TB, trachybasalt (host rock, sample 56-5A). Mineral abbreviations in the text and the tables are those of Kretz (1983)
and Cel, celsian; Crb, carbonate; Sulph, sulphide; Uvar, uvarovite.
Ultramafic xenoliths without obvious or with very little metasomatic alteration
Most of the studied mantle xenoliths are pristine or only slightly metasomatized spinel peridotites with coarse equant textures in the sense of Harte (1977). The modal compositions of these xenoliths vary widely from spinel dunites (SD) and spinel harzburgites with very low amounts of clinopyroxene (SH) to spinel lherzolites (SL). Metasomatism is observed in only a few locations, e.g. where phlogopite forms small rims around primary spinel. A remarkable feature of these xenoliths is the occurrence of numerous ultrafine spinel needles, which are included in parts of the olivine grains. These needles, which typically have a length of 5–25 µm and a width of 150–200 nm, are found in specific crystallographic orientations within the olivine and are interpreted to have formed by exsolution processes during cooling (Franz & Wirth, 2000).
Ultramafic xenoliths with modal metasomatism
Three main groups of metasomatized ultramafic xenoliths have been distinguished, as follows.
Veined ultramafic xenoliths
These are the most common type of metasomatized samples. Most of these display coarse equant textures, which are cut by isolated and, in some cases, network-forming veins. As already observed within the non- or poorly metasomatized ultramafic peridotites, veined peridotites commonly show inclusions of tiny spinel needles within olivine. The relationship between these inclusions and the metasomatic veins, however, cannot be established. On the basis of their mineral assemblages and textural characteristics, the following subgroups of veined xenoliths can be distinguished.
- Spinel harzburgites with orthopyroxene veins (SHOV; Fig. 2a) are coarse equant peridotites crosscut by veins of up to 1 mm width in which growth of secondary orthopyroxene crystals at the expense of olivine and primary orthopyroxene has occurred. The veins consist almost exclusively of secondary orthopyroxene, which predominantly forms randomly radiating, sheaf-like bundles but occasionally shows granoblastic shapes. In a few locations, small irregular grains of Ni–Fe monosulphide solid solution are intergrown with or enclosed in secondary orthopyroxene. Distinct changes have occurred to the other primary minerals within or near the veins: clinopyroxene is altered to a fibrous mass of hornblende, whereas spinel is embayed and corroded at its rims. Near the veins, olivine shows wavy extinction and sometimes dynamic recrystallization to small olivine neoblasts with triple junctions, providing evidence—at least in part—for ductile deformation during the formation of the veins.
- Spinel lherzolites with orthopyroxene–hornblende veins (SLOHV; Fig. 2b) have primary granular, coarse equant textures and are crosscut by small veinlets (usually <0·5 mm) that are filled with needle-shaped hornblende crystals. In some cases, coarse-grained secondary orthopyroxene prisms and rims are present on primary olivine. In some parts of veins, hornblende is recrystallized to millimetre-sized, euhedral grains. Spinel near the veins shows corrosion, whereas the opaque phase within the veins is Ni–Fe monosulphide solid solution. Although the wall rock has preserved its primary texture, metasomatic features can also be observed there. Secondary hornblende is present along the cleavage planes of clinopyroxene, whereas phlogopite forms rims around spinel and interstitial small flakes near spinel. At the contact of the veins with the host basalt, Ti-augite appears as panidiomorphic overgrowths on the amphibole needles of the veins.
- Spinel lherzolites with phlogopite-bearing clinopyroxene–olivine veins (SLPCOV; Fig. 2c) are characterized by primary, coarse equant textures crosscut by veins with prominent evidence of dynamic recrystallization. Especially olivine and sometimes secondary clinopyroxene form small neoblasts with triple junctions. Primary and secondary clinopyroxenes are altered to fine needles of hornblende, whereas primary orthopyroxene shows an overgrowth of cummingtonite. In many parts of the veins, small flakes of phlogopite are found, which are bent and bleached in sections of strong deformation.
- A single sample of coarse equant spinel lherzolite with a garnet-bearing phlogopite vein (SLGPV) is characterized by a unique secondary mineral assemblage and remarkable textural features. The vein has a width of 5 mm and mainly consists of xenoblastic phlogopite, which encloses idioblastic amphibole crystals and isolated grains of secondary clinopyroxene. In parts of the vein, a skeletal intergrowth of plagioclase and garnet is observed (Fig. 3a and b). Primary silicates appear as armoured relics in the vein, whereas spinel is missing. Ni–Fe monosulphide solid solution is the only opaque phase in the vein. Numerous fluid inclusions, with diameters of <1 µm, are found in primary mantle minerals near, as well as within, the vein. Metasomatic alterations are also observed in the wall rock, where primary clinopyroxene is rimmed by hornblende and small flakes of secondary phlogopite appear at the grain boundaries of olivine.
- The texture of olivine clinopyroxenite with phlogopite–hornblende veins (OCPHV; Fig. 2d) is characterized by an intergrowth of granular to short prismatic clinopyroxene grains (>2 mm in length) and subordinate, granular olivine (<0·5 mm in diameter) in the wall rock. Primary clinopyroxene contains numerous lamellae of hornblende and spinel. Veins crosscutting the rock typically have a width of 1–2 mm and are dominated by euhedral, olive–green hornblende grains and green phlogopite tablets. Phlogopite is also very common at grain boundaries of the primary minerals in the wall rock. Secondary carbonate mainly mantles phlogopite and hornblende in the veins, but also appears in microfractures and at grain boundaries of primary olivine and clinopyroxene in the wall rock.
|
|
Spinel harzburgites with irregular metasomatism (SHIM; Fig. 2e)
These are characterized by coarse equant textures and metasomatic alteration in centimetre-wide, irregular sections or isolated spots. There, primary pyroxenes are replaced along their rims and along cleavage planes by colourless hornblende. Dynamic recrystallization of primary olivine can also be found in these areas, whereas vein-like features are missing. In sections near the host basalt, needle-shaped secondary ortho- and clinopyroxene and a new generation of small olivine grains have grown at the expense of the primary silicates.
Plagioclase lherzolites with pervasive metasomatic features (PLPM; Fig. 2f)
These show a strong, penetrative alteration of the primary mantle minerals. The primary mineral assemblage of these coarse equant lherzolites consists of granular ortho- and clinopyroxene, equigranular olivine, plagioclase and angular grains of opaque spinel. Metasomatic alteration mainly affects primary plagioclase, which appears only as rare relics rimmed by faintly brownish hornblende. Plagioclase is commonly completely replaced by colourless hornblende and green spinel. Partial, and in some cases total, replacement of primary clinopyroxene by brownish hornblende is also observed.
|
MINERAL CHEMISTRY |
|---|
Mineral analyses were performed at the University of Freiberg (Germany) using a JEOL JXA-8900R electron microprobe with five spectrometers. Major and minor elements were determined at 15 kV acceleration voltage and a beam current of 20 nA with counting times of 20 s for Si, Al, Mg, Ca, Sr, Ba and K, and 30 s for Fe, Ni, Na, Cr, Mn and Ti. Ca in olivine was determined with an acceleration voltage of 20 kV, a beam current of 55 nA, and a counting time of 300 s. The accuracy of the analyses was checked with the SC/KA olivine standard, which contains 524 (±5) ppm Ca, and the N1 standard, which contains 19·6 (±3) ppm Ca (see Köhler & Brey, 1989). The JEOL microprobe reproduced these values within an error range of ±10 ppm. The standard sets of the Smithonian Institution (see Jarosewich et al., 1980
) and of MACTM (Micro-Analysis Consultants Ltd, UK) were used for reference. Selected microprobe analyses of the representative minerals are listed in Tables 1–7
. The entire dataset is available from the first author on request.
|
|
|
|
|
|
|
Primary orthopyroxene from the TUBAF peridotites has an enstatite (En) component of 84–92 mol % and contains low amounts of Al2O3 (0·5–2·2 wt %) and Cr2O3 (0·2–0·5 wt %). CaO contents range from 0·15 to 1·1 wt %, corresponding to a wollastonite component (Wo) of 0·2–2·1 mol %. In 50% of the peridotites, orthopyroxene contains numerous exsolution lamellae (1–5 µm thick) of clinopyroxene and subordinately of spinel. Analysing these grains using a 50 µm microprobe beam consequently leads to higher Al, Cr and Ca contents (Table 1). Chemical zoning is present in all types of orthopyroxene: microprobe profiles (avoiding exsolution lamellae) reveal in most cases decreasing Ca, Al and Cr contents and increasing Mg contents from core to rim. Only in the SHIM is a reverse trend in concentration observed. Remarkably, the composition of primary orthopyroxene is not influenced by the occurrence of metasomatic veins unless they display replacement textures at the expense of orthopyroxene.
Metasomatic orthopyroxene in veins always has a higher En component compared with the primary orthopyroxene in the same sample, i.e. 90–93 mol % vs 88–90 mol % (compare samples 56-2A and -2B in Table 5). Similarly, its Al, Cr and Ca contents are distinctly lower. It shows irregular zonation patterns, which are especially prominent with respect to the En component (ranging, for example, from 89 to 94 mol % in sample 56-2A). Exsolution lamellae and mineral inclusions are absent. Needle-shaped orthopyroxene in the SHIM (see sample 54-2H) has high Al and Cr contents, an En component of 88–91 mol % and a Wo component of 0·5–6·9 mol %. The composition varies from needle to needle.
Primary clinopyroxene from the TUBAF and EDISON xenoliths is Cr-diopside with Wo41–48, En49–56 and Fs2–7 [calculation after Lindsley (1983)]. In xenoliths with exsolved primary orthopyroxene, clinopyroxene also shows exsolution features (see integrated clinopyroxene in Table 2, which shows higher contents of Al, Cr, En and a lower Wo component). Clinopyroxene contains low amounts of Al2O3 (0·8–2·6 wt %) and Cr2O3 (0·2–1·1 wt %), and its total cation proportions point to a very low concentration or a complete absence of ferric iron. With the exception of the olivine clinopyroxenite and the spinel dunite (samples 33-2A and 54-2A), chemical zoning is observed in nearly all investigated clinopyroxene crystals. Toward the rims of the grains, a distinct increase in Al, Cr and Wo, and an increase in En is observed. Primary clinopyroxene adjacent to veins only rarely shows evidence for metasomatic alteration. In the SLGPV (sample 56-2X), relic clinopyroxene within the phlogopite vein yields higher contents of Al and Fe3+ and a slightly higher Wo content compared with the clinopyroxene in the wall rock. Similar features are recorded in clinopyroxene near the hornblende–phlogopite vein of the olivine clinopyroxenite.
The composition of dynamically recrystallized, metasomatic clinopyroxene neoblasts in the veins of the SLPCOV is Wo46–50, En48–51 and Fs1–5, and they are somewhat higher in Ca and lower in Mg than the primary clinopyroxene (Table 5). The most striking difference between primary and secondary clinopyroxene is the very low Cr2O3 contents of the neoblasts (0·0–0·2 wt % vs 0·4–0·6 wt %). The idiomorphic, short prismatic clinopyroxene crystals in the phlogopite vein of the SLGPV are Wo33–34, En58–59 and Fs7·7–7·9, and contain rather low Cr2O3 contents. Strongly variable compositions (Wo25–38, En58–71 and Fs3·2–4·5) are observed in different metasomatic clinopyroxene needles of the SHIM. They display Al2O3 contents of 0·5–3 wt % and high Cr2O3 contents (0·4–0·7 wt %). Similar to secondary orthopyroxene, metasomatic clinopyroxene does not show exsolution features.
Primary olivine crystals in the peridotites are Fo85–92; Fo increases with increasing En contents of the coexisting orthopyroxene (see Table 3). Zoning of the Fo component is not observed and olivine near the metasomatic veins usually shows the same composition as grains distant from the vein. The Ca content of olivine in the peridotites ranges from 0 to 250 ppm, with distinctly variable zonation patterns from sample to sample. Primary olivine crystals from most peridotites—regardless of whether or not they experienced modal metasomatism—show symmetrical zonation patterns with a decrease of Ca towards the rim section of the grains (Fig. 4). The Ca concentrations in olivine grains in direct contact with the metasomatic veins have been reset within a narrow band close to the vein, although the more typical zonation pattern is still recognizable in the parts of the grain that are most distant from the vein (Fig. 5). Olivine near the contact to the host basalt shows an abrupt increase in Ca content at the rim, which can be assigned to late thermal input and decompression. The lowest Mg contents (84 mol % Fo) and the highest Ca contents (500 ppm in the core) are found in olivine of the olivine clinopyroxenite. A strong increase in Ca towards the rim of these olivine grains has been related to secondary fluorescence from the neighbouring Cpx.
|
|
Olivine neoblasts, which were generated during syn-metasomatic deformation, mostly show Fo contents that are 0·4–0·6 mol % higher than in the primary grains. The Ca contents of the neoblasts are often high (i.e. in some cases up to three times higher than those of the primary olivine crystals). A distinctly different composition is displayed by the olivine grains that developed at the expense of primary orthopyroxene in the SHIM (sample 54-2H). They have distinctly lower Mg contents than the primary olivine grains (Fo88·3 vs Fo91·2) and show somewhat higher Ca contents (Table 5).
Primary spinel octahedra in the matrix of the peridotites are Cr rich with XCr [Cr/(Cr + Al)] values of 0·44–0·72 and XMg [Mg/(Mg + Fe2+)] values of 0·37–0·71 (Table 4). Most spinel crystals display chemical zoning, with decreasing XCr and increasing XMg values from core to rim. This pattern, however, is reversed in some of the xenoliths with cryptic metasomatism, which contain low ferric iron contents (12–24% Fe2O3). Spinel from xenoliths with modal metasomatism displays higher Fe3+ concentrations (30–51% Fe2O3) and is also characterized by an increase in Fe3+ from core to rim. The most prominent chemical zoning is present in spinel from metasomatic domains, where an increase in XMg and a decrease in XCr from core to rim is observed. These spinel grains contain the highest ferric iron contents (up to 60% Fe2O3). In metasomatic domains of the plagioclase lherzolites primary, Cr-rich, opaque spinel (XCr = 0·63) shows a continuous chemical transition to secondary, green hercynite. In the olivine clinopyroxenite, spinel occurs only as rare inclusions in clinopyroxene and is characterized by XCr values of 0·63–0·64, XMg values of 0·45–0·46 and elevated ferric iron contents (42–51% Fe2O3; Table 4).
Metasomatic spinel has a totally different chemical composition compared with the Cr-rich, primary types. Green spinel, formed at the expense of primary plagioclase in the plagioclase lherzolites, is hercynite. The small grains show no chemical zoning and contain virtually no Cr (XCr <0·03). Their XMg values range from 0·73 to 0·88; the variation is mainly due to the variable Fe3+ content (19–40% of total iron; Table 7). Within the phlogopite vein of the SLPCOV, secondary spinel displays high amounts of Fe2O3 (corresponding to a magnetite component of up to 45 mol %) and rather low concentrations of Al (XCr = 0·82–0·94; Table 7).
Primary plagioclase from the plagioclase lherzolites is Ca rich (An contents of 88–99·8 mol %; Table 7) with low amounts of orthoclase (up to 0·2 mol %) and celsian (up to 0·1 mol %). Chemical zoning is observed only within larger plagioclase grains, where An contents show a distinct decrease from core to rim.
Metasomatic plagioclase in veins of the SLGPV is almost pure anorthite (An97·4–99·9Ab0–2·4Or0–0·7Cs<0·1; Table 7).
Metasomatic amphibole is mainly calcic [classification of Leake et al. (1997)] with high amounts of Al (Table 6). In veined peridotites and olivine clinopyroxenite, magnesiohastingsite, magnesiohornblende and minor pargasite are observed. The composition of the clinoamphibole from veined xenoliths is often strongly variable in the same sample. The most prominent variations are observed in the olivine clinopyroxenite, where pargasite and magnesiohastingsite in veins display high Ti and Al contents. Hornblende lamellae, which formed in the cleavage planes of primary clinopyroxene, are distinctly lower in Ti and Al and higher in Si. Clinoamphibole with strongly variable composition is also observed within the plagioclase lherzolites, where brownish, Ti-rich magnesiohastingsite and pargasite occur mantling primary plagioclase, and colourless Ti- and Na-poor magnesiohornblende, tschermakite (sample 56-2M) or tremolite (sample 55-2C) are present in the core of the plagioclase. In sample 54-2D, cummingtonite was observed forming small rims around primary orthopyroxene. F and Cl concentrations are very low in all amphiboles (<0·04 wt %).
Metasomatic phlogopite in peridotites yields XMg values of 0·9–0·94 and contains low amounts of TiO2 (0·01–0·22 wt %; Table 6). In the olivine clinopyroxenite, phlogopite in the metasomatic veins yields an XMg of 0·84–0·85 and high amounts of TiO2 (1·0–1·2 wt %), whereas phlogopite occurring at grain boundaries of the primary minerals contains higher amounts of Mg (XMg = 0·88–0·89) and lower amounts of TiO2 (0·5–0·6 wt %). All investigated phlogopite crystals are almost free of F and Cl (<0·02 wt %).
Metasomatic garnet in veins of the SLGPV is Ca rich (Andr7·2–15·3Grs82·1–88·5Alm0·8–2·7Prp0·4–1·8Sps0·1–0·3; Table 7). Variations of their composition are almost entirely due to the strongly variable amount of ferric iron, which ranges from 76 to 89%.
Metasomatic carbonate in veins of the olivine clinopyroxenite is characterized by a high CaCO3 component (up to 99·95 mol %) and minor amounts of MgCO3 (up to 2·1 mol %) and FeCO3 (up to 0·5 mol %). BaO and SrO contents are very low (<0·1 wt %; Table 7).
Metasomatic sulphide, occurring in the veins crosscutting the peridotites, is mainly Ni–Fe monosulphide solid solution with elevated Ni contents (Ni0·38–0·49Cu0–0·08Co0–0·01Fe0·40–0·49S). Rare Ni-poor monosulphide solid solution (Ni0·04–0·07Cu0–0·01Co0–0·005Fe0·78–0·87S) appears in the phlogopite vein of the SLOHV. Charge balance constraints suggest ferric iron contents of 17–56% of the total iron in the sulphide minerals (Table 7).
|
EQUILIBRIUM CONDITIONS AND THERMOBAROMETRY OF THE PRIMARY MINERAL PARAGENESES |
|---|
Estimation of the P–T conditions of equilibration of the xenoliths based on the composition of the primary mantle minerals is problematic. Orthopyroxene commonly shows a continuous decrease in Ca, Al and Cr from core to rim, whereas clinopyroxene shows an increase of Ca and a decrease of Mg from core to rim. This suggests diffusion as a result of cooling processes (see Lindsley, 1983
). Cooling is also evident by the exsolution features in ortho- and clinopyroxene, by the spinel exsolution in olivine (see Franz & Wirth, 2000), and by the zoning in olivine with continuously decreasing Ca contents from core to rim. At constant pressure, temperatures calculated with the Opx–Cpx geothermometer [T BKN calibration of Brey & Köhler (1990)] using core analyses of pyroxenes are 60–150°C higher than temperatures calculated using rim analyses (for P–T estimates, errors and description of the thermobarometers, see Table 8). Textural constraints and zonation patterns in olivine (Fig. 5) indicate that these cooling processes clearly pre-date the metasomatic overprint. Therefore, it is most likely that cooling occurred in the early history of the oceanic mantle, when the newly formed lithosphere drifted away from the mid-ocean ridge (see Pollack & Chapman, 1977; Rabinowicz et al., 1984).
|
|
Analyses of primary mantle minerals from the central part of spinel peridotite xenoliths without obvious modal metasomatic alteration, as well as analyses of minerals remote from altered sections in modally metasomatized spinel peridotites, were used to unravel the early thermal history of the mantle lithosphere of which they formed a part. Thermometry of the spinel lherzolites was performed using analyses from the cores of ortho- and clinopyroxene crystals as well as integrated analyses of pyroxenes with exsolution lamellae (i.e. mapping of the exsolved pyroxene grain using a defocused microprobe beam with a diameter of 50 µm). The thermometric results of the T BKN, T Ca–Opx and T Na–Px calibrations of Brey & Köhler (1990)
were then combined with the P KB barometry of Köhler & Brey (1989) using the Ca contents of the cores of the largest olivine grains. These P–T calculations led to unrealistically high pressures at elevated temperatures. A combination of T BKN and P KB, for example, results in a P–T range from 881°C at 39 kbar to 1093°C at 122 kbar for the modally metasomatized spinel peridotites, and from 919°C at 43 kbar to 1016°C at 116 kbar for the spinel peridotites without modal metasomatic alteration (see Table 8). This is in contradiction to the spinel- and plagioclase-facies mineralogy of the peridotites, realistic lithosphere thickness and the geochemistry of the host trachybasalts. The trachybasalts show no depletion of heavy rare earth elements (HREE), indicating that they were not in equilibrium with garnet in their mantle source (M. Perfit, unpublished data, 2000; see also Fig. 8). Thus, to be in agreement with the occurrence of spinel and the absence of any high-pressure relics and symplectitic garnet breakdown textures, TUBAF and EDISON xenoliths should originate from the stability field of spinel lherzolite and therefore cannot have experienced pressures of re-equilibration higher than 18–20 kbar (e.g. O’Neill, 1981; O’Neill & Wall, 1987).
|
The unrealistically high pressure estimates are related to the strongly different diffusion rates of the elements used for thermobarometry in pyroxenes and olivine. The lowest diffusion rates are recorded for Ca and Mg in pyroxene, whereas diffusion of Ca (as well as Fe and Mg) in olivine is significantly faster (see Brady, 1995). During cooling, the fast diffusion of Ca in olivine results in a distinct shift of the steep KD line of the P KB calibration towards lower temperatures and higher pressures. Thermal re-equilibration with respect to the T BKN geothermometer is hampered by the slow diffusion of Ca and Mg in pyroxene and thus the thermometric memory of the pyroxenes is distinctly ‘older’ than the barometric information given by the Ca in olivine, leading to an intersection of the two KD lines at elevated pressures. The same holds true for most P–T calculations using other geothermometers (e.g. a combination of T CaOpx or T NaOpx with P KB). Similar observations of mineral disequilibria were made by Zipfel & Wörner (1992) in peridotite xenoliths from the Ross Sea.
Another approach was chosen to unravel the P–T history of the spinel peridotites. According to the compilation of regional geotherms of Pollack & Chapman (1977), the recent heat flow in the New Ireland area is 60 W/m2 and the mantle xenoliths should have re-equilibrated to that state before they were erupted with the trachybasalts. Therefore, analyses of the rims of the pyroxenes, which show maximum cooling, were used to calculate the KD lines for the recent equilibrium temperatures. Temperature and pressure estimates for the samples are given where those KD lines intersect the oceanic 60 mW/m2 geotherm. According to this procedure, the P–T estimates with the T BKN method for the spinel lherzolites plot between 665°C at 13 kbar and 860°C at 18 kbar, which can be considered as realistic given the plate tectonic situation and the depth of formation of the host magma (Fig. 6). Distinctly lower pressures are obtained if the rim compositions of spinel and olivine remote from metasomatic areas are chosen for thermometry. Because of the high diffusion rates for Fe and Mg in spinel and olivine, these minerals re-equilibrate much faster during cooling processes, which results in distinctly lower temperatures for the T BBG geothermometer (up to 190°C lower than obtained by the T BKN method). Therefore, an intersection of the KD lines of the T BBG geothermometer with the oceanic geotherm results in a decrease of pressure of 5 kbar.
For the plagioclase lherzolites, the T BKN thermometry was combined with the XMgTs barometer of Gasparik (1984), which resulted in P–T conditions of 773°C at 2·0 kbar (sample 55-2C) and 790°C at 3·0 kbar (sample 56-2M). This indicates a rather high geothermal gradient for these two samples, which is probably caused by the failure of the samples to achieve thermal equilibration in the upper levels of the lithosphere. The same applies to the olivine clinopyroxenite (sample 33-2A), whose P–T estimate of 953 (±30)°C at 1·3 (±4) kbar was obtained using a combination of T BBG and P KB (Fig. 6). The petrology, thermobarometry and geochemistry (see below) of the olivine clinopyroxenite classify this rock as a cumulate at the base of the oceanic crust or even within a crustal magma chamber.
|
THERMAL CONDITIONS AND MINERAL REACTIONS DURING THE METASOMATIC OVERPRINT |
|---|
 TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGPETROGRAPHY OF THE XENOLITHSMINERAL CHEMISTRYEQUILIBRIUM CONDITIONS AND... THERMAL CONDITIONS AND MINERAL... GEOCHEMISTRYDISCUSSIONCONCLUSIONSAPPENDIXREFERENCES |
|---|
Orthopyroxene and orthopyroxene–hornblende veins
There is textural and mineralogical evidence that the metasomatic veins formed via reaction of percolating fluids or melts with the wall rock. In case of the orthopyroxene veins, this reaction can be expressed as
 |
This reaction, however, is strongly dependent on the SiO2 activity of the fluid phase, for which we have no indication of composition. According to the experiments of Kitahara et al. (1966), the paragenesis forsterite + enstatite + H2O is stable above temperatures of 700°C in the stability field of spinel lherzolite. The upper temperature limit for the formation of the orthopyroxene veins is constrained by the wet solidus of peridotite, i.e. temperatures below 1000°C at pressures of 5–20 kbar (see Wyllie, 1981; Olafsson & Eggler, 1983). A more precise determination of the formation temperatures of the veins using the composition of the secondary orthopyroxene is not possible. There are no equilibrium grain boundaries between primary clinopyroxene and orthopyroxene from the vein; wherever primary clinopyroxene is in contact with the vein, it has been replaced by hornblende. A further argument for disequilibrium is given by the results of the three pyroxene geothermometers of Brey & Köhler (1990), which should be equal within the error of their calibrations in the case of mineral chemical equilibrium. Using the average composition of the secondary orthopyroxene and the average composition of the primary clinopyroxene of sample 56-2A, the calculation at a fixed pressure of 15 kbar yields 944°C for T BKN, 727°C for T CaOpx and 1260°C for T NaOpx. These values, as well as the wide variability in thermometric results, which are due to inhomogeneous mineral compositions, clearly highlight the absence of equilibrium (Table 8).
Another approach to determining the temperature of vein formation is to use Fe–Mg exchange between zoned spinel and adjacent olivine neoblasts, which recrystallized during vein formation. The application of the T BBG calibration yields temperatures of 900–950°C using the rim composition of spinel. Calculations based on the core compositions of spinel and primary olivine give much lower temperatures of 650–680°C, which are interpreted to represent the former state of cooling before the metasomatic overprint. This demonstrates a two-stage thermal history of the mantle lithosphere, i.e. cooling after its formation, possibly at the mid-ocean ridge or a back-arc basin, and subsequent heating related to the metasomatic event affecting limited areas adjacent to veins.
Another specific feature of the vein-forming metasomatic event is a distinct increase in oxygen fugacity. According to the oxygen geobarometer of Ballhaus et al. (1991), the (fO2)FMQ values (where FMQ is the fayalite–magnetite–quartz buffer) obtained using the rim section of the zoned spinel and the olivine neoblasts range between 0·8 and 1·4, whereas the core of the spinel combined with primary olivine yields log(fO2)FMQ values near the FMQ buffer (-0·4 to -0·2). In this context, the widespread occurrence of secondary sulphides (Ni–Fe monosulphide solid solution) in the veins is noteworthy. These minerals formed by a reaction of sulphur (from the percolating fluid) with Ni and Fe, which were released by the decomposition of primary olivine. The stability of Ni–Fe monosulphide solid solution at the observed high oxygen fugacity may be explained by the incorporation of prominent amounts of Fe3+ in the sulphide (Table 7).
The crystallization of hornblende in the orthopyroxene–hornblende veins, as observed in sample 56-2B, may be explained by the reactions
 |
 |
(mineral formulae used for the balancing are listed in the Appendix). Again, these reactions are strongly dependent on the activities of SiO2 and H2O in the fluid phase. An estimate of the temperature of formation for the orthopyroxene–hornblende veins using conventional pyroxene thermometry is hampered by evident mineral chemical disequilibria between primary and secondary pyroxene. The calculation of the Fe–Mg exchange equilibria of olivine and zoned spinel in the vicinity of the veins leads to temperatures of 750–780°C at P = 15 kbar using the T BBG calibration (Table 8). Another possibility for estimating the formation temperature uses the zoning of Ca in olivine near the metasomatic vein of sample 56-2B (Fig. 5). The slope of the P KB geobarometer in the P–T space is rather steep, which, in the case of well-known pressures, permits the use of this calibration as a geothermometer. At mantle pressures of 15–20 kbar, corresponding to the depth of formation of the host magma, temperatures of 730–760°C are calculated with P KB, which is very similar to the data gained with the T BBG calibration. An elevated oxygen fugacity with log(fO2)FMQ ranging between 1·2 and 1·4 is observed in the vicinity of the vein using the oxygen geobarometer of Ballhaus et al. (1991).
Phlogopite–clinopyroxene veins
The formation of phlogopite in these veins (SLPCOV, sample 54-2D) can be explained only by the addition of K2O from the fluid or melt phase according to the following reactions:
 |
 |
Secondary clinopyroxene (Cpx II, rich in Ca), which may have formed at the expense of the primary, Mg-rich clinopyroxene (Cpx I), points to much more complex cation exchange processes.
Similar to the other xenoliths with modal metasomatism, pronounced mineral disequilibria are observed in the SLPCOV. The Cpx II neoblasts display compositions varying from one part of the metasomatic veins to the other, which results in a wide temperature range of 260–720°C using T BKN thermometry. There is no equilibrium between Cpx II and primary orthopyroxene, as evidenced by the strongly differing temperatures obtained by T CaOpx (814–950°C) and T NaOpx (1270–1570°C). An application of T BBG to spinel and olivine neoblasts in the vein results in a small range of temperatures of 820–850°C for different mineral pairs. Temperatures in excess of 850°C are also calculated using the olivine neoblasts with the highest Ca contents and applying the P KB calibration as a geothermometer (P = 15 kbar). Again, an elevated oxygen fugacity with log(fO2)FMQ of 2·8–4·3 was observed.
Garnet-bearing phlogopite veins
These veins have the mineral assemblage phlogopite–amphibole–plagioclase–garnet–clinopyroxene II, which must have formed by complex cation exchange processes and strong metasomatic interaction. Phlogopite and amphibole may have formed according to reactions (2)–(5), whereas the growth of the grossular-rich garnet and plagioclase may be due to the reaction
 |
P–T conditions during the vein-forming event may be estimated using local chemical equilibria in the vein. Calculations using the average composition of the garnet and the composition of clinopyroxene II yield temperatures of 708–763°C (Krogh, 2000), whereas garnet–hornblende geothermometry gives temperatures of 700–800°C [Graham & Powell, 1986; Fe3+ in amphibole according to the midpoint method of Papike et al. (1974)]. These temperature estimates are in accord with the hornblende–plagioclase geothermometry of Spear (1980), which indicates temperatures in excess of 725°C. Temperatures of 690–740°C are obtained applying the T BBG thermometer on zoned spinel and olivine in the vicinity of the vein, whereas oxygen barometry indicates log(fO2)FMQ values of 1·3–2·5.
The occurrence of garnet in the vein enables us to estimate the pressure for the metasomatic event. Assuming a local equilibrium in the vein according to the reaction
 |
can be performed, which yields pressures of 7–8 kbar. As a result of the absence of quartz in the paragenesis and because there is no estimate for the activity of SiO2 in the metasomatic agent, this has to be regarded as a maximum pressure estimate. As this vein developed in a spinel lherzolite, the lower pressure limit for this paragenesis is the plagioclase-out reaction in peridotite, which is located at 4–5 kbar (T = 700–800°C; Gasparik, 1984). These data clearly indicate that the metasomatism took place at relatively high temperature and oxygen fugacity in the shallower part of the upper mantle.
Hornblende–phlogopite veins
Distinct alkali metasomatism must have taken place during the formation of the hornblende–phlogopite veins in the olivine clinopyroxenite. The occurrence of calcite in the veins provides evidence for the presence of CO2 in the fluid. A possible reaction for the formation of the hornblende (pargasite) and the phlogopite is
 |
This reaction reveals the presence of an aqueous fluid phase or melt with dissolved alkalis and silica as well as CO2 during metasomatism. The application of the Ti in hornblende thermometry of Colombi (1988) results in temperatures of 680–726°C. However, because of the absence of a Ti-bearing mineral phase such as rutile or ilmenite, which would be necessary for this geothermometer (see Colombi, 1988), this is only a minimum estimate.
Irregular metasomatism
The occurrence of the metasomatized sections near the host trachybasalt, as well as the felt-like texture of the secondary mineral assemblages indicate that the irregular, patchy metasomatism in sample 54-2H formed by the influence of the host magma. A strong thermal overprint also affected the whole sample, as evident from the mineral zoning in spinel and olivine. Furthermore, high temperatures can be estimated for small pyroxene prisms in the metasomatic sections. The T BKN calibration yields temperatures of 1188–1310°C for these pyroxenes, which can clearly be attributed to the thermal influence of the host trachybasalt. The small secondary hornblende needles and aggregates probably developed during a late stage of hydrothermal activity, as indicated by their formation at the expense of primary and secondary pyroxenes and their position near the contact to the host basalt.
Pervasive metasomatism
The xenoliths with pervasive metasomatism underwent the most extensive changes to their primary mineralogy. Almost all primary plagioclase broke down according to the reactions
 |
 |
A calculation of the P–T conditions for these reactions using the TWEEQ program of Berman (1991; activity models in Appendix) yields temperatures in excess of 850°C for the first reaction and of 700°C for the second reaction at 1 kbar with aH2O= 1. A reduction of the water activity results in lower temperatures. Conventional thermobarometry provides much more accurate data to constrain this event. The application of the T BBG calibration for secondary spinel and nearby olivine yields temperatures of 720–790°C. Similar results are obtained by the amphibole–plagioclase thermometer of Spear (1980), which points to temperatures of 725°C. Finally, the application of the P KB calibration to olivine neoblasts occurring in the metasomatic sections results in a temperature range of 680–790°C (P = 2–3 kbar).
|
GEOCHEMISTRY |
|---|
Normalized REE distributions (Fig. 7) and some extended incompatible element patterns (Fig. 8) for the ultramafic xenolith assemblages and their host rock from the TUBAF and EDISON seamounts outline the main geochemical features and provide information about the character of the mantle (see also Table 9). They vary within one (REE) and one and a half (extended patterns) orders of magnitude and are grouped in accordance with their metasomatic type and paragenetic sequences.
|
|
Peridotites without modal metasomatism and spinel peridotites with orthopyroxene veins have the most uniform REE abundance patterns (Fig. 7c), with mostly high (Ce/Yb)n from 3·04 to 7·33 (Table 9). Each pattern displays relatively high light REE (LREE) + Nd, low middle REE (MREE), and non- to weakly elevated heavy REE (HREE). With respect to the HREE and MREE, these peridotites, and even sample 56-2A with orthopyroxene veins, are similar to a serpentinized peridotite from the plutonic complex of the Troodos ophiolite (see Kay & Senechal, 1976). This ophiolite is believed to have been formed in a supra-subduction setting (Bloomer et al., 1995). These ‘highly depleted’ patterns of the refractory MREE, and in part the HREE, contrast markedly with the average patterns of mid-ocean-ridge-type peridotitic mantle residues of the Internal Ligurides (Rampone et al., 1996), and emphasize that the supra-subduction mantle is generally more strongly depleted in refractory REE than the mantle in a mid-ocean ridge setting (Davidson, 1996). Porphyroclastic peridotite xenoliths from the TUBAF and EDISON seamounts yield low LREE and are still more depleted in MREE and HREE (Grégoire et al., 2001) than the coarse equant peridotites of our study. In contrast to the porphyroclastic xenoliths, the patterns of most investigated coarse equant peridotites are kinked at Eu and ‘spoon-shaped’, so that they indicate a stronger enrichment in LREE (see Fig. 7a–c). They also display an enrichment in the large ion lithophile elements (LILE) Rb and Sr (Table 9). The high LILE and LREE contents may be due to hydrous fluid circulation, which is typical in a supra-subduction-zone mantle and may result in cryptic metasomatism (Dawson, 1984; Ryan et al., 1996; Harry & Green, 1999). The imprinted subduction margin setting of these peridotites is confirmed by low Al contents in the primary orthopyroxene and by medium to high Cr contents in spinel (Tables 1 and 4). The investigated minerals plot within or around the arc field within the Al2O3 Opx vs 100Cr/(Cr + Al)Spl diagram (Fig. 9).
|
Veined spinel peridotites that are penetrated by orthopyroxene–hornblende and phlogopite–clinopyroxene veins have the same or even a higher level of LREE enrichment (see sample 56-2X) and are clearly higher in MREE and HREE than non-veined lherzolites, whereas the formation of orthopyroxene veinlets (56-2A) had little effect on the REE distribution (Fig. 7b). As the HREE and part of the MREE are conservative (Pearce & Peate, 1995) and only weakly incompatible to compatible, they should have been virtually excluded from any volatile transport within the mantle wedge. This might be an argument for a silica-rich melt as metasomatic agent for the peridotites with phlogopite- and hornblende-bearing veins.
The phlogopite-bearing spinel lherzolite 56-2X has the highest REE concentrations of all investigated spinel lherzolites and the highest (Ce/Yb)n ratio (7·7). The high HREE content of this rock is consistent with the presence of garnet in the veinlet.
The REE pattern of the irregularly metasomatized harzburgite, 54-2H, is similar to those without modal metasomatism. This rock is strongly enriched in LREE [(Ce/Yb)n = 7·33; see Fig. 7a and Table 9].
The REE patterns of the plagioclase lherzolites and of olivine clinopyroxenite (Fig. 7a) are relatively flat and have chondrite-normalized ratios from 0·6 to 1·3. Only the plagioclase lherzolite 56-2M shows slight enrichments in La and Ce. The elevated Al2O3 and Na2O contents and the mg-number [mg-number =100Mg/(Mg + Fe2+) = 87·2] of the olivine clinopyroxenite are similar to those of the plagioclase lherzolites (see Table 9). These ultramafic rocks (55-2C, 56-2M) as well as the SLGPV (56-2X) have MREE and HREE contents similar to the REE characteristics of the mafic–ultramafic cumulate sequences of the Troodos ophiolite (Kay & Senechal, 1976). Those features are similar to, and suggest a possible genetic relationship with, members of mafic–ultramafic cumulate arc sequences (e.g. from the marginal sea ophiolites at the Kronozky arc, NW Pacific; see Kramer et al., 2000).
The basaltic rock 56-5A from the TUBAF seamount is part of the host rock of the ultramafic xenolith assemblage. This volcanic rock has been classified on the basis of an anhydrous bulk-rock analysis (Table 9) as potassic trachybasalt according to Le Maitre et al. (1989). Its REE pattern shows a regular decrease from LREE to HREE. Its extended incompatible element pattern is characterized by typical subduction-type features, particularly negative high field strength element (HFSE) anomalies as well as high alkali and alkaline earth metals (Fig. 8). This might be due to LILE (Rb, Ba, Sr) and U transportation via aqueous fluid phase at high pressures, and reduced solubility of HFSE-rich phases under high-pressure hydrous conditions (Tatsumi, 1991). Thus, the trachybasalt displays high Ba/Th (123) and U/Th (0·70), low Nb/Zr (0·02) and 0·77 wt % TiO2, comparable with 12–14 Ma calc-alkaline lavas from northern Sulawesi to the west of the recent Philippine plate (Elburg & Foden, 1998).
|
DISCUSSION |
|---|
The mantle xenoliths from TUBAF and EDISON seamounts underwent a complex thermal history, possibly starting at a mid-ocean ridge setting, where the initial formation of the mantle lithosphere occurred by uprise of asthenospheric material (e.g. Rabinowicz et al., 1984
; see Fig. 11a). During its transport away from the ridge, the lithosphere cooled, as evident from zoning and exsolution lamellae in ortho- and clinopyroxene as well as low Ca contents in olivine and high Fe contents in spinel. Cooling rates may be obtained by comparing measured Ca profiles in olivine with results of diffusion calculations. For these calculations, which are based on the exchange of Ca between olivine and clinopyroxene (Köhler & Brey, 1989), a linear cooling rate is assumed [e.g. Lasaga, 1983; for details of the arithmetic method, see Köhler et al. (1991)]. Using the diffusion coefficients for Ca in olivine according to Stahl et al. (1998; diffusion along the c-axis of olivine) the zonation patterns of Ca in olivine of our samples point to cooling rates of 0·1–1°C/Ma (see Fig. 10). The error in the diffusion data accounts for an increase or decrease of the cooling by a factor of 10, which means that even the fastest cooling rate never exceeds 10°C/Ma. Such low cooling rates are reached only when the oceanic lithosphere has drifted a considerable distance away from the mid-ocean ridge, i.e. almost 3000 km at a half-spreading rate of 6 cm/yr applying the thermal modelling of Parsons & Sclater (1977; see also Anderson et al., 1977). During cooling processes, pronounced mineral chemical disequilibria occurred as a consequence of the strongly differing diffusion rates for elements in the primary mantle minerals. Whereas slow diffusion of Mg in pyroxenes led to closing temperatures of 700–900°C, fast diffusion of Ca in olivine allowed a re-equilibration to much lower temperatures. A combination of the pyroxene thermometry and the Ca in olivine barometry therefore gives unrealistically high pressures. Thus, only an intersection of the recent geothermal gradient of the Bismarck Archipelago with the KD lines of the pyroxene thermometry yields realistic thermobarometric results for the spinel peridotites of 665–860°C at 13–18 kbar corresponding to a depth of 43–60 km.
|
|
Cooling of the mantle lithosphere was followed by strong metasomatism, which is evident from thermal and textural re-equilibration of the primary mantle mineral assemblages. Metasomatic processes mainly occurred via discrete veins and generated coarse-grained and granoblastic vein assemblages. The formation of granoblastic equilibrium textures takes a long time at the temperatures recorded for this event and thus rules out a thermal influence of the host basalt during the eruption of the xenoliths. The formation of the veins was, at least in part, accompanied by tectonic deformation as evident from dynamically recrystallized olivine and clinopyroxene neoblasts. According to the P–T estimates of the primary parageneses (i.e. 850°C at 17 kbar; see Fig. 6) and the high metasomatic temperatures of 900–950°C, the spinel lherzolite with orthopyroxene veins (SHOV) probably formed in the lower part of the lithospheric mantle above the subducted slab. According to thermal modelling of subduction heat flow (Peacock, 1990), metasomatic fluids or melts must have emerged in large amounts at least 5 my after the stalling of the subduction zone, when the slab was heated to temperatures far beyond 500°C (see Fig. 11b and c). The absence of alkali-bearing metasomatic minerals in these veins points to a low salinity of the fluids. At shallower levels, where temperatures during metasomatism were distinctly lower (660–800°C), alkali-bearing minerals such as amphibole and phlogopite, as well as carbonate, were generated. These observations are in good agreement with the model of Eggler (1987) for metasomatic processes in the mantle, which postulates a high water activity in lower levels of the lithospheric mantle, which decreases at the expense of CO2 in higher levels. The generation of alkali-bearing minerals in shallower levels of the mantle, as observed in the plagioclase lherzolites, the olivine clinopyroxenite and the spinel lherzolite with the garnet–phlogopite vein (PLPM, OCPHV, SLGPV) may be interpreted as the consequence of reactions of alkali-rich fluids with the surrounding peridotite. Such processes were predicted by a study of Schneider & Eggler (1986), which described the steady enrichment in alkalis of fluids emerging from subduction zones during their ascent. According to those workers, upward-moving solute-undersaturated hydrous fluids may leach deeper parts of the lithosphere by dissolving orthopyroxene and phlogopite. In the shallow mantle lithosphere, changing physical and chemical conditions trigger precipitation of minerals such as phlogopite or amphibole.
Mantle metasomatism was connected with a high oxygen fugacity as indicated by log(fO2)FMQ values of 0·4 to >4, which could be expected for a water-bearing silicate melt or more likely for a solute-rich hydrous fluid from the subducted slab (Parkinson & Arculus, 1999). Similarly high oxygen fugacities were recorded in cogenetic volcanics from the neighbouring Lihir island (Müller et al., 2001).
A remarkable feature of the metasomatic ultramafic rocks is the presence of distinct mineral chemical disequilibria between the vein parageneses and the host minerals. Only in very few cases was compositional re-equilibration observed in the immediate vicinity of the vein, which highlights the channelled flow regime of the metasomatic fluids. There was no indication of porous flow in the sense of Watson & Brenan (1987), and even transmission electron microscope studies did not give any indication for the presence of melt on grain boundaries near the metasomatic veins (R. Wirth, unpublished data, 1999). Metasomatic minerals often displayed numerous primary, mostly decrepitated, fluid inclusions. The only melt inclusions detected in the investigated xenoliths occurred in fractures and were derived from the trachybasaltic host rock (Franz & Wirth, 2000). These observations are in contradiction to current models of mantle metasomatism (e.g. Foley, 1992), which highlight as important the interaction of the veins with the wall rock. At least in our xenoliths sampled by trachybasalts, vein–host rock interactions are minor, which, however, does not have to be the case at deeper levels of the mantle. An estimate of the duration of the fluid flow and its thermal effects on neighbouring minerals is possible for the spinel lherzolite with orthopyroxene–hornblende veins (SLOHV, sample 56-2B), where the rim of an olivine grain adjacent to a vein shows a resetting of its Ca content (Fig. 5). Assuming diffusion along the c-axis of the olivine, the time for the formation of the zoning can be calculated using the diffusion data of Stahl et al. (1998). At a temperature of 800°C for the metasomatic event (as evidenced from the thermometry), a duration of 0·55 my is calculated using the diffusion equation of Crank (1975). Similar results were achieved for zoned olivine grains close to the garnet-bearing phlogopite vein of the SLGPV (sample 56-2X). However, the relatively large error of the method results in a range of 0·07–4·65 my for the formation of the rim zoning. Furthermore, the influence of the oxygen fugacity, which should increase the diffusion rate (see Jurewicz & Watson, 1988) was not taken into account. Despite this large error range, it is evident that the influence of the metasomatic agents was not a short-term process but prevailed at least for 70 ky. This is a further argument for a solvent-rich hydrous fluid phase as metasomatic agent; according to Eggler (1987), the formation of metasomatic veins occurs quickly in the presence of a melt, whereas it takes a considerably longer time to generate them in the presence of a hydrous fluid phase.
|
CONCLUSIONS |
|---|
- All investigated ultramafic xenoliths in the trachybasalt from TUBAF and EDISON seamounts are derived from modally or cryptically metasomatized upper-mantle sections. Modal metasomatism occurred in the mantle as a result of the intrusion of a solute-rich, hydrous metasomatic agent from the underlying mantle wedge. The hydration of the mantle in combination with Quaternary lithospheric spreading is also responsible for the formation of the mafic magma, which builds up Lihir island and the neighbouring seamounts (see Fig. 11).
- In most cases, mineral textures of the modally metasomatized xenoliths and temperature during the metasomatic event (660–950°C) rule out any thermal or metasomatic influence of the host trachybasalt. According to thermal modelling, the influence of the metasomatizing agents lasted for a relatively long time (at least 70 ky). Only in very few cases can an overprint, related to the activity of the host magma, be observed in the form of irregular and patchy metasomatism and contact metamorphism.
- Metasomatism can be observed from the formation depth of the host trachybasalt (60–80 km) to the shallow lithospheric mantle, i.e. the stability field of plagioclase lherzolite.
- Whereas the metasomatic agents penetrated the lower part of the lithospheric mantle in veins and thus were subjected to a channelled flow regime, metasomatic interactions in the upper lithospheric mantle occurred in a penetrative way. Remarkable also is the high oxygen fugacity, which is recorded near the metasomatic sections of all samples.
- The localized character of the metasomatism is most pronounced in the veined peridotites. In most cases, there is no equilibrium between the metasomatic minerals in the veins and the primary minerals of the hosting xenoliths. Mineral chemical re-equilibration is limited to the close vicinity of the veins and affected only minerals with fast diffusing elements (e.g. Fe and Mg in spinel and Fe, Mg and Ca in olivine).
- The ultramafic xenoliths can be divided into different metasomatic groups on the basis of their geochemical characteristics, particularly their REE patterns. These subdivisions are also reflected in the mineralogy and texture of the xenoliths. Olivine clinopyroxenites and pervasively metasomatized plagioclase lherzolites, which originate from a cumulate complex in the uppermost mantle, yield high contents of MREE and HREE. All investigated spinel peridotites are enriched in LREE. Whereas cryptically metasomatized peridotites display ‘spoon-shaped’ REE patterns, veined spinel peridotites are moderately enriched in MREE and HREE.
- Most mineralogical arguments point to a hydrous fluid phase as the metasomatic agent. The strongest line of evidence is given by the presence of primary, aqueous fluid inclusions in the metasomatic minerals and the absence of any melt inclusions as well as the lack of melt on grain boundaries. Balanced mineral reactions show that metasomatism occurred in the presence of a hydrous fluid phase with variable amounts of dissolved alkali cations. The lengthy time of metasomatic influence on the samples as well as results from LA ICP-MS studies support these arguments (Grégoire et al., 2001).
- Peridotite xenoliths from the TUBAF and EDISON seamounts reveal part of their early history by retrograde zoning and exsolution features. Thermal modelling based on Ca in olivine zoning gives evidence for very slow cooling processes, which are supposed to have taken place after the initial formation of peridotitic mantle residues at a mid-ocean ridge (see Fig. 11). Subsequent to this cooling, a depletion took place by partial melting in a supra-subduction setting, which is reflected by low MREE, Al decrease and Cr increase in spinel peridotites. Finally, a strong cryptic to modal metasomatism affected and re-enriched the mantle peridotites. These metasomatic processes, which are indicated by metasomatic mineral assemblages and significant LREE enrichment, terminated the three-stage history of the mantle peridotites.
|
APPENDIX |
|---|
Mineral formulae used for the balancing of the mineral reactions: albite, NaAlSi3O8; anorthite, CaAl2Si2O8; diopside, CaMgSi2O6; Ca-Tschermaks component in diopside, CaAl2SiO6; enstatite, Mg2Si2O6; Mg-Tschermaks component in enstatite, MgAl2SiO6; forsterite, Mg2SiO4; grossular, Ca3Al2Si3O12; pyrope, Mg3Al2Si3O12; pargasite, NaCa2Mg4Al3Si6O24H2; tschermakite, Ca2Mg3Al4Si6O24H2; tremolite, Ca2Mg5Si8O24H2; phlogopite, KMg3AlSi3O12H2; quartz, SiO2; calcite, CaCO3; spinel, MgAl2O4.
Activity models for the calculation of the reaction curves with the TWEEQ program of Berman (1991; dataset jun92.gsc):
- albite and anorthite are from Fuhrman & Lindsley (1988);
- diopside, enstatite, pargasite and tschermakite are ideal;
- forsterite is from O’Neill & Wall (1987);
- spinel is from Nichols et al. (1992).
|
ACKNOWLEDGEMENTS |
|---|
The authors thank the master, officers and crew of R.V. Sonne for the logistical support and good co-operation during our research activities in the New Ireland basin. Thanks are also due to the staff of the Geochemical Department of the GFZ Potsdam and Mario Drischel (Freiberg) for help with the geochemistry and the preparation of the samples. Gerhard Brey (Frankfurt), Horst Palme (Köln), Richard Wirth (Potsdam), Mike Perfit (Gainesville) and Roger Hekinian (Kiel–Brest) supported us with important information and fruitful discussions. We owe great thanks to Klaus Mezger (Münster), who supplied the computer program for the calculation of the cooling rates with the Lasaga algorithm. We appreciate the thorough and constructive reviews of Pamela Kempton, Martina Elburg, Michel Grégoire and an anonymous reviewer. The English language of this study was considerably improved by Randall Keller (Oregon State University). Principal funding for this project was provided by the German Federal Ministry for Education and Science (BMBF Grant 03G0133A to P.M.H.) and by the Leibniz Program of the German Research Foundation.
|
FOOTNOTES |
|---|
*Corresponding author. E-mail: lfranz{at}mineral.tu-freiberg.de.
|
REFERENCES |
|---|
Anderson, R. N., Langseth, M. F. & Scalter, J. G. (1977). The mechanisms of heat transfer through the flow of the Indian Ocean. Journal of Geophysical Research 82, 3391–3409.[Web of Science]
Ballhaus, C., Berry, R. F. & Green, D. H. (1991). High pressure experimental calibration of the olivine–orthopyroxene–spinel oxygen geobarometer: implications for the oxidation state of the upper mantle. Contributions to Mineralogy and Petrology 107, 27–40.
Berman, R. G. (1991). Thermobarometry using multiequilibrium calculations: a new technique with petrologic applications. Canadian Mineralogist 29, 833–855.[Web of Science]
Bloomer, S. H., Taylor, B., MacLeod, C. J., Stern, R. J., Fryer, P., Hawkins, J. W. & Johnson, L. (1995). Early arc volcanism and the ophiolite problem: a perspective from the drilling in the Western Pacific. In: Taylor, B. & Natland, J. (eds) Active Margins and Marginal Basins of the Western Pacific. Geophysical Monographs, American Geophysical Union 88, 1–30.
Bonatti, E. & Michael, P. J. (1989). Mantle peridotites from continental rifts to oceanic basins to subduction zones. Earth and Planetary Science Letters 91, 297–311.[Web of Science]
Bonatti, E., Hamlyn, P. & Ottonello, G. (1981). Upper mantle beneath a young ocean rift; peridotites from the island of Zabargad (Red Sea). Geology 9, 474–479.
Brady, J. B. (1995). Diffusion data for silicate minerals, glasses and liquids. In: Ahrens, T. H. (ed.) Handbook of Physical Constants. Washington, DC: American Geophysical Union, pp. 269–290.
Brey, G. P. & Köhler, T. (1990). Geothermobarometry in four-phase lherzolites II. New thermobarometers, and practical assessment of existing thermobarometers. Journal of Petrology 31, 1353–1378.
Churikova, T., Andre, L. & Koloskov, A. (1999). Across-arc variations for mantle xenoliths compositions in Kamchatka (Russia). Terra Abstracts 4, 359.
Coleman, P. J. & Kroenke, L. W. (1981). Subduction without volcanism in the Solomon Island arc. Geo-Marine Letters 1, 129–134.
Colombi, A. (1988). Métamorphisme et géochemie des roches mafiques des Alpes ouest-centrales (géoprofil Viège–Domodossola–Locarno). Ph.D. thesis, University of Lausanne, 216 pp.
Crank, J. (1975). The Mathematics of Diffusion. London: Clarendon Press, 414 pp.
Dautria, J. M., Liotard, J. M., Cabanes, N., Girod, M. & Briqueu, L. (1987). Amphibole-rich xenoliths and host alkali basalts: petrogenetic constraints and implications on the recent evolution of the upper mantle beneath Ahaggar (Central Sahara, Southern Algeria). Contributions to Mineralogy and Petrology 95, 133–144.
Davidson, J. P. (1996). Deciphering mantle and crustal signatures in subduction zone magmatism. In: Bebout, G. E., Scholl, D. W., Kirby, S. H. & Platt, J. P. (eds) Subduction Top to Bottom. Geophysical Monograph, American Geophysical Union 96, 251–262.
Dawson, J. B. (1984). Contrasting types of upper mantle metasomatism? In: Kornprobst, J. (ed.) Kimberlites II: The Mantle and Crust–Mantle Relations, Developments in Petrology 11b. Amsterdam: Elsevier, pp. 277–287.
Draper, D. S., Brandon, A. D. & Becker, H. (1999). Interaction between slab and sub-arc mantle: dehydration, melting and element transport in subduction zones. Special Issue, AGU Symposium, Session V13 Chemical Geology 160, 249–446.
Droop, G. T. B. (1987). A general equation for estimating Fe3+ concentrations in ferromagnesian silicates and oxides from microprobe analyses, using stoichiometric criteria. Mineralogical Magazine 51, 431–435.
Eggler, D. H. (1987). Solubility of major and trace elements in mantle metasomatic fluids: experimental constraints. In: Menzies, M. A. & Hawkesworth, C. J. (eds) Mantle Metasomatism. London: Academic Press, pp. 21–41.
Elburg, M. & Foden, J. (1998). Temporal changes in arc magma geochemistry, northern Sulawesi, Indonesia. Earth and Planetary Science Letters 163, 381–398.
Evensen, N. M., Hamilton, P. J. & O’Nions, R. K. (1978). Rare earth abundances in chondritic meteorites. Geochimica et Cosmochimica Acta 42, 1199–1212.[Web of Science]
Foley, S. F. (1992). Vein-plus-wall-rock melting mechanisms in the lithosphere and the origin of potassic alkaline magmas. Lithos 28, 435–453.[Web of Science]
Franz, L. & Wirth, R. (1997). Thin intergranular melt films and melt pockets in spinel peridotite xenoliths from the Rhön area (Germany): early stage of melt generation by grain boundary melting. Contributions to Mineralogy and Petrology 129, 268–283.
Franz, L. & Wirth, R. (2000). Spinel inclusion in olivine of peridotite xenoliths from TUBAF seamount (Bismarck Archipelago/Papua New Guinea): evidence for the thermal and tectonic evolution of the oceanic lithosphere. Contributions to Mineralogy and Petrology 140, 283–295.
Frey, F. A. & Prinz, M. (1978). Ultramafic inclusions from San Carlos, Arizona: petrologic and geochemical data bearing on their petrogenesis. Earth and Planetary Science Letters 38, 126–176.
Fuhrman, M. L. & Lindsley, D. H. (1988). Ternary feldspar modeling and thermometry. American Mineralogist 73, 201–215.[Abstract]
Gasparik, T. (1984). Two-pyroxene thermobarometry with new experimental data in the system CaO–MgO–Al2O3–SiO2. Contributions to Mineralogy and Petrology 87, 87–97.[Web of Science]
Graham, C. M. & Powell, R. (1986). A garnet–hornblende geothermometer: calibration, testing, and application to the Pelona schists, southern California. Journal of Metamorphic Geology 2, 13–31.
Grégoire, M., McInnes, B. & O’Reilly, S. (2001). Hydrous metasomatism of oceanic sub-arc mantle, Lihir, Papua New Guinea, Part 2: trace element characteristics of slab-derived fluids. Lithos (in press).
Harry, D. L. & Green, N. L. (1999). Slab dehydration and basalt petrogenesis in subduction systems involving very young oceanic lithosphere. Chemical Geology 160, 309–333.
Harte, B. (1977). Rock nomenclature with particular relation to deformation and recrystallization textures in olivine-bearing xenoliths. Journal of Geology 85, 279–288.
Harte, B. (1983). Mantle peridotites and processes: the kimberlite sample. In: Hawkesworth, C. J. & Norry, M. J. (eds) Continental Basalts and Mantle Xenoliths. Nantwich: Shiva, pp. 46–91.
Hawkesworth, C. J., van Calsteren, P., Rogers, N. W. & Menzies, M. A. (1987). Isotope variations in recent volcanics: a trace-element perspective. In: Menzies, M. A. & Hawkesworth, C. J. (eds) Mantle Metasomatism. London: Academic Press, pp. 365–388.
Herzig, P. M. & Becker, K.-P. (1996). Tectonics, hydrology, and hydrothermal processes in areas of alkaline island-arc volcanoes in the southwest Pacific: the Tabar–Lihir–Tanga–Feni Island chain, Papua New Guinea. Final Report Sonne SO-94, 289 pp.
Herzig, P. M., Hannington, M. D., McInnes, B., Stoffers, P., Villinger, H., Seifert, R., Binns, R., Liebe, T. & Scientific Party R.V. Sonne Cruise SO-94 (1994). Submarine volcanism and hydrothermal venting studied in Papua, New Guinea. EOS Transactions, American Geophysical Union 75, 513–516.
Herzig, P. M., Hannington, M. D., Stoffers, P., Becker, K.-P., Drischel, M., Franklin, J., Franz, L., Gemmel, J. B., Hoeppner, B., Horn, C., Horz, K., Jellineck, T., Jonasson, I. R., Kia, P., Nickelsen, S., Percival, J., Perfit, M., Petersen, S., Schmidt, M., Seifert, T., Thießen, O., Türkay, M., Tunnicliffe, V. & Winn, K. (1998). Petrology, gold mineralization and biological communities at shallow submarine volcanoes of the New Ireland fore-arc (Papua New Guinea): preliminary results of R/V Sonne cruise SO-133. Inter Ridge News 7, 34–38.
Hofmann, A. W. (1988). Chemical differentiation of the earth: the relationship between mantle, continental crust and oceanic crust. Earth and Planetary Science Letters 90, 243–262.
Ionov, D. A., Hofmann, A. W. & Shimizu, N. (1994). Metasomatism-induced melting in mantle xenoliths from Mongolia. Journal of Petrology 35, 753–785.
Jagoutz, E., Palme, H., Baddenhausen, H., Blum, K., Cendales, M., Dreibus, G., Spettel, B., Lorenz, V. & Wänke, H. (1979). The abundances of major, minor and trace elements in the Earth’s mantle as derived from primitive ultramafic nodules. In: Merrill, R. B., Bogard, D. D., Hoerz, F., McKay, D. S. & Robertson, P. C. (eds) Early Solar System and Lunar Regolith. Proceedings of the 10th Lunar and Planetary Science Conference, 2. Geochimica et Cosmochimica Acta Supplement 2031–2050.
Jarosewich, E., Nelen, J. A. & Norberg, J. A. (1980). Reference samples for electron microprobe analysis. Geostandards Newsletter 4, 43–47.
Jurewicz, A. J. G. & Watson, E. B. (1988). Cations in olivine, Part 2: diffusion in olivine xenocrysts with applications to petrology and mineral physics. Contributions to Mineralogy and Petrology 99, 186–201.
Kay, R. W. & Senechal, R. G. (1976). The rare earth element geochemistry of the Troodos ophiolite complex. Journal of Geophysical Research 81, 964–970.[Web of Science]
Kempton, P. D. (1987). Mineralogic and geochemical evidence for differing styles of metasomatism in spinel lherzolite xenoliths: enriched mantle source regions of basalts? In: Menzies, M. A. & Hawkesworth, C. J. (eds) Mantle Metasomatism. London: Academic Press, pp. 45–123.
Kempton, P. D., Lopez-Escobar, L., Hawkesworth, C. J., Pearson, D. G., Wright, D. W. & Ware, A. J. (1999a). Spinel ± garnet lherzolite xenoliths from Pali Aike, Part 1: petrography, mineral chemistry and geothermobarometry. In: Gurney, J. J., Gurney, J. L., Pascoe, M. D. & Richardson, S. H. (eds) Proceedings of the VIIth International Kimberlite Conference (Dawson Volume). University of Cape Town: Red Roof Design, pp. 403–414.
Kempton, P. D., Hawkesworth, C. J., Lopez-Escobar, L., Pearson, D. G. & Ware, A. J. (1999b). Spinel ± garnet lherzolite xenoliths from Pali Aike, Part 2: trace element and isotopic evidence bearing on the evolution of lithospheric mantle beneath southern Patagonia. In: Gurney, J. J., Gurney, J. L., Pascoe, M. D. & Richardson, S. H. (eds) Proceedings of the VIIth International Kimberlite Conference (Dawson volume). University of Cape Town: Red Roof Design, pp. 415–428.
Kennedy, A. K., Hart, S. R. & Frey, F. A. (1990). Composition and isotopic constraints on the petrogenesis of alkaline arc lavas: Lihir island, Papua New Guinea. Journal of Geophysical Research 95, 6929–6942.
Kepezhinskas, P. K., Taylor, R. N. & Tanaka, H. (1993). Geochemistry of plutonic spinels from the North Kamchatka Arc: comparison with spinels from other tectonic settings. Mineralogical Magazine 57, 575–589.
Kepezhinskas, P. K., Defant, M. J. & Drummond, M. S. (1995). Na metasomatism in the island-arc mantle by slab melt–peridotite interaction: evidence from mantle xenoliths in the North Kamchatka arc. Journal of Petrology 36, 1505–1527.
Kitahara, S., Takenouchi, S. & Kennedy, G. C. (1966). Phase relations in the system MgO–SiO2–H2O at high temperatures and pressures. American Journal of Science 264, 223–233.[Web of Science]
Köhler, T. & Brey, G. P. (1990). Calcium exchange between olivine and clinopyroxene calibrated as a geothermobarometer for natural peridotites from 2 to 60 kb with applications. Geochimica et Cosmochimica Acta 54, 2375–2388.[Web of Science]
Köhler, T., Palme, H. & Brey, G. P. (1991). Determination of cooling rates using calcium exchange between olivine and clinopyroxene. Neues Jahrbuch für Mineralogie, Monatshefte 9, 423–431.
Kramer, W., Zimmer, M., Skolotnev, S. G., Tsukanov, N. V., Gaedicke, C. & Freitag, R. (2000). Geochemistry of the mafic–ultramafic complexes of the Kamchatka Cape peninsula. Abstracts of the Scientific Meeting of the Kamchatka Division of the Russian Mineralogical Society, April 2000. Petropavlovsk-Kamchatsky: DVORAN (Russian Academy of Science).
Kretz, R. (1983). Symbols for rock-forming minerals. American Mineralogist 68, 277–279.[Abstract]
Krogh, E. J. (2000). The garnet–clinopyroxene geothermometer: an updated calibration. Journal of Metamorphic Geology 18, 211–219.[Web of Science]
Lasaga, A. C. (1983). Geospeedometry: an extension of geothermometry. In: Saxena, S. K. (ed.) Kinetics and Equilibrium in Mineral Reactions, Advances in Physics and Geochemistry, Vol. 3. Berlin: Springer, 81–114.
Leake, B. W., Wooley, A. R., Arps, C. E. S., Birch, W. D., Gilbert, M. C., Grice, J. D., Hawthorne, F. C., Kato, A., Kisch, H. J., Krivovichev, V. G., Linthout, K., Laird, J., Mandarino, J., Maresch, W. V., Nickel, E. H., Rock, N. M. S., Schumacher, J. C., Smith, D. C., Stephenson, N. C. N., Ungaretti, L., Whittaker, E. J. W. & Youzhi, G. (1997). Nomenclature of amphiboles. Report of the subcommittee on amphiboles of the International Mineralogical Association Commission on new minerals and mineral names. European Journal of Mineralogy 9, 623–651.[Web of Science]
Le Maitre, R. W. (1989). A Classification of Igneous Rocks and Glossary of Terms. Recommendations of the International Union of Geological Sciences Subcommission on the Systematics of Igneous Rocks. Oxford: Blackwell Scientific, 193 pp.
Licence, P. S., Terril, J. E. & Fergusson, L. J. (1987). Epithermal gold mineralisation, Ambittle Island, Papua New Guinea. Proceedings of the Pacific Rim Conference ’87. Melbourne: Australian Institute of Mining and Metallurgy, pp. 273–278.
Lindsley, D. H. (1983). Pyroxene thermometry. American Mineralogist 68, 477–493.[Abstract]
Maury, R. C., Defant, M. J. & Joron, J.-L. (1992). Metasomatism of the sub-arc mantle inferred from trace elements in Philippine xenoliths. Nature 360, 661–663.
McInnes, B. I. A. (1992). A glimpse of ephemeral subduction zone processes from Simberi island, Papua New Guinea. Ph.D. thesis, University of Ottawa, 235 pp.
McInnes, B. I. A. & Cameron, E. M. (1994). Carbonated, alkaline metasomatic melts from a sub-arc environment: mantle wedge samples from the Tabar–Lhir–Feni arc, Papua New Guinea. Earth and Planetary Science Letters 122, 125–141.
McInnes, B. I. A., Herzig, P. M., Hannington, M. D. & Binns, R. A. (1994). Modern oceanic fore-arc lithosphere resembles ophiolite sequences. EOS Transactions, American Geophysical Union 75, 747.
McInnes, B. I. A., McBride, J. S., Evans, N. J., Lambert, D. D. & Andrew, A. S. (1999). Osmium isotope constraints on ore metal recycling in subduction zones. Science 286, 512–517.
McInnes, B. I. A., Gregoire, M., Binns, R. A., Herzig, P. M. & Hannington, M. D. (2001). Hydrous metasomatism of oceanic sub-arc mantle, Lihir, Papua New Guinea: petrology and geochemistry of fluid-metasomatized mantle wedge xenoliths. Earth and Planetary Science Letters 188, 169–183.[Web of Science]
Müller, D., Franz, L., Herzig, P. & Hunt, S. (2001). Potassic igneous rocks from the vicinity to epithermal gold mineralization at Lihir Island, Papua New Guinea. Lithos 57, 163–186.
Newton, R. C. & Perkins, D. I. (1982). Thermodynamic calibration of geobarometers based on the assemblages garnet–plagioclase–orthopyroxene (clinopyroxene)–quartz. American Mineralogist 67, 203–222.[Abstract]
Nichols, G. T., Berry, R. F. & Green, D. (1992). Internally consistent gahnitic spinel–cordierite–garnet equilibria in the FMASHZn system: geothermobarometry and applications. Contributions to Mineralogy and Petrology 111, 362–377.
Olafsson, M. & Eggler, D. H. (1983). Phase relations of amphibole, amphibole–carbonate, and phlogopite–carbonate peridotite: petrologic constraints on the asthenosphere. Earth and Planetary Science Letters 64, 305–315.
O’Neill, H. St. C. (1981). The transition between spinel lherzolite and garnet lherzolite and its use as a geobarometer. Contributions to Mineralogy and Petrology 77, 185–194.[Web of Science]
O’Neill, H. St. C. & Wall, V. J. (1987). The olivine–orthopyroxene–spinel oxygen barometer, the nickel precipitation curve, and the oxygen fugacity of the earth’s upper mantle. Journal of Petrology 28, 1169–1191.
Papike, J. J., Ross, M. & Baldwin, K. (1974). Amphiboles and pyroxenes: characterization of other than quadrilateral components and estimates of ferric iron from microprobe data. Geological Society of America, Abstracts with Programs 6, 1053–1054.
Parkinson, I. J. & Arculus, R. J. (1999). The redox state of subduction zones: insight from arc peridotites. Chemical Geology 160, 409–423.
Parsons, B. & Sclater, J. G. (1977). An analysis of the variation of ocean floor bathymetry and heat flow with age. Journal of Geophysical Research 82, 803–827.
Peacock, S. M. (1990). Numerical simulation of metamorphic pressure–temperature time paths and fluid production in subducting slabs. Tectonics 9, 1197–1211.[Web of Science]
Pearce, J. A. & Peate, D. W. (1995). Tectonic implications of the composition of volcanic arc magmas. Annual Review of Earth and Planetary Sciences 24, 251–285.
Pollack, H. N. & Chapman, D. S. (1977). On the regional variation of heat flow, geotherms, and lithospheric thickness. Tectonophysics 38, 279–296.[Web of Science]
Rabinowicz, M., Nicolas, A. & Vigneresse, J. L. (1984). A rolling mill effect in asthenosphere beneath oceanic spreading centers. Earth and Planetary Science Letters 67, 97–108.
Rampone, E., Hofmann, A. W., Piccardo, G. B., Vannucci, R., Bottazzi, P. & Ottolini, L. (1996). Trace element and isotope geochemistry of depleted peridotites from an N-MORB type ophiolite (Internal Liguride, N. Italy). Contributions to Mineralogy and Petrology 123, 61–76.
Ryan, J., Morris, J., Bebout, G. & Leeman, B. (1996). Describing chemical fluxes in subduction zones: Insights from ‘depth-profiling’ studies of arc and forearc rocks. In: Bebout, G. E., Scholl, D. W., Kirby, S. H. & Platt, J. P. (eds) Subduction Top to Bottom, Geophysical Monograph, American Geophysical Union 96, 251–262.
Rytuba, J. J., McKee, E. H. & Cox, D. (1993). Geochronology and geochemistry of the Ladolam gold deposit, Lihir island, and gold deposits and volcanoes of Tabar and Tatau, Papua New Guinea. US Geological Survey Bulletin 2039, 119–126.
Schneider, M. E. & Eggler, D. H. (1986). Fluids in equilibrium with peridotite minerals; implications for mantle metasomatism. Geochimica et Cosmochimica Acta 50, 711–724.[Web of Science]
Spear, F. S. (1980). The NaSi = CaAl exchange equilibria between plagioclase and amphibole. Contributions to Mineralogy and Petrology 72, 33–41.
Stahl, S., Hain, A., Chakraborty, S., Laqua, W. & Palme, H. (1998). Tracer diffusion of 42Ca in olivine as a function of composition and pO2 between 800°C and 1500°C. EOS Transactions, American Geophysical Union 79(17, Suppl.), 370.
Steward, W. D. & Sandy, M. J. (1988). Geology of New Ireland and Djaul Islands, northeast Papua New Guinea. In: Marlow, M. S., Dadisman, S. V. & Exton, N. F. (eds) Geology and Offshore Resources of Pacific Island Arcs—New Ireland and Manus Region, Papua New Guinea. Circum Pacific Council for Energy and Mineral Resources, Earth Science Series 9, 12–30.
Stosch, H. G. & Seck, H. A. (1980). Geochemistry and mineralogy of two spinel peridotite suites from Dreiser Weiher, West Germany. Geochimica et Cosmochimica Acta 44, 457–470.[Web of Science]
Szabó, C., Bodnar, R. J. & Sobolev, A. V. (1996). Metasomatism associated with subduction related volatile-rich silicate melt in the upper mantle beneath the Nógrád–Gömör Volcanic Field, Northern Hungary/Southern Slovakia: evidence from silicate melt inclusions. European Journal of Mineralogy 8, 881–899.[Web of Science]
Tatsumi, Y. (1991). Origin of subduction zone magmas based on experimental petrology. In: Perchuk, L. L. & Kushiro, I. (eds) Physical Chemistry of Magmas. Advances in Physical Geochemistry 9, 268–301.
Taylor, B. (1979). Bismarck Sea: evolution of a back arc basin. Geology 7, 171–174.
Vaselli, O., Downes, H., Thirlwall, M., Dobosi, G., Coradossi, I., Seghedi, I., Szakacs, A. & Vannucci, R. (1995). Ultramafic xenoliths in Plio-Pleistocene alkali basalts from Eastern Transylvanian Basin: depleted mantle enriched by vein metasomatism. Journal of Petrology 36, 23–53.
Wallace, D. A., Johnson, R. W., Chappel, B. W., Arculus, R. J., Perfit, M. R. & Crick, I. H. (1983). Cainozoic Volcanism of the Tabar–Lihir, Tanga and Feni islands, Papua New Guinea. Report. Canberra, A.C.T.: Bureau of Mineral Resources, Geology and Geophysics, Australian Geological Survey Organisation, 243 pp.
Watson, E. B. & Brenan, J. M. (1987). Fluids in the lithosphere. I. Experimentally determined wetting characteristics of CO2–H2O fluids and their implications for fluid transport, host rock physical properties, and fluid inclusion formation. Earth and Planetary Science Letters 85, 497–515.[Web of Science]
Witt-Eickschen, G. & Kramm, U. (1997). Mantle upwelling and metasomatism beneath Central Europe: geochemical and isotopic constraints from mantle xenoliths from the Rhön (Germany). Journal of Petrology 38, 479–493.
Wulff-Pedersen, E., Neumann, E.-R. & Jensen, B. B. (1996). The upper mantle under La Palma, Canary Islands: formation of Si–K–Na-rich melt and its importance as a metasomatic agent. Contributions to Mineralogy and Petrology 125, 113–139.[Web of Science]
Wyllie, P. J. (1981). Plate tectonics and magma genesis. Geologische Rundschau 70, 128–153.
Zinngrebe, E. & Foley, S. F. (1995). Metasomatism in mantle xenoliths from Gees, West Eifel, Germany: evidence for the genesis of calc-alkaline glasses and metasomatic Ca-enrichment. Contributions to Mineralogy and Petrology 122, 79–96.[Web of Science]
Zipfel, J. & Wörner, G. (1992). Four- and five-phase peridotites from a continental rift system: evidence for upper mantle uplift and cooling at the Ross Sea margin (Antarctica). Contributions to Mineralogy and Petrology 111, 24–36.
Zuleger, E. & Erzinger, J. (1988). Determination of the REE and Y in silicate materials with ICP-AES. Fresenius Zeitschrift für Analytische Chemie 332, 140–143.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Facer, H. Downes, and A. Beard In situ Serpentinization and Hydrous Fluid Metasomatism in Spinel Dunite Xenoliths from the Bearpaw Mountains, Montana, USA J. Petrology, August 1, 2009; 50(8): 1443 - 1475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. Maria and J. F. Luhr Lamprophyres, Basanites, and Basalts of the Western Mexican Volcanic Belt: Volatile Contents and a Vein-Wallrock Melting Relationship J. Petrology, December 17, 2008; (2008) egn060v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. S. C. Simon, E.-R. Neumann, C. Bonadiman, M. Coltorti, G. Delpech, M. Gregoire, and E. Widom Ultra-refractory Domains in the Oceanic Mantle Lithosphere Sampled as Mantle Xenoliths at Ocean Islands J. Petrology, June 1, 2008; 49(6): 1223 - 1251. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Arai and S. Ishimaru Insights into Petrological Characteristics of the Lithosphere of Mantle Wedge beneath Arcs through Peridotite Xenoliths: a Review J. Petrology, April 1, 2008; 49(4): 665 - 695. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ishimaru and S. Arai Calcic amphiboles in peridotite xenoliths from Avacha volcano, Kamchatka, and their implications for metasomatic conditions in the mantle wedge Geological Society, London, Special Publications, January 1, 2008; 293(1): 35 - 55. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ishimaru, S. Arai, Y. Ishida, M. Shirasaka, and V. M. Okrugin Melting and Multi-stage Metasomatism in the Mantle Wedge beneath a Frontal Arc Inferred from Highly Depleted Peridotite Xenoliths from the Avacha Volcano, Southern Kamchatka J. Petrology, February 1, 2007; 48(2): 395 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Polat, C. Herzberg, C. Munker, R. Rodgers, T. Kusky, J. Li, B. Fryer, and J. Delaney Geochemical and petrological evidence for a suprasubduction zone origin of Neoarchean (ca. 2.5 Ga) peridotites, central orogenic belt, North China craton Geological Society of America Bulletin, July 1, 2006; 118(7-8): 771 - 784. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. MEDARD, M. W. SCHMIDT, P. SCHIANO, and L. OTTOLINI Melting of Amphibole-bearing Wehrlites: an Experimental Study on the Origin of Ultra-calcic Nepheline-normative Melts J. Petrology, March 1, 2006; 47(3): 481 - 504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ikehata, K. Ikehata, and S. Arai Metasomatic formation of kosmochlor-bearing diopside in peridotite xenoliths from North Island, New Zealand American Mineralogist, October 1, 2004; 89(10): 1396 - 1404. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. MELCHER and T. MEISEL A Metamorphosed Early Cambrian Crust-Mantle Transition in the Eastern Alps, Austria J. Petrology, August 1, 2004; 45(8): 1689 - 1723. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. D. Renno, L. Franz, T. Witzke, and P. M. Herzig THE COEXISTENCE OF MELTS OF HYDROUS COPPER CHLORIDE, SULFIDE AND SILICATE COMPOSITIONS IN A MAGNESIOHASTINGSITE CUMULATE, TUBAF SEAMOUNT, PAPUA NEW GUINEA Can Mineral, February 1, 2004; 42(1): 1 - 16. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. ARAI, S. TAKADA, K. MICHIBAYASHI, and M. KIDA Petrology of Peridotite Xenoliths from Iraya Volcano, Philippines, and its Implication for Dynamic Mantle-Wedge Processes J. Petrology, February 1, 2004; 45(2): 369 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Petersen, S. Petersen, P. M. Herzig, M. D. Hannington, I. R. Jonasson, and A. Arribas Jr. Submarine Gold Mineralization Near Lihir Island, New Ireland Fore-Arc, Papua New Guinea Economic Geology, December 1, 2002; 97(8): 1795 - 1813. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||